From ddb24fc525ddaf35130d96478f3f18682a9d5926 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 16:30:56 +0200
Subject: [PATCH 0001/1791] drm/i915/ttm: Add I915_BO_PREALLOC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a mechanism to preserve existing data when creating
a TTM object with the I915_BO_ALLOC_USER flag. This will
be used in the subsequent patch where the I915_BO_ALLOC_USER
flag will be applied to the framebuffer object. For a pre-allocated
framebuffer without the I915_BO_PREALLOC flag.

TTM would clear the content, which is not desirable.
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404143100.10452-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_object_types.h | 15 +++++++++++----
 drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c     |  5 +++--
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 19c9bdd8f905f..8540edff164b3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -328,6 +328,12 @@ struct drm_i915_gem_object {
  */
 #define I915_BO_ALLOC_GPU_ONLY	  BIT(6)
 #define I915_BO_ALLOC_CCS_AUX	  BIT(7)
+/*
+ * Object is allowed to retain its initial data and will not be cleared on first
+ * access if used along with I915_BO_ALLOC_USER. This is mainly to keep
+ * preallocated framebuffer data intact while transitioning it to i915drmfb.
+ */
+#define I915_BO_PREALLOC	  BIT(8)
 #define I915_BO_ALLOC_FLAGS (I915_BO_ALLOC_CONTIGUOUS | \
 			     I915_BO_ALLOC_VOLATILE | \
 			     I915_BO_ALLOC_CPU_CLEAR | \
@@ -335,10 +341,11 @@ struct drm_i915_gem_object {
 			     I915_BO_ALLOC_PM_VOLATILE | \
 			     I915_BO_ALLOC_PM_EARLY | \
 			     I915_BO_ALLOC_GPU_ONLY | \
-			     I915_BO_ALLOC_CCS_AUX)
-#define I915_BO_READONLY          BIT(8)
-#define I915_TILING_QUIRK_BIT     9 /* unknown swizzling; do not release! */
-#define I915_BO_PROTECTED         BIT(10)
+			     I915_BO_ALLOC_CCS_AUX | \
+			     I915_BO_PREALLOC)
+#define I915_BO_READONLY          BIT(9)
+#define I915_TILING_QUIRK_BIT     10 /* unknown swizzling; do not release! */
+#define I915_BO_PROTECTED         BIT(11)
 	/**
 	 * @mem_flags - Mutable placement-related flags
 	 *
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
index 6e3de0f33add9..46b398ccd2507 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
@@ -560,7 +560,7 @@ int i915_ttm_move(struct ttm_buffer_object *bo, bool evict,
 	struct dma_fence *migration_fence = NULL;
 	struct ttm_tt *ttm = bo->ttm;
 	struct i915_refct_sgt *dst_rsgt;
-	bool clear;
+	bool clear, prealloc_bo;
 	int ret;
 
 	if (GEM_WARN_ON(i915_ttm_is_ghost_object(bo))) {
@@ -590,7 +590,8 @@ int i915_ttm_move(struct ttm_buffer_object *bo, bool evict,
 		return PTR_ERR(dst_rsgt);
 
 	clear = !i915_ttm_cpu_maps_iomem(bo->resource) && (!ttm || !ttm_tt_is_populated(ttm));
-	if (!(clear && ttm && !(ttm->page_flags & TTM_TT_FLAG_ZERO_ALLOC))) {
+	prealloc_bo = obj->flags & I915_BO_PREALLOC;
+	if (!(clear && ttm && !((ttm->page_flags & TTM_TT_FLAG_ZERO_ALLOC) && !prealloc_bo))) {
 		struct i915_deps deps;
 
 		i915_deps_init(&deps, GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN);
-- 
GitLab


From ddb78a51fac65e8db2316ded59e27ab621aea856 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 16:30:57 +0200
Subject: [PATCH 0002/1791] drm/i915/display: Set I915_BO_ALLOC_USER for fb
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Framebuffer is exposed to userspace so make sure we set
proper flags for it. Set I915_BO_PREALLOC for prealloced
fb so that ttm won't clear existing data.

Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404143100.10452-2-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/display/intel_fbdev.c         | 3 ++-
 drivers/gpu/drm/i915/display/intel_plane_initial.c | 4 +++-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index 03ed4607a46d2..9ea49aaa33caa 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -163,7 +163,8 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 	obj = ERR_PTR(-ENODEV);
 	if (HAS_LMEM(dev_priv)) {
 		obj = i915_gem_object_create_lmem(dev_priv, size,
-						  I915_BO_ALLOC_CONTIGUOUS);
+						  I915_BO_ALLOC_CONTIGUOUS |
+						  I915_BO_ALLOC_USER);
 	} else {
 		/*
 		 * If the FB is too big, just don't use it since fbdev is not very
diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c
index 76be796df255d..3a95d24eab744 100644
--- a/drivers/gpu/drm/i915/display/intel_plane_initial.c
+++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c
@@ -110,7 +110,9 @@ initial_plane_vma(struct drm_i915_private *i915,
 	    size * 2 > i915->stolen_usable_size)
 		return NULL;
 
-	obj = i915_gem_object_create_region_at(mem, phys_base, size, 0);
+	obj = i915_gem_object_create_region_at(mem, phys_base, size,
+					       I915_BO_ALLOC_USER |
+					       I915_BO_PREALLOC);
 	if (IS_ERR(obj))
 		return NULL;
 
-- 
GitLab


From eaee1c08586395182e0004b3512a2f83570ea461 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 16:30:58 +0200
Subject: [PATCH 0003/1791] drm/i915: Add a function to mmap framebuffer obj
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement i915_gem_fb_mmap() to enable fb_ops.fb_mmap()
callback for i915's framebuffer objects.

v2: add a comment why i915_gem_object_get() needed(Andi).
v3: mmap also ttm objects.

Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404143100.10452-3-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_mman.c | 137 +++++++++++++++--------
 drivers/gpu/drm/i915/gem/i915_gem_mman.h |   2 +-
 2 files changed, 93 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index 4f69bff630682..d036ef2dcdba3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -927,53 +927,15 @@ static struct file *mmap_singleton(struct drm_i915_private *i915)
 	return file;
 }
 
-/*
- * This overcomes the limitation in drm_gem_mmap's assignment of a
- * drm_gem_object as the vma->vm_private_data. Since we need to
- * be able to resolve multiple mmap offsets which could be tied
- * to a single gem object.
- */
-int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma)
+static int
+i915_gem_object_mmap(struct drm_i915_gem_object *obj,
+		     struct i915_mmap_offset *mmo,
+		     struct vm_area_struct *vma)
 {
-	struct drm_vma_offset_node *node;
-	struct drm_file *priv = filp->private_data;
-	struct drm_device *dev = priv->minor->dev;
-	struct drm_i915_gem_object *obj = NULL;
-	struct i915_mmap_offset *mmo = NULL;
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+	struct drm_device *dev = &i915->drm;
 	struct file *anon;
 
-	if (drm_dev_is_unplugged(dev))
-		return -ENODEV;
-
-	rcu_read_lock();
-	drm_vma_offset_lock_lookup(dev->vma_offset_manager);
-	node = drm_vma_offset_exact_lookup_locked(dev->vma_offset_manager,
-						  vma->vm_pgoff,
-						  vma_pages(vma));
-	if (node && drm_vma_node_is_allowed(node, priv)) {
-		/*
-		 * Skip 0-refcnted objects as it is in the process of being
-		 * destroyed and will be invalid when the vma manager lock
-		 * is released.
-		 */
-		if (!node->driver_private) {
-			mmo = container_of(node, struct i915_mmap_offset, vma_node);
-			obj = i915_gem_object_get_rcu(mmo->obj);
-
-			GEM_BUG_ON(obj && obj->ops->mmap_ops);
-		} else {
-			obj = i915_gem_object_get_rcu
-				(container_of(node, struct drm_i915_gem_object,
-					      base.vma_node));
-
-			GEM_BUG_ON(obj && !obj->ops->mmap_ops);
-		}
-	}
-	drm_vma_offset_unlock_lookup(dev->vma_offset_manager);
-	rcu_read_unlock();
-	if (!obj)
-		return node ? -EACCES : -EINVAL;
-
 	if (i915_gem_object_is_readonly(obj)) {
 		if (vma->vm_flags & VM_WRITE) {
 			i915_gem_object_put(obj);
@@ -1005,7 +967,7 @@ int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 	if (obj->ops->mmap_ops) {
 		vma->vm_page_prot = pgprot_decrypted(vm_get_page_prot(vma->vm_flags));
 		vma->vm_ops = obj->ops->mmap_ops;
-		vma->vm_private_data = node->driver_private;
+		vma->vm_private_data = obj->base.vma_node.driver_private;
 		return 0;
 	}
 
@@ -1043,6 +1005,91 @@ int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 	return 0;
 }
 
+/*
+ * This overcomes the limitation in drm_gem_mmap's assignment of a
+ * drm_gem_object as the vma->vm_private_data. Since we need to
+ * be able to resolve multiple mmap offsets which could be tied
+ * to a single gem object.
+ */
+int i915_gem_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	struct drm_vma_offset_node *node;
+	struct drm_file *priv = filp->private_data;
+	struct drm_device *dev = priv->minor->dev;
+	struct drm_i915_gem_object *obj = NULL;
+	struct i915_mmap_offset *mmo = NULL;
+
+	if (drm_dev_is_unplugged(dev))
+		return -ENODEV;
+
+	rcu_read_lock();
+	drm_vma_offset_lock_lookup(dev->vma_offset_manager);
+	node = drm_vma_offset_exact_lookup_locked(dev->vma_offset_manager,
+						  vma->vm_pgoff,
+						  vma_pages(vma));
+	if (node && drm_vma_node_is_allowed(node, priv)) {
+		/*
+		 * Skip 0-refcnted objects as it is in the process of being
+		 * destroyed and will be invalid when the vma manager lock
+		 * is released.
+		 */
+		if (!node->driver_private) {
+			mmo = container_of(node, struct i915_mmap_offset, vma_node);
+			obj = i915_gem_object_get_rcu(mmo->obj);
+
+			GEM_BUG_ON(obj && obj->ops->mmap_ops);
+		} else {
+			obj = i915_gem_object_get_rcu
+				(container_of(node, struct drm_i915_gem_object,
+					      base.vma_node));
+
+			GEM_BUG_ON(obj && !obj->ops->mmap_ops);
+		}
+	}
+	drm_vma_offset_unlock_lookup(dev->vma_offset_manager);
+	rcu_read_unlock();
+	if (!obj)
+		return node ? -EACCES : -EINVAL;
+
+	return i915_gem_object_mmap(obj, mmo, vma);
+}
+
+int i915_gem_fb_mmap(struct drm_i915_gem_object *obj, struct vm_area_struct *vma)
+{
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+	struct drm_device *dev = &i915->drm;
+	struct i915_mmap_offset *mmo = NULL;
+	enum i915_mmap_type mmap_type;
+	struct i915_ggtt *ggtt = to_gt(i915)->ggtt;
+
+	if (drm_dev_is_unplugged(dev))
+		return -ENODEV;
+
+	/* handle ttm object */
+	if (obj->ops->mmap_ops) {
+		/*
+		 * ttm fault handler, ttm_bo_vm_fault_reserved() uses fake offset
+		 * to calculate page offset so set that up.
+		 */
+		vma->vm_pgoff += drm_vma_node_start(&obj->base.vma_node);
+	} else {
+		/* handle stolen and smem objects */
+		mmap_type = i915_ggtt_has_aperture(ggtt) ? I915_MMAP_TYPE_GTT : I915_MMAP_TYPE_WC;
+		mmo = mmap_offset_attach(obj, mmap_type, NULL);
+		if (!mmo)
+			return -ENODEV;
+	}
+
+	/*
+	 * When we install vm_ops for mmap we are too late for
+	 * the vm_ops->open() which increases the ref_count of
+	 * this obj and then it gets decreased by the vm_ops->close().
+	 * To balance this increase the obj ref_count here.
+	 */
+	obj = i915_gem_object_get(obj);
+	return i915_gem_object_mmap(obj, mmo, vma);
+}
+
 #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
 #include "selftests/i915_gem_mman.c"
 #endif
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.h b/drivers/gpu/drm/i915/gem/i915_gem_mman.h
index 1fa91b3033b35..196417fd0f5c4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.h
@@ -29,5 +29,5 @@ void i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj);
 
 void i915_gem_object_runtime_pm_release_mmap_offset(struct drm_i915_gem_object *obj);
 void i915_gem_object_release_mmap_offset(struct drm_i915_gem_object *obj);
-
+int i915_gem_fb_mmap(struct drm_i915_gem_object *obj, struct vm_area_struct *vma);
 #endif
-- 
GitLab


From 63b685efaa4d6b9db388857a2e6f5f5f11454f8d Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 16:30:59 +0200
Subject: [PATCH 0004/1791] drm/i915/display: Add helper func to get
 intel_fbdev from drm_fb_helper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a helper function to retrieve struct intel_fbdev from
struct drm_fb_helper.

Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404143100.10452-4-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/display/intel_fbdev.c | 23 ++++++++++------------
 1 file changed, 10 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index 9ea49aaa33caa..64971a8d5531e 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -67,6 +67,11 @@ struct intel_fbdev {
 	struct mutex hpd_lock;
 };
 
+static struct intel_fbdev *to_intel_fbdev(struct drm_fb_helper *fb_helper)
+{
+	return container_of(fb_helper, struct intel_fbdev, helper);
+}
+
 static struct intel_frontbuffer *to_frontbuffer(struct intel_fbdev *ifbdev)
 {
 	return ifbdev->fb->frontbuffer;
@@ -79,9 +84,7 @@ static void intel_fbdev_invalidate(struct intel_fbdev *ifbdev)
 
 static int intel_fbdev_set_par(struct fb_info *info)
 {
-	struct drm_fb_helper *fb_helper = info->par;
-	struct intel_fbdev *ifbdev =
-		container_of(fb_helper, struct intel_fbdev, helper);
+	struct intel_fbdev *ifbdev = to_intel_fbdev(info->par);
 	int ret;
 
 	ret = drm_fb_helper_set_par(info);
@@ -93,9 +96,7 @@ static int intel_fbdev_set_par(struct fb_info *info)
 
 static int intel_fbdev_blank(int blank, struct fb_info *info)
 {
-	struct drm_fb_helper *fb_helper = info->par;
-	struct intel_fbdev *ifbdev =
-		container_of(fb_helper, struct intel_fbdev, helper);
+	struct intel_fbdev *ifbdev = to_intel_fbdev(info->par);
 	int ret;
 
 	ret = drm_fb_helper_blank(blank, info);
@@ -108,9 +109,7 @@ static int intel_fbdev_blank(int blank, struct fb_info *info)
 static int intel_fbdev_pan_display(struct fb_var_screeninfo *var,
 				   struct fb_info *info)
 {
-	struct drm_fb_helper *fb_helper = info->par;
-	struct intel_fbdev *ifbdev =
-		container_of(fb_helper, struct intel_fbdev, helper);
+	struct intel_fbdev *ifbdev = to_intel_fbdev(info->par);
 	int ret;
 
 	ret = drm_fb_helper_pan_display(var, info);
@@ -136,8 +135,7 @@ static const struct fb_ops intelfb_ops = {
 static int intelfb_alloc(struct drm_fb_helper *helper,
 			 struct drm_fb_helper_surface_size *sizes)
 {
-	struct intel_fbdev *ifbdev =
-		container_of(helper, struct intel_fbdev, helper);
+	struct intel_fbdev *ifbdev = to_intel_fbdev(helper);
 	struct drm_framebuffer *fb;
 	struct drm_device *dev = helper->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -194,8 +192,7 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 static int intelfb_create(struct drm_fb_helper *helper,
 			  struct drm_fb_helper_surface_size *sizes)
 {
-	struct intel_fbdev *ifbdev =
-		container_of(helper, struct intel_fbdev, helper);
+	struct intel_fbdev *ifbdev = to_intel_fbdev(helper);
 	struct intel_framebuffer *intel_fb = ifbdev->fb;
 	struct drm_device *dev = helper->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
-- 
GitLab


From e24e6d695377ca70008ffc39695c3975b3e177b6 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 16:31:00 +0200
Subject: [PATCH 0005/1791] drm/i915/display: Implement fb_mmap callback
 function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If stolen memory allocation fails for fbdev, the driver will
fallback to system memory. Calculation of smem_start is wrong
for such framebuffer objs if the platform comes with no gmadr or
no aperture. Solve this by adding fb_mmap callback which will
use GTT if aperture is available otherwise will use cpu to access
the framebuffer.

v2: Use to_intel_fbdev() function(Jani)

Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404143100.10452-5-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/display/intel_fbdev.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index 64971a8d5531e..a44a5535db1ba 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -40,8 +40,10 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_gem_framebuffer_helper.h>
 
 #include "gem/i915_gem_lmem.h"
+#include "gem/i915_gem_mman.h"
 
 #include "i915_drv.h"
 #include "intel_display_types.h"
@@ -119,6 +121,15 @@ static int intel_fbdev_pan_display(struct fb_var_screeninfo *var,
 	return ret;
 }
 
+static int intel_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
+{
+	struct intel_fbdev *fbdev = to_intel_fbdev(info->par);
+	struct drm_gem_object *bo = drm_gem_fb_get_obj(&fbdev->fb->base, 0);
+	struct drm_i915_gem_object *obj = to_intel_bo(bo);
+
+	return i915_gem_fb_mmap(obj, vma);
+}
+
 static const struct fb_ops intelfb_ops = {
 	.owner = THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
@@ -130,6 +141,7 @@ static const struct fb_ops intelfb_ops = {
 	.fb_imageblit = drm_fb_helper_cfb_imageblit,
 	.fb_pan_display = intel_fbdev_pan_display,
 	.fb_blank = intel_fbdev_blank,
+	.fb_mmap = intel_fbdev_mmap,
 };
 
 static int intelfb_alloc(struct drm_fb_helper *helper,
-- 
GitLab


From 16fc9c08f0ec7b1c95f1ea4a16097acdb3fc943d Mon Sep 17 00:00:00 2001
From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Date: Fri, 7 Apr 2023 12:32:37 +0300
Subject: [PATCH 0006/1791] drm/i915: disable sampler indirect state in
 bindless heap

By default the indirect state sampler data (border colors) are stored
in the same heap as the SAMPLER_STATE structure. For userspace drivers
that can be 2 different heaps (dynamic state heap & bindless sampler
state heap). This means that border colors have to copied in 2
different places so that the same SAMPLER_STATE structure find the
right data.

This change is forcing the indirect state sampler data to only be in
the dynamic state pool (more convenient for userspace drivers, they
only have to have one copy of the border colors). This is reproducing
the behavior of the Windows drivers.

BSpec: 46052

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: stable@vger.kernel.org
Reviewed-by: Haridhar Kalvala <haridhar.kalvala@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230407093237.3296286-1-lionel.g.landwerlin@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_regs.h     |  1 +
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 19 +++++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index 6115cc1805de6..6903c66a4ed72 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -1147,6 +1147,7 @@
 #define   SC_DISABLE_POWER_OPTIMIZATION_EBB	REG_BIT(9)
 #define   GEN11_SAMPLER_ENABLE_HEADLESS_MSG	REG_BIT(5)
 #define   MTL_DISABLE_SAMPLER_SC_OOO		REG_BIT(3)
+#define   GEN11_INDIRECT_STATE_BASE_ADDR_OVERRIDE	REG_BIT(0)
 
 #define GEN9_HALF_SLICE_CHICKEN7		MCR_REG(0xe194)
 #define   DG2_DISABLE_ROUND_ENABLE_ALLOW_FOR_SSLA	REG_BIT(15)
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index c0de5276b3d32..4e3cbeca9eece 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -3051,6 +3051,25 @@ general_render_compute_wa_init(struct intel_engine_cs *engine, struct i915_wa_li
 
 	add_render_compute_tuning_settings(i915, wal);
 
+	if (GRAPHICS_VER(i915) >= 11) {
+		/* This is not a Wa (although referred to as
+		 * WaSetInidrectStateOverride in places), this allows
+		 * applications that reference sampler states through
+		 * the BindlessSamplerStateBaseAddress to have their
+		 * border color relative to DynamicStateBaseAddress
+		 * rather than BindlessSamplerStateBaseAddress.
+		 *
+		 * Otherwise SAMPLER_STATE border colors have to be
+		 * copied in multiple heaps (DynamicStateBaseAddress &
+		 * BindlessSamplerStateBaseAddress)
+		 *
+		 * BSpec: 46052
+		 */
+		wa_mcr_masked_en(wal,
+				 GEN10_SAMPLER_MODE,
+				 GEN11_INDIRECT_STATE_BASE_ADDR_OVERRIDE);
+	}
+
 	if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_B0, STEP_FOREVER) ||
 	    IS_MTL_GRAPHICS_STEP(i915, P, STEP_B0, STEP_FOREVER))
 		/* Wa_14017856879 */
-- 
GitLab


From 404c3acda4b65924c05bc63242e94f954f84c165 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:51 +0300
Subject: [PATCH 0007/1791] drm/i915: Fix limited range csc matrix
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Our current limited range matrix is a bit off. I think it
was originally calculated with rounding, as if we wanted
the normal pixel replication type of behaviour.
That is, since the 8bpc max value is 0xeb we assumed the
16bpc max value should be 0xebeb, but what the HDMI spec
actually says it should be is 0xeb00.

So to get what we want we make the formula
 out = in * (235-16) << (12-8) / in_max + 16 << (12-8),
with 12 being precision of the csc, 8 being the precision
of the constants we used.

The hardware takes its coefficients as floating point
values, but the (235−16)/255 = ~.86, so exponent 0
is what we want anyway, so it works out perfectly without
having to hardcode it in hex or start playing with floats.

In terms of raw numbers we are feeding the hardware the
post offset changes from 0x101 to 0x100, and the coefficient
changes from 0xdc0 to 0xdb0 (~.860->~.855). So this should
make everything come out just a tad darker.

I already used better constants in lut_limited_range() earlier
so the output of the two paths should be closer now.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-2-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 36aac88143ac1..3c3e2f5a5cdec 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -116,10 +116,9 @@ struct intel_color_funcs {
 #define ILK_CSC_COEFF_FP(coeff, fbits)	\
 	(clamp_val(((coeff) >> (32 - (fbits) - 3)) + 4, 0, 0xfff) & 0xff8)
 
-#define ILK_CSC_COEFF_LIMITED_RANGE 0x0dc0
 #define ILK_CSC_COEFF_1_0 0x7800
-
-#define ILK_CSC_POSTOFF_LIMITED_RANGE (16 * (1 << 12) / 255)
+#define ILK_CSC_COEFF_LIMITED_RANGE ((235 - 16) << (12 - 8)) /* exponent 0 */
+#define ILK_CSC_POSTOFF_LIMITED_RANGE (16 << (12 - 8))
 
 /* Nop pre/post offsets */
 static const u16 ilk_csc_off_zero[3] = {};
-- 
GitLab


From 57b5482bff9e4f60069a8c0de91bb397612ce059 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:52 +0300
Subject: [PATCH 0008/1791] drm/i915: Introduce intel_csc_matrix struct
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce a structure that can hold our CSC matrices. In there
we shall have the preoffsets, postoffsets, and coefficients,
all in platform specific format (at least for now).

We shall start by converting the ilk+ code to make use of
the new structure. chv will come later.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-3-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c    | 188 +++++++++---------
 .../drm/i915/display/intel_display_types.h    |   6 +
 2 files changed, 97 insertions(+), 97 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 3c3e2f5a5cdec..b1059e0c06651 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -120,40 +120,42 @@ struct intel_color_funcs {
 #define ILK_CSC_COEFF_LIMITED_RANGE ((235 - 16) << (12 - 8)) /* exponent 0 */
 #define ILK_CSC_POSTOFF_LIMITED_RANGE (16 << (12 - 8))
 
-/* Nop pre/post offsets */
-static const u16 ilk_csc_off_zero[3] = {};
-
-/* Identity matrix */
-static const u16 ilk_csc_coeff_identity[9] = {
-	ILK_CSC_COEFF_1_0, 0, 0,
-	0, ILK_CSC_COEFF_1_0, 0,
-	0, 0, ILK_CSC_COEFF_1_0,
-};
-
-/* Limited range RGB post offsets */
-static const u16 ilk_csc_postoff_limited_range[3] = {
-	ILK_CSC_POSTOFF_LIMITED_RANGE,
-	ILK_CSC_POSTOFF_LIMITED_RANGE,
-	ILK_CSC_POSTOFF_LIMITED_RANGE,
+static const struct intel_csc_matrix ilk_csc_matrix_identity = {
+	.preoff = {},
+	.coeff = {
+		ILK_CSC_COEFF_1_0, 0, 0,
+		0, ILK_CSC_COEFF_1_0, 0,
+		0, 0, ILK_CSC_COEFF_1_0,
+	},
+	.postoff = {},
 };
 
 /* Full range RGB -> limited range RGB matrix */
-static const u16 ilk_csc_coeff_limited_range[9] = {
-	ILK_CSC_COEFF_LIMITED_RANGE, 0, 0,
-	0, ILK_CSC_COEFF_LIMITED_RANGE, 0,
-	0, 0, ILK_CSC_COEFF_LIMITED_RANGE,
+static const struct intel_csc_matrix ilk_csc_matrix_limited_range = {
+	.preoff = {},
+	.coeff = {
+		ILK_CSC_COEFF_LIMITED_RANGE, 0, 0,
+		0, ILK_CSC_COEFF_LIMITED_RANGE, 0,
+		0, 0, ILK_CSC_COEFF_LIMITED_RANGE,
+	},
+	.postoff = {
+		ILK_CSC_POSTOFF_LIMITED_RANGE,
+		ILK_CSC_POSTOFF_LIMITED_RANGE,
+		ILK_CSC_POSTOFF_LIMITED_RANGE,
+	},
 };
 
 /* BT.709 full range RGB -> limited range YCbCr matrix */
-static const u16 ilk_csc_coeff_rgb_to_ycbcr[9] = {
-	0x1e08, 0x9cc0, 0xb528,
-	0x2ba8, 0x09d8, 0x37e8,
-	0xbce8, 0x9ad8, 0x1e08,
-};
-
-/* Limited range YCbCr post offsets */
-static const u16 ilk_csc_postoff_rgb_to_ycbcr[3] = {
-	0x0800, 0x0100, 0x0800,
+static const struct intel_csc_matrix ilk_csc_matrix_rgb_to_ycbcr = {
+	.preoff = {},
+	.coeff = {
+		0x1e08, 0x9cc0, 0xb528,
+		0x2ba8, 0x09d8, 0x37e8,
+		0xbce8, 0x9ad8, 0x1e08,
+	},
+	.postoff = {
+		0x0800, 0x0100, 0x0800,
+	},
 };
 
 static bool lut_is_legacy(const struct drm_property_blob *lut)
@@ -188,69 +190,66 @@ static u64 *ctm_mult_by_limited(u64 *result, const u64 *input)
 }
 
 static void ilk_update_pipe_csc(struct intel_crtc *crtc,
-				const u16 preoff[3],
-				const u16 coeff[9],
-				const u16 postoff[3])
+				const struct intel_csc_matrix *csc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	intel_de_write_fw(i915, PIPE_CSC_PREOFF_HI(pipe), preoff[0]);
-	intel_de_write_fw(i915, PIPE_CSC_PREOFF_ME(pipe), preoff[1]);
-	intel_de_write_fw(i915, PIPE_CSC_PREOFF_LO(pipe), preoff[2]);
+	intel_de_write_fw(i915, PIPE_CSC_PREOFF_HI(pipe), csc->preoff[0]);
+	intel_de_write_fw(i915, PIPE_CSC_PREOFF_ME(pipe), csc->preoff[1]);
+	intel_de_write_fw(i915, PIPE_CSC_PREOFF_LO(pipe), csc->preoff[2]);
 
 	intel_de_write_fw(i915, PIPE_CSC_COEFF_RY_GY(pipe),
-			  coeff[0] << 16 | coeff[1]);
-	intel_de_write_fw(i915, PIPE_CSC_COEFF_BY(pipe), coeff[2] << 16);
+			  csc->coeff[0] << 16 | csc->coeff[1]);
+	intel_de_write_fw(i915, PIPE_CSC_COEFF_BY(pipe),
+			  csc->coeff[2] << 16);
 
 	intel_de_write_fw(i915, PIPE_CSC_COEFF_RU_GU(pipe),
-			  coeff[3] << 16 | coeff[4]);
-	intel_de_write_fw(i915, PIPE_CSC_COEFF_BU(pipe), coeff[5] << 16);
+			  csc->coeff[3] << 16 | csc->coeff[4]);
+	intel_de_write_fw(i915, PIPE_CSC_COEFF_BU(pipe),
+			  csc->coeff[5] << 16);
 
 	intel_de_write_fw(i915, PIPE_CSC_COEFF_RV_GV(pipe),
-			  coeff[6] << 16 | coeff[7]);
-	intel_de_write_fw(i915, PIPE_CSC_COEFF_BV(pipe), coeff[8] << 16);
+			  csc->coeff[6] << 16 | csc->coeff[7]);
+	intel_de_write_fw(i915, PIPE_CSC_COEFF_BV(pipe),
+			  csc->coeff[8] << 16);
 
-	if (DISPLAY_VER(i915) >= 7) {
-		intel_de_write_fw(i915, PIPE_CSC_POSTOFF_HI(pipe),
-				  postoff[0]);
-		intel_de_write_fw(i915, PIPE_CSC_POSTOFF_ME(pipe),
-				  postoff[1]);
-		intel_de_write_fw(i915, PIPE_CSC_POSTOFF_LO(pipe),
-				  postoff[2]);
-	}
+	if (DISPLAY_VER(i915) < 7)
+		return;
+
+	intel_de_write_fw(i915, PIPE_CSC_POSTOFF_HI(pipe), csc->postoff[0]);
+	intel_de_write_fw(i915, PIPE_CSC_POSTOFF_ME(pipe), csc->postoff[1]);
+	intel_de_write_fw(i915, PIPE_CSC_POSTOFF_LO(pipe), csc->postoff[2]);
 }
 
 static void icl_update_output_csc(struct intel_crtc *crtc,
-				  const u16 preoff[3],
-				  const u16 coeff[9],
-				  const u16 postoff[3])
+				  const struct intel_csc_matrix *csc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe), preoff[0]);
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe), preoff[1]);
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe), preoff[2]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe), csc->preoff[0]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe), csc->preoff[1]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe), csc->preoff[2]);
 
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe),
-			  coeff[0] << 16 | coeff[1]);
+			  csc->coeff[0] << 16 | csc->coeff[1]);
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BY(pipe),
-			  coeff[2] << 16);
+			  csc->coeff[2] << 16);
 
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe),
-			  coeff[3] << 16 | coeff[4]);
+			  csc->coeff[3] << 16 | csc->coeff[4]);
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BU(pipe),
-			  coeff[5] << 16);
+			  csc->coeff[5] << 16);
 
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe),
-			  coeff[6] << 16 | coeff[7]);
+			  csc->coeff[6] << 16 | csc->coeff[7]);
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BV(pipe),
-			  coeff[8] << 16);
+			  csc->coeff[8] << 16);
 
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), postoff[0]);
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), postoff[1]);
-	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), postoff[2]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), csc->postoff[0]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), csc->postoff[1]);
+	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), csc->postoff[2]);
 }
 
 static bool ilk_limited_range(const struct intel_crtc_state *crtc_state)
@@ -294,13 +293,20 @@ static bool ilk_csc_limited_range(const struct intel_crtc_state *crtc_state)
 }
 
 static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
-				u16 coeffs[9], bool limited_color_range)
+				struct intel_csc_matrix *csc,
+				bool limited_color_range)
 {
 	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
 	const u64 *input;
 	u64 temp[9];
 	int i;
 
+	/* for preoff/postoff */
+	if (limited_color_range)
+		*csc = ilk_csc_matrix_limited_range;
+	else
+		*csc = ilk_csc_matrix_identity;
+
 	if (limited_color_range)
 		input = ctm_mult_by_limited(temp, ctm->matrix);
 	else
@@ -319,28 +325,28 @@ static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 		 */
 		abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_4_0 - 1);
 
-		coeffs[i] = 0;
+		csc->coeff[i] = 0;
 
 		/* sign bit */
 		if (CTM_COEFF_NEGATIVE(input[i]))
-			coeffs[i] |= 1 << 15;
+			csc->coeff[i] |= 1 << 15;
 
 		if (abs_coeff < CTM_COEFF_0_125)
-			coeffs[i] |= (3 << 12) |
+			csc->coeff[i] |= (3 << 12) |
 				ILK_CSC_COEFF_FP(abs_coeff, 12);
 		else if (abs_coeff < CTM_COEFF_0_25)
-			coeffs[i] |= (2 << 12) |
+			csc->coeff[i] |= (2 << 12) |
 				ILK_CSC_COEFF_FP(abs_coeff, 11);
 		else if (abs_coeff < CTM_COEFF_0_5)
-			coeffs[i] |= (1 << 12) |
+			csc->coeff[i] |= (1 << 12) |
 				ILK_CSC_COEFF_FP(abs_coeff, 10);
 		else if (abs_coeff < CTM_COEFF_1_0)
-			coeffs[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9);
+			csc->coeff[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9);
 		else if (abs_coeff < CTM_COEFF_2_0)
-			coeffs[i] |= (7 << 12) |
+			csc->coeff[i] |= (7 << 12) |
 				ILK_CSC_COEFF_FP(abs_coeff, 8);
 		else
-			coeffs[i] |= (6 << 12) |
+			csc->coeff[i] |= (6 << 12) |
 				ILK_CSC_COEFF_FP(abs_coeff, 7);
 	}
 }
@@ -352,21 +358,15 @@ static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 	bool limited_color_range = ilk_csc_limited_range(crtc_state);
 
 	if (crtc_state->hw.ctm) {
-		u16 coeff[9];
+		struct intel_csc_matrix tmp;
+
+		ilk_csc_convert_ctm(crtc_state, &tmp, limited_color_range);
 
-		ilk_csc_convert_ctm(crtc_state, coeff, limited_color_range);
-		ilk_update_pipe_csc(crtc, ilk_csc_off_zero, coeff,
-				    limited_color_range ?
-				    ilk_csc_postoff_limited_range :
-				    ilk_csc_off_zero);
+		ilk_update_pipe_csc(crtc, &tmp);
 	} else if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
-		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
-				    ilk_csc_coeff_rgb_to_ycbcr,
-				    ilk_csc_postoff_rgb_to_ycbcr);
+		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_rgb_to_ycbcr);
 	} else if (limited_color_range) {
-		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
-				    ilk_csc_coeff_limited_range,
-				    ilk_csc_postoff_limited_range);
+		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_limited_range);
 	} else if (crtc_state->csc_enable) {
 		/*
 		 * On GLK both pipe CSC and degamma LUT are controlled
@@ -376,9 +376,7 @@ static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 		 */
 		drm_WARN_ON(&i915->drm, !IS_GEMINILAKE(i915));
 
-		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
-				    ilk_csc_coeff_identity,
-				    ilk_csc_off_zero);
+		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_identity);
 	}
 }
 
@@ -387,21 +385,17 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 
 	if (crtc_state->hw.ctm) {
-		u16 coeff[9];
+		struct intel_csc_matrix tmp;
+
+		ilk_csc_convert_ctm(crtc_state, &tmp, false);
 
-		ilk_csc_convert_ctm(crtc_state, coeff, false);
-		ilk_update_pipe_csc(crtc, ilk_csc_off_zero,
-				    coeff, ilk_csc_off_zero);
+		ilk_update_pipe_csc(crtc, &tmp);
 	}
 
 	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
-		icl_update_output_csc(crtc, ilk_csc_off_zero,
-				      ilk_csc_coeff_rgb_to_ycbcr,
-				      ilk_csc_postoff_rgb_to_ycbcr);
+		icl_update_output_csc(crtc, &ilk_csc_matrix_rgb_to_ycbcr);
 	} else if (crtc_state->limited_color_range) {
-		icl_update_output_csc(crtc, ilk_csc_off_zero,
-				      ilk_csc_coeff_limited_range,
-				      ilk_csc_postoff_limited_range);
+		icl_update_output_csc(crtc, &ilk_csc_matrix_limited_range);
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 47395b39c8f49..c0ed71dafd5e1 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -980,6 +980,12 @@ struct intel_link_m_n {
 	u32 link_n;
 };
 
+struct intel_csc_matrix {
+	u16 coeff[9];
+	u16 preoff[3];
+	u16 postoff[3];
+};
+
 struct intel_crtc_state {
 	/*
 	 * uapi (drm) state. This is the software state shown to userspace.
-- 
GitLab


From ec280042094c3f700d7321e7163591c6eac72274 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:53 +0300
Subject: [PATCH 0009/1791] drm/i915: Split chv_load_cgm_csc() into pieces
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Split chv_cgm_csc_convert_ctm() out from chv_load_cgm_csc() so
that we have functions with clear jobs. This is also how the ilk+
code is already structured.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-4-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index b1059e0c06651..47af24e64a7ea 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -399,16 +399,13 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 	}
 }
 
-static void chv_load_cgm_csc(struct intel_crtc *crtc,
-			     const struct drm_property_blob *blob)
+static void chv_cgm_csc_convert_ctm(u16 coeffs[9],
+				    const struct drm_property_blob *blob)
 {
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	const struct drm_color_ctm *ctm = blob->data;
-	enum pipe pipe = crtc->pipe;
-	u16 coeffs[9];
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(coeffs); i++) {
+	for (i = 0; i < 9; i++) {
 		u64 abs_coeff = ((1ULL << 63) - 1) & ctm->matrix[i];
 
 		/* Round coefficient. */
@@ -425,6 +422,16 @@ static void chv_load_cgm_csc(struct intel_crtc *crtc,
 		coeffs[i] |= ((abs_coeff >> 32) & 7) << 12;
 		coeffs[i] |= (abs_coeff >> 20) & 0xfff;
 	}
+}
+
+static void chv_load_cgm_csc(struct intel_crtc *crtc,
+			     const struct drm_property_blob *blob)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u16 coeffs[9];
+
+	chv_cgm_csc_convert_ctm(coeffs, blob);
 
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF01(pipe),
 			  coeffs[1] << 16 | coeffs[0]);
-- 
GitLab


From 1dcd7aac31842028beb5fa2531fd54ce0d588b17 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:54 +0300
Subject: [PATCH 0010/1791] drm/i915: Start using struct intel_csc_matrix for
 chv cgm csc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Convert chv_cgm_csc_convert_ctm() over to using the new
intel_csc_matrix structure. No pre/post offsets on this
hardware so only the coefficients get filled out.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-5-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 24 +++++++++++-----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 47af24e64a7ea..a76cea4ab1ec9 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -399,7 +399,7 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 	}
 }
 
-static void chv_cgm_csc_convert_ctm(u16 coeffs[9],
+static void chv_cgm_csc_convert_ctm(struct intel_csc_matrix *csc,
 				    const struct drm_property_blob *blob)
 {
 	const struct drm_color_ctm *ctm = blob->data;
@@ -413,14 +413,14 @@ static void chv_cgm_csc_convert_ctm(u16 coeffs[9],
 		/* Clamp to hardware limits. */
 		abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1);
 
-		coeffs[i] = 0;
+		csc->coeff[i] = 0;
 
 		/* Write coefficients in S3.12 format. */
 		if (ctm->matrix[i] & (1ULL << 63))
-			coeffs[i] |= 1 << 15;
+			csc->coeff[i] |= 1 << 15;
 
-		coeffs[i] |= ((abs_coeff >> 32) & 7) << 12;
-		coeffs[i] |= (abs_coeff >> 20) & 0xfff;
+		csc->coeff[i] |= ((abs_coeff >> 32) & 7) << 12;
+		csc->coeff[i] |= (abs_coeff >> 20) & 0xfff;
 	}
 }
 
@@ -429,20 +429,20 @@ static void chv_load_cgm_csc(struct intel_crtc *crtc,
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
-	u16 coeffs[9];
+	struct intel_csc_matrix tmp;
 
-	chv_cgm_csc_convert_ctm(coeffs, blob);
+	chv_cgm_csc_convert_ctm(&tmp, blob);
 
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF01(pipe),
-			  coeffs[1] << 16 | coeffs[0]);
+			  tmp.coeff[1] << 16 | tmp.coeff[0]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF23(pipe),
-			  coeffs[3] << 16 | coeffs[2]);
+			  tmp.coeff[3] << 16 | tmp.coeff[2]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF45(pipe),
-			  coeffs[5] << 16 | coeffs[4]);
+			  tmp.coeff[5] << 16 | tmp.coeff[4]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF67(pipe),
-			  coeffs[7] << 16 | coeffs[6]);
+			  tmp.coeff[7] << 16 | tmp.coeff[6]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF8(pipe),
-			  coeffs[8]);
+			  tmp.coeff[8]);
 }
 
 /* convert hw value with given bit_precision to lut property val */
-- 
GitLab


From 68f5f78d0fe08f277a3aea9ad28508a7f243de6a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:55 +0300
Subject: [PATCH 0011/1791] drm/i915: Store ilk+ csc matrices in the crtc state
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Embed a pair of intel_csc_matrix structs in the crtc state,
and fill them out appropriately during atomic_check().

Since pre-ivb platforms don't have programmable post offsets
we shall leave those zeroed, mainly in preparation for
state readout+check.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-6-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c    | 86 +++++++++++++------
 .../drm/i915/display/intel_display_types.h    |  2 +
 2 files changed, 64 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index a76cea4ab1ec9..67e8b8f583f18 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -158,6 +158,11 @@ static const struct intel_csc_matrix ilk_csc_matrix_rgb_to_ycbcr = {
 	},
 };
 
+static void intel_csc_clear(struct intel_csc_matrix *csc)
+{
+	memset(csc, 0, sizeof(*csc));
+}
+
 static bool lut_is_legacy(const struct drm_property_blob *lut)
 {
 	return lut && drm_color_lut_size(lut) == LEGACY_LUT_LENGTH;
@@ -292,10 +297,21 @@ static bool ilk_csc_limited_range(const struct intel_crtc_state *crtc_state)
 	return !ilk_lut_limited_range(crtc_state);
 }
 
+static void ilk_csc_copy(struct drm_i915_private *i915,
+			 struct intel_csc_matrix *dst,
+			 const struct intel_csc_matrix *src)
+{
+	*dst = *src;
+
+	if (DISPLAY_VER(i915) < 7)
+		memset(dst->postoff, 0, sizeof(dst->postoff));
+}
+
 static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 				struct intel_csc_matrix *csc,
 				bool limited_color_range)
 {
+	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
 	const u64 *input;
 	u64 temp[9];
@@ -303,9 +319,9 @@ static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 
 	/* for preoff/postoff */
 	if (limited_color_range)
-		*csc = ilk_csc_matrix_limited_range;
+		ilk_csc_copy(i915, csc, &ilk_csc_matrix_limited_range);
 	else
-		*csc = ilk_csc_matrix_identity;
+		ilk_csc_copy(i915, csc, &ilk_csc_matrix_identity);
 
 	if (limited_color_range)
 		input = ctm_mult_by_limited(temp, ctm->matrix);
@@ -351,22 +367,17 @@ static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 	}
 }
 
-static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+static void ilk_assign_csc(struct intel_crtc_state *crtc_state)
 {
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 	bool limited_color_range = ilk_csc_limited_range(crtc_state);
 
 	if (crtc_state->hw.ctm) {
-		struct intel_csc_matrix tmp;
-
-		ilk_csc_convert_ctm(crtc_state, &tmp, limited_color_range);
-
-		ilk_update_pipe_csc(crtc, &tmp);
+		ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, limited_color_range);
 	} else if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
-		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_rgb_to_ycbcr);
+		ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_rgb_to_ycbcr);
 	} else if (limited_color_range) {
-		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_limited_range);
+		ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_limited_range);
 	} else if (crtc_state->csc_enable) {
 		/*
 		 * On GLK both pipe CSC and degamma LUT are controlled
@@ -376,27 +387,46 @@ static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 		 */
 		drm_WARN_ON(&i915->drm, !IS_GEMINILAKE(i915));
 
-		ilk_update_pipe_csc(crtc, &ilk_csc_matrix_identity);
+		ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_identity);
+	} else {
+		intel_csc_clear(&crtc_state->csc);
 	}
 }
 
-static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 
-	if (crtc_state->hw.ctm) {
-		struct intel_csc_matrix tmp;
+	if (crtc_state->csc_enable)
+		ilk_update_pipe_csc(crtc, &crtc_state->csc);
+}
 
-		ilk_csc_convert_ctm(crtc_state, &tmp, false);
+static void icl_assign_csc(struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
-		ilk_update_pipe_csc(crtc, &tmp);
-	}
+	if (crtc_state->hw.ctm)
+		ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, false);
+	else
+		intel_csc_clear(&crtc_state->csc);
 
-	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
-		icl_update_output_csc(crtc, &ilk_csc_matrix_rgb_to_ycbcr);
-	} else if (crtc_state->limited_color_range) {
-		icl_update_output_csc(crtc, &ilk_csc_matrix_limited_range);
-	}
+	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB)
+		ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_rgb_to_ycbcr);
+	else if (crtc_state->limited_color_range)
+		ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_limited_range);
+	else
+		intel_csc_clear(&crtc_state->output_csc);
+}
+
+static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	if (crtc_state->csc_mode & ICL_CSC_ENABLE)
+		ilk_update_pipe_csc(crtc, &crtc_state->csc);
+
+	if (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE)
+		icl_update_output_csc(crtc, &crtc_state->output_csc);
 }
 
 static void chv_cgm_csc_convert_ctm(struct intel_csc_matrix *csc,
@@ -1962,6 +1992,8 @@ static int ilk_color_check(struct intel_crtc_state *crtc_state)
 	if (ret)
 		return ret;
 
+	ilk_assign_csc(crtc_state);
+
 	crtc_state->preload_luts = intel_can_preload_luts(crtc_state);
 
 	return 0;
@@ -2068,6 +2100,8 @@ static int ivb_color_check(struct intel_crtc_state *crtc_state)
 	if (ret)
 		return ret;
 
+	ilk_assign_csc(crtc_state);
+
 	crtc_state->preload_luts = intel_can_preload_luts(crtc_state);
 
 	return 0;
@@ -2199,6 +2233,8 @@ static int glk_color_check(struct intel_crtc_state *crtc_state)
 	if (ret)
 		return ret;
 
+	ilk_assign_csc(crtc_state);
+
 	crtc_state->preload_luts = intel_can_preload_luts(crtc_state);
 
 	return 0;
@@ -2261,6 +2297,8 @@ static int icl_color_check(struct intel_crtc_state *crtc_state)
 
 	intel_assign_luts(crtc_state);
 
+	icl_assign_csc(crtc_state);
+
 	crtc_state->preload_luts = intel_can_preload_luts(crtc_state);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index c0ed71dafd5e1..ed02399ee41dc 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1027,6 +1027,8 @@ struct intel_crtc_state {
 	/* actual state of LUTs */
 	struct drm_property_blob *pre_csc_lut, *post_csc_lut;
 
+	struct intel_csc_matrix csc, output_csc;
+
 	/**
 	 * quirks - bitfield with hw state readout quirks
 	 *
-- 
GitLab


From e0980b8d82d1306251bcd67b693e008bb6c89512 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:56 +0300
Subject: [PATCH 0012/1791] drm/i915: Utilize crtc_state->csc on chv
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Store the chv cgm csc matrix in the crtc state as well. We
shall store it in the same place where we store the ilk+
pipe csc matrix (as opposed to the output csc matrix).

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-7-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 34 +++++++++++++---------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 67e8b8f583f18..54c8fa96efba9 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -429,10 +429,10 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 		icl_update_output_csc(crtc, &crtc_state->output_csc);
 }
 
-static void chv_cgm_csc_convert_ctm(struct intel_csc_matrix *csc,
-				    const struct drm_property_blob *blob)
+static void chv_cgm_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
+				    struct intel_csc_matrix *csc)
 {
-	const struct drm_color_ctm *ctm = blob->data;
+	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
 	int i;
 
 	for (i = 0; i < 9; i++) {
@@ -455,24 +455,29 @@ static void chv_cgm_csc_convert_ctm(struct intel_csc_matrix *csc,
 }
 
 static void chv_load_cgm_csc(struct intel_crtc *crtc,
-			     const struct drm_property_blob *blob)
+			     const struct intel_csc_matrix *csc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
-	struct intel_csc_matrix tmp;
-
-	chv_cgm_csc_convert_ctm(&tmp, blob);
 
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF01(pipe),
-			  tmp.coeff[1] << 16 | tmp.coeff[0]);
+			  csc->coeff[1] << 16 | csc->coeff[0]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF23(pipe),
-			  tmp.coeff[3] << 16 | tmp.coeff[2]);
+			  csc->coeff[3] << 16 | csc->coeff[2]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF45(pipe),
-			  tmp.coeff[5] << 16 | tmp.coeff[4]);
+			  csc->coeff[5] << 16 | csc->coeff[4]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF67(pipe),
-			  tmp.coeff[7] << 16 | tmp.coeff[6]);
+			  csc->coeff[7] << 16 | csc->coeff[6]);
 	intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF8(pipe),
-			  tmp.coeff[8]);
+			  csc->coeff[8]);
+}
+
+static void chv_assign_csc(struct intel_crtc_state *crtc_state)
+{
+	if (crtc_state->hw.ctm)
+		chv_cgm_csc_convert_ctm(crtc_state, &crtc_state->csc);
+	else
+		intel_csc_clear(&crtc_state->csc);
 }
 
 /* convert hw value with given bit_precision to lut property val */
@@ -1440,10 +1445,9 @@ static void chv_load_luts(const struct intel_crtc_state *crtc_state)
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	const struct drm_property_blob *pre_csc_lut = crtc_state->pre_csc_lut;
 	const struct drm_property_blob *post_csc_lut = crtc_state->post_csc_lut;
-	const struct drm_property_blob *ctm = crtc_state->hw.ctm;
 
 	if (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC)
-		chv_load_cgm_csc(crtc, ctm);
+		chv_load_cgm_csc(crtc, &crtc_state->csc);
 
 	if (crtc_state->cgm_mode & CGM_PIPE_MODE_DEGAMMA)
 		chv_load_cgm_degamma(crtc, pre_csc_lut);
@@ -1870,6 +1874,8 @@ static int chv_color_check(struct intel_crtc_state *crtc_state)
 
 	intel_assign_luts(crtc_state);
 
+	chv_assign_csc(crtc_state);
+
 	crtc_state->preload_luts = chv_can_preload_luts(crtc_state);
 
 	return 0;
-- 
GitLab


From e006df050606785aa7d04cd47f913d9c6ce4669d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 08:51:44 +0300
Subject: [PATCH 0013/1791] drm/i915: Sprinke a few sanity check WARNS during
 csc assignment
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make sure the csc enable bit(s) match the way we're about to
fill the csc matrices.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-8-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 39 ++++++++++++++++++----
 1 file changed, 32 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 54c8fa96efba9..2988c91d8ff6b 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -373,10 +373,16 @@ static void ilk_assign_csc(struct intel_crtc_state *crtc_state)
 	bool limited_color_range = ilk_csc_limited_range(crtc_state);
 
 	if (crtc_state->hw.ctm) {
+		drm_WARN_ON(&i915->drm, !crtc_state->csc_enable);
+
 		ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, limited_color_range);
 	} else if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
+		drm_WARN_ON(&i915->drm, !crtc_state->csc_enable);
+
 		ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_rgb_to_ycbcr);
 	} else if (limited_color_range) {
+		drm_WARN_ON(&i915->drm, !crtc_state->csc_enable);
+
 		ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_limited_range);
 	} else if (crtc_state->csc_enable) {
 		/*
@@ -405,17 +411,29 @@ static void icl_assign_csc(struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
-	if (crtc_state->hw.ctm)
+	if (crtc_state->hw.ctm) {
+		drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_CSC_ENABLE) == 0);
+
 		ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, false);
-	else
+	} else {
+		drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_CSC_ENABLE) != 0);
+
 		intel_csc_clear(&crtc_state->csc);
+	}
+
+	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) {
+		drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) == 0);
 
-	if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB)
 		ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_rgb_to_ycbcr);
-	else if (crtc_state->limited_color_range)
+	} else if (crtc_state->limited_color_range) {
+		drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) == 0);
+
 		ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_limited_range);
-	else
+	} else {
+		drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) != 0);
+
 		intel_csc_clear(&crtc_state->output_csc);
+	}
 }
 
 static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
@@ -474,10 +492,17 @@ static void chv_load_cgm_csc(struct intel_crtc *crtc,
 
 static void chv_assign_csc(struct intel_crtc_state *crtc_state)
 {
-	if (crtc_state->hw.ctm)
+	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+
+	if (crtc_state->hw.ctm) {
+		drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) == 0);
+
 		chv_cgm_csc_convert_ctm(crtc_state, &crtc_state->csc);
-	else
+	} else {
+		drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) != 0);
+
 		intel_csc_clear(&crtc_state->csc);
+	}
 }
 
 /* convert hw value with given bit_precision to lut property val */
-- 
GitLab


From 01c2be8e1b97ee4891d1e1ffb7758897d441bb3c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:58 +0300
Subject: [PATCH 0014/1791] drm/i915: Add hardware csc readout for ilk+
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Read out the pipe/output csc matrices on ilk+ and stash the results
(in the hardware specific format) into the appropriate place
in the crtc state.

Note that on skl/glk/icl the pipe csc unit suffers from an issue
where *reads* of the coefficient/offset registers also disarm
the double buffer update (if currently armed via CSC_MODE write).
So it's rather important that the readout only happens after the
csc registers have been latched. Fortunately the state checker
only runs after the start of vblank where the latching happens.

And on skl/glk the DMC + CSC register read has the potential to
corrupt the latched CSC register values, so let's add a comment
reminding us that the DC states should remain off until the
readout has been completed.

TODO: maybe we could somehow check to make sure PSR has in fact
latched the new register values already, and that DC states
have been off all along?

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-9-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c   | 132 +++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_display.c |   6 +
 2 files changed, 138 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 2988c91d8ff6b..efa8a6decd9d3 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -70,6 +70,11 @@ struct intel_color_funcs {
 			  const struct drm_property_blob *blob1,
 			  const struct drm_property_blob *blob2,
 			  bool is_pre_csc_lut);
+	/*
+	 * Read out the CSCs (if any) from the hardware into the
+	 * software state. Used by eg. the hardware state checker.
+	 */
+	void (*read_csc)(struct intel_crtc_state *crtc_state);
 };
 
 #define CTM_COEFF_SIGN	(1ULL << 63)
@@ -227,6 +232,72 @@ static void ilk_update_pipe_csc(struct intel_crtc *crtc,
 	intel_de_write_fw(i915, PIPE_CSC_POSTOFF_LO(pipe), csc->postoff[2]);
 }
 
+static void ilk_read_pipe_csc(struct intel_crtc *crtc,
+			      struct intel_csc_matrix *csc)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 tmp;
+
+	csc->preoff[0] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_HI(pipe));
+	csc->preoff[1] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_ME(pipe));
+	csc->preoff[2] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_LO(pipe));
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RY_GY(pipe));
+	csc->coeff[0] = tmp >> 16;
+	csc->coeff[1] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BY(pipe));
+	csc->coeff[2] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RU_GU(pipe));
+	csc->coeff[3] = tmp >> 16;
+	csc->coeff[4] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BU(pipe));
+	csc->coeff[5] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RV_GV(pipe));
+	csc->coeff[6] = tmp >> 16;
+	csc->coeff[7] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BV(pipe));
+	csc->coeff[8] = tmp >> 16;
+
+	if (DISPLAY_VER(i915) < 7)
+		return;
+
+	csc->postoff[0] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_HI(pipe));
+	csc->postoff[1] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_ME(pipe));
+	csc->postoff[2] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_LO(pipe));
+}
+
+static void ilk_read_csc(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	if (crtc_state->csc_enable)
+		ilk_read_pipe_csc(crtc, &crtc_state->csc);
+}
+
+static void skl_read_csc(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	/*
+	 * Display WA #1184: skl,glk
+	 * Wa_1406463849: icl
+	 *
+	 * Danger! On SKL-ICL *reads* from the CSC coeff/offset registers
+	 * will disarm an already armed CSC double buffer update.
+	 * So this must not be called while armed. Fortunately the state checker
+	 * readout happens only after the update has been already been latched.
+	 *
+	 * On earlier and later platforms only writes to said registers will
+	 * disarm the update. This is considered normal behavior and also
+	 * happens with various other hardware units.
+	 */
+	if (crtc_state->csc_enable)
+		ilk_read_pipe_csc(crtc, &crtc_state->csc);
+}
+
 static void icl_update_output_csc(struct intel_crtc *crtc,
 				  const struct intel_csc_matrix *csc)
 {
@@ -257,6 +328,56 @@ static void icl_update_output_csc(struct intel_crtc *crtc,
 	intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), csc->postoff[2]);
 }
 
+static void icl_read_output_csc(struct intel_crtc *crtc,
+				struct intel_csc_matrix *csc)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 tmp;
+
+	csc->preoff[0] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe));
+	csc->preoff[1] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe));
+	csc->preoff[2] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe));
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe));
+	csc->coeff[0] = tmp >> 16;
+	csc->coeff[1] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BY(pipe));
+	csc->coeff[2] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe));
+	csc->coeff[3] = tmp >> 16;
+	csc->coeff[4] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BU(pipe));
+	csc->coeff[5] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe));
+	csc->coeff[6] = tmp >> 16;
+	csc->coeff[7] = tmp & 0xffff;
+	tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BV(pipe));
+	csc->coeff[8] = tmp >> 16;
+
+	csc->postoff[0] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe));
+	csc->postoff[1] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe));
+	csc->postoff[2] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe));
+}
+
+static void icl_read_csc(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	/*
+	 * Wa_1406463849: icl
+	 *
+	 * See skl_read_csc()
+	 */
+	if (crtc_state->csc_mode & ICL_CSC_ENABLE)
+		ilk_read_pipe_csc(crtc, &crtc_state->csc);
+
+	if (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE)
+		icl_read_output_csc(crtc, &crtc_state->output_csc);
+}
+
 static bool ilk_limited_range(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
@@ -1581,6 +1702,9 @@ void intel_color_get_config(struct intel_crtc_state *crtc_state)
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
 	i915->display.funcs.color->read_luts(crtc_state);
+
+	if (i915->display.funcs.color->read_csc)
+		i915->display.funcs.color->read_csc(crtc_state);
 }
 
 bool intel_color_lut_equal(const struct intel_crtc_state *crtc_state,
@@ -3229,6 +3353,7 @@ static const struct intel_color_funcs tgl_color_funcs = {
 	.load_luts = icl_load_luts,
 	.read_luts = icl_read_luts,
 	.lut_equal = icl_lut_equal,
+	.read_csc = icl_read_csc,
 };
 
 static const struct intel_color_funcs icl_color_funcs = {
@@ -3239,6 +3364,7 @@ static const struct intel_color_funcs icl_color_funcs = {
 	.load_luts = icl_load_luts,
 	.read_luts = icl_read_luts,
 	.lut_equal = icl_lut_equal,
+	.read_csc = icl_read_csc,
 };
 
 static const struct intel_color_funcs glk_color_funcs = {
@@ -3248,6 +3374,7 @@ static const struct intel_color_funcs glk_color_funcs = {
 	.load_luts = glk_load_luts,
 	.read_luts = glk_read_luts,
 	.lut_equal = glk_lut_equal,
+	.read_csc = skl_read_csc,
 };
 
 static const struct intel_color_funcs skl_color_funcs = {
@@ -3257,6 +3384,7 @@ static const struct intel_color_funcs skl_color_funcs = {
 	.load_luts = bdw_load_luts,
 	.read_luts = bdw_read_luts,
 	.lut_equal = ivb_lut_equal,
+	.read_csc = skl_read_csc,
 };
 
 static const struct intel_color_funcs bdw_color_funcs = {
@@ -3266,6 +3394,7 @@ static const struct intel_color_funcs bdw_color_funcs = {
 	.load_luts = bdw_load_luts,
 	.read_luts = bdw_read_luts,
 	.lut_equal = ivb_lut_equal,
+	.read_csc = ilk_read_csc,
 };
 
 static const struct intel_color_funcs hsw_color_funcs = {
@@ -3275,6 +3404,7 @@ static const struct intel_color_funcs hsw_color_funcs = {
 	.load_luts = ivb_load_luts,
 	.read_luts = ivb_read_luts,
 	.lut_equal = ivb_lut_equal,
+	.read_csc = ilk_read_csc,
 };
 
 static const struct intel_color_funcs ivb_color_funcs = {
@@ -3284,6 +3414,7 @@ static const struct intel_color_funcs ivb_color_funcs = {
 	.load_luts = ivb_load_luts,
 	.read_luts = ivb_read_luts,
 	.lut_equal = ivb_lut_equal,
+	.read_csc = ilk_read_csc,
 };
 
 static const struct intel_color_funcs ilk_color_funcs = {
@@ -3293,6 +3424,7 @@ static const struct intel_color_funcs ilk_color_funcs = {
 	.load_luts = ilk_load_luts,
 	.read_luts = ilk_read_luts,
 	.lut_equal = ilk_lut_equal,
+	.read_csc = ilk_read_csc,
 };
 
 void intel_color_crtc_init(struct intel_crtc *crtc)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 010ee793e1ffd..270658e93141b 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7313,6 +7313,12 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 	 * 7. New _arm() registers are finally written
 	 * 8. Hardware finally latches a complete set of new
 	 *    register values, and subsequent frames will be OK again
+	 *
+	 * Also note that due to the pipe CSC hardware issues on
+	 * SKL/GLK DC states must remain off until the pipe CSC
+	 * state readout has happened. Otherwise we risk corrupting
+	 * the CSC latched register values with the readout (see
+	 * skl_read_csc() and skl_color_commit_noarm()).
 	 */
 	wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_DC_OFF);
 
-- 
GitLab


From b6f4b3a1474d8bed9fad2b4d681368710375bbe9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:49:59 +0300
Subject: [PATCH 0015/1791] drm/i915: Implement chv cgm csc readout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Read out the csc matrix on chv, and stash the result into the
correct spot in the crtc state.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-10-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 36 ++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index efa8a6decd9d3..07f1afe1d4067 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -611,6 +611,41 @@ static void chv_load_cgm_csc(struct intel_crtc *crtc,
 			  csc->coeff[8]);
 }
 
+static void chv_read_cgm_csc(struct intel_crtc *crtc,
+			     struct intel_csc_matrix *csc)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 tmp;
+
+	tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF01(pipe));
+	csc->coeff[0] = tmp & 0xffff;
+	csc->coeff[1] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF23(pipe));
+	csc->coeff[2] = tmp & 0xffff;
+	csc->coeff[3] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF45(pipe));
+	csc->coeff[4] = tmp & 0xffff;
+	csc->coeff[5] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF67(pipe));
+	csc->coeff[6] = tmp & 0xffff;
+	csc->coeff[7] = tmp >> 16;
+
+	tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF8(pipe));
+	csc->coeff[8] = tmp & 0xffff;
+}
+
+static void chv_read_csc(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	if (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC)
+		chv_read_cgm_csc(crtc, &crtc_state->csc);
+}
+
 static void chv_assign_csc(struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
@@ -3328,6 +3363,7 @@ static const struct intel_color_funcs chv_color_funcs = {
 	.load_luts = chv_load_luts,
 	.read_luts = chv_read_luts,
 	.lut_equal = chv_lut_equal,
+	.read_csc = chv_read_csc,
 };
 
 static const struct intel_color_funcs i965_color_funcs = {
-- 
GitLab


From 37c8cabfcc5ce2c06baf0a2d0176043b0b256e49 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:50:00 +0300
Subject: [PATCH 0016/1791] drm/i915: Include the csc matrices in the crtc
 state dump
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Include the csc matrices in the state dump. The format being
hardware specific we just dump as hex for now. Might have
to think of some way to get a bit more human readable
output...

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-11-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 .../drm/i915/display/intel_crtc_state_dump.c  | 47 +++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
index 54c8adc0702ee..0cdcaa49656f1 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
@@ -158,6 +158,45 @@ static void intel_dump_plane_state(const struct intel_plane_state *plane_state)
 			    DRM_RECT_ARG(&plane_state->uapi.dst));
 }
 
+static void
+ilk_dump_csc(struct drm_i915_private *i915, const char *name,
+	     const struct intel_csc_matrix *csc)
+{
+	int i;
+
+	drm_dbg_kms(&i915->drm,
+		    "%s: pre offsets: 0x%04x 0x%04x 0x%04x\n", name,
+		    csc->preoff[0], csc->preoff[1], csc->preoff[2]);
+
+	for (i = 0; i < 3; i++)
+		drm_dbg_kms(&i915->drm,
+			    "%s: coefficients: 0x%04x 0x%04x 0x%04x\n", name,
+			    csc->coeff[3 * i + 0],
+			    csc->coeff[3 * i + 1],
+			    csc->coeff[3 * i + 2]);
+
+	if (DISPLAY_VER(i915) < 7)
+		return;
+
+	drm_dbg_kms(&i915->drm,
+		    "%s: post offsets: 0x%04x 0x%04x 0x%04x\n", name,
+		    csc->postoff[0], csc->postoff[1], csc->postoff[2]);
+}
+
+static void
+chv_dump_csc(struct drm_i915_private *i915, const char *name,
+	     const struct intel_csc_matrix *csc)
+{
+	int i;
+
+	for (i = 0; i < 3; i++)
+		drm_dbg_kms(&i915->drm,
+			    "%s: coefficients: 0x%04x 0x%04x 0x%04x\n", name,
+			    csc->coeff[3 * i + 0],
+			    csc->coeff[3 * i + 1],
+			    csc->coeff[3 * i + 2]);
+}
+
 void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 			   struct intel_atomic_state *state,
 			   const char *context)
@@ -325,6 +364,14 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 		    pipe_config->post_csc_lut ?
 		    drm_color_lut_size(pipe_config->post_csc_lut) : 0);
 
+	if (DISPLAY_VER(i915) >= 11)
+		ilk_dump_csc(i915, "output csc", &pipe_config->output_csc);
+
+	if (!HAS_GMCH(i915))
+		ilk_dump_csc(i915, "pipe csc", &pipe_config->csc);
+	else if (IS_CHERRYVIEW(i915))
+		chv_dump_csc(i915, "cgm csc", &pipe_config->csc);
+
 dump_planes:
 	if (!state)
 		return;
-- 
GitLab


From d6fff836c0e0502a569811c7708aed0762e04337 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 29 Mar 2023 16:50:01 +0300
Subject: [PATCH 0017/1791] drm/i915: Hook up csc into state checker
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Have the state checker validate that the csc matrices
look correct when read back from the hardware.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329135002.3096-12-ville.syrjala@linux.intel.com
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 21 ++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 270658e93141b..e6f8f2d585f45 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -5573,6 +5573,24 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 	} \
 } while (0)
 
+#define PIPE_CONF_CHECK_CSC(name) do { \
+	PIPE_CONF_CHECK_X(name.preoff[0]); \
+	PIPE_CONF_CHECK_X(name.preoff[1]); \
+	PIPE_CONF_CHECK_X(name.preoff[2]); \
+	PIPE_CONF_CHECK_X(name.coeff[0]); \
+	PIPE_CONF_CHECK_X(name.coeff[1]); \
+	PIPE_CONF_CHECK_X(name.coeff[2]); \
+	PIPE_CONF_CHECK_X(name.coeff[3]); \
+	PIPE_CONF_CHECK_X(name.coeff[4]); \
+	PIPE_CONF_CHECK_X(name.coeff[5]); \
+	PIPE_CONF_CHECK_X(name.coeff[6]); \
+	PIPE_CONF_CHECK_X(name.coeff[7]); \
+	PIPE_CONF_CHECK_X(name.coeff[8]); \
+	PIPE_CONF_CHECK_X(name.postoff[0]); \
+	PIPE_CONF_CHECK_X(name.postoff[1]); \
+	PIPE_CONF_CHECK_X(name.postoff[2]); \
+} while (0)
+
 #define PIPE_CONF_QUIRK(quirk) \
 	((current_config->quirks | pipe_config->quirks) & (quirk))
 
@@ -5670,6 +5688,9 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 		PIPE_CONF_CHECK_COLOR_LUT(pre_csc_lut, true);
 		PIPE_CONF_CHECK_COLOR_LUT(post_csc_lut, false);
 
+		PIPE_CONF_CHECK_CSC(csc);
+		PIPE_CONF_CHECK_CSC(output_csc);
+
 		if (current_config->active_planes) {
 			PIPE_CONF_CHECK_BOOL(has_psr);
 			PIPE_CONF_CHECK_BOOL(has_psr2);
-- 
GitLab


From b90b044c64f669cb20919fb5e5673933de59c653 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Tue, 4 Apr 2023 20:13:42 +0200
Subject: [PATCH 0018/1791] drm/i915/mtl: Disable stolen memory backed FB for
 A0

Stolen memory is not usable for MTL A0 stepping beyond
certain access size and we have no control over userspace
access size of /dev/fb which can be backed by stolen memory.
So disable stolen memory backed fb by setting i915->dsm.usable_size
to zero.

v2: remove hsdes reference and fix commit message(Andi)
v3: use revid as we want to target SOC stepping(Radhakrishna)

Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404181342.23362-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_stolen.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
index 8ac376c24aa22..ee492d823f1bb 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
@@ -535,6 +535,14 @@ static int i915_gem_init_stolen(struct intel_memory_region *mem)
 	/* Basic memrange allocator for stolen space. */
 	drm_mm_init(&i915->mm.stolen, 0, i915->dsm.usable_size);
 
+	/*
+	 * Access to stolen lmem beyond certain size for MTL A0 stepping
+	 * would crash the machine. Disable stolen lmem for userspace access
+	 * by setting usable_size to zero.
+	 */
+	if (IS_METEORLAKE(i915) && INTEL_REVID(i915) == 0x0)
+		i915->dsm.usable_size = 0;
+
 	return 0;
 }
 
-- 
GitLab


From bfa8342c27c67e86a7b7022df06848709386e00d Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 12 Apr 2023 10:09:20 +0200
Subject: [PATCH 0019/1791] MAINTAINERS: add drm_bridge for drm bridge
 maintainers

Otherwise core changes don't get noticed by the right people. I
noticed this because a patch set from Jagan Teki seems to have fallen
through the cracks.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jagan Teki <jagan@amarulasolutions.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Robert Foss <rfoss@kernel.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
[narmstrong: fixed ordering & Daniel's SoB]
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412080921.10171-1-daniel.vetter@ffwll.ch
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 904500cbdcf05..1558eebdcdeac 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6841,6 +6841,7 @@ S:	Maintained
 T:	git git://anongit.freedesktop.org/drm/drm-misc
 F:	Documentation/devicetree/bindings/display/bridge/
 F:	drivers/gpu/drm/bridge/
+F:	drivers/gpu/drm/drm_bridge.c
 F:	include/drm/drm_bridge.h
 
 DRM DRIVERS FOR EXYNOS
-- 
GitLab


From 1bf3836383e6957ac848ee81eb691820c862b3d6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 11 Apr 2023 16:19:22 +0300
Subject: [PATCH 0020/1791] drm/i915/display: remove unnecessary i915_debugfs.h
 includes

Leftovers from before display debugfs was separated to its own file.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411131922.401602-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c   | 1 -
 drivers/gpu/drm/i915/display/intel_hdmi.c | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index f0bace9d98a18..48d43f7f0c58a 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -44,7 +44,6 @@
 #include <drm/drm_probe_helper.h>
 
 #include "g4x_dp.h"
-#include "i915_debugfs.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index c7e9e1fbed379..a73a315ac8ad6 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -40,7 +40,6 @@
 #include <drm/drm_edid.h>
 #include <drm/intel_lpe_audio.h>
 
-#include "i915_debugfs.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
-- 
GitLab


From 99cc528ebe923d04767e9979665a0824727376ba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Mon, 16 Jan 2023 17:58:00 -0300
Subject: [PATCH 0021/1791] drm/vkms: Use drmm_crtc_init_with_planes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use drmm_crtc_init_with_planes() instead of drm_crtc_init_with_planes()
to get rid of the explicit destroy hook in struct drm_crtc_funcs.

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230116205800.1266227-1-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_crtc.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c
index 57bbd32e9bebb..515f6772b8663 100644
--- a/drivers/gpu/drm/vkms/vkms_crtc.c
+++ b/drivers/gpu/drm/vkms/vkms_crtc.c
@@ -161,7 +161,6 @@ static void vkms_atomic_crtc_reset(struct drm_crtc *crtc)
 
 static const struct drm_crtc_funcs vkms_crtc_funcs = {
 	.set_config             = drm_atomic_helper_set_config,
-	.destroy                = drm_crtc_cleanup,
 	.page_flip              = drm_atomic_helper_page_flip,
 	.reset                  = vkms_atomic_crtc_reset,
 	.atomic_duplicate_state = vkms_atomic_crtc_duplicate_state,
@@ -282,8 +281,8 @@ int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc,
 	struct vkms_output *vkms_out = drm_crtc_to_vkms_output(crtc);
 	int ret;
 
-	ret = drm_crtc_init_with_planes(dev, crtc, primary, cursor,
-					&vkms_crtc_funcs, NULL);
+	ret = drmm_crtc_init_with_planes(dev, crtc, primary, cursor,
+					 &vkms_crtc_funcs, NULL);
 	if (ret) {
 		DRM_ERROR("Failed to init CRTC\n");
 		return ret;
-- 
GitLab


From 17e05aeb3b987a17b2ee5185264bec5db1d0c746 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Mon, 16 Jan 2023 17:58:01 -0300
Subject: [PATCH 0022/1791] drm/vkms: Use drmm_mode_config_init()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use drmm_mode_config_init() instead of drm_mode_config_init(), as it allows
us to assure that the resource will be properly cleaned.

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230116205800.1266227-2-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_drv.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c
index 6d3a2d57d9921..e3c9c9571c8d6 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.c
+++ b/drivers/gpu/drm/vkms/vkms_drv.c
@@ -133,8 +133,12 @@ static const struct drm_mode_config_helper_funcs vkms_mode_config_helpers = {
 static int vkms_modeset_init(struct vkms_device *vkmsdev)
 {
 	struct drm_device *dev = &vkmsdev->drm;
+	int ret;
+
+	ret = drmm_mode_config_init(dev);
+	if (ret)
+		return ret;
 
-	drm_mode_config_init(dev);
 	dev->mode_config.funcs = &vkms_mode_funcs;
 	dev->mode_config.min_width = XRES_MIN;
 	dev->mode_config.min_height = YRES_MIN;
-- 
GitLab


From 6a98560755636b07ca54bf9cea6435b8e82b57d5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 20 Mar 2023 22:33:47 +0200
Subject: [PATCH 0023/1791] drm/i915: Generalize planes_{enabling,disabling}()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I want to use the same logic that planes_{enabling,disabling}()
are using for other features as well. Generlize the thing
into a pair of macros.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230320203352.19515-2-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index e6f8f2d585f45..d93cb8c63296f 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1069,20 +1069,28 @@ static bool needs_async_flip_vtd_wa(const struct intel_crtc_state *crtc_state)
 		(DISPLAY_VER(i915) == 9 || IS_BROADWELL(i915) || IS_HASWELL(i915));
 }
 
+#define is_enabling(feature, old_crtc_state, new_crtc_state) \
+	((!(old_crtc_state)->feature || intel_crtc_needs_modeset(new_crtc_state)) && \
+	 (new_crtc_state)->feature)
+#define is_disabling(feature, old_crtc_state, new_crtc_state) \
+	((old_crtc_state)->feature && \
+	 (!(new_crtc_state)->feature || intel_crtc_needs_modeset(new_crtc_state)))
+
 static bool planes_enabling(const struct intel_crtc_state *old_crtc_state,
 			    const struct intel_crtc_state *new_crtc_state)
 {
-	return (!old_crtc_state->active_planes || intel_crtc_needs_modeset(new_crtc_state)) &&
-		new_crtc_state->active_planes;
+	return is_enabling(active_planes, old_crtc_state, new_crtc_state);
 }
 
 static bool planes_disabling(const struct intel_crtc_state *old_crtc_state,
 			     const struct intel_crtc_state *new_crtc_state)
 {
-	return old_crtc_state->active_planes &&
-		(!new_crtc_state->active_planes || intel_crtc_needs_modeset(new_crtc_state));
+	return is_disabling(active_planes, old_crtc_state, new_crtc_state);
 }
 
+#undef is_disabling
+#undef is_enabling
+
 static void intel_post_plane_update(struct intel_atomic_state *state,
 				    struct intel_crtc *crtc)
 {
-- 
GitLab


From b25e07419fee6e3be07e58cc64f50e11228987d3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 20 Mar 2023 22:33:48 +0200
Subject: [PATCH 0024/1791] drm/i915/vrr: Eliminate redundant function
 arguments
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some of the VRR functions take redundant arguments. Get rid
of them to make life simpler.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230320203352.19515-3-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c     |  2 +-
 drivers/gpu/drm/i915/display/intel_display.c |  2 +-
 drivers/gpu/drm/i915/display/intel_vrr.c     | 10 ++++------
 drivers/gpu/drm/i915/display/intel_vrr.h     |  9 ++-------
 4 files changed, 8 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index dc294717bcdf4..1f453362d079c 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2963,7 +2963,7 @@ static void intel_enable_ddi(struct intel_atomic_state *state,
 
 	intel_enable_transcoder(crtc_state);
 
-	intel_vrr_enable(encoder, crtc_state);
+	intel_vrr_enable(crtc_state);
 
 	intel_crtc_vblank_on(crtc_state);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index d93cb8c63296f..d8ef03637830b 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3909,7 +3909,7 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc,
 		intel_get_transcoder_timings(crtc, pipe_config);
 
 	if (HAS_VRR(dev_priv) && !transcoder_is_dsi(pipe_config->cpu_transcoder))
-		intel_vrr_get_config(crtc, pipe_config);
+		intel_vrr_get_config(pipe_config);
 
 	intel_get_pipe_src_size(crtc, pipe_config);
 
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
index 4228f26b4c11d..6d749de71058c 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.c
+++ b/drivers/gpu/drm/i915/display/intel_vrr.c
@@ -168,10 +168,9 @@ static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state)
 			VRR_CTL_PIPELINE_FULL_OVERRIDE;
 }
 
-void intel_vrr_enable(struct intel_encoder *encoder,
-		      const struct intel_crtc_state *crtc_state)
+void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 
 	if (!crtc_state->vrr.enable)
@@ -230,10 +229,9 @@ void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state)
 	intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), 0);
 }
 
-void intel_vrr_get_config(struct intel_crtc *crtc,
-			  struct intel_crtc_state *crtc_state)
+void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	u32 trans_vrr_ctl;
 
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.h b/drivers/gpu/drm/i915/display/intel_vrr.h
index 9fda1135b0dd5..ee636a5367c8b 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.h
+++ b/drivers/gpu/drm/i915/display/intel_vrr.h
@@ -11,22 +11,17 @@
 struct drm_connector_state;
 struct intel_atomic_state;
 struct intel_connector;
-struct intel_crtc;
 struct intel_crtc_state;
-struct intel_dp;
-struct intel_encoder;
 
 bool intel_vrr_is_capable(struct intel_connector *connector);
 void intel_vrr_check_modeset(struct intel_atomic_state *state);
 void intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 			      struct drm_connector_state *conn_state);
-void intel_vrr_enable(struct intel_encoder *encoder,
-		      const struct intel_crtc_state *crtc_state);
+void intel_vrr_enable(const struct intel_crtc_state *crtc_state);
 void intel_vrr_send_push(const struct intel_crtc_state *crtc_state);
 bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state);
 void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state);
-void intel_vrr_get_config(struct intel_crtc *crtc,
-			  struct intel_crtc_state *crtc_state);
+void intel_vrr_get_config(struct intel_crtc_state *crtc_state);
 int intel_vrr_vmax_vblank_start(const struct intel_crtc_state *crtc_state);
 int intel_vrr_vmin_vblank_start(const struct intel_crtc_state *crtc_state);
 
-- 
GitLab


From fa9e4fce52ec4ee45ddfc6747ecb2bc8856c4753 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 20 Mar 2023 22:33:49 +0200
Subject: [PATCH 0025/1791] drm/i915/vrr: Make delayed vblank operational in
 VRR mode on adl/dg2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On adl/dg2 a chicken bit needs to be set for TRANS_SET_CONTENXT_LATENCY
to take effect in VRR mode. Can't really think of a reason why we'd
ever disable that chicken bit, so let's just always set it.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230320203352.19515-4-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_vrr.c | 8 ++++++++
 drivers/gpu/drm/i915/i915_reg.h          | 3 +--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
index 6d749de71058c..348a7cc8e6200 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.c
+++ b/drivers/gpu/drm/i915/display/intel_vrr.c
@@ -173,6 +173,14 @@ void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 
+	/*
+	 * TRANS_SET_CONTEXT_LATENCY with VRR enabled
+	 * requires this chicken bit on ADL/DG2.
+	 */
+	if (DISPLAY_VER(dev_priv) == 13)
+		intel_de_rmw(dev_priv, CHICKEN_TRANS(cpu_transcoder),
+			     0, PIPE_VBLANK_WITH_DELAY);
+
 	if (!crtc_state->vrr.enable)
 		return;
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 0e6dfdfcda8a4..ae387c702da9e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4560,13 +4560,12 @@
 					    [TRANSCODER_B] = _CHICKEN_TRANS_B, \
 					    [TRANSCODER_C] = _CHICKEN_TRANS_C, \
 					    [TRANSCODER_D] = _CHICKEN_TRANS_D))
-
 #define _MTL_CHICKEN_TRANS_A	0x604e0
 #define _MTL_CHICKEN_TRANS_B	0x614e0
 #define MTL_CHICKEN_TRANS(trans)	_MMIO_TRANS((trans), \
 						    _MTL_CHICKEN_TRANS_A, \
 						    _MTL_CHICKEN_TRANS_B)
-
+#define  PIPE_VBLANK_WITH_DELAY		REG_BIT(31) /* ADL/DG2 */
 #define  HSW_FRAME_START_DELAY_MASK	REG_GENMASK(28, 27)
 #define  HSW_FRAME_START_DELAY(x)	REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x)
 #define  VSC_DATA_SEL_SOFTWARE_CONTROL	REG_BIT(25) /* GLK */
-- 
GitLab


From ecaeecea9263496ecbb287aac6545e8b3cd9257d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 20 Mar 2023 22:33:50 +0200
Subject: [PATCH 0026/1791] drm/i915/vrr: Tell
 intel_crtc_update_active_timings() about VRR explicitly
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In order to move VRR enable/disable to a place where it's also
applicable to fastsets we need to be prepared to configure
the pipe into non-VRR mode initially, and then later switch
to VRR mode. To that end allow the active timings to be configured
in non-VRR mode temporarily even when the crtc_state says we're
going to be using VRR.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230320203352.19515-5-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_crtc.c          |  3 ++-
 drivers/gpu/drm/i915/display/intel_display.c       |  3 ++-
 drivers/gpu/drm/i915/display/intel_modeset_setup.c |  3 ++-
 drivers/gpu/drm/i915/display/intel_vblank.c        | 12 +++++++++---
 drivers/gpu/drm/i915/display/intel_vblank.h        |  3 ++-
 5 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index ed45a69348548..b92df8814f247 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -692,7 +692,8 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state)
 	 * FIXME Should be synchronized with the start of vblank somehow...
 	 */
 	if (new_crtc_state->seamless_m_n && intel_crtc_needs_fastset(new_crtc_state))
-		intel_crtc_update_active_timings(new_crtc_state);
+		intel_crtc_update_active_timings(new_crtc_state,
+						 new_crtc_state->vrr.enable);
 
 	local_irq_enable();
 
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index d8ef03637830b..921cf469682f9 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -6910,7 +6910,8 @@ static void intel_enable_crtc(struct intel_atomic_state *state,
 	if (!intel_crtc_needs_modeset(new_crtc_state))
 		return;
 
-	intel_crtc_update_active_timings(new_crtc_state);
+	intel_crtc_update_active_timings(new_crtc_state,
+					 new_crtc_state->vrr.enable);
 
 	dev_priv->display.funcs.display->crtc_enable(state, crtc);
 
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 134b943f19533..bc0695c5bb6f8 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -559,7 +559,8 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 			 */
 			crtc_state->inherited = true;
 
-			intel_crtc_update_active_timings(crtc_state);
+			intel_crtc_update_active_timings(crtc_state,
+							 crtc_state->vrr.enable);
 
 			intel_crtc_copy_hw_to_uapi_state(crtc_state);
 		}
diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index f8bf9810527de..2e4f7de199d60 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -488,21 +488,27 @@ static int intel_crtc_scanline_offset(const struct intel_crtc_state *crtc_state)
 	}
 }
 
-void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state)
+void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state,
+				      bool vrr_enable)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	u8 mode_flags = crtc_state->mode_flags;
 	struct drm_display_mode adjusted_mode;
 	int vmax_vblank_start = 0;
 	unsigned long irqflags;
 
 	drm_mode_init(&adjusted_mode, &crtc_state->hw.adjusted_mode);
 
-	if (crtc_state->vrr.enable) {
+	if (vrr_enable) {
+		drm_WARN_ON(&i915->drm, (mode_flags & I915_MODE_FLAG_VRR) == 0);
+
 		adjusted_mode.crtc_vtotal = crtc_state->vrr.vmax;
 		adjusted_mode.crtc_vblank_end = crtc_state->vrr.vmax;
 		adjusted_mode.crtc_vblank_start = intel_vrr_vmin_vblank_start(crtc_state);
 		vmax_vblank_start = intel_vrr_vmax_vblank_start(crtc_state);
+	} else {
+		mode_flags &= ~I915_MODE_FLAG_VRR;
 	}
 
 	/*
@@ -524,7 +530,7 @@ void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state)
 
 	crtc->vmax_vblank_start = vmax_vblank_start;
 
-	crtc->mode_flags = crtc_state->mode_flags;
+	crtc->mode_flags = mode_flags;
 
 	crtc->scanline_offset = intel_crtc_scanline_offset(crtc_state);
 
diff --git a/drivers/gpu/drm/i915/display/intel_vblank.h b/drivers/gpu/drm/i915/display/intel_vblank.h
index 0884db7e76ae0..08e706b291495 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.h
+++ b/drivers/gpu/drm/i915/display/intel_vblank.h
@@ -20,6 +20,7 @@ bool intel_crtc_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error,
 int intel_get_crtc_scanline(struct intel_crtc *crtc);
 void intel_wait_for_pipe_scanline_stopped(struct intel_crtc *crtc);
 void intel_wait_for_pipe_scanline_moving(struct intel_crtc *crtc);
-void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state);
+void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state,
+				      bool vrr_enable);
 
 #endif /* __INTEL_VBLANK_H__ */
-- 
GitLab


From 99cfbed19d06dfe9c9929c436b5a768231c05b70 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 21 Mar 2023 15:56:15 +0200
Subject: [PATCH 0027/1791] drm/i915/vrr: Relocate VRR enable/disable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move VRR enabling/disabling into a place where it also works
for fastsets.

With this we always start the transcoder up in non-VRR mode.
Granted  we already did that but for a very short period of
time. But now that we might end up doing a bit more with the
transcoder in non-VRR mode it seems prudent to also update
the active timings as the transcoder changes its operating
mode.

crtc_state->vrr.enable still tracks whether VRR is actually
enabled or not, but now we configure all the other VRR timing
registers whenever VRR is possible (whether we actually enable
it or not). crtc_state->vrr.flipline can now serve as our
"is VRR possible" bit of state.

I decided to leave the MSA timing ignore bit set all the time
whether VRR is actually enabled or not. If the sink can figure
out the timings with that information when VRR is active then
surely it can also do it when VRR is inactive.

v2: Protect intel_vrr_set_transcoder_timings() with HAS_VRR()

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230321135615.27338-1-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c      |  5 --
 drivers/gpu/drm/i915/display/intel_display.c  | 29 ++++++++++-
 .../drm/i915/display/intel_dp_link_training.c |  2 +-
 drivers/gpu/drm/i915/display/intel_vrr.c      | 48 +++++++++++--------
 drivers/gpu/drm/i915/display/intel_vrr.h      |  1 +
 5 files changed, 58 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 1f453362d079c..d5652e3b2fdb6 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -68,7 +68,6 @@
 #include "intel_tc.h"
 #include "intel_vdsc.h"
 #include "intel_vdsc_regs.h"
-#include "intel_vrr.h"
 #include "skl_scaler.h"
 #include "skl_universal_plane.h"
 
@@ -2725,8 +2724,6 @@ static void intel_ddi_post_disable(struct intel_atomic_state *state,
 	if (!intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_DP_MST)) {
 		intel_crtc_vblank_off(old_crtc_state);
 
-		intel_vrr_disable(old_crtc_state);
-
 		intel_disable_transcoder(old_crtc_state);
 
 		intel_ddi_disable_transcoder_func(old_crtc_state);
@@ -2963,8 +2960,6 @@ static void intel_enable_ddi(struct intel_atomic_state *state,
 
 	intel_enable_transcoder(crtc_state);
 
-	intel_vrr_enable(crtc_state);
-
 	intel_crtc_vblank_on(crtc_state);
 
 	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 921cf469682f9..25ffa114a380d 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1088,6 +1088,18 @@ static bool planes_disabling(const struct intel_crtc_state *old_crtc_state,
 	return is_disabling(active_planes, old_crtc_state, new_crtc_state);
 }
 
+static bool vrr_enabling(const struct intel_crtc_state *old_crtc_state,
+			 const struct intel_crtc_state *new_crtc_state)
+{
+	return is_enabling(vrr.enable, old_crtc_state, new_crtc_state);
+}
+
+static bool vrr_disabling(const struct intel_crtc_state *old_crtc_state,
+			  const struct intel_crtc_state *new_crtc_state)
+{
+	return is_disabling(vrr.enable, old_crtc_state, new_crtc_state);
+}
+
 #undef is_disabling
 #undef is_enabling
 
@@ -1204,6 +1216,11 @@ static void intel_pre_plane_update(struct intel_atomic_state *state,
 		intel_atomic_get_new_crtc_state(state, crtc);
 	enum pipe pipe = crtc->pipe;
 
+	if (vrr_disabling(old_crtc_state, new_crtc_state)) {
+		intel_vrr_disable(old_crtc_state);
+		intel_crtc_update_active_timings(old_crtc_state, false);
+	}
+
 	intel_drrs_deactivate(old_crtc_state);
 
 	intel_psr_pre_plane_update(state, crtc);
@@ -1684,6 +1701,8 @@ static void hsw_configure_cpu_transcoder(const struct intel_crtc_state *crtc_sta
 	}
 
 	intel_set_transcoder_timings(crtc_state);
+	if (HAS_VRR(dev_priv))
+		intel_vrr_set_transcoder_timings(crtc_state);
 
 	if (cpu_transcoder != TRANSCODER_EDP)
 		intel_de_write(dev_priv, TRANS_MULT(cpu_transcoder),
@@ -6910,8 +6929,8 @@ static void intel_enable_crtc(struct intel_atomic_state *state,
 	if (!intel_crtc_needs_modeset(new_crtc_state))
 		return;
 
-	intel_crtc_update_active_timings(new_crtc_state,
-					 new_crtc_state->vrr.enable);
+	/* VRR will be enable later, if required */
+	intel_crtc_update_active_timings(new_crtc_state, false);
 
 	dev_priv->display.funcs.display->crtc_enable(state, crtc);
 
@@ -6938,6 +6957,12 @@ static void intel_update_crtc(struct intel_atomic_state *state,
 			intel_dpt_configure(crtc);
 	}
 
+	if (vrr_enabling(old_crtc_state, new_crtc_state)) {
+		intel_vrr_enable(new_crtc_state);
+		intel_crtc_update_active_timings(new_crtc_state,
+						 new_crtc_state->vrr.enable);
+	}
+
 	if (!modeset) {
 		if (new_crtc_state->preload_luts &&
 		    intel_crtc_needs_color_update(new_crtc_state))
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index d638054c74ac1..6aa4ae5e7ebe3 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -699,7 +699,7 @@ intel_dp_prepare_link_train(struct intel_dp *intel_dp,
 		drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET,
 				  &rate_select, 1);
 
-	link_config[0] = crtc_state->vrr.enable ? DP_MSA_TIMING_PAR_IGNORE_EN : 0;
+	link_config[0] = crtc_state->vrr.flipline ? DP_MSA_TIMING_PAR_IGNORE_EN : 0;
 	link_config[1] = intel_dp_is_uhbr(crtc_state) ?
 		DP_SET_ANSI_128B132B : DP_SET_ANSI_8B10B;
 	drm_dp_dpcd_write(&intel_dp->aux, DP_DOWNSPREAD_CTRL, link_config, 2);
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c
index 348a7cc8e6200..88e4759b538b6 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.c
+++ b/drivers/gpu/drm/i915/display/intel_vrr.c
@@ -114,9 +114,6 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE)
 		return;
 
-	if (!crtc_state->uapi.vrr_enabled)
-		return;
-
 	vmin = DIV_ROUND_UP(adjusted_mode->crtc_clock * 1000,
 			    adjusted_mode->crtc_htotal * info->monitor_range.max_vfreq);
 	vmax = adjusted_mode->crtc_clock * 1000 /
@@ -135,7 +132,6 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 	 */
 	crtc_state->vrr.vmin = vmin - 1;
 	crtc_state->vrr.vmax = vmax;
-	crtc_state->vrr.enable = true;
 
 	crtc_state->vrr.flipline = crtc_state->vrr.vmin + 1;
 
@@ -152,7 +148,10 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 			    crtc_state->framestart_delay - 1);
 	}
 
-	crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
+	if (crtc_state->uapi.vrr_enabled) {
+		crtc_state->vrr.enable = true;
+		crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
+	}
 }
 
 static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state)
@@ -168,7 +167,7 @@ static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state)
 			VRR_CTL_PIPELINE_FULL_OVERRIDE;
 }
 
-void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
+void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
@@ -181,17 +180,15 @@ void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
 		intel_de_rmw(dev_priv, CHICKEN_TRANS(cpu_transcoder),
 			     0, PIPE_VBLANK_WITH_DELAY);
 
-	if (!crtc_state->vrr.enable)
+	if (!crtc_state->vrr.flipline) {
+		intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), 0);
 		return;
+	}
 
 	intel_de_write(dev_priv, TRANS_VRR_VMIN(cpu_transcoder), crtc_state->vrr.vmin - 1);
 	intel_de_write(dev_priv, TRANS_VRR_VMAX(cpu_transcoder), crtc_state->vrr.vmax - 1);
 	intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), trans_vrr_ctl(crtc_state));
 	intel_de_write(dev_priv, TRANS_VRR_FLIPLINE(cpu_transcoder), crtc_state->vrr.flipline - 1);
-	intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), TRANS_PUSH_EN);
-
-	intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder),
-		       VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state));
 }
 
 void intel_vrr_send_push(const struct intel_crtc_state *crtc_state)
@@ -219,6 +216,19 @@ bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state)
 	return intel_de_read(dev_priv, TRANS_PUSH(cpu_transcoder)) & TRANS_PUSH_SEND;
 }
 
+void intel_vrr_enable(const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
+
+	if (!crtc_state->vrr.enable)
+		return;
+
+	intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), TRANS_PUSH_EN);
+	intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder),
+		       VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state));
+}
+
 void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->uapi.crtc);
@@ -232,9 +242,7 @@ void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state)
 		       trans_vrr_ctl(old_crtc_state));
 	intel_de_wait_for_clear(dev_priv, TRANS_VRR_STATUS(cpu_transcoder),
 				VRR_STATUS_VRR_EN_LIVE, 1000);
-
 	intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), 0);
-	intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), 0);
 }
 
 void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
@@ -244,9 +252,8 @@ void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 	u32 trans_vrr_ctl;
 
 	trans_vrr_ctl = intel_de_read(dev_priv, TRANS_VRR_CTL(cpu_transcoder));
+
 	crtc_state->vrr.enable = trans_vrr_ctl & VRR_CTL_VRR_ENABLE;
-	if (!crtc_state->vrr.enable)
-		return;
 
 	if (DISPLAY_VER(dev_priv) >= 13)
 		crtc_state->vrr.guardband =
@@ -255,10 +262,13 @@ void intel_vrr_get_config(struct intel_crtc_state *crtc_state)
 		if (trans_vrr_ctl & VRR_CTL_PIPELINE_FULL_OVERRIDE)
 			crtc_state->vrr.pipeline_full =
 				REG_FIELD_GET(VRR_CTL_PIPELINE_FULL_MASK, trans_vrr_ctl);
-	if (trans_vrr_ctl & VRR_CTL_FLIP_LINE_EN)
+
+	if (trans_vrr_ctl & VRR_CTL_FLIP_LINE_EN) {
 		crtc_state->vrr.flipline = intel_de_read(dev_priv, TRANS_VRR_FLIPLINE(cpu_transcoder)) + 1;
-	crtc_state->vrr.vmax = intel_de_read(dev_priv, TRANS_VRR_VMAX(cpu_transcoder)) + 1;
-	crtc_state->vrr.vmin = intel_de_read(dev_priv, TRANS_VRR_VMIN(cpu_transcoder)) + 1;
+		crtc_state->vrr.vmax = intel_de_read(dev_priv, TRANS_VRR_VMAX(cpu_transcoder)) + 1;
+		crtc_state->vrr.vmin = intel_de_read(dev_priv, TRANS_VRR_VMIN(cpu_transcoder)) + 1;
+	}
 
-	crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
+	if (crtc_state->vrr.enable)
+		crtc_state->mode_flags |= I915_MODE_FLAG_VRR;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_vrr.h b/drivers/gpu/drm/i915/display/intel_vrr.h
index ee636a5367c8b..de16960c4929f 100644
--- a/drivers/gpu/drm/i915/display/intel_vrr.h
+++ b/drivers/gpu/drm/i915/display/intel_vrr.h
@@ -17,6 +17,7 @@ bool intel_vrr_is_capable(struct intel_connector *connector);
 void intel_vrr_check_modeset(struct intel_atomic_state *state);
 void intel_vrr_compute_config(struct intel_crtc_state *crtc_state,
 			      struct drm_connector_state *conn_state);
+void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state);
 void intel_vrr_enable(const struct intel_crtc_state *crtc_state);
 void intel_vrr_send_push(const struct intel_crtc_state *crtc_state);
 bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state);
-- 
GitLab


From 1af1d18825d3a5d36b6a3e5049998c3f09321145 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 20 Mar 2023 22:33:52 +0200
Subject: [PATCH 0028/1791] drm/i915/vrr: Allow VRR to be toggled during
 fastsets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that VRR enable/disable are called from convenient places
it is trivial to allow it to change state during fastsets.
Make it so.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7542
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230320203352.19515-7-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 25ffa114a380d..27b47680573a2 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -5804,7 +5804,8 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 	PIPE_CONF_CHECK_I(splitter.link_count);
 	PIPE_CONF_CHECK_I(splitter.pixel_overlap);
 
-	PIPE_CONF_CHECK_BOOL(vrr.enable);
+	if (!fastset)
+		PIPE_CONF_CHECK_BOOL(vrr.enable);
 	PIPE_CONF_CHECK_I(vrr.vmin);
 	PIPE_CONF_CHECK_I(vrr.vmax);
 	PIPE_CONF_CHECK_I(vrr.flipline);
-- 
GitLab


From 76ec69272195317080c16b970d23aebdaf192883 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 28 Mar 2023 15:23:57 +0300
Subject: [PATCH 0029/1791] drm/i915: Flag purely internal commits to not clear
 crtc_state->inherited
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If we have to force the hardware to go through a full modeset
due to eg. cdclk reprogramming, we need to preserve
crtc_state->inherited for all crtcs that have not otherwise
gone through the whole compute_config() stuff after connectors
have been detected.

Otherwise eg. cdclk induced modeset glk_force_audio_cdclk()
will clear the inherited flag, and thus the first real commit
coming from userspace later on will not be forced through
the full .compute_config() path and so eg. audio state may
not get properly recomputed.

But instead of adding all kinds of ad-hoc crtc_state->inherited
preservation hacks all over, let's change things so that we
only clear it for the crtcs directly included in userspace/client
initiated commits.

Should be far less fragile since now we just need to remember
to flag the internal commits, and not worry about where new
crtcs might get pulled in.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5260
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230328122357.1697-1-ville.syrjala@linux.intel.com
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/gpu/drm/i915/display/i9xx_wm.c        |  3 ++-
 drivers/gpu/drm/i915/display/intel_atomic.c   |  3 ++-
 drivers/gpu/drm/i915/display/intel_audio.c    |  1 +
 drivers/gpu/drm/i915/display/intel_ddi.c      |  1 +
 drivers/gpu/drm/i915/display/intel_display.c  | 22 ++++++++++---------
 .../drm/i915/display/intel_display_types.h    |  3 +++
 .../drm/i915/display/intel_modeset_setup.c    |  1 +
 drivers/gpu/drm/i915/display/intel_pipe_crc.c |  1 +
 drivers/gpu/drm/i915/display/intel_psr.c      |  3 ++-
 .../gpu/drm/i915/display/intel_sprite_uapi.c  |  1 +
 10 files changed, 26 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index caef72d387981..6288826a94974 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -3447,9 +3447,10 @@ void ilk_wm_sanitize(struct drm_i915_private *dev_priv)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-retry:
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
+retry:
 	/*
 	 * Hardware readout is the only time we don't want to calculate
 	 * intermediate watermarks (since we don't trust the current
diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
index a9a3f3715279d..61011641f6ab8 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic.c
@@ -265,7 +265,6 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 	crtc_state->update_wm_post = false;
 	crtc_state->fifo_changed = false;
 	crtc_state->preload_luts = false;
-	crtc_state->inherited = false;
 	crtc_state->wm.need_postvbl_update = false;
 	crtc_state->do_async_flip = false;
 	crtc_state->fb_bits = 0;
@@ -599,6 +598,8 @@ void intel_atomic_state_clear(struct drm_atomic_state *s)
 	drm_atomic_state_default_clear(&state->base);
 	intel_atomic_clear_global_state(state);
 
+	/* state->internal not reset on purpose */
+
 	state->dpll_set = state->modeset = false;
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c
index 3d5a9bbc6fde3..3d9c9b4f27f80 100644
--- a/drivers/gpu/drm/i915/display/intel_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_audio.c
@@ -1039,6 +1039,7 @@ static void glk_force_audio_cdclk(struct drm_i915_private *i915,
 		return;
 
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 retry:
 	ret = glk_force_audio_cdclk_commit(to_intel_atomic_state(state), crtc,
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index d5652e3b2fdb6..d6fb07821cffb 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3903,6 +3903,7 @@ static int modeset_pipe(struct drm_crtc *crtc,
 		return -ENOMEM;
 
 	state->acquire_ctx = ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 	crtc_state = drm_atomic_get_crtc_state(state, crtc);
 	if (IS_ERR(crtc_state)) {
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 27b47680573a2..0334565cec824 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -4137,7 +4137,10 @@ int intel_get_load_detect_pipe(struct drm_connector *connector,
 	}
 
 	state->acquire_ctx = ctx;
+	to_intel_atomic_state(state)->internal = true;
+
 	restore_state->acquire_ctx = ctx;
+	to_intel_atomic_state(restore_state)->internal = true;
 
 	connector_state = drm_atomic_get_connector_state(state, connector);
 	if (IS_ERR(connector_state)) {
@@ -6585,6 +6588,13 @@ int intel_atomic_check(struct drm_device *dev,
 
 	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
 					    new_crtc_state, i) {
+		/*
+		 * crtc's state no longer considered to be inherited
+		 * after the first userspace/client initiated commit.
+		 */
+		if (!state->internal)
+			new_crtc_state->inherited = false;
+
 		if (new_crtc_state->inherited != old_crtc_state->inherited)
 			new_crtc_state->uapi.mode_changed = true;
 
@@ -8285,9 +8295,10 @@ static int intel_initial_commit(struct drm_device *dev)
 
 	drm_modeset_acquire_init(&ctx, 0);
 
-retry:
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
+retry:
 	for_each_intel_crtc(dev, crtc) {
 		struct intel_crtc_state *crtc_state =
 			intel_atomic_get_crtc_state(state, crtc);
@@ -8300,15 +8311,6 @@ static int intel_initial_commit(struct drm_device *dev)
 		if (crtc_state->hw.active) {
 			struct intel_encoder *encoder;
 
-			/*
-			 * We've not yet detected sink capabilities
-			 * (audio,infoframes,etc.) and thus we don't want to
-			 * force a full state recomputation yet. We want that to
-			 * happen only for the first real commit from userspace.
-			 * So preserve the inherited flag for the time being.
-			 */
-			crtc_state->inherited = true;
-
 			ret = drm_atomic_add_affected_planes(state, &crtc->base);
 			if (ret)
 				goto out;
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index ed02399ee41dc..b28c8dc48e244 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -643,6 +643,9 @@ struct intel_atomic_state {
 	struct __intel_global_objs_state *global_objs;
 	int num_global_objs;
 
+	/* Internal commit, as opposed to userspace/client initiated one */
+	bool internal;
+
 	bool dpll_set, modeset;
 
 	struct intel_shared_dpll_state shared_dpll[I915_NUM_PLLS];
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index bc0695c5bb6f8..eefa4018dc0c2 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -69,6 +69,7 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 	}
 
 	state->acquire_ctx = ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 	/* Everything's already locked, -EDEADLK can't happen. */
 	temp_crtc_state = intel_atomic_get_crtc_state(state, crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_pipe_crc.c b/drivers/gpu/drm/i915/display/intel_pipe_crc.c
index 8d3ea8d7b7372..5a468ed6e26c7 100644
--- a/drivers/gpu/drm/i915/display/intel_pipe_crc.c
+++ b/drivers/gpu/drm/i915/display/intel_pipe_crc.c
@@ -293,6 +293,7 @@ intel_crtc_crc_setup_workarounds(struct intel_crtc *crtc, bool enable)
 	}
 
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 retry:
 	pipe_config = intel_atomic_get_crtc_state(state, crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 6badfff2b4a28..3e2d9998566c7 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -2149,10 +2149,11 @@ static int intel_psr_fastset_force(struct drm_i915_private *dev_priv)
 		return -ENOMEM;
 
 	drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
+
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 retry:
-
 	drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter);
 	drm_for_each_connector_iter(conn, &conn_iter) {
 		struct drm_connector_state *conn_state;
diff --git a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
index 70a3910837519..a76b48ebc2d31 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c
@@ -86,6 +86,7 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
 		goto out;
 	}
 	state->acquire_ctx = &ctx;
+	to_intel_atomic_state(state)->internal = true;
 
 	while (1) {
 		plane_state = drm_atomic_get_plane_state(state, plane);
-- 
GitLab


From e39c76b2160bbd005587f978d29603ef790aefcd Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Thu, 30 Mar 2023 20:31:04 +0530
Subject: [PATCH 0030/1791] drm/i915/color: Fix typo for Plane CSC indexes

Replace _PLANE_INPUT_CSC_RY_GY_2_* with _PLANE_CSC_RY_GY_2_*
for Plane CSC

Fixes: 6eba56f64d5d ("drm/i915/pxp: black pixels on pxp disabled")

Cc: <stable@vger.kernel.org>

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330150104.2923519-1-chaitanya.kumar.borah@intel.com
---
 drivers/gpu/drm/i915/i915_reg.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ae387c702da9e..4511f01ca7056 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6338,8 +6338,8 @@ enum skl_power_gate {
 
 #define _PLANE_CSC_RY_GY_1(pipe)	_PIPE(pipe, _PLANE_CSC_RY_GY_1_A, \
 					      _PLANE_CSC_RY_GY_1_B)
-#define _PLANE_CSC_RY_GY_2(pipe)	_PIPE(pipe, _PLANE_INPUT_CSC_RY_GY_2_A, \
-					      _PLANE_INPUT_CSC_RY_GY_2_B)
+#define _PLANE_CSC_RY_GY_2(pipe)	_PIPE(pipe, _PLANE_CSC_RY_GY_2_A, \
+					      _PLANE_CSC_RY_GY_2_B)
 #define PLANE_CSC_COEFF(pipe, plane, index)	_MMIO_PLANE(plane, \
 							    _PLANE_CSC_RY_GY_1(pipe) +  (index) * 4, \
 							    _PLANE_CSC_RY_GY_2(pipe) + (index) * 4)
-- 
GitLab


From a2da67028cd05516343533c1609fcaf037237fed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 4 Apr 2023 20:54:29 +0300
Subject: [PATCH 0031/1791] drm/i915: Allow arbitrary refresh rates with VRR
 eDP panels
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If the panel supports VRR it must be capable of accepting
timings with arbitrary vblank length, within the valid VRR
range. Use that fact to allow the user to request any refresh
rate they like. We simply pick the next highest fixed mode
from our list, and adjust the vblank to get the desired refresh
rate in the end.

Of course currently everything to do with the vrefresh is
using 1Hz precision, so might not be exact. But we can improve
that in the future by just upping our vrefresh precision.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404175431.23064-1-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_panel.c | 80 ++++++++++++++++++----
 1 file changed, 66 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c
index ce2a34a252117..9acdd68b2dbcf 100644
--- a/drivers/gpu/drm/i915/display/intel_panel.c
+++ b/drivers/gpu/drm/i915/display/intel_panel.c
@@ -42,6 +42,7 @@
 #include "intel_lvds_regs.h"
 #include "intel_panel.h"
 #include "intel_quirks.h"
+#include "intel_vrr.h"
 
 bool intel_panel_use_ssc(struct drm_i915_private *i915)
 {
@@ -58,6 +59,38 @@ intel_panel_preferred_fixed_mode(struct intel_connector *connector)
 					struct drm_display_mode, head);
 }
 
+static bool is_in_vrr_range(struct intel_connector *connector, int vrefresh)
+{
+	const struct drm_display_info *info = &connector->base.display_info;
+
+	return intel_vrr_is_capable(connector) &&
+		vrefresh >= info->monitor_range.min_vfreq &&
+		vrefresh <= info->monitor_range.max_vfreq;
+}
+
+static bool is_best_fixed_mode(struct intel_connector *connector,
+			       int vrefresh, int fixed_mode_vrefresh,
+			       const struct drm_display_mode *best_mode)
+{
+	/* we want to always return something */
+	if (!best_mode)
+		return true;
+
+	/*
+	 * With VRR always pick a mode with equal/higher than requested
+	 * vrefresh, which we can then reduce to match the requested
+	 * vrefresh by extending the vblank length.
+	 */
+	if (is_in_vrr_range(connector, vrefresh) &&
+	    is_in_vrr_range(connector, fixed_mode_vrefresh) &&
+	    fixed_mode_vrefresh < vrefresh)
+		return false;
+
+	/* pick the fixed_mode that is closest in terms of vrefresh */
+	return abs(fixed_mode_vrefresh - vrefresh) <
+		abs(drm_mode_vrefresh(best_mode) - vrefresh);
+}
+
 const struct drm_display_mode *
 intel_panel_fixed_mode(struct intel_connector *connector,
 		       const struct drm_display_mode *mode)
@@ -65,11 +98,11 @@ intel_panel_fixed_mode(struct intel_connector *connector,
 	const struct drm_display_mode *fixed_mode, *best_mode = NULL;
 	int vrefresh = drm_mode_vrefresh(mode);
 
-	/* pick the fixed_mode that is closest in terms of vrefresh */
 	list_for_each_entry(fixed_mode, &connector->panel.fixed_modes, head) {
-		if (!best_mode ||
-		    abs(drm_mode_vrefresh(fixed_mode) - vrefresh) <
-		    abs(drm_mode_vrefresh(best_mode) - vrefresh))
+		int fixed_mode_vrefresh = drm_mode_vrefresh(fixed_mode);
+
+		if (is_best_fixed_mode(connector, vrefresh,
+				       fixed_mode_vrefresh, best_mode))
 			best_mode = fixed_mode;
 	}
 
@@ -178,27 +211,46 @@ int intel_panel_compute_config(struct intel_connector *connector,
 {
 	const struct drm_display_mode *fixed_mode =
 		intel_panel_fixed_mode(connector, adjusted_mode);
+	int vrefresh, fixed_mode_vrefresh;
+	bool is_vrr;
 
 	if (!fixed_mode)
 		return 0;
 
+	vrefresh = drm_mode_vrefresh(adjusted_mode);
+	fixed_mode_vrefresh = drm_mode_vrefresh(fixed_mode);
+
 	/*
-	 * We don't want to lie too much to the user about the refresh
-	 * rate they're going to get. But we have to allow a bit of latitude
-	 * for Xorg since it likes to automagically cook up modes with slightly
-	 * off refresh rates.
+	 * Assume that we shouldn't muck about with the
+	 * timings if they don't land in the VRR range.
 	 */
-	if (abs(drm_mode_vrefresh(adjusted_mode) - drm_mode_vrefresh(fixed_mode)) > 1) {
-		drm_dbg_kms(connector->base.dev,
-			    "[CONNECTOR:%d:%s] Requested mode vrefresh (%d Hz) does not match fixed mode vrefresh (%d Hz)\n",
-			    connector->base.base.id, connector->base.name,
-			    drm_mode_vrefresh(adjusted_mode), drm_mode_vrefresh(fixed_mode));
+	is_vrr = is_in_vrr_range(connector, vrefresh) &&
+		is_in_vrr_range(connector, fixed_mode_vrefresh);
 
-		return -EINVAL;
+	if (!is_vrr) {
+		/*
+		 * We don't want to lie too much to the user about the refresh
+		 * rate they're going to get. But we have to allow a bit of latitude
+		 * for Xorg since it likes to automagically cook up modes with slightly
+		 * off refresh rates.
+		 */
+		if (abs(vrefresh - fixed_mode_vrefresh) > 1) {
+			drm_dbg_kms(connector->base.dev,
+				    "[CONNECTOR:%d:%s] Requested mode vrefresh (%d Hz) does not match fixed mode vrefresh (%d Hz)\n",
+				    connector->base.base.id, connector->base.name,
+				    vrefresh, fixed_mode_vrefresh);
+
+			return -EINVAL;
+		}
 	}
 
 	drm_mode_copy(adjusted_mode, fixed_mode);
 
+	if (is_vrr && fixed_mode_vrefresh != vrefresh)
+		adjusted_mode->vtotal =
+			DIV_ROUND_CLOSEST(adjusted_mode->clock * 1000,
+					  adjusted_mode->htotal * vrefresh);
+
 	drm_mode_set_crtcinfo(adjusted_mode, 0);
 
 	return 0;
-- 
GitLab


From 435db526a68b6454a882eae7a3768c516d4b540e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 4 Apr 2023 20:54:30 +0300
Subject: [PATCH 0032/1791] drm/i915: Evade transcoder's vblank when doing
 seamless M/N changes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The transcoder M/N values are double buffered on the transcoder's
undelayed vblank. So when doing seamless M/N fastsets we need to
evade also that.

Note that currently the pipe's delayed vblank == transcoder's
undelayed vblank, so this is still a nop change. But in the
future when we may have to delay the pipe's vblank to create
a register programming window ("window2") for the DSB.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404175431.23064-2-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com
---
 drivers/gpu/drm/i915/display/intel_crtc.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index b92df8814f247..df7d05f1e14b3 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -510,6 +510,13 @@ void intel_pipe_update_start(struct intel_crtc_state *new_crtc_state)
 						      VBLANK_EVASION_TIME_US);
 	max = vblank_start - 1;
 
+	/*
+	 * M/N is double buffered on the transcoder's undelayed vblank,
+	 * so with seamless M/N we must evade both vblanks.
+	 */
+	if (new_crtc_state->seamless_m_n && intel_crtc_needs_fastset(new_crtc_state))
+		min -= adjusted_mode->crtc_vblank_start - adjusted_mode->crtc_vdisplay;
+
 	if (min <= 0 || max <= 0)
 		goto irq_disable;
 
-- 
GitLab


From e13b3f65af719e79046df2b1da6599eb387752e0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 4 Apr 2023 20:54:31 +0300
Subject: [PATCH 0033/1791] drm/i915: Use min() instead of hand rolling it
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Most places in the vblank code use min() to clamp scanline
counters below vtotal. But we missed one in the gen3/4
pixel counter based codepath.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404175431.23064-3-ville.syrjala@linux.intel.com
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
---
 drivers/gpu/drm/i915/display/intel_vblank.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index 2e4f7de199d60..f5659ebd08eb2 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -340,8 +340,7 @@ static bool i915_get_crtc_scanoutpos(struct drm_crtc *_crtc,
 		 * matches how the scanline counter based position works since
 		 * the scanline counter doesn't count the two half lines.
 		 */
-		if (position >= vtotal)
-			position = vtotal - 1;
+		position = min(position, vtotal - 1);
 
 		/*
 		 * Start of vblank interrupt is triggered at start of hsync,
-- 
GitLab


From d7c281eecec0699449ca9ecfff82fe056dddb488 Mon Sep 17 00:00:00 2001
From: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
Date: Thu, 13 Apr 2023 17:15:02 +0530
Subject: [PATCH 0034/1791] drm/i915/debugfs: New debugfs for display clock
 frequencies

Instead of mixing display & non-display stuff together, move
display specific clock info to new debugfs. This patch will
create a new debugfs "i915_cdclk_info" to expose Current & Max
cdclk and Max pixel clock frequency info.

Example:
$ cat /sys/kernel/debug/dri/0/i915_cdclk_info
Current CD clock frequency: 163200 kHz
Max CD clock frequency: 652800 kHz
Max pixel clock frequency: 1305600 kHz

V2: - s/i915_display_clock_info/i915_cdclk_info/ (Jani)
    - Move the logic to intel_cdclk.c (Jani)
    - Don't remove info from i915_frequency_info (Jani)
V3: - Drop locking (Jani)

Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413114502.1105288-1-bhanuprakash.modem@intel.com
---
 drivers/gpu/drm/i915/display/intel_cdclk.c    | 21 +++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_cdclk.h    |  1 +
 .../drm/i915/display/intel_display_debugfs.c  |  1 +
 3 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 084a483f97766..f6223d8f13b8f 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -3235,6 +3235,27 @@ u32 intel_read_rawclk(struct drm_i915_private *dev_priv)
 	return freq;
 }
 
+static int i915_cdclk_info_show(struct seq_file *m, void *unused)
+{
+	struct drm_i915_private *i915 = m->private;
+
+	seq_printf(m, "Current CD clock frequency: %d kHz\n", i915->display.cdclk.hw.cdclk);
+	seq_printf(m, "Max CD clock frequency: %d kHz\n", i915->display.cdclk.max_cdclk_freq);
+	seq_printf(m, "Max pixel clock frequency: %d kHz\n", i915->max_dotclk_freq);
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(i915_cdclk_info);
+
+void intel_cdclk_debugfs_register(struct drm_i915_private *i915)
+{
+	struct drm_minor *minor = i915->drm.primary;
+
+	debugfs_create_file("i915_cdclk_info", 0444, minor->debugfs_root,
+			    i915, &i915_cdclk_info_fops);
+}
+
 static const struct intel_cdclk_funcs mtl_cdclk_funcs = {
 	.get_cdclk = bxt_get_cdclk,
 	.set_cdclk = bxt_set_cdclk,
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.h b/drivers/gpu/drm/i915/display/intel_cdclk.h
index 51e2f6a11ce41..48fd7d39e0cd9 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.h
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.h
@@ -82,5 +82,6 @@ intel_atomic_get_cdclk_state(struct intel_atomic_state *state);
 	to_intel_cdclk_state(intel_atomic_get_new_global_obj_state(state, &to_i915(state->base.dev)->display.cdclk.obj))
 
 int intel_cdclk_init(struct drm_i915_private *dev_priv);
+void intel_cdclk_debugfs_register(struct drm_i915_private *i915);
 
 #endif /* __INTEL_CDCLK_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index 45113ae107ba8..abd16a2b1f7a0 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -1094,6 +1094,7 @@ void intel_display_debugfs_register(struct drm_i915_private *i915)
 				 ARRAY_SIZE(intel_display_debugfs_list),
 				 minor->debugfs_root, minor);
 
+	intel_cdclk_debugfs_register(i915);
 	intel_dmc_debugfs_register(i915);
 	intel_fbc_debugfs_register(i915);
 	intel_hpd_debugfs_register(i915);
-- 
GitLab


From 446a20c9ba622bb531f1705eab88b64d478ee434 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 11 Apr 2023 13:56:43 +0300
Subject: [PATCH 0035/1791] drm/i915: hide mkwrite_device_info() better

The goal has been to just make device info a pointer to static const
data, i.e. the static const structs in i915_pci.c. See [1]. However,
there were issues with intel_device_info_runtime_init() clearing the
display sub-struct of device info on the !HAS_DISPLAY() path, which
consequently disables a lot of display functionality, like it
should. Looks like we'd have to cover all those paths, and maybe
sprinkle HAS_DISPLAY() checks in them, which we haven't gotten around
to.

In the mean time, hide mkwrite_device_info() better within
intel_device_info.c by adding a intel_device_info_driver_create() for
the very early initialization of the device info and initial runtime
info. This also lets us declutter i915_drv.h a bit, and stops promoting
mkwrite_device_info() as something that could be used.

[1] https://lore.kernel.org/r/a0422f0a8ac055f65b7922bcd3119b180a41e79e.1655712106.git.jani.nikula@intel.com

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411105643.292416-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_driver.c       | 12 ++--------
 drivers/gpu/drm/i915/i915_drv.h          |  7 ------
 drivers/gpu/drm/i915/intel_device_info.c | 29 ++++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_device_info.h |  2 ++
 4 files changed, 33 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index fb5fbda70ad9d..e5576c440ca0a 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -718,8 +718,6 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	const struct intel_device_info *match_info =
 		(struct intel_device_info *)ent->driver_data;
-	struct intel_device_info *device_info;
-	struct intel_runtime_info *runtime;
 	struct drm_i915_private *i915;
 
 	i915 = devm_drm_dev_alloc(&pdev->dev, &i915_drm_driver,
@@ -732,14 +730,8 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* Device parameters start as a copy of module parameters. */
 	i915_params_copy(&i915->params, &i915_modparams);
 
-	/* Setup the write-once "constant" device info */
-	device_info = mkwrite_device_info(i915);
-	memcpy(device_info, match_info, sizeof(*device_info));
-
-	/* Initialize initial runtime info from static const data and pdev. */
-	runtime = RUNTIME_INFO(i915);
-	memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime));
-	runtime->device_id = pdev->device;
+	/* Set up device info and initial runtime info. */
+	intel_device_info_driver_create(i915, pdev->device, match_info);
 
 	return i915;
 }
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index de6a3cd3f46e0..672e4fdf8804e 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -929,11 +929,4 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define HAS_LMEMBAR_SMEM_STOLEN(i915) (!HAS_LMEM(i915) && \
 				       GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70))
 
-/* intel_device_info.c */
-static inline struct intel_device_info *
-mkwrite_device_info(struct drm_i915_private *dev_priv)
-{
-	return (struct intel_device_info *)INTEL_INFO(dev_priv);
-}
-
 #endif
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index fc5cd14adfccb..4e23be2995bf6 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -381,6 +381,13 @@ void intel_device_info_runtime_init_early(struct drm_i915_private *i915)
 	intel_device_info_subplatform_init(i915);
 }
 
+/* FIXME: Remove this, and make device info a const pointer to rodata. */
+static struct intel_device_info *
+mkwrite_device_info(struct drm_i915_private *i915)
+{
+	return (struct intel_device_info *)INTEL_INFO(i915);
+}
+
 /**
  * intel_device_info_runtime_init - initialize runtime info
  * @dev_priv: the i915 device
@@ -548,6 +555,28 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 		dev_priv->drm.driver_features &= ~DRIVER_ATOMIC;
 }
 
+/*
+ * Set up device info and initial runtime info at driver create.
+ *
+ * Note: i915 is only an allocated blob of memory at this point.
+ */
+void intel_device_info_driver_create(struct drm_i915_private *i915,
+				     u16 device_id,
+				     const struct intel_device_info *match_info)
+{
+	struct intel_device_info *info;
+	struct intel_runtime_info *runtime;
+
+	/* Setup the write-once "constant" device info */
+	info = mkwrite_device_info(i915);
+	memcpy(info, match_info, sizeof(*info));
+
+	/* Initialize initial runtime info from static const data and pdev. */
+	runtime = RUNTIME_INFO(i915);
+	memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime));
+	runtime->device_id = device_id;
+}
+
 void intel_driver_caps_print(const struct intel_driver_caps *caps,
 			     struct drm_printer *p)
 {
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index dd8b17c155669..30496966be610 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -316,6 +316,8 @@ struct intel_driver_caps {
 
 const char *intel_platform_name(enum intel_platform platform);
 
+void intel_device_info_driver_create(struct drm_i915_private *i915, u16 device_id,
+				     const struct intel_device_info *match_info);
 void intel_device_info_runtime_init_early(struct drm_i915_private *dev_priv);
 void intel_device_info_runtime_init(struct drm_i915_private *dev_priv);
 
-- 
GitLab


From 7cb3eb334b8c2a06f780abcf38bffbd9efa4cec1 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Thu, 13 Apr 2023 14:24:35 -0700
Subject: [PATCH 0036/1791] drm/i915/mtl: Add DP rates

Add DP rates for Meteorlake.

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-2-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 48d43f7f0c58a..db7b6eaf8c854 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -420,6 +420,11 @@ static int ehl_max_source_rate(struct intel_dp *intel_dp)
 	return 810000;
 }
 
+static int mtl_max_source_rate(struct intel_dp *intel_dp)
+{
+	return intel_dp_is_edp(intel_dp) ? 675000 : 810000;
+}
+
 static int vbt_max_link_rate(struct intel_dp *intel_dp)
 {
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
@@ -444,6 +449,10 @@ static void
 intel_dp_set_source_rates(struct intel_dp *intel_dp)
 {
 	/* The values must be in increasing order */
+	static const int mtl_rates[] = {
+		162000, 216000, 243000, 270000, 324000, 432000, 540000, 675000,
+		810000,
+	};
 	static const int icl_rates[] = {
 		162000, 216000, 270000, 324000, 432000, 540000, 648000, 810000,
 		1000000, 1350000,
@@ -469,7 +478,11 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp)
 	drm_WARN_ON(&dev_priv->drm,
 		    intel_dp->source_rates || intel_dp->num_source_rates);
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		source_rates = mtl_rates;
+		size = ARRAY_SIZE(mtl_rates);
+		max_rate = mtl_max_source_rate(intel_dp);
+	} else if (DISPLAY_VER(dev_priv) >= 11) {
 		source_rates = icl_rates;
 		size = ARRAY_SIZE(icl_rates);
 		if (IS_DG2(dev_priv))
-- 
GitLab


From a42e65f33c38e3b0191cf6a1bc8ebb6c8289127d Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Thu, 13 Apr 2023 14:24:36 -0700
Subject: [PATCH 0037/1791] drm/i915/mtl: Create separate reg file for PICA
 registers

Create a separate file to store registers for PICA chips
C10 and C20.

v2: Rename file (Jani)
v3: Use _PICK_EVEN_2RANGES() macro (Lucas)
    Coding style fixed (Lucas)
v4: Redefine macros (Imre)

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-3-radhakrishna.sripada@intel.com
---
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h | 133 ++++++++++++++++++
 1 file changed, 133 insertions(+)
 create mode 100644 drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
new file mode 100644
index 0000000000000..916002aad7463
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -0,0 +1,133 @@
+/* SPDX-License-Identifier: MIT
+ *
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_CX0_PHY_REGS_H__
+#define __INTEL_CX0_PHY_REGS_H__
+
+#include "i915_reg_defs.h"
+
+#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A		0x64040
+#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B		0x64140
+#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1		0x16F240
+#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2		0x16F440
+#define XELPDP_PORT_M2P_MSGBUS_CTL(port, lane)		_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4)
+#define   XELPDP_PORT_M2P_TRANSACTION_PENDING		REG_BIT(31)
+#define   XELPDP_PORT_M2P_COMMAND_TYPE_MASK		REG_GENMASK(30, 27)
+#define   XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED	REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x1)
+#define   XELPDP_PORT_M2P_COMMAND_WRITE_COMMITTED	REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x2)
+#define   XELPDP_PORT_M2P_COMMAND_READ			REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x3)
+#define   XELPDP_PORT_M2P_DATA_MASK			REG_GENMASK(23, 16)
+#define   XELPDP_PORT_M2P_DATA(val)			REG_FIELD_PREP(XELPDP_PORT_M2P_DATA_MASK, val)
+#define   XELPDP_PORT_M2P_TRANSACTION_RESET		REG_BIT(15)
+#define   XELPDP_PORT_M2P_ADDRESS_MASK			REG_GENMASK(11, 0)
+#define   XELPDP_PORT_M2P_ADDRESS(val)			REG_FIELD_PREP(XELPDP_PORT_M2P_ADDRESS_MASK, val)
+#define XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane)	_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \
+										 _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4 + 8)
+#define   XELPDP_PORT_P2M_RESPONSE_READY		REG_BIT(31)
+#define   XELPDP_PORT_P2M_COMMAND_TYPE_MASK		REG_GENMASK(30, 27)
+#define   XELPDP_PORT_P2M_COMMAND_READ_ACK		0x4
+#define   XELPDP_PORT_P2M_COMMAND_WRITE_ACK		0x5
+#define   XELPDP_PORT_P2M_DATA_MASK			REG_GENMASK(23, 16)
+#define   XELPDP_PORT_P2M_DATA(val)			REG_FIELD_PREP(XELPDP_PORT_P2M_DATA_MASK, val)
+#define   XELPDP_PORT_P2M_ERROR_SET			REG_BIT(15)
+
+#define XELPDP_MSGBUS_TIMEOUT_SLOW			1
+#define XELPDP_MSGBUS_TIMEOUT_FAST_US			2
+#define XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US		3200
+#define XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US		20
+#define XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US		100
+#define XELPDP_PORT_RESET_START_TIMEOUT_US		5
+#define XELPDP_PORT_POWERDOWN_UPDATE_TIMEOUT_US		100
+#define XELPDP_PORT_RESET_END_TIMEOUT			15
+#define XELPDP_REFCLK_ENABLE_TIMEOUT_US			1
+
+#define _XELPDP_PORT_BUF_CTL1_LN0_A			0x64004
+#define _XELPDP_PORT_BUF_CTL1_LN0_B			0x64104
+#define _XELPDP_PORT_BUF_CTL1_LN0_USBC1			0x16F200
+#define _XELPDP_PORT_BUF_CTL1_LN0_USBC2			0x16F400
+#define XELPDP_PORT_BUF_CTL1(port)			_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_A, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_B, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC2))
+#define   XELPDP_PORT_BUF_SOC_PHY_READY			REG_BIT(24)
+#define   XELPDP_PORT_REVERSAL				REG_BIT(16)
+#define   XELPDP_TC_PHY_OWNERSHIP			REG_BIT(6)
+#define   XELPDP_TCSS_POWER_REQUEST			REG_BIT(5)
+#define   XELPDP_TCSS_POWER_STATE			REG_BIT(4)
+#define   XELPDP_PORT_WIDTH_MASK			REG_GENMASK(3, 1)
+#define   XELPDP_PORT_WIDTH(val)			REG_FIELD_PREP(XELPDP_PORT_WIDTH_MASK, val)
+
+#define XELPDP_PORT_BUF_CTL2(port)			_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_A, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_B, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 4)
+
+#define   XELPDP_LANE_PIPE_RESET(lane)			_PICK(lane, REG_BIT(31), REG_BIT(30))
+#define   XELPDP_LANE_PHY_CURRENT_STATUS(lane)		_PICK(lane, REG_BIT(29), REG_BIT(28))
+#define   XELPDP_LANE_POWERDOWN_UPDATE(lane)		_PICK(lane, REG_BIT(25), REG_BIT(24))
+#define   _XELPDP_LANE0_POWERDOWN_NEW_STATE_MASK	REG_GENMASK(23, 20)
+#define   _XELPDP_LANE0_POWERDOWN_NEW_STATE(val)	REG_FIELD_PREP(_XELPDP_LANE0_POWERDOWN_NEW_STATE_MASK, val)
+#define   _XELPDP_LANE1_POWERDOWN_NEW_STATE_MASK	REG_GENMASK(19, 16)
+#define   _XELPDP_LANE1_POWERDOWN_NEW_STATE(val)	REG_FIELD_PREP(_XELPDP_LANE1_POWERDOWN_NEW_STATE_MASK, val)
+#define   XELPDP_LANE_POWERDOWN_NEW_STATE(lane, val)	_PICK(lane, \
+							      _XELPDP_LANE0_POWERDOWN_NEW_STATE(val), \
+							      _XELPDP_LANE1_POWERDOWN_NEW_STATE(val))
+#define   XELPDP_LANE_POWERDOWN_NEW_STATE_MASK		REG_GENMASK(3, 0)
+#define   XELPDP_POWER_STATE_READY_MASK			REG_GENMASK(7, 4)
+#define   XELPDP_POWER_STATE_READY(val)			REG_FIELD_PREP(XELPDP_POWER_STATE_READY_MASK, val)
+
+#define XELPDP_PORT_BUF_CTL3(port)			_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_A, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_B, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \
+										 _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 8)
+#define   XELPDP_PLL_LANE_STAGGERING_DELAY_MASK		REG_GENMASK(15, 8)
+#define   XELPDP_PLL_LANE_STAGGERING_DELAY(val)		REG_FIELD_PREP(XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, val)
+#define   XELPDP_POWER_STATE_ACTIVE_MASK		REG_GENMASK(3, 0)
+#define   XELPDP_POWER_STATE_ACTIVE(val)		REG_FIELD_PREP(XELPDP_POWER_STATE_ACTIVE_MASK, val)
+
+#define _XELPDP_PORT_CLOCK_CTL_A			0x640E0
+#define _XELPDP_PORT_CLOCK_CTL_B			0x641E0
+#define _XELPDP_PORT_CLOCK_CTL_USBC1			0x16F260
+#define _XELPDP_PORT_CLOCK_CTL_USBC2			0x16F460
+#define XELPDP_PORT_CLOCK_CTL(port)			_MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \
+										 _XELPDP_PORT_CLOCK_CTL_A, \
+										 _XELPDP_PORT_CLOCK_CTL_B, \
+										 _XELPDP_PORT_CLOCK_CTL_USBC1, \
+										 _XELPDP_PORT_CLOCK_CTL_USBC2))
+#define   XELPDP_LANE0_PCLK_PLL_REQUEST			REG_BIT(31)
+#define   XELPDP_LANE0_PCLK_PLL_ACK			REG_BIT(30)
+#define   XELPDP_LANE0_PCLK_REFCLK_REQUEST		REG_BIT(29)
+#define   XELPDP_LANE0_PCLK_REFCLK_ACK			REG_BIT(28)
+#define   XELPDP_LANE1_PCLK_PLL_REQUEST			REG_BIT(27)
+#define   XELPDP_LANE1_PCLK_PLL_ACK			REG_BIT(26)
+#define   XELPDP_LANE1_PCLK_REFCLK_REQUEST		REG_BIT(25)
+#define   XELPDP_LANE1_PCLK_REFCLK_ACK			REG_BIT(24)
+#define   XELPDP_TBT_CLOCK_REQUEST			REG_BIT(19)
+#define   XELPDP_TBT_CLOCK_ACK				REG_BIT(18)
+#define   XELPDP_DDI_CLOCK_SELECT_MASK			REG_GENMASK(15, 12)
+#define   XELPDP_DDI_CLOCK_SELECT(val)			REG_FIELD_PREP(XELPDP_DDI_CLOCK_SELECT_MASK, val)
+#define   XELPDP_DDI_CLOCK_SELECT_NONE			0x0
+#define   XELPDP_DDI_CLOCK_SELECT_MAXPCLK		0x8
+#define   XELPDP_DDI_CLOCK_SELECT_DIV18CLK		0x9
+#define   XELPDP_DDI_CLOCK_SELECT_TBT_162		0xc
+#define   XELPDP_DDI_CLOCK_SELECT_TBT_270		0xd
+#define   XELPDP_DDI_CLOCK_SELECT_TBT_540		0xe
+#define   XELPDP_DDI_CLOCK_SELECT_TBT_810		0xf
+#define   XELPDP_FORWARD_CLOCK_UNGATE			REG_BIT(10)
+#define   XELPDP_LANE1_PHY_CLOCK_SELECT			REG_BIT(8)
+#define   XELPDP_SSC_ENABLE_PLLA			REG_BIT(1)
+#define   XELPDP_SSC_ENABLE_PLLB			REG_BIT(0)
+
+#endif /* __INTEL_CX0_PHY_REGS_H__ */
-- 
GitLab


From 51390cc0e00a378b7c152bb6f63efc0a01b59d20 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Thu, 13 Apr 2023 14:24:37 -0700
Subject: [PATCH 0038/1791] drm/i915/mtl: Add Support for C10 PHY message bus
 and pll programming

XELPDP has C10 and C20 phys from Synopsys to drive displays. Each phy
has a dedicated PIPE 5.2 Message bus for configuration. This message
bus is used to configure the phy internal registers.

XELPDP has C10 phys to drive output to the EDP and the native output
from the display engine. Add structures, programming hardware state
readout logic. Port clock calculations are similar to DG2. Use the DG2
formulae to calculate the port clock but use the relevant pll signals.
Note: PHY lane 0 is always used for PLL programming.

Add sequences for C10 phy enable/disable phy lane reset,
powerdown change sequence and phy lane programming.

Bspec: 64539, 64568, 64599, 65100, 65101, 65450, 65451, 67610, 67636

v2: Squash patches related to C10 phy message bus and pll
    programming support (Jani)
    Move register definitions to a new file i.e. intel_cx0_reg_defs.h (Jani)
    Move macro definitions (Jani)
    DP rates as separate patch (Jani)
    Spin out xelpdp register definitions into a separate file (Jani)
    Replace macro to select registers based on phy lane with
    function calls (Jani)
    Fix styling issues (Jani)
    Call XELPDP_PORT_P2M_MSGBUS_STATUS() with port instead of phy (Lucas)
v3: Move clear request flag into try-loop
v4: On PHY idle change drm_err_once() as drm_dbg_kms() (Jani)
    use __intel_de_wait_for_register() instead of __intel_wait_for_register
    and uncomment intel_uncore.h (Jani)
    Add DP-alt support for PHY lane programming (Khaled)
v4: Add tx and cmn on c10mpllb_state (Imre)
    Add missing waits for pending transactions between two message bus
    writes (Imre)
    General cleanups and simplifications (Imre)
v5: Few nit cleanups from rev4 (imre)
    s/dev_priv/i915/ , s/c10mpllb/c10pll/ (RK)
    Rebase
v6: Move the mtl code from intel_c10pll_calc_port_clock to mtl function
    Fix typo in comment for REG_FIELD_PREP8 definition(Imre)

Cc: Mika Kahola <mika.kahola@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com> (v4)
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-4-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |    1 +
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 1207 +++++++++++++++++
 drivers/gpu/drm/i915/display/intel_cx0_phy.h  |   34 +
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |   47 +-
 drivers/gpu/drm/i915/display/intel_ddi.c      |   22 +-
 .../drm/i915/display/intel_display_types.h    |   13 +
 drivers/gpu/drm/i915/display/intel_dpll.c     |   33 +-
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c |    2 +-
 .../drm/i915/display/intel_modeset_verify.c   |    2 +
 drivers/gpu/drm/i915/i915_reg.h               |    5 +
 drivers/gpu/drm/i915/i915_reg_defs.h          |   57 +
 11 files changed, 1411 insertions(+), 12 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_cx0_phy.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_cx0_phy.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 97b0d4ae221ac..4ee3b5850dd06 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -298,6 +298,7 @@ i915-y += \
 	display/icl_dsi.o \
 	display/intel_backlight.o \
 	display/intel_crt.o \
+	display/intel_cx0_phy.o \
 	display/intel_ddi.o \
 	display/intel_ddi_buf_trans.o \
 	display/intel_display_trace.o \
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
new file mode 100644
index 0000000000000..9ab1e686a40b6
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -0,0 +1,1207 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include "i915_reg.h"
+#include "intel_cx0_phy.h"
+#include "intel_cx0_phy_regs.h"
+#include "intel_de.h"
+#include "intel_display_types.h"
+#include "intel_dp.h"
+#include "intel_panel.h"
+#include "intel_psr.h"
+#include "intel_tc.h"
+
+#define MB_WRITE_COMMITTED      true
+#define MB_WRITE_UNCOMMITTED    false
+
+#define for_each_cx0_lane_in_mask(__lane_mask, __lane) \
+	for ((__lane) = 0; (__lane) < 2; (__lane)++) \
+		for_each_if((__lane_mask) & BIT(__lane))
+
+#define INTEL_CX0_LANE0		BIT(0)
+#define INTEL_CX0_LANE1		BIT(1)
+#define INTEL_CX0_BOTH_LANES	(INTEL_CX0_LANE1 | INTEL_CX0_LANE0)
+
+bool intel_is_c10phy(struct drm_i915_private *i915, enum phy phy)
+{
+	if (IS_METEORLAKE(i915) && (phy < PHY_C))
+		return true;
+
+	return false;
+}
+
+static int lane_mask_to_lane(u8 lane_mask)
+{
+	if (WARN_ON((lane_mask & ~INTEL_CX0_BOTH_LANES) ||
+		    hweight8(lane_mask) != 1))
+		return 0;
+
+	return ilog2(lane_mask);
+}
+
+static void
+assert_dc_off(struct drm_i915_private *i915)
+{
+	bool enabled;
+
+	enabled = intel_display_power_is_enabled(i915, POWER_DOMAIN_DC_OFF);
+	drm_WARN_ON(&i915->drm, !enabled);
+}
+
+/*
+ * Prepare HW for CX0 phy transactions.
+ *
+ * It is required that PSR and DC5/6 are disabled before any CX0 message
+ * bus transaction is executed.
+ */
+static intel_wakeref_t intel_cx0_phy_transaction_begin(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+	intel_psr_pause(intel_dp);
+	return intel_display_power_get(i915, POWER_DOMAIN_DC_OFF);
+}
+
+static void intel_cx0_phy_transaction_end(struct intel_encoder *encoder, intel_wakeref_t wakeref)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+	intel_psr_resume(intel_dp);
+	intel_display_power_put(i915, POWER_DOMAIN_DC_OFF, wakeref);
+}
+
+static void intel_clear_response_ready_flag(struct drm_i915_private *i915,
+					    enum port port, int lane)
+{
+	intel_de_rmw(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane),
+		     0, XELPDP_PORT_P2M_RESPONSE_READY | XELPDP_PORT_P2M_ERROR_SET);
+}
+
+static void intel_cx0_bus_reset(struct drm_i915_private *i915, enum port port, int lane)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+
+	intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+		       XELPDP_PORT_M2P_TRANSACTION_RESET);
+
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+				    XELPDP_PORT_M2P_TRANSACTION_RESET,
+				    XELPDP_MSGBUS_TIMEOUT_SLOW)) {
+		drm_err_once(&i915->drm, "Failed to bring PHY %c to idle.\n", phy_name(phy));
+		return;
+	}
+
+	intel_clear_response_ready_flag(i915, port, lane);
+}
+
+static int intel_cx0_wait_for_ack(struct drm_i915_private *i915, enum port port,
+				  int command, int lane, u32 *val)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+
+	if (__intel_de_wait_for_register(i915,
+					 XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane),
+					 XELPDP_PORT_P2M_RESPONSE_READY,
+					 XELPDP_PORT_P2M_RESPONSE_READY,
+					 XELPDP_MSGBUS_TIMEOUT_FAST_US,
+					 XELPDP_MSGBUS_TIMEOUT_SLOW, val)) {
+		drm_dbg_kms(&i915->drm, "PHY %c Timeout waiting for message ACK. Status: 0x%x\n",
+			    phy_name(phy), *val);
+		return -ETIMEDOUT;
+	}
+
+	if (*val & XELPDP_PORT_P2M_ERROR_SET) {
+		drm_dbg_kms(&i915->drm, "PHY %c Error occurred during %s command. Status: 0x%x\n", phy_name(phy),
+			    command == XELPDP_PORT_P2M_COMMAND_READ_ACK ? "read" : "write", *val);
+		intel_cx0_bus_reset(i915, port, lane);
+		return -EINVAL;
+	}
+
+	if (REG_FIELD_GET(XELPDP_PORT_P2M_COMMAND_TYPE_MASK, *val) != command) {
+		drm_dbg_kms(&i915->drm, "PHY %c Not a %s response. MSGBUS Status: 0x%x.\n", phy_name(phy),
+			    command == XELPDP_PORT_P2M_COMMAND_READ_ACK ? "read" : "write", *val);
+		intel_cx0_bus_reset(i915, port, lane);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __intel_cx0_read_once(struct drm_i915_private *i915, enum port port,
+				 int lane, u16 addr)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	int ack;
+	u32 val;
+
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+				    XELPDP_PORT_M2P_TRANSACTION_PENDING,
+				    XELPDP_MSGBUS_TIMEOUT_SLOW)) {
+		drm_dbg_kms(&i915->drm,
+			    "PHY %c Timeout waiting for previous transaction to complete. Reset the bus and retry.\n", phy_name(phy));
+		intel_cx0_bus_reset(i915, port, lane);
+		return -ETIMEDOUT;
+	}
+
+	intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+		       XELPDP_PORT_M2P_TRANSACTION_PENDING |
+		       XELPDP_PORT_M2P_COMMAND_READ |
+		       XELPDP_PORT_M2P_ADDRESS(addr));
+
+	ack = intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_READ_ACK, lane, &val);
+	if (ack < 0) {
+		intel_cx0_bus_reset(i915, port, lane);
+		return ack;
+	}
+
+	intel_clear_response_ready_flag(i915, port, lane);
+
+	return REG_FIELD_GET(XELPDP_PORT_P2M_DATA_MASK, val);
+}
+
+static u8 __intel_cx0_read(struct drm_i915_private *i915, enum port port,
+			   int lane, u16 addr)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	int i, status;
+
+	assert_dc_off(i915);
+
+	/* 3 tries is assumed to be enough to read successfully */
+	for (i = 0; i < 3; i++) {
+		status = __intel_cx0_read_once(i915, port, lane, addr);
+
+		if (status >= 0)
+			return status;
+	}
+
+	drm_err_once(&i915->drm, "PHY %c Read %04x failed after %d retries.\n",
+		     phy_name(phy), addr, i);
+
+	return 0;
+}
+
+static u8 intel_cx0_read(struct drm_i915_private *i915, enum port port,
+			 u8 lane_mask, u16 addr)
+{
+	int lane = lane_mask_to_lane(lane_mask);
+
+	return __intel_cx0_read(i915, port, lane, addr);
+}
+
+static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port,
+				  int lane, u16 addr, u8 data, bool committed)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	u32 val;
+
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+				    XELPDP_PORT_M2P_TRANSACTION_PENDING,
+				    XELPDP_MSGBUS_TIMEOUT_SLOW)) {
+		drm_dbg_kms(&i915->drm,
+			    "PHY %c Timeout waiting for previous transaction to complete. Resetting the bus.\n", phy_name(phy));
+		intel_cx0_bus_reset(i915, port, lane);
+		return -ETIMEDOUT;
+	}
+
+	intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+		       XELPDP_PORT_M2P_TRANSACTION_PENDING |
+		       (committed ? XELPDP_PORT_M2P_COMMAND_WRITE_COMMITTED :
+				    XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED) |
+		       XELPDP_PORT_M2P_DATA(data) |
+		       XELPDP_PORT_M2P_ADDRESS(addr));
+
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+				    XELPDP_PORT_M2P_TRANSACTION_PENDING,
+				    XELPDP_MSGBUS_TIMEOUT_SLOW)) {
+		drm_dbg_kms(&i915->drm,
+			    "PHY %c Timeout waiting for write to complete. Resetting the bus.\n", phy_name(phy));
+		intel_cx0_bus_reset(i915, port, lane);
+		return -ETIMEDOUT;
+	}
+
+	if (committed) {
+		if (intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_WRITE_ACK, lane, &val) < 0) {
+			intel_cx0_bus_reset(i915, port, lane);
+			return -EINVAL;
+		}
+	} else if ((intel_de_read(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane)) &
+		    XELPDP_PORT_P2M_ERROR_SET)) {
+		drm_dbg_kms(&i915->drm,
+			    "PHY %c Error occurred during write command.\n", phy_name(phy));
+		intel_cx0_bus_reset(i915, port, lane);
+		return -EINVAL;
+	}
+
+	intel_clear_response_ready_flag(i915, port, lane);
+
+	return 0;
+}
+
+static void __intel_cx0_write(struct drm_i915_private *i915, enum port port,
+			      int lane, u16 addr, u8 data, bool committed)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	int i, status;
+
+	assert_dc_off(i915);
+
+	/* 3 tries is assumed to be enough to write successfully */
+	for (i = 0; i < 3; i++) {
+		status = __intel_cx0_write_once(i915, port, lane, addr, data, committed);
+
+		if (status == 0)
+			return;
+	}
+
+	drm_err_once(&i915->drm,
+		     "PHY %c Write %04x failed after %d retries.\n", phy_name(phy), addr, i);
+}
+
+static void intel_cx0_write(struct drm_i915_private *i915, enum port port,
+			    u8 lane_mask, u16 addr, u8 data, bool committed)
+{
+	int lane;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		__intel_cx0_write(i915, port, lane, addr, data, committed);
+}
+
+static void __intel_cx0_rmw(struct drm_i915_private *i915, enum port port,
+			    int lane, u16 addr, u8 clear, u8 set, bool committed)
+{
+	u8 old, val;
+
+	old = __intel_cx0_read(i915, port, lane, addr);
+	val = (old & ~clear) | set;
+
+	if (val != old)
+		__intel_cx0_write(i915, port, lane, addr, val, committed);
+}
+
+static void intel_cx0_rmw(struct drm_i915_private *i915, enum port port,
+			  u8 lane_mask, u16 addr, u8 clear, u8 set, bool committed)
+{
+	u8 lane;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		__intel_cx0_rmw(i915, port, lane, addr, clear, set, committed);
+}
+
+/*
+ * Basic DP link rates with 38.4 MHz reference clock.
+ * Note: The tables below are with SSC. In non-ssc
+ * registers 0xC04 to 0xC08(pll[4] to pll[8]) will be
+ * programmed 0.
+ */
+
+static const struct intel_c10pll_state mtl_c10_dp_rbr = {
+	.clock = 162000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0xB4,
+	.pll[1] = 0,
+	.pll[2] = 0x30,
+	.pll[3] = 0x1,
+	.pll[4] = 0x26,
+	.pll[5] = 0x0C,
+	.pll[6] = 0x98,
+	.pll[7] = 0x46,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xC0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x2,
+	.pll[16] = 0x84,
+	.pll[17] = 0x4F,
+	.pll[18] = 0xE5,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_edp_r216 = {
+	.clock = 216000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0x4,
+	.pll[1] = 0,
+	.pll[2] = 0xA2,
+	.pll[3] = 0x1,
+	.pll[4] = 0x33,
+	.pll[5] = 0x10,
+	.pll[6] = 0x75,
+	.pll[7] = 0xB3,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x2,
+	.pll[16] = 0x85,
+	.pll[17] = 0x0F,
+	.pll[18] = 0xE6,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_edp_r243 = {
+	.clock = 243000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0x34,
+	.pll[1] = 0,
+	.pll[2] = 0xDA,
+	.pll[3] = 0x1,
+	.pll[4] = 0x39,
+	.pll[5] = 0x12,
+	.pll[6] = 0xE3,
+	.pll[7] = 0xE9,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0x20,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x2,
+	.pll[16] = 0x85,
+	.pll[17] = 0x8F,
+	.pll[18] = 0xE6,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_dp_hbr1 = {
+	.clock = 270000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0xF4,
+	.pll[1] = 0,
+	.pll[2] = 0xF8,
+	.pll[3] = 0x0,
+	.pll[4] = 0x20,
+	.pll[5] = 0x0A,
+	.pll[6] = 0x29,
+	.pll[7] = 0x10,
+	.pll[8] = 0x1,   /* Verify */
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xA0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x1,
+	.pll[16] = 0x84,
+	.pll[17] = 0x4F,
+	.pll[18] = 0xE5,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_edp_r324 = {
+	.clock = 324000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0xB4,
+	.pll[1] = 0,
+	.pll[2] = 0x30,
+	.pll[3] = 0x1,
+	.pll[4] = 0x26,
+	.pll[5] = 0x0C,
+	.pll[6] = 0x98,
+	.pll[7] = 0x46,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xC0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x1,
+	.pll[16] = 0x85,
+	.pll[17] = 0x4F,
+	.pll[18] = 0xE6,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_edp_r432 = {
+	.clock = 432000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0x4,
+	.pll[1] = 0,
+	.pll[2] = 0xA2,
+	.pll[3] = 0x1,
+	.pll[4] = 0x33,
+	.pll[5] = 0x10,
+	.pll[6] = 0x75,
+	.pll[7] = 0xB3,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x1,
+	.pll[16] = 0x85,
+	.pll[17] = 0x0F,
+	.pll[18] = 0xE6,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_dp_hbr2 = {
+	.clock = 540000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0xF4,
+	.pll[1] = 0,
+	.pll[2] = 0xF8,
+	.pll[3] = 0,
+	.pll[4] = 0x20,
+	.pll[5] = 0x0A,
+	.pll[6] = 0x29,
+	.pll[7] = 0x10,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xA0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0,
+	.pll[16] = 0x84,
+	.pll[17] = 0x4F,
+	.pll[18] = 0xE5,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_edp_r675 = {
+	.clock = 675000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0xB4,
+	.pll[1] = 0,
+	.pll[2] = 0x3E,
+	.pll[3] = 0x1,
+	.pll[4] = 0xA8,
+	.pll[5] = 0x0C,
+	.pll[6] = 0x33,
+	.pll[7] = 0x54,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xC8,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0,
+	.pll[16] = 0x85,
+	.pll[17] = 0x8F,
+	.pll[18] = 0xE6,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_dp_hbr3 = {
+	.clock = 810000,
+	.tx = 0x10,
+	.cmn = 0x21,
+	.pll[0] = 0x34,
+	.pll[1] = 0,
+	.pll[2] = 0x84,
+	.pll[3] = 0x1,
+	.pll[4] = 0x30,
+	.pll[5] = 0x0F,
+	.pll[6] = 0x3D,
+	.pll[7] = 0x98,
+	.pll[8] = 0x1,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0xF0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0,
+	.pll[16] = 0x84,
+	.pll[17] = 0x0F,
+	.pll[18] = 0xE5,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state * const mtl_c10_dp_tables[] = {
+	&mtl_c10_dp_rbr,
+	&mtl_c10_dp_hbr1,
+	&mtl_c10_dp_hbr2,
+	&mtl_c10_dp_hbr3,
+	NULL,
+};
+
+static const struct intel_c10pll_state * const mtl_c10_edp_tables[] = {
+	&mtl_c10_dp_rbr,
+	&mtl_c10_edp_r216,
+	&mtl_c10_edp_r243,
+	&mtl_c10_dp_hbr1,
+	&mtl_c10_edp_r324,
+	&mtl_c10_edp_r432,
+	&mtl_c10_dp_hbr2,
+	&mtl_c10_edp_r675,
+	&mtl_c10_dp_hbr3,
+	NULL,
+};
+
+static const struct intel_c10pll_state * const *
+intel_c10pll_tables_get(struct intel_crtc_state *crtc_state,
+			struct intel_encoder *encoder)
+{
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
+		if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP))
+			return mtl_c10_edp_tables;
+		else
+			return mtl_c10_dp_tables;
+	}
+
+	/* TODO: Add HDMI Support */
+	MISSING_CASE(encoder->type);
+	return NULL;
+}
+
+static void intel_c10pll_update_pll(struct intel_crtc_state *crtc_state,
+				    struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_cx0pll_state *pll_state = &crtc_state->cx0pll_state;
+	int i;
+
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
+		if (intel_panel_use_ssc(i915)) {
+			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+			pll_state->ssc_enabled =
+				(intel_dp->dpcd[DP_MAX_DOWNSPREAD] & DP_MAX_DOWNSPREAD_0_5);
+		}
+	}
+
+	if (pll_state->ssc_enabled)
+		return;
+
+	drm_WARN_ON(&i915->drm, ARRAY_SIZE(pll_state->c10.pll) < 9);
+	for (i = 4; i < 9; i++)
+		pll_state->c10.pll[i] = 0;
+}
+
+static int intel_c10pll_calc_state(struct intel_crtc_state *crtc_state,
+				   struct intel_encoder *encoder)
+{
+	const struct intel_c10pll_state * const *tables;
+	int i;
+
+	tables = intel_c10pll_tables_get(crtc_state, encoder);
+	if (!tables)
+		return -EINVAL;
+
+	for (i = 0; tables[i]; i++) {
+		if (crtc_state->port_clock == tables[i]->clock) {
+			crtc_state->cx0pll_state.c10 = *tables[i];
+			intel_c10pll_update_pll(crtc_state, encoder);
+
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state,
+			    struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
+
+	return intel_c10pll_calc_state(crtc_state, encoder);
+}
+
+void intel_c10pll_readout_hw_state(struct intel_encoder *encoder,
+				   struct intel_c10pll_state *pll_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	u8 lane = INTEL_CX0_LANE0;
+	intel_wakeref_t wakeref;
+	int i;
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
+	/*
+	 * According to C10 VDR Register programming Sequence we need
+	 * to do this to read PHY internal registers from MsgBus.
+	 */
+	intel_cx0_rmw(i915, encoder->port, lane, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_MSGBUS_ACCESS,
+		      MB_WRITE_COMMITTED);
+
+	for (i = 0; i < ARRAY_SIZE(pll_state->pll); i++)
+		pll_state->pll[i] = intel_cx0_read(i915, encoder->port, lane,
+						   PHY_C10_VDR_PLL(i));
+
+	pll_state->cmn = intel_cx0_read(i915, encoder->port, lane, PHY_C10_VDR_CMN(0));
+	pll_state->tx = intel_cx0_read(i915, encoder->port, lane, PHY_C10_VDR_TX(0));
+
+	intel_cx0_phy_transaction_end(encoder, wakeref);
+}
+
+static void intel_c10_pll_program(struct drm_i915_private *i915,
+				  const struct intel_crtc_state *crtc_state,
+				  struct intel_encoder *encoder)
+{
+	const struct intel_c10pll_state *pll_state = &crtc_state->cx0pll_state.c10;
+	int i;
+
+	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_MSGBUS_ACCESS,
+		      MB_WRITE_COMMITTED);
+	/* Custom width needs to be programmed to 0 for both the phy lanes */
+	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CUSTOM_WIDTH,
+		      C10_VDR_CUSTOM_WIDTH_MASK, C10_VDR_CUSTOM_WIDTH_8_10,
+		      MB_WRITE_COMMITTED);
+	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_UPDATE_CFG,
+		      MB_WRITE_COMMITTED);
+
+	/* Program the pll values only for the master lane */
+	for (i = 0; i < ARRAY_SIZE(pll_state->pll); i++)
+		intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_PLL(i),
+				pll_state->pll[i],
+				(i % 4) ? MB_WRITE_UNCOMMITTED : MB_WRITE_COMMITTED);
+
+	intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_CMN(0), pll_state->cmn, MB_WRITE_COMMITTED);
+	intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_TX(0), pll_state->tx, MB_WRITE_COMMITTED);
+
+	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_MASTER_LANE | C10_VDR_CTRL_UPDATE_CFG,
+		      MB_WRITE_COMMITTED);
+}
+
+void intel_c10pll_dump_hw_state(struct drm_i915_private *i915,
+				const struct intel_c10pll_state *hw_state)
+{
+	bool fracen;
+	int i;
+	unsigned int frac_quot = 0, frac_rem = 0, frac_den = 1;
+	unsigned int multiplier, tx_clk_div;
+
+	fracen = hw_state->pll[0] & C10_PLL0_FRACEN;
+	drm_dbg_kms(&i915->drm, "c10pll_hw_state: fracen: %s, ",
+		    str_yes_no(fracen));
+
+	if (fracen) {
+		frac_quot = hw_state->pll[12] << 8 | hw_state->pll[11];
+		frac_rem =  hw_state->pll[14] << 8 | hw_state->pll[13];
+		frac_den =  hw_state->pll[10] << 8 | hw_state->pll[9];
+		drm_dbg_kms(&i915->drm, "quot: %u, rem: %u, den: %u,\n",
+			    frac_quot, frac_rem, frac_den);
+	}
+
+	multiplier = (REG_FIELD_GET8(C10_PLL3_MULTIPLIERH_MASK, hw_state->pll[3]) << 8 |
+		      hw_state->pll[2]) / 2 + 16;
+	tx_clk_div = REG_FIELD_GET8(C10_PLL15_TXCLKDIV_MASK, hw_state->pll[15]);
+	drm_dbg_kms(&i915->drm,
+		    "multiplier: %u, tx_clk_div: %u.\n", multiplier, tx_clk_div);
+
+	drm_dbg_kms(&i915->drm, "c10pll_rawhw_state:");
+	drm_dbg_kms(&i915->drm, "tx: 0x%x, cmn: 0x%x\n", hw_state->tx, hw_state->cmn);
+
+	BUILD_BUG_ON(ARRAY_SIZE(hw_state->pll) % 4);
+	for (i = 0; i < ARRAY_SIZE(hw_state->pll); i = i + 4)
+		drm_dbg_kms(&i915->drm, "pll[%d] = 0x%x, pll[%d] = 0x%x, pll[%d] = 0x%x, pll[%d] = 0x%x\n",
+			    i, hw_state->pll[i], i + 1, hw_state->pll[i + 1],
+			    i + 2, hw_state->pll[i + 2], i + 3, hw_state->pll[i + 3]);
+}
+
+int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
+				 const struct intel_c10pll_state *pll_state)
+{
+	unsigned int frac_quot = 0, frac_rem = 0, frac_den = 1;
+	unsigned int multiplier, tx_clk_div, refclk = 38400;
+
+	if (pll_state->pll[0] & C10_PLL0_FRACEN) {
+		frac_quot = pll_state->pll[12] << 8 | pll_state->pll[11];
+		frac_rem =  pll_state->pll[14] << 8 | pll_state->pll[13];
+		frac_den =  pll_state->pll[10] << 8 | pll_state->pll[9];
+	}
+
+	multiplier = (REG_FIELD_GET8(C10_PLL3_MULTIPLIERH_MASK, pll_state->pll[3]) << 8 |
+		      pll_state->pll[2]) / 2 + 16;
+
+	tx_clk_div = REG_FIELD_GET8(C10_PLL15_TXCLKDIV_MASK, pll_state->pll[15]);
+
+	return DIV_ROUND_CLOSEST_ULL(mul_u32_u32(refclk, (multiplier << 16) + frac_quot) +
+				     DIV_ROUND_CLOSEST(refclk * frac_rem, frac_den),
+				     10 << (tx_clk_div + 16));
+}
+
+static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
+					 const struct intel_crtc_state *crtc_state,
+					 bool lane_reversal)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	u32 val = 0;
+
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port), XELPDP_PORT_REVERSAL,
+		     lane_reversal ? XELPDP_PORT_REVERSAL : 0);
+
+	if (lane_reversal)
+		val |= XELPDP_LANE1_PHY_CLOCK_SELECT;
+
+	val |= XELPDP_FORWARD_CLOCK_UNGATE;
+	val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_MAXPCLK);
+
+	/* TODO: HDMI FRL */
+	/* TODO: DP2.0 10G and 20G rates enable MPLLA*/
+	val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0;
+
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     XELPDP_LANE1_PHY_CLOCK_SELECT |
+		     XELPDP_FORWARD_CLOCK_UNGATE |
+		     XELPDP_DDI_CLOCK_SELECT_MASK |
+		     XELPDP_SSC_ENABLE_PLLB, val);
+}
+
+static u32 intel_cx0_get_powerdown_update(u8 lane_mask)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_POWERDOWN_UPDATE(lane);
+
+	return val;
+}
+
+static u32 intel_cx0_get_powerdown_state(u8 lane_mask, u8 state)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_POWERDOWN_NEW_STATE(lane, state);
+
+	return val;
+}
+
+static void intel_cx0_powerdown_change_sequence(struct drm_i915_private *i915,
+						enum port port,
+						u8 lane_mask, u8 state)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	int lane;
+
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
+		     intel_cx0_get_powerdown_state(INTEL_CX0_BOTH_LANES, XELPDP_LANE_POWERDOWN_NEW_STATE_MASK),
+		     intel_cx0_get_powerdown_state(lane_mask, state));
+
+	/* Wait for pending transactions.*/
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane),
+					    XELPDP_PORT_M2P_TRANSACTION_PENDING,
+					    XELPDP_MSGBUS_TIMEOUT_SLOW)) {
+			drm_dbg_kms(&i915->drm,
+				    "PHY %c Timeout waiting for previous transaction to complete. Reset the bus.\n",
+				    phy_name(phy));
+			intel_cx0_bus_reset(i915, port, lane);
+		}
+
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
+		     intel_cx0_get_powerdown_update(INTEL_CX0_BOTH_LANES),
+		     intel_cx0_get_powerdown_update(lane_mask));
+
+	/* Update Timeout Value */
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port),
+					 intel_cx0_get_powerdown_update(lane_mask), 0,
+					 XELPDP_PORT_POWERDOWN_UPDATE_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n",
+			 phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US);
+}
+
+static void intel_cx0_setup_powerdown(struct drm_i915_private *i915, enum port port)
+{
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
+		     XELPDP_POWER_STATE_READY_MASK,
+		     XELPDP_POWER_STATE_READY(CX0_P2_STATE_READY));
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL3(port),
+		     XELPDP_POWER_STATE_ACTIVE_MASK |
+		     XELPDP_PLL_LANE_STAGGERING_DELAY_MASK,
+		     XELPDP_POWER_STATE_ACTIVE(CX0_P0_STATE_ACTIVE) |
+		     XELPDP_PLL_LANE_STAGGERING_DELAY(0));
+}
+
+static u32 intel_cx0_get_pclk_refclk_request(u8 lane_mask)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_PCLK_REFCLK_REQUEST(lane);
+
+	return val;
+}
+
+static u32 intel_cx0_get_pclk_refclk_ack(u8 lane_mask)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_PCLK_REFCLK_ACK(lane);
+
+	return val;
+}
+
+/* FIXME: Some Type-C cases need not reset both the lanes. Handle those cases. */
+static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, enum port port,
+				     bool lane_reversal)
+{
+	enum phy phy = intel_port_to_phy(i915, port);
+	u8 lane_mask = lane_reversal ? INTEL_CX0_LANE1 :
+				  INTEL_CX0_LANE0;
+
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL1(port),
+					 XELPDP_PORT_BUF_SOC_PHY_READY,
+					 XELPDP_PORT_BUF_SOC_PHY_READY,
+					 XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "PHY %c failed to bring out of SOC reset after %dus.\n",
+			 phy_name(phy), XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US);
+
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
+		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1),
+		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1));
+
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port),
+					 XELPDP_LANE_PHY_CURRENT_STATUS(0) |
+					 XELPDP_LANE_PHY_CURRENT_STATUS(1),
+					 XELPDP_LANE_PHY_CURRENT_STATUS(0) |
+					 XELPDP_LANE_PHY_CURRENT_STATUS(1),
+					 XELPDP_PORT_RESET_START_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n",
+			 phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US);
+
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(port),
+		     intel_cx0_get_pclk_refclk_request(INTEL_CX0_BOTH_LANES),
+		     intel_cx0_get_pclk_refclk_request(lane_mask));
+
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(port),
+					 intel_cx0_get_pclk_refclk_ack(INTEL_CX0_BOTH_LANES),
+					 intel_cx0_get_pclk_refclk_ack(lane_mask),
+					 XELPDP_REFCLK_ENABLE_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "PHY %c failed to request refclk after %dus.\n",
+			 phy_name(phy), XELPDP_REFCLK_ENABLE_TIMEOUT_US);
+
+	intel_cx0_powerdown_change_sequence(i915, port, INTEL_CX0_BOTH_LANES,
+					    CX0_P2_STATE_RESET);
+	intel_cx0_setup_powerdown(i915, port);
+
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
+		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1),
+		     0);
+
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(port),
+				    XELPDP_LANE_PHY_CURRENT_STATUS(0) |
+				    XELPDP_LANE_PHY_CURRENT_STATUS(1),
+				    XELPDP_PORT_RESET_END_TIMEOUT))
+		drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dms.\n",
+			 phy_name(phy), XELPDP_PORT_RESET_END_TIMEOUT);
+}
+
+static void intel_c10_program_phy_lane(struct drm_i915_private *i915,
+				       struct intel_encoder *encoder, int lane_count,
+				       bool lane_reversal)
+{
+	u8 l0t1, l0t2, l1t1, l1t2;
+	bool dp_alt_mode = intel_tc_port_in_dp_alt_mode(enc_to_dig_port(encoder));
+	enum port port = encoder->port;
+
+	intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_MSGBUS_ACCESS,
+		      MB_WRITE_COMMITTED);
+
+	/* TODO: DP-alt MFD case where only one PHY lane should be programmed. */
+	l0t1 = intel_cx0_read(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(1, 2));
+	l0t2 = intel_cx0_read(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(2, 2));
+	l1t1 = intel_cx0_read(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(1, 2));
+	l1t2 = intel_cx0_read(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(2, 2));
+
+	l0t1 |= CONTROL2_DISABLE_SINGLE_TX;
+	l0t2 |= CONTROL2_DISABLE_SINGLE_TX;
+	l1t1 |= CONTROL2_DISABLE_SINGLE_TX;
+	l1t2 |= CONTROL2_DISABLE_SINGLE_TX;
+
+	if (lane_reversal) {
+		switch (lane_count) {
+		case 4:
+			l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			fallthrough;
+		case 3:
+			l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			fallthrough;
+		case 2:
+			l1t1 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			fallthrough;
+		case 1:
+			l1t2 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			break;
+		default:
+			MISSING_CASE(lane_count);
+		}
+	} else {
+		switch (lane_count) {
+		case 4:
+			l1t2 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			fallthrough;
+		case 3:
+			l1t1 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			fallthrough;
+		case 2:
+			l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			break;
+		case 1:
+			if (dp_alt_mode)
+				l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			else
+				l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX;
+			break;
+		default:
+			MISSING_CASE(lane_count);
+		}
+	}
+
+	/* disable MLs */
+	intel_cx0_write(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(1, 2),
+			l0t1, MB_WRITE_COMMITTED);
+	intel_cx0_write(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(2, 2),
+			l0t2, MB_WRITE_COMMITTED);
+	intel_cx0_write(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(1, 2),
+			l1t1, MB_WRITE_COMMITTED);
+	intel_cx0_write(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(2, 2),
+			l1t2, MB_WRITE_COMMITTED);
+
+	intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+		      0, C10_VDR_CTRL_UPDATE_CFG,
+		      MB_WRITE_COMMITTED);
+}
+
+static u32 intel_cx0_get_pclk_pll_request(u8 lane_mask)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_PCLK_PLL_REQUEST(lane);
+
+	return val;
+}
+
+static u32 intel_cx0_get_pclk_pll_ack(u8 lane_mask)
+{
+	u32 val = 0;
+	int lane = 0;
+
+	for_each_cx0_lane_in_mask(lane_mask, lane)
+		val |= XELPDP_LANE_PCLK_PLL_ACK(lane);
+
+	return val;
+}
+
+static void intel_c10pll_enable(struct intel_encoder *encoder,
+				const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+	bool lane_reversal = dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL;
+	u8 maxpclk_lane = lane_reversal ? INTEL_CX0_LANE1 :
+					  INTEL_CX0_LANE0;
+
+	/*
+	 * 1. Program PORT_CLOCK_CTL REGISTER to configure
+	 * clock muxes, gating and SSC
+	 */
+	intel_program_port_clock_ctl(encoder, crtc_state, lane_reversal);
+
+	/* 2. Bring PHY out of reset. */
+	intel_cx0_phy_lane_reset(i915, encoder->port, lane_reversal);
+
+	/*
+	 * 3. Change Phy power state to Ready.
+	 * TODO: For DP alt mode use only one lane.
+	 */
+	intel_cx0_powerdown_change_sequence(i915, encoder->port, INTEL_CX0_BOTH_LANES,
+					    CX0_P2_STATE_READY);
+
+	/* 4. Program PHY internal PLL internal registers. */
+	intel_c10_pll_program(i915, crtc_state, encoder);
+
+	/*
+	 * 5. Program the enabled and disabled owned PHY lane
+	 * transmitters over message bus
+	 */
+	intel_c10_program_phy_lane(i915, encoder, crtc_state->lane_count, lane_reversal);
+
+	/*
+	 * 6. Follow the Display Voltage Frequency Switching - Sequence
+	 * Before Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 7. Program DDI_CLK_VALFREQ to match intended DDI
+	 * clock frequency.
+	 */
+	intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port),
+		       crtc_state->port_clock);
+
+	/*
+	 * 8. Set PORT_CLOCK_CTL register PCLK PLL Request
+	 * LN<Lane for maxPCLK> to "1" to enable PLL.
+	 */
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES),
+		     intel_cx0_get_pclk_pll_request(maxpclk_lane));
+
+	/* 9. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN<Lane for maxPCLK> == "1". */
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+					 intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES),
+					 intel_cx0_get_pclk_pll_ack(maxpclk_lane),
+					 XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "Port %c PLL not locked after %dus.\n",
+			 phy_name(phy), XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US);
+
+	/*
+	 * 10. Follow the Display Voltage Frequency Switching Sequence After
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+}
+
+void intel_cx0pll_enable(struct intel_encoder *encoder,
+			 const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	intel_wakeref_t wakeref;
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
+	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
+	intel_c10pll_enable(encoder, crtc_state);
+
+	/* TODO: enable TBT-ALT mode */
+	intel_cx0_phy_transaction_end(encoder, wakeref);
+}
+
+static void intel_c10pll_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	/* 1. Change owned PHY lane power to Disable state. */
+	intel_cx0_powerdown_change_sequence(i915, encoder->port, INTEL_CX0_BOTH_LANES,
+					    CX0_P2PG_STATE_DISABLE);
+
+	/*
+	 * 2. Follow the Display Voltage Frequency Switching Sequence Before
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 3. Set PORT_CLOCK_CTL register PCLK PLL Request LN<Lane for maxPCLK>
+	 * to "0" to disable PLL.
+	 */
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES) |
+		     intel_cx0_get_pclk_refclk_request(INTEL_CX0_BOTH_LANES), 0);
+
+	/* 4. Program DDI_CLK_VALFREQ to 0. */
+	intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), 0);
+
+	/*
+	 * 5. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN<Lane for maxPCLK**> == "0".
+	 */
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+					 intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES) |
+					 intel_cx0_get_pclk_refclk_ack(INTEL_CX0_BOTH_LANES), 0,
+					 XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US, 0, NULL))
+		drm_warn(&i915->drm, "Port %c PLL not unlocked after %dus.\n",
+			 phy_name(phy), XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US);
+
+	/*
+	 * 6. Follow the Display Voltage Frequency Switching Sequence After
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/* 7. Program PORT_CLOCK_CTL register to disable and gate clocks. */
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     XELPDP_DDI_CLOCK_SELECT_MASK |
+		     XELPDP_FORWARD_CLOCK_UNGATE, 0);
+}
+
+void intel_cx0pll_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	intel_wakeref_t wakeref;
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
+	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
+	intel_c10pll_disable(encoder);
+	intel_cx0_phy_transaction_end(encoder, wakeref);
+}
+
+void intel_c10pll_state_verify(struct intel_atomic_state *state,
+			       struct intel_crtc_state *new_crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_c10pll_state mpllb_hw_state = { 0 };
+	struct intel_c10pll_state *mpllb_sw_state = &new_crtc_state->cx0pll_state.c10;
+	struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc);
+	struct intel_encoder *encoder;
+	enum phy phy;
+	int i;
+
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	if (!new_crtc_state->hw.active)
+		return;
+
+	encoder = intel_get_crtc_new_encoder(state, new_crtc_state);
+	phy = intel_port_to_phy(i915, encoder->port);
+
+	if (!intel_is_c10phy(i915, phy))
+		return;
+
+	intel_c10pll_readout_hw_state(encoder, &mpllb_hw_state);
+
+	for (i = 0; i < ARRAY_SIZE(mpllb_sw_state->pll); i++) {
+		u8 expected = mpllb_sw_state->pll[i];
+
+		I915_STATE_WARN(mpllb_hw_state.pll[i] != expected,
+				"[CRTC:%d:%s] mismatch in C10MPLLB: Register[%d] (expected 0x%02x, found 0x%02x)",
+				crtc->base.base.id, crtc->base.name,
+				i, expected, mpllb_hw_state.pll[i]);
+	}
+
+	I915_STATE_WARN(mpllb_hw_state.tx != mpllb_sw_state->tx,
+			"[CRTC:%d:%s] mismatch in C10MPLLB: Register TX0 (expected 0x%02x, found 0x%02x)",
+			crtc->base.base.id, crtc->base.name,
+			mpllb_sw_state->tx, mpllb_hw_state.tx);
+
+	I915_STATE_WARN(mpllb_hw_state.cmn != mpllb_sw_state->cmn,
+			"[CRTC:%d:%s] mismatch in C10MPLLB: Register CMN0 (expected 0x%02x, found 0x%02x)",
+			crtc->base.base.id, crtc->base.name,
+			mpllb_sw_state->cmn, mpllb_hw_state.cmn);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
new file mode 100644
index 0000000000000..46fa0576ef0dd
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -0,0 +1,34 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_CX0_PHY_H__
+#define __INTEL_CX0_PHY_H__
+
+#include <linux/types.h>
+#include <linux/bitfield.h>
+#include <linux/bits.h>
+
+#include "i915_drv.h"
+#include "intel_display_types.h"
+
+struct drm_i915_private;
+struct intel_encoder;
+struct intel_crtc_state;
+enum phy;
+
+bool intel_is_c10phy(struct drm_i915_private *dev_priv, enum phy phy);
+void intel_cx0pll_enable(struct intel_encoder *encoder,
+			 const struct intel_crtc_state *crtc_state);
+void intel_cx0pll_disable(struct intel_encoder *encoder);
+void intel_c10pll_readout_hw_state(struct intel_encoder *encoder, struct intel_c10pll_state *pll_state);
+int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder);
+void intel_c10pll_dump_hw_state(struct drm_i915_private *dev_priv,
+				const struct intel_c10pll_state *hw_state);
+int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
+				 const struct intel_c10pll_state *pll_state);
+void intel_c10pll_state_verify(struct intel_atomic_state *state,
+			       struct intel_crtc_state *new_crtc_state);
+
+#endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 916002aad7463..21a699c678a1d 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -96,6 +96,11 @@
 #define   XELPDP_PLL_LANE_STAGGERING_DELAY(val)		REG_FIELD_PREP(XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, val)
 #define   XELPDP_POWER_STATE_ACTIVE_MASK		REG_GENMASK(3, 0)
 #define   XELPDP_POWER_STATE_ACTIVE(val)		REG_FIELD_PREP(XELPDP_POWER_STATE_ACTIVE_MASK, val)
+#define   CX0_P0_STATE_ACTIVE				0x0
+#define   CX0_P2_STATE_READY				0x2
+#define   CX0_P2PG_STATE_DISABLE			0x9
+#define   CX0_P4PG_STATE_DISABLE			0xC
+#define   CX0_P2_STATE_RESET				0x2
 
 #define _XELPDP_PORT_CLOCK_CTL_A			0x640E0
 #define _XELPDP_PORT_CLOCK_CTL_B			0x641E0
@@ -106,14 +111,11 @@
 										 _XELPDP_PORT_CLOCK_CTL_B, \
 										 _XELPDP_PORT_CLOCK_CTL_USBC1, \
 										 _XELPDP_PORT_CLOCK_CTL_USBC2))
-#define   XELPDP_LANE0_PCLK_PLL_REQUEST			REG_BIT(31)
-#define   XELPDP_LANE0_PCLK_PLL_ACK			REG_BIT(30)
-#define   XELPDP_LANE0_PCLK_REFCLK_REQUEST		REG_BIT(29)
-#define   XELPDP_LANE0_PCLK_REFCLK_ACK			REG_BIT(28)
-#define   XELPDP_LANE1_PCLK_PLL_REQUEST			REG_BIT(27)
-#define   XELPDP_LANE1_PCLK_PLL_ACK			REG_BIT(26)
-#define   XELPDP_LANE1_PCLK_REFCLK_REQUEST		REG_BIT(25)
-#define   XELPDP_LANE1_PCLK_REFCLK_ACK			REG_BIT(24)
+#define   XELPDP_LANE_PCLK_PLL_REQUEST(lane)		REG_BIT(31 - ((lane) * 4))
+#define   XELPDP_LANE_PCLK_PLL_ACK(lane)		REG_BIT(30 - ((lane) * 4))
+#define   XELPDP_LANE_PCLK_REFCLK_REQUEST(lane)		REG_BIT(29 - ((lane) * 4))
+#define   XELPDP_LANE_PCLK_REFCLK_ACK(lane)		REG_BIT(28 - ((lane) * 4))
+
 #define   XELPDP_TBT_CLOCK_REQUEST			REG_BIT(19)
 #define   XELPDP_TBT_CLOCK_ACK				REG_BIT(18)
 #define   XELPDP_DDI_CLOCK_SELECT_MASK			REG_GENMASK(15, 12)
@@ -130,4 +132,31 @@
 #define   XELPDP_SSC_ENABLE_PLLA			REG_BIT(1)
 #define   XELPDP_SSC_ENABLE_PLLB			REG_BIT(0)
 
-#endif /* __INTEL_CX0_PHY_REGS_H__ */
+/* C10 Vendor Registers */
+#define PHY_C10_VDR_PLL(idx)		(0xC00 + (idx))
+#define   C10_PLL0_FRACEN		REG_BIT8(4)
+#define   C10_PLL3_MULTIPLIERH_MASK	REG_GENMASK8(3, 0)
+#define   C10_PLL15_TXCLKDIV_MASK	REG_GENMASK8(2, 0)
+#define PHY_C10_VDR_CMN(idx)		(0xC20 + (idx))
+#define   C10_CMN0_REF_RANGE		REG_FIELD_PREP(REG_GENMASK(4, 0), 1)
+#define   C10_CMN0_REF_CLK_MPLLB_DIV	REG_FIELD_PREP(REG_GENMASK(7, 5), 1)
+#define   C10_CMN3_TXVBOOST_MASK	REG_GENMASK8(7, 5)
+#define   C10_CMN3_TXVBOOST(val)	REG_FIELD_PREP8(C10_CMN3_TXVBOOST_MASK, val)
+#define PHY_C10_VDR_TX(idx)		(0xC30 + (idx))
+#define   C10_TX0_TX_MPLLB_SEL		REG_BIT(4)
+#define PHY_C10_VDR_CONTROL(idx)	(0xC70 + (idx) - 1)
+#define   C10_VDR_CTRL_MSGBUS_ACCESS	REG_BIT8(2)
+#define   C10_VDR_CTRL_MASTER_LANE	REG_BIT8(1)
+#define   C10_VDR_CTRL_UPDATE_CFG	REG_BIT8(0)
+#define PHY_C10_VDR_CUSTOM_WIDTH	0xD02
+#define   C10_VDR_CUSTOM_WIDTH_MASK    REG_GENMASK(1, 0)
+#define   C10_VDR_CUSTOM_WIDTH_8_10    REG_FIELD_PREP(C10_VDR_CUSTOM_WIDTH_MASK, 0)
+
+/* PHY_C10_VDR_PLL0 */
+#define PLL_C10_MPLL_SSC_EN             REG_BIT8(0)
+
+/* PIPE SPEC Defined Registers */
+#define PHY_CX0_TX_CONTROL(tx, control)	(0x400 + ((tx) - 1) * 0x200 + (control))
+#define   CONTROL2_DISABLE_SINGLE_TX	REG_BIT(6)
+
+#endif /* __INTEL_CX0_REG_DEFS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index d6fb07821cffb..de7cf773440d1 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -39,6 +39,7 @@
 #include "intel_combo_phy_regs.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
+#include "intel_cx0_phy.h"
 #include "intel_ddi.h"
 #include "intel_ddi_buf_trans.h"
 #include "intel_de.h"
@@ -3488,6 +3489,21 @@ void intel_ddi_get_clock(struct intel_encoder *encoder,
 						     &crtc_state->dpll_hw_state);
 }
 
+static void mtl_ddi_get_config(struct intel_encoder *encoder,
+			       struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
+
+	intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10);
+	intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
+	crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
+
+	intel_ddi_get_config(encoder, crtc_state);
+}
+
 static void dg2_ddi_get_config(struct intel_encoder *encoder,
 				struct intel_crtc_state *crtc_state)
 {
@@ -4396,7 +4412,11 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	encoder->cloneable = 0;
 	encoder->pipe_mask = ~0;
 
-	if (IS_DG2(dev_priv)) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		encoder->enable_clock = intel_cx0pll_enable;
+		encoder->disable_clock = intel_cx0pll_disable;
+		encoder->get_config = mtl_ddi_get_config;
+	} else if (IS_DG2(dev_priv)) {
 		encoder->enable_clock = intel_mpllb_enable;
 		encoder->disable_clock = intel_mpllb_disable;
 		encoder->get_config = dg2_ddi_get_config;
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index b28c8dc48e244..64b6a5ec4b818 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -989,6 +989,18 @@ struct intel_csc_matrix {
 	u16 postoff[3];
 };
 
+struct intel_c10pll_state {
+	u32 clock; /* in KHz */
+	u8 tx;
+	u8 cmn;
+	u8 pll[20];
+};
+
+struct intel_cx0pll_state {
+	struct intel_c10pll_state c10;
+	bool ssc_enabled;
+};
+
 struct intel_crtc_state {
 	/*
 	 * uapi (drm) state. This is the software state shown to userspace.
@@ -1134,6 +1146,7 @@ struct intel_crtc_state {
 	union {
 		struct intel_dpll_hw_state dpll_hw_state;
 		struct intel_mpllb_state mpllb_state;
+		struct intel_cx0pll_state cx0pll_state;
 	};
 
 	/*
diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c
index 4e9c18be7e1fd..a9fbef0fa817c 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll.c
@@ -8,6 +8,7 @@
 
 #include "i915_reg.h"
 #include "intel_crtc.h"
+#include "intel_cx0_phy.h"
 #include "intel_de.h"
 #include "intel_display.h"
 #include "intel_display_types.h"
@@ -995,6 +996,30 @@ static int dg2_crtc_compute_clock(struct intel_atomic_state *state,
 	return 0;
 }
 
+static int mtl_crtc_compute_clock(struct intel_atomic_state *state,
+				  struct intel_crtc *crtc)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_crtc_state *crtc_state =
+		intel_atomic_get_new_crtc_state(state, crtc);
+	struct intel_encoder *encoder =
+		intel_get_crtc_new_encoder(state, crtc_state);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	int ret;
+
+	ret = intel_cx0pll_calc_state(crtc_state, encoder);
+	if (ret)
+		return ret;
+
+	/* TODO: Do the readback via intel_compute_shared_dplls() */
+	if (intel_is_c10phy(i915, phy))
+		crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
+
+	crtc_state->hw.adjusted_mode.crtc_clock = intel_crtc_dotclock(crtc_state);
+
+	return 0;
+}
+
 static bool ilk_needs_fb_cb_tune(const struct dpll *dpll, int factor)
 {
 	return dpll->m < factor * dpll->n;
@@ -1423,6 +1448,10 @@ static int i8xx_crtc_compute_clock(struct intel_atomic_state *state,
 	return 0;
 }
 
+static const struct intel_dpll_funcs mtl_dpll_funcs = {
+	.crtc_compute_clock = mtl_crtc_compute_clock,
+};
+
 static const struct intel_dpll_funcs dg2_dpll_funcs = {
 	.crtc_compute_clock = dg2_crtc_compute_clock,
 };
@@ -1517,7 +1546,9 @@ int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state,
 void
 intel_dpll_init_clock_hook(struct drm_i915_private *dev_priv)
 {
-	if (IS_DG2(dev_priv))
+	if (DISPLAY_VER(dev_priv) >= 14)
+		dev_priv->display.funcs.dpll = &mtl_dpll_funcs;
+	else if (IS_DG2(dev_priv))
 		dev_priv->display.funcs.dpll = &dg2_dpll_funcs;
 	else if (DISPLAY_VER(dev_priv) >= 9 || HAS_DDI(dev_priv))
 		dev_priv->display.funcs.dpll = &hsw_dpll_funcs;
diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index 22fc908b7e5d6..ed372d227aa73 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -4104,7 +4104,7 @@ void intel_shared_dpll_init(struct drm_i915_private *dev_priv)
 
 	mutex_init(&dev_priv->display.dpll.lock);
 
-	if (IS_DG2(dev_priv))
+	if (DISPLAY_VER(dev_priv) >= 14 || IS_DG2(dev_priv))
 		/* No shared DPLLs on DG2; port PLLs are part of the PHY */
 		dpll_mgr = NULL;
 	else if (IS_ALDERLAKE_P(dev_priv))
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_verify.c b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
index 842d70f0dfd2a..5e0ec15d9fd5e 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_verify.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
@@ -11,6 +11,7 @@
 #include "intel_atomic.h"
 #include "intel_crtc.h"
 #include "intel_crtc_state_dump.h"
+#include "intel_cx0_phy.h"
 #include "intel_display.h"
 #include "intel_display_types.h"
 #include "intel_fdi.h"
@@ -236,6 +237,7 @@ void intel_modeset_verify_crtc(struct intel_crtc *crtc,
 	verify_crtc_state(crtc, old_crtc_state, new_crtc_state);
 	intel_shared_dpll_state_verify(crtc, old_crtc_state, new_crtc_state);
 	intel_mpllb_state_verify(state, new_crtc_state);
+	intel_c10pll_state_verify(state, new_crtc_state);
 }
 
 void intel_modeset_verify_disabled(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 4511f01ca7056..924557294960c 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1799,6 +1799,11 @@
 #define CLKGATE_DIS_PSL_EXT(pipe) \
 	_MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_EXT_A, _CLKGATE_DIS_PSL_EXT_B)
 
+/* DDI Buffer Control */
+#define _DDI_CLK_VALFREQ_A		0x64030
+#define _DDI_CLK_VALFREQ_B		0x64130
+#define DDI_CLK_VALFREQ(port)		_MMIO_PORT(port, _DDI_CLK_VALFREQ_A, _DDI_CLK_VALFREQ_B)
+
 /*
  * Display engine regs
  */
diff --git a/drivers/gpu/drm/i915/i915_reg_defs.h b/drivers/gpu/drm/i915/i915_reg_defs.h
index db26de6b57bc8..622d603080f9e 100644
--- a/drivers/gpu/drm/i915/i915_reg_defs.h
+++ b/drivers/gpu/drm/i915/i915_reg_defs.h
@@ -22,6 +22,19 @@
 	       BUILD_BUG_ON_ZERO(__is_constexpr(__n) &&		\
 				 ((__n) < 0 || (__n) > 31))))
 
+/**
+ * REG_BIT8() - Prepare a u8 bit value
+ * @__n: 0-based bit number
+ *
+ * Local wrapper for BIT() to force u8, with compile time checks.
+ *
+ * @return: Value with bit @__n set.
+ */
+#define REG_BIT8(__n)                                                   \
+	((u8)(BIT(__n) +                                                \
+	       BUILD_BUG_ON_ZERO(__is_constexpr(__n) &&         \
+				 ((__n) < 0 || (__n) > 7))))
+
 /**
  * REG_GENMASK() - Prepare a continuous u32 bitmask
  * @__high: 0-based high bit
@@ -52,6 +65,21 @@
 				 __is_constexpr(__low) &&		\
 				 ((__low) < 0 || (__high) > 63 || (__low) > (__high)))))
 
+/**
+ * REG_GENMASK8() - Prepare a continuous u8 bitmask
+ * @__high: 0-based high bit
+ * @__low: 0-based low bit
+ *
+ * Local wrapper for GENMASK() to force u8, with compile time checks.
+ *
+ * @return: Continuous bitmask from @__high to @__low, inclusive.
+ */
+#define REG_GENMASK8(__high, __low)                                     \
+	((u8)(GENMASK(__high, __low) +                                  \
+	       BUILD_BUG_ON_ZERO(__is_constexpr(__high) &&      \
+				 __is_constexpr(__low) &&               \
+				 ((__low) < 0 || (__high) > 7 || (__low) > (__high)))))
+
 /*
  * Local integer constant expression version of is_power_of_2().
  */
@@ -74,6 +102,23 @@
 	       BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \
 	       BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0))))
 
+/**
+ * REG_FIELD_PREP8() - Prepare a u8 bitfield value
+ * @__mask: shifted mask defining the field's length and position
+ * @__val: value to put in the field
+ *
+ * Local copy of FIELD_PREP() to generate an integer constant expression, force
+ * u8 and for consistency with REG_FIELD_GET8(), REG_BIT8() and REG_GENMASK8().
+ *
+ * @return: @__val masked and shifted into the field defined by @__mask.
+ */
+#define REG_FIELD_PREP8(__mask, __val)                                          \
+	((u8)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) +      \
+	       BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) +             \
+	       BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U8_MAX) +          \
+	       BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \
+	       BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0))))
+
 /**
  * REG_FIELD_GET() - Extract a u32 bitfield value
  * @__mask: shifted mask defining the field's length and position
@@ -155,6 +200,18 @@
  */
 #define _PICK(__index, ...) (((const u32 []){ __VA_ARGS__ })[__index])
 
+/**
+ * REG_FIELD_GET8() - Extract a u8 bitfield value
+ * @__mask: shifted mask defining the field's length and position
+ * @__val: value to extract the bitfield value from
+ *
+ * Local wrapper for FIELD_GET() to force u8 and for consistency with
+ * REG_FIELD_PREP(), REG_BIT() and REG_GENMASK().
+ *
+ * @return: Masked and shifted value of the field defined by @__mask in @__val.
+ */
+#define REG_FIELD_GET8(__mask, __val)   ((u8)FIELD_GET(__mask, __val))
+
 typedef struct {
 	u32 reg;
 } i915_reg_t;
-- 
GitLab


From ea8af87ae6be578b3b633ad6aa9188b0ce4cd7ee Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Thu, 13 Apr 2023 14:24:38 -0700
Subject: [PATCH 0039/1791] drm/i915/mtl: Add vswing programming for C10 phys

C10 phys uses direct mapping internally for voltage and pre-emphasis levels.
Program the levels directly to the fields in the VDR Registers.

Bspec: 65449

v2: From table "C10: Tx EQ settings for DP 1.4x" it shows level 1
    and preemphasis 1 instead of two times of level 1 preemphasis 0.
    Fix this in the driver code as well.
v3: VSwing update (Clint)
v4: Add vboost termination ctl programming(Imre)
    Fix tx llogic and other nits
    Restrict C10 vdr ctl register access for C10 phy(RK)
v5: Program vboots, termination ctl for both lanes(Imre)

Cc: Imre Deak <imre.deak@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>(v3)
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-5-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 102 +++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_cx0_phy.h  |   2 +
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  14 ++-
 drivers/gpu/drm/i915/display/intel_ddi.c      |   4 +-
 .../drm/i915/display/intel_ddi_buf_trans.c    |  31 +++++-
 5 files changed, 143 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 9ab1e686a40b6..5ffd661fa5074 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -6,6 +6,8 @@
 #include "i915_reg.h"
 #include "intel_cx0_phy.h"
 #include "intel_cx0_phy_regs.h"
+#include "intel_ddi.h"
+#include "intel_ddi_buf_trans.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
@@ -292,6 +294,97 @@ static void intel_cx0_rmw(struct drm_i915_private *i915, enum port port,
 		__intel_cx0_rmw(i915, port, lane, addr, clear, set, committed);
 }
 
+static u8 intel_c10_get_tx_vboost_lvl(const struct intel_crtc_state *crtc_state)
+{
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
+		if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP) &&
+		    (crtc_state->port_clock == 540000 ||
+		     crtc_state->port_clock == 810000))
+			return 5;
+		else
+			return 4;
+	} else {
+		return 5;
+	}
+}
+
+static u8 intel_c10_get_tx_term_ctl(const struct intel_crtc_state *crtc_state)
+{
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
+		if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP) &&
+		    (crtc_state->port_clock == 540000 ||
+		     crtc_state->port_clock == 810000))
+			return 5;
+		else
+			return 2;
+	} else {
+		return 6;
+	}
+}
+
+void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
+				     const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	const struct intel_ddi_buf_trans *trans;
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	intel_wakeref_t wakeref;
+	int n_entries, ln;
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
+	trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries);
+	if (drm_WARN_ON_ONCE(&i915->drm, !trans)) {
+		intel_cx0_phy_transaction_end(encoder, wakeref);
+		return;
+	}
+
+	if (intel_is_c10phy(i915, phy)) {
+		intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+			      0, C10_VDR_CTRL_MSGBUS_ACCESS, MB_WRITE_COMMITTED);
+		intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CMN(3),
+			      C10_CMN3_TXVBOOST_MASK,
+			      C10_CMN3_TXVBOOST(intel_c10_get_tx_vboost_lvl(crtc_state)),
+			      MB_WRITE_UNCOMMITTED);
+		intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_TX(1),
+			      C10_TX1_TERMCTL_MASK,
+			      C10_TX1_TERMCTL(intel_c10_get_tx_term_ctl(crtc_state)),
+			      MB_WRITE_COMMITTED);
+	}
+
+	for (ln = 0; ln < crtc_state->lane_count; ln++) {
+		int level = intel_ddi_level(encoder, crtc_state, ln);
+		int lane, tx;
+
+		lane = ln / 2;
+		tx = ln % 2;
+
+		intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 0),
+			      C10_PHY_OVRD_LEVEL_MASK,
+			      C10_PHY_OVRD_LEVEL(trans->entries[level].snps.pre_cursor),
+			      MB_WRITE_COMMITTED);
+		intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 1),
+			      C10_PHY_OVRD_LEVEL_MASK,
+			      C10_PHY_OVRD_LEVEL(trans->entries[level].snps.vswing),
+			      MB_WRITE_COMMITTED);
+		intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 2),
+			      C10_PHY_OVRD_LEVEL_MASK,
+			      C10_PHY_OVRD_LEVEL(trans->entries[level].snps.post_cursor),
+			      MB_WRITE_COMMITTED);
+	}
+
+	/* Write Override enables in 0xD71 */
+	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_OVRD,
+		      0, PHY_C10_VDR_OVRD_TX1 | PHY_C10_VDR_OVRD_TX2,
+		      MB_WRITE_COMMITTED);
+
+	if (intel_is_c10phy(i915, phy))
+		intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
+			      0, C10_VDR_CTRL_UPDATE_CFG, MB_WRITE_COMMITTED);
+
+	intel_cx0_phy_transaction_end(encoder, wakeref);
+}
+
 /*
  * Basic DP link rates with 38.4 MHz reference clock.
  * Note: The tables below are with SSC. In non-ssc
@@ -766,10 +859,8 @@ static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
 	val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0;
 
 	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
-		     XELPDP_LANE1_PHY_CLOCK_SELECT |
-		     XELPDP_FORWARD_CLOCK_UNGATE |
-		     XELPDP_DDI_CLOCK_SELECT_MASK |
-		     XELPDP_SSC_ENABLE_PLLB, val);
+		     XELPDP_LANE1_PHY_CLOCK_SELECT | XELPDP_FORWARD_CLOCK_UNGATE |
+		     XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_SSC_ENABLE_PLLB, val);
 }
 
 static u32 intel_cx0_get_powerdown_update(u8 lane_mask)
@@ -1144,7 +1235,8 @@ static void intel_c10pll_disable(struct intel_encoder *encoder)
 
 	/* 7. Program PORT_CLOCK_CTL register to disable and gate clocks. */
 	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
-		     XELPDP_DDI_CLOCK_SELECT_MASK |
+		     XELPDP_DDI_CLOCK_SELECT_MASK, 0);
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
 		     XELPDP_FORWARD_CLOCK_UNGATE, 0);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 46fa0576ef0dd..6b736acb83e01 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -30,5 +30,7 @@ int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 				 const struct intel_c10pll_state *pll_state);
 void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state);
+void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
+				     const struct intel_crtc_state *crtc_state);
 
 #endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 21a699c678a1d..9cfa7f508c908 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -144,6 +144,8 @@
 #define   C10_CMN3_TXVBOOST(val)	REG_FIELD_PREP8(C10_CMN3_TXVBOOST_MASK, val)
 #define PHY_C10_VDR_TX(idx)		(0xC30 + (idx))
 #define   C10_TX0_TX_MPLLB_SEL		REG_BIT(4)
+#define   C10_TX1_TERMCTL_MASK		REG_GENMASK8(7, 5)
+#define   C10_TX1_TERMCTL(val)		REG_FIELD_PREP8(C10_TX1_TERMCTL_MASK, val)
 #define PHY_C10_VDR_CONTROL(idx)	(0xC70 + (idx) - 1)
 #define   C10_VDR_CTRL_MSGBUS_ACCESS	REG_BIT8(2)
 #define   C10_VDR_CTRL_MASTER_LANE	REG_BIT8(1)
@@ -151,9 +153,15 @@
 #define PHY_C10_VDR_CUSTOM_WIDTH	0xD02
 #define   C10_VDR_CUSTOM_WIDTH_MASK    REG_GENMASK(1, 0)
 #define   C10_VDR_CUSTOM_WIDTH_8_10    REG_FIELD_PREP(C10_VDR_CUSTOM_WIDTH_MASK, 0)
-
-/* PHY_C10_VDR_PLL0 */
-#define PLL_C10_MPLL_SSC_EN             REG_BIT8(0)
+#define PHY_C10_VDR_OVRD		0xD71
+#define   PHY_C10_VDR_OVRD_TX1		REG_BIT8(0)
+#define   PHY_C10_VDR_OVRD_TX2		REG_BIT8(2)
+#define PHY_C10_VDR_PRE_OVRD_TX1	0xD80
+#define C10_PHY_OVRD_LEVEL_MASK		REG_GENMASK8(5, 0)
+#define C10_PHY_OVRD_LEVEL(val)		REG_FIELD_PREP8(C10_PHY_OVRD_LEVEL_MASK, val)
+#define PHY_CX0_VDROVRD_CTL(lane, tx, control)				\
+					(PHY_C10_VDR_PRE_OVRD_TX1 +	\
+					 ((lane) ^ (tx)) * 0x10 + (control))
 
 /* PIPE SPEC Defined Registers */
 #define PHY_CX0_TX_CONTROL(tx, control)	(0x400 + ((tx) - 1) * 0x200 + (control))
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index de7cf773440d1..0a057624ae7c9 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4476,7 +4476,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 		encoder->get_config = hsw_ddi_get_config;
 	}
 
-	if (IS_DG2(dev_priv)) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		encoder->set_signal_levels = intel_cx0_phy_set_signal_levels;
+	} else if (IS_DG2(dev_priv)) {
 		encoder->set_signal_levels = intel_snps_phy_set_signal_levels;
 	} else if (DISPLAY_VER(dev_priv) >= 12) {
 		if (intel_phy_is_combo(dev_priv, phy))
diff --git a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
index 006a2e9790003..cd4becbae0988 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
@@ -1035,6 +1035,25 @@ static const struct intel_ddi_buf_trans dg2_snps_trans_uhbr = {
 	.num_entries = ARRAY_SIZE(_dg2_snps_trans_uhbr),
 };
 
+static const union intel_ddi_buf_trans_entry _mtl_c10_trans_dp14[] = {
+	{ .snps = { 26, 0, 0  } },      /* preset 0 */
+	{ .snps = { 33, 0, 6  } },      /* preset 1 */
+	{ .snps = { 38, 0, 11 } },      /* preset 2 */
+	{ .snps = { 43, 0, 19 } },      /* preset 3 */
+	{ .snps = { 39, 0, 0  } },      /* preset 4 */
+	{ .snps = { 45, 0, 7  } },      /* preset 5 */
+	{ .snps = { 46, 0, 13 } },      /* preset 6 */
+	{ .snps = { 46, 0, 0  } },      /* preset 7 */
+	{ .snps = { 55, 0, 7  } },      /* preset 8 */
+	{ .snps = { 62, 0, 0  } },      /* preset 9 */
+};
+
+static const struct intel_ddi_buf_trans mtl_cx0c10_trans = {
+	.entries = _mtl_c10_trans_dp14,
+	.num_entries = ARRAY_SIZE(_mtl_c10_trans_dp14),
+	.hdmi_default_entry = ARRAY_SIZE(_mtl_c10_trans_dp14) - 1,
+};
+
 bool is_hobl_buf_trans(const struct intel_ddi_buf_trans *table)
 {
 	return table == &tgl_combo_phy_trans_edp_hbr2_hobl;
@@ -1606,12 +1625,22 @@ dg2_get_snps_buf_trans(struct intel_encoder *encoder,
 		return intel_get_buf_trans(&dg2_snps_trans, n_entries);
 }
 
+static const struct intel_ddi_buf_trans *
+mtl_get_cx0_buf_trans(struct intel_encoder *encoder,
+		      const struct intel_crtc_state *crtc_state,
+		      int *n_entries)
+{
+	return intel_get_buf_trans(&mtl_cx0c10_trans, n_entries);
+}
+
 void intel_ddi_buf_trans_init(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
 
-	if (IS_DG2(i915)) {
+	if (DISPLAY_VER(i915) >= 14) {
+		encoder->get_buf_trans = mtl_get_cx0_buf_trans;
+	} else if (IS_DG2(i915)) {
 		encoder->get_buf_trans = dg2_get_snps_buf_trans;
 	} else if (IS_ALDERLAKE_P(i915)) {
 		if (intel_phy_is_combo(i915, phy))
-- 
GitLab


From babde06db8858a4fdb4ab3c64e442885487dbd8c Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Thu, 13 Apr 2023 14:24:39 -0700
Subject: [PATCH 0040/1791] drm/i915/mtl: MTL PICA hotplug detection

PICA is used for DP alt mode and TBT modes. Hotplug interruption is routed
from PICA chip to south display engine and from there to north display
engine. This patch adds functionality to enable hotplug detection for
all Type-C ports (4 ports available).

Differently from HPD in south display, PICA provides a dedicated HPD
control register for each supported port, so we loop over ports
ourselves instead of using intel_hpd_hotplug_enables() or
intel_get_hpd_pins().

BSpec: 49305, 55726, 65107, 65300

Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Madhumitha Tolakanahalli Pradeep <madhumitha.tolakanahalli.pradeep@intel.com>
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-6-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 237 +++++++++++++++++++++++++++++++-
 drivers/gpu/drm/i915/i915_reg.h |  31 ++++-
 2 files changed, 261 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index d24bdea65a3dc..b4dd6a5a536f9 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -162,6 +162,13 @@ static const u32 hpd_gen11[HPD_NUM_PINS] = {
 	[HPD_PORT_TC6] = GEN11_TC_HOTPLUG(HPD_PORT_TC6) | GEN11_TBT_HOTPLUG(HPD_PORT_TC6),
 };
 
+static const u32 hpd_xelpdp[HPD_NUM_PINS] = {
+	[HPD_PORT_TC1] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC1) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC2) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC3) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC4) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC4),
+};
+
 static const u32 hpd_icp[HPD_NUM_PINS] = {
 	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
 	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
@@ -182,6 +189,15 @@ static const u32 hpd_sde_dg1[HPD_NUM_PINS] = {
 	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_DG2(HPD_PORT_TC1),
 };
 
+static const u32 hpd_mtp[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
+	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
+	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
+};
+
 static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
 {
 	struct intel_hotplug *hpd = &dev_priv->display.hotplug;
@@ -195,7 +211,9 @@ static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
 		return;
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 11)
+	if (DISPLAY_VER(dev_priv) >= 14)
+		hpd->hpd = hpd_xelpdp;
+	else if (DISPLAY_VER(dev_priv) >= 11)
 		hpd->hpd = hpd_gen11;
 	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
 		hpd->hpd = hpd_bxt;
@@ -214,6 +232,8 @@ static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
 
 	if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
 		hpd->pch_hpd = hpd_sde_dg1;
+	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP)
+		hpd->pch_hpd = hpd_mtp;
 	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 		hpd->pch_hpd = hpd_icp;
 	else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv))
@@ -1559,6 +1579,44 @@ static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		cpt_serr_int_handler(dev_priv);
 }
 
+static void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir)
+{
+	enum hpd_pin pin;
+	u32 hotplug_trigger = iir & (XELPDP_DP_ALT_HOTPLUG_MASK | XELPDP_TBT_HOTPLUG_MASK);
+	u32 trigger_aux = iir & XELPDP_AUX_TC_MASK;
+	u32 pin_mask = 0, long_mask = 0;
+
+	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
+		u32 val;
+
+		if (!(i915->display.hotplug.hpd[pin] & hotplug_trigger))
+			continue;
+
+		pin_mask |= BIT(pin);
+
+		val = intel_de_read(i915, XELPDP_PORT_HOTPLUG_CTL(pin));
+		intel_de_write(i915, XELPDP_PORT_HOTPLUG_CTL(pin), val);
+
+		if (val & (XELPDP_DP_ALT_HPD_LONG_DETECT | XELPDP_TBT_HPD_LONG_DETECT))
+			long_mask |= BIT(pin);
+	}
+
+	if (pin_mask) {
+		drm_dbg(&i915->drm,
+			"pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n",
+			hotplug_trigger, pin_mask, long_mask);
+
+		intel_hpd_irq_handler(i915, pin_mask, long_mask);
+	}
+
+	if (trigger_aux)
+		dp_aux_irq_handler(i915);
+
+	if (!pin_mask && !trigger_aux)
+		drm_err(&i915->drm,
+			"Unexpected DE HPD/AUX interrupt 0x%08x\n", iir);
+}
+
 static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 {
 	u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP;
@@ -2029,6 +2087,34 @@ u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv)
 	return mask;
 }
 
+static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_iir, u32 *pica_iir)
+{
+	u32 pica_ier = 0;
+
+	*pica_iir = 0;
+	*pch_iir = intel_de_read(i915, SDEIIR);
+	if (!*pch_iir)
+		return;
+
+	/**
+	 * PICA IER must be disabled/re-enabled around clearing PICA IIR and
+	 * SDEIIR, to avoid losing PICA IRQs and to ensure that such IRQs set
+	 * their flags both in the PICA and SDE IIR.
+	 */
+	if (*pch_iir & SDE_PICAINTERRUPT) {
+		drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTP);
+
+		pica_ier = intel_de_rmw(i915, PICAINTERRUPT_IER, ~0, 0);
+		*pica_iir = intel_de_read(i915, PICAINTERRUPT_IIR);
+		intel_de_write(i915, PICAINTERRUPT_IIR, *pica_iir);
+	}
+
+	intel_de_write(i915, SDEIIR, *pch_iir);
+
+	if (pica_ier)
+		intel_de_write(i915, PICAINTERRUPT_IER, pica_ier);
+}
+
 static irqreturn_t
 gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 {
@@ -2153,16 +2239,20 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 
 	if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) &&
 	    master_ctl & GEN8_DE_PCH_IRQ) {
+		u32 pica_iir;
+
 		/*
 		 * FIXME(BDW): Assume for now that the new interrupt handling
 		 * scheme also closed the SDE interrupt handling race we've seen
 		 * on older pch-split platforms. But this needs testing.
 		 */
-		iir = intel_uncore_read(&dev_priv->uncore, SDEIIR);
+		gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir);
 		if (iir) {
-			intel_uncore_write(&dev_priv->uncore, SDEIIR, iir);
 			ret = IRQ_HANDLED;
 
+			if (pica_iir)
+				xelpdp_pica_irq_handler(dev_priv, pica_iir);
+
 			if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 				icp_irq_handler(dev_priv, iir);
 			else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
@@ -2740,7 +2830,11 @@ static void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
 
 	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
 	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
-	GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_);
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		GEN3_IRQ_RESET(uncore, PICAINTERRUPT_);
+	else
+		GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_);
 
 	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
 		GEN3_IRQ_RESET(uncore, SDE);
@@ -3031,6 +3125,116 @@ static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 		icp_hpd_irq_setup(dev_priv);
 }
 
+static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder)
+{
+	switch (encoder->hpd_pin) {
+	case HPD_PORT_A:
+	case HPD_PORT_B:
+		return SHOTPLUG_CTL_DDI_HPD_ENABLE(encoder->hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
+{
+	switch (encoder->hpd_pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+		return ICP_TC_HPD_ENABLE(encoder->hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     (SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_A) |
+		      SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_B)),
+		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
+}
+
+static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
+		     (ICP_TC_HPD_ENABLE(HPD_PORT_TC1) |
+		      ICP_TC_HPD_ENABLE(HPD_PORT_TC2) |
+		      ICP_TC_HPD_ENABLE(HPD_PORT_TC3) |
+		      ICP_TC_HPD_ENABLE(HPD_PORT_TC4)),
+		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
+}
+
+static void mtp_hpd_invert(struct drm_i915_private *i915)
+{
+	u32 val = (INVERT_DDIA_HPD |
+		   INVERT_DDIB_HPD |
+		   INVERT_DDIC_HPD |
+		   INVERT_TC1_HPD |
+		   INVERT_TC2_HPD |
+		   INVERT_TC3_HPD |
+		   INVERT_TC4_HPD |
+		   INVERT_DDID_HPD_MTP |
+		   INVERT_DDIE_HPD);
+	intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val);
+}
+
+static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd);
+
+	intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
+
+	mtp_hpd_invert(i915);
+	ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs);
+
+	mtp_ddi_hpd_detection_setup(i915);
+	mtp_tc_hpd_detection_setup(i915);
+}
+
+static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	struct intel_encoder *encoder;
+	enum hpd_pin pin;
+	u32 available_pins = 0;
+
+	BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS);
+
+	for_each_intel_encoder(&i915->drm, encoder)
+		available_pins |= BIT(encoder->hpd_pin);
+
+	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
+		u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
+			   XELPDP_DP_ALT_HOTPLUG_ENABLE;
+
+		intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(pin),
+			     mask,
+			     available_pins & BIT(pin) ?  mask : 0);
+	}
+}
+
+static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.hpd);
+
+	intel_de_rmw(i915, PICAINTERRUPT_IMR, hotplug_irqs,
+		     ~enabled_irqs & hotplug_irqs);
+	intel_uncore_posting_read(&i915->uncore, PICAINTERRUPT_IMR);
+
+	xelpdp_pica_hpd_detection_setup(i915);
+
+	if (INTEL_PCH_TYPE(i915) >= PCH_MTP)
+		mtp_hpd_irq_setup(i915);
+}
+
 static u32 spt_hotplug_enables(struct intel_encoder *encoder)
 {
 	switch (encoder->hpd_pin) {
@@ -3363,7 +3567,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
 	GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked);
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
+	if (IS_DISPLAY_VER(dev_priv, 11, 13)) {
 		u32 de_hpd_masked = 0;
 		u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK |
 				     GEN11_DE_TBT_HOTPLUG_MASK;
@@ -3373,6 +3577,20 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	}
 }
 
+static void mtp_irq_postinstall(struct drm_i915_private *i915)
+{
+	struct intel_uncore *uncore = &i915->uncore;
+	u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT;
+	u32 de_hpd_mask = XELPDP_AUX_TC_MASK;
+	u32 de_hpd_enables = de_hpd_mask | XELPDP_DP_ALT_HOTPLUG_MASK |
+			     XELPDP_TBT_HOTPLUG_MASK;
+
+	GEN3_IRQ_INIT(uncore, PICAINTERRUPT_, ~de_hpd_mask,
+		      de_hpd_enables);
+
+	GEN3_IRQ_INIT(uncore, SDE, ~sde_mask, 0xffffffff);
+}
+
 static void icp_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore *uncore = &dev_priv->uncore;
@@ -3434,7 +3652,11 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv)
 	GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked);
 
 	if (HAS_DISPLAY(dev_priv)) {
-		icp_irq_postinstall(dev_priv);
+		if (DISPLAY_VER(dev_priv) >= 14)
+			mtp_irq_postinstall(dev_priv);
+		else
+			icp_irq_postinstall(dev_priv);
+
 		gen8_de_irq_postinstall(dev_priv);
 		intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL,
 				   GEN11_DISPLAY_IRQ_ENABLE);
@@ -3920,6 +4142,7 @@ static const struct intel_hotplug_funcs platform##_hpd_funcs = { \
 }
 
 HPD_FUNCS(i915);
+HPD_FUNCS(xelpdp);
 HPD_FUNCS(dg1);
 HPD_FUNCS(gen11);
 HPD_FUNCS(bxt);
@@ -3980,6 +4203,8 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 			dev_priv->display.funcs.hotplug = &icp_hpd_funcs;
 		else if (HAS_PCH_DG1(dev_priv))
 			dev_priv->display.funcs.hotplug = &dg1_hpd_funcs;
+		else if (DISPLAY_VER(dev_priv) >= 14)
+			dev_priv->display.funcs.hotplug = &xelpdp_hpd_funcs;
 		else if (DISPLAY_VER(dev_priv) >= 11)
 			dev_priv->display.funcs.hotplug = &gen11_hpd_funcs;
 		else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 924557294960c..de95e81007c22 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4477,6 +4477,28 @@
 #define  GEN11_HOTPLUG_CTL_SHORT_DETECT(hpd_pin)	(1 << (_HPD_PIN_TC(hpd_pin) * 4))
 #define  GEN11_HOTPLUG_CTL_NO_DETECT(hpd_pin)		(0 << (_HPD_PIN_TC(hpd_pin) * 4))
 
+#define PICAINTERRUPT_ISR			_MMIO(0x16FE50)
+#define PICAINTERRUPT_IMR			_MMIO(0x16FE54)
+#define PICAINTERRUPT_IIR			_MMIO(0x16FE58)
+#define PICAINTERRUPT_IER			_MMIO(0x16FE5C)
+
+#define  XELPDP_DP_ALT_HOTPLUG(hpd_pin)		REG_BIT(16 + _HPD_PIN_TC(hpd_pin))
+#define  XELPDP_DP_ALT_HOTPLUG_MASK		REG_GENMASK(19, 16)
+
+#define  XELPDP_AUX_TC(hpd_pin)			REG_BIT(8 + _HPD_PIN_TC(hpd_pin))
+#define  XELPDP_AUX_TC_MASK			REG_GENMASK(11, 8)
+
+#define  XELPDP_TBT_HOTPLUG(hpd_pin)		REG_BIT(_HPD_PIN_TC(hpd_pin))
+#define  XELPDP_TBT_HOTPLUG_MASK		REG_GENMASK(3, 0)
+
+#define XELPDP_PORT_HOTPLUG_CTL(hpd_pin)	_MMIO(0x16F270 + (_HPD_PIN_TC(hpd_pin) * 0x200))
+#define  XELPDP_TBT_HOTPLUG_ENABLE		REG_BIT(6)
+#define  XELPDP_TBT_HPD_LONG_DETECT		REG_BIT(5)
+#define  XELPDP_TBT_HPD_SHORT_DETECT		REG_BIT(4)
+#define  XELPDP_DP_ALT_HOTPLUG_ENABLE		REG_BIT(2)
+#define  XELPDP_DP_ALT_HPD_LONG_DETECT		REG_BIT(1)
+#define  XELPDP_DP_ALT_HPD_SHORT_DETECT		REG_BIT(0)
+
 #define ILK_DISPLAY_CHICKEN2	_MMIO(0x42004)
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
 #define  ILK_ELPIN_409_SELECT	(1 << 25)
@@ -4762,7 +4784,8 @@
 				 SDE_FDI_RXB_CPT | \
 				 SDE_FDI_RXA_CPT)
 
-/* south display engine interrupt: ICP/TGP */
+/* south display engine interrupt: ICP/TGP/MTP */
+#define SDE_PICAINTERRUPT		REG_BIT(31)
 #define SDE_GMBUS_ICP			(1 << 23)
 #define SDE_TC_HOTPLUG_ICP(hpd_pin)	REG_BIT(24 + _HPD_PIN_TC(hpd_pin))
 #define SDE_TC_HOTPLUG_DG2(hpd_pin)	REG_BIT(25 + _HPD_PIN_TC(hpd_pin)) /* sigh */
@@ -5116,6 +5139,12 @@
 #define SOUTH_CHICKEN1		_MMIO(0xc2000)
 #define  FDIA_PHASE_SYNC_SHIFT_OVR	19
 #define  FDIA_PHASE_SYNC_SHIFT_EN	18
+#define  INVERT_DDIE_HPD			REG_BIT(28)
+#define  INVERT_DDID_HPD_MTP			REG_BIT(27)
+#define  INVERT_TC4_HPD				REG_BIT(26)
+#define  INVERT_TC3_HPD				REG_BIT(25)
+#define  INVERT_TC2_HPD				REG_BIT(24)
+#define  INVERT_TC1_HPD				REG_BIT(23)
 #define  INVERT_DDID_HPD			(1 << 18)
 #define  INVERT_DDIC_HPD			(1 << 17)
 #define  INVERT_DDIB_HPD			(1 << 16)
-- 
GitLab


From 23ef61946374a9ba52ae051cbc95e82f054ea16b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jos=C3=A9=20Roberto=20de=20Souza?= <jose.souza@intel.com>
Date: Thu, 13 Apr 2023 14:24:40 -0700
Subject: [PATCH 0041/1791] drm/i915/mtl/display: Implement DisplayPort
 sequences
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The differences between MTL and TGL DP sequences are big enough to
MTL have its own functions.

Also it is much easier to follow MTL sequences against spec with
its own functions.

One change worthy to mention is the move of
'intel_display_power_get(dev_priv, dig_port->ddi_io_power_domain)'.
This call is not necessary for MTL but we have _put() counter part in
intel_ddi_post_disable_dp() that needs to balanced.
We could add a display version check on it but instead here it is
moving it to intel_ddi_pre_enable_dp() so it is executed for all
platforms in a single place and this will not cause any harm in MTL
and newer platforms.

v2:
 - Fix logic to wait for buf idle.
 - Use the right register to wait for ddi active.(RK)
v3:
 - Increase wait timeout for ddi buf active (Mika)
v4:
 - Increase idle timeout for ddi buf idle (Mika)
v5: use rmw in mtl_disable_ddi_buf. Donot clear
    link training mask(Imre)

BSpec: 65448 65505
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Satyeshwar Singh <satyeshwar.singh@intel.com>
Cc: Clint Taylor <clinton.a.taylor@intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-7-radhakrishna.sripada@intel.com
---
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |   8 +
 drivers/gpu/drm/i915/display/intel_ddi.c      | 344 +++++++++++++++++-
 drivers/gpu/drm/i915/i915_reg.h               |   5 +
 3 files changed, 345 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 9cfa7f508c908..fe2e3edef69b0 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -59,8 +59,16 @@
 										 _XELPDP_PORT_BUF_CTL1_LN0_B, \
 										 _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \
 										 _XELPDP_PORT_BUF_CTL1_LN0_USBC2))
+#define   XELPDP_PORT_BUF_D2D_LINK_ENABLE		REG_BIT(29)
+#define   XELPDP_PORT_BUF_D2D_LINK_STATE		REG_BIT(28)
 #define   XELPDP_PORT_BUF_SOC_PHY_READY			REG_BIT(24)
+#define   XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK		REG_GENMASK(19, 18)
+#define   XELPDP_PORT_BUF_PORT_DATA_10BIT		REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 0)
+#define   XELPDP_PORT_BUF_PORT_DATA_20BIT		REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 1)
+#define   XELPDP_PORT_BUF_PORT_DATA_40BIT		REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 2)
 #define   XELPDP_PORT_REVERSAL				REG_BIT(16)
+#define   XELPDP_PORT_BUF_IO_SELECT_TBT			REG_BIT(11)
+#define   XELPDP_PORT_BUF_PHY_IDLE			REG_BIT(7)
 #define   XELPDP_TC_PHY_OWNERSHIP			REG_BIT(6)
 #define   XELPDP_TCSS_POWER_REQUEST			REG_BIT(5)
 #define   XELPDP_TCSS_POWER_STATE			REG_BIT(4)
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 0a057624ae7c9..842095ae3e0bc 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -40,6 +40,7 @@
 #include "intel_connector.h"
 #include "intel_crtc.h"
 #include "intel_cx0_phy.h"
+#include "intel_cx0_phy_regs.h"
 #include "intel_ddi.h"
 #include "intel_ddi_buf_trans.h"
 #include "intel_de.h"
@@ -169,6 +170,18 @@ static void hsw_prepare_hdmi_ddi_buffers(struct intel_encoder *encoder,
 		       trans->entries[level].hsw.trans2);
 }
 
+static void mtl_wait_ddi_buf_idle(struct drm_i915_private *i915, enum port port)
+{
+	int ret;
+
+	/* FIXME: find out why Bspec's 100us timeout is too short */
+	ret = wait_for_us((intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) &
+			   XELPDP_PORT_BUF_PHY_IDLE), 10000);
+	if (ret)
+		drm_err(&i915->drm, "Timeout waiting for DDI BUF %c to get idle\n",
+			port_name(port));
+}
+
 void intel_wait_ddi_buf_idle(struct drm_i915_private *dev_priv,
 			     enum port port)
 {
@@ -196,7 +209,9 @@ static void intel_wait_ddi_buf_active(struct drm_i915_private *dev_priv,
 		return;
 	}
 
-	if (IS_DG2(dev_priv)) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		timeout_us = 10000;
+	} else if (IS_DG2(dev_priv)) {
 		timeout_us = 1200;
 	} else if (DISPLAY_VER(dev_priv) >= 12) {
 		if (intel_phy_is_tc(dev_priv, phy))
@@ -207,8 +222,12 @@ static void intel_wait_ddi_buf_active(struct drm_i915_private *dev_priv,
 		timeout_us = 500;
 	}
 
-	ret = _wait_for(!(intel_de_read(dev_priv, DDI_BUF_CTL(port)) &
-			  DDI_BUF_IS_IDLE), timeout_us, 10, 10);
+	if (DISPLAY_VER(dev_priv) >= 14)
+		ret = _wait_for(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_PORT_BUF_PHY_IDLE),
+				timeout_us, 10, 10);
+	else
+		ret = _wait_for(!(intel_de_read(dev_priv, DDI_BUF_CTL(port)) & DDI_BUF_IS_IDLE),
+				timeout_us, 10, 10);
 
 	if (ret)
 		drm_err(&dev_priv->drm, "Timeout waiting for DDI BUF %c to get active\n",
@@ -313,6 +332,13 @@ static void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder,
 		DDI_PORT_WIDTH(crtc_state->lane_count) |
 		DDI_BUF_TRANS_SELECT(0);
 
+	if (DISPLAY_VER(i915) >= 14) {
+		if (intel_dp_is_uhbr(crtc_state))
+			intel_dp->DP |= DDI_BUF_PORT_DATA_40BIT;
+		else
+			intel_dp->DP |= DDI_BUF_PORT_DATA_10BIT;
+	}
+
 	if (IS_ALDERLAKE_P(i915) && intel_phy_is_tc(i915, phy)) {
 		intel_dp->DP |= ddi_buf_phy_link_rate(crtc_state->port_clock);
 		if (!intel_tc_port_in_tbt_alt_mode(dig_port))
@@ -2309,6 +2335,179 @@ static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state)
 		     OVERLAP_PIXELS_MASK, dss1);
 }
 
+static u8 mtl_get_port_width(u8 lane_count)
+{
+	switch (lane_count) {
+	case 1:
+		return 0;
+	case 2:
+		return 1;
+	case 3:
+		return 4;
+	case 4:
+		return 3;
+	default:
+		MISSING_CASE(lane_count);
+		return 4;
+	}
+}
+
+static void
+mtl_ddi_enable_d2d(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum port port = encoder->port;
+
+	intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), 0,
+		     XELPDP_PORT_BUF_D2D_LINK_ENABLE);
+
+	if (wait_for_us((intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) &
+			 XELPDP_PORT_BUF_D2D_LINK_STATE), 100)) {
+		drm_err(&dev_priv->drm, "Timeout waiting for D2D Link enable for PORT_BUF_CTL %c\n",
+			port_name(port));
+	}
+}
+
+static void mtl_port_buf_ctl_program(struct intel_encoder *encoder,
+				     const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+	enum port port = encoder->port;
+	u32 val;
+
+	val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port));
+	val &= ~XELPDP_PORT_WIDTH_MASK;
+	val |= XELPDP_PORT_WIDTH(mtl_get_port_width(crtc_state->lane_count));
+
+	val &= ~XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK;
+	if (intel_dp_is_uhbr(crtc_state))
+		val |= XELPDP_PORT_BUF_PORT_DATA_40BIT;
+	else
+		val |= XELPDP_PORT_BUF_PORT_DATA_10BIT;
+
+	if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL)
+		val |= XELPDP_PORT_REVERSAL;
+
+	intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val);
+}
+
+static void mtl_port_buf_ctl_io_selection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+	u32 val;
+
+	val = intel_tc_port_in_tbt_alt_mode(dig_port) ?
+	      XELPDP_PORT_BUF_IO_SELECT_TBT : 0;
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port),
+		     XELPDP_PORT_BUF_IO_SELECT_TBT, val);
+}
+
+static void mtl_ddi_pre_enable_dp(struct intel_atomic_state *state,
+				  struct intel_encoder *encoder,
+				  const struct intel_crtc_state *crtc_state,
+				  const struct drm_connector_state *conn_state)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+	bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST);
+
+	intel_dp_set_link_params(intel_dp,
+				 crtc_state->port_clock,
+				 crtc_state->lane_count);
+
+	/*
+	 * We only configure what the register value will be here.  Actual
+	 * enabling happens during link training farther down.
+	 */
+	intel_ddi_init_dp_buf_reg(encoder, crtc_state);
+
+	/*
+	 * 1. Enable Power Wells
+	 *
+	 * This was handled at the beginning of intel_atomic_commit_tail(),
+	 * before we called down into this function.
+	 */
+
+	/* 2. PMdemand was already set */
+
+	/* 3. Select Thunderbolt */
+	mtl_port_buf_ctl_io_selection(encoder);
+
+	/* 4. Enable Panel Power if PPS is required */
+	intel_pps_on(intel_dp);
+
+	/* 5. Enable the port PLL */
+	intel_ddi_enable_clock(encoder, crtc_state);
+
+	/*
+	 * 6.a Configure Transcoder Clock Select to direct the Port clock to the
+	 * Transcoder.
+	 */
+	intel_ddi_enable_transcoder_clock(encoder, crtc_state);
+
+	/*
+	 * 6.b If DP v2.0/128b mode - Configure TRANS_DP2_CTL register settings.
+	 */
+	intel_ddi_config_transcoder_dp2(encoder, crtc_state);
+
+	/*
+	 * 6.c Configure TRANS_DDI_FUNC_CTL DDI Select, DDI Mode Select & MST
+	 * Transport Select
+	 */
+	intel_ddi_config_transcoder_func(encoder, crtc_state);
+
+	/*
+	 * 6.e Program CoG/MSO configuration bits in DSS_CTL1 if selected.
+	 */
+	intel_ddi_mso_configure(crtc_state);
+
+	if (!is_mst)
+		intel_dp_set_power(intel_dp, DP_SET_POWER_D0);
+
+	intel_dp_configure_protocol_converter(intel_dp, crtc_state);
+	intel_dp_sink_set_decompression_state(intel_dp, crtc_state, true);
+	/*
+	 * DDI FEC: "anticipates enabling FEC encoding sets the FEC_READY bit
+	 * in the FEC_CONFIGURATION register to 1 before initiating link
+	 * training
+	 */
+	intel_dp_sink_set_fec_ready(intel_dp, crtc_state);
+
+	intel_dp_check_frl_training(intel_dp);
+	intel_dp_pcon_dsc_configure(intel_dp, crtc_state);
+
+	/*
+	 * 6. The rest of the below are substeps under the bspec's "Enable and
+	 * Train Display Port" step.  Note that steps that are specific to
+	 * MST will be handled by intel_mst_pre_enable_dp() before/after it
+	 * calls into this function.  Also intel_mst_pre_enable_dp() only calls
+	 * us when active_mst_links==0, so any steps designated for "single
+	 * stream or multi-stream master transcoder" can just be performed
+	 * unconditionally here.
+	 *
+	 * mtl_ddi_prepare_link_retrain() that is called by
+	 * intel_dp_start_link_train() will execute steps: 6.d, 6.f, 6.g, 6.h,
+	 * 6.i and 6.j
+	 *
+	 * 6.k Follow DisplayPort specification training sequence (see notes for
+	 *     failure handling)
+	 * 6.m If DisplayPort multi-stream - Set DP_TP_CTL link training to Idle
+	 *     Pattern, wait for 5 idle patterns (DP_TP_STATUS Min_Idles_Sent)
+	 *     (timeout after 800 us)
+	 */
+	intel_dp_start_link_train(intel_dp, crtc_state);
+
+	/* 6.n Set DP_TP_CTL link training to Normal */
+	if (!is_trans_port_sync_mode(crtc_state))
+		intel_dp_stop_link_train(intel_dp, crtc_state);
+
+	/* 6.o Configure and enable FEC if needed */
+	intel_ddi_enable_fec(encoder, crtc_state);
+
+	intel_dsc_dp_pps_write(encoder, crtc_state);
+}
+
 static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state,
 				  struct intel_encoder *encoder,
 				  const struct intel_crtc_state *crtc_state,
@@ -2523,7 +2722,9 @@ static void intel_ddi_pre_enable_dp(struct intel_atomic_state *state,
 		intel_dp_128b132b_sdp_crc16(enc_to_intel_dp(encoder),
 					    crtc_state);
 
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(dev_priv) >= 14)
+		mtl_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state);
+	else if (DISPLAY_VER(dev_priv) >= 12)
 		tgl_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state);
 	else
 		hsw_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state);
@@ -2604,8 +2805,50 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state,
 	}
 }
 
-static void intel_disable_ddi_buf(struct intel_encoder *encoder,
-				  const struct intel_crtc_state *crtc_state)
+static void
+mtl_ddi_disable_d2d_link(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum port port = encoder->port;
+
+	intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port),
+		     XELPDP_PORT_BUF_D2D_LINK_ENABLE, 0);
+
+	if (wait_for_us(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) &
+			  XELPDP_PORT_BUF_D2D_LINK_STATE), 100))
+		drm_err(&dev_priv->drm, "Timeout waiting for D2D Link disable for PORT_BUF_CTL %c\n",
+			port_name(port));
+}
+
+static void mtl_disable_ddi_buf(struct intel_encoder *encoder,
+				const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum port port = encoder->port;
+	u32 val;
+
+	/* 3.b Clear DDI_CTL_DE Enable to 0. */
+	val = intel_de_read(dev_priv, DDI_BUF_CTL(port));
+	if (val & DDI_BUF_CTL_ENABLE) {
+		val &= ~DDI_BUF_CTL_ENABLE;
+		intel_de_write(dev_priv, DDI_BUF_CTL(port), val);
+
+		/* 3.c Poll for PORT_BUF_CTL Idle Status == 1, timeout after 100us */
+		mtl_wait_ddi_buf_idle(dev_priv, port);
+	}
+
+	/* 3.d Disable D2D Link */
+	mtl_ddi_disable_d2d_link(encoder);
+
+	/* 3.e Disable DP_TP_CTL */
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
+		intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
+			     DP_TP_CTL_ENABLE, 0);
+	}
+}
+
+static void disable_ddi_buf(struct intel_encoder *encoder,
+			    const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	enum port port = encoder->port;
@@ -2630,6 +2873,21 @@ static void intel_disable_ddi_buf(struct intel_encoder *encoder,
 		intel_wait_ddi_buf_idle(dev_priv, port);
 }
 
+static void intel_disable_ddi_buf(struct intel_encoder *encoder,
+				  const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		mtl_disable_ddi_buf(encoder, crtc_state);
+
+		/* 3.f Disable DP_TP_CTL FEC Enable if it is needed */
+		intel_ddi_disable_fec_state(encoder, crtc_state);
+	} else {
+		disable_ddi_buf(encoder, crtc_state);
+	}
+}
+
 static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 				      struct intel_encoder *encoder,
 				      const struct intel_crtc_state *old_crtc_state,
@@ -2638,6 +2896,7 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct intel_dp *intel_dp = &dig_port->dp;
+	intel_wakeref_t wakeref;
 	bool is_mst = intel_crtc_has_type(old_crtc_state,
 					  INTEL_OUTPUT_DP_MST);
 
@@ -2677,12 +2936,19 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state,
 	intel_pps_vdd_on(intel_dp);
 	intel_pps_off(intel_dp);
 
-	if (!intel_tc_port_in_tbt_alt_mode(dig_port))
+	wakeref = fetch_and_zero(&dig_port->ddi_io_wakeref);
+
+	if (wakeref)
 		intel_display_power_put(dev_priv,
 					dig_port->ddi_io_power_domain,
-					fetch_and_zero(&dig_port->ddi_io_wakeref));
+					wakeref);
 
 	intel_ddi_disable_clock(encoder);
+
+	/* De-select Thunderbolt */
+	if (DISPLAY_VER(dev_priv) >= 14)
+		intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(encoder->port),
+			     XELPDP_PORT_BUF_IO_SELECT_TBT, 0);
 }
 
 static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state,
@@ -2693,6 +2959,7 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state,
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 	struct intel_hdmi *intel_hdmi = &dig_port->hdmi;
+	intel_wakeref_t wakeref;
 
 	dig_port->set_infoframes(encoder, false,
 				 old_crtc_state, old_conn_state);
@@ -2705,9 +2972,11 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state,
 	if (DISPLAY_VER(dev_priv) >= 12)
 		intel_ddi_disable_transcoder_clock(old_crtc_state);
 
-	intel_display_power_put(dev_priv,
-				dig_port->ddi_io_power_domain,
-				fetch_and_zero(&dig_port->ddi_io_wakeref));
+	wakeref = fetch_and_zero(&dig_port->ddi_io_wakeref);
+	if (wakeref)
+		intel_display_power_put(dev_priv,
+					dig_port->ddi_io_power_domain,
+					wakeref);
 
 	intel_ddi_disable_clock(encoder);
 
@@ -3104,6 +3373,53 @@ static void adlp_tbt_to_dp_alt_switch_wa(struct intel_encoder *encoder)
 		intel_dkl_phy_rmw(i915, DKL_PCS_DW5(tc_port, ln), DKL_PCS_DW5_CORE_SOFTRESET, 0);
 }
 
+static void mtl_ddi_prepare_link_retrain(struct intel_dp *intel_dp,
+					 const struct intel_crtc_state *crtc_state)
+{
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *encoder = &dig_port->base;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum port port = encoder->port;
+	u32 dp_tp_ctl;
+
+	/*
+	 * TODO: To train with only a different voltage swing entry is not
+	 * necessary disable and enable port
+	 */
+	dp_tp_ctl = intel_de_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+	if (dp_tp_ctl & DP_TP_CTL_ENABLE)
+		mtl_disable_ddi_buf(encoder, crtc_state);
+
+	/* 6.d Configure and enable DP_TP_CTL with link training pattern 1 selected */
+	dp_tp_ctl = DP_TP_CTL_ENABLE | DP_TP_CTL_LINK_TRAIN_PAT1;
+	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST)) {
+		dp_tp_ctl |= DP_TP_CTL_MODE_MST;
+	} else {
+		dp_tp_ctl |= DP_TP_CTL_MODE_SST;
+		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
+			dp_tp_ctl |= DP_TP_CTL_ENHANCED_FRAME_ENABLE;
+	}
+	intel_de_write(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), dp_tp_ctl);
+	intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
+
+	/* 6.f Enable D2D Link */
+	mtl_ddi_enable_d2d(encoder);
+
+	/* 6.g Configure voltage swing and related IO settings */
+	encoder->set_signal_levels(encoder, crtc_state);
+
+	/* 6.h Configure PORT_BUF_CTL1 */
+	mtl_port_buf_ctl_program(encoder, crtc_state);
+
+	/* 6.i Configure and enable DDI_CTL_DE to start sending valid data to port slice */
+	intel_dp->DP |= DDI_BUF_CTL_ENABLE;
+	intel_de_write(dev_priv, DDI_BUF_CTL(port), intel_dp->DP);
+	intel_de_posting_read(dev_priv, DDI_BUF_CTL(port));
+
+	/* 6.j Poll for PORT_BUF_CTL Idle Status == 0, timeout after 100 us */
+	intel_wait_ddi_buf_active(dev_priv, port);
+}
+
 static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp,
 					   const struct intel_crtc_state *crtc_state)
 {
@@ -3871,6 +4187,7 @@ static const struct drm_encoder_funcs intel_ddi_funcs = {
 static struct intel_connector *
 intel_ddi_init_dp_connector(struct intel_digital_port *dig_port)
 {
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	struct intel_connector *connector;
 	enum port port = dig_port->base.port;
 
@@ -3879,7 +4196,10 @@ intel_ddi_init_dp_connector(struct intel_digital_port *dig_port)
 		return NULL;
 
 	dig_port->dp.output_reg = DDI_BUF_CTL(port);
-	dig_port->dp.prepare_link_retrain = intel_ddi_prepare_link_retrain;
+	if (DISPLAY_VER(i915) >= 14)
+		dig_port->dp.prepare_link_retrain = mtl_ddi_prepare_link_retrain;
+	else
+		dig_port->dp.prepare_link_retrain = intel_ddi_prepare_link_retrain;
 	dig_port->dp.set_link_train = intel_ddi_set_link_train;
 	dig_port->dp.set_idle_link_train = intel_ddi_set_idle_link_train;
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index de95e81007c22..33bc060112355 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5645,11 +5645,16 @@ enum skl_power_gate {
 /* DDI Buffer Control */
 #define _DDI_BUF_CTL_A				0x64000
 #define _DDI_BUF_CTL_B				0x64100
+/* Known as DDI_CTL_DE in MTL+ */
 #define DDI_BUF_CTL(port) _MMIO_PORT(port, _DDI_BUF_CTL_A, _DDI_BUF_CTL_B)
 #define  DDI_BUF_CTL_ENABLE			(1 << 31)
 #define  DDI_BUF_TRANS_SELECT(n)	((n) << 24)
 #define  DDI_BUF_EMP_MASK			(0xf << 24)
 #define  DDI_BUF_PHY_LINK_RATE(r)		((r) << 20)
+#define  DDI_BUF_PORT_DATA_MASK			REG_GENMASK(19, 18)
+#define  DDI_BUF_PORT_DATA_10BIT		REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 0)
+#define  DDI_BUF_PORT_DATA_20BIT		REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 1)
+#define  DDI_BUF_PORT_DATA_40BIT		REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 2)
 #define  DDI_BUF_PORT_REVERSAL			(1 << 16)
 #define  DDI_BUF_IS_IDLE			(1 << 7)
 #define  DDI_BUF_CTL_TC_PHY_OWNERSHIP		REG_BIT(6)
-- 
GitLab


From 5836bc5f8d3113ccdda2a10fb86344a9f03698ca Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Thu, 13 Apr 2023 14:24:41 -0700
Subject: [PATCH 0042/1791] drm/i915/mtl: Add C10 phy programming for HDMI

Like DG2, we still don't have a proper algorithm that can be used
for calculating PHY settings, but we do have tables of register
values for a handful of the more common link rates. Some support is
better than none, so let's go ahead and add/use these tables when we
can, and also add some logic to hdmi_port_clock_valid() to filter the
modelist to just the modes we can actually support with these link
rates.

Hopefully we'll have a proper / non-encumbered algorithm to calculate
these registers by the time we upstream and we'll be able to replace
this patch with something more general purpose.

Bspec: 64568

v2: Rebasing with Clint's HDMI C10 PLL tables (Mika)
v3: Remove the extra hdmi clock check pruning.

Cc: Imre Deak <imre.deak@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-8-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 610 +++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_cx0_phy.h  |   1 +
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |   2 +
 drivers/gpu/drm/i915/display/intel_hdmi.c     |   5 +-
 4 files changed, 614 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 5ffd661fa5074..d46ff3401e5ef 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -647,6 +647,603 @@ static const struct intel_c10pll_state * const mtl_c10_edp_tables[] = {
 	NULL,
 };
 
+/*
+ * HDMI link rates with 38.4 MHz reference clock.
+ */
+
+static const struct intel_c10pll_state mtl_c10_hdmi_25_2 = {
+	.clock = 25200,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x4,
+	.pll[1] = 0,
+	.pll[2] = 0xB2,
+	.pll[3] = 0,
+	.pll[4] = 0,
+	.pll[5] = 0,
+	.pll[6] = 0,
+	.pll[7] = 0,
+	.pll[8] = 0x20,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0xD,
+	.pll[16] = 0x6,
+	.pll[17] = 0x8F,
+	.pll[18] = 0x84,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_27_0 = {
+	.clock = 27000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34,
+	.pll[1] = 0,
+	.pll[2] = 0xC0,
+	.pll[3] = 0,
+	.pll[4] = 0,
+	.pll[5] = 0,
+	.pll[6] = 0,
+	.pll[7] = 0,
+	.pll[8] = 0x20,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0x80,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0xD,
+	.pll[16] = 0x6,
+	.pll[17] = 0xCF,
+	.pll[18] = 0x84,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_74_25 = {
+	.clock = 74250,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4,
+	.pll[1] = 0,
+	.pll[2] = 0x7A,
+	.pll[3] = 0,
+	.pll[4] = 0,
+	.pll[5] = 0,
+	.pll[6] = 0,
+	.pll[7] = 0,
+	.pll[8] = 0x20,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0x58,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0xB,
+	.pll[16] = 0x6,
+	.pll[17] = 0xF,
+	.pll[18] = 0x85,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_148_5 = {
+	.clock = 148500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4,
+	.pll[1] = 0,
+	.pll[2] = 0x7A,
+	.pll[3] = 0,
+	.pll[4] = 0,
+	.pll[5] = 0,
+	.pll[6] = 0,
+	.pll[7] = 0,
+	.pll[8] = 0x20,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0x58,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0xA,
+	.pll[16] = 0x6,
+	.pll[17] = 0xF,
+	.pll[18] = 0x85,
+	.pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_594 = {
+	.clock = 594000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4,
+	.pll[1] = 0,
+	.pll[2] = 0x7A,
+	.pll[3] = 0,
+	.pll[4] = 0,
+	.pll[5] = 0,
+	.pll[6] = 0,
+	.pll[7] = 0,
+	.pll[8] = 0x20,
+	.pll[9] = 0x1,
+	.pll[10] = 0,
+	.pll[11] = 0,
+	.pll[12] = 0x58,
+	.pll[13] = 0,
+	.pll[14] = 0,
+	.pll[15] = 0x8,
+	.pll[16] = 0x6,
+	.pll[17] = 0xF,
+	.pll[18] = 0x85,
+	.pll[19] = 0x23,
+};
+
+/* Precomputed C10 HDMI PLL tables */
+static const struct intel_c10pll_state mtl_c10_hdmi_27027 = {
+	.clock = 27027,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xC0, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xCC, .pll[12] = 0x9C, .pll[13] = 0xCB, .pll[14] = 0xCC,
+	.pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_28320 = {
+	.clock = 28320,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x04, .pll[1] = 0x00, .pll[2] = 0xCC, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_30240 = {
+	.clock = 30240,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x04, .pll[1] = 0x00, .pll[2] = 0xDC, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_31500 = {
+	.clock = 31500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x62, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xA0, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0C, .pll[16] = 0x09, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_36000 = {
+	.clock = 36000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xC4, .pll[1] = 0x00, .pll[2] = 0x76, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_40000 = {
+	.clock = 40000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x86, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x55, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_49500 = {
+	.clock = 49500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_50000 = {
+	.clock = 50000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xB0, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x2A, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_57284 = {
+	.clock = 57284,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xCE, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x77, .pll[12] = 0x57, .pll[13] = 0x77, .pll[14] = 0x77,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_58000 = {
+	.clock = 58000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD0, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xD5, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_65000 = {
+	.clock = 65000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x66, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xB5, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0B, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_71000 = {
+	.clock = 71000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x72, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xF5, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_74176 = {
+	.clock = 74176,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x44, .pll[12] = 0x44, .pll[13] = 0x44, .pll[14] = 0x44,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_75000 = {
+	.clock = 75000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7C, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_78750 = {
+	.clock = 78750,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x84, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x08, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_85500 = {
+	.clock = 85500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x92, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x10, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_88750 = {
+	.clock = 88750,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0x98, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x72, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x0B, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_106500 = {
+	.clock = 106500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xBC, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xF0, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_108000 = {
+	.clock = 108000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xC0, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x80, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_115500 = {
+	.clock = 115500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD0, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x50, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_119000 = {
+	.clock = 119000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD6, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xF5, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_135000 = {
+	.clock = 135000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x6C, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x50, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0A, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_138500 = {
+	.clock = 138500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x70, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x22, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_147160 = {
+	.clock = 147160,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x78, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xA5, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_148352 = {
+	.clock = 148352,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x44, .pll[12] = 0x44, .pll[13] = 0x44, .pll[14] = 0x44,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_154000 = {
+	.clock = 154000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x80, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x35, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_162000 = {
+	.clock = 162000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x88, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x60, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_167000 = {
+	.clock = 167000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x8C, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0xFA, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_197802 = {
+	.clock = 197802,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x99, .pll[12] = 0x05, .pll[13] = 0x98, .pll[14] = 0x99,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_198000 = {
+	.clock = 198000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_209800 = {
+	.clock = 209800,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xBA, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x45, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_241500 = {
+	.clock = 241500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xDA, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xC8, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_262750 = {
+	.clock = 262750,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x68, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x6C, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x09, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_268500 = {
+	.clock = 268500,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x6A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xEC, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x09, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_296703 = {
+	.clock = 296703,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x33, .pll[12] = 0x44, .pll[13] = 0x33, .pll[14] = 0x33,
+	.pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_297000 = {
+	.clock = 297000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x58, .pll[13] = 0x00, .pll[14] = 0x00,
+	.pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_319750 = {
+	.clock = 319750,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x86, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x44, .pll[13] = 0xA9, .pll[14] = 0xAA,
+	.pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_497750 = {
+	.clock = 497750,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xE2, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x9F, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_592000 = {
+	.clock = 592000,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x15, .pll[13] = 0x55, .pll[14] = 0x55,
+	.pll[15] = 0x08, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state mtl_c10_hdmi_593407 = {
+	.clock = 593407,
+	.tx = 0x10,
+	.cmn = 0x1,
+	.pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00,
+	.pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF,
+	.pll[10] = 0xFF, .pll[11] = 0x3B, .pll[12] = 0x44, .pll[13] = 0xBA, .pll[14] = 0xBB,
+	.pll[15] = 0x08, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23,
+};
+
+static const struct intel_c10pll_state * const mtl_c10_hdmi_tables[] = {
+	&mtl_c10_hdmi_25_2, /* Consolidated Table */
+	&mtl_c10_hdmi_27_0, /* Consolidated Table */
+	&mtl_c10_hdmi_27027,
+	&mtl_c10_hdmi_28320,
+	&mtl_c10_hdmi_30240,
+	&mtl_c10_hdmi_31500,
+	&mtl_c10_hdmi_36000,
+	&mtl_c10_hdmi_40000,
+	&mtl_c10_hdmi_49500,
+	&mtl_c10_hdmi_50000,
+	&mtl_c10_hdmi_57284,
+	&mtl_c10_hdmi_58000,
+	&mtl_c10_hdmi_65000,
+	&mtl_c10_hdmi_71000,
+	&mtl_c10_hdmi_74176,
+	&mtl_c10_hdmi_74_25, /* Consolidated Table */
+	&mtl_c10_hdmi_75000,
+	&mtl_c10_hdmi_78750,
+	&mtl_c10_hdmi_85500,
+	&mtl_c10_hdmi_88750,
+	&mtl_c10_hdmi_106500,
+	&mtl_c10_hdmi_108000,
+	&mtl_c10_hdmi_115500,
+	&mtl_c10_hdmi_119000,
+	&mtl_c10_hdmi_135000,
+	&mtl_c10_hdmi_138500,
+	&mtl_c10_hdmi_147160,
+	&mtl_c10_hdmi_148352,
+	&mtl_c10_hdmi_148_5, /* Consolidated Table */
+	&mtl_c10_hdmi_154000,
+	&mtl_c10_hdmi_162000,
+	&mtl_c10_hdmi_167000,
+	&mtl_c10_hdmi_197802,
+	&mtl_c10_hdmi_198000,
+	&mtl_c10_hdmi_209800,
+	&mtl_c10_hdmi_241500,
+	&mtl_c10_hdmi_262750,
+	&mtl_c10_hdmi_268500,
+	&mtl_c10_hdmi_296703,
+	&mtl_c10_hdmi_297000,
+	&mtl_c10_hdmi_319750,
+	&mtl_c10_hdmi_497750,
+	&mtl_c10_hdmi_592000,
+	&mtl_c10_hdmi_593407,
+	&mtl_c10_hdmi_594, /* Consolidated Table */
+	NULL,
+};
+
+int intel_c10_phy_check_hdmi_link_rate(int clock)
+{
+	const struct intel_c10pll_state * const *tables = mtl_c10_hdmi_tables;
+	int i;
+
+	for (i = 0; tables[i]; i++) {
+		if (clock == tables[i]->clock)
+			return MODE_OK;
+	}
+
+	return MODE_CLOCK_RANGE;
+}
+
 static const struct intel_c10pll_state * const *
 intel_c10pll_tables_get(struct intel_crtc_state *crtc_state,
 			struct intel_encoder *encoder)
@@ -656,9 +1253,10 @@ intel_c10pll_tables_get(struct intel_crtc_state *crtc_state,
 			return mtl_c10_edp_tables;
 		else
 			return mtl_c10_dp_tables;
+	} else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
+		return mtl_c10_hdmi_tables;
 	}
 
-	/* TODO: Add HDMI Support */
 	MISSING_CASE(encoder->type);
 	return NULL;
 }
@@ -758,6 +1356,7 @@ static void intel_c10_pll_program(struct drm_i915_private *i915,
 	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
 		      0, C10_VDR_CTRL_MSGBUS_ACCESS,
 		      MB_WRITE_COMMITTED);
+
 	/* Custom width needs to be programmed to 0 for both the phy lanes */
 	intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CUSTOM_WIDTH,
 		      C10_VDR_CUSTOM_WIDTH_MASK, C10_VDR_CUSTOM_WIDTH_8_10,
@@ -820,7 +1419,8 @@ int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 				 const struct intel_c10pll_state *pll_state)
 {
 	unsigned int frac_quot = 0, frac_rem = 0, frac_den = 1;
-	unsigned int multiplier, tx_clk_div, refclk = 38400;
+	unsigned int multiplier, tx_clk_div, hdmi_div, refclk = 38400;
+	int tmpclk = 0;
 
 	if (pll_state->pll[0] & C10_PLL0_FRACEN) {
 		frac_quot = pll_state->pll[12] << 8 | pll_state->pll[11];
@@ -832,10 +1432,14 @@ int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 		      pll_state->pll[2]) / 2 + 16;
 
 	tx_clk_div = REG_FIELD_GET8(C10_PLL15_TXCLKDIV_MASK, pll_state->pll[15]);
+	hdmi_div = REG_FIELD_GET8(C10_PLL15_HDMIDIV_MASK, pll_state->pll[15]);
 
-	return DIV_ROUND_CLOSEST_ULL(mul_u32_u32(refclk, (multiplier << 16) + frac_quot) +
+	tmpclk = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(refclk, (multiplier << 16) + frac_quot) +
 				     DIV_ROUND_CLOSEST(refclk * frac_rem, frac_den),
 				     10 << (tx_clk_div + 16));
+	tmpclk *= (hdmi_div ? 2 : 1);
+
+	return tmpclk;
 }
 
 static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 6b736acb83e01..509d1d12776eb 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -32,5 +32,6 @@ void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state);
 void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
 				     const struct intel_crtc_state *crtc_state);
+int intel_c10_phy_check_hdmi_link_rate(int clock);
 
 #endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index fe2e3edef69b0..20024622d0eb0 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -145,6 +145,8 @@
 #define   C10_PLL0_FRACEN		REG_BIT8(4)
 #define   C10_PLL3_MULTIPLIERH_MASK	REG_GENMASK8(3, 0)
 #define   C10_PLL15_TXCLKDIV_MASK	REG_GENMASK8(2, 0)
+#define   C10_PLL15_HDMIDIV_MASK	REG_GENMASK8(5, 3)
+
 #define PHY_C10_VDR_CMN(idx)		(0xC20 + (idx))
 #define   C10_CMN0_REF_RANGE		REG_FIELD_PREP(REG_GENMASK(4, 0), 1)
 #define   C10_CMN0_REF_CLK_MPLLB_DIV	REG_FIELD_PREP(REG_GENMASK(7, 5), 1)
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index a73a315ac8ad6..b3f3f761cbb3f 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -45,6 +45,7 @@
 #include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_connector.h"
+#include "intel_cx0_phy.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
@@ -1864,7 +1865,9 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi,
 	 * FIXME: We will hopefully get an algorithmic way of programming
 	 * the MPLLB for HDMI in the future.
 	 */
-	if (IS_DG2(dev_priv))
+	if (IS_METEORLAKE(dev_priv))
+		return intel_c10_phy_check_hdmi_link_rate(clock);
+	else if (IS_DG2(dev_priv))
 		return intel_snps_phy_check_hdmi_link_rate(clock);
 
 	return MODE_OK;
-- 
GitLab


From b66a8abaa48accd3d4b93c1820bbd995fa26ed78 Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 13 Apr 2023 14:24:42 -0700
Subject: [PATCH 0043/1791] drm/i915/display/mtl: Fill port width in
 DDI_BUF_/TRANS_DDI_FUNC_/PORT_BUF_CTL for HDMI

MTL requires the PORT_CTL_WIDTH, TRANS_DDI_FUNC_CTL and DDI_BUF_CTL
to be filled with 4 lanes for TMDS mode.
This patch enables D2D link and fills PORT_WIDTH in appropriate
registers.

v2:
  - Added fixes from Clint's Add HDMI implementation changes.
  - Modified commit message.
v3:
  - Use TRANS_DDI_PORT_WIDTH() instead of DDI_PORT_WIDTH() for the value
    of TRANS_DDI_FUNC_CTL_*. (Gustavo)

Cc: Taylor, Clinton A <clinton.a.taylor@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-9-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 32 ++++++++++++++++++++++--
 drivers/gpu/drm/i915/i915_reg.h          |  2 ++
 2 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 842095ae3e0bc..2ef1af921a0b1 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -541,6 +541,8 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder,
 			temp |= TRANS_DDI_HDMI_SCRAMBLING;
 		if (crtc_state->hdmi_high_tmds_clock_ratio)
 			temp |= TRANS_DDI_HIGH_TMDS_CHAR_RATE;
+		if (DISPLAY_VER(dev_priv) >= 14)
+			temp |= TRANS_DDI_PORT_WIDTH(crtc_state->lane_count);
 	} else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) {
 		temp |= TRANS_DDI_MODE_SELECT_FDI_OR_128B132B;
 		temp |= (crtc_state->fdi_lanes - 1) << 1;
@@ -3157,6 +3159,10 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state,
 	if (has_buf_trans_select(dev_priv))
 		hsw_prepare_hdmi_ddi_buffers(encoder, crtc_state);
 
+	/* e. Enable D2D Link for C10/C20 Phy */
+	if (DISPLAY_VER(dev_priv) >= 14)
+		mtl_ddi_enable_d2d(encoder);
+
 	encoder->set_signal_levels(encoder, crtc_state);
 
 	/* Display WA #1143: skl,kbl,cfl */
@@ -3202,12 +3208,30 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state,
 	 *
 	 * On ADL_P the PHY link rate and lane count must be programmed but
 	 * these are both 0 for HDMI.
+	 *
+	 * But MTL onwards HDMI2.1 is supported and in TMDS mode this
+	 * is filled with lane count, already set in the crtc_state.
+	 * The same is required to be filled in PORT_BUF_CTL for C10/20 Phy.
 	 */
 	buf_ctl = dig_port->saved_port_bits | DDI_BUF_CTL_ENABLE;
-	if (IS_ALDERLAKE_P(dev_priv) && intel_phy_is_tc(dev_priv, phy)) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		u8  lane_count = mtl_get_port_width(crtc_state->lane_count);
+		u32 port_buf = 0;
+
+		port_buf |= XELPDP_PORT_WIDTH(lane_count);
+
+		if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL)
+			port_buf |= XELPDP_PORT_REVERSAL;
+
+		intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port),
+			     XELPDP_PORT_WIDTH_MASK | XELPDP_PORT_REVERSAL, port_buf);
+
+		buf_ctl |= DDI_PORT_WIDTH(lane_count);
+	} else if (IS_ALDERLAKE_P(dev_priv) && intel_phy_is_tc(dev_priv, phy)) {
 		drm_WARN_ON(&dev_priv->drm, !intel_tc_port_in_legacy_mode(dig_port));
 		buf_ctl |= DDI_BUF_CTL_TC_PHY_OWNERSHIP;
 	}
+
 	intel_de_write(dev_priv, DDI_BUF_CTL(port), buf_ctl);
 
 	intel_wait_ddi_buf_active(dev_priv, port);
@@ -3668,7 +3692,11 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder,
 		fallthrough;
 	case TRANS_DDI_MODE_SELECT_DVI:
 		pipe_config->output_types |= BIT(INTEL_OUTPUT_HDMI);
-		pipe_config->lane_count = 4;
+		if (DISPLAY_VER(dev_priv) >= 14)
+			pipe_config->lane_count =
+				((temp & DDI_PORT_WIDTH_MASK) >> DDI_PORT_WIDTH_SHIFT) + 1;
+		else
+			pipe_config->lane_count = 4;
 		break;
 	case TRANS_DDI_MODE_SELECT_DP_SST:
 		if (encoder->type == INTEL_OUTPUT_EDP)
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 33bc060112355..024a92f6bdba7 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5586,6 +5586,8 @@ enum skl_power_gate {
 #define  TRANS_DDI_HDCP_SELECT		REG_BIT(5)
 #define  TRANS_DDI_BFI_ENABLE		(1 << 4)
 #define  TRANS_DDI_HIGH_TMDS_CHAR_RATE	(1 << 4)
+#define  TRANS_DDI_PORT_WIDTH_MASK	REG_GENMASK(3, 1)
+#define  TRANS_DDI_PORT_WIDTH(width)	REG_FIELD_PREP(TRANS_DDI_PORT_WIDTH_MASK, (width) - 1)
 #define  TRANS_DDI_HDMI_SCRAMBLING	(1 << 0)
 #define  TRANS_DDI_HDMI_SCRAMBLING_MASK (TRANS_DDI_HDMI_SCRAMBLER_CTS_ENABLE \
 					| TRANS_DDI_HDMI_SCRAMBLER_RESET_FREQ \
-- 
GitLab


From 764739d8cef28a3c926bb58a63894a162d9997a2 Mon Sep 17 00:00:00 2001
From: Clint Taylor <clinton.a.taylor@intel.com>
Date: Thu, 13 Apr 2023 14:24:43 -0700
Subject: [PATCH 0044/1791] drm/i915/mtl: Initial DDI port setup

Initialization sequences and C10 phy are in place to be able to enable
the first 2 ports of MTL. The other ports use C20 phy that still need
to be properly added. Enable the first ports for now, keeping a TODO
comment about the others.

Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413212443.1504245-10-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 0334565cec824..2d4215862ed31 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7801,7 +7801,11 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
-	if (IS_DG2(dev_priv)) {
+	if (IS_METEORLAKE(dev_priv)) {
+		/* TODO: initialize TC ports as well */
+		intel_ddi_init(dev_priv, PORT_A);
+		intel_ddi_init(dev_priv, PORT_B);
+	} else if (IS_DG2(dev_priv)) {
 		intel_ddi_init(dev_priv, PORT_A);
 		intel_ddi_init(dev_priv, PORT_B);
 		intel_ddi_init(dev_priv, PORT_C);
-- 
GitLab


From adfbae9ffe339eed08d54a4eb87c93f4b35f214b Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 13 Apr 2023 13:03:49 -0700
Subject: [PATCH 0045/1791] drm/i915/gt: Avoid out-of-bounds access when
 loading HuC

When HuC is loaded by GSC, there is no header definition for the kernel
to look at and firmware is just handed to GSC. However when reading the
version, it should still check the size of the blob to guarantee it's not
incurring into out-of-bounds array access.

If firmware is smaller than expected, the following message is now
printed:

	# echo boom > /lib/firmware/i915/dg2_huc_gsc.bin
	# dmesg | grep -i huc
	[drm] GT0: HuC firmware i915/dg2_huc_gsc.bin: invalid size: 5 < 184
	[drm] *ERROR* GT0: HuC firmware i915/dg2_huc_gsc.bin: fetch failed -ENODATA
	...

Even without this change the size, header and signature are still
checked by GSC when loading, so this only avoids the out-of-bounds array
access.

Fixes: a7b516bd981f ("drm/i915/huc: Add fetch support for gsc-loaded HuC binary")
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413200349.3492571-1-lucas.demarchi@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 264c952f777bb..24765c30a0e1d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -488,12 +488,25 @@ static void __force_fw_fetch_failures(struct intel_uc_fw *uc_fw, int e)
 	}
 }
 
-static int check_gsc_manifest(const struct firmware *fw,
+static int check_gsc_manifest(struct intel_gt *gt,
+			      const struct firmware *fw,
 			      struct intel_uc_fw *uc_fw)
 {
 	u32 *dw = (u32 *)fw->data;
-	u32 version_hi = dw[HUC_GSC_VERSION_HI_DW];
-	u32 version_lo = dw[HUC_GSC_VERSION_LO_DW];
+	u32 version_hi, version_lo;
+	size_t min_size;
+
+	/* Check the size of the blob before examining buffer contents */
+	min_size = sizeof(u32) * (HUC_GSC_VERSION_LO_DW + 1);
+	if (unlikely(fw->size < min_size)) {
+		gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n",
+			intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
+			fw->size, min_size);
+		return -ENODATA;
+	}
+
+	version_hi = dw[HUC_GSC_VERSION_HI_DW];
+	version_lo = dw[HUC_GSC_VERSION_LO_DW];
 
 	uc_fw->file_selected.ver.major = FIELD_GET(HUC_GSC_MAJOR_VER_HI_MASK, version_hi);
 	uc_fw->file_selected.ver.minor = FIELD_GET(HUC_GSC_MINOR_VER_HI_MASK, version_hi);
@@ -664,7 +677,7 @@ static int check_fw_header(struct intel_gt *gt,
 		return 0;
 
 	if (uc_fw->loaded_via_gsc)
-		err = check_gsc_manifest(fw, uc_fw);
+		err = check_gsc_manifest(gt, fw, uc_fw);
 	else
 		err = check_ccs_header(gt, fw, uc_fw);
 	if (err)
-- 
GitLab


From 3b6692357f70498f617ea1b31a0378070a0acf1c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 13 Apr 2023 23:06:02 +0300
Subject: [PATCH 0046/1791] drm/i915: Make intel_get_crtc_new_encoder() less
 oopsy
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The point of the WARN was to print something, not oops
straight up. Currently that is precisely what happens
if we can't find the connector for the crtc in the atomic
state. Get the dev pointer from the atomic state instead
of the potentially NULL encoder to avoid that.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413200602.6037-2-ville.syrjala@linux.intel.com
Fixes: 3a47ae201e07 ("drm/i915/display: Make WARN* drm specific where encoder ptr is available")
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 2d4215862ed31..52a5857946aa7 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -959,7 +959,7 @@ intel_get_crtc_new_encoder(const struct intel_atomic_state *state,
 		num_encoders++;
 	}
 
-	drm_WARN(encoder->base.dev, num_encoders != 1,
+	drm_WARN(state->base.dev, num_encoders != 1,
 		 "%d encoders for pipe %c\n",
 		 num_encoders, pipe_name(master_crtc->pipe));
 
-- 
GitLab


From f85f2ce57783913a23f21b9d3d03a27357c07c3a Mon Sep 17 00:00:00 2001
From: Fabio Estevam <festevam@denx.de>
Date: Mon, 3 Apr 2023 23:01:28 -0300
Subject: [PATCH 0047/1791] dt-bindings: display: bridge: ldb: Add an i.MX6SX
 entry

i.MX6SX has a single LVDS port and share a similar LDB_CTRL register
layout with i.MX8MP and i.MX93.

Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404020129.509356-1-festevam@gmail.com
---
 .../devicetree/bindings/display/bridge/fsl,ldb.yaml          | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/fsl,ldb.yaml b/Documentation/devicetree/bindings/display/bridge/fsl,ldb.yaml
index 6e0e3ba9b49ee..07388bf2b90df 100644
--- a/Documentation/devicetree/bindings/display/bridge/fsl,ldb.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/fsl,ldb.yaml
@@ -17,6 +17,7 @@ description: |
 properties:
   compatible:
     enum:
+      - fsl,imx6sx-ldb
       - fsl,imx8mp-ldb
       - fsl,imx93-ldb
 
@@ -64,7 +65,9 @@ allOf:
       properties:
         compatible:
           contains:
-            const: fsl,imx93-ldb
+            enum:
+              - fsl,imx6sx-ldb
+              - fsl,imx93-ldb
     then:
       properties:
         ports:
-- 
GitLab


From ebb7619d3ac85d7aae0a5e8d7038d32211ca7dba Mon Sep 17 00:00:00 2001
From: Fabio Estevam <festevam@denx.de>
Date: Mon, 3 Apr 2023 23:01:29 -0300
Subject: [PATCH 0048/1791] drm/bridge: fsl-ldb: Add i.MX6SX support

i.MX6SX has a single LVDS port and share a similar LDB_CTRL register layout
with i.MX8MP and i.MX93.

There is no LVDS CTRL register on the i.MX6SX, so only write to
this register on the appropriate SoCs.

Add support for the i.MX6SX LDB.

Tested on a imx6sx-sdb board with a Hannstar HSD100PXN1 LVDS panel
and also on a custom i.MX6SX-based board.

Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230404020129.509356-2-festevam@gmail.com
---
 drivers/gpu/drm/bridge/fsl-ldb.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/fsl-ldb.c b/drivers/gpu/drm/bridge/fsl-ldb.c
index 682623369498e..b8e52156b07a9 100644
--- a/drivers/gpu/drm/bridge/fsl-ldb.c
+++ b/drivers/gpu/drm/bridge/fsl-ldb.c
@@ -56,6 +56,7 @@
 #define LVDS_CTRL_VBG_ADJ_MASK			GENMASK(19, 17)
 
 enum fsl_ldb_devtype {
+	IMX6SX_LDB,
 	IMX8MP_LDB,
 	IMX93_LDB,
 };
@@ -64,9 +65,14 @@ struct fsl_ldb_devdata {
 	u32 ldb_ctrl;
 	u32 lvds_ctrl;
 	bool lvds_en_bit;
+	bool single_ctrl_reg;
 };
 
 static const struct fsl_ldb_devdata fsl_ldb_devdata[] = {
+	[IMX6SX_LDB] = {
+		.ldb_ctrl = 0x18,
+		.single_ctrl_reg = true,
+	},
 	[IMX8MP_LDB] = {
 		.ldb_ctrl = 0x5c,
 		.lvds_ctrl = 0x128,
@@ -201,6 +207,9 @@ static void fsl_ldb_atomic_enable(struct drm_bridge *bridge,
 
 	regmap_write(fsl_ldb->regmap, fsl_ldb->devdata->ldb_ctrl, reg);
 
+	if (fsl_ldb->devdata->single_ctrl_reg)
+		return;
+
 	/* Program LVDS_CTRL */
 	reg = LVDS_CTRL_CC_ADJ(2) | LVDS_CTRL_PRE_EMPH_EN |
 	      LVDS_CTRL_PRE_EMPH_ADJ(3) | LVDS_CTRL_VBG_EN;
@@ -226,7 +235,8 @@ static void fsl_ldb_atomic_disable(struct drm_bridge *bridge,
 		regmap_write(fsl_ldb->regmap, fsl_ldb->devdata->lvds_ctrl,
 			     LVDS_CTRL_LVDS_EN);
 	else
-		regmap_write(fsl_ldb->regmap, fsl_ldb->devdata->lvds_ctrl, 0);
+		if (!fsl_ldb->devdata->single_ctrl_reg)
+			regmap_write(fsl_ldb->regmap, fsl_ldb->devdata->lvds_ctrl, 0);
 	regmap_write(fsl_ldb->regmap, fsl_ldb->devdata->ldb_ctrl, 0);
 
 	clk_disable_unprepare(fsl_ldb->clk);
@@ -372,6 +382,8 @@ static void fsl_ldb_remove(struct platform_device *pdev)
 }
 
 static const struct of_device_id fsl_ldb_match[] = {
+	{ .compatible = "fsl,imx6sx-ldb",
+	  .data = &fsl_ldb_devdata[IMX6SX_LDB], },
 	{ .compatible = "fsl,imx8mp-ldb",
 	  .data = &fsl_ldb_devdata[IMX8MP_LDB], },
 	{ .compatible = "fsl,imx93-ldb",
-- 
GitLab


From e920aabf8348de4160301b029b13d72eae0531d5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 14 Apr 2023 22:01:59 +0300
Subject: [PATCH 0049/1791] drm/i915: Make intel_{mpllb,c10pll}_state_verify()
 safer
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

intel_{mpllb,c10pll}_state_verify() blows up if you call them
for a non-modeset/fastset commit on account of the relevant
connector not being part of the overall atomic state.
Currently the state checker only runs for modeset/fastset
commits, but for testing purposes it is sometimes desirable
to run it for other commits too. Check for modeset/fastset
in intel_{mpllb,c10pll}_state_verify() itself to make this safe.

v2: Give the new intel_c10pll_state_verify() the same treatment
    Add comment to explain why we do this

Reviewed-by: Jani Nikula <jani.nikula@intel.com> #v1
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230414190159.7904-1-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 5 +++++
 drivers/gpu/drm/i915/display/intel_snps_phy.c | 5 +++++
 2 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index d46ff3401e5ef..83180074b5127 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -1874,6 +1874,11 @@ void intel_c10pll_state_verify(struct intel_atomic_state *state,
 	if (!new_crtc_state->hw.active)
 		return;
 
+	/* intel_get_crtc_new_encoder() only works for modeset/fastset commits */
+	if (!intel_crtc_needs_modeset(new_crtc_state) &&
+	    !intel_crtc_needs_fastset(new_crtc_state))
+		return;
+
 	encoder = intel_get_crtc_new_encoder(state, new_crtc_state);
 	phy = intel_port_to_phy(i915, encoder->port);
 
diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy.c b/drivers/gpu/drm/i915/display/intel_snps_phy.c
index 1cfb94b5cedbd..a72677bf617b6 100644
--- a/drivers/gpu/drm/i915/display/intel_snps_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_snps_phy.c
@@ -2007,6 +2007,11 @@ void intel_mpllb_state_verify(struct intel_atomic_state *state,
 	if (!new_crtc_state->hw.active)
 		return;
 
+	/* intel_get_crtc_new_encoder() only works for modeset/fastset commits */
+	if (!intel_crtc_needs_modeset(new_crtc_state) &&
+	    !intel_crtc_needs_fastset(new_crtc_state))
+		return;
+
 	encoder = intel_get_crtc_new_encoder(state, new_crtc_state);
 	intel_mpllb_readout_hw_state(encoder, &mpllb_hw_state);
 
-- 
GitLab


From 8bfbdadce85c4c51689da10f39c805a7106d4567 Mon Sep 17 00:00:00 2001
From: Cong Liu <liucong2@kylinos.cn>
Date: Sat, 15 Apr 2023 00:41:09 +0200
Subject: [PATCH 0050/1791] drm/i915: Fix memory leaks in i915 selftests

This patch fixes memory leaks on error escapes in function fake_get_pages

Fixes: c3bfba9a2225 ("drm/i915: Check for integer truncation on scatterlist creation")
Signed-off-by: Cong Liu <liucong2@kylinos.cn>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230414224109.1051922-1-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 5361ce70d3f29..154801f1c4685 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -69,8 +69,10 @@ static int fake_get_pages(struct drm_i915_gem_object *obj)
 
 	rem = round_up(obj->base.size, BIT(31)) >> 31;
 	/* restricted by sg_alloc_table */
-	if (overflows_type(rem, unsigned int))
+	if (overflows_type(rem, unsigned int)) {
+		kfree(pages);
 		return -E2BIG;
+	}
 
 	if (sg_alloc_table(pages, rem, GFP)) {
 		kfree(pages);
-- 
GitLab


From dae2f7b89a8436351a2982d0d96a8a56accca576 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <error27@gmail.com>
Date: Wed, 22 Mar 2023 12:06:55 +0300
Subject: [PATCH 0051/1791] drm/imx/lcdc: fix a NULL vs IS_ERR() bug in probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The devm_drm_dev_alloc() function returns error pointers.  It never
returns NULL.  Fix the check.

Fixes: c87e859cdeb5 ("drm/imx/lcdc: Implement DRM driver for imx25")
Signed-off-by: Dan Carpenter <error27@gmail.com>
Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/d0a1fc55-3ef6-444e-b3ef-fdc937d8d57a@kili.mountain
---
 drivers/gpu/drm/imx/lcdc/imx-lcdc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/imx/lcdc/imx-lcdc.c b/drivers/gpu/drm/imx/lcdc/imx-lcdc.c
index 8e6d457917daf..277ead6a459a4 100644
--- a/drivers/gpu/drm/imx/lcdc/imx-lcdc.c
+++ b/drivers/gpu/drm/imx/lcdc/imx-lcdc.c
@@ -400,8 +400,8 @@ static int imx_lcdc_probe(struct platform_device *pdev)
 
 	lcdc = devm_drm_dev_alloc(dev, &imx_lcdc_drm_driver,
 				  struct imx_lcdc, drm);
-	if (!lcdc)
-		return -ENOMEM;
+	if (IS_ERR(lcdc))
+		return PTR_ERR(lcdc);
 
 	drm = &lcdc->drm;
 
-- 
GitLab


From 80e993988b97fe794f3ec2be6db05fe30f9353c3 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:01 +0200
Subject: [PATCH 0052/1791] drm/gma500: Use
 drm_aperture_remove_conflicting_pci_framebuffers

This one nukes all framebuffers, which is a bit much. In reality
gma500 is igpu and never shipped with anything discrete, so there should
not be any difference.

v2: Unfortunately the framebuffer sits outside of the pci bars for
gma500, and so only using the pci helpers won't be enough. Otoh if we
only use non-pci helper, then we don't get the vga handling, and
subsequent refactoring to untangle these special cases won't work.

It's not pretty, but the simplest fix (since gma500 really is the only
quirky pci driver like this we have) is to just have both calls.

v4:
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-2-tzimmermann@suse.de
---
 drivers/gpu/drm/gma500/psb_drv.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index 2ce96b1b9c748..f1e0eed8fea47 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -422,12 +422,17 @@ static int psb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	/*
 	 * We cannot yet easily find the framebuffer's location in memory. So
-	 * remove all framebuffers here.
+	 * remove all framebuffers here. Note that we still want the pci special
+	 * handling to kick out vgacon.
 	 *
 	 * TODO: Refactor psb_driver_load() to map vdc_reg earlier. Then we
 	 *       might be able to read the framebuffer range from the device.
 	 */
-	ret = drm_aperture_remove_framebuffers(true, &driver);
+	ret = drm_aperture_remove_framebuffers(false, &driver);
+	if (ret)
+		return ret;
+
+	ret = drm_aperture_remove_conflicting_pci_framebuffers(pdev, &driver);
 	if (ret)
 		return ret;
 
-- 
GitLab


From db082219569e3392cd38c9e322151450c855eed4 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:02 +0200
Subject: [PATCH 0053/1791] video/aperture: use generic code to figure out the
 vga default device

Since vgaarb has been promoted to be a core piece of the pci subsystem
we don't have to open code random guesses anymore, we actually know
this in a platform agnostic way, and there's no need for an x86
specific hack. See also commit 1d38fe6ee6a8 ("PCI/VGA: Move vgaarb to
drivers/pci")

This should not result in any functional change, and the non-x86
multi-gpu pci systems are probably rare enough to not matter (I don't
know of any tbh). But it's a nice cleanup, so let's do it.

There's been a few questions on previous iterations on dri-devel and
irc:

- fb_is_primary_device() seems to be yet another implementation of
  this theme, and at least on x86 it checks for both
  vga_default_device OR rom shadowing. There shouldn't ever be a case
  where rom shadowing gives any additional hints about the boot vga
  device, but if there is then the default vga selection in vgaarb
  should probably be fixed. And not special-case checks replicated all
  over.

- Thomas also brought up that on most !x86 systems
  fb_is_primary_device() returns 0, except on sparc/parisc. But these
  2 special cases are about platform specific devices and not pci, so
  shouldn't have any interactions.

- Furthermore fb_is_primary_device() is a bit a red herring since it's
  only used to select the right fbdev driver for fbcon, and not for
  the fw handover dance which the aperture helpers handle. At least
  for x86 we might want to look into unifying them, but that's a
  separate thing.

v2: Extend commit message trying to summarize various discussions.

v4:
- make the test for the primary device easier to read (Javier)
- fix commit message style (i.e., commit 1234 ("..."))
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: linux-pci@vger.kernel.org
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-3-tzimmermann@suse.de
---
 drivers/video/aperture.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index b009468ffdff3..c56553d86fe77 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -328,9 +328,8 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 	resource_size_t base, size;
 	int bar, ret;
 
-#ifdef CONFIG_X86
-	primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
-#endif
+	if (pdev == vga_default_device())
+		primary = true;
 
 	for (bar = 0; bar < PCI_STD_NUM_BARS; ++bar) {
 		if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
-- 
GitLab


From 62aeaeaa1b267c5149abee6b45967a5df3feed58 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:03 +0200
Subject: [PATCH 0054/1791] drm/aperture: Remove primary argument

Only really pci devices have a business setting this - it's for
figuring out whether the legacy vga stuff should be nuked too. And
with the preceding two patches those are all using the pci version of
this.

Which means for all other callers primary == false and we can remove
it now.

v2:
- Reorder to avoid compile fail (Thomas)
- Include gma500, which retained it's called to the non-pci version.

v4:
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Deepak Rawat <drawat.floss@gmail.com>
Cc: Neil Armstrong <neil.armstrong@linaro.org>
Cc: Kevin Hilman <khilman@baylibre.com>
Cc: Jerome Brunet <jbrunet@baylibre.com>
Cc: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Emma Anholt <emma@anholt.net>
Cc: Helge Deller <deller@gmx.de>
Cc: David Airlie <airlied@gmail.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: linux-hyperv@vger.kernel.org
Cc: linux-amlogic@lists.infradead.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-tegra@vger.kernel.org
Cc: linux-fbdev@vger.kernel.org
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Acked-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-4-tzimmermann@suse.de
---
 drivers/gpu/drm/arm/hdlcd_drv.c             |  2 +-
 drivers/gpu/drm/armada/armada_drv.c         |  2 +-
 drivers/gpu/drm/drm_aperture.c              | 11 +++--------
 drivers/gpu/drm/gma500/psb_drv.c            |  2 +-
 drivers/gpu/drm/hyperv/hyperv_drm_drv.c     |  1 -
 drivers/gpu/drm/meson/meson_drv.c           |  2 +-
 drivers/gpu/drm/msm/msm_fbdev.c             |  2 +-
 drivers/gpu/drm/rockchip/rockchip_drm_drv.c |  2 +-
 drivers/gpu/drm/stm/drv.c                   |  2 +-
 drivers/gpu/drm/sun4i/sun4i_drv.c           |  2 +-
 drivers/gpu/drm/tegra/drm.c                 |  2 +-
 drivers/gpu/drm/vc4/vc4_drv.c               |  2 +-
 include/drm/drm_aperture.h                  |  7 +++----
 13 files changed, 16 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c
index 9020bf820bc8d..12f5a2c7f03d5 100644
--- a/drivers/gpu/drm/arm/hdlcd_drv.c
+++ b/drivers/gpu/drm/arm/hdlcd_drv.c
@@ -285,7 +285,7 @@ static int hdlcd_drm_bind(struct device *dev)
 	 */
 	if (hdlcd_read(hdlcd, HDLCD_REG_COMMAND)) {
 		hdlcd_write(hdlcd, HDLCD_REG_COMMAND, 0);
-		drm_aperture_remove_framebuffers(false, &hdlcd_driver);
+		drm_aperture_remove_framebuffers(&hdlcd_driver);
 	}
 
 	drm_mode_config_reset(drm);
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 0643887800b4d..c99ec70783013 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -95,7 +95,7 @@ static int armada_drm_bind(struct device *dev)
 	}
 
 	/* Remove early framebuffers */
-	ret = drm_aperture_remove_framebuffers(false, &armada_drm_driver);
+	ret = drm_aperture_remove_framebuffers(&armada_drm_driver);
 	if (ret) {
 		dev_err(dev, "[" DRM_NAME ":%s] can't kick out simple-fb: %d\n",
 			__func__, ret);
diff --git a/drivers/gpu/drm/drm_aperture.c b/drivers/gpu/drm/drm_aperture.c
index 3b8fdeeafd53a..697cffbfd6037 100644
--- a/drivers/gpu/drm/drm_aperture.c
+++ b/drivers/gpu/drm/drm_aperture.c
@@ -32,17 +32,13 @@
  *
  *	static int remove_conflicting_framebuffers(struct pci_dev *pdev)
  *	{
- *		bool primary = false;
  *		resource_size_t base, size;
  *		int ret;
  *
  *		base = pci_resource_start(pdev, 0);
  *		size = pci_resource_len(pdev, 0);
- *	#ifdef CONFIG_X86
- *		primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW;
- *	#endif
  *
- *		return drm_aperture_remove_conflicting_framebuffers(base, size, primary,
+ *		return drm_aperture_remove_conflicting_framebuffers(base, size,
  *		                                                    &example_driver);
  *	}
  *
@@ -161,7 +157,6 @@ EXPORT_SYMBOL(devm_aperture_acquire_from_firmware);
  * drm_aperture_remove_conflicting_framebuffers - remove existing framebuffers in the given range
  * @base: the aperture's base address in physical memory
  * @size: aperture size in bytes
- * @primary: also kick vga16fb if present
  * @req_driver: requesting DRM driver
  *
  * This function removes graphics device drivers which use the memory range described by
@@ -171,9 +166,9 @@ EXPORT_SYMBOL(devm_aperture_acquire_from_firmware);
  * 0 on success, or a negative errno code otherwise
  */
 int drm_aperture_remove_conflicting_framebuffers(resource_size_t base, resource_size_t size,
-						 bool primary, const struct drm_driver *req_driver)
+						 const struct drm_driver *req_driver)
 {
-	return aperture_remove_conflicting_devices(base, size, primary, req_driver->name);
+	return aperture_remove_conflicting_devices(base, size, false, req_driver->name);
 }
 EXPORT_SYMBOL(drm_aperture_remove_conflicting_framebuffers);
 
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index f1e0eed8fea47..4bb06a89e48da 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -428,7 +428,7 @@ static int psb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * TODO: Refactor psb_driver_load() to map vdc_reg earlier. Then we
 	 *       might be able to read the framebuffer range from the device.
 	 */
-	ret = drm_aperture_remove_framebuffers(false, &driver);
+	ret = drm_aperture_remove_framebuffers(&driver);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_drv.c b/drivers/gpu/drm/hyperv/hyperv_drm_drv.c
index f830d62a5ce60..a7d2c92d6c6a0 100644
--- a/drivers/gpu/drm/hyperv/hyperv_drm_drv.c
+++ b/drivers/gpu/drm/hyperv/hyperv_drm_drv.c
@@ -74,7 +74,6 @@ static int hyperv_setup_vram(struct hyperv_drm_device *hv,
 
 	drm_aperture_remove_conflicting_framebuffers(screen_info.lfb_base,
 						     screen_info.lfb_size,
-						     false,
 						     &hyperv_driver);
 
 	hv->fb_size = (unsigned long)hv->mmio_megabytes * 1024 * 1024;
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index bb72fda9106dd..ca6d1e59e5d9f 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -285,7 +285,7 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 	 * Remove early framebuffers (ie. simplefb). The framebuffer can be
 	 * located anywhere in RAM
 	 */
-	ret = drm_aperture_remove_framebuffers(false, &meson_driver);
+	ret = drm_aperture_remove_framebuffers(&meson_driver);
 	if (ret)
 		goto free_drm;
 
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index d26aa52217ce1..16652a5a7018d 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -155,7 +155,7 @@ struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev)
 	}
 
 	/* the fw fb could be anywhere in memory */
-	ret = drm_aperture_remove_framebuffers(false, dev->driver);
+	ret = drm_aperture_remove_framebuffers(dev->driver);
 	if (ret)
 		goto fini;
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index 6e0788d14c10f..d97f2edc646b7 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -140,7 +140,7 @@ static int rockchip_drm_bind(struct device *dev)
 	int ret;
 
 	/* Remove existing drivers that may own the framebuffer memory. */
-	ret = drm_aperture_remove_framebuffers(false, &rockchip_drm_driver);
+	ret = drm_aperture_remove_framebuffers(&rockchip_drm_driver);
 	if (ret) {
 		DRM_DEV_ERROR(dev,
 			      "Failed to remove existing framebuffers - %d.\n",
diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c
index 422220df7d8cd..cb4404b3ce62c 100644
--- a/drivers/gpu/drm/stm/drv.c
+++ b/drivers/gpu/drm/stm/drv.c
@@ -185,7 +185,7 @@ static int stm_drm_platform_probe(struct platform_device *pdev)
 
 	DRM_DEBUG("%s\n", __func__);
 
-	ret = drm_aperture_remove_framebuffers(false, &drv_driver);
+	ret = drm_aperture_remove_framebuffers(&drv_driver);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/sun4i/sun4i_drv.c b/drivers/gpu/drm/sun4i/sun4i_drv.c
index e49f78a6a8cf5..daa7faf72a4b7 100644
--- a/drivers/gpu/drm/sun4i/sun4i_drv.c
+++ b/drivers/gpu/drm/sun4i/sun4i_drv.c
@@ -98,7 +98,7 @@ static int sun4i_drv_bind(struct device *dev)
 		goto unbind_all;
 
 	/* Remove early framebuffers (ie. simplefb) */
-	ret = drm_aperture_remove_framebuffers(false, &sun4i_drv_driver);
+	ret = drm_aperture_remove_framebuffers(&sun4i_drv_driver);
 	if (ret)
 		goto unbind_all;
 
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 6ca9f396e55be..d11d259f9399f 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -1252,7 +1252,7 @@ static int host1x_drm_probe(struct host1x_device *dev)
 
 	drm_mode_config_reset(drm);
 
-	err = drm_aperture_remove_framebuffers(false, &tegra_drm_driver);
+	err = drm_aperture_remove_framebuffers(&tegra_drm_driver);
 	if (err < 0)
 		goto hub;
 
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index c8bf954042e0a..823395c23cc30 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -350,7 +350,7 @@ static int vc4_drm_bind(struct device *dev)
 			return -EPROBE_DEFER;
 	}
 
-	ret = drm_aperture_remove_framebuffers(false, driver);
+	ret = drm_aperture_remove_framebuffers(driver);
 	if (ret)
 		return ret;
 
diff --git a/include/drm/drm_aperture.h b/include/drm/drm_aperture.h
index 7096703c39493..cbe33b49fd5dc 100644
--- a/include/drm/drm_aperture.h
+++ b/include/drm/drm_aperture.h
@@ -13,14 +13,13 @@ int devm_aperture_acquire_from_firmware(struct drm_device *dev, resource_size_t
 					resource_size_t size);
 
 int drm_aperture_remove_conflicting_framebuffers(resource_size_t base, resource_size_t size,
-						 bool primary, const struct drm_driver *req_driver);
+						 const struct drm_driver *req_driver);
 
 int drm_aperture_remove_conflicting_pci_framebuffers(struct pci_dev *pdev,
 						     const struct drm_driver *req_driver);
 
 /**
  * drm_aperture_remove_framebuffers - remove all existing framebuffers
- * @primary: also kick vga16fb if present
  * @req_driver: requesting DRM driver
  *
  * This function removes all graphics device drivers. Use this function on systems
@@ -30,9 +29,9 @@ int drm_aperture_remove_conflicting_pci_framebuffers(struct pci_dev *pdev,
  * 0 on success, or a negative errno code otherwise
  */
 static inline int
-drm_aperture_remove_framebuffers(bool primary, const struct drm_driver *req_driver)
+drm_aperture_remove_framebuffers(const struct drm_driver *req_driver)
 {
-	return drm_aperture_remove_conflicting_framebuffers(0, (resource_size_t)-1, primary,
+	return drm_aperture_remove_conflicting_framebuffers(0, (resource_size_t)-1,
 							    req_driver);
 }
 
-- 
GitLab


From 7450cd235b45d43ee6f3c235f89e92623458175d Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:04 +0200
Subject: [PATCH 0055/1791] video/aperture: Only kick vgacon when the pdev is
 decoding vga

Otherwise it's a bit silly, and we might throw out the driver for the
screen the user is actually looking at. I haven't found a bug report
for this case yet, but we did get bug reports for the analog case
where we're throwing out the efifb driver.

v2: Flip the check around to make it clear it's a special case for
kicking out the vgacon driver only (Thomas)

v4:
- fixes to commit message
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Link: https://bugzilla.kernel.org/show_bug.cgi?id=216303
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-5-tzimmermann@suse.de
---
 drivers/video/aperture.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index c56553d86fe77..b85163e843f30 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -342,13 +342,15 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 			return ret;
 	}
 
-	/*
-	 * WARNING: Apparently we must kick fbdev drivers before vgacon,
-	 * otherwise the vga fbdev driver falls over.
-	 */
-	ret = vga_remove_vgacon(pdev);
-	if (ret)
-		return ret;
+	if (primary) {
+		/*
+		 * WARNING: Apparently we must kick fbdev drivers before vgacon,
+		 * otherwise the vga fbdev driver falls over.
+		 */
+		ret = vga_remove_vgacon(pdev);
+		if (ret)
+			return ret;
+	}
 
 	return 0;
 
-- 
GitLab


From f1d599d315fb7b7343cddaf365e671aaa8453aca Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:05 +0200
Subject: [PATCH 0056/1791] video/aperture: Move vga handling to pci function

A few reasons for this:

- It's really the only one where this matters. I tried looking around,
  and I didn't find any non-pci vga-compatible controllers for x86
  (since that's the only platform where we had this until a few
  patches ago), where a driver participating in the aperture claim
  dance would interfere.

- I also don't expect that any future bus anytime soon will
  not just look like pci towards the OS, that's been the case for like
  25+ years by now for practically everything (even non non-x86).

- Also it's a bit funny if we have one part of the vga removal in the
  pci function, and the other in the generic one.

v2: Rebase.

v4:
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-6-tzimmermann@suse.de
---
 drivers/video/aperture.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index b85163e843f30..b378cd1d44d04 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -298,14 +298,6 @@ int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t si
 
 	aperture_detach_devices(base, size);
 
-	/*
-	 * If this is the primary adapter, there could be a VGA device
-	 * that consumes the VGA framebuffer I/O range. Remove this device
-	 * as well.
-	 */
-	if (primary)
-		aperture_detach_devices(VGA_FB_PHYS_BASE, VGA_FB_PHYS_SIZE);
-
 	return 0;
 }
 EXPORT_SYMBOL(aperture_remove_conflicting_devices);
@@ -343,6 +335,13 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 	}
 
 	if (primary) {
+		/*
+		 * If this is the primary adapter, there could be a VGA device
+		 * that consumes the VGA framebuffer I/O range. Remove this
+		 * device as well.
+		 */
+		aperture_detach_devices(VGA_FB_PHYS_BASE, VGA_FB_PHYS_SIZE);
+
 		/*
 		 * WARNING: Apparently we must kick fbdev drivers before vgacon,
 		 * otherwise the vga fbdev driver falls over.
-- 
GitLab


From 5fbcc6708fe32ef80122cd2a59ddca9d18b24d6e Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:06 +0200
Subject: [PATCH 0057/1791] video/aperture: Drop primary argument

With the preceding patches it's become defunct. Also I'm about to add
a different boolean argument, so it's better to keep the confusion
down to the absolute minimum.

v2: Since the hypervfb patch got droppped (it's only a pci device for
gen1 vm, not for gen2) there is one leftover user in an actual driver
left to touch.

v4:
- fixes to commit message
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: Wei Liu <wei.liu@kernel.org>
Cc: Dexuan Cui <decui@microsoft.com>
Cc: linux-hyperv@vger.kernel.org
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-7-tzimmermann@suse.de
---
 drivers/gpu/drm/drm_aperture.c  | 2 +-
 drivers/video/aperture.c        | 7 +++----
 drivers/video/fbdev/hyperv_fb.c | 2 +-
 include/linux/aperture.h        | 9 ++++-----
 4 files changed, 9 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/drm_aperture.c b/drivers/gpu/drm/drm_aperture.c
index 697cffbfd6037..5729f3bb43989 100644
--- a/drivers/gpu/drm/drm_aperture.c
+++ b/drivers/gpu/drm/drm_aperture.c
@@ -168,7 +168,7 @@ EXPORT_SYMBOL(devm_aperture_acquire_from_firmware);
 int drm_aperture_remove_conflicting_framebuffers(resource_size_t base, resource_size_t size,
 						 const struct drm_driver *req_driver)
 {
-	return aperture_remove_conflicting_devices(base, size, false, req_driver->name);
+	return aperture_remove_conflicting_devices(base, size, req_driver->name);
 }
 EXPORT_SYMBOL(drm_aperture_remove_conflicting_framebuffers);
 
diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index b378cd1d44d04..a0945027e0dfb 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -43,7 +43,7 @@
  *		base = mem->start;
  *		size = resource_size(mem);
  *
- *		ret = aperture_remove_conflicting_devices(base, size, false, "example");
+ *		ret = aperture_remove_conflicting_devices(base, size, "example");
  *		if (ret)
  *			return ret;
  *
@@ -274,7 +274,6 @@ static void aperture_detach_devices(resource_size_t base, resource_size_t size)
  * aperture_remove_conflicting_devices - remove devices in the given range
  * @base: the aperture's base address in physical memory
  * @size: aperture size in bytes
- * @primary: also kick vga16fb if present; only relevant for VGA devices
  * @name: a descriptive name of the requesting driver
  *
  * This function removes devices that own apertures within @base and @size.
@@ -283,7 +282,7 @@ static void aperture_detach_devices(resource_size_t base, resource_size_t size)
  * 0 on success, or a negative errno code otherwise
  */
 int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t size,
-					bool primary, const char *name)
+					const char *name)
 {
 	/*
 	 * If a driver asked to unregister a platform device registered by
@@ -329,7 +328,7 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 
 		base = pci_resource_start(pdev, bar);
 		size = pci_resource_len(pdev, bar);
-		ret = aperture_remove_conflicting_devices(base, size, primary, name);
+		ret = aperture_remove_conflicting_devices(base, size, name);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c
index ec3f6cf05f8cf..54f433e09ab8d 100644
--- a/drivers/video/fbdev/hyperv_fb.c
+++ b/drivers/video/fbdev/hyperv_fb.c
@@ -1073,7 +1073,7 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
 	info->screen_size = dio_fb_size;
 
 getmem_done:
-	aperture_remove_conflicting_devices(base, size, false, KBUILD_MODNAME);
+	aperture_remove_conflicting_devices(base, size, KBUILD_MODNAME);
 
 	if (gen2vm) {
 		/* framebuffer is reallocated, clear screen_info to avoid misuse from kexec */
diff --git a/include/linux/aperture.h b/include/linux/aperture.h
index 442f15a57cadb..7248727753bed 100644
--- a/include/linux/aperture.h
+++ b/include/linux/aperture.h
@@ -14,7 +14,7 @@ int devm_aperture_acquire_for_platform_device(struct platform_device *pdev,
 					      resource_size_t size);
 
 int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t size,
-					bool primary, const char *name);
+					const char *name);
 
 int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name);
 #else
@@ -26,7 +26,7 @@ static inline int devm_aperture_acquire_for_platform_device(struct platform_devi
 }
 
 static inline int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t size,
-						      bool primary, const char *name)
+						      const char *name)
 {
 	return 0;
 }
@@ -39,7 +39,6 @@ static inline int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev,
 
 /**
  * aperture_remove_all_conflicting_devices - remove all existing framebuffers
- * @primary: also kick vga16fb if present; only relevant for VGA devices
  * @name: a descriptive name of the requesting driver
  *
  * This function removes all graphics device drivers. Use this function on systems
@@ -48,9 +47,9 @@ static inline int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev,
  * Returns:
  * 0 on success, or a negative errno code otherwise
  */
-static inline int aperture_remove_all_conflicting_devices(bool primary, const char *name)
+static inline int aperture_remove_all_conflicting_devices(const char *name)
 {
-	return aperture_remove_conflicting_devices(0, (resource_size_t)-1, primary, name);
+	return aperture_remove_conflicting_devices(0, (resource_size_t)-1, name);
 }
 
 #endif
-- 
GitLab


From 5ae3716cfdcd286268133867f67d0803847acefc Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:07 +0200
Subject: [PATCH 0058/1791] video/aperture: Only remove sysfb on the default
 vga pci device

Instead of calling aperture_remove_conflicting_devices() to remove the
conflicting devices, just call to aperture_detach_devices() to detach
the device that matches the same PCI BAR / aperture range. Since the
former is just a wrapper of the latter plus a sysfb_disable() call,
and now that's done in this function but only for the primary devices.

This fixes a regression introduced by commit ee7a69aa38d8 ("fbdev:
Disable sysfb device registration when removing conflicting FBs"),
where we remove the sysfb when loading a driver for an unrelated pci
device, resulting in the user losing their efifb console or similar.

Note that in practice this only is a problem with the nvidia blob,
because that's the only gpu driver people might install which does not
come with an fbdev driver of it's own. For everyone else the real gpu
driver will restore a working console.

Also note that in the referenced bug there's confusion that this same
bug also happens on amdgpu. But that was just another amdgpu specific
regression, which just happened to happen at roughly the same time and
with the same user-observable symptoms. That bug is fixed now, see
https://bugzilla.kernel.org/show_bug.cgi?id=216331#c15

Note that we should not have any such issues on non-pci multi-gpu
issues, because I could only find two such cases:
- SoC with some external panel over spi or similar. These panel
  drivers do not use drm_aperture_remove_conflicting_framebuffers(),
  so no problem.
- vga+mga, which is a direct console driver and entirely bypasses all
  this.

For the above reasons the cc: stable is just notionally, this patch
will need a backport and that's up to nvidia if they care enough.

v2:
- Explain a bit better why other multi-gpu that aren't pci shouldn't
  have any issues with making all this fully pci specific.

v3
- polish commit message (Javier)

v4:
- Fix commit message style (i.e., commit 1234 ("..."))
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Fixes: ee7a69aa38d8 ("fbdev: Disable sysfb device registration when removing conflicting FBs")
Tested-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216303#c28
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Aaron Plattner <aplattner@nvidia.com>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Helge Deller <deller@gmx.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: <stable@vger.kernel.org> # v5.19+ (if someone else does the backport)
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-8-tzimmermann@suse.de
---
 drivers/video/aperture.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index a0945027e0dfb..fa71f8257eedc 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -322,15 +322,16 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 	if (pdev == vga_default_device())
 		primary = true;
 
+	if (primary)
+		sysfb_disable();
+
 	for (bar = 0; bar < PCI_STD_NUM_BARS; ++bar) {
 		if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM))
 			continue;
 
 		base = pci_resource_start(pdev, bar);
 		size = pci_resource_len(pdev, bar);
-		ret = aperture_remove_conflicting_devices(base, size, name);
-		if (ret)
-			return ret;
+		aperture_detach_devices(base, size);
 	}
 
 	if (primary) {
-- 
GitLab


From 5ca1479cd35d9003040e6ac829380debe89b802b Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 6 Apr 2023 15:21:08 +0200
Subject: [PATCH 0059/1791] fbdev: Simplify fb_is_primary_device for x86

vga_default_device really is supposed to cover all corners, at least
for x86. Additionally checking for rom shadowing should be redundant,
because the bios/fw only does that for the boot vga device.

If this turns out to be wrong, then most likely that's a special case
which should be added to the vgaarb code, not replicated all over.

Patch motived by changes to the aperture helpers, which also have this
open code in a bunch of places, and which are all removed in a
clean-up series. This function here is just for selecting the default
fbdev device for fbcon, but I figured for consistency it might be good
to throw this patch in on top.

Note that the shadow rom check predates vgaarb, which was only wired
up in commit 88674088d10c ("x86: Use vga_default_device() when
determining whether an fb is primary"). That patch doesn't explain why
we still fall back to the shadow rom check.

v4:
- fix commit message style (i.e., commit 1234 ("..."))
- fix Daniel's S-o-b address

v5:
- add back an S-o-b tag with Daniel's Intel address

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Helge Deller <deller@gmx.de>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: x86@kernel.org
Cc: "H. Peter Anvin" <hpa@zytor.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-9-tzimmermann@suse.de
---
 arch/x86/video/fbdev.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/arch/x86/video/fbdev.c b/arch/x86/video/fbdev.c
index 9fd24846d0943..5ec4eafbb9811 100644
--- a/arch/x86/video/fbdev.c
+++ b/arch/x86/video/fbdev.c
@@ -14,26 +14,15 @@
 int fb_is_primary_device(struct fb_info *info)
 {
 	struct device *device = info->device;
-	struct pci_dev *default_device = vga_default_device();
 	struct pci_dev *pci_dev;
-	struct resource *res;
 
 	if (!device || !dev_is_pci(device))
 		return 0;
 
 	pci_dev = to_pci_dev(device);
 
-	if (default_device) {
-		if (pci_dev == default_device)
-			return 1;
-		return 0;
-	}
-
-	res = pci_dev->resource + PCI_ROM_RESOURCE;
-
-	if (res->flags & IORESOURCE_ROM_SHADOW)
+	if (pci_dev == vga_default_device())
 		return 1;
-
 	return 0;
 }
 EXPORT_SYMBOL(fb_is_primary_device);
-- 
GitLab


From 116b1c5a364bcbdc40be64d4f3ec9dbc32e264dd Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 6 Apr 2023 15:21:09 +0200
Subject: [PATCH 0060/1791] video/aperture: Provide a VGA helper for gma500 and
 internal use

The hardware for gma500 is different from the rest, as it uses stolen
framebuffer memory that is not available via PCI BAR. The regular PCI
removal helper cannot detect the framebuffer, while the non-PCI helper
misses possible conflicting VGA devices (i.e., a framebuffer or text
console).

Gma500 therefore calls both helpers to catch all cases. It's confusing
as it implies that there's something about the PCI device that requires
ownership management. The relationship between the PCI device and the
VGA devices is non-obvious. At worst, readers might assume that calling
two functions for clearing aperture ownership is a bug in the driver.

Hence, move the PCI removal helper's code for VGA functionality into
a separate function and call this function from gma500. Documents the
purpose of each call to aperture helpers. The change contains comments
and example code form the discussion at [1].

v5:
	* fix grammar in gma500 comment (Javier)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20230404201842.567344-1-daniel.vetter@ffwll.ch/ # 1
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230406132109.32050-10-tzimmermann@suse.de
---
 drivers/gpu/drm/gma500/psb_drv.c | 48 ++++++++++++++++++--------
 drivers/video/aperture.c         | 58 ++++++++++++++++++++++----------
 include/linux/aperture.h         |  7 ++++
 3 files changed, 81 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index 4bb06a89e48da..8b64f61ffaf9d 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -7,6 +7,7 @@
  *
  **************************************************************************/
 
+#include <linux/aperture.h>
 #include <linux/cpu.h>
 #include <linux/module.h>
 #include <linux/notifier.h>
@@ -19,7 +20,6 @@
 #include <acpi/video.h>
 
 #include <drm/drm.h>
-#include <drm/drm_aperture.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_file.h>
 #include <drm/drm_ioctl.h>
@@ -414,25 +414,45 @@ static int psb_driver_load(struct drm_device *dev, unsigned long flags)
 	return ret;
 }
 
-static int psb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+/*
+ * Hardware for gma500 is a hybrid device, which both acts as a PCI
+ * device (for legacy vga functionality) but also more like an
+ * integrated display on a SoC where the framebuffer simply
+ * resides in main memory and not in a special PCI bar (that
+ * internally redirects to a stolen range of main memory) like all
+ * other integrated PCI display devices implement it.
+ *
+ * To catch all cases we need to remove conflicting firmware devices
+ * for the stolen system memory and for the VGA functionality. As we
+ * currently cannot easily find the framebuffer's location in stolen
+ * memory, we remove all framebuffers here.
+ *
+ * TODO: Refactor psb_driver_load() to map vdc_reg earlier. Then
+ *       we might be able to read the framebuffer range from the
+ *       device.
+ */
+static int gma_remove_conflicting_framebuffers(struct pci_dev *pdev,
+					       const struct drm_driver *req_driver)
 {
-	struct drm_psb_private *dev_priv;
-	struct drm_device *dev;
+	resource_size_t base = 0;
+	resource_size_t size = U32_MAX; /* 4 GiB HW limit */
+	const char *name = req_driver->name;
 	int ret;
 
-	/*
-	 * We cannot yet easily find the framebuffer's location in memory. So
-	 * remove all framebuffers here. Note that we still want the pci special
-	 * handling to kick out vgacon.
-	 *
-	 * TODO: Refactor psb_driver_load() to map vdc_reg earlier. Then we
-	 *       might be able to read the framebuffer range from the device.
-	 */
-	ret = drm_aperture_remove_framebuffers(&driver);
+	ret = aperture_remove_conflicting_devices(base, size, name);
 	if (ret)
 		return ret;
 
-	ret = drm_aperture_remove_conflicting_pci_framebuffers(pdev, &driver);
+	return __aperture_remove_legacy_vga_devices(pdev);
+}
+
+static int psb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct drm_psb_private *dev_priv;
+	struct drm_device *dev;
+	int ret;
+
+	ret = gma_remove_conflicting_framebuffers(pdev, &driver);
 	if (ret)
 		return ret;
 
diff --git a/drivers/video/aperture.c b/drivers/video/aperture.c
index fa71f8257eedc..561be8feca96c 100644
--- a/drivers/video/aperture.c
+++ b/drivers/video/aperture.c
@@ -301,6 +301,37 @@ int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t si
 }
 EXPORT_SYMBOL(aperture_remove_conflicting_devices);
 
+/**
+ * __aperture_remove_legacy_vga_devices - remove legacy VGA devices of a PCI devices
+ * @pdev: PCI device
+ *
+ * This function removes VGA devices provided by @pdev, such as a VGA
+ * framebuffer or a console. This is useful if you have a VGA-compatible
+ * PCI graphics device with framebuffers in non-BAR locations. Drivers
+ * should acquire ownership of those memory areas and afterwards call
+ * this helper to release remaining VGA devices.
+ *
+ * If your hardware has its framebuffers accessible via PCI BARS, use
+ * aperture_remove_conflicting_pci_devices() instead. The function will
+ * release any VGA devices automatically.
+ *
+ * WARNING: Apparently we must remove graphics drivers before calling
+ *          this helper. Otherwise the vga fbdev driver falls over if
+ *          we have vgacon configured.
+ *
+ * Returns:
+ * 0 on success, or a negative errno code otherwise
+ */
+int __aperture_remove_legacy_vga_devices(struct pci_dev *pdev)
+{
+	/* VGA framebuffer */
+	aperture_detach_devices(VGA_FB_PHYS_BASE, VGA_FB_PHYS_SIZE);
+
+	/* VGA textmode console */
+	return vga_remove_vgacon(pdev);
+}
+EXPORT_SYMBOL(__aperture_remove_legacy_vga_devices);
+
 /**
  * aperture_remove_conflicting_pci_devices - remove existing framebuffers for PCI devices
  * @pdev: PCI device
@@ -317,7 +348,7 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 {
 	bool primary = false;
 	resource_size_t base, size;
-	int bar, ret;
+	int bar, ret = 0;
 
 	if (pdev == vga_default_device())
 		primary = true;
@@ -334,24 +365,15 @@ int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *na
 		aperture_detach_devices(base, size);
 	}
 
-	if (primary) {
-		/*
-		 * If this is the primary adapter, there could be a VGA device
-		 * that consumes the VGA framebuffer I/O range. Remove this
-		 * device as well.
-		 */
-		aperture_detach_devices(VGA_FB_PHYS_BASE, VGA_FB_PHYS_SIZE);
-
-		/*
-		 * WARNING: Apparently we must kick fbdev drivers before vgacon,
-		 * otherwise the vga fbdev driver falls over.
-		 */
-		ret = vga_remove_vgacon(pdev);
-		if (ret)
-			return ret;
-	}
+	/*
+	 * If this is the primary adapter, there could be a VGA device
+	 * that consumes the VGA framebuffer I/O range. Remove this
+	 * device as well.
+	 */
+	if (primary)
+		ret = __aperture_remove_legacy_vga_devices(pdev);
 
-	return 0;
+	return ret;
 
 }
 EXPORT_SYMBOL(aperture_remove_conflicting_pci_devices);
diff --git a/include/linux/aperture.h b/include/linux/aperture.h
index 7248727753bed..1a9a88b11584b 100644
--- a/include/linux/aperture.h
+++ b/include/linux/aperture.h
@@ -16,6 +16,8 @@ int devm_aperture_acquire_for_platform_device(struct platform_device *pdev,
 int aperture_remove_conflicting_devices(resource_size_t base, resource_size_t size,
 					const char *name);
 
+int __aperture_remove_legacy_vga_devices(struct pci_dev *pdev);
+
 int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name);
 #else
 static inline int devm_aperture_acquire_for_platform_device(struct platform_device *pdev,
@@ -31,6 +33,11 @@ static inline int aperture_remove_conflicting_devices(resource_size_t base, reso
 	return 0;
 }
 
+static inline int __aperture_remove_legacy_vga_devices(struct pci_dev *pdev)
+{
+	return 0;
+}
+
 static inline int aperture_remove_conflicting_pci_devices(struct pci_dev *pdev, const char *name)
 {
 	return 0;
-- 
GitLab


From e28f6966ac8021d8b677fe3839361db8c9344206 Mon Sep 17 00:00:00 2001
From: Sui Jingfeng <suijingfeng@loongson.cn>
Date: Sun, 16 Apr 2023 22:38:49 +0800
Subject: [PATCH 0061/1791] dma-buf/dma-resv.c: fix a typo
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The dma_resv_wait_timeout() function return a value greater than zero
on success.

Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230416143849.1142779-1-suijingfeng@loongson.cn
---
 drivers/dma-buf/dma-resv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/dma-buf/dma-resv.c b/drivers/dma-buf/dma-resv.c
index 2a594b754af14..b6f71eb00866b 100644
--- a/drivers/dma-buf/dma-resv.c
+++ b/drivers/dma-buf/dma-resv.c
@@ -660,7 +660,7 @@ EXPORT_SYMBOL_GPL(dma_resv_get_singleton);
  * dma_resv_lock() already
  * RETURNS
  * Returns -ERESTARTSYS if interrupted, 0 if the wait timed out, or
- * greater than zer on success.
+ * greater than zero on success.
  */
 long dma_resv_wait_timeout(struct dma_resv *obj, enum dma_resv_usage usage,
 			   bool intr, unsigned long timeout)
-- 
GitLab


From ae52dd7d6d933511c90d129addfc629feff147b5 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:53 +0300
Subject: [PATCH 0062/1791] drm/i915/display: remove
 intel_display_commit_duplicated_state()

This seems like an unnecessary wrapper layer. Removing it will be
helpful later.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a38f08e259221d71314ce6d764431147b0fba218.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 28 +++++++-------------
 1 file changed, 10 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 52a5857946aa7..32e939017b4ee 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -701,20 +701,6 @@ intel_plane_fence_y_offset(const struct intel_plane_state *plane_state)
 	return y;
 }
 
-static int
-intel_display_commit_duplicated_state(struct intel_atomic_state *state,
-				      struct drm_modeset_acquire_ctx *ctx)
-{
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
-	int ret;
-
-	ret = drm_atomic_helper_commit_duplicated_state(&state->base, ctx);
-
-	drm_WARN_ON(&i915->drm, ret == -EDEADLK);
-
-	return ret;
-}
-
 static int
 __intel_display_resume(struct drm_i915_private *i915,
 		       struct drm_atomic_state *state,
@@ -722,7 +708,7 @@ __intel_display_resume(struct drm_i915_private *i915,
 {
 	struct drm_crtc_state *crtc_state;
 	struct drm_crtc *crtc;
-	int i;
+	int ret, i;
 
 	intel_modeset_setup_hw_state(i915, ctx);
 	intel_vga_redisable(i915);
@@ -748,7 +734,11 @@ __intel_display_resume(struct drm_i915_private *i915,
 	if (!HAS_GMCH(i915))
 		to_intel_atomic_state(state)->skip_intermediate_wm = true;
 
-	return intel_display_commit_duplicated_state(to_intel_atomic_state(state), ctx);
+	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
+
+	drm_WARN_ON(&i915->drm, ret == -EDEADLK);
+
+	return ret;
 }
 
 static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv)
@@ -839,10 +829,12 @@ void intel_display_finish_reset(struct drm_i915_private *i915)
 	/* reset doesn't touch the display */
 	if (!gpu_reset_clobbers_display(i915)) {
 		/* for testing only restore the display */
-		ret = intel_display_commit_duplicated_state(to_intel_atomic_state(state), ctx);
-		if (ret)
+		ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
+		if (ret) {
+			drm_WARN_ON(&i915->drm, ret == -EDEADLK);
 			drm_err(&i915->drm,
 				"Restoring old state failed with %i\n", ret);
+		}
 	} else {
 		/*
 		 * The display has been reset as well,
-- 
GitLab


From 77316e7552137e1ed1ac3d719be844f36756eb1c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:54 +0300
Subject: [PATCH 0063/1791] drm/i915/display: start high level display driver
 file

The only way to truly clean up intel_display.[ch] is to move stuff out
of them until there's absolutely nothing left.

Start moving the high level display driver entry points, i.e. functions
called from top level driver code only, to a new file, which we'll call
intel_display_driver.c. The intention is that there's no low-level
display code or details here. This is an in-between layer.

Initially, move intel_display_driver_register() and
intel_display_driver_unregister() there.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e42cc037881a4c6042948a34bd4a9698f9e8487c.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |  1 +
 drivers/gpu/drm/i915/display/intel_display.c  | 53 --------------
 drivers/gpu/drm/i915/display/intel_display.h  |  3 -
 .../drm/i915/display/intel_display_driver.c   | 71 +++++++++++++++++++
 .../drm/i915/display/intel_display_driver.h   | 15 ++++
 drivers/gpu/drm/i915/i915_driver.c            |  1 +
 6 files changed, 88 insertions(+), 56 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_driver.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_driver.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 4ee3b5850dd06..627ec074df09b 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -237,6 +237,7 @@ i915-y += \
 	display/intel_crtc_state_dump.o \
 	display/intel_cursor.o \
 	display/intel_display.o \
+	display/intel_display_driver.o \
 	display/intel_display_power.o \
 	display/intel_display_power_map.o \
 	display/intel_display_power_well.o \
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 32e939017b4ee..41f8f6c18c06f 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -32,7 +32,6 @@
 #include <linux/slab.h>
 #include <linux/string_helpers.h>
 #include <linux/vga_switcheroo.h>
-#include <acpi/video.h>
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_atomic.h>
@@ -70,7 +69,6 @@
 #include "intel_crtc_state_dump.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
-#include "intel_display_debugfs.h"
 #include "intel_display_power.h"
 #include "intel_display_types.h"
 #include "intel_dmc.h"
@@ -8839,57 +8837,6 @@ bool intel_modeset_probe_defer(struct pci_dev *pdev)
 	return false;
 }
 
-void intel_display_driver_register(struct drm_i915_private *i915)
-{
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	/* Must be done after probing outputs */
-	intel_opregion_register(i915);
-	intel_acpi_video_register(i915);
-
-	intel_audio_init(i915);
-
-	intel_display_debugfs_register(i915);
-
-	/*
-	 * Some ports require correctly set-up hpd registers for
-	 * detection to work properly (leading to ghost connected
-	 * connector status), e.g. VGA on gm45.  Hence we can only set
-	 * up the initial fbdev config after hpd irqs are fully
-	 * enabled. We do it last so that the async config cannot run
-	 * before the connectors are registered.
-	 */
-	intel_fbdev_initial_config_async(i915);
-
-	/*
-	 * We need to coordinate the hotplugs with the asynchronous
-	 * fbdev configuration, for which we use the
-	 * fbdev->async_cookie.
-	 */
-	drm_kms_helper_poll_init(&i915->drm);
-}
-
-void intel_display_driver_unregister(struct drm_i915_private *i915)
-{
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	intel_fbdev_unregister(i915);
-	intel_audio_deinit(i915);
-
-	/*
-	 * After flushing the fbdev (incl. a late async config which
-	 * will have delayed queuing of a hotplug event), then flush
-	 * the hotplug events.
-	 */
-	drm_kms_helper_poll_fini(&i915->drm);
-	drm_atomic_helper_shutdown(&i915->drm);
-
-	acpi_video_unregister();
-	intel_opregion_unregister(i915);
-}
-
 bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915)
 {
 	return DISPLAY_VER(i915) >= 6 && i915_vtd_active(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 287159bdeb0d1..e46732d26b7c8 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -518,9 +518,6 @@ void intel_set_plane_visible(struct intel_crtc_state *crtc_state,
 			     bool visible);
 void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state);
 
-void intel_display_driver_register(struct drm_i915_private *i915);
-void intel_display_driver_unregister(struct drm_i915_private *i915);
-
 void intel_update_watermarks(struct drm_i915_private *i915);
 
 /* modesetting */
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
new file mode 100644
index 0000000000000..d4a1893e9218a
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -0,0 +1,71 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2022-2023 Intel Corporation
+ *
+ * High level display driver entry points. This is a layer between top level
+ * driver code and low level display functionality; no low level display code or
+ * details here.
+ */
+
+#include <acpi/video.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_probe_helper.h>
+
+#include "i915_drv.h"
+#include "intel_acpi.h"
+#include "intel_audio.h"
+#include "intel_display_debugfs.h"
+#include "intel_display_driver.h"
+#include "intel_fbdev.h"
+#include "intel_opregion.h"
+
+void intel_display_driver_register(struct drm_i915_private *i915)
+{
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	/* Must be done after probing outputs */
+	intel_opregion_register(i915);
+	intel_acpi_video_register(i915);
+
+	intel_audio_init(i915);
+
+	intel_display_debugfs_register(i915);
+
+	/*
+	 * Some ports require correctly set-up hpd registers for
+	 * detection to work properly (leading to ghost connected
+	 * connector status), e.g. VGA on gm45.  Hence we can only set
+	 * up the initial fbdev config after hpd irqs are fully
+	 * enabled. We do it last so that the async config cannot run
+	 * before the connectors are registered.
+	 */
+	intel_fbdev_initial_config_async(i915);
+
+	/*
+	 * We need to coordinate the hotplugs with the asynchronous
+	 * fbdev configuration, for which we use the
+	 * fbdev->async_cookie.
+	 */
+	drm_kms_helper_poll_init(&i915->drm);
+}
+
+void intel_display_driver_unregister(struct drm_i915_private *i915)
+{
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	intel_fbdev_unregister(i915);
+	intel_audio_deinit(i915);
+
+	/*
+	 * After flushing the fbdev (incl. a late async config which
+	 * will have delayed queuing of a hotplug event), then flush
+	 * the hotplug events.
+	 */
+	drm_kms_helper_poll_fini(&i915->drm);
+	drm_atomic_helper_shutdown(&i915->drm);
+
+	acpi_video_unregister();
+	intel_opregion_unregister(i915);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
new file mode 100644
index 0000000000000..4f6deef5a23ff
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2022-2023 Intel Corporation
+ */
+
+#ifndef __INTEL_DISPLAY_DRIVER_H__
+#define __INTEL_DISPLAY_DRIVER_H__
+
+struct drm_i915_private;
+
+void intel_display_driver_register(struct drm_i915_private *i915);
+void intel_display_driver_unregister(struct drm_i915_private *i915);
+
+#endif /* __INTEL_DISPLAY_DRIVER_H__ */
+
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index e5576c440ca0a..f7ebf499d0ce1 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -48,6 +48,7 @@
 #include "display/intel_acpi.h"
 #include "display/intel_bw.h"
 #include "display/intel_cdclk.h"
+#include "display/intel_display_driver.h"
 #include "display/intel_display_types.h"
 #include "display/intel_dmc.h"
 #include "display/intel_dp.h"
-- 
GitLab


From ff2c80be1a002ae268ee78e3cf183b89c1aef77c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:55 +0300
Subject: [PATCH 0064/1791] drm/i915/display: move intel_modeset_probe_defer()
 to intel_display_driver.[ch]

High level display functionality only called from driver top level code.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/831ed4d0356b51526704269038a6d2d72739e779.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  | 23 -------------------
 drivers/gpu/drm/i915/display/intel_display.h  |  1 -
 .../drm/i915/display/intel_display_driver.c   | 23 +++++++++++++++++++
 .../drm/i915/display/intel_display_driver.h   |  4 ++++
 drivers/gpu/drm/i915/i915_pci.c               |  1 +
 5 files changed, 28 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 41f8f6c18c06f..00faab0eefd56 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -31,7 +31,6 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/string_helpers.h>
-#include <linux/vga_switcheroo.h>
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_atomic.h>
@@ -40,7 +39,6 @@
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
-#include <drm/drm_privacy_screen_consumer.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_rect.h>
 
@@ -8816,27 +8814,6 @@ void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915)
 	intel_bios_driver_remove(i915);
 }
 
-bool intel_modeset_probe_defer(struct pci_dev *pdev)
-{
-	struct drm_privacy_screen *privacy_screen;
-
-	/*
-	 * apple-gmux is needed on dual GPU MacBook Pro
-	 * to probe the panel if we're the inactive GPU.
-	 */
-	if (vga_switcheroo_client_probe_defer(pdev))
-		return true;
-
-	/* If the LCD panel has a privacy-screen, wait for it */
-	privacy_screen = drm_privacy_screen_get(&pdev->dev, NULL);
-	if (IS_ERR(privacy_screen) && PTR_ERR(privacy_screen) == -EPROBE_DEFER)
-		return true;
-
-	drm_privacy_screen_put(privacy_screen);
-
-	return false;
-}
-
 bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915)
 {
 	return DISPLAY_VER(i915) >= 6 && i915_vtd_active(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index e46732d26b7c8..6ff8faa1b5ac8 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -521,7 +521,6 @@ void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state);
 void intel_update_watermarks(struct drm_i915_private *i915);
 
 /* modesetting */
-bool intel_modeset_probe_defer(struct pci_dev *pdev);
 void intel_modeset_init_hw(struct drm_i915_private *i915);
 int intel_modeset_init_noirq(struct drm_i915_private *i915);
 int intel_modeset_init_nogem(struct drm_i915_private *i915);
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index d4a1893e9218a..1386f2001613f 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -7,8 +7,10 @@
  * details here.
  */
 
+#include <linux/vga_switcheroo.h>
 #include <acpi/video.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_privacy_screen_consumer.h>
 #include <drm/drm_probe_helper.h>
 
 #include "i915_drv.h"
@@ -19,6 +21,27 @@
 #include "intel_fbdev.h"
 #include "intel_opregion.h"
 
+bool intel_modeset_probe_defer(struct pci_dev *pdev)
+{
+	struct drm_privacy_screen *privacy_screen;
+
+	/*
+	 * apple-gmux is needed on dual GPU MacBook Pro
+	 * to probe the panel if we're the inactive GPU.
+	 */
+	if (vga_switcheroo_client_probe_defer(pdev))
+		return true;
+
+	/* If the LCD panel has a privacy-screen, wait for it */
+	privacy_screen = drm_privacy_screen_get(&pdev->dev, NULL);
+	if (IS_ERR(privacy_screen) && PTR_ERR(privacy_screen) == -EPROBE_DEFER)
+		return true;
+
+	drm_privacy_screen_put(privacy_screen);
+
+	return false;
+}
+
 void intel_display_driver_register(struct drm_i915_private *i915)
 {
 	if (!HAS_DISPLAY(i915))
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 4f6deef5a23ff..4c18792fcafd1 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -6,8 +6,12 @@
 #ifndef __INTEL_DISPLAY_DRIVER_H__
 #define __INTEL_DISPLAY_DRIVER_H__
 
+#include <linux/types.h>
+
 struct drm_i915_private;
+struct pci_dev;
 
+bool intel_modeset_probe_defer(struct pci_dev *pdev);
 void intel_display_driver_register(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
 
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index bbc4d62e490e5..076d6147635e9 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -27,6 +27,7 @@
 #include <drm/i915_pciids.h>
 
 #include "display/intel_display.h"
+#include "display/intel_display_driver.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_sa_media.h"
 
-- 
GitLab


From 15e4f0b541d4c0b2144955111a1b37b5bfabbf63 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:56 +0300
Subject: [PATCH 0065/1791] drm/i915/display: rename
 intel_modeset_probe_defer() -> intel_display_driver_probe_defer()

Follow the usual naming conventions.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ab5023570d8ae55c0d9c98c78f588e51c0790b6c.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_driver.c | 2 +-
 drivers/gpu/drm/i915/display/intel_display_driver.h | 2 +-
 drivers/gpu/drm/i915/i915_pci.c                     | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 1386f2001613f..882a2586aba40 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -21,7 +21,7 @@
 #include "intel_fbdev.h"
 #include "intel_opregion.h"
 
-bool intel_modeset_probe_defer(struct pci_dev *pdev)
+bool intel_display_driver_probe_defer(struct pci_dev *pdev)
 {
 	struct drm_privacy_screen *privacy_screen;
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 4c18792fcafd1..744117b04ed42 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -11,7 +11,7 @@
 struct drm_i915_private;
 struct pci_dev;
 
-bool intel_modeset_probe_defer(struct pci_dev *pdev);
+bool intel_display_driver_probe_defer(struct pci_dev *pdev);
 void intel_display_driver_register(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
 
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 076d6147635e9..0de4d22a1291d 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1354,7 +1354,7 @@ static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENXIO;
 
 	/* Detect if we need to wait for other drivers early on */
-	if (intel_modeset_probe_defer(pdev))
+	if (intel_display_driver_probe_defer(pdev))
 		return -EPROBE_DEFER;
 
 	err = i915_driver_probe(pdev, ent);
-- 
GitLab


From 40053823baadce4e300cb011ac4e3d16be93bf6a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:57 +0300
Subject: [PATCH 0066/1791] drm/i915/display: move modeset probe/remove
 functions to intel_display_driver.c

High level display functionality only called from driver top level code.

v2:
- Remove unnecessary declarations for intel_sanitize_watermarks() and
  intel_atomic_check() (Gustavo)

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/c00df5cfc233520bc28dd972296197a8a619a6a0.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  | 373 +----------------
 drivers/gpu/drm/i915/display/intel_display.h  |  21 +-
 .../drm/i915/display/intel_display_driver.c   | 374 ++++++++++++++++++
 .../drm/i915/display/intel_display_driver.h   |   7 +
 4 files changed, 406 insertions(+), 369 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 00faab0eefd56..a0236260f4043 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -54,7 +54,6 @@
 #include "i9xx_plane.h"
 #include "i9xx_wm.h"
 #include "icl_dsi.h"
-#include "intel_acpi.h"
 #include "intel_atomic.h"
 #include "intel_atomic_plane.h"
 #include "intel_audio.h"
@@ -67,6 +66,7 @@
 #include "intel_crtc_state_dump.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
+#include "intel_display_driver.h"
 #include "intel_display_power.h"
 #include "intel_display_types.h"
 #include "intel_dmc.h"
@@ -86,11 +86,8 @@
 #include "intel_fdi.h"
 #include "intel_fifo_underrun.h"
 #include "intel_frontbuffer.h"
-#include "intel_gmbus.h"
-#include "intel_hdcp.h"
 #include "intel_hdmi.h"
 #include "intel_hotplug.h"
-#include "intel_hti.h"
 #include "intel_lvds.h"
 #include "intel_lvds_regs.h"
 #include "intel_modeset_setup.h"
@@ -104,7 +101,6 @@
 #include "intel_plane_initial.h"
 #include "intel_pps.h"
 #include "intel_psr.h"
-#include "intel_quirks.h"
 #include "intel_sdvo.h"
 #include "intel_snps_phy.h"
 #include "intel_tc.h"
@@ -174,7 +170,7 @@ int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv,
 	return hpll;
 }
 
-static void intel_update_czclk(struct drm_i915_private *dev_priv)
+void intel_update_czclk(struct drm_i915_private *dev_priv)
 {
 	if (!(IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)))
 		return;
@@ -2582,7 +2578,7 @@ intel_link_compute_m_n(u16 bits_per_pixel, int nlanes,
 		    0x80000);
 }
 
-static void intel_panel_sanitize_ssc(struct drm_i915_private *dev_priv)
+void intel_panel_sanitize_ssc(struct drm_i915_private *dev_priv)
 {
 	/*
 	 * There may be no VBT; and if the BIOS enabled SSC we can
@@ -7235,7 +7231,7 @@ static void intel_atomic_helper_free_state(struct drm_i915_private *dev_priv)
 		drm_atomic_state_put(&state->base);
 }
 
-static void intel_atomic_helper_free_state_worker(struct work_struct *work)
+void intel_atomic_helper_free_state_worker(struct work_struct *work)
 {
 	struct drm_i915_private *dev_priv =
 		container_of(work, typeof(*dev_priv), display.atomic_helper.free_work);
@@ -7583,9 +7579,8 @@ static void intel_atomic_track_fbs(struct intel_atomic_state *state)
 					plane->frontbuffer_bit);
 }
 
-static int intel_atomic_commit(struct drm_device *dev,
-			       struct drm_atomic_state *_state,
-			       bool nonblock)
+int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
+			bool nonblock)
 {
 	struct intel_atomic_state *state = to_intel_atomic_state(_state);
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -7687,19 +7682,6 @@ void intel_plane_destroy(struct drm_plane *plane)
 	kfree(to_intel_plane(plane));
 }
 
-static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv)
-{
-	struct intel_plane *plane;
-
-	for_each_intel_plane(&dev_priv->drm, plane) {
-		struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv,
-							      plane->pipe);
-
-		plane->base.possible_crtcs = drm_crtc_mask(&crtc->base);
-	}
-}
-
-
 int intel_get_pipe_from_crtc_id_ioctl(struct drm_device *dev, void *data,
 				      struct drm_file *file)
 {
@@ -7779,7 +7761,7 @@ static bool intel_ddi_crt_present(struct drm_i915_private *dev_priv)
 	return true;
 }
 
-static void intel_setup_outputs(struct drm_i915_private *dev_priv)
+void intel_setup_outputs(struct drm_i915_private *dev_priv)
 {
 	struct intel_encoder *encoder;
 	bool dpd_is_edp = false;
@@ -8035,9 +8017,8 @@ static int max_dotclock(struct drm_i915_private *i915)
 	return max_dotclock;
 }
 
-static enum drm_mode_status
-intel_mode_valid(struct drm_device *dev,
-		 const struct drm_display_mode *mode)
+enum drm_mode_status intel_mode_valid(struct drm_device *dev,
+				      const struct drm_display_mode *mode)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	int hdisplay_max, htotal_max;
@@ -8177,18 +8158,6 @@ intel_mode_valid_max_plane_size(struct drm_i915_private *dev_priv,
 	return MODE_OK;
 }
 
-static const struct drm_mode_config_funcs intel_mode_funcs = {
-	.fb_create = intel_user_framebuffer_create,
-	.get_format_info = intel_fb_get_format_info,
-	.output_poll_changed = intel_fbdev_output_poll_changed,
-	.mode_valid = intel_mode_valid,
-	.atomic_check = intel_atomic_check,
-	.atomic_commit = intel_atomic_commit,
-	.atomic_state_alloc = intel_atomic_state_alloc,
-	.atomic_state_clear = intel_atomic_state_clear,
-	.atomic_state_free = intel_atomic_state_free,
-};
-
 static const struct intel_display_funcs skl_display_funcs = {
 	.get_pipe_config = hsw_get_pipe_config,
 	.crtc_enable = hsw_crtc_enable,
@@ -8260,21 +8229,7 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv)
 	intel_fdi_init_hook(dev_priv);
 }
 
-void intel_modeset_init_hw(struct drm_i915_private *i915)
-{
-	struct intel_cdclk_state *cdclk_state;
-
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	cdclk_state = to_intel_cdclk_state(i915->display.cdclk.obj.state);
-
-	intel_update_cdclk(i915);
-	intel_cdclk_dump_config(i915, &i915->display.cdclk.hw, "Current CDCLK");
-	cdclk_state->logical = cdclk_state->actual = i915->display.cdclk.hw;
-}
-
-static int intel_initial_commit(struct drm_device *dev)
+int intel_initial_commit(struct drm_device *dev)
 {
 	struct drm_atomic_state *state = NULL;
 	struct drm_modeset_acquire_ctx ctx;
@@ -8345,246 +8300,6 @@ static int intel_initial_commit(struct drm_device *dev)
 	return ret;
 }
 
-static const struct drm_mode_config_helper_funcs intel_mode_config_funcs = {
-	.atomic_commit_setup = drm_dp_mst_atomic_setup_commit,
-};
-
-static void intel_mode_config_init(struct drm_i915_private *i915)
-{
-	struct drm_mode_config *mode_config = &i915->drm.mode_config;
-
-	drm_mode_config_init(&i915->drm);
-	INIT_LIST_HEAD(&i915->display.global.obj_list);
-
-	mode_config->min_width = 0;
-	mode_config->min_height = 0;
-
-	mode_config->preferred_depth = 24;
-	mode_config->prefer_shadow = 1;
-
-	mode_config->funcs = &intel_mode_funcs;
-	mode_config->helper_private = &intel_mode_config_funcs;
-
-	mode_config->async_page_flip = HAS_ASYNC_FLIPS(i915);
-
-	/*
-	 * Maximum framebuffer dimensions, chosen to match
-	 * the maximum render engine surface size on gen4+.
-	 */
-	if (DISPLAY_VER(i915) >= 7) {
-		mode_config->max_width = 16384;
-		mode_config->max_height = 16384;
-	} else if (DISPLAY_VER(i915) >= 4) {
-		mode_config->max_width = 8192;
-		mode_config->max_height = 8192;
-	} else if (DISPLAY_VER(i915) == 3) {
-		mode_config->max_width = 4096;
-		mode_config->max_height = 4096;
-	} else {
-		mode_config->max_width = 2048;
-		mode_config->max_height = 2048;
-	}
-
-	if (IS_I845G(i915) || IS_I865G(i915)) {
-		mode_config->cursor_width = IS_I845G(i915) ? 64 : 512;
-		mode_config->cursor_height = 1023;
-	} else if (IS_I830(i915) || IS_I85X(i915) ||
-		   IS_I915G(i915) || IS_I915GM(i915)) {
-		mode_config->cursor_width = 64;
-		mode_config->cursor_height = 64;
-	} else {
-		mode_config->cursor_width = 256;
-		mode_config->cursor_height = 256;
-	}
-}
-
-static void intel_mode_config_cleanup(struct drm_i915_private *i915)
-{
-	intel_atomic_global_obj_cleanup(i915);
-	drm_mode_config_cleanup(&i915->drm);
-}
-
-/* part #1: call before irq install */
-int intel_modeset_init_noirq(struct drm_i915_private *i915)
-{
-	int ret;
-
-	if (i915_inject_probe_failure(i915))
-		return -ENODEV;
-
-	if (HAS_DISPLAY(i915)) {
-		ret = drm_vblank_init(&i915->drm,
-				      INTEL_NUM_PIPES(i915));
-		if (ret)
-			return ret;
-	}
-
-	intel_bios_init(i915);
-
-	ret = intel_vga_register(i915);
-	if (ret)
-		goto cleanup_bios;
-
-	/* FIXME: completely on the wrong abstraction layer */
-	ret = intel_power_domains_init(i915);
-	if (ret < 0)
-		goto cleanup_vga;
-
-	intel_power_domains_init_hw(i915, false);
-
-	if (!HAS_DISPLAY(i915))
-		return 0;
-
-	intel_dmc_init(i915);
-
-	i915->display.wq.modeset = alloc_ordered_workqueue("i915_modeset", 0);
-	i915->display.wq.flip = alloc_workqueue("i915_flip", WQ_HIGHPRI |
-						WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE);
-
-	intel_mode_config_init(i915);
-
-	ret = intel_cdclk_init(i915);
-	if (ret)
-		goto cleanup_vga_client_pw_domain_dmc;
-
-	ret = intel_color_init(i915);
-	if (ret)
-		goto cleanup_vga_client_pw_domain_dmc;
-
-	ret = intel_dbuf_init(i915);
-	if (ret)
-		goto cleanup_vga_client_pw_domain_dmc;
-
-	ret = intel_bw_init(i915);
-	if (ret)
-		goto cleanup_vga_client_pw_domain_dmc;
-
-	init_llist_head(&i915->display.atomic_helper.free_list);
-	INIT_WORK(&i915->display.atomic_helper.free_work,
-		  intel_atomic_helper_free_state_worker);
-
-	intel_init_quirks(i915);
-
-	intel_fbc_init(i915);
-
-	return 0;
-
-cleanup_vga_client_pw_domain_dmc:
-	intel_dmc_fini(i915);
-	intel_power_domains_driver_remove(i915);
-cleanup_vga:
-	intel_vga_unregister(i915);
-cleanup_bios:
-	intel_bios_driver_remove(i915);
-
-	return ret;
-}
-
-/* part #2: call after irq install, but before gem init */
-int intel_modeset_init_nogem(struct drm_i915_private *i915)
-{
-	struct drm_device *dev = &i915->drm;
-	enum pipe pipe;
-	struct intel_crtc *crtc;
-	int ret;
-
-	if (!HAS_DISPLAY(i915))
-		return 0;
-
-	intel_wm_init(i915);
-
-	intel_panel_sanitize_ssc(i915);
-
-	intel_pps_setup(i915);
-
-	intel_gmbus_setup(i915);
-
-	drm_dbg_kms(&i915->drm, "%d display pipe%s available.\n",
-		    INTEL_NUM_PIPES(i915),
-		    INTEL_NUM_PIPES(i915) > 1 ? "s" : "");
-
-	for_each_pipe(i915, pipe) {
-		ret = intel_crtc_init(i915, pipe);
-		if (ret) {
-			intel_mode_config_cleanup(i915);
-			return ret;
-		}
-	}
-
-	intel_plane_possible_crtcs_init(i915);
-	intel_shared_dpll_init(i915);
-	intel_fdi_pll_freq_update(i915);
-
-	intel_update_czclk(i915);
-	intel_modeset_init_hw(i915);
-	intel_dpll_update_ref_clks(i915);
-
-	intel_hdcp_component_init(i915);
-
-	if (i915->display.cdclk.max_cdclk_freq == 0)
-		intel_update_max_cdclk(i915);
-
-	intel_hti_init(i915);
-
-	/* Just disable it once at startup */
-	intel_vga_disable(i915);
-	intel_setup_outputs(i915);
-
-	drm_modeset_lock_all(dev);
-	intel_modeset_setup_hw_state(i915, dev->mode_config.acquire_ctx);
-	intel_acpi_assign_connector_fwnodes(i915);
-	drm_modeset_unlock_all(dev);
-
-	for_each_intel_crtc(dev, crtc) {
-		if (!to_intel_crtc_state(crtc->base.state)->uapi.active)
-			continue;
-		intel_crtc_initial_plane_config(crtc);
-	}
-
-	/*
-	 * Make sure hardware watermarks really match the state we read out.
-	 * Note that we need to do this after reconstructing the BIOS fb's
-	 * since the watermark calculation done here will use pstate->fb.
-	 */
-	if (!HAS_GMCH(i915))
-		ilk_wm_sanitize(i915);
-
-	return 0;
-}
-
-/* part #3: call after gem init */
-int intel_modeset_init(struct drm_i915_private *i915)
-{
-	int ret;
-
-	if (!HAS_DISPLAY(i915))
-		return 0;
-
-	/*
-	 * Force all active planes to recompute their states. So that on
-	 * mode_setcrtc after probe, all the intel_plane_state variables
-	 * are already calculated and there is no assert_plane warnings
-	 * during bootup.
-	 */
-	ret = intel_initial_commit(&i915->drm);
-	if (ret)
-		drm_dbg_kms(&i915->drm, "Initial modeset failed, %d\n", ret);
-
-	intel_overlay_setup(i915);
-
-	ret = intel_fbdev_init(&i915->drm);
-	if (ret)
-		return ret;
-
-	/* Only enable hotplug handling once the fbdev is fully set up. */
-	intel_hpd_init(i915);
-	intel_hpd_poll_disable(i915);
-
-	skl_watermark_ipc_init(i915);
-
-	return 0;
-}
-
 void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 {
 	struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe);
@@ -8730,7 +8445,7 @@ void intel_display_resume(struct drm_device *dev)
 		drm_atomic_state_put(state);
 }
 
-static void intel_hpd_poll_fini(struct drm_i915_private *i915)
+void intel_hpd_poll_fini(struct drm_i915_private *i915)
 {
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
@@ -8748,72 +8463,6 @@ static void intel_hpd_poll_fini(struct drm_i915_private *i915)
 	drm_connector_list_iter_end(&conn_iter);
 }
 
-/* part #1: call before irq uninstall */
-void intel_modeset_driver_remove(struct drm_i915_private *i915)
-{
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	flush_workqueue(i915->display.wq.flip);
-	flush_workqueue(i915->display.wq.modeset);
-
-	flush_work(&i915->display.atomic_helper.free_work);
-	drm_WARN_ON(&i915->drm, !llist_empty(&i915->display.atomic_helper.free_list));
-
-	/*
-	 * MST topology needs to be suspended so we don't have any calls to
-	 * fbdev after it's finalized. MST will be destroyed later as part of
-	 * drm_mode_config_cleanup()
-	 */
-	intel_dp_mst_suspend(i915);
-}
-
-/* part #2: call after irq uninstall */
-void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
-{
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	/*
-	 * Due to the hpd irq storm handling the hotplug work can re-arm the
-	 * poll handlers. Hence disable polling after hpd handling is shut down.
-	 */
-	intel_hpd_poll_fini(i915);
-
-	/* poll work can call into fbdev, hence clean that up afterwards */
-	intel_fbdev_fini(i915);
-
-	intel_unregister_dsm_handler();
-
-	/* flush any delayed tasks or pending work */
-	flush_scheduled_work();
-
-	intel_hdcp_component_fini(i915);
-
-	intel_mode_config_cleanup(i915);
-
-	intel_overlay_cleanup(i915);
-
-	intel_gmbus_teardown(i915);
-
-	destroy_workqueue(i915->display.wq.flip);
-	destroy_workqueue(i915->display.wq.modeset);
-
-	intel_fbc_cleanup(i915);
-}
-
-/* part #3: call after gem init */
-void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915)
-{
-	intel_dmc_fini(i915);
-
-	intel_power_domains_driver_remove(i915);
-
-	intel_vga_unregister(i915);
-
-	intel_bios_driver_remove(i915);
-}
-
 bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915)
 {
 	return DISPLAY_VER(i915) >= 6 && i915_vtd_active(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 6ff8faa1b5ac8..01c27a4757a57 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -63,6 +63,7 @@ struct intel_power_domain_mask;
 struct intel_remapped_info;
 struct intel_rotation_info;
 struct pci_dev;
+struct work_struct;
 
 
 #define pipe_name(p) ((p) + 'A')
@@ -521,13 +522,6 @@ void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state);
 void intel_update_watermarks(struct drm_i915_private *i915);
 
 /* modesetting */
-void intel_modeset_init_hw(struct drm_i915_private *i915);
-int intel_modeset_init_noirq(struct drm_i915_private *i915);
-int intel_modeset_init_nogem(struct drm_i915_private *i915);
-int intel_modeset_init(struct drm_i915_private *i915);
-void intel_modeset_driver_remove(struct drm_i915_private *i915);
-void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915);
-void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_resume(struct drm_device *dev);
 int intel_modeset_all_pipes(struct intel_atomic_state *state,
 			    const char *reason);
@@ -536,6 +530,19 @@ void intel_modeset_get_crtc_power_domains(struct intel_crtc_state *crtc_state,
 void intel_modeset_put_crtc_power_domains(struct intel_crtc *crtc,
 					  struct intel_power_domain_mask *domains);
 
+/* interface for intel_display_driver.c */
+void intel_setup_outputs(struct drm_i915_private *i915);
+int intel_initial_commit(struct drm_device *dev);
+void intel_panel_sanitize_ssc(struct drm_i915_private *i915);
+void intel_update_czclk(struct drm_i915_private *i915);
+void intel_atomic_helper_free_state_worker(struct work_struct *work);
+enum drm_mode_status intel_mode_valid(struct drm_device *dev,
+				      const struct drm_display_mode *mode);
+int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
+			bool nonblock);
+
+void intel_hpd_poll_fini(struct drm_i915_private *i915);
+
 /* modesetting asserts */
 void assert_transcoder(struct drm_i915_private *dev_priv,
 		       enum transcoder cpu_transcoder, bool state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 882a2586aba40..f7805c6f1b63c 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -9,17 +9,47 @@
 
 #include <linux/vga_switcheroo.h>
 #include <acpi/video.h>
+#include <drm/display/drm_dp_mst_helper.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_mode_config.h>
 #include <drm/drm_privacy_screen_consumer.h>
 #include <drm/drm_probe_helper.h>
+#include <drm/drm_vblank.h>
 
 #include "i915_drv.h"
+#include "i9xx_wm.h"
 #include "intel_acpi.h"
+#include "intel_atomic.h"
 #include "intel_audio.h"
+#include "intel_bios.h"
+#include "intel_bw.h"
+#include "intel_cdclk.h"
+#include "intel_color.h"
+#include "intel_crtc.h"
 #include "intel_display_debugfs.h"
 #include "intel_display_driver.h"
+#include "intel_display_power.h"
+#include "intel_display_types.h"
+#include "intel_dmc.h"
+#include "intel_dp.h"
+#include "intel_dpll_mgr.h"
+#include "intel_fb.h"
+#include "intel_fbc.h"
 #include "intel_fbdev.h"
+#include "intel_fdi.h"
+#include "intel_gmbus.h"
+#include "intel_hdcp.h"
+#include "intel_hotplug.h"
+#include "intel_hti.h"
+#include "intel_modeset_setup.h"
 #include "intel_opregion.h"
+#include "intel_overlay.h"
+#include "intel_plane_initial.h"
+#include "intel_pps.h"
+#include "intel_quirks.h"
+#include "intel_vga.h"
+#include "intel_wm.h"
+#include "skl_watermark.h"
 
 bool intel_display_driver_probe_defer(struct pci_dev *pdev)
 {
@@ -42,6 +72,284 @@ bool intel_display_driver_probe_defer(struct pci_dev *pdev)
 	return false;
 }
 
+void intel_modeset_init_hw(struct drm_i915_private *i915)
+{
+	struct intel_cdclk_state *cdclk_state;
+
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	cdclk_state = to_intel_cdclk_state(i915->display.cdclk.obj.state);
+
+	intel_update_cdclk(i915);
+	intel_cdclk_dump_config(i915, &i915->display.cdclk.hw, "Current CDCLK");
+	cdclk_state->logical = cdclk_state->actual = i915->display.cdclk.hw;
+}
+
+static const struct drm_mode_config_funcs intel_mode_funcs = {
+	.fb_create = intel_user_framebuffer_create,
+	.get_format_info = intel_fb_get_format_info,
+	.output_poll_changed = intel_fbdev_output_poll_changed,
+	.mode_valid = intel_mode_valid,
+	.atomic_check = intel_atomic_check,
+	.atomic_commit = intel_atomic_commit,
+	.atomic_state_alloc = intel_atomic_state_alloc,
+	.atomic_state_clear = intel_atomic_state_clear,
+	.atomic_state_free = intel_atomic_state_free,
+};
+
+static const struct drm_mode_config_helper_funcs intel_mode_config_funcs = {
+	.atomic_commit_setup = drm_dp_mst_atomic_setup_commit,
+};
+
+static void intel_mode_config_init(struct drm_i915_private *i915)
+{
+	struct drm_mode_config *mode_config = &i915->drm.mode_config;
+
+	drm_mode_config_init(&i915->drm);
+	INIT_LIST_HEAD(&i915->display.global.obj_list);
+
+	mode_config->min_width = 0;
+	mode_config->min_height = 0;
+
+	mode_config->preferred_depth = 24;
+	mode_config->prefer_shadow = 1;
+
+	mode_config->funcs = &intel_mode_funcs;
+	mode_config->helper_private = &intel_mode_config_funcs;
+
+	mode_config->async_page_flip = HAS_ASYNC_FLIPS(i915);
+
+	/*
+	 * Maximum framebuffer dimensions, chosen to match
+	 * the maximum render engine surface size on gen4+.
+	 */
+	if (DISPLAY_VER(i915) >= 7) {
+		mode_config->max_width = 16384;
+		mode_config->max_height = 16384;
+	} else if (DISPLAY_VER(i915) >= 4) {
+		mode_config->max_width = 8192;
+		mode_config->max_height = 8192;
+	} else if (DISPLAY_VER(i915) == 3) {
+		mode_config->max_width = 4096;
+		mode_config->max_height = 4096;
+	} else {
+		mode_config->max_width = 2048;
+		mode_config->max_height = 2048;
+	}
+
+	if (IS_I845G(i915) || IS_I865G(i915)) {
+		mode_config->cursor_width = IS_I845G(i915) ? 64 : 512;
+		mode_config->cursor_height = 1023;
+	} else if (IS_I830(i915) || IS_I85X(i915) ||
+		   IS_I915G(i915) || IS_I915GM(i915)) {
+		mode_config->cursor_width = 64;
+		mode_config->cursor_height = 64;
+	} else {
+		mode_config->cursor_width = 256;
+		mode_config->cursor_height = 256;
+	}
+}
+
+static void intel_mode_config_cleanup(struct drm_i915_private *i915)
+{
+	intel_atomic_global_obj_cleanup(i915);
+	drm_mode_config_cleanup(&i915->drm);
+}
+
+static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv)
+{
+	struct intel_plane *plane;
+
+	for_each_intel_plane(&dev_priv->drm, plane) {
+		struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv,
+							      plane->pipe);
+
+		plane->base.possible_crtcs = drm_crtc_mask(&crtc->base);
+	}
+}
+
+/* part #1: call before irq install */
+int intel_modeset_init_noirq(struct drm_i915_private *i915)
+{
+	int ret;
+
+	if (i915_inject_probe_failure(i915))
+		return -ENODEV;
+
+	if (HAS_DISPLAY(i915)) {
+		ret = drm_vblank_init(&i915->drm,
+				      INTEL_NUM_PIPES(i915));
+		if (ret)
+			return ret;
+	}
+
+	intel_bios_init(i915);
+
+	ret = intel_vga_register(i915);
+	if (ret)
+		goto cleanup_bios;
+
+	/* FIXME: completely on the wrong abstraction layer */
+	ret = intel_power_domains_init(i915);
+	if (ret < 0)
+		goto cleanup_vga;
+
+	intel_power_domains_init_hw(i915, false);
+
+	if (!HAS_DISPLAY(i915))
+		return 0;
+
+	intel_dmc_init(i915);
+
+	i915->display.wq.modeset = alloc_ordered_workqueue("i915_modeset", 0);
+	i915->display.wq.flip = alloc_workqueue("i915_flip", WQ_HIGHPRI |
+						WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE);
+
+	intel_mode_config_init(i915);
+
+	ret = intel_cdclk_init(i915);
+	if (ret)
+		goto cleanup_vga_client_pw_domain_dmc;
+
+	ret = intel_color_init(i915);
+	if (ret)
+		goto cleanup_vga_client_pw_domain_dmc;
+
+	ret = intel_dbuf_init(i915);
+	if (ret)
+		goto cleanup_vga_client_pw_domain_dmc;
+
+	ret = intel_bw_init(i915);
+	if (ret)
+		goto cleanup_vga_client_pw_domain_dmc;
+
+	init_llist_head(&i915->display.atomic_helper.free_list);
+	INIT_WORK(&i915->display.atomic_helper.free_work,
+		  intel_atomic_helper_free_state_worker);
+
+	intel_init_quirks(i915);
+
+	intel_fbc_init(i915);
+
+	return 0;
+
+cleanup_vga_client_pw_domain_dmc:
+	intel_dmc_fini(i915);
+	intel_power_domains_driver_remove(i915);
+cleanup_vga:
+	intel_vga_unregister(i915);
+cleanup_bios:
+	intel_bios_driver_remove(i915);
+
+	return ret;
+}
+
+/* part #2: call after irq install, but before gem init */
+int intel_modeset_init_nogem(struct drm_i915_private *i915)
+{
+	struct drm_device *dev = &i915->drm;
+	enum pipe pipe;
+	struct intel_crtc *crtc;
+	int ret;
+
+	if (!HAS_DISPLAY(i915))
+		return 0;
+
+	intel_wm_init(i915);
+
+	intel_panel_sanitize_ssc(i915);
+
+	intel_pps_setup(i915);
+
+	intel_gmbus_setup(i915);
+
+	drm_dbg_kms(&i915->drm, "%d display pipe%s available.\n",
+		    INTEL_NUM_PIPES(i915),
+		    INTEL_NUM_PIPES(i915) > 1 ? "s" : "");
+
+	for_each_pipe(i915, pipe) {
+		ret = intel_crtc_init(i915, pipe);
+		if (ret) {
+			intel_mode_config_cleanup(i915);
+			return ret;
+		}
+	}
+
+	intel_plane_possible_crtcs_init(i915);
+	intel_shared_dpll_init(i915);
+	intel_fdi_pll_freq_update(i915);
+
+	intel_update_czclk(i915);
+	intel_modeset_init_hw(i915);
+	intel_dpll_update_ref_clks(i915);
+
+	intel_hdcp_component_init(i915);
+
+	if (i915->display.cdclk.max_cdclk_freq == 0)
+		intel_update_max_cdclk(i915);
+
+	intel_hti_init(i915);
+
+	/* Just disable it once at startup */
+	intel_vga_disable(i915);
+	intel_setup_outputs(i915);
+
+	drm_modeset_lock_all(dev);
+	intel_modeset_setup_hw_state(i915, dev->mode_config.acquire_ctx);
+	intel_acpi_assign_connector_fwnodes(i915);
+	drm_modeset_unlock_all(dev);
+
+	for_each_intel_crtc(dev, crtc) {
+		if (!to_intel_crtc_state(crtc->base.state)->uapi.active)
+			continue;
+		intel_crtc_initial_plane_config(crtc);
+	}
+
+	/*
+	 * Make sure hardware watermarks really match the state we read out.
+	 * Note that we need to do this after reconstructing the BIOS fb's
+	 * since the watermark calculation done here will use pstate->fb.
+	 */
+	if (!HAS_GMCH(i915))
+		ilk_wm_sanitize(i915);
+
+	return 0;
+}
+
+/* part #3: call after gem init */
+int intel_modeset_init(struct drm_i915_private *i915)
+{
+	int ret;
+
+	if (!HAS_DISPLAY(i915))
+		return 0;
+
+	/*
+	 * Force all active planes to recompute their states. So that on
+	 * mode_setcrtc after probe, all the intel_plane_state variables
+	 * are already calculated and there is no assert_plane warnings
+	 * during bootup.
+	 */
+	ret = intel_initial_commit(&i915->drm);
+	if (ret)
+		drm_dbg_kms(&i915->drm, "Initial modeset failed, %d\n", ret);
+
+	intel_overlay_setup(i915);
+
+	ret = intel_fbdev_init(&i915->drm);
+	if (ret)
+		return ret;
+
+	/* Only enable hotplug handling once the fbdev is fully set up. */
+	intel_hpd_init(i915);
+	intel_hpd_poll_disable(i915);
+
+	skl_watermark_ipc_init(i915);
+
+	return 0;
+}
+
 void intel_display_driver_register(struct drm_i915_private *i915)
 {
 	if (!HAS_DISPLAY(i915))
@@ -73,6 +381,72 @@ void intel_display_driver_register(struct drm_i915_private *i915)
 	drm_kms_helper_poll_init(&i915->drm);
 }
 
+/* part #1: call before irq uninstall */
+void intel_modeset_driver_remove(struct drm_i915_private *i915)
+{
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	flush_workqueue(i915->display.wq.flip);
+	flush_workqueue(i915->display.wq.modeset);
+
+	flush_work(&i915->display.atomic_helper.free_work);
+	drm_WARN_ON(&i915->drm, !llist_empty(&i915->display.atomic_helper.free_list));
+
+	/*
+	 * MST topology needs to be suspended so we don't have any calls to
+	 * fbdev after it's finalized. MST will be destroyed later as part of
+	 * drm_mode_config_cleanup()
+	 */
+	intel_dp_mst_suspend(i915);
+}
+
+/* part #2: call after irq uninstall */
+void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
+{
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	/*
+	 * Due to the hpd irq storm handling the hotplug work can re-arm the
+	 * poll handlers. Hence disable polling after hpd handling is shut down.
+	 */
+	intel_hpd_poll_fini(i915);
+
+	/* poll work can call into fbdev, hence clean that up afterwards */
+	intel_fbdev_fini(i915);
+
+	intel_unregister_dsm_handler();
+
+	/* flush any delayed tasks or pending work */
+	flush_scheduled_work();
+
+	intel_hdcp_component_fini(i915);
+
+	intel_mode_config_cleanup(i915);
+
+	intel_overlay_cleanup(i915);
+
+	intel_gmbus_teardown(i915);
+
+	destroy_workqueue(i915->display.wq.flip);
+	destroy_workqueue(i915->display.wq.modeset);
+
+	intel_fbc_cleanup(i915);
+}
+
+/* part #3: call after gem init */
+void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915)
+{
+	intel_dmc_fini(i915);
+
+	intel_power_domains_driver_remove(i915);
+
+	intel_vga_unregister(i915);
+
+	intel_bios_driver_remove(i915);
+}
+
 void intel_display_driver_unregister(struct drm_i915_private *i915)
 {
 	if (!HAS_DISPLAY(i915))
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 744117b04ed42..947b666575ee2 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -12,7 +12,14 @@ struct drm_i915_private;
 struct pci_dev;
 
 bool intel_display_driver_probe_defer(struct pci_dev *pdev);
+void intel_modeset_init_hw(struct drm_i915_private *i915);
+int intel_modeset_init_noirq(struct drm_i915_private *i915);
+int intel_modeset_init_nogem(struct drm_i915_private *i915);
+int intel_modeset_init(struct drm_i915_private *i915);
 void intel_display_driver_register(struct drm_i915_private *i915);
+void intel_modeset_driver_remove(struct drm_i915_private *i915);
+void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915);
+void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
 
 #endif /* __INTEL_DISPLAY_DRIVER_H__ */
-- 
GitLab


From 86a1758d751de03e8f3d8810fe22eaf571798871 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:58 +0300
Subject: [PATCH 0067/1791] drm/i915/display: rename intel_display_driver_*
 functions

Follow the usual naming conventions.

v2:
- Also rename references in comments (Gustavo)

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/6ff7b76018bf4cf611037d7bf027c975cddfe2af.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  |  2 +-
 .../drm/i915/display/intel_display_driver.c   | 16 ++++++------
 .../drm/i915/display/intel_display_driver.h   | 14 +++++------
 drivers/gpu/drm/i915/i915_driver.c            | 25 ++++++++++---------
 drivers/gpu/drm/i915/i915_irq.c               |  2 +-
 5 files changed, 30 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index a0236260f4043..8abce9a9a9000 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -833,7 +833,7 @@ void intel_display_finish_reset(struct drm_i915_private *i915)
 		 * so need a full re-initialization.
 		 */
 		intel_pps_unlock_regs_wa(i915);
-		intel_modeset_init_hw(i915);
+		intel_display_driver_init_hw(i915);
 		intel_clock_gating_init(i915);
 		intel_hpd_init(i915);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index f7805c6f1b63c..45e4251003852 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -72,7 +72,7 @@ bool intel_display_driver_probe_defer(struct pci_dev *pdev)
 	return false;
 }
 
-void intel_modeset_init_hw(struct drm_i915_private *i915)
+void intel_display_driver_init_hw(struct drm_i915_private *i915)
 {
 	struct intel_cdclk_state *cdclk_state;
 
@@ -170,7 +170,7 @@ static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv)
 }
 
 /* part #1: call before irq install */
-int intel_modeset_init_noirq(struct drm_i915_private *i915)
+int intel_display_driver_probe_noirq(struct drm_i915_private *i915)
 {
 	int ret;
 
@@ -246,7 +246,7 @@ int intel_modeset_init_noirq(struct drm_i915_private *i915)
 }
 
 /* part #2: call after irq install, but before gem init */
-int intel_modeset_init_nogem(struct drm_i915_private *i915)
+int intel_display_driver_probe_nogem(struct drm_i915_private *i915)
 {
 	struct drm_device *dev = &i915->drm;
 	enum pipe pipe;
@@ -281,7 +281,7 @@ int intel_modeset_init_nogem(struct drm_i915_private *i915)
 	intel_fdi_pll_freq_update(i915);
 
 	intel_update_czclk(i915);
-	intel_modeset_init_hw(i915);
+	intel_display_driver_init_hw(i915);
 	intel_dpll_update_ref_clks(i915);
 
 	intel_hdcp_component_init(i915);
@@ -318,7 +318,7 @@ int intel_modeset_init_nogem(struct drm_i915_private *i915)
 }
 
 /* part #3: call after gem init */
-int intel_modeset_init(struct drm_i915_private *i915)
+int intel_display_driver_probe(struct drm_i915_private *i915)
 {
 	int ret;
 
@@ -382,7 +382,7 @@ void intel_display_driver_register(struct drm_i915_private *i915)
 }
 
 /* part #1: call before irq uninstall */
-void intel_modeset_driver_remove(struct drm_i915_private *i915)
+void intel_display_driver_remove(struct drm_i915_private *i915)
 {
 	if (!HAS_DISPLAY(i915))
 		return;
@@ -402,7 +402,7 @@ void intel_modeset_driver_remove(struct drm_i915_private *i915)
 }
 
 /* part #2: call after irq uninstall */
-void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
+void intel_display_driver_remove_noirq(struct drm_i915_private *i915)
 {
 	if (!HAS_DISPLAY(i915))
 		return;
@@ -436,7 +436,7 @@ void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
 }
 
 /* part #3: call after gem init */
-void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915)
+void intel_display_driver_remove_nogem(struct drm_i915_private *i915)
 {
 	intel_dmc_fini(i915);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 947b666575ee2..aab498617b90e 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -12,14 +12,14 @@ struct drm_i915_private;
 struct pci_dev;
 
 bool intel_display_driver_probe_defer(struct pci_dev *pdev);
-void intel_modeset_init_hw(struct drm_i915_private *i915);
-int intel_modeset_init_noirq(struct drm_i915_private *i915);
-int intel_modeset_init_nogem(struct drm_i915_private *i915);
-int intel_modeset_init(struct drm_i915_private *i915);
+void intel_display_driver_init_hw(struct drm_i915_private *i915);
+int intel_display_driver_probe_noirq(struct drm_i915_private *i915);
+int intel_display_driver_probe_nogem(struct drm_i915_private *i915);
+int intel_display_driver_probe(struct drm_i915_private *i915);
 void intel_display_driver_register(struct drm_i915_private *i915);
-void intel_modeset_driver_remove(struct drm_i915_private *i915);
-void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915);
-void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915);
+void intel_display_driver_remove(struct drm_i915_private *i915);
+void intel_display_driver_remove_noirq(struct drm_i915_private *i915);
+void intel_display_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
 
 #endif /* __INTEL_DISPLAY_DRIVER_H__ */
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index f7ebf499d0ce1..d38ce3885c4d1 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -743,7 +743,7 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent)
  * @ent: matching PCI ID entry
  *
  * The driver probe routine has to do several things:
- *   - drive output discovery via intel_modeset_init()
+ *   - drive output discovery via intel_display_driver_probe()
  *   - initialize the memory manager
  *   - allocate initial config memory
  *   - setup the DRM framebuffer with the allocated memory
@@ -781,7 +781,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret < 0)
 		goto out_cleanup_mmio;
 
-	ret = intel_modeset_init_noirq(i915);
+	ret = intel_display_driver_probe_noirq(i915);
 	if (ret < 0)
 		goto out_cleanup_hw;
 
@@ -789,7 +789,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ret)
 		goto out_cleanup_modeset;
 
-	ret = intel_modeset_init_nogem(i915);
+	ret = intel_display_driver_probe_nogem(i915);
 	if (ret)
 		goto out_cleanup_irq;
 
@@ -799,7 +799,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	intel_pxp_init(i915);
 
-	ret = intel_modeset_init(i915);
+	ret = intel_display_driver_probe(i915);
 	if (ret)
 		goto out_cleanup_gem;
 
@@ -819,14 +819,14 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i915_gem_driver_release(i915);
 out_cleanup_modeset2:
 	/* FIXME clean up the error path */
-	intel_modeset_driver_remove(i915);
+	intel_display_driver_remove(i915);
 	intel_irq_uninstall(i915);
-	intel_modeset_driver_remove_noirq(i915);
+	intel_display_driver_remove_noirq(i915);
 	goto out_cleanup_modeset;
 out_cleanup_irq:
 	intel_irq_uninstall(i915);
 out_cleanup_modeset:
-	intel_modeset_driver_remove_nogem(i915);
+	intel_display_driver_remove_nogem(i915);
 out_cleanup_hw:
 	i915_driver_hw_remove(i915);
 	intel_memory_regions_driver_release(i915);
@@ -862,16 +862,16 @@ void i915_driver_remove(struct drm_i915_private *i915)
 
 	intel_gvt_driver_remove(i915);
 
-	intel_modeset_driver_remove(i915);
+	intel_display_driver_remove(i915);
 
 	intel_irq_uninstall(i915);
 
-	intel_modeset_driver_remove_noirq(i915);
+	intel_display_driver_remove_noirq(i915);
 
 	i915_reset_error_state(i915);
 	i915_gem_driver_remove(i915);
 
-	intel_modeset_driver_remove_nogem(i915);
+	intel_display_driver_remove_nogem(i915);
 
 	i915_driver_hw_remove(i915);
 
@@ -1224,7 +1224,7 @@ static int i915_drm_resume(struct drm_device *dev)
 	 *
 	 * drm_mode_config_reset() needs AUX interrupts.
 	 *
-	 * Modeset enabling in intel_modeset_init_hw() also needs working
+	 * Modeset enabling in intel_display_driver_init_hw() also needs working
 	 * interrupts.
 	 */
 	intel_runtime_pm_enable_interrupts(dev_priv);
@@ -1234,7 +1234,8 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	i915_gem_resume(dev_priv);
 
-	intel_modeset_init_hw(dev_priv);
+	intel_display_driver_init_hw(dev_priv);
+
 	intel_clock_gating_init(dev_priv);
 	intel_hpd_init(dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b4dd6a5a536f9..dea1a117f3fa1 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -4360,7 +4360,7 @@ void intel_irq_uninstall(struct drm_i915_private *dev_priv)
 	/*
 	 * FIXME we can get called twice during driver probe
 	 * error handling as well as during driver remove due to
-	 * intel_modeset_driver_remove() calling us out of sequence.
+	 * intel_display_driver_remove() calling us out of sequence.
 	 * Would be nice if it didn't do that...
 	 */
 	if (!dev_priv->irq_enabled)
-- 
GitLab


From 59c6106e274dbafd9d25357585ae5ede4b6673dd Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:41:59 +0300
Subject: [PATCH 0068/1791] drm/i915/display: add intel_display_reset.[ch]

Split out the display reset functionality to a separate file to
declutter intel_display.c. Rename the functions accordingly. The minor
downside is having to expose __intel_display_resume().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/5e98e2fc5f0c09490e02d22250c8201342852288.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |   1 +
 drivers/gpu/drm/i915/display/intel_display.c  | 123 +---------------
 drivers/gpu/drm/i915/display/intel_display.h  |   8 +-
 .../drm/i915/display/intel_display_reset.c    | 135 ++++++++++++++++++
 .../drm/i915/display/intel_display_reset.h    |  14 ++
 drivers/gpu/drm/i915/gt/intel_reset.c         |   6 +-
 6 files changed, 160 insertions(+), 127 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_reset.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_reset.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 627ec074df09b..154986359169e 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -241,6 +241,7 @@ i915-y += \
 	display/intel_display_power.o \
 	display/intel_display_power_map.o \
 	display/intel_display_power_well.o \
+	display/intel_display_reset.o \
 	display/intel_display_rps.o \
 	display/intel_dmc.o \
 	display/intel_dpio_phy.o \
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 8abce9a9a9000..47ec1bf46f719 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -693,7 +693,7 @@ intel_plane_fence_y_offset(const struct intel_plane_state *plane_state)
 	return y;
 }
 
-static int
+int
 __intel_display_resume(struct drm_i915_private *i915,
 		       struct drm_atomic_state *state,
 		       struct drm_modeset_acquire_ctx *ctx)
@@ -733,127 +733,6 @@ __intel_display_resume(struct drm_i915_private *i915,
 	return ret;
 }
 
-static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv)
-{
-	return (INTEL_INFO(dev_priv)->gpu_reset_clobbers_display &&
-		intel_has_gpu_reset(to_gt(dev_priv)));
-}
-
-void intel_display_prepare_reset(struct drm_i915_private *dev_priv)
-{
-	struct drm_modeset_acquire_ctx *ctx = &dev_priv->display.restore.reset_ctx;
-	struct drm_atomic_state *state;
-	int ret;
-
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	/* reset doesn't touch the display */
-	if (!dev_priv->params.force_reset_modeset_test &&
-	    !gpu_reset_clobbers_display(dev_priv))
-		return;
-
-	/* We have a modeset vs reset deadlock, defensively unbreak it. */
-	set_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags);
-	smp_mb__after_atomic();
-	wake_up_bit(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET);
-
-	if (atomic_read(&dev_priv->gpu_error.pending_fb_pin)) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "Modeset potentially stuck, unbreaking through wedging\n");
-		intel_gt_set_wedged(to_gt(dev_priv));
-	}
-
-	/*
-	 * Need mode_config.mutex so that we don't
-	 * trample ongoing ->detect() and whatnot.
-	 */
-	mutex_lock(&dev_priv->drm.mode_config.mutex);
-	drm_modeset_acquire_init(ctx, 0);
-	while (1) {
-		ret = drm_modeset_lock_all_ctx(&dev_priv->drm, ctx);
-		if (ret != -EDEADLK)
-			break;
-
-		drm_modeset_backoff(ctx);
-	}
-	/*
-	 * Disabling the crtcs gracefully seems nicer. Also the
-	 * g33 docs say we should at least disable all the planes.
-	 */
-	state = drm_atomic_helper_duplicate_state(&dev_priv->drm, ctx);
-	if (IS_ERR(state)) {
-		ret = PTR_ERR(state);
-		drm_err(&dev_priv->drm, "Duplicating state failed with %i\n",
-			ret);
-		return;
-	}
-
-	ret = drm_atomic_helper_disable_all(&dev_priv->drm, ctx);
-	if (ret) {
-		drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n",
-			ret);
-		drm_atomic_state_put(state);
-		return;
-	}
-
-	dev_priv->display.restore.modeset_state = state;
-	state->acquire_ctx = ctx;
-}
-
-void intel_display_finish_reset(struct drm_i915_private *i915)
-{
-	struct drm_modeset_acquire_ctx *ctx = &i915->display.restore.reset_ctx;
-	struct drm_atomic_state *state;
-	int ret;
-
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	/* reset doesn't touch the display */
-	if (!test_bit(I915_RESET_MODESET, &to_gt(i915)->reset.flags))
-		return;
-
-	state = fetch_and_zero(&i915->display.restore.modeset_state);
-	if (!state)
-		goto unlock;
-
-	/* reset doesn't touch the display */
-	if (!gpu_reset_clobbers_display(i915)) {
-		/* for testing only restore the display */
-		ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
-		if (ret) {
-			drm_WARN_ON(&i915->drm, ret == -EDEADLK);
-			drm_err(&i915->drm,
-				"Restoring old state failed with %i\n", ret);
-		}
-	} else {
-		/*
-		 * The display has been reset as well,
-		 * so need a full re-initialization.
-		 */
-		intel_pps_unlock_regs_wa(i915);
-		intel_display_driver_init_hw(i915);
-		intel_clock_gating_init(i915);
-		intel_hpd_init(i915);
-
-		ret = __intel_display_resume(i915, state, ctx);
-		if (ret)
-			drm_err(&i915->drm,
-				"Restoring old state failed with %i\n", ret);
-
-		intel_hpd_poll_disable(i915);
-	}
-
-	drm_atomic_state_put(state);
-unlock:
-	drm_modeset_drop_locks(ctx);
-	drm_modeset_acquire_fini(ctx);
-	mutex_unlock(&i915->drm.mode_config.mutex);
-
-	clear_bit_unlock(I915_RESET_MODESET, &to_gt(i915)->reset.flags);
-}
-
 static void icl_set_pipe_chicken(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 01c27a4757a57..346f6342e8edb 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -468,8 +468,6 @@ intel_framebuffer_create(struct drm_i915_gem_object *obj,
 
 bool intel_fuzzy_clock_check(int clock1, int clock2);
 
-void intel_display_prepare_reset(struct drm_i915_private *dev_priv);
-void intel_display_finish_reset(struct drm_i915_private *dev_priv);
 void intel_zero_m_n(struct intel_link_m_n *m_n);
 void intel_set_m_n(struct drm_i915_private *i915,
 		   const struct intel_link_m_n *m_n,
@@ -543,6 +541,12 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
 
 void intel_hpd_poll_fini(struct drm_i915_private *i915);
 
+/* interface for intel_display_reset.c */
+int
+__intel_display_resume(struct drm_i915_private *i915,
+		       struct drm_atomic_state *state,
+		       struct drm_modeset_acquire_ctx *ctx);
+
 /* modesetting asserts */
 void assert_transcoder(struct drm_i915_private *dev_priv,
 		       enum transcoder cpu_transcoder, bool state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.c b/drivers/gpu/drm/i915/display/intel_display_reset.c
new file mode 100644
index 0000000000000..166aa0cab1fc6
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_reset.c
@@ -0,0 +1,135 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <drm/drm_atomic_helper.h>
+
+#include "i915_drv.h"
+#include "intel_clock_gating.h"
+#include "intel_display_driver.h"
+#include "intel_display_reset.h"
+#include "intel_display_types.h"
+#include "intel_hotplug.h"
+#include "intel_pps.h"
+
+static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv)
+{
+	return (INTEL_INFO(dev_priv)->gpu_reset_clobbers_display &&
+		intel_has_gpu_reset(to_gt(dev_priv)));
+}
+
+void intel_display_reset_prepare(struct drm_i915_private *dev_priv)
+{
+	struct drm_modeset_acquire_ctx *ctx = &dev_priv->display.restore.reset_ctx;
+	struct drm_atomic_state *state;
+	int ret;
+
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	/* reset doesn't touch the display */
+	if (!dev_priv->params.force_reset_modeset_test &&
+	    !gpu_reset_clobbers_display(dev_priv))
+		return;
+
+	/* We have a modeset vs reset deadlock, defensively unbreak it. */
+	set_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags);
+	smp_mb__after_atomic();
+	wake_up_bit(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET);
+
+	if (atomic_read(&dev_priv->gpu_error.pending_fb_pin)) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "Modeset potentially stuck, unbreaking through wedging\n");
+		intel_gt_set_wedged(to_gt(dev_priv));
+	}
+
+	/*
+	 * Need mode_config.mutex so that we don't
+	 * trample ongoing ->detect() and whatnot.
+	 */
+	mutex_lock(&dev_priv->drm.mode_config.mutex);
+	drm_modeset_acquire_init(ctx, 0);
+	while (1) {
+		ret = drm_modeset_lock_all_ctx(&dev_priv->drm, ctx);
+		if (ret != -EDEADLK)
+			break;
+
+		drm_modeset_backoff(ctx);
+	}
+	/*
+	 * Disabling the crtcs gracefully seems nicer. Also the
+	 * g33 docs say we should at least disable all the planes.
+	 */
+	state = drm_atomic_helper_duplicate_state(&dev_priv->drm, ctx);
+	if (IS_ERR(state)) {
+		ret = PTR_ERR(state);
+		drm_err(&dev_priv->drm, "Duplicating state failed with %i\n",
+			ret);
+		return;
+	}
+
+	ret = drm_atomic_helper_disable_all(&dev_priv->drm, ctx);
+	if (ret) {
+		drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n",
+			ret);
+		drm_atomic_state_put(state);
+		return;
+	}
+
+	dev_priv->display.restore.modeset_state = state;
+	state->acquire_ctx = ctx;
+}
+
+void intel_display_reset_finish(struct drm_i915_private *i915)
+{
+	struct drm_modeset_acquire_ctx *ctx = &i915->display.restore.reset_ctx;
+	struct drm_atomic_state *state;
+	int ret;
+
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	/* reset doesn't touch the display */
+	if (!test_bit(I915_RESET_MODESET, &to_gt(i915)->reset.flags))
+		return;
+
+	state = fetch_and_zero(&i915->display.restore.modeset_state);
+	if (!state)
+		goto unlock;
+
+	/* reset doesn't touch the display */
+	if (!gpu_reset_clobbers_display(i915)) {
+		/* for testing only restore the display */
+		ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
+		if (ret) {
+			drm_WARN_ON(&i915->drm, ret == -EDEADLK);
+			drm_err(&i915->drm,
+				"Restoring old state failed with %i\n", ret);
+		}
+	} else {
+		/*
+		 * The display has been reset as well,
+		 * so need a full re-initialization.
+		 */
+		intel_pps_unlock_regs_wa(i915);
+		intel_display_driver_init_hw(i915);
+		intel_clock_gating_init(i915);
+		intel_hpd_init(i915);
+
+		ret = __intel_display_resume(i915, state, ctx);
+		if (ret)
+			drm_err(&i915->drm,
+				"Restoring old state failed with %i\n", ret);
+
+		intel_hpd_poll_disable(i915);
+	}
+
+	drm_atomic_state_put(state);
+unlock:
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+	mutex_unlock(&i915->drm.mode_config.mutex);
+
+	clear_bit_unlock(I915_RESET_MODESET, &to_gt(i915)->reset.flags);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.h b/drivers/gpu/drm/i915/display/intel_display_reset.h
new file mode 100644
index 0000000000000..f06d0d35b86b1
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_reset.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_RESET_H__
+#define __INTEL_RESET_H__
+
+struct drm_i915_private;
+
+void intel_display_reset_prepare(struct drm_i915_private *i915);
+void intel_display_reset_finish(struct drm_i915_private *i915);
+
+#endif /* __INTEL_RESET_H__ */
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 0bb9094fdacd5..c129138f666f0 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -7,7 +7,7 @@
 #include <linux/stop_machine.h>
 #include <linux/string_helpers.h>
 
-#include "display/intel_display.h"
+#include "display/intel_display_reset.h"
 #include "display/intel_overlay.h"
 
 #include "gem/i915_gem_context.h"
@@ -1291,11 +1291,11 @@ static void intel_gt_reset_global(struct intel_gt *gt,
 
 	/* Use a watchdog to ensure that our reset completes */
 	intel_wedge_on_timeout(&w, gt, 60 * HZ) {
-		intel_display_prepare_reset(gt->i915);
+		intel_display_reset_prepare(gt->i915);
 
 		intel_gt_reset(gt, engine_mask, reason);
 
-		intel_display_finish_reset(gt->i915);
+		intel_display_reset_finish(gt->i915);
 	}
 
 	if (!test_bit(I915_WEDGED, &gt->reset.flags))
-- 
GitLab


From 3183b9ebad7d3f1d711f152222577116ca08b299 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:42:00 +0300
Subject: [PATCH 0069/1791] drm/i915/display: move display suspend/resume to
 intel_display_driver.[ch]

High level display functionality only called from driver top level code.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/11fdd82437370d8f341cd546d546de5e934c000f.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  | 102 ------------------
 drivers/gpu/drm/i915/display/intel_display.h  |   8 --
 .../drm/i915/display/intel_display_driver.c   | 101 +++++++++++++++++
 .../drm/i915/display/intel_display_driver.h   |  10 ++
 4 files changed, 111 insertions(+), 110 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 47ec1bf46f719..9c9b0ef94598b 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -693,46 +693,6 @@ intel_plane_fence_y_offset(const struct intel_plane_state *plane_state)
 	return y;
 }
 
-int
-__intel_display_resume(struct drm_i915_private *i915,
-		       struct drm_atomic_state *state,
-		       struct drm_modeset_acquire_ctx *ctx)
-{
-	struct drm_crtc_state *crtc_state;
-	struct drm_crtc *crtc;
-	int ret, i;
-
-	intel_modeset_setup_hw_state(i915, ctx);
-	intel_vga_redisable(i915);
-
-	if (!state)
-		return 0;
-
-	/*
-	 * We've duplicated the state, pointers to the old state are invalid.
-	 *
-	 * Don't attempt to use the old state until we commit the duplicated state.
-	 */
-	for_each_new_crtc_in_state(state, crtc, crtc_state, i) {
-		/*
-		 * Force recalculation even if we restore
-		 * current state. With fast modeset this may not result
-		 * in a modeset when the state is compatible.
-		 */
-		crtc_state->mode_changed = true;
-	}
-
-	/* ignore any reset values/BIOS leftovers in the WM registers */
-	if (!HAS_GMCH(i915))
-		to_intel_atomic_state(state)->skip_intermediate_wm = true;
-
-	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
-
-	drm_WARN_ON(&i915->drm, ret == -EDEADLK);
-
-	return ret;
-}
-
 static void icl_set_pipe_chicken(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
@@ -2104,30 +2064,6 @@ static void i9xx_crtc_disable(struct intel_atomic_state *state,
 		i830_enable_pipe(dev_priv, pipe);
 }
 
-
-/*
- * turn all crtc's off, but do not adjust state
- * This has to be paired with a call to intel_modeset_setup_hw_state.
- */
-int intel_display_suspend(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_atomic_state *state;
-	int ret;
-
-	if (!HAS_DISPLAY(dev_priv))
-		return 0;
-
-	state = drm_atomic_helper_suspend(dev);
-	ret = PTR_ERR_OR_ZERO(state);
-	if (ret)
-		drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n",
-			ret);
-	else
-		dev_priv->display.restore.modeset_state = state;
-	return ret;
-}
-
 void intel_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
@@ -8286,44 +8222,6 @@ void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe)
 	intel_de_posting_read(dev_priv, DPLL(pipe));
 }
 
-void intel_display_resume(struct drm_device *dev)
-{
-	struct drm_i915_private *i915 = to_i915(dev);
-	struct drm_atomic_state *state = i915->display.restore.modeset_state;
-	struct drm_modeset_acquire_ctx ctx;
-	int ret;
-
-	if (!HAS_DISPLAY(i915))
-		return;
-
-	i915->display.restore.modeset_state = NULL;
-	if (state)
-		state->acquire_ctx = &ctx;
-
-	drm_modeset_acquire_init(&ctx, 0);
-
-	while (1) {
-		ret = drm_modeset_lock_all_ctx(dev, &ctx);
-		if (ret != -EDEADLK)
-			break;
-
-		drm_modeset_backoff(&ctx);
-	}
-
-	if (!ret)
-		ret = __intel_display_resume(i915, state, &ctx);
-
-	skl_watermark_ipc_update(i915);
-	drm_modeset_drop_locks(&ctx);
-	drm_modeset_acquire_fini(&ctx);
-
-	if (ret)
-		drm_err(&i915->drm,
-			"Restoring old state failed with %i\n", ret);
-	if (state)
-		drm_atomic_state_put(state);
-}
-
 void intel_hpd_poll_fini(struct drm_i915_private *i915)
 {
 	struct intel_connector *connector;
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 346f6342e8edb..fefad8cfc8a28 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -438,7 +438,6 @@ void intel_add_fb_offsets(int *x, int *y,
 unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info);
 unsigned int intel_remapped_info_size(const struct intel_remapped_info *rem_info);
 bool intel_has_pending_fb_unpin(struct drm_i915_private *dev_priv);
-int intel_display_suspend(struct drm_device *dev);
 void intel_encoder_destroy(struct drm_encoder *encoder);
 struct drm_display_mode *
 intel_encoder_current_mode(struct intel_encoder *encoder);
@@ -520,7 +519,6 @@ void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state);
 void intel_update_watermarks(struct drm_i915_private *i915);
 
 /* modesetting */
-void intel_display_resume(struct drm_device *dev);
 int intel_modeset_all_pipes(struct intel_atomic_state *state,
 			    const char *reason);
 void intel_modeset_get_crtc_power_domains(struct intel_crtc_state *crtc_state,
@@ -541,12 +539,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state,
 
 void intel_hpd_poll_fini(struct drm_i915_private *i915);
 
-/* interface for intel_display_reset.c */
-int
-__intel_display_resume(struct drm_i915_private *i915,
-		       struct drm_atomic_state *state,
-		       struct drm_modeset_acquire_ctx *ctx);
-
 /* modesetting asserts */
 void assert_transcoder(struct drm_i915_private *dev_priv,
 		       enum transcoder cpu_transcoder, bool state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 45e4251003852..eb03b0a87d5a5 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -466,3 +466,104 @@ void intel_display_driver_unregister(struct drm_i915_private *i915)
 	acpi_video_unregister();
 	intel_opregion_unregister(i915);
 }
+
+/*
+ * turn all crtc's off, but do not adjust state
+ * This has to be paired with a call to intel_modeset_setup_hw_state.
+ */
+int intel_display_suspend(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_atomic_state *state;
+	int ret;
+
+	if (!HAS_DISPLAY(dev_priv))
+		return 0;
+
+	state = drm_atomic_helper_suspend(dev);
+	ret = PTR_ERR_OR_ZERO(state);
+	if (ret)
+		drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n",
+			ret);
+	else
+		dev_priv->display.restore.modeset_state = state;
+	return ret;
+}
+
+int
+__intel_display_resume(struct drm_i915_private *i915,
+		       struct drm_atomic_state *state,
+		       struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_crtc_state *crtc_state;
+	struct drm_crtc *crtc;
+	int ret, i;
+
+	intel_modeset_setup_hw_state(i915, ctx);
+	intel_vga_redisable(i915);
+
+	if (!state)
+		return 0;
+
+	/*
+	 * We've duplicated the state, pointers to the old state are invalid.
+	 *
+	 * Don't attempt to use the old state until we commit the duplicated state.
+	 */
+	for_each_new_crtc_in_state(state, crtc, crtc_state, i) {
+		/*
+		 * Force recalculation even if we restore
+		 * current state. With fast modeset this may not result
+		 * in a modeset when the state is compatible.
+		 */
+		crtc_state->mode_changed = true;
+	}
+
+	/* ignore any reset values/BIOS leftovers in the WM registers */
+	if (!HAS_GMCH(i915))
+		to_intel_atomic_state(state)->skip_intermediate_wm = true;
+
+	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
+
+	drm_WARN_ON(&i915->drm, ret == -EDEADLK);
+
+	return ret;
+}
+
+void intel_display_resume(struct drm_device *dev)
+{
+	struct drm_i915_private *i915 = to_i915(dev);
+	struct drm_atomic_state *state = i915->display.restore.modeset_state;
+	struct drm_modeset_acquire_ctx ctx;
+	int ret;
+
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	i915->display.restore.modeset_state = NULL;
+	if (state)
+		state->acquire_ctx = &ctx;
+
+	drm_modeset_acquire_init(&ctx, 0);
+
+	while (1) {
+		ret = drm_modeset_lock_all_ctx(dev, &ctx);
+		if (ret != -EDEADLK)
+			break;
+
+		drm_modeset_backoff(&ctx);
+	}
+
+	if (!ret)
+		ret = __intel_display_resume(i915, state, &ctx);
+
+	skl_watermark_ipc_update(i915);
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	if (ret)
+		drm_err(&i915->drm,
+			"Restoring old state failed with %i\n", ret);
+	if (state)
+		drm_atomic_state_put(state);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index aab498617b90e..7b5ff4309dec7 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -8,7 +8,10 @@
 
 #include <linux/types.h>
 
+struct drm_atomic_state;
+struct drm_device;
 struct drm_i915_private;
+struct drm_modeset_acquire_ctx;
 struct pci_dev;
 
 bool intel_display_driver_probe_defer(struct pci_dev *pdev);
@@ -21,6 +24,13 @@ void intel_display_driver_remove(struct drm_i915_private *i915);
 void intel_display_driver_remove_noirq(struct drm_i915_private *i915);
 void intel_display_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
+int intel_display_suspend(struct drm_device *dev);
+void intel_display_resume(struct drm_device *dev);
+
+/* interface for intel_display_reset.c */
+int __intel_display_resume(struct drm_i915_private *i915,
+			   struct drm_atomic_state *state,
+			   struct drm_modeset_acquire_ctx *ctx);
 
 #endif /* __INTEL_DISPLAY_DRIVER_H__ */
 
-- 
GitLab


From cde4bd87863124fb6feee35b7f73552f5e75ea61 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:42:01 +0300
Subject: [PATCH 0070/1791] drm/i915/display: rename
 intel_display_driver_suspend/resume functions

Follow the usual naming conventions. Switch to i915 arguments and naming
while at it.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/cdd4228337678609967ed176dcfc9690de5d490b.1681465222.git.jani.nikula@intel.com
---
 .../drm/i915/display/intel_display_driver.c   | 24 +++++++++----------
 .../drm/i915/display/intel_display_driver.h   | 11 ++++-----
 .../drm/i915/display/intel_display_reset.c    |  2 +-
 drivers/gpu/drm/i915/i915_driver.c            |  6 ++---
 4 files changed, 20 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index eb03b0a87d5a5..d20a279fdf514 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -471,29 +471,28 @@ void intel_display_driver_unregister(struct drm_i915_private *i915)
  * turn all crtc's off, but do not adjust state
  * This has to be paired with a call to intel_modeset_setup_hw_state.
  */
-int intel_display_suspend(struct drm_device *dev)
+int intel_display_driver_suspend(struct drm_i915_private *i915)
 {
-	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_atomic_state *state;
 	int ret;
 
-	if (!HAS_DISPLAY(dev_priv))
+	if (!HAS_DISPLAY(i915))
 		return 0;
 
-	state = drm_atomic_helper_suspend(dev);
+	state = drm_atomic_helper_suspend(&i915->drm);
 	ret = PTR_ERR_OR_ZERO(state);
 	if (ret)
-		drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n",
+		drm_err(&i915->drm, "Suspending crtc's failed with %i\n",
 			ret);
 	else
-		dev_priv->display.restore.modeset_state = state;
+		i915->display.restore.modeset_state = state;
 	return ret;
 }
 
 int
-__intel_display_resume(struct drm_i915_private *i915,
-		       struct drm_atomic_state *state,
-		       struct drm_modeset_acquire_ctx *ctx)
+__intel_display_driver_resume(struct drm_i915_private *i915,
+			      struct drm_atomic_state *state,
+			      struct drm_modeset_acquire_ctx *ctx)
 {
 	struct drm_crtc_state *crtc_state;
 	struct drm_crtc *crtc;
@@ -530,9 +529,8 @@ __intel_display_resume(struct drm_i915_private *i915,
 	return ret;
 }
 
-void intel_display_resume(struct drm_device *dev)
+void intel_display_driver_resume(struct drm_i915_private *i915)
 {
-	struct drm_i915_private *i915 = to_i915(dev);
 	struct drm_atomic_state *state = i915->display.restore.modeset_state;
 	struct drm_modeset_acquire_ctx ctx;
 	int ret;
@@ -547,7 +545,7 @@ void intel_display_resume(struct drm_device *dev)
 	drm_modeset_acquire_init(&ctx, 0);
 
 	while (1) {
-		ret = drm_modeset_lock_all_ctx(dev, &ctx);
+		ret = drm_modeset_lock_all_ctx(&i915->drm, &ctx);
 		if (ret != -EDEADLK)
 			break;
 
@@ -555,7 +553,7 @@ void intel_display_resume(struct drm_device *dev)
 	}
 
 	if (!ret)
-		ret = __intel_display_resume(i915, state, &ctx);
+		ret = __intel_display_driver_resume(i915, state, &ctx);
 
 	skl_watermark_ipc_update(i915);
 	drm_modeset_drop_locks(&ctx);
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 7b5ff4309dec7..84e7977f265ac 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -9,7 +9,6 @@
 #include <linux/types.h>
 
 struct drm_atomic_state;
-struct drm_device;
 struct drm_i915_private;
 struct drm_modeset_acquire_ctx;
 struct pci_dev;
@@ -24,13 +23,13 @@ void intel_display_driver_remove(struct drm_i915_private *i915);
 void intel_display_driver_remove_noirq(struct drm_i915_private *i915);
 void intel_display_driver_remove_nogem(struct drm_i915_private *i915);
 void intel_display_driver_unregister(struct drm_i915_private *i915);
-int intel_display_suspend(struct drm_device *dev);
-void intel_display_resume(struct drm_device *dev);
+int intel_display_driver_suspend(struct drm_i915_private *i915);
+void intel_display_driver_resume(struct drm_i915_private *i915);
 
 /* interface for intel_display_reset.c */
-int __intel_display_resume(struct drm_i915_private *i915,
-			   struct drm_atomic_state *state,
-			   struct drm_modeset_acquire_ctx *ctx);
+int __intel_display_driver_resume(struct drm_i915_private *i915,
+				  struct drm_atomic_state *state,
+				  struct drm_modeset_acquire_ctx *ctx);
 
 #endif /* __INTEL_DISPLAY_DRIVER_H__ */
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.c b/drivers/gpu/drm/i915/display/intel_display_reset.c
index 166aa0cab1fc6..17178d5d7788a 100644
--- a/drivers/gpu/drm/i915/display/intel_display_reset.c
+++ b/drivers/gpu/drm/i915/display/intel_display_reset.c
@@ -117,7 +117,7 @@ void intel_display_reset_finish(struct drm_i915_private *i915)
 		intel_clock_gating_init(i915);
 		intel_hpd_init(i915);
 
-		ret = __intel_display_resume(i915, state, ctx);
+		ret = __intel_display_driver_resume(i915, state, ctx);
 		if (ret)
 			drm_err(&i915->drm,
 				"Restoring old state failed with %i\n", ret);
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index d38ce3885c4d1..d36e4b019cce0 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -1043,7 +1043,7 @@ static int i915_drm_prepare(struct drm_device *dev)
 	intel_pxp_suspend_prepare(i915->pxp);
 
 	/*
-	 * NB intel_display_suspend() may issue new requests after we've
+	 * NB intel_display_driver_suspend() may issue new requests after we've
 	 * ostensibly marked the GPU as ready-to-sleep here. We need to
 	 * split out that work and pull it forward so that after point,
 	 * the GPU is not woken again.
@@ -1067,7 +1067,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 
 	pci_save_state(pdev);
 
-	intel_display_suspend(dev);
+	intel_display_driver_suspend(dev_priv);
 
 	intel_dp_mst_suspend(dev_priv);
 
@@ -1241,7 +1241,7 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	/* MST sideband requires HPD interrupts enabled */
 	intel_dp_mst_resume(dev_priv);
-	intel_display_resume(dev);
+	intel_display_driver_resume(dev_priv);
 
 	intel_hpd_poll_disable(dev_priv);
 	if (HAS_DISPLAY(dev_priv))
-- 
GitLab


From 62bb6b4920ce9d9a7cc365c4e1cc13134cd1cc24 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Apr 2023 12:42:02 +0300
Subject: [PATCH 0071/1791] drm/i915/display: add
 intel_display_driver_early_probe()

Add intel_display_driver_early_probe() as the early probe call to
replace intel_init_display_hooks(). The latter will be "demoted" to
setting up hooks in intel_display.c only.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/0b02311e5343527b501b44671d2188f2a1b30a7d.1681465222.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c       | 11 -----------
 .../gpu/drm/i915/display/intel_display_driver.c    | 14 ++++++++++++++
 .../gpu/drm/i915/display/intel_display_driver.h    |  1 +
 drivers/gpu/drm/i915/i915_driver.c                 |  2 +-
 4 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 9c9b0ef94598b..1c264c17b6e4b 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -8019,15 +8019,6 @@ static const struct intel_display_funcs i9xx_display_funcs = {
  */
 void intel_init_display_hooks(struct drm_i915_private *dev_priv)
 {
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	intel_color_init_hooks(dev_priv);
-	intel_init_cdclk_hooks(dev_priv);
-	intel_audio_hooks_init(dev_priv);
-
-	intel_dpll_init_clock_hook(dev_priv);
-
 	if (DISPLAY_VER(dev_priv) >= 9) {
 		dev_priv->display.funcs.display = &skl_display_funcs;
 	} else if (HAS_DDI(dev_priv)) {
@@ -8040,8 +8031,6 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv)
 	} else {
 		dev_priv->display.funcs.display = &i9xx_display_funcs;
 	}
-
-	intel_fdi_init_hook(dev_priv);
 }
 
 int intel_initial_commit(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index d20a279fdf514..b3dbfe2a892e6 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -32,6 +32,7 @@
 #include "intel_display_types.h"
 #include "intel_dmc.h"
 #include "intel_dp.h"
+#include "intel_dpll.h"
 #include "intel_dpll_mgr.h"
 #include "intel_fb.h"
 #include "intel_fbc.h"
@@ -169,6 +170,19 @@ static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv)
 	}
 }
 
+void intel_display_driver_early_probe(struct drm_i915_private *i915)
+{
+	if (!HAS_DISPLAY(i915))
+		return;
+
+	intel_color_init_hooks(i915);
+	intel_init_cdclk_hooks(i915);
+	intel_audio_hooks_init(i915);
+	intel_dpll_init_clock_hook(i915);
+	intel_init_display_hooks(i915);
+	intel_fdi_init_hook(i915);
+}
+
 /* part #1: call before irq install */
 int intel_display_driver_probe_noirq(struct drm_i915_private *i915)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h
index 84e7977f265ac..c276a58ee3293 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.h
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.h
@@ -15,6 +15,7 @@ struct pci_dev;
 
 bool intel_display_driver_probe_defer(struct pci_dev *pdev);
 void intel_display_driver_init_hw(struct drm_i915_private *i915);
+void intel_display_driver_early_probe(struct drm_i915_private *i915);
 int intel_display_driver_probe_noirq(struct drm_i915_private *i915);
 int intel_display_driver_probe_nogem(struct drm_i915_private *i915);
 int intel_display_driver_probe(struct drm_i915_private *i915);
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index d36e4b019cce0..6ad4190ed46cf 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -252,7 +252,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 	intel_detect_pch(dev_priv);
 
 	intel_irq_init(dev_priv);
-	intel_init_display_hooks(dev_priv);
+	intel_display_driver_early_probe(dev_priv);
 	intel_clock_gating_hooks_init(dev_priv);
 
 	intel_detect_preproduction_hw(dev_priv);
-- 
GitLab


From b4b26ab24f8f9ad3aa9fbdfd211e51b051ddb91e Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Mon, 17 Apr 2023 16:39:03 +0200
Subject: [PATCH 0072/1791] dt-bindings: display: panel: nt36523: Add Lenovo
 J606F panel

Some Lenovo J606F tablets come with a 2K (2000x1200) 60Hz 11" 5:3
video mode display. Document it and allow rotation while at it (Lenovo
mounted it upside down!).

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412-topic-lenovopanel-v3-1-bcf9ba4de46f@linaro.org
---
 .../bindings/display/panel/novatek,nt36523.yaml  | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/panel/novatek,nt36523.yaml b/Documentation/devicetree/bindings/display/panel/novatek,nt36523.yaml
index 0039561ef04c8..5f7e4c4860944 100644
--- a/Documentation/devicetree/bindings/display/panel/novatek,nt36523.yaml
+++ b/Documentation/devicetree/bindings/display/panel/novatek,nt36523.yaml
@@ -19,11 +19,16 @@ allOf:
 
 properties:
   compatible:
-    items:
-      - enum:
-          - xiaomi,elish-boe-nt36523
-          - xiaomi,elish-csot-nt36523
-      - const: novatek,nt36523
+    oneOf:
+      - items:
+          - enum:
+              - xiaomi,elish-boe-nt36523
+              - xiaomi,elish-csot-nt36523
+          - const: novatek,nt36523
+      - items:
+          - enum:
+              - lenovo,j606f-boe-nt36523w
+          - const: novatek,nt36523w
 
   reset-gpios:
     maxItems: 1
@@ -34,6 +39,7 @@ properties:
 
   reg: true
   ports: true
+  rotation: true
   backlight: true
 
 required:
-- 
GitLab


From aecb583cb997935fb4f4a667a8013469528a8d53 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Mon, 17 Apr 2023 16:39:04 +0200
Subject: [PATCH 0073/1791] drm/panel: nt36523: Add DCS backlight support

This chip supports controlling the backlight via DCS commands, on at
least some panels. Add support for doing so.

Note this may only concern the NT36523*W* variant. Nobody knows, really,
there's no docs.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Jianhua Lu <lujianhua000@gmail.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412-topic-lenovopanel-v3-2-bcf9ba4de46f@linaro.org
---
 drivers/gpu/drm/panel/panel-novatek-nt36523.c | 68 ++++++++++++++++++-
 1 file changed, 65 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-novatek-nt36523.c b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
index d30dbbfb67b17..b0466abae8125 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt36523.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
@@ -5,6 +5,7 @@
  * Copyright (c) 2022, 2023 Jianhua Lu <lujianhua000@gmail.com>
  */
 
+#include <linux/backlight.h>
 #include <linux/delay.h>
 #include <linux/gpio/consumer.h>
 #include <linux/module.h>
@@ -53,6 +54,7 @@ struct panel_desc {
 	int (*init_sequence)(struct panel_info *pinfo);
 
 	bool is_dual_dsi;
+	bool has_dcs_backlight;
 };
 
 static inline struct panel_info *to_panel_info(struct drm_panel *panel)
@@ -679,6 +681,59 @@ static const struct drm_panel_funcs nt36523_panel_funcs = {
 	.get_modes = nt36523_get_modes,
 };
 
+static int nt36523_bl_update_status(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness = backlight_get_brightness(bl);
+	int ret;
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	ret = mipi_dsi_dcs_set_display_brightness_large(dsi, brightness);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	return 0;
+}
+
+static int nt36523_bl_get_brightness(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness;
+	int ret;
+
+	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
+
+	ret = mipi_dsi_dcs_get_display_brightness_large(dsi, &brightness);
+	if (ret < 0)
+		return ret;
+
+	dsi->mode_flags |= MIPI_DSI_MODE_LPM;
+
+	return brightness;
+}
+
+static const struct backlight_ops nt36523_bl_ops = {
+	.update_status = nt36523_bl_update_status,
+	.get_brightness = nt36523_bl_get_brightness,
+};
+
+static struct backlight_device *nt36523_create_backlight(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	const struct backlight_properties props = {
+		.type = BACKLIGHT_RAW,
+		.brightness = 512,
+		.max_brightness = 4095,
+		.scale = BACKLIGHT_SCALE_NON_LINEAR,
+	};
+
+	return devm_backlight_device_register(dev, dev_name(dev), dev, dsi,
+					      &nt36523_bl_ops, &props);
+}
+
 static int nt36523_probe(struct mipi_dsi_device *dsi)
 {
 	struct device *dev = &dsi->dev;
@@ -730,9 +785,16 @@ static int nt36523_probe(struct mipi_dsi_device *dsi)
 	mipi_dsi_set_drvdata(dsi, pinfo);
 	drm_panel_init(&pinfo->panel, dev, &nt36523_panel_funcs, DRM_MODE_CONNECTOR_DSI);
 
-	ret = drm_panel_of_backlight(&pinfo->panel);
-	if (ret)
-		return dev_err_probe(dev, ret, "failed to get backlight\n");
+	if (pinfo->desc->has_dcs_backlight) {
+		pinfo->panel.backlight = nt36523_create_backlight(dsi);
+		if (IS_ERR(pinfo->panel.backlight))
+			return dev_err_probe(dev, PTR_ERR(pinfo->panel.backlight),
+					     "Failed to create backlight\n");
+	} else {
+		ret = drm_panel_of_backlight(&pinfo->panel);
+		if (ret)
+			return dev_err_probe(dev, ret, "Failed to get backlight\n");
+	}
 
 	drm_panel_add(&pinfo->panel);
 
-- 
GitLab


From 1eae88fa7b56b3b02d0e003a737fc31d71f3f486 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Mon, 17 Apr 2023 16:39:05 +0200
Subject: [PATCH 0074/1791] drm/panel: nt36523: Get orientation from OF

Some bright vendors mount their display panels upside down. Add the
required pieces to allow for accounting for that.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Jianhua Lu <lujianhua000@gmail.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412-topic-lenovopanel-v3-3-bcf9ba4de46f@linaro.org
---
 drivers/gpu/drm/panel/panel-novatek-nt36523.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-novatek-nt36523.c b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
index b0466abae8125..2c5712ac30377 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt36523.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
@@ -31,6 +31,7 @@ struct panel_info {
 	struct drm_panel panel;
 	struct mipi_dsi_device *dsi[2];
 	const struct panel_desc *desc;
+	enum drm_panel_orientation orientation;
 
 	struct gpio_desc *reset_gpio;
 	struct backlight_device *backlight;
@@ -674,11 +675,19 @@ static int nt36523_get_modes(struct drm_panel *panel,
 	return pinfo->desc->num_modes;
 }
 
+static enum drm_panel_orientation nt36523_get_orientation(struct drm_panel *panel)
+{
+	struct panel_info *pinfo = to_panel_info(panel);
+
+	return pinfo->orientation;
+}
+
 static const struct drm_panel_funcs nt36523_panel_funcs = {
 	.disable = nt36523_disable,
 	.prepare = nt36523_prepare,
 	.unprepare = nt36523_unprepare,
 	.get_modes = nt36523_get_modes,
+	.get_orientation = nt36523_get_orientation,
 };
 
 static int nt36523_bl_update_status(struct backlight_device *bl)
@@ -785,6 +794,12 @@ static int nt36523_probe(struct mipi_dsi_device *dsi)
 	mipi_dsi_set_drvdata(dsi, pinfo);
 	drm_panel_init(&pinfo->panel, dev, &nt36523_panel_funcs, DRM_MODE_CONNECTOR_DSI);
 
+	ret = of_drm_get_panel_orientation(dev->of_node, &pinfo->orientation);
+	if (ret < 0) {
+		dev_err(dev, "%pOF: failed to get orientation %d\n", dev->of_node, ret);
+		return ret;
+	}
+
 	if (pinfo->desc->has_dcs_backlight) {
 		pinfo->panel.backlight = nt36523_create_backlight(dsi);
 		if (IS_ERR(pinfo->panel.backlight))
-- 
GitLab


From 4f048de28b90abefff2044f450e882576eb746e9 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Mon, 17 Apr 2023 16:39:06 +0200
Subject: [PATCH 0075/1791] drm/panel: nt36523: Add Lenovo J606F panel

Some Lenovo J606F tablets come with a 2K (2000x1200) 60Hz 11" 5:3
video mode display. Add support for these panels.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Jianhua Lu <lujianhua000@gmail.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412-topic-lenovopanel-v3-4-bcf9ba4de46f@linaro.org
---
 drivers/gpu/drm/panel/panel-novatek-nt36523.c | 486 ++++++++++++++++++
 1 file changed, 486 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-novatek-nt36523.c b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
index 2c5712ac30377..c3befa7f253dc 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt36523.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
@@ -13,6 +13,8 @@
 #include <linux/of_graph.h>
 #include <linux/regulator/consumer.h>
 
+#include <video/mipi_display.h>
+
 #include <drm/drm_connector.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_mipi_dsi.h>
@@ -481,6 +483,456 @@ static int elish_csot_init_sequence(struct panel_info *pinfo)
 	return 0;
 }
 
+static int j606f_boe_init_sequence(struct panel_info *pinfo)
+{
+	struct mipi_dsi_device *dsi = pinfo->dsi[0];
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x20);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x05, 0xd9);
+	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x78);
+	mipi_dsi_dcs_write_seq(dsi, 0x08, 0x5a);
+	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x63);
+	mipi_dsi_dcs_write_seq(dsi, 0x0e, 0x91);
+	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x73);
+	mipi_dsi_dcs_write_seq(dsi, 0x95, 0xeb);
+	mipi_dsi_dcs_write_seq(dsi, 0x96, 0xeb);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x11);
+	mipi_dsi_dcs_write_seq(dsi, 0x6d, 0x66);
+	mipi_dsi_dcs_write_seq(dsi, 0x75, 0xa2);
+	mipi_dsi_dcs_write_seq(dsi, 0x77, 0xb3);
+	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
+			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq(dsi, 0xb1, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
+			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
+			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
+			       0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xb4, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
+			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
+			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
+			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq(dsi, 0xb7, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
+			       0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xb8, 0x00, 0x08, 0x00, 0x23, 0x00, 0x4d, 0x00, 0x6d, 0x00,
+			       0x89, 0x00, 0xa1, 0x00, 0xb6, 0x00, 0xc9);
+	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x00, 0xda, 0x01, 0x13, 0x01, 0x3c, 0x01, 0x7e, 0x01,
+			       0xab, 0x01, 0xf7, 0x02, 0x2f, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq(dsi, 0xba, 0x02, 0x67, 0x02, 0xa6, 0x02, 0xd1, 0x03, 0x08, 0x03,
+			       0x2e, 0x03, 0x5b, 0x03, 0x6b, 0x03, 0x7b);
+	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x03, 0x8e, 0x03, 0xa2, 0x03, 0xb7, 0x03, 0xe7, 0x03,
+			       0xfd, 0x03, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x21);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0xb0, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
+			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq(dsi, 0xb1, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
+			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
+			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
+			       0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq(dsi, 0xb4, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
+			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
+			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
+			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq(dsi, 0xb7, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
+			       0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq(dsi, 0xb8, 0x00, 0x00, 0x00, 0x1b, 0x00, 0x45, 0x00, 0x65, 0x00,
+			       0x81, 0x00, 0x99, 0x00, 0xae, 0x00, 0xc1);
+	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x00, 0xd2, 0x01, 0x0b, 0x01, 0x34, 0x01, 0x76, 0x01,
+			       0xa3, 0x01, 0xef, 0x02, 0x27, 0x02, 0x29);
+	mipi_dsi_dcs_write_seq(dsi, 0xba, 0x02, 0x5f, 0x02, 0x9e, 0x02, 0xc9, 0x03, 0x00, 0x03,
+			       0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73);
+	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x03, 0x86, 0x03, 0x9a, 0x03, 0xaf, 0x03, 0xdf, 0x03,
+			       0xf5, 0x03, 0xf7);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x23);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x80);
+	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x77);
+	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x07);
+	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x07);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x24);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x01, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x02, 0x1c);
+	mipi_dsi_dcs_write_seq(dsi, 0x03, 0x1c);
+	mipi_dsi_dcs_write_seq(dsi, 0x04, 0x1d);
+	mipi_dsi_dcs_write_seq(dsi, 0x05, 0x1d);
+	mipi_dsi_dcs_write_seq(dsi, 0x06, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x07, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x08, 0x0f);
+	mipi_dsi_dcs_write_seq(dsi, 0x09, 0x0f);
+	mipi_dsi_dcs_write_seq(dsi, 0x0a, 0x0e);
+	mipi_dsi_dcs_write_seq(dsi, 0x0b, 0x0e);
+	mipi_dsi_dcs_write_seq(dsi, 0x0c, 0x0d);
+	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x0d);
+	mipi_dsi_dcs_write_seq(dsi, 0x0e, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, 0x10, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x13, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x17, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x1c);
+	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x1c);
+	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x1d);
+	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x1d);
+	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x1d, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0x0f);
+	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x0f);
+	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x0e);
+	mipi_dsi_dcs_write_seq(dsi, 0x21, 0x0e);
+	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x0d);
+	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x0d);
+	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x28, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x29, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_LUT, 0x20);
+	mipi_dsi_dcs_write_seq(dsi, 0x2f, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x44);
+	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x32);
+	mipi_dsi_dcs_write_seq(dsi, 0x37, 0x44);
+	mipi_dsi_dcs_write_seq(dsi, 0x38, 0x40);
+	mipi_dsi_dcs_write_seq(dsi, 0x39, 0x00);
+
+	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x9a);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set pixel format: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_3D_CONTROL, 0x42);
+	mipi_dsi_dcs_write_seq(dsi, 0x3f, 0x06);
+	mipi_dsi_dcs_write_seq(dsi, 0x43, 0x06);
+	mipi_dsi_dcs_write_seq(dsi, 0x47, 0x66);
+	mipi_dsi_dcs_write_seq(dsi, 0x4a, 0x9a);
+	mipi_dsi_dcs_write_seq(dsi, 0x4b, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0x4c, 0x91);
+	mipi_dsi_dcs_write_seq(dsi, 0x4d, 0x21);
+	mipi_dsi_dcs_write_seq(dsi, 0x4e, 0x43);
+
+	ret = mipi_dsi_dcs_set_display_brightness(dsi, 18);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display brightness: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0x52, 0x34);
+	mipi_dsi_dcs_write_seq(dsi, 0x55, 0x82, 0x02);
+	mipi_dsi_dcs_write_seq(dsi, 0x56, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x58, 0x21);
+	mipi_dsi_dcs_write_seq(dsi, 0x59, 0x30);
+	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x00, 0x06);
+	mipi_dsi_dcs_write_seq(dsi, 0x5f, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x65, 0x82);
+	mipi_dsi_dcs_write_seq(dsi, 0x7e, 0x20);
+	mipi_dsi_dcs_write_seq(dsi, 0x7f, 0x3c);
+	mipi_dsi_dcs_write_seq(dsi, 0x82, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0x97, 0xc0);
+	mipi_dsi_dcs_write_seq(dsi, 0xb6,
+			       0x05, 0x00, 0x05, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05,
+			       0x05, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x92, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0x93, 0x1a);
+	mipi_dsi_dcs_write_seq(dsi, 0x94, 0x5f);
+	mipi_dsi_dcs_write_seq(dsi, 0xd7, 0x55);
+	mipi_dsi_dcs_write_seq(dsi, 0xda, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0xde, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0xdb, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xdc, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0xdd, 0x22);
+	mipi_dsi_dcs_write_seq(dsi, 0xdf, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe0, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0xe1, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe2, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0xe3, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe4, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0xe5, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe6, 0xc4);
+	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x88);
+	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x8d, 0x88);
+	mipi_dsi_dcs_write_seq(dsi, 0x8e, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0xb5, 0x90);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x25);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x05, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x07);
+	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x20, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x27, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0x33, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x34, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0x3f, 0xe0);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_VSYNC_TIMING, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x44, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_GET_SCANLINE, 0x40);
+	mipi_dsi_dcs_write_seq(dsi, 0x48, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x49, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0xd0);
+	mipi_dsi_dcs_write_seq(dsi, 0x61, 0xba);
+	mipi_dsi_dcs_write_seq(dsi, 0x62, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0xf1, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x64, 0x16);
+	mipi_dsi_dcs_write_seq(dsi, 0x67, 0x16);
+	mipi_dsi_dcs_write_seq(dsi, 0x6a, 0x16);
+	mipi_dsi_dcs_write_seq(dsi, 0x70, 0x30);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_READ_PPS_START, 0xf3);
+	mipi_dsi_dcs_write_seq(dsi, 0xa3, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xa4, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xa5, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, 0xd6, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x26);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x00, 0xa1);
+	mipi_dsi_dcs_write_seq(dsi, 0x0a, 0xf2);
+	mipi_dsi_dcs_write_seq(dsi, 0x04, 0x28);
+	mipi_dsi_dcs_write_seq(dsi, 0x06, 0x30);
+	mipi_dsi_dcs_write_seq(dsi, 0x0c, 0x13);
+	mipi_dsi_dcs_write_seq(dsi, 0x0d, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0x0f, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0x11, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x12, 0x50);
+	mipi_dsi_dcs_write_seq(dsi, 0x13, 0x51);
+	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x65);
+	mipi_dsi_dcs_write_seq(dsi, 0x15, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x16, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0x17, 0xa0);
+	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x86);
+	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x11);
+	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x7b);
+	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0xbb);
+	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x11);
+	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x7b);
+	mipi_dsi_dcs_write_seq(dsi, 0x1d, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x25, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, 0x2f, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x32, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, 0x39, 0x00);
+
+	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0xc3);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set pixel format: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x11);
+	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x78);
+	mipi_dsi_dcs_write_seq(dsi, 0x35, 0x16);
+	mipi_dsi_dcs_write_seq(dsi, 0xc8, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0xc9, 0x82);
+	mipi_dsi_dcs_write_seq(dsi, 0xca, 0x4e);
+	mipi_dsi_dcs_write_seq(dsi, 0xcb, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_READ_PPS_CONTINUE, 0x4c);
+	mipi_dsi_dcs_write_seq(dsi, 0xaa, 0x47);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x27);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x56, 0x06);
+	mipi_dsi_dcs_write_seq(dsi, 0x58, 0x80);
+	mipi_dsi_dcs_write_seq(dsi, 0x59, 0x53);
+	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x14);
+	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_CABC_MIN_BRIGHTNESS, 0x20);
+	mipi_dsi_dcs_write_seq(dsi, 0x5f, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0x60, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x61, 0x1d);
+	mipi_dsi_dcs_write_seq(dsi, 0x62, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x63, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x64, 0x24);
+	mipi_dsi_dcs_write_seq(dsi, 0x65, 0x1c);
+	mipi_dsi_dcs_write_seq(dsi, 0x66, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x67, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x68, 0x25);
+	mipi_dsi_dcs_write_seq(dsi, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x78, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0xc3, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0xd1, 0x24);
+	mipi_dsi_dcs_write_seq(dsi, 0xd2, 0x30);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x2f);
+	mipi_dsi_dcs_write_seq(dsi, 0x23, 0x08);
+	mipi_dsi_dcs_write_seq(dsi, 0x24, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x25, 0xc3);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0xf8);
+	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x28, 0x1a);
+	mipi_dsi_dcs_write_seq(dsi, 0x29, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x1a);
+	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_LUT, 0x1a);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0xe0);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x14, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x16, 0xc0);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0xf0);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+
+	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x08);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set pixel format: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x24);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+
+	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x5d);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set pixel format: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x4a, 0x5d);
+	mipi_dsi_dcs_write_seq(dsi, 0x4b, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x5a, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x91, 0x44);
+	mipi_dsi_dcs_write_seq(dsi, 0x92, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xdb, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xdc, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xdd, 0x22);
+	mipi_dsi_dcs_write_seq(dsi, 0xdf, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe0, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xe1, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe2, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xe3, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe4, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xe5, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0xe6, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0x5c, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x5d, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x8d, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x8e, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x25);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x20, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_GAMMA_CURVE, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x27, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x33, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x34, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x48, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x49, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0x5b, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x61, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, 0x62, 0x60);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x26);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x02, 0x31);
+	mipi_dsi_dcs_write_seq(dsi, 0x19, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0x1a, 0x7f);
+	mipi_dsi_dcs_write_seq(dsi, 0x1b, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0x1c, 0x0c);
+	mipi_dsi_dcs_write_seq(dsi, 0x2a, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, 0x2b, 0x7f);
+	mipi_dsi_dcs_write_seq(dsi, 0x1e, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0x1f, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_ROWS, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_PARTIAL_COLUMNS, 0x05);
+	mipi_dsi_dcs_write_seq(dsi, 0x32, 0x8d);
+
+	ret = mipi_dsi_dcs_set_pixel_format(dsi, 0x75);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set pixel format: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x2a);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x25, 0x75);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x20);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0x18, 0x40);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+	mipi_dsi_dcs_write_seq(dsi, 0xb9, 0x02);
+
+	ret = mipi_dsi_dcs_set_tear_on(dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set tear on: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x13);
+	mipi_dsi_dcs_write_seq(dsi, 0x3b, 0x03, 0x5f, 0x1a, 0x04, 0x04);
+	mipi_dsi_dcs_write_seq(dsi, 0xff, 0x10);
+	usleep_range(10000, 11000);
+	mipi_dsi_dcs_write_seq(dsi, 0xfb, 0x01);
+
+	ret = mipi_dsi_dcs_set_display_brightness(dsi, 0);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display brightness: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x2c);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0x68, 0x05, 0x01);
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
+		return ret;
+	}
+	msleep(100);
+
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display on: %d\n", ret);
+		return ret;
+	}
+	msleep(30);
+
+	return 0;
+}
+
 static const struct drm_display_mode elish_boe_modes[] = {
 	{
 		/* There is only one 120 Hz timing, but it doesn't work perfectly, 104 Hz preferred */
@@ -511,6 +963,22 @@ static const struct drm_display_mode elish_csot_modes[] = {
 	},
 };
 
+static const struct drm_display_mode j606f_boe_modes[] = {
+	{
+		.clock = (1200 + 58 + 2 + 60) * (2000 + 26 + 2 + 93) * 60 / 1000,
+		.hdisplay = 1200,
+		.hsync_start = 1200 + 58,
+		.hsync_end = 1200 + 58 + 2,
+		.htotal = 1200 + 58 + 2 + 60,
+		.vdisplay = 2000,
+		.vsync_start = 2000 + 26,
+		.vsync_end = 2000 + 26 + 2,
+		.vtotal = 2000 + 26 + 2 + 93,
+		.width_mm = 143,
+		.height_mm = 235,
+	},
+};
+
 static const struct panel_desc elish_boe_desc = {
 	.modes = elish_boe_modes,
 	.num_modes = ARRAY_SIZE(elish_boe_modes),
@@ -547,6 +1015,20 @@ static const struct panel_desc elish_csot_desc = {
 	.is_dual_dsi = true,
 };
 
+static const struct panel_desc j606f_boe_desc = {
+	.modes = j606f_boe_modes,
+	.num_modes = ARRAY_SIZE(j606f_boe_modes),
+	.width_mm = 143,
+	.height_mm = 235,
+	.bpc = 8,
+	.lanes = 4,
+	.format = MIPI_DSI_FMT_RGB888,
+	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+		      MIPI_DSI_CLOCK_NON_CONTINUOUS | MIPI_DSI_MODE_LPM,
+	.init_sequence = j606f_boe_init_sequence,
+	.has_dcs_backlight = true,
+};
+
 static void nt36523_reset(struct panel_info *pinfo)
 {
 	gpiod_set_value_cansleep(pinfo->reset_gpio, 1);
@@ -827,6 +1309,10 @@ static int nt36523_probe(struct mipi_dsi_device *dsi)
 }
 
 static const struct of_device_id nt36523_of_match[] = {
+	{
+		.compatible = "lenovo,j606f-boe-nt36523w",
+		.data = &j606f_boe_desc,
+	},
 	{
 		.compatible = "xiaomi,elish-boe-nt36523",
 		.data = &elish_boe_desc,
-- 
GitLab


From 3778724bec53a3585cde6aed8c75ae3f085c3dcb Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sat, 15 Apr 2023 21:06:06 +0200
Subject: [PATCH 0076/1791] drm/nouveau/debugfs: Move an expression into a
 function call parameter in nouveau_debugfs_pstate_set()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “nouveau_debugfs_pstate_set”.

Thus avoid the risk for undefined behaviour by moving the usage
of an expression into a parameter for a function call at the end.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1903e057-57d3-e12e-da37-008e23afd2e2@web.de
---
 drivers/gpu/drm/nouveau/nouveau_debugfs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
index 2a36d1ca8fdab..44e26b6e74c7b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -144,7 +144,6 @@ nouveau_debugfs_pstate_set(struct file *file, const char __user *ubuf,
 	struct seq_file *m = file->private_data;
 	struct drm_device *drm = m->private;
 	struct nouveau_debugfs *debugfs = nouveau_debugfs(drm);
-	struct nvif_object *ctrl = &debugfs->ctrl;
 	struct nvif_control_pstate_user_v0 args = { .pwrsrc = -EINVAL };
 	char buf[32] = {}, *tmp, *cur = buf;
 	long value, ret;
@@ -188,7 +187,8 @@ nouveau_debugfs_pstate_set(struct file *file, const char __user *ubuf,
 		return ret;
 	}
 
-	ret = nvif_mthd(ctrl, NVIF_CONTROL_PSTATE_USER, &args, sizeof(args));
+	ret = nvif_mthd(&debugfs->ctrl, NVIF_CONTROL_PSTATE_USER,
+			&args, sizeof(args));
 	pm_runtime_put_autosuspend(drm->dev);
 	if (ret < 0)
 		return ret;
-- 
GitLab


From e1b570db8687c993e7a87516387a3bb3b5e63ac6 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sat, 15 Apr 2023 21:24:43 +0200
Subject: [PATCH 0077/1791] drm/nouveau/debugfs: Move a variable assignment
 behind a null pointer check in nouveau_debugfs_pstate_get()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “nouveau_debugfs_pstate_get”.

Thus avoid the risk for undefined behaviour by moving the assignment
for the variable “ctrl” behind the null pointer check.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/5b7b99f7-1692-74e5-4b1e-cfa14dad7c57@web.de
---
 drivers/gpu/drm/nouveau/nouveau_debugfs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
index 44e26b6e74c7b..a859a086f308c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -73,13 +73,14 @@ nouveau_debugfs_pstate_get(struct seq_file *m, void *data)
 {
 	struct drm_device *drm = m->private;
 	struct nouveau_debugfs *debugfs = nouveau_debugfs(drm);
-	struct nvif_object *ctrl = &debugfs->ctrl;
+	struct nvif_object *ctrl;
 	struct nvif_control_pstate_info_v0 info = {};
 	int ret, i;
 
 	if (!debugfs)
 		return -ENODEV;
 
+	ctrl = &debugfs->ctrl;
 	ret = nvif_mthd(ctrl, NVIF_CONTROL_PSTATE_INFO, &info, sizeof(info));
 	if (ret)
 		return ret;
-- 
GitLab


From df8133d0c3e53e98df3c78c30f31530f580f1bce Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sat, 15 Apr 2023 21:48:47 +0200
Subject: [PATCH 0078/1791] drm/nouveau/debugfs: Use seq_putc() in
 nouveau_debugfs_pstate_get()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A single character (line break) should be put into a sequence.
Thus use the corresponding function “seq_putc”.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e6f9b34e-58d9-b1aa-8b12-8d4a55153d1d@web.de
---
 drivers/gpu/drm/nouveau/nouveau_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
index a859a086f308c..13c82eea88286 100644
--- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -132,7 +132,7 @@ nouveau_debugfs_pstate_get(struct seq_file *m, void *data)
 				seq_printf(m, " DC");
 		}
 
-		seq_printf(m, "\n");
+		seq_putc(m, '\n');
 	}
 
 	return 0;
-- 
GitLab


From 6ef481d45f8b37dc9cca11d7126bd2c99e8a0943 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sat, 15 Apr 2023 22:02:31 +0200
Subject: [PATCH 0079/1791] drm/nouveau/debugfs: Replace five seq_printf()
 calls by seq_puts() in nouveau_debugfs_pstate_get()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Five strings which did not contain a data format specification should
be put into a sequence. Thus use the corresponding function “seq_puts”.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/bdee7e41-094a-91d1-3f6f-56df402d70b7@web.de
---
 drivers/gpu/drm/nouveau/nouveau_debugfs.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
index 13c82eea88286..99d022a91afc5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c
@@ -120,16 +120,16 @@ nouveau_debugfs_pstate_get(struct seq_file *m, void *data)
 
 		if (state >= 0) {
 			if (info.ustate_ac == state)
-				seq_printf(m, " AC");
+				seq_puts(m, " AC");
 			if (info.ustate_dc == state)
-				seq_printf(m, " DC");
+				seq_puts(m, " DC");
 			if (info.pstate == state)
-				seq_printf(m, " *");
+				seq_puts(m, " *");
 		} else {
 			if (info.ustate_ac < -1)
-				seq_printf(m, " AC");
+				seq_puts(m, " AC");
 			if (info.ustate_dc < -1)
-				seq_printf(m, " DC");
+				seq_puts(m, " DC");
 		}
 
 		seq_putc(m, '\n');
-- 
GitLab


From 5149545dbcbb15105312014201b9c0aa859c6044 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sat, 15 Apr 2023 22:30:30 +0200
Subject: [PATCH 0080/1791] drm/nouveau/bios/power_budget: Move an expression
 into a macro call parameter in nvbios_power_budget_header()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “nvbios_power_budget_header”.

Thus avoid the risk for undefined behaviour by moving the usage
of an expression into a parameter for a macro call in one if branch.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/529fe82e-59f0-71a2-1e33-b051f833b67c@web.de
---
 drivers/gpu/drm/nouveau/nvkm/subdev/bios/power_budget.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/power_budget.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/power_budget.c
index 03d2f970a29ff..2ba992bdb19dc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/power_budget.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/power_budget.c
@@ -59,7 +59,6 @@ int
 nvbios_power_budget_header(struct nvkm_bios *bios,
                            struct nvbios_power_budget *budget)
 {
-	struct nvkm_subdev *subdev = &bios->subdev;
 	u8 ver, hdr, cnt, len, cap_entry;
 	u32 header;
 
@@ -82,7 +81,7 @@ nvbios_power_budget_header(struct nvkm_bios *bios,
 	}
 
 	if (cap_entry >= cnt && cap_entry != 0xff) {
-		nvkm_warn(subdev,
+		nvkm_warn(&bios->subdev,
 		          "invalid cap_entry in power budget table found\n");
 		budget->cap_entry = 0xff;
 		return -EINVAL;
-- 
GitLab


From 41239aa4f751fbc39481a4af726faa347a26da83 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sun, 16 Apr 2023 07:45:54 +0200
Subject: [PATCH 0081/1791] drm/nouveau/clk: Move a variable assignment behind
 a null pointer check in nvkm_pstate_new()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “nvkm_pstate_new”.

Thus avoid the risk for undefined behaviour by moving the assignment
for the variable “cstate” behind the null pointer check.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/6435d829-2c9b-a3f7-c8cd-cf923508a092@web.de
---
 drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c
index da07a2fbef062..178dc56909c27 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/base.c
@@ -417,7 +417,6 @@ nvkm_pstate_new(struct nvkm_clk *clk, int idx)
 		return 0;
 
 	pstate = kzalloc(sizeof(*pstate), GFP_KERNEL);
-	cstate = &pstate->base;
 	if (!pstate)
 		return -ENOMEM;
 
@@ -427,6 +426,7 @@ nvkm_pstate_new(struct nvkm_clk *clk, int idx)
 	pstate->fanspeed = perfE.fanspeed;
 	pstate->pcie_speed = perfE.pcie_speed;
 	pstate->pcie_width = perfE.pcie_width;
+	cstate = &pstate->base;
 	cstate->voltage = perfE.voltage;
 	cstate->domain[nv_clk_src_core] = perfE.core;
 	cstate->domain[nv_clk_src_shader] = perfE.shader;
-- 
GitLab


From 488c1ce6a8f3d46e20440b2e240e096a20259fcc Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sun, 16 Apr 2023 08:18:40 +0200
Subject: [PATCH 0082/1791] drm/nouveau/pci: Move a variable assignment behind
 condition checks in nvkm_pcie_set_link()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “nvkm_pcie_set_link”.

Thus avoid the risk for undefined behaviour by moving the assignment
for the variable “subdev” behind some condition checks.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b31019b1-bb73-ed93-3378-e551e17a4f32@web.de
---
 drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
index d71e5db5028ab..b295f100e1c13 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
@@ -114,7 +114,7 @@ nvkm_pcie_init(struct nvkm_pci *pci)
 int
 nvkm_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 {
-	struct nvkm_subdev *subdev = &pci->subdev;
+	struct nvkm_subdev *subdev;
 	enum nvkm_pcie_speed cur_speed, max_speed;
 	struct pci_bus *pbus;
 	int ret;
@@ -126,6 +126,7 @@ nvkm_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 	if (!pci->func->pcie.set_link)
 		return -ENOSYS;
 
+	subdev = &pci->subdev;
 	nvkm_trace(subdev, "requested %s\n", nvkm_pcie_speeds[speed]);
 
 	if (pci->func->pcie.version(pci) < 2) {
-- 
GitLab


From 8a89e1323d7dc90fa551cf2167ab838013641788 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sun, 16 Apr 2023 08:45:31 +0200
Subject: [PATCH 0083/1791] drm/nouveau/pci: Move an expression into a function
 call parameter in nvkm_pcie_set_link()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The variable “pbus” was read only once in the implementation of
the function “nvkm_pcie_set_link”.
Thus move the usage of an expression into a parameter for a function call.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/04e2c1f5-26c7-9a23-5861-3d8757556cdd@web.de
---
 drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
index b295f100e1c13..dd18d9d0bade6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/pcie.c
@@ -116,12 +116,10 @@ nvkm_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 {
 	struct nvkm_subdev *subdev;
 	enum nvkm_pcie_speed cur_speed, max_speed;
-	struct pci_bus *pbus;
 	int ret;
 
 	if (!pci || !pci_is_pcie(pci->pdev))
 		return 0;
-	pbus = pci->pdev->bus;
 
 	if (!pci->func->pcie.set_link)
 		return -ENOSYS;
@@ -135,7 +133,7 @@ nvkm_pcie_set_link(struct nvkm_pci *pci, enum nvkm_pcie_speed speed, u8 width)
 	}
 
 	cur_speed = pci->func->pcie.cur_speed(pci);
-	max_speed = min(nvkm_pcie_speed(pbus->max_bus_speed),
+	max_speed = min(nvkm_pcie_speed(pci->pdev->bus->max_bus_speed),
 			pci->func->pcie.max_speed(pci));
 
 	nvkm_trace(subdev, "current speed: %s\n", nvkm_pcie_speeds[cur_speed]);
-- 
GitLab


From 4aa35a0130d6b8afbefc9ef530a521fb0fb9b8e1 Mon Sep 17 00:00:00 2001
From: Markus Elfring <Markus.Elfring@web.de>
Date: Sun, 16 Apr 2023 10:50:12 +0200
Subject: [PATCH 0084/1791] drm/nouveau/therm: Move an assignment statement
 behind a null pointer check in two functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the functions “nvkm_fanpwm_create” and “nvkm_fantog_create”.

Thus avoid the risk for undefined behaviour by moving the assignment
for the data structure member “fan” behind two null pointer checks.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/3d0215dc-74d4-2c42-2eee-7a5fcf62b9f1@web.de
---
 drivers/gpu/drm/nouveau/nvkm/subdev/therm/fanpwm.c | 2 +-
 drivers/gpu/drm/nouveau/nvkm/subdev/therm/fantog.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fanpwm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fanpwm.c
index 340f37a299dce..b13ba9b2f6be0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fanpwm.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fanpwm.c
@@ -98,10 +98,10 @@ nvkm_fanpwm_create(struct nvkm_therm *therm, struct dcb_gpio_func *func)
 		return -ENODEV;
 
 	fan = kzalloc(sizeof(*fan), GFP_KERNEL);
-	therm->fan = &fan->base;
 	if (!fan)
 		return -ENOMEM;
 
+	therm->fan = &fan->base;
 	fan->base.type = "PWM";
 	fan->base.get = nvkm_fanpwm_get;
 	fan->base.set = nvkm_fanpwm_set;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fantog.c b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fantog.c
index ff9fbe7950e5c..bfdf4ca5625c9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fantog.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/therm/fantog.c
@@ -100,10 +100,10 @@ nvkm_fantog_create(struct nvkm_therm *therm, struct dcb_gpio_func *func)
 	}
 
 	fan = kzalloc(sizeof(*fan), GFP_KERNEL);
-	therm->fan = &fan->base;
 	if (!fan)
 		return -ENOMEM;
 
+	therm->fan = &fan->base;
 	fan->base.type = "toggle";
 	fan->base.get = nvkm_fantog_get;
 	fan->base.set = nvkm_fantog_set;
-- 
GitLab


From 088248f4b0ee089cb78a318411d23566e3737dd1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 17 Apr 2023 11:57:42 +0300
Subject: [PATCH 0085/1791] drm/i915/pps: use intel_de_rmw() for panel unlock

Use rmw where needed.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417085742.793379-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_pps.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index 7f9926672a6a5..4f0b0cca03cca 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -1653,12 +1653,9 @@ void intel_pps_unlock_regs_wa(struct drm_i915_private *dev_priv)
 	 */
 	pps_num = intel_num_pps(dev_priv);
 
-	for (pps_idx = 0; pps_idx < pps_num; pps_idx++) {
-		u32 val = intel_de_read(dev_priv, PP_CONTROL(pps_idx));
-
-		val = (val & ~PANEL_UNLOCK_MASK) | PANEL_UNLOCK_REGS;
-		intel_de_write(dev_priv, PP_CONTROL(pps_idx), val);
-	}
+	for (pps_idx = 0; pps_idx < pps_num; pps_idx++)
+		intel_de_rmw(dev_priv, PP_CONTROL(pps_idx),
+			     PANEL_UNLOCK_MASK, PANEL_UNLOCK_REGS);
 }
 
 void intel_pps_setup(struct drm_i915_private *i915)
-- 
GitLab


From d1f3b5e92cbab10b4710ba33e20b264c852d19aa Mon Sep 17 00:00:00 2001
From: Andi Shyti <andi.shyti@linux.intel.com>
Date: Tue, 18 Apr 2023 01:53:56 +0200
Subject: [PATCH 0086/1791] drm/i915: Make IRQ reset and postinstall multi-gt
 aware

In multi-gt systems IRQs need to be reset and enabled per GT.

This might add some redundancy when handling interrupts for
engines that might not exist in every tile, but helps to keep the
code cleaner and more understandable.

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417235356.1291060-1-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index d24bdea65a3dc..53c83e2570553 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2762,12 +2762,15 @@ static void gen11_irq_reset(struct drm_i915_private *dev_priv)
 
 static void dg1_irq_reset(struct drm_i915_private *dev_priv)
 {
-	struct intel_gt *gt = to_gt(dev_priv);
-	struct intel_uncore *uncore = gt->uncore;
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	struct intel_gt *gt;
+	unsigned int i;
 
 	dg1_master_intr_disable(dev_priv->uncore.regs);
 
-	gen11_gt_irq_reset(gt);
+	for_each_gt(gt, dev_priv, i)
+		gen11_gt_irq_reset(gt);
+
 	gen11_display_irq_reset(dev_priv);
 
 	GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_);
@@ -3425,11 +3428,13 @@ static void gen11_irq_postinstall(struct drm_i915_private *dev_priv)
 
 static void dg1_irq_postinstall(struct drm_i915_private *dev_priv)
 {
-	struct intel_gt *gt = to_gt(dev_priv);
-	struct intel_uncore *uncore = gt->uncore;
+	struct intel_uncore *uncore = &dev_priv->uncore;
 	u32 gu_misc_masked = GEN11_GU_MISC_GSE;
+	struct intel_gt *gt;
+	unsigned int i;
 
-	gen11_gt_irq_postinstall(gt);
+	for_each_gt(gt, dev_priv, i)
+		gen11_gt_irq_postinstall(gt);
 
 	GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked);
 
-- 
GitLab


From 40d06b0fae915ec8bbfbde69ee1cfb2d9a2b7a53 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 17 Apr 2023 16:17:25 +0300
Subject: [PATCH 0087/1791] drm/i915: Introduce <platform>_hotplug_mask()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pair each <platform>_hotplug_enables() function with
a corresponding <platform>_hotplug_mask() function so that
we can determine right bits to clear on a per hpd_pin basis.
We'll need this for turning on HPD sense for a specific
encoder rather than just all of them.

v2: Drop the unused 'i915' param (Jani)
v3: Drop the _foo_hotplug_enables() redirection too
v4: Deal with mtp

Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com> #v3
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417131728.7705-2-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 201 ++++++++++++++++++++++----------
 1 file changed, 137 insertions(+), 64 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index dea1a117f3fa1..273ffce9807ab 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2931,6 +2931,22 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
+static u32 ibx_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return PORTA_HOTPLUG_ENABLE;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_MASK;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_MASK;
+	case HPD_PORT_D:
+		return PORTD_HOTPLUG_ENABLE | PORTD_PULSE_DURATION_MASK;
+	default:
+		return 0;
+	}
+}
+
 static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
@@ -2965,13 +2981,10 @@ static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
 	 * The pulse duration bits are reserved on LPT+.
 	 */
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 PORTA_HOTPLUG_ENABLE |
-			 PORTB_HOTPLUG_ENABLE |
-			 PORTC_HOTPLUG_ENABLE |
-			 PORTD_HOTPLUG_ENABLE |
-			 PORTB_PULSE_DURATION_MASK |
-			 PORTC_PULSE_DURATION_MASK |
-			 PORTD_PULSE_DURATION_MASK,
+			 ibx_hotplug_mask(HPD_PORT_A) |
+			 ibx_hotplug_mask(HPD_PORT_B) |
+			 ibx_hotplug_mask(HPD_PORT_C) |
+			 ibx_hotplug_mask(HPD_PORT_D),
 			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
 }
 
@@ -2987,53 +3000,63 @@ static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	ibx_hpd_detection_setup(dev_priv);
 }
 
-static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder)
+static u32 icp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
 {
-	switch (encoder->hpd_pin) {
+	switch (hpd_pin) {
 	case HPD_PORT_A:
 	case HPD_PORT_B:
 	case HPD_PORT_C:
 	case HPD_PORT_D:
-		return SHOTPLUG_CTL_DDI_HPD_ENABLE(encoder->hpd_pin);
+		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
 	default:
 		return 0;
 	}
 }
 
-static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
+static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder)
 {
-	switch (encoder->hpd_pin) {
+	return icp_ddi_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 icp_tc_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
 	case HPD_PORT_TC1:
 	case HPD_PORT_TC2:
 	case HPD_PORT_TC3:
 	case HPD_PORT_TC4:
 	case HPD_PORT_TC5:
 	case HPD_PORT_TC6:
-		return ICP_TC_HPD_ENABLE(encoder->hpd_pin);
+		return ICP_TC_HPD_ENABLE(hpd_pin);
 	default:
 		return 0;
 	}
 }
 
+static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
+{
+	return icp_tc_hotplug_mask(encoder->hpd_pin);
+}
+
 static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI,
-			 SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_A) |
-			 SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_B) |
-			 SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_C) |
-			 SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_D),
+			 icp_ddi_hotplug_mask(HPD_PORT_A) |
+			 icp_ddi_hotplug_mask(HPD_PORT_B) |
+			 icp_ddi_hotplug_mask(HPD_PORT_C) |
+			 icp_ddi_hotplug_mask(HPD_PORT_D),
 			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
 }
 
 static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC1) |
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC2) |
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC3) |
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC4) |
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC5) |
-			 ICP_TC_HPD_ENABLE(HPD_PORT_TC6),
+			 icp_tc_hotplug_mask(HPD_PORT_TC1) |
+			 icp_tc_hotplug_mask(HPD_PORT_TC2) |
+			 icp_tc_hotplug_mask(HPD_PORT_TC3) |
+			 icp_tc_hotplug_mask(HPD_PORT_TC4) |
+			 icp_tc_hotplug_mask(HPD_PORT_TC5) |
+			 icp_tc_hotplug_mask(HPD_PORT_TC6),
 			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
 }
 
@@ -3053,21 +3076,26 @@ static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	icp_tc_hpd_detection_setup(dev_priv);
 }
 
-static u32 gen11_hotplug_enables(struct intel_encoder *encoder)
+static u32 gen11_hotplug_mask(enum hpd_pin hpd_pin)
 {
-	switch (encoder->hpd_pin) {
+	switch (hpd_pin) {
 	case HPD_PORT_TC1:
 	case HPD_PORT_TC2:
 	case HPD_PORT_TC3:
 	case HPD_PORT_TC4:
 	case HPD_PORT_TC5:
 	case HPD_PORT_TC6:
-		return GEN11_HOTPLUG_CTL_ENABLE(encoder->hpd_pin);
+		return GEN11_HOTPLUG_CTL_ENABLE(hpd_pin);
 	default:
 		return 0;
 	}
 }
 
+static u32 gen11_hotplug_enables(struct intel_encoder *encoder)
+{
+	return gen11_hotplug_mask(encoder->hpd_pin);
+}
+
 static void dg1_hpd_invert(struct drm_i915_private *i915)
 {
 	u32 val = (INVERT_DDIA_HPD |
@@ -3086,24 +3114,24 @@ static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
 static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL,
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC1) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC2) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC3) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC4) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC5) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC6),
+			 gen11_hotplug_mask(HPD_PORT_TC1) |
+			 gen11_hotplug_mask(HPD_PORT_TC2) |
+			 gen11_hotplug_mask(HPD_PORT_TC3) |
+			 gen11_hotplug_mask(HPD_PORT_TC4) |
+			 gen11_hotplug_mask(HPD_PORT_TC5) |
+			 gen11_hotplug_mask(HPD_PORT_TC6),
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
 static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC1) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC2) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC3) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC4) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC5) |
-			 GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC6),
+			 gen11_hotplug_mask(HPD_PORT_TC1) |
+			 gen11_hotplug_mask(HPD_PORT_TC2) |
+			 gen11_hotplug_mask(HPD_PORT_TC3) |
+			 gen11_hotplug_mask(HPD_PORT_TC4) |
+			 gen11_hotplug_mask(HPD_PORT_TC5) |
+			 gen11_hotplug_mask(HPD_PORT_TC6),
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
@@ -3125,45 +3153,55 @@ static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 		icp_hpd_irq_setup(dev_priv);
 }
 
-static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder)
+static u32 mtp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
 {
-	switch (encoder->hpd_pin) {
+	switch (hpd_pin) {
 	case HPD_PORT_A:
 	case HPD_PORT_B:
-		return SHOTPLUG_CTL_DDI_HPD_ENABLE(encoder->hpd_pin);
+		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
 	default:
 		return 0;
 	}
 }
 
-static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
+static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder)
 {
-	switch (encoder->hpd_pin) {
+	return mtp_ddi_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 mtp_tc_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
 	case HPD_PORT_TC1:
 	case HPD_PORT_TC2:
 	case HPD_PORT_TC3:
 	case HPD_PORT_TC4:
-		return ICP_TC_HPD_ENABLE(encoder->hpd_pin);
+		return ICP_TC_HPD_ENABLE(hpd_pin);
 	default:
 		return 0;
 	}
 }
 
+static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
+{
+	return mtp_tc_hotplug_mask(encoder->hpd_pin);
+}
+
 static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     (SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_A) |
-		      SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_B)),
+		     mtp_ddi_hotplug_mask(HPD_PORT_A) |
+		     mtp_ddi_hotplug_mask(HPD_PORT_B),
 		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
 }
 
 static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
-		     (ICP_TC_HPD_ENABLE(HPD_PORT_TC1) |
-		      ICP_TC_HPD_ENABLE(HPD_PORT_TC2) |
-		      ICP_TC_HPD_ENABLE(HPD_PORT_TC3) |
-		      ICP_TC_HPD_ENABLE(HPD_PORT_TC4)),
+		     mtp_tc_hotplug_mask(HPD_PORT_TC1) |
+		     mtp_tc_hotplug_mask(HPD_PORT_TC2) |
+		     mtp_tc_hotplug_mask(HPD_PORT_TC3) |
+		     mtp_tc_hotplug_mask(HPD_PORT_TC4),
 		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
 }
 
@@ -3235,9 +3273,9 @@ static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
 		mtp_hpd_irq_setup(i915);
 }
 
-static u32 spt_hotplug_enables(struct intel_encoder *encoder)
+static u32 spt_hotplug_mask(enum hpd_pin hpd_pin)
 {
-	switch (encoder->hpd_pin) {
+	switch (hpd_pin) {
 	case HPD_PORT_A:
 		return PORTA_HOTPLUG_ENABLE;
 	case HPD_PORT_B:
@@ -3251,9 +3289,14 @@ static u32 spt_hotplug_enables(struct intel_encoder *encoder)
 	}
 }
 
-static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
+static u32 spt_hotplug_enables(struct intel_encoder *encoder)
 {
-	switch (encoder->hpd_pin) {
+	return spt_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 spt_hotplug2_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
 	case HPD_PORT_E:
 		return PORTE_HOTPLUG_ENABLE;
 	default:
@@ -3261,6 +3304,11 @@ static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
 	}
 }
 
+static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
+{
+	return spt_hotplug2_mask(encoder->hpd_pin);
+}
+
 static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	/* Display WA #1179 WaHardHangonHotPlug: cnp */
@@ -3271,13 +3319,14 @@ static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 
 	/* Enable digital hotplug on the PCH */
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 PORTA_HOTPLUG_ENABLE |
-			 PORTB_HOTPLUG_ENABLE |
-			 PORTC_HOTPLUG_ENABLE |
-			 PORTD_HOTPLUG_ENABLE,
+			 spt_hotplug_mask(HPD_PORT_A) |
+			 spt_hotplug_mask(HPD_PORT_B) |
+			 spt_hotplug_mask(HPD_PORT_C) |
+			 spt_hotplug_mask(HPD_PORT_D),
 			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables));
 
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, PORTE_HOTPLUG_ENABLE,
+	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2,
+			 spt_hotplug2_mask(HPD_PORT_E),
 			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
 }
 
@@ -3296,6 +3345,17 @@ static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	spt_hpd_detection_setup(dev_priv);
 }
 
+static u32 ilk_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return DIGITAL_PORTA_HOTPLUG_ENABLE |
+			DIGITAL_PORTA_PULSE_DURATION_MASK;
+	default:
+		return 0;
+	}
+}
+
 static u32 ilk_hotplug_enables(struct intel_encoder *encoder)
 {
 	switch (encoder->hpd_pin) {
@@ -3315,7 +3375,7 @@ static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
 	 * The pulse duration bits are reserved on HSW+.
 	 */
 	intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 DIGITAL_PORTA_HOTPLUG_ENABLE | DIGITAL_PORTA_PULSE_DURATION_MASK,
+			 ilk_hotplug_mask(HPD_PORT_A),
 			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
 }
 
@@ -3336,6 +3396,20 @@ static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
 	ibx_hpd_irq_setup(dev_priv);
 }
 
+static u32 bxt_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return PORTA_HOTPLUG_ENABLE | BXT_DDIA_HPD_INVERT;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE | BXT_DDIB_HPD_INVERT;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE | BXT_DDIC_HPD_INVERT;
+	default:
+		return 0;
+	}
+}
+
 static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
 {
 	u32 hotplug;
@@ -3364,10 +3438,9 @@ static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
 static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 PORTA_HOTPLUG_ENABLE |
-			 PORTB_HOTPLUG_ENABLE |
-			 PORTC_HOTPLUG_ENABLE |
-			 BXT_DDI_HPD_INVERT_MASK,
+			 bxt_hotplug_mask(HPD_PORT_A) |
+			 bxt_hotplug_mask(HPD_PORT_B) |
+			 bxt_hotplug_mask(HPD_PORT_C),
 			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
 }
 
-- 
GitLab


From d28cdc43b43b77287d7839ef1b94fdaa24d2a444 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 17 Apr 2023 16:17:26 +0300
Subject: [PATCH 0088/1791] drm/i915: Introduce intel_hpd_enable_detection()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a mechanism by which we can enable the HPD sense for
individual encoders.

This will be used during eDP probing to figure out if
anything is actually connected. The normal intel_hpd_irq_setup()
thing doesn't work since we only do that after probing the
outputs, and we only enable HPD sense for encoders that were
successfully probed.

The other idea that crossed my minds was to just turn on
HPD sense for all pins before output probing and let hpd_irq_setup()
clean it up afterwards. But that doesn't work for BXT/GLK where
the HPD invert information comes from the VBT child device.
So looks like this really needs to be per-encoder.

v2: Give it a better name (Jani)
v3: Deal with mtl

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com> #v2
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417131728.7705-3-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 199 ++++++++++++++++++++++++++++++--
 drivers/gpu/drm/i915/i915_irq.h |   2 +
 2 files changed, 193 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 273ffce9807ab..3505fafcfb7b0 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2988,6 +2988,15 @@ static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
 }
 
+static void ibx_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 ibx_hotplug_mask(encoder->hpd_pin),
+			 ibx_hotplug_enables(encoder));
+}
+
 static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3048,6 +3057,15 @@ static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
 }
 
+static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_DDI,
+			 icp_ddi_hotplug_mask(encoder->hpd_pin),
+			 icp_ddi_hotplug_enables(encoder));
+}
+
 static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
@@ -3060,6 +3078,21 @@ static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
 }
 
+static void icp_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_TC,
+			 icp_tc_hotplug_mask(encoder->hpd_pin),
+			 icp_tc_hotplug_enables(encoder));
+}
+
+static void icp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	icp_ddi_hpd_enable_detection(encoder);
+	icp_tc_hpd_enable_detection(encoder);
+}
+
 static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3105,6 +3138,14 @@ static void dg1_hpd_invert(struct drm_i915_private *i915)
 	intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val);
 }
 
+static void dg1_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	dg1_hpd_invert(i915);
+	icp_hpd_enable_detection(encoder);
+}
+
 static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	dg1_hpd_invert(dev_priv);
@@ -3123,6 +3164,15 @@ static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
+static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, GEN11_TC_HOTPLUG_CTL,
+			 gen11_hotplug_mask(encoder->hpd_pin),
+			 gen11_hotplug_enables(encoder));
+}
+
 static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
@@ -3135,6 +3185,26 @@ static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
+static void gen11_tbt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, GEN11_TBT_HOTPLUG_CTL,
+			 gen11_hotplug_mask(encoder->hpd_pin),
+			 gen11_hotplug_enables(encoder));
+}
+
+static void gen11_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	gen11_tc_hpd_enable_detection(encoder);
+	gen11_tbt_hpd_enable_detection(encoder);
+
+	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+		icp_hpd_enable_detection(encoder);
+}
+
 static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3195,6 +3265,15 @@ static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
 		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
 }
 
+static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     mtp_ddi_hotplug_mask(encoder->hpd_pin),
+		     mtp_ddi_hotplug_enables(encoder));
+}
+
 static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
@@ -3205,6 +3284,15 @@ static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
 		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
 }
 
+static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     mtp_tc_hotplug_mask(encoder->hpd_pin),
+		     mtp_tc_hotplug_enables(encoder));
+}
+
 static void mtp_hpd_invert(struct drm_i915_private *i915)
 {
 	u32 val = (INVERT_DDIA_HPD |
@@ -3219,6 +3307,15 @@ static void mtp_hpd_invert(struct drm_i915_private *i915)
 	intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val);
 }
 
+static void mtp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	mtp_hpd_invert(i915);
+	mtp_ddi_hpd_enable_detection(encoder);
+	mtp_tc_hpd_enable_detection(encoder);
+}
+
 static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3235,25 +3332,50 @@ static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
 	mtp_tc_hpd_detection_setup(i915);
 }
 
+static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin)
+{
+	return hpd_pin >= HPD_PORT_TC1 && hpd_pin <= HPD_PORT_TC4;
+}
+
+static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915,
+					     enum hpd_pin hpd_pin, bool enable)
+{
+	u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
+		XELPDP_DP_ALT_HOTPLUG_ENABLE;
+
+	if (!is_xelpdp_pica_hpd_pin(hpd_pin))
+		return;
+
+	intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(hpd_pin),
+		     mask, enable ? mask : 0);
+}
+
+static void xelpdp_pica_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	_xelpdp_pica_hpd_detection_setup(i915, encoder->hpd_pin, true);
+}
+
 static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	struct intel_encoder *encoder;
-	enum hpd_pin pin;
 	u32 available_pins = 0;
+	enum hpd_pin pin;
 
 	BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS);
 
 	for_each_intel_encoder(&i915->drm, encoder)
 		available_pins |= BIT(encoder->hpd_pin);
 
-	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
-		u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
-			   XELPDP_DP_ALT_HOTPLUG_ENABLE;
+	for_each_hpd_pin(pin)
+		_xelpdp_pica_hpd_detection_setup(i915, pin, available_pins & BIT(pin));
+}
 
-		intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(pin),
-			     mask,
-			     available_pins & BIT(pin) ?  mask : 0);
-	}
+static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	xelpdp_pica_hpd_enable_detection(encoder);
+	mtp_hpd_enable_detection(encoder);
 }
 
 static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
@@ -3330,6 +3452,26 @@ static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
 }
 
+static void spt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	/* Display WA #1179 WaHardHangonHotPlug: cnp */
+	if (HAS_PCH_CNP(i915)) {
+		intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1,
+				 CHASSIS_CLK_REQ_DURATION_MASK,
+				 CHASSIS_CLK_REQ_DURATION(0xf));
+	}
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 spt_hotplug_mask(encoder->hpd_pin),
+			 spt_hotplug_enables(encoder));
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG2,
+			 spt_hotplug2_mask(encoder->hpd_pin),
+			 spt_hotplug2_enables(encoder));
+}
+
 static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3379,6 +3521,17 @@ static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
 }
 
+static void ilk_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
+			 ilk_hotplug_mask(encoder->hpd_pin),
+			 ilk_hotplug_enables(encoder));
+
+	ibx_hpd_enable_detection(encoder);
+}
+
 static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -3444,6 +3597,15 @@ static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
 }
 
+static void bxt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 bxt_hotplug_mask(encoder->hpd_pin),
+			 bxt_hotplug_enables(encoder));
+}
+
 static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_irqs, enabled_irqs;
@@ -4121,6 +4283,15 @@ static void i965_irq_postinstall(struct drm_i915_private *dev_priv)
 	i915_enable_asle_pipestat(dev_priv);
 }
 
+static void i915_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	u32 hotplug_en = hpd_mask_i915[encoder->hpd_pin];
+
+	/* HPD sense and interrupt enable are one and the same */
+	i915_hotplug_interrupt_update(i915, hotplug_en, hotplug_en);
+}
+
 static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv)
 {
 	u32 hotplug_en;
@@ -4206,12 +4377,16 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 }
 
 struct intel_hotplug_funcs {
+	/* Enable HPD sense and interrupts for all present encoders */
 	void (*hpd_irq_setup)(struct drm_i915_private *i915);
+	/* Enable HPD sense for a single encoder */
+	void (*hpd_enable_detection)(struct intel_encoder *encoder);
 };
 
 #define HPD_FUNCS(platform)					 \
 static const struct intel_hotplug_funcs platform##_hpd_funcs = { \
 	.hpd_irq_setup = platform##_hpd_irq_setup,		 \
+	.hpd_enable_detection = platform##_hpd_enable_detection, \
 }
 
 HPD_FUNCS(i915);
@@ -4224,6 +4399,14 @@ HPD_FUNCS(spt);
 HPD_FUNCS(ilk);
 #undef HPD_FUNCS
 
+void intel_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	if (i915->display.funcs.hotplug)
+		i915->display.funcs.hotplug->hpd_enable_detection(encoder);
+}
+
 void intel_hpd_irq_setup(struct drm_i915_private *i915)
 {
 	if (i915->display_irqs_enabled && i915->display.funcs.hotplug)
diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h
index 03ee4c8b1ed38..3717a66f97c60 100644
--- a/drivers/gpu/drm/i915/i915_irq.h
+++ b/drivers/gpu/drm/i915/i915_irq.h
@@ -17,6 +17,7 @@ struct drm_device;
 struct drm_display_mode;
 struct drm_i915_private;
 struct intel_crtc;
+struct intel_encoder;
 struct intel_uncore;
 
 void intel_irq_init(struct drm_i915_private *dev_priv);
@@ -37,6 +38,7 @@ i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv);
 void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv);
 
+void intel_hpd_enable_detection(struct intel_encoder *encoder);
 void intel_hpd_irq_setup(struct drm_i915_private *i915);
 void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
 				   u32 mask,
-- 
GitLab


From cfe5bdfb27fa234505e96f7775d32415ed705f94 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 17 Apr 2023 16:17:27 +0300
Subject: [PATCH 0089/1791] drm/i915: Check HPD live state during eDP probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We need to untangle the mess where some SKL machines (at least)
declare both DDI A and DDI E to be present in their VBT, and
both using AUX A. DDI A is a ghost eDP, wheres DDI E may be a
real DP->VGA converter.

Currently that is handled by checking the VBT child devices
for conflicts before output probing. But that kind of solution
will not work for the ADL phantom dual eDP VBTs. I think on
those we just have to probe the eDP first. And would be nice
to use the same probe scheme for everything.

On these SKL systems if we probe DDI A first (which is only
natural given it's declared by VBT first) we will get an answer
via AUX, but it came from the DP->VGA converter hooked to the
DDI E, not DDI A. Thus we mistakenly register eDP on DDI A
and screw up the real DP device in DDI E.

To fix this let's check the HPD live state during the eDP probe.
If we got an answer via DPCD but HPD is still down let's assume
we got the answer from someone else.

Smoke tested on all my eDP machines (ilk,hsw-ult,tgl,adl) and
I also tested turning off all HPD hardware prior to loading
i915 to make sure it all comes up properly. And I simulated
the failure path too by not turning on HPD sense and that
correctly gave up on eDP.

I *think* Windows might just fully depend on HPD here. I
couldn't really find any other way they probe displays. And
I did find code where they also check the live state prior
to AUX transfers (something Imre and I have also talked
about perhaps doing). That would also solve this as we'd
not succeed in the eDP probe DPCD reads.

Other solutions I've considered:

- Reintrduce DDI strap checks on SKL. Unfortunately we just
  don't have any idea how reliable they are on real production
  hardware, and commit 5a2376d1360b ("drm/i915/skl: WaIgnoreDDIAStrap
  is forever, always init DDI A") does suggest that not very.
  Sadly that commit is very poor in details :/

  Also the systems (Asrock B250M-HDV at least) fixed by commit
  41e35ffb380b ("drm/i915: Favor last VBT child device with
  conflicting AUX ch/DDC pin") might still not work since we
  don't know what their straps indicate. Stupid me for not
  asking the reporter to check those at the time :(

  We have currently two CI machines (fi-cfl-guc,fi-cfl-8700k
  both MS-7B54/Z370M) that also declare both DDI A and DDI E
  in VBT to use AUX A, and on these the DDI A strap is also
  set. There doesn't seem to be anything hooked up to either
  DDI however. But given the DDI A strap is wrong on these it
  might well be wrong on the Asrock too.

  Most other CI machines seem to have straps that generally
  match the VBT. fi-kbl-soraka is an exception though as DDI D
  strap is not set, but it is declared in VBT as a DP++ port.
  No idea if there's a real physical port to go with it or not.

- Some kind of quirk just for the cases where both DDI A and DDI E
  are present in VBT. Might be feasible given we've ignored
  DDI A in these cases up to now successfully. But feels rather
  unsatisfactory, and not very future proof against funny VBTs.

References: https://bugs.freedesktop.org/show_bug.cgi?id=111966
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417131728.7705-4-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 28 +++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index db7b6eaf8c854..1d28a2560ae01 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -45,6 +45,7 @@
 
 #include "g4x_dp.h"
 #include "i915_drv.h"
+#include "i915_irq.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
@@ -5377,6 +5378,15 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 		goto out_vdd_off;
 	}
 
+	/*
+	 * Enable HPD sense for live status check.
+	 * intel_hpd_irq_setup() will turn it off again
+	 * if it's no longer needed later.
+	 *
+	 * The DPCD probe below will make sure VDD is on.
+	 */
+	intel_hpd_enable_detection(encoder);
+
 	/* Cache DPCD and EDID for edp. */
 	has_dpcd = intel_edp_init_dpcd(intel_dp);
 
@@ -5388,6 +5398,24 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 		goto out_vdd_off;
 	}
 
+	/*
+	 * VBT and straps are liars. Also check HPD as that seems
+	 * to be the most reliable piece of information available.
+	 */
+	if (!intel_digital_port_connected(encoder)) {
+		/*
+		 * If this fails, presume the DPCD answer came
+		 * from some other port using the same AUX CH.
+		 *
+		 * FIXME maybe cleaner to check this before the
+		 * DPCD read? Would need sort out the VDD handling...
+		 */
+		drm_info(&dev_priv->drm,
+			 "[ENCODER:%d:%s] HPD is down, disabling eDP\n",
+			 encoder->base.base.id, encoder->base.name);
+		goto out_vdd_off;
+	}
+
 	mutex_lock(&dev_priv->drm.mode_config.mutex);
 	drm_edid = drm_edid_read_ddc(connector, &intel_dp->aux.ddc);
 	if (!drm_edid) {
-- 
GitLab


From a823c5a825cf9e49b2e20011cd87d93f949c2a27 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 17 Apr 2023 16:17:28 +0300
Subject: [PATCH 0090/1791] drm/i915: Reuse <platform>_hotplug_mask() in
 .hpd_detection_setup()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace the hardcoded masks with just a loop over all hpd
pins using the <platform>_hotplug_mask() functions.

v2: Deal with mtp

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com> #v1
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417131728.7705-5-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 65 ++++++++++++---------------------
 1 file changed, 24 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 3505fafcfb7b0..4345931c65965 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -83,6 +83,7 @@ static inline void pmu_irq_stats(struct drm_i915_private *i915,
 
 typedef bool (*long_pulse_detect_func)(enum hpd_pin pin, u32 val);
 typedef u32 (*hotplug_enables_func)(struct intel_encoder *encoder);
+typedef u32 (*hotplug_mask_func)(enum hpd_pin pin);
 
 static const u32 hpd_ilk[HPD_NUM_PINS] = {
 	[HPD_PORT_A] = DE_DP_A_HOTPLUG,
@@ -899,6 +900,18 @@ static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv,
 	return hotplug_irqs;
 }
 
+static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915,
+				  hotplug_mask_func hotplug_mask)
+{
+	enum hpd_pin pin;
+	u32 hotplug = 0;
+
+	for_each_hpd_pin(pin)
+		hotplug |= hotplug_mask(pin);
+
+	return hotplug;
+}
+
 static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915,
 				     hotplug_enables_func hotplug_enables)
 {
@@ -2981,10 +2994,7 @@ static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
 	 * The pulse duration bits are reserved on LPT+.
 	 */
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 ibx_hotplug_mask(HPD_PORT_A) |
-			 ibx_hotplug_mask(HPD_PORT_B) |
-			 ibx_hotplug_mask(HPD_PORT_C) |
-			 ibx_hotplug_mask(HPD_PORT_D),
+			 intel_hpd_hotplug_mask(dev_priv, ibx_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
 }
 
@@ -3050,10 +3060,7 @@ static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
 static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI,
-			 icp_ddi_hotplug_mask(HPD_PORT_A) |
-			 icp_ddi_hotplug_mask(HPD_PORT_B) |
-			 icp_ddi_hotplug_mask(HPD_PORT_C) |
-			 icp_ddi_hotplug_mask(HPD_PORT_D),
+			 intel_hpd_hotplug_mask(dev_priv, icp_ddi_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
 }
 
@@ -3069,12 +3076,7 @@ static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
 static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
-			 icp_tc_hotplug_mask(HPD_PORT_TC1) |
-			 icp_tc_hotplug_mask(HPD_PORT_TC2) |
-			 icp_tc_hotplug_mask(HPD_PORT_TC3) |
-			 icp_tc_hotplug_mask(HPD_PORT_TC4) |
-			 icp_tc_hotplug_mask(HPD_PORT_TC5) |
-			 icp_tc_hotplug_mask(HPD_PORT_TC6),
+			 intel_hpd_hotplug_mask(dev_priv, icp_tc_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
 }
 
@@ -3155,12 +3157,7 @@ static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
 static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL,
-			 gen11_hotplug_mask(HPD_PORT_TC1) |
-			 gen11_hotplug_mask(HPD_PORT_TC2) |
-			 gen11_hotplug_mask(HPD_PORT_TC3) |
-			 gen11_hotplug_mask(HPD_PORT_TC4) |
-			 gen11_hotplug_mask(HPD_PORT_TC5) |
-			 gen11_hotplug_mask(HPD_PORT_TC6),
+			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
@@ -3176,12 +3173,7 @@ static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder)
 static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 gen11_hotplug_mask(HPD_PORT_TC1) |
-			 gen11_hotplug_mask(HPD_PORT_TC2) |
-			 gen11_hotplug_mask(HPD_PORT_TC3) |
-			 gen11_hotplug_mask(HPD_PORT_TC4) |
-			 gen11_hotplug_mask(HPD_PORT_TC5) |
-			 gen11_hotplug_mask(HPD_PORT_TC6),
+			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
 }
 
@@ -3260,8 +3252,7 @@ static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
 static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     mtp_ddi_hotplug_mask(HPD_PORT_A) |
-		     mtp_ddi_hotplug_mask(HPD_PORT_B),
+		     intel_hpd_hotplug_mask(i915, mtp_ddi_hotplug_mask),
 		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
 }
 
@@ -3277,10 +3268,7 @@ static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
 static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
 {
 	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
-		     mtp_tc_hotplug_mask(HPD_PORT_TC1) |
-		     mtp_tc_hotplug_mask(HPD_PORT_TC2) |
-		     mtp_tc_hotplug_mask(HPD_PORT_TC3) |
-		     mtp_tc_hotplug_mask(HPD_PORT_TC4),
+		     intel_hpd_hotplug_mask(i915, mtp_tc_hotplug_mask),
 		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
 }
 
@@ -3441,14 +3429,11 @@ static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 
 	/* Enable digital hotplug on the PCH */
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 spt_hotplug_mask(HPD_PORT_A) |
-			 spt_hotplug_mask(HPD_PORT_B) |
-			 spt_hotplug_mask(HPD_PORT_C) |
-			 spt_hotplug_mask(HPD_PORT_D),
+			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables));
 
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2,
-			 spt_hotplug2_mask(HPD_PORT_E),
+			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug2_mask),
 			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
 }
 
@@ -3517,7 +3502,7 @@ static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
 	 * The pulse duration bits are reserved on HSW+.
 	 */
 	intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 ilk_hotplug_mask(HPD_PORT_A),
+			 intel_hpd_hotplug_mask(dev_priv, ilk_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
 }
 
@@ -3591,9 +3576,7 @@ static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
 static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
 {
 	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 bxt_hotplug_mask(HPD_PORT_A) |
-			 bxt_hotplug_mask(HPD_PORT_B) |
-			 bxt_hotplug_mask(HPD_PORT_C),
+			 intel_hpd_hotplug_mask(dev_priv, bxt_hotplug_mask),
 			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
 }
 
-- 
GitLab


From b8fe56375f78835db47565d91ea9d21767fe3c08 Mon Sep 17 00:00:00 2001
From: Leon Huang <Leon.Huang1@amd.com>
Date: Tue, 8 Nov 2022 16:29:13 +0800
Subject: [PATCH 0091/1791] drm/amd/display: Refactor ABM feature

[Why]
Refactor ABM feature and implement inbox command for DMUB.

[How]
Implement the ioctl to send inbox command to DMUB.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Leon Huang <Leon.Huang1@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/Makefile   |   2 +-
 drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c | 264 +++++++---------
 .../gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c | 286 ++++++++++++++++++
 .../gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h |  46 +++
 drivers/gpu/drm/amd/display/dc/inc/hw/abm.h   |   6 +
 5 files changed, 442 insertions(+), 162 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h

diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile
index 0d7db132a20fe..01490c9ba9587 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile
@@ -29,7 +29,7 @@
 DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \
 dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \
 dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \
-dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dce_panel_cntl.o \
+dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dmub_abm_lcd.o dce_panel_cntl.o \
 dmub_hw_lock_mgr.o dmub_outbox.o
 
 AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE))
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
index 9fc48208c2e42..a66f83a61402d 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
@@ -24,212 +24,151 @@
  */
 
 #include "dmub_abm.h"
-#include "dce_abm.h"
+#include "dmub_abm_lcd.h"
 #include "dc.h"
-#include "dc_dmub_srv.h"
-#include "dmub/dmub_srv.h"
 #include "core_types.h"
-#include "dm_services.h"
-#include "reg_helper.h"
-#include "fixed31_32.h"
-
-#include "atom.h"
 
 #define TO_DMUB_ABM(abm)\
 	container_of(abm, struct dce_abm, base)
 
-#define REG(reg) \
-	(dce_abm->regs->reg)
+#define ABM_FEATURE_NO_SUPPORT	0
+#define ABM_LCD_SUPPORT			1
 
-#undef FN
-#define FN(reg_name, field_name) \
-	dce_abm->abm_shift->field_name, dce_abm->abm_mask->field_name
+static unsigned int abm_feature_support(struct abm *abm, unsigned int panel_inst)
+{
+	struct dc_context *dc = abm->ctx;
+	struct dc_link *edp_links[MAX_NUM_EDP];
+	int i;
+	int edp_num;
+	unsigned int ret = ABM_FEATURE_NO_SUPPORT;
 
-#define CTX \
-	dce_abm->base.ctx
+	dc_get_edp_links(dc->dc, edp_links, &edp_num);
 
-#define DISABLE_ABM_IMMEDIATELY 255
+	for (i = 0; i < edp_num; i++) {
+		if (panel_inst == i)
+			break;
+	}
 
+	if (i < edp_num) {
+		ret = ABM_LCD_SUPPORT;
+	}
 
+	return ret;
+}
 
-static void dmub_abm_enable_fractional_pwm(struct dc_context *dc)
+static void dmub_abm_init_ex(struct abm *abm, uint32_t backlight)
 {
-	union dmub_rb_cmd cmd;
-	uint32_t fractional_pwm = (dc->dc->config.disable_fractional_pwm == false) ? 1 : 0;
-	uint32_t edp_id_count = dc->dc_edp_id_count;
-	int i;
-	uint8_t panel_mask = 0;
-
-	for (i = 0; i < edp_id_count; i++)
-		panel_mask |= 0x01 << i;
-
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.abm_set_pwm_frac.header.type = DMUB_CMD__ABM;
-	cmd.abm_set_pwm_frac.header.sub_type = DMUB_CMD__ABM_SET_PWM_FRAC;
-	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.fractional_pwm = fractional_pwm;
-	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
-	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask;
-	cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data);
-
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dmub_abm_init(abm, backlight);
 }
 
-static void dmub_abm_init(struct abm *abm, uint32_t backlight)
+static unsigned int dmub_abm_get_current_backlight_ex(struct abm *abm)
 {
-	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
-
-	REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x3);
-	REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x1);
-	REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x3);
-	REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x1);
-	REG_WRITE(BL1_PWM_BL_UPDATE_SAMPLE_RATE, 0x1);
-
-	REG_SET_3(DC_ABM1_HG_MISC_CTRL, 0,
-			ABM1_HG_NUM_OF_BINS_SEL, 0,
-			ABM1_HG_VMAX_SEL, 1,
-			ABM1_HG_BIN_BITWIDTH_SIZE_SEL, 0);
-
-	REG_SET_3(DC_ABM1_IPCSC_COEFF_SEL, 0,
-			ABM1_IPCSC_COEFF_SEL_R, 2,
-			ABM1_IPCSC_COEFF_SEL_G, 4,
-			ABM1_IPCSC_COEFF_SEL_B, 2);
+	return dmub_abm_get_current_backlight(abm);
+}
 
-	REG_UPDATE(BL1_PWM_CURRENT_ABM_LEVEL,
-			BL1_PWM_CURRENT_ABM_LEVEL, backlight);
+static unsigned int dmub_abm_get_target_backlight_ex(struct abm *abm)
+{
+	return dmub_abm_get_target_backlight(abm);
+}
 
-	REG_UPDATE(BL1_PWM_TARGET_ABM_LEVEL,
-			BL1_PWM_TARGET_ABM_LEVEL, backlight);
+static bool dmub_abm_set_level_ex(struct abm *abm, uint32_t level)
+{
+	bool ret = false;
+	unsigned int feature_support, i;
+	uint8_t panel_mask0 = 0;
 
-	REG_UPDATE(BL1_PWM_USER_LEVEL,
-			BL1_PWM_USER_LEVEL, backlight);
+	for (i = 0; i < MAX_NUM_EDP; i++) {
+		feature_support = abm_feature_support(abm, i);
 
-	REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES,
-			ABM1_LS_MIN_PIXEL_VALUE_THRES, 0,
-			ABM1_LS_MAX_PIXEL_VALUE_THRES, 1000);
+		if (feature_support == ABM_LCD_SUPPORT)
+			panel_mask0 |= (0x01 << i);
+	}
 
-	REG_SET_3(DC_ABM1_HGLS_REG_READ_PROGRESS, 0,
-			ABM1_HG_REG_READ_MISSED_FRAME_CLEAR, 1,
-			ABM1_LS_REG_READ_MISSED_FRAME_CLEAR, 1,
-			ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1);
+	if (panel_mask0)
+		ret = dmub_abm_set_level(abm, level, panel_mask0);
 
-	dmub_abm_enable_fractional_pwm(abm->ctx);
+	return ret;
 }
 
-static unsigned int dmub_abm_get_current_backlight(struct abm *abm)
+static bool dmub_abm_init_config_ex(struct abm *abm,
+	const char *src,
+	unsigned int bytes,
+	unsigned int inst)
 {
-	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
-	unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL);
+	unsigned int feature_support;
 
-	/* return backlight in hardware format which is unsigned 17 bits, with
-	 * 1 bit integer and 16 bit fractional
-	 */
-	return backlight;
-}
+	feature_support = abm_feature_support(abm, inst);
 
-static unsigned int dmub_abm_get_target_backlight(struct abm *abm)
-{
-	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
-	unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL);
+	if (feature_support == ABM_LCD_SUPPORT)
+		dmub_abm_init_config(abm, src, bytes, inst);
 
-	/* return backlight in hardware format which is unsigned 17 bits, with
-	 * 1 bit integer and 16 bit fractional
-	 */
-	return backlight;
+	return true;
 }
 
-static bool dmub_abm_set_level(struct abm *abm, uint32_t level)
+static bool dmub_abm_set_pause_ex(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst)
 {
-	union dmub_rb_cmd cmd;
-	struct dc_context *dc = abm->ctx;
-	struct dc_link *edp_links[MAX_NUM_EDP];
-	int i;
-	int edp_num;
-	uint8_t panel_mask = 0;
+	bool ret = false;
+	unsigned int feature_support;
 
-	dc_get_edp_links(dc->dc, edp_links, &edp_num);
-
-	for (i = 0; i < edp_num; i++) {
-		if (edp_links[i]->link_status.link_active)
-			panel_mask |= (0x01 << i);
-	}
-
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.abm_set_level.header.type = DMUB_CMD__ABM;
-	cmd.abm_set_level.header.sub_type = DMUB_CMD__ABM_SET_LEVEL;
-	cmd.abm_set_level.abm_set_level_data.level = level;
-	cmd.abm_set_level.abm_set_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
-	cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask;
-	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data);
+	feature_support = abm_feature_support(abm, panel_inst);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	if (feature_support == ABM_LCD_SUPPORT)
+		ret = dmub_abm_set_pause(abm, pause, panel_inst, stream_inst);
 
-	return true;
+	return ret;
 }
 
-static bool dmub_abm_init_config(struct abm *abm,
-	const char *src,
-	unsigned int bytes,
-	unsigned int inst)
+static bool dmub_abm_set_pipe_ex(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst)
 {
-	union dmub_rb_cmd cmd;
-	struct dc_context *dc = abm->ctx;
-	uint8_t panel_mask = 0x01 << inst;
+	bool ret = false;
+	unsigned int feature_support;
 
-	// TODO: Optimize by only reading back final 4 bytes
-	dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb);
+	feature_support = abm_feature_support(abm, panel_inst);
 
-	// Copy iramtable into cw7
-	memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)src, bytes);
+	if (feature_support == ABM_LCD_SUPPORT)
+		ret = dmub_abm_set_pipe(abm, otg_inst, option, panel_inst);
 
-	memset(&cmd, 0, sizeof(cmd));
-	// Fw will copy from cw7 to fw_state
-	cmd.abm_init_config.header.type = DMUB_CMD__ABM;
-	cmd.abm_init_config.header.sub_type = DMUB_CMD__ABM_INIT_CONFIG;
-	cmd.abm_init_config.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr;
-	cmd.abm_init_config.abm_init_config_data.bytes = bytes;
-	cmd.abm_init_config.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
-	cmd.abm_init_config.abm_init_config_data.panel_mask = panel_mask;
+	return ret;
+}
 
-	cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data);
+static bool dmub_abm_set_event_ex(struct abm *abm, unsigned int full_screen, unsigned int video_mode,
+		unsigned int hdr_mode, unsigned int panel_inst)
+{
+	bool ret = false;
+	unsigned int feature_support;
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	feature_support = abm_feature_support(abm, panel_inst);
 
-	return true;
+	return ret;
 }
 
-static bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst)
+static bool dmub_abm_set_backlight_level_pwm_ex(struct abm *abm,
+		unsigned int backlight_pwm_u16_16,
+		unsigned int frame_ramp,
+		unsigned int controller_id,
+		unsigned int panel_inst)
 {
-	union dmub_rb_cmd cmd;
-	struct dc_context *dc = abm->ctx;
-	uint8_t panel_mask = 0x01 << panel_inst;
+	bool ret = false;
+	unsigned int feature_support;
 
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.abm_pause.header.type = DMUB_CMD__ABM;
-	cmd.abm_pause.header.sub_type = DMUB_CMD__ABM_PAUSE;
-	cmd.abm_pause.abm_pause_data.enable = pause;
-	cmd.abm_pause.abm_pause_data.panel_mask = panel_mask;
-	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data);
+	feature_support = abm_feature_support(abm, panel_inst);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	if (feature_support == ABM_LCD_SUPPORT)
+		ret = dmub_abm_set_backlight_level(abm, backlight_pwm_u16_16, frame_ramp, panel_inst);
 
-	return true;
+	return ret;
 }
 
 static const struct abm_funcs abm_funcs = {
-	.abm_init = dmub_abm_init,
-	.set_abm_level = dmub_abm_set_level,
-	.get_current_backlight = dmub_abm_get_current_backlight,
-	.get_target_backlight = dmub_abm_get_target_backlight,
-	.init_abm_config = dmub_abm_init_config,
-	.set_abm_pause = dmub_abm_set_pause,
+	.abm_init = dmub_abm_init_ex,
+	.set_abm_level = dmub_abm_set_level_ex,
+	.get_current_backlight = dmub_abm_get_current_backlight_ex,
+	.get_target_backlight = dmub_abm_get_target_backlight_ex,
+	.init_abm_config = dmub_abm_init_config_ex,
+	.set_abm_pause = dmub_abm_set_pause_ex,
+	.set_pipe_ex = dmub_abm_set_pipe_ex,
+	.set_abm_event = dmub_abm_set_event_ex,
+	.set_backlight_level_pwm = dmub_abm_set_backlight_level_pwm_ex,
 };
 
 static void dmub_abm_construct(
@@ -256,16 +195,19 @@ struct abm *dmub_abm_create(
 	const struct dce_abm_shift *abm_shift,
 	const struct dce_abm_mask *abm_mask)
 {
-	struct dce_abm *abm_dce = kzalloc(sizeof(*abm_dce), GFP_KERNEL);
+	if (ctx->dc->caps.dmcub_support) {
+		struct dce_abm *abm_dce = kzalloc(sizeof(*abm_dce), GFP_KERNEL);
 
-	if (abm_dce == NULL) {
-		BREAK_TO_DEBUGGER();
-		return NULL;
-	}
+		if (abm_dce == NULL) {
+			BREAK_TO_DEBUGGER();
+			return NULL;
+		}
 
-	dmub_abm_construct(abm_dce, ctx, regs, abm_shift, abm_mask);
+		dmub_abm_construct(abm_dce, ctx, regs, abm_shift, abm_mask);
 
-	return &abm_dce->base;
+		return &abm_dce->base;
+	}
+	return NULL;
 }
 
 void dmub_abm_destroy(struct abm **abm)
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
new file mode 100644
index 0000000000000..4055d271ac57f
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
@@ -0,0 +1,286 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include "dmub_abm.h"
+#include "dce_abm.h"
+#include "dc.h"
+#include "dc_dmub_srv.h"
+#include "dmub/dmub_srv.h"
+#include "core_types.h"
+#include "dm_services.h"
+#include "reg_helper.h"
+#include "fixed31_32.h"
+
+#ifdef _WIN32
+#include "atombios.h"
+#else
+#include "atom.h"
+#endif
+
+#define TO_DMUB_ABM(abm)\
+	container_of(abm, struct dce_abm, base)
+
+#define REG(reg) \
+	(dce_abm->regs->reg)
+
+#undef FN
+#define FN(reg_name, field_name) \
+	dce_abm->abm_shift->field_name, dce_abm->abm_mask->field_name
+
+#define CTX \
+	dce_abm->base.ctx
+
+#define DISABLE_ABM_IMMEDIATELY 255
+
+
+
+static void dmub_abm_enable_fractional_pwm(struct dc_context *dc)
+{
+	union dmub_rb_cmd cmd;
+	uint32_t fractional_pwm = (dc->dc->config.disable_fractional_pwm == false) ? 1 : 0;
+	uint32_t edp_id_count = dc->dc_edp_id_count;
+	int i;
+	uint8_t panel_mask = 0;
+
+	for (i = 0; i < edp_id_count; i++)
+		panel_mask |= 0x01 << i;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_pwm_frac.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_pwm_frac.header.sub_type = DMUB_CMD__ABM_SET_PWM_FRAC;
+	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.fractional_pwm = fractional_pwm;
+	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask;
+	cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+}
+
+void dmub_abm_init(struct abm *abm, uint32_t backlight)
+{
+	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
+
+	REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x3);
+	REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x1);
+	REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x3);
+	REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x1);
+	REG_WRITE(BL1_PWM_BL_UPDATE_SAMPLE_RATE, 0x1);
+
+	REG_SET_3(DC_ABM1_HG_MISC_CTRL, 0,
+			ABM1_HG_NUM_OF_BINS_SEL, 0,
+			ABM1_HG_VMAX_SEL, 1,
+			ABM1_HG_BIN_BITWIDTH_SIZE_SEL, 0);
+
+	REG_SET_3(DC_ABM1_IPCSC_COEFF_SEL, 0,
+			ABM1_IPCSC_COEFF_SEL_R, 2,
+			ABM1_IPCSC_COEFF_SEL_G, 4,
+			ABM1_IPCSC_COEFF_SEL_B, 2);
+
+	REG_UPDATE(BL1_PWM_CURRENT_ABM_LEVEL,
+			BL1_PWM_CURRENT_ABM_LEVEL, backlight);
+
+	REG_UPDATE(BL1_PWM_TARGET_ABM_LEVEL,
+			BL1_PWM_TARGET_ABM_LEVEL, backlight);
+
+	REG_UPDATE(BL1_PWM_USER_LEVEL,
+			BL1_PWM_USER_LEVEL, backlight);
+
+	REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES,
+			ABM1_LS_MIN_PIXEL_VALUE_THRES, 0,
+			ABM1_LS_MAX_PIXEL_VALUE_THRES, 1000);
+
+	REG_SET_3(DC_ABM1_HGLS_REG_READ_PROGRESS, 0,
+			ABM1_HG_REG_READ_MISSED_FRAME_CLEAR, 1,
+			ABM1_LS_REG_READ_MISSED_FRAME_CLEAR, 1,
+			ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1);
+
+	dmub_abm_enable_fractional_pwm(abm->ctx);
+}
+
+unsigned int dmub_abm_get_current_backlight(struct abm *abm)
+{
+	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
+	unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL);
+
+	/* return backlight in hardware format which is unsigned 17 bits, with
+	 * 1 bit integer and 16 bit fractional
+	 */
+	return backlight;
+}
+
+unsigned int dmub_abm_get_target_backlight(struct abm *abm)
+{
+	struct dce_abm *dce_abm = TO_DMUB_ABM(abm);
+	unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL);
+
+	/* return backlight in hardware format which is unsigned 17 bits, with
+	 * 1 bit integer and 16 bit fractional
+	 */
+	return backlight;
+}
+
+bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_level.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_level.header.sub_type = DMUB_CMD__ABM_SET_LEVEL;
+	cmd.abm_set_level.abm_set_level_data.level = level;
+	cmd.abm_set_level.abm_set_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask;
+	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+	return true;
+}
+
+#ifndef TRIM_AMBIENT_GAMMA
+void dmub_abm_set_ambient_level(struct abm *abm, unsigned int ambient_lux, uint8_t panel_mask)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+
+	if (ambient_lux > 0xFFFF)
+		ambient_lux = 0xFFFF;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_ambient_level.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_ambient_level.header.sub_type = DMUB_CMD__ABM_SET_AMBIENT_LEVEL;
+	cmd.abm_set_ambient_level.abm_set_ambient_level_data.ambient_lux = ambient_lux;
+	cmd.abm_set_ambient_level.abm_set_ambient_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_set_ambient_level.abm_set_ambient_level_data.panel_mask = panel_mask;
+	cmd.abm_set_ambient_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_ambient_level_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+}
+#endif
+
+void dmub_abm_init_config(struct abm *abm,
+	const char *src,
+	unsigned int bytes,
+	unsigned int inst)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+	uint8_t panel_mask = 0x01 << inst;
+
+	// TODO: Optimize by only reading back final 4 bytes
+	dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb);
+
+	// Copy iramtable into cw7
+	memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)src, bytes);
+
+	memset(&cmd, 0, sizeof(cmd));
+	// Fw will copy from cw7 to fw_state
+	cmd.abm_init_config.header.type = DMUB_CMD__ABM;
+	cmd.abm_init_config.header.sub_type = DMUB_CMD__ABM_INIT_CONFIG;
+	cmd.abm_init_config.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr;
+	cmd.abm_init_config.abm_init_config_data.bytes = bytes;
+	cmd.abm_init_config.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_init_config.abm_init_config_data.panel_mask = panel_mask;
+
+	cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+}
+
+bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+	uint8_t panel_mask = 0x01 << panel_inst;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_pause.header.type = DMUB_CMD__ABM;
+	cmd.abm_pause.header.sub_type = DMUB_CMD__ABM_PAUSE;
+	cmd.abm_pause.abm_pause_data.enable = pause;
+	cmd.abm_pause.abm_pause_data.panel_mask = panel_mask;
+	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+	return true;
+}
+
+bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+	uint32_t ramping_boundary = 0xFFFF;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_pipe.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_pipe.header.sub_type = DMUB_CMD__ABM_SET_PIPE;
+	cmd.abm_set_pipe.abm_set_pipe_data.otg_inst = otg_inst;
+	cmd.abm_set_pipe.abm_set_pipe_data.set_pipe_option = option;
+	cmd.abm_set_pipe.abm_set_pipe_data.panel_inst = panel_inst;
+	cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary;
+	cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+	return true;
+}
+
+bool dmub_abm_set_backlight_level(struct abm *abm,
+		unsigned int backlight_pwm_u16_16,
+		unsigned int frame_ramp,
+		unsigned int panel_inst)
+{
+	union dmub_rb_cmd cmd;
+	struct dc_context *dc = abm->ctx;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_backlight.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT;
+	cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp;
+	cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16;
+	cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst);
+	cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+
+	return true;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h
new file mode 100644
index 0000000000000..00b4e268768ef
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#ifndef __DMUB_ABM_LCD_H__
+#define __DMUB_ABM_LCD_H__
+
+#include "abm.h"
+
+void dmub_abm_init(struct abm *abm, uint32_t backlight);
+bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask);
+unsigned int dmub_abm_get_current_backlight(struct abm *abm);
+unsigned int dmub_abm_get_target_backlight(struct abm *abm);
+void dmub_abm_init_config(struct abm *abm,
+	const char *src,
+	unsigned int bytes,
+	unsigned int inst);
+
+bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst);
+bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst);
+bool dmub_abm_set_backlight_level(struct abm *abm,
+		unsigned int backlight_pwm_u16_16,
+		unsigned int frame_ramp,
+		unsigned int panel_inst);
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
index ecb4191b6e64f..db5cf9acafe6d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
@@ -55,6 +55,12 @@ struct abm_funcs {
 			unsigned int bytes,
 			unsigned int inst);
 	bool (*set_abm_pause)(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int otg_inst);
+	bool (*set_pipe_ex)(struct abm *abm,
+			unsigned int otg_inst,
+			unsigned int option,
+			unsigned int panel_inst);
+	bool (*set_abm_event)(struct abm *abm, unsigned int full_screen, unsigned int video_mode,
+			unsigned int hdr_mode, unsigned int panel_inst);
 };
 
 #endif
-- 
GitLab


From e4dfd94d5e3851df607b26ab5b20ad8d94f5ccff Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 17 Apr 2023 22:30:01 +0530
Subject: [PATCH 0092/1791] drm/amd/display: Unconditionally print when DP sink
 power state fails

The previous 'commit ca9beb8aac68 ("drm/amd/display: Add logging when
setting DP sink power state fails")', it is better to unconditionally
print "failed to power up sink", because we are returning
DC_ERROR_UNEXPECTED.

Fixes: ca9beb8aac68 ("drm/amd/display: Add logging when setting DP sink power state fails")
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/protocols/link_dp_capability.c    | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index ba98013fecd00..6d2d10da2b77c 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1043,9 +1043,7 @@ static enum dc_status wake_up_aux_channel(struct dc_link *link)
 				DP_SET_POWER,
 				&dpcd_power_state,
 				sizeof(dpcd_power_state));
-		if (status < 0)
-			DC_LOG_DC("%s: Failed to power up sink: %s\n", __func__,
-				  dpcd_power_state == DP_SET_POWER_D0 ? "D0" : "D3");
+		DC_LOG_DC("%s: Failed to power up sink\n", __func__);
 		return DC_ERROR_UNEXPECTED;
 	}
 
-- 
GitLab


From 9b035d089086deb75d6664b26d36e35853e58ce9 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 15 Apr 2023 21:03:16 +0530
Subject: [PATCH 0093/1791] drm/amd/display: Check & log if receiver supports
 MST, DSC & FEC.

After reading from receiver via DPCD, check & log if it supports MST,
DSC & FEC

Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/link/protocols/link_dp_capability.c       | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 6d2d10da2b77c..2914fca7dab30 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1552,6 +1552,9 @@ static bool retrieve_link_cap(struct dc_link *link)
 	int i;
 	struct dp_sink_hw_fw_revision dp_hw_fw_revision;
 	const uint32_t post_oui_delay = 30; // 30ms
+	bool is_fec_supported = false;
+	bool is_dsc_basic_supported = false;
+	bool is_dsc_passthrough_supported = false;
 
 	memset(dpcd_data, '\0', sizeof(dpcd_data));
 	memset(&down_strm_port_count,
@@ -1694,6 +1697,7 @@ static bool retrieve_link_cap(struct dc_link *link)
 
 	/* TODO - decouple raw mst capability from policy decision */
 	link->dpcd_caps.is_mst_capable = read_is_mst_supported(link);
+	DC_LOG_DC("%s: MST_Support: %s\n", __func__, str_yes_no(link->dpcd_caps.is_mst_capable));
 
 	get_active_converter_info(ds_port.byte, link);
 
@@ -1801,6 +1805,18 @@ static bool retrieve_link_cap(struct dc_link *link)
 				DP_DSC_SUPPORT,
 				link->dpcd_caps.dsc_caps.dsc_basic_caps.raw,
 				sizeof(link->dpcd_caps.dsc_caps.dsc_basic_caps.raw));
+		if (status == DC_OK) {
+			is_fec_supported = link->dpcd_caps.fec_cap.bits.FEC_CAPABLE;
+			is_dsc_basic_supported = link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT;
+			is_dsc_passthrough_supported = link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_PASSTHROUGH_SUPPORT;
+			DC_LOG_DC("%s: FEC_Sink_Support: %s\n", __func__,
+				  str_yes_no(is_fec_supported));
+			DC_LOG_DC("%s: DSC_Basic_Sink_Support: %s\n", __func__,
+				  str_yes_no(is_dsc_basic_supported));
+			if (link->dpcd_caps.is_mst_capable)
+				DC_LOG_DC("%s: DSC_Passthrough_Sink_Support: %s\n", __func__,
+					  str_yes_no(is_dsc_passthrough_supported));
+		}
 		if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) {
 			status = core_link_read_dpcd(
 					link,
-- 
GitLab


From 6f0ef80a00adfd51be22b6ab84acd48de1d3938d Mon Sep 17 00:00:00 2001
From: Leon Huang <Leon.Huang1@amd.com>
Date: Thu, 8 Dec 2022 16:06:07 +0800
Subject: [PATCH 0094/1791] drm/amd/display: Fix ABM pipe/backlight issues when
 change backlight

[Why]
set ABM pipe/backlight gets some issues when abm callback func pointers
are NULL. For some usecase, driver would like to control PWM level before
ABM resource is ready. However, recent flow refactor of ABM didn't
consider that use case.

[How]
Rollback flow that sending inbox command to dmub directly when ABM
function pointers aren't ready.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Leon Huang <Leon.Huang1@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn21/dcn21_hwseq.c    | 64 ++++++++++++++-----
 1 file changed, 47 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
index 2a182c2f57d6f..1c6477d73c8e1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
@@ -159,6 +159,25 @@ static bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t optio
 	return true;
 }
 
+static void dmub_abm_set_backlight(struct dc_context *dc, uint32_t backlight_pwm_u16_16,
+									uint32_t frame_ramp, uint32_t panel_inst)
+{
+	union dmub_rb_cmd cmd;
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.abm_set_backlight.header.type = DMUB_CMD__ABM;
+	cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT;
+	cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp;
+	cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16;
+	cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
+	cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst);
+	cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data);
+
+	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
+	dc_dmub_srv_cmd_execute(dc->dmub_srv);
+	dc_dmub_srv_wait_idle(dc->dmub_srv);
+}
+
 void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx)
 {
 	struct abm *abm = pipe_ctx->stream_res.abm;
@@ -173,8 +192,12 @@ void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx)
 	}
 
 	if (abm && panel_cntl) {
-		dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE,
-				panel_cntl->inst);
+		if (abm->funcs && abm->funcs->set_pipe_ex) {
+			abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE,
+			panel_cntl->inst);
+		} else {
+			dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE, panel_cntl->inst);
+		}
 		panel_cntl->funcs->store_backlight_level(panel_cntl);
 	}
 }
@@ -191,8 +214,13 @@ void dcn21_set_pipe(struct pipe_ctx *pipe_ctx)
 		return;
 	}
 
-	if (abm && panel_cntl)
-		dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+	if (abm && panel_cntl) {
+		if (abm->funcs && abm->funcs->set_pipe_ex) {
+			abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+		} else {
+			dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+		}
+	}
 }
 
 bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx,
@@ -210,21 +238,23 @@ bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx,
 		return true;
 	}
 
-	if (abm && panel_cntl)
-		dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+	if (abm != NULL) {
+		uint32_t otg_inst = pipe_ctx->stream_res.tg->inst;
 
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.abm_set_backlight.header.type = DMUB_CMD__ABM;
-	cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT;
-	cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp;
-	cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16;
-	cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
-	cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_cntl->inst);
-	cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data);
+		if (abm && panel_cntl) {
+			if (abm->funcs && abm->funcs->set_pipe_ex) {
+				abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+			} else {
+				dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst);
+			}
+		}
+	}
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	if (abm && abm->funcs && abm->funcs->set_backlight_level_pwm)
+		abm->funcs->set_backlight_level_pwm(abm, backlight_pwm_u16_16,
+			frame_ramp, 0, panel_cntl->inst);
+	else
+		dmub_abm_set_backlight(dc, backlight_pwm_u16_16, frame_ramp, panel_cntl->inst);
 
 	return true;
 }
-- 
GitLab


From e101bf95ea87ccc03ac2f48dfc0757c6364ff3c7 Mon Sep 17 00:00:00 2001
From: Wesley Chalmers <Wesley.Chalmers@amd.com>
Date: Thu, 3 Nov 2022 22:29:31 -0400
Subject: [PATCH 0095/1791] drm/amd/display: Do not set drr on pipe commit

[WHY]
Writing to DRR registers such as OTG_V_TOTAL_MIN on the same frame as a
pipe commit can cause underflow.

[HOW]
Move DMUB p-state delegate into optimze_bandwidth; enabling FAMS sets
optimized_required.

This change expects that Freesync requests are blocked when
optimized_required is true.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c | 7 +++++++
 2 files changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 5403e9399a465..6ce10fd4bb1a4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2113,6 +2113,12 @@ void dcn20_optimize_bandwidth(
 	if (hubbub->funcs->program_compbuf_size)
 		hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true);
 
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+		dc_dmub_srv_p_state_delegate(dc,
+			true, context);
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+	}
+
 	dc->clk_mgr->funcs->update_clocks(
 			dc->clk_mgr,
 			context,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index 0e071fbc9154a..0411867654dde 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -985,11 +985,18 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 void dcn30_prepare_bandwidth(struct dc *dc,
  	struct dc_state *context)
 {
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+		dc->optimized_required = true;
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
+	}
+
 	if (dc->clk_mgr->dc_mode_softmax_enabled)
 		if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 &&
 				context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000)
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
+
+	dc_dmub_srv_p_state_delegate(dc, false, context);
 }
 
-- 
GitLab


From 751e17147953bc30036b8fe0eaaf780b6951404c Mon Sep 17 00:00:00 2001
From: Wesley Chalmers <Wesley.Chalmers@amd.com>
Date: Tue, 28 Feb 2023 13:48:00 -0500
Subject: [PATCH 0096/1791] drm/amd/display: Block optimize on consecutive FAMS
 enables

[WHY]
It is possible to commit state multiple times in rapid succession with
FAMS enabled; if each of these commits were to set optimized_required,
then the user may see latency.

[HOW]
fw_based_mclk_switching is currently not used in dc->clk_mgr; use it
to track whether the current state has FAMS enabled;
if it has, then do not disable FAMS in prepare_bandwidth, and do not set
optimized_required.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.c    |  3 +++
 .../drm/amd/display/dc/dcn30/dcn30_hwseq.c    | 22 ++++++++++++++++---
 2 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 6ce10fd4bb1a4..422fbf79da64f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2117,6 +2117,9 @@ void dcn20_optimize_bandwidth(
 		dc_dmub_srv_p_state_delegate(dc,
 			true, context);
 		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+		dc->clk_mgr->clks.fw_based_mclk_switching = true;
+	} else {
+		dc->clk_mgr->clks.fw_based_mclk_switching = false;
 	}
 
 	dc->clk_mgr->funcs->update_clocks(
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index 0411867654dde..8263a07f265f2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -983,9 +983,13 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 }
 
 void dcn30_prepare_bandwidth(struct dc *dc,
- 	struct dc_state *context)
+	struct dc_state *context)
 {
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+	bool p_state_change_support = context->bw_ctx.bw.dcn.clk.p_state_change_support;
+	/* Any transition into an FPO config should disable MCLK switching first to avoid
+	 * driver and FW P-State synchronization issues.
+	 */
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) {
 		dc->optimized_required = true;
 		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
 	}
@@ -996,7 +1000,19 @@ void dcn30_prepare_bandwidth(struct dc *dc,
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
+	/*
+	 * enabled -> enabled: do not disable
+	 * enabled -> disabled: disable
+	 * disabled -> enabled: don't care
+	 * disabled -> disabled: don't care
+	 */
+	if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching)
+		dc_dmub_srv_p_state_delegate(dc, false, context);
 
-	dc_dmub_srv_p_state_delegate(dc, false, context);
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) {
+		/* After disabling P-State, restore the original value to ensure we get the correct P-State
+		 * on the next optimize. */
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = p_state_change_support;
+	}
 }
 
-- 
GitLab


From e0a77e09c707cf89317de00f87b94b1168f27acd Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Date: Tue, 4 Apr 2023 14:54:05 -0600
Subject: [PATCH 0097/1791] drm/amd/display: Add missing WA and MCLK validation

When the commit fff7eb56b376 ("drm/amd/display: Don't set dram clock
change requirement for SubVP") was merged, we missed some parts
associated with the MCLK switch. This commit adds all the missing parts.

Fixes: fff7eb56b376 ("drm/amd/display: Don't set dram clock change requirement for SubVP")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c |  1 +
 .../drm/amd/display/dc/dcn32/dcn32_resource.c  |  2 +-
 .../drm/amd/display/dc/dml/dcn30/dcn30_fpu.c   | 18 +++++++++++++++++-
 3 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index db0974fe58aba..1f5ee5cde6e1c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -948,6 +948,7 @@ void dcn32_init_hw(struct dc *dc)
 	if (dc->ctx->dmub_srv) {
 		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
 		dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
+		dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 0beb11d95eb76..a876e6eb6cd88 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -2023,7 +2023,7 @@ int dcn32_populate_dml_pipes_from_context(
 	// In general cases we want to keep the dram clock change requirement
 	// (prefer configs that support MCLK switch). Only override to false
 	// for SubVP
-	if (subvp_in_use)
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || subvp_in_use)
 		context->bw_ctx.dml.soc.dram_clock_change_requirement_final = false;
 	else
 		context->bw_ctx.dml.soc.dram_clock_change_requirement_final = true;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
index 80972ee5e55b6..a352c703e2589 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
@@ -368,7 +368,9 @@ void dcn30_fpu_update_soc_for_wm_a(struct dc *dc, struct dc_state *context)
 	dc_assert_fp_enabled();
 
 	if (dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid) {
-		context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
+		if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching ||
+				context->bw_ctx.dml.soc.dram_clock_change_latency_us == 0)
+			context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us;
 		context->bw_ctx.dml.soc.sr_enter_plus_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us;
 		context->bw_ctx.dml.soc.sr_exit_time_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us;
 	}
@@ -563,6 +565,20 @@ void dcn30_fpu_calculate_wm_and_dlg(
 		pipe_idx++;
 	}
 
+	// WA: restrict FPO to use first non-strobe mode (NV24 BW issue)
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching &&
+			dc->dml.soc.num_chans <= 4 &&
+			context->bw_ctx.dml.vba.DRAMSpeed <= 1700 &&
+			context->bw_ctx.dml.vba.DRAMSpeed >= 1500) {
+
+		for (i = 0; i < dc->dml.soc.num_states; i++) {
+			if (dc->dml.soc.clock_limits[i].dram_speed_mts > 1700) {
+				context->bw_ctx.dml.vba.DRAMSpeed = dc->dml.soc.clock_limits[i].dram_speed_mts;
+				break;
+			}
+		}
+	}
+
 	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
 
 	if (!pstate_en)
-- 
GitLab


From d63e31f66892f67d8f1e279c57c4c0aee789fc66 Mon Sep 17 00:00:00 2001
From: Josip Pavic <Josip.Pavic@amd.com>
Date: Fri, 14 May 2021 14:04:02 -0400
Subject: [PATCH 0098/1791] drm/amd/display: copy dmub caps to dc on dcn31

[Why & How]
Add code path to copy dmub caps to dc, which is missing on dcn31

Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Josip Pavic <Josip.Pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index 7ac6e69cff378..62ce36c75c4d2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -295,6 +295,10 @@ void dcn31_init_hw(struct dc *dc)
 	if (dc->res_pool->hubbub->funcs->init_crb)
 		dc->res_pool->hubbub->funcs->init_crb(dc->res_pool->hubbub);
 #endif
+
+	// Get DMCUB capabilities
+	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
+	dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
 }
 
 void dcn31_dsc_pg_control(
-- 
GitLab


From 4ad3ee5ccc77aa3f9d702f7b9ad4d9cfeca6c443 Mon Sep 17 00:00:00 2001
From: Alex Hung <alex.hung@amd.com>
Date: Wed, 15 Mar 2023 19:09:15 +0800
Subject: [PATCH 0099/1791] drm/amd/display: allow edp updates for virtual
 signal

[Why]
When IGT's kms_hdmi_inject forces EDID for HDMI audio, dc rejects the
request because virtual signal is not in dc_is_audio_capable_signal
function.

[How]
Includes SIGNAL_TYPE_VIRTUAL as audio capable.

Reviewed-by: Chao-kai Wang <Stylon.Wang@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Wenchieh Chien <wenchieh.chien@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/include/signal_types.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/include/signal_types.h b/drivers/gpu/drm/amd/display/include/signal_types.h
index beed70179bb5d..23a308c3eccbf 100644
--- a/drivers/gpu/drm/amd/display/include/signal_types.h
+++ b/drivers/gpu/drm/amd/display/include/signal_types.h
@@ -104,6 +104,7 @@ static inline bool dc_is_audio_capable_signal(enum signal_type signal)
 {
 	return (signal == SIGNAL_TYPE_DISPLAY_PORT ||
 		signal == SIGNAL_TYPE_DISPLAY_PORT_MST ||
+		signal == SIGNAL_TYPE_VIRTUAL ||
 		dc_is_hdmi_signal(signal));
 }
 
-- 
GitLab


From 785b250e33c7b1a9dcdb262eac691cd33ac7a53a Mon Sep 17 00:00:00 2001
From: Alan Liu <HaoPing.Liu@amd.com>
Date: Wed, 22 Mar 2023 13:43:28 +0800
Subject: [PATCH 0100/1791] drm/amd/display: Fix in disabling secure display

[Why]
Currently we don't check if secure display is enabled before we send
command to disable secure display in dmub. It will accidentally cause
some other igt tests to fail, eg, crtc-linear-degamma.

[How]
Refactor the code we reset the secure display state to check secure
display was enabled or not before stopping it in dmub.

Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alan Liu <HaoPing.Liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c | 31 ++++++++-----------
 1 file changed, 13 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
index 27711743c22ce..0802f8e8fac5f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
@@ -83,12 +83,15 @@ const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc,
 }
 
 #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY
-static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc)
+static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc, struct dc_stream_state *stream)
 {
 	struct drm_device *drm_dev = crtc->dev;
+	struct amdgpu_display_manager *dm = &drm_to_adev(drm_dev)->dm;
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
+	bool was_activated;
 
 	spin_lock_irq(&drm_dev->event_lock);
+	was_activated = acrtc->dm_irq_params.window_param.activated;
 	acrtc->dm_irq_params.window_param.x_start = 0;
 	acrtc->dm_irq_params.window_param.y_start = 0;
 	acrtc->dm_irq_params.window_param.x_end = 0;
@@ -97,6 +100,14 @@ static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc)
 	acrtc->dm_irq_params.window_param.update_win = false;
 	acrtc->dm_irq_params.window_param.skip_frame_cnt = 0;
 	spin_unlock_irq(&drm_dev->event_lock);
+
+	/* Disable secure_display if it was enabled */
+	if (was_activated) {
+		/* stop ROI update on this crtc */
+		flush_work(&dm->secure_display_ctxs[crtc->index].notify_ta_work);
+		flush_work(&dm->secure_display_ctxs[crtc->index].forward_roi_work);
+		dc_stream_forward_crc_window(stream, NULL, true);
+	}
 }
 
 static void amdgpu_dm_crtc_notify_ta_to_read(struct work_struct *work)
@@ -204,9 +215,6 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc,
 					struct dm_crtc_state *dm_crtc_state,
 					enum amdgpu_dm_pipe_crc_source source)
 {
-#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
-	int i;
-#endif
 	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct dc_stream_state *stream_state = dm_crtc_state->stream;
 	bool enable = amdgpu_dm_is_valid_crc_source(source);
@@ -220,19 +228,6 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc,
 
 	/* Enable or disable CRTC CRC generation */
 	if (dm_is_crc_source_crtc(source) || source == AMDGPU_DM_PIPE_CRC_SOURCE_NONE) {
-#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
-		/* Disable secure_display if it was enabled */
-		if (!enable) {
-			for (i = 0; i < adev->mode_info.num_crtc; i++) {
-				if (adev->dm.secure_display_ctxs[i].crtc == crtc) {
-					/* stop ROI update on this crtc */
-					flush_work(&adev->dm.secure_display_ctxs[i].notify_ta_work);
-					flush_work(&adev->dm.secure_display_ctxs[i].forward_roi_work);
-					dc_stream_forward_crc_window(stream_state, NULL, true);
-				}
-			}
-		}
-#endif
 		if (!dc_stream_configure_crc(stream_state->ctx->dc,
 					     stream_state, NULL, enable, enable)) {
 			ret = -EINVAL;
@@ -363,7 +358,7 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name)
 
 #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
 	/* Reset secure_display when we change crc source from debugfs */
-	amdgpu_dm_set_crc_window_default(crtc);
+	amdgpu_dm_set_crc_window_default(crtc, crtc_state->stream);
 #endif
 
 	if (amdgpu_dm_crtc_configure_crc_source(crtc, crtc_state, source)) {
-- 
GitLab


From 72529b683c6d94fa87f3a42efd9b92ccfb8def41 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Fri, 24 Mar 2023 10:42:37 -0400
Subject: [PATCH 0101/1791] drm/amd/display: Fix hang when skipping modeset

[Why&How]

When skipping full modeset since the only state change was a front porch
change, the DC commit sequence requires extra checks to handle non
existant plane states being asked to be removed from context.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++++-
 drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 3 +++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 6cacb76f389e1..c432436cd66d3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7972,6 +7972,8 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			continue;
 
 		dc_plane = dm_new_plane_state->dc_state;
+		if (!dc_plane)
+			continue;
 
 		bundle->surface_updates[planes_count].surface = dc_plane;
 		if (new_pcrtc_state->color_mgmt_changed) {
@@ -9620,8 +9622,9 @@ static int dm_update_plane_state(struct dc *dc,
 			return -EINVAL;
 		}
 
+		if (dm_old_plane_state->dc_state)
+			dc_plane_state_release(dm_old_plane_state->dc_state);
 
-		dc_plane_state_release(dm_old_plane_state->dc_state);
 		dm_new_plane_state->dc_state = NULL;
 
 		*lock_and_validation_needed = true;
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 85d54bfb595ce..117d80cb36fbb 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -1707,6 +1707,9 @@ bool dc_remove_plane_from_context(
 	struct dc_stream_status *stream_status = NULL;
 	struct resource_pool *pool = dc->res_pool;
 
+	if (!plane_state)
+		return true;
+
 	for (i = 0; i < context->stream_count; i++)
 		if (context->streams[i] == stream) {
 			stream_status = &context->stream_status[i];
-- 
GitLab


From c0162a05bd31129e29a23d4bd1d2321c9663d5bc Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Mon, 27 Mar 2023 09:10:48 -0400
Subject: [PATCH 0102/1791] drm/amd/display: fix memleak in
 aconnector->timing_requested

[Why]
when amdgpu_dm_update_connector_after_detect is called
two times successively with valid sink, memory allocated of
aconnector->timing_requested for the first call is not free.
this causes memeleak.

[How]
allocate memory only when aconnector->timing_requested
is null.

Reviewed-by: Qingqing Zhuo <Qingqing.Zhuo@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c432436cd66d3..8b03c8d8f0b8e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3128,9 +3128,12 @@ void amdgpu_dm_update_connector_after_detect(
 						    aconnector->edid);
 		}
 
-		aconnector->timing_requested = kzalloc(sizeof(struct dc_crtc_timing), GFP_KERNEL);
-		if (!aconnector->timing_requested)
-			dm_error("%s: failed to create aconnector->requested_timing\n", __func__);
+		if (!aconnector->timing_requested) {
+			aconnector->timing_requested =
+				kzalloc(sizeof(struct dc_crtc_timing), GFP_KERNEL);
+			if (!aconnector->timing_requested)
+				dm_error("failed to create aconnector->requested_timing\n");
+		}
 
 		drm_connector_update_edid_property(connector, aconnector->edid);
 		amdgpu_dm_update_freesync_caps(connector, aconnector->edid);
-- 
GitLab


From 3cfd03b79425c8c9e10d15434f0b017249372609 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Date: Mon, 27 Mar 2023 15:33:54 -0400
Subject: [PATCH 0103/1791] drm/amd/display: update max streams per surface

Increse to 6 as that is the max surfaces supported asics can have.
The is no practical use case yet, but this is valuable for pre-si
validation.

Reviewed-by: Ariel Bernstein <Eric.Bernstein@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_types.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 45ab48fe5d004..34c848311455a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -83,7 +83,7 @@ struct dc_perf_trace {
 	unsigned long last_entry_write;
 };
 
-#define MAX_SURFACE_NUM 4
+#define MAX_SURFACE_NUM 6
 #define NUM_PIXEL_FORMATS 10
 
 enum tiling_mode {
-- 
GitLab


From d11dfbecc3feed5916bbe6c10942d9db61a9e2de Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Mon, 27 Mar 2023 18:40:34 -0400
Subject: [PATCH 0104/1791] drm/amd/display: Only consider DISPCLK when using
 optimized boot path

[Description]
- Previous bug fix for audio issue included dtbclk and p-state
  on the optimized boot path which is incorarect
- We only care about DISPCLK in the optimized vs. non-optimized
  boot path to avoid audio issues

Reviewed-by: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 1f5ee5cde6e1c..26791e3d162f4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -721,6 +721,9 @@ static void dcn32_initialize_min_clocks(struct dc *dc)
 	clocks->socclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].socclk_mhz * 1000;
 	clocks->dramclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 1000;
 	clocks->dppclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dppclk_mhz * 1000;
+	clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000;
+	clocks->fclk_p_state_change_support = true;
+	clocks->p_state_change_support = true;
 	if (dc->debug.disable_boot_optimizations) {
 		clocks->dispclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dispclk_mhz * 1000;
 	} else {
@@ -730,9 +733,6 @@ static void dcn32_initialize_min_clocks(struct dc *dc)
 		 * freq to ensure that the timing is valid and unchanged.
 		 */
 		clocks->dispclk_khz = dc->clk_mgr->funcs->get_dispclk_from_dentist(dc->clk_mgr);
-		clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000;
-		clocks->fclk_p_state_change_support = true;
-		clocks->p_state_change_support = true;
 	}
 
 	dc->clk_mgr->funcs->update_clocks(
-- 
GitLab


From c4edb01374685a3ea195c7d31459448cabe5a34e Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Tue, 28 Mar 2023 09:53:33 -0400
Subject: [PATCH 0105/1791] drm/amd/display: Reduce SubVP + DRR stretch margin

[Description]
- Having excessively large margin causes failure in the static
  schedulability check in some cases for SubVP + DRR
- 100us of DRR margin is sufficient based on a weeks worth of
  stress testing on different display configs

Reviewed-by: Michael Strauss <Michael.Strauss@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_stream.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 181a3408cc61a..25284006019c3 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -144,7 +144,7 @@ struct test_pattern {
 	unsigned int cust_pattern_size;
 };
 
-#define SUBVP_DRR_MARGIN_US 600 // 600us for DRR margin (SubVP + DRR)
+#define SUBVP_DRR_MARGIN_US 100 // 100us for DRR margin (SubVP + DRR)
 
 enum mall_stream_type {
 	SUBVP_NONE, // subvp not in use
-- 
GitLab


From e97cc04fe0fb33e489583dff79f6b1d6919fcc66 Mon Sep 17 00:00:00 2001
From: Josip Pavic <Josip.Pavic@amd.com>
Date: Wed, 15 Feb 2023 15:47:59 -0500
Subject: [PATCH 0106/1791] drm/amd/display: refactor dmub commands into single
 function

[Why & How]
Consolidate dmub access to a single interface. This makes it easier to
add code in the future that needs to run every time a dmub command is
requested (e.g. instrumentation, locking etc).

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Josip Pavic <Josip.Pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  12 +-
 .../drm/amd/display/dc/bios/command_table2.c  |  25 +--
 .../display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c  |   4 +-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        |   4 +-
 .../dc/clk_mgr/dcn315/dcn315_clk_mgr.c        |   4 +-
 .../dc/clk_mgr/dcn316/dcn316_clk_mgr.c        |   4 +-
 drivers/gpu/drm/amd/display/dc/core/dc.c      |  23 +--
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c  | 164 +++++++-----------
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h  |  13 +-
 drivers/gpu/drm/amd/display/dc/dc_helper.c    |  28 +--
 .../gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c |  28 +--
 .../drm/amd/display/dc/dce/dmub_hw_lock_mgr.c |   4 +-
 .../gpu/drm/amd/display/dc/dce/dmub_outbox.c  |   4 +-
 drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c |  28 +--
 .../gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c |   7 +-
 .../drm/amd/display/dc/dcn21/dcn21_hwseq.c    |   8 +-
 .../drm/amd/display/dc/dcn30/dcn30_hwseq.c    |  16 +-
 .../display/dc/dcn31/dcn31_dio_link_encoder.c |   8 +-
 .../drm/amd/display/dc/dcn31/dcn31_hwseq.c    |  10 +-
 .../amd/display/dc/dcn31/dcn31_panel_cntl.c   |   4 +-
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  |   4 +-
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    |  12 +-
 drivers/gpu/drm/amd/display/dc/dm_services.h  |   7 +
 .../drm/amd/display/dc/dm_services_types.h    |   6 +
 .../dc/link/protocols/link_dp_capability.c    |   2 +-
 .../display/dc/link/protocols/link_dp_dpia.c  |   2 +-
 26 files changed, 148 insertions(+), 283 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 8b03c8d8f0b8e..af3efb2456103 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -10311,7 +10311,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm,
 	input->cea_total_length = total_length;
 	memcpy(input->payload, data, length);
 
-	res = dc_dmub_srv_cmd_with_reply_data(dm->dc->ctx->dmub_srv, &cmd);
+	res = dm_execute_dmub_cmd(dm->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY);
 	if (!res) {
 		DRM_ERROR("EDID CEA parser failed\n");
 		return false;
@@ -10761,3 +10761,13 @@ bool check_seamless_boot_capability(struct amdgpu_device *adev)
 
 	return false;
 }
+
+bool dm_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type)
+{
+	return dc_dmub_srv_cmd_run(ctx->dmub_srv, cmd, wait_type);
+}
+
+bool dm_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type)
+{
+	return dc_dmub_srv_cmd_run_list(ctx->dmub_srv, count, cmd, wait_type);
+}
diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
index 1ef9e4053bb71..90a02d7bd3da3 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c
@@ -123,9 +123,7 @@ static void encoder_control_dmcub(
 		sizeof(cmd.digx_encoder_control.header);
 	cmd.digx_encoder_control.encoder_control.dig.stream_param = *dig;
 
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result encoder_control_digx_v1_5(
@@ -261,9 +259,7 @@ static void transmitter_control_dmcub(
 		sizeof(cmd.dig1_transmitter_control.header);
 	cmd.dig1_transmitter_control.transmitter_control.dig = *dig;
 
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result transmitter_control_v1_6(
@@ -325,9 +321,7 @@ static void transmitter_control_dmcub_v1_7(
 		sizeof(cmd.dig1_transmitter_control.header);
 	cmd.dig1_transmitter_control.transmitter_control.dig_v1_7 = *dig;
 
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result transmitter_control_v1_7(
@@ -435,9 +429,7 @@ static void set_pixel_clock_dmcub(
 		sizeof(cmd.set_pixel_clock.header);
 	cmd.set_pixel_clock.pixel_clock.clk = *clk;
 
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result set_pixel_clock_v7(
@@ -804,9 +796,7 @@ static void enable_disp_power_gating_dmcub(
 		sizeof(cmd.enable_disp_power_gating.header);
 	cmd.enable_disp_power_gating.power_gating.pwr = *pwr;
 
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result enable_disp_power_gating_v2_1(
@@ -1016,10 +1006,7 @@ static void enable_lvtma_control_dmcub(
 			panel_instance;
 	cmd.lvtma_control.data.bypass_panel_control_wait =
 			bypass_panel_control_wait;
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	dc_dmub_srv_cmd_execute(dmcub);
-	dc_dmub_srv_wait_idle(dmcub);
-
+	dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static enum bp_result enable_lvtma_control(
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
index f9e2e0c3095e7..3c743cd3d3f05 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
@@ -250,9 +250,7 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base,
 	cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz;
 	cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 5cb44f838bde5..4d5cd59f6433a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -286,9 +286,7 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 	cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz;
 	cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
index b737cbc468f55..300c6a5872d0b 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
@@ -234,9 +234,7 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base,
 	cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz;
 	cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static void dcn315_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
index 93db4dbee713e..538126cefd4d4 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
@@ -254,9 +254,7 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base,
 	cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz;
 	cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static void dcn316_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 52564b93f7eb2..61eec5aa40671 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -515,8 +515,7 @@ dc_stream_forward_dmub_crc_window(struct dc_dmub_srv *dmub_srv,
 		cmd.secure_display.roi_info.y_end = rect->y + rect->height;
 	}
 
-	dc_dmub_srv_cmd_queue(dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
+	dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 }
 
 static inline void
@@ -3309,7 +3308,6 @@ void dc_dmub_update_dirty_rect(struct dc *dc,
 			       struct dc_state *context)
 {
 	union dmub_rb_cmd cmd;
-	struct dc_context *dc_ctx = dc->ctx;
 	struct dmub_cmd_update_dirty_rect_data *update_dirty_rect;
 	unsigned int i, j;
 	unsigned int panel_inst = 0;
@@ -3350,8 +3348,7 @@ void dc_dmub_update_dirty_rect(struct dc *dc,
 
 			update_dirty_rect->panel_inst = panel_inst;
 			update_dirty_rect->pipe_idx = j;
-			dc_dmub_srv_cmd_queue(dc_ctx->dmub_srv, &cmd);
-			dc_dmub_srv_cmd_execute(dc_ctx->dmub_srv);
+			dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 		}
 	}
 }
@@ -4606,7 +4603,6 @@ bool dc_process_dmub_aux_transfer_async(struct dc *dc,
 {
 	uint8_t action;
 	union dmub_rb_cmd cmd = {0};
-	struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv;
 
 	ASSERT(payload->length <= 16);
 
@@ -4654,9 +4650,7 @@ bool dc_process_dmub_aux_transfer_async(struct dc *dc,
 			);
 	}
 
-	dc_dmub_srv_cmd_queue(dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
-	dc_dmub_srv_wait_idle(dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -4700,7 +4694,6 @@ bool dc_process_dmub_set_config_async(struct dc *dc,
 				struct dmub_notification *notify)
 {
 	union dmub_rb_cmd cmd = {0};
-	struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv;
 	bool is_cmd_complete = true;
 
 	/* prepare SET_CONFIG command */
@@ -4711,7 +4704,7 @@ bool dc_process_dmub_set_config_async(struct dc *dc,
 	cmd.set_config_access.set_config_control.cmd_pkt.msg_type = payload->msg_type;
 	cmd.set_config_access.set_config_control.cmd_pkt.msg_data = payload->msg_data;
 
-	if (!dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd)) {
+	if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) {
 		/* command is not processed by dmub */
 		notify->sc_status = SET_CONFIG_UNKNOWN_ERROR;
 		return is_cmd_complete;
@@ -4746,7 +4739,6 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc,
 				uint8_t *mst_slots_in_use)
 {
 	union dmub_rb_cmd cmd = {0};
-	struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv;
 
 	/* prepare MST_ALLOC_SLOTS command */
 	cmd.set_mst_alloc_slots.header.type = DMUB_CMD__DPIA;
@@ -4755,7 +4747,7 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc,
 	cmd.set_mst_alloc_slots.mst_slots_control.instance = dc->links[link_index]->ddc_hw_inst;
 	cmd.set_mst_alloc_slots.mst_slots_control.mst_alloc_slots = mst_alloc_slots;
 
-	if (!dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd))
+	if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
 		/* command is not processed by dmub */
 		return DC_ERROR_UNEXPECTED;
 
@@ -4789,14 +4781,11 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc,
 				uint32_t hpd_int_enable)
 {
 	union dmub_rb_cmd cmd = {0};
-	struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv;
 
 	cmd.dpia_hpd_int_enable.header.type = DMUB_CMD__DPIA_HPD_INT_ENABLE;
 	cmd.dpia_hpd_int_enable.enable = hpd_int_enable;
 
-	dc_dmub_srv_cmd_queue(dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
-	dc_dmub_srv_wait_idle(dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	DC_LOG_DEBUG("%s: hpd_int_enable(%d)\n", __func__, hpd_int_enable);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index a9b9490a532c2..954cbfdbc3b6a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -65,47 +65,6 @@ void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv)
 	}
 }
 
-void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv,
-			   union dmub_rb_cmd *cmd)
-{
-	struct dmub_srv *dmub = dc_dmub_srv->dmub;
-	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
-	enum dmub_status status;
-
-	status = dmub_srv_cmd_queue(dmub, cmd);
-	if (status == DMUB_STATUS_OK)
-		return;
-
-	if (status != DMUB_STATUS_QUEUE_FULL)
-		goto error;
-
-	/* Execute and wait for queue to become empty again. */
-	dc_dmub_srv_cmd_execute(dc_dmub_srv);
-	dc_dmub_srv_wait_idle(dc_dmub_srv);
-
-	/* Requeue the command. */
-	status = dmub_srv_cmd_queue(dmub, cmd);
-	if (status == DMUB_STATUS_OK)
-		return;
-
-error:
-	DC_ERROR("Error queuing DMUB command: status=%d\n", status);
-	dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-}
-
-void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv)
-{
-	struct dmub_srv *dmub = dc_dmub_srv->dmub;
-	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
-	enum dmub_status status;
-
-	status = dmub_srv_cmd_execute(dmub);
-	if (status != DMUB_STATUS_OK) {
-		DC_ERROR("Error starting DMUB execution: status=%d\n", status);
-		dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-	}
-}
-
 void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv)
 {
 	struct dmub_srv *dmub = dc_dmub_srv->dmub;
@@ -159,22 +118,55 @@ void dc_dmub_srv_send_inbox0_cmd(struct dc_dmub_srv *dmub_srv,
 	}
 }
 
-bool dc_dmub_srv_cmd_with_reply_data(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd)
+bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type)
 {
+	return dc_dmub_srv_cmd_run_list(dc_dmub_srv, 1, cmd, wait_type);
+}
+
+bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type)
+{
+	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
 	struct dmub_srv *dmub;
 	enum dmub_status status;
+	int i;
 
 	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
 		return false;
 
 	dmub = dc_dmub_srv->dmub;
 
-	status = dmub_srv_cmd_with_reply_data(dmub, cmd);
+	for (i = 0 ; i < count; i++) {
+		// Queue command
+		status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+
+		if (status != DMUB_STATUS_OK) {
+			DC_ERROR("Error queueing DMUB command: status=%d\n", status);
+			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
+			return false;
+		}
+	}
+
+	status = dmub_srv_cmd_execute(dmub);
 	if (status != DMUB_STATUS_OK) {
-		DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
+		DC_ERROR("Error starting DMUB execution: status=%d\n", status);
+		dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
 		return false;
 	}
 
+	// Wait for DMUB to process command
+	if (wait_type != DM_DMUB_WAIT_TYPE_NO_WAIT) {
+		status = dmub_srv_wait_for_idle(dmub, 100000);
+
+		if (status != DMUB_STATUS_OK) {
+			DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
+			return false;
+		}
+
+		// Copy data back from ring buffer into command
+		if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)
+			dmub_rb_get_return_data(&dmub->inbox1_rb, cmd_list);
+	}
+
 	return true;
 }
 
@@ -267,9 +259,7 @@ void dc_dmub_srv_drr_update_cmd(struct dc *dc, uint32_t tg_inst, uint32_t vtotal
 	cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header);
 
 	// Send the command to the DMCUB.
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst)
@@ -283,9 +273,7 @@ void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst)
 	cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header);
 
 	// Send the command to the DMCUB.
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 static uint8_t dc_dmub_srv_get_pipes_for_stream(struct dc *dc, struct dc_stream_state *stream)
@@ -378,21 +366,14 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru
 		sizeof(cmd.fw_assisted_mclk_switch) - sizeof(cmd.fw_assisted_mclk_switch.header);
 
 	// Send the command to the DMCUB.
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
 
-void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub)
+void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv)
 {
 	union dmub_rb_cmd cmd = { 0 };
-	enum dmub_status status;
-
-	if (!dmub) {
-		return;
-	}
 
 	memset(&cmd, 0, sizeof(cmd));
 
@@ -402,15 +383,10 @@ void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub)
 	cmd.query_feature_caps.header.ret_status = 1;
 	cmd.query_feature_caps.header.payload_bytes = sizeof(struct dmub_cmd_query_feature_caps_data);
 
-	/* Send command to fw */
-	status = dmub_srv_cmd_with_reply_data(dmub, &cmd);
-
-	ASSERT(status == DMUB_STATUS_OK);
-
 	/* If command was processed, copy feature caps to dmub srv */
-	if (status == DMUB_STATUS_OK &&
+	if (dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) &&
 	    cmd.query_feature_caps.header.ret_status == 0) {
-		memcpy(&dmub->feature_caps,
+		memcpy(&dc_dmub_srv->dmub->feature_caps,
 		       &cmd.query_feature_caps.query_feature_caps_data,
 		       sizeof(struct dmub_feature_caps));
 	}
@@ -419,7 +395,6 @@ void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub)
 void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pipe_ctx)
 {
 	union dmub_rb_cmd cmd = { 0 };
-	enum dmub_status status;
 	unsigned int panel_inst = 0;
 
 	dc_get_edp_link_panel_inst(dc, pipe_ctx->stream->link, &panel_inst);
@@ -433,13 +408,8 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi
 	cmd.visual_confirm_color.header.payload_bytes = sizeof(struct dmub_cmd_visual_confirm_color_data);
 	cmd.visual_confirm_color.visual_confirm_color_data.visual_confirm_color.panel_inst = panel_inst;
 
-	// Send command to fw
-	status = dmub_srv_cmd_with_reply_data(dc->ctx->dmub_srv->dmub, &cmd);
-
-	ASSERT(status == DMUB_STATUS_OK);
-
 	// If command was processed, copy feature caps to dmub srv
-	if (status == DMUB_STATUS_OK &&
+	if (dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) &&
 		cmd.visual_confirm_color.header.ret_status == 0) {
 		memcpy(&dc->ctx->dmub_srv->dmub->visual_confirm_color,
 			&cmd.visual_confirm_color.visual_confirm_color_data,
@@ -797,9 +767,8 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc,
 
 		cmd.fw_assisted_mclk_switch_v2.config_data.watermark_a_cache = wm_val_refclk < 0xFFFF ? wm_val_refclk : 0xFFFF;
 	}
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 bool dc_dmub_srv_get_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv, struct dmub_diagnostic_data *diag_data)
@@ -982,14 +951,6 @@ static void dc_build_cursor_update_payload0(
 	payload->panel_inst  = panel_inst;
 }
 
-static void dc_send_cmd_to_dmu(struct dc_dmub_srv *dmub_srv,
-		union dmub_rb_cmd *cmd)
-{
-	dc_dmub_srv_cmd_queue(dmub_srv, cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
-	dc_dmub_srv_wait_idle(dmub_srv);
-}
-
 static void dc_build_cursor_position_update_payload0(
 		struct dmub_cmd_update_cursor_payload0 *pl, const uint8_t p_idx,
 		const struct hubp *hubp, const struct dpp *dpp)
@@ -1032,9 +993,11 @@ static void dc_build_cursor_attribute_update_payload1(
 void dc_send_update_cursor_info_to_dmu(
 		struct pipe_ctx *pCtx, uint8_t pipe_idx)
 {
-	union dmub_rb_cmd cmd = { 0 };
-	union dmub_cmd_update_cursor_info_data *update_cursor_info =
-					&cmd.update_cursor_info.update_cursor_info_data;
+	union dmub_rb_cmd cmd[2];
+	union dmub_cmd_update_cursor_info_data *update_cursor_info_0 =
+					&cmd[0].update_cursor_info.update_cursor_info_data;
+
+	memset(cmd, 0, sizeof(cmd));
 
 	if (!dc_dmub_should_update_cursor_data(pCtx))
 		return;
@@ -1051,31 +1014,28 @@ void dc_send_update_cursor_info_to_dmu(
 
 	{
 		/* Build Payload#0 Header */
-		cmd.update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO;
-		cmd.update_cursor_info.header.payload_bytes =
-				sizeof(cmd.update_cursor_info.update_cursor_info_data);
-		cmd.update_cursor_info.header.multi_cmd_pending = 1; /* To combine multi dmu cmd, 1st cmd */
+		cmd[0].update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO;
+		cmd[0].update_cursor_info.header.payload_bytes =
+				sizeof(cmd[0].update_cursor_info.update_cursor_info_data);
+		cmd[0].update_cursor_info.header.multi_cmd_pending = 1; //To combine multi dmu cmd, 1st cmd
 
 		/* Prepare Payload */
-		dc_build_cursor_update_payload0(pCtx, pipe_idx, &update_cursor_info->payload0);
+		dc_build_cursor_update_payload0(pCtx, pipe_idx, &update_cursor_info_0->payload0);
 
-		dc_build_cursor_position_update_payload0(&update_cursor_info->payload0, pipe_idx,
+		dc_build_cursor_position_update_payload0(&update_cursor_info_0->payload0, pipe_idx,
 				pCtx->plane_res.hubp, pCtx->plane_res.dpp);
-		/* Send update_curosr_info to queue */
-		dc_dmub_srv_cmd_queue(pCtx->stream->ctx->dmub_srv, &cmd);
-	}
+		}
 	{
 		/* Build Payload#1 Header */
-		memset(update_cursor_info, 0, sizeof(union dmub_cmd_update_cursor_info_data));
-		cmd.update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO;
-		cmd.update_cursor_info.header.payload_bytes = sizeof(struct cursor_attributes_cfg);
-		cmd.update_cursor_info.header.multi_cmd_pending = 0; /* Indicate it's the last command. */
+		cmd[1].update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO;
+		cmd[1].update_cursor_info.header.payload_bytes = sizeof(struct cursor_attributes_cfg);
+		cmd[1].update_cursor_info.header.multi_cmd_pending = 0; //Indicate it's the last command.
 
 		dc_build_cursor_attribute_update_payload1(
-				&cmd.update_cursor_info.update_cursor_info_data.payload1.attribute_cfg,
+				&cmd[1].update_cursor_info.update_cursor_info_data.payload1.attribute_cfg,
 				pipe_idx, pCtx->plane_res.hubp, pCtx->plane_res.dpp);
 
 		/* Combine 2nd cmds update_curosr_info to DMU */
-		dc_send_cmd_to_dmu(pCtx->stream->ctx->dmub_srv, &cmd);
+		dm_execute_dmub_cmd_list(pCtx->stream->ctx, 2, cmd, DM_DMUB_WAIT_TYPE_WAIT);
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index d34f5563df2ec..22f7b2704c8ec 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -26,7 +26,7 @@
 #ifndef _DMUB_DC_SRV_H_
 #define _DMUB_DC_SRV_H_
 
-#include "os_types.h"
+#include "dm_services_types.h"
 #include "dmub/dmub_srv.h"
 
 struct dmub_srv;
@@ -52,16 +52,13 @@ struct dc_dmub_srv {
 	void *dm;
 };
 
-void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv,
-			   union dmub_rb_cmd *cmd);
-
-void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv);
-
 void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv);
 
 void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv);
 
-bool dc_dmub_srv_cmd_with_reply_data(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd);
+bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type);
+
+bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type);
 
 bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
 				    unsigned int stream_mask);
@@ -77,7 +74,7 @@ void dc_dmub_srv_drr_update_cmd(struct dc *dc, uint32_t tg_inst, uint32_t vtotal
 void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst);
 bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool enable_pstate, struct dc_state *context);
 
-void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub);
+void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv);
 void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pipe_ctx);
 void dc_dmub_srv_clear_inbox0_ack(struct dc_dmub_srv *dmub_srv);
 void dc_dmub_srv_wait_for_inbox0_ack(struct dc_dmub_srv *dmub_srv);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index f43cce16bb6ce..a21948267c0f8 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -41,19 +41,13 @@ static inline void submit_dmub_read_modify_write(
 	const struct dc_context *ctx)
 {
 	struct dmub_rb_cmd_read_modify_write *cmd_buf = &offload->cmd_data.read_modify_write;
-	bool gather = false;
 
 	offload->should_burst_write =
 			(offload->same_addr_count == (DMUB_READ_MODIFY_WRITE_SEQ__MAX - 1));
 	cmd_buf->header.payload_bytes =
 			sizeof(struct dmub_cmd_read_modify_write_sequence) * offload->reg_seq_count;
 
-	gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress;
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = false;
-
-	dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data);
-
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather;
+	dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 	memset(cmd_buf, 0, sizeof(*cmd_buf));
 
@@ -66,17 +60,11 @@ static inline void submit_dmub_burst_write(
 	const struct dc_context *ctx)
 {
 	struct dmub_rb_cmd_burst_write *cmd_buf = &offload->cmd_data.burst_write;
-	bool gather = false;
 
 	cmd_buf->header.payload_bytes =
 			sizeof(uint32_t) * offload->reg_seq_count;
 
-	gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress;
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = false;
-
-	dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data);
-
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather;
+	dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 	memset(cmd_buf, 0, sizeof(*cmd_buf));
 
@@ -88,17 +76,11 @@ static inline void submit_dmub_reg_wait(
 		const struct dc_context *ctx)
 {
 	struct dmub_rb_cmd_reg_wait *cmd_buf = &offload->cmd_data.reg_wait;
-	bool gather = false;
 
-	gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress;
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = false;
-
-	dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data);
+	dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 	memset(cmd_buf, 0, sizeof(*cmd_buf));
 	offload->reg_seq_count = 0;
-
-	ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather;
 }
 
 struct dc_reg_value_masks {
@@ -151,7 +133,6 @@ static void dmub_flush_buffer_execute(
 		const struct dc_context *ctx)
 {
 	submit_dmub_read_modify_write(offload, ctx);
-	dc_dmub_srv_cmd_execute(ctx->dmub_srv);
 }
 
 static void dmub_flush_burst_write_buffer_execute(
@@ -159,7 +140,6 @@ static void dmub_flush_burst_write_buffer_execute(
 		const struct dc_context *ctx)
 {
 	submit_dmub_burst_write(offload, ctx);
-	dc_dmub_srv_cmd_execute(ctx->dmub_srv);
 }
 
 static bool dmub_reg_value_burst_set_pack(const struct dc_context *ctx, uint32_t addr,
@@ -691,8 +671,6 @@ void reg_sequence_start_execute(const struct dc_context *ctx)
 		default:
 			return;
 		}
-
-		dc_dmub_srv_cmd_execute(ctx->dmub_srv);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
index 4055d271ac57f..e152c68edfd1e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
@@ -75,9 +75,7 @@ static void dmub_abm_enable_fractional_pwm(struct dc_context *dc)
 	cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask;
 	cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dmub_abm_init(struct abm *abm, uint32_t backlight)
@@ -156,9 +154,7 @@ bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask)
 	cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask;
 	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -180,9 +176,7 @@ void dmub_abm_set_ambient_level(struct abm *abm, unsigned int ambient_lux, uint8
 	cmd.abm_set_ambient_level.abm_set_ambient_level_data.panel_mask = panel_mask;
 	cmd.abm_set_ambient_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_ambient_level_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 #endif
 
@@ -212,9 +206,7 @@ void dmub_abm_init_config(struct abm *abm,
 
 	cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 }
 
@@ -231,9 +223,7 @@ bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, un
 	cmd.abm_pause.abm_pause_data.panel_mask = panel_mask;
 	cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -253,9 +243,7 @@ bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint
 	cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary;
 	cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -277,9 +265,7 @@ bool dmub_abm_set_backlight_level(struct abm *abm,
 	cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst);
 	cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c
index 3f32e9c3fbaf4..2aa0e01a6891b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c
@@ -47,9 +47,7 @@ void dmub_hw_lock_mgr_cmd(struct dc_dmub_srv *dmub_srv,
 	if (!lock)
 		cmd.lock_hw.lock_hw_data.should_release = 1;
 
-	dc_dmub_srv_cmd_queue(dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
-	dc_dmub_srv_wait_idle(dmub_srv);
+	dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c
index fff1d07d865d7..d8009b2dc56a0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c
@@ -48,7 +48,5 @@ void dmub_enable_outbox_notification(struct dc_dmub_srv *dmub_srv)
 		sizeof(cmd.outbox1_enable.header);
 	cmd.outbox1_enable.enable = true;
 
-	dc_dmub_srv_cmd_queue(dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dmub_srv);
-	dc_dmub_srv_wait_idle(dmub_srv);
+	dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
index 9705d8f883825..4000a834592c0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@@ -168,9 +168,7 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state *
 	cmd.psr_set_version.psr_set_version_data.panel_inst = panel_inst;
 	cmd.psr_set_version.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -198,9 +196,7 @@ static void dmub_psr_enable(struct dmub_psr *dmub, bool enable, bool wait, uint8
 
 	cmd.psr_enable.header.payload_bytes = 0; // Send header only
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc->dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	/* Below loops 1000 x 500us = 500 ms.
 	 *  Exit PSR may need to wait 1-2 frames to power up. Timeout after at
@@ -248,9 +244,7 @@ static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level, uint8_
 	cmd.psr_set_level.psr_set_level_data.psr_level = psr_level;
 	cmd.psr_set_level.psr_set_level_data.cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1;
 	cmd.psr_set_level.psr_set_level_data.panel_inst = panel_inst;
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 /*
@@ -269,9 +263,7 @@ static void dmub_psr_set_sink_vtotal_in_psr_active(struct dmub_psr *dmub,
 	cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_idle = psr_vtotal_idle;
 	cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_su = psr_vtotal_su;
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 /*
@@ -290,9 +282,7 @@ static void dmub_psr_set_power_opt(struct dmub_psr *dmub, unsigned int power_opt
 	cmd.psr_set_power_opt.psr_set_power_opt_data.power_opt = power_opt;
 	cmd.psr_set_power_opt.psr_set_power_opt_data.panel_inst = panel_inst;
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 /*
@@ -422,9 +412,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub,
 		copy_settings_data->relock_delay_frame_cnt = 2;
 	copy_settings_data->dsc_slice_height = psr_context->dsc_slice_height;
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -445,9 +433,7 @@ static void dmub_psr_force_static(struct dmub_psr *dmub, uint8_t panel_inst)
 	cmd.psr_force_static.header.sub_type = DMUB_CMD__PSR_FORCE_STATIC;
 	cmd.psr_enable.header.payload_bytes = 0;
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c
index 58e459c7e7d35..f976fac8dc3f6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c
@@ -667,7 +667,6 @@ static void program_surface_flip_and_addr(struct hubp *hubp, struct surface_flip
 static void dmcub_PLAT_54186_wa(struct hubp *hubp,
 				struct surface_flip_registers *flip_regs)
 {
-	struct dc_dmub_srv *dmcub = hubp->ctx->dmub_srv;
 	struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp);
 	union dmub_rb_cmd cmd;
 
@@ -690,11 +689,7 @@ static void dmcub_PLAT_54186_wa(struct hubp *hubp,
 	cmd.PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid;
 
 	PERF_TRACE();  // TODO: remove after performance is stable.
-	dc_dmub_srv_cmd_queue(dmcub, &cmd);
-	PERF_TRACE();  // TODO: remove after performance is stable.
-	dc_dmub_srv_cmd_execute(dmcub);
-	PERF_TRACE();  // TODO: remove after performance is stable.
-	dc_dmub_srv_wait_idle(dmcub);
+	dm_execute_dmub_cmd(hubp->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 	PERF_TRACE();  // TODO: remove after performance is stable.
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
index 1c6477d73c8e1..55a464a39529b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
@@ -152,9 +152,7 @@ static bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t optio
 	cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary;
 	cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -173,9 +171,7 @@ static void dmub_abm_set_backlight(struct dc_context *dc, uint32_t backlight_pwm
 	cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst);
 	cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data);
 
-	dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->dmub_srv);
+	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index 8263a07f265f2..3303c9aae0686 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -632,7 +632,7 @@ void dcn30_init_hw(struct dc *dc)
 		dc->res_pool->hubbub->funcs->init_crb(dc->res_pool->hubbub);
 
 	// Get DMCUB capabilities
-	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
+	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 	dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
 	dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 }
@@ -736,8 +736,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable)
 				cmd.mall.header.sub_type = DMUB_CMD__MALL_ACTION_NO_DF_REQ;
 				cmd.mall.header.payload_bytes = sizeof(cmd.mall) - sizeof(cmd.mall.header);
 
-				dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-				dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
+				dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 				return true;
 			}
@@ -859,9 +858,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable)
 					cmd.mall.cursor_height = cursor_attr.height;
 					cmd.mall.cursor_pitch = cursor_attr.pitch;
 
-					dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-					dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-					dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+					dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 					/* Use copied cursor, and it's okay to not switch back */
 					cursor_attr.address.quad_part = cmd.mall.cursor_copy_dst.quad_part;
@@ -877,8 +874,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable)
 				cmd.mall.tmr_scale = tmr_scale;
 				cmd.mall.debug_bits = dc->debug.mall_error_as_fatal;
 
-				dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-				dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
+				dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 				return true;
 			}
@@ -895,9 +891,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable)
 	cmd.mall.header.payload_bytes =
 		sizeof(cmd.mall) - sizeof(cmd.mall.header);
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c
index 745a5d187a98d..bd62502380d8d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c
@@ -117,7 +117,6 @@ static bool query_dp_alt_from_dmub(struct link_encoder *enc,
 				   union dmub_rb_cmd *cmd)
 {
 	struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc);
-	struct dc_dmub_srv *dc_dmub_srv = enc->ctx->dmub_srv;
 
 	memset(cmd, 0, sizeof(*cmd));
 	cmd->query_dp_alt.header.type = DMUB_CMD__VBIOS;
@@ -126,7 +125,7 @@ static bool query_dp_alt_from_dmub(struct link_encoder *enc,
 	cmd->query_dp_alt.header.payload_bytes = sizeof(cmd->query_dp_alt.data);
 	cmd->query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter);
 
-	if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, cmd))
+	if (!dm_execute_dmub_cmd(enc->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
 		return false;
 
 	return true;
@@ -425,7 +424,6 @@ static bool link_dpia_control(struct dc_context *dc_ctx,
 	struct dmub_cmd_dig_dpia_control_data *dpia_control)
 {
 	union dmub_rb_cmd cmd;
-	struct dc_dmub_srv *dmub = dc_ctx->dmub_srv;
 
 	memset(&cmd, 0, sizeof(cmd));
 
@@ -438,9 +436,7 @@ static bool link_dpia_control(struct dc_context *dc_ctx,
 
 	cmd.dig1_dpia_control.dpia_control = *dpia_control;
 
-	dc_dmub_srv_cmd_queue(dmub, &cmd);
-	dc_dmub_srv_cmd_execute(dmub);
-	dc_dmub_srv_wait_idle(dmub);
+	dm_execute_dmub_cmd(dc_ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index 62ce36c75c4d2..e0c74868d2ee5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -297,7 +297,7 @@ void dcn31_init_hw(struct dc *dc)
 #endif
 
 	// Get DMCUB capabilities
-	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
+	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 	dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
 }
 
@@ -442,9 +442,7 @@ void dcn31_z10_save_init(struct dc *dc)
 	cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT;
 	cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dcn31_z10_restore(const struct dc *dc)
@@ -462,9 +460,7 @@ void dcn31_z10_restore(const struct dc *dc)
 	cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT;
 	cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_RESTORE;
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 }
 
 void dcn31_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c
index 11ea9d13e3128..217acd4e292a3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c
@@ -52,7 +52,7 @@ static bool dcn31_query_backlight_info(struct panel_cntl *panel_cntl, union dmub
 	cmd->panel_cntl.header.payload_bytes = sizeof(cmd->panel_cntl.data);
 	cmd->panel_cntl.data.inst = dcn31_panel_cntl->base.inst;
 
-	return dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, cmd);
+	return dm_execute_dmub_cmd(dc_dmub_srv->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY);
 }
 
 static uint32_t dcn31_get_16_bit_backlight_from_pwm(struct panel_cntl *panel_cntl)
@@ -85,7 +85,7 @@ static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl)
 		panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV;
 	cmd.panel_cntl.data.bl_pwm_ref_div2 =
 		panel_cntl->stored_backlight_registers.PANEL_PWRSEQ_REF_DIV2;
-	if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, &cmd))
+	if (!dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY))
 		return 0;
 
 	panel_cntl->stored_backlight_registers.BL_PWM_CNTL = cmd.panel_cntl.data.bl_pwm_cntl;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 40c488b26901b..6fb3f64e30576 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -417,9 +417,7 @@ void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool
 	cmd.domain_control.data.inst = hubp_inst;
 	cmd.domain_control.data.power_gate = !power_on;
 
-	dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(ctx->dmub_srv);
+	dm_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	PERF_TRACE();
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 26791e3d162f4..9ce11ed769a07 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -274,8 +274,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable)
 				cmd.cab.header.sub_type = DMUB_CMD__CAB_NO_DCN_REQ;
 				cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header);
 
-				dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-				dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
+				dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 				return true;
 			}
@@ -309,8 +308,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable)
 				cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header);
 				cmd.cab.cab_alloc_ways = ways;
 
-				dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-				dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
+				dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT);
 
 				return true;
 			}
@@ -326,9 +324,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable)
 	cmd.cab.header.payload_bytes =
 			sizeof(cmd.cab) - sizeof(cmd.cab.header);
 
-	dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd);
-	dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv);
-	dc_dmub_srv_wait_idle(dc->ctx->dmub_srv);
+	dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
 }
@@ -946,7 +942,7 @@ void dcn32_init_hw(struct dc *dc)
 
 	// Get DMCUB capabilities
 	if (dc->ctx->dmub_srv) {
-		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub);
+		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 		dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
 		dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index 9a3f2a44f8824..d33d595405a9d 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -40,6 +40,7 @@
 
 struct dmub_srv;
 struct dc_dmub_srv;
+union dmub_rb_cmd;
 
 irq_handler_idx dm_register_interrupt(
 	struct dc_context *ctx,
@@ -273,6 +274,12 @@ void dm_perf_trace_timestamp(const char *func_name, unsigned int line, struct dc
 #define PERF_TRACE()	dm_perf_trace_timestamp(__func__, __LINE__, CTX)
 #define PERF_TRACE_CTX(__CTX)	dm_perf_trace_timestamp(__func__, __LINE__, __CTX)
 
+/*
+ * DMUB Interfaces
+ */
+bool dm_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type);
+bool dm_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type);
+
 /*
  * Debug and verification hooks
  */
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
index b52ba6ffabe1c..facf269c4326d 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
@@ -269,4 +269,10 @@ struct dtn_min_clk_info {
 	uint32_t min_memory_clock_khz;
 };
 
+enum dm_dmub_wait_type {
+	DM_DMUB_WAIT_TYPE_NO_WAIT,
+	DM_DMUB_WAIT_TYPE_WAIT,
+	DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY,
+};
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 2914fca7dab30..5ff5d4e649022 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1394,7 +1394,7 @@ static bool get_usbc_cable_id(struct dc_link *link, union dp_cable_id *cable_id)
 	cmd.cable_id.header.payload_bytes = sizeof(cmd.cable_id.data);
 	cmd.cable_id.data.input.phy_inst = resource_transmitter_to_phy_idx(
 			link->dc, link->link_enc->transmitter);
-	if (dc_dmub_srv_cmd_with_reply_data(link->ctx->dmub_srv, &cmd) &&
+	if (dm_execute_dmub_cmd(link->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) &&
 			cmd.cable_id.header.ret_status == 1) {
 		cable_id->raw = cmd.cable_id.data.output_raw;
 		DC_LOG_DC("usbc_cable_id = %d.\n", cable_id->raw);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
index 4626fabc0a96b..0bb7491339098 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c
@@ -90,7 +90,7 @@ bool dpia_query_hpd_status(struct dc_link *link)
 	cmd.query_hpd.data.ch_type = AUX_CHANNEL_DPIA;
 
 	/* Return HPD status reported by DMUB if query successfully executed. */
-	if (dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd) && cmd.query_hpd.data.status == AUX_RET_SUCCESS)
+	if (dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.query_hpd.data.status == AUX_RET_SUCCESS)
 		is_hpd_high = cmd.query_hpd.data.result;
 
 	DC_LOG_DEBUG("%s: link(%d) dpia(%d) cmd_status(%d) result(%d)\n",
-- 
GitLab


From 522b9a5d5852f99e51fbc460054dc8af3b4b5b30 Mon Sep 17 00:00:00 2001
From: Josip Pavic <Josip.Pavic@amd.com>
Date: Fri, 24 Mar 2023 15:31:10 -0400
Subject: [PATCH 0107/1791] drm/amd/display: drain dmub inbox if queue is full

[Why & How]
If dmub command queuing fails due to the inbox being full, flush the
inbox and resubmit the comamnd. This was previously the default behavior
but was lost in a refactor.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Josip Pavic <Josip.Pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 954cbfdbc3b6a..eef43577508cf 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -139,6 +139,15 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun
 		// Queue command
 		status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
 
+		if (status == DMUB_STATUS_QUEUE_FULL) {
+			/* Execute and wait for queue to become empty again. */
+			dmub_srv_cmd_execute(dmub);
+			dmub_srv_wait_for_idle(dmub, 100000);
+
+			/* Requeue the command. */
+			status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+		}
+
 		if (status != DMUB_STATUS_OK) {
 			DC_ERROR("Error queueing DMUB command: status=%d\n", status);
 			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
-- 
GitLab


From cdff36a0217aadf5cbc167893ad1c0da869619cb Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Tue, 28 Mar 2023 10:45:24 -0400
Subject: [PATCH 0108/1791] drm/amd/display: fix access hdcp_workqueue assert

[Why] hdcp are enabled for asics from raven. for old asics
which hdcp are not enabled, hdcp_workqueue are null. some
access to hdcp work queue are not guarded with pointer check.

[How] add hdcp_workqueue pointer check before access workqueue.

Reviewed-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c    |  6 ++++++
 .../amd/display/amdgpu_dm/amdgpu_dm_mst_types.c  | 16 ++++++++++------
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index af3efb2456103..6e9d2e680d502 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8559,6 +8559,9 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
 		struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
 
+		if (!adev->dm.hdcp_workqueue)
+			continue;
+
 		pr_debug("[HDCP_DM] -------------- i : %x ----------\n", i);
 
 		if (!connector)
@@ -8607,6 +8610,9 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
 		struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
 
+		if (!adev->dm.hdcp_workqueue)
+			continue;
+
 		new_crtc_state = NULL;
 		old_crtc_state = NULL;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 994ba426ca66d..5dc79b753d5fb 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -379,13 +379,17 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector)
 		if (aconnector->dc_sink && connector->state) {
 			struct drm_device *dev = connector->dev;
 			struct amdgpu_device *adev = drm_to_adev(dev);
-			struct hdcp_workqueue *hdcp_work = adev->dm.hdcp_workqueue;
-			struct hdcp_workqueue *hdcp_w = &hdcp_work[aconnector->dc_link->link_index];
 
-			connector->state->hdcp_content_type =
-			hdcp_w->hdcp_content_type[connector->index];
-			connector->state->content_protection =
-			hdcp_w->content_protection[connector->index];
+			if (adev->dm.hdcp_workqueue) {
+				struct hdcp_workqueue *hdcp_work = adev->dm.hdcp_workqueue;
+				struct hdcp_workqueue *hdcp_w =
+					&hdcp_work[aconnector->dc_link->link_index];
+
+				connector->state->hdcp_content_type =
+				hdcp_w->hdcp_content_type[connector->index];
+				connector->state->content_protection =
+				hdcp_w->content_protection[connector->index];
+			}
 		}
 
 		if (aconnector->dc_sink) {
-- 
GitLab


From e3416e872f84086667df21daf166506fab97358d Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Date: Fri, 24 Mar 2023 14:29:52 -0600
Subject: [PATCH 0109/1791] drm/amd/display: Add FAMS validation before trying
 to use it

To ensure that FAMS can be used, DC must check if there is VRR support.
This commit adds the required configuration to ensure FAMS can be executed in the target system.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c          | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dc_stream.h        | 1 +
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c | 7 ++++++-
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 2 +-
 4 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 61eec5aa40671..105f705bd91a8 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2604,6 +2604,12 @@ static enum surface_update_type check_update_surfaces_for_stream(
 
 		if (stream_update->mst_bw_update)
 			su_flags->bits.mst_bw = 1;
+
+		if (stream_update->stream && stream_update->stream->freesync_on_desktop &&
+			(stream_update->vrr_infopacket || stream_update->allow_freesync ||
+				stream_update->vrr_active_variable))
+			su_flags->bits.fams_changed = 1;
+
 		if (stream_update->crtc_timing_adjust && dc_extended_blank_supported(dc))
 			su_flags->bits.crtc_timing_adjust = 1;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 25284006019c3..270282fbda4ab 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -131,6 +131,7 @@ union stream_update_flags {
 		uint32_t dsc_changed : 1;
 		uint32_t mst_bw : 1;
 		uint32_t crtc_timing_adjust : 1;
+		uint32_t fams_changed : 1;
 	} bits;
 
 	uint32_t raw;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
index c95f000b63b28..34b08d90dc1da 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
@@ -301,7 +301,12 @@ static void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *o
 
 void optc3_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max)
 {
-	optc1_set_vtotal_min_max(optc, vtotal_min, vtotal_max);
+	struct dc *dc = optc->ctx->dc;
+
+	if (dc->caps.dmub_caps.mclk_sw && !dc->debug.disable_fams)
+		dc_dmub_srv_drr_update_cmd(dc, optc->inst, vtotal_min, vtotal_max);
+	else
+		optc1_set_vtotal_min_max(optc, vtotal_min, vtotal_max);
 }
 
 void optc3_tg_init(struct timing_generator *optc)
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 598fa1de54ce3..1c55d3b01f53e 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -360,7 +360,7 @@ union dmub_fw_boot_status {
 		uint32_t optimized_init_done : 1; /**< 1 if optimized init done */
 		uint32_t restore_required : 1; /**< 1 if driver should call restore */
 		uint32_t defer_load : 1; /**< 1 if VBIOS data is deferred programmed */
-		uint32_t reserved : 1;
+		uint32_t fams_enabled : 1; /**< 1 if VBIOS data is deferred programmed */
 		uint32_t detection_required: 1; /**<  if detection need to be triggered by driver */
 		uint32_t hw_power_init_done: 1; /**< 1 if hw power init is completed */
 	} bits; /**< status bits */
-- 
GitLab


From 4d5f872dbc755114628c236e17421629ec522203 Mon Sep 17 00:00:00 2001
From: Iswara Nagulendran <Iswara.Nagulendran@amd.com>
Date: Thu, 16 Mar 2023 16:29:06 -0400
Subject: [PATCH 0110/1791] drm/amd/display: Adding support for VESA SCR

[HOW&WHY]
Write DPCD 721 bit 7 to high, and
the appropriate luminance level
to DPCD 734-736 if bit 4 from DPCD register
734 is high, indicating that the panel
luminance control is enabled from the panel side.

Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Iswara Nagulendran <Iswara.Nagulendran@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dp_types.h  |  7 +++++
 .../dc/link/protocols/link_dp_capability.c    |  9 +++++-
 .../link/protocols/link_edp_panel_control.c   | 29 ++++++++++++++++---
 include/drm/display/drm_dp.h                  |  3 ++
 4 files changed, 43 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 49aab1924665a..4a7f6497dc5a4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -566,6 +566,12 @@ struct dpcd_amd_device_id {
 	uint8_t dal_version_byte2;
 };
 
+struct target_luminance_value {
+	uint8_t byte0;
+	uint8_t byte1;
+	uint8_t byte2;
+};
+
 struct dpcd_source_backlight_set {
 	struct  {
 		uint8_t byte0;
@@ -1225,6 +1231,7 @@ struct dpcd_caps {
 	union dp_main_line_channel_coding_cap channel_coding_cap;
 	union dp_sink_video_fallback_formats fallback_formats;
 	union dp_fec_capability1 fec_cap1;
+	bool panel_luminance_control;
 	union dp_cable_id cable_id;
 	uint8_t edp_rev;
 	union edp_alpm_caps alpm_caps;
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 5ff5d4e649022..84265dc66bbab 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1450,7 +1450,8 @@ bool read_is_mst_supported(struct dc_link *link)
  */
 static bool dpcd_read_sink_ext_caps(struct dc_link *link)
 {
-	uint8_t dpcd_data;
+	uint8_t dpcd_data = 0;
+	uint8_t edp_general_cap2 = 0;
 
 	if (!link)
 		return false;
@@ -1459,6 +1460,12 @@ static bool dpcd_read_sink_ext_caps(struct dc_link *link)
 		return false;
 
 	link->dpcd_sink_ext_caps.raw = dpcd_data;
+
+	if (core_link_read_dpcd(link, DP_EDP_GENERAL_CAP_2, &edp_general_cap2, 1) != DC_OK)
+		return false;
+
+	link->dpcd_caps.panel_luminance_control = (edp_general_cap2 & DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE) != 0;
+
 	return true;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index d895046787bc4..5ab2de12ccf87 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -164,14 +164,35 @@ bool edp_set_backlight_level_nits(struct dc_link *link,
 	*(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms;
 
 
-	if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL,
+	if (!link->dpcd_caps.panel_luminance_control) {
+		if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL,
 			(uint8_t *)(&dpcd_backlight_set),
 			sizeof(dpcd_backlight_set)) != DC_OK)
-		return false;
+			return false;
 
-	if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL,
+		if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL,
 			&backlight_control, 1) != DC_OK)
-		return false;
+			return false;
+	} else {
+		const uint8_t backlight_enable = DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE;
+		struct target_luminance_value *target_luminance = NULL;
+
+		//if target luminance value is greater than 24 bits, clip the value to 24 bits
+		if (backlight_millinits > 0xFFFFFF)
+			backlight_millinits = 0xFFFFFF;
+
+		target_luminance = (struct target_luminance_value *)&backlight_millinits;
+
+		if (core_link_write_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER,
+			&backlight_enable,
+			sizeof(backlight_enable)) != DC_OK)
+			return false;
+
+		if (core_link_write_dpcd(link, DP_EDP_PANEL_TARGET_LUMINANCE_VALUE,
+			(uint8_t *)(target_luminance),
+			sizeof(struct target_luminance_value)) != DC_OK)
+			return false;
+	}
 
 	return true;
 }
diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h
index 358db4a9f167a..d735073fdd814 100644
--- a/include/drm/display/drm_dp.h
+++ b/include/drm/display/drm_dp.h
@@ -983,6 +983,7 @@
 
 #define DP_EDP_GENERAL_CAP_2		    0x703
 # define DP_EDP_OVERDRIVE_ENGINE_ENABLED		(1 << 0)
+# define DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE (1 << 4)
 
 #define DP_EDP_GENERAL_CAP_3		    0x704    /* eDP 1.4 */
 # define DP_EDP_X_REGION_CAP_MASK			(0xf << 0)
@@ -1008,6 +1009,7 @@
 # define DP_EDP_DYNAMIC_BACKLIGHT_ENABLE		(1 << 4)
 # define DP_EDP_REGIONAL_BACKLIGHT_ENABLE		(1 << 5)
 # define DP_EDP_UPDATE_REGION_BRIGHTNESS		(1 << 6) /* eDP 1.4 */
+# define DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE  (1 << 7)
 
 #define DP_EDP_BACKLIGHT_BRIGHTNESS_MSB     0x722
 #define DP_EDP_BACKLIGHT_BRIGHTNESS_LSB     0x723
@@ -1032,6 +1034,7 @@
 
 #define DP_EDP_DBC_MINIMUM_BRIGHTNESS_SET   0x732
 #define DP_EDP_DBC_MAXIMUM_BRIGHTNESS_SET   0x733
+#define DP_EDP_PANEL_TARGET_LUMINANCE_VALUE 0x734
 
 #define DP_EDP_REGIONAL_BACKLIGHT_BASE      0x740    /* eDP 1.4 */
 #define DP_EDP_REGIONAL_BACKLIGHT_0	    0x741    /* eDP 1.4 */
-- 
GitLab


From 6bfe9a23a8d6c7292d520747859a515fd429518d Mon Sep 17 00:00:00 2001
From: Nasir Osman <nasir.osman@amd.com>
Date: Thu, 30 Mar 2023 15:08:42 -0400
Subject: [PATCH 0111/1791] drm/amd/display: DSC policy override when ODM
 combine is forced

[why]
When we force ODM combine with DSC, we lose several
8 bit and 10 bit modes in validation and thus
not able to use HDR. This is due to the number of
horizontal slices used in DSC not properly being
accounted for currently when 2:1 ODM Combine is forced.

[how]
Enforce at least two horizontal slices are used for DSC when
ODM combine is forced.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Nasir Osman <nasir.osman@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dsc.h     |  1 +
 drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c | 10 +++++++++-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dsc.h b/drivers/gpu/drm/amd/display/dc/dc_dsc.h
index 0e92a322c2ed3..9491b76d61f58 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dsc.h
@@ -58,6 +58,7 @@ struct dc_dsc_config_options {
 	uint32_t dsc_min_slice_height_override;
 	uint32_t max_target_bpp_limit_override_x16;
 	uint32_t slice_height_granularity;
+	uint32_t dsc_force_odm_hslice_override;
 };
 
 bool dc_dsc_parse_dsc_dpcd(const struct dc *dc,
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index 2bdc47615543c..b9a05bb025db6 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -700,7 +700,7 @@ static int inc_num_slices(union dsc_enc_slice_caps slice_caps, int num_slices)
 		}
 	}
 
-	if (new_num_slices == num_slices) // No biger number of slices found
+	if (new_num_slices == num_slices) // No bigger number of slices found
 		new_num_slices++;
 
 	return new_num_slices;
@@ -952,6 +952,13 @@ static bool setup_dsc_config(
 		else
 			is_dsc_possible = false;
 	}
+	// When we force 2:1 ODM, we can't have 1 slice to divide amongst 2 separate DSC instances
+	// need to enforce at minimum 2 horizontal slices
+	if (options->dsc_force_odm_hslice_override) {
+		num_slices_h = fit_num_slices_up(dsc_common_caps.slice_caps, 2);
+		if (num_slices_h == 0)
+			is_dsc_possible = false;
+	}
 
 	if (!is_dsc_possible)
 		goto done;
@@ -1163,6 +1170,7 @@ void dc_dsc_policy_set_disable_dsc_stream_overhead(bool disable)
 void dc_dsc_get_default_config_option(const struct dc *dc, struct dc_dsc_config_options *options)
 {
 	options->dsc_min_slice_height_override = dc->debug.dsc_min_slice_height_override;
+	options->dsc_force_odm_hslice_override = dc->debug.force_odm_combine;
 	options->max_target_bpp_limit_override_x16 = 0;
 	options->slice_height_granularity = 1;
 }
-- 
GitLab


From 7052a801d6bc8cd203e1708313e4996630208a6e Mon Sep 17 00:00:00 2001
From: Michael Mityushkin <michael.mityushkin@amd.com>
Date: Thu, 30 Mar 2023 12:46:58 -0400
Subject: [PATCH 0112/1791] drm/amd/display: Correct output color space during
 HW reinitialize

[Why]
Doing core_link_disable_stream or set_dpms_off when reinitializing
hardware causes issue to repro with external display connected. This is
unnecessary, blanking pixel data should be sufficient.

[How]
Call disable_pixel_data while reinitializing hardware instead of
core_link_disable_stream or set_dpms_off.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Michael Mityushkin <michael.mityushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c  | 4 ++++
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c   | 1 +
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c | 1 +
 3 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 422fbf79da64f..5800acf6aae11 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -313,6 +313,10 @@ void dcn20_init_blank(
 	}
 	opp = dc->res_pool->opps[opp_id_src0];
 
+	/* don't override the blank pattern if already enabled with the correct one. */
+	if (opp->funcs->dpg_is_blanked && opp->funcs->dpg_is_blanked(opp))
+		return;
+
 	if (num_opps == 2) {
 		otg_active_width = otg_active_width / 2;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
index 3a32810bbe382..8598ea233ef3e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
@@ -58,6 +58,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = {
 	.enable_audio_stream = dce110_enable_audio_stream,
 	.disable_audio_stream = dce110_disable_audio_stream,
 	.disable_plane = dcn20_disable_plane,
+	.disable_pixel_data = dcn20_disable_pixel_data,
 	.pipe_control_lock = dcn20_pipe_control_lock,
 	.interdependent_update_lock = dcn10_lock_all_pipes,
 	.cursor_lock = dcn10_cursor_lock,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index 5267e901a35c1..ce53339b2e10d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -60,6 +60,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = {
 	.enable_audio_stream = dce110_enable_audio_stream,
 	.disable_audio_stream = dce110_disable_audio_stream,
 	.disable_plane = dcn20_disable_plane,
+	.disable_pixel_data = dcn20_disable_pixel_data,
 	.pipe_control_lock = dcn20_pipe_control_lock,
 	.interdependent_update_lock = dcn10_lock_all_pipes,
 	.cursor_lock = dcn10_cursor_lock,
-- 
GitLab


From 9fc6e4b36f2a748c853512d5ce4c8c4b98941c75 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Fri, 31 Mar 2023 17:38:21 -0400
Subject: [PATCH 0113/1791] drm/amd/display: Set watermarks set D equal to A

[Description]
- Since we do not use optimized watermark settings for MALL,
  set D = A
- PMFW uses Set D for d0i3.1, so driver should make D = A for the
  time being
- If we choose to optimize in the future we can set watermarks D
  correctly

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 23a972f2885fe..2f7723053042f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2062,6 +2062,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 	 * sr_enter_exit/sr_exit should be lower than used for DRAM (TBD after bringup or later, use as decided in Clk Mgr)
 	 */
 
+	/*
 	if (dcn3_2_soc.num_states > 2) {
 		vlevel_temp = 0;
 		dcfclk = dc->clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz;
@@ -2088,6 +2089,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 	context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.fclk_pstate_change_ns = get_fclk_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	context->bw_ctx.bw.dcn.watermarks.d.usr_retraining_ns = get_usr_retraining_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
+	*/
 
 	/* Set C, for Dummy P-State:
 	 * All clocks min.
@@ -2189,6 +2191,9 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 		context->bw_ctx.bw.dcn.watermarks.a.usr_retraining_ns = get_usr_retraining_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000;
 	}
 
+	/* Make set D = set A since we do not optimized watermarks for MALL */
+	context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a;
+
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		if (!context->res_ctx.pipe_ctx[i].stream)
 			continue;
-- 
GitLab


From d062de7b68f27546a45c063b046b66c0a73633db Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Fri, 31 Mar 2023 20:02:57 -0400
Subject: [PATCH 0114/1791] drm/amd/display: Enable FPO + Vactive

[Description]
- Enable FPO + Vactive

Reviewed-by: George Shen <George.Shen@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index a876e6eb6cd88..4f8286ae699b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -726,7 +726,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.override_dispclk_programming = true,
 	.disable_fpo_optimizations = false,
 	.fpo_vactive_margin_us = 2000, // 2000us
-	.disable_fpo_vactive = true,
+	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index e5ab7f3077c40..cf21b240fc55e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -725,7 +725,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.override_dispclk_programming = true,
 	.disable_fpo_optimizations = false,
 	.fpo_vactive_margin_us = 2000, // 2000us
-	.disable_fpo_vactive = true,
+	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
 };
 
-- 
GitLab


From 276641775848020c6e84166d1bc885e028a04680 Mon Sep 17 00:00:00 2001
From: Anthony Koo <Anthony.Koo@amd.com>
Date: Sat, 1 Apr 2023 11:45:31 -0400
Subject: [PATCH 0115/1791] drm/amd/display: [FW Promotion] Release 0.0.162.0

 - Add DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command
 - Remove d3 entry event and instead check for stream mask
 - dmu: Enable timeout recovery and detection for p-state

Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Anthony Koo <Anthony.Koo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 1c55d3b01f53e..54b7786f56819 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -419,7 +419,8 @@ union dmub_fw_boot_options {
 		uint32_t dpia_hpd_int_enable_supported: 1; /* 1 if dpia hpd int enable supported */
 		uint32_t usb4_dpia_bw_alloc_supported: 1; /* 1 if USB4 dpia BW allocation supported */
 		uint32_t disable_clk_ds: 1; /* 1 if disallow dispclk_ds and dppclk_ds*/
-		uint32_t reserved : 14; /**< reserved */
+		uint32_t disable_timeout_recovery : 1; /* 1 if timeout recovery should be disabled */
+		uint32_t reserved : 13; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
@@ -1125,8 +1126,6 @@ struct dmub_rb_cmd_idle_opt_dcn_restore {
  */
 struct dmub_dcn_notify_idle_cntl_data {
 	uint8_t driver_idle;
-	uint8_t d3_entry;
-	uint8_t trigger;
 	uint8_t pad[1];
 };
 
@@ -3550,6 +3549,10 @@ union dmub_rb_cmd {
 	 * Definition of a DMUB_CMD__DPIA_HPD_INT_ENABLE command.
 	 */
 	struct dmub_rb_cmd_dpia_hpd_int_enable dpia_hpd_int_enable;
+	/**
+	 * Definition of a DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command.
+	 */
+	struct dmub_rb_cmd_idle_opt_dcn_notify_idle idle_opt_notify_idle;
 };
 
 /**
-- 
GitLab


From 21fc0ff38f571debdba6aaff944addb50f49a7f7 Mon Sep 17 00:00:00 2001
From: Alex Hung <alex.hung@amd.com>
Date: Mon, 3 Apr 2023 17:45:41 +0800
Subject: [PATCH 0116/1791] drm/amd/display: fix a divided-by-zero error

[Why & How]

timing.dsc_cfg.num_slices_v can be zero and it is necessary to check
before using it.

This fixes the error "divide error: 0000 [#1] PREEMPT SMP NOPTI".

Reviewed-by: Aurabindo Pillai <Aurabindo.Pillai@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/modules/power/power_helpers.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 0d3a983cb9ec5..51e76bce92eaa 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -927,6 +927,10 @@ bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link,
 
 	pic_height = stream->timing.v_addressable +
 		stream->timing.v_border_top + stream->timing.v_border_bottom;
+
+	if (stream->timing.dsc_cfg.num_slices_v == 0)
+		return false;
+
 	slice_height = pic_height / stream->timing.dsc_cfg.num_slices_v;
 	config->dsc_slice_height = slice_height;
 
-- 
GitLab


From ec341e0f4a02040ee8d6ef156f8bf02c5aa5c511 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Date: Mon, 20 Mar 2023 16:47:21 -0400
Subject: [PATCH 0117/1791] drm/amd/display: add extra dc odm debug options

[Why & How]
Add options for dc odm debug.

Reviewed-by: Ariel Bernstein <Eric.Bernstein@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h        | 2 ++
 drivers/gpu/drm/amd/display/dc/dc_stream.h | 5 +++++
 2 files changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 23ee63b98dcde..3595149decebe 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -879,6 +879,8 @@ struct dc_debug_options {
 	uint32_t fpo_vactive_margin_us;
 	bool disable_fpo_vactive;
 	bool disable_boot_optimizations;
+	bool override_odm_optimization;
+	bool minimize_dispclk_using_odm;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 270282fbda4ab..0add5ecc895fa 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -172,6 +172,10 @@ struct mall_temp_config {
 	bool is_phantom_plane[MAX_PIPES];
 };
 
+struct dc_stream_debug_options {
+	char force_odm_combine_segments;
+};
+
 struct dc_stream_state {
 	// sink is deprecated, new code should not reference
 	// this pointer
@@ -182,6 +186,7 @@ struct dc_stream_state {
 	 * a stream via the volatile dc_state rather than the static dc_link.
 	 */
 	struct link_encoder *link_enc;
+	struct dc_stream_debug_options debug;
 	struct dc_panel_patch sink_patches;
 	union display_content_support content_support;
 	struct dc_crtc_timing timing;
-- 
GitLab


From f0c5f9ebbc4cfd6b9611b8a5d9fb208c5f60e3e6 Mon Sep 17 00:00:00 2001
From: Michael Mityushkin <michael.mityushkin@amd.com>
Date: Thu, 30 Mar 2023 11:35:08 -0400
Subject: [PATCH 0118/1791] drm/amd/display: Apply correct panel mode when
 reinitializing hardware

[Why]
When link training during engine recovery, ASSR might fail causing panel
mode to be reset to default. This should not happen for eDP as it
will prevent the panel from turning back on.

[How]
Added dp_panel_mode to struct dc_link to remember previously applied
panel mode. Do not reset panel mode to default while performing link
training if previously used panel mode = eDP.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Michael Mityushkin <michael.mityushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                          | 1 +
 .../gpu/drm/amd/display/dc/link/protocols/link_dp_training.c | 5 ++++-
 .../amd/display/dc/link/protocols/link_edp_panel_control.c   | 1 +
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 3595149decebe..3b53f36029d06 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1456,6 +1456,7 @@ struct dc_link {
 
 	struct ddc_service *ddc;
 
+	enum dp_panel_mode panel_mode;
 	bool aux_mode;
 
 	/* Private to DC core */
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index 170f338359301..579fa222810d4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -1596,7 +1596,10 @@ bool perform_link_training_with_retries(
 				 * Report and continue with eDP panel mode to
 				 * perform eDP link training with right settings
 				 */
-				cp_psp->funcs.enable_assr(cp_psp->handle, link);
+				bool result;
+				result = cp_psp->funcs.enable_assr(cp_psp->handle, link);
+				if (!result && link->panel_mode != DP_PANEL_MODE_EDP)
+					panel_mode = DP_PANEL_MODE_DEFAULT;
 			}
 		}
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index 5ab2de12ccf87..2039a345f23a1 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -83,6 +83,7 @@ void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode)
 			ASSERT(result == DC_OK);
 		}
 	}
+	link->panel_mode = panel_mode;
 	DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d "
 		 "eDP panel mode enabled: %d \n",
 		 link->link_index,
-- 
GitLab


From a6c0c9f56197fcb3418be82a7d9f9952be1b5598 Mon Sep 17 00:00:00 2001
From: Jingwen Zhu <Jingwen.Zhu@amd.com>
Date: Thu, 30 Mar 2023 16:38:59 +0800
Subject: [PATCH 0119/1791] drm/amd/display: Improvement for handling edp link
 training fails
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
The eDP retrain will cause the DPCD 300 to be reset to default.
And cause the brightness can't be set correctly.

[How]
delete the call to edp panel power control in both
enable_link_output/disable_link_output entirely and
only call edp panel control in enable_link_dp and 
disable_link_dp once.

Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Jingwen Zhu <Jingwen.Zhu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dce110/dce110_hw_sequencer.c   | 19 +++++++++++--------
 .../gpu/drm/amd/display/dc/link/link_dpms.c   |  5 +++++
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 9fe0ce91db002..8d2460d06bced 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -3031,10 +3031,12 @@ void dce110_enable_dp_link_output(
 	const struct link_hwss *link_hwss = get_link_hwss(link, link_res);
 	unsigned int i;
 
-
+	/*
+	 * Add the logic to extract BOTH power up and power down sequences
+	 * from enable/disable link output and only call edp panel control
+	 * in enable_link_dp and disable_link_dp once.
+	 */
 	if (link->connector_signal == SIGNAL_TYPE_EDP) {
-		if (!link->dc->config.edp_no_power_sequencing)
-			link->dc->hwss.edp_power_control(link, true);
 		link->dc->hwss.edp_wait_for_hpd_ready(link, true);
 	}
 
@@ -3096,11 +3098,12 @@ void dce110_disable_link_output(struct dc_link *link,
 
 	link_hwss->disable_link_output(link, link_res, signal);
 	link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF;
-
-	if (signal == SIGNAL_TYPE_EDP &&
-			link->dc->hwss.edp_backlight_control)
-		link->dc->hwss.edp_power_control(link, false);
-	else if (dmcu != NULL && dmcu->funcs->lock_phy)
+	/*
+	 * Add the logic to extract BOTH power up and power down sequences
+	 * from enable/disable link output and only call edp panel control
+	 * in enable_link_dp and disable_link_dp once.
+	 */
+	if (dmcu != NULL && dmcu->funcs->lock_phy)
 		dmcu->funcs->unlock_phy(dmcu);
 	dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 027ad1f0144d8..2267fb0978305 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -1927,6 +1927,11 @@ static void disable_link_dp(struct dc_link *link,
 
 	dp_disable_link_phy(link, link_res, signal);
 
+	if (link->connector_signal == SIGNAL_TYPE_EDP) {
+		if (!link->dc->config.edp_no_power_sequencing)
+			link->dc->hwss.edp_power_control(link, false);
+	}
+
 	if (signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
 		/* set the sink to SST mode after disabling the link */
 		enable_mst_on_sink(link, false);
-- 
GitLab


From 7ab269d54813e03eec8a5acc415b2aef55aaf916 Mon Sep 17 00:00:00 2001
From: Daniel Miess <Daniel.Miess@amd.com>
Date: Tue, 4 Apr 2023 14:04:11 -0400
Subject: [PATCH 0120/1791] drm/amd/display: limit timing for single dimm
 memory

[Why]
1. It could hit bandwidth limitdation under single dimm
memory when connecting 8K external monitor.
2. IsSupportedVidPn got validation failed with
2K240Hz eDP + 8K24Hz external monitor.
3. It's better to filter out such combination in
EnumVidPnCofuncModality
4. For short term, filter out in dc bandwidth validation.

[How]
Force 2K@240Hz+8K@24Hz timing validation false in dc.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Daniel Miess <Daniel.Miess@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn314/dcn314_resource.c   | 20 +++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 50ed7e09d5ba4..24806acc84385 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -1696,6 +1696,23 @@ static void dcn314_get_panel_config_defaults(struct dc_panel_config *panel_confi
 	*panel_config = panel_config_defaults;
 }
 
+static bool filter_modes_for_single_channel_workaround(struct dc *dc,
+		struct dc_state *context)
+{
+	// Filter 2K@240Hz+8K@24fps above combination timing if memory only has single dimm LPDDR
+	if (dc->clk_mgr->bw_params->vram_type == 34 && dc->clk_mgr->bw_params->num_channels < 2) {
+		int total_phy_pix_clk = 0;
+
+		for (int i = 0; i < context->stream_count; i++)
+			if (context->res_ctx.pipe_ctx[i].stream)
+				total_phy_pix_clk += context->res_ctx.pipe_ctx[i].stream->phy_pix_clk;
+
+		if (total_phy_pix_clk >= (1148928+826260)) //2K@240Hz+8K@24fps
+			return true;
+	}
+	return false;
+}
+
 bool dcn314_validate_bandwidth(struct dc *dc,
 		struct dc_state *context,
 		bool fast_validate)
@@ -1711,6 +1728,9 @@ bool dcn314_validate_bandwidth(struct dc *dc,
 
 	BW_VAL_TRACE_COUNT();
 
+	if (filter_modes_for_single_channel_workaround(dc, context))
+		goto validate_fail;
+
 	DC_FP_START();
 	// do not support self refresh only
 	out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, false);
-- 
GitLab


From bddb55ccbdc20dd7fd526c8dfb13f695637bf7b1 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Date: Mon, 3 Apr 2023 10:13:12 -0400
Subject: [PATCH 0121/1791] drm/amd/display: set dcn315 lb bpp to 48

[Why & How]
Fix a typo for dcn315 line buffer bpp.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
index b37d14369a622..59836570603ac 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
@@ -222,7 +222,7 @@ struct _vcs_dpi_ip_params_st dcn3_15_ip = {
 	.maximum_dsc_bits_per_component = 10,
 	.dsc422_native_support = false,
 	.is_line_buffer_bpp_fixed = true,
-	.line_buffer_fixed_bpp = 49,
+	.line_buffer_fixed_bpp = 48,
 	.line_buffer_size_bits = 789504,
 	.max_line_buffer_lines = 12,
 	.writeback_interface_buffer_size_kbytes = 90,
-- 
GitLab


From 499e4b1c722e0e2ca40c56342b766e95f6c31f4a Mon Sep 17 00:00:00 2001
From: Eric Yang <Eric.Yang2@amd.com>
Date: Fri, 29 May 2020 17:13:57 -0400
Subject: [PATCH 0122/1791] drm/amd/display: add mechanism to skip DCN init

[Why]
If optimized init is done in FW. DCN init can be skipped in driver. This
need to be communicated between driver and fw and maintain backwards
compatibility.

[How]
Use DMUB scratch 0 bit 2 to indicate optimized init done in fw and
use DMUB scatch 4 bit 0 to indicate drive supports the optimized flow
so FW will perform it.

Signed-off-by: Eric Yang <Eric.Yang2@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      |  3 --
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c  | 28 +++++--------------
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h  |  2 +-
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c |  4 ++-
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h   |  1 -
 .../gpu/drm/amd/display/dmub/src/dmub_dcn21.c |  6 ----
 .../gpu/drm/amd/display/dmub/src/dmub_dcn21.h |  4 ---
 .../gpu/drm/amd/display/dmub/src/dmub_srv.c   | 25 +----------------
 8 files changed, 12 insertions(+), 61 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 105f705bd91a8..631c6b10562e1 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -4267,9 +4267,6 @@ void dc_set_power_state(
 
 		dc_z10_restore(dc);
 
-		if (dc->ctx->dmub_srv)
-			dc_dmub_srv_wait_phy_init(dc->ctx->dmub_srv);
-
 		dc->hwss.init_hw(dc);
 
 		if (dc->hwss.init_sys_ctx != NULL &&
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index eef43577508cf..d15ec32243e22 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -179,31 +179,17 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun
 	return true;
 }
 
-void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv)
+bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv)
 {
-	struct dmub_srv *dmub = dc_dmub_srv->dmub;
-	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
-	enum dmub_status status;
-
-	for (;;) {
-		/* Wait up to a second for PHY init. */
-		status = dmub_srv_wait_for_phy_init(dmub, 1000000);
-		if (status == DMUB_STATUS_OK)
-			/* Initialization OK */
-			break;
+	struct dmub_srv *dmub;
+	union dmub_fw_boot_status status;
 
-		DC_ERROR("DMCUB PHY init failed: status=%d\n", status);
-		ASSERT(0);
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
 
-		if (status != DMUB_STATUS_TIMEOUT)
-			/*
-			 * Server likely initialized or we don't have
-			 * DMCUB HW support - this won't end.
-			 */
-			break;
+	dmub = dc_dmub_srv->dmub;
 
-		/* Continue spinning so we don't hang the ASIC. */
-	}
+	return status.bits.optimized_init_done;
 }
 
 bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index 22f7b2704c8ec..a5196a9292b35 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -54,7 +54,7 @@ struct dc_dmub_srv {
 
 void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv);
 
-void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv);
+bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv);
 
 bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 1c3b6f25a7825..a7ad1d7bc43e5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -1524,7 +1524,9 @@ void dcn10_init_hw(struct dc *dc)
 	if (!dcb->funcs->is_accelerated_mode(dcb))
 		hws->funcs.disable_vga(dc->hwseq);
 
-	hws->funcs.bios_golden_init(dc);
+	if (!dc_dmub_srv_optimized_init_done(dc->ctx->dmub_srv))
+		hws->funcs.bios_golden_init(dc);
+
 
 	if (dc->ctx->dc_bios->fw_info_valid) {
 		res_pool->ref_clocks.xtalin_clock_inKhz =
diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index ba1715e2d25a9..719bf9bb168ad 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -366,7 +366,6 @@ struct dmub_srv_hw_funcs {
 
 	bool (*is_hw_init)(struct dmub_srv *dmub);
 
-	bool (*is_phy_init)(struct dmub_srv *dmub);
 	void (*enable_dmub_boot_options)(struct dmub_srv *dmub,
 				const struct dmub_srv_hw_params *params);
 
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c
index 51bb9bceb1b10..2d212bc974cca 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c
@@ -54,9 +54,3 @@ const struct dmub_srv_common_regs dmub_srv_dcn21_regs = {
 #undef DMUB_SF
 };
 
-/* Shared functions. */
-
-bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub)
-{
-	return REG_READ(DMCUB_SCRATCH10) == 0;
-}
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h
index 6fd5b0cd4ef34..8c4033ae4007d 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h
@@ -32,8 +32,4 @@
 
 extern const struct dmub_srv_common_regs dmub_srv_dcn21_regs;
 
-/* Hardware functions. */
-
-bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub);
-
 #endif /* _DMUB_DCN21_H_ */
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index 92c18bfb98b3b..67c53f7e589c6 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -190,11 +190,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 
 		funcs->get_diagnostic_data = dmub_dcn20_get_diagnostic_data;
 
-		if (asic == DMUB_ASIC_DCN21) {
+		if (asic == DMUB_ASIC_DCN21)
 			dmub->regs = &dmub_srv_dcn21_regs;
 
-			funcs->is_phy_init = dmub_dcn21_is_phy_init;
-		}
 		if (asic == DMUB_ASIC_DCN30) {
 			dmub->regs = &dmub_srv_dcn30_regs;
 
@@ -721,27 +719,6 @@ enum dmub_status dmub_srv_wait_for_auto_load(struct dmub_srv *dmub,
 	return DMUB_STATUS_TIMEOUT;
 }
 
-enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub,
-					    uint32_t timeout_us)
-{
-	uint32_t i = 0;
-
-	if (!dmub->hw_init)
-		return DMUB_STATUS_INVALID;
-
-	if (!dmub->hw_funcs.is_phy_init)
-		return DMUB_STATUS_OK;
-
-	for (i = 0; i <= timeout_us; i += 10) {
-		if (dmub->hw_funcs.is_phy_init(dmub))
-			return DMUB_STATUS_OK;
-
-		udelay(10);
-	}
-
-	return DMUB_STATUS_TIMEOUT;
-}
-
 enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub,
 					uint32_t timeout_us)
 {
-- 
GitLab


From c18842a2e81a3e3d5c7401f061d7887b422aeebc Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Sun, 29 May 2022 10:54:30 -0400
Subject: [PATCH 0123/1791] drm/amd/display: Return error code on DSC atomic
 check failure

[Why&How]
We were not returning -EINVAL on DSC atomic check fail. Add it.

Fixes: 71be4b16d39a ("drm/amd/display: dsc validate fail not pass to atomic check")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c           | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 6e9d2e680d502..73b52bf76461f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -10170,6 +10170,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
 		ret = compute_mst_dsc_configs_for_state(state, dm_state->context, vars);
 		if (ret) {
 			DRM_DEBUG_DRIVER("compute_mst_dsc_configs_for_state() failed\n");
+			ret = -EINVAL;
 			goto fail;
 		}
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 5dc79b753d5fb..810ab682f424f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -1410,6 +1410,7 @@ int pre_validate_dsc(struct drm_atomic_state *state,
 	ret = pre_compute_mst_dsc_configs_for_state(state, local_dc_state, vars);
 	if (ret != 0) {
 		DRM_INFO_ONCE("pre_compute_mst_dsc_configs_for_state() failed\n");
+		ret = -EINVAL;
 		goto clean_exit;
 	}
 
-- 
GitLab


From 08c73e896836be1a1875c612c25a9ad81893ff98 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Wed, 5 Apr 2023 16:17:42 -0400
Subject: [PATCH 0124/1791] drm/amd/display: remove incorrect early return

[Why&How]
Remove incorrect early return in a device specific fifo reset workaround

Reviewed-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 330ab036c830f..c6ce2b7123b79 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -687,7 +687,6 @@ static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux)
 		return;
 
 	data[0] |= (1 << 1); // set bit 1 to 1
-		return;
 
 	if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x221198, data))
 		return;
-- 
GitLab


From 2a66c0c9d20238812172693b5bef28c6c659eff4 Mon Sep 17 00:00:00 2001
From: Mikita Lipski <mikita.lipski@amd.com>
Date: Thu, 17 May 2018 15:44:20 -0400
Subject: [PATCH 0125/1791] drm/amd/display: Write TEST_EDID_CHECKSUM_WRITE for
 EDID tests

Extract edid's checksum and send it back for verification if EDID_TEST
is requested.

Signed-off-by: Mikita Lipski <mikita.lipski@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 30 +++++++++++++++++--
 1 file changed, 27 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index c6ce2b7123b79..09e056a647087 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -885,10 +885,34 @@ enum dc_edid_status dm_helpers_read_local_edid(
 		DRM_ERROR("EDID err: %d, on connector: %s",
 				edid_status,
 				aconnector->base.name);
+	if (link->aux_mode) {
+		union test_request test_request = {0};
+		union test_response test_response = {0};
 
-	/* DP Compliance Test 4.2.2.3 */
-	if (link->aux_mode)
-		drm_dp_send_real_edid_checksum(&aconnector->dm_dp_aux.aux, sink->dc_edid.raw_edid[sink->dc_edid.length-1]);
+		dm_helpers_dp_read_dpcd(ctx,
+					link,
+					DP_TEST_REQUEST,
+					&test_request.raw,
+					sizeof(union test_request));
+
+		if (!test_request.bits.EDID_READ)
+			return edid_status;
+
+		test_response.bits.EDID_CHECKSUM_WRITE = 1;
+
+		dm_helpers_dp_write_dpcd(ctx,
+					link,
+					DP_TEST_EDID_CHECKSUM,
+					&sink->dc_edid.raw_edid[sink->dc_edid.length-1],
+					1);
+
+		dm_helpers_dp_write_dpcd(ctx,
+					link,
+					DP_TEST_RESPONSE,
+					&test_response.raw,
+					sizeof(test_response));
+
+	}
 
 	return edid_status;
 }
-- 
GitLab


From 0c316556d124916e1dc2be171b3414b764972802 Mon Sep 17 00:00:00 2001
From: Tianci Yin <tianci.yin@amd.com>
Date: Mon, 6 Feb 2023 15:58:46 +0800
Subject: [PATCH 0126/1791] drm/amd/display: Disable migration to ensure
 consistency of per-CPU variable

[why]
Since the variable fpu_recursion_depth is per-CPU type, it has one copy
on each CPU, thread migration causes data consistency issue, then the
call trace shows up. And preemption disabling can't prevent migration.

[how]
Disable migration to ensure consistency of fpu_recursion_depth.

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Tianci Yin <tianci.yin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
index 1743ca0a36419..c42aa947c969d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
@@ -89,6 +89,7 @@ void dc_fpu_begin(const char *function_name, const int line)
 
 	if (*pcpu == 1) {
 #if defined(CONFIG_X86)
+		migrate_disable();
 		kernel_fpu_begin();
 #elif defined(CONFIG_PPC64)
 		if (cpu_has_feature(CPU_FTR_VSX_COMP)) {
@@ -129,6 +130,7 @@ void dc_fpu_end(const char *function_name, const int line)
 	if (*pcpu <= 0) {
 #if defined(CONFIG_X86)
 		kernel_fpu_end();
+		migrate_enable();
 #elif defined(CONFIG_PPC64)
 		if (cpu_has_feature(CPU_FTR_VSX_COMP)) {
 			disable_kernel_vsx();
-- 
GitLab


From cd8f067a46d34dee3188da184912ae3d64d98444 Mon Sep 17 00:00:00 2001
From: Wesley Chalmers <Wesley.Chalmers@amd.com>
Date: Wed, 10 Jun 2020 11:49:16 -0400
Subject: [PATCH 0127/1791] drm/amd/display: Add logging for display MALL
 refresh setting

[WHY]
Add log entry for when display refresh from MALL
settings are sent to SMU.

Fixes: 1664641ea946 ("drm/amd/display: Add logger for SMU msg")
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c   | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
index 1fbf1c105dc12..bdbf183066981 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
@@ -312,6 +312,9 @@ void dcn30_smu_set_display_refresh_from_mall(struct clk_mgr_internal *clk_mgr, b
 	/* bits 8:7 for cache timer scale, bits 6:1 for cache timer delay, bit 0 = 1 for enable, = 0 for disable */
 	uint32_t param = (cache_timer_scale << 7) | (cache_timer_delay << 1) | (enable ? 1 : 0);
 
+	smu_print("SMU Set display refresh from mall: enable = %d, cache_timer_delay = %d, cache_timer_scale = %d\n",
+		enable, cache_timer_delay, cache_timer_scale);
+
 	dcn30_smu_send_msg_with_param(clk_mgr,
 			DALSMC_MSG_SetDisplayRefreshFromMall, param, NULL);
 }
-- 
GitLab


From 3306ba4b60b2f3d9ac6bddc587a4d702e1ba2224 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 18 Apr 2023 00:07:18 +0200
Subject: [PATCH 0128/1791] drm/amd/display: fix is_timing_changed() prototype

Three functions in the amdgpu display driver cause -Wmissing-prototype
warnings:

drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_resource.c:1858:6: error: no previous prototype for 'is_timing_changed' [-Werror=missing-prototypes]

is_timing_changed() is actually meant to be a global symbol, but needs
a proper name and prototype.

Fixes: 17ce8a6907f7 ("drm/amd/display: Add dsc pre-validation in atomic check")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 5 ++---
 drivers/gpu/drm/amd/display/dc/core/dc_resource.c           | 6 +++---
 drivers/gpu/drm/amd/display/dc/dc.h                         | 3 +++
 3 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 810ab682f424f..46d0a8f57e552 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -45,8 +45,7 @@
 #endif
 
 #include "dc/dcn20/dcn20_resource.h"
-bool is_timing_changed(struct dc_stream_state *cur_stream,
-		       struct dc_stream_state *new_stream);
+
 #define PEAK_FACTOR_X1000 1006
 
 static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux,
@@ -1422,7 +1421,7 @@ int pre_validate_dsc(struct drm_atomic_state *state,
 		struct dc_stream_state *stream = dm_state->context->streams[i];
 
 		if (local_dc_state->streams[i] &&
-		    is_timing_changed(stream, local_dc_state->streams[i])) {
+		    dc_is_timing_changed(stream, local_dc_state->streams[i])) {
 			DRM_INFO_ONCE("crtc[%d] needs mode_changed\n", i);
 		} else {
 			int ind = find_crtc_index_in_state_by_stream(state, stream);
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 117d80cb36fbb..2f704e26219c6 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -1858,7 +1858,7 @@ bool dc_add_all_planes_for_stream(
 	return add_all_planes_for_stream(dc, stream, &set, 1, context);
 }
 
-bool is_timing_changed(struct dc_stream_state *cur_stream,
+bool dc_is_timing_changed(struct dc_stream_state *cur_stream,
 		       struct dc_stream_state *new_stream)
 {
 	if (cur_stream == NULL)
@@ -1883,7 +1883,7 @@ static bool are_stream_backends_same(
 	if (stream_a == NULL || stream_b == NULL)
 		return false;
 
-	if (is_timing_changed(stream_a, stream_b))
+	if (dc_is_timing_changed(stream_a, stream_b))
 		return false;
 
 	if (stream_a->signal != stream_b->signal)
@@ -3508,7 +3508,7 @@ bool pipe_need_reprogram(
 	if (pipe_ctx_old->stream_res.stream_enc != pipe_ctx->stream_res.stream_enc)
 		return true;
 
-	if (is_timing_changed(pipe_ctx_old->stream, pipe_ctx->stream))
+	if (dc_is_timing_changed(pipe_ctx_old->stream, pipe_ctx->stream))
 		return true;
 
 	if (pipe_ctx_old->stream->dpms_off != pipe_ctx->stream->dpms_off)
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 3b53f36029d06..edef5d181590f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -2228,4 +2228,7 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc,
 /* Disable acc mode Interfaces */
 void dc_disable_accelerated_mode(struct dc *dc);
 
+bool dc_is_timing_changed(struct dc_stream_state *cur_stream,
+		       struct dc_stream_state *new_stream);
+
 #endif /* DC_INTERFACE_H_ */
-- 
GitLab


From 128c1ca0303fe764a4cde5f761e72810d9e40b6e Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Fri, 29 Apr 2022 20:41:10 -0400
Subject: [PATCH 0129/1791] drm/amd/display: Update DTBCLK for DCN32
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why&How]
- Implement interface to program DTBCLK DTO’s
  according to reference DTBCLK returned by PMFW
- This is required because DTO programming
  requires exact DTBCLK reference freq or it could
  result in underflow

Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index ea753f8fa175d..2b8a81b6d53b1 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -233,6 +233,32 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 	DC_FP_END();
 }
 
+static void dcn32_update_clocks_update_dtb_dto(struct clk_mgr_internal *clk_mgr,
+			struct dc_state *context,
+			int ref_dtbclk_khz)
+{
+	struct dccg *dccg = clk_mgr->dccg;
+	uint32_t tg_mask = 0;
+	int i;
+
+	for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+		struct dtbclk_dto_params dto_params = {0};
+
+		/* use mask to program DTO once per tg */
+		if (pipe_ctx->stream_res.tg &&
+				!(tg_mask & (1 << pipe_ctx->stream_res.tg->inst))) {
+			tg_mask |= (1 << pipe_ctx->stream_res.tg->inst);
+
+			dto_params.otg_inst = pipe_ctx->stream_res.tg->inst;
+			dto_params.ref_dtbclk_khz = ref_dtbclk_khz;
+
+			dccg->funcs->set_dtbclk_dto(clk_mgr->dccg, &dto_params);
+			//dccg->funcs->set_audio_dtbclk_dto(clk_mgr->dccg, &dto_params);
+		}
+	}
+}
+
 /* Since DPPCLK request to PMFW needs to be exact (due to DPP DTO programming),
  * update DPPCLK to be the exact frequency that will be set after the DPPCLK
  * divider is updated. This will prevent rounding issues that could cause DPP
@@ -570,6 +596,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 		/* DCCG requires KHz precision for DTBCLK */
 		clk_mgr_base->clks.ref_dtbclk_khz =
 				dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DTBCLK, khz_to_mhz_ceil(new_clocks->ref_dtbclk_khz));
+		dcn32_update_clocks_update_dtb_dto(clk_mgr, context, clk_mgr_base->clks.ref_dtbclk_khz);
 	}
 
 	if (dc->config.forced_clocks == false || (force_reset && safe_to_lower)) {
-- 
GitLab


From 97041ed37718dc9ba30aa23ca74093dc93ac89fb Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Thu, 13 Apr 2023 16:22:51 +0200
Subject: [PATCH 0130/1791] drm/amdgpu: Increase GFX6 graphics ring size.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To ensure it supports 192 IBs per submission, so we can keep a
simplified IB limit in the follow up patch without having to
look at IP or GPU version.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index c41219e23151c..d9ce4d1c50e49 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -3073,7 +3073,7 @@ static int gfx_v6_0_sw_init(void *handle)
 		ring = &adev->gfx.gfx_ring[i];
 		ring->ring_obj = NULL;
 		sprintf(ring->name, "gfx");
-		r = amdgpu_ring_init(adev, ring, 1024,
+		r = amdgpu_ring_init(adev, ring, 2048,
 				     &adev->gfx.eop_irq,
 				     AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP,
 				     AMDGPU_RING_PRIO_DEFAULT, NULL);
-- 
GitLab


From c30ddcece3a0a86853862a7d92678a79525ca1fb Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Thu, 13 Apr 2023 16:22:52 +0200
Subject: [PATCH 0131/1791] drm/amdgpu: Add a max ibs per submission limit.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

And ensure each ring supports that many submissions. This makes
sure that we don't get surprises after the submission has been
scheduled where the ring allocation actually gets rejected.

My calculations on the existing limits:
COMPUTE v10: 128
COMPUTE v11: 128
COMPUTE v6: 157
COMPUTE v7: 133
COMPUTE v8: 130
COMPUTE v9: 125
GFX v10: 208
GFX v11: 213
GFX v6: 154 (doubling this in the previous patch)
GFX v7: 226
GFX v8: 213
GFX v9: 208
GFX v9 (SW): 208
SDMA CIK: 87
SDMA SI: 97
SDMA v2.4: 74
SDMA v3.0: 74
SDMA v4.0: 72
SDMA v5.0: 51
SDMA v6.0: 52
UVD ENC v6.0: 98
UVD ENC v7.0: 92
UVD v3.1: 124
UVD v4.2: 124
UVD v5.0: 83
UVD v6.0  (VM): 55
UVD v7.0: 51
VCE v2.0: 126
VCE v3.0 (VM): 98
VCE v4.0: 93
VCN DEC v1.0: 49
VCN DEC v2.0: 51
VCN DEC v3.0: 51
VCN ENC v1.0: 58
VCN ENC v2.0: 93
VCN ENC v3.0: 93
VCN ENC v4.0: 93
VCN JPEG v1.0: 17
VCN JPEG v2.0: 16
VCN JPEG v2.5: 17
VCN JPEG v3.0: 17
VCN JPEG v4.0: 17

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2498
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c   |  3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 29 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h |  1 +
 3 files changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 08eced097bd8e..d8b3c9198d332 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -112,6 +112,9 @@ static int amdgpu_cs_p1_ib(struct amdgpu_cs_parser *p,
 	if (r < 0)
 		return r;
 
+	if (num_ibs[r] >= amdgpu_ring_max_ibs(chunk_ib->ip_type))
+		return -EINVAL;
+
 	++(num_ibs[r]);
 	p->gang_leader_idx = r;
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index dc474b8096040..f676c236b657b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -49,6 +49,26 @@
  * them until the pointers are equal again.
  */
 
+/**
+ * amdgpu_ring_max_ibs - Return max IBs that fit in a single submission.
+ *
+ * @type: ring type for which to return the limit.
+ */
+unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type)
+{
+	switch (type) {
+	case AMDGPU_RING_TYPE_GFX:
+		/* Need to keep at least 192 on GFX7+ for old radv. */
+		return 192;
+	case AMDGPU_RING_TYPE_COMPUTE:
+		return 125;
+	case AMDGPU_RING_TYPE_VCN_JPEG:
+		return 16;
+	default:
+		return 49;
+	}
+}
+
 /**
  * amdgpu_ring_alloc - allocate space on the ring buffer
  *
@@ -182,6 +202,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	int sched_hw_submission = amdgpu_sched_hw_submission;
 	u32 *num_sched;
 	u32 hw_ip;
+	unsigned int max_ibs_dw;
 
 	/* Set the hw submission limit higher for KIQ because
 	 * it's used for a number of gfx/compute tasks by both
@@ -290,6 +311,14 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		return r;
 	}
 
+	max_ibs_dw = ring->funcs->emit_frame_size +
+		     amdgpu_ring_max_ibs(ring->funcs->type) * ring->funcs->emit_ib_size;
+	max_ibs_dw = (max_ibs_dw + ring->funcs->align_mask) & ~ring->funcs->align_mask;
+
+	if (WARN_ON(max_ibs_dw > max_dw)) {
+		max_dw = max_ibs_dw;
+	}
+
 	ring->ring_size = roundup_pow_of_two(max_dw * 4 * sched_hw_submission);
 
 	ring->buf_mask = (ring->ring_size / 4) - 1;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index d8749444b6891..8eca6532ed196 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -319,6 +319,7 @@ struct amdgpu_ring {
 #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))
 #define amdgpu_ring_preempt_ib(r) (r)->funcs->preempt_ib(r)
 
+unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type);
 int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw);
 void amdgpu_ring_ib_begin(struct amdgpu_ring *ring);
 void amdgpu_ring_ib_end(struct amdgpu_ring *ring);
-- 
GitLab


From 4f18b9a6711adbc7c76993c734a94ee3f5c61791 Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Thu, 13 Apr 2023 16:22:53 +0200
Subject: [PATCH 0132/1791] drm/amdgpu: Add support for querying the max ibs in
 a submission. (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This info would be used by radv to figure out when we need to
split a submission into multiple submissions. radv currently has
a limit of 192 which seems to work for most gfx submissions, but
is way too high for e.g. compute or sdma.

Userspace is available at
https://gitlab.freedesktop.org/bnieuwenhuizen/mesa/-/commits/ib-rejection-v3

v3: Completely rewrote based on suggestion of making it a separate query.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2498
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 9 +++++++++
 include/uapi/drm/amdgpu_drm.h           | 2 ++
 2 files changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 0efb38539d70c..1a2e342af1c09 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1140,6 +1140,15 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		kfree(caps);
 		return r;
 	}
+	case AMDGPU_INFO_MAX_IBS: {
+		uint32_t max_ibs[AMDGPU_HW_IP_NUM];
+
+		for (i = 0; i < AMDGPU_HW_IP_NUM; ++i)
+			max_ibs[i] = amdgpu_ring_max_ibs(i);
+
+		return copy_to_user(out, max_ibs,
+				    min((size_t)size, sizeof(max_ibs))) ? -EFAULT : 0;
+	}
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->query);
 		return -EINVAL;
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index b6eb90df5d052..6981e59a94017 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -876,6 +876,8 @@ struct drm_amdgpu_cs_chunk_data {
 	#define AMDGPU_INFO_VIDEO_CAPS_DECODE		0
 	/* Subquery id: Encode */
 	#define AMDGPU_INFO_VIDEO_CAPS_ENCODE		1
+/* Query the max number of IBs per gang per submission */
+#define AMDGPU_INFO_MAX_IBS			0x22
 
 #define AMDGPU_INFO_MMR_SE_INDEX_SHIFT	0
 #define AMDGPU_INFO_MMR_SE_INDEX_MASK	0xff
-- 
GitLab


From 7f102a906681cddb8ababe53e0caa40a17f4cd11 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Mon, 17 Apr 2023 18:42:37 +0100
Subject: [PATCH 0133/1791] drm/amd/pm: Fix spelling mistake "aquire" ->
 "acquire"

There is a spelling mistake in the smu_i2c_bus_access prototype. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h
index 5ce433e2c16a5..f1580a26a8506 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h
@@ -359,7 +359,7 @@ struct pp_hwmgr_func {
 	int (*set_ppfeature_status)(struct pp_hwmgr *hwmgr, uint64_t ppfeature_masks);
 	int (*set_mp1_state)(struct pp_hwmgr *hwmgr, enum pp_mp1_state mp1_state);
 	int (*asic_reset)(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mode);
-	int (*smu_i2c_bus_access)(struct pp_hwmgr *hwmgr, bool aquire);
+	int (*smu_i2c_bus_access)(struct pp_hwmgr *hwmgr, bool acquire);
 	int (*set_df_cstate)(struct pp_hwmgr *hwmgr, enum pp_df_cstate state);
 	int (*set_xgmi_pstate)(struct pp_hwmgr *hwmgr, uint32_t pstate);
 	int (*disable_power_features_for_compute_performance)(struct pp_hwmgr *hwmgr,
-- 
GitLab


From 1fa8d710573f02ae9118bc5f53e7ede09d6920da Mon Sep 17 00:00:00 2001
From: Alan Liu <HaoPing.Liu@amd.com>
Date: Fri, 14 Apr 2023 18:39:52 +0800
Subject: [PATCH 0134/1791] drm/amdgpu: Fix desktop freezed after gpu-reset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
After gpu-reset, sometimes the driver fails to enable vblank irq,
causing flip_done timed out and the desktop freezed.

During gpu-reset, we disable and enable vblank irq in dm_suspend() and
dm_resume(). Later on in amdgpu_irq_gpu_reset_resume_helper(), we check
irqs' refcount and decide to enable or disable the irqs again.

However, we have 2 sets of API for controling vblank irq, one is
dm_vblank_get/put() and another is amdgpu_irq_get/put(). Each API has
its own refcount and flag to store the state of vblank irq, and they
are not synchronized.

In drm we use the first API to control vblank irq but in
amdgpu_irq_gpu_reset_resume_helper() we use the second set of API.

The failure happens when vblank irq was enabled by dm_vblank_get()
before gpu-reset, we have vblank->enabled true. However, during
gpu-reset, in amdgpu_irq_gpu_reset_resume_helper() vblank irq's state
checked from amdgpu_irq_update() is DISABLED. So finally it disables
vblank irq again. After gpu-reset, if there is a cursor plane commit,
the driver will try to enable vblank irq by calling drm_vblank_enable(),
but the vblank->enabled is still true, so it fails to turn on vblank
irq and causes flip_done can't be completed in vblank irq handler and
desktop become freezed.

[How]
Combining the 2 vblank control APIs by letting drm's API finally calls
amdgpu_irq's API, so the irq's refcount and state of both APIs can be
synchronized. Also add a check to prevent refcount from being less then
0 in amdgpu_irq_put().

v2:
- Add warning in amdgpu_irq_enable() if the irq is already disabled.
- Call dc_interrupt_set() in dm_set_vblank() to avoid refcount change
  if it is in gpu-reset.

v3:
- Improve commit message and code comments.

Signed-off-by: Alan Liu <HaoPing.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c         |  3 +++
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c  | 17 ++++++++++++++---
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index d0a1cc88832cc..fafebec5b7b66 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -596,6 +596,9 @@ int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 	if (!src->enabled_types || !src->funcs->set)
 		return -EINVAL;
 
+	if (WARN_ON(!amdgpu_irq_enabled(adev, src, type)))
+		return -EINVAL;
+
 	if (atomic_dec_and_test(&src->enabled_types[type]))
 		return amdgpu_irq_update(adev, src, type);
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 1d924dc51a3eb..e3762e806617c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -169,10 +169,21 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
 	if (rc)
 		return rc;
 
-	irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
+	if (amdgpu_in_reset(adev)) {
+		irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
+		/* During gpu-reset we disable and then enable vblank irq, so
+		 * don't use amdgpu_irq_get/put() to avoid refcount change.
+		 */
+		if (!dc_interrupt_set(adev->dm.dc, irq_source, enable))
+			rc = -EBUSY;
+	} else {
+		rc = (enable)
+			? amdgpu_irq_get(adev, &adev->crtc_irq, acrtc->crtc_id)
+			: amdgpu_irq_put(adev, &adev->crtc_irq, acrtc->crtc_id);
+	}
 
-	if (!dc_interrupt_set(adev->dm.dc, irq_source, enable))
-		return -EBUSY;
+	if (rc)
+		return rc;
 
 skip:
 	if (amdgpu_in_reset(adev))
-- 
GitLab


From 94344e62a9ce8abcf681390f9822a7b075cf98e2 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Fri, 14 Apr 2023 11:08:30 -0400
Subject: [PATCH 0135/1791] drm/amd/display: remove unused variable
 oldest_index

cpp_check reports
drivers/gpu/drm/amd/display/modules/freesync/freesync.c:1143:17: style: Variable
  'oldest_index' is assigned a value that is never used. [unreadVariable]
   oldest_index = 0;
                ^

This variable is not used so remove.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/modules/freesync/freesync.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 5c41a4751db44..5798c0eafa1fb 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -1137,10 +1137,6 @@ void mod_freesync_handle_preflip(struct mod_freesync *mod_freesync,
 
 	if (in_out_vrr->supported &&
 			in_out_vrr->state == VRR_STATE_ACTIVE_VARIABLE) {
-		unsigned int oldest_index = plane->time.index + 1;
-
-		if (oldest_index >= DC_PLANE_UPDATE_TIMES_MAX)
-			oldest_index = 0;
 
 		last_render_time_in_us = curr_time_stamp_in_us -
 				plane->time.prev_update_time_in_us;
-- 
GitLab


From 20c3dffdccbd494e0dd631d1660aeecbff6775f2 Mon Sep 17 00:00:00 2001
From: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Date: Thu, 13 Apr 2023 08:12:28 -0700
Subject: [PATCH 0136/1791] radeon: avoid double free in ci_dpm_init()

Several calls to ci_dpm_fini() will attempt to free resources that
either have been freed before or haven't been allocated yet. This
may lead to undefined or dangerous behaviour.

For instance, if r600_parse_extended_power_table() fails, it might
call r600_free_extended_power_table() as will ci_dpm_fini() later
during error handling.

Fix this by only freeing pointers to objects previously allocated.

Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.

Fixes: cc8dbbb4f62a ("drm/radeon: add dpm support for CI dGPUs (v2)")
Co-developed-by: Natalia Petrova <n.petrova@fintech.ru>
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/ci_dpm.c | 28 ++++++++++++++++++++--------
 1 file changed, 20 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 8ef25ab305ae7..b8f4dac68d850 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -5517,6 +5517,7 @@ static int ci_parse_power_table(struct radeon_device *rdev)
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct ci_ps *ps;
+	int ret;
 
 	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
 				   &frev, &crev, &data_offset))
@@ -5546,11 +5547,15 @@ static int ci_parse_power_table(struct radeon_device *rdev)
 		non_clock_array_index = power_state->v2.nonClockInfoIndex;
 		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
 			&non_clock_info_array->nonClockInfo[non_clock_array_index];
-		if (!rdev->pm.power_state[i].clock_info)
-			return -EINVAL;
+		if (!rdev->pm.power_state[i].clock_info) {
+			ret = -EINVAL;
+			goto err_free_ps;
+		}
 		ps = kzalloc(sizeof(struct ci_ps), GFP_KERNEL);
-		if (ps == NULL)
-			return -ENOMEM;
+		if (ps == NULL) {
+			ret = -ENOMEM;
+			goto err_free_ps;
+		}
 		rdev->pm.dpm.ps[i].ps_priv = ps;
 		ci_parse_pplib_non_clock_info(rdev, &rdev->pm.dpm.ps[i],
 					      non_clock_info,
@@ -5590,6 +5595,12 @@ static int ci_parse_power_table(struct radeon_device *rdev)
 	}
 
 	return 0;
+
+err_free_ps:
+	for (i = 0; i < rdev->pm.dpm.num_ps; i++)
+		kfree(rdev->pm.dpm.ps[i].ps_priv);
+	kfree(rdev->pm.dpm.ps);
+	return ret;
 }
 
 static int ci_get_vbios_boot_values(struct radeon_device *rdev,
@@ -5678,25 +5689,26 @@ int ci_dpm_init(struct radeon_device *rdev)
 
 	ret = ci_get_vbios_boot_values(rdev, &pi->vbios_boot_state);
 	if (ret) {
-		ci_dpm_fini(rdev);
+		kfree(rdev->pm.dpm.priv);
 		return ret;
 	}
 
 	ret = r600_get_platform_caps(rdev);
 	if (ret) {
-		ci_dpm_fini(rdev);
+		kfree(rdev->pm.dpm.priv);
 		return ret;
 	}
 
 	ret = r600_parse_extended_power_table(rdev);
 	if (ret) {
-		ci_dpm_fini(rdev);
+		kfree(rdev->pm.dpm.priv);
 		return ret;
 	}
 
 	ret = ci_parse_power_table(rdev);
 	if (ret) {
-		ci_dpm_fini(rdev);
+		kfree(rdev->pm.dpm.priv);
+		r600_free_extended_power_table(rdev);
 		return ret;
 	}
 
-- 
GitLab


From 277bd3371f11400d5b02df54f057569be4b10cea Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 24 May 2022 10:51:43 +0800
Subject: [PATCH 0137/1791] drm/amdgpu: convert gfx.kiq to array type (v3)

v1: more kiq instances are a available in SOC (Le)
v2: squash commits to avoid breaking the build (Le)
v3: make the conversion for gfx/mec v11_0 (Hawking)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    |  6 ++--
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |  6 ++--
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    |  6 ++--
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c |  6 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c       | 34 +++++++++----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        | 32 ++++++++---------
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        | 26 +++++++-------
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c         | 16 ++++-----
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         | 24 ++++++-------
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c        | 12 +++----
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c        | 12 +++----
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c         | 12 +++----
 drivers/gpu/drm/amd/amdgpu/mes_v10_1.c        | 22 ++++++------
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c        | 26 +++++++-------
 16 files changed, 122 insertions(+), 122 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 9378fc79e9ea6..f599e1e74fcc2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -288,7 +288,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
 			    uint32_t doorbell_off)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v10_compute_mqd *m;
 	uint32_t mec, pipe;
 	int r;
@@ -303,7 +303,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
 		 mec, pipe, queue_id);
 
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, 7);
 	if (r) {
 		pr_err("Failed to alloc KIQ (%d).\n", r);
@@ -330,7 +330,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	amdgpu_ring_commit(kiq_ring);
 
 out_unlock:
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 	release_queue(adev);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index ba21ec6b35e0c..5c4152ae44da0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -275,7 +275,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
 			    uint32_t doorbell_off)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v10_compute_mqd *m;
 	uint32_t mec, pipe;
 	int r;
@@ -290,7 +290,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 	pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
 		 mec, pipe, queue_id);
 
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, 7);
 	if (r) {
 		pr_err("Failed to alloc KIQ (%d).\n", r);
@@ -317,7 +317,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 	amdgpu_ring_commit(kiq_ring);
 
 out_unlock:
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 	release_queue(adev);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 7e80caa05060b..5cdb7289d35b8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -260,7 +260,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd,
 			      uint32_t pipe_id, uint32_t queue_id,
 			      uint32_t doorbell_off)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v11_compute_mqd *m;
 	uint32_t mec, pipe;
 	int r;
@@ -275,7 +275,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd,
 	pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
 		 mec, pipe, queue_id);
 
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, 7);
 	if (r) {
 		pr_err("Failed to alloc KIQ (%d).\n", r);
@@ -302,7 +302,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd,
 	amdgpu_ring_commit(kiq_ring);
 
 out_unlock:
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 	release_queue(adev);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index e92b93557c13f..bc944ae4fd5b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -300,7 +300,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
 			    uint32_t doorbell_off)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v9_mqd *m;
 	uint32_t mec, pipe;
 	int r;
@@ -315,7 +315,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
 		 mec, pipe, queue_id);
 
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, 7);
 	if (r) {
 		pr_err("Failed to alloc KIQ (%d).\n", r);
@@ -342,7 +342,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	amdgpu_ring_commit(kiq_ring);
 
 out_unlock:
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 	release_queue(adev);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 9d3a0542c9967..9b6071df1fa7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -296,7 +296,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 			     struct amdgpu_ring *ring,
 			     struct amdgpu_irq_src *irq)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	int r = 0;
 
 	spin_lock_init(&kiq->ring_lock);
@@ -329,7 +329,7 @@ void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring)
 
 void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 
 	amdgpu_bo_free_kernel(&kiq->eop_obj, &kiq->eop_gpu_addr, NULL);
 }
@@ -339,7 +339,7 @@ int amdgpu_gfx_kiq_init(struct amdgpu_device *adev,
 {
 	int r;
 	u32 *hpd;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 
 	r = amdgpu_bo_create_kernel(adev, hpd_size, PAGE_SIZE,
 				    AMDGPU_GEM_DOMAIN_GTT, &kiq->eop_obj,
@@ -368,7 +368,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 	int r, i;
 
 	/* create MQD for KIQ */
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 	if (!adev->enable_mes_kiq && !ring->mqd_obj) {
 		/* originaly the KIQ MQD is put in GTT domain, but for SRIOV VRAM domain is a must
 		 * otherwise hypervisor trigger SAVE_VF fail after driver unloaded which mean MQD
@@ -458,7 +458,7 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev)
 				      &ring->mqd_ptr);
 	}
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 	kfree(adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS]);
 	amdgpu_bo_free_kernel(&ring->mqd_obj,
 			      &ring->mqd_gpu_addr,
@@ -467,17 +467,17 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev)
 
 int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	int i, r = 0;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues)
 		return -EINVAL;
 
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
 					adev->gfx.num_compute_rings)) {
-		spin_unlock(&adev->gfx.kiq.ring_lock);
+		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		return -ENOMEM;
 	}
 
@@ -485,9 +485,9 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev)
 		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i],
 					   RESET_QUEUES, 0, 0);
 
-	if (adev->gfx.kiq.ring.sched.ready && !adev->job_hang)
+	if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang)
 		r = amdgpu_ring_test_helper(kiq_ring);
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 
 	return r;
 }
@@ -507,8 +507,8 @@ int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev,
 
 int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	uint64_t queue_mask = 0;
 	int r, i;
 
@@ -532,13 +532,13 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 
 	DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe,
 							kiq_ring->queue);
-	spin_lock(&adev->gfx.kiq.ring_lock);
+	spin_lock(&adev->gfx.kiq[0].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
 					adev->gfx.num_compute_rings +
 					kiq->pmf->set_resources_size);
 	if (r) {
 		DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-		spin_unlock(&adev->gfx.kiq.ring_lock);
+		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		return r;
 	}
 
@@ -550,7 +550,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.compute_ring[i]);
 
 	r = amdgpu_ring_test_helper(kiq_ring);
-	spin_unlock(&adev->gfx.kiq.ring_lock);
+	spin_unlock(&adev->gfx.kiq[0].ring_lock);
 	if (r)
 		DRM_ERROR("KCQ enable failed\n");
 
@@ -788,7 +788,7 @@ uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
 	signed long r, cnt = 0;
 	unsigned long flags;
 	uint32_t seq, reg_val_offs = 0, value = 0;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *ring = &kiq->ring;
 
 	if (amdgpu_device_skip_hw_access(adev))
@@ -856,7 +856,7 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
 	signed long r, cnt = 0;
 	unsigned long flags;
 	uint32_t seq;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *ring = &kiq->ring;
 
 	BUG_ON(!ring->funcs->emit_wreg);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index bfabea76d1667..c742b4a369799 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -296,7 +296,7 @@ struct amdgpu_gfx {
 	struct amdgpu_ce		ce;
 	struct amdgpu_me		me;
 	struct amdgpu_mec		mec;
-	struct amdgpu_kiq		kiq;
+	struct amdgpu_kiq		kiq[AMDGPU_MAX_GC_INSTANCES];
 	struct amdgpu_imu		imu;
 	bool				rs64_enable; /* firmware format */
 	const struct firmware		*me_fw;	/* ME firmware */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index f2e2cbaa7fde0..9dd474262c291 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -74,7 +74,7 @@ void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
 					uint32_t reg0, uint32_t reg1,
 					uint32_t ref, uint32_t mask)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *ring = &kiq->ring;
 	signed long r, cnt = 0;
 	unsigned long flags;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index f5b5ce1051a28..d4e7de8fd9da1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -3568,7 +3568,7 @@ static void gfx10_kiq_unmap_queues(struct amdgpu_ring *kiq_ring,
 	struct amdgpu_device *adev = kiq_ring->adev;
 	uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0;
 
-	if (adev->enable_mes && !adev->gfx.kiq.ring.sched.ready) {
+	if (adev->enable_mes && !adev->gfx.kiq[0].ring.sched.ready) {
 		amdgpu_mes_unmap_legacy_queue(adev, ring, action, gpu_addr, seq);
 		return;
 	}
@@ -3636,7 +3636,7 @@ static const struct kiq_pm4_funcs gfx_v10_0_kiq_pm4_funcs = {
 
 static void gfx_v10_0_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 {
-	adev->gfx.kiq.pmf = &gfx_v10_0_kiq_pm4_funcs;
+	adev->gfx.kiq[0].pmf = &gfx_v10_0_kiq_pm4_funcs;
 }
 
 static void gfx_v10_0_init_spm_golden_registers(struct amdgpu_device *adev)
@@ -4550,7 +4550,7 @@ static int gfx_v10_0_sw_init(void *handle)
 	/* KIQ event */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP,
 			      GFX_10_1__SRCID__CP_IB2_INTERRUPT_PKT,
-			      &adev->gfx.kiq.irq);
+			      &adev->gfx.kiq[0].irq);
 	if (r)
 		return r;
 
@@ -4635,7 +4635,7 @@ static int gfx_v10_0_sw_init(void *handle)
 			return r;
 		}
 
-		kiq = &adev->gfx.kiq;
+		kiq = &adev->gfx.kiq[0];
 		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
 		if (r)
 			return r;
@@ -4693,7 +4693,7 @@ static int gfx_v10_0_sw_fini(void *handle)
 	amdgpu_gfx_mqd_sw_fini(adev);
 
 	if (!adev->enable_mes_kiq) {
-		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring);
+		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
 		amdgpu_gfx_kiq_fini(adev);
 	}
 
@@ -6214,7 +6214,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 				      CP_MEC_CNTL__MEC_ME2_HALT_MASK));
 			break;
 		}
-		adev->gfx.kiq.ring.sched.ready = false;
+		adev->gfx.kiq[0].ring.sched.ready = false;
 	}
 	udelay(50);
 }
@@ -6524,8 +6524,8 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 #ifndef BRING_UP_DEBUG
 static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	int r, i;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
@@ -6885,7 +6885,7 @@ static int gfx_v10_0_kiq_resume(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring;
 	int r;
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	r = amdgpu_bo_reserve(ring->mqd_obj, false);
 	if (unlikely(r != 0))
@@ -7243,7 +7243,7 @@ static int gfx_v10_0_hw_init(void *handle)
 #ifndef BRING_UP_DEBUG
 static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	int i;
 
@@ -8640,7 +8640,7 @@ static int gfx_v10_0_ring_preempt_ib(struct amdgpu_ring *ring)
 {
 	int i, r = 0;
 	struct amdgpu_device *adev = ring->adev;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	unsigned long flags;
 
@@ -9148,7 +9148,7 @@ static int gfx_v10_0_kiq_set_interrupt_state(struct amdgpu_device *adev,
 					     enum amdgpu_interrupt_state state)
 {
 	uint32_t tmp, target;
-	struct amdgpu_ring *ring = &(adev->gfx.kiq.ring);
+	struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring);
 
 	if (ring->me == 1)
 		target = SOC15_REG_OFFSET(GC, 0, mmCP_ME1_PIPE0_INT_CNTL);
@@ -9192,7 +9192,7 @@ static int gfx_v10_0_kiq_irq(struct amdgpu_device *adev,
 			     struct amdgpu_iv_entry *entry)
 {
 	u8 me_id, pipe_id, queue_id;
-	struct amdgpu_ring *ring = &(adev->gfx.kiq.ring);
+	struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring);
 
 	me_id = (entry->ring_id & 0x0c) >> 2;
 	pipe_id = (entry->ring_id & 0x03) >> 0;
@@ -9369,7 +9369,7 @@ static void gfx_v10_0_set_ring_funcs(struct amdgpu_device *adev)
 {
 	int i;
 
-	adev->gfx.kiq.ring.funcs = &gfx_v10_0_ring_funcs_kiq;
+	adev->gfx.kiq[0].ring.funcs = &gfx_v10_0_ring_funcs_kiq;
 
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		adev->gfx.gfx_ring[i].funcs = &gfx_v10_0_ring_funcs_gfx;
@@ -9403,8 +9403,8 @@ static void gfx_v10_0_set_irq_funcs(struct amdgpu_device *adev)
 	adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST;
 	adev->gfx.eop_irq.funcs = &gfx_v10_0_eop_irq_funcs;
 
-	adev->gfx.kiq.irq.num_types = AMDGPU_CP_KIQ_IRQ_LAST;
-	adev->gfx.kiq.irq.funcs = &gfx_v10_0_kiq_irq_funcs;
+	adev->gfx.kiq[0].irq.num_types = AMDGPU_CP_KIQ_IRQ_LAST;
+	adev->gfx.kiq[0].irq.funcs = &gfx_v10_0_kiq_irq_funcs;
 
 	adev->gfx.priv_reg_irq.num_types = 1;
 	adev->gfx.priv_reg_irq.funcs = &gfx_v10_0_priv_reg_irq_funcs;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index a9da0486467ac..6a5435255e6d7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -192,7 +192,7 @@ static void gfx11_kiq_unmap_queues(struct amdgpu_ring *kiq_ring,
 	struct amdgpu_device *adev = kiq_ring->adev;
 	uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0;
 
-	if (adev->enable_mes && !adev->gfx.kiq.ring.sched.ready) {
+	if (adev->enable_mes && !adev->gfx.kiq[0].ring.sched.ready) {
 		amdgpu_mes_unmap_legacy_queue(adev, ring, action, gpu_addr, seq);
 		return;
 	}
@@ -260,7 +260,7 @@ static const struct kiq_pm4_funcs gfx_v11_0_kiq_pm4_funcs = {
 
 static void gfx_v11_0_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 {
-	adev->gfx.kiq.pmf = &gfx_v11_0_kiq_pm4_funcs;
+	adev->gfx.kiq[0].pmf = &gfx_v11_0_kiq_pm4_funcs;
 }
 
 static void gfx_v11_0_init_golden_registers(struct amdgpu_device *adev)
@@ -1395,7 +1395,7 @@ static int gfx_v11_0_sw_init(void *handle)
 			return r;
 		}
 
-		kiq = &adev->gfx.kiq;
+		kiq = &adev->gfx.kiq[0];
 		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
 		if (r)
 			return r;
@@ -1466,7 +1466,7 @@ static int gfx_v11_0_sw_fini(void *handle)
 	amdgpu_gfx_mqd_sw_fini(adev);
 
 	if (!adev->enable_mes_kiq) {
-		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring);
+		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
 		amdgpu_gfx_kiq_fini(adev);
 	}
 
@@ -3337,7 +3337,7 @@ static void gfx_v11_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, data);
 	}
 
-	adev->gfx.kiq.ring.sched.ready = enable;
+	adev->gfx.kiq[0].ring.sched.ready = enable;
 
 	udelay(50);
 }
@@ -3732,8 +3732,8 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 #ifndef BRING_UP_DEBUG
 static int gfx_v11_0_kiq_enable_kgq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	int r, i;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
@@ -4108,7 +4108,7 @@ static int gfx_v11_0_kiq_resume(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring;
 	int r;
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	r = amdgpu_bo_reserve(ring->mqd_obj, false);
 	if (unlikely(r != 0))
@@ -4417,7 +4417,7 @@ static int gfx_v11_0_hw_init(void *handle)
 #ifndef BRING_UP_DEBUG
 static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	int i, r = 0;
 
@@ -4432,7 +4432,7 @@ static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev)
 		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i],
 					   PREEMPT_QUEUES, 0, 0);
 
-	if (adev->gfx.kiq.ring.sched.ready)
+	if (adev->gfx.kiq[0].ring.sched.ready)
 		r = amdgpu_ring_test_helper(kiq_ring);
 
 	return r;
@@ -5622,7 +5622,7 @@ static int gfx_v11_0_ring_preempt_ib(struct amdgpu_ring *ring)
 {
 	int i, r = 0;
 	struct amdgpu_device *adev = ring->adev;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	unsigned long flags;
 
@@ -6120,7 +6120,7 @@ static int gfx_v11_0_kiq_set_interrupt_state(struct amdgpu_device *adev,
 					     enum amdgpu_interrupt_state state)
 {
 	uint32_t tmp, target;
-	struct amdgpu_ring *ring = &(adev->gfx.kiq.ring);
+	struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring);
 
 	target = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE0_INT_CNTL);
 	target += ring->pipe;
@@ -6317,7 +6317,7 @@ static void gfx_v11_0_set_ring_funcs(struct amdgpu_device *adev)
 {
 	int i;
 
-	adev->gfx.kiq.ring.funcs = &gfx_v11_0_ring_funcs_kiq;
+	adev->gfx.kiq[0].ring.funcs = &gfx_v11_0_ring_funcs_kiq;
 
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		adev->gfx.gfx_ring[i].funcs = &gfx_v11_0_ring_funcs_gfx;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index b1f2684d854ad..ed04bad8543da 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -2021,7 +2021,7 @@ static int gfx_v8_0_sw_init(void *handle)
 		return r;
 	}
 
-	kiq = &adev->gfx.kiq;
+	kiq = &adev->gfx.kiq[0];
 	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
 	if (r)
 		return r;
@@ -2051,7 +2051,7 @@ static int gfx_v8_0_sw_fini(void *handle)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
 	amdgpu_gfx_mqd_sw_fini(adev);
-	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring);
+	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
 	amdgpu_gfx_kiq_fini(adev);
 
 	gfx_v8_0_mec_fini(adev);
@@ -4292,7 +4292,7 @@ static void gfx_v8_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 		WREG32(mmCP_MEC_CNTL, 0);
 	} else {
 		WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
-		adev->gfx.kiq.ring.sched.ready = false;
+		adev->gfx.kiq[0].ring.sched.ready = false;
 	}
 	udelay(50);
 }
@@ -4314,7 +4314,7 @@ static void gfx_v8_0_kiq_setting(struct amdgpu_ring *ring)
 
 static int gfx_v8_0_kiq_kcq_enable(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	uint64_t queue_mask = 0;
 	int r, i;
 
@@ -4678,7 +4678,7 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring;
 	int r;
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	r = amdgpu_bo_reserve(ring->mqd_obj, false);
 	if (unlikely(r != 0))
@@ -4741,7 +4741,7 @@ static int gfx_v8_0_cp_test_all_rings(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 	r = amdgpu_ring_test_helper(ring);
 	if (r)
 		return r;
@@ -4808,7 +4808,7 @@ static int gfx_v8_0_hw_init(void *handle)
 static int gfx_v8_0_kcq_disable(struct amdgpu_device *adev)
 {
 	int r, i;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 
 	r = amdgpu_ring_alloc(kiq_ring, 6 * adev->gfx.num_compute_rings);
 	if (r)
@@ -7001,7 +7001,7 @@ static void gfx_v8_0_set_ring_funcs(struct amdgpu_device *adev)
 {
 	int i;
 
-	adev->gfx.kiq.ring.funcs = &gfx_v8_0_ring_funcs_kiq;
+	adev->gfx.kiq[0].ring.funcs = &gfx_v8_0_ring_funcs_kiq;
 
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		adev->gfx.gfx_ring[i].funcs = &gfx_v8_0_ring_funcs_gfx;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index adbcd8127c82e..adf86bc7ed369 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -898,7 +898,7 @@ static const struct kiq_pm4_funcs gfx_v9_0_kiq_pm4_funcs = {
 
 static void gfx_v9_0_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 {
-	adev->gfx.kiq.pmf = &gfx_v9_0_kiq_pm4_funcs;
+	adev->gfx.kiq[0].pmf = &gfx_v9_0_kiq_pm4_funcs;
 }
 
 static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev)
@@ -2174,7 +2174,7 @@ static int gfx_v9_0_sw_init(void *handle)
 		return r;
 	}
 
-	kiq = &adev->gfx.kiq;
+	kiq = &adev->gfx.kiq[0];
 	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
 	if (r)
 		return r;
@@ -2216,7 +2216,7 @@ static int gfx_v9_0_sw_fini(void *handle)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
 	amdgpu_gfx_mqd_sw_fini(adev);
-	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring);
+	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
 	amdgpu_gfx_kiq_fini(adev);
 
 	gfx_v9_0_mec_fini(adev);
@@ -3155,7 +3155,7 @@ static void gfx_v9_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 	} else {
 		WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL,
 			(CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
-		adev->gfx.kiq.ring.sched.ready = false;
+		adev->gfx.kiq[0].ring.sched.ready = false;
 	}
 	udelay(50);
 }
@@ -3610,7 +3610,7 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring;
 	int r;
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	r = amdgpu_bo_reserve(ring->mqd_obj, false);
 	if (unlikely(r != 0))
@@ -3789,10 +3789,10 @@ static int gfx_v9_0_hw_fini(void *handle)
 	 */
 	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, adev->gfx.kiq.ring.me,
-				adev->gfx.kiq.ring.pipe,
-				adev->gfx.kiq.ring.queue, 0);
-		gfx_v9_0_kiq_fini_register(&adev->gfx.kiq.ring);
+		soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me,
+				adev->gfx.kiq[0].ring.pipe,
+				adev->gfx.kiq[0].ring.queue, 0);
+		gfx_v9_0_kiq_fini_register(&adev->gfx.kiq[0].ring);
 		soc15_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	}
@@ -3913,7 +3913,7 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev)
 	unsigned long flags;
 	uint32_t seq, reg_val_offs = 0;
 	uint64_t value = 0;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *ring = &kiq->ring;
 
 	BUG_ON(!ring->funcs->emit_rreg);
@@ -5385,7 +5385,7 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
 {
 	int i, r = 0;
 	struct amdgpu_device *adev = ring->adev;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	unsigned long flags;
 
@@ -6964,7 +6964,7 @@ static void gfx_v9_0_set_ring_funcs(struct amdgpu_device *adev)
 {
 	int i;
 
-	adev->gfx.kiq.ring.funcs = &gfx_v9_0_ring_funcs_kiq;
+	adev->gfx.kiq[0].ring.funcs = &gfx_v9_0_ring_funcs_kiq;
 
 	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
 		adev->gfx.gfx_ring[i].funcs = &gfx_v9_0_ring_funcs_gfx;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 7d6f4a68f4166..23d4081eca006 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -343,7 +343,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	/* For SRIOV run time, driver shouldn't access the register through MMIO
 	 * Directly use kiq to do the vm invalidation instead
 	 */
-	if (adev->gfx.kiq.ring.sched.ready && !adev->enable_mes &&
+	if (adev->gfx.kiq[0].ring.sched.ready && !adev->enable_mes &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
 	    down_read_trylock(&adev->reset_domain->sem)) {
 		struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];
@@ -428,11 +428,11 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint16_t queried_pasid;
 	bool ret;
 	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout;
-	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 
 	if (amdgpu_emu_mode == 0 && ring->sched.ready) {
-		spin_lock(&adev->gfx.kiq.ring_lock);
+		spin_lock(&adev->gfx.kiq[0].ring_lock);
 		/* 2 dwords flush + 8 dwords fence */
 		amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8);
 		kiq->pmf->kiq_invalidate_tlbs(ring,
@@ -440,12 +440,12 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 		if (r) {
 			amdgpu_ring_undo(ring);
-			spin_unlock(&adev->gfx.kiq.ring_lock);
+			spin_unlock(&adev->gfx.kiq[0].ring_lock);
 			return -ETIME;
 		}
 
 		amdgpu_ring_commit(ring);
-		spin_unlock(&adev->gfx.kiq.ring_lock);
+		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index d809f2ed56002..3828ca95899fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -291,7 +291,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	/* For SRIOV run time, driver shouldn't access the register through MMIO
 	 * Directly use kiq to do the vm invalidation instead
 	 */
-	if ((adev->gfx.kiq.ring.sched.ready || adev->mes.ring.sched.ready) &&
+	if ((adev->gfx.kiq[0].ring.sched.ready || adev->mes.ring.sched.ready) &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) {
 		struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];
 		const unsigned eng = 17;
@@ -329,11 +329,11 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint32_t seq;
 	uint16_t queried_pasid;
 	bool ret;
-	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 
 	if (amdgpu_emu_mode == 0 && ring->sched.ready) {
-		spin_lock(&adev->gfx.kiq.ring_lock);
+		spin_lock(&adev->gfx.kiq[0].ring_lock);
 		/* 2 dwords flush + 8 dwords fence */
 		amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8);
 		kiq->pmf->kiq_invalidate_tlbs(ring,
@@ -341,12 +341,12 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 		if (r) {
 			amdgpu_ring_undo(ring);
-			spin_unlock(&adev->gfx.kiq.ring_lock);
+			spin_unlock(&adev->gfx.kiq[0].ring_lock);
 			return -ETIME;
 		}
 
 		amdgpu_ring_commit(ring);
-		spin_unlock(&adev->gfx.kiq.ring_lock);
+		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 64ab1a306dfe3..290804a06e058 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -824,7 +824,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	/* This is necessary for a HW workaround under SRIOV as well
 	 * as GFXOFF under bare metal
 	 */
-	if (adev->gfx.kiq.ring.sched.ready &&
+	if (adev->gfx.kiq[0].ring.sched.ready &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
 	    down_read_trylock(&adev->reset_domain->sem)) {
 		uint32_t req = hub->vm_inv_eng0_req + hub->eng_distance * eng;
@@ -934,8 +934,8 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint16_t queried_pasid;
 	bool ret;
 	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout;
-	struct amdgpu_ring *ring = &adev->gfx.kiq.ring;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+	struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
 
 	if (amdgpu_in_reset(adev))
 		return -EIO;
@@ -955,7 +955,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		if (vega20_xgmi_wa)
 			ndw += kiq->pmf->invalidate_tlbs_size;
 
-		spin_lock(&adev->gfx.kiq.ring_lock);
+		spin_lock(&adev->gfx.kiq[0].ring_lock);
 		/* 2 dwords flush + 8 dwords fence */
 		amdgpu_ring_alloc(ring, ndw);
 		if (vega20_xgmi_wa)
@@ -966,13 +966,13 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 		if (r) {
 			amdgpu_ring_undo(ring);
-			spin_unlock(&adev->gfx.kiq.ring_lock);
+			spin_unlock(&adev->gfx.kiq[0].ring_lock);
 			up_read(&adev->reset_domain->sem);
 			return -ETIME;
 		}
 
 		amdgpu_ring_commit(ring);
-		spin_unlock(&adev->gfx.kiq.ring_lock);
+		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index 2e2062636d5f6..0599f8a6813e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -797,8 +797,8 @@ static void mes_v10_1_queue_init_register(struct amdgpu_ring *ring)
 
 static int mes_v10_1_kiq_enable_queue(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	int r;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
@@ -863,9 +863,9 @@ static int mes_v10_1_kiq_ring_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
 
-	spin_lock_init(&adev->gfx.kiq.ring_lock);
+	spin_lock_init(&adev->gfx.kiq[0].ring_lock);
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	ring->me = 3;
 	ring->pipe = 1;
@@ -891,7 +891,7 @@ static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev,
 	struct amdgpu_ring *ring;
 
 	if (pipe == AMDGPU_MES_KIQ_PIPE)
-		ring = &adev->gfx.kiq.ring;
+		ring = &adev->gfx.kiq[0].ring;
 	else if (pipe == AMDGPU_MES_SCHED_PIPE)
 		ring = &adev->mes.ring;
 	else
@@ -974,15 +974,15 @@ static int mes_v10_1_sw_fini(void *handle)
 		amdgpu_ucode_release(&adev->mes.fw[pipe]);
 	}
 
-	amdgpu_bo_free_kernel(&adev->gfx.kiq.ring.mqd_obj,
-			      &adev->gfx.kiq.ring.mqd_gpu_addr,
-			      &adev->gfx.kiq.ring.mqd_ptr);
+	amdgpu_bo_free_kernel(&adev->gfx.kiq[0].ring.mqd_obj,
+			      &adev->gfx.kiq[0].ring.mqd_gpu_addr,
+			      &adev->gfx.kiq[0].ring.mqd_ptr);
 
 	amdgpu_bo_free_kernel(&adev->mes.ring.mqd_obj,
 			      &adev->mes.ring.mqd_gpu_addr,
 			      &adev->mes.ring.mqd_ptr);
 
-	amdgpu_ring_fini(&adev->gfx.kiq.ring);
+	amdgpu_ring_fini(&adev->gfx.kiq[0].ring);
 	amdgpu_ring_fini(&adev->mes.ring);
 
 	amdgpu_mes_fini(adev);
@@ -1038,7 +1038,7 @@ static int mes_v10_1_kiq_hw_init(struct amdgpu_device *adev)
 
 	mes_v10_1_enable(adev, true);
 
-	mes_v10_1_kiq_setting(&adev->gfx.kiq.ring);
+	mes_v10_1_kiq_setting(&adev->gfx.kiq[0].ring);
 
 	r = mes_v10_1_queue_init(adev);
 	if (r)
@@ -1090,7 +1090,7 @@ static int mes_v10_1_hw_init(void *handle)
 	 * MES uses KIQ ring exclusively so driver cannot access KIQ ring
 	 * with MES enabled.
 	 */
-	adev->gfx.kiq.ring.sched.ready = false;
+	adev->gfx.kiq[0].ring.sched.ready = false;
 	adev->mes.ring.sched.ready = true;
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 45280f047180a..e853bcb892fcb 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -864,8 +864,8 @@ static void mes_v11_0_queue_init_register(struct amdgpu_ring *ring)
 
 static int mes_v11_0_kiq_enable_queue(struct amdgpu_device *adev)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq;
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	int r;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
@@ -894,7 +894,7 @@ static int mes_v11_0_queue_init(struct amdgpu_device *adev,
 	int r;
 
 	if (pipe == AMDGPU_MES_KIQ_PIPE)
-		ring = &adev->gfx.kiq.ring;
+		ring = &adev->gfx.kiq[0].ring;
 	else if (pipe == AMDGPU_MES_SCHED_PIPE)
 		ring = &adev->mes.ring;
 	else
@@ -961,9 +961,9 @@ static int mes_v11_0_kiq_ring_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
 
-	spin_lock_init(&adev->gfx.kiq.ring_lock);
+	spin_lock_init(&adev->gfx.kiq[0].ring_lock);
 
-	ring = &adev->gfx.kiq.ring;
+	ring = &adev->gfx.kiq[0].ring;
 
 	ring->me = 3;
 	ring->pipe = 1;
@@ -989,7 +989,7 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev,
 	struct amdgpu_ring *ring;
 
 	if (pipe == AMDGPU_MES_KIQ_PIPE)
-		ring = &adev->gfx.kiq.ring;
+		ring = &adev->gfx.kiq[0].ring;
 	else if (pipe == AMDGPU_MES_SCHED_PIPE)
 		ring = &adev->mes.ring;
 	else
@@ -1074,15 +1074,15 @@ static int mes_v11_0_sw_fini(void *handle)
 		amdgpu_ucode_release(&adev->mes.fw[pipe]);
 	}
 
-	amdgpu_bo_free_kernel(&adev->gfx.kiq.ring.mqd_obj,
-			      &adev->gfx.kiq.ring.mqd_gpu_addr,
-			      &adev->gfx.kiq.ring.mqd_ptr);
+	amdgpu_bo_free_kernel(&adev->gfx.kiq[0].ring.mqd_obj,
+			      &adev->gfx.kiq[0].ring.mqd_gpu_addr,
+			      &adev->gfx.kiq[0].ring.mqd_ptr);
 
 	amdgpu_bo_free_kernel(&adev->mes.ring.mqd_obj,
 			      &adev->mes.ring.mqd_gpu_addr,
 			      &adev->mes.ring.mqd_ptr);
 
-	amdgpu_ring_fini(&adev->gfx.kiq.ring);
+	amdgpu_ring_fini(&adev->gfx.kiq[0].ring);
 	amdgpu_ring_fini(&adev->mes.ring);
 
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) {
@@ -1175,7 +1175,7 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev)
 
 	mes_v11_0_enable(adev, true);
 
-	mes_v11_0_kiq_setting(&adev->gfx.kiq.ring);
+	mes_v11_0_kiq_setting(&adev->gfx.kiq[0].ring);
 
 	r = mes_v11_0_queue_init(adev, AMDGPU_MES_KIQ_PIPE);
 	if (r)
@@ -1196,7 +1196,7 @@ static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_sriov_vf(adev)) {
-		mes_v11_0_kiq_dequeue(&adev->gfx.kiq.ring);
+		mes_v11_0_kiq_dequeue(&adev->gfx.kiq[0].ring);
 		mes_v11_0_kiq_clear(adev);
 	}
 
@@ -1244,7 +1244,7 @@ static int mes_v11_0_hw_init(void *handle)
 	 * MES uses KIQ ring exclusively so driver cannot access KIQ ring
 	 * with MES enabled.
 	 */
-	adev->gfx.kiq.ring.sched.ready = false;
+	adev->gfx.kiq[0].ring.sched.ready = false;
 	adev->mes.ring.sched.ready = true;
 
 	return 0;
-- 
GitLab


From be697aa3a78ef83a6b8d49e1f0671a002e502cd0 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 27 Jul 2022 14:35:49 +0800
Subject: [PATCH 0138/1791] drm/amdgpu: move queue_bitmap to an independent
 structure (v3)

To allocate independent queue_bitmap for each XCD,
then the old bitmap policy can be continued to use
with a clear logic.

Use mec_bitmap[0] as default for all non-GC 9.4.3 IPs.

v2: squash commits to avoid breaking the build
v3: unify naming style

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c    |  2 +-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c       | 41 +++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |  7 +++-
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        |  6 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        |  6 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c         |  5 ++-
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c         |  7 ++--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         |  5 ++-
 9 files changed, 48 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 0385f7f692785..fed8bb9a721f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -162,7 +162,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 		 * clear
 		 */
 		bitmap_complement(gpu_resources.cp_queue_bitmap,
-				  adev->gfx.mec.queue_bitmap,
+				  adev->gfx.mec_bitmap[0].queue_bitmap,
 				  KGD_MAX_QUEUES);
 
 		/* According to linux/bitmap.h we shouldn't use bitmap_clear if
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index bc944ae4fd5b7..03875b971ba65 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -778,7 +778,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	 * Iterate through the shader engines and arrays of the device
 	 * to get number of waves in flight
 	 */
-	bitmap_complement(cp_queue_bitmap, adev->gfx.mec.queue_bitmap,
+	bitmap_complement(cp_queue_bitmap, adev->gfx.mec_bitmap[0].queue_bitmap,
 			  KGD_MAX_QUEUES);
 	max_queue_cnt = adev->gfx.mec.num_pipe_per_mec *
 			adev->gfx.mec.num_queue_per_pipe;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 9b6071df1fa7a..b300b1784210b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -63,10 +63,10 @@ void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit,
 }
 
 bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev,
-				     int mec, int pipe, int queue)
+				     int xcc_id, int mec, int pipe, int queue)
 {
 	return test_bit(amdgpu_gfx_mec_queue_to_bit(adev, mec, pipe, queue),
-			adev->gfx.mec.queue_bitmap);
+			adev->gfx.mec_bitmap[xcc_id].queue_bitmap);
 }
 
 int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev,
@@ -204,29 +204,38 @@ bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev,
 
 void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev)
 {
-	int i, queue, pipe;
+	int i, j, queue, pipe;
 	bool multipipe_policy = amdgpu_gfx_is_compute_multipipe_capable(adev);
 	int max_queues_per_mec = min(adev->gfx.mec.num_pipe_per_mec *
 				     adev->gfx.mec.num_queue_per_pipe,
 				     adev->gfx.num_compute_rings);
+	int num_xcd = (adev->gfx.num_xcd > 1) ? adev->gfx.num_xcd : 1;
 
 	if (multipipe_policy) {
-		/* policy: make queues evenly cross all pipes on MEC1 only */
-		for (i = 0; i < max_queues_per_mec; i++) {
-			pipe = i % adev->gfx.mec.num_pipe_per_mec;
-			queue = (i / adev->gfx.mec.num_pipe_per_mec) %
-				adev->gfx.mec.num_queue_per_pipe;
-
-			set_bit(pipe * adev->gfx.mec.num_queue_per_pipe + queue,
-					adev->gfx.mec.queue_bitmap);
+		/* policy: make queues evenly cross all pipes on MEC1 only
+		 * for multiple xcc, just use the original policy for simplicity */
+		for (j = 0; j < num_xcd; j++) {
+			for (i = 0; i < max_queues_per_mec; i++) {
+				pipe = i % adev->gfx.mec.num_pipe_per_mec;
+				queue = (i / adev->gfx.mec.num_pipe_per_mec) %
+					 adev->gfx.mec.num_queue_per_pipe;
+
+				set_bit(pipe * adev->gfx.mec.num_queue_per_pipe + queue,
+					adev->gfx.mec_bitmap[j].queue_bitmap);
+			}
 		}
 	} else {
 		/* policy: amdgpu owns all queues in the given pipe */
-		for (i = 0; i < max_queues_per_mec; ++i)
-			set_bit(i, adev->gfx.mec.queue_bitmap);
+		for (j = 0; j < num_xcd; j++) {
+			for (i = 0; i < max_queues_per_mec; ++i)
+				set_bit(i, adev->gfx.mec_bitmap[j].queue_bitmap);
+		}
 	}
 
-	dev_dbg(adev->dev, "mec queue bitmap weight=%d\n", bitmap_weight(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES));
+	for (j = 0; j < num_xcd; j++) {
+		dev_dbg(adev->dev, "mec queue bitmap weight=%d\n",
+			bitmap_weight(adev->gfx.mec_bitmap[j].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES));
+	}
 }
 
 void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
@@ -268,7 +277,7 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev,
 		    * adev->gfx.mec.num_queue_per_pipe;
 
 	while (--queue_bit >= 0) {
-		if (test_bit(queue_bit, adev->gfx.mec.queue_bitmap))
+		if (test_bit(queue_bit, adev->gfx.mec_bitmap[0].queue_bitmap))
 			continue;
 
 		amdgpu_queue_mask_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue);
@@ -516,7 +525,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 		return -EINVAL;
 
 	for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) {
-		if (!test_bit(i, adev->gfx.mec.queue_bitmap))
+		if (!test_bit(i, adev->gfx.mec_bitmap[0].queue_bitmap))
 			continue;
 
 		/* This situation may be hit in the future if a new HW
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index c742b4a369799..8303233106941 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -76,7 +76,9 @@ struct amdgpu_mec {
 	u32 num_pipe_per_mec;
 	u32 num_queue_per_pipe;
 	void			*mqd_backup[AMDGPU_MAX_COMPUTE_RINGS + 1];
+};
 
+struct amdgpu_mec_bitmap {
 	/* These are the resources for which amdgpu takes ownership */
 	DECLARE_BITMAP(queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 };
@@ -296,6 +298,7 @@ struct amdgpu_gfx {
 	struct amdgpu_ce		ce;
 	struct amdgpu_me		me;
 	struct amdgpu_mec		mec;
+	struct amdgpu_mec_bitmap	mec_bitmap[AMDGPU_MAX_GC_INSTANCES];
 	struct amdgpu_kiq		kiq[AMDGPU_MAX_GC_INSTANCES];
 	struct amdgpu_imu		imu;
 	bool				rs64_enable; /* firmware format */
@@ -425,8 +428,8 @@ int amdgpu_gfx_mec_queue_to_bit(struct amdgpu_device *adev, int mec,
 				int pipe, int queue);
 void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit,
 				 int *mec, int *pipe, int *queue);
-bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int mec,
-				     int pipe, int queue);
+bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int inst,
+				     int mec, int pipe, int queue);
 bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev,
 					       struct amdgpu_ring *ring);
 bool amdgpu_gfx_is_high_priority_graphics_queue(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index d4e7de8fd9da1..88f8424ea1e4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4219,7 +4219,7 @@ static int gfx_v10_0_mec_init(struct amdgpu_device *adev)
 
 	const struct gfx_firmware_header_v1_0 *mec_hdr = NULL;
 
-	bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -4614,8 +4614,8 @@ static int gfx_v10_0_sw_init(void *handle)
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k,
-								     j))
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
 					continue;
 
 				r = gfx_v10_0_compute_ring_init(adev, ring_id,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 6a5435255e6d7..3e42a44f10a73 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -699,7 +699,7 @@ static int gfx_v11_0_mec_init(struct amdgpu_device *adev)
 	u32 *hpd;
 	size_t mec_hpd_size;
 
-	bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -1374,8 +1374,8 @@ static int gfx_v11_0_sw_init(void *handle)
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k,
-								     j))
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
 					continue;
 
 				r = gfx_v11_0_compute_ring_init(adev, ring_id,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 9d5c1e29b4a35..46740ad9a80ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -2728,7 +2728,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev)
 	u32 *hpd;
 	size_t mec_hpd_size;
 
-	bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -4456,7 +4456,8 @@ static int gfx_v7_0_sw_init(void *handle)
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j))
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
 					continue;
 
 				r = gfx_v7_0_compute_ring_init(adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index ed04bad8543da..18722450e2655 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1304,7 +1304,7 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
 	u32 *hpd;
 	size_t mec_hpd_size;
 
-	bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -2001,7 +2001,8 @@ static int gfx_v8_0_sw_init(void *handle)
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j))
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
 					continue;
 
 				r = gfx_v8_0_compute_ring_init(adev,
@@ -4319,7 +4320,7 @@ static int gfx_v8_0_kiq_kcq_enable(struct amdgpu_device *adev)
 	int r, i;
 
 	for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) {
-		if (!test_bit(i, adev->gfx.mec.queue_bitmap))
+		if (!test_bit(i, adev->gfx.mec_bitmap[0].queue_bitmap))
 			continue;
 
 		/* This situation may be hit in the future if a new HW
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index adf86bc7ed369..49adc36dcc6fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1713,7 +1713,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
 
 	const struct gfx_firmware_header_v1_0 *mec_hdr;
 
-	bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -2154,7 +2154,8 @@ static int gfx_v9_0_sw_init(void *handle)
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j))
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
 					continue;
 
 				r = gfx_v9_0_compute_ring_init(adev,
-- 
GitLab


From c38be07035bcb31274ce5f85e3b249f691c5b8db Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 16 Nov 2021 21:56:34 +0800
Subject: [PATCH 0139/1791] drm/amdgpu: separate the mqd_backup for kiq from
 kcq

This will benifit the mqd indexing for kiq/kcq in multi XCD case.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 8303233106941..d811cb038e947 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -75,7 +75,7 @@ struct amdgpu_mec {
 	u32 num_mec;
 	u32 num_pipe_per_mec;
 	u32 num_queue_per_pipe;
-	void			*mqd_backup[AMDGPU_MAX_COMPUTE_RINGS + 1];
+	void			*mqd_backup[AMDGPU_MAX_COMPUTE_RINGS * AMDGPU_MAX_GC_INSTANCES];
 };
 
 struct amdgpu_mec_bitmap {
@@ -122,6 +122,7 @@ struct amdgpu_kiq {
 	struct amdgpu_ring	ring;
 	struct amdgpu_irq_src	irq;
 	const struct kiq_pm4_funcs *pmf;
+	void			*mqd_backup;
 };
 
 /*
-- 
GitLab


From def799c6596d078112095c24c25e162cb5102d90 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 24 May 2022 12:23:03 +0800
Subject: [PATCH 0140/1791] drm/amdgpu: add multi-xcc support to amdgpu_gfx
 interfaces (v4)

v1: Modify kiq_init/fini, mqd_sw_init/fini and
enable/disable_kcq to adapt to multi-die case.
Pass 0 as default to all asics with single xcc (Le)
v2: squash commits to avoid breaking the build (Le)
v3: unify naming style (Le)
v4: apply the changes to gc v11_0 (Hawking)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 75 ++++++++++++++-----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 16 +++---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c  | 23 ++++----
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  | 23 ++++----
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c   | 19 +++----
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   | 25 ++++-----
 6 files changed, 93 insertions(+), 88 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index b300b1784210b..7f5c60381103a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -267,7 +267,7 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev)
 }
 
 static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev,
-				  struct amdgpu_ring *ring)
+				  struct amdgpu_ring *ring, int xcc_id)
 {
 	int queue_bit;
 	int mec, pipe, queue;
@@ -277,7 +277,7 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev,
 		    * adev->gfx.mec.num_queue_per_pipe;
 
 	while (--queue_bit >= 0) {
-		if (test_bit(queue_bit, adev->gfx.mec_bitmap[0].queue_bitmap))
+		if (test_bit(queue_bit, adev->gfx.mec_bitmap[xcc_id].queue_bitmap))
 			continue;
 
 		amdgpu_queue_mask_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue);
@@ -303,9 +303,9 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev,
 
 int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 			     struct amdgpu_ring *ring,
-			     struct amdgpu_irq_src *irq)
+			     struct amdgpu_irq_src *irq, int xcc_id)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	int r = 0;
 
 	spin_lock_init(&kiq->ring_lock);
@@ -314,15 +314,16 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
 	ring->doorbell_index = adev->doorbell_index.kiq;
+	ring->xcc_id = xcc_id;
 	ring->vm_hub = AMDGPU_GFXHUB_0;
 
-	r = amdgpu_gfx_kiq_acquire(adev, ring);
+	r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id);
 	if (r)
 		return r;
 
 	ring->eop_gpu_addr = kiq->eop_gpu_addr;
 	ring->no_scheduler = true;
-	sprintf(ring->name, "kiq_%d.%d.%d", ring->me, ring->pipe, ring->queue);
+	sprintf(ring->name, "kiq_%d.%d.%d.%d", xcc_id, ring->me, ring->pipe, ring->queue);
 	r = amdgpu_ring_init(adev, ring, 1024, irq, AMDGPU_CP_KIQ_IRQ_DRIVER0,
 			     AMDGPU_RING_PRIO_DEFAULT, NULL);
 	if (r)
@@ -336,19 +337,19 @@ void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring)
 	amdgpu_ring_fini(ring);
 }
 
-void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev)
+void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev, int xcc_id)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 
 	amdgpu_bo_free_kernel(&kiq->eop_obj, &kiq->eop_gpu_addr, NULL);
 }
 
 int amdgpu_gfx_kiq_init(struct amdgpu_device *adev,
-			unsigned hpd_size)
+			unsigned hpd_size, int xcc_id)
 {
 	int r;
 	u32 *hpd;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 
 	r = amdgpu_bo_create_kernel(adev, hpd_size, PAGE_SIZE,
 				    AMDGPU_GEM_DOMAIN_GTT, &kiq->eop_obj,
@@ -371,13 +372,13 @@ int amdgpu_gfx_kiq_init(struct amdgpu_device *adev,
 
 /* create MQD for each compute/gfx queue */
 int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
-			   unsigned mqd_size)
+			   unsigned mqd_size, int xcc_id)
 {
-	struct amdgpu_ring *ring = NULL;
 	int r, i;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
+	struct amdgpu_ring *ring = &kiq->ring;
 
 	/* create MQD for KIQ */
-	ring = &adev->gfx.kiq[0].ring;
 	if (!adev->enable_mes_kiq && !ring->mqd_obj) {
 		/* originaly the KIQ MQD is put in GTT domain, but for SRIOV VRAM domain is a must
 		 * otherwise hypervisor trigger SAVE_VF fail after driver unloaded which mean MQD
@@ -396,8 +397,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 		}
 
 		/* prepare MQD backup */
-		adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS] = kmalloc(mqd_size, GFP_KERNEL);
-		if (!adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS])
+		kiq->mqd_backup = kmalloc(mqd_size, GFP_KERNEL);
+		if (!kiq->mqd_backup)
 				dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name);
 	}
 
@@ -424,7 +425,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 
 	/* create MQD for each KCQ */
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		ring = &adev->gfx.compute_ring[i];
+		ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings];
 		if (!ring->mqd_obj) {
 			r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
 						    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
@@ -435,7 +436,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 			}
 
 			/* prepare MQD backup */
-			adev->gfx.mec.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL);
+			adev->gfx.mec.mqd_backup[i + xcc_id * adev->gfx.num_compute_rings] = kmalloc(mqd_size, GFP_KERNEL);
 			if (!adev->gfx.mec.mqd_backup[i])
 				dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name);
 		}
@@ -444,10 +445,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 	return 0;
 }
 
-void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev)
+void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring = NULL;
-	int i;
+	int i, j;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 
 	if (adev->asic_type >= CHIP_NAVI10 && amdgpu_async_gfx_ring) {
 		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
@@ -460,6 +462,7 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev)
 	}
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		j = i + xcc_id * adev->gfx.num_compute_rings;
 		ring = &adev->gfx.compute_ring[i];
 		kfree(adev->gfx.mec.mqd_backup[i]);
 		amdgpu_bo_free_kernel(&ring->mqd_obj,
@@ -467,36 +470,40 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev)
 				      &ring->mqd_ptr);
 	}
 
-	ring = &adev->gfx.kiq[0].ring;
+	ring = &kiq->ring;
+	kfree(kiq->mqd_backup);
 	kfree(adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS]);
 	amdgpu_bo_free_kernel(&ring->mqd_obj,
 			      &ring->mqd_gpu_addr,
 			      &ring->mqd_ptr);
 }
 
-int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev)
+int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	int i, r = 0;
+	int j;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues)
 		return -EINVAL;
 
-	spin_lock(&adev->gfx.kiq[0].ring_lock);
+	spin_lock(&kiq->ring_lock);
 	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
 					adev->gfx.num_compute_rings)) {
 		spin_unlock(&adev->gfx.kiq[0].ring_lock);
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < adev->gfx.num_compute_rings; i++)
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		j = i + xcc_id * adev->gfx.num_compute_rings;
 		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i],
 					   RESET_QUEUES, 0, 0);
+	}
 
 	if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang)
 		r = amdgpu_ring_test_helper(kiq_ring);
-	spin_unlock(&adev->gfx.kiq[0].ring_lock);
+	spin_unlock(&kiq->ring_lock);
 
 	return r;
 }
@@ -514,18 +521,18 @@ int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev,
 	return set_resource_bit;
 }
 
-int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
+int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
+	struct amdgpu_ring *kiq_ring = &kiq->ring;
 	uint64_t queue_mask = 0;
-	int r, i;
+	int r, i, j;
 
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues || !kiq->pmf->kiq_set_resources)
 		return -EINVAL;
 
 	for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) {
-		if (!test_bit(i, adev->gfx.mec_bitmap[0].queue_bitmap))
+		if (!test_bit(i, adev->gfx.mec_bitmap[xcc_id].queue_bitmap))
 			continue;
 
 		/* This situation may be hit in the future if a new HW
@@ -541,7 +548,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 
 	DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe,
 							kiq_ring->queue);
-	spin_lock(&adev->gfx.kiq[0].ring_lock);
+	spin_lock(&kiq->ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
 					adev->gfx.num_compute_rings +
 					kiq->pmf->set_resources_size);
@@ -555,11 +562,13 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev)
 		queue_mask = ~0ULL;
 
 	kiq->pmf->kiq_set_resources(kiq_ring, queue_mask);
-	for (i = 0; i < adev->gfx.num_compute_rings; i++)
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		j = i + xcc_id * adev->gfx.num_compute_rings;
 		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.compute_ring[i]);
+	}
 
 	r = amdgpu_ring_test_helper(kiq_ring);
-	spin_unlock(&adev->gfx.kiq[0].ring_lock);
+	spin_unlock(&kiq->ring_lock);
 	if (r)
 		DRM_ERROR("KCQ enable failed\n");
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index d811cb038e947..a9e41d7970ead 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -408,19 +408,19 @@ void amdgpu_gfx_parse_disable_cu(unsigned *mask, unsigned max_se,
 
 int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 			     struct amdgpu_ring *ring,
-			     struct amdgpu_irq_src *irq);
+			     struct amdgpu_irq_src *irq, int xcc_id);
 
 void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring);
 
-void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev);
+void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_kiq_init(struct amdgpu_device *adev,
-			unsigned hpd_size);
+			unsigned hpd_size, int xcc_id);
 
 int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
-			   unsigned mqd_size);
-void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev);
-int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev);
-int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev);
+			   unsigned mqd_size, int xcc_id);
+void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id);
+int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id);
+int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id);
 
 void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev);
 void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev);
@@ -429,7 +429,7 @@ int amdgpu_gfx_mec_queue_to_bit(struct amdgpu_device *adev, int mec,
 				int pipe, int queue);
 void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit,
 				 int *mec, int *pipe, int *queue);
-bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int inst,
+bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int xcc_id,
 				     int mec, int pipe, int queue);
 bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev,
 					       struct amdgpu_ring *ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 88f8424ea1e4e..accc0a7251b9a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4629,19 +4629,19 @@ static int gfx_v10_0_sw_init(void *handle)
 	}
 
 	if (!adev->enable_mes_kiq) {
-		r = amdgpu_gfx_kiq_init(adev, GFX10_MEC_HPD_SIZE);
+		r = amdgpu_gfx_kiq_init(adev, GFX10_MEC_HPD_SIZE, 0);
 		if (r) {
 			DRM_ERROR("Failed to init KIQ BOs!\n");
 			return r;
 		}
 
 		kiq = &adev->gfx.kiq[0];
-		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
+		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
 		if (r)
 			return r;
 	}
 
-	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v10_compute_mqd));
+	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v10_compute_mqd), 0);
 	if (r)
 		return r;
 
@@ -4690,11 +4690,11 @@ static int gfx_v10_0_sw_fini(void *handle)
 	for (i = 0; i < adev->gfx.num_compute_rings; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	amdgpu_gfx_mqd_sw_fini(adev);
+	amdgpu_gfx_mqd_sw_fini(adev, 0);
 
 	if (!adev->enable_mes_kiq) {
 		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
-		amdgpu_gfx_kiq_fini(adev);
+		amdgpu_gfx_kiq_fini(adev, 0);
 	}
 
 	gfx_v10_0_pfp_fini(adev);
@@ -6812,14 +6812,13 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v10_compute_mqd *mqd = ring->mqd_ptr;
-	int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS;
 
 	gfx_v10_0_kiq_setting(ring);
 
 	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(*mqd));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
@@ -6841,8 +6840,8 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring)
 		nv_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(*mqd));
 	}
 
 	return 0;
@@ -6927,7 +6926,7 @@ static int gfx_v10_0_kcq_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = amdgpu_gfx_enable_kcq(adev);
+	r = amdgpu_gfx_enable_kcq(adev, 0);
 done:
 	return r;
 }
@@ -7280,7 +7279,7 @@ static int gfx_v10_0_hw_fini(void *handle)
 				DRM_ERROR("KGQ disable failed\n");
 		}
 #endif
-		if (amdgpu_gfx_disable_kcq(adev))
+		if (amdgpu_gfx_disable_kcq(adev, 0))
 			DRM_ERROR("KCQ disable failed\n");
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 3e42a44f10a73..4fbefe236fc71 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1389,19 +1389,19 @@ static int gfx_v11_0_sw_init(void *handle)
 	}
 
 	if (!adev->enable_mes_kiq) {
-		r = amdgpu_gfx_kiq_init(adev, GFX11_MEC_HPD_SIZE);
+		r = amdgpu_gfx_kiq_init(adev, GFX11_MEC_HPD_SIZE, 0);
 		if (r) {
 			DRM_ERROR("Failed to init KIQ BOs!\n");
 			return r;
 		}
 
 		kiq = &adev->gfx.kiq[0];
-		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
+		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
 		if (r)
 			return r;
 	}
 
-	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v11_compute_mqd));
+	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v11_compute_mqd), 0);
 	if (r)
 		return r;
 
@@ -1463,11 +1463,11 @@ static int gfx_v11_0_sw_fini(void *handle)
 	for (i = 0; i < adev->gfx.num_compute_rings; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	amdgpu_gfx_mqd_sw_fini(adev);
+	amdgpu_gfx_mqd_sw_fini(adev, 0);
 
 	if (!adev->enable_mes_kiq) {
 		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
-		amdgpu_gfx_kiq_fini(adev);
+		amdgpu_gfx_kiq_fini(adev, 0);
 	}
 
 	gfx_v11_0_pfp_fini(adev);
@@ -4035,14 +4035,13 @@ static int gfx_v11_0_kiq_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v11_compute_mqd *mqd = ring->mqd_ptr;
-	int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS;
 
 	gfx_v11_0_kiq_setting(ring);
 
 	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(*mqd));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
@@ -4064,8 +4063,8 @@ static int gfx_v11_0_kiq_init_queue(struct amdgpu_ring *ring)
 		soc21_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(*mqd));
 	}
 
 	return 0;
@@ -4153,7 +4152,7 @@ static int gfx_v11_0_kcq_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = amdgpu_gfx_enable_kcq(adev);
+	r = amdgpu_gfx_enable_kcq(adev, 0);
 done:
 	return r;
 }
@@ -4456,7 +4455,7 @@ static int gfx_v11_0_hw_fini(void *handle)
 				DRM_ERROR("KGQ disable failed\n");
 		}
 #endif
-		if (amdgpu_gfx_disable_kcq(adev))
+		if (amdgpu_gfx_disable_kcq(adev, 0))
 			DRM_ERROR("KCQ disable failed\n");
 
 		amdgpu_mes_kiq_hw_fini(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 18722450e2655..8a43e87de49fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -2016,19 +2016,19 @@ static int gfx_v8_0_sw_init(void *handle)
 		}
 	}
 
-	r = amdgpu_gfx_kiq_init(adev, GFX8_MEC_HPD_SIZE);
+	r = amdgpu_gfx_kiq_init(adev, GFX8_MEC_HPD_SIZE, 0);
 	if (r) {
 		DRM_ERROR("Failed to init KIQ BOs!\n");
 		return r;
 	}
 
 	kiq = &adev->gfx.kiq[0];
-	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
+	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
 	if (r)
 		return r;
 
 	/* create MQD for all compute queues as well as KIQ for SRIOV case */
-	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct vi_mqd_allocation));
+	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct vi_mqd_allocation), 0);
 	if (r)
 		return r;
 
@@ -2051,9 +2051,9 @@ static int gfx_v8_0_sw_fini(void *handle)
 	for (i = 0; i < adev->gfx.num_compute_rings; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	amdgpu_gfx_mqd_sw_fini(adev);
+	amdgpu_gfx_mqd_sw_fini(adev, 0);
 	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
-	amdgpu_gfx_kiq_fini(adev);
+	amdgpu_gfx_kiq_fini(adev, 0);
 
 	gfx_v8_0_mec_fini(adev);
 	amdgpu_gfx_rlc_fini(adev);
@@ -4596,14 +4596,13 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct vi_mqd *mqd = ring->mqd_ptr;
-	int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS;
 
 	gfx_v8_0_kiq_setting(ring);
 
 	if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct vi_mqd_allocation));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
@@ -4626,8 +4625,8 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring)
 		vi_srbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct vi_mqd_allocation));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct vi_mqd_allocation));
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 49adc36dcc6fc..62af92e5be51c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2169,19 +2169,19 @@ static int gfx_v9_0_sw_init(void *handle)
 		}
 	}
 
-	r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE);
+	r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, 0);
 	if (r) {
 		DRM_ERROR("Failed to init KIQ BOs!\n");
 		return r;
 	}
 
 	kiq = &adev->gfx.kiq[0];
-	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq);
+	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
 	if (r)
 		return r;
 
 	/* create MQD for all compute queues as wel as KIQ for SRIOV case */
-	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation));
+	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation), 0);
 	if (r)
 		return r;
 
@@ -2216,9 +2216,9 @@ static int gfx_v9_0_sw_fini(void *handle)
 	for (i = 0; i < adev->gfx.num_compute_rings; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	amdgpu_gfx_mqd_sw_fini(adev);
+	amdgpu_gfx_mqd_sw_fini(adev, 0);
 	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
-	amdgpu_gfx_kiq_fini(adev);
+	amdgpu_gfx_kiq_fini(adev, 0);
 
 	gfx_v9_0_mec_fini(adev);
 	amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj,
@@ -3520,7 +3520,6 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
-	int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS;
 	struct v9_mqd *tmp_mqd;
 
 	gfx_v9_0_kiq_setting(ring);
@@ -3530,11 +3529,11 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 	 * driver need to re-init the mqd.
 	 * check mqd->cp_hqd_pq_control since this value should not be 0
 	 */
-	tmp_mqd = (struct v9_mqd *)adev->gfx.mec.mqd_backup[mqd_idx];
+	tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[0].mqd_backup;
 	if (amdgpu_in_reset(adev) && tmp_mqd->cp_hqd_pq_control){
 		/* for GPU_RESET case , reset MQD to a clean status */
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct v9_mqd_allocation));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
@@ -3558,8 +3557,8 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 		soc15_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
-		if (adev->gfx.mec.mqd_backup[mqd_idx])
-			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct v9_mqd_allocation));
 	}
 
 	return 0;
@@ -3653,7 +3652,7 @@ static int gfx_v9_0_kcq_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = amdgpu_gfx_enable_kcq(adev);
+	r = amdgpu_gfx_enable_kcq(adev, 0);
 done:
 	return r;
 }
@@ -3772,7 +3771,7 @@ static int gfx_v9_0_hw_fini(void *handle)
 	/* DF freeze and kcq disable will fail */
 	if (!amdgpu_ras_intr_triggered())
 		/* disable KCQ to avoid CPC touch memory not valid anymore */
-		amdgpu_gfx_disable_kcq(adev);
+		amdgpu_gfx_disable_kcq(adev, 0);
 
 	if (amdgpu_sriov_vf(adev)) {
 		gfx_v9_0_cp_gfx_enable(adev, false);
-- 
GitLab


From 86301129698be52f8398f92ea8564168f6bfcae1 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 24 May 2022 10:09:39 +0800
Subject: [PATCH 0141/1791] drm/amdgpu: split gc v9_4_3 functionality from gc
 v9_0

To prepare for gc v9_4_3 specific feature.

v2: fix exports (Alex)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2666 ++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h |    3 +-
 2 files changed, 2644 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 5f8500577c021..212d48114da8d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -25,16 +25,292 @@
 #include "amdgpu.h"
 #include "amdgpu_gfx.h"
 #include "soc15.h"
+#include "soc15d.h"
 #include "soc15_common.h"
 #include "vega10_enum.h"
 
+#include "clearstate_gfx9.h"
+#include "v9_structs.h"
+
+#include "ivsrcid/gfx/irqsrcs_gfx_9_0.h"
+
 #include "gc/gc_9_4_3_offset.h"
 #include "gc/gc_9_4_3_sh_mask.h"
 
 #include "gfx_v9_4_3.h"
 
+MODULE_FIRMWARE("amdgpu/gc_9_4_3_mec.bin");
+MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
+
+#define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
 
+static const struct soc15_reg_golden golden_settings_gc_9_4_3[] = {
+
+};
+
+static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev);
+static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev);
+static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev);
+static void gfx_v9_4_3_set_rlc_funcs(struct amdgpu_device *adev);
+static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
+				struct amdgpu_cu_info *cu_info);
+
+static void gfx_v9_4_3_kiq_set_resources(struct amdgpu_ring *kiq_ring,
+				uint64_t queue_mask)
+{
+	amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_SET_RESOURCES, 6));
+	amdgpu_ring_write(kiq_ring,
+		PACKET3_SET_RESOURCES_VMID_MASK(0) |
+		/* vmid_mask:0* queue_type:0 (KIQ) */
+		PACKET3_SET_RESOURCES_QUEUE_TYPE(0));
+	amdgpu_ring_write(kiq_ring,
+			lower_32_bits(queue_mask));	/* queue mask lo */
+	amdgpu_ring_write(kiq_ring,
+			upper_32_bits(queue_mask));	/* queue mask hi */
+	amdgpu_ring_write(kiq_ring, 0);	/* gws mask lo */
+	amdgpu_ring_write(kiq_ring, 0);	/* gws mask hi */
+	amdgpu_ring_write(kiq_ring, 0);	/* oac mask */
+	amdgpu_ring_write(kiq_ring, 0);	/* gds heap base:0, gds heap size:0 */
+}
+
+static void gfx_v9_4_3_kiq_map_queues(struct amdgpu_ring *kiq_ring,
+				 struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = kiq_ring->adev;
+	uint64_t mqd_addr = amdgpu_bo_gpu_offset(ring->mqd_obj);
+	uint64_t wptr_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
+	uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0;
+
+	amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_MAP_QUEUES, 5));
+	/* Q_sel:0, vmid:0, vidmem: 1, engine:0, num_Q:1*/
+	amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */
+			 PACKET3_MAP_QUEUES_QUEUE_SEL(0) | /* Queue_Sel */
+			 PACKET3_MAP_QUEUES_VMID(0) | /* VMID */
+			 PACKET3_MAP_QUEUES_QUEUE(ring->queue) |
+			 PACKET3_MAP_QUEUES_PIPE(ring->pipe) |
+			 PACKET3_MAP_QUEUES_ME((ring->me == 1 ? 0 : 1)) |
+			 /*queue_type: normal compute queue */
+			 PACKET3_MAP_QUEUES_QUEUE_TYPE(0) |
+			 /* alloc format: all_on_one_pipe */
+			 PACKET3_MAP_QUEUES_ALLOC_FORMAT(0) |
+			 PACKET3_MAP_QUEUES_ENGINE_SEL(eng_sel) |
+			 /* num_queues: must be 1 */
+			 PACKET3_MAP_QUEUES_NUM_QUEUES(1));
+	amdgpu_ring_write(kiq_ring,
+			PACKET3_MAP_QUEUES_DOORBELL_OFFSET(ring->doorbell_index));
+	amdgpu_ring_write(kiq_ring, lower_32_bits(mqd_addr));
+	amdgpu_ring_write(kiq_ring, upper_32_bits(mqd_addr));
+	amdgpu_ring_write(kiq_ring, lower_32_bits(wptr_addr));
+	amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr));
+}
+
+static void gfx_v9_4_3_kiq_unmap_queues(struct amdgpu_ring *kiq_ring,
+				   struct amdgpu_ring *ring,
+				   enum amdgpu_unmap_queues_action action,
+				   u64 gpu_addr, u64 seq)
+{
+	uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0;
+
+	amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_UNMAP_QUEUES, 4));
+	amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */
+			  PACKET3_UNMAP_QUEUES_ACTION(action) |
+			  PACKET3_UNMAP_QUEUES_QUEUE_SEL(0) |
+			  PACKET3_UNMAP_QUEUES_ENGINE_SEL(eng_sel) |
+			  PACKET3_UNMAP_QUEUES_NUM_QUEUES(1));
+	amdgpu_ring_write(kiq_ring,
+			PACKET3_UNMAP_QUEUES_DOORBELL_OFFSET0(ring->doorbell_index));
+
+	if (action == PREEMPT_QUEUES_NO_UNMAP) {
+		amdgpu_ring_write(kiq_ring, lower_32_bits(gpu_addr));
+		amdgpu_ring_write(kiq_ring, upper_32_bits(gpu_addr));
+		amdgpu_ring_write(kiq_ring, seq);
+	} else {
+		amdgpu_ring_write(kiq_ring, 0);
+		amdgpu_ring_write(kiq_ring, 0);
+		amdgpu_ring_write(kiq_ring, 0);
+	}
+}
+
+static void gfx_v9_4_3_kiq_query_status(struct amdgpu_ring *kiq_ring,
+				   struct amdgpu_ring *ring,
+				   u64 addr,
+				   u64 seq)
+{
+	uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0;
+
+	amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_QUERY_STATUS, 5));
+	amdgpu_ring_write(kiq_ring,
+			  PACKET3_QUERY_STATUS_CONTEXT_ID(0) |
+			  PACKET3_QUERY_STATUS_INTERRUPT_SEL(0) |
+			  PACKET3_QUERY_STATUS_COMMAND(2));
+	/* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */
+	amdgpu_ring_write(kiq_ring,
+			PACKET3_QUERY_STATUS_DOORBELL_OFFSET(ring->doorbell_index) |
+			PACKET3_QUERY_STATUS_ENG_SEL(eng_sel));
+	amdgpu_ring_write(kiq_ring, lower_32_bits(addr));
+	amdgpu_ring_write(kiq_ring, upper_32_bits(addr));
+	amdgpu_ring_write(kiq_ring, lower_32_bits(seq));
+	amdgpu_ring_write(kiq_ring, upper_32_bits(seq));
+}
+
+static void gfx_v9_4_3_kiq_invalidate_tlbs(struct amdgpu_ring *kiq_ring,
+				uint16_t pasid, uint32_t flush_type,
+				bool all_hub)
+{
+	amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_INVALIDATE_TLBS, 0));
+	amdgpu_ring_write(kiq_ring,
+			PACKET3_INVALIDATE_TLBS_DST_SEL(1) |
+			PACKET3_INVALIDATE_TLBS_ALL_HUB(all_hub) |
+			PACKET3_INVALIDATE_TLBS_PASID(pasid) |
+			PACKET3_INVALIDATE_TLBS_FLUSH_TYPE(flush_type));
+}
+
+static const struct kiq_pm4_funcs gfx_v9_4_3_kiq_pm4_funcs = {
+	.kiq_set_resources = gfx_v9_4_3_kiq_set_resources,
+	.kiq_map_queues = gfx_v9_4_3_kiq_map_queues,
+	.kiq_unmap_queues = gfx_v9_4_3_kiq_unmap_queues,
+	.kiq_query_status = gfx_v9_4_3_kiq_query_status,
+	.kiq_invalidate_tlbs = gfx_v9_4_3_kiq_invalidate_tlbs,
+	.set_resources_size = 8,
+	.map_queues_size = 7,
+	.unmap_queues_size = 6,
+	.query_status_size = 7,
+	.invalidate_tlbs_size = 2,
+};
+
+static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev)
+{
+	adev->gfx.kiq[0].pmf = &gfx_v9_4_3_kiq_pm4_funcs;
+}
+
+static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
+{
+
+}
+
+static void gfx_v9_4_3_write_data_to_reg(struct amdgpu_ring *ring, int eng_sel,
+				       bool wc, uint32_t reg, uint32_t val)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, WRITE_DATA_ENGINE_SEL(eng_sel) |
+				WRITE_DATA_DST_SEL(0) |
+				(wc ? WR_CONFIRM : 0));
+	amdgpu_ring_write(ring, reg);
+	amdgpu_ring_write(ring, 0);
+	amdgpu_ring_write(ring, val);
+}
+
+static void gfx_v9_4_3_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel,
+				  int mem_space, int opt, uint32_t addr0,
+				  uint32_t addr1, uint32_t ref, uint32_t mask,
+				  uint32_t inv)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5));
+	amdgpu_ring_write(ring,
+				 /* memory (1) or register (0) */
+				 (WAIT_REG_MEM_MEM_SPACE(mem_space) |
+				 WAIT_REG_MEM_OPERATION(opt) | /* wait */
+				 WAIT_REG_MEM_FUNCTION(3) |  /* equal */
+				 WAIT_REG_MEM_ENGINE(eng_sel)));
+
+	if (mem_space)
+		BUG_ON(addr0 & 0x3); /* Dword align */
+	amdgpu_ring_write(ring, addr0);
+	amdgpu_ring_write(ring, addr1);
+	amdgpu_ring_write(ring, ref);
+	amdgpu_ring_write(ring, mask);
+	amdgpu_ring_write(ring, inv); /* poll interval */
+}
+
+static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t tmp = 0;
+	unsigned i;
+	int r;
+
+	WREG32_SOC15(GC, 0, regSCRATCH_REG0, 0xCAFEDEAD);
+	r = amdgpu_ring_alloc(ring, 3);
+	if (r)
+		return r;
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
+	amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0) -
+			  PACKET3_SET_UCONFIG_REG_START);
+	amdgpu_ring_write(ring, 0xDEADBEEF);
+	amdgpu_ring_commit(ring);
+
+	for (i = 0; i < adev->usec_timeout; i++) {
+		tmp = RREG32_SOC15(GC, 0, regSCRATCH_REG0);
+		if (tmp == 0xDEADBEEF)
+			break;
+		udelay(1);
+	}
+
+	if (i >= adev->usec_timeout)
+		r = -ETIMEDOUT;
+	return r;
+}
+
+static int gfx_v9_4_3_ring_test_ib(struct amdgpu_ring *ring, long timeout)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_ib ib;
+	struct dma_fence *f = NULL;
+
+	unsigned index;
+	uint64_t gpu_addr;
+	uint32_t tmp;
+	long r;
+
+	r = amdgpu_device_wb_get(adev, &index);
+	if (r)
+		return r;
+
+	gpu_addr = adev->wb.gpu_addr + (index * 4);
+	adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
+	memset(&ib, 0, sizeof(ib));
+	r = amdgpu_ib_get(adev, NULL, 16,
+			  AMDGPU_IB_POOL_DIRECT, &ib);
+	if (r)
+		goto err1;
+
+	ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3);
+	ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM;
+	ib.ptr[2] = lower_32_bits(gpu_addr);
+	ib.ptr[3] = upper_32_bits(gpu_addr);
+	ib.ptr[4] = 0xDEADBEEF;
+	ib.length_dw = 5;
+
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
+	if (r)
+		goto err2;
+
+	r = dma_fence_wait_timeout(f, false, timeout);
+	if (r == 0) {
+		r = -ETIMEDOUT;
+		goto err2;
+	} else if (r < 0) {
+		goto err2;
+	}
+
+	tmp = adev->wb.wb[index];
+	if (tmp == 0xDEADBEEF)
+		r = 0;
+	else
+		r = -EINVAL;
+
+err2:
+	amdgpu_ib_free(adev, &ib, NULL);
+	dma_fence_put(f);
+err1:
+	amdgpu_device_wb_free(adev, index);
+	return r;
+}
+
+
+/* This value might differs per partition */
 static uint64_t gfx_v9_4_3_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
@@ -50,6 +326,241 @@ static uint64_t gfx_v9_4_3_get_gpu_clock_counter(struct amdgpu_device *adev)
 	return clock;
 }
 
+static void gfx_v9_4_3_free_microcode(struct amdgpu_device *adev)
+{
+	amdgpu_ucode_release(&adev->gfx.pfp_fw);
+	amdgpu_ucode_release(&adev->gfx.me_fw);
+	amdgpu_ucode_release(&adev->gfx.ce_fw);
+	amdgpu_ucode_release(&adev->gfx.rlc_fw);
+	amdgpu_ucode_release(&adev->gfx.mec_fw);
+	amdgpu_ucode_release(&adev->gfx.mec2_fw);
+
+	kfree(adev->gfx.rlc.register_list_format);
+}
+
+static int gfx_v9_4_3_init_rlc_microcode(struct amdgpu_device *adev,
+					  const char *chip_name)
+{
+	char fw_name[30];
+	int err;
+	const struct rlc_firmware_header_v2_0 *rlc_hdr;
+	uint16_t version_major;
+	uint16_t version_minor;
+
+	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name);
+
+	err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name);
+	if (err)
+		goto out;
+	rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data;
+
+	version_major = le16_to_cpu(rlc_hdr->header.header_version_major);
+	version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor);
+	err = amdgpu_gfx_rlc_init_microcode(adev, version_major, version_minor);
+out:
+	if (err)
+		amdgpu_ucode_release(&adev->gfx.rlc_fw);
+
+	return err;
+}
+
+static bool gfx_v9_4_3_should_disable_gfxoff(struct pci_dev *pdev)
+{
+	return true;
+}
+
+static void gfx_v9_4_3_check_if_need_gfxoff(struct amdgpu_device *adev)
+{
+	if (gfx_v9_4_3_should_disable_gfxoff(adev->pdev))
+		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
+}
+
+static int gfx_v9_4_3_init_cp_compute_microcode(struct amdgpu_device *adev,
+					  const char *chip_name)
+{
+	char fw_name[30];
+	int err;
+
+	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name);
+
+	err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name);
+	if (err)
+		goto out;
+	amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1);
+	amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1_JT);
+
+	adev->gfx.mec2_fw_version = adev->gfx.mec_fw_version;
+	adev->gfx.mec2_feature_version = adev->gfx.mec_feature_version;
+
+	gfx_v9_4_3_check_if_need_gfxoff(adev);
+
+out:
+	if (err)
+		amdgpu_ucode_release(&adev->gfx.mec_fw);
+	return err;
+}
+
+static int gfx_v9_4_3_init_microcode(struct amdgpu_device *adev)
+{
+	const char *chip_name;
+	int r;
+
+	chip_name = "gc_9_4_3";
+
+	r = gfx_v9_4_3_init_rlc_microcode(adev, chip_name);
+	if (r)
+		return r;
+
+	r = gfx_v9_4_3_init_cp_compute_microcode(adev, chip_name);
+	if (r)
+		return r;
+
+	return r;
+}
+
+static u32 gfx_v9_4_3_get_csb_size(struct amdgpu_device *adev)
+{
+	u32 count = 0;
+	const struct cs_section_def *sect = NULL;
+	const struct cs_extent_def *ext = NULL;
+
+	/* begin clear state */
+	count += 2;
+	/* context control state */
+	count += 3;
+
+	for (sect = gfx9_cs_data; sect->section != NULL; ++sect) {
+		for (ext = sect->section; ext->extent != NULL; ++ext) {
+			if (sect->id == SECT_CONTEXT)
+				count += 2 + ext->reg_count;
+			else
+				return 0;
+		}
+	}
+
+	/* end clear state */
+	count += 2;
+	/* clear state */
+	count += 2;
+
+	return count;
+}
+
+static void gfx_v9_4_3_get_csb_buffer(struct amdgpu_device *adev,
+				    volatile u32 *buffer)
+{
+	u32 count = 0, i;
+	const struct cs_section_def *sect = NULL;
+	const struct cs_extent_def *ext = NULL;
+
+	if (adev->gfx.rlc.cs_data == NULL)
+		return;
+	if (buffer == NULL)
+		return;
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
+	buffer[count++] = cpu_to_le32(0x80000000);
+	buffer[count++] = cpu_to_le32(0x80000000);
+
+	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
+		for (ext = sect->section; ext->extent != NULL; ++ext) {
+			if (sect->id == SECT_CONTEXT) {
+				buffer[count++] =
+					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
+				buffer[count++] = cpu_to_le32(ext->reg_index -
+						PACKET3_SET_CONTEXT_REG_START);
+				for (i = 0; i < ext->reg_count; i++)
+					buffer[count++] = cpu_to_le32(ext->extent[i]);
+			} else {
+				return;
+			}
+		}
+	}
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
+	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
+
+	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
+	buffer[count++] = cpu_to_le32(0);
+}
+
+static void gfx_v9_4_3_mec_fini(struct amdgpu_device *adev)
+{
+	amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
+	amdgpu_bo_free_kernel(&adev->gfx.mec.mec_fw_obj, NULL, NULL);
+}
+
+static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
+{
+	int r, i;
+	u32 *hpd;
+	const __le32 *fw_data;
+	unsigned fw_size;
+	u32 *fw;
+	size_t mec_hpd_size;
+
+	const struct gfx_firmware_header_v1_0 *mec_hdr;
+
+	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+
+	/* take ownership of the relevant compute queues */
+	amdgpu_gfx_compute_queue_acquire(adev);
+	mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
+	if (mec_hpd_size) {
+		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
+					      AMDGPU_GEM_DOMAIN_VRAM,
+					      &adev->gfx.mec.hpd_eop_obj,
+					      &adev->gfx.mec.hpd_eop_gpu_addr,
+					      (void **)&hpd);
+		if (r) {
+			dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r);
+			gfx_v9_4_3_mec_fini(adev);
+			return r;
+		}
+
+		if (amdgpu_emu_mode == 1) {
+			for (i = 0; i < mec_hpd_size / 4; i++) {
+				memset((void *)(hpd + i), 0, 4);
+				if (i % 50 == 0)
+					msleep(1);
+			}
+		} else {
+			memset(hpd, 0, mec_hpd_size);
+		}
+
+		amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj);
+		amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj);
+	}
+
+	mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
+
+	fw_data = (const __le32 *)
+		(adev->gfx.mec_fw->data +
+		 le32_to_cpu(mec_hdr->header.ucode_array_offset_bytes));
+	fw_size = le32_to_cpu(mec_hdr->header.ucode_size_bytes);
+
+	r = amdgpu_bo_create_reserved(adev, mec_hdr->header.ucode_size_bytes,
+				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT,
+				      &adev->gfx.mec.mec_fw_obj,
+				      &adev->gfx.mec.mec_fw_gpu_addr,
+				      (void **)&fw);
+	if (r) {
+		dev_warn(adev->dev, "(%d) create mec firmware bo failed\n", r);
+		gfx_v9_4_3_mec_fini(adev);
+		return r;
+	}
+
+	memcpy(fw, fw_data, fw_size);
+
+	amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_obj);
+	amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj);
+
+	return 0;
+}
+
 static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
 				    u32 se_num,
 				    u32 sh_num,
@@ -150,6 +661,400 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 	soc15_grbm_select(adev, me, pipe, q, vm);
 }
 
+static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
+	.get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter,
+	.select_se_sh = &gfx_v9_4_3_select_se_sh,
+	.read_wave_data = &gfx_v9_4_3_read_wave_data,
+	.read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs,
+	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
+	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
+};
+
+static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
+{
+	u32 gb_addr_config;
+
+	adev->gfx.funcs = &gfx_v9_4_3_gfx_funcs;
+
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 3):
+		adev->gfx.config.max_hw_contexts = 8;
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0;
+		gb_addr_config = RREG32_SOC15(GC, 0, regGB_ADDR_CONFIG);
+		break;
+	default:
+		BUG();
+		break;
+	}
+
+	adev->gfx.config.gb_addr_config = gb_addr_config;
+
+	adev->gfx.config.gb_addr_config_fields.num_pipes = 1 <<
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					NUM_PIPES);
+
+	adev->gfx.config.max_tile_pipes =
+		adev->gfx.config.gb_addr_config_fields.num_pipes;
+
+	adev->gfx.config.gb_addr_config_fields.num_banks = 1 <<
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					NUM_BANKS);
+	adev->gfx.config.gb_addr_config_fields.max_compress_frags = 1 <<
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					MAX_COMPRESSED_FRAGS);
+	adev->gfx.config.gb_addr_config_fields.num_rb_per_se = 1 <<
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					NUM_RB_PER_SE);
+	adev->gfx.config.gb_addr_config_fields.num_se = 1 <<
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					NUM_SHADER_ENGINES);
+	adev->gfx.config.gb_addr_config_fields.pipe_interleave_size = 1 << (8 +
+			REG_GET_FIELD(
+					adev->gfx.config.gb_addr_config,
+					GB_ADDR_CONFIG,
+					PIPE_INTERLEAVE_SIZE));
+
+	return 0;
+}
+
+static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
+				      int mec, int pipe, int queue)
+{
+	unsigned irq_type;
+	struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id];
+	unsigned int hw_prio;
+
+	ring = &adev->gfx.compute_ring[ring_id];
+
+	/* mec0 is me1 */
+	ring->me = mec + 1;
+	ring->pipe = pipe;
+	ring->queue = queue;
+
+	ring->ring_obj = NULL;
+	ring->use_doorbell = true;
+	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
+	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
+				+ (ring_id * GFX9_MEC_HPD_SIZE);
+	ring->vm_hub = AMDGPU_GFXHUB_0;
+	sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
+
+	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
+		+ ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec)
+		+ ring->pipe;
+	hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring) ?
+			AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL;
+	/* type-2 packets are deprecated on MEC, use type-3 instead */
+	return amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, irq_type,
+				hw_prio, NULL);
+}
+
+static int gfx_v9_4_3_sw_init(void *handle)
+{
+	int i, j, k, r, ring_id;
+	struct amdgpu_kiq *kiq;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	adev->gfx.mec.num_mec = 2;
+	adev->gfx.mec.num_pipe_per_mec = 4;
+	adev->gfx.mec.num_queue_per_pipe = 8;
+
+	/* EOP Event */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_EOP_INTERRUPT, &adev->gfx.eop_irq);
+	if (r)
+		return r;
+
+	/* Privileged reg */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_PRIV_REG_FAULT,
+			      &adev->gfx.priv_reg_irq);
+	if (r)
+		return r;
+
+	/* Privileged inst */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_PRIV_INSTR_FAULT,
+			      &adev->gfx.priv_inst_irq);
+	if (r)
+		return r;
+
+	adev->gfx.gfx_current_status = AMDGPU_GFX_NORMAL_MODE;
+
+	r = adev->gfx.rlc.funcs->init(adev);
+	if (r) {
+		DRM_ERROR("Failed to init rlc BOs!\n");
+		return r;
+	}
+
+	r = gfx_v9_4_3_mec_init(adev);
+	if (r) {
+		DRM_ERROR("Failed to init MEC BOs!\n");
+		return r;
+	}
+
+	/* set up the compute queues - allocate horizontally across pipes */
+	ring_id = 0;
+	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
+		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
+			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
+				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+								     k, j))
+					continue;
+
+				r = gfx_v9_4_3_compute_ring_init(adev,
+							       ring_id,
+							       i, k, j);
+				if (r)
+					return r;
+
+				ring_id++;
+			}
+		}
+	}
+
+	r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, 0);
+	if (r) {
+		DRM_ERROR("Failed to init KIQ BOs!\n");
+		return r;
+	}
+
+	kiq = &adev->gfx.kiq[0];
+	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
+	if (r)
+		return r;
+
+	/* create MQD for all compute queues as wel as KIQ for SRIOV case */
+	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation), 0);
+	if (r)
+		return r;
+
+	r = gfx_v9_4_3_gpu_early_init(adev);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+static int gfx_v9_4_3_sw_fini(void *handle)
+{
+	int i;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++)
+		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
+
+	amdgpu_gfx_mqd_sw_fini(adev, 0);
+	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
+	amdgpu_gfx_kiq_fini(adev, 0);
+
+	gfx_v9_4_3_mec_fini(adev);
+	amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
+	gfx_v9_4_3_free_microcode(adev);
+
+	return 0;
+}
+
+static u32 gfx_v9_4_3_get_rb_active_bitmap(struct amdgpu_device *adev)
+{
+	u32 data, mask;
+
+	data = RREG32_SOC15(GC, 0, regCC_RB_BACKEND_DISABLE);
+	data |= RREG32_SOC15(GC, 0, regGC_USER_RB_BACKEND_DISABLE);
+
+	data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK;
+	data >>= GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT;
+
+	mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_backends_per_se /
+					 adev->gfx.config.max_sh_per_se);
+
+	return (~data) & mask;
+}
+
+static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev)
+{
+	int i, j;
+	u32 data;
+	u32 active_rbs = 0;
+	u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se /
+					adev->gfx.config.max_sh_per_se;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
+		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
+			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff);
+			data = gfx_v9_4_3_get_rb_active_bitmap(adev);
+			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
+					       rb_bitmap_width_per_sh);
+		}
+	}
+	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	adev->gfx.config.backend_enable_mask = active_rbs;
+	adev->gfx.config.num_rbs = hweight32(active_rbs);
+}
+
+#define DEFAULT_SH_MEM_BASES	(0x6000)
+static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev)
+{
+	int i;
+	uint32_t sh_mem_config;
+	uint32_t sh_mem_bases;
+
+	/*
+	 * Configure apertures:
+	 * LDS:         0x60000000'00000000 - 0x60000001'00000000 (4GB)
+	 * Scratch:     0x60000001'00000000 - 0x60000002'00000000 (4GB)
+	 * GPUVM:       0x60010000'00000000 - 0x60020000'00000000 (1TB)
+	 */
+	sh_mem_bases = DEFAULT_SH_MEM_BASES | (DEFAULT_SH_MEM_BASES << 16);
+
+	sh_mem_config = SH_MEM_ADDRESS_MODE_64 |
+			SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
+			SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
+
+	mutex_lock(&adev->srbm_mutex);
+	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
+		soc15_grbm_select(adev, 0, 0, 0, i);
+		/* CP and shaders */
+		WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, sh_mem_config);
+		WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, sh_mem_bases);
+	}
+	soc15_grbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+
+	/* Initialize all compute VMIDs to have no GDS, GWS, or OA
+	   acccess. These should be enabled by FW for target VMIDs. */
+	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_BASE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_SIZE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_GWS_VMID0, i, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_OA_VMID0, i, 0);
+	}
+}
+
+static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev)
+{
+	int vmid;
+
+	/*
+	 * Initialize all compute and user-gfx VMIDs to have no GDS, GWS, or OA
+	 * access. Compute VMIDs should be enabled by FW for target VMIDs,
+	 * the driver can enable them for graphics. VMID0 should maintain
+	 * access so that HWS firmware can save/restore entries.
+	 */
+	for (vmid = 1; vmid < AMDGPU_NUM_VMID; vmid++) {
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_BASE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_SIZE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_GWS_VMID0, vmid, 0);
+		WREG32_SOC15_OFFSET(GC, 0, regGDS_OA_VMID0, vmid, 0);
+	}
+}
+
+static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
+{
+	u32 tmp;
+	int i;
+
+	WREG32_FIELD15_PREREG(GC, 0, GRBM_CNTL, READ_TIMEOUT, 0xff);
+
+	gfx_v9_4_3_setup_rb(adev);
+	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
+	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, 0, regDB_DEBUG2);
+
+	/* XXX SH_MEM regs */
+	/* where to put LDS, scratch, GPUVM in FSA64 space */
+	mutex_lock(&adev->srbm_mutex);
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
+		soc15_grbm_select(adev, 0, 0, 0, i);
+		/* CP and shaders */
+		if (i == 0) {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+					    !!adev->gmc.noretry);
+			WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, tmp);
+			WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, 0);
+		} else {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+					    !!adev->gmc.noretry);
+			WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, tmp);
+			tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
+				(adev->gmc.private_aperture_start >> 48));
+			tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
+				(adev->gmc.shared_aperture_start >> 48));
+			WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, tmp);
+		}
+	}
+	soc15_grbm_select(adev, 0, 0, 0, 0);
+
+	mutex_unlock(&adev->srbm_mutex);
+
+	gfx_v9_4_3_init_compute_vmid(adev);
+	gfx_v9_4_3_init_gds_vmid(adev);
+}
+
+static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev)
+{
+	WREG32_FIELD15_PREREG(GC, 0, RLC_SRM_CNTL, SRM_ENABLE, 1);
+}
+
+static void gfx_v9_4_3_init_csb(struct amdgpu_device *adev)
+{
+	adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr);
+	/* csib */
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_ADDR_HI),
+			adev->gfx.rlc.clear_state_gpu_addr >> 32);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_ADDR_LO),
+			adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_LENGTH),
+			adev->gfx.rlc.clear_state_size);
+}
+
+static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev)
+{
+	gfx_v9_4_3_init_csb(adev);
+
+	/*
+	 * Rlc save restore list is workable since v2_1.
+	 * And it's needed by gfxoff feature.
+	 */
+	if (adev->gfx.rlc.is_rlc_v2_1)
+		gfx_v9_4_3_enable_save_restore_machine(adev);
+
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
+			      AMD_PG_SUPPORT_GFX_SMG |
+			      AMD_PG_SUPPORT_GFX_DMG |
+			      AMD_PG_SUPPORT_CP |
+			      AMD_PG_SUPPORT_GDS |
+			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
+		WREG32_SOC15(GC, 0, regRLC_JUMP_TABLE_RESTORE,
+		       adev->gfx.rlc.cp_table_gpu_addr >> 8);
+	}
+}
+
+void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	data = RREG32_SOC15(GC, 0, regCPC_PSP_DEBUG);
+	data |= CPC_PSP_DEBUG__UTCL2IUGPAOVERRIDE_MASK;
+	WREG32_SOC15(GC, 0, regCPC_PSP_DEBUG, data);
+}
+
 static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
 {
 	uint32_t rlc_setting;
@@ -189,6 +1094,20 @@ static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev)
 
 static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
 {
+	const struct cs_section_def *cs_data;
+	int r;
+
+	adev->gfx.rlc.cs_data = gfx9_cs_data;
+
+	cs_data = adev->gfx.rlc.cs_data;
+
+	if (cs_data) {
+		/* init clear state block */
+		r = amdgpu_gfx_rlc_init_csb(adev);
+		if (r)
+			return r;
+	}
+
 	/* init spm vmid with 0xf */
 	if (adev->gfx.rlc.funcs->update_spm_vmid)
 		adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf);
@@ -246,8 +1165,6 @@ static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0);
-	if (adev->gfx.num_gfx_rings)
-		tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, GFX_IDLE_INT_ENABLE, enable ? 1 : 0);
 
 	WREG32_SOC15(GC, 0, regCP_INT_CNTL_RING0, tmp);
 }
@@ -339,8 +1256,7 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 	/* disable CG */
 	WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, 0);
 
-	/* TODO: revisit pg function */
-	/* gfx_v9_4_3_init_pg(adev);*/
+	gfx_v9_4_3_init_pg(adev);
 
 	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
 		/* legacy rlc firmware loading */
@@ -407,24 +1323,1728 @@ static bool gfx_v9_4_3_is_rlcg_access_range(struct amdgpu_device *adev, u32 offs
 					ARRAY_SIZE(rlcg_access_gc_9_4_3));
 }
 
-const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
-	.get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter,
-	.select_se_sh = &gfx_v9_4_3_select_se_sh,
-	.read_wave_data = &gfx_v9_4_3_read_wave_data,
-	.read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs,
-	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
-	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
-};
+static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev, bool enable)
+{
+	if (enable) {
+		WREG32_SOC15_RLC(GC, 0, regCP_MEC_CNTL, 0);
+	} else {
+		WREG32_SOC15_RLC(GC, 0, regCP_MEC_CNTL,
+			(CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
+		adev->gfx.kiq[0].ring.sched.ready = false;
+	}
+	udelay(50);
+}
 
-const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = {
-	.is_rlc_enabled = gfx_v9_4_3_is_rlc_enabled,
-	.set_safe_mode = gfx_v9_4_3_set_safe_mode,
-	.unset_safe_mode = gfx_v9_4_3_unset_safe_mode,
-	.init = gfx_v9_4_3_rlc_init,
-	.resume = gfx_v9_4_3_rlc_resume,
-	.stop = gfx_v9_4_3_rlc_stop,
-	.reset = gfx_v9_4_3_rlc_reset,
-	.start = gfx_v9_4_3_rlc_start,
-	.update_spm_vmid = gfx_v9_4_3_update_spm_vmid,
-	.is_rlcg_access_range = gfx_v9_4_3_is_rlcg_access_range,
+static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev)
+{
+	const struct gfx_firmware_header_v1_0 *mec_hdr;
+	const __le32 *fw_data;
+	unsigned i;
+	u32 tmp;
+	u32 mec_ucode_addr_offset;
+	u32 mec_ucode_data_offset;
+
+	if (!adev->gfx.mec_fw)
+		return -EINVAL;
+
+	gfx_v9_4_3_cp_compute_enable(adev, false);
+
+	mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
+	amdgpu_ucode_print_gfx_hdr(&mec_hdr->header);
+
+	fw_data = (const __le32 *)
+		(adev->gfx.mec_fw->data +
+		 le32_to_cpu(mec_hdr->header.ucode_array_offset_bytes));
+	tmp = 0;
+	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, VMID, 0);
+	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
+	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_CNTL, tmp);
+
+	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_LO,
+		adev->gfx.mec.mec_fw_gpu_addr & 0xFFFFF000);
+	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_HI,
+		upper_32_bits(adev->gfx.mec.mec_fw_gpu_addr));
+
+	mec_ucode_addr_offset =
+		SOC15_REG_OFFSET(GC, 0, regCP_MEC_ME1_UCODE_ADDR);
+	mec_ucode_data_offset =
+		SOC15_REG_OFFSET(GC, 0, regCP_MEC_ME1_UCODE_DATA);
+
+	/* MEC1 */
+	WREG32(mec_ucode_addr_offset, mec_hdr->jt_offset);
+	for (i = 0; i < mec_hdr->jt_size; i++)
+		WREG32(mec_ucode_data_offset,
+		       le32_to_cpup(fw_data + mec_hdr->jt_offset + i));
+
+	WREG32(mec_ucode_addr_offset, adev->gfx.mec_fw_version);
+	/* Todo : Loading MEC2 firmware is only necessary if MEC2 should run different microcode than MEC1. */
+
+	return 0;
+}
+
+/* KIQ functions */
+static void gfx_v9_4_3_kiq_setting(struct amdgpu_ring *ring)
+{
+	uint32_t tmp;
+	struct amdgpu_device *adev = ring->adev;
+
+	/* tell RLC which is KIQ queue */
+	tmp = RREG32_SOC15(GC, 0, regRLC_CP_SCHEDULERS);
+	tmp &= 0xffffff00;
+	tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue);
+	WREG32_SOC15_RLC(GC, 0, regRLC_CP_SCHEDULERS, tmp);
+	tmp |= 0x80;
+	WREG32_SOC15_RLC(GC, 0, regRLC_CP_SCHEDULERS, tmp);
+}
+
+static void gfx_v9_4_3_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) {
+		if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring)) {
+			mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH;
+			mqd->cp_hqd_queue_priority =
+				AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM;
+		}
+	}
+}
+
+static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct v9_mqd *mqd = ring->mqd_ptr;
+	uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr;
+	uint32_t tmp;
+
+	mqd->header = 0xC0310800;
+	mqd->compute_pipelinestat_enable = 0x00000001;
+	mqd->compute_static_thread_mgmt_se0 = 0xffffffff;
+	mqd->compute_static_thread_mgmt_se1 = 0xffffffff;
+	mqd->compute_static_thread_mgmt_se2 = 0xffffffff;
+	mqd->compute_static_thread_mgmt_se3 = 0xffffffff;
+	mqd->compute_misc_reserved = 0x00000003;
+
+	mqd->dynamic_cu_mask_addr_lo =
+		lower_32_bits(ring->mqd_gpu_addr
+			      + offsetof(struct v9_mqd_allocation, dynamic_cu_mask));
+	mqd->dynamic_cu_mask_addr_hi =
+		upper_32_bits(ring->mqd_gpu_addr
+			      + offsetof(struct v9_mqd_allocation, dynamic_cu_mask));
+
+	eop_base_addr = ring->eop_gpu_addr >> 8;
+	mqd->cp_hqd_eop_base_addr_lo = eop_base_addr;
+	mqd->cp_hqd_eop_base_addr_hi = upper_32_bits(eop_base_addr);
+
+	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
+	tmp = RREG32_SOC15(GC, 0, regCP_HQD_EOP_CONTROL);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_EOP_CONTROL, EOP_SIZE,
+			(order_base_2(GFX9_MEC_HPD_SIZE / 4) - 1));
+
+	mqd->cp_hqd_eop_control = tmp;
+
+	/* enable doorbell? */
+	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL);
+
+	if (ring->use_doorbell) {
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
+				    DOORBELL_OFFSET, ring->doorbell_index);
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
+				    DOORBELL_EN, 1);
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
+				    DOORBELL_SOURCE, 0);
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
+				    DOORBELL_HIT, 0);
+	} else {
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
+					 DOORBELL_EN, 0);
+	}
+
+	mqd->cp_hqd_pq_doorbell_control = tmp;
+
+	/* disable the queue if it's active */
+	ring->wptr = 0;
+	mqd->cp_hqd_dequeue_request = 0;
+	mqd->cp_hqd_pq_rptr = 0;
+	mqd->cp_hqd_pq_wptr_lo = 0;
+	mqd->cp_hqd_pq_wptr_hi = 0;
+
+	/* set the pointer to the MQD */
+	mqd->cp_mqd_base_addr_lo = ring->mqd_gpu_addr & 0xfffffffc;
+	mqd->cp_mqd_base_addr_hi = upper_32_bits(ring->mqd_gpu_addr);
+
+	/* set MQD vmid to 0 */
+	tmp = RREG32_SOC15(GC, 0, regCP_MQD_CONTROL);
+	tmp = REG_SET_FIELD(tmp, CP_MQD_CONTROL, VMID, 0);
+	mqd->cp_mqd_control = tmp;
+
+	/* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */
+	hqd_gpu_addr = ring->gpu_addr >> 8;
+	mqd->cp_hqd_pq_base_lo = hqd_gpu_addr;
+	mqd->cp_hqd_pq_base_hi = upper_32_bits(hqd_gpu_addr);
+
+	/* set up the HQD, this is similar to CP_RB0_CNTL */
+	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PQ_CONTROL);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, QUEUE_SIZE,
+			    (order_base_2(ring->ring_size / 4) - 1));
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE,
+			((order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1) << 8));
+#ifdef __BIG_ENDIAN
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, ENDIAN_SWAP, 1);
+#endif
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 0);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, ROQ_PQ_IB_FLIP, 0);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	mqd->cp_hqd_pq_control = tmp;
+
+	/* set the wb address whether it's enabled or not */
+	wb_gpu_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4);
+	mqd->cp_hqd_pq_rptr_report_addr_lo = wb_gpu_addr & 0xfffffffc;
+	mqd->cp_hqd_pq_rptr_report_addr_hi =
+		upper_32_bits(wb_gpu_addr) & 0xffff;
+
+	/* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */
+	wb_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
+	mqd->cp_hqd_pq_wptr_poll_addr_lo = wb_gpu_addr & 0xfffffffc;
+	mqd->cp_hqd_pq_wptr_poll_addr_hi = upper_32_bits(wb_gpu_addr) & 0xffff;
+
+	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
+	ring->wptr = 0;
+	mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, 0, regCP_HQD_PQ_RPTR);
+
+	/* set the vmid for the queue */
+	mqd->cp_hqd_vmid = 0;
+
+	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PERSISTENT_STATE);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PERSISTENT_STATE, PRELOAD_SIZE, 0x53);
+	mqd->cp_hqd_persistent_state = tmp;
+
+	/* set MIN_IB_AVAIL_SIZE */
+	tmp = RREG32_SOC15(GC, 0, regCP_HQD_IB_CONTROL);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3);
+	mqd->cp_hqd_ib_control = tmp;
+
+	/* set static priority for a queue/ring */
+	gfx_v9_4_3_mqd_set_priority(ring, mqd);
+	mqd->cp_hqd_quantum = RREG32(regCP_HQD_QUANTUM);
+
+	/* map_queues packet doesn't need activate the queue,
+	 * so only kiq need set this field.
+	 */
+	if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ)
+		mqd->cp_hqd_active = 1;
+
+	return 0;
+}
+
+static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct v9_mqd *mqd = ring->mqd_ptr;
+	int j;
+
+	/* disable wptr polling */
+	WREG32_FIELD15_PREREG(GC, 0, CP_PQ_WPTR_POLL_CNTL, EN, 0);
+
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_BASE_ADDR,
+	       mqd->cp_hqd_eop_base_addr_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_BASE_ADDR_HI,
+	       mqd->cp_hqd_eop_base_addr_hi);
+
+	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_CONTROL,
+	       mqd->cp_hqd_eop_control);
+
+	/* enable doorbell? */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL,
+	       mqd->cp_hqd_pq_doorbell_control);
+
+	/* disable the queue if it's active */
+	if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) {
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1);
+		for (j = 0; j < adev->usec_timeout; j++) {
+			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
+				break;
+			udelay(1);
+		}
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST,
+		       mqd->cp_hqd_dequeue_request);
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR,
+		       mqd->cp_hqd_pq_rptr);
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO,
+		       mqd->cp_hqd_pq_wptr_lo);
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI,
+		       mqd->cp_hqd_pq_wptr_hi);
+	}
+
+	/* set the pointer to the MQD */
+	WREG32_SOC15_RLC(GC, 0, regCP_MQD_BASE_ADDR,
+	       mqd->cp_mqd_base_addr_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_MQD_BASE_ADDR_HI,
+	       mqd->cp_mqd_base_addr_hi);
+
+	/* set MQD vmid to 0 */
+	WREG32_SOC15_RLC(GC, 0, regCP_MQD_CONTROL,
+	       mqd->cp_mqd_control);
+
+	/* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_BASE,
+	       mqd->cp_hqd_pq_base_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_BASE_HI,
+	       mqd->cp_hqd_pq_base_hi);
+
+	/* set up the HQD, this is similar to CP_RB0_CNTL */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_CONTROL,
+	       mqd->cp_hqd_pq_control);
+
+	/* set the wb address whether it's enabled or not */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR_REPORT_ADDR,
+				mqd->cp_hqd_pq_rptr_report_addr_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR_REPORT_ADDR_HI,
+				mqd->cp_hqd_pq_rptr_report_addr_hi);
+
+	/* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR,
+	       mqd->cp_hqd_pq_wptr_poll_addr_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
+	       mqd->cp_hqd_pq_wptr_poll_addr_hi);
+
+	/* enable the doorbell if requested */
+	if (ring->use_doorbell) {
+		WREG32_SOC15(GC, 0, regCP_MEC_DOORBELL_RANGE_LOWER,
+					(adev->doorbell_index.kiq * 2) << 2);
+		WREG32_SOC15(GC, 0, regCP_MEC_DOORBELL_RANGE_UPPER,
+				(adev->doorbell_index.userqueue_end * 2) << 2);
+	}
+
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL,
+	       mqd->cp_hqd_pq_doorbell_control);
+
+	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO,
+	       mqd->cp_hqd_pq_wptr_lo);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI,
+	       mqd->cp_hqd_pq_wptr_hi);
+
+	/* set the vmid for the queue */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_VMID, mqd->cp_hqd_vmid);
+
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PERSISTENT_STATE,
+	       mqd->cp_hqd_persistent_state);
+
+	/* activate the queue */
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_ACTIVE,
+	       mqd->cp_hqd_active);
+
+	if (ring->use_doorbell)
+		WREG32_FIELD15_PREREG(GC, 0, CP_PQ_STATUS, DOORBELL_ENABLE, 1);
+
+	return 0;
+}
+
+static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	int j;
+
+	/* disable the queue if it's active */
+	if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) {
+
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1);
+
+		for (j = 0; j < adev->usec_timeout; j++) {
+			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
+				break;
+			udelay(1);
+		}
+
+		if (j == AMDGPU_MAX_USEC_TIMEOUT) {
+			DRM_DEBUG("KIQ dequeue request failed.\n");
+
+			/* Manual disable if dequeue request times out */
+			WREG32_SOC15_RLC(GC, 0, regCP_HQD_ACTIVE, 0);
+		}
+
+		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST,
+		      0);
+	}
+
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_IQ_TIMER, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_IB_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PERSISTENT_STATE, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI, 0);
+	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO, 0);
+
+	return 0;
+}
+
+static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct v9_mqd *mqd = ring->mqd_ptr;
+	struct v9_mqd *tmp_mqd;
+
+	gfx_v9_4_3_kiq_setting(ring);
+
+	/* GPU could be in bad state during probe, driver trigger the reset
+	 * after load the SMU, in this case , the mqd is not be initialized.
+	 * driver need to re-init the mqd.
+	 * check mqd->cp_hqd_pq_control since this value should not be 0
+	 */
+	tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[0].mqd_backup;
+	if (amdgpu_in_reset(adev) && tmp_mqd->cp_hqd_pq_control) {
+		/* for GPU_RESET case , reset MQD to a clean status */
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct v9_mqd_allocation));
+
+		/* reset ring buffer */
+		ring->wptr = 0;
+		amdgpu_ring_clear_ring(ring);
+
+		mutex_lock(&adev->srbm_mutex);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		gfx_v9_4_3_kiq_init_register(ring);
+		soc15_grbm_select(adev, 0, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
+	} else {
+		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
+		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
+		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
+		mutex_lock(&adev->srbm_mutex);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		gfx_v9_4_3_mqd_init(ring);
+		gfx_v9_4_3_kiq_init_register(ring);
+		soc15_grbm_select(adev, 0, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
+
+		if (adev->gfx.kiq[0].mqd_backup)
+			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct v9_mqd_allocation));
+	}
+
+	return 0;
+}
+
+static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct v9_mqd *mqd = ring->mqd_ptr;
+	int mqd_idx = ring - &adev->gfx.compute_ring[0];
+	struct v9_mqd *tmp_mqd;
+
+	/* Same as above kiq init, driver need to re-init the mqd if mqd->cp_hqd_pq_control
+	 * is not be initialized before
+	 */
+	tmp_mqd = (struct v9_mqd *)adev->gfx.mec.mqd_backup[mqd_idx];
+
+	if (!tmp_mqd->cp_hqd_pq_control ||
+	    (!amdgpu_in_reset(adev) && !adev->in_suspend)) {
+		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
+		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
+		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
+		mutex_lock(&adev->srbm_mutex);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		gfx_v9_4_3_mqd_init(ring);
+		soc15_grbm_select(adev, 0, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
+
+		if (adev->gfx.mec.mqd_backup[mqd_idx])
+			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
+	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
+		/* reset MQD to a clean status */
+		if (adev->gfx.mec.mqd_backup[mqd_idx])
+			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
+
+		/* reset ring buffer */
+		ring->wptr = 0;
+		atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0);
+		amdgpu_ring_clear_ring(ring);
+	} else {
+		amdgpu_ring_clear_ring(ring);
+	}
+
+	return 0;
+}
+
+static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring;
+	int r;
+
+	ring = &adev->gfx.kiq[0].ring;
+
+	r = amdgpu_bo_reserve(ring->mqd_obj, false);
+	if (unlikely(r != 0))
+		return r;
+
+	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
+	if (unlikely(r != 0))
+		return r;
+
+	gfx_v9_4_3_kiq_init_queue(ring);
+	amdgpu_bo_kunmap(ring->mqd_obj);
+	ring->mqd_ptr = NULL;
+	amdgpu_bo_unreserve(ring->mqd_obj);
+	ring->sched.ready = true;
+	return 0;
+}
+
+static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring = NULL;
+	int r = 0, i;
+
+	gfx_v9_4_3_cp_compute_enable(adev, true);
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		ring = &adev->gfx.compute_ring[i];
+
+		r = amdgpu_bo_reserve(ring->mqd_obj, false);
+		if (unlikely(r != 0))
+			goto done;
+		r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
+		if (!r) {
+			r = gfx_v9_4_3_kcq_init_queue(ring);
+			amdgpu_bo_kunmap(ring->mqd_obj);
+			ring->mqd_ptr = NULL;
+		}
+		amdgpu_bo_unreserve(ring->mqd_obj);
+		if (r)
+			goto done;
+	}
+
+	r = amdgpu_gfx_enable_kcq(adev, 0);
+done:
+	return r;
+}
+
+static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
+{
+	int r, i;
+	struct amdgpu_ring *ring;
+
+	gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+
+	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+		gfx_v9_4_3_disable_gpa_mode(adev);
+
+		r = gfx_v9_4_3_cp_compute_load_microcode(adev);
+		if (r)
+			return r;
+	}
+
+	r = gfx_v9_4_3_kiq_resume(adev);
+	if (r)
+		return r;
+
+	r = gfx_v9_4_3_kcq_resume(adev);
+	if (r)
+		return r;
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		ring = &adev->gfx.compute_ring[i];
+		amdgpu_ring_test_helper(ring);
+	}
+
+	gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
+
+	return 0;
+}
+
+static void gfx_v9_4_3_cp_enable(struct amdgpu_device *adev, bool enable)
+{
+	gfx_v9_4_3_cp_compute_enable(adev, enable);
+}
+
+static int gfx_v9_4_3_hw_init(void *handle)
+{
+	int r;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	gfx_v9_4_3_init_golden_registers(adev);
+
+	gfx_v9_4_3_constants_init(adev);
+
+	r = adev->gfx.rlc.funcs->resume(adev);
+	if (r)
+		return r;
+
+	r = gfx_v9_4_3_cp_resume(adev);
+	if (r)
+		return r;
+
+	return r;
+}
+
+static int gfx_v9_4_3_hw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
+	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
+
+	if (amdgpu_gfx_disable_kcq(adev, 0))
+		DRM_ERROR("KCQ disable failed\n");
+
+	/* Use deinitialize sequence from CAIL when unbinding device from driver,
+	 * otherwise KIQ is hanging when binding back
+	 */
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
+		mutex_lock(&adev->srbm_mutex);
+		soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me,
+				adev->gfx.kiq[0].ring.pipe,
+				adev->gfx.kiq[0].ring.queue, 0);
+		gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[0].ring);
+		soc15_grbm_select(adev, 0, 0, 0, 0);
+		mutex_unlock(&adev->srbm_mutex);
+	}
+
+	gfx_v9_4_3_cp_enable(adev, false);
+
+	/* Skip suspend with A+A reset */
+	if (adev->gmc.xgmi.connected_to_cpu && amdgpu_in_reset(adev)) {
+		dev_dbg(adev->dev, "Device in reset. Skipping RLC halt\n");
+		return 0;
+	}
+
+	adev->gfx.rlc.funcs->stop(adev);
+	return 0;
+}
+
+static int gfx_v9_4_3_suspend(void *handle)
+{
+	return gfx_v9_4_3_hw_fini(handle);
+}
+
+static int gfx_v9_4_3_resume(void *handle)
+{
+	return gfx_v9_4_3_hw_init(handle);
+}
+
+static bool gfx_v9_4_3_is_idle(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (REG_GET_FIELD(RREG32_SOC15(GC, 0, regGRBM_STATUS),
+				GRBM_STATUS, GUI_ACTIVE))
+		return false;
+	else
+		return true;
+}
+
+static int gfx_v9_4_3_wait_for_idle(void *handle)
+{
+	unsigned i;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	for (i = 0; i < adev->usec_timeout; i++) {
+		if (gfx_v9_4_3_is_idle(handle))
+			return 0;
+		udelay(1);
+	}
+	return -ETIMEDOUT;
+}
+
+static int gfx_v9_4_3_soft_reset(void *handle)
+{
+	u32 grbm_soft_reset = 0;
+	u32 tmp;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	/* GRBM_STATUS */
+	tmp = RREG32_SOC15(GC, 0, regGRBM_STATUS);
+	if (tmp & (GRBM_STATUS__PA_BUSY_MASK | GRBM_STATUS__SC_BUSY_MASK |
+		   GRBM_STATUS__BCI_BUSY_MASK | GRBM_STATUS__SX_BUSY_MASK |
+		   GRBM_STATUS__TA_BUSY_MASK | GRBM_STATUS__VGT_BUSY_MASK |
+		   GRBM_STATUS__DB_BUSY_MASK | GRBM_STATUS__CB_BUSY_MASK |
+		   GRBM_STATUS__GDS_BUSY_MASK | GRBM_STATUS__SPI_BUSY_MASK |
+		   GRBM_STATUS__IA_BUSY_MASK | GRBM_STATUS__IA_BUSY_NO_DMA_MASK)) {
+		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
+						GRBM_SOFT_RESET, SOFT_RESET_CP, 1);
+		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
+						GRBM_SOFT_RESET, SOFT_RESET_GFX, 1);
+	}
+
+	if (tmp & (GRBM_STATUS__CP_BUSY_MASK | GRBM_STATUS__CP_COHERENCY_BUSY_MASK)) {
+		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
+						GRBM_SOFT_RESET, SOFT_RESET_CP, 1);
+	}
+
+	/* GRBM_STATUS2 */
+	tmp = RREG32_SOC15(GC, 0, regGRBM_STATUS2);
+	if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY))
+		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
+						GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
+
+
+	if (grbm_soft_reset) {
+		/* stop the rlc */
+		adev->gfx.rlc.funcs->stop(adev);
+
+		/* Disable MEC parsing/prefetching */
+		gfx_v9_4_3_cp_compute_enable(adev, false);
+
+		if (grbm_soft_reset) {
+			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+			tmp |= grbm_soft_reset;
+			dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+			WREG32_SOC15(GC, 0, regGRBM_SOFT_RESET, tmp);
+			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+
+			udelay(50);
+
+			tmp &= ~grbm_soft_reset;
+			WREG32_SOC15(GC, 0, regGRBM_SOFT_RESET, tmp);
+			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+		}
+
+		/* Wait a little for things to settle down */
+		udelay(50);
+	}
+	return 0;
+}
+
+static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring,
+					  uint32_t vmid,
+					  uint32_t gds_base, uint32_t gds_size,
+					  uint32_t gws_base, uint32_t gws_size,
+					  uint32_t oa_base, uint32_t oa_size)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	/* GDS Base */
+	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
+				   SOC15_REG_OFFSET(GC, 0, regGDS_VMID0_BASE) + 2 * vmid,
+				   gds_base);
+
+	/* GDS Size */
+	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
+				   SOC15_REG_OFFSET(GC, 0, regGDS_VMID0_SIZE) + 2 * vmid,
+				   gds_size);
+
+	/* GWS */
+	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
+				   SOC15_REG_OFFSET(GC, 0, regGDS_GWS_VMID0) + vmid,
+				   gws_size << GDS_GWS_VMID0__SIZE__SHIFT | gws_base);
+
+	/* OA */
+	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
+				   SOC15_REG_OFFSET(GC, 0, regGDS_OA_VMID0) + vmid,
+				   (1 << (oa_size + oa_base)) - (1 << oa_base));
+}
+
+static int gfx_v9_4_3_early_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
+					  AMDGPU_MAX_COMPUTE_RINGS);
+	gfx_v9_4_3_set_kiq_pm4_funcs(adev);
+	gfx_v9_4_3_set_ring_funcs(adev);
+	gfx_v9_4_3_set_irq_funcs(adev);
+	gfx_v9_4_3_set_gds_init(adev);
+	gfx_v9_4_3_set_rlc_funcs(adev);
+
+	return gfx_v9_4_3_init_microcode(adev);
+}
+
+static int gfx_v9_4_3_late_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0);
+	if (r)
+		return r;
+
+	r = amdgpu_irq_get(adev, &adev->gfx.priv_inst_irq, 0);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+						      bool enable)
+{
+	uint32_t data, def;
+
+	amdgpu_gfx_rlc_enter_safe_mode(adev);
+
+	/* It is disabled by HW by default */
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
+		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
+		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+
+		data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
+			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
+			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK);
+
+		/* only for Vega10 & Raven1 */
+		data |= RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK;
+
+		if (def != data)
+			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+
+		/* MGLS is a global flag to control all MGLS in GFX */
+		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) {
+			/* 2 - RLC memory Light sleep */
+			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_RLC_LS) {
+				def = data = RREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL);
+				data |= RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
+				if (def != data)
+					WREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL, data);
+			}
+			/* 3 - CP memory Light sleep */
+			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) {
+				def = data = RREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL);
+				data |= CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
+				if (def != data)
+					WREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL, data);
+			}
+		}
+	} else {
+		/* 1 - MGCG_OVERRIDE */
+		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+
+		data |= (RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK |
+			 RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
+			 RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
+			 RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK);
+
+		if (def != data)
+			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+
+		/* 2 - disable MGLS in RLC */
+		data = RREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL);
+		if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) {
+			data &= ~RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
+			WREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL, data);
+		}
+
+		/* 3 - disable MGLS in CP */
+		data = RREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL);
+		if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) {
+			data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
+			WREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL, data);
+		}
+	}
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev);
+}
+
+static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
+						      bool enable)
+{
+	uint32_t def, data;
+
+	amdgpu_gfx_rlc_enter_safe_mode(adev);
+
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
+		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+		/* unset CGCG override */
+		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK;
+		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS)
+			data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK;
+		else
+			data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK;
+		/* update CGCG and CGLS override bits */
+		if (def != data)
+			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+
+		/* enable cgcg FSM(0x0000363F) */
+		def = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL);
+
+		if (adev->asic_type == CHIP_ARCTURUS)
+			data = (0x2000 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
+				RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK;
+		else
+			data = (0x36 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
+				RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK;
+		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS)
+			data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) |
+				RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK;
+		if (def != data)
+			WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, data);
+
+		/* set IDLE_POLL_COUNT(0x00900100) */
+		def = RREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_CNTL);
+		data = (0x0100 << CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT) |
+			(0x0090 << CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT);
+		if (def != data)
+			WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_CNTL, data);
+	} else {
+		def = data = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL);
+		/* reset CGCG/CGLS bits */
+		data &= ~(RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK);
+		/* disable cgcg and cgls in FSM */
+		if (def != data)
+			WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, data);
+	}
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev);
+}
+
+static int gfx_v9_4_3_update_gfx_clock_gating(struct amdgpu_device *adev,
+					    bool enable)
+{
+	if (enable) {
+		/* CGCG/CGLS should be enabled after MGCG/MGLS
+		 * ===  MGCG + MGLS ===
+		 */
+		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable);
+		/* ===  CGCG + CGLS === */
+		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable);
+	} else {
+		/* CGCG/CGLS should be disabled before MGCG/MGLS
+		 * ===  CGCG + CGLS ===
+		 */
+		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable);
+		/* ===  MGCG + MGLS === */
+		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable);
+	}
+	return 0;
+}
+
+static const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = {
+	.is_rlc_enabled = gfx_v9_4_3_is_rlc_enabled,
+	.set_safe_mode = gfx_v9_4_3_set_safe_mode,
+	.unset_safe_mode = gfx_v9_4_3_unset_safe_mode,
+	.init = gfx_v9_4_3_rlc_init,
+	.get_csb_size = gfx_v9_4_3_get_csb_size,
+	.get_csb_buffer = gfx_v9_4_3_get_csb_buffer,
+	.resume = gfx_v9_4_3_rlc_resume,
+	.stop = gfx_v9_4_3_rlc_stop,
+	.reset = gfx_v9_4_3_rlc_reset,
+	.start = gfx_v9_4_3_rlc_start,
+	.update_spm_vmid = gfx_v9_4_3_update_spm_vmid,
+	.is_rlcg_access_range = gfx_v9_4_3_is_rlcg_access_range,
+};
+
+static int gfx_v9_4_3_set_powergating_state(void *handle,
+					  enum amd_powergating_state state)
+{
+	return 0;
+}
+
+static int gfx_v9_4_3_set_clockgating_state(void *handle,
+					  enum amd_clockgating_state state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 3):
+		gfx_v9_4_3_update_gfx_clock_gating(adev,
+						 state == AMD_CG_STATE_GATE);
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+static void gfx_v9_4_3_get_clockgating_state(void *handle, u64 *flags)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int data;
+
+	if (amdgpu_sriov_vf(adev))
+		*flags = 0;
+
+	/* AMD_CG_SUPPORT_GFX_MGCG */
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_CGTT_MGCG_OVERRIDE));
+	if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK))
+		*flags |= AMD_CG_SUPPORT_GFX_MGCG;
+
+	/* AMD_CG_SUPPORT_GFX_CGCG */
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_CGCG_CGLS_CTRL));
+	if (data & RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_GFX_CGCG;
+
+	/* AMD_CG_SUPPORT_GFX_CGLS */
+	if (data & RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_GFX_CGLS;
+
+	/* AMD_CG_SUPPORT_GFX_RLC_LS */
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_MEM_SLP_CNTL));
+	if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_GFX_RLC_LS | AMD_CG_SUPPORT_GFX_MGLS;
+
+	/* AMD_CG_SUPPORT_GFX_CP_LS */
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regCP_MEM_SLP_CNTL));
+	if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK)
+		*flags |= AMD_CG_SUPPORT_GFX_CP_LS | AMD_CG_SUPPORT_GFX_MGLS;
+}
+
+static void gfx_v9_4_3_ring_emit_hdp_flush(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	u32 ref_and_mask, reg_mem_engine;
+	const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio.hdp_flush_reg;
+
+	if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) {
+		switch (ring->me) {
+		case 1:
+			ref_and_mask = nbio_hf_reg->ref_and_mask_cp2 << ring->pipe;
+			break;
+		case 2:
+			ref_and_mask = nbio_hf_reg->ref_and_mask_cp6 << ring->pipe;
+			break;
+		default:
+			return;
+		}
+		reg_mem_engine = 0;
+	} else {
+		ref_and_mask = nbio_hf_reg->ref_and_mask_cp0;
+		reg_mem_engine = 1; /* pfp */
+	}
+
+	gfx_v9_4_3_wait_reg_mem(ring, reg_mem_engine, 0, 1,
+			      adev->nbio.funcs->get_hdp_flush_req_offset(adev),
+			      adev->nbio.funcs->get_hdp_flush_done_offset(adev),
+			      ref_and_mask, ref_and_mask, 0x20);
+}
+
+static void gfx_v9_4_3_ring_emit_ib_compute(struct amdgpu_ring *ring,
+					  struct amdgpu_job *job,
+					  struct amdgpu_ib *ib,
+					  uint32_t flags)
+{
+	unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+	u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
+
+	/* Currently, there is a high possibility to get wave ID mismatch
+	 * between ME and GDS, leading to a hw deadlock, because ME generates
+	 * different wave IDs than the GDS expects. This situation happens
+	 * randomly when at least 5 compute pipes use GDS ordered append.
+	 * The wave IDs generated by ME are also wrong after suspend/resume.
+	 * Those are probably bugs somewhere else in the kernel driver.
+	 *
+	 * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and
+	 * GDS to 0 for this ring (me/pipe).
+	 */
+	if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) {
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+		amdgpu_ring_write(ring, regGDS_COMPUTE_MAX_WAVE_ID);
+		amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id);
+	}
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
+	amdgpu_ring_write(ring,
+#ifdef __BIG_ENDIAN
+				(2 << 0) |
+#endif
+				lower_32_bits(ib->gpu_addr));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+	amdgpu_ring_write(ring, control);
+}
+
+static void gfx_v9_4_3_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
+				     u64 seq, unsigned flags)
+{
+	bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT;
+	bool int_sel = flags & AMDGPU_FENCE_FLAG_INT;
+	bool writeback = flags & AMDGPU_FENCE_FLAG_TC_WB_ONLY;
+
+	/* RELEASE_MEM - flush caches, send int */
+	amdgpu_ring_write(ring, PACKET3(PACKET3_RELEASE_MEM, 6));
+	amdgpu_ring_write(ring, ((writeback ? (EOP_TC_WB_ACTION_EN |
+					       EOP_TC_NC_ACTION_EN) :
+					      (EOP_TCL1_ACTION_EN |
+					       EOP_TC_ACTION_EN |
+					       EOP_TC_WB_ACTION_EN |
+					       EOP_TC_MD_ACTION_EN)) |
+				 EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) |
+				 EVENT_INDEX(5)));
+	amdgpu_ring_write(ring, DATA_SEL(write64bit ? 2 : 1) | INT_SEL(int_sel ? 2 : 0));
+
+	/*
+	 * the address should be Qword aligned if 64bit write, Dword
+	 * aligned if only send 32bit data low (discard data high)
+	 */
+	if (write64bit)
+		BUG_ON(addr & 0x7);
+	else
+		BUG_ON(addr & 0x3);
+	amdgpu_ring_write(ring, lower_32_bits(addr));
+	amdgpu_ring_write(ring, upper_32_bits(addr));
+	amdgpu_ring_write(ring, lower_32_bits(seq));
+	amdgpu_ring_write(ring, upper_32_bits(seq));
+	amdgpu_ring_write(ring, 0);
+}
+
+static void gfx_v9_4_3_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX);
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	gfx_v9_4_3_wait_reg_mem(ring, usepfp, 1, 0,
+			      lower_32_bits(addr), upper_32_bits(addr),
+			      seq, 0xffffffff, 4);
+}
+
+static void gfx_v9_4_3_ring_emit_vm_flush(struct amdgpu_ring *ring,
+					unsigned vmid, uint64_t pd_addr)
+{
+	amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
+}
+
+static u64 gfx_v9_4_3_ring_get_rptr_compute(struct amdgpu_ring *ring)
+{
+	return ring->adev->wb.wb[ring->rptr_offs]; /* gfx9 hardware is 32bit rptr */
+}
+
+static u64 gfx_v9_4_3_ring_get_wptr_compute(struct amdgpu_ring *ring)
+{
+	u64 wptr;
+
+	/* XXX check if swapping is necessary on BE */
+	if (ring->use_doorbell)
+		wptr = atomic64_read((atomic64_t *)&ring->adev->wb.wb[ring->wptr_offs]);
+	else
+		BUG();
+	return wptr;
+}
+
+static void gfx_v9_4_3_ring_set_wptr_compute(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	/* XXX check if swapping is necessary on BE */
+	if (ring->use_doorbell) {
+		atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], ring->wptr);
+		WDOORBELL64(ring->doorbell_index, ring->wptr);
+	} else {
+		BUG(); /* only DOORBELL method supported on gfx9 now */
+	}
+}
+
+static void gfx_v9_4_3_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
+					 u64 seq, unsigned int flags)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	/* we only allocate 32bit for each seq wb address */
+	BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
+
+	/* write fence seq to the "addr" */
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+				 WRITE_DATA_DST_SEL(5) | WR_CONFIRM));
+	amdgpu_ring_write(ring, lower_32_bits(addr));
+	amdgpu_ring_write(ring, upper_32_bits(addr));
+	amdgpu_ring_write(ring, lower_32_bits(seq));
+
+	if (flags & AMDGPU_FENCE_FLAG_INT) {
+		/* set register to trigger INT */
+		amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+		amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+					 WRITE_DATA_DST_SEL(0) | WR_CONFIRM));
+		amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, 0, regCPC_INT_STATUS));
+		amdgpu_ring_write(ring, 0);
+		amdgpu_ring_write(ring, 0x20000000); /* src_id is 178 */
+	}
+}
+
+static void gfx_v9_4_3_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg,
+				    uint32_t reg_val_offs)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4));
+	amdgpu_ring_write(ring, 0 |	/* src: register*/
+				(5 << 8) |	/* dst: memory */
+				(1 << 20));	/* write confirm */
+	amdgpu_ring_write(ring, reg);
+	amdgpu_ring_write(ring, 0);
+	amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr +
+				reg_val_offs * 4));
+	amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr +
+				reg_val_offs * 4));
+}
+
+static void gfx_v9_4_3_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg,
+				    uint32_t val)
+{
+	uint32_t cmd = 0;
+
+	switch (ring->funcs->type) {
+	case AMDGPU_RING_TYPE_GFX:
+		cmd = WRITE_DATA_ENGINE_SEL(1) | WR_CONFIRM;
+		break;
+	case AMDGPU_RING_TYPE_KIQ:
+		cmd = (1 << 16); /* no inc addr */
+		break;
+	default:
+		cmd = WR_CONFIRM;
+		break;
+	}
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, cmd);
+	amdgpu_ring_write(ring, reg);
+	amdgpu_ring_write(ring, 0);
+	amdgpu_ring_write(ring, val);
+}
+
+static void gfx_v9_4_3_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
+					uint32_t val, uint32_t mask)
+{
+	gfx_v9_4_3_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20);
+}
+
+static void gfx_v9_4_3_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
+						  uint32_t reg0, uint32_t reg1,
+						  uint32_t ref, uint32_t mask)
+{
+	amdgpu_ring_emit_reg_write_reg_wait_helper(ring, reg0, reg1,
+						   ref, mask);
+}
+
+static void gfx_v9_4_3_set_compute_eop_interrupt_state(struct amdgpu_device *adev,
+						     int me, int pipe,
+						     enum amdgpu_interrupt_state state)
+{
+	u32 mec_int_cntl, mec_int_cntl_reg;
+
+	/*
+	 * amdgpu controls only the first MEC. That's why this function only
+	 * handles the setting of interrupts for this specific MEC. All other
+	 * pipes' interrupts are set by amdkfd.
+	 */
+
+	if (me == 1) {
+		switch (pipe) {
+		case 0:
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE0_INT_CNTL);
+			break;
+		case 1:
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE1_INT_CNTL);
+			break;
+		case 2:
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE2_INT_CNTL);
+			break;
+		case 3:
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE3_INT_CNTL);
+			break;
+		default:
+			DRM_DEBUG("invalid pipe %d\n", pipe);
+			return;
+		}
+	} else {
+		DRM_DEBUG("invalid me %d\n", me);
+		return;
+	}
+
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+		mec_int_cntl = RREG32(mec_int_cntl_reg);
+		mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL,
+					     TIME_STAMP_INT_ENABLE, 0);
+		WREG32(mec_int_cntl_reg, mec_int_cntl);
+		break;
+	case AMDGPU_IRQ_STATE_ENABLE:
+		mec_int_cntl = RREG32(mec_int_cntl_reg);
+		mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL,
+					     TIME_STAMP_INT_ENABLE, 1);
+		WREG32(mec_int_cntl_reg, mec_int_cntl);
+		break;
+	default:
+		break;
+	}
+}
+
+static int gfx_v9_4_3_set_priv_reg_fault_state(struct amdgpu_device *adev,
+					     struct amdgpu_irq_src *source,
+					     unsigned type,
+					     enum amdgpu_interrupt_state state)
+{
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+	case AMDGPU_IRQ_STATE_ENABLE:
+		WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0,
+			       PRIV_REG_INT_ENABLE,
+			       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int gfx_v9_4_3_set_priv_inst_fault_state(struct amdgpu_device *adev,
+					      struct amdgpu_irq_src *source,
+					      unsigned type,
+					      enum amdgpu_interrupt_state state)
+{
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+	case AMDGPU_IRQ_STATE_ENABLE:
+		WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0,
+			       PRIV_INSTR_INT_ENABLE,
+			       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
+					    struct amdgpu_irq_src *src,
+					    unsigned type,
+					    enum amdgpu_interrupt_state state)
+{
+	switch (type) {
+	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 0, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 1, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 2, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 3, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 0, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 1, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 2, state);
+		break;
+	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP:
+		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 3, state);
+		break;
+	default:
+		break;
+	}
+	return 0;
+}
+
+static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
+			    struct amdgpu_irq_src *source,
+			    struct amdgpu_iv_entry *entry)
+{
+	int i;
+	u8 me_id, pipe_id, queue_id;
+	struct amdgpu_ring *ring;
+
+	DRM_DEBUG("IH: CP EOP\n");
+	me_id = (entry->ring_id & 0x0c) >> 2;
+	pipe_id = (entry->ring_id & 0x03) >> 0;
+	queue_id = (entry->ring_id & 0x70) >> 4;
+
+	switch (me_id) {
+	case 0:
+	case 1:
+	case 2:
+		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+			ring = &adev->gfx.compute_ring[i];
+			/* Per-queue interrupt is supported for MEC starting from VI.
+			  * The interrupt can only be enabled/disabled per pipe instead of per queue.
+			  */
+			if ((ring->me == me_id) && (ring->pipe == pipe_id) && (ring->queue == queue_id))
+				amdgpu_fence_process(ring);
+		}
+		break;
+	}
+	return 0;
+}
+
+static void gfx_v9_4_3_fault(struct amdgpu_device *adev,
+			   struct amdgpu_iv_entry *entry)
+{
+	u8 me_id, pipe_id, queue_id;
+	struct amdgpu_ring *ring;
+	int i;
+
+	me_id = (entry->ring_id & 0x0c) >> 2;
+	pipe_id = (entry->ring_id & 0x03) >> 0;
+	queue_id = (entry->ring_id & 0x70) >> 4;
+
+	switch (me_id) {
+	case 0:
+	case 1:
+	case 2:
+		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+			ring = &adev->gfx.compute_ring[i];
+			if (ring->me == me_id && ring->pipe == pipe_id &&
+			    ring->queue == queue_id)
+				drm_sched_fault(&ring->sched);
+		}
+		break;
+	}
+}
+
+static int gfx_v9_4_3_priv_reg_irq(struct amdgpu_device *adev,
+				 struct amdgpu_irq_src *source,
+				 struct amdgpu_iv_entry *entry)
+{
+	DRM_ERROR("Illegal register access in command stream\n");
+	gfx_v9_4_3_fault(adev, entry);
+	return 0;
+}
+
+static int gfx_v9_4_3_priv_inst_irq(struct amdgpu_device *adev,
+				  struct amdgpu_irq_src *source,
+				  struct amdgpu_iv_entry *entry)
+{
+	DRM_ERROR("Illegal instruction in command stream\n");
+	gfx_v9_4_3_fault(adev, entry);
+	return 0;
+}
+
+static void gfx_v9_4_3_emit_mem_sync(struct amdgpu_ring *ring)
+{
+	const unsigned int cp_coher_cntl =
+			PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_ICACHE_ACTION_ENA(1) |
+			PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_ACTION_ENA(1) |
+			PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_ACTION_ENA(1) |
+			PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TCL1_ACTION_ENA(1) |
+			PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_WB_ACTION_ENA(1);
+
+	/* ACQUIRE_MEM -make one or more surfaces valid for use by the subsequent operations */
+	amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 5));
+	amdgpu_ring_write(ring, cp_coher_cntl); /* CP_COHER_CNTL */
+	amdgpu_ring_write(ring, 0xffffffff);  /* CP_COHER_SIZE */
+	amdgpu_ring_write(ring, 0xffffff);  /* CP_COHER_SIZE_HI */
+	amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */
+	amdgpu_ring_write(ring, 0);  /* CP_COHER_BASE_HI */
+	amdgpu_ring_write(ring, 0x0000000A); /* POLL_INTERVAL */
+}
+
+static void gfx_v9_4_3_emit_wave_limit_cs(struct amdgpu_ring *ring,
+					uint32_t pipe, bool enable)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t val;
+	uint32_t wcl_cs_reg;
+
+	/* regSPI_WCL_PIPE_PERCENT_CS[0-7]_DEFAULT values are same */
+	val = enable ? 0x1 : 0x7f;
+
+	switch (pipe) {
+	case 0:
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS0);
+		break;
+	case 1:
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS1);
+		break;
+	case 2:
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS2);
+		break;
+	case 3:
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS3);
+		break;
+	default:
+		DRM_DEBUG("invalid pipe %d\n", pipe);
+		return;
+	}
+
+	amdgpu_ring_emit_wreg(ring, wcl_cs_reg, val);
+
+}
+static void gfx_v9_4_3_emit_wave_limit(struct amdgpu_ring *ring, bool enable)
+{
+	struct amdgpu_device *adev = ring->adev;
+	uint32_t val;
+	int i;
+
+	/* regSPI_WCL_PIPE_PERCENT_GFX is 7 bit multiplier register to limit
+	 * number of gfx waves. Setting 5 bit will make sure gfx only gets
+	 * around 25% of gpu resources.
+	 */
+	val = enable ? 0x1f : 0x07ffffff;
+	amdgpu_ring_emit_wreg(ring,
+			      SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_GFX),
+			      val);
+
+	/* Restrict waves for normal/low priority compute queues as well
+	 * to get best QoS for high priority compute jobs.
+	 *
+	 * amdgpu controls only 1st ME(0-3 CS pipes).
+	 */
+	for (i = 0; i < adev->gfx.mec.num_pipe_per_mec; i++) {
+		if (i != ring->pipe)
+			gfx_v9_4_3_emit_wave_limit_cs(ring, i, enable);
+
+	}
+}
+
+static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = {
+	.name = "gfx_v9_4_3",
+	.early_init = gfx_v9_4_3_early_init,
+	.late_init = gfx_v9_4_3_late_init,
+	.sw_init = gfx_v9_4_3_sw_init,
+	.sw_fini = gfx_v9_4_3_sw_fini,
+	.hw_init = gfx_v9_4_3_hw_init,
+	.hw_fini = gfx_v9_4_3_hw_fini,
+	.suspend = gfx_v9_4_3_suspend,
+	.resume = gfx_v9_4_3_resume,
+	.is_idle = gfx_v9_4_3_is_idle,
+	.wait_for_idle = gfx_v9_4_3_wait_for_idle,
+	.soft_reset = gfx_v9_4_3_soft_reset,
+	.set_clockgating_state = gfx_v9_4_3_set_clockgating_state,
+	.set_powergating_state = gfx_v9_4_3_set_powergating_state,
+	.get_clockgating_state = gfx_v9_4_3_get_clockgating_state,
+};
+
+static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_compute = {
+	.type = AMDGPU_RING_TYPE_COMPUTE,
+	.align_mask = 0xff,
+	.nop = PACKET3(PACKET3_NOP, 0x3FFF),
+	.support_64bit_ptrs = true,
+	.get_rptr = gfx_v9_4_3_ring_get_rptr_compute,
+	.get_wptr = gfx_v9_4_3_ring_get_wptr_compute,
+	.set_wptr = gfx_v9_4_3_ring_set_wptr_compute,
+	.emit_frame_size =
+		20 + /* gfx_v9_4_3_ring_emit_gds_switch */
+		7 + /* gfx_v9_4_3_ring_emit_hdp_flush */
+		5 + /* hdp invalidate */
+		7 + /* gfx_v9_4_3_ring_emit_pipeline_sync */
+		SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 +
+		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 +
+		2 + /* gfx_v9_4_3_ring_emit_vm_flush */
+		8 + 8 + 8 + /* gfx_v9_4_3_ring_emit_fence x3 for user fence, vm fence */
+		7 + /* gfx_v9_4_3_emit_mem_sync */
+		5 + /* gfx_v9_4_3_emit_wave_limit for updating regSPI_WCL_PIPE_PERCENT_GFX register */
+		15, /* for updating 3 regSPI_WCL_PIPE_PERCENT_CS registers */
+	.emit_ib_size =	7, /* gfx_v9_4_3_ring_emit_ib_compute */
+	.emit_ib = gfx_v9_4_3_ring_emit_ib_compute,
+	.emit_fence = gfx_v9_4_3_ring_emit_fence,
+	.emit_pipeline_sync = gfx_v9_4_3_ring_emit_pipeline_sync,
+	.emit_vm_flush = gfx_v9_4_3_ring_emit_vm_flush,
+	.emit_gds_switch = gfx_v9_4_3_ring_emit_gds_switch,
+	.emit_hdp_flush = gfx_v9_4_3_ring_emit_hdp_flush,
+	.test_ring = gfx_v9_4_3_ring_test_ring,
+	.test_ib = gfx_v9_4_3_ring_test_ib,
+	.insert_nop = amdgpu_ring_insert_nop,
+	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.emit_wreg = gfx_v9_4_3_ring_emit_wreg,
+	.emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait,
+	.emit_reg_write_reg_wait = gfx_v9_4_3_ring_emit_reg_write_reg_wait,
+	.emit_mem_sync = gfx_v9_4_3_emit_mem_sync,
+	.emit_wave_limit = gfx_v9_4_3_emit_wave_limit,
+};
+
+static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_kiq = {
+	.type = AMDGPU_RING_TYPE_KIQ,
+	.align_mask = 0xff,
+	.nop = PACKET3(PACKET3_NOP, 0x3FFF),
+	.support_64bit_ptrs = true,
+	.get_rptr = gfx_v9_4_3_ring_get_rptr_compute,
+	.get_wptr = gfx_v9_4_3_ring_get_wptr_compute,
+	.set_wptr = gfx_v9_4_3_ring_set_wptr_compute,
+	.emit_frame_size =
+		20 + /* gfx_v9_4_3_ring_emit_gds_switch */
+		7 + /* gfx_v9_4_3_ring_emit_hdp_flush */
+		5 + /* hdp invalidate */
+		7 + /* gfx_v9_4_3_ring_emit_pipeline_sync */
+		SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 +
+		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 +
+		2 + /* gfx_v9_4_3_ring_emit_vm_flush */
+		8 + 8 + 8, /* gfx_v9_4_3_ring_emit_fence_kiq x3 for user fence, vm fence */
+	.emit_ib_size =	7, /* gfx_v9_4_3_ring_emit_ib_compute */
+	.emit_fence = gfx_v9_4_3_ring_emit_fence_kiq,
+	.test_ring = gfx_v9_4_3_ring_test_ring,
+	.insert_nop = amdgpu_ring_insert_nop,
+	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.emit_rreg = gfx_v9_4_3_ring_emit_rreg,
+	.emit_wreg = gfx_v9_4_3_ring_emit_wreg,
+	.emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait,
+	.emit_reg_write_reg_wait = gfx_v9_4_3_ring_emit_reg_write_reg_wait,
+};
+
+static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev)
+{
+	int i;
+
+	adev->gfx.kiq[0].ring.funcs = &gfx_v9_4_3_ring_funcs_kiq;
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++)
+		adev->gfx.compute_ring[i].funcs = &gfx_v9_4_3_ring_funcs_compute;
+}
+
+static const struct amdgpu_irq_src_funcs gfx_v9_4_3_eop_irq_funcs = {
+	.set = gfx_v9_4_3_set_eop_interrupt_state,
+	.process = gfx_v9_4_3_eop_irq,
+};
+
+static const struct amdgpu_irq_src_funcs gfx_v9_4_3_priv_reg_irq_funcs = {
+	.set = gfx_v9_4_3_set_priv_reg_fault_state,
+	.process = gfx_v9_4_3_priv_reg_irq,
+};
+
+static const struct amdgpu_irq_src_funcs gfx_v9_4_3_priv_inst_irq_funcs = {
+	.set = gfx_v9_4_3_set_priv_inst_fault_state,
+	.process = gfx_v9_4_3_priv_inst_irq,
+};
+
+static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev)
+{
+	adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST;
+	adev->gfx.eop_irq.funcs = &gfx_v9_4_3_eop_irq_funcs;
+
+	adev->gfx.priv_reg_irq.num_types = 1;
+	adev->gfx.priv_reg_irq.funcs = &gfx_v9_4_3_priv_reg_irq_funcs;
+
+	adev->gfx.priv_inst_irq.num_types = 1;
+	adev->gfx.priv_inst_irq.funcs = &gfx_v9_4_3_priv_inst_irq_funcs;
+}
+
+static void gfx_v9_4_3_set_rlc_funcs(struct amdgpu_device *adev)
+{
+	adev->gfx.rlc.funcs = &gfx_v9_4_3_rlc_funcs;
+}
+
+
+static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev)
+{
+	/* init asci gds info */
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 3):
+		/* 9.4.3 removed all the GDS internal memory,
+		 * only support GWS opcode in kernel, like barrier
+		 * semaphore.etc */
+		adev->gds.gds_size = 0;
+		break;
+	default:
+		adev->gds.gds_size = 0x10000;
+		break;
+	}
+
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 3):
+		/* deprecated for 9.4.3, no usage at all */
+		adev->gds.gds_compute_max_wave_id = 0;
+		break;
+	default:
+		/* this really depends on the chip */
+		adev->gds.gds_compute_max_wave_id = 0x7ff;
+		break;
+	}
+
+	adev->gds.gws_size = 64;
+	adev->gds.oa_size = 16;
+}
+
+static void gfx_v9_4_3_set_user_cu_inactive_bitmap(struct amdgpu_device *adev,
+						 u32 bitmap)
+{
+	u32 data;
+
+	if (!bitmap)
+		return;
+
+	data = bitmap << GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
+	data &= GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
+
+	WREG32_SOC15(GC, 0, regGC_USER_SHADER_ARRAY_CONFIG, data);
+}
+
+static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev)
+{
+	u32 data, mask;
+
+	data = RREG32_SOC15(GC, 0, regCC_GC_SHADER_ARRAY_CONFIG);
+	data |= RREG32_SOC15(GC, 0, regGC_USER_SHADER_ARRAY_CONFIG);
+
+	data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
+	data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
+
+	mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_cu_per_sh);
+
+	return (~data) & mask;
+}
+
+static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
+				 struct amdgpu_cu_info *cu_info)
+{
+	int i, j, k, counter, active_cu_number = 0;
+	u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0;
+	unsigned disable_masks[4 * 4];
+
+	if (!adev || !cu_info)
+		return -EINVAL;
+
+	/*
+	 * 16 comes from bitmap array size 4*4, and it can cover all gfx9 ASICs
+	 */
+	if (adev->gfx.config.max_shader_engines *
+		adev->gfx.config.max_sh_per_se > 16)
+		return -EINVAL;
+
+	amdgpu_gfx_parse_disable_cu(disable_masks,
+				    adev->gfx.config.max_shader_engines,
+				    adev->gfx.config.max_sh_per_se);
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
+		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
+			mask = 1;
+			ao_bitmap = 0;
+			counter = 0;
+			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v9_4_3_set_user_cu_inactive_bitmap(
+				adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]);
+			bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev);
+
+			/*
+			 * The bitmap(and ao_cu_bitmap) in cu_info structure is
+			 * 4x4 size array, and it's usually suitable for Vega
+			 * ASICs which has 4*2 SE/SH layout.
+			 * But for Arcturus, SE/SH layout is changed to 8*1.
+			 * To mostly reduce the impact, we make it compatible
+			 * with current bitmap array as below:
+			 *    SE4,SH0 --> bitmap[0][1]
+			 *    SE5,SH0 --> bitmap[1][1]
+			 *    SE6,SH0 --> bitmap[2][1]
+			 *    SE7,SH0 --> bitmap[3][1]
+			 */
+			cu_info->bitmap[i % 4][j + i / 4] = bitmap;
+
+			for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) {
+				if (bitmap & mask) {
+					if (counter < adev->gfx.config.max_cu_per_sh)
+						ao_bitmap |= mask;
+					counter++;
+				}
+				mask <<= 1;
+			}
+			active_cu_number += counter;
+			if (i < 2 && j < 2)
+				ao_cu_mask |= (ao_bitmap << (i * 16 + j * 8));
+			cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap;
+		}
+	}
+	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	cu_info->number = active_cu_number;
+	cu_info->ao_cu_mask = ao_cu_mask;
+	cu_info->simd_per_cu = NUM_SIMD_PER_CU;
+
+	return 0;
+}
+
+const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block = {
+	.type = AMD_IP_BLOCK_TYPE_GFX,
+	.major = 9,
+	.minor = 4,
+	.rev = 0,
+	.funcs = &gfx_v9_4_3_ip_funcs,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
index 84e69701b81a6..4b530f4c12953 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
@@ -24,7 +24,6 @@
 #ifndef __GFX_V9_4_3_H__
 #define __GFX_V9_4_3_H__
 
-extern const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs;
-extern const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs;
+extern const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block;
 
 #endif /* __GFX_V9_4_3_H__ */
-- 
GitLab


From 5aa998baab3360d0f1b93d6aff3df924045f956c Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 17 Nov 2021 16:28:51 +0800
Subject: [PATCH 0142/1791] drm/amdgpu: add xcc index argument to
 soc15_grbm_select

To support grbm select for multiple XCD case.

v2: unify naming style

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 10 +++----
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         | 26 +++++++++----------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c       |  4 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 26 +++++++++----------
 drivers/gpu/drm/amd/amdgpu/soc15.c            |  4 +--
 drivers/gpu/drm/amd/amdgpu/soc15.h            |  2 +-
 6 files changed, 36 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 03875b971ba65..ebb35633058c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -50,12 +50,12 @@ static void lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe,
 			uint32_t queue, uint32_t vmid)
 {
 	mutex_lock(&adev->srbm_mutex);
-	soc15_grbm_select(adev, mec, pipe, queue, vmid);
+	soc15_grbm_select(adev, mec, pipe, queue, vmid, 0);
 }
 
 static void unlock_srbm(struct amdgpu_device *adev)
 {
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 }
 
@@ -700,7 +700,7 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
 	*wave_cnt = 0;
 	pipe_idx = queue_idx / adev->gfx.mec.num_queue_per_pipe;
 	queue_slot = queue_idx % adev->gfx.mec.num_queue_per_pipe;
-	soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0);
+	soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0, 0);
 	reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) +
 			 queue_slot);
 	*wave_cnt = reg_val & SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK;
@@ -772,7 +772,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	DECLARE_BITMAP(cp_queue_bitmap, KGD_MAX_QUEUES);
 
 	lock_spi_csq_mutexes(adev);
-	soc15_grbm_select(adev, 1, 0, 0, 0);
+	soc15_grbm_select(adev, 1, 0, 0, 0, 0);
 
 	/*
 	 * Iterate through the shader engines and arrays of the device
@@ -821,7 +821,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	}
 
 	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	unlock_spi_csq_mutexes(adev);
 
 	/* Update the output parameters and return */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 62af92e5be51c..4939fd61355be 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1831,7 +1831,7 @@ static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 static void gfx_v9_0_select_me_pipe_q(struct amdgpu_device *adev,
 				  u32 me, u32 pipe, u32 q, u32 vm)
 {
-	soc15_grbm_select(adev, me, pipe, q, vm);
+	soc15_grbm_select(adev, me, pipe, q, vm, 0);
 }
 
 static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = {
@@ -2324,12 +2324,12 @@ static void gfx_v9_0_init_compute_vmid(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->srbm_mutex);
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i);
+		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		/* CP and shaders */
 		WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, sh_mem_config);
 		WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, sh_mem_bases);
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 
 	/* Initialize all compute VMIDs to have no GDS, GWS, or OA
@@ -2394,7 +2394,7 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev)
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i);
+		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		/* CP and shaders */
 		if (i == 0) {
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
@@ -2416,7 +2416,7 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev)
 			WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, tmp);
 		}
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 
 	mutex_unlock(&adev->srbm_mutex);
 
@@ -3540,9 +3540,9 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 		amdgpu_ring_clear_ring(ring);
 
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_0_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
@@ -3551,10 +3551,10 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 		if (amdgpu_sriov_vf(adev) && adev->in_suspend)
 			amdgpu_ring_clear_ring(ring);
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_0_mqd_init(ring);
 		gfx_v9_0_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.kiq[0].mqd_backup)
@@ -3582,9 +3582,9 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_0_mqd_init(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
@@ -3791,9 +3791,9 @@ static int gfx_v9_0_hw_fini(void *handle)
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me,
 				adev->gfx.kiq[0].ring.pipe,
-				adev->gfx.kiq[0].ring.queue, 0);
+				adev->gfx.kiq[0].ring.queue, 0, 0);
 		gfx_v9_0_kiq_fini_register(&adev->gfx.kiq[0].ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index 3a797424579c5..93438770ca1a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -761,7 +761,7 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev,
 
 	for (i = first_vmid; i < last_vmid; i++) {
 		data = 0;
-		soc15_grbm_select(adev, 0, 0, 0, i);
+		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
 		data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0);
 		data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE,
@@ -769,7 +769,7 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev,
 		WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_PER_VMID_CNTL), data);
 	}
 
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 212d48114da8d..12185e7aac4e4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -658,7 +658,7 @@ static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd
 static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 					u32 me, u32 pipe, u32 q, u32 vm)
 {
-	soc15_grbm_select(adev, me, pipe, q, vm);
+	soc15_grbm_select(adev, me, pipe, q, vm, 0);
 }
 
 static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
@@ -926,12 +926,12 @@ static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->srbm_mutex);
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i);
+		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		/* CP and shaders */
 		WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, sh_mem_config);
 		WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, sh_mem_bases);
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 
 	/* Initialize all compute VMIDs to have no GDS, GWS, or OA
@@ -977,7 +977,7 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i);
+		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		/* CP and shaders */
 		if (i == 0) {
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
@@ -999,7 +999,7 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 			WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, tmp);
 		}
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 
 	mutex_unlock(&adev->srbm_mutex);
 
@@ -1706,19 +1706,19 @@ static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring)
 		amdgpu_ring_clear_ring(ring);
 
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_4_3_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_4_3_mqd_init(ring);
 		gfx_v9_4_3_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.kiq[0].mqd_backup)
@@ -1746,9 +1746,9 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring)
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
 		gfx_v9_4_3_mqd_init(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
@@ -1896,9 +1896,9 @@ static int gfx_v9_4_3_hw_fini(void *handle)
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me,
 				adev->gfx.kiq[0].ring.pipe,
-				adev->gfx.kiq[0].ring.queue, 0);
+				adev->gfx.kiq[0].ring.queue, 0, 0);
 		gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[0].ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index bc5dd80f10c1f..4b79a8933476c 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -311,7 +311,7 @@ static u32 soc15_get_xclk(struct amdgpu_device *adev)
 
 
 void soc15_grbm_select(struct amdgpu_device *adev,
-		     u32 me, u32 pipe, u32 queue, u32 vmid)
+		     u32 me, u32 pipe, u32 queue, u32 vmid, int xcc_id)
 {
 	u32 grbm_gfx_cntl = 0;
 	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, PIPEID, pipe);
@@ -319,7 +319,7 @@ void soc15_grbm_select(struct amdgpu_device *adev,
 	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, VMID, vmid);
 	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, QUEUEID, queue);
 
-	WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_CNTL, grbm_gfx_cntl);
+	WREG32_SOC15_RLC_SHADOW(GC, xcc_id, mmGRBM_GFX_CNTL, grbm_gfx_cntl);
 }
 
 static void soc15_vga_set_state(struct amdgpu_device *adev, bool state)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index efc2a253e8dbe..2b41ee968dd10 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -100,7 +100,7 @@ struct soc15_ras_field_entry {
 #define SOC15_RAS_REG_FIELD_VAL(val, entry, field) SOC15_REG_FIELD_VAL((val), (entry).field##_count_mask, (entry).field##_count_shift)
 
 void soc15_grbm_select(struct amdgpu_device *adev,
-		    u32 me, u32 pipe, u32 queue, u32 vmid);
+		    u32 me, u32 pipe, u32 queue, u32 vmid, int xcc_id);
 void soc15_set_virt_ops(struct amdgpu_device *adev);
 
 void soc15_program_register_sequence(struct amdgpu_device *adev,
-- 
GitLab


From 6f917fdc934518401ff2e166e6db1f6ac1ef1078 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 17 Nov 2021 17:24:02 +0800
Subject: [PATCH 0143/1791] drm/amdgpu: add multi-XCC initial support in
 gfx_v9_4_3.c

Each XCD needs to be initialized respectively. The major changes are:

  1. add iteration to do rlc/kiq/kcq init/fini for each xcd
  2. load rlc/mec microcode to each xcd
  3. add argument to specify xcc index in initialization functions

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 721 +++++++++++++-----------
 1 file changed, 394 insertions(+), 327 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 12185e7aac4e4..56999bb7ac26a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -181,7 +181,9 @@ static const struct kiq_pm4_funcs gfx_v9_4_3_kiq_pm4_funcs = {
 
 static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 {
-	adev->gfx.kiq[0].pmf = &gfx_v9_4_3_kiq_pm4_funcs;
+	int i;
+	for (i = 0; i < adev->gfx.num_xcd; i++)
+		adev->gfx.kiq[i].pmf = &gfx_v9_4_3_kiq_pm4_funcs;
 }
 
 static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
@@ -504,7 +506,9 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 
 	const struct gfx_firmware_header_v1_0 *mec_hdr;
 
-	bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES);
+	for (i = 0; i < adev->gfx.num_xcd; i++)
+		bitmap_zero(adev->gfx.mec_bitmap[i].queue_bitmap,
+			AMDGPU_MAX_COMPUTE_QUEUES);
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
@@ -731,7 +735,7 @@ static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
 }
 
 static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
-				      int mec, int pipe, int queue)
+				        int xcc_id, int mec, int pipe, int queue)
 {
 	unsigned irq_type;
 	struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id];
@@ -740,6 +744,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	ring = &adev->gfx.compute_ring[ring_id];
 
 	/* mec0 is me1 */
+	ring->xcc_id = xcc_id;
 	ring->me = mec + 1;
 	ring->pipe = pipe;
 	ring->queue = queue;
@@ -750,7 +755,8 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
 	ring->vm_hub = AMDGPU_GFXHUB_0;
-	sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
+	sprintf(ring->name, "comp_%d.%d.%d.%d",
+			ring->xcc_id, ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
 		+ ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec)
@@ -764,7 +770,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 
 static int gfx_v9_4_3_sw_init(void *handle)
 {
-	int i, j, k, r, ring_id;
+	int i, j, k, r, ring_id, xcc_id;
 	struct amdgpu_kiq *kiq;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
@@ -805,39 +811,45 @@ static int gfx_v9_4_3_sw_init(void *handle)
 
 	/* set up the compute queues - allocate horizontally across pipes */
 	ring_id = 0;
-	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
-		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
-			for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
-				if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
-								     k, j))
-					continue;
-
-				r = gfx_v9_4_3_compute_ring_init(adev,
-							       ring_id,
-							       i, k, j);
-				if (r)
-					return r;
-
-				ring_id++;
+	for (xcc_id = 0; xcc_id < adev->gfx.num_xcd; xcc_id++) {
+
+		for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
+			for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
+				for (k = 0; k < adev->gfx.mec.num_pipe_per_mec;
+				     k++) {
+					if (!amdgpu_gfx_is_mec_queue_enabled(
+							adev, xcc_id, i, k, j))
+						continue;
+
+					r = gfx_v9_4_3_compute_ring_init(adev,
+								       ring_id,
+								       xcc_id,
+								       i, k, j);
+					if (r)
+						return r;
+
+					ring_id++;
+				}
 			}
 		}
-	}
 
-	r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, 0);
-	if (r) {
-		DRM_ERROR("Failed to init KIQ BOs!\n");
-		return r;
-	}
+		r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, xcc_id);
+		if (r) {
+			DRM_ERROR("Failed to init KIQ BOs!\n");
+			return r;
+		}
 
-	kiq = &adev->gfx.kiq[0];
-	r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0);
-	if (r)
-		return r;
+		kiq = &adev->gfx.kiq[xcc_id];
+		r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, xcc_id);
+		if (r)
+			return r;
 
-	/* create MQD for all compute queues as wel as KIQ for SRIOV case */
-	r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation), 0);
-	if (r)
-		return r;
+		/* create MQD for all compute queues as wel as KIQ for SRIOV case */
+		r = amdgpu_gfx_mqd_sw_init(adev,
+				sizeof(struct v9_mqd_allocation), xcc_id);
+		if (r)
+			return r;
+	}
 
 	r = gfx_v9_4_3_gpu_early_init(adev);
 	if (r)
@@ -851,12 +863,15 @@ static int gfx_v9_4_3_sw_fini(void *handle)
 	int i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	for (i = 0; i < adev->gfx.num_compute_rings; i++)
+	for (i = 0; i < adev->gfx.num_compute_rings *
+		adev->gfx.num_xcd; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	amdgpu_gfx_mqd_sw_fini(adev, 0);
-	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring);
-	amdgpu_gfx_kiq_fini(adev, 0);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		amdgpu_gfx_mqd_sw_fini(adev, i);
+		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[i].ring);
+		amdgpu_gfx_kiq_fini(adev, i);
+	}
 
 	gfx_v9_4_3_mec_fini(adev);
 	amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
@@ -881,7 +896,7 @@ static u32 gfx_v9_4_3_get_rb_active_bitmap(struct amdgpu_device *adev)
 	return (~data) & mask;
 }
 
-static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev)
+static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev, int xcc_id)
 {
 	int i, j;
 	u32 data;
@@ -906,7 +921,7 @@ static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev)
 }
 
 #define DEFAULT_SH_MEM_BASES	(0x6000)
-static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev)
+static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev, int xcc_id)
 {
 	int i;
 	uint32_t sh_mem_config;
@@ -926,25 +941,25 @@ static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->srbm_mutex);
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i, 0);
+		soc15_grbm_select(adev, 0, 0, 0, i, xcc_id);
 		/* CP and shaders */
-		WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, sh_mem_config);
-		WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, sh_mem_bases);
+		WREG32_SOC15_RLC(GC, xcc_id, regSH_MEM_CONFIG, sh_mem_config);
+		WREG32_SOC15_RLC(GC, xcc_id, regSH_MEM_BASES, sh_mem_bases);
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
 	mutex_unlock(&adev->srbm_mutex);
 
 	/* Initialize all compute VMIDs to have no GDS, GWS, or OA
 	   acccess. These should be enabled by FW for target VMIDs. */
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_BASE, 2 * i, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_SIZE, 2 * i, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_GWS_VMID0, i, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_OA_VMID0, i, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_BASE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_SIZE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_GWS_VMID0, i, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_OA_VMID0, i, 0);
 	}
 }
 
-static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev)
+static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 {
 	int vmid;
 
@@ -955,21 +970,23 @@ static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev)
 	 * access so that HWS firmware can save/restore entries.
 	 */
 	for (vmid = 1; vmid < AMDGPU_NUM_VMID; vmid++) {
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_BASE, 2 * vmid, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_SIZE, 2 * vmid, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_GWS_VMID0, vmid, 0);
-		WREG32_SOC15_OFFSET(GC, 0, regGDS_OA_VMID0, vmid, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_BASE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_SIZE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_GWS_VMID0, vmid, 0);
+		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_OA_VMID0, vmid, 0);
 	}
 }
 
 static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 {
 	u32 tmp;
-	int i;
+	int i, j;
 
-	WREG32_FIELD15_PREREG(GC, 0, GRBM_CNTL, READ_TIMEOUT, 0xff);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_FIELD15_PREREG(GC, i, GRBM_CNTL, READ_TIMEOUT, 0xff);
+		gfx_v9_4_3_setup_rb(adev, i);
+	}
 
-	gfx_v9_4_3_setup_rb(adev);
 	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
 	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, 0, regDB_DEBUG2);
 
@@ -977,63 +994,68 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i, 0);
-		/* CP and shaders */
-		if (i == 0) {
-			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
-					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
-			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-					    !!adev->gmc.noretry);
-			WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, tmp);
-			WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, 0);
-		} else {
-			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
-					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
-			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-					    !!adev->gmc.noretry);
-			WREG32_SOC15_RLC(GC, 0, regSH_MEM_CONFIG, tmp);
-			tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
-				(adev->gmc.private_aperture_start >> 48));
-			tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
-				(adev->gmc.shared_aperture_start >> 48));
-			WREG32_SOC15_RLC(GC, 0, regSH_MEM_BASES, tmp);
+		for (j = 0; j < adev->gfx.num_xcd; j++) {
+			soc15_grbm_select(adev, 0, 0, 0, i, j);
+			/* CP and shaders */
+			if (i == 0) {
+				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+						    !!adev->gmc.noretry);
+				WREG32_SOC15_RLC(GC, j, regSH_MEM_CONFIG, tmp);
+				WREG32_SOC15_RLC(GC, j, regSH_MEM_BASES, 0);
+			} else {
+				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+						    !!adev->gmc.noretry);
+				WREG32_SOC15_RLC(GC, j, regSH_MEM_CONFIG, tmp);
+				tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
+					(adev->gmc.private_aperture_start >> 48));
+				tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
+					(adev->gmc.shared_aperture_start >> 48));
+				WREG32_SOC15_RLC(GC, j, regSH_MEM_BASES, tmp);
+			}
 		}
 	}
 	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 
 	mutex_unlock(&adev->srbm_mutex);
 
-	gfx_v9_4_3_init_compute_vmid(adev);
-	gfx_v9_4_3_init_gds_vmid(adev);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		gfx_v9_4_3_init_compute_vmid(adev, i);
+		gfx_v9_4_3_init_gds_vmid(adev, i);
+	}
 }
 
-static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev)
+static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev,
+						   int xcc_id)
 {
-	WREG32_FIELD15_PREREG(GC, 0, RLC_SRM_CNTL, SRM_ENABLE, 1);
+	WREG32_FIELD15_PREREG(GC, xcc_id, RLC_SRM_CNTL, SRM_ENABLE, 1);
 }
 
-static void gfx_v9_4_3_init_csb(struct amdgpu_device *adev)
+static void gfx_v9_4_3_init_csb(struct amdgpu_device *adev, int xcc_id)
 {
 	adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr);
 	/* csib */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_ADDR_HI),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_ADDR_HI),
 			adev->gfx.rlc.clear_state_gpu_addr >> 32);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_ADDR_LO),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_ADDR_LO),
 			adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regRLC_CSIB_LENGTH),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_LENGTH),
 			adev->gfx.rlc.clear_state_size);
 }
 
-static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev)
+static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 {
-	gfx_v9_4_3_init_csb(adev);
+	gfx_v9_4_3_init_csb(adev, xcc_id);
 
 	/*
 	 * Rlc save restore list is workable since v2_1.
 	 * And it's needed by gfxoff feature.
 	 */
 	if (adev->gfx.rlc.is_rlc_v2_1)
-		gfx_v9_4_3_enable_save_restore_machine(adev);
+		gfx_v9_4_3_enable_save_restore_machine(adev, xcc_id);
 
 	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
 			      AMD_PG_SUPPORT_GFX_SMG |
@@ -1115,7 +1137,8 @@ static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev)
+static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
+					   int xcc_id)
 {
 	u32 i, j, k;
 	u32 mask;
@@ -1171,17 +1194,25 @@ static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
 
 static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 {
-	WREG32_FIELD15_PREREG(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 0);
-	gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
-	gfx_v9_4_3_wait_for_rlc_serdes(adev);
+	int i;
+
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 0);
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+		gfx_v9_4_3_wait_for_rlc_serdes(adev, i);
+	}
 }
 
 static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev)
 {
-	WREG32_FIELD15_PREREG(GC, 0, GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
-	udelay(50);
-	WREG32_FIELD15_PREREG(GC, 0, GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
-	udelay(50);
+	int i;
+
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
+		udelay(50);
+		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
+		udelay(50);
+	}
 }
 
 static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
@@ -1189,35 +1220,38 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 #ifdef AMDGPU_RLC_DEBUG_RETRY
 	u32 rlc_ucode_ver;
 #endif
+	int i;
 
-	WREG32_FIELD15_PREREG(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 1);
-	udelay(50);
-
-	/* carrizo do enable cp interrupt after cp inited */
-	if (!(adev->flags & AMD_IS_APU)) {
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 1);
 		udelay(50);
-	}
+
+		/* carrizo do enable cp interrupt after cp inited */
+		if (!(adev->flags & AMD_IS_APU)) {
+			gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
+			udelay(50);
+		}
 
 #ifdef AMDGPU_RLC_DEBUG_RETRY
-	/* RLC_GPM_GENERAL_6 : RLC Ucode version */
-	rlc_ucode_ver = RREG32_SOC15(GC, 0, regRLC_GPM_GENERAL_6);
-	if (rlc_ucode_ver == 0x108) {
-		dev_info(adev->dev,
-			 "Using rlc debug ucode. regRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n",
-			 rlc_ucode_ver, adev->gfx.rlc_fw_version);
-		/* RLC_GPM_TIMER_INT_3 : Timer interval in RefCLK cycles,
-		 * default is 0x9C4 to create a 100us interval */
-		WREG32_SOC15(GC, 0, regRLC_GPM_TIMER_INT_3, 0x9C4);
-		/* RLC_GPM_GENERAL_12 : Minimum gap between wptr and rptr
-		 * to disable the page fault retry interrupts, default is
-		 * 0x100 (256) */
-		WREG32_SOC15(GC, 0, regRLC_GPM_GENERAL_12, 0x100);
-	}
+		/* RLC_GPM_GENERAL_6 : RLC Ucode version */
+		rlc_ucode_ver = RREG32_SOC15(GC, i, regRLC_GPM_GENERAL_6);
+		if (rlc_ucode_ver == 0x108) {
+			dev_info(adev->dev,
+				 "Using rlc debug ucode. regRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n",
+				 rlc_ucode_ver, adev->gfx.rlc_fw_version);
+			/* RLC_GPM_TIMER_INT_3 : Timer interval in RefCLK cycles,
+			 * default is 0x9C4 to create a 100us interval */
+			WREG32_SOC15(GC, i, regRLC_GPM_TIMER_INT_3, 0x9C4);
+			/* RLC_GPM_GENERAL_12 : Minimum gap between wptr and rptr
+			 * to disable the page fault retry interrupts, default is
+			 * 0x100 (256) */
+			WREG32_SOC15(GC, i, regRLC_GPM_GENERAL_12, 0x100);
+		}
 #endif
+	}
 }
 
-static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev)
+static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev, int xcc_id)
 {
 	const struct rlc_firmware_header_v2_0 *hdr;
 	const __le32 *fw_data;
@@ -1233,36 +1267,38 @@ static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev)
 			   le32_to_cpu(hdr->header.ucode_array_offset_bytes));
 	fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
 
-	WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_ADDR,
+	WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_ADDR,
 			RLCG_UCODE_LOADING_START_ADDRESS);
 	for (i = 0; i < fw_size; i++) {
 		if (amdgpu_emu_mode == 1 && i % 100 == 0) {
 			dev_info(adev->dev, "Write RLC ucode data %u DWs\n", i);
 			msleep(1);
 		}
-		WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++));
+		WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++));
 	}
-	WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version);
+	WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version);
 
 	return 0;
 }
 
 static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 {
-	int r;
+	int r, i;
 
 	adev->gfx.rlc.funcs->stop(adev);
 
-	/* disable CG */
-	WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, 0);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		/* disable CG */
+		WREG32_SOC15(GC, i, regRLC_CGCG_CGLS_CTRL, 0);
 
-	gfx_v9_4_3_init_pg(adev);
+		gfx_v9_4_3_init_pg(adev, i);
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-		/* legacy rlc firmware loading */
-		r = gfx_v9_4_3_rlc_load_microcode(adev);
-		if (r)
-			return r;
+		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+			/* legacy rlc firmware loading */
+			r = gfx_v9_4_3_rlc_load_microcode(adev, i);
+			if (r)
+				return r;
+		}
 	}
 
 	adev->gfx.rlc.funcs->start(adev);
@@ -1270,7 +1306,8 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
+static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev,
+				       unsigned vmid)
 {
 	u32 reg, data;
 
@@ -1323,19 +1360,21 @@ static bool gfx_v9_4_3_is_rlcg_access_range(struct amdgpu_device *adev, u32 offs
 					ARRAY_SIZE(rlcg_access_gc_9_4_3));
 }
 
-static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev, bool enable)
+static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev,
+					 bool enable, int xcc_id)
 {
 	if (enable) {
-		WREG32_SOC15_RLC(GC, 0, regCP_MEC_CNTL, 0);
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_MEC_CNTL, 0);
 	} else {
-		WREG32_SOC15_RLC(GC, 0, regCP_MEC_CNTL,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_MEC_CNTL,
 			(CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
-		adev->gfx.kiq[0].ring.sched.ready = false;
+		adev->gfx.kiq[xcc_id].ring.sched.ready = false;
 	}
 	udelay(50);
 }
 
-static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev)
+static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev,
+						int xcc_id)
 {
 	const struct gfx_firmware_header_v1_0 *mec_hdr;
 	const __le32 *fw_data;
@@ -1347,7 +1386,7 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev)
 	if (!adev->gfx.mec_fw)
 		return -EINVAL;
 
-	gfx_v9_4_3_cp_compute_enable(adev, false);
+	gfx_v9_4_3_cp_compute_enable(adev, false, xcc_id);
 
 	mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
 	amdgpu_ucode_print_gfx_hdr(&mec_hdr->header);
@@ -1358,17 +1397,17 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev)
 	tmp = 0;
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, VMID, 0);
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
-	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_CNTL, tmp);
+	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_CNTL, tmp);
 
-	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_LO,
+	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_LO,
 		adev->gfx.mec.mec_fw_gpu_addr & 0xFFFFF000);
-	WREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_HI,
+	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_HI,
 		upper_32_bits(adev->gfx.mec.mec_fw_gpu_addr));
 
 	mec_ucode_addr_offset =
-		SOC15_REG_OFFSET(GC, 0, regCP_MEC_ME1_UCODE_ADDR);
+		SOC15_REG_OFFSET(GC, xcc_id, regCP_MEC_ME1_UCODE_ADDR);
 	mec_ucode_data_offset =
-		SOC15_REG_OFFSET(GC, 0, regCP_MEC_ME1_UCODE_DATA);
+		SOC15_REG_OFFSET(GC, xcc_id, regCP_MEC_ME1_UCODE_DATA);
 
 	/* MEC1 */
 	WREG32(mec_ucode_addr_offset, mec_hdr->jt_offset);
@@ -1383,18 +1422,18 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev)
 }
 
 /* KIQ functions */
-static void gfx_v9_4_3_kiq_setting(struct amdgpu_ring *ring)
+static void gfx_v9_4_3_kiq_setting(struct amdgpu_ring *ring, int xcc_id)
 {
 	uint32_t tmp;
 	struct amdgpu_device *adev = ring->adev;
 
 	/* tell RLC which is KIQ queue */
-	tmp = RREG32_SOC15(GC, 0, regRLC_CP_SCHEDULERS);
+	tmp = RREG32_SOC15(GC, xcc_id, regRLC_CP_SCHEDULERS);
 	tmp &= 0xffffff00;
 	tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue);
-	WREG32_SOC15_RLC(GC, 0, regRLC_CP_SCHEDULERS, tmp);
+	WREG32_SOC15_RLC(GC, xcc_id, regRLC_CP_SCHEDULERS, tmp);
 	tmp |= 0x80;
-	WREG32_SOC15_RLC(GC, 0, regRLC_CP_SCHEDULERS, tmp);
+	WREG32_SOC15_RLC(GC, xcc_id, regRLC_CP_SCHEDULERS, tmp);
 }
 
 static void gfx_v9_4_3_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd)
@@ -1538,123 +1577,123 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring)
+static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
 	int j;
 
 	/* disable wptr polling */
-	WREG32_FIELD15_PREREG(GC, 0, CP_PQ_WPTR_POLL_CNTL, EN, 0);
+	WREG32_FIELD15_PREREG(GC, xcc_id, CP_PQ_WPTR_POLL_CNTL, EN, 0);
 
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_BASE_ADDR,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_BASE_ADDR,
 	       mqd->cp_hqd_eop_base_addr_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_BASE_ADDR_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_BASE_ADDR_HI,
 	       mqd->cp_hqd_eop_base_addr_hi);
 
 	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_EOP_CONTROL,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_CONTROL,
 	       mqd->cp_hqd_eop_control);
 
 	/* enable doorbell? */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL,
 	       mqd->cp_hqd_pq_doorbell_control);
 
 	/* disable the queue if it's active */
-	if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) {
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1);
+	if (RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1) {
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST, 1);
 		for (j = 0; j < adev->usec_timeout; j++) {
-			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
+			if (!(RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1))
 				break;
 			udelay(1);
 		}
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST,
 		       mqd->cp_hqd_dequeue_request);
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR,
 		       mqd->cp_hqd_pq_rptr);
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO,
 		       mqd->cp_hqd_pq_wptr_lo);
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI,
 		       mqd->cp_hqd_pq_wptr_hi);
 	}
 
 	/* set the pointer to the MQD */
-	WREG32_SOC15_RLC(GC, 0, regCP_MQD_BASE_ADDR,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_BASE_ADDR,
 	       mqd->cp_mqd_base_addr_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_MQD_BASE_ADDR_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_BASE_ADDR_HI,
 	       mqd->cp_mqd_base_addr_hi);
 
 	/* set MQD vmid to 0 */
-	WREG32_SOC15_RLC(GC, 0, regCP_MQD_CONTROL,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_CONTROL,
 	       mqd->cp_mqd_control);
 
 	/* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_BASE,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_BASE,
 	       mqd->cp_hqd_pq_base_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_BASE_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_BASE_HI,
 	       mqd->cp_hqd_pq_base_hi);
 
 	/* set up the HQD, this is similar to CP_RB0_CNTL */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_CONTROL,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_CONTROL,
 	       mqd->cp_hqd_pq_control);
 
 	/* set the wb address whether it's enabled or not */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR_REPORT_ADDR,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR_REPORT_ADDR,
 				mqd->cp_hqd_pq_rptr_report_addr_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR_REPORT_ADDR_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR_REPORT_ADDR_HI,
 				mqd->cp_hqd_pq_rptr_report_addr_hi);
 
 	/* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_POLL_ADDR,
 	       mqd->cp_hqd_pq_wptr_poll_addr_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
 	       mqd->cp_hqd_pq_wptr_poll_addr_hi);
 
 	/* enable the doorbell if requested */
 	if (ring->use_doorbell) {
-		WREG32_SOC15(GC, 0, regCP_MEC_DOORBELL_RANGE_LOWER,
+		WREG32_SOC15(GC, xcc_id, regCP_MEC_DOORBELL_RANGE_LOWER,
 					(adev->doorbell_index.kiq * 2) << 2);
-		WREG32_SOC15(GC, 0, regCP_MEC_DOORBELL_RANGE_UPPER,
+		WREG32_SOC15(GC, xcc_id, regCP_MEC_DOORBELL_RANGE_UPPER,
 				(adev->doorbell_index.userqueue_end * 2) << 2);
 	}
 
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL,
 	       mqd->cp_hqd_pq_doorbell_control);
 
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO,
 	       mqd->cp_hqd_pq_wptr_lo);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI,
 	       mqd->cp_hqd_pq_wptr_hi);
 
 	/* set the vmid for the queue */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_VMID, mqd->cp_hqd_vmid);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_VMID, mqd->cp_hqd_vmid);
 
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PERSISTENT_STATE,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PERSISTENT_STATE,
 	       mqd->cp_hqd_persistent_state);
 
 	/* activate the queue */
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_ACTIVE,
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_ACTIVE,
 	       mqd->cp_hqd_active);
 
 	if (ring->use_doorbell)
-		WREG32_FIELD15_PREREG(GC, 0, CP_PQ_STATUS, DOORBELL_ENABLE, 1);
+		WREG32_FIELD15_PREREG(GC, xcc_id, CP_PQ_STATUS, DOORBELL_ENABLE, 1);
 
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring)
+static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	int j;
 
 	/* disable the queue if it's active */
-	if (RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1) {
+	if (RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1) {
 
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST, 1);
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST, 1);
 
 		for (j = 0; j < adev->usec_timeout; j++) {
-			if (!(RREG32_SOC15(GC, 0, regCP_HQD_ACTIVE) & 1))
+			if (!(RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1))
 				break;
 			udelay(1);
 		}
@@ -1663,72 +1702,71 @@ static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring)
 			DRM_DEBUG("KIQ dequeue request failed.\n");
 
 			/* Manual disable if dequeue request times out */
-			WREG32_SOC15_RLC(GC, 0, regCP_HQD_ACTIVE, 0);
+			WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_ACTIVE, 0);
 		}
 
-		WREG32_SOC15_RLC(GC, 0, regCP_HQD_DEQUEUE_REQUEST,
+		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST,
 		      0);
 	}
 
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_IQ_TIMER, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_IB_CONTROL, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PERSISTENT_STATE, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_RPTR, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_HI, 0);
-	WREG32_SOC15_RLC(GC, 0, regCP_HQD_PQ_WPTR_LO, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_IQ_TIMER, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_IB_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PERSISTENT_STATE, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI, 0);
+	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO, 0);
 
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring)
+static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
 	struct v9_mqd *tmp_mqd;
 
-	gfx_v9_4_3_kiq_setting(ring);
+	gfx_v9_4_3_kiq_setting(ring, xcc_id);
 
 	/* GPU could be in bad state during probe, driver trigger the reset
 	 * after load the SMU, in this case , the mqd is not be initialized.
 	 * driver need to re-init the mqd.
 	 * check mqd->cp_hqd_pq_control since this value should not be 0
 	 */
-	tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[0].mqd_backup;
+	tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[xcc_id].mqd_backup;
 	if (amdgpu_in_reset(adev) && tmp_mqd->cp_hqd_pq_control) {
 		/* for GPU_RESET case , reset MQD to a clean status */
-		if (adev->gfx.kiq[0].mqd_backup)
-			memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct v9_mqd_allocation));
+		if (adev->gfx.kiq[xcc_id].mqd_backup)
+			memcpy(mqd, adev->gfx.kiq[xcc_id].mqd_backup, sizeof(struct v9_mqd_allocation));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
 		amdgpu_ring_clear_ring(ring);
-
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
-		gfx_v9_4_3_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
+		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
 		gfx_v9_4_3_mqd_init(ring);
-		gfx_v9_4_3_kiq_init_register(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
 		mutex_unlock(&adev->srbm_mutex);
 
-		if (adev->gfx.kiq[0].mqd_backup)
-			memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct v9_mqd_allocation));
+		if (adev->gfx.kiq[xcc_id].mqd_backup)
+			memcpy(adev->gfx.kiq[xcc_id].mqd_backup, mqd, sizeof(struct v9_mqd_allocation));
 	}
 
 	return 0;
 }
 
-static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring)
+static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
@@ -1746,9 +1784,9 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring)
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
 		gfx_v9_4_3_mqd_init(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
@@ -1769,12 +1807,12 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev)
+static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring;
 	int r;
 
-	ring = &adev->gfx.kiq[0].ring;
+	ring = &adev->gfx.kiq[xcc_id].ring;
 
 	r = amdgpu_bo_reserve(ring->mqd_obj, false);
 	if (unlikely(r != 0))
@@ -1784,7 +1822,7 @@ static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev)
 	if (unlikely(r != 0))
 		return r;
 
-	gfx_v9_4_3_kiq_init_queue(ring);
+	gfx_v9_4_3_kiq_init_queue(ring, xcc_id);
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
@@ -1792,22 +1830,22 @@ static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev)
 	return 0;
 }
 
-static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev)
+static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring = NULL;
 	int r = 0, i;
 
-	gfx_v9_4_3_cp_compute_enable(adev, true);
+	gfx_v9_4_3_cp_compute_enable(adev, true, xcc_id);
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		ring = &adev->gfx.compute_ring[i];
+		ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings];
 
 		r = amdgpu_bo_reserve(ring->mqd_obj, false);
 		if (unlikely(r != 0))
 			goto done;
 		r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
 		if (!r) {
-			r = gfx_v9_4_3_kcq_init_queue(ring);
+			r = gfx_v9_4_3_kcq_init_queue(ring, xcc_id);
 			amdgpu_bo_kunmap(ring->mqd_obj);
 			ring->mqd_ptr = NULL;
 		}
@@ -1816,47 +1854,50 @@ static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = amdgpu_gfx_enable_kcq(adev, 0);
+	r = amdgpu_gfx_enable_kcq(adev, xcc_id);
 done:
 	return r;
 }
 
 static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 {
-	int r, i;
+	int r, i, j;
 	struct amdgpu_ring *ring;
 
-	gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+
+		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+			gfx_v9_4_3_disable_gpa_mode(adev);
 
-	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-		gfx_v9_4_3_disable_gpa_mode(adev);
+			r = gfx_v9_4_3_cp_compute_load_microcode(adev, i);
+			if (r)
+				return r;
+		}
 
-		r = gfx_v9_4_3_cp_compute_load_microcode(adev);
+		r = gfx_v9_4_3_kiq_resume(adev, i);
 		if (r)
 			return r;
-	}
 
-	r = gfx_v9_4_3_kiq_resume(adev);
-	if (r)
-		return r;
+		r = gfx_v9_4_3_kcq_resume(adev, i);
+		if (r)
+			return r;
 
-	r = gfx_v9_4_3_kcq_resume(adev);
-	if (r)
-		return r;
+		for (j = 0; j < adev->gfx.num_compute_rings; j++) {
+			ring = &adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings];
+			amdgpu_ring_test_helper(ring);
+		}
 
-	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		ring = &adev->gfx.compute_ring[i];
-		amdgpu_ring_test_helper(ring);
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
 	}
 
-	gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
-
 	return 0;
 }
 
-static void gfx_v9_4_3_cp_enable(struct amdgpu_device *adev, bool enable)
+static void gfx_v9_4_3_cp_enable(struct amdgpu_device *adev, bool enable,
+				int xcc_id)
 {
-	gfx_v9_4_3_cp_compute_enable(adev, enable);
+	gfx_v9_4_3_cp_compute_enable(adev, enable, xcc_id);
 }
 
 static int gfx_v9_4_3_hw_init(void *handle)
@@ -1882,27 +1923,30 @@ static int gfx_v9_4_3_hw_init(void *handle)
 static int gfx_v9_4_3_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
-	if (amdgpu_gfx_disable_kcq(adev, 0))
-		DRM_ERROR("KCQ disable failed\n");
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		if (amdgpu_gfx_disable_kcq(adev, i))
+			DRM_ERROR("XCD %d KCQ disable failed\n", i);
 
-	/* Use deinitialize sequence from CAIL when unbinding device from driver,
-	 * otherwise KIQ is hanging when binding back
-	 */
-	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
-		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me,
-				adev->gfx.kiq[0].ring.pipe,
-				adev->gfx.kiq[0].ring.queue, 0, 0);
-		gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[0].ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0, 0);
-		mutex_unlock(&adev->srbm_mutex);
-	}
+		/* Use deinitialize sequence from CAIL when unbinding device
+		 * from driver, otherwise KIQ is hanging when binding back
+		 */
+		if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
+			mutex_lock(&adev->srbm_mutex);
+			soc15_grbm_select(adev, adev->gfx.kiq[i].ring.me,
+					adev->gfx.kiq[i].ring.pipe,
+					adev->gfx.kiq[i].ring.queue, 0, i);
+			gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[i].ring, i);
+			soc15_grbm_select(adev, 0, 0, 0, 0, i);
+			mutex_unlock(&adev->srbm_mutex);
+		}
 
-	gfx_v9_4_3_cp_enable(adev, false);
+		gfx_v9_4_3_cp_enable(adev, false, i);
+	}
 
 	/* Skip suspend with A+A reset */
 	if (adev->gmc.xgmi.connected_to_cpu && amdgpu_in_reset(adev)) {
@@ -1927,12 +1971,14 @@ static int gfx_v9_4_3_resume(void *handle)
 static bool gfx_v9_4_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
-	if (REG_GET_FIELD(RREG32_SOC15(GC, 0, regGRBM_STATUS),
-				GRBM_STATUS, GUI_ACTIVE))
-		return false;
-	else
-		return true;
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		if (REG_GET_FIELD(RREG32_SOC15(GC, i, regGRBM_STATUS),
+					GRBM_STATUS, GUI_ACTIVE))
+			return false;
+	}
+	return true;
 }
 
 static int gfx_v9_4_3_wait_for_idle(void *handle)
@@ -1985,7 +2031,7 @@ static int gfx_v9_4_3_soft_reset(void *handle)
 		adev->gfx.rlc.funcs->stop(adev);
 
 		/* Disable MEC parsing/prefetching */
-		gfx_v9_4_3_cp_compute_enable(adev, false);
+		gfx_v9_4_3_cp_compute_enable(adev, false, 0);
 
 		if (grbm_soft_reset) {
 			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
@@ -2040,6 +2086,11 @@ static int gfx_v9_4_3_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	/* hardcode in emulation phase */
+	adev->gfx.num_xcd = 1;
+	adev->gfx.num_xcc_per_xcp = 1;
+	adev->gfx.partition_mode = AMDGPU_SPX_PARTITION_MODE;
+
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
 	gfx_v9_4_3_set_kiq_pm4_funcs(adev);
@@ -2068,7 +2119,7 @@ static int gfx_v9_4_3_late_init(void *handle)
 }
 
 static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *adev,
-						      bool enable)
+						      bool enable, int xcc_id)
 {
 	uint32_t data, def;
 
@@ -2077,7 +2128,7 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
 		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
-		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
 
 		data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
@@ -2087,28 +2138,28 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 		data |= RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK;
 
 		if (def != data)
-			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* MGLS is a global flag to control all MGLS in GFX */
 		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) {
 			/* 2 - RLC memory Light sleep */
 			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_RLC_LS) {
-				def = data = RREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL);
+				def = data = RREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL);
 				data |= RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
 				if (def != data)
-					WREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL, data);
+					WREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL, data);
 			}
 			/* 3 - CP memory Light sleep */
 			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) {
-				def = data = RREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL);
+				def = data = RREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL);
 				data |= CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
 				if (def != data)
-					WREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL, data);
+					WREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL, data);
 			}
 		}
 	} else {
 		/* 1 - MGCG_OVERRIDE */
-		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
 
 		data |= (RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK |
 			 RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
@@ -2116,20 +2167,20 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 			 RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK);
 
 		if (def != data)
-			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* 2 - disable MGLS in RLC */
-		data = RREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL);
+		data = RREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL);
 		if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) {
 			data &= ~RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
-			WREG32_SOC15(GC, 0, regRLC_MEM_SLP_CNTL, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL, data);
 		}
 
 		/* 3 - disable MGLS in CP */
-		data = RREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL);
+		data = RREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL);
 		if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) {
 			data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
-			WREG32_SOC15(GC, 0, regCP_MEM_SLP_CNTL, data);
+			WREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL, data);
 		}
 	}
 
@@ -2137,14 +2188,14 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 }
 
 static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
-						      bool enable)
+						      bool enable, int xcc_id)
 {
 	uint32_t def, data;
 
 	amdgpu_gfx_rlc_enter_safe_mode(adev);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
-		def = data = RREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
 		/* unset CGCG override */
 		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK;
 		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS)
@@ -2153,10 +2204,10 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 			data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK;
 		/* update CGCG and CGLS override bits */
 		if (def != data)
-			WREG32_SOC15(GC, 0, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* enable cgcg FSM(0x0000363F) */
-		def = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL);
+		def = RREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL);
 
 		if (adev->asic_type == CHIP_ARCTURUS)
 			data = (0x2000 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
@@ -2168,43 +2219,43 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 			data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) |
 				RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK;
 		if (def != data)
-			WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL, data);
 
 		/* set IDLE_POLL_COUNT(0x00900100) */
-		def = RREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_CNTL);
+		def = RREG32_SOC15(GC, xcc_id, regCP_RB_WPTR_POLL_CNTL);
 		data = (0x0100 << CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT) |
 			(0x0090 << CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT);
 		if (def != data)
-			WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_CNTL, data);
+			WREG32_SOC15(GC, xcc_id, regCP_RB_WPTR_POLL_CNTL, data);
 	} else {
-		def = data = RREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL);
+		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL);
 		/* reset CGCG/CGLS bits */
 		data &= ~(RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK);
 		/* disable cgcg and cgls in FSM */
 		if (def != data)
-			WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, data);
+			WREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL, data);
 	}
 
 	amdgpu_gfx_rlc_exit_safe_mode(adev);
 }
 
 static int gfx_v9_4_3_update_gfx_clock_gating(struct amdgpu_device *adev,
-					    bool enable)
+					    bool enable, int xcc_id)
 {
 	if (enable) {
 		/* CGCG/CGLS should be enabled after MGCG/MGLS
 		 * ===  MGCG + MGLS ===
 		 */
-		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable);
+		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable, xcc_id);
 		/* ===  CGCG + CGLS === */
-		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable);
+		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable, xcc_id);
 	} else {
 		/* CGCG/CGLS should be disabled before MGCG/MGLS
 		 * ===  CGCG + CGLS ===
 		 */
-		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable);
+		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable, xcc_id);
 		/* ===  MGCG + MGLS === */
-		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable);
+		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable, xcc_id);
 	}
 	return 0;
 }
@@ -2234,14 +2285,16 @@ static int gfx_v9_4_3_set_clockgating_state(void *handle,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 3):
-		gfx_v9_4_3_update_gfx_clock_gating(adev,
-						 state == AMD_CG_STATE_GATE);
+		for (i = 0; i < adev->gfx.num_xcd; i++)
+			gfx_v9_4_3_update_gfx_clock_gating(adev,
+						state == AMD_CG_STATE_GATE, i);
 		break;
 	default:
 		break;
@@ -2509,8 +2562,9 @@ static void gfx_v9_4_3_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
 }
 
 static void gfx_v9_4_3_set_compute_eop_interrupt_state(struct amdgpu_device *adev,
-						     int me, int pipe,
-						     enum amdgpu_interrupt_state state)
+						       int me, int pipe,
+						       enum amdgpu_interrupt_state state,
+						       int xcc_id)
 {
 	u32 mec_int_cntl, mec_int_cntl_reg;
 
@@ -2523,16 +2577,16 @@ static void gfx_v9_4_3_set_compute_eop_interrupt_state(struct amdgpu_device *ade
 	if (me == 1) {
 		switch (pipe) {
 		case 0:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE0_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE0_INT_CNTL);
 			break;
 		case 1:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE1_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE1_INT_CNTL);
 			break;
 		case 2:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE2_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE2_INT_CNTL);
 			break;
 		case 3:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE3_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE3_INT_CNTL);
 			break;
 		default:
 			DRM_DEBUG("invalid pipe %d\n", pipe);
@@ -2566,12 +2620,15 @@ static int gfx_v9_4_3_set_priv_reg_fault_state(struct amdgpu_device *adev,
 					     unsigned type,
 					     enum amdgpu_interrupt_state state)
 {
+	int i;
+
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
-		WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0,
-			       PRIV_REG_INT_ENABLE,
-			       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+		for (i = 0; i < adev->gfx.num_xcd; i++)
+			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
+				PRIV_REG_INT_ENABLE,
+				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
 		break;
 	default:
 		break;
@@ -2585,12 +2642,15 @@ static int gfx_v9_4_3_set_priv_inst_fault_state(struct amdgpu_device *adev,
 					      unsigned type,
 					      enum amdgpu_interrupt_state state)
 {
+	int i;
+
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
-		WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0,
-			       PRIV_INSTR_INT_ENABLE,
-			       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+		for (i = 0; i < adev->gfx.num_xcd; i++)
+			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
+				PRIV_INSTR_INT_ENABLE,
+				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
 		break;
 	default:
 		break;
@@ -2604,34 +2664,38 @@ static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
 					    unsigned type,
 					    enum amdgpu_interrupt_state state)
 {
-	switch (type) {
-	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 0, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 1, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 2, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 3, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 0, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 1, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 2, state);
-		break;
-	case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP:
-		gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 3, state);
-		break;
-	default:
-		break;
+	int i;
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		switch (type) {
+		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 0, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 1, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 2, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 3, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 0, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 1, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 2, state, i);
+			break;
+		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP:
+			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 3, state, i);
+			break;
+		default:
+			break;
+		}
 	}
+
 	return 0;
 }
 
@@ -2871,12 +2935,15 @@ static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_kiq = {
 
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev)
 {
-	int i;
+	int i, j;
 
-	adev->gfx.kiq[0].ring.funcs = &gfx_v9_4_3_ring_funcs_kiq;
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		adev->gfx.kiq[i].ring.funcs = &gfx_v9_4_3_ring_funcs_kiq;
 
-	for (i = 0; i < adev->gfx.num_compute_rings; i++)
-		adev->gfx.compute_ring[i].funcs = &gfx_v9_4_3_ring_funcs_compute;
+		for (j = 0; j < adev->gfx.num_compute_rings; j++)
+			adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings].funcs
+					= &gfx_v9_4_3_ring_funcs_compute;
+	}
 }
 
 static const struct amdgpu_irq_src_funcs gfx_v9_4_3_eop_irq_funcs = {
-- 
GitLab


From ec08571aca7c5e6bf2d1820db9c8aaa104eb9b68 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 17 Nov 2021 17:51:17 +0800
Subject: [PATCH 0144/1791] drm/amdgpu: add xcc index argument to gfx v9_4_3
 functions

Change those v9_4_3 interfaces which are exposed in gfx_v9_0.c.
For some active single-xcc emu models, the code path in
gfx_v9_0.c is better to keep reserved for a while.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 56999bb7ac26a..f1c04140e7177 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1068,13 +1068,13 @@ static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev)
+void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
-	data = RREG32_SOC15(GC, 0, regCPC_PSP_DEBUG);
+	data = RREG32_SOC15(GC, xcc_id, regCPC_PSP_DEBUG);
 	data |= CPC_PSP_DEBUG__UTCL2IUGPAOVERRIDE_MASK;
-	WREG32_SOC15(GC, 0, regCPC_PSP_DEBUG, data);
+	WREG32_SOC15(GC, xcc_id, regCPC_PSP_DEBUG, data);
 }
 
 static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
@@ -1177,19 +1177,19 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 }
 
 static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
-					       bool enable)
+						 bool enable, int xcc_id)
 {
 	u32 tmp;
 
 	/* These interrupts should be enabled to drive DS clock */
 
-	tmp = RREG32_SOC15(GC, 0, regCP_INT_CNTL_RING0);
+	tmp = RREG32_SOC15(GC, xcc_id, regCP_INT_CNTL_RING0);
 
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0);
 
-	WREG32_SOC15(GC, 0, regCP_INT_CNTL_RING0, tmp);
+	WREG32_SOC15(GC, xcc_id, regCP_INT_CNTL_RING0, tmp);
 }
 
 static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
@@ -1198,7 +1198,7 @@ static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 
 	for (i = 0; i < adev->gfx.num_xcd; i++) {
 		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 0);
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
 		gfx_v9_4_3_wait_for_rlc_serdes(adev, i);
 	}
 }
@@ -1228,7 +1228,7 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 
 		/* carrizo do enable cp interrupt after cp inited */
 		if (!(adev->flags & AMD_IS_APU)) {
-			gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
+			gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
 			udelay(50);
 		}
 
@@ -1865,10 +1865,10 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 	struct amdgpu_ring *ring;
 
 	for (i = 0; i < adev->gfx.num_xcd; i++) {
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false);
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
 
 		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-			gfx_v9_4_3_disable_gpa_mode(adev);
+			gfx_v9_4_3_disable_gpa_mode(adev, i);
 
 			r = gfx_v9_4_3_cp_compute_load_microcode(adev, i);
 			if (r)
@@ -1888,7 +1888,7 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 			amdgpu_ring_test_helper(ring);
 		}
 
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true);
+		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
 	}
 
 	return 0;
-- 
GitLab


From d51ac6d0a23caf1005cb640f8533161c5d2dd0c0 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 24 May 2022 11:22:49 +0800
Subject: [PATCH 0145/1791] drm/amdgpu: add xcc index argument to select_sh_se
 function v2

v1: To support multiple XCD case (Le)
v2: introduce xcc index to gfx_v11_0_select_sh_se (Hawking)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c |  4 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c   | 16 ++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |  4 +--
 drivers/gpu/drm/amd/amdgpu/cik.c              |  4 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        | 16 ++++-----
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        |  8 ++---
 drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c         | 24 +++++++-------
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c         | 33 ++++++++++---------
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c         | 33 ++++++++++---------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         | 30 ++++++++---------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h         |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c       |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 20 ++++++-----
 drivers/gpu/drm/amd/amdgpu/nv.c               |  4 +--
 drivers/gpu/drm/amd/amdgpu/si.c               |  4 +--
 drivers/gpu/drm/amd/amdgpu/soc15.c            |  4 +--
 drivers/gpu/drm/amd/amdgpu/soc21.c            |  4 +--
 drivers/gpu/drm/amd/amdgpu/vi.c               |  4 +--
 18 files changed, 110 insertions(+), 106 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index ebb35633058c2..ae06d1f2af932 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -787,7 +787,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	for (se_idx = 0; se_idx < se_cnt; se_idx++) {
 		for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) {
 
-			amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff, 0);
 			queue_map = RREG32_SOC15(GC, 0, mmSPI_CSQ_WF_ACTIVE_STATUS);
 
 			/*
@@ -820,7 +820,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 		}
 	}
 
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	unlock_spi_csq_mutexes(adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index f60753f97ac59..cc64ae550bc1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -136,7 +136,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 		}
 		mutex_lock(&adev->grbm_idx_mutex);
 		amdgpu_gfx_select_se_sh(adev, se_bank,
-					sh_bank, instance_bank);
+					sh_bank, instance_bank, 0);
 	} else if (use_ring) {
 		mutex_lock(&adev->srbm_mutex);
 		amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid);
@@ -169,7 +169,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 
 end:
 	if (use_bank) {
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 	} else if (use_ring) {
 		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0);
@@ -263,7 +263,7 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off
 		mutex_lock(&adev->grbm_idx_mutex);
 		amdgpu_gfx_select_se_sh(adev, rd->id.grbm.se,
 								rd->id.grbm.sh,
-								rd->id.grbm.instance);
+								rd->id.grbm.instance, 0);
 	}
 
 	if (rd->id.use_srbm) {
@@ -295,7 +295,7 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off
 	}
 end:
 	if (rd->id.use_grbm) {
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 	}
 
@@ -907,13 +907,13 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf,
 
 	/* switch to the specific se/sh/cu */
 	mutex_lock(&adev->grbm_idx_mutex);
-	amdgpu_gfx_select_se_sh(adev, se, sh, cu);
+	amdgpu_gfx_select_se_sh(adev, se, sh, cu, 0);
 
 	x = 0;
 	if (adev->gfx.funcs->read_wave_data)
 		adev->gfx.funcs->read_wave_data(adev, simd, wave, data, &x);
 
-	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF);
+	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
@@ -1001,7 +1001,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 
 	/* switch to the specific se/sh/cu */
 	mutex_lock(&adev->grbm_idx_mutex);
-	amdgpu_gfx_select_se_sh(adev, se, sh, cu);
+	amdgpu_gfx_select_se_sh(adev, se, sh, cu, 0);
 
 	if (bank == 0) {
 		if (adev->gfx.funcs->read_wave_vgprs)
@@ -1011,7 +1011,7 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 			adev->gfx.funcs->read_wave_sgprs(adev, simd, wave, offset, size>>2, data);
 	}
 
-	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF);
+	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index a9e41d7970ead..77e2f714e357d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -237,7 +237,7 @@ struct amdgpu_gfx_funcs {
 	/* get the gpu clock counter */
 	uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev);
 	void (*select_se_sh)(struct amdgpu_device *adev, u32 se_num,
-			     u32 sh_num, u32 instance);
+			     u32 sh_num, u32 instance, int xcc_id);
 	void (*read_wave_data)(struct amdgpu_device *adev, uint32_t simd,
 			       uint32_t wave, uint32_t *dst, int *no_fields);
 	void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t simd,
@@ -386,7 +386,7 @@ struct amdgpu_gfx {
 };
 
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
-#define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance))
+#define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id)))
 #define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid))
 #define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev))
 
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index de6d10390ab2f..5641cf05d856b 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -1141,12 +1141,12 @@ static uint32_t cik_get_register_value(struct amdgpu_device *adev,
 
 		mutex_lock(&adev->grbm_idx_mutex);
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 		val = RREG32(reg_offset);
 
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 		return val;
 	} else {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index accc0a7251b9a..323f5b8927ad1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -3490,7 +3490,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev,
 				 struct amdgpu_cu_info *cu_info);
 static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev);
 static void gfx_v10_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
-				   u32 sh_num, u32 instance);
+				   u32 sh_num, u32 instance, int xcc_id);
 static u32 gfx_v10_0_get_wgp_active_bitmap_per_sh(struct amdgpu_device *adev);
 
 static int gfx_v10_0_rlc_backdoor_autoload_buffer_init(struct amdgpu_device *adev);
@@ -4712,7 +4712,7 @@ static int gfx_v10_0_sw_fini(void *handle)
 }
 
 static void gfx_v10_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
-				   u32 sh_num, u32 instance)
+				   u32 sh_num, u32 instance, int xcc_id)
 {
 	u32 data;
 
@@ -4772,13 +4772,13 @@ static void gfx_v10_0_setup_rb(struct amdgpu_device *adev)
 				(adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6))) &&
 			    ((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1))
 				continue;
-			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = gfx_v10_0_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
 					       rb_bitmap_width_per_sh);
 		}
 	}
-	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
@@ -4907,7 +4907,7 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			wgp_active_bitmap = gfx_v10_0_get_wgp_active_bitmap_per_sh(adev);
 			/*
 			 * Set corresponding TCP bits for the inactive WGPs in
@@ -4940,7 +4940,7 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev)
 		}
 	}
 
-	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -9540,7 +9540,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev,
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			if (i < 4 && j < 2)
 				gfx_v10_0_set_user_wgp_inactive_bitmap_per_sh(
 					adev, disable_masks[i * 2 + j]);
@@ -9561,7 +9561,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev,
 			cu_info->ao_cu_bitmap[i][j] = ao_bitmap;
 		}
 	}
-	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 4fbefe236fc71..50d0ff9ca2599 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -112,7 +112,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev,
                                  struct amdgpu_cu_info *cu_info);
 static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev);
 static void gfx_v11_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
-				   u32 sh_num, u32 instance);
+				   u32 sh_num, u32 instance, int xcc_id);
 static u32 gfx_v11_0_get_wgp_active_bitmap_per_sh(struct amdgpu_device *adev);
 
 static void gfx_v11_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume);
@@ -1484,7 +1484,7 @@ static int gfx_v11_0_sw_fini(void *handle)
 }
 
 static void gfx_v11_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
-				   u32 sh_num, u32 instance)
+				   u32 sh_num, u32 instance, int xcc_id)
 {
 	u32 data;
 
@@ -6473,7 +6473,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev,
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
 			mask = 1;
 			counter = 0;
-			gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			if (i < 8 && j < 2)
 				gfx_v11_0_set_user_wgp_inactive_bitmap_per_sh(
 					adev, disable_masks[i * 2 + j]);
@@ -6505,7 +6505,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev,
 			active_cu_number += counter;
 		}
 	}
-	gfx_v11_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v11_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index d9ce4d1c50e49..7cb72bf1acdd3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -1285,7 +1285,7 @@ static void gfx_v6_0_tiling_mode_table_init(struct amdgpu_device *adev)
 }
 
 static void gfx_v6_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
-				  u32 sh_num, u32 instance)
+				  u32 sh_num, u32 instance, int xcc_id)
 {
 	u32 data;
 
@@ -1438,12 +1438,12 @@ static void gfx_v6_0_write_harvested_raster_configs(struct amdgpu_device *adev,
 		}
 
 		/* GRBM_GFX_INDEX has a different offset on SI */
-		gfx_v6_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff);
+		gfx_v6_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se);
 	}
 
 	/* GRBM_GFX_INDEX has a different offset on SI */
-	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 }
 
 static void gfx_v6_0_setup_rb(struct amdgpu_device *adev)
@@ -1459,14 +1459,14 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = gfx_v6_0_get_rb_active_bitmap(adev);
 			active_rbs |= data <<
 				((i * adev->gfx.config.max_sh_per_se + j) *
 				 rb_bitmap_width_per_sh);
 		}
 	}
-	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
 	adev->gfx.config.num_rbs = hweight32(active_rbs);
@@ -1487,7 +1487,7 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev)
 	/* cache the values for userspace */
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			adev->gfx.config.rb_config[i][j].rb_backend_disable =
 				RREG32(mmCC_RB_BACKEND_DISABLE);
 			adev->gfx.config.rb_config[i][j].user_rb_backend_disable =
@@ -1496,7 +1496,7 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev)
 				RREG32(mmPA_SC_RASTER_CONFIG);
 		}
 	}
-	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -1535,7 +1535,7 @@ static void gfx_v6_0_setup_spi(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = RREG32(mmSPI_STATIC_THREAD_MGMT_3);
 			active_cu = gfx_v6_0_get_cu_enabled(adev);
 
@@ -1550,7 +1550,7 @@ static void gfx_v6_0_setup_spi(struct amdgpu_device *adev)
 			}
 		}
 	}
-	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -2391,7 +2391,7 @@ static void gfx_v6_0_enable_lbpw(struct amdgpu_device *adev, bool enable)
 	WREG32_FIELD(RLC_LB_CNTL, LOAD_BALANCE_ENABLE, enable ? 1 : 0);
 
 	if (!enable) {
-		gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmSPI_LB_CU_MASK, 0x00ff);
 	}
 }
@@ -3571,7 +3571,7 @@ static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev)
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			if (i < 4 && j < 2)
 				gfx_v6_0_set_user_cu_inactive_bitmap(
 					adev, disable_masks[i * 2 + j]);
@@ -3593,7 +3593,7 @@ static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev)
 		}
 	}
 
-	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 46740ad9a80ff..d055e44eee1d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -1552,7 +1552,8 @@ static void gfx_v7_0_tiling_mode_table_init(struct amdgpu_device *adev)
  * Select which SE, SH combinations to address.
  */
 static void gfx_v7_0_select_se_sh(struct amdgpu_device *adev,
-				  u32 se_num, u32 sh_num, u32 instance)
+				  u32 se_num, u32 sh_num, u32 instance,
+				  int xcc_id)
 {
 	u32 data;
 
@@ -1732,13 +1733,13 @@ gfx_v7_0_write_harvested_raster_configs(struct amdgpu_device *adev,
 		}
 
 		/* GRBM_GFX_INDEX has a different offset on CI+ */
-		gfx_v7_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff);
+		gfx_v7_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se);
 		WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1);
 	}
 
 	/* GRBM_GFX_INDEX has a different offset on CI+ */
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 }
 
 /**
@@ -1761,13 +1762,13 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = gfx_v7_0_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
 					       rb_bitmap_width_per_sh);
 		}
 	}
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
 	adev->gfx.config.num_rbs = hweight32(active_rbs);
@@ -1790,7 +1791,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 	/* cache the values for userspace */
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			adev->gfx.config.rb_config[i][j].rb_backend_disable =
 				RREG32(mmCC_RB_BACKEND_DISABLE);
 			adev->gfx.config.rb_config[i][j].user_rb_backend_disable =
@@ -1801,7 +1802,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 				RREG32(mmPA_SC_RASTER_CONFIG_1);
 		}
 	}
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -1911,7 +1912,7 @@ static void gfx_v7_0_constants_init(struct amdgpu_device *adev)
 	 * making sure that the following register writes will be broadcasted
 	 * to all the shaders
 	 */
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
@@ -3301,7 +3302,7 @@ static void gfx_v7_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			for (k = 0; k < adev->usec_timeout; k++) {
 				if (RREG32(mmRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
@@ -3309,7 +3310,7 @@ static void gfx_v7_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 			}
 		}
 	}
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK |
@@ -3474,7 +3475,7 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev)
 	WREG32(mmRLC_LB_CNTR_MAX, 0x00008000);
 
 	mutex_lock(&adev->grbm_idx_mutex);
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	WREG32(mmRLC_LB_INIT_CU_MASK, 0xffffffff);
 	WREG32(mmRLC_LB_PARAMS, 0x00600408);
 	WREG32(mmRLC_LB_CNTL, 0x80000004);
@@ -3530,7 +3531,7 @@ static void gfx_v7_0_enable_cgcg(struct amdgpu_device *adev, bool enable)
 		tmp = gfx_v7_0_halt_rlc(adev);
 
 		mutex_lock(&adev->grbm_idx_mutex);
-		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
 		WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
 		tmp2 = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK |
@@ -3584,7 +3585,7 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
 		tmp = gfx_v7_0_halt_rlc(adev);
 
 		mutex_lock(&adev->grbm_idx_mutex);
-		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
 		WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
 		data = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK |
@@ -3635,7 +3636,7 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
 		tmp = gfx_v7_0_halt_rlc(adev);
 
 		mutex_lock(&adev->grbm_idx_mutex);
-		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
 		WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
 		data = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK | RLC_SERDES_WR_CTRL__MGCG_OVERRIDE_1_MASK;
@@ -5115,7 +5116,7 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev)
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			if (i < 4 && j < 2)
 				gfx_v7_0_set_user_cu_inactive_bitmap(
 					adev, disable_masks[i * 2 + j]);
@@ -5136,7 +5137,7 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev)
 			cu_info->ao_cu_bitmap[i][j] = ao_bitmap;
 		}
 	}
-	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 8a43e87de49fa..b604808761499 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -3395,7 +3395,8 @@ static void gfx_v8_0_tiling_mode_table_init(struct amdgpu_device *adev)
 }
 
 static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev,
-				  u32 se_num, u32 sh_num, u32 instance)
+				  u32 se_num, u32 sh_num, u32 instance,
+				  int xcc_id)
 {
 	u32 data;
 
@@ -3579,13 +3580,13 @@ gfx_v8_0_write_harvested_raster_configs(struct amdgpu_device *adev,
 		}
 
 		/* GRBM_GFX_INDEX has a different offset on VI */
-		gfx_v8_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff);
+		gfx_v8_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0);
 		WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se);
 		WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1);
 	}
 
 	/* GRBM_GFX_INDEX has a different offset on VI */
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 }
 
 static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
@@ -3601,13 +3602,13 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = gfx_v8_0_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
 					       rb_bitmap_width_per_sh);
 		}
 	}
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
 	adev->gfx.config.num_rbs = hweight32(active_rbs);
@@ -3630,7 +3631,7 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
 	/* cache the values for userspace */
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			adev->gfx.config.rb_config[i][j].rb_backend_disable =
 				RREG32(mmCC_RB_BACKEND_DISABLE);
 			adev->gfx.config.rb_config[i][j].user_rb_backend_disable =
@@ -3641,7 +3642,7 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
 				RREG32(mmPA_SC_RASTER_CONFIG_1);
 		}
 	}
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -3788,7 +3789,7 @@ static void gfx_v8_0_constants_init(struct amdgpu_device *adev)
 	 * making sure that the following register writes will be broadcasted
 	 * to all the shaders
 	 */
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	WREG32(mmPA_SC_FIFO_SIZE,
 		   (adev->gfx.config.sc_prim_fifo_size_frontend <<
@@ -3819,7 +3820,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			for (k = 0; k < adev->usec_timeout; k++) {
 				if (RREG32(mmRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
@@ -3827,7 +3828,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 			}
 			if (k == adev->usec_timeout) {
 				gfx_v8_0_select_se_sh(adev, 0xffffffff,
-						      0xffffffff, 0xffffffff);
+						      0xffffffff, 0xffffffff, 0);
 				mutex_unlock(&adev->grbm_idx_mutex);
 				DRM_INFO("Timeout wait for RLC serdes %u,%u\n",
 					 i, j);
@@ -3835,7 +3836,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 			}
 		}
 	}
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK |
@@ -5481,7 +5482,7 @@ static void gfx_v8_0_send_serdes_cmd(struct amdgpu_device *adev,
 {
 	uint32_t data;
 
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff);
 	WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff);
@@ -6723,11 +6724,11 @@ static void gfx_v8_0_parse_sq_irq(struct amdgpu_device *adev, unsigned ih_data,
 			 */
 			if (from_wq) {
 				mutex_lock(&adev->grbm_idx_mutex);
-				gfx_v8_0_select_se_sh(adev, se_id, sh_id, cu_id);
+				gfx_v8_0_select_se_sh(adev, se_id, sh_id, cu_id, 0);
 
 				sq_edc_source = REG_GET_FIELD(RREG32(mmSQ_EDC_INFO), SQ_EDC_INFO, SOURCE);
 
-				gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+				gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 				mutex_unlock(&adev->grbm_idx_mutex);
 			}
 
@@ -7116,7 +7117,7 @@ static void gfx_v8_0_get_cu_info(struct amdgpu_device *adev)
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0);
 			if (i < 4 && j < 2)
 				gfx_v8_0_set_user_cu_inactive_bitmap(
 					adev, disable_masks[i * 2 + j]);
@@ -7137,7 +7138,7 @@ static void gfx_v8_0_get_cu_info(struct amdgpu_device *adev)
 			cu_info->ao_cu_bitmap[i][j] = ao_bitmap;
 		}
 	}
-	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 4939fd61355be..2fa7adef18a91 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1504,7 +1504,7 @@ static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev)
 			mask = 1;
 			cu_bitmap = 0;
 			counter = 0;
-			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0);
 
 			for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) {
 				if (cu_info->bitmap[i][j] & mask) {
@@ -1523,7 +1523,7 @@ static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev)
 			cu_info->ao_cu_bitmap[i][j] = cu_bitmap;
 		}
 	}
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
@@ -1545,7 +1545,7 @@ static void gfx_v9_0_init_lbpw(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	/* set mmRLC_LB_INIT_CU_MASK thru broadcast mode to enable all SE/SH*/
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	WREG32_SOC15(GC, 0, mmRLC_LB_INIT_CU_MASK, 0xffffffff);
 
 	/* set mmRLC_LB_PARAMS = 0x003F_1006 */
@@ -1594,7 +1594,7 @@ static void gfx_v9_4_init_lbpw(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	/* set mmRLC_LB_INIT_CU_MASK thru broadcast mode to enable all SE/SH*/
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	WREG32_SOC15(GC, 0, mmRLC_LB_INIT_CU_MASK, 0xffffffff);
 
 	/* set mmRLC_LB_PARAMS = 0x003F_1006 */
@@ -2241,7 +2241,7 @@ static void gfx_v9_0_tiling_mode_table_init(struct amdgpu_device *adev)
 }
 
 void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num,
-			   u32 instance)
+			   u32 instance, int xcc_id)
 {
 	u32 data;
 
@@ -2290,13 +2290,13 @@ static void gfx_v9_0_setup_rb(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0);
 			data = gfx_v9_0_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
 					       rb_bitmap_width_per_sh);
 		}
 	}
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
@@ -2433,7 +2433,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0);
 			for (k = 0; k < adev->usec_timeout; k++) {
 				if (RREG32_SOC15(GC, 0, mmRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
@@ -2441,7 +2441,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 			}
 			if (k == adev->usec_timeout) {
 				amdgpu_gfx_select_se_sh(adev, 0xffffffff,
-						      0xffffffff, 0xffffffff);
+						      0xffffffff, 0xffffffff, 0);
 				mutex_unlock(&adev->grbm_idx_mutex);
 				DRM_INFO("Timeout wait for RLC serdes %u,%u\n",
 					 i, j);
@@ -2449,7 +2449,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 			}
 		}
 	}
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK |
@@ -6608,7 +6608,7 @@ static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev)
 	for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) {
 		for (j = 0; j < gfx_v9_0_edc_counter_regs[i].se_num; j++) {
 			for (k = 0; k < gfx_v9_0_edc_counter_regs[i].instance; k++) {
-				amdgpu_gfx_select_se_sh(adev, j, 0x0, k);
+				amdgpu_gfx_select_se_sh(adev, j, 0x0, k, 0);
 				RREG32(SOC15_REG_ENTRY_OFFSET(gfx_v9_0_edc_counter_regs[i]));
 			}
 		}
@@ -6670,7 +6670,7 @@ static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
 	for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) {
 		for (j = 0; j < gfx_v9_0_edc_counter_regs[i].se_num; j++) {
 			for (k = 0; k < gfx_v9_0_edc_counter_regs[i].instance; k++) {
-				amdgpu_gfx_select_se_sh(adev, j, 0, k);
+				amdgpu_gfx_select_se_sh(adev, j, 0, k, 0);
 				reg_value =
 					RREG32(SOC15_REG_ENTRY_OFFSET(gfx_v9_0_edc_counter_regs[i]));
 				if (reg_value)
@@ -6685,7 +6685,7 @@ static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
 	err_data->ce_count += sec_count;
 	err_data->ue_count += ded_count;
 
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	gfx_v9_0_query_utc_edc_status(adev, err_data);
@@ -7145,7 +7145,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev,
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0);
 			gfx_v9_0_set_user_cu_inactive_bitmap(
 				adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]);
 			bitmap = gfx_v9_0_get_cu_active_bitmap(adev);
@@ -7178,7 +7178,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev,
 			cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap;
 		}
 	}
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
index dfe8d4841f582..f9f6edc5e5589 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h
@@ -27,6 +27,6 @@
 extern const struct amdgpu_ip_block_version gfx_v9_0_ip_block;
 
 void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num,
-			   u32 instance);
+			   u32 instance, int xcc_id);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index 93438770ca1a1..d648a29c33e0a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -777,7 +777,7 @@ void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev)
 {
 	u32 tmp;
 
-	gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 
 	tmp = 0;
 	tmp = REG_SET_FIELD(tmp, GC_THROTTLE_CTRL, PATTERN_MODE, 1);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index f1c04140e7177..b67be666f38a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -568,7 +568,8 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
 				    u32 se_num,
 				    u32 sh_num,
-				    u32 instance)
+				    u32 instance,
+				    int xcc_id)
 {
 	u32 data;
 
@@ -591,7 +592,7 @@ static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
 	else
 		data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SH_INDEX, sh_num);
 
-	WREG32_SOC15_RLC_SHADOW_EX(reg, GC, 0, regGRBM_GFX_INDEX, data);
+	WREG32_SOC15_RLC_SHADOW_EX(reg, GC, xcc_id, regGRBM_GFX_INDEX, data);
 }
 
 static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address)
@@ -907,13 +908,13 @@ static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev, int xcc_id)
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, xcc_id);
 			data = gfx_v9_4_3_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
 					       rb_bitmap_width_per_sh);
 		}
 	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
@@ -1146,7 +1147,7 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, xcc_id);
 			for (k = 0; k < adev->usec_timeout; k++) {
 				if (RREG32_SOC15(GC, 0, regRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
@@ -1154,7 +1155,8 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 			}
 			if (k == adev->usec_timeout) {
 				gfx_v9_4_3_select_se_sh(adev, 0xffffffff,
-						      0xffffffff, 0xffffffff);
+							0xffffffff, 0xffffffff,
+							xcc_id);
 				mutex_unlock(&adev->grbm_idx_mutex);
 				DRM_INFO("Timeout wait for RLC serdes %u,%u\n",
 					 i, j);
@@ -1162,7 +1164,7 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 			}
 		}
 	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK |
@@ -3065,7 +3067,7 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff);
+			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, 0);
 			gfx_v9_4_3_set_user_cu_inactive_bitmap(
 				adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]);
 			bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev);
@@ -3098,7 +3100,7 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
 			cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap;
 		}
 	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index 47420b403871c..148049782f50d 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -371,12 +371,12 @@ static uint32_t nv_read_indexed_register(struct amdgpu_device *adev, u32 se_num,
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 	val = RREG32(reg_offset);
 
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 	return val;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c
index 7f99e130acd06..f64b87b11b1b5 100644
--- a/drivers/gpu/drm/amd/amdgpu/si.c
+++ b/drivers/gpu/drm/amd/amdgpu/si.c
@@ -1181,12 +1181,12 @@ static uint32_t si_get_register_value(struct amdgpu_device *adev,
 
 		mutex_lock(&adev->grbm_idx_mutex);
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 		val = RREG32(reg_offset);
 
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 		return val;
 	} else {
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 4b79a8933476c..4d1487a9836c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -363,12 +363,12 @@ static uint32_t soc15_read_indexed_register(struct amdgpu_device *adev, u32 se_n
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 	val = RREG32(reg_offset);
 
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 	return val;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 514bfc705d5ab..6ef4be9322d95 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -288,12 +288,12 @@ static uint32_t soc21_read_indexed_register(struct amdgpu_device *adev, u32 se_n
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 	val = RREG32(reg_offset);
 
 	if (se_num != 0xffffffff || sh_num != 0xffffffff)
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 	return val;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 531f173ade2d6..8e70581960fbb 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -762,12 +762,12 @@ static uint32_t vi_get_register_value(struct amdgpu_device *adev,
 
 		mutex_lock(&adev->grbm_idx_mutex);
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0);
 
 		val = RREG32(reg_offset);
 
 		if (se_num != 0xffffffff || sh_num != 0xffffffff)
-			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 		return val;
 	} else {
-- 
GitLab


From 86b20703e4c5a3c39891def0a68e7438aeca9db9 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 27 Jul 2022 14:24:05 +0800
Subject: [PATCH 0146/1791] drm/amdgpu: add xcc index argument to rlc safe_mode
 func (v4)

v1: To support multple XCD case (Le)
v2: unify naming style (Le)
v3: apply the changes to gc v11_0 (Hawking)
v4: apply the changes to gc SOC21 (Morris)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Morris Zhang <Shiwu.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c       | 16 ++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h       | 10 +++---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        | 12 +++----
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        | 20 +++++------
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c         |  4 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c         | 20 +++++------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         | 20 +++++------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 16 ++++-----
 drivers/gpu/drm/amd/amdgpu/nv.c               |  4 +--
 drivers/gpu/drm/amd/amdgpu/soc21.c            |  4 +--
 drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c    |  6 ++--
 .../amd/pm/powerplay/hwmgr/smu7_powertune.c   | 12 +++----
 .../amd/pm/powerplay/hwmgr/vega10_powertune.c | 36 +++++++++----------
 13 files changed, 90 insertions(+), 90 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
index 85fb730d9fc84..d3bed9a3e61fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
@@ -34,9 +34,9 @@
  *
  * Set RLC enter into safe mode if RLC is enabled and haven't in safe mode.
  */
-void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev)
+void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
-	if (adev->gfx.rlc.in_safe_mode)
+	if (adev->gfx.rlc.in_safe_mode[xcc_id])
 		return;
 
 	/* if RLC is not enabled, do nothing */
@@ -46,8 +46,8 @@ void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev)
 	if (adev->cg_flags &
 	    (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG |
 	     AMD_CG_SUPPORT_GFX_3D_CGCG)) {
-		adev->gfx.rlc.funcs->set_safe_mode(adev);
-		adev->gfx.rlc.in_safe_mode = true;
+		adev->gfx.rlc.funcs->set_safe_mode(adev, xcc_id);
+		adev->gfx.rlc.in_safe_mode[xcc_id] = true;
 	}
 }
 
@@ -58,9 +58,9 @@ void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev)
  *
  * Set RLC exit safe mode if RLC is enabled and have entered into safe mode.
  */
-void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev)
+void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
-	if (!(adev->gfx.rlc.in_safe_mode))
+	if (!(adev->gfx.rlc.in_safe_mode[xcc_id]))
 		return;
 
 	/* if RLC is not enabled, do nothing */
@@ -70,8 +70,8 @@ void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev)
 	if (adev->cg_flags &
 	    (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG |
 	     AMD_CG_SUPPORT_GFX_3D_CGCG)) {
-		adev->gfx.rlc.funcs->unset_safe_mode(adev);
-		adev->gfx.rlc.in_safe_mode = false;
+		adev->gfx.rlc.funcs->unset_safe_mode(adev, xcc_id);
+		adev->gfx.rlc.in_safe_mode[xcc_id] = false;
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
index 23f060db9255c..80b263646966e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -157,8 +157,8 @@ typedef struct _RLC_TABLE_OF_CONTENT {
 
 struct amdgpu_rlc_funcs {
 	bool (*is_rlc_enabled)(struct amdgpu_device *adev);
-	void (*set_safe_mode)(struct amdgpu_device *adev);
-	void (*unset_safe_mode)(struct amdgpu_device *adev);
+	void (*set_safe_mode)(struct amdgpu_device *adev, int xcc_id);
+	void (*unset_safe_mode)(struct amdgpu_device *adev, int xcc_id);
 	int  (*init)(struct amdgpu_device *adev);
 	u32  (*get_csb_size)(struct amdgpu_device *adev);
 	void (*get_csb_buffer)(struct amdgpu_device *adev, volatile u32 *buffer);
@@ -201,7 +201,7 @@ struct amdgpu_rlc {
 	u32                     cp_table_size;
 
 	/* safe mode for updating CG/PG state */
-	bool in_safe_mode;
+	bool in_safe_mode[8];
 	const struct amdgpu_rlc_funcs *funcs;
 
 	/* for firmware data */
@@ -260,8 +260,8 @@ struct amdgpu_rlc {
 	struct amdgpu_rlcg_reg_access_ctrl reg_access_ctrl;
 };
 
-void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev);
-void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev);
+void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id);
+void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws);
 int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev);
 int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 323f5b8927ad1..8bd07ff596718 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -7571,7 +7571,7 @@ static bool gfx_v10_0_is_rlc_enabled(struct amdgpu_device *adev)
 	return (REG_GET_FIELD(rlc_cntl, RLC_CNTL, RLC_ENABLE_F32)) ? true : false;
 }
 
-static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -7612,7 +7612,7 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
@@ -7959,7 +7959,7 @@ static void gfx_v10_0_apply_medium_grain_clock_gating_workaround(struct amdgpu_d
 static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 					    bool enable)
 {
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	if (enable) {
 		/* enable FGCG firstly*/
@@ -7998,7 +7998,7 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 	     AMD_CG_SUPPORT_GFX_3D_CGLS))
 		gfx_v10_0_enable_gui_idle_interrupt(adev, enable);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -8092,11 +8092,11 @@ static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable)
 
 static void gfx_v10_cntl_pg(struct amdgpu_device *adev, bool enable)
 {
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	gfx_v10_cntl_power_gating(adev, enable);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 50d0ff9ca2599..d3c89e6c0c035 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -123,8 +123,8 @@ static int gfx_v11_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev);
 static void gfx_v11_0_ring_invalidate_tlbs(struct amdgpu_ring *ring,
 					   uint16_t pasid, uint32_t flush_type,
 					   bool all_hub, uint8_t dst_sel);
-static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev);
-static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev);
+static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id);
+static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id);
 static void gfx_v11_0_update_perf_clk(struct amdgpu_device *adev,
 				      bool enable);
 
@@ -4532,7 +4532,7 @@ static int gfx_v11_0_soft_reset(void *handle)
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL, GFX_IDLE_INT_ENABLE, 0);
 	WREG32_SOC15(GC, 0, regCP_INT_CNTL, tmp);
 
-	gfx_v11_0_set_safe_mode(adev);
+	gfx_v11_0_set_safe_mode(adev, 0);
 
 	for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 		for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
@@ -4632,7 +4632,7 @@ static int gfx_v11_0_soft_reset(void *handle)
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL, GFX_IDLE_INT_ENABLE, 1);
 	WREG32_SOC15(GC, 0, regCP_INT_CNTL, tmp);
 
-	gfx_v11_0_unset_safe_mode(adev);
+	gfx_v11_0_unset_safe_mode(adev, 0);
 
 	return gfx_v11_0_cp_resume(adev);
 }
@@ -4798,7 +4798,7 @@ static bool gfx_v11_0_is_rlc_enabled(struct amdgpu_device *adev)
 	return (REG_GET_FIELD(rlc_cntl, RLC_CNTL, RLC_ENABLE_F32)) ? true : false;
 }
 
-static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -4817,7 +4817,7 @@ static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, RLC_SAFE_MODE__CMD_MASK);
 }
@@ -5045,7 +5045,7 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade
 static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 					    bool enable)
 {
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	gfx_v11_0_update_coarse_grain_clock_gating(adev, enable);
 
@@ -5065,7 +5065,7 @@ static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 	     AMD_CG_SUPPORT_GFX_3D_CGLS))
 	        gfx_v11_0_enable_gui_idle_interrupt(adev, enable);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -5133,11 +5133,11 @@ static void gfx_v11_cntl_power_gating(struct amdgpu_device *adev, bool enable)
 
 static void gfx_v11_cntl_pg(struct amdgpu_device *adev, bool enable)
 {
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	gfx_v11_cntl_power_gating(adev, enable);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static int gfx_v11_0_set_powergating_state(void *handle,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index d055e44eee1d8..d56dda5fc5883 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -3362,7 +3362,7 @@ static bool gfx_v7_0_is_rlc_enabled(struct amdgpu_device *adev)
 	return true;
 }
 
-static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	u32 tmp, i, mask;
 
@@ -3384,7 +3384,7 @@ static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v7_0_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v7_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	u32 tmp;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index b604808761499..278416acf0608 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4903,7 +4903,7 @@ static int gfx_v8_0_hw_fini(void *handle)
 		pr_debug("For SRIOV client, shouldn't do anything.\n");
 		return 0;
 	}
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	if (!gfx_v8_0_wait_for_idle(adev))
 		gfx_v8_0_cp_enable(adev, false);
 	else
@@ -4912,7 +4912,7 @@ static int gfx_v8_0_hw_fini(void *handle)
 		adev->gfx.rlc.funcs->stop(adev);
 	else
 		pr_err("rlc is busy, skip halt rlc\n");
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -5377,7 +5377,7 @@ static int gfx_v8_0_set_powergating_state(void *handle,
 				AMD_PG_SUPPORT_RLC_SMU_HS |
 				AMD_PG_SUPPORT_CP |
 				AMD_PG_SUPPORT_GFX_DMG))
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	switch (adev->asic_type) {
 	case CHIP_CARRIZO:
 	case CHIP_STONEY:
@@ -5431,7 +5431,7 @@ static int gfx_v8_0_set_powergating_state(void *handle,
 				AMD_PG_SUPPORT_RLC_SMU_HS |
 				AMD_PG_SUPPORT_CP |
 				AMD_PG_SUPPORT_GFX_DMG))
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return 0;
 }
 
@@ -5536,7 +5536,7 @@ static bool gfx_v8_0_is_rlc_enabled(struct amdgpu_device *adev)
 	return true;
 }
 
-static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -5563,7 +5563,7 @@ static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -5622,7 +5622,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 {
 	uint32_t temp, data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
@@ -5718,7 +5718,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 		gfx_v8_0_wait_for_rlc_serdes(adev);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
@@ -5728,7 +5728,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
 
 	temp = data = RREG32(mmRLC_CGCG_CGLS_CTRL);
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
 		temp1 = data1 =	RREG32(mmRLC_CGTT_MGCG_OVERRIDE);
@@ -5811,7 +5811,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
 
 	gfx_v8_0_wait_for_rlc_serdes(adev);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 static int gfx_v8_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 					    bool enable)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 2fa7adef18a91..bce6919d666ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4623,7 +4623,7 @@ static bool gfx_v9_0_is_rlc_enabled(struct amdgpu_device *adev)
 	return true;
 }
 
-static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -4640,7 +4640,7 @@ static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
@@ -4651,7 +4651,7 @@ static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev)
 static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev,
 						bool enable)
 {
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	if ((adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) && enable) {
 		gfx_v9_0_enable_gfx_cg_power_gating(adev, true);
@@ -4663,7 +4663,7 @@ static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev,
 			gfx_v9_0_enable_gfx_pipeline_powergating(adev, false);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static void gfx_v9_0_update_gfx_mg_power_gating(struct amdgpu_device *adev,
@@ -4690,7 +4690,7 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 {
 	uint32_t data, def;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
@@ -4757,7 +4757,7 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
 		}
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
@@ -4768,7 +4768,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
 	if (!adev->gfx.num_gfx_rings)
 		return;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	/* Enable 3D CGCG/CGLS */
 	if (enable) {
@@ -4812,7 +4812,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
 			WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL_3D, data);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
@@ -4820,7 +4820,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
 {
 	uint32_t def, data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
 		def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE);
@@ -4864,7 +4864,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
 			WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, data);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 }
 
 static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index b67be666f38a1..baa10ee8ec697 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1090,14 +1090,14 @@ static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
 	return true;
 }
 
-static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev)
+static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
 
 	data = RLC_SAFE_MODE__CMD_MASK;
 	data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
-	WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, data);
+	WREG32_SOC15(GC, xcc_id, regRLC_SAFE_MODE, data);
 
 	/* wait for RLC_SAFE_MODE */
 	for (i = 0; i < adev->usec_timeout; i++) {
@@ -1107,12 +1107,12 @@ static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev)
 	}
 }
 
-static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev)
+static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
 	data = RLC_SAFE_MODE__CMD_MASK;
-	WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, data);
+	WREG32_SOC15(GC, xcc_id, regRLC_SAFE_MODE, data);
 }
 
 static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
@@ -2125,7 +2125,7 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 {
 	uint32_t data, def;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
 
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
@@ -2186,7 +2186,7 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 		}
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
 static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
@@ -2194,7 +2194,7 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 {
 	uint32_t def, data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
 		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
@@ -2238,7 +2238,7 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 			WREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL, data);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
 static int gfx_v9_4_3_update_gfx_clock_gating(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index 148049782f50d..dabeeab2f2ad8 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -629,9 +629,9 @@ static int nv_update_umd_stable_pstate(struct amdgpu_device *adev,
 				       bool enter)
 {
 	if (enter)
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	else
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	if (adev->gfx.funcs->update_perfmon_mgcg)
 		adev->gfx.funcs->update_perfmon_mgcg(adev, !enter);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 6ef4be9322d95..7d59303ca2f95 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -549,9 +549,9 @@ static int soc21_update_umd_stable_pstate(struct amdgpu_device *adev,
 					  bool enter)
 {
 	if (enter)
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 	else
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	if (adev->gfx.funcs->update_perfmon_mgcg)
 		adev->gfx.funcs->update_perfmon_mgcg(adev, !enter);
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
index f5e08b60f66ef..36c831b280ed9 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
@@ -508,19 +508,19 @@ static int kv_enable_didt(struct amdgpu_device *adev, bool enable)
 	    pi->caps_db_ramping ||
 	    pi->caps_td_ramping ||
 	    pi->caps_tcp_ramping) {
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 		if (enable) {
 			ret = kv_program_pt_config_registers(adev, didt_config_kv);
 			if (ret) {
-				amdgpu_gfx_rlc_exit_safe_mode(adev);
+				amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 				return ret;
 			}
 		}
 
 		kv_do_enable_didt(adev, enable);
 
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c
index 32a5a00fd8ae7..21be23ec3c79e 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c
@@ -973,7 +973,7 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr)
 	    PP_CAP(PHM_PlatformCaps_TDRamping) ||
 	    PP_CAP(PHM_PlatformCaps_TCPRamping)) {
 
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 		mutex_lock(&adev->grbm_idx_mutex);
 		value = 0;
 		value2 = cgs_read_register(hwmgr->device, mmGRBM_GFX_INDEX);
@@ -1048,13 +1048,13 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr)
 		}
 
 		mutex_unlock(&adev->grbm_idx_mutex);
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	}
 
 	return 0;
 error:
 	mutex_unlock(&adev->grbm_idx_mutex);
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return result;
 }
 
@@ -1068,7 +1068,7 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr)
 	    PP_CAP(PHM_PlatformCaps_TDRamping) ||
 	    PP_CAP(PHM_PlatformCaps_TCPRamping)) {
 
-		amdgpu_gfx_rlc_enter_safe_mode(adev);
+		amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 		result = smu7_enable_didt(hwmgr, false);
 		PP_ASSERT_WITH_CODE((result == 0),
@@ -1081,12 +1081,12 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr)
 			PP_ASSERT_WITH_CODE((0 == result),
 					"Failed to disable DPM DIDT.", goto error);
 		}
-		amdgpu_gfx_rlc_exit_safe_mode(adev);
+		amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	}
 
 	return 0;
 error:
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 	return result;
 }
 
diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c
index 9757d47dd6b81..309a9d3bc1b79 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c
@@ -915,7 +915,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
 
 	num_se = adev->gfx.config.max_shader_engines;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (count = 0; count < num_se; count++) {
@@ -940,7 +940,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
 
 	vega10_didt_set_mask(hwmgr, true);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -949,11 +949,11 @@ static int vega10_disable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
 {
 	struct amdgpu_device *adev = hwmgr->adev;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	vega10_didt_set_mask(hwmgr, false);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -966,7 +966,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
 
 	num_se = adev->gfx.config.max_shader_engines;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (count = 0; count < num_se; count++) {
@@ -985,7 +985,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
 
 	vega10_didt_set_mask(hwmgr, true);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	vega10_program_gc_didt_config_registers(hwmgr, GCDiDtDroopCtrlConfig_vega10);
 	if (PP_CAP(PHM_PlatformCaps_GCEDC))
@@ -1002,11 +1002,11 @@ static int vega10_disable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
 	struct amdgpu_device *adev = hwmgr->adev;
 	uint32_t data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	vega10_didt_set_mask(hwmgr, false);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	if (PP_CAP(PHM_PlatformCaps_GCEDC)) {
 		data = 0x00000000;
@@ -1027,7 +1027,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr)
 
 	num_se = adev->gfx.config.max_shader_engines;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (count = 0; count < num_se; count++) {
@@ -1048,7 +1048,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr)
 
 	vega10_didt_set_mask(hwmgr, true);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -1057,11 +1057,11 @@ static int vega10_disable_se_edc_config(struct pp_hwmgr *hwmgr)
 {
 	struct amdgpu_device *adev = hwmgr->adev;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	vega10_didt_set_mask(hwmgr, false);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
@@ -1075,7 +1075,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
 
 	num_se = adev->gfx.config.max_shader_engines;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	vega10_program_gc_didt_config_registers(hwmgr, AvfsPSMResetConfig_vega10);
 
@@ -1096,7 +1096,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
 
 	vega10_didt_set_mask(hwmgr, true);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	vega10_program_gc_didt_config_registers(hwmgr, PSMGCEDCDroopCtrlConfig_vega10);
 
@@ -1116,11 +1116,11 @@ static int vega10_disable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
 	struct amdgpu_device *adev = hwmgr->adev;
 	uint32_t data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	vega10_didt_set_mask(hwmgr, false);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	if (PP_CAP(PHM_PlatformCaps_GCEDC)) {
 		data = 0x00000000;
@@ -1138,7 +1138,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr)
 	struct amdgpu_device *adev = hwmgr->adev;
 	int result;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, 0);
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, 0xE0000000);
@@ -1151,7 +1151,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr)
 
 	vega10_didt_set_mask(hwmgr, false);
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, 0);
 
 	return 0;
 }
-- 
GitLab


From 96c7c2f4d5bd94b15fe63448c087f01607b56f4a Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@redhat.com>
Date: Tue, 18 Apr 2023 12:04:53 +0200
Subject: [PATCH 0147/1791] drm/scheduler: set entity to NULL in
 drm_sched_entity_pop_job()

It already happend a few times that patches slipped through which
implemented access to an entity through a job that was already removed
from the entities queue. Since jobs and entities might have different
lifecycles, this can potentially cause UAF bugs.

In order to make it obvious that a jobs entity pointer shouldn't be
accessed after drm_sched_entity_pop_job() was called successfully, set
the jobs entity pointer to NULL once the job is removed from the entity
queue.

Moreover, debugging a potential NULL pointer dereference is way easier
than potentially corrupted memory through a UAF.

Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Link: https://lore.kernel.org/r/20230418100453.4433-1-dakr@redhat.com
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
---
 drivers/gpu/drm/scheduler/sched_entity.c | 6 ++++++
 drivers/gpu/drm/scheduler/sched_main.c   | 4 ++++
 2 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index e0a8890a62e23..3e2a31d8190eb 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -448,6 +448,12 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 			drm_sched_rq_update_fifo(entity, next->submit_ts);
 	}
 
+	/* Jobs and entities might have different lifecycles. Since we're
+	 * removing the job from the entities queue, set the jobs entity pointer
+	 * to NULL to prevent any future access of the entity through this job.
+	 */
+	sched_job->entity = NULL;
+
 	return sched_job;
 }
 
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index f6801eb21820d..c3582c4fc2992 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -42,6 +42,10 @@
  *    the hardware.
  *
  * The jobs in a entity are always scheduled in the order that they were pushed.
+ *
+ * Note that once a job was taken from the entities queue and pushed to the
+ * hardware, i.e. the pending queue, the entity must not be referenced anymore
+ * through the jobs entity pointer.
  */
 
 #include <linux/kthread.h>
-- 
GitLab


From 27ac123b454417ea92d77c13a5d94655f53b759c Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 14 Apr 2023 20:38:00 +0300
Subject: [PATCH 0148/1791] drm/i915/dp_mst: Fix active port PLL selection for
 secondary MST streams

The port PLL selection needs to be up-to-date in the CRTC state of both
the primary and all secondary MST streams. The commit removing the
encoder update_prepare/complete hooks (see Fixes: below), stopped doing
this for secondary streams, fix this up.

Fixes: 0f752b2178c9 ("drm/i915: Remove the encoder update_prepare()/complete() hooks")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8336
Cc: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230414173800.590790-1-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c    | 27 +++++++++++++++------
 drivers/gpu/drm/i915/display/intel_ddi.h    |  3 +++
 drivers/gpu/drm/i915/display/intel_dp_mst.c |  7 ++++++
 3 files changed, 30 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 2ef1af921a0b1..8fb0f205b553e 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3349,6 +3349,25 @@ void intel_ddi_update_pipe(struct intel_atomic_state *state,
 	intel_hdcp_update_pipe(state, encoder, crtc_state, conn_state);
 }
 
+void intel_ddi_update_active_dpll(struct intel_atomic_state *state,
+				  struct intel_encoder *encoder,
+				  struct intel_crtc *crtc)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_crtc_state *crtc_state =
+		intel_atomic_get_new_crtc_state(state, crtc);
+	struct intel_crtc *slave_crtc;
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	if (!intel_phy_is_tc(i915, phy))
+		return;
+
+	intel_update_active_dpll(state, crtc, encoder);
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, slave_crtc,
+					 intel_crtc_bigjoiner_slave_pipes(crtc_state))
+		intel_update_active_dpll(state, slave_crtc, encoder);
+}
+
 static void
 intel_ddi_pre_pll_enable(struct intel_atomic_state *state,
 			 struct intel_encoder *encoder,
@@ -3363,15 +3382,9 @@ intel_ddi_pre_pll_enable(struct intel_atomic_state *state,
 	if (is_tc_port) {
 		struct intel_crtc *master_crtc =
 			to_intel_crtc(crtc_state->uapi.crtc);
-		struct intel_crtc *slave_crtc;
 
 		intel_tc_port_get_link(dig_port, crtc_state->lane_count);
-
-		intel_update_active_dpll(state, master_crtc, encoder);
-
-		for_each_intel_crtc_in_pipe_mask(&dev_priv->drm, slave_crtc,
-						 intel_crtc_bigjoiner_slave_pipes(crtc_state))
-			intel_update_active_dpll(state, slave_crtc, encoder);
+		intel_ddi_update_active_dpll(state, encoder, master_crtc);
 	}
 
 	main_link_aux_power_domain_get(dig_port, crtc_state);
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h
index c85e74ae68e4d..2bc034042a937 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.h
+++ b/drivers/gpu/drm/i915/display/intel_ddi.h
@@ -72,5 +72,8 @@ void intel_ddi_sanitize_encoder_pll_mapping(struct intel_encoder *encoder);
 int intel_ddi_level(struct intel_encoder *encoder,
 		    const struct intel_crtc_state *crtc_state,
 		    int lane);
+void intel_ddi_update_active_dpll(struct intel_atomic_state *state,
+				  struct intel_encoder *encoder,
+				  struct intel_crtc *crtc);
 
 #endif /* __INTEL_DDI_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index a88b852c437c4..2c49d9ab86a2a 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -674,6 +674,13 @@ static void intel_mst_pre_pll_enable_dp(struct intel_atomic_state *state,
 	if (intel_dp->active_mst_links == 0)
 		dig_port->base.pre_pll_enable(state, &dig_port->base,
 						    pipe_config, NULL);
+	else
+		/*
+		 * The port PLL state needs to get updated for secondary
+		 * streams as for the primary stream.
+		 */
+		intel_ddi_update_active_dpll(state, &dig_port->base,
+					     to_intel_crtc(pipe_config->uapi.crtc));
 }
 
 static void intel_mst_pre_enable_dp(struct intel_atomic_state *state,
-- 
GitLab


From 476f62b8a597202a7c97bf50a7f6ece0925ce6f0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 19 Apr 2023 12:42:43 +0300
Subject: [PATCH 0149/1791] drm/i915: use explicit includes for i915_reg.h and
 i915_irq.h

A lot of places include i915_reg.h implicitly via i915_irq.h, which gets
included implicitly via intel_display_trace.h. Remove the includes from
the headers, and include i915_reg.h and i915_irq.h explicitly where
needed.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419094243.366821-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/i9xx_wm.c                  | 1 +
 drivers/gpu/drm/i915/display/intel_atomic_plane.c       | 1 +
 drivers/gpu/drm/i915/display/intel_display_power.c      | 1 +
 drivers/gpu/drm/i915/display/intel_display_power_well.c | 1 +
 drivers/gpu/drm/i915/display/intel_display_trace.h      | 1 -
 drivers/gpu/drm/i915/display/intel_fbc.c                | 1 +
 drivers/gpu/drm/i915/display/intel_fifo_underrun.c      | 2 ++
 drivers/gpu/drm/i915/gt/intel_gt_irq.c                  | 1 +
 drivers/gpu/drm/i915/gt/intel_reset.c                   | 1 +
 drivers/gpu/drm/i915/gt/intel_rps.c                     | 1 +
 drivers/gpu/drm/i915/gt/selftest_rps.c                  | 1 +
 drivers/gpu/drm/i915/gt/uc/intel_guc.c                  | 1 +
 drivers/gpu/drm/i915/i915_debugfs.c                     | 1 +
 drivers/gpu/drm/i915/i915_irq.c                         | 1 +
 drivers/gpu/drm/i915/i915_irq.h                         | 2 +-
 drivers/gpu/drm/i915/intel_clock_gating.c               | 1 +
 16 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index 6288826a94974..af0c79a4c9a4f 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -4,6 +4,7 @@
  */
 
 #include "i915_drv.h"
+#include "i915_reg.h"
 #include "i9xx_wm.h"
 #include "intel_atomic.h"
 #include "intel_display.h"
diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
index 40de9f0f171b4..9f670dcfe76e4 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
@@ -36,6 +36,7 @@
 #include <drm/drm_fourcc.h>
 
 #include "i915_config.h"
+#include "i915_reg.h"
 #include "intel_atomic_plane.h"
 #include "intel_cdclk.h"
 #include "intel_display_rps.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 7c9f4288329ed..5150069f3f821 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -7,6 +7,7 @@
 
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "intel_backlight_regs.h"
 #include "intel_cdclk.h"
 #include "intel_combo_phy.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 62bafcbc7937c..1a19fd3bf103e 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -5,6 +5,7 @@
 
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "intel_backlight_regs.h"
 #include "intel_combo_phy.h"
 #include "intel_combo_phy_regs.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_trace.h b/drivers/gpu/drm/i915/display/intel_display_trace.h
index 651ea8564e1b7..99bdb833591ce 100644
--- a/drivers/gpu/drm/i915/display/intel_display_trace.h
+++ b/drivers/gpu/drm/i915/display/intel_display_trace.h
@@ -14,7 +14,6 @@
 #include <linux/tracepoint.h>
 
 #include "i915_drv.h"
-#include "i915_irq.h"
 #include "intel_crtc.h"
 #include "intel_display_types.h"
 #include "intel_vblank.h"
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index b507ff944864e..11bb8cf9c9d0d 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -44,6 +44,7 @@
 #include <drm/drm_fourcc.h>
 
 #include "i915_drv.h"
+#include "i915_reg.h"
 #include "i915_utils.h"
 #include "i915_vgpu.h"
 #include "intel_cdclk.h"
diff --git a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
index b708a62e509a0..e7f77a225739c 100644
--- a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
+++ b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
@@ -26,6 +26,8 @@
  */
 
 #include "i915_drv.h"
+#include "i915_reg.h"
+#include "i915_irq.h"
 #include "intel_de.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index 1b25a60391522..c0f3ff4746add 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -7,6 +7,7 @@
 
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "intel_breadcrumbs.h"
 #include "intel_gt.h"
 #include "intel_gt_irq.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index c129138f666f0..6be28fe7f8f8f 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -18,6 +18,7 @@
 #include "i915_file_private.h"
 #include "i915_gpu_error.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "intel_breadcrumbs.h"
 #include "intel_engine_pm.h"
 #include "intel_engine_regs.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 4d0dc9de23f96..bc424d956cced 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -10,6 +10,7 @@
 #include "display/intel_display.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "intel_breadcrumbs.h"
 #include "intel_gt.h"
 #include "intel_gt_clock_utils.h"
diff --git a/drivers/gpu/drm/i915/gt/selftest_rps.c b/drivers/gpu/drm/i915/gt/selftest_rps.c
index 84e77e8dbba19..fb30f733b0366 100644
--- a/drivers/gpu/drm/i915/gt/selftest_rps.c
+++ b/drivers/gpu/drm/i915/gt/selftest_rps.c
@@ -8,6 +8,7 @@
 
 #include "gem/i915_gem_internal.h"
 
+#include "i915_reg.h"
 #include "intel_engine_heartbeat.h"
 #include "intel_engine_pm.h"
 #include "intel_engine_regs.h"
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
index d76508fa3af7f..e89f16ecf1ae4 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
@@ -16,6 +16,7 @@
 #include "intel_guc_submission.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 
 /**
  * DOC: GuC
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 16011c0286ada..7545de191a405 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -50,6 +50,7 @@
 #include "i915_debugfs_params.h"
 #include "i915_driver.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 #include "i915_scheduler.h"
 #include "intel_mchbar_regs.h"
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 4345931c65965..b7a19db75c9cb 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -54,6 +54,7 @@
 #include "i915_driver.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_reg.h"
 
 /**
  * DOC: interrupt handling
diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h
index 3717a66f97c60..dd47e473ba4f7 100644
--- a/drivers/gpu/drm/i915/i915_irq.h
+++ b/drivers/gpu/drm/i915/i915_irq.h
@@ -9,7 +9,7 @@
 #include <linux/ktime.h>
 #include <linux/types.h>
 
-#include "i915_reg.h"
+#include "i915_reg_defs.h"
 
 enum pipe;
 struct drm_crtc;
diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index 2c5302bcba19c..1d796b76f71b9 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -36,6 +36,7 @@
 #include "gt/intel_gt_regs.h"
 
 #include "i915_drv.h"
+#include "i915_reg.h"
 #include "intel_clock_gating.h"
 #include "intel_mchbar_regs.h"
 #include "vlv_sideband.h"
-- 
GitLab


From a6704f4a5452950e7ff22cab0cff23477dbdf0e0 Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Wed, 19 Apr 2023 11:30:34 +0530
Subject: [PATCH 0150/1791] drm/i915/gt: Consider multi-gt instead of to_gt()

In order to enable complete multi-GT, use the GT
reference obtained directly from the engine, rather
than relying on the to_gt(), which only provides a
reference to the primary GT.

Problem appear when it runs on platform like MTL
where different set of engines are possible on
different GTs.

Cc: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419060036.3422635-2-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gt/intel_engine_user.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_user.c b/drivers/gpu/drm/i915/gt/intel_engine_user.c
index cd4f1b126f753..dcedff41a825f 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_user.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_user.c
@@ -117,7 +117,7 @@ static void set_scheduler_caps(struct drm_i915_private *i915)
 			disabled |= (I915_SCHEDULER_CAP_ENABLED |
 				     I915_SCHEDULER_CAP_PRIORITY);
 
-		if (intel_uc_uses_guc_submission(&to_gt(i915)->uc))
+		if (intel_uc_uses_guc_submission(&engine->gt->uc))
 			enabled |= I915_SCHEDULER_CAP_STATIC_PRIORITY_MAP;
 
 		for (i = 0; i < ARRAY_SIZE(map); i++) {
-- 
GitLab


From a347279dec1eb68e995f864ae1fd41ab57cbcd21 Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Wed, 19 Apr 2023 11:30:35 +0530
Subject: [PATCH 0151/1791] drm/i915/gem: Consider multi-gt instead of to_gt()

In order to enable complete multi-GT, use the GT
reference obtained directly from the engine, rather
than relying on the to_gt(), which only provides a
reference to the primary GT.

Problem appear when it runs on platform like MTL
where different set of engines are possible on
different GTs.

Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419060036.3422635-3-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
index a81fa6a20f5aa..2697fbaa2cebb 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
@@ -93,7 +93,7 @@ static int live_nop_switch(void *arg)
 		}
 		if (i915_request_wait(rq, 0, 10 * HZ) < 0) {
 			pr_err("Failed to populated %d contexts\n", nctx);
-			intel_gt_set_wedged(to_gt(i915));
+			intel_gt_set_wedged(engine->gt);
 			i915_request_put(rq);
 			err = -EIO;
 			goto out_file;
@@ -149,7 +149,7 @@ static int live_nop_switch(void *arg)
 			if (i915_request_wait(rq, 0, HZ / 5) < 0) {
 				pr_err("Switching between %ld contexts timed out\n",
 				       prime);
-				intel_gt_set_wedged(to_gt(i915));
+				intel_gt_set_wedged(engine->gt);
 				i915_request_put(rq);
 				break;
 			}
-- 
GitLab


From 0c29efa23f5c2f51b744856200af0cff3e287e9e Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Wed, 19 Apr 2023 11:30:36 +0530
Subject: [PATCH 0152/1791] drm/i915/selftests: Consider multi-gt instead of
 to_gt()

In order to enable complete multi-GT, loop through all
the GTs, rather than relying on the to_gt(), which only
provides a reference to the primary GT.

Problem appear when it runs on platform like MTL where
different set of engines are possible on different GTs.

Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419060036.3422635-4-tejas.upadhyay@intel.com
---
 .../gpu/drm/i915/selftests/igt_live_test.c    | 46 +++++++++++--------
 1 file changed, 27 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/i915/selftests/igt_live_test.c b/drivers/gpu/drm/i915/selftests/igt_live_test.c
index 72b58b66692ad..714b7afc490b3 100644
--- a/drivers/gpu/drm/i915/selftests/igt_live_test.c
+++ b/drivers/gpu/drm/i915/selftests/igt_live_test.c
@@ -16,27 +16,31 @@ int igt_live_test_begin(struct igt_live_test *t,
 			const char *func,
 			const char *name)
 {
-	struct intel_gt *gt = to_gt(i915);
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
+	struct intel_gt *gt;
+	unsigned int i;
 	int err;
 
 	t->i915 = i915;
 	t->func = func;
 	t->name = name;
 
-	err = intel_gt_wait_for_idle(gt, MAX_SCHEDULE_TIMEOUT);
-	if (err) {
-		pr_err("%s(%s): failed to idle before, with err=%d!",
-		       func, name, err);
-		return err;
-	}
+	for_each_gt(gt, i915, i) {
 
-	t->reset_global = i915_reset_count(&i915->gpu_error);
+		err = intel_gt_wait_for_idle(gt, MAX_SCHEDULE_TIMEOUT);
+		if (err) {
+			pr_err("%s(%s): failed to idle before, with err=%d!",
+			       func, name, err);
+			return err;
+		}
 
-	for_each_engine(engine, gt, id)
-		t->reset_engine[id] =
+		for_each_engine(engine, gt, id)
+			t->reset_engine[id] =
 			i915_reset_engine_count(&i915->gpu_error, engine);
+	}
+
+	t->reset_global = i915_reset_count(&i915->gpu_error);
 
 	return 0;
 }
@@ -46,6 +50,8 @@ int igt_live_test_end(struct igt_live_test *t)
 	struct drm_i915_private *i915 = t->i915;
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
+	struct intel_gt *gt;
+	unsigned int i;
 
 	if (igt_flush_test(i915))
 		return -EIO;
@@ -57,16 +63,18 @@ int igt_live_test_end(struct igt_live_test *t)
 		return -EIO;
 	}
 
-	for_each_engine(engine, to_gt(i915), id) {
-		if (t->reset_engine[id] ==
-		    i915_reset_engine_count(&i915->gpu_error, engine))
-			continue;
+	for_each_gt(gt, i915, i) {
+		for_each_engine(engine, gt, id) {
+			if (t->reset_engine[id] ==
+			    i915_reset_engine_count(&i915->gpu_error, engine))
+				continue;
 
-		pr_err("%s(%s): engine '%s' was reset %d times!\n",
-		       t->func, t->name, engine->name,
-		       i915_reset_engine_count(&i915->gpu_error, engine) -
-		       t->reset_engine[id]);
-		return -EIO;
+			pr_err("%s(%s): engine '%s' was reset %d times!\n",
+			       t->func, t->name, engine->name,
+			       i915_reset_engine_count(&i915->gpu_error, engine) -
+			       t->reset_engine[id]);
+			return -EIO;
+		}
 	}
 
 	return 0;
-- 
GitLab


From 88c487938414c519fdb1c7e55211d8778d3367d0 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 18 Apr 2023 15:04:43 -0700
Subject: [PATCH 0153/1791] drm/i915: Use separate "DC off" power well for
 ADL-P and DG2

Although ADL-P and DG2 both use the same general power well setup, the
DC5/DC6 requirements are slightly different which means each platform
should have its own "DC off" power well.

DG2 (i.e., Xe_HPD IP) requires that DC5 be disabled whenever PG2 is
active.  However ADL-P (i.e., Xe_LPD IP) only requires DC5/DC6 to be
disabled when the PGC or PGD subwells are active; we should be able to
remain in these DC states when PGB and general PG2 functionality is in
use.

v2: Use dc_of as power well name.
    Move xehpd power domain definitions near power well definition.(Imre)

Bspec: 49193
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418220446.2205509-2-radhakrishna.sripada@intel.com
---
 .../i915/display/intel_display_power_map.c    | 41 +++++++++++++++++--
 1 file changed, 38 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index 6645eb1911d85..5906b62e79f15 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1301,7 +1301,8 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_2,
  */
 
 I915_DECL_PW_DOMAINS(xelpd_pwdoms_dc_off,
-	XELPD_PW_2_POWER_DOMAINS,
+	XELPD_PW_C_POWER_DOMAINS,
+	XELPD_PW_D_POWER_DOMAINS,
 	POWER_DOMAIN_PORT_DSI,
 	POWER_DOMAIN_AUDIO_MMIO,
 	POWER_DOMAIN_AUX_A,
@@ -1310,14 +1311,18 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_dc_off,
 	POWER_DOMAIN_DC_OFF,
 	POWER_DOMAIN_INIT);
 
-static const struct i915_power_well_desc xelpd_power_wells_main[] = {
+static const struct i915_power_well_desc xelpd_power_wells_dc_off[] = {
 	{
 		.instances = &I915_PW_INSTANCES(
 			I915_PW("DC_off", &xelpd_pwdoms_dc_off,
 				.id = SKL_DISP_DC_OFF),
 		),
 		.ops = &gen9_dc_off_power_well_ops,
-	}, {
+	}
+};
+
+static const struct i915_power_well_desc xelpd_power_wells_main[] = {
+	{
 		.instances = &I915_PW_INSTANCES(
 			I915_PW("PW_2", &xelpd_pwdoms_pw_2,
 				.hsw.idx = ICL_PW_CTL_IDX_PW_2,
@@ -1400,6 +1405,34 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = {
 static const struct i915_power_well_desc_list xelpd_power_wells[] = {
 	I915_PW_DESCRIPTORS(i9xx_power_wells_always_on),
 	I915_PW_DESCRIPTORS(icl_power_wells_pw_1),
+	I915_PW_DESCRIPTORS(xelpd_power_wells_dc_off),
+	I915_PW_DESCRIPTORS(xelpd_power_wells_main),
+};
+
+I915_DECL_PW_DOMAINS(xehpd_pwdoms_dc_off,
+	XELPD_PW_2_POWER_DOMAINS,
+	POWER_DOMAIN_PORT_DSI,
+	POWER_DOMAIN_AUDIO_MMIO,
+	POWER_DOMAIN_AUX_A,
+	POWER_DOMAIN_AUX_B,
+	POWER_DOMAIN_MODESET,
+	POWER_DOMAIN_DC_OFF,
+	POWER_DOMAIN_INIT);
+
+static const struct i915_power_well_desc xehpd_power_wells_dc_off[] = {
+	{
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("DC_off", &xehpd_pwdoms_dc_off,
+				.id = SKL_DISP_DC_OFF),
+		),
+		.ops = &gen9_dc_off_power_well_ops,
+	}
+};
+
+static const struct i915_power_well_desc_list xehpd_power_wells[] = {
+	I915_PW_DESCRIPTORS(i9xx_power_wells_always_on),
+	I915_PW_DESCRIPTORS(icl_power_wells_pw_1),
+	I915_PW_DESCRIPTORS(xehpd_power_wells_dc_off),
 	I915_PW_DESCRIPTORS(xelpd_power_wells_main),
 };
 
@@ -1624,6 +1657,8 @@ int intel_display_power_map_init(struct i915_power_domains *power_domains)
 
 	if (DISPLAY_VER(i915) >= 14)
 		return set_power_wells(power_domains, xelpdp_power_wells);
+	else if (IS_DG2(i915))
+		return set_power_wells(power_domains, xehpd_power_wells);
 	else if (DISPLAY_VER(i915) >= 13)
 		return set_power_wells(power_domains, xelpd_power_wells);
 	else if (IS_DG1(i915))
-- 
GitLab


From 0c8d9870177a2d7c9e88a2e79dc20950ec84328f Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 18 Apr 2023 15:04:44 -0700
Subject: [PATCH 0154/1791] drm/i915/mtl: Re-use ADL-P's "DC off" power well

As with ADL-P, MTL's "DC off" power well should be a dependency of the
PGC and PGD power wells, not the entire PG2 well.  In fact, the DC5/DC6
requirements between the two platforms are the same, so the Xe_LPD "DC
off" well definition can just be re-used for Xe_LPD+.

Bspec: 49193
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418220446.2205509-3-radhakrishna.sripada@intel.com
---
 .../drm/i915/display/intel_display_power_map.c   | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index 5906b62e79f15..100582f105905 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1456,15 +1456,6 @@ I915_DECL_PW_DOMAINS(xelpdp_pwdoms_pw_2,
 	XELPDP_PW_2_POWER_DOMAINS,
 	POWER_DOMAIN_INIT);
 
-I915_DECL_PW_DOMAINS(xelpdp_pwdoms_dc_off,
-	XELPDP_PW_2_POWER_DOMAINS,
-	POWER_DOMAIN_AUDIO_MMIO,
-	POWER_DOMAIN_MODESET,
-	POWER_DOMAIN_AUX_A,
-	POWER_DOMAIN_AUX_B,
-	POWER_DOMAIN_DC_OFF,
-	POWER_DOMAIN_INIT);
-
 I915_DECL_PW_DOMAINS(xelpdp_pwdoms_aux_tc1,
 	POWER_DOMAIN_AUX_USBC1,
 	POWER_DOMAIN_AUX_TBT1);
@@ -1483,12 +1474,6 @@ I915_DECL_PW_DOMAINS(xelpdp_pwdoms_aux_tc4,
 
 static const struct i915_power_well_desc xelpdp_power_wells_main[] = {
 	{
-		.instances = &I915_PW_INSTANCES(
-			I915_PW("DC_off", &xelpdp_pwdoms_dc_off,
-				.id = SKL_DISP_DC_OFF),
-		),
-		.ops = &gen9_dc_off_power_well_ops,
-	}, {
 		.instances = &I915_PW_INSTANCES(
 			I915_PW("PW_2", &xelpdp_pwdoms_pw_2,
 				.hsw.idx = ICL_PW_CTL_IDX_PW_2,
@@ -1545,6 +1530,7 @@ static const struct i915_power_well_desc xelpdp_power_wells_main[] = {
 static const struct i915_power_well_desc_list xelpdp_power_wells[] = {
 	I915_PW_DESCRIPTORS(i9xx_power_wells_always_on),
 	I915_PW_DESCRIPTORS(icl_power_wells_pw_1),
+	I915_PW_DESCRIPTORS(xelpd_power_wells_dc_off),
 	I915_PW_DESCRIPTORS(xelpdp_power_wells_main),
 };
 
-- 
GitLab


From 514b8a79aa85d800458cfb7909ed0e1a1c1bffa2 Mon Sep 17 00:00:00 2001
From: Madhumitha Tolakanahalli Pradeep
 <madhumitha.tolakanahalli.pradeep@intel.com>
Date: Tue, 18 Apr 2023 15:04:45 -0700
Subject: [PATCH 0155/1791] drm/i915/mtl: Extend Wa_22011802037 to MTL A-step

Wa_22011802037 was being applied to all graphics_ver 11 & 12. This patch
updates the if statement to apply the W/A to right platforms and extends
it to MTL-M:A step.

v1.1: Fix checkpatch warning.
v2: Change the check to reflect the wa at other places(Lucas)

Bspec: 66622
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Madhumitha Tolakanahalli Pradeep <madhumitha.tolakanahalli.pradeep@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418220446.2205509-4-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index 88e881b100cf0..ee3e8352637f2 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -1629,16 +1629,16 @@ static void guc_reset_state(struct intel_context *ce, u32 head, bool scrub)
 
 static void guc_engine_reset_prepare(struct intel_engine_cs *engine)
 {
-	if (!IS_GRAPHICS_VER(engine->i915, 11, 12))
-		return;
-
-	intel_engine_stop_cs(engine);
-
 	/*
 	 * Wa_22011802037: In addition to stopping the cs, we need
 	 * to wait for any pending mi force wakeups
 	 */
-	intel_engine_wait_for_pending_mi_fw(engine);
+	if (IS_MTL_GRAPHICS_STEP(engine->i915, M, STEP_A0, STEP_B0) ||
+	    (GRAPHICS_VER(engine->i915) >= 11 &&
+	     GRAPHICS_VER_FULL(engine->i915) < IP_VER(12, 70))) {
+		intel_engine_stop_cs(engine);
+		intel_engine_wait_for_pending_mi_fw(engine);
+	}
 }
 
 static void guc_reset_nop(struct intel_engine_cs *engine)
-- 
GitLab


From c91acda3a380bcaf41b67c8fbab668ef8ddf91c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Wed, 12 Apr 2023 11:29:23 -0300
Subject: [PATCH 0156/1791] drm/gem: Check for valid formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, drm_gem_fb_create() doesn't check if the pixel format is
supported, which can lead to the acceptance of invalid pixel formats
e.g. the acceptance of invalid modifiers. Therefore, add a check for
valid formats on drm_gem_fb_create().

Note that this check is only valid for atomic drivers, because, for
non-atomic drivers, checking drm_any_plane_has_format() is not
possible since the format list for the primary plane is fake, and we'd
therefore reject valid formats.

Adding this check to drm_gem_fb_create() will guarantee that the
igt@kms_addfb_basic@addfb25-bad-modifier IGT test passes for drivers
using this callback.

This commit is a recapture of a series sent a while ago. Initially,
I sent a patch [1] similar to this one in which I introduced the
format check to drm_gem_fb_create().

Based on the feedback on the patch, I placed the check inside
framebuffer_check() [2] so that it wouldn't be needed to hit any
driver-specific code path when the check fails. Therefore, we could
remove the check from the specific drivers (i915, amdgpu, and vmwgfx).

But, with some new feedback, it was shown that introducing this check
inside framebuffer_check() is problematic for the i915 driver [3].
For the i915 driver, in the legacy case, in which we don't get the
modifier from the userspace, i915's fb_create hook computes the right
modifier, which isn't necessarily linear.  Therefore, if we check the
modifier before that point, we might get wrong answers.

So, I kept the check inside the i915 driver and removed the check from
amdgpu and vmwgfx [4]. But, this yet hasn't solved the i915 problem [5].

As we cannot add the check inside framebuffer_check() without
affecting the i915 behavior, this commit went back to the original
patch. This way we can guarantee a more uniform behavior from the
drivers that use the drm_gem_fb_create() callback.

[1] https://lore.kernel.org/dri-devel/20230103125322.855089-1-mcanal@igalia.com/T/
[2] https://lore.kernel.org/dri-devel/20230109105807.18172-1-mcanal@igalia.com/T/
[3] https://lore.kernel.org/dri-devel/Y8AAdW2y7zN7DCUZ@intel.com/
[4] https://lore.kernel.org/dri-devel/20230113112743.188486-1-mcanal@igalia.com/T/
[5] https://lore.kernel.org/dri-devel/Y8FXWvEhO7GCRKVJ@intel.com/

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412142923.136707-1-mcanal@igalia.com
---
 Documentation/gpu/todo.rst                   | 7 ++-----
 drivers/gpu/drm/drm_gem_framebuffer_helper.c | 9 +++++++++
 2 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst
index 1f8a5ebe188eb..68bdafa0284f5 100644
--- a/Documentation/gpu/todo.rst
+++ b/Documentation/gpu/todo.rst
@@ -276,11 +276,8 @@ Various hold-ups:
 - Need to switch to drm_fbdev_generic_setup(), otherwise a lot of the custom fb
   setup code can't be deleted.
 
-- Many drivers wrap drm_gem_fb_create() only to check for valid formats. For
-  atomic drivers we could check for valid formats by calling
-  drm_plane_check_pixel_format() against all planes, and pass if any plane
-  supports the format. For non-atomic that's not possible since like the format
-  list for the primary plane is fake and we'd therefor reject valid formats.
+- Need to switch to drm_gem_fb_create(), as now drm_gem_fb_create() checks for
+  valid formats for atomic drivers.
 
 - Many drivers subclass drm_framebuffer, we'd need a embedding compatible
   version of the varios drm_gem_fb_create functions. Maybe called
diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
index e93533b86037f..b8a615a138cd6 100644
--- a/drivers/gpu/drm/drm_gem_framebuffer_helper.c
+++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 
 #include <drm/drm_damage_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_framebuffer.h>
 #include <drm/drm_gem.h>
@@ -164,6 +165,14 @@ int drm_gem_fb_init_with_funcs(struct drm_device *dev,
 		return -EINVAL;
 	}
 
+	if (drm_drv_uses_atomic_modeset(dev) &&
+	    !drm_any_plane_has_format(dev, mode_cmd->pixel_format,
+				      mode_cmd->modifier[0])) {
+		drm_dbg(dev, "Unsupported pixel format %p4cc / modifier 0x%llx\n",
+			&mode_cmd->pixel_format, mode_cmd->modifier[0]);
+		return -EINVAL;
+	}
+
 	for (i = 0; i < info->num_planes; i++) {
 		unsigned int width = mode_cmd->width / (i ? info->hsub : 1);
 		unsigned int height = mode_cmd->height / (i ? info->vsub : 1);
-- 
GitLab


From 3bece767dab2ffbd6f77be1a0ceb3e214f0144f8 Mon Sep 17 00:00:00 2001
From: Haridhar Kalvala <haridhar.kalvala@intel.com>
Date: Tue, 18 Apr 2023 15:04:46 -0700
Subject: [PATCH 0157/1791] drm/i915/mtl: WA to clear RDOP clock gating

Workaround implementation to clear RDOP clock gating.

Bspec: 66622
Signed-off-by: Haridhar Kalvala <haridhar.kalvala@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418220446.2205509-5-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index b925ef47304b6..312eb8b5f9491 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -1703,6 +1703,9 @@ xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 		/* Wa_18018781329 */
 		wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB);
 		wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB);
+
+		/* Wa_14015795083 */
+		wa_write_clr(wal, GEN7_MISCCPCTL, GEN12_DOP_CLOCK_GATE_RENDER_ENABLE);
 	}
 
 	/*
-- 
GitLab


From ba1fb317bbcb67d8024f0e83e509e8c57da35ac5 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 19 Apr 2023 14:54:49 +0300
Subject: [PATCH 0158/1791] drm/i915/wm: remove stale and unused
 ilk_wm_max_level() declaration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The declaration was removed earlier, but got accidentally resurrected in
i915xx_wm.[ch] refactoring. Remove harder.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419115449.422214-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/i9xx_wm.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.h b/drivers/gpu/drm/i915/display/i9xx_wm.h
index a7875cbcd05aa..b87ae369685a9 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.h
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.h
@@ -12,7 +12,6 @@ struct drm_i915_private;
 struct intel_crtc_state;
 struct intel_plane_state;
 
-int ilk_wm_max_level(const struct drm_i915_private *i915);
 bool ilk_disable_lp_wm(struct drm_i915_private *i915);
 void ilk_wm_sanitize(struct drm_i915_private *i915);
 bool intel_set_memory_cxsr(struct drm_i915_private *i915, bool enable);
-- 
GitLab


From 91254a4d2e01b8a91c6f0929b77e5fff676a0751 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:33 +0200
Subject: [PATCH 0159/1791] fbdev: Prepare generic architecture helpers

Generic implementations of fb_pgprotect() and fb_is_primary_device()
have been in the source code for a long time. Prepare the header file
to make use of them.

Improve the code by using an inline function for fb_pgprotect()
and by removing include statements. The default mode set by
fb_pgprotect() is now writecombine, which is what most platforms
want.

Symbols are protected by preprocessor guards. Architectures that
provide a symbol need to define a preprocessor token of the same
name and value. Otherwise the header file will provide a generic
implementation. This pattern has been taken from <asm/io.h>.

v3:
	* include the correct header files
v2:
	*  use writecombine mappings by default (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-2-tzimmermann@suse.de
---
 include/asm-generic/fb.h | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/include/asm-generic/fb.h b/include/asm-generic/fb.h
index f9f18101ed362..c8af99f5a5358 100644
--- a/include/asm-generic/fb.h
+++ b/include/asm-generic/fb.h
@@ -1,13 +1,33 @@
 /* SPDX-License-Identifier: GPL-2.0 */
+
 #ifndef __ASM_GENERIC_FB_H_
 #define __ASM_GENERIC_FB_H_
-#include <linux/fb.h>
 
-#define fb_pgprotect(...) do {} while (0)
+/*
+ * Only include this header file from your architecture's <asm/fb.h>.
+ */
+
+#include <linux/mm_types.h>
+#include <linux/pgtable.h>
+
+struct fb_info;
+struct file;
+
+#ifndef fb_pgprotect
+#define fb_pgprotect fb_pgprotect
+static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
+				unsigned long off)
+{
+	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
+}
+#endif
 
+#ifndef fb_is_primary_device
+#define fb_is_primary_device fb_is_primary_device
 static inline int fb_is_primary_device(struct fb_info *info)
 {
 	return 0;
 }
+#endif
 
 #endif /* __ASM_GENERIC_FB_H_ */
-- 
GitLab


From 3a6155052b07157f3af49e1e63f68cb1943f7de9 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:34 +0200
Subject: [PATCH 0160/1791] arch/arc: Implement <asm/fb.h> with generic helpers

Replace the architecture's fbdev helpers with the generic ones from
<asm-generic/fb.h>. On arc, pgprot_writecombine() and pgprot_noncached()
are the same; hence no functional changes.

v3:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Vineet Gupta <vgupta@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-3-tzimmermann@suse.de
---
 arch/arc/include/asm/fb.h | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/arch/arc/include/asm/fb.h b/arch/arc/include/asm/fb.h
index dc2e303cdbbb4..9c2383d29cbb9 100644
--- a/arch/arc/include/asm/fb.h
+++ b/arch/arc/include/asm/fb.h
@@ -1,20 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0 */
+
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 9cfefba8daabb8058122fbcc74ff37b9e515df07 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:35 +0200
Subject: [PATCH 0161/1791] arch/arm: Implement <asm/fb.h> with generic helpers

Replace the architecture's fbdev helpers with the generic
ones from <asm-generic/fb.h>. No functional changes.

v2:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Russell King <linux@armlinux.org.uk>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-4-tzimmermann@suse.de
---
 arch/arm/include/asm/fb.h | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)

diff --git a/arch/arm/include/asm/fb.h b/arch/arm/include/asm/fb.h
index d92e99cd8c8ab..ce20a43c30339 100644
--- a/arch/arm/include/asm/fb.h
+++ b/arch/arm/include/asm/fb.h
@@ -1,19 +1,6 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 10b707b4a2022ee901e6801c35c7eee0de62c168 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:36 +0200
Subject: [PATCH 0162/1791] arch/arm64: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fbdev helpers with the generic
ones from <asm-generic/fb.h>. No functional changes.

v2:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-5-tzimmermann@suse.de
---
 arch/arm64/include/asm/fb.h | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)

diff --git a/arch/arm64/include/asm/fb.h b/arch/arm64/include/asm/fb.h
index bdc735ee1f679..1a495d8fb2ce0 100644
--- a/arch/arm64/include/asm/fb.h
+++ b/arch/arm64/include/asm/fb.h
@@ -5,19 +5,6 @@
 #ifndef __ASM_FB_H_
 #define __ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* __ASM_FB_H_ */
-- 
GitLab


From 81ea5144f0cb88ae417f66448e66fd41cbe1e2c6 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:37 +0200
Subject: [PATCH 0163/1791] arch/ia64: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fb_is_primary_device() with the generic
one from <asm-generic/fb.h>. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-6-tzimmermann@suse.de
---
 arch/ia64/include/asm/fb.h | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/arch/ia64/include/asm/fb.h b/arch/ia64/include/asm/fb.h
index 5f95782bfa46d..0208f64a0da07 100644
--- a/arch/ia64/include/asm/fb.h
+++ b/arch/ia64/include/asm/fb.h
@@ -2,11 +2,12 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
 #include <linux/efi.h>
+
 #include <asm/page.h>
 
+struct file;
+
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
 {
@@ -15,10 +16,8 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 	else
 		vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 }
+#define fb_pgprotect fb_pgprotect
 
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 84998fc1c39270b8e20223d1b10c00464eec09dc Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:38 +0200
Subject: [PATCH 0164/1791] arch/loongarch: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fbdev helpers with the generic
ones from <asm-generic/fb.h>. No functional changes.

v2:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Huacai Chen <chenhuacai@kernel.org>
Cc: WANG Xuerui <kernel@xen0n.name>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-7-tzimmermann@suse.de
---
 arch/loongarch/include/asm/fb.h | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)

diff --git a/arch/loongarch/include/asm/fb.h b/arch/loongarch/include/asm/fb.h
index 3116bde8772dd..ff82f20685c8b 100644
--- a/arch/loongarch/include/asm/fb.h
+++ b/arch/loongarch/include/asm/fb.h
@@ -5,19 +5,6 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From d847517950f07353c1195de098c25024f948fe37 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:39 +0200
Subject: [PATCH 0165/1791] arch/m68k: Merge variants of fb_pgprotect() into
 single function

Merge all variants of fb_pgprotect() into a single function body.
There are two different cases for MMU systems. For non-MMU systems,
the function body will be empty. No functional changes, but this
will help with the switch to <asm-generic/fb.h>.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-8-tzimmermann@suse.de
---
 arch/m68k/include/asm/fb.h | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/arch/m68k/include/asm/fb.h b/arch/m68k/include/asm/fb.h
index b86c6e2e26ddc..4f96989922af0 100644
--- a/arch/m68k/include/asm/fb.h
+++ b/arch/m68k/include/asm/fb.h
@@ -7,17 +7,13 @@
 #include <asm/page.h>
 #include <asm/setup.h>
 
-#ifdef CONFIG_MMU
-#ifdef CONFIG_SUN3
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
 {
+#ifdef CONFIG_MMU
+#ifdef CONFIG_SUN3
 	pgprot_val(vma->vm_page_prot) |= SUN3_PAGE_NOCACHE;
-}
 #else
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
 	if (CPU_IS_020_OR_030)
 		pgprot_val(vma->vm_page_prot) |= _PAGE_NOCACHE030;
 	if (CPU_IS_040_OR_060) {
@@ -25,11 +21,9 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 		/* Use no-cache mode, serialized */
 		pgprot_val(vma->vm_page_prot) |= _PAGE_NOCACHE_S;
 	}
-}
 #endif /* CONFIG_SUN3 */
-#else
-#define fb_pgprotect(...) do {} while (0)
 #endif /* CONFIG_MMU */
+}
 
 static inline int fb_is_primary_device(struct fb_info *info)
 {
-- 
GitLab


From db21647767db268cebc4a244c61700aeb4a0b17d Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:40 +0200
Subject: [PATCH 0166/1791] arch/m68k: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fb_is_primary_device() with the generic
one from <asm-generic/fb.h>. No functional changes.

v2:
	* provide empty fb_pgprotect() on non-MMU systems

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-9-tzimmermann@suse.de
---
 arch/m68k/include/asm/fb.h | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/arch/m68k/include/asm/fb.h b/arch/m68k/include/asm/fb.h
index 4f96989922af0..24273fc7ad917 100644
--- a/arch/m68k/include/asm/fb.h
+++ b/arch/m68k/include/asm/fb.h
@@ -2,11 +2,11 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
 #include <asm/page.h>
 #include <asm/setup.h>
 
+struct file;
+
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
 {
@@ -24,10 +24,8 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 #endif /* CONFIG_SUN3 */
 #endif /* CONFIG_MMU */
 }
+#define fb_pgprotect fb_pgprotect
 
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 910994c0f622e5d2e02ec20e4a0dc342be5bca42 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:41 +0200
Subject: [PATCH 0167/1791] arch/mips: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fb_is_primary_device() with the generic
one from <asm-generic/fb.h>. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-10-tzimmermann@suse.de
---
 arch/mips/include/asm/fb.h | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/arch/mips/include/asm/fb.h b/arch/mips/include/asm/fb.h
index bd3f68c9ddfca..6bda0a81d8ca0 100644
--- a/arch/mips/include/asm/fb.h
+++ b/arch/mips/include/asm/fb.h
@@ -1,19 +1,17 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
 #include <asm/page.h>
 
+struct file;
+
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
 {
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 }
+#define fb_pgprotect fb_pgprotect
 
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 0d556f1f0e01ccab9c6bca8099e45170beb01297 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:42 +0200
Subject: [PATCH 0168/1791] video: Remove trailing whitespaces

Fix trailing whitespaces. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-11-tzimmermann@suse.de
---
 drivers/video/console/sticon.c  |   4 +-
 drivers/video/console/sticore.c | 102 ++++++++++-----------
 drivers/video/fbdev/sticore.h   |  14 +--
 drivers/video/fbdev/stifb.c     | 156 ++++++++++++++++----------------
 4 files changed, 138 insertions(+), 138 deletions(-)

diff --git a/drivers/video/console/sticon.c b/drivers/video/console/sticon.c
index 2cea69418a83c..89ad7ade6cf9f 100644
--- a/drivers/video/console/sticon.c
+++ b/drivers/video/console/sticon.c
@@ -282,7 +282,7 @@ static void sticon_init(struct vc_data *c, int init)
     vc_cols = sti_onscreen_x(sti) / sti->font->width;
     vc_rows = sti_onscreen_y(sti) / sti->font->height;
     c->vc_can_do_color = 1;
-    
+
     if (init) {
 	c->vc_cols = vc_cols;
 	c->vc_rows = vc_rows;
@@ -374,7 +374,7 @@ static const struct consw sti_con = {
 	.con_font_set		= sticon_font_set,
 	.con_font_default	= sticon_font_default,
 	.con_build_attr		= sticon_build_attr,
-	.con_invert_region	= sticon_invert_region, 
+	.con_invert_region	= sticon_invert_region,
 };
 
 
diff --git a/drivers/video/console/sticore.c b/drivers/video/console/sticore.c
index db568f67e4dc2..6ea9596a3c4b0 100644
--- a/drivers/video/console/sticore.c
+++ b/drivers/video/console/sticore.c
@@ -6,12 +6,12 @@
  *	Copyright (C) 2000 Philipp Rumpf <prumpf@tux.org>
  *	Copyright (C) 2001-2020 Helge Deller <deller@gmx.de>
  *	Copyright (C) 2001-2002 Thomas Bogendoerfer <tsbogend@alpha.franken.de>
- * 
+ *
  * TODO:
  * - call STI in virtual mode rather than in real mode
- * - screen blanking with state_mgmt() in text mode STI ? 
+ * - screen blanking with state_mgmt() in text mode STI ?
  * - try to make it work on m68k hp workstations ;)
- * 
+ *
  */
 
 #define pr_fmt(fmt) "%s: " fmt, KBUILD_MODNAME
@@ -66,12 +66,12 @@ static const u8 col_trans[8] = {
 #define c_index(sti, c) ((c) & 0xff)
 
 static const struct sti_init_flags default_init_flags = {
-	.wait	= STI_WAIT, 
+	.wait	= STI_WAIT,
 	.reset	= 1,
-	.text	= 1, 
+	.text	= 1,
 	.nontext = 1,
-	.no_chg_bet = 1, 
-	.no_chg_bei = 1, 
+	.no_chg_bet = 1,
+	.no_chg_bei = 1,
 	.init_cmap_tx = 1,
 };
 
@@ -104,7 +104,7 @@ static int sti_init_graph(struct sti_struct *sti)
 		pr_err("STI init_graph failed (ret %d, errno %d)\n", ret, err);
 		return -1;
 	}
-	
+
 	return 0;
 }
 
@@ -120,7 +120,7 @@ static void sti_inq_conf(struct sti_struct *sti)
 	s32 ret;
 
 	outptr->ext_ptr = STI_PTR(&sti->sti_data->inq_outptr_ext);
-	
+
 	do {
 		spin_lock_irqsave(&sti->lock, flags);
 		memset(inptr, 0, sizeof(*inptr));
@@ -162,9 +162,9 @@ sti_putc(struct sti_struct *sti, int c, int y, int x,
 }
 
 static const struct sti_blkmv_flags clear_blkmv_flags = {
-	.wait	= STI_WAIT, 
-	.color	= 1, 
-	.clear	= 1, 
+	.wait	= STI_WAIT,
+	.color	= 1,
+	.clear	= 1,
 };
 
 void
@@ -185,7 +185,7 @@ sti_set(struct sti_struct *sti, int src_y, int src_x,
 	struct sti_blkmv_outptr *outptr = &sti->sti_data->blkmv_outptr;
 	s32 ret;
 	unsigned long flags;
-	
+
 	do {
 		spin_lock_irqsave(&sti->lock, flags);
 		*inptr = inptr_default;
@@ -224,7 +224,7 @@ sti_clear(struct sti_struct *sti, int src_y, int src_x,
 }
 
 static const struct sti_blkmv_flags default_blkmv_flags = {
-	.wait = STI_WAIT, 
+	.wait = STI_WAIT,
 };
 
 void
@@ -291,14 +291,14 @@ static int __init sti_setup(char *str)
 {
 	if (str)
 		strscpy(default_sti_path, str, sizeof(default_sti_path));
-	
+
 	return 1;
 }
 
 /*	Assuming the machine has multiple STI consoles (=graphic cards) which
  *	all get detected by sticon, the user may define with the linux kernel
  *	parameter sti=<x> which of them will be the initial boot-console.
- *	<x> is a number between 0 and MAX_STI_ROMS, with 0 as the default 
+ *	<x> is a number between 0 and MAX_STI_ROMS, with 0 as the default
  *	STI screen.
  */
 __setup("sti=", sti_setup);
@@ -341,13 +341,13 @@ static int sti_font_setup(char *str)
  *	should be used by the sticon driver to draw characters to the screen.
  *	Possible values are:
  *	- sti_font=<fb_fontname>:
- *		<fb_fontname> is the name of one of the linux-kernel built-in 
- *		framebuffer font names (e.g. VGA8x16, SUN22x18). 
- *		This is only available if the fonts have been statically compiled 
+ *		<fb_fontname> is the name of one of the linux-kernel built-in
+ *		framebuffer font names (e.g. VGA8x16, SUN22x18).
+ *		This is only available if the fonts have been statically compiled
  *		in with e.g. the CONFIG_FONT_8x16 or CONFIG_FONT_SUN12x22 options.
  *	- sti_font=<number>	(<number> = 1,2,3,...)
  *		most STI ROMs have built-in HP specific fonts, which can be selected
- *		by giving the desired number to the sticon driver. 
+ *		by giving the desired number to the sticon driver.
  *		NOTE: This number is machine and STI ROM dependend.
  *	- sti_font=<height>x<width>  (e.g. sti_font=16x8)
  *		<height> and <width> gives hints to the height and width of the
@@ -359,12 +359,12 @@ __setup("sti_font=", sti_font_setup);
 #endif
 
 
-	
+
 static void sti_dump_globcfg(struct sti_glob_cfg *glob_cfg,
 			     unsigned int sti_mem_request)
 {
 	struct sti_glob_cfg_ext *cfg;
-	
+
 	pr_debug("%d text planes\n"
 		"%4d x %4d screen resolution\n"
 		"%4d x %4d offscreen\n"
@@ -384,7 +384,7 @@ static void sti_dump_globcfg(struct sti_glob_cfg *glob_cfg,
 		glob_cfg->reent_lvl,
 		glob_cfg->save_addr);
 
-	/* dump extended cfg */ 
+	/* dump extended cfg */
 	cfg = PTR_STI((unsigned long)glob_cfg->ext_ptr);
 	pr_debug("monitor %d\n"
 		"in friendly mode: %d\n"
@@ -437,10 +437,10 @@ static int sti_init_glob_cfg(struct sti_struct *sti, unsigned long rom_address,
 	glob_cfg->save_addr = STI_PTR(save_addr);
 	for (i=0; i<8; i++) {
 		unsigned long newhpa, len;
-	       
+
 		if (sti->pd) {
 			unsigned char offs = sti->rm_entry[i];
-				
+
 			if (offs == 0)
 				continue;
 			if (offs != PCI_ROM_ADDRESS &&
@@ -456,18 +456,18 @@ static int sti_init_glob_cfg(struct sti_struct *sti, unsigned long rom_address,
 
 		sti->regions_phys[i] =
 			REGION_OFFSET_TO_PHYS(sti->regions[i], newhpa);
-		
+
 		len = sti->regions[i].region_desc.length * 4096;
 		if (len)
 			glob_cfg->region_ptrs[i] = sti->regions_phys[i];
-		
+
 		pr_debug("region #%d: phys %08lx, region_ptr %08x, len=%lukB, "
 			 "btlb=%d, sysonly=%d, cache=%d, last=%d\n",
 			i, sti->regions_phys[i], glob_cfg->region_ptrs[i],
 			len/1024,
 			sti->regions[i].region_desc.btlb,
 			sti->regions[i].region_desc.sys_only,
-			sti->regions[i].region_desc.cache, 
+			sti->regions[i].region_desc.cache,
 			sti->regions[i].region_desc.last);
 
 		/* last entry reached ? */
@@ -482,7 +482,7 @@ static int sti_init_glob_cfg(struct sti_struct *sti, unsigned long rom_address,
 	glob_cfg_ext->sti_mem_addr = STI_PTR(sti_mem_addr);
 
 	sti->glob_cfg = glob_cfg;
-	
+
 	return 0;
 }
 
@@ -495,7 +495,7 @@ sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 	void *dest;
 	struct sti_rom_font *nf;
 	struct sti_cooked_font *cooked_font;
-	
+
 	if (fbfont_name && strlen(fbfont_name))
 		fbfont = find_font(fbfont_name);
 	if (!fbfont)
@@ -505,8 +505,8 @@ sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 
 	pr_info("    using %ux%u framebuffer font %s\n",
 			fbfont->width, fbfont->height, fbfont->name);
-			
-	bpc = ((fbfont->width+7)/8) * fbfont->height; 
+
+	bpc = ((fbfont->width+7)/8) * fbfont->height;
 	size = bpc * fbfont->charcount;
 	size += sizeof(struct sti_rom_font);
 
@@ -533,7 +533,7 @@ sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 		kfree(nf);
 		return NULL;
 	}
-	
+
 	cooked_font->raw = nf;
 	cooked_font->raw_ptr = nf;
 	cooked_font->next_font = NULL;
@@ -617,9 +617,9 @@ static void sti_dump_rom(struct sti_struct *sti)
 	int nr;
 
 	pr_info("  id %04x-%04x, conforms to spec rev. %d.%02x\n",
-		rom->graphics_id[0], 
+		rom->graphics_id[0],
 		rom->graphics_id[1],
-		rom->revno[0] >> 4, 
+		rom->revno[0] >> 4,
 		rom->revno[0] & 0x0f);
 	pr_debug("  supports %d monitors\n", rom->num_mons);
 	pr_debug("  font start %08x\n", rom->font_start);
@@ -647,7 +647,7 @@ static int sti_cook_fonts(struct sti_cooked_rom *cooked_rom,
 {
 	struct sti_rom_font *raw_font, *font_start;
 	struct sti_cooked_font *cooked_font;
-	
+
 	cooked_font = kzalloc(sizeof(*cooked_font), GFP_KERNEL);
 	if (!cooked_font)
 		return 0;
@@ -745,7 +745,7 @@ static struct sti_rom *sti_get_bmode_rom (unsigned long address)
 
 		raw_font = ((void *)raw) + raw->font_start;
 		font_start = raw_font;
-		
+
 		while (raw_font->next_font) {
 			BMODE_RELOCATE (raw_font->next_font);
 			raw_font = ((void *)font_start) + raw_font->next_font;
@@ -759,7 +759,7 @@ static struct sti_rom *sti_get_wmode_rom(unsigned long address)
 	struct sti_rom *raw;
 	unsigned long size;
 
-	/* read the ROM size directly from the struct in ROM */ 
+	/* read the ROM size directly from the struct in ROM */
 	size = gsc_readl(address + offsetof(struct sti_rom,last_addr));
 
 	raw = kmalloc(size, STI_LOWMEM);
@@ -869,7 +869,7 @@ static struct sti_struct *sti_try_rom_generic(unsigned long address,
 		pr_warn("maximum number of STI ROMS reached !\n");
 		return NULL;
 	}
-	
+
 	sti = kzalloc(sizeof(*sti), GFP_KERNEL);
 	if (!sti)
 		return NULL;
@@ -890,19 +890,19 @@ static struct sti_struct *sti_try_rom_generic(unsigned long address,
 		u32 *rm;
 		i = gsc_readl(address+0x04);
 		if (i != 1) {
-			/* The ROM could have multiple architecture 
+			/* The ROM could have multiple architecture
 			 * dependent images (e.g. i386, parisc,...) */
 			pr_warn("PCI ROM is not a STI ROM type image (0x%8x)\n", i);
 			goto out_err;
 		}
-		
+
 		sti->pd = pd;
 
 		i = gsc_readl(address+0x0c);
 		pr_debug("PCI ROM size (from header) = %d kB\n",
 			le16_to_cpu(i>>16)*512/1024);
 		rm_offset = le16_to_cpu(i & 0xffff);
-		if (rm_offset) { 
+		if (rm_offset) {
 			/* read 16 bytes from the pci region mapper array */
 			rm = (u32*) &sti->rm_entry;
 			*rm++ = gsc_readl(address+rm_offset+0x00);
@@ -915,9 +915,9 @@ static struct sti_struct *sti_try_rom_generic(unsigned long address,
 		pr_debug("sig %04x, PCI STI ROM at %08lx\n", sig, address);
 		goto test_rom;
 	}
-	
+
 	ok = 0;
-	
+
 	if ((sig & 0xff) == 0x01) {
 		pr_debug("    byte mode ROM at %08lx, hpa at %08lx\n",
 		       address, hpa);
@@ -941,7 +941,7 @@ static struct sti_struct *sti_try_rom_generic(unsigned long address,
 	 */
 	if (sti->pd) {
 		unsigned long rom_base;
-		rom_base = pci_resource_start(sti->pd, PCI_ROM_RESOURCE);	
+		rom_base = pci_resource_start(sti->pd, PCI_ROM_RESOURCE);
 		pci_write_config_dword(sti->pd, PCI_ROM_ADDRESS, rom_base & ~PCI_ROM_ADDRESS_ENABLE);
 		pr_debug("STI PCI ROM disabled\n");
 	}
@@ -952,13 +952,13 @@ static struct sti_struct *sti_try_rom_generic(unsigned long address,
 	sti_inq_conf(sti);
 	sti_dump_globcfg(sti->glob_cfg, sti->sti_mem_request);
 	sti_dump_outptr(sti);
-	
+
 	pr_info("    graphics card name: %s\n",
 		sti->sti_data->inq_outptr.dev_name);
 
 	sti_roms[num_sti_roms] = sti;
 	num_sti_roms++;
-	
+
 	return sti;
 
 out_err:
@@ -974,9 +974,9 @@ static void sticore_check_for_default_sti(struct sti_struct *sti, char *path)
 }
 
 /*
- * on newer systems PDC gives the address of the ROM 
+ * on newer systems PDC gives the address of the ROM
  * in the additional address field addr[1] while on
- * older Systems the PDC stores it in page0->proc_sti 
+ * older Systems the PDC stores it in page0->proc_sti
  */
 static int __init sticore_pa_init(struct parisc_device *dev)
 {
@@ -1005,7 +1005,7 @@ static int sticore_pci_init(struct pci_dev *pd, const struct pci_device_id *ent)
 	unsigned int fb_len, rom_len;
 	int err;
 	struct sti_struct *sti;
-	
+
 	err = pci_enable_device(pd);
 	if (err < 0) {
 		dev_err(&pd->dev, "Cannot enable PCI device\n");
@@ -1032,7 +1032,7 @@ static int sticore_pci_init(struct pci_dev *pd, const struct pci_device_id *ent)
 		print_pci_hwpath(pd, sti->pa_path);
 		sticore_check_for_default_sti(sti, sti->pa_path);
 	}
-	
+
 	if (!sti) {
 		pr_warn("Unable to handle STI device '%s'\n", pci_name(pd));
 		return -ENODEV;
diff --git a/drivers/video/fbdev/sticore.h b/drivers/video/fbdev/sticore.h
index 0ebdd28a0b813..c0879352cde44 100644
--- a/drivers/video/fbdev/sticore.h
+++ b/drivers/video/fbdev/sticore.h
@@ -27,11 +27,11 @@
  *
  * Probably the best solution to all this is have the generic code manage
  * the screen buffer and a kernel thread to call STI occasionally.
- * 
+ *
  * Luckily, the frame buffer guys have the same problem so we can just wait
  * for them to fix it and steal their solution.   prumpf
  */
- 
+
 #include <asm/io.h>
 
 #define STI_WAIT 1
@@ -56,7 +56,7 @@
 /* STI function configuration structs */
 
 typedef union region {
-	struct { 
+	struct {
 		u32 offset	: 14;	/* offset in 4kbyte page */
 		u32 sys_only	: 1;	/* don't map to user space */
 		u32 cache	: 1;	/* map to data cache */
@@ -154,7 +154,7 @@ struct sti_conf_inptr {
 };
 
 struct sti_conf_outptr_ext {
-	u32 crt_config[3];	/* hardware specific X11/OGL information */	
+	u32 crt_config[3];	/* hardware specific X11/OGL information */
 	u32 crt_hdw[3];
 	u32 future_ptr;
 };
@@ -211,7 +211,7 @@ struct sti_rom {
 	u32 set_cm_entry;
 	u32 dma_ctrl;
 	 u8 res040[7 * 4];
-	
+
 	u32 init_graph_addr;
 	u32 state_mgmt_addr;
 	u32 font_unp_addr;
@@ -271,7 +271,7 @@ struct sti_font_flags {
 	u32 pad : 30;		/* pad to word boundary */
 	u32 future_ptr; 	/* pointer to future data */
 };
-	
+
 struct sti_font_outptr {
 	s32 errno;		/* error number on failure */
 	u32 future_ptr; 	/* pointer to future data */
@@ -338,7 +338,7 @@ struct sti_all_data {
 
 struct sti_struct {
 	spinlock_t lock;
-		
+
 	/* char **mon_strings; */
 	int sti_mem_request;
 	u32 graphics_id[2];
diff --git a/drivers/video/fbdev/stifb.c b/drivers/video/fbdev/stifb.c
index ef8a4c5fc6875..99996bc7e6d97 100644
--- a/drivers/video/fbdev/stifb.c
+++ b/drivers/video/fbdev/stifb.c
@@ -1,11 +1,11 @@
 /*
- * linux/drivers/video/stifb.c - 
- * Low level Frame buffer driver for HP workstations with 
+ * linux/drivers/video/stifb.c -
+ * Low level Frame buffer driver for HP workstations with
  * STI (standard text interface) video firmware.
  *
  * Copyright (C) 2001-2006 Helge Deller <deller@gmx.de>
  * Portions Copyright (C) 2001 Thomas Bogendoerfer <tsbogend@alpha.franken.de>
- * 
+ *
  * Based on:
  * - linux/drivers/video/artistfb.c -- Artist frame buffer driver
  *	Copyright (C) 2000 Philipp Rumpf <prumpf@tux.org>
@@ -14,7 +14,7 @@
  * - HP Xhp cfb-based X11 window driver for XFree86
  *	(c)Copyright 1992 Hewlett-Packard Co.
  *
- * 
+ *
  *  The following graphics display devices (NGLE family) are supported by this driver:
  *
  *  HPA4070A	known as "HCRX", a 1280x1024 color device with 8 planes
@@ -30,7 +30,7 @@
  *		supports 1280x1024 color displays with 8 planes.
  *  HP710G	same as HP710C, 1280x1024 grayscale only
  *  HP710L	same as HP710C, 1024x768 color only
- *  HP712	internal graphics support on HP9000s712 SPU, supports 640x480, 
+ *  HP712	internal graphics support on HP9000s712 SPU, supports 640x480,
  *		1024x768 or 1280x1024 color displays on 8 planes (Artist)
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -92,7 +92,7 @@ typedef struct {
 	__s32	misc_video_end;
 } video_setup_t;
 
-typedef struct {                  
+typedef struct {
 	__s16	sizeof_ngle_data;
 	__s16	x_size_visible;	    /* visible screen dim in pixels  */
 	__s16	y_size_visible;
@@ -177,10 +177,10 @@ static int __initdata stifb_bpp_pref[MAX_STI_ROMS];
 #endif /* DEBUG_STIFB_REGS */
 
 
-#define ENABLE	1	/* for enabling/disabling screen */	
+#define ENABLE	1	/* for enabling/disabling screen */
 #define DISABLE 0
 
-#define NGLE_LOCK(fb_info)	do { } while (0) 
+#define NGLE_LOCK(fb_info)	do { } while (0)
 #define NGLE_UNLOCK(fb_info)	do { } while (0)
 
 static void
@@ -198,9 +198,9 @@ SETUP_HW(struct stifb_info *fb)
 
 static void
 SETUP_FB(struct stifb_info *fb)
-{	
+{
 	unsigned int reg10_value = 0;
-	
+
 	SETUP_HW(fb);
 	switch (fb->id)
 	{
@@ -210,15 +210,15 @@ SETUP_FB(struct stifb_info *fb)
 			reg10_value = 0x13601000;
 			break;
 		case S9000_ID_A1439A:
-			if (fb->info.var.bits_per_pixel == 32)						
+			if (fb->info.var.bits_per_pixel == 32)
 				reg10_value = 0xBBA0A000;
-			else 
+			else
 				reg10_value = 0x13601000;
 			break;
 		case S9000_ID_HCRX:
 			if (fb->info.var.bits_per_pixel == 32)
 				reg10_value = 0xBBA0A000;
-			else					
+			else
 				reg10_value = 0x13602000;
 			break;
 		case S9000_ID_TIMBER:
@@ -243,7 +243,7 @@ START_IMAGE_COLORMAP_ACCESS(struct stifb_info *fb)
 }
 
 static void
-WRITE_IMAGE_COLOR(struct stifb_info *fb, int index, int color) 
+WRITE_IMAGE_COLOR(struct stifb_info *fb, int index, int color)
 {
 	SETUP_HW(fb);
 	WRITE_WORD(((0x100+index)<<2), fb, REG_3);
@@ -251,30 +251,30 @@ WRITE_IMAGE_COLOR(struct stifb_info *fb, int index, int color)
 }
 
 static void
-FINISH_IMAGE_COLORMAP_ACCESS(struct stifb_info *fb) 
-{		
+FINISH_IMAGE_COLORMAP_ACCESS(struct stifb_info *fb)
+{
 	WRITE_WORD(0x400, fb, REG_2);
 	if (fb->info.var.bits_per_pixel == 32) {
 		WRITE_WORD(0x83000100, fb, REG_1);
 	} else {
 		if (fb->id == S9000_ID_ARTIST || fb->id == CRT_ID_VISUALIZE_EG)
 			WRITE_WORD(0x80000100, fb, REG_26);
-		else							
+		else
 			WRITE_WORD(0x80000100, fb, REG_1);
 	}
 	SETUP_FB(fb);
 }
 
 static void
-SETUP_RAMDAC(struct stifb_info *fb) 
+SETUP_RAMDAC(struct stifb_info *fb)
 {
 	SETUP_HW(fb);
 	WRITE_WORD(0x04000000, fb, 0x1020);
 	WRITE_WORD(0xff000000, fb, 0x1028);
 }
 
-static void 
-CRX24_SETUP_RAMDAC(struct stifb_info *fb) 
+static void
+CRX24_SETUP_RAMDAC(struct stifb_info *fb)
 {
 	SETUP_HW(fb);
 	WRITE_WORD(0x04000000, fb, 0x1000);
@@ -286,14 +286,14 @@ CRX24_SETUP_RAMDAC(struct stifb_info *fb)
 }
 
 #if 0
-static void 
+static void
 HCRX_SETUP_RAMDAC(struct stifb_info *fb)
 {
 	WRITE_WORD(0xffffffff, fb, REG_32);
 }
 #endif
 
-static void 
+static void
 CRX24_SET_OVLY_MASK(struct stifb_info *fb)
 {
 	SETUP_HW(fb);
@@ -314,7 +314,7 @@ ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable)
         WRITE_WORD(value, 	fb, 0x1038);
 }
 
-static void 
+static void
 CRX24_ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable)
 {
 	unsigned int value = enable ? 0x10000000 : 0x30000000;
@@ -325,11 +325,11 @@ CRX24_ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable)
 }
 
 static void
-ARTIST_ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable) 
+ARTIST_ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable)
 {
 	u32 DregsMiscVideo = REG_21;
 	u32 DregsMiscCtl = REG_27;
-	
+
 	SETUP_HW(fb);
 	if (enable) {
 	  WRITE_WORD(READ_WORD(fb, DregsMiscVideo) | 0x0A000000, fb, DregsMiscVideo);
@@ -344,7 +344,7 @@ ARTIST_ENABLE_DISABLE_DISPLAY(struct stifb_info *fb, int enable)
 	(READ_BYTE(fb, REG_16b3) - 1)
 
 #define HYPER_CONFIG_PLANES_24 0x00000100
-	
+
 #define IS_24_DEVICE(fb) \
 	(fb->deviceSpecificConfig & HYPER_CONFIG_PLANES_24)
 
@@ -470,15 +470,15 @@ SETUP_ATTR_ACCESS(struct stifb_info *fb, unsigned BufferNumber)
 }
 
 static void
-SET_ATTR_SIZE(struct stifb_info *fb, int width, int height) 
+SET_ATTR_SIZE(struct stifb_info *fb, int width, int height)
 {
-	/* REG_6 seems to have special values when run on a 
+	/* REG_6 seems to have special values when run on a
 	   RDI precisionbook parisc laptop (INTERNAL_EG_DX1024 or
 	   INTERNAL_EG_X1024).  The values are:
 		0x2f0: internal (LCD) & external display enabled
 		0x2a0: external display only
 		0x000: zero on standard artist graphic cards
-	*/ 
+	*/
 	WRITE_WORD(0x00000000, fb, REG_6);
 	WRITE_WORD((width<<16) | height, fb, REG_9);
 	WRITE_WORD(0x05000000, fb, REG_6);
@@ -486,7 +486,7 @@ SET_ATTR_SIZE(struct stifb_info *fb, int width, int height)
 }
 
 static void
-FINISH_ATTR_ACCESS(struct stifb_info *fb) 
+FINISH_ATTR_ACCESS(struct stifb_info *fb)
 {
 	SETUP_HW(fb);
 	WRITE_WORD(0x00000000, fb, REG_12);
@@ -499,7 +499,7 @@ elkSetupPlanes(struct stifb_info *fb)
 	SETUP_FB(fb);
 }
 
-static void 
+static void
 ngleSetupAttrPlanes(struct stifb_info *fb, int BufferNumber)
 {
 	SETUP_ATTR_ACCESS(fb, BufferNumber);
@@ -519,7 +519,7 @@ rattlerSetupPlanes(struct stifb_info *fb)
 	 * read mask register for overlay planes, not image planes).
 	 */
 	CRX24_SETUP_RAMDAC(fb);
-    
+
 	/* change fb->id temporarily to fool SETUP_FB() */
 	saved_id = fb->id;
 	fb->id = CRX24_OVERLAY_PLANES;
@@ -565,7 +565,7 @@ setNgleLutBltCtl(struct stifb_info *fb, int offsetWithinLut, int length)
 	lutBltCtl.all           = 0x80000000;
 	lutBltCtl.fields.length = length;
 
-	switch (fb->id) 
+	switch (fb->id)
 	{
 	case S9000_ID_A1439A:		/* CRX24 */
 		if (fb->var.bits_per_pixel == 8) {
@@ -576,12 +576,12 @@ setNgleLutBltCtl(struct stifb_info *fb, int offsetWithinLut, int length)
 			lutBltCtl.fields.lutOffset = 0 * 256;
 		}
 		break;
-		
+
 	case S9000_ID_ARTIST:
 		lutBltCtl.fields.lutType = NGLE_CMAP_INDEXED0_TYPE;
 		lutBltCtl.fields.lutOffset = 0 * 256;
 		break;
-		
+
 	default:
 		lutBltCtl.fields.lutType = NGLE_CMAP_INDEXED0_TYPE;
 		lutBltCtl.fields.lutOffset = 0;
@@ -596,7 +596,7 @@ setNgleLutBltCtl(struct stifb_info *fb, int offsetWithinLut, int length)
 #endif
 
 static NgleLutBltCtl
-setHyperLutBltCtl(struct stifb_info *fb, int offsetWithinLut, int length) 
+setHyperLutBltCtl(struct stifb_info *fb, int offsetWithinLut, int length)
 {
 	NgleLutBltCtl lutBltCtl;
 
@@ -633,7 +633,7 @@ static void hyperUndoITE(struct stifb_info *fb)
 
 	/* Hardware setup for full-depth write to "magic" location */
 	GET_FIFO_SLOTS(fb, nFreeFifoSlots, 7);
-	NGLE_QUICK_SET_DST_BM_ACCESS(fb, 
+	NGLE_QUICK_SET_DST_BM_ACCESS(fb,
 		BA(IndexedDcd, Otc04, Ots08, AddrLong,
 		BAJustPoint(0), BINovly, BAIndexBase(0)));
 	NGLE_QUICK_SET_IMAGE_BITMAP_OP(fb,
@@ -653,13 +653,13 @@ static void hyperUndoITE(struct stifb_info *fb)
 	NGLE_UNLOCK(fb);
 }
 
-static void 
+static void
 ngleDepth8_ClearImagePlanes(struct stifb_info *fb)
 {
 	/* FIXME! */
 }
 
-static void 
+static void
 ngleDepth24_ClearImagePlanes(struct stifb_info *fb)
 {
 	/* FIXME! */
@@ -675,7 +675,7 @@ ngleResetAttrPlanes(struct stifb_info *fb, unsigned int ctlPlaneReg)
 	NGLE_LOCK(fb);
 
 	GET_FIFO_SLOTS(fb, nFreeFifoSlots, 4);
-	NGLE_QUICK_SET_DST_BM_ACCESS(fb, 
+	NGLE_QUICK_SET_DST_BM_ACCESS(fb,
 				     BA(IndexedDcd, Otc32, OtsIndirect,
 					AddrLong, BAJustPoint(0),
 					BINattr, BAIndexBase(0)));
@@ -713,22 +713,22 @@ ngleResetAttrPlanes(struct stifb_info *fb, unsigned int ctlPlaneReg)
 	/**** Finally, set the Control Plane Register back to zero: ****/
 	GET_FIFO_SLOTS(fb, nFreeFifoSlots, 1);
 	NGLE_QUICK_SET_CTL_PLN_REG(fb, 0);
-	
+
 	NGLE_UNLOCK(fb);
 }
-    
+
 static void
 ngleClearOverlayPlanes(struct stifb_info *fb, int mask, int data)
 {
 	int nFreeFifoSlots = 0;
 	u32 packed_dst;
 	u32 packed_len;
-    
+
 	NGLE_LOCK(fb);
 
 	/* Hardware setup */
 	GET_FIFO_SLOTS(fb, nFreeFifoSlots, 8);
-	NGLE_QUICK_SET_DST_BM_ACCESS(fb, 
+	NGLE_QUICK_SET_DST_BM_ACCESS(fb,
 				     BA(IndexedDcd, Otc04, Ots08, AddrLong,
 					BAJustPoint(0), BINovly, BAIndexBase(0)));
 
@@ -736,23 +736,23 @@ ngleClearOverlayPlanes(struct stifb_info *fb, int mask, int data)
 
         NGLE_REALLY_SET_IMAGE_FG_COLOR(fb, data);
         NGLE_REALLY_SET_IMAGE_PLANEMASK(fb, mask);
-    
+
         packed_dst = 0;
         packed_len = (fb->info.var.xres << 16) | fb->info.var.yres;
         NGLE_SET_DSTXY(fb, packed_dst);
-    
-        /* Write zeroes to overlay planes */		       
+
+	/* Write zeroes to overlay planes */
 	NGLE_QUICK_SET_IMAGE_BITMAP_OP(fb,
 				       IBOvals(RopSrc, MaskAddrOffset(0),
 					       BitmapExtent08, StaticReg(0),
 					       DataDynamic, MaskOtc, BGx(0), FGx(0)));
-		       
+
         SET_LENXY_START_RECFILL(fb, packed_len);
 
 	NGLE_UNLOCK(fb);
 }
 
-static void 
+static void
 hyperResetPlanes(struct stifb_info *fb, int enable)
 {
 	unsigned int controlPlaneReg;
@@ -783,7 +783,7 @@ hyperResetPlanes(struct stifb_info *fb, int enable)
 	        ngleClearOverlayPlanes(fb, 0xff, 255);
 
 		/**************************************************
-		 ** Also need to counteract ITE settings 
+		 ** Also need to counteract ITE settings
 		 **************************************************/
 		hyperUndoITE(fb);
 		break;
@@ -803,13 +803,13 @@ hyperResetPlanes(struct stifb_info *fb, int enable)
 		ngleResetAttrPlanes(fb, controlPlaneReg);
 		break;
     	}
-	
+
 	NGLE_UNLOCK(fb);
 }
 
 /* Return pointer to in-memory structure holding ELK device-dependent ROM values. */
 
-static void 
+static void
 ngleGetDeviceRomData(struct stifb_info *fb)
 {
 #if 0
@@ -821,7 +821,7 @@ XXX: FIXME: !!!
 	char	*pCard8;
 	int	i;
 	char	*mapOrigin = NULL;
-    
+
 	int romTableIdx;
 
 	pPackedDevRomData = fb->ngle_rom;
@@ -888,7 +888,7 @@ SETUP_HCRX(struct stifb_info *fb)
 
 	/* Initialize Hyperbowl registers */
 	GET_FIFO_SLOTS(fb, nFreeFifoSlots, 7);
-	
+
 	if (IS_24_DEVICE(fb)) {
 		hyperbowl = (fb->info.var.bits_per_pixel == 32) ?
 			HYPERBOWL_MODE01_8_24_LUT0_TRANSPARENT_LUT1_OPAQUE :
@@ -897,9 +897,9 @@ SETUP_HCRX(struct stifb_info *fb)
 		/* First write to Hyperbowl must happen twice (bug) */
 		WRITE_WORD(hyperbowl, fb, REG_40);
 		WRITE_WORD(hyperbowl, fb, REG_40);
-		
+
 		WRITE_WORD(HYPERBOWL_MODE2_8_24, fb, REG_39);
-		
+
 		WRITE_WORD(0x014c0148, fb, REG_42); /* Set lut 0 to be the direct color */
 		WRITE_WORD(0x404c4048, fb, REG_43);
 		WRITE_WORD(0x034c0348, fb, REG_44);
@@ -990,7 +990,7 @@ stifb_setcolreg(u_int regno, u_int red, u_int green,
 				0,	/* Offset w/i LUT */
 				256);	/* Load entire LUT */
 		NGLE_BINC_SET_SRCADDR(fb,
-				NGLE_LONG_FB_ADDRESS(0, 0x100, 0)); 
+				NGLE_LONG_FB_ADDRESS(0, 0x100, 0));
 				/* 0x100 is same as used in WRITE_IMAGE_COLOR() */
 		START_COLORMAPLOAD(fb, lutBltCtl.all);
 		SETUP_FB(fb);
@@ -1028,7 +1028,7 @@ stifb_blank(int blank_mode, struct fb_info *info)
 		ENABLE_DISABLE_DISPLAY(fb, enable);
 		break;
 	}
-	
+
 	SETUP_FB(fb);
 	return 0;
 }
@@ -1114,15 +1114,15 @@ stifb_init_display(struct stifb_info *fb)
 
 	/* HCRX specific initialization */
 	SETUP_HCRX(fb);
-	
+
 	/*
 	if (id == S9000_ID_HCRX)
 		hyperInitSprite(fb);
 	else
 		ngleInitSprite(fb);
 	*/
-	
-	/* Initialize the image planes. */ 
+
+	/* Initialize the image planes. */
         switch (id) {
 	 case S9000_ID_HCRX:
 	    hyperResetPlanes(fb, ENABLE);
@@ -1194,7 +1194,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 	fb = kzalloc(sizeof(*fb), GFP_ATOMIC);
 	if (!fb)
 		return -ENOMEM;
-	
+
 	info = &fb->info;
 
 	/* set struct to a known state */
@@ -1235,7 +1235,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 			dev_name, fb->id);
 		goto out_err0;
 	}
-	
+
 	/* default to 8 bpp on most graphic chips */
 	bpp = 8;
 	xres = sti_onscreen_x(fb->sti);
@@ -1256,7 +1256,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 		fb->id = S9000_ID_A1659A;
 		break;
 	case S9000_ID_TIMBER:	/* HP9000/710 Any (may be a grayscale device) */
-		if (strstr(dev_name, "GRAYSCALE") || 
+		if (strstr(dev_name, "GRAYSCALE") ||
 		    strstr(dev_name, "Grayscale") ||
 		    strstr(dev_name, "grayscale"))
 			var->grayscale = 1;
@@ -1295,16 +1295,16 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 	case CRT_ID_VISUALIZE_EG:
 	case S9000_ID_ARTIST:	/* Artist */
 		break;
-	default: 
+	default:
 #ifdef FALLBACK_TO_1BPP
-	       	printk(KERN_WARNING 
+		printk(KERN_WARNING
 			"stifb: Unsupported graphics card (id=0x%08x) "
 				"- now trying 1bpp mode instead\n",
 			fb->id);
 		bpp = 1;	/* default to 1 bpp */
 		break;
 #else
-	       	printk(KERN_WARNING 
+		printk(KERN_WARNING
 			"stifb: Unsupported graphics card (id=0x%08x) "
 				"- skipping.\n",
 			fb->id);
@@ -1320,11 +1320,11 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 	fix->line_length = (fb->sti->glob_cfg->total_x * bpp) / 8;
 	if (!fix->line_length)
 		fix->line_length = 2048; /* default */
-	
+
 	/* limit fbsize to max visible screen size */
 	if (fix->smem_len > yres*fix->line_length)
 		fix->smem_len = ALIGN(yres*fix->line_length, 4*1024*1024);
-	
+
 	fix->accel = FB_ACCEL_NONE;
 
 	switch (bpp) {
@@ -1350,7 +1350,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 	    default:
 		break;
 	}
-	
+
 	var->xres = var->xres_virtual = xres;
 	var->yres = var->yres_virtual = yres;
 	var->bits_per_pixel = bpp;
@@ -1379,7 +1379,7 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 				fix->smem_start, fix->smem_start+fix->smem_len);
 		goto out_err2;
 	}
-		
+
 	if (!request_mem_region(fix->mmio_start, fix->mmio_len, "stifb mmio")) {
 		printk(KERN_ERR "stifb: cannot reserve sti mmio region 0x%04lx-0x%04lx\n",
 				fix->mmio_start, fix->mmio_start+fix->mmio_len);
@@ -1393,11 +1393,11 @@ static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 
 	fb_info(&fb->info, "%s %dx%d-%d frame buffer device, %s, id: %04x, mmio: 0x%04lx\n",
 		fix->id,
-		var->xres, 
+		var->xres,
 		var->yres,
 		var->bits_per_pixel,
 		dev_name,
-		fb->id, 
+		fb->id,
 		fix->mmio_start);
 
 	return 0;
@@ -1426,7 +1426,7 @@ static int __init stifb_init(void)
 	struct sti_struct *sti;
 	struct sti_struct *def_sti;
 	int i;
-	
+
 #ifndef MODULE
 	char *option = NULL;
 
@@ -1438,7 +1438,7 @@ static int __init stifb_init(void)
 		printk(KERN_INFO "stifb: disabled by \"stifb=off\" kernel parameter\n");
 		return -ENXIO;
 	}
-	
+
 	def_sti = sti_get_rom(0);
 	if (def_sti) {
 		for (i = 1; i <= MAX_STI_ROMS; i++) {
@@ -1472,7 +1472,7 @@ stifb_cleanup(void)
 {
 	struct sti_struct *sti;
 	int i;
-	
+
 	for (i = 1; i <= MAX_STI_ROMS; i++) {
 		sti = sti_get_rom(i);
 		if (!sti)
@@ -1495,10 +1495,10 @@ int __init
 stifb_setup(char *options)
 {
 	int i;
-	
+
 	if (!options || !*options)
 		return 1;
-	
+
 	if (strncmp(options, "off", 3) == 0) {
 		stifb_disabled = 1;
 		options += 3;
-- 
GitLab


From 7470849745e6cd746ae773a6e59b309867310181 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:43 +0200
Subject: [PATCH 0169/1791] video: Move HP PARISC STI core code to shared
 location

STI core files have been located in console and fbdev code. Move
the source code and header to the directories for video helpers.
Also update the config and build rules such that the code depends
on the config symbol CONFIG_STI_CORE, which STI console and STI
framebuffer select automatically.

Cleans up the console makefile and prepares PARISC to implement
fb_is_primary_device() within the arch/ directory. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-12-tzimmermann@suse.de
---
 drivers/video/Kconfig                            | 7 +++++++
 drivers/video/Makefile                           | 1 +
 drivers/video/console/Kconfig                    | 1 +
 drivers/video/console/Makefile                   | 4 +---
 drivers/video/console/sticon.c                   | 2 +-
 drivers/video/fbdev/Kconfig                      | 3 +--
 drivers/video/fbdev/stifb.c                      | 2 +-
 drivers/video/{console => }/sticore.c            | 2 +-
 {drivers/video/fbdev => include/video}/sticore.h | 0
 9 files changed, 14 insertions(+), 8 deletions(-)
 rename drivers/video/{console => }/sticore.c (99%)
 rename {drivers/video/fbdev => include/video}/sticore.h (100%)

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index bf05363d89062..8b2b9ac37c3df 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -11,6 +11,13 @@ config APERTURE_HELPERS
 	  Support tracking and hand-over of aperture ownership. Required
 	  by graphics drivers for firmware-provided framebuffers.
 
+config STI_CORE
+	bool
+	depends on PARISC
+	help
+	  STI refers to the HP "Standard Text Interface" which is a set of
+	  BIOS routines contained in a ROM chip in HP PA-RISC based machines.
+
 config VIDEO_CMDLINE
 	bool
 
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 831c9fa57a6c9..6bbc039508995 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 
 obj-$(CONFIG_APERTURE_HELPERS)    += aperture.o
+obj-$(CONFIG_STI_CORE)            += sticore.o
 obj-$(CONFIG_VGASTATE)            += vgastate.o
 obj-$(CONFIG_VIDEO_CMDLINE)       += cmdline.o
 obj-$(CONFIG_VIDEO_NOMODESET)     += nomodeset.o
diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig
index 22cea5082ac46..a2a88d42edf0c 100644
--- a/drivers/video/console/Kconfig
+++ b/drivers/video/console/Kconfig
@@ -141,6 +141,7 @@ config STI_CONSOLE
 	depends on PARISC && HAS_IOMEM
 	select FONT_SUPPORT
 	select CRC32
+	select STI_CORE
 	default y
 	help
 	  The STI console is the builtin display/keyboard on HP-PARISC
diff --git a/drivers/video/console/Makefile b/drivers/video/console/Makefile
index db07b784bd2cc..fd79016a0d95b 100644
--- a/drivers/video/console/Makefile
+++ b/drivers/video/console/Makefile
@@ -5,8 +5,6 @@
 
 obj-$(CONFIG_DUMMY_CONSOLE)       += dummycon.o
 obj-$(CONFIG_SGI_NEWPORT_CONSOLE) += newport_con.o
-obj-$(CONFIG_STI_CONSOLE)         += sticon.o sticore.o
+obj-$(CONFIG_STI_CONSOLE)         += sticon.o
 obj-$(CONFIG_VGA_CONSOLE)         += vgacon.o
 obj-$(CONFIG_MDA_CONSOLE)         += mdacon.o
-
-obj-$(CONFIG_FB_STI)              += sticore.o
diff --git a/drivers/video/console/sticon.c b/drivers/video/console/sticon.c
index 89ad7ade6cf9f..d11cfd2d68b54 100644
--- a/drivers/video/console/sticon.c
+++ b/drivers/video/console/sticon.c
@@ -50,7 +50,7 @@
 
 #include <asm/io.h>
 
-#include "../fbdev/sticore.h"
+#include <video/sticore.h>
 
 /* switching to graphics mode */
 #define BLANK 0
diff --git a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig
index 96e91570cdd32..485e8c35d5c68 100644
--- a/drivers/video/fbdev/Kconfig
+++ b/drivers/video/fbdev/Kconfig
@@ -551,10 +551,9 @@ config FB_STI
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
+	select STI_CORE
 	default y
 	help
-	  STI refers to the HP "Standard Text Interface" which is a set of
-	  BIOS routines contained in a ROM chip in HP PA-RISC based machines.
 	  Enabling this option will implement the linux framebuffer device
 	  using calls to the STI BIOS routines for initialisation.
 
diff --git a/drivers/video/fbdev/stifb.c b/drivers/video/fbdev/stifb.c
index 99996bc7e6d97..baca6974e288e 100644
--- a/drivers/video/fbdev/stifb.c
+++ b/drivers/video/fbdev/stifb.c
@@ -69,7 +69,7 @@
 #include <asm/grfioctl.h>	/* for HP-UX compatibility */
 #include <linux/uaccess.h>
 
-#include "sticore.h"
+#include <video/sticore.h>
 
 /* REGION_BASE(fb_info, index) returns the virtual address for region <index> */
 #define REGION_BASE(fb_info, index) \
diff --git a/drivers/video/console/sticore.c b/drivers/video/sticore.c
similarity index 99%
rename from drivers/video/console/sticore.c
rename to drivers/video/sticore.c
index 6ea9596a3c4b0..f8aaedea437d1 100644
--- a/drivers/video/console/sticore.c
+++ b/drivers/video/sticore.c
@@ -32,7 +32,7 @@
 #include <asm/grfioctl.h>
 #include <asm/fb.h>
 
-#include "../fbdev/sticore.h"
+#include <video/sticore.h>
 
 #define STI_DRIVERVERSION "Version 0.9c"
 
diff --git a/drivers/video/fbdev/sticore.h b/include/video/sticore.h
similarity index 100%
rename from drivers/video/fbdev/sticore.h
rename to include/video/sticore.h
-- 
GitLab


From d77f0ecdac62968ec99a18deff3d91b453974b9c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:44 +0200
Subject: [PATCH 0170/1791] arch/parisc: Remove trailing whitespaces

Fix trailing whitespaces. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Helge Deller <deller@gmx.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-13-tzimmermann@suse.de
---
 arch/parisc/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/parisc/Makefile b/arch/parisc/Makefile
index a2d8600521f93..0d049a6f6a604 100644
--- a/arch/parisc/Makefile
+++ b/arch/parisc/Makefile
@@ -11,7 +11,7 @@
 # Copyright (C) 1994 by Linus Torvalds
 # Portions Copyright (C) 1999 The Puffin Group
 #
-# Modified for PA-RISC Linux by Paul Lahaie, Alex deVries, 
+# Modified for PA-RISC Linux by Paul Lahaie, Alex deVries,
 # Mike Shaver, Helge Deller and Martin K. Petersen
 #
 
-- 
GitLab


From cf41d18b72c8b79f617c59529171faf49aa74dca Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:45 +0200
Subject: [PATCH 0171/1791] arch/parisc: Implement fb_is_primary_device() under
 arch/parisc

Move PARISC's implementation of fb_is_primary_device() into the
architecture directory. This the place of the declaration and
where other architectures implement this function. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Helge Deller <deller@gmx.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-14-tzimmermann@suse.de
---
 arch/parisc/Makefile         |  2 ++
 arch/parisc/include/asm/fb.h |  2 +-
 arch/parisc/video/Makefile   |  3 +++
 arch/parisc/video/fbdev.c    | 27 +++++++++++++++++++++++++++
 drivers/video/sticore.c      | 19 -------------------
 include/video/sticore.h      |  2 ++
 6 files changed, 35 insertions(+), 20 deletions(-)
 create mode 100644 arch/parisc/video/Makefile
 create mode 100644 arch/parisc/video/fbdev.c

diff --git a/arch/parisc/Makefile b/arch/parisc/Makefile
index 0d049a6f6a604..968ebe17494c5 100644
--- a/arch/parisc/Makefile
+++ b/arch/parisc/Makefile
@@ -119,6 +119,8 @@ export LIBGCC
 
 libs-y	+= arch/parisc/lib/ $(LIBGCC)
 
+drivers-y += arch/parisc/video/
+
 boot	:= arch/parisc/boot
 
 PALO := $(shell if (which palo 2>&1); then : ; \
diff --git a/arch/parisc/include/asm/fb.h b/arch/parisc/include/asm/fb.h
index 55d29c4f716e6..0b9a38ced5c8e 100644
--- a/arch/parisc/include/asm/fb.h
+++ b/arch/parisc/include/asm/fb.h
@@ -12,7 +12,7 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 	pgprot_val(vma->vm_page_prot) |= _PAGE_NO_CACHE;
 }
 
-#if defined(CONFIG_FB_STI)
+#if defined(CONFIG_STI_CORE)
 int fb_is_primary_device(struct fb_info *info);
 #else
 static inline int fb_is_primary_device(struct fb_info *info)
diff --git a/arch/parisc/video/Makefile b/arch/parisc/video/Makefile
new file mode 100644
index 0000000000000..16a73cce46612
--- /dev/null
+++ b/arch/parisc/video/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_STI_CORE) += fbdev.o
diff --git a/arch/parisc/video/fbdev.c b/arch/parisc/video/fbdev.c
new file mode 100644
index 0000000000000..4a0ae08fc75b3
--- /dev/null
+++ b/arch/parisc/video/fbdev.c
@@ -0,0 +1,27 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2000 Philipp Rumpf <prumpf@tux.org>
+ * Copyright (C) 2001-2020 Helge Deller <deller@gmx.de>
+ * Copyright (C) 2001-2002 Thomas Bogendoerfer <tsbogend@alpha.franken.de>
+ */
+
+#include <linux/module.h>
+
+#include <asm/fb.h>
+
+#include <video/sticore.h>
+
+int fb_is_primary_device(struct fb_info *info)
+{
+	struct sti_struct *sti;
+
+	sti = sti_get_rom(0);
+
+	/* if no built-in graphics card found, allow any fb driver as default */
+	if (!sti)
+		return true;
+
+	/* return true if it's the default built-in framebuffer driver */
+	return (sti->info == info);
+}
+EXPORT_SYMBOL(fb_is_primary_device);
diff --git a/drivers/video/sticore.c b/drivers/video/sticore.c
index f8aaedea437d1..7eb925f2ba9c9 100644
--- a/drivers/video/sticore.c
+++ b/drivers/video/sticore.c
@@ -30,7 +30,6 @@
 #include <asm/pdc.h>
 #include <asm/cacheflush.h>
 #include <asm/grfioctl.h>
-#include <asm/fb.h>
 
 #include <video/sticore.h>
 
@@ -1148,24 +1147,6 @@ int sti_call(const struct sti_struct *sti, unsigned long func,
 	return ret;
 }
 
-#if defined(CONFIG_FB_STI)
-/* check if given fb_info is the primary device */
-int fb_is_primary_device(struct fb_info *info)
-{
-	struct sti_struct *sti;
-
-	sti = sti_get_rom(0);
-
-	/* if no built-in graphics card found, allow any fb driver as default */
-	if (!sti)
-		return true;
-
-	/* return true if it's the default built-in framebuffer driver */
-	return (sti->info == info);
-}
-EXPORT_SYMBOL(fb_is_primary_device);
-#endif
-
 MODULE_AUTHOR("Philipp Rumpf, Helge Deller, Thomas Bogendoerfer");
 MODULE_DESCRIPTION("Core STI driver for HP's NGLE series graphics cards in HP PARISC machines");
 MODULE_LICENSE("GPL v2");
diff --git a/include/video/sticore.h b/include/video/sticore.h
index c0879352cde44..fbb78d7e7565e 100644
--- a/include/video/sticore.h
+++ b/include/video/sticore.h
@@ -2,6 +2,8 @@
 #ifndef STICORE_H
 #define STICORE_H
 
+struct fb_info;
+
 /* generic STI structures & functions */
 
 #define MAX_STI_ROMS 4		/* max no. of ROMs which this driver handles */
-- 
GitLab


From f3424d287626f223cd9851e4d14a5c37b8d04028 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:46 +0200
Subject: [PATCH 0172/1791] arch/parisc: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fbdev helpers with the generic ones
from <asm-generic/fb.h>. On PARISC, pgprot_writecombine() and
pgprot_noncached() are the same; hence no functional changes.

v3:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Helge Deller <deller@gmx.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-15-tzimmermann@suse.de
---
 arch/parisc/include/asm/fb.h | 18 ++++--------------
 1 file changed, 4 insertions(+), 14 deletions(-)

diff --git a/arch/parisc/include/asm/fb.h b/arch/parisc/include/asm/fb.h
index 0b9a38ced5c8e..658a8a7dc5312 100644
--- a/arch/parisc/include/asm/fb.h
+++ b/arch/parisc/include/asm/fb.h
@@ -2,23 +2,13 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	pgprot_val(vma->vm_page_prot) |= _PAGE_NO_CACHE;
-}
+struct fb_info;
 
 #if defined(CONFIG_STI_CORE)
 int fb_is_primary_device(struct fb_info *info);
-#else
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#define fb_is_primary_device fb_is_primary_device
 #endif
 
+#include <asm-generic/fb.h>
+
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From b50f6d465ac1fb37cab7bfb39ac6c9fb9d8e3716 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:47 +0200
Subject: [PATCH 0173/1791] arch/powerpc: Implement <asm/fb.h> with generic
 helpers

Replace the architecture's fb_is_primary_device() with the generic
one from <asm-generic/fb.h>. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-16-tzimmermann@suse.de
---
 arch/powerpc/include/asm/fb.h | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/include/asm/fb.h b/arch/powerpc/include/asm/fb.h
index 6541ab77c5b9f..5f1a2e5f76548 100644
--- a/arch/powerpc/include/asm/fb.h
+++ b/arch/powerpc/include/asm/fb.h
@@ -2,8 +2,8 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
 #include <linux/fs.h>
+
 #include <asm/page.h>
 
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
@@ -13,10 +13,8 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 						 vma->vm_end - vma->vm_start,
 						 vma->vm_page_prot);
 }
+#define fb_pgprotect fb_pgprotect
 
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From b6cf29637be638a6846d2d297d80cf7c6fb74faf Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:48 +0200
Subject: [PATCH 0174/1791] arch/sh: Implement <asm/fb.h> with generic helpers

Replace the architecture's fbdev helpers with the generic
ones from <asm-generic/fb.h>. No functional changes.

v2:
	* use default implementation for fb_pgprotect() (Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Rich Felker <dalias@libc.org>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Acked-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-17-tzimmermann@suse.de
---
 arch/sh/include/asm/fb.h | 15 +--------------
 1 file changed, 1 insertion(+), 14 deletions(-)

diff --git a/arch/sh/include/asm/fb.h b/arch/sh/include/asm/fb.h
index 9a0bca2686fd3..19df13ee9ca73 100644
--- a/arch/sh/include/asm/fb.h
+++ b/arch/sh/include/asm/fb.h
@@ -2,19 +2,6 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
-#include <linux/fb.h>
-#include <linux/fs.h>
-#include <asm/page.h>
-
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-}
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
-- 
GitLab


From 4eec0b3048fcd74b504c2a6828a07f133a8ab508 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:49 +0200
Subject: [PATCH 0175/1791] arch/sparc: Implement fb_is_primary_device() in
 source file

Other architectures implment fb_is_primary_device() in a source
file. Do the same on sparc. No functional changes, but allows to
remove several include statement from <asm/fb.h>.

v2:
	* don't include <asm/prom.h> in header file

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-18-tzimmermann@suse.de
---
 arch/sparc/Makefile         |  1 +
 arch/sparc/include/asm/fb.h | 23 +++++------------------
 arch/sparc/video/Makefile   |  3 +++
 arch/sparc/video/fbdev.c    | 24 ++++++++++++++++++++++++
 4 files changed, 33 insertions(+), 18 deletions(-)
 create mode 100644 arch/sparc/video/Makefile
 create mode 100644 arch/sparc/video/fbdev.c

diff --git a/arch/sparc/Makefile b/arch/sparc/Makefile
index a4ea5b05f2885..95a9211e48e36 100644
--- a/arch/sparc/Makefile
+++ b/arch/sparc/Makefile
@@ -60,6 +60,7 @@ libs-y                 += arch/sparc/prom/
 libs-y                 += arch/sparc/lib/
 
 drivers-$(CONFIG_PM) += arch/sparc/power/
+drivers-$(CONFIG_FB) += arch/sparc/video/
 
 boot := arch/sparc/boot
 
diff --git a/arch/sparc/include/asm/fb.h b/arch/sparc/include/asm/fb.h
index f699962e9ddf0..28609f7a965c2 100644
--- a/arch/sparc/include/asm/fb.h
+++ b/arch/sparc/include/asm/fb.h
@@ -1,11 +1,12 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 #ifndef _SPARC_FB_H_
 #define _SPARC_FB_H_
-#include <linux/console.h>
-#include <linux/fb.h>
+
 #include <linux/fs.h>
+
 #include <asm/page.h>
-#include <asm/prom.h>
+
+struct fb_info;
 
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
@@ -15,20 +16,6 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 #endif
 }
 
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	struct device *dev = info->device;
-	struct device_node *node;
-
-	if (console_set_on_cmdline)
-		return 0;
-
-	node = dev->of_node;
-	if (node &&
-	    node == of_console_device)
-		return 1;
-
-	return 0;
-}
+int fb_is_primary_device(struct fb_info *info);
 
 #endif /* _SPARC_FB_H_ */
diff --git a/arch/sparc/video/Makefile b/arch/sparc/video/Makefile
new file mode 100644
index 0000000000000..6baddbd58e4db
--- /dev/null
+++ b/arch/sparc/video/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_FB) += fbdev.o
diff --git a/arch/sparc/video/fbdev.c b/arch/sparc/video/fbdev.c
new file mode 100644
index 0000000000000..dadd5799fbb32
--- /dev/null
+++ b/arch/sparc/video/fbdev.c
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/console.h>
+#include <linux/fb.h>
+#include <linux/module.h>
+
+#include <asm/fb.h>
+#include <asm/prom.h>
+
+int fb_is_primary_device(struct fb_info *info)
+{
+	struct device *dev = info->device;
+	struct device_node *node;
+
+	if (console_set_on_cmdline)
+		return 0;
+
+	node = dev->of_node;
+	if (node && node == of_console_device)
+		return 1;
+
+	return 0;
+}
+EXPORT_SYMBOL(fb_is_primary_device);
-- 
GitLab


From db76f19a607d6cf4a9d5868d05675c355e99809c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:50 +0200
Subject: [PATCH 0176/1791] arch/sparc: Implement <asm/fb.h> with generic
 helpers

Replace sparc64's fb_pgprotect() with the generic one
from <asm-generic/fb.h>. On sparc, pgprot_writecombine() and
pgprot_noncached() are the same; hence no functional changes

v3:
	* use default implementation for fb_pgprotect() on
	  sparc64 (Arnd)
v2:
	* restore the original fb_pgprotect()

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-19-tzimmermann@suse.de
---
 arch/sparc/include/asm/fb.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/sparc/include/asm/fb.h b/arch/sparc/include/asm/fb.h
index 28609f7a965c2..689ee5c60054c 100644
--- a/arch/sparc/include/asm/fb.h
+++ b/arch/sparc/include/asm/fb.h
@@ -2,20 +2,20 @@
 #ifndef _SPARC_FB_H_
 #define _SPARC_FB_H_
 
-#include <linux/fs.h>
-
-#include <asm/page.h>
-
 struct fb_info;
+struct file;
+struct vm_area_struct;
 
+#ifdef CONFIG_SPARC32
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
-{
-#ifdef CONFIG_SPARC64
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+{ }
+#define fb_pgprotect fb_pgprotect
 #endif
-}
 
 int fb_is_primary_device(struct fb_info *info);
+#define fb_is_primary_device fb_is_primary_device
+
+#include <asm-generic/fb.h>
 
 #endif /* _SPARC_FB_H_ */
-- 
GitLab


From 91c249b2b9f6a80c744387b6713adf275ffd296b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 17 Apr 2023 14:56:51 +0200
Subject: [PATCH 0177/1791] arch/x86: Implement <asm/fb.h> with generic helpers

Include <asm-generic/fb.h> and set the required preprocessor tokens
correctly. x86 now implements its own set of fb helpers, but still
follows the overall pattern of the other <asm/fb.h> files.

v3:
	* clarified commit message

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417125651.25126-20-tzimmermann@suse.de
---
 arch/x86/include/asm/fb.h | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/arch/x86/include/asm/fb.h b/arch/x86/include/asm/fb.h
index ab4c960146e3d..a3fb801f12f18 100644
--- a/arch/x86/include/asm/fb.h
+++ b/arch/x86/include/asm/fb.h
@@ -2,10 +2,11 @@
 #ifndef _ASM_X86_FB_H
 #define _ASM_X86_FB_H
 
-#include <linux/fb.h>
-#include <linux/fs.h>
 #include <asm/page.h>
 
+struct fb_info;
+struct file;
+
 static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 				unsigned long off)
 {
@@ -16,7 +17,11 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 		pgprot_val(vma->vm_page_prot) =
 			prot | cachemode2protval(_PAGE_CACHE_MODE_UC_MINUS);
 }
+#define fb_pgprotect fb_pgprotect
+
+int fb_is_primary_device(struct fb_info *info);
+#define fb_is_primary_device fb_is_primary_device
 
-extern int fb_is_primary_device(struct fb_info *info);
+#include <asm-generic/fb.h>
 
 #endif /* _ASM_X86_FB_H */
-- 
GitLab


From b13604c0ff26ecde5f708ef5e1fd233f8e89e1f6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 17 Apr 2023 18:37:40 +0300
Subject: [PATCH 0178/1791] drm/i915/display: split out load detect to a
 separate file

Load detect is shared between tv and crt but otherwise isolated in
intel_display.c.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417153741.1074692-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |   1 +
 drivers/gpu/drm/i915/display/intel_crt.c      |   5 +-
 drivers/gpu/drm/i915/display/intel_display.c  | 215 ----------------
 drivers/gpu/drm/i915/display/intel_display.h  |   7 -
 .../drm/i915/display/intel_display_types.h    |   4 -
 .../gpu/drm/i915/display/intel_load_detect.c  | 229 ++++++++++++++++++
 .../gpu/drm/i915/display/intel_load_detect.h  |  24 ++
 drivers/gpu/drm/i915/display/intel_tv.c       |   5 +-
 8 files changed, 260 insertions(+), 230 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_load_detect.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_load_detect.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 154986359169e..9af76e376ca91 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -261,6 +261,7 @@ i915-y += \
 	display/intel_hdcp_gsc.o \
 	display/intel_hotplug.o \
 	display/intel_hti.o \
+	display/intel_load_detect.o \
 	display/intel_lpe_audio.o \
 	display/intel_modeset_verify.o \
 	display/intel_modeset_setup.o \
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 38e9c61c2344e..96acdf98a0c03 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -48,6 +48,7 @@
 #include "intel_fifo_underrun.h"
 #include "intel_gmbus.h"
 #include "intel_hotplug.h"
+#include "intel_load_detect.h"
 #include "intel_pch_display.h"
 #include "intel_pch_refclk.h"
 
@@ -881,7 +882,7 @@ intel_crt_detect(struct drm_connector *connector,
 	}
 
 	/* for pre-945g platforms use load detect */
-	ret = intel_get_load_detect_pipe(connector, &tmp, ctx);
+	ret = intel_load_detect_get_pipe(connector, &tmp, ctx);
 	if (ret > 0) {
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
@@ -892,7 +893,7 @@ intel_crt_detect(struct drm_connector *connector,
 			status = connector_status_disconnected;
 		else
 			status = connector_status_unknown;
-		intel_release_load_detect_pipe(connector, &tmp, ctx);
+		intel_load_detect_release_pipe(connector, &tmp, ctx);
 	} else if (ret == 0) {
 		status = connector_status_unknown;
 	} else {
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 1c264c17b6e4b..2d4a6936ab2bf 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3821,221 +3821,6 @@ bool intel_crtc_get_pipe_config(struct intel_crtc_state *crtc_state)
 	return true;
 }
 
-/* VESA 640x480x72Hz mode to set on the pipe */
-static const struct drm_display_mode load_detect_mode = {
-	DRM_MODE("640x480", DRM_MODE_TYPE_DEFAULT, 31500, 640, 664,
-		 704, 832, 0, 480, 489, 491, 520, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC),
-};
-
-static int intel_modeset_disable_planes(struct drm_atomic_state *state,
-					struct drm_crtc *crtc)
-{
-	struct drm_plane *plane;
-	struct drm_plane_state *plane_state;
-	int ret, i;
-
-	ret = drm_atomic_add_affected_planes(state, crtc);
-	if (ret)
-		return ret;
-
-	for_each_new_plane_in_state(state, plane, plane_state, i) {
-		if (plane_state->crtc != crtc)
-			continue;
-
-		ret = drm_atomic_set_crtc_for_plane(plane_state, NULL);
-		if (ret)
-			return ret;
-
-		drm_atomic_set_fb_for_plane(plane_state, NULL);
-	}
-
-	return 0;
-}
-
-int intel_get_load_detect_pipe(struct drm_connector *connector,
-			       struct intel_load_detect_pipe *old,
-			       struct drm_modeset_acquire_ctx *ctx)
-{
-	struct intel_encoder *encoder =
-		intel_attached_encoder(to_intel_connector(connector));
-	struct intel_crtc *possible_crtc;
-	struct intel_crtc *crtc = NULL;
-	struct drm_device *dev = encoder->base.dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct drm_mode_config *config = &dev->mode_config;
-	struct drm_atomic_state *state = NULL, *restore_state = NULL;
-	struct drm_connector_state *connector_state;
-	struct intel_crtc_state *crtc_state;
-	int ret;
-
-	drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		    connector->base.id, connector->name,
-		    encoder->base.base.id, encoder->base.name);
-
-	old->restore_state = NULL;
-
-	drm_WARN_ON(dev, !drm_modeset_is_locked(&config->connection_mutex));
-
-	/*
-	 * Algorithm gets a little messy:
-	 *
-	 *   - if the connector already has an assigned crtc, use it (but make
-	 *     sure it's on first)
-	 *
-	 *   - try to find the first unused crtc that can drive this connector,
-	 *     and use that if we find one
-	 */
-
-	/* See if we already have a CRTC for this connector */
-	if (connector->state->crtc) {
-		crtc = to_intel_crtc(connector->state->crtc);
-
-		ret = drm_modeset_lock(&crtc->base.mutex, ctx);
-		if (ret)
-			goto fail;
-
-		/* Make sure the crtc and connector are running */
-		goto found;
-	}
-
-	/* Find an unused one (if possible) */
-	for_each_intel_crtc(dev, possible_crtc) {
-		if (!(encoder->base.possible_crtcs &
-		      drm_crtc_mask(&possible_crtc->base)))
-			continue;
-
-		ret = drm_modeset_lock(&possible_crtc->base.mutex, ctx);
-		if (ret)
-			goto fail;
-
-		if (possible_crtc->base.state->enable) {
-			drm_modeset_unlock(&possible_crtc->base.mutex);
-			continue;
-		}
-
-		crtc = possible_crtc;
-		break;
-	}
-
-	/*
-	 * If we didn't find an unused CRTC, don't use any.
-	 */
-	if (!crtc) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "no pipe available for load-detect\n");
-		ret = -ENODEV;
-		goto fail;
-	}
-
-found:
-	state = drm_atomic_state_alloc(dev);
-	restore_state = drm_atomic_state_alloc(dev);
-	if (!state || !restore_state) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
-	state->acquire_ctx = ctx;
-	to_intel_atomic_state(state)->internal = true;
-
-	restore_state->acquire_ctx = ctx;
-	to_intel_atomic_state(restore_state)->internal = true;
-
-	connector_state = drm_atomic_get_connector_state(state, connector);
-	if (IS_ERR(connector_state)) {
-		ret = PTR_ERR(connector_state);
-		goto fail;
-	}
-
-	ret = drm_atomic_set_crtc_for_connector(connector_state, &crtc->base);
-	if (ret)
-		goto fail;
-
-	crtc_state = intel_atomic_get_crtc_state(state, crtc);
-	if (IS_ERR(crtc_state)) {
-		ret = PTR_ERR(crtc_state);
-		goto fail;
-	}
-
-	crtc_state->uapi.active = true;
-
-	ret = drm_atomic_set_mode_for_crtc(&crtc_state->uapi,
-					   &load_detect_mode);
-	if (ret)
-		goto fail;
-
-	ret = intel_modeset_disable_planes(state, &crtc->base);
-	if (ret)
-		goto fail;
-
-	ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(restore_state, connector));
-	if (!ret)
-		ret = PTR_ERR_OR_ZERO(drm_atomic_get_crtc_state(restore_state, &crtc->base));
-	if (!ret)
-		ret = drm_atomic_add_affected_planes(restore_state, &crtc->base);
-	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "Failed to create a copy of old state to restore: %i\n",
-			    ret);
-		goto fail;
-	}
-
-	ret = drm_atomic_commit(state);
-	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
-			    "failed to set mode on load-detect pipe\n");
-		goto fail;
-	}
-
-	old->restore_state = restore_state;
-	drm_atomic_state_put(state);
-
-	/* let the connector get through one full cycle before testing */
-	intel_crtc_wait_for_next_vblank(crtc);
-
-	return true;
-
-fail:
-	if (state) {
-		drm_atomic_state_put(state);
-		state = NULL;
-	}
-	if (restore_state) {
-		drm_atomic_state_put(restore_state);
-		restore_state = NULL;
-	}
-
-	if (ret == -EDEADLK)
-		return ret;
-
-	return false;
-}
-
-void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
-				    struct drm_modeset_acquire_ctx *ctx)
-{
-	struct intel_encoder *intel_encoder =
-		intel_attached_encoder(to_intel_connector(connector));
-	struct drm_i915_private *i915 = to_i915(intel_encoder->base.dev);
-	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_atomic_state *state = old->restore_state;
-	int ret;
-
-	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		    connector->base.id, connector->name,
-		    encoder->base.id, encoder->name);
-
-	if (!state)
-		return;
-
-	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
-	if (ret)
-		drm_dbg_kms(&i915->drm,
-			    "Couldn't release load detect pipe: %i\n", ret);
-	drm_atomic_state_put(state);
-}
-
 static int i9xx_pll_refclk(struct drm_device *dev,
 			   const struct intel_crtc_state *pipe_config)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index fefad8cfc8a28..ac95961f68ba7 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -56,7 +56,6 @@ struct intel_dp;
 struct intel_encoder;
 struct intel_initial_plane_config;
 struct intel_link_m_n;
-struct intel_load_detect_pipe;
 struct intel_plane;
 struct intel_plane_state;
 struct intel_power_domain_mask;
@@ -455,12 +454,6 @@ int ilk_get_lanes_required(int target_clock, int link_bw, int bpp);
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 			 struct intel_digital_port *dig_port,
 			 unsigned int expected_mask);
-int intel_get_load_detect_pipe(struct drm_connector *connector,
-			       struct intel_load_detect_pipe *old,
-			       struct drm_modeset_acquire_ctx *ctx);
-void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
-				    struct drm_modeset_acquire_ctx *ctx);
 struct drm_framebuffer *
 intel_framebuffer_create(struct drm_i915_gem_object *obj,
 			 struct drm_mode_fb_cmd2 *mode_cmd);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 64b6a5ec4b818..35c260bd14618 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1838,10 +1838,6 @@ struct intel_dp_mst_encoder {
 	struct intel_connector *connector;
 };
 
-struct intel_load_detect_pipe {
-	struct drm_atomic_state *restore_state;
-};
-
 static inline struct intel_encoder *
 intel_attached_encoder(struct intel_connector *connector)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.c b/drivers/gpu/drm/i915/display/intel_load_detect.c
new file mode 100644
index 0000000000000..5d6bb6d712bc4
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_load_detect.c
@@ -0,0 +1,229 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic_uapi.h>
+
+#include "i915_drv.h"
+#include "intel_atomic.h"
+#include "intel_crtc.h"
+#include "intel_display_types.h"
+#include "intel_load_detect.h"
+
+/* VESA 640x480x72Hz mode to set on the pipe */
+static const struct drm_display_mode load_detect_mode = {
+	DRM_MODE("640x480", DRM_MODE_TYPE_DEFAULT, 31500, 640, 664,
+		 704, 832, 0, 480, 489, 491, 520, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC),
+};
+
+static int intel_modeset_disable_planes(struct drm_atomic_state *state,
+					struct drm_crtc *crtc)
+{
+	struct drm_plane *plane;
+	struct drm_plane_state *plane_state;
+	int ret, i;
+
+	ret = drm_atomic_add_affected_planes(state, crtc);
+	if (ret)
+		return ret;
+
+	for_each_new_plane_in_state(state, plane, plane_state, i) {
+		if (plane_state->crtc != crtc)
+			continue;
+
+		ret = drm_atomic_set_crtc_for_plane(plane_state, NULL);
+		if (ret)
+			return ret;
+
+		drm_atomic_set_fb_for_plane(plane_state, NULL);
+	}
+
+	return 0;
+}
+
+int intel_load_detect_get_pipe(struct drm_connector *connector,
+			       struct intel_load_detect_pipe *old,
+			       struct drm_modeset_acquire_ctx *ctx)
+{
+	struct intel_encoder *encoder =
+		intel_attached_encoder(to_intel_connector(connector));
+	struct intel_crtc *possible_crtc;
+	struct intel_crtc *crtc = NULL;
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_atomic_state *state = NULL, *restore_state = NULL;
+	struct drm_connector_state *connector_state;
+	struct intel_crtc_state *crtc_state;
+	int ret;
+
+	drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
+		    connector->base.id, connector->name,
+		    encoder->base.base.id, encoder->base.name);
+
+	old->restore_state = NULL;
+
+	drm_WARN_ON(dev, !drm_modeset_is_locked(&config->connection_mutex));
+
+	/*
+	 * Algorithm gets a little messy:
+	 *
+	 *   - if the connector already has an assigned crtc, use it (but make
+	 *     sure it's on first)
+	 *
+	 *   - try to find the first unused crtc that can drive this connector,
+	 *     and use that if we find one
+	 */
+
+	/* See if we already have a CRTC for this connector */
+	if (connector->state->crtc) {
+		crtc = to_intel_crtc(connector->state->crtc);
+
+		ret = drm_modeset_lock(&crtc->base.mutex, ctx);
+		if (ret)
+			goto fail;
+
+		/* Make sure the crtc and connector are running */
+		goto found;
+	}
+
+	/* Find an unused one (if possible) */
+	for_each_intel_crtc(dev, possible_crtc) {
+		if (!(encoder->base.possible_crtcs &
+		      drm_crtc_mask(&possible_crtc->base)))
+			continue;
+
+		ret = drm_modeset_lock(&possible_crtc->base.mutex, ctx);
+		if (ret)
+			goto fail;
+
+		if (possible_crtc->base.state->enable) {
+			drm_modeset_unlock(&possible_crtc->base.mutex);
+			continue;
+		}
+
+		crtc = possible_crtc;
+		break;
+	}
+
+	/*
+	 * If we didn't find an unused CRTC, don't use any.
+	 */
+	if (!crtc) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "no pipe available for load-detect\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+
+found:
+	state = drm_atomic_state_alloc(dev);
+	restore_state = drm_atomic_state_alloc(dev);
+	if (!state || !restore_state) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	state->acquire_ctx = ctx;
+	to_intel_atomic_state(state)->internal = true;
+
+	restore_state->acquire_ctx = ctx;
+	to_intel_atomic_state(restore_state)->internal = true;
+
+	connector_state = drm_atomic_get_connector_state(state, connector);
+	if (IS_ERR(connector_state)) {
+		ret = PTR_ERR(connector_state);
+		goto fail;
+	}
+
+	ret = drm_atomic_set_crtc_for_connector(connector_state, &crtc->base);
+	if (ret)
+		goto fail;
+
+	crtc_state = intel_atomic_get_crtc_state(state, crtc);
+	if (IS_ERR(crtc_state)) {
+		ret = PTR_ERR(crtc_state);
+		goto fail;
+	}
+
+	crtc_state->uapi.active = true;
+
+	ret = drm_atomic_set_mode_for_crtc(&crtc_state->uapi,
+					   &load_detect_mode);
+	if (ret)
+		goto fail;
+
+	ret = intel_modeset_disable_planes(state, &crtc->base);
+	if (ret)
+		goto fail;
+
+	ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(restore_state, connector));
+	if (!ret)
+		ret = PTR_ERR_OR_ZERO(drm_atomic_get_crtc_state(restore_state, &crtc->base));
+	if (!ret)
+		ret = drm_atomic_add_affected_planes(restore_state, &crtc->base);
+	if (ret) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "Failed to create a copy of old state to restore: %i\n",
+			    ret);
+		goto fail;
+	}
+
+	ret = drm_atomic_commit(state);
+	if (ret) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "failed to set mode on load-detect pipe\n");
+		goto fail;
+	}
+
+	old->restore_state = restore_state;
+	drm_atomic_state_put(state);
+
+	/* let the connector get through one full cycle before testing */
+	intel_crtc_wait_for_next_vblank(crtc);
+
+	return true;
+
+fail:
+	if (state) {
+		drm_atomic_state_put(state);
+		state = NULL;
+	}
+	if (restore_state) {
+		drm_atomic_state_put(restore_state);
+		restore_state = NULL;
+	}
+
+	if (ret == -EDEADLK)
+		return ret;
+
+	return false;
+}
+
+void intel_load_detect_release_pipe(struct drm_connector *connector,
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx)
+{
+	struct intel_encoder *intel_encoder =
+		intel_attached_encoder(to_intel_connector(connector));
+	struct drm_i915_private *i915 = to_i915(intel_encoder->base.dev);
+	struct drm_encoder *encoder = &intel_encoder->base;
+	struct drm_atomic_state *state = old->restore_state;
+	int ret;
+
+	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
+		    connector->base.id, connector->name,
+		    encoder->base.id, encoder->name);
+
+	if (!state)
+		return;
+
+	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
+	if (ret)
+		drm_dbg_kms(&i915->drm,
+			    "Couldn't release load detect pipe: %i\n", ret);
+	drm_atomic_state_put(state);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.h b/drivers/gpu/drm/i915/display/intel_load_detect.h
new file mode 100644
index 0000000000000..9b69da1867a58
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_load_detect.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_LOAD_DETECT_H__
+#define __INTEL_LOAD_DETECT_H__
+
+struct drm_atomic_state;
+struct drm_connector;
+struct drm_modeset_acquire_ctx;
+
+struct intel_load_detect_pipe {
+	struct drm_atomic_state *restore_state;
+};
+
+int intel_load_detect_get_pipe(struct drm_connector *connector,
+			       struct intel_load_detect_pipe *old,
+			       struct drm_modeset_acquire_ctx *ctx);
+void intel_load_detect_release_pipe(struct drm_connector *connector,
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx);
+
+#endif /* __INTEL_LOAD_DETECT_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index 557ec5b62afa2..07e7f7cdd9615 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -43,6 +43,7 @@
 #include "intel_display_types.h"
 #include "intel_dpll.h"
 #include "intel_hotplug.h"
+#include "intel_load_detect.h"
 #include "intel_tv.h"
 #include "intel_tv_regs.h"
 
@@ -1725,13 +1726,13 @@ intel_tv_detect(struct drm_connector *connector,
 		struct intel_load_detect_pipe tmp;
 		int ret;
 
-		ret = intel_get_load_detect_pipe(connector, &tmp, ctx);
+		ret = intel_load_detect_get_pipe(connector, &tmp, ctx);
 		if (ret < 0)
 			return ret;
 
 		if (ret > 0) {
 			type = intel_tv_detect_type(intel_tv, connector);
-			intel_release_load_detect_pipe(connector, &tmp, ctx);
+			intel_load_detect_release_pipe(connector, &tmp, ctx);
 			status = type < 0 ?
 				connector_status_disconnected :
 				connector_status_connected;
-- 
GitLab


From 8902a55dfcff7add5d8cc77781ecc311fba8855a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 17 Apr 2023 18:37:41 +0300
Subject: [PATCH 0179/1791] drm/i915/display: throw out struct
 intel_load_detect_pipe

An error-valued pointer can handle all in one without the wrapper
struct.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417153741.1074692-2-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_crt.c      | 18 ++++++++---------
 .../gpu/drm/i915/display/intel_load_detect.c  | 20 ++++++++-----------
 .../gpu/drm/i915/display/intel_load_detect.h  | 12 ++++-------
 drivers/gpu/drm/i915/display/intel_tv.c       | 16 +++++++--------
 4 files changed, 29 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 96acdf98a0c03..13519f78cf9fe 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -822,9 +822,9 @@ intel_crt_detect(struct drm_connector *connector,
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector));
 	struct intel_encoder *intel_encoder = &crt->base;
+	struct drm_atomic_state *state;
 	intel_wakeref_t wakeref;
-	int status, ret;
-	struct intel_load_detect_pipe tmp;
+	int status;
 
 	drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s] force=%d\n",
 		    connector->base.id, connector->name,
@@ -882,8 +882,12 @@ intel_crt_detect(struct drm_connector *connector,
 	}
 
 	/* for pre-945g platforms use load detect */
-	ret = intel_load_detect_get_pipe(connector, &tmp, ctx);
-	if (ret > 0) {
+	state = intel_load_detect_get_pipe(connector, ctx);
+	if (IS_ERR(state)) {
+		status = PTR_ERR(state);
+	} else if (!state) {
+		status = connector_status_unknown;
+	} else {
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
 		else if (DISPLAY_VER(dev_priv) < 4)
@@ -893,11 +897,7 @@ intel_crt_detect(struct drm_connector *connector,
 			status = connector_status_disconnected;
 		else
 			status = connector_status_unknown;
-		intel_load_detect_release_pipe(connector, &tmp, ctx);
-	} else if (ret == 0) {
-		status = connector_status_unknown;
-	} else {
-		status = ret;
+		intel_load_detect_release_pipe(connector, state, ctx);
 	}
 
 out:
diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.c b/drivers/gpu/drm/i915/display/intel_load_detect.c
index 5d6bb6d712bc4..d5a0aecf3e8ff 100644
--- a/drivers/gpu/drm/i915/display/intel_load_detect.c
+++ b/drivers/gpu/drm/i915/display/intel_load_detect.c
@@ -44,9 +44,9 @@ static int intel_modeset_disable_planes(struct drm_atomic_state *state,
 	return 0;
 }
 
-int intel_load_detect_get_pipe(struct drm_connector *connector,
-			       struct intel_load_detect_pipe *old,
-			       struct drm_modeset_acquire_ctx *ctx)
+struct drm_atomic_state *
+intel_load_detect_get_pipe(struct drm_connector *connector,
+			   struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_encoder *encoder =
 		intel_attached_encoder(to_intel_connector(connector));
@@ -64,8 +64,6 @@ int intel_load_detect_get_pipe(struct drm_connector *connector,
 		    connector->base.id, connector->name,
 		    encoder->base.base.id, encoder->base.name);
 
-	old->restore_state = NULL;
-
 	drm_WARN_ON(dev, !drm_modeset_is_locked(&config->connection_mutex));
 
 	/*
@@ -179,13 +177,12 @@ int intel_load_detect_get_pipe(struct drm_connector *connector,
 		goto fail;
 	}
 
-	old->restore_state = restore_state;
 	drm_atomic_state_put(state);
 
 	/* let the connector get through one full cycle before testing */
 	intel_crtc_wait_for_next_vblank(crtc);
 
-	return true;
+	return restore_state;
 
 fail:
 	if (state) {
@@ -198,27 +195,26 @@ int intel_load_detect_get_pipe(struct drm_connector *connector,
 	}
 
 	if (ret == -EDEADLK)
-		return ret;
+		return ERR_PTR(ret);
 
-	return false;
+	return NULL;
 }
 
 void intel_load_detect_release_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
+				    struct drm_atomic_state *state,
 				    struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(to_intel_connector(connector));
 	struct drm_i915_private *i915 = to_i915(intel_encoder->base.dev);
 	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_atomic_state *state = old->restore_state;
 	int ret;
 
 	drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 		    connector->base.id, connector->name,
 		    encoder->base.id, encoder->name);
 
-	if (!state)
+	if (IS_ERR_OR_NULL(state))
 		return;
 
 	ret = drm_atomic_helper_commit_duplicated_state(state, ctx);
diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.h b/drivers/gpu/drm/i915/display/intel_load_detect.h
index 9b69da1867a58..aed51901b9ba8 100644
--- a/drivers/gpu/drm/i915/display/intel_load_detect.h
+++ b/drivers/gpu/drm/i915/display/intel_load_detect.h
@@ -10,15 +10,11 @@ struct drm_atomic_state;
 struct drm_connector;
 struct drm_modeset_acquire_ctx;
 
-struct intel_load_detect_pipe {
-	struct drm_atomic_state *restore_state;
-};
-
-int intel_load_detect_get_pipe(struct drm_connector *connector,
-			       struct intel_load_detect_pipe *old,
-			       struct drm_modeset_acquire_ctx *ctx);
+struct drm_atomic_state *
+intel_load_detect_get_pipe(struct drm_connector *connector,
+			   struct drm_modeset_acquire_ctx *ctx);
 void intel_load_detect_release_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old,
+				    struct drm_atomic_state *old,
 				    struct drm_modeset_acquire_ctx *ctx);
 
 #endif /* __INTEL_LOAD_DETECT_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index 07e7f7cdd9615..e3ccface0c9d9 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -1723,21 +1723,21 @@ intel_tv_detect(struct drm_connector *connector,
 		return connector_status_disconnected;
 
 	if (force) {
-		struct intel_load_detect_pipe tmp;
-		int ret;
+		struct drm_atomic_state *state;
 
-		ret = intel_load_detect_get_pipe(connector, &tmp, ctx);
-		if (ret < 0)
-			return ret;
+		state = intel_load_detect_get_pipe(connector, ctx);
+		if (IS_ERR(state))
+			return PTR_ERR(state);
 
-		if (ret > 0) {
+		if (state) {
 			type = intel_tv_detect_type(intel_tv, connector);
-			intel_load_detect_release_pipe(connector, &tmp, ctx);
+			intel_load_detect_release_pipe(connector, state, ctx);
 			status = type < 0 ?
 				connector_status_disconnected :
 				connector_status_connected;
-		} else
+		} else {
 			status = connector_status_unknown;
+		}
 
 		if (status == connector_status_connected) {
 			intel_tv->type = type;
-- 
GitLab


From 691248d4135fe3fae64b4ee0676bc96a7fd6950c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:14 +0300
Subject: [PATCH 0180/1791] drm/i915: Check pipe source size when using skl+
 scalers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The skl+ scalers only sample 12 bits of PIPESRC so we can't
do any plane scaling at all when the pipe source size is >4k.

Make sure the pipe source size is also below the scaler's src
size limits. Might not be 100% accurate, but should at least be
safe. We can refine the limits later if we discover that recent
hw is less restricted.

Cc: stable@vger.kernel.org
Tested-by: Ross Zwisler <zwisler@google.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8357
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 473d53610b924..0e7e014fcc717 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -111,6 +111,8 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct drm_display_mode *adjusted_mode =
 		&crtc_state->hw.adjusted_mode;
+	int pipe_src_w = drm_rect_width(&crtc_state->pipe_src);
+	int pipe_src_h = drm_rect_height(&crtc_state->pipe_src);
 	int min_src_w, min_src_h, min_dst_w, min_dst_h;
 	int max_src_w, max_src_h, max_dst_w, max_dst_h;
 
@@ -207,6 +209,21 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 		return -EINVAL;
 	}
 
+	/*
+	 * The pipe scaler does not use all the bits of PIPESRC, at least
+	 * on the earlier platforms. So even when we're scaling a plane
+	 * the *pipe* source size must not be too large. For simplicity
+	 * we assume the limits match the scaler source size limits. Might
+	 * not be 100% accurate on all platforms, but good enough for now.
+	 */
+	if (pipe_src_w > max_src_w || pipe_src_h > max_src_h) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "scaler_user index %u.%u: pipe src size %ux%u "
+			    "is out of scaler range\n",
+			    crtc->pipe, scaler_user, pipe_src_w, pipe_src_h);
+		return -EINVAL;
+	}
+
 	/* mark this plane as a scaler user in crtc_state */
 	scaler_state->scaler_users |= (1 << scaler_user);
 	drm_dbg_kms(&dev_priv->drm, "scaler_user index %u.%u: "
-- 
GitLab


From 8637e1c84c5f2c3ea4a1e145ca226fef90a148a2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:15 +0300
Subject: [PATCH 0181/1791] drm/i915: Relocate VBLANK_EVASION_TIME_US
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move the VBLANK_EVASION_TIME_US definition to a slightly
better place.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-3-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_crtc.h            | 10 ++++++++++
 drivers/gpu/drm/i915/display/intel_display_debugfs.c |  2 +-
 drivers/gpu/drm/i915/display/intel_sprite.h          | 10 ----------
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc.h b/drivers/gpu/drm/i915/display/intel_crtc.h
index 73077137fb990..51a4c8df9e657 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.h
+++ b/drivers/gpu/drm/i915/display/intel_crtc.h
@@ -16,6 +16,16 @@ struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
 
+/*
+ * FIXME: We should instead only take spinlocks once for the entire update
+ * instead of once per mmio.
+ */
+#if IS_ENABLED(CONFIG_PROVE_LOCKING)
+#define VBLANK_EVASION_TIME_US 250
+#else
+#define VBLANK_EVASION_TIME_US 100
+#endif
+
 int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
 			     int usecs);
 u32 intel_crtc_max_vblank_count(const struct intel_crtc_state *crtc_state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index abd16a2b1f7a0..e72288662f022 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -12,6 +12,7 @@
 #include "i915_debugfs.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "intel_crtc.h"
 #include "intel_de.h"
 #include "intel_crtc_state_dump.h"
 #include "intel_display_debugfs.h"
@@ -30,7 +31,6 @@
 #include "intel_panel.h"
 #include "intel_psr.h"
 #include "intel_psr_regs.h"
-#include "intel_sprite.h"
 #include "intel_wm.h"
 
 static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node)
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.h b/drivers/gpu/drm/i915/display/intel_sprite.h
index 4635c7ad23f95..91c6dca342b2d 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.h
+++ b/drivers/gpu/drm/i915/display/intel_sprite.h
@@ -16,16 +16,6 @@ struct intel_crtc_state;
 struct intel_plane_state;
 enum pipe;
 
-/*
- * FIXME: We should instead only take spinlocks once for the entire update
- * instead of once per mmio.
- */
-#if IS_ENABLED(CONFIG_PROVE_LOCKING)
-#define VBLANK_EVASION_TIME_US 250
-#else
-#define VBLANK_EVASION_TIME_US 100
-#endif
-
 struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv,
 					      enum pipe pipe, int plane);
 int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data,
-- 
GitLab


From 8976b18249407df8bf6ea18ecae0640a15341a50 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:16 +0300
Subject: [PATCH 0182/1791] drm/i915: Relocate intel_atomic_setup_scalers()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move intel_atomic_setup_scalers() next to the other scaler
code in skl_scaler.c.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-4-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_atomic.c | 256 -------------------
 drivers/gpu/drm/i915/display/intel_atomic.h |   4 -
 drivers/gpu/drm/i915/display/skl_scaler.c   | 257 ++++++++++++++++++++
 drivers/gpu/drm/i915/display/skl_scaler.h   |  10 +-
 4 files changed, 265 insertions(+), 262 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c
index 61011641f6ab8..7cf51dd8c0567 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic.c
@@ -310,262 +310,6 @@ intel_crtc_destroy_state(struct drm_crtc *crtc,
 	kfree(crtc_state);
 }
 
-static int intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_state,
-				     int num_scalers_need, struct intel_crtc *intel_crtc,
-				     const char *name, int idx,
-				     struct intel_plane_state *plane_state,
-				     int *scaler_id)
-{
-	struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev);
-	int j;
-	u32 mode;
-
-	if (*scaler_id < 0) {
-		/* find a free scaler */
-		for (j = 0; j < intel_crtc->num_scalers; j++) {
-			if (scaler_state->scalers[j].in_use)
-				continue;
-
-			*scaler_id = j;
-			scaler_state->scalers[*scaler_id].in_use = 1;
-			break;
-		}
-	}
-
-	if (drm_WARN(&dev_priv->drm, *scaler_id < 0,
-		     "Cannot find scaler for %s:%d\n", name, idx))
-		return -EINVAL;
-
-	/* set scaler mode */
-	if (plane_state && plane_state->hw.fb &&
-	    plane_state->hw.fb->format->is_yuv &&
-	    plane_state->hw.fb->format->num_planes > 1) {
-		struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
-		if (DISPLAY_VER(dev_priv) == 9) {
-			mode = SKL_PS_SCALER_MODE_NV12;
-		} else if (icl_is_hdr_plane(dev_priv, plane->id)) {
-			/*
-			 * On gen11+'s HDR planes we only use the scaler for
-			 * scaling. They have a dedicated chroma upsampler, so
-			 * we don't need the scaler to upsample the UV plane.
-			 */
-			mode = PS_SCALER_MODE_NORMAL;
-		} else {
-			struct intel_plane *linked =
-				plane_state->planar_linked_plane;
-
-			mode = PS_SCALER_MODE_PLANAR;
-
-			if (linked)
-				mode |= PS_PLANE_Y_SEL(linked->id);
-		}
-	} else if (DISPLAY_VER(dev_priv) >= 10) {
-		mode = PS_SCALER_MODE_NORMAL;
-	} else if (num_scalers_need == 1 && intel_crtc->num_scalers > 1) {
-		/*
-		 * when only 1 scaler is in use on a pipe with 2 scalers
-		 * scaler 0 operates in high quality (HQ) mode.
-		 * In this case use scaler 0 to take advantage of HQ mode
-		 */
-		scaler_state->scalers[*scaler_id].in_use = 0;
-		*scaler_id = 0;
-		scaler_state->scalers[0].in_use = 1;
-		mode = SKL_PS_SCALER_MODE_HQ;
-	} else {
-		mode = SKL_PS_SCALER_MODE_DYN;
-	}
-
-	/*
-	 * FIXME: we should also check the scaler factors for pfit, so
-	 * this shouldn't be tied directly to planes.
-	 */
-	if (plane_state && plane_state->hw.fb) {
-		const struct drm_framebuffer *fb = plane_state->hw.fb;
-		const struct drm_rect *src = &plane_state->uapi.src;
-		const struct drm_rect *dst = &plane_state->uapi.dst;
-		int hscale, vscale, max_vscale, max_hscale;
-
-		/*
-		 * FIXME: When two scalers are needed, but only one of
-		 * them needs to downscale, we should make sure that
-		 * the one that needs downscaling support is assigned
-		 * as the first scaler, so we don't reject downscaling
-		 * unnecessarily.
-		 */
-
-		if (DISPLAY_VER(dev_priv) >= 14) {
-			/*
-			 * On versions 14 and up, only the first
-			 * scaler supports a vertical scaling factor
-			 * of more than 1.0, while a horizontal
-			 * scaling factor of 3.0 is supported.
-			 */
-			max_hscale = 0x30000 - 1;
-			if (*scaler_id == 0)
-				max_vscale = 0x30000 - 1;
-			else
-				max_vscale = 0x10000;
-
-		} else if (DISPLAY_VER(dev_priv) >= 10 ||
-			   !intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier)) {
-			max_hscale = 0x30000 - 1;
-			max_vscale = 0x30000 - 1;
-		} else {
-			max_hscale = 0x20000 - 1;
-			max_vscale = 0x20000 - 1;
-		}
-
-		/*
-		 * FIXME: We should change the if-else block above to
-		 * support HQ vs dynamic scaler properly.
-		 */
-
-		/* Check if required scaling is within limits */
-		hscale = drm_rect_calc_hscale(src, dst, 1, max_hscale);
-		vscale = drm_rect_calc_vscale(src, dst, 1, max_vscale);
-
-		if (hscale < 0 || vscale < 0) {
-			drm_dbg_kms(&dev_priv->drm,
-				    "Scaler %d doesn't support required plane scaling\n",
-				    *scaler_id);
-			drm_rect_debug_print("src: ", src, true);
-			drm_rect_debug_print("dst: ", dst, false);
-
-			return -EINVAL;
-		}
-	}
-
-	drm_dbg_kms(&dev_priv->drm, "Attached scaler id %u.%u to %s:%d\n",
-		    intel_crtc->pipe, *scaler_id, name, idx);
-	scaler_state->scalers[*scaler_id].mode = mode;
-
-	return 0;
-}
-
-/**
- * intel_atomic_setup_scalers() - setup scalers for crtc per staged requests
- * @dev_priv: i915 device
- * @intel_crtc: intel crtc
- * @crtc_state: incoming crtc_state to validate and setup scalers
- *
- * This function sets up scalers based on staged scaling requests for
- * a @crtc and its planes. It is called from crtc level check path. If request
- * is a supportable request, it attaches scalers to requested planes and crtc.
- *
- * This function takes into account the current scaler(s) in use by any planes
- * not being part of this atomic state
- *
- *  Returns:
- *         0 - scalers were setup succesfully
- *         error code - otherwise
- */
-int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
-			       struct intel_crtc *intel_crtc,
-			       struct intel_crtc_state *crtc_state)
-{
-	struct drm_plane *plane = NULL;
-	struct intel_plane *intel_plane;
-	struct intel_plane_state *plane_state = NULL;
-	struct intel_crtc_scaler_state *scaler_state =
-		&crtc_state->scaler_state;
-	struct drm_atomic_state *drm_state = crtc_state->uapi.state;
-	struct intel_atomic_state *intel_state = to_intel_atomic_state(drm_state);
-	int num_scalers_need;
-	int i;
-
-	num_scalers_need = hweight32(scaler_state->scaler_users);
-
-	/*
-	 * High level flow:
-	 * - staged scaler requests are already in scaler_state->scaler_users
-	 * - check whether staged scaling requests can be supported
-	 * - add planes using scalers that aren't in current transaction
-	 * - assign scalers to requested users
-	 * - as part of plane commit, scalers will be committed
-	 *   (i.e., either attached or detached) to respective planes in hw
-	 * - as part of crtc_commit, scaler will be either attached or detached
-	 *   to crtc in hw
-	 */
-
-	/* fail if required scalers > available scalers */
-	if (num_scalers_need > intel_crtc->num_scalers){
-		drm_dbg_kms(&dev_priv->drm,
-			    "Too many scaling requests %d > %d\n",
-			    num_scalers_need, intel_crtc->num_scalers);
-		return -EINVAL;
-	}
-
-	/* walkthrough scaler_users bits and start assigning scalers */
-	for (i = 0; i < sizeof(scaler_state->scaler_users) * 8; i++) {
-		int *scaler_id;
-		const char *name;
-		int idx, ret;
-
-		/* skip if scaler not required */
-		if (!(scaler_state->scaler_users & (1 << i)))
-			continue;
-
-		if (i == SKL_CRTC_INDEX) {
-			name = "CRTC";
-			idx = intel_crtc->base.base.id;
-
-			/* panel fitter case: assign as a crtc scaler */
-			scaler_id = &scaler_state->scaler_id;
-		} else {
-			name = "PLANE";
-
-			/* plane scaler case: assign as a plane scaler */
-			/* find the plane that set the bit as scaler_user */
-			plane = drm_state->planes[i].ptr;
-
-			/*
-			 * to enable/disable hq mode, add planes that are using scaler
-			 * into this transaction
-			 */
-			if (!plane) {
-				struct drm_plane_state *state;
-
-				/*
-				 * GLK+ scalers don't have a HQ mode so it
-				 * isn't necessary to change between HQ and dyn mode
-				 * on those platforms.
-				 */
-				if (DISPLAY_VER(dev_priv) >= 10)
-					continue;
-
-				plane = drm_plane_from_index(&dev_priv->drm, i);
-				state = drm_atomic_get_plane_state(drm_state, plane);
-				if (IS_ERR(state)) {
-					drm_dbg_kms(&dev_priv->drm,
-						    "Failed to add [PLANE:%d] to drm_state\n",
-						    plane->base.id);
-					return PTR_ERR(state);
-				}
-			}
-
-			intel_plane = to_intel_plane(plane);
-			idx = plane->base.id;
-
-			/* plane on different crtc cannot be a scaler user of this crtc */
-			if (drm_WARN_ON(&dev_priv->drm,
-					intel_plane->pipe != intel_crtc->pipe))
-				continue;
-
-			plane_state = intel_atomic_get_new_plane_state(intel_state,
-								       intel_plane);
-			scaler_id = &plane_state->scaler_id;
-		}
-
-		ret = intel_atomic_setup_scaler(scaler_state, num_scalers_need,
-						intel_crtc, name, idx,
-						plane_state, scaler_id);
-		if (ret < 0)
-			return ret;
-	}
-
-	return 0;
-}
-
 struct drm_atomic_state *
 intel_atomic_state_alloc(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_atomic.h b/drivers/gpu/drm/i915/display/intel_atomic.h
index 1dc439983dd94..e506f6a873447 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic.h
+++ b/drivers/gpu/drm/i915/display/intel_atomic.h
@@ -52,8 +52,4 @@ struct intel_crtc_state *
 intel_atomic_get_crtc_state(struct drm_atomic_state *state,
 			    struct intel_crtc *crtc);
 
-int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
-			       struct intel_crtc *intel_crtc,
-			       struct intel_crtc_state *crtc_state);
-
 #endif /* __INTEL_ATOMIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 0e7e014fcc717..62443834f64e5 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -348,6 +348,263 @@ int skl_update_scaler_plane(struct intel_crtc_state *crtc_state,
 	return 0;
 }
 
+static int intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_state,
+				     int num_scalers_need, struct intel_crtc *intel_crtc,
+				     const char *name, int idx,
+				     struct intel_plane_state *plane_state,
+				     int *scaler_id)
+{
+	struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev);
+	int j;
+	u32 mode;
+
+	if (*scaler_id < 0) {
+		/* find a free scaler */
+		for (j = 0; j < intel_crtc->num_scalers; j++) {
+			if (scaler_state->scalers[j].in_use)
+				continue;
+
+			*scaler_id = j;
+			scaler_state->scalers[*scaler_id].in_use = 1;
+			break;
+		}
+	}
+
+	if (drm_WARN(&dev_priv->drm, *scaler_id < 0,
+		     "Cannot find scaler for %s:%d\n", name, idx))
+		return -EINVAL;
+
+	/* set scaler mode */
+	if (plane_state && plane_state->hw.fb &&
+	    plane_state->hw.fb->format->is_yuv &&
+	    plane_state->hw.fb->format->num_planes > 1) {
+		struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
+
+		if (DISPLAY_VER(dev_priv) == 9) {
+			mode = SKL_PS_SCALER_MODE_NV12;
+		} else if (icl_is_hdr_plane(dev_priv, plane->id)) {
+			/*
+			 * On gen11+'s HDR planes we only use the scaler for
+			 * scaling. They have a dedicated chroma upsampler, so
+			 * we don't need the scaler to upsample the UV plane.
+			 */
+			mode = PS_SCALER_MODE_NORMAL;
+		} else {
+			struct intel_plane *linked =
+				plane_state->planar_linked_plane;
+
+			mode = PS_SCALER_MODE_PLANAR;
+
+			if (linked)
+				mode |= PS_PLANE_Y_SEL(linked->id);
+		}
+	} else if (DISPLAY_VER(dev_priv) >= 10) {
+		mode = PS_SCALER_MODE_NORMAL;
+	} else if (num_scalers_need == 1 && intel_crtc->num_scalers > 1) {
+		/*
+		 * when only 1 scaler is in use on a pipe with 2 scalers
+		 * scaler 0 operates in high quality (HQ) mode.
+		 * In this case use scaler 0 to take advantage of HQ mode
+		 */
+		scaler_state->scalers[*scaler_id].in_use = 0;
+		*scaler_id = 0;
+		scaler_state->scalers[0].in_use = 1;
+		mode = SKL_PS_SCALER_MODE_HQ;
+	} else {
+		mode = SKL_PS_SCALER_MODE_DYN;
+	}
+
+	/*
+	 * FIXME: we should also check the scaler factors for pfit, so
+	 * this shouldn't be tied directly to planes.
+	 */
+	if (plane_state && plane_state->hw.fb) {
+		const struct drm_framebuffer *fb = plane_state->hw.fb;
+		const struct drm_rect *src = &plane_state->uapi.src;
+		const struct drm_rect *dst = &plane_state->uapi.dst;
+		int hscale, vscale, max_vscale, max_hscale;
+
+		/*
+		 * FIXME: When two scalers are needed, but only one of
+		 * them needs to downscale, we should make sure that
+		 * the one that needs downscaling support is assigned
+		 * as the first scaler, so we don't reject downscaling
+		 * unnecessarily.
+		 */
+
+		if (DISPLAY_VER(dev_priv) >= 14) {
+			/*
+			 * On versions 14 and up, only the first
+			 * scaler supports a vertical scaling factor
+			 * of more than 1.0, while a horizontal
+			 * scaling factor of 3.0 is supported.
+			 */
+			max_hscale = 0x30000 - 1;
+			if (*scaler_id == 0)
+				max_vscale = 0x30000 - 1;
+			else
+				max_vscale = 0x10000;
+
+		} else if (DISPLAY_VER(dev_priv) >= 10 ||
+			   !intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier)) {
+			max_hscale = 0x30000 - 1;
+			max_vscale = 0x30000 - 1;
+		} else {
+			max_hscale = 0x20000 - 1;
+			max_vscale = 0x20000 - 1;
+		}
+
+		/*
+		 * FIXME: We should change the if-else block above to
+		 * support HQ vs dynamic scaler properly.
+		 */
+
+		/* Check if required scaling is within limits */
+		hscale = drm_rect_calc_hscale(src, dst, 1, max_hscale);
+		vscale = drm_rect_calc_vscale(src, dst, 1, max_vscale);
+
+		if (hscale < 0 || vscale < 0) {
+			drm_dbg_kms(&dev_priv->drm,
+				    "Scaler %d doesn't support required plane scaling\n",
+				    *scaler_id);
+			drm_rect_debug_print("src: ", src, true);
+			drm_rect_debug_print("dst: ", dst, false);
+
+			return -EINVAL;
+		}
+	}
+
+	drm_dbg_kms(&dev_priv->drm, "Attached scaler id %u.%u to %s:%d\n",
+		    intel_crtc->pipe, *scaler_id, name, idx);
+	scaler_state->scalers[*scaler_id].mode = mode;
+
+	return 0;
+}
+
+/**
+ * intel_atomic_setup_scalers() - setup scalers for crtc per staged requests
+ * @dev_priv: i915 device
+ * @intel_crtc: intel crtc
+ * @crtc_state: incoming crtc_state to validate and setup scalers
+ *
+ * This function sets up scalers based on staged scaling requests for
+ * a @crtc and its planes. It is called from crtc level check path. If request
+ * is a supportable request, it attaches scalers to requested planes and crtc.
+ *
+ * This function takes into account the current scaler(s) in use by any planes
+ * not being part of this atomic state
+ *
+ *  Returns:
+ *         0 - scalers were setup successfully
+ *         error code - otherwise
+ */
+int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
+			       struct intel_crtc *intel_crtc,
+			       struct intel_crtc_state *crtc_state)
+{
+	struct drm_plane *plane = NULL;
+	struct intel_plane *intel_plane;
+	struct intel_plane_state *plane_state = NULL;
+	struct intel_crtc_scaler_state *scaler_state =
+		&crtc_state->scaler_state;
+	struct drm_atomic_state *drm_state = crtc_state->uapi.state;
+	struct intel_atomic_state *intel_state = to_intel_atomic_state(drm_state);
+	int num_scalers_need;
+	int i;
+
+	num_scalers_need = hweight32(scaler_state->scaler_users);
+
+	/*
+	 * High level flow:
+	 * - staged scaler requests are already in scaler_state->scaler_users
+	 * - check whether staged scaling requests can be supported
+	 * - add planes using scalers that aren't in current transaction
+	 * - assign scalers to requested users
+	 * - as part of plane commit, scalers will be committed
+	 *   (i.e., either attached or detached) to respective planes in hw
+	 * - as part of crtc_commit, scaler will be either attached or detached
+	 *   to crtc in hw
+	 */
+
+	/* fail if required scalers > available scalers */
+	if (num_scalers_need > intel_crtc->num_scalers) {
+		drm_dbg_kms(&dev_priv->drm,
+			    "Too many scaling requests %d > %d\n",
+			    num_scalers_need, intel_crtc->num_scalers);
+		return -EINVAL;
+	}
+
+	/* walkthrough scaler_users bits and start assigning scalers */
+	for (i = 0; i < sizeof(scaler_state->scaler_users) * 8; i++) {
+		int *scaler_id;
+		const char *name;
+		int idx, ret;
+
+		/* skip if scaler not required */
+		if (!(scaler_state->scaler_users & (1 << i)))
+			continue;
+
+		if (i == SKL_CRTC_INDEX) {
+			name = "CRTC";
+			idx = intel_crtc->base.base.id;
+
+			/* panel fitter case: assign as a crtc scaler */
+			scaler_id = &scaler_state->scaler_id;
+		} else {
+			name = "PLANE";
+
+			/* plane scaler case: assign as a plane scaler */
+			/* find the plane that set the bit as scaler_user */
+			plane = drm_state->planes[i].ptr;
+
+			/*
+			 * to enable/disable hq mode, add planes that are using scaler
+			 * into this transaction
+			 */
+			if (!plane) {
+				struct drm_plane_state *state;
+
+				/*
+				 * GLK+ scalers don't have a HQ mode so it
+				 * isn't necessary to change between HQ and dyn mode
+				 * on those platforms.
+				 */
+				if (DISPLAY_VER(dev_priv) >= 10)
+					continue;
+
+				plane = drm_plane_from_index(&dev_priv->drm, i);
+				state = drm_atomic_get_plane_state(drm_state, plane);
+				if (IS_ERR(state)) {
+					drm_dbg_kms(&dev_priv->drm,
+						    "Failed to add [PLANE:%d] to drm_state\n",
+						    plane->base.id);
+					return PTR_ERR(state);
+				}
+			}
+
+			intel_plane = to_intel_plane(plane);
+			idx = plane->base.id;
+
+			/* plane on different crtc cannot be a scaler user of this crtc */
+			if (drm_WARN_ON(&dev_priv->drm,
+					intel_plane->pipe != intel_crtc->pipe))
+				continue;
+
+			plane_state = intel_atomic_get_new_plane_state(intel_state,
+								       intel_plane);
+			scaler_id = &plane_state->scaler_id;
+		}
+
+		ret = intel_atomic_setup_scaler(scaler_state, num_scalers_need,
+						intel_crtc, name, idx,
+						plane_state, scaler_id);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
 static int glk_coef_tap(int i)
 {
 	return i % 7;
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h
index 0097d5d08e102..f040f6ac061f2 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.h
+++ b/drivers/gpu/drm/i915/display/skl_scaler.h
@@ -8,17 +8,22 @@
 #include <linux/types.h>
 
 enum drm_scaling_filter;
+enum pipe;
 struct drm_i915_private;
+struct intel_crtc;
 struct intel_crtc_state;
-struct intel_plane_state;
 struct intel_plane;
-enum pipe;
+struct intel_plane_state;
 
 int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state);
 
 int skl_update_scaler_plane(struct intel_crtc_state *crtc_state,
 			    struct intel_plane_state *plane_state);
 
+int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv,
+			       struct intel_crtc *intel_crtc,
+			       struct intel_crtc_state *crtc_state);
+
 void skl_pfit_enable(const struct intel_crtc_state *crtc_state);
 
 void skl_program_plane_scaler(struct intel_plane *plane,
@@ -26,4 +31,5 @@ void skl_program_plane_scaler(struct intel_plane *plane,
 			      const struct intel_plane_state *plane_state);
 void skl_detach_scalers(const struct intel_crtc_state *crtc_state);
 void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state);
+
 #endif
-- 
GitLab


From 597421a806c7485b91ee4cf7ec2ce3618223dc92 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:17 +0300
Subject: [PATCH 0183/1791] drm/i915: Relocate skl_get_pfit_config()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move skl_get_pfit_config() next to the other skl+ scaler code
and rename it to skl_scaler_get_config() so that it has a consistnet
namespace.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-5-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 49 ++------------------
 drivers/gpu/drm/i915/display/skl_scaler.c    | 37 +++++++++++++++
 drivers/gpu/drm/i915/display/skl_scaler.h    |  2 +
 3 files changed, 43 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 2d4a6936ab2bf..9c017e9bae266 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3224,49 +3224,6 @@ void intel_cpu_transcoder_get_m2_n2(struct intel_crtc *crtc,
 		      PIPE_LINK_M2(transcoder), PIPE_LINK_N2(transcoder));
 }
 
-static void ilk_get_pfit_pos_size(struct intel_crtc_state *crtc_state,
-				  u32 pos, u32 size)
-{
-	drm_rect_init(&crtc_state->pch_pfit.dst,
-		      pos >> 16, pos & 0xffff,
-		      size >> 16, size & 0xffff);
-}
-
-static void skl_get_pfit_config(struct intel_crtc_state *crtc_state)
-{
-	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	struct intel_crtc_scaler_state *scaler_state = &crtc_state->scaler_state;
-	int id = -1;
-	int i;
-
-	/* find scaler attached to this pipe */
-	for (i = 0; i < crtc->num_scalers; i++) {
-		u32 ctl, pos, size;
-
-		ctl = intel_de_read(dev_priv, SKL_PS_CTRL(crtc->pipe, i));
-		if ((ctl & (PS_SCALER_EN | PS_PLANE_SEL_MASK)) != PS_SCALER_EN)
-			continue;
-
-		id = i;
-		crtc_state->pch_pfit.enabled = true;
-
-		pos = intel_de_read(dev_priv, SKL_PS_WIN_POS(crtc->pipe, i));
-		size = intel_de_read(dev_priv, SKL_PS_WIN_SZ(crtc->pipe, i));
-
-		ilk_get_pfit_pos_size(crtc_state, pos, size);
-
-		scaler_state->scalers[i].in_use = true;
-		break;
-	}
-
-	scaler_state->scaler_id = id;
-	if (id >= 0)
-		scaler_state->scaler_users |= (1 << SKL_CRTC_INDEX);
-	else
-		scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX);
-}
-
 static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
@@ -3282,7 +3239,9 @@ static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state)
 	pos = intel_de_read(dev_priv, PF_WIN_POS(crtc->pipe));
 	size = intel_de_read(dev_priv, PF_WIN_SZ(crtc->pipe));
 
-	ilk_get_pfit_pos_size(crtc_state, pos, size);
+	drm_rect_init(&crtc_state->pch_pfit.dst,
+		      pos >> 16, pos & 0xffff,
+		      size >> 16, size & 0xffff);
 
 	/*
 	 * We currently do not free assignements of panel fitters on
@@ -3773,7 +3732,7 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc,
 	if (intel_display_power_get_in_set_if_enabled(dev_priv, &crtc->hw_readout_power_domains,
 						      POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe))) {
 		if (DISPLAY_VER(dev_priv) >= 9)
-			skl_get_pfit_config(pipe_config);
+			skl_scaler_get_config(pipe_config);
 		else
 			ilk_get_pfit_config(pipe_config);
 	}
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 62443834f64e5..ec930aec21c4c 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -856,3 +856,40 @@ void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state)
 	for (i = 0; i < crtc->num_scalers; i++)
 		skl_detach_scaler(crtc, i);
 }
+
+void skl_scaler_get_config(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	struct intel_crtc_scaler_state *scaler_state = &crtc_state->scaler_state;
+	int id = -1;
+	int i;
+
+	/* find scaler attached to this pipe */
+	for (i = 0; i < crtc->num_scalers; i++) {
+		u32 ctl, pos, size;
+
+		ctl = intel_de_read(dev_priv, SKL_PS_CTRL(crtc->pipe, i));
+		if ((ctl & (PS_SCALER_EN | PS_PLANE_SEL_MASK)) != PS_SCALER_EN)
+			continue;
+
+		id = i;
+		crtc_state->pch_pfit.enabled = true;
+
+		pos = intel_de_read(dev_priv, SKL_PS_WIN_POS(crtc->pipe, i));
+		size = intel_de_read(dev_priv, SKL_PS_WIN_SZ(crtc->pipe, i));
+
+		drm_rect_init(&crtc_state->pch_pfit.dst,
+			      pos >> 16, pos & 0xffff,
+			      size >> 16, size & 0xffff);
+
+		scaler_state->scalers[i].in_use = true;
+		break;
+	}
+
+	scaler_state->scaler_id = id;
+	if (id >= 0)
+		scaler_state->scaler_users |= (1 << SKL_CRTC_INDEX);
+	else
+		scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX);
+}
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h
index f040f6ac061f2..63f93ca03c89e 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.h
+++ b/drivers/gpu/drm/i915/display/skl_scaler.h
@@ -32,4 +32,6 @@ void skl_program_plane_scaler(struct intel_plane *plane,
 void skl_detach_scalers(const struct intel_crtc_state *crtc_state);
 void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state);
 
+void skl_scaler_get_config(struct intel_crtc_state *crtc_state);
+
 #endif
-- 
GitLab


From 08df6d30c1022e738dd26e732a36f3f7642594d2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:18 +0300
Subject: [PATCH 0184/1791] drm/i915: Use REG_BIT() & co for the pre-ilk pfit
 registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Modernize the gmch pfit register definitions using REG_BIT/etc.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-6-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 15 +++---
 drivers/gpu/drm/i915/display/intel_overlay.c | 16 +++---
 drivers/gpu/drm/i915/display/intel_panel.c   |  8 +--
 drivers/gpu/drm/i915/i915_reg.h              | 54 ++++++++++----------
 4 files changed, 48 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 9c017e9bae266..11bf7f7b10a5b 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2749,6 +2749,7 @@ static void i9xx_get_pfit_config(struct intel_crtc_state *crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe;
 	u32 tmp;
 
 	if (!i9xx_has_pfit(dev_priv))
@@ -2759,13 +2760,13 @@ static void i9xx_get_pfit_config(struct intel_crtc_state *crtc_state)
 		return;
 
 	/* Check whether the pfit is attached to our pipe. */
-	if (DISPLAY_VER(dev_priv) < 4) {
-		if (crtc->pipe != PIPE_B)
-			return;
-	} else {
-		if ((tmp & PFIT_PIPE_MASK) != (crtc->pipe << PFIT_PIPE_SHIFT))
-			return;
-	}
+	if (DISPLAY_VER(dev_priv) >= 4)
+		pipe = REG_FIELD_GET(PFIT_PIPE_MASK, tmp);
+	else
+		pipe = PIPE_B;
+
+	if (pipe != crtc->pipe)
+		return;
 
 	crtc_state->gmch_pfit.control = tmp;
 	crtc_state->gmch_pfit.pgm_ratios =
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index c12bdca8da9ba..1813ab5056a1b 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -935,21 +935,25 @@ static int check_overlay_possible_on_crtc(struct intel_overlay *overlay,
 static void update_pfit_vscale_ratio(struct intel_overlay *overlay)
 {
 	struct drm_i915_private *dev_priv = overlay->i915;
-	u32 pfit_control = intel_de_read(dev_priv, PFIT_CONTROL);
 	u32 ratio;
 
 	/* XXX: This is not the same logic as in the xorg driver, but more in
 	 * line with the intel documentation for the i965
 	 */
 	if (DISPLAY_VER(dev_priv) >= 4) {
+		u32 tmp = intel_de_read(dev_priv, PFIT_PGM_RATIOS);
+
 		/* on i965 use the PGM reg to read out the autoscaler values */
-		ratio = intel_de_read(dev_priv, PFIT_PGM_RATIOS) >> PFIT_VERT_SCALE_SHIFT_965;
+		ratio = REG_FIELD_GET(PFIT_VERT_SCALE_MASK_965, tmp);
 	} else {
-		if (pfit_control & VERT_AUTO_SCALE)
-			ratio = intel_de_read(dev_priv, PFIT_AUTO_RATIOS);
+		u32 tmp;
+
+		if (intel_de_read(dev_priv, PFIT_CONTROL) & VERT_AUTO_SCALE)
+			tmp = intel_de_read(dev_priv, PFIT_AUTO_RATIOS);
 		else
-			ratio = intel_de_read(dev_priv, PFIT_PGM_RATIOS);
-		ratio >>= PFIT_VERT_SCALE_SHIFT;
+			tmp = intel_de_read(dev_priv, PFIT_PGM_RATIOS);
+
+		ratio = REG_FIELD_GET(PFIT_VERT_SCALE_MASK, tmp);
 	}
 
 	overlay->pfit_vscale_ratio = ratio;
diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c
index 9acdd68b2dbcf..71cd08f44ed02 100644
--- a/drivers/gpu/drm/i915/display/intel_panel.c
+++ b/drivers/gpu/drm/i915/display/intel_panel.c
@@ -564,8 +564,8 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state,
 			bits = panel_fitter_scaling(pipe_src_h,
 						    adjusted_mode->crtc_vdisplay);
 
-			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
-					     bits << PFIT_VERT_SCALE_SHIFT);
+			*pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) |
+					     PFIT_VERT_SCALE(bits));
 			*pfit_control |= (PFIT_ENABLE |
 					  VERT_INTERP_BILINEAR |
 					  HORIZ_INTERP_BILINEAR);
@@ -579,8 +579,8 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state,
 			bits = panel_fitter_scaling(pipe_src_w,
 						    adjusted_mode->crtc_hdisplay);
 
-			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
-					     bits << PFIT_VERT_SCALE_SHIFT);
+			*pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) |
+					     PFIT_VERT_SCALE(bits));
 			*pfit_control |= (PFIT_ENABLE |
 					  VERT_INTERP_BILINEAR |
 					  HORIZ_INTERP_BILINEAR);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 024a92f6bdba7..f1e0fa8c17acd 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2327,35 +2327,33 @@
 
 /* Panel fitting */
 #define PFIT_CONTROL	_MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61230)
-#define   PFIT_ENABLE		(1 << 31)
-#define   PFIT_PIPE_MASK	(3 << 29)
-#define   PFIT_PIPE_SHIFT	29
-#define   PFIT_PIPE(pipe)	((pipe) << 29)
-#define   VERT_INTERP_DISABLE	(0 << 10)
-#define   VERT_INTERP_BILINEAR	(1 << 10)
-#define   VERT_INTERP_MASK	(3 << 10)
-#define   VERT_AUTO_SCALE	(1 << 9)
-#define   HORIZ_INTERP_DISABLE	(0 << 6)
-#define   HORIZ_INTERP_BILINEAR	(1 << 6)
-#define   HORIZ_INTERP_MASK	(3 << 6)
-#define   HORIZ_AUTO_SCALE	(1 << 5)
-#define   PANEL_8TO6_DITHER_ENABLE (1 << 3)
-#define   PFIT_FILTER_FUZZY	(0 << 24)
-#define   PFIT_SCALING_AUTO	(0 << 26)
-#define   PFIT_SCALING_PROGRAMMED (1 << 26)
-#define   PFIT_SCALING_PILLAR	(2 << 26)
-#define   PFIT_SCALING_LETTER	(3 << 26)
+#define   PFIT_ENABLE			REG_BIT(31)
+#define   PFIT_PIPE_MASK		REG_GENMASK(30, 29) /* 965+ */
+#define   PFIT_PIPE(pipe)		REG_FIELD_PREP(PFIT_PIPE_MASK, (pipe))
+#define   PFIT_SCALING_MASK		REG_GENMASK(28, 26) /* 965+ */
+#define   PFIT_SCALING_AUTO		REG_FIELD_PREP(PFIT_SCALING_MASK, 0)
+#define   PFIT_SCALING_PROGRAMMED	REG_FIELD_PREP(PFIT_SCALING_MASK, 1)
+#define   PFIT_SCALING_PILLAR		REG_FIELD_PREP(PFIT_SCALING_MASK, 2)
+#define   PFIT_SCALING_LETTER		REG_FIELD_PREP(PFIT_SCALING_MASK, 3)
+#define   PFIT_FILTER_MASK		REG_GENMASK(25, 24) /* 965+ */
+#define   PFIT_FILTER_FUZZY		REG_FIELD_PREP(PFIT_FILTER_MASK, 0)
+#define   PFIT_FILTER_CRISP		REG_FIELD_PREP(PFIT_FILTER_MASK, 1)
+#define   PFIT_FILTER_MEDIAN		REG_FIELD_PREP(PFIT_FILTER_MASK, 2)
+#define   VERT_INTERP_MASK		REG_GENMASK(11, 10) /* pre-965 */
+#define   VERT_INTERP_BILINEAR		REG_FIELD_PREP(VERT_INTERP_MASK, 1)
+#define   VERT_AUTO_SCALE		REG_BIT(9) /* pre-965 */
+#define   HORIZ_INTERP_MASK		REG_GENMASK(7, 6) /* pre-965 */
+#define   HORIZ_INTERP_BILINEAR		REG_FIELD_PREP(HORIZ_INTERP_MASK, 1)
+#define   HORIZ_AUTO_SCALE		REG_BIT(5) /* pre-965 */
+#define   PANEL_8TO6_DITHER_ENABLE	REG_BIT(3) /* pre-965 */
+
 #define PFIT_PGM_RATIOS _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61234)
-/* Pre-965 */
-#define		PFIT_VERT_SCALE_SHIFT		20
-#define		PFIT_VERT_SCALE_MASK		0xfff00000
-#define		PFIT_HORIZ_SCALE_SHIFT		4
-#define		PFIT_HORIZ_SCALE_MASK		0x0000fff0
-/* 965+ */
-#define		PFIT_VERT_SCALE_SHIFT_965	16
-#define		PFIT_VERT_SCALE_MASK_965	0x1fff0000
-#define		PFIT_HORIZ_SCALE_SHIFT_965	0
-#define		PFIT_HORIZ_SCALE_MASK_965	0x00001fff
+#define   PFIT_VERT_SCALE_MASK		REG_GENMASK(31, 20) /* pre-965 */
+#define   PFIT_VERT_SCALE(x)		REG_FIELD_PREP(PFIT_VERT_SCALE_MASK, (x))
+#define   PFIT_HORIZ_SCALE_MASK		REG_GENMASK(15, 4) /* pre-965 */
+#define   PFIT_HORIZ_SCALE(x)		REG_FIELD_PREP(PFIT_HORIZ_SCALE_MASK, (x))
+#define   PFIT_VERT_SCALE_MASK_965	REG_GENMASK(28, 16) /* 965+ */
+#define   PFIT_HORIZ_SCALE_MASK_965	REG_GENMASK(12, 0) /* 965+ */
 
 #define PFIT_AUTO_RATIOS _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61238)
 
-- 
GitLab


From e27525cc805548eaaa9d0cb8e8f0f181e9cd5390 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:19 +0300
Subject: [PATCH 0185/1791] drm/i915: Namespace pfit registers properly
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Give the PFIT_CONTROL bits a consistent namespace.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-7-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_lvds.c    |  2 +-
 drivers/gpu/drm/i915/display/intel_overlay.c |  2 +-
 drivers/gpu/drm/i915/display/intel_panel.c   | 25 ++++++++++----------
 drivers/gpu/drm/i915/i915_reg.h              | 14 +++++------
 4 files changed, 22 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 0de44b3631cde..8e9a3d72b83b7 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -150,7 +150,7 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	if (DISPLAY_VER(dev_priv) < 4) {
 		tmp = intel_de_read(dev_priv, PFIT_CONTROL);
 
-		crtc_state->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE;
+		crtc_state->gmch_pfit.control |= tmp & PFIT_PANEL_8TO6_DITHER_ENABLE;
 	}
 
 	crtc_state->hw.adjusted_mode.crtc_clock = crtc_state->port_clock;
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index 1813ab5056a1b..d6fe2bbabe559 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -948,7 +948,7 @@ static void update_pfit_vscale_ratio(struct intel_overlay *overlay)
 	} else {
 		u32 tmp;
 
-		if (intel_de_read(dev_priv, PFIT_CONTROL) & VERT_AUTO_SCALE)
+		if (intel_de_read(dev_priv, PFIT_CONTROL) & PFIT_VERT_AUTO_SCALE)
 			tmp = intel_de_read(dev_priv, PFIT_AUTO_RATIOS);
 		else
 			tmp = intel_de_read(dev_priv, PFIT_PGM_RATIOS);
diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c
index 71cd08f44ed02..9232a305b1e6f 100644
--- a/drivers/gpu/drm/i915/display/intel_panel.c
+++ b/drivers/gpu/drm/i915/display/intel_panel.c
@@ -567,8 +567,8 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state,
 			*pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) |
 					     PFIT_VERT_SCALE(bits));
 			*pfit_control |= (PFIT_ENABLE |
-					  VERT_INTERP_BILINEAR |
-					  HORIZ_INTERP_BILINEAR);
+					  PFIT_VERT_INTERP_BILINEAR |
+					  PFIT_HORIZ_INTERP_BILINEAR);
 		}
 	} else if (scaled_width < scaled_height) { /* letter */
 		centre_vertically(adjusted_mode,
@@ -582,15 +582,16 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state,
 			*pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) |
 					     PFIT_VERT_SCALE(bits));
 			*pfit_control |= (PFIT_ENABLE |
-					  VERT_INTERP_BILINEAR |
-					  HORIZ_INTERP_BILINEAR);
+					  PFIT_VERT_INTERP_BILINEAR |
+					  PFIT_HORIZ_INTERP_BILINEAR);
 		}
 	} else {
 		/* Aspects match, Let hw scale both directions */
 		*pfit_control |= (PFIT_ENABLE |
-				  VERT_AUTO_SCALE | HORIZ_AUTO_SCALE |
-				  VERT_INTERP_BILINEAR |
-				  HORIZ_INTERP_BILINEAR);
+				  PFIT_VERT_AUTO_SCALE |
+				  PFIT_HORIZ_AUTO_SCALE |
+				  PFIT_VERT_INTERP_BILINEAR |
+				  PFIT_HORIZ_INTERP_BILINEAR);
 	}
 }
 
@@ -638,10 +639,10 @@ static int gmch_panel_fitting(struct intel_crtc_state *crtc_state,
 			if (DISPLAY_VER(dev_priv) >= 4)
 				pfit_control |= PFIT_SCALING_AUTO;
 			else
-				pfit_control |= (VERT_AUTO_SCALE |
-						 VERT_INTERP_BILINEAR |
-						 HORIZ_AUTO_SCALE |
-						 HORIZ_INTERP_BILINEAR);
+				pfit_control |= (PFIT_VERT_AUTO_SCALE |
+						 PFIT_VERT_INTERP_BILINEAR |
+						 PFIT_HORIZ_AUTO_SCALE |
+						 PFIT_HORIZ_INTERP_BILINEAR);
 		}
 		break;
 	default:
@@ -662,7 +663,7 @@ static int gmch_panel_fitting(struct intel_crtc_state *crtc_state,
 
 	/* Make sure pre-965 set dither correctly for 18bpp panels. */
 	if (DISPLAY_VER(dev_priv) < 4 && crtc_state->pipe_bpp == 18)
-		pfit_control |= PANEL_8TO6_DITHER_ENABLE;
+		pfit_control |= PFIT_PANEL_8TO6_DITHER_ENABLE;
 
 	crtc_state->gmch_pfit.control = pfit_control;
 	crtc_state->gmch_pfit.pgm_ratios = pfit_pgm_ratios;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f1e0fa8c17acd..639a150110c2d 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2339,13 +2339,13 @@
 #define   PFIT_FILTER_FUZZY		REG_FIELD_PREP(PFIT_FILTER_MASK, 0)
 #define   PFIT_FILTER_CRISP		REG_FIELD_PREP(PFIT_FILTER_MASK, 1)
 #define   PFIT_FILTER_MEDIAN		REG_FIELD_PREP(PFIT_FILTER_MASK, 2)
-#define   VERT_INTERP_MASK		REG_GENMASK(11, 10) /* pre-965 */
-#define   VERT_INTERP_BILINEAR		REG_FIELD_PREP(VERT_INTERP_MASK, 1)
-#define   VERT_AUTO_SCALE		REG_BIT(9) /* pre-965 */
-#define   HORIZ_INTERP_MASK		REG_GENMASK(7, 6) /* pre-965 */
-#define   HORIZ_INTERP_BILINEAR		REG_FIELD_PREP(HORIZ_INTERP_MASK, 1)
-#define   HORIZ_AUTO_SCALE		REG_BIT(5) /* pre-965 */
-#define   PANEL_8TO6_DITHER_ENABLE	REG_BIT(3) /* pre-965 */
+#define   PFIT_VERT_INTERP_MASK		REG_GENMASK(11, 10) /* pre-965 */
+#define   PFIT_VERT_INTERP_BILINEAR	REG_FIELD_PREP(PFIT_VERT_INTERP_MASK, 1)
+#define   PFIT_VERT_AUTO_SCALE		REG_BIT(9) /* pre-965 */
+#define   PFIT_HORIZ_INTERP_MASK	REG_GENMASK(7, 6) /* pre-965 */
+#define   PFIT_HORIZ_INTERP_BILINEAR	REG_FIELD_PREP(PFIT_HORIZ_INTERP_MASK, 1)
+#define   PFIT_HORIZ_AUTO_SCALE		REG_BIT(5) /* pre-965 */
+#define   PFIT_PANEL_8TO6_DITHER_ENABLE	REG_BIT(3) /* pre-965 */
 
 #define PFIT_PGM_RATIOS _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61234)
 #define   PFIT_VERT_SCALE_MASK		REG_GENMASK(31, 20) /* pre-965 */
-- 
GitLab


From 75d020db38b8ec325fe5a28336646e93eac3b033 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:20 +0300
Subject: [PATCH 0186/1791] drm/i915: Use REG_BIT() & co. for ilk+ pfit
 registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Polish the ilk+ pfit registers with REG_BIT() & co., and
also take the opportunity to unify the ivb/hsw vs. not checks
in ilk_pfit_enable() and ilk_get_pfit_config().

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-8-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c |  9 +++++++--
 drivers/gpu/drm/i915/i915_reg.h              | 20 ++++++++++----------
 2 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 11bf7f7b10a5b..4aef71640aba0 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3230,11 +3230,17 @@ static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state)
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	u32 ctl, pos, size;
+	enum pipe pipe;
 
 	ctl = intel_de_read(dev_priv, PF_CTL(crtc->pipe));
 	if ((ctl & PF_ENABLE) == 0)
 		return;
 
+	if (IS_IVYBRIDGE(dev_priv) || IS_HASWELL(dev_priv))
+		pipe = REG_FIELD_GET(PF_PIPE_SEL_MASK_IVB, ctl);
+	else
+		pipe = crtc->pipe;
+
 	crtc_state->pch_pfit.enabled = true;
 
 	pos = intel_de_read(dev_priv, PF_WIN_POS(crtc->pipe));
@@ -3249,8 +3255,7 @@ static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state)
 	 * ivb/hsw (since we don't use the higher upscaling modes which
 	 * differentiates them) so just WARN about this case for now.
 	 */
-	drm_WARN_ON(&dev_priv->drm, DISPLAY_VER(dev_priv) == 7 &&
-		    (ctl & PF_PIPE_SEL_MASK_IVB) != PF_PIPE_SEL_IVB(crtc->pipe));
+	drm_WARN_ON(&dev_priv->drm, pipe != crtc->pipe);
 }
 
 static bool ilk_get_pipe_config(struct intel_crtc *crtc,
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 639a150110c2d..a89e7a1cdcfeb 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3998,16 +3998,16 @@
 
 /* CPU panel fitter */
 /* IVB+ has 3 fitters, 0 is 7x5 capable, the other two only 3x3 */
-#define _PFA_CTL_1               0x68080
-#define _PFB_CTL_1               0x68880
-#define  PF_ENABLE              (1 << 31)
-#define  PF_PIPE_SEL_MASK_IVB	(3 << 29)
-#define  PF_PIPE_SEL_IVB(pipe)	((pipe) << 29)
-#define  PF_FILTER_MASK		(3 << 23)
-#define  PF_FILTER_PROGRAMMED	(0 << 23)
-#define  PF_FILTER_MED_3x3	(1 << 23)
-#define  PF_FILTER_EDGE_ENHANCE	(2 << 23)
-#define  PF_FILTER_EDGE_SOFTEN	(3 << 23)
+#define _PFA_CTL_1		0x68080
+#define _PFB_CTL_1		0x68880
+#define   PF_ENABLE			REG_BIT(31)
+#define   PF_PIPE_SEL_MASK_IVB		REG_GENMASK(30, 29) /* ivb/hsw */
+#define   PF_PIPE_SEL_IVB(pipe)		REG_FIELD_PREP(PF_PIPE_SEL_MASK_IVB, (pipe))
+#define   PF_FILTER_MASK		REG_GENMASK(24, 23)
+#define   PF_FILTER_PROGRAMMED		REG_FIELD_PREP(PF_FILTER_MASK, 0)
+#define   PF_FILTER_MED_3x3		REG_FIELD_PREP(PF_FILTER_MASK, 1)
+#define   PF_FILTER_EDGE_ENHANCE	REG_FIELD_PREP(PF_FILTER_EDGE_MASK, 2)
+#define   PF_FILTER_EDGE_SOFTEN		REG_FIELD_PREP(PF_FILTER_EDGE_MASK, 3)
 #define _PFA_WIN_SZ		0x68074
 #define _PFB_WIN_SZ		0x68874
 #define _PFA_WIN_POS		0x68070
-- 
GitLab


From 12de65d9c6e3b7b0b05ec03cc19669268ee1318f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 18 Apr 2023 20:55:21 +0300
Subject: [PATCH 0187/1791] drm/i915: Drop a useless forward declararion
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

ilk_pfit_enable() is defined before the first use. No need
for a forwared declaration.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418175528.13117-9-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 4aef71640aba0..bf391a6cd8d68 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -123,7 +123,6 @@ static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_sta
 static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state);
 static void hsw_set_transconf(const struct intel_crtc_state *crtc_state);
 static void bdw_set_pipe_misc(const struct intel_crtc_state *crtc_state);
-static void ilk_pfit_enable(const struct intel_crtc_state *crtc_state);
 
 /* returns HPLL frequency in kHz */
 int vlv_get_hpll_vco(struct drm_i915_private *dev_priv)
-- 
GitLab


From 3d0d33363cdf4bdc41227cccc28848d6b2fc2998 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:22 +0300
Subject: [PATCH 0188/1791] drm/i915: Fix up whitespace in some display chicken
 registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix a bunch of whitespace issues in some display register
definitons. Only touching the bits alerayd using REG_BIT() &
co. here. The rest will come later.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-2-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 72 ++++++++++++++++-----------------
 1 file changed, 36 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index a89e7a1cdcfeb..f6ca0d82a43c8 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1334,8 +1334,8 @@
 
 #define ILK_DISPLAY_CHICKEN1	_MMIO(0x42000)
 #define   ILK_FBCQ_DIS		(1 << 22)
-#define   ILK_PABSTRETCH_DIS	REG_BIT(21)
-#define   ILK_SABSTRETCH_DIS	REG_BIT(20)
+#define   ILK_PABSTRETCH_DIS		REG_BIT(21)
+#define   ILK_SABSTRETCH_DIS		REG_BIT(20)
 #define   IVB_PRI_STRETCH_MAX_MASK	REG_GENMASK(21, 20)
 #define   IVB_PRI_STRETCH_MAX_X8	REG_FIELD_PREP(IVB_PRI_STRETCH_MAX_MASK, 0)
 #define   IVB_PRI_STRETCH_MAX_X4	REG_FIELD_PREP(IVB_PRI_STRETCH_MAX_MASK, 1)
@@ -2649,7 +2649,7 @@
 #define   PIPE_MISC_YUV420_MODE_FULL_BLEND	REG_BIT(26) /* glk+ */
 #define   PIPE_MISC_HDR_MODE_PRECISION		REG_BIT(23) /* icl+ */
 #define   PIPE_MISC_OUTPUT_COLORSPACE_YUV	REG_BIT(11)
-#define   PIPE_MISC_PIXEL_ROUNDING_TRUNC		REG_BIT(8) /* tgl+ */
+#define   PIPE_MISC_PIXEL_ROUNDING_TRUNC	REG_BIT(8) /* tgl+ */
 /*
  * For Display < 13, Bits 5-7 of PIPE MISC represent DITHER BPC with
  * valid values of: 6, 8, 10 BPC.
@@ -4528,8 +4528,8 @@
 # define CHICKEN3_DGMG_DONE_FIX_DISABLE		(1 << 2)
 
 #define CHICKEN_PAR1_1			_MMIO(0x42080)
-#define  IGNORE_KVMR_PIPE_A		REG_BIT(23)
-#define  KBL_ARB_FILL_SPARE_22		REG_BIT(22)
+#define   IGNORE_KVMR_PIPE_A		REG_BIT(23)
+#define   KBL_ARB_FILL_SPARE_22		REG_BIT(22)
 #define  DIS_RAM_BYPASS_PSR2_MAN_TRACK	(1 << 16)
 #define  SKL_DE_COMPRESSED_HASH_MODE	(1 << 15)
 #define  DPA_MASK_VBLANK_SRD		(1 << 15)
@@ -4541,9 +4541,9 @@
 #define  KVM_CONFIG_CHANGE_NOTIFICATION_SELECT	(1 << 14)
 
 #define CHICKEN_MISC_2		_MMIO(0x42084)
-#define  CHICKEN_MISC_DISABLE_DPT	REG_BIT(30) /* adl,dg2 */
-#define  KBL_ARB_FILL_SPARE_14	REG_BIT(14)
-#define  KBL_ARB_FILL_SPARE_13	REG_BIT(13)
+#define   CHICKEN_MISC_DISABLE_DPT	REG_BIT(30) /* adl,dg2 */
+#define   KBL_ARB_FILL_SPARE_14		REG_BIT(14)
+#define   KBL_ARB_FILL_SPARE_13		REG_BIT(13)
 #define  GLK_CL2_PWR_DOWN	(1 << 12)
 #define  GLK_CL1_PWR_DOWN	(1 << 11)
 #define  GLK_CL0_PWR_DOWN	(1 << 10)
@@ -4555,24 +4555,24 @@
 
 #define _CHICKEN_PIPESL_1_A	0x420b0
 #define _CHICKEN_PIPESL_1_B	0x420b4
-#define  HSW_PRI_STRETCH_MAX_MASK	REG_GENMASK(28, 27)
-#define  HSW_PRI_STRETCH_MAX_X8		REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 0)
-#define  HSW_PRI_STRETCH_MAX_X4		REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 1)
-#define  HSW_PRI_STRETCH_MAX_X2		REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 2)
-#define  HSW_PRI_STRETCH_MAX_X1		REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 3)
-#define  HSW_SPR_STRETCH_MAX_MASK	REG_GENMASK(26, 25)
-#define  HSW_SPR_STRETCH_MAX_X8		REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 0)
-#define  HSW_SPR_STRETCH_MAX_X4		REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 1)
-#define  HSW_SPR_STRETCH_MAX_X2		REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2)
-#define  HSW_SPR_STRETCH_MAX_X1		REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3)
+#define CHICKEN_PIPESL_1(pipe)	_MMIO_PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B)
+#define   HSW_PRI_STRETCH_MAX_MASK	REG_GENMASK(28, 27)
+#define   HSW_PRI_STRETCH_MAX_X8	REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 0)
+#define   HSW_PRI_STRETCH_MAX_X4	REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 1)
+#define   HSW_PRI_STRETCH_MAX_X2	REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 2)
+#define   HSW_PRI_STRETCH_MAX_X1	REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 3)
+#define   HSW_SPR_STRETCH_MAX_MASK	REG_GENMASK(26, 25)
+#define   HSW_SPR_STRETCH_MAX_X8	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 0)
+#define   HSW_SPR_STRETCH_MAX_X4	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 1)
+#define   HSW_SPR_STRETCH_MAX_X2	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2)
+#define   HSW_SPR_STRETCH_MAX_X1	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3)
 #define  HSW_FBCQ_DIS			(1 << 22)
 #define  BDW_DPRS_MASK_VBLANK_SRD	(1 << 0)
-#define  SKL_PLANE1_STRETCH_MAX_MASK	REG_GENMASK(1, 0)
-#define  SKL_PLANE1_STRETCH_MAX_X8	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0)
-#define  SKL_PLANE1_STRETCH_MAX_X4	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1)
-#define  SKL_PLANE1_STRETCH_MAX_X2	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2)
-#define  SKL_PLANE1_STRETCH_MAX_X1	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3)
-#define CHICKEN_PIPESL_1(pipe) _MMIO_PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B)
+#define   SKL_PLANE1_STRETCH_MAX_MASK	REG_GENMASK(1, 0)
+#define   SKL_PLANE1_STRETCH_MAX_X8	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0)
+#define   SKL_PLANE1_STRETCH_MAX_X4	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1)
+#define   SKL_PLANE1_STRETCH_MAX_X2	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2)
+#define   SKL_PLANE1_STRETCH_MAX_X1	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3)
 
 #define _CHICKEN_TRANS_A	0x420c0
 #define _CHICKEN_TRANS_B	0x420c4
@@ -4590,18 +4590,18 @@
 #define MTL_CHICKEN_TRANS(trans)	_MMIO_TRANS((trans), \
 						    _MTL_CHICKEN_TRANS_A, \
 						    _MTL_CHICKEN_TRANS_B)
-#define  PIPE_VBLANK_WITH_DELAY		REG_BIT(31) /* ADL/DG2 */
-#define  HSW_FRAME_START_DELAY_MASK	REG_GENMASK(28, 27)
-#define  HSW_FRAME_START_DELAY(x)	REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x)
-#define  VSC_DATA_SEL_SOFTWARE_CONTROL	REG_BIT(25) /* GLK */
-#define  FECSTALL_DIS_DPTSTREAM_DPTTG	REG_BIT(23)
-#define  DDI_TRAINING_OVERRIDE_ENABLE	REG_BIT(19)
-#define  ADLP_1_BASED_X_GRANULARITY	REG_BIT(18)
-#define  DDI_TRAINING_OVERRIDE_VALUE	REG_BIT(18)
-#define  DDIE_TRAINING_OVERRIDE_ENABLE	REG_BIT(17) /* CHICKEN_TRANS_A only */
-#define  DDIE_TRAINING_OVERRIDE_VALUE	REG_BIT(16) /* CHICKEN_TRANS_A only */
-#define  PSR2_ADD_VERTICAL_LINE_COUNT	REG_BIT(15)
-#define  PSR2_VSC_ENABLE_PROG_HEADER	REG_BIT(12)
+#define   PIPE_VBLANK_WITH_DELAY	REG_BIT(31) /* ADL/DG2 */
+#define   HSW_FRAME_START_DELAY_MASK	REG_GENMASK(28, 27)
+#define   HSW_FRAME_START_DELAY(x)	REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x)
+#define   VSC_DATA_SEL_SOFTWARE_CONTROL	REG_BIT(25) /* GLK */
+#define   FECSTALL_DIS_DPTSTREAM_DPTTG	REG_BIT(23)
+#define   DDI_TRAINING_OVERRIDE_ENABLE	REG_BIT(19)
+#define   ADLP_1_BASED_X_GRANULARITY	REG_BIT(18)
+#define   DDI_TRAINING_OVERRIDE_VALUE	REG_BIT(18)
+#define   DDIE_TRAINING_OVERRIDE_ENABLE	REG_BIT(17) /* CHICKEN_TRANS_A only */
+#define   DDIE_TRAINING_OVERRIDE_VALUE	REG_BIT(16) /* CHICKEN_TRANS_A only */
+#define   PSR2_ADD_VERTICAL_LINE_COUNT	REG_BIT(15)
+#define   PSR2_VSC_ENABLE_PROG_HEADER	REG_BIT(12)
 
 #define DISP_ARB_CTL	_MMIO(0x45000)
 #define  DISP_FBC_MEMORY_WAKE		(1 << 31)
-- 
GitLab


From 3df3c589cc8f4ff1f08302029d6ddef96ea1d449 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:23 +0300
Subject: [PATCH 0189/1791] drm/i915: Clean up various display chicken
 registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Modernize a bunch of display chicken registers by using
REG_BIT() & co.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-3-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 114 +++++++++++++++++---------------
 1 file changed, 59 insertions(+), 55 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f6ca0d82a43c8..366efba8796f7 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1333,7 +1333,7 @@
 #define   SNB_FBC_FRONT_BUFFER	REG_BIT(1)
 
 #define ILK_DISPLAY_CHICKEN1	_MMIO(0x42000)
-#define   ILK_FBCQ_DIS		(1 << 22)
+#define   ILK_FBCQ_DIS			REG_BIT(22)
 #define   ILK_PABSTRETCH_DIS		REG_BIT(21)
 #define   ILK_SABSTRETCH_DIS		REG_BIT(20)
 #define   IVB_PRI_STRETCH_MAX_MASK	REG_GENMASK(21, 20)
@@ -4499,54 +4499,55 @@
 
 #define ILK_DISPLAY_CHICKEN2	_MMIO(0x42004)
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
-#define  ILK_ELPIN_409_SELECT	(1 << 25)
-#define  ILK_DPARB_GATE	(1 << 22)
-#define  ILK_VSDPFD_FULL	(1 << 21)
-#define FUSE_STRAP			_MMIO(0x42014)
-#define  ILK_INTERNAL_GRAPHICS_DISABLE	(1 << 31)
-#define  ILK_INTERNAL_DISPLAY_DISABLE	(1 << 30)
-#define  ILK_DISPLAY_DEBUG_DISABLE	(1 << 29)
-#define  IVB_PIPE_C_DISABLE		(1 << 28)
-#define  ILK_HDCP_DISABLE		(1 << 25)
-#define  ILK_eDP_A_DISABLE		(1 << 24)
-#define  HSW_CDCLK_LIMIT		(1 << 24)
-#define  ILK_DESKTOP			(1 << 23)
-#define  HSW_CPU_SSC_ENABLE		(1 << 21)
-
-#define FUSE_STRAP3			_MMIO(0x42020)
-#define  HSW_REF_CLK_SELECT		(1 << 1)
-
-#define ILK_DSPCLK_GATE_D			_MMIO(0x42020)
-#define   ILK_VRHUNIT_CLOCK_GATE_DISABLE	(1 << 28)
-#define   ILK_DPFCUNIT_CLOCK_GATE_DISABLE	(1 << 9)
-#define   ILK_DPFCRUNIT_CLOCK_GATE_DISABLE	(1 << 8)
-#define   ILK_DPFDUNIT_CLOCK_GATE_ENABLE	(1 << 7)
-#define   ILK_DPARBUNIT_CLOCK_GATE_ENABLE	(1 << 5)
-
-#define IVB_CHICKEN3	_MMIO(0x4200c)
-# define CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE	(1 << 5)
-# define CHICKEN3_DGMG_DONE_FIX_DISABLE		(1 << 2)
-
-#define CHICKEN_PAR1_1			_MMIO(0x42080)
+#define   ILK_ELPIN_409_SELECT	REG_BIT(25)
+#define   ILK_DPARB_GATE	REG_BIT(22)
+#define   ILK_VSDPFD_FULL	REG_BIT(21)
+
+#define FUSE_STRAP		_MMIO(0x42014)
+#define   ILK_INTERNAL_GRAPHICS_DISABLE	REG_BIT(31)
+#define   ILK_INTERNAL_DISPLAY_DISABLE	REG_BIT(30)
+#define   ILK_DISPLAY_DEBUG_DISABLE	REG_BIT(29)
+#define   IVB_PIPE_C_DISABLE		REG_BIT(28)
+#define   ILK_HDCP_DISABLE		REG_BIT(25)
+#define   ILK_eDP_A_DISABLE		REG_BIT(24)
+#define   HSW_CDCLK_LIMIT		REG_BIT(24)
+#define   ILK_DESKTOP			REG_BIT(23)
+#define   HSW_CPU_SSC_ENABLE		REG_BIT(21)
+
+#define FUSE_STRAP3		_MMIO(0x42020)
+#define   HSW_REF_CLK_SELECT		REG_BIT(1)
+
+#define ILK_DSPCLK_GATE_D	_MMIO(0x42020)
+#define   ILK_VRHUNIT_CLOCK_GATE_DISABLE	REG_BIT(28)
+#define   ILK_DPFCUNIT_CLOCK_GATE_DISABLE	REG_BIT(9)
+#define   ILK_DPFCRUNIT_CLOCK_GATE_DISABLE	REG_BIT(8)
+#define   ILK_DPFDUNIT_CLOCK_GATE_ENABLE	REG_BIT(7)
+#define   ILK_DPARBUNIT_CLOCK_GATE_ENABLE	REG_BIT(5)
+
+#define IVB_CHICKEN3		_MMIO(0x4200c)
+#define   CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE	REG_BIT(5)
+#define   CHICKEN3_DGMG_DONE_FIX_DISABLE	REG_BIT(2)
+
+#define CHICKEN_PAR1_1		_MMIO(0x42080)
 #define   IGNORE_KVMR_PIPE_A		REG_BIT(23)
 #define   KBL_ARB_FILL_SPARE_22		REG_BIT(22)
-#define  DIS_RAM_BYPASS_PSR2_MAN_TRACK	(1 << 16)
-#define  SKL_DE_COMPRESSED_HASH_MODE	(1 << 15)
-#define  DPA_MASK_VBLANK_SRD		(1 << 15)
-#define  FORCE_ARB_IDLE_PLANES		(1 << 14)
-#define  SKL_EDP_PSR_FIX_RDWRAP		(1 << 3)
-#define  IGNORE_PSR2_HW_TRACKING	(1 << 1)
+#define   DIS_RAM_BYPASS_PSR2_MAN_TRACK	REG_BIT(16)
+#define   SKL_DE_COMPRESSED_HASH_MODE	REG_BIT(15)
+#define   DPA_MASK_VBLANK_SRD		REG_BIT(15)
+#define   FORCE_ARB_IDLE_PLANES		REG_BIT(14)
+#define   SKL_EDP_PSR_FIX_RDWRAP	REG_BIT(3)
+#define   IGNORE_PSR2_HW_TRACKING	REG_BIT(1)
 
 #define CHICKEN_PAR2_1		_MMIO(0x42090)
-#define  KVM_CONFIG_CHANGE_NOTIFICATION_SELECT	(1 << 14)
+#define   KVM_CONFIG_CHANGE_NOTIFICATION_SELECT	REG_BIT(14)
 
 #define CHICKEN_MISC_2		_MMIO(0x42084)
 #define   CHICKEN_MISC_DISABLE_DPT	REG_BIT(30) /* adl,dg2 */
 #define   KBL_ARB_FILL_SPARE_14		REG_BIT(14)
 #define   KBL_ARB_FILL_SPARE_13		REG_BIT(13)
-#define  GLK_CL2_PWR_DOWN	(1 << 12)
-#define  GLK_CL1_PWR_DOWN	(1 << 11)
-#define  GLK_CL0_PWR_DOWN	(1 << 10)
+#define   GLK_CL2_PWR_DOWN		REG_BIT(12)
+#define   GLK_CL1_PWR_DOWN		REG_BIT(11)
+#define   GLK_CL0_PWR_DOWN		REG_BIT(10)
 
 #define CHICKEN_MISC_4		_MMIO(0x4208c)
 #define   CHICKEN_FBC_STRIDE_OVERRIDE	REG_BIT(13)
@@ -4566,13 +4567,13 @@
 #define   HSW_SPR_STRETCH_MAX_X4	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 1)
 #define   HSW_SPR_STRETCH_MAX_X2	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2)
 #define   HSW_SPR_STRETCH_MAX_X1	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3)
-#define  HSW_FBCQ_DIS			(1 << 22)
-#define  BDW_DPRS_MASK_VBLANK_SRD	(1 << 0)
+#define   HSW_FBCQ_DIS			REG_BIT(22)
 #define   SKL_PLANE1_STRETCH_MAX_MASK	REG_GENMASK(1, 0)
 #define   SKL_PLANE1_STRETCH_MAX_X8	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0)
 #define   SKL_PLANE1_STRETCH_MAX_X4	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1)
 #define   SKL_PLANE1_STRETCH_MAX_X2	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2)
 #define   SKL_PLANE1_STRETCH_MAX_X1	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3)
+#define   BDW_DPRS_MASK_VBLANK_SRD	REG_BIT(0)
 
 #define _CHICKEN_TRANS_A	0x420c0
 #define _CHICKEN_TRANS_B	0x420c4
@@ -4604,12 +4605,13 @@
 #define   PSR2_VSC_ENABLE_PROG_HEADER	REG_BIT(12)
 
 #define DISP_ARB_CTL	_MMIO(0x45000)
-#define  DISP_FBC_MEMORY_WAKE		(1 << 31)
-#define  DISP_TILE_SURFACE_SWIZZLING	(1 << 13)
-#define  DISP_FBC_WM_DIS		(1 << 15)
+#define   DISP_FBC_MEMORY_WAKE		REG_BIT(31)
+#define   DISP_TILE_SURFACE_SWIZZLING	REG_BIT(13)
+#define   DISP_FBC_WM_DIS		REG_BIT(15)
+
 #define DISP_ARB_CTL2	_MMIO(0x45004)
-#define  DISP_DATA_PARTITION_5_6	(1 << 6)
-#define  DISP_IPC_ENABLE		(1 << 3)
+#define   DISP_DATA_PARTITION_5_6	REG_BIT(6)
+#define   DISP_IPC_ENABLE		REG_BIT(3)
 
 #define GEN7_MSG_CTL	_MMIO(0x45010)
 #define  WAIT_FOR_PCH_RESET_ACK		(1 << 1)
@@ -5119,20 +5121,22 @@
 #define  TRANS_BPC_10			REG_FIELD_PREP(TRANS_BPC_MASK, 1)
 #define  TRANS_BPC_6			REG_FIELD_PREP(TRANS_BPC_MASK, 2)
 #define  TRANS_BPC_12			REG_FIELD_PREP(TRANS_BPC_MASK, 3)
+
 #define _TRANSA_CHICKEN1	 0xf0060
 #define _TRANSB_CHICKEN1	 0xf1060
 #define TRANS_CHICKEN1(pipe)	_MMIO_PIPE(pipe, _TRANSA_CHICKEN1, _TRANSB_CHICKEN1)
-#define  TRANS_CHICKEN1_HDMIUNIT_GC_DISABLE	(1 << 10)
-#define  TRANS_CHICKEN1_DP0UNIT_GC_DISABLE	(1 << 4)
+#define   TRANS_CHICKEN1_HDMIUNIT_GC_DISABLE	REG_BIT(10)
+#define   TRANS_CHICKEN1_DP0UNIT_GC_DISABLE	REG_BIT(4)
+
 #define _TRANSA_CHICKEN2	 0xf0064
 #define _TRANSB_CHICKEN2	 0xf1064
 #define TRANS_CHICKEN2(pipe)	_MMIO_PIPE(pipe, _TRANSA_CHICKEN2, _TRANSB_CHICKEN2)
-#define  TRANS_CHICKEN2_TIMING_OVERRIDE			(1 << 31)
-#define  TRANS_CHICKEN2_FDI_POLARITY_REVERSED		(1 << 29)
-#define  TRANS_CHICKEN2_FRAME_START_DELAY_MASK		(3 << 27)
-#define  TRANS_CHICKEN2_FRAME_START_DELAY(x)		((x) << 27) /* 0-3 */
-#define  TRANS_CHICKEN2_DISABLE_DEEP_COLOR_COUNTER	(1 << 26)
-#define  TRANS_CHICKEN2_DISABLE_DEEP_COLOR_MODESWITCH	(1 << 25)
+#define   TRANS_CHICKEN2_TIMING_OVERRIDE		REG_BIT(31)
+#define   TRANS_CHICKEN2_FDI_POLARITY_REVERSED		REG_BIT(29)
+#define   TRANS_CHICKEN2_FRAME_START_DELAY_MASK		REG_GENMASK(28, 27)
+#define   TRANS_CHICKEN2_FRAME_START_DELAY(x)		REG_FIELD_PREP(TRANS_CHICKEN2_FRAME_START_DELAY_MASK, (x)) /* 0-3 */
+#define   TRANS_CHICKEN2_DISABLE_DEEP_COLOR_COUNTER	REG_BIT(26)
+#define   TRANS_CHICKEN2_DISABLE_DEEP_COLOR_MODESWITCH	REG_BIT(25)
 
 #define SOUTH_CHICKEN1		_MMIO(0xc2000)
 #define  FDIA_PHASE_SYNC_SHIFT_OVR	19
-- 
GitLab


From 1d9582dc2b5f5c82dc9471c8d97c0712d5a04c5c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:24 +0300
Subject: [PATCH 0190/1791] drm/i915/psr: Clean up PSR register defininitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use REG_BIT() & co. to make the PSR register definitions
nicer.

v2: Rebase due to intel_psr_regs.h

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-4-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_psr.c      |  34 ++-
 drivers/gpu/drm/i915/display/intel_psr_regs.h | 197 +++++++++---------
 2 files changed, 116 insertions(+), 115 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 3e2d9998566c7..04bac99e25900 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -419,7 +419,7 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
 	u32 val = 0;
 
 	if (DISPLAY_VER(dev_priv) >= 11)
-		val |= EDP_PSR_TP4_TIME_0US;
+		val |= EDP_PSR_TP4_TIME_0us;
 
 	if (dev_priv->params.psr_safest_params) {
 		val |= EDP_PSR_TP1_TIME_2500us;
@@ -448,9 +448,9 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
 check_tp3_sel:
 	if (intel_dp_source_supports_tps3(dev_priv) &&
 	    drm_dp_tps3_supported(intel_dp->dpcd))
-		val |= EDP_PSR_TP1_TP3_SEL;
+		val |= EDP_PSR_TP_TP1_TP3;
 	else
-		val |= EDP_PSR_TP1_TP2_SEL;
+		val |= EDP_PSR_TP_TP1_TP2;
 
 	return val;
 }
@@ -479,9 +479,9 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	u32 max_sleep_time = 0x1f;
 	u32 val = EDP_PSR_ENABLE;
 
-	val |= psr_compute_idle_frames(intel_dp) << EDP_PSR_IDLE_FRAME_SHIFT;
+	val |= EDP_PSR_IDLE_FRAMES(psr_compute_idle_frames(intel_dp));
 
-	val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT;
+	val |= EDP_PSR_MAX_SLEEP_TIME(max_sleep_time);
 	if (IS_HASWELL(dev_priv))
 		val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
 
@@ -536,7 +536,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
 	u32 val = EDP_PSR2_ENABLE;
 
-	val |= psr_compute_idle_frames(intel_dp) << EDP_PSR2_IDLE_FRAME_SHIFT;
+	val |= EDP_PSR2_IDLE_FRAMES(psr_compute_idle_frames(intel_dp));
 
 	if (DISPLAY_VER(dev_priv) <= 13 && !IS_ALDERLAKE_P(dev_priv))
 		val |= EDP_SU_TRACK_ENABLE;
@@ -570,15 +570,13 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 		 * Still using the default IO_BUFFER_WAKE and FAST_WAKE, see
 		 * comments bellow for more information
 		 */
-		u32 tmp;
+		int tmp;
 
 		tmp = map[intel_dp->psr.io_wake_lines - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES];
-		tmp = tmp << TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT;
-		val |= tmp;
+		val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(tmp + TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES);
 
 		tmp = map[intel_dp->psr.fast_wake_lines - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES];
-		tmp = tmp << TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT;
-		val |= tmp;
+		val |= TGL_EDP_PSR2_FAST_WAKE(tmp + TGL_EDP_PSR2_FAST_WAKE_MIN_LINES);
 	} else if (DISPLAY_VER(dev_priv) >= 12) {
 		val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(intel_dp->psr.io_wake_lines);
 		val |= TGL_EDP_PSR2_FAST_WAKE(intel_dp->psr.fast_wake_lines);
@@ -634,9 +632,9 @@ static void psr2_program_idle_frames(struct intel_dp *intel_dp,
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
 
-	idle_frames <<=  EDP_PSR2_IDLE_FRAME_SHIFT;
 	intel_de_rmw(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder),
-		     EDP_PSR2_IDLE_FRAME_MASK, idle_frames);
+		     EDP_PSR2_IDLE_FRAMES_MASK,
+		     EDP_PSR2_IDLE_FRAMES(idle_frames));
 }
 
 static void tgl_psr2_enable_dc3co(struct intel_dp *intel_dp)
@@ -1107,8 +1105,7 @@ void intel_psr_get_config(struct intel_encoder *encoder,
 
 	if (DISPLAY_VER(dev_priv) >= 12) {
 		val = intel_de_read(dev_priv, TRANS_EXITLINE(intel_dp->psr.transcoder));
-		val &= EXITLINE_MASK;
-		pipe_config->dc3co_exitline = val;
+		pipe_config->dc3co_exitline = REG_FIELD_GET(EXITLINE_MASK, val);
 	}
 unlock:
 	mutex_unlock(&intel_dp->psr.lock);
@@ -2738,8 +2735,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 		};
 		val = intel_de_read(dev_priv,
 				    EDP_PSR_STATUS(intel_dp->psr.transcoder));
-		status_val = (val & EDP_PSR_STATUS_STATE_MASK) >>
-			      EDP_PSR_STATUS_STATE_SHIFT;
+		status_val = REG_FIELD_GET(EDP_PSR_STATUS_STATE_MASK, val);
 		if (status_val < ARRAY_SIZE(live_status))
 			status = live_status[status_val];
 	}
@@ -2801,8 +2797,8 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		val = intel_de_read(dev_priv,
 				    EDP_PSR_PERF_CNT(intel_dp->psr.transcoder));
-		val &= EDP_PSR_PERF_CNT_MASK;
-		seq_printf(m, "Performance counter: %u\n", val);
+		seq_printf(m, "Performance counter: %u\n",
+			   REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val));
 	}
 
 	if (psr->debug & I915_PSR_DEBUG_IRQ) {
diff --git a/drivers/gpu/drm/i915/display/intel_psr_regs.h b/drivers/gpu/drm/i915/display/intel_psr_regs.h
index 958d8cabc44b5..4cef32f47402e 100644
--- a/drivers/gpu/drm/i915/display/intel_psr_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_psr_regs.h
@@ -22,30 +22,36 @@
 #define _SRD_CTL_A				0x60800
 #define _SRD_CTL_EDP				0x6f800
 #define EDP_PSR_CTL(tran)			_MMIO_TRANS2(tran, _SRD_CTL_A)
-#define   EDP_PSR_ENABLE			(1 << 31)
-#define   BDW_PSR_SINGLE_FRAME			(1 << 30)
-#define   EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK	(1 << 29) /* SW can't modify */
-#define   EDP_PSR_LINK_STANDBY			(1 << 27)
-#define   EDP_PSR_MIN_LINK_ENTRY_TIME_MASK	(3 << 25)
-#define   EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES	(0 << 25)
-#define   EDP_PSR_MIN_LINK_ENTRY_TIME_4_LINES	(1 << 25)
-#define   EDP_PSR_MIN_LINK_ENTRY_TIME_2_LINES	(2 << 25)
-#define   EDP_PSR_MIN_LINK_ENTRY_TIME_0_LINES	(3 << 25)
-#define   EDP_PSR_MAX_SLEEP_TIME_SHIFT		20
-#define   EDP_PSR_SKIP_AUX_EXIT			(1 << 12)
-#define   EDP_PSR_TP1_TP2_SEL			(0 << 11)
-#define   EDP_PSR_TP1_TP3_SEL			(1 << 11)
-#define   EDP_PSR_CRC_ENABLE			(1 << 10) /* BDW+ */
-#define   EDP_PSR_TP2_TP3_TIME_500us		(0 << 8)
-#define   EDP_PSR_TP2_TP3_TIME_100us		(1 << 8)
-#define   EDP_PSR_TP2_TP3_TIME_2500us		(2 << 8)
-#define   EDP_PSR_TP2_TP3_TIME_0us		(3 << 8)
-#define   EDP_PSR_TP4_TIME_0US			(3 << 6) /* ICL+ */
-#define   EDP_PSR_TP1_TIME_500us		(0 << 4)
-#define   EDP_PSR_TP1_TIME_100us		(1 << 4)
-#define   EDP_PSR_TP1_TIME_2500us		(2 << 4)
-#define   EDP_PSR_TP1_TIME_0us			(3 << 4)
-#define   EDP_PSR_IDLE_FRAME_SHIFT		0
+#define   EDP_PSR_ENABLE			REG_BIT(31)
+#define   BDW_PSR_SINGLE_FRAME			REG_BIT(30)
+#define   EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK	REG_BIT(29) /* SW can't modify */
+#define   EDP_PSR_LINK_STANDBY			REG_BIT(27)
+#define   EDP_PSR_MIN_LINK_ENTRY_TIME_MASK	REG_GENMASK(26, 25)
+#define   EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES	REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 0)
+#define   EDP_PSR_MIN_LINK_ENTRY_TIME_4_LINES	REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 1)
+#define   EDP_PSR_MIN_LINK_ENTRY_TIME_2_LINES	REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 2)
+#define   EDP_PSR_MIN_LINK_ENTRY_TIME_0_LINES	REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 3)
+#define   EDP_PSR_MAX_SLEEP_TIME_MASK		REG_GENMASK(24, 20)
+#define   EDP_PSR_MAX_SLEEP_TIME(x)		REG_FIELD_PREP(EDP_PSR_MAX_SLEEP_TIME_MASK, (x))
+#define   EDP_PSR_SKIP_AUX_EXIT			REG_BIT(12)
+#define   EDP_PSR_TP_MASK			REG_BIT(11)
+#define   EDP_PSR_TP_TP1_TP2			REG_FIELD_PREP(EDP_PSR_TP_MASK, 0)
+#define   EDP_PSR_TP_TP1_TP3			REG_FIELD_PREP(EDP_PSR_TP_MASK, 1)
+#define   EDP_PSR_CRC_ENABLE			REG_BIT(10) /* BDW+ */
+#define   EDP_PSR_TP2_TP3_TIME_MASK		REG_GENMASK(9, 8)
+#define   EDP_PSR_TP2_TP3_TIME_500us		REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 0)
+#define   EDP_PSR_TP2_TP3_TIME_100us		REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 1)
+#define   EDP_PSR_TP2_TP3_TIME_2500us		REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 2)
+#define   EDP_PSR_TP2_TP3_TIME_0us		REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 3)
+#define   EDP_PSR_TP4_TIME_MASK			REG_GENMASK(7, 6)
+#define   EDP_PSR_TP4_TIME_0us			REG_FIELD_PREP(EDP_PSR_TP4_TIME_MASK, 3) /* ICL+ */
+#define   EDP_PSR_TP1_TIME_MASK			REG_GENMASK(5, 4)
+#define   EDP_PSR_TP1_TIME_500us		REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 0)
+#define   EDP_PSR_TP1_TIME_100us		REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 1)
+#define   EDP_PSR_TP1_TIME_2500us		REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 2)
+#define   EDP_PSR_TP1_TIME_0us			REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 3)
+#define   EDP_PSR_IDLE_FRAMES_MASK		REG_GENMASK(3, 0)
+#define   EDP_PSR_IDLE_FRAMES(x)		REG_FIELD_PREP(EDP_PSR_IDLE_FRAMES_MASK, (x))
 
 /*
  * Until TGL, IMR/IIR are fixed at 0x648xx. On TGL+ those registers are relative
@@ -80,81 +86,80 @@
 #define _SRD_STATUS_A				0x60840
 #define _SRD_STATUS_EDP				0x6f840
 #define EDP_PSR_STATUS(tran)			_MMIO_TRANS2(tran, _SRD_STATUS_A)
-#define   EDP_PSR_STATUS_STATE_MASK		(7 << 29)
-#define   EDP_PSR_STATUS_STATE_SHIFT		29
-#define   EDP_PSR_STATUS_STATE_IDLE		(0 << 29)
-#define   EDP_PSR_STATUS_STATE_SRDONACK		(1 << 29)
-#define   EDP_PSR_STATUS_STATE_SRDENT		(2 << 29)
-#define   EDP_PSR_STATUS_STATE_BUFOFF		(3 << 29)
-#define   EDP_PSR_STATUS_STATE_BUFON		(4 << 29)
-#define   EDP_PSR_STATUS_STATE_AUXACK		(5 << 29)
-#define   EDP_PSR_STATUS_STATE_SRDOFFACK	(6 << 29)
-#define   EDP_PSR_STATUS_LINK_MASK		(3 << 26)
-#define   EDP_PSR_STATUS_LINK_FULL_OFF		(0 << 26)
-#define   EDP_PSR_STATUS_LINK_FULL_ON		(1 << 26)
-#define   EDP_PSR_STATUS_LINK_STANDBY		(2 << 26)
-#define   EDP_PSR_STATUS_MAX_SLEEP_TIMER_SHIFT	20
-#define   EDP_PSR_STATUS_MAX_SLEEP_TIMER_MASK	0x1f
-#define   EDP_PSR_STATUS_COUNT_SHIFT		16
-#define   EDP_PSR_STATUS_COUNT_MASK		0xf
-#define   EDP_PSR_STATUS_AUX_ERROR		(1 << 15)
-#define   EDP_PSR_STATUS_AUX_SENDING		(1 << 12)
-#define   EDP_PSR_STATUS_SENDING_IDLE		(1 << 9)
-#define   EDP_PSR_STATUS_SENDING_TP2_TP3	(1 << 8)
-#define   EDP_PSR_STATUS_SENDING_TP1		(1 << 4)
-#define   EDP_PSR_STATUS_IDLE_MASK		0xf
+#define   EDP_PSR_STATUS_STATE_MASK		REG_GENMASK(31, 29)
+#define   EDP_PSR_STATUS_STATE_IDLE		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 0)
+#define   EDP_PSR_STATUS_STATE_SRDONACK		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 1)
+#define   EDP_PSR_STATUS_STATE_SRDENT		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 2)
+#define   EDP_PSR_STATUS_STATE_BUFOFF		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 3)
+#define   EDP_PSR_STATUS_STATE_BUFON		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 4)
+#define   EDP_PSR_STATUS_STATE_AUXACK		REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 5)
+#define   EDP_PSR_STATUS_STATE_SRDOFFACK	REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 6)
+#define   EDP_PSR_STATUS_LINK_MASK		REG_GENMASK(27, 26)
+#define   EDP_PSR_STATUS_LINK_FULL_OFF		REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 0)
+#define   EDP_PSR_STATUS_LINK_FULL_ON		REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 1)
+#define   EDP_PSR_STATUS_LINK_STANDBY		REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 2)
+#define   EDP_PSR_STATUS_MAX_SLEEP_TIMER_MASK	REG_GENMASK(24, 20)
+#define   EDP_PSR_STATUS_COUNT_MASK		REG_GENMASK(19, 16)
+#define   EDP_PSR_STATUS_AUX_ERROR		REG_BIT(15)
+#define   EDP_PSR_STATUS_AUX_SENDING		REG_BIT(12)
+#define   EDP_PSR_STATUS_SENDING_IDLE		REG_BIT(9)
+#define   EDP_PSR_STATUS_SENDING_TP2_TP3	REG_BIT(8)
+#define   EDP_PSR_STATUS_SENDING_TP1		REG_BIT(4)
+#define   EDP_PSR_STATUS_IDLE_MASK		REG_GENMASK(3, 0)
 
 #define _SRD_PERF_CNT_A			0x60844
 #define _SRD_PERF_CNT_EDP		0x6f844
 #define EDP_PSR_PERF_CNT(tran)		_MMIO_TRANS2(tran, _SRD_PERF_CNT_A)
-#define   EDP_PSR_PERF_CNT_MASK		0xffffff
+#define   EDP_PSR_PERF_CNT_MASK		REG_GENMASK(23, 0)
 
 /* PSR_MASK on SKL+ */
 #define _SRD_DEBUG_A				0x60860
 #define _SRD_DEBUG_EDP				0x6f860
 #define EDP_PSR_DEBUG(tran)			_MMIO_TRANS2(tran, _SRD_DEBUG_A)
-#define   EDP_PSR_DEBUG_MASK_MAX_SLEEP         (1 << 28)
-#define   EDP_PSR_DEBUG_MASK_LPSP              (1 << 27)
-#define   EDP_PSR_DEBUG_MASK_MEMUP             (1 << 26)
-#define   EDP_PSR_DEBUG_MASK_HPD               (1 << 25)
-#define   EDP_PSR_DEBUG_MASK_DISP_REG_WRITE    (1 << 16) /* Reserved in ICL+ */
-#define   EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN (1 << 15) /* SKL+ */
+#define   EDP_PSR_DEBUG_MASK_MAX_SLEEP		REG_BIT(28)
+#define   EDP_PSR_DEBUG_MASK_LPSP		REG_BIT(27)
+#define   EDP_PSR_DEBUG_MASK_MEMUP		REG_BIT(26)
+#define   EDP_PSR_DEBUG_MASK_HPD		REG_BIT(25)
+#define   EDP_PSR_DEBUG_MASK_DISP_REG_WRITE	REG_BIT(16) /* Reserved in ICL+ */
+#define   EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN	REG_BIT(15) /* SKL+ */
 
 #define _PSR2_CTL_A				0x60900
 #define _PSR2_CTL_EDP				0x6f900
 #define EDP_PSR2_CTL(tran)			_MMIO_TRANS2(tran, _PSR2_CTL_A)
-#define   EDP_PSR2_ENABLE			(1 << 31)
-#define   EDP_SU_TRACK_ENABLE			(1 << 30) /* up to adl-p */
-#define   TGL_EDP_PSR2_BLOCK_COUNT_NUM_2	(0 << 28)
-#define   TGL_EDP_PSR2_BLOCK_COUNT_NUM_3	(1 << 28)
+#define   EDP_PSR2_ENABLE			REG_BIT(31)
+#define   EDP_SU_TRACK_ENABLE			REG_BIT(30) /* up to adl-p */
+#define   TGL_EDP_PSR2_BLOCK_COUNT_MASK		REG_BIT(28)
+#define   TGL_EDP_PSR2_BLOCK_COUNT_NUM_2	REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 0)
+#define   TGL_EDP_PSR2_BLOCK_COUNT_NUM_3	REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 1)
 #define   EDP_Y_COORDINATE_ENABLE		REG_BIT(25) /* display 10, 11 and 12 */
 #define   EDP_PSR2_SU_SDP_SCANLINE		REG_BIT(25) /* display 13+ */
-#define   EDP_MAX_SU_DISABLE_TIME(t)		((t) << 20)
-#define   EDP_MAX_SU_DISABLE_TIME_MASK		(0x1f << 20)
+#define   EDP_MAX_SU_DISABLE_TIME_MASK		REG_GENMASK(24, 20)
+#define   EDP_MAX_SU_DISABLE_TIME(t)		REG_FIELD_PREP(EDP_MAX_SU_DISABLE_TIME, (t))
+#define   EDP_PSR2_IO_BUFFER_WAKE_MASK		REG_GENMASK(14, 13)
 #define   EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES	8
-#define   EDP_PSR2_IO_BUFFER_WAKE(lines)	((EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES - (lines)) << 13)
-#define   EDP_PSR2_IO_BUFFER_WAKE_MASK		(3 << 13)
+#define   EDP_PSR2_IO_BUFFER_WAKE(lines)	REG_FIELD_PREP(EDP_PSR2_IO_BUFFER_WAKE_MASK, \
+							       EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES - (lines))
+#define   TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK	REG_GENMASK(15, 13)
 #define   TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES	5
-#define   TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT	13
-#define   TGL_EDP_PSR2_IO_BUFFER_WAKE(lines)	(((lines) - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES) << TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT)
-#define   TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK	(7 << 13)
+#define   TGL_EDP_PSR2_IO_BUFFER_WAKE(lines)	REG_FIELD_PREP(TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK, \
+							       (lines) - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES)
+#define   EDP_PSR2_FAST_WAKE_MASK		REG_GENMASK(12, 11)
 #define   EDP_PSR2_FAST_WAKE_MAX_LINES		8
-#define   EDP_PSR2_FAST_WAKE(lines)		((EDP_PSR2_FAST_WAKE_MAX_LINES - (lines)) << 11)
-#define   EDP_PSR2_FAST_WAKE_MASK		(3 << 11)
+#define   EDP_PSR2_FAST_WAKE(lines)		REG_FIELD_PREP(EDP_PSR2_FAST_WAKE_MASK, \
+							       EDP_PSR2_FAST_WAKE_MAX_LINES - (lines))
+#define   TGL_EDP_PSR2_FAST_WAKE_MASK		REG_GENMASK(12, 10)
 #define   TGL_EDP_PSR2_FAST_WAKE_MIN_LINES	5
-#define   TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT	10
-#define   TGL_EDP_PSR2_FAST_WAKE(lines)		(((lines) - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES) << TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT)
-#define   TGL_EDP_PSR2_FAST_WAKE_MASK		(7 << 10)
-#define   EDP_PSR2_TP2_TIME_500us		(0 << 8)
-#define   EDP_PSR2_TP2_TIME_100us		(1 << 8)
-#define   EDP_PSR2_TP2_TIME_2500us		(2 << 8)
-#define   EDP_PSR2_TP2_TIME_50us		(3 << 8)
-#define   EDP_PSR2_TP2_TIME_MASK		(3 << 8)
-#define   EDP_PSR2_FRAME_BEFORE_SU_SHIFT	4
-#define   EDP_PSR2_FRAME_BEFORE_SU_MASK		(0xf << 4)
-#define   EDP_PSR2_FRAME_BEFORE_SU(a)		((a) << 4)
-#define   EDP_PSR2_IDLE_FRAME_MASK		0xf
-#define   EDP_PSR2_IDLE_FRAME_SHIFT		0
+#define   TGL_EDP_PSR2_FAST_WAKE(lines)		REG_FIELD_PREP(TGL_EDP_PSR2_FAST_WAKE_MASK, \
+							       (lines) - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES)
+#define   EDP_PSR2_TP2_TIME_MASK		REG_GENMASK(9, 8)
+#define   EDP_PSR2_TP2_TIME_500us		REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 0)
+#define   EDP_PSR2_TP2_TIME_100us		REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 1)
+#define   EDP_PSR2_TP2_TIME_2500us		REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 2)
+#define   EDP_PSR2_TP2_TIME_50us		REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 3)
+#define   EDP_PSR2_FRAME_BEFORE_SU_MASK		REG_GENMASK(7, 4)
+#define   EDP_PSR2_FRAME_BEFORE_SU(a)		REG_FIELD_PREP(EDP_PSR2_FRAME_BEFORE_SU_MASK, (a))
+#define   EDP_PSR2_IDLE_FRAMES_MASK		REG_GENMASK(3, 0)
+#define   EDP_PSR2_IDLE_FRAMES(x)		REG_FIELD_PREP(EDP_PSR2_IDLE_FRAMES_MASK, (x))
 
 #define _PSR_EVENT_TRANS_A			0x60848
 #define _PSR_EVENT_TRANS_B			0x61848
@@ -162,22 +167,22 @@
 #define _PSR_EVENT_TRANS_D			0x63848
 #define _PSR_EVENT_TRANS_EDP			0x6f848
 #define PSR_EVENT(tran)				_MMIO_TRANS2(tran, _PSR_EVENT_TRANS_A)
-#define  PSR_EVENT_PSR2_WD_TIMER_EXPIRE		(1 << 17)
-#define  PSR_EVENT_PSR2_DISABLED		(1 << 16)
-#define  PSR_EVENT_SU_DIRTY_FIFO_UNDERRUN	(1 << 15)
-#define  PSR_EVENT_SU_CRC_FIFO_UNDERRUN		(1 << 14)
-#define  PSR_EVENT_GRAPHICS_RESET		(1 << 12)
-#define  PSR_EVENT_PCH_INTERRUPT		(1 << 11)
-#define  PSR_EVENT_MEMORY_UP			(1 << 10)
-#define  PSR_EVENT_FRONT_BUFFER_MODIFY		(1 << 9)
-#define  PSR_EVENT_WD_TIMER_EXPIRE		(1 << 8)
-#define  PSR_EVENT_PIPE_REGISTERS_UPDATE	(1 << 6)
-#define  PSR_EVENT_REGISTER_UPDATE		(1 << 5) /* Reserved in ICL+ */
-#define  PSR_EVENT_HDCP_ENABLE			(1 << 4)
-#define  PSR_EVENT_KVMR_SESSION_ENABLE		(1 << 3)
-#define  PSR_EVENT_VBI_ENABLE			(1 << 2)
-#define  PSR_EVENT_LPSP_MODE_EXIT		(1 << 1)
-#define  PSR_EVENT_PSR_DISABLE			(1 << 0)
+#define  PSR_EVENT_PSR2_WD_TIMER_EXPIRE		REG_BIT(17)
+#define  PSR_EVENT_PSR2_DISABLED		REG_BIT(16)
+#define  PSR_EVENT_SU_DIRTY_FIFO_UNDERRUN	REG_BIT(15)
+#define  PSR_EVENT_SU_CRC_FIFO_UNDERRUN		REG_BIT(14)
+#define  PSR_EVENT_GRAPHICS_RESET		REG_BIT(12)
+#define  PSR_EVENT_PCH_INTERRUPT		REG_BIT(11)
+#define  PSR_EVENT_MEMORY_UP			REG_BIT(10)
+#define  PSR_EVENT_FRONT_BUFFER_MODIFY		REG_BIT(9)
+#define  PSR_EVENT_WD_TIMER_EXPIRE		REG_BIT(8)
+#define  PSR_EVENT_PIPE_REGISTERS_UPDATE	REG_BIT(6)
+#define  PSR_EVENT_REGISTER_UPDATE		REG_BIT(5) /* Reserved in ICL+ */
+#define  PSR_EVENT_HDCP_ENABLE			REG_BIT(4)
+#define  PSR_EVENT_KVMR_SESSION_ENABLE		REG_BIT(3)
+#define  PSR_EVENT_VBI_ENABLE			REG_BIT(2)
+#define  PSR_EVENT_LPSP_MODE_EXIT		REG_BIT(1)
+#define  PSR_EVENT_PSR_DISABLE			REG_BIT(0)
 
 #define _PSR2_STATUS_A				0x60940
 #define _PSR2_STATUS_EDP			0x6f940
-- 
GitLab


From 2624ee66359b24062b6e8f6216c4ff2a778237a6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:25 +0300
Subject: [PATCH 0191/1791] drm/i915/psr: Use intel_de_rmw()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace some hand rolled RMW stuff with intel_de_rmw().

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-5-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_psr.c | 32 ++++++++++--------------
 1 file changed, 13 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 04bac99e25900..05cb65b988255 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -232,13 +232,11 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir)
 			    transcoder_name(cpu_transcoder));
 
 		if (DISPLAY_VER(dev_priv) >= 9) {
-			u32 val = intel_de_read(dev_priv,
-						PSR_EVENT(cpu_transcoder));
-			bool psr2_enabled = intel_dp->psr.psr2_enabled;
+			u32 val;
 
-			intel_de_write(dev_priv, PSR_EVENT(cpu_transcoder),
-				       val);
-			psr_event_print(dev_priv, val, psr2_enabled);
+			val = intel_de_rmw(dev_priv, PSR_EVENT(cpu_transcoder), 0, 0);
+
+			psr_event_print(dev_priv, val, intel_dp->psr.psr2_enabled);
 		}
 	}
 
@@ -493,9 +491,8 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	if (DISPLAY_VER(dev_priv) >= 8)
 		val |= EDP_PSR_CRC_ENABLE;
 
-	val |= (intel_de_read(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder)) &
-		EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK);
-	intel_de_write(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder), val);
+	intel_de_rmw(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder),
+		     ~EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK, val);
 }
 
 static u32 intel_psr2_get_tp_time(struct intel_dp *intel_dp)
@@ -1342,19 +1339,16 @@ static void intel_psr_exit(struct intel_dp *intel_dp)
 
 	if (intel_dp->psr.psr2_enabled) {
 		tgl_disallow_dc3co_on_psr2_exit(intel_dp);
-		val = intel_de_read(dev_priv,
-				    EDP_PSR2_CTL(intel_dp->psr.transcoder));
+
+		val = intel_de_rmw(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder),
+				   EDP_PSR2_ENABLE, 0);
+
 		drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR2_ENABLE));
-		val &= ~EDP_PSR2_ENABLE;
-		intel_de_write(dev_priv,
-			       EDP_PSR2_CTL(intel_dp->psr.transcoder), val);
 	} else {
-		val = intel_de_read(dev_priv,
-				    EDP_PSR_CTL(intel_dp->psr.transcoder));
+		val = intel_de_rmw(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder),
+				   EDP_PSR_ENABLE, 0);
+
 		drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR_ENABLE));
-		val &= ~EDP_PSR_ENABLE;
-		intel_de_write(dev_priv,
-			       EDP_PSR_CTL(intel_dp->psr.transcoder), val);
 	}
 	intel_dp->psr.active = false;
 }
-- 
GitLab


From 34c1aeb579ddd36392e18e2fc354db441556811e Mon Sep 17 00:00:00 2001
From: Nikita Travkin <nikita@trvn.ru>
Date: Sat, 8 Apr 2023 13:20:14 +0500
Subject: [PATCH 0192/1791] drm/bridge: ti-sn65dsi86: Implement
 wait_hpd_asserted

This bridge doesn't actually implement HPD due to it being way too slow
but instead expects the panel driver to wait enough to assume HPD is
asserted. However some panels (such as the generic 'edp-panel') expect
the bridge to deal with the delay and pass maximum delay to the aux
instead.

In order to support such panels, add a dummy implementation of wait
that would just sleep the maximum delay and assume no failure has
happened.

Signed-off-by: Nikita Travkin <nikita@trvn.ru>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230408082014.235425-1-nikita@trvn.ru
---
 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 7a748785c5459..260cad1fd1da3 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -618,6 +618,24 @@ static ssize_t ti_sn_aux_transfer(struct drm_dp_aux *aux,
 	return len;
 }
 
+static int ti_sn_aux_wait_hpd_asserted(struct drm_dp_aux *aux, unsigned long wait_us)
+{
+	/*
+	 * The HPD in this chip is a bit useless (See comment in
+	 * ti_sn65dsi86_enable_comms) so if our driver is expected to wait
+	 * for HPD, we just assume it's asserted after the wait_us delay.
+	 *
+	 * In case we are asked to wait forever (wait_us=0) take conservative
+	 * 500ms delay.
+	 */
+	if (wait_us == 0)
+		wait_us = 500000;
+
+	usleep_range(wait_us, wait_us + 1000);
+
+	return 0;
+}
+
 static int ti_sn_aux_probe(struct auxiliary_device *adev,
 			   const struct auxiliary_device_id *id)
 {
@@ -627,6 +645,7 @@ static int ti_sn_aux_probe(struct auxiliary_device *adev,
 	pdata->aux.name = "ti-sn65dsi86-aux";
 	pdata->aux.dev = &adev->dev;
 	pdata->aux.transfer = ti_sn_aux_transfer;
+	pdata->aux.wait_hpd_asserted = ti_sn_aux_wait_hpd_asserted;
 	drm_dp_aux_init(&pdata->aux);
 
 	ret = devm_of_dp_aux_populate_ep_devices(&pdata->aux);
-- 
GitLab


From 2590ef92cdff017ac98012593db54c4d5b96388d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:26 +0300
Subject: [PATCH 0193/1791] drm/i915/psr: Define more PSR mask bits
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Define more of the PSR mask bits, and describe in detail
what some of them do. Even if we don't set them all from
the driver they can be very useful during PSR debugging.
Having to trawl through bspec every time to find them is
not fun, and re-reverse engineering the behaviour every
time is time consuming (even if a bit more fun than spec
trawling).

v2: Moar bits
    Put the description into a comment to be easily available
v2: Fix the BDW_UNMASK_VBL_TO_REGS_IN_SRD/HSW_UNMASK_VBL_TO_REGS_IN_SRD
    description
    Rebase due to intel_psr_regs.h

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-6-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_psr.c      | 85 +++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_psr_regs.h | 14 ++-
 drivers/gpu/drm/i915/i915_reg.h               | 12 ++-
 drivers/gpu/drm/i915/intel_clock_gating.c     |  4 +-
 4 files changed, 109 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 05cb65b988255..c981881cacb67 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -85,6 +85,91 @@
  * use page flips.
  */
 
+/*
+ * Description of PSR mask bits:
+ *
+ * EDP_PSR_DEBUG[16]/EDP_PSR_DEBUG_MASK_DISP_REG_WRITE (hsw-skl):
+ *
+ *  When unmasked (nearly) all display register writes (eg. even
+ *  SWF) trigger a PSR exit. Some registers are excluded from this
+ *  and they have a more specific mask (described below). On icl+
+ *  this bit no longer exists and is effectively always set.
+ *
+ * PIPE_MISC[21]/PIPE_MISC_PSR_MASK_PIPE_REG_WRITE (skl+):
+ *
+ *  When unmasked (nearly) all pipe/plane register writes
+ *  trigger a PSR exit. Some plane registers are excluded from this
+ *  and they have a more specific mask (described below).
+ *
+ * CHICKEN_PIPESL_1[11]/SKL_PSR_MASK_PLANE_FLIP (skl+):
+ * PIPE_MISC[23]/PIPE_MISC_PSR_MASK_PRIMARY_FLIP (bdw):
+ * EDP_PSR_DEBUG[23]/EDP_PSR_DEBUG_MASK_PRIMARY_FLIP (hsw):
+ *
+ *  When unmasked PRI_SURF/PLANE_SURF writes trigger a PSR exit.
+ *  SPR_SURF/CURBASE are not included in this and instead are
+ *  controlled by PIPE_MISC_PSR_MASK_PIPE_REG_WRITE (skl+) or
+ *  EDP_PSR_DEBUG_MASK_DISP_REG_WRITE (hsw/bdw).
+ *
+ * PIPE_MISC[22]/PIPE_MISC_PSR_MASK_SPRITE_ENABLE (bdw):
+ * EDP_PSR_DEBUG[21]/EDP_PSR_DEBUG_MASK_SPRITE_ENABLE (hsw):
+ *
+ *  When unmasked PSR is blocked as long as the sprite
+ *  plane is enabled. skl+ with their universal planes no
+ *  longer have a mask bit like this, and no plane being
+ *  enabledb blocks PSR.
+ *
+ * PIPE_MISC[21]/PIPE_MISC_PSR_MASK_CURSOR_MOVE (bdw):
+ * EDP_PSR_DEBUG[20]/EDP_PSR_DEBUG_MASK_CURSOR_MOVE (hsw):
+ *
+ *  When umasked CURPOS writes trigger a PSR exit. On skl+
+ *  this doesn't exit but CURPOS is included in the
+ *  PIPE_MISC_PSR_MASK_PIPE_REG_WRITE mask.
+ *
+ * PIPE_MISC[20]/PIPE_MISC_PSR_MASK_VBLANK_VSYNC_INT (bdw+):
+ * EDP_PSR_DEBUG[19]/EDP_PSR_DEBUG_MASK_VBLANK_VSYNC_INT (hsw):
+ *
+ *  When unmasked PSR is blocked as long as vblank and/or vsync
+ *  interrupt is unmasked in IMR *and* enabled in IER.
+ *
+ * CHICKEN_TRANS[30]/SKL_UNMASK_VBL_TO_PIPE_IN_SRD (skl+):
+ * CHICKEN_PAR1_1[15]/HSW_MASK_VBL_TO_PIPE_IN_SRD (hsw/bdw):
+ *
+ *  Selectcs whether PSR exit generates an extra vblank before
+ *  the first frame is transmitted. Also note the opposite polarity
+ *  if the bit on hsw/bdw vs. skl+ (masked==generate the extra vblank,
+ *  unmasked==do not generate the extra vblank).
+ *
+ *  With DC states enabled the extra vblank happens after link training,
+ *  with DC states disabled it happens immediately upuon PSR exit trigger.
+ *  No idea as of now why there is a difference. HSW/BDW (which don't
+ *  even have DMC) always generate it after link training. Go figure.
+ *
+ *  Unfortunately CHICKEN_TRANS itself seems to be double buffered
+ *  and thus won't latch until the first vblank. So with DC states
+ *  enabled the register effctively uses the reset value during DC5
+ *  exit+PSR exit sequence, and thus the bit does nothing until
+ *  latched by the vblank that it was trying to prevent from being
+ *  generated in the first place. So we should probably call this
+ *  one a chicken/egg bit instead on skl+.
+ *
+ *  In standby mode (as opposed to link-off) this makes no difference
+ *  as the timing generator keeps running the whole time generating
+ *  normal periodic vblanks.
+ *
+ *  WaPsrDPAMaskVBlankInSRD asks us to set the bit on hsw/bdw,
+ *  and doing so makes the behaviour match the skl+ reset value.
+ *
+ * CHICKEN_PIPESL_1[0]/BDW_UNMASK_VBL_TO_REGS_IN_SRD (bdw):
+ * CHICKEN_PIPESL_1[15]/HSW_UNMASK_VBL_TO_REGS_IN_SRD (hsw):
+ *
+ *  On BDW without this bit is no vblanks whatsoever are
+ *  generated after PSR exit. On HSW this has no apparant effect.
+ *  WaPsrDPRSUnmaskVBlankInSRD says to set this.
+ *
+ * The rest of the bits are more self-explanatory and/or
+ * irrelevant for normal operation.
+ */
+
 static bool psr_global_enabled(struct intel_dp *intel_dp)
 {
 	struct intel_connector *connector = intel_dp->attached_connector;
diff --git a/drivers/gpu/drm/i915/display/intel_psr_regs.h b/drivers/gpu/drm/i915/display/intel_psr_regs.h
index 4cef32f47402e..0f7db617425a3 100644
--- a/drivers/gpu/drm/i915/display/intel_psr_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_psr_regs.h
@@ -120,8 +120,18 @@
 #define   EDP_PSR_DEBUG_MASK_LPSP		REG_BIT(27)
 #define   EDP_PSR_DEBUG_MASK_MEMUP		REG_BIT(26)
 #define   EDP_PSR_DEBUG_MASK_HPD		REG_BIT(25)
-#define   EDP_PSR_DEBUG_MASK_DISP_REG_WRITE	REG_BIT(16) /* Reserved in ICL+ */
-#define   EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN	REG_BIT(15) /* SKL+ */
+#define   EDP_PSR_DEBUG_MASK_FBC_MODIFY		REG_BIT(24)
+#define   EDP_PSR_DEBUG_MASK_PRIMARY_FLIP	REG_BIT(23)  /* hsw */
+#define   EDP_PSR_DEBUG_MASK_HDCP_ENABLE	REG_BIT(22)  /* hsw/bdw */
+#define   EDP_PSR_DEBUG_MASK_SPRITE_ENABLE	REG_BIT(21)  /* hsw */
+#define   EDP_PSR_DEBUG_MASK_CURSOR_MOVE	REG_BIT(20)  /* hsw */
+#define   EDP_PSR_DEBUG_MASK_VBLANK_VSYNC_INT	REG_BIT(19)  /* hsw */
+#define   EDP_PSR_DEBUG_MASK_DPST_PHASE_IN	REG_BIT(18)  /* hsw */
+#define   EDP_PSR_DEBUG_MASK_KVMR_SESSION_EN	REG_BIT(17)
+#define   EDP_PSR_DEBUG_MASK_DISP_REG_WRITE	REG_BIT(16)  /* hsw-skl */
+#define   EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN	REG_BIT(15)  /* skl+ */
+#define   EDP_PSR_DEBUG_RFB_UPDATE_SENT		REG_BIT(2)  /* bdw */
+#define   EDP_PSR_DEBUG_ENTRY_COMPLETION	REG_BIT(1)  /* hsw/bdw */
 
 #define _PSR2_CTL_A				0x60900
 #define _PSR2_CTL_EDP				0x6f900
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 366efba8796f7..831a89841934e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2648,6 +2648,11 @@
 #define   PIPE_MISC_YUV420_ENABLE		REG_BIT(27) /* glk+ */
 #define   PIPE_MISC_YUV420_MODE_FULL_BLEND	REG_BIT(26) /* glk+ */
 #define   PIPE_MISC_HDR_MODE_PRECISION		REG_BIT(23) /* icl+ */
+#define   PIPE_MISC_PSR_MASK_PRIMARY_FLIP	REG_BIT(23) /* bdw */
+#define   PIPE_MISC_PSR_MASK_SPRITE_ENABLE	REG_BIT(22) /* bdw */
+#define   PIPE_MISC_PSR_MASK_PIPE_REG_WRITE	REG_BIT(21) /* skl+ */
+#define   PIPE_MISC_PSR_MASK_CURSOR_MOVE	REG_BIT(21) /* bdw */
+#define   PIPE_MISC_PSR_MASK_VBLANK_VSYNC_INT	REG_BIT(20)
 #define   PIPE_MISC_OUTPUT_COLORSPACE_YUV	REG_BIT(11)
 #define   PIPE_MISC_PIXEL_ROUNDING_TRUNC	REG_BIT(8) /* tgl+ */
 /*
@@ -4533,7 +4538,7 @@
 #define   KBL_ARB_FILL_SPARE_22		REG_BIT(22)
 #define   DIS_RAM_BYPASS_PSR2_MAN_TRACK	REG_BIT(16)
 #define   SKL_DE_COMPRESSED_HASH_MODE	REG_BIT(15)
-#define   DPA_MASK_VBLANK_SRD		REG_BIT(15)
+#define   HSW_MASK_VBL_TO_PIPE_IN_SRD	REG_BIT(15) /* hsw/bdw */
 #define   FORCE_ARB_IDLE_PLANES		REG_BIT(14)
 #define   SKL_EDP_PSR_FIX_RDWRAP	REG_BIT(3)
 #define   IGNORE_PSR2_HW_TRACKING	REG_BIT(1)
@@ -4568,12 +4573,14 @@
 #define   HSW_SPR_STRETCH_MAX_X2	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2)
 #define   HSW_SPR_STRETCH_MAX_X1	REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3)
 #define   HSW_FBCQ_DIS			REG_BIT(22)
+#define   HSW_UNMASK_VBL_TO_REGS_IN_SRD REG_BIT(15) /* hsw */
+#define   SKL_PSR_MASK_PLANE_FLIP	REG_BIT(11) /* skl+ */
 #define   SKL_PLANE1_STRETCH_MAX_MASK	REG_GENMASK(1, 0)
 #define   SKL_PLANE1_STRETCH_MAX_X8	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0)
 #define   SKL_PLANE1_STRETCH_MAX_X4	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1)
 #define   SKL_PLANE1_STRETCH_MAX_X2	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2)
 #define   SKL_PLANE1_STRETCH_MAX_X1	REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3)
-#define   BDW_DPRS_MASK_VBLANK_SRD	REG_BIT(0)
+#define   BDW_UNMASK_VBL_TO_REGS_IN_SRD	REG_BIT(0) /* bdw */
 
 #define _CHICKEN_TRANS_A	0x420c0
 #define _CHICKEN_TRANS_B	0x420c4
@@ -4592,6 +4599,7 @@
 						    _MTL_CHICKEN_TRANS_A, \
 						    _MTL_CHICKEN_TRANS_B)
 #define   PIPE_VBLANK_WITH_DELAY	REG_BIT(31) /* ADL/DG2 */
+#define   SKL_UNMASK_VBL_TO_PIPE_IN_SRD	REG_BIT(30) /* skl+ */
 #define   HSW_FRAME_START_DELAY_MASK	REG_GENMASK(28, 27)
 #define   HSW_FRAME_START_DELAY(x)	REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x)
 #define   VSC_DATA_SEL_SOFTWARE_CONTROL	REG_BIT(25) /* GLK */
diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index 1d796b76f71b9..a27600bc59766 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -521,12 +521,12 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915)
 	intel_uncore_rmw(&i915->uncore, GAM_ECOCHK, 0, HSW_ECOCHK_ARB_PRIO_SOL);
 
 	/* WaPsrDPAMaskVBlankInSRD:bdw */
-	intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, DPA_MASK_VBLANK_SRD);
+	intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD);
 
 	for_each_pipe(i915, pipe) {
 		/* WaPsrDPRSUnmaskVBlankInSRD:bdw */
 		intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe),
-				 0, BDW_DPRS_MASK_VBLANK_SRD);
+				 0, BDW_UNMASK_VBL_TO_REGS_IN_SRD);
 	}
 
 	/* WaVSRefCountFullforceMissDisable:bdw */
-- 
GitLab


From da57e3d6ed433bbfc52730aec111afb6856f9441 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:27 +0300
Subject: [PATCH 0194/1791] drm/i915/psr: Add a FIXME for the PSR vs. AUX usage
 conflict
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We need to disable PSR when we are doing AUX by hand, otherwise
it's possible that the PSR hardware could be using the AUX CH
while we try to do our manual stuff. Add a FIXME for now.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-7-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp_aux.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c
index 705915d505652..abf77ba769728 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c
@@ -268,6 +268,11 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp,
 
 	intel_pps_check_power_unlocked(intel_dp);
 
+	/*
+	 * FIXME PSR should be disabled here to prevent
+	 * it using the same AUX CH simultaneously
+	 */
+
 	/* Try to wait for any previous AUX channel activity */
 	for (try = 0; try < 3; try++) {
 		status = intel_de_read_notrace(i915, ch_ctl);
-- 
GitLab


From 4fb5f569055ccead1807139791b368b31f1e7301 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:28 +0300
Subject: [PATCH 0195/1791] drm/i915/psr: Include PSR_PERF_CNT in debugfs
 output on all platforms
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The fact that DC states reset the PSR perofrmance counter
is no reason not to include it in the debug output.
But let's keep the comment there to remind people about
that caveat.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-8-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_psr.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index c981881cacb67..3a94f40304919 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -2873,12 +2873,10 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 	/*
 	 * SKL+ Perf counter is reset to 0 everytime DC state is entered
 	 */
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
-		val = intel_de_read(dev_priv,
-				    EDP_PSR_PERF_CNT(intel_dp->psr.transcoder));
-		seq_printf(m, "Performance counter: %u\n",
-			   REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val));
-	}
+	val = intel_de_read(dev_priv,
+			    EDP_PSR_PERF_CNT(intel_dp->psr.transcoder));
+	seq_printf(m, "Performance counter: %u\n",
+		   REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val));
 
 	if (psr->debug & I915_PSR_DEBUG_IRQ) {
 		seq_printf(m, "Last attempted entry at: %lld\n",
-- 
GitLab


From a82e0b6fb2cb369672ff9ffb383d7d402aa54163 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Apr 2023 22:14:29 +0300
Subject: [PATCH 0196/1791] drm/i915/psr: Sprinkle cpu_transcoder variables
 around
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Reduce the 'intel_dp' stuff a bit by introducing local
cpu_transcoder variables.

Ideally I'd like the whole PSR code to stop using intel_dp
except during a full modeset, but dunno yet if that's
possible. But the less 'intel_dp' we have sprad around
the easier that should be to figure out eventually.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411191429.29895-9-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_psr.c | 111 ++++++++++++-----------
 1 file changed, 59 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 3a94f40304919..ea0389c5f656b 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -293,13 +293,13 @@ static void psr_event_print(struct drm_i915_private *i915,
 
 void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir)
 {
-	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	ktime_t time_ns =  ktime_get();
 	i915_reg_t imr_reg;
 
 	if (DISPLAY_VER(dev_priv) >= 12)
-		imr_reg = TRANS_PSR_IMR(intel_dp->psr.transcoder);
+		imr_reg = TRANS_PSR_IMR(cpu_transcoder);
 	else
 		imr_reg = EDP_PSR_IMR;
 
@@ -559,6 +559,7 @@ static u8 psr_compute_idle_frames(struct intel_dp *intel_dp)
 static void hsw_activate_psr1(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 max_sleep_time = 0x1f;
 	u32 val = EDP_PSR_ENABLE;
 
@@ -576,7 +577,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	if (DISPLAY_VER(dev_priv) >= 8)
 		val |= EDP_PSR_CRC_ENABLE;
 
-	intel_de_rmw(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder),
+	intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder),
 		     ~EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK, val);
 }
 
@@ -616,6 +617,7 @@ static int psr2_block_count(struct intel_dp *intel_dp)
 static void hsw_activate_psr2(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 val = EDP_PSR2_ENABLE;
 
 	val |= EDP_PSR2_IDLE_FRAMES(psr_compute_idle_frames(intel_dp));
@@ -673,31 +675,30 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 	if (intel_dp->psr.psr2_sel_fetch_enabled) {
 		u32 tmp;
 
-		tmp = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder));
+		tmp = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder));
 		drm_WARN_ON(&dev_priv->drm, !(tmp & PSR2_MAN_TRK_CTL_ENABLE));
 	} else if (HAS_PSR2_SEL_FETCH(dev_priv)) {
-		intel_de_write(dev_priv,
-			       PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), 0);
+		intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), 0);
 	}
 
 	/*
 	 * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is
 	 * recommending keep this bit unset while PSR2 is enabled.
 	 */
-	intel_de_write(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder), 0);
+	intel_de_write(dev_priv, EDP_PSR_CTL(cpu_transcoder), 0);
 
-	intel_de_write(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder), val);
+	intel_de_write(dev_priv, EDP_PSR2_CTL(cpu_transcoder), val);
 }
 
 static bool
-transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder trans)
+transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder cpu_transcoder)
 {
 	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(dev_priv) >= 14)
-		return trans == TRANSCODER_A || trans == TRANSCODER_B;
+		return cpu_transcoder == TRANSCODER_A || cpu_transcoder == TRANSCODER_B;
 	else if (DISPLAY_VER(dev_priv) >= 12)
-		return trans == TRANSCODER_A;
+		return cpu_transcoder == TRANSCODER_A;
 	else
-		return trans == TRANSCODER_EDP;
+		return cpu_transcoder == TRANSCODER_EDP;
 }
 
 static u32 intel_get_frame_time_us(const struct intel_crtc_state *cstate)
@@ -713,8 +714,9 @@ static void psr2_program_idle_frames(struct intel_dp *intel_dp,
 				     u32 idle_frames)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
-	intel_de_rmw(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder),
+	intel_de_rmw(dev_priv, EDP_PSR2_CTL(cpu_transcoder),
 		     EDP_PSR2_IDLE_FRAMES_MASK,
 		     EDP_PSR2_IDLE_FRAMES(idle_frames));
 }
@@ -1154,6 +1156,7 @@ void intel_psr_get_config(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+	enum transcoder cpu_transcoder = pipe_config->cpu_transcoder;
 	struct intel_dp *intel_dp;
 	u32 val;
 
@@ -1180,13 +1183,13 @@ void intel_psr_get_config(struct intel_encoder *encoder,
 		goto unlock;
 
 	if (HAS_PSR2_SEL_FETCH(dev_priv)) {
-		val = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder));
 		if (val & PSR2_MAN_TRK_CTL_ENABLE)
 			pipe_config->enable_psr2_sel_fetch = true;
 	}
 
 	if (DISPLAY_VER(dev_priv) >= 12) {
-		val = intel_de_read(dev_priv, TRANS_EXITLINE(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, TRANS_EXITLINE(cpu_transcoder));
 		pipe_config->dc3co_exitline = REG_FIELD_GET(EXITLINE_MASK, val);
 	}
 unlock:
@@ -1196,14 +1199,14 @@ void intel_psr_get_config(struct intel_encoder *encoder,
 static void intel_psr_activate(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
-	enum transcoder transcoder = intel_dp->psr.transcoder;
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
-	if (transcoder_has_psr2(dev_priv, transcoder))
+	if (transcoder_has_psr2(dev_priv, cpu_transcoder))
 		drm_WARN_ON(&dev_priv->drm,
-			    intel_de_read(dev_priv, EDP_PSR2_CTL(transcoder)) & EDP_PSR2_ENABLE);
+			    intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)) & EDP_PSR2_ENABLE);
 
 	drm_WARN_ON(&dev_priv->drm,
-		    intel_de_read(dev_priv, EDP_PSR_CTL(transcoder)) & EDP_PSR_ENABLE);
+		    intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)) & EDP_PSR_ENABLE);
 	drm_WARN_ON(&dev_priv->drm, intel_dp->psr.active);
 	lockdep_assert_held(&intel_dp->psr.lock);
 
@@ -1282,7 +1285,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 	if (DISPLAY_VER(dev_priv) < 11)
 		mask |= EDP_PSR_DEBUG_MASK_DISP_REG_WRITE;
 
-	intel_de_write(dev_priv, EDP_PSR_DEBUG(intel_dp->psr.transcoder),
+	intel_de_write(dev_priv, EDP_PSR_DEBUG(cpu_transcoder),
 		       mask);
 
 	psr_irq_control(intel_dp);
@@ -1338,6 +1341,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp,
 static bool psr_interrupt_error_check(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 val;
 
 	/*
@@ -1349,8 +1353,7 @@ static bool psr_interrupt_error_check(struct intel_dp *intel_dp)
 	 * to avoid any rendering problems.
 	 */
 	if (DISPLAY_VER(dev_priv) >= 12)
-		val = intel_de_read(dev_priv,
-				    TRANS_PSR_IIR(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, TRANS_PSR_IIR(cpu_transcoder));
 	else
 		val = intel_de_read(dev_priv, EDP_PSR_IIR);
 	val &= psr_irq_psr_error_bit_get(intel_dp);
@@ -1406,17 +1409,16 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp,
 static void intel_psr_exit(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	u32 val;
 
 	if (!intel_dp->psr.active) {
-		if (transcoder_has_psr2(dev_priv, intel_dp->psr.transcoder)) {
-			val = intel_de_read(dev_priv,
-					    EDP_PSR2_CTL(intel_dp->psr.transcoder));
+		if (transcoder_has_psr2(dev_priv, cpu_transcoder)) {
+			val = intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder));
 			drm_WARN_ON(&dev_priv->drm, val & EDP_PSR2_ENABLE);
 		}
 
-		val = intel_de_read(dev_priv,
-				    EDP_PSR_CTL(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder));
 		drm_WARN_ON(&dev_priv->drm, val & EDP_PSR_ENABLE);
 
 		return;
@@ -1425,12 +1427,12 @@ static void intel_psr_exit(struct intel_dp *intel_dp)
 	if (intel_dp->psr.psr2_enabled) {
 		tgl_disallow_dc3co_on_psr2_exit(intel_dp);
 
-		val = intel_de_rmw(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder),
+		val = intel_de_rmw(dev_priv, EDP_PSR2_CTL(cpu_transcoder),
 				   EDP_PSR2_ENABLE, 0);
 
 		drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR2_ENABLE));
 	} else {
-		val = intel_de_rmw(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder),
+		val = intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder),
 				   EDP_PSR_ENABLE, 0);
 
 		drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR_ENABLE));
@@ -1441,14 +1443,15 @@ static void intel_psr_exit(struct intel_dp *intel_dp)
 static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	i915_reg_t psr_status;
 	u32 psr_status_mask;
 
 	if (intel_dp->psr.psr2_enabled) {
-		psr_status = EDP_PSR2_STATUS(intel_dp->psr.transcoder);
+		psr_status = EDP_PSR2_STATUS(cpu_transcoder);
 		psr_status_mask = EDP_PSR2_STATUS_STATE_MASK;
 	} else {
-		psr_status = EDP_PSR_STATUS(intel_dp->psr.transcoder);
+		psr_status = EDP_PSR_STATUS(cpu_transcoder);
 		psr_status_mask = EDP_PSR_STATUS_STATE_MASK;
 	}
 
@@ -1461,6 +1464,7 @@ static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp)
 static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	enum phy phy = intel_port_to_phy(dev_priv,
 					 dp_to_dig_port(intel_dp)->base.port);
 
@@ -1487,7 +1491,7 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 		/* Wa_16012604467:adlp,mtl[a0,b0] */
 		if (IS_MTL_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0))
 			intel_de_rmw(dev_priv,
-				     MTL_CLKGATE_DIS_TRANS(intel_dp->psr.transcoder),
+				     MTL_CLKGATE_DIS_TRANS(cpu_transcoder),
 				     MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS, 0);
 		else if (IS_ALDERLAKE_P(dev_priv))
 			intel_de_rmw(dev_priv, CLKGATE_DIS_MISC,
@@ -1624,10 +1628,11 @@ static u32 man_trk_ctl_continuos_full_frame(struct drm_i915_private *dev_priv)
 static void psr_force_hw_tracking_exit(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	if (intel_dp->psr.psr2_sel_fetch_enabled)
 		intel_de_write(dev_priv,
-			       PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder),
+			       PSR2_MAN_TRK_CTL(cpu_transcoder),
 			       man_trk_ctl_enable_bit_get(dev_priv) |
 			       man_trk_ctl_partial_frame_bit_get(dev_priv) |
 			       man_trk_ctl_single_full_frame_bit_get(dev_priv) |
@@ -1727,6 +1732,7 @@ void intel_psr2_program_plane_sel_fetch_noarm(struct intel_plane *plane,
 void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
+	enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
 	struct intel_encoder *encoder;
 
 	if (!crtc_state->enable_psr2_sel_fetch)
@@ -1742,7 +1748,7 @@ void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_st
 		break;
 	}
 
-	intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(crtc_state->cpu_transcoder),
+	intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder),
 		       crtc_state->psr2_man_track_ctl);
 }
 
@@ -2121,6 +2127,7 @@ void intel_psr_post_plane_update(const struct intel_atomic_state *state)
 static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	/*
 	 * Any state lower than EDP_PSR2_STATUS_STATE_DEEP_SLEEP is enough.
@@ -2128,13 +2135,14 @@ static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
 	 * EDP_PSR2_STATUS_STATE_DEEP_SLEEP to be cleared.
 	 */
 	return intel_de_wait_for_clear(dev_priv,
-				       EDP_PSR2_STATUS(intel_dp->psr.transcoder),
+				       EDP_PSR2_STATUS(cpu_transcoder),
 				       EDP_PSR2_STATUS_STATE_DEEP_SLEEP, 50);
 }
 
 static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	/*
 	 * From bspec: Panel Self Refresh (BDW+)
@@ -2143,7 +2151,7 @@ static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
 	 * defensive enough to cover everything.
 	 */
 	return intel_de_wait_for_clear(dev_priv,
-				       EDP_PSR_STATUS(intel_dp->psr.transcoder),
+				       EDP_PSR_STATUS(cpu_transcoder),
 				       EDP_PSR_STATUS_STATE_MASK, 50);
 }
 
@@ -2185,6 +2193,7 @@ void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_stat
 static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	i915_reg_t reg;
 	u32 mask;
 	int err;
@@ -2193,10 +2202,10 @@ static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp)
 		return false;
 
 	if (intel_dp->psr.psr2_enabled) {
-		reg = EDP_PSR2_STATUS(intel_dp->psr.transcoder);
+		reg = EDP_PSR2_STATUS(cpu_transcoder);
 		mask = EDP_PSR2_STATUS_STATE_MASK;
 	} else {
-		reg = EDP_PSR_STATUS(intel_dp->psr.transcoder);
+		reg = EDP_PSR_STATUS(cpu_transcoder);
 		mask = EDP_PSR_STATUS_STATE_MASK;
 	}
 
@@ -2358,6 +2367,7 @@ static void intel_psr_work(struct work_struct *work)
 static void _psr_invalidate_handle(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	if (intel_dp->psr.psr2_sel_fetch_enabled) {
 		u32 val;
@@ -2371,7 +2381,7 @@ static void _psr_invalidate_handle(struct intel_dp *intel_dp)
 		val = man_trk_ctl_enable_bit_get(dev_priv) |
 		      man_trk_ctl_partial_frame_bit_get(dev_priv) |
 		      man_trk_ctl_continuos_full_frame(dev_priv);
-		intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), val);
+		intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), val);
 		intel_de_write(dev_priv, CURSURFLIVE(intel_dp->psr.pipe), 0);
 		intel_dp->psr.psr2_sel_fetch_cff_enabled = true;
 	} else {
@@ -2450,6 +2460,7 @@ tgl_dc3co_flush_locked(struct intel_dp *intel_dp, unsigned int frontbuffer_bits,
 static void _psr_flush_handle(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 
 	if (intel_dp->psr.psr2_sel_fetch_enabled) {
 		if (intel_dp->psr.psr2_sel_fetch_cff_enabled) {
@@ -2466,7 +2477,7 @@ static void _psr_flush_handle(struct intel_dp *intel_dp)
 				 * SU configuration in case update is sent for any reason after
 				 * sff bit gets cleared by the HW on next vblank.
 				 */
-				intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder),
+				intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder),
 					       val);
 				intel_de_write(dev_priv, CURSURFLIVE(intel_dp->psr.pipe), 0);
 				intel_dp->psr.psr2_sel_fetch_cff_enabled = false;
@@ -2779,6 +2790,7 @@ static void
 psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	const char *status = "unknown";
 	u32 val, status_val;
 
@@ -2796,8 +2808,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 			"BUF_ON",
 			"TG_ON"
 		};
-		val = intel_de_read(dev_priv,
-				    EDP_PSR2_STATUS(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, EDP_PSR2_STATUS(cpu_transcoder));
 		status_val = REG_FIELD_GET(EDP_PSR2_STATUS_STATE_MASK, val);
 		if (status_val < ARRAY_SIZE(live_status))
 			status = live_status[status_val];
@@ -2812,8 +2823,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 			"SRDOFFACK",
 			"SRDENT_ON",
 		};
-		val = intel_de_read(dev_priv,
-				    EDP_PSR_STATUS(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, EDP_PSR_STATUS(cpu_transcoder));
 		status_val = REG_FIELD_GET(EDP_PSR_STATUS_STATE_MASK, val);
 		if (status_val < ARRAY_SIZE(live_status))
 			status = live_status[status_val];
@@ -2825,6 +2835,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m)
 static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
+	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
 	struct intel_psr *psr = &intel_dp->psr;
 	intel_wakeref_t wakeref;
 	const char *status;
@@ -2856,12 +2867,10 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 	}
 
 	if (psr->psr2_enabled) {
-		val = intel_de_read(dev_priv,
-				    EDP_PSR2_CTL(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder));
 		enabled = val & EDP_PSR2_ENABLE;
 	} else {
-		val = intel_de_read(dev_priv,
-				    EDP_PSR_CTL(intel_dp->psr.transcoder));
+		val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder));
 		enabled = val & EDP_PSR_ENABLE;
 	}
 	seq_printf(m, "Source PSR ctl: %s [0x%08x]\n",
@@ -2873,8 +2882,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 	/*
 	 * SKL+ Perf counter is reset to 0 everytime DC state is entered
 	 */
-	val = intel_de_read(dev_priv,
-			    EDP_PSR_PERF_CNT(intel_dp->psr.transcoder));
+	val = intel_de_read(dev_priv, EDP_PSR_PERF_CNT(cpu_transcoder));
 	seq_printf(m, "Performance counter: %u\n",
 		   REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val));
 
@@ -2893,8 +2901,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp)
 		 * frame boundary between register reads
 		 */
 		for (frame = 0; frame < PSR2_SU_STATUS_FRAMES; frame += 3) {
-			val = intel_de_read(dev_priv,
-					    PSR2_SU_STATUS(intel_dp->psr.transcoder, frame));
+			val = intel_de_read(dev_priv, PSR2_SU_STATUS(cpu_transcoder, frame));
 			su_frames_val[frame / 3] = val;
 		}
 
-- 
GitLab


From 79b6e265d92092b49252f546e1a0f63ae8851f83 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 12:28:59 -0400
Subject: [PATCH 0197/1791] drm/amd/display: Fixes for dcn32_clk_mgr
 implementation

[Why&How]
Fix CLK MGR early initialization and add logging.

Fixes: 265280b99822 ("drm/amd/display: add CLKMGR changes for DCN32/321")
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 2b8a81b6d53b1..eea103908b09f 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -905,6 +905,8 @@ void dcn32_clk_mgr_construct(
 		struct pp_smu_funcs *pp_smu,
 		struct dccg *dccg)
 {
+	struct clk_log_info log_info = {0};
+
 	clk_mgr->base.ctx = ctx;
 	clk_mgr->base.funcs = &dcn32_funcs;
 	if (ASICREV_IS_GC_11_0_2(clk_mgr->base.ctx->asic_id.hw_internal_rev)) {
@@ -938,6 +940,7 @@ void dcn32_clk_mgr_construct(
 			clk_mgr->base.clks.ref_dtbclk_khz = 268750;
 	}
 
+
 	/* integer part is now VCO frequency in kHz */
 	clk_mgr->base.dentist_vco_freq_khz = dcn32_get_vco_frequency_from_reg(clk_mgr);
 
@@ -945,6 +948,8 @@ void dcn32_clk_mgr_construct(
 	if (clk_mgr->base.dentist_vco_freq_khz == 0)
 		clk_mgr->base.dentist_vco_freq_khz = 4300000; /* Updated as per HW docs */
 
+	dcn32_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info);
+
 	if (ctx->dc->debug.disable_dtb_ref_clk_switch &&
 			clk_mgr->base.clks.ref_dtbclk_khz != clk_mgr->base.boot_snapshot.dtbclk) {
 		clk_mgr->base.clks.ref_dtbclk_khz = clk_mgr->base.boot_snapshot.dtbclk;
-- 
GitLab


From 224d3df954c184826657bbacd2a562dc99478cb3 Mon Sep 17 00:00:00 2001
From: Samson Tam <Samson.Tam@amd.com>
Date: Wed, 23 Feb 2022 10:23:25 -0500
Subject: [PATCH 0198/1791] drm/amd/display: Clear GPINT1 before taking DMCUB
 out of reset

[Why]
Workaround for DMCUB front door load

[How]
Clear GPINT after reset so its consistent

Signed-off-by: Samson Tam <Samson.Tam@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index a76da0131addd..568a2702d5f75 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -116,10 +116,6 @@ void dmub_dcn32_reset(struct dmub_srv *dmub)
 				break;
 		}
 
-		/* Clear the GPINT command manually so we don't reset again. */
-		cmd.all = 0;
-		dmub->hw_funcs.set_gpint(dmub, cmd);
-
 		/* Force reset in case we timed out, DMCUB is likely hung. */
 	}
 
@@ -131,6 +127,10 @@ void dmub_dcn32_reset(struct dmub_srv *dmub)
 	REG_WRITE(DMCUB_OUTBOX1_RPTR, 0);
 	REG_WRITE(DMCUB_OUTBOX1_WPTR, 0);
 	REG_WRITE(DMCUB_SCRATCH0, 0);
+
+	/* Clear the GPINT command manually so we don't reset again. */
+	cmd.all = 0;
+	dmub->hw_funcs.set_gpint(dmub, cmd);
 }
 
 void dmub_dcn32_reset_release(struct dmub_srv *dmub)
-- 
GitLab


From 66daccde429611530db82605c197be01adadb389 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 1 Dec 2021 16:44:18 +0800
Subject: [PATCH 0199/1791] drm/amdgpu: add master/slave check in init phase

Skip KCQ setup on slave xcc as there's no use case.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 59 +++++++++++++++----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  2 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c  |  5 +++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 10 +++--
 4 files changed, 50 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 7f5c60381103a..c83fb42772339 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -489,16 +489,19 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id)
 		return -EINVAL;
 
 	spin_lock(&kiq->ring_lock);
-	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
-					adev->gfx.num_compute_rings)) {
-		spin_unlock(&adev->gfx.kiq[0].ring_lock);
-		return -ENOMEM;
-	}
+	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
+		if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
+						adev->gfx.num_compute_rings)) {
+			spin_unlock(&kiq->ring_lock);
+			return -ENOMEM;
+		}
 
-	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		j = i + xcc_id * adev->gfx.num_compute_rings;
-		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i],
-					   RESET_QUEUES, 0, 0);
+		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+			j = i + xcc_id * adev->gfx.num_compute_rings;
+			kiq->pmf->kiq_unmap_queues(kiq_ring,
+						   &adev->gfx.compute_ring[i],
+						   RESET_QUEUES, 0, 0);
+		}
 	}
 
 	if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang)
@@ -549,22 +552,26 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 	DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe,
 							kiq_ring->queue);
 	spin_lock(&kiq->ring_lock);
-	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
-					adev->gfx.num_compute_rings +
-					kiq->pmf->set_resources_size);
-	if (r) {
-		DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-		spin_unlock(&adev->gfx.kiq[0].ring_lock);
-		return r;
-	}
+	/* No need to map kcq on the slave */
+	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
+		r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
+						adev->gfx.num_compute_rings +
+						kiq->pmf->set_resources_size);
+		if (r) {
+			DRM_ERROR("Failed to lock KIQ (%d).\n", r);
+			spin_unlock(&adev->gfx.kiq[0].ring_lock);
+			return r;
+		}
 
-	if (adev->enable_mes)
-		queue_mask = ~0ULL;
+		if (adev->enable_mes)
+			queue_mask = ~0ULL;
 
-	kiq->pmf->kiq_set_resources(kiq_ring, queue_mask);
-	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		j = i + xcc_id * adev->gfx.num_compute_rings;
-		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.compute_ring[i]);
+		kiq->pmf->kiq_set_resources(kiq_ring, queue_mask);
+		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+			j = i + xcc_id * adev->gfx.num_compute_rings;
+			kiq->pmf->kiq_map_queues(kiq_ring,
+						 &adev->gfx.compute_ring[i]);
+		}
 	}
 
 	r = amdgpu_ring_test_helper(kiq_ring);
@@ -1078,3 +1085,9 @@ void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev,
 		adev->firmware.fw_size += ALIGN(fw_size, PAGE_SIZE);
 	}
 }
+
+bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id)
+{
+	return !(xcc_id % (adev->gfx.num_xcc_per_xcp ?
+			adev->gfx.num_xcc_per_xcp : 1));
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 77e2f714e357d..a2d311a4da5ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -462,4 +462,6 @@ void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev, uint32_t ucode_id)
 int amdgpu_gfx_ras_sw_init(struct amdgpu_device *adev);
 int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev,
 						struct amdgpu_iv_entry *entry);
+
+bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 4ff348e10e4d4..ef96ff2f42729 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -406,6 +406,11 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
 		else
 			tmo = tmo_gfx;
 
+		/* skip ib test on the slave kcq */
+		if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE &&
+		    !amdgpu_gfx_is_master_xcc(adev, ring->xcc_id))
+			continue;
+
 		r = amdgpu_ring_test_ib(ring, tmo);
 		if (!r) {
 			DRM_DEV_DEBUG(adev->dev, "ib test on %s succeeded\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index baa10ee8ec697..572f84f487cd3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1885,9 +1885,13 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 		if (r)
 			return r;
 
-		for (j = 0; j < adev->gfx.num_compute_rings; j++) {
-			ring = &adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings];
-			amdgpu_ring_test_helper(ring);
+		/* skip ring test on slave kcq */
+		if (amdgpu_gfx_is_master_xcc(adev, i)) {
+			for (j = 0; j < adev->gfx.num_compute_rings; j++) {
+				ring = &adev->gfx.compute_ring[j +
+					i * adev->gfx.num_compute_rings];
+				amdgpu_ring_test_helper(ring);
+			}
 		}
 
 		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
-- 
GitLab


From 4e7f84ec068cec6a9a72fe0f558e0ae4cf765c51 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 16 Dec 2021 15:35:25 +0800
Subject: [PATCH 0200/1791] drm/amdgpu: initialize num_xcd to 1 for gfx v9_0

Assign value here as the num_xcd is referenced in some gfx9 common path.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index bce6919d666ad..8fb027cf1bfb7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4543,6 +4543,7 @@ static int gfx_v9_0_early_init(void *handle)
 		adev->gfx.num_gfx_rings = 0;
 	else
 		adev->gfx.num_gfx_rings = GFX9_NUM_GFX_RINGS;
+	adev->gfx.num_xcd = 1;
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
 	gfx_v9_0_set_kiq_pm4_funcs(adev);
-- 
GitLab


From 20dfae3ca28ff874e506c48704eed2a465a82585 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Thu, 20 Apr 2023 12:23:49 +0200
Subject: [PATCH 0201/1791] drm/i915/mtl: Set has_llc=0

On MTL, LLC is not shared between GT and CPU, set has_llc=0.

Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420102349.15302-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/i915_pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 0de4d22a1291d..6d431489131ee 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1146,6 +1146,7 @@ static const struct intel_device_info mtl_info = {
 	.has_flat_ccs = 0,
 	.has_gmd_id = 1,
 	.has_guc_deprivilege = 1,
+	.has_llc = 0,
 	.has_mslice_steering = 0,
 	.has_snoop = 1,
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM,
-- 
GitLab


From bfa010f608491036327db20aad1d15e28da0189e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jos=C3=A9=20Roberto=20de=20Souza?= <jose.souza@intel.com>
Date: Thu, 20 Apr 2023 10:05:58 -0700
Subject: [PATCH 0202/1791] drm/i915: Initialize dkl_phy spin lock from display
 code path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drm/i915: Initialize dkl_phy spin lock from display code path

Start moving the initialization of display locks from
i915_driver_early_probe().
Display locks should be initialized from display-only code paths.

It was also agreed that if a variable is only used in one file, it
should be initialized only in that file, so intel_dkl_phy_init() was
added.

v2:
- added intel_display_locks_init()

v3:
- rebased

v4:
- dropped intel_display_locks_init()

v5:
- moved intel_dkl_phy_init() to the beginning of file

Cc: intel-gfx@lists.freedesktop.org
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420170558.35398-1-jose.souza@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_driver.c | 2 ++
 drivers/gpu/drm/i915/display/intel_dkl_phy.c        | 9 +++++++++
 drivers/gpu/drm/i915/display/intel_dkl_phy.h        | 1 +
 drivers/gpu/drm/i915/i915_driver.c                  | 1 -
 4 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index b3dbfe2a892e6..60ce10fc72058 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -30,6 +30,7 @@
 #include "intel_display_driver.h"
 #include "intel_display_power.h"
 #include "intel_display_types.h"
+#include "intel_dkl_phy.h"
 #include "intel_dmc.h"
 #include "intel_dp.h"
 #include "intel_dpll.h"
@@ -175,6 +176,7 @@ void intel_display_driver_early_probe(struct drm_i915_private *i915)
 	if (!HAS_DISPLAY(i915))
 		return;
 
+	intel_dkl_phy_init(i915);
 	intel_color_init_hooks(i915);
 	intel_init_cdclk_hooks(i915);
 	intel_audio_hooks_init(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.c b/drivers/gpu/drm/i915/display/intel_dkl_phy.c
index 57cc3edba0163..a001232ad445e 100644
--- a/drivers/gpu/drm/i915/display/intel_dkl_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.c
@@ -11,6 +11,15 @@
 #include "intel_dkl_phy.h"
 #include "intel_dkl_phy_regs.h"
 
+/**
+ * intel_dkl_phy_init - initialize Dekel PHY
+ * @i915: i915 device instance
+ */
+void intel_dkl_phy_init(struct drm_i915_private *i915)
+{
+	spin_lock_init(&i915->display.dkl.phy_lock);
+}
+
 static void
 dkl_phy_set_hip_idx(struct drm_i915_private *i915, struct intel_dkl_phy_reg reg)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.h b/drivers/gpu/drm/i915/display/intel_dkl_phy.h
index 570ee36f9386f..5956ec3e940b1 100644
--- a/drivers/gpu/drm/i915/display/intel_dkl_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.h
@@ -12,6 +12,7 @@
 
 struct drm_i915_private;
 
+void intel_dkl_phy_init(struct drm_i915_private *i915);
 u32
 intel_dkl_phy_read(struct drm_i915_private *i915, struct intel_dkl_phy_reg reg);
 void
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 6ad4190ed46cf..bf3af32fa962d 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -223,7 +223,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 	mutex_init(&dev_priv->display.wm.wm_mutex);
 	mutex_init(&dev_priv->display.pps.mutex);
 	mutex_init(&dev_priv->display.hdcp.comp_mutex);
-	spin_lock_init(&dev_priv->display.dkl.phy_lock);
 
 	i915_memcpy_init_early(dev_priv);
 	intel_runtime_pm_init_early(&dev_priv->runtime_pm);
-- 
GitLab


From 1c9da7db8899fa0e891997118d75b46d38102497 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 30 Mar 2023 09:30:43 +0200
Subject: [PATCH 0203/1791] drm/armada: Include <linux/of.h>

Include <linux/of.h> to get declarations of of_node_put() and
of_device_is_available(). No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330073046.7150-2-tzimmermann@suse.de
---
 drivers/gpu/drm/armada/armada_drv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index c99ec70783013..d8b8e80503d73 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -6,6 +6,7 @@
 #include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_graph.h>
 #include <linux/platform_device.h>
 
-- 
GitLab


From 169c85edcabcc1a7f01b7b5bee1f918ca49ac14f Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 30 Mar 2023 09:30:44 +0200
Subject: [PATCH 0204/1791] drm/armada: Hide fbdev support behind config option

Only build armada's fbdev emulation if CONFIG_DRM_FBDEV_EMULATION
has been enabled. No functional changes, but allows to build armada
without fbdev.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330073046.7150-3-tzimmermann@suse.de
---
 drivers/gpu/drm/armada/Makefile     |  3 ++-
 drivers/gpu/drm/armada/armada_drm.h | 14 ++++++++++++--
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/armada/Makefile b/drivers/gpu/drm/armada/Makefile
index 9bc3c32137248..dc75a7db9ed30 100644
--- a/drivers/gpu/drm/armada/Makefile
+++ b/drivers/gpu/drm/armada/Makefile
@@ -1,7 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
-armada-y	:= armada_crtc.o armada_drv.o armada_fb.o armada_fbdev.o \
+armada-y	:= armada_crtc.o armada_drv.o armada_fb.o \
 		   armada_gem.o armada_overlay.o armada_plane.o armada_trace.o
 armada-y	+= armada_510.o
 armada-$(CONFIG_DEBUG_FS) += armada_debugfs.o
+armada-$(CONFIG_DRM_FBDEV_EMULATION) += armada_fbdev.o
 
 obj-$(CONFIG_DRM_ARMADA) := armada.o
diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
index 6a5a879325766..1475146b1c472 100644
--- a/drivers/gpu/drm/armada/armada_drm.h
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -75,8 +75,18 @@ struct armada_private {
 
 #define drm_to_armada_dev(dev) container_of(dev, struct armada_private, drm)
 
-int armada_fbdev_init(struct drm_device *);
-void armada_fbdev_fini(struct drm_device *);
+#if defined(CONFIG_DRM_FBDEV_EMULATION)
+int armada_fbdev_init(struct drm_device *dev);
+void armada_fbdev_fini(struct drm_device *dev);
+#else
+static inline int armada_fbdev_init(struct drm_device *dev)
+{
+	return 0;
+}
+
+static inline void armada_fbdev_fini(struct drm_device *dev)
+{ }
+#endif
 
 int armada_overlay_plane_create(struct drm_device *, unsigned long);
 
-- 
GitLab


From 174c3c38e3a2a37bccf56e537e3703ce020a50ff Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 30 Mar 2023 09:30:45 +0200
Subject: [PATCH 0205/1791] drm/armada: Initialize fbdev DRM client

Initialize the fbdev client in the fbdev code with empty helper
functions. Also clean up the client. The helpers will later
implement various functionality of the DRM client. No functional
changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330073046.7150-4-tzimmermann@suse.de
---
 drivers/gpu/drm/armada/armada_fbdev.c | 43 ++++++++++++++++++++++++---
 1 file changed, 39 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 0e44f53e9fa40..303c557042250 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -117,6 +117,32 @@ static const struct drm_fb_helper_funcs armada_fb_helper_funcs = {
 	.fb_probe	= armada_fb_probe,
 };
 
+/*
+ * Fbdev client and struct drm_client_funcs
+ */
+
+static void armada_fbdev_client_unregister(struct drm_client_dev *client)
+{ }
+
+static int armada_fbdev_client_restore(struct drm_client_dev *client)
+{
+	drm_fb_helper_lastclose(client->dev);
+
+	return 0;
+}
+
+static int armada_fbdev_client_hotplug(struct drm_client_dev *client)
+{
+	return 0;
+}
+
+static const struct drm_client_funcs armada_fbdev_client_funcs = {
+	.owner		= THIS_MODULE,
+	.unregister	= armada_fbdev_client_unregister,
+	.restore	= armada_fbdev_client_restore,
+	.hotplug	= armada_fbdev_client_hotplug,
+};
+
 int armada_fbdev_init(struct drm_device *dev)
 {
 	struct armada_private *priv = drm_to_armada_dev(dev);
@@ -131,22 +157,30 @@ int armada_fbdev_init(struct drm_device *dev)
 
 	drm_fb_helper_prepare(dev, fbh, 32, &armada_fb_helper_funcs);
 
+	ret = drm_client_init(dev, &fbh->client, "armada-fbdev",
+			      &armada_fbdev_client_funcs);
+	if (ret)
+		goto err_drm_fb_helper_unprepare;
+
 	ret = drm_fb_helper_init(dev, fbh);
 	if (ret) {
 		DRM_ERROR("failed to initialize drm fb helper\n");
-		goto err_fb_helper;
+		goto err_drm_client_release;
 	}
 
 	ret = drm_fb_helper_initial_config(fbh);
 	if (ret) {
 		DRM_ERROR("failed to set initial config\n");
-		goto err_fb_setup;
+		goto err_drm_fb_helper_fini;
 	}
 
 	return 0;
- err_fb_setup:
+
+err_drm_fb_helper_fini:
 	drm_fb_helper_fini(fbh);
- err_fb_helper:
+err_drm_client_release:
+	drm_client_release(&fbh->client);
+err_drm_fb_helper_unprepare:
 	drm_fb_helper_unprepare(fbh);
 	priv->fbdev = NULL;
 	return ret;
@@ -161,6 +195,7 @@ void armada_fbdev_fini(struct drm_device *dev)
 		drm_fb_helper_unregister_info(fbh);
 
 		drm_fb_helper_fini(fbh);
+		drm_client_release(&fbh->client);
 
 		if (fbh->fb)
 			fbh->fb->funcs->destroy(fbh->fb);
-- 
GitLab


From 63c381552f69d188fe1e705515251c52bb07449c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 30 Mar 2023 09:30:46 +0200
Subject: [PATCH 0206/1791] drm/armada: Implement fbdev emulation as in-kernel
 client

Move code from ad-hoc fbdev callbacks into DRM client functions
and remove the old callbacks. The functions instruct the client
to poll for changed output or restore the display. The DRM core
calls both, the old callbacks and the new client helpers, from
the same places. The new functions perform the same operation as
before, so there's no change in functionality.

Replace all code that initializes or releases fbdev emulation
throughout the driver. Instead initialize the fbdev client by a
single call to armada_fbdev_setup() after armada has registered its
DRM device. As in most drivers, aramda's fbdev emulation now acts
like a regular DRM client.

The fbdev client setup consists of the initial preparation and the
hot-plugging of the display. The latter creates the fbdev device
and sets up the fbdev framebuffer. The setup performs display
hot-plugging once. If no display can be detected, DRM probe helpers
re-run the detection on each hotplug event.

A call to drm_dev_unregister() releases the client automatically.
No further action is required within armada. If the fbdev framebuffer
has been fully set up, struct fb_ops.fb_destroy implements the
release. For partially initialized emulation, the fbdev client
reverts the initial setup.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330073046.7150-5-tzimmermann@suse.de
---
 drivers/gpu/drm/armada/armada_drm.h   |  12 +--
 drivers/gpu/drm/armada/armada_drv.c   |  11 +--
 drivers/gpu/drm/armada/armada_fb.c    |   1 -
 drivers/gpu/drm/armada/armada_fbdev.c | 118 +++++++++++++++-----------
 4 files changed, 73 insertions(+), 69 deletions(-)

diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
index 1475146b1c472..c303e8c7ff6c7 100644
--- a/drivers/gpu/drm/armada/armada_drm.h
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -16,7 +16,6 @@ struct armada_crtc;
 struct armada_gem_object;
 struct clk;
 struct drm_display_mode;
-struct drm_fb_helper;
 
 static inline void
 armada_updatel(uint32_t val, uint32_t mask, void __iomem *ptr)
@@ -55,7 +54,6 @@ extern const struct armada_variant armada510_ops;
 
 struct armada_private {
 	struct drm_device	drm;
-	struct drm_fb_helper	*fbdev;
 	struct armada_crtc	*dcrtc[2];
 	struct drm_mm		linear; /* protected by linear_lock */
 	struct mutex		linear_lock;
@@ -76,15 +74,9 @@ struct armada_private {
 #define drm_to_armada_dev(dev) container_of(dev, struct armada_private, drm)
 
 #if defined(CONFIG_DRM_FBDEV_EMULATION)
-int armada_fbdev_init(struct drm_device *dev);
-void armada_fbdev_fini(struct drm_device *dev);
+void armada_fbdev_setup(struct drm_device *dev);
 #else
-static inline int armada_fbdev_init(struct drm_device *dev)
-{
-	return 0;
-}
-
-static inline void armada_fbdev_fini(struct drm_device *dev)
+static inline void armada_fbdev_setup(struct drm_device *dev)
 { }
 #endif
 
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index d8b8e80503d73..22efd0f295218 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -17,7 +17,6 @@
 #include <drm/drm_managed.h>
 #include <drm/drm_prime.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_of.h>
 #include <drm/drm_vblank.h>
 
@@ -38,7 +37,6 @@ static const struct drm_ioctl_desc armada_ioctls[] = {
 DEFINE_DRM_GEM_FOPS(armada_drm_fops);
 
 static const struct drm_driver armada_drm_driver = {
-	.lastclose		= drm_fb_helper_lastclose,
 	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
 	.gem_prime_import	= armada_gem_prime_import,
@@ -56,7 +54,6 @@ static const struct drm_driver armada_drm_driver = {
 
 static const struct drm_mode_config_funcs armada_drm_mode_config_funcs = {
 	.fb_create		= armada_fb_create,
-	.output_poll_changed	= drm_fb_helper_output_poll_changed,
 	.atomic_check		= drm_atomic_helper_check,
 	.atomic_commit		= drm_atomic_helper_commit,
 };
@@ -133,10 +130,6 @@ static int armada_drm_bind(struct device *dev)
 
 	drm_mode_config_reset(&priv->drm);
 
-	ret = armada_fbdev_init(&priv->drm);
-	if (ret)
-		goto err_comp;
-
 	drm_kms_helper_poll_init(&priv->drm);
 
 	ret = drm_dev_register(&priv->drm, 0);
@@ -147,11 +140,12 @@ static int armada_drm_bind(struct device *dev)
 	armada_drm_debugfs_init(priv->drm.primary);
 #endif
 
+	armada_fbdev_setup(&priv->drm);
+
 	return 0;
 
  err_poll:
 	drm_kms_helper_poll_fini(&priv->drm);
-	armada_fbdev_fini(&priv->drm);
  err_comp:
 	component_unbind_all(dev, &priv->drm);
  err_kms:
@@ -166,7 +160,6 @@ static void armada_drm_unbind(struct device *dev)
 	struct armada_private *priv = drm_to_armada_dev(drm);
 
 	drm_kms_helper_poll_fini(&priv->drm);
-	armada_fbdev_fini(&priv->drm);
 
 	drm_dev_unregister(&priv->drm);
 
diff --git a/drivers/gpu/drm/armada/armada_fb.c b/drivers/gpu/drm/armada/armada_fb.c
index b87c71703c850..cf2e88218dc0b 100644
--- a/drivers/gpu/drm/armada/armada_fb.c
+++ b/drivers/gpu/drm/armada/armada_fb.c
@@ -4,7 +4,6 @@
  */
 
 #include <drm/drm_modeset_helper.h>
-#include <drm/drm_fb_helper.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 303c557042250..0a5fd1aa86ebd 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -8,6 +8,8 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 
+#include <drm/drm_crtc_helper.h>
+#include <drm/drm_drv.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_fourcc.h>
 
@@ -16,6 +18,19 @@
 #include "armada_fb.h"
 #include "armada_gem.h"
 
+static void armada_fbdev_fb_destroy(struct fb_info *info)
+{
+	struct drm_fb_helper *fbh = info->par;
+
+	drm_fb_helper_fini(fbh);
+
+	fbh->fb->funcs->destroy(fbh->fb);
+
+	drm_client_release(&fbh->client);
+	drm_fb_helper_unprepare(fbh);
+	kfree(fbh);
+}
+
 static const struct fb_ops armada_fb_ops = {
 	.owner		= THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
@@ -24,6 +39,7 @@ static const struct fb_ops armada_fb_ops = {
 	.fb_fillrect	= drm_fb_helper_cfb_fillrect,
 	.fb_copyarea	= drm_fb_helper_cfb_copyarea,
 	.fb_imageblit	= drm_fb_helper_cfb_imageblit,
+	.fb_destroy	= armada_fbdev_fb_destroy,
 };
 
 static int armada_fbdev_create(struct drm_fb_helper *fbh,
@@ -122,7 +138,17 @@ static const struct drm_fb_helper_funcs armada_fb_helper_funcs = {
  */
 
 static void armada_fbdev_client_unregister(struct drm_client_dev *client)
-{ }
+{
+	struct drm_fb_helper *fbh = drm_fb_helper_from_client(client);
+
+	if (fbh->info) {
+		drm_fb_helper_unregister_info(fbh);
+	} else {
+		drm_client_release(&fbh->client);
+		drm_fb_helper_unprepare(fbh);
+		kfree(fbh);
+	}
+}
 
 static int armada_fbdev_client_restore(struct drm_client_dev *client)
 {
@@ -133,7 +159,31 @@ static int armada_fbdev_client_restore(struct drm_client_dev *client)
 
 static int armada_fbdev_client_hotplug(struct drm_client_dev *client)
 {
+	struct drm_fb_helper *fbh = drm_fb_helper_from_client(client);
+	struct drm_device *dev = client->dev;
+	int ret;
+
+	if (dev->fb_helper)
+		return drm_fb_helper_hotplug_event(dev->fb_helper);
+
+	ret = drm_fb_helper_init(dev, fbh);
+	if (ret)
+		goto err_drm_err;
+
+	if (!drm_drv_uses_atomic_modeset(dev))
+		drm_helper_disable_unused_functions(dev);
+
+	ret = drm_fb_helper_initial_config(fbh);
+	if (ret)
+		goto err_drm_fb_helper_fini;
+
 	return 0;
+
+err_drm_fb_helper_fini:
+	drm_fb_helper_fini(fbh);
+err_drm_err:
+	drm_err(dev, "armada: Failed to setup fbdev emulation (ret=%d)\n", ret);
+	return ret;
 }
 
 static const struct drm_client_funcs armada_fbdev_client_funcs = {
@@ -143,65 +193,35 @@ static const struct drm_client_funcs armada_fbdev_client_funcs = {
 	.hotplug	= armada_fbdev_client_hotplug,
 };
 
-int armada_fbdev_init(struct drm_device *dev)
+void armada_fbdev_setup(struct drm_device *dev)
 {
-	struct armada_private *priv = drm_to_armada_dev(dev);
 	struct drm_fb_helper *fbh;
 	int ret;
 
-	fbh = devm_kzalloc(dev->dev, sizeof(*fbh), GFP_KERNEL);
-	if (!fbh)
-		return -ENOMEM;
-
-	priv->fbdev = fbh;
+	drm_WARN(dev, !dev->registered, "Device has not been registered.\n");
+	drm_WARN(dev, dev->fb_helper, "fb_helper is already set!\n");
 
+	fbh = kzalloc(sizeof(*fbh), GFP_KERNEL);
+	if (!fbh)
+		return;
 	drm_fb_helper_prepare(dev, fbh, 32, &armada_fb_helper_funcs);
 
-	ret = drm_client_init(dev, &fbh->client, "armada-fbdev",
-			      &armada_fbdev_client_funcs);
-	if (ret)
-		goto err_drm_fb_helper_unprepare;
-
-	ret = drm_fb_helper_init(dev, fbh);
+	ret = drm_client_init(dev, &fbh->client, "fbdev", &armada_fbdev_client_funcs);
 	if (ret) {
-		DRM_ERROR("failed to initialize drm fb helper\n");
-		goto err_drm_client_release;
+		drm_err(dev, "Failed to register client: %d\n", ret);
+		goto err_drm_client_init;
 	}
 
-	ret = drm_fb_helper_initial_config(fbh);
-	if (ret) {
-		DRM_ERROR("failed to set initial config\n");
-		goto err_drm_fb_helper_fini;
-	}
-
-	return 0;
-
-err_drm_fb_helper_fini:
-	drm_fb_helper_fini(fbh);
-err_drm_client_release:
-	drm_client_release(&fbh->client);
-err_drm_fb_helper_unprepare:
-	drm_fb_helper_unprepare(fbh);
-	priv->fbdev = NULL;
-	return ret;
-}
-
-void armada_fbdev_fini(struct drm_device *dev)
-{
-	struct armada_private *priv = drm_to_armada_dev(dev);
-	struct drm_fb_helper *fbh = priv->fbdev;
-
-	if (fbh) {
-		drm_fb_helper_unregister_info(fbh);
+	ret = armada_fbdev_client_hotplug(&fbh->client);
+	if (ret)
+		drm_dbg_kms(dev, "client hotplug ret=%d\n", ret);
 
-		drm_fb_helper_fini(fbh);
-		drm_client_release(&fbh->client);
+	drm_client_register(&fbh->client);
 
-		if (fbh->fb)
-			fbh->fb->funcs->destroy(fbh->fb);
+	return;
 
-		drm_fb_helper_unprepare(fbh);
-
-		priv->fbdev = NULL;
-	}
+err_drm_client_init:
+	drm_fb_helper_unprepare(fbh);
+	kfree(fbh);
+	return;
 }
-- 
GitLab


From 6152aec1ddb40620cd8d2b36b45171c2d1bd82d1 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Fri, 17 Mar 2023 17:58:52 -0700
Subject: [PATCH 0207/1791] drm/i915/mtl: Skip pcode qgv restrictions for MTL

Communicating QGV points restriction to PUnit happens via PM Demand
instead of the Pcode mailbox in the previous platforms. GV point
restriction is handled by the PM demand code.

Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230318005852.2303937-5-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 202321ffbe2ac..597d5816ad1b0 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -150,6 +150,9 @@ int icl_pcode_restrict_qgv_points(struct drm_i915_private *dev_priv,
 {
 	int ret;
 
+	if (DISPLAY_VER(dev_priv) >= 14)
+		return 0;
+
 	/* bspec says to keep retrying for at least 1 ms */
 	ret = skl_pcode_request(&dev_priv->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG,
 				points_mask,
-- 
GitLab


From b0dc10428460ac2408cf5f82fc3562e9e57324e1 Mon Sep 17 00:00:00 2001
From: Cruise Hung <Cruise.Hung@amd.com>
Date: Fri, 13 May 2022 09:16:42 +0800
Subject: [PATCH 0208/1791] drm/amd/display: Reset OUTBOX0 r/w pointer on DMUB
 reset

[Why & How]
We missed resetting OUTBOX0 mailbox r/w pointer on DMUB reset.
Fix it.

Fixes: 6ecf9773a503 ("drm/amd/display: Fix DMUB outbox trace in S4 (#4465)")
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index 568a2702d5f75..b45ac31ba5557 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -126,6 +126,8 @@ void dmub_dcn32_reset(struct dmub_srv *dmub)
 	REG_WRITE(DMCUB_INBOX1_WPTR, 0);
 	REG_WRITE(DMCUB_OUTBOX1_RPTR, 0);
 	REG_WRITE(DMCUB_OUTBOX1_WPTR, 0);
+	REG_WRITE(DMCUB_OUTBOX0_RPTR, 0);
+	REG_WRITE(DMCUB_OUTBOX0_WPTR, 0);
 	REG_WRITE(DMCUB_SCRATCH0, 0);
 
 	/* Clear the GPINT command manually so we don't reset again. */
-- 
GitLab


From 01382501509871d0799bab6bd412c228486af5bf Mon Sep 17 00:00:00 2001
From: Longlong Yao <Longlong.Yao@amd.com>
Date: Thu, 13 Apr 2023 13:21:45 +0800
Subject: [PATCH 0209/1791] drm/amdgpu: fix calltrace warning in
 amddrm_buddy_fini

The following call trace is observed when removing the amdgpu driver, which
is caused by that BOs allocated for psp are not freed until removing.

[61811.450562] RIP: 0010:amddrm_buddy_fini.cold+0x29/0x47 [amddrm_buddy]
[61811.450577] Call Trace:
[61811.450577]  <TASK>
[61811.450579]  amdgpu_vram_mgr_fini+0x135/0x1c0 [amdgpu]
[61811.450728]  amdgpu_ttm_fini+0x207/0x290 [amdgpu]
[61811.450870]  amdgpu_bo_fini+0x27/0xa0 [amdgpu]
[61811.451012]  gmc_v9_0_sw_fini+0x4a/0x60 [amdgpu]
[61811.451166]  amdgpu_device_fini_sw+0x117/0x520 [amdgpu]
[61811.451306]  amdgpu_driver_release_kms+0x16/0x30 [amdgpu]
[61811.451447]  devm_drm_dev_init_release+0x4d/0x80 [drm]
[61811.451466]  devm_action_release+0x15/0x20
[61811.451469]  release_nodes+0x40/0xb0
[61811.451471]  devres_release_all+0x9b/0xd0
[61811.451473]  __device_release_driver+0x1bb/0x2a0
[61811.451476]  driver_detach+0xf3/0x140
[61811.451479]  bus_remove_driver+0x6c/0xf0
[61811.451481]  driver_unregister+0x31/0x60
[61811.451483]  pci_unregister_driver+0x40/0x90
[61811.451486]  amdgpu_exit+0x15/0x447 [amdgpu]

For smu v13_0_2, if the GPU supports xgmi, refer to

commit f5c7e7797060 ("drm/amdgpu: Adjust removal control flow for smu v13_0_2"),

it will run gpu recover in AMDGPU_RESET_FOR_DEVICE_REMOVE mode when removing,
which makes all devices in hive list have hw reset but no resume except the
basic ip blocks, then other ip blocks will not call .hw_fini according to
ip_block.status.hw.

Since psp_free_shared_bufs just includes some software operations, so move
it to psp_sw_fini.

Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Longlong Yao <Longlong.Yao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 9d7e6e0e73edb..a496bf2fb1991 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -520,6 +520,8 @@ static int psp_sw_fini(void *handle)
 	kfree(cmd);
 	cmd = NULL;
 
+	psp_free_shared_bufs(psp);
+
 	if (psp->km_ring.ring_mem)
 		amdgpu_bo_free_kernel(&adev->firmware.rbuf,
 				      &psp->km_ring.ring_mem_mc_addr,
@@ -2655,8 +2657,6 @@ static int psp_hw_fini(void *handle)
 
 	psp_ring_destroy(psp, PSP_RING_TYPE__KM);
 
-	psp_free_shared_bufs(psp);
-
 	return 0;
 }
 
-- 
GitLab


From 8eba72053c682d8ae652bed4a62546239e58390b Mon Sep 17 00:00:00 2001
From: Candice Li <candice.li@amd.com>
Date: Wed, 19 Apr 2023 17:28:19 +0800
Subject: [PATCH 0210/1791] drm/amdgpu: Drop pcie_bif ras check from fatal
 error handler

Some ASICs support fatal error event but do not
support pcie_bif ras.

Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 3ab8a88789c8f..22f401fd1901c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1597,8 +1597,7 @@ static int amdgpu_ras_fs_fini(struct amdgpu_device *adev)
 void amdgpu_ras_interrupt_fatal_error_handler(struct amdgpu_device *adev)
 {
 	/* Fatal error events are handled on host side */
-	if (amdgpu_sriov_vf(adev) ||
-		!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__PCIE_BIF))
+	if (amdgpu_sriov_vf(adev))
 		return;
 
 	if (adev->nbio.ras &&
-- 
GitLab


From 5c65a4b8e65e3247f86ec5152da138500edcf50c Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 19 Apr 2023 14:55:03 +0530
Subject: [PATCH 0211/1791] drm/amd/amdgpu: Fix spaces in array indexing and
 indentations in amdgpu_kms.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following errors reported by checkpatch:

ERROR: space prohibited before open square bracket '['
+#define TA_FW_NAME(type) [TA_FW_TYPE_PSP_##type] = #type

ERROR: code indent should use tabs where possible
+        query_fw.fw_type = AMDGPU_INFO_FW_GFX_RLCV;$

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 1a2e342af1c09..a5bae7eb993a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1458,7 +1458,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused)
 	int ret, i;
 
 	static const char *ta_fw_name[TA_FW_TYPE_MAX_INDEX] = {
-#define TA_FW_NAME(type) [TA_FW_TYPE_PSP_##type] = #type
+#define TA_FW_NAME(type)[TA_FW_TYPE_PSP_##type] = #type
 		TA_FW_NAME(XGMI),
 		TA_FW_NAME(RAS),
 		TA_FW_NAME(HDCP),
@@ -1557,7 +1557,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused)
 		   fw_info.feature, fw_info.ver);
 
 	/* RLCV */
-        query_fw.fw_type = AMDGPU_INFO_FW_GFX_RLCV;
+	query_fw.fw_type = AMDGPU_INFO_FW_GFX_RLCV;
 	ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);
 	if (ret)
 		return ret;
-- 
GitLab


From 47fc644f801e4414753a9b7e87ed41f991cd68c3 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 19 Apr 2023 16:12:45 +0530
Subject: [PATCH 0212/1791] drm/amd/amdgpu: Fix style errors in amdgpu_drv.c &
 amdgpu_device.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix following checkpatch style errors in amdgpu_drv.c &
amdgpu_device.c

ERROR: exactly one space required after that #ifdef
ERROR: spaces required around that '+=' (ctx:WxV)
ERROR: space required before the open brace '{'
ERROR: spaces required around that '||' (ctx:VxE)
ERROR: space prohibited before that close parenthesis ')'
ERROR: space required before the open parenthesis '('
ERROR: space required before the open brace '{'
ERROR: code indent should use tabs where possible

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 65 +++++++++++-----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c    |  2 +-
 2 files changed, 34 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 9b1eaba85bbd8..ac78caa7cba87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -998,7 +998,7 @@ void amdgpu_device_program_register_sequence(struct amdgpu_device *adev,
 	if (array_size % 3)
 		return;
 
-	for (i = 0; i < array_size; i +=3) {
+	for (i = 0; i < array_size; i += 3) {
 		reg = registers[i + 0];
 		and_mask = registers[i + 1];
 		or_mask = registers[i + 2];
@@ -1547,7 +1547,7 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
 		dev_warn(adev->dev, "sched jobs (%d) must be at least 4\n",
 			 amdgpu_sched_jobs);
 		amdgpu_sched_jobs = 4;
-	} else if (!is_power_of_2(amdgpu_sched_jobs)){
+	} else if (!is_power_of_2(amdgpu_sched_jobs)) {
 		dev_warn(adev->dev, "sched jobs (%d) must be a power of 2\n",
 			 amdgpu_sched_jobs);
 		amdgpu_sched_jobs = roundup_pow_of_two(amdgpu_sched_jobs);
@@ -2759,8 +2759,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 		DRM_ERROR("enable mgpu fan boost failed (%d).\n", r);
 
 	/* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */
-	if (amdgpu_passthrough(adev) && ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1)||
-			       adev->asic_type == CHIP_ALDEBARAN ))
+	if (amdgpu_passthrough(adev) &&
+	    ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1) ||
+	     adev->asic_type == CHIP_ALDEBARAN))
 		amdgpu_dpm_handle_passthrough_sbr(adev, true);
 
 	if (adev->gmc.xgmi.num_physical_nodes > 1) {
@@ -3089,7 +3090,7 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev)
 		}
 		adev->ip_blocks[i].status.hw = false;
 		/* handle putting the SMC in the appropriate state */
-		if(!amdgpu_sriov_vf(adev)){
+		if (!amdgpu_sriov_vf(adev)) {
 			if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
 				r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state);
 				if (r) {
@@ -4036,7 +4037,7 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev)
 
 	/* disable all interrupts */
 	amdgpu_irq_disable_all(adev);
-	if (adev->mode_info.mode_config_initialized){
+	if (adev->mode_info.mode_config_initialized) {
 		if (!drm_drv_uses_atomic_modeset(adev_to_drm(adev)))
 			drm_helper_force_disable_all(adev_to_drm(adev));
 		else
@@ -4697,42 +4698,42 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev)
 
 int amdgpu_device_mode1_reset(struct amdgpu_device *adev)
 {
-        u32 i;
-        int ret = 0;
+	u32 i;
+	int ret = 0;
 
-        amdgpu_atombios_scratch_regs_engine_hung(adev, true);
+	amdgpu_atombios_scratch_regs_engine_hung(adev, true);
 
-        dev_info(adev->dev, "GPU mode1 reset\n");
+	dev_info(adev->dev, "GPU mode1 reset\n");
 
-        /* disable BM */
-        pci_clear_master(adev->pdev);
+	/* disable BM */
+	pci_clear_master(adev->pdev);
 
-        amdgpu_device_cache_pci_state(adev->pdev);
+	amdgpu_device_cache_pci_state(adev->pdev);
 
-        if (amdgpu_dpm_is_mode1_reset_supported(adev)) {
-                dev_info(adev->dev, "GPU smu mode1 reset\n");
-                ret = amdgpu_dpm_mode1_reset(adev);
-        } else {
-                dev_info(adev->dev, "GPU psp mode1 reset\n");
-                ret = psp_gpu_reset(adev);
-        }
+	if (amdgpu_dpm_is_mode1_reset_supported(adev)) {
+		dev_info(adev->dev, "GPU smu mode1 reset\n");
+		ret = amdgpu_dpm_mode1_reset(adev);
+	} else {
+		dev_info(adev->dev, "GPU psp mode1 reset\n");
+		ret = psp_gpu_reset(adev);
+	}
 
-        if (ret)
-                dev_err(adev->dev, "GPU mode1 reset failed\n");
+	if (ret)
+		dev_err(adev->dev, "GPU mode1 reset failed\n");
 
-        amdgpu_device_load_pci_state(adev->pdev);
+	amdgpu_device_load_pci_state(adev->pdev);
 
-        /* wait for asic to come out of reset */
-        for (i = 0; i < adev->usec_timeout; i++) {
-                u32 memsize = adev->nbio.funcs->get_memsize(adev);
+	/* wait for asic to come out of reset */
+	for (i = 0; i < adev->usec_timeout; i++) {
+		u32 memsize = adev->nbio.funcs->get_memsize(adev);
 
-                if (memsize != 0xffffffff)
-                        break;
-                udelay(1);
-        }
+		if (memsize != 0xffffffff)
+			break;
+		udelay(1);
+	}
 
-        amdgpu_atombios_scratch_regs_engine_hung(adev, false);
-        return ret;
+	amdgpu_atombios_scratch_regs_engine_hung(adev, false);
+	return ret;
 }
 
 int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b1ca1ab6d6ad3..b400d598b75ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -1660,7 +1660,7 @@ static const u16 amdgpu_unsupported_pciidlist[] = {
 };
 
 static const struct pci_device_id pciidlist[] = {
-#ifdef  CONFIG_DRM_AMDGPU_SI
+#ifdef CONFIG_DRM_AMDGPU_SI
 	{0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
 	{0x1002, 0x6784, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
 	{0x1002, 0x6788, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
-- 
GitLab


From 47659738fbd2f06730635a487605002ea9b11f3d Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 19 Nov 2021 15:35:30 +0800
Subject: [PATCH 0213/1791] drm/amdgpu: allocate doorbell index for multi-die
 case

Allocate different doorbell index for kiq/kcq rings
on each die

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h | 9 ++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c      | 5 +++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c       | 1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h      | 1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c      | 8 +++++++-
 5 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index 8fd11497fabaa..ffb75d23d2fc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -83,6 +83,8 @@ struct amdgpu_doorbell_index {
 	};
 	uint32_t first_non_cp;
 	uint32_t last_non_cp;
+	uint32_t xcc1_kiq_start;
+	uint32_t xcc1_mec_ring0_start;
 	uint32_t max_assignment;
 	/* Per engine SDMA doorbell size in dword */
 	uint32_t sdma_doorbell_range;
@@ -164,7 +166,12 @@ typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT
 	AMDGPU_VEGA20_DOORBELL64_FIRST_NON_CP            = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0,
 	AMDGPU_VEGA20_DOORBELL64_LAST_NON_CP             = AMDGPU_VEGA20_DOORBELL64_VCE_RING6_7,
 
-	AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT            = 0x18F,
+	/* kiq/kcq from second XCD. Max 8 XCDs */
+	AMDGPU_VEGA20_DOORBELL_XCC1_KIQ_START             = 0x190,
+	/* 8 compute rings per GC. Max to 0x1CE */
+	AMDGPU_VEGA20_DOORBELL_XCC1_MEC_RING0_START       = 0x197,
+
+	AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT            = 0x1CE,
 	AMDGPU_VEGA20_DOORBELL_INVALID                   = 0xFFFF
 } AMDGPU_VEGA20_DOORBELL_ASSIGNMENT;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index c83fb42772339..465ad0b7cddb0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -316,6 +316,11 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->doorbell_index = adev->doorbell_index.kiq;
 	ring->xcc_id = xcc_id;
 	ring->vm_hub = AMDGPU_GFXHUB_0;
+	if (xcc_id >= 1)
+		ring->doorbell_index = adev->doorbell_index.xcc1_kiq_start +
+					xcc_id - 1;
+	else
+		ring->doorbell_index = adev->doorbell_index.kiq;
 
 	r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id);
 	if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index 1d5af50331e41..d58353c89e59a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -270,6 +270,7 @@ void amdgpu_ih_decode_iv_helper(struct amdgpu_device *adev,
 	entry->timestamp = dw[1] | ((u64)(dw[2] & 0xffff) << 32);
 	entry->timestamp_src = dw[2] >> 31;
 	entry->pasid = dw[3] & 0xffff;
+	entry->node_id = (dw[3] >> 16) & 0xff;
 	entry->pasid_src = dw[3] >> 31;
 	entry->src_data[0] = dw[4];
 	entry->src_data[1] = dw[5];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
index be243adf3e657..7a8e686bdd412 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
@@ -53,6 +53,7 @@ struct amdgpu_iv_entry {
 	uint64_t timestamp;
 	unsigned timestamp_src;
 	unsigned pasid;
+	unsigned node_id;
 	unsigned pasid_src;
 	unsigned src_data[AMDGPU_IRQ_SRC_DATA_MAX_SIZE_DW];
 	const uint32_t *iv_entry;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 572f84f487cd3..56a415e151d4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -752,7 +752,13 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
-	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
+	if (xcc_id >= 1)
+		ring->doorbell_index =
+				(adev->doorbell_index.xcc1_mec_ring0_start +
+				ring_id - adev->gfx.num_compute_rings) << 1;
+	else
+		ring->doorbell_index =
+				(adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
 	ring->vm_hub = AMDGPU_GFXHUB_0;
-- 
GitLab


From 948ca54c424be395402624ca0e21ad5ddf77cb6a Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 19 Apr 2023 17:35:15 +0530
Subject: [PATCH 0214/1791] drm/amd/display: Remove unused variables in
 dcn21_hwseq.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the below compiler error:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn21/dcn21_hwseq.c:229:11: error: unused variable 'otg_inst' [-Werror,-Wunused-variable]
        uint32_t otg_inst = pipe_ctx->stream_res.tg->inst;
                 ^
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn21/dcn21_hwseq.c:226:20: error: unused variable 'cmd' [-Werror,-Wunused-variable]
        union dmub_rb_cmd cmd;

Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Jerry Zuo <jerry.zuo@amd.com>
Cc: Yongqiang Sun <yongqiang.sun@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Yongqiang Sun <Yongqiang.Sun@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
index 55a464a39529b..43463d08f21ba 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c
@@ -223,10 +223,8 @@ bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx,
 		uint32_t backlight_pwm_u16_16,
 		uint32_t frame_ramp)
 {
-	union dmub_rb_cmd cmd;
 	struct dc_context *dc = pipe_ctx->stream->ctx;
 	struct abm *abm = pipe_ctx->stream_res.abm;
-	uint32_t otg_inst = pipe_ctx->stream_res.tg->inst;
 	struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl;
 
 	if (dc->dc->res_pool->dmcu) {
-- 
GitLab


From 8fa33bd8d327ae2f9b602cd883f32efc4662bea0 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 15:48:48 -0400
Subject: [PATCH 0215/1791] drm/amd/display: Do not clear GPINT register when
 releasing DMUB from reset

[Why & How]
There's no need to clear GPINT register for DMUB
when releasing it from reset. Fix that.

Fixes: ac2e555e0a7f ("drm/amd/display: Add DMCUB source files and changes for DCN32/321")
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index b45ac31ba5557..a7d5607459edb 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -137,7 +137,6 @@ void dmub_dcn32_reset(struct dmub_srv *dmub)
 
 void dmub_dcn32_reset_release(struct dmub_srv *dmub)
 {
-	REG_WRITE(DMCUB_GPINT_DATAIN1, 0);
 	REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 0);
 	REG_WRITE(DMCUB_SCRATCH15, dmub->psp_version & 0x001100FF);
 	REG_UPDATE_2(DMCUB_CNTL, DMCUB_ENABLE, 1, DMCUB_TRACEPORT_EN, 1);
-- 
GitLab


From 8b6a6aa5d6d2be6a0669a29deb89184aaa0bad65 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 15:59:45 -0400
Subject: [PATCH 0216/1791] drm/amd/display: Update bounding box values for
 DCN321

[Why&how]

Update bounding box values as per hardware spec

Fixes: 197485c69543 ("drm/amd/display: Create dcn321_fpu file")
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index 57b9bd8966789..342a1bcb49273 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -106,16 +106,16 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = {
 	.clock_limits = {
 		{
 			.state = 0,
-			.dcfclk_mhz = 1564.0,
-			.fabricclk_mhz = 400.0,
-			.dispclk_mhz = 2150.0,
-			.dppclk_mhz = 2150.0,
+			.dcfclk_mhz = 1434.0,
+			.fabricclk_mhz = 2250.0,
+			.dispclk_mhz = 1720.0,
+			.dppclk_mhz = 1720.0,
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
-			.phyclk_d32_mhz = 625.0,
+			.phyclk_d32_mhz = 313.0,
 			.socclk_mhz = 1200.0,
-			.dscclk_mhz = 716.667,
-			.dram_speed_mts = 1600.0,
+			.dscclk_mhz = 573.333,
+			.dram_speed_mts = 16000.0,
 			.dtbclk_mhz = 1564.0,
 		},
 	},
@@ -125,14 +125,14 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_21_soc = {
 	.sr_exit_z8_time_us = 285.0,
 	.sr_enter_plus_exit_z8_time_us = 320,
 	.writeback_latency_us = 12.0,
-	.round_trip_ping_latency_dcfclk_cycles = 263,
+	.round_trip_ping_latency_dcfclk_cycles = 207,
 	.urgent_latency_pixel_data_only_us = 4,
 	.urgent_latency_pixel_mixed_with_vm_data_us = 4,
 	.urgent_latency_vm_data_only_us = 4,
-	.fclk_change_latency_us = 20,
-	.usr_retraining_latency_us = 2,
-	.smn_latency_us = 2,
-	.mall_allocated_for_dcn_mbytes = 64,
+	.fclk_change_latency_us = 7,
+	.usr_retraining_latency_us = 0,
+	.smn_latency_us = 0,
+	.mall_allocated_for_dcn_mbytes = 32,
 	.urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096,
 	.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096,
 	.urgent_out_of_order_return_per_channel_vm_only_bytes = 4096,
-- 
GitLab


From 7abac457ba75e3c6f6468036ad6f424bc56b2750 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 19 Apr 2023 18:40:00 +0530
Subject: [PATCH 0217/1791] drm/amd/amdgpu: Fix style issues in
 amdgpu_discovery.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix following checkpatch errors in amdgpu_discovery.c

ERROR: space required after that ',' (ctx:VxV)
ERROR: space required before the open parenthesis '('
ERROR: code indent should use tabs where possible

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 0ecce0b92b82e..0ba013275dc17 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -311,7 +311,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin) || amdgpu_discovery == 2) {
 		/* ignore the discovery binary from vram if discovery=2 in kernel module parameter */
 		if (amdgpu_discovery == 2)
-			dev_info(adev->dev,"force read ip discovery binary from file");
+			dev_info(adev->dev, "force read ip discovery binary from file");
 		else
 			dev_warn(adev->dev, "get invalid ip discovery binary signature from vram\n");
 
@@ -323,7 +323,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 			goto out;
 		}
 		/* check the ip discovery binary signature */
-		if(!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) {
+		if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) {
 			dev_warn(adev->dev, "get invalid ip discovery binary signature from file\n");
 			r = -EINVAL;
 			goto out;
@@ -529,8 +529,8 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev,
 					break;
 				default:
 					break;
-                                }
-                        }
+				}
+			}
 next_ip:
 			ip_offset += struct_size(ip, base_address, ip->num_base_address);
 		}
-- 
GitLab


From 92d1fe5954dc28c4e0260b730dd79d2acfdfa29f Mon Sep 17 00:00:00 2001
From: Dillon Varone <dillon.varone@amd.com>
Date: Thu, 6 Apr 2023 16:10:04 -0400
Subject: [PATCH 0218/1791] drm/amd/display: add support for low bpc

[WHY&HOW]
Low bpc timings are failing validation, port a patch to allow them to pass.

Signed-off-by: Dillon Varone <dillon.varone@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/dml/dcn32/display_mode_vba_util_32.c        | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index 61cc4904ade41..cad2bc3aea672 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -1595,7 +1595,6 @@ double dml32_TruncToValidBPP(
 	unsigned int   NonDSCBPP0;
 	unsigned int   NonDSCBPP1;
 	unsigned int   NonDSCBPP2;
-	unsigned int   NonDSCBPP3;
 
 	if (Format == dm_420) {
 		NonDSCBPP0 = 12;
@@ -1604,10 +1603,9 @@ double dml32_TruncToValidBPP(
 		MinDSCBPP = 6;
 		MaxDSCBPP = 1.5 * DSCInputBitPerComponent - 1 / 16;
 	} else if (Format == dm_444) {
-		NonDSCBPP0 = 18;
-		NonDSCBPP1 = 24;
-		NonDSCBPP2 = 30;
-		NonDSCBPP3 = 36;
+		NonDSCBPP0 = 24;
+		NonDSCBPP1 = 30;
+		NonDSCBPP2 = 36;
 		MinDSCBPP = 8;
 		MaxDSCBPP = 3 * DSCInputBitPerComponent - 1.0 / 16;
 	} else {
@@ -1661,9 +1659,7 @@ double dml32_TruncToValidBPP(
 			else
 				return dml_floor(16.0 * MaxLinkBPP, 1.0) / 16.0;
 		} else {
-			if (MaxLinkBPP >= NonDSCBPP3)
-				return NonDSCBPP3;
-			else if (MaxLinkBPP >= NonDSCBPP2)
+			if (MaxLinkBPP >= NonDSCBPP2)
 				return NonDSCBPP2;
 			else if (MaxLinkBPP >= NonDSCBPP1)
 				return NonDSCBPP1;
@@ -1674,7 +1670,7 @@ double dml32_TruncToValidBPP(
 		}
 	} else {
 		if (!((DSCEnable == false && (DesiredBPP == NonDSCBPP2 || DesiredBPP == NonDSCBPP1 ||
-				DesiredBPP == NonDSCBPP0 || DesiredBPP == NonDSCBPP3)) ||
+				DesiredBPP <= NonDSCBPP0)) ||
 				(DSCEnable && DesiredBPP >= MinDSCBPP && DesiredBPP <= MaxDSCBPP)))
 			return BPP_INVALID;
 		else
-- 
GitLab


From a1f1fecd04f0b9ef600898c7f9b2094504127fd7 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 16:16:33 -0400
Subject: [PATCH 0219/1791] drm/amd/display: Set DRAM clock if retraining is
 required

Set DRAM clock change state if retraining is required.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index cad2bc3aea672..d39e77d95fc30 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -4338,7 +4338,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport(
 				+ v->WritebackChunkSize * 1024.0 / 32.0 / SOCCLK;
 	}
 	if (v->USRRetrainingRequiredFinal)
-		v->Watermark.WritebackUrgentWatermark = v->Watermark.WritebackUrgentWatermark
+		v->Watermark.WritebackDRAMClockChangeWatermark = v->Watermark.WritebackDRAMClockChangeWatermark
 				+ mmSOCParameters.USRRetrainingLatency;
 
 	if (TotalActiveWriteback <= 1) {
-- 
GitLab


From 974ce18160bb16c8addf6889fff977564404b87b Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 16:19:16 -0400
Subject: [PATCH 0220/1791] drm/amd/display: Add check for PState change in
 DCN32

For pstate change, allow DML to loop through
all possible prefetch combinations so as to
support more display configurations. Set the max
and min prefetch modes to enable the sequence.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c   | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index d39e77d95fc30..a50e7f4dce421 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -4656,6 +4656,10 @@ void dml32_CalculateMinAndMaxPrefetchMode(
 	} else if (AllowForPStateChangeOrStutterInVBlankFinal == dm_prefetch_support_uclk_fclk_and_stutter) {
 		*MinPrefetchMode = 0;
 		*MaxPrefetchMode = 0;
+	} else if (AllowForPStateChangeOrStutterInVBlankFinal ==
+			dm_prefetch_support_uclk_fclk_and_stutter_if_possible) {
+		*MinPrefetchMode = 0;
+		*MaxPrefetchMode = 3;
 	} else {
 		*MinPrefetchMode = 0;
 		*MaxPrefetchMode = 3;
-- 
GitLab


From 2c30f85551211a48f7be57499a6fcb4ad00875da Mon Sep 17 00:00:00 2001
From: Jasdeep Dhillon <jasdeep.dhillon@amd.com>
Date: Tue, 28 Feb 2023 11:46:31 -0500
Subject: [PATCH 0221/1791] drm/amd/display: Isolate remaining FPU code in
 DCN32

[Why]
DCN32 resource contains code that uses FPU.

[How]
Moved code into DCN32 FPU

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Jasdeep Dhillon <jasdeep.dhillon@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h | 2 ++
 2 files changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 2f7723053042f..b8a2518faecc7 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2881,3 +2881,9 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint
 	}
 	return vactive_found;
 }
+
+void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb)
+{
+	dc_assert_fp_enabled();
+	dcn3_2_soc.clock_limits[0].dcfclk_mhz = 1200.0;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
index 9a0806a0e2ef0..dcf512cd30721 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
@@ -80,4 +80,6 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co
 
 bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req);
 
+void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb);
+
 #endif
-- 
GitLab


From 44243719117171e947317a1aca4625c317f7dcb0 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Date: Thu, 30 Apr 2020 15:38:04 -0400
Subject: [PATCH 0222/1791] drm/amd/display: Limit nv21 dst_y

Dst_y can become negative in extreme odm 4to1 cases. While not strictly
invalid, this should be limited to 0 for rq/dlg/ttu calculation.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
index cd3cfcb2a2b01..0497a5d74a62f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
@@ -980,7 +980,7 @@ static void dml_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 
 	unsigned int vstartup_start = 0;
 	unsigned int dst_x_after_scaler = 0;
-	unsigned int dst_y_after_scaler = 0;
+	int dst_y_after_scaler = 0;
 	double line_wait = 0;
 	double dst_y_prefetch = 0;
 	double dst_y_per_vm_vblank = 0;
@@ -1171,6 +1171,8 @@ static void dml_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib,
 
 	dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
 	dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);
+	if (dst_y_after_scaler < 0)
+		dst_y_after_scaler = 0;
 
 	// do some adjustment on the dst_after scaler to account for odm combine mode
 	dml_print("DML_DLG: %s: input dst_x_after_scaler                     = %d\n",
-- 
GitLab


From 05ffbdf4dbd55b9526535bddddf40dafdc2b27d4 Mon Sep 17 00:00:00 2001
From: Sherry Wang <Yao.Wang1@amd.com>
Date: Sun, 9 Oct 2022 17:05:51 +0800
Subject: [PATCH 0223/1791] drm/amd/display: correct DML calc error

[Why]
DML calculation is different from HW formula.

[How]
Correct the bug to keep it same as HW formula.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Sherry Wang <Yao.Wang1@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c    | 4 ++--
 .../gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c    | 4 ++--
 .../gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c  | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index 7d0626e42ea68..dea2b84e5ebee 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -4939,8 +4939,8 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 					}
 					v->TotImmediateFlipBytes = 0.0;
 					for (k = 0; k <= v->NumberOfActivePlanes - 1; k++) {
-						v->TotImmediateFlipBytes = v->TotImmediateFlipBytes + v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k]
-								+ v->MetaRowBytes[i][j][k] + v->DPTEBytesPerRow[i][j][k];
+						v->TotImmediateFlipBytes = v->TotImmediateFlipBytes + v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k]
+								+ v->MetaRowBytes[i][j][k] + v->DPTEBytesPerRow[i][j][k]);
 					}
 
 					for (k = 0; k <= v->NumberOfActivePlanes - 1; k++) {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index bd674dc30df33..330b089d6a86f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -5274,8 +5274,8 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 					v->TotImmediateFlipBytes = 0.0;
 					for (k = 0; k < v->NumberOfActivePlanes; k++) {
 						v->TotImmediateFlipBytes = v->TotImmediateFlipBytes
-								+ v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k]
-								+ v->DPTEBytesPerRow[i][j][k];
+								+ v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k]
+								+ v->DPTEBytesPerRow[i][j][k]);
 					}
 
 					for (k = 0; k < v->NumberOfActivePlanes; k++) {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 7eb2173b7691e..27b83162ae452 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -5371,8 +5371,8 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_
 					v->TotImmediateFlipBytes = 0.0;
 					for (k = 0; k < v->NumberOfActivePlanes; k++) {
 						v->TotImmediateFlipBytes = v->TotImmediateFlipBytes
-								+ v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k]
-								+ v->DPTEBytesPerRow[i][j][k];
+								+ v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k]
+								+ v->DPTEBytesPerRow[i][j][k]);
 					}
 
 					for (k = 0; k < v->NumberOfActivePlanes; k++) {
-- 
GitLab


From 0244b0f7d5ac2b6a8d5716f6dd2fb2631f13a3b5 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 16:46:30 -0400
Subject: [PATCH 0224/1791] drm/amd/display: Add extra check for 444 16 format

DCN30 is missing a check for the pixel format 444 when using 16bits
before setting the flag that Viewport exceeds the surface.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index dea2b84e5ebee..9af1a43c042bc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -5130,7 +5130,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 			ViewportExceedsSurface = true;
 
 		if (v->SourcePixelFormat[k] != dm_444_64 && v->SourcePixelFormat[k] != dm_444_32 && v->SourcePixelFormat[k] != dm_444_16
-				&& v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
+				&& v->SourcePixelFormat[k] != dm_444_16 && v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
 			if (v->ViewportWidthChroma[k] > v->SurfaceWidthC[k] || v->ViewportHeightChroma[k] > v->SurfaceHeightC[k]) {
 				ViewportExceedsSurface = true;
 			}
-- 
GitLab


From 72a9be2f444a9fd4ab34e6329fa1889dbd3a7ef3 Mon Sep 17 00:00:00 2001
From: Krunoslav Kovac <Krunoslav.Kovac@amd.com>
Date: Fri, 11 Mar 2022 16:12:58 -0500
Subject: [PATCH 0225/1791] drm/amd/display: 3-plane MPO enablement for DCN321

Enable 3-planes MPO for DCN321 by reporting max_slave_planes in DC caps
for each ASIC.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Krunoslav Kovac <Krunoslav.Kovac@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index cf21b240fc55e..138657c38afe4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -1727,9 +1727,9 @@ static bool dcn321_resource_construct(
 	dc->caps.subvp_pstate_allow_width_us = 20;
 	dc->caps.subvp_vertical_int_margin_us = 30;
 	dc->caps.subvp_drr_vblank_start_margin_us = 100; // 100us margin
-	dc->caps.max_slave_planes = 1;
-	dc->caps.max_slave_yuv_planes = 1;
-	dc->caps.max_slave_rgb_planes = 1;
+	dc->caps.max_slave_planes = 2;
+	dc->caps.max_slave_yuv_planes = 2;
+	dc->caps.max_slave_rgb_planes = 2;
 	dc->caps.post_blend_color_processing = true;
 	dc->caps.force_dp_tps4_for_cp2520 = true;
 	dc->caps.dp_hpo = true;
-- 
GitLab


From 4335077a76095ff75dc0ffb031aeae93f9f5e80f Mon Sep 17 00:00:00 2001
From: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Date: Wed, 12 Jan 2022 19:58:04 -0500
Subject: [PATCH 0226/1791] drm/amd/display: Adjust dmub outbox notification
 enable

[Why]
Currently driver enables dmub outbox notification before oubox ISR is
registered. During boot scenario, sometimes dmub issues hpd outbox
message before driver registers ISR and those messages are missed.

[How]
Enable dmub outbox notification after outbox ISR is registered. Also,
restructured outbox enable code to call from dm layer and renamed APIs.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index e0c74868d2ee5..890268d954958 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -197,10 +197,6 @@ void dcn31_init_hw(struct dc *dc)
 		}
 	}
 
-	/* Enables outbox notifications for usb4 dpia */
-	if (dc->res_pool->usb4_dpia_count)
-		dmub_enable_outbox_notification(dc->ctx->dmub_srv);
-
 	/* we want to turn off all dp displays before doing detection */
 	dc->link_srv->blank_all_dp_displays(dc);
 
-- 
GitLab


From 3ac73f1ef2b403048c98fdf0f29ba32571efb849 Mon Sep 17 00:00:00 2001
From: Igor Kravchenko <Igor.Kravchenko@amd.com>
Date: Fri, 10 Jul 2020 16:24:30 -0400
Subject: [PATCH 0227/1791] drm/amd/display: Set min_width and min_height
 capability for DCN30

Add min_width, min_height fields to dc_plane_cap structure. Set values
to 16x16 for discrete ASICs, and 64x64 for others.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Igor Kravchenko <Igor.Kravchenko@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index 965f5ceb33f76..67a34cda3774a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -701,7 +701,9 @@ static const struct dc_plane_cap plane_cap = {
 			.argb8888 = 167,
 			.nv12 = 167,
 			.fp16 = 167
-	}
+	},
+	16,
+	16
 };
 
 static const struct dc_debug_options debug_defaults_drv = {
-- 
GitLab


From 255ce8f7c169fa385ad04f01c7babab5f636a3ce Mon Sep 17 00:00:00 2001
From: Po-Ting Chen <robin.chen@amd.com>
Date: Fri, 26 Feb 2021 15:48:02 +0800
Subject: [PATCH 0228/1791] drm/amd/display: update GSP1 generic info packet
 for PSRSU

Base on PSRSU specification, every seletive update frame need to use two
SDP to indicate the frame active range. So we occupy another GSP1 for
PSRSU execution.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Po-Ting Chen <robin.chen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dcn30/dcn30_dio_stream_encoder.c   | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
index 9d08127d209b8..005dbe099a7a7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c
@@ -436,6 +436,21 @@ void enc3_stream_encoder_update_dp_info_packets(
 				&info_frame->vsc,
 				true);
 	}
+	/* TODO: VSC SDP at packetIndex 1 should be retricted only if PSR-SU on.
+	 * There should have another Infopacket type (e.g. vsc_psrsu) for PSR_SU.
+	 * In addition, currently the driver check the valid bit then update and
+	 * send the corresponding Infopacket. For PSR-SU, the SDP only be sent
+	 * while entering PSR-SU mode. So we need another parameter(e.g. send)
+	 * in dc_info_packet to indicate which infopacket should be enabled by
+	 * default here.
+	 */
+	if (info_frame->vsc.valid) {
+		enc->vpg->funcs->update_generic_info_packet(
+				enc->vpg,
+				1,  /* packetIndex */
+				&info_frame->vsc,
+				true);
+	}
 	/* TODO: VSC SDP at packetIndex 1 should be restricted only if PSR-SU on.
 	 * There should have another Infopacket type (e.g. vsc_psrsu) for PSR_SU.
 	 * In addition, currently the driver check the valid bit then update and
-- 
GitLab


From 58e67bb3c131da5ee14e4842b08e53f4888dce0a Mon Sep 17 00:00:00 2001
From: Zhongwei <Zhongwei.Zhang@amd.com>
Date: Fri, 28 Oct 2022 13:40:31 +0800
Subject: [PATCH 0229/1791] drm/amd/display: fix dpms_off issue when disabling
 bios mode

[Why]
disable_vbios_mode_if_required() will set dpms_off to false during boot
when pixel clk dismatches with driver requires. This will cause extra
backlight on and off if OS call 2 times setmode.

[How]
Set dpms_off to true to keep power_off and let OS control backlight by
display's powerState.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Zhongwei <Zhongwei.Zhang@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 631c6b10562e1..308b604cb7919 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1268,7 +1268,7 @@ static void disable_vbios_mode_if_required(
 
 					if (pix_clk_100hz != requested_pix_clk_100hz) {
 						dc->link_srv->set_dpms_off(pipe);
-						pipe->stream->dpms_off = false;
+						pipe->stream->dpms_off = true;
 					}
 				}
 			}
-- 
GitLab


From 710cc1e7cd461446a9325c9bd1e9a54daa462952 Mon Sep 17 00:00:00 2001
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Date: Thu, 2 May 2019 13:21:48 -0400
Subject: [PATCH 0230/1791] drm/amd/display: Explicitly specify update type per
 plane info change

[Why]
The bit for flip addr is being set causing the determination for
FAST vs MEDIUM to always return MEDIUM when plane info is provided
as a surface update. This causes extreme stuttering for the typical
atomic update path on Linux.

[How]
Don't use update_flags->raw for determining FAST vs MEDIUM. It's too
fragile to changes like this.

Explicitly specify the update type per update flag instead. It's not
as clever as checking the bits itself but at least it's correct.

Fixes: aa5fdb1ab5b6 ("drm/amd/display: Explicitly specify update type per plane info change")
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 308b604cb7919..c10243ef93b6b 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2482,9 +2482,6 @@ static enum surface_update_type det_surface_update(const struct dc *dc,
 	enum surface_update_type overall_type = UPDATE_TYPE_FAST;
 	union surface_update_flags *update_flags = &u->surface->update_flags;
 
-	if (u->flip_addr)
-		update_flags->bits.addr_update = 1;
-
 	if (!is_surface_in_context(context, u->surface) || u->surface->force_full_update) {
 		update_flags->raw = 0xFFFFFFFF;
 		return UPDATE_TYPE_FULL;
-- 
GitLab


From 4f63b7a59926eb7fb50091e796170a10a8ef4091 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 17:08:53 -0400
Subject: [PATCH 0231/1791] drm/amd/display: Add FAMS capability to DCN31

DCN31 supports FAMS, but this was not correctly set to the hardware
setup sequence. This commit fixes this issue by setting the MCLK switch
capability based on the feature capability retrieved from the DMUB.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index 890268d954958..55494730e5008 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -295,6 +295,7 @@ void dcn31_init_hw(struct dc *dc)
 	// Get DMCUB capabilities
 	dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 	dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
+	dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 }
 
 void dcn31_dsc_pg_control(
-- 
GitLab


From f828b681d0cd566f86351c0b913e6cb6ed8c7b9c Mon Sep 17 00:00:00 2001
From: hackyzh002 <hackyzh002@gmail.com>
Date: Wed, 19 Apr 2023 20:20:58 +0800
Subject: [PATCH 0232/1791] drm/radeon: Fix integer overflow in
 radeon_cs_parser_init
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The type of size is unsigned, if size is 0x40000000, there will be an
integer overflow, size will be zero after size *= sizeof(uint32_t),
will cause uninitialized memory to be referenced later

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: hackyzh002 <hackyzh002@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_cs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 46a27ebf4588a..a6700d7278bf3 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -270,7 +270,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 {
 	struct drm_radeon_cs *cs = data;
 	uint64_t *chunk_array_ptr;
-	unsigned size, i;
+	u64 size;
+	unsigned i;
 	u32 ring = RADEON_CS_RING_GFX;
 	s32 priority = 0;
 
-- 
GitLab


From 87c2213e85bd81e4a9a4d0880c256568794ae388 Mon Sep 17 00:00:00 2001
From: hackyzh002 <hackyzh002@gmail.com>
Date: Wed, 19 Apr 2023 20:22:33 +0800
Subject: [PATCH 0233/1791] drm/amdgpu: Fix integer overflow in amdgpu_cs_pass1
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The type of size is unsigned int, if size is 0x40000000, there will
be an integer overflow, size will be zero after size *= sizeof(uint32_t),
will cause uninitialized memory to be referenced later.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: hackyzh002 <hackyzh002@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index d8b3c9198d332..a8ec5ff41fc05 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -195,7 +195,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 	uint64_t *chunk_array_user;
 	uint64_t *chunk_array;
 	uint32_t uf_offset = 0;
-	unsigned int size;
+	size_t size;
 	int ret;
 	int i;
 
-- 
GitLab


From 053065a43ca6466575d9d3c9110e305fdcf303d3 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 6 Apr 2023 15:01:32 -0400
Subject: [PATCH 0234/1791] drm/amd/display: Add FAMS related definitions and
 documenation for enum fields

[Why&How]
Add Enum and documenation related to FAMS (Firmware Assisted Memclk
Switching) and CAB (Cache As Buffer)

Reviewed-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 54b7786f56819..b32a5c977d171 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -376,6 +376,7 @@ enum dmub_fw_boot_status_bit {
 	DMUB_FW_BOOT_STATUS_BIT_OPTIMIZED_INIT_DONE = (1 << 2), /**< 1 if init done */
 	DMUB_FW_BOOT_STATUS_BIT_RESTORE_REQUIRED = (1 << 3), /**< 1 if driver should call restore */
 	DMUB_FW_BOOT_STATUS_BIT_DEFERRED_LOADED = (1 << 4), /**< 1 if VBIOS data is deferred programmed */
+	DMUB_FW_BOOT_STATUS_BIT_FAMS_ENABLED = (1 << 5), /**< 1 if FAMS is enabled*/
 	DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED = (1 << 6), /**< 1 if detection need to be triggered by driver*/
 	DMUB_FW_BOOT_STATUS_BIT_HW_POWER_INIT_DONE = (1 << 7), /**< 1 if hw power init is completed */
 };
@@ -989,16 +990,25 @@ struct dmub_rb_cmd_mall {
 };
 
 /**
- * enum dmub_cmd_cab_type - TODO:
+ * enum dmub_cmd_cab_type - CAB command data.
  */
 enum dmub_cmd_cab_type {
+	/**
+	 * No idle optimizations (i.e. no CAB)
+	 */
 	DMUB_CMD__CAB_NO_IDLE_OPTIMIZATION = 0,
+	/**
+	 * No DCN requests for memory
+	 */
 	DMUB_CMD__CAB_NO_DCN_REQ = 1,
+	/**
+	 * Fit surfaces in CAB (i.e. CAB enable)
+	 */
 	DMUB_CMD__CAB_DCN_SS_FIT_IN_CAB = 2,
 };
 
 /**
- * struct dmub_rb_cmd_cab_for_ss - TODO:
+ * struct dmub_rb_cmd_cab - CAB command data.
  */
 struct dmub_rb_cmd_cab_for_ss {
 	struct dmub_cmd_header header;
@@ -1006,6 +1016,9 @@ struct dmub_rb_cmd_cab_for_ss {
 	uint8_t debug_bits;     /* debug bits */
 };
 
+/**
+ * Enum for indicating which MCLK switch mode per pipe
+ */
 enum mclk_switch_mode {
 	NONE = 0,
 	FPO = 1,
-- 
GitLab


From 4371fa7795dabb422171de5e8d9fe7aa3e2edb86 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 2 Apr 2023 22:20:20 -0400
Subject: [PATCH 0235/1791] drm/amd/display: 3.2.231

This DC version brings along:
- FW Release 0.0.162.0
- Enable FPO+Vactivate
- Support for VESA SCR
- Refactor DMUB commands
- Fixes in secure display, modeset, memleak and more
- Picked up missed patches in history

Acked-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index edef5d181590f..b45974a2dec34 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.230"
+#define DC_VER "3.2.231"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From 0026c273e68ee82a7843f5de26147357bc7e6551 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 20 Apr 2023 10:47:01 +0200
Subject: [PATCH 0236/1791] drm/amd/display: dumb_abm_lcd: avoid
 missing-prototype warnings

The dmub_abm_set_ambient_level() function has no caller and can
just be removed, the other ones have a declaration in the
header file and just need to see the prototype:

drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:122:14: error: no previous prototype for function 'dmub_abm_get_current_backlight' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:133:14: error: no previous prototype for function 'dmub_abm_get_target_backlight' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:144:6: error: no previous prototype for function 'dmub_abm_set_level' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:163:6: error: no previous prototype for function 'dmub_abm_set_ambient_level' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:183:6: error: no previous prototype for function 'dmub_abm_init_config' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:213:6: error: no previous prototype for function 'dmub_abm_set_pause' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:231:6: error: no previous prototype for function 'dmub_abm_set_pipe' [-Werror,-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dmub_abm_lcd.c:251:6: error: no previous prototype for function 'dmub_abm_set_backlight_level' [-Werror,-Wmissing-prototypes]

Fixes: b8fe56375f78 ("drm/amd/display: Refactor ABM feature")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c | 22 +------------------
 1 file changed, 1 insertion(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
index e152c68edfd1e..39da73eba86e5 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c
@@ -24,6 +24,7 @@
  */
 
 #include "dmub_abm.h"
+#include "dmub_abm_lcd.h"
 #include "dce_abm.h"
 #include "dc.h"
 #include "dc_dmub_srv.h"
@@ -159,27 +160,6 @@ bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask)
 	return true;
 }
 
-#ifndef TRIM_AMBIENT_GAMMA
-void dmub_abm_set_ambient_level(struct abm *abm, unsigned int ambient_lux, uint8_t panel_mask)
-{
-	union dmub_rb_cmd cmd;
-	struct dc_context *dc = abm->ctx;
-
-	if (ambient_lux > 0xFFFF)
-		ambient_lux = 0xFFFF;
-
-	memset(&cmd, 0, sizeof(cmd));
-	cmd.abm_set_ambient_level.header.type = DMUB_CMD__ABM;
-	cmd.abm_set_ambient_level.header.sub_type = DMUB_CMD__ABM_SET_AMBIENT_LEVEL;
-	cmd.abm_set_ambient_level.abm_set_ambient_level_data.ambient_lux = ambient_lux;
-	cmd.abm_set_ambient_level.abm_set_ambient_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1;
-	cmd.abm_set_ambient_level.abm_set_ambient_level_data.panel_mask = panel_mask;
-	cmd.abm_set_ambient_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_ambient_level_data);
-
-	dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
-}
-#endif
-
 void dmub_abm_init_config(struct abm *abm,
 	const char *src,
 	unsigned int bytes,
-- 
GitLab


From 8f7f1b020ea641f21e7c88db55170a4d15834668 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Fri, 14 Apr 2023 14:26:27 -0400
Subject: [PATCH 0237/1791] drm/amd/display: fix flickering caused by S/G mode

Currently, on a handful of ASICs. We allow the framebuffer for a given
plane to exist in either VRAM or GTT. However, if the plane's new
framebuffer is in a different memory domain than it's previous
framebuffer, flipping between them can cause the screen to flicker. So,
to fix this, don't perform an immediate flip in the aforementioned case.

Cc: stable@vger.kernel.org
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2354
Reviewed-by: Roman Li <Roman.Li@amd.com>
Fixes: 81d0bcf99009 ("drm/amdgpu: make display pinning more flexible (v2)")
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 73b52bf76461f..8a3c04ff9bffd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7901,6 +7901,13 @@ static void amdgpu_dm_commit_cursors(struct drm_atomic_state *state)
 			amdgpu_dm_plane_handle_cursor_update(plane, old_plane_state);
 }
 
+static inline uint32_t get_mem_type(struct drm_framebuffer *fb)
+{
+	struct amdgpu_bo *abo = gem_to_amdgpu_bo(fb->obj[0]);
+
+	return abo->tbo.resource ? abo->tbo.resource->mem_type : 0;
+}
+
 static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 				    struct dc_state *dc_state,
 				    struct drm_device *dev,
@@ -8043,11 +8050,13 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 
 		/*
 		 * Only allow immediate flips for fast updates that don't
-		 * change FB pitch, DCC state, rotation or mirroing.
+		 * change memory domain, FB pitch, DCC state, rotation or
+		 * mirroring.
 		 */
 		bundle->flip_addrs[planes_count].flip_immediate =
 			crtc->state->async_flip &&
-			acrtc_state->update_type == UPDATE_TYPE_FAST;
+			acrtc_state->update_type == UPDATE_TYPE_FAST &&
+			get_mem_type(old_plane_state->fb) == get_mem_type(fb);
 
 		timestamp_ns = ktime_get_ns();
 		bundle->flip_addrs[planes_count].flip_timestamp_in_us = div_u64(timestamp_ns, 1000);
-- 
GitLab


From 7787af256504b13f0ab6c311ed7870a895b762b8 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Thu, 20 Apr 2023 12:23:49 +0200
Subject: [PATCH 0238/1791] drm/i915/mtl: Set has_llc=0

On MTL, LLC is not shared between GT and CPU, set has_llc=0.

Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420102349.15302-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/i915_pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index cddb6e1979723..025d32c0b161b 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1146,6 +1146,7 @@ static const struct intel_device_info mtl_info = {
 	.has_flat_ccs = 0,
 	.has_gmd_id = 1,
 	.has_guc_deprivilege = 1,
+	.has_llc = 0,
 	.has_mslice_steering = 0,
 	.has_snoop = 1,
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM,
-- 
GitLab


From 4b4b96826ba980fef5bbb9caabcff005c6a60294 Mon Sep 17 00:00:00 2001
From: Richard Leitner <richard.leitner@skidata.com>
Date: Mon, 13 Mar 2023 08:50:16 +0100
Subject: [PATCH 0239/1791] dt-bindings: display: simple: add support for
 InnoLux G070ACE-L01

Add Innolux G070ACE-L01 7" WVGA (800x480) TFT LCD panel compatible
string.

Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230201-innolux-g070ace-v2-1-2371e251dd40@skidata.com
---
 .../devicetree/bindings/display/panel/panel-simple.yaml         | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index 01560fe226dd1..ec50dd2683149 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -174,6 +174,8 @@ properties:
       - innolux,at043tn24
         # Innolux AT070TN92 7.0" WQVGA TFT LCD panel
       - innolux,at070tn92
+        # Innolux G070ACE-L01 7" WVGA (800x480) TFT LCD panel
+      - innolux,g070ace-l01
         # Innolux G070Y2-L01 7" WVGA (800x480) TFT LCD panel
       - innolux,g070y2-l01
         # Innolux G070Y2-T02 7" WVGA (800x480) TFT LCD TTL panel
-- 
GitLab


From 1993f598998d7419f5c9fa8459f4919cf5f3643e Mon Sep 17 00:00:00 2001
From: Richard Leitner <richard.leitner@skidata.com>
Date: Mon, 13 Mar 2023 08:50:17 +0100
Subject: [PATCH 0240/1791] drm/panel: simple: Add InnoLux G070ACE-L01

Add InnoLux G070ACE-L01 7" 800x480 TFT LCD with WLED backlight panel
support. Timing data was extracted from datasheet and vendor provided
EDID file.

Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230201-innolux-g070ace-v2-2-2371e251dd40@skidata.com
---
 drivers/gpu/drm/panel/panel-simple.c | 35 ++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 065f378bba9d2..2a9c1a785a5cb 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -2142,6 +2142,38 @@ static const struct panel_desc innolux_at070tn92 = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
 };
 
+static const struct display_timing innolux_g070ace_l01_timing = {
+	.pixelclock = { 25200000, 35000000, 35700000 },
+	.hactive = { 800, 800, 800 },
+	.hfront_porch = { 30, 32, 87 },
+	.hback_porch = { 30, 32, 87 },
+	.hsync_len = { 1, 1, 1 },
+	.vactive = { 480, 480, 480 },
+	.vfront_porch = { 3, 3, 3 },
+	.vback_porch = { 13, 13, 13 },
+	.vsync_len = { 1, 1, 4 },
+	.flags = DISPLAY_FLAGS_DE_HIGH,
+};
+
+static const struct panel_desc innolux_g070ace_l01 = {
+	.timings = &innolux_g070ace_l01_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.delay = {
+		.prepare = 10,
+		.enable = 50,
+		.disable = 50,
+		.unprepare = 500,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct display_timing innolux_g070y2_l01_timing = {
 	.pixelclock = { 28000000, 29500000, 32000000 },
 	.hactive = { 800, 800, 800 },
@@ -4098,6 +4130,9 @@ static const struct of_device_id platform_of_match[] = {
 	}, {
 		.compatible = "innolux,at070tn92",
 		.data = &innolux_at070tn92,
+	}, {
+		.compatible = "innolux,g070ace-l01",
+		.data = &innolux_g070ace_l01,
 	}, {
 		.compatible = "innolux,g070y2-l01",
 		.data = &innolux_g070y2_l01,
-- 
GitLab


From 55b24786b74863b8f10e4d262e642749911cb3bb Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Date: Tue, 11 Apr 2023 18:09:43 -0600
Subject: [PATCH 0241/1791] drm/display: Add missing OLED Vesa brightnesses
 definitions

This commit adds missing luminance control registers to enable a more
standard way (VESA) to deal with eDP luminance control.

Cc: Anthony Koo <anthony.koo@amd.com>
Cc: Iswara Negulendran <iswara.nagulendran@amd.com>
Cc: Felipe Clark <felipe.clark@amd.com>
Cc: Harry Wentland <Harry.Wentland@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230412000943.429031-1-Rodrigo.Siqueira@amd.com
---
 include/drm/display/drm_dp.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h
index 358db4a9f167a..5caa99e410995 100644
--- a/include/drm/display/drm_dp.h
+++ b/include/drm/display/drm_dp.h
@@ -983,6 +983,7 @@
 
 #define DP_EDP_GENERAL_CAP_2		    0x703
 # define DP_EDP_OVERDRIVE_ENGINE_ENABLED		(1 << 0)
+# define DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE		(1 << 4)
 
 #define DP_EDP_GENERAL_CAP_3		    0x704    /* eDP 1.4 */
 # define DP_EDP_X_REGION_CAP_MASK			(0xf << 0)
@@ -1008,6 +1009,7 @@
 # define DP_EDP_DYNAMIC_BACKLIGHT_ENABLE		(1 << 4)
 # define DP_EDP_REGIONAL_BACKLIGHT_ENABLE		(1 << 5)
 # define DP_EDP_UPDATE_REGION_BRIGHTNESS		(1 << 6) /* eDP 1.4 */
+# define DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE		(1 << 7)
 
 #define DP_EDP_BACKLIGHT_BRIGHTNESS_MSB     0x722
 #define DP_EDP_BACKLIGHT_BRIGHTNESS_LSB     0x723
@@ -1032,6 +1034,7 @@
 
 #define DP_EDP_DBC_MINIMUM_BRIGHTNESS_SET   0x732
 #define DP_EDP_DBC_MAXIMUM_BRIGHTNESS_SET   0x733
+#define DP_EDP_PANEL_TARGET_LUMINANCE_VALUE 0x734
 
 #define DP_EDP_REGIONAL_BACKLIGHT_BASE      0x740    /* eDP 1.4 */
 #define DP_EDP_REGIONAL_BACKLIGHT_0	    0x741    /* eDP 1.4 */
-- 
GitLab


From 6b99658c93ce0ab84c4682a22df245ddefb0ebc5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mairacanal@riseup.net>
Date: Fri, 12 Aug 2022 17:57:42 -0300
Subject: [PATCH 0242/1791] drm/bridge: anx7625: Drop of_gpio header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This driver includes the deprecated OF GPIO header <linux/of_gpio.h>
yet fail to use symbols from it, so drop the include.

Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Cc: Neil Armstrong <narmstrong@baylibre.com>
Cc: Robert Foss <robert.foss@linaro.org>
Cc: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Reviewed-by: Robert Foss <robert.foss@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20220812205746.609107-2-mairacanal@riseup.net
---
 drivers/gpu/drm/bridge/analogix/anx7625.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c
index 6846199a2ee14..093e7b9a570f0 100644
--- a/drivers/gpu/drm/bridge/analogix/anx7625.c
+++ b/drivers/gpu/drm/bridge/analogix/anx7625.c
@@ -17,7 +17,6 @@
 #include <linux/types.h>
 #include <linux/workqueue.h>
 
-#include <linux/of_gpio.h>
 #include <linux/of_graph.h>
 #include <linux/of_platform.h>
 
-- 
GitLab


From e1964d892fe5710c16e55d2eb153954f8407fb9a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mairacanal@riseup.net>
Date: Fri, 12 Aug 2022 17:57:45 -0300
Subject: [PATCH 0243/1791] drm/sti: Drop of_gpio header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This driver includes the deprecated OF GPIO header <linux/of_gpio.h>
yet fail to use symbols from it, so drop this include.

Cc: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Acked-by: Alain Volmat <avolmat@me.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220812205746.609107-5-mairacanal@riseup.net
---
 drivers/gpu/drm/sti/sti_dvo.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index 577c477b5f467..0c6679e361c82 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -8,7 +8,7 @@
 #include <linux/component.h>
 #include <linux/debugfs.h>
 #include <linux/module.h>
-#include <linux/of_gpio.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 
 #include <drm/drm_atomic_helper.h>
-- 
GitLab


From b76c0deef6273609c02ed5053209f6397cd1b0fb Mon Sep 17 00:00:00 2001
From: Madhumitha Tolakanahalli Pradeep
 <madhumitha.tolakanahalli.pradeep@intel.com>
Date: Fri, 21 Apr 2023 20:25:34 +0200
Subject: [PATCH 0244/1791] drm/i915/mtl: Define MOCS and PAT tables for MTL

On MTL, GT can no longer allocate on LLC - only the CPU can.
This, along with programming new register bits that MTL
requires calls for a MOCS/PAT table update.
Also the PAT index registers are multicasted for primary GT,
and there is an address jump from index 7 to 8. This patch
makes sure that these registers are programmed in the proper
way.

BSpec: 44509, 45101, 44235

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Madhumitha Tolakanahalli Pradeep <madhumitha.tolakanahalli.pradeep@intel.com>
Signed-off-by: Aravind Iddamsetty <aravind.iddamsetty@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421182535.292670-2-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_regs.h |  6 ++-
 drivers/gpu/drm/i915/gt/intel_gtt.c     | 47 ++++++++++++++++-
 drivers/gpu/drm/i915/gt/intel_gtt.h     |  8 +++
 drivers/gpu/drm/i915/gt/intel_mocs.c    | 70 ++++++++++++++++++++++++-
 4 files changed, 128 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index fd1f9cd35e9d7..e8c3b762a92a3 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -356,7 +356,11 @@
 #define GEN7_TLB_RD_ADDR			_MMIO(0x4700)
 
 #define GEN12_PAT_INDEX(index)			_MMIO(0x4800 + (index) * 4)
-#define XEHP_PAT_INDEX(index)			MCR_REG(0x4800 + (index) * 4)
+#define _PAT_INDEX(index)			_PICK_EVEN_2RANGES(index, 8, \
+								   0x4800, 0x4804, \
+								   0x4848, 0x484c)
+#define XEHP_PAT_INDEX(index)			MCR_REG(_PAT_INDEX(index))
+#define XELPMP_PAT_INDEX(index)			_MMIO(_PAT_INDEX(index))
 
 #define XEHP_TILE0_ADDR_RANGE			MCR_REG(0x4900)
 #define   XEHP_TILE_LMEM_RANGE_SHIFT		8
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c
index 4f436ba7a3c83..2f6a9be0ffe61 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.c
@@ -468,6 +468,44 @@ void gtt_write_workarounds(struct intel_gt *gt)
 	}
 }
 
+static void xelpmp_setup_private_ppat(struct intel_uncore *uncore)
+{
+	intel_uncore_write(uncore, XELPMP_PAT_INDEX(0),
+			   MTL_PPAT_L4_0_WB);
+	intel_uncore_write(uncore, XELPMP_PAT_INDEX(1),
+			   MTL_PPAT_L4_1_WT);
+	intel_uncore_write(uncore, XELPMP_PAT_INDEX(2),
+			   MTL_PPAT_L4_3_UC);
+	intel_uncore_write(uncore, XELPMP_PAT_INDEX(3),
+			   MTL_PPAT_L4_0_WB | MTL_2_COH_1W);
+	intel_uncore_write(uncore, XELPMP_PAT_INDEX(4),
+			   MTL_PPAT_L4_0_WB | MTL_3_COH_2W);
+
+	/*
+	 * Remaining PAT entries are left at the hardware-default
+	 * fully-cached setting
+	 */
+}
+
+static void xelpg_setup_private_ppat(struct intel_gt *gt)
+{
+	intel_gt_mcr_multicast_write(gt, XEHP_PAT_INDEX(0),
+				     MTL_PPAT_L4_0_WB);
+	intel_gt_mcr_multicast_write(gt, XEHP_PAT_INDEX(1),
+				     MTL_PPAT_L4_1_WT);
+	intel_gt_mcr_multicast_write(gt, XEHP_PAT_INDEX(2),
+				     MTL_PPAT_L4_3_UC);
+	intel_gt_mcr_multicast_write(gt, XEHP_PAT_INDEX(3),
+				     MTL_PPAT_L4_0_WB | MTL_2_COH_1W);
+	intel_gt_mcr_multicast_write(gt, XEHP_PAT_INDEX(4),
+				     MTL_PPAT_L4_0_WB | MTL_3_COH_2W);
+
+	/*
+	 * Remaining PAT entries are left at the hardware-default
+	 * fully-cached setting
+	 */
+}
+
 static void tgl_setup_private_ppat(struct intel_uncore *uncore)
 {
 	/* TGL doesn't support LLC or AGE settings */
@@ -603,7 +641,14 @@ void setup_private_pat(struct intel_gt *gt)
 
 	GEM_BUG_ON(GRAPHICS_VER(i915) < 8);
 
-	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50))
+	if (gt->type == GT_MEDIA) {
+		xelpmp_setup_private_ppat(gt->uncore);
+		return;
+	}
+
+	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70))
+		xelpg_setup_private_ppat(gt);
+	else if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50))
 		xehp_setup_private_ppat(gt);
 	else if (GRAPHICS_VER(i915) >= 12)
 		tgl_setup_private_ppat(uncore);
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index 69ce55f517f56..ea17849e7a5c8 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -147,6 +147,14 @@ typedef u64 gen8_pte_t;
 #define GEN8_PDE_IPS_64K BIT(11)
 #define GEN8_PDE_PS_2M   BIT(7)
 
+#define MTL_PPAT_L4_CACHE_POLICY_MASK	REG_GENMASK(3, 2)
+#define MTL_PAT_INDEX_COH_MODE_MASK	REG_GENMASK(1, 0)
+#define MTL_PPAT_L4_3_UC	REG_FIELD_PREP(MTL_PPAT_L4_CACHE_POLICY_MASK, 3)
+#define MTL_PPAT_L4_1_WT	REG_FIELD_PREP(MTL_PPAT_L4_CACHE_POLICY_MASK, 1)
+#define MTL_PPAT_L4_0_WB	REG_FIELD_PREP(MTL_PPAT_L4_CACHE_POLICY_MASK, 0)
+#define MTL_3_COH_2W	REG_FIELD_PREP(MTL_PAT_INDEX_COH_MODE_MASK, 3)
+#define MTL_2_COH_1W	REG_FIELD_PREP(MTL_PAT_INDEX_COH_MODE_MASK, 2)
+
 enum i915_cache_level;
 
 struct drm_i915_gem_object;
diff --git a/drivers/gpu/drm/i915/gt/intel_mocs.c b/drivers/gpu/drm/i915/gt/intel_mocs.c
index 69b489e8dfeda..2c014407225cc 100644
--- a/drivers/gpu/drm/i915/gt/intel_mocs.c
+++ b/drivers/gpu/drm/i915/gt/intel_mocs.c
@@ -40,6 +40,10 @@ struct drm_i915_mocs_table {
 #define LE_COS(value)		((value) << 15)
 #define LE_SSE(value)		((value) << 17)
 
+/* Defines for the tables (GLOB_MOCS_0 - GLOB_MOCS_16) */
+#define _L4_CACHEABILITY(value)	((value) << 2)
+#define IG_PAT(value)		((value) << 8)
+
 /* Defines for the tables (LNCFMOCS0 - LNCFMOCS31) - two entries per word */
 #define L3_ESC(value)		((value) << 0)
 #define L3_SCC(value)		((value) << 1)
@@ -50,6 +54,7 @@ struct drm_i915_mocs_table {
 /* Helper defines */
 #define GEN9_NUM_MOCS_ENTRIES	64  /* 63-64 are reserved, but configured. */
 #define PVC_NUM_MOCS_ENTRIES	3
+#define MTL_NUM_MOCS_ENTRIES	16
 
 /* (e)LLC caching options */
 /*
@@ -73,6 +78,12 @@ struct drm_i915_mocs_table {
 #define L3_2_RESERVED		_L3_CACHEABILITY(2)
 #define L3_3_WB			_L3_CACHEABILITY(3)
 
+/* L4 caching options */
+#define L4_0_WB			_L4_CACHEABILITY(0)
+#define L4_1_WT			_L4_CACHEABILITY(1)
+#define L4_2_RESERVED		_L4_CACHEABILITY(2)
+#define L4_3_UC			_L4_CACHEABILITY(3)
+
 #define MOCS_ENTRY(__idx, __control_value, __l3cc_value) \
 	[__idx] = { \
 		.control_value = __control_value, \
@@ -416,6 +427,57 @@ static const struct drm_i915_mocs_entry pvc_mocs_table[] = {
 	MOCS_ENTRY(2, 0, L3_3_WB),
 };
 
+static const struct drm_i915_mocs_entry mtl_mocs_table[] = {
+	/* Error - Reserved for Non-Use */
+	MOCS_ENTRY(0,
+		   IG_PAT(0),
+		   L3_LKUP(1) | L3_3_WB),
+	/* Cached - L3 + L4 */
+	MOCS_ENTRY(1,
+		   IG_PAT(1),
+		   L3_LKUP(1) | L3_3_WB),
+	/* L4 - GO:L3 */
+	MOCS_ENTRY(2,
+		   IG_PAT(1),
+		   L3_LKUP(1) | L3_1_UC),
+	/* Uncached - GO:L3 */
+	MOCS_ENTRY(3,
+		   IG_PAT(1) | L4_3_UC,
+		   L3_LKUP(1) | L3_1_UC),
+	/* L4 - GO:Mem */
+	MOCS_ENTRY(4,
+		   IG_PAT(1),
+		   L3_LKUP(1) | L3_GLBGO(1) | L3_1_UC),
+	/* Uncached - GO:Mem */
+	MOCS_ENTRY(5,
+		   IG_PAT(1) | L4_3_UC,
+		   L3_LKUP(1) | L3_GLBGO(1) | L3_1_UC),
+	/* L4 - L3:NoLKUP; GO:L3 */
+	MOCS_ENTRY(6,
+		   IG_PAT(1),
+		   L3_1_UC),
+	/* Uncached - L3:NoLKUP; GO:L3 */
+	MOCS_ENTRY(7,
+		   IG_PAT(1) | L4_3_UC,
+		   L3_1_UC),
+	/* L4 - L3:NoLKUP; GO:Mem */
+	MOCS_ENTRY(8,
+		   IG_PAT(1),
+		   L3_GLBGO(1) | L3_1_UC),
+	/* Uncached - L3:NoLKUP; GO:Mem */
+	MOCS_ENTRY(9,
+		   IG_PAT(1) | L4_3_UC,
+		   L3_GLBGO(1) | L3_1_UC),
+	/* Display - L3; L4:WT */
+	MOCS_ENTRY(14,
+		   IG_PAT(1) | L4_1_WT,
+		   L3_LKUP(1) | L3_3_WB),
+	/* CCS - Non-Displayable */
+	MOCS_ENTRY(15,
+		   IG_PAT(1),
+		   L3_GLBGO(1) | L3_1_UC),
+};
+
 enum {
 	HAS_GLOBAL_MOCS = BIT(0),
 	HAS_ENGINE_MOCS = BIT(1),
@@ -445,7 +507,13 @@ static unsigned int get_mocs_settings(const struct drm_i915_private *i915,
 	memset(table, 0, sizeof(struct drm_i915_mocs_table));
 
 	table->unused_entries_index = I915_MOCS_PTE;
-	if (IS_PONTEVECCHIO(i915)) {
+	if (IS_METEORLAKE(i915)) {
+		table->size = ARRAY_SIZE(mtl_mocs_table);
+		table->table = mtl_mocs_table;
+		table->n_entries = MTL_NUM_MOCS_ENTRIES;
+		table->uc_index = 9;
+		table->unused_entries_index = 1;
+	} else if (IS_PONTEVECCHIO(i915)) {
 		table->size = ARRAY_SIZE(pvc_mocs_table);
 		table->table = pvc_mocs_table;
 		table->n_entries = PVC_NUM_MOCS_ENTRIES;
-- 
GitLab


From faca6aaa4838c3c234caa619d3c7d1f09da0d303 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Fri, 21 Apr 2023 20:25:35 +0200
Subject: [PATCH 0245/1791] drm/i915/mtl: fix mocs selftest

Media GT has a different base for MOCS register, need to apply
gsi_offset to the mmio address if not using the intel_uncore_r/w
functions for register access.

Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421182535.292670-3-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/gt/selftest_mocs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_mocs.c b/drivers/gpu/drm/i915/gt/selftest_mocs.c
index ca009a6a13bdb..a8446ab825012 100644
--- a/drivers/gpu/drm/i915/gt/selftest_mocs.c
+++ b/drivers/gpu/drm/i915/gt/selftest_mocs.c
@@ -131,13 +131,14 @@ static int read_mocs_table(struct i915_request *rq,
 			   const struct drm_i915_mocs_table *table,
 			   u32 *offset)
 {
+	struct intel_gt *gt = rq->engine->gt;
 	u32 addr;
 
 	if (!table)
 		return 0;
 
 	if (HAS_GLOBAL_MOCS_REGISTERS(rq->engine->i915))
-		addr = global_mocs_offset();
+		addr = global_mocs_offset() + gt->uncore->gsi_offset;
 	else
 		addr = mocs_offset(rq->engine);
 
-- 
GitLab


From 64e22551b64c694de428a9d3e210587df2831ef3 Mon Sep 17 00:00:00 2001
From: Andi Shyti <andi.shyti@linux.intel.com>
Date: Fri, 21 Apr 2023 21:00:25 +0200
Subject: [PATCH 0246/1791] drm/i915/i915_drv: Use proper parameter naming in
 for_each_engine()

for_each_engine() loops through engines in the GT, not in
dev_priv.  Because it's misleading, call it "gt__" instead of
"dev_priv__".

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421190026.294208-2-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/i915_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e771fdc3099c2..d95d354c4cb88 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -381,11 +381,11 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 }
 
 /* Simple iterator over all initialised engines */
-#define for_each_engine(engine__, dev_priv__, id__) \
+#define for_each_engine(engine__, gt__, id__) \
 	for ((id__) = 0; \
 	     (id__) < I915_NUM_ENGINES; \
 	     (id__)++) \
-		for_each_if ((engine__) = (dev_priv__)->engine[(id__)])
+		for_each_if ((engine__) = (gt__)->engine[(id__)])
 
 /* Iterator over subset of engines selected by mask */
 #define for_each_engine_masked(engine__, gt__, mask__, tmp__) \
-- 
GitLab


From 66ca1d8f222bdb1c9470e44131e12d753622ab08 Mon Sep 17 00:00:00 2001
From: Andi Shyti <andi.shyti@linux.intel.com>
Date: Fri, 21 Apr 2023 21:00:26 +0200
Subject: [PATCH 0247/1791] drm/i915/i915_drv: Use i915 instead of dev_priv
 insied the file_priv structure

In the process of renaming all instances of 'dev_priv' to 'i915',
start using 'i915' within the i915_drv.h file.

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421190026.294208-3-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/i915_drv.h | 458 ++++++++++++++++----------------
 1 file changed, 229 insertions(+), 229 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index d95d354c4cb88..f23b030aaf09f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -407,11 +407,11 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 	     (engine__) && (engine__)->uabi_class == (class__); \
 	     (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node)))
 
-#define INTEL_INFO(dev_priv)	(&(dev_priv)->__info)
-#define RUNTIME_INFO(dev_priv)	(&(dev_priv)->__runtime)
-#define DRIVER_CAPS(dev_priv)	(&(dev_priv)->caps)
+#define INTEL_INFO(i915)	(&(i915)->__info)
+#define RUNTIME_INFO(i915)	(&(i915)->__runtime)
+#define DRIVER_CAPS(i915)	(&(i915)->caps)
 
-#define INTEL_DEVID(dev_priv)	(RUNTIME_INFO(dev_priv)->device_id)
+#define INTEL_DEVID(i915)	(RUNTIME_INFO(i915)->device_id)
 
 #define IP_VER(ver, rel)		((ver) << 8 | (rel))
 
@@ -431,7 +431,7 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 #define IS_DISPLAY_VER(i915, from, until) \
 	(DISPLAY_VER(i915) >= (from) && DISPLAY_VER(i915) <= (until))
 
-#define INTEL_REVID(dev_priv)	(to_pci_dev((dev_priv)->drm.dev)->revision)
+#define INTEL_REVID(i915)	(to_pci_dev((i915)->drm.dev)->revision)
 
 #define INTEL_DISPLAY_STEP(__i915) (RUNTIME_INFO(__i915)->step.display_step)
 #define INTEL_GRAPHICS_STEP(__i915) (RUNTIME_INFO(__i915)->step.graphics_step)
@@ -516,135 +516,135 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 	return ((mask << (msb - pb)) & (mask << (msb - s))) & BIT(msb);
 }
 
-#define IS_MOBILE(dev_priv)	(INTEL_INFO(dev_priv)->is_mobile)
-#define IS_DGFX(dev_priv)   (INTEL_INFO(dev_priv)->is_dgfx)
-
-#define IS_I830(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I830)
-#define IS_I845G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I845G)
-#define IS_I85X(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I85X)
-#define IS_I865G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I865G)
-#define IS_I915G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I915G)
-#define IS_I915GM(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I915GM)
-#define IS_I945G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I945G)
-#define IS_I945GM(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I945GM)
-#define IS_I965G(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I965G)
-#define IS_I965GM(dev_priv)	IS_PLATFORM(dev_priv, INTEL_I965GM)
-#define IS_G45(dev_priv)	IS_PLATFORM(dev_priv, INTEL_G45)
-#define IS_GM45(dev_priv)	IS_PLATFORM(dev_priv, INTEL_GM45)
-#define IS_G4X(dev_priv)	(IS_G45(dev_priv) || IS_GM45(dev_priv))
-#define IS_PINEVIEW(dev_priv)	IS_PLATFORM(dev_priv, INTEL_PINEVIEW)
-#define IS_G33(dev_priv)	IS_PLATFORM(dev_priv, INTEL_G33)
-#define IS_IRONLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_IRONLAKE)
-#define IS_IRONLAKE_M(dev_priv) \
-	(IS_PLATFORM(dev_priv, INTEL_IRONLAKE) && IS_MOBILE(dev_priv))
-#define IS_SANDYBRIDGE(dev_priv) IS_PLATFORM(dev_priv, INTEL_SANDYBRIDGE)
-#define IS_IVYBRIDGE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_IVYBRIDGE)
-#define IS_IVB_GT1(dev_priv)	(IS_IVYBRIDGE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 1)
-#define IS_VALLEYVIEW(dev_priv)	IS_PLATFORM(dev_priv, INTEL_VALLEYVIEW)
-#define IS_CHERRYVIEW(dev_priv)	IS_PLATFORM(dev_priv, INTEL_CHERRYVIEW)
-#define IS_HASWELL(dev_priv)	IS_PLATFORM(dev_priv, INTEL_HASWELL)
-#define IS_BROADWELL(dev_priv)	IS_PLATFORM(dev_priv, INTEL_BROADWELL)
-#define IS_SKYLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_SKYLAKE)
-#define IS_BROXTON(dev_priv)	IS_PLATFORM(dev_priv, INTEL_BROXTON)
-#define IS_KABYLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_KABYLAKE)
-#define IS_GEMINILAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_GEMINILAKE)
-#define IS_COFFEELAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_COFFEELAKE)
-#define IS_COMETLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_COMETLAKE)
-#define IS_ICELAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_ICELAKE)
-#define IS_JSL_EHL(dev_priv)	(IS_PLATFORM(dev_priv, INTEL_JASPERLAKE) || \
-				IS_PLATFORM(dev_priv, INTEL_ELKHARTLAKE))
-#define IS_TIGERLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_TIGERLAKE)
-#define IS_ROCKETLAKE(dev_priv)	IS_PLATFORM(dev_priv, INTEL_ROCKETLAKE)
-#define IS_DG1(dev_priv)        IS_PLATFORM(dev_priv, INTEL_DG1)
-#define IS_ALDERLAKE_S(dev_priv) IS_PLATFORM(dev_priv, INTEL_ALDERLAKE_S)
-#define IS_ALDERLAKE_P(dev_priv) IS_PLATFORM(dev_priv, INTEL_ALDERLAKE_P)
-#define IS_XEHPSDV(dev_priv) IS_PLATFORM(dev_priv, INTEL_XEHPSDV)
-#define IS_DG2(dev_priv)	IS_PLATFORM(dev_priv, INTEL_DG2)
-#define IS_PONTEVECCHIO(dev_priv) IS_PLATFORM(dev_priv, INTEL_PONTEVECCHIO)
-#define IS_METEORLAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_METEORLAKE)
-
-#define IS_METEORLAKE_M(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_METEORLAKE, INTEL_SUBPLATFORM_M)
-#define IS_METEORLAKE_P(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_METEORLAKE, INTEL_SUBPLATFORM_P)
-#define IS_DG2_G10(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G10)
-#define IS_DG2_G11(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G11)
-#define IS_DG2_G12(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G12)
-#define IS_ADLS_RPLS(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_S, INTEL_SUBPLATFORM_RPL)
-#define IS_ADLP_N(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_N)
-#define IS_ADLP_RPLP(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPL)
-#define IS_ADLP_RPLU(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPLU)
-#define IS_HSW_EARLY_SDV(dev_priv) (IS_HASWELL(dev_priv) && \
-				    (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0C00)
-#define IS_BDW_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULT)
-#define IS_BDW_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULX)
-#define IS_BDW_GT3(dev_priv)	(IS_BROADWELL(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_HSW_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULT)
-#define IS_HSW_GT3(dev_priv)	(IS_HASWELL(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_HSW_GT1(dev_priv)	(IS_HASWELL(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 1)
+#define IS_MOBILE(i915)	(INTEL_INFO(i915)->is_mobile)
+#define IS_DGFX(i915)   (INTEL_INFO(i915)->is_dgfx)
+
+#define IS_I830(i915)	IS_PLATFORM(i915, INTEL_I830)
+#define IS_I845G(i915)	IS_PLATFORM(i915, INTEL_I845G)
+#define IS_I85X(i915)	IS_PLATFORM(i915, INTEL_I85X)
+#define IS_I865G(i915)	IS_PLATFORM(i915, INTEL_I865G)
+#define IS_I915G(i915)	IS_PLATFORM(i915, INTEL_I915G)
+#define IS_I915GM(i915)	IS_PLATFORM(i915, INTEL_I915GM)
+#define IS_I945G(i915)	IS_PLATFORM(i915, INTEL_I945G)
+#define IS_I945GM(i915)	IS_PLATFORM(i915, INTEL_I945GM)
+#define IS_I965G(i915)	IS_PLATFORM(i915, INTEL_I965G)
+#define IS_I965GM(i915)	IS_PLATFORM(i915, INTEL_I965GM)
+#define IS_G45(i915)	IS_PLATFORM(i915, INTEL_G45)
+#define IS_GM45(i915)	IS_PLATFORM(i915, INTEL_GM45)
+#define IS_G4X(i915)	(IS_G45(i915) || IS_GM45(i915))
+#define IS_PINEVIEW(i915)	IS_PLATFORM(i915, INTEL_PINEVIEW)
+#define IS_G33(i915)	IS_PLATFORM(i915, INTEL_G33)
+#define IS_IRONLAKE(i915)	IS_PLATFORM(i915, INTEL_IRONLAKE)
+#define IS_IRONLAKE_M(i915) \
+	(IS_PLATFORM(i915, INTEL_IRONLAKE) && IS_MOBILE(i915))
+#define IS_SANDYBRIDGE(i915) IS_PLATFORM(i915, INTEL_SANDYBRIDGE)
+#define IS_IVYBRIDGE(i915)	IS_PLATFORM(i915, INTEL_IVYBRIDGE)
+#define IS_IVB_GT1(i915)	(IS_IVYBRIDGE(i915) && \
+				 INTEL_INFO(i915)->gt == 1)
+#define IS_VALLEYVIEW(i915)	IS_PLATFORM(i915, INTEL_VALLEYVIEW)
+#define IS_CHERRYVIEW(i915)	IS_PLATFORM(i915, INTEL_CHERRYVIEW)
+#define IS_HASWELL(i915)	IS_PLATFORM(i915, INTEL_HASWELL)
+#define IS_BROADWELL(i915)	IS_PLATFORM(i915, INTEL_BROADWELL)
+#define IS_SKYLAKE(i915)	IS_PLATFORM(i915, INTEL_SKYLAKE)
+#define IS_BROXTON(i915)	IS_PLATFORM(i915, INTEL_BROXTON)
+#define IS_KABYLAKE(i915)	IS_PLATFORM(i915, INTEL_KABYLAKE)
+#define IS_GEMINILAKE(i915)	IS_PLATFORM(i915, INTEL_GEMINILAKE)
+#define IS_COFFEELAKE(i915)	IS_PLATFORM(i915, INTEL_COFFEELAKE)
+#define IS_COMETLAKE(i915)	IS_PLATFORM(i915, INTEL_COMETLAKE)
+#define IS_ICELAKE(i915)	IS_PLATFORM(i915, INTEL_ICELAKE)
+#define IS_JSL_EHL(i915)	(IS_PLATFORM(i915, INTEL_JASPERLAKE) || \
+				IS_PLATFORM(i915, INTEL_ELKHARTLAKE))
+#define IS_TIGERLAKE(i915)	IS_PLATFORM(i915, INTEL_TIGERLAKE)
+#define IS_ROCKETLAKE(i915)	IS_PLATFORM(i915, INTEL_ROCKETLAKE)
+#define IS_DG1(i915)        IS_PLATFORM(i915, INTEL_DG1)
+#define IS_ALDERLAKE_S(i915) IS_PLATFORM(i915, INTEL_ALDERLAKE_S)
+#define IS_ALDERLAKE_P(i915) IS_PLATFORM(i915, INTEL_ALDERLAKE_P)
+#define IS_XEHPSDV(i915) IS_PLATFORM(i915, INTEL_XEHPSDV)
+#define IS_DG2(i915)	IS_PLATFORM(i915, INTEL_DG2)
+#define IS_PONTEVECCHIO(i915) IS_PLATFORM(i915, INTEL_PONTEVECCHIO)
+#define IS_METEORLAKE(i915) IS_PLATFORM(i915, INTEL_METEORLAKE)
+
+#define IS_METEORLAKE_M(i915) \
+	IS_SUBPLATFORM(i915, INTEL_METEORLAKE, INTEL_SUBPLATFORM_M)
+#define IS_METEORLAKE_P(i915) \
+	IS_SUBPLATFORM(i915, INTEL_METEORLAKE, INTEL_SUBPLATFORM_P)
+#define IS_DG2_G10(i915) \
+	IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G10)
+#define IS_DG2_G11(i915) \
+	IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G11)
+#define IS_DG2_G12(i915) \
+	IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G12)
+#define IS_ADLS_RPLS(i915) \
+	IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_S, INTEL_SUBPLATFORM_RPL)
+#define IS_ADLP_N(i915) \
+	IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_N)
+#define IS_ADLP_RPLP(i915) \
+	IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPL)
+#define IS_ADLP_RPLU(i915) \
+	IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPLU)
+#define IS_HSW_EARLY_SDV(i915) (IS_HASWELL(i915) && \
+				    (INTEL_DEVID(i915) & 0xFF00) == 0x0C00)
+#define IS_BDW_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULT)
+#define IS_BDW_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULX)
+#define IS_BDW_GT3(i915)	(IS_BROADWELL(i915) && \
+				 INTEL_INFO(i915)->gt == 3)
+#define IS_HSW_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_HASWELL, INTEL_SUBPLATFORM_ULT)
+#define IS_HSW_GT3(i915)	(IS_HASWELL(i915) && \
+				 INTEL_INFO(i915)->gt == 3)
+#define IS_HSW_GT1(i915)	(IS_HASWELL(i915) && \
+				 INTEL_INFO(i915)->gt == 1)
 /* ULX machines are also considered ULT. */
-#define IS_HSW_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_HASWELL, INTEL_SUBPLATFORM_ULX)
-#define IS_SKL_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULT)
-#define IS_SKL_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULX)
-#define IS_KBL_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULT)
-#define IS_KBL_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULX)
-#define IS_SKL_GT2(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 2)
-#define IS_SKL_GT3(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_SKL_GT4(dev_priv)	(IS_SKYLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 4)
-#define IS_KBL_GT2(dev_priv)	(IS_KABYLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 2)
-#define IS_KBL_GT3(dev_priv)	(IS_KABYLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 3)
-#define IS_CFL_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULT)
-#define IS_CFL_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULX)
-#define IS_CFL_GT2(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 2)
-#define IS_CFL_GT3(dev_priv)	(IS_COFFEELAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 3)
-
-#define IS_CML_ULT(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULT)
-#define IS_CML_ULX(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULX)
-#define IS_CML_GT2(dev_priv)	(IS_COMETLAKE(dev_priv) && \
-				 INTEL_INFO(dev_priv)->gt == 2)
-
-#define IS_ICL_WITH_PORT_F(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF)
-
-#define IS_TGL_UY(dev_priv) \
-	IS_SUBPLATFORM(dev_priv, INTEL_TIGERLAKE, INTEL_SUBPLATFORM_UY)
+#define IS_HSW_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_HASWELL, INTEL_SUBPLATFORM_ULX)
+#define IS_SKL_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_SKL_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_KBL_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_KBL_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_SKL_GT2(i915)	(IS_SKYLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 2)
+#define IS_SKL_GT3(i915)	(IS_SKYLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 3)
+#define IS_SKL_GT4(i915)	(IS_SKYLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 4)
+#define IS_KBL_GT2(i915)	(IS_KABYLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 2)
+#define IS_KBL_GT3(i915)	(IS_KABYLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 3)
+#define IS_CFL_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_CFL_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_CFL_GT2(i915)	(IS_COFFEELAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 2)
+#define IS_CFL_GT3(i915)	(IS_COFFEELAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 3)
+
+#define IS_CML_ULT(i915) \
+	IS_SUBPLATFORM(i915, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULT)
+#define IS_CML_ULX(i915) \
+	IS_SUBPLATFORM(i915, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULX)
+#define IS_CML_GT2(i915)	(IS_COMETLAKE(i915) && \
+				 INTEL_INFO(i915)->gt == 2)
+
+#define IS_ICL_WITH_PORT_F(i915) \
+	IS_SUBPLATFORM(i915, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF)
+
+#define IS_TGL_UY(i915) \
+	IS_SUBPLATFORM(i915, INTEL_TIGERLAKE, INTEL_SUBPLATFORM_UY)
 
 #define IS_SKL_GRAPHICS_STEP(p, since, until) (IS_SKYLAKE(p) && IS_GRAPHICS_STEP(p, since, until))
 
-#define IS_KBL_GRAPHICS_STEP(dev_priv, since, until) \
-	(IS_KABYLAKE(dev_priv) && IS_GRAPHICS_STEP(dev_priv, since, until))
-#define IS_KBL_DISPLAY_STEP(dev_priv, since, until) \
-	(IS_KABYLAKE(dev_priv) && IS_DISPLAY_STEP(dev_priv, since, until))
+#define IS_KBL_GRAPHICS_STEP(i915, since, until) \
+	(IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, since, until))
+#define IS_KBL_DISPLAY_STEP(i915, since, until) \
+	(IS_KABYLAKE(i915) && IS_DISPLAY_STEP(i915, since, until))
 
 #define IS_JSL_EHL_GRAPHICS_STEP(p, since, until) \
 	(IS_JSL_EHL(p) && IS_GRAPHICS_STEP(p, since, until))
@@ -720,9 +720,9 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 	(IS_PONTEVECCHIO(__i915) && \
 	 IS_GRAPHICS_STEP(__i915, since, until))
 
-#define IS_LP(dev_priv)		(INTEL_INFO(dev_priv)->is_lp)
-#define IS_GEN9_LP(dev_priv)	(GRAPHICS_VER(dev_priv) == 9 && IS_LP(dev_priv))
-#define IS_GEN9_BC(dev_priv)	(GRAPHICS_VER(dev_priv) == 9 && !IS_LP(dev_priv))
+#define IS_LP(i915)		(INTEL_INFO(i915)->is_lp)
+#define IS_GEN9_LP(i915)	(GRAPHICS_VER(i915) == 9 && IS_LP(i915))
+#define IS_GEN9_BC(i915)	(GRAPHICS_VER(i915) == 9 && !IS_LP(i915))
 
 #define __HAS_ENGINE(engine_mask, id) ((engine_mask) & BIT(id))
 #define HAS_ENGINE(gt, id) __HAS_ENGINE((gt)->info.engine_mask, id)
@@ -747,180 +747,180 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define CCS_MASK(gt) \
 	ENGINE_INSTANCES_MASK(gt, CCS0, I915_MAX_CCS)
 
-#define HAS_MEDIA_RATIO_MODE(dev_priv) (INTEL_INFO(dev_priv)->has_media_ratio_mode)
+#define HAS_MEDIA_RATIO_MODE(i915) (INTEL_INFO(i915)->has_media_ratio_mode)
 
 /*
  * The Gen7 cmdparser copies the scanned buffer to the ggtt for execution
  * All later gens can run the final buffer from the ppgtt
  */
-#define CMDPARSER_USES_GGTT(dev_priv) (GRAPHICS_VER(dev_priv) == 7)
+#define CMDPARSER_USES_GGTT(i915) (GRAPHICS_VER(i915) == 7)
 
-#define HAS_LLC(dev_priv)	(INTEL_INFO(dev_priv)->has_llc)
-#define HAS_4TILE(dev_priv)	(INTEL_INFO(dev_priv)->has_4tile)
-#define HAS_SNOOP(dev_priv)	(INTEL_INFO(dev_priv)->has_snoop)
-#define HAS_EDRAM(dev_priv)	((dev_priv)->edram_size_mb)
-#define HAS_SECURE_BATCHES(dev_priv) (GRAPHICS_VER(dev_priv) < 6)
-#define HAS_WT(dev_priv)	HAS_EDRAM(dev_priv)
+#define HAS_LLC(i915)	(INTEL_INFO(i915)->has_llc)
+#define HAS_4TILE(i915)	(INTEL_INFO(i915)->has_4tile)
+#define HAS_SNOOP(i915)	(INTEL_INFO(i915)->has_snoop)
+#define HAS_EDRAM(i915)	((i915)->edram_size_mb)
+#define HAS_SECURE_BATCHES(i915) (GRAPHICS_VER(i915) < 6)
+#define HAS_WT(i915)	HAS_EDRAM(i915)
 
-#define HWS_NEEDS_PHYSICAL(dev_priv)	(INTEL_INFO(dev_priv)->hws_needs_physical)
+#define HWS_NEEDS_PHYSICAL(i915)	(INTEL_INFO(i915)->hws_needs_physical)
 
-#define HAS_LOGICAL_RING_CONTEXTS(dev_priv) \
-		(INTEL_INFO(dev_priv)->has_logical_ring_contexts)
-#define HAS_LOGICAL_RING_ELSQ(dev_priv) \
-		(INTEL_INFO(dev_priv)->has_logical_ring_elsq)
+#define HAS_LOGICAL_RING_CONTEXTS(i915) \
+		(INTEL_INFO(i915)->has_logical_ring_contexts)
+#define HAS_LOGICAL_RING_ELSQ(i915) \
+		(INTEL_INFO(i915)->has_logical_ring_elsq)
 
-#define HAS_EXECLISTS(dev_priv) HAS_LOGICAL_RING_CONTEXTS(dev_priv)
+#define HAS_EXECLISTS(i915) HAS_LOGICAL_RING_CONTEXTS(i915)
 
-#define INTEL_PPGTT(dev_priv) (RUNTIME_INFO(dev_priv)->ppgtt_type)
-#define HAS_PPGTT(dev_priv) \
-	(INTEL_PPGTT(dev_priv) != INTEL_PPGTT_NONE)
-#define HAS_FULL_PPGTT(dev_priv) \
-	(INTEL_PPGTT(dev_priv) >= INTEL_PPGTT_FULL)
+#define INTEL_PPGTT(i915) (RUNTIME_INFO(i915)->ppgtt_type)
+#define HAS_PPGTT(i915) \
+	(INTEL_PPGTT(i915) != INTEL_PPGTT_NONE)
+#define HAS_FULL_PPGTT(i915) \
+	(INTEL_PPGTT(i915) >= INTEL_PPGTT_FULL)
 
-#define HAS_PAGE_SIZES(dev_priv, sizes) ({ \
+#define HAS_PAGE_SIZES(i915, sizes) ({ \
 	GEM_BUG_ON((sizes) == 0); \
-	((sizes) & ~RUNTIME_INFO(dev_priv)->page_sizes) == 0; \
+	((sizes) & ~RUNTIME_INFO(i915)->page_sizes) == 0; \
 })
 
-#define HAS_OVERLAY(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_overlay)
-#define OVERLAY_NEEDS_PHYSICAL(dev_priv) \
-		(INTEL_INFO(dev_priv)->display.overlay_needs_physical)
+#define HAS_OVERLAY(i915)		 (INTEL_INFO(i915)->display.has_overlay)
+#define OVERLAY_NEEDS_PHYSICAL(i915) \
+		(INTEL_INFO(i915)->display.overlay_needs_physical)
 
 /* Early gen2 have a totally busted CS tlb and require pinned batches. */
-#define HAS_BROKEN_CS_TLB(dev_priv)	(IS_I830(dev_priv) || IS_I845G(dev_priv))
+#define HAS_BROKEN_CS_TLB(i915)	(IS_I830(i915) || IS_I845G(i915))
 
-#define NEEDS_RC6_CTX_CORRUPTION_WA(dev_priv)	\
-	(IS_BROADWELL(dev_priv) || GRAPHICS_VER(dev_priv) == 9)
+#define NEEDS_RC6_CTX_CORRUPTION_WA(i915)	\
+	(IS_BROADWELL(i915) || GRAPHICS_VER(i915) == 9)
 
 /* WaRsDisableCoarsePowerGating:skl,cnl */
-#define NEEDS_WaRsDisableCoarsePowerGating(dev_priv)			\
-	(IS_SKL_GT3(dev_priv) || IS_SKL_GT4(dev_priv))
+#define NEEDS_WaRsDisableCoarsePowerGating(i915)			\
+	(IS_SKL_GT3(i915) || IS_SKL_GT4(i915))
 
-#define HAS_GMBUS_IRQ(dev_priv) (DISPLAY_VER(dev_priv) >= 4)
-#define HAS_GMBUS_BURST_READ(dev_priv) (DISPLAY_VER(dev_priv) >= 11 || \
-					IS_GEMINILAKE(dev_priv) || \
-					IS_KABYLAKE(dev_priv))
+#define HAS_GMBUS_IRQ(i915) (DISPLAY_VER(i915) >= 4)
+#define HAS_GMBUS_BURST_READ(i915) (DISPLAY_VER(i915) >= 11 || \
+					IS_GEMINILAKE(i915) || \
+					IS_KABYLAKE(i915))
 
 /* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte
  * rows, which changed the alignment requirements and fence programming.
  */
-#define HAS_128_BYTE_Y_TILING(dev_priv) (GRAPHICS_VER(dev_priv) != 2 && \
-					 !(IS_I915G(dev_priv) || IS_I915GM(dev_priv)))
-#define SUPPORTS_TV(dev_priv)		(INTEL_INFO(dev_priv)->display.supports_tv)
-#define I915_HAS_HOTPLUG(dev_priv)	(INTEL_INFO(dev_priv)->display.has_hotplug)
+#define HAS_128_BYTE_Y_TILING(i915) (GRAPHICS_VER(i915) != 2 && \
+					 !(IS_I915G(i915) || IS_I915GM(i915)))
+#define SUPPORTS_TV(i915)		(INTEL_INFO(i915)->display.supports_tv)
+#define I915_HAS_HOTPLUG(i915)	(INTEL_INFO(i915)->display.has_hotplug)
 
-#define HAS_FW_BLC(dev_priv)	(DISPLAY_VER(dev_priv) > 2)
-#define HAS_FBC(dev_priv)	(RUNTIME_INFO(dev_priv)->fbc_mask != 0)
-#define HAS_CUR_FBC(dev_priv)	(!HAS_GMCH(dev_priv) && DISPLAY_VER(dev_priv) >= 7)
+#define HAS_FW_BLC(i915)	(DISPLAY_VER(i915) > 2)
+#define HAS_FBC(i915)	(RUNTIME_INFO(i915)->fbc_mask != 0)
+#define HAS_CUR_FBC(i915)	(!HAS_GMCH(i915) && DISPLAY_VER(i915) >= 7)
 
-#define HAS_DPT(dev_priv)	(DISPLAY_VER(dev_priv) >= 13)
+#define HAS_DPT(i915)	(DISPLAY_VER(i915) >= 13)
 
-#define HAS_IPS(dev_priv)	(IS_HSW_ULT(dev_priv) || IS_BROADWELL(dev_priv))
+#define HAS_IPS(i915)	(IS_HSW_ULT(i915) || IS_BROADWELL(i915))
 
-#define HAS_DP_MST(dev_priv)	(INTEL_INFO(dev_priv)->display.has_dp_mst)
-#define HAS_DP20(dev_priv)	(IS_DG2(dev_priv) || DISPLAY_VER(dev_priv) >= 14)
+#define HAS_DP_MST(i915)	(INTEL_INFO(i915)->display.has_dp_mst)
+#define HAS_DP20(i915)	(IS_DG2(i915) || DISPLAY_VER(i915) >= 14)
 
-#define HAS_DOUBLE_BUFFERED_M_N(dev_priv)	(DISPLAY_VER(dev_priv) >= 9 || IS_BROADWELL(dev_priv))
+#define HAS_DOUBLE_BUFFERED_M_N(i915)	(DISPLAY_VER(i915) >= 9 || IS_BROADWELL(i915))
 
-#define HAS_CDCLK_CRAWL(dev_priv)	 (INTEL_INFO(dev_priv)->display.has_cdclk_crawl)
-#define HAS_CDCLK_SQUASH(dev_priv)	 (INTEL_INFO(dev_priv)->display.has_cdclk_squash)
-#define HAS_DDI(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_ddi)
-#define HAS_FPGA_DBG_UNCLAIMED(dev_priv) (INTEL_INFO(dev_priv)->display.has_fpga_dbg)
-#define HAS_PSR(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_psr)
-#define HAS_PSR_HW_TRACKING(dev_priv) \
-	(INTEL_INFO(dev_priv)->display.has_psr_hw_tracking)
-#define HAS_PSR2_SEL_FETCH(dev_priv)	 (DISPLAY_VER(dev_priv) >= 12)
-#define HAS_TRANSCODER(dev_priv, trans)	 ((RUNTIME_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
+#define HAS_CDCLK_CRAWL(i915)	 (INTEL_INFO(i915)->display.has_cdclk_crawl)
+#define HAS_CDCLK_SQUASH(i915)	 (INTEL_INFO(i915)->display.has_cdclk_squash)
+#define HAS_DDI(i915)		 (INTEL_INFO(i915)->display.has_ddi)
+#define HAS_FPGA_DBG_UNCLAIMED(i915) (INTEL_INFO(i915)->display.has_fpga_dbg)
+#define HAS_PSR(i915)		 (INTEL_INFO(i915)->display.has_psr)
+#define HAS_PSR_HW_TRACKING(i915) \
+	(INTEL_INFO(i915)->display.has_psr_hw_tracking)
+#define HAS_PSR2_SEL_FETCH(i915)	 (DISPLAY_VER(i915) >= 12)
+#define HAS_TRANSCODER(i915, trans)	 ((RUNTIME_INFO(i915)->cpu_transcoder_mask & BIT(trans)) != 0)
 
-#define HAS_RC6(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6)
-#define HAS_RC6p(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6p)
-#define HAS_RC6pp(dev_priv)		 (false) /* HW was never validated */
+#define HAS_RC6(i915)		 (INTEL_INFO(i915)->has_rc6)
+#define HAS_RC6p(i915)		 (INTEL_INFO(i915)->has_rc6p)
+#define HAS_RC6pp(i915)		 (false) /* HW was never validated */
 
-#define HAS_RPS(dev_priv)	(INTEL_INFO(dev_priv)->has_rps)
+#define HAS_RPS(i915)	(INTEL_INFO(i915)->has_rps)
 
-#define HAS_DMC(dev_priv)	(RUNTIME_INFO(dev_priv)->has_dmc)
-#define HAS_DSB(dev_priv)	(INTEL_INFO(dev_priv)->display.has_dsb)
+#define HAS_DMC(i915)	(RUNTIME_INFO(i915)->has_dmc)
+#define HAS_DSB(i915)	(INTEL_INFO(i915)->display.has_dsb)
 #define HAS_DSC(__i915)		(RUNTIME_INFO(__i915)->has_dsc)
 #define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915))
 
-#define HAS_HECI_PXP(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_heci_pxp)
+#define HAS_HECI_PXP(i915) \
+	(INTEL_INFO(i915)->has_heci_pxp)
 
-#define HAS_HECI_GSCFI(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_heci_gscfi)
+#define HAS_HECI_GSCFI(i915) \
+	(INTEL_INFO(i915)->has_heci_gscfi)
 
-#define HAS_HECI_GSC(dev_priv) (HAS_HECI_PXP(dev_priv) || HAS_HECI_GSCFI(dev_priv))
+#define HAS_HECI_GSC(i915) (HAS_HECI_PXP(i915) || HAS_HECI_GSCFI(i915))
 
 #define HAS_MSO(i915)		(DISPLAY_VER(i915) >= 12)
 
-#define HAS_RUNTIME_PM(dev_priv) (INTEL_INFO(dev_priv)->has_runtime_pm)
-#define HAS_64BIT_RELOC(dev_priv) (INTEL_INFO(dev_priv)->has_64bit_reloc)
+#define HAS_RUNTIME_PM(i915) (INTEL_INFO(i915)->has_runtime_pm)
+#define HAS_64BIT_RELOC(i915) (INTEL_INFO(i915)->has_64bit_reloc)
 
-#define HAS_OA_BPC_REPORTING(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_oa_bpc_reporting)
-#define HAS_OA_SLICE_CONTRIB_LIMITS(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_oa_slice_contrib_limits)
-#define HAS_OAM(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_oam)
+#define HAS_OA_BPC_REPORTING(i915) \
+	(INTEL_INFO(i915)->has_oa_bpc_reporting)
+#define HAS_OA_SLICE_CONTRIB_LIMITS(i915) \
+	(INTEL_INFO(i915)->has_oa_slice_contrib_limits)
+#define HAS_OAM(i915) \
+	(INTEL_INFO(i915)->has_oam)
 
 /*
  * Set this flag, when platform requires 64K GTT page sizes or larger for
  * device local memory access.
  */
-#define HAS_64K_PAGES(dev_priv) (INTEL_INFO(dev_priv)->has_64k_pages)
+#define HAS_64K_PAGES(i915) (INTEL_INFO(i915)->has_64k_pages)
 
-#define HAS_IPC(dev_priv)		(INTEL_INFO(dev_priv)->display.has_ipc)
-#define HAS_SAGV(dev_priv)		(DISPLAY_VER(dev_priv) >= 9 && !IS_LP(dev_priv))
+#define HAS_IPC(i915)		(INTEL_INFO(i915)->display.has_ipc)
+#define HAS_SAGV(i915)		(DISPLAY_VER(i915) >= 9 && !IS_LP(i915))
 
 #define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i))
 #define HAS_LMEM(i915) HAS_REGION(i915, REGION_LMEM)
 
-#define HAS_EXTRA_GT_LIST(dev_priv)   (INTEL_INFO(dev_priv)->extra_gt_list)
+#define HAS_EXTRA_GT_LIST(i915)   (INTEL_INFO(i915)->extra_gt_list)
 
 /*
  * Platform has the dedicated compression control state for each lmem surfaces
  * stored in lmem to support the 3D and media compression formats.
  */
-#define HAS_FLAT_CCS(dev_priv)   (INTEL_INFO(dev_priv)->has_flat_ccs)
+#define HAS_FLAT_CCS(i915)   (INTEL_INFO(i915)->has_flat_ccs)
 
-#define HAS_GT_UC(dev_priv)	(INTEL_INFO(dev_priv)->has_gt_uc)
+#define HAS_GT_UC(i915)	(INTEL_INFO(i915)->has_gt_uc)
 
-#define HAS_POOLED_EU(dev_priv)	(RUNTIME_INFO(dev_priv)->has_pooled_eu)
+#define HAS_POOLED_EU(i915)	(RUNTIME_INFO(i915)->has_pooled_eu)
 
-#define HAS_GLOBAL_MOCS_REGISTERS(dev_priv)	(INTEL_INFO(dev_priv)->has_global_mocs)
+#define HAS_GLOBAL_MOCS_REGISTERS(i915)	(INTEL_INFO(i915)->has_global_mocs)
 
-#define HAS_GMCH(dev_priv) (INTEL_INFO(dev_priv)->display.has_gmch)
+#define HAS_GMCH(i915) (INTEL_INFO(i915)->display.has_gmch)
 
 #define HAS_GMD_ID(i915)	(INTEL_INFO(i915)->has_gmd_id)
 
-#define HAS_LSPCON(dev_priv) (IS_DISPLAY_VER(dev_priv, 9, 10))
+#define HAS_LSPCON(i915) (IS_DISPLAY_VER(i915, 9, 10))
 
 #define HAS_L3_CCS_READ(i915) (INTEL_INFO(i915)->has_l3_ccs_read)
 
 /* DPF == dynamic parity feature */
-#define HAS_L3_DPF(dev_priv) (INTEL_INFO(dev_priv)->has_l3_dpf)
-#define NUM_L3_SLICES(dev_priv) (IS_HSW_GT3(dev_priv) ? \
-				 2 : HAS_L3_DPF(dev_priv))
+#define HAS_L3_DPF(i915) (INTEL_INFO(i915)->has_l3_dpf)
+#define NUM_L3_SLICES(i915) (IS_HSW_GT3(i915) ? \
+				 2 : HAS_L3_DPF(i915))
 
-#define INTEL_NUM_PIPES(dev_priv) (hweight8(RUNTIME_INFO(dev_priv)->pipe_mask))
+#define INTEL_NUM_PIPES(i915) (hweight8(RUNTIME_INFO(i915)->pipe_mask))
 
-#define HAS_DISPLAY(dev_priv) (RUNTIME_INFO(dev_priv)->pipe_mask != 0)
+#define HAS_DISPLAY(i915) (RUNTIME_INFO(i915)->pipe_mask != 0)
 
 #define HAS_VRR(i915)	(DISPLAY_VER(i915) >= 11)
 
 #define HAS_ASYNC_FLIPS(i915)		(DISPLAY_VER(i915) >= 5)
 
 /* Only valid when HAS_DISPLAY() is true */
-#define INTEL_DISPLAY_ENABLED(dev_priv) \
-	(drm_WARN_ON(&(dev_priv)->drm, !HAS_DISPLAY(dev_priv)),		\
-	 !(dev_priv)->params.disable_display &&				\
-	 !intel_opregion_headless_sku(dev_priv))
+#define INTEL_DISPLAY_ENABLED(i915) \
+	(drm_WARN_ON(&(i915)->drm, !HAS_DISPLAY(i915)),		\
+	 !(i915)->params.disable_display &&				\
+	 !intel_opregion_headless_sku(i915))
 
-#define HAS_GUC_DEPRIVILEGE(dev_priv) \
-	(INTEL_INFO(dev_priv)->has_guc_deprivilege)
+#define HAS_GUC_DEPRIVILEGE(i915) \
+	(INTEL_INFO(i915)->has_guc_deprivilege)
 
-#define HAS_D12_PLANE_MINIMIZATION(dev_priv) (IS_ROCKETLAKE(dev_priv) || \
-					      IS_ALDERLAKE_S(dev_priv))
+#define HAS_D12_PLANE_MINIMIZATION(i915) (IS_ROCKETLAKE(i915) || \
+					      IS_ALDERLAKE_S(i915))
 
 #define HAS_MBUS_JOINING(i915) (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14)
 
-- 
GitLab


From 15f5ee6b77c0953b3d668eb3f3bb3c0be0a280f6 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Fri, 21 Apr 2023 17:13:19 +0800
Subject: [PATCH 0248/1791] drm/tve200: Use devm_platform_ioremap_resource()

Convert platform_get_resource(),devm_ioremap_resource() to a single
call to devm_platform_ioremap_resource(), as this is exactly what this
function does.

Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421091319.79744-1-yang.lee@linux.alibaba.com
---
 drivers/gpu/drm/tve200/tve200_drv.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/tve200/tve200_drv.c b/drivers/gpu/drm/tve200/tve200_drv.c
index abd557332b288..40b1168ad671f 100644
--- a/drivers/gpu/drm/tve200/tve200_drv.c
+++ b/drivers/gpu/drm/tve200/tve200_drv.c
@@ -156,7 +156,6 @@ static int tve200_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct tve200_drm_dev_private *priv;
 	struct drm_device *drm;
-	struct resource *res;
 	int irq;
 	int ret;
 
@@ -192,8 +191,7 @@ static int tve200_probe(struct platform_device *pdev)
 		goto clk_disable;
 	}
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	priv->regs = devm_ioremap_resource(dev, res);
+	priv->regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(priv->regs)) {
 		dev_err(dev, "%s failed mmio\n", __func__);
 		ret = -EINVAL;
-- 
GitLab


From 539f9ee4b52a8bec95ff064e22dd2fb1e258e818 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 17 Apr 2023 13:36:02 +0200
Subject: [PATCH 0249/1791] drm/scheduler: properly forward fence errors
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When a hw fence is signaled with an error properly forward that to the
finished fence.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420115752.31470-1-christian.koenig@amd.com
---
 drivers/gpu/drm/scheduler/sched_entity.c |  4 +---
 drivers/gpu/drm/scheduler/sched_fence.c  |  4 +++-
 drivers/gpu/drm/scheduler/sched_main.c   | 18 ++++++++----------
 include/drm/gpu_scheduler.h              |  2 +-
 4 files changed, 13 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 3e2a31d8190eb..1795cd7e42ed7 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -144,7 +144,7 @@ static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
 {
 	struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
 
-	drm_sched_fence_finished(job->s_fence);
+	drm_sched_fence_finished(job->s_fence, -ESRCH);
 	WARN_ON(job->s_fence->parent);
 	job->sched->ops->free_job(job);
 }
@@ -195,8 +195,6 @@ static void drm_sched_entity_kill(struct drm_sched_entity *entity)
 	while ((job = to_drm_sched_job(spsc_queue_pop(&entity->job_queue)))) {
 		struct drm_sched_fence *s_fence = job->s_fence;
 
-		dma_fence_set_error(&s_fence->finished, -ESRCH);
-
 		dma_fence_get(&s_fence->finished);
 		if (!prev || dma_fence_add_callback(prev, &job->finish_cb,
 					   drm_sched_entity_kill_jobs_cb))
diff --git a/drivers/gpu/drm/scheduler/sched_fence.c b/drivers/gpu/drm/scheduler/sched_fence.c
index fe9c6468e4401..ef120475e7c61 100644
--- a/drivers/gpu/drm/scheduler/sched_fence.c
+++ b/drivers/gpu/drm/scheduler/sched_fence.c
@@ -53,8 +53,10 @@ void drm_sched_fence_scheduled(struct drm_sched_fence *fence)
 	dma_fence_signal(&fence->scheduled);
 }
 
-void drm_sched_fence_finished(struct drm_sched_fence *fence)
+void drm_sched_fence_finished(struct drm_sched_fence *fence, int result)
 {
+	if (result)
+		dma_fence_set_error(&fence->finished, result);
 	dma_fence_signal(&fence->finished);
 }
 
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index c3582c4fc2992..8d248ce233c84 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -262,7 +262,7 @@ drm_sched_rq_select_entity_fifo(struct drm_sched_rq *rq)
  *
  * Finish the job's fence and wake up the worker thread.
  */
-static void drm_sched_job_done(struct drm_sched_job *s_job)
+static void drm_sched_job_done(struct drm_sched_job *s_job, int result)
 {
 	struct drm_sched_fence *s_fence = s_job->s_fence;
 	struct drm_gpu_scheduler *sched = s_fence->sched;
@@ -273,7 +273,7 @@ static void drm_sched_job_done(struct drm_sched_job *s_job)
 	trace_drm_sched_process_job(s_fence);
 
 	dma_fence_get(&s_fence->finished);
-	drm_sched_fence_finished(s_fence);
+	drm_sched_fence_finished(s_fence, result);
 	dma_fence_put(&s_fence->finished);
 	wake_up_interruptible(&sched->wake_up_worker);
 }
@@ -287,7 +287,7 @@ static void drm_sched_job_done_cb(struct dma_fence *f, struct dma_fence_cb *cb)
 {
 	struct drm_sched_job *s_job = container_of(cb, struct drm_sched_job, cb);
 
-	drm_sched_job_done(s_job);
+	drm_sched_job_done(s_job, f->error);
 }
 
 /**
@@ -537,12 +537,12 @@ void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery)
 			r = dma_fence_add_callback(fence, &s_job->cb,
 						   drm_sched_job_done_cb);
 			if (r == -ENOENT)
-				drm_sched_job_done(s_job);
+				drm_sched_job_done(s_job, fence->error);
 			else if (r)
 				DRM_DEV_ERROR(sched->dev, "fence add callback failed (%d)\n",
 					  r);
 		} else
-			drm_sched_job_done(s_job);
+			drm_sched_job_done(s_job, 0);
 	}
 
 	if (full_recovery) {
@@ -1059,15 +1059,13 @@ static int drm_sched_main(void *param)
 			r = dma_fence_add_callback(fence, &sched_job->cb,
 						   drm_sched_job_done_cb);
 			if (r == -ENOENT)
-				drm_sched_job_done(sched_job);
+				drm_sched_job_done(sched_job, fence->error);
 			else if (r)
 				DRM_DEV_ERROR(sched->dev, "fence add callback failed (%d)\n",
 					  r);
 		} else {
-			if (IS_ERR(fence))
-				dma_fence_set_error(&s_fence->finished, PTR_ERR(fence));
-
-			drm_sched_job_done(sched_job);
+			drm_sched_job_done(sched_job, IS_ERR(fence) ?
+					   PTR_ERR(fence) : 0);
 		}
 
 		wake_up(&sched->job_scheduled);
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index c0ec6719282af..3b4800e0b24b0 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -598,7 +598,7 @@ void drm_sched_fence_init(struct drm_sched_fence *fence,
 void drm_sched_fence_free(struct drm_sched_fence *fence);
 
 void drm_sched_fence_scheduled(struct drm_sched_fence *fence);
-void drm_sched_fence_finished(struct drm_sched_fence *fence);
+void drm_sched_fence_finished(struct drm_sched_fence *fence, int result);
 
 unsigned long drm_sched_suspend_timeout(struct drm_gpu_scheduler *sched);
 void drm_sched_resume_timeout(struct drm_gpu_scheduler *sched,
-- 
GitLab


From 70102d77ff22dd88a0111b1c3bac5099ac5d0425 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 17 Apr 2023 17:32:11 +0200
Subject: [PATCH 0250/1791] drm/scheduler: add drm_sched_entity_error and use
 rcu for last_scheduled
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Switch to using RCU handling for the last scheduled job and add a
function to return the error code of it.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420115752.31470-2-christian.koenig@amd.com
---
 drivers/gpu/drm/scheduler/sched_entity.c | 39 +++++++++++++++++++-----
 include/drm/gpu_scheduler.h              |  3 +-
 2 files changed, 33 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 1795cd7e42ed7..cfb433e920058 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -72,7 +72,7 @@ int drm_sched_entity_init(struct drm_sched_entity *entity,
 	entity->num_sched_list = num_sched_list;
 	entity->priority = priority;
 	entity->sched_list = num_sched_list > 1 ? sched_list : NULL;
-	entity->last_scheduled = NULL;
+	RCU_INIT_POINTER(entity->last_scheduled, NULL);
 	RB_CLEAR_NODE(&entity->rb_tree_node);
 
 	if(num_sched_list)
@@ -140,6 +140,27 @@ bool drm_sched_entity_is_ready(struct drm_sched_entity *entity)
 	return true;
 }
 
+/**
+ * drm_sched_entity_error - return error of last scheduled job
+ * @entity: scheduler entity to check
+ *
+ * Opportunistically return the error of the last scheduled job. Result can
+ * change any time when new jobs are pushed to the hw.
+ */
+int drm_sched_entity_error(struct drm_sched_entity *entity)
+{
+	struct dma_fence *fence;
+	int r;
+
+	rcu_read_lock();
+	fence = rcu_dereference(entity->last_scheduled);
+	r = fence ? fence->error : 0;
+	rcu_read_unlock();
+
+	return r;
+}
+EXPORT_SYMBOL(drm_sched_entity_error);
+
 static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
 {
 	struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
@@ -191,7 +212,9 @@ static void drm_sched_entity_kill(struct drm_sched_entity *entity)
 	/* Make sure this entity is not used by the scheduler at the moment */
 	wait_for_completion(&entity->entity_idle);
 
-	prev = dma_fence_get(entity->last_scheduled);
+	/* The entity is guaranteed to not be used by the scheduler */
+	prev = rcu_dereference_check(entity->last_scheduled, true);
+	dma_fence_get(prev);
 	while ((job = to_drm_sched_job(spsc_queue_pop(&entity->job_queue)))) {
 		struct drm_sched_fence *s_fence = job->s_fence;
 
@@ -278,8 +301,8 @@ void drm_sched_entity_fini(struct drm_sched_entity *entity)
 		entity->dependency = NULL;
 	}
 
-	dma_fence_put(entity->last_scheduled);
-	entity->last_scheduled = NULL;
+	dma_fence_put(rcu_dereference_check(entity->last_scheduled, true));
+	RCU_INIT_POINTER(entity->last_scheduled, NULL);
 }
 EXPORT_SYMBOL(drm_sched_entity_fini);
 
@@ -421,9 +444,9 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 	if (entity->guilty && atomic_read(entity->guilty))
 		dma_fence_set_error(&sched_job->s_fence->finished, -ECANCELED);
 
-	dma_fence_put(entity->last_scheduled);
-
-	entity->last_scheduled = dma_fence_get(&sched_job->s_fence->finished);
+	dma_fence_put(rcu_dereference_check(entity->last_scheduled, true));
+	rcu_assign_pointer(entity->last_scheduled,
+			   dma_fence_get(&sched_job->s_fence->finished));
 
 	/*
 	 * If the queue is empty we allow drm_sched_entity_select_rq() to
@@ -477,7 +500,7 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity)
 	 */
 	smp_rmb();
 
-	fence = entity->last_scheduled;
+	fence = rcu_dereference_check(entity->last_scheduled, true);
 
 	/* stay on the same engine if the previous job hasn't finished */
 	if (fence && !dma_fence_is_signaled(fence))
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index 3b4800e0b24b0..fd15f8ae0c3fc 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -201,7 +201,7 @@ struct drm_sched_entity {
 	 * by the scheduler thread, can be accessed locklessly from
 	 * drm_sched_job_arm() iff the queue is empty.
 	 */
-	struct dma_fence                *last_scheduled;
+	struct dma_fence __rcu		*last_scheduled;
 
 	/**
 	 * @last_user: last group leader pushing a job into the entity.
@@ -588,6 +588,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job);
 void drm_sched_entity_set_priority(struct drm_sched_entity *entity,
 				   enum drm_sched_priority priority);
 bool drm_sched_entity_is_ready(struct drm_sched_entity *entity);
+int drm_sched_entity_error(struct drm_sched_entity *entity);
 
 void drm_sched_fence_set_parent(struct drm_sched_fence *s_fence,
 				struct dma_fence *fence);
-- 
GitLab


From 96c3161926c6aa81365f074e9c408ca31b7ca756 Mon Sep 17 00:00:00 2001
From: Arthur Grillo <arthurgrillo@riseup.net>
Date: Tue, 18 Apr 2023 20:01:42 -0300
Subject: [PATCH 0251/1791] drm/tests: Add test cases for drm_rect_intersect()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Insert parameterized tests for the drm_rect_intersect() to ensure
correctness and prevent future regressions.

Also, create a helper for testing if two drm_rects are equal.

Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418230146.461129-2-arthurgrillo@riseup.net
---
 drivers/gpu/drm/tests/drm_rect_test.c | 148 ++++++++++++++++++++++++++
 1 file changed, 148 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_rect_test.c b/drivers/gpu/drm/tests/drm_rect_test.c
index e9809ea326968..33584f63be378 100644
--- a/drivers/gpu/drm/tests/drm_rect_test.c
+++ b/drivers/gpu/drm/tests/drm_rect_test.c
@@ -9,6 +9,17 @@
 
 #include <drm/drm_rect.h>
 
+#include <linux/string_helpers.h>
+
+static void drm_rect_compare(struct kunit *test, const struct drm_rect *r,
+			     const struct drm_rect *expected)
+{
+	KUNIT_EXPECT_EQ(test, r->x1, expected->x1);
+	KUNIT_EXPECT_EQ(test, r->y1, expected->y1);
+	KUNIT_EXPECT_EQ(test, drm_rect_width(r), drm_rect_width(expected));
+	KUNIT_EXPECT_EQ(test, drm_rect_height(r), drm_rect_height(expected));
+}
+
 static void drm_test_rect_clip_scaled_div_by_zero(struct kunit *test)
 {
 	struct drm_rect src, dst, clip;
@@ -196,11 +207,148 @@ static void drm_test_rect_clip_scaled_signed_vs_unsigned(struct kunit *test)
 	KUNIT_EXPECT_FALSE_MSG(test, drm_rect_visible(&src), "Source should not be visible\n");
 }
 
+struct drm_rect_intersect_case {
+	const char *description;
+	struct drm_rect r1, r2;
+	bool should_be_visible;
+	struct drm_rect expected_intersection;
+};
+
+static const struct drm_rect_intersect_case drm_rect_intersect_cases[] = {
+	{
+		.description = "top-left x bottom-right",
+		.r1 = DRM_RECT_INIT(1, 1, 2, 2),
+		.r2 = DRM_RECT_INIT(0, 0, 2, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 1, 1, 1),
+	},
+	{
+		.description = "top-right x bottom-left",
+		.r1 = DRM_RECT_INIT(0, 0, 2, 2),
+		.r2 = DRM_RECT_INIT(1, -1, 2, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 0, 1, 1),
+	},
+	{
+		.description = "bottom-left x top-right",
+		.r1 = DRM_RECT_INIT(1, -1, 2, 2),
+		.r2 = DRM_RECT_INIT(0, 0, 2, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 0, 1, 1),
+	},
+	{
+		.description = "bottom-right x top-left",
+		.r1 = DRM_RECT_INIT(0, 0, 2, 2),
+		.r2 = DRM_RECT_INIT(1, 1, 2, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 1, 1, 1),
+	},
+	{
+		.description = "right x left",
+		.r1 = DRM_RECT_INIT(0, 0, 2, 1),
+		.r2 = DRM_RECT_INIT(1, 0, 3, 1),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 0, 1, 1),
+	},
+	{
+		.description = "left x right",
+		.r1 = DRM_RECT_INIT(1, 0, 3, 1),
+		.r2 = DRM_RECT_INIT(0, 0, 2, 1),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 0, 1, 1),
+	},
+	{
+		.description = "up x bottom",
+		.r1 = DRM_RECT_INIT(0, 0, 1, 2),
+		.r2 = DRM_RECT_INIT(0, -1, 1, 3),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(0, 0, 1, 2),
+	},
+	{
+		.description = "bottom x up",
+		.r1 = DRM_RECT_INIT(0, -1, 1, 3),
+		.r2 = DRM_RECT_INIT(0, 0, 1, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(0, 0, 1, 2),
+	},
+	{
+		.description = "touching corner",
+		.r1 = DRM_RECT_INIT(0, 0, 1, 1),
+		.r2 = DRM_RECT_INIT(1, 1, 2, 2),
+		.should_be_visible = false,
+		.expected_intersection = DRM_RECT_INIT(1, 1, 0, 0),
+	},
+	{
+		.description = "touching side",
+		.r1 = DRM_RECT_INIT(0, 0, 1, 1),
+		.r2 = DRM_RECT_INIT(1, 0, 1, 1),
+		.should_be_visible = false,
+		.expected_intersection = DRM_RECT_INIT(1, 0, 0, 1),
+	},
+	{
+		.description = "equal rects",
+		.r1 = DRM_RECT_INIT(0, 0, 2, 2),
+		.r2 = DRM_RECT_INIT(0, 0, 2, 2),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(0, 0, 2, 2),
+	},
+	{
+		.description = "inside another",
+		.r1 = DRM_RECT_INIT(0, 0, 2, 2),
+		.r2 = DRM_RECT_INIT(1, 1, 1, 1),
+		.should_be_visible = true,
+		.expected_intersection = DRM_RECT_INIT(1, 1, 1, 1),
+	},
+	{
+		.description = "far away",
+		.r1 = DRM_RECT_INIT(0, 0, 1, 1),
+		.r2 = DRM_RECT_INIT(3, 6, 1, 1),
+		.should_be_visible = false,
+		.expected_intersection = DRM_RECT_INIT(3, 6, -2, -5),
+	},
+	{
+		.description = "points intersecting",
+		.r1 = DRM_RECT_INIT(5, 10, 0, 0),
+		.r2 = DRM_RECT_INIT(5, 10, 0, 0),
+		.should_be_visible = false,
+		.expected_intersection = DRM_RECT_INIT(5, 10, 0, 0),
+	},
+	{
+		.description = "points not intersecting",
+		.r1 = DRM_RECT_INIT(0, 0, 0, 0),
+		.r2 = DRM_RECT_INIT(5, 10, 0, 0),
+		.should_be_visible = false,
+		.expected_intersection = DRM_RECT_INIT(5, 10, -5, -10),
+	},
+};
+
+static void drm_rect_intersect_case_desc(const struct drm_rect_intersect_case *t, char *desc)
+{
+	snprintf(desc, KUNIT_PARAM_DESC_SIZE,
+		 "%s: " DRM_RECT_FMT " x " DRM_RECT_FMT,
+		 t->description, DRM_RECT_ARG(&t->r1), DRM_RECT_ARG(&t->r2));
+}
+
+KUNIT_ARRAY_PARAM(drm_rect_intersect, drm_rect_intersect_cases, drm_rect_intersect_case_desc);
+
+static void drm_test_rect_intersect(struct kunit *test)
+{
+	const struct drm_rect_intersect_case *params = test->param_value;
+	struct drm_rect r1_aux = params->r1;
+	bool visible;
+
+	visible = drm_rect_intersect(&r1_aux, &params->r2);
+
+	KUNIT_EXPECT_EQ(test, visible, params->should_be_visible);
+	drm_rect_compare(test, &r1_aux, &params->expected_intersection);
+}
+
 static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_div_by_zero),
 	KUNIT_CASE(drm_test_rect_clip_scaled_not_clipped),
 	KUNIT_CASE(drm_test_rect_clip_scaled_clipped),
 	KUNIT_CASE(drm_test_rect_clip_scaled_signed_vs_unsigned),
+	KUNIT_CASE_PARAM(drm_test_rect_intersect, drm_rect_intersect_gen_params),
 	{ }
 };
 
-- 
GitLab


From 1f01f2247e3c4ed55c96afccd0072cd31793c604 Mon Sep 17 00:00:00 2001
From: Arthur Grillo <arthurgrillo@riseup.net>
Date: Tue, 18 Apr 2023 20:01:43 -0300
Subject: [PATCH 0252/1791] drm/tests: Add test cases for
 drm_rect_calc_hscale()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Insert parameterized test for the drm_rect_calc_hscale() to ensure
correctness and prevent future regressions. Besides the test for the
usual case, tests the exceptions.

Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418230146.461129-3-arthurgrillo@riseup.net
---
 drivers/gpu/drm/tests/drm_rect_test.c | 72 +++++++++++++++++++++++++++
 1 file changed, 72 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_rect_test.c b/drivers/gpu/drm/tests/drm_rect_test.c
index 33584f63be378..aae682ad9ee1b 100644
--- a/drivers/gpu/drm/tests/drm_rect_test.c
+++ b/drivers/gpu/drm/tests/drm_rect_test.c
@@ -10,6 +10,7 @@
 #include <drm/drm_rect.h>
 
 #include <linux/string_helpers.h>
+#include <linux/errno.h>
 
 static void drm_rect_compare(struct kunit *test, const struct drm_rect *r,
 			     const struct drm_rect *expected)
@@ -343,12 +344,83 @@ static void drm_test_rect_intersect(struct kunit *test)
 	drm_rect_compare(test, &r1_aux, &params->expected_intersection);
 }
 
+struct drm_rect_scale_case {
+	const char *name;
+	struct drm_rect src, dst;
+	int min_range, max_range;
+	int expected_scaling_factor;
+};
+
+static const struct drm_rect_scale_case drm_rect_scale_cases[] = {
+	{
+		.name = "normal use",
+		.src = DRM_RECT_INIT(0, 0, 2 << 16, 2 << 16),
+		.dst = DRM_RECT_INIT(0, 0, 1 << 16, 1 << 16),
+		.min_range = 0, .max_range = INT_MAX,
+		.expected_scaling_factor = 2,
+	},
+	{
+		.name = "out of max range",
+		.src = DRM_RECT_INIT(0, 0, 10 << 16, 10 << 16),
+		.dst = DRM_RECT_INIT(0, 0, 1 << 16, 1 << 16),
+		.min_range = 3, .max_range = 5,
+		.expected_scaling_factor = -ERANGE,
+	},
+	{
+		.name = "out of min range",
+		.src = DRM_RECT_INIT(0, 0, 2 << 16, 2 << 16),
+		.dst = DRM_RECT_INIT(0, 0, 1 << 16, 1 << 16),
+		.min_range = 3, .max_range = 5,
+		.expected_scaling_factor = -ERANGE,
+	},
+	{
+		.name = "zero dst",
+		.src = DRM_RECT_INIT(0, 0, 2 << 16, 2 << 16),
+		.dst = DRM_RECT_INIT(0, 0, 0 << 16, 0 << 16),
+		.min_range = 0, .max_range = INT_MAX,
+		.expected_scaling_factor = 0,
+	},
+	{
+		.name = "negative src",
+		.src = DRM_RECT_INIT(0, 0, -(1 << 16), -(1 << 16)),
+		.dst = DRM_RECT_INIT(0, 0, 1 << 16, 1 << 16),
+		.min_range = 0, .max_range = INT_MAX,
+		.expected_scaling_factor = -EINVAL,
+	},
+	{
+		.name = "negative dst",
+		.src = DRM_RECT_INIT(0, 0, 1 << 16, 1 << 16),
+		.dst = DRM_RECT_INIT(0, 0, -(1 << 16), -(1 << 16)),
+		.min_range = 0, .max_range = INT_MAX,
+		.expected_scaling_factor = -EINVAL,
+	},
+};
+
+static void drm_rect_scale_case_desc(const struct drm_rect_scale_case *t, char *desc)
+{
+	strscpy(desc, t->name, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(drm_rect_scale, drm_rect_scale_cases, drm_rect_scale_case_desc);
+
+static void drm_test_rect_calc_hscale(struct kunit *test)
+{
+	const struct drm_rect_scale_case *params = test->param_value;
+	int scaling_factor;
+
+	scaling_factor = drm_rect_calc_hscale(&params->src, &params->dst,
+					      params->min_range, params->max_range);
+
+	KUNIT_EXPECT_EQ(test, scaling_factor, params->expected_scaling_factor);
+}
+
 static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_div_by_zero),
 	KUNIT_CASE(drm_test_rect_clip_scaled_not_clipped),
 	KUNIT_CASE(drm_test_rect_clip_scaled_clipped),
 	KUNIT_CASE(drm_test_rect_clip_scaled_signed_vs_unsigned),
 	KUNIT_CASE_PARAM(drm_test_rect_intersect, drm_rect_intersect_gen_params),
+	KUNIT_CASE_PARAM(drm_test_rect_calc_hscale, drm_rect_scale_gen_params),
 	{ }
 };
 
-- 
GitLab


From 881e2a9cc11fb53b1de10d951a11a69d23035bbf Mon Sep 17 00:00:00 2001
From: Arthur Grillo <arthurgrillo@riseup.net>
Date: Tue, 18 Apr 2023 20:01:44 -0300
Subject: [PATCH 0253/1791] drm/tests: Add test cases for
 drm_rect_calc_vscale()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Insert parameterized test for the drm_rect_calc_vscale() to ensure
correctness and prevent future regressions. Besides the test for the
usual case, tests the exceptions.

It uses the same struct from drm_rect_calc_hscale().

Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418230146.461129-4-arthurgrillo@riseup.net
---
 drivers/gpu/drm/tests/drm_rect_test.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_rect_test.c b/drivers/gpu/drm/tests/drm_rect_test.c
index aae682ad9ee1b..fd89e62ad2cb8 100644
--- a/drivers/gpu/drm/tests/drm_rect_test.c
+++ b/drivers/gpu/drm/tests/drm_rect_test.c
@@ -414,6 +414,17 @@ static void drm_test_rect_calc_hscale(struct kunit *test)
 	KUNIT_EXPECT_EQ(test, scaling_factor, params->expected_scaling_factor);
 }
 
+static void drm_test_rect_calc_vscale(struct kunit *test)
+{
+	const struct drm_rect_scale_case *params = test->param_value;
+	int scaling_factor;
+
+	scaling_factor = drm_rect_calc_vscale(&params->src, &params->dst,
+					      params->min_range, params->max_range);
+
+	KUNIT_EXPECT_EQ(test, scaling_factor, params->expected_scaling_factor);
+}
+
 static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_div_by_zero),
 	KUNIT_CASE(drm_test_rect_clip_scaled_not_clipped),
@@ -421,6 +432,7 @@ static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_signed_vs_unsigned),
 	KUNIT_CASE_PARAM(drm_test_rect_intersect, drm_rect_intersect_gen_params),
 	KUNIT_CASE_PARAM(drm_test_rect_calc_hscale, drm_rect_scale_gen_params),
+	KUNIT_CASE_PARAM(drm_test_rect_calc_vscale, drm_rect_scale_gen_params),
 	{ }
 };
 
-- 
GitLab


From b5d88ec0ee3d0c75d2eb515efc335d031134bd75 Mon Sep 17 00:00:00 2001
From: Arthur Grillo <arthurgrillo@riseup.net>
Date: Tue, 18 Apr 2023 20:01:45 -0300
Subject: [PATCH 0254/1791] drm/tests: Add test cases for drm_rect_rotate()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Insert a parameterized test for the drm_rect_rotate() to ensure
correctness and prevent future regressions.

All possible rotation modes are covered by the test.

Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418230146.461129-5-arthurgrillo@riseup.net
---
 drivers/gpu/drm/tests/drm_rect_test.c | 72 +++++++++++++++++++++++++++
 1 file changed, 72 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_rect_test.c b/drivers/gpu/drm/tests/drm_rect_test.c
index fd89e62ad2cb8..514b9b3337e99 100644
--- a/drivers/gpu/drm/tests/drm_rect_test.c
+++ b/drivers/gpu/drm/tests/drm_rect_test.c
@@ -8,6 +8,7 @@
 #include <kunit/test.h>
 
 #include <drm/drm_rect.h>
+#include <drm/drm_mode.h>
 
 #include <linux/string_helpers.h>
 #include <linux/errno.h>
@@ -425,6 +426,76 @@ static void drm_test_rect_calc_vscale(struct kunit *test)
 	KUNIT_EXPECT_EQ(test, scaling_factor, params->expected_scaling_factor);
 }
 
+struct drm_rect_rotate_case {
+	const char *name;
+	unsigned int rotation;
+	struct drm_rect rect;
+	int width, height;
+	struct drm_rect expected;
+};
+
+static const struct drm_rect_rotate_case drm_rect_rotate_cases[] = {
+	{
+		.name = "reflect-x",
+		.rotation = DRM_MODE_REFLECT_X,
+		.rect = DRM_RECT_INIT(0, 0, 5, 5),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(0, 0, 5, 5),
+	},
+	{
+		.name = "reflect-y",
+		.rotation = DRM_MODE_REFLECT_Y,
+		.rect = DRM_RECT_INIT(2, 0, 5, 5),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(2, 5, 5, 5),
+	},
+	{
+		.name = "rotate-0",
+		.rotation = DRM_MODE_ROTATE_0,
+		.rect = DRM_RECT_INIT(0, 2, 5, 5),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(0, 2, 5, 5),
+	},
+	{
+		.name = "rotate-90",
+		.rotation = DRM_MODE_ROTATE_90,
+		.rect = DRM_RECT_INIT(0, 0, 5, 10),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(0, 0, 10, 5),
+	},
+	{
+		.name = "rotate-180",
+		.rotation = DRM_MODE_ROTATE_180,
+		.rect = DRM_RECT_INIT(11, 3, 5, 10),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(-11, -3, 5, 10),
+	},
+	{
+		.name = "rotate-270",
+		.rotation = DRM_MODE_ROTATE_270,
+		.rect = DRM_RECT_INIT(6, 3, 5, 10),
+		.width = 5, .height = 10,
+		.expected = DRM_RECT_INIT(-3, 6, 10, 5),
+	},
+};
+
+static void drm_rect_rotate_case_desc(const struct drm_rect_rotate_case *t, char *desc)
+{
+	strscpy(desc, t->name, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(drm_rect_rotate, drm_rect_rotate_cases, drm_rect_rotate_case_desc);
+
+static void drm_test_rect_rotate(struct kunit *test)
+{
+	const struct drm_rect_rotate_case *params = test->param_value;
+	struct drm_rect r = params->rect;
+
+	drm_rect_rotate(&r, params->width, params->height, params->rotation);
+
+	drm_rect_compare(test, &r, &params->expected);
+}
+
 static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_div_by_zero),
 	KUNIT_CASE(drm_test_rect_clip_scaled_not_clipped),
@@ -433,6 +504,7 @@ static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE_PARAM(drm_test_rect_intersect, drm_rect_intersect_gen_params),
 	KUNIT_CASE_PARAM(drm_test_rect_calc_hscale, drm_rect_scale_gen_params),
 	KUNIT_CASE_PARAM(drm_test_rect_calc_vscale, drm_rect_scale_gen_params),
+	KUNIT_CASE_PARAM(drm_test_rect_rotate, drm_rect_rotate_gen_params),
 	{ }
 };
 
-- 
GitLab


From 8356b97906503a02125c8d03c9b88a61ea46a05a Mon Sep 17 00:00:00 2001
From: Arthur Grillo <arthurgrillo@riseup.net>
Date: Tue, 18 Apr 2023 20:01:46 -0300
Subject: [PATCH 0255/1791] drm/test: Add test cases for drm_rect_rotate_inv()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Insert a parameterized test for the drm_rect_rotate_inv() to ensure its
correctness and prevent future regressions. The test covers all rotation
modes.

It uses the same test cases from drm_test_rect_rotate().

Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418230146.461129-6-arthurgrillo@riseup.net
---
 drivers/gpu/drm/tests/drm_rect_test.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/tests/drm_rect_test.c b/drivers/gpu/drm/tests/drm_rect_test.c
index 514b9b3337e99..76332cd2ead83 100644
--- a/drivers/gpu/drm/tests/drm_rect_test.c
+++ b/drivers/gpu/drm/tests/drm_rect_test.c
@@ -496,6 +496,16 @@ static void drm_test_rect_rotate(struct kunit *test)
 	drm_rect_compare(test, &r, &params->expected);
 }
 
+static void drm_test_rect_rotate_inv(struct kunit *test)
+{
+	const struct drm_rect_rotate_case *params = test->param_value;
+	struct drm_rect r = params->expected;
+
+	drm_rect_rotate_inv(&r, params->width, params->height, params->rotation);
+
+	drm_rect_compare(test, &r, &params->rect);
+}
+
 static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE(drm_test_rect_clip_scaled_div_by_zero),
 	KUNIT_CASE(drm_test_rect_clip_scaled_not_clipped),
@@ -505,6 +515,7 @@ static struct kunit_case drm_rect_tests[] = {
 	KUNIT_CASE_PARAM(drm_test_rect_calc_hscale, drm_rect_scale_gen_params),
 	KUNIT_CASE_PARAM(drm_test_rect_calc_vscale, drm_rect_scale_gen_params),
 	KUNIT_CASE_PARAM(drm_test_rect_rotate, drm_rect_rotate_gen_params),
+	KUNIT_CASE_PARAM(drm_test_rect_rotate_inv, drm_rect_rotate_gen_params),
 	{ }
 };
 
-- 
GitLab


From 322d716a3e8a74fb75cd0f657647be4df253fd2f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:21 -0300
Subject: [PATCH 0256/1791] drm/vkms: isolate pixel conversion functionality
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, the pixel conversion functions repeat the same loop to
iterate the rows. Instead of repeating the same code for each pixel
format, create a function to wrap the loop and isolate the pixel
conversion functionality.

Suggested-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-2-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_composer.c |   4 +-
 drivers/gpu/drm/vkms/vkms_drv.h      |   4 +-
 drivers/gpu/drm/vkms/vkms_formats.c  | 125 +++++++++++----------------
 drivers/gpu/drm/vkms/vkms_formats.h  |   2 +-
 drivers/gpu/drm/vkms/vkms_plane.c    |   2 +-
 5 files changed, 56 insertions(+), 81 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index 8e53fa80742b2..80164e79af006 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -99,7 +99,7 @@ static void blend(struct vkms_writeback_job *wb,
 			if (!check_y_limit(plane[i]->frame_info, y))
 				continue;
 
-			plane[i]->plane_read(stage_buffer, plane[i]->frame_info, y);
+			vkms_compose_row(stage_buffer, plane[i], y);
 			pre_mul_alpha_blend(plane[i]->frame_info, stage_buffer,
 					    output_buffer);
 		}
@@ -118,7 +118,7 @@ static int check_format_funcs(struct vkms_crtc_state *crtc_state,
 	u32 n_active_planes = crtc_state->num_active_planes;
 
 	for (size_t i = 0; i < n_active_planes; i++)
-		if (!planes[i]->plane_read)
+		if (!planes[i]->pixel_read)
 			return -1;
 
 	if (active_wb && !active_wb->wb_write)
diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h
index 4a248567efb26..f152d54baf769 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.h
+++ b/drivers/gpu/drm/vkms/vkms_drv.h
@@ -56,8 +56,7 @@ struct vkms_writeback_job {
 struct vkms_plane_state {
 	struct drm_shadow_plane_state base;
 	struct vkms_frame_info *frame_info;
-	void (*plane_read)(struct line_buffer *buffer,
-			   const struct vkms_frame_info *frame_info, int y);
+	void (*pixel_read)(u8 *src_buffer, struct pixel_argb_u16 *out_pixel);
 };
 
 struct vkms_plane {
@@ -155,6 +154,7 @@ int vkms_verify_crc_source(struct drm_crtc *crtc, const char *source_name,
 /* Composer Support */
 void vkms_composer_worker(struct work_struct *work);
 void vkms_set_composer(struct vkms_output *out, bool enabled);
+void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state *plane, int y);
 
 /* Writeback */
 int vkms_enable_writeback_connector(struct vkms_device *vkmsdev);
diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index d4950688b3f17..8d948c73741ef 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -42,100 +42,75 @@ static void *get_packed_src_addr(const struct vkms_frame_info *frame_info, int y
 	return packed_pixels_addr(frame_info, x_src, y_src);
 }
 
-static void ARGB8888_to_argb_u16(struct line_buffer *stage_buffer,
-				 const struct vkms_frame_info *frame_info, int y)
+static void ARGB8888_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
-	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
-	u8 *src_pixels = get_packed_src_addr(frame_info, y);
-	int x_limit = min_t(size_t, drm_rect_width(&frame_info->dst),
-			    stage_buffer->n_pixels);
-
-	for (size_t x = 0; x < x_limit; x++, src_pixels += 4) {
-		/*
-		 * The 257 is the "conversion ratio". This number is obtained by the
-		 * (2^16 - 1) / (2^8 - 1) division. Which, in this case, tries to get
-		 * the best color value in a pixel format with more possibilities.
-		 * A similar idea applies to others RGB color conversions.
-		 */
-		out_pixels[x].a = (u16)src_pixels[3] * 257;
-		out_pixels[x].r = (u16)src_pixels[2] * 257;
-		out_pixels[x].g = (u16)src_pixels[1] * 257;
-		out_pixels[x].b = (u16)src_pixels[0] * 257;
-	}
+	/*
+	 * The 257 is the "conversion ratio". This number is obtained by the
+	 * (2^16 - 1) / (2^8 - 1) division. Which, in this case, tries to get
+	 * the best color value in a pixel format with more possibilities.
+	 * A similar idea applies to others RGB color conversions.
+	 */
+	out_pixel->a = (u16)src_pixels[3] * 257;
+	out_pixel->r = (u16)src_pixels[2] * 257;
+	out_pixel->g = (u16)src_pixels[1] * 257;
+	out_pixel->b = (u16)src_pixels[0] * 257;
 }
 
-static void XRGB8888_to_argb_u16(struct line_buffer *stage_buffer,
-				 const struct vkms_frame_info *frame_info, int y)
+static void XRGB8888_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
-	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
-	u8 *src_pixels = get_packed_src_addr(frame_info, y);
-	int x_limit = min_t(size_t, drm_rect_width(&frame_info->dst),
-			    stage_buffer->n_pixels);
-
-	for (size_t x = 0; x < x_limit; x++, src_pixels += 4) {
-		out_pixels[x].a = (u16)0xffff;
-		out_pixels[x].r = (u16)src_pixels[2] * 257;
-		out_pixels[x].g = (u16)src_pixels[1] * 257;
-		out_pixels[x].b = (u16)src_pixels[0] * 257;
-	}
+	out_pixel->a = (u16)0xffff;
+	out_pixel->r = (u16)src_pixels[2] * 257;
+	out_pixel->g = (u16)src_pixels[1] * 257;
+	out_pixel->b = (u16)src_pixels[0] * 257;
 }
 
-static void ARGB16161616_to_argb_u16(struct line_buffer *stage_buffer,
-				     const struct vkms_frame_info *frame_info,
-				     int y)
+static void ARGB16161616_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
-	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
-	u16 *src_pixels = get_packed_src_addr(frame_info, y);
-	int x_limit = min_t(size_t, drm_rect_width(&frame_info->dst),
-			    stage_buffer->n_pixels);
+	u16 *pixels = (u16 *)src_pixels;
 
-	for (size_t x = 0; x < x_limit; x++, src_pixels += 4) {
-		out_pixels[x].a = le16_to_cpu(src_pixels[3]);
-		out_pixels[x].r = le16_to_cpu(src_pixels[2]);
-		out_pixels[x].g = le16_to_cpu(src_pixels[1]);
-		out_pixels[x].b = le16_to_cpu(src_pixels[0]);
-	}
+	out_pixel->a = le16_to_cpu(pixels[3]);
+	out_pixel->r = le16_to_cpu(pixels[2]);
+	out_pixel->g = le16_to_cpu(pixels[1]);
+	out_pixel->b = le16_to_cpu(pixels[0]);
 }
 
-static void XRGB16161616_to_argb_u16(struct line_buffer *stage_buffer,
-				     const struct vkms_frame_info *frame_info,
-				     int y)
+static void XRGB16161616_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
-	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
-	u16 *src_pixels = get_packed_src_addr(frame_info, y);
-	int x_limit = min_t(size_t, drm_rect_width(&frame_info->dst),
-			    stage_buffer->n_pixels);
+	u16 *pixels = (u16 *)src_pixels;
 
-	for (size_t x = 0; x < x_limit; x++, src_pixels += 4) {
-		out_pixels[x].a = (u16)0xffff;
-		out_pixels[x].r = le16_to_cpu(src_pixels[2]);
-		out_pixels[x].g = le16_to_cpu(src_pixels[1]);
-		out_pixels[x].b = le16_to_cpu(src_pixels[0]);
-	}
+	out_pixel->a = (u16)0xffff;
+	out_pixel->r = le16_to_cpu(pixels[2]);
+	out_pixel->g = le16_to_cpu(pixels[1]);
+	out_pixel->b = le16_to_cpu(pixels[0]);
 }
 
-static void RGB565_to_argb_u16(struct line_buffer *stage_buffer,
-			       const struct vkms_frame_info *frame_info, int y)
+static void RGB565_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
-	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
-	u16 *src_pixels = get_packed_src_addr(frame_info, y);
-	int x_limit = min_t(size_t, drm_rect_width(&frame_info->dst),
-			       stage_buffer->n_pixels);
+	u16 *pixels = (u16 *)src_pixels;
 
 	s64 fp_rb_ratio = drm_fixp_div(drm_int2fixp(65535), drm_int2fixp(31));
 	s64 fp_g_ratio = drm_fixp_div(drm_int2fixp(65535), drm_int2fixp(63));
 
-	for (size_t x = 0; x < x_limit; x++, src_pixels++) {
-		u16 rgb_565 = le16_to_cpu(*src_pixels);
-		s64 fp_r = drm_int2fixp((rgb_565 >> 11) & 0x1f);
-		s64 fp_g = drm_int2fixp((rgb_565 >> 5) & 0x3f);
-		s64 fp_b = drm_int2fixp(rgb_565 & 0x1f);
+	u16 rgb_565 = le16_to_cpu(*pixels);
+	s64 fp_r = drm_int2fixp((rgb_565 >> 11) & 0x1f);
+	s64 fp_g = drm_int2fixp((rgb_565 >> 5) & 0x3f);
+	s64 fp_b = drm_int2fixp(rgb_565 & 0x1f);
 
-		out_pixels[x].a = (u16)0xffff;
-		out_pixels[x].r = drm_fixp2int(drm_fixp_mul(fp_r, fp_rb_ratio));
-		out_pixels[x].g = drm_fixp2int(drm_fixp_mul(fp_g, fp_g_ratio));
-		out_pixels[x].b = drm_fixp2int(drm_fixp_mul(fp_b, fp_rb_ratio));
-	}
+	out_pixel->a = (u16)0xffff;
+	out_pixel->r = drm_fixp2int(drm_fixp_mul(fp_r, fp_rb_ratio));
+	out_pixel->g = drm_fixp2int(drm_fixp_mul(fp_g, fp_g_ratio));
+	out_pixel->b = drm_fixp2int(drm_fixp_mul(fp_b, fp_rb_ratio));
+}
+
+void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state *plane, int y)
+{
+	struct pixel_argb_u16 *out_pixels = stage_buffer->pixels;
+	struct vkms_frame_info *frame_info = plane->frame_info;
+	u8 *src_pixels = get_packed_src_addr(frame_info, y);
+	int limit = min_t(size_t, drm_rect_width(&frame_info->dst), stage_buffer->n_pixels);
+
+	for (size_t x = 0; x < limit; x++, src_pixels += frame_info->cpp)
+		plane->pixel_read(src_pixels, &out_pixels[x]);
 }
 
 /*
@@ -249,7 +224,7 @@ static void argb_u16_to_RGB565(struct vkms_frame_info *frame_info,
 	}
 }
 
-void *get_frame_to_line_function(u32 format)
+void *get_pixel_conversion_function(u32 format)
 {
 	switch (format) {
 	case DRM_FORMAT_ARGB8888:
diff --git a/drivers/gpu/drm/vkms/vkms_formats.h b/drivers/gpu/drm/vkms/vkms_formats.h
index 43b7c19790181..c5b113495d0c0 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.h
+++ b/drivers/gpu/drm/vkms/vkms_formats.h
@@ -5,7 +5,7 @@
 
 #include "vkms_drv.h"
 
-void *get_frame_to_line_function(u32 format);
+void *get_pixel_conversion_function(u32 format);
 
 void *get_line_to_frame_function(u32 format);
 
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index c41cec7dcb703..0a23875900ec5 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -123,7 +123,7 @@ static void vkms_plane_atomic_update(struct drm_plane *plane,
 	frame_info->offset = fb->offsets[0];
 	frame_info->pitch = fb->pitches[0];
 	frame_info->cpp = fb->format->cpp[0];
-	vkms_plane_state->plane_read = get_frame_to_line_function(fmt);
+	vkms_plane_state->pixel_read = get_pixel_conversion_function(fmt);
 }
 
 static int vkms_plane_atomic_check(struct drm_plane *plane,
-- 
GitLab


From 89d8445e4f5cfaf5c017c0bafb05f8bbaac59bf9 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 20 Mar 2023 12:22:49 -0400
Subject: [PATCH 0257/1791] drm/amdgpu/gfx11: add FW version check for new CP
 GFX shadow feature

Use this to determine if we support the new SET_Q_PREEMPTION_MODE
packet.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  2 ++
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  | 13 +++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index a2d311a4da5ae..6c5f58a34e842 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -380,6 +380,8 @@ struct amdgpu_gfx {
 	struct amdgpu_ring		sw_gfx_ring[AMDGPU_MAX_SW_GFX_RINGS];
 	struct amdgpu_ring_mux          muxer;
 
+	bool				cp_gfx_shadow; /* for gfx11 */
+
 	enum amdgpu_gfx_partition	partition_mode;
 	uint32_t			num_xcd;
 	uint32_t			num_xcc_per_xcp;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index d3c89e6c0c035..f1ad6bd1eda80 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -463,6 +463,18 @@ static int gfx_v11_0_init_toc_microcode(struct amdgpu_device *adev, const char *
 	return err;
 }
 
+static void gfx_v11_0_check_fw_cp_gfx_shadow(struct amdgpu_device *adev)
+{
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(11, 0, 0):
+	case IP_VERSION(11, 0, 2):
+	case IP_VERSION(11, 0, 3):
+	default:
+		adev->gfx.cp_gfx_shadow = false;
+		break;
+	}
+}
+
 static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 {
 	char fw_name[40];
@@ -539,6 +551,7 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev)
 	/* only one MEC for gfx 11.0.0. */
 	adev->gfx.mec2_fw = NULL;
 
+	gfx_v11_0_check_fw_cp_gfx_shadow(adev);
 out:
 	if (err) {
 		amdgpu_ucode_release(&adev->gfx.pfp_fw);
-- 
GitLab


From 31d7c3a4fc3d312a0646990767647925d5bde540 Mon Sep 17 00:00:00 2001
From: Jack Xiao <Jack.Xiao@amd.com>
Date: Fri, 21 Apr 2023 14:20:38 +0800
Subject: [PATCH 0258/1791] drm/amdgpu: fix memory leak in mes self test

The fences associated with mes queue have to be freed
up during amdgpu_ring_fini.

Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index f676c236b657b..eaf0f82757ef1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -390,6 +390,8 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 		amdgpu_bo_free_kernel(&ring->ring_obj,
 				      &ring->gpu_addr,
 				      (void **)&ring->ring);
+	} else {
+		kfree(ring->fence_drv.fences);
 	}
 
 	dma_fence_put(ring->vmid_wait);
-- 
GitLab


From b418e7193965a5f064a0e422ccf25e6203f265b5 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 20 Mar 2023 13:22:44 -0400
Subject: [PATCH 0259/1791] drm/amdgpu/gfx11: check the CP FW version CP GFX
 shadow support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Only set the supported flag if we have new enough CP FW.

v2: update to the final firmware versions

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index f1ad6bd1eda80..2b63380ec75aa 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -469,6 +469,11 @@ static void gfx_v11_0_check_fw_cp_gfx_shadow(struct amdgpu_device *adev)
 	case IP_VERSION(11, 0, 0):
 	case IP_VERSION(11, 0, 2):
 	case IP_VERSION(11, 0, 3):
+		if ((adev->gfx.me_fw_version >= 1505) &&
+		    (adev->gfx.pfp_fw_version >= 1600) &&
+		    (adev->gfx.mec_fw_version >= 512))
+			adev->gfx.cp_gfx_shadow = true;
+		break;
 	default:
 		adev->gfx.cp_gfx_shadow = false;
 		break;
-- 
GitLab


From 043dc33f443fd7abaf3fe076897503ce3d5dbc26 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 9 Mar 2023 13:48:11 -0500
Subject: [PATCH 0260/1791] drm/amdgpu/UAPI: add new CS chunk for GFX shadow
 buffers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For GFX11, the UMD needs to allocate some shadow buffers
to be used for preemption.  The UMD allocates the buffers
and passes the GPU virtual address to the kernel since the
kernel will program the packet that specified these
addresses as part of its IB submission frame.

v2: UMD passes shadow init to tell kernel when to initialize
    the shadow

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/uapi/drm/amdgpu_drm.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 6981e59a94017..fc094653b13f2 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -592,6 +592,7 @@ struct drm_amdgpu_gem_va {
 #define AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES	0x07
 #define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT    0x08
 #define AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL  0x09
+#define AMDGPU_CHUNK_ID_CP_GFX_SHADOW   0x0a
 
 struct drm_amdgpu_cs_chunk {
 	__u32		chunk_id;
@@ -708,6 +709,15 @@ struct drm_amdgpu_cs_chunk_data {
 	};
 };
 
+#define AMDGPU_CS_CHUNK_CP_GFX_SHADOW_FLAGS_INIT_SHADOW         0x1
+
+struct drm_amdgpu_cs_chunk_cp_gfx_shadow {
+	__u64 shadow_va;
+	__u64 csa_va;
+	__u64 gds_va;
+	__u64 flags;
+};
+
 /*
  *  Query h/w info: Flag that this is integrated (a.h.a. fusion) GPU
  *
-- 
GitLab


From ac9287055ff16a092416c76a19006764e4c6a978 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 9 Mar 2023 15:40:48 -0500
Subject: [PATCH 0261/1791] drm/amdgpu: add gfx shadow CS IOCTL support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add support for submitting the shadow update packet
when submitting an IB.  Needed for MCBP on GFX11.

v2: update API for CSA (Alex)
v3: fix ordering; SET_Q_PREEMPTION_MODE most come before COND_EXEC
    Add missing check for AMDGPU_CHUNK_ID_CP_GFX_SHADOW in
    amdgpu_cs_pass1()
    Only initialize shadow on first use
    (Alex)
v4: Pass parameters rather than job to new ring callback (Alex)
v5: squash in change to call SET_Q_PREEMPTION_MODE/COND_EXEC
    before RELEASE_MEM to complete the UMDs use of the shadow (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c   | 26 +++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c   | 29 +++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.h  |  6 +++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h |  3 +++
 4 files changed, 63 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index a8ec5ff41fc05..c3b3287dc29ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -288,6 +288,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 		case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT:
 		case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL:
+		case AMDGPU_CHUNK_ID_CP_GFX_SHADOW:
 			break;
 
 		default:
@@ -578,6 +579,26 @@ static int amdgpu_cs_p2_syncobj_timeline_signal(struct amdgpu_cs_parser *p,
 	return 0;
 }
 
+static int amdgpu_cs_p2_shadow(struct amdgpu_cs_parser *p,
+			       struct amdgpu_cs_chunk *chunk)
+{
+	struct drm_amdgpu_cs_chunk_cp_gfx_shadow *shadow = chunk->kdata;
+	int i;
+
+	if (shadow->flags & ~AMDGPU_CS_CHUNK_CP_GFX_SHADOW_FLAGS_INIT_SHADOW)
+		return -EINVAL;
+
+	for (i = 0; i < p->gang_size; ++i) {
+		p->jobs[i]->shadow_va = shadow->shadow_va;
+		p->jobs[i]->csa_va = shadow->csa_va;
+		p->jobs[i]->gds_va = shadow->gds_va;
+		p->jobs[i]->init_shadow =
+			shadow->flags & AMDGPU_CS_CHUNK_CP_GFX_SHADOW_FLAGS_INIT_SHADOW;
+	}
+
+	return 0;
+}
+
 static int amdgpu_cs_pass2(struct amdgpu_cs_parser *p)
 {
 	unsigned int ce_preempt = 0, de_preempt = 0;
@@ -620,6 +641,11 @@ static int amdgpu_cs_pass2(struct amdgpu_cs_parser *p)
 			if (r)
 				return r;
 			break;
+		case AMDGPU_CHUNK_ID_CP_GFX_SHADOW:
+			r = amdgpu_cs_p2_shadow(p, chunk);
+			if (r)
+				return r;
+			break;
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index ef96ff2f42729..aebc0e5bddc6c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -136,7 +136,9 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	uint64_t fence_ctx;
 	uint32_t status = 0, alloc_size;
 	unsigned fence_flags = 0;
-	bool secure;
+	bool secure, init_shadow;
+	u64 shadow_va, csa_va, gds_va;
+	int vmid = AMDGPU_JOB_GET_VMID(job);
 
 	unsigned i;
 	int r = 0;
@@ -150,9 +152,17 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		vm = job->vm;
 		fence_ctx = job->base.s_fence ?
 			job->base.s_fence->scheduled.context : 0;
+		shadow_va = job->shadow_va;
+		csa_va = job->csa_va;
+		gds_va = job->gds_va;
+		init_shadow = job->init_shadow;
 	} else {
 		vm = NULL;
 		fence_ctx = 0;
+		shadow_va = 0;
+		csa_va = 0;
+		gds_va = 0;
+		init_shadow = false;
 	}
 
 	if (!ring->sched.ready && !ring->is_mes_queue) {
@@ -212,6 +222,11 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	}
 
 	amdgpu_ring_ib_begin(ring);
+
+	if (job && ring->funcs->emit_gfx_shadow)
+		amdgpu_ring_emit_gfx_shadow(ring, shadow_va, csa_va, gds_va,
+					    init_shadow, vmid);
+
 	if (job && ring->funcs->init_cond_exec)
 		patch_offset = amdgpu_ring_init_cond_exec(ring);
 
@@ -263,6 +278,18 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 				       fence_flags | AMDGPU_FENCE_FLAG_64BIT);
 	}
 
+	if (ring->funcs->emit_gfx_shadow) {
+		amdgpu_ring_emit_gfx_shadow(ring, 0, 0, 0, false, 0);
+
+		if (ring->funcs->init_cond_exec) {
+			unsigned ce_offset = ~0;
+
+			ce_offset = amdgpu_ring_init_cond_exec(ring);
+			if (ce_offset != ~0 && ring->funcs->patch_cond_exec)
+				amdgpu_ring_patch_cond_exec(ring, ce_offset);
+		}
+	}
+
 	r = amdgpu_fence_emit(ring, f, job, fence_flags);
 	if (r) {
 		dev_err(adev->dev, "failed to emit fence (%d)\n", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index 52f2e313ea17f..3f9804f956c9c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -67,6 +67,12 @@ struct amdgpu_job {
 	uint64_t		uf_addr;
 	uint64_t		uf_sequence;
 
+	/* virtual addresses for shadow/GDS/CSA */
+	uint64_t		shadow_va;
+	uint64_t		csa_va;
+	uint64_t		gds_va;
+	bool			init_shadow;
+
 	/* job_run_counter >= 1 means a resubmit job */
 	uint32_t		job_run_counter;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 8eca6532ed196..2aa6cc1c4212c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -212,6 +212,8 @@ struct amdgpu_ring_funcs {
 	void (*end_use)(struct amdgpu_ring *ring);
 	void (*emit_switch_buffer) (struct amdgpu_ring *ring);
 	void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags);
+	void (*emit_gfx_shadow)(struct amdgpu_ring *ring, u64 shadow_va, u64 csa_va,
+				u64 gds_va, bool init_shadow, int vmid);
 	void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg,
 			  uint32_t reg_val_offs);
 	void (*emit_wreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
@@ -309,6 +311,7 @@ struct amdgpu_ring {
 #define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r))
 #define amdgpu_ring_emit_switch_buffer(r) (r)->funcs->emit_switch_buffer((r))
 #define amdgpu_ring_emit_cntxcntl(r, d) (r)->funcs->emit_cntxcntl((r), (d))
+#define amdgpu_ring_emit_gfx_shadow(r, s, c, g, i, v) ((r)->funcs->emit_gfx_shadow((r), (s), (c), (g), (i), (v)))
 #define amdgpu_ring_emit_rreg(r, d, o) (r)->funcs->emit_rreg((r), (d), (o))
 #define amdgpu_ring_emit_wreg(r, d, v) (r)->funcs->emit_wreg((r), (d), (v))
 #define amdgpu_ring_emit_reg_wait(r, d, v, m) (r)->funcs->emit_reg_wait((r), (d), (v), (m))
-- 
GitLab


From 46c1282e5a31c33a6973affbcc765eac1f3f5d1c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 8 Mar 2023 21:17:59 +0100
Subject: [PATCH 0262/1791] drm/amdgpu: add gfx11 emit shadow callback
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add ring callback for gfx to update the CP firmware
with the new shadow information before we process the
IB.

v2: add implementation for new packet (Alex)
v3: add current FW version checks (Alex)
v4: only initialize shadow on first use
    Only set IB_VMID when a valid shadow buffer is present
    (Alex)
v5: Pass parameters rather than job to new ring callback (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 25 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/nvd.h       |  5 ++++-
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 2b63380ec75aa..b8b2886aac92f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -5608,6 +5608,29 @@ static void gfx_v11_0_ring_emit_cntxcntl(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, 0);
 }
 
+static void gfx_v11_0_ring_emit_gfx_shadow(struct amdgpu_ring *ring,
+					   u64 shadow_va, u64 csa_va,
+					   u64 gds_va, bool init_shadow,
+					   int vmid)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (!adev->gfx.cp_gfx_shadow)
+		return;
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_SET_Q_PREEMPTION_MODE, 7));
+	amdgpu_ring_write(ring, lower_32_bits(shadow_va));
+	amdgpu_ring_write(ring, upper_32_bits(shadow_va));
+	amdgpu_ring_write(ring, lower_32_bits(gds_va));
+	amdgpu_ring_write(ring, upper_32_bits(gds_va));
+	amdgpu_ring_write(ring, lower_32_bits(csa_va));
+	amdgpu_ring_write(ring, upper_32_bits(csa_va));
+	amdgpu_ring_write(ring, shadow_va ?
+			  PACKET3_SET_Q_PREEMPTION_MODE_IB_VMID(vmid) : 0);
+	amdgpu_ring_write(ring, init_shadow ?
+			  PACKET3_SET_Q_PREEMPTION_MODE_INIT_SHADOW_MEM : 0);
+}
+
 static unsigned gfx_v11_0_ring_emit_init_cond_exec(struct amdgpu_ring *ring)
 {
 	unsigned ret;
@@ -6228,6 +6251,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = {
 	.set_wptr = gfx_v11_0_ring_set_wptr_gfx,
 	.emit_frame_size = /* totally 242 maximum if 16 IBs */
 		5 + /* COND_EXEC */
+		9 + /* SET_Q_PREEMPTION_MODE */
 		7 + /* PIPELINE_SYNC */
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 +
@@ -6254,6 +6278,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = {
 	.insert_nop = amdgpu_ring_insert_nop,
 	.pad_ib = amdgpu_ring_generic_pad_ib,
 	.emit_cntxcntl = gfx_v11_0_ring_emit_cntxcntl,
+	.emit_gfx_shadow = gfx_v11_0_ring_emit_gfx_shadow,
 	.init_cond_exec = gfx_v11_0_ring_emit_init_cond_exec,
 	.patch_cond_exec = gfx_v11_0_ring_emit_patch_cond_exec,
 	.preempt_ib = gfx_v11_0_ring_preempt_ib,
diff --git a/drivers/gpu/drm/amd/amdgpu/nvd.h b/drivers/gpu/drm/amd/amdgpu/nvd.h
index fd6b58243b032..631dafb922990 100644
--- a/drivers/gpu/drm/amd/amdgpu/nvd.h
+++ b/drivers/gpu/drm/amd/amdgpu/nvd.h
@@ -462,6 +462,9 @@
 #              define PACKET3_QUERY_STATUS_ENG_SEL(x)          ((x) << 25)
 #define	PACKET3_RUN_LIST				0xA5
 #define	PACKET3_MAP_PROCESS_VM				0xA6
-
+/* GFX11 */
+#define	PACKET3_SET_Q_PREEMPTION_MODE			0xF0
+#              define PACKET3_SET_Q_PREEMPTION_MODE_IB_VMID(x)  ((x) << 0)
+#              define PACKET3_SET_Q_PREEMPTION_MODE_INIT_SHADOW_MEM    (1 << 0)
 
 #endif
-- 
GitLab


From 38be7796f310cd2bc84dcc40c4fd1964df39a5b0 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 16 Mar 2023 11:33:43 -0400
Subject: [PATCH 0263/1791] drm/amdgpu: don't require a job for cond_exec and
 shadow
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We need to reset the shadow state every time we submit an
IB and there needs to be a COND_EXEC packet after the
SET_Q_PREEMPTION_MODE packet for it to work properly, so
we should emit both of these packets regardless of whether
there is a job present or not.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index aebc0e5bddc6c..c955c3f060cd2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -223,11 +223,11 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 
 	amdgpu_ring_ib_begin(ring);
 
-	if (job && ring->funcs->emit_gfx_shadow)
+	if (ring->funcs->emit_gfx_shadow)
 		amdgpu_ring_emit_gfx_shadow(ring, shadow_va, csa_va, gds_va,
 					    init_shadow, vmid);
 
-	if (job && ring->funcs->init_cond_exec)
+	if (ring->funcs->init_cond_exec)
 		patch_offset = amdgpu_ring_init_cond_exec(ring);
 
 	amdgpu_device_flush_hdp(adev, ring);
-- 
GitLab


From edd9038000352ba846cba9dfb84d8c397c3b6499 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 9 Mar 2023 13:43:13 -0500
Subject: [PATCH 0264/1791] drm/amdgpu: add UAPI to query GFX shadow sizes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add UAPI to query the GFX shadow buffer requirements
for preemption on GFX11.  UMDs need to specify the shadow
areas for preemption.

v2: move into existing asic info query
    drop GDS as its use is determined by the UMD (Marek)
v3: Update comments to note that alignment is base
    virtual alignment (Alex)

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/uapi/drm/amdgpu_drm.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index fc094653b13f2..cc78528c3b4bc 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -1138,6 +1138,14 @@ struct drm_amdgpu_info_device {
 	__u64 mall_size;            /* AKA infinity cache */
 	/* high 32 bits of the rb pipes mask */
 	__u32 enabled_rb_pipes_mask_hi;
+	/* shadow area size for gfx11 */
+	__u32 shadow_size;
+	/* shadow area base virtual alignment for gfx11 */
+	__u32 shadow_alignment;
+	/* context save area size for gfx11 */
+	__u32 csa_size;
+	/* context save area base virtual alignment for gfx11 */
+	__u32 csa_alignment;
 };
 
 struct drm_amdgpu_info_hw_ip {
-- 
GitLab


From 0db0c0379d15cd811214bdb631a0b6bdcdd22c84 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 9 Mar 2023 14:56:07 -0500
Subject: [PATCH 0265/1791] drm/amdgpu: add gfx shadow callback
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To provide IP specific shadow sizes.  UMDs will use
this to query the kernel driver for the size of the
shadow buffers.

v2: make callback return an int (Alex)
v3: drop GDS (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 6c5f58a34e842..f1a2ce7b6acaf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -233,6 +233,13 @@ struct amdgpu_gfx_ras {
 						struct amdgpu_iv_entry *entry);
 };
 
+struct amdgpu_gfx_shadow_info {
+	u32 shadow_size;
+	u32 shadow_alignment;
+	u32 csa_size;
+	u32 csa_alignment;
+};
+
 struct amdgpu_gfx_funcs {
 	/* get the gpu clock counter */
 	uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev);
@@ -250,6 +257,8 @@ struct amdgpu_gfx_funcs {
 				 u32 queue, u32 vmid);
 	void (*init_spm_golden)(struct amdgpu_device *adev);
 	void (*update_perfmon_mgcg)(struct amdgpu_device *adev, bool enable);
+	int (*get_gfx_shadow_info)(struct amdgpu_device *adev,
+				   struct amdgpu_gfx_shadow_info *shadow_info);
 };
 
 struct sq_work {
@@ -391,6 +400,7 @@ struct amdgpu_gfx {
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id)))
 #define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid))
 #define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev))
+#define amdgpu_gfx_get_gfx_shadow_info(adev, si) ((adev)->gfx.funcs->get_gfx_shadow_info((adev), (si)))
 
 /**
  * amdgpu_gfx_create_bitmask - create a bitmask
-- 
GitLab


From 02527099ddc74244b9d94c93ec54e123fcee5899 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 9 Mar 2023 15:10:14 -0500
Subject: [PATCH 0266/1791] drm/amdgpu: add get_gfx_shadow_info callback for
 gfx11

Used to get the size and alignment requirements for
the gfx shadow buffer for preemption.

v2: use FW version check to determine whether to
    return a valid size here
    return an error if not supported (Alex)
v3: drop GDS (Alex)
v4: make amdgpu_gfx_shadow_info mandatory (Alex)

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index b8b2886aac92f..543af07ff102a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -818,6 +818,27 @@ static void gfx_v11_0_select_me_pipe_q(struct amdgpu_device *adev,
 	soc21_grbm_select(adev, me, pipe, q, vm);
 }
 
+/* all sizes are in bytes */
+#define MQD_SHADOW_BASE_SIZE      73728
+#define MQD_SHADOW_BASE_ALIGNMENT 256
+#define MQD_FWWORKAREA_SIZE       484
+#define MQD_FWWORKAREA_ALIGNMENT  256
+
+static int gfx_v11_0_get_gfx_shadow_info(struct amdgpu_device *adev,
+					 struct amdgpu_gfx_shadow_info *shadow_info)
+{
+	if (adev->gfx.cp_gfx_shadow) {
+		shadow_info->shadow_size = MQD_SHADOW_BASE_SIZE;
+		shadow_info->shadow_alignment = MQD_SHADOW_BASE_ALIGNMENT;
+		shadow_info->csa_size = MQD_FWWORKAREA_SIZE;
+		shadow_info->csa_alignment = MQD_FWWORKAREA_ALIGNMENT;
+		return 0;
+	} else {
+		memset(shadow_info, 0, sizeof(struct amdgpu_gfx_shadow_info));
+		return -ENOTSUPP;
+	}
+}
+
 static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v11_0_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v11_0_select_se_sh,
@@ -826,6 +847,7 @@ static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = {
 	.read_wave_vgprs = &gfx_v11_0_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v11_0_select_me_pipe_q,
 	.update_perfmon_mgcg = &gfx_v11_0_update_perf_clk,
+	.get_gfx_shadow_info = &gfx_v11_0_get_gfx_shadow_info,
 };
 
 static int gfx_v11_0_gpu_early_init(struct amdgpu_device *adev)
-- 
GitLab


From 1ba91b54a9051205c2110ed43a7dc5650d49ca0e Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 9 Mar 2023 15:28:25 -0500
Subject: [PATCH 0267/1791] drm/amdgpu: add support for new GFX shadow size
 query

Use the new callback to fetch the data.  Return an error if
not supported.  UMDs should use this query to check whether
shadow buffers are supported and if so what size they
should be.

v2: return an error rather than a zerod structure.
v3: drop GDS, move into dev_info structure.  Data will be
    0 if not supported.
v4: drop local variable r

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index a5bae7eb993a0..1d3b224b8b289 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -876,6 +876,19 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		dev_info->gl2c_cache_size = adev->gfx.config.gc_gl2c_per_gpu;
 		dev_info->mall_size = adev->gmc.mall_size;
 
+
+		if (adev->gfx.funcs->get_gfx_shadow_info) {
+			struct amdgpu_gfx_shadow_info shadow_info;
+
+			ret = amdgpu_gfx_get_gfx_shadow_info(adev, &shadow_info);
+			if (!ret) {
+				dev_info->shadow_size = shadow_info.shadow_size;
+				dev_info->shadow_alignment = shadow_info.shadow_alignment;
+				dev_info->csa_size = shadow_info.csa_size;
+				dev_info->csa_alignment = shadow_info.csa_alignment;
+			}
+		}
+
 		ret = copy_to_user(out, dev_info,
 				   min((size_t)size, sizeof(*dev_info))) ? -EFAULT : 0;
 		kfree(dev_info);
-- 
GitLab


From 550e5d23f14784e2a625c25fe0c9d498589c9256 Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Tue, 4 Apr 2023 13:44:18 -0400
Subject: [PATCH 0268/1791] drm/amd/display: assign edid_blob_ptr with edid
 from debugfs

[Why] implementation change of drm_edid_override_set since linux
kernel 6.1, edid from debugfs is saved into connector->edid_override
immediatey, not saved to connector->edid_blob_ptr at the same time.

[How] call new drm_edid function drm_connector_update_edid_property
to assign connector->edid_blob_ptr with override edid from debugfs.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Chao-kai Wang <Stylon.Wang@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c   | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 8a3c04ff9bffd..deb740466b196 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6373,11 +6373,20 @@ static void create_eml_sink(struct amdgpu_dm_connector *aconnector)
 	struct edid *edid;
 
 	if (!aconnector->base.edid_blob_ptr) {
-		DRM_ERROR("No EDID firmware found on connector: %s ,forcing to OFF!\n",
-				aconnector->base.name);
+		/* if connector->edid_override valid, pass
+		 * it to edid_override to edid_blob_ptr
+		 */
+		int count;
 
-		aconnector->base.force = DRM_FORCE_OFF;
-		return;
+		count = drm_edid_override_connector_update(&aconnector->base);
+
+		if (!aconnector->base.edid_blob_ptr) {
+			DRM_ERROR("No EDID firmware found on connector: %s ,forcing to OFF!\n",
+					aconnector->base.name);
+
+			aconnector->base.force = DRM_FORCE_OFF;
+			return;
+		}
 	}
 
 	edid = (struct edid *) aconnector->base.edid_blob_ptr->data;
-- 
GitLab


From 9fa8cc0c444562fa19e20ca20f1c70e15b9d8c13 Mon Sep 17 00:00:00 2001
From: Michael Strauss <michael.strauss@amd.com>
Date: Mon, 3 Apr 2023 11:43:50 -0400
Subject: [PATCH 0269/1791] drm/amd/display: Convert Delaying Aux-I Disable To
 Monitor Patch

[WHY]
32ms delay was added to resolve issue with a specific sink, however this same
delay also introduces erroneous link training failures with certain sink
devices.

[HOW]
Only apply the 32ms delay for offending devices instead of globally.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h             |  1 -
 drivers/gpu/drm/amd/display/dc/dc_types.h       |  1 +
 .../link_dp_training_fixed_vs_pe_retimer.c      | 17 +++++++++++------
 3 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index b45974a2dec34..97747f5fde56a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -855,7 +855,6 @@ struct dc_debug_options {
 	bool force_usr_allow;
 	/* uses value at boot and disables switch */
 	bool disable_dtb_ref_clk_switch;
-	uint32_t fixed_vs_aux_delay_config_wa;
 	bool extended_blank_optimization;
 	union aux_wake_wa_options aux_wake_wa;
 	uint32_t mst_start_top_delay;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 34c848311455a..150c19286d67f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -196,6 +196,7 @@ struct dc_panel_patch {
 	unsigned int disable_fams;
 	unsigned int skip_avmute;
 	unsigned int mst_start_top_delay;
+	unsigned int delay_disable_aux_intercept_ms;
 };
 
 struct dc_edid_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
index 5731c4b61f9f0..fb6c938c6dab1 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
@@ -233,7 +233,8 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 			link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
 	const uint8_t vendor_lttpr_write_data_intercept_en[4] = {0x1, 0x55, 0x63, 0x0};
 	const uint8_t vendor_lttpr_write_data_intercept_dis[4] = {0x1, 0x55, 0x63, 0x68};
-	uint32_t pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa;
+	uint32_t pre_disable_intercept_delay_ms =
+			link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
 	uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0};
 	uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0};
 	uint32_t vendor_lttpr_write_address = 0xF004F;
@@ -259,7 +260,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 
 		/* Certain display and cable configuration require extra delay */
 		if (offset > 2)
-			pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2;
+			pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2;
 	}
 
 	/* Vendor specific: Reset lane settings */
@@ -380,7 +381,8 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 						0);
 				/* Vendor specific: Disable intercept */
 				for (i = 0; i < max_vendor_dpcd_retries; i++) {
-					msleep(pre_disable_intercept_delay_ms);
+					if (pre_disable_intercept_delay_ms != 0)
+						msleep(pre_disable_intercept_delay_ms);
 					dpcd_status = core_link_write_dpcd(
 							link,
 							vendor_lttpr_write_address,
@@ -591,9 +593,11 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 	const uint8_t vendor_lttpr_write_data_adicora_eq1[4] = {0x1, 0x55, 0x63, 0x2E};
 	const uint8_t vendor_lttpr_write_data_adicora_eq2[4] = {0x1, 0x55, 0x63, 0x01};
 	const uint8_t vendor_lttpr_write_data_adicora_eq3[4] = {0x1, 0x55, 0x63, 0x68};
-	uint32_t pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa;
 	uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0};
 	uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0};
+	uint32_t pre_disable_intercept_delay_ms =
+			link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
+
 
 	uint32_t vendor_lttpr_write_address = 0xF004F;
 	enum link_training_result status = LINK_TRAINING_SUCCESS;
@@ -618,7 +622,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 
 		/* Certain display and cable configuration require extra delay */
 		if (offset > 2)
-			pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2;
+			pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2;
 	}
 
 	/* Vendor specific: Reset lane settings */
@@ -739,7 +743,8 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 						0);
 				/* Vendor specific: Disable intercept */
 				for (i = 0; i < max_vendor_dpcd_retries; i++) {
-					msleep(pre_disable_intercept_delay_ms);
+					if (pre_disable_intercept_delay_ms != 0)
+						msleep(pre_disable_intercept_delay_ms);
 					dpcd_status = core_link_write_dpcd(
 							link,
 							vendor_lttpr_write_address,
-- 
GitLab


From 3af470cbcc9f40e47fe9b16882f60cd20b438095 Mon Sep 17 00:00:00 2001
From: Xiaogang Chen <Xiaogang.Chen@amd.com>
Date: Fri, 21 Apr 2023 13:35:01 -0500
Subject: [PATCH 0270/1791] drm/amdkfd: Fix an issue at userptr buffer
 validation process.

amdgpu_ttm_tt_get_user_pages can fail(-EFAULT). If it failed mem has no associated
hmm range or user_pages associated. Keep it at process_info->userptr_inval_list and
mark mem->invalid until following scheduled attempts can valid it.

Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 22 +++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 83a83ced2439f..4432e169fae8f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -2445,7 +2445,9 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info,
 			ret = -EAGAIN;
 			goto unlock_out;
 		}
-		mem->invalid = 0;
+		 /* set mem valid if mem has hmm range associated */
+		if (mem->range)
+			mem->invalid = 0;
 	}
 
 unlock_out:
@@ -2577,8 +2579,15 @@ static int confirm_valid_user_pages_locked(struct amdkfd_process_info *process_i
 	list_for_each_entry_safe(mem, tmp_mem,
 				 &process_info->userptr_inval_list,
 				 validate_list.head) {
-		bool valid = amdgpu_ttm_tt_get_user_pages_done(
-				mem->bo->tbo.ttm, mem->range);
+		bool valid;
+
+		/* keep mem without hmm range at userptr_inval_list */
+		if (!mem->range)
+			 continue;
+
+		/* Only check mem with hmm range associated */
+		valid = amdgpu_ttm_tt_get_user_pages_done(
+					mem->bo->tbo.ttm, mem->range);
 
 		mem->range = NULL;
 		if (!valid) {
@@ -2586,7 +2595,12 @@ static int confirm_valid_user_pages_locked(struct amdkfd_process_info *process_i
 			ret = -EAGAIN;
 			continue;
 		}
-		WARN(mem->invalid, "Valid BO is marked invalid");
+
+		if (mem->invalid) {
+			WARN(1, "Valid BO is marked invalid");
+			ret = -EAGAIN;
+			continue;
+		}
 
 		list_move_tail(&mem->validate_list.head,
 			       &process_info->userptr_valid_list);
-- 
GitLab


From 8a93c691248e7ff2b3944107a1ead2671b6854f2 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 10 Mar 2023 13:17:11 -0500
Subject: [PATCH 0271/1791] drm/amdgpu: bump driver version number for CP GFX
 shadow

So UMDs can determine whether the kernel supports this.

Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21986

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b400d598b75ad..b987022e11b4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -110,9 +110,10 @@
  *   3.52.0 - Add AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD, add device_info fields:
  *            tcp_cache_size, num_sqc_per_wgp, sqc_data_cache_size, sqc_inst_cache_size,
  *            gl1c_cache_size, gl2c_cache_size, mall_size, enabled_rb_pipes_mask_hi
+ *   3.53.0 - Support for GFX11 CP GFX shadowing
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	52
+#define KMS_DRIVER_MINOR	53
 #define KMS_DRIVER_PATCHLEVEL	0
 
 unsigned int amdgpu_vram_limit = UINT_MAX;
-- 
GitLab


From b185c31847856d9fa3008f727a824db163df0801 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 21 Mar 2023 13:28:33 -0400
Subject: [PATCH 0272/1791] drm/amdgpu: track MQD size for gfx and compute
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It varies by generation and we need to know the size
to expose this via debugfs.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c  | 2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 465ad0b7cddb0..60bb4bba19947 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -420,6 +420,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 					return r;
 				}
 
+				ring->mqd_size = mqd_size;
 				/* prepare MQD backup */
 				adev->gfx.me.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL);
 				if (!adev->gfx.me.mqd_backup[i])
@@ -440,6 +441,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 				return r;
 			}
 
+			ring->mqd_size = mqd_size;
 			/* prepare MQD backup */
 			adev->gfx.mec.mqd_backup[i + xcc_id * adev->gfx.num_compute_rings] = kmalloc(mqd_size, GFP_KERNEL);
 			if (!adev->gfx.mec.mqd_backup[i])
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 2aa6cc1c4212c..b0dc0a0c2631f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -258,6 +258,7 @@ struct amdgpu_ring {
 	struct amdgpu_bo	*mqd_obj;
 	uint64_t                mqd_gpu_addr;
 	void                    *mqd_ptr;
+	unsigned                mqd_size;
 	uint64_t                eop_gpu_addr;
 	u32			doorbell_index;
 	bool			use_doorbell;
-- 
GitLab


From 445d85e3c1dfd8c45b24be6f1527f1e117256d0e Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 21 Mar 2023 13:59:13 -0400
Subject: [PATCH 0273/1791] drm/amdgpu: add debugfs interface for reading MQDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Provide a debugfs interface to access the MQD.  Useful for
debugging issues with the CP and MES hardware scheduler.

v2: fix missing unreserve/unmap when pos >= size (Alex)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 59 ++++++++++++++++++++++++
 1 file changed, 59 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index eaf0f82757ef1..1e9ae3bc59d39 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -509,6 +509,59 @@ static const struct file_operations amdgpu_debugfs_ring_fops = {
 	.llseek = default_llseek
 };
 
+static ssize_t amdgpu_debugfs_mqd_read(struct file *f, char __user *buf,
+				       size_t size, loff_t *pos)
+{
+	struct amdgpu_ring *ring = file_inode(f)->i_private;
+	volatile u32 *mqd;
+	int r;
+	uint32_t value, result;
+
+	if (*pos & 3 || size & 3)
+		return -EINVAL;
+
+	result = 0;
+
+	r = amdgpu_bo_reserve(ring->mqd_obj, false);
+	if (unlikely(r != 0))
+		return r;
+
+	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&mqd);
+	if (r) {
+		amdgpu_bo_unreserve(ring->mqd_obj);
+		return r;
+	}
+
+	while (size) {
+		if (*pos >= ring->mqd_size)
+			goto done;
+
+		value = mqd[*pos/4];
+		r = put_user(value, (uint32_t *)buf);
+		if (r)
+			goto done;
+		buf += 4;
+		result += 4;
+		size -= 4;
+		*pos += 4;
+	}
+
+done:
+	amdgpu_bo_kunmap(ring->mqd_obj);
+	mqd = NULL;
+	amdgpu_bo_unreserve(ring->mqd_obj);
+	if (r)
+		return r;
+
+	return result;
+}
+
+static const struct file_operations amdgpu_debugfs_mqd_fops = {
+	.owner = THIS_MODULE,
+	.read = amdgpu_debugfs_mqd_read,
+	.llseek = default_llseek
+};
+
 #endif
 
 void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
@@ -524,6 +577,12 @@ void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
 				 &amdgpu_debugfs_ring_fops,
 				 ring->ring_size + 12);
 
+	if (ring->mqd_obj) {
+		sprintf(name, "amdgpu_mqd_%s", ring->name);
+		debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, ring,
+					 &amdgpu_debugfs_mqd_fops,
+					 ring->mqd_size);
+	}
 #endif
 }
 
-- 
GitLab


From 93aac179a44be000aa8a025963011c752c23d92e Mon Sep 17 00:00:00 2001
From: Anthony Koo <Anthony.Koo@amd.com>
Date: Sat, 8 Apr 2023 02:51:46 -0400
Subject: [PATCH 0274/1791] drm/amd/display: [FW Promotion] Release 0.0.163.0

Add feature caps for Subvp + PSR so driver is back compatible

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Anthony Koo <Anthony.Koo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index b32a5c977d171..dcec8955bde42 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -257,7 +257,8 @@ struct dmub_feature_caps {
 	 */
 	uint8_t psr;
 	uint8_t fw_assisted_mclk_switch;
-	uint8_t reserved[6];
+	uint8_t subvp_psr_support;
+	uint8_t reserved[5];
 };
 
 struct dmub_visual_confirm_color {
-- 
GitLab


From 267e2d8e8e60ed59e74cf888f6b74bd5202e7384 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 9 Apr 2023 23:08:25 -0400
Subject: [PATCH 0275/1791] drm/amd/display: 3.2.232

Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 97747f5fde56a..d6f0ecb708cd9 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.231"
+#define DC_VER "3.2.232"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From 0ba4a784a14592abed41873e339eab78ceb6e230 Mon Sep 17 00:00:00 2001
From: Alex Hung <alex.hung@amd.com>
Date: Wed, 5 Apr 2023 13:47:41 -0600
Subject: [PATCH 0276/1791] drm/amd/display: implement force function in
 amdgpu_dm_connector_funcs

[Why]
When userspace (IGT) inserts EDID with audio data (ELD), EDID
is not updated.

[How]
Implements force function (amdgpu_dm_connector_funcs_force) in
amdgpu_dm_connector_funcs to create emulated sink and to handle
EDID.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Hersen Wu <hersenxs.wu@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Wenchieh Chien <wenchieh.chien@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 28 ++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index deb740466b196..b265434c77e14 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6346,6 +6346,31 @@ amdgpu_dm_connector_late_register(struct drm_connector *connector)
 	return 0;
 }
 
+void amdgpu_dm_connector_funcs_force(struct drm_connector *connector)
+{
+	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
+	struct dc_link *dc_link = aconnector->dc_link;
+	struct dc_sink *dc_em_sink = aconnector->dc_em_sink;
+	struct edid *edid;
+
+	if (!connector->edid_override)
+		return;
+
+	drm_edid_override_connector_update(&aconnector->base);
+	edid = aconnector->base.edid_blob_ptr->data;
+	aconnector->edid = edid;
+
+	/* Update emulated (virtual) sink's EDID */
+	if (dc_em_sink && dc_link) {
+		memset(&dc_em_sink->edid_caps, 0, sizeof(struct dc_edid_caps));
+		memmove(dc_em_sink->dc_edid.raw_edid, edid, (edid->extensions + 1) * EDID_LENGTH);
+		dm_helpers_parse_edid_caps(
+			dc_link,
+			&dc_em_sink->dc_edid,
+			&dc_em_sink->edid_caps);
+	}
+}
+
 static const struct drm_connector_funcs amdgpu_dm_connector_funcs = {
 	.reset = amdgpu_dm_connector_funcs_reset,
 	.detect = amdgpu_dm_connector_detect,
@@ -6356,7 +6381,8 @@ static const struct drm_connector_funcs amdgpu_dm_connector_funcs = {
 	.atomic_set_property = amdgpu_dm_connector_atomic_set_property,
 	.atomic_get_property = amdgpu_dm_connector_atomic_get_property,
 	.late_register = amdgpu_dm_connector_late_register,
-	.early_unregister = amdgpu_dm_connector_unregister
+	.early_unregister = amdgpu_dm_connector_unregister,
+	.force = amdgpu_dm_connector_funcs_force
 };
 
 static int get_modes(struct drm_connector *connector)
-- 
GitLab


From 278d3de6754e778cb676b7e1b10782eff1971010 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Mon, 10 Apr 2023 13:52:21 -0400
Subject: [PATCH 0277/1791] drm/amd/display: Update FW feature caps struct

Reorder FW feature caps struct variable order to ensure backwards
compatability is maintained for older FW

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Sung Lee <Sung.Lee@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index dcec8955bde42..09d4d0d9b92c3 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -257,8 +257,9 @@ struct dmub_feature_caps {
 	 */
 	uint8_t psr;
 	uint8_t fw_assisted_mclk_switch;
+	uint8_t reserved[4];
 	uint8_t subvp_psr_support;
-	uint8_t reserved[5];
+	uint8_t gecc_enable;
 };
 
 struct dmub_visual_confirm_color {
-- 
GitLab


From 8f3589bb6fcea397775398cba4fbcc46829a60ed Mon Sep 17 00:00:00 2001
From: "JinZe.Xu" <JinZe.Xu@amd.com>
Date: Mon, 10 Apr 2023 23:23:37 +0800
Subject: [PATCH 0278/1791] drm/amd/display: Restore rptr/wptr for DMCUB as
 workaround

[Why]
States may be desync after resume.

[How]
Sync sw state with hw state.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: JinZe.Xu <JinZe.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h     | 14 ++++++++++++++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn20.c   |  5 +++++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn20.h   |  2 ++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn31.c   |  5 +++++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn31.h   |  2 ++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn32.c   |  5 +++++
 .../gpu/drm/amd/display/dmub/src/dmub_dcn32.h   |  2 ++
 drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c | 17 +++++++++++++++++
 8 files changed, 52 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 719bf9bb168ad..d35432c218564 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -340,6 +340,8 @@ struct dmub_srv_hw_funcs {
 	void (*setup_mailbox)(struct dmub_srv *dmub,
 			      const struct dmub_region *inbox1);
 
+	uint32_t (*get_inbox1_wptr)(struct dmub_srv *dmub);
+
 	uint32_t (*get_inbox1_rptr)(struct dmub_srv *dmub);
 
 	void (*set_inbox1_wptr)(struct dmub_srv *dmub, uint32_t wptr_offset);
@@ -600,6 +602,18 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
  */
 enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub);
 
+/**
+ * dmub_srv_sync_inbox1() - sync sw state with hw state
+ * @dmub: the dmub service
+ *
+ * Sync sw state with hw state when resume from S0i3
+ *
+ * Return:
+ *   DMUB_STATUS_OK - success
+ *   DMUB_STATUS_INVALID - unspecified error
+ */
+enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub);
+
 /**
  * dmub_srv_cmd_queue() - queues a command to the DMUB
  * @dmub: the dmub service
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c
index a6540e27044d2..98dad0d47e72c 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c
@@ -282,6 +282,11 @@ void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub,
 	REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base);
 }
 
+uint32_t dmub_dcn20_get_inbox1_wptr(struct dmub_srv *dmub)
+{
+	return REG_READ(DMCUB_INBOX1_WPTR);
+}
+
 uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub)
 {
 	return REG_READ(DMCUB_INBOX1_RPTR);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h
index c2e5831ac52cc..1df128e57ed3b 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h
@@ -202,6 +202,8 @@ void dmub_dcn20_setup_windows(struct dmub_srv *dmub,
 void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub,
 			      const struct dmub_region *inbox1);
 
+uint32_t dmub_dcn20_get_inbox1_wptr(struct dmub_srv *dmub);
+
 uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub);
 
 void dmub_dcn20_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
index c90b9ee42e126..ebf7aeec4029c 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
@@ -242,6 +242,11 @@ void dmub_dcn31_setup_mailbox(struct dmub_srv *dmub,
 	REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base);
 }
 
+uint32_t dmub_dcn31_get_inbox1_wptr(struct dmub_srv *dmub)
+{
+	return REG_READ(DMCUB_INBOX1_WPTR);
+}
+
 uint32_t dmub_dcn31_get_inbox1_rptr(struct dmub_srv *dmub)
 {
 	return REG_READ(DMCUB_INBOX1_RPTR);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h
index f6db6f89d45dc..7d5c10ee539b4 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h
@@ -204,6 +204,8 @@ void dmub_dcn31_setup_windows(struct dmub_srv *dmub,
 void dmub_dcn31_setup_mailbox(struct dmub_srv *dmub,
 			      const struct dmub_region *inbox1);
 
+uint32_t dmub_dcn31_get_inbox1_wptr(struct dmub_srv *dmub);
+
 uint32_t dmub_dcn31_get_inbox1_rptr(struct dmub_srv *dmub);
 
 void dmub_dcn31_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index a7d5607459edb..21dd6cbdb1063 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -266,6 +266,11 @@ void dmub_dcn32_setup_mailbox(struct dmub_srv *dmub,
 	REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base);
 }
 
+uint32_t dmub_dcn32_get_inbox1_wptr(struct dmub_srv *dmub)
+{
+	return REG_READ(DMCUB_INBOX1_WPTR);
+}
+
 uint32_t dmub_dcn32_get_inbox1_rptr(struct dmub_srv *dmub)
 {
 	return REG_READ(DMCUB_INBOX1_RPTR);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
index 7d1a6eb4d6657..f15336b6e22be 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
@@ -206,6 +206,8 @@ void dmub_dcn32_setup_windows(struct dmub_srv *dmub,
 void dmub_dcn32_setup_mailbox(struct dmub_srv *dmub,
 			      const struct dmub_region *inbox1);
 
+uint32_t dmub_dcn32_get_inbox1_wptr(struct dmub_srv *dmub);
+
 uint32_t dmub_dcn32_get_inbox1_rptr(struct dmub_srv *dmub);
 
 void dmub_dcn32_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset);
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index 67c53f7e589c6..ea3bed70a2297 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -166,6 +166,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 		funcs->backdoor_load = dmub_dcn20_backdoor_load;
 		funcs->setup_windows = dmub_dcn20_setup_windows;
 		funcs->setup_mailbox = dmub_dcn20_setup_mailbox;
+		funcs->get_inbox1_wptr = dmub_dcn20_get_inbox1_wptr;
 		funcs->get_inbox1_rptr = dmub_dcn20_get_inbox1_rptr;
 		funcs->set_inbox1_wptr = dmub_dcn20_set_inbox1_wptr;
 		funcs->is_supported = dmub_dcn20_is_supported;
@@ -235,6 +236,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 		funcs->backdoor_load = dmub_dcn31_backdoor_load;
 		funcs->setup_windows = dmub_dcn31_setup_windows;
 		funcs->setup_mailbox = dmub_dcn31_setup_mailbox;
+		funcs->get_inbox1_wptr = dmub_dcn31_get_inbox1_wptr;
 		funcs->get_inbox1_rptr = dmub_dcn31_get_inbox1_rptr;
 		funcs->set_inbox1_wptr = dmub_dcn31_set_inbox1_wptr;
 		funcs->setup_out_mailbox = dmub_dcn31_setup_out_mailbox;
@@ -273,6 +275,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 		funcs->backdoor_load_zfb_mode = dmub_dcn32_backdoor_load_zfb_mode;
 		funcs->setup_windows = dmub_dcn32_setup_windows;
 		funcs->setup_mailbox = dmub_dcn32_setup_mailbox;
+		funcs->get_inbox1_wptr = dmub_dcn32_get_inbox1_wptr;
 		funcs->get_inbox1_rptr = dmub_dcn32_get_inbox1_rptr;
 		funcs->set_inbox1_wptr = dmub_dcn32_set_inbox1_wptr;
 		funcs->setup_out_mailbox = dmub_dcn32_setup_out_mailbox;
@@ -642,6 +645,20 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 	return DMUB_STATUS_OK;
 }
 
+enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub)
+{
+	if (!dmub->sw_init)
+		return DMUB_STATUS_INVALID;
+
+	if (dmub->hw_funcs.get_inbox1_rptr && dmub->hw_funcs.get_inbox1_wptr) {
+		dmub->inbox1_rb.rptr = dmub->hw_funcs.get_inbox1_rptr(dmub);
+		dmub->inbox1_rb.wrpt = dmub->hw_funcs.get_inbox1_wptr(dmub);
+		dmub->inbox1_last_wptr = dmub->inbox1_rb.wrpt;
+	}
+
+	return DMUB_STATUS_OK;
+}
+
 enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub)
 {
 	if (!dmub->sw_init)
-- 
GitLab


From ee7be8f3de1ccc9665281fe996f9b6d45191ec1a Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Mon, 10 Apr 2023 14:37:27 -0400
Subject: [PATCH 0279/1791] drm/amd/display: Limit DCN32 8 channel or less
 parts to DPM1 for FPO

- Due to hardware related QoS issues, we need to limit certain
  SKUs with less memory channels to DPM1 and above.
- At DPM0 + workload running, the urgent return latency can
  exceed 15us (the expected maximum is 4us) which results in underflow

Cc: stable@vger.kernel.org
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Saaem Rizvi <SyedSaaem.Rizvi@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn32/dcn32_resource.c |  2 ++
 .../gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 15 +++++++++++++++
 .../gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h  |  2 ++
 3 files changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 4f8286ae699b0..0085ea78ea317 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -1888,6 +1888,8 @@ bool dcn32_validate_bandwidth(struct dc *dc,
 
 	dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel);
 
+	dcn32_override_min_req_memclk(dc, context);
+
 	BW_VAL_TRACE_END_WATERMARKS();
 
 	goto validate_out;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index b8a2518faecc7..ed7ea4c42412b 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2887,3 +2887,18 @@ void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb)
 	dc_assert_fp_enabled();
 	dcn3_2_soc.clock_limits[0].dcfclk_mhz = 1200.0;
 }
+
+void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context)
+{
+	// WA: restrict FPO and SubVP to use first non-strobe mode (DCN32 BW issue)
+	if ((context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dcn32_subvp_in_use(dc, context)) &&
+			dc->dml.soc.num_chans <= 8) {
+		int num_mclk_levels = dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels;
+
+		if (context->bw_ctx.dml.vba.DRAMSpeed <= dc->clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 16 &&
+				num_mclk_levels > 1) {
+			context->bw_ctx.dml.vba.DRAMSpeed = dc->clk_mgr->bw_params->clk_table.entries[1].memclk_mhz * 16;
+			context->bw_ctx.bw.dcn.clk.dramclk_khz = context->bw_ctx.dml.vba.DRAMSpeed * 1000 / 16;
+		}
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
index dcf512cd30721..a4206b71d650a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
@@ -80,6 +80,8 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co
 
 bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req);
 
+void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context);
+
 void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb);
 
 #endif
-- 
GitLab


From 9ba90d760e9354c124fa9bbea08017d96699a82c Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Date: Fri, 9 Dec 2022 15:13:18 -0500
Subject: [PATCH 0280/1791] drm/amd/display: add pixel rate based CRB
 allocation support

This feature is meant to unblock PSTATE for certain high end display
configs on dcn315. This is achieved by allocating CRB to detile buffer
based on display requirements to meet pstate latency hiding needs.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn31/dcn31_hubbub.c   |  1 +
 .../amd/display/dc/dcn315/dcn315_resource.c   | 97 ++++++++++++++++++-
 .../drm/amd/display/dc/dml/dcn31/dcn31_fpu.c  | 25 ++++-
 .../drm/amd/display/dc/dml/dcn31/dcn31_fpu.h  |  3 +
 .../dc/dml/dcn31/display_mode_vba_31.c        | 39 +++++---
 .../drm/amd/display/dc/dml/display_mode_vba.c |  6 ++
 6 files changed, 154 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c
index 7e7cd5b64e6a1..7445ed27852a1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c
@@ -103,6 +103,7 @@ static void dcn31_program_det_size(struct hubbub *hubbub, int hubp_inst, unsigne
 	default:
 		break;
 	}
+	DC_LOG_DEBUG("Set DET%d to %d segments\n", hubp_inst, det_size_segments);
 	/* Should never be hit, if it is we have an erroneous hw config*/
 	ASSERT(hubbub2->det0_size + hubbub2->det1_size + hubbub2->det2_size
 			+ hubbub2->det3_size + hubbub2->compbuf_size_segments <= hubbub2->crb_size_segs);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index 41c972c8eb198..42a0157fd8133 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -136,6 +136,9 @@
 
 #define DCN3_15_MAX_DET_SIZE 384
 #define DCN3_15_CRB_SEGMENT_SIZE_KB 64
+#define DCN3_15_MAX_DET_SEGS (DCN3_15_MAX_DET_SIZE / DCN3_15_CRB_SEGMENT_SIZE_KB)
+/* Minimum 2 extra segments need to be in compbuf and claimable to guarantee seamless mpo transitions */
+#define MIN_RESERVED_DET_SEGS 2
 
 enum dcn31_clk_src_array_id {
 	DCN31_CLK_SRC_PLL0,
@@ -1636,21 +1639,57 @@ static bool is_dual_plane(enum surface_pixel_format format)
 	return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA;
 }
 
+static int source_format_to_bpp (enum source_format_class SourcePixelFormat)
+{
+	if (SourcePixelFormat == dm_444_64)
+		return 8;
+	else if (SourcePixelFormat == dm_444_16 || SourcePixelFormat == dm_444_16)
+		return 2;
+	else if (SourcePixelFormat == dm_444_8)
+		return 1;
+	else if (SourcePixelFormat == dm_rgbe_alpha)
+		return 5;
+	else if (SourcePixelFormat == dm_420_8)
+		return 3;
+	else if (SourcePixelFormat == dm_420_12)
+		return 6;
+	else
+		return 4;
+}
+
+static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context)
+{
+	int i;
+	struct resource_context *res_ctx = &context->res_ctx;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!res_ctx->pipe_ctx[i].stream)
+			continue;
+
+		/*Don't apply if MPO to avoid transition issues*/
+		if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state != res_ctx->pipe_ctx[i].plane_state)
+			return false;
+	}
+	return true;
+}
+
 static int dcn315_populate_dml_pipes_from_context(
 	struct dc *dc, struct dc_state *context,
 	display_e2e_pipe_params_st *pipes,
 	bool fast_validate)
 {
-	int i, pipe_cnt;
+	int i, pipe_cnt, crb_idx, crb_pipes;
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *pipe;
 	const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB;
+	int remaining_det_segs = max_usable_det / DCN3_15_CRB_SEGMENT_SIZE_KB;
+	bool pixel_rate_crb = allow_pixel_rate_crb(dc, context);
 
 	DC_FP_START();
 	dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate);
 	DC_FP_END();
 
-	for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) {
+	for (i = 0, pipe_cnt = 0, crb_pipes = 0; i < dc->res_pool->pipe_count; i++) {
 		struct dc_crtc_timing *timing;
 
 		if (!res_ctx->pipe_ctx[i].stream)
@@ -1671,6 +1710,23 @@ static int dcn315_populate_dml_pipes_from_context(
 		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
 		DC_FP_START();
 		dcn31_zero_pipe_dcc_fraction(pipes, pipe_cnt);
+		if (pixel_rate_crb && !pipe->top_pipe && !pipe->prev_odm_pipe) {
+			int bpp = source_format_to_bpp(pipes[pipe_cnt].pipe.src.source_format);
+			/* Ceil to crb segment size */
+			int approx_det_segs_required_for_pstate = dcn_get_approx_det_segs_required_for_pstate(
+					&context->bw_ctx.dml.soc, timing->pix_clk_100hz, bpp, DCN3_15_CRB_SEGMENT_SIZE_KB);
+			if (approx_det_segs_required_for_pstate <= 2 * DCN3_15_MAX_DET_SEGS) {
+				bool split_required = approx_det_segs_required_for_pstate > DCN3_15_MAX_DET_SEGS;
+				split_required = split_required || timing->pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc);
+				split_required = split_required || (pipe->plane_state && pipe->plane_state->src_rect.width > 5120);
+				if (split_required)
+					approx_det_segs_required_for_pstate += approx_det_segs_required_for_pstate % 2;
+				pipes[pipe_cnt].pipe.src.det_size_override = approx_det_segs_required_for_pstate;
+				remaining_det_segs -= approx_det_segs_required_for_pstate;
+			} else
+				remaining_det_segs = -1;
+			crb_pipes++;
+		}
 		DC_FP_END();
 
 		if (pipes[pipe_cnt].dout.dsc_enable) {
@@ -1689,16 +1745,49 @@ static int dcn315_populate_dml_pipes_from_context(
 				break;
 			}
 		}
-
 		pipe_cnt++;
 	}
 
+	/* Spread remaining unreserved crb evenly among all pipes, use default policy if not enough det or single pipe */
+	if (pixel_rate_crb) {
+		for (i = 0, pipe_cnt = 0, crb_idx = 0; i < dc->res_pool->pipe_count; i++) {
+			pipe = &res_ctx->pipe_ctx[i];
+			if (!pipe->stream)
+				continue;
+
+			if (!pipe->top_pipe && !pipe->prev_odm_pipe) {
+				bool split_required = pipe->stream->timing.pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc)
+						|| (pipe->plane_state && pipe->plane_state->src_rect.width > 5120);
+
+				if (remaining_det_segs < 0 || crb_pipes == 1)
+					pipes[pipe_cnt].pipe.src.det_size_override = 0;
+				if (remaining_det_segs > MIN_RESERVED_DET_SEGS)
+					pipes[pipe_cnt].pipe.src.det_size_override += (remaining_det_segs - MIN_RESERVED_DET_SEGS) / crb_pipes +
+							(crb_idx < (remaining_det_segs - MIN_RESERVED_DET_SEGS) % crb_pipes ? 1 : 0);
+				if (pipes[pipe_cnt].pipe.src.det_size_override > 2 * DCN3_15_MAX_DET_SEGS) {
+					/* Clamp to 2 pipe split max det segments */
+					remaining_det_segs += pipes[pipe_cnt].pipe.src.det_size_override - 2 * (DCN3_15_MAX_DET_SEGS);
+					pipes[pipe_cnt].pipe.src.det_size_override = 2 * DCN3_15_MAX_DET_SEGS;
+				}
+				if (pipes[pipe_cnt].pipe.src.det_size_override > DCN3_15_MAX_DET_SEGS || split_required) {
+					/* If we are splitting we must have an even number of segments */
+					remaining_det_segs += pipes[pipe_cnt].pipe.src.det_size_override % 2;
+					pipes[pipe_cnt].pipe.src.det_size_override -= pipes[pipe_cnt].pipe.src.det_size_override % 2;
+				}
+				/* Convert segments into size for DML use */
+				pipes[pipe_cnt].pipe.src.det_size_override *= DCN3_15_CRB_SEGMENT_SIZE_KB;
+				crb_idx++;
+			}
+			pipe_cnt++;
+		}
+	}
+
 	if (pipe_cnt)
 		context->bw_ctx.dml.ip.det_buffer_size_kbytes =
 				(max_usable_det / DCN3_15_CRB_SEGMENT_SIZE_KB / pipe_cnt) * DCN3_15_CRB_SEGMENT_SIZE_KB;
 	if (context->bw_ctx.dml.ip.det_buffer_size_kbytes > DCN3_15_MAX_DET_SIZE)
 		context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_15_MAX_DET_SIZE;
-	ASSERT(context->bw_ctx.dml.ip.det_buffer_size_kbytes >= DCN3_15_DEFAULT_DET_SIZE);
+
 	dc->config.enable_4to1MPC = false;
 	if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) {
 		if (is_dual_plane(pipe->plane_state->format)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
index 59836570603ac..19d034341e640 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
@@ -483,7 +483,7 @@ void dcn31_calculate_wm_and_dlg_fp(
 		int pipe_cnt,
 		int vlevel)
 {
-	int i, pipe_idx, active_hubp_count = 0;
+	int i, pipe_idx, total_det = 0, active_hubp_count = 0;
 	double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
 
 	dc_assert_fp_enabled();
@@ -563,6 +563,18 @@ void dcn31_calculate_wm_and_dlg_fp(
 			if (context->res_ctx.pipe_ctx[i].stream)
 				context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = 0;
 	}
+	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+
+		context->res_ctx.pipe_ctx[i].det_buffer_size_kb =
+				get_det_buffer_size_kbytes(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
+		if (context->res_ctx.pipe_ctx[i].det_buffer_size_kb > 384)
+			context->res_ctx.pipe_ctx[i].det_buffer_size_kb /= 2;
+		total_det += context->res_ctx.pipe_ctx[i].det_buffer_size_kb;
+		pipe_idx++;
+	}
+	context->bw_ctx.bw.dcn.compbuf_size_kb = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - total_det;
 }
 
 void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
@@ -815,3 +827,14 @@ int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc)
 {
 	return soc->clock_limits[0].dispclk_mhz * 10000.0 / (1.0 + soc->dcn_downspread_percent / 100.0);
 }
+
+int dcn_get_approx_det_segs_required_for_pstate(
+		struct _vcs_dpi_soc_bounding_box_st *soc,
+		int pix_clk_100hz, int bpp, int seg_size_kb)
+{
+	/* Roughly calculate required crb to hide latency. In practice there is slightly
+	 * more buffer available for latency hiding
+	 */
+	return (int)(soc->dram_clock_change_latency_us * pix_clk_100hz * bpp
+					/ 10240000 + seg_size_kb - 1) /	seg_size_kb;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
index 687d3522cc33e..8f9c8faed2605 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h
@@ -47,6 +47,9 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params
 void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
 void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params);
 int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc);
+int dcn_get_approx_det_segs_required_for_pstate(
+		struct _vcs_dpi_soc_bounding_box_st *soc,
+		int pix_clk_100hz, int bpp, int seg_size_kb);
 
 int dcn31x_populate_dml_pipes_from_context(struct dc *dc,
 					  struct dc_state *context,
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index 330b089d6a86f..01603abd75bba 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -532,7 +532,8 @@ static void CalculateStutterEfficiency(
 static void CalculateSwathAndDETConfiguration(
 		bool ForceSingleDPP,
 		int NumberOfActivePlanes,
-		unsigned int DETBufferSizeInKByte,
+		bool DETSharedByAllDPP,
+		unsigned int DETBufferSizeInKByte[],
 		double MaximumSwathWidthLuma[],
 		double MaximumSwathWidthChroma[],
 		enum scan_direction_class SourceScan[],
@@ -3118,7 +3119,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 				v->SurfaceWidthC[k],
 				v->SurfaceHeightY[k],
 				v->SurfaceHeightC[k],
-				v->DETBufferSizeInKByte[0] * 1024,
+				v->DETBufferSizeInKByte[k] * 1024,
 				v->BlockHeight256BytesY[k],
 				v->BlockHeight256BytesC[k],
 				v->SurfaceTiling[k],
@@ -3313,7 +3314,8 @@ static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib)
 	CalculateSwathAndDETConfiguration(
 			false,
 			v->NumberOfActivePlanes,
-			v->DETBufferSizeInKByte[0],
+			mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0],
+			v->DETBufferSizeInKByte,
 			dummy1,
 			dummy2,
 			v->SourceScan,
@@ -3779,14 +3781,16 @@ static noinline void CalculatePrefetchSchedulePerPlane(
 		&v->VReadyOffsetPix[k]);
 }
 
-static void PatchDETBufferSizeInKByte(unsigned int NumberOfActivePlanes, int NoOfDPPThisState[], unsigned int config_return_buffer_size_in_kbytes, unsigned int *DETBufferSizeInKByte)
+static void PatchDETBufferSizeInKByte(unsigned int NumberOfActivePlanes, int NoOfDPPThisState[], unsigned int config_return_buffer_size_in_kbytes, unsigned int DETBufferSizeInKByte[])
 {
 	int i, total_pipes = 0;
 	for (i = 0; i < NumberOfActivePlanes; i++)
 		total_pipes += NoOfDPPThisState[i];
-	*DETBufferSizeInKByte = ((config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB) / 64 / total_pipes) * 64;
-	if (*DETBufferSizeInKByte > DCN3_15_MAX_DET_SIZE)
-		*DETBufferSizeInKByte = DCN3_15_MAX_DET_SIZE;
+	DETBufferSizeInKByte[0] = ((config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB) / 64 / total_pipes) * 64;
+	if (DETBufferSizeInKByte[0] > DCN3_15_MAX_DET_SIZE)
+		DETBufferSizeInKByte[0] = DCN3_15_MAX_DET_SIZE;
+	for (i = 1; i < NumberOfActivePlanes; i++)
+		DETBufferSizeInKByte[i] = DETBufferSizeInKByte[0];
 }
 
 
@@ -4026,7 +4030,8 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 	CalculateSwathAndDETConfiguration(
 			true,
 			v->NumberOfActivePlanes,
-			v->DETBufferSizeInKByte[0],
+			mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0],
+			v->DETBufferSizeInKByte,
 			v->MaximumSwathWidthLuma,
 			v->MaximumSwathWidthChroma,
 			v->SourceScan,
@@ -4166,6 +4171,10 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 						|| (v->PlaneRequiredDISPCLK > v->MaxDispclkRoundedDownToDFSGranularity)) {
 					v->DISPCLK_DPPCLK_Support[i][j] = false;
 				}
+				if (mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[k] > DCN3_15_MAX_DET_SIZE && v->NoOfDPP[i][j][k] < 2) {
+					v->MPCCombine[i][j][k] = true;
+					v->NoOfDPP[i][j][k] = 2;
+				}
 			}
 			v->TotalNumberOfActiveDPP[i][j] = 0;
 			v->TotalNumberOfSingleDPPPlanes[i][j] = 0;
@@ -4642,12 +4651,13 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 				v->ODMCombineEnableThisState[k] = v->ODMCombineEnablePerState[i][k];
 			}
 
-			if (v->NumberOfActivePlanes > 1 && mode_lib->project == DML_PROJECT_DCN315)
-				PatchDETBufferSizeInKByte(v->NumberOfActivePlanes, v->NoOfDPPThisState, v->ip.config_return_buffer_size_in_kbytes, &v->DETBufferSizeInKByte[0]);
+			if (v->NumberOfActivePlanes > 1 && mode_lib->project == DML_PROJECT_DCN315 && !v->DETSizeOverride[0])
+				PatchDETBufferSizeInKByte(v->NumberOfActivePlanes, v->NoOfDPPThisState, v->ip.config_return_buffer_size_in_kbytes, v->DETBufferSizeInKByte);
 			CalculateSwathAndDETConfiguration(
 					false,
 					v->NumberOfActivePlanes,
-					v->DETBufferSizeInKByte[0],
+					mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0],
+					v->DETBufferSizeInKByte,
 					v->MaximumSwathWidthLuma,
 					v->MaximumSwathWidthChroma,
 					v->SourceScan,
@@ -6611,7 +6621,8 @@ static void CalculateStutterEfficiency(
 static void CalculateSwathAndDETConfiguration(
 		bool ForceSingleDPP,
 		int NumberOfActivePlanes,
-		unsigned int DETBufferSizeInKByte,
+		bool DETSharedByAllDPP,
+		unsigned int DETBufferSizeInKByteA[],
 		double MaximumSwathWidthLuma[],
 		double MaximumSwathWidthChroma[],
 		enum scan_direction_class SourceScan[],
@@ -6695,6 +6706,10 @@ static void CalculateSwathAndDETConfiguration(
 
 	*ViewportSizeSupport = true;
 	for (k = 0; k < NumberOfActivePlanes; ++k) {
+		unsigned int DETBufferSizeInKByte = DETBufferSizeInKByteA[k];
+
+		if (DETSharedByAllDPP && DPPPerPlane[k])
+			DETBufferSizeInKByte /= DPPPerPlane[k];
 		if ((SourcePixelFormat[k] == dm_444_64 || SourcePixelFormat[k] == dm_444_32 || SourcePixelFormat[k] == dm_444_16 || SourcePixelFormat[k] == dm_mono_16
 				|| SourcePixelFormat[k] == dm_mono_8 || SourcePixelFormat[k] == dm_rgbe)) {
 			if (SurfaceTiling[k] == dm_sw_linear
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
index f9653f511baa3..2f63ae954826c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
@@ -571,6 +571,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib)
 		mode_lib->vba.OutputLinkDPRate[mode_lib->vba.NumberOfActivePlanes] = dout->dp_rate;
 		mode_lib->vba.ODMUse[mode_lib->vba.NumberOfActivePlanes] = dst->odm_combine_policy;
 		mode_lib->vba.DETSizeOverride[mode_lib->vba.NumberOfActivePlanes] = src->det_size_override;
+		if (src->det_size_override)
+			mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = src->det_size_override;
+		else
+			mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = ip->det_buffer_size_kbytes;
 		//TODO: Need to assign correct values to dp_multistream vars
 		mode_lib->vba.OutputMultistreamEn[mode_lib->vba.NumberOfActiveSurfaces] = dout->dp_multistream_en;
 		mode_lib->vba.OutputMultistreamId[mode_lib->vba.NumberOfActiveSurfaces] = dout->dp_multistream_id;
@@ -785,6 +789,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib)
 					mode_lib->vba.pipe_plane[k] =
 							mode_lib->vba.NumberOfActivePlanes;
 					mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes]++;
+					if (src_k->det_size_override)
+						mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = src_k->det_size_override;
 					if (mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes]
 							== dm_horz) {
 						mode_lib->vba.ViewportWidth[mode_lib->vba.NumberOfActivePlanes] +=
-- 
GitLab


From b058e3999021e04cd16d1e487732e20ad1f4b4f6 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Mon, 10 Apr 2023 17:17:07 -0400
Subject: [PATCH 0281/1791] drm/amd/display: Enable SubVP on PSR panels if
 single stream

Enable SubVP on PSR panels now that we have FW support

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                  | 1 +
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c   | 1 +
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 3 ++-
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index d6f0ecb708cd9..dc0a61ca99d01 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -209,6 +209,7 @@ struct dc_color_caps {
 struct dc_dmub_caps {
 	bool psr;
 	bool mclk_sw;
+	bool subvp_psr;
 };
 
 struct dc_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 9ce11ed769a07..1e9ada8b131ce 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -944,6 +944,7 @@ void dcn32_init_hw(struct dc *dc)
 	if (dc->ctx->dmub_srv) {
 		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 		dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
+		dc->caps.dmub_caps.subvp_psr = dc->ctx->dmub_srv->dmub->feature_caps.subvp_psr_support;
 		dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index ed7ea4c42412b..e1e82ad4c583c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -692,7 +692,8 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 		 *   to combine this with SubVP can cause issues with the scheduling).
 		 * - Not TMZ surface
 		 */
-		if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) && !dcn32_is_psr_capable(pipe) &&
+		if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) &&
+				(!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) &&
 				pipe->stream->mall_stream_config.type == SUBVP_NONE && refresh_rate < 120 && !pipe->plane_state->address.tmz_surface &&
 				(vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0 ||
 				(vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 &&
-- 
GitLab


From b5389eca7b39026452dfc1d2cd5b05448c631298 Mon Sep 17 00:00:00 2001
From: Leo Chen <sancchen@amd.com>
Date: Tue, 11 Apr 2023 10:49:38 -0400
Subject: [PATCH 0282/1791] drm/amd/display: Lowering min Z8 residency time

[Why & How]
Per HW team request, we're lowering the minimum Z8
residency time to 2000us. This enables Z8 support for additional
modes we were previously blocking like 2k>60hz

Cc: stable@vger.kernel.org
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 24806acc84385..abeeede38fb39 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -885,7 +885,7 @@ static const struct dc_plane_cap plane_cap = {
 static const struct dc_debug_options debug_defaults_drv = {
 	.disable_z10 = false,
 	.enable_z9_disable_interface = true,
-	.minimum_z8_residency_time = 3080,
+	.minimum_z8_residency_time = 2000,
 	.psr_skip_crtc_disable = true,
 	.disable_dmcu = true,
 	.force_abm_enable = false,
-- 
GitLab


From 5e9252d8415f50095c854c85cf9ebcc894e9ac0d Mon Sep 17 00:00:00 2001
From: Josip Pavic <Josip.Pavic@amd.com>
Date: Mon, 3 Apr 2023 10:42:06 -0400
Subject: [PATCH 0283/1791] drm/amd/display: add option to use custom backlight
 caps

[Why & How]
Provide option for vendors to specify a custom brightness-to-backlight
conversion profile.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Josip Pavic <Josip.Pavic@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/modules/power/power_helpers.c | 43 +++++++++++++++++++
 .../amd/display/modules/power/power_helpers.h |  3 ++
 2 files changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 51e76bce92eaa..68d95b92df76b 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -116,6 +116,27 @@ static const struct abm_parameters * const abm_settings[] = {
 	abm_settings_config2,
 };
 
+const struct dm_bl_data_point custom_backlight_curve0[] = {
+		{2, 14}, {4, 16}, {6, 18}, {8, 21}, {10, 23}, {12, 26}, {14, 29}, {16, 32}, {18, 35},
+		{20, 38}, {22, 41}, {24, 44}, {26, 48}, {28, 52}, {30, 55}, {32, 59}, {34, 62},
+		{36, 67}, {38, 71}, {40, 75}, {42, 80}, {44, 84}, {46, 88}, {48, 93}, {50, 98},
+		{52, 103}, {54, 108}, {56, 113}, {58, 118}, {60, 123}, {62, 129}, {64, 135}, {66, 140},
+		{68, 146}, {70, 152}, {72, 158}, {74, 164}, {76, 171}, {78, 177}, {80, 183}, {82, 190},
+		{84, 197}, {86, 204}, {88, 211}, {90, 218}, {92, 225}, {94, 232}, {96, 240}, {98, 247}};
+
+struct custom_backlight_profile {
+	uint8_t  ac_level_percentage;
+	uint8_t  dc_level_percentage;
+	uint8_t  min_input_signal;
+	uint8_t  max_input_signal;
+	uint8_t  num_data_points;
+	const struct dm_bl_data_point *data_points;
+};
+
+static const struct custom_backlight_profile custom_backlight_profiles[] = {
+		{100, 32, 12, 255, ARRAY_SIZE(custom_backlight_curve0), custom_backlight_curve0},
+};
+
 #define NUM_AMBI_LEVEL    5
 #define NUM_AGGR_LEVEL    4
 #define NUM_POWER_FN_SEGS 8
@@ -944,3 +965,25 @@ bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link,
 
 	return true;
 }
+
+bool fill_custom_backlight_caps(unsigned int config_no, struct dm_acpi_atif_backlight_caps *caps)
+{
+	unsigned int data_points_size;
+
+	if (config_no >= ARRAY_SIZE(custom_backlight_profiles))
+		return false;
+
+	data_points_size = custom_backlight_profiles[config_no].num_data_points
+			* sizeof(custom_backlight_profiles[config_no].data_points[0]);
+
+	caps->size = sizeof(struct dm_acpi_atif_backlight_caps) - sizeof(caps->data_points) + data_points_size;
+	caps->flags = 0;
+	caps->error_code = 0;
+	caps->ac_level_percentage = custom_backlight_profiles[config_no].ac_level_percentage;
+	caps->dc_level_percentage = custom_backlight_profiles[config_no].dc_level_percentage;
+	caps->min_input_signal = custom_backlight_profiles[config_no].min_input_signal;
+	caps->max_input_signal = custom_backlight_profiles[config_no].max_input_signal;
+	caps->num_data_points = custom_backlight_profiles[config_no].num_data_points;
+	memcpy(caps->data_points, custom_backlight_profiles[config_no].data_points, data_points_size);
+	return true;
+}
diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h
index 1d3079e56799f..ffc924c9991bf 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h
@@ -62,4 +62,7 @@ bool mod_power_only_edp(const struct dc_state *context,
 bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link,
 			      struct dc_stream_state *stream,
 			      struct psr_config *config);
+
+bool fill_custom_backlight_caps(unsigned int config_no,
+		struct dm_acpi_atif_backlight_caps *caps);
 #endif /* MODULES_POWER_POWER_HELPERS_H_ */
-- 
GitLab


From 7a1187eab0111ac52ec216f2c18cb7822fec4a4c Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Tue, 21 Mar 2023 11:31:22 -0400
Subject: [PATCH 0284/1791] drm/amd/display: Program OTG vtotal min/max
 selectors unconditionally

OTG_V_TOTAL_MIN/MAX_SEL bits are required to be programmed to 1 if
writes to OTG timing registers need to be honoured. This is usually
needed only when freesync is active. However, SubVP + DRR requires that
we're able to change timing even without freesync being active (but
supported). By unconditionally writing this bit to 1, we remove an
unnecessary dependency so that DMCUB can change OTG timing whenever it wants.

Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
index 2ee798965bc2b..6ef56fb32131d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
@@ -245,16 +245,9 @@ static void optc32_set_drr(
 		}
 
 		optc->funcs->set_vtotal_min_max(optc, params->vertical_total_min - 1, params->vertical_total_max - 1);
-		optc32_setup_manual_trigger(optc);
-	} else {
-		REG_UPDATE_4(OTG_V_TOTAL_CONTROL,
-				OTG_SET_V_TOTAL_MIN_MASK, 0,
-				OTG_V_TOTAL_MIN_SEL, 0,
-				OTG_V_TOTAL_MAX_SEL, 0,
-				OTG_FORCE_LOCK_ON_EVENT, 0);
-
-		optc->funcs->set_vtotal_min_max(optc, 0, 0);
 	}
+
+	optc32_setup_manual_trigger(optc);
 }
 
 static struct timing_generator_funcs dcn32_tg_funcs = {
-- 
GitLab


From 1068e987ad0be83a109147fe7fa0891700e8d80e Mon Sep 17 00:00:00 2001
From: "Leo (Hanghong) Ma" <hanghong.ma@amd.com>
Date: Tue, 28 Mar 2023 16:26:29 -0400
Subject: [PATCH 0285/1791] drm/amd/display: Update scaler recout data for
 visual confirm

[Why]
Our QA found visual confirm color is not as expected for Auto
Color Management feature test after enable it.

[How]
Calculate scaler recout data when visual confirm enabled to update
the visual confirm bar on the display.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Leo (Hanghong) Ma <hanghong.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/core/dc_resource.c  | 18 ++++++++++++++++++
 .../drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c  | 18 +-----------------
 2 files changed, 19 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 2f704e26219c6..193e09b05f5a4 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -69,6 +69,10 @@
 #include "../dcn32/dcn32_resource.h"
 #include "../dcn321/dcn321_resource.h"
 
+#define VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT 3
+#define VISUAL_CONFIRM_RECT_HEIGHT_MIN 1
+#define VISUAL_CONFIRM_RECT_HEIGHT_MAX 10
+
 #define DC_LOGGER_INIT(logger)
 
 enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id)
@@ -807,6 +811,8 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx)
 	struct rect surf_clip = plane_state->clip_rect;
 	bool split_tb = stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM;
 	int split_count, split_idx;
+	struct dpp *dpp = pipe_ctx->plane_res.dpp;
+	unsigned short visual_confirm_rect_height = VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT;
 
 	calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx);
 	if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE)
@@ -875,6 +881,18 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx)
 				data->recout.width = data->h_active - data->recout.x;
 		}
 	}
+
+	/* Check bounds to ensure the VC bar height was set to a sane value */
+	if (dpp != NULL) {
+		if ((dpp->ctx->dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_RECT_HEIGHT_MIN) &&
+			(dpp->ctx->dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_RECT_HEIGHT_MAX)) {
+			visual_confirm_rect_height = dpp->ctx->dc->debug.visual_confirm_rect_height;
+		}
+
+		if (dpp->ctx->dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE)
+			data->recout.height = data->recout.height -
+					2 * (dpp->inst + visual_confirm_rect_height);
+	}
 }
 
 static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
index b33955928bd0b..7e140c35a0ced 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
@@ -39,9 +39,6 @@
 #define BLACK_OFFSET_RGB_Y 0x0
 #define BLACK_OFFSET_CBCR  0x8000
 
-#define VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT 3
-#define VISUAL_CONFIRM_RECT_HEIGHT_MIN 1
-#define VISUAL_CONFIRM_RECT_HEIGHT_MAX 10
 
 #define REG(reg)\
 	dpp->tf_regs->reg
@@ -591,18 +588,6 @@ static void dpp1_dscl_set_manual_ratio_init(
 static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp,
 				 const struct rect *recout)
 {
-	int visual_confirm_on = 0;
-	unsigned short visual_confirm_rect_height = VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT;
-
-	if (dpp->base.ctx->dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE)
-		visual_confirm_on = 1;
-
-	/* Check bounds to ensure the VC bar height was set to a sane value */
-	if ((dpp->base.ctx->dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_RECT_HEIGHT_MIN) &&
-			(dpp->base.ctx->dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_RECT_HEIGHT_MAX)) {
-		visual_confirm_rect_height = dpp->base.ctx->dc->debug.visual_confirm_rect_height;
-	}
-
 	REG_SET_2(RECOUT_START, 0,
 		  /* First pixel of RECOUT in the active OTG area */
 		  RECOUT_START_X, recout->x,
@@ -613,8 +598,7 @@ static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp,
 		  /* Number of RECOUT horizontal pixels */
 		  RECOUT_WIDTH, recout->width,
 		  /* Number of RECOUT vertical lines */
-		  RECOUT_HEIGHT, recout->height
-			 - visual_confirm_on * 2 * (dpp->base.inst + visual_confirm_rect_height));
+		  RECOUT_HEIGHT, recout->height);
 }
 
 /**
-- 
GitLab


From f477c7b5ec3e4ef87606671b340abf3bdb0cccff Mon Sep 17 00:00:00 2001
From: Alan Liu <HaoPing.Liu@amd.com>
Date: Mon, 10 Apr 2023 11:35:44 +0800
Subject: [PATCH 0286/1791] drm/amd/display: Fix in secure display context
 creation

[Why & How]
We need to store CRTC information in secure_display_ctx, so postpone
the call to amdgpu_dm_crtc_secure_display_create_contexts() until we
initialize all CRTCs.

Cc: stable@vger.kernel.org
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c     | 11 +++++------
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h |  2 +-
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index b265434c77e14..7f0dcd7af4c74 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1776,12 +1776,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 
 		dc_init_callbacks(adev->dm.dc, &init_params);
 	}
-#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
-	adev->dm.secure_display_ctxs = amdgpu_dm_crtc_secure_display_create_contexts(adev);
-	if (!adev->dm.secure_display_ctxs) {
-		DRM_ERROR("amdgpu: failed to initialize secure_display_ctxs.\n");
-	}
-#endif
 	if (dc_is_dmub_outbox_supported(adev->dm.dc)) {
 		init_completion(&adev->dm.dmub_aux_transfer_done);
 		adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL);
@@ -1840,6 +1834,11 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		goto error;
 	}
 
+#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
+	adev->dm.secure_display_ctxs = amdgpu_dm_crtc_secure_display_create_contexts(adev);
+	if (!adev->dm.secure_display_ctxs)
+		DRM_ERROR("amdgpu: failed to initialize secure display contexts.\n");
+#endif
 
 	DRM_DEBUG_DRIVER("KMS initialized.\n");
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
index 935adca6f0486..748e80ef40d0a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
@@ -100,7 +100,7 @@ struct secure_display_context *amdgpu_dm_crtc_secure_display_create_contexts(
 #else
 #define amdgpu_dm_crc_window_is_activated(x)
 #define amdgpu_dm_crtc_handle_crc_window_irq(x)
-#define amdgpu_dm_crtc_secure_display_create_contexts()
+#define amdgpu_dm_crtc_secure_display_create_contexts(x)
 #endif
 
 #endif /* AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ */
-- 
GitLab


From 469a62938a45ef382c9cb7b9fec6c6c1fcd781c0 Mon Sep 17 00:00:00 2001
From: Gabe Teeger <gabe.teeger@amd.com>
Date: Thu, 6 Apr 2023 17:03:06 -0400
Subject: [PATCH 0287/1791] drm/amd/display: update extended blank for dcn314
 onwards

[Why]
Flickering and underflow was observed when testing extended
blank on dcn314.

[What]
Vstartup is contrainted by vblank_nom, so adjusting it to include
non-adjusted vtotal in its calculation during freesync video
means that Vstartup is not changed when vtotal changes.
This fixed the flickering + underflow.

dc_extended_blank_supported function was removed
because extended blank is only relevant to when
zstate is supported. The increased vtotal during
freesync can be passed to dml regardless of whether
extended blank is supported or not, so this function is
not needed.

Updates were made recently in dml to the calculation of
min_dst_y_next_start. Dml input for dcn314 will now
always use the newer calculation for min_dst_y_next_start.
Dml input for older dcn versions remains untouched.

The variable optimized_min_dst_y_next_start
is replaced everywhere with min_dst_y_next_start,
and the updated dml allows min_dst_y_next_start to
increase to an optimized value during freesync video,
then return to default when freesync is disengaged.

Also removed registry key for controlling
extended blank feature.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Gabe Teeger <gabe.teeger@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 21 -----------------
 drivers/gpu/drm/amd/display/dc/dc.h           |  2 --
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.c    |  4 ++--
 .../drm/amd/display/dc/dml/dcn20/dcn20_fpu.c  | 23 +++++++++----------
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     |  3 +--
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 14 +++++++----
 .../dc/dml/dcn314/display_rq_dlg_calc_314.c   | 16 ++++---------
 .../amd/display/dc/dml/display_mode_structs.h |  3 +--
 8 files changed, 29 insertions(+), 57 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index c10243ef93b6b..89a245e3c9ac9 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2607,9 +2607,6 @@ static enum surface_update_type check_update_surfaces_for_stream(
 				stream_update->vrr_active_variable))
 			su_flags->bits.fams_changed = 1;
 
-		if (stream_update->crtc_timing_adjust && dc_extended_blank_supported(dc))
-			su_flags->bits.crtc_timing_adjust = 1;
-
 		if (su_flags->raw != 0)
 			overall_type = UPDATE_TYPE_FULL;
 
@@ -4849,21 +4846,3 @@ void dc_notify_vsync_int_state(struct dc *dc, struct dc_stream_state *stream, bo
 	if (pipe->stream_res.abm && pipe->stream_res.abm->funcs->set_abm_pause)
 		pipe->stream_res.abm->funcs->set_abm_pause(pipe->stream_res.abm, !enable, i, pipe->stream_res.tg->inst);
 }
-
-/**
- * dc_extended_blank_supported - Decide whether extended blank is supported
- *
- * @dc: [in] Current DC state
- *
- * Extended blank is a freesync optimization feature to be enabled in the
- * future.  During the extra vblank period gained from freesync, we have the
- * ability to enter z9/z10.
- *
- * Return:
- * Indicate whether extended blank is supported (%true or %false)
- */
-bool dc_extended_blank_supported(struct dc *dc)
-{
-	return dc->debug.extended_blank_optimization && !dc->debug.disable_z10
-		&& dc->caps.zstate_support && dc->caps.is_apu;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index dc0a61ca99d01..5268d98b96dc4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -2128,8 +2128,6 @@ struct dc_sink_init_data {
 	bool converter_disable_audio;
 };
 
-bool dc_extended_blank_supported(struct dc *dc);
-
 struct dc_sink *dc_sink_create(const struct dc_sink_init_data *init_params);
 
 /* Newer interfaces  */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 5800acf6aae11..f49c1c0d62747 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2130,7 +2130,7 @@ void dcn20_optimize_bandwidth(
 			dc->clk_mgr,
 			context,
 			true);
-	if (dc_extended_blank_supported(dc) && context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW) {
+	if (context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW) {
 		for (i = 0; i < dc->res_pool->pipe_count; ++i) {
 			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
 
@@ -2138,7 +2138,7 @@ void dcn20_optimize_bandwidth(
 				&& pipe_ctx->stream->adjust.v_total_min == pipe_ctx->stream->adjust.v_total_max
 				&& pipe_ctx->stream->adjust.v_total_max > pipe_ctx->stream->timing.v_total)
 					pipe_ctx->plane_res.hubp->funcs->program_extended_blank(pipe_ctx->plane_res.hubp,
-						pipe_ctx->dlg_regs.optimized_min_dst_y_next_start);
+						pipe_ctx->dlg_regs.min_dst_y_next_start);
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index f1c1a4b5fcac3..7661f8946aa31 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -948,10 +948,10 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 {
 	int plane_count;
 	int i;
-	unsigned int optimized_min_dst_y_next_start_us;
+	unsigned int min_dst_y_next_start_us;
 
 	plane_count = 0;
-	optimized_min_dst_y_next_start_us = 0;
+	min_dst_y_next_start_us = 0;
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		if (context->res_ctx.pipe_ctx[i].plane_state)
 			plane_count++;
@@ -973,19 +973,18 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 	else if (context->stream_count == 1 &&  context->streams[0]->signal == SIGNAL_TYPE_EDP) {
 		struct dc_link *link = context->streams[0]->sink->link;
 		struct dc_stream_status *stream_status = &context->stream_status[0];
+		struct dc_stream_state *current_stream = context->streams[0];
 		int minmum_z8_residency = dc->debug.minimum_z8_residency_time > 0 ? dc->debug.minimum_z8_residency_time : 1000;
 		bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z8_residency;
 		bool is_pwrseq0 = link->link_index == 0;
+		bool isFreesyncVideo;
 
-		if (dc_extended_blank_supported(dc)) {
-			for (i = 0; i < dc->res_pool->pipe_count; i++) {
-				if (context->res_ctx.pipe_ctx[i].stream == context->streams[0]
-					&& context->res_ctx.pipe_ctx[i].stream->adjust.v_total_min == context->res_ctx.pipe_ctx[i].stream->adjust.v_total_max
-					&& context->res_ctx.pipe_ctx[i].stream->adjust.v_total_min > context->res_ctx.pipe_ctx[i].stream->timing.v_total) {
-						optimized_min_dst_y_next_start_us =
-							context->res_ctx.pipe_ctx[i].dlg_regs.optimized_min_dst_y_next_start_us;
-						break;
-				}
+		isFreesyncVideo = current_stream->adjust.v_total_min == current_stream->adjust.v_total_max;
+		isFreesyncVideo = isFreesyncVideo && current_stream->timing.v_total < current_stream->adjust.v_total_min;
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			if (context->res_ctx.pipe_ctx[i].stream == current_stream && isFreesyncVideo) {
+				min_dst_y_next_start_us = context->res_ctx.pipe_ctx[i].dlg_regs.min_dst_y_next_start_us;
+				break;
 			}
 		}
 
@@ -993,7 +992,7 @@ static enum dcn_zstate_support_state  decide_zstate_support(struct dc *dc, struc
 		if (stream_status->plane_count > 1)
 			return DCN_ZSTATE_SUPPORT_DISALLOW;
 
-		if (is_pwrseq0 && (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || optimized_min_dst_y_next_start_us > 5000))
+		if (is_pwrseq0 && (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || min_dst_y_next_start_us > 5000))
 			return DCN_ZSTATE_SUPPORT_ALLOW;
 		else if (is_pwrseq0 && link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr)
 			return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
index 2244e4fb8c96d..fcde8f21b8be0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
@@ -987,8 +987,7 @@ static void dml_rq_dlg_get_dlg_params(
 
 	dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start;
 	disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2));
-	disp_dlg_regs->optimized_min_dst_y_next_start_us = 0;
-	disp_dlg_regs->optimized_min_dst_y_next_start = disp_dlg_regs->min_dst_y_next_start;
+	disp_dlg_regs->min_dst_y_next_start_us = 0;
 	ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18));
 
 	dml_print("DML_DLG: %s: min_ttu_vblank (us)         = %3.2f\n", __func__, min_ttu_vblank);
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 44082f65de1fa..19370b872a91d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -286,6 +286,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *pipe;
 	bool upscaled = false;
+	bool isFreesyncVideo = false;
 
 	dc_assert_fp_enabled();
 
@@ -299,9 +300,16 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 		pipe = &res_ctx->pipe_ctx[i];
 		timing = &pipe->stream->timing;
 
-		if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min
-			&& pipe->stream->adjust.v_total_min > timing->v_total)
+		isFreesyncVideo = pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min;
+		isFreesyncVideo = isFreesyncVideo && pipe->stream->adjust.v_total_min > timing->v_total;
+
+		if (!isFreesyncVideo) {
+			pipes[pipe_cnt].pipe.dest.vblank_nom =
+				dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0));
+		} else {
 			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		}
 
 		if (pipe->plane_state &&
 				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
@@ -323,8 +331,6 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0;
 		pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0;
 		pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch;
-		pipes[pipe_cnt].pipe.dest.vblank_nom =
-				dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0));
 		pipes[pipe_cnt].pipe.src.dcc_rate = 3;
 		pipes[pipe_cnt].dout.dsc_input_bpc = 0;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
index ea4eb66066c42..4f945458b2b7e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
@@ -1051,7 +1051,6 @@ static void dml_rq_dlg_get_dlg_params(
 
 	float vba__refcyc_per_req_delivery_pre_l = get_refcyc_per_req_delivery_pre_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;  // From VBA
 	float vba__refcyc_per_req_delivery_l = get_refcyc_per_req_delivery_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz;  // From VBA
-	int blank_lines = 0;
 
 	memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs));
 	memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs));
@@ -1075,17 +1074,10 @@ static void dml_rq_dlg_get_dlg_params(
 	min_ttu_vblank = get_min_ttu_vblank_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx);	// From VBA
 
 	dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start;
-	disp_dlg_regs->optimized_min_dst_y_next_start = disp_dlg_regs->min_dst_y_next_start;
-	disp_dlg_regs->optimized_min_dst_y_next_start_us = 0;
-	disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2));
-	blank_lines = (dst->vblank_end + dst->vtotal_min - dst->vblank_start - dst->vstartup_start - 1);
-	if (blank_lines < 0)
-		blank_lines = 0;
-	if (blank_lines != 0) {
-		disp_dlg_regs->optimized_min_dst_y_next_start = vba__min_dst_y_next_start;
-		disp_dlg_regs->optimized_min_dst_y_next_start_us = (disp_dlg_regs->optimized_min_dst_y_next_start * dst->hactive) / (unsigned int) dst->pixel_rate_mhz;
-		disp_dlg_regs->min_dst_y_next_start = disp_dlg_regs->optimized_min_dst_y_next_start;
-	}
+	disp_dlg_regs->min_dst_y_next_start_us =
+		(vba__min_dst_y_next_start * dst->hactive) / (unsigned int) dst->pixel_rate_mhz;
+	disp_dlg_regs->min_dst_y_next_start = vba__min_dst_y_next_start * dml_pow(2, 2);
+
 	ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18));
 
 	dml_print("DML_DLG: %s: min_ttu_vblank (us)         = %3.2f\n", __func__, min_ttu_vblank);
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
index 3c077164f3620..ff0246a9458fd 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
@@ -619,8 +619,7 @@ struct _vcs_dpi_display_dlg_regs_st {
 	unsigned int refcyc_h_blank_end;
 	unsigned int dlg_vblank_end;
 	unsigned int min_dst_y_next_start;
-	unsigned int optimized_min_dst_y_next_start;
-	unsigned int optimized_min_dst_y_next_start_us;
+	unsigned int min_dst_y_next_start_us;
 	unsigned int refcyc_per_htotal;
 	unsigned int refcyc_x_after_scaler;
 	unsigned int dst_y_after_scaler;
-- 
GitLab


From 9c25ab167df412a5474dedfd0e7743e76bc89cbe Mon Sep 17 00:00:00 2001
From: Sung Lee <sunglee@amd.com>
Date: Mon, 10 Apr 2023 14:15:14 -0400
Subject: [PATCH 0288/1791] drm/amd/display: Add p-state debugging

[WHY]
P-State related issues are fairly common but currently
there is no way to debug these issues after the fact.

[HOW]
Add helpful registers to HW state queries

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Sung Lee <sunglee@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c | 5 +++++
 drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h    | 1 +
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h         | 3 ++-
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c   | 3 +++
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h   | 1 +
 5 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
index 24bd932199366..0ddd310cc9710 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
@@ -623,6 +623,11 @@ void hubbub2_read_state(struct hubbub *hubbub, struct dcn_hubbub_state *hubbub_s
 		 REG_GET(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_VMID, &hubbub_state->vm_error_vmid);
 		 REG_GET(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_PIPE, &hubbub_state->vm_error_pipe);
 	}
+
+	if (REG(DCHUBBUB_TEST_DEBUG_INDEX) && REG(DCHUBBUB_TEST_DEBUG_DATA)) {
+		REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, 0x6);
+		hubbub_state->test_debug_data = REG_READ(DCHUBBUB_TEST_DEBUG_DATA);
+	}
 }
 
 static const struct hubbub_funcs hubbub2_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index beb26dc8a07fa..aa80b3f2ca3f8 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -111,6 +111,7 @@ struct dcn_hubbub_state {
 	uint32_t vm_error_vmid;
 	uint32_t vm_error_pipe;
 	uint32_t vm_error_mode;
+	uint32_t test_debug_data;
 };
 
 struct hubbub_funcs {
diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index d35432c218564..7c9a2b34bd05c 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -271,7 +271,7 @@ struct dmub_srv_hw_params {
  */
 struct dmub_diagnostic_data {
 	uint32_t dmcub_version;
-	uint32_t scratch[16];
+	uint32_t scratch[17];
 	uint32_t pc;
 	uint32_t undefined_address_fault_addr;
 	uint32_t inst_fetch_fault_addr;
@@ -282,6 +282,7 @@ struct dmub_diagnostic_data {
 	uint32_t inbox0_rptr;
 	uint32_t inbox0_wptr;
 	uint32_t inbox0_size;
+	uint32_t gpint_datain0;
 	uint8_t is_dmcub_enabled : 1;
 	uint8_t is_dmcub_soft_reset : 1;
 	uint8_t is_dmcub_secure_reset : 1;
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index 21dd6cbdb1063..bf5994e292d94 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -439,6 +439,7 @@ void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnosti
 	diag_data->scratch[13] = REG_READ(DMCUB_SCRATCH13);
 	diag_data->scratch[14] = REG_READ(DMCUB_SCRATCH14);
 	diag_data->scratch[15] = REG_READ(DMCUB_SCRATCH15);
+	diag_data->scratch[16] = REG_READ(DMCUB_SCRATCH16);
 
 	diag_data->undefined_address_fault_addr = REG_READ(DMCUB_UNDEFINED_ADDRESS_FAULT_ADDR);
 	diag_data->inst_fetch_fault_addr = REG_READ(DMCUB_INST_FETCH_FAULT_ADDR);
@@ -469,6 +470,8 @@ void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnosti
 
 	REG_GET(DMCUB_REGION3_CW6_TOP_ADDRESS, DMCUB_REGION3_CW6_ENABLE, &is_cw6_enabled);
 	diag_data->is_cw6_enabled = is_cw6_enabled;
+
+	diag_data->gpint_datain0 = REG_READ(DMCUB_GPINT_DATAIN0);
 }
 void dmub_dcn32_configure_dmub_in_system_memory(struct dmub_srv *dmub)
 {
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
index f15336b6e22be..d58a1e4b9f1c1 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
@@ -107,6 +107,7 @@ struct dmub_srv;
 	DMUB_SR(DMCUB_SCRATCH15) \
 	DMUB_SR(DMCUB_SCRATCH16) \
 	DMUB_SR(DMCUB_SCRATCH17) \
+	DMUB_SR(DMCUB_GPINT_DATAIN0) \
 	DMUB_SR(DMCUB_GPINT_DATAIN1) \
 	DMUB_SR(DMCUB_GPINT_DATAOUT) \
 	DMUB_SR(CC_DC_PIPE_DIS) \
-- 
GitLab


From c8cefb99fc811304fff9590677994531ff0ac992 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Wed, 12 Apr 2023 15:42:41 -0400
Subject: [PATCH 0289/1791] drm/amd/display: For no plane case set pstate
 support in validation

- Previously update_clocks was overriding pstate support if
  it checked that there were no planes
- However, P-State support should be determined in validation
  phase instead
- This fixes an issue where a transition from FPO -> no planes
  expects UCLK MAX, but update_clocks was overriding to set
  UCLK to min

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c  | 4 +---
 .../gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  | 7 ++-----
 drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c      | 8 +++++++-
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c      | 1 +
 4 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
index 694a9d3d92aee..3908e7cfd6cbf 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
@@ -206,7 +206,6 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base,
 	bool force_reset = false;
 	bool update_uclk = false;
 	bool p_state_change_support;
-	int total_plane_count;
 
 	if (dc->work_arounds.skip_clock_update || !clk_mgr->smu_present)
 		return;
@@ -247,8 +246,7 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base,
 		clk_mgr_base->clks.socclk_khz = new_clocks->socclk_khz;
 
 	clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support;
-	total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context);
-	p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0);
+	p_state_change_support = new_clocks->p_state_change_support;
 
 	// invalidate the current P-State forced min in certain dc_mode_softmax situations
 	if (dc->clk_mgr->dc_mode_softmax_enabled && safe_to_lower && !p_state_change_support) {
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index eea103908b09f..85e963ec25ab1 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -459,7 +459,6 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 	bool update_uclk = false, update_fclk = false;
 	bool p_state_change_support;
 	bool fclk_p_state_change_support;
-	int total_plane_count;
 
 	if (dc->work_arounds.skip_clock_update)
 		return;
@@ -488,8 +487,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 
 		clk_mgr_base->clks.fclk_prev_p_state_change_support = clk_mgr_base->clks.fclk_p_state_change_support;
 
-		total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context);
-		fclk_p_state_change_support = new_clocks->fclk_p_state_change_support || (total_plane_count == 0);
+		fclk_p_state_change_support = new_clocks->fclk_p_state_change_support;
 
 		if (should_update_pstate_support(safe_to_lower, fclk_p_state_change_support, clk_mgr_base->clks.fclk_p_state_change_support)) {
 			clk_mgr_base->clks.fclk_p_state_change_support = fclk_p_state_change_support;
@@ -528,8 +526,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 			dcn32_smu_send_cab_for_uclk_message(clk_mgr, clk_mgr_base->clks.num_ways);
 		}
 
-
-		p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0);
+		p_state_change_support = new_clocks->p_state_change_support;
 		if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support)) {
 			clk_mgr_base->clks.p_state_change_support = p_state_change_support;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index 7661f8946aa31..a5b1e4bb1a222 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -1042,7 +1042,7 @@ void dcn20_calculate_dlg_params(struct dc *dc,
 				int pipe_cnt,
 				int vlevel)
 {
-	int i, pipe_idx;
+	int i, pipe_idx, active_hubp_count = 0;
 
 	dc_assert_fp_enabled();
 
@@ -1078,6 +1078,8 @@ void dcn20_calculate_dlg_params(struct dc *dc,
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		if (!context->res_ctx.pipe_ctx[i].stream)
 			continue;
+		if (context->res_ctx.pipe_ctx[i].plane_state)
+			active_hubp_count++;
 		pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
 		pipes[pipe_idx].pipe.dest.vupdate_offset = get_vupdate_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
 		pipes[pipe_idx].pipe.dest.vupdate_width = get_vupdate_width(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
@@ -1104,6 +1106,10 @@ void dcn20_calculate_dlg_params(struct dc *dc,
 
 		pipe_idx++;
 	}
+	/* If DCN isn't making memory requests we can allow pstate change */
+	if (!active_hubp_count) {
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+	}
 	/*save a original dppclock copy*/
 	context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz;
 	context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index e1e82ad4c583c..2624236d20d06 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -1433,6 +1433,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context,
 		context->bw_ctx.bw.dcn.clk.dramclk_khz = 0;
 		context->bw_ctx.bw.dcn.clk.fclk_khz = 0;
 		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+		context->bw_ctx.bw.dcn.clk.fclk_p_state_change_support = true;
 	}
 	/*save a original dppclock copy*/
 	context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz;
-- 
GitLab


From 87f0c16e0eeb672fb888b4e173edff0252e02757 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Mon, 17 Apr 2023 10:23:05 -0400
Subject: [PATCH 0290/1791] drm/amd/display: Enable SubVP for high refresh rate
 displays

[Description]
- Add debug option to enable SubVP for high refresh rate displays
- For now limit the enabled modes based on a table in debug options
- Currently disabled by default

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           |  1 +
 .../drm/amd/display/dc/dcn32/dcn32_resource.c |  1 +
 .../drm/amd/display/dc/dcn32/dcn32_resource.h | 14 +++++
 .../display/dc/dcn32/dcn32_resource_helpers.c | 15 +++++
 .../amd/display/dc/dcn321/dcn321_resource.c   |  1 +
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 62 ++++++++++++++++++-
 .../dc/dml/dcn32/display_mode_vba_32.c        | 12 ++--
 7 files changed, 101 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 5268d98b96dc4..e41107098ffe2 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -881,6 +881,7 @@ struct dc_debug_options {
 	bool disable_boot_optimizations;
 	bool override_odm_optimization;
 	bool minimize_dispclk_using_odm;
+	bool disable_subvp_high_refresh;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 0085ea78ea317..7feeba78c0c9e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -728,6 +728,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_margin_us = 2000, // 2000us
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
+	.disable_subvp_high_refresh = true,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 3937dbc1e5526..04be01ae1ecfd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -40,6 +40,7 @@
 #define DCN3_2_VMIN_DISPCLK_HZ 717000000
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
 #define DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US 100 // Only allow FPO + Vactive if active margin >= 100
+#define SUBVP_HIGH_REFRESH_LIST_LEN 3
 
 #define TO_DCN32_RES_POOL(pool)\
 	container_of(pool, struct dcn32_resource_pool, base)
@@ -47,6 +48,15 @@
 extern struct _vcs_dpi_ip_params_st dcn3_2_ip;
 extern struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc;
 
+struct subvp_high_refresh_list {
+	int min_refresh;
+	int max_refresh;
+	struct resolution {
+		int width;
+		int height;
+	} res[SUBVP_HIGH_REFRESH_LIST_LEN];
+};
+
 struct dcn32_resource_pool {
 	struct resource_pool base;
 };
@@ -151,10 +161,14 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre
 
 bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe);
 
+bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe);
+
 unsigned int dcn32_calc_num_avail_chans_for_mall(struct dc *dc, int num_chans);
 
 double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *context);
 
+bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int width, unsigned int height);
+
 /* definitions for run time init of reg offsets */
 
 /* CLK SRC */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index eeca16faf31af..df912c333bbdb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -656,3 +656,18 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre
 
 	return fpo_candidate_stream;
 }
+
+bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int width, unsigned int height)
+{
+	bool is_native_scaling = false;
+
+	if (pipe->stream->timing.h_addressable == width &&
+			pipe->stream->timing.v_addressable == height &&
+			pipe->plane_state->src_rect.width == width &&
+			pipe->plane_state->src_rect.height == height &&
+			pipe->plane_state->dst_rect.width == width &&
+			pipe->plane_state->dst_rect.height == height)
+		is_native_scaling = true;
+
+	return is_native_scaling;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index 138657c38afe4..63bd6928c82f0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -727,6 +727,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_margin_us = 2000, // 2000us
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
+	.disable_subvp_high_refresh = true,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 2624236d20d06..9cc0e60890325 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -35,6 +35,15 @@
 
 #define DC_LOGGER_INIT(logger)
 
+static const struct subvp_high_refresh_list subvp_high_refresh_list = {
+			.min_refresh = 120,
+			.max_refresh = 165,
+			.res = {
+				{.width = 3840, .height = 2160, },
+				{.width = 3440, .height = 1440, },
+				{.width = 2560, .height = 1440, }},
+};
+
 struct _vcs_dpi_ip_params_st dcn3_2_ip = {
 	.gpuvm_enable = 0,
 	.gpuvm_max_page_table_levels = 4,
@@ -694,7 +703,9 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 		 */
 		if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) &&
 				(!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) &&
-				pipe->stream->mall_stream_config.type == SUBVP_NONE && refresh_rate < 120 && !pipe->plane_state->address.tmz_surface &&
+				pipe->stream->mall_stream_config.type == SUBVP_NONE &&
+				(refresh_rate < 120 || dcn32_allow_subvp_high_refresh_rate(dc, context, pipe)) &&
+				!pipe->plane_state->address.tmz_surface &&
 				(vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0 ||
 				(vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 &&
 						dcn32_allow_subvp_with_active_margin(pipe)))) {
@@ -2789,6 +2800,55 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe)
 	return allow;
 }
 
+/**
+ * ************************************************************************************************
+ * dcn32_allow_subvp_high_refresh_rate: Determine if the high refresh rate config will allow subvp
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ * @param [in]: pipe: Pipe to be considered for use in subvp
+ *
+ * On high refresh rate display configs, we will allow subvp under the following conditions:
+ * 1. Resolution is 3840x2160, 3440x1440, or 2560x1440
+ * 2. Refresh rate is between 120hz - 165hz
+ * 3. No scaling
+ * 4. Freesync is inactive
+ * 5. For single display cases, freesync must be disabled
+ *
+ * @return: True if pipe can be used for subvp, false otherwise
+ *
+ * ************************************************************************************************
+ */
+bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe)
+{
+	bool allow = false;
+	uint32_t refresh_rate = 0;
+	uint32_t min_refresh = subvp_high_refresh_list.min_refresh;
+	uint32_t max_refresh = subvp_high_refresh_list.max_refresh;
+	uint32_t i;
+
+	if (!dc->debug.disable_subvp_high_refresh && pipe->stream &&
+			pipe->plane_state && !pipe->stream->vrr_active_variable) {
+		refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 +
+						pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1)
+						/ (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total);
+		if (refresh_rate >= min_refresh && refresh_rate <= max_refresh) {
+			for (i = 0; i < SUBVP_HIGH_REFRESH_LIST_LEN; i++) {
+				uint32_t width = subvp_high_refresh_list.res[i].width;
+				uint32_t height = subvp_high_refresh_list.res[i].height;
+
+				if (dcn32_check_native_scaling_for_res(pipe, width, height)) {
+					if ((context->stream_count == 1 && !pipe->stream->allow_freesync) || context->stream_count > 1) {
+						allow = true;
+						break;
+					}
+				}
+			}
+		}
+	}
+	return allow;
+}
+
 /**
  * *******************************************************************************************
  * dcn32_determine_max_vratio_prefetch: Determine max Vratio for prefetch by driver policy
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
index 13c7e7394b1c7..66f44a013fe5d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
@@ -2322,10 +2322,14 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 		mode_lib->vba.LinkCapacitySupport[i] = true;
 		for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) {
 			if (mode_lib->vba.BlendingAndTiming[k] == k
-					&& (mode_lib->vba.Output[k] == dm_dp || mode_lib->vba.Output[k] == dm_dp2p0
-							|| mode_lib->vba.Output[k] == dm_edp
-							|| mode_lib->vba.Output[k] == dm_hdmi)
-					&& mode_lib->vba.OutputBppPerState[i][k] == 0) {
+				&& (mode_lib->vba.Output[k] == dm_dp || mode_lib->vba.Output[k] == dm_dp2p0
+					|| mode_lib->vba.Output[k] == dm_edp
+					|| mode_lib->vba.Output[k] == dm_hdmi)
+				&& mode_lib->vba.OutputBppPerState[i][k] == 0 &&
+				(mode_lib->vba.UsesMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe)) {
+				/* Phantom pipes don't consider DSC in DML, so it could fail link check.
+				 * However, we don't care about the link for phantom pipes.
+				 */
 				mode_lib->vba.LinkCapacitySupport[i] = false;
 			}
 		}
-- 
GitLab


From 1938bcdc4b530d6413e03f238c2df56f99f17220 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Wed, 12 Apr 2023 17:06:46 -0400
Subject: [PATCH 0291/1791] drm/amd/display: Query GECC enable for SubVP
 disable

- We want to disable SubVP if Graphics Error Correction/Correcting Code
  (GECC) is enabled.
- After reading feature caps from DMCUB during init, use the GECC
  enable/disable info to determine if SubVP can be enabled or not.

Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                  | 1 +
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c   | 1 +
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 2 +-
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index e41107098ffe2..f52ab717ac7d7 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -210,6 +210,7 @@ struct dc_dmub_caps {
 	bool psr;
 	bool mclk_sw;
 	bool subvp_psr;
+	bool gecc_enable;
 };
 
 struct dc_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 1e9ada8b131ce..4950eaa4406b2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -945,6 +945,7 @@ void dcn32_init_hw(struct dc *dc)
 		dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv);
 		dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr;
 		dc->caps.dmub_caps.subvp_psr = dc->ctx->dmub_srv->dmub->feature_caps.subvp_psr_support;
+		dc->caps.dmub_caps.gecc_enable = dc->ctx->dmub_srv->dmub->feature_caps.gecc_enable;
 		dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch;
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 9cc0e60890325..a73cacef6e48d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -1141,7 +1141,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 	 * 4. Display configuration passes validation
 	 * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch)
 	 */
-	if (!dc->debug.force_disable_subvp && dcn32_all_pipes_have_stream_and_plane(dc, context) &&
+	if (!dc->debug.force_disable_subvp && !dc->caps.dmub_caps.gecc_enable && dcn32_all_pipes_have_stream_and_plane(dc, context) &&
 	    !dcn32_mpo_in_use(context) && !dcn32_any_surfaces_rotated(dc, context) &&
 		(*vlevel == context->bw_ctx.dml.soc.num_states ||
 	    vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported ||
-- 
GitLab


From 5a096b73c8fed3a9987ba15378285df360e2284b Mon Sep 17 00:00:00 2001
From: Michael Strauss <michael.strauss@amd.com>
Date: Tue, 11 Apr 2023 12:44:54 -0400
Subject: [PATCH 0292/1791] drm/amd/display: Keep disable aux-i delay as 0

[WHY]
Current Aux-I sequence checks for local_sink which isn't populated on
MST links

[HOW]
Leave disable aux-i delay as 0 for MST cases

Cc: stable@vger.kernel.org
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Reviewed-by: George Shen <George.Shen@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../link_dp_training_fixed_vs_pe_retimer.c       | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
index fb6c938c6dab1..15faaf645b145 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c
@@ -233,8 +233,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 			link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
 	const uint8_t vendor_lttpr_write_data_intercept_en[4] = {0x1, 0x55, 0x63, 0x0};
 	const uint8_t vendor_lttpr_write_data_intercept_dis[4] = {0x1, 0x55, 0x63, 0x68};
-	uint32_t pre_disable_intercept_delay_ms =
-			link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
+	uint32_t pre_disable_intercept_delay_ms = 0;
 	uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0};
 	uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0};
 	uint32_t vendor_lttpr_write_address = 0xF004F;
@@ -245,6 +244,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy(
 	uint8_t toggle_rate;
 	uint8_t rate;
 
+	if (link->local_sink)
+		pre_disable_intercept_delay_ms =
+				link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
+
 	/* Only 8b/10b is supported */
 	ASSERT(link_dp_get_encoding_format(&lt_settings->link_settings) ==
 			DP_8b_10b_ENCODING);
@@ -595,10 +598,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 	const uint8_t vendor_lttpr_write_data_adicora_eq3[4] = {0x1, 0x55, 0x63, 0x68};
 	uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0};
 	uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0};
-	uint32_t pre_disable_intercept_delay_ms =
-			link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
-
-
+	uint32_t pre_disable_intercept_delay_ms = 0;
 	uint32_t vendor_lttpr_write_address = 0xF004F;
 	enum link_training_result status = LINK_TRAINING_SUCCESS;
 	uint8_t lane = 0;
@@ -607,6 +607,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence(
 	uint8_t toggle_rate;
 	uint8_t rate;
 
+	if (link->local_sink)
+		pre_disable_intercept_delay_ms =
+				link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms;
+
 	/* Only 8b/10b is supported */
 	ASSERT(link_dp_get_encoding_format(&lt_settings->link_settings) ==
 			DP_8b_10b_ENCODING);
-- 
GitLab


From 612c5ad50c3e5505e674b7cc50bf6527bc0adee6 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 16 Apr 2023 20:35:01 -0400
Subject: [PATCH 0293/1791] drm/amd/display: 3.2.233

This DC version brings along:
- Improvements in the SubVP feature
- Keep disable aux-i delay as 0
- Add p-state debugging and improvements
- Fix in secure display context creation
- add option to use custom backlight caps
- Lowering min Z8 residency time
- Restore rptr/wptr for DMCUB as workaround
- Update FW feature caps struct

Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index f52ab717ac7d7..4424e7abb8010 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.232"
+#define DC_VER "3.2.233"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From f38129bb081758176dd78304faaee95007fb8838 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Tue, 21 Mar 2023 10:25:04 -0400
Subject: [PATCH 0294/1791] Revert "drm/amd/display: disable SubVP + DRR to
 prevent underflow"

This reverts commit 80c6d6804f31451848a3956a70c2bcb1f07cfcb0.
The orignal commit was intended as a workaround to prevent underflow and
flickering when using one normal monitor and the other high refresh rate
monitor (> 120Hz).

This patch is being reverted in favour of a software solution to enable
SubVP+DRR

Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c    | 5 -----
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 4 ----
 drivers/gpu/drm/amd/include/amd_shared.h             | 1 -
 3 files changed, 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 7f0dcd7af4c74..dd56e1512b0ae 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1646,11 +1646,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	if (amdgpu_dc_feature_mask & DC_DISABLE_LTTPR_DP2_0)
 		init_data.flags.allow_lttpr_non_transparent_mode.bits.DP2_0 = true;
 
-	/* Disable SubVP + DRR config by default */
-	init_data.flags.disable_subvp_drr = true;
-	if (amdgpu_dc_feature_mask & DC_ENABLE_SUBVP_DRR)
-		init_data.flags.disable_subvp_drr = false;
-
 	init_data.flags.seamless_boot_edp_requested = false;
 
 	if (check_seamless_boot_capability(adev)) {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index a73cacef6e48d..826059d5b3675 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -892,10 +892,6 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struc
 	int16_t stretched_drr_us = 0;
 	int16_t drr_stretched_vblank_us = 0;
 	int16_t max_vblank_mallregion = 0;
-	const struct dc_config *config = &dc->config;
-
-	if (config->disable_subvp_drr)
-		return false;
 
 	// Find SubVP pipe
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index e4a22c68517d1..f175e65b853a0 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -240,7 +240,6 @@ enum DC_FEATURE_MASK {
 	DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default
 	DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default
 	DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default
-	DC_ENABLE_SUBVP_DRR = (1 << 9), // 0x200, disabled by default
 };
 
 enum DC_DEBUG_MASK {
-- 
GitLab


From 179661ad45cb18ba7702ebafde5f22aea47be5e0 Mon Sep 17 00:00:00 2001
From: Likun Gao <Likun.Gao@amd.com>
Date: Fri, 21 Apr 2023 11:30:18 +0800
Subject: [PATCH 0295/1791] drm/amdgpu: support psp vbflash sysfs for MP0
 13_0_10

Add support for PSP vbflash sysfs interface with MP0 version v13.0.10.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index a496bf2fb1991..c58654a8b6c5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -3618,6 +3618,7 @@ int amdgpu_psp_sysfs_init(struct amdgpu_device *adev)
 	switch (adev->ip_versions[MP0_HWIP][0]) {
 	case IP_VERSION(13, 0, 0):
 	case IP_VERSION(13, 0, 7):
+	case IP_VERSION(13, 0, 10):
 		if (!psp->adev) {
 			psp->adev = adev;
 			psp_v13_0_set_psp_funcs(psp);
-- 
GitLab


From b91075866e58e6b073689958f246834ad0b2c79c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 10 Apr 2023 12:02:29 -0400
Subject: [PATCH 0296/1791] drm/amdgpu/gfx11: update gpu_clock_counter logic

This code was written prior to previous updates to this
logic for other chips.  The RSC registers are part of
SMUIO which is an always on block so there is no need
to disable gfxoff.  Additionally add the carryover and
preemption checks.

v2: rebase

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 543af07ff102a..8a4c4769e6074 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -4714,24 +4714,27 @@ static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 	uint64_t clock;
 	uint64_t clock_counter_lo, clock_counter_hi_pre, clock_counter_hi_after;
 
-	amdgpu_gfx_off_ctrl(adev, false);
-	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_gfx_off_ctrl(adev, false);
+		mutex_lock(&adev->gfx.gpu_clock_mutex);
 		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
 		clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
 		clock_counter_hi_after = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
 		if (clock_counter_hi_pre != clock_counter_hi_after)
 			clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
+		mutex_unlock(&adev->gfx.gpu_clock_mutex);
+		amdgpu_gfx_off_ctrl(adev, true);
 	} else {
+		preempt_disable();
 		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
 		clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
 		clock_counter_hi_after = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
 		if (clock_counter_hi_pre != clock_counter_hi_after)
 			clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
+		preempt_enable();
 	}
 	clock = clock_counter_lo | (clock_counter_hi_after << 32ULL);
-	mutex_unlock(&adev->gfx.gpu_clock_mutex);
-	amdgpu_gfx_off_ctrl(adev, true);
+
 	return clock;
 }
 
-- 
GitLab


From 341ad0e8e254267704e0b87e35ad23aba5c02359 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Mon, 24 Apr 2023 11:29:01 -0700
Subject: [PATCH 0297/1791] drm/i915/mtl: Add PTE encode function

PTE encode functions are platform dependent. This patch implements
PTE functions for MTL, and ensures the correct PTE encode function
is used by calling pte_encode function pointer instead of the
hardcoded gen8 version of PTE encode.

Fixes: b76c0deef627 ("drm/i915/mtl: Define MOCS and PAT tables for MTL")
Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424182902.3663500-2-fei.yang@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpt.c |  2 +-
 drivers/gpu/drm/i915/gt/gen8_ppgtt.c     | 45 ++++++++++++++++++++----
 drivers/gpu/drm/i915/gt/intel_ggtt.c     | 36 +++++++++++++++++--
 drivers/gpu/drm/i915/gt/intel_gtt.h      | 12 +++++--
 4 files changed, 82 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c
index b8027392144de..c5eacfdba1a53 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt.c
@@ -300,7 +300,7 @@ intel_dpt_create(struct intel_framebuffer *fb)
 	vm->vma_ops.bind_vma    = dpt_bind_vma;
 	vm->vma_ops.unbind_vma  = dpt_unbind_vma;
 
-	vm->pte_encode = gen8_ggtt_pte_encode;
+	vm->pte_encode = vm->gt->ggtt->vm.pte_encode;
 
 	dpt->obj = dpt_obj;
 	dpt->obj->is_dpt = true;
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 4daaa6f556688..4c9a2f2db9081 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -55,6 +55,34 @@ static u64 gen8_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
+static u64 mtl_pte_encode(dma_addr_t addr,
+			  enum i915_cache_level level,
+			  u32 flags)
+{
+	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT | GEN8_PAGE_RW;
+
+	if (unlikely(flags & PTE_READ_ONLY))
+		pte &= ~GEN8_PAGE_RW;
+
+	if (flags & PTE_LM)
+		pte |= GEN12_PPGTT_PTE_LM;
+
+	switch (level) {
+	case I915_CACHE_NONE:
+		pte |= GEN12_PPGTT_PTE_PAT1;
+		break;
+	case I915_CACHE_LLC:
+	case I915_CACHE_L3_LLC:
+		pte |= GEN12_PPGTT_PTE_PAT0 | GEN12_PPGTT_PTE_PAT1;
+		break;
+	case I915_CACHE_WT:
+		pte |= GEN12_PPGTT_PTE_PAT0;
+		break;
+	}
+
+	return pte;
+}
+
 static void gen8_ppgtt_notify_vgt(struct i915_ppgtt *ppgtt, bool create)
 {
 	struct drm_i915_private *i915 = ppgtt->vm.i915;
@@ -427,7 +455,7 @@ gen8_ppgtt_insert_pte(struct i915_ppgtt *ppgtt,
 		      u32 flags)
 {
 	struct i915_page_directory *pd;
-	const gen8_pte_t pte_encode = gen8_pte_encode(0, cache_level, flags);
+	const gen8_pte_t pte_encode = ppgtt->vm.pte_encode(0, cache_level, flags);
 	gen8_pte_t *vaddr;
 
 	pd = i915_pd_entry(pdp, gen8_pd_index(idx, 2));
@@ -580,7 +608,7 @@ static void gen8_ppgtt_insert_huge(struct i915_address_space *vm,
 				   enum i915_cache_level cache_level,
 				   u32 flags)
 {
-	const gen8_pte_t pte_encode = gen8_pte_encode(0, cache_level, flags);
+	const gen8_pte_t pte_encode = vm->pte_encode(0, cache_level, flags);
 	unsigned int rem = sg_dma_len(iter->sg);
 	u64 start = vma_res->start;
 
@@ -743,7 +771,7 @@ static void gen8_ppgtt_insert_entry(struct i915_address_space *vm,
 	GEM_BUG_ON(pt->is_compact);
 
 	vaddr = px_vaddr(pt);
-	vaddr[gen8_pd_index(idx, 0)] = gen8_pte_encode(addr, level, flags);
+	vaddr[gen8_pd_index(idx, 0)] = vm->pte_encode(addr, level, flags);
 	drm_clflush_virt_range(&vaddr[gen8_pd_index(idx, 0)], sizeof(*vaddr));
 }
 
@@ -773,7 +801,7 @@ static void __xehpsdv_ppgtt_insert_entry_lm(struct i915_address_space *vm,
 	}
 
 	vaddr = px_vaddr(pt);
-	vaddr[gen8_pd_index(idx, 0) / 16] = gen8_pte_encode(addr, level, flags);
+	vaddr[gen8_pd_index(idx, 0) / 16] = vm->pte_encode(addr, level, flags);
 }
 
 static void xehpsdv_ppgtt_insert_entry(struct i915_address_space *vm,
@@ -820,8 +848,8 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 		pte_flags |= PTE_LM;
 
 	vm->scratch[0]->encode =
-		gen8_pte_encode(px_dma(vm->scratch[0]),
-				I915_CACHE_NONE, pte_flags);
+		vm->pte_encode(px_dma(vm->scratch[0]),
+			       I915_CACHE_NONE, pte_flags);
 
 	for (i = 1; i <= vm->top; i++) {
 		struct drm_i915_gem_object *obj;
@@ -963,7 +991,10 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt,
 	 */
 	ppgtt->vm.alloc_scratch_dma = alloc_pt_dma;
 
-	ppgtt->vm.pte_encode = gen8_pte_encode;
+	if (GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 70))
+		ppgtt->vm.pte_encode = mtl_pte_encode;
+	else
+		ppgtt->vm.pte_encode = gen8_pte_encode;
 
 	ppgtt->vm.bind_async_flags = I915_VMA_LOCAL_BIND;
 	ppgtt->vm.insert_entries = gen8_ppgtt_insert;
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index 3c7f1ed92f5bb..20915edc8bd90 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -220,6 +220,33 @@ static void guc_ggtt_invalidate(struct i915_ggtt *ggtt)
 	}
 }
 
+static u64 mtl_ggtt_pte_encode(dma_addr_t addr,
+			       enum i915_cache_level level,
+			       u32 flags)
+{
+	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT;
+
+	WARN_ON_ONCE(addr & ~GEN12_GGTT_PTE_ADDR_MASK);
+
+	if (flags & PTE_LM)
+		pte |= GEN12_GGTT_PTE_LM;
+
+	switch (level) {
+	case I915_CACHE_NONE:
+		pte |= MTL_GGTT_PTE_PAT1;
+		break;
+	case I915_CACHE_LLC:
+	case I915_CACHE_L3_LLC:
+		pte |= MTL_GGTT_PTE_PAT0 | MTL_GGTT_PTE_PAT1;
+		break;
+	case I915_CACHE_WT:
+		pte |= MTL_GGTT_PTE_PAT0;
+		break;
+	}
+
+	return pte;
+}
+
 u64 gen8_ggtt_pte_encode(dma_addr_t addr,
 			 enum i915_cache_level level,
 			 u32 flags)
@@ -247,7 +274,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm,
 	gen8_pte_t __iomem *pte =
 		(gen8_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE;
 
-	gen8_set_pte(pte, gen8_ggtt_pte_encode(addr, level, flags));
+	gen8_set_pte(pte, ggtt->vm.pte_encode(addr, level, flags));
 
 	ggtt->invalidate(ggtt);
 }
@@ -257,8 +284,8 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 				     enum i915_cache_level level,
 				     u32 flags)
 {
-	const gen8_pte_t pte_encode = gen8_ggtt_pte_encode(0, level, flags);
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
+	const gen8_pte_t pte_encode = ggtt->vm.pte_encode(0, level, flags);
 	gen8_pte_t __iomem *gte;
 	gen8_pte_t __iomem *end;
 	struct sgt_iter iter;
@@ -981,7 +1008,10 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	ggtt->vm.vma_ops.bind_vma    = intel_ggtt_bind_vma;
 	ggtt->vm.vma_ops.unbind_vma  = intel_ggtt_unbind_vma;
 
-	ggtt->vm.pte_encode = gen8_ggtt_pte_encode;
+	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70))
+		ggtt->vm.pte_encode = mtl_ggtt_pte_encode;
+	else
+		ggtt->vm.pte_encode = gen8_ggtt_pte_encode;
 
 	return ggtt_probe_common(ggtt, size);
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index ea17849e7a5c8..1910683f03b4d 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -88,9 +88,17 @@ typedef u64 gen8_pte_t;
 #define BYT_PTE_SNOOPED_BY_CPU_CACHES	REG_BIT(2)
 #define BYT_PTE_WRITEABLE		REG_BIT(1)
 
+#define MTL_PPGTT_PTE_PAT3	BIT_ULL(62)
 #define GEN12_PPGTT_PTE_LM	BIT_ULL(11)
-
-#define GEN12_GGTT_PTE_LM	BIT_ULL(1)
+#define GEN12_PPGTT_PTE_PAT2	BIT_ULL(7)
+#define GEN12_PPGTT_PTE_PAT1	BIT_ULL(4)
+#define GEN12_PPGTT_PTE_PAT0	BIT_ULL(3)
+
+#define GEN12_GGTT_PTE_LM		BIT_ULL(1)
+#define MTL_GGTT_PTE_PAT0		BIT_ULL(52)
+#define MTL_GGTT_PTE_PAT1		BIT_ULL(53)
+#define GEN12_GGTT_PTE_ADDR_MASK	GENMASK_ULL(45, 12)
+#define MTL_GGTT_PTE_PAT_MASK		GENMASK_ULL(53, 52)
 
 #define GEN12_PDE_64K BIT(6)
 #define GEN12_PTE_PS64 BIT(8)
-- 
GitLab


From a161b6dba6e0e8132921a6b948d311cdf67f3476 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Mon, 24 Apr 2023 11:29:02 -0700
Subject: [PATCH 0298/1791] drm/i915/mtl: workaround coherency issue for Media

This patch implements Wa_22016122933.

In MTL, memory writes initiated by the Media tile update the whole
cache line, even for partial writes. This creates a coherency
problem for cacheable memory if both CPU and GPU are writing data
to different locations within a single cache line.
This patch circumvents the issue by making CPU/GPU shared memory
uncacheable (WC on CPU side, and PAT index 2 for GPU).  Additionally,
it ensures that CPU writes are visible to the GPU with an
intel_guc_write_barrier().

While fixing the CTB issue, we noticed some random GSC firmware
loading failure because the share buffers are cacheable (WB) on CPU
side but uncached on GPU side. To fix these issues we need to map
such shared buffers as WC on CPU side. Since such allocations are
not all done through GuC allocator, to avoid too many code changes,
the i915_coherent_map_type() is now hard coded to return WC for MTL.

v2: Simplify the commit message(Matt).

BSpec: 45101

Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424182902.3663500-3-fei.yang@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_pages.c |  5 ++++-
 drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c | 13 +++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_guc.c    |  7 +++++++
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c |  6 ++++++
 4 files changed, 30 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index ecd86130b74f1..89fc8ea6bcfc4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -469,7 +469,10 @@ enum i915_map_type i915_coherent_map_type(struct drm_i915_private *i915,
 					  struct drm_i915_gem_object *obj,
 					  bool always_coherent)
 {
-	if (i915_gem_object_is_lmem(obj))
+	/*
+	 * Wa_22016122933: always return I915_MAP_WC for MTL
+	 */
+	if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(i915))
 		return I915_MAP_WC;
 	if (HAS_LLC(i915) || always_coherent)
 		return I915_MAP_WB;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
index 1d9fdfb112681..236673c02f9ad 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
@@ -110,6 +110,13 @@ static int gsc_fw_load_prepare(struct intel_gsc_uc *gsc)
 	if (obj->base.size < gsc->fw.size)
 		return -ENOSPC;
 
+	/*
+	 * Wa_22016122933: For MTL the shared memory needs to be mapped
+	 * as WC on CPU side and UC (PAT index 2) on GPU side
+	 */
+	if (IS_METEORLAKE(i915))
+		i915_gem_object_set_cache_coherency(obj, I915_CACHE_NONE);
+
 	dst = i915_gem_object_pin_map_unlocked(obj,
 					       i915_coherent_map_type(i915, obj, true));
 	if (IS_ERR(dst))
@@ -125,6 +132,12 @@ static int gsc_fw_load_prepare(struct intel_gsc_uc *gsc)
 	memset(dst, 0, obj->base.size);
 	memcpy(dst, src, gsc->fw.size);
 
+	/*
+	 * Wa_22016122933: Making sure the data in dst is
+	 * visible to GSC right away
+	 */
+	intel_guc_write_barrier(&gt->uc.guc);
+
 	i915_gem_object_unpin_map(gsc->fw.obj);
 	i915_gem_object_unpin_map(obj);
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
index d76508fa3af7f..f9bddaa876d9b 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
@@ -743,6 +743,13 @@ struct i915_vma *intel_guc_allocate_vma(struct intel_guc *guc, u32 size)
 	if (IS_ERR(obj))
 		return ERR_CAST(obj);
 
+	/*
+	 * Wa_22016122933: For MTL the shared memory needs to be mapped
+	 * as WC on CPU side and UC (PAT index 2) on GPU side
+	 */
+	if (IS_METEORLAKE(gt->i915))
+		i915_gem_object_set_cache_coherency(obj, I915_CACHE_NONE);
+
 	vma = i915_vma_instance(obj, &gt->ggtt->vm, NULL);
 	if (IS_ERR(vma))
 		goto err;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 1803a633ed648..99a0a89091e7f 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -902,6 +902,12 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
 	/* now update descriptor */
 	WRITE_ONCE(desc->head, head);
 
+	/*
+	 * Wa_22016122933: Making sure the head update is
+	 * visible to GuC right away
+	 */
+	intel_guc_write_barrier(ct_to_guc(ct));
+
 	return available - len;
 
 corrupted:
-- 
GitLab


From 5a46e490e47e4d821c382288108f193fc88a9282 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <maxime@cerno.tech>
Date: Mon, 6 Mar 2023 11:46:42 +0100
Subject: [PATCH 0299/1791] drm/vc4: Switch to container_of_const

container_of_const() allows to preserve the pointer constness and is
thus more flexible than inline functions.

Let's switch all our instances of container_of() to
container_of_const().

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-1-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/tests/vc4_mock.h        |  3 +
 drivers/gpu/drm/vc4/tests/vc4_mock_output.c |  4 +-
 drivers/gpu/drm/vc4/vc4_dpi.c               |  7 +--
 drivers/gpu/drm/vc4/vc4_drv.h               | 65 ++++++---------------
 drivers/gpu/drm/vc4/vc4_dsi.c               | 17 ++----
 drivers/gpu/drm/vc4/vc4_hdmi.h              | 16 ++---
 drivers/gpu/drm/vc4/vc4_kms.c               | 16 ++---
 drivers/gpu/drm/vc4/vc4_plane.c             |  3 +-
 drivers/gpu/drm/vc4/vc4_txp.c               | 12 ++--
 drivers/gpu/drm/vc4/vc4_vec.c               | 14 ++---
 10 files changed, 51 insertions(+), 106 deletions(-)

diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock.h b/drivers/gpu/drm/vc4/tests/vc4_mock.h
index db8e9a141ef84..2d0b339bd9f39 100644
--- a/drivers/gpu/drm/vc4/tests/vc4_mock.h
+++ b/drivers/gpu/drm/vc4/tests/vc4_mock.h
@@ -43,6 +43,9 @@ struct vc4_dummy_output {
 	struct drm_connector connector;
 };
 
+#define encoder_to_vc4_dummy_output(_enc)				\
+	container_of_const(_enc, struct vc4_dummy_output, encoder.base)
+
 struct vc4_dummy_output *vc4_dummy_output(struct kunit *test,
 					  struct drm_device *drm,
 					  struct drm_crtc *crtc,
diff --git a/drivers/gpu/drm/vc4/tests/vc4_mock_output.c b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
index 8d33be828d9a1..6e11fcc9ef45e 100644
--- a/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
+++ b/drivers/gpu/drm/vc4/tests/vc4_mock_output.c
@@ -80,7 +80,7 @@ int vc4_mock_atomic_add_output(struct kunit *test,
 	crtc = vc4_find_crtc_for_encoder(test, drm, encoder);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc);
 
-	output = container_of(encoder, struct vc4_dummy_output, encoder.base);
+	output = encoder_to_vc4_dummy_output(encoder);
 	conn = &output->connector;
 	conn_state = drm_atomic_get_connector_state(state, conn);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state);
@@ -126,7 +126,7 @@ int vc4_mock_atomic_del_output(struct kunit *test,
 	ret = drm_atomic_set_mode_for_crtc(crtc_state, NULL);
 	KUNIT_ASSERT_EQ(test, ret, 0);
 
-	output = container_of(encoder, struct vc4_dummy_output, encoder.base);
+	output = encoder_to_vc4_dummy_output(encoder);
 	conn = &output->connector;
 	conn_state = drm_atomic_get_connector_state(state, conn);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state);
diff --git a/drivers/gpu/drm/vc4/vc4_dpi.c b/drivers/gpu/drm/vc4/vc4_dpi.c
index f518d6e59ed69..e68c07d86040d 100644
--- a/drivers/gpu/drm/vc4/vc4_dpi.c
+++ b/drivers/gpu/drm/vc4/vc4_dpi.c
@@ -97,11 +97,8 @@ struct vc4_dpi {
 	struct debugfs_regset32 regset;
 };
 
-static inline struct vc4_dpi *
-to_vc4_dpi(struct drm_encoder *encoder)
-{
-	return container_of(encoder, struct vc4_dpi, encoder.base);
-}
+#define to_vc4_dpi(_encoder)						\
+	container_of_const(_encoder, struct vc4_dpi, encoder.base)
 
 #define DPI_READ(offset)								\
 	({										\
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 8768566c610b3..bf66499765fbb 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -232,11 +232,8 @@ struct vc4_dev {
 	struct kref bin_bo_kref;
 };
 
-static inline struct vc4_dev *
-to_vc4_dev(const struct drm_device *dev)
-{
-	return container_of(dev, struct vc4_dev, base);
-}
+#define to_vc4_dev(_dev)			\
+	container_of_const(_dev, struct vc4_dev, base)
 
 struct vc4_bo {
 	struct drm_gem_dma_object base;
@@ -285,11 +282,8 @@ struct vc4_bo {
 	struct mutex madv_lock;
 };
 
-static inline struct vc4_bo *
-to_vc4_bo(const struct drm_gem_object *bo)
-{
-	return container_of(to_drm_gem_dma_obj(bo), struct vc4_bo, base);
-}
+#define to_vc4_bo(_bo)							\
+	container_of_const(to_drm_gem_dma_obj(_bo), struct vc4_bo, base)
 
 struct vc4_fence {
 	struct dma_fence base;
@@ -298,11 +292,8 @@ struct vc4_fence {
 	uint64_t seqno;
 };
 
-static inline struct vc4_fence *
-to_vc4_fence(const struct dma_fence *fence)
-{
-	return container_of(fence, struct vc4_fence, base);
-}
+#define to_vc4_fence(_fence)					\
+	container_of_const(_fence, struct vc4_fence, base)
 
 struct vc4_seqno_cb {
 	struct work_struct work;
@@ -368,11 +359,8 @@ struct vc4_hvs_state {
 	} fifo_state[HVS_NUM_CHANNELS];
 };
 
-static inline struct vc4_hvs_state *
-to_vc4_hvs_state(const struct drm_private_state *priv)
-{
-	return container_of(priv, struct vc4_hvs_state, base);
-}
+#define to_vc4_hvs_state(_state)				\
+	container_of_const(_state, struct vc4_hvs_state, base)
 
 struct vc4_hvs_state *vc4_hvs_get_global_state(struct drm_atomic_state *state);
 struct vc4_hvs_state *vc4_hvs_get_old_global_state(const struct drm_atomic_state *state);
@@ -382,11 +370,8 @@ struct vc4_plane {
 	struct drm_plane base;
 };
 
-static inline struct vc4_plane *
-to_vc4_plane(const struct drm_plane *plane)
-{
-	return container_of(plane, struct vc4_plane, base);
-}
+#define to_vc4_plane(_plane)					\
+	container_of_const(_plane, struct vc4_plane, base)
 
 enum vc4_scaling_mode {
 	VC4_SCALING_NONE,
@@ -458,11 +443,8 @@ struct vc4_plane_state {
 	u64 membus_load;
 };
 
-static inline struct vc4_plane_state *
-to_vc4_plane_state(const struct drm_plane_state *state)
-{
-	return container_of(state, struct vc4_plane_state, base);
-}
+#define to_vc4_plane_state(_state)				\
+	container_of_const(_state, struct vc4_plane_state, base)
 
 enum vc4_encoder_type {
 	VC4_ENCODER_TYPE_NONE,
@@ -489,11 +471,8 @@ struct vc4_encoder {
 	void (*post_crtc_powerdown)(struct drm_encoder *encoder, struct drm_atomic_state *state);
 };
 
-static inline struct vc4_encoder *
-to_vc4_encoder(const struct drm_encoder *encoder)
-{
-	return container_of(encoder, struct vc4_encoder, base);
-}
+#define to_vc4_encoder(_encoder)				\
+	container_of_const(_encoder, struct vc4_encoder, base)
 
 static inline
 struct drm_encoder *vc4_find_encoder_by_type(struct drm_device *drm,
@@ -591,11 +570,8 @@ struct vc4_crtc {
 	unsigned int current_hvs_channel;
 };
 
-static inline struct vc4_crtc *
-to_vc4_crtc(const struct drm_crtc *crtc)
-{
-	return container_of(crtc, struct vc4_crtc, base);
-}
+#define to_vc4_crtc(_crtc)					\
+	container_of_const(_crtc, struct vc4_crtc, base)
 
 static inline const struct vc4_crtc_data *
 vc4_crtc_to_vc4_crtc_data(const struct vc4_crtc *crtc)
@@ -608,7 +584,7 @@ vc4_crtc_to_vc4_pv_data(const struct vc4_crtc *crtc)
 {
 	const struct vc4_crtc_data *data = vc4_crtc_to_vc4_crtc_data(crtc);
 
-	return container_of(data, struct vc4_pv_data, base);
+	return container_of_const(data, struct vc4_pv_data, base);
 }
 
 struct drm_encoder *vc4_get_crtc_encoder(struct drm_crtc *crtc,
@@ -636,11 +612,8 @@ struct vc4_crtc_state {
 
 #define VC4_HVS_CHANNEL_DISABLED ((unsigned int)-1)
 
-static inline struct vc4_crtc_state *
-to_vc4_crtc_state(const struct drm_crtc_state *crtc_state)
-{
-	return container_of(crtc_state, struct vc4_crtc_state, base);
-}
+#define to_vc4_crtc_state(_state)				\
+	container_of_const(_state, struct vc4_crtc_state, base)
 
 #define V3D_READ(offset)								\
 	({										\
diff --git a/drivers/gpu/drm/vc4/vc4_dsi.c b/drivers/gpu/drm/vc4/vc4_dsi.c
index a5c075f802e42..9e0c355b236f5 100644
--- a/drivers/gpu/drm/vc4/vc4_dsi.c
+++ b/drivers/gpu/drm/vc4/vc4_dsi.c
@@ -600,19 +600,14 @@ struct vc4_dsi {
 	struct debugfs_regset32 regset;
 };
 
-#define host_to_dsi(host) container_of(host, struct vc4_dsi, dsi_host)
+#define host_to_dsi(host)					\
+	container_of_const(host, struct vc4_dsi, dsi_host)
 
-static inline struct vc4_dsi *
-to_vc4_dsi(struct drm_encoder *encoder)
-{
-	return container_of(encoder, struct vc4_dsi, encoder.base);
-}
+#define to_vc4_dsi(_encoder)					\
+	container_of_const(_encoder, struct vc4_dsi, encoder.base)
 
-static inline struct vc4_dsi *
-bridge_to_vc4_dsi(struct drm_bridge *bridge)
-{
-	return container_of(bridge, struct vc4_dsi, bridge);
-}
+#define bridge_to_vc4_dsi(_bridge)				\
+	container_of_const(_bridge, struct vc4_dsi, bridge)
 
 static inline void
 dsi_dma_workaround_write(struct vc4_dsi *dsi, u32 offset, u32 val)
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
index e3619836ca17a..b57ffbca5721a 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.h
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
@@ -222,17 +222,14 @@ struct vc4_hdmi {
 	enum vc4_hdmi_output_format output_format;
 };
 
-static inline struct vc4_hdmi *
-connector_to_vc4_hdmi(struct drm_connector *connector)
-{
-	return container_of(connector, struct vc4_hdmi, connector);
-}
+#define connector_to_vc4_hdmi(_connector)				\
+	container_of_const(_connector, struct vc4_hdmi, connector)
 
 static inline struct vc4_hdmi *
 encoder_to_vc4_hdmi(struct drm_encoder *encoder)
 {
 	struct vc4_encoder *_encoder = to_vc4_encoder(encoder);
-	return container_of(_encoder, struct vc4_hdmi, encoder);
+	return container_of_const(_encoder, struct vc4_hdmi, encoder);
 }
 
 struct vc4_hdmi_connector_state {
@@ -242,11 +239,8 @@ struct vc4_hdmi_connector_state {
 	enum vc4_hdmi_output_format	output_format;
 };
 
-static inline struct vc4_hdmi_connector_state *
-conn_state_to_vc4_hdmi_conn_state(struct drm_connector_state *conn_state)
-{
-	return container_of(conn_state, struct vc4_hdmi_connector_state, base);
-}
+#define conn_state_to_vc4_hdmi_conn_state(_state)			\
+	container_of_const(_state, struct vc4_hdmi_connector_state, base)
 
 void vc4_hdmi_phy_init(struct vc4_hdmi *vc4_hdmi,
 		       struct vc4_hdmi_connector_state *vc4_conn_state);
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index a7e3d47c50f4a..5495f2a94fa92 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -31,11 +31,8 @@ struct vc4_ctm_state {
 	int fifo;
 };
 
-static struct vc4_ctm_state *
-to_vc4_ctm_state(const struct drm_private_state *priv)
-{
-	return container_of(priv, struct vc4_ctm_state, base);
-}
+#define to_vc4_ctm_state(_state)				\
+	container_of_const(_state, struct vc4_ctm_state, base)
 
 struct vc4_load_tracker_state {
 	struct drm_private_state base;
@@ -43,11 +40,8 @@ struct vc4_load_tracker_state {
 	u64 membus_load;
 };
 
-static struct vc4_load_tracker_state *
-to_vc4_load_tracker_state(const struct drm_private_state *priv)
-{
-	return container_of(priv, struct vc4_load_tracker_state, base);
-}
+#define to_vc4_load_tracker_state(_state)				\
+	container_of_const(_state, struct vc4_load_tracker_state, base)
 
 static struct vc4_ctm_state *vc4_get_ctm_state(struct drm_atomic_state *state,
 					       struct drm_private_obj *manager)
@@ -717,7 +711,7 @@ static void vc4_hvs_channels_destroy_state(struct drm_private_obj *obj,
 static void vc4_hvs_channels_print_state(struct drm_printer *p,
 					 const struct drm_private_state *state)
 {
-	struct vc4_hvs_state *hvs_state = to_vc4_hvs_state(state);
+	const struct vc4_hvs_state *hvs_state = to_vc4_hvs_state(state);
 	unsigned int i;
 
 	drm_printf(p, "HVS State\n");
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
index 97c84a3f5a461..00e713faecd5a 100644
--- a/drivers/gpu/drm/vc4/vc4_plane.c
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -1334,8 +1334,7 @@ u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist)
 
 u32 vc4_plane_dlist_size(const struct drm_plane_state *state)
 {
-	const struct vc4_plane_state *vc4_state =
-		container_of(state, typeof(*vc4_state), base);
+	const struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
 
 	return vc4_state->dlist_count;
 }
diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c
index ef5cab2a3aa97..c5abdec031039 100644
--- a/drivers/gpu/drm/vc4/vc4_txp.c
+++ b/drivers/gpu/drm/vc4/vc4_txp.c
@@ -168,15 +168,11 @@ struct vc4_txp {
 	void __iomem *regs;
 };
 
-static inline struct vc4_txp *encoder_to_vc4_txp(struct drm_encoder *encoder)
-{
-	return container_of(encoder, struct vc4_txp, encoder.base);
-}
+#define encoder_to_vc4_txp(_encoder)					\
+	container_of_const(_encoder, struct vc4_txp, encoder.base)
 
-static inline struct vc4_txp *connector_to_vc4_txp(struct drm_connector *conn)
-{
-	return container_of(conn, struct vc4_txp, connector.base);
-}
+#define connector_to_vc4_txp(_connector)				\
+	container_of_const(_connector, struct vc4_txp, connector.base)
 
 static const struct debugfs_reg32 txp_regs[] = {
 	VC4_REG32(TXP_DST_PTR),
diff --git a/drivers/gpu/drm/vc4/vc4_vec.c b/drivers/gpu/drm/vc4/vc4_vec.c
index a3782d05cd669..d6e6a1a22eba7 100644
--- a/drivers/gpu/drm/vc4/vc4_vec.c
+++ b/drivers/gpu/drm/vc4/vc4_vec.c
@@ -219,17 +219,11 @@ struct vc4_vec {
 		writel(val, vec->regs + (offset));					\
 	} while (0)
 
-static inline struct vc4_vec *
-encoder_to_vc4_vec(struct drm_encoder *encoder)
-{
-	return container_of(encoder, struct vc4_vec, encoder.base);
-}
+#define encoder_to_vc4_vec(_encoder)					\
+	container_of_const(_encoder, struct vc4_vec, encoder.base)
 
-static inline struct vc4_vec *
-connector_to_vc4_vec(struct drm_connector *connector)
-{
-	return container_of(connector, struct vc4_vec, connector);
-}
+#define connector_to_vc4_vec(_connector)				\
+	container_of_const(_connector, struct vc4_vec, connector)
 
 enum vc4_vec_tv_mode_id {
 	VC4_VEC_TV_MODE_NTSC,
-- 
GitLab


From bb1b6094c099668e21b1b47c9327c01f5c395997 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <maxime@cerno.tech>
Date: Mon, 6 Mar 2023 11:46:43 +0100
Subject: [PATCH 0300/1791] drm/vc4: hdmi: Update all the planes if the TV
 margins are changed

On VC4, the TV margins on the HDMI connector are implemented by scaling
the planes.

However, if only the TV margins or the connector are changed by a new
state, the planes ending up on that connector won't be. Thus, they won't
be updated properly and we'll effectively ignore that change until the
next commit affecting these planes.

Let's make sure to add all the planes attached to the connector so that
we can update them properly.

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-2-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 06713d8b82b5d..dd1ad03de4e08 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -535,6 +535,32 @@ static int vc4_hdmi_connector_atomic_check(struct drm_connector *connector,
 	if (!crtc)
 		return 0;
 
+	if (old_state->tv.margins.left != new_state->tv.margins.left ||
+	    old_state->tv.margins.right != new_state->tv.margins.right ||
+	    old_state->tv.margins.top != new_state->tv.margins.top ||
+	    old_state->tv.margins.bottom != new_state->tv.margins.bottom) {
+		struct drm_crtc_state *crtc_state;
+		int ret;
+
+		crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+
+		/*
+		 * Strictly speaking, we should be calling
+		 * drm_atomic_helper_check_planes() after our call to
+		 * drm_atomic_add_affected_planes(). However, the
+		 * connector atomic_check is called as part of
+		 * drm_atomic_helper_check_modeset() that already
+		 * happens before a call to
+		 * drm_atomic_helper_check_planes() in
+		 * drm_atomic_helper_check().
+		 */
+		ret = drm_atomic_add_affected_planes(state, crtc);
+		if (ret)
+			return ret;
+	}
+
 	if (old_state->colorspace != new_state->colorspace ||
 	    !drm_connector_atomic_hdr_metadata_equal(old_state, new_state)) {
 		struct drm_crtc_state *crtc_state;
-- 
GitLab


From d6020f4b26179481c7cb13aa94d7abcdfd8a4326 Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:44 +0100
Subject: [PATCH 0301/1791] drm/vc4: hdmi: Add Broadcast RGB property to allow
 override of RGB range

Copy Intel's "Broadcast RGB" property semantics to add manual override
of the HDMI pixel range for monitors that don't abide by the content
of the AVI Infoframe.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Hans Verkuil <hverkuil-cisco at xs4all.nl>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-3-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 97 ++++++++++++++++++++++++++++++++--
 drivers/gpu/drm/vc4/vc4_hdmi.h |  9 ++++
 2 files changed, 102 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index dd1ad03de4e08..fd265388b5068 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -154,10 +154,16 @@ static bool vc4_hdmi_mode_needs_scrambling(const struct drm_display_mode *mode,
 }
 
 static bool vc4_hdmi_is_full_range_rgb(struct vc4_hdmi *vc4_hdmi,
-				       const struct drm_display_mode *mode)
+				       struct vc4_hdmi_connector_state *vc4_state)
 {
+	const struct drm_display_mode *mode = &vc4_hdmi->saved_adjusted_mode;
 	struct drm_display_info *display = &vc4_hdmi->connector.display_info;
 
+	if (vc4_state->broadcast_rgb == VC4_HDMI_BROADCAST_RGB_LIMITED)
+		return false;
+	else if (vc4_state->broadcast_rgb == VC4_HDMI_BROADCAST_RGB_FULL)
+		return true;
+
 	return !display->is_hdmi ||
 		drm_default_rgb_quant_range(mode) == HDMI_QUANTIZATION_RANGE_FULL;
 }
@@ -528,8 +534,12 @@ static int vc4_hdmi_connector_atomic_check(struct drm_connector *connector,
 {
 	struct drm_connector_state *old_state =
 		drm_atomic_get_old_connector_state(state, connector);
+	struct vc4_hdmi_connector_state *old_vc4_state =
+		conn_state_to_vc4_hdmi_conn_state(old_state);
 	struct drm_connector_state *new_state =
 		drm_atomic_get_new_connector_state(state, connector);
+	struct vc4_hdmi_connector_state *new_vc4_state =
+		conn_state_to_vc4_hdmi_conn_state(new_state);
 	struct drm_crtc *crtc = new_state->crtc;
 
 	if (!crtc)
@@ -562,6 +572,7 @@ static int vc4_hdmi_connector_atomic_check(struct drm_connector *connector,
 	}
 
 	if (old_state->colorspace != new_state->colorspace ||
+	    old_vc4_state->broadcast_rgb != new_vc4_state->broadcast_rgb ||
 	    !drm_connector_atomic_hdr_metadata_equal(old_state, new_state)) {
 		struct drm_crtc_state *crtc_state;
 
@@ -575,6 +586,49 @@ static int vc4_hdmi_connector_atomic_check(struct drm_connector *connector,
 	return 0;
 }
 
+static int vc4_hdmi_connector_get_property(struct drm_connector *connector,
+					   const struct drm_connector_state *state,
+					   struct drm_property *property,
+					   uint64_t *val)
+{
+	struct drm_device *drm = connector->dev;
+	struct vc4_hdmi *vc4_hdmi =
+		connector_to_vc4_hdmi(connector);
+	const struct vc4_hdmi_connector_state *vc4_conn_state =
+		conn_state_to_vc4_hdmi_conn_state(state);
+
+	if (property == vc4_hdmi->broadcast_rgb_property) {
+		*val = vc4_conn_state->broadcast_rgb;
+	} else {
+		drm_dbg(drm, "Unknown property [PROP:%d:%s]\n",
+			property->base.id, property->name);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int vc4_hdmi_connector_set_property(struct drm_connector *connector,
+					   struct drm_connector_state *state,
+					   struct drm_property *property,
+					   uint64_t val)
+{
+	struct drm_device *drm = connector->dev;
+	struct vc4_hdmi *vc4_hdmi =
+		connector_to_vc4_hdmi(connector);
+	struct vc4_hdmi_connector_state *vc4_conn_state =
+		conn_state_to_vc4_hdmi_conn_state(state);
+
+	if (property == vc4_hdmi->broadcast_rgb_property) {
+		vc4_conn_state->broadcast_rgb = val;
+		return 0;
+	}
+
+	drm_dbg(drm, "Unknown property [PROP:%d:%s]\n",
+		property->base.id, property->name);
+	return -EINVAL;
+}
+
 static void vc4_hdmi_connector_reset(struct drm_connector *connector)
 {
 	struct vc4_hdmi_connector_state *old_state =
@@ -594,6 +648,7 @@ static void vc4_hdmi_connector_reset(struct drm_connector *connector)
 	new_state->base.max_bpc = 8;
 	new_state->base.max_requested_bpc = 8;
 	new_state->output_format = VC4_HDMI_OUTPUT_RGB;
+	new_state->broadcast_rgb = VC4_HDMI_BROADCAST_RGB_AUTO;
 	drm_atomic_helper_connector_tv_margins_reset(connector);
 }
 
@@ -611,6 +666,7 @@ vc4_hdmi_connector_duplicate_state(struct drm_connector *connector)
 	new_state->tmds_char_rate = vc4_state->tmds_char_rate;
 	new_state->output_bpc = vc4_state->output_bpc;
 	new_state->output_format = vc4_state->output_format;
+	new_state->broadcast_rgb = vc4_state->broadcast_rgb;
 	__drm_atomic_helper_connector_duplicate_state(connector, &new_state->base);
 
 	return &new_state->base;
@@ -621,6 +677,8 @@ static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
 	.reset = vc4_hdmi_connector_reset,
 	.atomic_duplicate_state = vc4_hdmi_connector_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+	.atomic_get_property = vc4_hdmi_connector_get_property,
+	.atomic_set_property = vc4_hdmi_connector_set_property,
 };
 
 static const struct drm_connector_helper_funcs vc4_hdmi_connector_helper_funcs = {
@@ -629,6 +687,33 @@ static const struct drm_connector_helper_funcs vc4_hdmi_connector_helper_funcs =
 	.atomic_check = vc4_hdmi_connector_atomic_check,
 };
 
+static const struct drm_prop_enum_list broadcast_rgb_names[] = {
+	{ VC4_HDMI_BROADCAST_RGB_AUTO, "Automatic" },
+	{ VC4_HDMI_BROADCAST_RGB_FULL, "Full" },
+	{ VC4_HDMI_BROADCAST_RGB_LIMITED, "Limited 16:235" },
+};
+
+static void
+vc4_hdmi_attach_broadcast_rgb_property(struct drm_device *dev,
+				       struct vc4_hdmi *vc4_hdmi)
+{
+	struct drm_property *prop = vc4_hdmi->broadcast_rgb_property;
+
+	if (!prop) {
+		prop = drm_property_create_enum(dev, DRM_MODE_PROP_ENUM,
+						"Broadcast RGB",
+						broadcast_rgb_names,
+						ARRAY_SIZE(broadcast_rgb_names));
+		if (!prop)
+			return;
+
+		vc4_hdmi->broadcast_rgb_property = prop;
+	}
+
+	drm_object_attach_property(&vc4_hdmi->connector.base, prop,
+				   VC4_HDMI_BROADCAST_RGB_AUTO);
+}
+
 static int vc4_hdmi_connector_init(struct drm_device *dev,
 				   struct vc4_hdmi *vc4_hdmi)
 {
@@ -675,6 +760,8 @@ static int vc4_hdmi_connector_init(struct drm_device *dev,
 	if (vc4_hdmi->variant->supports_hdr)
 		drm_connector_attach_hdr_output_metadata_property(connector);
 
+	vc4_hdmi_attach_broadcast_rgb_property(dev, vc4_hdmi);
+
 	drm_connector_attach_encoder(connector, encoder);
 
 	return 0;
@@ -829,7 +916,7 @@ static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder)
 
 	drm_hdmi_avi_infoframe_quant_range(&frame.avi,
 					   connector, mode,
-					   vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode) ?
+					   vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state) ?
 					   HDMI_QUANTIZATION_RANGE_FULL :
 					   HDMI_QUANTIZATION_RANGE_LIMITED);
 	drm_hdmi_avi_infoframe_colorimetry(&frame.avi, cstate);
@@ -1072,6 +1159,8 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 			       struct drm_connector_state *state,
 			       const struct drm_display_mode *mode)
 {
+	struct vc4_hdmi_connector_state *vc4_state =
+		conn_state_to_vc4_hdmi_conn_state(state);
 	struct drm_device *drm = vc4_hdmi->connector.dev;
 	unsigned long flags;
 	u32 csc_ctl;
@@ -1085,7 +1174,7 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR,
 				VC4_HD_CSC_CTL_ORDER);
 
-	if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode)) {
+	if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state)) {
 		/* CEA VICs other than #1 requre limited range RGB
 		 * output unless overridden by an AVI infoframe.
 		 * Apply a colorspace conversion to squash 0-255 down
@@ -1238,7 +1327,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	case VC4_HDMI_OUTPUT_RGB:
 		if_xbar = 0x354021;
 
-		if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode))
+		if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state))
 			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_rgb);
 		else
 			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_unity);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h
index b57ffbca5721a..934d5d61485a5 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.h
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.h
@@ -117,6 +117,12 @@ enum vc4_hdmi_output_format {
 	VC4_HDMI_OUTPUT_YUV420,
 };
 
+enum vc4_hdmi_broadcast_rgb {
+	VC4_HDMI_BROADCAST_RGB_AUTO,
+	VC4_HDMI_BROADCAST_RGB_FULL,
+	VC4_HDMI_BROADCAST_RGB_LIMITED,
+};
+
 /* General HDMI hardware state. */
 struct vc4_hdmi {
 	struct vc4_hdmi_audio audio;
@@ -129,6 +135,8 @@ struct vc4_hdmi {
 
 	struct delayed_work scrambling_work;
 
+	struct drm_property *broadcast_rgb_property;
+
 	struct i2c_adapter *ddc;
 	void __iomem *hdmicore_regs;
 	void __iomem *hd_regs;
@@ -237,6 +245,7 @@ struct vc4_hdmi_connector_state {
 	unsigned long long		tmds_char_rate;
 	unsigned int 			output_bpc;
 	enum vc4_hdmi_output_format	output_format;
+	enum vc4_hdmi_broadcast_rgb	broadcast_rgb;
 };
 
 #define conn_state_to_vc4_hdmi_conn_state(_state)			\
-- 
GitLab


From 7186004485c9b737faae5aad6b8118dd0bc0face Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:45 +0100
Subject: [PATCH 0302/1791] drm/vc4: hdmi: Rename full range helper

The VC4 HDMI driver has a helper function to figure out whether full
range or limited range RGB is being used called
vc4_hdmi_is_full_range_rgb().

We'll need it to support other colorspaces, so let's rename it to
vc4_hdmi_is_full_range().

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-4-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index fd265388b5068..328e5294f7594 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -153,8 +153,8 @@ static bool vc4_hdmi_mode_needs_scrambling(const struct drm_display_mode *mode,
 	return clock > HDMI_14_MAX_TMDS_CLK;
 }
 
-static bool vc4_hdmi_is_full_range_rgb(struct vc4_hdmi *vc4_hdmi,
-				       struct vc4_hdmi_connector_state *vc4_state)
+static bool vc4_hdmi_is_full_range(struct vc4_hdmi *vc4_hdmi,
+				   struct vc4_hdmi_connector_state *vc4_state)
 {
 	const struct drm_display_mode *mode = &vc4_hdmi->saved_adjusted_mode;
 	struct drm_display_info *display = &vc4_hdmi->connector.display_info;
@@ -916,7 +916,7 @@ static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder)
 
 	drm_hdmi_avi_infoframe_quant_range(&frame.avi,
 					   connector, mode,
-					   vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state) ?
+					   vc4_hdmi_is_full_range(vc4_hdmi, vc4_state) ?
 					   HDMI_QUANTIZATION_RANGE_FULL :
 					   HDMI_QUANTIZATION_RANGE_LIMITED);
 	drm_hdmi_avi_infoframe_colorimetry(&frame.avi, cstate);
@@ -1174,7 +1174,7 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR,
 				VC4_HD_CSC_CTL_ORDER);
 
-	if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state)) {
+	if (!vc4_hdmi_is_full_range(vc4_hdmi, vc4_state)) {
 		/* CEA VICs other than #1 requre limited range RGB
 		 * output unless overridden by an AVI infoframe.
 		 * Apply a colorspace conversion to squash 0-255 down
@@ -1327,7 +1327,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	case VC4_HDMI_OUTPUT_RGB:
 		if_xbar = 0x354021;
 
-		if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, vc4_state))
+		if (!vc4_hdmi_is_full_range(vc4_hdmi, vc4_state))
 			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_rgb);
 		else
 			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_unity);
-- 
GitLab


From e3337aea097ffac21ccebd18114656eb88937517 Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:46 +0100
Subject: [PATCH 0303/1791] drm/vc4: hdmi: Swap CSC matrix channels for YUV444

YUV444 and YUV422 actually require the same matrix, but programmed
differently.

We've dealt with it in the past by having two matrices, with the one
for YUV444 reordered to accomodate the hardware.

This gets in the way of subsequent reworks so let's define a function
that will take the coefficients swap into account, and remove the now
redundant YUV444 matrix.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-5-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 39 +++++++++++++++++-----------------
 1 file changed, 19 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 328e5294f7594..d218b9dc465fb 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -1239,7 +1239,7 @@ static const u16 vc5_hdmi_csc_full_rgb_to_limited_rgb[3][4] = {
 };
 
 /*
- * Conversion between Full Range RGB and Full Range YUV422 using the
+ * Conversion between Full Range RGB and Limited Range YUV using the
  * BT.709 Colorspace
  *
  *
@@ -1249,28 +1249,12 @@ static const u16 vc5_hdmi_csc_full_rgb_to_limited_rgb[3][4] = {
  *
  * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
  */
-static const u16 vc5_hdmi_csc_full_rgb_to_limited_yuv422_bt709[3][4] = {
+static const u16 vc5_hdmi_csc_full_rgb_to_limited_bt709[3][4] = {
 	{ 0x05d2, 0x1394, 0x01fa, 0x0400 },
 	{ 0xfccc, 0xf536, 0x0e00, 0x2000 },
 	{ 0x0e00, 0xf34a, 0xfeb8, 0x2000 },
 };
 
-/*
- * Conversion between Full Range RGB and Full Range YUV444 using the
- * BT.709 Colorspace
- *
- * [ -0.100268 -0.337232  0.437500  128 ]
- * [  0.437500 -0.397386 -0.040114  128 ]
- * [  0.181906  0.611804  0.061758  16  ]
- *
- * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
- */
-static const u16 vc5_hdmi_csc_full_rgb_to_limited_yuv444_bt709[3][4] = {
-	{ 0xfccc, 0xf536, 0x0e00, 0x2000 },
-	{ 0x0e00, 0xf34a, 0xfeb8, 0x2000 },
-	{ 0x05d2, 0x1394, 0x01fa, 0x0400 },
-};
-
 static void vc5_hdmi_set_csc_coeffs(struct vc4_hdmi *vc4_hdmi,
 				    const u16 coeffs[3][4])
 {
@@ -1284,6 +1268,20 @@ static void vc5_hdmi_set_csc_coeffs(struct vc4_hdmi *vc4_hdmi,
 	HDMI_WRITE(HDMI_CSC_34_33, (coeffs[2][3] << 16) | coeffs[2][2]);
 }
 
+static void vc5_hdmi_set_csc_coeffs_swap(struct vc4_hdmi *vc4_hdmi,
+					 const u16 coeffs[3][4])
+{
+	lockdep_assert_held(&vc4_hdmi->hw_lock);
+
+	/* YUV444 needs the CSC matrices using the channels in a different order */
+	HDMI_WRITE(HDMI_CSC_12_11, (coeffs[1][1] << 16) | coeffs[1][0]);
+	HDMI_WRITE(HDMI_CSC_14_13, (coeffs[1][3] << 16) | coeffs[1][2]);
+	HDMI_WRITE(HDMI_CSC_22_21, (coeffs[2][1] << 16) | coeffs[2][0]);
+	HDMI_WRITE(HDMI_CSC_24_23, (coeffs[2][3] << 16) | coeffs[2][2]);
+	HDMI_WRITE(HDMI_CSC_32_31, (coeffs[0][1] << 16) | coeffs[0][0]);
+	HDMI_WRITE(HDMI_CSC_34_33, (coeffs[0][3] << 16) | coeffs[0][2]);
+}
+
 static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 			       struct drm_connector_state *state,
 			       const struct drm_display_mode *mode)
@@ -1306,7 +1304,8 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 
 	switch (vc4_state->output_format) {
 	case VC4_HDMI_OUTPUT_YUV444:
-		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_yuv444_bt709);
+		vc5_hdmi_set_csc_coeffs_swap(vc4_hdmi,
+					     vc5_hdmi_csc_full_rgb_to_limited_bt709);
 		break;
 
 	case VC4_HDMI_OUTPUT_YUV422:
@@ -1321,7 +1320,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 		if_cfg |= VC4_SET_FIELD(VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422_FORMAT_422_LEGACY,
 					VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422);
 
-		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_yuv422_bt709);
+		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_bt709);
 		break;
 
 	case VC4_HDMI_OUTPUT_RGB:
-- 
GitLab


From 1d239901770ef8c31ed4ab7c45d6e8f5715783b6 Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:47 +0100
Subject: [PATCH 0304/1791] drm/vc4: hdmi: Rework the CSC matrices organization

The CSC matrices were stored as separate matrix for each colorspace, and
if we wanted a limited or full RGB output.

This created some gaps in our support and we would not always pick the
relevant matrix.

Let's rework our data structure to store one per colorspace, and then a
matrix for limited range and one for full range. This makes us add a new
matrix to support full range BT709 YUV output.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-6-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 108 +++++++++++++++++++--------------
 1 file changed, 63 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index d218b9dc465fb..97ac87ac57769 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -1207,52 +1207,72 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 }
 
 /*
- * If we need to output Full Range RGB, then use the unity matrix
+ * Matrices for (internal) RGB to RGB output.
  *
- * [ 1      0      0      0]
- * [ 0      1      0      0]
- * [ 0      0      1      0]
- *
- * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
- */
-static const u16 vc5_hdmi_csc_full_rgb_unity[3][4] = {
-	{ 0x2000, 0x0000, 0x0000, 0x0000 },
-	{ 0x0000, 0x2000, 0x0000, 0x0000 },
-	{ 0x0000, 0x0000, 0x2000, 0x0000 },
-};
-
-/*
- * CEA VICs other than #1 require limited range RGB output unless
- * overridden by an AVI infoframe. Apply a colorspace conversion to
- * squash 0-255 down to 16-235. The matrix here is:
- *
- * [ 0.8594 0      0      16]
- * [ 0      0.8594 0      16]
- * [ 0      0      0.8594 16]
- *
- * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
+ * Matrices are signed 2p13 fixed point, with signed 9p6 offsets
  */
-static const u16 vc5_hdmi_csc_full_rgb_to_limited_rgb[3][4] = {
-	{ 0x1b80, 0x0000, 0x0000, 0x0400 },
-	{ 0x0000, 0x1b80, 0x0000, 0x0400 },
-	{ 0x0000, 0x0000, 0x1b80, 0x0400 },
+static const u16 vc5_hdmi_csc_full_rgb_to_rgb[2][3][4] = {
+	{
+		/*
+		 * Full range - unity
+		 *
+		 * [ 1      0      0      0]
+		 * [ 0      1      0      0]
+		 * [ 0      0      1      0]
+		 */
+		{ 0x2000, 0x0000, 0x0000, 0x0000 },
+		{ 0x0000, 0x2000, 0x0000, 0x0000 },
+		{ 0x0000, 0x0000, 0x2000, 0x0000 },
+	},
+	{
+		/*
+		 * Limited range
+		 *
+		 * CEA VICs other than #1 require limited range RGB
+		 * output unless overridden by an AVI infoframe. Apply a
+		 * colorspace conversion to squash 0-255 down to 16-235.
+		 * The matrix here is:
+		 *
+		 * [ 0.8594 0      0      16]
+		 * [ 0      0.8594 0      16]
+		 * [ 0      0      0.8594 16]
+		 */
+		{ 0x1b80, 0x0000, 0x0000, 0x0400 },
+		{ 0x0000, 0x1b80, 0x0000, 0x0400 },
+		{ 0x0000, 0x0000, 0x1b80, 0x0400 },
+	},
 };
 
 /*
- * Conversion between Full Range RGB and Limited Range YUV using the
- * BT.709 Colorspace
+ * Conversion between Full Range RGB and YUV using the BT.709 Colorspace
  *
- *
- * [  0.181906  0.611804  0.061758  16  ]
- * [ -0.100268 -0.337232  0.437500  128 ]
- * [  0.437500 -0.397386 -0.040114  128 ]
- *
- * Matrix is signed 2p13 fixed point, with signed 9p6 offsets
+ * Matrices are signed 2p13 fixed point, with signed 9p6 offsets
  */
-static const u16 vc5_hdmi_csc_full_rgb_to_limited_bt709[3][4] = {
-	{ 0x05d2, 0x1394, 0x01fa, 0x0400 },
-	{ 0xfccc, 0xf536, 0x0e00, 0x2000 },
-	{ 0x0e00, 0xf34a, 0xfeb8, 0x2000 },
+static const u16 vc5_hdmi_csc_full_rgb_to_yuv_bt709[2][3][4] = {
+	{
+		/*
+		 * Full Range
+		 *
+		 * [  0.212600  0.715200  0.072200  0   ]
+		 * [ -0.114572 -0.385428  0.500000  128 ]
+		 * [  0.500000 -0.454153 -0.045847  128 ]
+		 */
+		{ 0x06ce, 0x16e3, 0x024f, 0x0000 },
+		{ 0xfc56, 0xf3ac, 0x1000, 0x2000 },
+		{ 0x1000, 0xf179, 0xfe89, 0x2000 },
+	},
+	{
+		/*
+		 * Limited Range
+		 *
+		 * [  0.181906  0.611804  0.061758  16  ]
+		 * [ -0.100268 -0.337232  0.437500  128 ]
+		 * [  0.437500 -0.397386 -0.040114  128 ]
+		 */
+		{ 0x05d2, 0x1394, 0x01fa, 0x0400 },
+		{ 0xfccc, 0xf536, 0x0e00, 0x2000 },
+		{ 0x0e00, 0xf34a, 0xfeb8, 0x2000 },
+	},
 };
 
 static void vc5_hdmi_set_csc_coeffs(struct vc4_hdmi *vc4_hdmi,
@@ -1289,6 +1309,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	struct drm_device *drm = vc4_hdmi->connector.dev;
 	struct vc4_hdmi_connector_state *vc4_state =
 		conn_state_to_vc4_hdmi_conn_state(state);
+	unsigned int lim_range = vc4_hdmi_is_full_range(vc4_hdmi, vc4_state) ? 0 : 1;
 	unsigned long flags;
 	u32 if_cfg = 0;
 	u32 if_xbar = 0x543210;
@@ -1305,7 +1326,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 	switch (vc4_state->output_format) {
 	case VC4_HDMI_OUTPUT_YUV444:
 		vc5_hdmi_set_csc_coeffs_swap(vc4_hdmi,
-					     vc5_hdmi_csc_full_rgb_to_limited_bt709);
+					     vc5_hdmi_csc_full_rgb_to_yuv_bt709[lim_range]);
 		break;
 
 	case VC4_HDMI_OUTPUT_YUV422:
@@ -1320,16 +1341,13 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 		if_cfg |= VC4_SET_FIELD(VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422_FORMAT_422_LEGACY,
 					VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422);
 
-		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_bt709);
+		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_yuv_bt709[lim_range]);
 		break;
 
 	case VC4_HDMI_OUTPUT_RGB:
 		if_xbar = 0x354021;
 
-		if (!vc4_hdmi_is_full_range(vc4_hdmi, vc4_state))
-			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_rgb);
-		else
-			vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_unity);
+		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_rgb[lim_range]);
 		break;
 
 	default:
-- 
GitLab


From 1eaa6eedb8cf602476d3bf602e090fb26cbfaf1e Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:48 +0100
Subject: [PATCH 0305/1791] drm/vc4: hdmi: Add a function to retrieve the CSC
 matrix

The CSC matrix to use depends on the output format, its range and the
colorspace.

Since we're going to add more colorspaces, let's move the CSC matrix
retrieval to a function.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-7-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 24 +++++++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 97ac87ac57769..5c1c13025a93c 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -1302,6 +1302,20 @@ static void vc5_hdmi_set_csc_coeffs_swap(struct vc4_hdmi *vc4_hdmi,
 	HDMI_WRITE(HDMI_CSC_34_33, (coeffs[0][3] << 16) | coeffs[0][2]);
 }
 
+static const u16
+(*vc5_hdmi_find_yuv_csc_coeffs(struct vc4_hdmi *vc4_hdmi, u32 colorspace, bool limited))[4]
+{
+	switch (colorspace) {
+	default:
+	case DRM_MODE_COLORIMETRY_NO_DATA:
+	case DRM_MODE_COLORIMETRY_BT709_YCC:
+	case DRM_MODE_COLORIMETRY_XVYCC_709:
+	case DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED:
+	case DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT:
+		return vc5_hdmi_csc_full_rgb_to_yuv_bt709[limited];
+	}
+}
+
 static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 			       struct drm_connector_state *state,
 			       const struct drm_display_mode *mode)
@@ -1311,6 +1325,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 		conn_state_to_vc4_hdmi_conn_state(state);
 	unsigned int lim_range = vc4_hdmi_is_full_range(vc4_hdmi, vc4_state) ? 0 : 1;
 	unsigned long flags;
+	const u16 (*csc)[4];
 	u32 if_cfg = 0;
 	u32 if_xbar = 0x543210;
 	u32 csc_chan_ctl = 0;
@@ -1325,11 +1340,14 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 
 	switch (vc4_state->output_format) {
 	case VC4_HDMI_OUTPUT_YUV444:
-		vc5_hdmi_set_csc_coeffs_swap(vc4_hdmi,
-					     vc5_hdmi_csc_full_rgb_to_yuv_bt709[lim_range]);
+		csc = vc5_hdmi_find_yuv_csc_coeffs(vc4_hdmi, state->colorspace, !!lim_range);
+
+		vc5_hdmi_set_csc_coeffs_swap(vc4_hdmi, csc);
 		break;
 
 	case VC4_HDMI_OUTPUT_YUV422:
+		csc = vc5_hdmi_find_yuv_csc_coeffs(vc4_hdmi, state->colorspace, !!lim_range);
+
 		csc_ctl |= VC4_SET_FIELD(VC5_MT_CP_CSC_CTL_FILTER_MODE_444_TO_422_STANDARD,
 					 VC5_MT_CP_CSC_CTL_FILTER_MODE_444_TO_422) |
 			VC5_MT_CP_CSC_CTL_USE_444_TO_422 |
@@ -1341,7 +1359,7 @@ static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi,
 		if_cfg |= VC4_SET_FIELD(VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422_FORMAT_422_LEGACY,
 					VC5_DVP_HT_VEC_INTERFACE_CFG_SEL_422);
 
-		vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_yuv_bt709[lim_range]);
+		vc5_hdmi_set_csc_coeffs(vc4_hdmi, csc);
 		break;
 
 	case VC4_HDMI_OUTPUT_RGB:
-- 
GitLab


From 8297ee3f2d1277ddaa6de14c9520ddd9a1afa610 Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:49 +0100
Subject: [PATCH 0306/1791] drm/vc4: hdmi: Add BT.601 Support

Even though we report that we support the BT601 Colorspace, we were
always using the BT.709 conversion matrices. Let's add the BT601 ones.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-8-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 38 ++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 5c1c13025a93c..53c3d4f092960 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -1243,6 +1243,37 @@ static const u16 vc5_hdmi_csc_full_rgb_to_rgb[2][3][4] = {
 	},
 };
 
+/*
+ * Conversion between Full Range RGB and YUV using the BT.601 Colorspace
+ *
+ * Matrices are signed 2p13 fixed point, with signed 9p6 offsets
+ */
+static const u16 vc5_hdmi_csc_full_rgb_to_yuv_bt601[2][3][4] = {
+	{
+		/*
+		 * Full Range
+		 *
+		 * [  0.299000  0.587000  0.114000  0   ]
+		 * [ -0.168736 -0.331264  0.500000  128 ]
+		 * [  0.500000 -0.418688 -0.081312  128 ]
+		 */
+		{ 0x0991, 0x12c9, 0x03a6, 0x0000 },
+		{ 0xfa9b, 0xf567, 0x1000, 0x2000 },
+		{ 0x1000, 0xf29b, 0xfd67, 0x2000 },
+	},
+	{
+		/* Limited Range
+		 *
+		 * [  0.255785  0.502160  0.097523  16  ]
+		 * [ -0.147644 -0.289856  0.437500  128 ]
+		 * [  0.437500 -0.366352 -0.071148  128 ]
+		 */
+		{ 0x082f, 0x1012, 0x031f, 0x0400 },
+		{ 0xfb48, 0xf6ba, 0x0e00, 0x2000 },
+		{ 0x0e00, 0xf448, 0xfdba, 0x2000 },
+	},
+};
+
 /*
  * Conversion between Full Range RGB and YUV using the BT.709 Colorspace
  *
@@ -1306,6 +1337,13 @@ static const u16
 (*vc5_hdmi_find_yuv_csc_coeffs(struct vc4_hdmi *vc4_hdmi, u32 colorspace, bool limited))[4]
 {
 	switch (colorspace) {
+	case DRM_MODE_COLORIMETRY_SMPTE_170M_YCC:
+	case DRM_MODE_COLORIMETRY_XVYCC_601:
+	case DRM_MODE_COLORIMETRY_SYCC_601:
+	case DRM_MODE_COLORIMETRY_OPYCC_601:
+	case DRM_MODE_COLORIMETRY_BT601_YCC:
+		return vc5_hdmi_csc_full_rgb_to_yuv_bt601[limited];
+
 	default:
 	case DRM_MODE_COLORIMETRY_NO_DATA:
 	case DRM_MODE_COLORIMETRY_BT709_YCC:
-- 
GitLab


From b2751a4c5b5a7dcdfcea437f403c94a018cda5c9 Mon Sep 17 00:00:00 2001
From: Dave Stevenson <dave.stevenson@raspberrypi.com>
Date: Mon, 6 Mar 2023 11:46:50 +0100
Subject: [PATCH 0307/1791] drm/vc4: hdmi: Add BT.2020 Support

Even though we report that we support the BT.2020 Colorspace, we were
always using the BT.709 conversion matrices. Let's add the BT.2020 ones.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20221207-rpi-hdmi-improvements-v3-9-bdd54f66884e@cerno.tech
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 38 ++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 53c3d4f092960..6da41ea1250ab 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -1306,6 +1306,37 @@ static const u16 vc5_hdmi_csc_full_rgb_to_yuv_bt709[2][3][4] = {
 	},
 };
 
+/*
+ * Conversion between Full Range RGB and YUV using the BT.2020 Colorspace
+ *
+ * Matrices are signed 2p13 fixed point, with signed 9p6 offsets
+ */
+static const u16 vc5_hdmi_csc_full_rgb_to_yuv_bt2020[2][3][4] = {
+	{
+		/*
+		 * Full Range
+		 *
+		 * [  0.262700  0.678000  0.059300  0   ]
+		 * [ -0.139630 -0.360370  0.500000  128 ]
+		 * [  0.500000 -0.459786 -0.040214  128 ]
+		 */
+		{ 0x0868, 0x15b2, 0x01e6, 0x0000 },
+		{ 0xfb89, 0xf479, 0x1000, 0x2000 },
+		{ 0x1000, 0xf14a, 0xfeb8, 0x2000 },
+	},
+	{
+		/* Limited Range
+		 *
+		 * [  0.224732  0.580008  0.050729  16  ]
+		 * [ -0.122176 -0.315324  0.437500  128 ]
+		 * [  0.437500 -0.402312 -0.035188  128 ]
+		 */
+		{ 0x082f, 0x1012, 0x031f, 0x0400 },
+		{ 0xfb48, 0xf6ba, 0x0e00, 0x2000 },
+		{ 0x0e00, 0xf448, 0xfdba, 0x2000 },
+	},
+};
+
 static void vc5_hdmi_set_csc_coeffs(struct vc4_hdmi *vc4_hdmi,
 				    const u16 coeffs[3][4])
 {
@@ -1351,6 +1382,13 @@ static const u16
 	case DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED:
 	case DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT:
 		return vc5_hdmi_csc_full_rgb_to_yuv_bt709[limited];
+
+	case DRM_MODE_COLORIMETRY_BT2020_CYCC:
+	case DRM_MODE_COLORIMETRY_BT2020_YCC:
+	case DRM_MODE_COLORIMETRY_BT2020_RGB:
+	case DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65:
+	case DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER:
+		return vc5_hdmi_csc_full_rgb_to_yuv_bt2020[limited];
 	}
 }
 
-- 
GitLab


From cec5ccef85bd0128cf895612de54a9d21d2015d0 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 30 Mar 2023 11:59:41 +0200
Subject: [PATCH 0308/1791] drm/bridge: tc358768: Add
 atomic_get_input_bus_fmts() implementation

Add atomic_get_input_bus_fmts() implementation, tc358768 has a parallel
RGB input interface with the actual bus format depending on the amount
of parallel input data lines.

Without this change when the tc358768 is used with less than 24bit the
color mapping is completely wrong.

Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230330095941.428122-7-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 44 +++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 7c0cbe84611b9..a4ba93a629f0b 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -9,6 +9,7 @@
 #include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
 #include <linux/kernel.h>
+#include <linux/media-bus-format.h>
 #include <linux/module.h>
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
@@ -913,6 +914,44 @@ static void tc358768_bridge_enable(struct drm_bridge *bridge)
 	}
 }
 
+#define MAX_INPUT_SEL_FORMATS	1
+
+static u32 *
+tc358768_atomic_get_input_bus_fmts(struct drm_bridge *bridge,
+				   struct drm_bridge_state *bridge_state,
+				   struct drm_crtc_state *crtc_state,
+				   struct drm_connector_state *conn_state,
+				   u32 output_fmt,
+				   unsigned int *num_input_fmts)
+{
+	struct tc358768_priv *priv = bridge_to_tc358768(bridge);
+	u32 *input_fmts;
+
+	*num_input_fmts = 0;
+
+	input_fmts = kcalloc(MAX_INPUT_SEL_FORMATS, sizeof(*input_fmts),
+			     GFP_KERNEL);
+	if (!input_fmts)
+		return NULL;
+
+	switch (priv->pd_lines) {
+	case 16:
+		input_fmts[0] = MEDIA_BUS_FMT_RGB565_1X16;
+		break;
+	case 18:
+		input_fmts[0] = MEDIA_BUS_FMT_RGB666_1X18;
+		break;
+	default:
+	case 24:
+		input_fmts[0] = MEDIA_BUS_FMT_RGB888_1X24;
+		break;
+	};
+
+	*num_input_fmts = MAX_INPUT_SEL_FORMATS;
+
+	return input_fmts;
+}
+
 static const struct drm_bridge_funcs tc358768_bridge_funcs = {
 	.attach = tc358768_bridge_attach,
 	.mode_valid = tc358768_bridge_mode_valid,
@@ -920,6 +959,11 @@ static const struct drm_bridge_funcs tc358768_bridge_funcs = {
 	.enable = tc358768_bridge_enable,
 	.disable = tc358768_bridge_disable,
 	.post_disable = tc358768_bridge_post_disable,
+
+	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+	.atomic_get_input_bus_fmts = tc358768_atomic_get_input_bus_fmts,
 };
 
 static const struct drm_bridge_timings default_tc358768_timings = {
-- 
GitLab


From bca774c387548421efb5b533434b8408be0517b3 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 24 Apr 2023 23:02:05 +0300
Subject: [PATCH 0309/1791] drm/i915/adlp+: Disable DC5/6 states for TC port
 DDI/AUX and for combo port AUX

On ADLP+ Bspec allows DC5/6 to be enabled while power well 2 is enabled.
Since the AUX and DDI power wells (except for port A/B) are also backed
by power well 2, this would suggest that DC5/6 can be enabled while any
of these AUX or DDI port functionalities are used. As opposed to this
AUX transfers will time out on ADLP TypeC ports while DC6 is enabled.

Until the restriction for DC5/6 is clarified in Bspec let's assume that
the intention is to allow for using these power states while pipe A/B is
enabled, but only for combo ports which can be used with eDP outputs.
Similarly assume that AUX transaction initiated by the driver on any port
requires DC states to be disabled.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Fixes: 88c487938414 ("drm/i915: Use separate "DC off" power well for ADL-P and DG2")
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424200205.1732941-1-imre.deak@intel.com
---
 .../i915/display/intel_display_power_map.c    | 28 +++++++++++--------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index 100582f105905..ca448359a8226 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1251,22 +1251,11 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_a,
 	POWER_DOMAIN_PIPE_PANEL_FITTER_A,
 	POWER_DOMAIN_INIT);
 
-#define XELPD_PW_2_POWER_DOMAINS \
-	XELPD_PW_B_POWER_DOMAINS, \
-	XELPD_PW_C_POWER_DOMAINS, \
-	XELPD_PW_D_POWER_DOMAINS, \
-	POWER_DOMAIN_PORT_DDI_LANES_C, \
-	POWER_DOMAIN_PORT_DDI_LANES_D, \
-	POWER_DOMAIN_PORT_DDI_LANES_E, \
+#define XELPD_DC_OFF_PORT_POWER_DOMAINS \
 	POWER_DOMAIN_PORT_DDI_LANES_TC1, \
 	POWER_DOMAIN_PORT_DDI_LANES_TC2, \
 	POWER_DOMAIN_PORT_DDI_LANES_TC3, \
 	POWER_DOMAIN_PORT_DDI_LANES_TC4, \
-	POWER_DOMAIN_VGA, \
-	POWER_DOMAIN_AUDIO_PLAYBACK, \
-	POWER_DOMAIN_AUX_IO_C, \
-	POWER_DOMAIN_AUX_IO_D, \
-	POWER_DOMAIN_AUX_IO_E, \
 	POWER_DOMAIN_AUX_C, \
 	POWER_DOMAIN_AUX_D, \
 	POWER_DOMAIN_AUX_E, \
@@ -1279,6 +1268,20 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_a,
 	POWER_DOMAIN_AUX_TBT3, \
 	POWER_DOMAIN_AUX_TBT4
 
+#define XELPD_PW_2_POWER_DOMAINS \
+	XELPD_PW_B_POWER_DOMAINS, \
+	XELPD_PW_C_POWER_DOMAINS, \
+	XELPD_PW_D_POWER_DOMAINS, \
+	POWER_DOMAIN_PORT_DDI_LANES_C, \
+	POWER_DOMAIN_PORT_DDI_LANES_D, \
+	POWER_DOMAIN_PORT_DDI_LANES_E, \
+	POWER_DOMAIN_VGA, \
+	POWER_DOMAIN_AUDIO_PLAYBACK, \
+	POWER_DOMAIN_AUX_IO_C, \
+	POWER_DOMAIN_AUX_IO_D, \
+	POWER_DOMAIN_AUX_IO_E, \
+	XELPD_DC_OFF_PORT_POWER_DOMAINS
+
 I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_2,
 	XELPD_PW_2_POWER_DOMAINS,
 	POWER_DOMAIN_INIT);
@@ -1301,6 +1304,7 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_2,
  */
 
 I915_DECL_PW_DOMAINS(xelpd_pwdoms_dc_off,
+	XELPD_DC_OFF_PORT_POWER_DOMAINS,
 	XELPD_PW_C_POWER_DOMAINS,
 	XELPD_PW_D_POWER_DOMAINS,
 	POWER_DOMAIN_PORT_DSI,
-- 
GitLab


From e991b5244d89096943af7dcd506faecb57b9377d Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Mon, 24 Apr 2023 19:06:07 +0530
Subject: [PATCH 0310/1791] drm/i915/selftest: Record GT error for gt failure

igt_live_test has pr_err dumped in case of some
GT failures. It will be more informative regarding
GT if we use gt_err instead.

Cc: Andi Shyti <andi.shyti@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424133607.3736825-1-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/selftests/igt_live_test.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/selftests/igt_live_test.c b/drivers/gpu/drm/i915/selftests/igt_live_test.c
index 714b7afc490b3..4ddc6d902752a 100644
--- a/drivers/gpu/drm/i915/selftests/igt_live_test.c
+++ b/drivers/gpu/drm/i915/selftests/igt_live_test.c
@@ -6,6 +6,7 @@
 
 #include "i915_drv.h"
 #include "gt/intel_gt.h"
+#include "gt/intel_gt_print.h"
 
 #include "../i915_selftest.h"
 #include "igt_flush_test.h"
@@ -30,7 +31,7 @@ int igt_live_test_begin(struct igt_live_test *t,
 
 		err = intel_gt_wait_for_idle(gt, MAX_SCHEDULE_TIMEOUT);
 		if (err) {
-			pr_err("%s(%s): failed to idle before, with err=%d!",
+			gt_err(gt, "%s(%s): GT failed to idle before, with err=%d!",
 			       func, name, err);
 			return err;
 		}
@@ -69,7 +70,7 @@ int igt_live_test_end(struct igt_live_test *t)
 			    i915_reset_engine_count(&i915->gpu_error, engine))
 				continue;
 
-			pr_err("%s(%s): engine '%s' was reset %d times!\n",
+			gt_err(gt, "%s(%s): engine '%s' was reset %d times!\n",
 			       t->func, t->name, engine->name,
 			       i915_reset_engine_count(&i915->gpu_error, engine) -
 			       t->reset_engine[id]);
-- 
GitLab


From 47d8b3029684ce257e23022286b0861f72ac3f03 Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Mon, 24 Apr 2023 15:47:49 +0530
Subject: [PATCH 0311/1791] drm/i915/mtl: Add workaround 14018778641

WA 18018781329 is applicable now across all MTL
steppings.

V2:
  - Remove IS_MTL check, code already running for MTL - Matt

Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424101749.3719600-1-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 30 ++++++++++-----------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 312eb8b5f9491..de4f8e2e8e8c2 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -1695,19 +1695,18 @@ pvc_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 static void
 xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 {
+	/* Wa_14018778641 / Wa_18018781329 */
+	wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB);
+	wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB);
+
 	if (IS_MTL_GRAPHICS_STEP(gt->i915, M, STEP_A0, STEP_B0) ||
 	    IS_MTL_GRAPHICS_STEP(gt->i915, P, STEP_A0, STEP_B0)) {
 		/* Wa_14014830051 */
 		wa_mcr_write_clr(wal, SARB_CHICKEN1, COMP_CKN_IN);
 
-		/* Wa_18018781329 */
-		wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB);
-		wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB);
-
 		/* Wa_14015795083 */
 		wa_write_clr(wal, GEN7_MISCCPCTL, GEN12_DOP_CLOCK_GATE_RENDER_ENABLE);
 	}
-
 	/*
 	 * Unlike older platforms, we no longer setup implicit steering here;
 	 * all MCR accesses are explicitly steered.
@@ -1718,17 +1717,16 @@ xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 static void
 xelpmp_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 {
-	if (IS_MTL_MEDIA_STEP(gt->i915, STEP_A0, STEP_B0)) {
-		/*
-		 * Wa_18018781329
-		 *
-		 * Note that although these registers are MCR on the primary
-		 * GT, the media GT's versions are regular singleton registers.
-		 */
-		wa_write_or(wal, XELPMP_GSC_MOD_CTRL, FORCE_MISS_FTLB);
-		wa_write_or(wal, XELPMP_VDBX_MOD_CTRL, FORCE_MISS_FTLB);
-		wa_write_or(wal, XELPMP_VEBX_MOD_CTRL, FORCE_MISS_FTLB);
-	}
+	/*
+	 * Wa_14018778641
+	 * Wa_18018781329
+	 *
+	 * Note that although these registers are MCR on the primary
+	 * GT, the media GT's versions are regular singleton registers.
+	 */
+	wa_write_or(wal, XELPMP_GSC_MOD_CTRL, FORCE_MISS_FTLB);
+	wa_write_or(wal, XELPMP_VDBX_MOD_CTRL, FORCE_MISS_FTLB);
+	wa_write_or(wal, XELPMP_VEBX_MOD_CTRL, FORCE_MISS_FTLB);
 
 	debug_dump_steering(gt);
 }
-- 
GitLab


From 94545f27155df50d2db7a1c896ca6b8e81fb62d1 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 24 Apr 2023 10:47:51 +0200
Subject: [PATCH 0312/1791] arch/x86: Include <asm/fb.h> in fbdev source file

Move the implementation of fb_pgprotect() to fbdev.c and include
<asm/fb.h>. Fixes the following warning:

  ../arch/x86/video/fbdev.c:14:5: warning: no previous prototype for 'fb_is_primary_device' [-Wmissing-prototypes]
     14 | int fb_is_primary_device(struct fb_info *info)
        |     ^~~~~~~~~~~~~~~~~~~~

Just including <asm/fb.h> results in a number of built-in errors
about undefined function. Moving fb_pgprotect() to the source file
avoids the required include statements in the header. The function
is only called occasionally from fb_mmap(), [1] so having it as static
inline had no benefit.

While at it, fix the codying style in fbdev.c.

Link: https://elixir.bootlin.com/linux/v6.3-rc7/source/drivers/video/fbdev/core/fbmem.c#L1404 # 1
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230424084751.14641-1-tzimmermann@suse.de
---
 arch/x86/include/asm/fb.h | 14 ++------------
 arch/x86/video/fbdev.c    | 17 ++++++++++++++++-
 2 files changed, 18 insertions(+), 13 deletions(-)

diff --git a/arch/x86/include/asm/fb.h b/arch/x86/include/asm/fb.h
index a3fb801f12f18..23873da8fb77c 100644
--- a/arch/x86/include/asm/fb.h
+++ b/arch/x86/include/asm/fb.h
@@ -2,21 +2,11 @@
 #ifndef _ASM_X86_FB_H
 #define _ASM_X86_FB_H
 
-#include <asm/page.h>
-
 struct fb_info;
 struct file;
+struct vm_area_struct;
 
-static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
-				unsigned long off)
-{
-	unsigned long prot;
-
-	prot = pgprot_val(vma->vm_page_prot) & ~_PAGE_CACHE_MASK;
-	if (boot_cpu_data.x86 > 3)
-		pgprot_val(vma->vm_page_prot) =
-			prot | cachemode2protval(_PAGE_CACHE_MODE_UC_MINUS);
-}
+void fb_pgprotect(struct file *file, struct vm_area_struct *vma, unsigned long off);
 #define fb_pgprotect fb_pgprotect
 
 int fb_is_primary_device(struct fb_info *info);
diff --git a/arch/x86/video/fbdev.c b/arch/x86/video/fbdev.c
index 5ec4eafbb9811..57ee3c158f976 100644
--- a/arch/x86/video/fbdev.c
+++ b/arch/x86/video/fbdev.c
@@ -6,11 +6,25 @@
  * for more details.
  *
  */
+
+#include <asm/fb.h>
+
 #include <linux/fb.h>
-#include <linux/pci.h>
 #include <linux/module.h>
+#include <linux/pci.h>
 #include <linux/vgaarb.h>
 
+void fb_pgprotect(struct file *file, struct vm_area_struct *vma, unsigned long off)
+{
+	unsigned long prot;
+
+	prot = pgprot_val(vma->vm_page_prot) & ~_PAGE_CACHE_MASK;
+	if (boot_cpu_data.x86 > 3)
+		pgprot_val(vma->vm_page_prot) =
+			prot | cachemode2protval(_PAGE_CACHE_MODE_UC_MINUS);
+}
+EXPORT_SYMBOL(fb_pgprotect);
+
 int fb_is_primary_device(struct fb_info *info)
 {
 	struct device *device = info->device;
@@ -26,4 +40,5 @@ int fb_is_primary_device(struct fb_info *info)
 	return 0;
 }
 EXPORT_SYMBOL(fb_is_primary_device);
+
 MODULE_LICENSE("GPL");
-- 
GitLab


From 80ab31799002166ac7c660bacfbff4f85bc29107 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 21 Apr 2023 15:47:42 -0700
Subject: [PATCH 0313/1791] drm/i915/guc: Actually return an error if GuC
 version range check fails

Dan Carpenter pointed out that 'err' was not being set in the case
where the GuC firmware version range check fails. Fix that.

Note that while this is a bug fix for a previous patch (see Fixes tag
below). It is an exceedingly low risk bug. The range check is
asserting that the GuC firmware version is within spec. So it should
not be possible to ever have a firmware file that fails this check. If
larger version numbers are required in the future, that would be a
backwards breaking spec change and thus require a major version bump,
in which case an old i915 driver would not load that new version anyway.

Fixes: 9bbba0667f37 ("drm/i915/guc: Use GuC submission API version number")
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421224742.2357198-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 24765c30a0e1d..c36e68e23a141 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -635,9 +635,10 @@ static bool is_ver_8bit(struct intel_uc_fw_ver *ver)
 	return ver->major < 0xFF && ver->minor < 0xFF && ver->patch < 0xFF;
 }
 
-static bool guc_check_version_range(struct intel_uc_fw *uc_fw)
+static int guc_check_version_range(struct intel_uc_fw *uc_fw)
 {
 	struct intel_guc *guc = container_of(uc_fw, struct intel_guc, fw);
+	struct intel_gt *gt = __uc_fw_to_gt(uc_fw);
 
 	/*
 	 * GuC version number components are defined as being 8-bits.
@@ -646,24 +647,24 @@ static bool guc_check_version_range(struct intel_uc_fw *uc_fw)
 	 */
 
 	if (!is_ver_8bit(&uc_fw->file_selected.ver)) {
-		gt_warn(__uc_fw_to_gt(uc_fw), "%s firmware: invalid file version: 0x%02X:%02X:%02X\n",
+		gt_warn(gt, "%s firmware: invalid file version: 0x%02X:%02X:%02X\n",
 			intel_uc_fw_type_repr(uc_fw->type),
 			uc_fw->file_selected.ver.major,
 			uc_fw->file_selected.ver.minor,
 			uc_fw->file_selected.ver.patch);
-		return false;
+		return -EINVAL;
 	}
 
 	if (!is_ver_8bit(&guc->submission_version)) {
-		gt_warn(__uc_fw_to_gt(uc_fw), "%s firmware: invalid submit version: 0x%02X:%02X:%02X\n",
+		gt_warn(gt, "%s firmware: invalid submit version: 0x%02X:%02X:%02X\n",
 			intel_uc_fw_type_repr(uc_fw->type),
 			guc->submission_version.major,
 			guc->submission_version.minor,
 			guc->submission_version.patch);
-		return false;
+		return -EINVAL;
 	}
 
-	return true;
+	return i915_inject_probe_error(gt->i915, -EINVAL);
 }
 
 static int check_fw_header(struct intel_gt *gt,
@@ -772,8 +773,11 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
 	if (err)
 		goto fail;
 
-	if (uc_fw->type == INTEL_UC_FW_TYPE_GUC && !guc_check_version_range(uc_fw))
-		goto fail;
+	if (uc_fw->type == INTEL_UC_FW_TYPE_GUC) {
+		err = guc_check_version_range(uc_fw);
+		if (err)
+			goto fail;
+	}
 
 	if (uc_fw->file_wanted.ver.major && uc_fw->file_selected.ver.major) {
 		/* Check the file's major version was as it claimed */
-- 
GitLab


From a03f028542779fbdb0130b6a5e15445c94a859e8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Thu, 20 Apr 2023 20:22:27 -0300
Subject: [PATCH 0314/1791] drm/vkms: allow full alpha blending on all planes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Before commit bc0d7fdefec6 ("drm: vkms: Supports to the case where
primary plane doesn't match the CRTC"), the composition was executed on
top of the primary plane. Therefore, the primary plane was not able to
support the alpha channel. After commit bc0d7fdefec6, this is possible,
as the composition is now executed on top of the CRTC.

So, allow all planes to support the alpha channel, making full alpha
blending possible in vkms.

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420232228.273340-1-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_plane.c | 34 +++----------------------------
 1 file changed, 3 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 0a23875900ec5..87b2bccaf0839 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -12,12 +12,6 @@
 #include "vkms_formats.h"
 
 static const u32 vkms_formats[] = {
-	DRM_FORMAT_XRGB8888,
-	DRM_FORMAT_XRGB16161616,
-	DRM_FORMAT_RGB565
-};
-
-static const u32 vkms_plane_formats[] = {
 	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_XRGB16161616,
@@ -185,7 +179,7 @@ static void vkms_cleanup_fb(struct drm_plane *plane,
 	drm_gem_fb_vunmap(fb, shadow_plane_state->map);
 }
 
-static const struct drm_plane_helper_funcs vkms_primary_helper_funcs = {
+static const struct drm_plane_helper_funcs vkms_plane_helper_funcs = {
 	.atomic_update		= vkms_plane_atomic_update,
 	.atomic_check		= vkms_plane_atomic_check,
 	.prepare_fb		= vkms_prepare_fb,
@@ -196,38 +190,16 @@ struct vkms_plane *vkms_plane_init(struct vkms_device *vkmsdev,
 				   enum drm_plane_type type, int index)
 {
 	struct drm_device *dev = &vkmsdev->drm;
-	const struct drm_plane_helper_funcs *funcs;
 	struct vkms_plane *plane;
-	const u32 *formats;
-	int nformats;
-
-	switch (type) {
-	case DRM_PLANE_TYPE_PRIMARY:
-		formats = vkms_formats;
-		nformats = ARRAY_SIZE(vkms_formats);
-		funcs = &vkms_primary_helper_funcs;
-		break;
-	case DRM_PLANE_TYPE_CURSOR:
-	case DRM_PLANE_TYPE_OVERLAY:
-		formats = vkms_plane_formats;
-		nformats = ARRAY_SIZE(vkms_plane_formats);
-		funcs = &vkms_primary_helper_funcs;
-		break;
-	default:
-		formats = vkms_formats;
-		nformats = ARRAY_SIZE(vkms_formats);
-		funcs = &vkms_primary_helper_funcs;
-		break;
-	}
 
 	plane = drmm_universal_plane_alloc(dev, struct vkms_plane, base, 1 << index,
 					   &vkms_plane_funcs,
-					   formats, nformats,
+					   vkms_formats, ARRAY_SIZE(vkms_formats),
 					   NULL, type, NULL);
 	if (IS_ERR(plane))
 		return plane;
 
-	drm_plane_helper_add(&plane->base, funcs);
+	drm_plane_helper_add(&plane->base, &vkms_plane_helper_funcs);
 
 	return plane;
 }
-- 
GitLab


From 09c2c8c94b3d3b3c259b05bdd5f67f3a0a8e2063 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Thu, 20 Apr 2023 20:22:28 -0300
Subject: [PATCH 0315/1791] drm/vkms: drop full alpha blending TODO
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that VKMS supports full alpha blending on all planes, drop the
"ARGB format on primary plane" and "Full alpha blending on all planes"
tasks from the TODO list.

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420232228.273340-2-mcanal@igalia.com
---
 Documentation/gpu/vkms.rst | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/Documentation/gpu/vkms.rst b/Documentation/gpu/vkms.rst
index 49db221c0f528..0f599c8976148 100644
--- a/Documentation/gpu/vkms.rst
+++ b/Documentation/gpu/vkms.rst
@@ -118,13 +118,8 @@ Add Plane Features
 
 There's lots of plane features we could add support for:
 
-- ARGB format on primary plane: blend the primary plane into background with
-  translucent alpha.
-
 - Add background color KMS property[Good to get started].
 
-- Full alpha blending on all planes.
-
 - Rotation, scaling.
 
 - Additional buffer formats, especially YUV formats for video like NV12.
-- 
GitLab


From bddc18913bd44adae5c828fd514d570f43ba1576 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Thu, 20 Apr 2023 15:12:47 -0700
Subject: [PATCH 0316/1791] drm/i915/mtl: Add the missing CPU transcoder mask
 in intel_device_info
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CPU transcoder mask is used to iterate over the available
CPU transcoders in the macro for_each_cpu_transcoder().

The macro is broken on MTL and got highlighted when audio
state was being tracked for each transcoder added in [1].

Add the missing CPU transcoder mask which is similar to ADL-P
mask but without DSI transcoders.

[1]: https://patchwork.freedesktop.org/patch/523723/

Fixes: 7835303982d1 ("drm/i915/mtl: Add MeteorLake PCI IDs")
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Acked-by: Haridhar Kalvala <haridhar.kalvala@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420221248.2511314-1-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/i915_pci.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 6d431489131ee..6b5c119a7d829 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1134,6 +1134,8 @@ static const struct intel_gt_definition xelpmp_extra_gt[] = {
 static const struct intel_device_info mtl_info = {
 	XE_HP_FEATURES,
 	XE_LPDP_FEATURES,
+	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 	/*
 	 * Real graphics IP version will be obtained from hardware GMD_ID
 	 * register.  Value provided here is just for sanity checking.
-- 
GitLab


From d81268ee1c1073471cf0e8adb0fc9d026b602c3b Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Thu, 20 Apr 2023 09:40:39 -0700
Subject: [PATCH 0317/1791] drm/i915/hwmon: Get mutex and rpm ref just once in
 hwm_power_max_write

In preparation for follow-on patches, refactor hwm_power_max_write to take
hwmon_lock and runtime pm wakeref at start of the function and release them
at the end, therefore acquiring these just once each.

Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420164041.1428455-2-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/i915_hwmon.c | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c
index 8e7dccc8d3a0e..7f44e809ca155 100644
--- a/drivers/gpu/drm/i915/i915_hwmon.c
+++ b/drivers/gpu/drm/i915/i915_hwmon.c
@@ -396,31 +396,33 @@ hwm_power_max_write(struct hwm_drvdata *ddat, long val)
 {
 	struct i915_hwmon *hwmon = ddat->hwmon;
 	intel_wakeref_t wakeref;
+	int ret = 0;
 	u32 nval;
 
+	mutex_lock(&hwmon->hwmon_lock);
+	wakeref = intel_runtime_pm_get(ddat->uncore->rpm);
+
 	/* Disable PL1 limit and verify, because the limit cannot be disabled on all platforms */
 	if (val == PL1_DISABLE) {
-		mutex_lock(&hwmon->hwmon_lock);
-		with_intel_runtime_pm(ddat->uncore->rpm, wakeref) {
-			intel_uncore_rmw(ddat->uncore, hwmon->rg.pkg_rapl_limit,
-					 PKG_PWR_LIM_1_EN, 0);
-			nval = intel_uncore_read(ddat->uncore, hwmon->rg.pkg_rapl_limit);
-		}
-		mutex_unlock(&hwmon->hwmon_lock);
+		intel_uncore_rmw(ddat->uncore, hwmon->rg.pkg_rapl_limit,
+				 PKG_PWR_LIM_1_EN, 0);
+		nval = intel_uncore_read(ddat->uncore, hwmon->rg.pkg_rapl_limit);
 
 		if (nval & PKG_PWR_LIM_1_EN)
-			return -ENODEV;
-		return 0;
+			ret = -ENODEV;
+		goto exit;
 	}
 
 	/* Computation in 64-bits to avoid overflow. Round to nearest. */
 	nval = DIV_ROUND_CLOSEST_ULL((u64)val << hwmon->scl_shift_power, SF_POWER);
 	nval = PKG_PWR_LIM_1_EN | REG_FIELD_PREP(PKG_PWR_LIM_1, nval);
 
-	hwm_locked_with_pm_intel_uncore_rmw(ddat, hwmon->rg.pkg_rapl_limit,
-					    PKG_PWR_LIM_1_EN | PKG_PWR_LIM_1,
-					    nval);
-	return 0;
+	intel_uncore_rmw(ddat->uncore, hwmon->rg.pkg_rapl_limit,
+			 PKG_PWR_LIM_1_EN | PKG_PWR_LIM_1, nval);
+exit:
+	intel_runtime_pm_put(ddat->uncore->rpm, wakeref);
+	mutex_unlock(&hwmon->hwmon_lock);
+	return ret;
 }
 
 static int
-- 
GitLab


From 1b44019a93e2bc6088d777b3d3072df5f40f9324 Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Thu, 20 Apr 2023 09:40:40 -0700
Subject: [PATCH 0318/1791] drm/i915/guc: Disable PL1 power limit when loading
 GuC firmware

On dGfx, the PL1 power limit being enabled and set to a low value results
in a low GPU operating freq. It also negates the freq raise operation which
is done before GuC firmware load. As a result GuC firmware load can time
out. Such timeouts were seen in the GL #8062 bug below (where the PL1 power
limit was enabled and set to a low value). Therefore disable the PL1 power
limit when allowed by HW when loading GuC firmware.

v2:
 - Take mutex (to disallow writes to power1_max) across GuC reset/fw load
 - Add hwm_power_max_restore to error return code path

v3 (Jani N):
 - Add/remove explanatory comments
 - Function renames
 - Type corrections
 - Locking annotation

v4:
 - Don't hold the lock across GuC reset (Rodrigo)
 - New locking scheme (suggested by Rodrigo)
 - Eliminate rpm_get in power_max_disable/restore, not needed (Tvrtko)

v5:
 - Fix uninitialized pl1en variable compile warning reported by kernel
   build robot by creating new err_rps label

Link: https://gitlab.freedesktop.org/drm/intel/-/issues/8062
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420164041.1428455-3-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc.c | 13 +++++++--
 drivers/gpu/drm/i915/i915_hwmon.c     | 40 +++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_hwmon.h     |  7 +++++
 3 files changed, 58 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index 4ccb4be4c9cba..996168312340e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -18,6 +18,7 @@
 #include "intel_uc.h"
 
 #include "i915_drv.h"
+#include "i915_hwmon.h"
 
 static const struct intel_uc_ops uc_ops_off;
 static const struct intel_uc_ops uc_ops_on;
@@ -461,6 +462,7 @@ static int __uc_init_hw(struct intel_uc *uc)
 	struct intel_guc *guc = &uc->guc;
 	struct intel_huc *huc = &uc->huc;
 	int ret, attempts;
+	bool pl1en;
 
 	GEM_BUG_ON(!intel_uc_supports_guc(uc));
 	GEM_BUG_ON(!intel_uc_wants_guc(uc));
@@ -491,6 +493,9 @@ static int __uc_init_hw(struct intel_uc *uc)
 	else
 		attempts = 1;
 
+	/* Disable a potentially low PL1 power limit to allow freq to be raised */
+	i915_hwmon_power_max_disable(gt->i915, &pl1en);
+
 	intel_rps_raise_unslice(&uc_to_gt(uc)->rps);
 
 	while (attempts--) {
@@ -500,7 +505,7 @@ static int __uc_init_hw(struct intel_uc *uc)
 		 */
 		ret = __uc_sanitize(uc);
 		if (ret)
-			goto err_out;
+			goto err_rps;
 
 		intel_huc_fw_upload(huc);
 		intel_guc_ads_reset(guc);
@@ -547,6 +552,8 @@ static int __uc_init_hw(struct intel_uc *uc)
 		intel_rps_lower_unslice(&uc_to_gt(uc)->rps);
 	}
 
+	i915_hwmon_power_max_restore(gt->i915, pl1en);
+
 	guc_info(guc, "submission %s\n", str_enabled_disabled(intel_uc_uses_guc_submission(uc)));
 	guc_info(guc, "SLPC %s\n", str_enabled_disabled(intel_uc_uses_guc_slpc(uc)));
 
@@ -559,10 +566,12 @@ static int __uc_init_hw(struct intel_uc *uc)
 	intel_guc_submission_disable(guc);
 err_log_capture:
 	__uc_capture_load_err_log(uc);
-err_out:
+err_rps:
 	/* Return GT back to RPn */
 	intel_rps_lower_unslice(&uc_to_gt(uc)->rps);
 
+	i915_hwmon_power_max_restore(gt->i915, pl1en);
+err_out:
 	__uc_sanitize(uc);
 
 	if (!ret) {
diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c
index 7f44e809ca155..9ab8971679fe3 100644
--- a/drivers/gpu/drm/i915/i915_hwmon.c
+++ b/drivers/gpu/drm/i915/i915_hwmon.c
@@ -50,6 +50,7 @@ struct hwm_drvdata {
 	struct hwm_energy_info ei;		/*  Energy info for energy1_input */
 	char name[12];
 	int gt_n;
+	bool reset_in_progress;
 };
 
 struct i915_hwmon {
@@ -400,6 +401,10 @@ hwm_power_max_write(struct hwm_drvdata *ddat, long val)
 	u32 nval;
 
 	mutex_lock(&hwmon->hwmon_lock);
+	if (hwmon->ddat.reset_in_progress) {
+		ret = -EAGAIN;
+		goto unlock;
+	}
 	wakeref = intel_runtime_pm_get(ddat->uncore->rpm);
 
 	/* Disable PL1 limit and verify, because the limit cannot be disabled on all platforms */
@@ -421,6 +426,7 @@ hwm_power_max_write(struct hwm_drvdata *ddat, long val)
 			 PKG_PWR_LIM_1_EN | PKG_PWR_LIM_1, nval);
 exit:
 	intel_runtime_pm_put(ddat->uncore->rpm, wakeref);
+unlock:
 	mutex_unlock(&hwmon->hwmon_lock);
 	return ret;
 }
@@ -472,6 +478,40 @@ hwm_power_write(struct hwm_drvdata *ddat, u32 attr, int chan, long val)
 	}
 }
 
+void i915_hwmon_power_max_disable(struct drm_i915_private *i915, bool *old)
+{
+	struct i915_hwmon *hwmon = i915->hwmon;
+	u32 r;
+
+	if (!hwmon || !i915_mmio_reg_valid(hwmon->rg.pkg_rapl_limit))
+		return;
+
+	mutex_lock(&hwmon->hwmon_lock);
+
+	hwmon->ddat.reset_in_progress = true;
+	r = intel_uncore_rmw(hwmon->ddat.uncore, hwmon->rg.pkg_rapl_limit,
+			     PKG_PWR_LIM_1_EN, 0);
+	*old = !!(r & PKG_PWR_LIM_1_EN);
+
+	mutex_unlock(&hwmon->hwmon_lock);
+}
+
+void i915_hwmon_power_max_restore(struct drm_i915_private *i915, bool old)
+{
+	struct i915_hwmon *hwmon = i915->hwmon;
+
+	if (!hwmon || !i915_mmio_reg_valid(hwmon->rg.pkg_rapl_limit))
+		return;
+
+	mutex_lock(&hwmon->hwmon_lock);
+
+	intel_uncore_rmw(hwmon->ddat.uncore, hwmon->rg.pkg_rapl_limit,
+			 PKG_PWR_LIM_1_EN, old ? PKG_PWR_LIM_1_EN : 0);
+	hwmon->ddat.reset_in_progress = false;
+
+	mutex_unlock(&hwmon->hwmon_lock);
+}
+
 static umode_t
 hwm_energy_is_visible(const struct hwm_drvdata *ddat, u32 attr)
 {
diff --git a/drivers/gpu/drm/i915/i915_hwmon.h b/drivers/gpu/drm/i915/i915_hwmon.h
index 7ca9cf2c34c96..0fcb7de844061 100644
--- a/drivers/gpu/drm/i915/i915_hwmon.h
+++ b/drivers/gpu/drm/i915/i915_hwmon.h
@@ -7,14 +7,21 @@
 #ifndef __I915_HWMON_H__
 #define __I915_HWMON_H__
 
+#include <linux/types.h>
+
 struct drm_i915_private;
+struct intel_gt;
 
 #if IS_REACHABLE(CONFIG_HWMON)
 void i915_hwmon_register(struct drm_i915_private *i915);
 void i915_hwmon_unregister(struct drm_i915_private *i915);
+void i915_hwmon_power_max_disable(struct drm_i915_private *i915, bool *old);
+void i915_hwmon_power_max_restore(struct drm_i915_private *i915, bool old);
 #else
 static inline void i915_hwmon_register(struct drm_i915_private *i915) { };
 static inline void i915_hwmon_unregister(struct drm_i915_private *i915) { };
+static inline void i915_hwmon_power_max_disable(struct drm_i915_private *i915, bool *old) { };
+static inline void i915_hwmon_power_max_restore(struct drm_i915_private *i915, bool old) { };
 #endif
 
 #endif /* __I915_HWMON_H__ */
-- 
GitLab


From 655bd3b954cf18ae4e1ff69ebbf4a20b562f580b Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Thu, 20 Apr 2023 09:40:41 -0700
Subject: [PATCH 0319/1791] drm/i915/hwmon: Block waiting for GuC reset to
 complete

Instead of erroring out when GuC reset is in progress, block waiting for
GuC reset to complete which is a more reasonable uapi behavior.

v2: Avoid race between wake_up_all and waiting for wakeup (Rodrigo)
v3: Remove timeout when blocked (Tvrtko)

Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230420164041.1428455-4-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/i915_hwmon.c | 29 +++++++++++++++++++++++++----
 1 file changed, 25 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c
index 9ab8971679fe3..a3bdd9f68a458 100644
--- a/drivers/gpu/drm/i915/i915_hwmon.c
+++ b/drivers/gpu/drm/i915/i915_hwmon.c
@@ -51,6 +51,7 @@ struct hwm_drvdata {
 	char name[12];
 	int gt_n;
 	bool reset_in_progress;
+	wait_queue_head_t waitq;
 };
 
 struct i915_hwmon {
@@ -397,14 +398,32 @@ hwm_power_max_write(struct hwm_drvdata *ddat, long val)
 {
 	struct i915_hwmon *hwmon = ddat->hwmon;
 	intel_wakeref_t wakeref;
+	DEFINE_WAIT(wait);
 	int ret = 0;
 	u32 nval;
 
-	mutex_lock(&hwmon->hwmon_lock);
-	if (hwmon->ddat.reset_in_progress) {
-		ret = -EAGAIN;
-		goto unlock;
+	/* Block waiting for GuC reset to complete when needed */
+	for (;;) {
+		mutex_lock(&hwmon->hwmon_lock);
+
+		prepare_to_wait(&ddat->waitq, &wait, TASK_INTERRUPTIBLE);
+
+		if (!hwmon->ddat.reset_in_progress)
+			break;
+
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+
+		mutex_unlock(&hwmon->hwmon_lock);
+
+		schedule();
 	}
+	finish_wait(&ddat->waitq, &wait);
+	if (ret)
+		goto unlock;
+
 	wakeref = intel_runtime_pm_get(ddat->uncore->rpm);
 
 	/* Disable PL1 limit and verify, because the limit cannot be disabled on all platforms */
@@ -508,6 +527,7 @@ void i915_hwmon_power_max_restore(struct drm_i915_private *i915, bool old)
 	intel_uncore_rmw(hwmon->ddat.uncore, hwmon->rg.pkg_rapl_limit,
 			 PKG_PWR_LIM_1_EN, old ? PKG_PWR_LIM_1_EN : 0);
 	hwmon->ddat.reset_in_progress = false;
+	wake_up_all(&hwmon->ddat.waitq);
 
 	mutex_unlock(&hwmon->hwmon_lock);
 }
@@ -784,6 +804,7 @@ void i915_hwmon_register(struct drm_i915_private *i915)
 	ddat->uncore = &i915->uncore;
 	snprintf(ddat->name, sizeof(ddat->name), "i915");
 	ddat->gt_n = -1;
+	init_waitqueue_head(&ddat->waitq);
 
 	for_each_gt(gt, i915, i) {
 		ddat_gt = hwmon->ddat_gt + i;
-- 
GitLab


From a7fa1537b791cfb2ea3895ac448ab1455a4405f8 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Tue, 25 Apr 2023 11:30:11 -0700
Subject: [PATCH 0320/1791] drm/i915/mtl: Implement Wa_14019141245

Enable strict RAR to prevent spurious GPU hangs.

v1.1: Rebase
Bspec: 51762
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by:Haridhar Kalvala <haridhar.kalvala@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425183011.865085-1-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_regs.h     | 5 +++++
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 4 ++++
 drivers/gpu/drm/i915/i915_perf_oa_regs.h    | 4 ----
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index e8c3b762a92a3..af80d2fe739b0 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -529,6 +529,11 @@
 
 #define GEN8_RC6_CTX_INFO			_MMIO(0x8504)
 
+#define GEN12_SQCNT1				_MMIO(0x8718)
+#define   GEN12_SQCNT1_PMON_ENABLE		REG_BIT(30)
+#define   GEN12_SQCNT1_OABPC			REG_BIT(29)
+#define   GEN12_STRICT_RAR_ENABLE		REG_BIT(23)
+
 #define XEHP_SQCM				MCR_REG(0x8724)
 #define   EN_32B_ACCESS				REG_BIT(30)
 
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index de4f8e2e8e8c2..ad9e7f49a6fa4 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -1699,6 +1699,9 @@ xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 	wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB);
 	wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB);
 
+	/* Wa_14019141245 */
+	wa_write_or(wal, GEN12_SQCNT1, GEN12_STRICT_RAR_ENABLE);
+
 	if (IS_MTL_GRAPHICS_STEP(gt->i915, M, STEP_A0, STEP_B0) ||
 	    IS_MTL_GRAPHICS_STEP(gt->i915, P, STEP_A0, STEP_B0)) {
 		/* Wa_14014830051 */
@@ -1707,6 +1710,7 @@ xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 		/* Wa_14015795083 */
 		wa_write_clr(wal, GEN7_MISCCPCTL, GEN12_DOP_CLOCK_GATE_RENDER_ENABLE);
 	}
+
 	/*
 	 * Unlike older platforms, we no longer setup implicit steering here;
 	 * all MCR accesses are explicitly steered.
diff --git a/drivers/gpu/drm/i915/i915_perf_oa_regs.h b/drivers/gpu/drm/i915/i915_perf_oa_regs.h
index ba103875e19f5..e5ac7a8b5cb6e 100644
--- a/drivers/gpu/drm/i915/i915_perf_oa_regs.h
+++ b/drivers/gpu/drm/i915/i915_perf_oa_regs.h
@@ -134,10 +134,6 @@
 #define GDT_CHICKEN_BITS    _MMIO(0x9840)
 #define   GT_NOA_ENABLE	    0x00000080
 
-#define GEN12_SQCNT1				_MMIO(0x8718)
-#define   GEN12_SQCNT1_PMON_ENABLE		REG_BIT(30)
-#define   GEN12_SQCNT1_OABPC			REG_BIT(29)
-
 /* Gen12 OAM unit */
 #define GEN12_OAM_HEAD_POINTER_OFFSET   (0x1a0)
 #define  GEN12_OAM_HEAD_POINTER_MASK    0xffffffc0
-- 
GitLab


From 40aeab044a352b9af106960542257c1b9b652e8a Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Wed, 19 Apr 2023 15:19:13 -0400
Subject: [PATCH 0321/1791] drm/doc/rfc: Introduce the merge plan for the Xe
 driver.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Let’s establish a merge plan for Xe, by writing down clear pre-merge goals, in
order to avoid unnecessary delays.

This initial document starts with a TODO list containing items with clear and
measurable key results. Xe’s initial pull request should only be sent to
dri-devel after all the items are clearly resolved.

Since many of them involve some level of a community consensus, in many cases,
the consensus will be reached in follow-up patches to this document with more
details of the API or helpers that will be developed or modified.

Besides of the items that are highlighted in this document, it is important
to highlight that Oded, has been volunteered to give the overall ack on Xe
driver as the way to confirm that it looks good for upstream.

v2: Incorporated Daniel's feedback:
    - Do not make long-running compute a blocker.
    - Add a mention to drm-exec that that ties to vm_bind and long-running
      compute jobs. Then I also added GPUVA since I recently noticed that this
      ties also to the work Matt is doing on that front.
    - Added the devcoredump section.
    - Add the mention to Oded being volunteered for the overall ack.

v3: Reword a bit the Async VM_BIND to incorporate Daniel's feedback on ensuring
    the async vmbind consensus explicitly include Mesa, besides other kernel
    drivers.

Cc: Dave Airlie <airlied@redhat.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
Signed-off-by: Luis Strano <luis.strano@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230419191913.158807-1-rodrigo.vivi@intel.com
---
 Documentation/gpu/rfc/index.rst |   4 +
 Documentation/gpu/rfc/xe.rst    | 235 ++++++++++++++++++++++++++++++++
 2 files changed, 239 insertions(+)
 create mode 100644 Documentation/gpu/rfc/xe.rst

diff --git a/Documentation/gpu/rfc/index.rst b/Documentation/gpu/rfc/index.rst
index 476719771eef3..e4f7b005138d0 100644
--- a/Documentation/gpu/rfc/index.rst
+++ b/Documentation/gpu/rfc/index.rst
@@ -31,3 +31,7 @@ host such documentation:
 .. toctree::
 
     i915_vm_bind.rst
+
+.. toctree::
+
+   xe.rst
diff --git a/Documentation/gpu/rfc/xe.rst b/Documentation/gpu/rfc/xe.rst
new file mode 100644
index 0000000000000..2516fe141db64
--- /dev/null
+++ b/Documentation/gpu/rfc/xe.rst
@@ -0,0 +1,235 @@
+==========================
+Xe – Merge Acceptance Plan
+==========================
+Xe is a new driver for Intel GPUs that supports both integrated and
+discrete platforms starting with Tiger Lake (first Intel Xe Architecture).
+
+This document aims to establish a merge plan for the Xe, by writing down clear
+pre-merge goals, in order to avoid unnecessary delays.
+
+Xe – Overview
+=============
+The main motivation of Xe is to have a fresh base to work from that is
+unencumbered by older platforms, whilst also taking the opportunity to
+rearchitect our driver to increase sharing across the drm subsystem, both
+leveraging and allowing us to contribute more towards other shared components
+like TTM and drm/scheduler.
+
+This is also an opportunity to start from the beginning with a clean uAPI that is
+extensible by design and already aligned with the modern userspace needs. For
+this reason, the memory model is solely based on GPU Virtual Address space
+bind/unbind (‘VM_BIND’) of GEM buffer objects (BOs) and execution only supporting
+explicit synchronization. With persistent mapping across the execution, the
+userspace does not need to provide a list of all required mappings during each
+submission.
+
+The new driver leverages a lot from i915. As for display, the intent is to share
+the display code with the i915 driver so that there is maximum reuse there.
+
+As for the power management area, the goal is to have a much-simplified support
+for the system suspend states (S-states), PCI device suspend states (D-states),
+GPU/Render suspend states (R-states) and frequency management. It should leverage
+as much as possible all the existent PCI-subsystem infrastructure (pm and
+runtime_pm) and underlying firmware components such PCODE and GuC for the power
+states and frequency decisions.
+
+Repository:
+
+https://gitlab.freedesktop.org/drm/xe/kernel (branch drm-xe-next)
+
+Xe – Platforms
+==============
+Currently, Xe is already functional and has experimental support for multiple
+platforms starting from Tiger Lake, with initial support in userspace implemented
+in Mesa (for Iris and Anv, our OpenGL and Vulkan drivers), as well as in NEO
+(for OpenCL and Level0).
+
+During a transition period, platforms will be supported by both Xe and i915.
+However, the force_probe mechanism existent in both drivers will allow only one
+official and by-default probe at a given time.
+
+For instance, in order to probe a DG2 which PCI ID is 0x5690 by Xe instead of
+i915, the following set of parameters need to be used:
+
+```
+i915.force_probe=!5690 xe.force_probe=5690
+```
+
+In both drivers, the ‘.require_force_probe’ protection forces the user to use the
+force_probe parameter while the driver is under development. This protection is
+only removed when the support for the platform and the uAPI are stable. Stability
+which needs to be demonstrated by CI results.
+
+In order to avoid user space regressions, i915 will continue to support all the
+current platforms that are already out of this protection. Xe support will be
+forever experimental and dependent on the usage of force_probe for these
+platforms.
+
+When the time comes for Xe, the protection will be lifted on Xe and kept in i915.
+
+Xe driver will be protected with both STAGING Kconfig and force_probe. Changes in
+the uAPI are expected while the driver is behind these protections. STAGING will
+be removed when the driver uAPI gets to a mature state where we can guarantee the
+‘no regression’ rule. Then force_probe will be lifted only for future platforms
+that will be productized with Xe driver, but not with i915.
+
+Xe – Pre-Merge Goals
+====================
+
+Drm_scheduler
+-------------
+Xe primarily uses Firmware based scheduling (GuC FW). However, it will use
+drm_scheduler as the scheduler ‘frontend’ for userspace submission in order to
+resolve syncobj and dma-buf implicit sync dependencies. However, drm_scheduler is
+not yet prepared to handle the 1-to-1 relationship between drm_gpu_scheduler and
+drm_sched_entity.
+
+Deeper changes to drm_scheduler should *not* be required to get Xe accepted, but
+some consensus needs to be reached between Xe and other community drivers that
+could also benefit from this work, for coupling FW based/assisted submission such
+as the ARM’s new Mali GPU driver, and others.
+
+As a key measurable result, the patch series introducing Xe itself shall not
+depend on any other patch touching drm_scheduler itself that was not yet merged
+through drm-misc. This, by itself, already includes the reach of an agreement for
+uniform 1 to 1 relationship implementation / usage across drivers.
+
+GPU VA
+------
+Two main goals of Xe are meeting together here:
+
+1) Have an uAPI that aligns with modern UMD needs.
+
+2) Early upstream engagement.
+
+RedHat engineers working on Nouveau proposed a new DRM feature to handle keeping
+track of GPU virtual address mappings. This is still not merged upstream, but
+this aligns very well with our goals and with our VM_BIND. The engagement with
+upstream and the port of Xe towards GPUVA is already ongoing.
+
+As a key measurable result, Xe needs to be aligned with the GPU VA and working in
+our tree. Missing Nouveau patches should *not* block Xe and any needed GPUVA
+related patch should be independent and present on dri-devel or acked by
+maintainers to go along with the first Xe pull request towards drm-next.
+
+DRM_VM_BIND
+-----------
+Nouveau, and Xe are all implementing ‘VM_BIND’ and new ‘Exec’ uAPIs in order to
+fulfill the needs of the modern uAPI. Xe merge should *not* be blocked on the
+development of a common new drm_infrastructure. However, the Xe team needs to
+engage with the community to explore the options of a common API.
+
+As a key measurable result, the DRM_VM_BIND needs to be documented in this file
+below, or this entire block deleted if the consensus is for independent drivers
+vm_bind ioctls.
+
+Although having a common DRM level IOCTL for VM_BIND is not a requirement to get
+Xe merged, it is mandatory to enforce the overall locking scheme for all major
+structs and list (so vm and vma). So, a consensus is needed, and possibly some
+common helpers. If helpers are needed, they should be also documented in this
+document.
+
+ASYNC VM_BIND
+-------------
+Although having a common DRM level IOCTL for VM_BIND is not a requirement to get
+Xe merged, it is mandatory to have a consensus with other drivers and Mesa.
+It needs to be clear how to handle async VM_BIND and interactions with userspace
+memory fences. Ideally with helper support so people don't get it wrong in all
+possible ways.
+
+As a key measurable result, the benefits of ASYNC VM_BIND and a discussion of
+various flavors, error handling and a sample API should be documented here or in
+a separate document pointed to by this document.
+
+Userptr integration and vm_bind
+-------------------------------
+Different drivers implement different ways of dealing with execution of userptr.
+With multiple drivers currently introducing support to VM_BIND, the goal is to
+aim for a DRM consensus on what’s the best way to have that support. To some
+extent this is already getting addressed itself with the GPUVA where likely the
+userptr will be a GPUVA with a NULL GEM call VM bind directly on the userptr.
+However, there are more aspects around the rules for that and the usage of
+mmu_notifiers, locking and other aspects.
+
+This task here has the goal of introducing a documentation of the basic rules.
+
+The documentation *needs* to first live in this document (API session below) and
+then moved to another more specific document or at Xe level or at DRM level.
+
+Documentation should include:
+
+ * The userptr part of the VM_BIND api.
+
+ * Locking, including the page-faulting case.
+
+ * O(1) complexity under VM_BIND.
+
+Some parts of userptr like mmu_notifiers should become GPUVA or DRM helpers when
+the second driver supporting VM_BIND+userptr appears. Details to be defined when
+the time comes.
+
+Long running compute: minimal data structure/scaffolding
+--------------------------------------------------------
+The generic scheduler code needs to include the handling of endless compute
+contexts, with the minimal scaffolding for preempt-ctx fences (probably on the
+drm_sched_entity) and making sure drm_scheduler can cope with the lack of job
+completion fence.
+
+The goal is to achieve a consensus ahead of Xe initial pull-request, ideally with
+this minimal drm/scheduler work, if needed, merged to drm-misc in a way that any
+drm driver, including Xe, could re-use and add their own individual needs on top
+in a next stage. However, this should not block the initial merge.
+
+This is a non-blocker item since the driver without the support for the long
+running compute enabled is not a showstopper.
+
+Display integration with i915
+-----------------------------
+In order to share the display code with the i915 driver so that there is maximum
+reuse, the i915/display/ code is built twice, once for i915.ko and then for
+xe.ko. Currently, the i915/display code in Xe tree is polluted with many 'ifdefs'
+depending on the build target. The goal is to refactor both Xe and i915/display
+code simultaneously in order to get a clean result before they land upstream, so
+that display can already be part of the initial pull request towards drm-next.
+
+However, display code should not gate the acceptance of Xe in upstream. Xe
+patches will be refactored in a way that display code can be removed, if needed,
+from the first pull request of Xe towards drm-next. The expectation is that when
+both drivers are part of the drm-tip, the introduction of cleaner patches will be
+easier and speed up.
+
+Drm_exec
+--------
+Helper to make dma_resv locking for a big number of buffers is getting removed in
+the drm_exec series proposed in https://patchwork.freedesktop.org/patch/524376/
+If that happens, Xe needs to change and incorporate the changes in the driver.
+The goal is to engage with the Community to understand if the best approach is to
+move that to the drivers that are using it or if we should keep the helpers in
+place waiting for Xe to get merged.
+
+This item ties into the GPUVA, VM_BIND, and even long-running compute support.
+
+As a key measurable result, we need to have a community consensus documented in
+this document and the Xe driver prepared for the changes, if necessary.
+
+Dev_coredump
+------------
+
+Xe needs to align with other drivers on the way that the error states are
+dumped, avoiding a Xe only error_state solution. The goal is to use devcoredump
+infrastructure to report error states, since it produces a standardized way
+by exposing a virtual and temporary /sys/class/devcoredump device.
+
+As the key measurable result, Xe driver needs to provide GPU snapshots captured
+at hang time through devcoredump, but without depending on any core modification
+of devcoredump infrastructure itself.
+
+Later, when we are in-tree, the goal is to collaborate with devcoredump
+infrastructure with overall possible improvements, like multiple file support
+for better organization of the dumps, snapshot support, dmesg extra print,
+and whatever may make sense and help the overall infrastructure.
+
+Xe – uAPI high level overview
+=============================
+
+...Warning: To be done in follow up patches after/when/where the main consensus in various items are individually reached.
-- 
GitLab


From 0602d59376d391c460925daa3f8ba2b286cdcb4b Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Tue, 18 Apr 2023 18:44:25 +0530
Subject: [PATCH 0322/1791] drm/i915/display: Increase AUX timeout for Type-C

Type-C PHYs are taking longer than expected for Aux IO Power Enabling.
Workaround: Increase the timeout.

---v2
-change style on how we mention WA [Ankit]
-fix bat error by creating new func that is only called for aux power
well scenarios so we can avoid null pointer error as it is called
everywhere.

--v3
-Add non-default enable_timeout to power well descriptor which avoids
adding more platform checks [Imre]

--v4
-Remove Bspec link from top to bottom remove WA link from commit put it
on comment [Jani]
-enable_timeout in ms and add .fixed_enable_delay too [Imre]

--v5
-move power_wells instead of duplicating them [Imre]

Bspec: 55480

Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418131425.1285088-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_power_map.c  | 7 +++++++
 drivers/gpu/drm/i915/display/intel_display_power_well.c | 3 ++-
 drivers/gpu/drm/i915/display/intel_display_power_well.h | 2 ++
 3 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index ca448359a8226..1118ee9d224ca 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1387,6 +1387,11 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = {
 			I915_PW("AUX_C", &icl_pwdoms_aux_c, .hsw.idx = ICL_PW_CTL_IDX_AUX_C),
 			I915_PW("AUX_D", &icl_pwdoms_aux_d, .hsw.idx = XELPD_PW_CTL_IDX_AUX_D),
 			I915_PW("AUX_E", &icl_pwdoms_aux_e, .hsw.idx = XELPD_PW_CTL_IDX_AUX_E),
+		),
+		.ops = &icl_aux_power_well_ops,
+		.fixed_enable_delay = true,
+	}, {
+		.instances = &I915_PW_INSTANCES(
 			I915_PW("AUX_USBC1", &tgl_pwdoms_aux_usbc1, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC1),
 			I915_PW("AUX_USBC2", &tgl_pwdoms_aux_usbc2, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC2),
 			I915_PW("AUX_USBC3", &tgl_pwdoms_aux_usbc3, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC3),
@@ -1394,6 +1399,8 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = {
 		),
 		.ops = &icl_aux_power_well_ops,
 		.fixed_enable_delay = true,
+		/* WA_14017248603: adlp */
+		.enable_timeout = 500,
 	}, {
 		.instances = &I915_PW_INSTANCES(
 			I915_PW("AUX_TBT1", &icl_pwdoms_aux_tbt1, .hsw.idx = TGL_PW_CTL_IDX_AUX_TBT1),
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 1a19fd3bf103e..41eabdf3e8710 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -254,6 +254,7 @@ static void hsw_wait_for_power_well_enable(struct drm_i915_private *dev_priv,
 {
 	const struct i915_power_well_regs *regs = power_well->desc->ops->regs;
 	int pw_idx = i915_power_well_instance(power_well)->hsw.idx;
+	int timeout = power_well->desc->enable_timeout ? : 1;
 
 	/*
 	 * For some power wells we're not supposed to watch the status bit for
@@ -267,7 +268,7 @@ static void hsw_wait_for_power_well_enable(struct drm_i915_private *dev_priv,
 
 	/* Timeout for PW1:10 us, AUX:not specified, other PWs:20 us. */
 	if (intel_de_wait_for_set(dev_priv, regs->driver,
-				  HSW_PWR_WELL_CTL_STATE(pw_idx), 1)) {
+				  HSW_PWR_WELL_CTL_STATE(pw_idx), timeout)) {
 		drm_dbg_kms(&dev_priv->drm, "%s power well enable timeout\n",
 			    intel_power_well_name(power_well));
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.h b/drivers/gpu/drm/i915/display/intel_display_power_well.h
index ba7cb977e7c7f..e494df379e6c2 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.h
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.h
@@ -110,6 +110,8 @@ struct i915_power_well_desc {
 	 * Thunderbolt mode.
 	 */
 	u16 is_tc_tbt:1;
+	/* Enable timeout if greater than the default 1ms */
+	u16 enable_timeout;
 };
 
 struct i915_power_well {
-- 
GitLab


From e41fef1e43ae86aecc466e8c9e449d7bf71596a7 Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Wed, 26 Apr 2023 09:32:11 -0500
Subject: [PATCH 0323/1791] dt-bindings: panel: Add Anbernic RG353V-V2 panel
 compatible
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The Anbernic RG353V-V2 panel is a 3.5 inch 640x480 MIPI-DSI LCD panel.
It's based on the ST7703 LCD controller just like rocktech,jh057n00900.
It's used in a 2nd revision of the Anbernic RG353V handheld gaming
device. Like the first revision of the RG353V the control chip is known
but the panel itself is unknown, so name it for the device.

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Guido Günther <agx@sigxcpu.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426143213.4178586-2-macroalpha82@gmail.com
---
 .../devicetree/bindings/display/panel/rocktech,jh057n00900.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/rocktech,jh057n00900.yaml b/Documentation/devicetree/bindings/display/panel/rocktech,jh057n00900.yaml
index 09b5eb7542f8a..150e81090af2d 100644
--- a/Documentation/devicetree/bindings/display/panel/rocktech,jh057n00900.yaml
+++ b/Documentation/devicetree/bindings/display/panel/rocktech,jh057n00900.yaml
@@ -20,6 +20,8 @@ allOf:
 properties:
   compatible:
     enum:
+      # Anberic RG353V-V2 5.0" 640x480 TFT LCD panel
+      - anbernic,rg353v-panel-v2
       # Rocktech JH057N00900 5.5" 720x1440 TFT LCD panel
       - rocktech,jh057n00900
       # Xingbangda XBD599 5.99" 720x1440 TFT LCD panel
-- 
GitLab


From e0310564a7e31d7359a6831226aa9b559c6728a0 Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Wed, 26 Apr 2023 09:32:12 -0500
Subject: [PATCH 0324/1791] drm/panel: st7703: Rename CMD_UNKNOWN_C6 to
 CMD_SETECO
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A later revision of the datasheet for the ST7703 refers to this command
as "SETECO".

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Reviewed-by: Guido Günther <agx@sigxcpu.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426143213.4178586-3-macroalpha82@gmail.com
---
 drivers/gpu/drm/panel/panel-sitronix-st7703.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7703.c b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
index 6747ca237ced0..fc55b5fbb67f0 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7703.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
@@ -41,8 +41,8 @@
 #define ST7703_CMD_UNKNOWN_BF	 0xBF
 #define ST7703_CMD_SETSCR	 0xC0
 #define ST7703_CMD_SETPOWER	 0xC1
+#define ST7703_CMD_SETECO	 0xC6
 #define ST7703_CMD_SETPANEL	 0xCC
-#define ST7703_CMD_UNKNOWN_C6	 0xC6
 #define ST7703_CMD_SETGAMMA	 0xE0
 #define ST7703_CMD_SETEQ	 0xE3
 #define ST7703_CMD_SETGIP1	 0xE9
@@ -249,8 +249,7 @@ static int xbd599_init_sequence(struct st7703 *ctx)
 				      * ESD_DET_TIME_SEL = 0 frames
 				      */);
 
-	/* Undocumented command. */
-	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_UNKNOWN_C6, 0x01, 0x00, 0xFF, 0xFF, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETECO, 0x01, 0x00, 0xFF, 0xFF, 0x00);
 
 	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETPOWER,
 			       0x74, /* VBTHS, VBTLS: VGH = 17V, VBL = -11V */
-- 
GitLab


From 0ed9208cb120086f1b600199c548f00544406fa5 Mon Sep 17 00:00:00 2001
From: Chris Morgan <macromorgan@hotmail.com>
Date: Wed, 26 Apr 2023 09:32:13 -0500
Subject: [PATCH 0325/1791] drm/panel: st7703: Add Anbernic RG353V-V2 Panel
 Support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The Anbernic RG353V-V2 is a 5 inch panel used in a new revision of the
Anbernic RG353V handheld gaming device. Add support for it.

Unfortunately it appears this controller is not able to support 120hz
or 100hz mode like the first revision panel.

Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Reviewed-by: Guido Günther <agx@sigxcpu.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426143213.4178586-4-macroalpha82@gmail.com
---
 drivers/gpu/drm/panel/panel-sitronix-st7703.c | 97 +++++++++++++++++++
 1 file changed, 97 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7703.c b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
index fc55b5fbb67f0..3aa31f3d61574 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7703.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
@@ -28,6 +28,7 @@
 /* Manufacturer specific Commands send via DSI */
 #define ST7703_CMD_ALL_PIXEL_OFF 0x22
 #define ST7703_CMD_ALL_PIXEL_ON	 0x23
+#define ST7703_CMD_SETAPID	 0xB1
 #define ST7703_CMD_SETDISP	 0xB2
 #define ST7703_CMD_SETRGBIF	 0xB3
 #define ST7703_CMD_SETCYC	 0xB4
@@ -42,11 +43,14 @@
 #define ST7703_CMD_SETSCR	 0xC0
 #define ST7703_CMD_SETPOWER	 0xC1
 #define ST7703_CMD_SETECO	 0xC6
+#define ST7703_CMD_SETIO	 0xC7
+#define ST7703_CMD_SETCABC	 0xC8
 #define ST7703_CMD_SETPANEL	 0xCC
 #define ST7703_CMD_SETGAMMA	 0xE0
 #define ST7703_CMD_SETEQ	 0xE3
 #define ST7703_CMD_SETGIP1	 0xE9
 #define ST7703_CMD_SETGIP2	 0xEA
+#define ST7703_CMD_UNKNOWN_EF	 0xEF
 
 struct st7703 {
 	struct device *dev;
@@ -337,6 +341,98 @@ static const struct st7703_panel_desc xbd599_desc = {
 	.init_sequence = xbd599_init_sequence,
 };
 
+static int rg353v2_init_sequence(struct st7703 *ctx)
+{
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev);
+
+	/*
+	 * Init sequence was supplied by the panel vendor.
+	 */
+
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETEXTC, 0xf1, 0x12, 0x83);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETAPID, 0x00, 0x00, 0x00,
+			       0xda, 0x80);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETDISP, 0x00, 0x13, 0x70);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETRGBIF, 0x10, 0x10, 0x28,
+			       0x28, 0x03, 0xff, 0x00, 0x00, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETCYC, 0x80);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETBGP, 0x0a, 0x0a);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETVCOM, 0x92, 0x92);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETPOWER_EXT, 0x25, 0x22,
+			       0xf0, 0x63);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETMIPI, 0x33, 0x81, 0x05,
+			       0xf9, 0x0e, 0x0e, 0x20, 0x00, 0x00, 0x00, 0x00,
+			       0x00, 0x00, 0x00, 0x44, 0x25, 0x00, 0x90, 0x0a,
+			       0x00, 0x00, 0x01, 0x4f, 0x01, 0x00, 0x00, 0x37);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETVDC, 0x47);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_UNKNOWN_BF, 0x02, 0x11, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETSCR, 0x73, 0x73, 0x50, 0x50,
+			       0x00, 0x00, 0x12, 0x50, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETPOWER, 0x53, 0xc0, 0x32,
+			       0x32, 0x77, 0xe1, 0xdd, 0xdd, 0x77, 0x77, 0x33,
+			       0x33);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETECO, 0x82, 0x00, 0xbf, 0xff,
+			       0x00, 0xff);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETIO, 0xb8, 0x00, 0x0a, 0x00,
+			       0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETCABC, 0x10, 0x40, 0x1e,
+			       0x02);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETPANEL, 0x0b);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETGAMMA, 0x00, 0x07, 0x0d,
+			       0x37, 0x35, 0x3f, 0x41, 0x44, 0x06, 0x0c, 0x0d,
+			       0x0f, 0x11, 0x10, 0x12, 0x14, 0x1a, 0x00, 0x07,
+			       0x0d, 0x37, 0x35, 0x3f, 0x41, 0x44, 0x06, 0x0c,
+			       0x0d, 0x0f, 0x11, 0x10, 0x12, 0x14, 0x1a);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETEQ, 0x07, 0x07, 0x0b, 0x0b,
+			       0x0b, 0x0b, 0x00, 0x00, 0x00, 0x00, 0xff, 0x00,
+			       0xc0, 0x10);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETGIP1, 0xc8, 0x10, 0x02, 0x00,
+			       0x00, 0xb0, 0xb1, 0x11, 0x31, 0x23, 0x28, 0x80,
+			       0xb0, 0xb1, 0x27, 0x08, 0x00, 0x04, 0x02, 0x00,
+			       0x00, 0x00, 0x00, 0x04, 0x02, 0x00, 0x00, 0x00,
+			       0x88, 0x88, 0xba, 0x60, 0x24, 0x08, 0x88, 0x88,
+			       0x88, 0x88, 0x88, 0x88, 0x88, 0xba, 0x71, 0x35,
+			       0x18, 0x88, 0x88, 0x88, 0x88, 0x88, 0x00, 0x00,
+			       0x00, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x00, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_SETGIP2, 0x97, 0x0a, 0x82, 0x02,
+			       0x03, 0x07, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x81, 0x88, 0xba, 0x17, 0x53, 0x88, 0x88, 0x88,
+			       0x88, 0x88, 0x88, 0x80, 0x88, 0xba, 0x06, 0x42,
+			       0x88, 0x88, 0x88, 0x88, 0x88, 0x88, 0x23, 0x00,
+			       0x00, 0x02, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x00);
+	mipi_dsi_dcs_write_seq(dsi, ST7703_CMD_UNKNOWN_EF, 0xff, 0xff, 0x01);
+
+	return 0;
+}
+
+static const struct drm_display_mode rg353v2_mode = {
+	.hdisplay	= 640,
+	.hsync_start	= 640 + 40,
+	.hsync_end	= 640 + 40 + 2,
+	.htotal		= 640 + 40 + 2 + 80,
+	.vdisplay	= 480,
+	.vsync_start	= 480 + 18,
+	.vsync_end	= 480 + 18 + 2,
+	.vtotal		= 480 + 18 + 2 + 28,
+	.clock		= 24150,
+	.flags		= DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+	.width_mm	= 70,
+	.height_mm	= 57,
+};
+
+static const struct st7703_panel_desc rg353v2_desc = {
+	.mode = &rg353v2_mode,
+	.lanes = 4,
+	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+		      MIPI_DSI_MODE_NO_EOT_PACKET | MIPI_DSI_MODE_LPM,
+	.format = MIPI_DSI_FMT_RGB888,
+	.init_sequence = rg353v2_init_sequence,
+};
+
 static int st7703_enable(struct drm_panel *panel)
 {
 	struct st7703 *ctx = panel_to_st7703(panel);
@@ -597,6 +693,7 @@ static void st7703_remove(struct mipi_dsi_device *dsi)
 }
 
 static const struct of_device_id st7703_of_match[] = {
+	{ .compatible = "anbernic,rg353v-panel-v2", .data = &rg353v2_desc },
 	{ .compatible = "rocktech,jh057n00900", .data = &jh057n00900_panel_desc },
 	{ .compatible = "xingbangda,xbd599", .data = &xbd599_desc },
 	{ /* sentinel */ }
-- 
GitLab


From 4b733b2c38b7c9304bbb92cf88f1b23f43ca575d Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Sun, 23 Apr 2023 10:53:00 -0400
Subject: [PATCH 0326/1791] drm/meson: set variables meson_hdmi_*
 storage-class-specifier to static

smatch has several similar warnings to
drivers/gpu/drm/meson/meson_venc.c:189:28: warning: symbol
  'meson_hdmi_enci_mode_480i' was not declared. Should it be static?

These variables are only used in their defining file so should be static

Signed-off-by: Tom Rix <trix@redhat.com>
Acked-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230423145300.3937831-1-trix@redhat.com
---
 drivers/gpu/drm/meson/meson_venc.c | 32 +++++++++++++++---------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index fcd532db19c18..27ef9f88e4ff8 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -186,7 +186,7 @@ union meson_hdmi_venc_mode {
 	} encp;
 };
 
-union meson_hdmi_venc_mode meson_hdmi_enci_mode_480i = {
+static union meson_hdmi_venc_mode meson_hdmi_enci_mode_480i = {
 	.enci = {
 		.hso_begin = 5,
 		.hso_end = 129,
@@ -206,7 +206,7 @@ union meson_hdmi_venc_mode meson_hdmi_enci_mode_480i = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_enci_mode_576i = {
+static union meson_hdmi_venc_mode meson_hdmi_enci_mode_576i = {
 	.enci = {
 		.hso_begin = 3,
 		.hso_end = 129,
@@ -226,7 +226,7 @@ union meson_hdmi_venc_mode meson_hdmi_enci_mode_576i = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_480p = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_480p = {
 	.encp = {
 		.dvi_settings = 0x21,
 		.video_mode = 0x4000,
@@ -272,7 +272,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_480p = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_576p = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_576p = {
 	.encp = {
 		.dvi_settings = 0x21,
 		.video_mode = 0x4000,
@@ -318,7 +318,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_576p = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p60 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p60 = {
 	.encp = {
 		.dvi_settings = 0x2029,
 		.video_mode = 0x4040,
@@ -360,7 +360,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p60 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p50 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p50 = {
 	.encp = {
 		.dvi_settings = 0x202d,
 		.video_mode = 0x4040,
@@ -405,7 +405,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_720p50 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i60 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i60 = {
 	.encp = {
 		.dvi_settings = 0x2029,
 		.video_mode = 0x5ffc,
@@ -454,7 +454,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i60 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i50 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i50 = {
 	.encp = {
 		.dvi_settings = 0x202d,
 		.video_mode = 0x5ffc,
@@ -503,7 +503,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080i50 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p24 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p24 = {
 	.encp = {
 		.dvi_settings = 0xd,
 		.video_mode = 0x4040,
@@ -552,7 +552,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p24 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p30 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p30 = {
 	.encp = {
 		.dvi_settings = 0x1,
 		.video_mode = 0x4040,
@@ -596,7 +596,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p30 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p50 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p50 = {
 	.encp = {
 		.dvi_settings = 0xd,
 		.video_mode = 0x4040,
@@ -644,7 +644,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p50 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p60 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p60 = {
 	.encp = {
 		.dvi_settings = 0x1,
 		.video_mode = 0x4040,
@@ -688,7 +688,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_1080p60 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p24 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p24 = {
 	.encp = {
 		.dvi_settings = 0x1,
 		.video_mode = 0x4040,
@@ -730,7 +730,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p24 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p25 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p25 = {
 	.encp = {
 		.dvi_settings = 0x1,
 		.video_mode = 0x4040,
@@ -772,7 +772,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p25 = {
 	},
 };
 
-union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p30 = {
+static union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p30 = {
 	.encp = {
 		.dvi_settings = 0x1,
 		.video_mode = 0x4040,
@@ -814,7 +814,7 @@ union meson_hdmi_venc_mode meson_hdmi_encp_mode_2160p30 = {
 	},
 };
 
-struct meson_hdmi_venc_vic_mode {
+static struct meson_hdmi_venc_vic_mode {
 	unsigned int vic;
 	union meson_hdmi_venc_mode *mode;
 } meson_hdmi_venc_vic_modes[] = {
-- 
GitLab


From fa83c12132f71302f7d4b02758dc0d46048d3f5f Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Tue, 25 Apr 2023 21:44:41 +0200
Subject: [PATCH 0327/1791] drm/i915/dsi: Use unconditional msleep() instead of
 intel_dsi_msleep()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The intel_dsi_msleep() helper skips sleeping if the MIPI-sequences have
a version of 3 or newer and the panel is in vid-mode.

This is based on the big comment around line 730 which starts with
"Panel enable/disable sequences from the VBT spec.", where
the "v3 video mode seq" column does not have any wait t# entries.

Checking the Windows driver shows that it does always honor
the VBT delays independent of the version of the VBT sequences.

Commit 6fdb335f1c9c ("drm/i915/dsi: Use unconditional msleep for
the panel_on_delay when there is no reset-deassert MIPI-sequence")
switched to a direct msleep() instead of intel_dsi_msleep()
when there is no MIPI_SEQ_DEASSERT_RESET sequence, to fix
the panel on an Acer Aspire Switch 10 E SW3-016 not turning on.

And now testing on a Nextbook Ares 8A shows that panel_on_delay
must always be honored otherwise the panel will not turn on.

Instead of only always using regular msleep() for panel_on_delay
do as Windows does and always use regular msleep() everywhere
were intel_dsi_msleep() is used and drop the intel_dsi_msleep()
helper.

Changes in v2:
- Replace all intel_dsi_msleep() calls instead of just
  the intel_dsi_msleep(panel_on_delay) call

Cc: stable@vger.kernel.org
Fixes: 6fdb335f1c9c ("drm/i915/dsi: Use unconditional msleep for the panel_on_delay when there is no reset-deassert MIPI-sequence")
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425194441.68086-1-hdegoede@redhat.com
---
 drivers/gpu/drm/i915/display/icl_dsi.c       |  2 +-
 drivers/gpu/drm/i915/display/intel_dsi_vbt.c | 11 ----------
 drivers/gpu/drm/i915/display/intel_dsi_vbt.h |  1 -
 drivers/gpu/drm/i915/display/vlv_dsi.c       | 22 +++++---------------
 4 files changed, 6 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index ad78148e07889..c9aeba0ecf911 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -1140,7 +1140,7 @@ static void gen11_dsi_powerup_panel(struct intel_encoder *encoder)
 
 	/* panel power on related mipi dsi vbt sequences */
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_ON);
-	intel_dsi_msleep(intel_dsi, intel_dsi->panel_on_delay);
+	msleep(intel_dsi->panel_on_delay);
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET);
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_INIT_OTP);
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DISPLAY_ON);
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
index 695b0d69a4cba..c7935ea498c44 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
@@ -763,17 +763,6 @@ void intel_dsi_vbt_exec_sequence(struct intel_dsi *intel_dsi,
 		gpiod_set_value_cansleep(intel_dsi->gpio_backlight, 0);
 }
 
-void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec)
-{
-	struct intel_connector *connector = intel_dsi->attached_connector;
-
-	/* For v3 VBTs in vid-mode the delays are part of the VBT sequences */
-	if (is_vid_mode(intel_dsi) && connector->panel.vbt.dsi.seq_version >= 3)
-		return;
-
-	msleep(msec);
-}
-
 void intel_dsi_log_params(struct intel_dsi *intel_dsi)
 {
 	struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev);
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt.h
index dc642c1fe7efd..468d873fab1ae 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.h
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.h
@@ -16,7 +16,6 @@ void intel_dsi_vbt_gpio_init(struct intel_dsi *intel_dsi, bool panel_is_on);
 void intel_dsi_vbt_gpio_cleanup(struct intel_dsi *intel_dsi);
 void intel_dsi_vbt_exec_sequence(struct intel_dsi *intel_dsi,
 				 enum mipi_seq seq_id);
-void intel_dsi_msleep(struct intel_dsi *intel_dsi, int msec);
 void intel_dsi_log_params(struct intel_dsi *intel_dsi);
 
 #endif /* __INTEL_DSI_VBT_H__ */
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index 028965ab442dc..61d008d4e5f1f 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -737,7 +737,6 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 {
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(pipe_config->uapi.crtc);
-	struct intel_connector *connector = to_intel_connector(conn_state->connector);
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
 	enum port port;
@@ -779,21 +778,10 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 	if (!IS_GEMINILAKE(dev_priv))
 		intel_dsi_prepare(encoder, pipe_config);
 
+	/* Give the panel time to power-on and then deassert its reset */
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_ON);
-
-	/*
-	 * Give the panel time to power-on and then deassert its reset.
-	 * Depending on the VBT MIPI sequences version the deassert-seq
-	 * may contain the necessary delay, intel_dsi_msleep() will skip
-	 * the delay in that case. If there is no deassert-seq, then an
-	 * unconditional msleep is used to give the panel time to power-on.
-	 */
-	if (connector->panel.vbt.dsi.sequence[MIPI_SEQ_DEASSERT_RESET]) {
-		intel_dsi_msleep(intel_dsi, intel_dsi->panel_on_delay);
-		intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET);
-	} else {
-		msleep(intel_dsi->panel_on_delay);
-	}
+	msleep(intel_dsi->panel_on_delay);
+	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET);
 
 	if (IS_GEMINILAKE(dev_priv)) {
 		glk_cold_boot = glk_dsi_enable_io(encoder);
@@ -827,7 +815,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 		msleep(20); /* XXX */
 		for_each_dsi_port(port, intel_dsi->ports)
 			dpi_send_cmd(intel_dsi, TURN_ON, false, port);
-		intel_dsi_msleep(intel_dsi, 100);
+		msleep(100);
 
 		intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DISPLAY_ON);
 
@@ -949,7 +937,7 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state,
 	/* Assert reset */
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_ASSERT_RESET);
 
-	intel_dsi_msleep(intel_dsi, intel_dsi->panel_off_delay);
+	msleep(intel_dsi->panel_off_delay);
 	intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_OFF);
 
 	intel_dsi->panel_power_off_time = ktime_get_boottime();
-- 
GitLab


From d59ce1ff8e6830fab60b688f33a89ef48c94b5d8 Mon Sep 17 00:00:00 2001
From: Jonathan Cavitt <jonathan.cavitt@intel.com>
Date: Wed, 26 Apr 2023 23:28:48 +0200
Subject: [PATCH 0328/1791] drm/i915: Migrate platform-dependent mock hugepage
 selftests to live

Convert the igt_mock_ppgtt_huge_fill and igt_mock_ppgtt_64K mock selftests
into live selftests as their requirements have recently become
platform-dependent. Additionally, apply necessary platform dependency
checks to these tests.

v8:
- handle properly 64K and 2M pages
v9:
- do not expect 64K pages if 2M are present
- fix hex printing
- obey commit message line limit

Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Co-developed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425-hugepage-migrate-v8-1-7868d54eaa27@intel.com
---
 .../gpu/drm/i915/gem/selftests/huge_pages.c   | 106 +++++++++++++-----
 1 file changed, 81 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
index defece0bcb811..cb5863f37f9d3 100644
--- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
@@ -695,8 +695,7 @@ static int igt_mock_ppgtt_misaligned_dma(void *arg)
 	return err;
 }
 
-static void close_object_list(struct list_head *objects,
-			      struct i915_ppgtt *ppgtt)
+static void close_object_list(struct list_head *objects)
 {
 	struct drm_i915_gem_object *obj, *on;
 
@@ -710,17 +709,36 @@ static void close_object_list(struct list_head *objects,
 	}
 }
 
-static int igt_mock_ppgtt_huge_fill(void *arg)
+static int igt_ppgtt_huge_fill(void *arg)
 {
-	struct i915_ppgtt *ppgtt = arg;
-	struct drm_i915_private *i915 = ppgtt->vm.i915;
-	unsigned long max_pages = ppgtt->vm.total >> PAGE_SHIFT;
+	struct drm_i915_private *i915 = arg;
+	unsigned int supported = RUNTIME_INFO(i915)->page_sizes;
+	bool has_pte64 = GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50);
+	struct i915_address_space *vm;
+	struct i915_gem_context *ctx;
+	unsigned long max_pages;
 	unsigned long page_num;
+	struct file *file;
 	bool single = false;
 	LIST_HEAD(objects);
 	IGT_TIMEOUT(end_time);
 	int err = -ENODEV;
 
+	if (supported == I915_GTT_PAGE_SIZE_4K)
+		return 0;
+
+	file = mock_file(i915);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	ctx = hugepage_ctx(i915, file);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out;
+	}
+	vm = i915_gem_context_get_eb_vm(ctx);
+	max_pages = vm->total >> PAGE_SHIFT;
+
 	for_each_prime_number_from(page_num, 1, max_pages) {
 		struct drm_i915_gem_object *obj;
 		u64 size = page_num << PAGE_SHIFT;
@@ -750,13 +768,14 @@ static int igt_mock_ppgtt_huge_fill(void *arg)
 
 		list_add(&obj->st_link, &objects);
 
-		vma = i915_vma_instance(obj, &ppgtt->vm, NULL);
+		vma = i915_vma_instance(obj, vm, NULL);
 		if (IS_ERR(vma)) {
 			err = PTR_ERR(vma);
 			break;
 		}
 
-		err = i915_vma_pin(vma, 0, 0, PIN_USER);
+		/* vma start must be aligned to BIT(21) to allow 2M PTEs */
+		err = i915_vma_pin(vma, 0, BIT(21), PIN_USER);
 		if (err)
 			break;
 
@@ -784,12 +803,13 @@ static int igt_mock_ppgtt_huge_fill(void *arg)
 		GEM_BUG_ON(!expected_gtt);
 		GEM_BUG_ON(size);
 
-		if (expected_gtt & I915_GTT_PAGE_SIZE_4K)
+		if (!has_pte64 && (obj->base.size < I915_GTT_PAGE_SIZE_2M ||
+				   expected_gtt & I915_GTT_PAGE_SIZE_2M))
 			expected_gtt &= ~I915_GTT_PAGE_SIZE_64K;
 
 		i915_vma_unpin(vma);
 
-		if (vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K) {
+		if (!has_pte64 && vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K) {
 			if (!IS_ALIGNED(vma->node.start,
 					I915_GTT_PAGE_SIZE_2M)) {
 				pr_err("node.start(%llx) not aligned to 2M\n",
@@ -808,7 +828,7 @@ static int igt_mock_ppgtt_huge_fill(void *arg)
 		}
 
 		if (vma->resource->page_sizes_gtt != expected_gtt) {
-			pr_err("gtt=%u, expected=%u, size=%zd, single=%s\n",
+			pr_err("gtt=%#x, expected=%#x, size=0x%zx, single=%s\n",
 			       vma->resource->page_sizes_gtt, expected_gtt,
 			       obj->base.size, str_yes_no(!!single));
 			err = -EINVAL;
@@ -823,19 +843,25 @@ static int igt_mock_ppgtt_huge_fill(void *arg)
 		single = !single;
 	}
 
-	close_object_list(&objects, ppgtt);
+	close_object_list(&objects);
 
 	if (err == -ENOMEM || err == -ENOSPC)
 		err = 0;
 
+	i915_vm_put(vm);
+out:
+	fput(file);
 	return err;
 }
 
-static int igt_mock_ppgtt_64K(void *arg)
+static int igt_ppgtt_64K(void *arg)
 {
-	struct i915_ppgtt *ppgtt = arg;
-	struct drm_i915_private *i915 = ppgtt->vm.i915;
+	struct drm_i915_private *i915 = arg;
+	bool has_pte64 = GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50);
 	struct drm_i915_gem_object *obj;
+	struct i915_address_space *vm;
+	struct i915_gem_context *ctx;
+	struct file *file;
 	const struct object_info {
 		unsigned int size;
 		unsigned int gtt;
@@ -907,16 +933,41 @@ static int igt_mock_ppgtt_64K(void *arg)
 	if (!HAS_PAGE_SIZES(i915, I915_GTT_PAGE_SIZE_64K))
 		return 0;
 
+	file = mock_file(i915);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
+	ctx = hugepage_ctx(i915, file);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out;
+	}
+	vm = i915_gem_context_get_eb_vm(ctx);
+
 	for (i = 0; i < ARRAY_SIZE(objects); ++i) {
 		unsigned int size = objects[i].size;
 		unsigned int expected_gtt = objects[i].gtt;
 		unsigned int offset = objects[i].offset;
 		unsigned int flags = PIN_USER;
 
+		/*
+		 * For modern GTT models, the requirements for marking a page-table
+		 * as 64K have been relaxed.  Account for this.
+		 */
+		if (has_pte64) {
+			expected_gtt = 0;
+			if (size >= SZ_64K)
+				expected_gtt |= I915_GTT_PAGE_SIZE_64K;
+			if (size & (SZ_64K - 1))
+				expected_gtt |= I915_GTT_PAGE_SIZE_4K;
+		}
+
 		for (single = 0; single <= 1; single++) {
 			obj = fake_huge_pages_object(i915, size, !!single);
-			if (IS_ERR(obj))
-				return PTR_ERR(obj);
+			if (IS_ERR(obj)) {
+				err = PTR_ERR(obj);
+				goto out_vm;
+			}
 
 			err = i915_gem_object_pin_pages_unlocked(obj);
 			if (err)
@@ -928,7 +979,7 @@ static int igt_mock_ppgtt_64K(void *arg)
 			 */
 			obj->mm.page_sizes.sg &= ~I915_GTT_PAGE_SIZE_2M;
 
-			vma = i915_vma_instance(obj, &ppgtt->vm, NULL);
+			vma = i915_vma_instance(obj, vm, NULL);
 			if (IS_ERR(vma)) {
 				err = PTR_ERR(vma);
 				goto out_object_unpin;
@@ -945,7 +996,8 @@ static int igt_mock_ppgtt_64K(void *arg)
 			if (err)
 				goto out_vma_unpin;
 
-			if (!offset && vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K) {
+			if (!has_pte64 && !offset &&
+			    vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K) {
 				if (!IS_ALIGNED(vma->node.start,
 						I915_GTT_PAGE_SIZE_2M)) {
 					pr_err("node.start(%llx) not aligned to 2M\n",
@@ -964,9 +1016,10 @@ static int igt_mock_ppgtt_64K(void *arg)
 			}
 
 			if (vma->resource->page_sizes_gtt != expected_gtt) {
-				pr_err("gtt=%u, expected=%u, i=%d, single=%s\n",
+				pr_err("gtt=%#x, expected=%#x, i=%d, single=%s offset=%#x size=%#x\n",
 				       vma->resource->page_sizes_gtt,
-				       expected_gtt, i, str_yes_no(!!single));
+				       expected_gtt, i, str_yes_no(!!single),
+				       offset, size);
 				err = -EINVAL;
 				goto out_vma_unpin;
 			}
@@ -982,7 +1035,7 @@ static int igt_mock_ppgtt_64K(void *arg)
 		}
 	}
 
-	return 0;
+	goto out_vm;
 
 out_vma_unpin:
 	i915_vma_unpin(vma);
@@ -992,7 +1045,10 @@ static int igt_mock_ppgtt_64K(void *arg)
 	i915_gem_object_unlock(obj);
 out_object_put:
 	i915_gem_object_put(obj);
-
+out_vm:
+	i915_vm_put(vm);
+out:
+	fput(file);
 	return err;
 }
 
@@ -1910,8 +1966,6 @@ int i915_gem_huge_page_mock_selftests(void)
 		SUBTEST(igt_mock_exhaust_device_supported_pages),
 		SUBTEST(igt_mock_memory_region_huge_pages),
 		SUBTEST(igt_mock_ppgtt_misaligned_dma),
-		SUBTEST(igt_mock_ppgtt_huge_fill),
-		SUBTEST(igt_mock_ppgtt_64K),
 	};
 	struct drm_i915_private *dev_priv;
 	struct i915_ppgtt *ppgtt;
@@ -1962,6 +2016,8 @@ int i915_gem_huge_page_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_ppgtt_sanity_check),
 		SUBTEST(igt_ppgtt_compact),
 		SUBTEST(igt_ppgtt_mixed),
+		SUBTEST(igt_ppgtt_huge_fill),
+		SUBTEST(igt_ppgtt_64K),
 	};
 
 	if (!HAS_PPGTT(i915)) {
-- 
GitLab


From eee44c2903e5222632536f86ad5cf0ff9bbfc61b Mon Sep 17 00:00:00 2001
From: Jonathan Cavitt <jonathan.cavitt@intel.com>
Date: Wed, 26 Apr 2023 23:28:49 +0200
Subject: [PATCH 0329/1791] drm/i915: Use correct huge page manager for MTL

MTL currently uses gen8_ppgtt_insert_huge when managing huge pages.
This is because MTL reports as not supporting 64K pages, or more
accurately, the system that reports whether a platform has 64K pages
reports false for MTL.  This is only half correct, as the 64K page support
reporting system only cares about 64K page support for LMEM, which MTL
doesn't have.

MTL should be using xehpsdv_ppgtt_insert_huge.  However, simply changing
over to using that manager doesn't resolve the issue because MTL is
expecting the virtual address space for the page table to be flushed after
initialization, so we must also add a flush statement there.

Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425-hugepage-migrate-v8-2-7868d54eaa27@intel.com
---
 drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 4c9a2f2db9081..22ec1566d2a7b 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -598,6 +598,7 @@ xehpsdv_ppgtt_insert_huge(struct i915_address_space *vm,
 			}
 		} while (rem >= page_size && index < max);
 
+		drm_clflush_virt_range(vaddr, PAGE_SIZE);
 		vma_res->page_sizes_gtt |= page_size;
 	} while (iter->sg && sg_dma_len(iter->sg));
 }
@@ -735,7 +736,7 @@ static void gen8_ppgtt_insert(struct i915_address_space *vm,
 	struct sgt_dma iter = sgt_dma(vma_res);
 
 	if (vma_res->bi.page_sizes.sg > I915_GTT_PAGE_SIZE) {
-		if (HAS_64K_PAGES(vm->i915))
+		if (GRAPHICS_VER_FULL(vm->i915) >= IP_VER(12, 50))
 			xehpsdv_ppgtt_insert_huge(vm, vma_res, &iter, cache_level, flags);
 		else
 			gen8_ppgtt_insert_huge(vm, vma_res, &iter, cache_level, flags);
-- 
GitLab


From 99e7e3b60080d913ff2f94943f4af1f1b76a1c06 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 12 Apr 2023 01:29:26 +0300
Subject: [PATCH 0330/1791] drm/uapi: Document CTM matrix better
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Document in which order the CTM matrix elements are stored.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230411222931.15127-2-ville.syrjala@linux.intel.com
Reviewed-by: Xaver Hugl <xaver.hugl@gmail.com>
Acked-by: Simon Ser <contact@emersion.fr>
---
 include/uapi/drm/drm_mode.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index 46becedf5b2fc..43691058d28fb 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -834,6 +834,11 @@ struct drm_color_ctm {
 	/*
 	 * Conversion matrix in S31.32 sign-magnitude
 	 * (not two's complement!) format.
+	 *
+	 * out   matrix    in
+	 * |R|   |0 1 2|   |R|
+	 * |G| = |3 4 5| x |G|
+	 * |B|   |6 7 8|   |B|
 	 */
 	__u64 matrix[9];
 };
-- 
GitLab


From 55f9720dbf23ed640a51ea5564c22305efa8a467 Mon Sep 17 00:00:00 2001
From: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Date: Tue, 25 Apr 2023 17:39:41 -0700
Subject: [PATCH 0331/1791] drm/i915/guc/slpc: Provide sysfs for efficient freq

SLPC enables use of efficient freq at init by default. It is
possible for GuC to request frequencies that are higher than
the 'software' max if user has set it lower than the efficient
level.

Scenarios/tests that require strict fixing of freq below the efficient
level will need to disable it through this interface.

v2: Keep just one interface to toggle sysfs. With this, user will
be completely responsible for toggling efficient frequency if need
be. There will be no implicit disabling when user sets min < RP1 (Ashutosh)

v3: Remove unused label, review comments (Ashutosh)

v4: Toggle efficient freq usage in SLPC selftest and checkpatch fixes

v5: Review comments (Andi) and add a separate patch for selftest updates

Fixes: 95ccf312a1e4 ("drm/i915/guc/slpc: Allow SLPC to use efficient frequency")
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426003942.1924347-1-vinay.belgaumkar@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c   | 35 +++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c   | 38 +++++++++++++------
 drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.h   |  1 +
 .../gpu/drm/i915/gt/uc/intel_guc_slpc_types.h |  1 +
 4 files changed, 64 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c
index 28f27091cd3b7..ee2b44f896a27 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c
@@ -451,6 +451,33 @@ static ssize_t punit_req_freq_mhz_show(struct kobject *kobj,
 	return sysfs_emit(buff, "%u\n", preq);
 }
 
+static ssize_t slpc_ignore_eff_freq_show(struct kobject *kobj,
+					 struct kobj_attribute *attr,
+					 char *buff)
+{
+	struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name);
+	struct intel_guc_slpc *slpc = &gt->uc.guc.slpc;
+
+	return sysfs_emit(buff, "%u\n", slpc->ignore_eff_freq);
+}
+
+static ssize_t slpc_ignore_eff_freq_store(struct kobject *kobj,
+					  struct kobj_attribute *attr,
+					  const char *buff, size_t count)
+{
+	struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name);
+	struct intel_guc_slpc *slpc = &gt->uc.guc.slpc;
+	int err;
+	u32 val;
+
+	err = kstrtou32(buff, 0, &val);
+	if (err)
+		return err;
+
+	err = intel_guc_slpc_set_ignore_eff_freq(slpc, val);
+	return err ?: count;
+}
+
 struct intel_gt_bool_throttle_attr {
 	struct attribute attr;
 	ssize_t (*show)(struct kobject *kobj, struct kobj_attribute *attr,
@@ -663,6 +690,8 @@ static struct kobj_attribute attr_media_freq_factor_scale =
 INTEL_GT_ATTR_RO(media_RP0_freq_mhz);
 INTEL_GT_ATTR_RO(media_RPn_freq_mhz);
 
+INTEL_GT_ATTR_RW(slpc_ignore_eff_freq);
+
 static const struct attribute *media_perf_power_attrs[] = {
 	&attr_media_freq_factor.attr,
 	&attr_media_freq_factor_scale.attr,
@@ -744,6 +773,12 @@ void intel_gt_sysfs_pm_init(struct intel_gt *gt, struct kobject *kobj)
 	if (ret)
 		gt_warn(gt, "failed to create punit_req_freq_mhz sysfs (%pe)", ERR_PTR(ret));
 
+	if (intel_uc_uses_guc_slpc(&gt->uc)) {
+		ret = sysfs_create_file(kobj, &attr_slpc_ignore_eff_freq.attr);
+		if (ret)
+			gt_warn(gt, "failed to create ignore_eff_freq sysfs (%pe)", ERR_PTR(ret));
+	}
+
 	if (i915_mmio_reg_valid(intel_gt_perf_limit_reasons_reg(gt))) {
 		ret = sysfs_create_files(kobj, throttle_reason_attrs);
 		if (ret)
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
index 026d73855f36c..56dbba1ef6684 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
@@ -277,6 +277,7 @@ int intel_guc_slpc_init(struct intel_guc_slpc *slpc)
 
 	slpc->max_freq_softlimit = 0;
 	slpc->min_freq_softlimit = 0;
+	slpc->ignore_eff_freq = false;
 	slpc->min_is_rpmax = false;
 
 	slpc->boost_freq = 0;
@@ -457,6 +458,29 @@ int intel_guc_slpc_get_max_freq(struct intel_guc_slpc *slpc, u32 *val)
 	return ret;
 }
 
+int intel_guc_slpc_set_ignore_eff_freq(struct intel_guc_slpc *slpc, bool val)
+{
+	struct drm_i915_private *i915 = slpc_to_i915(slpc);
+	intel_wakeref_t wakeref;
+	int ret;
+
+	mutex_lock(&slpc->lock);
+	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
+
+	ret = slpc_set_param(slpc,
+			     SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY,
+			     val);
+	if (ret)
+		guc_probe_error(slpc_to_guc(slpc), "Failed to set efficient freq(%d): %pe\n",
+				val, ERR_PTR(ret));
+	else
+		slpc->ignore_eff_freq = val;
+
+	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
+	mutex_unlock(&slpc->lock);
+	return ret;
+}
+
 /**
  * intel_guc_slpc_set_min_freq() - Set min frequency limit for SLPC.
  * @slpc: pointer to intel_guc_slpc.
@@ -482,16 +506,6 @@ int intel_guc_slpc_set_min_freq(struct intel_guc_slpc *slpc, u32 val)
 	mutex_lock(&slpc->lock);
 	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
 
-	/* Ignore efficient freq if lower min freq is requested */
-	ret = slpc_set_param(slpc,
-			     SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY,
-			     val < slpc->rp1_freq);
-	if (ret) {
-		guc_probe_error(slpc_to_guc(slpc), "Failed to toggle efficient freq: %pe\n",
-				ERR_PTR(ret));
-		goto out;
-	}
-
 	ret = slpc_set_param(slpc,
 			     SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ,
 			     val);
@@ -499,7 +513,6 @@ int intel_guc_slpc_set_min_freq(struct intel_guc_slpc *slpc, u32 val)
 	if (!ret)
 		slpc->min_freq_softlimit = val;
 
-out:
 	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 	mutex_unlock(&slpc->lock);
 
@@ -752,6 +765,9 @@ int intel_guc_slpc_enable(struct intel_guc_slpc *slpc)
 	/* Set cached media freq ratio mode */
 	intel_guc_slpc_set_media_ratio_mode(slpc, slpc->media_ratio_mode);
 
+	/* Set cached value of ignore efficient freq */
+	intel_guc_slpc_set_ignore_eff_freq(slpc, slpc->ignore_eff_freq);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.h
index 17ed515f6a852..597eb5413ddf2 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.h
@@ -46,5 +46,6 @@ void intel_guc_slpc_boost(struct intel_guc_slpc *slpc);
 void intel_guc_slpc_dec_waiters(struct intel_guc_slpc *slpc);
 int intel_guc_slpc_unset_gucrc_mode(struct intel_guc_slpc *slpc);
 int intel_guc_slpc_override_gucrc_mode(struct intel_guc_slpc *slpc, u32 mode);
+int intel_guc_slpc_set_ignore_eff_freq(struct intel_guc_slpc *slpc, bool val);
 
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc_types.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc_types.h
index a6ef53b04e047..a886513314977 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc_types.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc_types.h
@@ -31,6 +31,7 @@ struct intel_guc_slpc {
 	/* frequency softlimits */
 	u32 min_freq_softlimit;
 	u32 max_freq_softlimit;
+	bool ignore_eff_freq;
 
 	/* cached media ratio mode */
 	u32 media_ratio_mode;
-- 
GitLab


From c73bd1706c0cdb5627d312efdc656baf48e0247f Mon Sep 17 00:00:00 2001
From: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Date: Tue, 25 Apr 2023 17:39:42 -0700
Subject: [PATCH 0332/1791] drm/i915/selftest: Update the SLPC selftest

Use the new efficient frequency toggling interface. Also
create a helper function to restore the frequencies after
the test is done.

v2: Restore max freq first and then min.

Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426003942.1924347-2-vinay.belgaumkar@intel.com
---
 drivers/gpu/drm/i915/gt/selftest_slpc.c | 42 ++++++++++++++++++++++---
 1 file changed, 37 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_slpc.c b/drivers/gpu/drm/i915/gt/selftest_slpc.c
index bd44ce73a5044..952c8d52d68a2 100644
--- a/drivers/gpu/drm/i915/gt/selftest_slpc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_slpc.c
@@ -70,6 +70,31 @@ static int slpc_set_freq(struct intel_gt *gt, u32 freq)
 	return err;
 }
 
+static int slpc_restore_freq(struct intel_guc_slpc *slpc, u32 min, u32 max)
+{
+	int err;
+
+	err = slpc_set_max_freq(slpc, max);
+	if (err) {
+		pr_err("Unable to restore max freq");
+		return err;
+	}
+
+	err = slpc_set_min_freq(slpc, min);
+	if (err) {
+		pr_err("Unable to restore min freq");
+		return err;
+	}
+
+	err = intel_guc_slpc_set_ignore_eff_freq(slpc, false);
+	if (err) {
+		pr_err("Unable to restore efficient freq");
+		return err;
+	}
+
+	return 0;
+}
+
 static u64 measure_power_at_freq(struct intel_gt *gt, int *freq, u64 *power)
 {
 	int err = 0;
@@ -268,8 +293,7 @@ static int run_test(struct intel_gt *gt, int test_type)
 
 	/*
 	 * Set min frequency to RPn so that we can test the whole
-	 * range of RPn-RP0. This also turns off efficient freq
-	 * usage and makes results more predictable.
+	 * range of RPn-RP0.
 	 */
 	err = slpc_set_min_freq(slpc, slpc->min_freq);
 	if (err) {
@@ -277,6 +301,15 @@ static int run_test(struct intel_gt *gt, int test_type)
 		return err;
 	}
 
+	/*
+	 * Turn off efficient frequency so RPn/RP0 ranges are obeyed.
+	 */
+	err = intel_guc_slpc_set_ignore_eff_freq(slpc, true);
+	if (err) {
+		pr_err("Unable to turn off efficient freq!");
+		return err;
+	}
+
 	intel_gt_pm_wait_for_idle(gt);
 	intel_gt_pm_get(gt);
 	for_each_engine(engine, gt, id) {
@@ -358,9 +391,8 @@ static int run_test(struct intel_gt *gt, int test_type)
 			break;
 	}
 
-	/* Restore min/max frequencies */
-	slpc_set_max_freq(slpc, slpc_max_freq);
-	slpc_set_min_freq(slpc, slpc_min_freq);
+	/* Restore min/max/efficient frequencies */
+	err = slpc_restore_freq(slpc, slpc_min_freq, slpc_max_freq);
 
 	if (igt_flush_test(gt->i915))
 		err = -EIO;
-- 
GitLab


From 0be05a75de2916421e88e0d64b001984f54df0bd Mon Sep 17 00:00:00 2001
From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sun, 16 Apr 2023 17:30:46 +0200
Subject: [PATCH 0333/1791] drm/bridge: it6505: Move a variable assignment
 behind a null pointer check in receive_timing_debugfs_show()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The address of a data structure member was determined before
a corresponding null pointer check in the implementation of
the function “receive_timing_debugfs_show”.

Thus avoid the risk for undefined behaviour by moving the assignment
for the variable “vid” behind the null pointer check.

This issue was detected by using the Coccinelle software.

Fixes: b5c84a9edcd4 ("drm/bridge: add it6505 driver")
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Link: https://patchwork.freedesktop.org/patch/msgid/fa69384f-1485-142b-c4ee-3df54ac68a89@web.de
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
---
 drivers/gpu/drm/bridge/ite-it6505.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c
index abaf6e23775eb..45f579c365e7f 100644
--- a/drivers/gpu/drm/bridge/ite-it6505.c
+++ b/drivers/gpu/drm/bridge/ite-it6505.c
@@ -3207,7 +3207,7 @@ static ssize_t receive_timing_debugfs_show(struct file *file, char __user *buf,
 					   size_t len, loff_t *ppos)
 {
 	struct it6505 *it6505 = file->private_data;
-	struct drm_display_mode *vid = &it6505->video_info;
+	struct drm_display_mode *vid;
 	u8 read_buf[READ_BUFFER_SIZE];
 	u8 *str = read_buf, *end = read_buf + READ_BUFFER_SIZE;
 	ssize_t ret, count;
@@ -3216,6 +3216,7 @@ static ssize_t receive_timing_debugfs_show(struct file *file, char __user *buf,
 		return -ENODEV;
 
 	it6505_calc_video_info(it6505);
+	vid = &it6505->video_info;
 	str += scnprintf(str, end - str, "---video timing---\n");
 	str += scnprintf(str, end - str, "PCLK:%d.%03dMHz\n",
 			 vid->clock / 1000, vid->clock % 1000);
-- 
GitLab


From f0038cffa843697501c442eff4468e3ab37031f1 Mon Sep 17 00:00:00 2001
From: Won Chung <wonchung@google.com>
Date: Wed, 29 Mar 2023 01:44:55 +0000
Subject: [PATCH 0334/1791] drm/sysfs: Expose DRM connector id in each
 connector sysfs

Expose DRM connector id in device sysfs so that we can map the connector
id to the connector syspath. Currently, even if we can derive the
connector id from modeset, we do not have a way to find the
corresponding connector's syspath.

This is helpful when determining the root connector of MST tree. When a
tree of multiple MST hub is connected to the system, modeset describes
the tree in the PATH blob. For example, consider the following scenario.

+-------------+
| Source      |    +-------------+
| (Device)    |    | BranchX     |
|             |    | (MST)       |
|       [conn6]--->|       [port1]--->DisplayA
+-------------+    |             |
                   |             |    +-------------+
                   |             |    | BranchY     |
                   |             |    | (MST)       |
                   |       [port2]--->|       [port1]----->DisplayB
                   +-------------+    |             |
                                      |       [port2]----->DisplayC
                                      +-------------+

DPMST connector of DisplayA would have "mst:6-1" PATH.
DPMST connector of DisplayB would have "mst:6-2-1" PATH.
DPMST connector of DisplayC would have "mst:6-2-2" PATH.

Given that connector id of 6 is the root of the MST connector tree, we
can utilize this patch to parse through DRM connectors sysfs and find
which connector syspath corresponds to the root connector (id == 6).

ChromeOS intend to use this information for metrics collection. For
example, we want to tell which port is deriving which displays even with
a MST hub. Chromium patch for parsing DRM connector id from the kernel
is at http://crrev.com/c/4317207.

Signed-off-by: Won Chung <wonchung@google.com>
Reviewed-by: Manasi Navare <navaremanasi@chromium.org>
Signed-off-by: Manasi Navare <navaremanasi@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230329014455.1990104-1-wonchung@google.com
---
 drivers/gpu/drm/drm_sysfs.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 1831303559972..11f98c5d61032 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -282,16 +282,27 @@ static ssize_t modes_show(struct device *device,
 	return written;
 }
 
+static ssize_t connector_id_show(struct device *device,
+				 struct device_attribute *attr,
+				 char *buf)
+{
+	struct drm_connector *connector = to_drm_connector(device);
+
+	return sysfs_emit(buf, "%d\n", connector->base.id);
+}
+
 static DEVICE_ATTR_RW(status);
 static DEVICE_ATTR_RO(enabled);
 static DEVICE_ATTR_RO(dpms);
 static DEVICE_ATTR_RO(modes);
+static DEVICE_ATTR_RO(connector_id);
 
 static struct attribute *connector_dev_attrs[] = {
 	&dev_attr_status.attr,
 	&dev_attr_enabled.attr,
 	&dev_attr_dpms.attr,
 	&dev_attr_modes.attr,
+	&dev_attr_connector_id.attr,
 	NULL
 };
 
-- 
GitLab


From c5c51b2420625faa1f0e363f21dba1de53806ff7 Mon Sep 17 00:00:00 2001
From: Won Chung <wonchung@google.com>
Date: Thu, 27 Apr 2023 16:58:13 +0000
Subject: [PATCH 0335/1791] drm/sysfs: Link DRM connectors to corresponding
 Type-C connectors

Create a symlink pointing to USB Type-C connector for DRM connectors
when they are created. The link will be created only if the firmware is
able to describe the connection beween the two connectors.

Currently, even if a display uses a USB Type-C port, there is no way for
the userspace to find which port is used for which display. With the
symlink, display information would be accessible from Type-C connectors
and port information would be accessible from DRM connectors.

Associating the two subsystems, userspace would have potential to expose
and utilize more complex information. ChromeOS intend to use this
information for metrics collection. For example, we want to tell which
port is deriving which displays. Also, combined with USB PD information,
we can tell whether user is charging their device through display.
Chromium patch for parsing the symlink from the kernel is at
http://crrev.com/c/4317207.

We already have a framework in typec port-mapper.c where it goes through
component devices and runs the bind functions for those with matching
_PLD (physical location of device).
https://elixir.bootlin.com/linux/v5.18.1/source/drivers/usb/typec/
port-mapper.c
Since _PLD is ACPI specific field, this linking would only work on ACPI
x86 as long as _PLD field for Type-C connectors and DRM connectors are
correctly added to the firmware.

Currently, USB ports and USB4 ports are added as components to create a
symlink with Type C connector.
USB:
https://lore.kernel.org/all/20211223082349.45616-1-heikki.krogerus
@linux.intel.com/
USB4:
https://lore.kernel.org/all/20220418175932.1809770-3-wonchung@google.com/
So, we follow the same pattern in this patch.

Signed-off-by: Won Chung <wonchung@google.com>
Acked-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: Manasi Navare <navaremanasi@chromium.org>
Signed-off-by: Manasi Navare <navaremanasi@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427165813.2844530-1-wonchung@google.com
---
 drivers/gpu/drm/drm_sysfs.c | 40 +++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 11f98c5d61032..62ce80b3eebce 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -11,12 +11,14 @@
  */
 
 #include <linux/acpi.h>
+#include <linux/component.h>
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/export.h>
 #include <linux/gfp.h>
 #include <linux/i2c.h>
 #include <linux/kdev_t.h>
+#include <linux/property.h>
 #include <linux/slab.h>
 
 #include <drm/drm_accel.h>
@@ -96,6 +98,34 @@ static char *drm_devnode(const struct device *dev, umode_t *mode)
 	return kasprintf(GFP_KERNEL, "dri/%s", dev_name(dev));
 }
 
+static int typec_connector_bind(struct device *dev,
+				struct device *typec_connector, void *data)
+{
+	int ret;
+
+	ret = sysfs_create_link(&dev->kobj, &typec_connector->kobj, "typec_connector");
+	if (ret)
+		return ret;
+
+	ret = sysfs_create_link(&typec_connector->kobj, &dev->kobj, "drm_connector");
+	if (ret)
+		sysfs_remove_link(&dev->kobj, "typec_connector");
+
+	return ret;
+}
+
+static void typec_connector_unbind(struct device *dev,
+				   struct device *typec_connector, void *data)
+{
+	sysfs_remove_link(&typec_connector->kobj, "drm_connector");
+	sysfs_remove_link(&dev->kobj, "typec_connector");
+}
+
+static const struct component_ops typec_connector_ops = {
+	.bind = typec_connector_bind,
+	.unbind = typec_connector_unbind,
+};
+
 static CLASS_ATTR_STRING(version, S_IRUGO, "drm 1.1.0 20060810");
 
 /**
@@ -364,9 +394,16 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 
 	connector->kdev = kdev;
 
+	if (dev_fwnode(kdev)) {
+		r = component_add(kdev, &typec_connector_ops);
+		if (r)
+			drm_err(dev, "failed to add component to create link to typec connector\n");
+	}
+
 	if (connector->ddc)
 		return sysfs_create_link(&connector->kdev->kobj,
 				 &connector->ddc->dev.kobj, "ddc");
+
 	return 0;
 
 err_free:
@@ -382,6 +419,9 @@ void drm_sysfs_connector_remove(struct drm_connector *connector)
 	if (connector->ddc)
 		sysfs_remove_link(&connector->kdev->kobj, "ddc");
 
+	if (dev_fwnode(connector->kdev))
+		component_del(connector->kdev, &typec_connector_ops);
+
 	DRM_DEBUG("removing \"%s\" from sysfs\n",
 		  connector->name);
 
-- 
GitLab


From 62618c7f117eedfd99b2f857885ed004d31df739 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:21 +0300
Subject: [PATCH 0336/1791] drm/i915/mtl: C20 PLL programming
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

C20 phy PLL programming sequence for DP, DP2.0, HDMI2.x non-FRL and
HDMI2.x FRL. This enables C20 MPLLA and MPLLB programming sequence. add
4 lane support for c20.

v2: Add 6.48Gbps and 6.75Gbps modes for eDP (RK)
    Fix lane check (RK)
    Fix multiline commenting (Arun)
    use usleep_range() instead of msleep() (Andi)

Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-2-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 288 +++++++++++++++---
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  33 ++
 drivers/gpu/drm/i915/display/intel_ddi.c      |   3 +-
 .../drm/i915/display/intel_display_types.h    |  15 +-
 drivers/gpu/drm/i915/display/intel_dp.c       |  12 +-
 5 files changed, 309 insertions(+), 42 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 83180074b5127..71163bc5bbf58 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -273,6 +273,18 @@ static void intel_cx0_write(struct drm_i915_private *i915, enum port port,
 		__intel_cx0_write(i915, port, lane, addr, data, committed);
 }
 
+static void intel_c20_sram_write(struct drm_i915_private *i915, enum port port,
+				 int lane, u16 addr, u16 data)
+{
+	assert_dc_off(i915);
+
+	intel_cx0_write(i915, port, lane, PHY_C20_WR_ADDRESS_H, addr >> 8, 0);
+	intel_cx0_write(i915, port, lane, PHY_C20_WR_ADDRESS_L, addr & 0xff, 0);
+
+	intel_cx0_write(i915, port, lane, PHY_C20_WR_DATA_H, data >> 8, 0);
+	intel_cx0_write(i915, port, lane, PHY_C20_WR_DATA_L, data & 0xff, 1);
+}
+
 static void __intel_cx0_rmw(struct drm_i915_private *i915, enum port port,
 			    int lane, u16 addr, u8 clear, u8 set, bool committed)
 {
@@ -1415,6 +1427,215 @@ void intel_c10pll_dump_hw_state(struct drm_i915_private *i915,
 			    i + 2, hw_state->pll[i + 2], i + 3, hw_state->pll[i + 3]);
 }
 
+static bool intel_c20_use_mplla(u32 clock)
+{
+	/* 10G and 20G rates use MPLLA */
+	if (clock == 312500 || clock == 625000)
+		return true;
+
+	return false;
+}
+
+static u8 intel_c20_get_dp_rate(u32 clock)
+{
+	switch (clock) {
+	case 162000: /* 1.62 Gbps DP1.4 */
+		return 0;
+	case 270000: /* 2.7 Gbps DP1.4 */
+		return 1;
+	case 540000: /* 5.4 Gbps DP 1.4 */
+		return 2;
+	case 810000: /* 8.1 Gbps DP1.4 */
+		return 3;
+	case 216000: /* 2.16 Gbps eDP */
+		return 4;
+	case 243000: /* 2.43 Gbps eDP */
+		return 5;
+	case 324000: /* 3.24 Gbps eDP */
+		return 6;
+	case 432000: /* 4.32 Gbps eDP */
+		return 7;
+	case 312500: /* 10 Gbps DP2.0 */
+		return 8;
+	case 421875: /* 13.5 Gbps DP2.0 */
+		return 9;
+	case 625000: /* 20 Gbps DP2.0*/
+		return 10;
+	case 648000: /* 6.48 Gbps eDP*/
+		return 11;
+	case 675000: /* 6.75 Gbps eDP*/
+		return 12;
+	default:
+		MISSING_CASE(clock);
+		return 0;
+	}
+}
+
+static u8 intel_c20_get_hdmi_rate(u32 clock)
+{
+	switch (clock) {
+	case 25175:
+	case 27000:
+	case 74250:
+	case 148500:
+	case 594000:
+		return 0;
+	case 166670: /* 3 Gbps */
+	case 333330: /* 6 Gbps */
+	case 666670: /* 12 Gbps */
+		return 1;
+	case 444440: /* 8 Gbps */
+		return 2;
+	case 555560: /* 10 Gbps */
+		return 3;
+	default:
+		MISSING_CASE(clock);
+		return 0;
+	}
+}
+
+static bool is_dp2(u32 clock)
+{
+	/* DP2.0 clock rates */
+	if (clock == 312500 || clock == 421875 || clock  == 625000)
+		return true;
+
+	return false;
+}
+
+static bool is_hdmi_frl(u32 clock)
+{
+	switch (clock) {
+	case 166670: /* 3 Gbps */
+	case 333330: /* 6 Gbps */
+	case 444440: /* 8 Gbps */
+	case 555560: /* 10 Gbps */
+	case 666670: /* 12 Gbps */
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool intel_c20_protocol_switch_valid(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
+
+	/* banks should not be cleared for DPALT/USB4/TBT modes */
+	/* TODO: optimize re-calibration in legacy mode */
+	return intel_tc_port_in_legacy_mode(intel_dig_port);
+}
+
+static int intel_get_c20_custom_width(u32 clock, bool dp)
+{
+	if (dp && is_dp2(clock))
+		return 2;
+	else if (is_hdmi_frl(clock))
+		return 1;
+	else
+		return 0;
+}
+
+static void intel_c20_pll_program(struct drm_i915_private *i915,
+				  const struct intel_crtc_state *crtc_state,
+				  struct intel_encoder *encoder)
+{
+	const struct intel_c20pll_state *pll_state = &crtc_state->cx0pll_state.c20;
+	bool dp = false;
+	int lane = crtc_state->lane_count > 2 ? INTEL_CX0_BOTH_LANES : INTEL_CX0_LANE0;
+	bool cntx;
+	int i;
+
+	if (intel_crtc_has_dp_encoder(crtc_state))
+		dp = true;
+
+	/* 1. Read current context selection */
+	cntx = intel_cx0_read(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_VDR_CUSTOM_SERDES_RATE) & BIT(0);
+
+	/*
+	 * 2. If there is a protocol switch from HDMI to DP or vice versa, clear
+	 * the lane #0 MPLLB CAL_DONE_BANK DP2.0 10G and 20G rates enable MPLLA.
+	 * Protocol switch is only applicable for MPLLA
+	 */
+	if (intel_c20_protocol_switch_valid(encoder)) {
+		for (i = 0; i < 4; i++)
+			intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(i), 0);
+		usleep_range(4000, 4100);
+	}
+
+	/* 3. Write SRAM configuration context. If A in use, write configuration to B context */
+	/* 3.1 Tx configuration */
+	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
+		if (cntx)
+			intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_A_TX_CNTX_CFG(i), pll_state->tx[i]);
+		else
+			intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_B_TX_CNTX_CFG(i), pll_state->tx[i]);
+	}
+
+	/* 3.2 common configuration */
+	for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) {
+		if (cntx)
+			intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_A_CMN_CNTX_CFG(i), pll_state->cmn[i]);
+		else
+			intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_B_CMN_CNTX_CFG(i), pll_state->cmn[i]);
+	}
+
+	/* 3.3 mpllb or mplla configuration */
+	if (intel_c20_use_mplla(pll_state->clock)) {
+		for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) {
+			if (cntx)
+				intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0,
+						     PHY_C20_A_MPLLA_CNTX_CFG(i),
+						     pll_state->mplla[i]);
+			else
+				intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0,
+						     PHY_C20_B_MPLLA_CNTX_CFG(i),
+						     pll_state->mplla[i]);
+		}
+	} else {
+		for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) {
+			if (cntx)
+				intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0,
+						     PHY_C20_A_MPLLB_CNTX_CFG(i),
+						     pll_state->mpllb[i]);
+			else
+				intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0,
+						     PHY_C20_B_MPLLB_CNTX_CFG(i),
+						     pll_state->mpllb[i]);
+		}
+	}
+
+	/* 4. Program custom width to match the link protocol */
+	intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_WIDTH,
+		      PHY_C20_CUSTOM_WIDTH_MASK,
+		      PHY_C20_CUSTOM_WIDTH(intel_get_c20_custom_width(pll_state->clock, dp)),
+		      MB_WRITE_COMMITTED);
+
+	/* 5. For DP or 6. For HDMI */
+	if (dp) {
+		intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+			      BIT(6) | PHY_C20_CUSTOM_SERDES_MASK,
+			      BIT(6) | PHY_C20_CUSTOM_SERDES(intel_c20_get_dp_rate(pll_state->clock)),
+			      MB_WRITE_COMMITTED);
+	} else {
+		intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+			      BIT(7) | PHY_C20_CUSTOM_SERDES_MASK,
+			      is_hdmi_frl(pll_state->clock) ? BIT(7) : 0,
+			      MB_WRITE_COMMITTED);
+
+		intel_cx0_write(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C20_VDR_HDMI_RATE,
+				intel_c20_get_hdmi_rate(pll_state->clock),
+				MB_WRITE_COMMITTED);
+	}
+
+	/*
+	 * 7. Write Vendor specific registers to toggle context setting to load
+	 * the updated programming toggle context bit
+	 */
+	intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+		      BIT(0), cntx ? 0 : 1, MB_WRITE_COMMITTED);
+}
+
 int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 				 const struct intel_c10pll_state *pll_state)
 {
@@ -1456,7 +1677,11 @@ static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
 		val |= XELPDP_LANE1_PHY_CLOCK_SELECT;
 
 	val |= XELPDP_FORWARD_CLOCK_UNGATE;
-	val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_MAXPCLK);
+
+	if (is_hdmi_frl(crtc_state->port_clock))
+		val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_DIV18CLK);
+	else
+		val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_MAXPCLK);
 
 	/* TODO: HDMI FRL */
 	/* TODO: DP2.0 10G and 20G rates enable MPLLA*/
@@ -1612,7 +1837,7 @@ static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, enum port po
 			 phy_name(phy), XELPDP_PORT_RESET_END_TIMEOUT);
 }
 
-static void intel_c10_program_phy_lane(struct drm_i915_private *i915,
+static void intel_cx0_program_phy_lane(struct drm_i915_private *i915,
 				       struct intel_encoder *encoder, int lane_count,
 				       bool lane_reversal)
 {
@@ -1620,9 +1845,11 @@ static void intel_c10_program_phy_lane(struct drm_i915_private *i915,
 	bool dp_alt_mode = intel_tc_port_in_dp_alt_mode(enc_to_dig_port(encoder));
 	enum port port = encoder->port;
 
-	intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
-		      0, C10_VDR_CTRL_MSGBUS_ACCESS,
-		      MB_WRITE_COMMITTED);
+	if (intel_is_c10phy(i915, intel_port_to_phy(i915, port)))
+		intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES,
+			      PHY_C10_VDR_CONTROL(1), 0,
+			      C10_VDR_CTRL_MSGBUS_ACCESS,
+			      MB_WRITE_COMMITTED);
 
 	/* TODO: DP-alt MFD case where only one PHY lane should be programmed. */
 	l0t1 = intel_cx0_read(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(1, 2));
@@ -1685,9 +1912,11 @@ static void intel_c10_program_phy_lane(struct drm_i915_private *i915,
 	intel_cx0_write(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(2, 2),
 			l1t2, MB_WRITE_COMMITTED);
 
-	intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1),
-		      0, C10_VDR_CTRL_UPDATE_CFG,
-		      MB_WRITE_COMMITTED);
+	if (intel_is_c10phy(i915, intel_port_to_phy(i915, port)))
+		intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES,
+			      PHY_C10_VDR_CONTROL(1), 0,
+			      C10_VDR_CTRL_UPDATE_CFG,
+			      MB_WRITE_COMMITTED);
 }
 
 static u32 intel_cx0_get_pclk_pll_request(u8 lane_mask)
@@ -1712,8 +1941,8 @@ static u32 intel_cx0_get_pclk_pll_ack(u8 lane_mask)
 	return val;
 }
 
-static void intel_c10pll_enable(struct intel_encoder *encoder,
-				const struct intel_crtc_state *crtc_state)
+void intel_cx0pll_enable(struct intel_encoder *encoder,
+			 const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
@@ -1721,6 +1950,7 @@ static void intel_c10pll_enable(struct intel_encoder *encoder,
 	bool lane_reversal = dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL;
 	u8 maxpclk_lane = lane_reversal ? INTEL_CX0_LANE1 :
 					  INTEL_CX0_LANE0;
+	intel_wakeref_t wakeref = intel_cx0_phy_transaction_begin(encoder);
 
 	/*
 	 * 1. Program PORT_CLOCK_CTL REGISTER to configure
@@ -1739,13 +1969,16 @@ static void intel_c10pll_enable(struct intel_encoder *encoder,
 					    CX0_P2_STATE_READY);
 
 	/* 4. Program PHY internal PLL internal registers. */
-	intel_c10_pll_program(i915, crtc_state, encoder);
+	if (intel_is_c10phy(i915, phy))
+		intel_c10_pll_program(i915, crtc_state, encoder);
+	else
+		intel_c20_pll_program(i915, crtc_state, encoder);
 
 	/*
 	 * 5. Program the enabled and disabled owned PHY lane
 	 * transmitters over message bus
 	 */
-	intel_c10_program_phy_lane(i915, encoder, crtc_state->lane_count, lane_reversal);
+	intel_cx0_program_phy_lane(i915, encoder, crtc_state->lane_count, lane_reversal);
 
 	/*
 	 * 6. Follow the Display Voltage Frequency Switching - Sequence
@@ -1779,32 +2012,22 @@ static void intel_c10pll_enable(struct intel_encoder *encoder,
 	 * 10. Follow the Display Voltage Frequency Switching Sequence After
 	 * Frequency Change. We handle this step in bxt_set_cdclk().
 	 */
-}
-
-void intel_cx0pll_enable(struct intel_encoder *encoder,
-			 const struct intel_crtc_state *crtc_state)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-	enum phy phy = intel_port_to_phy(i915, encoder->port);
-	intel_wakeref_t wakeref;
-
-	wakeref = intel_cx0_phy_transaction_begin(encoder);
-
-	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
-	intel_c10pll_enable(encoder, crtc_state);
 
 	/* TODO: enable TBT-ALT mode */
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
-static void intel_c10pll_disable(struct intel_encoder *encoder)
+void intel_cx0pll_disable(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	bool is_c10 = intel_is_c10phy(i915, phy);
+	intel_wakeref_t wakeref = intel_cx0_phy_transaction_begin(encoder);
 
 	/* 1. Change owned PHY lane power to Disable state. */
 	intel_cx0_powerdown_change_sequence(i915, encoder->port, INTEL_CX0_BOTH_LANES,
-					    CX0_P2PG_STATE_DISABLE);
+					    is_c10 ? CX0_P2PG_STATE_DISABLE :
+					    CX0_P4PG_STATE_DISABLE);
 
 	/*
 	 * 2. Follow the Display Voltage Frequency Switching Sequence Before
@@ -1842,18 +2065,7 @@ static void intel_c10pll_disable(struct intel_encoder *encoder)
 		     XELPDP_DDI_CLOCK_SELECT_MASK, 0);
 	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
 		     XELPDP_FORWARD_CLOCK_UNGATE, 0);
-}
-
-void intel_cx0pll_disable(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-	enum phy phy = intel_port_to_phy(i915, encoder->port);
-	intel_wakeref_t wakeref;
 
-	wakeref = intel_cx0_phy_transaction_begin(encoder);
-
-	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
-	intel_c10pll_disable(encoder);
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 20024622d0eb0..3aa334b15eafe 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -177,4 +177,37 @@
 #define PHY_CX0_TX_CONTROL(tx, control)	(0x400 + ((tx) - 1) * 0x200 + (control))
 #define   CONTROL2_DISABLE_SINGLE_TX	REG_BIT(6)
 
+/* C20 Registers */
+#define PHY_C20_WR_ADDRESS_L		0xC02
+#define PHY_C20_WR_ADDRESS_H		0xC03
+#define PHY_C20_WR_DATA_L		0xC04
+#define PHY_C20_WR_DATA_H		0xC05
+#define PHY_C20_RD_ADDRESS_L		0xC06
+#define PHY_C20_RD_ADDRESS_H		0xC07
+#define PHY_C20_RD_DATA_L		0xC08
+#define PHY_C20_RD_DATA_H		0xC09
+#define PHY_C20_VDR_CUSTOM_SERDES_RATE	0xD00
+#define PHY_C20_VDR_HDMI_RATE		0xD01
+#define   PHY_C20_CONTEXT_TOGGLE	REG_BIT8(0)
+#define   PHY_C20_CUSTOM_SERDES_MASK	REG_GENMASK8(4, 1)
+#define   PHY_C20_CUSTOM_SERDES(val)	REG_FIELD_PREP8(PHY_C20_CUSTOM_SERDES_MASK, val)
+#define PHY_C20_VDR_CUSTOM_WIDTH	0xD02
+#define   PHY_C20_CUSTOM_WIDTH_MASK	REG_GENMASK(1, 0)
+#define   PHY_C20_CUSTOM_WIDTH(val)	REG_FIELD_PREP8(PHY_C20_CUSTOM_WIDTH_MASK, val)
+#define PHY_C20_A_TX_CNTX_CFG(idx)	(0xCF2E - (idx))
+#define PHY_C20_B_TX_CNTX_CFG(idx)	(0xCF2A - (idx))
+#define PHY_C20_A_CMN_CNTX_CFG(idx)	(0xCDAA - (idx))
+#define PHY_C20_B_CMN_CNTX_CFG(idx)	(0xCDA5 - (idx))
+#define PHY_C20_A_MPLLA_CNTX_CFG(idx)	(0xCCF0 - (idx))
+#define PHY_C20_B_MPLLA_CNTX_CFG(idx)	(0xCCE5 - (idx))
+#define   C20_MPLLA_FRACEN		REG_BIT(14)
+#define   C20_MPLLA_TX_CLK_DIV_MASK	REG_GENMASK(10, 8)
+#define PHY_C20_A_MPLLB_CNTX_CFG(idx)	(0xCB5A - (idx))
+#define PHY_C20_B_MPLLB_CNTX_CFG(idx)	(0xCB4E - (idx))
+#define   C20_MPLLB_TX_CLK_DIV_MASK	REG_GENMASK(15, 13)
+#define   C20_MPLLB_FRACEN		REG_BIT(13)
+#define   C20_MULTIPLIER_MASK		REG_GENMASK(11, 0)
+
+#define RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(idx) (0x303D + (idx))
+
 #endif /* __INTEL_CX0_REG_DEFS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 8fb0f205b553e..e4cc7a3cd1379 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3359,7 +3359,8 @@ void intel_ddi_update_active_dpll(struct intel_atomic_state *state,
 	struct intel_crtc *slave_crtc;
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
 
-	if (!intel_phy_is_tc(i915, phy))
+	/* FIXME: Add MTL pll_mgr */
+	if (DISPLAY_VER(i915) >= 14 || !intel_phy_is_tc(i915, phy))
 		return;
 
 	intel_update_active_dpll(state, crtc, encoder);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 35c260bd14618..99b187f2f4edb 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -996,8 +996,21 @@ struct intel_c10pll_state {
 	u8 pll[20];
 };
 
+struct intel_c20pll_state {
+	u32 clock; /* in kHz */
+	u16 tx[3];
+	u16 cmn[4];
+	union {
+		u16 mplla[10];
+		u16 mpllb[11];
+	};
+};
+
 struct intel_cx0pll_state {
-	struct intel_c10pll_state c10;
+	union {
+		struct intel_c10pll_state c10;
+		struct intel_c20pll_state c20;
+	};
 	bool ssc_enabled;
 };
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 1d28a2560ae01..4361c1ac65c3a 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -53,6 +53,7 @@
 #include "intel_combo_phy_regs.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
+#include "intel_cx0_phy.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
@@ -423,7 +424,14 @@ static int ehl_max_source_rate(struct intel_dp *intel_dp)
 
 static int mtl_max_source_rate(struct intel_dp *intel_dp)
 {
-	return intel_dp_is_edp(intel_dp) ? 675000 : 810000;
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	enum phy phy = intel_port_to_phy(i915, dig_port->base.port);
+
+	if (intel_is_c10phy(i915, phy))
+		return intel_dp_is_edp(intel_dp) ? 675000 : 810000;
+
+	return 2000000;
 }
 
 static int vbt_max_link_rate(struct intel_dp *intel_dp)
@@ -452,7 +460,7 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp)
 	/* The values must be in increasing order */
 	static const int mtl_rates[] = {
 		162000, 216000, 243000, 270000, 324000, 432000, 540000, 675000,
-		810000,
+		810000,	1000000, 1350000, 2000000,
 	};
 	static const int icl_rates[] = {
 		162000, 216000, 270000, 324000, 432000, 540000, 648000, 810000,
-- 
GitLab


From 929f527a7b70a5a7810f83c5e8941657413596c3 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:22 +0300
Subject: [PATCH 0337/1791] drm/i915/mtl: C20 HW readout

Create a table for C20 DP1.4, DP2.0 and HDMI2.1 rates.
The PLL settings are based on table, not for algorithmic alternative.
For DP 1.4 only MPLLB is in use.

Once register settings are done, we read back C20 HW state.

BSpec: 64568

v2: Updated pll tables (RK)
    MPLLB selection fix (RK)

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-3-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 624 +++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_cx0_phy.h  |   8 +-
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |   1 +
 drivers/gpu/drm/i915/display/intel_ddi.c      |   9 +-
 .../drm/i915/display/intel_display_types.h    |   1 +
 drivers/gpu/drm/i915/display/intel_hdmi.c     |   6 +-
 drivers/gpu/drm/i915/display/intel_hdmi.h     |   1 +
 7 files changed, 630 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 71163bc5bbf58..651239006220c 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -11,6 +11,7 @@
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
+#include "intel_hdmi.h"
 #include "intel_panel.h"
 #include "intel_psr.h"
 #include "intel_tc.h"
@@ -285,6 +286,23 @@ static void intel_c20_sram_write(struct drm_i915_private *i915, enum port port,
 	intel_cx0_write(i915, port, lane, PHY_C20_WR_DATA_L, data & 0xff, 1);
 }
 
+static u16 intel_c20_sram_read(struct drm_i915_private *i915, enum port port,
+			       int lane, u16 addr)
+{
+	u16 val;
+
+	assert_dc_off(i915);
+
+	intel_cx0_write(i915, port, lane, PHY_C20_RD_ADDRESS_H, addr >> 8, 0);
+	intel_cx0_write(i915, port, lane, PHY_C20_RD_ADDRESS_L, addr & 0xff, 1);
+
+	val = intel_cx0_read(i915, port, lane, PHY_C20_RD_DATA_H);
+	val <<= 8;
+	val |= intel_cx0_read(i915, port, lane, PHY_C20_RD_DATA_L);
+
+	return val;
+}
+
 static void __intel_cx0_rmw(struct drm_i915_private *i915, enum port port,
 			    int lane, u16 addr, u8 clear, u8 set, bool committed)
 {
@@ -659,6 +677,199 @@ static const struct intel_c10pll_state * const mtl_c10_edp_tables[] = {
 	NULL,
 };
 
+/* C20 basic DP 1.4 tables */
+static const struct intel_c20pll_state mtl_c20_dp_rbr = {
+	.link_bit_rate = 162000,
+	.clock = 162000,
+	.tx = {	0xbe88, /* tx cfg0 */
+		0x5800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x50a8,	/* mpllb cfg0 */
+		0x2120,		/* mpllb cfg1 */
+		0xcd9a,		/* mpllb cfg2 */
+		0xbfc1,		/* mpllb cfg3 */
+		0x5ab8,         /* mpllb cfg4 */
+		0x4c34,         /* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x6000,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_dp_hbr1 = {
+	.link_bit_rate = 270000,
+	.clock = 270000,
+	.tx = {	0xbe88, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x308c,	/* mpllb cfg0 */
+		0x2110,		/* mpllb cfg1 */
+		0xcc9c,		/* mpllb cfg2 */
+		0xbfc1,		/* mpllb cfg3 */
+		0x4b9a,         /* mpllb cfg4 */
+		0x3f81,         /* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x5000,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_dp_hbr2 = {
+	.link_bit_rate = 540000,
+	.clock = 540000,
+	.tx = {	0xbe88, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x108c,	/* mpllb cfg0 */
+		0x2108,		/* mpllb cfg1 */
+		0xcc9c,		/* mpllb cfg2 */
+		0xbfc1,		/* mpllb cfg3 */
+		0x4b9a,         /* mpllb cfg4 */
+		0x3f81,         /* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x5000,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_dp_hbr3 = {
+	.link_bit_rate = 810000,
+	.clock = 810000,
+	.tx = {	0xbe88, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x10d2,	/* mpllb cfg0 */
+		0x2108,		/* mpllb cfg1 */
+		0x8d98,		/* mpllb cfg2 */
+		0xbfc1,		/* mpllb cfg3 */
+		0x7166,         /* mpllb cfg4 */
+		0x5f42,         /* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x7800,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+/* C20 basic DP 2.0 tables */
+static const struct intel_c20pll_state mtl_c20_dp_uhbr10 = {
+	.link_bit_rate = 1000000, /* 10 Gbps */
+	.clock = 312500,
+	.tx = {	0xbe21, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mplla = { 0x3104,	/* mplla cfg0 */
+		0xd105,		/* mplla cfg1 */
+		0xc025,		/* mplla cfg2 */
+		0xc025,		/* mplla cfg3 */
+		0x8c00,		/* mplla cfg4 */
+		0x759a,		/* mplla cfg5 */
+		0x4000,		/* mplla cfg6 */
+		0x0003,		/* mplla cfg7 */
+		0x3555,		/* mplla cfg8 */
+		0x0001,		/* mplla cfg9 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_dp_uhbr13_5 = {
+	.link_bit_rate = 1350000, /* 13.5 Gbps */
+	.clock = 421875,
+	.tx = {	0xbea0, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x015f,	/* mpllb cfg0 */
+		0x2205,		/* mpllb cfg1 */
+		0x1b17,		/* mpllb cfg2 */
+		0xffc1,		/* mpllb cfg3 */
+		0xe100,		/* mpllb cfg4 */
+		0xbd00,		/* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x4800,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_dp_uhbr20 = {
+	.link_bit_rate = 2000000, /* 20 Gbps */
+	.clock = 625000,
+	.tx = {	0xbe20, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mplla = { 0x3104,	/* mplla cfg0 */
+		0xd105,		/* mplla cfg1 */
+		0xc025,		/* mplla cfg2 */
+		0xc025,		/* mplla cfg3 */
+		0xa6ab,		/* mplla cfg4 */
+		0x8c00,		/* mplla cfg5 */
+		0x4000,		/* mplla cfg6 */
+		0x0003,		/* mplla cfg7 */
+		0x3555,		/* mplla cfg8 */
+		0x0001,		/* mplla cfg9 */
+		},
+};
+
+static const struct intel_c20pll_state * const mtl_c20_dp_tables[] = {
+	&mtl_c20_dp_rbr,
+	&mtl_c20_dp_hbr1,
+	&mtl_c20_dp_hbr2,
+	&mtl_c20_dp_hbr3,
+	&mtl_c20_dp_uhbr10,
+	&mtl_c20_dp_uhbr13_5,
+	&mtl_c20_dp_uhbr20,
+	NULL,
+};
+
 /*
  * HDMI link rates with 38.4 MHz reference clock.
  */
@@ -1243,7 +1454,281 @@ static const struct intel_c10pll_state * const mtl_c10_hdmi_tables[] = {
 	NULL,
 };
 
-int intel_c10_phy_check_hdmi_link_rate(int clock)
+static const struct intel_c20pll_state mtl_c20_hdmi_25_175 = {
+	.link_bit_rate = 25175,
+	.clock = 25175,
+	.tx = {  0xbe88, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0xa0d2,	/* mpllb cfg0 */
+		   0x7d80,	/* mpllb cfg1 */
+		   0x0906,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x0200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x0000,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0001,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_27_0 = {
+	.link_bit_rate = 27000,
+	.clock = 27000,
+	.tx = {  0xbe88, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0xa0e0,	/* mpllb cfg0 */
+		   0x7d80,	/* mpllb cfg1 */
+		   0x0906,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x8000,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0001,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_74_25 = {
+	.link_bit_rate = 74250,
+	.clock = 74250,
+	.tx = {  0xbe88, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x609a,	/* mpllb cfg0 */
+		   0x7d40,	/* mpllb cfg1 */
+		   0xca06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x5800,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0001,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_148_5 = {
+	.link_bit_rate = 148500,
+	.clock = 148500,
+	.tx = {  0xbe88, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x409a,	/* mpllb cfg0 */
+		   0x7d20,	/* mpllb cfg1 */
+		   0xca06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x5800,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0001,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_594 = {
+	.link_bit_rate = 594000,
+	.clock = 594000,
+	.tx = {  0xbe88, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x009a,	/* mpllb cfg0 */
+		   0x7d08,	/* mpllb cfg1 */
+		   0xca06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x5800,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0001,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_300 = {
+	.link_bit_rate = 3000000,
+	.clock = 166670,
+	.tx = {  0xbe98, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x209c,	/* mpllb cfg0 */
+		   0x7d10,	/* mpllb cfg1 */
+		   0xca06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x2000,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0004,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_600 = {
+	.link_bit_rate = 6000000,
+	.clock = 333330,
+	.tx = {  0xbe98, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x009c,	/* mpllb cfg0 */
+		   0x7d08,	/* mpllb cfg1 */
+		   0xca06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x2000,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0004,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_800 = {
+	.link_bit_rate = 8000000,
+	.clock = 444440,
+	.tx = {  0xbe98, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x00d0,	/* mpllb cfg0 */
+		   0x7d08,	/* mpllb cfg1 */
+		   0x4a06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0003,	/* mpllb cfg7 */
+		   0x2aaa,	/* mpllb cfg8 */
+		   0x0002,	/* mpllb cfg9 */
+		   0x0004,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_1000 = {
+	.link_bit_rate = 10000000,
+	.clock = 555560,
+	.tx = {  0xbe98, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x1104,	/* mpllb cfg0 */
+		   0x7d08,	/* mpllb cfg1 */
+		   0x0a06,	/* mpllb cfg2 */
+		   0xbe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0003,	/* mpllb cfg7 */
+		   0x3555,	/* mpllb cfg8 */
+		   0x0001,	/* mpllb cfg9 */
+		   0x0004,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state mtl_c20_hdmi_1200 = {
+	.link_bit_rate = 12000000,
+	.clock = 666670,
+	.tx = {  0xbe98, /* tx cfg0 */
+		  0x9800, /* tx cfg1 */
+		  0x0000, /* tx cfg2 */
+		},
+	.cmn = { 0x0500, /* cmn cfg0*/
+		  0x0005, /* cmn cfg1 */
+		  0x0000, /* cmn cfg2 */
+		  0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x0138,	/* mpllb cfg0 */
+		   0x7d08,	/* mpllb cfg1 */
+		   0x5486,	/* mpllb cfg2 */
+		   0xfe40,	/* mpllb cfg3 */
+		   0x0000,	/* mpllb cfg4 */
+		   0x0000,	/* mpllb cfg5 */
+		   0x2200,	/* mpllb cfg6 */
+		   0x0001,	/* mpllb cfg7 */
+		   0x4000,	/* mpllb cfg8 */
+		   0x0000,	/* mpllb cfg9 */
+		   0x0004,	/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state * const mtl_c20_hdmi_tables[] = {
+	&mtl_c20_hdmi_25_175,
+	&mtl_c20_hdmi_27_0,
+	&mtl_c20_hdmi_74_25,
+	&mtl_c20_hdmi_148_5,
+	&mtl_c20_hdmi_594,
+	&mtl_c20_hdmi_300,
+	&mtl_c20_hdmi_600,
+	&mtl_c20_hdmi_800,
+	&mtl_c20_hdmi_1000,
+	&mtl_c20_hdmi_1200,
+	NULL,
+};
+
+static int intel_c10_phy_check_hdmi_link_rate(int clock)
 {
 	const struct intel_c10pll_state * const *tables = mtl_c10_hdmi_tables;
 	int i;
@@ -1319,17 +1804,6 @@ static int intel_c10pll_calc_state(struct intel_crtc_state *crtc_state,
 	return -EINVAL;
 }
 
-int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state,
-			    struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-	enum phy phy = intel_port_to_phy(i915, encoder->port);
-
-	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
-
-	return intel_c10pll_calc_state(crtc_state, encoder);
-}
-
 void intel_c10pll_readout_hw_state(struct intel_encoder *encoder,
 				   struct intel_c10pll_state *pll_state)
 {
@@ -1427,6 +1901,74 @@ void intel_c10pll_dump_hw_state(struct drm_i915_private *i915,
 			    i + 2, hw_state->pll[i + 2], i + 3, hw_state->pll[i + 3]);
 }
 
+static int intel_c20_phy_check_hdmi_link_rate(int clock)
+{
+	const struct intel_c20pll_state * const *tables = mtl_c20_hdmi_tables;
+	int i;
+
+	for (i = 0; tables[i]; i++) {
+		if (clock == tables[i]->link_bit_rate)
+			return MODE_OK;
+	}
+
+	return MODE_CLOCK_RANGE;
+}
+
+int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock)
+{
+	struct intel_digital_port *dig_port = hdmi_to_dig_port(hdmi);
+	struct drm_i915_private *i915 = intel_hdmi_to_i915(hdmi);
+	enum phy phy = intel_port_to_phy(i915, dig_port->base.port);
+
+	if (intel_is_c10phy(i915, phy))
+		return intel_c10_phy_check_hdmi_link_rate(clock);
+	return intel_c20_phy_check_hdmi_link_rate(clock);
+}
+
+static const struct intel_c20pll_state * const *
+intel_c20_pll_tables_get(struct intel_crtc_state *crtc_state,
+			 struct intel_encoder *encoder)
+{
+	if (intel_crtc_has_dp_encoder(crtc_state))
+		return mtl_c20_dp_tables;
+	else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
+		return mtl_c20_hdmi_tables;
+
+	MISSING_CASE(encoder->type);
+	return NULL;
+}
+
+static int intel_c20pll_calc_state(struct intel_crtc_state *crtc_state,
+				   struct intel_encoder *encoder)
+{
+	const struct intel_c20pll_state * const *tables;
+	int i;
+
+	tables = intel_c20_pll_tables_get(crtc_state, encoder);
+	if (!tables)
+		return -EINVAL;
+
+	for (i = 0; tables[i]; i++) {
+		if (crtc_state->port_clock == tables[i]->link_bit_rate) {
+			crtc_state->cx0pll_state.c20 = *tables[i];
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state,
+			    struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	if (intel_is_c10phy(i915, phy))
+		return intel_c10pll_calc_state(crtc_state, encoder);
+	return intel_c20pll_calc_state(crtc_state, encoder);
+}
+
 static bool intel_c20_use_mplla(u32 clock)
 {
 	/* 10G and 20G rates use MPLLA */
@@ -1436,6 +1978,64 @@ static bool intel_c20_use_mplla(u32 clock)
 	return false;
 }
 
+void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
+				   struct intel_c20pll_state *pll_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	bool cntx;
+	intel_wakeref_t wakeref;
+	int i;
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
+	/* 1. Read current context selection */
+	cntx = intel_cx0_read(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_VDR_CUSTOM_SERDES_RATE) & PHY_C20_CONTEXT_TOGGLE;
+
+	/* Read Tx configuration */
+	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
+		if (cntx)
+			pll_state->tx[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+							       PHY_C20_B_TX_CNTX_CFG(i));
+		else
+			pll_state->tx[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+							       PHY_C20_A_TX_CNTX_CFG(i));
+	}
+
+	/* Read common configuration */
+	for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) {
+		if (cntx)
+			pll_state->cmn[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+								PHY_C20_B_CMN_CNTX_CFG(i));
+		else
+			pll_state->cmn[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+								PHY_C20_A_CMN_CNTX_CFG(i));
+	}
+
+	if (pll_state->tx[0] & C20_PHY_USE_MPLLB) {
+		/* MPLLB configuration */
+		for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) {
+			if (cntx)
+				pll_state->mpllb[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+									  PHY_C20_B_MPLLB_CNTX_CFG(i));
+			else
+				pll_state->mpllb[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+									  PHY_C20_A_MPLLB_CNTX_CFG(i));
+		}
+	} else {
+		/* MPLLA configuration */
+		for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) {
+			if (cntx)
+				pll_state->mplla[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+									  PHY_C20_B_MPLLA_CNTX_CFG(i));
+			else
+				pll_state->mplla[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0,
+									  PHY_C20_A_MPLLA_CNTX_CFG(i));
+		}
+	}
+
+	intel_cx0_phy_transaction_end(encoder, wakeref);
+}
+
 static u8 intel_c20_get_dp_rate(u32 clock)
 {
 	switch (clock) {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 509d1d12776eb..9760c6292c810 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -30,8 +30,12 @@ int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 				 const struct intel_c10pll_state *pll_state);
 void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state);
+void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
+				   struct intel_c20pll_state *pll_state);
 void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
 				     const struct intel_crtc_state *crtc_state);
-int intel_c10_phy_check_hdmi_link_rate(int clock);
-
+int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock);
+void intel_cx0_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
+				       const struct intel_crtc_state *crtc_state,
+				       u32 level);
 #endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 3aa334b15eafe..bfb39bce3b044 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -207,6 +207,7 @@
 #define   C20_MPLLB_TX_CLK_DIV_MASK	REG_GENMASK(15, 13)
 #define   C20_MPLLB_FRACEN		REG_BIT(13)
 #define   C20_MULTIPLIER_MASK		REG_GENMASK(11, 0)
+#define   C20_PHY_USE_MPLLB		REG_BIT(7)
 
 #define RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(idx) (0x303D + (idx))
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index e4cc7a3cd1379..bdd4c5871f862 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3853,10 +3853,13 @@ static void mtl_ddi_get_config(struct intel_encoder *encoder,
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
 
-	drm_WARN_ON(&i915->drm, !intel_is_c10phy(i915, phy));
+	if (intel_is_c10phy(i915, phy)) {
+		intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10);
+		intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
+	} else {
+		intel_c20pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c20);
+	}
 
-	intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10);
-	intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
 	crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
 
 	intel_ddi_get_config(encoder, crtc_state);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 99b187f2f4edb..96a3183675bee 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -997,6 +997,7 @@ struct intel_c10pll_state {
 };
 
 struct intel_c20pll_state {
+	u32 link_bit_rate;
 	u32 clock; /* in kHz */
 	u16 tx[3];
 	u16 cmn[4];
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index b3f3f761cbb3f..6ebdea722c845 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -58,7 +58,7 @@
 #include "intel_panel.h"
 #include "intel_snps_phy.h"
 
-static struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi)
+inline struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi)
 {
 	return to_i915(hdmi_to_dig_port(intel_hdmi)->base.base.dev);
 }
@@ -1865,8 +1865,8 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi,
 	 * FIXME: We will hopefully get an algorithmic way of programming
 	 * the MPLLB for HDMI in the future.
 	 */
-	if (IS_METEORLAKE(dev_priv))
-		return intel_c10_phy_check_hdmi_link_rate(clock);
+	if (DISPLAY_VER(dev_priv) >= 14)
+		return intel_cx0_phy_check_hdmi_link_rate(hdmi, clock);
 	else if (IS_DG2(dev_priv))
 		return intel_snps_phy_check_hdmi_link_rate(clock);
 
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h
index 774dda2376ed2..492bd39213850 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.h
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.h
@@ -54,5 +54,6 @@ int intel_hdmi_dsc_get_num_slices(const struct intel_crtc_state *crtc_state,
 				  int src_max_slices, int src_max_slice_width,
 				  int hdmi_max_slices, int hdmi_throughput);
 int intel_hdmi_dsc_get_slice_height(int vactive);
+struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi);
 
 #endif /* __INTEL_HDMI_H__ */
-- 
GitLab


From f968a25381acbe147c675fdd7a048f170e6cf44b Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:23 +0300
Subject: [PATCH 0338/1791] drm/i915/mtl: Dump C20 pll hw state

As we already do with C10 chip, let's dump the pll
hw state for C20 as well.

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-4-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 20 ++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_cx0_phy.h |  2 ++
 drivers/gpu/drm/i915/display/intel_ddi.c     |  1 +
 3 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 651239006220c..8a014f4edc0d4 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2036,6 +2036,26 @@ void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
+void intel_c20pll_dump_hw_state(struct drm_i915_private *i915,
+				const struct intel_c20pll_state *hw_state)
+{
+	int i;
+
+	drm_dbg_kms(&i915->drm, "c20pll_hw_state:\n");
+	drm_dbg_kms(&i915->drm, "tx[0] = 0x%.4x, tx[1] = 0x%.4x, tx[2] = 0x%.4x\n",
+		    hw_state->tx[0], hw_state->tx[1], hw_state->tx[2]);
+	drm_dbg_kms(&i915->drm, "cmn[0] = 0x%.4x, cmn[1] = 0x%.4x, cmn[2] = 0x%.4x, cmn[3] = 0x%.4x\n",
+		    hw_state->cmn[0], hw_state->cmn[1], hw_state->cmn[2], hw_state->cmn[3]);
+
+	if (intel_c20_use_mplla(hw_state->clock)) {
+		for (i = 0; i < ARRAY_SIZE(hw_state->mplla); i++)
+			drm_dbg_kms(&i915->drm, "mplla[%d] = 0x%.4x\n", i, hw_state->mplla[i]);
+	} else {
+		for (i = 0; i < ARRAY_SIZE(hw_state->mpllb); i++)
+			drm_dbg_kms(&i915->drm, "mpllb[%d] = 0x%.4x\n", i, hw_state->mpllb[i]);
+	}
+}
+
 static u8 intel_c20_get_dp_rate(u32 clock)
 {
 	switch (clock) {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 9760c6292c810..c643aae27bac1 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -32,6 +32,8 @@ void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state);
 void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 				   struct intel_c20pll_state *pll_state);
+void intel_c20pll_dump_hw_state(struct drm_i915_private *i915,
+				const struct intel_c20pll_state *hw_state);
 void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
 				     const struct intel_crtc_state *crtc_state);
 int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock);
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index bdd4c5871f862..2a0aa421422f6 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3858,6 +3858,7 @@ static void mtl_ddi_get_config(struct intel_encoder *encoder,
 		intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
 	} else {
 		intel_c20pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c20);
+		intel_c20pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c20);
 	}
 
 	crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
-- 
GitLab


From f1f9e62737abc4b6481a9a814526bd3e2cfd4d2f Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:24 +0300
Subject: [PATCH 0339/1791] drm/i915/mtl: C20 port clock calculation

Calculate port clock with C20 phy.

BSpec: 64568

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-5-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 45 +++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_cx0_phy.h  |  2 +
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  3 ++
 drivers/gpu/drm/i915/display/intel_ddi.c      |  4 +-
 drivers/gpu/drm/i915/display/intel_dpll.c     |  2 +
 5 files changed, 54 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 8a014f4edc0d4..aa8fc7beaafaf 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2283,6 +2283,51 @@ int intel_c10pll_calc_port_clock(struct intel_encoder *encoder,
 	return tmpclk;
 }
 
+int intel_c20pll_calc_port_clock(struct intel_encoder *encoder,
+				 const struct intel_c20pll_state *pll_state)
+{
+	unsigned int frac, frac_en, frac_quot, frac_rem, frac_den;
+	unsigned int multiplier, refclk = 38400;
+	unsigned int tx_clk_div;
+	unsigned int ref_clk_mpllb_div;
+	unsigned int fb_clk_div4_en;
+	unsigned int ref, vco;
+	unsigned int tx_rate_mult;
+	unsigned int tx_rate = REG_FIELD_GET(C20_PHY_TX_RATE, pll_state->tx[0]);
+
+	if (pll_state->tx[0] & C20_PHY_USE_MPLLB) {
+		tx_rate_mult = 1;
+		frac_en = REG_FIELD_GET(C20_MPLLB_FRACEN, pll_state->mpllb[6]);
+		frac_quot = pll_state->mpllb[8];
+		frac_rem =  pll_state->mpllb[9];
+		frac_den =  pll_state->mpllb[7];
+		multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mpllb[0]);
+		tx_clk_div = REG_FIELD_GET(C20_MPLLB_TX_CLK_DIV_MASK, pll_state->mpllb[0]);
+		ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mpllb[6]);
+		fb_clk_div4_en = 0;
+	} else {
+		tx_rate_mult = 2;
+		frac_en = REG_FIELD_GET(C20_MPLLA_FRACEN, pll_state->mplla[6]);
+		frac_quot = pll_state->mplla[8];
+		frac_rem =  pll_state->mplla[9];
+		frac_den =  pll_state->mplla[7];
+		multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mplla[0]);
+		tx_clk_div = REG_FIELD_GET(C20_MPLLA_TX_CLK_DIV_MASK, pll_state->mplla[1]);
+		ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mplla[6]);
+		fb_clk_div4_en = REG_FIELD_GET(C20_FB_CLK_DIV4_EN, pll_state->mplla[0]);
+	}
+
+	if (frac_en)
+		frac = frac_quot + DIV_ROUND_CLOSEST(frac_rem, frac_den);
+	else
+		frac = 0;
+
+	ref = DIV_ROUND_CLOSEST(refclk * (1 << (1 + fb_clk_div4_en)), 1 << ref_clk_mpllb_div);
+	vco = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(ref, (multiplier << (17 - 2)) + frac) >> 17, 10);
+
+	return vco << tx_rate_mult >> tx_clk_div >> tx_rate;
+}
+
 static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
 					 const struct intel_crtc_state *crtc_state,
 					 bool lane_reversal)
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index c643aae27bac1..83bd3500091b2 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -34,6 +34,8 @@ void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 				   struct intel_c20pll_state *pll_state);
 void intel_c20pll_dump_hw_state(struct drm_i915_private *i915,
 				const struct intel_c20pll_state *hw_state);
+int intel_c20pll_calc_port_clock(struct intel_encoder *encoder,
+				 const struct intel_c20pll_state *pll_state);
 void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder,
 				     const struct intel_crtc_state *crtc_state);
 int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock);
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index bfb39bce3b044..d3de4df2b682d 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -196,16 +196,19 @@
 #define   PHY_C20_CUSTOM_WIDTH(val)	REG_FIELD_PREP8(PHY_C20_CUSTOM_WIDTH_MASK, val)
 #define PHY_C20_A_TX_CNTX_CFG(idx)	(0xCF2E - (idx))
 #define PHY_C20_B_TX_CNTX_CFG(idx)	(0xCF2A - (idx))
+#define   C20_PHY_TX_RATE		REG_GENMASK(2, 0)
 #define PHY_C20_A_CMN_CNTX_CFG(idx)	(0xCDAA - (idx))
 #define PHY_C20_B_CMN_CNTX_CFG(idx)	(0xCDA5 - (idx))
 #define PHY_C20_A_MPLLA_CNTX_CFG(idx)	(0xCCF0 - (idx))
 #define PHY_C20_B_MPLLA_CNTX_CFG(idx)	(0xCCE5 - (idx))
 #define   C20_MPLLA_FRACEN		REG_BIT(14)
+#define   C20_FB_CLK_DIV4_EN		REG_BIT(13)
 #define   C20_MPLLA_TX_CLK_DIV_MASK	REG_GENMASK(10, 8)
 #define PHY_C20_A_MPLLB_CNTX_CFG(idx)	(0xCB5A - (idx))
 #define PHY_C20_B_MPLLB_CNTX_CFG(idx)	(0xCB4E - (idx))
 #define   C20_MPLLB_TX_CLK_DIV_MASK	REG_GENMASK(15, 13)
 #define   C20_MPLLB_FRACEN		REG_BIT(13)
+#define   C20_REF_CLK_MPLLB_DIV_MASK	REG_GENMASK(12, 10)
 #define   C20_MULTIPLIER_MASK		REG_GENMASK(11, 0)
 #define   C20_PHY_USE_MPLLB		REG_BIT(7)
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 2a0aa421422f6..ff4d8d6d4bc35 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3856,13 +3856,13 @@ static void mtl_ddi_get_config(struct intel_encoder *encoder,
 	if (intel_is_c10phy(i915, phy)) {
 		intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10);
 		intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
+		crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
 	} else {
 		intel_c20pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c20);
 		intel_c20pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c20);
+		crtc_state->port_clock = intel_c20pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c20);
 	}
 
-	crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
-
 	intel_ddi_get_config(encoder, crtc_state);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c
index a9fbef0fa817c..ca0f362a40e32 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll.c
@@ -1014,6 +1014,8 @@ static int mtl_crtc_compute_clock(struct intel_atomic_state *state,
 	/* TODO: Do the readback via intel_compute_shared_dplls() */
 	if (intel_is_c10phy(i915, phy))
 		crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
+	else
+		crtc_state->port_clock = intel_c20pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c20);
 
 	crtc_state->hw.adjusted_mode.crtc_clock = intel_crtc_dotclock(crtc_state);
 
-- 
GitLab


From ac50317e2659143bd1111b1785b5c0fa0ac1f6bf Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:25 +0300
Subject: [PATCH 0340/1791] drm/i915/mtl: Add voltage swing sequence for C20

DP1.4 and DP20 voltage swing sequence for C20 phy.

Bspec: 65449, 67636, 67610

Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-6-mika.kahola@intel.com
---
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  4 ++
 .../drm/i915/display/intel_ddi_buf_trans.c    | 53 ++++++++++++++++++-
 2 files changed, 55 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index d3de4df2b682d..ab9d1d983b881 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -212,6 +212,10 @@
 #define   C20_MULTIPLIER_MASK		REG_GENMASK(11, 0)
 #define   C20_PHY_USE_MPLLB		REG_BIT(7)
 
+/* C20 Phy VSwing Masks */
+#define C20_PHY_VSWING_PREEMPH_MASK	REG_GENMASK8(5, 0)
+#define C20_PHY_VSWING_PREEMPH(val)	REG_FIELD_PREP8(C20_PHY_VSWING_PREEMPH_MASK, val)
+
 #define RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(idx) (0x303D + (idx))
 
 #endif /* __INTEL_CX0_REG_DEFS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
index cd4becbae0988..b7d20485bde50 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c
@@ -9,6 +9,7 @@
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
+#include "intel_cx0_phy.h"
 
 /* HDMI/DVI modes ignore everything but the last 2 items. So we share
  * them for both DP and FDI transports, allowing those ports to
@@ -1048,12 +1049,52 @@ static const union intel_ddi_buf_trans_entry _mtl_c10_trans_dp14[] = {
 	{ .snps = { 62, 0, 0  } },      /* preset 9 */
 };
 
-static const struct intel_ddi_buf_trans mtl_cx0c10_trans = {
+static const struct intel_ddi_buf_trans mtl_cx0_trans = {
 	.entries = _mtl_c10_trans_dp14,
 	.num_entries = ARRAY_SIZE(_mtl_c10_trans_dp14),
 	.hdmi_default_entry = ARRAY_SIZE(_mtl_c10_trans_dp14) - 1,
 };
 
+/* DP2.0 */
+static const union intel_ddi_buf_trans_entry _mtl_c20_trans_uhbr[] = {
+	{ .snps = { 48, 0, 0 } },       /* preset 0 */
+	{ .snps = { 43, 0, 5 } },       /* preset 1 */
+	{ .snps = { 40, 0, 8 } },       /* preset 2 */
+	{ .snps = { 37, 0, 11 } },      /* preset 3 */
+	{ .snps = { 33, 0, 15 } },      /* preset 4 */
+	{ .snps = { 46, 2, 0 } },       /* preset 5 */
+	{ .snps = { 42, 2, 4 } },       /* preset 6 */
+	{ .snps = { 38, 2, 8 } },       /* preset 7 */
+	{ .snps = { 35, 2, 11 } },      /* preset 8 */
+	{ .snps = { 33, 2, 13 } },      /* preset 9 */
+	{ .snps = { 44, 4, 0 } },       /* preset 10 */
+	{ .snps = { 40, 4, 4 } },       /* preset 11 */
+	{ .snps = { 37, 4, 7 } },       /* preset 12 */
+	{ .snps = { 33, 4, 11 } },      /* preset 13 */
+	{ .snps = { 40, 8, 0 } },	/* preset 14 */
+	{ .snps = { 28, 2, 2 } },	/* preset 15 */
+};
+
+/* HDMI2.0 */
+static const union intel_ddi_buf_trans_entry _mtl_c20_trans_hdmi[] = {
+	{ .snps = { 48, 0, 0 } },       /* preset 0 */
+	{ .snps = { 38, 4, 6 } },       /* preset 1 */
+	{ .snps = { 36, 4, 8 } },       /* preset 2 */
+	{ .snps = { 34, 4, 10 } },      /* preset 3 */
+	{ .snps = { 32, 4, 12 } },      /* preset 4 */
+};
+
+static const struct intel_ddi_buf_trans mtl_c20_trans_hdmi = {
+	.entries = _mtl_c20_trans_hdmi,
+	.num_entries = ARRAY_SIZE(_mtl_c20_trans_hdmi),
+	.hdmi_default_entry = 0,
+};
+
+static const struct intel_ddi_buf_trans mtl_c20_trans_uhbr = {
+	.entries = _mtl_c20_trans_uhbr,
+	.num_entries = ARRAY_SIZE(_mtl_c20_trans_uhbr),
+};
+
 bool is_hobl_buf_trans(const struct intel_ddi_buf_trans *table)
 {
 	return table == &tgl_combo_phy_trans_edp_hbr2_hobl;
@@ -1630,7 +1671,15 @@ mtl_get_cx0_buf_trans(struct intel_encoder *encoder,
 		      const struct intel_crtc_state *crtc_state,
 		      int *n_entries)
 {
-	return intel_get_buf_trans(&mtl_cx0c10_trans, n_entries);
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	if (intel_crtc_has_dp_encoder(crtc_state) && crtc_state->port_clock >= 1000000)
+		return intel_get_buf_trans(&mtl_c20_trans_uhbr, n_entries);
+	else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) && !(intel_is_c10phy(i915, phy)))
+		return intel_get_buf_trans(&mtl_c20_trans_hdmi, n_entries);
+	else
+		return intel_get_buf_trans(&mtl_cx0_trans, n_entries);
 }
 
 void intel_ddi_buf_trans_init(struct intel_encoder *encoder)
-- 
GitLab


From 237e7be0bf57c2dd36f074d3b03a3291db3b3abf Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:26 +0300
Subject: [PATCH 0341/1791] drm/i915/mtl: For DP2.0 10G and 20G rates use MPLLA

Use MPLLA for DP2.0 rates 10G and 20G, when ssc is enabled.

v2: Fix typo in commit message (Animesh)

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-7-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index aa8fc7beaafaf..507ae0e1d80d4 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2349,8 +2349,11 @@ static void intel_program_port_clock_ctl(struct intel_encoder *encoder,
 		val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_MAXPCLK);
 
 	/* TODO: HDMI FRL */
-	/* TODO: DP2.0 10G and 20G rates enable MPLLA*/
-	val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0;
+	/* DP2.0 10G and 20G rates enable MPLLA*/
+	if (crtc_state->port_clock == 1000000 || crtc_state->port_clock == 2000000)
+		val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLA : 0;
+	else
+		val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0;
 
 	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
 		     XELPDP_LANE1_PHY_CLOCK_SELECT | XELPDP_FORWARD_CLOCK_UNGATE |
-- 
GitLab


From 73fc3abcb79732c2c9d4745d61b6859f587f01a3 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:27 +0300
Subject: [PATCH 0342/1791] drm/i915/mtl: Enabling/disabling sequence
 Thunderbolt pll

Enabling and disabling sequence for Thunderbolt PLL.

Bspec: 64568

v2: Use intel_de_wait_for_register() (RK)

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-8-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 135 ++++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_cx0_phy.h |   7 +-
 drivers/gpu/drm/i915/display/intel_ddi.c     |   4 +-
 3 files changed, 138 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 507ae0e1d80d4..c389af1a2d7bb 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2609,8 +2609,8 @@ static u32 intel_cx0_get_pclk_pll_ack(u8 lane_mask)
 	return val;
 }
 
-void intel_cx0pll_enable(struct intel_encoder *encoder,
-			 const struct intel_crtc_state *crtc_state)
+static void intel_cx0pll_enable(struct intel_encoder *encoder,
+				const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
@@ -2685,7 +2685,86 @@ void intel_cx0pll_enable(struct intel_encoder *encoder,
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
-void intel_cx0pll_disable(struct intel_encoder *encoder)
+static int intel_mtl_tbt_clock_select(struct drm_i915_private *i915, int clock)
+{
+	switch (clock) {
+	case 162000:
+		return XELPDP_DDI_CLOCK_SELECT_TBT_162;
+	case 270000:
+		return XELPDP_DDI_CLOCK_SELECT_TBT_270;
+	case 540000:
+		return XELPDP_DDI_CLOCK_SELECT_TBT_540;
+	case 810000:
+		return XELPDP_DDI_CLOCK_SELECT_TBT_810;
+	default:
+		MISSING_CASE(clock);
+		return XELPDP_DDI_CLOCK_SELECT_TBT_162;
+	}
+}
+
+static void intel_mtl_tbt_pll_enable(struct intel_encoder *encoder,
+				     const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	u32 val = 0;
+
+	/*
+	 * 1. Program PORT_CLOCK_CTL REGISTER to configure
+	 * clock muxes, gating and SSC
+	 */
+	val |= XELPDP_DDI_CLOCK_SELECT(intel_mtl_tbt_clock_select(i915, crtc_state->port_clock));
+	val |= XELPDP_FORWARD_CLOCK_UNGATE;
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_FORWARD_CLOCK_UNGATE, val);
+
+	/* 2. Read back PORT_CLOCK_CTL REGISTER */
+	val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port));
+
+	/*
+	 * 3. Follow the Display Voltage Frequency Switching - Sequence
+	 * Before Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 4. Set PORT_CLOCK_CTL register TBT CLOCK Request to "1" to enable PLL.
+	 */
+	val |= XELPDP_TBT_CLOCK_REQUEST;
+	intel_de_write(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), val);
+
+	/* 5. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "1". */
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+					 XELPDP_TBT_CLOCK_ACK,
+					 XELPDP_TBT_CLOCK_ACK,
+					 100, 0, NULL))
+		drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not locked after 100us.\n",
+			 encoder->base.base.id, encoder->base.name, phy_name(phy));
+
+	/*
+	 * 6. Follow the Display Voltage Frequency Switching Sequence After
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 7. Program DDI_CLK_VALFREQ to match intended DDI
+	 * clock frequency.
+	 */
+	intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port),
+		       crtc_state->port_clock);
+}
+
+void intel_mtl_pll_enable(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *crtc_state)
+{
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+
+	if (intel_tc_port_in_tbt_alt_mode(dig_port))
+		intel_mtl_tbt_pll_enable(encoder, crtc_state);
+	else
+		intel_cx0pll_enable(encoder, crtc_state);
+}
+
+static void intel_cx0pll_disable(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
@@ -2737,6 +2816,56 @@ void intel_cx0pll_disable(struct intel_encoder *encoder)
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
+static void intel_mtl_tbt_pll_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	enum phy phy = intel_port_to_phy(i915, encoder->port);
+
+	/*
+	 * 1. Follow the Display Voltage Frequency Switching Sequence Before
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 2. Set PORT_CLOCK_CTL register TBT CLOCK Request to "0" to disable PLL.
+	 */
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     XELPDP_TBT_CLOCK_REQUEST, 0);
+
+	/* 3. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "0". */
+	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+					 XELPDP_TBT_CLOCK_ACK,
+					 ~XELPDP_TBT_CLOCK_ACK,
+					 10, 0, NULL))
+		drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not unlocked after 10us.\n",
+			 encoder->base.base.id, encoder->base.name, phy_name(phy));
+
+	/*
+	 * 4. Follow the Display Voltage Frequency Switching Sequence After
+	 * Frequency Change. We handle this step in bxt_set_cdclk().
+	 */
+
+	/*
+	 * 5. Program PORT CLOCK CTRL register to disable and gate clocks
+	 */
+	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
+		     XELPDP_DDI_CLOCK_SELECT_MASK |
+		     XELPDP_FORWARD_CLOCK_UNGATE, 0);
+
+	/* 6. Program DDI_CLK_VALFREQ to 0. */
+	intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), 0);
+}
+
+void intel_mtl_pll_disable(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+
+	if (intel_tc_port_in_tbt_alt_mode(dig_port))
+		intel_mtl_tbt_pll_disable(encoder);
+	else
+		intel_cx0pll_disable(encoder);
+}
+
 void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 83bd3500091b2..9ea6310b6d794 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -19,9 +19,9 @@ struct intel_crtc_state;
 enum phy;
 
 bool intel_is_c10phy(struct drm_i915_private *dev_priv, enum phy phy);
-void intel_cx0pll_enable(struct intel_encoder *encoder,
-			 const struct intel_crtc_state *crtc_state);
-void intel_cx0pll_disable(struct intel_encoder *encoder);
+void intel_mtl_pll_enable(struct intel_encoder *encoder,
+			  const struct intel_crtc_state *crtc_state);
+void intel_mtl_pll_disable(struct intel_encoder *encoder);
 void intel_c10pll_readout_hw_state(struct intel_encoder *encoder, struct intel_c10pll_state *pll_state);
 int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder);
 void intel_c10pll_dump_hw_state(struct drm_i915_private *dev_priv,
@@ -42,4 +42,5 @@ int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock);
 void intel_cx0_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 				       const struct intel_crtc_state *crtc_state,
 				       u32 level);
+int intel_mtl_tbt_readout_hw_state(struct intel_encoder *encoder);
 #endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index ff4d8d6d4bc35..56cffd8394bc1 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4779,8 +4779,8 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	encoder->pipe_mask = ~0;
 
 	if (DISPLAY_VER(dev_priv) >= 14) {
-		encoder->enable_clock = intel_cx0pll_enable;
-		encoder->disable_clock = intel_cx0pll_disable;
+		encoder->enable_clock = intel_mtl_pll_enable;
+		encoder->disable_clock = intel_mtl_pll_disable;
 		encoder->get_config = mtl_ddi_get_config;
 	} else if (IS_DG2(dev_priv)) {
 		encoder->enable_clock = intel_mpllb_enable;
-- 
GitLab


From 7dee06bc0ff5ac7341ee3cde8161e58dc43dd9b0 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:28 +0300
Subject: [PATCH 0343/1791] drm/i915/mtl: Readout Thunderbolt HW state

Readout hw state for Thunderbolt.

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-9-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 27 ++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_cx0_phy.h |  2 +-
 drivers/gpu/drm/i915/display/intel_ddi.c     |  5 +++-
 3 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index c389af1a2d7bb..7f7b63ae76511 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2685,6 +2685,33 @@ static void intel_cx0pll_enable(struct intel_encoder *encoder,
 	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
+int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	u32 clock;
+	u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port));
+
+	clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val);
+
+	drm_WARN_ON(&i915->drm, !(val & XELPDP_FORWARD_CLOCK_UNGATE));
+	drm_WARN_ON(&i915->drm, !(val & XELPDP_TBT_CLOCK_REQUEST));
+	drm_WARN_ON(&i915->drm, !(val & XELPDP_TBT_CLOCK_ACK));
+
+	switch (clock) {
+	case XELPDP_DDI_CLOCK_SELECT_TBT_162:
+		return 162000;
+	case XELPDP_DDI_CLOCK_SELECT_TBT_270:
+		return 270000;
+	case XELPDP_DDI_CLOCK_SELECT_TBT_540:
+		return 540000;
+	case XELPDP_DDI_CLOCK_SELECT_TBT_810:
+		return 810000;
+	default:
+		MISSING_CASE(clock);
+		return 162000;
+	}
+}
+
 static int intel_mtl_tbt_clock_select(struct drm_i915_private *i915, int clock)
 {
 	switch (clock) {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index 9ea6310b6d794..c1b8f7980f693 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -42,5 +42,5 @@ int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock);
 void intel_cx0_phy_ddi_vswing_sequence(struct intel_encoder *encoder,
 				       const struct intel_crtc_state *crtc_state,
 				       u32 level);
-int intel_mtl_tbt_readout_hw_state(struct intel_encoder *encoder);
+int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder);
 #endif /* __INTEL_CX0_PHY_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 56cffd8394bc1..10d7327ba3fcc 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3852,8 +3852,11 @@ static void mtl_ddi_get_config(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_port_to_phy(i915, encoder->port);
+	struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
 
-	if (intel_is_c10phy(i915, phy)) {
+	if (intel_tc_port_in_tbt_alt_mode(dig_port)) {
+		crtc_state->port_clock = intel_mtl_tbt_calc_port_clock(encoder);
+	} else if (intel_is_c10phy(i915, phy)) {
 		intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10);
 		intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10);
 		crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10);
-- 
GitLab


From c0f3faaf936b67e124e9159fb1223a689cf9e15e Mon Sep 17 00:00:00 2001
From: Gustavo Sousa <gustavo.sousa@intel.com>
Date: Fri, 28 Apr 2023 12:54:29 +0300
Subject: [PATCH 0344/1791] drm/i915/mtl: Define mask for DDI AUX interrupts

Xe_LPD+ defines interrupt bits for only DDI ports in the DE Port
Interrupt registers. The bits for Type-C ports are defined in the PICA
interrupt registers.

BSpec: 50064

Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-10-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b7a19db75c9cb..e0a4fddccac5b 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1938,7 +1938,10 @@ static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
 {
 	u32 mask;
 
-	if (DISPLAY_VER(dev_priv) >= 13)
+	if (DISPLAY_VER(dev_priv) >= 14)
+		return TGL_DE_PORT_AUX_DDIA |
+			TGL_DE_PORT_AUX_DDIB;
+	else if (DISPLAY_VER(dev_priv) >= 13)
 		return TGL_DE_PORT_AUX_DDIA |
 			TGL_DE_PORT_AUX_DDIB |
 			TGL_DE_PORT_AUX_DDIC |
-- 
GitLab


From 6f0423b06a0cd55133bafb7a69fd2ac1ff7a6750 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:30 +0300
Subject: [PATCH 0345/1791] drm/i915/mtl: Power up TCSS

Add register writes to enable powering up Type-C subsystem i.e. TCSS.
For MeteorLake we need to request TCSS to power up and check the TCSS
power state after 500 us.

In addition, for PICA we need to set/clear the Type-C PHY ownnership
bit when Type-C device is connected/disconnected.

Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-11-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c |  19 ++
 drivers/gpu/drm/i915/display/intel_cx0_phy.h |   4 +
 drivers/gpu/drm/i915/display/intel_ddi.c     |   1 +
 drivers/gpu/drm/i915/display/intel_display.c |   2 +-
 drivers/gpu/drm/i915/display/intel_tc.c      | 199 ++++++++++++++++++-
 5 files changed, 216 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 7f7b63ae76511..d94127e7448b1 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2893,6 +2893,25 @@ void intel_mtl_pll_disable(struct intel_encoder *encoder)
 		intel_cx0pll_disable(encoder);
 }
 
+enum icl_port_dpll_id
+intel_mtl_port_pll_type(struct intel_encoder *encoder,
+			const struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	/*
+	 * TODO: Determine the PLL type from the SW state, once MTL PLL
+	 * handling is done via the standard shared DPLL framework.
+	 */
+	u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port));
+	u32 clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val);
+
+	if (clock == XELPDP_DDI_CLOCK_SELECT_MAXPCLK ||
+	    clock == XELPDP_DDI_CLOCK_SELECT_DIV18CLK)
+		return ICL_PORT_DPLL_MG_PHY;
+	else
+		return ICL_PORT_DPLL_DEFAULT;
+}
+
 void intel_c10pll_state_verify(struct intel_atomic_state *state,
 			       struct intel_crtc_state *new_crtc_state)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
index c1b8f7980f693..f99809af257dc 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h
@@ -16,12 +16,16 @@
 struct drm_i915_private;
 struct intel_encoder;
 struct intel_crtc_state;
+enum icl_port_dpll_id;
 enum phy;
 
 bool intel_is_c10phy(struct drm_i915_private *dev_priv, enum phy phy);
 void intel_mtl_pll_enable(struct intel_encoder *encoder,
 			  const struct intel_crtc_state *crtc_state);
 void intel_mtl_pll_disable(struct intel_encoder *encoder);
+enum icl_port_dpll_id
+intel_mtl_port_pll_type(struct intel_encoder *encoder,
+			const struct intel_crtc_state *crtc_state);
 void intel_c10pll_readout_hw_state(struct intel_encoder *encoder, struct intel_c10pll_state *pll_state);
 int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder);
 void intel_c10pll_dump_hw_state(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 10d7327ba3fcc..74d6544b7f88f 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4784,6 +4784,7 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	if (DISPLAY_VER(dev_priv) >= 14) {
 		encoder->enable_clock = intel_mtl_pll_enable;
 		encoder->disable_clock = intel_mtl_pll_disable;
+		encoder->port_pll_type = intel_mtl_port_pll_type;
 		encoder->get_config = mtl_ddi_get_config;
 	} else if (IS_DG2(dev_priv)) {
 		encoder->enable_clock = intel_mpllb_enable;
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index bf391a6cd8d68..0b9ae5759ffa4 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1756,7 +1756,7 @@ bool intel_phy_is_tc(struct drm_i915_private *dev_priv, enum phy phy)
 	if (IS_DG2(dev_priv))
 		/* DG2's "TC1" output uses a SNPS PHY */
 		return false;
-	else if (IS_ALDERLAKE_P(dev_priv))
+	else if (IS_ALDERLAKE_P(dev_priv) || IS_METEORLAKE(dev_priv))
 		return phy >= PHY_F && phy <= PHY_I;
 	else if (IS_TIGERLAKE(dev_priv))
 		return phy >= PHY_D && phy <= PHY_I;
diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 3b60995e9dfb3..951b12ac51dc1 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -5,6 +5,7 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "intel_cx0_phy_regs.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display.h"
@@ -59,6 +60,7 @@ static enum intel_display_power_domain
 tc_phy_cold_off_domain(struct intel_tc_port *);
 static u32 tc_phy_hpd_live_status(struct intel_tc_port *tc);
 static bool tc_phy_is_ready(struct intel_tc_port *tc);
+static bool tc_phy_wait_for_ready(struct intel_tc_port *tc);
 static enum tc_port_mode tc_phy_get_current_mode(struct intel_tc_port *tc);
 
 static const char *tc_port_mode_name(enum tc_port_mode mode)
@@ -141,15 +143,23 @@ bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port)
  *
  * POWER_DOMAIN_TC_COLD_OFF:
  * -------------------------
- * TGL/legacy, DP-alt modes:
+ * ICL/DP-alt, TBT mode:
+ *   - TCSS/TBT: block TC-cold power state for using the (direct or
+ *     TBT DP-IN) AUX and main lanes.
+ *
+ * TGL/all modes:
  *   - TCSS/IOM,FIA access for PHY ready, owned and HPD live state
- *   - TCSS/PHY: block TC-cold power state for using the PHY AUX and
- *     main lanes.
+ *   - TCSS/PHY: block TC-cold power state for using the (direct or
+ *     TBT DP-IN) AUX and main lanes.
  *
- * ICL, TGL, ADLP/TBT mode:
- *   - TCSS/IOM,FIA access for HPD live state
+ * ADLP/TBT mode:
  *   - TCSS/TBT: block TC-cold power state for using the (TBT DP-IN)
  *     AUX and main lanes.
+ *
+ * XELPDP+/all modes:
+ *   - TCSS/IOM,FIA access for PHY ready, owned state
+ *   - TCSS/PHY: block TC-cold power state for using the (direct or
+ *     TBT DP-IN) AUX and main lanes.
  */
 bool intel_tc_cold_requires_aux_pw(struct intel_digital_port *dig_port)
 {
@@ -872,6 +882,172 @@ static const struct intel_tc_phy_ops adlp_tc_phy_ops = {
 	.init = adlp_tc_phy_init,
 };
 
+/*
+ * XELPDP TC PHY handlers
+ * ----------------------
+ */
+static bool
+xelpdp_tc_phy_tcss_power_is_enabled(struct intel_tc_port *tc)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	enum port port = tc->dig_port->base.port;
+
+	assert_tc_cold_blocked(tc);
+
+	return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TCSS_POWER_STATE;
+}
+
+static bool
+xelpdp_tc_phy_wait_for_tcss_power(struct intel_tc_port *tc, bool enabled)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+
+	if (wait_for(xelpdp_tc_phy_tcss_power_is_enabled(tc) == enabled, 5)) {
+		drm_dbg_kms(&i915->drm,
+			    "Port %s: timeout waiting for TCSS power to get %s\n",
+			    enabled ? "enabled" : "disabled",
+			    tc->port_name);
+		return false;
+	}
+
+	return true;
+}
+
+static void __xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	enum port port = tc->dig_port->base.port;
+	u32 val;
+
+	assert_tc_cold_blocked(tc);
+
+	val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port));
+	if (enable)
+		val |= XELPDP_TCSS_POWER_REQUEST;
+	else
+		val &= ~XELPDP_TCSS_POWER_REQUEST;
+	intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val);
+}
+
+static bool xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+
+	__xelpdp_tc_phy_enable_tcss_power(tc, enable);
+
+	if ((!tc_phy_wait_for_ready(tc) ||
+	     !xelpdp_tc_phy_wait_for_tcss_power(tc, enable)) &&
+	    !drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY)) {
+		if (enable) {
+			__xelpdp_tc_phy_enable_tcss_power(tc, false);
+			xelpdp_tc_phy_wait_for_tcss_power(tc, false);
+		}
+
+		return false;
+	}
+
+	return true;
+}
+
+static void xelpdp_tc_phy_take_ownership(struct intel_tc_port *tc, bool take)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	enum port port = tc->dig_port->base.port;
+	u32 val;
+
+	assert_tc_cold_blocked(tc);
+
+	val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port));
+	if (take)
+		val |= XELPDP_TC_PHY_OWNERSHIP;
+	else
+		val &= ~XELPDP_TC_PHY_OWNERSHIP;
+	intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val);
+}
+
+static bool xelpdp_tc_phy_is_owned(struct intel_tc_port *tc)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	enum port port = tc->dig_port->base.port;
+
+	assert_tc_cold_blocked(tc);
+
+	return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TC_PHY_OWNERSHIP;
+}
+
+static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	intel_wakeref_t tc_cold_wref;
+	enum intel_display_power_domain domain;
+
+	tc_cold_wref = __tc_cold_block(tc, &domain);
+
+	tc->mode = tc_phy_get_current_mode(tc);
+	if (tc->mode != TC_PORT_DISCONNECTED)
+		tc->lock_wakeref = tc_cold_block(tc);
+
+	drm_WARN_ON(&i915->drm,
+		    (tc->mode == TC_PORT_DP_ALT || tc->mode == TC_PORT_LEGACY) &&
+		    !xelpdp_tc_phy_tcss_power_is_enabled(tc));
+
+	__tc_cold_unblock(tc, domain, tc_cold_wref);
+}
+
+static bool xelpdp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
+{
+	tc->lock_wakeref = tc_cold_block(tc);
+
+	if (tc->mode == TC_PORT_TBT_ALT)
+		return true;
+
+	if (!xelpdp_tc_phy_enable_tcss_power(tc, true))
+		goto out_unblock_tccold;
+
+	xelpdp_tc_phy_take_ownership(tc, true);
+
+	if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes))
+		goto out_release_phy;
+
+	return true;
+
+out_release_phy:
+	xelpdp_tc_phy_take_ownership(tc, false);
+	xelpdp_tc_phy_wait_for_tcss_power(tc, false);
+
+out_unblock_tccold:
+	tc_cold_unblock(tc, fetch_and_zero(&tc->lock_wakeref));
+
+	return false;
+}
+
+static void xelpdp_tc_phy_disconnect(struct intel_tc_port *tc)
+{
+	switch (tc->mode) {
+	case TC_PORT_LEGACY:
+	case TC_PORT_DP_ALT:
+		xelpdp_tc_phy_take_ownership(tc, false);
+		xelpdp_tc_phy_enable_tcss_power(tc, false);
+		fallthrough;
+	case TC_PORT_TBT_ALT:
+		tc_cold_unblock(tc, fetch_and_zero(&tc->lock_wakeref));
+		break;
+	default:
+		MISSING_CASE(tc->mode);
+	}
+}
+
+static const struct intel_tc_phy_ops xelpdp_tc_phy_ops = {
+	.cold_off_domain = tgl_tc_phy_cold_off_domain,
+	.hpd_live_status = adlp_tc_phy_hpd_live_status,
+	.is_ready = adlp_tc_phy_is_ready,
+	.is_owned = xelpdp_tc_phy_is_owned,
+	.get_hw_state = xelpdp_tc_phy_get_hw_state,
+	.connect = xelpdp_tc_phy_connect,
+	.disconnect = xelpdp_tc_phy_disconnect,
+	.init = adlp_tc_phy_init,
+};
+
 /*
  * Generic TC PHY handlers
  * -----------------------
@@ -945,13 +1121,18 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc,
 	return is_connected;
 }
 
-static void tc_phy_wait_for_ready(struct intel_tc_port *tc)
+static bool tc_phy_wait_for_ready(struct intel_tc_port *tc)
 {
 	struct drm_i915_private *i915 = tc_to_i915(tc);
 
-	if (wait_for(tc_phy_is_ready(tc), 100))
+	if (wait_for(tc_phy_is_ready(tc), 500)) {
 		drm_err(&i915->drm, "Port %s: timeout waiting for PHY ready\n",
 			tc->port_name);
+
+		return false;
+	}
+
+	return true;
 }
 
 static enum tc_port_mode
@@ -1442,7 +1623,9 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
 	dig_port->tc = tc;
 	tc->dig_port = dig_port;
 
-	if (DISPLAY_VER(i915) >= 13)
+	if (DISPLAY_VER(i915) >= 14)
+		tc->phy_ops = &xelpdp_tc_phy_ops;
+	else if (DISPLAY_VER(i915) >= 13)
 		tc->phy_ops = &adlp_tc_phy_ops;
 	else if (DISPLAY_VER(i915) >= 12)
 		tc->phy_ops = &tgl_tc_phy_ops;
-- 
GitLab


From 4366750a0d2d587ae8335944d723eb43a6c0d94a Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 28 Apr 2023 12:54:31 +0300
Subject: [PATCH 0346/1791] drm/i915/mtl: TypeC HPD live status query

The HPD live status for MTL has to be read from different set of
registers. MTL deserves a new function for this purpose
and cannot reuse the existing  HPD live status detection

Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-12-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 30 ++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 951b12ac51dc1..b192265a3d78b 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -886,6 +886,34 @@ static const struct intel_tc_phy_ops adlp_tc_phy_ops = {
  * XELPDP TC PHY handlers
  * ----------------------
  */
+static u32 xelpdp_tc_phy_hpd_live_status(struct intel_tc_port *tc)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_digital_port *dig_port = tc->dig_port;
+	enum hpd_pin hpd_pin = dig_port->base.hpd_pin;
+	u32 pica_isr_bits = i915->display.hotplug.hpd[hpd_pin];
+	u32 pch_isr_bit = i915->display.hotplug.pch_hpd[hpd_pin];
+	intel_wakeref_t wakeref;
+	u32 pica_isr;
+	u32 pch_isr;
+	u32 mask = 0;
+
+	with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref) {
+		pica_isr = intel_de_read(i915, PICAINTERRUPT_ISR);
+		pch_isr = intel_de_read(i915, SDEISR);
+	}
+
+	if (pica_isr & (pica_isr_bits & XELPDP_DP_ALT_HOTPLUG_MASK))
+		mask |= BIT(TC_PORT_DP_ALT);
+	if (pica_isr & (pica_isr_bits & XELPDP_TBT_HOTPLUG_MASK))
+		mask |= BIT(TC_PORT_TBT_ALT);
+
+	if (tc->legacy_port && (pch_isr & pch_isr_bit))
+		mask |= BIT(TC_PORT_LEGACY);
+
+	return mask;
+}
+
 static bool
 xelpdp_tc_phy_tcss_power_is_enabled(struct intel_tc_port *tc)
 {
@@ -1039,7 +1067,7 @@ static void xelpdp_tc_phy_disconnect(struct intel_tc_port *tc)
 
 static const struct intel_tc_phy_ops xelpdp_tc_phy_ops = {
 	.cold_off_domain = tgl_tc_phy_cold_off_domain,
-	.hpd_live_status = adlp_tc_phy_hpd_live_status,
+	.hpd_live_status = xelpdp_tc_phy_hpd_live_status,
 	.is_ready = adlp_tc_phy_is_ready,
 	.is_owned = xelpdp_tc_phy_is_owned,
 	.get_hw_state = xelpdp_tc_phy_get_hw_state,
-- 
GitLab


From dac6ce66db3e3bdb5dc66f6713929b125612ef01 Mon Sep 17 00:00:00 2001
From: Anusha Srivatsa <anusha.srivatsa@intel.com>
Date: Fri, 28 Apr 2023 12:54:32 +0300
Subject: [PATCH 0347/1791] drm/i915/mtl: Pin assignment for TypeC

Unlike previous platforms that used PORT_TX_DFLEXDPSP
for max_lane calculation, MTL uses only PORT_TX_DFLEXPA1
from which the max_lanes has to be calculated.

Bspec: 50235, 65380

Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Signed-off-by: Jose Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-13-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 28 +++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index b192265a3d78b..4fca711a58bce 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -16,6 +16,10 @@
 #include "intel_mg_phy_regs.h"
 #include "intel_tc.h"
 
+#define DP_PIN_ASSIGNMENT_C	0x3
+#define DP_PIN_ASSIGNMENT_D	0x4
+#define DP_PIN_ASSIGNMENT_E	0x5
+
 enum tc_port_mode {
 	TC_PORT_DISCONNECTED,
 	TC_PORT_TBT_ALT,
@@ -281,6 +285,27 @@ u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
 	       DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
 }
 
+static int mtl_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	intel_wakeref_t wakeref;
+	u32 pin_mask;
+
+	with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref)
+		pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port);
+
+	switch (pin_mask) {
+	default:
+		MISSING_CASE(pin_mask);
+		fallthrough;
+	case DP_PIN_ASSIGNMENT_D:
+		return 2;
+	case DP_PIN_ASSIGNMENT_C:
+	case DP_PIN_ASSIGNMENT_E:
+		return 4;
+	}
+}
+
 int intel_tc_port_fia_max_lane_count(struct intel_digital_port *dig_port)
 {
 	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
@@ -294,6 +319,9 @@ int intel_tc_port_fia_max_lane_count(struct intel_digital_port *dig_port)
 
 	assert_tc_cold_blocked(tc);
 
+	if (DISPLAY_VER(i915) >= 14)
+		return mtl_tc_port_get_pin_assignment_mask(dig_port);
+
 	lane_mask = 0;
 	with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref)
 		lane_mask = intel_tc_port_get_lane_mask(dig_port);
-- 
GitLab


From ee9634282d875083b2a172f0181f5fe6be50c524 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 28 Apr 2023 12:54:33 +0300
Subject: [PATCH 0348/1791] drm/i915/mtl: Enable TC ports

Finally, we can enable TC ports for Meteorlake.

Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428095433.4109054-14-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 0b9ae5759ffa4..3d3483e6f8363 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7336,9 +7336,12 @@ void intel_setup_outputs(struct drm_i915_private *dev_priv)
 		return;
 
 	if (IS_METEORLAKE(dev_priv)) {
-		/* TODO: initialize TC ports as well */
 		intel_ddi_init(dev_priv, PORT_A);
 		intel_ddi_init(dev_priv, PORT_B);
+		intel_ddi_init(dev_priv, PORT_TC1);
+		intel_ddi_init(dev_priv, PORT_TC2);
+		intel_ddi_init(dev_priv, PORT_TC3);
+		intel_ddi_init(dev_priv, PORT_TC4);
 	} else if (IS_DG2(dev_priv)) {
 		intel_ddi_init(dev_priv, PORT_A);
 		intel_ddi_init(dev_priv, PORT_B);
-- 
GitLab


From da3a99afd2575e744fe52cf661ce3f12feedd675 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 21 Apr 2023 16:59:48 +0300
Subject: [PATCH 0349/1791] drm/i915/rc6: throw out set() wrapper

Remove useless indirection that's just misdirection for the readers.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421135948.2029121-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_rc6.c | 157 ++++++++++++++--------------
 1 file changed, 76 insertions(+), 81 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
index 8f3cd68d14f8a..908a3d0f2343f 100644
--- a/drivers/gpu/drm/i915/gt/intel_rc6.c
+++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
@@ -53,11 +53,6 @@ static struct drm_i915_private *rc6_to_i915(struct intel_rc6 *rc)
 	return rc6_to_gt(rc)->i915;
 }
 
-static void set(struct intel_uncore *uncore, i915_reg_t reg, u32 val)
-{
-	intel_uncore_write_fw(uncore, reg, val);
-}
-
 static void gen11_rc6_enable(struct intel_rc6 *rc6)
 {
 	struct intel_gt *gt = rc6_to_gt(rc6);
@@ -72,19 +67,19 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6)
 	 */
 	if (!intel_uc_uses_guc_rc(&gt->uc)) {
 		/* 2b: Program RC6 thresholds.*/
-		set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85);
-		set(uncore, GEN10_MEDIA_WAKE_RATE_LIMIT, 150);
+		intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85);
+		intel_uncore_write_fw(uncore, GEN10_MEDIA_WAKE_RATE_LIMIT, 150);
 
-		set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
-		set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+		intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+		intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 		for_each_engine(engine, rc6_to_gt(rc6), id)
-			set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+			intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
 
-		set(uncore, GUC_MAX_IDLE_COUNT, 0xA);
+		intel_uncore_write_fw(uncore, GUC_MAX_IDLE_COUNT, 0xA);
 
-		set(uncore, GEN6_RC_SLEEP, 0);
+		intel_uncore_write_fw(uncore, GEN6_RC_SLEEP, 0);
 
-		set(uncore, GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
+		intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 50000); /* 50/125ms per EI */
 	}
 
 	/*
@@ -105,8 +100,8 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6)
 	 * Broadwell+, To be conservative, we want to factor in a context
 	 * switch on top (due to ksoftirqd).
 	 */
-	set(uncore, GEN9_MEDIA_PG_IDLE_HYSTERESIS, 60);
-	set(uncore, GEN9_RENDER_PG_IDLE_HYSTERESIS, 60);
+	intel_uncore_write_fw(uncore, GEN9_MEDIA_PG_IDLE_HYSTERESIS, 60);
+	intel_uncore_write_fw(uncore, GEN9_RENDER_PG_IDLE_HYSTERESIS, 60);
 
 	/* 3a: Enable RC6
 	 *
@@ -141,7 +136,7 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6)
 					      VDN_MFX_POWERGATE_ENABLE(i));
 	}
 
-	set(uncore, GEN9_PG_ENABLE, pg_enable);
+	intel_uncore_write_fw(uncore, GEN9_PG_ENABLE, pg_enable);
 }
 
 static void gen9_rc6_enable(struct intel_rc6 *rc6)
@@ -152,26 +147,26 @@ static void gen9_rc6_enable(struct intel_rc6 *rc6)
 
 	/* 2b: Program RC6 thresholds.*/
 	if (GRAPHICS_VER(rc6_to_i915(rc6)) >= 11) {
-		set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85);
-		set(uncore, GEN10_MEDIA_WAKE_RATE_LIMIT, 150);
+		intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16 | 85);
+		intel_uncore_write_fw(uncore, GEN10_MEDIA_WAKE_RATE_LIMIT, 150);
 	} else if (IS_SKYLAKE(rc6_to_i915(rc6))) {
 		/*
 		 * WaRsDoubleRc6WrlWithCoarsePowerGating:skl Doubling WRL only
 		 * when CPG is enabled
 		 */
-		set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 108 << 16);
+		intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 108 << 16);
 	} else {
-		set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16);
+		intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16);
 	}
 
-	set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
-	set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 	for_each_engine(engine, rc6_to_gt(rc6), id)
-		set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+		intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
 
-	set(uncore, GUC_MAX_IDLE_COUNT, 0xA);
+	intel_uncore_write_fw(uncore, GUC_MAX_IDLE_COUNT, 0xA);
 
-	set(uncore, GEN6_RC_SLEEP, 0);
+	intel_uncore_write_fw(uncore, GEN6_RC_SLEEP, 0);
 
 	/*
 	 * 2c: Program Coarse Power Gating Policies.
@@ -194,11 +189,11 @@ static void gen9_rc6_enable(struct intel_rc6 *rc6)
 	 * conservative, we have to factor in a context switch on top (due
 	 * to ksoftirqd).
 	 */
-	set(uncore, GEN9_MEDIA_PG_IDLE_HYSTERESIS, 250);
-	set(uncore, GEN9_RENDER_PG_IDLE_HYSTERESIS, 250);
+	intel_uncore_write_fw(uncore, GEN9_MEDIA_PG_IDLE_HYSTERESIS, 250);
+	intel_uncore_write_fw(uncore, GEN9_RENDER_PG_IDLE_HYSTERESIS, 250);
 
 	/* 3a: Enable RC6 */
-	set(uncore, GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */
+	intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */
 
 	rc6->ctl_enable =
 		GEN6_RC_CTL_HW_ENABLE |
@@ -210,8 +205,8 @@ static void gen9_rc6_enable(struct intel_rc6 *rc6)
 	 *   - Render/Media PG need to be disabled with RC6.
 	 */
 	if (!NEEDS_WaRsDisableCoarsePowerGating(rc6_to_i915(rc6)))
-		set(uncore, GEN9_PG_ENABLE,
-		    GEN9_RENDER_PG_ENABLE | GEN9_MEDIA_PG_ENABLE);
+		intel_uncore_write_fw(uncore, GEN9_PG_ENABLE,
+				      GEN9_RENDER_PG_ENABLE | GEN9_MEDIA_PG_ENABLE);
 }
 
 static void gen8_rc6_enable(struct intel_rc6 *rc6)
@@ -221,13 +216,13 @@ static void gen8_rc6_enable(struct intel_rc6 *rc6)
 	enum intel_engine_id id;
 
 	/* 2b: Program RC6 thresholds.*/
-	set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
-	set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
-	set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
+	intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 	for_each_engine(engine, rc6_to_gt(rc6), id)
-		set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
-	set(uncore, GEN6_RC_SLEEP, 0);
-	set(uncore, GEN6_RC6_THRESHOLD, 625); /* 800us/1.28 for TO */
+		intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+	intel_uncore_write_fw(uncore, GEN6_RC_SLEEP, 0);
+	intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 625); /* 800us/1.28 for TO */
 
 	/* 3: Enable RC6 */
 	rc6->ctl_enable =
@@ -245,20 +240,20 @@ static void gen6_rc6_enable(struct intel_rc6 *rc6)
 	u32 rc6vids, rc6_mask;
 	int ret;
 
-	set(uncore, GEN6_RC1_WAKE_RATE_LIMIT, 1000 << 16);
-	set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16 | 30);
-	set(uncore, GEN6_RC6pp_WAKE_RATE_LIMIT, 30);
-	set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000);
-	set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25);
+	intel_uncore_write_fw(uncore, GEN6_RC1_WAKE_RATE_LIMIT, 1000 << 16);
+	intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16 | 30);
+	intel_uncore_write_fw(uncore, GEN6_RC6pp_WAKE_RATE_LIMIT, 30);
+	intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000);
+	intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25);
 
 	for_each_engine(engine, rc6_to_gt(rc6), id)
-		set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+		intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
 
-	set(uncore, GEN6_RC_SLEEP, 0);
-	set(uncore, GEN6_RC1e_THRESHOLD, 1000);
-	set(uncore, GEN6_RC6_THRESHOLD, 50000);
-	set(uncore, GEN6_RC6p_THRESHOLD, 150000);
-	set(uncore, GEN6_RC6pp_THRESHOLD, 64000); /* unused */
+	intel_uncore_write_fw(uncore, GEN6_RC_SLEEP, 0);
+	intel_uncore_write_fw(uncore, GEN6_RC1e_THRESHOLD, 1000);
+	intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 50000);
+	intel_uncore_write_fw(uncore, GEN6_RC6p_THRESHOLD, 150000);
+	intel_uncore_write_fw(uncore, GEN6_RC6pp_THRESHOLD, 64000); /* unused */
 
 	/* We don't use those on Haswell */
 	rc6_mask = GEN6_RC_CTL_RC6_ENABLE;
@@ -372,22 +367,22 @@ static void chv_rc6_enable(struct intel_rc6 *rc6)
 	enum intel_engine_id id;
 
 	/* 2a: Program RC6 thresholds.*/
-	set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
-	set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
-	set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
+	intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
+	intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 
 	for_each_engine(engine, rc6_to_gt(rc6), id)
-		set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
-	set(uncore, GEN6_RC_SLEEP, 0);
+		intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+	intel_uncore_write_fw(uncore, GEN6_RC_SLEEP, 0);
 
 	/* TO threshold set to 500 us (0x186 * 1.28 us) */
-	set(uncore, GEN6_RC6_THRESHOLD, 0x186);
+	intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 0x186);
 
 	/* Allows RC6 residency counter to work */
-	set(uncore, VLV_COUNTER_CONTROL,
-	    _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
-			       VLV_MEDIA_RC6_COUNT_EN |
-			       VLV_RENDER_RC6_COUNT_EN));
+	intel_uncore_write_fw(uncore, VLV_COUNTER_CONTROL,
+			      _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
+						 VLV_MEDIA_RC6_COUNT_EN |
+						 VLV_RENDER_RC6_COUNT_EN));
 
 	/* 3: Enable RC6 */
 	rc6->ctl_enable = GEN7_RC_CTL_TO_MODE;
@@ -399,22 +394,22 @@ static void vlv_rc6_enable(struct intel_rc6 *rc6)
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
-	set(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 0x00280000);
-	set(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000);
-	set(uncore, GEN6_RC_IDLE_HYSTERSIS, 25);
+	intel_uncore_write_fw(uncore, GEN6_RC6_WAKE_RATE_LIMIT, 0x00280000);
+	intel_uncore_write_fw(uncore, GEN6_RC_EVALUATION_INTERVAL, 125000);
+	intel_uncore_write_fw(uncore, GEN6_RC_IDLE_HYSTERSIS, 25);
 
 	for_each_engine(engine, rc6_to_gt(rc6), id)
-		set(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
+		intel_uncore_write_fw(uncore, RING_MAX_IDLE(engine->mmio_base), 10);
 
-	set(uncore, GEN6_RC6_THRESHOLD, 0x557);
+	intel_uncore_write_fw(uncore, GEN6_RC6_THRESHOLD, 0x557);
 
 	/* Allows RC6 residency counter to work */
-	set(uncore, VLV_COUNTER_CONTROL,
-	    _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
-			       VLV_MEDIA_RC0_COUNT_EN |
-			       VLV_RENDER_RC0_COUNT_EN |
-			       VLV_MEDIA_RC6_COUNT_EN |
-			       VLV_RENDER_RC6_COUNT_EN));
+	intel_uncore_write_fw(uncore, VLV_COUNTER_CONTROL,
+			      _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH |
+						 VLV_MEDIA_RC0_COUNT_EN |
+						 VLV_RENDER_RC0_COUNT_EN |
+						 VLV_MEDIA_RC6_COUNT_EN |
+						 VLV_RENDER_RC6_COUNT_EN));
 
 	rc6->ctl_enable =
 	    GEN7_RC_CTL_TO_MODE | VLV_RC_CTL_CTX_RST_PARALLEL;
@@ -575,9 +570,9 @@ static void __intel_rc6_disable(struct intel_rc6 *rc6)
 
 	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 	if (GRAPHICS_VER(i915) >= 9)
-		set(uncore, GEN9_PG_ENABLE, 0);
-	set(uncore, GEN6_RC_CONTROL, 0);
-	set(uncore, GEN6_RC_STATE, 0);
+		intel_uncore_write_fw(uncore, GEN9_PG_ENABLE, 0);
+	intel_uncore_write_fw(uncore, GEN6_RC_CONTROL, 0);
+	intel_uncore_write_fw(uncore, GEN6_RC_STATE, 0);
 	intel_uncore_forcewake_put(uncore, FORCEWAKE_ALL);
 }
 
@@ -684,7 +679,7 @@ void intel_rc6_unpark(struct intel_rc6 *rc6)
 		return;
 
 	/* Restore HW timers for automatic RC6 entry while busy */
-	set(uncore, GEN6_RC_CONTROL, rc6->ctl_enable);
+	intel_uncore_write_fw(uncore, GEN6_RC_CONTROL, rc6->ctl_enable);
 }
 
 void intel_rc6_park(struct intel_rc6 *rc6)
@@ -704,7 +699,7 @@ void intel_rc6_park(struct intel_rc6 *rc6)
 		return;
 
 	/* Turn off the HW timers and go directly to rc6 */
-	set(uncore, GEN6_RC_CONTROL, GEN6_RC_CTL_RC6_ENABLE);
+	intel_uncore_write_fw(uncore, GEN6_RC_CONTROL, GEN6_RC_CTL_RC6_ENABLE);
 
 	if (HAS_RC6pp(rc6_to_i915(rc6)))
 		target = 0x6; /* deepest rc6 */
@@ -712,7 +707,7 @@ void intel_rc6_park(struct intel_rc6 *rc6)
 		target = 0x5; /* deep rc6 */
 	else
 		target = 0x4; /* normal rc6 */
-	set(uncore, GEN6_RC_STATE, target << RC_SW_TARGET_STATE_SHIFT);
+	intel_uncore_write_fw(uncore, GEN6_RC_STATE, target << RC_SW_TARGET_STATE_SHIFT);
 }
 
 void intel_rc6_disable(struct intel_rc6 *rc6)
@@ -735,7 +730,7 @@ void intel_rc6_fini(struct intel_rc6 *rc6)
 
 	/* We want the BIOS C6 state preserved across loads for MTL */
 	if (IS_METEORLAKE(rc6_to_i915(rc6)) && rc6->bios_state_captured)
-		set(uncore, GEN6_RC_STATE, rc6->bios_rc_state);
+		intel_uncore_write_fw(uncore, GEN6_RC_STATE, rc6->bios_rc_state);
 
 	pctx = fetch_and_zero(&rc6->pctx);
 	if (pctx)
@@ -766,18 +761,18 @@ static u64 vlv_residency_raw(struct intel_uncore *uncore, const i915_reg_t reg)
 	 * before we have set the default VLV_COUNTER_CONTROL value. So always
 	 * set the high bit to be safe.
 	 */
-	set(uncore, VLV_COUNTER_CONTROL,
-	    _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH));
+	intel_uncore_write_fw(uncore, VLV_COUNTER_CONTROL,
+			      _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH));
 	upper = intel_uncore_read_fw(uncore, reg);
 	do {
 		tmp = upper;
 
-		set(uncore, VLV_COUNTER_CONTROL,
-		    _MASKED_BIT_DISABLE(VLV_COUNT_RANGE_HIGH));
+		intel_uncore_write_fw(uncore, VLV_COUNTER_CONTROL,
+				      _MASKED_BIT_DISABLE(VLV_COUNT_RANGE_HIGH));
 		lower = intel_uncore_read_fw(uncore, reg);
 
-		set(uncore, VLV_COUNTER_CONTROL,
-		    _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH));
+		intel_uncore_write_fw(uncore, VLV_COUNTER_CONTROL,
+				      _MASKED_BIT_ENABLE(VLV_COUNT_RANGE_HIGH));
 		upper = intel_uncore_read_fw(uncore, reg);
 	} while (upper != tmp && --loop);
 
-- 
GitLab


From f60500f31e99fe5e641071d2ed4a8164a8050701 Mon Sep 17 00:00:00 2001
From: Arun R Murthy <arun.r.murthy@intel.com>
Date: Tue, 25 Apr 2023 08:29:44 +0530
Subject: [PATCH 0350/1791] drm/i915/display/dp: 128/132b LT requirement

For 128b/132b LT prior to LT DPTX should set power state, DP channel
coding and then link rate.

v2: added separate function to avoid code duplication(Jani N)
v3: DP2.1 section 3.5.2.16 is ordered, 3.5.1.2 is unordered and hence
    discarding <Ville>

Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425025944.151744-1-arun.r.murthy@intel.com
---
 .../drm/i915/display/intel_dp_link_training.c | 57 +++++++++++++------
 1 file changed, 39 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 6aa4ae5e7ebe3..e92c62bcc9b85 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -637,6 +637,38 @@ static bool intel_dp_link_max_vswing_reached(struct intel_dp *intel_dp,
 	return true;
 }
 
+static void
+intel_dp_update_downspread_ctrl(struct intel_dp *intel_dp,
+				const struct intel_crtc_state *crtc_state)
+{
+	u8 link_config[2];
+
+	link_config[0] = crtc_state->vrr.flipline ? DP_MSA_TIMING_PAR_IGNORE_EN : 0;
+	link_config[1] = intel_dp_is_uhbr(crtc_state) ?
+			 DP_SET_ANSI_128B132B : DP_SET_ANSI_8B10B;
+	drm_dp_dpcd_write(&intel_dp->aux, DP_DOWNSPREAD_CTRL, link_config, 2);
+}
+
+static void
+intel_dp_update_link_bw_set(struct intel_dp *intel_dp,
+			    const struct intel_crtc_state *crtc_state,
+			    u8 link_bw, u8 rate_select)
+{
+	u8 link_config[2];
+
+	/* Write the link configuration data */
+	link_config[0] = link_bw;
+	link_config[1] = crtc_state->lane_count;
+	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
+		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+	drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2);
+
+	/* eDP 1.4 rate select method. */
+	if (!link_bw)
+		drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET,
+				  &rate_select, 1);
+}
+
 /*
  * Prepare link training by configuring the link parameters. On DDI platforms
  * also enable the port here.
@@ -647,7 +679,6 @@ intel_dp_prepare_link_train(struct intel_dp *intel_dp,
 {
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-	u8 link_config[2];
 	u8 link_bw, rate_select;
 
 	if (intel_dp->prepare_link_retrain)
@@ -686,23 +717,13 @@ intel_dp_prepare_link_train(struct intel_dp *intel_dp,
 		drm_dbg_kms(&i915->drm,
 			    "[ENCODER:%d:%s] Using LINK_RATE_SET value %02x\n",
 			    encoder->base.base.id, encoder->base.name, rate_select);
-
-	/* Write the link configuration data */
-	link_config[0] = link_bw;
-	link_config[1] = crtc_state->lane_count;
-	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
-		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
-	drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2);
-
-	/* eDP 1.4 rate select method. */
-	if (!link_bw)
-		drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET,
-				  &rate_select, 1);
-
-	link_config[0] = crtc_state->vrr.flipline ? DP_MSA_TIMING_PAR_IGNORE_EN : 0;
-	link_config[1] = intel_dp_is_uhbr(crtc_state) ?
-		DP_SET_ANSI_128B132B : DP_SET_ANSI_8B10B;
-	drm_dp_dpcd_write(&intel_dp->aux, DP_DOWNSPREAD_CTRL, link_config, 2);
+	/*
+	 * Spec DP2.1 Section 3.5.2.16
+	 * Prior to LT DPTX should set 128b/132b DP Channel coding and then set link rate
+	 */
+	intel_dp_update_downspread_ctrl(intel_dp, crtc_state);
+	intel_dp_update_link_bw_set(intel_dp, crtc_state, link_bw,
+				    rate_select);
 
 	return true;
 }
-- 
GitLab


From 851c1aaed86260a8b5de9d03a31a5f8b0cd18dc0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 1 Sep 2022 15:47:09 +0300
Subject: [PATCH 0351/1791] drm/sti/sti_hdmi: convert to using is_hdmi from
 display info

Prefer the parsed results for is_hdmi in display info over calling
drm_detect_hdmi_monitor(). Remove the now redundant hdmi_monitor member
from struct sti_hdmi.

Cc: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Alain Volmat <avolmat@me.com>
Link: https://patchwork.freedesktop.org/patch/msgid/74fbd7a83712009734534d92e5499d4d87f0c53b.1662036058.git.jani.nikula@intel.com
---
 drivers/gpu/drm/sti/sti_hdmi.c | 11 ++++++-----
 drivers/gpu/drm/sti/sti_hdmi.h |  2 --
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index 8539fe1fedc4c..dc1562f14ceb0 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -266,6 +266,7 @@ static void hdmi_active_area(struct sti_hdmi *hdmi)
  */
 static void hdmi_config(struct sti_hdmi *hdmi)
 {
+	struct drm_connector *connector = hdmi->drm_connector;
 	u32 conf;
 
 	DRM_DEBUG_DRIVER("\n");
@@ -275,7 +276,7 @@ static void hdmi_config(struct sti_hdmi *hdmi)
 
 	/* Select encryption type and the framing mode */
 	conf |= HDMI_CFG_ESS_NOT_OESS;
-	if (hdmi->hdmi_monitor)
+	if (connector->display_info.is_hdmi)
 		conf |= HDMI_CFG_HDMI_NOT_DVI;
 
 	/* Set Hsync polarity */
@@ -985,15 +986,15 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector)
 	if (!edid)
 		goto fail;
 
-	hdmi->hdmi_monitor = drm_detect_hdmi_monitor(edid);
-	DRM_DEBUG_KMS("%s : %dx%d cm\n",
-		      (hdmi->hdmi_monitor ? "hdmi monitor" : "dvi monitor"),
-		      edid->width_cm, edid->height_cm);
 	cec_notifier_set_phys_addr_from_edid(hdmi->notifier, edid);
 
 	count = drm_add_edid_modes(connector, edid);
 	drm_connector_update_edid_property(connector, edid);
 
+	DRM_DEBUG_KMS("%s : %dx%d cm\n",
+		      (connector->display_info.is_hdmi ? "hdmi monitor" : "dvi monitor"),
+		      edid->width_cm, edid->height_cm);
+
 	kfree(edid);
 	return count;
 
diff --git a/drivers/gpu/drm/sti/sti_hdmi.h b/drivers/gpu/drm/sti/sti_hdmi.h
index 05b2f3d0d48d3..6d4c3f57bc46e 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.h
+++ b/drivers/gpu/drm/sti/sti_hdmi.h
@@ -57,7 +57,6 @@ struct hdmi_audio_params {
  * @reset: reset control of the hdmi phy
  * @ddc_adapt: i2c ddc adapter
  * @colorspace: current colorspace selected
- * @hdmi_monitor: true if HDMI monitor detected else DVI monitor assumed
  * @audio_pdev: ASoC hdmi-codec platform device
  * @audio: hdmi audio parameters.
  * @drm_connector: hdmi connector
@@ -83,7 +82,6 @@ struct sti_hdmi {
 	struct reset_control *reset;
 	struct i2c_adapter *ddc_adapt;
 	enum hdmi_colorspace colorspace;
-	bool hdmi_monitor;
 	struct platform_device *audio_pdev;
 	struct hdmi_audio_params audio;
 	struct drm_connector *drm_connector;
-- 
GitLab


From bfdede3a58ea970333d77a05144a7bcec13cf515 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 1 Sep 2022 15:47:10 +0300
Subject: [PATCH 0352/1791] drm/rockchip: cdn-dp: call
 drm_connector_update_edid_property() unconditionally
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Calling drm_connector_update_edid_property() should be done
unconditionally instead of depending on the number of modes added. Also
match the call order in inno_hdmi and rk3066_hdmi.

Cc: Sandy Huang <hjc@rock-chips.com>
Cc: Heiko Stübner <heiko@sntech.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Link: https://patchwork.freedesktop.org/patch/msgid/bb733eccfb389533cc6e207689be845164a1ed91.1662036058.git.jani.nikula@intel.com
---
 drivers/gpu/drm/rockchip/cdn-dp-core.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c
index 8526dda919317..b6afe3786b742 100644
--- a/drivers/gpu/drm/rockchip/cdn-dp-core.c
+++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c
@@ -273,10 +273,9 @@ static int cdn_dp_connector_get_modes(struct drm_connector *connector)
 				  edid->width_cm, edid->height_cm);
 
 		dp->sink_has_audio = drm_detect_monitor_audio(edid);
+
+		drm_connector_update_edid_property(connector, edid);
 		ret = drm_add_edid_modes(connector, edid);
-		if (ret)
-			drm_connector_update_edid_property(connector,
-								edid);
 	}
 	mutex_unlock(&dp->lock);
 
-- 
GitLab


From 5aa857db54ad2e963d65e234f7e4c183cd89c993 Mon Sep 17 00:00:00 2001
From: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Date: Thu, 27 Apr 2023 15:47:04 -0700
Subject: [PATCH 0353/1791] i915/pmu: Add support for total context runtime for
 GuC back-end

GPU accumulates the context runtime in a 32 bit counter - CTX_TIMESTAMP
in the context image. This value is saved/restored on context switches.
KMD accumulates these values into a 64 bit counter taking care of any
overflows as needed. This count provides the basis for client specific
busyness in the fdinfo interface.

KMD accumulation happens just before the context is unpinned and when
context switches out. This works for execlist back-end since execlist
scheduling has visibility into context switches. With GuC mode, KMD does
not have visibility into context switches and this counter is
accumulated only when context is unpinned. Context is unpinned once the
context scheduling is successfully disabled. Disabling context
scheduling is an asynchronous operation. Also if a context is servicing
frequent requests, scheduling may never be disabled on it.

For GuC mode, since updates to the context runtime may be delayed, add
hooks to update the context runtime in a worker thread as well as when
a user queries for it.

Limitation:
- If a context is never switched out or runs for a long period of time,
  the runtime value of CTX_TIMESTAMP may never be updated, so the
  counter value may be unreliable. This patch does not support such
  cases. Such support must be available from the GuC FW and it is WIP.

This patch is an extract from previous work authored by John/Umesh here -
https://patchwork.freedesktop.org/patch/496441/?series=105085&rev=4

v2: (Ashutosh)
- Drop COPS_RUNTIME_ACTIVE_TOTAL
- s/guc_context_update_clks/__guc_context_update_stats
- Pin context before accessing in guc_timestamp_ping
- In guc_context_unpin, use spinlock to serialize access to runtime stats

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Co-developed-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427224705.2785566-2-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/gt/intel_context.c       |  5 ++-
 drivers/gpu/drm/i915/gt/intel_context.h       |  2 +-
 drivers/gpu/drm/i915/gt/intel_context_types.h |  2 ++
 .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 31 +++++++++++++++++++
 4 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c
index 2aa63ec521b89..a53b26178f0ad 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.c
+++ b/drivers/gpu/drm/i915/gt/intel_context.c
@@ -578,10 +578,13 @@ void intel_context_bind_parent_child(struct intel_context *parent,
 	child->parallel.parent = parent;
 }
 
-u64 intel_context_get_total_runtime_ns(const struct intel_context *ce)
+u64 intel_context_get_total_runtime_ns(struct intel_context *ce)
 {
 	u64 total, active;
 
+	if (ce->ops->update_stats)
+		ce->ops->update_stats(ce);
+
 	total = ce->stats.runtime.total;
 	if (ce->ops->flags & COPS_RUNTIME_CYCLES)
 		total *= ce->engine->gt->clock_period_ns;
diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h
index 48f888c3da083..43ed92f8dc83c 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.h
+++ b/drivers/gpu/drm/i915/gt/intel_context.h
@@ -375,7 +375,7 @@ intel_context_clear_nopreempt(struct intel_context *ce)
 	clear_bit(CONTEXT_NOPREEMPT, &ce->flags);
 }
 
-u64 intel_context_get_total_runtime_ns(const struct intel_context *ce);
+u64 intel_context_get_total_runtime_ns(struct intel_context *ce);
 u64 intel_context_get_avg_runtime_ns(struct intel_context *ce);
 
 static inline u64 intel_context_clock(void)
diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
index e36670f2e6260..aceaac28a33eb 100644
--- a/drivers/gpu/drm/i915/gt/intel_context_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
@@ -58,6 +58,8 @@ struct intel_context_ops {
 
 	void (*sched_disable)(struct intel_context *ce);
 
+	void (*update_stats)(struct intel_context *ce);
+
 	void (*reset)(struct intel_context *ce);
 	void (*destroy)(struct kref *kref);
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index ee3e8352637f2..c869ddc73e695 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -1402,13 +1402,34 @@ static void __update_guc_busyness_stats(struct intel_guc *guc)
 	spin_unlock_irqrestore(&guc->timestamp.lock, flags);
 }
 
+static void __guc_context_update_stats(struct intel_context *ce)
+{
+	struct intel_guc *guc = ce_to_guc(ce);
+	unsigned long flags;
+
+	spin_lock_irqsave(&guc->timestamp.lock, flags);
+	lrc_update_runtime(ce);
+	spin_unlock_irqrestore(&guc->timestamp.lock, flags);
+}
+
+static void guc_context_update_stats(struct intel_context *ce)
+{
+	if (!intel_context_pin_if_active(ce))
+		return;
+
+	__guc_context_update_stats(ce);
+	intel_context_unpin(ce);
+}
+
 static void guc_timestamp_ping(struct work_struct *wrk)
 {
 	struct intel_guc *guc = container_of(wrk, typeof(*guc),
 					     timestamp.work.work);
 	struct intel_uc *uc = container_of(guc, typeof(*uc), guc);
 	struct intel_gt *gt = guc_to_gt(guc);
+	struct intel_context *ce;
 	intel_wakeref_t wakeref;
+	unsigned long index;
 	int srcu, ret;
 
 	/*
@@ -1424,6 +1445,10 @@ static void guc_timestamp_ping(struct work_struct *wrk)
 	with_intel_runtime_pm(&gt->i915->runtime_pm, wakeref)
 		__update_guc_busyness_stats(guc);
 
+	/* adjust context stats for overflow */
+	xa_for_each(&guc->context_lookup, index, ce)
+		guc_context_update_stats(ce);
+
 	intel_gt_reset_unlock(gt, srcu);
 
 	guc_enable_busyness_worker(guc);
@@ -2774,6 +2799,7 @@ static void guc_context_unpin(struct intel_context *ce)
 {
 	struct intel_guc *guc = ce_to_guc(ce);
 
+	__guc_context_update_stats(ce);
 	unpin_guc_id(guc, ce);
 	lrc_unpin(ce);
 
@@ -3455,6 +3481,7 @@ static void remove_from_context(struct i915_request *rq)
 }
 
 static const struct intel_context_ops guc_context_ops = {
+	.flags = COPS_RUNTIME_CYCLES,
 	.alloc = guc_context_alloc,
 
 	.close = guc_context_close,
@@ -3473,6 +3500,8 @@ static const struct intel_context_ops guc_context_ops = {
 
 	.sched_disable = guc_context_sched_disable,
 
+	.update_stats = guc_context_update_stats,
+
 	.reset = lrc_reset,
 	.destroy = guc_context_destroy,
 
@@ -3728,6 +3757,7 @@ static int guc_virtual_context_alloc(struct intel_context *ce)
 }
 
 static const struct intel_context_ops virtual_guc_context_ops = {
+	.flags = COPS_RUNTIME_CYCLES,
 	.alloc = guc_virtual_context_alloc,
 
 	.close = guc_context_close,
@@ -3745,6 +3775,7 @@ static const struct intel_context_ops virtual_guc_context_ops = {
 	.exit = guc_virtual_context_exit,
 
 	.sched_disable = guc_context_sched_disable,
+	.update_stats = guc_context_update_stats,
 
 	.destroy = guc_context_destroy,
 
-- 
GitLab


From 1324680a80ebf6eaba112bd54efa19e23081797d Mon Sep 17 00:00:00 2001
From: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Date: Thu, 27 Apr 2023 15:47:05 -0700
Subject: [PATCH 0354/1791] drm/i915/fdinfo: Enable fdinfo for GuC backends

Enable fdinfo for GuC based platforms with the exception that long
running contexts will not provide reliable busyness data unless they
switch out at some reasonable point in time.

Link: https://gitlab.freedesktop.org/drm/intel/issues/8303
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427224705.2785566-3-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_drm_client.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drm_client.c b/drivers/gpu/drm/i915/i915_drm_client.c
index e8fa172ebe5ee..d18d0a3ed905b 100644
--- a/drivers/gpu/drm/i915/i915_drm_client.c
+++ b/drivers/gpu/drm/i915/i915_drm_client.c
@@ -147,11 +147,7 @@ void i915_drm_client_fdinfo(struct seq_file *m, struct file *f)
 		   PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn));
 	seq_printf(m, "drm-client-id:\t%u\n", client->id);
 
-	/*
-	 * Temporarily skip showing client engine information with GuC submission till
-	 * fetching engine busyness is implemented in the GuC submission backend
-	 */
-	if (GRAPHICS_VER(i915) < 8 || intel_uc_uses_guc_submission(&i915->gt0.uc))
+	if (GRAPHICS_VER(i915) < 8)
 		return;
 
 	for (i = 0; i < ARRAY_SIZE(uabi_class_names); i++)
-- 
GitLab


From 03877d621db082610c9b7602c6e8cd6ebcb75a8f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 27 Apr 2023 14:05:43 +0200
Subject: [PATCH 0355/1791] drm/scheduler: mark jobs without fence as canceled
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When no hw fence is provided for a job that means that the job didn't executed.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427122726.1290170-1-christian.koenig@amd.com
---
 drivers/gpu/drm/scheduler/sched_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 8d248ce233c84..3e2d453296dd8 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -542,7 +542,7 @@ void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery)
 				DRM_DEV_ERROR(sched->dev, "fence add callback failed (%d)\n",
 					  r);
 		} else
-			drm_sched_job_done(s_job, 0);
+			drm_sched_job_done(s_job, -ECANCELED);
 	}
 
 	if (full_recovery) {
-- 
GitLab


From 860cf3bd715b3455e1c708386e773a669a2350ee Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Fri, 28 Apr 2023 18:29:51 +0530
Subject: [PATCH 0356/1791] drm/i915/gt: Use gt_err for GT info

It will be more informative regarding
GT if we use gt_err instead.

Cc: Andi Shyti <andi.shyti@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428125952.4037964-2-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gt/selftest_engine_pm.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
index 87c94314cf674..10e556a7eac45 100644
--- a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c
@@ -5,6 +5,7 @@
 
 #include <linux/sort.h>
 
+#include "gt/intel_gt_print.h"
 #include "i915_selftest.h"
 #include "intel_engine_regs.h"
 #include "intel_gpu_commands.h"
@@ -402,7 +403,7 @@ static int live_engine_pm(void *arg)
 
 			/* gt wakeref is async (deferred to workqueue) */
 			if (intel_gt_pm_wait_for_idle(gt)) {
-				pr_err("GT failed to idle\n");
+				gt_err(gt, "GT failed to idle\n");
 				return -EINVAL;
 			}
 		}
-- 
GitLab


From 725859b98a8eeb50c9a711b323e0cffc367873ea Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Fri, 28 Apr 2023 18:29:52 +0530
Subject: [PATCH 0357/1791] drm/i915/selftests: Use gt_err for GT info

It will be more informative regarding
GT if we use gt_err instead.

Cc: Andi Shyti <andi.shyti@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428125952.4037964-3-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/selftests/i915_gem_evict.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
index 37068542aafe7..f68ef4074088a 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
@@ -27,6 +27,7 @@
 #include "gem/selftests/igt_gem_utils.h"
 #include "gem/selftests/mock_context.h"
 #include "gt/intel_gt.h"
+#include "gt/intel_gt_print.h"
 
 #include "i915_selftest.h"
 
@@ -507,7 +508,8 @@ static int igt_evict_contexts(void *arg)
 		}
 		err = intel_gt_wait_for_idle(engine->gt, HZ * 3);
 		if (err) {
-			pr_err("Failed to idle GT (on %s)", engine->name);
+			gt_err(engine->gt, "Failed to idle GT (on %s)",
+			       engine->name);
 			break;
 		}
 	}
-- 
GitLab


From 629f59ade8fc399beee839cb81369510c9c982b6 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 2 May 2023 15:59:56 +0300
Subject: [PATCH 0358/1791] drm/udl: delete dead code

The "unode" pointer cannot be NULL here and checking for it causes
Smatch warnings:

   drivers/gpu/drm/udl/udl_main.c:259 udl_get_urb_locked()
   warn: can 'unode' even be NULL?

Fortunately, it's just harmless dead code which can be removed.  It's
left over from commit c5c354a3a472 ("drm/udl: Fix inconsistent urbs.count
value during udl_free_urb_list()").

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/e0e35421-8746-43b6-971e-e25d1cd1d6a7@kili.mountain
---
 drivers/gpu/drm/udl/udl_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index 061cb88c08a21..3ebe2ce55dfdf 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -255,7 +255,7 @@ static struct urb *udl_get_urb_locked(struct udl_device *udl, long timeout)
 	list_del_init(&unode->entry);
 	udl->urbs.available--;
 
-	return unode ? unode->urb : NULL;
+	return unode->urb;
 }
 
 #define GET_URB_TIMEOUT	HZ
-- 
GitLab


From b049132d61336f643d8faf2f6574b063667088cf Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 28 Apr 2023 11:56:33 -0700
Subject: [PATCH 0359/1791] drm/i915/guc: Don't capture Gen8 regs on Xe devices

A pair of pre-Xe registers were being included in the Xe capture list.
GuC was rejecting those as being invalid and logging errors about
them. So, stop doing it.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Fixes: dce2bd542337 ("drm/i915/guc: Add Gen9 registers for GuC error state capture.")
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428185636.457407-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index cf49188db6a6e..e0e793167d61b 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -31,12 +31,14 @@
 	{ FORCEWAKE_MT,             0,      0, "FORCEWAKE" }
 
 #define COMMON_GEN9BASE_GLOBAL \
-	{ GEN8_FAULT_TLB_DATA0,     0,      0, "GEN8_FAULT_TLB_DATA0" }, \
-	{ GEN8_FAULT_TLB_DATA1,     0,      0, "GEN8_FAULT_TLB_DATA1" }, \
 	{ ERROR_GEN6,               0,      0, "ERROR_GEN6" }, \
 	{ DONE_REG,                 0,      0, "DONE_REG" }, \
 	{ HSW_GTT_CACHE_EN,         0,      0, "HSW_GTT_CACHE_EN" }
 
+#define GEN9_GLOBAL \
+	{ GEN8_FAULT_TLB_DATA0,     0,      0, "GEN8_FAULT_TLB_DATA0" }, \
+	{ GEN8_FAULT_TLB_DATA1,     0,      0, "GEN8_FAULT_TLB_DATA1" }
+
 #define COMMON_GEN12BASE_GLOBAL \
 	{ GEN12_FAULT_TLB_DATA0,    0,      0, "GEN12_FAULT_TLB_DATA0" }, \
 	{ GEN12_FAULT_TLB_DATA1,    0,      0, "GEN12_FAULT_TLB_DATA1" }, \
@@ -142,6 +144,7 @@ static const struct __guc_mmio_reg_descr xe_lpd_gsc_inst_regs[] = {
 static const struct __guc_mmio_reg_descr default_global_regs[] = {
 	COMMON_BASE_GLOBAL,
 	COMMON_GEN9BASE_GLOBAL,
+	GEN9_GLOBAL,
 };
 
 static const struct __guc_mmio_reg_descr default_rc_class_regs[] = {
-- 
GitLab


From 684ee005d670a1b6360ba53db24b808f7584c89b Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 28 Apr 2023 11:56:34 -0700
Subject: [PATCH 0360/1791] drm/i915/guc: Consolidate duplicated capture list
 code

Remove 99% duplicated steered register list code. Also, include the
pre-Xe steered registers in the pre-Xe list generation.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428185636.457407-3-John.C.Harrison@Intel.com
---
 .../gpu/drm/i915/gt/uc/intel_guc_capture.c    | 112 +++++-------------
 1 file changed, 29 insertions(+), 83 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index e0e793167d61b..9184d2595e4ce 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -260,11 +260,15 @@ struct __ext_steer_reg {
 	i915_mcr_reg_t reg;
 };
 
-static const struct __ext_steer_reg xe_extregs[] = {
+static const struct __ext_steer_reg gen8_extregs[] = {
 	{"GEN8_SAMPLER_INSTDONE", GEN8_SAMPLER_INSTDONE},
 	{"GEN8_ROW_INSTDONE", GEN8_ROW_INSTDONE}
 };
 
+static const struct __ext_steer_reg xehpg_extregs[] = {
+	{"XEHPG_INSTDONE_GEOM_SVG", XEHPG_INSTDONE_GEOM_SVG}
+};
+
 static void __fill_ext_reg(struct __guc_mmio_reg_descr *ext,
 			   const struct __ext_steer_reg *extlist,
 			   int slice_id, int subslice_id)
@@ -295,8 +299,8 @@ __alloc_ext_regs(struct __guc_mmio_reg_descr_group *newlist,
 }
 
 static void
-guc_capture_alloc_steered_lists_xe_lpd(struct intel_guc *guc,
-				       const struct __guc_mmio_reg_descr_group *lists)
+guc_capture_alloc_steered_lists(struct intel_guc *guc,
+				const struct __guc_mmio_reg_descr_group *lists)
 {
 	struct intel_gt *gt = guc_to_gt(guc);
 	int slice, subslice, iter, i, num_steer_regs, num_tot_regs = 0;
@@ -304,74 +308,19 @@ guc_capture_alloc_steered_lists_xe_lpd(struct intel_guc *guc,
 	struct __guc_mmio_reg_descr_group *extlists;
 	struct __guc_mmio_reg_descr *extarray;
 	struct sseu_dev_info *sseu;
+	bool has_xehpg_extregs;
 
-	/* In XE_LPD we only have steered registers for the render-class */
+	/* steered registers currently only exist for the render-class */
 	list = guc_capture_get_one_list(lists, GUC_CAPTURE_LIST_INDEX_PF,
 					GUC_CAPTURE_LIST_TYPE_ENGINE_CLASS, GUC_RENDER_CLASS);
 	/* skip if extlists was previously allocated */
 	if (!list || guc->capture->extlists)
 		return;
 
-	num_steer_regs = ARRAY_SIZE(xe_extregs);
-
-	sseu = &gt->info.sseu;
-	for_each_ss_steering(iter, gt, slice, subslice)
-		num_tot_regs += num_steer_regs;
-
-	if (!num_tot_regs)
-		return;
-
-	/* allocate an extra for an end marker */
-	extlists = kcalloc(2, sizeof(struct __guc_mmio_reg_descr_group), GFP_KERNEL);
-	if (!extlists)
-		return;
-
-	if (__alloc_ext_regs(&extlists[0], list, num_tot_regs)) {
-		kfree(extlists);
-		return;
-	}
-
-	extarray = extlists[0].extlist;
-	for_each_ss_steering(iter, gt, slice, subslice) {
-		for (i = 0; i < num_steer_regs; ++i) {
-			__fill_ext_reg(extarray, &xe_extregs[i], slice, subslice);
-			++extarray;
-		}
-	}
-
-	guc->capture->extlists = extlists;
-}
-
-static const struct __ext_steer_reg xehpg_extregs[] = {
-	{"XEHPG_INSTDONE_GEOM_SVG", XEHPG_INSTDONE_GEOM_SVG}
-};
-
-static bool __has_xehpg_extregs(u32 ipver)
-{
-	return (ipver >= IP_VER(12, 55));
-}
-
-static void
-guc_capture_alloc_steered_lists_xe_hpg(struct intel_guc *guc,
-				       const struct __guc_mmio_reg_descr_group *lists,
-				       u32 ipver)
-{
-	struct intel_gt *gt = guc_to_gt(guc);
-	struct sseu_dev_info *sseu;
-	int slice, subslice, i, iter, num_steer_regs, num_tot_regs = 0;
-	const struct __guc_mmio_reg_descr_group *list;
-	struct __guc_mmio_reg_descr_group *extlists;
-	struct __guc_mmio_reg_descr *extarray;
-
-	/* In XE_LP / HPG we only have render-class steering registers during error-capture */
-	list = guc_capture_get_one_list(lists, GUC_CAPTURE_LIST_INDEX_PF,
-					GUC_CAPTURE_LIST_TYPE_ENGINE_CLASS, GUC_RENDER_CLASS);
-	/* skip if extlists was previously allocated */
-	if (!list || guc->capture->extlists)
-		return;
+	has_xehpg_extregs = GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 55);
 
-	num_steer_regs = ARRAY_SIZE(xe_extregs);
-	if (__has_xehpg_extregs(ipver))
+	num_steer_regs = ARRAY_SIZE(gen8_extregs);
+	if (has_xehpg_extregs)
 		num_steer_regs += ARRAY_SIZE(xehpg_extregs);
 
 	sseu = &gt->info.sseu;
@@ -393,11 +342,12 @@ guc_capture_alloc_steered_lists_xe_hpg(struct intel_guc *guc,
 
 	extarray = extlists[0].extlist;
 	for_each_ss_steering(iter, gt, slice, subslice) {
-		for (i = 0; i < ARRAY_SIZE(xe_extregs); ++i) {
-			__fill_ext_reg(extarray, &xe_extregs[i], slice, subslice);
+		for (i = 0; i < ARRAY_SIZE(gen8_extregs); ++i) {
+			__fill_ext_reg(extarray, &gen8_extregs[i], slice, subslice);
 			++extarray;
 		}
-		if (__has_xehpg_extregs(ipver)) {
+
+		if (has_xehpg_extregs) {
 			for (i = 0; i < ARRAY_SIZE(xehpg_extregs); ++i) {
 				__fill_ext_reg(extarray, &xehpg_extregs[i], slice, subslice);
 				++extarray;
@@ -413,26 +363,22 @@ static const struct __guc_mmio_reg_descr_group *
 guc_capture_get_device_reglist(struct intel_guc *guc)
 {
 	struct drm_i915_private *i915 = guc_to_gt(guc)->i915;
+	const struct __guc_mmio_reg_descr_group *lists;
 
-	if (GRAPHICS_VER(i915) > 11) {
-		/*
-		 * For certain engine classes, there are slice and subslice
-		 * level registers requiring steering. We allocate and populate
-		 * these at init time based on hw config add it as an extension
-		 * list at the end of the pre-populated render list.
-		 */
-		if (IS_DG2(i915))
-			guc_capture_alloc_steered_lists_xe_hpg(guc, xe_lpd_lists, IP_VER(12, 55));
-		else if (IS_XEHPSDV(i915))
-			guc_capture_alloc_steered_lists_xe_hpg(guc, xe_lpd_lists, IP_VER(12, 50));
-		else
-			guc_capture_alloc_steered_lists_xe_lpd(guc, xe_lpd_lists);
+	if (GRAPHICS_VER(i915) >= 12)
+		lists = xe_lpd_lists;
+	else
+		lists = default_lists;
 
-		return xe_lpd_lists;
-	}
+	/*
+	 * For certain engine classes, there are slice and subslice
+	 * level registers requiring steering. We allocate and populate
+	 * these at init time based on hw config add it as an extension
+	 * list at the end of the pre-populated render list.
+	 */
+	guc_capture_alloc_steered_lists(guc, lists);
 
-	/* if GuC submission is enabled on a non-POR platform, just use a common baseline */
-	return default_lists;
+	return lists;
 }
 
 static const char *
-- 
GitLab


From 44e36855bdb72cc2ade0153dce4a08a5b830a3a6 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 28 Apr 2023 11:56:35 -0700
Subject: [PATCH 0361/1791] drm/i915/guc: Capture list naming clean up

Don't use 'xe_lp*' prefixes for register lists that are common with
Gen8.

Don't add Xe only GSC registers to pre-Xe devices that don't
even have a GSC engine.

Fix Xe_LP name.

Don't use GEN9 as a prefix for register lists that contain all GEN8
registers.

Rename the 'default_' register list prefix to 'gen8_' as that is the
more accurate name.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428185636.457407-4-John.C.Harrison@Intel.com
---
 .../gpu/drm/i915/gt/uc/intel_guc_capture.c    | 100 +++++++++---------
 1 file changed, 49 insertions(+), 51 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index 9184d2595e4ce..729a8fcf20dda 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -30,12 +30,12 @@
 #define COMMON_BASE_GLOBAL \
 	{ FORCEWAKE_MT,             0,      0, "FORCEWAKE" }
 
-#define COMMON_GEN9BASE_GLOBAL \
+#define COMMON_GEN8BASE_GLOBAL \
 	{ ERROR_GEN6,               0,      0, "ERROR_GEN6" }, \
 	{ DONE_REG,                 0,      0, "DONE_REG" }, \
 	{ HSW_GTT_CACHE_EN,         0,      0, "HSW_GTT_CACHE_EN" }
 
-#define GEN9_GLOBAL \
+#define GEN8_GLOBAL \
 	{ GEN8_FAULT_TLB_DATA0,     0,      0, "GEN8_FAULT_TLB_DATA0" }, \
 	{ GEN8_FAULT_TLB_DATA1,     0,      0, "GEN8_FAULT_TLB_DATA1" }
 
@@ -96,67 +96,65 @@
 	{ GEN12_SFC_DONE(2),        0,      0, "SFC_DONE[2]" }, \
 	{ GEN12_SFC_DONE(3),        0,      0, "SFC_DONE[3]" }
 
-/* XE_LPD - Global */
-static const struct __guc_mmio_reg_descr xe_lpd_global_regs[] = {
+/* XE_LP Global */
+static const struct __guc_mmio_reg_descr xe_lp_global_regs[] = {
 	COMMON_BASE_GLOBAL,
-	COMMON_GEN9BASE_GLOBAL,
+	COMMON_GEN8BASE_GLOBAL,
 	COMMON_GEN12BASE_GLOBAL,
 };
 
-/* XE_LPD - Render / Compute Per-Class */
-static const struct __guc_mmio_reg_descr xe_lpd_rc_class_regs[] = {
+/* XE_LP Render / Compute Per-Class */
+static const struct __guc_mmio_reg_descr xe_lp_rc_class_regs[] = {
 	COMMON_BASE_HAS_EU,
 	COMMON_BASE_RENDER,
 	COMMON_GEN12BASE_RENDER,
 };
 
-/* GEN9/XE_LPD - Render / Compute Per-Engine-Instance */
-static const struct __guc_mmio_reg_descr xe_lpd_rc_inst_regs[] = {
+/* GEN8+ Render / Compute Per-Engine-Instance */
+static const struct __guc_mmio_reg_descr gen8_rc_inst_regs[] = {
 	COMMON_BASE_ENGINE_INSTANCE,
 };
 
-/* GEN9/XE_LPD - Media Decode/Encode Per-Engine-Instance */
-static const struct __guc_mmio_reg_descr xe_lpd_vd_inst_regs[] = {
+/* GEN8+ Media Decode/Encode Per-Engine-Instance */
+static const struct __guc_mmio_reg_descr gen8_vd_inst_regs[] = {
 	COMMON_BASE_ENGINE_INSTANCE,
 };
 
-/* XE_LPD - Video Enhancement Per-Class */
-static const struct __guc_mmio_reg_descr xe_lpd_vec_class_regs[] = {
+/* XE_LP Video Enhancement Per-Class */
+static const struct __guc_mmio_reg_descr xe_lp_vec_class_regs[] = {
 	COMMON_GEN12BASE_VEC,
 };
 
-/* GEN9/XE_LPD - Video Enhancement Per-Engine-Instance */
-static const struct __guc_mmio_reg_descr xe_lpd_vec_inst_regs[] = {
+/* GEN8+ Video Enhancement Per-Engine-Instance */
+static const struct __guc_mmio_reg_descr gen8_vec_inst_regs[] = {
 	COMMON_BASE_ENGINE_INSTANCE,
 };
 
-/* GEN9/XE_LPD - Blitter Per-Engine-Instance */
-static const struct __guc_mmio_reg_descr xe_lpd_blt_inst_regs[] = {
+/* GEN8+ Blitter Per-Engine-Instance */
+static const struct __guc_mmio_reg_descr gen8_blt_inst_regs[] = {
 	COMMON_BASE_ENGINE_INSTANCE,
 };
 
-/* XE_LPD - GSC Per-Engine-Instance */
-static const struct __guc_mmio_reg_descr xe_lpd_gsc_inst_regs[] = {
+/* XE_LP - GSC Per-Engine-Instance */
+static const struct __guc_mmio_reg_descr xe_lp_gsc_inst_regs[] = {
 	COMMON_BASE_ENGINE_INSTANCE,
 };
 
-/* GEN9 - Global */
-static const struct __guc_mmio_reg_descr default_global_regs[] = {
+/* GEN8 - Global */
+static const struct __guc_mmio_reg_descr gen8_global_regs[] = {
 	COMMON_BASE_GLOBAL,
-	COMMON_GEN9BASE_GLOBAL,
-	GEN9_GLOBAL,
+	COMMON_GEN8BASE_GLOBAL,
+	GEN8_GLOBAL,
 };
 
-static const struct __guc_mmio_reg_descr default_rc_class_regs[] = {
+static const struct __guc_mmio_reg_descr gen8_rc_class_regs[] = {
 	COMMON_BASE_HAS_EU,
 	COMMON_BASE_RENDER,
 };
 
 /*
- * Empty lists:
- * GEN9/XE_LPD - Blitter Per-Class
- * GEN9/XE_LPD - Media Decode/Encode Per-Class
- * GEN9 - VEC Class
+ * Empty list to prevent warnings about unknown class/instance types
+ * as not all class/instanace types have entries on all platforms.
  */
 static const struct __guc_mmio_reg_descr empty_regs_list[] = {
 };
@@ -174,37 +172,37 @@ static const struct __guc_mmio_reg_descr empty_regs_list[] = {
 	}
 
 /* List of lists */
-static const struct __guc_mmio_reg_descr_group default_lists[] = {
-	MAKE_REGLIST(default_global_regs, PF, GLOBAL, 0),
-	MAKE_REGLIST(default_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
-	MAKE_REGLIST(xe_lpd_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
-	MAKE_REGLIST(default_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(xe_lpd_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
+static const struct __guc_mmio_reg_descr_group gen8_lists[] = {
+	MAKE_REGLIST(gen8_global_regs, PF, GLOBAL, 0),
+	MAKE_REGLIST(gen8_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
+	MAKE_REGLIST(gen8_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(xe_lpd_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
+	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(xe_lpd_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
+	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(xe_lpd_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
+	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_GSC_OTHER_CLASS),
-	MAKE_REGLIST(xe_lpd_gsc_inst_regs, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
 	{}
 };
 
-static const struct __guc_mmio_reg_descr_group xe_lpd_lists[] = {
-	MAKE_REGLIST(xe_lpd_global_regs, PF, GLOBAL, 0),
-	MAKE_REGLIST(xe_lpd_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
-	MAKE_REGLIST(xe_lpd_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
-	MAKE_REGLIST(xe_lpd_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(xe_lpd_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
+static const struct __guc_mmio_reg_descr_group xe_lp_lists[] = {
+	MAKE_REGLIST(xe_lp_global_regs, PF, GLOBAL, 0),
+	MAKE_REGLIST(xe_lp_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
+	MAKE_REGLIST(xe_lp_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(xe_lpd_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(xe_lpd_vec_class_regs, PF, ENGINE_CLASS, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(xe_lpd_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
+	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
+	MAKE_REGLIST(xe_lp_vec_class_regs, PF, ENGINE_CLASS, GUC_VIDEOENHANCE_CLASS),
+	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(xe_lpd_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
+	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
 	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_GSC_OTHER_CLASS),
-	MAKE_REGLIST(xe_lpd_gsc_inst_regs, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
+	MAKE_REGLIST(xe_lp_gsc_inst_regs, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
 	{}
 };
 
@@ -366,9 +364,9 @@ guc_capture_get_device_reglist(struct intel_guc *guc)
 	const struct __guc_mmio_reg_descr_group *lists;
 
 	if (GRAPHICS_VER(i915) >= 12)
-		lists = xe_lpd_lists;
+		lists = xe_lp_lists;
 	else
-		lists = default_lists;
+		lists = gen8_lists;
 
 	/*
 	 * For certain engine classes, there are slice and subslice
-- 
GitLab


From e4730ae44240ae1a2e247bfc7f916813567acb92 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 28 Apr 2023 11:56:36 -0700
Subject: [PATCH 0362/1791] drm/i915/guc: Fix error capture for virtual engines

GuC based register dumps in error capture logs were basically broken
for virtual engines. This can be seen in igt@gem_exec_balancer@hang:
  [IGT] gem_exec_balancer: starting subtest hang
  [drm] GPU HANG: ecode 12:4:e1524110, in gem_exec_balanc [6388]
  [drm] GT0: GUC: No register capture node found for 0x1005 / 0xFEDC311D
  [drm] GPU HANG: ecode 12:4:00000000, in gem_exec_balanc [6388]
  [IGT] gem_exec_balancer: exiting, ret=0

The test causes a hang on both engines of a virtual engine context.
The engine instance zero hang gets a valid error capture but the
non-instance-zero hang does not.

Fix that by scanning through the list of pending register captures
when a hang notification for a virtual engine is received. That way,
the hang can be assigned to the correct physical engine prior to
starting the error capture process. So later on, when the error capture
handler tries to find the engine register list, it looks for one on
the correct engine.

Also, sneak in a missing blank line before a comment in the node
search code.

v2: Fix null pointer deref on non-GuC platforms.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428185636.457407-5-John.C.Harrison@Intel.com
---
 .../gpu/drm/i915/gt/uc/intel_guc_capture.c    | 31 ++++++++++++++++++
 .../gpu/drm/i915/gt/uc/intel_guc_capture.h    |  3 ++
 .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 32 ++++++++++++++++---
 drivers/gpu/drm/i915/i915_gpu_error.c         | 11 +++++--
 4 files changed, 70 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index 729a8fcf20dda..1def0b6467c79 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -1540,6 +1540,36 @@ void intel_guc_capture_free_node(struct intel_engine_coredump *ee)
 	ee->guc_capture_node = NULL;
 }
 
+bool intel_guc_capture_is_matching_engine(struct intel_gt *gt,
+					  struct intel_context *ce,
+					  struct intel_engine_cs *engine)
+{
+	struct __guc_capture_parsed_output *n;
+	struct intel_guc *guc;
+
+	if (!gt || !ce || !engine)
+		return false;
+
+	guc = &gt->uc.guc;
+	if (!guc->capture)
+		return false;
+
+	/*
+	 * Look for a matching GuC reported error capture node from
+	 * the internal output link-list based on lrca, guc-id and engine
+	 * identification.
+	 */
+	list_for_each_entry(n, &guc->capture->outlist, link) {
+		if (n->eng_inst == GUC_ID_TO_ENGINE_INSTANCE(engine->guc_id) &&
+		    n->eng_class == GUC_ID_TO_ENGINE_CLASS(engine->guc_id) &&
+		    n->guc_id == ce->guc_id.id &&
+		    (n->lrca & CTX_GTT_ADDRESS_MASK) == (ce->lrc.lrca & CTX_GTT_ADDRESS_MASK))
+			return true;
+	}
+
+	return false;
+}
+
 void intel_guc_capture_get_matching_node(struct intel_gt *gt,
 					 struct intel_engine_coredump *ee,
 					 struct intel_context *ce)
@@ -1555,6 +1585,7 @@ void intel_guc_capture_get_matching_node(struct intel_gt *gt,
 		return;
 
 	GEM_BUG_ON(ee->guc_capture_node);
+
 	/*
 	 * Look for a matching GuC reported error capture node from
 	 * the internal output link-list based on lrca, guc-id and engine
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.h
index fbd3713c7832d..302256d45431d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.h
@@ -11,6 +11,7 @@
 struct drm_i915_error_state_buf;
 struct guc_gt_system_info;
 struct intel_engine_coredump;
+struct intel_engine_cs;
 struct intel_context;
 struct intel_gt;
 struct intel_guc;
@@ -20,6 +21,8 @@ int intel_guc_capture_print_engine_node(struct drm_i915_error_state_buf *m,
 					const struct intel_engine_coredump *ee);
 void intel_guc_capture_get_matching_node(struct intel_gt *gt, struct intel_engine_coredump *ee,
 					 struct intel_context *ce);
+bool intel_guc_capture_is_matching_engine(struct intel_gt *gt, struct intel_context *ce,
+					  struct intel_engine_cs *engine);
 void intel_guc_capture_process(struct intel_guc *guc);
 int intel_guc_capture_getlist(struct intel_guc *guc, u32 owner, u32 type, u32 classid,
 			      void **outptr);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index c869ddc73e695..a0e3ef1c65d24 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -4728,13 +4728,37 @@ static void capture_error_state(struct intel_guc *guc,
 {
 	struct intel_gt *gt = guc_to_gt(guc);
 	struct drm_i915_private *i915 = gt->i915;
-	struct intel_engine_cs *engine = __context_to_physical_engine(ce);
 	intel_wakeref_t wakeref;
+	intel_engine_mask_t engine_mask;
+
+	if (intel_engine_is_virtual(ce->engine)) {
+		struct intel_engine_cs *e;
+		intel_engine_mask_t tmp, virtual_mask = ce->engine->mask;
+
+		engine_mask = 0;
+		for_each_engine_masked(e, ce->engine->gt, virtual_mask, tmp) {
+			bool match = intel_guc_capture_is_matching_engine(gt, ce, e);
+
+			if (match) {
+				intel_engine_set_hung_context(e, ce);
+				engine_mask |= e->mask;
+				atomic_inc(&i915->gpu_error.reset_engine_count[e->uabi_class]);
+			}
+		}
+
+		if (!engine_mask) {
+			guc_warn(guc, "No matching physical engine capture for virtual engine context 0x%04X / %s",
+				 ce->guc_id.id, ce->engine->name);
+			engine_mask = ~0U;
+		}
+	} else {
+		intel_engine_set_hung_context(ce->engine, ce);
+		engine_mask = ce->engine->mask;
+		atomic_inc(&i915->gpu_error.reset_engine_count[ce->engine->uabi_class]);
+	}
 
-	intel_engine_set_hung_context(engine, ce);
 	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
-		i915_capture_error_state(gt, engine->mask, CORE_DUMP_FLAG_IS_GUC_CAPTURE);
-	atomic_inc(&i915->gpu_error.reset_engine_count[engine->uabi_class]);
+		i915_capture_error_state(gt, engine_mask, CORE_DUMP_FLAG_IS_GUC_CAPTURE);
 }
 
 static void guc_context_replay(struct intel_context *ce)
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index f020c0086fbcd..7360046b99455 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -808,10 +808,15 @@ static void err_print_gt_engines(struct drm_i915_error_state_buf *m,
 	for (ee = gt->engine; ee; ee = ee->next) {
 		const struct i915_vma_coredump *vma;
 
-		if (ee->guc_capture_node)
-			intel_guc_capture_print_engine_node(m, ee);
-		else
+		if (gt->uc && gt->uc->guc.is_guc_capture) {
+			if (ee->guc_capture_node)
+				intel_guc_capture_print_engine_node(m, ee);
+			else
+				err_printf(m, "  Missing GuC capture node for %s\n",
+					   ee->engine->name);
+		} else {
 			error_print_engine(m, ee);
+		}
 
 		err_printf(m, "  hung: %u\n", ee->hung);
 		err_printf(m, "  engine reset count: %u\n", ee->reset_count);
-- 
GitLab


From 8a91b29f1f50ce7742cdbe5cf11d17f128511f3f Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Thu, 4 May 2023 08:53:16 +0200
Subject: [PATCH 0363/1791] drm/bridge: ti-sn65dsi83: Fix enable error path

If PLL locking failed, the regulator needs to be disabled again.

Fixes: 5664e3c907e2 ("drm/bridge: ti-sn65dsi83: Add vcc supply regulator support")
Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504065316.2640739-1-alexander.stein@ew.tq-group.com
---
 drivers/gpu/drm/bridge/ti-sn65dsi83.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
index 75286c9afbb96..1f5c07989e2bf 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
@@ -478,6 +478,7 @@ static void sn65dsi83_atomic_enable(struct drm_bridge *bridge,
 		dev_err(ctx->dev, "failed to lock PLL, ret=%i\n", ret);
 		/* On failure, disable PLL again and exit. */
 		regmap_write(ctx->regmap, REG_RC_PLL_EN, 0x00);
+		regulator_disable(ctx->vcc);
 		return;
 	}
 
-- 
GitLab


From 67f2dd9f38abf4d994a5bc5131ce684e594d66a8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:18 +0300
Subject: [PATCH 0364/1791] drm/i915/gvt: fix intel_vgpu_alloc_resource()
 kernel-doc parameter

drivers/gpu/drm/i915/gvt/aperture_gm.c:344: warning: Function parameter or member 'conf' not described in 'intel_vgpu_alloc_resource'
drivers/gpu/drm/i915/gvt/aperture_gm.c:344: warning: Excess function parameter 'param' description in 'intel_vgpu_alloc_resource'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9213214c9caa296ebd349a5d5b44c2bbb45cdf99.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gvt/aperture_gm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c
index 076c779f776a6..eedd1865bb98c 100644
--- a/drivers/gpu/drm/i915/gvt/aperture_gm.c
+++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c
@@ -330,7 +330,7 @@ void intel_vgpu_reset_resource(struct intel_vgpu *vgpu)
 /**
  * intel_vgpu_alloc_resource() - allocate HW resource for a vGPU
  * @vgpu: vGPU
- * @param: vGPU creation params
+ * @conf: vGPU creation params
  *
  * This function is used to allocate HW resource for a vGPU. User specifies
  * the resource configuration through the creation params.
-- 
GitLab


From e1172b617a93c2133750c4ff586f20f46e1e42c0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:19 +0300
Subject: [PATCH 0365/1791] drm/i915/vma: fix kernel-doc function name for
 i915_vma_size()

drivers/gpu/drm/i915/i915_vma.h:145: warning: expecting prototype for i915_vma_offset(). Prototype was for i915_vma_size() instead

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9fd58398ba6f86c55cc7a7c62efeab70c3311d59.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_vma.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
index ed5c9d682a1b2..38c8c66ed724a 100644
--- a/drivers/gpu/drm/i915/i915_vma.h
+++ b/drivers/gpu/drm/i915/i915_vma.h
@@ -132,7 +132,7 @@ static inline u64 __i915_vma_size(const struct i915_vma *vma)
 }
 
 /**
- * i915_vma_offset - Obtain the va range size of the vma
+ * i915_vma_size - Obtain the va range size of the vma
  * @vma: The vma
  *
  * GPU virtual address space may be allocated with padding. This
-- 
GitLab


From 144c3f7b190981a659b8fe9179ddd1f46baadee7 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:20 +0300
Subject: [PATCH 0366/1791] drm/i915/utils: drop kernel-doc from __wait_for()

The parameters aren't documented, and the file isn't included in Sphinx
build anyway, so demote the kernel-doc to a regular comment.

drivers/gpu/drm/i915/i915_utils.h:284: warning: Function parameter or member 'OP' not described in '__wait_for'
drivers/gpu/drm/i915/i915_utils.h:284: warning: Function parameter or member 'COND' not described in '__wait_for'
drivers/gpu/drm/i915/i915_utils.h:284: warning: Function parameter or member 'US' not described in '__wait_for'
drivers/gpu/drm/i915/i915_utils.h:284: warning: Function parameter or member 'Wmin' not described in '__wait_for'
drivers/gpu/drm/i915/i915_utils.h:284: warning: Function parameter or member 'Wmax' not described in '__wait_for'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e6e33a7f03c4a78739fa96e6ae74eb272ae147e7.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_utils.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 2c430c0c3badd..c61066498bf2f 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -250,7 +250,7 @@ wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms)
 	}
 }
 
-/**
+/*
  * __wait_for - magic wait macro
  *
  * Macro to help avoid open coding check/wait/timeout patterns. Note that it's
-- 
GitLab


From 28487ecb96b0bd1ff27a3e8ae8bc027af6e448e1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:21 +0300
Subject: [PATCH 0367/1791] drm/i915/vma: document struct i915_vma_resource
 wakeref member

drivers/gpu/drm/i915/i915_vma_resource.h:129: warning: Function parameter or member 'wakeref' not described in 'i915_vma_resource'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/50389365e095dc564ab5f1f1e3647934163ffefa.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_vma_resource.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/i915_vma_resource.h b/drivers/gpu/drm/i915/i915_vma_resource.h
index c1864e3d0b43e..2053c037dbfb1 100644
--- a/drivers/gpu/drm/i915/i915_vma_resource.h
+++ b/drivers/gpu/drm/i915/i915_vma_resource.h
@@ -47,6 +47,7 @@ struct i915_page_sizes {
  * @chain: Pointer to struct i915_sw_fence used to await dependencies.
  * @rb: Rb node for the vm's pending unbind interval tree.
  * @__subtree_last: Interval tree private member.
+ * @wakeref: wakeref.
  * @vm: non-refcounted pointer to the vm. This is for internal use only and
  * this member is cleared after vm_resource unbind.
  * @mr: The memory region of the object pointed to by the vma.
-- 
GitLab


From 9570b039075192923d3bd26a7892a708eef1a483 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:23 +0300
Subject: [PATCH 0368/1791] drm/i915/perf: fix i915_perf_ioctl_version()
 kernel-doc

drivers/gpu/drm/i915/i915_perf.c:5307: warning: Function parameter or member 'i915' not described in 'i915_perf_ioctl_version'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b93ddb95a15d1376936349b32c7facb35c76be82.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_perf.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 050b8ae7b8e70..19d5652300eeb 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -5300,6 +5300,7 @@ void i915_perf_fini(struct drm_i915_private *i915)
 
 /**
  * i915_perf_ioctl_version - Version of the i915-perf subsystem
+ * @i915: The i915 device
  *
  * This version number is used by userspace to detect available features.
  */
-- 
GitLab


From 88629feedcc4678ac21991a1154477d1c1dca19f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:24 +0300
Subject: [PATCH 0369/1791] drm/i915/error: fix i915_capture_error_state()
 kernel-doc

drivers/gpu/drm/i915/i915_gpu_error.c:2174: warning: Function parameter or member 'dump_flags' not described in 'i915_capture_error_state'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20524292b002800975d82d23b5bd47da878f1733.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_gpu_error.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 7360046b99455..8ffdd7f47e933 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -2167,7 +2167,7 @@ void i915_error_state_store(struct i915_gpu_coredump *error)
  * i915_capture_error_state - capture an error record for later analysis
  * @gt: intel_gt which originated the hang
  * @engine_mask: hung engines
- *
+ * @dump_flags: dump flags
  *
  * Should be called when an error is detected (either a hang or an error
  * interrupt) to capture error state from the time of the error.  Fills
-- 
GitLab


From d7b7332c0e9901ec7e6e73ca75b6c4cd398d5aaf Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:25 +0300
Subject: [PATCH 0370/1791] drm/i915/request: drop kernel-doc

The documentation is closer to not being kernel-doc. Just drop the
kernel-doc /** indicators.

drivers/gpu/drm/i915/i915_request.h:176: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Request queue structure.
drivers/gpu/drm/i915/i915_request.h:477: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Returns true if seq1 is later than seq2.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/62a7a0f6e5b33e59ef4339304d01e7ef7ee04430.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_request.h | 52 ++++++++++++++---------------
 1 file changed, 26 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index f5e1bb5e857aa..0ac55b2e4223e 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -172,7 +172,7 @@ enum {
 	I915_FENCE_FLAG_COMPOSITE,
 };
 
-/**
+/*
  * Request queue structure.
  *
  * The request queue allows us to note sequence numbers that have been emitted
@@ -198,7 +198,7 @@ struct i915_request {
 
 	struct drm_i915_private *i915;
 
-	/**
+	/*
 	 * Context and ring buffer related to this request
 	 * Contexts are refcounted, so when this request is associated with a
 	 * context, we must increment the context's refcount, to guarantee that
@@ -251,9 +251,9 @@ struct i915_request {
 	};
 	struct llist_head execute_cb;
 	struct i915_sw_fence semaphore;
-	/**
-	 * @submit_work: complete submit fence from an IRQ if needed for
-	 * locking hierarchy reasons.
+	/*
+	 * complete submit fence from an IRQ if needed for locking hierarchy
+	 * reasons.
 	 */
 	struct irq_work submit_work;
 
@@ -277,35 +277,35 @@ struct i915_request {
 	 */
 	const u32 *hwsp_seqno;
 
-	/** Position in the ring of the start of the request */
+	/* Position in the ring of the start of the request */
 	u32 head;
 
-	/** Position in the ring of the start of the user packets */
+	/* Position in the ring of the start of the user packets */
 	u32 infix;
 
-	/**
+	/*
 	 * Position in the ring of the start of the postfix.
 	 * This is required to calculate the maximum available ring space
 	 * without overwriting the postfix.
 	 */
 	u32 postfix;
 
-	/** Position in the ring of the end of the whole request */
+	/* Position in the ring of the end of the whole request */
 	u32 tail;
 
-	/** Position in the ring of the end of any workarounds after the tail */
+	/* Position in the ring of the end of any workarounds after the tail */
 	u32 wa_tail;
 
-	/** Preallocate space in the ring for the emitting the request */
+	/* Preallocate space in the ring for the emitting the request */
 	u32 reserved_space;
 
-	/** Batch buffer pointer for selftest internal use. */
+	/* Batch buffer pointer for selftest internal use. */
 	I915_SELFTEST_DECLARE(struct i915_vma *batch);
 
 	struct i915_vma_resource *batch_res;
 
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
-	/**
+	/*
 	 * Additional buffers requested by userspace to be captured upon
 	 * a GPU hang. The vma/obj on this list are protected by their
 	 * active reference - all objects on this list must also be
@@ -314,29 +314,29 @@ struct i915_request {
 	struct i915_capture_list *capture_list;
 #endif
 
-	/** Time at which this request was emitted, in jiffies. */
+	/* Time at which this request was emitted, in jiffies. */
 	unsigned long emitted_jiffies;
 
-	/** timeline->request entry for this request */
+	/* timeline->request entry for this request */
 	struct list_head link;
 
-	/** Watchdog support fields. */
+	/* Watchdog support fields. */
 	struct i915_request_watchdog {
 		struct llist_node link;
 		struct hrtimer timer;
 	} watchdog;
 
-	/**
-	 * @guc_fence_link: Requests may need to be stalled when using GuC
-	 * submission waiting for certain GuC operations to complete. If that is
-	 * the case, stalled requests are added to a per context list of stalled
-	 * requests. The below list_head is the link in that list. Protected by
+	/*
+	 * Requests may need to be stalled when using GuC submission waiting for
+	 * certain GuC operations to complete. If that is the case, stalled
+	 * requests are added to a per context list of stalled requests. The
+	 * below list_head is the link in that list. Protected by
 	 * ce->guc_state.lock.
 	 */
 	struct list_head guc_fence_link;
 
-	/**
-	 * @guc_prio: Priority level while the request is in flight. Differs
+	/*
+	 * Priority level while the request is in flight. Differs
 	 * from i915 scheduler priority. See comment above
 	 * I915_SCHEDULER_CAP_STATIC_PRIORITY_MAP for details. Protected by
 	 * ce->guc_active.lock. Two special values (GUC_PRIO_INIT and
@@ -348,8 +348,8 @@ struct i915_request {
 #define	GUC_PRIO_FINI	0xfe
 	u8 guc_prio;
 
-	/**
-	 * @hucq: wait queue entry used to wait on the HuC load to complete
+	/*
+	 * wait queue entry used to wait on the HuC load to complete
 	 */
 	wait_queue_entry_t hucq;
 
@@ -473,7 +473,7 @@ i915_request_has_initial_breadcrumb(const struct i915_request *rq)
 	return test_bit(I915_FENCE_FLAG_INITIAL_BREADCRUMB, &rq->fence.flags);
 }
 
-/**
+/*
  * Returns true if seq1 is later than seq2.
  */
 static inline bool i915_seqno_passed(u32 seq1, u32 seq2)
-- 
GitLab


From faa19ce89b0b2eb91c37e2c26b6be93e2e2cb47a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:26 +0300
Subject: [PATCH 0371/1791] drm/i915/gem: fix i915_gem_object_lookup_rcu()
 kernel-doc parameter name

drivers/gpu/drm/i915/gem/i915_gem_object.h:94: warning: Function parameter or member 'file' not described in 'i915_gem_object_lookup_rcu'
drivers/gpu/drm/i915/gem/i915_gem_object.h:94: warning: Excess function parameter 'filp' description in 'i915_gem_object_lookup_rcu'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/80c96863b7a0755aaa07efb49bcccd9ba620c6d7.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_object.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 885ccde9dc3c0..bc1291887d4f9 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -80,7 +80,7 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj);
 
 /**
  * i915_gem_object_lookup_rcu - look up a temporary GEM object from its handle
- * @filp: DRM file private date
+ * @file: DRM file private date
  * @handle: userspace handle
  *
  * Returns:
-- 
GitLab


From 9c55105be0a78942328882224c23965483e578a9 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:29 +0300
Subject: [PATCH 0372/1791] drm/i915/engine: fix kernel-doc function name for
 intel_engine_cleanup_common()

drivers/gpu/drm/i915/gt/intel_engine_cs.c:1525: warning: expecting prototype for intel_engines_cleanup_common(). Prototype was for intel_engine_cleanup_common() instead

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/33f8dfdf38be3e16675971e6983e3e300d4301a6.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_engine_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 5c6c9a6d469c5..0aff5bb13c538 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -1515,7 +1515,7 @@ int intel_engines_init(struct intel_gt *gt)
 }
 
 /**
- * intel_engines_cleanup_common - cleans up the engine state created by
+ * intel_engine_cleanup_common - cleans up the engine state created by
  *                                the common initiailizers.
  * @engine: Engine to cleanup.
  *
-- 
GitLab


From 71ca9b87a72bdc1dbe434d25c97eb14be58b4e00 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:30 +0300
Subject: [PATCH 0373/1791] drm/i915/context: fix kernel-doc parameter
 descriptions

drivers/gpu/drm/i915/gt/intel_context.h:108: warning: Function parameter or member 'ce' not described in 'intel_context_lock_pinned'
drivers/gpu/drm/i915/gt/intel_context.h:123: warning: Function parameter or member 'ce' not described in 'intel_context_is_pinned'
drivers/gpu/drm/i915/gt/intel_context.h:142: warning: Function parameter or member 'ce' not described in 'intel_context_unlock_pinned'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/8359a1cef6b5ab268a9dcc1a382281b6e39cfa64.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_context.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h
index 43ed92f8dc83c..a80e3b7c24ff9 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.h
+++ b/drivers/gpu/drm/i915/gt/intel_context.h
@@ -97,7 +97,7 @@ void intel_context_bind_parent_child(struct intel_context *parent,
 
 /**
  * intel_context_lock_pinned - Stablises the 'pinned' status of the HW context
- * @ce - the context
+ * @ce: the context
  *
  * Acquire a lock on the pinned status of the HW context, such that the context
  * can neither be bound to the GPU or unbound whilst the lock is held, i.e.
@@ -111,7 +111,7 @@ static inline int intel_context_lock_pinned(struct intel_context *ce)
 
 /**
  * intel_context_is_pinned - Reports the 'pinned' status
- * @ce - the context
+ * @ce: the context
  *
  * While in use by the GPU, the context, along with its ring and page
  * tables is pinned into memory and the GTT.
@@ -133,7 +133,7 @@ static inline void intel_context_cancel_request(struct intel_context *ce,
 
 /**
  * intel_context_unlock_pinned - Releases the earlier locking of 'pinned' status
- * @ce - the context
+ * @ce: the context
  *
  * Releases the lock earlier acquired by intel_context_unlock_pinned().
  */
-- 
GitLab


From e798a3b30dcb729fa9b9dc88de00984dc1f5aa04 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:31 +0300
Subject: [PATCH 0374/1791] drm/i915/gtt: fix i915_vm_resv_put() kernel-doc
 parameter name

drivers/gpu/drm/i915/gt/intel_gtt.h:515: warning: Function parameter or member 'vm' not described in 'i915_vm_resv_put'
drivers/gpu/drm/i915/gt/intel_gtt.h:515: warning: Excess function parameter 'resv' description in 'i915_vm_resv_put'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/c22e58e770019667980b3617f6e963b76d7e79a7.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gtt.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index 1910683f03b4d..9aff343beaa8b 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -509,7 +509,7 @@ static inline void i915_vm_put(struct i915_address_space *vm)
 
 /**
  * i915_vm_resv_put - Release a reference on the vm's reservation lock
- * @resv: Pointer to a reservation lock obtained from i915_vm_resv_get()
+ * @vm: The vm whose reservation lock reference we want to release
  */
 static inline void i915_vm_resv_put(struct i915_address_space *vm)
 {
-- 
GitLab


From 1f5cf999bedcdff85f84b7dc4d780cbb515b6c2f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:32 +0300
Subject: [PATCH 0375/1791] drm/i915/engine: hide preempt_hang selftest member
 from kernel-doc

drivers/gpu/drm/i915/gt/intel_engine_types.h:293: warning: Function parameter or member 'preempt_hang' not described in 'intel_engine_execlists'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/dafd771bb75cf14965dd3b666987c58a438de134.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_engine_types.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 960291f88fd6c..e99a6fa03d453 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -289,6 +289,7 @@ struct intel_engine_execlists {
 	 */
 	u8 csb_head;
 
+	/* private: selftest */
 	I915_SELFTEST_DECLARE(struct st_preempt_hang preempt_hang;)
 };
 
-- 
GitLab


From 3a21c6b4298d9d0c99c8dee28edcf317d68cd93e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:33 +0300
Subject: [PATCH 0376/1791] drm/i915/guc: add dbgfs_node member kernel-doc

drivers/gpu/drm/i915/gt/uc/intel_guc.h:274: warning: Function parameter or member 'dbgfs_node' not described in 'intel_guc'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b0f681dd82289dd86da78c6242411e8d812e51a1.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.h b/drivers/gpu/drm/i915/gt/uc/intel_guc.h
index e46aac1a41e6d..8dc291ff00935 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.h
@@ -42,6 +42,7 @@ struct intel_guc {
 	/** @capture: the error-state-capture module's data and objects */
 	struct intel_guc_state_capture *capture;
 
+	/** @dbgfs_node: debugfs node */
 	struct dentry *dbgfs_node;
 
 	/** @sched_engine: Global engine used to submit requests to GuC */
-- 
GitLab


From f05e526e44cb11e70c5b7d8bb55d7e0c6fb10990 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:34 +0300
Subject: [PATCH 0377/1791] drm/i915/guc: drop lots of kernel-doc markers

The documentation is closer to not being kernel-doc, so just drop the
kernel-doc markers.

drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:27: warning: Function parameter or member 'size' not described in '__guc_capture_bufstate'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:27: warning: Function parameter or member 'data' not described in '__guc_capture_bufstate'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:27: warning: Function parameter or member 'rd' not described in '__guc_capture_bufstate'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:27: warning: Function parameter or member 'wr' not described in '__guc_capture_bufstate'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'link' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'is_partial' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'eng_class' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'eng_inst' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'guc_id' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'lrca' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:59: warning: Function parameter or member 'reginfo' not described in '__guc_capture_parsed_output'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:62: warning: wrong kernel-doc identifier on line:
 * struct guc_debug_capture_list_header / struct guc_debug_capture_list
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:80: warning: wrong kernel-doc identifier on line:
 * struct __guc_mmio_reg_descr / struct __guc_mmio_reg_descr_group
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:105: warning: wrong kernel-doc identifier on line:
 * struct guc_state_capture_header_t / struct guc_state_capture_t /
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:163: warning: Function parameter or member 'is_valid' not described in '__guc_capture_ads_cache'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:163: warning: Function parameter or member 'ptr' not described in '__guc_capture_ads_cache'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:163: warning: Function parameter or member 'size' not described in '__guc_capture_ads_cache'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:163: warning: Function parameter or member 'status' not described in '__guc_capture_ads_cache'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'marker' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'read_ptr' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'write_ptr' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'size' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'sampled_write_ptr' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'wrap_offset' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'flush_to_file' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'buffer_full_cnt' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'reserved' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'flags' not described in 'guc_log_buffer_state'
drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h:491: warning: Function parameter or member 'version' not described in 'guc_log_buffer_state'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9c210d53fdbd6da5fac42e435855d269504919d7.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h | 12 ++++++------
 drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h   |  2 +-
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h b/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
index 9d589c28f40fb..1b6c219e1675c 100644
--- a/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
+++ b/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
@@ -12,7 +12,7 @@
 struct intel_guc;
 struct file;
 
-/**
+/*
  * struct __guc_capture_bufstate
  *
  * Book-keeping structure used to track read and write pointers
@@ -26,7 +26,7 @@ struct __guc_capture_bufstate {
 	u32 wr;
 };
 
-/**
+/*
  * struct __guc_capture_parsed_output - extracted error capture node
  *
  * A single unit of extracted error-capture output data grouped together
@@ -58,7 +58,7 @@ struct __guc_capture_parsed_output {
 #define GCAP_PARSED_REGLIST_INDEX_ENGINST  BIT(GUC_CAPTURE_LIST_TYPE_ENGINE_INSTANCE)
 };
 
-/**
+/*
  * struct guc_debug_capture_list_header / struct guc_debug_capture_list
  *
  * As part of ADS registration, these header structures (followed by
@@ -76,7 +76,7 @@ struct guc_debug_capture_list {
 	struct guc_mmio_reg regs[];
 } __packed;
 
-/**
+/*
  * struct __guc_mmio_reg_descr / struct __guc_mmio_reg_descr_group
  *
  * intel_guc_capture module uses these structures to maintain static
@@ -101,7 +101,7 @@ struct __guc_mmio_reg_descr_group {
 	struct __guc_mmio_reg_descr *extlist; /* only used for steered registers */
 };
 
-/**
+/*
  * struct guc_state_capture_header_t / struct guc_state_capture_t /
  * guc_state_capture_group_header_t / guc_state_capture_group_t
  *
@@ -148,7 +148,7 @@ struct guc_state_capture_group_t {
 	struct guc_state_capture_t capture_entries[];
 } __packed;
 
-/**
+/*
  * struct __guc_capture_ads_cache
  *
  * A structure to cache register lists that were populated and registered
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
index 4ae5fc2f60023..8bb9ed7c7b4dd 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
@@ -451,7 +451,7 @@ enum guc_log_buffer_type {
 	GUC_MAX_LOG_BUFFER
 };
 
-/**
+/*
  * struct guc_log_buffer_state - GuC log buffer state
  *
  * Below state structure is used for coordination of retrieval of GuC firmware
-- 
GitLab


From 1c519e0628f19e2790ddfa4cb512921047e8f65a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:35 +0300
Subject: [PATCH 0378/1791] drm/i915/guc: add intel_guc_state_capture member
 docs for ads_null_cache and max_mmio_per_node

drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:216: warning: Function parameter or member 'ads_null_cache' not described in 'intel_guc_state_capture'
drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h:216: warning: Function parameter or member 'max_mmio_per_node' not described in 'intel_guc_state_capture'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/c83878163221ed3684a6de5d5e1c5373ddd5c06f.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h b/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
index 1b6c219e1675c..1fc0c17b12309 100644
--- a/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
+++ b/drivers/gpu/drm/i915/gt/uc/guc_capture_fwif.h
@@ -187,6 +187,10 @@ struct intel_guc_state_capture {
 	struct __guc_capture_ads_cache ads_cache[GUC_CAPTURE_LIST_INDEX_MAX]
 						[GUC_CAPTURE_LIST_TYPE_MAX]
 						[GUC_MAX_ENGINE_CLASSES];
+
+	/**
+	 * @ads_null_cache: ADS null cache.
+	 */
 	void *ads_null_cache;
 
 	/**
@@ -202,6 +206,10 @@ struct intel_guc_state_capture {
 	struct list_head cachelist;
 #define PREALLOC_NODES_MAX_COUNT (3 * GUC_MAX_ENGINE_CLASSES * GUC_MAX_INSTANCES_PER_CLASS)
 #define PREALLOC_NODES_DEFAULT_NUMREGS 64
+
+	/**
+	 * @max_mmio_per_node: Max MMIO per node.
+	 */
 	int max_mmio_per_node;
 
 	/**
-- 
GitLab


From 8802628237ac73bf5a6f878ea0cbd8a4c39a55a1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:36 +0300
Subject: [PATCH 0379/1791] drm/i915/active: fix kernel-doc for function
 parameters

drivers/gpu/drm/i915/i915_active.h:66: warning: Function parameter or member 'active' not described in '__i915_active_fence_init'
drivers/gpu/drm/i915/i915_active.h:66: warning: Function parameter or member 'fence' not described in '__i915_active_fence_init'
drivers/gpu/drm/i915/i915_active.h:66: warning: Function parameter or member 'fn' not described in '__i915_active_fence_init'
drivers/gpu/drm/i915/i915_active.h:89: warning: Function parameter or member 'active' not described in 'i915_active_fence_set'
drivers/gpu/drm/i915/i915_active.h:89: warning: Function parameter or member 'rq' not described in 'i915_active_fence_set'
drivers/gpu/drm/i915/i915_active.h:102: warning: Function parameter or member 'active' not described in 'i915_active_fence_get'
drivers/gpu/drm/i915/i915_active.h:122: warning: Function parameter or member 'active' not described in 'i915_active_fence_isset'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/7f201c6cb715e4b3cece78ffa893a75610ecd27d.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_active.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_active.h b/drivers/gpu/drm/i915/i915_active.h
index 7eb44132183aa..77c676ecc2633 100644
--- a/drivers/gpu/drm/i915/i915_active.h
+++ b/drivers/gpu/drm/i915/i915_active.h
@@ -49,9 +49,9 @@ void i915_active_noop(struct dma_fence *fence, struct dma_fence_cb *cb);
 
 /**
  * __i915_active_fence_init - prepares the activity tracker for use
- * @active - the active tracker
- * @fence - initial fence to track, can be NULL
- * @func - a callback when then the tracker is retired (becomes idle),
+ * @active: the active tracker
+ * @fence: initial fence to track, can be NULL
+ * @fn: a callback when then the tracker is retired (becomes idle),
  *         can be NULL
  *
  * i915_active_fence_init() prepares the embedded @active struct for use as
@@ -77,8 +77,8 @@ __i915_active_fence_set(struct i915_active_fence *active,
 
 /**
  * i915_active_fence_set - updates the tracker to watch the current fence
- * @active - the active tracker
- * @rq - the request to watch
+ * @active: the active tracker
+ * @rq: the request to watch
  *
  * i915_active_fence_set() watches the given @rq for completion. While
  * that @rq is busy, the @active reports busy. When that @rq is signaled
@@ -89,7 +89,7 @@ i915_active_fence_set(struct i915_active_fence *active,
 		      struct i915_request *rq);
 /**
  * i915_active_fence_get - return a reference to the active fence
- * @active - the active tracker
+ * @active: the active tracker
  *
  * i915_active_fence_get() returns a reference to the active fence,
  * or NULL if the active tracker is idle. The reference is obtained under RCU,
@@ -111,7 +111,7 @@ i915_active_fence_get(struct i915_active_fence *active)
 
 /**
  * i915_active_fence_isset - report whether the active tracker is assigned
- * @active - the active tracker
+ * @active: the active tracker
  *
  * i915_active_fence_isset() returns true if the active tracker is currently
  * assigned to a fence. Due to the lazy retiring, that fence may be idle
-- 
GitLab


From 7bc30374f84c5588c952e47248595b1c38d06906 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:37 +0300
Subject: [PATCH 0380/1791] drm/i915/pmu: drop kernel-doc

The comments are closer to not being kernel-doc.

drivers/gpu/drm/i915/i915_pmu.h:21: warning: cannot understand function prototype: 'enum i915_pmu_tracked_events '
drivers/gpu/drm/i915/i915_pmu.h:32: warning: cannot understand function prototype: 'enum '
drivers/gpu/drm/i915/i915_pmu.h:41: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * How many different events we track in the global PMU mask.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/7656b8f58b088c108a2a32f7089329740efabba1.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index 449057648f39b..c30f43319a783 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -14,7 +14,7 @@
 
 struct drm_i915_private;
 
-/**
+/*
  * Non-engine events that we need to track enabled-disabled transition and
  * current state.
  */
@@ -25,7 +25,7 @@ enum i915_pmu_tracked_events {
 	__I915_PMU_TRACKED_EVENT_COUNT, /* count marker */
 };
 
-/**
+/*
  * Slots used from the sampling timer (non-engine events) with some extras for
  * convenience.
  */
@@ -37,7 +37,7 @@ enum {
 	__I915_NUM_PMU_SAMPLERS
 };
 
-/**
+/*
  * How many different events we track in the global PMU mask.
  *
  * It is also used to know to needed number of event reference counters.
-- 
GitLab


From 70b9933c09b8687edff5c2f833dc2a72c1b40362 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:38 +0300
Subject: [PATCH 0381/1791] drm/i915/pxp: fix kernel-doc for member dev_link

Add /** to make it a kernel-doc.

drivers/gpu/drm/i915/pxp/intel_pxp_types.h:96: warning: Function parameter or member 'dev_link' not described in 'intel_pxp'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/84f89a3332fa323888f1e3241fb51ae10417ecd7.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/pxp/intel_pxp_types.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
index 007de49e1ea42..c445f7f2f47ac 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
@@ -33,7 +33,9 @@ struct intel_pxp {
 	 */
 	struct i915_pxp_component *pxp_component;
 
-	/* @dev_link: Enforce module relationship for power management ordering. */
+	/**
+	 * @dev_link: Enforce module relationship for power management ordering.
+	 */
 	struct device_link *dev_link;
 	/**
 	 * @pxp_component_added: track if the pxp component has been added.
-- 
GitLab


From 8b2a7394e51d6a49edeae8fc640b8c23a4ab649d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:40 +0300
Subject: [PATCH 0382/1791] drm/i915/scatterlist: fix kernel-doc parameter
 documentation

drivers/gpu/drm/i915/i915_scatterlist.h:164: warning: Function parameter or member 'release' not described in 'i915_refct_sgt_ops'
drivers/gpu/drm/i915/i915_scatterlist.h:187: warning: Function parameter or member 'rsgt' not described in 'i915_refct_sgt_put'
drivers/gpu/drm/i915/i915_scatterlist.h:198: warning: Function parameter or member 'rsgt' not described in 'i915_refct_sgt_get'
drivers/gpu/drm/i915/i915_scatterlist.h:214: warning: Function parameter or member 'rsgt' not described in '__i915_refct_sgt_init'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/0b31edbb33116c8002dd1e72d3ad25efe5dd0176.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_scatterlist.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_scatterlist.h b/drivers/gpu/drm/i915/i915_scatterlist.h
index b0a1db44f8950..12c1fd63e9b8d 100644
--- a/drivers/gpu/drm/i915/i915_scatterlist.h
+++ b/drivers/gpu/drm/i915/i915_scatterlist.h
@@ -181,7 +181,7 @@ struct i915_refct_sgt {
 
 /**
  * i915_refct_sgt_put - Put a refcounted sg-table
- * @rsgt the struct i915_refct_sgt to put.
+ * @rsgt: the struct i915_refct_sgt to put.
  */
 static inline void i915_refct_sgt_put(struct i915_refct_sgt *rsgt)
 {
@@ -191,7 +191,7 @@ static inline void i915_refct_sgt_put(struct i915_refct_sgt *rsgt)
 
 /**
  * i915_refct_sgt_get - Get a refcounted sg-table
- * @rsgt the struct i915_refct_sgt to get.
+ * @rsgt: the struct i915_refct_sgt to get.
  */
 static inline struct i915_refct_sgt *
 i915_refct_sgt_get(struct i915_refct_sgt *rsgt)
@@ -203,7 +203,7 @@ i915_refct_sgt_get(struct i915_refct_sgt *rsgt)
 /**
  * __i915_refct_sgt_init - Initialize a refcounted sg-list with a custom
  * operations structure
- * @rsgt The struct i915_refct_sgt to initialize.
+ * @rsgt: The struct i915_refct_sgt to initialize.
  * @size: Size in bytes of the underlying memory buffer.
  * @ops: A customized operations structure in case the refcounted sg-list
  * is embedded into another structure.
-- 
GitLab


From e7b05d95cc1bfa1e61557358d936ebb33b0ae4be Mon Sep 17 00:00:00 2001
From: Alexander Usyskin <alexander.usyskin@intel.com>
Date: Tue, 2 May 2023 09:38:51 -0700
Subject: [PATCH 0383/1791] drm/i915/mtl: Define GSC Proxy component interface

GSC Proxy component is used for communication between the
Intel graphics driver and MEI driver.

Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502163854.317653-2-daniele.ceraolospurio@intel.com
---
 include/drm/i915_component.h               |  3 +-
 include/drm/i915_gsc_proxy_mei_interface.h | 53 ++++++++++++++++++++++
 2 files changed, 55 insertions(+), 1 deletion(-)
 create mode 100644 include/drm/i915_gsc_proxy_mei_interface.h

diff --git a/include/drm/i915_component.h b/include/drm/i915_component.h
index c1e2a43d2d1e4..56a84ee1c64c6 100644
--- a/include/drm/i915_component.h
+++ b/include/drm/i915_component.h
@@ -29,7 +29,8 @@
 enum i915_component_type {
 	I915_COMPONENT_AUDIO = 1,
 	I915_COMPONENT_HDCP,
-	I915_COMPONENT_PXP
+	I915_COMPONENT_PXP,
+	I915_COMPONENT_GSC_PROXY,
 };
 
 /* MAX_PORT is the number of port
diff --git a/include/drm/i915_gsc_proxy_mei_interface.h b/include/drm/i915_gsc_proxy_mei_interface.h
new file mode 100644
index 0000000000000..9462341d3ae1b
--- /dev/null
+++ b/include/drm/i915_gsc_proxy_mei_interface.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright (c) 2022-2023 Intel Corporation
+ */
+
+#ifndef _I915_GSC_PROXY_MEI_INTERFACE_H_
+#define _I915_GSC_PROXY_MEI_INTERFACE_H_
+
+#include <linux/types.h>
+
+struct device;
+struct module;
+
+/**
+ * struct i915_gsc_proxy_component_ops - ops for GSC Proxy services.
+ * @owner: Module providing the ops
+ * @send: sends a proxy message from GSC FW to ME FW
+ * @recv: receives a proxy message for GSC FW from ME FW
+ */
+struct i915_gsc_proxy_component_ops {
+	struct module *owner;
+
+	/**
+	 * send - Sends a proxy message to ME FW.
+	 * @dev: device struct corresponding to the mei device
+	 * @buf: message buffer to send
+	 * @size: size of the message
+	 * Return: bytes sent on success, negative errno value on failure
+	 */
+	int (*send)(struct device *dev, const void *buf, size_t size);
+
+	/**
+	 * recv - Receives a proxy message from ME FW.
+	 * @dev: device struct corresponding to the mei device
+	 * @buf: message buffer to contain the received message
+	 * @size: size of the buffer
+	 * Return: bytes received on success, negative errno value on failure
+	 */
+	int (*recv)(struct device *dev, void *buf, size_t size);
+};
+
+/**
+ * struct i915_gsc_proxy_component - Used for communication between i915 and
+ * MEI drivers for GSC proxy services
+ * @mei_dev: device that provide the GSC proxy service.
+ * @ops: Ops implemented by GSC proxy driver, used by i915 driver.
+ */
+struct i915_gsc_proxy_component {
+	struct device *mei_dev;
+	const struct i915_gsc_proxy_component_ops *ops;
+};
+
+#endif /* _I915_GSC_PROXY_MEI_INTERFACE_H_ */
-- 
GitLab


From 1dd924f6885b1f3bddc586c39467dc9471729c5e Mon Sep 17 00:00:00 2001
From: Alexander Usyskin <alexander.usyskin@intel.com>
Date: Tue, 2 May 2023 09:38:52 -0700
Subject: [PATCH 0384/1791] mei: gsc_proxy: add gsc proxy driver

Add GSC proxy driver. It to allows messaging between GSC component
on Intel graphics card and CSE device.

Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502163854.317653-3-daniele.ceraolospurio@intel.com
---
 drivers/misc/mei/Kconfig                   |   2 +-
 drivers/misc/mei/Makefile                  |   1 +
 drivers/misc/mei/gsc_proxy/Kconfig         |  14 ++
 drivers/misc/mei/gsc_proxy/Makefile        |   7 +
 drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c | 208 +++++++++++++++++++++
 5 files changed, 231 insertions(+), 1 deletion(-)
 create mode 100644 drivers/misc/mei/gsc_proxy/Kconfig
 create mode 100644 drivers/misc/mei/gsc_proxy/Makefile
 create mode 100644 drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c

diff --git a/drivers/misc/mei/Kconfig b/drivers/misc/mei/Kconfig
index d21486d69df27..37db142de413e 100644
--- a/drivers/misc/mei/Kconfig
+++ b/drivers/misc/mei/Kconfig
@@ -62,4 +62,4 @@ config INTEL_MEI_GSC
 
 source "drivers/misc/mei/hdcp/Kconfig"
 source "drivers/misc/mei/pxp/Kconfig"
-
+source "drivers/misc/mei/gsc_proxy/Kconfig"
diff --git a/drivers/misc/mei/Makefile b/drivers/misc/mei/Makefile
index fb740d7549008..14aee253ae482 100644
--- a/drivers/misc/mei/Makefile
+++ b/drivers/misc/mei/Makefile
@@ -30,3 +30,4 @@ CFLAGS_mei-trace.o = -I$(src)
 
 obj-$(CONFIG_INTEL_MEI_HDCP) += hdcp/
 obj-$(CONFIG_INTEL_MEI_PXP) += pxp/
+obj-$(CONFIG_INTEL_MEI_GSC_PROXY) += gsc_proxy/
diff --git a/drivers/misc/mei/gsc_proxy/Kconfig b/drivers/misc/mei/gsc_proxy/Kconfig
new file mode 100644
index 0000000000000..5f68d9f3d691b
--- /dev/null
+++ b/drivers/misc/mei/gsc_proxy/Kconfig
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2022-2023, Intel Corporation. All rights reserved.
+#
+config INTEL_MEI_GSC_PROXY
+	tristate "Intel GSC Proxy services of ME Interface"
+	select INTEL_MEI_ME
+	depends on DRM_I915
+	help
+         MEI Support for GSC Proxy Services on Intel platforms.
+
+         MEI GSC proxy enables messaging between GSC service on
+         Intel graphics card and services on CSE (MEI) firmware
+         residing SoC or PCH.
+
diff --git a/drivers/misc/mei/gsc_proxy/Makefile b/drivers/misc/mei/gsc_proxy/Makefile
new file mode 100644
index 0000000000000..358847e9aaa93
--- /dev/null
+++ b/drivers/misc/mei/gsc_proxy/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (c) 2022-2023, Intel Corporation. All rights reserved.
+#
+# Makefile - GSC Proxy client driver for Intel MEI Bus Driver.
+
+obj-$(CONFIG_INTEL_MEI_GSC_PROXY) += mei_gsc_proxy.o
diff --git a/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c b/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c
new file mode 100644
index 0000000000000..be52b113aea93
--- /dev/null
+++ b/drivers/misc/mei/gsc_proxy/mei_gsc_proxy.c
@@ -0,0 +1,208 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2022-2023 Intel Corporation
+ */
+
+/**
+ * DOC: MEI_GSC_PROXY Client Driver
+ *
+ * The mei_gsc_proxy driver acts as a translation layer between
+ * proxy user (I915) and ME FW by proxying messages to ME FW
+ */
+
+#include <linux/component.h>
+#include <linux/mei_cl_bus.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/uuid.h>
+#include <drm/drm_connector.h>
+#include <drm/i915_component.h>
+#include <drm/i915_gsc_proxy_mei_interface.h>
+
+/**
+ * mei_gsc_proxy_send - Sends a proxy message to ME FW.
+ * @dev: device corresponding to the mei_cl_device
+ * @buf: a message buffer to send
+ * @size: size of the message
+ * Return: bytes sent on Success, <0 on Failure
+ */
+static int mei_gsc_proxy_send(struct device *dev, const void *buf, size_t size)
+{
+	ssize_t ret;
+
+	if (!dev || !buf)
+		return -EINVAL;
+
+	ret = mei_cldev_send(to_mei_cl_device(dev), buf, size);
+	if (ret < 0)
+		dev_dbg(dev, "mei_cldev_send failed. %zd\n", ret);
+
+	return ret;
+}
+
+/**
+ * mei_gsc_proxy_recv - Receives a proxy message from ME FW.
+ * @dev: device corresponding to the mei_cl_device
+ * @buf: a message buffer to contain the received message
+ * @size: size of the buffer
+ * Return: bytes received on Success, <0 on Failure
+ */
+static int mei_gsc_proxy_recv(struct device *dev, void *buf, size_t size)
+{
+	ssize_t ret;
+
+	if (!dev || !buf)
+		return -EINVAL;
+
+	ret = mei_cldev_recv(to_mei_cl_device(dev), buf, size);
+	if (ret < 0)
+		dev_dbg(dev, "mei_cldev_recv failed. %zd\n", ret);
+
+	return ret;
+}
+
+static const struct i915_gsc_proxy_component_ops mei_gsc_proxy_ops = {
+	.owner = THIS_MODULE,
+	.send = mei_gsc_proxy_send,
+	.recv = mei_gsc_proxy_recv,
+};
+
+static int mei_component_master_bind(struct device *dev)
+{
+	struct mei_cl_device *cldev = to_mei_cl_device(dev);
+	struct i915_gsc_proxy_component *comp_master = mei_cldev_get_drvdata(cldev);
+
+	comp_master->ops = &mei_gsc_proxy_ops;
+	comp_master->mei_dev = dev;
+	return component_bind_all(dev, comp_master);
+}
+
+static void mei_component_master_unbind(struct device *dev)
+{
+	struct mei_cl_device *cldev = to_mei_cl_device(dev);
+	struct i915_gsc_proxy_component *comp_master = mei_cldev_get_drvdata(cldev);
+
+	component_unbind_all(dev, comp_master);
+}
+
+static const struct component_master_ops mei_component_master_ops = {
+	.bind = mei_component_master_bind,
+	.unbind = mei_component_master_unbind,
+};
+
+/**
+ * mei_gsc_proxy_component_match - compare function for matching mei.
+ *
+ *    The function checks if the device is pci device and
+ *    Intel VGA adapter, the subcomponent is SW Proxy
+ *    and the parent of MEI PCI and the parent of VGA are the same PCH device.
+ *
+ * @dev: master device
+ * @subcomponent: subcomponent to match (I915_COMPONENT_SWPROXY)
+ * @data: compare data (mei pci parent)
+ *
+ * Return:
+ * * 1 - if components match
+ * * 0 - otherwise
+ */
+static int mei_gsc_proxy_component_match(struct device *dev, int subcomponent,
+					 void *data)
+{
+	struct pci_dev *pdev;
+
+	if (!dev_is_pci(dev))
+		return 0;
+
+	pdev = to_pci_dev(dev);
+
+	if (pdev->class != (PCI_CLASS_DISPLAY_VGA << 8) ||
+	    pdev->vendor != PCI_VENDOR_ID_INTEL)
+		return 0;
+
+	if (subcomponent != I915_COMPONENT_GSC_PROXY)
+		return 0;
+
+	return component_compare_dev(dev->parent, ((struct device *)data)->parent);
+}
+
+static int mei_gsc_proxy_probe(struct mei_cl_device *cldev,
+			       const struct mei_cl_device_id *id)
+{
+	struct i915_gsc_proxy_component *comp_master;
+	struct component_match *master_match = NULL;
+	int ret;
+
+	ret = mei_cldev_enable(cldev);
+	if (ret < 0) {
+		dev_err(&cldev->dev, "mei_cldev_enable Failed. %d\n", ret);
+		goto enable_err_exit;
+	}
+
+	comp_master = kzalloc(sizeof(*comp_master), GFP_KERNEL);
+	if (!comp_master) {
+		ret = -ENOMEM;
+		goto err_exit;
+	}
+
+	component_match_add_typed(&cldev->dev, &master_match,
+				  mei_gsc_proxy_component_match, cldev->dev.parent);
+	if (IS_ERR_OR_NULL(master_match)) {
+		ret = -ENOMEM;
+		goto err_exit;
+	}
+
+	mei_cldev_set_drvdata(cldev, comp_master);
+	ret = component_master_add_with_match(&cldev->dev,
+					      &mei_component_master_ops,
+					      master_match);
+	if (ret < 0) {
+		dev_err(&cldev->dev, "Master comp add failed %d\n", ret);
+		goto err_exit;
+	}
+
+	return 0;
+
+err_exit:
+	mei_cldev_set_drvdata(cldev, NULL);
+	kfree(comp_master);
+	mei_cldev_disable(cldev);
+enable_err_exit:
+	return ret;
+}
+
+static void mei_gsc_proxy_remove(struct mei_cl_device *cldev)
+{
+	struct i915_gsc_proxy_component *comp_master = mei_cldev_get_drvdata(cldev);
+	int ret;
+
+	component_master_del(&cldev->dev, &mei_component_master_ops);
+	kfree(comp_master);
+	mei_cldev_set_drvdata(cldev, NULL);
+
+	ret = mei_cldev_disable(cldev);
+	if (ret)
+		dev_warn(&cldev->dev, "mei_cldev_disable() failed %d\n", ret);
+}
+
+#define MEI_UUID_GSC_PROXY UUID_LE(0xf73db04, 0x97ab, 0x4125, \
+				   0xb8, 0x93, 0xe9, 0x4, 0xad, 0xd, 0x54, 0x64)
+
+static struct mei_cl_device_id mei_gsc_proxy_tbl[] = {
+	{ .uuid = MEI_UUID_GSC_PROXY, .version = MEI_CL_VERSION_ANY },
+	{ }
+};
+MODULE_DEVICE_TABLE(mei, mei_gsc_proxy_tbl);
+
+static struct mei_cl_driver mei_gsc_proxy_driver = {
+	.id_table = mei_gsc_proxy_tbl,
+	.name = KBUILD_MODNAME,
+	.probe = mei_gsc_proxy_probe,
+	.remove	= mei_gsc_proxy_remove,
+};
+
+module_mei_cl_driver(mei_gsc_proxy_driver);
+
+MODULE_AUTHOR("Intel Corporation");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("MEI GSC PROXY");
-- 
GitLab


From 8a9bf29546a13efd3ed7784b890c2534e995348f Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Tue, 2 May 2023 09:38:53 -0700
Subject: [PATCH 0385/1791] drm/i915/gsc: add initial support for GSC proxy

The GSC uC needs to communicate with the CSME to perform certain
operations. Since the GSC can't perform this communication directly
on platforms where it is integrated in GT, i915 needs to transfer the
messages from GSC to CSME and back.
The proxy flow is as follow:
1 - i915 submits a request to GSC asking for the message to CSME
2 - GSC replies with the proxy header + payload for CSME
3 - i915 sends the reply from GSC as-is to CSME via the mei proxy
    component
4 - CSME replies with the proxy header + payload for GSC
5 - i915 submits a request to GSC with the reply from CSME
6 - GSC replies either with a new header + payload (same as step 2,
    so we restart from there) or with an end message.

After GSC load, i915 is expected to start the first proxy message chain,
while all subsequent ones will be triggered by the GSC via interrupt.

To communicate with the CSME, we use a dedicated mei component, which
means that we need to wait for it to bind before we can initialize the
proxies. This usually happens quite fast, but given that there is a
chance that we'll have to wait a few seconds the GSC work has been moved
to a dedicated WQ to not stall other processes.

v2: fix code style, includes and variable naming (Alan)
v3: add extra check for proxy status, fix includes and comments

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502163854.317653-4-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |   1 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c     |  10 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h     |   1 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c  | 384 ++++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h  |  17 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c     |  49 ++-
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h     |  14 +-
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h |   1 +
 8 files changed, 472 insertions(+), 5 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
 create mode 100644 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 97b0d4ae221ac..1f9071ee6c83f 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -194,6 +194,7 @@ i915-y += \
 # general-purpose microcontroller (GuC) support
 i915-y += \
 	  gt/uc/intel_gsc_fw.o \
+	  gt/uc/intel_gsc_proxy.o \
 	  gt/uc/intel_gsc_uc.o \
 	  gt/uc/intel_gsc_uc_heci_cmd_submit.o\
 	  gt/uc/intel_guc.o \
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
index 236673c02f9ad..f46eb17a7a987 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
@@ -13,6 +13,7 @@
 #define GSC_FW_STATUS_REG			_MMIO(0x116C40)
 #define GSC_FW_CURRENT_STATE			REG_GENMASK(3, 0)
 #define   GSC_FW_CURRENT_STATE_RESET		0
+#define   GSC_FW_PROXY_STATE_NORMAL		5
 #define GSC_FW_INIT_COMPLETE_BIT		REG_BIT(9)
 
 static bool gsc_is_in_reset(struct intel_uncore *uncore)
@@ -23,6 +24,15 @@ static bool gsc_is_in_reset(struct intel_uncore *uncore)
 	       GSC_FW_CURRENT_STATE_RESET;
 }
 
+bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc)
+{
+	struct intel_uncore *uncore = gsc_uc_to_gt(gsc)->uncore;
+	u32 fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG);
+
+	return REG_FIELD_GET(GSC_FW_CURRENT_STATE, fw_status) ==
+	       GSC_FW_PROXY_STATE_NORMAL;
+}
+
 bool intel_gsc_uc_fw_init_done(struct intel_gsc_uc *gsc)
 {
 	struct intel_uncore *uncore = gsc_uc_to_gt(gsc)->uncore;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h
index f4c1106bb2a93..fff8928218df8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h
@@ -13,5 +13,6 @@ struct intel_uncore;
 
 int intel_gsc_uc_fw_upload(struct intel_gsc_uc *gsc);
 bool intel_gsc_uc_fw_init_done(struct intel_gsc_uc *gsc);
+bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc);
 
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
new file mode 100644
index 0000000000000..65c9575863ea6
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
@@ -0,0 +1,384 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <linux/component.h>
+
+#include "drm/i915_component.h"
+#include "drm/i915_gsc_proxy_mei_interface.h"
+
+#include "gt/intel_gt.h"
+#include "gt/intel_gt_print.h"
+#include "intel_gsc_proxy.h"
+#include "intel_gsc_uc.h"
+#include "intel_gsc_uc_heci_cmd_submit.h"
+#include "i915_drv.h"
+
+/*
+ * GSC proxy:
+ * The GSC uC needs to communicate with the CSME to perform certain operations.
+ * Since the GSC can't perform this communication directly on platforms where it
+ * is integrated in GT, i915 needs to transfer the messages from GSC to CSME
+ * and back. i915 must manually start the proxy flow after the GSC is loaded to
+ * signal to GSC that we're ready to handle its messages and allow it to query
+ * its init data from CSME; GSC will then trigger an HECI2 interrupt if it needs
+ * to send messages to CSME again.
+ * The proxy flow is as follow:
+ * 1 - i915 submits a request to GSC asking for the message to CSME
+ * 2 - GSC replies with the proxy header + payload for CSME
+ * 3 - i915 sends the reply from GSC as-is to CSME via the mei proxy component
+ * 4 - CSME replies with the proxy header + payload for GSC
+ * 5 - i915 submits a request to GSC with the reply from CSME
+ * 6 - GSC replies either with a new header + payload (same as step 2, so we
+ *     restart from there) or with an end message.
+ */
+
+/*
+ * The component should load quite quickly in most cases, but it could take
+ * a bit. Using a very big timeout just to cover the worst case scenario
+ */
+#define GSC_PROXY_INIT_TIMEOUT_MS 20000
+
+/* the protocol supports up to 32K in each direction */
+#define GSC_PROXY_BUFFER_SIZE SZ_32K
+#define GSC_PROXY_CHANNEL_SIZE (GSC_PROXY_BUFFER_SIZE * 2)
+#define GSC_PROXY_MAX_MSG_SIZE (GSC_PROXY_BUFFER_SIZE - sizeof(struct intel_gsc_mtl_header))
+
+/* FW-defined proxy header */
+struct intel_gsc_proxy_header {
+	/*
+	 * hdr:
+	 * Bits 0-7: type of the proxy message (see enum intel_gsc_proxy_type)
+	 * Bits 8-15: rsvd
+	 * Bits 16-31: length in bytes of the payload following the proxy header
+	 */
+	u32 hdr;
+#define GSC_PROXY_TYPE		 GENMASK(7, 0)
+#define GSC_PROXY_PAYLOAD_LENGTH GENMASK(31, 16)
+
+	u32 source;		/* Source of the Proxy message */
+	u32 destination;	/* Destination of the Proxy message */
+#define GSC_PROXY_ADDRESSING_KMD  0x10000
+#define GSC_PROXY_ADDRESSING_GSC  0x20000
+#define GSC_PROXY_ADDRESSING_CSME 0x30000
+
+	u32 status;		/* Command status */
+} __packed;
+
+/* FW-defined proxy types */
+enum intel_gsc_proxy_type {
+	GSC_PROXY_MSG_TYPE_PROXY_INVALID = 0,
+	GSC_PROXY_MSG_TYPE_PROXY_QUERY = 1,
+	GSC_PROXY_MSG_TYPE_PROXY_PAYLOAD = 2,
+	GSC_PROXY_MSG_TYPE_PROXY_END = 3,
+	GSC_PROXY_MSG_TYPE_PROXY_NOTIFICATION = 4,
+};
+
+struct gsc_proxy_msg {
+	struct intel_gsc_mtl_header header;
+	struct intel_gsc_proxy_header proxy_header;
+} __packed;
+
+static int proxy_send_to_csme(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	struct i915_gsc_proxy_component *comp = gsc->proxy.component;
+	struct intel_gsc_mtl_header *hdr;
+	void *in = gsc->proxy.to_csme;
+	void *out = gsc->proxy.to_gsc;
+	u32 in_size;
+	int ret;
+
+	/* CSME msg only includes the proxy */
+	hdr = in;
+	in += sizeof(struct intel_gsc_mtl_header);
+	out += sizeof(struct intel_gsc_mtl_header);
+
+	in_size = hdr->message_size - sizeof(struct intel_gsc_mtl_header);
+
+	/* the message must contain at least the proxy header */
+	if (in_size < sizeof(struct intel_gsc_proxy_header) ||
+	    in_size > GSC_PROXY_MAX_MSG_SIZE) {
+		gt_err(gt, "Invalid CSME message size: %u\n", in_size);
+		return -EINVAL;
+	}
+
+	ret = comp->ops->send(comp->mei_dev, in, in_size);
+	if (ret < 0) {
+		gt_err(gt, "Failed to send CSME message\n");
+		return ret;
+	}
+
+	ret = comp->ops->recv(comp->mei_dev, out, GSC_PROXY_MAX_MSG_SIZE);
+	if (ret < 0) {
+		gt_err(gt, "Failed to receive CSME message\n");
+		return ret;
+	}
+
+	return ret;
+}
+
+static int proxy_send_to_gsc(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	u32 *marker = gsc->proxy.to_csme; /* first dw of the reply header */
+	u64 addr_in = i915_ggtt_offset(gsc->proxy.vma);
+	u64 addr_out = addr_in + GSC_PROXY_BUFFER_SIZE;
+	u32 size = ((struct gsc_proxy_msg *)gsc->proxy.to_gsc)->header.message_size;
+	int err;
+
+	/* the message must contain at least the gsc and proxy headers */
+	if (size < sizeof(struct gsc_proxy_msg) || size > GSC_PROXY_BUFFER_SIZE) {
+		gt_err(gt, "Invalid GSC proxy message size: %u\n", size);
+		return -EINVAL;
+	}
+
+	/* clear the message marker */
+	*marker = 0;
+
+	/* make sure the marker write is flushed */
+	wmb();
+
+	/* send the request */
+	err = intel_gsc_uc_heci_cmd_submit_packet(gsc, addr_in, size,
+						  addr_out, GSC_PROXY_BUFFER_SIZE);
+
+	if (!err) {
+		/* wait for the reply to show up */
+		err = wait_for(*marker != 0, 300);
+		if (err)
+			gt_err(gt, "Failed to get a proxy reply from gsc\n");
+	}
+
+	return err;
+}
+
+static int validate_proxy_header(struct intel_gsc_proxy_header *header,
+				 u32 source, u32 dest)
+{
+	u32 type = FIELD_GET(GSC_PROXY_TYPE, header->hdr);
+	u32 length = FIELD_GET(GSC_PROXY_PAYLOAD_LENGTH, header->hdr);
+	int ret = 0;
+
+	if (header->destination != dest || header->source != source) {
+		ret = -ENOEXEC;
+		goto fail;
+	}
+
+	switch (type) {
+	case GSC_PROXY_MSG_TYPE_PROXY_PAYLOAD:
+		if (length > 0)
+			break;
+		fallthrough;
+	case GSC_PROXY_MSG_TYPE_PROXY_INVALID:
+		ret = -EIO;
+		goto fail;
+	default:
+		break;
+	}
+
+fail:
+	return ret;
+}
+
+static int proxy_query(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	struct gsc_proxy_msg *to_gsc = gsc->proxy.to_gsc;
+	struct gsc_proxy_msg *to_csme = gsc->proxy.to_csme;
+	int ret;
+
+	intel_gsc_uc_heci_cmd_emit_mtl_header(&to_gsc->header,
+					      HECI_MEADDRESS_PROXY,
+					      sizeof(struct gsc_proxy_msg),
+					      0);
+
+	to_gsc->proxy_header.hdr =
+		FIELD_PREP(GSC_PROXY_TYPE, GSC_PROXY_MSG_TYPE_PROXY_QUERY) |
+		FIELD_PREP(GSC_PROXY_PAYLOAD_LENGTH, 0);
+
+	to_gsc->proxy_header.source = GSC_PROXY_ADDRESSING_KMD;
+	to_gsc->proxy_header.destination = GSC_PROXY_ADDRESSING_GSC;
+	to_gsc->proxy_header.status = 0;
+
+	while (1) {
+		/* clear the GSC response header space */
+		memset(gsc->proxy.to_csme, 0, sizeof(struct gsc_proxy_msg));
+
+		/* send proxy message to GSC */
+		ret = proxy_send_to_gsc(gsc);
+		if (ret) {
+			gt_err(gt, "failed to send proxy message to GSC! %d\n", ret);
+			goto proxy_error;
+		}
+
+		/* stop if this was the last message */
+		if (FIELD_GET(GSC_PROXY_TYPE, to_csme->proxy_header.hdr) ==
+				GSC_PROXY_MSG_TYPE_PROXY_END)
+			break;
+
+		/* make sure the GSC-to-CSME proxy header is sane */
+		ret = validate_proxy_header(&to_csme->proxy_header,
+					    GSC_PROXY_ADDRESSING_GSC,
+					    GSC_PROXY_ADDRESSING_CSME);
+		if (ret) {
+			gt_err(gt, "invalid GSC to CSME proxy header! %d\n", ret);
+			goto proxy_error;
+		}
+
+		/* send the GSC message to the CSME */
+		ret = proxy_send_to_csme(gsc);
+		if (ret < 0) {
+			gt_err(gt, "failed to send proxy message to CSME! %d\n", ret);
+			goto proxy_error;
+		}
+
+		/* update the GSC message size with the returned value from CSME */
+		to_gsc->header.message_size = ret + sizeof(struct intel_gsc_mtl_header);
+
+		/* make sure the CSME-to-GSC proxy header is sane */
+		ret = validate_proxy_header(&to_gsc->proxy_header,
+					    GSC_PROXY_ADDRESSING_CSME,
+					    GSC_PROXY_ADDRESSING_GSC);
+		if (ret) {
+			gt_err(gt, "invalid CSME to GSC proxy header! %d\n", ret);
+			goto proxy_error;
+		}
+	}
+
+proxy_error:
+	return ret < 0 ? ret : 0;
+}
+
+int intel_gsc_proxy_request_handler(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	int err;
+
+	if (!gsc->proxy.component_added)
+		return -ENODEV;
+
+	assert_rpm_wakelock_held(gt->uncore->rpm);
+
+	/* when GSC is loaded, we can queue this before the component is bound */
+	err = wait_for(gsc->proxy.component, GSC_PROXY_INIT_TIMEOUT_MS);
+	if (err) {
+		gt_err(gt, "GSC proxy component didn't bind within the expected timeout\n");
+		return -EIO;
+	}
+
+	mutex_lock(&gsc->proxy.mutex);
+	if (!gsc->proxy.component) {
+		gt_err(gt, "GSC proxy worker called without the component being bound!\n");
+		err = -EIO;
+	} else {
+		err = proxy_query(gsc);
+	}
+	mutex_unlock(&gsc->proxy.mutex);
+	return err;
+}
+
+static int i915_gsc_proxy_component_bind(struct device *i915_kdev,
+					 struct device *mei_kdev, void *data)
+{
+	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
+	struct intel_gsc_uc *gsc = &i915->media_gt->uc.gsc;
+
+	mutex_lock(&gsc->proxy.mutex);
+	gsc->proxy.component = data;
+	gsc->proxy.component->mei_dev = mei_kdev;
+	mutex_unlock(&gsc->proxy.mutex);
+
+	return 0;
+}
+
+static void i915_gsc_proxy_component_unbind(struct device *i915_kdev,
+					    struct device *mei_kdev, void *data)
+{
+	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
+	struct intel_gsc_uc *gsc = &i915->media_gt->uc.gsc;
+
+	mutex_lock(&gsc->proxy.mutex);
+	gsc->proxy.component = NULL;
+	mutex_unlock(&gsc->proxy.mutex);
+}
+
+static const struct component_ops i915_gsc_proxy_component_ops = {
+	.bind   = i915_gsc_proxy_component_bind,
+	.unbind = i915_gsc_proxy_component_unbind,
+};
+
+static int proxy_channel_alloc(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	struct i915_vma *vma;
+	void *vaddr;
+	int err;
+
+	err = intel_guc_allocate_and_map_vma(&gt->uc.guc, GSC_PROXY_CHANNEL_SIZE,
+					     &vma, &vaddr);
+	if (err)
+		return err;
+
+	gsc->proxy.vma = vma;
+	gsc->proxy.to_gsc = vaddr;
+	gsc->proxy.to_csme = vaddr + GSC_PROXY_BUFFER_SIZE;
+
+	return 0;
+}
+
+static void proxy_channel_free(struct intel_gsc_uc *gsc)
+{
+	if (!gsc->proxy.vma)
+		return;
+
+	gsc->proxy.to_gsc = NULL;
+	gsc->proxy.to_csme = NULL;
+	i915_vma_unpin_and_release(&gsc->proxy.vma, I915_VMA_RELEASE_MAP);
+}
+
+void intel_gsc_proxy_fini(struct intel_gsc_uc *gsc)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	struct drm_i915_private *i915 = gt->i915;
+
+	if (fetch_and_zero(&gsc->proxy.component_added))
+		component_del(i915->drm.dev, &i915_gsc_proxy_component_ops);
+
+	proxy_channel_free(gsc);
+}
+
+int intel_gsc_proxy_init(struct intel_gsc_uc *gsc)
+{
+	int err;
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+	struct drm_i915_private *i915 = gt->i915;
+
+	mutex_init(&gsc->proxy.mutex);
+
+	if (!IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY)) {
+		gt_info(gt, "can't init GSC proxy due to missing mei component\n");
+		return -ENODEV;
+	}
+
+	err = proxy_channel_alloc(gsc);
+	if (err)
+		return err;
+
+	err = component_add_typed(i915->drm.dev, &i915_gsc_proxy_component_ops,
+				  I915_COMPONENT_GSC_PROXY);
+	if (err < 0) {
+		gt_err(gt, "Failed to add GSC_PROXY component (%d)\n", err);
+		goto out_free;
+	}
+
+	gsc->proxy.component_added = true;
+
+	return 0;
+
+out_free:
+	proxy_channel_free(gsc);
+	return err;
+}
+
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h
new file mode 100644
index 0000000000000..1f27cb6e62ba9
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef _INTEL_GSC_PROXY_H_
+#define _INTEL_GSC_PROXY_H_
+
+#include <linux/types.h>
+
+struct intel_gsc_uc;
+
+int intel_gsc_proxy_init(struct intel_gsc_uc *gsc);
+void intel_gsc_proxy_fini(struct intel_gsc_uc *gsc);
+int intel_gsc_proxy_request_handler(struct intel_gsc_uc *gsc);
+
+#endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
index 2d5b70b3384cb..b234bb585ad3a 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
@@ -10,15 +10,39 @@
 #include "intel_gsc_uc.h"
 #include "intel_gsc_fw.h"
 #include "i915_drv.h"
+#include "intel_gsc_proxy.h"
 
 static void gsc_work(struct work_struct *work)
 {
 	struct intel_gsc_uc *gsc = container_of(work, typeof(*gsc), work);
 	struct intel_gt *gt = gsc_uc_to_gt(gsc);
 	intel_wakeref_t wakeref;
+	int ret;
 
-	with_intel_runtime_pm(gt->uncore->rpm, wakeref)
-		intel_gsc_uc_fw_upload(gsc);
+	wakeref = intel_runtime_pm_get(gt->uncore->rpm);
+
+	ret = intel_gsc_uc_fw_upload(gsc);
+	if (ret)
+		goto out_put;
+
+	ret = intel_gsc_proxy_request_handler(gsc);
+	if (ret)
+		goto out_put;
+
+	/*
+	 * If there is a proxy establishment error, the GSC might still
+	 * complete the request handling cleanly, so we need to check the
+	 * status register to check if the proxy init was actually successful
+	 */
+	if (intel_gsc_uc_fw_proxy_init_done(gsc)) {
+		drm_dbg(&gt->i915->drm, "GSC Proxy initialized\n");
+		intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_RUNNING);
+	} else {
+		drm_err(&gt->i915->drm, "GSC status reports proxy init not complete\n");
+	}
+
+out_put:
+	intel_runtime_pm_put(gt->uncore->rpm, wakeref);
 }
 
 static bool gsc_engine_supported(struct intel_gt *gt)
@@ -43,6 +67,8 @@ static bool gsc_engine_supported(struct intel_gt *gt)
 
 void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc)
 {
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+
 	intel_uc_fw_init_early(&gsc->fw, INTEL_UC_FW_TYPE_GSC);
 	INIT_WORK(&gsc->work, gsc_work);
 
@@ -50,10 +76,16 @@ void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc)
 	 * GT with it being not fully setup hence check device info's
 	 * engine mask
 	 */
-	if (!gsc_engine_supported(gsc_uc_to_gt(gsc))) {
+	if (!gsc_engine_supported(gt)) {
 		intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_NOT_SUPPORTED);
 		return;
 	}
+
+	gsc->wq = alloc_ordered_workqueue("i915_gsc", 0);
+	if (!gsc->wq) {
+		gt_err(gt, "failed to allocate WQ for GSC, disabling FW\n");
+		intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_NOT_SUPPORTED);
+	}
 }
 
 int intel_gsc_uc_init(struct intel_gsc_uc *gsc)
@@ -88,6 +120,9 @@ int intel_gsc_uc_init(struct intel_gsc_uc *gsc)
 
 	gsc->ce = ce;
 
+	/* if we fail to init proxy we still want to load GSC for PM */
+	intel_gsc_proxy_init(gsc);
+
 	intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_LOADABLE);
 
 	return 0;
@@ -107,6 +142,12 @@ void intel_gsc_uc_fini(struct intel_gsc_uc *gsc)
 		return;
 
 	flush_work(&gsc->work);
+	if (gsc->wq) {
+		destroy_workqueue(gsc->wq);
+		gsc->wq = NULL;
+	}
+
+	intel_gsc_proxy_fini(gsc);
 
 	if (gsc->ce)
 		intel_engine_destroy_pinned_context(fetch_and_zero(&gsc->ce));
@@ -151,5 +192,5 @@ void intel_gsc_uc_load_start(struct intel_gsc_uc *gsc)
 	if (intel_gsc_uc_fw_init_done(gsc))
 		return;
 
-	queue_work(system_unbound_wq, &gsc->work);
+	queue_work(gsc->wq, &gsc->work);
 }
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
index 5f50fa1ff8b93..023bded10ddef 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
@@ -10,6 +10,7 @@
 
 struct i915_vma;
 struct intel_context;
+struct i915_gsc_proxy_component;
 
 struct intel_gsc_uc {
 	/* Generic uC firmware management */
@@ -19,7 +20,18 @@ struct intel_gsc_uc {
 	struct i915_vma *local; /* private memory for GSC usage */
 	struct intel_context *ce; /* for submission to GSC FW via GSC engine */
 
-	struct work_struct work; /* for delayed load */
+	/* for delayed load and proxy handling */
+	struct workqueue_struct *wq;
+	struct work_struct work;
+
+	struct {
+		struct i915_gsc_proxy_component *component;
+		bool component_added;
+		struct i915_vma *vma;
+		void *to_gsc;
+		void *to_csme;
+		struct mutex mutex; /* protects the tee channel binding */
+	} proxy;
 };
 
 void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
index 3d56ae501991d..8f199d5f963e5 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
@@ -14,6 +14,7 @@ struct intel_gsc_mtl_header {
 #define GSC_HECI_VALIDITY_MARKER 0xA578875A
 
 	u8 heci_client_id;
+#define HECI_MEADDRESS_PROXY 10
 #define HECI_MEADDRESS_PXP 17
 #define HECI_MEADDRESS_HDCP 18
 
-- 
GitLab


From 31cc65b4c7e37eaf645be2d5fdcade56e3fdf8f2 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Tue, 2 May 2023 09:38:54 -0700
Subject: [PATCH 0386/1791] drm/i915/gsc: add support for GSC proxy interrupt

The GSC notifies us of a proxy request via the HECI2 interrupt. The
interrupt must be enabled both in the HECI layer and in our usual gt irq
programming; for the latter, the interrupt is enabled via the same enable
register as the GSC CS, but it does have its own mask register. When the
interrupt is received, we also need to de-assert it in both layers.

The handling of the proxy request is deferred to the same worker that we
use for GSC load. New flags have been added to distinguish between the
init case and the proxy interrupt.

v2: Make sure not to set the reset bit when enabling/disabling the GSC
interrupts, fix defines (Alan)

v3: rebase on proxy status register check

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502163854.317653-5-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_irq.c       | 22 ++++++-
 drivers/gpu/drm/i915/gt/intel_gt_regs.h      |  3 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c | 44 +++++++++++++-
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h |  1 +
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c    | 61 ++++++++++++++------
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h    |  3 +
 6 files changed, 112 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index 1b25a60391522..e13cfbe2daf55 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -15,6 +15,7 @@
 #include "intel_uncore.h"
 #include "intel_rps.h"
 #include "pxp/intel_pxp_irq.h"
+#include "uc/intel_gsc_proxy.h"
 
 static void guc_irq_handler(struct intel_guc *guc, u16 iir)
 {
@@ -81,6 +82,9 @@ gen11_other_irq_handler(struct intel_gt *gt, const u8 instance,
 	if (instance == OTHER_GSC_INSTANCE)
 		return intel_gsc_irq_handler(gt, iir);
 
+	if (instance == OTHER_GSC_HECI_2_INSTANCE)
+		return intel_gsc_proxy_irq_handler(&gt->uc.gsc, iir);
+
 	WARN_ONCE(1, "unhandled other interrupt instance=0x%x, iir=0x%x\n",
 		  instance, iir);
 }
@@ -100,6 +104,8 @@ static struct intel_gt *pick_gt(struct intel_gt *gt, u8 class, u8 instance)
 	case VIDEO_ENHANCEMENT_CLASS:
 		return media_gt;
 	case OTHER_CLASS:
+		if (instance == OTHER_GSC_HECI_2_INSTANCE)
+			return media_gt;
 		if (instance == OTHER_GSC_INSTANCE && HAS_ENGINE(media_gt, GSC0))
 			return media_gt;
 		fallthrough;
@@ -256,6 +262,7 @@ void gen11_gt_irq_postinstall(struct intel_gt *gt)
 	u32 irqs = GT_RENDER_USER_INTERRUPT;
 	u32 guc_mask = intel_uc_wants_guc(&gt->uc) ? GUC_INTR_GUC2HOST : 0;
 	u32 gsc_mask = 0;
+	u32 heci_mask = 0;
 	u32 dmask;
 	u32 smask;
 
@@ -267,10 +274,16 @@ void gen11_gt_irq_postinstall(struct intel_gt *gt)
 	dmask = irqs << 16 | irqs;
 	smask = irqs << 16;
 
-	if (HAS_ENGINE(gt, GSC0))
+	if (HAS_ENGINE(gt, GSC0)) {
+		/*
+		 * the heci2 interrupt is enabled via the same register as the
+		 * GSC interrupt, but it has its own mask register.
+		 */
 		gsc_mask = irqs;
-	else if (HAS_HECI_GSC(gt->i915))
+		heci_mask = GSC_IRQ_INTF(1); /* HECI2 IRQ for SW Proxy*/
+	} else if (HAS_HECI_GSC(gt->i915)) {
 		gsc_mask = GSC_IRQ_INTF(0) | GSC_IRQ_INTF(1);
+	}
 
 	BUILD_BUG_ON(irqs & 0xffff0000);
 
@@ -280,7 +293,7 @@ void gen11_gt_irq_postinstall(struct intel_gt *gt)
 	if (CCS_MASK(gt))
 		intel_uncore_write(uncore, GEN12_CCS_RSVD_INTR_ENABLE, smask);
 	if (gsc_mask)
-		intel_uncore_write(uncore, GEN11_GUNIT_CSME_INTR_ENABLE, gsc_mask);
+		intel_uncore_write(uncore, GEN11_GUNIT_CSME_INTR_ENABLE, gsc_mask | heci_mask);
 
 	/* Unmask irqs on RCS, BCS, VCS and VECS engines. */
 	intel_uncore_write(uncore, GEN11_RCS0_RSVD_INTR_MASK, ~smask);
@@ -308,6 +321,9 @@ void gen11_gt_irq_postinstall(struct intel_gt *gt)
 		intel_uncore_write(uncore, GEN12_CCS2_CCS3_INTR_MASK, ~dmask);
 	if (gsc_mask)
 		intel_uncore_write(uncore, GEN11_GUNIT_CSME_INTR_MASK, ~gsc_mask);
+	if (heci_mask)
+		intel_uncore_write(uncore, GEN12_HECI2_RSVD_INTR_MASK,
+				   ~REG_FIELD_PREP(ENGINE1_MASK, heci_mask));
 
 	if (guc_mask) {
 		/* the enable bit is common for both GTs but the masks are separate */
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index af80d2fe739b0..b8a39c219b60e 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -1596,6 +1596,7 @@
 
 #define GEN11_GT_INTR_DW(x)			_MMIO(0x190018 + ((x) * 4))
 #define   GEN11_CSME				(31)
+#define   GEN12_HECI_2				(30)
 #define   GEN11_GUNIT				(28)
 #define   GEN11_GUC				(25)
 #define   MTL_MGUC				(24)
@@ -1637,6 +1638,7 @@
 /* irq instances for OTHER_CLASS */
 #define   OTHER_GUC_INSTANCE			0
 #define   OTHER_GTPM_INSTANCE			1
+#define   OTHER_GSC_HECI_2_INSTANCE		3
 #define   OTHER_KCR_INSTANCE			4
 #define   OTHER_GSC_INSTANCE			6
 #define   OTHER_MEDIA_GUC_INSTANCE		16
@@ -1652,6 +1654,7 @@
 #define GEN12_VCS6_VCS7_INTR_MASK		_MMIO(0x1900b4)
 #define GEN11_VECS0_VECS1_INTR_MASK		_MMIO(0x1900d0)
 #define GEN12_VECS2_VECS3_INTR_MASK		_MMIO(0x1900d4)
+#define GEN12_HECI2_RSVD_INTR_MASK		_MMIO(0x1900e4)
 #define GEN11_GUC_SG_INTR_MASK			_MMIO(0x1900e8)
 #define MTL_GUC_MGUC_INTR_MASK			_MMIO(0x1900e8) /* MTL+ */
 #define GEN11_GPM_WGBOXPERF_INTR_MASK		_MMIO(0x1900ec)
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
index 65c9575863ea6..ebee0b5a2c1dd 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
@@ -14,6 +14,7 @@
 #include "intel_gsc_uc.h"
 #include "intel_gsc_uc_heci_cmd_submit.h"
 #include "i915_drv.h"
+#include "i915_reg.h"
 
 /*
  * GSC proxy:
@@ -273,17 +274,49 @@ int intel_gsc_proxy_request_handler(struct intel_gsc_uc *gsc)
 		gt_err(gt, "GSC proxy worker called without the component being bound!\n");
 		err = -EIO;
 	} else {
+		/*
+		 * write the status bit to clear it and allow new proxy
+		 * interrupts to be generated while we handle the current
+		 * request, but be sure not to write the reset bit
+		 */
+		intel_uncore_rmw(gt->uncore, HECI_H_CSR(MTL_GSC_HECI2_BASE),
+				 HECI_H_CSR_RST, HECI_H_CSR_IS);
 		err = proxy_query(gsc);
 	}
 	mutex_unlock(&gsc->proxy.mutex);
 	return err;
 }
 
+void intel_gsc_proxy_irq_handler(struct intel_gsc_uc *gsc, u32 iir)
+{
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+
+	if (unlikely(!iir))
+		return;
+
+	lockdep_assert_held(gt->irq_lock);
+
+	if (!gsc->proxy.component) {
+		gt_err(gt, "GSC proxy irq received without the component being bound!\n");
+		return;
+	}
+
+	gsc->gsc_work_actions |= GSC_ACTION_SW_PROXY;
+	queue_work(gsc->wq, &gsc->work);
+}
+
 static int i915_gsc_proxy_component_bind(struct device *i915_kdev,
 					 struct device *mei_kdev, void *data)
 {
 	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
-	struct intel_gsc_uc *gsc = &i915->media_gt->uc.gsc;
+	struct intel_gt *gt = i915->media_gt;
+	struct intel_gsc_uc *gsc = &gt->uc.gsc;
+	intel_wakeref_t wakeref;
+
+	/* enable HECI2 IRQs */
+	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+		intel_uncore_rmw(gt->uncore, HECI_H_CSR(MTL_GSC_HECI2_BASE),
+				 HECI_H_CSR_RST, HECI_H_CSR_IE);
 
 	mutex_lock(&gsc->proxy.mutex);
 	gsc->proxy.component = data;
@@ -297,11 +330,18 @@ static void i915_gsc_proxy_component_unbind(struct device *i915_kdev,
 					    struct device *mei_kdev, void *data)
 {
 	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
-	struct intel_gsc_uc *gsc = &i915->media_gt->uc.gsc;
+	struct intel_gt *gt = i915->media_gt;
+	struct intel_gsc_uc *gsc = &gt->uc.gsc;
+	intel_wakeref_t wakeref;
 
 	mutex_lock(&gsc->proxy.mutex);
 	gsc->proxy.component = NULL;
 	mutex_unlock(&gsc->proxy.mutex);
+
+	/* disable HECI2 IRQs */
+	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+		intel_uncore_rmw(gt->uncore, HECI_H_CSR(MTL_GSC_HECI2_BASE),
+				 HECI_H_CSR_IE | HECI_H_CSR_RST, 0);
 }
 
 static const struct component_ops i915_gsc_proxy_component_ops = {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h
index 1f27cb6e62ba9..fc5aef10bfb4e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.h
@@ -13,5 +13,6 @@ struct intel_gsc_uc;
 int intel_gsc_proxy_init(struct intel_gsc_uc *gsc);
 void intel_gsc_proxy_fini(struct intel_gsc_uc *gsc);
 int intel_gsc_proxy_request_handler(struct intel_gsc_uc *gsc);
+void intel_gsc_proxy_irq_handler(struct intel_gsc_uc *gsc, u32 iir);
 
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
index b234bb585ad3a..fb0984f875f9e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
@@ -17,28 +17,49 @@ static void gsc_work(struct work_struct *work)
 	struct intel_gsc_uc *gsc = container_of(work, typeof(*gsc), work);
 	struct intel_gt *gt = gsc_uc_to_gt(gsc);
 	intel_wakeref_t wakeref;
+	u32 actions;
 	int ret;
 
 	wakeref = intel_runtime_pm_get(gt->uncore->rpm);
 
-	ret = intel_gsc_uc_fw_upload(gsc);
-	if (ret)
-		goto out_put;
-
-	ret = intel_gsc_proxy_request_handler(gsc);
-	if (ret)
-		goto out_put;
+	spin_lock_irq(gt->irq_lock);
+	actions = gsc->gsc_work_actions;
+	gsc->gsc_work_actions = 0;
+	spin_unlock_irq(gt->irq_lock);
+
+	if (actions & GSC_ACTION_FW_LOAD) {
+		ret = intel_gsc_uc_fw_upload(gsc);
+		if (ret == -EEXIST) /* skip proxy if not a new load */
+			actions &= ~GSC_ACTION_FW_LOAD;
+		else if (ret)
+			goto out_put;
+	}
 
-	/*
-	 * If there is a proxy establishment error, the GSC might still
-	 * complete the request handling cleanly, so we need to check the
-	 * status register to check if the proxy init was actually successful
-	 */
-	if (intel_gsc_uc_fw_proxy_init_done(gsc)) {
-		drm_dbg(&gt->i915->drm, "GSC Proxy initialized\n");
-		intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_RUNNING);
-	} else {
-		drm_err(&gt->i915->drm, "GSC status reports proxy init not complete\n");
+	if (actions & (GSC_ACTION_FW_LOAD | GSC_ACTION_SW_PROXY)) {
+		if (!intel_gsc_uc_fw_init_done(gsc)) {
+			gt_err(gt, "Proxy request received with GSC not loaded!\n");
+			goto out_put;
+		}
+
+		ret = intel_gsc_proxy_request_handler(gsc);
+		if (ret)
+			goto out_put;
+
+		/* mark the GSC FW init as done the first time we run this */
+		if (actions & GSC_ACTION_FW_LOAD) {
+			/*
+			 * If there is a proxy establishment error, the GSC might still
+			 * complete the request handling cleanly, so we need to check the
+			 * status register to check if the proxy init was actually successful
+			 */
+			if (intel_gsc_uc_fw_proxy_init_done(gsc)) {
+				drm_dbg(&gt->i915->drm, "GSC Proxy initialized\n");
+				intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_RUNNING);
+			} else {
+				drm_err(&gt->i915->drm,
+					"GSC status reports proxy init not complete\n");
+			}
+		}
 	}
 
 out_put:
@@ -186,11 +207,17 @@ void intel_gsc_uc_resume(struct intel_gsc_uc *gsc)
 
 void intel_gsc_uc_load_start(struct intel_gsc_uc *gsc)
 {
+	struct intel_gt *gt = gsc_uc_to_gt(gsc);
+
 	if (!intel_uc_fw_is_loadable(&gsc->fw))
 		return;
 
 	if (intel_gsc_uc_fw_init_done(gsc))
 		return;
 
+	spin_lock_irq(gt->irq_lock);
+	gsc->gsc_work_actions |= GSC_ACTION_FW_LOAD;
+	spin_unlock_irq(gt->irq_lock);
+
 	queue_work(gsc->wq, &gsc->work);
 }
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
index 023bded10ddef..a2a0813b8a76d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h
@@ -23,6 +23,9 @@ struct intel_gsc_uc {
 	/* for delayed load and proxy handling */
 	struct workqueue_struct *wq;
 	struct work_struct work;
+	u32 gsc_work_actions; /* protected by gt->irq_lock */
+#define GSC_ACTION_FW_LOAD BIT(0)
+#define GSC_ACTION_SW_PROXY BIT(1)
 
 	struct {
 		struct i915_gsc_proxy_component *component;
-- 
GitLab


From 4883c804fc794dbd9f9668e73b3d266a52ae6c0c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:13 +0300
Subject: [PATCH 0387/1791] drm/i915: Define bitmasks for ilk pfit window
 pos/size
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Define and use the bitmasks for the x/y components
of the ilk+ panel filter window pos/size registers.

Note that we stick to the full 16 bit mask even though
the actual hardware limits are lower (and somewhat
platform dependent). BDW is actually limited to
13 bits horizontal and 12 bits vertical, with the high
bits being hardwired to zero. HSW should have the same
limits as BDW. And pre-HSW should be limited to 12bits
in both directions as that's already the limit of the
transcoder timing registers. Curiously on HSW and earlier
platforms all 16 bits can actually be set, but presumably
the hardware ignores the high bits.

v2: Switch back to full 16bit masks since that's what
    we use transcoder timign regs and PIPESRC as well

Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 12 ++++++++----
 drivers/gpu/drm/i915/i915_reg.h              |  8 ++++++++
 2 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 3d3483e6f8363..07443e283de94 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -812,8 +812,10 @@ static void ilk_pfit_enable(const struct intel_crtc_state *crtc_state)
 	else
 		intel_de_write_fw(dev_priv, PF_CTL(pipe), PF_ENABLE |
 				  PF_FILTER_MED_3x3);
-	intel_de_write_fw(dev_priv, PF_WIN_POS(pipe), x << 16 | y);
-	intel_de_write_fw(dev_priv, PF_WIN_SZ(pipe), width << 16 | height);
+	intel_de_write_fw(dev_priv, PF_WIN_POS(pipe),
+			  PF_WIN_XPOS(x) | PF_WIN_YPOS(y));
+	intel_de_write_fw(dev_priv, PF_WIN_SZ(pipe),
+			  PF_WIN_XSIZE(width) | PF_WIN_YSIZE(height));
 }
 
 static void intel_crtc_dpms_overlay_disable(struct intel_crtc *crtc)
@@ -3246,8 +3248,10 @@ static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state)
 	size = intel_de_read(dev_priv, PF_WIN_SZ(crtc->pipe));
 
 	drm_rect_init(&crtc_state->pch_pfit.dst,
-		      pos >> 16, pos & 0xffff,
-		      size >> 16, size & 0xffff);
+		      REG_FIELD_GET(PF_WIN_XPOS_MASK, pos),
+		      REG_FIELD_GET(PF_WIN_YPOS_MASK, pos),
+		      REG_FIELD_GET(PF_WIN_XSIZE_MASK, size),
+		      REG_FIELD_GET(PF_WIN_YSIZE_MASK, size));
 
 	/*
 	 * We currently do not free assignements of panel fitters on
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 831a89841934e..faf8ea6db609d 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4015,8 +4015,16 @@
 #define   PF_FILTER_EDGE_SOFTEN		REG_FIELD_PREP(PF_FILTER_EDGE_MASK, 3)
 #define _PFA_WIN_SZ		0x68074
 #define _PFB_WIN_SZ		0x68874
+#define   PF_WIN_XSIZE_MASK	REG_GENMASK(31, 16)
+#define   PF_WIN_XSIZE(w)	REG_FIELD_PREP(PF_WIN_XSIZE_MASK, (w))
+#define   PF_WIN_YSIZE_MASK	REG_GENMASK(15, 0)
+#define   PF_WIN_YSIZE(h)	REG_FIELD_PREP(PF_WIN_YSIZE_MASK, (h))
 #define _PFA_WIN_POS		0x68070
 #define _PFB_WIN_POS		0x68870
+#define   PF_WIN_XPOS_MASK	REG_GENMASK(31, 16)
+#define   PF_WIN_XPOS(x)	REG_FIELD_PREP(PF_WIN_XPOS_MASK, (x))
+#define   PF_WIN_YPOS_MASK	REG_GENMASK(15, 0)
+#define   PF_WIN_YPOS(y)	REG_FIELD_PREP(PF_WIN_YPOS_MASK, (y))
 #define _PFA_VSCALE		0x68084
 #define _PFB_VSCALE		0x68884
 #define _PFA_HSCALE		0x68090
-- 
GitLab


From 33d233f5cf9b9f1ce8f38480c45317dc0eecafa2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:14 +0300
Subject: [PATCH 0388/1791] drm/i915: Remove dead scaler register defines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We have some duplicated scaler register defines that are
never used. Remove them.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-3-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/i915_reg.h | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index faf8ea6db609d..c050b9b996213 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4036,18 +4036,6 @@
 #define PF_VSCALE(pipe)		_MMIO_PIPE(pipe, _PFA_VSCALE, _PFB_VSCALE)
 #define PF_HSCALE(pipe)		_MMIO_PIPE(pipe, _PFA_HSCALE, _PFB_HSCALE)
 
-#define _PSA_CTL		0x68180
-#define _PSB_CTL		0x68980
-#define PS_ENABLE		(1 << 31)
-#define _PSA_WIN_SZ		0x68174
-#define _PSB_WIN_SZ		0x68974
-#define _PSA_WIN_POS		0x68170
-#define _PSB_WIN_POS		0x68970
-
-#define PS_CTL(pipe)		_MMIO_PIPE(pipe, _PSA_CTL, _PSB_CTL)
-#define PS_WIN_SZ(pipe)		_MMIO_PIPE(pipe, _PSA_WIN_SZ, _PSB_WIN_SZ)
-#define PS_WIN_POS(pipe)	_MMIO_PIPE(pipe, _PSA_WIN_POS, _PSB_WIN_POS)
-
 /*
  * Skylake scalers
  */
-- 
GitLab


From ebb4e2b03acc3a9b58a8bc32b239e6af2166e84f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:15 +0300
Subject: [PATCH 0389/1791] drm/i915: Rename skl+ scaler binding bits
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename the scaler binding bits to match the spec more closely.
Also call the parameters 'plane_id' to make it a bit more clear
what to pass in.

v2: Don't break gvt

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-4-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 12 ++++++------
 drivers/gpu/drm/i915/gvt/handlers.c       |  2 +-
 drivers/gpu/drm/i915/i915_reg.h           |  9 +++++----
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index ec930aec21c4c..a96f8ecbeec13 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -396,7 +396,7 @@ static int intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_stat
 			mode = PS_SCALER_MODE_PLANAR;
 
 			if (linked)
-				mode |= PS_PLANE_Y_SEL(linked->id);
+				mode |= PS_BINDING_Y_PLANE(linked->id);
 		}
 	} else if (DISPLAY_VER(dev_priv) >= 10) {
 		mode = PS_SCALER_MODE_NORMAL;
@@ -741,8 +741,8 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state)
 
 	id = scaler_state->scaler_id;
 
-	ps_ctrl = skl_scaler_get_filter_select(crtc_state->hw.scaling_filter, 0);
-	ps_ctrl |=  PS_SCALER_EN | scaler_state->scalers[id].mode;
+	ps_ctrl = PS_SCALER_EN | PS_BINDING_PIPE | scaler_state->scalers[id].mode |
+		skl_scaler_get_filter_select(crtc_state->hw.scaling_filter, 0);
 
 	skl_scaler_setup_filter(dev_priv, pipe, id, 0,
 				crtc_state->hw.scaling_filter);
@@ -804,8 +804,8 @@ skl_program_plane_scaler(struct intel_plane *plane,
 		uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false);
 	}
 
-	ps_ctrl = skl_scaler_get_filter_select(plane_state->hw.scaling_filter, 0);
-	ps_ctrl |= PS_SCALER_EN | PS_PLANE_SEL(plane->id) | scaler->mode;
+	ps_ctrl = PS_SCALER_EN | PS_BINDING_PLANE(plane->id) | scaler->mode |
+		skl_scaler_get_filter_select(plane_state->hw.scaling_filter, 0);
 
 	skl_scaler_setup_filter(dev_priv, pipe, scaler_id, 0,
 				plane_state->hw.scaling_filter);
@@ -870,7 +870,7 @@ void skl_scaler_get_config(struct intel_crtc_state *crtc_state)
 		u32 ctl, pos, size;
 
 		ctl = intel_de_read(dev_priv, SKL_PS_CTRL(crtc->pipe, i));
-		if ((ctl & (PS_SCALER_EN | PS_PLANE_SEL_MASK)) != PS_SCALER_EN)
+		if ((ctl & (PS_SCALER_EN | PS_BINDING_MASK)) != (PS_SCALER_EN | PS_BINDING_PIPE))
 			continue;
 
 		id = i;
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 4b45a041ac5c9..a9f7fa9b90bda 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1562,7 +1562,7 @@ static int pf_write(struct intel_vgpu *vgpu,
 
 	if ((offset == _PS_1A_CTRL || offset == _PS_2A_CTRL ||
 	   offset == _PS_1B_CTRL || offset == _PS_2B_CTRL ||
-	   offset == _PS_1C_CTRL) && (val & PS_PLANE_SEL_MASK) != 0) {
+	   offset == _PS_1C_CTRL) && (val & PS_BINDING_MASK) != PS_BINDING_PIPE) {
 		drm_WARN_ONCE(&i915->drm, true,
 			      "VM(%d): guest is trying to scaling a plane\n",
 			      vgpu->id);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c050b9b996213..5225f497ab89a 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4051,8 +4051,9 @@
 #define SKL_PS_SCALER_MODE_NV12 (2 << 28)
 #define PS_SCALER_MODE_PLANAR (1 << 29)
 #define PS_SCALER_MODE_NORMAL (0 << 29)
-#define PS_PLANE_SEL_MASK  (7 << 25)
-#define PS_PLANE_SEL(plane) (((plane) + 1) << 25)
+#define PS_BINDING_MASK (7 << 25)
+#define PS_BINDING_PIPE (0 << 25)
+#define PS_BINDING_PLANE(plane_id) (((plane_id) + 1) << 25)
 #define PS_FILTER_MASK         (3 << 23)
 #define PS_FILTER_MEDIUM       (0 << 23)
 #define PS_FILTER_PROGRAMMED   (1 << 23)
@@ -4068,8 +4069,8 @@
 #define PS_VADAPT_MODE_LEAST_ADAPT (0 << 5)
 #define PS_VADAPT_MODE_MOD_ADAPT   (1 << 5)
 #define PS_VADAPT_MODE_MOST_ADAPT  (3 << 5)
-#define PS_PLANE_Y_SEL_MASK  (7 << 5)
-#define PS_PLANE_Y_SEL(plane) (((plane) + 1) << 5)
+#define PS_BINDING_Y_MASK  (7 << 5)
+#define PS_BINDING_Y_PLANE(plane_id) (((plane_id) + 1) << 5)
 #define PS_Y_VERT_FILTER_SELECT(set)   ((set) << 4)
 #define PS_Y_HORZ_FILTER_SELECT(set)   ((set) << 3)
 #define PS_UV_VERT_FILTER_SELECT(set)  ((set) << 2)
-- 
GitLab


From 576032f2b3db7ed5994a165ca0e4b6808c637f0f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:16 +0300
Subject: [PATCH 0390/1791] drm/i915:
 s/PS_COEE_INDEX_AUTO_INC/PS_COEF_INDEX_AUTO_INC/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix a typo in the PS_COEF_INDEX_AUTO_INC define.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-5-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 2 +-
 drivers/gpu/drm/i915/i915_reg.h           | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index a96f8ecbeec13..4437d130293a3 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -658,7 +658,7 @@ static void glk_program_nearest_filter_coefs(struct drm_i915_private *dev_priv,
 	int i;
 
 	intel_de_write_fw(dev_priv, GLK_PS_COEF_INDEX_SET(pipe, id, set),
-			  PS_COEE_INDEX_AUTO_INC);
+			  PS_COEF_INDEX_AUTO_INC);
 
 	for (i = 0; i < 17 * 7; i += 2) {
 		u32 tmp;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5225f497ab89a..ab6ca915101b5 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4141,7 +4141,7 @@
 #define _PS_COEF_SET0_INDEX_2A	   0x68298
 #define _PS_COEF_SET0_INDEX_1B	   0x68998
 #define _PS_COEF_SET0_INDEX_2B	   0x68A98
-#define PS_COEE_INDEX_AUTO_INC	   (1 << 10)
+#define PS_COEF_INDEX_AUTO_INC	   (1 << 10)
 
 #define _PS_COEF_SET0_DATA_1A	   0x6819C
 #define _PS_COEF_SET0_DATA_2A	   0x6829C
-- 
GitLab


From 6ec91794ce4aad0e92a16028885a7b139fe363ae Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:17 +0300
Subject: [PATCH 0391/1791] drm/i915: Define bitmasks for skl+ scaler window
 pos/size
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Define and use the bitmasks for the x/y components
of the skl+ scaler window pos/size registers.

We stick to the full 16 bits mask here even though the
hardware limits are actually lower. The current (ADL)
hardware maximums are in fact: 14 bits for X size, 13 bits
for X pos, 13 bits for Y size/pos. Yes, that is correct,
X pos has less bits than the X size for some reason. But
that  doesn't actually matter for now as we don't currently
even support such wide displays without the use of bigjoiner
(due to max plane width limit).

v2: Switch back to full 16bit masks since that's what
    we use transcoder timign regs and PIPESRC as well

Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-6-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 14 ++++++++------
 drivers/gpu/drm/i915/i915_reg.h           |  8 ++++++++
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 4437d130293a3..1e7c97243fcf5 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -754,9 +754,9 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state)
 	intel_de_write_fw(dev_priv, SKL_PS_HPHASE(pipe, id),
 			  PS_Y_PHASE(0) | PS_UV_RGB_PHASE(uv_rgb_hphase));
 	intel_de_write_fw(dev_priv, SKL_PS_WIN_POS(pipe, id),
-			  x << 16 | y);
+			  PS_WIN_XPOS(x) | PS_WIN_YPOS(y));
 	intel_de_write_fw(dev_priv, SKL_PS_WIN_SZ(pipe, id),
-			  width << 16 | height);
+			  PS_WIN_XSIZE(width) | PS_WIN_YSIZE(height));
 }
 
 void
@@ -816,9 +816,9 @@ skl_program_plane_scaler(struct intel_plane *plane,
 	intel_de_write_fw(dev_priv, SKL_PS_HPHASE(pipe, scaler_id),
 			  PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase));
 	intel_de_write_fw(dev_priv, SKL_PS_WIN_POS(pipe, scaler_id),
-			  (crtc_x << 16) | crtc_y);
+			  PS_WIN_XPOS(crtc_x) | PS_WIN_YPOS(crtc_y));
 	intel_de_write_fw(dev_priv, SKL_PS_WIN_SZ(pipe, scaler_id),
-			  (crtc_w << 16) | crtc_h);
+			  PS_WIN_XSIZE(crtc_w) | PS_WIN_YSIZE(crtc_h));
 }
 
 static void skl_detach_scaler(struct intel_crtc *crtc, int id)
@@ -880,8 +880,10 @@ void skl_scaler_get_config(struct intel_crtc_state *crtc_state)
 		size = intel_de_read(dev_priv, SKL_PS_WIN_SZ(crtc->pipe, i));
 
 		drm_rect_init(&crtc_state->pch_pfit.dst,
-			      pos >> 16, pos & 0xffff,
-			      size >> 16, size & 0xffff);
+			      REG_FIELD_GET(PS_WIN_XPOS_MASK, pos),
+			      REG_FIELD_GET(PS_WIN_YPOS_MASK, pos),
+			      REG_FIELD_GET(PS_WIN_XSIZE_MASK, size),
+			      REG_FIELD_GET(PS_WIN_YSIZE_MASK, size));
 
 		scaler_state->scalers[i].in_use = true;
 		break;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ab6ca915101b5..20041b00fbc3b 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4096,12 +4096,20 @@
 #define _PS_WIN_POS_1B      0x68970
 #define _PS_WIN_POS_2B      0x68A70
 #define _PS_WIN_POS_1C      0x69170
+#define   PS_WIN_XPOS_MASK			REG_GENMASK(31, 16)
+#define   PS_WIN_XPOS(x)			REG_FIELD_PREP(PS_WIN_XPOS_MASK, (x))
+#define   PS_WIN_YPOS_MASK			REG_GENMASK(15, 0)
+#define   PS_WIN_YPOS(y)			REG_FIELD_PREP(PS_WIN_YPOS_MASK, (y))
 
 #define _PS_WIN_SZ_1A       0x68174
 #define _PS_WIN_SZ_2A       0x68274
 #define _PS_WIN_SZ_1B       0x68974
 #define _PS_WIN_SZ_2B       0x68A74
 #define _PS_WIN_SZ_1C       0x69174
+#define   PS_WIN_XSIZE_MASK			REG_GENMASK(31, 16)
+#define   PS_WIN_XSIZE(w)			REG_FIELD_PREP(PS_WIN_XSIZE_MASK, (w))
+#define   PS_WIN_YSIZE_MASK			REG_GENMASK(15, 0)
+#define   PS_WIN_YSIZE(h)			REG_FIELD_PREP(PS_WIN_YSIZE_MASK, (h))
 
 #define _PS_VSCALE_1A       0x68184
 #define _PS_VSCALE_2A       0x68284
-- 
GitLab


From e62206682a2f51a8f4d37a204791e5e8ae4d9eab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:18 +0300
Subject: [PATCH 0392/1791] drm/i915: Use REG_BIT() & co. for pipe scaler
 registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pimp the skl+ scaler register bits with REG_BIT()/etc.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-7-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/i915_reg.h | 98 ++++++++++++++++++---------------
 1 file changed, 53 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 20041b00fbc3b..adad3de3dfa54 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4044,52 +4044,58 @@
 #define _PS_1B_CTRL      0x68980
 #define _PS_2B_CTRL      0x68A80
 #define _PS_1C_CTRL      0x69180
-#define PS_SCALER_EN        (1 << 31)
-#define SKL_PS_SCALER_MODE_MASK (3 << 28)
-#define SKL_PS_SCALER_MODE_DYN  (0 << 28)
-#define SKL_PS_SCALER_MODE_HQ  (1 << 28)
-#define SKL_PS_SCALER_MODE_NV12 (2 << 28)
-#define PS_SCALER_MODE_PLANAR (1 << 29)
-#define PS_SCALER_MODE_NORMAL (0 << 29)
-#define PS_BINDING_MASK (7 << 25)
-#define PS_BINDING_PIPE (0 << 25)
-#define PS_BINDING_PLANE(plane_id) (((plane_id) + 1) << 25)
-#define PS_FILTER_MASK         (3 << 23)
-#define PS_FILTER_MEDIUM       (0 << 23)
-#define PS_FILTER_PROGRAMMED   (1 << 23)
-#define PS_FILTER_EDGE_ENHANCE (2 << 23)
-#define PS_FILTER_BILINEAR     (3 << 23)
-#define PS_VERT3TAP            (1 << 21)
-#define PS_VERT_INT_INVERT_FIELD1 (0 << 20)
-#define PS_VERT_INT_INVERT_FIELD0 (1 << 20)
-#define PS_PWRUP_PROGRESS         (1 << 17)
-#define PS_V_FILTER_BYPASS        (1 << 8)
-#define PS_VADAPT_EN              (1 << 7)
-#define PS_VADAPT_MODE_MASK        (3 << 5)
-#define PS_VADAPT_MODE_LEAST_ADAPT (0 << 5)
-#define PS_VADAPT_MODE_MOD_ADAPT   (1 << 5)
-#define PS_VADAPT_MODE_MOST_ADAPT  (3 << 5)
-#define PS_BINDING_Y_MASK  (7 << 5)
-#define PS_BINDING_Y_PLANE(plane_id) (((plane_id) + 1) << 5)
-#define PS_Y_VERT_FILTER_SELECT(set)   ((set) << 4)
-#define PS_Y_HORZ_FILTER_SELECT(set)   ((set) << 3)
-#define PS_UV_VERT_FILTER_SELECT(set)  ((set) << 2)
-#define PS_UV_HORZ_FILTER_SELECT(set)  ((set) << 1)
+#define   PS_SCALER_EN				REG_BIT(31)
+#define   SKL_PS_SCALER_MODE_MASK		REG_GENMASK(29, 28) /* skl/bxt */
+#define   SKL_PS_SCALER_MODE_DYN		REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 0)
+#define   SKL_PS_SCALER_MODE_HQ			REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 1)
+#define   SKL_PS_SCALER_MODE_NV12		REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 2)
+#define   PS_SCALER_MODE_MASK			REG_BIT(29) /* glk-tgl */
+#define   PS_SCALER_MODE_NORMAL			REG_FIELD_PREP(PS_SCALER_MODE_MASK, 0)
+#define   PS_SCALER_MODE_PLANAR			REG_FIELD_PREP(PS_SCALER_MODE_MASK, 1)
+#define   PS_BINDING_MASK			REG_GENMASK(27, 25)
+#define   PS_BINDING_PIPE			REG_FIELD_PREP(PS_BINDING_MASK, 0)
+#define   PS_BINDING_PLANE(plane_id)		REG_FIELD_PREP(PS_BINDING_MASK, (plane_id) + 1)
+#define   PS_FILTER_MASK			REG_GENMASK(24, 23)
+#define   PS_FILTER_MEDIUM			REG_FIELD_PREP(PS_FILTER_MASK, 0)
+#define   PS_FILTER_PROGRAMMED			REG_FIELD_PREP(PS_FILTER_MASK, 1)
+#define   PS_FILTER_EDGE_ENHANCE		REG_FIELD_PREP(PS_FILTER_MASK, 2)
+#define   PS_FILTER_BILINEAR			REG_FIELD_PREP(PS_FILTER_MASK, 3)
+#define   PS_VERT3TAP				REG_BIT(21) /* skl/bxt */
+#define   PS_VERT_INT_INVERT_FIELD		REG_BIT(20)
+#define   PS_PWRUP_PROGRESS			REG_BIT(17)
+#define   PS_V_FILTER_BYPASS			REG_BIT(8)
+#define   PS_VADAPT_EN				REG_BIT(7) /* skl/bxt */
+#define   PS_VADAPT_MODE_MASK			REG_GENMASK(6, 5) /* skl/bxt */
+#define   PS_VADAPT_MODE_LEAST_ADAPT		REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 0)
+#define   PS_VADAPT_MODE_MOD_ADAPT		REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 1)
+#define   PS_VADAPT_MODE_MOST_ADAPT		REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 3)
+#define   PS_BINDING_Y_MASK			REG_GENMASK(7, 5) /* icl-tgl */
+#define   PS_BINDING_Y_PLANE(plane_id)		REG_FIELD_PREP(PS_BINDING_Y_MASK, (plane_id) + 1)
+#define   PS_Y_VERT_FILTER_SELECT_MASK		REG_BIT(4) /* glk+ */
+#define   PS_Y_VERT_FILTER_SELECT(set)		REG_FIELD_PREP(PS_Y_VERT_FILTER_SELECT_MASK, (set))
+#define   PS_Y_HORZ_FILTER_SELECT_MASK		REG_BIT(3) /* glk+ */
+#define   PS_Y_HORZ_FILTER_SELECT(set)		REG_FIELD_PREP(PS_Y_HORZ_FILTER_SELECT_MASK, (set))
+#define   PS_UV_VERT_FILTER_SELECT_MASK		REG_BIT(2) /* glk+ */
+#define   PS_UV_VERT_FILTER_SELECT(set)		REG_FIELD_PREP(PS_UV_VERT_FILTER_SELECT_MASK, (set))
+#define   PS_UV_HORZ_FILTER_SELECT_MASK		REG_BIT(1) /* glk+ */
+#define   PS_UV_HORZ_FILTER_SELECT(set)		REG_FIELD_PREP(PS_UV_HORZ_FILTER_SELECT_MASK, (set))
 
 #define _PS_PWR_GATE_1A     0x68160
 #define _PS_PWR_GATE_2A     0x68260
 #define _PS_PWR_GATE_1B     0x68960
 #define _PS_PWR_GATE_2B     0x68A60
 #define _PS_PWR_GATE_1C     0x69160
-#define PS_PWR_GATE_DIS_OVERRIDE       (1 << 31)
-#define PS_PWR_GATE_SETTLING_TIME_32   (0 << 3)
-#define PS_PWR_GATE_SETTLING_TIME_64   (1 << 3)
-#define PS_PWR_GATE_SETTLING_TIME_96   (2 << 3)
-#define PS_PWR_GATE_SETTLING_TIME_128  (3 << 3)
-#define PS_PWR_GATE_SLPEN_8             0
-#define PS_PWR_GATE_SLPEN_16            1
-#define PS_PWR_GATE_SLPEN_24            2
-#define PS_PWR_GATE_SLPEN_32            3
+#define   PS_PWR_GATE_DIS_OVERRIDE		REG_BIT(31)
+#define   PS_PWR_GATE_SETTLING_TIME_MASK	REG_GENMASK(4, 3)
+#define   PS_PWR_GATE_SETTLING_TIME_32		REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 0)
+#define   PS_PWR_GATE_SETTLING_TIME_64		REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 1)
+#define   PS_PWR_GATE_SETTLING_TIME_96		REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 2)
+#define   PS_PWR_GATE_SETTLING_TIME_128		REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 3)
+#define   PS_PWR_GATE_SLPEN_MASK		REG_GENMASK(1, 0)
+#define   PS_PWR_GATE_SLPEN_8			REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 0)
+#define   PS_PWR_GATE_SLPEN_16			REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 1)
+#define   PS_PWR_GATE_SLPEN_24			REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 2)
+#define   PS_PWR_GATE_SLPEN_32			REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 3)
 
 #define _PS_WIN_POS_1A      0x68170
 #define _PS_WIN_POS_2A      0x68270
@@ -4128,10 +4134,12 @@
 #define _PS_VPHASE_1B       0x68988
 #define _PS_VPHASE_2B       0x68A88
 #define _PS_VPHASE_1C       0x69188
-#define  PS_Y_PHASE(x)		((x) << 16)
-#define  PS_UV_RGB_PHASE(x)	((x) << 0)
-#define   PS_PHASE_MASK	(0x7fff << 1) /* u2.13 */
-#define   PS_PHASE_TRIP	(1 << 0)
+#define   PS_Y_PHASE_MASK			REG_GENMASK(31, 16)
+#define   PS_Y_PHASE(x)				REG_FIELD_PREP(PS_Y_PHASE_MASK, (x))
+#define   PS_UV_RGB_PHASE_MASK			REG_GENMASK(15, 0)
+#define   PS_UV_RGB_PHASE(x)			REG_FIELD_PREP(PS_UV_RGB_PHASE_MASK, (x))
+#define   PS_PHASE_MASK				(0x7fff << 1) /* u2.13 */
+#define   PS_PHASE_TRIP				(1 << 0)
 
 #define _PS_HPHASE_1A       0x68194
 #define _PS_HPHASE_2A       0x68294
@@ -4149,7 +4157,7 @@
 #define _PS_COEF_SET0_INDEX_2A	   0x68298
 #define _PS_COEF_SET0_INDEX_1B	   0x68998
 #define _PS_COEF_SET0_INDEX_2B	   0x68A98
-#define PS_COEF_INDEX_AUTO_INC	   (1 << 10)
+#define   PS_COEF_INDEX_AUTO_INC		REG_BIT(10)
 
 #define _PS_COEF_SET0_DATA_1A	   0x6819C
 #define _PS_COEF_SET0_DATA_2A	   0x6829C
-- 
GitLab


From e971121350e72ff34a0c1d140127703cc2c085c2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:22 +0300
Subject: [PATCH 0393/1791] drm/i915/vma: fix struct i915_vma_bindinfo
 kernel-doc

You can't document both a sub-struct type and a struct member at the
same time. Separate them.

drivers/gpu/drm/i915/i915_vma_resource.h:91: warning: Incorrect use of kernel-doc format:          * struct i915_vma_bindinfo - Information needed for async bind
drivers/gpu/drm/i915/i915_vma_resource.h:129: warning: Function parameter or member 'bi' not described in 'i915_vma_resource'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/89ab5cf179566c429383cc57db746038f75cba0d.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_vma_resource.h | 45 ++++++++++++++----------
 1 file changed, 27 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_vma_resource.h b/drivers/gpu/drm/i915/i915_vma_resource.h
index 2053c037dbfb1..ca2b0f7f59bce 100644
--- a/drivers/gpu/drm/i915/i915_vma_resource.h
+++ b/drivers/gpu/drm/i915/i915_vma_resource.h
@@ -33,6 +33,27 @@ struct i915_page_sizes {
 	unsigned int sg;
 };
 
+/**
+ * struct i915_vma_bindinfo - Information needed for async bind
+ * only but that can be dropped after the bind has taken place.
+ * Consider making this a separate argument to the bind_vma
+ * op, coalescing with other arguments like vm, stash, cache_level
+ * and flags
+ * @pages: The pages sg-table.
+ * @page_sizes: Page sizes of the pages.
+ * @pages_rsgt: Refcounted sg-table when delayed object destruction
+ * is supported. May be NULL.
+ * @readonly: Whether the vma should be bound read-only.
+ * @lmem: Whether the vma points to lmem.
+ */
+struct i915_vma_bindinfo {
+	struct sg_table *pages;
+	struct i915_page_sizes page_sizes;
+	struct i915_refct_sgt *pages_rsgt;
+	bool readonly:1;
+	bool lmem:1;
+};
+
 /**
  * struct i915_vma_resource - Snapshotted unbind information.
  * @unbind_fence: Fence to mark unbinding complete. Note that this fence
@@ -89,25 +110,13 @@ struct i915_vma_resource {
 	intel_wakeref_t wakeref;
 
 	/**
-	 * struct i915_vma_bindinfo - Information needed for async bind
-	 * only but that can be dropped after the bind has taken place.
-	 * Consider making this a separate argument to the bind_vma
-	 * op, coalescing with other arguments like vm, stash, cache_level
-	 * and flags
-	 * @pages: The pages sg-table.
-	 * @page_sizes: Page sizes of the pages.
-	 * @pages_rsgt: Refcounted sg-table when delayed object destruction
-	 * is supported. May be NULL.
-	 * @readonly: Whether the vma should be bound read-only.
-	 * @lmem: Whether the vma points to lmem.
+	 * @bi: Information needed for async bind only but that can be dropped
+	 * after the bind has taken place.
+	 *
+	 * Consider making this a separate argument to the bind_vma op,
+	 * coalescing with other arguments like vm, stash, cache_level and flags
 	 */
-	struct i915_vma_bindinfo {
-		struct sg_table *pages;
-		struct i915_page_sizes page_sizes;
-		struct i915_refct_sgt *pages_rsgt;
-		bool readonly:1;
-		bool lmem:1;
-	} bi;
+	struct i915_vma_bindinfo bi;
 
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
 	struct intel_memory_region *mr;
-- 
GitLab


From 08272a5a41876af38209def142de272e580d446e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:27 +0300
Subject: [PATCH 0394/1791] drm/i915/gem: fix function pointer member
 kernel-doc

You can't document function pointer member as functions.

drivers/gpu/drm/i915/gem/i915_gem_region.h:25: warning: Incorrect use of kernel-doc format:          * process_obj - Process the current object
drivers/gpu/drm/i915/gem/i915_gem_region.h:35: warning: Function parameter or member 'process_obj' not described in 'i915_gem_apply_to_region_ops'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a2a81f9ee78591def0534c81a63dbc652c44bbd3.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_region.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_region.h b/drivers/gpu/drm/i915/gem/i915_gem_region.h
index 2dfcc41c0170c..8a7650b27cc2e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_region.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_region.h
@@ -22,9 +22,7 @@ struct i915_gem_apply_to_region;
  */
 struct i915_gem_apply_to_region_ops {
 	/**
-	 * process_obj - Process the current object
-	 * @apply: Embed this for private data.
-	 * @obj: The current object.
+	 * @process_obj: Process the current object
 	 *
 	 * Note that if this function is part of a ww transaction, and
 	 * if returns -EDEADLK for one of the objects, it may be
-- 
GitLab


From 9d7fe94d35522115734b3c2c37cad3257c88c43b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:39 +0300
Subject: [PATCH 0395/1791] drm/i915/scatterlist: fix kernel-doc

Can't document function pointer members as if they are functions.

drivers/gpu/drm/i915/i915_scatterlist.h:160: warning: Incorrect use of kernel-doc format:          * release() - Free the memory of the struct i915_refct_sgt

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a0c099707ec4f1911b14b0f286848a298b2b29e0.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_scatterlist.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_scatterlist.h b/drivers/gpu/drm/i915/i915_scatterlist.h
index 12c1fd63e9b8d..5a10c1a311831 100644
--- a/drivers/gpu/drm/i915/i915_scatterlist.h
+++ b/drivers/gpu/drm/i915/i915_scatterlist.h
@@ -157,8 +157,7 @@ bool i915_sg_trim(struct sg_table *orig_st);
  */
 struct i915_refct_sgt_ops {
 	/**
-	 * release() - Free the memory of the struct i915_refct_sgt
-	 * @ref: struct kref that is embedded in the struct i915_refct_sgt
+	 * @release: Free the memory of the struct i915_refct_sgt
 	 */
 	void (*release)(struct kref *ref);
 };
-- 
GitLab


From c6948d8c221a8a61ebeb69567fd1a6c820127b53 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 4 May 2023 12:23:20 +0300
Subject: [PATCH 0396/1791] drm/i915/ttm: fix i915_ttm_to_gem() kernel-doc

drivers/gpu/drm/i915/gem/i915_gem_ttm.h:50: warning: Function parameter or member 'bo' not described in 'i915_ttm_to_gem'

v2: Fix return value documentation while at it (Rodrigo)

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504092320.1787627-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_ttm.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.h b/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
index f8f6bed1b2974..67347e62e29b3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.h
@@ -42,8 +42,9 @@ static inline bool i915_ttm_is_ghost_object(struct ttm_buffer_object *bo)
 /**
  * i915_ttm_to_gem - Convert a struct ttm_buffer_object to an embedding
  * struct drm_i915_gem_object.
+ * @bo: Pointer to the ttm buffer object
  *
- * Return: Pointer to the embedding struct ttm_buffer_object.
+ * Return: Pointer to the embedding struct drm_i915_gem_object.
  */
 static inline struct drm_i915_gem_object *
 i915_ttm_to_gem(struct ttm_buffer_object *bo)
-- 
GitLab


From ceb0cc3b428825fefae0123d2061ad3e06e2a0f7 Mon Sep 17 00:00:00 2001
From: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Date: Thu, 4 May 2023 12:39:59 +0300
Subject: [PATCH 0397/1791] drm/i915: Communicate display power demands to
 pcode

Display to communicate display pipe count/CDCLK/voltage configuration
to Pcode for more accurate power accounting for DG2.
Existing sequence is only sending the voltage value to the Pcode.
Adding new sequence with current cdclk associate with voltage value masking.
Adding pcode request when any pipe power well will disable or enable.

v2: - Make intel_cdclk_need_serialize static to make CI compiler happy.
v3: - Removed redundant return(Jani Nikula)
    - Changed intel_cdclk_power_usage_to_pcode_(pre|post)_notification to be
      static and also naming to intel_cdclk_pcode_(pre|post)_notify(Jani Nikula)
    - Changed u8 to be u16 for cdclk parameter in intel_pcode_notify function,
      as according to BSpec it requires 10 bits(Jani Nikula)
    - Replaced dev_priv's with i915's(Jani Nikula)
    - Simplified expression in intel_cdclk_need_serialize(Jani Nikula)
    - Removed redundant kernel-doc and indentation(Jani Nikula)
v4: - Fixed some checkpatch warnings
v5: - According to HW team comments that change should affect only DG2,
      fix correspodent platform check to account this.
v6: - Added one more missing IS_DG2 check(Vinod Govindapillai)

Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504093959.12085-1-stanislav.lisovskiy@intel.com
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 158 +++++++++++++++++++--
 drivers/gpu/drm/i915/i915_reg.h            |  14 ++
 2 files changed, 160 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index f6223d8f13b8f..a11092deaba67 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -1898,7 +1898,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 	 */
 	if (DISPLAY_VER(dev_priv) >= 14)
 		/* NOOP */;
-	else if (DISPLAY_VER(dev_priv) >= 11)
+	else if (DISPLAY_VER(dev_priv) >= 11 && !IS_DG2(dev_priv))
 		ret = skl_pcode_request(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL,
 					SKL_CDCLK_PREPARE_FOR_CHANGE,
 					SKL_CDCLK_READY_FOR_CHANGE,
@@ -1932,10 +1932,10 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 		 * NOOP - No Pcode communication needed for
 		 * Display versions 14 and beyond
 		 */;
-	else if (DISPLAY_VER(dev_priv) >= 11)
+	else if (DISPLAY_VER(dev_priv) >= 11 && !IS_DG2(dev_priv))
 		ret = snb_pcode_write(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL,
 				      cdclk_config->voltage_level);
-	else
+	if (DISPLAY_VER(dev_priv) < 11) {
 		/*
 		 * The timeout isn't specified, the 2ms used here is based on
 		 * experiment.
@@ -1946,7 +1946,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 					      HSW_PCODE_DE_WRITE_FREQ_REQ,
 					      cdclk_config->voltage_level,
 					      150, 2);
-
+	}
 	if (ret) {
 		drm_err(&dev_priv->drm,
 			"PCode CDCLK freq set failed, (err %d, freq %d)\n",
@@ -2242,6 +2242,38 @@ void intel_cdclk_dump_config(struct drm_i915_private *i915,
 		    cdclk_config->voltage_level);
 }
 
+static void intel_pcode_notify(struct drm_i915_private *i915,
+			       u8 voltage_level,
+			       u8 active_pipe_count,
+			       u16 cdclk,
+			       bool cdclk_update_valid,
+			       bool pipe_count_update_valid)
+{
+	int ret;
+	u32 update_mask = 0;
+
+	if (!IS_DG2(i915))
+		return;
+
+	update_mask = DISPLAY_TO_PCODE_UPDATE_MASK(cdclk, active_pipe_count, voltage_level);
+
+	if (cdclk_update_valid)
+		update_mask |= DISPLAY_TO_PCODE_CDCLK_VALID;
+
+	if (pipe_count_update_valid)
+		update_mask |= DISPLAY_TO_PCODE_PIPE_COUNT_VALID;
+
+	ret = skl_pcode_request(&i915->uncore, SKL_PCODE_CDCLK_CONTROL,
+				SKL_CDCLK_PREPARE_FOR_CHANGE |
+				update_mask,
+				SKL_CDCLK_READY_FOR_CHANGE,
+				SKL_CDCLK_READY_FOR_CHANGE, 3);
+	if (ret)
+		drm_err(&i915->drm,
+			"Failed to inform PCU about display config (err %d)\n",
+			ret);
+}
+
 /**
  * intel_set_cdclk - Push the CDCLK configuration to the hardware
  * @dev_priv: i915 device
@@ -2311,6 +2343,88 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv,
 	}
 }
 
+static void intel_cdclk_pcode_pre_notify(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_cdclk_state *old_cdclk_state =
+		intel_atomic_get_old_cdclk_state(state);
+	const struct intel_cdclk_state *new_cdclk_state =
+		intel_atomic_get_new_cdclk_state(state);
+	unsigned int cdclk = 0; u8 voltage_level, num_active_pipes = 0;
+	bool change_cdclk, update_pipe_count;
+
+	if (!intel_cdclk_changed(&old_cdclk_state->actual,
+				 &new_cdclk_state->actual) &&
+				 new_cdclk_state->active_pipes ==
+				 old_cdclk_state->active_pipes)
+		return;
+
+	/* According to "Sequence Before Frequency Change", voltage level set to 0x3 */
+	voltage_level = DISPLAY_TO_PCODE_VOLTAGE_MAX;
+
+	change_cdclk = new_cdclk_state->actual.cdclk != old_cdclk_state->actual.cdclk;
+	update_pipe_count = hweight8(new_cdclk_state->active_pipes) >
+			    hweight8(old_cdclk_state->active_pipes);
+
+	/*
+	 * According to "Sequence Before Frequency Change",
+	 * if CDCLK is increasing, set bits 25:16 to upcoming CDCLK,
+	 * if CDCLK is decreasing or not changing, set bits 25:16 to current CDCLK,
+	 * which basically means we choose the maximum of old and new CDCLK, if we know both
+	 */
+	if (change_cdclk)
+		cdclk = max(new_cdclk_state->actual.cdclk, old_cdclk_state->actual.cdclk);
+
+	/*
+	 * According to "Sequence For Pipe Count Change",
+	 * if pipe count is increasing, set bits 25:16 to upcoming pipe count
+	 * (power well is enabled)
+	 * no action if it is decreasing, before the change
+	 */
+	if (update_pipe_count)
+		num_active_pipes = hweight8(new_cdclk_state->active_pipes);
+
+	intel_pcode_notify(i915, voltage_level, num_active_pipes, cdclk,
+			   change_cdclk, update_pipe_count);
+}
+
+static void intel_cdclk_pcode_post_notify(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_cdclk_state *new_cdclk_state =
+		intel_atomic_get_new_cdclk_state(state);
+	const struct intel_cdclk_state *old_cdclk_state =
+		intel_atomic_get_old_cdclk_state(state);
+	unsigned int cdclk = 0; u8 voltage_level, num_active_pipes = 0;
+	bool update_cdclk, update_pipe_count;
+
+	/* According to "Sequence After Frequency Change", set voltage to used level */
+	voltage_level = new_cdclk_state->actual.voltage_level;
+
+	update_cdclk = new_cdclk_state->actual.cdclk != old_cdclk_state->actual.cdclk;
+	update_pipe_count = hweight8(new_cdclk_state->active_pipes) <
+			    hweight8(old_cdclk_state->active_pipes);
+
+	/*
+	 * According to "Sequence After Frequency Change",
+	 * set bits 25:16 to current CDCLK
+	 */
+	if (update_cdclk)
+		cdclk = new_cdclk_state->actual.cdclk;
+
+	/*
+	 * According to "Sequence For Pipe Count Change",
+	 * if pipe count is decreasing, set bits 25:16 to current pipe count,
+	 * after the change(power well is disabled)
+	 * no action if it is increasing, after the change
+	 */
+	if (update_pipe_count)
+		num_active_pipes = hweight8(new_cdclk_state->active_pipes);
+
+	intel_pcode_notify(i915, voltage_level, num_active_pipes, cdclk,
+			   update_cdclk, update_pipe_count);
+}
+
 /**
  * intel_set_cdclk_pre_plane_update - Push the CDCLK state to the hardware
  * @state: intel atomic state
@@ -2321,7 +2435,7 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv,
 void
 intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	const struct intel_cdclk_state *old_cdclk_state =
 		intel_atomic_get_old_cdclk_state(state);
 	const struct intel_cdclk_state *new_cdclk_state =
@@ -2332,11 +2446,14 @@ intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state)
 				 &new_cdclk_state->actual))
 		return;
 
+	if (IS_DG2(i915))
+		intel_cdclk_pcode_pre_notify(state);
+
 	if (pipe == INVALID_PIPE ||
 	    old_cdclk_state->actual.cdclk <= new_cdclk_state->actual.cdclk) {
-		drm_WARN_ON(&dev_priv->drm, !new_cdclk_state->base.changed);
+		drm_WARN_ON(&i915->drm, !new_cdclk_state->base.changed);
 
-		intel_set_cdclk(dev_priv, &new_cdclk_state->actual, pipe);
+		intel_set_cdclk(i915, &new_cdclk_state->actual, pipe);
 	}
 }
 
@@ -2350,7 +2467,7 @@ intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state)
 void
 intel_set_cdclk_post_plane_update(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	const struct intel_cdclk_state *old_cdclk_state =
 		intel_atomic_get_old_cdclk_state(state);
 	const struct intel_cdclk_state *new_cdclk_state =
@@ -2361,11 +2478,14 @@ intel_set_cdclk_post_plane_update(struct intel_atomic_state *state)
 				 &new_cdclk_state->actual))
 		return;
 
+	if (IS_DG2(i915))
+		intel_cdclk_pcode_post_notify(state);
+
 	if (pipe != INVALID_PIPE &&
 	    old_cdclk_state->actual.cdclk > new_cdclk_state->actual.cdclk) {
-		drm_WARN_ON(&dev_priv->drm, !new_cdclk_state->base.changed);
+		drm_WARN_ON(&i915->drm, !new_cdclk_state->base.changed);
 
-		intel_set_cdclk(dev_priv, &new_cdclk_state->actual, pipe);
+		intel_set_cdclk(i915, &new_cdclk_state->actual, pipe);
 	}
 }
 
@@ -2871,6 +2991,21 @@ int intel_cdclk_init(struct drm_i915_private *dev_priv)
 	return 0;
 }
 
+static bool intel_cdclk_need_serialize(struct drm_i915_private *i915,
+				       const struct intel_cdclk_state *old_cdclk_state,
+				       const struct intel_cdclk_state *new_cdclk_state)
+{
+	bool power_well_cnt_changed = hweight8(old_cdclk_state->active_pipes) !=
+				      hweight8(new_cdclk_state->active_pipes);
+	bool cdclk_changed = intel_cdclk_changed(&old_cdclk_state->actual,
+						 &new_cdclk_state->actual);
+	/*
+	 * We need to poke hw for gen >= 12, because we notify PCode if
+	 * pipe power well count changes.
+	 */
+	return cdclk_changed || (IS_DG2(i915) && power_well_cnt_changed);
+}
+
 int intel_modeset_calc_cdclk(struct intel_atomic_state *state)
 {
 	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
@@ -2892,8 +3027,7 @@ int intel_modeset_calc_cdclk(struct intel_atomic_state *state)
 	if (ret)
 		return ret;
 
-	if (intel_cdclk_changed(&old_cdclk_state->actual,
-				&new_cdclk_state->actual)) {
+	if (intel_cdclk_need_serialize(dev_priv, old_cdclk_state, new_cdclk_state)) {
 		/*
 		 * Also serialize commits across all crtcs
 		 * if the actual hw needs to be poked.
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index adad3de3dfa54..6a82c707bf9da 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5349,6 +5349,20 @@
 #define     ICL_PCODE_MEM_SS_READ_GLOBAL_INFO	(0x0 << 8)
 #define     ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point)	(((point) << 16) | (0x1 << 8))
 #define     ADL_PCODE_MEM_SS_READ_PSF_GV_INFO	((0) | (0x2 << 8))
+#define   DISPLAY_TO_PCODE_CDCLK_MAX		0x28D
+#define   DISPLAY_TO_PCODE_VOLTAGE_MASK		REG_GENMASK(1, 0)
+#define	  DISPLAY_TO_PCODE_VOLTAGE_MAX		DISPLAY_TO_PCODE_VOLTAGE_MASK
+#define   DISPLAY_TO_PCODE_CDCLK_VALID		REG_BIT(27)
+#define   DISPLAY_TO_PCODE_PIPE_COUNT_VALID	REG_BIT(31)
+#define   DISPLAY_TO_PCODE_CDCLK_MASK		REG_GENMASK(25, 16)
+#define   DISPLAY_TO_PCODE_PIPE_COUNT_MASK	REG_GENMASK(30, 28)
+#define   DISPLAY_TO_PCODE_CDCLK(x)		REG_FIELD_PREP(DISPLAY_TO_PCODE_CDCLK_MASK, (x))
+#define   DISPLAY_TO_PCODE_PIPE_COUNT(x)	REG_FIELD_PREP(DISPLAY_TO_PCODE_PIPE_COUNT_MASK, (x))
+#define   DISPLAY_TO_PCODE_VOLTAGE(x)		REG_FIELD_PREP(DISPLAY_TO_PCODE_VOLTAGE_MASK, (x))
+#define   DISPLAY_TO_PCODE_UPDATE_MASK(cdclk, num_pipes, voltage_level) \
+		((DISPLAY_TO_PCODE_CDCLK(cdclk)) | \
+		(DISPLAY_TO_PCODE_PIPE_COUNT(num_pipes)) | \
+		(DISPLAY_TO_PCODE_VOLTAGE(voltage_level)))
 #define   ICL_PCODE_SAGV_DE_MEM_SS_CONFIG	0xe
 #define     ICL_PCODE_REP_QGV_MASK		REG_GENMASK(1, 0)
 #define     ICL_PCODE_REP_QGV_SAFE		REG_FIELD_PREP(ICL_PCODE_REP_QGV_MASK, 0)
-- 
GitLab


From 1d5b09f8daf859247a1ea65b0d732a24d88980d8 Mon Sep 17 00:00:00 2001
From: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Date: Fri, 5 May 2023 11:22:12 +0300
Subject: [PATCH 0398/1791] drm/i915: Fix NULL ptr deref by checking
 new_crtc_state

intel_atomic_get_new_crtc_state can return NULL, unless crtc state wasn't
obtained previously with intel_atomic_get_crtc_state, so we must check it
for NULLness here, just as in many other places, where we can't guarantee
that intel_atomic_get_crtc_state was called.
We are currently getting NULL ptr deref because of that, so this fix was
confirmed to help.

Fixes: 74a75dc90869 ("drm/i915/display: move plane prepare/cleanup to intel_atomic_plane.c")
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505082212.27089-1-stanislav.lisovskiy@intel.com
---
 drivers/gpu/drm/i915/display/intel_atomic_plane.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
index 9f670dcfe76e4..4125ee07a2717 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
@@ -1029,7 +1029,7 @@ intel_prepare_plane_fb(struct drm_plane *_plane,
 	int ret;
 
 	if (old_obj) {
-		const struct intel_crtc_state *crtc_state =
+		const struct intel_crtc_state *new_crtc_state =
 			intel_atomic_get_new_crtc_state(state,
 							to_intel_crtc(old_plane_state->hw.crtc));
 
@@ -1044,7 +1044,7 @@ intel_prepare_plane_fb(struct drm_plane *_plane,
 		 * This should only fail upon a hung GPU, in which case we
 		 * can safely continue.
 		 */
-		if (intel_crtc_needs_modeset(crtc_state)) {
+		if (new_crtc_state && intel_crtc_needs_modeset(new_crtc_state)) {
 			ret = i915_sw_fence_await_reservation(&state->commit_ready,
 							      old_obj->base.resv,
 							      false, 0,
-- 
GitLab


From 83ac5457998ed464032665375dea56da8776a861 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 May 2023 18:37:41 +0300
Subject: [PATCH 0399/1791] drm/i915: use kernel-doc -Werror when
 CONFIG_DRM_I915_WERROR=y

With CONFIG_DRM_I915_WERROR=y, we enable kernel-doc check for both
objects and headers. Now that the kernel-doc warnings have been fixed,
also enable kernel-doc -Werror to fail the build on kernel-doc warnings.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/68ffcad0e6ff2b6cd70c6df28822f967898ce197.1683041799.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 1f9071ee6c83f..4494dcd2eb5d1 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -374,7 +374,7 @@ obj-$(CONFIG_DRM_I915_GVT_KVMGT) += kvmgt.o
 #
 # Enable locally for CONFIG_DRM_I915_WERROR=y. See also scripts/Makefile.build
 ifdef CONFIG_DRM_I915_WERROR
-    cmd_checkdoc = $(srctree)/scripts/kernel-doc -none $<
+    cmd_checkdoc = $(srctree)/scripts/kernel-doc -none -Werror $<
 endif
 
 # header test
@@ -389,7 +389,7 @@ always-$(CONFIG_DRM_I915_WERROR) += \
 
 quiet_cmd_hdrtest = HDRTEST $(patsubst %.hdrtest,%.h,$@)
       cmd_hdrtest = $(CC) $(filter-out $(CFLAGS_GCOV), $(c_flags)) -S -o /dev/null -x c /dev/null -include $<; \
-		$(srctree)/scripts/kernel-doc -none $<; touch $@
+		$(srctree)/scripts/kernel-doc -none -Werror $<; touch $@
 
 $(obj)/%.hdrtest: $(src)/%.h FORCE
 	$(call if_changed_dep,hdrtest)
-- 
GitLab


From 51f7008239de011370c5067bbba07f0207f06b72 Mon Sep 17 00:00:00 2001
From: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Date: Tue, 18 Apr 2023 07:04:30 -0700
Subject: [PATCH 0400/1791] drm/i915/dp: prevent potential div-by-zero

drm_dp_dsc_sink_max_slice_count() may return 0 if something goes
wrong on the part of the DSC sink and its DPCD register. This null
value may be later used as a divisor in intel_dsc_compute_params(),
which will lead to an error.
In the unlikely event that this issue occurs, fix it by testing the
return value of drm_dp_dsc_sink_max_slice_count() against zero.

Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.

Fixes: a4a157777c80 ("drm/i915/dp: Compute DSC pipe config in atomic check")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418140430.69902-1-n.zhandarovich@fintech.ru
---
 drivers/gpu/drm/i915/display/intel_dp.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 4361c1ac65c3a..0ac0315fe8eaa 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1622,6 +1622,11 @@ int intel_dp_dsc_compute_config(struct intel_dp *intel_dp,
 		pipe_config->dsc.slice_count =
 			drm_dp_dsc_sink_max_slice_count(intel_dp->dsc_dpcd,
 							true);
+		if (!pipe_config->dsc.slice_count) {
+			drm_dbg_kms(&dev_priv->drm, "Unsupported Slice Count %d\n",
+				    pipe_config->dsc.slice_count);
+			return -EINVAL;
+		}
 	} else {
 		u16 dsc_max_output_bpp = 0;
 		u8 dsc_dp_slice_count;
-- 
GitLab


From 9d0e3cac3517942a6e00eeecfe583a98715edb16 Mon Sep 17 00:00:00 2001
From: Brian Norris <briannorris@chromium.org>
Date: Mon, 9 Jan 2023 17:18:16 -0800
Subject: [PATCH 0401/1791] drm/atomic: Allow vblank-enabled + self-refresh
 "disable"

The self-refresh helper framework overloads "disable" to sometimes mean
"go into self-refresh mode," and this mode activates automatically
(e.g., after some period of unchanging display output). In such cases,
the display pipe is still considered "on", and user-space is not aware
that we went into self-refresh mode. Thus, users may expect that
vblank-related features (such as DRM_IOCTL_WAIT_VBLANK) still work
properly.

However, we trigger the WARN_ONCE() here if a CRTC driver tries to leave
vblank enabled.

Add a different expectation: that CRTCs *should* leave vblank enabled
when going into self-refresh.

This patch is preparation for another patch -- "drm/rockchip: vop: Leave
vblank enabled in self-refresh" -- which resolves conflicts between the
above self-refresh behavior and the API tests in IGT's kms_vblank test
module.

== Some alternatives discussed: ==

It's likely that on many display controllers, vblank interrupts will
turn off when the CRTC is disabled, and so in some cases, self-refresh
may not support vblank. To support such cases, we might consider
additions to the generic helpers such that we fire vblank events based
on a timer.

However, there is currently only one driver using the common
self-refresh helpers (i.e., rockchip), and at least as of commit
bed030a49f3e ("drm/rockchip: Don't fully disable vop on self refresh"),
the CRTC hardware is powered enough to continue to generate vblank
interrupts.

So we chose the simpler option of leaving vblank interrupts enabled. We
can reevaluate this decision and perhaps augment the helpers if/when we
gain a second driver that has different requirements.

v3:
 * include discussion summary

v2:
 * add 'ret != 0' warning case for self-refresh
 * describe failing test case and relation to drm/rockchip patch better

Cc: <stable@vger.kernel.org> # dependency for "drm/rockchip: vop: Leave
                             # vblank enabled in self-refresh"
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230109171809.v3.1.I3904f697863649eb1be540ecca147a66e42bfad7@changeid
---
 drivers/gpu/drm/drm_atomic_helper.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 2c2c9caf0be5e..e0ab555aad2cf 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1209,7 +1209,16 @@ disable_outputs(struct drm_device *dev, struct drm_atomic_state *old_state)
 			continue;
 
 		ret = drm_crtc_vblank_get(crtc);
-		WARN_ONCE(ret != -EINVAL, "driver forgot to call drm_crtc_vblank_off()\n");
+		/*
+		 * Self-refresh is not a true "disable"; ensure vblank remains
+		 * enabled.
+		 */
+		if (new_crtc_state->self_refresh_active)
+			WARN_ONCE(ret != 0,
+				  "driver disabled vblank in self-refresh\n");
+		else
+			WARN_ONCE(ret != -EINVAL,
+				  "driver forgot to call drm_crtc_vblank_off()\n");
 		if (ret == 0)
 			drm_crtc_vblank_put(crtc);
 	}
-- 
GitLab


From 2bdba9d4a3baa758c2ca7f5b37b35c7b3391dc42 Mon Sep 17 00:00:00 2001
From: Brian Norris <briannorris@chromium.org>
Date: Mon, 9 Jan 2023 17:18:17 -0800
Subject: [PATCH 0402/1791] drm/rockchip: vop: Leave vblank enabled in
 self-refresh

If we disable vblank when entering self-refresh, vblank APIs (like
DRM_IOCTL_WAIT_VBLANK) no longer work. But user space is not aware when
we enter self-refresh, so this appears to be an API violation -- that
DRM_IOCTL_WAIT_VBLANK fails with EINVAL whenever the display is idle and
enters self-refresh.

The downstream driver used by many of these systems never used to
disable vblank for PSR, and in fact, even upstream, we didn't do that
until radically redesigning the state machine in commit 6c836d965bad
("drm/rockchip: Use the helpers for PSR").

Thus, it seems like a reasonable API fix to simply restore that
behavior, and leave vblank enabled.

Note that this appears to potentially unbalance the
drm_crtc_vblank_{off,on}() calls in some cases, but:
(a) drm_crtc_vblank_on() documents this as OK and
(b) if I do the naive balancing, I find state machine issues such that
    we're not in sync properly; so it's easier to take advantage of (a).

This issue was exposed by IGT's kms_vblank tests, and reported by
KernelCI. The bug has been around a while (longer than KernelCI
noticed), but was only exposed once self-refresh was bugfixed more
recently, and so KernelCI could properly test it. Some other notes in:

  https://lore.kernel.org/dri-devel/Y6OCg9BPnJvimQLT@google.com/
  Re: renesas/master bisection: igt-kms-rockchip.kms_vblank.pipe-A-wait-forked on rk3399-gru-kevin

== Backporting notes: ==

Marking as 'Fixes' commit 6c836d965bad ("drm/rockchip: Use the helpers
for PSR"), but it probably depends on commit bed030a49f3e
("drm/rockchip: Don't fully disable vop on self refresh") as well.

We also need the previous patch ("drm/atomic: Allow vblank-enabled +
self-refresh "disable""), of course.

v3:
 * no update

v2:
 * skip unnecessary lock/unlock

Fixes: 6c836d965bad ("drm/rockchip: Use the helpers for PSR")
Cc: <stable@vger.kernel.org>
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Link: https://lore.kernel.org/dri-devel/Y5itf0+yNIQa6fU4@sirena.org.uk/
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230109171809.v3.2.Ic07cba4ab9a7bd3618a9e4258b8f92ea7d10ae5a@changeid
---
 drivers/gpu/drm/rockchip/rockchip_drm_vop.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index d8f5e064a1baa..a530ecc4d207c 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -714,13 +714,13 @@ static void vop_crtc_atomic_disable(struct drm_crtc *crtc,
 	if (crtc->state->self_refresh_active)
 		rockchip_drm_set_win_enabled(crtc, false);
 
+	if (crtc->state->self_refresh_active)
+		goto out;
+
 	mutex_lock(&vop->vop_lock);
 
 	drm_crtc_vblank_off(crtc);
 
-	if (crtc->state->self_refresh_active)
-		goto out;
-
 	/*
 	 * Vop standby will take effect at end of current frame,
 	 * if dsp hold valid irq happen, it means standby complete.
@@ -754,9 +754,9 @@ static void vop_crtc_atomic_disable(struct drm_crtc *crtc,
 	vop_core_clks_disable(vop);
 	pm_runtime_put(vop->dev);
 
-out:
 	mutex_unlock(&vop->vop_lock);
 
+out:
 	if (crtc->state->event && !crtc->state->active) {
 		spin_lock_irq(&crtc->dev->event_lock);
 		drm_crtc_send_vblank_event(crtc, crtc->state->event);
-- 
GitLab


From a04d27cdafb1caf95e6dc15ac72374c36e38acad Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:00 +0530
Subject: [PATCH 0403/1791] drm/i915/display: Add new member to configure PCON
 color conversion
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The decision to use DFP output format conversion capabilities should be
during compute_config phase.

This patch adds new member to crtc_state to represent the final
output_format to the sink. In case of a DFP this can be different than
the output_format, as per the format conversion done via the PCON.

This will help to store only the format conversion capabilities of the
DP device in intel_dp->dfp, and use crtc_state to compute and store the
configuration for color/format conversion for a given mode.

v2: modified the new member to crtc_state to represent the final
output_format that eaches the sink, after possible conversion by
PCON kind of devices. (Ville)

v3: Addressed comments from Ville:
-Added comments to clarify difference between sink_format and
output_format.
-Corrected the order of setting sink_format and output_format.
-Added readout for sink_format in get_pipe_config hooks.

v4: Set sink_format for intel_sdvo too. (Ville)

v5: Rebased.

v6: Fixed condition to go for YCbCr420 format for dp and hdmi. (Ville)

v7: Fix the condition to set sink_format for HDMI.
Set hdmi output_format simply as sink_format. (Ville)

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> (v3)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-2-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/icl_dsi.c        |  1 +
 drivers/gpu/drm/i915/display/intel_crt.c      |  1 +
 .../drm/i915/display/intel_crtc_state_dump.c  |  5 +--
 drivers/gpu/drm/i915/display/intel_display.c  |  5 +++
 .../drm/i915/display/intel_display_types.h    | 11 ++++++-
 drivers/gpu/drm/i915/display/intel_dp.c       | 33 +++++++++++++------
 drivers/gpu/drm/i915/display/intel_dp_mst.c   |  1 +
 drivers/gpu/drm/i915/display/intel_dvo.c      |  1 +
 drivers/gpu/drm/i915/display/intel_hdmi.c     | 27 ++++++++++-----
 drivers/gpu/drm/i915/display/intel_lvds.c     |  1 +
 drivers/gpu/drm/i915/display/intel_sdvo.c     |  1 +
 drivers/gpu/drm/i915/display/intel_tv.c       |  1 +
 drivers/gpu/drm/i915/display/vlv_dsi.c        |  1 +
 13 files changed, 67 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index c9aeba0ecf911..c133928a0655c 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -1591,6 +1591,7 @@ static int gen11_dsi_compute_config(struct intel_encoder *encoder,
 		&pipe_config->hw.adjusted_mode;
 	int ret;
 
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	ret = intel_panel_compute_config(intel_connector, adjusted_mode);
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 13519f78cf9fe..f0f4897b3c3c6 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -395,6 +395,7 @@ static int intel_crt_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
index 0cdcaa49656f1..27d7bab464274 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
@@ -217,10 +217,11 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 
 	snprintf_output_types(buf, sizeof(buf), pipe_config->output_types);
 	drm_dbg_kms(&i915->drm,
-		    "active: %s, output_types: %s (0x%x), output format: %s\n",
+		    "active: %s, output_types: %s (0x%x), output format: %s, sink format: %s\n",
 		    str_yes_no(pipe_config->hw.active),
 		    buf, pipe_config->output_types,
-		    intel_output_format_name(pipe_config->output_format));
+		    intel_output_format_name(pipe_config->output_format),
+		    intel_output_format_name(pipe_config->sink_format));
 
 	drm_dbg_kms(&i915->drm,
 		    "cpu_transcoder: %s, pipe bpp: %i, dithering: %i\n",
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 07443e283de94..1d5d42a408035 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2889,6 +2889,7 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		return false;
 
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
+	pipe_config->sink_format = pipe_config->output_format;
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = NULL;
 
@@ -3314,6 +3315,8 @@ static bool ilk_get_pipe_config(struct intel_crtc *crtc,
 		break;
 	}
 
+	pipe_config->sink_format = pipe_config->output_format;
+
 	pipe_config->gamma_mode = REG_FIELD_GET(TRANSCONF_GAMMA_MODE_MASK_ILK, tmp);
 
 	pipe_config->framestart_delay = REG_FIELD_GET(TRANSCONF_FRAME_START_DELAY_MASK, tmp) + 1;
@@ -3712,6 +3715,8 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc,
 			bdw_get_pipe_misc_output_format(crtc);
 	}
 
+	pipe_config->sink_format = pipe_config->output_format;
+
 	pipe_config->gamma_mode = intel_de_read(dev_priv,
 						GAMMA_MODE(crtc->pipe));
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 96a3183675bee..e19d7b5d8e328 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1313,9 +1313,18 @@ struct intel_crtc_state {
 	/* HDMI High TMDS char rate ratio */
 	bool hdmi_high_tmds_clock_ratio;
 
-	/* Output format RGB/YCBCR etc */
+	/*
+	 * Output format RGB/YCBCR etc., that is coming out
+	 * at the end of the pipe.
+	 */
 	enum intel_output_format output_format;
 
+	/*
+	 * Sink output format RGB/YCBCR etc., that is going
+	 * into the sink.
+	 */
+	enum intel_output_format sink_format;
+
 	/* enable pipe gamma? */
 	bool gamma_enable;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 0ac0315fe8eaa..88f8b00d775ac 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -851,14 +851,15 @@ u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp,
 
 static enum intel_output_format
 intel_dp_output_format(struct intel_connector *connector,
-		       bool ycbcr_420_output)
+		       enum intel_output_format sink_format)
 {
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 
 	if (intel_dp->force_dsc_output_format)
 		return intel_dp->force_dsc_output_format;
 
-	if (!connector->base.ycbcr_420_allowed || !ycbcr_420_output)
+	if (!connector->base.ycbcr_420_allowed ||
+	    sink_format != INTEL_OUTPUT_FORMAT_YCBCR420)
 		return INTEL_OUTPUT_FORMAT_RGB;
 
 	if (intel_dp->dfp.rgb_to_ycbcr &&
@@ -897,8 +898,14 @@ intel_dp_mode_min_output_bpp(struct intel_connector *connector,
 			     const struct drm_display_mode *mode)
 {
 	const struct drm_display_info *info = &connector->base.display_info;
-	enum intel_output_format output_format =
-		intel_dp_output_format(connector, drm_mode_is_420_only(info, mode));
+	enum intel_output_format output_format, sink_format;
+
+	if (drm_mode_is_420_only(info, mode))
+		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+	else
+		sink_format = INTEL_OUTPUT_FORMAT_RGB;
+
+	output_format = intel_dp_output_format(connector, sink_format);
 
 	return intel_dp_output_bpp(output_format, intel_dp_min_bpp(output_format));
 }
@@ -2118,23 +2125,29 @@ intel_dp_compute_output_format(struct intel_encoder *encoder,
 
 	ycbcr_420_only = drm_mode_is_420_only(info, adjusted_mode);
 
-	crtc_state->output_format = intel_dp_output_format(connector, ycbcr_420_only);
-
-	if (ycbcr_420_only && !intel_dp_is_ycbcr420(intel_dp, crtc_state)) {
+	if (ycbcr_420_only && !connector->base.ycbcr_420_allowed) {
 		drm_dbg_kms(&i915->drm,
 			    "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n");
-		crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
+	} else if (ycbcr_420_only) {
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+	} else {
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	}
 
+	crtc_state->output_format = intel_dp_output_format(connector, crtc_state->sink_format);
+
 	ret = intel_dp_compute_link_config(encoder, crtc_state, conn_state,
 					   respect_downstream_limits);
 	if (ret) {
-		if (intel_dp_is_ycbcr420(intel_dp, crtc_state) ||
+		if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
 		    !connector->base.ycbcr_420_allowed ||
 		    !drm_mode_is_420_also(info, adjusted_mode))
 			return ret;
 
-		crtc_state->output_format = intel_dp_output_format(connector, true);
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+		crtc_state->output_format = intel_dp_output_format(connector,
+								   crtc_state->sink_format);
 		ret = intel_dp_compute_link_config(encoder, crtc_state, conn_state,
 						   respect_downstream_limits);
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 2c49d9ab86a2a..63d61e6102102 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -318,6 +318,7 @@ static int intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->has_pch_encoder = false;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c
index eb2dcd866cc87..9884678743b66 100644
--- a/drivers/gpu/drm/i915/display/intel_dvo.c
+++ b/drivers/gpu/drm/i915/display/intel_dvo.c
@@ -271,6 +271,7 @@ static int intel_dvo_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 6ebdea722c845..dea45c2ecde5e 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -2172,9 +2172,9 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
 }
 
 static enum intel_output_format
-intel_hdmi_output_format(const struct intel_crtc_state *crtc_state,
-			 struct intel_connector *connector,
-			 bool ycbcr_420_output)
+intel_hdmi_sink_format(const struct intel_crtc_state *crtc_state,
+		       struct intel_connector *connector,
+		       bool ycbcr_420_output)
 {
 	if (!crtc_state->has_hdmi_sink)
 		return INTEL_OUTPUT_FORMAT_RGB;
@@ -2185,6 +2185,12 @@ intel_hdmi_output_format(const struct intel_crtc_state *crtc_state,
 		return INTEL_OUTPUT_FORMAT_RGB;
 }
 
+static enum intel_output_format
+intel_hdmi_output_format(const struct intel_crtc_state *crtc_state)
+{
+	return crtc_state->sink_format;
+}
+
 static int intel_hdmi_compute_output_format(struct intel_encoder *encoder,
 					    struct intel_crtc_state *crtc_state,
 					    const struct drm_connector_state *conn_state,
@@ -2197,23 +2203,26 @@ static int intel_hdmi_compute_output_format(struct intel_encoder *encoder,
 	bool ycbcr_420_only = drm_mode_is_420_only(info, adjusted_mode);
 	int ret;
 
-	crtc_state->output_format =
-		intel_hdmi_output_format(crtc_state, connector, ycbcr_420_only);
+	crtc_state->sink_format =
+		intel_hdmi_sink_format(crtc_state, connector, ycbcr_420_only);
 
-	if (ycbcr_420_only && !intel_hdmi_is_ycbcr420(crtc_state)) {
+	if (ycbcr_420_only && crtc_state->sink_format != INTEL_OUTPUT_FORMAT_YCBCR420) {
 		drm_dbg_kms(&i915->drm,
 			    "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n");
-		crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	}
 
+	crtc_state->output_format = intel_hdmi_output_format(crtc_state);
 	ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits);
 	if (ret) {
-		if (intel_hdmi_is_ycbcr420(crtc_state) ||
+		if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
+		    !crtc_state->has_hdmi_sink ||
 		    !connector->base.ycbcr_420_allowed ||
 		    !drm_mode_is_420_also(info, adjusted_mode))
 			return ret;
 
-		crtc_state->output_format = intel_hdmi_output_format(crtc_state, connector, true);
+		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+		crtc_state->output_format = intel_hdmi_output_format(crtc_state);
 		ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits);
 	}
 
diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 8e9a3d72b83b7..1f4349a12a026 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -437,6 +437,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder,
 		crtc_state->pipe_bpp = lvds_bpp;
 	}
 
+	crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	/*
diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index e12ba458636c1..34ee9dd82a785 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -1351,6 +1351,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder,
 
 	DRM_DEBUG_KMS("forcing bpc to 8 for SDVO\n");
 	pipe_config->pipe_bpp = 8*3;
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	if (HAS_PCH_SPLIT(to_i915(encoder->base.dev)))
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index e3ccface0c9d9..96fe4a2800770 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -1206,6 +1206,7 @@ intel_tv_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return -EINVAL;
 
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	drm_dbg_kms(&dev_priv->drm, "forcing bpc to 8 for TV\n");
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index 61d008d4e5f1f..cd90a30e04d84 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -280,6 +280,7 @@ static int intel_dsi_compute_config(struct intel_encoder *encoder,
 	int ret;
 
 	drm_dbg_kms(&dev_priv->drm, "\n");
+	pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB;
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
 
 	ret = intel_panel_compute_config(intel_connector, adjusted_mode);
-- 
GitLab


From 68910c2a903d518b3f7386901cf0d6a053d1c028 Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:01 +0530
Subject: [PATCH 0404/1791] drm/i915/dp: Replace intel_dp.dfp members with the
 new crtc_state sink_format
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The decision to use DFP output format conversion capabilities should be
during compute_config phase.

This patch uses the members of intel_dp->dfp to only store the
format conversion capabilities of the DP device and uses the crtc_state
sink_format member, to program the protocol-converter for
colorspace/format conversion.

v2: Use sink_format to determine the color conversion config for the
pcon (Ville).

v3: Fix typo: missing 'break' in switch case (lkp kernel test robot).

v4: Add helper to check if DP supports YCBCR420.

v5: Simplify logic for computing output_format, based on the given
sink_format. (Ville).
Added scaler constraint for YCbCr420 output.

v6: Split the patch for Scaler constraint for Ycbcr420.

v7: Simplify the policy for selecting output_format:
Always try for RGB first, followed by YCBCR444, and finally by YCBCR420.

v8: Removed redundant comments, minor refactoring. (Ville)

v9: Added member for ycbcr420 passthrough cap, fixed minor issues. (Ville)

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-3-ankit.k.nautiyal@intel.com
---
 .../drm/i915/display/intel_display_types.h    |   1 +
 drivers/gpu/drm/i915/display/intel_dp.c       | 171 ++++++++++++------
 2 files changed, 119 insertions(+), 53 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index e19d7b5d8e328..270c4c84a2920 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1775,6 +1775,7 @@ struct intel_dp {
 		int pcon_max_frl_bw;
 		u8 max_bpc;
 		bool ycbcr_444_to_420;
+		bool ycbcr420_passthrough;
 		bool rgb_to_ycbcr;
 	} dfp;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 88f8b00d775ac..6967ecfe9452b 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -849,27 +849,88 @@ u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp,
 	return 0;
 }
 
+static bool source_can_output(struct intel_dp *intel_dp,
+			      enum intel_output_format format)
+{
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+
+	switch (format) {
+	case INTEL_OUTPUT_FORMAT_RGB:
+		return true;
+
+	case INTEL_OUTPUT_FORMAT_YCBCR444:
+		/*
+		 * No YCbCr output support on gmch platforms.
+		 * Also, ILK doesn't seem capable of DP YCbCr output.
+		 * The displayed image is severly corrupted. SNB+ is fine.
+		 */
+		return !HAS_GMCH(i915) && !IS_IRONLAKE(i915);
+
+	case INTEL_OUTPUT_FORMAT_YCBCR420:
+		/* Platform < Gen 11 cannot output YCbCr420 format */
+		return DISPLAY_VER(i915) >= 11;
+
+	default:
+		MISSING_CASE(format);
+		return false;
+	}
+}
+
+static bool
+dfp_can_convert_from_rgb(struct intel_dp *intel_dp,
+			 enum intel_output_format sink_format)
+{
+	if (!drm_dp_is_branch(intel_dp->dpcd))
+		return false;
+
+	if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR444)
+		return intel_dp->dfp.rgb_to_ycbcr;
+
+	if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420)
+		return intel_dp->dfp.rgb_to_ycbcr &&
+			intel_dp->dfp.ycbcr_444_to_420;
+
+	return false;
+}
+
+static bool
+dfp_can_convert_from_ycbcr444(struct intel_dp *intel_dp,
+			      enum intel_output_format sink_format)
+{
+	if (!drm_dp_is_branch(intel_dp->dpcd))
+		return false;
+
+	if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420)
+		return intel_dp->dfp.ycbcr_444_to_420;
+
+	return false;
+}
+
 static enum intel_output_format
 intel_dp_output_format(struct intel_connector *connector,
 		       enum intel_output_format sink_format)
 {
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+	enum intel_output_format output_format;
 
 	if (intel_dp->force_dsc_output_format)
 		return intel_dp->force_dsc_output_format;
 
-	if (!connector->base.ycbcr_420_allowed ||
-	    sink_format != INTEL_OUTPUT_FORMAT_YCBCR420)
-		return INTEL_OUTPUT_FORMAT_RGB;
+	if (sink_format == INTEL_OUTPUT_FORMAT_RGB ||
+	    dfp_can_convert_from_rgb(intel_dp, sink_format))
+		output_format = INTEL_OUTPUT_FORMAT_RGB;
 
-	if (intel_dp->dfp.rgb_to_ycbcr &&
-	    intel_dp->dfp.ycbcr_444_to_420)
-		return INTEL_OUTPUT_FORMAT_RGB;
+	else if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR444 ||
+		 dfp_can_convert_from_ycbcr444(intel_dp, sink_format))
+		output_format = INTEL_OUTPUT_FORMAT_YCBCR444;
 
-	if (intel_dp->dfp.ycbcr_444_to_420)
-		return INTEL_OUTPUT_FORMAT_YCBCR444;
 	else
-		return INTEL_OUTPUT_FORMAT_YCBCR420;
+		output_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+
+	drm_WARN_ON(&i915->drm, !source_can_output(intel_dp, output_format));
+
+	return output_format;
 }
 
 int intel_dp_min_bpp(enum intel_output_format output_format)
@@ -2829,6 +2890,8 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp,
 					   const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+	bool ycbcr444_to_420 = false;
+	bool rgb_to_ycbcr = false;
 	u8 tmp;
 
 	if (intel_dp->dpcd[DP_DPCD_REV] < 0x13)
@@ -2845,8 +2908,24 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp,
 		drm_dbg_kms(&i915->drm, "Failed to %s protocol converter HDMI mode\n",
 			    str_enable_disable(intel_dp->has_hdmi_sink));
 
-	tmp = crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 &&
-		intel_dp->dfp.ycbcr_444_to_420 ? DP_CONVERSION_TO_YCBCR420_ENABLE : 0;
+	if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) {
+		switch (crtc_state->output_format) {
+		case INTEL_OUTPUT_FORMAT_YCBCR420:
+			break;
+		case INTEL_OUTPUT_FORMAT_YCBCR444:
+			ycbcr444_to_420 = true;
+			break;
+		case INTEL_OUTPUT_FORMAT_RGB:
+			rgb_to_ycbcr = true;
+			ycbcr444_to_420 = true;
+			break;
+		default:
+			MISSING_CASE(crtc_state->output_format);
+			break;
+		}
+	}
+
+	tmp = ycbcr444_to_420 ? DP_CONVERSION_TO_YCBCR420_ENABLE : 0;
 
 	if (drm_dp_dpcd_writeb(&intel_dp->aux,
 			       DP_PROTOCOL_CONVERTER_CONTROL_1, tmp) != 1)
@@ -2854,13 +2933,12 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp,
 			    "Failed to %s protocol converter YCbCr 4:2:0 conversion mode\n",
 			    str_enable_disable(intel_dp->dfp.ycbcr_444_to_420));
 
-	tmp = intel_dp->dfp.rgb_to_ycbcr ?
-		DP_CONVERSION_BT709_RGB_YCBCR_ENABLE : 0;
+	tmp = rgb_to_ycbcr ? DP_CONVERSION_BT709_RGB_YCBCR_ENABLE : 0;
 
 	if (drm_dp_pcon_convert_rgb_to_ycbcr(&intel_dp->aux, tmp) < 0)
 		drm_dbg_kms(&i915->drm,
-			   "Failed to %s protocol converter RGB->YCbCr conversion mode\n",
-			   str_enable_disable(tmp));
+			    "Failed to %s protocol converter RGB->YCbCr conversion mode\n",
+			    str_enable_disable(tmp));
 }
 
 bool intel_dp_get_colorimetry_status(struct intel_dp *intel_dp)
@@ -4650,57 +4728,44 @@ intel_dp_update_dfp(struct intel_dp *intel_dp,
 	intel_dp_get_pcon_dsc_cap(intel_dp);
 }
 
+static bool
+intel_dp_can_ycbcr420(struct intel_dp *intel_dp)
+{
+	if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420) &&
+	    (!drm_dp_is_branch(intel_dp->dpcd) || intel_dp->dfp.ycbcr420_passthrough))
+		return true;
+
+	if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_RGB) &&
+	    dfp_can_convert_from_rgb(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420))
+		return true;
+
+	if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR444) &&
+	    dfp_can_convert_from_ycbcr444(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420))
+		return true;
+
+	return false;
+}
+
 static void
 intel_dp_update_420(struct intel_dp *intel_dp)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	struct intel_connector *connector = intel_dp->attached_connector;
-	bool is_branch, ycbcr_420_passthrough, ycbcr_444_to_420, rgb_to_ycbcr;
-
-	/* No YCbCr output support on gmch platforms */
-	if (HAS_GMCH(i915))
-		return;
 
-	/*
-	 * ILK doesn't seem capable of DP YCbCr output. The
-	 * displayed image is severly corrupted. SNB+ is fine.
-	 */
-	if (IS_IRONLAKE(i915))
-		return;
-
-	is_branch = drm_dp_is_branch(intel_dp->dpcd);
-	ycbcr_420_passthrough =
+	intel_dp->dfp.ycbcr420_passthrough =
 		drm_dp_downstream_420_passthrough(intel_dp->dpcd,
 						  intel_dp->downstream_ports);
 	/* on-board LSPCON always assumed to support 4:4:4->4:2:0 conversion */
-	ycbcr_444_to_420 =
+	intel_dp->dfp.ycbcr_444_to_420 =
 		dp_to_dig_port(intel_dp)->lspcon.active ||
 		drm_dp_downstream_444_to_420_conversion(intel_dp->dpcd,
 							intel_dp->downstream_ports);
-	rgb_to_ycbcr = drm_dp_downstream_rgb_to_ycbcr_conversion(intel_dp->dpcd,
-								 intel_dp->downstream_ports,
-								 DP_DS_HDMI_BT709_RGB_YCBCR_CONV);
-
-	if (DISPLAY_VER(i915) >= 11) {
-		/* Let PCON convert from RGB->YCbCr if possible */
-		if (is_branch && rgb_to_ycbcr && ycbcr_444_to_420) {
-			intel_dp->dfp.rgb_to_ycbcr = true;
-			intel_dp->dfp.ycbcr_444_to_420 = true;
-			connector->base.ycbcr_420_allowed = true;
-		} else {
-		/* Prefer 4:2:0 passthrough over 4:4:4->4:2:0 conversion */
-			intel_dp->dfp.ycbcr_444_to_420 =
-				ycbcr_444_to_420 && !ycbcr_420_passthrough;
+	intel_dp->dfp.rgb_to_ycbcr =
+		drm_dp_downstream_rgb_to_ycbcr_conversion(intel_dp->dpcd,
+							  intel_dp->downstream_ports,
+							  DP_DS_HDMI_BT709_RGB_YCBCR_CONV);
 
-			connector->base.ycbcr_420_allowed =
-				!is_branch || ycbcr_444_to_420 || ycbcr_420_passthrough;
-		}
-	} else {
-		/* 4:4:4->4:2:0 conversion is the only way */
-		intel_dp->dfp.ycbcr_444_to_420 = ycbcr_444_to_420;
-
-		connector->base.ycbcr_420_allowed = ycbcr_444_to_420;
-	}
+	connector->base.ycbcr_420_allowed = intel_dp_can_ycbcr420(intel_dp);
 
 	drm_dbg_kms(&i915->drm,
 		    "[CONNECTOR:%d:%s] RGB->YcbCr conversion? %s, YCbCr 4:2:0 allowed? %s, YCbCr 4:4:4->4:2:0 conversion? %s\n",
-- 
GitLab


From 9cca0fe329599109ed69e3c9583a8c4287feaf9b Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:02 +0530
Subject: [PATCH 0405/1791] drm/i915/dp: Configure PCON for conversion of
 output_format to YCbCr444
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Handle the case with DP to HDMI PCON, where sink_format is set to YCbCr444.
In that case PCON is required to be configured to convert from given
output_format to YCbCR444.

v2: Drop drm_WARN for invalid case, let MISSING_CASE catch it. (Ville)

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-4-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 6967ecfe9452b..23bfa5e9de54e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -2923,6 +2923,17 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp,
 			MISSING_CASE(crtc_state->output_format);
 			break;
 		}
+	} else if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR444) {
+		switch (crtc_state->output_format) {
+		case INTEL_OUTPUT_FORMAT_YCBCR444:
+			break;
+		case INTEL_OUTPUT_FORMAT_RGB:
+			rgb_to_ycbcr = true;
+			break;
+		default:
+			MISSING_CASE(crtc_state->output_format);
+			break;
+		}
 	}
 
 	tmp = ycbcr444_to_420 ? DP_CONVERSION_TO_YCBCR420_ENABLE : 0;
-- 
GitLab


From 523156799441f2625d4acbb0ba0904128e06622e Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:03 +0530
Subject: [PATCH 0406/1791] drm/i915/display: Use sink_format instead of
 ycbcr420_output flag
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Start passing the sink_format, to all functions that take a bool
ycbcr420_output as parameter. This will make the functions generic,
and will serve as a slight step towards 4:2:2 support later.

v2: Rebased.

v3: Correct the checks in places concerned with pipe output. (Ville)
Other minor styling and refactoring fixes, as suggested by Ville.

Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-5-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c   | 32 ++++++++---------
 drivers/gpu/drm/i915/display/intel_hdmi.c | 44 ++++++++++++++---------
 drivers/gpu/drm/i915/display/intel_hdmi.h |  5 +--
 3 files changed, 45 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 23bfa5e9de54e..012d7af72cdf4 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1005,7 +1005,8 @@ static int intel_dp_max_tmds_clock(struct intel_dp *intel_dp)
 
 static enum drm_mode_status
 intel_dp_tmds_clock_valid(struct intel_dp *intel_dp,
-			  int clock, int bpc, bool ycbcr420_output,
+			  int clock, int bpc,
+			  enum intel_output_format sink_format,
 			  bool respect_downstream_limits)
 {
 	int tmds_clock, min_tmds_clock, max_tmds_clock;
@@ -1013,7 +1014,7 @@ intel_dp_tmds_clock_valid(struct intel_dp *intel_dp,
 	if (!respect_downstream_limits)
 		return MODE_OK;
 
-	tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output);
+	tmds_clock = intel_hdmi_tmds_clock(clock, bpc, sink_format);
 
 	min_tmds_clock = intel_dp->dfp.min_tmds_clock;
 	max_tmds_clock = intel_dp_max_tmds_clock(intel_dp);
@@ -1036,6 +1037,7 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector,
 	const struct drm_display_info *info = &connector->base.display_info;
 	enum drm_mode_status status;
 	bool ycbcr_420_only;
+	enum intel_output_format sink_format;
 
 	/* If PCON supports FRL MODE, check FRL bandwidth constraints */
 	if (intel_dp->dfp.pcon_max_frl_bw) {
@@ -1062,18 +1064,23 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector,
 
 	ycbcr_420_only = drm_mode_is_420_only(info, mode);
 
+	if (ycbcr_420_only)
+		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+	else
+		sink_format = INTEL_OUTPUT_FORMAT_RGB;
+
 	/* Assume 8bpc for the DP++/HDMI/DVI TMDS clock check */
 	status = intel_dp_tmds_clock_valid(intel_dp, target_clock,
-					   8, ycbcr_420_only, true);
+					   8, sink_format, true);
 
 	if (status != MODE_OK) {
-		if (ycbcr_420_only ||
+		if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
 		    !connector->base.ycbcr_420_allowed ||
 		    !drm_mode_is_420_also(info, mode))
 			return status;
-
+		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
 		status = intel_dp_tmds_clock_valid(intel_dp, target_clock,
-						   8, true, true);
+						   8, sink_format, true);
 		if (status != MODE_OK)
 			return status;
 	}
@@ -1309,19 +1316,10 @@ static bool intel_dp_supports_dsc(struct intel_dp *intel_dp,
 		drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd);
 }
 
-static bool intel_dp_is_ycbcr420(struct intel_dp *intel_dp,
-				 const struct intel_crtc_state *crtc_state)
-{
-	return crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 ||
-		(crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 &&
-		 intel_dp->dfp.ycbcr_444_to_420);
-}
-
 static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp,
 				     const struct intel_crtc_state *crtc_state,
 				     int bpc, bool respect_downstream_limits)
 {
-	bool ycbcr420_output = intel_dp_is_ycbcr420(intel_dp, crtc_state);
 	int clock = crtc_state->hw.adjusted_mode.crtc_clock;
 
 	/*
@@ -1341,8 +1339,8 @@ static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp,
 
 	for (; bpc >= 8; bpc -= 2) {
 		if (intel_hdmi_bpc_possible(crtc_state, bpc,
-					    intel_dp->has_hdmi_sink, ycbcr420_output) &&
-		    intel_dp_tmds_clock_valid(intel_dp, clock, bpc, ycbcr420_output,
+					    intel_dp->has_hdmi_sink) &&
+		    intel_dp_tmds_clock_valid(intel_dp, clock, bpc, crtc_state->sink_format,
 					      respect_downstream_limits) == MODE_OK)
 			return bpc;
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index dea45c2ecde5e..7d796c099fb68 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -1873,10 +1873,11 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi,
 	return MODE_OK;
 }
 
-int intel_hdmi_tmds_clock(int clock, int bpc, bool ycbcr420_output)
+int intel_hdmi_tmds_clock(int clock, int bpc,
+			  enum intel_output_format sink_format)
 {
 	/* YCBCR420 TMDS rate requirement is half the pixel clock */
-	if (ycbcr420_output)
+	if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420)
 		clock /= 2;
 
 	/*
@@ -1903,7 +1904,8 @@ static bool intel_hdmi_source_bpc_possible(struct drm_i915_private *i915, int bp
 }
 
 static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
-					 int bpc, bool has_hdmi_sink, bool ycbcr420_output)
+					 int bpc, bool has_hdmi_sink,
+					 enum intel_output_format sink_format)
 {
 	const struct drm_display_info *info = &connector->display_info;
 	const struct drm_hdmi_info *hdmi = &info->hdmi;
@@ -1913,7 +1915,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
 		if (!has_hdmi_sink)
 			return false;
 
-		if (ycbcr420_output)
+		if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420)
 			return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_36;
 		else
 			return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_36;
@@ -1921,7 +1923,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
 		if (!has_hdmi_sink)
 			return false;
 
-		if (ycbcr420_output)
+		if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420)
 			return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_30;
 		else
 			return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_30;
@@ -1935,7 +1937,8 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector,
 
 static enum drm_mode_status
 intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock,
-			    bool has_hdmi_sink, bool ycbcr420_output)
+			    bool has_hdmi_sink,
+			    enum intel_output_format sink_format)
 {
 	struct drm_i915_private *i915 = to_i915(connector->dev);
 	struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector));
@@ -1948,12 +1951,12 @@ intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock,
 	 * least one color depth is accepted.
 	 */
 	for (bpc = 12; bpc >= 8; bpc -= 2) {
-		int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output);
+		int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, sink_format);
 
 		if (!intel_hdmi_source_bpc_possible(i915, bpc))
 			continue;
 
-		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, ycbcr420_output))
+		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, sink_format))
 			continue;
 
 		status = hdmi_port_clock_valid(hdmi, tmds_clock, true, has_hdmi_sink);
@@ -1978,6 +1981,7 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 	bool has_hdmi_sink = intel_has_hdmi_sink(hdmi, connector->state);
 	bool ycbcr_420_only;
+	enum intel_output_format sink_format;
 
 	if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING)
 		clock *= 2;
@@ -2002,14 +2006,20 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 
 	ycbcr_420_only = drm_mode_is_420_only(&connector->display_info, mode);
 
-	status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, ycbcr_420_only);
+	if (ycbcr_420_only)
+		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+	else
+		sink_format = INTEL_OUTPUT_FORMAT_RGB;
+
+	status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format);
 	if (status != MODE_OK) {
 		if (ycbcr_420_only ||
 		    !connector->ycbcr_420_allowed ||
 		    !drm_mode_is_420_also(&connector->display_info, mode))
 			return status;
 
-		status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, true);
+		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
+		status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format);
 		if (status != MODE_OK)
 			return status;
 	}
@@ -2018,7 +2028,7 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 }
 
 bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state,
-			     int bpc, bool has_hdmi_sink, bool ycbcr420_output)
+			     int bpc, bool has_hdmi_sink)
 {
 	struct drm_atomic_state *state = crtc_state->uapi.state;
 	struct drm_connector_state *connector_state;
@@ -2029,7 +2039,8 @@ bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state,
 		if (connector_state->crtc != crtc_state->uapi.crtc)
 			continue;
 
-		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, ycbcr420_output))
+		if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink,
+						  crtc_state->sink_format))
 			return false;
 	}
 
@@ -2053,8 +2064,7 @@ static bool hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, int bpc
 	     adjusted_mode->crtc_hblank_start) % 8 == 2)
 		return false;
 
-	return intel_hdmi_bpc_possible(crtc_state, bpc, crtc_state->has_hdmi_sink,
-				       intel_hdmi_is_ycbcr420(crtc_state));
+	return intel_hdmi_bpc_possible(crtc_state, bpc, crtc_state->has_hdmi_sink);
 }
 
 static int intel_hdmi_compute_bpc(struct intel_encoder *encoder,
@@ -2062,7 +2072,6 @@ static int intel_hdmi_compute_bpc(struct intel_encoder *encoder,
 				  int clock, bool respect_downstream_limits)
 {
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
-	bool ycbcr420_output = intel_hdmi_is_ycbcr420(crtc_state);
 	int bpc;
 
 	/*
@@ -2080,7 +2089,8 @@ static int intel_hdmi_compute_bpc(struct intel_encoder *encoder,
 		bpc = 8;
 
 	for (; bpc >= 8; bpc -= 2) {
-		int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output);
+		int tmds_clock = intel_hdmi_tmds_clock(clock, bpc,
+						       crtc_state->sink_format);
 
 		if (hdmi_bpc_possible(crtc_state, bpc) &&
 		    hdmi_port_clock_valid(intel_hdmi, tmds_clock,
@@ -2110,7 +2120,7 @@ static int intel_hdmi_compute_clock(struct intel_encoder *encoder,
 		return bpc;
 
 	crtc_state->port_clock =
-		intel_hdmi_tmds_clock(clock, bpc, intel_hdmi_is_ycbcr420(crtc_state));
+		intel_hdmi_tmds_clock(clock, bpc, crtc_state->sink_format);
 
 	/*
 	 * pipe_bpp could already be below 8bpc due to
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h
index 492bd39213850..08a2745729d00 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.h
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.h
@@ -9,6 +9,7 @@
 #include <linux/types.h>
 
 enum hdmi_infoframe_type;
+enum intel_output_format;
 enum port;
 struct drm_connector;
 struct drm_connector_state;
@@ -45,8 +46,8 @@ void intel_read_infoframe(struct intel_encoder *encoder,
 bool intel_hdmi_limited_color_range(const struct intel_crtc_state *crtc_state,
 				    const struct drm_connector_state *conn_state);
 bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state,
-			     int bpc, bool has_hdmi_sink, bool ycbcr420_output);
-int intel_hdmi_tmds_clock(int clock, int bpc, bool ycbcr420_output);
+			     int bpc, bool has_hdmi_sink);
+int intel_hdmi_tmds_clock(int clock, int bpc, enum intel_output_format sink_format);
 int intel_hdmi_dsc_get_bpp(int src_fractional_bpp, int slice_width,
 			   int num_slices, int output_format, bool hdmi_all_bpp,
 			   int hdmi_max_chunk_bytes);
-- 
GitLab


From 5814227de13333463ace7146d2455ecabcc8e657 Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:04 +0530
Subject: [PATCH 0407/1791] drm/i915/dp: Add helper to get sink_format
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Common function to get the sink format for a given mode for DP.

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-6-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 30 ++++++++++++-------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 012d7af72cdf4..11a7d6cd9d410 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -954,17 +954,25 @@ static int intel_dp_output_bpp(enum intel_output_format output_format, int bpp)
 	return bpp;
 }
 
+static enum intel_output_format
+intel_dp_sink_format(struct intel_connector *connector,
+		     const struct drm_display_mode *mode)
+{
+	const struct drm_display_info *info = &connector->base.display_info;
+
+	if (drm_mode_is_420_only(info, mode))
+		return INTEL_OUTPUT_FORMAT_YCBCR420;
+
+	return INTEL_OUTPUT_FORMAT_RGB;
+}
+
 static int
 intel_dp_mode_min_output_bpp(struct intel_connector *connector,
 			     const struct drm_display_mode *mode)
 {
-	const struct drm_display_info *info = &connector->base.display_info;
 	enum intel_output_format output_format, sink_format;
 
-	if (drm_mode_is_420_only(info, mode))
-		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
-	else
-		sink_format = INTEL_OUTPUT_FORMAT_RGB;
+	sink_format = intel_dp_sink_format(connector, mode);
 
 	output_format = intel_dp_output_format(connector, sink_format);
 
@@ -1036,7 +1044,6 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector,
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	const struct drm_display_info *info = &connector->base.display_info;
 	enum drm_mode_status status;
-	bool ycbcr_420_only;
 	enum intel_output_format sink_format;
 
 	/* If PCON supports FRL MODE, check FRL bandwidth constraints */
@@ -1062,12 +1069,7 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector,
 	    target_clock > intel_dp->dfp.max_dotclock)
 		return MODE_CLOCK_HIGH;
 
-	ycbcr_420_only = drm_mode_is_420_only(info, mode);
-
-	if (ycbcr_420_only)
-		sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
-	else
-		sink_format = INTEL_OUTPUT_FORMAT_RGB;
+	sink_format = intel_dp_sink_format(connector, mode);
 
 	/* Assume 8bpc for the DP++/HDMI/DVI TMDS clock check */
 	status = intel_dp_tmds_clock_valid(intel_dp, target_clock,
@@ -2188,10 +2190,8 @@ intel_dp_compute_output_format(struct intel_encoder *encoder,
 		drm_dbg_kms(&i915->drm,
 			    "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n");
 		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
-	} else if (ycbcr_420_only) {
-		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420;
 	} else {
-		crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB;
+		crtc_state->sink_format = intel_dp_sink_format(connector, adjusted_mode);
 	}
 
 	crtc_state->output_format = intel_dp_output_format(connector, crtc_state->sink_format);
-- 
GitLab


From 1dc565764dc7a4b6477db0bc0202d26f74ad2a0e Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 27 Apr 2023 18:26:05 +0530
Subject: [PATCH 0408/1791] drm/i915/dp: Rearrange check for illegal mode and
 comments in mode_valid
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Check for MODE_H_ILLEGAL before calculating max rates, lanes etc.
Move comments about compressed bpp U6.4 format closer to where it is used.

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427125605.487769-7-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 11a7d6cd9d410..0cc57681dc4d4 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1139,6 +1139,9 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 	if (target_clock > max_dotclk)
 		return MODE_CLOCK_HIGH;
 
+	if (intel_dp_hdisplay_bad(dev_priv, mode->hdisplay))
+		return MODE_H_ILLEGAL;
+
 	max_link_clock = intel_dp_max_link_rate(intel_dp);
 	max_lanes = intel_dp_max_lane_count(intel_dp);
 
@@ -1146,13 +1149,6 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 	mode_rate = intel_dp_link_required(target_clock,
 					   intel_dp_mode_min_output_bpp(connector, mode));
 
-	if (intel_dp_hdisplay_bad(dev_priv, mode->hdisplay))
-		return MODE_H_ILLEGAL;
-
-	/*
-	 * Output bpp is stored in 6.4 format so right shift by 4 to get the
-	 * integer value since we support only integer values of bpp.
-	 */
 	if (HAS_DSC(dev_priv) &&
 	    drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd)) {
 		/*
@@ -1161,6 +1157,10 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 		 */
 		int pipe_bpp = intel_dp_dsc_compute_bpp(intel_dp, U8_MAX);
 
+		/*
+		 * Output bpp is stored in 6.4 format so right shift by 4 to get the
+		 * integer value since we support only integer values of bpp.
+		 */
 		if (intel_dp_is_edp(intel_dp)) {
 			dsc_max_output_bpp =
 				drm_edp_dsc_sink_output_bpp(intel_dp->dsc_dpcd) >> 4;
-- 
GitLab


From d41e14fabc16043664f153b89d3d5d26f7069ff7 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Thu, 4 May 2023 13:22:51 -0700
Subject: [PATCH 0409/1791] drm/i915/uc: Track patch level versions on reduced
 version firmware files

When reduced version firmware files were added (matching major
component being the only strict requirement), the minor version was
still tracked and a notification reported if it was older. However,
the patch version should really be tracked as well for the same
reasons. The KMD can work without the change but if the effort has
been taken to release a new firmware with the change then there must
be a valid reason for doing so - important bug fix, security fix, etc.
And in that case it would be good to alert the user if they are
missing out on that new fix.

v2: Use correct patch version number and drop redunant debug print
(review by Daniele / CI results).

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504202252.1104212-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 30 +++++++++++++++---------
 1 file changed, 19 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index c36e68e23a141..d26eef39f332a 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -79,14 +79,14 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw,
  * security fixes, etc. to be enabled.
  */
 #define INTEL_GUC_FIRMWARE_DEFS(fw_def, guc_maj, guc_mmp) \
-	fw_def(DG2,          0, guc_maj(dg2,  70, 5)) \
-	fw_def(ALDERLAKE_P,  0, guc_maj(adlp, 70, 5)) \
+	fw_def(DG2,          0, guc_maj(dg2,  70, 5, 1)) \
+	fw_def(ALDERLAKE_P,  0, guc_maj(adlp, 70, 5, 1)) \
 	fw_def(ALDERLAKE_P,  0, guc_mmp(adlp, 70, 1, 1)) \
 	fw_def(ALDERLAKE_P,  0, guc_mmp(adlp, 69, 0, 3)) \
-	fw_def(ALDERLAKE_S,  0, guc_maj(tgl,  70, 5)) \
+	fw_def(ALDERLAKE_S,  0, guc_maj(tgl,  70, 5, 1)) \
 	fw_def(ALDERLAKE_S,  0, guc_mmp(tgl,  70, 1, 1)) \
 	fw_def(ALDERLAKE_S,  0, guc_mmp(tgl,  69, 0, 3)) \
-	fw_def(DG1,          0, guc_maj(dg1,  70, 5)) \
+	fw_def(DG1,          0, guc_maj(dg1,  70, 5, 1)) \
 	fw_def(ROCKETLAKE,   0, guc_mmp(tgl,  70, 1, 1)) \
 	fw_def(TIGERLAKE,    0, guc_mmp(tgl,  70, 1, 1)) \
 	fw_def(JASPERLAKE,   0, guc_mmp(ehl,  70, 1, 1)) \
@@ -140,7 +140,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw,
 	__stringify(patch_) ".bin"
 
 /* Minor for internal driver use, not part of file name */
-#define MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_) \
+#define MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_, patch_) \
 	__MAKE_UC_FW_PATH_MAJOR(prefix_, "guc", major_)
 
 #define MAKE_GUC_FW_PATH_MMP(prefix_, major_, minor_, patch_) \
@@ -196,9 +196,9 @@ struct __packed uc_fw_blob {
 	{ UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \
 	  .legacy = true }
 
-#define GUC_FW_BLOB(prefix_, major_, minor_) \
-	UC_FW_BLOB_NEW(major_, minor_, 0, false, \
-		       MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_))
+#define GUC_FW_BLOB(prefix_, major_, minor_, patch_) \
+	UC_FW_BLOB_NEW(major_, minor_, patch_, false, \
+		       MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_, patch_))
 
 #define GUC_FW_BLOB_MMP(prefix_, major_, minor_, patch_) \
 	UC_FW_BLOB_OLD(major_, minor_, patch_, \
@@ -295,6 +295,7 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)
 		uc_fw->file_wanted.path = blob->path;
 		uc_fw->file_wanted.ver.major = blob->major;
 		uc_fw->file_wanted.ver.minor = blob->minor;
+		uc_fw->file_wanted.ver.patch = blob->patch;
 		uc_fw->loaded_via_gsc = blob->loaded_via_gsc;
 		found = true;
 		break;
@@ -793,6 +794,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
 		} else {
 			if (uc_fw->file_selected.ver.minor < uc_fw->file_wanted.ver.minor)
 				old_ver = true;
+			else if ((uc_fw->file_selected.ver.minor == uc_fw->file_wanted.ver.minor) &&
+				 (uc_fw->file_selected.ver.patch < uc_fw->file_wanted.ver.patch))
+				old_ver = true;
 		}
 	}
 
@@ -800,12 +804,16 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
 		/* Preserve the version that was really wanted */
 		memcpy(&uc_fw->file_wanted, &file_ideal, sizeof(uc_fw->file_wanted));
 
-		gt_notice(gt, "%s firmware %s (%d.%d) is recommended, but only %s (%d.%d) was found\n",
+		gt_notice(gt, "%s firmware %s (%d.%d.%d) is recommended, but only %s (%d.%d.%d) was found\n",
 			  intel_uc_fw_type_repr(uc_fw->type),
 			  uc_fw->file_wanted.path,
-			  uc_fw->file_wanted.ver.major, uc_fw->file_wanted.ver.minor,
+			  uc_fw->file_wanted.ver.major,
+			  uc_fw->file_wanted.ver.minor,
+			  uc_fw->file_wanted.ver.patch,
 			  uc_fw->file_selected.path,
-			  uc_fw->file_selected.ver.major, uc_fw->file_selected.ver.minor);
+			  uc_fw->file_selected.ver.major,
+			  uc_fw->file_selected.ver.minor,
+			  uc_fw->file_selected.ver.patch);
 		gt_info(gt, "Consider updating your linux-firmware pkg or downloading from %s\n",
 			INTEL_UC_FIRMWARE_URL);
 	}
-- 
GitLab


From 34682d6006ca31e8a4961104dba15d556db533b4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 7 Nov 2022 21:46:03 +0200
Subject: [PATCH 0410/1791] drm/i915: Move has_hdmi_sink out from
 intel_hdmi_compute_config()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We'll be wanting to compute has_hdmi_sink a bit differently
for some platforms. To that end compute it in the encoder
.compute_config_hook() before we call intel_hdmi_compute_config().
intel_hdmi_compute_has_hdmi_sink() will do the basic lifting
beyond any platform specific stuff.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221107194604.15227-5-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/g4x_hdmi.c   |  3 +++
 drivers/gpu/drm/i915/display/intel_ddi.c  |  3 +++
 drivers/gpu/drm/i915/display/intel_hdmi.c | 14 ++++++++++----
 drivers/gpu/drm/i915/display/intel_hdmi.h |  3 +++
 4 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c
index 448ea26786e0f..8498d2380d159 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.c
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c
@@ -89,6 +89,9 @@ static int g4x_hdmi_compute_config(struct intel_encoder *encoder,
 	if (HAS_PCH_SPLIT(i915))
 		crtc_state->has_pch_encoder = true;
 
+	crtc_state->has_hdmi_sink =
+		intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state);
+
 	return intel_hdmi_compute_config(encoder, crtc_state, conn_state);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 74d6544b7f88f..b1caf521a3486 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4063,6 +4063,9 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder,
 		pipe_config->cpu_transcoder = TRANSCODER_EDP;
 
 	if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_HDMI)) {
+		pipe_config->has_hdmi_sink =
+			intel_hdmi_compute_has_hdmi_sink(encoder, pipe_config, conn_state);
+
 		ret = intel_hdmi_compute_config(encoder, pipe_config, conn_state);
 	} else {
 		ret = intel_dp_compute_config(encoder, pipe_config, conn_state);
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 7d796c099fb68..fd20834523051 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -2264,11 +2264,20 @@ static bool source_supports_scrambling(struct intel_encoder *encoder)
 	return intel_hdmi_source_max_tmds_clock(encoder) > 340000;
 }
 
+bool intel_hdmi_compute_has_hdmi_sink(struct intel_encoder *encoder,
+				      const struct intel_crtc_state *crtc_state,
+				      const struct drm_connector_state *conn_state)
+{
+	struct intel_hdmi *hdmi = enc_to_intel_hdmi(encoder);
+
+	return intel_has_hdmi_sink(hdmi, conn_state) &&
+		!intel_hdmi_is_cloned(crtc_state);
+}
+
 int intel_hdmi_compute_config(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config,
 			      struct drm_connector_state *conn_state)
 {
-	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
 	struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode;
 	struct drm_connector *connector = conn_state->connector;
@@ -2283,9 +2292,6 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 		return -EINVAL;
 
 	pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB;
-	pipe_config->has_hdmi_sink =
-		intel_has_hdmi_sink(intel_hdmi, conn_state) &&
-		!intel_hdmi_is_cloned(pipe_config);
 
 	if (pipe_config->has_hdmi_sink)
 		pipe_config->has_infoframe = true;
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h
index 08a2745729d00..6b39df38d57ab 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.h
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.h
@@ -24,6 +24,9 @@ union hdmi_infoframe;
 
 void intel_hdmi_init_connector(struct intel_digital_port *dig_port,
 			       struct intel_connector *intel_connector);
+bool intel_hdmi_compute_has_hdmi_sink(struct intel_encoder *encoder,
+				      const struct intel_crtc_state *crtc_state,
+				      const struct drm_connector_state *conn_state);
 int intel_hdmi_compute_config(struct intel_encoder *encoder,
 			      struct intel_crtc_state *pipe_config,
 			      struct drm_connector_state *conn_state);
-- 
GitLab


From dafa65d185fa0e946e602b53cb7f9694b7584456 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 7 Nov 2022 21:46:04 +0200
Subject: [PATCH 0411/1791] drm/i915: Pick one HDMI port for infoframe/audio
 transmission on g4x
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On g4x the hardware has only one audio/video DIP block. Thus
only one HDMI port can transmit audio/infoframes at a time.
Currently we pretend that multiple ports can do it at the same
time, but that doesn't actually work for two reasons:
- the state of the single hw block will get clobbered by
  the multiple ports, leading to state checker failures
- the hardware will automagically disable audio/infoframe
  transmission when enabled on multiple ports

To fix this let's allow only one of the ports to transmit
audio/infoframes at a time. We'll just go over all the HDMI
ports and pick the first one that is otherwise capable of
audio/infoframes. Further HDMI ports will be treated as if
they had a DVI sink connected.

In order to compute this consistently we'll also need to
always add all HDMI ports to the atomic state.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221107194604.15227-6-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
---
 drivers/gpu/drm/i915/display/g4x_hdmi.c   | 114 +++++++++++++++++++++-
 drivers/gpu/drm/i915/display/g4x_hdmi.h   |   4 +
 drivers/gpu/drm/i915/display/intel_hdmi.c |  14 ++-
 3 files changed, 129 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c
index 8498d2380d159..5c187e6e0472a 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.c
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c
@@ -7,6 +7,7 @@
 
 #include "g4x_hdmi.h"
 #include "i915_reg.h"
+#include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
@@ -80,17 +81,66 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 	return ret;
 }
 
+static bool connector_is_hdmi(struct drm_connector *connector)
+{
+	struct intel_encoder *encoder =
+		intel_attached_encoder(to_intel_connector(connector));
+
+	return encoder && encoder->type == INTEL_OUTPUT_HDMI;
+}
+
+static bool g4x_compute_has_hdmi_sink(struct intel_atomic_state *state,
+				      struct intel_crtc *this_crtc)
+{
+	const struct drm_connector_state *conn_state;
+	struct drm_connector *connector;
+	int i;
+
+	/*
+	 * On g4x only one HDMI port can transmit infoframes/audio at
+	 * any given time. Select the first suitable port for this duty.
+	 *
+	 * See also g4x_hdmi_connector_atomic_check().
+	 */
+	for_each_new_connector_in_state(&state->base, connector, conn_state, i) {
+		struct intel_encoder *encoder = to_intel_encoder(conn_state->best_encoder);
+		const struct intel_crtc_state *crtc_state;
+		struct intel_crtc *crtc;
+
+		if (!connector_is_hdmi(connector))
+			continue;
+
+		crtc = to_intel_crtc(conn_state->crtc);
+		if (!crtc)
+			continue;
+
+		crtc_state = intel_atomic_get_new_crtc_state(state, crtc);
+
+		if (!intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state))
+			continue;
+
+		return crtc == this_crtc;
+	}
+
+	return false;
+}
+
 static int g4x_hdmi_compute_config(struct intel_encoder *encoder,
 				   struct intel_crtc_state *crtc_state,
 				   struct drm_connector_state *conn_state)
 {
+	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 
 	if (HAS_PCH_SPLIT(i915))
 		crtc_state->has_pch_encoder = true;
 
-	crtc_state->has_hdmi_sink =
-		intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state);
+	if (IS_G4X(i915))
+		crtc_state->has_hdmi_sink = g4x_compute_has_hdmi_sink(state, crtc);
+	else
+		crtc_state->has_hdmi_sink =
+			intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state);
 
 	return intel_hdmi_compute_config(encoder, crtc_state, conn_state);
 }
@@ -549,6 +599,66 @@ intel_hdmi_hotplug(struct intel_encoder *encoder,
 	return state;
 }
 
+int g4x_hdmi_connector_atomic_check(struct drm_connector *connector,
+				    struct drm_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->dev);
+	struct drm_connector_list_iter conn_iter;
+	struct drm_connector *conn;
+	int ret;
+
+	ret = intel_digital_connector_atomic_check(connector, state);
+	if (ret)
+		return ret;
+
+	if (!IS_G4X(i915))
+		return 0;
+
+	if (!intel_connector_needs_modeset(to_intel_atomic_state(state), connector))
+		return 0;
+
+	/*
+	 * On g4x only one HDMI port can transmit infoframes/audio
+	 * at any given time. Make sure all enabled HDMI ports are
+	 * included in the state so that it's possible to select
+	 * one of them for this duty.
+	 *
+	 * See also g4x_compute_has_hdmi_sink().
+	 */
+	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	drm_for_each_connector_iter(conn, &conn_iter) {
+		struct drm_connector_state *conn_state;
+		struct drm_crtc_state *crtc_state;
+		struct drm_crtc *crtc;
+
+		if (!connector_is_hdmi(conn))
+			continue;
+
+		drm_dbg_kms(&i915->drm, "Adding [CONNECTOR:%d:%s]\n",
+			    conn->base.id, conn->name);
+
+		conn_state = drm_atomic_get_connector_state(state, conn);
+		if (IS_ERR(conn_state)) {
+			ret = PTR_ERR(conn_state);
+			break;
+		}
+
+		crtc = conn_state->crtc;
+		if (!crtc)
+			continue;
+
+		crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+		crtc_state->mode_changed = true;
+
+		ret = drm_atomic_add_affected_planes(state, crtc);
+		if (ret)
+			break;
+	}
+	drm_connector_list_iter_end(&conn_iter);
+
+	return ret;
+}
+
 void g4x_hdmi_init(struct drm_i915_private *dev_priv,
 		   i915_reg_t hdmi_reg, enum port port)
 {
diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.h b/drivers/gpu/drm/i915/display/g4x_hdmi.h
index db9a93bc9321e..1e3ea7f3c846f 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.h
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.h
@@ -11,9 +11,13 @@
 #include "i915_reg_defs.h"
 
 enum port;
+struct drm_atomic_state;
+struct drm_connector;
 struct drm_i915_private;
 
 void g4x_hdmi_init(struct drm_i915_private *dev_priv,
 		   i915_reg_t hdmi_reg, enum port port);
+int g4x_hdmi_connector_atomic_check(struct drm_connector *connector,
+				    struct drm_atomic_state *state);
 
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index fd20834523051..7e6b28cfb4db8 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -40,6 +40,7 @@
 #include <drm/drm_edid.h>
 #include <drm/intel_lpe_audio.h>
 
+#include "g4x_hdmi.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "intel_atomic.h"
@@ -2624,10 +2625,21 @@ static const struct drm_connector_funcs intel_hdmi_connector_funcs = {
 	.atomic_duplicate_state = intel_digital_connector_duplicate_state,
 };
 
+static int intel_hdmi_connector_atomic_check(struct drm_connector *connector,
+					     struct drm_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->dev);
+
+	if (HAS_DDI(i915))
+		return intel_digital_connector_atomic_check(connector, state);
+	else
+		return g4x_hdmi_connector_atomic_check(connector, state);
+}
+
 static const struct drm_connector_helper_funcs intel_hdmi_connector_helper_funcs = {
 	.get_modes = intel_hdmi_get_modes,
 	.mode_valid = intel_hdmi_mode_valid,
-	.atomic_check = intel_digital_connector_atomic_check,
+	.atomic_check = intel_hdmi_connector_atomic_check,
 };
 
 static void
-- 
GitLab


From e541022b1ce2fe598b90cd96c88cf6fd2f7550d7 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Thu, 4 May 2023 13:22:52 -0700
Subject: [PATCH 0412/1791] drm/i915/mtl: Define GuC firmware version for MTL

First release of GuC for Meteorlake.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504202252.1104212-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index d26eef39f332a..10e48cbcf494a 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -79,6 +79,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw,
  * security fixes, etc. to be enabled.
  */
 #define INTEL_GUC_FIRMWARE_DEFS(fw_def, guc_maj, guc_mmp) \
+	fw_def(METEORLAKE,   0, guc_maj(mtl,  70, 6, 6)) \
 	fw_def(DG2,          0, guc_maj(dg2,  70, 5, 1)) \
 	fw_def(ALDERLAKE_P,  0, guc_maj(adlp, 70, 5, 1)) \
 	fw_def(ALDERLAKE_P,  0, guc_mmp(adlp, 70, 1, 1)) \
-- 
GitLab


From 75a8aeac2573ab258c53676eba9b3796ea691988 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:26 +0200
Subject: [PATCH 0413/1791] drm/bridge: tc358768: always enable HS video mode

Always enable HS video mode setting the TXMD bit, without this change no
video output is present with DSI sinks that are setting
MIPI_DSI_MODE_LPM flag (tested with LT8912B DSI-HDMI bridge).

Previously the driver was enabling HS mode only when the DSI sink was
not explicitly setting the MIPI_DSI_MODE_LPM, however this is not
correct.

The MIPI_DSI_MODE_LPM is supposed to indicate that the sink is willing
to receive data in low power mode, however clearing the
TC358768_DSI_CONTROL_TXMD bit will make the TC358768 send video in
LP mode that is not the intended behavior.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-2-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index a4ba93a629f0b..2484b5a321376 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -867,8 +867,7 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	val = TC358768_DSI_CONFW_MODE_SET | TC358768_DSI_CONFW_ADDR_DSI_CONTROL;
 	val |= (dsi_dev->lanes - 1) << 1;
 
-	if (!(dsi_dev->mode_flags & MIPI_DSI_MODE_LPM))
-		val |= TC358768_DSI_CONTROL_TXMD;
+	val |= TC358768_DSI_CONTROL_TXMD;
 
 	if (!(mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS))
 		val |= TC358768_DSI_CONTROL_HSCKMD;
-- 
GitLab


From 6a4020b4c63911977aaf8047f904a300d15de739 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:27 +0200
Subject: [PATCH 0414/1791] drm/bridge: tc358768: fix PLL parameters
 computation

According to Toshiba documentation the PLL input clock after the divider
should be not less than 4MHz, fix the PLL parameters computation
accordingly.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-3-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 2484b5a321376..8820a7d6131ab 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -335,13 +335,17 @@ static int tc358768_calc_pll(struct tc358768_priv *priv,
 		u32 fbd;
 
 		for (fbd = 0; fbd < 512; ++fbd) {
-			u32 pll, diff;
+			u32 pll, diff, pll_in;
 
 			pll = (u32)div_u64((u64)refclk * (fbd + 1), divisor);
 
 			if (pll >= max_pll || pll < min_pll)
 				continue;
 
+			pll_in = (u32)div_u64((u64)refclk, prd + 1);
+			if (pll_in < 4000000)
+				continue;
+
 			diff = max(pll, target_pll) - min(pll, target_pll);
 
 			if (diff < best_diff) {
-- 
GitLab


From ffd2e4bbea626d565b9817312b0fcfb382fecb88 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:28 +0200
Subject: [PATCH 0415/1791] drm/bridge: tc358768: fix PLL target frequency

Correctly compute the PLL target frequency, the current formula works
correctly only when the input bus width is 24bit, actually to properly
compute the PLL target frequency what is relevant is the bits-per-pixel
on the DSI link.

No regression expected since the DSI format is currently hard-coded as
MIPI_DSI_FMT_RGB888.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-4-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 8820a7d6131ab..429bd49c53c4e 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -147,6 +147,7 @@ struct tc358768_priv {
 
 	u32 pd_lines; /* number of Parallel Port Input Data Lines */
 	u32 dsi_lanes; /* number of DSI Lanes */
+	u32 dsi_bpp; /* number of Bits Per Pixel over DSI */
 
 	/* Parameters for PLL programming */
 	u32 fbd;	/* PLL feedback divider */
@@ -285,12 +286,12 @@ static void tc358768_hw_disable(struct tc358768_priv *priv)
 
 static u32 tc358768_pll_to_pclk(struct tc358768_priv *priv, u32 pll_clk)
 {
-	return (u32)div_u64((u64)pll_clk * priv->dsi_lanes, priv->pd_lines);
+	return (u32)div_u64((u64)pll_clk * priv->dsi_lanes, priv->dsi_bpp);
 }
 
 static u32 tc358768_pclk_to_pll(struct tc358768_priv *priv, u32 pclk)
 {
-	return (u32)div_u64((u64)pclk * priv->pd_lines, priv->dsi_lanes);
+	return (u32)div_u64((u64)pclk * priv->dsi_bpp, priv->dsi_lanes);
 }
 
 static int tc358768_calc_pll(struct tc358768_priv *priv,
@@ -427,6 +428,7 @@ static int tc358768_dsi_host_attach(struct mipi_dsi_host *host,
 	priv->output.panel = panel;
 
 	priv->dsi_lanes = dev->lanes;
+	priv->dsi_bpp = mipi_dsi_pixel_format_to_bpp(dev->format);
 
 	/* get input ep (port0/endpoint0) */
 	ret = -EINVAL;
@@ -438,7 +440,7 @@ static int tc358768_dsi_host_attach(struct mipi_dsi_host *host,
 	}
 
 	if (ret)
-		priv->pd_lines = mipi_dsi_pixel_format_to_bpp(dev->format);
+		priv->pd_lines = priv->dsi_bpp;
 
 	drm_bridge_add(&priv->bridge);
 
-- 
GitLab


From f9cf811374f42fca31ac34aaf59ee2ae72b89879 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:29 +0200
Subject: [PATCH 0416/1791] drm/bridge: tc358768: fix TCLK_ZEROCNT computation

Correct computation of TCLK_ZEROCNT register.

This register must be set to a value that ensure that
(TCLK-PREPARECNT + TCLK-ZERO) > 300ns

with the actual value of (TCLK-PREPARECNT + TCLK-ZERO) being

(1 to 2) + (TCLK_ZEROCNT + 1)) x HSByteClkCycle + (PHY output delay)

with PHY output delay being about

(2 to 3) x MIPIBitClk cycle in the BitClk conversion.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-5-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 429bd49c53c4e..487bfe33edc88 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -743,10 +743,10 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 
 	/* 38ns < TCLK_PREPARE < 95ns */
 	val = tc358768_ns_to_cnt(65, dsibclk_nsk) - 1;
-	/* TCLK_PREPARE > 300ns */
-	val2 = tc358768_ns_to_cnt(300 + tc358768_to_ns(3 * ui_nsk),
-				  dsibclk_nsk);
-	val |= (val2 - tc358768_to_ns(phy_delay_nsk - dsibclk_nsk)) << 8;
+	/* TCLK_PREPARE + TCLK_ZERO > 300ns */
+	val2 = tc358768_ns_to_cnt(300 - tc358768_to_ns(2 * ui_nsk),
+				  dsibclk_nsk) - 2;
+	val |= val2 << 8;
 	dev_dbg(priv->dev, "TCLK_HEADERCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_TCLK_HEADERCNT, val);
 
-- 
GitLab


From ee18698e212b1659dd0850d7e2ae0f22e16ed3d3 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:30 +0200
Subject: [PATCH 0417/1791] drm/bridge: tc358768: fix TCLK_TRAILCNT computation

Correct computation of TCLK_TRAILCNT register.

The driver does not implement non-continuous clock mode, so the actual
value doesn't make a practical difference yet. However this change also
ensures that the value does not write to reserved registers bits in case
of under/overflow.

This register must be set to a value that ensures that

TCLK-TRAIL > 60ns
 and
TEOT <= (105 ns + 12 x UI)

with the actual value of TCLK-TRAIL being

(TCLK_TRAILCNT + (1 to 2)) xHSByteClkCycle +
 (2 + (1 to 2)) * HSBYTECLKCycle - (PHY output delay)

with PHY output delay being about

(2 to 3) x MIPIBitClk cycle in the BitClk conversion.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-2-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-3-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-4-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-5-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-2-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-3-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-4-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-5-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-2-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-3-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-4-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-5-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-2-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-3-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-4-francesco@dolcini.it
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-5-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 487bfe33edc88..4cb46a3e6be8c 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -10,6 +10,7 @@
 #include <linux/i2c.h>
 #include <linux/kernel.h>
 #include <linux/media-bus-format.h>
+#include <linux/minmax.h>
 #include <linux/module.h>
 #include <linux/regmap.h>
 #include <linux/regulator/consumer.h>
@@ -639,6 +640,7 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	struct mipi_dsi_device *dsi_dev = priv->output.dev;
 	unsigned long mode_flags = dsi_dev->mode_flags;
 	u32 val, val2, lptxcnt, hact, data_type;
+	s32 raw_val;
 	const struct drm_display_mode *mode;
 	u32 dsibclk_nsk, dsiclk_nsk, ui_nsk, phy_delay_nsk;
 	u32 dsiclk, dsibclk, video_start;
@@ -750,9 +752,9 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	dev_dbg(priv->dev, "TCLK_HEADERCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_TCLK_HEADERCNT, val);
 
-	/* TCLK_TRAIL > 60ns + 3*UI */
-	val = 60 + tc358768_to_ns(3 * ui_nsk);
-	val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 5;
+	/* TCLK_TRAIL > 60ns AND TEOT <= 105 ns + 12*UI */
+	raw_val = tc358768_ns_to_cnt(60 + tc358768_to_ns(2 * ui_nsk), dsibclk_nsk) - 5;
+	val = clamp(raw_val, 0, 127);
 	dev_dbg(priv->dev, "TCLK_TRAILCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_TCLK_TRAILCNT, val);
 
-- 
GitLab


From 77a089328da791118af9692543a5eedc79eb5fd4 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:31 +0200
Subject: [PATCH 0418/1791] drm/bridge: tc358768: fix THS_ZEROCNT computation

Correct computation of THS_ZEROCNT register.

This register must be set to a value that ensure that
THS_PREPARE + THS_ZERO > 145ns + 10*UI

with the actual value of (THS_PREPARE + THS_ZERO) being

((1 to 2) + 1 + (TCLK_ZEROCNT + 1) + (3 to 4)) x ByteClk cycle +
  + HSByteClk x (2 + (1 to 2)) + (PHY delay)

with PHY delay being about

(8 + (5 to 6)) x MIPIBitClk cycle in the BitClk conversion.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-7-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 4cb46a3e6be8c..e3f456bfb90c1 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -761,9 +761,10 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	/* 40ns + 4*UI < THS_PREPARE < 85ns + 6*UI */
 	val = 50 + tc358768_to_ns(4 * ui_nsk);
 	val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 1;
-	/* THS_ZERO > 145ns + 10*UI */
-	val2 = tc358768_ns_to_cnt(145 - tc358768_to_ns(ui_nsk), dsibclk_nsk);
-	val |= (val2 - tc358768_to_ns(phy_delay_nsk)) << 8;
+	/* THS_PREPARE + THS_ZERO > 145ns + 10*UI */
+	raw_val = tc358768_ns_to_cnt(145 - tc358768_to_ns(3 * ui_nsk), dsibclk_nsk) - 10;
+	val2 = clamp(raw_val, 0, 127);
+	val |= val2 << 8;
 	dev_dbg(priv->dev, "THS_HEADERCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_THS_HEADERCNT, val);
 
-- 
GitLab


From 3666aad8185af8d0ce164fd3c4974235417d6d0b Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:32 +0200
Subject: [PATCH 0419/1791] drm/bridge: tc358768: fix TXTAGOCNT computation

Correct computation of TXTAGOCNT register.

This register must be set to a value that ensure that the
TTA-GO period = (4 x TLPX)

with the actual value of TTA-GO being

4 x (TXTAGOCNT + 1) x (HSByteClk cycle)

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-8-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index e3f456bfb90c1..d8209433d5f43 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -796,7 +796,7 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 
 	/* TXTAGOCNT[26:16] RXTASURECNT[10:0] */
 	val = tc358768_to_ns((lptxcnt + 1) * dsibclk_nsk * 4);
-	val = tc358768_ns_to_cnt(val, dsibclk_nsk) - 1;
+	val = tc358768_ns_to_cnt(val, dsibclk_nsk) / 4 - 1;
 	val2 = tc358768_ns_to_cnt(tc358768_to_ns((lptxcnt + 1) * dsibclk_nsk),
 				  dsibclk_nsk) - 2;
 	val = val << 16 | val2;
-- 
GitLab


From bac7842cd179572e8e0fc2d7b5254e40c6e9e057 Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:33 +0200
Subject: [PATCH 0420/1791] drm/bridge: tc358768: fix THS_TRAILCNT computation

Correct computation of THS_TRAILCNT register.

This register must be set to a value that ensure that
THS_TRAIL > 60 ns + 4 x UI
 and
THS_TRAIL > 8 x UI
 and
THS_TRAIL < TEOT
 with
TEOT = 105 ns + (12 x UI)

with the actual value of THS_TRAIL being

(1 + THS_TRAILCNT) x ByteClk cycle + ((1 to 2) + 2) xHSBYTECLK cycle +
 - (PHY output delay)

with PHY output delay being about

(8 + (5 to 6)) x MIPIBitClk cycle in the BitClk conversion.

Fixes: ff1ca6397b1d ("drm/bridge: Add tc358768 driver")
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-9-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index d8209433d5f43..966a25cb0b108 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -780,9 +780,10 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	dev_dbg(priv->dev, "TCLK_POSTCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_TCLK_POSTCNT, val);
 
-	/* 60ns + 4*UI < THS_PREPARE < 105ns + 12*UI */
-	val = tc358768_ns_to_cnt(60 + tc358768_to_ns(15 * ui_nsk),
-				 dsibclk_nsk) - 5;
+	/* max(60ns + 4*UI, 8*UI) < THS_TRAILCNT < 105ns + 12*UI */
+	raw_val = tc358768_ns_to_cnt(60 + tc358768_to_ns(18 * ui_nsk),
+				     dsibclk_nsk) - 4;
+	val = clamp(raw_val, 0, 15);
 	dev_dbg(priv->dev, "THS_TRAILCNT: 0x%x\n", val);
 	tc358768_write(priv, TC358768_THS_TRAILCNT, val);
 
-- 
GitLab


From e4a5e4442a8065c6959e045c061de801d545226d Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Thu, 27 Apr 2023 16:29:34 +0200
Subject: [PATCH 0421/1791] drm/bridge: tc358768: remove unused variable

Remove the unused phy_delay_nsk variable, before it was wrongly used
to compute some register value, the fixed computation is no longer using
it and therefore can be removed.

Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230427142934.55435-10-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 966a25cb0b108..03c7e82e41090 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -642,7 +642,7 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 	u32 val, val2, lptxcnt, hact, data_type;
 	s32 raw_val;
 	const struct drm_display_mode *mode;
-	u32 dsibclk_nsk, dsiclk_nsk, ui_nsk, phy_delay_nsk;
+	u32 dsibclk_nsk, dsiclk_nsk, ui_nsk;
 	u32 dsiclk, dsibclk, video_start;
 	const u32 internal_delay = 40;
 	int ret, i;
@@ -726,11 +726,9 @@ static void tc358768_bridge_pre_enable(struct drm_bridge *bridge)
 				  dsibclk);
 	dsiclk_nsk = (u32)div_u64((u64)1000000000 * TC358768_PRECISION, dsiclk);
 	ui_nsk = dsiclk_nsk / 2;
-	phy_delay_nsk = dsibclk_nsk + 2 * dsiclk_nsk;
 	dev_dbg(priv->dev, "dsiclk_nsk: %u\n", dsiclk_nsk);
 	dev_dbg(priv->dev, "ui_nsk: %u\n", ui_nsk);
 	dev_dbg(priv->dev, "dsibclk_nsk: %u\n", dsibclk_nsk);
-	dev_dbg(priv->dev, "phy_delay_nsk: %u\n", phy_delay_nsk);
 
 	/* LP11 > 100us for D-PHY Rx Init */
 	val = tc358768_ns_to_cnt(100 * 1000, dsibclk_nsk) - 1;
-- 
GitLab


From 6b8bfff56b2140396fd28088ad2dca4463aef9c9 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 2 May 2023 16:40:02 -0700
Subject: [PATCH 0422/1791] drm/i915/guc: Decode another GuC load failure case

Explain another potential firmware failure mode and early exit the
long wait if hit.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502234007.1762014-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/abi/guc_errors_abi.h | 1 +
 drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c       | 6 ++++++
 2 files changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_errors_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_errors_abi.h
index bcb1129b36102..dabeaf4f245f3 100644
--- a/drivers/gpu/drm/i915/gt/uc/abi/guc_errors_abi.h
+++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_errors_abi.h
@@ -44,6 +44,7 @@ enum intel_guc_load_status {
 enum intel_bootrom_load_status {
 	INTEL_BOOTROM_STATUS_NO_KEY_FOUND                 = 0x13,
 	INTEL_BOOTROM_STATUS_AES_PROD_KEY_FOUND           = 0x1A,
+	INTEL_BOOTROM_STATUS_PROD_KEY_CHECK_FAILURE       = 0x2B,
 	INTEL_BOOTROM_STATUS_RSA_FAILED                   = 0x50,
 	INTEL_BOOTROM_STATUS_PAVPC_FAILED                 = 0x73,
 	INTEL_BOOTROM_STATUS_WOPCM_FAILED                 = 0x74,
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
index 6fda3aec5c66a..0ff088a5e51a8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
@@ -129,6 +129,7 @@ static inline bool guc_load_done(struct intel_uncore *uncore, u32 *status, bool
 	case INTEL_BOOTROM_STATUS_RC6CTXCONFIG_FAILED:
 	case INTEL_BOOTROM_STATUS_MPUMAP_INCORRECT:
 	case INTEL_BOOTROM_STATUS_EXCEPTION:
+	case INTEL_BOOTROM_STATUS_PROD_KEY_CHECK_FAILURE:
 		*success = false;
 		return true;
 	}
@@ -219,6 +220,11 @@ static int guc_wait_ucode(struct intel_guc *guc)
 			guc_info(guc, "firmware signature verification failed\n");
 			ret = -ENOEXEC;
 			break;
+
+		case INTEL_BOOTROM_STATUS_PROD_KEY_CHECK_FAILURE:
+			guc_info(guc, "firmware production part check failure\n");
+			ret = -ENOEXEC;
+			break;
 		}
 
 		switch (ukernel) {
-- 
GitLab


From c354feb5895fe80fd3f896443015cf53d5cf98e8 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 2 May 2023 16:40:03 -0700
Subject: [PATCH 0423/1791] drm/i915/guc: Print status register when waiting
 for GuC to load

If the GuC load is taking an excessively long time, the wait loop
currently prints the GT frequency. Extend that to include the GuC
status as well so we can see if the GuC is actually making progress or
not.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502234007.1762014-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
index 0ff088a5e51a8..364d0d546ec82 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
@@ -191,8 +191,10 @@ static int guc_wait_ucode(struct intel_guc *guc)
 		if (!ret || !success)
 			break;
 
-		guc_dbg(guc, "load still in progress, count = %d, freq = %dMHz\n",
-			count, intel_rps_read_actual_frequency(&uncore->gt->rps));
+		guc_dbg(guc, "load still in progress, count = %d, freq = %dMHz, status = 0x%08X [0x%02X/%02X]\n",
+			count, intel_rps_read_actual_frequency(&uncore->gt->rps), status,
+			REG_FIELD_GET(GS_BOOTROM_MASK, status),
+			REG_FIELD_GET(GS_UKERNEL_MASK, status));
 	}
 	after = ktime_get();
 	delta = ktime_sub(after, before);
-- 
GitLab


From edba77891b58a1f0626daf8598717b8efb307bc8 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 2 May 2023 16:40:05 -0700
Subject: [PATCH 0424/1791] drm/i915/uc: Enhancements to firmware table
 validation

The validation of the firmware table was being done inside the code
for scanning the table for the next available firmware blob. Which is
unnecessary. So pull it out into a separate function that is only
called once per blob type at init time.

Also, drop the CONFIG_SELFTEST requirement and make errors terminal.
It was mentioned that potential issues with backports would not be
caught by regular pre-merge CI as that only occurs on tip not stable
branches. Making the validation unconditional and failing driver load
on detecting of a problem ensures that such backports will also be
validated correctly.

This requires adding a firmware global flag to indicate an issue with
any of the per firmware tables. This is done rather than adding a new
state enum as a new enum value would be a much more invasive change -
lots of places would need updating to support the new error state.

Note also that this change means that a table error will cause the
driver to wedge even on platforms that don't require firmware files.
This is intentional as per the above backport concern - someone doing
backports is not guaranteed to test on every platform that they may
potential affect. So forcing a failure on all platforms ensures that
the problem will be noticed and corrected immediately.

v2: Change to unconditionally fail module load on a validation error
(review feedback/discussion with Daniele).
v3: Add a new flag to track table validation errors (review
feedback/discussion with Daniele).

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502234007.1762014-5-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc.c    |   3 +
 drivers/gpu/drm/i915/gt/uc/intel_uc.h    |   1 +
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 161 +++++++++++++----------
 3 files changed, 99 insertions(+), 66 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index 996168312340e..1381943b8973d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -432,6 +432,9 @@ static bool uc_is_wopcm_locked(struct intel_uc *uc)
 
 static int __uc_check_hw(struct intel_uc *uc)
 {
+	if (uc->fw_table_invalid)
+		return -EIO;
+
 	if (!intel_uc_supports_guc(uc))
 		return 0;
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_uc.h
index 5d0f1bcc381e8..d585524d94deb 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.h
@@ -36,6 +36,7 @@ struct intel_uc {
 	struct drm_i915_gem_object *load_err_log;
 
 	bool reset_in_progress;
+	bool fw_table_invalid;
 };
 
 void intel_uc_init_early(struct intel_uc *uc);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 10e48cbcf494a..b9ab370ce9848 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -233,20 +233,22 @@ struct fw_blobs_by_type {
 	u32 count;
 };
 
+static const struct uc_fw_platform_requirement blobs_guc[] = {
+	INTEL_GUC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB, GUC_FW_BLOB_MMP)
+};
+
+static const struct uc_fw_platform_requirement blobs_huc[] = {
+	INTEL_HUC_FIRMWARE_DEFS(MAKE_FW_LIST, HUC_FW_BLOB, HUC_FW_BLOB_MMP, HUC_FW_BLOB_GSC)
+};
+
+static const struct fw_blobs_by_type blobs_all[INTEL_UC_FW_NUM_TYPES] = {
+	[INTEL_UC_FW_TYPE_GUC] = { blobs_guc, ARRAY_SIZE(blobs_guc) },
+	[INTEL_UC_FW_TYPE_HUC] = { blobs_huc, ARRAY_SIZE(blobs_huc) },
+};
+
 static void
 __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)
 {
-	static const struct uc_fw_platform_requirement blobs_guc[] = {
-		INTEL_GUC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB, GUC_FW_BLOB_MMP)
-	};
-	static const struct uc_fw_platform_requirement blobs_huc[] = {
-		INTEL_HUC_FIRMWARE_DEFS(MAKE_FW_LIST, HUC_FW_BLOB, HUC_FW_BLOB_MMP, HUC_FW_BLOB_GSC)
-	};
-	static const struct fw_blobs_by_type blobs_all[INTEL_UC_FW_NUM_TYPES] = {
-		[INTEL_UC_FW_TYPE_GUC] = { blobs_guc, ARRAY_SIZE(blobs_guc) },
-		[INTEL_UC_FW_TYPE_HUC] = { blobs_huc, ARRAY_SIZE(blobs_huc) },
-	};
-	static bool verified[INTEL_UC_FW_NUM_TYPES];
 	const struct uc_fw_platform_requirement *fw_blobs;
 	enum intel_platform p = INTEL_INFO(i915)->platform;
 	u32 fw_count;
@@ -286,6 +288,11 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)
 			continue;
 
 		if (uc_fw->file_selected.path) {
+			/*
+			 * Continuing an earlier search after a found blob failed to load.
+			 * Once the previously chosen path has been found, clear it out
+			 * and let the search continue from there.
+			 */
 			if (uc_fw->file_selected.path == blob->path)
 				uc_fw->file_selected.path = NULL;
 
@@ -306,76 +313,91 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)
 		/* Failed to find a match for the last attempt?! */
 		uc_fw->file_selected.path = NULL;
 	}
+}
 
-	/* make sure the list is ordered as expected */
-	if (IS_ENABLED(CONFIG_DRM_I915_SELFTEST) && !verified[uc_fw->type]) {
-		verified[uc_fw->type] = true;
+static bool validate_fw_table_type(struct drm_i915_private *i915, enum intel_uc_fw_type type)
+{
+	const struct uc_fw_platform_requirement *fw_blobs;
+	u32 fw_count;
+	int i;
 
-		for (i = 1; i < fw_count; i++) {
-			/* Next platform is good: */
-			if (fw_blobs[i].p < fw_blobs[i - 1].p)
-				continue;
+	if (type >= ARRAY_SIZE(blobs_all)) {
+		drm_err(&i915->drm, "No blob array for %s\n", intel_uc_fw_type_repr(type));
+		return false;
+	}
 
-			/* Next platform revision is good: */
-			if (fw_blobs[i].p == fw_blobs[i - 1].p &&
-			    fw_blobs[i].rev < fw_blobs[i - 1].rev)
-				continue;
+	fw_blobs = blobs_all[type].blobs;
+	fw_count = blobs_all[type].count;
 
-			/* Platform/revision must be in order: */
-			if (fw_blobs[i].p != fw_blobs[i - 1].p ||
-			    fw_blobs[i].rev != fw_blobs[i - 1].rev)
-				goto bad;
+	if (!fw_count)
+		return true;
 
-			/* Next major version is good: */
-			if (fw_blobs[i].blob.major < fw_blobs[i - 1].blob.major)
-				continue;
+	/* make sure the list is ordered as expected */
+	for (i = 1; i < fw_count; i++) {
+		/* Next platform is good: */
+		if (fw_blobs[i].p < fw_blobs[i - 1].p)
+			continue;
 
-			/* New must be before legacy: */
-			if (!fw_blobs[i].blob.legacy && fw_blobs[i - 1].blob.legacy)
-				goto bad;
+		/* Next platform revision is good: */
+		if (fw_blobs[i].p == fw_blobs[i - 1].p &&
+		    fw_blobs[i].rev < fw_blobs[i - 1].rev)
+			continue;
 
-			/* New to legacy also means 0.0 to X.Y (HuC), or X.0 to X.Y (GuC) */
-			if (fw_blobs[i].blob.legacy && !fw_blobs[i - 1].blob.legacy) {
-				if (!fw_blobs[i - 1].blob.major)
-					continue;
+		/* Platform/revision must be in order: */
+		if (fw_blobs[i].p != fw_blobs[i - 1].p ||
+		    fw_blobs[i].rev != fw_blobs[i - 1].rev)
+			goto bad;
 
-				if (fw_blobs[i].blob.major == fw_blobs[i - 1].blob.major)
-					continue;
-			}
+		/* Next major version is good: */
+		if (fw_blobs[i].blob.major < fw_blobs[i - 1].blob.major)
+			continue;
 
-			/* Major versions must be in order: */
-			if (fw_blobs[i].blob.major != fw_blobs[i - 1].blob.major)
-				goto bad;
+		/* New must be before legacy: */
+		if (!fw_blobs[i].blob.legacy && fw_blobs[i - 1].blob.legacy)
+			goto bad;
 
-			/* Next minor version is good: */
-			if (fw_blobs[i].blob.minor < fw_blobs[i - 1].blob.minor)
+		/* New to legacy also means 0.0 to X.Y (HuC), or X.0 to X.Y (GuC) */
+		if (fw_blobs[i].blob.legacy && !fw_blobs[i - 1].blob.legacy) {
+			if (!fw_blobs[i - 1].blob.major)
 				continue;
 
-			/* Minor versions must be in order: */
-			if (fw_blobs[i].blob.minor != fw_blobs[i - 1].blob.minor)
-				goto bad;
-
-			/* Patch versions must be in order: */
-			if (fw_blobs[i].blob.patch <= fw_blobs[i - 1].blob.patch)
+			if (fw_blobs[i].blob.major == fw_blobs[i - 1].blob.major)
 				continue;
+		}
+
+		/* Major versions must be in order: */
+		if (fw_blobs[i].blob.major != fw_blobs[i - 1].blob.major)
+			goto bad;
+
+		/* Next minor version is good: */
+		if (fw_blobs[i].blob.minor < fw_blobs[i - 1].blob.minor)
+			continue;
+
+		/* Minor versions must be in order: */
+		if (fw_blobs[i].blob.minor != fw_blobs[i - 1].blob.minor)
+			goto bad;
+
+		/* Patch versions must be in order: */
+		if (fw_blobs[i].blob.patch <= fw_blobs[i - 1].blob.patch)
+			continue;
 
 bad:
-			drm_err(&i915->drm, "Invalid %s blob order: %s r%u %s%d.%d.%d comes before %s r%u %s%d.%d.%d\n",
-				intel_uc_fw_type_repr(uc_fw->type),
-				intel_platform_name(fw_blobs[i - 1].p), fw_blobs[i - 1].rev,
-				fw_blobs[i - 1].blob.legacy ? "L" : "v",
-				fw_blobs[i - 1].blob.major,
-				fw_blobs[i - 1].blob.minor,
-				fw_blobs[i - 1].blob.patch,
-				intel_platform_name(fw_blobs[i].p), fw_blobs[i].rev,
-				fw_blobs[i].blob.legacy ? "L" : "v",
-				fw_blobs[i].blob.major,
-				fw_blobs[i].blob.minor,
-				fw_blobs[i].blob.patch);
-
-			uc_fw->file_selected.path = NULL;
-		}
+		drm_err(&i915->drm, "Invalid %s blob order: %s r%u %s%d.%d.%d comes before %s r%u %s%d.%d.%d\n",
+			intel_uc_fw_type_repr(type),
+			intel_platform_name(fw_blobs[i - 1].p), fw_blobs[i - 1].rev,
+			fw_blobs[i - 1].blob.legacy ? "L" : "v",
+			fw_blobs[i - 1].blob.major,
+			fw_blobs[i - 1].blob.minor,
+			fw_blobs[i - 1].blob.patch,
+			intel_platform_name(fw_blobs[i].p), fw_blobs[i].rev,
+			fw_blobs[i].blob.legacy ? "L" : "v",
+			fw_blobs[i].blob.major,
+			fw_blobs[i].blob.minor,
+			fw_blobs[i].blob.patch);
+		return false;
 	}
+
+	return true;
 }
 
 static const char *__override_guc_firmware_path(struct drm_i915_private *i915)
@@ -430,7 +452,8 @@ static void __uc_fw_user_override(struct drm_i915_private *i915, struct intel_uc
 void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw,
 			    enum intel_uc_fw_type type)
 {
-	struct drm_i915_private *i915 = ____uc_fw_to_gt(uc_fw, type)->i915;
+	struct intel_gt *gt = ____uc_fw_to_gt(uc_fw, type);
+	struct drm_i915_private *i915 = gt->i915;
 
 	/*
 	 * we use FIRMWARE_UNINITIALIZED to detect checks against uc_fw->status
@@ -443,6 +466,12 @@ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw,
 	uc_fw->type = type;
 
 	if (HAS_GT_UC(i915)) {
+		if (!validate_fw_table_type(i915, type)) {
+			gt->uc.fw_table_invalid = true;
+			intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_NOT_SUPPORTED);
+			return;
+		}
+
 		__uc_fw_auto_select(i915, uc_fw);
 		__uc_fw_user_override(i915, uc_fw);
 	}
-- 
GitLab


From bef550c952a71df6a6e86b11bdac5b10dac29163 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 2 May 2023 16:40:06 -0700
Subject: [PATCH 0425/1791] drm/i915/uc: Reject duplicate entries in firmware
 table

It was noticed that duplicate entries in the firmware table could cause
an infinite loop in the firmware loading code if that entry failed to
load. Duplicate entries are a bug anyway and so should never happen.
Ensure they don't by tweaking the table validation code to reject
duplicates.

For full m/m/p files, that can be done by simply tweaking the patch
level check to reject matching values. For reduced version entries,
the filename itself must be compared.

v2: Improve comment (review by Daniele)

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502234007.1762014-6-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 26 +++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index b9ab370ce9848..63e0049411d29 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -319,7 +319,7 @@ static bool validate_fw_table_type(struct drm_i915_private *i915, enum intel_uc_
 {
 	const struct uc_fw_platform_requirement *fw_blobs;
 	u32 fw_count;
-	int i;
+	int i, j;
 
 	if (type >= ARRAY_SIZE(blobs_all)) {
 		drm_err(&i915->drm, "No blob array for %s\n", intel_uc_fw_type_repr(type));
@@ -334,6 +334,26 @@ static bool validate_fw_table_type(struct drm_i915_private *i915, enum intel_uc_
 
 	/* make sure the list is ordered as expected */
 	for (i = 1; i < fw_count; i++) {
+		/* Versionless file names must be unique per platform: */
+		for (j = i + 1; j < fw_count; j++) {
+			/* Same platform? */
+			if (fw_blobs[i].p != fw_blobs[j].p)
+				continue;
+
+			if (fw_blobs[i].blob.path != fw_blobs[j].blob.path)
+				continue;
+
+			drm_err(&i915->drm, "Duplicate %s blobs: %s r%u %s%d.%d.%d [%s] matches %s%d.%d.%d [%s]\n",
+				intel_uc_fw_type_repr(type),
+				intel_platform_name(fw_blobs[j].p), fw_blobs[j].rev,
+				fw_blobs[j].blob.legacy ? "L" : "v",
+				fw_blobs[j].blob.major, fw_blobs[j].blob.minor,
+				fw_blobs[j].blob.patch, fw_blobs[j].blob.path,
+				fw_blobs[i].blob.legacy ? "L" : "v",
+				fw_blobs[i].blob.major, fw_blobs[i].blob.minor,
+				fw_blobs[i].blob.patch, fw_blobs[i].blob.path);
+		}
+
 		/* Next platform is good: */
 		if (fw_blobs[i].p < fw_blobs[i - 1].p)
 			continue;
@@ -377,8 +397,8 @@ static bool validate_fw_table_type(struct drm_i915_private *i915, enum intel_uc_
 		if (fw_blobs[i].blob.minor != fw_blobs[i - 1].blob.minor)
 			goto bad;
 
-		/* Patch versions must be in order: */
-		if (fw_blobs[i].blob.patch <= fw_blobs[i - 1].blob.patch)
+		/* Patch versions must be in order and unique: */
+		if (fw_blobs[i].blob.patch < fw_blobs[i - 1].blob.patch)
 			continue;
 
 bad:
-- 
GitLab


From 760133d42f0adc92b8408ad5544c98f8aefbc75f Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 2 May 2023 16:40:07 -0700
Subject: [PATCH 0426/1791] drm/i915/uc: Make unexpected firmware versions an
 error in debug builds

If the DEBUG_GEM config option is set then escalate the 'unexpected
firmware version' message from a notice to an error. This will ensure
that the CI system treats such occurences as a failure and logs a bug
about it (or fails the pre-merge testing).

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230502234007.1762014-7-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 34 ++++++++++++++----------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 63e0049411d29..4ec7df9ed5ff3 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -17,6 +17,12 @@
 #include "i915_drv.h"
 #include "i915_reg.h"
 
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+#define UNEXPECTED	gt_err
+#else
+#define UNEXPECTED	gt_notice
+#endif
+
 static inline struct intel_gt *
 ____uc_fw_to_gt(struct intel_uc_fw *uc_fw, enum intel_uc_fw_type type)
 {
@@ -833,10 +839,10 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
 	if (uc_fw->file_wanted.ver.major && uc_fw->file_selected.ver.major) {
 		/* Check the file's major version was as it claimed */
 		if (uc_fw->file_selected.ver.major != uc_fw->file_wanted.ver.major) {
-			gt_notice(gt, "%s firmware %s: unexpected version: %u.%u != %u.%u\n",
-				  intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
-				  uc_fw->file_selected.ver.major, uc_fw->file_selected.ver.minor,
-				  uc_fw->file_wanted.ver.major, uc_fw->file_wanted.ver.minor);
+			UNEXPECTED(gt, "%s firmware %s: unexpected version: %u.%u != %u.%u\n",
+				   intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
+				   uc_fw->file_selected.ver.major, uc_fw->file_selected.ver.minor,
+				   uc_fw->file_wanted.ver.major, uc_fw->file_wanted.ver.minor);
 			if (!intel_uc_fw_is_overridden(uc_fw)) {
 				err = -ENOEXEC;
 				goto fail;
@@ -854,16 +860,16 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw)
 		/* Preserve the version that was really wanted */
 		memcpy(&uc_fw->file_wanted, &file_ideal, sizeof(uc_fw->file_wanted));
 
-		gt_notice(gt, "%s firmware %s (%d.%d.%d) is recommended, but only %s (%d.%d.%d) was found\n",
-			  intel_uc_fw_type_repr(uc_fw->type),
-			  uc_fw->file_wanted.path,
-			  uc_fw->file_wanted.ver.major,
-			  uc_fw->file_wanted.ver.minor,
-			  uc_fw->file_wanted.ver.patch,
-			  uc_fw->file_selected.path,
-			  uc_fw->file_selected.ver.major,
-			  uc_fw->file_selected.ver.minor,
-			  uc_fw->file_selected.ver.patch);
+		UNEXPECTED(gt, "%s firmware %s (%d.%d.%d) is recommended, but only %s (%d.%d.%d) was found\n",
+			   intel_uc_fw_type_repr(uc_fw->type),
+			   uc_fw->file_wanted.path,
+			   uc_fw->file_wanted.ver.major,
+			   uc_fw->file_wanted.ver.minor,
+			   uc_fw->file_wanted.ver.patch,
+			   uc_fw->file_selected.path,
+			   uc_fw->file_selected.ver.major,
+			   uc_fw->file_selected.ver.minor,
+			   uc_fw->file_selected.ver.patch);
 		gt_info(gt, "Consider updating your linux-firmware pkg or downloading from %s\n",
 			INTEL_UC_FIRMWARE_URL);
 	}
-- 
GitLab


From b18f1ecaaf4aa8704a02de1accf8c94fa06f1ceb Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 4 Apr 2023 00:12:32 +0200
Subject: [PATCH 0427/1791] drm/bridge: lt9211: Do not generate HFP/HBP/HSA and
 EOT packet

Do not generate the HS front and back porch gaps, the HSA gap and
EOT packet, as these packets are not required. This makes the bridge
work with Samsung DSIM on i.MX8MM and i.MX8MP.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403221233.500485-1-marex@denx.de
---
 drivers/gpu/drm/bridge/lontium-lt9211.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/lontium-lt9211.c b/drivers/gpu/drm/bridge/lontium-lt9211.c
index 3e19fff6547a2..00db681512385 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9211.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9211.c
@@ -709,7 +709,9 @@ static int lt9211_host_attach(struct lt9211 *ctx)
 	dsi->lanes = dsi_lanes;
 	dsi->format = MIPI_DSI_FMT_RGB888;
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
-			  MIPI_DSI_MODE_VIDEO_HSE;
+			  MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO_NO_HSA |
+			  MIPI_DSI_MODE_VIDEO_NO_HFP | MIPI_DSI_MODE_VIDEO_NO_HBP |
+			  MIPI_DSI_MODE_NO_EOT_PACKET;
 
 	ret = devm_mipi_dsi_attach(dev, dsi);
 	if (ret < 0) {
-- 
GitLab


From 8ddce13ae696bf40397eac7ba2e9003113f94030 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 4 Apr 2023 00:12:33 +0200
Subject: [PATCH 0428/1791] drm/bridge: lt9611: Do not generate HFP/HBP/HSA and
 EOT packet

Do not generate the HS front and back porch gaps, the HSA gap and
EOT packet, as these packets are not required. This makes the bridge
work with Samsung DSIM on i.MX8MM and i.MX8MP.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230403221233.500485-2-marex@denx.de
---
 drivers/gpu/drm/bridge/lontium-lt9611.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c
index a25d21a7d5c19..151efe92711c4 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9611.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9611.c
@@ -774,7 +774,9 @@ static struct mipi_dsi_device *lt9611_attach_dsi(struct lt9611 *lt9611,
 	dsi->lanes = 4;
 	dsi->format = MIPI_DSI_FMT_RGB888;
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
-			  MIPI_DSI_MODE_VIDEO_HSE;
+			  MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO_NO_HSA |
+			  MIPI_DSI_MODE_VIDEO_NO_HFP | MIPI_DSI_MODE_VIDEO_NO_HBP |
+			  MIPI_DSI_MODE_NO_EOT_PACKET;
 
 	ret = devm_mipi_dsi_attach(dev, dsi);
 	if (ret < 0) {
-- 
GitLab


From b8e392245105b50706f18418054821e71e637288 Mon Sep 17 00:00:00 2001
From: Luc Ma <luc@sietium.com>
Date: Mon, 8 May 2023 08:09:16 +0800
Subject: [PATCH 0429/1791] drm/vram-helper: fix function names in vram helper
 doc

Refer to drmm_vram_helper_init() instead of the non-existent
drmm_vram_helper_alloc_mm().

Fixes: a5f23a72355d ("drm/vram-helper: Managed vram helpers")
Signed-off-by: Luc Ma <luc@sietium.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/64583db2.630a0220.eb75d.8f51@mx.google.com
---
 drivers/gpu/drm/drm_gem_vram_helper.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c
index 0bea3df2a16dc..b67eafa557159 100644
--- a/drivers/gpu/drm/drm_gem_vram_helper.c
+++ b/drivers/gpu/drm/drm_gem_vram_helper.c
@@ -45,7 +45,7 @@ static const struct drm_gem_object_funcs drm_gem_vram_object_funcs;
  * the frame's scanout buffer or the cursor image. If there's no more space
  * left in VRAM, inactive GEM objects can be moved to system memory.
  *
- * To initialize the VRAM helper library call drmm_vram_helper_alloc_mm().
+ * To initialize the VRAM helper library call drmm_vram_helper_init().
  * The function allocates and initializes an instance of &struct drm_vram_mm
  * in &struct drm_device.vram_mm . Use &DRM_GEM_VRAM_DRIVER to initialize
  * &struct drm_driver and  &DRM_VRAM_MM_FILE_OPERATIONS to initialize
@@ -73,7 +73,7 @@ static const struct drm_gem_object_funcs drm_gem_vram_object_funcs;
  *		// setup device, vram base and size
  *		// ...
  *
- *		ret = drmm_vram_helper_alloc_mm(dev, vram_base, vram_size);
+ *		ret = drmm_vram_helper_init(dev, vram_base, vram_size);
  *		if (ret)
  *			return ret;
  *		return 0;
@@ -86,7 +86,7 @@ static const struct drm_gem_object_funcs drm_gem_vram_object_funcs;
  * to userspace.
  *
  * You don't have to clean up the instance of VRAM MM.
- * drmm_vram_helper_alloc_mm() is a managed interface that installs a
+ * drmm_vram_helper_init() is a managed interface that installs a
  * clean-up handler to run during the DRM device's release.
  *
  * For drawing or scanout operations, rsp. buffer objects have to be pinned
-- 
GitLab


From 00b9dd061eb31c8619878f187e509eabbbdb681a Mon Sep 17 00:00:00 2001
From: Pallavi Mishra <pallavi.mishra@intel.com>
Date: Fri, 5 May 2023 16:40:04 +0200
Subject: [PATCH 0430/1791] drm/i915/mtl: Drop FLAT CCS check

Remove FLAT CCS check from XY_FAST_COLOR_BLT usage, thus
enabling MTL to use it.

Signed-off-by: Pallavi Mishra <pallavi.mishra@intel.com>
Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505144005.23480-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/gt/intel_migrate.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c
index 3f638f1987968..e0998879a0e1f 100644
--- a/drivers/gpu/drm/i915/gt/intel_migrate.c
+++ b/drivers/gpu/drm/i915/gt/intel_migrate.c
@@ -920,7 +920,7 @@ static int emit_clear(struct i915_request *rq, u32 offset, int size,
 
 	GEM_BUG_ON(size >> PAGE_SHIFT > S16_MAX);
 
-	if (HAS_FLAT_CCS(i915) && ver >= 12)
+	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50))
 		ring_sz = XY_FAST_COLOR_BLT_DW;
 	else if (ver >= 8)
 		ring_sz = 8;
@@ -931,7 +931,7 @@ static int emit_clear(struct i915_request *rq, u32 offset, int size,
 	if (IS_ERR(cs))
 		return PTR_ERR(cs);
 
-	if (HAS_FLAT_CCS(i915) && ver >= 12) {
+	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50)) {
 		*cs++ = XY_FAST_COLOR_BLT_CMD | XY_FAST_COLOR_BLT_DEPTH_32 |
 			(XY_FAST_COLOR_BLT_DW - 2);
 		*cs++ = FIELD_PREP(XY_FAST_COLOR_BLT_MOCS_MASK, mocs) |
-- 
GitLab


From 0da6bfe857ea9399498876cbe6ef428637b6e475 Mon Sep 17 00:00:00 2001
From: Clint Taylor <clinton.a.taylor@intel.com>
Date: Fri, 5 May 2023 16:40:05 +0200
Subject: [PATCH 0431/1791] drm/i915/mtl: Add MTL for remapping CCS FBs

Add support for remapping CCS FBs on MTL to remove the restriction
of the power-of-two sized stride and the 2MB surface offset alignment
for these FBs.

Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505144005.23480-2-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/display/intel_fb.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index e5f637897b5e4..c004f08fcfe15 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -1190,7 +1190,8 @@ bool intel_fb_needs_pot_stride_remap(const struct intel_framebuffer *fb)
 {
 	struct drm_i915_private *i915 = to_i915(fb->base.dev);
 
-	return IS_ALDERLAKE_P(i915) && intel_fb_uses_dpt(&fb->base);
+	return (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14) &&
+		intel_fb_uses_dpt(&fb->base);
 }
 
 static int intel_fb_pitch(const struct intel_framebuffer *fb, int color_plane, unsigned int rotation)
@@ -1326,9 +1327,10 @@ plane_view_scanout_stride(const struct intel_framebuffer *fb, int color_plane,
 			  unsigned int tile_width,
 			  unsigned int src_stride_tiles, unsigned int dst_stride_tiles)
 {
+	struct drm_i915_private *i915 = to_i915(fb->base.dev);
 	unsigned int stride_tiles;
 
-	if (IS_ALDERLAKE_P(to_i915(fb->base.dev)))
+	if (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14)
 		stride_tiles = src_stride_tiles;
 	else
 		stride_tiles = dst_stride_tiles;
@@ -1522,7 +1524,8 @@ static void intel_fb_view_init(struct drm_i915_private *i915, struct intel_fb_vi
 	memset(view, 0, sizeof(*view));
 	view->gtt.type = view_type;
 
-	if (view_type == I915_GTT_VIEW_REMAPPED && IS_ALDERLAKE_P(i915))
+	if (view_type == I915_GTT_VIEW_REMAPPED &&
+	    (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14))
 		view->gtt.remapped.plane_alignment = SZ_2M / PAGE_SIZE;
 }
 
-- 
GitLab


From b9279e9b28bb8754495f66d70902fa35c8fa7f53 Mon Sep 17 00:00:00 2001
From: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Date: Sat, 6 May 2023 17:42:17 +0300
Subject: [PATCH 0432/1791] drm/i915: Fix wrong condition in bxt_set_cdclk for
 DG2

By my own mistake, after adding !IS_DG2 into wrong branch,
bxt_set_cdclk started to execute code intended for platforms
gen < 11, which is wrong.
Move IS_DG2 check to better place.

Fixes: ceb0cc3b4288 ("drm/i915: Communicate display power demands to pcode")
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230506144217.26075-1-stanislav.lisovskiy@intel.com
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index a11092deaba67..6bed75f1541a7 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -1896,9 +1896,9 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv,
 	 * mailbox communication, skip
 	 * this step.
 	 */
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER(dev_priv) >= 14 || IS_DG2(dev_priv))
 		/* NOOP */;
-	else if (DISPLAY_VER(dev_priv) >= 11 && !IS_DG2(dev_priv))
+	else if (DISPLAY_VER(dev_priv) >= 11)
 		ret = skl_pcode_request(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL,
 					SKL_CDCLK_PREPARE_FOR_CHANGE,
 					SKL_CDCLK_READY_FOR_CHANGE,
-- 
GitLab


From 4a9820343579f65635500d9f9b268ec6d07496e0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:22 -0300
Subject: [PATCH 0433/1791] drm/vkms: add rotate-0 and reflect-x property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, vkms doesn't support any reflection property. Therefore, add
the reflect-x property to vkms through a software implementation of the
operation. This is possible by reverse reading the x axis during the
blending.

Tested with igt@kms_rotation_crc@primary-reflect-x [1] and
igt@kms_rotation_crc@sprite-reflect-x [1].

[1] https://patchwork.freedesktop.org/series/116025/

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-3-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_composer.c |  2 +-
 drivers/gpu/drm/vkms/vkms_drv.h      |  2 ++
 drivers/gpu/drm/vkms/vkms_formats.c  | 16 +++++++++++++---
 drivers/gpu/drm/vkms/vkms_plane.c    | 12 ++++++++++++
 4 files changed, 28 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index 80164e79af006..a4436981cbcde 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -55,7 +55,7 @@ static void pre_mul_alpha_blend(struct vkms_frame_info *frame_info,
 
 static bool check_y_limit(struct vkms_frame_info *frame_info, int y)
 {
-	if (y >= frame_info->dst.y1 && y < frame_info->dst.y2)
+	if (y >= frame_info->rotated.y1 && y < frame_info->rotated.y2)
 		return true;
 
 	return false;
diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h
index f152d54baf769..5f1a0a44a78cf 100644
--- a/drivers/gpu/drm/vkms/vkms_drv.h
+++ b/drivers/gpu/drm/vkms/vkms_drv.h
@@ -26,7 +26,9 @@
 struct vkms_frame_info {
 	struct drm_framebuffer *fb;
 	struct drm_rect src, dst;
+	struct drm_rect rotated;
 	struct iosys_map map[DRM_FORMAT_MAX_PLANES];
+	unsigned int rotation;
 	unsigned int offset;
 	unsigned int pitch;
 	unsigned int cpp;
diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index 8d948c73741ef..f59b1c48a563a 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -37,11 +37,18 @@ static void *packed_pixels_addr(const struct vkms_frame_info *frame_info,
 static void *get_packed_src_addr(const struct vkms_frame_info *frame_info, int y)
 {
 	int x_src = frame_info->src.x1 >> 16;
-	int y_src = y - frame_info->dst.y1 + (frame_info->src.y1 >> 16);
+	int y_src = y - frame_info->rotated.y1 + (frame_info->src.y1 >> 16);
 
 	return packed_pixels_addr(frame_info, x_src, y_src);
 }
 
+static int get_x_position(const struct vkms_frame_info *frame_info, int limit, int x)
+{
+	if (frame_info->rotation & DRM_MODE_REFLECT_X)
+		return limit - x - 1;
+	return x;
+}
+
 static void ARGB8888_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 {
 	/*
@@ -109,8 +116,11 @@ void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state
 	u8 *src_pixels = get_packed_src_addr(frame_info, y);
 	int limit = min_t(size_t, drm_rect_width(&frame_info->dst), stage_buffer->n_pixels);
 
-	for (size_t x = 0; x < limit; x++, src_pixels += frame_info->cpp)
-		plane->pixel_read(src_pixels, &out_pixels[x]);
+	for (size_t x = 0; x < limit; x++, src_pixels += frame_info->cpp) {
+		int x_pos = get_x_position(frame_info, limit, x);
+
+		plane->pixel_read(src_pixels, &out_pixels[x_pos]);
+	}
 }
 
 /*
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 87b2bccaf0839..d5392e3daf922 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -4,6 +4,7 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_blend.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
@@ -111,9 +112,16 @@ static void vkms_plane_atomic_update(struct drm_plane *plane,
 	frame_info = vkms_plane_state->frame_info;
 	memcpy(&frame_info->src, &new_state->src, sizeof(struct drm_rect));
 	memcpy(&frame_info->dst, &new_state->dst, sizeof(struct drm_rect));
+	memcpy(&frame_info->rotated, &new_state->dst, sizeof(struct drm_rect));
 	frame_info->fb = fb;
 	memcpy(&frame_info->map, &shadow_plane_state->data, sizeof(frame_info->map));
 	drm_framebuffer_get(frame_info->fb);
+	frame_info->rotation = drm_rotation_simplify(new_state->rotation, DRM_MODE_ROTATE_0 |
+						     DRM_MODE_REFLECT_X);
+
+	drm_rect_rotate(&frame_info->rotated, drm_rect_width(&frame_info->rotated),
+			drm_rect_height(&frame_info->rotated), frame_info->rotation);
+
 	frame_info->offset = fb->offsets[0];
 	frame_info->pitch = fb->pitches[0];
 	frame_info->cpp = fb->format->cpp[0];
@@ -201,5 +209,9 @@ struct vkms_plane *vkms_plane_init(struct vkms_device *vkmsdev,
 
 	drm_plane_helper_add(&plane->base, &vkms_plane_helper_funcs);
 
+	drm_plane_create_rotation_property(&plane->base, DRM_MODE_ROTATE_0,
+					   DRM_MODE_ROTATE_0 |
+					   DRM_MODE_REFLECT_X);
+
 	return plane;
 }
-- 
GitLab


From 1ce76fae2bc14dd776b0c237456d8bd117c0e617 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:23 -0300
Subject: [PATCH 0434/1791] drm/vkms: add reflect-y and rotate-180 property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, vkms only supports the reflect-x property. Therefore, add the
reflect-y property to vkms through a software implementation of the
operation. This is possible by reverse reading the y axis during the
blending.

Note that, by implementing the reflect-x and reflect-y properties, it is
also possible to add the rotate-180 property, as it is a combination
of those two properties.

Tested with igt@kms_rotation_crc@primary-reflect-y [1],
igt@kms_rotation_crc@sprite-reflect-y [1],
igt@kms_rotation_crc@primary-rotation-180,
igt@kms_rotation_crc@sprite-rotation-180,
and igt@kms_rotation_crc@cursor-rotation-180.

[1] https://patchwork.freedesktop.org/series/116025/

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-4-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_composer.c | 16 +++++++++++++---
 drivers/gpu/drm/vkms/vkms_plane.c    |  7 +++++--
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index a4436981cbcde..6c5ef11b39430 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -53,6 +53,13 @@ static void pre_mul_alpha_blend(struct vkms_frame_info *frame_info,
 	}
 }
 
+static int get_y_pos(struct vkms_frame_info *frame_info, int y)
+{
+	if (frame_info->rotation & DRM_MODE_REFLECT_Y)
+		return drm_rect_height(&frame_info->rotated) - y - 1;
+	return y;
+}
+
 static bool check_y_limit(struct vkms_frame_info *frame_info, int y)
 {
 	if (y >= frame_info->rotated.y1 && y < frame_info->rotated.y2)
@@ -86,6 +93,7 @@ static void blend(struct vkms_writeback_job *wb,
 {
 	struct vkms_plane_state **plane = crtc_state->active_planes;
 	u32 n_active_planes = crtc_state->num_active_planes;
+	int y_pos;
 
 	const struct pixel_argb_u16 background_color = { .a = 0xffff };
 
@@ -96,10 +104,12 @@ static void blend(struct vkms_writeback_job *wb,
 
 		/* The active planes are composed associatively in z-order. */
 		for (size_t i = 0; i < n_active_planes; i++) {
-			if (!check_y_limit(plane[i]->frame_info, y))
+			y_pos = get_y_pos(plane[i]->frame_info, y);
+
+			if (!check_y_limit(plane[i]->frame_info, y_pos))
 				continue;
 
-			vkms_compose_row(stage_buffer, plane[i], y);
+			vkms_compose_row(stage_buffer, plane[i], y_pos);
 			pre_mul_alpha_blend(plane[i]->frame_info, stage_buffer,
 					    output_buffer);
 		}
@@ -107,7 +117,7 @@ static void blend(struct vkms_writeback_job *wb,
 		*crc32 = crc32_le(*crc32, (void *)output_buffer->pixels, row_size);
 
 		if (wb)
-			wb->wb_write(&wb->wb_frame_info, output_buffer, y);
+			wb->wb_write(&wb->wb_frame_info, output_buffer, y_pos);
 	}
 }
 
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index d5392e3daf922..074b9c067d066 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -117,7 +117,8 @@ static void vkms_plane_atomic_update(struct drm_plane *plane,
 	memcpy(&frame_info->map, &shadow_plane_state->data, sizeof(frame_info->map));
 	drm_framebuffer_get(frame_info->fb);
 	frame_info->rotation = drm_rotation_simplify(new_state->rotation, DRM_MODE_ROTATE_0 |
-						     DRM_MODE_REFLECT_X);
+						     DRM_MODE_REFLECT_X |
+						     DRM_MODE_REFLECT_Y);
 
 	drm_rect_rotate(&frame_info->rotated, drm_rect_width(&frame_info->rotated),
 			drm_rect_height(&frame_info->rotated), frame_info->rotation);
@@ -211,7 +212,9 @@ struct vkms_plane *vkms_plane_init(struct vkms_device *vkmsdev,
 
 	drm_plane_create_rotation_property(&plane->base, DRM_MODE_ROTATE_0,
 					   DRM_MODE_ROTATE_0 |
-					   DRM_MODE_REFLECT_X);
+					   DRM_MODE_ROTATE_180 |
+					   DRM_MODE_REFLECT_X |
+					   DRM_MODE_REFLECT_Y);
 
 	return plane;
 }
-- 
GitLab


From cf7f8c671c830302bd382ac640591ee3d6381ad3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:24 -0300
Subject: [PATCH 0435/1791] drm/vkms: add rotate-90 property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, vkms only supports the rotate-180, reflect-x and reflect-y
properties. Therefore, improve the vkms IGT test coverage by adding the
rotate-90 property to vkms. The rotation was implement by software: rotate
the way the blending occurs by making the source x axis be the destination
y axis and the source y axis be the destination x axis.

Tested with igt@kms_rotation_crc@primary-rotation-90 [1],
igt@kms_rotation_crc@sprite-rotation-90 [1], and
igt@kms_rotation_crc@sprite-rotation-90-pos-100-0 [1].

[1] https://patchwork.freedesktop.org/series/116025/

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-5-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_composer.c | 21 ++++++++++++++++-----
 drivers/gpu/drm/vkms/vkms_formats.c  |  4 ++++
 drivers/gpu/drm/vkms/vkms_plane.c    |  2 ++
 3 files changed, 22 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index 6c5ef11b39430..491850ffeac9a 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -57,13 +57,24 @@ static int get_y_pos(struct vkms_frame_info *frame_info, int y)
 {
 	if (frame_info->rotation & DRM_MODE_REFLECT_Y)
 		return drm_rect_height(&frame_info->rotated) - y - 1;
-	return y;
+
+	switch (frame_info->rotation & DRM_MODE_ROTATE_MASK) {
+	case DRM_MODE_ROTATE_90:
+		return frame_info->rotated.x2 - y - 1;
+	default:
+		return y;
+	}
 }
 
-static bool check_y_limit(struct vkms_frame_info *frame_info, int y)
+static bool check_limit(struct vkms_frame_info *frame_info, int pos)
 {
-	if (y >= frame_info->rotated.y1 && y < frame_info->rotated.y2)
-		return true;
+	if (frame_info->rotation & DRM_MODE_ROTATE_90) {
+		if (pos >= 0 && pos < drm_rect_width(&frame_info->rotated))
+			return true;
+	} else {
+		if (pos >= frame_info->rotated.y1 && pos < frame_info->rotated.y2)
+			return true;
+	}
 
 	return false;
 }
@@ -106,7 +117,7 @@ static void blend(struct vkms_writeback_job *wb,
 		for (size_t i = 0; i < n_active_planes; i++) {
 			y_pos = get_y_pos(plane[i]->frame_info, y);
 
-			if (!check_y_limit(plane[i]->frame_info, y_pos))
+			if (!check_limit(plane[i]->frame_info, y_pos))
 				continue;
 
 			vkms_compose_row(stage_buffer, plane[i], y_pos);
diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index f59b1c48a563a..f56f359e0d2d7 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -119,6 +119,10 @@ void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state
 	for (size_t x = 0; x < limit; x++, src_pixels += frame_info->cpp) {
 		int x_pos = get_x_position(frame_info, limit, x);
 
+		if (frame_info->rotation & DRM_MODE_ROTATE_90)
+			src_pixels = get_packed_src_addr(frame_info, x + frame_info->rotated.y1)
+				+ frame_info->cpp * y;
+
 		plane->pixel_read(src_pixels, &out_pixels[x_pos]);
 	}
 }
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 074b9c067d066..1186c00de391b 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -117,6 +117,7 @@ static void vkms_plane_atomic_update(struct drm_plane *plane,
 	memcpy(&frame_info->map, &shadow_plane_state->data, sizeof(frame_info->map));
 	drm_framebuffer_get(frame_info->fb);
 	frame_info->rotation = drm_rotation_simplify(new_state->rotation, DRM_MODE_ROTATE_0 |
+						     DRM_MODE_ROTATE_90 |
 						     DRM_MODE_REFLECT_X |
 						     DRM_MODE_REFLECT_Y);
 
@@ -212,6 +213,7 @@ struct vkms_plane *vkms_plane_init(struct vkms_device *vkmsdev,
 
 	drm_plane_create_rotation_property(&plane->base, DRM_MODE_ROTATE_0,
 					   DRM_MODE_ROTATE_0 |
+					   DRM_MODE_ROTATE_90 |
 					   DRM_MODE_ROTATE_180 |
 					   DRM_MODE_REFLECT_X |
 					   DRM_MODE_REFLECT_Y);
-- 
GitLab


From cd0755508d81736d026bc25647a89dfc79162d95 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:25 -0300
Subject: [PATCH 0436/1791] drm/vkms: add rotate-270 property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, vkms supports the rotate-90, rotate-180, reflect-x and
reflect-y properties. Therefore, improve the vkms IGT test coverage by
adding the rotate-270 property to vkms. The rotation was implement by
software: rotate the way the blending occurs by making the source x axis
be the destination y axis and the source y axis be the destination x
axis and reverse-read the axis.

Now, vkms supports all possible rotation values.

Tested with igt@kms_rotation_crc@primary-rotation-270 [1],
and igt@kms_rotation_crc@sprite-rotation-270 [1].

[1] https://patchwork.freedesktop.org/series/116025/

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-6-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_composer.c | 5 ++++-
 drivers/gpu/drm/vkms/vkms_formats.c  | 6 ++++--
 drivers/gpu/drm/vkms/vkms_plane.c    | 7 ++-----
 3 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c
index 491850ffeac9a..906d3df40cdbe 100644
--- a/drivers/gpu/drm/vkms/vkms_composer.c
+++ b/drivers/gpu/drm/vkms/vkms_composer.c
@@ -4,6 +4,7 @@
 
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_blend.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_vblank.h>
@@ -61,6 +62,8 @@ static int get_y_pos(struct vkms_frame_info *frame_info, int y)
 	switch (frame_info->rotation & DRM_MODE_ROTATE_MASK) {
 	case DRM_MODE_ROTATE_90:
 		return frame_info->rotated.x2 - y - 1;
+	case DRM_MODE_ROTATE_270:
+		return y + frame_info->rotated.x1;
 	default:
 		return y;
 	}
@@ -68,7 +71,7 @@ static int get_y_pos(struct vkms_frame_info *frame_info, int y)
 
 static bool check_limit(struct vkms_frame_info *frame_info, int pos)
 {
-	if (frame_info->rotation & DRM_MODE_ROTATE_90) {
+	if (drm_rotation_90_or_270(frame_info->rotation)) {
 		if (pos >= 0 && pos < drm_rect_width(&frame_info->rotated))
 			return true;
 	} else {
diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index f56f359e0d2d7..ebacb8efa0553 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -2,6 +2,8 @@
 
 #include <linux/kernel.h>
 #include <linux/minmax.h>
+
+#include <drm/drm_blend.h>
 #include <drm/drm_rect.h>
 #include <drm/drm_fixed.h>
 
@@ -44,7 +46,7 @@ static void *get_packed_src_addr(const struct vkms_frame_info *frame_info, int y
 
 static int get_x_position(const struct vkms_frame_info *frame_info, int limit, int x)
 {
-	if (frame_info->rotation & DRM_MODE_REFLECT_X)
+	if (frame_info->rotation & (DRM_MODE_REFLECT_X | DRM_MODE_ROTATE_270))
 		return limit - x - 1;
 	return x;
 }
@@ -119,7 +121,7 @@ void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state
 	for (size_t x = 0; x < limit; x++, src_pixels += frame_info->cpp) {
 		int x_pos = get_x_position(frame_info, limit, x);
 
-		if (frame_info->rotation & DRM_MODE_ROTATE_90)
+		if (drm_rotation_90_or_270(frame_info->rotation))
 			src_pixels = get_packed_src_addr(frame_info, x + frame_info->rotated.y1)
 				+ frame_info->cpp * y;
 
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 1186c00de391b..e5c625ab8e3e0 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -118,6 +118,7 @@ static void vkms_plane_atomic_update(struct drm_plane *plane,
 	drm_framebuffer_get(frame_info->fb);
 	frame_info->rotation = drm_rotation_simplify(new_state->rotation, DRM_MODE_ROTATE_0 |
 						     DRM_MODE_ROTATE_90 |
+						     DRM_MODE_ROTATE_270 |
 						     DRM_MODE_REFLECT_X |
 						     DRM_MODE_REFLECT_Y);
 
@@ -212,11 +213,7 @@ struct vkms_plane *vkms_plane_init(struct vkms_device *vkmsdev,
 	drm_plane_helper_add(&plane->base, &vkms_plane_helper_funcs);
 
 	drm_plane_create_rotation_property(&plane->base, DRM_MODE_ROTATE_0,
-					   DRM_MODE_ROTATE_0 |
-					   DRM_MODE_ROTATE_90 |
-					   DRM_MODE_ROTATE_180 |
-					   DRM_MODE_REFLECT_X |
-					   DRM_MODE_REFLECT_Y);
+					   DRM_MODE_ROTATE_MASK | DRM_MODE_REFLECT_MASK);
 
 	return plane;
 }
-- 
GitLab


From 5aaa12873d9b610a052bafea5d4c8593ad66e390 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 18 Apr 2023 10:05:26 -0300
Subject: [PATCH 0437/1791] drm/vkms: drop "Rotation" TODO
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that VKMS supports all values of rotation and reflection, drop the
"Rotation" task from the TODO list.

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418130525.128733-7-mcanal@igalia.com
---
 Documentation/gpu/vkms.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/gpu/vkms.rst b/Documentation/gpu/vkms.rst
index 0f599c8976148..ba04ac7c2167a 100644
--- a/Documentation/gpu/vkms.rst
+++ b/Documentation/gpu/vkms.rst
@@ -120,7 +120,7 @@ There's lots of plane features we could add support for:
 
 - Add background color KMS property[Good to get started].
 
-- Rotation, scaling.
+- Scaling.
 
 - Additional buffer formats, especially YUV formats for video like NV12.
   Low/high bpp RGB formats would also be interesting.
-- 
GitLab


From 11f373c3c1ef04faadcdf6abfcc403b65a20412e Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:34 +0200
Subject: [PATCH 0438/1791] auxdisplay/cfag12864bfb: Use struct
 fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-2-tzimmermann@suse.de
---
 drivers/auxdisplay/cfag12864bfb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/auxdisplay/cfag12864bfb.c b/drivers/auxdisplay/cfag12864bfb.c
index 0df474506fb92..c2cab7e2b1268 100644
--- a/drivers/auxdisplay/cfag12864bfb.c
+++ b/drivers/auxdisplay/cfag12864bfb.c
@@ -72,7 +72,7 @@ static int cfag12864bfb_probe(struct platform_device *device)
 	if (!info)
 		goto none;
 
-	info->screen_base = (char __iomem *) cfag12864b_buffer;
+	info->screen_buffer = cfag12864b_buffer;
 	info->screen_size = CFAG12864B_SIZE;
 	info->fbops = &cfag12864bfb_ops;
 	info->fix = cfag12864bfb_fix;
-- 
GitLab


From 029cd092f23fb1b372adf8697be1846a1df0492d Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:35 +0200
Subject: [PATCH 0439/1791] auxdisplay/ht16k33: Use struct
 fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-3-tzimmermann@suse.de
---
 drivers/auxdisplay/ht16k33.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/auxdisplay/ht16k33.c b/drivers/auxdisplay/ht16k33.c
index 02425991c1590..edaf92b7ea770 100644
--- a/drivers/auxdisplay/ht16k33.c
+++ b/drivers/auxdisplay/ht16k33.c
@@ -640,7 +640,7 @@ static int ht16k33_fbdev_probe(struct device *dev, struct ht16k33_priv *priv,
 
 	INIT_DELAYED_WORK(&priv->work, ht16k33_fb_update);
 	fbdev->info->fbops = &ht16k33_fb_ops;
-	fbdev->info->screen_base = (char __iomem *) fbdev->buffer;
+	fbdev->info->screen_buffer = fbdev->buffer;
 	fbdev->info->screen_size = HT16K33_FB_SIZE;
 	fbdev->info->fix = ht16k33_fb_fix;
 	fbdev->info->var = ht16k33_fb_var;
-- 
GitLab


From cb18dfa930acaced019bfb8b9f1b5be0b66ded96 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:36 +0200
Subject: [PATCH 0440/1791] hid/hid-picolcd_fb: Use struct
 fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-4-tzimmermann@suse.de
---
 drivers/hid/hid-picolcd_fb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/hid/hid-picolcd_fb.c b/drivers/hid/hid-picolcd_fb.c
index de61c08fabea9..dabcd054dad94 100644
--- a/drivers/hid/hid-picolcd_fb.c
+++ b/drivers/hid/hid-picolcd_fb.c
@@ -221,7 +221,7 @@ int picolcd_fb_reset(struct picolcd_data *data, int clear)
 	return 0;
 }
 
-/* Update fb_vbitmap from the screen_base and send changed tiles to device */
+/* Update fb_vbitmap from the screen_buffer and send changed tiles to device */
 static void picolcd_fb_update(struct fb_info *info)
 {
 	int chip, tile, n;
@@ -541,7 +541,7 @@ int picolcd_init_framebuffer(struct picolcd_data *data)
 		dev_err(dev, "can't get a free page for framebuffer\n");
 		goto err_nomem;
 	}
-	info->screen_base = (char __force __iomem *)fbdata->bitmap;
+	info->screen_buffer = fbdata->bitmap;
 	info->fix.smem_start = (unsigned long)fbdata->bitmap;
 	memset(fbdata->vbitmap, 0xff, PICOLCDFB_SIZE);
 	data->fb_info = info;
-- 
GitLab


From 8112bafc3403196efff5b8b63a62a52ae1bd3e4c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:37 +0200
Subject: [PATCH 0441/1791] fbdev/arcfb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-5-tzimmermann@suse.de
---
 drivers/video/fbdev/arcfb.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/video/fbdev/arcfb.c b/drivers/video/fbdev/arcfb.c
index 45e64016db328..088c4b30fd314 100644
--- a/drivers/video/fbdev/arcfb.c
+++ b/drivers/video/fbdev/arcfb.c
@@ -260,7 +260,7 @@ static void arcfb_lcd_update_page(struct arcfb_par *par, unsigned int upper,
 	ks108_set_yaddr(par, chipindex, upper/8);
 
 	linesize = par->info->var.xres/8;
-	src = (unsigned char __force *) par->info->screen_base + (left/8) +
+	src = (unsigned char *)par->info->screen_buffer + (left/8) +
 		(upper * linesize);
 	ks108_set_xaddr(par, chipindex, left);
 
@@ -468,7 +468,7 @@ static ssize_t arcfb_write(struct fb_info *info, const char __user *buf,
 	if (count) {
 		char *base_addr;
 
-		base_addr = (char __force *)info->screen_base;
+		base_addr = info->screen_buffer;
 		count -= copy_from_user(base_addr + p, buf, count);
 		*ppos += count;
 		err = -EFAULT;
@@ -525,7 +525,7 @@ static int arcfb_probe(struct platform_device *dev)
 	if (!info)
 		goto err;
 
-	info->screen_base = (char __iomem *)videomemory;
+	info->screen_buffer = videomemory;
 	info->fbops = &arcfb_ops;
 
 	info->var = arcfb_var;
@@ -595,7 +595,7 @@ static int arcfb_remove(struct platform_device *dev)
 		unregister_framebuffer(info);
 		if (irq)
 			free_irq(((struct arcfb_par *)(info->par))->irq, info);
-		vfree((void __force *)info->screen_base);
+		vfree(info->screen_buffer);
 		framebuffer_release(info);
 	}
 	return 0;
-- 
GitLab


From f5eccc49d88ba516cc3433ae4f5f3ace3b125cbd Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:38 +0200
Subject: [PATCH 0442/1791] fbdev/au1200fb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-6-tzimmermann@suse.de
---
 drivers/video/fbdev/au1200fb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/fbdev/au1200fb.c b/drivers/video/fbdev/au1200fb.c
index b6b22fa4a8a01..847f82ed52ff3 100644
--- a/drivers/video/fbdev/au1200fb.c
+++ b/drivers/video/fbdev/au1200fb.c
@@ -1568,7 +1568,7 @@ static int au1200fb_init_fbinfo(struct au1200fb_device *fbdev)
 	fbi->fix.mmio_len = 0;
 	fbi->fix.accel = FB_ACCEL_NONE;
 
-	fbi->screen_base = (char __iomem *) fbdev->fb_mem;
+	fbi->screen_buffer = fbdev->fb_mem;
 
 	au1200fb_update_fbinfo(fbi);
 
-- 
GitLab


From ada45eda25f28115bd14680b15aa1144e268f266 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:39 +0200
Subject: [PATCH 0443/1791] fbdev/broadsheetfb: Use struct
 fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-7-tzimmermann@suse.de
---
 drivers/video/fbdev/broadsheetfb.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/video/fbdev/broadsheetfb.c b/drivers/video/fbdev/broadsheetfb.c
index 55e62dd96f9be..65dc86b7081e3 100644
--- a/drivers/video/fbdev/broadsheetfb.c
+++ b/drivers/video/fbdev/broadsheetfb.c
@@ -824,7 +824,7 @@ static void broadsheet_init_display(struct broadsheetfb_par *par)
 
 	broadsheet_burst_write(par, (panel_table[par->panel_index].w *
 					panel_table[par->panel_index].h)/2,
-					(u16 *) par->info->screen_base);
+					(u16 *)par->info->screen_buffer);
 
 	broadsheet_send_command(par, BS_CMD_LD_IMG_END);
 
@@ -865,7 +865,7 @@ static void broadsheetfb_dpy_update_pages(struct broadsheetfb_par *par,
 						u16 y1, u16 y2)
 {
 	u16 args[5];
-	unsigned char *buf = (unsigned char *)par->info->screen_base;
+	unsigned char *buf = par->info->screen_buffer;
 
 	mutex_lock(&(par->io_lock));
 	/* y1 must be a multiple of 4 so drop the lower bits */
@@ -913,7 +913,7 @@ static void broadsheetfb_dpy_update(struct broadsheetfb_par *par)
 	broadsheet_send_cmdargs(par, BS_CMD_WR_REG, 1, args);
 	broadsheet_burst_write(par, (panel_table[par->panel_index].w *
 					panel_table[par->panel_index].h)/2,
-					(u16 *) par->info->screen_base);
+					(u16 *)par->info->screen_buffer);
 
 	broadsheet_send_command(par, BS_CMD_LD_IMG_END);
 
@@ -1033,7 +1033,7 @@ static ssize_t broadsheetfb_write(struct fb_info *info, const char __user *buf,
 		count = total_size - p;
 	}
 
-	dst = (void *)(info->screen_base + p);
+	dst = info->screen_buffer + p;
 
 	if (copy_from_user(dst, buf, count))
 		err = -EFAULT;
@@ -1109,7 +1109,7 @@ static int broadsheetfb_probe(struct platform_device *dev)
 	if (!videomemory)
 		goto err_fb_rel;
 
-	info->screen_base = (char *)videomemory;
+	info->screen_buffer = videomemory;
 	info->fbops = &broadsheetfb_ops;
 
 	broadsheetfb_var.xres = dpyw;
@@ -1205,7 +1205,7 @@ static int broadsheetfb_remove(struct platform_device *dev)
 		fb_deferred_io_cleanup(info);
 		par->board->cleanup(par);
 		fb_dealloc_cmap(&info->cmap);
-		vfree((void *)info->screen_base);
+		vfree(info->screen_buffer);
 		module_put(par->board->owner);
 		framebuffer_release(info);
 	}
-- 
GitLab


From 3907e683c09e36e52a6faf5ff91e8834e1bb2a85 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:40 +0200
Subject: [PATCH 0444/1791] fbdev/hecubafb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-8-tzimmermann@suse.de
---
 drivers/video/fbdev/hecubafb.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/video/fbdev/hecubafb.c b/drivers/video/fbdev/hecubafb.c
index eb1eaadc1bbb4..ddfa2853cc419 100644
--- a/drivers/video/fbdev/hecubafb.c
+++ b/drivers/video/fbdev/hecubafb.c
@@ -102,7 +102,7 @@ static void apollo_send_command(struct hecubafb_par *par, unsigned char data)
 static void hecubafb_dpy_update(struct hecubafb_par *par)
 {
 	int i;
-	unsigned char *buf = (unsigned char __force *)par->info->screen_base;
+	unsigned char *buf = par->info->screen_buffer;
 
 	apollo_send_command(par, APOLLO_START_NEW_IMG);
 
@@ -183,7 +183,7 @@ static ssize_t hecubafb_write(struct fb_info *info, const char __user *buf,
 		count = total_size - p;
 	}
 
-	dst = (void __force *) (info->screen_base + p);
+	dst = info->screen_buffer + p;
 
 	if (copy_from_user(dst, buf, count))
 		err = -EFAULT;
@@ -239,7 +239,7 @@ static int hecubafb_probe(struct platform_device *dev)
 	if (!info)
 		goto err_fballoc;
 
-	info->screen_base = (char __force __iomem *)videomemory;
+	info->screen_buffer = videomemory;
 	info->fbops = &hecubafb_ops;
 
 	info->var = hecubafb_var;
@@ -287,7 +287,7 @@ static int hecubafb_remove(struct platform_device *dev)
 		struct hecubafb_par *par = info->par;
 		fb_deferred_io_cleanup(info);
 		unregister_framebuffer(info);
-		vfree((void __force *)info->screen_base);
+		vfree(info->screen_buffer);
 		if (par->board->remove)
 			par->board->remove(par);
 		module_put(par->board->owner);
-- 
GitLab


From 867187821e5eb02f37e03d98d5f3c9cb65faaa11 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:41 +0200
Subject: [PATCH 0445/1791] fbdev/metronomefb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-9-tzimmermann@suse.de
---
 drivers/video/fbdev/metronomefb.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/video/fbdev/metronomefb.c b/drivers/video/fbdev/metronomefb.c
index 9fd4bb85d7356..afa9b41f5a87d 100644
--- a/drivers/video/fbdev/metronomefb.c
+++ b/drivers/video/fbdev/metronomefb.c
@@ -438,7 +438,7 @@ static void metronomefb_dpy_update(struct metronomefb_par *par)
 {
 	int fbsize;
 	u16 cksum;
-	unsigned char *buf = (unsigned char __force *)par->info->screen_base;
+	unsigned char *buf = par->info->screen_buffer;
 
 	fbsize = par->info->fix.smem_len;
 	/* copy from vm to metromem */
@@ -453,7 +453,7 @@ static u16 metronomefb_dpy_update_page(struct metronomefb_par *par, int index)
 {
 	int i;
 	u16 csum = 0;
-	u16 *buf = (u16 __force *)(par->info->screen_base + index);
+	u16 *buf = (u16 *)(par->info->screen_buffer + index);
 	u16 *img = (u16 *)(par->metromem_img + index);
 
 	/* swizzle from vm to metromem and recalc cksum at the same time*/
@@ -543,7 +543,7 @@ static ssize_t metronomefb_write(struct fb_info *info, const char __user *buf,
 		count = total_size - p;
 	}
 
-	dst = (void __force *)(info->screen_base + p);
+	dst = info->screen_buffer + p;
 
 	if (copy_from_user(dst, buf, count))
 		err = -EFAULT;
@@ -599,7 +599,7 @@ static int metronomefb_probe(struct platform_device *dev)
 		goto err;
 
 	/* we have two blocks of memory.
-	info->screen_base which is vm, and is the fb used by apps.
+	info->screen_buffer which is vm, and is the fb used by apps.
 	par->metromem which is physically contiguous memory and
 	contains the display controller commands, waveform,
 	processed image data and padding. this is the data pulled
@@ -634,7 +634,7 @@ static int metronomefb_probe(struct platform_device *dev)
 	if (!videomemory)
 		goto err_fb_rel;
 
-	info->screen_base = (char __force __iomem *)videomemory;
+	info->screen_buffer = videomemory;
 	info->fbops = &metronomefb_ops;
 
 	metronomefb_fix.line_length = fw;
@@ -756,7 +756,7 @@ static int metronomefb_remove(struct platform_device *dev)
 		fb_dealloc_cmap(&info->cmap);
 		par->board->cleanup(par);
 		vfree(par->csum_table);
-		vfree((void __force *)info->screen_base);
+		vfree(info->screen_buffer);
 		module_put(par->board->owner);
 		dev_dbg(&dev->dev, "calling release\n");
 		framebuffer_release(info);
-- 
GitLab


From bdb616479eff419d5f881efbf1485ed098cbdbce Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:42 +0200
Subject: [PATCH 0446/1791] fbdev/ps3fb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-10-tzimmermann@suse.de
---
 drivers/video/fbdev/ps3fb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/fbdev/ps3fb.c b/drivers/video/fbdev/ps3fb.c
index 2fe08b67eda71..8dfa9f4048164 100644
--- a/drivers/video/fbdev/ps3fb.c
+++ b/drivers/video/fbdev/ps3fb.c
@@ -650,7 +650,7 @@ static int ps3fb_set_par(struct fb_info *info)
 	}
 
 	/* Clear XDR frame buffer memory */
-	memset((void __force *)info->screen_base, 0, info->fix.smem_len);
+	memset(info->screen_buffer, 0, info->fix.smem_len);
 
 	/* Clear DDR frame buffer memory */
 	lines = vmode->yres * par->num_frames;
@@ -1139,7 +1139,7 @@ static int ps3fb_probe(struct ps3_system_bus_device *dev)
 	 * memory
 	 */
 	fb_start = ps3fb_videomemory.address + GPU_FB_START;
-	info->screen_base = (char __force __iomem *)fb_start;
+	info->screen_buffer = fb_start;
 	info->fix.smem_start = __pa(fb_start);
 	info->fix.smem_len = ps3fb_videomemory.size - GPU_FB_START;
 
-- 
GitLab


From 6ca49268e0c328b441cd0f3387ee8568e33ffdc5 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:43 +0200
Subject: [PATCH 0447/1791] fbdev/smscufx: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-11-tzimmermann@suse.de
---
 drivers/video/fbdev/smscufx.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/video/fbdev/smscufx.c b/drivers/video/fbdev/smscufx.c
index 2ad6e98ce10d5..17cec62cc65db 100644
--- a/drivers/video/fbdev/smscufx.c
+++ b/drivers/video/fbdev/smscufx.c
@@ -1150,7 +1150,7 @@ static void ufx_free_framebuffer(struct ufx_data *dev)
 		fb_dealloc_cmap(&info->cmap);
 	if (info->monspecs.modedb)
 		fb_destroy_modedb(info->monspecs.modedb);
-	vfree(info->screen_base);
+	vfree(info->screen_buffer);
 
 	fb_destroy_modelist(&info->modelist);
 
@@ -1257,7 +1257,7 @@ static int ufx_ops_set_par(struct fb_info *info)
 
 	if ((result == 0) && (dev->fb_count == 0)) {
 		/* paint greenscreen */
-		pix_framebuffer = (u16 *) info->screen_base;
+		pix_framebuffer = (u16 *)info->screen_buffer;
 		for (i = 0; i < info->fix.smem_len / 2; i++)
 			pix_framebuffer[i] = 0x37e6;
 
@@ -1303,7 +1303,7 @@ static int ufx_realloc_framebuffer(struct ufx_data *dev, struct fb_info *info)
 {
 	int old_len = info->fix.smem_len;
 	int new_len;
-	unsigned char *old_fb = info->screen_base;
+	unsigned char *old_fb = info->screen_buffer;
 	unsigned char *new_fb;
 
 	pr_debug("Reallocating framebuffer. Addresses will change!");
@@ -1318,12 +1318,12 @@ static int ufx_realloc_framebuffer(struct ufx_data *dev, struct fb_info *info)
 		if (!new_fb)
 			return -ENOMEM;
 
-		if (info->screen_base) {
+		if (info->screen_buffer) {
 			memcpy(new_fb, old_fb, old_len);
-			vfree(info->screen_base);
+			vfree(info->screen_buffer);
 		}
 
-		info->screen_base = new_fb;
+		info->screen_buffer = new_fb;
 		info->fix.smem_len = PAGE_ALIGN(new_len);
 		info->fix.smem_start = (unsigned long) new_fb;
 		info->flags = smscufx_info_flags;
@@ -1746,7 +1746,7 @@ static int ufx_usb_probe(struct usb_interface *interface,
 	atomic_set(&dev->usb_active, 0);
 setup_modes:
 	fb_destroy_modedb(info->monspecs.modedb);
-	vfree(info->screen_base);
+	vfree(info->screen_buffer);
 	fb_destroy_modelist(&info->modelist);
 error:
 	fb_dealloc_cmap(&info->cmap);
-- 
GitLab


From 962a3faf32666a31c5a05c2a69023a4abab6389f Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:44 +0200
Subject: [PATCH 0448/1791] fbdev/udlfb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-12-tzimmermann@suse.de
---
 drivers/video/fbdev/udlfb.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/video/fbdev/udlfb.c b/drivers/video/fbdev/udlfb.c
index 216d49c9d47e5..09cf9381075aa 100644
--- a/drivers/video/fbdev/udlfb.c
+++ b/drivers/video/fbdev/udlfb.c
@@ -1006,7 +1006,7 @@ static void dlfb_ops_destroy(struct fb_info *info)
 		fb_dealloc_cmap(&info->cmap);
 	if (info->monspecs.modedb)
 		fb_destroy_modedb(info->monspecs.modedb);
-	vfree(info->screen_base);
+	vfree(info->screen_buffer);
 
 	fb_destroy_modelist(&info->modelist);
 
@@ -1120,7 +1120,7 @@ static int dlfb_ops_set_par(struct fb_info *info)
 
 		/* paint greenscreen */
 
-		pix_framebuffer = (u16 *) info->screen_base;
+		pix_framebuffer = (u16 *)info->screen_buffer;
 		for (i = 0; i < info->fix.smem_len / 2; i++)
 			pix_framebuffer[i] = 0x37e6;
 	}
@@ -1219,7 +1219,7 @@ static void dlfb_deferred_vfree(struct dlfb_data *dlfb, void *mem)
 static int dlfb_realloc_framebuffer(struct dlfb_data *dlfb, struct fb_info *info, u32 new_len)
 {
 	u32 old_len = info->fix.smem_len;
-	const void *old_fb = (const void __force *)info->screen_base;
+	const void *old_fb = info->screen_buffer;
 	unsigned char *new_fb;
 	unsigned char *new_back = NULL;
 
@@ -1236,12 +1236,12 @@ static int dlfb_realloc_framebuffer(struct dlfb_data *dlfb, struct fb_info *info
 		}
 		memset(new_fb, 0xff, new_len);
 
-		if (info->screen_base) {
+		if (info->screen_buffer) {
 			memcpy(new_fb, old_fb, old_len);
-			dlfb_deferred_vfree(dlfb, (void __force *)info->screen_base);
+			dlfb_deferred_vfree(dlfb, info->screen_buffer);
 		}
 
-		info->screen_base = (char __iomem *)new_fb;
+		info->screen_buffer = new_fb;
 		info->fix.smem_len = new_len;
 		info->fix.smem_start = (unsigned long) new_fb;
 		info->flags = udlfb_info_flags;
-- 
GitLab


From 5a2c399d609bff499fdebfc5d80d8a5fc720630e Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:45 +0200
Subject: [PATCH 0449/1791] fbdev/vfb: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead. This also gets
rid of casting needed due to not using the correct data type.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-13-tzimmermann@suse.de
---
 drivers/video/fbdev/vfb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/fbdev/vfb.c b/drivers/video/fbdev/vfb.c
index 95d3c59867d0a..0ea611dd27599 100644
--- a/drivers/video/fbdev/vfb.c
+++ b/drivers/video/fbdev/vfb.c
@@ -439,7 +439,7 @@ static int vfb_probe(struct platform_device *dev)
 	if (!info)
 		goto err;
 
-	info->screen_base = (char __iomem *)videomemory;
+	info->screen_buffer = videomemory;
 	info->fbops = &vfb_ops;
 
 	if (!fb_find_mode(&info->var, info, mode_option,
-- 
GitLab


From 254a4fda50590c5458f012220df3d0dc2f11fd6e Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:46 +0200
Subject: [PATCH 0450/1791] fbdev/xen-fbfront: Use struct fb_info.screen_buffer

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

As the driver operates on the latter address space, it is wrong to use
.screen_base and .screen_buffer must be used instead.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-14-tzimmermann@suse.de
---
 drivers/video/fbdev/xen-fbfront.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/fbdev/xen-fbfront.c b/drivers/video/fbdev/xen-fbfront.c
index d7f3e6281ce48..9b2a786621a63 100644
--- a/drivers/video/fbdev/xen-fbfront.c
+++ b/drivers/video/fbdev/xen-fbfront.c
@@ -429,7 +429,7 @@ static int xenfb_probe(struct xenbus_device *dev,
 	fb_info->pseudo_palette = fb_info->par;
 	fb_info->par = info;
 
-	fb_info->screen_base = info->fb;
+	fb_info->screen_buffer = info->fb;
 
 	fb_info->fbops = &xenfb_fb_ops;
 	fb_info->var.xres_virtual = fb_info->var.xres = video[KPARAM_WIDTH];
-- 
GitLab


From 921b7383f34802726a7f45b6b38ef62bae0df4d6 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:47 +0200
Subject: [PATCH 0451/1791] fbdev: Return number of bytes read or written

Always return the number of bytes read or written within the
framebuffer. Only return an errno code if framebuffer memory
was not touched. This is the semantics required by POSIX and
makes fb_read() and fb_write() compatible with IGT tests. [1]

This bug has been fixed for fb_write() long ago by
commit 6a2a88668e90 ("[PATCH] fbdev: Fix return error of
fb_write"). The code in fb_read() and the corresponding fb_sys_()
helpers was forgotten.

It can happen that copy_{from, to}_user() only partially copies
the given buffer. Take this into account when calculating the
number of bytes.

v2:
	* consider return value from copy_{from,to}_user() (Geert)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/blob/master/tests/fbdev.c # 1
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-15-tzimmermann@suse.de
---
 drivers/video/fbdev/core/fb_sys_fops.c | 24 ++++++++++++++----------
 drivers/video/fbdev/core/fbmem.c       | 15 ++++++++++-----
 2 files changed, 24 insertions(+), 15 deletions(-)

diff --git a/drivers/video/fbdev/core/fb_sys_fops.c b/drivers/video/fbdev/core/fb_sys_fops.c
index ff275d7f3eaf1..cefb77b9546d0 100644
--- a/drivers/video/fbdev/core/fb_sys_fops.c
+++ b/drivers/video/fbdev/core/fb_sys_fops.c
@@ -19,7 +19,8 @@ ssize_t fb_sys_read(struct fb_info *info, char __user *buf, size_t count,
 	unsigned long p = *ppos;
 	void *src;
 	int err = 0;
-	unsigned long total_size;
+	unsigned long total_size, c;
+	ssize_t ret;
 
 	if (info->state != FBINFO_STATE_RUNNING)
 		return -EPERM;
@@ -43,13 +44,14 @@ ssize_t fb_sys_read(struct fb_info *info, char __user *buf, size_t count,
 	if (info->fbops->fb_sync)
 		info->fbops->fb_sync(info);
 
-	if (copy_to_user(buf, src, count))
+	c = copy_to_user(buf, src, count);
+	if (c)
 		err = -EFAULT;
+	ret = count - c;
 
-	if  (!err)
-		*ppos += count;
+	*ppos += ret;
 
-	return (err) ? err : count;
+	return ret ? ret : err;
 }
 EXPORT_SYMBOL_GPL(fb_sys_read);
 
@@ -59,7 +61,8 @@ ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 	unsigned long p = *ppos;
 	void *dst;
 	int err = 0;
-	unsigned long total_size;
+	unsigned long total_size, c;
+	size_t ret;
 
 	if (info->state != FBINFO_STATE_RUNNING)
 		return -EPERM;
@@ -89,13 +92,14 @@ ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 	if (info->fbops->fb_sync)
 		info->fbops->fb_sync(info);
 
-	if (copy_from_user(dst, buf, count))
+	c = copy_from_user(dst, buf, count);
+	if (c)
 		err = -EFAULT;
+	ret = count - c;
 
-	if  (!err)
-		*ppos += count;
+	*ppos += ret;
 
-	return (err) ? err : count;
+	return ret ? ret : err;
 }
 EXPORT_SYMBOL_GPL(fb_sys_write);
 
diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c
index 875541ff185bf..9c79fb076c6d9 100644
--- a/drivers/video/fbdev/core/fbmem.c
+++ b/drivers/video/fbdev/core/fbmem.c
@@ -766,7 +766,7 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	u8 *buffer, *dst;
 	u8 __iomem *src;
 	int c, cnt = 0, err = 0;
-	unsigned long total_size;
+	unsigned long total_size, trailing;
 
 	if (!info || ! info->screen_base)
 		return -ENODEV;
@@ -808,10 +808,13 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 		dst += c;
 		src += c;
 
-		if (copy_to_user(buf, buffer, c)) {
+		trailing = copy_to_user(buf, buffer, c);
+		if (trailing == c) {
 			err = -EFAULT;
 			break;
 		}
+		c -= trailing;
+
 		*ppos += c;
 		buf += c;
 		cnt += c;
@@ -820,7 +823,7 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 
 	kfree(buffer);
 
-	return (err) ? err : cnt;
+	return cnt ? cnt : err;
 }
 
 static ssize_t
@@ -831,7 +834,7 @@ fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	u8 *buffer, *src;
 	u8 __iomem *dst;
 	int c, cnt = 0, err = 0;
-	unsigned long total_size;
+	unsigned long total_size, trailing;
 
 	if (!info || !info->screen_base)
 		return -ENODEV;
@@ -876,10 +879,12 @@ fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 		c = (count > PAGE_SIZE) ? PAGE_SIZE : count;
 		src = buffer;
 
-		if (copy_from_user(src, buf, c)) {
+		trailing = copy_from_user(src, buf, c);
+		if (trailing == c) {
 			err = -EFAULT;
 			break;
 		}
+		c -= trailing;
 
 		fb_memcpy_tofb(dst, src, c);
 		dst += c;
-- 
GitLab


From f06875dc4c4f5d8895519857e441d407e88280e4 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:48 +0200
Subject: [PATCH 0452/1791] fbdev: Use screen_buffer in fb_sys_{read,write}()

Use info->screen_buffer when reading and writing framebuffers in
system memory. It's the correct pointer for this address space.

The struct fb_info has a union to store the framebuffer memory. This can
either be info->screen_base if the framebuffer is stored in I/O memory,
or info->screen_buffer if the framebuffer is stored in system memory.

Since the fb_sys_{read,write}() functions operate on the latter address
space, it is wrong to use .screen_base and .screen_buffer must be used
instead. This also gets rid of all the casting needed due to not using
the correct data type.

v2:
	* add detailed commit message (Javier)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-16-tzimmermann@suse.de
---
 drivers/video/fbdev/core/fb_sys_fops.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/fbdev/core/fb_sys_fops.c b/drivers/video/fbdev/core/fb_sys_fops.c
index cefb77b9546d0..6589123f4127a 100644
--- a/drivers/video/fbdev/core/fb_sys_fops.c
+++ b/drivers/video/fbdev/core/fb_sys_fops.c
@@ -39,7 +39,7 @@ ssize_t fb_sys_read(struct fb_info *info, char __user *buf, size_t count,
 	if (count + p > total_size)
 		count = total_size - p;
 
-	src = (void __force *)(info->screen_base + p);
+	src = info->screen_buffer + p;
 
 	if (info->fbops->fb_sync)
 		info->fbops->fb_sync(info);
@@ -87,7 +87,7 @@ ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 		count = total_size - p;
 	}
 
-	dst = (void __force *) (info->screen_base + p);
+	dst = info->screen_buffer + p;
 
 	if (info->fbops->fb_sync)
 		info->fbops->fb_sync(info);
-- 
GitLab


From 453bd91302f16ddb8a2f5eb3df9fda3206dbacef Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:49 +0200
Subject: [PATCH 0453/1791] fbdev: Don't re-validate info->state in fb_ops
 implementations

The file-op entry points fb_read() and fb_write() verify that
info->state has been set to FBINFO_STATE_RUNNING. Remove the same
test from the implementations of struct fb_ops.{fb_read,fb_write}.

v2:
	* also remove test from ivtvfb, braodsheetfb, hecubafb and
	  metronomefb (Geert)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-17-tzimmermann@suse.de
---
 drivers/media/pci/ivtv/ivtvfb.c        | 3 ---
 drivers/video/fbdev/broadsheetfb.c     | 3 ---
 drivers/video/fbdev/core/fb_sys_fops.c | 6 ------
 drivers/video/fbdev/hecubafb.c         | 3 ---
 drivers/video/fbdev/metronomefb.c      | 3 ---
 drivers/video/fbdev/sm712fb.c          | 6 ------
 6 files changed, 24 deletions(-)

diff --git a/drivers/media/pci/ivtv/ivtvfb.c b/drivers/media/pci/ivtv/ivtvfb.c
index 00ac94d4ab19d..22123a25daea9 100644
--- a/drivers/media/pci/ivtv/ivtvfb.c
+++ b/drivers/media/pci/ivtv/ivtvfb.c
@@ -378,9 +378,6 @@ static ssize_t ivtvfb_write(struct fb_info *info, const char __user *buf,
 	unsigned long dma_size;
 	u16 lead = 0, tail = 0;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->screen_size;
 
 	if (total_size == 0)
diff --git a/drivers/video/fbdev/broadsheetfb.c b/drivers/video/fbdev/broadsheetfb.c
index 65dc86b7081e3..691de5df581b8 100644
--- a/drivers/video/fbdev/broadsheetfb.c
+++ b/drivers/video/fbdev/broadsheetfb.c
@@ -1013,9 +1013,6 @@ static ssize_t broadsheetfb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/core/fb_sys_fops.c b/drivers/video/fbdev/core/fb_sys_fops.c
index 6589123f4127a..7dee5d3c7fb14 100644
--- a/drivers/video/fbdev/core/fb_sys_fops.c
+++ b/drivers/video/fbdev/core/fb_sys_fops.c
@@ -22,9 +22,6 @@ ssize_t fb_sys_read(struct fb_info *info, char __user *buf, size_t count,
 	unsigned long total_size, c;
 	ssize_t ret;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->screen_size;
 
 	if (total_size == 0)
@@ -64,9 +61,6 @@ ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 	unsigned long total_size, c;
 	size_t ret;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->screen_size;
 
 	if (total_size == 0)
diff --git a/drivers/video/fbdev/hecubafb.c b/drivers/video/fbdev/hecubafb.c
index ddfa2853cc419..a2996d39f918c 100644
--- a/drivers/video/fbdev/hecubafb.c
+++ b/drivers/video/fbdev/hecubafb.c
@@ -163,9 +163,6 @@ static ssize_t hecubafb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/metronomefb.c b/drivers/video/fbdev/metronomefb.c
index afa9b41f5a87d..2bb068cadac63 100644
--- a/drivers/video/fbdev/metronomefb.c
+++ b/drivers/video/fbdev/metronomefb.c
@@ -523,9 +523,6 @@ static ssize_t metronomefb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/sm712fb.c b/drivers/video/fbdev/sm712fb.c
index b528776c76125..6f852cd756c5f 100644
--- a/drivers/video/fbdev/sm712fb.c
+++ b/drivers/video/fbdev/sm712fb.c
@@ -1031,9 +1031,6 @@ static ssize_t smtcfb_read(struct fb_info *info, char __user *buf,
 	if (!info || !info->screen_base)
 		return -ENODEV;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->screen_size;
 
 	if (total_size == 0)
@@ -1097,9 +1094,6 @@ static ssize_t smtcfb_write(struct fb_info *info, const char __user *buf,
 	if (!info || !info->screen_base)
 		return -ENODEV;
 
-	if (info->state != FBINFO_STATE_RUNNING)
-		return -EPERM;
-
 	total_size = info->screen_size;
 
 	if (total_size == 0)
-- 
GitLab


From 3f8974f68f408f4983443b745a3cf95b06112469 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:50 +0200
Subject: [PATCH 0454/1791] fbdev: Validate info->screen_{base, buffer} in
 fb_ops implementations

Push the test for info->screen_base from fb_read() and fb_write() into
the implementations of struct fb_ops.{fb_read,fb_write}. In cases where
the driver operates on info->screen_buffer, test this field instead.

While bothi fields, screen_base and screen_buffer, are stored in the
same location, they refer to different address spaces. For correctness,
we want to test each field in exactly the code that uses it.

v2:
	* also test screen_base in pvr2fb (Geert)
	* also test screen_buffer in ivtvfb, arcfb, broadsheetfb,
	  hecubafb, metronomefb and ssd1307fb (Geert)
	* give a rational for the change (Geert)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-18-tzimmermann@suse.de
---
 drivers/media/pci/ivtv/ivtvfb.c        |  3 +++
 drivers/video/fbdev/arcfb.c            |  3 +++
 drivers/video/fbdev/broadsheetfb.c     |  3 +++
 drivers/video/fbdev/cobalt_lcdfb.c     |  6 ++++++
 drivers/video/fbdev/core/fb_sys_fops.c |  6 ++++++
 drivers/video/fbdev/core/fbmem.c       | 10 ++++++++--
 drivers/video/fbdev/hecubafb.c         |  3 +++
 drivers/video/fbdev/metronomefb.c      |  3 +++
 drivers/video/fbdev/pvr2fb.c           |  3 +++
 drivers/video/fbdev/sm712fb.c          |  4 ++--
 drivers/video/fbdev/ssd1307fb.c        |  3 +++
 11 files changed, 43 insertions(+), 4 deletions(-)

diff --git a/drivers/media/pci/ivtv/ivtvfb.c b/drivers/media/pci/ivtv/ivtvfb.c
index 22123a25daea9..0aeb9daaee4cb 100644
--- a/drivers/media/pci/ivtv/ivtvfb.c
+++ b/drivers/media/pci/ivtv/ivtvfb.c
@@ -378,6 +378,9 @@ static ssize_t ivtvfb_write(struct fb_info *info, const char __user *buf,
 	unsigned long dma_size;
 	u16 lead = 0, tail = 0;
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	total_size = info->screen_size;
 
 	if (total_size == 0)
diff --git a/drivers/video/fbdev/arcfb.c b/drivers/video/fbdev/arcfb.c
index 088c4b30fd314..7750e020839ec 100644
--- a/drivers/video/fbdev/arcfb.c
+++ b/drivers/video/fbdev/arcfb.c
@@ -451,6 +451,9 @@ static ssize_t arcfb_write(struct fb_info *info, const char __user *buf,
 	struct arcfb_par *par;
 	unsigned int xres;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	p = *ppos;
 	par = info->par;
 	xres = info->var.xres;
diff --git a/drivers/video/fbdev/broadsheetfb.c b/drivers/video/fbdev/broadsheetfb.c
index 691de5df581b8..e9c5d5c04062b 100644
--- a/drivers/video/fbdev/broadsheetfb.c
+++ b/drivers/video/fbdev/broadsheetfb.c
@@ -1013,6 +1013,9 @@ static ssize_t broadsheetfb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/cobalt_lcdfb.c b/drivers/video/fbdev/cobalt_lcdfb.c
index 5f8b6324d2e8c..26dbd1c78195c 100644
--- a/drivers/video/fbdev/cobalt_lcdfb.c
+++ b/drivers/video/fbdev/cobalt_lcdfb.c
@@ -129,6 +129,9 @@ static ssize_t cobalt_lcdfb_read(struct fb_info *info, char __user *buf,
 	unsigned long pos;
 	int len, retval = 0;
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	pos = *ppos;
 	if (pos >= LCD_CHARS_MAX || count == 0)
 		return 0;
@@ -175,6 +178,9 @@ static ssize_t cobalt_lcdfb_write(struct fb_info *info, const char __user *buf,
 	unsigned long pos;
 	int len, retval = 0;
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	pos = *ppos;
 	if (pos >= LCD_CHARS_MAX || count == 0)
 		return 0;
diff --git a/drivers/video/fbdev/core/fb_sys_fops.c b/drivers/video/fbdev/core/fb_sys_fops.c
index 7dee5d3c7fb14..0cb0989abda61 100644
--- a/drivers/video/fbdev/core/fb_sys_fops.c
+++ b/drivers/video/fbdev/core/fb_sys_fops.c
@@ -22,6 +22,9 @@ ssize_t fb_sys_read(struct fb_info *info, char __user *buf, size_t count,
 	unsigned long total_size, c;
 	ssize_t ret;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->screen_size;
 
 	if (total_size == 0)
@@ -61,6 +64,9 @@ ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 	unsigned long total_size, c;
 	size_t ret;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->screen_size;
 
 	if (total_size == 0)
diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c
index 9c79fb076c6d9..bf5a0780457e1 100644
--- a/drivers/video/fbdev/core/fbmem.c
+++ b/drivers/video/fbdev/core/fbmem.c
@@ -768,7 +768,7 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	int c, cnt = 0, err = 0;
 	unsigned long total_size, trailing;
 
-	if (!info || ! info->screen_base)
+	if (!info)
 		return -ENODEV;
 
 	if (info->state != FBINFO_STATE_RUNNING)
@@ -777,6 +777,9 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	if (info->fbops->fb_read)
 		return info->fbops->fb_read(info, buf, count, ppos);
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	total_size = info->screen_size;
 
 	if (total_size == 0)
@@ -836,7 +839,7 @@ fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	int c, cnt = 0, err = 0;
 	unsigned long total_size, trailing;
 
-	if (!info || !info->screen_base)
+	if (!info)
 		return -ENODEV;
 
 	if (info->state != FBINFO_STATE_RUNNING)
@@ -845,6 +848,9 @@ fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	if (info->fbops->fb_write)
 		return info->fbops->fb_write(info, buf, count, ppos);
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	total_size = info->screen_size;
 
 	if (total_size == 0)
diff --git a/drivers/video/fbdev/hecubafb.c b/drivers/video/fbdev/hecubafb.c
index a2996d39f918c..72308d4e0c22d 100644
--- a/drivers/video/fbdev/hecubafb.c
+++ b/drivers/video/fbdev/hecubafb.c
@@ -163,6 +163,9 @@ static ssize_t hecubafb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/metronomefb.c b/drivers/video/fbdev/metronomefb.c
index 2bb068cadac63..7fc59466fe6c9 100644
--- a/drivers/video/fbdev/metronomefb.c
+++ b/drivers/video/fbdev/metronomefb.c
@@ -523,6 +523,9 @@ static ssize_t metronomefb_write(struct fb_info *info, const char __user *buf,
 	int err = 0;
 	unsigned long total_size;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
diff --git a/drivers/video/fbdev/pvr2fb.c b/drivers/video/fbdev/pvr2fb.c
index 6888127a5eb88..550fdb5b4d41e 100644
--- a/drivers/video/fbdev/pvr2fb.c
+++ b/drivers/video/fbdev/pvr2fb.c
@@ -647,6 +647,9 @@ static ssize_t pvr2fb_write(struct fb_info *info, const char *buf,
 	struct page **pages;
 	int ret, i;
 
+	if (!info->screen_base)
+		return -ENODEV;
+
 	nr_pages = (count + PAGE_SIZE - 1) >> PAGE_SHIFT;
 
 	pages = kmalloc_array(nr_pages, sizeof(struct page *), GFP_KERNEL);
diff --git a/drivers/video/fbdev/sm712fb.c b/drivers/video/fbdev/sm712fb.c
index 6f852cd756c5f..b7ad3c644e138 100644
--- a/drivers/video/fbdev/sm712fb.c
+++ b/drivers/video/fbdev/sm712fb.c
@@ -1028,7 +1028,7 @@ static ssize_t smtcfb_read(struct fb_info *info, char __user *buf,
 	int c, i, cnt = 0, err = 0;
 	unsigned long total_size;
 
-	if (!info || !info->screen_base)
+	if (!info->screen_base)
 		return -ENODEV;
 
 	total_size = info->screen_size;
@@ -1091,7 +1091,7 @@ static ssize_t smtcfb_write(struct fb_info *info, const char __user *buf,
 	int c, i, cnt = 0, err = 0;
 	unsigned long total_size;
 
-	if (!info || !info->screen_base)
+	if (!info->screen_base)
 		return -ENODEV;
 
 	total_size = info->screen_size;
diff --git a/drivers/video/fbdev/ssd1307fb.c b/drivers/video/fbdev/ssd1307fb.c
index 046b9990d27cd..a8f2975de76be 100644
--- a/drivers/video/fbdev/ssd1307fb.c
+++ b/drivers/video/fbdev/ssd1307fb.c
@@ -301,6 +301,9 @@ static ssize_t ssd1307fb_write(struct fb_info *info, const char __user *buf,
 	void *dst;
 	int ret;
 
+	if (!info->screen_buffer)
+		return -ENODEV;
+
 	total_size = info->fix.smem_len;
 
 	if (p > total_size)
-- 
GitLab


From 6121cd9ef911432b14c2a17aefaf8cd2f3cfcdff Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:51 +0200
Subject: [PATCH 0455/1791] fbdev: Move I/O read and write code into helper
 functions

Move the existing I/O read and write code for I/O memory into
the new helpers fb_cfb_read() and fb_cfb_write(). Make them the
default fp_ops. No functional changes.

In the near term, the new functions will be useful to the DRM
subsystem, which currently provides it's own implementation. It
can then use the shared code. In the longer term, it might make
sense to revise the I/O helper's default status and make them
opt-in by the driver. Systems that don't use them would not
contain the code any longer.

v2:
	* add detailed commit message (Javier)
	* rename fb_cfb_() to fb_io_() (Geert)
	* add fixes that got lost while moving the code (Geert)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Helge Deller <deller@gmx.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-19-tzimmermann@suse.de
---
 drivers/video/fbdev/core/Makefile     |   2 +-
 drivers/video/fbdev/core/fb_io_fops.c | 133 ++++++++++++++++++++++++++
 drivers/video/fbdev/core/fbmem.c      | 118 +----------------------
 include/linux/fb.h                    |  10 ++
 4 files changed, 146 insertions(+), 117 deletions(-)
 create mode 100644 drivers/video/fbdev/core/fb_io_fops.c

diff --git a/drivers/video/fbdev/core/Makefile b/drivers/video/fbdev/core/Makefile
index 08fabce76b749..8f0060160ffb7 100644
--- a/drivers/video/fbdev/core/Makefile
+++ b/drivers/video/fbdev/core/Makefile
@@ -2,7 +2,7 @@
 obj-$(CONFIG_FB_NOTIFY)           += fb_notify.o
 obj-$(CONFIG_FB)                  += fb.o
 fb-y                              := fbmem.o fbmon.o fbcmap.o fbsysfs.o \
-                                     modedb.o fbcvt.o fb_cmdline.o
+                                     modedb.o fbcvt.o fb_cmdline.o fb_io_fops.o
 fb-$(CONFIG_FB_DEFERRED_IO)       += fb_defio.o
 
 ifeq ($(CONFIG_FRAMEBUFFER_CONSOLE),y)
diff --git a/drivers/video/fbdev/core/fb_io_fops.c b/drivers/video/fbdev/core/fb_io_fops.c
new file mode 100644
index 0000000000000..f5299d50f33b9
--- /dev/null
+++ b/drivers/video/fbdev/core/fb_io_fops.c
@@ -0,0 +1,133 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/fb.h>
+#include <linux/module.h>
+#include <linux/uaccess.h>
+
+ssize_t fb_io_read(struct fb_info *info, char __user *buf, size_t count, loff_t *ppos)
+{
+	unsigned long p = *ppos;
+	u8 *buffer, *dst;
+	u8 __iomem *src;
+	int c, cnt = 0, err = 0;
+	unsigned long total_size, trailing;
+
+	if (!info->screen_base)
+		return -ENODEV;
+
+	total_size = info->screen_size;
+
+	if (total_size == 0)
+		total_size = info->fix.smem_len;
+
+	if (p >= total_size)
+		return 0;
+
+	if (count >= total_size)
+		count = total_size;
+
+	if (count + p > total_size)
+		count = total_size - p;
+
+	buffer = kmalloc((count > PAGE_SIZE) ? PAGE_SIZE : count,
+			 GFP_KERNEL);
+	if (!buffer)
+		return -ENOMEM;
+
+	src = (u8 __iomem *) (info->screen_base + p);
+
+	if (info->fbops->fb_sync)
+		info->fbops->fb_sync(info);
+
+	while (count) {
+		c  = (count > PAGE_SIZE) ? PAGE_SIZE : count;
+		dst = buffer;
+		fb_memcpy_fromfb(dst, src, c);
+		dst += c;
+		src += c;
+
+		trailing = copy_to_user(buf, buffer, c);
+		if (trailing == c) {
+			err = -EFAULT;
+			break;
+		}
+		c -= trailing;
+
+		*ppos += c;
+		buf += c;
+		cnt += c;
+		count -= c;
+	}
+
+	kfree(buffer);
+
+	return cnt ? cnt : err;
+}
+EXPORT_SYMBOL(fb_io_read);
+
+ssize_t fb_io_write(struct fb_info *info, const char __user *buf, size_t count, loff_t *ppos)
+{
+	unsigned long p = *ppos;
+	u8 *buffer, *src;
+	u8 __iomem *dst;
+	int c, cnt = 0, err = 0;
+	unsigned long total_size, trailing;
+
+	if (!info->screen_base)
+		return -ENODEV;
+
+	total_size = info->screen_size;
+
+	if (total_size == 0)
+		total_size = info->fix.smem_len;
+
+	if (p > total_size)
+		return -EFBIG;
+
+	if (count > total_size) {
+		err = -EFBIG;
+		count = total_size;
+	}
+
+	if (count + p > total_size) {
+		if (!err)
+			err = -ENOSPC;
+
+		count = total_size - p;
+	}
+
+	buffer = kmalloc((count > PAGE_SIZE) ? PAGE_SIZE : count,
+			 GFP_KERNEL);
+	if (!buffer)
+		return -ENOMEM;
+
+	dst = (u8 __iomem *) (info->screen_base + p);
+
+	if (info->fbops->fb_sync)
+		info->fbops->fb_sync(info);
+
+	while (count) {
+		c = (count > PAGE_SIZE) ? PAGE_SIZE : count;
+		src = buffer;
+
+		trailing = copy_from_user(src, buf, c);
+		if (trailing == c) {
+			err = -EFAULT;
+			break;
+		}
+		c -= trailing;
+
+		fb_memcpy_tofb(dst, src, c);
+		dst += c;
+		src += c;
+		*ppos += c;
+		buf += c;
+		cnt += c;
+		count -= c;
+	}
+
+	kfree(buffer);
+
+	return (cnt) ? cnt : err;
+}
+EXPORT_SYMBOL(fb_io_write);
diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c
index bf5a0780457e1..63af40831d7d6 100644
--- a/drivers/video/fbdev/core/fbmem.c
+++ b/drivers/video/fbdev/core/fbmem.c
@@ -761,12 +761,7 @@ static struct fb_info *file_fb_info(struct file *file)
 static ssize_t
 fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 {
-	unsigned long p = *ppos;
 	struct fb_info *info = file_fb_info(file);
-	u8 *buffer, *dst;
-	u8 __iomem *src;
-	int c, cnt = 0, err = 0;
-	unsigned long total_size, trailing;
 
 	if (!info)
 		return -ENODEV;
@@ -777,67 +772,13 @@ fb_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	if (info->fbops->fb_read)
 		return info->fbops->fb_read(info, buf, count, ppos);
 
-	if (!info->screen_base)
-		return -ENODEV;
-
-	total_size = info->screen_size;
-
-	if (total_size == 0)
-		total_size = info->fix.smem_len;
-
-	if (p >= total_size)
-		return 0;
-
-	if (count >= total_size)
-		count = total_size;
-
-	if (count + p > total_size)
-		count = total_size - p;
-
-	buffer = kmalloc((count > PAGE_SIZE) ? PAGE_SIZE : count,
-			 GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	src = (u8 __iomem *) (info->screen_base + p);
-
-	if (info->fbops->fb_sync)
-		info->fbops->fb_sync(info);
-
-	while (count) {
-		c  = (count > PAGE_SIZE) ? PAGE_SIZE : count;
-		dst = buffer;
-		fb_memcpy_fromfb(dst, src, c);
-		dst += c;
-		src += c;
-
-		trailing = copy_to_user(buf, buffer, c);
-		if (trailing == c) {
-			err = -EFAULT;
-			break;
-		}
-		c -= trailing;
-
-		*ppos += c;
-		buf += c;
-		cnt += c;
-		count -= c;
-	}
-
-	kfree(buffer);
-
-	return cnt ? cnt : err;
+	return fb_io_read(info, buf, count, ppos);
 }
 
 static ssize_t
 fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 {
-	unsigned long p = *ppos;
 	struct fb_info *info = file_fb_info(file);
-	u8 *buffer, *src;
-	u8 __iomem *dst;
-	int c, cnt = 0, err = 0;
-	unsigned long total_size, trailing;
 
 	if (!info)
 		return -ENODEV;
@@ -848,62 +789,7 @@ fb_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 	if (info->fbops->fb_write)
 		return info->fbops->fb_write(info, buf, count, ppos);
 
-	if (!info->screen_base)
-		return -ENODEV;
-
-	total_size = info->screen_size;
-
-	if (total_size == 0)
-		total_size = info->fix.smem_len;
-
-	if (p > total_size)
-		return -EFBIG;
-
-	if (count > total_size) {
-		err = -EFBIG;
-		count = total_size;
-	}
-
-	if (count + p > total_size) {
-		if (!err)
-			err = -ENOSPC;
-
-		count = total_size - p;
-	}
-
-	buffer = kmalloc((count > PAGE_SIZE) ? PAGE_SIZE : count,
-			 GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	dst = (u8 __iomem *) (info->screen_base + p);
-
-	if (info->fbops->fb_sync)
-		info->fbops->fb_sync(info);
-
-	while (count) {
-		c = (count > PAGE_SIZE) ? PAGE_SIZE : count;
-		src = buffer;
-
-		trailing = copy_from_user(src, buf, c);
-		if (trailing == c) {
-			err = -EFAULT;
-			break;
-		}
-		c -= trailing;
-
-		fb_memcpy_tofb(dst, src, c);
-		dst += c;
-		src += c;
-		*ppos += c;
-		buf += c;
-		cnt += c;
-		count -= c;
-	}
-
-	kfree(buffer);
-
-	return (cnt) ? cnt : err;
+	return fb_io_write(info, buf, count, ppos);
 }
 
 int
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 08cb47da71f8b..ec978a4969a97 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -576,9 +576,19 @@ struct fb_info {
 extern int fb_set_var(struct fb_info *info, struct fb_var_screeninfo *var);
 extern int fb_pan_display(struct fb_info *info, struct fb_var_screeninfo *var);
 extern int fb_blank(struct fb_info *info, int blank);
+
+/*
+ * Drawing operations where framebuffer is in I/O memory
+ */
+
 extern void cfb_fillrect(struct fb_info *info, const struct fb_fillrect *rect);
 extern void cfb_copyarea(struct fb_info *info, const struct fb_copyarea *area);
 extern void cfb_imageblit(struct fb_info *info, const struct fb_image *image);
+extern ssize_t fb_io_read(struct fb_info *info, char __user *buf,
+			  size_t count, loff_t *ppos);
+extern ssize_t fb_io_write(struct fb_info *info, const char __user *buf,
+			   size_t count, loff_t *ppos);
+
 /*
  * Drawing operations where framebuffer is in system RAM
  */
-- 
GitLab


From 04408952fe2aa60db266cfa8f6db22a6ff3e1971 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 28 Apr 2023 14:24:52 +0200
Subject: [PATCH 0456/1791] drm/fb-helper: Use fb_{cfb,sys}_{read, write}()

Implement DRM fbdev helpers for reading and writing framebuffer
memory with the respective fbdev functions. Removes duplicate
code.

v2:
	* rename fb_cfb_() to fb_io_() (Geert)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Helge Deller <deller@gmx.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230428122452.4856-20-tzimmermann@suse.de
---
 drivers/gpu/drm/drm_fb_helper.c | 174 +-------------------------------
 1 file changed, 4 insertions(+), 170 deletions(-)

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 64458982be40c..933fc78c373bd 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -706,95 +706,6 @@ void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagerefli
 }
 EXPORT_SYMBOL(drm_fb_helper_deferred_io);
 
-typedef ssize_t (*drm_fb_helper_read_screen)(struct fb_info *info, char __user *buf,
-					     size_t count, loff_t pos);
-
-static ssize_t __drm_fb_helper_read(struct fb_info *info, char __user *buf, size_t count,
-				    loff_t *ppos, drm_fb_helper_read_screen read_screen)
-{
-	loff_t pos = *ppos;
-	size_t total_size;
-	ssize_t ret;
-
-	if (info->screen_size)
-		total_size = info->screen_size;
-	else
-		total_size = info->fix.smem_len;
-
-	if (pos >= total_size)
-		return 0;
-	if (count >= total_size)
-		count = total_size;
-	if (total_size - count < pos)
-		count = total_size - pos;
-
-	if (info->fbops->fb_sync)
-		info->fbops->fb_sync(info);
-
-	ret = read_screen(info, buf, count, pos);
-	if (ret > 0)
-		*ppos += ret;
-
-	return ret;
-}
-
-typedef ssize_t (*drm_fb_helper_write_screen)(struct fb_info *info, const char __user *buf,
-					      size_t count, loff_t pos);
-
-static ssize_t __drm_fb_helper_write(struct fb_info *info, const char __user *buf, size_t count,
-				     loff_t *ppos, drm_fb_helper_write_screen write_screen)
-{
-	loff_t pos = *ppos;
-	size_t total_size;
-	ssize_t ret;
-	int err = 0;
-
-	if (info->screen_size)
-		total_size = info->screen_size;
-	else
-		total_size = info->fix.smem_len;
-
-	if (pos > total_size)
-		return -EFBIG;
-	if (count > total_size) {
-		err = -EFBIG;
-		count = total_size;
-	}
-	if (total_size - count < pos) {
-		if (!err)
-			err = -ENOSPC;
-		count = total_size - pos;
-	}
-
-	if (info->fbops->fb_sync)
-		info->fbops->fb_sync(info);
-
-	/*
-	 * Copy to framebuffer even if we already logged an error. Emulates
-	 * the behavior of the original fbdev implementation.
-	 */
-	ret = write_screen(info, buf, count, pos);
-	if (ret < 0)
-		return ret; /* return last error, if any */
-	else if (!ret)
-		return err; /* return previous error, if any */
-
-	*ppos += ret;
-
-	return ret;
-}
-
-static ssize_t drm_fb_helper_read_screen_buffer(struct fb_info *info, char __user *buf,
-						size_t count, loff_t pos)
-{
-	const char *src = info->screen_buffer + pos;
-
-	if (copy_to_user(buf, src, count))
-		return -EFAULT;
-
-	return count;
-}
-
 /**
  * drm_fb_helper_sys_read - Implements struct &fb_ops.fb_read for system memory
  * @info: fb_info struct pointer
@@ -808,21 +719,10 @@ static ssize_t drm_fb_helper_read_screen_buffer(struct fb_info *info, char __use
 ssize_t drm_fb_helper_sys_read(struct fb_info *info, char __user *buf,
 			       size_t count, loff_t *ppos)
 {
-	return __drm_fb_helper_read(info, buf, count, ppos, drm_fb_helper_read_screen_buffer);
+	return fb_sys_read(info, buf, count, ppos);
 }
 EXPORT_SYMBOL(drm_fb_helper_sys_read);
 
-static ssize_t drm_fb_helper_write_screen_buffer(struct fb_info *info, const char __user *buf,
-						 size_t count, loff_t pos)
-{
-	char *dst = info->screen_buffer + pos;
-
-	if (copy_from_user(dst, buf, count))
-		return -EFAULT;
-
-	return count;
-}
-
 /**
  * drm_fb_helper_sys_write - Implements struct &fb_ops.fb_write for system memory
  * @info: fb_info struct pointer
@@ -841,7 +741,7 @@ ssize_t drm_fb_helper_sys_write(struct fb_info *info, const char __user *buf,
 	ssize_t ret;
 	struct drm_rect damage_area;
 
-	ret = __drm_fb_helper_write(info, buf, count, ppos, drm_fb_helper_write_screen_buffer);
+	ret = fb_sys_write(info, buf, count, ppos);
 	if (ret <= 0)
 		return ret;
 
@@ -913,39 +813,6 @@ void drm_fb_helper_sys_imageblit(struct fb_info *info,
 }
 EXPORT_SYMBOL(drm_fb_helper_sys_imageblit);
 
-static ssize_t fb_read_screen_base(struct fb_info *info, char __user *buf, size_t count,
-				   loff_t pos)
-{
-	const char __iomem *src = info->screen_base + pos;
-	size_t alloc_size = min_t(size_t, count, PAGE_SIZE);
-	ssize_t ret = 0;
-	int err = 0;
-	char *tmp;
-
-	tmp = kmalloc(alloc_size, GFP_KERNEL);
-	if (!tmp)
-		return -ENOMEM;
-
-	while (count) {
-		size_t c = min_t(size_t, count, alloc_size);
-
-		memcpy_fromio(tmp, src, c);
-		if (copy_to_user(buf, tmp, c)) {
-			err = -EFAULT;
-			break;
-		}
-
-		src += c;
-		buf += c;
-		ret += c;
-		count -= c;
-	}
-
-	kfree(tmp);
-
-	return ret ? ret : err;
-}
-
 /**
  * drm_fb_helper_cfb_read - Implements struct &fb_ops.fb_read for I/O memory
  * @info: fb_info struct pointer
@@ -959,43 +826,10 @@ static ssize_t fb_read_screen_base(struct fb_info *info, char __user *buf, size_
 ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
 			       size_t count, loff_t *ppos)
 {
-	return __drm_fb_helper_read(info, buf, count, ppos, fb_read_screen_base);
+	return fb_io_read(info, buf, count, ppos);
 }
 EXPORT_SYMBOL(drm_fb_helper_cfb_read);
 
-static ssize_t fb_write_screen_base(struct fb_info *info, const char __user *buf, size_t count,
-				    loff_t pos)
-{
-	char __iomem *dst = info->screen_base + pos;
-	size_t alloc_size = min_t(size_t, count, PAGE_SIZE);
-	ssize_t ret = 0;
-	int err = 0;
-	u8 *tmp;
-
-	tmp = kmalloc(alloc_size, GFP_KERNEL);
-	if (!tmp)
-		return -ENOMEM;
-
-	while (count) {
-		size_t c = min_t(size_t, count, alloc_size);
-
-		if (copy_from_user(tmp, buf, c)) {
-			err = -EFAULT;
-			break;
-		}
-		memcpy_toio(dst, tmp, c);
-
-		dst += c;
-		buf += c;
-		ret += c;
-		count -= c;
-	}
-
-	kfree(tmp);
-
-	return ret ? ret : err;
-}
-
 /**
  * drm_fb_helper_cfb_write - Implements struct &fb_ops.fb_write for I/O memory
  * @info: fb_info struct pointer
@@ -1014,7 +848,7 @@ ssize_t drm_fb_helper_cfb_write(struct fb_info *info, const char __user *buf,
 	ssize_t ret;
 	struct drm_rect damage_area;
 
-	ret = __drm_fb_helper_write(info, buf, count, ppos, fb_write_screen_base);
+	ret = fb_io_write(info, buf, count, ppos);
 	if (ret <= 0)
 		return ret;
 
-- 
GitLab


From 6a98a6e40fdf0e105b5f25a58641f30901acdb34 Mon Sep 17 00:00:00 2001
From: Brandon Pollack <brpol@chromium.org>
Date: Tue, 25 Apr 2023 08:02:40 +0000
Subject: [PATCH 0457/1791] Documentation: vkms: clarify devres managed
 reference cleanup

added documentation to drm_dev_unregister clarifying that devres managed
devices allocated with devm_drm_dev_alloc do not require calls to
drm_dev_put.

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Brandon Pollack <brpol@chromium.org>
Signed-off-by: Sean Paul <seanpaul@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230425080240.3582324-1-brpol@chromium.org
---
 drivers/gpu/drm/drm_drv.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index cee0cc522ed9b..12687dd9e1ac5 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -969,7 +969,9 @@ EXPORT_SYMBOL(drm_dev_register);
  *
  * Unregister the DRM device from the system. This does the reverse of
  * drm_dev_register() but does not deallocate the device. The caller must call
- * drm_dev_put() to drop their final reference.
+ * drm_dev_put() to drop their final reference, unless it is managed with devres
+ * (as devices allocated with devm_drm_dev_alloc() are), in which case there is
+ * already an unwind action registered.
  *
  * A special form of unregistering for hotpluggable devices is drm_dev_unplug(),
  * which can be called while there are still open users of @dev.
-- 
GitLab


From 5b8ff071b57e0a50a0e6eb8db55d159a68853f6b Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Fri, 5 May 2023 16:45:44 -0700
Subject: [PATCH 0458/1791] drm/i915/mtl: Fix the wa number for Wa_22016670082

Fixes the right lineage number for the workaround.

Fixes: a7fa1537b791 ("drm/i915/mtl: Implement Wa_14019141245")
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505234544.4029535-1-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index ad9e7f49a6fa4..786349e954873 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -1699,7 +1699,7 @@ xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
 	wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB);
 	wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB);
 
-	/* Wa_14019141245 */
+	/* Wa_22016670082 */
 	wa_write_or(wal, GEN12_SQCNT1, GEN12_STRICT_RAR_ENABLE);
 
 	if (IS_MTL_GRAPHICS_STEP(gt->i915, M, STEP_A0, STEP_B0) ||
-- 
GitLab


From 6fa9df2b5e563674e0b22d740e27b0df8960ba1c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 4 May 2023 15:34:44 +0300
Subject: [PATCH 0459/1791] drm/connector: document enum drm_connector_tv_mode
 DRM_MODE_TV_MODE_MAX

Add documentation for the DRM_MODE_TV_MODE_MAX enumerator to fix the
kernel-doc warning:

include/drm/drm_connector.h:204: warning: Enum value 'DRM_MODE_TV_MODE_MAX' not described in enum 'drm_connector_tv_mode'

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504123444.1843795-1-jani.nikula@intel.com
---
 include/drm/drm_connector.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index 7b5048516185c..e6478fafa6b04 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -199,6 +199,11 @@ enum drm_connector_tv_mode {
 	 */
 	DRM_MODE_TV_MODE_SECAM,
 
+	/**
+	 * @DRM_MODE_TV_MODE_MAX: Number of analog TV output modes.
+	 *
+	 * Internal implementation detail; this is not uABI.
+	 */
 	DRM_MODE_TV_MODE_MAX,
 };
 
-- 
GitLab


From 3312bb4ad09ca6423bd4a5b15a94588a8962fb8e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 4 May 2023 13:35:08 +0300
Subject: [PATCH 0460/1791] drm/i915: taint kernel when force probing
 unsupported devices

For development and testing purposes, the i915.force_probe module
parameter and DRM_I915_FORCE_PROBE kconfig option allow probing of
devices that aren't supported by the driver.

The i915.force_probe module parameter is "unsafe" and setting it taints
the kernel. However, using the kconfig option does not.

Always taint the kernel when force probing a device that is not
supported.

v2: Drop "depends on EXPERT" to avoid build breakage (kernel test robot)

Fixes: 7ef5ef5cdead ("drm/i915: add force_probe module parameter to replace alpha_support")
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Dave Airlie <airlied@gmail.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230504103508.1818540-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Kconfig    | 12 +++++++-----
 drivers/gpu/drm/i915/i915_pci.c |  6 ++++++
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index 06a0ca157e89e..e4f4d2e3fdfeb 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -62,10 +62,11 @@ config DRM_I915_FORCE_PROBE
 	  This is the default value for the i915.force_probe module
 	  parameter. Using the module parameter overrides this option.
 
-	  Force probe the i915 for Intel graphics devices that are
-	  recognized but not properly supported by this kernel version. It is
-	  recommended to upgrade to a kernel version with proper support as soon
-	  as it is available.
+	  Force probe the i915 driver for Intel graphics devices that are
+	  recognized but not properly supported by this kernel version. Force
+	  probing an unsupported device taints the kernel. It is recommended to
+	  upgrade to a kernel version with proper support as soon as it is
+	  available.
 
 	  It can also be used to block the probe of recognized and fully
 	  supported devices.
@@ -75,7 +76,8 @@ config DRM_I915_FORCE_PROBE
 	  Use "<pci-id>[,<pci-id>,...]" to force probe the i915 for listed
 	  devices. For example, "4500" or "4500,4571".
 
-	  Use "*" to force probe the driver for all known devices.
+	  Use "*" to force probe the driver for all known devices. Not
+	  recommended.
 
 	  Use "!" right before the ID to block the probe of the device. For
 	  example, "4500,!4571" forces the probe of 4500 and blocks the probe of
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 6b5c119a7d829..ccc315782d22b 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1345,6 +1345,12 @@ static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENODEV;
 	}
 
+	if (intel_info->require_force_probe) {
+		dev_info(&pdev->dev, "Force probing unsupported Device ID %04x, tainting kernel\n",
+			 pdev->device);
+		add_taint(TAINT_USER, LOCKDEP_STILL_OK);
+	}
+
 	/* Only bind to function 0 of the device. Early generations
 	 * used function 1 as a placeholder for multi-head. This causes
 	 * us confusion instead, especially on the systems where both
-- 
GitLab


From 79c87edd18ec49f5b6fb40175bd1b1fea9398fdb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mairacanal@riseup.net>
Date: Mon, 8 May 2023 11:10:39 -0300
Subject: [PATCH 0461/1791] MAINTAINERS: Add Maira to VKMS maintainers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I've been contributing to VKMS with improvements, reviews, testing and
debugging. Therefore, add myself as a co-maintainer of the VKMS driver.

Acked-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230508141038.327160-1-mairacanal@riseup.net
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 276298cfc7ee4..63e9b1e8c1132 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6721,6 +6721,7 @@ F:	drivers/gpu/drm/udl/
 DRM DRIVER FOR VIRTUAL KERNEL MODESETTING (VKMS)
 M:	Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>
 M:	Melissa Wen <melissa.srw@gmail.com>
+M:	Maíra Canal <mairacanal@riseup.net>
 R:	Haneen Mohammed <hamohammed.sa@gmail.com>
 R:	Daniel Vetter <daniel@ffwll.ch>
 L:	dri-devel@lists.freedesktop.org
-- 
GitLab


From 71ffeafb13380b73f41bc16495962371c0fda596 Mon Sep 17 00:00:00 2001
From: Roman Beranek <me@crly.cz>
Date: Fri, 5 May 2023 07:21:09 +0200
Subject: [PATCH 0462/1791] drm: sun4i: rename sun4i_dotclock to
 sun4i_tcon_dclk

While the rate of TCON0's DCLK matches dotclock for parallel and LVDS
outputs, this doesn't hold for DSI. The 'D' in DCLK actually stands for
'Data' according to Allwinner's manuals. The clock is mostly referred to
as dclk throughout this driver already anyway, so stick with that.

Signed-off-by: Roman Beranek <me@crly.cz>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505052110.67514-4-me@crly.cz
---
 drivers/gpu/drm/sun4i/Makefile                         |  2 +-
 drivers/gpu/drm/sun4i/sun4i_tcon.c                     | 10 +++++-----
 .../drm/sun4i/{sun4i_dotclock.c => sun4i_tcon_dclk.c}  |  2 +-
 .../drm/sun4i/{sun4i_dotclock.h => sun4i_tcon_dclk.h}  |  0
 4 files changed, 7 insertions(+), 7 deletions(-)
 rename drivers/gpu/drm/sun4i/{sun4i_dotclock.c => sun4i_tcon_dclk.c} (99%)
 rename drivers/gpu/drm/sun4i/{sun4i_dotclock.h => sun4i_tcon_dclk.h} (100%)

diff --git a/drivers/gpu/drm/sun4i/Makefile b/drivers/gpu/drm/sun4i/Makefile
index 0d04f2447b01c..bad7497a0d11e 100644
--- a/drivers/gpu/drm/sun4i/Makefile
+++ b/drivers/gpu/drm/sun4i/Makefile
@@ -19,7 +19,7 @@ sun8i-mixer-y			+= sun8i_mixer.o sun8i_ui_layer.o \
 				   sun8i_vi_scaler.o sun8i_csc.o
 
 sun4i-tcon-y			+= sun4i_crtc.o
-sun4i-tcon-y			+= sun4i_dotclock.o
+sun4i-tcon-y			+= sun4i_tcon_dclk.o
 sun4i-tcon-y			+= sun4i_lvds.o
 sun4i-tcon-y			+= sun4i_tcon.o
 sun4i-tcon-y			+= sun4i_rgb.o
diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index 523a6d7879210..eec26b1faa4ba 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -31,12 +31,12 @@
 #include <uapi/drm/drm_mode.h>
 
 #include "sun4i_crtc.h"
-#include "sun4i_dotclock.h"
 #include "sun4i_drv.h"
 #include "sun4i_lvds.h"
 #include "sun4i_rgb.h"
 #include "sun4i_tcon.h"
 #include "sun6i_mipi_dsi.h"
+#include "sun4i_tcon_dclk.h"
 #include "sun8i_tcon_top.h"
 #include "sunxi_engine.h"
 
@@ -1237,14 +1237,14 @@ static int sun4i_tcon_bind(struct device *dev, struct device *master,
 	ret = sun4i_tcon_init_irq(dev, tcon);
 	if (ret) {
 		dev_err(dev, "Couldn't init our TCON interrupts\n");
-		goto err_free_dotclock;
+		goto err_free_dclk;
 	}
 
 	tcon->crtc = sun4i_crtc_init(drm, engine, tcon);
 	if (IS_ERR(tcon->crtc)) {
 		dev_err(dev, "Couldn't create our CRTC\n");
 		ret = PTR_ERR(tcon->crtc);
-		goto err_free_dotclock;
+		goto err_free_dclk;
 	}
 
 	if (tcon->quirks->has_channel_0) {
@@ -1264,7 +1264,7 @@ static int sun4i_tcon_bind(struct device *dev, struct device *master,
 		of_node_put(remote);
 
 		if (ret < 0)
-			goto err_free_dotclock;
+			goto err_free_dclk;
 	}
 
 	if (tcon->quirks->needs_de_be_mux) {
@@ -1290,7 +1290,7 @@ static int sun4i_tcon_bind(struct device *dev, struct device *master,
 
 	return 0;
 
-err_free_dotclock:
+err_free_dclk:
 	if (tcon->quirks->has_channel_0)
 		sun4i_dclk_free(tcon);
 err_free_clocks:
diff --git a/drivers/gpu/drm/sun4i/sun4i_dotclock.c b/drivers/gpu/drm/sun4i/sun4i_tcon_dclk.c
similarity index 99%
rename from drivers/gpu/drm/sun4i/sun4i_dotclock.c
rename to drivers/gpu/drm/sun4i/sun4i_tcon_dclk.c
index 417ade3d2565d..03d7de1911cd6 100644
--- a/drivers/gpu/drm/sun4i/sun4i_dotclock.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon_dclk.c
@@ -10,7 +10,7 @@
 #include <linux/regmap.h>
 
 #include "sun4i_tcon.h"
-#include "sun4i_dotclock.h"
+#include "sun4i_tcon_dclk.h"
 
 struct sun4i_dclk {
 	struct clk_hw		hw;
diff --git a/drivers/gpu/drm/sun4i/sun4i_dotclock.h b/drivers/gpu/drm/sun4i/sun4i_tcon_dclk.h
similarity index 100%
rename from drivers/gpu/drm/sun4i/sun4i_dotclock.h
rename to drivers/gpu/drm/sun4i/sun4i_tcon_dclk.h
-- 
GitLab


From 4795c78768bcbd58d4ffab650674d314dc6dd772 Mon Sep 17 00:00:00 2001
From: Roman Beranek <me@crly.cz>
Date: Fri, 5 May 2023 07:21:10 +0200
Subject: [PATCH 0463/1791] drm: sun4i: calculate proper DCLK rate for DSI

In DSI mode, TCON0's data clock is required to run at 1/4 the per-lane
bit rate.

Signed-off-by: Roman Beranek <me@crly.cz>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505052110.67514-5-me@crly.cz
---
 drivers/gpu/drm/sun4i/sun4i_tcon.c | 36 +++++++++++++++++-------------
 1 file changed, 21 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index eec26b1faa4ba..b263de7a82375 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -291,18 +291,6 @@ static int sun4i_tcon_get_clk_delay(const struct drm_display_mode *mode,
 	return delay;
 }
 
-static void sun4i_tcon0_mode_set_common(struct sun4i_tcon *tcon,
-					const struct drm_display_mode *mode)
-{
-	/* Configure the dot clock */
-	clk_set_rate(tcon->dclk, mode->crtc_clock * 1000);
-
-	/* Set the resolution */
-	regmap_write(tcon->regs, SUN4I_TCON0_BASIC0_REG,
-		     SUN4I_TCON0_BASIC0_X(mode->crtc_hdisplay) |
-		     SUN4I_TCON0_BASIC0_Y(mode->crtc_vdisplay));
-}
-
 static void sun4i_tcon0_mode_set_dithering(struct sun4i_tcon *tcon,
 					   const struct drm_connector *connector)
 {
@@ -367,10 +355,18 @@ static void sun4i_tcon0_mode_set_cpu(struct sun4i_tcon *tcon,
 	u32 block_space, start_delay;
 	u32 tcon_div;
 
+	/*
+	 * dclk is required to run at 1/4 the DSI per-lane bit rate.
+	 */
 	tcon->dclk_min_div = SUN6I_DSI_TCON_DIV;
 	tcon->dclk_max_div = SUN6I_DSI_TCON_DIV;
+	clk_set_rate(tcon->dclk, mode->crtc_clock * 1000 * (bpp / lanes)
+						  / SUN6I_DSI_TCON_DIV);
 
-	sun4i_tcon0_mode_set_common(tcon, mode);
+	/* Set the resolution */
+	regmap_write(tcon->regs, SUN4I_TCON0_BASIC0_REG,
+		     SUN4I_TCON0_BASIC0_X(mode->crtc_hdisplay) |
+		     SUN4I_TCON0_BASIC0_Y(mode->crtc_vdisplay));
 
 	/* Set dithering if needed */
 	sun4i_tcon0_mode_set_dithering(tcon, sun4i_tcon_get_connector(encoder));
@@ -438,7 +434,12 @@ static void sun4i_tcon0_mode_set_lvds(struct sun4i_tcon *tcon,
 
 	tcon->dclk_min_div = 7;
 	tcon->dclk_max_div = 7;
-	sun4i_tcon0_mode_set_common(tcon, mode);
+	clk_set_rate(tcon->dclk, mode->crtc_clock * 1000);
+
+	/* Set the resolution */
+	regmap_write(tcon->regs, SUN4I_TCON0_BASIC0_REG,
+		     SUN4I_TCON0_BASIC0_X(mode->crtc_hdisplay) |
+		     SUN4I_TCON0_BASIC0_Y(mode->crtc_vdisplay));
 
 	/* Set dithering if needed */
 	sun4i_tcon0_mode_set_dithering(tcon, sun4i_tcon_get_connector(encoder));
@@ -515,7 +516,12 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon,
 
 	tcon->dclk_min_div = tcon->quirks->dclk_min_div;
 	tcon->dclk_max_div = 127;
-	sun4i_tcon0_mode_set_common(tcon, mode);
+	clk_set_rate(tcon->dclk, mode->crtc_clock * 1000);
+
+	/* Set the resolution */
+	regmap_write(tcon->regs, SUN4I_TCON0_BASIC0_REG,
+		     SUN4I_TCON0_BASIC0_X(mode->crtc_hdisplay) |
+		     SUN4I_TCON0_BASIC0_Y(mode->crtc_vdisplay));
 
 	/* Set dithering if needed */
 	sun4i_tcon0_mode_set_dithering(tcon, connector);
-- 
GitLab


From 123ee07ba5b7123e0ce0e0f9d64938026c16a2ce Mon Sep 17 00:00:00 2001
From: XuDong Liu <m202071377@hust.edu.cn>
Date: Sun, 30 Apr 2023 19:23:46 +0800
Subject: [PATCH 0464/1791] drm: sun4i_tcon: use devm_clk_get_enabled in
 `sun4i_tcon_init_clocks`

Smatch reports:
drivers/gpu/drm/sun4i/sun4i_tcon.c:805 sun4i_tcon_init_clocks() warn:
'tcon->clk' from clk_prepare_enable() not released on lines: 792,801.

In the function sun4i_tcon_init_clocks(), tcon->clk and tcon->sclk0 are
not disabled in the error handling, which affects the release of
these variable. Although sun4i_tcon_bind(), which calls
sun4i_tcon_init_clocks(), use sun4i_tcon_free_clocks to disable the
variables mentioned, but the error handling branch of
sun4i_tcon_init_clocks() ignores the required disable process.

To fix this issue, use the devm_clk_get_enabled to automatically
balance enable and disabled calls. As original implementation use
sun4i_tcon_free_clocks() to disable clk explicitly, we delete the
related calls and error handling that are no longer needed.

Fixes: 9026e0d122ac ("drm: Add Allwinner A10 Display Engine support")
Fixes: b14e945bda8a ("drm/sun4i: tcon: Prepare and enable TCON channel 0 clock at init")
Fixes: 8e9240472522 ("drm/sun4i: support TCONs without channel 1")
Fixes: 34d698f6e349 ("drm/sun4i: Add has_channel_0 TCON quirk")
Signed-off-by: XuDong Liu <m202071377@hust.edu.cn>
Reviewed-by: Dongliang Mu <dzm91@hust.edu.cn>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20230430112347.4689-1-m202071377@hust.edu.cn
---
 drivers/gpu/drm/sun4i/sun4i_tcon.c | 19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c
index b263de7a82375..6a52fb12cbfbc 100644
--- a/drivers/gpu/drm/sun4i/sun4i_tcon.c
+++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c
@@ -784,21 +784,19 @@ static irqreturn_t sun4i_tcon_handler(int irq, void *private)
 static int sun4i_tcon_init_clocks(struct device *dev,
 				  struct sun4i_tcon *tcon)
 {
-	tcon->clk = devm_clk_get(dev, "ahb");
+	tcon->clk = devm_clk_get_enabled(dev, "ahb");
 	if (IS_ERR(tcon->clk)) {
 		dev_err(dev, "Couldn't get the TCON bus clock\n");
 		return PTR_ERR(tcon->clk);
 	}
-	clk_prepare_enable(tcon->clk);
 
 	if (tcon->quirks->has_channel_0) {
-		tcon->sclk0 = devm_clk_get(dev, "tcon-ch0");
+		tcon->sclk0 = devm_clk_get_enabled(dev, "tcon-ch0");
 		if (IS_ERR(tcon->sclk0)) {
 			dev_err(dev, "Couldn't get the TCON channel 0 clock\n");
 			return PTR_ERR(tcon->sclk0);
 		}
 	}
-	clk_prepare_enable(tcon->sclk0);
 
 	if (tcon->quirks->has_channel_1) {
 		tcon->sclk1 = devm_clk_get(dev, "tcon-ch1");
@@ -811,12 +809,6 @@ static int sun4i_tcon_init_clocks(struct device *dev,
 	return 0;
 }
 
-static void sun4i_tcon_free_clocks(struct sun4i_tcon *tcon)
-{
-	clk_disable_unprepare(tcon->sclk0);
-	clk_disable_unprepare(tcon->clk);
-}
-
 static int sun4i_tcon_init_irq(struct device *dev,
 			       struct sun4i_tcon *tcon)
 {
@@ -1229,14 +1221,14 @@ static int sun4i_tcon_bind(struct device *dev, struct device *master,
 	ret = sun4i_tcon_init_regmap(dev, tcon);
 	if (ret) {
 		dev_err(dev, "Couldn't init our TCON regmap\n");
-		goto err_free_clocks;
+		goto err_assert_reset;
 	}
 
 	if (tcon->quirks->has_channel_0) {
 		ret = sun4i_dclk_create(dev, tcon);
 		if (ret) {
 			dev_err(dev, "Couldn't create our TCON dot clock\n");
-			goto err_free_clocks;
+			goto err_assert_reset;
 		}
 	}
 
@@ -1299,8 +1291,6 @@ static int sun4i_tcon_bind(struct device *dev, struct device *master,
 err_free_dclk:
 	if (tcon->quirks->has_channel_0)
 		sun4i_dclk_free(tcon);
-err_free_clocks:
-	sun4i_tcon_free_clocks(tcon);
 err_assert_reset:
 	reset_control_assert(tcon->lcd_rst);
 	return ret;
@@ -1314,7 +1304,6 @@ static void sun4i_tcon_unbind(struct device *dev, struct device *master,
 	list_del(&tcon->list);
 	if (tcon->quirks->has_channel_0)
 		sun4i_dclk_free(tcon);
-	sun4i_tcon_free_clocks(tcon);
 }
 
 static const struct component_ops sun4i_tcon_ops = {
-- 
GitLab


From dee23b2c9e3ff46d59c5d45e1436eceb878e7c9a Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 7 May 2023 20:26:38 +0300
Subject: [PATCH 0465/1791] drm/panel: sharp-ls043t1le01: adjust mode settings

Using current settings causes panel flickering on APQ8074 dragonboard.
Adjust panel settings to follow the vendor-provided mode. This also
enables MIPI_DSI_MODE_VIDEO_SYNC_PULSE, which is also specified by the
vendor dtsi for the mentioned dragonboard.

Fixes: ee0172383190 ("drm/panel: Add Sharp LS043T1LE01 MIPI DSI panel")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230507172639.2320934-1-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
index d1ec80a3e3c72..ef148504cf24a 100644
--- a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
@@ -192,15 +192,15 @@ static int sharp_nt_panel_enable(struct drm_panel *panel)
 }
 
 static const struct drm_display_mode default_mode = {
-	.clock = 41118,
+	.clock = (540 + 48 + 32 + 80) * (960 + 3 + 10 + 15) * 60 / 1000,
 	.hdisplay = 540,
 	.hsync_start = 540 + 48,
-	.hsync_end = 540 + 48 + 80,
-	.htotal = 540 + 48 + 80 + 32,
+	.hsync_end = 540 + 48 + 32,
+	.htotal = 540 + 48 + 32 + 80,
 	.vdisplay = 960,
 	.vsync_start = 960 + 3,
-	.vsync_end = 960 + 3 + 15,
-	.vtotal = 960 + 3 + 15 + 1,
+	.vsync_end = 960 + 3 + 10,
+	.vtotal = 960 + 3 + 10 + 15,
 };
 
 static int sharp_nt_panel_get_modes(struct drm_panel *panel,
@@ -280,6 +280,7 @@ static int sharp_nt_panel_probe(struct mipi_dsi_device *dsi)
 	dsi->lanes = 2;
 	dsi->format = MIPI_DSI_FMT_RGB888;
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO |
+			MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
 			MIPI_DSI_MODE_VIDEO_HSE |
 			MIPI_DSI_CLOCK_NON_CONTINUOUS |
 			MIPI_DSI_MODE_NO_EOT_PACKET;
-- 
GitLab


From 7e8ba34d357127e2c93f18123d09b5c817156512 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 7 May 2023 20:26:39 +0300
Subject: [PATCH 0466/1791] drm/panel: sharp-ls043t1le01: drop dummy functions
 and data fields

Drop sharp_nt_panel_disable() and sharp_nt_panel_enable(), which bear no
useful code. Also drop sharp_nt_panel::enable and sharp_nt_panel::mode
fields which also provide no use now.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230507172639.2320934-2-dmitry.baryshkov@linaro.org
---
 .../gpu/drm/panel/panel-sharp-ls043t1le01.c   | 32 -------------------
 1 file changed, 32 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
index ef148504cf24a..855e64444daa6 100644
--- a/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
+++ b/drivers/gpu/drm/panel/panel-sharp-ls043t1le01.c
@@ -28,9 +28,6 @@ struct sharp_nt_panel {
 	struct gpio_desc *reset_gpio;
 
 	bool prepared;
-	bool enabled;
-
-	const struct drm_display_mode *mode;
 };
 
 static inline struct sharp_nt_panel *to_sharp_nt_panel(struct drm_panel *panel)
@@ -97,19 +94,6 @@ static int sharp_nt_panel_off(struct sharp_nt_panel *sharp_nt)
 	return 0;
 }
 
-
-static int sharp_nt_panel_disable(struct drm_panel *panel)
-{
-	struct sharp_nt_panel *sharp_nt = to_sharp_nt_panel(panel);
-
-	if (!sharp_nt->enabled)
-		return 0;
-
-	sharp_nt->enabled = false;
-
-	return 0;
-}
-
 static int sharp_nt_panel_unprepare(struct drm_panel *panel)
 {
 	struct sharp_nt_panel *sharp_nt = to_sharp_nt_panel(panel);
@@ -179,18 +163,6 @@ static int sharp_nt_panel_prepare(struct drm_panel *panel)
 	return ret;
 }
 
-static int sharp_nt_panel_enable(struct drm_panel *panel)
-{
-	struct sharp_nt_panel *sharp_nt = to_sharp_nt_panel(panel);
-
-	if (sharp_nt->enabled)
-		return 0;
-
-	sharp_nt->enabled = true;
-
-	return 0;
-}
-
 static const struct drm_display_mode default_mode = {
 	.clock = (540 + 48 + 32 + 80) * (960 + 3 + 10 + 15) * 60 / 1000,
 	.hdisplay = 540,
@@ -227,10 +199,8 @@ static int sharp_nt_panel_get_modes(struct drm_panel *panel,
 }
 
 static const struct drm_panel_funcs sharp_nt_panel_funcs = {
-	.disable = sharp_nt_panel_disable,
 	.unprepare = sharp_nt_panel_unprepare,
 	.prepare = sharp_nt_panel_prepare,
-	.enable = sharp_nt_panel_enable,
 	.get_modes = sharp_nt_panel_get_modes,
 };
 
@@ -239,8 +209,6 @@ static int sharp_nt_panel_add(struct sharp_nt_panel *sharp_nt)
 	struct device *dev = &sharp_nt->dsi->dev;
 	int ret;
 
-	sharp_nt->mode = &default_mode;
-
 	sharp_nt->supply = devm_regulator_get(dev, "avdd");
 	if (IS_ERR(sharp_nt->supply))
 		return PTR_ERR(sharp_nt->supply);
-- 
GitLab


From 0bd5bd65cd2e4d1335ea6c17cd2c8664decbc630 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Mon, 8 May 2023 16:38:25 +0800
Subject: [PATCH 0467/1791] dt-bindings: display: simple: Add BOE
 EV121WXM-N10-1850 panel

Add BOE EV121WXM-N10-1850 12.1" WXGA (1280x800) TFT LCD panel
compatible string.  The panel has a LVDS display interface.

The panel's product specification can be found at:
http://www.onetech.com.tw/files/EV121WXM-N10-1850ProductSpecification_20180801.pdf

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230508083826.1016206-2-victor.liu@nxp.com
---
 .../devicetree/bindings/display/panel/panel-simple.yaml         | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index ec50dd2683149..733e47da36e80 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -77,6 +77,8 @@ properties:
       - auo,t215hvn01
         # Shanghai AVIC Optoelectronics 7" 1024x600 color TFT-LCD panel
       - avic,tm070ddh03
+        # BOE EV121WXM-N10-1850 12.1" WXGA (1280x800) TFT LCD panel
+      - boe,ev121wxm-n10-1850
         # BOE HV070WSA-100 7.01" WSVGA TFT LCD panel
       - boe,hv070wsa-100
         # BOE OPTOELECTRONICS TECHNOLOGY 10.1" WXGA TFT LCD panel
-- 
GitLab


From 8bb7c7bca5b70f3cd22d95b4d36029295c4274f6 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Mon, 8 May 2023 16:38:26 +0800
Subject: [PATCH 0468/1791] drm/panel: panel-simple: Add BOE EV121WXM-N10-1850
 panel support

Add BOE EV121WXM-N10-1850 12.1" WXGA (1280x800) TFT LCD panel support.
The panel has a LVDS display interface.

The panel's product specification can be found at:
http://www.onetech.com.tw/files/EV121WXM-N10-1850ProductSpecification_20180801.pdf

Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230508083826.1016206-3-victor.liu@nxp.com
---
 drivers/gpu/drm/panel/panel-simple.c | 34 ++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 2a9c1a785a5cb..5778824dffd47 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -1211,6 +1211,37 @@ static const struct panel_desc bananapi_s070wv20_ct16 = {
 	},
 };
 
+static const struct display_timing boe_ev121wxm_n10_1850_timing = {
+	.pixelclock = { 69922000, 71000000, 72293000 },
+	.hactive = { 1280, 1280, 1280 },
+	.hfront_porch = { 48, 48, 48 },
+	.hback_porch = { 80, 80, 80 },
+	.hsync_len = { 32, 32, 32 },
+	.vactive = { 800, 800, 800 },
+	.vfront_porch = { 3, 3, 3 },
+	.vback_porch = { 14, 14, 14 },
+	.vsync_len = { 6, 6, 6 },
+};
+
+static const struct panel_desc boe_ev121wxm_n10_1850 = {
+	.timings = &boe_ev121wxm_n10_1850_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 261,
+		.height = 163,
+	},
+	.delay = {
+		.prepare = 9,
+		.enable = 300,
+		.unprepare = 300,
+		.disable = 560,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+	.connector_type = DRM_MODE_CONNECTOR_LVDS,
+};
+
 static const struct drm_display_mode boe_hv070wsa_mode = {
 	.clock = 42105,
 	.hdisplay = 1024,
@@ -4016,6 +4047,9 @@ static const struct of_device_id platform_of_match[] = {
 	}, {
 		.compatible = "bananapi,s070wv20-ct16",
 		.data = &bananapi_s070wv20_ct16,
+	}, {
+		.compatible = "boe,ev121wxm-n10-1850",
+		.data = &boe_ev121wxm_n10_1850,
 	}, {
 		.compatible = "boe,hv070wsa-100",
 		.data = &boe_hv070wsa
-- 
GitLab


From 413d2e2fd052c57e66ebce988882fb58d33966ab Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 May 2023 11:55:53 +0300
Subject: [PATCH 0469/1791] drm/i915/hdcp: drop display/ prefix from include

The display prefix is unnecessary within the display sub-directory.

Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511085553.799321-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp_gsc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
index 7e52aea6aa177..4056bb2323ca5 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
@@ -5,11 +5,11 @@
 
 #include <drm/i915_hdcp_interface.h>
 
-#include "display/intel_hdcp_gsc.h"
 #include "gem/i915_gem_region.h"
 #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h"
 #include "i915_drv.h"
 #include "i915_utils.h"
+#include "intel_hdcp_gsc.h"
 
 bool intel_hdcp_gsc_cs_required(struct drm_i915_private *i915)
 {
-- 
GitLab


From 9f8c1fe320e8014b1d2cd0d8bfcca299223f7604 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Apr 2023 16:50:19 +0300
Subject: [PATCH 0470/1791] drm/i915: Define more PS_CTRL bits
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To avoid annoying spec lookups let's define more PS_CTRL
bits in the header.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230426135019.7603-8-ville.syrjala@linux.intel.com
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 6a82c707bf9da..6a91564964281 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4045,6 +4045,9 @@
 #define _PS_2B_CTRL      0x68A80
 #define _PS_1C_CTRL      0x69180
 #define   PS_SCALER_EN				REG_BIT(31)
+#define   PS_SCALER_TYPE_MASK			REG_BIT(30) /* icl+ */
+#define   PS_SCALER_TYPE_NON_LINEAR		REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 0)
+#define   PS_SCALER_TYPE_LINEAR			REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 1)
 #define   SKL_PS_SCALER_MODE_MASK		REG_GENMASK(29, 28) /* skl/bxt */
 #define   SKL_PS_SCALER_MODE_DYN		REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 0)
 #define   SKL_PS_SCALER_MODE_HQ			REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 1)
@@ -4052,6 +4055,7 @@
 #define   PS_SCALER_MODE_MASK			REG_BIT(29) /* glk-tgl */
 #define   PS_SCALER_MODE_NORMAL			REG_FIELD_PREP(PS_SCALER_MODE_MASK, 0)
 #define   PS_SCALER_MODE_PLANAR			REG_FIELD_PREP(PS_SCALER_MODE_MASK, 1)
+#define   PS_ADAPTIVE_FILTERING_EN		REG_BIT(28) /* icl+ */
 #define   PS_BINDING_MASK			REG_GENMASK(27, 25)
 #define   PS_BINDING_PIPE			REG_FIELD_PREP(PS_BINDING_MASK, 0)
 #define   PS_BINDING_PLANE(plane_id)		REG_FIELD_PREP(PS_BINDING_MASK, (plane_id) + 1)
@@ -4060,8 +4064,15 @@
 #define   PS_FILTER_PROGRAMMED			REG_FIELD_PREP(PS_FILTER_MASK, 1)
 #define   PS_FILTER_EDGE_ENHANCE		REG_FIELD_PREP(PS_FILTER_MASK, 2)
 #define   PS_FILTER_BILINEAR			REG_FIELD_PREP(PS_FILTER_MASK, 3)
+#define   PS_ADAPTIVE_FILTER_MASK		REG_BIT(22) /* icl+ */
+#define   PS_ADAPTIVE_FILTER_MEDIUM		REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 0)
+#define   PS_ADAPTIVE_FILTER_EDGE_ENHANCE	REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 1)
+#define   PS_PIPE_SCALER_LOC_MASK		REG_BIT(21) /* icl+ */
+#define   PS_PIPE_SCALER_LOC_AFTER_OUTPUT_CSC	REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 0) /* non-linear */
+#define   PS_PIPE_SCALER_LOC_AFTER_CSC		REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 1) /* linear */
 #define   PS_VERT3TAP				REG_BIT(21) /* skl/bxt */
 #define   PS_VERT_INT_INVERT_FIELD		REG_BIT(20)
+#define   PS_PROG_SCALE_FACTOR			REG_BIT(19) /* tgl+ */
 #define   PS_PWRUP_PROGRESS			REG_BIT(17)
 #define   PS_V_FILTER_BYPASS			REG_BIT(8)
 #define   PS_VADAPT_EN				REG_BIT(7) /* skl/bxt */
-- 
GitLab


From 0cad796a226997e176c71bd4e04a3d426ba4fcc3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 9 May 2023 20:14:11 +0300
Subject: [PATCH 0471/1791] drm/i915: Use REG_BIT() & co. for AUX CH registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Modernize the DP AUX CH register definitions with REG_BIT() & co.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509171411.7184-1-ville.syrjala@linux.intel.com
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp_aux.c   | 35 +++++------
 .../gpu/drm/i915/display/intel_dp_aux_regs.h  | 62 ++++++++++---------
 drivers/gpu/drm/i915/gvt/edid.c               | 10 +--
 3 files changed, 52 insertions(+), 55 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c
index abf77ba769728..25e36bdc4adb9 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c
@@ -161,14 +161,14 @@ static u32 g4x_get_aux_send_ctl(struct intel_dp *intel_dp,
 		timeout = DP_AUX_CH_CTL_TIME_OUT_400us;
 
 	return DP_AUX_CH_CTL_SEND_BUSY |
-	       DP_AUX_CH_CTL_DONE |
-	       DP_AUX_CH_CTL_INTERRUPT |
-	       DP_AUX_CH_CTL_TIME_OUT_ERROR |
-	       timeout |
-	       DP_AUX_CH_CTL_RECEIVE_ERROR |
-	       (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
-	       (g4x_dp_aux_precharge_len() << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
-	       (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT);
+		DP_AUX_CH_CTL_DONE |
+		DP_AUX_CH_CTL_INTERRUPT |
+		DP_AUX_CH_CTL_TIME_OUT_ERROR |
+		timeout |
+		DP_AUX_CH_CTL_RECEIVE_ERROR |
+		DP_AUX_CH_CTL_MESSAGE_SIZE(send_bytes) |
+		DP_AUX_CH_CTL_PRECHARGE_2US(g4x_dp_aux_precharge_len()) |
+		DP_AUX_CH_CTL_BIT_CLOCK_2X(aux_clock_divider);
 }
 
 static u32 skl_get_aux_send_ctl(struct intel_dp *intel_dp,
@@ -185,14 +185,14 @@ static u32 skl_get_aux_send_ctl(struct intel_dp *intel_dp,
 	 * ICL+: 4ms
 	 */
 	ret = DP_AUX_CH_CTL_SEND_BUSY |
-	      DP_AUX_CH_CTL_DONE |
-	      DP_AUX_CH_CTL_INTERRUPT |
-	      DP_AUX_CH_CTL_TIME_OUT_ERROR |
-	      DP_AUX_CH_CTL_TIME_OUT_MAX |
-	      DP_AUX_CH_CTL_RECEIVE_ERROR |
-	      (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
-	      DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(intel_dp_aux_fw_sync_len()) |
-	      DP_AUX_CH_CTL_SYNC_PULSE_SKL(intel_dp_aux_sync_len());
+		DP_AUX_CH_CTL_DONE |
+		DP_AUX_CH_CTL_INTERRUPT |
+		DP_AUX_CH_CTL_TIME_OUT_ERROR |
+		DP_AUX_CH_CTL_TIME_OUT_MAX |
+		DP_AUX_CH_CTL_RECEIVE_ERROR |
+		DP_AUX_CH_CTL_MESSAGE_SIZE(send_bytes) |
+		DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(intel_dp_aux_fw_sync_len()) |
+		DP_AUX_CH_CTL_SYNC_PULSE_SKL(intel_dp_aux_sync_len());
 
 	if (intel_tc_port_in_tbt_alt_mode(dig_port))
 		ret |= DP_AUX_CH_CTL_TBT_IO;
@@ -378,8 +378,7 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp,
 	}
 
 	/* Unload any bytes sent back from the other side */
-	recv_bytes = ((status & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >>
-		      DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT);
+	recv_bytes = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, status);
 
 	/*
 	 * By BSpec: "Message sizes of 0 or >20 are not allowed."
diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h b/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h
index 5702f318d5378..5185345277c7c 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h
@@ -50,35 +50,37 @@
 						       _XELPDP_USBC3_AUX_CH_DATA1, \
 						       _XELPDP_USBC4_AUX_CH_DATA1) + (i) * 4)
 
-#define   DP_AUX_CH_CTL_SEND_BUSY	    (1 << 31)
-#define   DP_AUX_CH_CTL_DONE		    (1 << 30)
-#define   DP_AUX_CH_CTL_INTERRUPT	    (1 << 29)
-#define   DP_AUX_CH_CTL_TIME_OUT_ERROR	    (1 << 28)
-#define   DP_AUX_CH_CTL_TIME_OUT_400us	    (0 << 26)
-#define   DP_AUX_CH_CTL_TIME_OUT_600us	    (1 << 26)
-#define   DP_AUX_CH_CTL_TIME_OUT_800us	    (2 << 26)
-#define   DP_AUX_CH_CTL_TIME_OUT_MAX	    (3 << 26) /* Varies per platform */
-#define   DP_AUX_CH_CTL_TIME_OUT_MASK	    (3 << 26)
-#define   DP_AUX_CH_CTL_RECEIVE_ERROR	    (1 << 25)
-#define   DP_AUX_CH_CTL_MESSAGE_SIZE_MASK    (0x1f << 20)
-#define   DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT   20
-#define   XELPDP_DP_AUX_CH_CTL_POWER_REQUEST REG_BIT(19)
-#define   XELPDP_DP_AUX_CH_CTL_POWER_STATUS  REG_BIT(18)
-#define   DP_AUX_CH_CTL_PRECHARGE_2US_MASK   (0xf << 16)
-#define   DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT  16
-#define   DP_AUX_CH_CTL_AUX_AKSV_SELECT	    (1 << 15)
-#define   DP_AUX_CH_CTL_MANCHESTER_TEST	    (1 << 14)
-#define   DP_AUX_CH_CTL_SYNC_TEST	    (1 << 13)
-#define   DP_AUX_CH_CTL_DEGLITCH_TEST	    (1 << 12)
-#define   DP_AUX_CH_CTL_PRECHARGE_TEST	    (1 << 11)
-#define   DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK    (0x7ff)
-#define   DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT   0
-#define   DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL	(1 << 14)
-#define   DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL	(1 << 13)
-#define   DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL	(1 << 12)
-#define   DP_AUX_CH_CTL_TBT_IO			(1 << 11)
-#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK (0x1f << 5)
-#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(c) (((c) - 1) << 5)
-#define   DP_AUX_CH_CTL_SYNC_PULSE_SKL(c)   ((c) - 1)
+#define   DP_AUX_CH_CTL_SEND_BUSY		REG_BIT(31)
+#define   DP_AUX_CH_CTL_DONE			REG_BIT(30)
+#define   DP_AUX_CH_CTL_INTERRUPT		REG_BIT(29)
+#define   DP_AUX_CH_CTL_TIME_OUT_ERROR		REG_BIT(28)
+
+#define   DP_AUX_CH_CTL_TIME_OUT_MASK		REG_GENMASK(27, 26)
+#define   DP_AUX_CH_CTL_TIME_OUT_400us		REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 0)
+#define   DP_AUX_CH_CTL_TIME_OUT_600us		REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 1)
+#define   DP_AUX_CH_CTL_TIME_OUT_800us		REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 2)
+#define   DP_AUX_CH_CTL_TIME_OUT_MAX		REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 3) /* Varies per platform */
+#define   DP_AUX_CH_CTL_RECEIVE_ERROR		REG_BIT(25)
+#define   DP_AUX_CH_CTL_MESSAGE_SIZE_MASK	REG_GENMASK(24, 20)
+#define   DP_AUX_CH_CTL_MESSAGE_SIZE(x)		REG_FIELD_PREP(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, (x))
+#define   DP_AUX_CH_CTL_PRECHARGE_2US_MASK	REG_GENMASK(19, 16) /* pre-skl */
+#define   DP_AUX_CH_CTL_PRECHARGE_2US(x)	REG_FIELD_PREP(DP_AUX_CH_CTL_PRECHARGE_2US_MASK, (x))
+#define   XELPDP_DP_AUX_CH_CTL_POWER_REQUEST	REG_BIT(19) /* mtl+ */
+#define   XELPDP_DP_AUX_CH_CTL_POWER_STATUS	REG_BIT(18) /* mtl+ */
+#define   DP_AUX_CH_CTL_AUX_AKSV_SELECT		REG_BIT(15)
+#define   DP_AUX_CH_CTL_MANCHESTER_TEST		REG_BIT(14) /* pre-hsw */
+#define   DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL	REG_BIT(14) /* skl+ */
+#define   DP_AUX_CH_CTL_SYNC_TEST		REG_BIT(13) /* pre-hsw */
+#define   DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL	REG_BIT(13) /* skl+ */
+#define   DP_AUX_CH_CTL_DEGLITCH_TEST		REG_BIT(12) /* pre-hsw */
+#define   DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL	REG_BIT(12) /* skl+ */
+#define   DP_AUX_CH_CTL_PRECHARGE_TEST		REG_BIT(11) /* pre-hsw */
+#define   DP_AUX_CH_CTL_TBT_IO			REG_BIT(11) /* icl+ */
+#define   DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK	REG_GENMASK(10, 0) /* pre-skl */
+#define   DP_AUX_CH_CTL_BIT_CLOCK_2X(x)		REG_FIELD_PREP(DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK, (x))
+#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK	REG_GENMASK(9, 5) /* skl+ */
+#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(c)	REG_FIELD_PREP(DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK, (c) - 1)
+#define   DP_AUX_CH_CTL_SYNC_PULSE_SKL_MASK	REG_GENMASK(4, 0) /* skl+ */
+#define   DP_AUX_CH_CTL_SYNC_PULSE_SKL(c)	REG_FIELD_PREP(DP_AUX_CH_CTL_SYNC_PULSE_SKL_MASK, (c) - 1)
 
 #endif /* __INTEL_DP_AUX_REGS_H__ */
diff --git a/drivers/gpu/drm/i915/gvt/edid.c b/drivers/gpu/drm/i915/gvt/edid.c
index 7c49a3d673a54..2a0438f12a149 100644
--- a/drivers/gpu/drm/i915/gvt/edid.c
+++ b/drivers/gpu/drm/i915/gvt/edid.c
@@ -463,10 +463,6 @@ static inline int get_aux_ch_reg(unsigned int offset)
 	return reg;
 }
 
-#define AUX_CTL_MSG_LENGTH(reg) \
-	((reg & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >> \
-		DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT)
-
 /**
  * intel_gvt_i2c_handle_aux_ch_write - emulate AUX channel register write
  * @vgpu: a vGPU
@@ -495,7 +491,8 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu,
 		return;
 	}
 
-	msg_length = AUX_CTL_MSG_LENGTH(value);
+	msg_length = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, reg);
+
 	// check the msg in DATA register.
 	msg = vgpu_vreg(vgpu, offset + 4);
 	addr = (msg >> 8) & 0xffff;
@@ -510,8 +507,7 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu,
 	ret_msg_size = (((op & 0x1) == GVT_AUX_I2C_READ) ? 2 : 1);
 	vgpu_vreg(vgpu, offset) =
 		DP_AUX_CH_CTL_DONE |
-		((ret_msg_size << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) &
-		DP_AUX_CH_CTL_MESSAGE_SIZE_MASK);
+		DP_AUX_CH_CTL_MESSAGE_SIZE(ret_msg_size);
 
 	if (msg_length == 3) {
 		if (!(op & GVT_AUX_I2C_MOT)) {
-- 
GitLab


From 5a0fc7a0df6b3525bf3b33fb27eb8d47f7826dc2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 9 May 2023 19:02:04 +0300
Subject: [PATCH 0472/1791] drm/i915: Split map_aux_ch() into per-platform
 arrays
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The big switch+if statement mess in map_aux_ch() is
illegible. Split up into cleaner per-platform arrays
like we already have for the gmbus pins.

We use enum aux_ch as the index and the VBT thing as
the value. Slightly non-intuitive perhaps but if we
did it the other way around we'd have problems with
AUX_CH_A being zero, and thus any non-populated
element would look like AUX_CH_A.

v2: flip the index vs. value around

TODO: Didn't bother with the platform variants beyond the
      ones that really need remapping, which means if the
      VBT is bogus we end up with a nonexistent aux ch.
      Might be nice to check this a bit better.
      Yet another bitmask in device info?

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509160206.25971-1-ville.syrjala@linux.intel.com
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c | 144 +++++++++++-----------
 1 file changed, 71 insertions(+), 73 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 75e69dffc5e90..c77d40535fc47 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -3578,84 +3578,82 @@ bool intel_bios_get_dsc_params(struct intel_encoder *encoder,
 	return false;
 }
 
+static const u8 adlp_aux_ch_map[] = {
+	[AUX_CH_A] = DP_AUX_A,
+	[AUX_CH_B] = DP_AUX_B,
+	[AUX_CH_C] = DP_AUX_C,
+	[AUX_CH_D_XELPD] = DP_AUX_D,
+	[AUX_CH_E_XELPD] = DP_AUX_E,
+	[AUX_CH_USBC1] = DP_AUX_F,
+	[AUX_CH_USBC2] = DP_AUX_G,
+	[AUX_CH_USBC3] = DP_AUX_H,
+	[AUX_CH_USBC4] = DP_AUX_I,
+};
+
+/*
+ * ADL-S VBT uses PHY based mapping. Combo PHYs A,B,C,D,E
+ * map to DDI A,TC1,TC2,TC3,TC4 respectively.
+ */
+static const u8 adls_aux_ch_map[] = {
+	[AUX_CH_A] = DP_AUX_A,
+	[AUX_CH_USBC1] = DP_AUX_B,
+	[AUX_CH_USBC2] = DP_AUX_C,
+	[AUX_CH_USBC3] = DP_AUX_D,
+	[AUX_CH_USBC4] = DP_AUX_E,
+};
+
+/*
+ * RKL/DG1 VBT uses PHY based mapping. Combo PHYs A,B,C,D
+ * map to DDI A,B,TC1,TC2 respectively.
+ */
+static const u8 rkl_aux_ch_map[] = {
+	[AUX_CH_A] = DP_AUX_A,
+	[AUX_CH_B] = DP_AUX_B,
+	[AUX_CH_USBC1] = DP_AUX_C,
+	[AUX_CH_USBC2] = DP_AUX_D,
+};
+
+static const u8 direct_aux_ch_map[] = {
+	[AUX_CH_A] = DP_AUX_A,
+	[AUX_CH_B] = DP_AUX_B,
+	[AUX_CH_C] = DP_AUX_C,
+	[AUX_CH_D] = DP_AUX_D, /* aka AUX_CH_USBC1 */
+	[AUX_CH_E] = DP_AUX_E, /* aka AUX_CH_USBC2 */
+	[AUX_CH_F] = DP_AUX_F, /* aka AUX_CH_USBC3 */
+	[AUX_CH_G] = DP_AUX_G, /* aka AUX_CH_USBC4 */
+	[AUX_CH_H] = DP_AUX_H, /* aka AUX_CH_USBC5 */
+	[AUX_CH_I] = DP_AUX_I, /* aka AUX_CH_USBC6 */
+};
+
 static enum aux_ch map_aux_ch(struct drm_i915_private *i915, u8 aux_channel)
 {
-	enum aux_ch aux_ch;
+	const u8 *aux_ch_map;
+	int i, n_entries;
 
-	/*
-	 * RKL/DG1 VBT uses PHY based mapping. Combo PHYs A,B,C,D
-	 * map to DDI A,B,TC1,TC2 respectively.
-	 *
-	 * ADL-S VBT uses PHY based mapping. Combo PHYs A,B,C,D,E
-	 * map to DDI A,TC1,TC2,TC3,TC4 respectively.
-	 */
-	switch (aux_channel) {
-	case DP_AUX_A:
-		aux_ch = AUX_CH_A;
-		break;
-	case DP_AUX_B:
-		if (IS_ALDERLAKE_S(i915))
-			aux_ch = AUX_CH_USBC1;
-		else
-			aux_ch = AUX_CH_B;
-		break;
-	case DP_AUX_C:
-		if (IS_ALDERLAKE_S(i915))
-			aux_ch = AUX_CH_USBC2;
-		else if (IS_DG1(i915) || IS_ROCKETLAKE(i915))
-			aux_ch = AUX_CH_USBC1;
-		else
-			aux_ch = AUX_CH_C;
-		break;
-	case DP_AUX_D:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_D_XELPD;
-		else if (IS_ALDERLAKE_S(i915))
-			aux_ch = AUX_CH_USBC3;
-		else if (IS_DG1(i915) || IS_ROCKETLAKE(i915))
-			aux_ch = AUX_CH_USBC2;
-		else
-			aux_ch = AUX_CH_D;
-		break;
-	case DP_AUX_E:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_E_XELPD;
-		else if (IS_ALDERLAKE_S(i915))
-			aux_ch = AUX_CH_USBC4;
-		else
-			aux_ch = AUX_CH_E;
-		break;
-	case DP_AUX_F:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_USBC1;
-		else
-			aux_ch = AUX_CH_F;
-		break;
-	case DP_AUX_G:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_USBC2;
-		else
-			aux_ch = AUX_CH_G;
-		break;
-	case DP_AUX_H:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_USBC3;
-		else
-			aux_ch = AUX_CH_H;
-		break;
-	case DP_AUX_I:
-		if (DISPLAY_VER(i915) >= 13)
-			aux_ch = AUX_CH_USBC4;
-		else
-			aux_ch = AUX_CH_I;
-		break;
-	default:
-		MISSING_CASE(aux_channel);
-		aux_ch = AUX_CH_A;
-		break;
+	if (DISPLAY_VER(i915) >= 13) {
+		aux_ch_map = adlp_aux_ch_map;
+		n_entries = ARRAY_SIZE(adlp_aux_ch_map);
+	} else if (IS_ALDERLAKE_S(i915)) {
+		aux_ch_map = adls_aux_ch_map;
+		n_entries = ARRAY_SIZE(adls_aux_ch_map);
+	} else if (IS_DG1(i915) || IS_ROCKETLAKE(i915)) {
+		aux_ch_map = rkl_aux_ch_map;
+		n_entries = ARRAY_SIZE(rkl_aux_ch_map);
+	} else {
+		aux_ch_map = direct_aux_ch_map;
+		n_entries = ARRAY_SIZE(direct_aux_ch_map);
 	}
 
-	return aux_ch;
+	for (i = 0; i < n_entries; i++) {
+		if (aux_ch_map[i] == aux_channel)
+			return i;
+	}
+
+	drm_dbg_kms(&i915->drm,
+		    "Ignoring alternate AUX CH: VBT claims AUX 0x%x, which is not valid for this platform\n",
+		    aux_channel);
+
+	return AUX_CH_NONE;
 }
 
 enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata)
-- 
GitLab


From 3d7af6cfed6c2df11388b7ac0630ad8015f9e915 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 9 May 2023 19:02:05 +0300
Subject: [PATCH 0473/1791] drm/i915: Flip VBT DDC pin maps around
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Swap the roles of the index vs. value for the VBT DDC pin
mapping tables. This is not strictly necessary for DDC pins
but it will make this work exactly like the AUX CH mapping
tables where the role reversal is necessary (or at least makes
things easier). Consistency is good.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509160206.25971-2-ville.syrjala@linux.intel.com
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c | 70 ++++++++++++-----------
 1 file changed, 36 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index c77d40535fc47..ac0fc1993376a 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -2141,58 +2141,58 @@ static u8 translate_iboost(u8 val)
 
 static const u8 cnp_ddc_pin_map[] = {
 	[0] = 0, /* N/A */
-	[DDC_BUS_DDI_B] = GMBUS_PIN_1_BXT,
-	[DDC_BUS_DDI_C] = GMBUS_PIN_2_BXT,
-	[DDC_BUS_DDI_D] = GMBUS_PIN_4_CNP, /* sic */
-	[DDC_BUS_DDI_F] = GMBUS_PIN_3_BXT, /* sic */
+	[GMBUS_PIN_1_BXT] = DDC_BUS_DDI_B,
+	[GMBUS_PIN_2_BXT] = DDC_BUS_DDI_C,
+	[GMBUS_PIN_4_CNP] = DDC_BUS_DDI_D, /* sic */
+	[GMBUS_PIN_3_BXT] = DDC_BUS_DDI_F, /* sic */
 };
 
 static const u8 icp_ddc_pin_map[] = {
-	[ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT,
-	[ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT,
-	[TGL_DDC_BUS_DDI_C] = GMBUS_PIN_3_BXT,
-	[ICL_DDC_BUS_PORT_1] = GMBUS_PIN_9_TC1_ICP,
-	[ICL_DDC_BUS_PORT_2] = GMBUS_PIN_10_TC2_ICP,
-	[ICL_DDC_BUS_PORT_3] = GMBUS_PIN_11_TC3_ICP,
-	[ICL_DDC_BUS_PORT_4] = GMBUS_PIN_12_TC4_ICP,
-	[TGL_DDC_BUS_PORT_5] = GMBUS_PIN_13_TC5_TGP,
-	[TGL_DDC_BUS_PORT_6] = GMBUS_PIN_14_TC6_TGP,
+	[GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A,
+	[GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B,
+	[GMBUS_PIN_3_BXT] = TGL_DDC_BUS_DDI_C,
+	[GMBUS_PIN_9_TC1_ICP] = ICL_DDC_BUS_PORT_1,
+	[GMBUS_PIN_10_TC2_ICP] = ICL_DDC_BUS_PORT_2,
+	[GMBUS_PIN_11_TC3_ICP] = ICL_DDC_BUS_PORT_3,
+	[GMBUS_PIN_12_TC4_ICP] = ICL_DDC_BUS_PORT_4,
+	[GMBUS_PIN_13_TC5_TGP] = TGL_DDC_BUS_PORT_5,
+	[GMBUS_PIN_14_TC6_TGP] = TGL_DDC_BUS_PORT_6,
 };
 
 static const u8 rkl_pch_tgp_ddc_pin_map[] = {
-	[ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT,
-	[ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT,
-	[RKL_DDC_BUS_DDI_D] = GMBUS_PIN_9_TC1_ICP,
-	[RKL_DDC_BUS_DDI_E] = GMBUS_PIN_10_TC2_ICP,
+	[GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A,
+	[GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B,
+	[GMBUS_PIN_9_TC1_ICP] = RKL_DDC_BUS_DDI_D,
+	[GMBUS_PIN_10_TC2_ICP] = RKL_DDC_BUS_DDI_E,
 };
 
 static const u8 adls_ddc_pin_map[] = {
-	[ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT,
-	[ADLS_DDC_BUS_PORT_TC1] = GMBUS_PIN_9_TC1_ICP,
-	[ADLS_DDC_BUS_PORT_TC2] = GMBUS_PIN_10_TC2_ICP,
-	[ADLS_DDC_BUS_PORT_TC3] = GMBUS_PIN_11_TC3_ICP,
-	[ADLS_DDC_BUS_PORT_TC4] = GMBUS_PIN_12_TC4_ICP,
+	[GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A,
+	[GMBUS_PIN_9_TC1_ICP] = ADLS_DDC_BUS_PORT_TC1,
+	[GMBUS_PIN_10_TC2_ICP] = ADLS_DDC_BUS_PORT_TC2,
+	[GMBUS_PIN_11_TC3_ICP] = ADLS_DDC_BUS_PORT_TC3,
+	[GMBUS_PIN_12_TC4_ICP] = ADLS_DDC_BUS_PORT_TC4,
 };
 
 static const u8 gen9bc_tgp_ddc_pin_map[] = {
-	[DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT,
-	[DDC_BUS_DDI_C] = GMBUS_PIN_9_TC1_ICP,
-	[DDC_BUS_DDI_D] = GMBUS_PIN_10_TC2_ICP,
+	[GMBUS_PIN_2_BXT] = DDC_BUS_DDI_B,
+	[GMBUS_PIN_9_TC1_ICP] = DDC_BUS_DDI_C,
+	[GMBUS_PIN_10_TC2_ICP] = DDC_BUS_DDI_D,
 };
 
 static const u8 adlp_ddc_pin_map[] = {
-	[ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT,
-	[ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT,
-	[ADLP_DDC_BUS_PORT_TC1] = GMBUS_PIN_9_TC1_ICP,
-	[ADLP_DDC_BUS_PORT_TC2] = GMBUS_PIN_10_TC2_ICP,
-	[ADLP_DDC_BUS_PORT_TC3] = GMBUS_PIN_11_TC3_ICP,
-	[ADLP_DDC_BUS_PORT_TC4] = GMBUS_PIN_12_TC4_ICP,
+	[GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A,
+	[GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B,
+	[GMBUS_PIN_9_TC1_ICP] = ADLP_DDC_BUS_PORT_TC1,
+	[GMBUS_PIN_10_TC2_ICP] = ADLP_DDC_BUS_PORT_TC2,
+	[GMBUS_PIN_11_TC3_ICP] = ADLP_DDC_BUS_PORT_TC3,
+	[GMBUS_PIN_12_TC4_ICP] = ADLP_DDC_BUS_PORT_TC4,
 };
 
 static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin)
 {
 	const u8 *ddc_pin_map;
-	int n_entries;
+	int i, n_entries;
 
 	if (HAS_PCH_MTP(i915) || IS_ALDERLAKE_P(i915)) {
 		ddc_pin_map = adlp_ddc_pin_map;
@@ -2219,8 +2219,10 @@ static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin)
 		return vbt_pin;
 	}
 
-	if (vbt_pin < n_entries && ddc_pin_map[vbt_pin] != 0)
-		return ddc_pin_map[vbt_pin];
+	for (i = 0; i < n_entries; i++) {
+		if (ddc_pin_map[i] == vbt_pin)
+			return i;
+	}
 
 	drm_dbg_kms(&i915->drm,
 		    "Ignoring alternate pin: VBT claims DDC pin %d, which is not valid for this platform\n",
-- 
GitLab


From 2bea1d7c594dd0643db23a8131c689384d0e5d8c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 9 May 2023 19:02:06 +0300
Subject: [PATCH 0474/1791] drm/i915: Nuke intel_bios_is_port_dp_dual_mode()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that we have encoder->devdata everwhere we don't need
the intel_bios_is_port_dp_dual_mode() wrapper any more.

And while at it let's include it in the child device log
dump as well since the logic in there is a bit more complex
than just DP&&HDMI.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509160206.25971-3-ville.syrjala@linux.intel.com
Acked-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c | 14 +++-----------
 drivers/gpu/drm/i915/display/intel_bios.h |  2 +-
 drivers/gpu/drm/i915/display/intel_hdmi.c |  6 +++---
 3 files changed, 7 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index ac0fc1993376a..64eb11a452659 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -2677,8 +2677,9 @@ static void print_ddi_port(const struct intel_bios_encoder_data *devdata,
 	supports_tbt = intel_bios_encoder_supports_tbt(devdata);
 
 	drm_dbg_kms(&i915->drm,
-		    "Port %c VBT info: CRT:%d DVI:%d HDMI:%d DP:%d eDP:%d DSI:%d LSPCON:%d USB-Type-C:%d TBT:%d DSC:%d\n",
+		    "Port %c VBT info: CRT:%d DVI:%d HDMI:%d DP:%d eDP:%d DSI:%d DP++:%d LSPCON:%d USB-Type-C:%d TBT:%d DSC:%d\n",
 		    port_name(port), is_crt, is_dvi, is_hdmi, is_dp, is_edp, is_dsi,
+		    intel_bios_encoder_supports_dp_dual_mode(devdata),
 		    intel_bios_encoder_is_lspcon(devdata),
 		    supports_typec_usb, supports_tbt,
 		    devdata->dsc != NULL);
@@ -3426,7 +3427,7 @@ bool intel_bios_is_port_present(struct drm_i915_private *i915, enum port port)
 	return false;
 }
 
-static bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata)
+bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata)
 {
 	const struct child_device_config *child = &devdata->child;
 
@@ -3445,15 +3446,6 @@ static bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_enc
 	return false;
 }
 
-bool intel_bios_is_port_dp_dual_mode(struct drm_i915_private *i915,
-				     enum port port)
-{
-	const struct intel_bios_encoder_data *devdata =
-		intel_bios_encoder_data_lookup(i915, port);
-
-	return devdata && intel_bios_encoder_supports_dp_dual_mode(devdata);
-}
-
 /**
  * intel_bios_is_dsi_present - is DSI present in VBT
  * @i915:	i915 device instance
diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h
index 8a0730c9b48cf..45fae97d97192 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.h
+++ b/drivers/gpu/drm/i915/display/intel_bios.h
@@ -247,7 +247,6 @@ bool intel_bios_is_tv_present(struct drm_i915_private *dev_priv);
 bool intel_bios_is_lvds_present(struct drm_i915_private *dev_priv, u8 *i2c_pin);
 bool intel_bios_is_port_present(struct drm_i915_private *dev_priv, enum port port);
 bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port);
-bool intel_bios_is_port_dp_dual_mode(struct drm_i915_private *dev_priv, enum port port);
 bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv, enum port *port);
 bool intel_bios_get_dsc_params(struct intel_encoder *encoder,
 			       struct intel_crtc_state *crtc_state,
@@ -264,6 +263,7 @@ bool intel_bios_encoder_supports_dp(const struct intel_bios_encoder_data *devdat
 bool intel_bios_encoder_supports_edp(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_supports_typec_usb(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_supports_tbt(const struct intel_bios_encoder_data *devdata);
+bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_lane_reversal(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata);
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 7e6b28cfb4db8..b4605010b29f3 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -2400,7 +2400,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector));
-	enum port port = hdmi_to_dig_port(hdmi)->base.port;
+	struct intel_encoder *encoder = &hdmi_to_dig_port(hdmi)->base;
 	struct i2c_adapter *adapter =
 		intel_gmbus_get_adapter(dev_priv, hdmi->ddc_bus);
 	enum drm_dp_dual_mode_type type = drm_dp_dual_mode_detect(&dev_priv->drm, adapter);
@@ -2416,7 +2416,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 	 */
 	if (type == DRM_DP_DUAL_MODE_UNKNOWN) {
 		if (!connector->force &&
-		    intel_bios_is_port_dp_dual_mode(dev_priv, port)) {
+		    intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) {
 			drm_dbg_kms(&dev_priv->drm,
 				    "Assuming DP dual mode adaptor presence based on VBT\n");
 			type = DRM_DP_DUAL_MODE_TYPE1_DVI;
@@ -2439,7 +2439,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector)
 
 	/* Older VBTs are often buggy and can't be trusted :( Play it safe. */
 	if ((DISPLAY_VER(dev_priv) >= 8 || IS_HASWELL(dev_priv)) &&
-	    !intel_bios_is_port_dp_dual_mode(dev_priv, port)) {
+	    !intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) {
 		drm_dbg_kms(&dev_priv->drm,
 			    "Ignoring DP dual mode adaptor max TMDS clock for native HDMI port\n");
 		hdmi->dp_dual_mode.max_tmds_clock = 0;
-- 
GitLab


From 5e352e32aec23570ea948f039e25faf9b9ba362b Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Tue, 9 May 2023 09:51:59 -0700
Subject: [PATCH 0475/1791] drm/i915: preparation for using PAT index

This patch is a preparation for replacing enum i915_cache_level with PAT
index. Caching policy for buffer objects is set through the PAT index in
PTE, the old i915_cache_level is not sufficient to represent all caching
modes supported by the hardware.

Preparing the transition by adding some platform dependent data structures
and helper functions to translate the cache_level to pat_index.

cachelevel_to_pat: a platform dependent array mapping cache_level to
                   pat_index.

max_pat_index: the maximum PAT index recommended in hardware specification
               Needed for validating the PAT index passed in from user
               space.

i915_gem_get_pat_index: function to convert cache_level to PAT index.

obj_to_i915(obj): macro moved to header file for wider usage.

I915_MAX_CACHE_LEVEL: upper bound of i915_cache_level for the
                      convenience of coding.

Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509165200.1740-2-fei.yang@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_object.c    |  9 +++
 drivers/gpu/drm/i915/gem/i915_gem_object.h    |  4 +
 .../gpu/drm/i915/gem/i915_gem_object_types.h  |  7 ++
 drivers/gpu/drm/i915/gem/i915_gem_shrinker.c  |  2 -
 drivers/gpu/drm/i915/gt/gen8_ppgtt.c          |  6 ++
 drivers/gpu/drm/i915/gt/intel_ggtt.c          |  6 ++
 drivers/gpu/drm/i915/i915_pci.c               | 79 ++++++++++++++++---
 drivers/gpu/drm/i915/intel_device_info.h      |  5 ++
 .../gpu/drm/i915/selftests/mock_gem_device.c  |  9 +++
 9 files changed, 116 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 4666bb82f312c..8c70a0ec7d2fa 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -45,6 +45,15 @@ static struct kmem_cache *slab_objects;
 
 static const struct drm_gem_object_funcs i915_gem_object_funcs;
 
+unsigned int i915_gem_get_pat_index(struct drm_i915_private *i915,
+				    enum i915_cache_level level)
+{
+	if (drm_WARN_ON(&i915->drm, level >= I915_MAX_CACHE_LEVEL))
+		return 0;
+
+	return INTEL_INFO(i915)->cachelevel_to_pat[level];
+}
+
 struct drm_i915_gem_object *i915_gem_object_alloc(void)
 {
 	struct drm_i915_gem_object *obj;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index bc1291887d4f9..284e1aa396cdc 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -20,6 +20,8 @@
 
 enum intel_region_id;
 
+#define obj_to_i915(obj__) to_i915((obj__)->base.dev)
+
 static inline bool i915_gem_object_size_2big(u64 size)
 {
 	struct drm_i915_gem_object *obj;
@@ -30,6 +32,8 @@ static inline bool i915_gem_object_size_2big(u64 size)
 	return false;
 }
 
+unsigned int i915_gem_get_pat_index(struct drm_i915_private *i915,
+				    enum i915_cache_level level);
 void i915_gem_init__objects(struct drm_i915_private *i915);
 
 void i915_objects_module_exit(void);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index 830c11431ee8b..bf0bd8e113559 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -194,6 +194,13 @@ enum i915_cache_level {
 	 * engine.
 	 */
 	I915_CACHE_WT,
+	/**
+	 * @I915_MAX_CACHE_LEVEL:
+	 *
+	 * Mark the last entry in the enum. Used for defining cachelevel_to_pat
+	 * array for cache_level to pat translation table.
+	 */
+	I915_MAX_CACHE_LEVEL,
 };
 
 enum i915_map_type {
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index b1672e054b21e..214763942aa2b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -460,8 +460,6 @@ void i915_gem_shrinker_taints_mutex(struct drm_i915_private *i915,
 	fs_reclaim_release(GFP_KERNEL);
 }
 
-#define obj_to_i915(obj__) to_i915((obj__)->base.dev)
-
 /**
  * i915_gem_object_make_unshrinkable - Hide the object from the shrinker. By
  * default all object types that support shrinking(see IS_SHRINKABLE), will also
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index 22ec1566d2a7b..bb6998d671331 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -78,6 +78,12 @@ static u64 mtl_pte_encode(dma_addr_t addr,
 	case I915_CACHE_WT:
 		pte |= GEN12_PPGTT_PTE_PAT0;
 		break;
+	default:
+		/* This should never happen. Added to deal with the compile
+		 * error due to the addition of I915_MAX_CACHE_LEVEL. Will
+		 * be removed by the pat_index patch.
+		 */
+		break;
 	}
 
 	return pte;
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index 20915edc8bd90..c8390d03fce27 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -242,6 +242,12 @@ static u64 mtl_ggtt_pte_encode(dma_addr_t addr,
 	case I915_CACHE_WT:
 		pte |= MTL_GGTT_PTE_PAT0;
 		break;
+	default:
+		/* This should never happen. Added to deal with the compile
+		 * error due to the addition of I915_MAX_CACHE_LEVEL. Will
+		 * be removed by the pat_index patch.
+		 */
+		break;
 	}
 
 	return pte;
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 025d32c0b161b..1a3247451d0f8 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -29,6 +29,7 @@
 #include "display/intel_display.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_sa_media.h"
+#include "gem/i915_gem_object_types.h"
 
 #include "i915_driver.h"
 #include "i915_drv.h"
@@ -163,6 +164,38 @@
 		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 	}
 
+#define LEGACY_CACHELEVEL \
+	.cachelevel_to_pat = { \
+		[I915_CACHE_NONE]   = 0, \
+		[I915_CACHE_LLC]    = 1, \
+		[I915_CACHE_L3_LLC] = 2, \
+		[I915_CACHE_WT]     = 3, \
+	}
+
+#define TGL_CACHELEVEL \
+	.cachelevel_to_pat = { \
+		[I915_CACHE_NONE]   = 3, \
+		[I915_CACHE_LLC]    = 0, \
+		[I915_CACHE_L3_LLC] = 0, \
+		[I915_CACHE_WT]     = 2, \
+	}
+
+#define PVC_CACHELEVEL \
+	.cachelevel_to_pat = { \
+		[I915_CACHE_NONE]   = 0, \
+		[I915_CACHE_LLC]    = 3, \
+		[I915_CACHE_L3_LLC] = 3, \
+		[I915_CACHE_WT]     = 2, \
+	}
+
+#define MTL_CACHELEVEL \
+	.cachelevel_to_pat = { \
+		[I915_CACHE_NONE]   = 2, \
+		[I915_CACHE_LLC]    = 3, \
+		[I915_CACHE_L3_LLC] = 3, \
+		[I915_CACHE_WT]     = 1, \
+	}
+
 /* Keep in gen based order, and chronological order within a gen */
 
 #define GEN_DEFAULT_PAGE_SIZES \
@@ -188,11 +221,13 @@
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.dma_mask_size = 32, \
+	.max_pat_index = 3, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
 	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 #define I845_FEATURES \
 	GEN(2), \
@@ -209,11 +244,13 @@
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.dma_mask_size = 32, \
+	.max_pat_index = 3, \
 	I845_PIPE_OFFSETS, \
 	I845_CURSOR_OFFSETS, \
 	I845_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 static const struct intel_device_info i830_info = {
 	I830_FEATURES,
@@ -248,11 +285,13 @@ static const struct intel_device_info i865g_info = {
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
 	.dma_mask_size = 32, \
+	.max_pat_index = 3, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
 	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 static const struct intel_device_info i915g_info = {
 	GEN3_FEATURES,
@@ -340,11 +379,13 @@ static const struct intel_device_info pnv_m_info = {
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
 	.dma_mask_size = 36, \
+	.max_pat_index = 3, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
 	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 static const struct intel_device_info i965g_info = {
 	GEN4_FEATURES,
@@ -394,11 +435,13 @@ static const struct intel_device_info gm45_info = {
 	/* ilk does support rc6, but we do not implement [power] contexts */ \
 	.has_rc6 = 0, \
 	.dma_mask_size = 36, \
+	.max_pat_index = 3, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
 	ILK_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 static const struct intel_device_info ilk_d_info = {
 	GEN5_FEATURES,
@@ -428,13 +471,15 @@ static const struct intel_device_info ilk_m_info = {
 	.has_rc6p = 0, \
 	.has_rps = true, \
 	.dma_mask_size = 40, \
+	.max_pat_index = 3, \
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \
 	.__runtime.ppgtt_size = 31, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
 	ILK_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 #define SNB_D_PLATFORM \
 	GEN6_FEATURES, \
@@ -481,13 +526,15 @@ static const struct intel_device_info snb_m_gt2_info = {
 	.has_reset_engine = true, \
 	.has_rps = true, \
 	.dma_mask_size = 40, \
+	.max_pat_index = 3, \
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \
 	.__runtime.ppgtt_size = 31, \
 	IVB_PIPE_OFFSETS, \
 	IVB_CURSOR_OFFSETS, \
 	IVB_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 #define IVB_D_PLATFORM \
 	GEN7_FEATURES, \
@@ -541,6 +588,7 @@ static const struct intel_device_info vlv_info = {
 	.display.has_gmch = 1,
 	.display.has_hotplug = 1,
 	.dma_mask_size = 40,
+	.max_pat_index = 3,
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING,
 	.__runtime.ppgtt_size = 31,
 	.has_snoop = true,
@@ -552,6 +600,7 @@ static const struct intel_device_info vlv_info = {
 	I9XX_COLORS,
 	GEN_DEFAULT_PAGE_SIZES,
 	GEN_DEFAULT_REGIONS,
+	LEGACY_CACHELEVEL,
 };
 
 #define G75_FEATURES  \
@@ -639,6 +688,7 @@ static const struct intel_device_info chv_info = {
 	.has_logical_ring_contexts = 1,
 	.display.has_gmch = 1,
 	.dma_mask_size = 39,
+	.max_pat_index = 3,
 	.__runtime.ppgtt_type = INTEL_PPGTT_FULL,
 	.__runtime.ppgtt_size = 32,
 	.has_reset_engine = 1,
@@ -650,6 +700,7 @@ static const struct intel_device_info chv_info = {
 	CHV_COLORS,
 	GEN_DEFAULT_PAGE_SIZES,
 	GEN_DEFAULT_REGIONS,
+	LEGACY_CACHELEVEL,
 };
 
 #define GEN9_DEFAULT_PAGE_SIZES \
@@ -731,11 +782,13 @@ static const struct intel_device_info skl_gt4_info = {
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.display.has_ipc = 1, \
+	.max_pat_index = 3, \
 	HSW_PIPE_OFFSETS, \
 	IVB_CURSOR_OFFSETS, \
 	IVB_COLORS, \
 	GEN9_DEFAULT_PAGE_SIZES, \
-	GEN_DEFAULT_REGIONS
+	GEN_DEFAULT_REGIONS, \
+	LEGACY_CACHELEVEL
 
 static const struct intel_device_info bxt_info = {
 	GEN9_LP_FEATURES,
@@ -889,9 +942,11 @@ static const struct intel_device_info jsl_info = {
 		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
 	}, \
 	TGL_CURSOR_OFFSETS, \
+	TGL_CACHELEVEL, \
 	.has_global_mocs = 1, \
 	.has_pxp = 1, \
-	.display.has_dsb = 1
+	.display.has_dsb = 1, \
+	.max_pat_index = 3
 
 static const struct intel_device_info tgl_info = {
 	GEN12_FEATURES,
@@ -1013,6 +1068,7 @@ static const struct intel_device_info adl_p_info = {
 	.__runtime.graphics.ip.ver = 12, \
 	.__runtime.graphics.ip.rel = 50, \
 	XE_HP_PAGE_SIZES, \
+	TGL_CACHELEVEL, \
 	.dma_mask_size = 46, \
 	.has_3d_pipeline = 1, \
 	.has_64bit_reloc = 1, \
@@ -1031,6 +1087,7 @@ static const struct intel_device_info adl_p_info = {
 	.has_reset_engine = 1, \
 	.has_rps = 1, \
 	.has_runtime_pm = 1, \
+	.max_pat_index = 3, \
 	.__runtime.ppgtt_size = 48, \
 	.__runtime.ppgtt_type = INTEL_PPGTT_FULL
 
@@ -1107,11 +1164,13 @@ static const struct intel_device_info pvc_info = {
 	PLATFORM(INTEL_PONTEVECCHIO),
 	NO_DISPLAY,
 	.has_flat_ccs = 0,
+	.max_pat_index = 7,
 	.__runtime.platform_engine_mask =
 		BIT(BCS0) |
 		BIT(VCS0) |
 		BIT(CCS0) | BIT(CCS1) | BIT(CCS2) | BIT(CCS3),
 	.require_force_probe = 1,
+	PVC_CACHELEVEL,
 };
 
 #define XE_LPDP_FEATURES	\
@@ -1149,9 +1208,11 @@ static const struct intel_device_info mtl_info = {
 	.has_llc = 0,
 	.has_mslice_steering = 0,
 	.has_snoop = 1,
+	.max_pat_index = 4,
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(CCS0),
 	.require_force_probe = 1,
+	MTL_CACHELEVEL,
 };
 
 #undef PLATFORM
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index 080a4557899b6..d737dd601a6e3 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -35,6 +35,8 @@
 #include "gt/intel_context_types.h"
 #include "gt/intel_sseu.h"
 
+#include "gem/i915_gem_object_types.h"
+
 struct drm_printer;
 struct drm_i915_private;
 struct intel_gt_definition;
@@ -308,6 +310,9 @@ struct intel_device_info {
 	 * Initial runtime info. Do not access outside of i915_driver_create().
 	 */
 	const struct intel_runtime_info __runtime;
+
+	u32 cachelevel_to_pat[I915_MAX_CACHE_LEVEL];
+	u32 max_pat_index;
 };
 
 struct intel_driver_caps {
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
index f6a7c0bd2955c..0eda8b4ee17f5 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
@@ -123,7 +123,9 @@ struct drm_i915_private *mock_gem_device(void)
 	static struct dev_iommu fake_iommu = { .priv = (void *)-1 };
 #endif
 	struct drm_i915_private *i915;
+	struct intel_device_info *i915_info;
 	struct pci_dev *pdev;
+	unsigned int i;
 	int ret;
 
 	pdev = kzalloc(sizeof(*pdev), GFP_KERNEL);
@@ -180,6 +182,13 @@ struct drm_i915_private *mock_gem_device(void)
 		I915_GTT_PAGE_SIZE_2M;
 
 	RUNTIME_INFO(i915)->memory_regions = REGION_SMEM;
+
+	/* simply use legacy cache level for mock device */
+	i915_info = (struct intel_device_info *)INTEL_INFO(i915);
+	i915_info->max_pat_index = 3;
+	for (i = 0; i < I915_MAX_CACHE_LEVEL; i++)
+		i915_info->cachelevel_to_pat[i] = i;
+
 	intel_memory_regions_hw_probe(i915);
 
 	spin_lock_init(&i915->gpu_error.lock);
-- 
GitLab


From 9275277d53248d3f529d7ce66a6954241ae4d5cb Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Tue, 9 May 2023 09:52:00 -0700
Subject: [PATCH 0476/1791] drm/i915: use pat_index instead of cache_level

Currently the KMD is using enum i915_cache_level to set caching policy for
buffer objects. This is flaky because the PAT index which really controls
the caching behavior in PTE has far more levels than what's defined in the
enum. In addition, the PAT index is platform dependent, having to translate
between i915_cache_level and PAT index is not reliable, and makes the code
more complicated.

From UMD's perspective there is also a necessity to set caching policy for
performance fine tuning. It's much easier for the UMD to directly use PAT
index because the behavior of each PAT index is clearly defined in Bspec.
Having the abstracted i915_cache_level sitting in between would only cause
more ambiguity. PAT is expected to work much like MOCS already works today,
and by design userspace is expected to select the index that exactly
matches the desired behavior described in the hardware specification.

For these reasons this patch replaces i915_cache_level with PAT index. Also
note, the cache_level is not completely removed yet, because the KMD still
has the need of creating buffer objects with simple cache settings such as
cached, uncached, or writethrough. For kernel objects, cache_level is used
for simplicity and backward compatibility. For Pre-gen12 platforms PAT can
have 1:1 mapping to i915_cache_level, so these two are interchangeable. see
the use of LEGACY_CACHELEVEL.

One consequence of this change is that gen8_pte_encode is no longer working
for gen12 platforms due to the fact that gen12 platforms has different PAT
definitions. In the meantime the mtl_pte_encode introduced specfically for
MTL becomes generic for all gen12 platforms. This patch renames the MTL
PTE encode function into gen12_pte_encode and apply it to all gen12. Even
though this change looks unrelated, but separating them would temporarily
break gen12 PTE encoding, thus squash them in one patch.

Special note: this patch changes the way caching behavior is controlled in
the sense that some objects are left to be managed by userspace. For such
objects we need to be careful not to change the userspace settings.There
are kerneldoc and comments added around obj->cache_coherent, cache_dirty,
and how to bypass the checkings by i915_gem_object_has_cache_level. For
full understanding, these changes need to be looked at together with the
two follow-up patches, one disables the {set|get}_caching ioctl's and the
other adds set_pat extension to the GEM_CREATE uAPI.

Bspec: 63019

Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509165200.1740-3-fei.yang@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpt.c      | 12 +--
 drivers/gpu/drm/i915/gem/i915_gem_domain.c    | 58 +++++++++----
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    | 15 +++-
 drivers/gpu/drm/i915/gem/i915_gem_mman.c      | 11 ++-
 drivers/gpu/drm/i915/gem/i915_gem_object.c    | 51 ++++++++++-
 drivers/gpu/drm/i915/gem/i915_gem_object.h    |  4 +
 .../gpu/drm/i915/gem/i915_gem_object_types.h  | 46 +++++++++-
 drivers/gpu/drm/i915/gem/i915_gem_stolen.c    |  4 +-
 drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c  |  8 +-
 .../gpu/drm/i915/gem/selftests/huge_pages.c   |  2 +-
 .../drm/i915/gem/selftests/i915_gem_migrate.c |  2 +-
 .../drm/i915/gem/selftests/i915_gem_mman.c    |  2 +-
 drivers/gpu/drm/i915/gt/gen6_ppgtt.c          | 10 ++-
 drivers/gpu/drm/i915/gt/gen8_ppgtt.c          | 84 +++++++++----------
 drivers/gpu/drm/i915/gt/gen8_ppgtt.h          |  3 +-
 drivers/gpu/drm/i915/gt/intel_ggtt.c          | 82 +++++++++---------
 drivers/gpu/drm/i915/gt/intel_gtt.h           | 18 ++--
 drivers/gpu/drm/i915/gt/intel_migrate.c       | 47 ++++++-----
 drivers/gpu/drm/i915/gt/intel_migrate.h       | 13 ++-
 drivers/gpu/drm/i915/gt/intel_ppgtt.c         |  4 +-
 drivers/gpu/drm/i915/gt/selftest_migrate.c    | 47 ++++++-----
 drivers/gpu/drm/i915/gt/selftest_reset.c      |  8 +-
 drivers/gpu/drm/i915/gt/selftest_timeline.c   |  2 +-
 drivers/gpu/drm/i915/gt/selftest_tlb.c        |  4 +-
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c      | 10 ++-
 drivers/gpu/drm/i915/i915_debugfs.c           | 53 +++++++++---
 drivers/gpu/drm/i915/i915_gem.c               | 27 +++++-
 drivers/gpu/drm/i915/i915_gpu_error.c         |  8 +-
 drivers/gpu/drm/i915/i915_vma.c               | 16 ++--
 drivers/gpu/drm/i915/i915_vma.h               |  2 +-
 drivers/gpu/drm/i915/i915_vma_types.h         |  2 -
 drivers/gpu/drm/i915/selftests/i915_gem.c     |  5 +-
 .../gpu/drm/i915/selftests/i915_gem_evict.c   |  4 +-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 15 ++--
 .../drm/i915/selftests/intel_memory_region.c  |  4 +-
 drivers/gpu/drm/i915/selftests/mock_gtt.c     |  8 +-
 36 files changed, 453 insertions(+), 238 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c
index c5eacfdba1a53..7c5fddb203ba0 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt.c
@@ -43,24 +43,24 @@ static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
 static void dpt_insert_page(struct i915_address_space *vm,
 			    dma_addr_t addr,
 			    u64 offset,
-			    enum i915_cache_level level,
+			    unsigned int pat_index,
 			    u32 flags)
 {
 	struct i915_dpt *dpt = i915_vm_to_dpt(vm);
 	gen8_pte_t __iomem *base = dpt->iomem;
 
 	gen8_set_pte(base + offset / I915_GTT_PAGE_SIZE,
-		     vm->pte_encode(addr, level, flags));
+		     vm->pte_encode(addr, pat_index, flags));
 }
 
 static void dpt_insert_entries(struct i915_address_space *vm,
 			       struct i915_vma_resource *vma_res,
-			       enum i915_cache_level level,
+			       unsigned int pat_index,
 			       u32 flags)
 {
 	struct i915_dpt *dpt = i915_vm_to_dpt(vm);
 	gen8_pte_t __iomem *base = dpt->iomem;
-	const gen8_pte_t pte_encode = vm->pte_encode(0, level, flags);
+	const gen8_pte_t pte_encode = vm->pte_encode(0, pat_index, flags);
 	struct sgt_iter sgt_iter;
 	dma_addr_t addr;
 	int i;
@@ -83,7 +83,7 @@ static void dpt_clear_range(struct i915_address_space *vm,
 static void dpt_bind_vma(struct i915_address_space *vm,
 			 struct i915_vm_pt_stash *stash,
 			 struct i915_vma_resource *vma_res,
-			 enum i915_cache_level cache_level,
+			 unsigned int pat_index,
 			 u32 flags)
 {
 	u32 pte_flags;
@@ -98,7 +98,7 @@ static void dpt_bind_vma(struct i915_address_space *vm,
 	if (vma_res->bi.lmem)
 		pte_flags |= PTE_LM;
 
-	vm->insert_entries(vm, vma_res, cache_level, pte_flags);
+	vm->insert_entries(vm, vma_res, pat_index, pte_flags);
 
 	vma_res->page_sizes_gtt = I915_GTT_PAGE_SIZE;
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index d2d5a24301b25..05107a6efe450 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -27,8 +27,15 @@ static bool gpu_write_needs_clflush(struct drm_i915_gem_object *obj)
 	if (IS_DGFX(i915))
 		return false;
 
-	return !(obj->cache_level == I915_CACHE_NONE ||
-		 obj->cache_level == I915_CACHE_WT);
+	/*
+	 * For objects created by userspace through GEM_CREATE with pat_index
+	 * set by set_pat extension, i915_gem_object_has_cache_level() will
+	 * always return true, because the coherency of such object is managed
+	 * by userspace. Othereise the call here would fall back to checking
+	 * whether the object is un-cached or write-through.
+	 */
+	return !(i915_gem_object_has_cache_level(obj, I915_CACHE_NONE) ||
+		 i915_gem_object_has_cache_level(obj, I915_CACHE_WT));
 }
 
 bool i915_gem_cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
@@ -267,7 +274,13 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 {
 	int ret;
 
-	if (obj->cache_level == cache_level)
+	/*
+	 * For objects created by userspace through GEM_CREATE with pat_index
+	 * set by set_pat extension, simply return 0 here without touching
+	 * the cache setting, because such objects should have an immutable
+	 * cache setting by desgin and always managed by userspace.
+	 */
+	if (i915_gem_object_has_cache_level(obj, cache_level))
 		return 0;
 
 	ret = i915_gem_object_wait(obj,
@@ -278,10 +291,8 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		return ret;
 
 	/* Always invalidate stale cachelines */
-	if (obj->cache_level != cache_level) {
-		i915_gem_object_set_cache_coherency(obj, cache_level);
-		obj->cache_dirty = true;
-	}
+	i915_gem_object_set_cache_coherency(obj, cache_level);
+	obj->cache_dirty = true;
 
 	/* The cache-level will be applied when each vma is rebound. */
 	return i915_gem_object_unbind(obj,
@@ -306,20 +317,22 @@ int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data,
 		goto out;
 	}
 
-	switch (obj->cache_level) {
-	case I915_CACHE_LLC:
-	case I915_CACHE_L3_LLC:
-		args->caching = I915_CACHING_CACHED;
-		break;
+	/*
+	 * This ioctl should be disabled for the objects with pat_index
+	 * set by user space.
+	 */
+	if (obj->pat_set_by_user) {
+		err = -EOPNOTSUPP;
+		goto out;
+	}
 
-	case I915_CACHE_WT:
+	if (i915_gem_object_has_cache_level(obj, I915_CACHE_LLC) ||
+	    i915_gem_object_has_cache_level(obj, I915_CACHE_L3_LLC))
+		args->caching = I915_CACHING_CACHED;
+	else if (i915_gem_object_has_cache_level(obj, I915_CACHE_WT))
 		args->caching = I915_CACHING_DISPLAY;
-		break;
-
-	default:
+	else
 		args->caching = I915_CACHING_NONE;
-		break;
-	}
 out:
 	rcu_read_unlock();
 	return err;
@@ -364,6 +377,15 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 	if (!obj)
 		return -ENOENT;
 
+	/*
+	 * This ioctl should be disabled for the objects with pat_index
+	 * set by user space.
+	 */
+	if (obj->pat_set_by_user) {
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+
 	/*
 	 * The caching mode of proxy object is handled by its generator, and
 	 * not allowed to be changed by userspace.
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 3aeede6aee4dc..5fb459ea42945 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -640,9 +640,15 @@ static inline int use_cpu_reloc(const struct reloc_cache *cache,
 	if (DBG_FORCE_RELOC == FORCE_GTT_RELOC)
 		return false;
 
+	/*
+	 * For objects created by userspace through GEM_CREATE with pat_index
+	 * set by set_pat extension, i915_gem_object_has_cache_level() always
+	 * return true, otherwise the call would fall back to checking whether
+	 * the object is un-cached.
+	 */
 	return (cache->has_llc ||
 		obj->cache_dirty ||
-		obj->cache_level != I915_CACHE_NONE);
+		!i915_gem_object_has_cache_level(obj, I915_CACHE_NONE));
 }
 
 static int eb_reserve_vma(struct i915_execbuffer *eb,
@@ -1324,7 +1330,10 @@ static void *reloc_iomap(struct i915_vma *batch,
 	if (drm_mm_node_allocated(&cache->node)) {
 		ggtt->vm.insert_page(&ggtt->vm,
 				     i915_gem_object_get_dma_address(obj, page),
-				     offset, I915_CACHE_NONE, 0);
+				     offset,
+				     i915_gem_get_pat_index(ggtt->vm.i915,
+							    I915_CACHE_NONE),
+				     0);
 	} else {
 		offset += page << PAGE_SHIFT;
 	}
@@ -1464,7 +1473,7 @@ eb_relocate_entry(struct i915_execbuffer *eb,
 			reloc_cache_unmap(&eb->reloc_cache);
 			mutex_lock(&vma->vm->mutex);
 			err = i915_vma_bind(target->vma,
-					    target->vma->obj->cache_level,
+					    target->vma->obj->pat_index,
 					    PIN_GLOBAL, NULL, NULL);
 			mutex_unlock(&vma->vm->mutex);
 			reloc_cache_remap(&eb->reloc_cache, ev->vma->obj);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index 3dbacdf0911a5..4e7a838ab7bd9 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -383,7 +383,16 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf)
 	}
 
 	/* Access to snoopable pages through the GTT is incoherent. */
-	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(i915)) {
+	/*
+	 * For objects created by userspace through GEM_CREATE with pat_index
+	 * set by set_pat extension, coherency is managed by userspace, make
+	 * sure we don't fail handling the vm fault by calling
+	 * i915_gem_object_has_cache_level() which always return true for such
+	 * objects. Otherwise this helper function would fall back to checking
+	 * whether the object is un-cached.
+	 */
+	if (!(i915_gem_object_has_cache_level(obj, I915_CACHE_NONE) ||
+	      HAS_LLC(i915))) {
 		ret = -EFAULT;
 		goto err_unpin;
 	}
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 8c70a0ec7d2fa..46a19b099ec88 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -54,6 +54,24 @@ unsigned int i915_gem_get_pat_index(struct drm_i915_private *i915,
 	return INTEL_INFO(i915)->cachelevel_to_pat[level];
 }
 
+bool i915_gem_object_has_cache_level(const struct drm_i915_gem_object *obj,
+				     enum i915_cache_level lvl)
+{
+	/*
+	 * In case the pat_index is set by user space, this kernel mode
+	 * driver should leave the coherency to be managed by user space,
+	 * simply return true here.
+	 */
+	if (obj->pat_set_by_user)
+		return true;
+
+	/*
+	 * Otherwise the pat_index should have been converted from cache_level
+	 * so that the following comparison is valid.
+	 */
+	return obj->pat_index == i915_gem_get_pat_index(obj_to_i915(obj), lvl);
+}
+
 struct drm_i915_gem_object *i915_gem_object_alloc(void)
 {
 	struct drm_i915_gem_object *obj;
@@ -133,7 +151,7 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
 
-	obj->cache_level = cache_level;
+	obj->pat_index = i915_gem_get_pat_index(i915, cache_level);
 
 	if (cache_level != I915_CACHE_NONE)
 		obj->cache_coherent = (I915_BO_CACHE_COHERENT_FOR_READ |
@@ -148,6 +166,37 @@ void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 		!IS_DGFX(i915);
 }
 
+/**
+ * i915_gem_object_set_pat_index - set PAT index to be used in PTE encode
+ * @obj: #drm_i915_gem_object
+ * @pat_index: PAT index
+ *
+ * This is a clone of i915_gem_object_set_cache_coherency taking pat index
+ * instead of cache_level as its second argument.
+ */
+void i915_gem_object_set_pat_index(struct drm_i915_gem_object *obj,
+				   unsigned int pat_index)
+{
+	struct drm_i915_private *i915 = to_i915(obj->base.dev);
+
+	if (obj->pat_index == pat_index)
+		return;
+
+	obj->pat_index = pat_index;
+
+	if (pat_index != i915_gem_get_pat_index(i915, I915_CACHE_NONE))
+		obj->cache_coherent = (I915_BO_CACHE_COHERENT_FOR_READ |
+				       I915_BO_CACHE_COHERENT_FOR_WRITE);
+	else if (HAS_LLC(i915))
+		obj->cache_coherent = I915_BO_CACHE_COHERENT_FOR_READ;
+	else
+		obj->cache_coherent = 0;
+
+	obj->cache_dirty =
+		!(obj->cache_coherent & I915_BO_CACHE_COHERENT_FOR_WRITE) &&
+		!IS_DGFX(i915);
+}
+
 bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *i915 = to_i915(obj->base.dev);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 284e1aa396cdc..884a17275b3a4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -34,6 +34,8 @@ static inline bool i915_gem_object_size_2big(u64 size)
 
 unsigned int i915_gem_get_pat_index(struct drm_i915_private *i915,
 				    enum i915_cache_level level);
+bool i915_gem_object_has_cache_level(const struct drm_i915_gem_object *obj,
+				     enum i915_cache_level lvl);
 void i915_gem_init__objects(struct drm_i915_private *i915);
 
 void i915_objects_module_exit(void);
@@ -764,6 +766,8 @@ bool i915_gem_object_has_unknown_state(struct drm_i915_gem_object *obj);
 
 void i915_gem_object_set_cache_coherency(struct drm_i915_gem_object *obj,
 					 unsigned int cache_level);
+void i915_gem_object_set_pat_index(struct drm_i915_gem_object *obj,
+				   unsigned int pat_index);
 bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj);
 void i915_gem_object_flush_if_display(struct drm_i915_gem_object *obj);
 void i915_gem_object_flush_if_display_locked(struct drm_i915_gem_object *obj);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
index bf0bd8e113559..e72c57716beec 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h
@@ -364,15 +364,43 @@ struct drm_i915_gem_object {
 #define I915_BO_FLAG_STRUCT_PAGE BIT(0) /* Object backed by struct pages */
 #define I915_BO_FLAG_IOMEM       BIT(1) /* Object backed by IO memory */
 	/**
-	 * @cache_level: The desired GTT caching level.
+	 * @pat_index: The desired PAT index.
+	 *
+	 * See hardware specification for valid PAT indices for each platform.
+	 * This field replaces the @cache_level that contains a value of enum
+	 * i915_cache_level since PAT indices are being used by both userspace
+	 * and kernel mode driver for caching policy control after GEN12.
+	 * In the meantime platform specific tables are created to translate
+	 * i915_cache_level into pat index, for more details check the macros
+	 * defined i915/i915_pci.c, e.g. PVC_CACHELEVEL.
+	 * For backward compatibility, this field contains values exactly match
+	 * the entries of enum i915_cache_level for pre-GEN12 platforms (See
+	 * LEGACY_CACHELEVEL), so that the PTE encode functions for these
+	 * legacy platforms can stay the same.
+	 */
+	unsigned int pat_index:6;
+	/**
+	 * @pat_set_by_user: Indicate whether pat_index is set by user space
 	 *
-	 * See enum i915_cache_level for possible values, along with what
-	 * each does.
+	 * This field is set to false by default, only set to true if the
+	 * pat_index is set by user space. By design, user space is capable of
+	 * managing caching behavior by setting pat_index, in which case this
+	 * kernel mode driver should never touch the pat_index.
 	 */
-	unsigned int cache_level:3;
+	unsigned int pat_set_by_user:1;
 	/**
 	 * @cache_coherent:
 	 *
+	 * Note: with the change above which replaced @cache_level with pat_index,
+	 * the use of @cache_coherent is limited to the objects created by kernel
+	 * or by userspace without pat index specified.
+	 * Check for @pat_set_by_user to find out if an object has pat index set
+	 * by userspace. The ioctl's to change cache settings have also been
+	 * disabled for the objects with pat index set by userspace. Please don't
+	 * assume @cache_coherent having the flags set as describe here. A helper
+	 * function i915_gem_object_has_cache_level() provides one way to bypass
+	 * the use of this field.
+	 *
 	 * Track whether the pages are coherent with the GPU if reading or
 	 * writing through the CPU caches. The largely depends on the
 	 * @cache_level setting.
@@ -446,6 +474,16 @@ struct drm_i915_gem_object {
 	/**
 	 * @cache_dirty:
 	 *
+	 * Note: with the change above which replaced cache_level with pat_index,
+	 * the use of @cache_dirty is limited to the objects created by kernel
+	 * or by userspace without pat index specified.
+	 * Check for @pat_set_by_user to find out if an object has pat index set
+	 * by userspace. The ioctl's to change cache settings have also been
+	 * disabled for the objects with pat_index set by userspace. Please don't
+	 * assume @cache_dirty is set as describe here. Also see helper function
+	 * i915_gem_object_has_cache_level() for possible ways to bypass the use
+	 * of this field.
+	 *
 	 * Track if we are we dirty with writes through the CPU cache for this
 	 * object. As a result reading directly from main memory might yield
 	 * stale data.
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
index ee492d823f1bb..3b094d36a0b04 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c
@@ -565,7 +565,9 @@ static void dbg_poison(struct i915_ggtt *ggtt,
 
 		ggtt->vm.insert_page(&ggtt->vm, addr,
 				     ggtt->error_capture.start,
-				     I915_CACHE_NONE, 0);
+				     i915_gem_get_pat_index(ggtt->vm.i915,
+							    I915_CACHE_NONE),
+				     0);
 		mb();
 
 		s = io_mapping_map_wc(&ggtt->iomap,
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
index 69eb20ed4d47c..7078af2f8f795 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c
@@ -214,7 +214,8 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo,
 
 		intel_engine_pm_get(to_gt(i915)->migrate.context->engine);
 		ret = intel_context_migrate_clear(to_gt(i915)->migrate.context, deps,
-						  dst_st->sgl, dst_level,
+						  dst_st->sgl,
+						  i915_gem_get_pat_index(i915, dst_level),
 						  i915_ttm_gtt_binds_lmem(dst_mem),
 						  0, &rq);
 	} else {
@@ -228,9 +229,10 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo,
 		intel_engine_pm_get(to_gt(i915)->migrate.context->engine);
 		ret = intel_context_migrate_copy(to_gt(i915)->migrate.context,
 						 deps, src_rsgt->table.sgl,
-						 src_level,
+						 i915_gem_get_pat_index(i915, src_level),
 						 i915_ttm_gtt_binds_lmem(bo->resource),
-						 dst_st->sgl, dst_level,
+						 dst_st->sgl,
+						 i915_gem_get_pat_index(i915, dst_level),
 						 i915_ttm_gtt_binds_lmem(dst_mem),
 						 &rq);
 
diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
index cb5863f37f9d3..763df437c20c4 100644
--- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
@@ -354,7 +354,7 @@ fake_huge_pages_object(struct drm_i915_private *i915, u64 size, bool single)
 
 	obj->write_domain = I915_GEM_DOMAIN_CPU;
 	obj->read_domains = I915_GEM_DOMAIN_CPU;
-	obj->cache_level = I915_CACHE_NONE;
+	obj->pat_index = i915_gem_get_pat_index(i915, I915_CACHE_NONE);
 
 	return obj;
 }
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c
index fe6c37fd7859a..a93a90b15907b 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c
@@ -219,7 +219,7 @@ static int __igt_lmem_pages_migrate(struct intel_gt *gt,
 			continue;
 
 		err = intel_migrate_clear(&gt->migrate, &ww, deps,
-					  obj->mm.pages->sgl, obj->cache_level,
+					  obj->mm.pages->sgl, obj->pat_index,
 					  i915_gem_object_is_lmem(obj),
 					  0xdeadbeaf, &rq);
 		if (rq) {
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index 56279908ed305..a93d8f9f8bc14 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -1222,7 +1222,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 	}
 
 	err = intel_context_migrate_clear(to_gt(i915)->migrate.context, NULL,
-					  obj->mm.pages->sgl, obj->cache_level,
+					  obj->mm.pages->sgl, obj->pat_index,
 					  i915_gem_object_is_lmem(obj),
 					  expand32(POISON_INUSE), &rq);
 	i915_gem_object_unpin_pages(obj);
diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
index 5aaacc53fa4ca..c2bdc133c89ac 100644
--- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c
@@ -109,7 +109,7 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm,
 
 static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 				      struct i915_vma_resource *vma_res,
-				      enum i915_cache_level cache_level,
+				      unsigned int pat_index,
 				      u32 flags)
 {
 	struct i915_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
@@ -117,7 +117,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,
 	unsigned int first_entry = vma_res->start / I915_GTT_PAGE_SIZE;
 	unsigned int act_pt = first_entry / GEN6_PTES;
 	unsigned int act_pte = first_entry % GEN6_PTES;
-	const u32 pte_encode = vm->pte_encode(0, cache_level, flags);
+	const u32 pte_encode = vm->pte_encode(0, pat_index, flags);
 	struct sgt_dma iter = sgt_dma(vma_res);
 	gen6_pte_t *vaddr;
 
@@ -227,7 +227,9 @@ static int gen6_ppgtt_init_scratch(struct gen6_ppgtt *ppgtt)
 
 	vm->scratch[0]->encode =
 		vm->pte_encode(px_dma(vm->scratch[0]),
-			       I915_CACHE_NONE, PTE_READ_ONLY);
+			       i915_gem_get_pat_index(vm->i915,
+						      I915_CACHE_NONE),
+			       PTE_READ_ONLY);
 
 	vm->scratch[1] = vm->alloc_pt_dma(vm, I915_GTT_PAGE_SIZE_4K);
 	if (IS_ERR(vm->scratch[1])) {
@@ -278,7 +280,7 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 static void pd_vma_bind(struct i915_address_space *vm,
 			struct i915_vm_pt_stash *stash,
 			struct i915_vma_resource *vma_res,
-			enum i915_cache_level cache_level,
+			unsigned int pat_index,
 			u32 unused)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
index bb6998d671331..f948d33e5ec5a 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
@@ -29,7 +29,7 @@ static u64 gen8_pde_encode(const dma_addr_t addr,
 }
 
 static u64 gen8_pte_encode(dma_addr_t addr,
-			   enum i915_cache_level level,
+			   unsigned int pat_index,
 			   u32 flags)
 {
 	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT | GEN8_PAGE_RW;
@@ -40,7 +40,12 @@ static u64 gen8_pte_encode(dma_addr_t addr,
 	if (flags & PTE_LM)
 		pte |= GEN12_PPGTT_PTE_LM;
 
-	switch (level) {
+	/*
+	 * For pre-gen12 platforms pat_index is the same as enum
+	 * i915_cache_level, so the switch-case here is still valid.
+	 * See translation table defined by LEGACY_CACHELEVEL.
+	 */
+	switch (pat_index) {
 	case I915_CACHE_NONE:
 		pte |= PPAT_UNCACHED;
 		break;
@@ -55,9 +60,9 @@ static u64 gen8_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
-static u64 mtl_pte_encode(dma_addr_t addr,
-			  enum i915_cache_level level,
-			  u32 flags)
+static u64 gen12_pte_encode(dma_addr_t addr,
+			    unsigned int pat_index,
+			    u32 flags)
 {
 	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT | GEN8_PAGE_RW;
 
@@ -67,24 +72,17 @@ static u64 mtl_pte_encode(dma_addr_t addr,
 	if (flags & PTE_LM)
 		pte |= GEN12_PPGTT_PTE_LM;
 
-	switch (level) {
-	case I915_CACHE_NONE:
-		pte |= GEN12_PPGTT_PTE_PAT1;
-		break;
-	case I915_CACHE_LLC:
-	case I915_CACHE_L3_LLC:
-		pte |= GEN12_PPGTT_PTE_PAT0 | GEN12_PPGTT_PTE_PAT1;
-		break;
-	case I915_CACHE_WT:
+	if (pat_index & BIT(0))
 		pte |= GEN12_PPGTT_PTE_PAT0;
-		break;
-	default:
-		/* This should never happen. Added to deal with the compile
-		 * error due to the addition of I915_MAX_CACHE_LEVEL. Will
-		 * be removed by the pat_index patch.
-		 */
-		break;
-	}
+
+	if (pat_index & BIT(1))
+		pte |= GEN12_PPGTT_PTE_PAT1;
+
+	if (pat_index & BIT(2))
+		pte |= GEN12_PPGTT_PTE_PAT2;
+
+	if (pat_index & BIT(3))
+		pte |= MTL_PPGTT_PTE_PAT3;
 
 	return pte;
 }
@@ -457,11 +455,11 @@ gen8_ppgtt_insert_pte(struct i915_ppgtt *ppgtt,
 		      struct i915_page_directory *pdp,
 		      struct sgt_dma *iter,
 		      u64 idx,
-		      enum i915_cache_level cache_level,
+		      unsigned int pat_index,
 		      u32 flags)
 {
 	struct i915_page_directory *pd;
-	const gen8_pte_t pte_encode = ppgtt->vm.pte_encode(0, cache_level, flags);
+	const gen8_pte_t pte_encode = ppgtt->vm.pte_encode(0, pat_index, flags);
 	gen8_pte_t *vaddr;
 
 	pd = i915_pd_entry(pdp, gen8_pd_index(idx, 2));
@@ -504,10 +502,10 @@ static void
 xehpsdv_ppgtt_insert_huge(struct i915_address_space *vm,
 			  struct i915_vma_resource *vma_res,
 			  struct sgt_dma *iter,
-			  enum i915_cache_level cache_level,
+			  unsigned int pat_index,
 			  u32 flags)
 {
-	const gen8_pte_t pte_encode = vm->pte_encode(0, cache_level, flags);
+	const gen8_pte_t pte_encode = vm->pte_encode(0, pat_index, flags);
 	unsigned int rem = sg_dma_len(iter->sg);
 	u64 start = vma_res->start;
 	u64 end = start + vma_res->vma_size;
@@ -612,10 +610,10 @@ xehpsdv_ppgtt_insert_huge(struct i915_address_space *vm,
 static void gen8_ppgtt_insert_huge(struct i915_address_space *vm,
 				   struct i915_vma_resource *vma_res,
 				   struct sgt_dma *iter,
-				   enum i915_cache_level cache_level,
+				   unsigned int pat_index,
 				   u32 flags)
 {
-	const gen8_pte_t pte_encode = vm->pte_encode(0, cache_level, flags);
+	const gen8_pte_t pte_encode = vm->pte_encode(0, pat_index, flags);
 	unsigned int rem = sg_dma_len(iter->sg);
 	u64 start = vma_res->start;
 
@@ -735,7 +733,7 @@ static void gen8_ppgtt_insert_huge(struct i915_address_space *vm,
 
 static void gen8_ppgtt_insert(struct i915_address_space *vm,
 			      struct i915_vma_resource *vma_res,
-			      enum i915_cache_level cache_level,
+			      unsigned int pat_index,
 			      u32 flags)
 {
 	struct i915_ppgtt * const ppgtt = i915_vm_to_ppgtt(vm);
@@ -743,9 +741,9 @@ static void gen8_ppgtt_insert(struct i915_address_space *vm,
 
 	if (vma_res->bi.page_sizes.sg > I915_GTT_PAGE_SIZE) {
 		if (GRAPHICS_VER_FULL(vm->i915) >= IP_VER(12, 50))
-			xehpsdv_ppgtt_insert_huge(vm, vma_res, &iter, cache_level, flags);
+			xehpsdv_ppgtt_insert_huge(vm, vma_res, &iter, pat_index, flags);
 		else
-			gen8_ppgtt_insert_huge(vm, vma_res, &iter, cache_level, flags);
+			gen8_ppgtt_insert_huge(vm, vma_res, &iter, pat_index, flags);
 	} else  {
 		u64 idx = vma_res->start >> GEN8_PTE_SHIFT;
 
@@ -754,7 +752,7 @@ static void gen8_ppgtt_insert(struct i915_address_space *vm,
 				gen8_pdp_for_page_index(vm, idx);
 
 			idx = gen8_ppgtt_insert_pte(ppgtt, pdp, &iter, idx,
-						    cache_level, flags);
+						    pat_index, flags);
 		} while (idx);
 
 		vma_res->page_sizes_gtt = I915_GTT_PAGE_SIZE;
@@ -764,7 +762,7 @@ static void gen8_ppgtt_insert(struct i915_address_space *vm,
 static void gen8_ppgtt_insert_entry(struct i915_address_space *vm,
 				    dma_addr_t addr,
 				    u64 offset,
-				    enum i915_cache_level level,
+				    unsigned int pat_index,
 				    u32 flags)
 {
 	u64 idx = offset >> GEN8_PTE_SHIFT;
@@ -778,14 +776,14 @@ static void gen8_ppgtt_insert_entry(struct i915_address_space *vm,
 	GEM_BUG_ON(pt->is_compact);
 
 	vaddr = px_vaddr(pt);
-	vaddr[gen8_pd_index(idx, 0)] = vm->pte_encode(addr, level, flags);
+	vaddr[gen8_pd_index(idx, 0)] = vm->pte_encode(addr, pat_index, flags);
 	drm_clflush_virt_range(&vaddr[gen8_pd_index(idx, 0)], sizeof(*vaddr));
 }
 
 static void __xehpsdv_ppgtt_insert_entry_lm(struct i915_address_space *vm,
 					    dma_addr_t addr,
 					    u64 offset,
-					    enum i915_cache_level level,
+					    unsigned int pat_index,
 					    u32 flags)
 {
 	u64 idx = offset >> GEN8_PTE_SHIFT;
@@ -808,20 +806,20 @@ static void __xehpsdv_ppgtt_insert_entry_lm(struct i915_address_space *vm,
 	}
 
 	vaddr = px_vaddr(pt);
-	vaddr[gen8_pd_index(idx, 0) / 16] = vm->pte_encode(addr, level, flags);
+	vaddr[gen8_pd_index(idx, 0) / 16] = vm->pte_encode(addr, pat_index, flags);
 }
 
 static void xehpsdv_ppgtt_insert_entry(struct i915_address_space *vm,
 				       dma_addr_t addr,
 				       u64 offset,
-				       enum i915_cache_level level,
+				       unsigned int pat_index,
 				       u32 flags)
 {
 	if (flags & PTE_LM)
 		return __xehpsdv_ppgtt_insert_entry_lm(vm, addr, offset,
-						       level, flags);
+						       pat_index, flags);
 
-	return gen8_ppgtt_insert_entry(vm, addr, offset, level, flags);
+	return gen8_ppgtt_insert_entry(vm, addr, offset, pat_index, flags);
 }
 
 static int gen8_init_scratch(struct i915_address_space *vm)
@@ -856,7 +854,9 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 
 	vm->scratch[0]->encode =
 		vm->pte_encode(px_dma(vm->scratch[0]),
-			       I915_CACHE_NONE, pte_flags);
+			       i915_gem_get_pat_index(vm->i915,
+						      I915_CACHE_NONE),
+			       pte_flags);
 
 	for (i = 1; i <= vm->top; i++) {
 		struct drm_i915_gem_object *obj;
@@ -998,8 +998,8 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt,
 	 */
 	ppgtt->vm.alloc_scratch_dma = alloc_pt_dma;
 
-	if (GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 70))
-		ppgtt->vm.pte_encode = mtl_pte_encode;
+	if (GRAPHICS_VER(gt->i915) >= 12)
+		ppgtt->vm.pte_encode = gen12_pte_encode;
 	else
 		ppgtt->vm.pte_encode = gen8_pte_encode;
 
diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.h b/drivers/gpu/drm/i915/gt/gen8_ppgtt.h
index f541d19264b4e..19c6354416427 100644
--- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.h
+++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.h
@@ -10,13 +10,12 @@
 
 struct i915_address_space;
 struct intel_gt;
-enum i915_cache_level;
 
 struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt,
 				     unsigned long lmem_pt_obj_flags);
 
 u64 gen8_ggtt_pte_encode(dma_addr_t addr,
-			 enum i915_cache_level level,
+			 unsigned int pat_index,
 			 u32 flags);
 
 #endif
diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index c8390d03fce27..2a7942fac7980 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -221,7 +221,7 @@ static void guc_ggtt_invalidate(struct i915_ggtt *ggtt)
 }
 
 static u64 mtl_ggtt_pte_encode(dma_addr_t addr,
-			       enum i915_cache_level level,
+			       unsigned int pat_index,
 			       u32 flags)
 {
 	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT;
@@ -231,30 +231,17 @@ static u64 mtl_ggtt_pte_encode(dma_addr_t addr,
 	if (flags & PTE_LM)
 		pte |= GEN12_GGTT_PTE_LM;
 
-	switch (level) {
-	case I915_CACHE_NONE:
-		pte |= MTL_GGTT_PTE_PAT1;
-		break;
-	case I915_CACHE_LLC:
-	case I915_CACHE_L3_LLC:
-		pte |= MTL_GGTT_PTE_PAT0 | MTL_GGTT_PTE_PAT1;
-		break;
-	case I915_CACHE_WT:
+	if (pat_index & BIT(0))
 		pte |= MTL_GGTT_PTE_PAT0;
-		break;
-	default:
-		/* This should never happen. Added to deal with the compile
-		 * error due to the addition of I915_MAX_CACHE_LEVEL. Will
-		 * be removed by the pat_index patch.
-		 */
-		break;
-	}
+
+	if (pat_index & BIT(1))
+		pte |= MTL_GGTT_PTE_PAT1;
 
 	return pte;
 }
 
 u64 gen8_ggtt_pte_encode(dma_addr_t addr,
-			 enum i915_cache_level level,
+			 unsigned int pat_index,
 			 u32 flags)
 {
 	gen8_pte_t pte = addr | GEN8_PAGE_PRESENT;
@@ -273,25 +260,25 @@ static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
 static void gen8_ggtt_insert_page(struct i915_address_space *vm,
 				  dma_addr_t addr,
 				  u64 offset,
-				  enum i915_cache_level level,
+				  unsigned int pat_index,
 				  u32 flags)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 	gen8_pte_t __iomem *pte =
 		(gen8_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE;
 
-	gen8_set_pte(pte, ggtt->vm.pte_encode(addr, level, flags));
+	gen8_set_pte(pte, ggtt->vm.pte_encode(addr, pat_index, flags));
 
 	ggtt->invalidate(ggtt);
 }
 
 static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct i915_vma_resource *vma_res,
-				     enum i915_cache_level level,
+				     unsigned int pat_index,
 				     u32 flags)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
-	const gen8_pte_t pte_encode = ggtt->vm.pte_encode(0, level, flags);
+	const gen8_pte_t pte_encode = ggtt->vm.pte_encode(0, pat_index, flags);
 	gen8_pte_t __iomem *gte;
 	gen8_pte_t __iomem *end;
 	struct sgt_iter iter;
@@ -348,14 +335,14 @@ static void gen8_ggtt_clear_range(struct i915_address_space *vm,
 static void gen6_ggtt_insert_page(struct i915_address_space *vm,
 				  dma_addr_t addr,
 				  u64 offset,
-				  enum i915_cache_level level,
+				  unsigned int pat_index,
 				  u32 flags)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
 	gen6_pte_t __iomem *pte =
 		(gen6_pte_t __iomem *)ggtt->gsm + offset / I915_GTT_PAGE_SIZE;
 
-	iowrite32(vm->pte_encode(addr, level, flags), pte);
+	iowrite32(vm->pte_encode(addr, pat_index, flags), pte);
 
 	ggtt->invalidate(ggtt);
 }
@@ -368,7 +355,7 @@ static void gen6_ggtt_insert_page(struct i915_address_space *vm,
  */
 static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct i915_vma_resource *vma_res,
-				     enum i915_cache_level level,
+				     unsigned int pat_index,
 				     u32 flags)
 {
 	struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm);
@@ -385,7 +372,7 @@ static void gen6_ggtt_insert_entries(struct i915_address_space *vm,
 		iowrite32(vm->scratch[0]->encode, gte++);
 	end += (vma_res->node_size + vma_res->guard) / I915_GTT_PAGE_SIZE;
 	for_each_sgt_daddr(addr, iter, vma_res->bi.pages)
-		iowrite32(vm->pte_encode(addr, level, flags), gte++);
+		iowrite32(vm->pte_encode(addr, pat_index, flags), gte++);
 	GEM_BUG_ON(gte > end);
 
 	/* Fill the allocated but "unused" space beyond the end of the buffer */
@@ -420,14 +407,15 @@ struct insert_page {
 	struct i915_address_space *vm;
 	dma_addr_t addr;
 	u64 offset;
-	enum i915_cache_level level;
+	unsigned int pat_index;
 };
 
 static int bxt_vtd_ggtt_insert_page__cb(void *_arg)
 {
 	struct insert_page *arg = _arg;
 
-	gen8_ggtt_insert_page(arg->vm, arg->addr, arg->offset, arg->level, 0);
+	gen8_ggtt_insert_page(arg->vm, arg->addr, arg->offset,
+			      arg->pat_index, 0);
 	bxt_vtd_ggtt_wa(arg->vm);
 
 	return 0;
@@ -436,10 +424,10 @@ static int bxt_vtd_ggtt_insert_page__cb(void *_arg)
 static void bxt_vtd_ggtt_insert_page__BKL(struct i915_address_space *vm,
 					  dma_addr_t addr,
 					  u64 offset,
-					  enum i915_cache_level level,
+					  unsigned int pat_index,
 					  u32 unused)
 {
-	struct insert_page arg = { vm, addr, offset, level };
+	struct insert_page arg = { vm, addr, offset, pat_index };
 
 	stop_machine(bxt_vtd_ggtt_insert_page__cb, &arg, NULL);
 }
@@ -447,7 +435,7 @@ static void bxt_vtd_ggtt_insert_page__BKL(struct i915_address_space *vm,
 struct insert_entries {
 	struct i915_address_space *vm;
 	struct i915_vma_resource *vma_res;
-	enum i915_cache_level level;
+	unsigned int pat_index;
 	u32 flags;
 };
 
@@ -455,7 +443,8 @@ static int bxt_vtd_ggtt_insert_entries__cb(void *_arg)
 {
 	struct insert_entries *arg = _arg;
 
-	gen8_ggtt_insert_entries(arg->vm, arg->vma_res, arg->level, arg->flags);
+	gen8_ggtt_insert_entries(arg->vm, arg->vma_res,
+				 arg->pat_index, arg->flags);
 	bxt_vtd_ggtt_wa(arg->vm);
 
 	return 0;
@@ -463,10 +452,10 @@ static int bxt_vtd_ggtt_insert_entries__cb(void *_arg)
 
 static void bxt_vtd_ggtt_insert_entries__BKL(struct i915_address_space *vm,
 					     struct i915_vma_resource *vma_res,
-					     enum i915_cache_level level,
+					     unsigned int pat_index,
 					     u32 flags)
 {
-	struct insert_entries arg = { vm, vma_res, level, flags };
+	struct insert_entries arg = { vm, vma_res, pat_index, flags };
 
 	stop_machine(bxt_vtd_ggtt_insert_entries__cb, &arg, NULL);
 }
@@ -495,7 +484,7 @@ static void gen6_ggtt_clear_range(struct i915_address_space *vm,
 void intel_ggtt_bind_vma(struct i915_address_space *vm,
 			 struct i915_vm_pt_stash *stash,
 			 struct i915_vma_resource *vma_res,
-			 enum i915_cache_level cache_level,
+			 unsigned int pat_index,
 			 u32 flags)
 {
 	u32 pte_flags;
@@ -512,7 +501,7 @@ void intel_ggtt_bind_vma(struct i915_address_space *vm,
 	if (vma_res->bi.lmem)
 		pte_flags |= PTE_LM;
 
-	vm->insert_entries(vm, vma_res, cache_level, pte_flags);
+	vm->insert_entries(vm, vma_res, pat_index, pte_flags);
 	vma_res->page_sizes_gtt = I915_GTT_PAGE_SIZE;
 }
 
@@ -661,7 +650,7 @@ static int init_ggtt(struct i915_ggtt *ggtt)
 static void aliasing_gtt_bind_vma(struct i915_address_space *vm,
 				  struct i915_vm_pt_stash *stash,
 				  struct i915_vma_resource *vma_res,
-				  enum i915_cache_level cache_level,
+				  unsigned int pat_index,
 				  u32 flags)
 {
 	u32 pte_flags;
@@ -673,10 +662,10 @@ static void aliasing_gtt_bind_vma(struct i915_address_space *vm,
 
 	if (flags & I915_VMA_LOCAL_BIND)
 		ppgtt_bind_vma(&i915_vm_to_ggtt(vm)->alias->vm,
-			       stash, vma_res, cache_level, flags);
+			       stash, vma_res, pat_index, flags);
 
 	if (flags & I915_VMA_GLOBAL_BIND)
-		vm->insert_entries(vm, vma_res, cache_level, pte_flags);
+		vm->insert_entries(vm, vma_res, pat_index, pte_flags);
 
 	vma_res->bound_flags |= flags;
 }
@@ -933,7 +922,9 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size)
 
 	ggtt->vm.scratch[0]->encode =
 		ggtt->vm.pte_encode(px_dma(ggtt->vm.scratch[0]),
-				    I915_CACHE_NONE, pte_flags);
+				    i915_gem_get_pat_index(i915,
+							   I915_CACHE_NONE),
+				    pte_flags);
 
 	return 0;
 }
@@ -1022,6 +1013,11 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 	return ggtt_probe_common(ggtt, size);
 }
 
+/*
+ * For pre-gen8 platforms pat_index is the same as enum i915_cache_level,
+ * so these PTE encode functions are left with using cache_level.
+ * See translation table LEGACY_CACHELEVEL.
+ */
 static u64 snb_pte_encode(dma_addr_t addr,
 			  enum i915_cache_level level,
 			  u32 flags)
@@ -1302,7 +1298,9 @@ bool i915_ggtt_resume_vm(struct i915_address_space *vm)
 		 */
 		vma->resource->bound_flags = 0;
 		vma->ops->bind_vma(vm, NULL, vma->resource,
-				   obj ? obj->cache_level : 0,
+				   obj ? obj->pat_index :
+					 i915_gem_get_pat_index(vm->i915,
+								I915_CACHE_NONE),
 				   was_bound);
 
 		if (obj) { /* only used during resume => exclusive access */
diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h
index 9aff343beaa8b..4d6296cdbcfdd 100644
--- a/drivers/gpu/drm/i915/gt/intel_gtt.h
+++ b/drivers/gpu/drm/i915/gt/intel_gtt.h
@@ -163,8 +163,6 @@ typedef u64 gen8_pte_t;
 #define MTL_3_COH_2W	REG_FIELD_PREP(MTL_PAT_INDEX_COH_MODE_MASK, 3)
 #define MTL_2_COH_1W	REG_FIELD_PREP(MTL_PAT_INDEX_COH_MODE_MASK, 2)
 
-enum i915_cache_level;
-
 struct drm_i915_gem_object;
 struct i915_fence_reg;
 struct i915_vma;
@@ -232,7 +230,7 @@ struct i915_vma_ops {
 	void (*bind_vma)(struct i915_address_space *vm,
 			 struct i915_vm_pt_stash *stash,
 			 struct i915_vma_resource *vma_res,
-			 enum i915_cache_level cache_level,
+			 unsigned int pat_index,
 			 u32 flags);
 	/*
 	 * Unmap an object from an address space. This usually consists of
@@ -304,7 +302,7 @@ struct i915_address_space {
 		(*alloc_scratch_dma)(struct i915_address_space *vm, int sz);
 
 	u64 (*pte_encode)(dma_addr_t addr,
-			  enum i915_cache_level level,
+			  unsigned int pat_index,
 			  u32 flags); /* Create a valid PTE */
 #define PTE_READ_ONLY	BIT(0)
 #define PTE_LM		BIT(1)
@@ -319,20 +317,20 @@ struct i915_address_space {
 	void (*insert_page)(struct i915_address_space *vm,
 			    dma_addr_t addr,
 			    u64 offset,
-			    enum i915_cache_level cache_level,
+			    unsigned int pat_index,
 			    u32 flags);
 	void (*insert_entries)(struct i915_address_space *vm,
 			       struct i915_vma_resource *vma_res,
-			       enum i915_cache_level cache_level,
+			       unsigned int pat_index,
 			       u32 flags);
 	void (*raw_insert_page)(struct i915_address_space *vm,
 				dma_addr_t addr,
 				u64 offset,
-				enum i915_cache_level cache_level,
+				unsigned int pat_index,
 				u32 flags);
 	void (*raw_insert_entries)(struct i915_address_space *vm,
 				   struct i915_vma_resource *vma_res,
-				   enum i915_cache_level cache_level,
+				   unsigned int pat_index,
 				   u32 flags);
 	void (*cleanup)(struct i915_address_space *vm);
 
@@ -579,7 +577,7 @@ void ppgtt_init(struct i915_ppgtt *ppgtt, struct intel_gt *gt,
 void intel_ggtt_bind_vma(struct i915_address_space *vm,
 			 struct i915_vm_pt_stash *stash,
 			 struct i915_vma_resource *vma_res,
-			 enum i915_cache_level cache_level,
+			 unsigned int pat_index,
 			 u32 flags);
 void intel_ggtt_unbind_vma(struct i915_address_space *vm,
 			   struct i915_vma_resource *vma_res);
@@ -657,7 +655,7 @@ void gen6_ggtt_invalidate(struct i915_ggtt *ggtt);
 void ppgtt_bind_vma(struct i915_address_space *vm,
 		    struct i915_vm_pt_stash *stash,
 		    struct i915_vma_resource *vma_res,
-		    enum i915_cache_level cache_level,
+		    unsigned int pat_index,
 		    u32 flags);
 void ppgtt_unbind_vma(struct i915_address_space *vm,
 		      struct i915_vma_resource *vma_res);
diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c
index e0998879a0e1f..6023288b0e2dd 100644
--- a/drivers/gpu/drm/i915/gt/intel_migrate.c
+++ b/drivers/gpu/drm/i915/gt/intel_migrate.c
@@ -45,7 +45,9 @@ static void xehpsdv_toggle_pdes(struct i915_address_space *vm,
 	 * Insert a dummy PTE into every PT that will map to LMEM to ensure
 	 * we have a correctly setup PDE structure for later use.
 	 */
-	vm->insert_page(vm, 0, d->offset, I915_CACHE_NONE, PTE_LM);
+	vm->insert_page(vm, 0, d->offset,
+			i915_gem_get_pat_index(vm->i915, I915_CACHE_NONE),
+			PTE_LM);
 	GEM_BUG_ON(!pt->is_compact);
 	d->offset += SZ_2M;
 }
@@ -63,7 +65,9 @@ static void xehpsdv_insert_pte(struct i915_address_space *vm,
 	 * alignment is 64K underneath for the pt, and we are careful
 	 * not to access the space in the void.
 	 */
-	vm->insert_page(vm, px_dma(pt), d->offset, I915_CACHE_NONE, PTE_LM);
+	vm->insert_page(vm, px_dma(pt), d->offset,
+			i915_gem_get_pat_index(vm->i915, I915_CACHE_NONE),
+			PTE_LM);
 	d->offset += SZ_64K;
 }
 
@@ -73,7 +77,8 @@ static void insert_pte(struct i915_address_space *vm,
 {
 	struct insert_pte_data *d = data;
 
-	vm->insert_page(vm, px_dma(pt), d->offset, I915_CACHE_NONE,
+	vm->insert_page(vm, px_dma(pt), d->offset,
+			i915_gem_get_pat_index(vm->i915, I915_CACHE_NONE),
 			i915_gem_object_is_lmem(pt->base) ? PTE_LM : 0);
 	d->offset += PAGE_SIZE;
 }
@@ -356,13 +361,13 @@ static int max_pte_pkt_size(struct i915_request *rq, int pkt)
 
 static int emit_pte(struct i915_request *rq,
 		    struct sgt_dma *it,
-		    enum i915_cache_level cache_level,
+		    unsigned int pat_index,
 		    bool is_lmem,
 		    u64 offset,
 		    int length)
 {
 	bool has_64K_pages = HAS_64K_PAGES(rq->engine->i915);
-	const u64 encode = rq->context->vm->pte_encode(0, cache_level,
+	const u64 encode = rq->context->vm->pte_encode(0, pat_index,
 						       is_lmem ? PTE_LM : 0);
 	struct intel_ring *ring = rq->ring;
 	int pkt, dword_length;
@@ -673,17 +678,17 @@ int
 intel_context_migrate_copy(struct intel_context *ce,
 			   const struct i915_deps *deps,
 			   struct scatterlist *src,
-			   enum i915_cache_level src_cache_level,
+			   unsigned int src_pat_index,
 			   bool src_is_lmem,
 			   struct scatterlist *dst,
-			   enum i915_cache_level dst_cache_level,
+			   unsigned int dst_pat_index,
 			   bool dst_is_lmem,
 			   struct i915_request **out)
 {
 	struct sgt_dma it_src = sg_sgt(src), it_dst = sg_sgt(dst), it_ccs;
 	struct drm_i915_private *i915 = ce->engine->i915;
 	u64 ccs_bytes_to_cpy = 0, bytes_to_cpy;
-	enum i915_cache_level ccs_cache_level;
+	unsigned int ccs_pat_index;
 	u32 src_offset, dst_offset;
 	u8 src_access, dst_access;
 	struct i915_request *rq;
@@ -707,12 +712,12 @@ intel_context_migrate_copy(struct intel_context *ce,
 		dst_sz = scatter_list_length(dst);
 		if (src_is_lmem) {
 			it_ccs = it_dst;
-			ccs_cache_level = dst_cache_level;
+			ccs_pat_index = dst_pat_index;
 			ccs_is_src = false;
 		} else if (dst_is_lmem) {
 			bytes_to_cpy = dst_sz;
 			it_ccs = it_src;
-			ccs_cache_level = src_cache_level;
+			ccs_pat_index = src_pat_index;
 			ccs_is_src = true;
 		}
 
@@ -773,7 +778,7 @@ intel_context_migrate_copy(struct intel_context *ce,
 		src_sz = calculate_chunk_sz(i915, src_is_lmem,
 					    bytes_to_cpy, ccs_bytes_to_cpy);
 
-		len = emit_pte(rq, &it_src, src_cache_level, src_is_lmem,
+		len = emit_pte(rq, &it_src, src_pat_index, src_is_lmem,
 			       src_offset, src_sz);
 		if (!len) {
 			err = -EINVAL;
@@ -784,7 +789,7 @@ intel_context_migrate_copy(struct intel_context *ce,
 			goto out_rq;
 		}
 
-		err = emit_pte(rq, &it_dst, dst_cache_level, dst_is_lmem,
+		err = emit_pte(rq, &it_dst, dst_pat_index, dst_is_lmem,
 			       dst_offset, len);
 		if (err < 0)
 			goto out_rq;
@@ -811,7 +816,7 @@ intel_context_migrate_copy(struct intel_context *ce,
 				goto out_rq;
 
 			ccs_sz = GET_CCS_BYTES(i915, len);
-			err = emit_pte(rq, &it_ccs, ccs_cache_level, false,
+			err = emit_pte(rq, &it_ccs, ccs_pat_index, false,
 				       ccs_is_src ? src_offset : dst_offset,
 				       ccs_sz);
 			if (err < 0)
@@ -979,7 +984,7 @@ int
 intel_context_migrate_clear(struct intel_context *ce,
 			    const struct i915_deps *deps,
 			    struct scatterlist *sg,
-			    enum i915_cache_level cache_level,
+			    unsigned int pat_index,
 			    bool is_lmem,
 			    u32 value,
 			    struct i915_request **out)
@@ -1027,7 +1032,7 @@ intel_context_migrate_clear(struct intel_context *ce,
 		if (err)
 			goto out_rq;
 
-		len = emit_pte(rq, &it, cache_level, is_lmem, offset, CHUNK_SZ);
+		len = emit_pte(rq, &it, pat_index, is_lmem, offset, CHUNK_SZ);
 		if (len <= 0) {
 			err = len;
 			goto out_rq;
@@ -1074,10 +1079,10 @@ int intel_migrate_copy(struct intel_migrate *m,
 		       struct i915_gem_ww_ctx *ww,
 		       const struct i915_deps *deps,
 		       struct scatterlist *src,
-		       enum i915_cache_level src_cache_level,
+		       unsigned int src_pat_index,
 		       bool src_is_lmem,
 		       struct scatterlist *dst,
-		       enum i915_cache_level dst_cache_level,
+		       unsigned int dst_pat_index,
 		       bool dst_is_lmem,
 		       struct i915_request **out)
 {
@@ -1098,8 +1103,8 @@ int intel_migrate_copy(struct intel_migrate *m,
 		goto out;
 
 	err = intel_context_migrate_copy(ce, deps,
-					 src, src_cache_level, src_is_lmem,
-					 dst, dst_cache_level, dst_is_lmem,
+					 src, src_pat_index, src_is_lmem,
+					 dst, dst_pat_index, dst_is_lmem,
 					 out);
 
 	intel_context_unpin(ce);
@@ -1113,7 +1118,7 @@ intel_migrate_clear(struct intel_migrate *m,
 		    struct i915_gem_ww_ctx *ww,
 		    const struct i915_deps *deps,
 		    struct scatterlist *sg,
-		    enum i915_cache_level cache_level,
+		    unsigned int pat_index,
 		    bool is_lmem,
 		    u32 value,
 		    struct i915_request **out)
@@ -1134,7 +1139,7 @@ intel_migrate_clear(struct intel_migrate *m,
 	if (err)
 		goto out;
 
-	err = intel_context_migrate_clear(ce, deps, sg, cache_level,
+	err = intel_context_migrate_clear(ce, deps, sg, pat_index,
 					  is_lmem, value, out);
 
 	intel_context_unpin(ce);
diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.h b/drivers/gpu/drm/i915/gt/intel_migrate.h
index ccc677ec4aa33..11fc09a00c4be 100644
--- a/drivers/gpu/drm/i915/gt/intel_migrate.h
+++ b/drivers/gpu/drm/i915/gt/intel_migrate.h
@@ -16,7 +16,6 @@ struct i915_request;
 struct i915_gem_ww_ctx;
 struct intel_gt;
 struct scatterlist;
-enum i915_cache_level;
 
 int intel_migrate_init(struct intel_migrate *m, struct intel_gt *gt);
 
@@ -26,20 +25,20 @@ int intel_migrate_copy(struct intel_migrate *m,
 		       struct i915_gem_ww_ctx *ww,
 		       const struct i915_deps *deps,
 		       struct scatterlist *src,
-		       enum i915_cache_level src_cache_level,
+		       unsigned int src_pat_index,
 		       bool src_is_lmem,
 		       struct scatterlist *dst,
-		       enum i915_cache_level dst_cache_level,
+		       unsigned int dst_pat_index,
 		       bool dst_is_lmem,
 		       struct i915_request **out);
 
 int intel_context_migrate_copy(struct intel_context *ce,
 			       const struct i915_deps *deps,
 			       struct scatterlist *src,
-			       enum i915_cache_level src_cache_level,
+			       unsigned int src_pat_index,
 			       bool src_is_lmem,
 			       struct scatterlist *dst,
-			       enum i915_cache_level dst_cache_level,
+			       unsigned int dst_pat_index,
 			       bool dst_is_lmem,
 			       struct i915_request **out);
 
@@ -48,7 +47,7 @@ intel_migrate_clear(struct intel_migrate *m,
 		    struct i915_gem_ww_ctx *ww,
 		    const struct i915_deps *deps,
 		    struct scatterlist *sg,
-		    enum i915_cache_level cache_level,
+		    unsigned int pat_index,
 		    bool is_lmem,
 		    u32 value,
 		    struct i915_request **out);
@@ -56,7 +55,7 @@ int
 intel_context_migrate_clear(struct intel_context *ce,
 			    const struct i915_deps *deps,
 			    struct scatterlist *sg,
-			    enum i915_cache_level cache_level,
+			    unsigned int pat_index,
 			    bool is_lmem,
 			    u32 value,
 			    struct i915_request **out);
diff --git a/drivers/gpu/drm/i915/gt/intel_ppgtt.c b/drivers/gpu/drm/i915/gt/intel_ppgtt.c
index 7ecfa672f738d..436756bfbb1a4 100644
--- a/drivers/gpu/drm/i915/gt/intel_ppgtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ppgtt.c
@@ -181,7 +181,7 @@ struct i915_ppgtt *i915_ppgtt_create(struct intel_gt *gt,
 void ppgtt_bind_vma(struct i915_address_space *vm,
 		    struct i915_vm_pt_stash *stash,
 		    struct i915_vma_resource *vma_res,
-		    enum i915_cache_level cache_level,
+		    unsigned int pat_index,
 		    u32 flags)
 {
 	u32 pte_flags;
@@ -199,7 +199,7 @@ void ppgtt_bind_vma(struct i915_address_space *vm,
 	if (vma_res->bi.lmem)
 		pte_flags |= PTE_LM;
 
-	vm->insert_entries(vm, vma_res, cache_level, pte_flags);
+	vm->insert_entries(vm, vma_res, pat_index, pte_flags);
 	wmb();
 }
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_migrate.c b/drivers/gpu/drm/i915/gt/selftest_migrate.c
index e677f2da093df..3def5ca72decf 100644
--- a/drivers/gpu/drm/i915/gt/selftest_migrate.c
+++ b/drivers/gpu/drm/i915/gt/selftest_migrate.c
@@ -137,7 +137,7 @@ static int copy(struct intel_migrate *migrate,
 static int intel_context_copy_ccs(struct intel_context *ce,
 				  const struct i915_deps *deps,
 				  struct scatterlist *sg,
-				  enum i915_cache_level cache_level,
+				  unsigned int pat_index,
 				  bool write_to_ccs,
 				  struct i915_request **out)
 {
@@ -185,7 +185,7 @@ static int intel_context_copy_ccs(struct intel_context *ce,
 		if (err)
 			goto out_rq;
 
-		len = emit_pte(rq, &it, cache_level, true, offset, CHUNK_SZ);
+		len = emit_pte(rq, &it, pat_index, true, offset, CHUNK_SZ);
 		if (len <= 0) {
 			err = len;
 			goto out_rq;
@@ -223,7 +223,7 @@ intel_migrate_ccs_copy(struct intel_migrate *m,
 		       struct i915_gem_ww_ctx *ww,
 		       const struct i915_deps *deps,
 		       struct scatterlist *sg,
-		       enum i915_cache_level cache_level,
+		       unsigned int pat_index,
 		       bool write_to_ccs,
 		       struct i915_request **out)
 {
@@ -243,7 +243,7 @@ intel_migrate_ccs_copy(struct intel_migrate *m,
 	if (err)
 		goto out;
 
-	err = intel_context_copy_ccs(ce, deps, sg, cache_level,
+	err = intel_context_copy_ccs(ce, deps, sg, pat_index,
 				     write_to_ccs, out);
 
 	intel_context_unpin(ce);
@@ -300,7 +300,7 @@ static int clear(struct intel_migrate *migrate,
 			/* Write the obj data into ccs surface */
 			err = intel_migrate_ccs_copy(migrate, &ww, NULL,
 						     obj->mm.pages->sgl,
-						     obj->cache_level,
+						     obj->pat_index,
 						     true, &rq);
 			if (rq && !err) {
 				if (i915_request_wait(rq, 0, HZ) < 0) {
@@ -351,7 +351,7 @@ static int clear(struct intel_migrate *migrate,
 
 			err = intel_migrate_ccs_copy(migrate, &ww, NULL,
 						     obj->mm.pages->sgl,
-						     obj->cache_level,
+						     obj->pat_index,
 						     false, &rq);
 			if (rq && !err) {
 				if (i915_request_wait(rq, 0, HZ) < 0) {
@@ -414,9 +414,9 @@ static int __migrate_copy(struct intel_migrate *migrate,
 			  struct i915_request **out)
 {
 	return intel_migrate_copy(migrate, ww, NULL,
-				  src->mm.pages->sgl, src->cache_level,
+				  src->mm.pages->sgl, src->pat_index,
 				  i915_gem_object_is_lmem(src),
-				  dst->mm.pages->sgl, dst->cache_level,
+				  dst->mm.pages->sgl, dst->pat_index,
 				  i915_gem_object_is_lmem(dst),
 				  out);
 }
@@ -428,9 +428,9 @@ static int __global_copy(struct intel_migrate *migrate,
 			 struct i915_request **out)
 {
 	return intel_context_migrate_copy(migrate->context, NULL,
-					  src->mm.pages->sgl, src->cache_level,
+					  src->mm.pages->sgl, src->pat_index,
 					  i915_gem_object_is_lmem(src),
-					  dst->mm.pages->sgl, dst->cache_level,
+					  dst->mm.pages->sgl, dst->pat_index,
 					  i915_gem_object_is_lmem(dst),
 					  out);
 }
@@ -455,7 +455,7 @@ static int __migrate_clear(struct intel_migrate *migrate,
 {
 	return intel_migrate_clear(migrate, ww, NULL,
 				   obj->mm.pages->sgl,
-				   obj->cache_level,
+				   obj->pat_index,
 				   i915_gem_object_is_lmem(obj),
 				   value, out);
 }
@@ -468,7 +468,7 @@ static int __global_clear(struct intel_migrate *migrate,
 {
 	return intel_context_migrate_clear(migrate->context, NULL,
 					   obj->mm.pages->sgl,
-					   obj->cache_level,
+					   obj->pat_index,
 					   i915_gem_object_is_lmem(obj),
 					   value, out);
 }
@@ -648,7 +648,7 @@ static int live_emit_pte_full_ring(void *arg)
 	 */
 	pr_info("%s emite_pte ring space=%u\n", __func__, rq->ring->space);
 	it = sg_sgt(obj->mm.pages->sgl);
-	len = emit_pte(rq, &it, obj->cache_level, false, 0, CHUNK_SZ);
+	len = emit_pte(rq, &it, obj->pat_index, false, 0, CHUNK_SZ);
 	if (!len) {
 		err = -EINVAL;
 		goto out_rq;
@@ -844,7 +844,7 @@ static int wrap_ktime_compare(const void *A, const void *B)
 
 static int __perf_clear_blt(struct intel_context *ce,
 			    struct scatterlist *sg,
-			    enum i915_cache_level cache_level,
+			    unsigned int pat_index,
 			    bool is_lmem,
 			    size_t sz)
 {
@@ -858,7 +858,7 @@ static int __perf_clear_blt(struct intel_context *ce,
 
 		t0 = ktime_get();
 
-		err = intel_context_migrate_clear(ce, NULL, sg, cache_level,
+		err = intel_context_migrate_clear(ce, NULL, sg, pat_index,
 						  is_lmem, 0, &rq);
 		if (rq) {
 			if (i915_request_wait(rq, 0, MAX_SCHEDULE_TIMEOUT) < 0)
@@ -904,7 +904,8 @@ static int perf_clear_blt(void *arg)
 
 		err = __perf_clear_blt(gt->migrate.context,
 				       dst->mm.pages->sgl,
-				       I915_CACHE_NONE,
+				       i915_gem_get_pat_index(gt->i915,
+							      I915_CACHE_NONE),
 				       i915_gem_object_is_lmem(dst),
 				       sizes[i]);
 
@@ -919,10 +920,10 @@ static int perf_clear_blt(void *arg)
 
 static int __perf_copy_blt(struct intel_context *ce,
 			   struct scatterlist *src,
-			   enum i915_cache_level src_cache_level,
+			   unsigned int src_pat_index,
 			   bool src_is_lmem,
 			   struct scatterlist *dst,
-			   enum i915_cache_level dst_cache_level,
+			   unsigned int dst_pat_index,
 			   bool dst_is_lmem,
 			   size_t sz)
 {
@@ -937,9 +938,9 @@ static int __perf_copy_blt(struct intel_context *ce,
 		t0 = ktime_get();
 
 		err = intel_context_migrate_copy(ce, NULL,
-						 src, src_cache_level,
+						 src, src_pat_index,
 						 src_is_lmem,
-						 dst, dst_cache_level,
+						 dst, dst_pat_index,
 						 dst_is_lmem,
 						 &rq);
 		if (rq) {
@@ -994,10 +995,12 @@ static int perf_copy_blt(void *arg)
 
 		err = __perf_copy_blt(gt->migrate.context,
 				      src->mm.pages->sgl,
-				      I915_CACHE_NONE,
+				      i915_gem_get_pat_index(gt->i915,
+							     I915_CACHE_NONE),
 				      i915_gem_object_is_lmem(src),
 				      dst->mm.pages->sgl,
-				      I915_CACHE_NONE,
+				      i915_gem_get_pat_index(gt->i915,
+							     I915_CACHE_NONE),
 				      i915_gem_object_is_lmem(dst),
 				      sz);
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_reset.c b/drivers/gpu/drm/i915/gt/selftest_reset.c
index a9e0a91bc0e02..79aa6ac66ad2f 100644
--- a/drivers/gpu/drm/i915/gt/selftest_reset.c
+++ b/drivers/gpu/drm/i915/gt/selftest_reset.c
@@ -86,7 +86,9 @@ __igt_reset_stolen(struct intel_gt *gt,
 
 		ggtt->vm.insert_page(&ggtt->vm, dma,
 				     ggtt->error_capture.start,
-				     I915_CACHE_NONE, 0);
+				     i915_gem_get_pat_index(gt->i915,
+							    I915_CACHE_NONE),
+				     0);
 		mb();
 
 		s = io_mapping_map_wc(&ggtt->iomap,
@@ -127,7 +129,9 @@ __igt_reset_stolen(struct intel_gt *gt,
 
 		ggtt->vm.insert_page(&ggtt->vm, dma,
 				     ggtt->error_capture.start,
-				     I915_CACHE_NONE, 0);
+				     i915_gem_get_pat_index(gt->i915,
+							    I915_CACHE_NONE),
+				     0);
 		mb();
 
 		s = io_mapping_map_wc(&ggtt->iomap,
diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c
index 9f536c251179a..39c3ec12df1ab 100644
--- a/drivers/gpu/drm/i915/gt/selftest_timeline.c
+++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c
@@ -836,7 +836,7 @@ static int setup_watcher(struct hwsp_watcher *w, struct intel_gt *gt,
 		return PTR_ERR(obj);
 
 	/* keep the same cache settings as timeline */
-	i915_gem_object_set_cache_coherency(obj, tl->hwsp_ggtt->obj->cache_level);
+	i915_gem_object_set_pat_index(obj, tl->hwsp_ggtt->obj->pat_index);
 	w->map = i915_gem_object_pin_map_unlocked(obj,
 						  page_unmask_bits(tl->hwsp_ggtt->obj->mm.mapping));
 	if (IS_ERR(w->map)) {
diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c
index e6cac1f15d6ed..4493c8518e91b 100644
--- a/drivers/gpu/drm/i915/gt/selftest_tlb.c
+++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c
@@ -36,6 +36,8 @@ pte_tlbinv(struct intel_context *ce,
 	   u64 length,
 	   struct rnd_state *prng)
 {
+	const unsigned int pat_index =
+		i915_gem_get_pat_index(ce->vm->i915, I915_CACHE_NONE);
 	struct drm_i915_gem_object *batch;
 	struct drm_mm_node vb_node;
 	struct i915_request *rq;
@@ -155,7 +157,7 @@ pte_tlbinv(struct intel_context *ce,
 		/* Flip the PTE between A and B */
 		if (i915_gem_object_is_lmem(vb->obj))
 			pte_flags |= PTE_LM;
-		ce->vm->insert_entries(ce->vm, &vb_res, 0, pte_flags);
+		ce->vm->insert_entries(ce->vm, &vb_res, pat_index, pte_flags);
 
 		/* Flush the PTE update to concurrent HW */
 		tlbinv(ce->vm, addr & -length, length);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 4ec7df9ed5ff3..e34c791200903 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -957,9 +957,15 @@ static void uc_fw_bind_ggtt(struct intel_uc_fw *uc_fw)
 		pte_flags |= PTE_LM;
 
 	if (ggtt->vm.raw_insert_entries)
-		ggtt->vm.raw_insert_entries(&ggtt->vm, dummy, I915_CACHE_NONE, pte_flags);
+		ggtt->vm.raw_insert_entries(&ggtt->vm, dummy,
+					    i915_gem_get_pat_index(ggtt->vm.i915,
+								   I915_CACHE_NONE),
+					    pte_flags);
 	else
-		ggtt->vm.insert_entries(&ggtt->vm, dummy, I915_CACHE_NONE, pte_flags);
+		ggtt->vm.insert_entries(&ggtt->vm, dummy,
+					i915_gem_get_pat_index(ggtt->vm.i915,
+							       I915_CACHE_NONE),
+					pte_flags);
 }
 
 static void uc_fw_unbind_ggtt(struct intel_uc_fw *uc_fw)
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 80c2bf98e341c..e39937c82d730 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -138,21 +138,54 @@ static const char *stringify_vma_type(const struct i915_vma *vma)
 	return "ppgtt";
 }
 
-static const char *i915_cache_level_str(struct drm_i915_private *i915, int type)
-{
-	switch (type) {
-	case I915_CACHE_NONE: return " uncached";
-	case I915_CACHE_LLC: return HAS_LLC(i915) ? " LLC" : " snooped";
-	case I915_CACHE_L3_LLC: return " L3+LLC";
-	case I915_CACHE_WT: return " WT";
-	default: return "";
+static const char *i915_cache_level_str(struct drm_i915_gem_object *obj)
+{
+	struct drm_i915_private *i915 = obj_to_i915(obj);
+
+	if (IS_METEORLAKE(i915)) {
+		switch (obj->pat_index) {
+		case 0: return " WB";
+		case 1: return " WT";
+		case 2: return " UC";
+		case 3: return " WB (1-Way Coh)";
+		case 4: return " WB (2-Way Coh)";
+		default: return " not defined";
+		}
+	} else if (IS_PONTEVECCHIO(i915)) {
+		switch (obj->pat_index) {
+		case 0: return " UC";
+		case 1: return " WC";
+		case 2: return " WT";
+		case 3: return " WB";
+		case 4: return " WT (CLOS1)";
+		case 5: return " WB (CLOS1)";
+		case 6: return " WT (CLOS2)";
+		case 7: return " WT (CLOS2)";
+		default: return " not defined";
+		}
+	} else if (GRAPHICS_VER(i915) >= 12) {
+		switch (obj->pat_index) {
+		case 0: return " WB";
+		case 1: return " WC";
+		case 2: return " WT";
+		case 3: return " UC";
+		default: return " not defined";
+		}
+	} else {
+		switch (obj->pat_index) {
+		case 0: return " UC";
+		case 1: return HAS_LLC(i915) ?
+			       " LLC" : " snooped";
+		case 2: return " L3+LLC";
+		case 3: return " WT";
+		default: return " not defined";
+		}
 	}
 }
 
 void
 i915_debugfs_describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 {
-	struct drm_i915_private *dev_priv = to_i915(obj->base.dev);
 	struct i915_vma *vma;
 	int pin_count = 0;
 
@@ -164,7 +197,7 @@ i915_debugfs_describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		   obj->base.size / 1024,
 		   obj->read_domains,
 		   obj->write_domain,
-		   i915_cache_level_str(dev_priv, obj->cache_level),
+		   i915_cache_level_str(obj),
 		   obj->mm.dirty ? " dirty" : "",
 		   obj->mm.madv == I915_MADV_DONTNEED ? " purgeable" : "");
 	if (obj->base.name)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 0a78bdbd36b19..e70b762f0b032 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -420,8 +420,11 @@ i915_gem_gtt_pread(struct drm_i915_gem_object *obj,
 		page_length = remain < page_length ? remain : page_length;
 		if (drm_mm_node_allocated(&node)) {
 			ggtt->vm.insert_page(&ggtt->vm,
-					     i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
-					     node.start, I915_CACHE_NONE, 0);
+					     i915_gem_object_get_dma_address(obj,
+									     offset >> PAGE_SHIFT),
+					     node.start,
+					     i915_gem_get_pat_index(i915,
+								    I915_CACHE_NONE), 0);
 		} else {
 			page_base += offset & PAGE_MASK;
 		}
@@ -598,8 +601,11 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj,
 			/* flush the write before we modify the GGTT */
 			intel_gt_flush_ggtt_writes(ggtt->vm.gt);
 			ggtt->vm.insert_page(&ggtt->vm,
-					     i915_gem_object_get_dma_address(obj, offset >> PAGE_SHIFT),
-					     node.start, I915_CACHE_NONE, 0);
+					     i915_gem_object_get_dma_address(obj,
+									     offset >> PAGE_SHIFT),
+					     node.start,
+					     i915_gem_get_pat_index(i915,
+								    I915_CACHE_NONE), 0);
 			wmb(); /* flush modifications to the GGTT (insert_page) */
 		} else {
 			page_base += offset & PAGE_MASK;
@@ -1142,6 +1148,19 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 	unsigned int i;
 	int ret;
 
+	/*
+	 * In the proccess of replacing cache_level with pat_index a tricky
+	 * dependency is created on the definition of the enum i915_cache_level.
+	 * in case this enum is changed, PTE encode would be broken.
+	 * Add a WARNING here. And remove when we completely quit using this
+	 * enum
+	 */
+	BUILD_BUG_ON(I915_CACHE_NONE != 0 ||
+		     I915_CACHE_LLC != 1 ||
+		     I915_CACHE_L3_LLC != 2 ||
+		     I915_CACHE_WT != 3 ||
+		     I915_MAX_CACHE_LEVEL != 4);
+
 	/* We need to fallback to 4K pages if host doesn't support huge gtt. */
 	if (intel_vgpu_active(dev_priv) && !intel_vgpu_has_huge_gtt(dev_priv))
 		RUNTIME_INFO(dev_priv)->page_sizes = I915_GTT_PAGE_SIZE_4K;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 8ffdd7f47e933..5665364999543 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1122,10 +1122,14 @@ i915_vma_coredump_create(const struct intel_gt *gt,
 			mutex_lock(&ggtt->error_mutex);
 			if (ggtt->vm.raw_insert_page)
 				ggtt->vm.raw_insert_page(&ggtt->vm, dma, slot,
-							 I915_CACHE_NONE, 0);
+							 i915_gem_get_pat_index(gt->i915,
+										I915_CACHE_NONE),
+							 0);
 			else
 				ggtt->vm.insert_page(&ggtt->vm, dma, slot,
-						     I915_CACHE_NONE, 0);
+						     i915_gem_get_pat_index(gt->i915,
+									    I915_CACHE_NONE),
+						     0);
 			mb();
 
 			s = io_mapping_map_wc(&ggtt->iomap, slot, PAGE_SIZE);
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 20a44788999e5..a814775a363df 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -315,7 +315,7 @@ struct i915_vma_work {
 	struct i915_vma_resource *vma_res;
 	struct drm_i915_gem_object *obj;
 	struct i915_sw_dma_fence_cb cb;
-	enum i915_cache_level cache_level;
+	unsigned int pat_index;
 	unsigned int flags;
 };
 
@@ -334,7 +334,7 @@ static void __vma_bind(struct dma_fence_work *work)
 		return;
 
 	vma_res->ops->bind_vma(vma_res->vm, &vw->stash,
-			       vma_res, vw->cache_level, vw->flags);
+			       vma_res, vw->pat_index, vw->flags);
 }
 
 static void __vma_release(struct dma_fence_work *work)
@@ -426,7 +426,7 @@ i915_vma_resource_init_from_vma(struct i915_vma_resource *vma_res,
 /**
  * i915_vma_bind - Sets up PTEs for an VMA in it's corresponding address space.
  * @vma: VMA to map
- * @cache_level: mapping cache level
+ * @pat_index: PAT index to set in PTE
  * @flags: flags like global or local mapping
  * @work: preallocated worker for allocating and binding the PTE
  * @vma_res: pointer to a preallocated vma resource. The resource is either
@@ -437,7 +437,7 @@ i915_vma_resource_init_from_vma(struct i915_vma_resource *vma_res,
  * Note that DMA addresses are also the only part of the SG table we care about.
  */
 int i915_vma_bind(struct i915_vma *vma,
-		  enum i915_cache_level cache_level,
+		  unsigned int pat_index,
 		  u32 flags,
 		  struct i915_vma_work *work,
 		  struct i915_vma_resource *vma_res)
@@ -507,7 +507,7 @@ int i915_vma_bind(struct i915_vma *vma,
 		struct dma_fence *prev;
 
 		work->vma_res = i915_vma_resource_get(vma->resource);
-		work->cache_level = cache_level;
+		work->pat_index = pat_index;
 		work->flags = bind_flags;
 
 		/*
@@ -537,7 +537,7 @@ int i915_vma_bind(struct i915_vma *vma,
 
 			return ret;
 		}
-		vma->ops->bind_vma(vma->vm, NULL, vma->resource, cache_level,
+		vma->ops->bind_vma(vma->vm, NULL, vma->resource, pat_index,
 				   bind_flags);
 	}
 
@@ -814,7 +814,7 @@ i915_vma_insert(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
 	color = 0;
 
 	if (i915_vm_has_cache_coloring(vma->vm))
-		color = vma->obj->cache_level;
+		color = vma->obj->pat_index;
 
 	if (flags & PIN_OFFSET_FIXED) {
 		u64 offset = flags & PIN_OFFSET_MASK;
@@ -1518,7 +1518,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
 
 	GEM_BUG_ON(!vma->pages);
 	err = i915_vma_bind(vma,
-			    vma->obj->cache_level,
+			    vma->obj->pat_index,
 			    flags, work, vma_res);
 	vma_res = NULL;
 	if (err)
diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
index 38c8c66ed724a..9a9729205d5be 100644
--- a/drivers/gpu/drm/i915/i915_vma.h
+++ b/drivers/gpu/drm/i915/i915_vma.h
@@ -250,7 +250,7 @@ i915_vma_compare(struct i915_vma *vma,
 
 struct i915_vma_work *i915_vma_work(void);
 int i915_vma_bind(struct i915_vma *vma,
-		  enum i915_cache_level cache_level,
+		  unsigned int pat_index,
 		  u32 flags,
 		  struct i915_vma_work *work,
 		  struct i915_vma_resource *vma_res);
diff --git a/drivers/gpu/drm/i915/i915_vma_types.h b/drivers/gpu/drm/i915/i915_vma_types.h
index 77fda2244d161..64472b7f0e770 100644
--- a/drivers/gpu/drm/i915/i915_vma_types.h
+++ b/drivers/gpu/drm/i915/i915_vma_types.h
@@ -32,8 +32,6 @@
 
 #include "gem/i915_gem_object_types.h"
 
-enum i915_cache_level;
-
 /**
  * DOC: Global GTT views
  *
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem.c b/drivers/gpu/drm/i915/selftests/i915_gem.c
index d91d0ade8abdd..61da4ed9d521b 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem.c
@@ -57,7 +57,10 @@ static void trash_stolen(struct drm_i915_private *i915)
 		u32 __iomem *s;
 		int x;
 
-		ggtt->vm.insert_page(&ggtt->vm, dma, slot, I915_CACHE_NONE, 0);
+		ggtt->vm.insert_page(&ggtt->vm, dma, slot,
+				     i915_gem_get_pat_index(i915,
+							    I915_CACHE_NONE),
+				     0);
 
 		s = io_mapping_map_atomic_wc(&ggtt->iomap, slot);
 		for (x = 0; x < PAGE_SIZE / sizeof(u32); x++) {
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
index f68ef4074088a..f8fe3681c3dc8 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c
@@ -246,7 +246,7 @@ static int igt_evict_for_cache_color(void *arg)
 	struct drm_mm_node target = {
 		.start = I915_GTT_PAGE_SIZE * 2,
 		.size = I915_GTT_PAGE_SIZE,
-		.color = I915_CACHE_LLC,
+		.color = i915_gem_get_pat_index(gt->i915, I915_CACHE_LLC),
 	};
 	struct drm_i915_gem_object *obj;
 	struct i915_vma *vma;
@@ -309,7 +309,7 @@ static int igt_evict_for_cache_color(void *arg)
 	/* Attempt to remove the first *pinned* vma, by removing the (empty)
 	 * neighbour -- this should fail.
 	 */
-	target.color = I915_CACHE_L3_LLC;
+	target.color = i915_gem_get_pat_index(gt->i915, I915_CACHE_L3_LLC);
 
 	mutex_lock(&ggtt->vm.mutex);
 	err = i915_gem_evict_for_node(&ggtt->vm, NULL, &target, 0);
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 154801f1c4685..36940ef10108d 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -135,7 +135,7 @@ fake_dma_object(struct drm_i915_private *i915, u64 size)
 
 	obj->write_domain = I915_GEM_DOMAIN_CPU;
 	obj->read_domains = I915_GEM_DOMAIN_CPU;
-	obj->cache_level = I915_CACHE_NONE;
+	obj->pat_index = i915_gem_get_pat_index(i915, I915_CACHE_NONE);
 
 	/* Preallocate the "backing storage" */
 	if (i915_gem_object_pin_pages_unlocked(obj))
@@ -359,7 +359,9 @@ static int lowlevel_hole(struct i915_address_space *vm,
 
 			with_intel_runtime_pm(vm->gt->uncore->rpm, wakeref)
 			  vm->insert_entries(vm, mock_vma_res,
-						   I915_CACHE_NONE, 0);
+					     i915_gem_get_pat_index(vm->i915,
+								    I915_CACHE_NONE),
+					     0);
 		}
 		count = n;
 
@@ -1377,7 +1379,10 @@ static int igt_ggtt_page(void *arg)
 
 		ggtt->vm.insert_page(&ggtt->vm,
 				     i915_gem_object_get_dma_address(obj, 0),
-				     offset, I915_CACHE_NONE, 0);
+				     offset,
+				     i915_gem_get_pat_index(i915,
+							    I915_CACHE_NONE),
+				     0);
 	}
 
 	order = i915_random_order(count, &prng);
@@ -1510,7 +1515,7 @@ static int reserve_gtt_with_resource(struct i915_vma *vma, u64 offset)
 	mutex_lock(&vm->mutex);
 	err = i915_gem_gtt_reserve(vm, NULL, &vma->node, obj->base.size,
 				   offset,
-				   obj->cache_level,
+				   obj->pat_index,
 				   0);
 	if (!err) {
 		i915_vma_resource_init_from_vma(vma_res, vma);
@@ -1690,7 +1695,7 @@ static int insert_gtt_with_resource(struct i915_vma *vma)
 
 	mutex_lock(&vm->mutex);
 	err = i915_gem_gtt_insert(vm, NULL, &vma->node, obj->base.size, 0,
-				  obj->cache_level, 0, vm->total, 0);
+				  obj->pat_index, 0, vm->total, 0);
 	if (!err) {
 		i915_vma_resource_init_from_vma(vma_res, vma);
 		vma->resource = vma_res;
diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
index 3b18e5905c86b..d985d9bae2e8a 100644
--- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c
+++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c
@@ -1070,7 +1070,9 @@ static int igt_lmem_write_cpu(void *arg)
 	/* Put the pages into a known state -- from the gpu for added fun */
 	intel_engine_pm_get(engine);
 	err = intel_context_migrate_clear(engine->gt->migrate.context, NULL,
-					  obj->mm.pages->sgl, I915_CACHE_NONE,
+					  obj->mm.pages->sgl,
+					  i915_gem_get_pat_index(i915,
+								 I915_CACHE_NONE),
 					  true, 0xdeadbeaf, &rq);
 	if (rq) {
 		dma_resv_add_fence(obj->base.resv, &rq->fence,
diff --git a/drivers/gpu/drm/i915/selftests/mock_gtt.c b/drivers/gpu/drm/i915/selftests/mock_gtt.c
index ece97e4faacb9..a516c0aa88fd3 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gtt.c
@@ -27,21 +27,21 @@
 static void mock_insert_page(struct i915_address_space *vm,
 			     dma_addr_t addr,
 			     u64 offset,
-			     enum i915_cache_level level,
+			     unsigned int pat_index,
 			     u32 flags)
 {
 }
 
 static void mock_insert_entries(struct i915_address_space *vm,
 				struct i915_vma_resource *vma_res,
-				enum i915_cache_level level, u32 flags)
+				unsigned int pat_index, u32 flags)
 {
 }
 
 static void mock_bind_ppgtt(struct i915_address_space *vm,
 			    struct i915_vm_pt_stash *stash,
 			    struct i915_vma_resource *vma_res,
-			    enum i915_cache_level cache_level,
+			    unsigned int pat_index,
 			    u32 flags)
 {
 	GEM_BUG_ON(flags & I915_VMA_GLOBAL_BIND);
@@ -94,7 +94,7 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name)
 static void mock_bind_ggtt(struct i915_address_space *vm,
 			   struct i915_vm_pt_stash *stash,
 			   struct i915_vma_resource *vma_res,
-			   enum i915_cache_level cache_level,
+			   unsigned int pat_index,
 			   u32 flags)
 {
 }
-- 
GitLab


From 685282a3b39be5be000d28c8b88a3e0cae195104 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 4 May 2023 19:57:29 +0300
Subject: [PATCH 0477/1791] drm/i915/irq: relocate gmbus and dp aux irq
 handlers

Move gmbus and dp aux irq handlers to their respective files. It should
be up to them what to do with the irq, not the generic irq code.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e825385fc03cb3d53c1f0b66712eea42dad69d59.1683219363.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_aux.c |  5 +++
 drivers/gpu/drm/i915/display/intel_dp_aux.h |  3 ++
 drivers/gpu/drm/i915/display/intel_gmbus.c  |  5 +++
 drivers/gpu/drm/i915/display/intel_gmbus.h  |  2 ++
 drivers/gpu/drm/i915/i915_irq.c             | 40 +++++++++------------
 5 files changed, 31 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c
index 25e36bdc4adb9..0c27db8ae4f18 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c
@@ -814,3 +814,8 @@ enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder)
 
 	return aux_ch;
 }
+
+void intel_dp_aux_irq_handler(struct drm_i915_private *i915)
+{
+	wake_up_all(&i915->display.gmbus.wait_queue);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.h b/drivers/gpu/drm/i915/display/intel_dp_aux.h
index 138e340f94ee0..5b608f9d34993 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.h
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.h
@@ -7,6 +7,7 @@
 #define __INTEL_DP_AUX_H__
 
 enum aux_ch;
+struct drm_i915_private;
 struct intel_dp;
 struct intel_encoder;
 
@@ -15,4 +16,6 @@ void intel_dp_aux_init(struct intel_dp *intel_dp);
 
 enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder);
 
+void intel_dp_aux_irq_handler(struct drm_i915_private *i915);
+
 #endif /* __INTEL_DP_AUX_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index 3ddfc8080ee89..e95ddb580ef6c 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -991,3 +991,8 @@ void intel_gmbus_teardown(struct drm_i915_private *i915)
 		i915->display.gmbus.bus[pin] = NULL;
 	}
 }
+
+void intel_gmbus_irq_handler(struct drm_i915_private *i915)
+{
+	wake_up_all(&i915->display.gmbus.wait_queue);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.h b/drivers/gpu/drm/i915/display/intel_gmbus.h
index 20f704bd4e700..8111eb23e2af2 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.h
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.h
@@ -46,4 +46,6 @@ void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit);
 bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter);
 void intel_gmbus_reset(struct drm_i915_private *dev_priv);
 
+void intel_gmbus_irq_handler(struct drm_i915_private *i915);
+
 #endif /* __INTEL_GMBUS_H__ */
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index e0a4fddccac5b..e54090a680299 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -37,8 +37,10 @@
 #include "display/intel_de.h"
 #include "display/intel_display_trace.h"
 #include "display/intel_display_types.h"
+#include "display/intel_dp_aux.h"
 #include "display/intel_fdi_regs.h"
 #include "display/intel_fifo_underrun.h"
+#include "display/intel_gmbus.h"
 #include "display/intel_hotplug.h"
 #include "display/intel_lpe_audio.h"
 #include "display/intel_psr.h"
@@ -925,16 +927,6 @@ static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915,
 	return hotplug;
 }
 
-static void gmbus_irq_handler(struct drm_i915_private *dev_priv)
-{
-	wake_up_all(&dev_priv->display.gmbus.wait_queue);
-}
-
-static void dp_aux_irq_handler(struct drm_i915_private *dev_priv)
-{
-	wake_up_all(&dev_priv->display.gmbus.wait_queue);
-}
-
 #if defined(CONFIG_DEBUG_FS)
 static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 					 enum pipe pipe,
@@ -1181,7 +1173,7 @@ static void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 		intel_opregion_asle_intr(dev_priv);
 
 	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 }
 
 static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
@@ -1204,7 +1196,7 @@ static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 	}
 
 	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 }
 
 static u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
@@ -1269,7 +1261,7 @@ static void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	if ((IS_G4X(dev_priv) ||
 	     IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
 	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
-		dp_aux_irq_handler(dev_priv);
+		intel_dp_aux_irq_handler(dev_priv);
 }
 
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
@@ -1483,10 +1475,10 @@ static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 	}
 
 	if (pch_iir & SDE_AUX_MASK)
-		dp_aux_irq_handler(dev_priv);
+		intel_dp_aux_irq_handler(dev_priv);
 
 	if (pch_iir & SDE_GMBUS)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 
 	if (pch_iir & SDE_AUDIO_HDCP_MASK)
 		drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n");
@@ -1571,10 +1563,10 @@ static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 	}
 
 	if (pch_iir & SDE_AUX_MASK_CPT)
-		dp_aux_irq_handler(dev_priv);
+		intel_dp_aux_irq_handler(dev_priv);
 
 	if (pch_iir & SDE_GMBUS_CPT)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 
 	if (pch_iir & SDE_AUDIO_CP_REQ_CPT)
 		drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n");
@@ -1624,7 +1616,7 @@ static void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir)
 	}
 
 	if (trigger_aux)
-		dp_aux_irq_handler(i915);
+		intel_dp_aux_irq_handler(i915);
 
 	if (!pin_mask && !trigger_aux)
 		drm_err(&i915->drm,
@@ -1666,7 +1658,7 @@ static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
 
 	if (pch_iir & SDE_GMBUS_ICP)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 }
 
 static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
@@ -1702,7 +1694,7 @@ static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
 
 	if (pch_iir & SDE_GMBUS_CPT)
-		gmbus_irq_handler(dev_priv);
+		intel_gmbus_irq_handler(dev_priv);
 }
 
 static void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv,
@@ -1730,7 +1722,7 @@ static void ilk_display_irq_handler(struct drm_i915_private *dev_priv,
 		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
 
 	if (de_iir & DE_AUX_CHANNEL_A)
-		dp_aux_irq_handler(dev_priv);
+		intel_dp_aux_irq_handler(dev_priv);
 
 	if (de_iir & DE_GSE)
 		intel_opregion_asle_intr(dev_priv);
@@ -1782,7 +1774,7 @@ static void ivb_display_irq_handler(struct drm_i915_private *dev_priv,
 		ivb_err_int_handler(dev_priv);
 
 	if (de_iir & DE_AUX_CHANNEL_A_IVB)
-		dp_aux_irq_handler(dev_priv);
+		intel_dp_aux_irq_handler(dev_priv);
 
 	if (de_iir & DE_GSE_IVB)
 		intel_opregion_asle_intr(dev_priv);
@@ -2174,7 +2166,7 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 			ret = IRQ_HANDLED;
 
 			if (iir & gen8_de_port_aux_mask(dev_priv)) {
-				dp_aux_irq_handler(dev_priv);
+				intel_dp_aux_irq_handler(dev_priv);
 				found = true;
 			}
 
@@ -2196,7 +2188,7 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 
 			if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
 			    (iir & BXT_DE_PORT_GMBUS)) {
-				gmbus_irq_handler(dev_priv);
+				intel_gmbus_irq_handler(dev_priv);
 				found = true;
 			}
 
-- 
GitLab


From d6bf74de0772e233f97a82cc4dad2ec14b14fb28 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:31 -0700
Subject: [PATCH 0478/1791] drm/i915/pxp: Add GSC-CS back-end resource init and
 cleanup

For MTL, the PXP back-end transport uses the GSC engine to submit
HECI packets through the HW to the GSC firmware for PXP arb
session management. This submission uses a non-priveleged
batch buffer, a buffer for the command packet and of course
a context targeting the GSC-CS.

Thus for MTL, we need to allocate and free a set of execution
submission resources for the management of the arbitration session.
Lets start with the context creation first since that object and
its usage is very straight-forward. We'll add the buffer allocation
and freeing later when we introduce the gsccs' send-message function.

Do this one time allocation of gsccs specific resources in
a new gsccs source file with intel_pxp_gsccs_init / fini functions
and hook them up from the PXP front-end.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-2-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/Makefile              |  1 +
 drivers/gpu/drm/i915/pxp/intel_pxp.c       | 20 +++++--
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c | 63 ++++++++++++++++++++++
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h | 29 ++++++++++
 drivers/gpu/drm/i915/pxp/intel_pxp_tee.c   |  2 -
 drivers/gpu/drm/i915/pxp/intel_pxp_types.h |  8 +++
 6 files changed, 117 insertions(+), 6 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
 create mode 100644 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 4494dcd2eb5d1..c58d7b1936644 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -339,6 +339,7 @@ i915-y += \
 i915-$(CONFIG_DRM_I915_PXP) += \
 	pxp/intel_pxp_cmd.o \
 	pxp/intel_pxp_debugfs.o \
+	pxp/intel_pxp_gsccs.o \
 	pxp/intel_pxp_irq.o \
 	pxp/intel_pxp_pm.o \
 	pxp/intel_pxp_session.o
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index 9d4c7724e98ed..f93aa171aa1e0 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -12,6 +12,7 @@
 #include "i915_drv.h"
 
 #include "intel_pxp.h"
+#include "intel_pxp_gsccs.h"
 #include "intel_pxp_irq.h"
 #include "intel_pxp_session.h"
 #include "intel_pxp_tee.h"
@@ -132,7 +133,10 @@ static void pxp_init_full(struct intel_pxp *pxp)
 	if (ret)
 		return;
 
-	ret = intel_pxp_tee_component_init(pxp);
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
+		ret = intel_pxp_gsccs_init(pxp);
+	else
+		ret = intel_pxp_tee_component_init(pxp);
 	if (ret)
 		goto out_context;
 
@@ -165,9 +169,12 @@ static struct intel_gt *find_gt_for_required_protected_content(struct drm_i915_p
 	/*
 	 * For MTL onwards, PXP-controller-GT needs to have a valid GSC engine
 	 * on the media GT. NOTE: if we have a media-tile with a GSC-engine,
-	 * the VDBOX is already present so skip that check
+	 * the VDBOX is already present so skip that check. We also have to
+	 * ensure the GSC and HUC firmware are coming online
 	 */
-	if (i915->media_gt && HAS_ENGINE(i915->media_gt, GSC0))
+	if (i915->media_gt && HAS_ENGINE(i915->media_gt, GSC0) &&
+	    intel_uc_fw_is_loadable(&i915->media_gt->uc.gsc.fw) &&
+	    intel_uc_fw_is_loadable(&i915->media_gt->uc.huc.fw))
 		return i915->media_gt;
 
 	/*
@@ -207,7 +214,9 @@ int intel_pxp_init(struct drm_i915_private *i915)
 	if (!i915->pxp)
 		return -ENOMEM;
 
+	/* init common info used by all feature-mode usages*/
 	i915->pxp->ctrl_gt = gt;
+	mutex_init(&i915->pxp->tee_mutex);
 
 	/*
 	 * If full PXP feature is not available but HuC is loaded by GSC on pre-MTL
@@ -229,7 +238,10 @@ void intel_pxp_fini(struct drm_i915_private *i915)
 
 	i915->pxp->arb_is_valid = false;
 
-	intel_pxp_tee_component_fini(i915->pxp);
+	if (HAS_ENGINE(i915->pxp->ctrl_gt, GSC0))
+		intel_pxp_gsccs_fini(i915->pxp);
+	else
+		intel_pxp_tee_component_fini(i915->pxp);
 
 	destroy_vcs_context(i915->pxp);
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
new file mode 100644
index 0000000000000..bad55719a7acc
--- /dev/null
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright(c) 2023 Intel Corporation.
+ */
+
+#include "gem/i915_gem_internal.h"
+
+#include "gt/intel_context.h"
+
+#include "i915_drv.h"
+#include "intel_pxp_cmd_interface_43.h"
+#include "intel_pxp_gsccs.h"
+#include "intel_pxp_types.h"
+
+static void
+gsccs_destroy_execution_resource(struct intel_pxp *pxp)
+{
+	struct gsccs_session_resources *exec_res = &pxp->gsccs_res;
+
+	if (exec_res->ce)
+		intel_context_put(exec_res->ce);
+
+	memset(exec_res, 0, sizeof(*exec_res));
+}
+
+static int
+gsccs_allocate_execution_resource(struct intel_pxp *pxp)
+{
+	struct intel_gt *gt = pxp->ctrl_gt;
+	struct gsccs_session_resources *exec_res = &pxp->gsccs_res;
+	struct intel_engine_cs *engine = gt->engine[GSC0];
+	struct intel_context *ce;
+
+	/*
+	 * First, ensure the GSC engine is present.
+	 * NOTE: Backend would only be called with the correct gt.
+	 */
+	if (!engine)
+		return -ENODEV;
+
+	/* Finally, create an intel_context to be used during the submission */
+	ce = intel_context_create(engine);
+	if (IS_ERR(ce)) {
+		drm_err(&gt->i915->drm, "Failed creating gsccs backend ctx\n");
+		return PTR_ERR(ce);
+	}
+
+	i915_vm_put(ce->vm);
+	ce->vm = i915_vm_get(pxp->ctrl_gt->vm);
+	exec_res->ce = ce;
+
+	return 0;
+}
+
+void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
+{
+	gsccs_destroy_execution_resource(pxp);
+}
+
+int intel_pxp_gsccs_init(struct intel_pxp *pxp)
+{
+	return gsccs_allocate_execution_resource(pxp);
+}
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
new file mode 100644
index 0000000000000..354ea9a8f9403
--- /dev/null
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright(c) 2022, Intel Corporation. All rights reserved.
+ */
+
+#ifndef __INTEL_PXP_GSCCS_H__
+#define __INTEL_PXP_GSCCS_H__
+
+#include <linux/types.h>
+
+struct intel_pxp;
+
+#ifdef CONFIG_DRM_I915_PXP
+void intel_pxp_gsccs_fini(struct intel_pxp *pxp);
+int intel_pxp_gsccs_init(struct intel_pxp *pxp);
+
+#else
+static inline void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
+{
+}
+
+static inline int intel_pxp_gsccs_init(struct intel_pxp *pxp)
+{
+	return 0;
+}
+
+#endif
+
+#endif /*__INTEL_PXP_GSCCS_H__ */
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c
index a2846b1dbbee4..1ce07d7e87690 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c
@@ -284,8 +284,6 @@ int intel_pxp_tee_component_init(struct intel_pxp *pxp)
 	struct intel_gt *gt = pxp->ctrl_gt;
 	struct drm_i915_private *i915 = gt->i915;
 
-	mutex_init(&pxp->tee_mutex);
-
 	ret = alloc_streaming_command(pxp);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
index c445f7f2f47ac..1759106b9a8e1 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
@@ -26,6 +26,14 @@ struct intel_pxp {
 	 */
 	struct intel_gt *ctrl_gt;
 
+	/**
+	 * @gsccs_res: resources for request submission for platforms that have a GSC engine.
+	 */
+	struct gsccs_session_resources {
+		u64 host_session_handle; /* used by firmware to link commands to sessions */
+		struct intel_context *ce; /* context for gsc command submission */
+	} gsccs_res;
+
 	/**
 	 * @pxp_component: i915_pxp_component struct of the bound mei_pxp
 	 * module. Only set and cleared inside component bind/unbind functions,
-- 
GitLab


From 5adacf19f6c3fa7d6119878246e9a253867e14c9 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:32 -0700
Subject: [PATCH 0479/1791] drm/i915/pxp: Add MTL hw-plumbing enabling for KCR
 operation

Add MTL hw-plumbing enabling for KCR operation under PXP
which includes:

1. Updating 'pick-gt' to get the media tile for
   KCR interrupt handling
2. Adding MTL's KCR registers for PXP operation
   (init, status-checking, etc.).

While doing #2, lets create a separate registers header file for PXP
to be consistent with other i915 global subsystems.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-3-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_irq.c       |  3 +-
 drivers/gpu/drm/i915/pxp/intel_pxp.c         | 32 ++++++++++++--------
 drivers/gpu/drm/i915/pxp/intel_pxp_regs.h    | 27 +++++++++++++++++
 drivers/gpu/drm/i915/pxp/intel_pxp_session.c | 12 +++-----
 drivers/gpu/drm/i915/pxp/intel_pxp_types.h   |  6 ++++
 5 files changed, 58 insertions(+), 22 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/pxp/intel_pxp_regs.h

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index e13cfbe2daf55..acec6566b9145 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -106,7 +106,8 @@ static struct intel_gt *pick_gt(struct intel_gt *gt, u8 class, u8 instance)
 	case OTHER_CLASS:
 		if (instance == OTHER_GSC_HECI_2_INSTANCE)
 			return media_gt;
-		if (instance == OTHER_GSC_INSTANCE && HAS_ENGINE(media_gt, GSC0))
+		if ((instance == OTHER_GSC_INSTANCE || instance == OTHER_KCR_INSTANCE) &&
+		    HAS_ENGINE(media_gt, GSC0))
 			return media_gt;
 		fallthrough;
 	default:
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index f93aa171aa1e0..8949d4be78828 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -14,6 +14,7 @@
 #include "intel_pxp.h"
 #include "intel_pxp_gsccs.h"
 #include "intel_pxp_irq.h"
+#include "intel_pxp_regs.h"
 #include "intel_pxp_session.h"
 #include "intel_pxp_tee.h"
 #include "intel_pxp_types.h"
@@ -61,21 +62,22 @@ bool intel_pxp_is_active(const struct intel_pxp *pxp)
 	return IS_ENABLED(CONFIG_DRM_I915_PXP) && pxp && pxp->arb_is_valid;
 }
 
-/* KCR register definitions */
-#define KCR_INIT _MMIO(0x320f0)
-/* Setting KCR Init bit is required after system boot */
-#define KCR_INIT_ALLOW_DISPLAY_ME_WRITES REG_BIT(14)
+static void kcr_pxp_set_status(const struct intel_pxp *pxp, bool enable)
+{
+	u32 val = enable ? _MASKED_BIT_ENABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES) :
+		  _MASKED_BIT_DISABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES);
+
+	intel_uncore_write(pxp->ctrl_gt->uncore, KCR_INIT(pxp->kcr_base), val);
+}
 
-static void kcr_pxp_enable(struct intel_gt *gt)
+static void kcr_pxp_enable(const struct intel_pxp *pxp)
 {
-	intel_uncore_write(gt->uncore, KCR_INIT,
-			   _MASKED_BIT_ENABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES));
+	kcr_pxp_set_status(pxp, true);
 }
 
-static void kcr_pxp_disable(struct intel_gt *gt)
+static void kcr_pxp_disable(const struct intel_pxp *pxp)
 {
-	intel_uncore_write(gt->uncore, KCR_INIT,
-			   _MASKED_BIT_DISABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES));
+	kcr_pxp_set_status(pxp, false);
 }
 
 static int create_vcs_context(struct intel_pxp *pxp)
@@ -127,6 +129,11 @@ static void pxp_init_full(struct intel_pxp *pxp)
 	init_completion(&pxp->termination);
 	complete_all(&pxp->termination);
 
+	if (pxp->ctrl_gt->type == GT_MEDIA)
+		pxp->kcr_base = MTL_KCR_BASE;
+	else
+		pxp->kcr_base = GEN12_KCR_BASE;
+
 	intel_pxp_session_management_init(pxp);
 
 	ret = create_vcs_context(pxp);
@@ -369,14 +376,13 @@ int intel_pxp_start(struct intel_pxp *pxp)
 
 void intel_pxp_init_hw(struct intel_pxp *pxp)
 {
-	kcr_pxp_enable(pxp->ctrl_gt);
+	kcr_pxp_enable(pxp);
 	intel_pxp_irq_enable(pxp);
 }
 
 void intel_pxp_fini_hw(struct intel_pxp *pxp)
 {
-	kcr_pxp_disable(pxp->ctrl_gt);
-
+	kcr_pxp_disable(pxp);
 	intel_pxp_irq_disable(pxp);
 }
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_regs.h b/drivers/gpu/drm/i915/pxp/intel_pxp_regs.h
new file mode 100644
index 0000000000000..a9e7e6efa4c79
--- /dev/null
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_regs.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright(c) 2023, Intel Corporation. All rights reserved.
+ */
+
+#ifndef __INTEL_PXP_REGS_H__
+#define __INTEL_PXP_REGS_H__
+
+#include "i915_reg_defs.h"
+
+/* KCR subsystem register base address */
+#define GEN12_KCR_BASE 0x32000
+#define MTL_KCR_BASE 0x386000
+
+/* KCR enable/disable control */
+#define KCR_INIT(base) _MMIO((base) + 0xf0)
+
+/* Setting KCR Init bit is required after system boot */
+#define KCR_INIT_ALLOW_DISPLAY_ME_WRITES REG_BIT(14)
+
+/* KCR hwdrm session in play status 0-31 */
+#define KCR_SIP(base) _MMIO((base) + 0x260)
+
+/* PXP global terminate register for session termination */
+#define KCR_GLOBAL_TERMINATE(base) _MMIO((base) + 0xf8)
+
+#endif /* __INTEL_PXP_REGS_H__ */
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
index 7de849cb6c47b..7899079e17b06 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
@@ -10,14 +10,10 @@
 #include "intel_pxp_session.h"
 #include "intel_pxp_tee.h"
 #include "intel_pxp_types.h"
+#include "intel_pxp_regs.h"
 
 #define ARB_SESSION I915_PROTECTED_CONTENT_DEFAULT_SESSION /* shorter define */
 
-#define GEN12_KCR_SIP _MMIO(0x32260) /* KCR hwdrm session in play 0-31 */
-
-/* PXP global terminate register for session termination */
-#define PXP_GLOBAL_TERMINATE _MMIO(0x320f8)
-
 static bool intel_pxp_session_is_in_play(struct intel_pxp *pxp, u32 id)
 {
 	struct intel_uncore *uncore = pxp->ctrl_gt->uncore;
@@ -26,7 +22,7 @@ static bool intel_pxp_session_is_in_play(struct intel_pxp *pxp, u32 id)
 
 	/* if we're suspended the session is considered off */
 	with_intel_runtime_pm_if_in_use(uncore->rpm, wakeref)
-		sip = intel_uncore_read(uncore, GEN12_KCR_SIP);
+		sip = intel_uncore_read(uncore, KCR_SIP(pxp->kcr_base));
 
 	return sip & BIT(id);
 }
@@ -44,7 +40,7 @@ static int pxp_wait_for_session_state(struct intel_pxp *pxp, u32 id, bool in_pla
 		return in_play ? -ENODEV : 0;
 
 	ret = intel_wait_for_register(uncore,
-				      GEN12_KCR_SIP,
+				      KCR_SIP(pxp->kcr_base),
 				      mask,
 				      in_play ? mask : 0,
 				      100);
@@ -108,7 +104,7 @@ static int pxp_terminate_arb_session_and_global(struct intel_pxp *pxp)
 		return ret;
 	}
 
-	intel_uncore_write(gt->uncore, PXP_GLOBAL_TERMINATE, 1);
+	intel_uncore_write(gt->uncore, KCR_GLOBAL_TERMINATE(pxp->kcr_base), 1);
 
 	intel_pxp_tee_end_arb_fw_session(pxp, ARB_SESSION);
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
index 1759106b9a8e1..87d0d8da98cd5 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
@@ -26,6 +26,12 @@ struct intel_pxp {
 	 */
 	struct intel_gt *ctrl_gt;
 
+	/**
+	 * @kcr_base: base mmio offset for the KCR engine which is different on legacy platforms
+	 * vs newer platforms where the KCR is inside the media-tile.
+	 */
+	u32 kcr_base;
+
 	/**
 	 * @gsccs_res: resources for request submission for platforms that have a GSC engine.
 	 */
-- 
GitLab


From e5e1e6d28ebcc0fe52567b1301c23f05d4c79df7 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:33 -0700
Subject: [PATCH 0480/1791] drm/i915/pxp: Add MTL helpers to submit
 Heci-Cmd-Packet to GSC

Add helper functions into a new file for heci-packet-submission.
The helpers will handle generating the MTL GSC-CS Memory-Header
and submission of the Heci-Cmd-Packet instructions to the engine.

NOTE1: These common functions for heci-packet-submission will be used
by different i915 callers:
     1- GSC-SW-Proxy: This is pending upstream publication awaiting
        a few remaining opens
     2- MTL-HDCP: An equivalent patch has also been published at:
        https://patchwork.freedesktop.org/series/111876/. (Patch 1)
     3- PXP: This series.

NOTE2: A difference in this patch vs what is appearing is in bullet 2
above is that HDCP (and SW-Proxy) will be using priveleged submission
(GGTT and common gsc-uc-context) while PXP will be using non-priveleged
PPGTT, context and batch buffer. Therefore this patch will only slightly
overlap with the MTL-HDCP patches despite have very similar function
names (emit_foo vs emit_nonpriv_foo). This is because HECI_CMD_PKT
instructions require different flows and hw-specific code when done
via PPGTT based submission (not different from other engines). MTL-HDCP
contains the same intel_gsc_mtl_header_t structures as this but the
helpers there are different. Both add the same new file names.

NOTE3: Additional clarity about the heci-cmd-pkt layout and where the
       common helpers come in:
     - On MTL, when an i915 subsystem needs to send a command request
       to the security firmware, it will send that via the GSC-
       engine-command-streamer.
     - However those commands, (lets call them "gsc_specific_fw_api"
       calls), are not understood by the GSC command streamer hw.
     - The GSC CS only looks at the GSC_HECI_CMD_PKT instruction and
       passes it along to the GSC firmware.
     - The GSC FW on the other hand needs additional metadata to know
       which usage service is being called (PXP, HDCP, proxy, etc) along
       with session specific info. Thus an extra header called GSC-CS
       HECI Memory Header, (C) in below diagram is prepended before
       the FW specific API, (D).
     - Thus, the structural layout of the request submitted would
       need to look like the diagram below (for non-priv PXP).
     - In the diagram, the common helper for HDCP, (GSC-Sw-Proxy) and
       PXP (i.e. new function intel_gsc_uc_heci_cmd_emit_mtl_header)
       will populate blob (C) while additional helpers, different for
       PPGGTT (this patch) vs GGTT (HDCP series) will populate
       blobs (A) and (B) below.
      ___________________________________________________________
 (A)  |  MI_BATCH_BUFFER_START (ppgtt, batchbuff-addr, ...)     |
      |     |                                                   |
      |    _|________________________________________________   |
      | (B)| GSC_HECI_CMD_PKT (pkt-addr-in, pkt-size-in,    |   |
      |    |                   pkt-addr-out, pkt-size-out)  |--------
      |    | MI_BATCH_BUFFER_END                            |   |   |
      |    |________________________________________________|   |   |
      |                                                         |   |
      |_________________________________________________________|   |
                                                                    |
            ---------------------------------------------------------
            |
           \|/
      ______V___________________________________________
      |   _________________________________________    |
      |(C)|                                       |    |
      |   | struct intel_gsc_mtl_header {         |    |
      |   |   validity marker                     |    |
      |   |   heci_clent_id                       |    |
      |   |   ...                                 |    |
      |   |  }                                    |    |
      |   |_______________________________________|    |
      |(D)|                                       |    |
      |   | struct gsc_fw_specific_api_foobar {   |    |
      |   |     ...                               |    |
      |   |     For an example, see               |    |
      |   |     'struct pxp43_create_arb_in' at   |    |
      |   |     intel_pxp_cmd_interface_43.h      |    |
      |   |                                       |    |
      |   | }                                     |    |
      |   |  Struture depends on command type     |    |
      |   | struct gsc_fw_specific_api_foobar {   |    |
      |   |_______________________________________|    |
      |________________________________________________|

That said, this patch provides basic helpers but leaves the
PXP subsystem (i.e. the caller) to handle (D) and everything
else such as input/output size verification or handling the
responses from security firmware (for example, requiring a retry).

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-4-alan.previn.teres.alexis@intel.com
---
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c | 102 ++++++++++++++++++
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h |  23 ++++
 2 files changed, 125 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
index ea0da06e2f399..579c0f5a14384 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
@@ -3,6 +3,7 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include "gt/intel_context.h"
 #include "gt/intel_engine_pm.h"
 #include "gt/intel_gpu_commands.h"
 #include "gt/intel_gt.h"
@@ -107,3 +108,104 @@ void intel_gsc_uc_heci_cmd_emit_mtl_header(struct intel_gsc_mtl_header *header,
 	header->header_version = MTL_GSC_HEADER_VERSION;
 	header->message_size = message_size;
 }
+
+static void
+emit_gsc_heci_pkt_nonpriv(u32 *cmd, struct intel_gsc_heci_non_priv_pkt *pkt)
+{
+	*cmd++ = GSC_HECI_CMD_PKT;
+	*cmd++ = lower_32_bits(pkt->addr_in);
+	*cmd++ = upper_32_bits(pkt->addr_in);
+	*cmd++ = pkt->size_in;
+	*cmd++ = lower_32_bits(pkt->addr_out);
+	*cmd++ = upper_32_bits(pkt->addr_out);
+	*cmd++ = pkt->size_out;
+	*cmd++ = 0;
+	*cmd++ = MI_BATCH_BUFFER_END;
+}
+
+int
+intel_gsc_uc_heci_cmd_submit_nonpriv(struct intel_gsc_uc *gsc,
+				     struct intel_context *ce,
+				     struct intel_gsc_heci_non_priv_pkt *pkt,
+				     u32 *cmd, int timeout_ms)
+{
+	struct intel_engine_cs *engine;
+	struct i915_gem_ww_ctx ww;
+	struct i915_request *rq;
+	int err, trials = 0;
+
+	i915_gem_ww_ctx_init(&ww, false);
+retry:
+	err = i915_gem_object_lock(pkt->bb_vma->obj, &ww);
+	if (err)
+		goto out_ww;
+	err = i915_gem_object_lock(pkt->heci_pkt_vma->obj, &ww);
+	if (err)
+		goto out_ww;
+	err = intel_context_pin_ww(ce, &ww);
+	if (err)
+		goto out_ww;
+
+	rq = i915_request_create(ce);
+	if (IS_ERR(rq)) {
+		err = PTR_ERR(rq);
+		goto out_unpin_ce;
+	}
+
+	emit_gsc_heci_pkt_nonpriv(cmd, pkt);
+
+	err = i915_vma_move_to_active(pkt->bb_vma, rq, 0);
+	if (err)
+		goto out_rq;
+	err = i915_vma_move_to_active(pkt->heci_pkt_vma, rq, EXEC_OBJECT_WRITE);
+	if (err)
+		goto out_rq;
+
+	engine = rq->context->engine;
+	if (engine->emit_init_breadcrumb) {
+		err = engine->emit_init_breadcrumb(rq);
+		if (err)
+			goto out_rq;
+	}
+
+	err = engine->emit_bb_start(rq, i915_vma_offset(pkt->bb_vma), PAGE_SIZE, 0);
+	if (err)
+		goto out_rq;
+
+	err = ce->engine->emit_flush(rq, 0);
+	if (err)
+		drm_err(&gsc_uc_to_gt(gsc)->i915->drm,
+			"Failed emit-flush for gsc-heci-non-priv-pkterr=%d\n", err);
+
+out_rq:
+	i915_request_get(rq);
+
+	if (unlikely(err))
+		i915_request_set_error_once(rq, err);
+
+	i915_request_add(rq);
+
+	if (!err) {
+		if (i915_request_wait(rq, I915_WAIT_INTERRUPTIBLE,
+				      msecs_to_jiffies(timeout_ms)) < 0)
+			err = -ETIME;
+	}
+
+	i915_request_put(rq);
+
+out_unpin_ce:
+	intel_context_unpin(ce);
+out_ww:
+	if (err == -EDEADLK) {
+		err = i915_gem_ww_ctx_backoff(&ww);
+		if (!err) {
+			if (++trials < 10)
+				goto retry;
+			else
+				err = EAGAIN;
+		}
+	}
+	i915_gem_ww_ctx_fini(&ww);
+
+	return err;
+}
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
index 8f199d5f963e5..e790608745d40 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
@@ -8,7 +8,10 @@
 
 #include <linux/types.h>
 
+struct i915_vma;
+struct intel_context;
 struct intel_gsc_uc;
+
 struct intel_gsc_mtl_header {
 	u32 validity_marker;
 #define GSC_HECI_VALIDITY_MARKER 0xA578875A
@@ -59,4 +62,24 @@ int intel_gsc_uc_heci_cmd_submit_packet(struct intel_gsc_uc *gsc,
 void intel_gsc_uc_heci_cmd_emit_mtl_header(struct intel_gsc_mtl_header *header,
 					   u8 heci_client_id, u32 message_size,
 					   u64 host_session_id);
+
+struct intel_gsc_heci_non_priv_pkt {
+	u64 addr_in;
+	u32 size_in;
+	u64 addr_out;
+	u32 size_out;
+	struct i915_vma *heci_pkt_vma;
+	struct i915_vma *bb_vma;
+};
+
+void
+intel_gsc_uc_heci_cmd_emit_mtl_header(struct intel_gsc_mtl_header *header,
+				      u8 heci_client_id, u32 msg_size,
+				      u64 host_session_id);
+
+int
+intel_gsc_uc_heci_cmd_submit_nonpriv(struct intel_gsc_uc *gsc,
+				     struct intel_context *ce,
+				     struct intel_gsc_heci_non_priv_pkt *pkt,
+				     u32 *cs, int timeout_ms);
 #endif
-- 
GitLab


From dc9ac125d81faf4761574a9f613ebc8eb35717e1 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:34 -0700
Subject: [PATCH 0481/1791] drm/i915/pxp: Add GSC-CS backend to send GSC fw
 messages

Add GSC engine based method for sending PXP firmware packets
to the GSC firmware for MTL (and future) products.

Use the newly added helpers to populate the GSC-CS memory
header and send the message packet to the FW by dispatching
the GSC_HECI_CMD_PKT instruction on the GSC engine.

We use non-priveleged batches for submission to GSC engine
which require two buffers for the request:
     - a buffer for the HECI packet that contains PXP FW commands
     - a batch-buffer that contains the engine instruction for
       sending the HECI packet to the GSC firmware.

Thus, add the allocation and freeing of these buffers in gsccs
init and fini.

The GSC-fw may reply to commands with a SUCCESS but with an
additional pending-bit set in the reply packet. This bit
means the GSC-FW is currently busy and the caller needs to
try again with the gsc_message_handle the fw returned. Thus,
add a wrapper to continuously retry send_message while
replaying the gsc_message_handle. Retries need to follow the
arch-spec count and delay until GSC-FW replies with the real
SUCCESS or timeout after that spec'd delay.

The GSC-fw requires a non-zero host_session_handle provided
by the caller to enable gsc_message_handle tracking. Thus,
allocate the host_session_handle at init and destroy it
at fini (the latter requiring an FYI to the gsc-firmware).

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-5-alan.previn.teres.alexis@intel.com
---
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h |   3 +-
 .../drm/i915/pxp/intel_pxp_cmd_interface_43.h |   3 +
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c    | 240 +++++++++++++++++-
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h    |   4 +
 drivers/gpu/drm/i915/pxp/intel_pxp_types.h    |   6 +
 5 files changed, 254 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
index e790608745d40..ef70e304904a3 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h
@@ -51,7 +51,8 @@ struct intel_gsc_mtl_header {
 	 * we distinguish the flags using OUTFLAG or INFLAG
 	 */
 	u32 flags;
-#define GSC_OUTFLAG_MSG_PENDING	1
+#define GSC_OUTFLAG_MSG_PENDING	BIT(0)
+#define GSC_INFLAG_MSG_CLEANUP	BIT(1)
 
 	u32 status;
 } __packed;
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
index ad67e3f49c207..c65ada99e54fd 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
@@ -12,6 +12,9 @@
 /* PXP-Cmd-Op definitions */
 #define PXP43_CMDID_START_HUC_AUTH 0x0000003A
 
+/* PXP-Packet sizes for MTL's GSCCS-HECI instruction */
+#define PXP43_MAX_HECI_INOUT_SIZE (SZ_32K)
+
 /* PXP-Input-Packet: HUC-Authentication */
 struct pxp43_start_huc_auth_in {
 	struct pxp_cmd_header header;
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
index bad55719a7acc..16e3b73d06538 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -6,23 +6,232 @@
 #include "gem/i915_gem_internal.h"
 
 #include "gt/intel_context.h"
+#include "gt/uc/intel_gsc_uc_heci_cmd_submit.h"
 
 #include "i915_drv.h"
 #include "intel_pxp_cmd_interface_43.h"
 #include "intel_pxp_gsccs.h"
 #include "intel_pxp_types.h"
 
+static int
+gsccs_send_message(struct intel_pxp *pxp,
+		   void *msg_in, size_t msg_in_size,
+		   void *msg_out, size_t msg_out_size_max,
+		   size_t *msg_out_len,
+		   u64 *gsc_msg_handle_retry)
+{
+	struct intel_gt *gt = pxp->ctrl_gt;
+	struct drm_i915_private *i915 = gt->i915;
+	struct gsccs_session_resources *exec_res =  &pxp->gsccs_res;
+	struct intel_gsc_mtl_header *header = exec_res->pkt_vaddr;
+	struct intel_gsc_heci_non_priv_pkt pkt;
+	size_t max_msg_size;
+	u32 reply_size;
+	int ret;
+
+	if (!exec_res->ce)
+		return -ENODEV;
+
+	max_msg_size = PXP43_MAX_HECI_INOUT_SIZE - sizeof(*header);
+
+	if (msg_in_size > max_msg_size || msg_out_size_max > max_msg_size)
+		return -ENOSPC;
+
+	if (!exec_res->pkt_vma || !exec_res->bb_vma)
+		return -ENOENT;
+
+	GEM_BUG_ON(exec_res->pkt_vma->size < (2 * PXP43_MAX_HECI_INOUT_SIZE));
+
+	mutex_lock(&pxp->tee_mutex);
+
+	memset(header, 0, sizeof(*header));
+	intel_gsc_uc_heci_cmd_emit_mtl_header(header, HECI_MEADDRESS_PXP,
+					      msg_in_size + sizeof(*header),
+					      exec_res->host_session_handle);
+
+	/* check if this is a host-session-handle cleanup call (empty packet) */
+	if (!msg_in && !msg_out)
+		header->flags |= GSC_INFLAG_MSG_CLEANUP;
+
+	/* copy caller provided gsc message handle if this is polling for a prior msg completion */
+	header->gsc_message_handle = *gsc_msg_handle_retry;
+
+	/* NOTE: zero size packets are used for session-cleanups */
+	if (msg_in && msg_in_size)
+		memcpy(exec_res->pkt_vaddr + sizeof(*header), msg_in, msg_in_size);
+
+	pkt.addr_in = i915_vma_offset(exec_res->pkt_vma);
+	pkt.size_in = header->message_size;
+	pkt.addr_out = pkt.addr_in + PXP43_MAX_HECI_INOUT_SIZE;
+	pkt.size_out = msg_out_size_max + sizeof(*header);
+	pkt.heci_pkt_vma = exec_res->pkt_vma;
+	pkt.bb_vma = exec_res->bb_vma;
+
+	/*
+	 * Before submitting, let's clear-out the validity marker on the reply offset.
+	 * We use offset PXP43_MAX_HECI_INOUT_SIZE for reply location so point header there.
+	 */
+	header = exec_res->pkt_vaddr + PXP43_MAX_HECI_INOUT_SIZE;
+	header->validity_marker = 0;
+
+	ret = intel_gsc_uc_heci_cmd_submit_nonpriv(&gt->uc.gsc,
+						   exec_res->ce, &pkt, exec_res->bb_vaddr,
+						   GSC_REPLY_LATENCY_MS);
+	if (ret) {
+		drm_err(&i915->drm, "failed to send gsc PXP msg (%d)\n", ret);
+		goto unlock;
+	}
+
+	/* Response validity marker, status and busyness */
+	if (header->validity_marker != GSC_HECI_VALIDITY_MARKER) {
+		drm_err(&i915->drm, "gsc PXP reply with invalid validity marker\n");
+		ret = -EINVAL;
+		goto unlock;
+	}
+	if (header->status != 0) {
+		drm_dbg(&i915->drm, "gsc PXP reply status has error = 0x%08x\n",
+			header->status);
+		ret = -EINVAL;
+		goto unlock;
+	}
+	if (header->flags & GSC_OUTFLAG_MSG_PENDING) {
+		drm_dbg(&i915->drm, "gsc PXP reply is busy\n");
+		/*
+		 * When the GSC firmware replies with pending bit, it means that the requested
+		 * operation has begun but the completion is pending and the caller needs
+		 * to re-request with the gsc_message_handle that was returned by the firmware.
+		 * until the pending bit is turned off.
+		 */
+		*gsc_msg_handle_retry = header->gsc_message_handle;
+		ret = -EAGAIN;
+		goto unlock;
+	}
+
+	reply_size = header->message_size - sizeof(*header);
+	if (reply_size > msg_out_size_max) {
+		drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%ld)\n",
+			 reply_size, msg_out_size_max);
+		reply_size = msg_out_size_max;
+	}
+
+	if (msg_out)
+		memcpy(msg_out, exec_res->pkt_vaddr + PXP43_MAX_HECI_INOUT_SIZE + sizeof(*header),
+		       reply_size);
+	if (msg_out_len)
+		*msg_out_len = reply_size;
+
+unlock:
+	mutex_unlock(&pxp->tee_mutex);
+	return ret;
+}
+
+static int
+gsccs_send_message_retry_complete(struct intel_pxp *pxp,
+				  void *msg_in, size_t msg_in_size,
+				  void *msg_out, size_t msg_out_size_max,
+				  size_t *msg_out_len)
+{
+	u64 gsc_session_retry = 0;
+	int ret, tries = 0;
+
+	/*
+	 * Keep sending request if GSC firmware was busy. Based on fw specs +
+	 * sw overhead (and testing) we expect a worst case pending-bit delay of
+	 * GSC_PENDING_RETRY_MAXCOUNT x GSC_PENDING_RETRY_PAUSE_MS millisecs.
+	 */
+	do {
+		ret = gsccs_send_message(pxp, msg_in, msg_in_size, msg_out, msg_out_size_max,
+					 msg_out_len, &gsc_session_retry);
+		/* Only try again if gsc says so */
+		if (ret != -EAGAIN)
+			break;
+
+		msleep(GSC_PENDING_RETRY_PAUSE_MS);
+	} while (++tries < GSC_PENDING_RETRY_MAXCOUNT);
+
+	return ret;
+}
+
+static void
+gsccs_cleanup_fw_host_session_handle(struct intel_pxp *pxp)
+{
+	struct drm_i915_private *i915 = pxp->ctrl_gt->i915;
+	int ret;
+
+	ret = gsccs_send_message_retry_complete(pxp, NULL, 0, NULL, 0, NULL);
+	if (ret)
+		drm_dbg(&i915->drm, "Failed to send gsccs msg host-session-cleanup: ret=[%d]\n",
+			ret);
+}
+
 static void
 gsccs_destroy_execution_resource(struct intel_pxp *pxp)
 {
 	struct gsccs_session_resources *exec_res = &pxp->gsccs_res;
 
+	if (exec_res->host_session_handle)
+		gsccs_cleanup_fw_host_session_handle(pxp);
 	if (exec_res->ce)
 		intel_context_put(exec_res->ce);
+	if (exec_res->bb_vma)
+		i915_vma_unpin_and_release(&exec_res->bb_vma, I915_VMA_RELEASE_MAP);
+	if (exec_res->pkt_vma)
+		i915_vma_unpin_and_release(&exec_res->pkt_vma, I915_VMA_RELEASE_MAP);
 
 	memset(exec_res, 0, sizeof(*exec_res));
 }
 
+static int
+gsccs_create_buffer(struct intel_gt *gt,
+		    const char *bufname, size_t size,
+		    struct i915_vma **vma, void **map)
+{
+	struct drm_i915_private *i915 = gt->i915;
+	struct drm_i915_gem_object *obj;
+	int err = 0;
+
+	obj = i915_gem_object_create_internal(i915, size);
+	if (IS_ERR(obj)) {
+		drm_err(&i915->drm, "Failed to allocate gsccs backend %s.\n", bufname);
+		err = PTR_ERR(obj);
+		goto out_none;
+	}
+
+	*vma = i915_vma_instance(obj, gt->vm, NULL);
+	if (IS_ERR(*vma)) {
+		drm_err(&i915->drm, "Failed to vma-instance gsccs backend %s.\n", bufname);
+		err = PTR_ERR(*vma);
+		goto out_put;
+	}
+
+	/* return a virtual pointer */
+	*map = i915_gem_object_pin_map_unlocked(obj, i915_coherent_map_type(i915, obj, true));
+	if (IS_ERR(*map)) {
+		drm_err(&i915->drm, "Failed to map gsccs backend %s.\n", bufname);
+		err = PTR_ERR(*map);
+		goto out_put;
+	}
+
+	/* all PXP sessions commands are treated as non-privileged */
+	err = i915_vma_pin(*vma, 0, 0, PIN_USER);
+	if (err) {
+		drm_err(&i915->drm, "Failed to vma-pin gsccs backend %s.\n", bufname);
+		goto out_unmap;
+	}
+
+	return 0;
+
+out_unmap:
+	i915_gem_object_unpin_map(obj);
+out_put:
+	i915_gem_object_put(obj);
+out_none:
+	*vma = NULL;
+	*map = NULL;
+
+	return err;
+}
+
 static int
 gsccs_allocate_execution_resource(struct intel_pxp *pxp)
 {
@@ -30,6 +239,7 @@ gsccs_allocate_execution_resource(struct intel_pxp *pxp)
 	struct gsccs_session_resources *exec_res = &pxp->gsccs_res;
 	struct intel_engine_cs *engine = gt->engine[GSC0];
 	struct intel_context *ce;
+	int err = 0;
 
 	/*
 	 * First, ensure the GSC engine is present.
@@ -38,18 +248,46 @@ gsccs_allocate_execution_resource(struct intel_pxp *pxp)
 	if (!engine)
 		return -ENODEV;
 
+	/*
+	 * Now, allocate, pin and map two objects, one for the heci message packet
+	 * and another for the batch buffer we submit into GSC engine (that includes the packet).
+	 * NOTE: GSC-CS backend is currently only supported on MTL, so we allocate shmem.
+	 */
+	err = gsccs_create_buffer(pxp->ctrl_gt, "Heci Packet",
+				  2 * PXP43_MAX_HECI_INOUT_SIZE,
+				  &exec_res->pkt_vma, &exec_res->pkt_vaddr);
+	if (err)
+		return err;
+
+	err = gsccs_create_buffer(pxp->ctrl_gt, "Batch Buffer", PAGE_SIZE,
+				  &exec_res->bb_vma, &exec_res->bb_vaddr);
+	if (err)
+		goto free_pkt;
+
 	/* Finally, create an intel_context to be used during the submission */
 	ce = intel_context_create(engine);
 	if (IS_ERR(ce)) {
 		drm_err(&gt->i915->drm, "Failed creating gsccs backend ctx\n");
-		return PTR_ERR(ce);
+		err = PTR_ERR(ce);
+		goto free_batch;
 	}
 
 	i915_vm_put(ce->vm);
 	ce->vm = i915_vm_get(pxp->ctrl_gt->vm);
 	exec_res->ce = ce;
 
+	/* initialize host-session-handle (for all i915-to-gsc-firmware PXP cmds) */
+	get_random_bytes(&exec_res->host_session_handle, sizeof(exec_res->host_session_handle));
+
 	return 0;
+
+free_batch:
+	i915_vma_unpin_and_release(&exec_res->bb_vma, I915_VMA_RELEASE_MAP);
+free_pkt:
+	i915_vma_unpin_and_release(&exec_res->pkt_vma, I915_VMA_RELEASE_MAP);
+	memset(exec_res, 0, sizeof(*exec_res));
+
+	return err;
 }
 
 void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
index 354ea9a8f9403..bd1c028bc80f0 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
@@ -10,6 +10,10 @@
 
 struct intel_pxp;
 
+#define GSC_REPLY_LATENCY_MS 200
+#define GSC_PENDING_RETRY_MAXCOUNT 40
+#define GSC_PENDING_RETRY_PAUSE_MS 50
+
 #ifdef CONFIG_DRM_I915_PXP
 void intel_pxp_gsccs_fini(struct intel_pxp *pxp);
 int intel_pxp_gsccs_init(struct intel_pxp *pxp);
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
index 87d0d8da98cd5..1a8765866b8b6 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_types.h
@@ -38,6 +38,12 @@ struct intel_pxp {
 	struct gsccs_session_resources {
 		u64 host_session_handle; /* used by firmware to link commands to sessions */
 		struct intel_context *ce; /* context for gsc command submission */
+
+		struct i915_vma *pkt_vma; /* GSC FW cmd packet vma */
+		void *pkt_vaddr;  /* GSC FW cmd packet virt pointer */
+
+		struct i915_vma *bb_vma; /* HECI_PKT batch buffer vma */
+		void *bb_vaddr; /* HECI_PKT batch buffer virt pointer */
 	} gsccs_res;
 
 	/**
-- 
GitLab


From 99afb7cc8c44578615200ea4806b183e1e35a81d Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:35 -0700
Subject: [PATCH 0482/1791] drm/i915/pxp: Add ARB session creation and cleanup

Add MTL's function for ARB session creation using PXP firmware
version 4.3 ABI structure format.

While relooking at the ARB session creation flow in intel_pxp_start,
let's address missing UAPI documentation. Without actually changing
backward compatible behavior, update i915's drm-uapi comments
that describe the possible error values when creating a context
with I915_CONTEXT_PARAM_PROTECTED_CONTENT:
   Since the first merge of PXP support on ADL, i915 returns -ENXIO
   if a dependency such as firmware or component driver was yet to
   be loaded or returns -EIO if the creation attempt failed when
   requested by the PXP firmware (specific firmware error responses
   are reported in dmesg).

Add MTL's function for ARB session invalidation but this
reuses PXP firmware version 4.2 ABI structure format.

For both cases, in the back-end gsccs functions for sending messages
to the firmware inspect the GSC-CS-Mem-Header's pending-bit which
means the GSC firmware is busy and we should retry.

Given the last hw requirement, lets also update functions in
front-end layer that wait for session creation or teardown
completion to use new worst case timeout periods.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-6-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/pxp/intel_pxp.c          |  27 +++-
 .../drm/i915/pxp/intel_pxp_cmd_interface_43.h |  21 +++
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c    | 130 ++++++++++++++++++
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h    |  12 +-
 drivers/gpu/drm/i915/pxp/intel_pxp_session.c  |  11 +-
 include/uapi/drm/i915_drm.h                   |  15 ++
 6 files changed, 209 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index 8949d4be78828..b600d68de2a44 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -291,6 +291,8 @@ static bool pxp_component_bound(struct intel_pxp *pxp)
 
 static int __pxp_global_teardown_final(struct intel_pxp *pxp)
 {
+	int timeout;
+
 	if (!pxp->arb_is_valid)
 		return 0;
 	/*
@@ -300,7 +302,12 @@ static int __pxp_global_teardown_final(struct intel_pxp *pxp)
 	intel_pxp_mark_termination_in_progress(pxp);
 	intel_pxp_terminate(pxp, false);
 
-	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(250)))
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
+		timeout = GSCFW_MAX_ROUND_TRIP_LATENCY_MS;
+	else
+		timeout = 250;
+
+	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(timeout)))
 		return -ETIMEDOUT;
 
 	return 0;
@@ -308,6 +315,8 @@ static int __pxp_global_teardown_final(struct intel_pxp *pxp)
 
 static int __pxp_global_teardown_restart(struct intel_pxp *pxp)
 {
+	int timeout;
+
 	if (pxp->arb_is_valid)
 		return 0;
 	/*
@@ -316,7 +325,12 @@ static int __pxp_global_teardown_restart(struct intel_pxp *pxp)
 	 */
 	pxp_queue_termination(pxp);
 
-	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(250)))
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
+		timeout = GSCFW_MAX_ROUND_TRIP_LATENCY_MS;
+	else
+		timeout = 250;
+
+	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(timeout)))
 		return -ETIMEDOUT;
 
 	return 0;
@@ -354,8 +368,13 @@ int intel_pxp_start(struct intel_pxp *pxp)
 	if (!intel_pxp_is_enabled(pxp))
 		return -ENODEV;
 
-	if (wait_for(pxp_component_bound(pxp), 250))
-		return -ENXIO;
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0)) {
+		if (wait_for(intel_pxp_gsccs_is_ready_for_sessions(pxp), 250))
+			return -ENXIO;
+	} else {
+		if (wait_for(pxp_component_bound(pxp), 250))
+			return -ENXIO;
+	}
 
 	mutex_lock(&pxp->arb_mutex);
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
index c65ada99e54fd..09777719cd847 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
@@ -11,6 +11,7 @@
 
 /* PXP-Cmd-Op definitions */
 #define PXP43_CMDID_START_HUC_AUTH 0x0000003A
+#define PXP43_CMDID_INIT_SESSION 0x00000036
 
 /* PXP-Packet sizes for MTL's GSCCS-HECI instruction */
 #define PXP43_MAX_HECI_INOUT_SIZE (SZ_32K)
@@ -26,4 +27,24 @@ struct pxp43_start_huc_auth_out {
 	struct pxp_cmd_header header;
 } __packed;
 
+/* PXP-Input-Packet: Init PXP session */
+struct pxp43_create_arb_in {
+	struct pxp_cmd_header header;
+		/* header.stream_id fields for vesion 4.3 of Init PXP session: */
+		#define PXP43_INIT_SESSION_VALID BIT(0)
+		#define PXP43_INIT_SESSION_APPTYPE BIT(1)
+		#define PXP43_INIT_SESSION_APPID GENMASK(17, 2)
+	u32 protection_mode;
+		#define PXP43_INIT_SESSION_PROTECTION_ARB 0x2
+	u32 sub_session_id;
+	u32 init_flags;
+	u32 rsvd[12];
+} __packed;
+
+/* PXP-Input-Packet: Init PXP session */
+struct pxp43_create_arb_out {
+	struct pxp_cmd_header header;
+	u32 rsvd[8];
+} __packed;
+
 #endif /* __INTEL_PXP_FW_INTERFACE_43_H__ */
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
index 16e3b73d06538..4bc276daca16e 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -6,13 +6,46 @@
 #include "gem/i915_gem_internal.h"
 
 #include "gt/intel_context.h"
+#include "gt/uc/intel_gsc_fw.h"
 #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h"
 
 #include "i915_drv.h"
+#include "intel_pxp_cmd_interface_42.h"
 #include "intel_pxp_cmd_interface_43.h"
 #include "intel_pxp_gsccs.h"
 #include "intel_pxp_types.h"
 
+static bool
+is_fw_err_platform_config(u32 type)
+{
+	switch (type) {
+	case PXP_STATUS_ERROR_API_VERSION:
+	case PXP_STATUS_PLATFCONFIG_KF1_NOVERIF:
+	case PXP_STATUS_PLATFCONFIG_KF1_BAD:
+		return true;
+	default:
+		break;
+	}
+	return false;
+}
+
+static const char *
+fw_err_to_string(u32 type)
+{
+	switch (type) {
+	case PXP_STATUS_ERROR_API_VERSION:
+		return "ERR_API_VERSION";
+	case PXP_STATUS_NOT_READY:
+		return "ERR_NOT_READY";
+	case PXP_STATUS_PLATFCONFIG_KF1_NOVERIF:
+	case PXP_STATUS_PLATFCONFIG_KF1_BAD:
+		return "ERR_PLATFORM_CONFIG";
+	default:
+		break;
+	}
+	return NULL;
+}
+
 static int
 gsccs_send_message(struct intel_pxp *pxp,
 		   void *msg_in, size_t msg_in_size,
@@ -152,6 +185,103 @@ gsccs_send_message_retry_complete(struct intel_pxp *pxp,
 	return ret;
 }
 
+bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp)
+{
+	/*
+	 * GSC-fw loading, HuC-fw loading, HuC-fw authentication and
+	 * GSC-proxy init flow (requiring an mei component driver)
+	 * must all occur first before we can start requesting for PXP
+	 * sessions. Checking for completion on HuC authentication and
+	 * gsc-proxy init flow (the last set of dependencies that
+	 * are out of order) will suffice.
+	 */
+	if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc) &&
+	    intel_gsc_uc_fw_proxy_init_done(&pxp->ctrl_gt->uc.gsc))
+		return true;
+
+	return false;
+}
+
+int intel_pxp_gsccs_create_session(struct intel_pxp *pxp,
+				   int arb_session_id)
+{
+	struct drm_i915_private *i915 = pxp->ctrl_gt->i915;
+	struct pxp43_create_arb_in msg_in = {0};
+	struct pxp43_create_arb_out msg_out = {0};
+	int ret;
+
+	msg_in.header.api_version = PXP_APIVER(4, 3);
+	msg_in.header.command_id = PXP43_CMDID_INIT_SESSION;
+	msg_in.header.stream_id = (FIELD_PREP(PXP43_INIT_SESSION_APPID, arb_session_id) |
+				   FIELD_PREP(PXP43_INIT_SESSION_VALID, 1) |
+				   FIELD_PREP(PXP43_INIT_SESSION_APPTYPE, 0));
+	msg_in.header.buffer_len = sizeof(msg_in) - sizeof(msg_in.header);
+	msg_in.protection_mode = PXP43_INIT_SESSION_PROTECTION_ARB;
+
+	ret = gsccs_send_message_retry_complete(pxp,
+						&msg_in, sizeof(msg_in),
+						&msg_out, sizeof(msg_out), NULL);
+	if (ret) {
+		drm_err(&i915->drm, "Failed to init session %d, ret=[%d]\n", arb_session_id, ret);
+	} else if (msg_out.header.status != 0) {
+		if (is_fw_err_platform_config(msg_out.header.status)) {
+			drm_info_once(&i915->drm,
+				      "PXP init-session-%d failed due to BIOS/SOC:0x%08x:%s\n",
+				      arb_session_id, msg_out.header.status,
+				      fw_err_to_string(msg_out.header.status));
+		} else {
+			drm_dbg(&i915->drm, "PXP init-session-%d failed 0x%08x:%st:\n",
+				arb_session_id, msg_out.header.status,
+				fw_err_to_string(msg_out.header.status));
+			drm_dbg(&i915->drm, "     cmd-detail: ID=[0x%08x],API-Ver-[0x%08x]\n",
+				msg_in.header.command_id, msg_in.header.api_version);
+		}
+	}
+
+	return ret;
+}
+
+void intel_pxp_gsccs_end_arb_fw_session(struct intel_pxp *pxp, u32 session_id)
+{
+	struct drm_i915_private *i915 = pxp->ctrl_gt->i915;
+	struct pxp42_inv_stream_key_in msg_in = {0};
+	struct pxp42_inv_stream_key_out msg_out = {0};
+	int ret = 0;
+
+	/*
+	 * Stream key invalidation reuses the same version 4.2 input/output
+	 * command format but firmware requires 4.3 API interaction
+	 */
+	msg_in.header.api_version = PXP_APIVER(4, 3);
+	msg_in.header.command_id = PXP42_CMDID_INVALIDATE_STREAM_KEY;
+	msg_in.header.buffer_len = sizeof(msg_in) - sizeof(msg_in.header);
+
+	msg_in.header.stream_id = FIELD_PREP(PXP_CMDHDR_EXTDATA_SESSION_VALID, 1);
+	msg_in.header.stream_id |= FIELD_PREP(PXP_CMDHDR_EXTDATA_APP_TYPE, 0);
+	msg_in.header.stream_id |= FIELD_PREP(PXP_CMDHDR_EXTDATA_SESSION_ID, session_id);
+
+	ret = gsccs_send_message_retry_complete(pxp,
+						&msg_in, sizeof(msg_in),
+						&msg_out, sizeof(msg_out), NULL);
+	if (ret) {
+		drm_err(&i915->drm, "Failed to inv-stream-key-%u, ret=[%d]\n",
+			session_id, ret);
+	} else if (msg_out.header.status != 0) {
+		if (is_fw_err_platform_config(msg_out.header.status)) {
+			drm_info_once(&i915->drm,
+				      "PXP inv-stream-key-%u failed due to BIOS/SOC :0x%08x:%s\n",
+				      session_id, msg_out.header.status,
+				      fw_err_to_string(msg_out.header.status));
+		} else {
+			drm_dbg(&i915->drm, "PXP inv-stream-key-%u failed 0x%08x:%s:\n",
+				session_id, msg_out.header.status,
+				fw_err_to_string(msg_out.header.status));
+			drm_dbg(&i915->drm, "     cmd-detail: ID=[0x%08x],API-Ver-[0x%08x]\n",
+				msg_in.header.command_id, msg_in.header.api_version);
+		}
+	}
+}
+
 static void
 gsccs_cleanup_fw_host_session_handle(struct intel_pxp *pxp)
 {
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
index bd1c028bc80f0..298ad38e6c7df 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.h
@@ -10,14 +10,22 @@
 
 struct intel_pxp;
 
-#define GSC_REPLY_LATENCY_MS 200
+#define GSC_REPLY_LATENCY_MS 210
+/*
+ * Max FW response time is 200ms, to which we add 10ms to account for overhead
+ * such as request preparation, GuC submission to hw and pipeline completion times.
+ */
 #define GSC_PENDING_RETRY_MAXCOUNT 40
 #define GSC_PENDING_RETRY_PAUSE_MS 50
+#define GSCFW_MAX_ROUND_TRIP_LATENCY_MS (GSC_PENDING_RETRY_MAXCOUNT * GSC_PENDING_RETRY_PAUSE_MS)
 
 #ifdef CONFIG_DRM_I915_PXP
 void intel_pxp_gsccs_fini(struct intel_pxp *pxp);
 int intel_pxp_gsccs_init(struct intel_pxp *pxp);
 
+int intel_pxp_gsccs_create_session(struct intel_pxp *pxp, int arb_session_id);
+void intel_pxp_gsccs_end_arb_fw_session(struct intel_pxp *pxp, u32 arb_session_id);
+
 #else
 static inline void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
 {
@@ -30,4 +38,6 @@ static inline int intel_pxp_gsccs_init(struct intel_pxp *pxp)
 
 #endif
 
+bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp);
+
 #endif /*__INTEL_PXP_GSCCS_H__ */
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
index 7899079e17b06..e4d8242302c53 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
@@ -7,6 +7,7 @@
 
 #include "intel_pxp.h"
 #include "intel_pxp_cmd.h"
+#include "intel_pxp_gsccs.h"
 #include "intel_pxp_session.h"
 #include "intel_pxp_tee.h"
 #include "intel_pxp_types.h"
@@ -62,7 +63,10 @@ static int pxp_create_arb_session(struct intel_pxp *pxp)
 		return -EEXIST;
 	}
 
-	ret = intel_pxp_tee_cmd_create_arb_session(pxp, ARB_SESSION);
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
+		ret = intel_pxp_gsccs_create_session(pxp, ARB_SESSION);
+	else
+		ret = intel_pxp_tee_cmd_create_arb_session(pxp, ARB_SESSION);
 	if (ret) {
 		drm_err(&gt->i915->drm, "tee cmd for arb session creation failed\n");
 		return ret;
@@ -106,7 +110,10 @@ static int pxp_terminate_arb_session_and_global(struct intel_pxp *pxp)
 
 	intel_uncore_write(gt->uncore, KCR_GLOBAL_TERMINATE(pxp->kcr_base), 1);
 
-	intel_pxp_tee_end_arb_fw_session(pxp, ARB_SESSION);
+	if (HAS_ENGINE(gt, GSC0))
+		intel_pxp_gsccs_end_arb_fw_session(pxp, ARB_SESSION);
+	else
+		intel_pxp_tee_end_arb_fw_session(pxp, ARB_SESSION);
 
 	return ret;
 }
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index dba7c5a5b25e9..0aa3190e76540 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -2096,6 +2096,21 @@ struct drm_i915_gem_context_param {
  *
  * -ENODEV: feature not available
  * -EPERM: trying to mark a recoverable or not bannable context as protected
+ * -ENXIO: A dependency such as a component driver or firmware is not yet
+ *         loaded so user space may need to attempt again. Depending on the
+ *         device, this error may be reported if protected context creation is
+ *         attempted very early after kernel start because the internal timeout
+ *         waiting for such dependencies is not guaranteed to be larger than
+ *         required (numbers differ depending on system and kernel config):
+ *            - ADL/RPL: dependencies may take up to 3 seconds from kernel start
+ *                       while context creation internal timeout is 250 milisecs
+ *            - MTL: dependencies may take up to 8 seconds from kernel start
+ *                   while context creation internal timeout is 250 milisecs
+ *         NOTE: such dependencies happen once, so a subsequent call to create a
+ *         protected context after a prior successful call will not experience
+ *         such timeouts and will not return -ENXIO (unless the driver is reloaded,
+ *         or, depending on the device, resumes from a suspended state).
+ * -EIO: The firmware did not succeed in creating the protected context.
  */
 #define I915_CONTEXT_PARAM_PROTECTED_CONTENT    0xd
 /* Must be kept compact -- no holes and well documented */
-- 
GitLab


From d1da138f245d4fb46b21d2ddb19504a2831d813f Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:36 -0700
Subject: [PATCH 0483/1791] drm/i915/uapi/pxp: Add a GET_PARAM for PXP

Because of the additional firmware, component-driver and
initialization depedencies required on MTL platform before a
PXP context can be created, UMD calling for PXP creation as a
way to get-caps can take a long time. An actual real world
customer stack has seen this happen in the 4-to-8 second range
after the kernel starts (which sees MESA's init appear in the
middle of this range as the compositor comes up). To avoid
unncessary delays experienced by the UMD for get-caps purposes,
add a GET_PARAM for I915_PARAM_PXP_SUPPORT.

However, some failures can still occur after all the depedencies
are met (such as firmware init flow failure, bios configurations
or SOC fusing not allowing PXP enablement). Those scenarios will
only be known to user space when it attempts creating a PXP context
and is documented in the GEM UAPI headers.

While making this change, create a helper that is common to both
GET_PARAM caller and intel_pxp_start since the latter does
similar checks.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Acked-by: Jordan Justen <jordan.l.justen@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-7-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/i915_getparam.c |  7 +++++++
 drivers/gpu/drm/i915/pxp/intel_pxp.c | 29 +++++++++++++++++++++-------
 drivers/gpu/drm/i915/pxp/intel_pxp.h |  1 +
 include/uapi/drm/i915_drm.h          | 19 ++++++++++++++++++
 4 files changed, 49 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c
index 2238e096c9579..6f11d7eaa91a9 100644
--- a/drivers/gpu/drm/i915/i915_getparam.c
+++ b/drivers/gpu/drm/i915/i915_getparam.c
@@ -5,6 +5,8 @@
 #include "gem/i915_gem_mman.h"
 #include "gt/intel_engine_user.h"
 
+#include "pxp/intel_pxp.h"
+
 #include "i915_cmd_parser.h"
 #include "i915_drv.h"
 #include "i915_getparam.h"
@@ -102,6 +104,11 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		if (value < 0)
 			return value;
 		break;
+	case I915_PARAM_PXP_STATUS:
+		value = intel_pxp_get_readiness_status(i915->pxp);
+		if (value < 0)
+			return value;
+		break;
 	case I915_PARAM_MMAP_GTT_VERSION:
 		/* Though we've started our numbering from 1, and so class all
 		 * earlier versions as 0, in effect their value is undefined as
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index b600d68de2a44..f143eadbc253e 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -358,23 +358,38 @@ void intel_pxp_end(struct intel_pxp *pxp)
 }
 
 /*
- * the arb session is restarted from the irq work when we receive the
- * termination completion interrupt
+ * this helper is used by both intel_pxp_start and by
+ * the GET_PARAM IOCTL that user space calls. Thus, the
+ * return values here should match the UAPI spec.
  */
-int intel_pxp_start(struct intel_pxp *pxp)
+int intel_pxp_get_readiness_status(struct intel_pxp *pxp)
 {
-	int ret = 0;
-
 	if (!intel_pxp_is_enabled(pxp))
 		return -ENODEV;
 
 	if (HAS_ENGINE(pxp->ctrl_gt, GSC0)) {
 		if (wait_for(intel_pxp_gsccs_is_ready_for_sessions(pxp), 250))
-			return -ENXIO;
+			return 2;
 	} else {
 		if (wait_for(pxp_component_bound(pxp), 250))
-			return -ENXIO;
+			return 2;
 	}
+	return 1;
+}
+
+/*
+ * the arb session is restarted from the irq work when we receive the
+ * termination completion interrupt
+ */
+int intel_pxp_start(struct intel_pxp *pxp)
+{
+	int ret = 0;
+
+	ret = intel_pxp_get_readiness_status(pxp);
+	if (ret < 0)
+		return ret;
+	else if (ret > 1)
+		return -EIO; /* per UAPI spec, user may retry later */
 
 	mutex_lock(&pxp->arb_mutex);
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.h b/drivers/gpu/drm/i915/pxp/intel_pxp.h
index 3ded0890cd277..17e6dbc86b388 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.h
@@ -26,6 +26,7 @@ void intel_pxp_fini_hw(struct intel_pxp *pxp);
 void intel_pxp_mark_termination_in_progress(struct intel_pxp *pxp);
 void intel_pxp_tee_end_arb_fw_session(struct intel_pxp *pxp, u32 arb_session_id);
 
+int intel_pxp_get_readiness_status(struct intel_pxp *pxp);
 int intel_pxp_start(struct intel_pxp *pxp);
 void intel_pxp_end(struct intel_pxp *pxp);
 
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 0aa3190e76540..ba40855dbc931 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -771,6 +771,25 @@ typedef struct drm_i915_irq_wait {
  */
 #define I915_PARAM_OA_TIMESTAMP_FREQUENCY 57
 
+/*
+ * Query the status of PXP support in i915.
+ *
+ * The query can fail in the following scenarios with the listed error codes:
+ *     -ENODEV = PXP support is not available on the GPU device or in the
+ *               kernel due to missing component drivers or kernel configs.
+ *
+ * If the IOCTL is successful, the returned parameter will be set to one of
+ * the following values:
+ *     1 = PXP feature is supported and is ready for use.
+ *     2 = PXP feature is supported but should be ready soon (pending
+ *         initialization of non-i915 system dependencies).
+ *
+ * NOTE: When param is supported (positive return values), user space should
+ *       still refer to the GEM PXP context-creation UAPI header specs to be
+ *       aware of possible failure due to system state machine at the time.
+ */
+#define I915_PARAM_PXP_STATUS		 58
+
 /* Must be kept compact -- no holes and well documented */
 
 /**
-- 
GitLab


From 9e134ed720b6e69a5c857743daedd403101ca078 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:37 -0700
Subject: [PATCH 0484/1791] drm/i915/pxp: On MTL, KCR enabling doesn't wait on
 tee component

On legacy platforms, KCR HW enabling is done at the time the mei
component interface is bound. It's also disabled during unbind.
However, for MTL onwards, we don't depend on a tee component
to start sending GSC-CS firmware messages.

Thus, immediately enable (or disable) KCR HW on PXP's init,
fini and resume.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-8-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c | 15 ++++++++++++++-
 drivers/gpu/drm/i915/pxp/intel_pxp_pm.c    |  3 ++-
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
index 4bc276daca16e..8dc41de3f6f74 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -10,6 +10,7 @@
 #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h"
 
 #include "i915_drv.h"
+#include "intel_pxp.h"
 #include "intel_pxp_cmd_interface_42.h"
 #include "intel_pxp_cmd_interface_43.h"
 #include "intel_pxp_gsccs.h"
@@ -422,10 +423,22 @@ gsccs_allocate_execution_resource(struct intel_pxp *pxp)
 
 void intel_pxp_gsccs_fini(struct intel_pxp *pxp)
 {
+	intel_wakeref_t wakeref;
+
 	gsccs_destroy_execution_resource(pxp);
+	with_intel_runtime_pm(&pxp->ctrl_gt->i915->runtime_pm, wakeref)
+		intel_pxp_fini_hw(pxp);
 }
 
 int intel_pxp_gsccs_init(struct intel_pxp *pxp)
 {
-	return gsccs_allocate_execution_resource(pxp);
+	int ret;
+	intel_wakeref_t wakeref;
+
+	ret = gsccs_allocate_execution_resource(pxp);
+	if (!ret) {
+		with_intel_runtime_pm(&pxp->ctrl_gt->i915->runtime_pm, wakeref)
+			intel_pxp_init_hw(pxp);
+	}
+	return ret;
 }
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_pm.c b/drivers/gpu/drm/i915/pxp/intel_pxp_pm.c
index 4f836b3174240..1a04067f61fc7 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_pm.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_pm.c
@@ -43,8 +43,9 @@ void intel_pxp_resume_complete(struct intel_pxp *pxp)
 	 * The PXP component gets automatically unbound when we go into S3 and
 	 * re-bound after we come out, so in that scenario we can defer the
 	 * hw init to the bind call.
+	 * NOTE: GSC-CS backend doesn't rely on components.
 	 */
-	if (!pxp->pxp_component)
+	if (!HAS_ENGINE(pxp->ctrl_gt, GSC0) && !pxp->pxp_component)
 		return;
 
 	intel_pxp_init_hw(pxp);
-- 
GitLab


From 41e65d8790bd96d1caa3cb136a57ac45a131e66d Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 11 May 2023 16:17:38 -0700
Subject: [PATCH 0485/1791] drm/i915/pxp: Enable PXP with MTL-GSC-CS

Enable PXP with MTL-GSC-CS: add the has_pxp into device info
and increase the debugfs teardown timeouts to align with
new GSC-CS + firmware specs.

Now that we have 3 places that are selecting pxp timeouts
based on tee vs gsccs back-end, let's add a helper.

Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511231738.1077674-9-alan.previn.teres.alexis@intel.com
---
 drivers/gpu/drm/i915/i915_pci.c              |  1 +
 drivers/gpu/drm/i915/pxp/intel_pxp.c         | 18 ++++++++++--------
 drivers/gpu/drm/i915/pxp/intel_pxp.h         |  1 +
 drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c |  6 +++++-
 drivers/gpu/drm/i915/pxp/intel_pxp_session.c |  2 +-
 5 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 1a3247451d0f8..3e38283cbe8f7 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -1209,6 +1209,7 @@ static const struct intel_device_info mtl_info = {
 	.has_mslice_steering = 0,
 	.has_snoop = 1,
 	.max_pat_index = 4,
+	.has_pxp = 1,
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(CCS0),
 	.require_force_probe = 1,
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index f143eadbc253e..bb2e15329f346 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -289,6 +289,14 @@ static bool pxp_component_bound(struct intel_pxp *pxp)
 	return bound;
 }
 
+int intel_pxp_get_backend_timeout_ms(struct intel_pxp *pxp)
+{
+	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
+		return GSCFW_MAX_ROUND_TRIP_LATENCY_MS;
+	else
+		return 250;
+}
+
 static int __pxp_global_teardown_final(struct intel_pxp *pxp)
 {
 	int timeout;
@@ -302,10 +310,7 @@ static int __pxp_global_teardown_final(struct intel_pxp *pxp)
 	intel_pxp_mark_termination_in_progress(pxp);
 	intel_pxp_terminate(pxp, false);
 
-	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
-		timeout = GSCFW_MAX_ROUND_TRIP_LATENCY_MS;
-	else
-		timeout = 250;
+	timeout = intel_pxp_get_backend_timeout_ms(pxp);
 
 	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(timeout)))
 		return -ETIMEDOUT;
@@ -325,10 +330,7 @@ static int __pxp_global_teardown_restart(struct intel_pxp *pxp)
 	 */
 	pxp_queue_termination(pxp);
 
-	if (HAS_ENGINE(pxp->ctrl_gt, GSC0))
-		timeout = GSCFW_MAX_ROUND_TRIP_LATENCY_MS;
-	else
-		timeout = 250;
+	timeout = intel_pxp_get_backend_timeout_ms(pxp);
 
 	if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(timeout)))
 		return -ETIMEDOUT;
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.h b/drivers/gpu/drm/i915/pxp/intel_pxp.h
index 17e6dbc86b388..17254c3f1267a 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.h
@@ -27,6 +27,7 @@ void intel_pxp_mark_termination_in_progress(struct intel_pxp *pxp);
 void intel_pxp_tee_end_arb_fw_session(struct intel_pxp *pxp, u32 arb_session_id);
 
 int intel_pxp_get_readiness_status(struct intel_pxp *pxp);
+int intel_pxp_get_backend_timeout_ms(struct intel_pxp *pxp);
 int intel_pxp_start(struct intel_pxp *pxp);
 void intel_pxp_end(struct intel_pxp *pxp);
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
index 4b8e70caa3adc..e07c5b3807891 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
@@ -14,6 +14,7 @@
 
 #include "intel_pxp.h"
 #include "intel_pxp_debugfs.h"
+#include "intel_pxp_gsccs.h"
 #include "intel_pxp_irq.h"
 #include "intel_pxp_types.h"
 
@@ -45,6 +46,7 @@ static int pxp_terminate_set(void *data, u64 val)
 {
 	struct intel_pxp *pxp = data;
 	struct intel_gt *gt = pxp->ctrl_gt;
+	int timeout_ms;
 
 	if (!intel_pxp_is_active(pxp))
 		return -ENODEV;
@@ -54,8 +56,10 @@ static int pxp_terminate_set(void *data, u64 val)
 	intel_pxp_irq_handler(pxp, GEN12_DISPLAY_PXP_STATE_TERMINATED_INTERRUPT);
 	spin_unlock_irq(gt->irq_lock);
 
+	timeout_ms = intel_pxp_get_backend_timeout_ms(pxp);
+
 	if (!wait_for_completion_timeout(&pxp->termination,
-					 msecs_to_jiffies(100)))
+					 msecs_to_jiffies(timeout_ms)))
 		return -ETIMEDOUT;
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
index e4d8242302c53..0a3e66b0265e8 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c
@@ -44,7 +44,7 @@ static int pxp_wait_for_session_state(struct intel_pxp *pxp, u32 id, bool in_pla
 				      KCR_SIP(pxp->kcr_base),
 				      mask,
 				      in_play ? mask : 0,
-				      100);
+				      250);
 
 	intel_runtime_pm_put(uncore->rpm, wakeref);
 
-- 
GitLab


From 6e4e9fbd5ba01eed13cb086ea645f8513370761d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 May 2023 18:21:53 +0300
Subject: [PATCH 0486/1791] drm/i915/gt: drop dependency on VLV_DISPLAY_BASE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CHV_FUSE_GT (0x182168) is purely about GT fuses, therefore belongs in
intel_gt_regs.h, is in the gcfgmmio unit, but is technically in the VLV
display base area.

Add VLV_GUNIT_BASE to drop dependency on VLV_DISPLAY_BASE and thus
display/intel_display_reg_defs.h in intel_gt_regs.h.

v2: Add VLV_GUNIT_BASE (Ville)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511152153.986676-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_regs.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
index 4aecb5a7b6318..3bb4098fe97a0 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
@@ -7,7 +7,8 @@
 #define __INTEL_GT_REGS__
 
 #include "i915_reg_defs.h"
-#include "display/intel_display_reg_defs.h"	/* VLV_DISPLAY_BASE */
+
+#define VLV_GUNIT_BASE			0x180000
 
 /*
  * The perf control registers are technically multicast registers, but the
@@ -1450,7 +1451,7 @@
 #define GEN12_RCU_MODE				_MMIO(0x14800)
 #define   GEN12_RCU_MODE_CCS_ENABLE		REG_BIT(0)
 
-#define CHV_FUSE_GT				_MMIO(VLV_DISPLAY_BASE + 0x2168)
+#define CHV_FUSE_GT				_MMIO(VLV_GUNIT_BASE + 0x2168)
 #define   CHV_FGT_DISABLE_SS0			(1 << 10)
 #define   CHV_FGT_DISABLE_SS1			(1 << 11)
 #define   CHV_FGT_EU_DIS_SS0_R0_SHIFT		16
-- 
GitLab


From db2ce1ab0508cd95efb4be938a146472c56c9461 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 10 May 2023 13:55:56 -0700
Subject: [PATCH 0487/1791] drm/i1915/guc: Fix probe injection CI failures
 after recent change

A recent change bumped a 'notice' message up to 'error' level for
debug builds to help trap incorrect configurations in CI systems.
Unfortunately, the error condition in question is triggered by the
error injection probe test. So change the message again to be 'probe
error' level instead.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Fixes: 760133d42f0a ("drm/i915/uc: Make unexpected firmware versions an error in debug builds")
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510205556.312999-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index e34c791200903..dc5c96c503a92 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -18,7 +18,7 @@
 #include "i915_reg.h"
 
 #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
-#define UNEXPECTED	gt_err
+#define UNEXPECTED	gt_probe_error
 #else
 #define UNEXPECTED	gt_notice
 #endif
-- 
GitLab


From 8ba3ba992fc2e456f4211ac4dc80dcb7775e722f Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Thu, 11 May 2023 18:35:44 -0700
Subject: [PATCH 0488/1791] drm/i915/guc: Fix confused register capture list
 creation

The GuC has a completely separate engine class enum when referring to
register capture lists, which combines render and compute. The driver
was using the 'normal' GuC specific engine class enum instead. That
meant that it thought it was defining a capture list for compute
engines, the list was actually being applied to the GSC engine. And if
a platform didn't have a render engine, then it would get no compute
register captures at all.

Fix that.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512013544.3367606-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c    | 36 ++++++++++-
 .../gpu/drm/i915/gt/uc/intel_guc_capture.c    | 61 +++++++++----------
 drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h   |  9 +++
 3 files changed, 72 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c
index 69ce06faf8cda..63724e17829a7 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c
@@ -643,6 +643,39 @@ static void guc_init_golden_context(struct intel_guc *guc)
 	GEM_BUG_ON(guc->ads_golden_ctxt_size != total_size);
 }
 
+static u32 guc_get_capture_engine_mask(struct iosys_map *info_map, u32 capture_class)
+{
+	u32 mask;
+
+	switch (capture_class) {
+	case GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE:
+		mask = info_map_read(info_map, engine_enabled_masks[GUC_RENDER_CLASS]);
+		mask |= info_map_read(info_map, engine_enabled_masks[GUC_COMPUTE_CLASS]);
+		break;
+
+	case GUC_CAPTURE_LIST_CLASS_VIDEO:
+		mask = info_map_read(info_map, engine_enabled_masks[GUC_VIDEO_CLASS]);
+		break;
+
+	case GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE:
+		mask = info_map_read(info_map, engine_enabled_masks[GUC_VIDEOENHANCE_CLASS]);
+		break;
+
+	case GUC_CAPTURE_LIST_CLASS_BLITTER:
+		mask = info_map_read(info_map, engine_enabled_masks[GUC_BLITTER_CLASS]);
+		break;
+
+	case GUC_CAPTURE_LIST_CLASS_GSC_OTHER:
+		mask = info_map_read(info_map, engine_enabled_masks[GUC_GSC_OTHER_CLASS]);
+		break;
+
+	default:
+		mask = 0;
+	}
+
+	return mask;
+}
+
 static int
 guc_capture_prep_lists(struct intel_guc *guc)
 {
@@ -678,9 +711,10 @@ guc_capture_prep_lists(struct intel_guc *guc)
 
 	for (i = 0; i < GUC_CAPTURE_LIST_INDEX_MAX; i++) {
 		for (j = 0; j < GUC_MAX_ENGINE_CLASSES; j++) {
+			u32 engine_mask = guc_get_capture_engine_mask(&info_map, j);
 
 			/* null list if we dont have said engine or list */
-			if (!info_map_read(&info_map, engine_enabled_masks[j])) {
+			if (!engine_mask) {
 				if (ads_is_mapped) {
 					ads_blob_write(guc, ads.capture_class[i][j], null_ggtt);
 					ads_blob_write(guc, ads.capture_instance[i][j], null_ggtt);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index 1def0b6467c79..0ff864da92dfe 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -174,35 +174,31 @@ static const struct __guc_mmio_reg_descr empty_regs_list[] = {
 /* List of lists */
 static const struct __guc_mmio_reg_descr_group gen8_lists[] = {
 	MAKE_REGLIST(gen8_global_regs, PF, GLOBAL, 0),
-	MAKE_REGLIST(gen8_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
-	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
-	MAKE_REGLIST(gen8_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_GSC_OTHER_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
+	MAKE_REGLIST(gen8_rc_class_regs, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_VIDEO),
+	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_VIDEO),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE),
+	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_BLITTER),
+	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_BLITTER),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_GSC_OTHER),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_GSC_OTHER),
 	{}
 };
 
 static const struct __guc_mmio_reg_descr_group xe_lp_lists[] = {
 	MAKE_REGLIST(xe_lp_global_regs, PF, GLOBAL, 0),
-	MAKE_REGLIST(xe_lp_rc_class_regs, PF, ENGINE_CLASS, GUC_RENDER_CLASS),
-	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_RENDER_CLASS),
-	MAKE_REGLIST(xe_lp_rc_class_regs, PF, ENGINE_CLASS, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_COMPUTE_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEO_CLASS),
-	MAKE_REGLIST(xe_lp_vec_class_regs, PF, ENGINE_CLASS, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_VIDEOENHANCE_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_BLITTER_CLASS),
-	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_GSC_OTHER_CLASS),
-	MAKE_REGLIST(xe_lp_gsc_inst_regs, PF, ENGINE_INSTANCE, GUC_GSC_OTHER_CLASS),
+	MAKE_REGLIST(xe_lp_rc_class_regs, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE),
+	MAKE_REGLIST(gen8_rc_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_VIDEO),
+	MAKE_REGLIST(gen8_vd_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_VIDEO),
+	MAKE_REGLIST(xe_lp_vec_class_regs, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE),
+	MAKE_REGLIST(gen8_vec_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_BLITTER),
+	MAKE_REGLIST(gen8_blt_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_BLITTER),
+	MAKE_REGLIST(empty_regs_list, PF, ENGINE_CLASS, GUC_CAPTURE_LIST_CLASS_GSC_OTHER),
+	MAKE_REGLIST(xe_lp_gsc_inst_regs, PF, ENGINE_INSTANCE, GUC_CAPTURE_LIST_CLASS_GSC_OTHER),
 	{}
 };
 
@@ -310,7 +306,8 @@ guc_capture_alloc_steered_lists(struct intel_guc *guc,
 
 	/* steered registers currently only exist for the render-class */
 	list = guc_capture_get_one_list(lists, GUC_CAPTURE_LIST_INDEX_PF,
-					GUC_CAPTURE_LIST_TYPE_ENGINE_CLASS, GUC_RENDER_CLASS);
+					GUC_CAPTURE_LIST_TYPE_ENGINE_CLASS,
+					GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE);
 	/* skip if extlists was previously allocated */
 	if (!list || guc->capture->extlists)
 		return;
@@ -400,17 +397,15 @@ static const char *
 __stringify_engclass(u32 class)
 {
 	switch (class) {
-	case GUC_RENDER_CLASS:
-		return "Render";
-	case GUC_VIDEO_CLASS:
+	case GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE:
+		return "Render/Compute";
+	case GUC_CAPTURE_LIST_CLASS_VIDEO:
 		return "Video";
-	case GUC_VIDEOENHANCE_CLASS:
+	case GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE:
 		return "VideoEnhance";
-	case GUC_BLITTER_CLASS:
+	case GUC_CAPTURE_LIST_CLASS_BLITTER:
 		return "Blitter";
-	case GUC_COMPUTE_CLASS:
-		return "Compute";
-	case GUC_GSC_OTHER_CLASS:
+	case GUC_CAPTURE_LIST_CLASS_GSC_OTHER:
 		return "GSC-Other";
 	default:
 		break;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
index 8bb9ed7c7b4dd..4e57bd09d50d9 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
@@ -411,6 +411,15 @@ enum guc_capture_type {
 	GUC_CAPTURE_LIST_TYPE_MAX,
 };
 
+/* Class indecies for capture_class and capture_instance arrays */
+enum {
+	GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE = 0,
+	GUC_CAPTURE_LIST_CLASS_VIDEO = 1,
+	GUC_CAPTURE_LIST_CLASS_VIDEOENHANCE = 2,
+	GUC_CAPTURE_LIST_CLASS_BLITTER = 3,
+	GUC_CAPTURE_LIST_CLASS_GSC_OTHER = 4,
+};
+
 /* GuC Additional Data Struct */
 struct guc_ads {
 	struct guc_mmio_reg_set reg_state_list[GUC_MAX_ENGINE_CLASSES][GUC_MAX_INSTANCES_PER_CLASS];
-- 
GitLab


From 6986f05b557bf1efea5bac8dbdffb7ee89d0cf77 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Fri, 5 May 2023 09:45:12 +0530
Subject: [PATCH 0489/1791] drm/i915/hdcp: Check if media_gt exists

Check if media_gt exists if we are using gsc cs

--v2
-correct typo [Ankit]
-assign gsc variable if gt exists [Ankit]

--v3
-declare gsc and gt variables in if block [Ankit]

--v4
-add fixes tag [Ankit]

Fixes: 883631771038 ("drm/i915/mtl: Add HDCP GSC interface")
Cc: Tomas Winkler <tomas.winkler@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Anshuman Gupta <anshuman.gupta@intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230505041512.585486-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 650232c4892b1..b183efab04a1d 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -204,8 +204,6 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
-	struct intel_gt *gt = dev_priv->media_gt;
-	struct intel_gsc_uc *gsc = &gt->uc.gsc;
 	bool capable = false;
 
 	/* I915 support for HDCP2.2 */
@@ -213,9 +211,13 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 		return false;
 
 	/* If MTL+ make sure gsc is loaded and proxy is setup */
-	if (intel_hdcp_gsc_cs_required(dev_priv))
-		if (!intel_uc_fw_is_running(&gsc->fw))
+	if (intel_hdcp_gsc_cs_required(dev_priv)) {
+		struct intel_gt *gt = dev_priv->media_gt;
+		struct intel_gsc_uc *gsc = gt ? &gt->uc.gsc : NULL;
+
+		if (!gsc || !intel_uc_fw_is_running(&gsc->fw))
 			return false;
+	}
 
 	/* MEI/GSC interface is solid depending on which is used */
 	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-- 
GitLab


From 3631c363b7c1d203a84d35e1f3d76160d8bd1961 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 12 May 2023 14:14:46 +0300
Subject: [PATCH 0490/1791] drm/i915/bios: add helper for reading SPI

Add helper for reading SPI to not duplicate the write&read combo
everywhere.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512111446.1524038-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_bios.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 64eb11a452659..34a397adbd6b0 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -3033,6 +3033,13 @@ bool intel_bios_is_valid_vbt(const void *buf, size_t size)
 	return vbt;
 }
 
+static u32 intel_spi_read(struct intel_uncore *uncore, u32 offset)
+{
+	intel_uncore_write(uncore, PRIMARY_SPI_ADDRESS, offset);
+
+	return intel_uncore_read(uncore, PRIMARY_SPI_TRIGGER);
+}
+
 static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915)
 {
 	u32 count, data, found, store = 0;
@@ -3049,9 +3056,7 @@ static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915)
 	oprom_offset &= OROM_OFFSET_MASK;
 
 	for (count = 0; count < oprom_size; count += 4) {
-		intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, oprom_offset + count);
-		data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER);
-
+		data = intel_spi_read(&i915->uncore, oprom_offset + count);
 		if (data == *((const u32 *)"$VBT")) {
 			found = oprom_offset + count;
 			break;
@@ -3062,20 +3067,16 @@ static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915)
 		goto err_not_found;
 
 	/* Get VBT size and allocate space for the VBT */
-	intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found +
-		   offsetof(struct vbt_header, vbt_size));
-	vbt_size = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER);
+	vbt_size = intel_spi_read(&i915->uncore,
+				  found + offsetof(struct vbt_header, vbt_size));
 	vbt_size &= 0xffff;
 
 	vbt = kzalloc(round_up(vbt_size, 4), GFP_KERNEL);
 	if (!vbt)
 		goto err_not_found;
 
-	for (count = 0; count < vbt_size; count += 4) {
-		intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found + count);
-		data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER);
-		*(vbt + store++) = data;
-	}
+	for (count = 0; count < vbt_size; count += 4)
+		*(vbt + store++) = intel_spi_read(&i915->uncore, found + count);
 
 	if (!intel_bios_is_valid_vbt(vbt, vbt_size))
 		goto err_free_vbt;
-- 
GitLab


From c7c12de893f808bd7c1215fe9056262295e5203b Mon Sep 17 00:00:00 2001
From: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Date: Sun, 14 May 2023 21:42:39 +0300
Subject: [PATCH 0491/1791] drm/fourcc: define Intel Meteorlake related ccs
 modifiers

Add Tile4 type ccs modifiers with aux buffer needed for MTL

Bspec: 49251, 49252, 49253
Cc: dri-devel@lists.freedesktop.org
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230514184240.6184-1-juhapekka.heikkila@gmail.com
---
 include/uapi/drm/drm_fourcc.h | 43 +++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h
index de703c6be9698..8db7fd3f743e4 100644
--- a/include/uapi/drm/drm_fourcc.h
+++ b/include/uapi/drm/drm_fourcc.h
@@ -657,6 +657,49 @@ extern "C" {
  */
 #define I915_FORMAT_MOD_4_TILED_DG2_RC_CCS_CC fourcc_mod_code(INTEL, 12)
 
+/*
+ * Intel Color Control Surfaces (CCS) for display ver. 14 render compression.
+ *
+ * The main surface is tile4 and at plane index 0, the CCS is linear and
+ * at index 1. A 64B CCS cache line corresponds to an area of 4x1 tiles in
+ * main surface. In other words, 4 bits in CCS map to a main surface cache
+ * line pair. The main surface pitch is required to be a multiple of four
+ * tile4 widths.
+ */
+#define I915_FORMAT_MOD_4_TILED_MTL_RC_CCS fourcc_mod_code(INTEL, 13)
+
+/*
+ * Intel Color Control Surfaces (CCS) for display ver. 14 media compression
+ *
+ * The main surface is tile4 and at plane index 0, the CCS is linear and
+ * at index 1. A 64B CCS cache line corresponds to an area of 4x1 tiles in
+ * main surface. In other words, 4 bits in CCS map to a main surface cache
+ * line pair. The main surface pitch is required to be a multiple of four
+ * tile4 widths. For semi-planar formats like NV12, CCS planes follow the
+ * Y and UV planes i.e., planes 0 and 1 are used for Y and UV surfaces,
+ * planes 2 and 3 for the respective CCS.
+ */
+#define I915_FORMAT_MOD_4_TILED_MTL_MC_CCS fourcc_mod_code(INTEL, 14)
+
+/*
+ * Intel Color Control Surface with Clear Color (CCS) for display ver. 14 render
+ * compression.
+ *
+ * The main surface is tile4 and is at plane index 0 whereas CCS is linear
+ * and at index 1. The clear color is stored at index 2, and the pitch should
+ * be ignored. The clear color structure is 256 bits. The first 128 bits
+ * represents Raw Clear Color Red, Green, Blue and Alpha color each represented
+ * by 32 bits. The raw clear color is consumed by the 3d engine and generates
+ * the converted clear color of size 64 bits. The first 32 bits store the Lower
+ * Converted Clear Color value and the next 32 bits store the Higher Converted
+ * Clear Color value when applicable. The Converted Clear Color values are
+ * consumed by the DE. The last 64 bits are used to store Color Discard Enable
+ * and Depth Clear Value Valid which are ignored by the DE. A CCS cache line
+ * corresponds to an area of 4x1 tiles in the main surface. The main surface
+ * pitch is required to be a multiple of 4 tile widths.
+ */
+#define I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC fourcc_mod_code(INTEL, 15)
+
 /*
  * Tiled, NV12MT, grouped in 64 (pixels) x 32 (lines) -sized macroblocks
  *
-- 
GitLab


From f2eb43f0b8c25e58b299ae2593d805c35c98f82f Mon Sep 17 00:00:00 2001
From: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Date: Sun, 14 May 2023 21:42:40 +0300
Subject: [PATCH 0492/1791] drm/i915/mtl: Add handling for MTL ccs modifiers

Add Tile4 ccs modifiers w/ auxbuffer handling

Implement Wa_14017240301

Bspec: 49251, 49252, 49253
Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230514184240.6184-2-juhapekka.heikkila@gmail.com
---
 drivers/gpu/drm/i915/display/intel_fb.c       | 42 ++++++++++++++++++-
 .../drm/i915/display/skl_universal_plane.c    | 22 +++++++++-
 2 files changed, 61 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index e5f637897b5e4..9db7aabef9b67 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -157,6 +157,32 @@ struct intel_modifier_desc {
 
 static const struct intel_modifier_desc intel_modifiers[] = {
 	{
+		.modifier = I915_FORMAT_MOD_4_TILED_MTL_MC_CCS,
+		.display_ver = { 14, 14 },
+		.plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_MC,
+
+		.ccs.packed_aux_planes = BIT(1),
+		.ccs.planar_aux_planes = BIT(2) | BIT(3),
+
+		FORMAT_OVERRIDE(gen12_ccs_formats),
+	}, {
+		.modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS,
+		.display_ver = { 14, 14 },
+		.plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_RC,
+
+		.ccs.packed_aux_planes = BIT(1),
+
+		FORMAT_OVERRIDE(gen12_ccs_formats),
+	}, {
+		.modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC,
+		.display_ver = { 14, 14 },
+		.plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_RC_CC,
+
+		.ccs.cc_planes = BIT(2),
+		.ccs.packed_aux_planes = BIT(1),
+
+		FORMAT_OVERRIDE(gen12_ccs_cc_formats),
+	}, {
 		.modifier = I915_FORMAT_MOD_4_TILED_DG2_MC_CCS,
 		.display_ver = { 13, 13 },
 		.plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_MC,
@@ -370,6 +396,14 @@ static bool plane_has_modifier(struct drm_i915_private *i915,
 	if (!plane_caps_contain_all(plane_caps, md->plane_caps))
 		return false;
 
+	/*
+	 * Separate AuxCCS and Flat CCS modifiers to be run only on platforms
+	 * where supported.
+	 */
+	if (intel_fb_is_ccs_modifier(md->modifier) &&
+	    HAS_FLAT_CCS(i915) != !md->ccs.packed_aux_planes)
+		return false;
+
 	return true;
 }
 
@@ -489,7 +523,7 @@ static bool intel_fb_is_gen12_ccs_aux_plane(const struct drm_framebuffer *fb, in
 {
 	const struct intel_modifier_desc *md = lookup_modifier(fb->modifier);
 
-	return check_modifier_display_ver_range(md, 12, 13) &&
+	return check_modifier_display_ver_range(md, 12, 14) &&
 	       ccs_aux_plane_mask(md, fb->format) & BIT(color_plane);
 }
 
@@ -605,6 +639,9 @@ intel_tile_width_bytes(const struct drm_framebuffer *fb, int color_plane)
 		if (intel_fb_is_ccs_aux_plane(fb, color_plane))
 			return 128;
 		fallthrough;
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS:
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC:
+	case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS:
@@ -791,6 +828,9 @@ unsigned int intel_surf_alignment(const struct drm_framebuffer *fb,
 	case I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC:
+	case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS:
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS:
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC:
 		return 16 * 1024;
 	case I915_FORMAT_MOD_Y_TILED_CCS:
 	case I915_FORMAT_MOD_Yf_TILED_CCS:
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 8ea0598a5a07e..f6f760e59c9ef 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -789,6 +789,14 @@ static u32 skl_plane_ctl_tiling(u64 fb_modifier)
 			PLANE_CTL_CLEAR_COLOR_DISABLE;
 	case I915_FORMAT_MOD_4_TILED_DG2_RC_CCS_CC:
 		return PLANE_CTL_TILED_4 | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE;
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS:
+		return PLANE_CTL_TILED_4 |
+			PLANE_CTL_RENDER_DECOMPRESSION_ENABLE |
+			PLANE_CTL_CLEAR_COLOR_DISABLE;
+	case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC:
+		return PLANE_CTL_TILED_4 | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE;
+	case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS:
+		return PLANE_CTL_TILED_4 | PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE;
 	case I915_FORMAT_MOD_Y_TILED_CCS:
 	case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC:
 		return PLANE_CTL_TILED_Y | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE;
@@ -2160,6 +2168,11 @@ skl_plane_disable_flip_done(struct intel_plane *plane)
 static bool skl_plane_has_rc_ccs(struct drm_i915_private *i915,
 				 enum pipe pipe, enum plane_id plane_id)
 {
+	/* Wa_14017240301 */
+	if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_A0, STEP_B0) ||
+	    IS_MTL_GRAPHICS_STEP(i915, P, STEP_A0, STEP_B0))
+		return false;
+
 	/* Wa_22011186057 */
 	if (IS_ADLP_DISPLAY_STEP(i915, STEP_A0, STEP_B0))
 		return false;
@@ -2441,12 +2454,17 @@ skl_get_initial_plane_config(struct intel_crtc *crtc,
 	case PLANE_CTL_TILED_Y:
 		plane_config->tiling = I915_TILING_Y;
 		if (val & PLANE_CTL_RENDER_DECOMPRESSION_ENABLE)
-			if (DISPLAY_VER(dev_priv) >= 12)
+			if (DISPLAY_VER(dev_priv) >= 14)
+				fb->modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS;
+			else if (DISPLAY_VER(dev_priv) >= 12)
 				fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS;
 			else
 				fb->modifier = I915_FORMAT_MOD_Y_TILED_CCS;
 		else if (val & PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE)
-			fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS;
+			if (DISPLAY_VER(dev_priv) >= 14)
+				fb->modifier = I915_FORMAT_MOD_4_TILED_MTL_MC_CCS;
+			else
+				fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS;
 		else
 			fb->modifier = I915_FORMAT_MOD_Y_TILED;
 		break;
-- 
GitLab


From a44bebf6afa8114c1bf0f69f5623b7394737a0c8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 12 May 2023 14:04:41 +0300
Subject: [PATCH 0493/1791] drm/i915/dpll: drop a useless I915_STATE_WARN_ON()

In general, we don't do assertions that a function gets called on the
right platforms, and if we did, it should not be a state warn.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512110444.1448231-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index ed372d227aa73..936b8de9e439b 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -464,8 +464,6 @@ static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv)
 	u32 val;
 	bool enabled;
 
-	I915_STATE_WARN_ON(!(HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv)));
-
 	val = intel_de_read(dev_priv, PCH_DREF_CONTROL);
 	enabled = !!(val & (DREF_SSC_SOURCE_MASK | DREF_NONSPREAD_SOURCE_MASK |
 			    DREF_SUPERSPREAD_SOURCE_MASK));
-- 
GitLab


From fe735c34ef06a1d08218569370875967e337608f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 12 May 2023 14:04:42 +0300
Subject: [PATCH 0494/1791] drm/i915/crtc: replace I915_STATE_WARN_ON() with
 I915_STATE_WARN()

Describe the assertion better.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512110444.1448231-2-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_crtc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index df7d05f1e14b3..1e3f88d006094 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -35,7 +35,9 @@
 
 static void assert_vblank_disabled(struct drm_crtc *crtc)
 {
-	if (I915_STATE_WARN_ON(drm_crtc_vblank_get(crtc) == 0))
+	if (I915_STATE_WARN(drm_crtc_vblank_get(crtc) == 0,
+			    "[CRTC:%d:%s] vblank assertion failure (expected off, current on)\n",
+			    crtc->base.id, crtc->name))
 		drm_crtc_vblank_put(crtc);
 }
 
-- 
GitLab


From b8e6185bcf5cf41c0195d496bbfe48e5f0ffb3a6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 12 May 2023 14:04:43 +0300
Subject: [PATCH 0495/1791] drm/i915/display: remove I915_STATE_WARN_ON()

Remove the unused macro.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512110444.1448231-3-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.h | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index ac95961f68ba7..8f451aaf5760d 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -538,12 +538,12 @@ void assert_transcoder(struct drm_i915_private *dev_priv,
 #define assert_transcoder_enabled(d, t) assert_transcoder(d, t, true)
 #define assert_transcoder_disabled(d, t) assert_transcoder(d, t, false)
 
-/* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and
- * WARN_ON()) for hw state sanity checks to check for unexpected conditions
- * which may not necessarily be a user visible problem.  This will either
- * WARN() or DRM_ERROR() depending on the verbose_checks moduleparam, to
- * enable distros and users to tailor their preferred amount of i915 abrt
- * spam.
+/*
+ * Use I915_STATE_WARN(x) (rather than WARN() and WARN_ON()) for hw state sanity
+ * checks to check for unexpected conditions which may not necessarily be a user
+ * visible problem. This will either WARN() or DRM_ERROR() depending on the
+ * verbose_state_checks module param, to enable distros and users to tailor
+ * their preferred amount of i915 abrt spam.
  */
 #define I915_STATE_WARN(condition, format...) ({			\
 	int __ret_warn_on = !!(condition);				\
@@ -553,9 +553,6 @@ void assert_transcoder(struct drm_i915_private *dev_priv,
 	unlikely(__ret_warn_on);					\
 })
 
-#define I915_STATE_WARN_ON(x)						\
-	I915_STATE_WARN((x), "%s", "WARN_ON(" __stringify(x) ")")
-
 bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915);
 
 #endif
-- 
GitLab


From 6b9bd7c35d32be464591ba704a80ab2db4f2ea4a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 12 May 2023 21:16:58 +0300
Subject: [PATCH 0496/1791] drm/i915/display: add i915 parameter to
 I915_STATE_WARN()

Add i915 parameter to I915_STATE_WARN() and use device based logging.

Done using cocci + hand edited where there was no i915 local variable
ready.

v2: avoid null deref in verify_connector_state()

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512181658.1735594-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/g4x_dp.c         |  4 +-
 drivers/gpu/drm/i915/display/intel_crtc.c     |  4 +-
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 10 ++---
 drivers/gpu/drm/i915/display/intel_display.c  |  9 +++--
 drivers/gpu/drm/i915/display/intel_display.h  |  7 ++--
 .../drm/i915/display/intel_display_power.c    | 37 ++++++++++++-------
 drivers/gpu/drm/i915/display/intel_dpll.c     |  2 +-
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c | 30 ++++++++-------
 drivers/gpu/drm/i915/display/intel_fdi.c      |  9 +++--
 .../drm/i915/display/intel_modeset_verify.c   | 34 +++++++++--------
 .../gpu/drm/i915/display/intel_pch_display.c  | 20 +++++-----
 drivers/gpu/drm/i915/display/intel_pps.c      |  7 ++--
 drivers/gpu/drm/i915/display/intel_snps_phy.c |  2 +-
 drivers/gpu/drm/i915/display/vlv_dsi_pll.c    |  2 +-
 14 files changed, 99 insertions(+), 78 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c
index 920d570f75943..112d91d81fdc4 100644
--- a/drivers/gpu/drm/i915/display/g4x_dp.c
+++ b/drivers/gpu/drm/i915/display/g4x_dp.c
@@ -169,7 +169,7 @@ static void assert_dp_port(struct intel_dp *intel_dp, bool state)
 	struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev);
 	bool cur_state = intel_de_read(dev_priv, intel_dp->output_reg) & DP_PORT_EN;
 
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"[ENCODER:%d:%s] state assertion failure (expected %s, current %s)\n",
 			dig_port->base.base.base.id, dig_port->base.base.name,
 			str_on_off(state), str_on_off(cur_state));
@@ -180,7 +180,7 @@ static void assert_edp_pll(struct drm_i915_private *dev_priv, bool state)
 {
 	bool cur_state = intel_de_read(dev_priv, DP_A) & DP_PLL_ENABLE;
 
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"eDP PLL state assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index 1e3f88d006094..ecae9bf05269b 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -35,7 +35,9 @@
 
 static void assert_vblank_disabled(struct drm_crtc *crtc)
 {
-	if (I915_STATE_WARN(drm_crtc_vblank_get(crtc) == 0,
+	struct drm_i915_private *i915 = to_i915(crtc->dev);
+
+	if (I915_STATE_WARN(i915, drm_crtc_vblank_get(crtc) == 0,
 			    "[CRTC:%d:%s] vblank assertion failure (expected off, current on)\n",
 			    crtc->base.id, crtc->name))
 		drm_crtc_vblank_put(crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index d94127e7448b1..ef0615cdc8a0c 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2945,18 +2945,18 @@ void intel_c10pll_state_verify(struct intel_atomic_state *state,
 	for (i = 0; i < ARRAY_SIZE(mpllb_sw_state->pll); i++) {
 		u8 expected = mpllb_sw_state->pll[i];
 
-		I915_STATE_WARN(mpllb_hw_state.pll[i] != expected,
+		I915_STATE_WARN(i915, mpllb_hw_state.pll[i] != expected,
 				"[CRTC:%d:%s] mismatch in C10MPLLB: Register[%d] (expected 0x%02x, found 0x%02x)",
-				crtc->base.base.id, crtc->base.name,
-				i, expected, mpllb_hw_state.pll[i]);
+				crtc->base.base.id, crtc->base.name, i,
+				expected, mpllb_hw_state.pll[i]);
 	}
 
-	I915_STATE_WARN(mpllb_hw_state.tx != mpllb_sw_state->tx,
+	I915_STATE_WARN(i915, mpllb_hw_state.tx != mpllb_sw_state->tx,
 			"[CRTC:%d:%s] mismatch in C10MPLLB: Register TX0 (expected 0x%02x, found 0x%02x)",
 			crtc->base.base.id, crtc->base.name,
 			mpllb_sw_state->tx, mpllb_hw_state.tx);
 
-	I915_STATE_WARN(mpllb_hw_state.cmn != mpllb_sw_state->cmn,
+	I915_STATE_WARN(i915, mpllb_hw_state.cmn != mpllb_sw_state->cmn,
 			"[CRTC:%d:%s] mismatch in C10MPLLB: Register CMN0 (expected 0x%02x, found 0x%02x)",
 			crtc->base.base.id, crtc->base.name,
 			mpllb_sw_state->cmn, mpllb_hw_state.cmn);
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 1d5d42a408035..4b70b389e0cb1 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -322,20 +322,21 @@ void assert_transcoder(struct drm_i915_private *dev_priv,
 		cur_state = false;
 	}
 
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"transcoder %s assertion failure (expected %s, current %s)\n",
-			transcoder_name(cpu_transcoder),
-			str_on_off(state), str_on_off(cur_state));
+			transcoder_name(cpu_transcoder), str_on_off(state),
+			str_on_off(cur_state));
 }
 
 static void assert_plane(struct intel_plane *plane, bool state)
 {
+	struct drm_i915_private *i915 = to_i915(plane->base.dev);
 	enum pipe pipe;
 	bool cur_state;
 
 	cur_state = plane->get_hw_state(plane, &pipe);
 
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(i915, cur_state != state,
 			"%s assertion failure (expected %s, current %s)\n",
 			plane->base.name, str_on_off(state),
 			str_on_off(cur_state));
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 8f451aaf5760d..ee3def6e14a8a 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -545,11 +545,12 @@ void assert_transcoder(struct drm_i915_private *dev_priv,
  * verbose_state_checks module param, to enable distros and users to tailor
  * their preferred amount of i915 abrt spam.
  */
-#define I915_STATE_WARN(condition, format...) ({			\
+#define I915_STATE_WARN(__i915, condition, format...) ({		\
+	struct drm_device *drm = &(__i915)->drm;			\
 	int __ret_warn_on = !!(condition);				\
 	if (unlikely(__ret_warn_on))					\
-		if (!WARN(i915_modparams.verbose_state_checks, format))	\
-			DRM_ERROR(format);				\
+		if (!drm_WARN(drm, i915_modparams.verbose_state_checks, format)) \
+			drm_err(drm, format);				\
 	unlikely(__ret_warn_on);					\
 })
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 5150069f3f821..6ed2ece89c3fd 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -1165,31 +1165,39 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	struct intel_crtc *crtc;
 
 	for_each_intel_crtc(&dev_priv->drm, crtc)
-		I915_STATE_WARN(crtc->active, "CRTC for pipe %c enabled\n",
+		I915_STATE_WARN(dev_priv, crtc->active,
+				"CRTC for pipe %c enabled\n",
 				pipe_name(crtc->pipe));
 
-	I915_STATE_WARN(intel_de_read(dev_priv, HSW_PWR_WELL_CTL2),
+	I915_STATE_WARN(dev_priv, intel_de_read(dev_priv, HSW_PWR_WELL_CTL2),
 			"Display power well on\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, SPLL_CTL) & SPLL_PLL_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, SPLL_CTL) & SPLL_PLL_ENABLE,
 			"SPLL enabled\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, WRPLL_CTL(0)) & WRPLL_PLL_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, WRPLL_CTL(0)) & WRPLL_PLL_ENABLE,
 			"WRPLL1 enabled\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, WRPLL_CTL(1)) & WRPLL_PLL_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, WRPLL_CTL(1)) & WRPLL_PLL_ENABLE,
 			"WRPLL2 enabled\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, PP_STATUS(0)) & PP_ON,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, PP_STATUS(0)) & PP_ON,
 			"Panel power on\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE,
 			"CPU PWM1 enabled\n");
 	if (IS_HASWELL(dev_priv))
-		I915_STATE_WARN(intel_de_read(dev_priv, HSW_BLC_PWM2_CTL) & BLM_PWM_ENABLE,
+		I915_STATE_WARN(dev_priv,
+				intel_de_read(dev_priv, HSW_BLC_PWM2_CTL) & BLM_PWM_ENABLE,
 				"CPU PWM2 enabled\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, BLC_PWM_PCH_CTL1) & BLM_PCH_PWM_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, BLC_PWM_PCH_CTL1) & BLM_PCH_PWM_ENABLE,
 			"PCH PWM1 enabled\n");
-	I915_STATE_WARN((intel_de_read(dev_priv, UTIL_PIN_CTL) &
-			 (UTIL_PIN_ENABLE | UTIL_PIN_MODE_MASK)) ==
-			(UTIL_PIN_ENABLE | UTIL_PIN_MODE_PWM),
+	I915_STATE_WARN(dev_priv,
+			(intel_de_read(dev_priv, UTIL_PIN_CTL) & (UTIL_PIN_ENABLE | UTIL_PIN_MODE_MASK)) == (UTIL_PIN_ENABLE | UTIL_PIN_MODE_PWM),
 			"Utility pin enabled in PWM mode\n");
-	I915_STATE_WARN(intel_de_read(dev_priv, PCH_GTC_CTL) & PCH_GTC_ENABLE,
+	I915_STATE_WARN(dev_priv,
+			intel_de_read(dev_priv, PCH_GTC_CTL) & PCH_GTC_ENABLE,
 			"PCH GTC enabled\n");
 
 	/*
@@ -1198,7 +1206,8 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	 * gen-specific and since we only disable LCPLL after we fully disable
 	 * the interrupts, the check below should be enough.
 	 */
-	I915_STATE_WARN(intel_irqs_enabled(dev_priv), "IRQs enabled\n");
+	I915_STATE_WARN(dev_priv, intel_irqs_enabled(dev_priv),
+			"IRQs enabled\n");
 }
 
 static u32 hsw_read_dcomp(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c
index ca0f362a40e32..824be7f037246 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll.c
@@ -2080,7 +2080,7 @@ static void assert_pll(struct drm_i915_private *dev_priv,
 	bool cur_state;
 
 	cur_state = intel_de_read(dev_priv, DPLL(pipe)) & DPLL_VCO_ENABLE;
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"PLL state assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index 936b8de9e439b..ee55ab3095684 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -169,8 +169,8 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv,
 		return;
 
 	cur_state = intel_dpll_get_hw_state(dev_priv, pll, &hw_state);
-	I915_STATE_WARN(cur_state != state,
-	     "%s assertion failure (expected %s, current %s)\n",
+	I915_STATE_WARN(dev_priv, cur_state != state,
+			"%s assertion failure (expected %s, current %s)\n",
 			pll->info->name, str_on_off(state),
 			str_on_off(cur_state));
 }
@@ -467,7 +467,8 @@ static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv)
 	val = intel_de_read(dev_priv, PCH_DREF_CONTROL);
 	enabled = !!(val & (DREF_SSC_SOURCE_MASK | DREF_NONSPREAD_SOURCE_MASK |
 			    DREF_SUPERSPREAD_SOURCE_MASK));
-	I915_STATE_WARN(!enabled, "PCH refclk assertion failure, should be active but is disabled\n");
+	I915_STATE_WARN(dev_priv, !enabled,
+			"PCH refclk assertion failure, should be active but is disabled\n");
 }
 
 static void ibx_pch_dpll_enable(struct drm_i915_private *dev_priv,
@@ -4405,17 +4406,18 @@ verify_single_dpll_state(struct drm_i915_private *dev_priv,
 	active = intel_dpll_get_hw_state(dev_priv, pll, &dpll_hw_state);
 
 	if (!(pll->info->flags & INTEL_DPLL_ALWAYS_ON)) {
-		I915_STATE_WARN(!pll->on && pll->active_mask,
+		I915_STATE_WARN(dev_priv, !pll->on && pll->active_mask,
 				"pll in active use but not on in sw tracking\n");
-		I915_STATE_WARN(pll->on && !pll->active_mask,
+		I915_STATE_WARN(dev_priv, pll->on && !pll->active_mask,
 				"pll is on but not used by any active pipe\n");
-		I915_STATE_WARN(pll->on != active,
+		I915_STATE_WARN(dev_priv, pll->on != active,
 				"pll on state mismatch (expected %i, found %i)\n",
 				pll->on, active);
 	}
 
 	if (!crtc) {
-		I915_STATE_WARN(pll->active_mask & ~pll->state.pipe_mask,
+		I915_STATE_WARN(dev_priv,
+				pll->active_mask & ~pll->state.pipe_mask,
 				"more active pll users than references: 0x%x vs 0x%x\n",
 				pll->active_mask, pll->state.pipe_mask);
 
@@ -4425,20 +4427,20 @@ verify_single_dpll_state(struct drm_i915_private *dev_priv,
 	pipe_mask = BIT(crtc->pipe);
 
 	if (new_crtc_state->hw.active)
-		I915_STATE_WARN(!(pll->active_mask & pipe_mask),
+		I915_STATE_WARN(dev_priv, !(pll->active_mask & pipe_mask),
 				"pll active mismatch (expected pipe %c in active mask 0x%x)\n",
 				pipe_name(crtc->pipe), pll->active_mask);
 	else
-		I915_STATE_WARN(pll->active_mask & pipe_mask,
+		I915_STATE_WARN(dev_priv, pll->active_mask & pipe_mask,
 				"pll active mismatch (didn't expect pipe %c in active mask 0x%x)\n",
 				pipe_name(crtc->pipe), pll->active_mask);
 
-	I915_STATE_WARN(!(pll->state.pipe_mask & pipe_mask),
+	I915_STATE_WARN(dev_priv, !(pll->state.pipe_mask & pipe_mask),
 			"pll enabled crtcs mismatch (expected 0x%x in 0x%x)\n",
 			pipe_mask, pll->state.pipe_mask);
 
-	I915_STATE_WARN(pll->on && memcmp(&pll->state.hw_state,
-					  &dpll_hw_state,
+	I915_STATE_WARN(dev_priv,
+			pll->on && memcmp(&pll->state.hw_state, &dpll_hw_state,
 					  sizeof(dpll_hw_state)),
 			"pll hw state mismatch\n");
 }
@@ -4458,10 +4460,10 @@ void intel_shared_dpll_state_verify(struct intel_crtc *crtc,
 		u8 pipe_mask = BIT(crtc->pipe);
 		struct intel_shared_dpll *pll = old_crtc_state->shared_dpll;
 
-		I915_STATE_WARN(pll->active_mask & pipe_mask,
+		I915_STATE_WARN(dev_priv, pll->active_mask & pipe_mask,
 				"pll active mismatch (didn't expect pipe %c in active mask (0x%x))\n",
 				pipe_name(crtc->pipe), pll->active_mask);
-		I915_STATE_WARN(pll->state.pipe_mask & pipe_mask,
+		I915_STATE_WARN(dev_priv, pll->state.pipe_mask & pipe_mask,
 				"pll enabled crtcs mismatch (found %x in enabled mask (0x%x))\n",
 				pipe_name(crtc->pipe), pll->state.pipe_mask);
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_fdi.c b/drivers/gpu/drm/i915/display/intel_fdi.c
index 55283677c45a4..e12b46a84fa11 100644
--- a/drivers/gpu/drm/i915/display/intel_fdi.c
+++ b/drivers/gpu/drm/i915/display/intel_fdi.c
@@ -36,7 +36,7 @@ static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 	} else {
 		cur_state = intel_de_read(dev_priv, FDI_TX_CTL(pipe)) & FDI_TX_ENABLE;
 	}
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"FDI TX state assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
@@ -57,7 +57,7 @@ static void assert_fdi_rx(struct drm_i915_private *dev_priv,
 	bool cur_state;
 
 	cur_state = intel_de_read(dev_priv, FDI_RX_CTL(pipe)) & FDI_RX_ENABLE;
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(dev_priv, cur_state != state,
 			"FDI RX state assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
@@ -86,7 +86,8 @@ void assert_fdi_tx_pll_enabled(struct drm_i915_private *i915,
 		return;
 
 	cur_state = intel_de_read(i915, FDI_TX_CTL(pipe)) & FDI_TX_PLL_ENABLE;
-	I915_STATE_WARN(!cur_state, "FDI TX PLL assertion failure, should be active but is disabled\n");
+	I915_STATE_WARN(i915, !cur_state,
+			"FDI TX PLL assertion failure, should be active but is disabled\n");
 }
 
 static void assert_fdi_rx_pll(struct drm_i915_private *i915,
@@ -95,7 +96,7 @@ static void assert_fdi_rx_pll(struct drm_i915_private *i915,
 	bool cur_state;
 
 	cur_state = intel_de_read(i915, FDI_RX_CTL(pipe)) & FDI_RX_PLL_ENABLE;
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(i915, cur_state != state,
 			"FDI RX PLL assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_verify.c b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
index 5e0ec15d9fd5e..138144a65a457 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_verify.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_verify.c
@@ -35,27 +35,28 @@ static void intel_connector_verify_state(struct intel_crtc_state *crtc_state,
 	if (connector->get_hw_state(connector)) {
 		struct intel_encoder *encoder = intel_attached_encoder(connector);
 
-		I915_STATE_WARN(!crtc_state,
+		I915_STATE_WARN(i915, !crtc_state,
 				"connector enabled without attached crtc\n");
 
 		if (!crtc_state)
 			return;
 
-		I915_STATE_WARN(!crtc_state->hw.active,
+		I915_STATE_WARN(i915, !crtc_state->hw.active,
 				"connector is active, but attached crtc isn't\n");
 
 		if (!encoder || encoder->type == INTEL_OUTPUT_DP_MST)
 			return;
 
-		I915_STATE_WARN(conn_state->best_encoder != &encoder->base,
+		I915_STATE_WARN(i915,
+				conn_state->best_encoder != &encoder->base,
 				"atomic encoder doesn't match attached encoder\n");
 
-		I915_STATE_WARN(conn_state->crtc != encoder->base.crtc,
+		I915_STATE_WARN(i915, conn_state->crtc != encoder->base.crtc,
 				"attached encoder crtc differs from connector crtc\n");
 	} else {
-		I915_STATE_WARN(crtc_state && crtc_state->hw.active,
+		I915_STATE_WARN(i915, crtc_state && crtc_state->hw.active,
 				"attached crtc is active, but connector isn't\n");
-		I915_STATE_WARN(!crtc_state && conn_state->best_encoder,
+		I915_STATE_WARN(i915, !crtc_state && conn_state->best_encoder,
 				"best encoder set without crtc!\n");
 	}
 }
@@ -80,7 +81,7 @@ verify_connector_state(struct intel_atomic_state *state,
 
 		intel_connector_verify_state(crtc_state, new_conn_state);
 
-		I915_STATE_WARN(new_conn_state->best_encoder != encoder,
+		I915_STATE_WARN(to_i915(connector->dev), new_conn_state->best_encoder != encoder,
 				"connector's atomic encoder doesn't match legacy encoder\n");
 	}
 }
@@ -131,15 +132,15 @@ verify_encoder_state(struct drm_i915_private *dev_priv, struct intel_atomic_stat
 			found = true;
 			enabled = true;
 
-			I915_STATE_WARN(new_conn_state->crtc !=
-					encoder->base.crtc,
+			I915_STATE_WARN(dev_priv,
+					new_conn_state->crtc != encoder->base.crtc,
 					"connector's crtc doesn't match encoder crtc\n");
 		}
 
 		if (!found)
 			continue;
 
-		I915_STATE_WARN(!!encoder->base.crtc != enabled,
+		I915_STATE_WARN(dev_priv, !!encoder->base.crtc != enabled,
 				"encoder's enabled state mismatch (expected %i, found %i)\n",
 				!!encoder->base.crtc, enabled);
 
@@ -147,7 +148,7 @@ verify_encoder_state(struct drm_i915_private *dev_priv, struct intel_atomic_stat
 			bool active;
 
 			active = encoder->get_hw_state(encoder, &pipe);
-			I915_STATE_WARN(active,
+			I915_STATE_WARN(dev_priv, active,
 					"encoder detached but still enabled on pipe %c.\n",
 					pipe_name(pipe));
 		}
@@ -182,11 +183,12 @@ verify_crtc_state(struct intel_crtc *crtc,
 	if (IS_I830(dev_priv) && pipe_config->hw.active)
 		pipe_config->hw.active = new_crtc_state->hw.active;
 
-	I915_STATE_WARN(new_crtc_state->hw.active != pipe_config->hw.active,
+	I915_STATE_WARN(dev_priv,
+			new_crtc_state->hw.active != pipe_config->hw.active,
 			"crtc active state doesn't match with hw state (expected %i, found %i)\n",
 			new_crtc_state->hw.active, pipe_config->hw.active);
 
-	I915_STATE_WARN(crtc->active != new_crtc_state->hw.active,
+	I915_STATE_WARN(dev_priv, crtc->active != new_crtc_state->hw.active,
 			"transitional active state does not match atomic hw state (expected %i, found %i)\n",
 			new_crtc_state->hw.active, crtc->active);
 
@@ -197,12 +199,12 @@ verify_crtc_state(struct intel_crtc *crtc,
 		bool active;
 
 		active = encoder->get_hw_state(encoder, &pipe);
-		I915_STATE_WARN(active != new_crtc_state->hw.active,
+		I915_STATE_WARN(dev_priv, active != new_crtc_state->hw.active,
 				"[ENCODER:%i] active %i with crtc active %i\n",
 				encoder->base.base.id, active,
 				new_crtc_state->hw.active);
 
-		I915_STATE_WARN(active && master_crtc->pipe != pipe,
+		I915_STATE_WARN(dev_priv, active && master_crtc->pipe != pipe,
 				"Encoder connected to wrong pipe %c\n",
 				pipe_name(pipe));
 
@@ -217,7 +219,7 @@ verify_crtc_state(struct intel_crtc *crtc,
 
 	if (!intel_pipe_config_compare(new_crtc_state,
 				       pipe_config, false)) {
-		I915_STATE_WARN(1, "pipe state doesn't match!\n");
+		I915_STATE_WARN(dev_priv, 1, "pipe state doesn't match!\n");
 		intel_crtc_state_dump(pipe_config, NULL, "hw state");
 		intel_crtc_state_dump(new_crtc_state, NULL, "sw state");
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.c b/drivers/gpu/drm/i915/display/intel_pch_display.c
index 2411fe4dee8b2..866786e6b32f6 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_display.c
+++ b/drivers/gpu/drm/i915/display/intel_pch_display.c
@@ -43,11 +43,12 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv,
 
 	state = g4x_dp_port_enabled(dev_priv, dp_reg, port, &port_pipe);
 
-	I915_STATE_WARN(state && port_pipe == pipe,
+	I915_STATE_WARN(dev_priv, state && port_pipe == pipe,
 			"PCH DP %c enabled on transcoder %c, should be disabled\n",
 			port_name(port), pipe_name(pipe));
 
-	I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
+	I915_STATE_WARN(dev_priv,
+			HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
 			"IBX PCH DP %c still using transcoder B\n",
 			port_name(port));
 }
@@ -61,11 +62,12 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
 
 	state = intel_sdvo_port_enabled(dev_priv, hdmi_reg, &port_pipe);
 
-	I915_STATE_WARN(state && port_pipe == pipe,
+	I915_STATE_WARN(dev_priv, state && port_pipe == pipe,
 			"PCH HDMI %c enabled on transcoder %c, should be disabled\n",
 			port_name(port), pipe_name(pipe));
 
-	I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
+	I915_STATE_WARN(dev_priv,
+			HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B,
 			"IBX PCH HDMI %c still using transcoder B\n",
 			port_name(port));
 }
@@ -79,13 +81,13 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 	assert_pch_dp_disabled(dev_priv, pipe, PORT_C, PCH_DP_C);
 	assert_pch_dp_disabled(dev_priv, pipe, PORT_D, PCH_DP_D);
 
-	I915_STATE_WARN(intel_crt_port_enabled(dev_priv, PCH_ADPA, &port_pipe) &&
-			port_pipe == pipe,
+	I915_STATE_WARN(dev_priv,
+			intel_crt_port_enabled(dev_priv, PCH_ADPA, &port_pipe) && port_pipe == pipe,
 			"PCH VGA enabled on transcoder %c, should be disabled\n",
 			pipe_name(pipe));
 
-	I915_STATE_WARN(intel_lvds_port_enabled(dev_priv, PCH_LVDS, &port_pipe) &&
-			port_pipe == pipe,
+	I915_STATE_WARN(dev_priv,
+			intel_lvds_port_enabled(dev_priv, PCH_LVDS, &port_pipe) && port_pipe == pipe,
 			"PCH LVDS enabled on transcoder %c, should be disabled\n",
 			pipe_name(pipe));
 
@@ -103,7 +105,7 @@ static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
 
 	val = intel_de_read(dev_priv, PCH_TRANSCONF(pipe));
 	enabled = !!(val & TRANS_ENABLE);
-	I915_STATE_WARN(enabled,
+	I915_STATE_WARN(dev_priv, enabled,
 			"transcoder assertion failed, should be off on pipe %c but is still active\n",
 			pipe_name(pipe));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index 4f0b0cca03cca..5e7ba594e7e74 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -787,7 +787,7 @@ void intel_pps_vdd_on(struct intel_dp *intel_dp)
 	vdd = false;
 	with_intel_pps_lock(intel_dp, wakeref)
 		vdd = intel_pps_vdd_on_unlocked(intel_dp);
-	I915_STATE_WARN(!vdd, "[ENCODER:%d:%s] %s VDD already requested on\n",
+	I915_STATE_WARN(i915, !vdd, "[ENCODER:%d:%s] %s VDD already requested on\n",
 			dp_to_dig_port(intel_dp)->base.base.base.id,
 			dp_to_dig_port(intel_dp)->base.base.name,
 			pps_name(i915, &intel_dp->pps));
@@ -899,7 +899,8 @@ void intel_pps_vdd_off_unlocked(struct intel_dp *intel_dp, bool sync)
 	if (!intel_dp_is_edp(intel_dp))
 		return;
 
-	I915_STATE_WARN(!intel_dp->pps.want_panel_vdd, "[ENCODER:%d:%s] %s VDD not forced on",
+	I915_STATE_WARN(dev_priv, !intel_dp->pps.want_panel_vdd,
+			"[ENCODER:%d:%s] %s VDD not forced on",
 			dp_to_dig_port(intel_dp)->base.base.base.id,
 			dp_to_dig_port(intel_dp)->base.base.name,
 			pps_name(dev_priv, &intel_dp->pps));
@@ -1721,7 +1722,7 @@ void assert_pps_unlocked(struct drm_i915_private *dev_priv, enum pipe pipe)
 	    ((val & PANEL_UNLOCK_MASK) == PANEL_UNLOCK_REGS))
 		locked = false;
 
-	I915_STATE_WARN(panel_pipe == pipe && locked,
+	I915_STATE_WARN(dev_priv, panel_pipe == pipe && locked,
 			"panel assertion failure, pipe %c regs locked\n",
 			pipe_name(pipe));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy.c b/drivers/gpu/drm/i915/display/intel_snps_phy.c
index a72677bf617b6..88ef56b6e0fd0 100644
--- a/drivers/gpu/drm/i915/display/intel_snps_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_snps_phy.c
@@ -2016,7 +2016,7 @@ void intel_mpllb_state_verify(struct intel_atomic_state *state,
 	intel_mpllb_readout_hw_state(encoder, &mpllb_hw_state);
 
 #define MPLLB_CHECK(__name)						\
-	I915_STATE_WARN(mpllb_sw_state->__name != mpllb_hw_state.__name,	\
+	I915_STATE_WARN(i915, mpllb_sw_state->__name != mpllb_hw_state.__name, \
 			"[CRTC:%d:%s] mismatch in MPLLB: %s (expected 0x%08x, found 0x%08x)", \
 			crtc->base.base.id, crtc->base.name,		\
 			__stringify(__name),				\
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index b697badbbe711..ae0a0b11bae3b 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -598,7 +598,7 @@ static void assert_dsi_pll(struct drm_i915_private *i915, bool state)
 	cur_state = vlv_cck_read(i915, CCK_REG_DSI_PLL_CONTROL) & DSI_PLL_VCO_EN;
 	vlv_cck_put(i915);
 
-	I915_STATE_WARN(cur_state != state,
+	I915_STATE_WARN(i915, cur_state != state,
 			"DSI PLL state assertion failure (expected %s, current %s)\n",
 			str_on_off(state), str_on_off(cur_state));
 }
-- 
GitLab


From 621b6783c73100067c844c3be0b254c2f350e8ec Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Fri, 12 May 2023 13:37:35 -0700
Subject: [PATCH 0497/1791] drm/i915/hwmon: Silence UBSAN uninitialized bool
 variable warning

Loading i915 on UBSAN enabled kernels (CONFIG_UBSAN/CONFIG_UBSAN_BOOL)
causes the following warning:

  UBSAN: invalid-load in drivers/gpu/drm/i915/gt/uc/intel_uc.c:558:2
  load of value 255 is not a valid value for type '_Bool'
  Call Trace:
   dump_stack_lvl+0x57/0x7d
   ubsan_epilogue+0x5/0x40
   __ubsan_handle_load_invalid_value.cold+0x43/0x48
   __uc_init_hw+0x76a/0x903 [i915]
   ...
   i915_driver_probe+0xfb1/0x1eb0 [i915]
   i915_pci_probe+0xbe/0x2d0 [i915]

The warning happens because during probe i915_hwmon is still not available
which results in the output boolean variable *old remaining
uninitialized. Silence the warning by initializing the variable to an
arbitrary value.

v2: Move variable initialization to the declaration (Andi)

Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512203735.2635237-1-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index 1381943b8973d..c8b9cbb7ba3a9 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -465,7 +465,7 @@ static int __uc_init_hw(struct intel_uc *uc)
 	struct intel_guc *guc = &uc->guc;
 	struct intel_huc *huc = &uc->huc;
 	int ret, attempts;
-	bool pl1en;
+	bool pl1en = false;
 
 	GEM_BUG_ON(!intel_uc_supports_guc(uc));
 	GEM_BUG_ON(!intel_uc_wants_guc(uc));
-- 
GitLab


From 60aebc9559492cea6a9625f514a8041717e3a2e4 Mon Sep 17 00:00:00 2001
From: Huacai Chen <chenhuacai@loongson.cn>
Date: Mon, 4 Jul 2022 09:17:04 +0800
Subject: [PATCH 0498/1791] drivers/firmware: Move sysfb_init() from
 device_initcall to subsys_initcall_sync

Consider a configuration like this:
1, efifb (or simpledrm) is built-in;
2, a native display driver (such as radeon) is also built-in.

As Javier said, this is not a common configuration (the native display
driver is usually built as a module), but it can happen and cause some
trouble.

In this case, since efifb, radeon and sysfb are all in device_initcall()
level, the order in practise is like this:

efifb registered at first, but no "efi-framebuffer" device yet. radeon
registered later, and /dev/fb0 created. sysfb_init() comes at last, it
registers "efi-framebuffer" and then causes an error message "efifb: a
framebuffer is already registered". Make sysfb_init() to be subsys_
initcall_sync() can avoid this. And Javier Martinez Canillas is trying
to make a more general solution in commit 873eb3b11860 ("fbdev: Disable
sysfb device registration when removing conflicting FBs").

However, this patch still makes sense because it can make the screen
display as early as possible (We cannot move to subsys_initcall, since
sysfb_init() should be executed after PCI enumeration).

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220704011704.1418055-1-chenhuacai@loongson.cn
---
 drivers/firmware/sysfb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/firmware/sysfb.c b/drivers/firmware/sysfb.c
index 3c197db42c9d9..82fcfd29bc4d2 100644
--- a/drivers/firmware/sysfb.c
+++ b/drivers/firmware/sysfb.c
@@ -128,4 +128,4 @@ static __init int sysfb_init(void)
 }
 
 /* must execute after PCI subsystem for EFI quirks */
-device_initcall(sysfb_init);
+subsys_initcall_sync(sysfb_init);
-- 
GitLab


From 0f1cb4d777281ca3360dbc8959befc488e0c327e Mon Sep 17 00:00:00 2001
From: Javier Martinez Canillas <javierm@redhat.com>
Date: Fri, 12 May 2023 14:02:31 +0200
Subject: [PATCH 0499/1791] drm/ssd130x: Fix include guard name

This is a leftover from an early iteration of the driver when it was still
named ssd1307 instead of ssd130x. Change it for consistency with the rest.

Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512120232.304603-1-javierm@redhat.com
---
 drivers/gpu/drm/solomon/ssd130x.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/solomon/ssd130x.h b/drivers/gpu/drm/solomon/ssd130x.h
index 03038c1b6476b..db03ee5db3921 100644
--- a/drivers/gpu/drm/solomon/ssd130x.h
+++ b/drivers/gpu/drm/solomon/ssd130x.h
@@ -10,8 +10,8 @@
  * Copyright 2012 Free Electrons
  */
 
-#ifndef __SSD1307X_H__
-#define __SSD1307X_H__
+#ifndef __SSD130X_H__
+#define __SSD130X_H__
 
 #include <drm/drm_connector.h>
 #include <drm/drm_crtc.h>
@@ -94,4 +94,4 @@ struct ssd130x_device *ssd130x_probe(struct device *dev, struct regmap *regmap);
 void ssd130x_remove(struct ssd130x_device *ssd130x);
 void ssd130x_shutdown(struct ssd130x_device *ssd130x);
 
-#endif /* __SSD1307X_H__ */
+#endif /* __SSD130X_H__ */
-- 
GitLab


From fe7f4e8d496552f880d7368b482d2ccac33780b7 Mon Sep 17 00:00:00 2001
From: Shuijing Li <shuijing.li@mediatek.com>
Date: Mon, 15 May 2023 17:49:54 +0800
Subject: [PATCH 0500/1791] drm/panel: boe-tv101wum-nl6: Remove extra delay

Reduce the delay after LCM reset by removing an extra delay in the
initialization commands array. The required delay of at least 6ms after
reset is guaranteed by boe_panel_prepare().

Signed-off-by: Shuijing Li <shuijing.li@mediatek.com>
Signed-off-by: Xinlei Lee <xinlei.lee@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515094955.15982-2-shuijing.li@mediatek.com
---
 drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index 783234ae0f576..a5652d38acdaa 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -780,7 +780,6 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 };
 
 static const struct panel_init_cmd boe_init_cmd[] = {
-	_INIT_DELAY_CMD(24),
 	_INIT_DCS_CMD(0xB0, 0x05),
 	_INIT_DCS_CMD(0xB1, 0xE5),
 	_INIT_DCS_CMD(0xB3, 0x52),
-- 
GitLab


From 812562b8d881ce6d33fed8052b3a10b718430fb5 Mon Sep 17 00:00:00 2001
From: Shuijing Li <shuijing.li@mediatek.com>
Date: Mon, 15 May 2023 17:49:55 +0800
Subject: [PATCH 0501/1791] drm/panel: boe-tv101wum-nl6: Fine tune the panel
 power sequence

For "boe,tv105wum-nw0" this special panel, it is stipulated in
the panel spec that MIPI needs to keep the LP11 state before
the lcm_reset pin is pulled high.

Signed-off-by: Shuijing Li <shuijing.li@mediatek.com>
Signed-off-by: Xinlei Lee <xinlei.lee@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515094955.15982-3-shuijing.li@mediatek.com
---
 drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index a5652d38acdaa..64e462f49c603 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -36,6 +36,7 @@ struct panel_desc {
 	const struct panel_init_cmd *init_cmds;
 	unsigned int lanes;
 	bool discharge_on_disable;
+	bool lp11_before_reset;
 };
 
 struct boe_panel {
@@ -1365,6 +1366,10 @@ static int boe_panel_prepare(struct drm_panel *panel)
 
 	usleep_range(10000, 11000);
 
+	if (boe->desc->lp11_before_reset) {
+		mipi_dsi_dcs_nop(boe->dsi);
+		usleep_range(1000, 2000);
+	}
 	gpiod_set_value(boe->enable_gpio, 1);
 	usleep_range(1000, 2000);
 	gpiod_set_value(boe->enable_gpio, 0);
@@ -1591,6 +1596,7 @@ static const struct panel_desc boe_tv105wum_nw0_desc = {
 	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
 		      MIPI_DSI_MODE_LPM,
 	.init_cmds = boe_init_cmd,
+	.lp11_before_reset = true,
 };
 
 static const struct drm_display_mode starry_qfh032011_53g_default_mode = {
-- 
GitLab


From 8b25320887d7feac98875546ea0f521628b745bb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Fri, 12 May 2023 07:40:44 -0300
Subject: [PATCH 0502/1791] drm: Add fixed-point helper to get rounded integer
 values
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Create a new fixed-point helper to allow us to return the rounded value
of our fixed point value.

[v2]:
    * Create the function drm_fixp2int_round() (Melissa Wen).
[v3]:
    * Use drm_fixp2int() instead of shifting manually (Arthur Grillo).

Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512104044.65034-1-mcanal@igalia.com
---
 include/drm/drm_fixed.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/drm/drm_fixed.h b/include/drm/drm_fixed.h
index 255645c1f9a89..6ea339d5de088 100644
--- a/include/drm/drm_fixed.h
+++ b/include/drm/drm_fixed.h
@@ -71,6 +71,7 @@ static inline u32 dfixed_div(fixed20_12 A, fixed20_12 B)
 }
 
 #define DRM_FIXED_POINT		32
+#define DRM_FIXED_POINT_HALF	16
 #define DRM_FIXED_ONE		(1ULL << DRM_FIXED_POINT)
 #define DRM_FIXED_DECIMAL_MASK	(DRM_FIXED_ONE - 1)
 #define DRM_FIXED_DIGITS_MASK	(~DRM_FIXED_DECIMAL_MASK)
@@ -87,6 +88,11 @@ static inline int drm_fixp2int(s64 a)
 	return ((s64)a) >> DRM_FIXED_POINT;
 }
 
+static inline int drm_fixp2int_round(s64 a)
+{
+	return drm_fixp2int(a + (1 << (DRM_FIXED_POINT_HALF - 1)));
+}
+
 static inline int drm_fixp2int_ceil(s64 a)
 {
 	if (a > 0)
-- 
GitLab


From ab87f558dcfb2562c3497e89600dec798a446665 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Fri, 12 May 2023 07:40:45 -0300
Subject: [PATCH 0503/1791] drm/vkms: Fix RGB565 pixel conversion
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, the pixel conversion isn't rounding the fixed-point values
before assigning it to the RGB coefficients, which is causing the IGT
pixel-format tests to fail. So, use the drm_fixp2int_round() fixed-point
helper to round the values when assigning it to the RGB coefficients.

Tested with igt@kms_plane@pixel-format and igt@kms_plane@pixel-format-source-clamping.

[v2]:
    * Use drm_fixp2int_round() to fix the pixel conversion instead of
      casting the values to s32 (Melissa Wen).

Fixes: 89b03aeaef16 ("drm/vkms: fix 32bit compilation error by replacing macros")
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Arthur Grillo <arthurgrillo@riseup.net>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512104044.65034-2-mcanal@igalia.com
---
 drivers/gpu/drm/vkms/vkms_formats.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index ebacb8efa0553..5945da0beba6f 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -106,9 +106,9 @@ static void RGB565_to_argb_u16(u8 *src_pixels, struct pixel_argb_u16 *out_pixel)
 	s64 fp_b = drm_int2fixp(rgb_565 & 0x1f);
 
 	out_pixel->a = (u16)0xffff;
-	out_pixel->r = drm_fixp2int(drm_fixp_mul(fp_r, fp_rb_ratio));
-	out_pixel->g = drm_fixp2int(drm_fixp_mul(fp_g, fp_g_ratio));
-	out_pixel->b = drm_fixp2int(drm_fixp_mul(fp_b, fp_rb_ratio));
+	out_pixel->r = drm_fixp2int_round(drm_fixp_mul(fp_r, fp_rb_ratio));
+	out_pixel->g = drm_fixp2int_round(drm_fixp_mul(fp_g, fp_g_ratio));
+	out_pixel->b = drm_fixp2int_round(drm_fixp_mul(fp_b, fp_rb_ratio));
 }
 
 void vkms_compose_row(struct line_buffer *stage_buffer, struct vkms_plane_state *plane, int y)
@@ -232,9 +232,9 @@ static void argb_u16_to_RGB565(struct vkms_frame_info *frame_info,
 		s64 fp_g = drm_int2fixp(in_pixels[x].g);
 		s64 fp_b = drm_int2fixp(in_pixels[x].b);
 
-		u16 r = drm_fixp2int(drm_fixp_div(fp_r, fp_rb_ratio));
-		u16 g = drm_fixp2int(drm_fixp_div(fp_g, fp_g_ratio));
-		u16 b = drm_fixp2int(drm_fixp_div(fp_b, fp_rb_ratio));
+		u16 r = drm_fixp2int_round(drm_fixp_div(fp_r, fp_rb_ratio));
+		u16 g = drm_fixp2int_round(drm_fixp_div(fp_g, fp_g_ratio));
+		u16 b = drm_fixp2int_round(drm_fixp_div(fp_b, fp_rb_ratio));
 
 		*dst_pixels = cpu_to_le16(r << 11 | g << 5 | b);
 	}
-- 
GitLab


From 08d8f43045af8589671d4462190d27765653cdf8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 15 May 2023 13:17:36 +0300
Subject: [PATCH 0504/1791] drm/i915/irq: convert gen8_de_irq_handler() to void

The return value is not used for anything.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515101738.2399816-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/i915_irq.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index e54090a680299..fe8bef6ecdae7 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2124,10 +2124,8 @@ static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_i
 		intel_de_write(i915, PICAINTERRUPT_IER, pica_ier);
 }
 
-static irqreturn_t
-gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
+static void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 {
-	irqreturn_t ret = IRQ_NONE;
 	u32 iir;
 	enum pipe pipe;
 
@@ -2137,7 +2135,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_MISC_IIR);
 		if (iir) {
 			intel_uncore_write(&dev_priv->uncore, GEN8_DE_MISC_IIR, iir);
-			ret = IRQ_HANDLED;
 			gen8_de_misc_irq_handler(dev_priv, iir);
 		} else {
 			drm_err_ratelimited(&dev_priv->drm,
@@ -2149,7 +2146,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 		iir = intel_uncore_read(&dev_priv->uncore, GEN11_DE_HPD_IIR);
 		if (iir) {
 			intel_uncore_write(&dev_priv->uncore, GEN11_DE_HPD_IIR, iir);
-			ret = IRQ_HANDLED;
 			gen11_hpd_irq_handler(dev_priv, iir);
 		} else {
 			drm_err_ratelimited(&dev_priv->drm,
@@ -2163,7 +2159,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 			bool found = false;
 
 			intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IIR, iir);
-			ret = IRQ_HANDLED;
 
 			if (iir & gen8_de_port_aux_mask(dev_priv)) {
 				intel_dp_aux_irq_handler(dev_priv);
@@ -2223,7 +2218,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 			continue;
 		}
 
-		ret = IRQ_HANDLED;
 		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe), iir);
 
 		if (iir & GEN8_PIPE_VBLANK)
@@ -2257,8 +2251,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 		 */
 		gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir);
 		if (iir) {
-			ret = IRQ_HANDLED;
-
 			if (pica_iir)
 				xelpdp_pica_irq_handler(dev_priv, pica_iir);
 
@@ -2277,8 +2269,6 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
 				"The master control interrupt lied (SDE)!\n");
 		}
 	}
-
-	return ret;
 }
 
 static inline u32 gen8_master_intr_disable(void __iomem * const regs)
-- 
GitLab


From da38ba98645d789ddda2a584d40e2de00139e98b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 15 May 2023 13:17:37 +0300
Subject: [PATCH 0505/1791] drm/i915/irq: split out hotplug irq handling

Split hotplug irq handling out of i915_irq.[ch] into
display/intel_hotplug_irq.[ch].

The line between the new intel_hotplug_irq.[ch] and the existing
intel_hotplug.[ch] needs further clarification, but the first step is to
move the stuff out of i915_irq.[ch].

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515101738.2399816-2-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |    1 +
 drivers/gpu/drm/i915/display/intel_crt.c      |    1 +
 drivers/gpu/drm/i915/display/intel_dp.c       |    1 +
 drivers/gpu/drm/i915/display/intel_hotplug.c  |    1 +
 .../gpu/drm/i915/display/intel_hotplug_irq.c  | 1442 ++++++++++++++++
 .../gpu/drm/i915/display/intel_hotplug_irq.h  |   35 +
 drivers/gpu/drm/i915/i915_irq.c               | 1504 +----------------
 drivers/gpu/drm/i915/i915_irq.h               |   12 +-
 8 files changed, 1525 insertions(+), 1472 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_hotplug_irq.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_hotplug_irq.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 9af76e376ca91..cc80d483fd6fd 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -260,6 +260,7 @@ i915-y += \
 	display/intel_hdcp.o \
 	display/intel_hdcp_gsc.o \
 	display/intel_hotplug.o \
+	display/intel_hotplug_irq.o \
 	display/intel_hti.o \
 	display/intel_load_detect.o \
 	display/intel_lpe_audio.o \
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index f0f4897b3c3c6..673c036466960 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -48,6 +48,7 @@
 #include "intel_fifo_underrun.h"
 #include "intel_gmbus.h"
 #include "intel_hotplug.h"
+#include "intel_hotplug_irq.h"
 #include "intel_load_detect.h"
 #include "intel_pch_display.h"
 #include "intel_pch_refclk.h"
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 0cc57681dc4d4..b35ab251f5435 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -68,6 +68,7 @@
 #include "intel_hdcp.h"
 #include "intel_hdmi.h"
 #include "intel_hotplug.h"
+#include "intel_hotplug_irq.h"
 #include "intel_lspcon.h"
 #include "intel_lvds.h"
 #include "intel_panel.h"
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index b12900446828a..23a5e1a875f13 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -27,6 +27,7 @@
 #include "i915_irq.h"
 #include "intel_display_types.h"
 #include "intel_hotplug.h"
+#include "intel_hotplug_irq.h"
 
 /**
  * DOC: Hotplug
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
new file mode 100644
index 0000000000000..1d7ae49e073ee
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
@@ -0,0 +1,1442 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include "i915_drv.h"
+#include "i915_irq.h"
+#include "i915_reg.h"
+#include "intel_de.h"
+#include "intel_display_types.h"
+#include "intel_dp_aux.h"
+#include "intel_gmbus.h"
+#include "intel_hotplug.h"
+#include "intel_hotplug_irq.h"
+
+typedef bool (*long_pulse_detect_func)(enum hpd_pin pin, u32 val);
+typedef u32 (*hotplug_enables_func)(struct intel_encoder *encoder);
+typedef u32 (*hotplug_mask_func)(enum hpd_pin pin);
+
+static const u32 hpd_ilk[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = DE_DP_A_HOTPLUG,
+};
+
+static const u32 hpd_ivb[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB,
+};
+
+static const u32 hpd_bdw[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A),
+};
+
+static const u32 hpd_ibx[HPD_NUM_PINS] = {
+	[HPD_CRT] = SDE_CRT_HOTPLUG,
+	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG,
+	[HPD_PORT_B] = SDE_PORTB_HOTPLUG,
+	[HPD_PORT_C] = SDE_PORTC_HOTPLUG,
+	[HPD_PORT_D] = SDE_PORTD_HOTPLUG,
+};
+
+static const u32 hpd_cpt[HPD_NUM_PINS] = {
+	[HPD_CRT] = SDE_CRT_HOTPLUG_CPT,
+	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG_CPT,
+	[HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT,
+	[HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT,
+	[HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT,
+};
+
+static const u32 hpd_spt[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_PORTA_HOTPLUG_SPT,
+	[HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT,
+	[HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT,
+	[HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT,
+	[HPD_PORT_E] = SDE_PORTE_HOTPLUG_SPT,
+};
+
+static const u32 hpd_mask_i915[HPD_NUM_PINS] = {
+	[HPD_CRT] = CRT_HOTPLUG_INT_EN,
+	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_EN,
+	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_EN,
+	[HPD_PORT_B] = PORTB_HOTPLUG_INT_EN,
+	[HPD_PORT_C] = PORTC_HOTPLUG_INT_EN,
+	[HPD_PORT_D] = PORTD_HOTPLUG_INT_EN,
+};
+
+static const u32 hpd_status_g4x[HPD_NUM_PINS] = {
+	[HPD_CRT] = CRT_HOTPLUG_INT_STATUS,
+	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_G4X,
+	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_G4X,
+	[HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS,
+	[HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS,
+	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS,
+};
+
+static const u32 hpd_status_i915[HPD_NUM_PINS] = {
+	[HPD_CRT] = CRT_HOTPLUG_INT_STATUS,
+	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_I915,
+	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_I915,
+	[HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS,
+	[HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS,
+	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS,
+};
+
+static const u32 hpd_bxt[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A),
+	[HPD_PORT_B] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_B),
+	[HPD_PORT_C] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_C),
+};
+
+static const u32 hpd_gen11[HPD_NUM_PINS] = {
+	[HPD_PORT_TC1] = GEN11_TC_HOTPLUG(HPD_PORT_TC1) | GEN11_TBT_HOTPLUG(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = GEN11_TC_HOTPLUG(HPD_PORT_TC2) | GEN11_TBT_HOTPLUG(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = GEN11_TC_HOTPLUG(HPD_PORT_TC3) | GEN11_TBT_HOTPLUG(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = GEN11_TC_HOTPLUG(HPD_PORT_TC4) | GEN11_TBT_HOTPLUG(HPD_PORT_TC4),
+	[HPD_PORT_TC5] = GEN11_TC_HOTPLUG(HPD_PORT_TC5) | GEN11_TBT_HOTPLUG(HPD_PORT_TC5),
+	[HPD_PORT_TC6] = GEN11_TC_HOTPLUG(HPD_PORT_TC6) | GEN11_TBT_HOTPLUG(HPD_PORT_TC6),
+};
+
+static const u32 hpd_xelpdp[HPD_NUM_PINS] = {
+	[HPD_PORT_TC1] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC1) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC2) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC3) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC4) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC4),
+};
+
+static const u32 hpd_icp[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
+	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
+	[HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C),
+	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
+	[HPD_PORT_TC5] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC5),
+	[HPD_PORT_TC6] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC6),
+};
+
+static const u32 hpd_sde_dg1[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
+	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
+	[HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C),
+	[HPD_PORT_D] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_D),
+	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_DG2(HPD_PORT_TC1),
+};
+
+static const u32 hpd_mtp[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
+	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
+	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1),
+	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2),
+	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3),
+	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
+};
+
+static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
+{
+	struct intel_hotplug *hpd = &dev_priv->display.hotplug;
+
+	if (HAS_GMCH(dev_priv)) {
+		if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) ||
+		    IS_CHERRYVIEW(dev_priv))
+			hpd->hpd = hpd_status_g4x;
+		else
+			hpd->hpd = hpd_status_i915;
+		return;
+	}
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		hpd->hpd = hpd_xelpdp;
+	else if (DISPLAY_VER(dev_priv) >= 11)
+		hpd->hpd = hpd_gen11;
+	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+		hpd->hpd = hpd_bxt;
+	else if (DISPLAY_VER(dev_priv) == 9)
+		hpd->hpd = NULL; /* no north HPD on SKL */
+	else if (DISPLAY_VER(dev_priv) >= 8)
+		hpd->hpd = hpd_bdw;
+	else if (DISPLAY_VER(dev_priv) >= 7)
+		hpd->hpd = hpd_ivb;
+	else
+		hpd->hpd = hpd_ilk;
+
+	if ((INTEL_PCH_TYPE(dev_priv) < PCH_DG1) &&
+	    (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv)))
+		return;
+
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
+		hpd->pch_hpd = hpd_sde_dg1;
+	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP)
+		hpd->pch_hpd = hpd_mtp;
+	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+		hpd->pch_hpd = hpd_icp;
+	else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv))
+		hpd->pch_hpd = hpd_spt;
+	else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_CPT(dev_priv))
+		hpd->pch_hpd = hpd_cpt;
+	else if (HAS_PCH_IBX(dev_priv))
+		hpd->pch_hpd = hpd_ibx;
+	else
+		MISSING_CASE(INTEL_PCH_TYPE(dev_priv));
+}
+
+/* For display hotplug interrupt */
+void i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv,
+					  u32 mask, u32 bits)
+{
+	lockdep_assert_held(&dev_priv->irq_lock);
+	drm_WARN_ON(&dev_priv->drm, bits & ~mask);
+
+	intel_uncore_rmw(&dev_priv->uncore, PORT_HOTPLUG_EN, mask, bits);
+}
+
+/**
+ * i915_hotplug_interrupt_update - update hotplug interrupt enable
+ * @dev_priv: driver private
+ * @mask: bits to update
+ * @bits: bits to enable
+ * NOTE: the HPD enable bits are modified both inside and outside
+ * of an interrupt context. To avoid that read-modify-write cycles
+ * interfer, these bits are protected by a spinlock. Since this
+ * function is usually not called from a context where the lock is
+ * held already, this function acquires the lock itself. A non-locking
+ * version is also available.
+ */
+void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
+				   u32 mask,
+				   u32 bits)
+{
+	spin_lock_irq(&dev_priv->irq_lock);
+	i915_hotplug_interrupt_update_locked(dev_priv, mask, bits);
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
+static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+	case HPD_PORT_TC5:
+	case HPD_PORT_TC6:
+		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(pin);
+	default:
+		return false;
+	}
+}
+
+static bool bxt_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_A:
+		return val & PORTA_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_B:
+		return val & PORTB_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_C:
+		return val & PORTC_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
+static bool icp_ddi_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_A:
+	case HPD_PORT_B:
+	case HPD_PORT_C:
+	case HPD_PORT_D:
+		return val & SHOTPLUG_CTL_DDI_HPD_LONG_DETECT(pin);
+	default:
+		return false;
+	}
+}
+
+static bool icp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+	case HPD_PORT_TC5:
+	case HPD_PORT_TC6:
+		return val & ICP_TC_HPD_LONG_DETECT(pin);
+	default:
+		return false;
+	}
+}
+
+static bool spt_port_hotplug2_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_E:
+		return val & PORTE_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
+static bool spt_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_A:
+		return val & PORTA_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_B:
+		return val & PORTB_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_C:
+		return val & PORTC_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_D:
+		return val & PORTD_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
+static bool ilk_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_A:
+		return val & DIGITAL_PORTA_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
+static bool pch_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_B:
+		return val & PORTB_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_C:
+		return val & PORTC_HOTPLUG_LONG_DETECT;
+	case HPD_PORT_D:
+		return val & PORTD_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
+static bool i9xx_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
+{
+	switch (pin) {
+	case HPD_PORT_B:
+		return val & PORTB_HOTPLUG_INT_LONG_PULSE;
+	case HPD_PORT_C:
+		return val & PORTC_HOTPLUG_INT_LONG_PULSE;
+	case HPD_PORT_D:
+		return val & PORTD_HOTPLUG_INT_LONG_PULSE;
+	default:
+		return false;
+	}
+}
+
+/*
+ * Get a bit mask of pins that have triggered, and which ones may be long.
+ * This can be called multiple times with the same masks to accumulate
+ * hotplug detection results from several registers.
+ *
+ * Note that the caller is expected to zero out the masks initially.
+ */
+static void intel_get_hpd_pins(struct drm_i915_private *dev_priv,
+			       u32 *pin_mask, u32 *long_mask,
+			       u32 hotplug_trigger, u32 dig_hotplug_reg,
+			       const u32 hpd[HPD_NUM_PINS],
+			       bool long_pulse_detect(enum hpd_pin pin, u32 val))
+{
+	enum hpd_pin pin;
+
+	BUILD_BUG_ON(BITS_PER_TYPE(*pin_mask) < HPD_NUM_PINS);
+
+	for_each_hpd_pin(pin) {
+		if ((hpd[pin] & hotplug_trigger) == 0)
+			continue;
+
+		*pin_mask |= BIT(pin);
+
+		if (long_pulse_detect(pin, dig_hotplug_reg))
+			*long_mask |= BIT(pin);
+	}
+
+	drm_dbg(&dev_priv->drm,
+		"hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n",
+		hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask);
+}
+
+static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv,
+				  const u32 hpd[HPD_NUM_PINS])
+{
+	struct intel_encoder *encoder;
+	u32 enabled_irqs = 0;
+
+	for_each_intel_encoder(&dev_priv->drm, encoder)
+		if (dev_priv->display.hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED)
+			enabled_irqs |= hpd[encoder->hpd_pin];
+
+	return enabled_irqs;
+}
+
+static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv,
+				  const u32 hpd[HPD_NUM_PINS])
+{
+	struct intel_encoder *encoder;
+	u32 hotplug_irqs = 0;
+
+	for_each_intel_encoder(&dev_priv->drm, encoder)
+		hotplug_irqs |= hpd[encoder->hpd_pin];
+
+	return hotplug_irqs;
+}
+
+static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915,
+				  hotplug_mask_func hotplug_mask)
+{
+	enum hpd_pin pin;
+	u32 hotplug = 0;
+
+	for_each_hpd_pin(pin)
+		hotplug |= hotplug_mask(pin);
+
+	return hotplug;
+}
+
+static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915,
+				     hotplug_enables_func hotplug_enables)
+{
+	struct intel_encoder *encoder;
+	u32 hotplug = 0;
+
+	for_each_intel_encoder(&i915->drm, encoder)
+		hotplug |= hotplug_enables(encoder);
+
+	return hotplug;
+}
+
+u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_status = 0, hotplug_status_mask;
+	int i;
+
+	if (IS_G4X(dev_priv) ||
+	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		hotplug_status_mask = HOTPLUG_INT_STATUS_G4X |
+			DP_AUX_CHANNEL_MASK_INT_STATUS_G4X;
+	else
+		hotplug_status_mask = HOTPLUG_INT_STATUS_I915;
+
+	/*
+	 * We absolutely have to clear all the pending interrupt
+	 * bits in PORT_HOTPLUG_STAT. Otherwise the ISR port
+	 * interrupt bit won't have an edge, and the i965/g4x
+	 * edge triggered IIR will not notice that an interrupt
+	 * is still pending. We can't use PORT_HOTPLUG_EN to
+	 * guarantee the edge as the act of toggling the enable
+	 * bits can itself generate a new hotplug interrupt :(
+	 */
+	for (i = 0; i < 10; i++) {
+		u32 tmp = intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT) & hotplug_status_mask;
+
+		if (tmp == 0)
+			return hotplug_status;
+
+		hotplug_status |= tmp;
+		intel_uncore_write(&dev_priv->uncore, PORT_HOTPLUG_STAT, hotplug_status);
+	}
+
+	drm_WARN_ONCE(&dev_priv->drm, 1,
+		      "PORT_HOTPLUG_STAT did not clear (0x%08x)\n",
+		      intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT));
+
+	return hotplug_status;
+}
+
+void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_status)
+{
+	u32 pin_mask = 0, long_mask = 0;
+	u32 hotplug_trigger;
+
+	if (IS_G4X(dev_priv) ||
+	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
+	else
+		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
+
+	if (hotplug_trigger) {
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   hotplug_trigger, hotplug_trigger,
+				   dev_priv->display.hotplug.hpd,
+				   i9xx_port_hotplug_long_detect);
+
+		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+	}
+
+	if ((IS_G4X(dev_priv) ||
+	     IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
+		intel_dp_aux_irq_handler(dev_priv);
+}
+
+void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+{
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+	/*
+	 * Somehow the PCH doesn't seem to really ack the interrupt to the CPU
+	 * unless we touch the hotplug register, even if hotplug_trigger is
+	 * zero. Not acking leads to "The master control interrupt lied (SDE)!"
+	 * errors.
+	 */
+	dig_hotplug_reg = intel_uncore_read(&dev_priv->uncore, PCH_PORT_HOTPLUG);
+	if (!hotplug_trigger) {
+		u32 mask = PORTA_HOTPLUG_STATUS_MASK |
+			PORTD_HOTPLUG_STATUS_MASK |
+			PORTC_HOTPLUG_STATUS_MASK |
+			PORTB_HOTPLUG_STATUS_MASK;
+		dig_hotplug_reg &= ~mask;
+	}
+
+	intel_uncore_write(&dev_priv->uncore, PCH_PORT_HOTPLUG, dig_hotplug_reg);
+	if (!hotplug_trigger)
+		return;
+
+	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+			   hotplug_trigger, dig_hotplug_reg,
+			   dev_priv->display.hotplug.pch_hpd,
+			   pch_port_hotplug_long_detect);
+
+	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+}
+
+void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir)
+{
+	enum hpd_pin pin;
+	u32 hotplug_trigger = iir & (XELPDP_DP_ALT_HOTPLUG_MASK | XELPDP_TBT_HOTPLUG_MASK);
+	u32 trigger_aux = iir & XELPDP_AUX_TC_MASK;
+	u32 pin_mask = 0, long_mask = 0;
+
+	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
+		u32 val;
+
+		if (!(i915->display.hotplug.hpd[pin] & hotplug_trigger))
+			continue;
+
+		pin_mask |= BIT(pin);
+
+		val = intel_de_read(i915, XELPDP_PORT_HOTPLUG_CTL(pin));
+		intel_de_write(i915, XELPDP_PORT_HOTPLUG_CTL(pin), val);
+
+		if (val & (XELPDP_DP_ALT_HPD_LONG_DETECT | XELPDP_TBT_HPD_LONG_DETECT))
+			long_mask |= BIT(pin);
+	}
+
+	if (pin_mask) {
+		drm_dbg(&i915->drm,
+			"pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n",
+			hotplug_trigger, pin_mask, long_mask);
+
+		intel_hpd_irq_handler(i915, pin_mask, long_mask);
+	}
+
+	if (trigger_aux)
+		intel_dp_aux_irq_handler(i915);
+
+	if (!pin_mask && !trigger_aux)
+		drm_err(&i915->drm,
+			"Unexpected DE HPD/AUX interrupt 0x%08x\n", iir);
+}
+
+void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+{
+	u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP;
+	u32 tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_MASK_ICP;
+	u32 pin_mask = 0, long_mask = 0;
+
+	if (ddi_hotplug_trigger) {
+		u32 dig_hotplug_reg;
+
+		/* Locking due to DSI native GPIO sequences */
+		spin_lock(&dev_priv->irq_lock);
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, 0, 0);
+		spin_unlock(&dev_priv->irq_lock);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   ddi_hotplug_trigger, dig_hotplug_reg,
+				   dev_priv->display.hotplug.pch_hpd,
+				   icp_ddi_port_hotplug_long_detect);
+	}
+
+	if (tc_hotplug_trigger) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, 0, 0);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   tc_hotplug_trigger, dig_hotplug_reg,
+				   dev_priv->display.hotplug.pch_hpd,
+				   icp_tc_port_hotplug_long_detect);
+	}
+
+	if (pin_mask)
+		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+
+	if (pch_iir & SDE_GMBUS_ICP)
+		intel_gmbus_irq_handler(dev_priv);
+}
+
+void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+{
+	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT &
+		~SDE_PORTE_HOTPLUG_SPT;
+	u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT;
+	u32 pin_mask = 0, long_mask = 0;
+
+	if (hotplug_trigger) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   hotplug_trigger, dig_hotplug_reg,
+				   dev_priv->display.hotplug.pch_hpd,
+				   spt_port_hotplug_long_detect);
+	}
+
+	if (hotplug2_trigger) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, 0, 0);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   hotplug2_trigger, dig_hotplug_reg,
+				   dev_priv->display.hotplug.pch_hpd,
+				   spt_port_hotplug2_long_detect);
+	}
+
+	if (pin_mask)
+		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+
+	if (pch_iir & SDE_GMBUS_CPT)
+		intel_gmbus_irq_handler(dev_priv);
+}
+
+void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+{
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0);
+
+	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+			   hotplug_trigger, dig_hotplug_reg,
+			   dev_priv->display.hotplug.hpd,
+			   ilk_port_hotplug_long_detect);
+
+	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+}
+
+void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger)
+{
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
+
+	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+			   hotplug_trigger, dig_hotplug_reg,
+			   dev_priv->display.hotplug.hpd,
+			   bxt_port_hotplug_long_detect);
+
+	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+}
+
+void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
+{
+	u32 pin_mask = 0, long_mask = 0;
+	u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK;
+	u32 trigger_tbt = iir & GEN11_DE_TBT_HOTPLUG_MASK;
+
+	if (trigger_tc) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, 0, 0);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   trigger_tc, dig_hotplug_reg,
+				   dev_priv->display.hotplug.hpd,
+				   gen11_port_hotplug_long_detect);
+	}
+
+	if (trigger_tbt) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, 0, 0);
+
+		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
+				   trigger_tbt, dig_hotplug_reg,
+				   dev_priv->display.hotplug.hpd,
+				   gen11_port_hotplug_long_detect);
+	}
+
+	if (pin_mask)
+		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
+	else
+		drm_err(&dev_priv->drm,
+			"Unexpected DE HPD interrupt 0x%08x\n", iir);
+}
+
+static u32 ibx_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return PORTA_HOTPLUG_ENABLE;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_MASK;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_MASK;
+	case HPD_PORT_D:
+		return PORTD_HOTPLUG_ENABLE | PORTD_PULSE_DURATION_MASK;
+	default:
+		return 0;
+	}
+}
+
+static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	switch (encoder->hpd_pin) {
+	case HPD_PORT_A:
+		/*
+		 * When CPU and PCH are on the same package, port A
+		 * HPD must be enabled in both north and south.
+		 */
+		return HAS_PCH_LPT_LP(i915) ?
+			PORTA_HOTPLUG_ENABLE : 0;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE |
+			PORTB_PULSE_DURATION_2ms;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE |
+			PORTC_PULSE_DURATION_2ms;
+	case HPD_PORT_D:
+		return PORTD_HOTPLUG_ENABLE |
+			PORTD_PULSE_DURATION_2ms;
+	default:
+		return 0;
+	}
+}
+
+static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	/*
+	 * Enable digital hotplug on the PCH, and configure the DP short pulse
+	 * duration to 2ms (which is the minimum in the Display Port spec).
+	 * The pulse duration bits are reserved on LPT+.
+	 */
+	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
+			 intel_hpd_hotplug_mask(dev_priv, ibx_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
+}
+
+static void ibx_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 ibx_hotplug_mask(encoder->hpd_pin),
+			 ibx_hotplug_enables(encoder));
+}
+
+static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+
+	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+
+	ibx_hpd_detection_setup(dev_priv);
+}
+
+static u32 icp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+	case HPD_PORT_B:
+	case HPD_PORT_C:
+	case HPD_PORT_D:
+		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder)
+{
+	return icp_ddi_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 icp_tc_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+	case HPD_PORT_TC5:
+	case HPD_PORT_TC6:
+		return ICP_TC_HPD_ENABLE(hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
+{
+	return icp_tc_hotplug_mask(encoder->hpd_pin);
+}
+
+static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI,
+			 intel_hpd_hotplug_mask(dev_priv, icp_ddi_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
+}
+
+static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_DDI,
+			 icp_ddi_hotplug_mask(encoder->hpd_pin),
+			 icp_ddi_hotplug_enables(encoder));
+}
+
+static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
+			 intel_hpd_hotplug_mask(dev_priv, icp_tc_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
+}
+
+static void icp_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_TC,
+			 icp_tc_hotplug_mask(encoder->hpd_pin),
+			 icp_tc_hotplug_enables(encoder));
+}
+
+static void icp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	icp_ddi_hpd_enable_detection(encoder);
+	icp_tc_hpd_enable_detection(encoder);
+}
+
+static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+
+	if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP)
+		intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
+
+	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+
+	icp_ddi_hpd_detection_setup(dev_priv);
+	icp_tc_hpd_detection_setup(dev_priv);
+}
+
+static u32 gen11_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+	case HPD_PORT_TC5:
+	case HPD_PORT_TC6:
+		return GEN11_HOTPLUG_CTL_ENABLE(hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 gen11_hotplug_enables(struct intel_encoder *encoder)
+{
+	return gen11_hotplug_mask(encoder->hpd_pin);
+}
+
+static void dg1_hpd_invert(struct drm_i915_private *i915)
+{
+	u32 val = (INVERT_DDIA_HPD |
+		   INVERT_DDIB_HPD |
+		   INVERT_DDIC_HPD |
+		   INVERT_DDID_HPD);
+	intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val);
+}
+
+static void dg1_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	dg1_hpd_invert(i915);
+	icp_hpd_enable_detection(encoder);
+}
+
+static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	dg1_hpd_invert(dev_priv);
+	icp_hpd_irq_setup(dev_priv);
+}
+
+static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL,
+			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
+}
+
+static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, GEN11_TC_HOTPLUG_CTL,
+			 gen11_hotplug_mask(encoder->hpd_pin),
+			 gen11_hotplug_enables(encoder));
+}
+
+static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
+			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
+}
+
+static void gen11_tbt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, GEN11_TBT_HOTPLUG_CTL,
+			 gen11_hotplug_mask(encoder->hpd_pin),
+			 gen11_hotplug_enables(encoder));
+}
+
+static void gen11_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	gen11_tc_hpd_enable_detection(encoder);
+	gen11_tbt_hpd_enable_detection(encoder);
+
+	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+		icp_hpd_enable_detection(encoder);
+}
+
+static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+
+	intel_uncore_rmw(&dev_priv->uncore, GEN11_DE_HPD_IMR, hotplug_irqs,
+			 ~enabled_irqs & hotplug_irqs);
+	intel_uncore_posting_read(&dev_priv->uncore, GEN11_DE_HPD_IMR);
+
+	gen11_tc_hpd_detection_setup(dev_priv);
+	gen11_tbt_hpd_detection_setup(dev_priv);
+
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+		icp_hpd_irq_setup(dev_priv);
+}
+
+static u32 mtp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+	case HPD_PORT_B:
+		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder)
+{
+	return mtp_ddi_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 mtp_tc_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_TC1:
+	case HPD_PORT_TC2:
+	case HPD_PORT_TC3:
+	case HPD_PORT_TC4:
+		return ICP_TC_HPD_ENABLE(hpd_pin);
+	default:
+		return 0;
+	}
+}
+
+static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
+{
+	return mtp_tc_hotplug_mask(encoder->hpd_pin);
+}
+
+static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     intel_hpd_hotplug_mask(i915, mtp_ddi_hotplug_mask),
+		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
+}
+
+static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     mtp_ddi_hotplug_mask(encoder->hpd_pin),
+		     mtp_ddi_hotplug_enables(encoder));
+}
+
+static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
+		     intel_hpd_hotplug_mask(i915, mtp_tc_hotplug_mask),
+		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
+}
+
+static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
+		     mtp_tc_hotplug_mask(encoder->hpd_pin),
+		     mtp_tc_hotplug_enables(encoder));
+}
+
+static void mtp_hpd_invert(struct drm_i915_private *i915)
+{
+	u32 val = (INVERT_DDIA_HPD |
+		   INVERT_DDIB_HPD |
+		   INVERT_DDIC_HPD |
+		   INVERT_TC1_HPD |
+		   INVERT_TC2_HPD |
+		   INVERT_TC3_HPD |
+		   INVERT_TC4_HPD |
+		   INVERT_DDID_HPD_MTP |
+		   INVERT_DDIE_HPD);
+	intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val);
+}
+
+static void mtp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	mtp_hpd_invert(i915);
+	mtp_ddi_hpd_enable_detection(encoder);
+	mtp_tc_hpd_enable_detection(encoder);
+}
+
+static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd);
+
+	intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
+
+	mtp_hpd_invert(i915);
+	ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs);
+
+	mtp_ddi_hpd_detection_setup(i915);
+	mtp_tc_hpd_detection_setup(i915);
+}
+
+static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin)
+{
+	return hpd_pin >= HPD_PORT_TC1 && hpd_pin <= HPD_PORT_TC4;
+}
+
+static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915,
+					     enum hpd_pin hpd_pin, bool enable)
+{
+	u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
+		XELPDP_DP_ALT_HOTPLUG_ENABLE;
+
+	if (!is_xelpdp_pica_hpd_pin(hpd_pin))
+		return;
+
+	intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(hpd_pin),
+		     mask, enable ? mask : 0);
+}
+
+static void xelpdp_pica_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	_xelpdp_pica_hpd_detection_setup(i915, encoder->hpd_pin, true);
+}
+
+static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
+{
+	struct intel_encoder *encoder;
+	u32 available_pins = 0;
+	enum hpd_pin pin;
+
+	BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS);
+
+	for_each_intel_encoder(&i915->drm, encoder)
+		available_pins |= BIT(encoder->hpd_pin);
+
+	for_each_hpd_pin(pin)
+		_xelpdp_pica_hpd_detection_setup(i915, pin, available_pins & BIT(pin));
+}
+
+static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	xelpdp_pica_hpd_enable_detection(encoder);
+	mtp_hpd_enable_detection(encoder);
+}
+
+static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.hpd);
+
+	intel_de_rmw(i915, PICAINTERRUPT_IMR, hotplug_irqs,
+		     ~enabled_irqs & hotplug_irqs);
+	intel_uncore_posting_read(&i915->uncore, PICAINTERRUPT_IMR);
+
+	xelpdp_pica_hpd_detection_setup(i915);
+
+	if (INTEL_PCH_TYPE(i915) >= PCH_MTP)
+		mtp_hpd_irq_setup(i915);
+}
+
+static u32 spt_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return PORTA_HOTPLUG_ENABLE;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE;
+	case HPD_PORT_D:
+		return PORTD_HOTPLUG_ENABLE;
+	default:
+		return 0;
+	}
+}
+
+static u32 spt_hotplug_enables(struct intel_encoder *encoder)
+{
+	return spt_hotplug_mask(encoder->hpd_pin);
+}
+
+static u32 spt_hotplug2_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_E:
+		return PORTE_HOTPLUG_ENABLE;
+	default:
+		return 0;
+	}
+}
+
+static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
+{
+	return spt_hotplug2_mask(encoder->hpd_pin);
+}
+
+static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	/* Display WA #1179 WaHardHangonHotPlug: cnp */
+	if (HAS_PCH_CNP(dev_priv)) {
+		intel_uncore_rmw(&dev_priv->uncore, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK,
+				 CHASSIS_CLK_REQ_DURATION(0xf));
+	}
+
+	/* Enable digital hotplug on the PCH */
+	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
+			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables));
+
+	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2,
+			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug2_mask),
+			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
+}
+
+static void spt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	/* Display WA #1179 WaHardHangonHotPlug: cnp */
+	if (HAS_PCH_CNP(i915)) {
+		intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1,
+				 CHASSIS_CLK_REQ_DURATION_MASK,
+				 CHASSIS_CLK_REQ_DURATION(0xf));
+	}
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 spt_hotplug_mask(encoder->hpd_pin),
+			 spt_hotplug_enables(encoder));
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG2,
+			 spt_hotplug2_mask(encoder->hpd_pin),
+			 spt_hotplug2_enables(encoder));
+}
+
+static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
+		intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
+
+	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+
+	spt_hpd_detection_setup(dev_priv);
+}
+
+static u32 ilk_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return DIGITAL_PORTA_HOTPLUG_ENABLE |
+			DIGITAL_PORTA_PULSE_DURATION_MASK;
+	default:
+		return 0;
+	}
+}
+
+static u32 ilk_hotplug_enables(struct intel_encoder *encoder)
+{
+	switch (encoder->hpd_pin) {
+	case HPD_PORT_A:
+		return DIGITAL_PORTA_HOTPLUG_ENABLE |
+			DIGITAL_PORTA_PULSE_DURATION_2ms;
+	default:
+		return 0;
+	}
+}
+
+static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	/*
+	 * Enable digital hotplug on the CPU, and configure the DP short pulse
+	 * duration to 2ms (which is the minimum in the Display Port spec)
+	 * The pulse duration bits are reserved on HSW+.
+	 */
+	intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
+			 intel_hpd_hotplug_mask(dev_priv, ilk_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
+}
+
+static void ilk_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
+			 ilk_hotplug_mask(encoder->hpd_pin),
+			 ilk_hotplug_enables(encoder));
+
+	ibx_hpd_enable_detection(encoder);
+}
+
+static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+
+	if (DISPLAY_VER(dev_priv) >= 8)
+		bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
+	else
+		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
+
+	ilk_hpd_detection_setup(dev_priv);
+
+	ibx_hpd_irq_setup(dev_priv);
+}
+
+static u32 bxt_hotplug_mask(enum hpd_pin hpd_pin)
+{
+	switch (hpd_pin) {
+	case HPD_PORT_A:
+		return PORTA_HOTPLUG_ENABLE | BXT_DDIA_HPD_INVERT;
+	case HPD_PORT_B:
+		return PORTB_HOTPLUG_ENABLE | BXT_DDIB_HPD_INVERT;
+	case HPD_PORT_C:
+		return PORTC_HOTPLUG_ENABLE | BXT_DDIC_HPD_INVERT;
+	default:
+		return 0;
+	}
+}
+
+static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
+{
+	u32 hotplug;
+
+	switch (encoder->hpd_pin) {
+	case HPD_PORT_A:
+		hotplug = PORTA_HOTPLUG_ENABLE;
+		if (intel_bios_encoder_hpd_invert(encoder->devdata))
+			hotplug |= BXT_DDIA_HPD_INVERT;
+		return hotplug;
+	case HPD_PORT_B:
+		hotplug = PORTB_HOTPLUG_ENABLE;
+		if (intel_bios_encoder_hpd_invert(encoder->devdata))
+			hotplug |= BXT_DDIB_HPD_INVERT;
+		return hotplug;
+	case HPD_PORT_C:
+		hotplug = PORTC_HOTPLUG_ENABLE;
+		if (intel_bios_encoder_hpd_invert(encoder->devdata))
+			hotplug |= BXT_DDIC_HPD_INVERT;
+		return hotplug;
+	default:
+		return 0;
+	}
+}
+
+static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
+{
+	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
+			 intel_hpd_hotplug_mask(dev_priv, bxt_hotplug_mask),
+			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
+}
+
+static void bxt_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
+			 bxt_hotplug_mask(encoder->hpd_pin),
+			 bxt_hotplug_enables(encoder));
+}
+
+static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_irqs, enabled_irqs;
+
+	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
+
+	bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
+
+	bxt_hpd_detection_setup(dev_priv);
+}
+
+static void i915_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	u32 hotplug_en = hpd_mask_i915[encoder->hpd_pin];
+
+	/* HPD sense and interrupt enable are one and the same */
+	i915_hotplug_interrupt_update(i915, hotplug_en, hotplug_en);
+}
+
+static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv)
+{
+	u32 hotplug_en;
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	/*
+	 * Note HDMI and DP share hotplug bits. Enable bits are the same for all
+	 * generations.
+	 */
+	hotplug_en = intel_hpd_enabled_irqs(dev_priv, hpd_mask_i915);
+	/*
+	 * Programming the CRT detection parameters tends to generate a spurious
+	 * hotplug event about three seconds later. So just do it once.
+	 */
+	if (IS_G4X(dev_priv))
+		hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
+	hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
+
+	/* Ignore TV since it's buggy */
+	i915_hotplug_interrupt_update_locked(dev_priv,
+					     HOTPLUG_INT_EN_MASK |
+					     CRT_HOTPLUG_VOLTAGE_COMPARE_MASK |
+					     CRT_HOTPLUG_ACTIVATION_PERIOD_64,
+					     hotplug_en);
+}
+
+struct intel_hotplug_funcs {
+	/* Enable HPD sense and interrupts for all present encoders */
+	void (*hpd_irq_setup)(struct drm_i915_private *i915);
+	/* Enable HPD sense for a single encoder */
+	void (*hpd_enable_detection)(struct intel_encoder *encoder);
+};
+
+#define HPD_FUNCS(platform)					 \
+static const struct intel_hotplug_funcs platform##_hpd_funcs = { \
+	.hpd_irq_setup = platform##_hpd_irq_setup,		 \
+	.hpd_enable_detection = platform##_hpd_enable_detection, \
+}
+
+HPD_FUNCS(i915);
+HPD_FUNCS(xelpdp);
+HPD_FUNCS(dg1);
+HPD_FUNCS(gen11);
+HPD_FUNCS(bxt);
+HPD_FUNCS(icp);
+HPD_FUNCS(spt);
+HPD_FUNCS(ilk);
+#undef HPD_FUNCS
+
+void intel_hpd_enable_detection(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	if (i915->display.funcs.hotplug)
+		i915->display.funcs.hotplug->hpd_enable_detection(encoder);
+}
+
+void intel_hpd_irq_setup(struct drm_i915_private *i915)
+{
+	if (i915->display_irqs_enabled && i915->display.funcs.hotplug)
+		i915->display.funcs.hotplug->hpd_irq_setup(i915);
+}
+
+void intel_hotplug_irq_init(struct drm_i915_private *i915)
+{
+	intel_hpd_init_pins(i915);
+
+	intel_hpd_init_early(i915);
+
+	if (HAS_GMCH(i915)) {
+		if (I915_HAS_HOTPLUG(i915))
+			i915->display.funcs.hotplug = &i915_hpd_funcs;
+	} else {
+		if (HAS_PCH_DG2(i915))
+			i915->display.funcs.hotplug = &icp_hpd_funcs;
+		else if (HAS_PCH_DG1(i915))
+			i915->display.funcs.hotplug = &dg1_hpd_funcs;
+		else if (DISPLAY_VER(i915) >= 14)
+			i915->display.funcs.hotplug = &xelpdp_hpd_funcs;
+		else if (DISPLAY_VER(i915) >= 11)
+			i915->display.funcs.hotplug = &gen11_hpd_funcs;
+		else if (IS_GEMINILAKE(i915) || IS_BROXTON(i915))
+			i915->display.funcs.hotplug = &bxt_hpd_funcs;
+		else if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
+			i915->display.funcs.hotplug = &icp_hpd_funcs;
+		else if (INTEL_PCH_TYPE(i915) >= PCH_SPT)
+			i915->display.funcs.hotplug = &spt_hpd_funcs;
+		else
+			i915->display.funcs.hotplug = &ilk_hpd_funcs;
+	}
+}
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.h b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h
new file mode 100644
index 0000000000000..e4db752df0967
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_HOTPLUG_IRQ_H__
+#define __INTEL_HOTPLUG_IRQ_H__
+
+#include <linux/types.h>
+
+struct drm_i915_private;
+struct intel_encoder;
+
+u32 i9xx_hpd_irq_ack(struct drm_i915_private *i915);
+
+void i9xx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_status);
+void ibx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
+void ilk_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
+void gen11_hpd_irq_handler(struct drm_i915_private *i915, u32 iir);
+void bxt_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger);
+void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir);
+void icp_irq_handler(struct drm_i915_private *i915, u32 pch_iir);
+void spt_irq_handler(struct drm_i915_private *i915, u32 pch_iir);
+
+void i915_hotplug_interrupt_update_locked(struct drm_i915_private *i915,
+					  u32 mask, u32 bits);
+void i915_hotplug_interrupt_update(struct drm_i915_private *i915,
+				   u32 mask, u32 bits);
+
+void intel_hpd_enable_detection(struct intel_encoder *encoder);
+void intel_hpd_irq_setup(struct drm_i915_private *i915);
+
+void intel_hotplug_irq_init(struct drm_i915_private *i915);
+
+#endif /* __INTEL_HOTPLUG_IRQ_H__ */
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index fe8bef6ecdae7..238f5a5a1527f 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -42,6 +42,7 @@
 #include "display/intel_fifo_underrun.h"
 #include "display/intel_gmbus.h"
 #include "display/intel_hotplug.h"
+#include "display/intel_hotplug_irq.h"
 #include "display/intel_lpe_audio.h"
 #include "display/intel_psr.h"
 #include "display/intel_psr_regs.h"
@@ -84,172 +85,6 @@ static inline void pmu_irq_stats(struct drm_i915_private *i915,
 	WRITE_ONCE(i915->pmu.irq_count, i915->pmu.irq_count + 1);
 }
 
-typedef bool (*long_pulse_detect_func)(enum hpd_pin pin, u32 val);
-typedef u32 (*hotplug_enables_func)(struct intel_encoder *encoder);
-typedef u32 (*hotplug_mask_func)(enum hpd_pin pin);
-
-static const u32 hpd_ilk[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = DE_DP_A_HOTPLUG,
-};
-
-static const u32 hpd_ivb[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB,
-};
-
-static const u32 hpd_bdw[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A),
-};
-
-static const u32 hpd_ibx[HPD_NUM_PINS] = {
-	[HPD_CRT] = SDE_CRT_HOTPLUG,
-	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG,
-	[HPD_PORT_B] = SDE_PORTB_HOTPLUG,
-	[HPD_PORT_C] = SDE_PORTC_HOTPLUG,
-	[HPD_PORT_D] = SDE_PORTD_HOTPLUG,
-};
-
-static const u32 hpd_cpt[HPD_NUM_PINS] = {
-	[HPD_CRT] = SDE_CRT_HOTPLUG_CPT,
-	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG_CPT,
-	[HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT,
-	[HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT,
-	[HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT,
-};
-
-static const u32 hpd_spt[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = SDE_PORTA_HOTPLUG_SPT,
-	[HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT,
-	[HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT,
-	[HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT,
-	[HPD_PORT_E] = SDE_PORTE_HOTPLUG_SPT,
-};
-
-static const u32 hpd_mask_i915[HPD_NUM_PINS] = {
-	[HPD_CRT] = CRT_HOTPLUG_INT_EN,
-	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_EN,
-	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_EN,
-	[HPD_PORT_B] = PORTB_HOTPLUG_INT_EN,
-	[HPD_PORT_C] = PORTC_HOTPLUG_INT_EN,
-	[HPD_PORT_D] = PORTD_HOTPLUG_INT_EN,
-};
-
-static const u32 hpd_status_g4x[HPD_NUM_PINS] = {
-	[HPD_CRT] = CRT_HOTPLUG_INT_STATUS,
-	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_G4X,
-	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_G4X,
-	[HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS,
-	[HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS,
-	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS,
-};
-
-static const u32 hpd_status_i915[HPD_NUM_PINS] = {
-	[HPD_CRT] = CRT_HOTPLUG_INT_STATUS,
-	[HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_I915,
-	[HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_I915,
-	[HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS,
-	[HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS,
-	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS,
-};
-
-static const u32 hpd_bxt[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A),
-	[HPD_PORT_B] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_B),
-	[HPD_PORT_C] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_C),
-};
-
-static const u32 hpd_gen11[HPD_NUM_PINS] = {
-	[HPD_PORT_TC1] = GEN11_TC_HOTPLUG(HPD_PORT_TC1) | GEN11_TBT_HOTPLUG(HPD_PORT_TC1),
-	[HPD_PORT_TC2] = GEN11_TC_HOTPLUG(HPD_PORT_TC2) | GEN11_TBT_HOTPLUG(HPD_PORT_TC2),
-	[HPD_PORT_TC3] = GEN11_TC_HOTPLUG(HPD_PORT_TC3) | GEN11_TBT_HOTPLUG(HPD_PORT_TC3),
-	[HPD_PORT_TC4] = GEN11_TC_HOTPLUG(HPD_PORT_TC4) | GEN11_TBT_HOTPLUG(HPD_PORT_TC4),
-	[HPD_PORT_TC5] = GEN11_TC_HOTPLUG(HPD_PORT_TC5) | GEN11_TBT_HOTPLUG(HPD_PORT_TC5),
-	[HPD_PORT_TC6] = GEN11_TC_HOTPLUG(HPD_PORT_TC6) | GEN11_TBT_HOTPLUG(HPD_PORT_TC6),
-};
-
-static const u32 hpd_xelpdp[HPD_NUM_PINS] = {
-	[HPD_PORT_TC1] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC1) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC1),
-	[HPD_PORT_TC2] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC2) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC2),
-	[HPD_PORT_TC3] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC3) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC3),
-	[HPD_PORT_TC4] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC4) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC4),
-};
-
-static const u32 hpd_icp[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
-	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
-	[HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C),
-	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1),
-	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2),
-	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3),
-	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
-	[HPD_PORT_TC5] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC5),
-	[HPD_PORT_TC6] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC6),
-};
-
-static const u32 hpd_sde_dg1[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
-	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
-	[HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C),
-	[HPD_PORT_D] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_D),
-	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_DG2(HPD_PORT_TC1),
-};
-
-static const u32 hpd_mtp[HPD_NUM_PINS] = {
-	[HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A),
-	[HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B),
-	[HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1),
-	[HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2),
-	[HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3),
-	[HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4),
-};
-
-static void intel_hpd_init_pins(struct drm_i915_private *dev_priv)
-{
-	struct intel_hotplug *hpd = &dev_priv->display.hotplug;
-
-	if (HAS_GMCH(dev_priv)) {
-		if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) ||
-		    IS_CHERRYVIEW(dev_priv))
-			hpd->hpd = hpd_status_g4x;
-		else
-			hpd->hpd = hpd_status_i915;
-		return;
-	}
-
-	if (DISPLAY_VER(dev_priv) >= 14)
-		hpd->hpd = hpd_xelpdp;
-	else if (DISPLAY_VER(dev_priv) >= 11)
-		hpd->hpd = hpd_gen11;
-	else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
-		hpd->hpd = hpd_bxt;
-	else if (DISPLAY_VER(dev_priv) == 9)
-		hpd->hpd = NULL; /* no north HPD on SKL */
-	else if (DISPLAY_VER(dev_priv) >= 8)
-		hpd->hpd = hpd_bdw;
-	else if (DISPLAY_VER(dev_priv) >= 7)
-		hpd->hpd = hpd_ivb;
-	else
-		hpd->hpd = hpd_ilk;
-
-	if ((INTEL_PCH_TYPE(dev_priv) < PCH_DG1) &&
-	    (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv)))
-		return;
-
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1)
-		hpd->pch_hpd = hpd_sde_dg1;
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP)
-		hpd->pch_hpd = hpd_mtp;
-	else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		hpd->pch_hpd = hpd_icp;
-	else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv))
-		hpd->pch_hpd = hpd_spt;
-	else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_CPT(dev_priv))
-		hpd->pch_hpd = hpd_cpt;
-	else if (HAS_PCH_IBX(dev_priv))
-		hpd->pch_hpd = hpd_ibx;
-	else
-		MISSING_CASE(INTEL_PCH_TYPE(dev_priv));
-}
-
 static void
 intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
 {
@@ -344,47 +179,14 @@ static void gen2_irq_init(struct intel_uncore *uncore,
 	intel_uncore_posting_read16(uncore, GEN2_IMR);
 }
 
-/* For display hotplug interrupt */
-static inline void
-i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv,
-				     u32 mask,
-				     u32 bits)
-{
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, bits & ~mask);
-
-	intel_uncore_rmw(&dev_priv->uncore, PORT_HOTPLUG_EN, mask, bits);
-}
-
-/**
- * i915_hotplug_interrupt_update - update hotplug interrupt enable
- * @dev_priv: driver private
- * @mask: bits to update
- * @bits: bits to enable
- * NOTE: the HPD enable bits are modified both inside and outside
- * of an interrupt context. To avoid that read-modify-write cycles
- * interfer, these bits are protected by a spinlock. Since this
- * function is usually not called from a context where the lock is
- * held already, this function acquires the lock itself. A non-locking
- * version is also available.
- */
-void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
-				   u32 mask,
-				   u32 bits)
-{
-	spin_lock_irq(&dev_priv->irq_lock);
-	i915_hotplug_interrupt_update_locked(dev_priv, mask, bits);
-	spin_unlock_irq(&dev_priv->irq_lock);
-}
-
 /**
  * ilk_update_display_irq - update DEIMR
  * @dev_priv: driver private
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-static void ilk_update_display_irq(struct drm_i915_private *dev_priv,
-				   u32 interrupt_mask, u32 enabled_irq_mask)
+void ilk_update_display_irq(struct drm_i915_private *dev_priv,
+			    u32 interrupt_mask, u32 enabled_irq_mask)
 {
 	u32 new_val;
 
@@ -419,9 +221,8 @@ void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits)
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-static void bdw_update_port_irq(struct drm_i915_private *dev_priv,
-				u32 interrupt_mask,
-				u32 enabled_irq_mask)
+void bdw_update_port_irq(struct drm_i915_private *dev_priv,
+			 u32 interrupt_mask, u32 enabled_irq_mask)
 {
 	u32 new_val;
 	u32 old_val;
@@ -494,9 +295,9 @@ void bdw_disable_pipe_irq(struct drm_i915_private *i915,
  * @interrupt_mask: mask of interrupt bits to update
  * @enabled_irq_mask: mask of interrupt bits to enable
  */
-static void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
-					 u32 interrupt_mask,
-					 u32 enabled_irq_mask)
+void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
+				  u32 interrupt_mask,
+				  u32 enabled_irq_mask)
 {
 	u32 sdeimr = intel_uncore_read(&dev_priv->uncore, SDEIMR);
 	sdeimr &= ~interrupt_mask;
@@ -724,209 +525,6 @@ static void ivb_parity_work(struct work_struct *work)
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 }
 
-static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_TC1:
-	case HPD_PORT_TC2:
-	case HPD_PORT_TC3:
-	case HPD_PORT_TC4:
-	case HPD_PORT_TC5:
-	case HPD_PORT_TC6:
-		return val & GEN11_HOTPLUG_CTL_LONG_DETECT(pin);
-	default:
-		return false;
-	}
-}
-
-static bool bxt_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_A:
-		return val & PORTA_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_B:
-		return val & PORTB_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_C:
-		return val & PORTC_HOTPLUG_LONG_DETECT;
-	default:
-		return false;
-	}
-}
-
-static bool icp_ddi_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_A:
-	case HPD_PORT_B:
-	case HPD_PORT_C:
-	case HPD_PORT_D:
-		return val & SHOTPLUG_CTL_DDI_HPD_LONG_DETECT(pin);
-	default:
-		return false;
-	}
-}
-
-static bool icp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_TC1:
-	case HPD_PORT_TC2:
-	case HPD_PORT_TC3:
-	case HPD_PORT_TC4:
-	case HPD_PORT_TC5:
-	case HPD_PORT_TC6:
-		return val & ICP_TC_HPD_LONG_DETECT(pin);
-	default:
-		return false;
-	}
-}
-
-static bool spt_port_hotplug2_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_E:
-		return val & PORTE_HOTPLUG_LONG_DETECT;
-	default:
-		return false;
-	}
-}
-
-static bool spt_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_A:
-		return val & PORTA_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_B:
-		return val & PORTB_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_C:
-		return val & PORTC_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_D:
-		return val & PORTD_HOTPLUG_LONG_DETECT;
-	default:
-		return false;
-	}
-}
-
-static bool ilk_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_A:
-		return val & DIGITAL_PORTA_HOTPLUG_LONG_DETECT;
-	default:
-		return false;
-	}
-}
-
-static bool pch_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_B:
-		return val & PORTB_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_C:
-		return val & PORTC_HOTPLUG_LONG_DETECT;
-	case HPD_PORT_D:
-		return val & PORTD_HOTPLUG_LONG_DETECT;
-	default:
-		return false;
-	}
-}
-
-static bool i9xx_port_hotplug_long_detect(enum hpd_pin pin, u32 val)
-{
-	switch (pin) {
-	case HPD_PORT_B:
-		return val & PORTB_HOTPLUG_INT_LONG_PULSE;
-	case HPD_PORT_C:
-		return val & PORTC_HOTPLUG_INT_LONG_PULSE;
-	case HPD_PORT_D:
-		return val & PORTD_HOTPLUG_INT_LONG_PULSE;
-	default:
-		return false;
-	}
-}
-
-/*
- * Get a bit mask of pins that have triggered, and which ones may be long.
- * This can be called multiple times with the same masks to accumulate
- * hotplug detection results from several registers.
- *
- * Note that the caller is expected to zero out the masks initially.
- */
-static void intel_get_hpd_pins(struct drm_i915_private *dev_priv,
-			       u32 *pin_mask, u32 *long_mask,
-			       u32 hotplug_trigger, u32 dig_hotplug_reg,
-			       const u32 hpd[HPD_NUM_PINS],
-			       bool long_pulse_detect(enum hpd_pin pin, u32 val))
-{
-	enum hpd_pin pin;
-
-	BUILD_BUG_ON(BITS_PER_TYPE(*pin_mask) < HPD_NUM_PINS);
-
-	for_each_hpd_pin(pin) {
-		if ((hpd[pin] & hotplug_trigger) == 0)
-			continue;
-
-		*pin_mask |= BIT(pin);
-
-		if (long_pulse_detect(pin, dig_hotplug_reg))
-			*long_mask |= BIT(pin);
-	}
-
-	drm_dbg(&dev_priv->drm,
-		"hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n",
-		hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask);
-
-}
-
-static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv,
-				  const u32 hpd[HPD_NUM_PINS])
-{
-	struct intel_encoder *encoder;
-	u32 enabled_irqs = 0;
-
-	for_each_intel_encoder(&dev_priv->drm, encoder)
-		if (dev_priv->display.hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED)
-			enabled_irqs |= hpd[encoder->hpd_pin];
-
-	return enabled_irqs;
-}
-
-static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv,
-				  const u32 hpd[HPD_NUM_PINS])
-{
-	struct intel_encoder *encoder;
-	u32 hotplug_irqs = 0;
-
-	for_each_intel_encoder(&dev_priv->drm, encoder)
-		hotplug_irqs |= hpd[encoder->hpd_pin];
-
-	return hotplug_irqs;
-}
-
-static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915,
-				  hotplug_mask_func hotplug_mask)
-{
-	enum hpd_pin pin;
-	u32 hotplug = 0;
-
-	for_each_hpd_pin(pin)
-		hotplug |= hotplug_mask(pin);
-
-	return hotplug;
-}
-
-static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915,
-				     hotplug_enables_func hotplug_enables)
-{
-	struct intel_encoder *encoder;
-	u32 hotplug = 0;
-
-	for_each_intel_encoder(&i915->drm, encoder)
-		hotplug |= hotplug_enables(encoder);
-
-	return hotplug;
-}
-
 #if defined(CONFIG_DEBUG_FS)
 static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
 					 enum pipe pipe,
@@ -1199,71 +797,6 @@ static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
 		intel_gmbus_irq_handler(dev_priv);
 }
 
-static u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_status = 0, hotplug_status_mask;
-	int i;
-
-	if (IS_G4X(dev_priv) ||
-	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		hotplug_status_mask = HOTPLUG_INT_STATUS_G4X |
-			DP_AUX_CHANNEL_MASK_INT_STATUS_G4X;
-	else
-		hotplug_status_mask = HOTPLUG_INT_STATUS_I915;
-
-	/*
-	 * We absolutely have to clear all the pending interrupt
-	 * bits in PORT_HOTPLUG_STAT. Otherwise the ISR port
-	 * interrupt bit won't have an edge, and the i965/g4x
-	 * edge triggered IIR will not notice that an interrupt
-	 * is still pending. We can't use PORT_HOTPLUG_EN to
-	 * guarantee the edge as the act of toggling the enable
-	 * bits can itself generate a new hotplug interrupt :(
-	 */
-	for (i = 0; i < 10; i++) {
-		u32 tmp = intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT) & hotplug_status_mask;
-
-		if (tmp == 0)
-			return hotplug_status;
-
-		hotplug_status |= tmp;
-		intel_uncore_write(&dev_priv->uncore, PORT_HOTPLUG_STAT, hotplug_status);
-	}
-
-	drm_WARN_ONCE(&dev_priv->drm, 1,
-		      "PORT_HOTPLUG_STAT did not clear (0x%08x)\n",
-		      intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT));
-
-	return hotplug_status;
-}
-
-static void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv,
-				 u32 hotplug_status)
-{
-	u32 pin_mask = 0, long_mask = 0;
-	u32 hotplug_trigger;
-
-	if (IS_G4X(dev_priv) ||
-	    IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
-	else
-		hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
-
-	if (hotplug_trigger) {
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   hotplug_trigger, hotplug_trigger,
-				   dev_priv->display.hotplug.hpd,
-				   i9xx_port_hotplug_long_detect);
-
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-	}
-
-	if ((IS_G4X(dev_priv) ||
-	     IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
-	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
-		intel_dp_aux_irq_handler(dev_priv);
-}
-
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
@@ -1428,38 +961,6 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-static void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv,
-				u32 hotplug_trigger)
-{
-	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-	/*
-	 * Somehow the PCH doesn't seem to really ack the interrupt to the CPU
-	 * unless we touch the hotplug register, even if hotplug_trigger is
-	 * zero. Not acking leads to "The master control interrupt lied (SDE)!"
-	 * errors.
-	 */
-	dig_hotplug_reg = intel_uncore_read(&dev_priv->uncore, PCH_PORT_HOTPLUG);
-	if (!hotplug_trigger) {
-		u32 mask = PORTA_HOTPLUG_STATUS_MASK |
-			PORTD_HOTPLUG_STATUS_MASK |
-			PORTC_HOTPLUG_STATUS_MASK |
-			PORTB_HOTPLUG_STATUS_MASK;
-		dig_hotplug_reg &= ~mask;
-	}
-
-	intel_uncore_write(&dev_priv->uncore, PCH_PORT_HOTPLUG, dig_hotplug_reg);
-	if (!hotplug_trigger)
-		return;
-
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.pch_hpd,
-			   pch_port_hotplug_long_detect);
-
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-}
-
 static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 {
 	enum pipe pipe;
@@ -1585,133 +1086,6 @@ static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
 		cpt_serr_int_handler(dev_priv);
 }
 
-static void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir)
-{
-	enum hpd_pin pin;
-	u32 hotplug_trigger = iir & (XELPDP_DP_ALT_HOTPLUG_MASK | XELPDP_TBT_HOTPLUG_MASK);
-	u32 trigger_aux = iir & XELPDP_AUX_TC_MASK;
-	u32 pin_mask = 0, long_mask = 0;
-
-	for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) {
-		u32 val;
-
-		if (!(i915->display.hotplug.hpd[pin] & hotplug_trigger))
-			continue;
-
-		pin_mask |= BIT(pin);
-
-		val = intel_de_read(i915, XELPDP_PORT_HOTPLUG_CTL(pin));
-		intel_de_write(i915, XELPDP_PORT_HOTPLUG_CTL(pin), val);
-
-		if (val & (XELPDP_DP_ALT_HPD_LONG_DETECT | XELPDP_TBT_HPD_LONG_DETECT))
-			long_mask |= BIT(pin);
-	}
-
-	if (pin_mask) {
-		drm_dbg(&i915->drm,
-			"pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n",
-			hotplug_trigger, pin_mask, long_mask);
-
-		intel_hpd_irq_handler(i915, pin_mask, long_mask);
-	}
-
-	if (trigger_aux)
-		intel_dp_aux_irq_handler(i915);
-
-	if (!pin_mask && !trigger_aux)
-		drm_err(&i915->drm,
-			"Unexpected DE HPD/AUX interrupt 0x%08x\n", iir);
-}
-
-static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
-{
-	u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP;
-	u32 tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_MASK_ICP;
-	u32 pin_mask = 0, long_mask = 0;
-
-	if (ddi_hotplug_trigger) {
-		u32 dig_hotplug_reg;
-
-		/* Locking due to DSI native GPIO sequences */
-		spin_lock(&dev_priv->irq_lock);
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, 0, 0);
-		spin_unlock(&dev_priv->irq_lock);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   ddi_hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
-				   icp_ddi_port_hotplug_long_detect);
-	}
-
-	if (tc_hotplug_trigger) {
-		u32 dig_hotplug_reg;
-
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, 0, 0);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   tc_hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
-				   icp_tc_port_hotplug_long_detect);
-	}
-
-	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-
-	if (pch_iir & SDE_GMBUS_ICP)
-		intel_gmbus_irq_handler(dev_priv);
-}
-
-static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
-{
-	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT &
-		~SDE_PORTE_HOTPLUG_SPT;
-	u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT;
-	u32 pin_mask = 0, long_mask = 0;
-
-	if (hotplug_trigger) {
-		u32 dig_hotplug_reg;
-
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   hotplug_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
-				   spt_port_hotplug_long_detect);
-	}
-
-	if (hotplug2_trigger) {
-		u32 dig_hotplug_reg;
-
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, 0, 0);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   hotplug2_trigger, dig_hotplug_reg,
-				   dev_priv->display.hotplug.pch_hpd,
-				   spt_port_hotplug2_long_detect);
-	}
-
-	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-
-	if (pch_iir & SDE_GMBUS_CPT)
-		intel_gmbus_irq_handler(dev_priv);
-}
-
-static void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv,
-				u32 hotplug_trigger)
-{
-	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0);
-
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.hpd,
-			   ilk_port_hotplug_long_detect);
-
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-}
-
 static void ilk_display_irq_handler(struct drm_i915_private *dev_priv,
 				    u32 de_iir)
 {
@@ -1876,56 +1250,6 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-static void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv,
-				u32 hotplug_trigger)
-{
-	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-	dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0);
-
-	intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-			   hotplug_trigger, dig_hotplug_reg,
-			   dev_priv->display.hotplug.hpd,
-			   bxt_port_hotplug_long_detect);
-
-	intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-}
-
-static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
-{
-	u32 pin_mask = 0, long_mask = 0;
-	u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK;
-	u32 trigger_tbt = iir & GEN11_DE_TBT_HOTPLUG_MASK;
-
-	if (trigger_tc) {
-		u32 dig_hotplug_reg;
-
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, 0, 0);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   trigger_tc, dig_hotplug_reg,
-				   dev_priv->display.hotplug.hpd,
-				   gen11_port_hotplug_long_detect);
-	}
-
-	if (trigger_tbt) {
-		u32 dig_hotplug_reg;
-
-		dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, 0, 0);
-
-		intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask,
-				   trigger_tbt, dig_hotplug_reg,
-				   dev_priv->display.hotplug.hpd,
-				   gen11_port_hotplug_long_detect);
-	}
-
-	if (pin_mask)
-		intel_hpd_irq_handler(dev_priv, pin_mask, long_mask);
-	else
-		drm_err(&dev_priv->drm,
-			"Unexpected DE HPD interrupt 0x%08x\n", iir);
-}
-
 static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
 {
 	u32 mask;
@@ -2930,696 +2254,39 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
-static u32 ibx_hotplug_mask(enum hpd_pin hpd_pin)
+/*
+ * SDEIER is also touched by the interrupt handler to work around missed PCH
+ * interrupts. Hence we can't update it after the interrupt handler is enabled -
+ * instead we unconditionally enable all PCH interrupt sources here, but then
+ * only unmask them as needed with SDEIMR.
+ *
+ * Note that we currently do this after installing the interrupt handler,
+ * but before we enable the master interrupt. That should be sufficient
+ * to avoid races with the irq handler, assuming we have MSI. Shared legacy
+ * interrupts could still race.
+ */
+static void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
 {
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-		return PORTA_HOTPLUG_ENABLE;
-	case HPD_PORT_B:
-		return PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_MASK;
-	case HPD_PORT_C:
-		return PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_MASK;
-	case HPD_PORT_D:
-		return PORTD_HOTPLUG_ENABLE | PORTD_PULSE_DURATION_MASK;
-	default:
-		return 0;
-	}
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	u32 mask;
+
+	if (HAS_PCH_NOP(dev_priv))
+		return;
+
+	if (HAS_PCH_IBX(dev_priv))
+		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
+	else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
+		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
+	else
+		mask = SDE_GMBUS_CPT;
+
+	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
 }
 
-static u32 ibx_hotplug_enables(struct intel_encoder *encoder)
+static void ilk_irq_postinstall(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	switch (encoder->hpd_pin) {
-	case HPD_PORT_A:
-		/*
-		 * When CPU and PCH are on the same package, port A
-		 * HPD must be enabled in both north and south.
-		 */
-		return HAS_PCH_LPT_LP(i915) ?
-			PORTA_HOTPLUG_ENABLE : 0;
-	case HPD_PORT_B:
-		return PORTB_HOTPLUG_ENABLE |
-			PORTB_PULSE_DURATION_2ms;
-	case HPD_PORT_C:
-		return PORTC_HOTPLUG_ENABLE |
-			PORTC_PULSE_DURATION_2ms;
-	case HPD_PORT_D:
-		return PORTD_HOTPLUG_ENABLE |
-			PORTD_PULSE_DURATION_2ms;
-	default:
-		return 0;
-	}
-}
-
-static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	/*
-	 * Enable digital hotplug on the PCH, and configure the DP short pulse
-	 * duration to 2ms (which is the minimum in the Display Port spec).
-	 * The pulse duration bits are reserved on LPT+.
-	 */
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, ibx_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables));
-}
-
-static void ibx_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 ibx_hotplug_mask(encoder->hpd_pin),
-			 ibx_hotplug_enables(encoder));
-}
-
-static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
-
-	ibx_hpd_detection_setup(dev_priv);
-}
-
-static u32 icp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-	case HPD_PORT_B:
-	case HPD_PORT_C:
-	case HPD_PORT_D:
-		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
-	default:
-		return 0;
-	}
-}
-
-static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder)
-{
-	return icp_ddi_hotplug_mask(encoder->hpd_pin);
-}
-
-static u32 icp_tc_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_TC1:
-	case HPD_PORT_TC2:
-	case HPD_PORT_TC3:
-	case HPD_PORT_TC4:
-	case HPD_PORT_TC5:
-	case HPD_PORT_TC6:
-		return ICP_TC_HPD_ENABLE(hpd_pin);
-	default:
-		return 0;
-	}
-}
-
-static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder)
-{
-	return icp_tc_hotplug_mask(encoder->hpd_pin);
-}
-
-static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI,
-			 intel_hpd_hotplug_mask(dev_priv, icp_ddi_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables));
-}
-
-static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_DDI,
-			 icp_ddi_hotplug_mask(encoder->hpd_pin),
-			 icp_ddi_hotplug_enables(encoder));
-}
-
-static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC,
-			 intel_hpd_hotplug_mask(dev_priv, icp_tc_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables));
-}
-
-static void icp_tc_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_TC,
-			 icp_tc_hotplug_mask(encoder->hpd_pin),
-			 icp_tc_hotplug_enables(encoder));
-}
-
-static void icp_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	icp_ddi_hpd_enable_detection(encoder);
-	icp_tc_hpd_enable_detection(encoder);
-}
-
-static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-
-	if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP)
-		intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
-
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
-
-	icp_ddi_hpd_detection_setup(dev_priv);
-	icp_tc_hpd_detection_setup(dev_priv);
-}
-
-static u32 gen11_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_TC1:
-	case HPD_PORT_TC2:
-	case HPD_PORT_TC3:
-	case HPD_PORT_TC4:
-	case HPD_PORT_TC5:
-	case HPD_PORT_TC6:
-		return GEN11_HOTPLUG_CTL_ENABLE(hpd_pin);
-	default:
-		return 0;
-	}
-}
-
-static u32 gen11_hotplug_enables(struct intel_encoder *encoder)
-{
-	return gen11_hotplug_mask(encoder->hpd_pin);
-}
-
-static void dg1_hpd_invert(struct drm_i915_private *i915)
-{
-	u32 val = (INVERT_DDIA_HPD |
-		   INVERT_DDIB_HPD |
-		   INVERT_DDIC_HPD |
-		   INVERT_DDID_HPD);
-	intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val);
-}
-
-static void dg1_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	dg1_hpd_invert(i915);
-	icp_hpd_enable_detection(encoder);
-}
-
-static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	dg1_hpd_invert(dev_priv);
-	icp_hpd_irq_setup(dev_priv);
-}
-
-static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL,
-			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
-}
-
-static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, GEN11_TC_HOTPLUG_CTL,
-			 gen11_hotplug_mask(encoder->hpd_pin),
-			 gen11_hotplug_enables(encoder));
-}
-
-static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables));
-}
-
-static void gen11_tbt_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, GEN11_TBT_HOTPLUG_CTL,
-			 gen11_hotplug_mask(encoder->hpd_pin),
-			 gen11_hotplug_enables(encoder));
-}
-
-static void gen11_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	gen11_tc_hpd_enable_detection(encoder);
-	gen11_tbt_hpd_enable_detection(encoder);
-
-	if (INTEL_PCH_TYPE(i915) >= PCH_ICP)
-		icp_hpd_enable_detection(encoder);
-}
-
-static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-
-	intel_uncore_rmw(&dev_priv->uncore, GEN11_DE_HPD_IMR, hotplug_irqs,
-			 ~enabled_irqs & hotplug_irqs);
-	intel_uncore_posting_read(&dev_priv->uncore, GEN11_DE_HPD_IMR);
-
-	gen11_tc_hpd_detection_setup(dev_priv);
-	gen11_tbt_hpd_detection_setup(dev_priv);
-
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		icp_hpd_irq_setup(dev_priv);
-}
-
-static u32 mtp_ddi_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-	case HPD_PORT_B:
-		return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin);
-	default:
-		return 0;
-	}
-}
-
-static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder)
-{
-	return mtp_ddi_hotplug_mask(encoder->hpd_pin);
-}
-
-static u32 mtp_tc_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_TC1:
-	case HPD_PORT_TC2:
-	case HPD_PORT_TC3:
-	case HPD_PORT_TC4:
-		return ICP_TC_HPD_ENABLE(hpd_pin);
-	default:
-		return 0;
-	}
-}
-
-static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder)
-{
-	return mtp_tc_hotplug_mask(encoder->hpd_pin);
-}
-
-static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915)
-{
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     intel_hpd_hotplug_mask(i915, mtp_ddi_hotplug_mask),
-		     intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables));
-}
-
-static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     mtp_ddi_hotplug_mask(encoder->hpd_pin),
-		     mtp_ddi_hotplug_enables(encoder));
-}
-
-static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915)
-{
-	intel_de_rmw(i915, SHOTPLUG_CTL_TC,
-		     intel_hpd_hotplug_mask(i915, mtp_tc_hotplug_mask),
-		     intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables));
-}
-
-static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_de_rmw(i915, SHOTPLUG_CTL_DDI,
-		     mtp_tc_hotplug_mask(encoder->hpd_pin),
-		     mtp_tc_hotplug_enables(encoder));
-}
-
-static void mtp_hpd_invert(struct drm_i915_private *i915)
-{
-	u32 val = (INVERT_DDIA_HPD |
-		   INVERT_DDIB_HPD |
-		   INVERT_DDIC_HPD |
-		   INVERT_TC1_HPD |
-		   INVERT_TC2_HPD |
-		   INVERT_TC3_HPD |
-		   INVERT_TC4_HPD |
-		   INVERT_DDID_HPD_MTP |
-		   INVERT_DDIE_HPD);
-	intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val);
-}
-
-static void mtp_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	mtp_hpd_invert(i915);
-	mtp_ddi_hpd_enable_detection(encoder);
-	mtp_tc_hpd_enable_detection(encoder);
-}
-
-static void mtp_hpd_irq_setup(struct drm_i915_private *i915)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd);
-
-	intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
-
-	mtp_hpd_invert(i915);
-	ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs);
-
-	mtp_ddi_hpd_detection_setup(i915);
-	mtp_tc_hpd_detection_setup(i915);
-}
-
-static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin)
-{
-	return hpd_pin >= HPD_PORT_TC1 && hpd_pin <= HPD_PORT_TC4;
-}
-
-static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915,
-					     enum hpd_pin hpd_pin, bool enable)
-{
-	u32 mask = XELPDP_TBT_HOTPLUG_ENABLE |
-		XELPDP_DP_ALT_HOTPLUG_ENABLE;
-
-	if (!is_xelpdp_pica_hpd_pin(hpd_pin))
-		return;
-
-	intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(hpd_pin),
-		     mask, enable ? mask : 0);
-}
-
-static void xelpdp_pica_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	_xelpdp_pica_hpd_detection_setup(i915, encoder->hpd_pin, true);
-}
-
-static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915)
-{
-	struct intel_encoder *encoder;
-	u32 available_pins = 0;
-	enum hpd_pin pin;
-
-	BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS);
-
-	for_each_intel_encoder(&i915->drm, encoder)
-		available_pins |= BIT(encoder->hpd_pin);
-
-	for_each_hpd_pin(pin)
-		_xelpdp_pica_hpd_detection_setup(i915, pin, available_pins & BIT(pin));
-}
-
-static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	xelpdp_pica_hpd_enable_detection(encoder);
-	mtp_hpd_enable_detection(encoder);
-}
-
-static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.hpd);
-
-	intel_de_rmw(i915, PICAINTERRUPT_IMR, hotplug_irqs,
-		     ~enabled_irqs & hotplug_irqs);
-	intel_uncore_posting_read(&i915->uncore, PICAINTERRUPT_IMR);
-
-	xelpdp_pica_hpd_detection_setup(i915);
-
-	if (INTEL_PCH_TYPE(i915) >= PCH_MTP)
-		mtp_hpd_irq_setup(i915);
-}
-
-static u32 spt_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-		return PORTA_HOTPLUG_ENABLE;
-	case HPD_PORT_B:
-		return PORTB_HOTPLUG_ENABLE;
-	case HPD_PORT_C:
-		return PORTC_HOTPLUG_ENABLE;
-	case HPD_PORT_D:
-		return PORTD_HOTPLUG_ENABLE;
-	default:
-		return 0;
-	}
-}
-
-static u32 spt_hotplug_enables(struct intel_encoder *encoder)
-{
-	return spt_hotplug_mask(encoder->hpd_pin);
-}
-
-static u32 spt_hotplug2_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_E:
-		return PORTE_HOTPLUG_ENABLE;
-	default:
-		return 0;
-	}
-}
-
-static u32 spt_hotplug2_enables(struct intel_encoder *encoder)
-{
-	return spt_hotplug2_mask(encoder->hpd_pin);
-}
-
-static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	/* Display WA #1179 WaHardHangonHotPlug: cnp */
-	if (HAS_PCH_CNP(dev_priv)) {
-		intel_uncore_rmw(&dev_priv->uncore, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK,
-				 CHASSIS_CLK_REQ_DURATION(0xf));
-	}
-
-	/* Enable digital hotplug on the PCH */
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables));
-
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2,
-			 intel_hpd_hotplug_mask(dev_priv, spt_hotplug2_mask),
-			 intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables));
-}
-
-static void spt_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	/* Display WA #1179 WaHardHangonHotPlug: cnp */
-	if (HAS_PCH_CNP(i915)) {
-		intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1,
-				 CHASSIS_CLK_REQ_DURATION_MASK,
-				 CHASSIS_CLK_REQ_DURATION(0xf));
-	}
-
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 spt_hotplug_mask(encoder->hpd_pin),
-			 spt_hotplug_enables(encoder));
-
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG2,
-			 spt_hotplug2_mask(encoder->hpd_pin),
-			 spt_hotplug2_enables(encoder));
-}
-
-static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP)
-		intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ);
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd);
-
-	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
-
-	spt_hpd_detection_setup(dev_priv);
-}
-
-static u32 ilk_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-		return DIGITAL_PORTA_HOTPLUG_ENABLE |
-			DIGITAL_PORTA_PULSE_DURATION_MASK;
-	default:
-		return 0;
-	}
-}
-
-static u32 ilk_hotplug_enables(struct intel_encoder *encoder)
-{
-	switch (encoder->hpd_pin) {
-	case HPD_PORT_A:
-		return DIGITAL_PORTA_HOTPLUG_ENABLE |
-			DIGITAL_PORTA_PULSE_DURATION_2ms;
-	default:
-		return 0;
-	}
-}
-
-static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	/*
-	 * Enable digital hotplug on the CPU, and configure the DP short pulse
-	 * duration to 2ms (which is the minimum in the Display Port spec)
-	 * The pulse duration bits are reserved on HSW+.
-	 */
-	intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 intel_hpd_hotplug_mask(dev_priv, ilk_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables));
-}
-
-static void ilk_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, DIGITAL_PORT_HOTPLUG_CNTRL,
-			 ilk_hotplug_mask(encoder->hpd_pin),
-			 ilk_hotplug_enables(encoder));
-
-	ibx_hpd_enable_detection(encoder);
-}
-
-static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-
-	if (DISPLAY_VER(dev_priv) >= 8)
-		bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
-	else
-		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
-
-	ilk_hpd_detection_setup(dev_priv);
-
-	ibx_hpd_irq_setup(dev_priv);
-}
-
-static u32 bxt_hotplug_mask(enum hpd_pin hpd_pin)
-{
-	switch (hpd_pin) {
-	case HPD_PORT_A:
-		return PORTA_HOTPLUG_ENABLE | BXT_DDIA_HPD_INVERT;
-	case HPD_PORT_B:
-		return PORTB_HOTPLUG_ENABLE | BXT_DDIB_HPD_INVERT;
-	case HPD_PORT_C:
-		return PORTC_HOTPLUG_ENABLE | BXT_DDIC_HPD_INVERT;
-	default:
-		return 0;
-	}
-}
-
-static u32 bxt_hotplug_enables(struct intel_encoder *encoder)
-{
-	u32 hotplug;
-
-	switch (encoder->hpd_pin) {
-	case HPD_PORT_A:
-		hotplug = PORTA_HOTPLUG_ENABLE;
-		if (intel_bios_encoder_hpd_invert(encoder->devdata))
-			hotplug |= BXT_DDIA_HPD_INVERT;
-		return hotplug;
-	case HPD_PORT_B:
-		hotplug = PORTB_HOTPLUG_ENABLE;
-		if (intel_bios_encoder_hpd_invert(encoder->devdata))
-			hotplug |= BXT_DDIB_HPD_INVERT;
-		return hotplug;
-	case HPD_PORT_C:
-		hotplug = PORTC_HOTPLUG_ENABLE;
-		if (intel_bios_encoder_hpd_invert(encoder->devdata))
-			hotplug |= BXT_DDIC_HPD_INVERT;
-		return hotplug;
-	default:
-		return 0;
-	}
-}
-
-static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv)
-{
-	intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG,
-			 intel_hpd_hotplug_mask(dev_priv, bxt_hotplug_mask),
-			 intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables));
-}
-
-static void bxt_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG,
-			 bxt_hotplug_mask(encoder->hpd_pin),
-			 bxt_hotplug_enables(encoder));
-}
-
-static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_irqs, enabled_irqs;
-
-	enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-	hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd);
-
-	bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
-
-	bxt_hpd_detection_setup(dev_priv);
-}
-
-/*
- * SDEIER is also touched by the interrupt handler to work around missed PCH
- * interrupts. Hence we can't update it after the interrupt handler is enabled -
- * instead we unconditionally enable all PCH interrupt sources here, but then
- * only unmask them as needed with SDEIMR.
- *
- * Note that we currently do this after installing the interrupt handler,
- * but before we enable the master interrupt. That should be sufficient
- * to avoid races with the irq handler, assuming we have MSI. Shared legacy
- * interrupts could still race.
- */
-static void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	u32 mask;
-
-	if (HAS_PCH_NOP(dev_priv))
-		return;
-
-	if (HAS_PCH_IBX(dev_priv))
-		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
-	else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
-		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
-	else
-		mask = SDE_GMBUS_CPT;
-
-	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
-}
-
-static void ilk_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	u32 display_mask, extra_mask;
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	u32 display_mask, extra_mask;
 
 	if (GRAPHICS_VER(dev_priv) >= 7) {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE_IVB |
@@ -4252,40 +2919,6 @@ static void i965_irq_postinstall(struct drm_i915_private *dev_priv)
 	i915_enable_asle_pipestat(dev_priv);
 }
 
-static void i915_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-	u32 hotplug_en = hpd_mask_i915[encoder->hpd_pin];
-
-	/* HPD sense and interrupt enable are one and the same */
-	i915_hotplug_interrupt_update(i915, hotplug_en, hotplug_en);
-}
-
-static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv)
-{
-	u32 hotplug_en;
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	/* Note HDMI and DP share hotplug bits */
-	/* enable bits are the same for all generations */
-	hotplug_en = intel_hpd_enabled_irqs(dev_priv, hpd_mask_i915);
-	/* Programming the CRT detection parameters tends
-	   to generate a spurious hotplug event about three
-	   seconds later.  So just do it once.
-	*/
-	if (IS_G4X(dev_priv))
-		hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
-	hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
-
-	/* Ignore TV since it's buggy */
-	i915_hotplug_interrupt_update_locked(dev_priv,
-					     HOTPLUG_INT_EN_MASK |
-					     CRT_HOTPLUG_VOLTAGE_COMPARE_MASK |
-					     CRT_HOTPLUG_ACTIVATION_PERIOD_64,
-					     hotplug_en);
-}
-
 static irqreturn_t i965_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
@@ -4345,43 +2978,6 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-struct intel_hotplug_funcs {
-	/* Enable HPD sense and interrupts for all present encoders */
-	void (*hpd_irq_setup)(struct drm_i915_private *i915);
-	/* Enable HPD sense for a single encoder */
-	void (*hpd_enable_detection)(struct intel_encoder *encoder);
-};
-
-#define HPD_FUNCS(platform)					 \
-static const struct intel_hotplug_funcs platform##_hpd_funcs = { \
-	.hpd_irq_setup = platform##_hpd_irq_setup,		 \
-	.hpd_enable_detection = platform##_hpd_enable_detection, \
-}
-
-HPD_FUNCS(i915);
-HPD_FUNCS(xelpdp);
-HPD_FUNCS(dg1);
-HPD_FUNCS(gen11);
-HPD_FUNCS(bxt);
-HPD_FUNCS(icp);
-HPD_FUNCS(spt);
-HPD_FUNCS(ilk);
-#undef HPD_FUNCS
-
-void intel_hpd_enable_detection(struct intel_encoder *encoder)
-{
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	if (i915->display.funcs.hotplug)
-		i915->display.funcs.hotplug->hpd_enable_detection(encoder);
-}
-
-void intel_hpd_irq_setup(struct drm_i915_private *i915)
-{
-	if (i915->display_irqs_enabled && i915->display.funcs.hotplug)
-		i915->display.funcs.hotplug->hpd_irq_setup(i915);
-}
-
 /**
  * intel_irq_init - initializes irq support
  * @dev_priv: i915 device instance
@@ -4404,10 +3000,6 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
-	intel_hpd_init_pins(dev_priv);
-
-	intel_hpd_init_early(dev_priv);
-
 	dev_priv->drm.vblank_disable_immediate = true;
 
 	/* Most platforms treat the display irq block as an always-on
@@ -4420,27 +3012,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
 		dev_priv->display_irqs_enabled = false;
 
-	if (HAS_GMCH(dev_priv)) {
-		if (I915_HAS_HOTPLUG(dev_priv))
-			dev_priv->display.funcs.hotplug = &i915_hpd_funcs;
-	} else {
-		if (HAS_PCH_DG2(dev_priv))
-			dev_priv->display.funcs.hotplug = &icp_hpd_funcs;
-		else if (HAS_PCH_DG1(dev_priv))
-			dev_priv->display.funcs.hotplug = &dg1_hpd_funcs;
-		else if (DISPLAY_VER(dev_priv) >= 14)
-			dev_priv->display.funcs.hotplug = &xelpdp_hpd_funcs;
-		else if (DISPLAY_VER(dev_priv) >= 11)
-			dev_priv->display.funcs.hotplug = &gen11_hpd_funcs;
-		else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
-			dev_priv->display.funcs.hotplug = &bxt_hpd_funcs;
-		else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-			dev_priv->display.funcs.hotplug = &icp_hpd_funcs;
-		else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
-			dev_priv->display.funcs.hotplug = &spt_hpd_funcs;
-		else
-			dev_priv->display.funcs.hotplug = &ilk_hpd_funcs;
-	}
+	intel_hotplug_irq_init(dev_priv);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h
index dd47e473ba4f7..913c854f873de 100644
--- a/drivers/gpu/drm/i915/i915_irq.h
+++ b/drivers/gpu/drm/i915/i915_irq.h
@@ -38,18 +38,18 @@ i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv);
 void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv);
 
-void intel_hpd_enable_detection(struct intel_encoder *encoder);
-void intel_hpd_irq_setup(struct drm_i915_private *i915);
-void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
-				   u32 mask,
-				   u32 bits);
-
+void ilk_update_display_irq(struct drm_i915_private *i915,
+			    u32 interrupt_mask, u32 enabled_irq_mask);
 void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits);
 void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits);
 
+void bdw_update_port_irq(struct drm_i915_private *i915,
+			 u32 interrupt_mask, u32 enabled_irq_mask);
 void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
 void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
 
+void ibx_display_interrupt_update(struct drm_i915_private *i915,
+				  u32 interrupt_mask, u32 enabled_irq_mask);
 void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits);
 void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits);
 
-- 
GitLab


From 2b874a027810d50b627408f51c59b9648f778a19 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 15 May 2023 13:17:38 +0300
Subject: [PATCH 0506/1791] drm/i915/irq: split out display irq handling

Split (non-hotplug) display irq handling out of i915_irq.[ch] into
display/intel_display_irq.[ch].

v3:
- Preserve [I915_MAX_PIPES] harder (kernel test robot)

v2:
- Rebase
- Preserve [I915_MAX_PIPES] in functions (kernel test robot)

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515101738.2399816-3-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |    1 +
 drivers/gpu/drm/i915/display/i9xx_plane.c     |    2 +-
 drivers/gpu/drm/i915/display/intel_crtc.c     |    2 +-
 .../gpu/drm/i915/display/intel_display_irq.c  | 1668 +++++++++++++++++
 .../gpu/drm/i915/display/intel_display_irq.h  |   81 +
 .../i915/display/intel_display_power_well.c   |    1 +
 .../drm/i915/display/intel_fifo_underrun.c    |    2 +-
 .../gpu/drm/i915/display/intel_hotplug_irq.c  |    2 +-
 drivers/gpu/drm/i915/display/intel_tv.c       |    2 +-
 .../drm/i915/display/skl_universal_plane.c    |    2 +-
 drivers/gpu/drm/i915/gt/intel_rps.c           |    1 +
 drivers/gpu/drm/i915/i915_irq.c               | 1666 +---------------
 drivers/gpu/drm/i915/i915_irq.h               |   46 +-
 13 files changed, 1763 insertions(+), 1713 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_irq.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_irq.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index cc80d483fd6fd..1f4c9b99c2c96 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -238,6 +238,7 @@ i915-y += \
 	display/intel_cursor.o \
 	display/intel_display.o \
 	display/intel_display_driver.o \
+	display/intel_display_irq.o \
 	display/intel_display_power.o \
 	display/intel_display_power_map.o \
 	display/intel_display_power_well.o \
diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index ecaeb7dc196bc..616654adbfb8f 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -8,12 +8,12 @@
 #include <drm/drm_blend.h>
 #include <drm/drm_fourcc.h>
 
-#include "i915_irq.h"
 #include "i915_reg.h"
 #include "i9xx_plane.h"
 #include "intel_atomic.h"
 #include "intel_atomic_plane.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_types.h"
 #include "intel_fb.h"
 #include "intel_fbc.h"
diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index ecae9bf05269b..93c3226b98c90 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -11,7 +11,6 @@
 #include <drm/drm_plane.h>
 #include <drm/drm_vblank_work.h>
 
-#include "i915_irq.h"
 #include "i915_vgpu.h"
 #include "i9xx_plane.h"
 #include "icl_dsi.h"
@@ -21,6 +20,7 @@
 #include "intel_crtc.h"
 #include "intel_cursor.h"
 #include "intel_display_debugfs.h"
+#include "intel_display_irq.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
 #include "intel_drrs.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
new file mode 100644
index 0000000000000..0eedd1ebb3890
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -0,0 +1,1668 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include "i915_drv.h"
+#include "i915_irq.h"
+#include "i915_reg.h"
+#include "icl_dsi_regs.h"
+#include "intel_display_irq.h"
+#include "intel_display_types.h"
+#include "intel_hotplug_irq.h"
+#include "intel_psr_regs.h"
+#include "intel_crtc.h"
+#include "intel_display_trace.h"
+#include "intel_dp_aux.h"
+#include "intel_gmbus.h"
+#include "intel_fifo_underrun.h"
+#include "intel_psr.h"
+#include "intel_fdi_regs.h"
+#include "gt/intel_rps.h"
+#include "intel_de.h"
+
+static void
+intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe);
+
+	drm_crtc_handle_vblank(&crtc->base);
+}
+
+/**
+ * ilk_update_display_irq - update DEIMR
+ * @dev_priv: driver private
+ * @interrupt_mask: mask of interrupt bits to update
+ * @enabled_irq_mask: mask of interrupt bits to enable
+ */
+void ilk_update_display_irq(struct drm_i915_private *dev_priv,
+			    u32 interrupt_mask, u32 enabled_irq_mask)
+{
+	u32 new_val;
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+
+	new_val = dev_priv->irq_mask;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != dev_priv->irq_mask &&
+	    !drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) {
+		dev_priv->irq_mask = new_val;
+		intel_uncore_write(&dev_priv->uncore, DEIMR, dev_priv->irq_mask);
+		intel_uncore_posting_read(&dev_priv->uncore, DEIMR);
+	}
+}
+
+void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits)
+{
+	ilk_update_display_irq(i915, bits, bits);
+}
+
+void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits)
+{
+	ilk_update_display_irq(i915, bits, 0);
+}
+
+/**
+ * bdw_update_port_irq - update DE port interrupt
+ * @dev_priv: driver private
+ * @interrupt_mask: mask of interrupt bits to update
+ * @enabled_irq_mask: mask of interrupt bits to enable
+ */
+void bdw_update_port_irq(struct drm_i915_private *dev_priv,
+			 u32 interrupt_mask, u32 enabled_irq_mask)
+{
+	u32 new_val;
+	u32 old_val;
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+
+	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+		return;
+
+	old_val = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IMR);
+
+	new_val = old_val;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != old_val) {
+		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IMR, new_val);
+		intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PORT_IMR);
+	}
+}
+
+/**
+ * bdw_update_pipe_irq - update DE pipe interrupt
+ * @dev_priv: driver private
+ * @pipe: pipe whose interrupt to update
+ * @interrupt_mask: mask of interrupt bits to update
+ * @enabled_irq_mask: mask of interrupt bits to enable
+ */
+static void bdw_update_pipe_irq(struct drm_i915_private *dev_priv,
+				enum pipe pipe, u32 interrupt_mask,
+				u32 enabled_irq_mask)
+{
+	u32 new_val;
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+
+	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+		return;
+
+	new_val = dev_priv->de_irq_mask[pipe];
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != dev_priv->de_irq_mask[pipe]) {
+		dev_priv->de_irq_mask[pipe] = new_val;
+		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
+		intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe));
+	}
+}
+
+void bdw_enable_pipe_irq(struct drm_i915_private *i915,
+			 enum pipe pipe, u32 bits)
+{
+	bdw_update_pipe_irq(i915, pipe, bits, bits);
+}
+
+void bdw_disable_pipe_irq(struct drm_i915_private *i915,
+			  enum pipe pipe, u32 bits)
+{
+	bdw_update_pipe_irq(i915, pipe, bits, 0);
+}
+
+/**
+ * ibx_display_interrupt_update - update SDEIMR
+ * @dev_priv: driver private
+ * @interrupt_mask: mask of interrupt bits to update
+ * @enabled_irq_mask: mask of interrupt bits to enable
+ */
+void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
+				  u32 interrupt_mask,
+				  u32 enabled_irq_mask)
+{
+	u32 sdeimr = intel_uncore_read(&dev_priv->uncore, SDEIMR);
+
+	sdeimr &= ~interrupt_mask;
+	sdeimr |= (~enabled_irq_mask & interrupt_mask);
+
+	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
+		return;
+
+	intel_uncore_write(&dev_priv->uncore, SDEIMR, sdeimr);
+	intel_uncore_posting_read(&dev_priv->uncore, SDEIMR);
+}
+
+void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits)
+{
+	ibx_display_interrupt_update(i915, bits, bits);
+}
+
+void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits)
+{
+	ibx_display_interrupt_update(i915, bits, 0);
+}
+
+u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv,
+			      enum pipe pipe)
+{
+	u32 status_mask = dev_priv->pipestat_irq_mask[pipe];
+	u32 enable_mask = status_mask << 16;
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	if (DISPLAY_VER(dev_priv) < 5)
+		goto out;
+
+	/*
+	 * On pipe A we don't support the PSR interrupt yet,
+	 * on pipe B and C the same bit MBZ.
+	 */
+	if (drm_WARN_ON_ONCE(&dev_priv->drm,
+			     status_mask & PIPE_A_PSR_STATUS_VLV))
+		return 0;
+	/*
+	 * On pipe B and C we don't support the PSR interrupt yet, on pipe
+	 * A the same bit is for perf counters which we don't use either.
+	 */
+	if (drm_WARN_ON_ONCE(&dev_priv->drm,
+			     status_mask & PIPE_B_PSR_STATUS_VLV))
+		return 0;
+
+	enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS |
+			 SPRITE0_FLIP_DONE_INT_EN_VLV |
+			 SPRITE1_FLIP_DONE_INT_EN_VLV);
+	if (status_mask & SPRITE0_FLIP_DONE_INT_STATUS_VLV)
+		enable_mask |= SPRITE0_FLIP_DONE_INT_EN_VLV;
+	if (status_mask & SPRITE1_FLIP_DONE_INT_STATUS_VLV)
+		enable_mask |= SPRITE1_FLIP_DONE_INT_EN_VLV;
+
+out:
+	drm_WARN_ONCE(&dev_priv->drm,
+		      enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
+		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
+		      pipe_name(pipe), enable_mask, status_mask);
+
+	return enable_mask;
+}
+
+void i915_enable_pipestat(struct drm_i915_private *dev_priv,
+			  enum pipe pipe, u32 status_mask)
+{
+	i915_reg_t reg = PIPESTAT(pipe);
+	u32 enable_mask;
+
+	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: status_mask=0x%x\n",
+		      pipe_name(pipe), status_mask);
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
+
+	if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == status_mask)
+		return;
+
+	dev_priv->pipestat_irq_mask[pipe] |= status_mask;
+	enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
+
+	intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask);
+	intel_uncore_posting_read(&dev_priv->uncore, reg);
+}
+
+void i915_disable_pipestat(struct drm_i915_private *dev_priv,
+			   enum pipe pipe, u32 status_mask)
+{
+	i915_reg_t reg = PIPESTAT(pipe);
+	u32 enable_mask;
+
+	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: status_mask=0x%x\n",
+		      pipe_name(pipe), status_mask);
+
+	lockdep_assert_held(&dev_priv->irq_lock);
+	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
+
+	if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == 0)
+		return;
+
+	dev_priv->pipestat_irq_mask[pipe] &= ~status_mask;
+	enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
+
+	intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask);
+	intel_uncore_posting_read(&dev_priv->uncore, reg);
+}
+
+static bool i915_has_asle(struct drm_i915_private *dev_priv)
+{
+	if (!dev_priv->display.opregion.asle)
+		return false;
+
+	return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
+}
+
+/**
+ * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion
+ * @dev_priv: i915 device private
+ */
+void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv)
+{
+	if (!i915_has_asle(dev_priv))
+		return;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+
+	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS);
+	if (DISPLAY_VER(dev_priv) >= 4)
+		i915_enable_pipestat(dev_priv, PIPE_A,
+				     PIPE_LEGACY_BLC_EVENT_STATUS);
+
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
+#if defined(CONFIG_DEBUG_FS)
+static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+					 enum pipe pipe,
+					 u32 crc0, u32 crc1,
+					 u32 crc2, u32 crc3,
+					 u32 crc4)
+{
+	struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe);
+	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
+	u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 };
+
+	trace_intel_pipe_crc(crtc, crcs);
+
+	spin_lock(&pipe_crc->lock);
+	/*
+	 * For some not yet identified reason, the first CRC is
+	 * bonkers. So let's just wait for the next vblank and read
+	 * out the buggy result.
+	 *
+	 * On GEN8+ sometimes the second CRC is bonkers as well, so
+	 * don't trust that one either.
+	 */
+	if (pipe_crc->skipped <= 0 ||
+	    (DISPLAY_VER(dev_priv) >= 8 && pipe_crc->skipped == 1)) {
+		pipe_crc->skipped++;
+		spin_unlock(&pipe_crc->lock);
+		return;
+	}
+	spin_unlock(&pipe_crc->lock);
+
+	drm_crtc_add_crc_entry(&crtc->base, true,
+			       drm_crtc_accurate_vblank_count(&crtc->base),
+			       crcs);
+}
+#else
+static inline void
+display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+			     enum pipe pipe,
+			     u32 crc0, u32 crc1,
+			     u32 crc2, u32 crc3,
+			     u32 crc4) {}
+#endif
+
+static void flip_done_handler(struct drm_i915_private *i915,
+			      enum pipe pipe)
+{
+	struct intel_crtc *crtc = intel_crtc_for_pipe(i915, pipe);
+	struct drm_crtc_state *crtc_state = crtc->base.state;
+	struct drm_pending_vblank_event *e = crtc_state->event;
+	struct drm_device *dev = &i915->drm;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev->event_lock, irqflags);
+
+	crtc_state->event = NULL;
+
+	drm_crtc_send_vblank_event(&crtc->base, e);
+
+	spin_unlock_irqrestore(&dev->event_lock, irqflags);
+}
+
+static void hsw_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+				     enum pipe pipe)
+{
+	display_pipe_crc_irq_handler(dev_priv, pipe,
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)),
+				     0, 0, 0, 0);
+}
+
+static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+				     enum pipe pipe)
+{
+	display_pipe_crc_irq_handler(dev_priv, pipe,
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_2_IVB(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_3_IVB(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_4_IVB(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_5_IVB(pipe)));
+}
+
+static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
+				      enum pipe pipe)
+{
+	u32 res1, res2;
+
+	if (DISPLAY_VER(dev_priv) >= 3)
+		res1 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES1_I915(pipe));
+	else
+		res1 = 0;
+
+	if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv))
+		res2 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES2_G4X(pipe));
+	else
+		res2 = 0;
+
+	display_pipe_crc_irq_handler(dev_priv, pipe,
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RED(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_GREEN(pipe)),
+				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_BLUE(pipe)),
+				     res1, res2);
+}
+
+void i9xx_pipestat_irq_reset(struct drm_i915_private *dev_priv)
+{
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe) {
+		intel_uncore_write(&dev_priv->uncore, PIPESTAT(pipe),
+				   PIPESTAT_INT_STATUS_MASK |
+				   PIPE_FIFO_UNDERRUN_STATUS);
+
+		dev_priv->pipestat_irq_mask[pipe] = 0;
+	}
+}
+
+void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv,
+			   u32 iir, u32 pipe_stats[I915_MAX_PIPES])
+{
+	enum pipe pipe;
+
+	spin_lock(&dev_priv->irq_lock);
+
+	if (!dev_priv->display_irqs_enabled) {
+		spin_unlock(&dev_priv->irq_lock);
+		return;
+	}
+
+	for_each_pipe(dev_priv, pipe) {
+		i915_reg_t reg;
+		u32 status_mask, enable_mask, iir_bit = 0;
+
+		/*
+		 * PIPESTAT bits get signalled even when the interrupt is
+		 * disabled with the mask bits, and some of the status bits do
+		 * not generate interrupts at all (like the underrun bit). Hence
+		 * we need to be careful that we only handle what we want to
+		 * handle.
+		 */
+
+		/* fifo underruns are filterered in the underrun handler. */
+		status_mask = PIPE_FIFO_UNDERRUN_STATUS;
+
+		switch (pipe) {
+		default:
+		case PIPE_A:
+			iir_bit = I915_DISPLAY_PIPE_A_EVENT_INTERRUPT;
+			break;
+		case PIPE_B:
+			iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
+			break;
+		case PIPE_C:
+			iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+			break;
+		}
+		if (iir & iir_bit)
+			status_mask |= dev_priv->pipestat_irq_mask[pipe];
+
+		if (!status_mask)
+			continue;
+
+		reg = PIPESTAT(pipe);
+		pipe_stats[pipe] = intel_uncore_read(&dev_priv->uncore, reg) & status_mask;
+		enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
+
+		/*
+		 * Clear the PIPE*STAT regs before the IIR
+		 *
+		 * Toggle the enable bits to make sure we get an
+		 * edge in the ISR pipe event bit if we don't clear
+		 * all the enabled status bits. Otherwise the edge
+		 * triggered IIR on i965/g4x wouldn't notice that
+		 * an interrupt is still pending.
+		 */
+		if (pipe_stats[pipe]) {
+			intel_uncore_write(&dev_priv->uncore, reg, pipe_stats[pipe]);
+			intel_uncore_write(&dev_priv->uncore, reg, enable_mask);
+		}
+	}
+	spin_unlock(&dev_priv->irq_lock);
+}
+
+void i8xx_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+			       u16 iir, u32 pipe_stats[I915_MAX_PIPES])
+{
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe) {
+		if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+	}
+}
+
+void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+			       u32 iir, u32 pipe_stats[I915_MAX_PIPES])
+{
+	bool blc_event = false;
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe) {
+		if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
+			blc_event = true;
+
+		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+	}
+
+	if (blc_event || (iir & I915_ASLE_INTERRUPT))
+		intel_opregion_asle_intr(dev_priv);
+}
+
+void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+			       u32 iir, u32 pipe_stats[I915_MAX_PIPES])
+{
+	bool blc_event = false;
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe) {
+		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
+			blc_event = true;
+
+		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+	}
+
+	if (blc_event || (iir & I915_ASLE_INTERRUPT))
+		intel_opregion_asle_intr(dev_priv);
+
+	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
+		intel_gmbus_irq_handler(dev_priv);
+}
+
+void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
+				     u32 pipe_stats[I915_MAX_PIPES])
+{
+	enum pipe pipe;
+
+	for_each_pipe(dev_priv, pipe) {
+		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV)
+			flip_done_handler(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
+			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+
+		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+	}
+
+	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
+		intel_gmbus_irq_handler(dev_priv);
+}
+
+static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+{
+	enum pipe pipe;
+	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
+
+	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
+
+	if (pch_iir & SDE_AUDIO_POWER_MASK) {
+		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >>
+			       SDE_AUDIO_POWER_SHIFT);
+		drm_dbg(&dev_priv->drm, "PCH audio power change on port %d\n",
+			port_name(port));
+	}
+
+	if (pch_iir & SDE_AUX_MASK)
+		intel_dp_aux_irq_handler(dev_priv);
+
+	if (pch_iir & SDE_GMBUS)
+		intel_gmbus_irq_handler(dev_priv);
+
+	if (pch_iir & SDE_AUDIO_HDCP_MASK)
+		drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n");
+
+	if (pch_iir & SDE_AUDIO_TRANS_MASK)
+		drm_dbg(&dev_priv->drm, "PCH transcoder audio interrupt\n");
+
+	if (pch_iir & SDE_POISON)
+		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
+
+	if (pch_iir & SDE_FDI_MASK) {
+		for_each_pipe(dev_priv, pipe)
+			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
+				pipe_name(pipe),
+				intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe)));
+	}
+
+	if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE))
+		drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n");
+
+	if (pch_iir & (SDE_TRANSB_CRC_ERR | SDE_TRANSA_CRC_ERR))
+		drm_dbg(&dev_priv->drm,
+			"PCH transcoder CRC error interrupt\n");
+
+	if (pch_iir & SDE_TRANSA_FIFO_UNDER)
+		intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_A);
+
+	if (pch_iir & SDE_TRANSB_FIFO_UNDER)
+		intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_B);
+}
+
+static void ivb_err_int_handler(struct drm_i915_private *dev_priv)
+{
+	u32 err_int = intel_uncore_read(&dev_priv->uncore, GEN7_ERR_INT);
+	enum pipe pipe;
+
+	if (err_int & ERR_INT_POISON)
+		drm_err(&dev_priv->drm, "Poison interrupt\n");
+
+	for_each_pipe(dev_priv, pipe) {
+		if (err_int & ERR_INT_FIFO_UNDERRUN(pipe))
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+
+		if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) {
+			if (IS_IVYBRIDGE(dev_priv))
+				ivb_pipe_crc_irq_handler(dev_priv, pipe);
+			else
+				hsw_pipe_crc_irq_handler(dev_priv, pipe);
+		}
+	}
+
+	intel_uncore_write(&dev_priv->uncore, GEN7_ERR_INT, err_int);
+}
+
+static void cpt_serr_int_handler(struct drm_i915_private *dev_priv)
+{
+	u32 serr_int = intel_uncore_read(&dev_priv->uncore, SERR_INT);
+	enum pipe pipe;
+
+	if (serr_int & SERR_INT_POISON)
+		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
+
+	for_each_pipe(dev_priv, pipe)
+		if (serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pipe))
+			intel_pch_fifo_underrun_irq_handler(dev_priv, pipe);
+
+	intel_uncore_write(&dev_priv->uncore, SERR_INT, serr_int);
+}
+
+static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
+{
+	enum pipe pipe;
+	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
+
+	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
+
+	if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) {
+		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >>
+			       SDE_AUDIO_POWER_SHIFT_CPT);
+		drm_dbg(&dev_priv->drm, "PCH audio power change on port %c\n",
+			port_name(port));
+	}
+
+	if (pch_iir & SDE_AUX_MASK_CPT)
+		intel_dp_aux_irq_handler(dev_priv);
+
+	if (pch_iir & SDE_GMBUS_CPT)
+		intel_gmbus_irq_handler(dev_priv);
+
+	if (pch_iir & SDE_AUDIO_CP_REQ_CPT)
+		drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n");
+
+	if (pch_iir & SDE_AUDIO_CP_CHG_CPT)
+		drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n");
+
+	if (pch_iir & SDE_FDI_MASK_CPT) {
+		for_each_pipe(dev_priv, pipe)
+			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
+				pipe_name(pipe),
+				intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe)));
+	}
+
+	if (pch_iir & SDE_ERROR_CPT)
+		cpt_serr_int_handler(dev_priv);
+}
+
+void ilk_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
+{
+	enum pipe pipe;
+	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG;
+
+	if (hotplug_trigger)
+		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+
+	if (de_iir & DE_AUX_CHANNEL_A)
+		intel_dp_aux_irq_handler(dev_priv);
+
+	if (de_iir & DE_GSE)
+		intel_opregion_asle_intr(dev_priv);
+
+	if (de_iir & DE_POISON)
+		drm_err(&dev_priv->drm, "Poison interrupt\n");
+
+	for_each_pipe(dev_priv, pipe) {
+		if (de_iir & DE_PIPE_VBLANK(pipe))
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (de_iir & DE_PLANE_FLIP_DONE(pipe))
+			flip_done_handler(dev_priv, pipe);
+
+		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+
+		if (de_iir & DE_PIPE_CRC_DONE(pipe))
+			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
+	}
+
+	/* check event from PCH */
+	if (de_iir & DE_PCH_EVENT) {
+		u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR);
+
+		if (HAS_PCH_CPT(dev_priv))
+			cpt_irq_handler(dev_priv, pch_iir);
+		else
+			ibx_irq_handler(dev_priv, pch_iir);
+
+		/* should clear PCH hotplug event before clear CPU irq */
+		intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir);
+	}
+
+	if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT)
+		gen5_rps_irq_handler(&to_gt(dev_priv)->rps);
+}
+
+void ivb_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir)
+{
+	enum pipe pipe;
+	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB;
+
+	if (hotplug_trigger)
+		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+
+	if (de_iir & DE_ERR_INT_IVB)
+		ivb_err_int_handler(dev_priv);
+
+	if (de_iir & DE_AUX_CHANNEL_A_IVB)
+		intel_dp_aux_irq_handler(dev_priv);
+
+	if (de_iir & DE_GSE_IVB)
+		intel_opregion_asle_intr(dev_priv);
+
+	for_each_pipe(dev_priv, pipe) {
+		if (de_iir & DE_PIPE_VBLANK_IVB(pipe))
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe))
+			flip_done_handler(dev_priv, pipe);
+	}
+
+	/* check event from PCH */
+	if (!HAS_PCH_NOP(dev_priv) && (de_iir & DE_PCH_EVENT_IVB)) {
+		u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR);
+
+		cpt_irq_handler(dev_priv, pch_iir);
+
+		/* clear PCH hotplug event before clear CPU irq */
+		intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir);
+	}
+}
+
+static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
+{
+	u32 mask;
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		return TGL_DE_PORT_AUX_DDIA |
+			TGL_DE_PORT_AUX_DDIB;
+	else if (DISPLAY_VER(dev_priv) >= 13)
+		return TGL_DE_PORT_AUX_DDIA |
+			TGL_DE_PORT_AUX_DDIB |
+			TGL_DE_PORT_AUX_DDIC |
+			XELPD_DE_PORT_AUX_DDID |
+			XELPD_DE_PORT_AUX_DDIE |
+			TGL_DE_PORT_AUX_USBC1 |
+			TGL_DE_PORT_AUX_USBC2 |
+			TGL_DE_PORT_AUX_USBC3 |
+			TGL_DE_PORT_AUX_USBC4;
+	else if (DISPLAY_VER(dev_priv) >= 12)
+		return TGL_DE_PORT_AUX_DDIA |
+			TGL_DE_PORT_AUX_DDIB |
+			TGL_DE_PORT_AUX_DDIC |
+			TGL_DE_PORT_AUX_USBC1 |
+			TGL_DE_PORT_AUX_USBC2 |
+			TGL_DE_PORT_AUX_USBC3 |
+			TGL_DE_PORT_AUX_USBC4 |
+			TGL_DE_PORT_AUX_USBC5 |
+			TGL_DE_PORT_AUX_USBC6;
+
+	mask = GEN8_AUX_CHANNEL_A;
+	if (DISPLAY_VER(dev_priv) >= 9)
+		mask |= GEN9_AUX_CHANNEL_B |
+			GEN9_AUX_CHANNEL_C |
+			GEN9_AUX_CHANNEL_D;
+
+	if (DISPLAY_VER(dev_priv) == 11) {
+		mask |= ICL_AUX_CHANNEL_F;
+		mask |= ICL_AUX_CHANNEL_E;
+	}
+
+	return mask;
+}
+
+static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv)
+{
+	if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv))
+		return RKL_DE_PIPE_IRQ_FAULT_ERRORS;
+	else if (DISPLAY_VER(dev_priv) >= 11)
+		return GEN11_DE_PIPE_IRQ_FAULT_ERRORS;
+	else if (DISPLAY_VER(dev_priv) >= 9)
+		return GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
+	else
+		return GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
+}
+
+static void
+gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
+{
+	bool found = false;
+
+	if (iir & GEN8_DE_MISC_GSE) {
+		intel_opregion_asle_intr(dev_priv);
+		found = true;
+	}
+
+	if (iir & GEN8_DE_EDP_PSR) {
+		struct intel_encoder *encoder;
+		u32 psr_iir;
+		i915_reg_t iir_reg;
+
+		for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) {
+			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+
+			if (DISPLAY_VER(dev_priv) >= 12)
+				iir_reg = TRANS_PSR_IIR(intel_dp->psr.transcoder);
+			else
+				iir_reg = EDP_PSR_IIR;
+
+			psr_iir = intel_uncore_rmw(&dev_priv->uncore, iir_reg, 0, 0);
+
+			if (psr_iir)
+				found = true;
+
+			intel_psr_irq_handler(intel_dp, psr_iir);
+
+			/* prior GEN12 only have one EDP PSR */
+			if (DISPLAY_VER(dev_priv) < 12)
+				break;
+		}
+	}
+
+	if (!found)
+		drm_err(&dev_priv->drm, "Unexpected DE Misc interrupt\n");
+}
+
+static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
+					   u32 te_trigger)
+{
+	enum pipe pipe = INVALID_PIPE;
+	enum transcoder dsi_trans;
+	enum port port;
+	u32 val, tmp;
+
+	/*
+	 * Incase of dual link, TE comes from DSI_1
+	 * this is to check if dual link is enabled
+	 */
+	val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL2(TRANSCODER_DSI_0));
+	val &= PORT_SYNC_MODE_ENABLE;
+
+	/*
+	 * if dual link is enabled, then read DSI_0
+	 * transcoder registers
+	 */
+	port = ((te_trigger & DSI1_TE && val) || (te_trigger & DSI0_TE)) ?
+						  PORT_A : PORT_B;
+	dsi_trans = (port == PORT_A) ? TRANSCODER_DSI_0 : TRANSCODER_DSI_1;
+
+	/* Check if DSI configured in command mode */
+	val = intel_uncore_read(&dev_priv->uncore, DSI_TRANS_FUNC_CONF(dsi_trans));
+	val = val & OP_MODE_MASK;
+
+	if (val != CMD_MODE_NO_GATE && val != CMD_MODE_TE_GATE) {
+		drm_err(&dev_priv->drm, "DSI trancoder not configured in command mode\n");
+		return;
+	}
+
+	/* Get PIPE for handling VBLANK event */
+	val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL(dsi_trans));
+	switch (val & TRANS_DDI_EDP_INPUT_MASK) {
+	case TRANS_DDI_EDP_INPUT_A_ON:
+		pipe = PIPE_A;
+		break;
+	case TRANS_DDI_EDP_INPUT_B_ONOFF:
+		pipe = PIPE_B;
+		break;
+	case TRANS_DDI_EDP_INPUT_C_ONOFF:
+		pipe = PIPE_C;
+		break;
+	default:
+		drm_err(&dev_priv->drm, "Invalid PIPE\n");
+		return;
+	}
+
+	intel_handle_vblank(dev_priv, pipe);
+
+	/* clear TE in dsi IIR */
+	port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A;
+	tmp = intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
+}
+
+static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915)
+{
+	if (DISPLAY_VER(i915) >= 9)
+		return GEN9_PIPE_PLANE1_FLIP_DONE;
+	else
+		return GEN8_PIPE_PRIMARY_FLIP_DONE;
+}
+
+u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv)
+{
+	u32 mask = GEN8_PIPE_FIFO_UNDERRUN;
+
+	if (DISPLAY_VER(dev_priv) >= 13)
+		mask |= XELPD_PIPE_SOFT_UNDERRUN |
+			XELPD_PIPE_HARD_UNDERRUN;
+
+	return mask;
+}
+
+static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_iir, u32 *pica_iir)
+{
+	u32 pica_ier = 0;
+
+	*pica_iir = 0;
+	*pch_iir = intel_de_read(i915, SDEIIR);
+	if (!*pch_iir)
+		return;
+
+	/**
+	 * PICA IER must be disabled/re-enabled around clearing PICA IIR and
+	 * SDEIIR, to avoid losing PICA IRQs and to ensure that such IRQs set
+	 * their flags both in the PICA and SDE IIR.
+	 */
+	if (*pch_iir & SDE_PICAINTERRUPT) {
+		drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTP);
+
+		pica_ier = intel_de_rmw(i915, PICAINTERRUPT_IER, ~0, 0);
+		*pica_iir = intel_de_read(i915, PICAINTERRUPT_IIR);
+		intel_de_write(i915, PICAINTERRUPT_IIR, *pica_iir);
+	}
+
+	intel_de_write(i915, SDEIIR, *pch_iir);
+
+	if (pica_ier)
+		intel_de_write(i915, PICAINTERRUPT_IER, pica_ier);
+}
+
+void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
+{
+	u32 iir;
+	enum pipe pipe;
+
+	drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_DISPLAY(dev_priv));
+
+	if (master_ctl & GEN8_DE_MISC_IRQ) {
+		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_MISC_IIR);
+		if (iir) {
+			intel_uncore_write(&dev_priv->uncore, GEN8_DE_MISC_IIR, iir);
+			gen8_de_misc_irq_handler(dev_priv, iir);
+		} else {
+			drm_err_ratelimited(&dev_priv->drm,
+					    "The master control interrupt lied (DE MISC)!\n");
+		}
+	}
+
+	if (DISPLAY_VER(dev_priv) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) {
+		iir = intel_uncore_read(&dev_priv->uncore, GEN11_DE_HPD_IIR);
+		if (iir) {
+			intel_uncore_write(&dev_priv->uncore, GEN11_DE_HPD_IIR, iir);
+			gen11_hpd_irq_handler(dev_priv, iir);
+		} else {
+			drm_err_ratelimited(&dev_priv->drm,
+					    "The master control interrupt lied, (DE HPD)!\n");
+		}
+	}
+
+	if (master_ctl & GEN8_DE_PORT_IRQ) {
+		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IIR);
+		if (iir) {
+			bool found = false;
+
+			intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IIR, iir);
+
+			if (iir & gen8_de_port_aux_mask(dev_priv)) {
+				intel_dp_aux_irq_handler(dev_priv);
+				found = true;
+			}
+
+			if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
+				u32 hotplug_trigger = iir & BXT_DE_PORT_HOTPLUG_MASK;
+
+				if (hotplug_trigger) {
+					bxt_hpd_irq_handler(dev_priv, hotplug_trigger);
+					found = true;
+				}
+			} else if (IS_BROADWELL(dev_priv)) {
+				u32 hotplug_trigger = iir & BDW_DE_PORT_HOTPLUG_MASK;
+
+				if (hotplug_trigger) {
+					ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
+					found = true;
+				}
+			}
+
+			if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
+			    (iir & BXT_DE_PORT_GMBUS)) {
+				intel_gmbus_irq_handler(dev_priv);
+				found = true;
+			}
+
+			if (DISPLAY_VER(dev_priv) >= 11) {
+				u32 te_trigger = iir & (DSI0_TE | DSI1_TE);
+
+				if (te_trigger) {
+					gen11_dsi_te_interrupt_handler(dev_priv, te_trigger);
+					found = true;
+				}
+			}
+
+			if (!found)
+				drm_err_ratelimited(&dev_priv->drm,
+						    "Unexpected DE Port interrupt\n");
+		} else {
+			drm_err_ratelimited(&dev_priv->drm,
+					    "The master control interrupt lied (DE PORT)!\n");
+		}
+	}
+
+	for_each_pipe(dev_priv, pipe) {
+		u32 fault_errors;
+
+		if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe)))
+			continue;
+
+		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe));
+		if (!iir) {
+			drm_err_ratelimited(&dev_priv->drm,
+					    "The master control interrupt lied (DE PIPE)!\n");
+			continue;
+		}
+
+		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe), iir);
+
+		if (iir & GEN8_PIPE_VBLANK)
+			intel_handle_vblank(dev_priv, pipe);
+
+		if (iir & gen8_de_pipe_flip_done_mask(dev_priv))
+			flip_done_handler(dev_priv, pipe);
+
+		if (iir & GEN8_PIPE_CDCLK_CRC_DONE)
+			hsw_pipe_crc_irq_handler(dev_priv, pipe);
+
+		if (iir & gen8_de_pipe_underrun_mask(dev_priv))
+			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
+
+		fault_errors = iir & gen8_de_pipe_fault_mask(dev_priv);
+		if (fault_errors)
+			drm_err_ratelimited(&dev_priv->drm,
+					    "Fault errors on pipe %c: 0x%08x\n",
+					    pipe_name(pipe),
+					    fault_errors);
+	}
+
+	if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) &&
+	    master_ctl & GEN8_DE_PCH_IRQ) {
+		u32 pica_iir;
+
+		/*
+		 * FIXME(BDW): Assume for now that the new interrupt handling
+		 * scheme also closed the SDE interrupt handling race we've seen
+		 * on older pch-split platforms. But this needs testing.
+		 */
+		gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir);
+		if (iir) {
+			if (pica_iir)
+				xelpdp_pica_irq_handler(dev_priv, pica_iir);
+
+			if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+				icp_irq_handler(dev_priv, iir);
+			else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
+				spt_irq_handler(dev_priv, iir);
+			else
+				cpt_irq_handler(dev_priv, iir);
+		} else {
+			/*
+			 * Like on previous PCH there seems to be something
+			 * fishy going on with forwarding PCH interrupts.
+			 */
+			drm_dbg(&dev_priv->drm,
+				"The master control interrupt lied (SDE)!\n");
+		}
+	}
+}
+
+u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl)
+{
+	void __iomem * const regs = i915->uncore.regs;
+	u32 iir;
+
+	if (!(master_ctl & GEN11_GU_MISC_IRQ))
+		return 0;
+
+	iir = raw_reg_read(regs, GEN11_GU_MISC_IIR);
+	if (likely(iir))
+		raw_reg_write(regs, GEN11_GU_MISC_IIR, iir);
+
+	return iir;
+}
+
+void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir)
+{
+	if (iir & GEN11_GU_MISC_GSE)
+		intel_opregion_asle_intr(i915);
+}
+
+void gen11_display_irq_handler(struct drm_i915_private *i915)
+{
+	void __iomem * const regs = i915->uncore.regs;
+	const u32 disp_ctl = raw_reg_read(regs, GEN11_DISPLAY_INT_CTL);
+
+	disable_rpm_wakeref_asserts(&i915->runtime_pm);
+	/*
+	 * GEN11_DISPLAY_INT_CTL has same format as GEN8_MASTER_IRQ
+	 * for the display related bits.
+	 */
+	raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, 0x0);
+	gen8_de_irq_handler(i915, disp_ctl);
+	raw_reg_write(regs, GEN11_DISPLAY_INT_CTL,
+		      GEN11_DISPLAY_IRQ_ENABLE);
+
+	enable_rpm_wakeref_asserts(&i915->runtime_pm);
+}
+
+/* Called from drm generic code, passed 'crtc' which
+ * we use as a pipe index
+ */
+int i8xx_enable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	return 0;
+}
+
+int i915gm_enable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+
+	/*
+	 * Vblank interrupts fail to wake the device up from C2+.
+	 * Disabling render clock gating during C-states avoids
+	 * the problem. There is a small power cost so we do this
+	 * only when vblank interrupts are actually enabled.
+	 */
+	if (dev_priv->vblank_enabled++ == 0)
+		intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_ENABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
+
+	return i8xx_enable_vblank(crtc);
+}
+
+int i965_enable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, pipe,
+			     PIPE_START_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	return 0;
+}
+
+int ilk_enable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
+		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	ilk_enable_display_irq(dev_priv, bit);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	/* Even though there is no DMC, frame counter can get stuck when
+	 * PSR is active as no frames are generated.
+	 */
+	if (HAS_PSR(dev_priv))
+		drm_crtc_vblank_restore(crtc);
+
+	return 0;
+}
+
+static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc,
+				   bool enable)
+{
+	struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev);
+	enum port port;
+
+	if (!(intel_crtc->mode_flags &
+	    (I915_MODE_FLAG_DSI_USE_TE1 | I915_MODE_FLAG_DSI_USE_TE0)))
+		return false;
+
+	/* for dual link cases we consider TE from slave */
+	if (intel_crtc->mode_flags & I915_MODE_FLAG_DSI_USE_TE1)
+		port = PORT_B;
+	else
+		port = PORT_A;
+
+	intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_MASK_REG(port), DSI_TE_EVENT,
+			 enable ? 0 : DSI_TE_EVENT);
+
+	intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
+
+	return true;
+}
+
+int bdw_enable_vblank(struct drm_crtc *_crtc)
+{
+	struct intel_crtc *crtc = to_intel_crtc(_crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	unsigned long irqflags;
+
+	if (gen11_dsi_configure_te(crtc, true))
+		return 0;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+
+	/* Even if there is no DMC, frame counter can get stuck when
+	 * PSR is active as no frames are generated, so check only for PSR.
+	 */
+	if (HAS_PSR(dev_priv))
+		drm_crtc_vblank_restore(&crtc->base);
+
+	return 0;
+}
+
+/* Called from drm generic code, passed 'crtc' which
+ * we use as a pipe index
+ */
+void i8xx_disable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+}
+
+void i915gm_disable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+
+	i8xx_disable_vblank(crtc);
+
+	if (--dev_priv->vblank_enabled == 0)
+		intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_DISABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
+}
+
+void i965_disable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_disable_pipestat(dev_priv, pipe,
+			      PIPE_START_VBLANK_INTERRUPT_STATUS);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+}
+
+void ilk_disable_vblank(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	enum pipe pipe = to_intel_crtc(crtc)->pipe;
+	unsigned long irqflags;
+	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
+		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	ilk_disable_display_irq(dev_priv, bit);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+}
+
+void bdw_disable_vblank(struct drm_crtc *_crtc)
+{
+	struct intel_crtc *crtc = to_intel_crtc(_crtc);
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	unsigned long irqflags;
+
+	if (gen11_dsi_configure_te(crtc, false))
+		return;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+}
+
+void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
+	if (IS_CHERRYVIEW(dev_priv))
+		intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
+	else
+		intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_VLV);
+
+	i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0);
+	intel_uncore_rmw(uncore, PORT_HOTPLUG_STAT, 0, 0);
+
+	i9xx_pipestat_irq_reset(dev_priv);
+
+	GEN3_IRQ_RESET(uncore, VLV_);
+	dev_priv->irq_mask = ~0u;
+}
+
+void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
+	u32 pipestat_mask;
+	u32 enable_mask;
+	enum pipe pipe;
+
+	pipestat_mask = PIPE_CRC_DONE_INTERRUPT_STATUS;
+
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	for_each_pipe(dev_priv, pipe)
+		i915_enable_pipestat(dev_priv, pipe, pipestat_mask);
+
+	enable_mask = I915_DISPLAY_PORT_INTERRUPT |
+		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
+		I915_LPE_PIPE_A_INTERRUPT |
+		I915_LPE_PIPE_B_INTERRUPT;
+
+	if (IS_CHERRYVIEW(dev_priv))
+		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT |
+			I915_LPE_PIPE_C_INTERRUPT;
+
+	drm_WARN_ON(&dev_priv->drm, dev_priv->irq_mask != ~0u);
+
+	dev_priv->irq_mask = ~enable_mask;
+
+	GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask);
+}
+
+void gen8_display_irq_reset(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	enum pipe pipe;
+
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff);
+	intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff);
+
+	for_each_pipe(dev_priv, pipe)
+		if (intel_display_power_is_enabled(dev_priv,
+						   POWER_DOMAIN_PIPE(pipe)))
+			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
+
+	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
+}
+
+void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	enum pipe pipe;
+	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
+
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	intel_uncore_write(uncore, GEN11_DISPLAY_INT_CTL, 0);
+
+	if (DISPLAY_VER(dev_priv) >= 12) {
+		enum transcoder trans;
+
+		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
+			enum intel_display_power_domain domain;
+
+			domain = POWER_DOMAIN_TRANSCODER(trans);
+			if (!intel_display_power_is_enabled(dev_priv, domain))
+				continue;
+
+			intel_uncore_write(uncore, TRANS_PSR_IMR(trans), 0xffffffff);
+			intel_uncore_write(uncore, TRANS_PSR_IIR(trans), 0xffffffff);
+		}
+	} else {
+		intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff);
+		intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff);
+	}
+
+	for_each_pipe(dev_priv, pipe)
+		if (intel_display_power_is_enabled(dev_priv,
+						   POWER_DOMAIN_PIPE(pipe)))
+			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
+
+	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
+	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		GEN3_IRQ_RESET(uncore, PICAINTERRUPT_);
+	else
+		GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_);
+
+	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
+		GEN3_IRQ_RESET(uncore, SDE);
+}
+
+void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
+				     u8 pipe_mask)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	u32 extra_ier = GEN8_PIPE_VBLANK |
+		gen8_de_pipe_underrun_mask(dev_priv) |
+		gen8_de_pipe_flip_done_mask(dev_priv);
+	enum pipe pipe;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+
+	if (!intel_irqs_enabled(dev_priv)) {
+		spin_unlock_irq(&dev_priv->irq_lock);
+		return;
+	}
+
+	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+		GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
+				  dev_priv->de_irq_mask[pipe],
+				  ~dev_priv->de_irq_mask[pipe] | extra_ier);
+
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
+void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
+				     u8 pipe_mask)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	enum pipe pipe;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+
+	if (!intel_irqs_enabled(dev_priv)) {
+		spin_unlock_irq(&dev_priv->irq_lock);
+		return;
+	}
+
+	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+		GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
+
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	/* make sure we're done processing display irqs */
+	intel_synchronize_irq(dev_priv);
+}
+
+/*
+ * SDEIER is also touched by the interrupt handler to work around missed PCH
+ * interrupts. Hence we can't update it after the interrupt handler is enabled -
+ * instead we unconditionally enable all PCH interrupt sources here, but then
+ * only unmask them as needed with SDEIMR.
+ *
+ * Note that we currently do this after installing the interrupt handler,
+ * but before we enable the master interrupt. That should be sufficient
+ * to avoid races with the irq handler, assuming we have MSI. Shared legacy
+ * interrupts could still race.
+ */
+void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	u32 mask;
+
+	if (HAS_PCH_NOP(dev_priv))
+		return;
+
+	if (HAS_PCH_IBX(dev_priv))
+		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
+	else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
+		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
+	else
+		mask = SDE_GMBUS_CPT;
+
+	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
+}
+
+void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
+{
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	if (dev_priv->display_irqs_enabled)
+		return;
+
+	dev_priv->display_irqs_enabled = true;
+
+	if (intel_irqs_enabled(dev_priv)) {
+		vlv_display_irq_reset(dev_priv);
+		vlv_display_irq_postinstall(dev_priv);
+	}
+}
+
+void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
+{
+	lockdep_assert_held(&dev_priv->irq_lock);
+
+	if (!dev_priv->display_irqs_enabled)
+		return;
+
+	dev_priv->display_irqs_enabled = false;
+
+	if (intel_irqs_enabled(dev_priv))
+		vlv_display_irq_reset(dev_priv);
+}
+
+void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+
+	u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) |
+		GEN8_PIPE_CDCLK_CRC_DONE;
+	u32 de_pipe_enables;
+	u32 de_port_masked = gen8_de_port_aux_mask(dev_priv);
+	u32 de_port_enables;
+	u32 de_misc_masked = GEN8_DE_EDP_PSR;
+	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
+	enum pipe pipe;
+
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	if (DISPLAY_VER(dev_priv) <= 10)
+		de_misc_masked |= GEN8_DE_MISC_GSE;
+
+	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+		de_port_masked |= BXT_DE_PORT_GMBUS;
+
+	if (DISPLAY_VER(dev_priv) >= 11) {
+		enum port port;
+
+		if (intel_bios_is_dsi_present(dev_priv, &port))
+			de_port_masked |= DSI0_TE | DSI1_TE;
+	}
+
+	de_pipe_enables = de_pipe_masked |
+		GEN8_PIPE_VBLANK |
+		gen8_de_pipe_underrun_mask(dev_priv) |
+		gen8_de_pipe_flip_done_mask(dev_priv);
+
+	de_port_enables = de_port_masked;
+	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
+		de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK;
+	else if (IS_BROADWELL(dev_priv))
+		de_port_enables |= BDW_DE_PORT_HOTPLUG_MASK;
+
+	if (DISPLAY_VER(dev_priv) >= 12) {
+		enum transcoder trans;
+
+		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
+			enum intel_display_power_domain domain;
+
+			domain = POWER_DOMAIN_TRANSCODER(trans);
+			if (!intel_display_power_is_enabled(dev_priv, domain))
+				continue;
+
+			gen3_assert_iir_is_zero(uncore, TRANS_PSR_IIR(trans));
+		}
+	} else {
+		gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR);
+	}
+
+	for_each_pipe(dev_priv, pipe) {
+		dev_priv->de_irq_mask[pipe] = ~de_pipe_masked;
+
+		if (intel_display_power_is_enabled(dev_priv,
+						   POWER_DOMAIN_PIPE(pipe)))
+			GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
+					  dev_priv->de_irq_mask[pipe],
+					  de_pipe_enables);
+	}
+
+	GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
+	GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked);
+
+	if (IS_DISPLAY_VER(dev_priv, 11, 13)) {
+		u32 de_hpd_masked = 0;
+		u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK |
+				     GEN11_DE_TBT_HOTPLUG_MASK;
+
+		GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked,
+			      de_hpd_enables);
+	}
+}
+
+void mtp_irq_postinstall(struct drm_i915_private *i915)
+{
+	struct intel_uncore *uncore = &i915->uncore;
+	u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT;
+	u32 de_hpd_mask = XELPDP_AUX_TC_MASK;
+	u32 de_hpd_enables = de_hpd_mask | XELPDP_DP_ALT_HOTPLUG_MASK |
+			     XELPDP_TBT_HOTPLUG_MASK;
+
+	GEN3_IRQ_INIT(uncore, PICAINTERRUPT_, ~de_hpd_mask,
+		      de_hpd_enables);
+
+	GEN3_IRQ_INIT(uncore, SDE, ~sde_mask, 0xffffffff);
+}
+
+void icp_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	struct intel_uncore *uncore = &dev_priv->uncore;
+	u32 mask = SDE_GMBUS_ICP;
+
+	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
+}
+
+void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv)
+{
+	if (!HAS_DISPLAY(dev_priv))
+		return;
+
+	gen8_de_irq_postinstall(dev_priv);
+
+	intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL,
+			   GEN11_DISPLAY_IRQ_ENABLE);
+}
+
diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.h b/drivers/gpu/drm/i915/display/intel_display_irq.h
new file mode 100644
index 0000000000000..874893f4f16db
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.h
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_DISPLAY_IRQ_H__
+#define __INTEL_DISPLAY_IRQ_H__
+
+#include <linux/types.h>
+
+#include "intel_display_limits.h"
+
+enum pipe;
+struct drm_i915_private;
+struct drm_crtc;
+
+void valleyview_enable_display_irqs(struct drm_i915_private *i915);
+void valleyview_disable_display_irqs(struct drm_i915_private *i915);
+
+void ilk_update_display_irq(struct drm_i915_private *i915,
+			    u32 interrupt_mask, u32 enabled_irq_mask);
+void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits);
+void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits);
+
+void bdw_update_port_irq(struct drm_i915_private *i915, u32 interrupt_mask, u32 enabled_irq_mask);
+void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
+void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
+
+void ibx_display_interrupt_update(struct drm_i915_private *i915,
+				  u32 interrupt_mask, u32 enabled_irq_mask);
+void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits);
+void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits);
+
+void gen8_irq_power_well_post_enable(struct drm_i915_private *i915, u8 pipe_mask);
+void gen8_irq_power_well_pre_disable(struct drm_i915_private *i915, u8 pipe_mask);
+u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *i915);
+
+int i8xx_enable_vblank(struct drm_crtc *crtc);
+int i915gm_enable_vblank(struct drm_crtc *crtc);
+int i965_enable_vblank(struct drm_crtc *crtc);
+int ilk_enable_vblank(struct drm_crtc *crtc);
+int bdw_enable_vblank(struct drm_crtc *crtc);
+void i8xx_disable_vblank(struct drm_crtc *crtc);
+void i915gm_disable_vblank(struct drm_crtc *crtc);
+void i965_disable_vblank(struct drm_crtc *crtc);
+void ilk_disable_vblank(struct drm_crtc *crtc);
+void bdw_disable_vblank(struct drm_crtc *crtc);
+
+void ivb_display_irq_handler(struct drm_i915_private *i915, u32 de_iir);
+void ilk_display_irq_handler(struct drm_i915_private *i915, u32 de_iir);
+void gen8_de_irq_handler(struct drm_i915_private *i915, u32 master_ctl);
+void gen11_display_irq_handler(struct drm_i915_private *i915);
+
+u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl);
+void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir);
+
+void vlv_display_irq_reset(struct drm_i915_private *i915);
+void gen8_display_irq_reset(struct drm_i915_private *i915);
+void gen11_display_irq_reset(struct drm_i915_private *i915);
+
+void ibx_irq_postinstall(struct drm_i915_private *i915);
+void vlv_display_irq_postinstall(struct drm_i915_private *i915);
+void icp_irq_postinstall(struct drm_i915_private *i915);
+void gen8_de_irq_postinstall(struct drm_i915_private *i915);
+void mtp_irq_postinstall(struct drm_i915_private *i915);
+void gen11_de_irq_postinstall(struct drm_i915_private *i915);
+
+u32 i915_pipestat_enable_mask(struct drm_i915_private *i915, enum pipe pipe);
+void i915_enable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask);
+void i915_disable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask);
+void i915_enable_asle_pipestat(struct drm_i915_private *i915);
+void i9xx_pipestat_irq_reset(struct drm_i915_private *i915);
+
+void i9xx_pipestat_irq_ack(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+
+void i915_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+void i965_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]);
+void valleyview_pipestat_irq_handler(struct drm_i915_private *i915, u32 pipe_stats[I915_MAX_PIPES]);
+void i8xx_pipestat_irq_handler(struct drm_i915_private *i915, u16 iir, u32 pipe_stats[I915_MAX_PIPES]);
+
+#endif /* __INTEL_DISPLAY_IRQ_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 41eabdf3e8710..916009894d89c 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -11,6 +11,7 @@
 #include "intel_combo_phy_regs.h"
 #include "intel_crt.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_power_well.h"
 #include "intel_display_types.h"
 #include "intel_dkl_phy.h"
diff --git a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
index e7f77a225739c..09a7fa6c0c375 100644
--- a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
+++ b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c
@@ -27,8 +27,8 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
-#include "i915_irq.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
 #include "intel_fbc.h"
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
index 1d7ae49e073ee..f95fa793fabba 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
@@ -4,9 +4,9 @@
  */
 
 #include "i915_drv.h"
-#include "i915_irq.h"
 #include "i915_reg.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_types.h"
 #include "intel_dp_aux.h"
 #include "intel_gmbus.h"
diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c
index 96fe4a2800770..36b479b46b600 100644
--- a/drivers/gpu/drm/i915/display/intel_tv.c
+++ b/drivers/gpu/drm/i915/display/intel_tv.c
@@ -35,11 +35,11 @@
 #include <drm/drm_edid.h>
 
 #include "i915_drv.h"
-#include "i915_irq.h"
 #include "i915_reg.h"
 #include "intel_connector.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_types.h"
 #include "intel_dpll.h"
 #include "intel_hotplug.h"
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index f6f760e59c9ef..110401aab038b 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -9,10 +9,10 @@
 #include <drm/drm_fourcc.h>
 
 #include "i915_drv.h"
-#include "i915_irq.h"
 #include "i915_reg.h"
 #include "intel_atomic_plane.h"
 #include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_types.h"
 #include "intel_fb.h"
 #include "intel_fbc.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index bc424d956cced..460090f066d4a 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -8,6 +8,7 @@
 #include <drm/i915_drm.h>
 
 #include "display/intel_display.h"
+#include "display/intel_display_irq.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 238f5a5a1527f..c2de0ae2d347f 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -33,18 +33,11 @@
 
 #include <drm/drm_drv.h>
 
-#include "display/icl_dsi_regs.h"
-#include "display/intel_de.h"
-#include "display/intel_display_trace.h"
+#include "display/intel_display_irq.h"
 #include "display/intel_display_types.h"
-#include "display/intel_dp_aux.h"
-#include "display/intel_fdi_regs.h"
-#include "display/intel_fifo_underrun.h"
-#include "display/intel_gmbus.h"
 #include "display/intel_hotplug.h"
 #include "display/intel_hotplug_irq.h"
 #include "display/intel_lpe_audio.h"
-#include "display/intel_psr.h"
 #include "display/intel_psr_regs.h"
 
 #include "gt/intel_breadcrumbs.h"
@@ -85,14 +78,6 @@ static inline void pmu_irq_stats(struct drm_i915_private *i915,
 	WRITE_ONCE(i915->pmu.irq_count, i915->pmu.irq_count + 1);
 }
 
-static void
-intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
-{
-	struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe);
-
-	drm_crtc_handle_vblank(&crtc->base);
-}
-
 void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr,
 		    i915_reg_t iir, i915_reg_t ier)
 {
@@ -125,7 +110,7 @@ static void gen2_irq_reset(struct intel_uncore *uncore)
 /*
  * We should clear IMR at preinstall/uninstall, and just check at postinstall.
  */
-static void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg)
+void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg)
 {
 	u32 val = intel_uncore_read(uncore, reg);
 
@@ -179,268 +164,6 @@ static void gen2_irq_init(struct intel_uncore *uncore,
 	intel_uncore_posting_read16(uncore, GEN2_IMR);
 }
 
-/**
- * ilk_update_display_irq - update DEIMR
- * @dev_priv: driver private
- * @interrupt_mask: mask of interrupt bits to update
- * @enabled_irq_mask: mask of interrupt bits to enable
- */
-void ilk_update_display_irq(struct drm_i915_private *dev_priv,
-			    u32 interrupt_mask, u32 enabled_irq_mask)
-{
-	u32 new_val;
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
-
-	new_val = dev_priv->irq_mask;
-	new_val &= ~interrupt_mask;
-	new_val |= (~enabled_irq_mask & interrupt_mask);
-
-	if (new_val != dev_priv->irq_mask &&
-	    !drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) {
-		dev_priv->irq_mask = new_val;
-		intel_uncore_write(&dev_priv->uncore, DEIMR, dev_priv->irq_mask);
-		intel_uncore_posting_read(&dev_priv->uncore, DEIMR);
-	}
-}
-
-void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits)
-{
-	ilk_update_display_irq(i915, bits, bits);
-}
-
-void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits)
-{
-	ilk_update_display_irq(i915, bits, 0);
-}
-
-/**
- * bdw_update_port_irq - update DE port interrupt
- * @dev_priv: driver private
- * @interrupt_mask: mask of interrupt bits to update
- * @enabled_irq_mask: mask of interrupt bits to enable
- */
-void bdw_update_port_irq(struct drm_i915_private *dev_priv,
-			 u32 interrupt_mask, u32 enabled_irq_mask)
-{
-	u32 new_val;
-	u32 old_val;
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
-
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
-		return;
-
-	old_val = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IMR);
-
-	new_val = old_val;
-	new_val &= ~interrupt_mask;
-	new_val |= (~enabled_irq_mask & interrupt_mask);
-
-	if (new_val != old_val) {
-		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IMR, new_val);
-		intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PORT_IMR);
-	}
-}
-
-/**
- * bdw_update_pipe_irq - update DE pipe interrupt
- * @dev_priv: driver private
- * @pipe: pipe whose interrupt to update
- * @interrupt_mask: mask of interrupt bits to update
- * @enabled_irq_mask: mask of interrupt bits to enable
- */
-static void bdw_update_pipe_irq(struct drm_i915_private *dev_priv,
-				enum pipe pipe, u32 interrupt_mask,
-				u32 enabled_irq_mask)
-{
-	u32 new_val;
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
-
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
-		return;
-
-	new_val = dev_priv->de_irq_mask[pipe];
-	new_val &= ~interrupt_mask;
-	new_val |= (~enabled_irq_mask & interrupt_mask);
-
-	if (new_val != dev_priv->de_irq_mask[pipe]) {
-		dev_priv->de_irq_mask[pipe] = new_val;
-		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
-		intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe));
-	}
-}
-
-void bdw_enable_pipe_irq(struct drm_i915_private *i915,
-			 enum pipe pipe, u32 bits)
-{
-	bdw_update_pipe_irq(i915, pipe, bits, bits);
-}
-
-void bdw_disable_pipe_irq(struct drm_i915_private *i915,
-			  enum pipe pipe, u32 bits)
-{
-	bdw_update_pipe_irq(i915, pipe, bits, 0);
-}
-
-/**
- * ibx_display_interrupt_update - update SDEIMR
- * @dev_priv: driver private
- * @interrupt_mask: mask of interrupt bits to update
- * @enabled_irq_mask: mask of interrupt bits to enable
- */
-void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
-				  u32 interrupt_mask,
-				  u32 enabled_irq_mask)
-{
-	u32 sdeimr = intel_uncore_read(&dev_priv->uncore, SDEIMR);
-	sdeimr &= ~interrupt_mask;
-	sdeimr |= (~enabled_irq_mask & interrupt_mask);
-
-	drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask);
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)))
-		return;
-
-	intel_uncore_write(&dev_priv->uncore, SDEIMR, sdeimr);
-	intel_uncore_posting_read(&dev_priv->uncore, SDEIMR);
-}
-
-void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits)
-{
-	ibx_display_interrupt_update(i915, bits, bits);
-}
-
-void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits)
-{
-	ibx_display_interrupt_update(i915, bits, 0);
-}
-
-u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv,
-			      enum pipe pipe)
-{
-	u32 status_mask = dev_priv->pipestat_irq_mask[pipe];
-	u32 enable_mask = status_mask << 16;
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	if (DISPLAY_VER(dev_priv) < 5)
-		goto out;
-
-	/*
-	 * On pipe A we don't support the PSR interrupt yet,
-	 * on pipe B and C the same bit MBZ.
-	 */
-	if (drm_WARN_ON_ONCE(&dev_priv->drm,
-			     status_mask & PIPE_A_PSR_STATUS_VLV))
-		return 0;
-	/*
-	 * On pipe B and C we don't support the PSR interrupt yet, on pipe
-	 * A the same bit is for perf counters which we don't use either.
-	 */
-	if (drm_WARN_ON_ONCE(&dev_priv->drm,
-			     status_mask & PIPE_B_PSR_STATUS_VLV))
-		return 0;
-
-	enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS |
-			 SPRITE0_FLIP_DONE_INT_EN_VLV |
-			 SPRITE1_FLIP_DONE_INT_EN_VLV);
-	if (status_mask & SPRITE0_FLIP_DONE_INT_STATUS_VLV)
-		enable_mask |= SPRITE0_FLIP_DONE_INT_EN_VLV;
-	if (status_mask & SPRITE1_FLIP_DONE_INT_STATUS_VLV)
-		enable_mask |= SPRITE1_FLIP_DONE_INT_EN_VLV;
-
-out:
-	drm_WARN_ONCE(&dev_priv->drm,
-		      enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
-		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
-		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
-		      pipe_name(pipe), enable_mask, status_mask);
-
-	return enable_mask;
-}
-
-void i915_enable_pipestat(struct drm_i915_private *dev_priv,
-			  enum pipe pipe, u32 status_mask)
-{
-	i915_reg_t reg = PIPESTAT(pipe);
-	u32 enable_mask;
-
-	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
-		      "pipe %c: status_mask=0x%x\n",
-		      pipe_name(pipe), status_mask);
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
-
-	if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == status_mask)
-		return;
-
-	dev_priv->pipestat_irq_mask[pipe] |= status_mask;
-	enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
-
-	intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask);
-	intel_uncore_posting_read(&dev_priv->uncore, reg);
-}
-
-void i915_disable_pipestat(struct drm_i915_private *dev_priv,
-			   enum pipe pipe, u32 status_mask)
-{
-	i915_reg_t reg = PIPESTAT(pipe);
-	u32 enable_mask;
-
-	drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK,
-		      "pipe %c: status_mask=0x%x\n",
-		      pipe_name(pipe), status_mask);
-
-	lockdep_assert_held(&dev_priv->irq_lock);
-	drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv));
-
-	if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == 0)
-		return;
-
-	dev_priv->pipestat_irq_mask[pipe] &= ~status_mask;
-	enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
-
-	intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask);
-	intel_uncore_posting_read(&dev_priv->uncore, reg);
-}
-
-static bool i915_has_asle(struct drm_i915_private *dev_priv)
-{
-	if (!dev_priv->display.opregion.asle)
-		return false;
-
-	return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv);
-}
-
-/**
- * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion
- * @dev_priv: i915 device private
- */
-static void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv)
-{
-	if (!i915_has_asle(dev_priv))
-		return;
-
-	spin_lock_irq(&dev_priv->irq_lock);
-
-	i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS);
-	if (DISPLAY_VER(dev_priv) >= 4)
-		i915_enable_pipestat(dev_priv, PIPE_A,
-				     PIPE_LEGACY_BLC_EVENT_STATUS);
-
-	spin_unlock_irq(&dev_priv->irq_lock);
-}
-
 /**
  * ivb_parity_work - Workqueue called when a parity error interrupt
  * occurred.
@@ -525,278 +248,6 @@ static void ivb_parity_work(struct work_struct *work)
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 }
 
-#if defined(CONFIG_DEBUG_FS)
-static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
-					 enum pipe pipe,
-					 u32 crc0, u32 crc1,
-					 u32 crc2, u32 crc3,
-					 u32 crc4)
-{
-	struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe);
-	struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc;
-	u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 };
-
-	trace_intel_pipe_crc(crtc, crcs);
-
-	spin_lock(&pipe_crc->lock);
-	/*
-	 * For some not yet identified reason, the first CRC is
-	 * bonkers. So let's just wait for the next vblank and read
-	 * out the buggy result.
-	 *
-	 * On GEN8+ sometimes the second CRC is bonkers as well, so
-	 * don't trust that one either.
-	 */
-	if (pipe_crc->skipped <= 0 ||
-	    (DISPLAY_VER(dev_priv) >= 8 && pipe_crc->skipped == 1)) {
-		pipe_crc->skipped++;
-		spin_unlock(&pipe_crc->lock);
-		return;
-	}
-	spin_unlock(&pipe_crc->lock);
-
-	drm_crtc_add_crc_entry(&crtc->base, true,
-				drm_crtc_accurate_vblank_count(&crtc->base),
-				crcs);
-}
-#else
-static inline void
-display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
-			     enum pipe pipe,
-			     u32 crc0, u32 crc1,
-			     u32 crc2, u32 crc3,
-			     u32 crc4) {}
-#endif
-
-static void flip_done_handler(struct drm_i915_private *i915,
-			      enum pipe pipe)
-{
-	struct intel_crtc *crtc = intel_crtc_for_pipe(i915, pipe);
-	struct drm_crtc_state *crtc_state = crtc->base.state;
-	struct drm_pending_vblank_event *e = crtc_state->event;
-	struct drm_device *dev = &i915->drm;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev->event_lock, irqflags);
-
-	crtc_state->event = NULL;
-
-	drm_crtc_send_vblank_event(&crtc->base, e);
-
-	spin_unlock_irqrestore(&dev->event_lock, irqflags);
-}
-
-static void hsw_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
-				     enum pipe pipe)
-{
-	display_pipe_crc_irq_handler(dev_priv, pipe,
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)),
-				     0, 0, 0, 0);
-}
-
-static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
-				     enum pipe pipe)
-{
-	display_pipe_crc_irq_handler(dev_priv, pipe,
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_2_IVB(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_3_IVB(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_4_IVB(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_5_IVB(pipe)));
-}
-
-static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv,
-				      enum pipe pipe)
-{
-	u32 res1, res2;
-
-	if (DISPLAY_VER(dev_priv) >= 3)
-		res1 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES1_I915(pipe));
-	else
-		res1 = 0;
-
-	if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv))
-		res2 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES2_G4X(pipe));
-	else
-		res2 = 0;
-
-	display_pipe_crc_irq_handler(dev_priv, pipe,
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RED(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_GREEN(pipe)),
-				     intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_BLUE(pipe)),
-				     res1, res2);
-}
-
-static void i9xx_pipestat_irq_reset(struct drm_i915_private *dev_priv)
-{
-	enum pipe pipe;
-
-	for_each_pipe(dev_priv, pipe) {
-		intel_uncore_write(&dev_priv->uncore, PIPESTAT(pipe),
-			   PIPESTAT_INT_STATUS_MASK |
-			   PIPE_FIFO_UNDERRUN_STATUS);
-
-		dev_priv->pipestat_irq_mask[pipe] = 0;
-	}
-}
-
-static void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv,
-				  u32 iir, u32 pipe_stats[I915_MAX_PIPES])
-{
-	enum pipe pipe;
-
-	spin_lock(&dev_priv->irq_lock);
-
-	if (!dev_priv->display_irqs_enabled) {
-		spin_unlock(&dev_priv->irq_lock);
-		return;
-	}
-
-	for_each_pipe(dev_priv, pipe) {
-		i915_reg_t reg;
-		u32 status_mask, enable_mask, iir_bit = 0;
-
-		/*
-		 * PIPESTAT bits get signalled even when the interrupt is
-		 * disabled with the mask bits, and some of the status bits do
-		 * not generate interrupts at all (like the underrun bit). Hence
-		 * we need to be careful that we only handle what we want to
-		 * handle.
-		 */
-
-		/* fifo underruns are filterered in the underrun handler. */
-		status_mask = PIPE_FIFO_UNDERRUN_STATUS;
-
-		switch (pipe) {
-		default:
-		case PIPE_A:
-			iir_bit = I915_DISPLAY_PIPE_A_EVENT_INTERRUPT;
-			break;
-		case PIPE_B:
-			iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
-			break;
-		case PIPE_C:
-			iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
-			break;
-		}
-		if (iir & iir_bit)
-			status_mask |= dev_priv->pipestat_irq_mask[pipe];
-
-		if (!status_mask)
-			continue;
-
-		reg = PIPESTAT(pipe);
-		pipe_stats[pipe] = intel_uncore_read(&dev_priv->uncore, reg) & status_mask;
-		enable_mask = i915_pipestat_enable_mask(dev_priv, pipe);
-
-		/*
-		 * Clear the PIPE*STAT regs before the IIR
-		 *
-		 * Toggle the enable bits to make sure we get an
-		 * edge in the ISR pipe event bit if we don't clear
-		 * all the enabled status bits. Otherwise the edge
-		 * triggered IIR on i965/g4x wouldn't notice that
-		 * an interrupt is still pending.
-		 */
-		if (pipe_stats[pipe]) {
-			intel_uncore_write(&dev_priv->uncore, reg, pipe_stats[pipe]);
-			intel_uncore_write(&dev_priv->uncore, reg, enable_mask);
-		}
-	}
-	spin_unlock(&dev_priv->irq_lock);
-}
-
-static void i8xx_pipestat_irq_handler(struct drm_i915_private *dev_priv,
-				      u16 iir, u32 pipe_stats[I915_MAX_PIPES])
-{
-	enum pipe pipe;
-
-	for_each_pipe(dev_priv, pipe) {
-		if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-	}
-}
-
-static void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv,
-				      u32 iir, u32 pipe_stats[I915_MAX_PIPES])
-{
-	bool blc_event = false;
-	enum pipe pipe;
-
-	for_each_pipe(dev_priv, pipe) {
-		if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
-			blc_event = true;
-
-		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-	}
-
-	if (blc_event || (iir & I915_ASLE_INTERRUPT))
-		intel_opregion_asle_intr(dev_priv);
-}
-
-static void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv,
-				      u32 iir, u32 pipe_stats[I915_MAX_PIPES])
-{
-	bool blc_event = false;
-	enum pipe pipe;
-
-	for_each_pipe(dev_priv, pipe) {
-		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
-			blc_event = true;
-
-		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-	}
-
-	if (blc_event || (iir & I915_ASLE_INTERRUPT))
-		intel_opregion_asle_intr(dev_priv);
-
-	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
-		intel_gmbus_irq_handler(dev_priv);
-}
-
-static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv,
-					    u32 pipe_stats[I915_MAX_PIPES])
-{
-	enum pipe pipe;
-
-	for_each_pipe(dev_priv, pipe) {
-		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV)
-			flip_done_handler(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS)
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
-
-		if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS)
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-	}
-
-	if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS)
-		intel_gmbus_irq_handler(dev_priv);
-}
-
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *dev_priv = arg;
@@ -961,217 +412,6 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
-{
-	enum pipe pipe;
-	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
-
-	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
-
-	if (pch_iir & SDE_AUDIO_POWER_MASK) {
-		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >>
-			       SDE_AUDIO_POWER_SHIFT);
-		drm_dbg(&dev_priv->drm, "PCH audio power change on port %d\n",
-			port_name(port));
-	}
-
-	if (pch_iir & SDE_AUX_MASK)
-		intel_dp_aux_irq_handler(dev_priv);
-
-	if (pch_iir & SDE_GMBUS)
-		intel_gmbus_irq_handler(dev_priv);
-
-	if (pch_iir & SDE_AUDIO_HDCP_MASK)
-		drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n");
-
-	if (pch_iir & SDE_AUDIO_TRANS_MASK)
-		drm_dbg(&dev_priv->drm, "PCH transcoder audio interrupt\n");
-
-	if (pch_iir & SDE_POISON)
-		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
-
-	if (pch_iir & SDE_FDI_MASK) {
-		for_each_pipe(dev_priv, pipe)
-			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
-				pipe_name(pipe),
-				intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe)));
-	}
-
-	if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE))
-		drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n");
-
-	if (pch_iir & (SDE_TRANSB_CRC_ERR | SDE_TRANSA_CRC_ERR))
-		drm_dbg(&dev_priv->drm,
-			"PCH transcoder CRC error interrupt\n");
-
-	if (pch_iir & SDE_TRANSA_FIFO_UNDER)
-		intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_A);
-
-	if (pch_iir & SDE_TRANSB_FIFO_UNDER)
-		intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_B);
-}
-
-static void ivb_err_int_handler(struct drm_i915_private *dev_priv)
-{
-	u32 err_int = intel_uncore_read(&dev_priv->uncore, GEN7_ERR_INT);
-	enum pipe pipe;
-
-	if (err_int & ERR_INT_POISON)
-		drm_err(&dev_priv->drm, "Poison interrupt\n");
-
-	for_each_pipe(dev_priv, pipe) {
-		if (err_int & ERR_INT_FIFO_UNDERRUN(pipe))
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-
-		if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) {
-			if (IS_IVYBRIDGE(dev_priv))
-				ivb_pipe_crc_irq_handler(dev_priv, pipe);
-			else
-				hsw_pipe_crc_irq_handler(dev_priv, pipe);
-		}
-	}
-
-	intel_uncore_write(&dev_priv->uncore, GEN7_ERR_INT, err_int);
-}
-
-static void cpt_serr_int_handler(struct drm_i915_private *dev_priv)
-{
-	u32 serr_int = intel_uncore_read(&dev_priv->uncore, SERR_INT);
-	enum pipe pipe;
-
-	if (serr_int & SERR_INT_POISON)
-		drm_err(&dev_priv->drm, "PCH poison interrupt\n");
-
-	for_each_pipe(dev_priv, pipe)
-		if (serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pipe))
-			intel_pch_fifo_underrun_irq_handler(dev_priv, pipe);
-
-	intel_uncore_write(&dev_priv->uncore, SERR_INT, serr_int);
-}
-
-static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir)
-{
-	enum pipe pipe;
-	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
-
-	ibx_hpd_irq_handler(dev_priv, hotplug_trigger);
-
-	if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) {
-		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >>
-			       SDE_AUDIO_POWER_SHIFT_CPT);
-		drm_dbg(&dev_priv->drm, "PCH audio power change on port %c\n",
-			port_name(port));
-	}
-
-	if (pch_iir & SDE_AUX_MASK_CPT)
-		intel_dp_aux_irq_handler(dev_priv);
-
-	if (pch_iir & SDE_GMBUS_CPT)
-		intel_gmbus_irq_handler(dev_priv);
-
-	if (pch_iir & SDE_AUDIO_CP_REQ_CPT)
-		drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n");
-
-	if (pch_iir & SDE_AUDIO_CP_CHG_CPT)
-		drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n");
-
-	if (pch_iir & SDE_FDI_MASK_CPT) {
-		for_each_pipe(dev_priv, pipe)
-			drm_dbg(&dev_priv->drm, "  pipe %c FDI IIR: 0x%08x\n",
-				pipe_name(pipe),
-				intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe)));
-	}
-
-	if (pch_iir & SDE_ERROR_CPT)
-		cpt_serr_int_handler(dev_priv);
-}
-
-static void ilk_display_irq_handler(struct drm_i915_private *dev_priv,
-				    u32 de_iir)
-{
-	enum pipe pipe;
-	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG;
-
-	if (hotplug_trigger)
-		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
-
-	if (de_iir & DE_AUX_CHANNEL_A)
-		intel_dp_aux_irq_handler(dev_priv);
-
-	if (de_iir & DE_GSE)
-		intel_opregion_asle_intr(dev_priv);
-
-	if (de_iir & DE_POISON)
-		drm_err(&dev_priv->drm, "Poison interrupt\n");
-
-	for_each_pipe(dev_priv, pipe) {
-		if (de_iir & DE_PIPE_VBLANK(pipe))
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (de_iir & DE_PLANE_FLIP_DONE(pipe))
-			flip_done_handler(dev_priv, pipe);
-
-		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-
-		if (de_iir & DE_PIPE_CRC_DONE(pipe))
-			i9xx_pipe_crc_irq_handler(dev_priv, pipe);
-	}
-
-	/* check event from PCH */
-	if (de_iir & DE_PCH_EVENT) {
-		u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR);
-
-		if (HAS_PCH_CPT(dev_priv))
-			cpt_irq_handler(dev_priv, pch_iir);
-		else
-			ibx_irq_handler(dev_priv, pch_iir);
-
-		/* should clear PCH hotplug event before clear CPU irq */
-		intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir);
-	}
-
-	if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT)
-		gen5_rps_irq_handler(&to_gt(dev_priv)->rps);
-}
-
-static void ivb_display_irq_handler(struct drm_i915_private *dev_priv,
-				    u32 de_iir)
-{
-	enum pipe pipe;
-	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB;
-
-	if (hotplug_trigger)
-		ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
-
-	if (de_iir & DE_ERR_INT_IVB)
-		ivb_err_int_handler(dev_priv);
-
-	if (de_iir & DE_AUX_CHANNEL_A_IVB)
-		intel_dp_aux_irq_handler(dev_priv);
-
-	if (de_iir & DE_GSE_IVB)
-		intel_opregion_asle_intr(dev_priv);
-
-	for_each_pipe(dev_priv, pipe) {
-		if (de_iir & DE_PIPE_VBLANK_IVB(pipe))
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe))
-			flip_done_handler(dev_priv, pipe);
-	}
-
-	/* check event from PCH */
-	if (!HAS_PCH_NOP(dev_priv) && (de_iir & DE_PCH_EVENT_IVB)) {
-		u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR);
-
-		cpt_irq_handler(dev_priv, pch_iir);
-
-		/* clear PCH hotplug event before clear CPU irq */
-		intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir);
-	}
-}
-
 /*
  * To handle irqs with the minimum potential races with fresh interrupts, we:
  * 1 - Disable Master Interrupt Control.
@@ -1246,353 +486,8 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg)
 
 	/* IRQs are synced during runtime_suspend, we don't require a wakeref */
 	enable_rpm_wakeref_asserts(&i915->runtime_pm);
-
-	return ret;
-}
-
-static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv)
-{
-	u32 mask;
-
-	if (DISPLAY_VER(dev_priv) >= 14)
-		return TGL_DE_PORT_AUX_DDIA |
-			TGL_DE_PORT_AUX_DDIB;
-	else if (DISPLAY_VER(dev_priv) >= 13)
-		return TGL_DE_PORT_AUX_DDIA |
-			TGL_DE_PORT_AUX_DDIB |
-			TGL_DE_PORT_AUX_DDIC |
-			XELPD_DE_PORT_AUX_DDID |
-			XELPD_DE_PORT_AUX_DDIE |
-			TGL_DE_PORT_AUX_USBC1 |
-			TGL_DE_PORT_AUX_USBC2 |
-			TGL_DE_PORT_AUX_USBC3 |
-			TGL_DE_PORT_AUX_USBC4;
-	else if (DISPLAY_VER(dev_priv) >= 12)
-		return TGL_DE_PORT_AUX_DDIA |
-			TGL_DE_PORT_AUX_DDIB |
-			TGL_DE_PORT_AUX_DDIC |
-			TGL_DE_PORT_AUX_USBC1 |
-			TGL_DE_PORT_AUX_USBC2 |
-			TGL_DE_PORT_AUX_USBC3 |
-			TGL_DE_PORT_AUX_USBC4 |
-			TGL_DE_PORT_AUX_USBC5 |
-			TGL_DE_PORT_AUX_USBC6;
-
-
-	mask = GEN8_AUX_CHANNEL_A;
-	if (DISPLAY_VER(dev_priv) >= 9)
-		mask |= GEN9_AUX_CHANNEL_B |
-			GEN9_AUX_CHANNEL_C |
-			GEN9_AUX_CHANNEL_D;
-
-	if (DISPLAY_VER(dev_priv) == 11) {
-		mask |= ICL_AUX_CHANNEL_F;
-		mask |= ICL_AUX_CHANNEL_E;
-	}
-
-	return mask;
-}
-
-static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv)
-{
-	if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv))
-		return RKL_DE_PIPE_IRQ_FAULT_ERRORS;
-	else if (DISPLAY_VER(dev_priv) >= 11)
-		return GEN11_DE_PIPE_IRQ_FAULT_ERRORS;
-	else if (DISPLAY_VER(dev_priv) >= 9)
-		return GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
-	else
-		return GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
-}
-
-static void
-gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
-{
-	bool found = false;
-
-	if (iir & GEN8_DE_MISC_GSE) {
-		intel_opregion_asle_intr(dev_priv);
-		found = true;
-	}
-
-	if (iir & GEN8_DE_EDP_PSR) {
-		struct intel_encoder *encoder;
-		u32 psr_iir;
-		i915_reg_t iir_reg;
-
-		for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) {
-			struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-
-			if (DISPLAY_VER(dev_priv) >= 12)
-				iir_reg = TRANS_PSR_IIR(intel_dp->psr.transcoder);
-			else
-				iir_reg = EDP_PSR_IIR;
-
-			psr_iir = intel_uncore_rmw(&dev_priv->uncore, iir_reg, 0, 0);
-
-			if (psr_iir)
-				found = true;
-
-			intel_psr_irq_handler(intel_dp, psr_iir);
-
-			/* prior GEN12 only have one EDP PSR */
-			if (DISPLAY_VER(dev_priv) < 12)
-				break;
-		}
-	}
-
-	if (!found)
-		drm_err(&dev_priv->drm, "Unexpected DE Misc interrupt\n");
-}
-
-static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
-					   u32 te_trigger)
-{
-	enum pipe pipe = INVALID_PIPE;
-	enum transcoder dsi_trans;
-	enum port port;
-	u32 val, tmp;
-
-	/*
-	 * Incase of dual link, TE comes from DSI_1
-	 * this is to check if dual link is enabled
-	 */
-	val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL2(TRANSCODER_DSI_0));
-	val &= PORT_SYNC_MODE_ENABLE;
-
-	/*
-	 * if dual link is enabled, then read DSI_0
-	 * transcoder registers
-	 */
-	port = ((te_trigger & DSI1_TE && val) || (te_trigger & DSI0_TE)) ?
-						  PORT_A : PORT_B;
-	dsi_trans = (port == PORT_A) ? TRANSCODER_DSI_0 : TRANSCODER_DSI_1;
-
-	/* Check if DSI configured in command mode */
-	val = intel_uncore_read(&dev_priv->uncore, DSI_TRANS_FUNC_CONF(dsi_trans));
-	val = val & OP_MODE_MASK;
-
-	if (val != CMD_MODE_NO_GATE && val != CMD_MODE_TE_GATE) {
-		drm_err(&dev_priv->drm, "DSI trancoder not configured in command mode\n");
-		return;
-	}
-
-	/* Get PIPE for handling VBLANK event */
-	val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL(dsi_trans));
-	switch (val & TRANS_DDI_EDP_INPUT_MASK) {
-	case TRANS_DDI_EDP_INPUT_A_ON:
-		pipe = PIPE_A;
-		break;
-	case TRANS_DDI_EDP_INPUT_B_ONOFF:
-		pipe = PIPE_B;
-		break;
-	case TRANS_DDI_EDP_INPUT_C_ONOFF:
-		pipe = PIPE_C;
-		break;
-	default:
-		drm_err(&dev_priv->drm, "Invalid PIPE\n");
-		return;
-	}
-
-	intel_handle_vblank(dev_priv, pipe);
-
-	/* clear TE in dsi IIR */
-	port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A;
-	tmp = intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
-}
-
-static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915)
-{
-	if (DISPLAY_VER(i915) >= 9)
-		return GEN9_PIPE_PLANE1_FLIP_DONE;
-	else
-		return GEN8_PIPE_PRIMARY_FLIP_DONE;
-}
-
-u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv)
-{
-	u32 mask = GEN8_PIPE_FIFO_UNDERRUN;
-
-	if (DISPLAY_VER(dev_priv) >= 13)
-		mask |= XELPD_PIPE_SOFT_UNDERRUN |
-			XELPD_PIPE_HARD_UNDERRUN;
-
-	return mask;
-}
-
-static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_iir, u32 *pica_iir)
-{
-	u32 pica_ier = 0;
-
-	*pica_iir = 0;
-	*pch_iir = intel_de_read(i915, SDEIIR);
-	if (!*pch_iir)
-		return;
-
-	/**
-	 * PICA IER must be disabled/re-enabled around clearing PICA IIR and
-	 * SDEIIR, to avoid losing PICA IRQs and to ensure that such IRQs set
-	 * their flags both in the PICA and SDE IIR.
-	 */
-	if (*pch_iir & SDE_PICAINTERRUPT) {
-		drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTP);
-
-		pica_ier = intel_de_rmw(i915, PICAINTERRUPT_IER, ~0, 0);
-		*pica_iir = intel_de_read(i915, PICAINTERRUPT_IIR);
-		intel_de_write(i915, PICAINTERRUPT_IIR, *pica_iir);
-	}
-
-	intel_de_write(i915, SDEIIR, *pch_iir);
-
-	if (pica_ier)
-		intel_de_write(i915, PICAINTERRUPT_IER, pica_ier);
-}
-
-static void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl)
-{
-	u32 iir;
-	enum pipe pipe;
-
-	drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_DISPLAY(dev_priv));
-
-	if (master_ctl & GEN8_DE_MISC_IRQ) {
-		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_MISC_IIR);
-		if (iir) {
-			intel_uncore_write(&dev_priv->uncore, GEN8_DE_MISC_IIR, iir);
-			gen8_de_misc_irq_handler(dev_priv, iir);
-		} else {
-			drm_err_ratelimited(&dev_priv->drm,
-					    "The master control interrupt lied (DE MISC)!\n");
-		}
-	}
-
-	if (DISPLAY_VER(dev_priv) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) {
-		iir = intel_uncore_read(&dev_priv->uncore, GEN11_DE_HPD_IIR);
-		if (iir) {
-			intel_uncore_write(&dev_priv->uncore, GEN11_DE_HPD_IIR, iir);
-			gen11_hpd_irq_handler(dev_priv, iir);
-		} else {
-			drm_err_ratelimited(&dev_priv->drm,
-					    "The master control interrupt lied, (DE HPD)!\n");
-		}
-	}
-
-	if (master_ctl & GEN8_DE_PORT_IRQ) {
-		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IIR);
-		if (iir) {
-			bool found = false;
-
-			intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IIR, iir);
-
-			if (iir & gen8_de_port_aux_mask(dev_priv)) {
-				intel_dp_aux_irq_handler(dev_priv);
-				found = true;
-			}
-
-			if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) {
-				u32 hotplug_trigger = iir & BXT_DE_PORT_HOTPLUG_MASK;
-
-				if (hotplug_trigger) {
-					bxt_hpd_irq_handler(dev_priv, hotplug_trigger);
-					found = true;
-				}
-			} else if (IS_BROADWELL(dev_priv)) {
-				u32 hotplug_trigger = iir & BDW_DE_PORT_HOTPLUG_MASK;
-
-				if (hotplug_trigger) {
-					ilk_hpd_irq_handler(dev_priv, hotplug_trigger);
-					found = true;
-				}
-			}
-
-			if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) &&
-			    (iir & BXT_DE_PORT_GMBUS)) {
-				intel_gmbus_irq_handler(dev_priv);
-				found = true;
-			}
-
-			if (DISPLAY_VER(dev_priv) >= 11) {
-				u32 te_trigger = iir & (DSI0_TE | DSI1_TE);
-
-				if (te_trigger) {
-					gen11_dsi_te_interrupt_handler(dev_priv, te_trigger);
-					found = true;
-				}
-			}
-
-			if (!found)
-				drm_err_ratelimited(&dev_priv->drm,
-						    "Unexpected DE Port interrupt\n");
-		}
-		else
-			drm_err_ratelimited(&dev_priv->drm,
-					    "The master control interrupt lied (DE PORT)!\n");
-	}
-
-	for_each_pipe(dev_priv, pipe) {
-		u32 fault_errors;
-
-		if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe)))
-			continue;
-
-		iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe));
-		if (!iir) {
-			drm_err_ratelimited(&dev_priv->drm,
-					    "The master control interrupt lied (DE PIPE)!\n");
-			continue;
-		}
-
-		intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe), iir);
-
-		if (iir & GEN8_PIPE_VBLANK)
-			intel_handle_vblank(dev_priv, pipe);
-
-		if (iir & gen8_de_pipe_flip_done_mask(dev_priv))
-			flip_done_handler(dev_priv, pipe);
-
-		if (iir & GEN8_PIPE_CDCLK_CRC_DONE)
-			hsw_pipe_crc_irq_handler(dev_priv, pipe);
-
-		if (iir & gen8_de_pipe_underrun_mask(dev_priv))
-			intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe);
-
-		fault_errors = iir & gen8_de_pipe_fault_mask(dev_priv);
-		if (fault_errors)
-			drm_err_ratelimited(&dev_priv->drm,
-					    "Fault errors on pipe %c: 0x%08x\n",
-					    pipe_name(pipe),
-					    fault_errors);
-	}
-
-	if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) &&
-	    master_ctl & GEN8_DE_PCH_IRQ) {
-		u32 pica_iir;
-
-		/*
-		 * FIXME(BDW): Assume for now that the new interrupt handling
-		 * scheme also closed the SDE interrupt handling race we've seen
-		 * on older pch-split platforms. But this needs testing.
-		 */
-		gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir);
-		if (iir) {
-			if (pica_iir)
-				xelpdp_pica_irq_handler(dev_priv, pica_iir);
-
-			if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-				icp_irq_handler(dev_priv, iir);
-			else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT)
-				spt_irq_handler(dev_priv, iir);
-			else
-				cpt_irq_handler(dev_priv, iir);
-		} else {
-			/*
-			 * Like on previous PCH there seems to be something
-			 * fishy going on with forwarding PCH interrupts.
-			 */
-			drm_dbg(&dev_priv->drm,
-				"The master control interrupt lied (SDE)!\n");
-		}
-	}
+
+	return ret;
 }
 
 static inline u32 gen8_master_intr_disable(void __iomem * const regs)
@@ -1645,29 +540,6 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
-static u32
-gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl)
-{
-	void __iomem * const regs = i915->uncore.regs;
-	u32 iir;
-
-	if (!(master_ctl & GEN11_GU_MISC_IRQ))
-		return 0;
-
-	iir = raw_reg_read(regs, GEN11_GU_MISC_IIR);
-	if (likely(iir))
-		raw_reg_write(regs, GEN11_GU_MISC_IIR, iir);
-
-	return iir;
-}
-
-static void
-gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir)
-{
-	if (iir & GEN11_GU_MISC_GSE)
-		intel_opregion_asle_intr(i915);
-}
-
 static inline u32 gen11_master_intr_disable(void __iomem * const regs)
 {
 	raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, 0);
@@ -1686,25 +558,6 @@ static inline void gen11_master_intr_enable(void __iomem * const regs)
 	raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, GEN11_MASTER_IRQ);
 }
 
-static void
-gen11_display_irq_handler(struct drm_i915_private *i915)
-{
-	void __iomem * const regs = i915->uncore.regs;
-	const u32 disp_ctl = raw_reg_read(regs, GEN11_DISPLAY_INT_CTL);
-
-	disable_rpm_wakeref_asserts(&i915->runtime_pm);
-	/*
-	 * GEN11_DISPLAY_INT_CTL has same format as GEN8_MASTER_IRQ
-	 * for the display related bits.
-	 */
-	raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, 0x0);
-	gen8_de_irq_handler(i915, disp_ctl);
-	raw_reg_write(regs, GEN11_DISPLAY_INT_CTL,
-		      GEN11_DISPLAY_IRQ_ENABLE);
-
-	enable_rpm_wakeref_asserts(&i915->runtime_pm);
-}
-
 static irqreturn_t gen11_irq_handler(int irq, void *arg)
 {
 	struct drm_i915_private *i915 = arg;
@@ -1806,184 +659,6 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
-/* Called from drm generic code, passed 'crtc' which
- * we use as a pipe index
- */
-int i8xx_enable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-
-	return 0;
-}
-
-int i915gm_enable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-
-	/*
-	 * Vblank interrupts fail to wake the device up from C2+.
-	 * Disabling render clock gating during C-states avoids
-	 * the problem. There is a small power cost so we do this
-	 * only when vblank interrupts are actually enabled.
-	 */
-	if (dev_priv->vblank_enabled++ == 0)
-		intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_ENABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
-
-	return i8xx_enable_vblank(crtc);
-}
-
-int i965_enable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_enable_pipestat(dev_priv, pipe,
-			     PIPE_START_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-
-	return 0;
-}
-
-int ilk_enable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
-		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	ilk_enable_display_irq(dev_priv, bit);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-
-	/* Even though there is no DMC, frame counter can get stuck when
-	 * PSR is active as no frames are generated.
-	 */
-	if (HAS_PSR(dev_priv))
-		drm_crtc_vblank_restore(crtc);
-
-	return 0;
-}
-
-static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc,
-				   bool enable)
-{
-	struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev);
-	enum port port;
-
-	if (!(intel_crtc->mode_flags &
-	    (I915_MODE_FLAG_DSI_USE_TE1 | I915_MODE_FLAG_DSI_USE_TE0)))
-		return false;
-
-	/* for dual link cases we consider TE from slave */
-	if (intel_crtc->mode_flags & I915_MODE_FLAG_DSI_USE_TE1)
-		port = PORT_B;
-	else
-		port = PORT_A;
-
-	intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_MASK_REG(port), DSI_TE_EVENT,
-			 enable ? 0 : DSI_TE_EVENT);
-
-	intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
-
-	return true;
-}
-
-int bdw_enable_vblank(struct drm_crtc *_crtc)
-{
-	struct intel_crtc *crtc = to_intel_crtc(_crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	enum pipe pipe = crtc->pipe;
-	unsigned long irqflags;
-
-	if (gen11_dsi_configure_te(crtc, true))
-		return 0;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-
-	/* Even if there is no DMC, frame counter can get stuck when
-	 * PSR is active as no frames are generated, so check only for PSR.
-	 */
-	if (HAS_PSR(dev_priv))
-		drm_crtc_vblank_restore(&crtc->base);
-
-	return 0;
-}
-
-/* Called from drm generic code, passed 'crtc' which
- * we use as a pipe index
- */
-void i8xx_disable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-}
-
-void i915gm_disable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-
-	i8xx_disable_vblank(crtc);
-
-	if (--dev_priv->vblank_enabled == 0)
-		intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_DISABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE));
-}
-
-void i965_disable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	i915_disable_pipestat(dev_priv, pipe,
-			      PIPE_START_VBLANK_INTERRUPT_STATUS);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-}
-
-void ilk_disable_vblank(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	enum pipe pipe = to_intel_crtc(crtc)->pipe;
-	unsigned long irqflags;
-	u32 bit = DISPLAY_VER(dev_priv) >= 7 ?
-		DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe);
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	ilk_disable_display_irq(dev_priv, bit);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-}
-
-void bdw_disable_vblank(struct drm_crtc *_crtc)
-{
-	struct intel_crtc *crtc = to_intel_crtc(_crtc);
-	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	enum pipe pipe = crtc->pipe;
-	unsigned long irqflags;
-
-	if (gen11_dsi_configure_te(crtc, false))
-		return;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
-}
-
 static void ibx_irq_reset(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore *uncore = &dev_priv->uncore;
@@ -1997,55 +672,6 @@ static void ibx_irq_reset(struct drm_i915_private *dev_priv)
 		intel_uncore_write(&dev_priv->uncore, SERR_INT, 0xffffffff);
 }
 
-static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-
-	if (IS_CHERRYVIEW(dev_priv))
-		intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
-	else
-		intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_VLV);
-
-	i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0);
-	intel_uncore_rmw(uncore, PORT_HOTPLUG_STAT, 0, 0);
-
-	i9xx_pipestat_irq_reset(dev_priv);
-
-	GEN3_IRQ_RESET(uncore, VLV_);
-	dev_priv->irq_mask = ~0u;
-}
-
-static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-
-	u32 pipestat_mask;
-	u32 enable_mask;
-	enum pipe pipe;
-
-	pipestat_mask = PIPE_CRC_DONE_INTERRUPT_STATUS;
-
-	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
-	for_each_pipe(dev_priv, pipe)
-		i915_enable_pipestat(dev_priv, pipe, pipestat_mask);
-
-	enable_mask = I915_DISPLAY_PORT_INTERRUPT |
-		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
-		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
-		I915_LPE_PIPE_A_INTERRUPT |
-		I915_LPE_PIPE_B_INTERRUPT;
-
-	if (IS_CHERRYVIEW(dev_priv))
-		enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT |
-			I915_LPE_PIPE_C_INTERRUPT;
-
-	drm_WARN_ON(&dev_priv->drm, dev_priv->irq_mask != ~0u);
-
-	dev_priv->irq_mask = ~enable_mask;
-
-	GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask);
-}
-
 /* drm_dma.h hooks
 */
 static void ilk_irq_reset(struct drm_i915_private *dev_priv)
@@ -2081,26 +707,6 @@ static void valleyview_irq_reset(struct drm_i915_private *dev_priv)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
-static void gen8_display_irq_reset(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	enum pipe pipe;
-
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff);
-	intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff);
-
-	for_each_pipe(dev_priv, pipe)
-		if (intel_display_power_is_enabled(dev_priv,
-						   POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
-
-	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
-	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
-}
-
 static void gen8_irq_reset(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore *uncore = &dev_priv->uncore;
@@ -2116,53 +722,6 @@ static void gen8_irq_reset(struct drm_i915_private *dev_priv)
 
 }
 
-static void gen11_display_irq_reset(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	enum pipe pipe;
-	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
-
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	intel_uncore_write(uncore, GEN11_DISPLAY_INT_CTL, 0);
-
-	if (DISPLAY_VER(dev_priv) >= 12) {
-		enum transcoder trans;
-
-		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
-			enum intel_display_power_domain domain;
-
-			domain = POWER_DOMAIN_TRANSCODER(trans);
-			if (!intel_display_power_is_enabled(dev_priv, domain))
-				continue;
-
-			intel_uncore_write(uncore, TRANS_PSR_IMR(trans), 0xffffffff);
-			intel_uncore_write(uncore, TRANS_PSR_IIR(trans), 0xffffffff);
-		}
-	} else {
-		intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff);
-		intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff);
-	}
-
-	for_each_pipe(dev_priv, pipe)
-		if (intel_display_power_is_enabled(dev_priv,
-						   POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
-
-	GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_);
-	GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_);
-
-	if (DISPLAY_VER(dev_priv) >= 14)
-		GEN3_IRQ_RESET(uncore, PICAINTERRUPT_);
-	else
-		GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_);
-
-	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
-		GEN3_IRQ_RESET(uncore, SDE);
-}
-
 static void gen11_irq_reset(struct drm_i915_private *dev_priv)
 {
 	struct intel_gt *gt = to_gt(dev_priv);
@@ -2191,52 +750,6 @@ static void dg1_irq_reset(struct drm_i915_private *dev_priv)
 	GEN3_IRQ_RESET(uncore, GEN8_PCU_);
 }
 
-void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
-				     u8 pipe_mask)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	u32 extra_ier = GEN8_PIPE_VBLANK |
-		gen8_de_pipe_underrun_mask(dev_priv) |
-		gen8_de_pipe_flip_done_mask(dev_priv);
-	enum pipe pipe;
-
-	spin_lock_irq(&dev_priv->irq_lock);
-
-	if (!intel_irqs_enabled(dev_priv)) {
-		spin_unlock_irq(&dev_priv->irq_lock);
-		return;
-	}
-
-	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
-		GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
-				  dev_priv->de_irq_mask[pipe],
-				  ~dev_priv->de_irq_mask[pipe] | extra_ier);
-
-	spin_unlock_irq(&dev_priv->irq_lock);
-}
-
-void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
-				     u8 pipe_mask)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	enum pipe pipe;
-
-	spin_lock_irq(&dev_priv->irq_lock);
-
-	if (!intel_irqs_enabled(dev_priv)) {
-		spin_unlock_irq(&dev_priv->irq_lock);
-		return;
-	}
-
-	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
-		GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe);
-
-	spin_unlock_irq(&dev_priv->irq_lock);
-
-	/* make sure we're done processing display irqs */
-	intel_synchronize_irq(dev_priv);
-}
-
 static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore *uncore = &dev_priv->uncore;
@@ -2254,35 +767,6 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
-/*
- * SDEIER is also touched by the interrupt handler to work around missed PCH
- * interrupts. Hence we can't update it after the interrupt handler is enabled -
- * instead we unconditionally enable all PCH interrupt sources here, but then
- * only unmask them as needed with SDEIMR.
- *
- * Note that we currently do this after installing the interrupt handler,
- * but before we enable the master interrupt. That should be sufficient
- * to avoid races with the irq handler, assuming we have MSI. Shared legacy
- * interrupts could still race.
- */
-static void ibx_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	u32 mask;
-
-	if (HAS_PCH_NOP(dev_priv))
-		return;
-
-	if (HAS_PCH_IBX(dev_priv))
-		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
-	else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv))
-		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
-	else
-		mask = SDE_GMBUS_CPT;
-
-	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
-}
-
 static void ilk_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	struct intel_uncore *uncore = &dev_priv->uncore;
@@ -2326,35 +810,6 @@ static void ilk_irq_postinstall(struct drm_i915_private *dev_priv)
 		      display_mask | extra_mask);
 }
 
-void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv)
-{
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	if (dev_priv->display_irqs_enabled)
-		return;
-
-	dev_priv->display_irqs_enabled = true;
-
-	if (intel_irqs_enabled(dev_priv)) {
-		vlv_display_irq_reset(dev_priv);
-		vlv_display_irq_postinstall(dev_priv);
-	}
-}
-
-void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv)
-{
-	lockdep_assert_held(&dev_priv->irq_lock);
-
-	if (!dev_priv->display_irqs_enabled)
-		return;
-
-	dev_priv->display_irqs_enabled = false;
-
-	if (intel_irqs_enabled(dev_priv))
-		vlv_display_irq_reset(dev_priv);
-}
-
-
 static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	gen5_gt_irq_postinstall(to_gt(dev_priv));
@@ -2368,108 +823,6 @@ static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv)
 	intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER);
 }
 
-static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-
-	u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) |
-		GEN8_PIPE_CDCLK_CRC_DONE;
-	u32 de_pipe_enables;
-	u32 de_port_masked = gen8_de_port_aux_mask(dev_priv);
-	u32 de_port_enables;
-	u32 de_misc_masked = GEN8_DE_EDP_PSR;
-	u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D);
-	enum pipe pipe;
-
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	if (DISPLAY_VER(dev_priv) <= 10)
-		de_misc_masked |= GEN8_DE_MISC_GSE;
-
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
-		de_port_masked |= BXT_DE_PORT_GMBUS;
-
-	if (DISPLAY_VER(dev_priv) >= 11) {
-		enum port port;
-
-		if (intel_bios_is_dsi_present(dev_priv, &port))
-			de_port_masked |= DSI0_TE | DSI1_TE;
-	}
-
-	de_pipe_enables = de_pipe_masked |
-		GEN8_PIPE_VBLANK |
-		gen8_de_pipe_underrun_mask(dev_priv) |
-		gen8_de_pipe_flip_done_mask(dev_priv);
-
-	de_port_enables = de_port_masked;
-	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
-		de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK;
-	else if (IS_BROADWELL(dev_priv))
-		de_port_enables |= BDW_DE_PORT_HOTPLUG_MASK;
-
-	if (DISPLAY_VER(dev_priv) >= 12) {
-		enum transcoder trans;
-
-		for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) {
-			enum intel_display_power_domain domain;
-
-			domain = POWER_DOMAIN_TRANSCODER(trans);
-			if (!intel_display_power_is_enabled(dev_priv, domain))
-				continue;
-
-			gen3_assert_iir_is_zero(uncore, TRANS_PSR_IIR(trans));
-		}
-	} else {
-		gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR);
-	}
-
-	for_each_pipe(dev_priv, pipe) {
-		dev_priv->de_irq_mask[pipe] = ~de_pipe_masked;
-
-		if (intel_display_power_is_enabled(dev_priv,
-				POWER_DOMAIN_PIPE(pipe)))
-			GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe,
-					  dev_priv->de_irq_mask[pipe],
-					  de_pipe_enables);
-	}
-
-	GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
-	GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked);
-
-	if (IS_DISPLAY_VER(dev_priv, 11, 13)) {
-		u32 de_hpd_masked = 0;
-		u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK |
-				     GEN11_DE_TBT_HOTPLUG_MASK;
-
-		GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked,
-			      de_hpd_enables);
-	}
-}
-
-static void mtp_irq_postinstall(struct drm_i915_private *i915)
-{
-	struct intel_uncore *uncore = &i915->uncore;
-	u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT;
-	u32 de_hpd_mask = XELPDP_AUX_TC_MASK;
-	u32 de_hpd_enables = de_hpd_mask | XELPDP_DP_ALT_HOTPLUG_MASK |
-			     XELPDP_TBT_HOTPLUG_MASK;
-
-	GEN3_IRQ_INIT(uncore, PICAINTERRUPT_, ~de_hpd_mask,
-		      de_hpd_enables);
-
-	GEN3_IRQ_INIT(uncore, SDE, ~sde_mask, 0xffffffff);
-}
-
-static void icp_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	struct intel_uncore *uncore = &dev_priv->uncore;
-	u32 mask = SDE_GMBUS_ICP;
-
-	GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff);
-}
-
 static void gen8_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP)
@@ -2483,17 +836,6 @@ static void gen8_irq_postinstall(struct drm_i915_private *dev_priv)
 	gen8_master_intr_enable(dev_priv->uncore.regs);
 }
 
-static void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv)
-{
-	if (!HAS_DISPLAY(dev_priv))
-		return;
-
-	gen8_de_irq_postinstall(dev_priv);
-
-	intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL,
-			   GEN11_DISPLAY_IRQ_ENABLE);
-}
-
 static void gen11_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	struct intel_gt *gt = to_gt(dev_priv);
diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h
index 913c854f873de..e665a1b007dcc 100644
--- a/drivers/gpu/drm/i915/i915_irq.h
+++ b/drivers/gpu/drm/i915/i915_irq.h
@@ -25,34 +25,6 @@ void intel_irq_fini(struct drm_i915_private *dev_priv);
 int intel_irq_install(struct drm_i915_private *dev_priv);
 void intel_irq_uninstall(struct drm_i915_private *dev_priv);
 
-u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv,
-			      enum pipe pipe);
-void
-i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
-		     u32 status_mask);
-
-void
-i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
-		      u32 status_mask);
-
-void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv);
-void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv);
-
-void ilk_update_display_irq(struct drm_i915_private *i915,
-			    u32 interrupt_mask, u32 enabled_irq_mask);
-void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits);
-void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits);
-
-void bdw_update_port_irq(struct drm_i915_private *i915,
-			 u32 interrupt_mask, u32 enabled_irq_mask);
-void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
-void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits);
-
-void ibx_display_interrupt_update(struct drm_i915_private *i915,
-				  u32 interrupt_mask, u32 enabled_irq_mask);
-void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits);
-void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits);
-
 void gen5_enable_gt_irq(struct drm_i915_private *dev_priv, u32 mask);
 void gen5_disable_gt_irq(struct drm_i915_private *dev_priv, u32 mask);
 void gen11_reset_rps_interrupts(struct drm_i915_private *dev_priv);
@@ -68,23 +40,7 @@ bool intel_irqs_enabled(struct drm_i915_private *dev_priv);
 void intel_synchronize_irq(struct drm_i915_private *i915);
 void intel_synchronize_hardirq(struct drm_i915_private *i915);
 
-void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
-				     u8 pipe_mask);
-void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
-				     u8 pipe_mask);
-u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv);
-
-
-int i8xx_enable_vblank(struct drm_crtc *crtc);
-int i915gm_enable_vblank(struct drm_crtc *crtc);
-int i965_enable_vblank(struct drm_crtc *crtc);
-int ilk_enable_vblank(struct drm_crtc *crtc);
-int bdw_enable_vblank(struct drm_crtc *crtc);
-void i8xx_disable_vblank(struct drm_crtc *crtc);
-void i915gm_disable_vblank(struct drm_crtc *crtc);
-void i965_disable_vblank(struct drm_crtc *crtc);
-void ilk_disable_vblank(struct drm_crtc *crtc);
-void bdw_disable_vblank(struct drm_crtc *crtc);
+void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg);
 
 void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr,
 		    i915_reg_t iir, i915_reg_t ier);
-- 
GitLab


From 4c4279a8d58d146dc39b150226ee974652de4eef Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 15 May 2023 16:02:22 +0530
Subject: [PATCH 0507/1791] drm/i915/hdcp: add intel_atomic_state argument to
 hdcp_enable function

Pass all the parameter in intel_encoder->enable()
to intel_hdcp_enable as we need intel_atomic_state
later down to get acquire_ctx.

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515103225.688830-2-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c    |  4 +---
 drivers/gpu/drm/i915/display/intel_dp_mst.c |  4 +---
 drivers/gpu/drm/i915/display/intel_hdcp.c   | 16 +++++++++-------
 drivers/gpu/drm/i915/display/intel_hdcp.h   |  6 ++++--
 4 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index b1caf521a3486..db056bb084d16 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3264,9 +3264,7 @@ static void intel_enable_ddi(struct intel_atomic_state *state,
 	/* Enable hdcp if it's desired */
 	if (conn_state->content_protection ==
 	    DRM_MODE_CONTENT_PROTECTION_DESIRED)
-		intel_hdcp_enable(to_intel_connector(conn_state->connector),
-				  crtc_state,
-				  (u8)conn_state->hdcp_content_type);
+		intel_hdcp_enable(state, encoder, crtc_state, conn_state);
 }
 
 static void intel_disable_ddi_dp(struct intel_atomic_state *state,
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 63d61e6102102..6df8519b3347e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -801,9 +801,7 @@ static void intel_mst_enable_dp(struct intel_atomic_state *state,
 	/* Enable hdcp if it's desired */
 	if (conn_state->content_protection ==
 	    DRM_MODE_CONTENT_PROTECTION_DESIRED)
-		intel_hdcp_enable(to_intel_connector(conn_state->connector),
-				  pipe_config,
-				  (u8)conn_state->hdcp_content_type);
+		intel_hdcp_enable(state, encoder, pipe_config, conn_state);
 }
 
 static bool intel_dp_mst_enc_get_hw_state(struct intel_encoder *encoder,
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index b183efab04a1d..bce0fa2e45a1a 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -2332,10 +2332,14 @@ int intel_hdcp_init(struct intel_connector *connector,
 	return 0;
 }
 
-int intel_hdcp_enable(struct intel_connector *connector,
-		      const struct intel_crtc_state *pipe_config, u8 content_type)
+int intel_hdcp_enable(struct intel_atomic_state *state,
+		      struct intel_encoder *encoder,
+		      const struct intel_crtc_state *pipe_config,
+		      const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct intel_connector *connector =
+		to_intel_connector(conn_state->connector);
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS;
@@ -2354,7 +2358,7 @@ int intel_hdcp_enable(struct intel_connector *connector,
 	mutex_lock(&dig_port->hdcp_mutex);
 	drm_WARN_ON(&dev_priv->drm,
 		    hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED);
-	hdcp->content_type = content_type;
+	hdcp->content_type = (u8)conn_state->content_type;
 
 	if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_DP_MST)) {
 		hdcp->cpu_transcoder = pipe_config->mst_master_transcoder;
@@ -2485,9 +2489,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 	}
 
 	if (desired_and_not_enabled || content_protection_type_changed)
-		intel_hdcp_enable(connector,
-				  crtc_state,
-				  (u8)conn_state->hdcp_content_type);
+		intel_hdcp_enable(state, encoder, crtc_state, conn_state);
 }
 
 void intel_hdcp_component_fini(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h
index 8f53b0c7fe5cf..ce283f4f69fd9 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.h
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.h
@@ -28,8 +28,10 @@ void intel_hdcp_atomic_check(struct drm_connector *connector,
 int intel_hdcp_init(struct intel_connector *connector,
 		    struct intel_digital_port *dig_port,
 		    const struct intel_hdcp_shim *hdcp_shim);
-int intel_hdcp_enable(struct intel_connector *connector,
-		      const struct intel_crtc_state *pipe_config, u8 content_type);
+int intel_hdcp_enable(struct intel_atomic_state *state,
+		      struct intel_encoder *encoder,
+		      const struct intel_crtc_state *pipe_config,
+		      const struct drm_connector_state *conn_state);
 int intel_hdcp_disable(struct intel_connector *connector);
 void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 			    struct intel_encoder *encoder,
-- 
GitLab


From 415beb1fa63cf06e3b696822e94f977171921a2a Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 15 May 2023 16:02:23 +0530
Subject: [PATCH 0508/1791] drm/i915/hdcp: Remove enforce_type0 check outside
 loop

Remove enforce_type0 check outside the loop since it
does not make sense to keep it there as we use the same
digport and continue checking it again and again

Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515103225.688830-3-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index bce0fa2e45a1a..73b7e7e62705b 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -84,6 +84,9 @@ intel_hdcp_required_content_stream(struct intel_digital_port *dig_port)
 	if (dig_port->hdcp_auth_status)
 		return 0;
 
+	if (!dig_port->hdcp_mst_type1_capable)
+		enforce_type0 = true;
+
 	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		if (connector->base.status == connector_status_disconnected)
@@ -96,9 +99,6 @@ intel_hdcp_required_content_stream(struct intel_digital_port *dig_port)
 		if (conn_dig_port != dig_port)
 			continue;
 
-		if (!enforce_type0 && !dig_port->hdcp_mst_type1_capable)
-			enforce_type0 = true;
-
 		data->streams[data->k].stream_id = intel_conn_to_vcpi(connector);
 		data->k++;
 
-- 
GitLab


From a6d82f1b7ada6154668f1b5ff751a845aa540637 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 15 May 2023 16:02:24 +0530
Subject: [PATCH 0509/1791] drm/i915/hdcp: Fix modeset locking issue in hdcp
 mst

Since topology state is being added to drm_atomic_state now all
drm_modeset_lock required are being taken from core. This raises
an issue when we try to loop over connector and assign vcpi id to
our streams as we did not have atomic state to derive acquire_ctx
from. We fill in stream info if dpmst encoder is found before
enabling hdcp. intel_hdcp_required_stream will be broken which
will only set the content type.

--v2
-move prepare streams to beginning of intel_hdcp_enable to avoid
checking of mst encoder twice [Ankit]

--v3
-break intel_required_content_stream to two part and set the stream_id
at the beginning [Ankit]

--v4
-change return types for intel_hdcp_prepare_stream and
intel_hdcp_required content_stream [Ankit]
-rename intel_hdcp_set_content_stream to
intel_hdcp_set_stream [Ankit]
-place intel_hdcp_set_streams above caller [Ankit]

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515103225.688830-4-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 116 ++++++++++++----------
 1 file changed, 62 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 73b7e7e62705b..6d3ea68593286 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -30,7 +30,8 @@
 #define KEY_LOAD_TRIES	5
 #define HDCP2_LC_RETRY_CNT			3
 
-static int intel_conn_to_vcpi(struct intel_connector *connector)
+static int intel_conn_to_vcpi(struct drm_atomic_state *state,
+			      struct intel_connector *connector)
 {
 	struct drm_dp_mst_topology_mgr *mgr;
 	struct drm_dp_mst_atomic_payload *payload;
@@ -42,7 +43,7 @@ static int intel_conn_to_vcpi(struct intel_connector *connector)
 		return 0;
 	mgr = connector->port->mgr;
 
-	drm_modeset_lock(&mgr->base.lock, NULL);
+	drm_modeset_lock(&mgr->base.lock, state->acquire_ctx);
 	mst_state = to_drm_dp_mst_topology_state(mgr->base.state);
 	payload = drm_atomic_get_mst_payload_state(mst_state, connector->port);
 	if (drm_WARN_ON(mgr->dev, !payload))
@@ -54,7 +55,6 @@ static int intel_conn_to_vcpi(struct intel_connector *connector)
 		goto out;
 	}
 out:
-	drm_modeset_unlock(&mgr->base.lock);
 	return vcpi;
 }
 
@@ -68,49 +68,19 @@ static int intel_conn_to_vcpi(struct intel_connector *connector)
  * DP MST topology. Though it is not compulsory, security fw should change its
  * policy to mark different content_types for different streams.
  */
-static int
+static void
 intel_hdcp_required_content_stream(struct intel_digital_port *dig_port)
 {
-	struct drm_connector_list_iter conn_iter;
-	struct intel_digital_port *conn_dig_port;
-	struct intel_connector *connector;
-	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	bool enforce_type0 = false;
 	int k;
 
-	data->k = 0;
-
 	if (dig_port->hdcp_auth_status)
-		return 0;
+		return;
 
 	if (!dig_port->hdcp_mst_type1_capable)
 		enforce_type0 = true;
 
-	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
-	for_each_intel_connector_iter(connector, &conn_iter) {
-		if (connector->base.status == connector_status_disconnected)
-			continue;
-
-		if (!intel_encoder_is_mst(intel_attached_encoder(connector)))
-			continue;
-
-		conn_dig_port = intel_attached_dig_port(connector);
-		if (conn_dig_port != dig_port)
-			continue;
-
-		data->streams[data->k].stream_id = intel_conn_to_vcpi(connector);
-		data->k++;
-
-		/* if there is only one active stream */
-		if (dig_port->dp.active_mst_links <= 1)
-			break;
-	}
-	drm_connector_list_iter_end(&conn_iter);
-
-	if (drm_WARN_ON(&i915->drm, data->k > INTEL_NUM_PIPES(i915) || data->k == 0))
-		return -EINVAL;
-
 	/*
 	 * Apply common protection level across all streams in DP MST Topology.
 	 * Use highest supported content type for all streams in DP MST Topology.
@@ -118,27 +88,20 @@ intel_hdcp_required_content_stream(struct intel_digital_port *dig_port)
 	for (k = 0; k < data->k; k++)
 		data->streams[k].stream_type =
 			enforce_type0 ? DRM_MODE_HDCP_CONTENT_TYPE0 : DRM_MODE_HDCP_CONTENT_TYPE1;
-
-	return 0;
 }
 
-static int intel_hdcp_prepare_streams(struct intel_connector *connector)
+static void intel_hdcp_prepare_streams(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct intel_hdcp *hdcp = &connector->hdcp;
-	int ret;
 
 	if (!intel_encoder_is_mst(intel_attached_encoder(connector))) {
 		data->k = 1;
 		data->streams[0].stream_type = hdcp->content_type;
 	} else {
-		ret = intel_hdcp_required_content_stream(dig_port);
-		if (ret)
-			return ret;
+		intel_hdcp_required_content_stream(dig_port);
 	}
-
-	return 0;
 }
 
 static
@@ -1919,13 +1882,7 @@ static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector)
 	for (i = 0; i < tries && !dig_port->hdcp_auth_status; i++) {
 		ret = hdcp2_authenticate_sink(connector);
 		if (!ret) {
-			ret = intel_hdcp_prepare_streams(connector);
-			if (ret) {
-				drm_dbg_kms(&i915->drm,
-					    "Prepare streams failed.(%d)\n",
-					    ret);
-				break;
-			}
+			intel_hdcp_prepare_streams(connector);
 
 			ret = hdcp2_propagate_stream_management_info(connector);
 			if (ret) {
@@ -2332,6 +2289,49 @@ int intel_hdcp_init(struct intel_connector *connector,
 	return 0;
 }
 
+static int
+intel_hdcp_set_streams(struct intel_digital_port *dig_port,
+		       struct intel_atomic_state *state)
+{
+	struct drm_connector_list_iter conn_iter;
+	struct intel_digital_port *conn_dig_port;
+	struct intel_connector *connector;
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
+
+	if (!intel_encoder_is_mst(&dig_port->base))
+		return 0;
+
+	data->k = 0;
+
+	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	for_each_intel_connector_iter(connector, &conn_iter) {
+		if (connector->base.status == connector_status_disconnected)
+			continue;
+
+		if (!intel_encoder_is_mst(intel_attached_encoder(connector)))
+			continue;
+
+		conn_dig_port = intel_attached_dig_port(connector);
+		if (conn_dig_port != dig_port)
+			continue;
+
+		data->streams[data->k].stream_id =
+			intel_conn_to_vcpi(&state->base, connector);
+		data->k++;
+
+		/* if there is only one active stream */
+		if (dig_port->dp.active_mst_links <= 1)
+			break;
+	}
+	drm_connector_list_iter_end(&conn_iter);
+
+	if (drm_WARN_ON(&i915->drm, data->k > INTEL_NUM_PIPES(i915) || data->k == 0))
+		return -EINVAL;
+
+	return 0;
+}
+
 int intel_hdcp_enable(struct intel_atomic_state *state,
 		      struct intel_encoder *encoder,
 		      const struct intel_crtc_state *pipe_config,
@@ -2377,9 +2377,17 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 	 * is capable of HDCP2.2, it is preferred to use HDCP2.2.
 	 */
 	if (intel_hdcp2_capable(connector)) {
-		ret = _intel_hdcp2_enable(connector);
-		if (!ret)
-			check_link_interval = DRM_HDCP2_CHECK_PERIOD_MS;
+		ret = intel_hdcp_set_streams(dig_port, state);
+		if (!ret) {
+			ret = _intel_hdcp2_enable(connector);
+			if (!ret)
+				check_link_interval =
+					DRM_HDCP2_CHECK_PERIOD_MS;
+		} else {
+			drm_dbg_kms(&dev_priv->drm,
+				    "Set content streams failed: (%d)\n",
+				    ret);
+		}
 	}
 
 	/*
-- 
GitLab


From e45c5f1ad76968798b62add0396f85d9838aaca2 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 15 May 2023 16:57:26 +0530
Subject: [PATCH 0510/1791] drm/i915/hdcp: Fill hdcp2_streamid_type and k in
 appropriate places

stream_id and k(no of streams) should be set in
intel_hdcp_set_content_streams. stream_type should be set in
intel_hdcp_required_content_stream.

--v5
-add missing stream_id assignment [Ankit]

Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515112726.689116-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 6d3ea68593286..dd539106ee5ad 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -97,7 +97,6 @@ static void intel_hdcp_prepare_streams(struct intel_connector *connector)
 	struct intel_hdcp *hdcp = &connector->hdcp;
 
 	if (!intel_encoder_is_mst(intel_attached_encoder(connector))) {
-		data->k = 1;
 		data->streams[0].stream_type = hdcp->content_type;
 	} else {
 		intel_hdcp_required_content_stream(dig_port);
@@ -2161,7 +2160,6 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 
 	if (DISPLAY_VER(dev_priv) < 12)
@@ -2191,9 +2189,6 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 		drm_err(&dev_priv->drm, "Out of Memory\n");
 		return -ENOMEM;
 	}
-	/* For SST */
-	data->streams[0].stream_id = 0;
-	data->streams[0].stream_type = hdcp->content_type;
 
 	return 0;
 }
@@ -2299,8 +2294,11 @@ intel_hdcp_set_streams(struct intel_digital_port *dig_port,
 	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 
-	if (!intel_encoder_is_mst(&dig_port->base))
+	if (!intel_encoder_is_mst(&dig_port->base)) {
+		data->k = 1;
+		data->streams[0].stream_id = 0;
 		return 0;
+	}
 
 	data->k = 0;
 
-- 
GitLab


From adbcfcc92b5aaffdee4c5e62c077919014c483d9 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 15 May 2023 18:20:33 +0200
Subject: [PATCH 0511/1791] dt-bindings: display: bridge: tc358867: Document
 TC358867/TC9595 compatible

The TC358867/TC9595 devices are compatible with the predecessor TC358767.
Document compatible strings for the new devices, so they can be discerned
in board DTs. Update the title to match description in the process.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515162033.66986-1-marex@denx.de
---
 .../bindings/display/bridge/toshiba,tc358767.yaml  | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358767.yaml b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358767.yaml
index e1494b5007cba..0521261b04a9c 100644
--- a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358767.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358767.yaml
@@ -4,16 +4,24 @@
 $id: http://devicetree.org/schemas/display/bridge/toshiba,tc358767.yaml#
 $schema: http://devicetree.org/meta-schemas/core.yaml#
 
-title: Toshiba TC358767 eDP bridge
+title: Toshiba TC358767/TC358867/TC9595 DSI/DPI/eDP bridge
 
 maintainers:
   - Andrey Gusakov <andrey.gusakov@cogentembedded.com>
 
-description: The TC358767 is bridge device which converts DSI/DPI to eDP/DP
+description: |
+  The TC358767/TC358867/TC9595 is bridge device which
+  converts DSI/DPI to eDP/DP .
 
 properties:
   compatible:
-    const: toshiba,tc358767
+    oneOf:
+      - items:
+          - enum:
+              - toshiba,tc358867
+              - toshiba,tc9595
+          - const: toshiba,tc358767
+      - const: toshiba,tc358767
 
   reg:
     enum:
-- 
GitLab


From 728cb3f061e2b3a002fd76d91c2449b1497b6640 Mon Sep 17 00:00:00 2001
From: Anup Sharma <anupnewsmail@gmail.com>
Date: Sat, 13 May 2023 15:42:17 +0530
Subject: [PATCH 0512/1791] gpu: drm: bridge: No need to set device_driver
 owner

There is no need to exclusively set the .owner member of the struct
device_driver when defining the platform_driver struct. The Linux core
takes care of setting the .owner member as part of the call to
module_platform_driver() helper function.

Issue identified using the platform_no_drv_owner.cocci Coccinelle
semantic patch as:
drivers/gpu/drm/bridge/samsung-dsim.c:1957:6-11: No need to set .owner here.
The core will do it.

No functional changes are intended.

Signed-off-by: Anup Sharma <anupnewsmail@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/ZF9igb/nvL6GRBsq@yoga
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index e0a402a85787c..10dc3315e69eb 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -1954,7 +1954,6 @@ static struct platform_driver samsung_dsim_driver = {
 	.remove = samsung_dsim_remove,
 	.driver = {
 		   .name = "samsung-dsim",
-		   .owner = THIS_MODULE,
 		   .pm = &samsung_dsim_pm_ops,
 		   .of_match_table = samsung_dsim_of_match,
 	},
-- 
GitLab


From 81d6b37b69b8b5d1a4c81d2e208b41888d4283df Mon Sep 17 00:00:00 2001
From: Francesco Dolcini <francesco.dolcini@toradex.com>
Date: Mon, 15 May 2023 20:44:08 +0200
Subject: [PATCH 0513/1791] drm/bridge: tc358768: remove unneeded semicolon

Remove unneeded stray semicolon.

Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202305152341.oiSjRpv6-lkp@intel.com/
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515184408.9421-1-francesco@dolcini.it
---
 drivers/gpu/drm/bridge/tc358768.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 03c7e82e41090..97ae3a9c90daa 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -952,7 +952,7 @@ tc358768_atomic_get_input_bus_fmts(struct drm_bridge *bridge,
 	case 24:
 		input_fmts[0] = MEDIA_BUS_FMT_RGB888_1X24;
 		break;
-	};
+	}
 
 	*num_input_fmts = MAX_INPUT_SEL_FORMATS;
 
-- 
GitLab


From 7eeef32719f6af935a1554813e6bc206446339cd Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:18 +0300
Subject: [PATCH 0514/1791] drm/i915: Fix PIPEDMC disabling for a bigjoiner
 configuration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For a bigjoiner configuration display->crtc_disable() will be called
first for the slave CRTCs and then for the master CRTC. However slave
CRTCs will be actually disabled only after the master CRTC is disabled
(from the encoder disable hooks called with the master CRTC state).
Hence the slave PIPEDMCs can be disabled only after the master CRTC is
disabled, make this so.

intel_encoders_post_pll_disable() must be called only for the master
CRTC, as for the other two encoder disable hooks. While at it fix this
up as well. This didn't cause a problem, since
intel_encoders_post_pll_disable() will call the corresponding hook only
for an encoder/connector connected to the given CRTC, however slave
CRTCs will have no associated encoder/connector.

Fixes: 3af2ff0840be ("drm/i915: Enable a PIPEDMC whenever its corresponding pipe is enabled")
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-2-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 4b70b389e0cb1..e7ed5afd1d429 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1703,9 +1703,17 @@ static void hsw_crtc_disable(struct intel_atomic_state *state,
 
 	intel_disable_shared_dpll(old_crtc_state);
 
-	intel_encoders_post_pll_disable(state, crtc);
+	if (!intel_crtc_is_bigjoiner_slave(old_crtc_state)) {
+		struct intel_crtc *slave_crtc;
+
+		intel_encoders_post_pll_disable(state, crtc);
 
-	intel_dmc_disable_pipe(i915, crtc->pipe);
+		intel_dmc_disable_pipe(i915, crtc->pipe);
+
+		for_each_intel_crtc_in_pipe_mask(&i915->drm, slave_crtc,
+						 intel_crtc_bigjoiner_slave_pipes(old_crtc_state))
+			intel_dmc_disable_pipe(i915, slave_crtc->pipe);
+	}
 }
 
 static void i9xx_pfit_enable(const struct intel_crtc_state *crtc_state)
-- 
GitLab


From 0694cab1d5a425139d9cda08332f59fb79248ff3 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:19 +0300
Subject: [PATCH 0515/1791] drm/i915: Add helpers to reference/unreference a
 DPLL for a CRTC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add helpers to reference/unreference a shared DPLL tracking the use of
it by a given CRTC.

This prepares for the next patch, which unreferences a DPLL during CRTC
HW-readout/sanitization.

Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-3-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c | 58 +++++++++++++++----
 1 file changed, 46 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index ee55ab3095684..98bde2cf4ac69 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -351,13 +351,35 @@ intel_find_shared_dpll(struct intel_atomic_state *state,
 	return NULL;
 }
 
+/**
+ * intel_reference_shared_dpll_crtc - Get a DPLL reference for a CRTC
+ * @crtc: CRTC on which behalf the reference is taken
+ * @pll: DPLL for which the reference is taken
+ * @shared_dpll_state: the DPLL atomic state in which the reference is tracked
+ *
+ * Take a reference for @pll tracking the use of it by @crtc.
+ */
+static void
+intel_reference_shared_dpll_crtc(const struct intel_crtc *crtc,
+				 const struct intel_shared_dpll *pll,
+				 struct intel_shared_dpll_state *shared_dpll_state)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+
+	drm_WARN_ON(&i915->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) != 0);
+
+	shared_dpll_state->pipe_mask |= BIT(crtc->pipe);
+
+	drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] reserving %s\n",
+		    crtc->base.base.id, crtc->base.name, pll->info->name);
+}
+
 static void
 intel_reference_shared_dpll(struct intel_atomic_state *state,
 			    const struct intel_crtc *crtc,
 			    const struct intel_shared_dpll *pll,
 			    const struct intel_dpll_hw_state *pll_state)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	struct intel_shared_dpll_state *shared_dpll;
 	const enum intel_dpll_id id = pll->info->id;
 
@@ -366,11 +388,29 @@ intel_reference_shared_dpll(struct intel_atomic_state *state,
 	if (shared_dpll[id].pipe_mask == 0)
 		shared_dpll[id].hw_state = *pll_state;
 
-	drm_WARN_ON(&i915->drm, (shared_dpll[id].pipe_mask & BIT(crtc->pipe)) != 0);
+	intel_reference_shared_dpll_crtc(crtc, pll, &shared_dpll[id]);
+}
+
+/**
+ * intel_unreference_shared_dpll_crtc - Drop a DPLL reference for a CRTC
+ * @crtc: CRTC on which behalf the reference is dropped
+ * @pll: DPLL for which the reference is dropped
+ * @shared_dpll_state: the DPLL atomic state in which the reference is tracked
+ *
+ * Drop a reference for @pll tracking the end of use of it by @crtc.
+ */
+static void
+intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc,
+				   const struct intel_shared_dpll *pll,
+				   struct intel_shared_dpll_state *shared_dpll_state)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 
-	shared_dpll[id].pipe_mask |= BIT(crtc->pipe);
+	drm_WARN_ON(&i915->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) == 0);
 
-	drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] reserving %s\n",
+	shared_dpll_state->pipe_mask &= ~BIT(crtc->pipe);
+
+	drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] releasing %s\n",
 		    crtc->base.base.id, crtc->base.name, pll->info->name);
 }
 
@@ -378,18 +418,12 @@ static void intel_unreference_shared_dpll(struct intel_atomic_state *state,
 					  const struct intel_crtc *crtc,
 					  const struct intel_shared_dpll *pll)
 {
-	struct drm_i915_private *i915 = to_i915(state->base.dev);
 	struct intel_shared_dpll_state *shared_dpll;
 	const enum intel_dpll_id id = pll->info->id;
 
 	shared_dpll = intel_atomic_get_shared_dpll_state(&state->base);
 
-	drm_WARN_ON(&i915->drm, (shared_dpll[id].pipe_mask & BIT(crtc->pipe)) == 0);
-
-	shared_dpll[id].pipe_mask &= ~BIT(crtc->pipe);
-
-	drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] releasing %s\n",
-		    crtc->base.base.id, crtc->base.name, pll->info->name);
+	intel_unreference_shared_dpll_crtc(crtc, pll, &shared_dpll[id]);
 }
 
 static void intel_put_dpll(struct intel_atomic_state *state,
@@ -4313,7 +4347,7 @@ static void readout_dpll_hw_state(struct drm_i915_private *i915,
 			to_intel_crtc_state(crtc->base.state);
 
 		if (crtc_state->hw.active && crtc_state->shared_dpll == pll)
-			pll->state.pipe_mask |= BIT(crtc->pipe);
+			intel_reference_shared_dpll_crtc(crtc, pll, &pll->state);
 	}
 	pll->active_mask = pll->state.pipe_mask;
 
-- 
GitLab


From 7ff9a17e372e5152a77ec657949a3c9250720fa4 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:20 +0300
Subject: [PATCH 0516/1791] drm/i915: Make the CRTC state consistent during
 sanitize-disabling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make sure that the CRTC state is reset correctly, as expected after
disabling the CRTC.

In particular this change will:
- Zero all the CSC blob pointers after intel_crtc_free_hw_state()
  has freed them.
- Zero the shared DPLL and port PLL pointers and clear the
  corresponding CRTC reference flag in the PLL state.
- Reset all the transcoder and pipe fields.

v2:
- Reset fully the CRTC state. (Ville)
- Clear pipe active flags in the DPLL state.

v3:
- Clear only the CRTC reference flag and add a helper for this.
  (Ville)

v4:
- Rebased on previous patch, adding
  intel_unreference_shared_dpll_crtc() separately. (Ville)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-4-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c      |  2 +-
 drivers/gpu/drm/i915/display/intel_dpll_mgr.h      |  3 +++
 drivers/gpu/drm/i915/display/intel_modeset_setup.c | 13 +++++++------
 3 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index 98bde2cf4ac69..6b2d8a1e2aa90 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -399,7 +399,7 @@ intel_reference_shared_dpll(struct intel_atomic_state *state,
  *
  * Drop a reference for @pll tracking the end of use of it by @crtc.
  */
-static void
+void
 intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc,
 				   const struct intel_shared_dpll *pll,
 				   struct intel_shared_dpll_state *shared_dpll_state)
diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h
index 3854f1b4299ac..ba62eb5d7c517 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h
@@ -341,6 +341,9 @@ int intel_reserve_shared_dplls(struct intel_atomic_state *state,
 			       struct intel_encoder *encoder);
 void intel_release_shared_dplls(struct intel_atomic_state *state,
 				struct intel_crtc *crtc);
+void intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc,
+					const struct intel_shared_dpll *pll,
+					struct intel_shared_dpll_state *shared_dpll_state);
 void icl_set_active_port_dpll(struct intel_crtc_state *crtc_state,
 			      enum icl_port_dpll_id port_dpll_id);
 void intel_update_active_dpll(struct intel_atomic_state *state,
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index eefa4018dc0c2..6e55806bbe066 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -88,13 +88,14 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 	crtc->active = false;
 	crtc->base.enabled = false;
 
-	drm_WARN_ON(&i915->drm,
-		    drm_atomic_set_mode_for_crtc(&crtc_state->uapi, NULL) < 0);
-	crtc_state->uapi.active = false;
-	crtc_state->uapi.connector_mask = 0;
-	crtc_state->uapi.encoder_mask = 0;
+	if (crtc_state->shared_dpll)
+		intel_unreference_shared_dpll_crtc(crtc,
+						   crtc_state->shared_dpll,
+						   &crtc_state->shared_dpll->state);
+
+	__drm_atomic_helper_crtc_destroy_state(&crtc_state->uapi);
 	intel_crtc_free_hw_state(crtc_state);
-	memset(&crtc_state->hw, 0, sizeof(crtc_state->hw));
+	intel_crtc_state_reset(crtc_state, crtc);
 
 	for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder)
 		encoder->base.crtc = NULL;
-- 
GitLab


From db4069fcbdc5c8bc03424934a3395b39b71d9dc6 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:21 +0300
Subject: [PATCH 0517/1791] drm/i915: Update connector atomic state before crtc
 sanitize-disabling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During HW state readout/sanitization an up-to-date connector atomic
state will be required by a follow-up patch, which can disable CRTCs
with an encoder (and calling the correct encoder hooks happens via the
connector atomic state encoder pointer). So update the connector state
already before the CRTC sanitize/disable step. For now this doesn't make
a difference, since intel_modeset_update_connector_atomic_state() will
update/enable the atomic state only for connectors that have an enabled
encoder/CRTC. Such CRTCs/encoders will not be affected by
intel_sanitize_crtc().

v2: Add comment about why the connector state needs to be up-to-date.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-5-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_modeset_setup.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 6e55806bbe066..66796e8eef90a 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -701,6 +701,12 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
 	for_each_intel_encoder(&i915->drm, encoder)
 		intel_sanitize_encoder(encoder);
 
+	/*
+	 * Sanitizing CRTCs needs their connector atomic state to be
+	 * up-to-date, so ensure that already here.
+	 */
+	intel_modeset_update_connector_atomic_state(i915);
+
 	for_each_intel_crtc(&i915->drm, crtc) {
 		struct intel_crtc_state *crtc_state =
 			to_intel_crtc_state(crtc->base.state);
@@ -709,8 +715,6 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
 		intel_crtc_state_dump(crtc_state, NULL, "setup_hw_state");
 	}
 
-	intel_modeset_update_connector_atomic_state(i915);
-
 	intel_dpll_sanitize_state(i915);
 
 	intel_wm_get_hw_state(i915);
-- 
GitLab


From 3ad41442d7bf5b3af0de927e14ed92b39da68224 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:22 +0300
Subject: [PATCH 0518/1791] drm/i915: Separate
 intel_crtc_disable_noatomic_begin/complete()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Split calling the CRTC/encoder disabling hooks and updating the CRTC and
DPLL object states from updating the CRTC and atomic state and other
global state (BW, CDCLK, DBUF) into separate functions. When disabling a
bigjoiner configuration the latter step can be done only after all the
linked pipes are disabled, so this change prepares for that.

No functional changes.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-6-imre.deak@intel.com
---
 .../drm/i915/display/intel_modeset_setup.c    | 34 +++++++++++++------
 1 file changed, 24 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 66796e8eef90a..2c93f4c5dc8cf 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -30,23 +30,15 @@
 #include "intel_wm.h"
 #include "skl_watermark.h"
 
-static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
-					struct drm_modeset_acquire_ctx *ctx)
+static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
+					      struct drm_modeset_acquire_ctx *ctx)
 {
-	struct intel_encoder *encoder;
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	struct intel_bw_state *bw_state =
-		to_intel_bw_state(i915->display.bw.obj.state);
-	struct intel_cdclk_state *cdclk_state =
-		to_intel_cdclk_state(i915->display.cdclk.obj.state);
-	struct intel_dbuf_state *dbuf_state =
-		to_intel_dbuf_state(i915->display.dbuf.obj.state);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	struct intel_plane *plane;
 	struct drm_atomic_state *state;
 	struct intel_crtc_state *temp_crtc_state;
-	enum pipe pipe = crtc->pipe;
 	int ret;
 
 	if (!crtc_state->hw.active)
@@ -92,6 +84,21 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 		intel_unreference_shared_dpll_crtc(crtc,
 						   crtc_state->shared_dpll,
 						   &crtc_state->shared_dpll->state);
+}
+
+static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
+{
+	struct intel_encoder *encoder;
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_bw_state *bw_state =
+		to_intel_bw_state(i915->display.bw.obj.state);
+	struct intel_cdclk_state *cdclk_state =
+		to_intel_cdclk_state(i915->display.cdclk.obj.state);
+	struct intel_dbuf_state *dbuf_state =
+		to_intel_dbuf_state(i915->display.dbuf.obj.state);
+	struct intel_crtc_state *crtc_state =
+		to_intel_crtc_state(crtc->base.state);
+	enum pipe pipe = crtc->pipe;
 
 	__drm_atomic_helper_crtc_destroy_state(&crtc_state->uapi);
 	intel_crtc_free_hw_state(crtc_state);
@@ -115,6 +122,13 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 	bw_state->num_active_planes[pipe] = 0;
 }
 
+static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
+					struct drm_modeset_acquire_ctx *ctx)
+{
+	intel_crtc_disable_noatomic_begin(crtc, ctx);
+	intel_crtc_disable_noatomic_complete(crtc);
+}
+
 static void intel_modeset_update_connector_atomic_state(struct drm_i915_private *i915)
 {
 	struct intel_connector *connector;
-- 
GitLab


From 26281d3597254546161fb5d1dce80e1111a282fe Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:23 +0300
Subject: [PATCH 0519/1791] drm/i915: Factor out set_encoder_for_connector()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Factor out a function setting the encoder and CRTC in the connector
atomic state, required by a follow up patch.

No functional changes.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-7-imre.deak@intel.com
---
 .../drm/i915/display/intel_modeset_setup.c    | 28 +++++++++++++------
 1 file changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 2c93f4c5dc8cf..6f59654ea0261 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -86,6 +86,24 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 						   &crtc_state->shared_dpll->state);
 }
 
+static void set_encoder_for_connector(struct intel_connector *connector,
+				      struct intel_encoder *encoder)
+{
+	struct drm_connector_state *conn_state = connector->base.state;
+
+	if (conn_state->crtc)
+		drm_connector_put(&connector->base);
+
+	if (encoder) {
+		conn_state->best_encoder = &encoder->base;
+		conn_state->crtc = encoder->base.crtc;
+		drm_connector_get(&connector->base);
+	} else {
+		conn_state->best_encoder = NULL;
+		conn_state->crtc = NULL;
+	}
+}
+
 static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 {
 	struct intel_encoder *encoder;
@@ -140,8 +158,7 @@ static void intel_modeset_update_connector_atomic_state(struct drm_i915_private
 		struct intel_encoder *encoder =
 			to_intel_encoder(connector->base.encoder);
 
-		if (conn_state->crtc)
-			drm_connector_put(&connector->base);
+		set_encoder_for_connector(connector, encoder);
 
 		if (encoder) {
 			struct intel_crtc *crtc =
@@ -149,14 +166,7 @@ static void intel_modeset_update_connector_atomic_state(struct drm_i915_private
 			const struct intel_crtc_state *crtc_state =
 				to_intel_crtc_state(crtc->base.state);
 
-			conn_state->best_encoder = &encoder->base;
-			conn_state->crtc = &crtc->base;
 			conn_state->max_bpc = (crtc_state->pipe_bpp ?: 24) / 3;
-
-			drm_connector_get(&connector->base);
-		} else {
-			conn_state->best_encoder = NULL;
-			conn_state->crtc = NULL;
 		}
 	}
 	drm_connector_list_iter_end(&conn_iter);
-- 
GitLab


From e826839e18b77edb9be622a505d34e883985df48 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:24 +0300
Subject: [PATCH 0520/1791] drm/i915: Add support for disabling any CRTCs
 during HW readout/sanitization
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During HW readout/sanitization CRTCs can be disabled only if they don't
have an attached encoder (and so the encoder disable hooks don't need to
be called). An upcoming patch will need to disable CRTCs also with an
attached encoder, so add support for this.

For bigjoiner configs the encoder disabling hooks require the slave CRTC
states, so add these too to the atomic state. Since the connector atomic
state is already up-to-date when the CRTC is disabled the connector
state needs to be updated (reset) after the CRTC is disabled, make this
so. Follow the proper order of disabling first all bigjoiner slaves,
then any port synced CRTC slaves followed by the CRTC originally
requested to be disabled.

v2:
- Fix calculating the bigjoiner_masters mask in a port sync config,
  (Ville)
- Keep _noatomic suffix in intel_crtc_disable_noatomic(). (Ville)
- Rebase on full CRTC state reset in this patchset, not requiring
  resetting the bigjoiner state separately and (instead) resetting
  the full atomic CRTC and related global state after all linked
  pipes got disabled.
- Disable portsync slaves before a portsync master.
- Disable a portsync master if a linked portsync slave is disabled.

v3: (Ville)
- Use s/u32/u8 for transcoder and pipe masks.
- Use is_power_of_2() instead of hweight()==1.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-8-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  |   2 +-
 drivers/gpu/drm/i915/display/intel_display.h  |   1 +
 .../drm/i915/display/intel_modeset_setup.c    | 160 ++++++++++++++++--
 3 files changed, 152 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index e7ed5afd1d429..b3bac5dae20fa 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -225,7 +225,7 @@ is_trans_port_sync_slave(const struct intel_crtc_state *crtc_state)
 	return crtc_state->master_transcoder != INVALID_TRANSCODER;
 }
 
-static bool
+bool
 is_trans_port_sync_master(const struct intel_crtc_state *crtc_state)
 {
 	return crtc_state->sync_mode_slaves_mask != 0;
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index ee3def6e14a8a..205b3929b8614 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -407,6 +407,7 @@ intel_mode_valid_max_plane_size(struct drm_i915_private *dev_priv,
 				bool bigjoiner);
 enum phy intel_port_to_phy(struct drm_i915_private *i915, enum port port);
 bool is_trans_port_sync_mode(const struct intel_crtc_state *state);
+bool is_trans_port_sync_master(const struct intel_crtc_state *state);
 bool intel_crtc_is_bigjoiner_slave(const struct intel_crtc_state *crtc_state);
 bool intel_crtc_is_bigjoiner_master(const struct intel_crtc_state *crtc_state);
 u8 intel_crtc_bigjoiner_slave_pipes(const struct intel_crtc_state *crtc_state);
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 6f59654ea0261..75b4dea1e442b 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -38,8 +38,8 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 		to_intel_crtc_state(crtc->base.state);
 	struct intel_plane *plane;
 	struct drm_atomic_state *state;
-	struct intel_crtc_state *temp_crtc_state;
-	int ret;
+	struct intel_crtc *temp_crtc;
+	enum pipe pipe = crtc->pipe;
 
 	if (!crtc_state->hw.active)
 		return;
@@ -64,10 +64,17 @@ static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc,
 	to_intel_atomic_state(state)->internal = true;
 
 	/* Everything's already locked, -EDEADLK can't happen. */
-	temp_crtc_state = intel_atomic_get_crtc_state(state, crtc);
-	ret = drm_atomic_add_affected_connectors(state, &crtc->base);
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc,
+					 BIT(pipe) |
+					 intel_crtc_bigjoiner_slave_pipes(crtc_state)) {
+		struct intel_crtc_state *temp_crtc_state =
+			intel_atomic_get_crtc_state(state, temp_crtc);
+		int ret;
+
+		ret = drm_atomic_add_affected_connectors(state, &temp_crtc->base);
 
-	drm_WARN_ON(&i915->drm, IS_ERR(temp_crtc_state) || ret);
+		drm_WARN_ON(&i915->drm, IS_ERR(temp_crtc_state) || ret);
+	}
 
 	i915->display.funcs.display->crtc_disable(to_intel_atomic_state(state), crtc);
 
@@ -104,9 +111,38 @@ static void set_encoder_for_connector(struct intel_connector *connector,
 	}
 }
 
-static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
+static void reset_encoder_connector_state(struct intel_encoder *encoder)
 {
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_connector *connector;
+	struct drm_connector_list_iter conn_iter;
+
+	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
+	for_each_intel_connector_iter(connector, &conn_iter) {
+		if (connector->base.encoder != &encoder->base)
+			continue;
+
+		set_encoder_for_connector(connector, NULL);
+
+		connector->base.dpms = DRM_MODE_DPMS_OFF;
+		connector->base.encoder = NULL;
+	}
+	drm_connector_list_iter_end(&conn_iter);
+}
+
+static void reset_crtc_encoder_state(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_encoder *encoder;
+
+	for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder) {
+		reset_encoder_connector_state(encoder);
+		encoder->base.crtc = NULL;
+	}
+}
+
+static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
+{
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_bw_state *bw_state =
 		to_intel_bw_state(i915->display.bw.obj.state);
@@ -122,8 +158,7 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 	intel_crtc_free_hw_state(crtc_state);
 	intel_crtc_state_reset(crtc_state, crtc);
 
-	for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder)
-		encoder->base.crtc = NULL;
+	reset_crtc_encoder_state(crtc);
 
 	intel_fbc_disable(crtc);
 	intel_update_watermarks(i915);
@@ -140,11 +175,116 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 	bw_state->num_active_planes[pipe] = 0;
 }
 
+/*
+ * Return all the pipes using a transcoder in @transcoder_mask.
+ * For bigjoiner configs return only the bigjoiner master.
+ */
+static u8 get_transcoder_pipes(struct drm_i915_private *i915,
+			       u8 transcoder_mask)
+{
+	struct intel_crtc *temp_crtc;
+	u8 pipes = 0;
+
+	for_each_intel_crtc(&i915->drm, temp_crtc) {
+		struct intel_crtc_state *temp_crtc_state =
+			to_intel_crtc_state(temp_crtc->base.state);
+
+		if (temp_crtc_state->cpu_transcoder == INVALID_TRANSCODER)
+			continue;
+
+		if (intel_crtc_is_bigjoiner_slave(temp_crtc_state))
+			continue;
+
+		if (transcoder_mask & BIT(temp_crtc_state->cpu_transcoder))
+			pipes |= BIT(temp_crtc->pipe);
+	}
+
+	return pipes;
+}
+
+/*
+ * Return the port sync master and slave pipes linked to @crtc.
+ * For bigjoiner configs return only the bigjoiner master pipes.
+ */
+static void get_portsync_pipes(struct intel_crtc *crtc,
+			       u8 *master_pipe_mask, u8 *slave_pipes_mask)
+{
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	struct intel_crtc_state *crtc_state =
+		to_intel_crtc_state(crtc->base.state);
+	struct intel_crtc *master_crtc;
+	struct intel_crtc_state *master_crtc_state;
+	enum transcoder master_transcoder;
+
+	if (!is_trans_port_sync_mode(crtc_state)) {
+		*master_pipe_mask = BIT(crtc->pipe);
+		*slave_pipes_mask = 0;
+
+		return;
+	}
+
+	if (is_trans_port_sync_master(crtc_state))
+		master_transcoder = crtc_state->cpu_transcoder;
+	else
+		master_transcoder = crtc_state->master_transcoder;
+
+	*master_pipe_mask = get_transcoder_pipes(i915, BIT(master_transcoder));
+	drm_WARN_ON(&i915->drm, !is_power_of_2(*master_pipe_mask));
+
+	master_crtc = intel_crtc_for_pipe(i915, ffs(*master_pipe_mask) - 1);
+	master_crtc_state = to_intel_crtc_state(master_crtc->base.state);
+	*slave_pipes_mask = get_transcoder_pipes(i915, master_crtc_state->sync_mode_slaves_mask);
+}
+
+static u8 get_bigjoiner_slave_pipes(struct drm_i915_private *i915, u8 master_pipes_mask)
+{
+	struct intel_crtc *master_crtc;
+	u8 pipes = 0;
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, master_crtc, master_pipes_mask) {
+		struct intel_crtc_state *master_crtc_state =
+			to_intel_crtc_state(master_crtc->base.state);
+
+		pipes |= intel_crtc_bigjoiner_slave_pipes(master_crtc_state);
+	}
+
+	return pipes;
+}
+
 static void intel_crtc_disable_noatomic(struct intel_crtc *crtc,
 					struct drm_modeset_acquire_ctx *ctx)
 {
-	intel_crtc_disable_noatomic_begin(crtc, ctx);
-	intel_crtc_disable_noatomic_complete(crtc);
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+	u8 portsync_master_mask;
+	u8 portsync_slaves_mask;
+	u8 bigjoiner_slaves_mask;
+	struct intel_crtc *temp_crtc;
+
+	/* TODO: Add support for MST */
+	get_portsync_pipes(crtc, &portsync_master_mask, &portsync_slaves_mask);
+	bigjoiner_slaves_mask = get_bigjoiner_slave_pipes(i915,
+							  portsync_master_mask |
+							  portsync_slaves_mask);
+
+	drm_WARN_ON(&i915->drm,
+		    portsync_master_mask & portsync_slaves_mask ||
+		    portsync_master_mask & bigjoiner_slaves_mask ||
+		    portsync_slaves_mask & bigjoiner_slaves_mask);
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, bigjoiner_slaves_mask)
+		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_slaves_mask)
+		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_master_mask)
+		intel_crtc_disable_noatomic_begin(temp_crtc, ctx);
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc,
+					 bigjoiner_slaves_mask |
+					 portsync_slaves_mask |
+					 portsync_master_mask)
+		intel_crtc_disable_noatomic_complete(temp_crtc);
 }
 
 static void intel_modeset_update_connector_atomic_state(struct drm_i915_private *i915)
-- 
GitLab


From f48eab29028769bb83e5fc9ff0b0ece56b1ba5b2 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:25 +0300
Subject: [PATCH 0521/1791] drm/i915/dp: Add link training debug and error
 printing helpers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add functions for printing link training debug and error messages, both
to prepare for the next patch, which downgrades an error to a debug
message if the sink is disconnected and to remove some code duplication.

v2: (Ville)
- Always print the connector prefix.
- Preserve the drm_dbg_kms() debug category.
v3:
- Keep printing the name of functions calling the helpers. (Jani)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> (v2)
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-9-imre.deak@intel.com
---
 .../drm/i915/display/intel_dp_link_training.c | 367 ++++++------------
 1 file changed, 120 insertions(+), 247 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index e92c62bcc9b85..4f33b79b23db0 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -26,6 +26,23 @@
 #include "intel_dp.h"
 #include "intel_dp_link_training.h"
 
+#define LT_MSG_PREFIX			"[CONNECTOR:%d:%s][ENCODER:%d:%s][%s] "
+#define LT_MSG_ARGS(_intel_dp, _dp_phy)	(_intel_dp)->attached_connector->base.base.id, \
+					(_intel_dp)->attached_connector->base.name, \
+					dp_to_dig_port(_intel_dp)->base.base.base.id, \
+					dp_to_dig_port(_intel_dp)->base.base.name, \
+					drm_dp_phy_name(_dp_phy)
+
+#define lt_dbg(_intel_dp, _dp_phy, _format, ...) \
+	drm_dbg_kms(&dp_to_i915(_intel_dp)->drm, \
+		    LT_MSG_PREFIX _format, \
+		    LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__)
+
+#define lt_err(_intel_dp, _dp_phy, _format, ...) \
+	drm_err(&dp_to_i915(_intel_dp)->drm, \
+		LT_MSG_PREFIX _format, \
+		LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__)
+
 static void intel_dp_reset_lttpr_common_caps(struct intel_dp *intel_dp)
 {
 	memset(intel_dp->lttpr_common_caps, 0, sizeof(intel_dp->lttpr_common_caps));
@@ -47,29 +64,21 @@ static void intel_dp_read_lttpr_phy_caps(struct intel_dp *intel_dp,
 					 const u8 dpcd[DP_RECEIVER_CAP_SIZE],
 					 enum drm_dp_phy dp_phy)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	u8 *phy_caps = intel_dp_lttpr_phy_caps(intel_dp, dp_phy);
 
 	if (drm_dp_read_lttpr_phy_caps(&intel_dp->aux, dpcd, dp_phy, phy_caps) < 0) {
-		drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
-			    "[ENCODER:%d:%s][%s] failed to read the PHY caps\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy));
+		lt_dbg(intel_dp, dp_phy, "failed to read the PHY caps\n");
 		return;
 	}
 
-	drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
-		    "[ENCODER:%d:%s][%s] PHY capabilities: %*ph\n",
-		    encoder->base.base.id, encoder->base.name,
-		    drm_dp_phy_name(dp_phy),
-		    (int)sizeof(intel_dp->lttpr_phy_caps[0]),
-		    phy_caps);
+	lt_dbg(intel_dp, dp_phy, "PHY capabilities: %*ph\n",
+	       (int)sizeof(intel_dp->lttpr_phy_caps[0]),
+	       phy_caps);
 }
 
 static bool intel_dp_read_lttpr_common_caps(struct intel_dp *intel_dp,
 					    const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	int ret;
 
 	ret = drm_dp_read_lttpr_common_caps(&intel_dp->aux, dpcd,
@@ -77,11 +86,9 @@ static bool intel_dp_read_lttpr_common_caps(struct intel_dp *intel_dp,
 	if (ret < 0)
 		goto reset_caps;
 
-	drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
-		    "[ENCODER:%d:%s] LTTPR common capabilities: %*ph\n",
-		    encoder->base.base.id, encoder->base.name,
-		    (int)sizeof(intel_dp->lttpr_common_caps),
-		    intel_dp->lttpr_common_caps);
+	lt_dbg(intel_dp, DP_PHY_DPRX, "LTTPR common capabilities: %*ph\n",
+	       (int)sizeof(intel_dp->lttpr_common_caps),
+	       intel_dp->lttpr_common_caps);
 
 	/* The minimum value of LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV is 1.4 */
 	if (intel_dp->lttpr_common_caps[0] < 0x14)
@@ -105,8 +112,6 @@ intel_dp_set_lttpr_transparent_mode(struct intel_dp *intel_dp, bool enable)
 
 static int intel_dp_init_lttpr(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	int lttpr_count;
 	int i;
 
@@ -138,9 +143,8 @@ static int intel_dp_init_lttpr(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEI
 		return 0;
 
 	if (!intel_dp_set_lttpr_transparent_mode(intel_dp, false)) {
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s] Switching to LTTPR non-transparent LT mode failed, fall-back to transparent mode\n",
-			    encoder->base.base.id, encoder->base.name);
+		lt_dbg(intel_dp, DP_PHY_DPRX,
+		       "Switching to LTTPR non-transparent LT mode failed, fall-back to transparent mode\n");
 
 		intel_dp_set_lttpr_transparent_mode(intel_dp, true);
 		intel_dp_reset_lttpr_count(intel_dp);
@@ -409,26 +413,22 @@ intel_dp_get_adjust_train(struct intel_dp *intel_dp,
 			  enum drm_dp_phy dp_phy,
 			  const u8 link_status[DP_LINK_STATUS_SIZE])
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	int lane;
 
 	if (intel_dp_is_uhbr(crtc_state)) {
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 128b/132b, lanes: %d, "
-			    "TX FFE request: " TRAIN_REQ_FMT "\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy),
-			    crtc_state->lane_count,
-			    TRAIN_REQ_TX_FFE_ARGS(link_status));
+		lt_dbg(intel_dp, dp_phy,
+		       "128b/132b, lanes: %d, "
+		       "TX FFE request: " TRAIN_REQ_FMT "\n",
+		       crtc_state->lane_count,
+		       TRAIN_REQ_TX_FFE_ARGS(link_status));
 	} else {
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 8b/10b, lanes: %d, "
-			    "vswing request: " TRAIN_REQ_FMT ", "
-			    "pre-emphasis request: " TRAIN_REQ_FMT "\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy),
-			    crtc_state->lane_count,
-			    TRAIN_REQ_VSWING_ARGS(link_status),
-			    TRAIN_REQ_PREEMPH_ARGS(link_status));
+		lt_dbg(intel_dp, dp_phy,
+		       "8b/10b, lanes: %d, "
+		       "vswing request: " TRAIN_REQ_FMT ", "
+		       "pre-emphasis request: " TRAIN_REQ_FMT "\n",
+		       crtc_state->lane_count,
+		       TRAIN_REQ_VSWING_ARGS(link_status),
+		       TRAIN_REQ_PREEMPH_ARGS(link_status));
 	}
 
 	for (lane = 0; lane < 4; lane++)
@@ -487,16 +487,11 @@ intel_dp_program_link_training_pattern(struct intel_dp *intel_dp,
 				       enum drm_dp_phy dp_phy,
 				       u8 dp_train_pat)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	u8 train_pat = intel_dp_training_pattern_symbol(dp_train_pat);
 
 	if (train_pat != DP_TRAINING_PATTERN_DISABLE)
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s][%s] Using DP training pattern TPS%c\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy),
-			    dp_training_pattern_name(train_pat));
+		lt_dbg(intel_dp, dp_phy, "Using DP training pattern TPS%c\n",
+		       dp_training_pattern_name(train_pat));
 
 	intel_dp->set_link_train(intel_dp, crtc_state, dp_train_pat);
 }
@@ -531,24 +526,21 @@ void intel_dp_set_signal_levels(struct intel_dp *intel_dp,
 				enum drm_dp_phy dp_phy)
 {
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 
 	if (intel_dp_is_uhbr(crtc_state)) {
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 128b/132b, lanes: %d, "
-			    "TX FFE presets: " TRAIN_SET_FMT "\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy),
-			    crtc_state->lane_count,
-			    TRAIN_SET_TX_FFE_ARGS(intel_dp->train_set));
+		lt_dbg(intel_dp, dp_phy,
+		       "128b/132b, lanes: %d, "
+		       "TX FFE presets: " TRAIN_SET_FMT "\n",
+		       crtc_state->lane_count,
+		       TRAIN_SET_TX_FFE_ARGS(intel_dp->train_set));
 	} else {
-		drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 8b/10b, lanes: %d, "
-			    "vswing levels: " TRAIN_SET_FMT ", "
-			    "pre-emphasis levels: " TRAIN_SET_FMT "\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy),
-			    crtc_state->lane_count,
-			    TRAIN_SET_VSWING_ARGS(intel_dp->train_set),
-			    TRAIN_SET_PREEMPH_ARGS(intel_dp->train_set));
+		lt_dbg(intel_dp, dp_phy,
+		       "8b/10b, lanes: %d, "
+		       "vswing levels: " TRAIN_SET_FMT ", "
+		       "pre-emphasis levels: " TRAIN_SET_FMT "\n",
+		       crtc_state->lane_count,
+		       TRAIN_SET_VSWING_ARGS(intel_dp->train_set),
+		       TRAIN_SET_PREEMPH_ARGS(intel_dp->train_set));
 	}
 
 	if (intel_dp_phy_is_downstream_of_source(intel_dp, dp_phy))
@@ -677,8 +669,6 @@ static bool
 intel_dp_prepare_link_train(struct intel_dp *intel_dp,
 			    const struct intel_crtc_state *crtc_state)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	u8 link_bw, rate_select;
 
 	if (intel_dp->prepare_link_retrain)
@@ -699,24 +689,21 @@ intel_dp_prepare_link_train(struct intel_dp *intel_dp,
 	 * link rates are not stable.
 	 */
 	if (!link_bw) {
-		struct intel_connector *connector = intel_dp->attached_connector;
 		__le16 sink_rates[DP_MAX_SUPPORTED_RATES];
 
-		drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] Reloading eDP link rates\n",
-			    connector->base.base.id, connector->base.name);
+		lt_dbg(intel_dp, DP_PHY_DPRX, "Reloading eDP link rates\n");
 
 		drm_dp_dpcd_read(&intel_dp->aux, DP_SUPPORTED_LINK_RATES,
 				 sink_rates, sizeof(sink_rates));
 	}
 
 	if (link_bw)
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s] Using LINK_BW_SET value %02x\n",
-			    encoder->base.base.id, encoder->base.name, link_bw);
+		lt_dbg(intel_dp, DP_PHY_DPRX, "Using LINK_BW_SET value %02x\n",
+		       link_bw);
 	else
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s] Using LINK_RATE_SET value %02x\n",
-			    encoder->base.base.id, encoder->base.name, rate_select);
+		lt_dbg(intel_dp, DP_PHY_DPRX,
+		       "Using LINK_RATE_SET value %02x\n",
+		       rate_select);
 	/*
 	 * Spec DP2.1 Section 3.5.2.16
 	 * Prior to LT DPTX should set 128b/132b DP Channel coding and then set link rate
@@ -758,15 +745,10 @@ void
 intel_dp_dump_link_status(struct intel_dp *intel_dp, enum drm_dp_phy dp_phy,
 			  const u8 link_status[DP_LINK_STATUS_SIZE])
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
-
-	drm_dbg_kms(&i915->drm,
-		    "[ENCODER:%d:%s][%s] ln0_1:0x%x ln2_3:0x%x align:0x%x sink:0x%x adj_req0_1:0x%x adj_req2_3:0x%x\n",
-		    encoder->base.base.id, encoder->base.name,
-		    drm_dp_phy_name(dp_phy),
-		    link_status[0], link_status[1], link_status[2],
-		    link_status[3], link_status[4], link_status[5]);
+	lt_dbg(intel_dp, dp_phy,
+	       "ln0_1:0x%x ln2_3:0x%x align:0x%x sink:0x%x adj_req0_1:0x%x adj_req2_3:0x%x\n",
+	       link_status[0], link_status[1], link_status[2],
+	       link_status[3], link_status[4], link_status[5]);
 }
 
 /*
@@ -778,8 +760,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 				      const struct intel_crtc_state *crtc_state,
 				      enum drm_dp_phy dp_phy)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	u8 old_link_status[DP_LINK_STATUS_SIZE] = {};
 	int voltage_tries, cr_tries, max_cr_tries;
 	u8 link_status[DP_LINK_STATUS_SIZE];
@@ -794,9 +774,7 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 	if (!intel_dp_reset_link_train(intel_dp, crtc_state, dp_phy,
 				       DP_TRAINING_PATTERN_1 |
 				       DP_LINK_SCRAMBLING_DISABLE)) {
-		drm_err(&i915->drm, "[ENCODER:%d:%s][%s] Failed to enable link training\n",
-			encoder->base.base.id, encoder->base.name,
-			drm_dp_phy_name(dp_phy));
+		lt_err(intel_dp, dp_phy, "Failed to enable link training\n");
 		return false;
 	}
 
@@ -819,35 +797,24 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 
 		if (drm_dp_dpcd_read_phy_link_status(&intel_dp->aux, dp_phy,
 						     link_status) < 0) {
-			drm_err(&i915->drm, "[ENCODER:%d:%s][%s] Failed to get link status\n",
-				encoder->base.base.id, encoder->base.name,
-				drm_dp_phy_name(dp_phy));
+			lt_err(intel_dp, dp_phy, "Failed to get link status\n");
 			return false;
 		}
 
 		if (drm_dp_clock_recovery_ok(link_status, crtc_state->lane_count)) {
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s][%s] Clock recovery OK\n",
-				    encoder->base.base.id, encoder->base.name,
-				    drm_dp_phy_name(dp_phy));
+			lt_dbg(intel_dp, dp_phy, "Clock recovery OK\n");
 			return true;
 		}
 
 		if (voltage_tries == 5) {
 			intel_dp_dump_link_status(intel_dp, dp_phy, link_status);
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s][%s] Same voltage tried 5 times\n",
-				    encoder->base.base.id, encoder->base.name,
-				    drm_dp_phy_name(dp_phy));
+			lt_dbg(intel_dp, dp_phy, "Same voltage tried 5 times\n");
 			return false;
 		}
 
 		if (max_vswing_reached) {
 			intel_dp_dump_link_status(intel_dp, dp_phy, link_status);
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s][%s] Max Voltage Swing reached\n",
-				    encoder->base.base.id, encoder->base.name,
-				    drm_dp_phy_name(dp_phy));
+			lt_dbg(intel_dp, dp_phy, "Max Voltage Swing reached\n");
 			return false;
 		}
 
@@ -855,10 +822,7 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 		intel_dp_get_adjust_train(intel_dp, crtc_state, dp_phy,
 					  link_status);
 		if (!intel_dp_update_link_train(intel_dp, crtc_state, dp_phy)) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s][%s] Failed to update link training\n",
-				encoder->base.base.id, encoder->base.name,
-				drm_dp_phy_name(dp_phy));
+			lt_err(intel_dp, dp_phy, "Failed to update link training\n");
 			return false;
 		}
 
@@ -874,10 +838,8 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 	}
 
 	intel_dp_dump_link_status(intel_dp, dp_phy, link_status);
-	drm_err(&i915->drm,
-		"[ENCODER:%d:%s][%s] Failed clock recovery %d times, giving up!\n",
-		encoder->base.base.id, encoder->base.name,
-		drm_dp_phy_name(dp_phy), max_cr_tries);
+	lt_err(intel_dp, dp_phy, "Failed clock recovery %d times, giving up!\n",
+	       max_cr_tries);
 
 	return false;
 }
@@ -911,11 +873,11 @@ static u32 intel_dp_training_pattern(struct intel_dp *intel_dp,
 		return DP_TRAINING_PATTERN_4;
 	} else if (crtc_state->port_clock == 810000) {
 		if (!source_tps4)
-			drm_dbg_kms(&i915->drm,
-				    "8.1 Gbps link rate without source TPS4 support\n");
+			lt_dbg(intel_dp, dp_phy,
+			       "8.1 Gbps link rate without source TPS4 support\n");
 		if (!sink_tps4)
-			drm_dbg_kms(&i915->drm,
-				    "8.1 Gbps link rate without sink TPS4 support\n");
+			lt_dbg(intel_dp, dp_phy,
+			       "8.1 Gbps link rate without sink TPS4 support\n");
 	}
 
 	/*
@@ -929,11 +891,11 @@ static u32 intel_dp_training_pattern(struct intel_dp *intel_dp,
 		return  DP_TRAINING_PATTERN_3;
 	} else if (crtc_state->port_clock >= 540000) {
 		if (!source_tps3)
-			drm_dbg_kms(&i915->drm,
-				    ">=5.4/6.48 Gbps link rate without source TPS3 support\n");
+			lt_dbg(intel_dp, dp_phy,
+			       ">=5.4/6.48 Gbps link rate without source TPS3 support\n");
 		if (!sink_tps3)
-			drm_dbg_kms(&i915->drm,
-				    ">=5.4/6.48 Gbps link rate without sink TPS3 support\n");
+			lt_dbg(intel_dp, dp_phy,
+			       ">=5.4/6.48 Gbps link rate without sink TPS3 support\n");
 	}
 
 	return DP_TRAINING_PATTERN_2;
@@ -949,8 +911,6 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 					    const struct intel_crtc_state *crtc_state,
 					    enum drm_dp_phy dp_phy)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	int tries;
 	u32 training_pattern;
 	u8 link_status[DP_LINK_STATUS_SIZE];
@@ -969,10 +929,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 	/* channel equalization */
 	if (!intel_dp_set_link_train(intel_dp, crtc_state, dp_phy,
 				     training_pattern)) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s][%s] Failed to start channel equalization\n",
-			encoder->base.base.id, encoder->base.name,
-			drm_dp_phy_name(dp_phy));
+		lt_err(intel_dp, dp_phy, "Failed to start channel equalization\n");
 		return false;
 	}
 
@@ -981,10 +938,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 
 		if (drm_dp_dpcd_read_phy_link_status(&intel_dp->aux, dp_phy,
 						     link_status) < 0) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s][%s] Failed to get link status\n",
-				encoder->base.base.id, encoder->base.name,
-				drm_dp_phy_name(dp_phy));
+			lt_err(intel_dp, dp_phy, "Failed to get link status\n");
 			break;
 		}
 
@@ -992,21 +946,15 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 		if (!drm_dp_clock_recovery_ok(link_status,
 					      crtc_state->lane_count)) {
 			intel_dp_dump_link_status(intel_dp, dp_phy, link_status);
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s][%s] Clock recovery check failed, cannot "
-				    "continue channel equalization\n",
-				    encoder->base.base.id, encoder->base.name,
-				    drm_dp_phy_name(dp_phy));
+			lt_dbg(intel_dp, dp_phy,
+			       "Clock recovery check failed, cannot continue channel equalization\n");
 			break;
 		}
 
 		if (drm_dp_channel_eq_ok(link_status,
 					 crtc_state->lane_count)) {
 			channel_eq = true;
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s][%s] Channel EQ done. DP Training successful\n",
-				    encoder->base.base.id, encoder->base.name,
-				    drm_dp_phy_name(dp_phy));
+			lt_dbg(intel_dp, dp_phy, "Channel EQ done. DP Training successful\n");
 			break;
 		}
 
@@ -1014,10 +962,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 		intel_dp_get_adjust_train(intel_dp, crtc_state, dp_phy,
 					  link_status);
 		if (!intel_dp_update_link_train(intel_dp, crtc_state, dp_phy)) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s][%s] Failed to update link training\n",
-				encoder->base.base.id, encoder->base.name,
-				drm_dp_phy_name(dp_phy));
+			lt_err(intel_dp, dp_phy, "Failed to update link training\n");
 			break;
 		}
 	}
@@ -1025,10 +970,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp,
 	/* Try 5 times, else fail and try at lower BW */
 	if (tries == 5) {
 		intel_dp_dump_link_status(intel_dp, dp_phy, link_status);
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s][%s] Channel equalization failed 5 times\n",
-			    encoder->base.base.id, encoder->base.name,
-			    drm_dp_phy_name(dp_phy));
+		lt_dbg(intel_dp, dp_phy, "Channel equalization failed 5 times\n");
 	}
 
 	return channel_eq;
@@ -1047,13 +989,12 @@ static int
 intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp,
 			    const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	u8 sink_status;
 	int ret;
 
 	ret = drm_dp_dpcd_readb(&intel_dp->aux, DP_SINK_STATUS, &sink_status);
 	if (ret != 1) {
-		drm_dbg_kms(&i915->drm, "Failed to read sink status\n");
+		lt_dbg(intel_dp, DP_PHY_DPRX, "Failed to read sink status\n");
 		return ret < 0 ? ret : -EIO;
 	}
 
@@ -1079,9 +1020,6 @@ intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp,
 void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 			      const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-
 	intel_dp->link_trained = true;
 
 	intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
@@ -1090,9 +1028,7 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 
 	if (intel_dp_is_uhbr(crtc_state) &&
 	    wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) {
-		drm_dbg_kms(&i915->drm,
-			    "[ENCODER:%d:%s] 128b/132b intra-hop not clearing\n",
-			    encoder->base.base.id, encoder->base.name);
+		lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n");
 	}
 }
 
@@ -1101,8 +1037,6 @@ intel_dp_link_train_phy(struct intel_dp *intel_dp,
 			const struct intel_crtc_state *crtc_state,
 			enum drm_dp_phy dp_phy)
 {
-	struct intel_connector *connector = intel_dp->attached_connector;
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	bool ret = false;
 
 	if (!intel_dp_link_training_clock_recovery(intel_dp, crtc_state, dp_phy))
@@ -1114,13 +1048,10 @@ intel_dp_link_train_phy(struct intel_dp *intel_dp,
 	ret = true;
 
 out:
-	drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
-		    "[CONNECTOR:%d:%s][ENCODER:%d:%s][%s] Link Training %s at link rate = %d, lane count = %d\n",
-		    connector->base.base.id, connector->base.name,
-		    encoder->base.base.id, encoder->base.name,
-		    drm_dp_phy_name(dp_phy),
-		    ret ? "passed" : "failed",
-		    crtc_state->port_clock, crtc_state->lane_count);
+	lt_dbg(intel_dp, dp_phy,
+	       "Link Training %s at link rate = %d, lane count = %d\n",
+	       ret ? "passed" : "failed",
+	       crtc_state->port_clock, crtc_state->lane_count);
 
 	return ret;
 }
@@ -1129,13 +1060,10 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp,
 						     const struct intel_crtc_state *crtc_state)
 {
 	struct intel_connector *intel_connector = intel_dp->attached_connector;
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 
 	if (intel_dp->hobl_active) {
-		drm_dbg_kms(&dp_to_i915(intel_dp)->drm,
-			    "[ENCODER:%d:%s] Link Training failed with HOBL active, "
-			    "not enabling it from now on",
-			    encoder->base.base.id, encoder->base.name);
+		lt_dbg(intel_dp, DP_PHY_DPRX,
+		       "Link Training failed with HOBL active, not enabling it from now on\n");
 		intel_dp->hobl_failed = true;
 	} else if (intel_dp_get_link_train_fallback_values(intel_dp,
 							   crtc_state->port_clock,
@@ -1182,8 +1110,6 @@ static bool
 intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 			  const struct intel_crtc_state *crtc_state)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	int delay_us;
 	int try, max_tries = 20;
@@ -1198,9 +1124,7 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 	 */
 	if (!intel_dp_reset_link_train(intel_dp, crtc_state, DP_PHY_DPRX,
 				       DP_TRAINING_PATTERN_1)) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Failed to start 128b/132b TPS1\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS1\n");
 		return false;
 	}
 
@@ -1208,27 +1132,21 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 
 	/* Read the initial TX FFE settings. */
 	if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Failed to read TX FFE presets\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Failed to read TX FFE presets\n");
 		return false;
 	}
 
 	/* Update signal levels and training set as requested. */
 	intel_dp_get_adjust_train(intel_dp, crtc_state, DP_PHY_DPRX, link_status);
 	if (!intel_dp_update_link_train(intel_dp, crtc_state, DP_PHY_DPRX)) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Failed to set initial TX FFE settings\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Failed to set initial TX FFE settings\n");
 		return false;
 	}
 
 	/* Start transmitting 128b/132b TPS2. */
 	if (!intel_dp_set_link_train(intel_dp, crtc_state, DP_PHY_DPRX,
 				     DP_TRAINING_PATTERN_2)) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Failed to start 128b/132b TPS2\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS2\n");
 		return false;
 	}
 
@@ -1245,32 +1163,25 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 		delay_us = drm_dp_128b132b_read_aux_rd_interval(&intel_dp->aux);
 
 		if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Failed to read link status\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n");
 			return false;
 		}
 
 		if (drm_dp_128b132b_link_training_failed(link_status)) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Downstream link training failure\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX,
+			       "Downstream link training failure\n");
 			return false;
 		}
 
 		if (drm_dp_128b132b_lane_channel_eq_done(link_status, crtc_state->lane_count)) {
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s] Lane channel eq done\n",
-				    encoder->base.base.id, encoder->base.name);
+			lt_dbg(intel_dp, DP_PHY_DPRX, "Lane channel eq done\n");
 			break;
 		}
 
 		if (timeout) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Lane channel eq timeout\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Lane channel eq timeout\n");
 			return false;
 		}
 
@@ -1280,18 +1191,14 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 		/* Update signal levels and training set as requested. */
 		intel_dp_get_adjust_train(intel_dp, crtc_state, DP_PHY_DPRX, link_status);
 		if (!intel_dp_update_link_train(intel_dp, crtc_state, DP_PHY_DPRX)) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Failed to update TX FFE settings\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Failed to update TX FFE settings\n");
 			return false;
 		}
 	}
 
 	if (try == max_tries) {
 		intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Max loop count reached\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Max loop count reached\n");
 		return false;
 	}
 
@@ -1300,32 +1207,24 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp,
 			timeout = true; /* try one last time after deadline */
 
 		if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Failed to read link status\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n");
 			return false;
 		}
 
 		if (drm_dp_128b132b_link_training_failed(link_status)) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Downstream link training failure\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Downstream link training failure\n");
 			return false;
 		}
 
 		if (drm_dp_128b132b_eq_interlane_align_done(link_status)) {
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s] Interlane align done\n",
-				    encoder->base.base.id, encoder->base.name);
+			lt_dbg(intel_dp, DP_PHY_DPRX, "Interlane align done\n");
 			break;
 		}
 
 		if (timeout) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Interlane align timeout\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Interlane align timeout\n");
 			return false;
 		}
 
@@ -1343,16 +1242,12 @@ intel_dp_128b132b_lane_cds(struct intel_dp *intel_dp,
 			   const struct intel_crtc_state *crtc_state,
 			   int lttpr_count)
 {
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	unsigned long deadline;
 
 	if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TRAINING_PATTERN_SET,
 			       DP_TRAINING_PATTERN_2_CDS) != 1) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] Failed to start 128b/132b TPS2 CDS\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS2 CDS\n");
 		return false;
 	}
 
@@ -1368,34 +1263,26 @@ intel_dp_128b132b_lane_cds(struct intel_dp *intel_dp,
 		usleep_range(2000, 3000);
 
 		if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) {
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Failed to read link status\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n");
 			return false;
 		}
 
 		if (drm_dp_128b132b_eq_interlane_align_done(link_status) &&
 		    drm_dp_128b132b_cds_interlane_align_done(link_status) &&
 		    drm_dp_128b132b_lane_symbol_locked(link_status, crtc_state->lane_count)) {
-			drm_dbg_kms(&i915->drm,
-				    "[ENCODER:%d:%s] CDS interlane align done\n",
-				    encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "CDS interlane align done\n");
 			break;
 		}
 
 		if (drm_dp_128b132b_link_training_failed(link_status)) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] Downstream link training failure\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "Downstream link training failure\n");
 			return false;
 		}
 
 		if (timeout) {
 			intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status);
-			drm_err(&i915->drm,
-				"[ENCODER:%d:%s] CDS timeout\n",
-				encoder->base.base.id, encoder->base.name);
+			lt_err(intel_dp, DP_PHY_DPRX, "CDS timeout\n");
 			return false;
 		}
 	}
@@ -1411,15 +1298,10 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp,
 			     const struct intel_crtc_state *crtc_state,
 			     int lttpr_count)
 {
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	struct intel_connector *connector = intel_dp->attached_connector;
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	bool passed = false;
 
 	if (wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) {
-		drm_err(&i915->drm,
-			"[ENCODER:%d:%s] 128b/132b intra-hop not clear\n",
-			encoder->base.base.id, encoder->base.name);
+		lt_err(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clear\n");
 		return false;
 	}
 
@@ -1427,12 +1309,10 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp,
 	    intel_dp_128b132b_lane_cds(intel_dp, crtc_state, lttpr_count))
 		passed = true;
 
-	drm_dbg_kms(&i915->drm,
-		    "[CONNECTOR:%d:%s][ENCODER:%d:%s] 128b/132b Link Training %s at link rate = %d, lane count = %d\n",
-		    connector->base.base.id, connector->base.name,
-		    encoder->base.base.id, encoder->base.name,
-		    passed ? "passed" : "failed",
-		    crtc_state->port_clock, crtc_state->lane_count);
+	lt_dbg(intel_dp, DP_PHY_DPRX,
+	       "128b/132b Link Training %s at link rate = %d, lane count = %d\n",
+	       passed ? "passed" : "failed",
+	       crtc_state->port_clock, crtc_state->lane_count);
 
 	return passed;
 }
@@ -1451,8 +1331,6 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp,
 			       const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	struct intel_connector *connector = intel_dp->attached_connector;
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
 	bool passed;
 
 	/*
@@ -1485,10 +1363,7 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp,
 	 * ignore_long_hpd flag can unset from the testcase.
 	 */
 	if (!passed && i915->display.hotplug.ignore_long_hpd) {
-		drm_dbg_kms(&i915->drm,
-			    "[CONNECTOR:%d:%s][ENCODER:%d:%s] Ignore the link failure\n",
-			    connector->base.base.id, connector->base.name,
-			    encoder->base.base.id, encoder->base.name);
+		lt_dbg(intel_dp, DP_PHY_DPRX, "Ignore the link failure\n");
 		return;
 	}
 
@@ -1499,8 +1374,6 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp,
 void intel_dp_128b132b_sdp_crc16(struct intel_dp *intel_dp,
 				 const struct intel_crtc_state *crtc_state)
 {
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-
 	/*
 	 * VIDEO_DIP_CTL register bit 31 should be set to '0' to not
 	 * disable SDP CRC. This is applicable for Display version 13.
@@ -1513,5 +1386,5 @@ void intel_dp_128b132b_sdp_crc16(struct intel_dp *intel_dp,
 				   DP_SDP_ERROR_DETECTION_CONFIGURATION,
 				   DP_SDP_CRC16_128B132B_EN);
 
-	drm_dbg_kms(&i915->drm, "DP2.0 SDP CRC16 for 128b/132b enabled\n");
+	lt_dbg(intel_dp, DP_PHY_DPRX, "DP2.0 SDP CRC16 for 128b/132b enabled\n");
 }
-- 
GitLab


From 9031c6d4f78a1e801695d0c4d97724455c6e2a61 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:26 +0300
Subject: [PATCH 0522/1791] drm/i915/dp: Convert link training error to debug
 message on disconnected sink
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If a sink is disconnected it's expected that link training actions will
fail on it, so downgrade the error messages about such actions to be a
debug message. Such - expected - link training failures are more
frequent after a follow up patch, after which an active TypeC link is
reset after the sink is disconnected which also involves a link
training.

v2:
- Check the actual HPD state to handle the forced connector state case.
  (Vinod)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> (v1)
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-10-imre.deak@intel.com
---
 .../gpu/drm/i915/display/intel_dp_link_training.c    | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 4f33b79b23db0..51d1e4b4b2f19 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -38,10 +38,14 @@
 		    LT_MSG_PREFIX _format, \
 		    LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__)
 
-#define lt_err(_intel_dp, _dp_phy, _format, ...) \
-	drm_err(&dp_to_i915(_intel_dp)->drm, \
-		LT_MSG_PREFIX _format, \
-		LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__)
+#define lt_err(_intel_dp, _dp_phy, _format, ...) do { \
+	if (intel_digital_port_connected(&dp_to_dig_port(_intel_dp)->base)) \
+		drm_err(&dp_to_i915(_intel_dp)->drm, \
+			LT_MSG_PREFIX _format, \
+			LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__); \
+	else \
+		lt_dbg(_intel_dp, _dp_phy, "Sink disconnected: " _format, ## __VA_ARGS__); \
+} while (0)
 
 static void intel_dp_reset_lttpr_common_caps(struct intel_dp *intel_dp)
 {
-- 
GitLab


From f45156ff18bae00ee56ed6aa2a937a8e93e56d7f Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:27 +0300
Subject: [PATCH 0523/1791] drm/i915/dp: Prevent link training fallback on
 disconnected port
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prevent downgrading the link training maximum lane count/rate if the
sink is disconnected - and so the link training failure is expected. In
such cases modeset failures due to the reduced max link params would be
just confusing for user space (instead of which the correct thing it
should act on is the sink disconnect signaled by a hotplug event,
requiring a disabling modeset).

v2:
- Check the actual HPD state to handle the forced connector state case.
  (Vinod)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> (v1)
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-11-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_link_training.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 51d1e4b4b2f19..0952a707358c1 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1065,6 +1065,11 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp,
 {
 	struct intel_connector *intel_connector = intel_dp->attached_connector;
 
+	if (!intel_digital_port_connected(&dp_to_dig_port(intel_dp)->base)) {
+		lt_dbg(intel_dp, DP_PHY_DPRX, "Link Training failed on disconnected sink.\n");
+		return;
+	}
+
 	if (intel_dp->hobl_active) {
 		lt_dbg(intel_dp, DP_PHY_DPRX,
 		       "Link Training failed with HOBL active, not enabling it from now on\n");
-- 
GitLab


From 7e4460c34b011ae15a898256ad0682e3f34a94d1 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:28 +0300
Subject: [PATCH 0524/1791] drm/i915/dp: Factor out intel_dp_get_active_pipes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Factor out a helper used by a follow up patch to reset an active DP
link.

No functional changes.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-12-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index b35ab251f5435..088d67c448248 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -4232,9 +4232,9 @@ static bool intel_dp_has_connector(struct intel_dp *intel_dp,
 	return false;
 }
 
-static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp,
-				      struct drm_modeset_acquire_ctx *ctx,
-				      u8 *pipe_mask)
+static int intel_dp_get_active_pipes(struct intel_dp *intel_dp,
+				     struct drm_modeset_acquire_ctx *ctx,
+				     u8 *pipe_mask)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	struct drm_connector_list_iter conn_iter;
@@ -4243,9 +4243,6 @@ static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp,
 
 	*pipe_mask = 0;
 
-	if (!intel_dp_needs_link_retrain(intel_dp))
-		return 0;
-
 	drm_connector_list_iter_begin(&i915->drm, &conn_iter);
 	for_each_intel_connector_iter(connector, &conn_iter) {
 		struct drm_connector_state *conn_state =
@@ -4279,9 +4276,6 @@ static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp,
 	}
 	drm_connector_list_iter_end(&conn_iter);
 
-	if (!intel_dp_needs_link_retrain(intel_dp))
-		*pipe_mask = 0;
-
 	return ret;
 }
 
@@ -4310,13 +4304,19 @@ int intel_dp_retrain_link(struct intel_encoder *encoder,
 	if (ret)
 		return ret;
 
-	ret = intel_dp_prep_link_retrain(intel_dp, ctx, &pipe_mask);
+	if (!intel_dp_needs_link_retrain(intel_dp))
+		return 0;
+
+	ret = intel_dp_get_active_pipes(intel_dp, ctx, &pipe_mask);
 	if (ret)
 		return ret;
 
 	if (pipe_mask == 0)
 		return 0;
 
+	if (!intel_dp_needs_link_retrain(intel_dp))
+		return 0;
+
 	drm_dbg_kms(&dev_priv->drm, "[ENCODER:%d:%s] retraining link\n",
 		    encoder->base.base.id, encoder->base.name);
 
-- 
GitLab


From 60ded7cc86f363161e37dc41c548b2ab3e1af5ce Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 10 May 2023 13:31:29 +0300
Subject: [PATCH 0525/1791] drm/i915: Factor out a helper for handling atomic
 modeset locks/state
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch simplifying the handling of modeset locks and atomic state
for an atomic commit is based on

https://lore.kernel.org/all/20210715184954.7794-2-ville.syrjala@linux.intel.com/

adding the helper to i915. I find this approach preferrable than
open-coding the corresponding steps (fixed for me an atomic
state reset during a DEADLK retry, which I missed in the open-coded
version) and also better than the existing
DRM_MODESET_LOCK_ALL_BEGIN/END macros for the reasons described in the
above original patchset.

This change takes the helper into use only for atomic commits during DDI
hotplug handling, as a preparation for a follow-up patch adding a
similar commit started from the same spot. Other places doing a
driver-internal atomic commit is to be converted by a follow-up
patchset.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510103131.1618266-13-imre.deak@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |  1 +
 drivers/gpu/drm/i915/display/intel_ddi.c      | 17 ++-----
 .../gpu/drm/i915/display/intel_modeset_lock.c | 50 +++++++++++++++++++
 .../gpu/drm/i915/display/intel_modeset_lock.h | 33 ++++++++++++
 4 files changed, 87 insertions(+), 14 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_modeset_lock.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_modeset_lock.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 1f4c9b99c2c96..e3b699e7a755f 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -265,6 +265,7 @@ i915-y += \
 	display/intel_hti.o \
 	display/intel_load_detect.o \
 	display/intel_lpe_audio.o \
+	display/intel_modeset_lock.o \
 	display/intel_modeset_verify.o \
 	display/intel_modeset_setup.o \
 	display/intel_overlay.o \
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index db056bb084d16..b8ffbbbb46426 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -63,6 +63,7 @@
 #include "intel_hti.h"
 #include "intel_lspcon.h"
 #include "intel_mg_phy_regs.h"
+#include "intel_modeset_lock.h"
 #include "intel_pps.h"
 #include "intel_psr.h"
 #include "intel_quirks.h"
@@ -4401,26 +4402,14 @@ intel_ddi_hotplug(struct intel_encoder *encoder,
 
 	state = intel_encoder_hotplug(encoder, connector);
 
-	drm_modeset_acquire_init(&ctx, 0);
-
-	for (;;) {
+	intel_modeset_lock_ctx_retry(&ctx, NULL, 0, ret) {
 		if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA)
 			ret = intel_hdmi_reset_link(encoder, &ctx);
 		else
 			ret = intel_dp_retrain_link(encoder, &ctx);
-
-		if (ret == -EDEADLK) {
-			drm_modeset_backoff(&ctx);
-			continue;
-		}
-
-		break;
 	}
 
-	drm_modeset_drop_locks(&ctx);
-	drm_modeset_acquire_fini(&ctx);
-	drm_WARN(encoder->base.dev, ret,
-		 "Acquiring modeset locks failed with %i\n", ret);
+	drm_WARN_ON(encoder->base.dev, ret);
 
 	/*
 	 * Unpowered type-c dongles can take some time to boot and be
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_lock.c b/drivers/gpu/drm/i915/display/intel_modeset_lock.c
new file mode 100644
index 0000000000000..8fb6fd849a75d
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_modeset_lock.c
@@ -0,0 +1,50 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <drm/drm_modeset_lock.h>
+
+#include "intel_display_types.h"
+#include "intel_modeset_lock.h"
+
+void _intel_modeset_lock_begin(struct drm_modeset_acquire_ctx *ctx,
+			       struct intel_atomic_state *state,
+			       unsigned int flags, int *ret)
+{
+	drm_modeset_acquire_init(ctx, flags);
+
+	if (state)
+		state->base.acquire_ctx = ctx;
+
+	*ret = -EDEADLK;
+}
+
+bool _intel_modeset_lock_loop(int *ret)
+{
+	if (*ret == -EDEADLK) {
+		*ret = 0;
+		return true;
+	}
+
+	return false;
+}
+
+void _intel_modeset_lock_end(struct drm_modeset_acquire_ctx *ctx,
+			     struct intel_atomic_state *state,
+			     int *ret)
+{
+	if (*ret == -EDEADLK) {
+		if (state)
+			drm_atomic_state_clear(&state->base);
+
+		*ret = drm_modeset_backoff(ctx);
+		if (*ret == 0) {
+			*ret = -EDEADLK;
+			return;
+		}
+	}
+
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_lock.h b/drivers/gpu/drm/i915/display/intel_modeset_lock.h
new file mode 100644
index 0000000000000..edb5099bcd99c
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_modeset_lock.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_MODESET_LOCK_H__
+#define __INTEL_MODESET_LOCK_H__
+
+#include <linux/types.h>
+
+struct drm_modeset_acquire_ctx;
+struct intel_atomic_state;
+
+void _intel_modeset_lock_begin(struct drm_modeset_acquire_ctx *ctx,
+			       struct intel_atomic_state *state,
+			       unsigned int flags,
+			       int *ret);
+bool _intel_modeset_lock_loop(int *ret);
+void _intel_modeset_lock_end(struct drm_modeset_acquire_ctx *ctx,
+			     struct intel_atomic_state *state,
+			     int *ret);
+
+/*
+ * Note that one must always use "continue" rather than
+ * "break" or "return" to handle errors within the
+ * intel_modeset_lock_ctx_retry() block.
+ */
+#define intel_modeset_lock_ctx_retry(ctx, state, flags, ret) \
+	for (_intel_modeset_lock_begin((ctx), (state), (flags), &(ret)); \
+	     _intel_modeset_lock_loop(&(ret)); \
+	     _intel_modeset_lock_end((ctx), (state), &(ret)))
+
+#endif /* __INTEL_MODESET_LOCK_H__ */
-- 
GitLab


From b61fad5f7e5d859d95a413c3a57f59d007951fa6 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 12 May 2023 22:55:12 +0300
Subject: [PATCH 0526/1791] drm/i915/tc: Call TypeC port flush_work/cleanup
 without modeset locks held
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Call the TypeC port flush_work and cleanup handlers without the modeset
locks held. These don't require the locks, as the work takes - as it
should be able to at any point in time - any locks it needs and by the
time cleanup is called and after cleanup returns the encoder is not in
use.

This is required by the next patch canceling a TypeC port work
synchronously during encoder suspend and shutdown, where the work can
take modeset locks as well, hence the canceling must be done without
holding the locks.

I also considered moving the modeset locking down to each encoder
suspend()/shutdown() hook instead, however locking the full modeset
state for each encoder separately would be odd, and the bigger change -
affecting all encoders - is beyond the scope of this patchset.

v2:
- Add a TODO: comment to remove modeset locks if no encoder depends
  on this. (Ville)

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512195513.2699-1-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c      | 27 +++++++++----------
 .../drm/i915/display/intel_display_types.h    | 12 +++++++++
 drivers/gpu/drm/i915/i915_driver.c            | 16 +++++++++++
 3 files changed, 41 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index b8ffbbbb46426..6566518eeb7f2 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -4615,31 +4615,27 @@ static bool intel_ddi_is_tc(struct drm_i915_private *i915, enum port port)
 
 static void intel_ddi_encoder_suspend(struct intel_encoder *encoder)
 {
-	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	enum phy phy = intel_port_to_phy(i915, encoder->port);
-
 	intel_dp_encoder_suspend(encoder);
+}
 
-	if (!intel_phy_is_tc(i915, phy))
-		return;
+static void intel_ddi_tc_encoder_suspend_complete(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 
 	intel_tc_port_flush_work(dig_port);
 }
 
 static void intel_ddi_encoder_shutdown(struct intel_encoder *encoder)
 {
-	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
-	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
-	enum phy phy = intel_port_to_phy(i915, encoder->port);
-
 	intel_dp_encoder_shutdown(encoder);
 	intel_hdmi_encoder_shutdown(encoder);
+}
 
-	if (!intel_phy_is_tc(i915, phy))
-		return;
+static void intel_ddi_tc_encoder_shutdown_complete(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 
 	intel_tc_port_cleanup(dig_port);
 }
@@ -4906,6 +4902,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 				    is_legacy ? "legacy" : "non-legacy");
 		}
 
+		encoder->suspend_complete = intel_ddi_tc_encoder_suspend_complete;
+		encoder->shutdown_complete = intel_ddi_tc_encoder_shutdown_complete;
+
 		if (intel_tc_port_init(dig_port, is_legacy) < 0)
 			goto err;
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 270c4c84a2920..88b2a55d19f21 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -233,13 +233,25 @@ struct intel_encoder {
 	 * Called during system suspend after all pending requests for the
 	 * encoder are flushed (for example for DP AUX transactions) and
 	 * device interrupts are disabled.
+	 * All modeset locks are held while the hook is called.
 	 */
 	void (*suspend)(struct intel_encoder *);
+	/*
+	 * Called without the modeset locks held after the suspend() hook for
+	 * all encoders have been called.
+	 */
+	void (*suspend_complete)(struct intel_encoder *encoder);
 	/*
 	 * Called during system reboot/shutdown after all the
 	 * encoders have been disabled and suspended.
+	 * All modeset locks are held while the hook is called.
 	 */
 	void (*shutdown)(struct intel_encoder *encoder);
+	/*
+	 * Called without the modeset locks held after the shutdown() hook for
+	 * all encoders have been called.
+	 */
+	void (*shutdown_complete)(struct intel_encoder *encoder);
 	/*
 	 * Enable/disable the clock to the port.
 	 */
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index bf3af32fa962d..31d4847eb21cc 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -957,11 +957,19 @@ static void intel_suspend_encoders(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
+	/*
+	 * TODO: check and remove holding the modeset locks if none of
+	 * the encoders depends on this.
+	 */
 	drm_modeset_lock_all(&dev_priv->drm);
 	for_each_intel_encoder(&dev_priv->drm, encoder)
 		if (encoder->suspend)
 			encoder->suspend(encoder);
 	drm_modeset_unlock_all(&dev_priv->drm);
+
+	for_each_intel_encoder(&dev_priv->drm, encoder)
+		if (encoder->suspend_complete)
+			encoder->suspend_complete(encoder);
 }
 
 static void intel_shutdown_encoders(struct drm_i915_private *dev_priv)
@@ -971,11 +979,19 @@ static void intel_shutdown_encoders(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
+	/*
+	 * TODO: check and remove holding the modeset locks if none of
+	 * the encoders depends on this.
+	 */
 	drm_modeset_lock_all(&dev_priv->drm);
 	for_each_intel_encoder(&dev_priv->drm, encoder)
 		if (encoder->shutdown)
 			encoder->shutdown(encoder);
 	drm_modeset_unlock_all(&dev_priv->drm);
+
+	for_each_intel_encoder(&dev_priv->drm, encoder)
+		if (encoder->shutdown_complete)
+			encoder->shutdown_complete(encoder);
 }
 
 void i915_driver_shutdown(struct drm_i915_private *i915)
-- 
GitLab


From c598c335da420715670b1adac846e4f3ebd01e40 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 12 May 2023 22:55:13 +0300
Subject: [PATCH 0527/1791] drm/i915/tc: Reset TypeC PHYs left enabled in
 DP-alt mode after the sink disconnects
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If the output on a DP-alt link with its sink disconnected is kept
enabled for too long (about 20 sec), then some IOM/TCSS firmware timeout
will cause havoc on the PCI bus, at least for other GFX devices on it
which will stop powering up. Since user space is not guaranteed to do a
disabling modeset in time, switch such disconnected but active links to
TBT mode - which is without such shortcomings - with a 2 second delay.

If the above condition is detected already during the driver load/system
resume sanitization step disable the output instead, as at that point no
user space or kernel client depends on a consistent output state yet and
because subsequent atomic modeset on such connectors - without the
actual sink capabilities available - can fail.

An active/disconnected port as above will also block the HPD status of
other active/disconnected ports to get updated (stuck in the connected
state), until the former port is disabled, its PHY is disconnected and
a ~10 ms delay has elapsed. This means the link state for all TypeC
ports/CRTCs must be rechecked after a CRTC is disabled due to the above
reason. For this disconnect the PHY synchronously after the CRTC/port is
disabled and recheck all CRTCs for the above condition whenever such a
port is disabled.

To account for a race condition during driver loading where the sink is
disconnected after the above sanitization step and before the HPD
interrupts get enabled, do an explicit check/link reset if needed from
the encoder's late_register hook, which is called after the HPD
interrupts are enabled already.

v2:
- Handle an active/disconnected port blocking the HPD state update of
  another active/disconnected port.
- Cancel the delayed work resetting the link also from the encoder
  enable/suspend/shutdown hooks.
- Rebase on the earlier intel_modeset_lock_ctx_retry() addition,
  fixing here the missed atomic state reset in case of a retry.
- Fix handling of an error return from intel_atomic_get_crtc_state().
- Recheck if the port needs to be reset after all the atomic state
  is locked and async commits are waited on.

v3:
- Add intel_crtc_needs_link_reset(), instead of open-coding it,
  keep intel_crtc_has_encoders(). (Ville)
- Fix state dumping and use a bitmask to track disabled CRTCs in
  intel_sanitize_all_crtcs(). (Ville)
- Set internal in intel_atomic_state right after allocating it.
  (Ville)
- Recheck all CRTCs (not yet force-disabled) after a CRTC is
  force-disabled for any reason (not only due to a link state)
  in intel_sanitize_all_crtcs().
- Reduce delay after CRTC disabling to 20ms, and use the simpler
  msleep().
- Clarify code comment about HPD behaviour in
  intel_sanitize_all_crtcs().
- Move all the TC link reset logic to intel_tc.c .
- Cancel the link reset work synchronously during system suspend,
  driver unload and shutdown.

v4:
- Rebased on previous patch, which allows calling the TC port
  suspend/cleanup handlers without modeset locks held; remove the
  display driver suspended assert from the link reset work
  accordingly.

v5: (Ville)
- Remove reset work canceling from intel_ddi_pre_pll_enable().
- Track a crtc vs. pipe mask in intel_sanitize_all_crtcs().
- Add reset_link_commit() to clarify the
  intel_modeset_lock_ctx_retry loop.

Cc: Kai-Heng Feng <kai.heng.feng@canonical.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5860
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512195513.2699-2-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c      |  30 +++-
 drivers/gpu/drm/i915/display/intel_dp.c       |   6 +-
 drivers/gpu/drm/i915/display/intel_dp.h       |   3 +
 .../drm/i915/display/intel_modeset_setup.c    |  87 ++++++++--
 drivers/gpu/drm/i915/display/intel_tc.c       | 159 +++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_tc.h       |   5 +-
 6 files changed, 265 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 6566518eeb7f2..9c2b7f7deb635 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -3311,6 +3311,8 @@ static void intel_disable_ddi(struct intel_atomic_state *state,
 			      const struct intel_crtc_state *old_crtc_state,
 			      const struct drm_connector_state *old_conn_state)
 {
+	intel_tc_port_link_cancel_reset_work(enc_to_dig_port(encoder));
+
 	intel_hdcp_disable(to_intel_connector(old_conn_state->connector));
 
 	if (intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_HDMI))
@@ -4230,9 +4232,19 @@ static void intel_ddi_encoder_reset(struct drm_encoder *encoder)
 		intel_tc_port_init_mode(dig_port);
 }
 
+static int intel_ddi_encoder_late_register(struct drm_encoder *_encoder)
+{
+	struct intel_encoder *encoder = to_intel_encoder(_encoder);
+
+	intel_tc_port_link_reset(enc_to_dig_port(encoder));
+
+	return 0;
+}
+
 static const struct drm_encoder_funcs intel_ddi_funcs = {
 	.reset = intel_ddi_encoder_reset,
 	.destroy = intel_ddi_encoder_destroy,
+	.late_register = intel_ddi_encoder_late_register,
 };
 
 static struct intel_connector *
@@ -4402,14 +4414,16 @@ intel_ddi_hotplug(struct intel_encoder *encoder,
 
 	state = intel_encoder_hotplug(encoder, connector);
 
-	intel_modeset_lock_ctx_retry(&ctx, NULL, 0, ret) {
-		if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA)
-			ret = intel_hdmi_reset_link(encoder, &ctx);
-		else
-			ret = intel_dp_retrain_link(encoder, &ctx);
-	}
+	if (!intel_tc_port_link_reset(dig_port)) {
+		intel_modeset_lock_ctx_retry(&ctx, NULL, 0, ret) {
+			if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA)
+				ret = intel_hdmi_reset_link(encoder, &ctx);
+			else
+				ret = intel_dp_retrain_link(encoder, &ctx);
+		}
 
-	drm_WARN_ON(encoder->base.dev, ret);
+		drm_WARN_ON(encoder->base.dev, ret);
+	}
 
 	/*
 	 * Unpowered type-c dongles can take some time to boot and be
@@ -4623,7 +4637,7 @@ static void intel_ddi_tc_encoder_suspend_complete(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 
-	intel_tc_port_flush_work(dig_port);
+	intel_tc_port_suspend(dig_port);
 }
 
 static void intel_ddi_encoder_shutdown(struct intel_encoder *encoder)
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 088d67c448248..4bec8cd7979fd 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -4232,9 +4232,9 @@ static bool intel_dp_has_connector(struct intel_dp *intel_dp,
 	return false;
 }
 
-static int intel_dp_get_active_pipes(struct intel_dp *intel_dp,
-				     struct drm_modeset_acquire_ctx *ctx,
-				     u8 *pipe_mask)
+int intel_dp_get_active_pipes(struct intel_dp *intel_dp,
+			      struct drm_modeset_acquire_ctx *ctx,
+			      u8 *pipe_mask)
 {
 	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	struct drm_connector_list_iter conn_iter;
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index ef39e4f7a329e..5f86157a10d2d 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -42,6 +42,9 @@ void intel_dp_set_link_params(struct intel_dp *intel_dp,
 			      int link_rate, int lane_count);
 int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp,
 					    int link_rate, u8 lane_count);
+int intel_dp_get_active_pipes(struct intel_dp *intel_dp,
+			      struct drm_modeset_acquire_ctx *ctx,
+			      u8 *pipe_mask);
 int intel_dp_retrain_link(struct intel_encoder *encoder,
 			  struct drm_modeset_acquire_ctx *ctx);
 void intel_dp_set_power(struct intel_dp *intel_dp, u8 mode);
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 75b4dea1e442b..5ff99ca7f1de0 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -26,6 +26,7 @@
 #include "intel_fifo_underrun.h"
 #include "intel_modeset_setup.h"
 #include "intel_pch_display.h"
+#include "intel_tc.h"
 #include "intel_vblank.h"
 #include "intel_wm.h"
 #include "skl_watermark.h"
@@ -379,6 +380,21 @@ static bool intel_crtc_has_encoders(struct intel_crtc *crtc)
 	return false;
 }
 
+static bool intel_crtc_needs_link_reset(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct intel_encoder *encoder;
+
+	for_each_encoder_on_crtc(dev, &crtc->base, encoder) {
+		struct intel_digital_port *dig_port = enc_to_dig_port(encoder);
+
+		if (dig_port && intel_tc_port_link_needs_reset(dig_port))
+			return true;
+	}
+
+	return false;
+}
+
 static struct intel_connector *intel_encoder_find_connector(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
@@ -421,11 +437,12 @@ static void intel_sanitize_fifo_underrun_reporting(const struct intel_crtc_state
 					   !HAS_GMCH(i915));
 }
 
-static void intel_sanitize_crtc(struct intel_crtc *crtc,
+static bool intel_sanitize_crtc(struct intel_crtc *crtc,
 				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
 	struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state);
+	bool needs_link_reset;
 
 	if (crtc_state->hw.active) {
 		struct intel_plane *plane;
@@ -445,13 +462,67 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc,
 		intel_color_commit_arm(crtc_state);
 	}
 
+	if (!crtc_state->hw.active ||
+	    intel_crtc_is_bigjoiner_slave(crtc_state))
+		return false;
+
+	needs_link_reset = intel_crtc_needs_link_reset(crtc);
+
 	/*
 	 * Adjust the state of the output pipe according to whether we have
 	 * active connectors/encoders.
 	 */
-	if (crtc_state->hw.active && !intel_crtc_has_encoders(crtc) &&
-	    !intel_crtc_is_bigjoiner_slave(crtc_state))
-		intel_crtc_disable_noatomic(crtc, ctx);
+	if (!needs_link_reset && intel_crtc_has_encoders(crtc))
+		return false;
+
+	intel_crtc_disable_noatomic(crtc, ctx);
+
+	/*
+	 * The HPD state on other active/disconnected TC ports may be stuck in
+	 * the connected state until this port is disabled and a ~10ms delay has
+	 * passed, wait here for that so that sanitizing other CRTCs will see the
+	 * up-to-date HPD state.
+	 */
+	if (needs_link_reset)
+		msleep(20);
+
+	return true;
+}
+
+static void intel_sanitize_all_crtcs(struct drm_i915_private *i915,
+				     struct drm_modeset_acquire_ctx *ctx)
+{
+	struct intel_crtc *crtc;
+	u32 crtcs_forced_off = 0;
+
+	/*
+	 * An active and disconnected TypeC port prevents the HPD live state
+	 * to get updated on other active/disconnected TypeC ports, so after
+	 * a port gets disabled the CRTCs using other TypeC ports must be
+	 * rechecked wrt. their link status.
+	 */
+	for (;;) {
+		u32 old_mask = crtcs_forced_off;
+
+		for_each_intel_crtc(&i915->drm, crtc) {
+			u32 crtc_mask = drm_crtc_mask(&crtc->base);
+
+			if (crtcs_forced_off & crtc_mask)
+				continue;
+
+			if (intel_sanitize_crtc(crtc, ctx))
+				crtcs_forced_off |= crtc_mask;
+		}
+		if (crtcs_forced_off == old_mask)
+			break;
+	}
+
+	for_each_intel_crtc(&i915->drm, crtc) {
+		struct intel_crtc_state *crtc_state =
+			to_intel_crtc_state(crtc->base.state);
+
+		intel_crtc_state_dump(crtc_state, NULL, "setup_hw_state");
+	}
 }
 
 static bool has_bogus_dpll_config(const struct intel_crtc_state *crtc_state)
@@ -871,13 +942,7 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915,
 	 */
 	intel_modeset_update_connector_atomic_state(i915);
 
-	for_each_intel_crtc(&i915->drm, crtc) {
-		struct intel_crtc_state *crtc_state =
-			to_intel_crtc_state(crtc->base.state);
-
-		intel_sanitize_crtc(crtc, ctx);
-		intel_crtc_state_dump(crtc_state, NULL, "setup_hw_state");
-	}
+	intel_sanitize_all_crtcs(i915, ctx);
 
 	intel_dpll_sanitize_state(i915);
 
diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 4fca711a58bce..3ebf41859043e 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -5,15 +5,19 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "intel_atomic.h"
 #include "intel_cx0_phy_regs.h"
 #include "intel_ddi.h"
 #include "intel_de.h"
 #include "intel_display.h"
+#include "intel_display_driver.h"
 #include "intel_display_power_map.h"
 #include "intel_display_types.h"
 #include "intel_dkl_phy_regs.h"
+#include "intel_dp.h"
 #include "intel_dp_mst.h"
 #include "intel_mg_phy_regs.h"
+#include "intel_modeset_lock.h"
 #include "intel_tc.h"
 
 #define DP_PIN_ASSIGNMENT_C	0x3
@@ -51,6 +55,7 @@ struct intel_tc_port {
 	enum intel_display_power_domain lock_power_domain;
 #endif
 	struct delayed_work disconnect_phy_work;
+	struct delayed_work link_reset_work;
 	int link_refcount;
 	bool legacy_port:1;
 	char port_name[8];
@@ -1572,6 +1577,138 @@ bool intel_tc_port_connected(struct intel_encoder *encoder)
 	return is_connected;
 }
 
+static bool __intel_tc_port_link_needs_reset(struct intel_tc_port *tc)
+{
+	bool ret;
+
+	mutex_lock(&tc->lock);
+
+	ret = tc->link_refcount &&
+	      tc->mode == TC_PORT_DP_ALT &&
+	      intel_tc_port_needs_reset(tc);
+
+	mutex_unlock(&tc->lock);
+
+	return ret;
+}
+
+bool intel_tc_port_link_needs_reset(struct intel_digital_port *dig_port)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	enum phy phy = intel_port_to_phy(i915, dig_port->base.port);
+
+	if (!intel_phy_is_tc(i915, phy))
+		return false;
+
+	return __intel_tc_port_link_needs_reset(to_tc_port(dig_port));
+}
+
+static int reset_link_commit(struct intel_tc_port *tc,
+			     struct intel_atomic_state *state,
+			     struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct intel_digital_port *dig_port = tc->dig_port;
+	struct intel_dp *intel_dp = enc_to_intel_dp(&dig_port->base);
+	struct intel_crtc *crtc;
+	u8 pipe_mask;
+	int ret;
+
+	ret = drm_modeset_lock(&i915->drm.mode_config.connection_mutex, ctx);
+	if (ret)
+		return ret;
+
+	ret = intel_dp_get_active_pipes(intel_dp, ctx, &pipe_mask);
+	if (ret)
+		return ret;
+
+	if (!pipe_mask)
+		return 0;
+
+	for_each_intel_crtc_in_pipe_mask(&i915->drm, crtc, pipe_mask) {
+		struct intel_crtc_state *crtc_state;
+
+		crtc_state = intel_atomic_get_crtc_state(&state->base, crtc);
+		if (IS_ERR(crtc_state))
+			return PTR_ERR(crtc_state);
+
+		crtc_state->uapi.connectors_changed = true;
+	}
+
+	if (!__intel_tc_port_link_needs_reset(tc))
+		return 0;
+
+	return drm_atomic_commit(&state->base);
+}
+
+static int reset_link(struct intel_tc_port *tc)
+{
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	struct drm_modeset_acquire_ctx ctx;
+	struct drm_atomic_state *_state;
+	struct intel_atomic_state *state;
+	int ret;
+
+	_state = drm_atomic_state_alloc(&i915->drm);
+	if (!_state)
+		return -ENOMEM;
+
+	state = to_intel_atomic_state(_state);
+	state->internal = true;
+
+	intel_modeset_lock_ctx_retry(&ctx, state, 0, ret)
+		ret = reset_link_commit(tc, state, &ctx);
+
+	drm_atomic_state_put(&state->base);
+
+	return ret;
+}
+
+static void intel_tc_port_link_reset_work(struct work_struct *work)
+{
+	struct intel_tc_port *tc =
+		container_of(work, struct intel_tc_port, link_reset_work.work);
+	struct drm_i915_private *i915 = tc_to_i915(tc);
+	int ret;
+
+	if (!__intel_tc_port_link_needs_reset(tc))
+		return;
+
+	mutex_lock(&i915->drm.mode_config.mutex);
+
+	drm_dbg_kms(&i915->drm,
+		    "Port %s: TypeC DP-alt sink disconnected, resetting link\n",
+		    tc->port_name);
+	ret = reset_link(tc);
+	drm_WARN_ON(&i915->drm, ret);
+
+	mutex_unlock(&i915->drm.mode_config.mutex);
+}
+
+bool intel_tc_port_link_reset(struct intel_digital_port *dig_port)
+{
+	if (!intel_tc_port_link_needs_reset(dig_port))
+		return false;
+
+	queue_delayed_work(system_unbound_wq,
+			   &to_tc_port(dig_port)->link_reset_work,
+			   msecs_to_jiffies(2000));
+
+	return true;
+}
+
+void intel_tc_port_link_cancel_reset_work(struct intel_digital_port *dig_port)
+{
+	struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev);
+	enum phy phy = intel_port_to_phy(i915, dig_port->base.port);
+	struct intel_tc_port *tc = to_tc_port(dig_port);
+
+	if (!intel_phy_is_tc(i915, phy))
+		return;
+
+	cancel_delayed_work(&tc->link_reset_work);
+}
+
 static void __intel_tc_port_lock(struct intel_tc_port *tc,
 				 int required_lanes)
 {
@@ -1619,11 +1756,19 @@ static void intel_tc_port_disconnect_phy_work(struct work_struct *work)
  *
  * Flush the delayed work disconnecting an idle PHY.
  */
-void intel_tc_port_flush_work(struct intel_digital_port *dig_port)
+static void intel_tc_port_flush_work(struct intel_digital_port *dig_port)
 {
 	flush_delayed_work(&to_tc_port(dig_port)->disconnect_phy_work);
 }
 
+void intel_tc_port_suspend(struct intel_digital_port *dig_port)
+{
+	struct intel_tc_port *tc = to_tc_port(dig_port);
+
+	cancel_delayed_work_sync(&tc->link_reset_work);
+	intel_tc_port_flush_work(dig_port);
+}
+
 void intel_tc_port_unlock(struct intel_digital_port *dig_port)
 {
 	struct intel_tc_port *tc = to_tc_port(dig_port);
@@ -1660,6 +1805,14 @@ void intel_tc_port_put_link(struct intel_digital_port *dig_port)
 	intel_tc_port_lock(dig_port);
 	__intel_tc_port_put_link(tc);
 	intel_tc_port_unlock(dig_port);
+
+	/*
+	 * The firmware will not update the HPD status of other TypeC ports
+	 * that are active in DP-alt mode with their sink disconnected, until
+	 * this port is disabled and its PHY gets disconnected. Make sure this
+	 * happens in a timely manner by disconnecting the PHY synchronously.
+	 */
+	intel_tc_port_flush_work(dig_port);
 }
 
 int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
@@ -1692,7 +1845,9 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
 		 "%c/TC#%d", port_name(port), tc_port + 1);
 
 	mutex_init(&tc->lock);
+	/* TODO: Combine the two works */
 	INIT_DELAYED_WORK(&tc->disconnect_phy_work, intel_tc_port_disconnect_phy_work);
+	INIT_DELAYED_WORK(&tc->link_reset_work, intel_tc_port_link_reset_work);
 	tc->legacy_port = is_legacy;
 	tc->mode = TC_PORT_DISCONNECTED;
 	tc->link_refcount = 0;
@@ -1706,7 +1861,7 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy)
 
 void intel_tc_port_cleanup(struct intel_digital_port *dig_port)
 {
-	intel_tc_port_flush_work(dig_port);
+	intel_tc_port_suspend(dig_port);
 
 	kfree(dig_port->tc);
 	dig_port->tc = NULL;
diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h
index dd0810f9ea95e..3b16491925fa9 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.h
+++ b/drivers/gpu/drm/i915/display/intel_tc.h
@@ -30,11 +30,14 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
 				 const struct intel_crtc_state *crtc_state);
 void intel_tc_port_lock(struct intel_digital_port *dig_port);
 void intel_tc_port_unlock(struct intel_digital_port *dig_port);
-void intel_tc_port_flush_work(struct intel_digital_port *dig_port);
+void intel_tc_port_suspend(struct intel_digital_port *dig_port);
 void intel_tc_port_get_link(struct intel_digital_port *dig_port,
 			    int required_lanes);
 void intel_tc_port_put_link(struct intel_digital_port *dig_port);
 bool intel_tc_port_ref_held(struct intel_digital_port *dig_port);
+bool intel_tc_port_link_needs_reset(struct intel_digital_port *dig_port);
+bool intel_tc_port_link_reset(struct intel_digital_port *dig_port);
+void intel_tc_port_link_cancel_reset_work(struct intel_digital_port *dig_port);
 
 int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy);
 void intel_tc_port_cleanup(struct intel_digital_port *dig_port);
-- 
GitLab


From 519ce291168af247c7c0fd122c754b74bcf08117 Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Tue, 9 May 2023 10:51:53 +0800
Subject: [PATCH 0528/1791] drm/panel: Modify innolux hj110iz panel initial
 code

There is a problem of screen shake on the old panel. So increase the
panel GOP component pull-down circuit size in hardware, and update the
initialization code at the same time. The new initialization code mainly
modifles the following.

a)adjusted for GOP timing. When Display sleep in, raise all GOP signals
to VGHO and then drop to GND.
b)Increased the Vertical back Porch and Vertical pulse width, so need to
update vsync_end and vtotal and CLK in drm_display_mode.

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
[dianders: fixed spelling mistake in subject]
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509025153.1321446-1-yangcong5@huaqin.corp-partner.google.com
---
 .../gpu/drm/panel/panel-boe-tv101wum-nl6.c    | 397 +++++++++++-------
 1 file changed, 235 insertions(+), 162 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index 64e462f49c603..f5a6046f1d19b 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -452,11 +452,14 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 	_INIT_DCS_CMD(0xFF, 0x20),
 	_INIT_DCS_CMD(0xFB, 0x01),
 	_INIT_DCS_CMD(0x05, 0xD1),
-	_INIT_DCS_CMD(0x0D, 0x63),
-	_INIT_DCS_CMD(0x07, 0x8C),
+	_INIT_DCS_CMD(0x06, 0xC0),
+	_INIT_DCS_CMD(0x07, 0x87),
 	_INIT_DCS_CMD(0x08, 0x4B),
+
+	_INIT_DCS_CMD(0x0D, 0x63),
 	_INIT_DCS_CMD(0x0E, 0x91),
 	_INIT_DCS_CMD(0x0F, 0x69),
+	_INIT_DCS_CMD(0x94, 0x00),
 	_INIT_DCS_CMD(0x95, 0xF5),
 	_INIT_DCS_CMD(0x96, 0xF5),
 	_INIT_DCS_CMD(0x9D, 0x00),
@@ -464,98 +467,96 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 	_INIT_DCS_CMD(0x69, 0x98),
 	_INIT_DCS_CMD(0x75, 0xA2),
 	_INIT_DCS_CMD(0x77, 0xB3),
+
+	_INIT_DCS_CMD(0x58, 0x43),
 	_INIT_DCS_CMD(0xFF, 0x24),
 	_INIT_DCS_CMD(0xFB, 0x01),
 	_INIT_DCS_CMD(0x91, 0x44),
-	_INIT_DCS_CMD(0x92, 0x7A),
-	_INIT_DCS_CMD(0x93, 0x1A),
-	_INIT_DCS_CMD(0x94, 0x40),
-	_INIT_DCS_CMD(0x9A, 0x08),
+	_INIT_DCS_CMD(0x92, 0x4C),
+	_INIT_DCS_CMD(0x94, 0x86),
 	_INIT_DCS_CMD(0x60, 0x96),
 	_INIT_DCS_CMD(0x61, 0xD0),
 	_INIT_DCS_CMD(0x63, 0x70),
-	_INIT_DCS_CMD(0xC2, 0xCF),
-	_INIT_DCS_CMD(0x9B, 0x0F),
-	_INIT_DCS_CMD(0x9A, 0x08),
+	_INIT_DCS_CMD(0xC2, 0xCA),
+
 	_INIT_DCS_CMD(0x00, 0x03),
 	_INIT_DCS_CMD(0x01, 0x03),
 	_INIT_DCS_CMD(0x02, 0x03),
-	_INIT_DCS_CMD(0x03, 0x03),
-	_INIT_DCS_CMD(0x04, 0x03),
-	_INIT_DCS_CMD(0x05, 0x03),
-	_INIT_DCS_CMD(0x06, 0x22),
-	_INIT_DCS_CMD(0x07, 0x06),
-	_INIT_DCS_CMD(0x08, 0x00),
-	_INIT_DCS_CMD(0x09, 0x1D),
-	_INIT_DCS_CMD(0x0A, 0x1C),
-	_INIT_DCS_CMD(0x0B, 0x13),
-	_INIT_DCS_CMD(0x0C, 0x12),
-	_INIT_DCS_CMD(0x0D, 0x11),
-	_INIT_DCS_CMD(0x0E, 0x10),
-	_INIT_DCS_CMD(0x0F, 0x0F),
-	_INIT_DCS_CMD(0x10, 0x0E),
-	_INIT_DCS_CMD(0x11, 0x0D),
-	_INIT_DCS_CMD(0x12, 0x0C),
+	_INIT_DCS_CMD(0x03, 0x29),
+	_INIT_DCS_CMD(0x04, 0x22),
+	_INIT_DCS_CMD(0x05, 0x22),
+	_INIT_DCS_CMD(0x06, 0x0B),
+	_INIT_DCS_CMD(0x07, 0x1D),
+	_INIT_DCS_CMD(0x08, 0x1C),
+	_INIT_DCS_CMD(0x09, 0x05),
+	_INIT_DCS_CMD(0x0A, 0x08),
+	_INIT_DCS_CMD(0x0B, 0x09),
+	_INIT_DCS_CMD(0x0C, 0x0A),
+	_INIT_DCS_CMD(0x0D, 0x0C),
+	_INIT_DCS_CMD(0x0E, 0x0D),
+	_INIT_DCS_CMD(0x0F, 0x0E),
+	_INIT_DCS_CMD(0x10, 0x0F),
+	_INIT_DCS_CMD(0x11, 0x10),
+	_INIT_DCS_CMD(0x12, 0x11),
 	_INIT_DCS_CMD(0x13, 0x04),
-	_INIT_DCS_CMD(0x14, 0x03),
+	_INIT_DCS_CMD(0x14, 0x00),
 	_INIT_DCS_CMD(0x15, 0x03),
 	_INIT_DCS_CMD(0x16, 0x03),
 	_INIT_DCS_CMD(0x17, 0x03),
 	_INIT_DCS_CMD(0x18, 0x03),
-	_INIT_DCS_CMD(0x19, 0x03),
-	_INIT_DCS_CMD(0x1A, 0x03),
-	_INIT_DCS_CMD(0x1B, 0x03),
-	_INIT_DCS_CMD(0x1C, 0x22),
-	_INIT_DCS_CMD(0x1D, 0x06),
-	_INIT_DCS_CMD(0x1E, 0x00),
-	_INIT_DCS_CMD(0x1F, 0x1D),
-	_INIT_DCS_CMD(0x20, 0x1C),
-	_INIT_DCS_CMD(0x21, 0x13),
-	_INIT_DCS_CMD(0x22, 0x12),
-	_INIT_DCS_CMD(0x23, 0x11),
-	_INIT_DCS_CMD(0x24, 0x10),
-	_INIT_DCS_CMD(0x25, 0x0F),
-	_INIT_DCS_CMD(0x26, 0x0E),
-	_INIT_DCS_CMD(0x27, 0x0D),
-	_INIT_DCS_CMD(0x28, 0x0C),
+	_INIT_DCS_CMD(0x19, 0x29),
+	_INIT_DCS_CMD(0x1A, 0x22),
+	_INIT_DCS_CMD(0x1B, 0x22),
+	_INIT_DCS_CMD(0x1C, 0x0B),
+	_INIT_DCS_CMD(0x1D, 0x1D),
+	_INIT_DCS_CMD(0x1E, 0x1C),
+	_INIT_DCS_CMD(0x1F, 0x05),
+	_INIT_DCS_CMD(0x20, 0x08),
+	_INIT_DCS_CMD(0x21, 0x09),
+	_INIT_DCS_CMD(0x22, 0x0A),
+	_INIT_DCS_CMD(0x23, 0x0C),
+	_INIT_DCS_CMD(0x24, 0x0D),
+	_INIT_DCS_CMD(0x25, 0x0E),
+	_INIT_DCS_CMD(0x26, 0x0F),
+	_INIT_DCS_CMD(0x27, 0x10),
+	_INIT_DCS_CMD(0x28, 0x11),
 	_INIT_DCS_CMD(0x29, 0x04),
-	_INIT_DCS_CMD(0x2A, 0x03),
+	_INIT_DCS_CMD(0x2A, 0x00),
 	_INIT_DCS_CMD(0x2B, 0x03),
 
-	_INIT_DCS_CMD(0x2F, 0x05),
-	_INIT_DCS_CMD(0x30, 0x32),
-	_INIT_DCS_CMD(0x31, 0x43),
-	_INIT_DCS_CMD(0x33, 0x05),
-	_INIT_DCS_CMD(0x34, 0x32),
-	_INIT_DCS_CMD(0x35, 0x43),
-	_INIT_DCS_CMD(0x37, 0x44),
-	_INIT_DCS_CMD(0x38, 0x40),
+	_INIT_DCS_CMD(0x2F, 0x0A),
+	_INIT_DCS_CMD(0x30, 0x35),
+	_INIT_DCS_CMD(0x37, 0xA7),
 	_INIT_DCS_CMD(0x39, 0x00),
-	_INIT_DCS_CMD(0x3A, 0x18),
-	_INIT_DCS_CMD(0x3B, 0x00),
-	_INIT_DCS_CMD(0x3D, 0x93),
-	_INIT_DCS_CMD(0xAB, 0x44),
-	_INIT_DCS_CMD(0xAC, 0x40),
+	_INIT_DCS_CMD(0x3A, 0x46),
+	_INIT_DCS_CMD(0x3B, 0x32),
+	_INIT_DCS_CMD(0x3D, 0x12),
+
+	_INIT_DCS_CMD(0x3F, 0x33),
+	_INIT_DCS_CMD(0x40, 0x31),
+	_INIT_DCS_CMD(0x41, 0x40),
+	_INIT_DCS_CMD(0x42, 0x42),
+	_INIT_DCS_CMD(0x47, 0x77),
+	_INIT_DCS_CMD(0x48, 0x77),
+	_INIT_DCS_CMD(0x4A, 0x45),
+	_INIT_DCS_CMD(0x4B, 0x45),
+	_INIT_DCS_CMD(0x4C, 0x14),
 
 	_INIT_DCS_CMD(0x4D, 0x21),
 	_INIT_DCS_CMD(0x4E, 0x43),
 	_INIT_DCS_CMD(0x4F, 0x65),
-	_INIT_DCS_CMD(0x50, 0x87),
-	_INIT_DCS_CMD(0x51, 0x78),
-	_INIT_DCS_CMD(0x52, 0x56),
-	_INIT_DCS_CMD(0x53, 0x34),
-	_INIT_DCS_CMD(0x54, 0x21),
-	_INIT_DCS_CMD(0x55, 0x83),
-	_INIT_DCS_CMD(0x56, 0x08),
+	_INIT_DCS_CMD(0x55, 0x06),
+	_INIT_DCS_CMD(0x56, 0x06),
 	_INIT_DCS_CMD(0x58, 0x21),
-	_INIT_DCS_CMD(0x59, 0x40),
-	_INIT_DCS_CMD(0x5A, 0x00),
-	_INIT_DCS_CMD(0x5B, 0x2C),
-	_INIT_DCS_CMD(0x5E, 0x00, 0x10),
+	_INIT_DCS_CMD(0x59, 0x70),
+	_INIT_DCS_CMD(0x5A, 0x46),
+	_INIT_DCS_CMD(0x5B, 0x32),
+	_INIT_DCS_CMD(0x5C, 0x88),
+	_INIT_DCS_CMD(0x5E, 0x00, 0x00),
 	_INIT_DCS_CMD(0x5F, 0x00),
 
-	_INIT_DCS_CMD(0x7A, 0x00),
-	_INIT_DCS_CMD(0x7B, 0x00),
+	_INIT_DCS_CMD(0x7A, 0xFF),
+	_INIT_DCS_CMD(0x7B, 0xFF),
 	_INIT_DCS_CMD(0x7C, 0x00),
 	_INIT_DCS_CMD(0x7D, 0x00),
 	_INIT_DCS_CMD(0x7E, 0x20),
@@ -565,152 +566,183 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 	_INIT_DCS_CMD(0x82, 0x08),
 	_INIT_DCS_CMD(0x97, 0x02),
 	_INIT_DCS_CMD(0xC5, 0x10),
+
+	_INIT_DCS_CMD(0xD7, 0x55),
+	_INIT_DCS_CMD(0xD8, 0x55),
+	_INIT_DCS_CMD(0xD9, 0x23),
 	_INIT_DCS_CMD(0xDA, 0x05),
 	_INIT_DCS_CMD(0xDB, 0x01),
-	_INIT_DCS_CMD(0xDC, 0x7A),
+	_INIT_DCS_CMD(0xDC, 0x65),
 	_INIT_DCS_CMD(0xDD, 0x55),
 	_INIT_DCS_CMD(0xDE, 0x27),
 	_INIT_DCS_CMD(0xDF, 0x01),
-	_INIT_DCS_CMD(0xE0, 0x7A),
+	_INIT_DCS_CMD(0xE0, 0x65),
 	_INIT_DCS_CMD(0xE1, 0x01),
-	_INIT_DCS_CMD(0xE2, 0x7A),
+	_INIT_DCS_CMD(0xE2, 0x65),
 	_INIT_DCS_CMD(0xE3, 0x01),
-	_INIT_DCS_CMD(0xE4, 0x7A),
+	_INIT_DCS_CMD(0xE4, 0x65),
 	_INIT_DCS_CMD(0xE5, 0x01),
-	_INIT_DCS_CMD(0xE6, 0x7A),
+	_INIT_DCS_CMD(0xE6, 0x65),
 	_INIT_DCS_CMD(0xE7, 0x00),
 	_INIT_DCS_CMD(0xE8, 0x00),
 	_INIT_DCS_CMD(0xE9, 0x01),
-	_INIT_DCS_CMD(0xEA, 0x7A),
+	_INIT_DCS_CMD(0xEA, 0x65),
 	_INIT_DCS_CMD(0xEB, 0x01),
-	_INIT_DCS_CMD(0xEE, 0x7A),
+	_INIT_DCS_CMD(0xEE, 0x65),
 	_INIT_DCS_CMD(0xEF, 0x01),
-	_INIT_DCS_CMD(0xF0, 0x7A),
-
+	_INIT_DCS_CMD(0xF0, 0x65),
 	_INIT_DCS_CMD(0xB6, 0x05, 0x00, 0x05, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05, 0x05, 0x00, 0x00),
+
 	_INIT_DCS_CMD(0xFF, 0x25),
-	_INIT_DCS_CMD(0xFB, 0x01),
 
+	_INIT_DCS_CMD(0xFB, 0x01),
 	_INIT_DCS_CMD(0x05, 0x00),
-
-	_INIT_DCS_CMD(0x13, 0x02),
-	_INIT_DCS_CMD(0x14, 0xDF),
 	_INIT_DCS_CMD(0xF1, 0x10),
+
 	_INIT_DCS_CMD(0x1E, 0x00),
-	_INIT_DCS_CMD(0x1F, 0x00),
-	_INIT_DCS_CMD(0x20, 0x2C),
+	_INIT_DCS_CMD(0x1F, 0x46),
+	_INIT_DCS_CMD(0x20, 0x32),
+
 	_INIT_DCS_CMD(0x25, 0x00),
-	_INIT_DCS_CMD(0x26, 0x00),
-	_INIT_DCS_CMD(0x27, 0x2C),
+	_INIT_DCS_CMD(0x26, 0x46),
+	_INIT_DCS_CMD(0x27, 0x32),
+
 	_INIT_DCS_CMD(0x3F, 0x80),
 	_INIT_DCS_CMD(0x40, 0x00),
 	_INIT_DCS_CMD(0x43, 0x00),
 
-	_INIT_DCS_CMD(0x44, 0x18),
-	_INIT_DCS_CMD(0x45, 0x00),
+	_INIT_DCS_CMD(0x44, 0x46),
+	_INIT_DCS_CMD(0x45, 0x46),
+
+	_INIT_DCS_CMD(0x48, 0x46),
+	_INIT_DCS_CMD(0x49, 0x32),
 
-	_INIT_DCS_CMD(0x48, 0x00),
-	_INIT_DCS_CMD(0x49, 0x2C),
 	_INIT_DCS_CMD(0x5B, 0x80),
+
 	_INIT_DCS_CMD(0x5C, 0x00),
-	_INIT_DCS_CMD(0x5D, 0x00),
-	_INIT_DCS_CMD(0x5E, 0x00),
-	_INIT_DCS_CMD(0x61, 0x00),
-	_INIT_DCS_CMD(0x62, 0x2C),
-	_INIT_DCS_CMD(0x68, 0x10),
+	_INIT_DCS_CMD(0x5D, 0x46),
+	_INIT_DCS_CMD(0x5E, 0x32),
+
+	_INIT_DCS_CMD(0x5F, 0x46),
+	_INIT_DCS_CMD(0x60, 0x32),
+
+	_INIT_DCS_CMD(0x61, 0x46),
+	_INIT_DCS_CMD(0x62, 0x32),
+	_INIT_DCS_CMD(0x68, 0x0C),
+
+	_INIT_DCS_CMD(0x6C, 0x0D),
+	_INIT_DCS_CMD(0x6E, 0x0D),
+	_INIT_DCS_CMD(0x78, 0x00),
+	_INIT_DCS_CMD(0x79, 0xC5),
+	_INIT_DCS_CMD(0x7A, 0x0C),
+	_INIT_DCS_CMD(0x7B, 0xB0),
+
 	_INIT_DCS_CMD(0xFF, 0x26),
 	_INIT_DCS_CMD(0xFB, 0x01),
 
 	_INIT_DCS_CMD(0x00, 0xA1),
 	_INIT_DCS_CMD(0x02, 0x31),
-	_INIT_DCS_CMD(0x0A, 0xF2),
-	_INIT_DCS_CMD(0x04, 0x28),
+	_INIT_DCS_CMD(0x0A, 0xF4),
+	_INIT_DCS_CMD(0x04, 0x50),
 	_INIT_DCS_CMD(0x06, 0x30),
 	_INIT_DCS_CMD(0x0C, 0x16),
 	_INIT_DCS_CMD(0x0D, 0x0D),
 	_INIT_DCS_CMD(0x0F, 0x00),
 	_INIT_DCS_CMD(0x11, 0x00),
 	_INIT_DCS_CMD(0x12, 0x50),
-	_INIT_DCS_CMD(0x13, 0x56),
-	_INIT_DCS_CMD(0x14, 0x57),
+	_INIT_DCS_CMD(0x13, 0x40),
+	_INIT_DCS_CMD(0x14, 0x58),
 	_INIT_DCS_CMD(0x15, 0x00),
 	_INIT_DCS_CMD(0x16, 0x10),
 	_INIT_DCS_CMD(0x17, 0xA0),
 	_INIT_DCS_CMD(0x18, 0x86),
 	_INIT_DCS_CMD(0x22, 0x00),
 	_INIT_DCS_CMD(0x23, 0x00),
-	_INIT_DCS_CMD(0x19, 0x0D),
-	_INIT_DCS_CMD(0x1A, 0x7F),
-	_INIT_DCS_CMD(0x1B, 0x0C),
-	_INIT_DCS_CMD(0x1C, 0xBF),
-	_INIT_DCS_CMD(0x2A, 0x0D),
-	_INIT_DCS_CMD(0x2B, 0x7F),
-	_INIT_DCS_CMD(0x20, 0x00),
+
+	_INIT_DCS_CMD(0x19, 0x0E),
+	_INIT_DCS_CMD(0x1A, 0x31),
+	_INIT_DCS_CMD(0x1B, 0x0D),
+	_INIT_DCS_CMD(0x1C, 0x29),
+	_INIT_DCS_CMD(0x2A, 0x0E),
+	_INIT_DCS_CMD(0x2B, 0x31),
 
 	_INIT_DCS_CMD(0x1D, 0x00),
-	_INIT_DCS_CMD(0x1E, 0x78),
-	_INIT_DCS_CMD(0x1F, 0x78),
+	_INIT_DCS_CMD(0x1E, 0x62),
+	_INIT_DCS_CMD(0x1F, 0x62),
 
-	_INIT_DCS_CMD(0x2F, 0x03),
-	_INIT_DCS_CMD(0x30, 0x78),
-	_INIT_DCS_CMD(0x33, 0x78),
-	_INIT_DCS_CMD(0x34, 0x66),
-	_INIT_DCS_CMD(0x35, 0x11),
+	_INIT_DCS_CMD(0x2F, 0x06),
+	_INIT_DCS_CMD(0x30, 0x62),
+	_INIT_DCS_CMD(0x31, 0x06),
+	_INIT_DCS_CMD(0x32, 0x7F),
+	_INIT_DCS_CMD(0x33, 0x11),
+	_INIT_DCS_CMD(0x34, 0x89),
+	_INIT_DCS_CMD(0x35, 0x67),
 
-	_INIT_DCS_CMD(0x39, 0x10),
-	_INIT_DCS_CMD(0x3A, 0x78),
+	_INIT_DCS_CMD(0x39, 0x0B),
+	_INIT_DCS_CMD(0x3A, 0x62),
 	_INIT_DCS_CMD(0x3B, 0x06),
 
 	_INIT_DCS_CMD(0xC8, 0x04),
-	_INIT_DCS_CMD(0xC9, 0x84),
+	_INIT_DCS_CMD(0xC9, 0x89),
 	_INIT_DCS_CMD(0xCA, 0x4E),
 	_INIT_DCS_CMD(0xCB, 0x00),
+	_INIT_DCS_CMD(0xA9, 0x3F),
+	_INIT_DCS_CMD(0xAA, 0x3E),
+	_INIT_DCS_CMD(0xAB, 0x3D),
+	_INIT_DCS_CMD(0xAC, 0x3C),
+	_INIT_DCS_CMD(0xAD, 0x3B),
+	_INIT_DCS_CMD(0xAE, 0x3A),
+	_INIT_DCS_CMD(0xAF, 0x39),
+	_INIT_DCS_CMD(0xB0, 0x38),
 
-	_INIT_DCS_CMD(0xA9, 0x50),
-	_INIT_DCS_CMD(0xAA, 0x4F),
-	_INIT_DCS_CMD(0xAB, 0x4D),
-	_INIT_DCS_CMD(0xAC, 0x4A),
-	_INIT_DCS_CMD(0xAD, 0x48),
-	_INIT_DCS_CMD(0xAE, 0x46),
 	_INIT_DCS_CMD(0xFF, 0x27),
 	_INIT_DCS_CMD(0xFB, 0x01),
+
+	_INIT_DCS_CMD(0xD0, 0x11),
+	_INIT_DCS_CMD(0xD1, 0x54),
+	_INIT_DCS_CMD(0xDE, 0x43),
+	_INIT_DCS_CMD(0xDF, 0x02),
+
 	_INIT_DCS_CMD(0xC0, 0x18),
 	_INIT_DCS_CMD(0xC1, 0x00),
 	_INIT_DCS_CMD(0xC2, 0x00),
+	_INIT_DCS_CMD(0x00, 0x00),
+	_INIT_DCS_CMD(0xC3, 0x00),
 	_INIT_DCS_CMD(0x56, 0x06),
+
 	_INIT_DCS_CMD(0x58, 0x80),
-	_INIT_DCS_CMD(0x59, 0x75),
+	_INIT_DCS_CMD(0x59, 0x78),
 	_INIT_DCS_CMD(0x5A, 0x00),
-	_INIT_DCS_CMD(0x5B, 0x02),
+	_INIT_DCS_CMD(0x5B, 0x18),
 	_INIT_DCS_CMD(0x5C, 0x00),
-	_INIT_DCS_CMD(0x5D, 0x00),
+	_INIT_DCS_CMD(0x5D, 0x01),
 	_INIT_DCS_CMD(0x5E, 0x20),
 	_INIT_DCS_CMD(0x5F, 0x10),
 	_INIT_DCS_CMD(0x60, 0x00),
-	_INIT_DCS_CMD(0x61, 0x2E),
+	_INIT_DCS_CMD(0x61, 0x1C),
 	_INIT_DCS_CMD(0x62, 0x00),
 	_INIT_DCS_CMD(0x63, 0x01),
-	_INIT_DCS_CMD(0x64, 0x43),
-	_INIT_DCS_CMD(0x65, 0x2D),
+	_INIT_DCS_CMD(0x64, 0x44),
+	_INIT_DCS_CMD(0x65, 0x1B),
 	_INIT_DCS_CMD(0x66, 0x00),
 	_INIT_DCS_CMD(0x67, 0x01),
-	_INIT_DCS_CMD(0x68, 0x43),
+	_INIT_DCS_CMD(0x68, 0x44),
+
 	_INIT_DCS_CMD(0x98, 0x01),
 	_INIT_DCS_CMD(0xB4, 0x03),
-	_INIT_DCS_CMD(0x9B, 0xBD),
-	_INIT_DCS_CMD(0xA0, 0x90),
-	_INIT_DCS_CMD(0xAB, 0x1B),
-	_INIT_DCS_CMD(0xBC, 0x0C),
+	_INIT_DCS_CMD(0x9B, 0xBE),
+
+	_INIT_DCS_CMD(0xAB, 0x14),
+	_INIT_DCS_CMD(0xBC, 0x08),
 	_INIT_DCS_CMD(0xBD, 0x28),
 
 	_INIT_DCS_CMD(0xFF, 0x2A),
 	_INIT_DCS_CMD(0xFB, 0x01),
-
 	_INIT_DCS_CMD(0x22, 0x2F),
 	_INIT_DCS_CMD(0x23, 0x08),
 
 	_INIT_DCS_CMD(0x24, 0x00),
-	_INIT_DCS_CMD(0x25, 0x65),
+	_INIT_DCS_CMD(0x25, 0x62),
 	_INIT_DCS_CMD(0x26, 0xF8),
 	_INIT_DCS_CMD(0x27, 0x00),
 	_INIT_DCS_CMD(0x28, 0x1A),
@@ -720,18 +752,29 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 	_INIT_DCS_CMD(0x2D, 0x1A),
 
 	_INIT_DCS_CMD(0x64, 0x96),
-	_INIT_DCS_CMD(0x65, 0x00),
+	_INIT_DCS_CMD(0x65, 0x10),
 	_INIT_DCS_CMD(0x66, 0x00),
+	_INIT_DCS_CMD(0x67, 0x96),
+	_INIT_DCS_CMD(0x68, 0x10),
+	_INIT_DCS_CMD(0x69, 0x00),
 	_INIT_DCS_CMD(0x6A, 0x96),
-	_INIT_DCS_CMD(0x6B, 0x00),
+	_INIT_DCS_CMD(0x6B, 0x10),
 	_INIT_DCS_CMD(0x6C, 0x00),
 	_INIT_DCS_CMD(0x70, 0x92),
-	_INIT_DCS_CMD(0x71, 0x00),
+	_INIT_DCS_CMD(0x71, 0x10),
 	_INIT_DCS_CMD(0x72, 0x00),
-	_INIT_DCS_CMD(0xA2, 0x33),
+	_INIT_DCS_CMD(0x79, 0x96),
+	_INIT_DCS_CMD(0x7A, 0x10),
+	_INIT_DCS_CMD(0x88, 0x96),
+	_INIT_DCS_CMD(0x89, 0x10),
+
+	_INIT_DCS_CMD(0xA2, 0x3F),
 	_INIT_DCS_CMD(0xA3, 0x30),
 	_INIT_DCS_CMD(0xA4, 0xC0),
+	_INIT_DCS_CMD(0xA5, 0x03),
+
 	_INIT_DCS_CMD(0xE8, 0x00),
+
 	_INIT_DCS_CMD(0x97, 0x3C),
 	_INIT_DCS_CMD(0x98, 0x02),
 	_INIT_DCS_CMD(0x99, 0x95),
@@ -740,38 +783,68 @@ static const struct panel_init_cmd inx_hj110iz_init_cmd[] = {
 	_INIT_DCS_CMD(0x9C, 0x0B),
 	_INIT_DCS_CMD(0x9D, 0x0A),
 	_INIT_DCS_CMD(0x9E, 0x90),
+
+	_INIT_DCS_CMD(0xFF, 0x25),
+	_INIT_DCS_CMD(0x13, 0x02),
+	_INIT_DCS_CMD(0x14, 0xD7),
+	_INIT_DCS_CMD(0xDB, 0x02),
+	_INIT_DCS_CMD(0xDC, 0xD7),
+	_INIT_DCS_CMD(0x17, 0xCF),
+	_INIT_DCS_CMD(0x19, 0x0F),
+	_INIT_DCS_CMD(0x1B, 0x5B),
+
+	_INIT_DCS_CMD(0xFF, 0x20),
+
+	_INIT_DCS_CMD(0xB0, 0x00, 0x00, 0x00, 0x0C, 0x00, 0x24, 0x00, 0x38, 0x00, 0x4C, 0x00, 0x5E, 0x00, 0x6F, 0x00, 0x7E),
+	_INIT_DCS_CMD(0xB1, 0x00, 0x8C, 0x00, 0xBE, 0x00, 0xE5, 0x01, 0x27, 0x01, 0x58, 0x01, 0xA8, 0x01, 0xE8, 0x01, 0xEA),
+	_INIT_DCS_CMD(0xB2, 0x02, 0x28, 0x02, 0x71, 0x02, 0x9E, 0x02, 0xDA, 0x03, 0x00, 0x03, 0x31, 0x03, 0x40, 0x03, 0x51),
+	_INIT_DCS_CMD(0xB3, 0x03, 0x62, 0x03, 0x75, 0x03, 0x89, 0x03, 0x9C, 0x03, 0xAA, 0x03, 0xB2),
+
+	_INIT_DCS_CMD(0xB4, 0x00, 0x00, 0x00, 0x0D, 0x00, 0x27, 0x00, 0x3D, 0x00, 0x52, 0x00, 0x64, 0x00, 0x75, 0x00, 0x84),
+	_INIT_DCS_CMD(0xB5, 0x00, 0x93, 0x00, 0xC5, 0x00, 0xEC, 0x01, 0x2C, 0x01, 0x5D, 0x01, 0xAC, 0x01, 0xEC, 0x01, 0xEE),
+	_INIT_DCS_CMD(0xB6, 0x02, 0x2B, 0x02, 0x73, 0x02, 0xA0, 0x02, 0xDB, 0x03, 0x01, 0x03, 0x31, 0x03, 0x41, 0x03, 0x51),
+	_INIT_DCS_CMD(0xB7, 0x03, 0x63, 0x03, 0x75, 0x03, 0x89, 0x03, 0x9C, 0x03, 0xAA, 0x03, 0xB2),
+
+	_INIT_DCS_CMD(0xB8, 0x00, 0x00, 0x00, 0x0E, 0x00, 0x2A, 0x00, 0x40, 0x00, 0x56, 0x00, 0x68, 0x00, 0x7A, 0x00, 0x89),
+	_INIT_DCS_CMD(0xB9, 0x00, 0x98, 0x00, 0xC9, 0x00, 0xF1, 0x01, 0x30, 0x01, 0x61, 0x01, 0xB0, 0x01, 0xEF, 0x01, 0xF1),
+	_INIT_DCS_CMD(0xBA, 0x02, 0x2E, 0x02, 0x76, 0x02, 0xA3, 0x02, 0xDD, 0x03, 0x02, 0x03, 0x32, 0x03, 0x42, 0x03, 0x53),
+	_INIT_DCS_CMD(0xBB, 0x03, 0x66, 0x03, 0x75, 0x03, 0x89, 0x03, 0x9C, 0x03, 0xAA, 0x03, 0xB2),
+
+	_INIT_DCS_CMD(0xFF, 0x21),
+	_INIT_DCS_CMD(0xB0, 0x00, 0x00, 0x00, 0x0C, 0x00, 0x24, 0x00, 0x38, 0x00, 0x4C, 0x00, 0x5E, 0x00, 0x6F, 0x00, 0x7E),
+	_INIT_DCS_CMD(0xB1, 0x00, 0x8C, 0x00, 0xBE, 0x00, 0xE5, 0x01, 0x27, 0x01, 0x58, 0x01, 0xA8, 0x01, 0xE8, 0x01, 0xEA),
+	_INIT_DCS_CMD(0xB2, 0x02, 0x28, 0x02, 0x71, 0x02, 0x9E, 0x02, 0xDA, 0x03, 0x00, 0x03, 0x31, 0x03, 0x40, 0x03, 0x51),
+	_INIT_DCS_CMD(0xB3, 0x03, 0x62, 0x03, 0x77, 0x03, 0x90, 0x03, 0xAC, 0x03, 0xCA, 0x03, 0xDA),
+
+	_INIT_DCS_CMD(0xB4, 0x00, 0x00, 0x00, 0x0D, 0x00, 0x27, 0x00, 0x3D, 0x00, 0x52, 0x00, 0x64, 0x00, 0x75, 0x00, 0x84),
+	_INIT_DCS_CMD(0xB5, 0x00, 0x93, 0x00, 0xC5, 0x00, 0xEC, 0x01, 0x2C, 0x01, 0x5D, 0x01, 0xAC, 0x01, 0xEC, 0x01, 0xEE),
+	_INIT_DCS_CMD(0xB6, 0x02, 0x2B, 0x02, 0x73, 0x02, 0xA0, 0x02, 0xDB, 0x03, 0x01, 0x03, 0x31, 0x03, 0x41, 0x03, 0x51),
+	_INIT_DCS_CMD(0xB7, 0x03, 0x63, 0x03, 0x77, 0x03, 0x90, 0x03, 0xAC, 0x03, 0xCA, 0x03, 0xDA),
+
+	_INIT_DCS_CMD(0xB8, 0x00, 0x00, 0x00, 0x0E, 0x00, 0x2A, 0x00, 0x40, 0x00, 0x56, 0x00, 0x68, 0x00, 0x7A, 0x00, 0x89),
+	_INIT_DCS_CMD(0xB9, 0x00, 0x98, 0x00, 0xC9, 0x00, 0xF1, 0x01, 0x30, 0x01, 0x61, 0x01, 0xB0, 0x01, 0xEF, 0x01, 0xF1),
+	_INIT_DCS_CMD(0xBA, 0x02, 0x2E, 0x02, 0x76, 0x02, 0xA3, 0x02, 0xDD, 0x03, 0x02, 0x03, 0x32, 0x03, 0x42, 0x03, 0x53),
+	_INIT_DCS_CMD(0xBB, 0x03, 0x66, 0x03, 0x77, 0x03, 0x90, 0x03, 0xAC, 0x03, 0xCA, 0x03, 0xDA),
+
 	_INIT_DCS_CMD(0xFF, 0xF0),
 	_INIT_DCS_CMD(0xFB, 0x01),
 	_INIT_DCS_CMD(0x3A, 0x08),
-	_INIT_DCS_CMD(0xFF, 0xD0),
-	_INIT_DCS_CMD(0xFB, 0x01),
-	_INIT_DCS_CMD(0x00, 0x33),
-	_INIT_DCS_CMD(0x08, 0x01),
-	_INIT_DCS_CMD(0x09, 0xBF),
-	_INIT_DCS_CMD(0x2F, 0x33),
-	_INIT_DCS_CMD(0xFF, 0x23),
-	_INIT_DCS_CMD(0xFB, 0x01),
-	_INIT_DCS_CMD(0x00, 0x80),
-	_INIT_DCS_CMD(0x07, 0x00),
-	_INIT_DCS_CMD(0xFF, 0x20),
-	_INIT_DCS_CMD(0xFB, 0x01),
-	_INIT_DCS_CMD(0x30, 0x00),
-	_INIT_DCS_CMD(0xFF, 0x24),
-	_INIT_DCS_CMD(0x5C, 0x88),
-	_INIT_DCS_CMD(0x5D, 0x08),
+
 	_INIT_DCS_CMD(0xFF, 0x10),
 	_INIT_DCS_CMD(0xB9, 0x01),
+
 	_INIT_DCS_CMD(0xFF, 0x20),
+
 	_INIT_DCS_CMD(0x18, 0x40),
 	_INIT_DCS_CMD(0xFF, 0x10),
+
 	_INIT_DCS_CMD(0xB9, 0x02),
 	_INIT_DCS_CMD(0xFF, 0x10),
+
 	_INIT_DCS_CMD(0xFB, 0x01),
-	_INIT_DCS_CMD(0xBB, 0x13),
-	_INIT_DCS_CMD(0x3B, 0x03, 0x96, 0x1A, 0x04, 0x04),
+	_INIT_DCS_CMD(0xB0, 0x01),
 	_INIT_DCS_CMD(0x35, 0x00),
-	_INIT_DCS_CMD(0x51, 0x0F, 0xFF),
-	_INIT_DCS_CMD(0x53, 0x24),
+	_INIT_DCS_CMD(0x3B, 0x03, 0xAE, 0x1A, 0x04, 0x04),
 	_INIT_DELAY_CMD(100),
 	_INIT_DCS_CMD(0x11),
 	_INIT_DELAY_CMD(200),
@@ -1435,15 +1508,15 @@ static const struct panel_desc boe_tv110c9m_desc = {
 };
 
 static const struct drm_display_mode inx_hj110iz_default_mode = {
-	.clock = 166594,
+	.clock = 168432,
 	.hdisplay = 1200,
 	.hsync_start = 1200 + 40,
 	.hsync_end = 1200 + 40 + 8,
 	.htotal = 1200 + 40 + 8 + 28,
 	.vdisplay = 2000,
 	.vsync_start = 2000 + 26,
-	.vsync_end = 2000 + 26 + 1,
-	.vtotal = 2000 + 26 + 1 + 149,
+	.vsync_end = 2000 + 26 + 2,
+	.vtotal = 2000 + 26 + 2 + 172,
 	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
 };
 
-- 
GitLab


From 6197cff30df44e4db85fed545fecb7df00ff8cd0 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 18 Apr 2023 11:17:43 -0700
Subject: [PATCH 0529/1791] drm/i915: Dump error capture to kernel log

This is useful for getting debug information out in certain
situations, such as failing kernel selftests and CI runs that don't
log error captures. It is especially useful for things like retrieving
GuC logs as GuC operation can't be tracked by adding printk or ftrace
entries.

v2: Add CONFIG_DRM_I915_DEBUG_GEM wrapper (review feedback by Rodrigo).

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418181744.3251240-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/i915_gpu_error.c | 132 ++++++++++++++++++++++++++
 drivers/gpu/drm/i915/i915_gpu_error.h |  10 ++
 2 files changed, 142 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 5665364999543..ec368e7002354 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -2228,3 +2228,135 @@ void i915_disable_error_state(struct drm_i915_private *i915, int err)
 		i915->gpu_error.first_error = ERR_PTR(err);
 	spin_unlock_irq(&i915->gpu_error.lock);
 }
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+void intel_klog_error_capture(struct intel_gt *gt,
+			      intel_engine_mask_t engine_mask)
+{
+	static int g_count;
+	struct drm_i915_private *i915 = gt->i915;
+	struct i915_gpu_coredump *error;
+	intel_wakeref_t wakeref;
+	size_t buf_size = PAGE_SIZE * 128;
+	size_t pos_err;
+	char *buf, *ptr, *next;
+	int l_count = g_count++;
+	int line = 0;
+
+	/* Can't allocate memory during a reset */
+	if (test_bit(I915_RESET_BACKOFF, &gt->reset.flags)) {
+		drm_err(&gt->i915->drm, "[Capture/%d.%d] Inside GT reset, skipping error capture :(\n",
+			l_count, line++);
+		return;
+	}
+
+	error = READ_ONCE(i915->gpu_error.first_error);
+	if (error) {
+		drm_err(&i915->drm, "[Capture/%d.%d] Clearing existing error capture first...\n",
+			l_count, line++);
+		i915_reset_error_state(i915);
+	}
+
+	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+		error = i915_gpu_coredump(gt, engine_mask, CORE_DUMP_FLAG_NONE);
+
+	if (IS_ERR(error)) {
+		drm_err(&i915->drm, "[Capture/%d.%d] Failed to capture error capture: %ld!\n",
+			l_count, line++, PTR_ERR(error));
+		return;
+	}
+
+	buf = kvmalloc(buf_size, GFP_KERNEL);
+	if (!buf) {
+		drm_err(&i915->drm, "[Capture/%d.%d] Failed to allocate buffer for error capture!\n",
+			l_count, line++);
+		i915_gpu_coredump_put(error);
+		return;
+	}
+
+	drm_info(&i915->drm, "[Capture/%d.%d] Dumping i915 error capture for %ps...\n",
+		 l_count, line++, __builtin_return_address(0));
+
+	/* Largest string length safe to print via dmesg */
+#	define MAX_CHUNK	800
+
+	pos_err = 0;
+	while (1) {
+		ssize_t got = i915_gpu_coredump_copy_to_buffer(error, buf, pos_err, buf_size - 1);
+
+		if (got <= 0)
+			break;
+
+		buf[got] = 0;
+		pos_err += got;
+
+		ptr = buf;
+		while (got > 0) {
+			size_t count;
+			char tag[2];
+
+			next = strnchr(ptr, got, '\n');
+			if (next) {
+				count = next - ptr;
+				*next = 0;
+				tag[0] = '>';
+				tag[1] = '<';
+			} else {
+				count = got;
+				tag[0] = '}';
+				tag[1] = '{';
+			}
+
+			if (count > MAX_CHUNK) {
+				size_t pos;
+				char *ptr2 = ptr;
+
+				for (pos = MAX_CHUNK; pos < count; pos += MAX_CHUNK) {
+					char chr = ptr[pos];
+
+					ptr[pos] = 0;
+					drm_info(&i915->drm, "[Capture/%d.%d] }%s{\n",
+						 l_count, line++, ptr2);
+					ptr[pos] = chr;
+					ptr2 = ptr + pos;
+
+					/*
+					 * If spewing large amounts of data via a serial console,
+					 * this can be a very slow process. So be friendly and try
+					 * not to cause 'softlockup on CPU' problems.
+					 */
+					cond_resched();
+				}
+
+				if (ptr2 < (ptr + count))
+					drm_info(&i915->drm, "[Capture/%d.%d] %c%s%c\n",
+						 l_count, line++, tag[0], ptr2, tag[1]);
+				else if (tag[0] == '>')
+					drm_info(&i915->drm, "[Capture/%d.%d] ><\n",
+						 l_count, line++);
+			} else {
+				drm_info(&i915->drm, "[Capture/%d.%d] %c%s%c\n",
+					 l_count, line++, tag[0], ptr, tag[1]);
+			}
+
+			ptr = next;
+			got -= count;
+			if (next) {
+				ptr++;
+				got--;
+			}
+
+			/* As above. */
+			cond_resched();
+		}
+
+		if (got)
+			drm_info(&i915->drm, "[Capture/%d.%d] Got %zd bytes remaining!\n",
+				 l_count, line++, got);
+	}
+
+	kvfree(buf);
+
+	drm_info(&i915->drm, "[Capture/%d.%d] Dumped %zd bytes\n", l_count, line++, pos_err);
+}
+#endif
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h
index a91932cc65317..a78c061ce26fb 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.h
+++ b/drivers/gpu/drm/i915/i915_gpu_error.h
@@ -258,6 +258,16 @@ static inline u32 i915_reset_engine_count(struct i915_gpu_error *error,
 #define CORE_DUMP_FLAG_NONE           0x0
 #define CORE_DUMP_FLAG_IS_GUC_CAPTURE BIT(0)
 
+#if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) && IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+void intel_klog_error_capture(struct intel_gt *gt,
+			      intel_engine_mask_t engine_mask);
+#else
+static inline void intel_klog_error_capture(struct intel_gt *gt,
+					    intel_engine_mask_t engine_mask)
+{
+}
+#endif
+
 #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)
 
 __printf(2, 3)
-- 
GitLab


From f6eeea8d7097a82d1460537146dee670d5014f13 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 18 Apr 2023 11:17:44 -0700
Subject: [PATCH 0530/1791] drm/i915/guc: Dump error capture to dmesg on CTB
 error

In the past, There have been sporadic CTB failures which proved hard
to reproduce manually. The most effective solution was to dump the GuC
log at the point of failure and let the CI system do the repro. It is
preferable not to dump the GuC log via dmesg for all issues as it is
not always necessary and is not helpful for end users. But rather than
trying to re-invent the code to do this each time it is wanted, commit
the code but for DEBUG_GUC builds only.

v2: Use IS_ENABLED for testing config options.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230418181744.3251240-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 53 +++++++++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h |  6 +++
 2 files changed, 59 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 99a0a89091e7f..a22e33f37cae6 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -13,6 +13,30 @@
 #include "intel_guc_ct.h"
 #include "intel_guc_print.h"
 
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+enum {
+	CT_DEAD_ALIVE = 0,
+	CT_DEAD_SETUP,
+	CT_DEAD_WRITE,
+	CT_DEAD_DEADLOCK,
+	CT_DEAD_H2G_HAS_ROOM,
+	CT_DEAD_READ,
+	CT_DEAD_PROCESS_FAILED,
+};
+
+static void ct_dead_ct_worker_func(struct work_struct *w);
+
+#define CT_DEAD(ct, reason)	\
+	do { \
+		if (!(ct)->dead_ct_reported) { \
+			(ct)->dead_ct_reason |= 1 << CT_DEAD_##reason; \
+			queue_work(system_unbound_wq, &(ct)->dead_ct_worker); \
+		} \
+	} while (0)
+#else
+#define CT_DEAD(ct, reason)	do { } while (0)
+#endif
+
 static inline struct intel_guc *ct_to_guc(struct intel_guc_ct *ct)
 {
 	return container_of(ct, struct intel_guc, ct);
@@ -93,6 +117,9 @@ void intel_guc_ct_init_early(struct intel_guc_ct *ct)
 	spin_lock_init(&ct->requests.lock);
 	INIT_LIST_HEAD(&ct->requests.pending);
 	INIT_LIST_HEAD(&ct->requests.incoming);
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+	INIT_WORK(&ct->dead_ct_worker, ct_dead_ct_worker_func);
+#endif
 	INIT_WORK(&ct->requests.worker, ct_incoming_request_worker_func);
 	tasklet_setup(&ct->receive_tasklet, ct_receive_tasklet_func);
 	init_waitqueue_head(&ct->wq);
@@ -319,11 +346,16 @@ int intel_guc_ct_enable(struct intel_guc_ct *ct)
 
 	ct->enabled = true;
 	ct->stall_time = KTIME_MAX;
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+	ct->dead_ct_reported = false;
+	ct->dead_ct_reason = CT_DEAD_ALIVE;
+#endif
 
 	return 0;
 
 err_out:
 	CT_PROBE_ERROR(ct, "Failed to enable CTB (%pe)\n", ERR_PTR(err));
+	CT_DEAD(ct, SETUP);
 	return err;
 }
 
@@ -434,6 +466,7 @@ static int ct_write(struct intel_guc_ct *ct,
 corrupted:
 	CT_ERROR(ct, "Corrupted descriptor head=%u tail=%u status=%#x\n",
 		 desc->head, desc->tail, desc->status);
+	CT_DEAD(ct, WRITE);
 	ctb->broken = true;
 	return -EPIPE;
 }
@@ -504,6 +537,7 @@ static inline bool ct_deadlocked(struct intel_guc_ct *ct)
 		CT_ERROR(ct, "Head: %u\n (Dwords)", ct->ctbs.recv.desc->head);
 		CT_ERROR(ct, "Tail: %u\n (Dwords)", ct->ctbs.recv.desc->tail);
 
+		CT_DEAD(ct, DEADLOCK);
 		ct->ctbs.send.broken = true;
 	}
 
@@ -552,6 +586,7 @@ static inline bool h2g_has_room(struct intel_guc_ct *ct, u32 len_dw)
 			 head, ctb->size);
 		desc->status |= GUC_CTB_STATUS_OVERFLOW;
 		ctb->broken = true;
+		CT_DEAD(ct, H2G_HAS_ROOM);
 		return false;
 	}
 
@@ -914,6 +949,7 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
 	CT_ERROR(ct, "Corrupted descriptor head=%u tail=%u status=%#x\n",
 		 desc->head, desc->tail, desc->status);
 	ctb->broken = true;
+	CT_DEAD(ct, READ);
 	return -EPIPE;
 }
 
@@ -1063,6 +1099,7 @@ static bool ct_process_incoming_requests(struct intel_guc_ct *ct)
 	if (unlikely(err)) {
 		CT_ERROR(ct, "Failed to process CT message (%pe) %*ph\n",
 			 ERR_PTR(err), 4 * request->size, request->msg);
+		CT_DEAD(ct, PROCESS_FAILED);
 		ct_free_msg(request);
 	}
 
@@ -1239,3 +1276,19 @@ void intel_guc_ct_print_info(struct intel_guc_ct *ct,
 	drm_printf(p, "Tail: %u\n",
 		   ct->ctbs.recv.desc->tail);
 }
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+static void ct_dead_ct_worker_func(struct work_struct *w)
+{
+	struct intel_guc_ct *ct = container_of(w, struct intel_guc_ct, dead_ct_worker);
+	struct intel_guc *guc = ct_to_guc(ct);
+
+	if (ct->dead_ct_reported)
+		return;
+
+	ct->dead_ct_reported = true;
+
+	guc_info(guc, "CTB is dead - reason=0x%X\n", ct->dead_ct_reason);
+	intel_klog_error_capture(guc_to_gt(guc), (intel_engine_mask_t)~0U);
+}
+#endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
index f709a19c7e214..818415b64f4d1 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
@@ -85,6 +85,12 @@ struct intel_guc_ct {
 
 	/** @stall_time: time of first time a CTB submission is stalled */
 	ktime_t stall_time;
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+	int dead_ct_reason;
+	bool dead_ct_reported;
+	struct work_struct dead_ct_worker;
+#endif
 };
 
 void intel_guc_ct_init_early(struct intel_guc_ct *ct);
-- 
GitLab


From 788557fb23702dbd21767a69894a53c1aa58bb60 Mon Sep 17 00:00:00 2001
From: Sui Jingfeng <15330273260@189.cn>
Date: Wed, 17 May 2023 12:16:02 +0800
Subject: [PATCH 0531/1791] drm/drm_atomic_helper.c: fix a typo

Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517041602.3225325-1-suijingfeng@loongson.cn
---
 drivers/gpu/drm/drm_atomic_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index e0ab555aad2cf..41b8066f61fff 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -3154,7 +3154,7 @@ int drm_atomic_helper_update_plane(struct drm_plane *plane,
 EXPORT_SYMBOL(drm_atomic_helper_update_plane);
 
 /**
- * drm_atomic_helper_disable_plane - Helper for primary plane disable using * atomic
+ * drm_atomic_helper_disable_plane - Helper for primary plane disable using atomic
  * @plane: plane to disable
  * @ctx: lock acquire context
  *
-- 
GitLab


From d705a8c73acbe523cb602228d411a73f288bf34c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 16 May 2023 15:29:26 +0300
Subject: [PATCH 0532/1791] drm/i915: fix intel_display_irq.c include order

I meant to sort the includes before submitting commit 2b874a027810
("drm/i915/irq: split out display irq handling") but forgot, and it
wasn't noticed in review either. Sort the includes.

Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230516122926.2720581-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_irq.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index 0eedd1ebb3890..3b2a287d20418 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -3,23 +3,23 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include "gt/intel_rps.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
 #include "icl_dsi_regs.h"
-#include "intel_display_irq.h"
-#include "intel_display_types.h"
-#include "intel_hotplug_irq.h"
-#include "intel_psr_regs.h"
 #include "intel_crtc.h"
+#include "intel_de.h"
+#include "intel_display_irq.h"
 #include "intel_display_trace.h"
+#include "intel_display_types.h"
 #include "intel_dp_aux.h"
-#include "intel_gmbus.h"
+#include "intel_fdi_regs.h"
 #include "intel_fifo_underrun.h"
+#include "intel_gmbus.h"
+#include "intel_hotplug_irq.h"
 #include "intel_psr.h"
-#include "intel_fdi_regs.h"
-#include "gt/intel_rps.h"
-#include "intel_de.h"
+#include "intel_psr_regs.h"
 
 static void
 intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe)
-- 
GitLab


From 1ad797597a80ebe1c62b12403460d71e215f417b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 9 Mar 2023 14:37:00 +0200
Subject: [PATCH 0533/1791] drm/ttm: let struct ttm_device_funcs be placed in
 rodata
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make the struct ttm_device_funcs pointers const so the data can be placed in rodata.

Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Huang Rui <ray.huang@amd.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230309123700.528641-1-jani.nikula@intel.com
---
 drivers/gpu/drm/ttm/ttm_device.c | 2 +-
 include/drm/ttm/ttm_device.h     | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c
index 64a59f46f6c34..df4cf5468e7fd 100644
--- a/drivers/gpu/drm/ttm/ttm_device.c
+++ b/drivers/gpu/drm/ttm/ttm_device.c
@@ -189,7 +189,7 @@ EXPORT_SYMBOL(ttm_device_swapout);
  * Returns:
  * !0: Failure.
  */
-int ttm_device_init(struct ttm_device *bdev, struct ttm_device_funcs *funcs,
+int ttm_device_init(struct ttm_device *bdev, const struct ttm_device_funcs *funcs,
 		    struct device *dev, struct address_space *mapping,
 		    struct drm_vma_offset_manager *vma_manager,
 		    bool use_dma_alloc, bool use_dma32)
diff --git a/include/drm/ttm/ttm_device.h b/include/drm/ttm/ttm_device.h
index 56e82ba2d0462..c22f30535c848 100644
--- a/include/drm/ttm/ttm_device.h
+++ b/include/drm/ttm/ttm_device.h
@@ -223,7 +223,7 @@ struct ttm_device {
 	 * @funcs: Function table for the device.
 	 * Constant after bo device init
 	 */
-	struct ttm_device_funcs *funcs;
+	const struct ttm_device_funcs *funcs;
 
 	/**
 	 * @sysman: Resource manager for the system domain.
@@ -287,7 +287,7 @@ static inline void ttm_set_driver_manager(struct ttm_device *bdev, int type,
 	bdev->man_drv[type] = manager;
 }
 
-int ttm_device_init(struct ttm_device *bdev, struct ttm_device_funcs *funcs,
+int ttm_device_init(struct ttm_device *bdev, const struct ttm_device_funcs *funcs,
 		    struct device *dev, struct address_space *mapping,
 		    struct drm_vma_offset_manager *vma_manager,
 		    bool use_dma_alloc, bool use_dma32);
-- 
GitLab


From 86634ba3dfc789497805da3d5774fcc1953cbe41 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 16 May 2023 12:44:06 +0300
Subject: [PATCH 0534/1791] drm/i915: tweak language in fastset pipe config
 compare logging
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The "fastset mismatch" debug logging has been slightly confusing,
leading people to believe some error happened. Change it to the more
informative "fastset requirement not met", and add a final message about
this leading to full modeset.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230516094406.2675307-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index b3bac5dae20fa..09320e14d75cd 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -4865,7 +4865,7 @@ pipe_config_infoframe_mismatch(struct drm_i915_private *dev_priv,
 			return;
 
 		drm_dbg_kms(&dev_priv->drm,
-			    "fastset mismatch in %s infoframe\n", name);
+			    "fastset requirement not met in %s infoframe\n", name);
 		drm_dbg_kms(&dev_priv->drm, "expected:\n");
 		hdmi_infoframe_log(KERN_DEBUG, dev_priv->drm.dev, a);
 		drm_dbg_kms(&dev_priv->drm, "found:\n");
@@ -4890,7 +4890,7 @@ pipe_config_dp_vsc_sdp_mismatch(struct drm_i915_private *dev_priv,
 			return;
 
 		drm_dbg_kms(&dev_priv->drm,
-			    "fastset mismatch in %s dp sdp\n", name);
+			    "fastset requirement not met in %s dp sdp\n", name);
 		drm_dbg_kms(&dev_priv->drm, "expected:\n");
 		drm_dp_vsc_sdp_log(KERN_DEBUG, dev_priv->drm.dev, a);
 		drm_dbg_kms(&dev_priv->drm, "found:\n");
@@ -4931,7 +4931,7 @@ pipe_config_buffer_mismatch(struct drm_i915_private *dev_priv,
 		len = memcmp_diff_len(a, b, len);
 
 		drm_dbg_kms(&dev_priv->drm,
-			    "fastset mismatch in %s buffer\n", name);
+			    "fastset requirement not met in %s buffer\n", name);
 		print_hex_dump(KERN_DEBUG, "expected: ", DUMP_PREFIX_NONE,
 			       16, 0, a, len, false);
 		print_hex_dump(KERN_DEBUG, "found: ", DUMP_PREFIX_NONE,
@@ -4962,7 +4962,7 @@ pipe_config_mismatch(bool fastset, const struct intel_crtc *crtc,
 
 	if (fastset)
 		drm_dbg_kms(&i915->drm,
-			    "[CRTC:%d:%s] fastset mismatch in %s %pV\n",
+			    "[CRTC:%d:%s] fastset requirement not met in %s %pV\n",
 			    crtc->base.base.id, crtc->base.name, name, &vaf);
 	else
 		drm_err(&i915->drm, "[CRTC:%d:%s] mismatch in %s %pV\n",
@@ -5550,8 +5550,13 @@ static int intel_modeset_checks(struct intel_atomic_state *state)
 static void intel_crtc_check_fastset(const struct intel_crtc_state *old_crtc_state,
 				     struct intel_crtc_state *new_crtc_state)
 {
-	if (!intel_pipe_config_compare(old_crtc_state, new_crtc_state, true))
+	struct drm_i915_private *i915 = to_i915(old_crtc_state->uapi.crtc->dev);
+
+	if (!intel_pipe_config_compare(old_crtc_state, new_crtc_state, true)) {
+		drm_dbg_kms(&i915->drm, "fastset requirement not met, forcing full modeset\n");
+
 		return;
+	}
 
 	new_crtc_state->uapi.mode_changed = false;
 	if (!intel_crtc_needs_modeset(new_crtc_state))
-- 
GitLab


From f47d6140b7a4c858d82d263e7577ff6fb5279a9c Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Wed, 17 May 2023 14:21:06 +0200
Subject: [PATCH 0535/1791] drm/bridge: tc358767: Switch to devm MIPI-DSI
 helpers

DSI device registering and attaching needs to be undone upon
deregistration. This fixes module unload/load.

Fixes: bbfd3190b656 ("drm/bridge: tc358767: Add DSI-to-DPI mode support")
Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517122107.1766673-1-alexander.stein@ew.tq-group.com
---
 drivers/gpu/drm/bridge/tc358767.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 91f7cb56a654d..d6349af4f1b62 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -1890,7 +1890,7 @@ static int tc_mipi_dsi_host_attach(struct tc_data *tc)
 	if (dsi_lanes < 0)
 		return dsi_lanes;
 
-	dsi = mipi_dsi_device_register_full(host, &info);
+	dsi = devm_mipi_dsi_device_register_full(dev, host, &info);
 	if (IS_ERR(dsi))
 		return dev_err_probe(dev, PTR_ERR(dsi),
 				     "failed to create dsi device\n");
@@ -1901,7 +1901,7 @@ static int tc_mipi_dsi_host_attach(struct tc_data *tc)
 	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
 			  MIPI_DSI_MODE_LPM | MIPI_DSI_CLOCK_NON_CONTINUOUS;
 
-	ret = mipi_dsi_attach(dsi);
+	ret = devm_mipi_dsi_attach(dev, dsi);
 	if (ret < 0) {
 		dev_err(dev, "failed to attach dsi to host: %d\n", ret);
 		return ret;
-- 
GitLab


From 6f22587c915c34a4ee02c314cfdb708b11b5eafb Mon Sep 17 00:00:00 2001
From: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Date: Tue, 16 May 2023 08:49:05 -0700
Subject: [PATCH 0536/1791] drm/i915/guc/slpc: Disable rps_boost debugfs

rps_boost debugfs shows host turbo related info. This is not valid
when SLPC is enabled. guc_slpc_info already shows the number of boosts.
Add num_waiters there as well and disable rps_boost when SLPC is
enabled.

v2: Replace Bug with Link to resolve checkpatch warning

Link: https://gitlab.freedesktop.org/drm/intel/-/issues/7632
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230516154905.1048006-1-vinay.belgaumkar@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c | 5 ++++-
 drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c   | 2 ++
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c b/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c
index 80dbbef86b1db..357e2f865727c 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c
@@ -539,7 +539,10 @@ static bool rps_eval(void *data)
 {
 	struct intel_gt *gt = data;
 
-	return HAS_RPS(gt->i915);
+	if (intel_guc_slpc_is_used(&gt->uc.guc))
+		return false;
+	else
+		return HAS_RPS(gt->i915);
 }
 
 DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(rps_boost);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
index 56dbba1ef6684..01b75529311c8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
@@ -837,6 +837,8 @@ int intel_guc_slpc_print_info(struct intel_guc_slpc *slpc, struct drm_printer *p
 				   slpc_decode_min_freq(slpc));
 			drm_printf(p, "\twaitboosts: %u\n",
 				   slpc->num_boosts);
+			drm_printf(p, "\tBoosts outstanding: %u\n",
+				   atomic_read(&slpc->num_waiters));
 		}
 	}
 
-- 
GitLab


From bb47f218fd015b8530364271463f6cbc4d15b814 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:38 +0200
Subject: [PATCH 0537/1791] fbdev/hitfb: Cast I/O offset to address

Cast I/O offsets to pointers to use them with I/O functions. The I/O
functions expect pointers of type 'volatile void __iomem *', but the
offsets are plain integers. Build warnings are

  ../drivers/video/fbdev/hitfb.c: In function 'hitfb_accel_wait':
  ../arch/x86/include/asm/hd64461.h:18:33: warning: passing argument 1 of 'fb_readw' makes pointer from integer without a cast [-Wint-conversion]
   18 | #define HD64461_IO_OFFSET(x)    (HD64461_IOBASE + (x))
      |                                 ^~~~~~~~~~~~~~~~~~~~~~
      |                                 |
      |                                 unsigned int
  ../arch/x86/include/asm/hd64461.h:93:33: note: in expansion of macro 'HD64461_IO_OFFSET'
   93 | #define HD64461_GRCFGR          HD64461_IO_OFFSET(0x1044)       /* Accelerator Configuration Register */
      |                                 ^~~~~~~~~~~~~~~~~
  ../drivers/video/fbdev/hitfb.c:47:25: note: in expansion of macro 'HD64461_GRCFGR'
   47 |         while (fb_readw(HD64461_GRCFGR) & HD64461_GRCFGR_ACCSTATUS) ;
      |                         ^~~~~~~~~~~~~~
  In file included from ../arch/x86/include/asm/fb.h:15,
  from ../include/linux/fb.h:19,
  from ../drivers/video/fbdev/hitfb.c:22:
  ../include/asm-generic/fb.h:52:57: note: expected 'const volatile void *' but argument is of type 'unsigned int'
   52 | static inline u16 fb_readw(const volatile void __iomem *addr)
      |                            ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~

This patch only fixes the build warnings. It's not clear if the I/O
offsets can legally be passed to the I/O helpers. It was apparently
broken in 2007 when custom inw()/outw() helpers got removed by
commit 34a780a0afeb ("sh: hp6xx pata_platform support."). Fixing the
driver would require setting the I/O base address.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202305102136.eMjTSPwH-lkp@intel.com/
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Artur Rojek <contact@artur-rojek.eu>
Acked-by: Helge Deller <deller@gmx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-2-tzimmermann@suse.de
---
 drivers/video/fbdev/hitfb.c | 122 ++++++++++++++++++++----------------
 1 file changed, 69 insertions(+), 53 deletions(-)

diff --git a/drivers/video/fbdev/hitfb.c b/drivers/video/fbdev/hitfb.c
index 3033f50569767..7737923b7a0a0 100644
--- a/drivers/video/fbdev/hitfb.c
+++ b/drivers/video/fbdev/hitfb.c
@@ -42,17 +42,33 @@ static struct fb_fix_screeninfo hitfb_fix = {
 	.accel		= FB_ACCEL_NONE,
 };
 
+static volatile void __iomem *hitfb_offset_to_addr(unsigned int offset)
+{
+	return (__force volatile void __iomem *)(uintptr_t)offset;
+}
+
+static u16 hitfb_readw(unsigned int offset)
+{
+	return fb_readw(hitfb_offset_to_addr(offset));
+}
+
+static void hitfb_writew(u16 value, unsigned int offset)
+{
+	fb_writew(value, hitfb_offset_to_addr(offset));
+}
+
 static inline void hitfb_accel_wait(void)
 {
-	while (fb_readw(HD64461_GRCFGR) & HD64461_GRCFGR_ACCSTATUS) ;
+	while (hitfb_readw(HD64461_GRCFGR) & HD64461_GRCFGR_ACCSTATUS)
+		;
 }
 
 static inline void hitfb_accel_start(int truecolor)
 {
 	if (truecolor) {
-		fb_writew(6, HD64461_GRCFGR);
+		hitfb_writew(6, HD64461_GRCFGR);
 	} else {
-		fb_writew(7, HD64461_GRCFGR);
+		hitfb_writew(7, HD64461_GRCFGR);
 	}
 }
 
@@ -63,11 +79,11 @@ static inline void hitfb_accel_set_dest(int truecolor, u16 dx, u16 dy,
 	if (truecolor)
 		saddr <<= 1;
 
-	fb_writew(width-1, HD64461_BBTDWR);
-	fb_writew(height-1, HD64461_BBTDHR);
+	hitfb_writew(width-1, HD64461_BBTDWR);
+	hitfb_writew(height-1, HD64461_BBTDHR);
 
-	fb_writew(saddr & 0xffff, HD64461_BBTDSARL);
-	fb_writew(saddr >> 16, HD64461_BBTDSARH);
+	hitfb_writew(saddr & 0xffff, HD64461_BBTDSARL);
+	hitfb_writew(saddr >> 16, HD64461_BBTDSARH);
 
 }
 
@@ -80,7 +96,7 @@ static inline void hitfb_accel_bitblt(int truecolor, u16 sx, u16 sy, u16 dx,
 
 	height--;
 	width--;
-	fb_writew(rop, HD64461_BBTROPR);
+	hitfb_writew(rop, HD64461_BBTROPR);
 	if ((sy < dy) || ((sy == dy) && (sx <= dx))) {
 		saddr = WIDTH * (sy + height) + sx + width;
 		daddr = WIDTH * (dy + height) + dx + width;
@@ -91,32 +107,32 @@ static inline void hitfb_accel_bitblt(int truecolor, u16 sx, u16 sy, u16 dx,
 				maddr =
 				    (((width >> 4) + 1) * (height + 1) - 1) * 2;
 
-			fb_writew((1 << 5) | 1, HD64461_BBTMDR);
+			hitfb_writew((1 << 5) | 1, HD64461_BBTMDR);
 		} else
-			fb_writew(1, HD64461_BBTMDR);
+			hitfb_writew(1, HD64461_BBTMDR);
 	} else {
 		saddr = WIDTH * sy + sx;
 		daddr = WIDTH * dy + dx;
 		if (mask_addr) {
-			fb_writew((1 << 5), HD64461_BBTMDR);
+			hitfb_writew((1 << 5), HD64461_BBTMDR);
 		} else {
-			fb_writew(0, HD64461_BBTMDR);
+			hitfb_writew(0, HD64461_BBTMDR);
 		}
 	}
 	if (truecolor) {
 		saddr <<= 1;
 		daddr <<= 1;
 	}
-	fb_writew(width, HD64461_BBTDWR);
-	fb_writew(height, HD64461_BBTDHR);
-	fb_writew(saddr & 0xffff, HD64461_BBTSSARL);
-	fb_writew(saddr >> 16, HD64461_BBTSSARH);
-	fb_writew(daddr & 0xffff, HD64461_BBTDSARL);
-	fb_writew(daddr >> 16, HD64461_BBTDSARH);
+	hitfb_writew(width, HD64461_BBTDWR);
+	hitfb_writew(height, HD64461_BBTDHR);
+	hitfb_writew(saddr & 0xffff, HD64461_BBTSSARL);
+	hitfb_writew(saddr >> 16, HD64461_BBTSSARH);
+	hitfb_writew(daddr & 0xffff, HD64461_BBTDSARL);
+	hitfb_writew(daddr >> 16, HD64461_BBTDSARH);
 	if (mask_addr) {
 		maddr += mask_addr;
-		fb_writew(maddr & 0xffff, HD64461_BBTMARL);
-		fb_writew(maddr >> 16, HD64461_BBTMARH);
+		hitfb_writew(maddr & 0xffff, HD64461_BBTMARL);
+		hitfb_writew(maddr >> 16, HD64461_BBTMARH);
 	}
 	hitfb_accel_start(truecolor);
 }
@@ -127,17 +143,17 @@ static void hitfb_fillrect(struct fb_info *p, const struct fb_fillrect *rect)
 		cfb_fillrect(p, rect);
 	else {
 		hitfb_accel_wait();
-		fb_writew(0x00f0, HD64461_BBTROPR);
-		fb_writew(16, HD64461_BBTMDR);
+		hitfb_writew(0x00f0, HD64461_BBTROPR);
+		hitfb_writew(16, HD64461_BBTMDR);
 
 		if (p->var.bits_per_pixel == 16) {
-			fb_writew(((u32 *) (p->pseudo_palette))[rect->color],
+			hitfb_writew(((u32 *) (p->pseudo_palette))[rect->color],
 				  HD64461_GRSCR);
 			hitfb_accel_set_dest(1, rect->dx, rect->dy, rect->width,
 					     rect->height);
 			hitfb_accel_start(1);
 		} else {
-			fb_writew(rect->color, HD64461_GRSCR);
+			hitfb_writew(rect->color, HD64461_GRSCR);
 			hitfb_accel_set_dest(0, rect->dx, rect->dy, rect->width,
 					     rect->height);
 			hitfb_accel_start(0);
@@ -162,7 +178,7 @@ static int hitfb_pan_display(struct fb_var_screeninfo *var,
 	if (xoffset != 0)
 		return -EINVAL;
 
-	fb_writew((yoffset*info->fix.line_length)>>10, HD64461_LCDCBAR);
+	hitfb_writew((yoffset*info->fix.line_length)>>10, HD64461_LCDCBAR);
 
 	return 0;
 }
@@ -172,33 +188,33 @@ int hitfb_blank(int blank_mode, struct fb_info *info)
 	unsigned short v;
 
 	if (blank_mode) {
-		v = fb_readw(HD64461_LDR1);
+		v = hitfb_readw(HD64461_LDR1);
 		v &= ~HD64461_LDR1_DON;
-		fb_writew(v, HD64461_LDR1);
+		hitfb_writew(v, HD64461_LDR1);
 
-		v = fb_readw(HD64461_LCDCCR);
+		v = hitfb_readw(HD64461_LCDCCR);
 		v |= HD64461_LCDCCR_MOFF;
-		fb_writew(v, HD64461_LCDCCR);
+		hitfb_writew(v, HD64461_LCDCCR);
 
-		v = fb_readw(HD64461_STBCR);
+		v = hitfb_readw(HD64461_STBCR);
 		v |= HD64461_STBCR_SLCDST;
-		fb_writew(v, HD64461_STBCR);
+		hitfb_writew(v, HD64461_STBCR);
 	} else {
-		v = fb_readw(HD64461_STBCR);
+		v = hitfb_readw(HD64461_STBCR);
 		v &= ~HD64461_STBCR_SLCDST;
-		fb_writew(v, HD64461_STBCR);
+		hitfb_writew(v, HD64461_STBCR);
 
-		v = fb_readw(HD64461_LCDCCR);
+		v = hitfb_readw(HD64461_LCDCCR);
 		v &= ~(HD64461_LCDCCR_MOFF | HD64461_LCDCCR_STREQ);
-		fb_writew(v, HD64461_LCDCCR);
+		hitfb_writew(v, HD64461_LCDCCR);
 
 		do {
-		    v = fb_readw(HD64461_LCDCCR);
+		    v = hitfb_readw(HD64461_LCDCCR);
 		} while(v&HD64461_LCDCCR_STBACK);
 
-		v = fb_readw(HD64461_LDR1);
+		v = hitfb_readw(HD64461_LDR1);
 		v |= HD64461_LDR1_DON;
-		fb_writew(v, HD64461_LDR1);
+		hitfb_writew(v, HD64461_LDR1);
 	}
 	return 0;
 }
@@ -211,10 +227,10 @@ static int hitfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 
 	switch (info->var.bits_per_pixel) {
 	case 8:
-		fb_writew(regno << 8, HD64461_CPTWAR);
-		fb_writew(red >> 10, HD64461_CPTWDR);
-		fb_writew(green >> 10, HD64461_CPTWDR);
-		fb_writew(blue >> 10, HD64461_CPTWDR);
+		hitfb_writew(regno << 8, HD64461_CPTWAR);
+		hitfb_writew(red >> 10, HD64461_CPTWDR);
+		hitfb_writew(green >> 10, HD64461_CPTWDR);
+		hitfb_writew(blue >> 10, HD64461_CPTWDR);
 		break;
 	case 16:
 		if (regno >= 16)
@@ -302,11 +318,11 @@ static int hitfb_set_par(struct fb_info *info)
 		break;
 	}
 
-	fb_writew(info->fix.line_length, HD64461_LCDCLOR);
-	ldr3 = fb_readw(HD64461_LDR3);
+	hitfb_writew(info->fix.line_length, HD64461_LCDCLOR);
+	ldr3 = hitfb_readw(HD64461_LDR3);
 	ldr3 &= ~15;
 	ldr3 |= (info->var.bits_per_pixel == 8) ? 4 : 8;
-	fb_writew(ldr3, HD64461_LDR3);
+	hitfb_writew(ldr3, HD64461_LDR3);
 	return 0;
 }
 
@@ -337,9 +353,9 @@ static int hitfb_probe(struct platform_device *dev)
 	hitfb_fix.smem_start = HD64461_IO_OFFSET(0x02000000);
 	hitfb_fix.smem_len = 512 * 1024;
 
-	lcdclor = fb_readw(HD64461_LCDCLOR);
-	ldvndr = fb_readw(HD64461_LDVNDR);
-	ldr3 = fb_readw(HD64461_LDR3);
+	lcdclor = hitfb_readw(HD64461_LCDCLOR);
+	ldvndr = hitfb_readw(HD64461_LDVNDR);
+	ldr3 = hitfb_readw(HD64461_LDR3);
 
 	switch (ldr3 & 15) {
 	default:
@@ -429,9 +445,9 @@ static int hitfb_suspend(struct device *dev)
 	u16 v;
 
 	hitfb_blank(1,0);
-	v = fb_readw(HD64461_STBCR);
+	v = hitfb_readw(HD64461_STBCR);
 	v |= HD64461_STBCR_SLCKE_IST;
-	fb_writew(v, HD64461_STBCR);
+	hitfb_writew(v, HD64461_STBCR);
 
 	return 0;
 }
@@ -440,12 +456,12 @@ static int hitfb_resume(struct device *dev)
 {
 	u16 v;
 
-	v = fb_readw(HD64461_STBCR);
+	v = hitfb_readw(HD64461_STBCR);
 	v &= ~HD64461_STBCR_SLCKE_OST;
 	msleep(100);
-	v = fb_readw(HD64461_STBCR);
+	v = hitfb_readw(HD64461_STBCR);
 	v &= ~HD64461_STBCR_SLCKE_IST;
-	fb_writew(v, HD64461_STBCR);
+	hitfb_writew(v, HD64461_STBCR);
 	hitfb_blank(0,0);
 
 	return 0;
-- 
GitLab


From 2df418ff8be6c5e96bf92d74ce341342dc9080fe Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:39 +0200
Subject: [PATCH 0538/1791] fbdev/matrox: Remove trailing whitespaces

Fix coding style. No functional changes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-3-tzimmermann@suse.de
---
 drivers/video/fbdev/matrox/matroxfb_accel.c | 6 +++---
 drivers/video/fbdev/matrox/matroxfb_base.h  | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/video/fbdev/matrox/matroxfb_accel.c b/drivers/video/fbdev/matrox/matroxfb_accel.c
index 9cb0685feddd7..ce51227798a10 100644
--- a/drivers/video/fbdev/matrox/matroxfb_accel.c
+++ b/drivers/video/fbdev/matrox/matroxfb_accel.c
@@ -88,7 +88,7 @@
 
 static inline void matrox_cfb4_pal(u_int32_t* pal) {
 	unsigned int i;
-	
+
 	for (i = 0; i < 16; i++) {
 		pal[i] = i * 0x11111111U;
 	}
@@ -96,7 +96,7 @@ static inline void matrox_cfb4_pal(u_int32_t* pal) {
 
 static inline void matrox_cfb8_pal(u_int32_t* pal) {
 	unsigned int i;
-	
+
 	for (i = 0; i < 16; i++) {
 		pal[i] = i * 0x01010101U;
 	}
@@ -482,7 +482,7 @@ static void matroxfb_1bpp_imageblit(struct matrox_fb_info *minfo, u_int32_t fgx,
 			/* Tell... well, why bother... */
 			while (height--) {
 				size_t i;
-				
+
 				for (i = 0; i < step; i += 4) {
 					/* Hope that there are at least three readable bytes beyond the end of bitmap */
 					fb_writel(get_unaligned((u_int32_t*)(chardata + i)),mmio.vaddr);
diff --git a/drivers/video/fbdev/matrox/matroxfb_base.h b/drivers/video/fbdev/matrox/matroxfb_base.h
index 958be6805f876..c93c69bbcd579 100644
--- a/drivers/video/fbdev/matrox/matroxfb_base.h
+++ b/drivers/video/fbdev/matrox/matroxfb_base.h
@@ -301,9 +301,9 @@ struct matrox_altout {
 	int		(*verifymode)(void* altout_dev, u_int32_t mode);
 	int		(*getqueryctrl)(void* altout_dev,
 					struct v4l2_queryctrl* ctrl);
-	int		(*getctrl)(void* altout_dev, 
+	int		(*getctrl)(void *altout_dev,
 				   struct v4l2_control* ctrl);
-	int		(*setctrl)(void* altout_dev, 
+	int		(*setctrl)(void *altout_dev,
 				   struct v4l2_control* ctrl);
 };
 
-- 
GitLab


From 30745abe9bb8c67f4ac5486cd810cbb72e3afa16 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:40 +0200
Subject: [PATCH 0539/1791] ipu-v3: Include <linux/io.h>

The code uses readl() and writel(). Include the header file to
get the declarations.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-4-tzimmermann@suse.de
---
 drivers/gpu/ipu-v3/ipu-prv.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/ipu-v3/ipu-prv.h b/drivers/gpu/ipu-v3/ipu-prv.h
index 291ac1bab66d4..d4621b1ea7f12 100644
--- a/drivers/gpu/ipu-v3/ipu-prv.h
+++ b/drivers/gpu/ipu-v3/ipu-prv.h
@@ -8,6 +8,7 @@
 
 struct ipu_soc;
 
+#include <linux/io.h>
 #include <linux/types.h>
 #include <linux/device.h>
 #include <linux/clk.h>
-- 
GitLab


From e1d534ac5b764a573789d4e1d79be6fb7abb1d78 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:41 +0200
Subject: [PATCH 0540/1791] fbdev: Include <linux/io.h> in various drivers

The code uses writel() and similar I/O-memory helpers. Include
the header file to get the declarations.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-5-tzimmermann@suse.de
---
 drivers/video/fbdev/arcfb.c       | 1 +
 drivers/video/fbdev/aty/atyfb.h   | 2 ++
 drivers/video/fbdev/wmt_ge_rops.c | 2 ++
 3 files changed, 5 insertions(+)

diff --git a/drivers/video/fbdev/arcfb.c b/drivers/video/fbdev/arcfb.c
index 7750e020839ec..ace104850e50b 100644
--- a/drivers/video/fbdev/arcfb.c
+++ b/drivers/video/fbdev/arcfb.c
@@ -41,6 +41,7 @@
 #include <linux/vmalloc.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
+#include <linux/io.h>
 #include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/arcfb.h>
diff --git a/drivers/video/fbdev/aty/atyfb.h b/drivers/video/fbdev/aty/atyfb.h
index 465f55beb97f9..30da3e82ed3c2 100644
--- a/drivers/video/fbdev/aty/atyfb.h
+++ b/drivers/video/fbdev/aty/atyfb.h
@@ -3,8 +3,10 @@
  *  ATI Frame Buffer Device Driver Core Definitions
  */
 
+#include <linux/io.h>
 #include <linux/spinlock.h>
 #include <linux/wait.h>
+
     /*
      *  Elements of the hardware specific atyfb_par structure
      */
diff --git a/drivers/video/fbdev/wmt_ge_rops.c b/drivers/video/fbdev/wmt_ge_rops.c
index 3ed143457d228..b70961901683f 100644
--- a/drivers/video/fbdev/wmt_ge_rops.c
+++ b/drivers/video/fbdev/wmt_ge_rops.c
@@ -9,7 +9,9 @@
 
 #include <linux/module.h>
 #include <linux/fb.h>
+#include <linux/io.h>
 #include <linux/platform_device.h>
+
 #include "core/fb_draw.h"
 #include "wmt_ge_rops.h"
 
-- 
GitLab


From 8ff1541da3908b504cb53e5384d5deae2b9c6e1a Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:42 +0200
Subject: [PATCH 0541/1791] fbdev: Include <linux/fb.h> instead of <asm/fb.h>

Replace include statements for <asm/fb.h> with <linux/fb.h>. Fixes
the coding style: if a header is available in asm/ and linux/, it
is preferable to include the header from linux/. This only affects
a few source files, most of which already include <linux/fb.h>.

Suggested-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-6-tzimmermann@suse.de
---
 arch/parisc/video/fbdev.c        | 3 +--
 arch/sparc/video/fbdev.c         | 1 -
 arch/x86/video/fbdev.c           | 2 --
 drivers/staging/sm750fb/sm750.c  | 2 +-
 drivers/video/fbdev/core/fbcon.c | 1 -
 drivers/video/fbdev/core/fbmem.c | 2 --
 include/linux/fb.h               | 2 ++
 7 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/arch/parisc/video/fbdev.c b/arch/parisc/video/fbdev.c
index 4a0ae08fc75b3..137561d982460 100644
--- a/arch/parisc/video/fbdev.c
+++ b/arch/parisc/video/fbdev.c
@@ -5,10 +5,9 @@
  * Copyright (C) 2001-2002 Thomas Bogendoerfer <tsbogend@alpha.franken.de>
  */
 
+#include <linux/fb.h>
 #include <linux/module.h>
 
-#include <asm/fb.h>
-
 #include <video/sticore.h>
 
 int fb_is_primary_device(struct fb_info *info)
diff --git a/arch/sparc/video/fbdev.c b/arch/sparc/video/fbdev.c
index dadd5799fbb32..25837f128132d 100644
--- a/arch/sparc/video/fbdev.c
+++ b/arch/sparc/video/fbdev.c
@@ -4,7 +4,6 @@
 #include <linux/fb.h>
 #include <linux/module.h>
 
-#include <asm/fb.h>
 #include <asm/prom.h>
 
 int fb_is_primary_device(struct fb_info *info)
diff --git a/arch/x86/video/fbdev.c b/arch/x86/video/fbdev.c
index 57ee3c158f976..f41a17ebac48d 100644
--- a/arch/x86/video/fbdev.c
+++ b/arch/x86/video/fbdev.c
@@ -7,8 +7,6 @@
  *
  */
 
-#include <asm/fb.h>
-
 #include <linux/fb.h>
 #include <linux/module.h>
 #include <linux/pci.h>
diff --git a/drivers/staging/sm750fb/sm750.c b/drivers/staging/sm750fb/sm750.c
index 22ace31687233..55e302a27847d 100644
--- a/drivers/staging/sm750fb/sm750.c
+++ b/drivers/staging/sm750fb/sm750.c
@@ -16,7 +16,7 @@
 #include <linux/pagemap.h>
 #include <linux/screen_info.h>
 #include <linux/console.h>
-#include <asm/fb.h>
+
 #include "sm750.h"
 #include "sm750_accel.h"
 #include "sm750_cursor.h"
diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c
index eb565a10e5cda..c6c9d040bdec7 100644
--- a/drivers/video/fbdev/core/fbcon.c
+++ b/drivers/video/fbdev/core/fbcon.c
@@ -75,7 +75,6 @@
 #include <linux/interrupt.h>
 #include <linux/crc32.h> /* For counting font checksums */
 #include <linux/uaccess.h>
-#include <asm/fb.h>
 #include <asm/irq.h>
 
 #include "fbcon.h"
diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c
index 700b9f7e1bb8a..d5f897b5ba54f 100644
--- a/drivers/video/fbdev/core/fbmem.c
+++ b/drivers/video/fbdev/core/fbmem.c
@@ -37,8 +37,6 @@
 #include <linux/mem_encrypt.h>
 #include <linux/pci.h>
 
-#include <asm/fb.h>
-
 #include <video/nomodeset.h>
 #include <video/vga.h>
 
diff --git a/include/linux/fb.h b/include/linux/fb.h
index ec978a4969a97..4b4d9a5d200a7 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -15,6 +15,8 @@
 #include <linux/list.h>
 #include <linux/backlight.h>
 #include <linux/slab.h>
+
+#include <asm/fb.h>
 #include <asm/io.h>
 
 struct vm_area_struct;
-- 
GitLab


From 8f8eaa1b023580f7dce7fe8d73539b093edea65b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:43 +0200
Subject: [PATCH 0542/1791] fbdev: Move framebuffer I/O helpers into <asm/fb.h>

Implement framebuffer I/O helpers, such as fb_read*() and fb_write*(),
in the architecture's <asm/fb.h> header file or the generic one.

The common case has been the use of regular I/O functions, such as
__raw_readb() or memset_io(). A few architectures used plain system-
memory reads and writes. Sparc used helpers for its SBus.

The architectures that used special cases provide the same code in
their __raw_*() I/O helpers. So the patch replaces this code with the
__raw_*() functions and moves it to <asm-generic/fb.h> for all
architectures.

v8:
	* remove garbage after commit-message tags
v6:
	* fix fb_readq()/fb_writeq() on 64-bit mips (kernel test robot)
v5:
	* include <linux/io.h> in <asm-generic/fb>; fix s390 build
v4:
	* ia64, loongarch, sparc64: add fb_mem*() to arch headers
	  to keep current semantics (Arnd)
v3:
	* implement all architectures with generic helpers
	* support reordering and native byte order (Geert, Arnd)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-7-tzimmermann@suse.de
---
 arch/ia64/include/asm/fb.h      |  20 +++++++
 arch/loongarch/include/asm/fb.h |  21 +++++++
 arch/mips/include/asm/fb.h      |  22 +++++++
 arch/sparc/include/asm/fb.h     |  20 +++++++
 include/asm-generic/fb.h        | 102 ++++++++++++++++++++++++++++++++
 include/linux/fb.h              |  53 -----------------
 6 files changed, 185 insertions(+), 53 deletions(-)

diff --git a/arch/ia64/include/asm/fb.h b/arch/ia64/include/asm/fb.h
index 0208f64a0da07..bcf982043a5c2 100644
--- a/arch/ia64/include/asm/fb.h
+++ b/arch/ia64/include/asm/fb.h
@@ -2,7 +2,9 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
+#include <linux/compiler.h>
 #include <linux/efi.h>
+#include <linux/string.h>
 
 #include <asm/page.h>
 
@@ -18,6 +20,24 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 }
 #define fb_pgprotect fb_pgprotect
 
+static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+{
+	memcpy(to, (void __force *)from, n);
+}
+#define fb_memcpy_fromfb fb_memcpy_fromfb
+
+static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+{
+	memcpy((void __force *)to, from, n);
+}
+#define fb_memcpy_tofb fb_memcpy_tofb
+
+static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+{
+	memset((void __force *)addr, c, n);
+}
+#define fb_memset fb_memset
+
 #include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
diff --git a/arch/loongarch/include/asm/fb.h b/arch/loongarch/include/asm/fb.h
index ff82f20685c8b..c6fc7ef374a4a 100644
--- a/arch/loongarch/include/asm/fb.h
+++ b/arch/loongarch/include/asm/fb.h
@@ -5,6 +5,27 @@
 #ifndef _ASM_FB_H_
 #define _ASM_FB_H_
 
+#include <linux/compiler.h>
+#include <linux/string.h>
+
+static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+{
+	memcpy(to, (void __force *)from, n);
+}
+#define fb_memcpy_fromfb fb_memcpy_fromfb
+
+static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+{
+	memcpy((void __force *)to, from, n);
+}
+#define fb_memcpy_tofb fb_memcpy_tofb
+
+static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+{
+	memset((void __force *)addr, c, n);
+}
+#define fb_memset fb_memset
+
 #include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
diff --git a/arch/mips/include/asm/fb.h b/arch/mips/include/asm/fb.h
index 6bda0a81d8ca0..18b7226403bad 100644
--- a/arch/mips/include/asm/fb.h
+++ b/arch/mips/include/asm/fb.h
@@ -12,6 +12,28 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 }
 #define fb_pgprotect fb_pgprotect
 
+/*
+ * MIPS doesn't define __raw_ I/O macros, so the helpers
+ * in <asm-generic/fb.h> don't generate fb_readq() and
+ * fb_write(). We have to provide them here.
+ *
+ * TODO: Convert MIPS to generic I/O. The helpers below can
+ *       then be removed.
+ */
+#ifdef CONFIG_64BIT
+static inline u64 fb_readq(const volatile void __iomem *addr)
+{
+	return __raw_readq(addr);
+}
+#define fb_readq fb_readq
+
+static inline void fb_writeq(u64 b, volatile void __iomem *addr)
+{
+	__raw_writeq(b, addr);
+}
+#define fb_writeq fb_writeq
+#endif
+
 #include <asm-generic/fb.h>
 
 #endif /* _ASM_FB_H_ */
diff --git a/arch/sparc/include/asm/fb.h b/arch/sparc/include/asm/fb.h
index 689ee5c60054c..077da91aeba1a 100644
--- a/arch/sparc/include/asm/fb.h
+++ b/arch/sparc/include/asm/fb.h
@@ -2,6 +2,8 @@
 #ifndef _SPARC_FB_H_
 #define _SPARC_FB_H_
 
+#include <linux/io.h>
+
 struct fb_info;
 struct file;
 struct vm_area_struct;
@@ -16,6 +18,24 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 int fb_is_primary_device(struct fb_info *info);
 #define fb_is_primary_device fb_is_primary_device
 
+static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+{
+	sbus_memcpy_fromio(to, from, n);
+}
+#define fb_memcpy_fromfb fb_memcpy_fromfb
+
+static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+{
+	sbus_memcpy_toio(to, from, n);
+}
+#define fb_memcpy_tofb fb_memcpy_tofb
+
+static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+{
+	sbus_memset_io(addr, c, n);
+}
+#define fb_memset fb_memset
+
 #include <asm-generic/fb.h>
 
 #endif /* _SPARC_FB_H_ */
diff --git a/include/asm-generic/fb.h b/include/asm-generic/fb.h
index c8af99f5a5358..0540eccdbeca1 100644
--- a/include/asm-generic/fb.h
+++ b/include/asm-generic/fb.h
@@ -7,6 +7,7 @@
  * Only include this header file from your architecture's <asm/fb.h>.
  */
 
+#include <linux/io.h>
 #include <linux/mm_types.h>
 #include <linux/pgtable.h>
 
@@ -30,4 +31,105 @@ static inline int fb_is_primary_device(struct fb_info *info)
 }
 #endif
 
+/*
+ * I/O helpers for the framebuffer. Prefer these functions over their
+ * regular counterparts. The regular I/O functions provide in-order
+ * access and swap bytes to/from little-endian ordering. Neither is
+ * required for framebuffers. Instead, the helpers read and write
+ * raw framebuffer data. Independent operations can be reordered for
+ * improved performance.
+ */
+
+#ifndef fb_readb
+static inline u8 fb_readb(const volatile void __iomem *addr)
+{
+	return __raw_readb(addr);
+}
+#define fb_readb fb_readb
+#endif
+
+#ifndef fb_readw
+static inline u16 fb_readw(const volatile void __iomem *addr)
+{
+	return __raw_readw(addr);
+}
+#define fb_readw fb_readw
+#endif
+
+#ifndef fb_readl
+static inline u32 fb_readl(const volatile void __iomem *addr)
+{
+	return __raw_readl(addr);
+}
+#define fb_readl fb_readl
+#endif
+
+#ifndef fb_readq
+#if defined(__raw_readq)
+static inline u64 fb_readq(const volatile void __iomem *addr)
+{
+	return __raw_readq(addr);
+}
+#define fb_readq fb_readq
+#endif
+#endif
+
+#ifndef fb_writeb
+static inline void fb_writeb(u8 b, volatile void __iomem *addr)
+{
+	__raw_writeb(b, addr);
+}
+#define fb_writeb fb_writeb
+#endif
+
+#ifndef fb_writew
+static inline void fb_writew(u16 b, volatile void __iomem *addr)
+{
+	__raw_writew(b, addr);
+}
+#define fb_writew fb_writew
+#endif
+
+#ifndef fb_writel
+static inline void fb_writel(u32 b, volatile void __iomem *addr)
+{
+	__raw_writel(b, addr);
+}
+#define fb_writel fb_writel
+#endif
+
+#ifndef fb_writeq
+#if defined(__raw_writeq)
+static inline void fb_writeq(u64 b, volatile void __iomem *addr)
+{
+	__raw_writeq(b, addr);
+}
+#define fb_writeq fb_writeq
+#endif
+#endif
+
+#ifndef fb_memcpy_fromfb
+static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+{
+	memcpy_fromio(to, from, n);
+}
+#define fb_memcpy_fromfb fb_memcpy_fromfb
+#endif
+
+#ifndef fb_memcpy_tofb
+static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+{
+	memcpy_toio(to, from, n);
+}
+#define fb_memcpy_tofb fb_memcpy_tofb
+#endif
+
+#ifndef fb_memset
+static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+{
+	memset_io(addr, c, n);
+}
+#define fb_memset fb_memset
+#endif
+
 #endif /* __ASM_GENERIC_FB_H_ */
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 4b4d9a5d200a7..2cf8efcb9e327 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -17,7 +17,6 @@
 #include <linux/slab.h>
 
 #include <asm/fb.h>
-#include <asm/io.h>
 
 struct vm_area_struct;
 struct fb_info;
@@ -513,58 +512,6 @@ struct fb_info {
  */
 #define STUPID_ACCELF_TEXT_SHIT
 
-// This will go away
-#if defined(__sparc__)
-
-/* We map all of our framebuffers such that big-endian accesses
- * are what we want, so the following is sufficient.
- */
-
-// This will go away
-#define fb_readb sbus_readb
-#define fb_readw sbus_readw
-#define fb_readl sbus_readl
-#define fb_readq sbus_readq
-#define fb_writeb sbus_writeb
-#define fb_writew sbus_writew
-#define fb_writel sbus_writel
-#define fb_writeq sbus_writeq
-#define fb_memset sbus_memset_io
-#define fb_memcpy_fromfb sbus_memcpy_fromio
-#define fb_memcpy_tofb sbus_memcpy_toio
-
-#elif defined(__i386__) || defined(__alpha__) || defined(__x86_64__) ||	\
-	defined(__hppa__) || defined(__sh__) || defined(__powerpc__) ||	\
-	defined(__arm__) || defined(__aarch64__) || defined(__mips__)
-
-#define fb_readb __raw_readb
-#define fb_readw __raw_readw
-#define fb_readl __raw_readl
-#define fb_readq __raw_readq
-#define fb_writeb __raw_writeb
-#define fb_writew __raw_writew
-#define fb_writel __raw_writel
-#define fb_writeq __raw_writeq
-#define fb_memset memset_io
-#define fb_memcpy_fromfb memcpy_fromio
-#define fb_memcpy_tofb memcpy_toio
-
-#else
-
-#define fb_readb(addr) (*(volatile u8 *) (addr))
-#define fb_readw(addr) (*(volatile u16 *) (addr))
-#define fb_readl(addr) (*(volatile u32 *) (addr))
-#define fb_readq(addr) (*(volatile u64 *) (addr))
-#define fb_writeb(b,addr) (*(volatile u8 *) (addr) = (b))
-#define fb_writew(b,addr) (*(volatile u16 *) (addr) = (b))
-#define fb_writel(b,addr) (*(volatile u32 *) (addr) = (b))
-#define fb_writeq(b,addr) (*(volatile u64 *) (addr) = (b))
-#define fb_memset memset
-#define fb_memcpy_fromfb memcpy
-#define fb_memcpy_tofb memcpy
-
-#endif
-
 #define FB_LEFT_POS(p, bpp)          (fb_be_math(p) ? (32 - (bpp)) : 0)
 #define FB_SHIFT_HIGH(p, val, bits)  (fb_be_math(p) ? (val) >> (bits) : \
 						      (val) << (bits))
-- 
GitLab


From 20d54e48d9c705091a025afff5839da2ea606f6b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 12 May 2023 12:24:44 +0200
Subject: [PATCH 0543/1791] fbdev: Rename fb_mem*() helpers

Update the names of the fb_mem*() helpers to be consistent with their
regular counterparts. Hence, fb_memset() now becomes fb_memset_io(),
fb_memcpy_fromfb() now becomes fb_memcpy_fromio() and fb_memcpy_tofb()
becomes fb_memcpy_toio(). No functional changes.

v6:
	* update new file fb_io_fops.c

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512102444.5438-8-tzimmermann@suse.de
---
 arch/ia64/include/asm/fb.h              | 12 ++++++------
 arch/loongarch/include/asm/fb.h         | 12 ++++++------
 arch/sparc/include/asm/fb.h             | 12 ++++++------
 drivers/video/fbdev/aty/mach64_cursor.c |  2 +-
 drivers/video/fbdev/chipsfb.c           |  2 +-
 drivers/video/fbdev/core/fb_io_fops.c   |  4 ++--
 drivers/video/fbdev/kyro/fbdev.c        |  2 +-
 drivers/video/fbdev/pvr2fb.c            |  2 +-
 drivers/video/fbdev/sstfb.c             |  2 +-
 drivers/video/fbdev/stifb.c             |  4 ++--
 drivers/video/fbdev/tdfxfb.c            |  2 +-
 include/asm-generic/fb.h                | 16 ++++++++--------
 12 files changed, 36 insertions(+), 36 deletions(-)

diff --git a/arch/ia64/include/asm/fb.h b/arch/ia64/include/asm/fb.h
index bcf982043a5c2..1717b26fd423f 100644
--- a/arch/ia64/include/asm/fb.h
+++ b/arch/ia64/include/asm/fb.h
@@ -20,23 +20,23 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 }
 #define fb_pgprotect fb_pgprotect
 
-static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+static inline void fb_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
 	memcpy(to, (void __force *)from, n);
 }
-#define fb_memcpy_fromfb fb_memcpy_fromfb
+#define fb_memcpy_fromio fb_memcpy_fromio
 
-static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+static inline void fb_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
 {
 	memcpy((void __force *)to, from, n);
 }
-#define fb_memcpy_tofb fb_memcpy_tofb
+#define fb_memcpy_toio fb_memcpy_toio
 
-static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+static inline void fb_memset_io(volatile void __iomem *addr, int c, size_t n)
 {
 	memset((void __force *)addr, c, n);
 }
-#define fb_memset fb_memset
+#define fb_memset fb_memset_io
 
 #include <asm-generic/fb.h>
 
diff --git a/arch/loongarch/include/asm/fb.h b/arch/loongarch/include/asm/fb.h
index c6fc7ef374a4a..0b218b10a9ec3 100644
--- a/arch/loongarch/include/asm/fb.h
+++ b/arch/loongarch/include/asm/fb.h
@@ -8,23 +8,23 @@
 #include <linux/compiler.h>
 #include <linux/string.h>
 
-static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+static inline void fb_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
 	memcpy(to, (void __force *)from, n);
 }
-#define fb_memcpy_fromfb fb_memcpy_fromfb
+#define fb_memcpy_fromio fb_memcpy_fromio
 
-static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+static inline void fb_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
 {
 	memcpy((void __force *)to, from, n);
 }
-#define fb_memcpy_tofb fb_memcpy_tofb
+#define fb_memcpy_toio fb_memcpy_toio
 
-static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+static inline void fb_memset_io(volatile void __iomem *addr, int c, size_t n)
 {
 	memset((void __force *)addr, c, n);
 }
-#define fb_memset fb_memset
+#define fb_memset fb_memset_io
 
 #include <asm-generic/fb.h>
 
diff --git a/arch/sparc/include/asm/fb.h b/arch/sparc/include/asm/fb.h
index 077da91aeba1a..572ecd3e1cc48 100644
--- a/arch/sparc/include/asm/fb.h
+++ b/arch/sparc/include/asm/fb.h
@@ -18,23 +18,23 @@ static inline void fb_pgprotect(struct file *file, struct vm_area_struct *vma,
 int fb_is_primary_device(struct fb_info *info);
 #define fb_is_primary_device fb_is_primary_device
 
-static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+static inline void fb_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
 	sbus_memcpy_fromio(to, from, n);
 }
-#define fb_memcpy_fromfb fb_memcpy_fromfb
+#define fb_memcpy_fromio fb_memcpy_fromio
 
-static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+static inline void fb_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
 {
 	sbus_memcpy_toio(to, from, n);
 }
-#define fb_memcpy_tofb fb_memcpy_tofb
+#define fb_memcpy_toio fb_memcpy_toio
 
-static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+static inline void fb_memset_io(volatile void __iomem *addr, int c, size_t n)
 {
 	sbus_memset_io(addr, c, n);
 }
-#define fb_memset fb_memset
+#define fb_memset fb_memset_io
 
 #include <asm-generic/fb.h>
 
diff --git a/drivers/video/fbdev/aty/mach64_cursor.c b/drivers/video/fbdev/aty/mach64_cursor.c
index 4ad0331a8c57d..971355c2cd7e1 100644
--- a/drivers/video/fbdev/aty/mach64_cursor.c
+++ b/drivers/video/fbdev/aty/mach64_cursor.c
@@ -153,7 +153,7 @@ static int atyfb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 	    u8 m, b;
 
 	    // Clear cursor image with 1010101010...
-	    fb_memset(dst, 0xaa, 1024);
+	    fb_memset_io(dst, 0xaa, 1024);
 
 	    offset = align - width*2;
 
diff --git a/drivers/video/fbdev/chipsfb.c b/drivers/video/fbdev/chipsfb.c
index 7799d52a651f3..2a27ba94f652b 100644
--- a/drivers/video/fbdev/chipsfb.c
+++ b/drivers/video/fbdev/chipsfb.c
@@ -332,7 +332,7 @@ static const struct fb_var_screeninfo chipsfb_var = {
 
 static void init_chips(struct fb_info *p, unsigned long addr)
 {
-	fb_memset(p->screen_base, 0, 0x100000);
+	fb_memset_io(p->screen_base, 0, 0x100000);
 
 	p->fix = chipsfb_fix;
 	p->fix.smem_start = addr;
diff --git a/drivers/video/fbdev/core/fb_io_fops.c b/drivers/video/fbdev/core/fb_io_fops.c
index f5299d50f33b9..5985e5e1b040c 100644
--- a/drivers/video/fbdev/core/fb_io_fops.c
+++ b/drivers/video/fbdev/core/fb_io_fops.c
@@ -42,7 +42,7 @@ ssize_t fb_io_read(struct fb_info *info, char __user *buf, size_t count, loff_t
 	while (count) {
 		c  = (count > PAGE_SIZE) ? PAGE_SIZE : count;
 		dst = buffer;
-		fb_memcpy_fromfb(dst, src, c);
+		fb_memcpy_fromio(dst, src, c);
 		dst += c;
 		src += c;
 
@@ -117,7 +117,7 @@ ssize_t fb_io_write(struct fb_info *info, const char __user *buf, size_t count,
 		}
 		c -= trailing;
 
-		fb_memcpy_tofb(dst, src, c);
+		fb_memcpy_toio(dst, src, c);
 		dst += c;
 		src += c;
 		*ppos += c;
diff --git a/drivers/video/fbdev/kyro/fbdev.c b/drivers/video/fbdev/kyro/fbdev.c
index 0596573ef1400..3f277bdb3a321 100644
--- a/drivers/video/fbdev/kyro/fbdev.c
+++ b/drivers/video/fbdev/kyro/fbdev.c
@@ -737,7 +737,7 @@ static int kyrofb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			       info->var.bits_per_pixel);
 	size *= info->var.yres_virtual;
 
-	fb_memset(info->screen_base, 0, size);
+	fb_memset_io(info->screen_base, 0, size);
 
 	if (register_framebuffer(info) < 0)
 		goto out_unmap;
diff --git a/drivers/video/fbdev/pvr2fb.c b/drivers/video/fbdev/pvr2fb.c
index 550fdb5b4d41e..c692cd597ce32 100644
--- a/drivers/video/fbdev/pvr2fb.c
+++ b/drivers/video/fbdev/pvr2fb.c
@@ -801,7 +801,7 @@ static int __maybe_unused pvr2fb_common_init(void)
 		goto out_err;
 	}
 
-	fb_memset(fb_info->screen_base, 0, pvr2_fix.smem_len);
+	fb_memset_io(fb_info->screen_base, 0, pvr2_fix.smem_len);
 
 	pvr2_fix.ypanstep	= nopan  ? 0 : 1;
 	pvr2_fix.ywrapstep	= nowrap ? 0 : 1;
diff --git a/drivers/video/fbdev/sstfb.c b/drivers/video/fbdev/sstfb.c
index da296b2ab54a3..582324f5d869a 100644
--- a/drivers/video/fbdev/sstfb.c
+++ b/drivers/video/fbdev/sstfb.c
@@ -335,7 +335,7 @@ static int sst_calc_pll(const int freq, int *freq_out, struct pll_timing *t)
 static void sstfb_clear_screen(struct fb_info *info)
 {
 	/* clear screen */
-	fb_memset(info->screen_base, 0, info->fix.smem_len);
+	fb_memset_io(info->screen_base, 0, info->fix.smem_len);
 }
 
 
diff --git a/drivers/video/fbdev/stifb.c b/drivers/video/fbdev/stifb.c
index baca6974e288e..01363dccfdaf7 100644
--- a/drivers/video/fbdev/stifb.c
+++ b/drivers/video/fbdev/stifb.c
@@ -527,8 +527,8 @@ rattlerSetupPlanes(struct stifb_info *fb)
 	fb->id = saved_id;
 
 	for (y = 0; y < fb->info.var.yres; ++y)
-		fb_memset(fb->info.screen_base + y * fb->info.fix.line_length,
-			0xff, fb->info.var.xres * fb->info.var.bits_per_pixel/8);
+		fb_memset_io(fb->info.screen_base + y * fb->info.fix.line_length,
+			     0xff, fb->info.var.xres * fb->info.var.bits_per_pixel/8);
 
 	CRX24_SET_OVLY_MASK(fb);
 	SETUP_FB(fb);
diff --git a/drivers/video/fbdev/tdfxfb.c b/drivers/video/fbdev/tdfxfb.c
index d17e5e1472aa6..cdf8e9fe99487 100644
--- a/drivers/video/fbdev/tdfxfb.c
+++ b/drivers/video/fbdev/tdfxfb.c
@@ -1116,7 +1116,7 @@ static int tdfxfb_cursor(struct fb_info *info, struct fb_cursor *cursor)
 		u8 *mask = (u8 *)cursor->mask;
 		int i;
 
-		fb_memset(cursorbase, 0, 1024);
+		fb_memset_io(cursorbase, 0, 1024);
 
 		for (i = 0; i < cursor->image.height; i++) {
 			int h = 0;
diff --git a/include/asm-generic/fb.h b/include/asm-generic/fb.h
index 0540eccdbeca1..bb7ee9c70e603 100644
--- a/include/asm-generic/fb.h
+++ b/include/asm-generic/fb.h
@@ -108,28 +108,28 @@ static inline void fb_writeq(u64 b, volatile void __iomem *addr)
 #endif
 #endif
 
-#ifndef fb_memcpy_fromfb
-static inline void fb_memcpy_fromfb(void *to, const volatile void __iomem *from, size_t n)
+#ifndef fb_memcpy_fromio
+static inline void fb_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
 	memcpy_fromio(to, from, n);
 }
-#define fb_memcpy_fromfb fb_memcpy_fromfb
+#define fb_memcpy_fromio fb_memcpy_fromio
 #endif
 
-#ifndef fb_memcpy_tofb
-static inline void fb_memcpy_tofb(volatile void __iomem *to, const void *from, size_t n)
+#ifndef fb_memcpy_toio
+static inline void fb_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
 {
 	memcpy_toio(to, from, n);
 }
-#define fb_memcpy_tofb fb_memcpy_tofb
+#define fb_memcpy_toio fb_memcpy_toio
 #endif
 
 #ifndef fb_memset
-static inline void fb_memset(volatile void __iomem *addr, int c, size_t n)
+static inline void fb_memset_io(volatile void __iomem *addr, int c, size_t n)
 {
 	memset_io(addr, c, n);
 }
-#define fb_memset fb_memset
+#define fb_memset fb_memset_io
 #endif
 
 #endif /* __ASM_GENERIC_FB_H_ */
-- 
GitLab


From 615ed9ece01814a94fb544226cb3f4e03f06851d Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 12 May 2023 15:00:03 +0300
Subject: [PATCH 0544/1791] drm/i915/mtl: Fix expected reg value for
 Thunderbolt PLL disabling

While disabling Thunderbolt PLL, we request PLL to be stopped and
wait for ACK bit to be cleared. The expected value should be '0'
instead of '~XELPDP_TBT_CLOCK_ACK' or otherwise we incorrectly
receive dmesg warn "PHY PLL not unlocked in 10us".

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512120003.587360-1-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index ef0615cdc8a0c..c0755ac796ee0 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2861,9 +2861,7 @@ static void intel_mtl_tbt_pll_disable(struct intel_encoder *encoder)
 
 	/* 3. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "0". */
 	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port),
-					 XELPDP_TBT_CLOCK_ACK,
-					 ~XELPDP_TBT_CLOCK_ACK,
-					 10, 0, NULL))
+					 XELPDP_TBT_CLOCK_ACK, 0, 10, 0, NULL))
 		drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not unlocked after 10us.\n",
 			 encoder->base.base.id, encoder->base.name, phy_name(phy));
 
-- 
GitLab


From 25e7976db86bed5f1826bdd5c59e5be424a9d91f Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <andrzej.hajda@intel.com>
Date: Thu, 18 May 2023 16:50:52 +0200
Subject: [PATCH 0545/1791] drm/i915/mtl: do not enable render power-gating on
 MTL

Multiple CI tests fails with forcewake ack timeouts if render
power gating is enabled.
BSpec 52698 states it should be 0 for MTL, but apparently
this info is outdated. Anyway since the patch makes MTL pass basic
tests added FIXME tag informing this is temporary workaround.

v2: added FIXME tag

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/4983
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517-mtl_disable_render_pg-v2-1-0b51180a43f0@intel.com
---
 drivers/gpu/drm/i915/gt/intel_rc6.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
index 908a3d0f2343f..58bb1c55294c9 100644
--- a/drivers/gpu/drm/i915/gt/intel_rc6.c
+++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
@@ -117,8 +117,14 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6)
 			GEN6_RC_CTL_RC6_ENABLE |
 			GEN6_RC_CTL_EI_MODE(1);
 
-	/* Wa_16011777198 - Render powergating must remain disabled */
-	if (IS_DG2_GRAPHICS_STEP(gt->i915, G10, STEP_A0, STEP_C0) ||
+	/*
+	 * Wa_16011777198 and BSpec 52698 - Render powergating must be off.
+	 * FIXME BSpec is outdated, disabling powergating for MTL is just
+	 * temporary wa and should be removed after fixing real cause
+	 * of forcewake timeouts.
+	 */
+	if (IS_METEORLAKE(gt->i915) ||
+	    IS_DG2_GRAPHICS_STEP(gt->i915, G10, STEP_A0, STEP_C0) ||
 	    IS_DG2_GRAPHICS_STEP(gt->i915, G11, STEP_A0, STEP_B0))
 		pg_enable =
 			GEN9_MEDIA_PG_ENABLE |
-- 
GitLab


From 5f12c8d0a761af5dda2e798b1af56ff967442b83 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Wed, 17 May 2023 16:31:10 -0700
Subject: [PATCH 0546/1791] drm/i915/mtl: Add MTL performance tuning changes

MTL reuses the tuning parameters for DG2. Extend the dg2
performance tuning parameters to MTL.

v2: Add DRAW_WATERMARK tuning parameter.
v3: Limit DRAW_WATERMARK tuning to non A0 step.
v4: Reorder platform checks.
    Restrict Blend fill caching optimization to Render GT.
v5: Move mtl tuning params to its own function

Bspec: 68331
Cc: Haridhar Kalvala <haridhar.kalvala@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517233111.297542-1-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 23 ++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 786349e954873..4d2dece960115 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -812,11 +812,25 @@ static void dg2_ctx_workarounds_init(struct intel_engine_cs *engine,
 	wa_masked_en(wal, CACHE_MODE_1, MSAA_OPTIMIZATION_REDUC_DISABLE);
 }
 
+static void mtl_ctx_gt_tuning_init(struct intel_engine_cs *engine,
+				   struct i915_wa_list *wal)
+{
+	struct drm_i915_private *i915 = engine->i915;
+
+	dg2_ctx_gt_tuning_init(engine, wal);
+
+	if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_B0, STEP_FOREVER) ||
+	    IS_MTL_GRAPHICS_STEP(i915, P, STEP_B0, STEP_FOREVER))
+		wa_add(wal, DRAW_WATERMARK, VERT_WM_VAL, 0x3FF, 0, false);
+}
+
 static void mtl_ctx_workarounds_init(struct intel_engine_cs *engine,
 				     struct i915_wa_list *wal)
 {
 	struct drm_i915_private *i915 = engine->i915;
 
+	mtl_ctx_gt_tuning_init(engine, wal);
+
 	if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_A0, STEP_B0) ||
 	    IS_MTL_GRAPHICS_STEP(i915, P, STEP_A0, STEP_B0)) {
 		/* Wa_14014947963 */
@@ -1748,6 +1762,13 @@ xelpmp_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal)
  */
 static void gt_tuning_settings(struct intel_gt *gt, struct i915_wa_list *wal)
 {
+	if (IS_METEORLAKE(gt->i915)) {
+		if (gt->type != GT_MEDIA)
+			wa_mcr_write_or(wal, XEHP_L3SCQREG7, BLEND_FILL_CACHING_OPT_DIS);
+
+		wa_mcr_write_or(wal, XEHP_SQCM, EN_32B_ACCESS);
+	}
+
 	if (IS_PONTEVECCHIO(gt->i915)) {
 		wa_mcr_write(wal, XEHPC_L3SCRUB,
 			     SCRUB_CL_DWNGRADE_SHARED | SCRUB_RATE_4B_PER_CLK);
@@ -2944,7 +2965,7 @@ static void
 add_render_compute_tuning_settings(struct drm_i915_private *i915,
 				   struct i915_wa_list *wal)
 {
-	if (IS_DG2(i915))
+	if (IS_METEORLAKE(i915) || IS_DG2(i915))
 		wa_mcr_write_clr_set(wal, RT_CTRL, STACKID_CTRL, STACKID_CTRL_512);
 
 	/*
-- 
GitLab


From 1a365a2b24cda48ff8d441e91663a6c0ab1353a9 Mon Sep 17 00:00:00 2001
From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Date: Wed, 17 May 2023 16:31:11 -0700
Subject: [PATCH 0547/1791] drm/i915/mtl: Extend Wa_16014892111 to MTL A-step

Like DG2, MTL a-step hardware is subject to Wa_16014892111 which
requires that any changes made to the DRAW_WATERMARK register be
done via an INDIRECT_CTX batch buffer rather than through a regular
context workaround.

The bspec gives the same non-default recommended tuning value
for DRAW_WATERMARK as DG2, so we can re-use the INDIRECT_CTX code
to apply that tuning setting on A-step hardware.

Application of the tuning setting on B-step and later does not
need INDIRECT_CTX handling and is already done in
mtl_ctx_workarounds_init() as usual.

v2: Limit the WA for A-step
v3: Update the commit message.
v4: Reorder platform checks and update commit message.

Bspec: 68331
Cc: Haridhar Kalvala <haridhar.kalvala@intel.com>
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517233111.297542-2-radhakrishna.sripada@intel.com
---
 drivers/gpu/drm/i915/gt/intel_lrc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index 81a96c52a92b3..a4ec20aaafe28 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1370,7 +1370,9 @@ gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs)
 					      cs, GEN12_GFX_CCS_AUX_NV);
 
 	/* Wa_16014892111 */
-	if (IS_DG2(ce->engine->i915))
+	if (IS_MTL_GRAPHICS_STEP(ce->engine->i915, M, STEP_A0, STEP_B0) ||
+	    IS_MTL_GRAPHICS_STEP(ce->engine->i915, P, STEP_A0, STEP_B0) ||
+	    IS_DG2(ce->engine->i915))
 		cs = dg2_emit_draw_watermark_setting(cs);
 
 	return cs;
-- 
GitLab


From e072700869dd96405a9c3752d3741a79bca6e2e2 Mon Sep 17 00:00:00 2001
From: Luben Tuikov <luben.tuikov@amd.com>
Date: Wed, 17 May 2023 19:35:49 -0400
Subject: [PATCH 0548/1791] drm/sched: Rename to drm_sched_can_queue()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename drm_sched_ready() to drm_sched_can_queue(). "ready" can mean many
things and is thus meaningless in this context. Instead, rename to a name
which precisely conveys what is being checked.

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
Link: https://lore.kernel.org/r/20230517233550.377847-1-luben.tuikov@amd.com
---
 drivers/gpu/drm/scheduler/sched_main.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 172e63c87bfcf..8739322c30321 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -848,13 +848,12 @@ void drm_sched_job_cleanup(struct drm_sched_job *job)
 EXPORT_SYMBOL(drm_sched_job_cleanup);
 
 /**
- * drm_sched_ready - is the scheduler ready
- *
+ * drm_sched_can_queue -- Can we queue more to the hardware?
  * @sched: scheduler instance
  *
  * Return true if we can push more jobs to the hw, otherwise false.
  */
-static bool drm_sched_ready(struct drm_gpu_scheduler *sched)
+static bool drm_sched_can_queue(struct drm_gpu_scheduler *sched)
 {
 	return atomic_read(&sched->hw_rq_count) <
 		sched->hw_submission_limit;
@@ -868,7 +867,7 @@ static bool drm_sched_ready(struct drm_gpu_scheduler *sched)
  */
 void drm_sched_wakeup(struct drm_gpu_scheduler *sched)
 {
-	if (drm_sched_ready(sched))
+	if (drm_sched_can_queue(sched))
 		wake_up_interruptible(&sched->wake_up_worker);
 }
 
@@ -885,7 +884,7 @@ drm_sched_select_entity(struct drm_gpu_scheduler *sched)
 	struct drm_sched_entity *entity;
 	int i;
 
-	if (!drm_sched_ready(sched))
+	if (!drm_sched_can_queue(sched))
 		return NULL;
 
 	/* Kernel run queue has higher priority than normal run queue*/
-- 
GitLab


From 3655c5900f4d49881ad09e3893e5f5516b06a9f1 Mon Sep 17 00:00:00 2001
From: Luben Tuikov <luben.tuikov@amd.com>
Date: Wed, 17 May 2023 19:35:50 -0400
Subject: [PATCH 0549/1791] drm/sched: Rename to
 drm_sched_wakeup_if_can_queue()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename drm_sched_wakeup() to drm_sched_wakeup_if_canqueue() since the former
is misleading, as it wakes up the GPU scheduler _only if_ more jobs can be
queued to the underlying hardware.

This distinction is important to make, since the wake conditional in the GPU
scheduler thread wakes up when other conditions are also true, e.g. when there
are jobs to be cleaned. For instance, a user might want to wake up the
scheduler only because there are more jobs to clean, but whether we can queue
more jobs is irrelevant.

v2: Separate "canqueue" to "can_queue". (Alex D.)

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://lore.kernel.org/r/20230517233550.377847-2-luben.tuikov@amd.com
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
---
 drivers/gpu/drm/scheduler/sched_entity.c | 4 ++--
 drivers/gpu/drm/scheduler/sched_main.c   | 6 +++---
 include/drm/gpu_scheduler.h              | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index cfb433e920058..68e807ae136ad 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -342,7 +342,7 @@ static void drm_sched_entity_wakeup(struct dma_fence *f,
 		container_of(cb, struct drm_sched_entity, cb);
 
 	drm_sched_entity_clear_dep(f, cb);
-	drm_sched_wakeup(entity->rq->sched);
+	drm_sched_wakeup_if_can_queue(entity->rq->sched);
 }
 
 /**
@@ -565,7 +565,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 		if (drm_sched_policy == DRM_SCHED_POLICY_FIFO)
 			drm_sched_rq_update_fifo(entity, submit_ts);
 
-		drm_sched_wakeup(entity->rq->sched);
+		drm_sched_wakeup_if_can_queue(entity->rq->sched);
 	}
 }
 EXPORT_SYMBOL(drm_sched_entity_push_job);
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 8739322c30321..b352227a60553 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -860,12 +860,12 @@ static bool drm_sched_can_queue(struct drm_gpu_scheduler *sched)
 }
 
 /**
- * drm_sched_wakeup - Wake up the scheduler when it is ready
- *
+ * drm_sched_wakeup_if_can_queue - Wake up the scheduler
  * @sched: scheduler instance
  *
+ * Wake up the scheduler if we can queue jobs.
  */
-void drm_sched_wakeup(struct drm_gpu_scheduler *sched)
+void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched)
 {
 	if (drm_sched_can_queue(sched))
 		wake_up_interruptible(&sched->wake_up_worker);
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index 31d1f5166c79f..e95b4837e5a37 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -549,7 +549,7 @@ void drm_sched_entity_modify_sched(struct drm_sched_entity *entity,
                                    unsigned int num_sched_list);
 
 void drm_sched_job_cleanup(struct drm_sched_job *job);
-void drm_sched_wakeup(struct drm_gpu_scheduler *sched);
+void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched);
 void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad);
 void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recovery);
 void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched);
-- 
GitLab


From 4f96bb37e645cafbfde2f6f40aabc5dcd188295d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 9 Feb 2023 15:38:34 +0200
Subject: [PATCH 0550/1791] dt-bindings: display/msm/gmu: add Adreno 660
 support

Add Adreno A660 to the A635 clause to define all version-specific
properties. There is no need to add it to the top-level clause, since
top-level compatible uses pattern to define compatible strings.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/521835/
Link: https://lore.kernel.org/r/20230209133839.762631-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 Documentation/devicetree/bindings/display/msm/gmu.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/msm/gmu.yaml b/Documentation/devicetree/bindings/display/msm/gmu.yaml
index ab14e81cb0506..d5ce0dff4220c 100644
--- a/Documentation/devicetree/bindings/display/msm/gmu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gmu.yaml
@@ -122,6 +122,7 @@ allOf:
           contains:
             enum:
               - qcom,adreno-gmu-635.0
+              - qcom,adreno-gmu-660.1
     then:
       properties:
         reg:
-- 
GitLab


From 58fab79731db95776ff743035a06094bb528547d Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 21 Mar 2023 22:35:56 +0100
Subject: [PATCH 0551/1791] dt-bindings: display: msm: sm8350-mdss: Fix DSI
 compatible

The DSI compatible changed between patchset revisions, but that wasn't
reflected in the bindings. Fix it.

Fixes: 430e11f42bff ("dt-bindings: display: msm: Add qcom, sm8350-mdss binding")
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: David Heidelberg <david@ixit.cz>
Patchwork: https://patchwork.freedesktop.org/patch/528037/
Link: https://lore.kernel.org/r/20230321213557.1737905-1-konrad.dybcio@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml       | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
index 4d94dbff30541..79a226e4cc6a2 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml
@@ -64,7 +64,7 @@ patternProperties:
     type: object
     properties:
       compatible:
-        const: qcom,dsi-phy-5nm-8350
+        const: qcom,sm8350-dsi-phy-5nm
 
 unevaluatedProperties: false
 
-- 
GitLab


From f26e0acc3e07ebec228503a995492925690380db Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 8 Apr 2023 03:27:47 +0300
Subject: [PATCH 0552/1791] drm/msm/dpu: enable DPU_CTL_SPLIT_DISPLAY for
 sc8280xp

Theoretically, since sm8150 we should be using a single CTL for the
split panel case, but since we do not support it for now, fallback to
DPU_CTL_SPLIT_DISPLAY.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/531489/
Link: https://lore.kernel.org/r/20230408002750.2722304-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index 706d0f13b598e..88b8226e6f75d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -42,17 +42,18 @@ static const struct dpu_mdp_cfg sc8280xp_mdp[] = {
 	},
 };
 
+/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */
 static const struct dpu_ctl_cfg sc8280xp_ctl[] = {
 	{
 	.name = "ctl_0", .id = CTL_0,
 	.base = 0x15000, .len = 0x204,
-	.features = CTL_SC7280_MASK,
+	.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
 	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	},
 	{
 	.name = "ctl_1", .id = CTL_1,
 	.base = 0x16000, .len = 0x204,
-	.features = CTL_SC7280_MASK,
+	.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
 	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
 	},
 	{
-- 
GitLab


From 800d28a4988d842d11e95d3dea491520ebf4991d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 8 Apr 2023 03:27:48 +0300
Subject: [PATCH 0553/1791] drm/msm/dpu: enable DSPP_2/3 for LM_2/3 on sm8450

Mark DSPP_2 and DSPP_3 as used for LM_2 and LM_3

Fixes: 100d7ef6995d ("drm/msm/dpu: add support for SM8450")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/531491/
Link: https://lore.kernel.org/r/20230408002750.2722304-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 4ecb3df5cbc02..8bd4bb97e639c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -107,9 +107,9 @@ static const struct dpu_lm_cfg sm8450_lm[] = {
 	LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1),
 	LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_2, LM_3, 0),
+		&sdm845_lm_sblk, PINGPONG_2, LM_3, DSPP_2),
 	LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_3, LM_2, 0),
+		&sdm845_lm_sblk, PINGPONG_3, LM_2, DSPP_3),
 	LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_4, LM_5, 0),
 	LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK,
-- 
GitLab


From f5abecfe339e4d8183fbb208da674f4a49b7f722 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 8 Apr 2023 03:27:49 +0300
Subject: [PATCH 0554/1791] drm/msm/dpu: enable DSPP and DSC on sc8180x

Enable DSPP and DSC hardware blocks on sc8180x platform.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/531490/
Link: https://lore.kernel.org/r/20230408002750.2722304-4-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h   | 28 +++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index e3bdfe7b30f1f..3911ae4925404 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -102,9 +102,9 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = {
 
 static const struct dpu_lm_cfg sc8180x_lm[] = {
 	LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_0, LM_1, 0),
+		&sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0),
 	LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_1, LM_0, 0),
+		&sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1),
 	LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_2, LM_3, 0),
 	LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK,
@@ -115,6 +115,17 @@ static const struct dpu_lm_cfg sc8180x_lm[] = {
 		&sdm845_lm_sblk, PINGPONG_5, LM_4, 0),
 };
 
+static const struct dpu_dspp_cfg sc8180x_dspp[] = {
+	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+};
+
 static const struct dpu_pingpong_cfg sc8180x_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
@@ -142,6 +153,15 @@ static const struct dpu_merge_3d_cfg sc8180x_merge_3d[] = {
 	MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x83200),
 };
 
+static const struct dpu_dsc_cfg sc8180x_dsc[] = {
+	DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)),
+	DSC_BLK("dsc_1", DSC_1, 0x80400, BIT(DPU_DSC_OUTPUT_CTRL)),
+	DSC_BLK("dsc_2", DSC_2, 0x80800, BIT(DPU_DSC_OUTPUT_CTRL)),
+	DSC_BLK("dsc_3", DSC_3, 0x80c00, BIT(DPU_DSC_OUTPUT_CTRL)),
+	DSC_BLK("dsc_4", DSC_4, 0x81000, BIT(DPU_DSC_OUTPUT_CTRL)),
+	DSC_BLK("dsc_5", DSC_5, 0x81400, BIT(DPU_DSC_OUTPUT_CTRL)),
+};
+
 static const struct dpu_intf_cfg sc8180x_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
 	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
@@ -190,6 +210,10 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = {
 	.sspp = sc8180x_sspp,
 	.mixer_count = ARRAY_SIZE(sc8180x_lm),
 	.mixer = sc8180x_lm,
+	.dspp_count = ARRAY_SIZE(sc8180x_dspp),
+	.dspp = sc8180x_dspp,
+	.dsc_count = ARRAY_SIZE(sc8180x_dsc),
+	.dsc = sc8180x_dsc,
 	.pingpong_count = ARRAY_SIZE(sc8180x_pp),
 	.pingpong = sc8180x_pp,
 	.merge_3d_count = ARRAY_SIZE(sc8180x_merge_3d),
-- 
GitLab


From 23bdc972daf436412d55ad7fb2f1bf430a001ffd Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 8 Apr 2023 03:27:50 +0300
Subject: [PATCH 0555/1791] drm/msm/dpu: use CTL_SC7280_MASK for sm8450's ctl_0

On sm8450 platform the CTL_0 doesn't differ from the rest of CTL blocks,
so switch it to CTL_SC7280_MASK too.

Some background: original commit 100d7ef6995d ("drm/msm/dpu: add support
for SM8450") had all (relevant at that time) bit spelled individually.
Then commit 0e91bcbb0016 ("drm/msm/dpu: Add SM8350 to hw catalog"),
despite being a mismerge, correctly changed all other CTL entries to use
CTL_SC7280_MASK, except CTL_0.

While the current BLOCK_SOC_MASK style is not ideal (and while we are
working on a better scheme), let's follow its usage as a least minimal
surprise. For example, sc8280xp, a close associate of sm8450, also uses
CTL_SC7280_MASK.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/531493/
Link: https://lore.kernel.org/r/20230408002750.2722304-5-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 8bd4bb97e639c..cb83d25834ceb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -47,7 +47,7 @@ static const struct dpu_ctl_cfg sm8450_ctl[] = {
 	{
 	.name = "ctl_0", .id = CTL_0,
 	.base = 0x15000, .len = 0x204,
-	.features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY) | BIT(DPU_CTL_FETCH_ACTIVE),
+	.features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK,
 	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
 	},
 	{
-- 
GitLab


From 5a7a86bfa2cd79c80e4f3bdb01feb15edd08b92b Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 15 Apr 2023 20:19:25 +0300
Subject: [PATCH 0556/1791] drm/msm/dpu: simplify intf allocation code

Rather than passing DRM_MODE_ENCODER_* and letting dpu_encoder to guess,
which intf type we mean, pass INTF_DSI/INTF_DP directly. This is
required to support HDMI output in DPU, as both DP and HDMI encoders are
DRM_MODE_ENCODER_TMDS. Thus dpu_encoder code can not make a difference
between HDMI and DP outputs.

Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Arnaud Vrac <rawoul@gmail.com>
Tested-by: Arnaud Vrac <rawoul@gmail.com>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/532370/
Link: https://lore.kernel.org/r/20230415171926.85774-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 39 +++++++--------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h |  4 +--
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c     |  6 ++--
 3 files changed, 18 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 1dc5dbe585723..b34416cbd0f5f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -495,7 +495,7 @@ void dpu_encoder_helper_split_config(
 	hw_mdptop = phys_enc->hw_mdptop;
 	disp_info = &dpu_enc->disp_info;
 
-	if (disp_info->intf_type != DRM_MODE_ENCODER_DSI)
+	if (disp_info->intf_type != INTF_DSI)
 		return;
 
 	/**
@@ -1127,7 +1127,7 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc)
 	}
 
 
-	if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_TMDS &&
+	if (dpu_enc->disp_info.intf_type == INTF_DP &&
 		dpu_enc->cur_master->hw_mdptop &&
 		dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select)
 		dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select(
@@ -1135,7 +1135,7 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc)
 
 	_dpu_encoder_update_vsync_source(dpu_enc, &dpu_enc->disp_info);
 
-	if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_DSI &&
+	if (dpu_enc->disp_info.intf_type == INTF_DSI &&
 			!WARN_ON(dpu_enc->num_phys_encs == 0)) {
 		unsigned bpc = dpu_enc->connector->display_info.bpc;
 		for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
@@ -1977,7 +1977,7 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc)
 			phys->ops.handle_post_kickoff(phys);
 	}
 
-	if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_DSI &&
+	if (dpu_enc->disp_info.intf_type == INTF_DSI &&
 			!dpu_encoder_vsync_time(drm_enc, &wakeup_time)) {
 		trace_dpu_enc_early_kickoff(DRMID(drm_enc),
 					    ktime_to_ms(wakeup_time));
@@ -2182,7 +2182,7 @@ static int dpu_encoder_virt_add_phys_encs(
 	}
 
 
-	if (disp_info->intf_type == DRM_MODE_ENCODER_VIRTUAL) {
+	if (disp_info->intf_type == INTF_WB) {
 		enc = dpu_encoder_phys_wb_init(params);
 
 		if (IS_ERR(enc)) {
@@ -2231,7 +2231,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 {
 	int ret = 0;
 	int i = 0;
-	enum dpu_intf_type intf_type = INTF_NONE;
 	struct dpu_enc_phys_init_params phys_params;
 
 	if (!dpu_enc) {
@@ -2246,23 +2245,11 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 	phys_params.parent = &dpu_enc->base;
 	phys_params.enc_spinlock = &dpu_enc->enc_spinlock;
 
-	switch (disp_info->intf_type) {
-	case DRM_MODE_ENCODER_DSI:
-		intf_type = INTF_DSI;
-		break;
-	case DRM_MODE_ENCODER_TMDS:
-		intf_type = INTF_DP;
-		break;
-	case DRM_MODE_ENCODER_VIRTUAL:
-		intf_type = INTF_WB;
-		break;
-	}
-
 	WARN_ON(disp_info->num_of_h_tiles < 1);
 
 	DPU_DEBUG("dsi_info->num_of_h_tiles %d\n", disp_info->num_of_h_tiles);
 
-	if (disp_info->intf_type != DRM_MODE_ENCODER_VIRTUAL)
+	if (disp_info->intf_type != INTF_WB)
 		dpu_enc->idle_pc_supported =
 				dpu_kms->catalog->caps->has_idle_pc;
 
@@ -2290,11 +2277,11 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 				i, controller_id, phys_params.split_role);
 
 		phys_params.intf_idx = dpu_encoder_get_intf(dpu_kms->catalog,
-													intf_type,
-													controller_id);
+							    disp_info->intf_type,
+							    controller_id);
 
 		phys_params.wb_idx = dpu_encoder_get_wb(dpu_kms->catalog,
-				intf_type, controller_id);
+				disp_info->intf_type, controller_id);
 		/*
 		 * The phys_params might represent either an INTF or a WB unit, but not
 		 * both of them at the same time.
@@ -2302,14 +2289,14 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		if ((phys_params.intf_idx == INTF_MAX) &&
 				(phys_params.wb_idx == WB_MAX)) {
 			DPU_ERROR_ENC(dpu_enc, "could not get intf or wb: type %d, id %d\n",
-						  intf_type, controller_id);
+						  disp_info->intf_type, controller_id);
 			ret = -EINVAL;
 		}
 
 		if ((phys_params.intf_idx != INTF_MAX) &&
 				(phys_params.wb_idx != WB_MAX)) {
 			DPU_ERROR_ENC(dpu_enc, "both intf and wb present: type %d, id %d\n",
-						  intf_type, controller_id);
+						  disp_info->intf_type, controller_id);
 			ret = -EINVAL;
 		}
 
@@ -2409,11 +2396,11 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
 	timer_setup(&dpu_enc->frame_done_timer,
 			dpu_encoder_frame_done_timeout, 0);
 
-	if (disp_info->intf_type == DRM_MODE_ENCODER_DSI)
+	if (disp_info->intf_type == INTF_DSI)
 		timer_setup(&dpu_enc->vsync_event_timer,
 				dpu_encoder_vsync_event_handler,
 				0);
-	else if (disp_info->intf_type == DRM_MODE_ENCODER_TMDS)
+	else if (disp_info->intf_type == INTF_DP)
 		dpu_enc->wide_bus_en = msm_dp_wide_bus_available(
 				priv->dp[disp_info->h_tile_instance[0]]);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
index 2c9ef8d1b8779..6d14f84dd43fb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
@@ -21,7 +21,7 @@
 
 /**
  * struct msm_display_info - defines display properties
- * @intf_type:          DRM_MODE_ENCODER_ type
+ * @intf_type:          INTF_ type
  * @num_of_h_tiles:     Number of horizontal tiles in case of split interface
  * @h_tile_instance:    Controller instance used per tile. Number of elements is
  *                      based on num_of_h_tiles
@@ -31,7 +31,7 @@
  * @dsc:		DSC configuration data for DSC-enabled displays
  */
 struct msm_display_info {
-	int intf_type;
+	enum dpu_intf_type intf_type;
 	uint32_t num_of_h_tiles;
 	uint32_t h_tile_instance[MAX_H_TILES_PER_DISPLAY];
 	bool is_cmd_mode;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 0e7a68714e9e1..e85e3721d2c71 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -542,7 +542,7 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 		}
 
 		memset(&info, 0, sizeof(info));
-		info.intf_type = encoder->encoder_type;
+		info.intf_type = INTF_DSI;
 
 		rc = msm_dsi_modeset_init(priv->dsi[i], dev, encoder);
 		if (rc) {
@@ -605,7 +605,7 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev,
 
 		info.num_of_h_tiles = 1;
 		info.h_tile_instance[0] = i;
-		info.intf_type = encoder->encoder_type;
+		info.intf_type = INTF_DP;
 		rc = dpu_encoder_setup(dev, encoder, &info);
 		if (rc) {
 			DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
@@ -644,7 +644,7 @@ static int _dpu_kms_initialize_writeback(struct drm_device *dev,
 	info.num_of_h_tiles = 1;
 	/* use only WB idx 2 instance for DPU */
 	info.h_tile_instance[0] = WB_2;
-	info.intf_type = encoder->encoder_type;
+	info.intf_type = INTF_WB;
 
 	rc = dpu_encoder_setup(dev, encoder, &info);
 	if (rc) {
-- 
GitLab


From cb7b04c83e9006c39af6d806761fc628573920e8 Mon Sep 17 00:00:00 2001
From: Clint Taylor <clinton.a.taylor@intel.com>
Date: Mon, 15 May 2023 16:17:24 -0700
Subject: [PATCH 0557/1791] drm/i915: Add 16bit register/mask operators

Add the support macros to define/extract bits as 16bits.

v2: checkpatch fixes
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515231725.3815199-2-clinton.a.taylor@intel.com
---
 drivers/gpu/drm/i915/i915_reg_defs.h | 48 ++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_reg_defs.h b/drivers/gpu/drm/i915/i915_reg_defs.h
index 622d603080f9e..a685db1e815d8 100644
--- a/drivers/gpu/drm/i915/i915_reg_defs.h
+++ b/drivers/gpu/drm/i915/i915_reg_defs.h
@@ -143,6 +143,54 @@
  */
 #define REG_FIELD_GET64(__mask, __val)	((u64)FIELD_GET(__mask, __val))
 
+/**
+ * REG_BIT16() - Prepare a u16 bit value
+ * @__n: 0-based bit number
+ *
+ * Local wrapper for BIT() to force u16, with compile time
+ * checks.
+ *
+ * @return: Value with bit @__n set.
+ */
+#define REG_BIT16(__n)                                                   \
+	((u16)(BIT(__n) +                                                \
+	       BUILD_BUG_ON_ZERO(__is_constexpr(__n) &&         \
+				 ((__n) < 0 || (__n) > 15))))
+
+/**
+ * REG_GENMASK16() - Prepare a continuous u8 bitmask
+ * @__high: 0-based high bit
+ * @__low: 0-based low bit
+ *
+ * Local wrapper for GENMASK() to force u16, with compile time
+ * checks.
+ *
+ * @return: Continuous bitmask from @__high to @__low, inclusive.
+ */
+#define REG_GENMASK16(__high, __low)                                     \
+	((u16)(GENMASK(__high, __low) +                                  \
+	       BUILD_BUG_ON_ZERO(__is_constexpr(__high) &&      \
+				 __is_constexpr(__low) &&               \
+				 ((__low) < 0 || (__high) > 15 || (__low) > (__high)))))
+
+/**
+ * REG_FIELD_PREP16() - Prepare a u16 bitfield value
+ * @__mask: shifted mask defining the field's length and position
+ * @__val: value to put in the field
+ *
+ * Local copy of FIELD_PREP16() to generate an integer constant
+ * expression, force u8 and for consistency with
+ * REG_FIELD_GET16(), REG_BIT16() and REG_GENMASK16().
+ *
+ * @return: @__val masked and shifted into the field defined by @__mask.
+ */
+#define REG_FIELD_PREP16(__mask, __val)                                          \
+	((u16)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) +      \
+	       BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) +             \
+	       BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U16_MAX) +          \
+	       BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \
+	       BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0))))
+
 #define __MASKED_FIELD(mask, value) ((mask) << 16 | (value))
 #define _MASKED_FIELD(mask, value) ({					   \
 	if (__builtin_constant_p(mask))					   \
-- 
GitLab


From 234fcb978f61f53bc05c276f6204332fde2b4951 Mon Sep 17 00:00:00 2001
From: Clint Taylor <clinton.a.taylor@intel.com>
Date: Mon, 15 May 2023 16:17:25 -0700
Subject: [PATCH 0558/1791] drm/i915/hdmi: C20 computed PLL frequencies

Use algorithm to generate HDMI C20 PLL clock frequencies.

v2: checkpatch fixes

BSPEC: 64568
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Cc: Mika Kahola <mika.kahola@intel.com>
Cc: Anusha Srivatsa <anusha.srivatsa@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
[mattrope: Wrapped one overly long line]
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515231725.3815199-3-clinton.a.taylor@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 90 +++++++++++++++++--
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h | 53 +++++++++++
 2 files changed, 137 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index c0755ac796ee0..ee6902118860e 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -3,6 +3,8 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include <linux/log2.h>
+#include <linux/math64.h>
 #include "i915_reg.h"
 #include "intel_cx0_phy.h"
 #include "intel_cx0_phy_regs.h"
@@ -1901,6 +1903,75 @@ void intel_c10pll_dump_hw_state(struct drm_i915_private *i915,
 			    i + 2, hw_state->pll[i + 2], i + 3, hw_state->pll[i + 3]);
 }
 
+static int intel_c20_compute_hdmi_tmds_pll(u64 pixel_clock, struct intel_c20pll_state *pll_state)
+{
+	u64 datarate;
+	u64 mpll_tx_clk_div;
+	u64 vco_freq_shift;
+	u64 vco_freq;
+	u64 multiplier;
+	u64 mpll_multiplier;
+	u64 mpll_fracn_quot;
+	u64 mpll_fracn_rem;
+	u8  mpllb_ana_freq_vco;
+	u8  mpll_div_multiplier;
+
+	if (pixel_clock < 25175 || pixel_clock > 600000)
+		return -EINVAL;
+
+	datarate = ((u64)pixel_clock * 1000) * 10;
+	mpll_tx_clk_div = ilog2(div64_u64((u64)CLOCK_9999MHZ, (u64)datarate));
+	vco_freq_shift = ilog2(div64_u64((u64)CLOCK_4999MHZ * (u64)256, (u64)datarate));
+	vco_freq = (datarate << vco_freq_shift) >> 8;
+	multiplier = div64_u64((vco_freq << 28), (REFCLK_38_4_MHZ >> 4));
+	mpll_multiplier = 2 * (multiplier >> 32);
+
+	mpll_fracn_quot = (multiplier >> 16) & 0xFFFF;
+	mpll_fracn_rem  = multiplier & 0xFFFF;
+
+	mpll_div_multiplier = min_t(u8, div64_u64((vco_freq * 16 + (datarate >> 1)),
+						  datarate), 255);
+
+	if (vco_freq <= DATARATE_3000000000)
+		mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_3;
+	else if (vco_freq <= DATARATE_3500000000)
+		mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_2;
+	else if (vco_freq <= DATARATE_4000000000)
+		mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_1;
+	else
+		mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_0;
+
+	pll_state->link_bit_rate	= pixel_clock;
+	pll_state->clock	= pixel_clock;
+	pll_state->tx[0]	= 0xbe88;
+	pll_state->tx[1]	= 0x9800;
+	pll_state->tx[2]	= 0x0000;
+	pll_state->cmn[0]	= 0x0500;
+	pll_state->cmn[1]	= 0x0005;
+	pll_state->cmn[2]	= 0x0000;
+	pll_state->cmn[3]	= 0x0000;
+	pll_state->mpllb[0]	= (MPLL_TX_CLK_DIV(mpll_tx_clk_div) |
+				   MPLL_MULTIPLIER(mpll_multiplier));
+	pll_state->mpllb[1]	= (CAL_DAC_CODE(CAL_DAC_CODE_31) |
+				   WORD_CLK_DIV |
+				   MPLL_DIV_MULTIPLIER(mpll_div_multiplier));
+	pll_state->mpllb[2]	= (MPLLB_ANA_FREQ_VCO(mpllb_ana_freq_vco) |
+				   CP_PROP(CP_PROP_20) |
+				   CP_INT(CP_INT_6));
+	pll_state->mpllb[3]	= (V2I(V2I_2) |
+				   CP_PROP_GS(CP_PROP_GS_30) |
+				   CP_INT_GS(CP_INT_GS_28));
+	pll_state->mpllb[4]	= 0x0000;
+	pll_state->mpllb[5]	= 0x0000;
+	pll_state->mpllb[6]	= (C20_MPLLB_FRACEN | SSC_UP_SPREAD);
+	pll_state->mpllb[7]	= MPLL_FRACN_DEN;
+	pll_state->mpllb[8]	= mpll_fracn_quot;
+	pll_state->mpllb[9]	= mpll_fracn_rem;
+	pll_state->mpllb[10]	= HDMI_DIV(HDMI_DIV_1);
+
+	return 0;
+}
+
 static int intel_c20_phy_check_hdmi_link_rate(int clock)
 {
 	const struct intel_c20pll_state * const *tables = mtl_c20_hdmi_tables;
@@ -1911,6 +1982,9 @@ static int intel_c20_phy_check_hdmi_link_rate(int clock)
 			return MODE_OK;
 	}
 
+	if (clock >= 25175 && clock <= 594000)
+		return MODE_OK;
+
 	return MODE_CLOCK_RANGE;
 }
 
@@ -1944,6 +2018,13 @@ static int intel_c20pll_calc_state(struct intel_crtc_state *crtc_state,
 	const struct intel_c20pll_state * const *tables;
 	int i;
 
+	/* try computed C20 HDMI tables before using consolidated tables */
+	if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
+		if (intel_c20_compute_hdmi_tmds_pll(crtc_state->port_clock,
+						    &crtc_state->cx0pll_state.c20) == 0)
+			return 0;
+	}
+
 	tables = intel_c20_pll_tables_get(crtc_state, encoder);
 	if (!tables)
 		return -EINVAL;
@@ -2093,13 +2174,10 @@ static u8 intel_c20_get_dp_rate(u32 clock)
 
 static u8 intel_c20_get_hdmi_rate(u32 clock)
 {
-	switch (clock) {
-	case 25175:
-	case 27000:
-	case 74250:
-	case 148500:
-	case 594000:
+	if (clock >= 25175 && clock <= 600000)
 		return 0;
+
+	switch (clock) {
 	case 166670: /* 3 Gbps */
 	case 333330: /* 6 Gbps */
 	case 666670: /* 12 Gbps */
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index ab9d1d983b881..cb5d1be2ba191 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -218,4 +218,57 @@
 
 #define RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(idx) (0x303D + (idx))
 
+/* C20 HDMI computed pll definitions */
+#define REFCLK_38_4_MHZ		38400000
+#define CLOCK_4999MHZ		4999999999
+#define CLOCK_9999MHZ		9999999999
+#define DATARATE_3000000000	3000000000
+#define DATARATE_3500000000	3500000000
+#define DATARATE_4000000000	4000000000
+#define MPLL_FRACN_DEN		0xFFFF
+
+#define SSC_UP_SPREAD		REG_BIT16(9)
+#define WORD_CLK_DIV		REG_BIT16(8)
+
+#define MPLL_TX_CLK_DIV(val)	REG_FIELD_PREP16(C20_MPLLB_TX_CLK_DIV_MASK, val)
+#define MPLL_MULTIPLIER(val)	REG_FIELD_PREP16(C20_MULTIPLIER_MASK, val)
+
+#define MPLLB_ANA_FREQ_VCO_0	0
+#define MPLLB_ANA_FREQ_VCO_1	1
+#define MPLLB_ANA_FREQ_VCO_2	2
+#define MPLLB_ANA_FREQ_VCO_3	3
+#define MPLLB_ANA_FREQ_VCO_MASK	REG_GENMASK16(15, 14)
+#define MPLLB_ANA_FREQ_VCO(val)	REG_FIELD_PREP16(MPLLB_ANA_FREQ_VCO_MASK, val)
+
+#define MPLL_DIV_MULTIPLIER_MASK	REG_GENMASK16(7, 0)
+#define MPLL_DIV_MULTIPLIER(val)	REG_FIELD_PREP16(MPLL_DIV_MULTIPLIER_MASK, val)
+
+#define CAL_DAC_CODE_31		31
+#define CAL_DAC_CODE_MASK	REG_GENMASK16(14, 10)
+#define CAL_DAC_CODE(val)	REG_FIELD_PREP16(CAL_DAC_CODE_MASK, val)
+
+#define CP_INT_GS_28		28
+#define CP_INT_GS_MASK		REG_GENMASK16(6, 0)
+#define CP_INT_GS(val)		REG_FIELD_PREP16(CP_INT_GS_MASK, val)
+
+#define CP_PROP_GS_30		30
+#define CP_PROP_GS_MASK		REG_GENMASK16(13, 7)
+#define CP_PROP_GS(val)		REG_FIELD_PREP16(CP_PROP_GS_MASK, val)
+
+#define CP_INT_6		6
+#define CP_INT_MASK		REG_GENMASK16(6, 0)
+#define CP_INT(val)		REG_FIELD_PREP16(CP_INT_MASK, val)
+
+#define CP_PROP_20		20
+#define CP_PROP_MASK		REG_GENMASK16(13, 7)
+#define CP_PROP(val)		REG_FIELD_PREP16(CP_PROP_MASK, val)
+
+#define V2I_2			2
+#define V2I_MASK		REG_GENMASK16(15, 14)
+#define V2I(val)		REG_FIELD_PREP16(V2I_MASK, val)
+
+#define HDMI_DIV_1		1
+#define HDMI_DIV_MASK		REG_GENMASK16(2, 0)
+#define HDMI_DIV(val)		REG_FIELD_PREP16(HDMI_DIV_MASK, val)
+
 #endif /* __INTEL_CX0_REG_DEFS_H__ */
-- 
GitLab


From 3bc7fb9ac9307583f11c991ae72369499a604621 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Wed, 12 Apr 2023 00:43:08 +0200
Subject: [PATCH 0559/1791] MAINTAINERS: Add Marijn Suijten as drm/msm reviewer

As I get more and more active in the drm/msm space, yet sometimes miss
out on patches (where I was involved in previous discussions), add
myself as reviewer to make this involvement clear.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Rob Clark <robdclark@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/531770/
Link: https://lore.kernel.org/r/20230411224308.440550-1-marijn.suijten@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index dd389b88c512a..633bd47c4abab 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6514,6 +6514,7 @@ M:	Rob Clark <robdclark@gmail.com>
 M:	Abhinav Kumar <quic_abhinavk@quicinc.com>
 M:	Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
 R:	Sean Paul <sean@poorly.run>
+R:	Marijn Suijten <marijn.suijten@somainline.org>
 L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
 L:	freedreno@lists.freedesktop.org
-- 
GitLab


From 25553494b275a1a4cf06e4a7aa4073817cb2b846 Mon Sep 17 00:00:00 2001
From: Cong Liu <liucong2@kylinos.cn>
Date: Wed, 17 May 2023 13:02:03 +0800
Subject: [PATCH 0560/1791] drm/i915: Fix memory leaks in function
 live_nop_switch

Be sure to properly free the allocated memory before exiting
the live_nop_switch function.

Signed-off-by: Cong Liu <liucong2@kylinos.cn>
Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517050204.4111874-1-liucong2@kylinos.cn
---
 .../gpu/drm/i915/gem/selftests/i915_gem_context.c  | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
index 2697fbaa2cebb..ad6a3b2fb387b 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
@@ -66,7 +66,7 @@ static int live_nop_switch(void *arg)
 		ctx[n] = live_context(i915, file);
 		if (IS_ERR(ctx[n])) {
 			err = PTR_ERR(ctx[n]);
-			goto out_file;
+			goto out_ctx;
 		}
 	}
 
@@ -82,7 +82,7 @@ static int live_nop_switch(void *arg)
 			this = igt_request_alloc(ctx[n], engine);
 			if (IS_ERR(this)) {
 				err = PTR_ERR(this);
-				goto out_file;
+				goto out_ctx;
 			}
 			if (rq) {
 				i915_request_await_dma_fence(this, &rq->fence);
@@ -96,7 +96,7 @@ static int live_nop_switch(void *arg)
 			intel_gt_set_wedged(engine->gt);
 			i915_request_put(rq);
 			err = -EIO;
-			goto out_file;
+			goto out_ctx;
 		}
 		i915_request_put(rq);
 
@@ -107,7 +107,7 @@ static int live_nop_switch(void *arg)
 
 		err = igt_live_test_begin(&t, i915, __func__, engine->name);
 		if (err)
-			goto out_file;
+			goto out_ctx;
 
 		end_time = jiffies + i915_selftest.timeout_jiffies;
 		for_each_prime_number_from(prime, 2, 8192) {
@@ -120,7 +120,7 @@ static int live_nop_switch(void *arg)
 				this = igt_request_alloc(ctx[n % nctx], engine);
 				if (IS_ERR(this)) {
 					err = PTR_ERR(this);
-					goto out_file;
+					goto out_ctx;
 				}
 
 				if (rq) { /* Force submission order */
@@ -165,7 +165,7 @@ static int live_nop_switch(void *arg)
 
 		err = igt_live_test_end(&t);
 		if (err)
-			goto out_file;
+			goto out_ctx;
 
 		pr_info("Switch latencies on %s: 1 = %lluns, %lu = %lluns\n",
 			engine->name,
@@ -173,6 +173,8 @@ static int live_nop_switch(void *arg)
 			prime - 1, div64_u64(ktime_to_ns(times[1]), prime - 1));
 	}
 
+out_ctx:
+	kfree(ctx);
 out_file:
 	fput(file);
 	return err;
-- 
GitLab


From a8c4b36ecac1d71bfde8d1f64a30ce694c4fc44d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:00 +0300
Subject: [PATCH 0561/1791] drm/i915/dsc: change DSC param tables to follow the
 DSC model

After cross-checking DSC models (20150914, 20161212, 20210623) change
values in rc_parameters tables to follow config files present inside
the DSC model. Handle two places, where i915 tables diverged from the
model, by patching the rc values in the code.

Note: I left one case uncorrected, 8bpp/10bpc/range_max_qp[0], because
the table in the VESA DSC 1.1 sets it to 4.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-2-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/i915/display/intel_vdsc.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index 8e787c13d26d0..7003ae9f683ac 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -87,7 +87,7 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 		}
 	},
 	/* 6BPP/14BPC */
-	{ 768, 15, 6144, 15, 25, 23, 27, {
+	{ 768, 15, 6144, 15, 25, 23, 23, {
 		{ 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 },
 		{ 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 },
 		{ 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 },
@@ -116,6 +116,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 	},
 	/* 8BPP/10BPC */
 	{ 512, 12, 6144, 7, 16, 15, 15, {
+		/*
+		 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
+		 * VESA DSC 1.1 Table E-5 sets it to 4.
+		 */
 		{ 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
 		{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
 		{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
@@ -133,7 +137,7 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 	},
 	/* 8BPP/14BPC */
 	{ 512, 12, 6144, 15, 24, 23, 23, {
-		{ 0, 12, 0 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
+		{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
 		{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
 		{ 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 },
 		{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
@@ -598,6 +602,20 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 			DSC_RANGE_BPG_OFFSET_MASK;
 	}
 
+	if (DISPLAY_VER(dev_priv) < 13) {
+		/*
+		 * FIXME: verify that the hardware actually needs these
+		 * modifications rather than them being simple typos.
+		 */
+		if (compressed_bpp == 6 &&
+		    vdsc_cfg->bits_per_component == 8)
+			vdsc_cfg->rc_quant_incr_limit1 = 23;
+
+		if (compressed_bpp == 8 &&
+		    vdsc_cfg->bits_per_component == 14)
+			vdsc_cfg->rc_range_params[0].range_bpg_offset = 0;
+	}
+
 	/*
 	 * BitsPerComponent value determines mux_word_size:
 	 * When BitsPerComponent is less than or 10bpc, muxWordSize will be equal to
-- 
GitLab


From 169219213c144abf0f2cc86886df218159dbe4b7 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:01 +0300
Subject: [PATCH 0562/1791] drm/i915/dsc: move rc_buf_thresh values to common
 helper

The rc_buf_thresh values are common to all DSC implementations. Move
them to the common helper together with the code to propagate them to
the drm_dsc_config.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-3-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c  | 35 +++++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_vdsc.c | 24 +---------------
 include/drm/display/drm_dsc_helper.h      |  1 +
 3 files changed, 37 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index c869c6e51e2b7..be91abe2cfb28 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -270,6 +270,41 @@ void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload,
 }
 EXPORT_SYMBOL(drm_dsc_pps_payload_pack);
 
+/* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */
+static const u16 drm_dsc_rc_buf_thresh[] = {
+	896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616,
+	7744, 7872, 8000, 8064
+};
+
+/**
+ * drm_dsc_set_rc_buf_thresh() - Set thresholds for the RC model
+ * in accordance with the DSC 1.2 specification.
+ *
+ * @vdsc_cfg: DSC Configuration data partially filled by driver
+ */
+void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg)
+{
+	int i;
+
+	BUILD_BUG_ON(ARRAY_SIZE(drm_dsc_rc_buf_thresh) !=
+		     DSC_NUM_BUF_RANGES - 1);
+	BUILD_BUG_ON(ARRAY_SIZE(drm_dsc_rc_buf_thresh) !=
+		     ARRAY_SIZE(vdsc_cfg->rc_buf_thresh));
+
+	for (i = 0; i < ARRAY_SIZE(drm_dsc_rc_buf_thresh); i++)
+		vdsc_cfg->rc_buf_thresh[i] = drm_dsc_rc_buf_thresh[i] >> 6;
+
+	/*
+	 * For 6bpp, RC Buffer threshold 12 and 13 need a different value
+	 * as per C Model
+	 */
+	if (vdsc_cfg->bits_per_pixel == 6 << 4) {
+		vdsc_cfg->rc_buf_thresh[12] = 7936 >> 6;
+		vdsc_cfg->rc_buf_thresh[13] = 8000 >> 6;
+	}
+}
+EXPORT_SYMBOL(drm_dsc_set_rc_buf_thresh);
+
 /**
  * drm_dsc_compute_rc_parameters() - Write rate control
  * parameters to the dsc configuration defined in
diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index 7003ae9f683ac..2fd08375bbe30 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -37,12 +37,6 @@ enum COLUMN_INDEX_BPC {
 	MAX_COLUMN_INDEX
 };
 
-/* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */
-static const u16 rc_buf_thresh[] = {
-	896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616,
-	7744, 7872, 8000, 8064
-};
-
 struct rc_parameters {
 	u16 initial_xmit_delay;
 	u8 first_line_bpg_offset;
@@ -543,23 +537,7 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 
 	vdsc_cfg->bits_per_component = pipe_config->pipe_bpp / 3;
 
-	for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) {
-		/*
-		 * six 0s are appended to the lsb of each threshold value
-		 * internally in h/w.
-		 * Only 8 bits are allowed for programming RcBufThreshold
-		 */
-		vdsc_cfg->rc_buf_thresh[i] = rc_buf_thresh[i] >> 6;
-	}
-
-	/*
-	 * For 6bpp, RC Buffer threshold 12 and 13 need a different value
-	 * as per C Model
-	 */
-	if (compressed_bpp == 6) {
-		vdsc_cfg->rc_buf_thresh[12] = 0x7C;
-		vdsc_cfg->rc_buf_thresh[13] = 0x7D;
-	}
+	drm_dsc_set_rc_buf_thresh(vdsc_cfg);
 
 	/*
 	 * From XE_LPD onwards we supports compression bpps in steps of 1
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index 8b41edbbabab7..706ba1d347427 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -14,6 +14,7 @@ void drm_dsc_dp_pps_header_init(struct dp_sdp_header *pps_header);
 int drm_dsc_dp_rc_buffer_size(u8 rc_buffer_block_size, u8 rc_buffer_size);
 void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_sdp,
 			      const struct drm_dsc_config *dsc_cfg);
+void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg);
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
 
 #endif /* _DRM_DSC_HELPER_H_ */
-- 
GitLab


From 2b470e5531f57c1b9bfa129cca0ee17a2ecd2183 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:02 +0300
Subject: [PATCH 0563/1791] drm/i915/dsc: move DSC tables to DRM DSC helper

Move DSC RC tables to DRM DSC helper. No additional code changes
and/or cleanups are a part of this commit, it will be cleaned up in the
followup commits.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-4-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c  | 372 ++++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_vdsc.c | 319 +------------------
 include/drm/display/drm_dsc_helper.h      |   1 +
 3 files changed, 380 insertions(+), 312 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index be91abe2cfb28..122a292bbc8fb 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -305,6 +305,378 @@ void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg)
 }
 EXPORT_SYMBOL(drm_dsc_set_rc_buf_thresh);
 
+enum ROW_INDEX_BPP {
+	ROW_INDEX_6BPP = 0,
+	ROW_INDEX_8BPP,
+	ROW_INDEX_10BPP,
+	ROW_INDEX_12BPP,
+	ROW_INDEX_15BPP,
+	MAX_ROW_INDEX
+};
+
+enum COLUMN_INDEX_BPC {
+	COLUMN_INDEX_8BPC = 0,
+	COLUMN_INDEX_10BPC,
+	COLUMN_INDEX_12BPC,
+	COLUMN_INDEX_14BPC,
+	COLUMN_INDEX_16BPC,
+	MAX_COLUMN_INDEX
+};
+
+struct rc_parameters {
+	u16 initial_xmit_delay;
+	u8 first_line_bpg_offset;
+	u16 initial_offset;
+	u8 flatness_min_qp;
+	u8 flatness_max_qp;
+	u8 rc_quant_incr_limit0;
+	u8 rc_quant_incr_limit1;
+	struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES];
+};
+
+/*
+ * Selected Rate Control Related Parameter Recommended Values
+ * from DSC_v1.11 spec & C Model release: DSC_model_20161212
+ */
+static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
+	{
+		/* 6BPP/8BPC */
+		{ 768, 15, 6144, 3, 13, 11, 11, {
+			{ 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 },
+			{ 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 },
+			{ 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 },
+			{ 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 }
+			}
+		},
+		/* 6BPP/10BPC */
+		{ 768, 15, 6144, 7, 17, 15, 15, {
+			{ 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 },
+			{ 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 },
+			{ 11, 15, -8 }, { 12, 16, -10 }, { 13, 16, -10 },
+			{ 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 },
+			{ 17, 18, -12 }
+			}
+		},
+		/* 6BPP/12BPC */
+		{ 768, 15, 6144, 11, 21, 19, 19, {
+			{ 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 },
+			{ 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 },
+			{ 15, 19, -8 }, { 16, 20, -10 }, { 17, 20, -10 },
+			{ 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 },
+			{ 21, 22, -12 }
+			}
+		},
+		/* 6BPP/14BPC */
+		{ 768, 15, 6144, 15, 25, 23, 23, {
+			{ 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 },
+			{ 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 },
+			{ 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 },
+			{ 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 },
+			{ 25, 26, -12 }
+			}
+		},
+		/* 6BPP/16BPC */
+		{ 768, 15, 6144, 19, 29, 27, 27, {
+			{ 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 },
+			{ 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 },
+			{ 23, 27, -8 }, { 24, 28, -10 }, { 25, 28, -10 },
+			{ 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 },
+			{ 29, 30, -12 }
+			}
+		},
+	},
+	{
+		/* 8BPP/8BPC */
+		{ 512, 12, 6144, 3, 12, 11, 11, {
+			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 },
+			{ 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		},
+		/* 8BPP/10BPC */
+		{ 512, 12, 6144, 7, 16, 15, 15, {
+			/*
+			 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
+			 * VESA DSC 1.1 Table E-5 sets it to 4.
+			 */
+			{ 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
+			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		},
+		/* 8BPP/12BPC */
+		{ 512, 12, 6144, 11, 20, 19, 19, {
+			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
+			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		},
+		/* 8BPP/14BPC */
+		{ 512, 12, 6144, 15, 24, 23, 23, {
+			{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 },
+			{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		},
+		/* 8BPP/16BPC */
+		{ 512, 12, 6144, 19, 28, 27, 27, {
+			{ 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 },
+			{ 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 },
+			{ 28, 29, -12 }
+			}
+		},
+	},
+	{
+		/* 10BPP/8BPC */
+		{ 410, 15, 5632, 3, 12, 11, 11, {
+			{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 },
+			{ 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 }
+			}
+		},
+		/* 10BPP/10BPC */
+		{ 410, 15, 5632, 7, 16, 15, 15, {
+			{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 },
+			{ 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 }
+			}
+		},
+		/* 10BPP/12BPC */
+		{ 410, 15, 5632, 11, 20, 19, 19, {
+			{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 },
+			{ 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 },
+			{ 19, 20, -12 }
+			}
+		},
+		/* 10BPP/14BPC */
+		{ 410, 15, 5632, 15, 24, 23, 23, {
+			{ 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 },
+			{ 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 },
+			{ 23, 24, -12 }
+			}
+		},
+		/* 10BPP/16BPC */
+		{ 410, 15, 5632, 19, 28, 27, 27, {
+			{ 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 },
+			{ 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 },
+			{ 27, 28, -12 }
+			}
+		},
+	},
+	{
+		/* 12BPP/8BPC */
+		{ 341, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 },
+			{ 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		},
+		/* 12BPP/10BPC */
+		{ 341, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
+			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
+			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		},
+		/* 12BPP/12BPC */
+		{ 341, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
+			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
+			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		},
+		/* 12BPP/14BPC */
+		{ 341, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 },
+			{ 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 },
+			{ 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 },
+			{ 22, 23, -12 }
+			}
+		},
+		/* 12BPP/16BPC */
+		{ 341, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 },
+			{ 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 },
+			{ 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 },
+			{ 26, 27, -12 }
+			}
+		},
+	},
+	{
+		/* 15BPP/8BPC */
+		{ 273, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
+			{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 },
+			{ 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 },
+			{ 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 }
+			}
+		},
+		/* 15BPP/10BPC */
+		{ 273, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
+			{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 },
+			{ 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 },
+			{ 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 }
+			}
+		},
+		/* 15BPP/12BPC */
+		{ 273, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
+			{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
+			{ 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 },
+			{ 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 },
+			{ 16, 17, -12 }
+			}
+		},
+		/* 15BPP/14BPC */
+		{ 273, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 },
+			{ 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 },
+			{ 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 },
+			{ 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 },
+			{ 20, 21, -12 }
+			}
+		},
+		/* 15BPP/16BPC */
+		{ 273, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 },
+			{ 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 },
+			{ 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 },
+			{ 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		}
+	}
+};
+
+static int get_row_index_for_rc_params(u16 compressed_bpp)
+{
+	switch (compressed_bpp) {
+	case 6:
+		return ROW_INDEX_6BPP;
+	case 8:
+		return ROW_INDEX_8BPP;
+	case 10:
+		return ROW_INDEX_10BPP;
+	case 12:
+		return ROW_INDEX_12BPP;
+	case 15:
+		return ROW_INDEX_15BPP;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int get_column_index_for_rc_params(u8 bits_per_component)
+{
+	switch (bits_per_component) {
+	case 8:
+		return COLUMN_INDEX_8BPC;
+	case 10:
+		return COLUMN_INDEX_10BPC;
+	case 12:
+		return COLUMN_INDEX_12BPC;
+	case 14:
+		return COLUMN_INDEX_14BPC;
+	case 16:
+		return COLUMN_INDEX_16BPC;
+	default:
+		return -EINVAL;
+	}
+}
+
+static const struct rc_parameters *get_rc_params(u16 compressed_bpp,
+						 u8 bits_per_component)
+{
+	int row_index, column_index;
+
+	row_index = get_row_index_for_rc_params(compressed_bpp);
+	if (row_index < 0)
+		return NULL;
+
+	column_index = get_column_index_for_rc_params(bits_per_component);
+	if (column_index < 0)
+		return NULL;
+
+	return &rc_parameters[row_index][column_index];
+}
+
+/**
+ * drm_dsc_setup_rc_params() - Set parameters and limits for RC model in
+ * accordance with the DSC 1.1 or 1.2 specification and DSC C Model
+ * Required bits_per_pixel and bits_per_component to be set before calling this
+ * function.
+ *
+ * @vdsc_cfg: DSC Configuration data partially filled by driver
+ *
+ * Return: 0 or -error code in case of an error
+ */
+int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg)
+{
+	const struct rc_parameters *rc_params;
+	int i;
+
+	if (WARN_ON_ONCE(!vdsc_cfg->bits_per_pixel ||
+			 !vdsc_cfg->bits_per_component))
+		return -EINVAL;
+
+	/* fractional BPP is not supported */
+	if (vdsc_cfg->bits_per_pixel & 0xf)
+		return -EINVAL;
+
+	rc_params = get_rc_params(vdsc_cfg->bits_per_pixel >> 4,
+				  vdsc_cfg->bits_per_component);
+	if (!rc_params)
+		return -EINVAL;
+
+	vdsc_cfg->first_line_bpg_offset = rc_params->first_line_bpg_offset;
+	vdsc_cfg->initial_xmit_delay = rc_params->initial_xmit_delay;
+	vdsc_cfg->initial_offset = rc_params->initial_offset;
+	vdsc_cfg->flatness_min_qp = rc_params->flatness_min_qp;
+	vdsc_cfg->flatness_max_qp = rc_params->flatness_max_qp;
+	vdsc_cfg->rc_quant_incr_limit0 = rc_params->rc_quant_incr_limit0;
+	vdsc_cfg->rc_quant_incr_limit1 = rc_params->rc_quant_incr_limit1;
+
+	for (i = 0; i < DSC_NUM_BUF_RANGES; i++) {
+		vdsc_cfg->rc_range_params[i].range_min_qp =
+			rc_params->rc_range_params[i].range_min_qp;
+		vdsc_cfg->rc_range_params[i].range_max_qp =
+			rc_params->rc_range_params[i].range_max_qp;
+		/*
+		 * Range BPG Offset uses 2's complement and is only a 6 bits. So
+		 * mask it to get only 6 bits.
+		 */
+		vdsc_cfg->rc_range_params[i].range_bpg_offset =
+			rc_params->rc_range_params[i].range_bpg_offset &
+			DSC_RANGE_BPG_OFFSET_MASK;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_dsc_setup_rc_params);
+
 /**
  * drm_dsc_compute_rc_parameters() - Write rate control
  * parameters to the dsc configuration defined in
diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index 2fd08375bbe30..d0536582e4b9a 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -19,24 +19,6 @@
 #include "intel_vdsc.h"
 #include "intel_vdsc_regs.h"
 
-enum ROW_INDEX_BPP {
-	ROW_INDEX_6BPP = 0,
-	ROW_INDEX_8BPP,
-	ROW_INDEX_10BPP,
-	ROW_INDEX_12BPP,
-	ROW_INDEX_15BPP,
-	MAX_ROW_INDEX
-};
-
-enum COLUMN_INDEX_BPC {
-	COLUMN_INDEX_8BPC = 0,
-	COLUMN_INDEX_10BPC,
-	COLUMN_INDEX_12BPC,
-	COLUMN_INDEX_14BPC,
-	COLUMN_INDEX_16BPC,
-	MAX_COLUMN_INDEX
-};
-
 struct rc_parameters {
 	u16 initial_xmit_delay;
 	u8 first_line_bpg_offset;
@@ -48,296 +30,6 @@ struct rc_parameters {
 	struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES];
 };
 
-/*
- * Selected Rate Control Related Parameter Recommended Values
- * from DSC_v1.11 spec & C Model release: DSC_model_20161212
- */
-static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
-{
-	/* 6BPP/8BPC */
-	{ 768, 15, 6144, 3, 13, 11, 11, {
-		{ 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 },
-		{ 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 },
-		{ 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 },
-		{ 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 }
-		}
-	},
-	/* 6BPP/10BPC */
-	{ 768, 15, 6144, 7, 17, 15, 15, {
-		{ 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 },
-		{ 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 },
-		{ 11, 15, -8 }, { 12, 16, -10 }, { 13, 16, -10 },
-		{ 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 },
-		{ 17, 18, -12 }
-		}
-	},
-	/* 6BPP/12BPC */
-	{ 768, 15, 6144, 11, 21, 19, 19, {
-		{ 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 },
-		{ 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 },
-		{ 15, 19, -8 }, { 16, 20, -10 }, { 17, 20, -10 },
-		{ 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 },
-		{ 21, 22, -12 }
-		}
-	},
-	/* 6BPP/14BPC */
-	{ 768, 15, 6144, 15, 25, 23, 23, {
-		{ 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 },
-		{ 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 },
-		{ 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 },
-		{ 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 },
-		{ 25, 26, -12 }
-		}
-	},
-	/* 6BPP/16BPC */
-	{ 768, 15, 6144, 19, 29, 27, 27, {
-		{ 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 },
-		{ 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 },
-		{ 23, 27, -8 }, { 24, 28, -10 }, { 25, 28, -10 },
-		{ 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 },
-		{ 29, 30, -12 }
-		}
-	},
-},
-{
-	/* 8BPP/8BPC */
-	{ 512, 12, 6144, 3, 12, 11, 11, {
-		{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
-		{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
-		{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 },
-		{ 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
-		}
-	},
-	/* 8BPP/10BPC */
-	{ 512, 12, 6144, 7, 16, 15, 15, {
-		/*
-		 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
-		 * VESA DSC 1.1 Table E-5 sets it to 4.
-		 */
-		{ 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
-		{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
-		{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
-		{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
-		}
-	},
-	/* 8BPP/12BPC */
-	{ 512, 12, 6144, 11, 20, 19, 19, {
-		{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
-		{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
-		{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
-		{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
-		{ 21, 23, -12 }
-		}
-	},
-	/* 8BPP/14BPC */
-	{ 512, 12, 6144, 15, 24, 23, 23, {
-		{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
-		{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
-		{ 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 },
-		{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
-		{ 24, 25, -12 }
-		}
-	},
-	/* 8BPP/16BPC */
-	{ 512, 12, 6144, 19, 28, 27, 27, {
-		{ 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 },
-		{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
-		{ 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 },
-		{ 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 },
-		{ 28, 29, -12 }
-		}
-	},
-},
-{
-	/* 10BPP/8BPC */
-	{ 410, 15, 5632, 3, 12, 11, 11, {
-		{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
-		{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
-		{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 },
-		{ 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 }
-		}
-	},
-	/* 10BPP/10BPC */
-	{ 410, 15, 5632, 7, 16, 15, 15, {
-		{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
-		{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
-		{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 },
-		{ 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 }
-		}
-	},
-	/* 10BPP/12BPC */
-	{ 410, 15, 5632, 11, 20, 19, 19, {
-		{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
-		{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
-		{ 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 },
-		{ 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 },
-		{ 19, 20, -12 }
-		}
-	},
-	/* 10BPP/14BPC */
-	{ 410, 15, 5632, 15, 24, 23, 23, {
-		{ 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 },
-		{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
-		{ 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 },
-		{ 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 },
-		{ 23, 24, -12 }
-		}
-	},
-	/* 10BPP/16BPC */
-	{ 410, 15, 5632, 19, 28, 27, 27, {
-		{ 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 },
-		{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
-		{ 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 },
-		{ 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 },
-		{ 27, 28, -12 }
-		}
-	},
-},
-{
-	/* 12BPP/8BPC */
-	{ 341, 15, 2048, 3, 12, 11, 11, {
-		{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
-		{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
-		{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 },
-		{ 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
-		}
-	},
-	/* 12BPP/10BPC */
-	{ 341, 15, 2048, 7, 16, 15, 15, {
-		{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
-		{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
-		{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
-		{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
-		}
-	},
-	/* 12BPP/12BPC */
-	{ 341, 15, 2048, 11, 20, 19, 19, {
-		{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
-		{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
-		{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
-		{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
-		{ 21, 23, -12 }
-		}
-	},
-	/* 12BPP/14BPC */
-	{ 341, 15, 2048, 15, 24, 23, 23, {
-		{ 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 },
-		{ 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
-		{ 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 },
-		{ 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 },
-		{ 22, 23, -12 }
-		}
-	},
-	/* 12BPP/16BPC */
-	{ 341, 15, 2048, 19, 28, 27, 27, {
-		{ 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 },
-		{ 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
-		{ 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 },
-		{ 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 },
-		{ 26, 27, -12 }
-		}
-	},
-},
-{
-	/* 15BPP/8BPC */
-	{ 273, 15, 2048, 3, 12, 11, 11, {
-		{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
-		{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 },
-		{ 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 },
-		{ 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 }
-		}
-	},
-	/* 15BPP/10BPC */
-	{ 273, 15, 2048, 7, 16, 15, 15, {
-		{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
-		{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 },
-		{ 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 },
-		{ 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 }
-		}
-	},
-	/* 15BPP/12BPC */
-	{ 273, 15, 2048, 11, 20, 19, 19, {
-		{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
-		{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
-		{ 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 },
-		{ 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 },
-		{ 16, 17, -12 }
-		}
-	},
-	/* 15BPP/14BPC */
-	{ 273, 15, 2048, 15, 24, 23, 23, {
-		{ 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 },
-		{ 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 },
-		{ 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 },
-		{ 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 },
-		{ 20, 21, -12 }
-		}
-	},
-	/* 15BPP/16BPC */
-	{ 273, 15, 2048, 19, 28, 27, 27, {
-		{ 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 },
-		{ 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 },
-		{ 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 },
-		{ 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 },
-		{ 24, 25, -12 }
-		}
-	}
-}
-
-};
-
-static int get_row_index_for_rc_params(u16 compressed_bpp)
-{
-	switch (compressed_bpp) {
-	case 6:
-		return ROW_INDEX_6BPP;
-	case 8:
-		return ROW_INDEX_8BPP;
-	case 10:
-		return ROW_INDEX_10BPP;
-	case 12:
-		return ROW_INDEX_12BPP;
-	case 15:
-		return ROW_INDEX_15BPP;
-	default:
-		return -EINVAL;
-	}
-}
-
-static int get_column_index_for_rc_params(u8 bits_per_component)
-{
-	switch (bits_per_component) {
-	case 8:
-		return COLUMN_INDEX_8BPC;
-	case 10:
-		return COLUMN_INDEX_10BPC;
-	case 12:
-		return COLUMN_INDEX_12BPC;
-	case 14:
-		return COLUMN_INDEX_14BPC;
-	case 16:
-		return COLUMN_INDEX_16BPC;
-	default:
-		return -EINVAL;
-	}
-}
-
-static const struct rc_parameters *get_rc_params(u16 compressed_bpp,
-						 u8 bits_per_component)
-{
-	int row_index, column_index;
-
-	row_index = get_row_index_for_rc_params(compressed_bpp);
-	if (row_index < 0)
-		return NULL;
-
-	column_index = get_column_index_for_rc_params(bits_per_component);
-	if (column_index < 0)
-		return NULL;
-
-	return &rc_parameters[row_index][column_index];
-}
-
 bool intel_dsc_source_support(const struct intel_crtc_state *crtc_state)
 {
 	const struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
@@ -479,6 +171,7 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	struct rc_parameters *rc = NULL;
 	int err;
 	u8 i = 0;
+	int ret;
 
 	vdsc_cfg->pic_width = pipe_config->hw.adjusted_mode.crtc_hdisplay;
 	vdsc_cfg->slice_width = DIV_ROUND_UP(vdsc_cfg->pic_width,
@@ -552,10 +245,11 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 		calculate_rc_params(rc, vdsc_cfg);
 		rc_params = rc;
 	} else {
-		rc_params = get_rc_params(compressed_bpp,
-					  vdsc_cfg->bits_per_component);
-		if (!rc_params)
-			return -EINVAL;
+		ret = drm_dsc_setup_rc_params(vdsc_cfg);
+		if (ret)
+			return ret;
+
+		goto out;
 	}
 
 	vdsc_cfg->first_line_bpg_offset = rc_params->first_line_bpg_offset;
@@ -594,6 +288,7 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 			vdsc_cfg->rc_range_params[0].range_bpg_offset = 0;
 	}
 
+out:
 	/*
 	 * BitsPerComponent value determines mux_word_size:
 	 * When BitsPerComponent is less than or 10bpc, muxWordSize will be equal to
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index 706ba1d347427..1681791f65a55 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -15,6 +15,7 @@ int drm_dsc_dp_rc_buffer_size(u8 rc_buffer_block_size, u8 rc_buffer_size);
 void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_sdp,
 			      const struct drm_dsc_config *dsc_cfg);
 void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg);
+int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg);
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
 
 #endif /* _DRM_DSC_HELPER_H_ */
-- 
GitLab


From 9ff6b5256de3843ee6a2ee91737554d96852963d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:03 +0300
Subject: [PATCH 0564/1791] drm/i915/dsc: stop using interim structure for
 calculated params

Stop using an interim structure rc_parameters for storing calculated
params and then setting drm_dsc_config using that structure. Instead put
calculated params into the struct drm_dsc_config directly.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-5-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/i915/display/intel_vdsc.c | 100 ++++++----------------
 1 file changed, 26 insertions(+), 74 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index d0536582e4b9a..d4340b18c18d1 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -19,17 +19,6 @@
 #include "intel_vdsc.h"
 #include "intel_vdsc_regs.h"
 
-struct rc_parameters {
-	u16 initial_xmit_delay;
-	u8 first_line_bpg_offset;
-	u16 initial_offset;
-	u8 flatness_min_qp;
-	u8 flatness_max_qp;
-	u8 rc_quant_incr_limit0;
-	u8 rc_quant_incr_limit1;
-	struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES];
-};
-
 bool intel_dsc_source_support(const struct intel_crtc_state *crtc_state)
 {
 	const struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
@@ -64,8 +53,7 @@ static bool is_pipe_dsc(struct intel_crtc *crtc, enum transcoder cpu_transcoder)
 }
 
 static void
-calculate_rc_params(struct rc_parameters *rc,
-		    struct drm_dsc_config *vdsc_cfg)
+calculate_rc_params(struct drm_dsc_config *vdsc_cfg)
 {
 	int bpc = vdsc_cfg->bits_per_component;
 	int bpp = vdsc_cfg->bits_per_pixel >> 4;
@@ -85,56 +73,57 @@ calculate_rc_params(struct rc_parameters *rc,
 	u32 res, buf_i, bpp_i;
 
 	if (vdsc_cfg->slice_height >= 8)
-		rc->first_line_bpg_offset =
+		vdsc_cfg->first_line_bpg_offset =
 			12 + DIV_ROUND_UP((9 * min(34, vdsc_cfg->slice_height - 8)), 100);
 	else
-		rc->first_line_bpg_offset = 2 * (vdsc_cfg->slice_height - 1);
+		vdsc_cfg->first_line_bpg_offset = 2 * (vdsc_cfg->slice_height - 1);
 
 	/* Our hw supports only 444 modes as of today */
 	if (bpp >= 12)
-		rc->initial_offset = 2048;
+		vdsc_cfg->initial_offset = 2048;
 	else if (bpp >= 10)
-		rc->initial_offset = 5632 - DIV_ROUND_UP(((bpp - 10) * 3584), 2);
+		vdsc_cfg->initial_offset = 5632 - DIV_ROUND_UP(((bpp - 10) * 3584), 2);
 	else if (bpp >= 8)
-		rc->initial_offset = 6144 - DIV_ROUND_UP(((bpp - 8) * 512), 2);
+		vdsc_cfg->initial_offset = 6144 - DIV_ROUND_UP(((bpp - 8) * 512), 2);
 	else
-		rc->initial_offset = 6144;
+		vdsc_cfg->initial_offset = 6144;
 
 	/* initial_xmit_delay = rc_model_size/2/compression_bpp */
-	rc->initial_xmit_delay = DIV_ROUND_UP(DSC_RC_MODEL_SIZE_CONST, 2 * bpp);
+	vdsc_cfg->initial_xmit_delay = DIV_ROUND_UP(DSC_RC_MODEL_SIZE_CONST, 2 * bpp);
 
-	rc->flatness_min_qp = 3 + qp_bpc_modifier;
-	rc->flatness_max_qp = 12 + qp_bpc_modifier;
+	vdsc_cfg->flatness_min_qp = 3 + qp_bpc_modifier;
+	vdsc_cfg->flatness_max_qp = 12 + qp_bpc_modifier;
 
-	rc->rc_quant_incr_limit0 = 11 + qp_bpc_modifier;
-	rc->rc_quant_incr_limit1 = 11 + qp_bpc_modifier;
+	vdsc_cfg->rc_quant_incr_limit0 = 11 + qp_bpc_modifier;
+	vdsc_cfg->rc_quant_incr_limit1 = 11 + qp_bpc_modifier;
 
 	bpp_i  = (2 * (bpp - 6));
 	for (buf_i = 0; buf_i < DSC_NUM_BUF_RANGES; buf_i++) {
+		u8 range_bpg_offset;
+
 		/* Read range_minqp and range_max_qp from qp tables */
-		rc->rc_range_params[buf_i].range_min_qp =
+		vdsc_cfg->rc_range_params[buf_i].range_min_qp =
 			intel_lookup_range_min_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420);
-		rc->rc_range_params[buf_i].range_max_qp =
+		vdsc_cfg->rc_range_params[buf_i].range_max_qp =
 			intel_lookup_range_max_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420);
 
-		/* Calculate range_bgp_offset */
+		/* Calculate range_bpg_offset */
 		if (bpp <= 6) {
-			rc->rc_range_params[buf_i].range_bpg_offset = ofs_und6[buf_i];
+			range_bpg_offset = ofs_und6[buf_i];
 		} else if (bpp <= 8) {
 			res = DIV_ROUND_UP(((bpp - 6) * (ofs_und8[buf_i] - ofs_und6[buf_i])), 2);
-			rc->rc_range_params[buf_i].range_bpg_offset =
-								ofs_und6[buf_i] + res;
+			range_bpg_offset = ofs_und6[buf_i] + res;
 		} else if (bpp <= 12) {
-			rc->rc_range_params[buf_i].range_bpg_offset =
-								ofs_und8[buf_i];
+			range_bpg_offset = ofs_und8[buf_i];
 		} else if (bpp <= 15) {
 			res = DIV_ROUND_UP(((bpp - 12) * (ofs_und15[buf_i] - ofs_und12[buf_i])), 3);
-			rc->rc_range_params[buf_i].range_bpg_offset =
-								ofs_und12[buf_i] + res;
+			range_bpg_offset = ofs_und12[buf_i] + res;
 		} else {
-			rc->rc_range_params[buf_i].range_bpg_offset =
-								ofs_und15[buf_i];
+			range_bpg_offset = ofs_und15[buf_i];
 		}
+
+		vdsc_cfg->rc_range_params[buf_i].range_bpg_offset =
+			range_bpg_offset & DSC_RANGE_BPG_OFFSET_MASK;
 	}
 }
 
@@ -167,10 +156,7 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	struct drm_dsc_config *vdsc_cfg = &pipe_config->dsc.config;
 	u16 compressed_bpp = pipe_config->dsc.compressed_bpp;
-	const struct rc_parameters *rc_params;
-	struct rc_parameters *rc = NULL;
 	int err;
-	u8 i = 0;
 	int ret;
 
 	vdsc_cfg->pic_width = pipe_config->hw.adjusted_mode.crtc_hdisplay;
@@ -238,43 +224,12 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	 * parameters
 	 */
 	if (DISPLAY_VER(dev_priv) >= 13) {
-		rc = kmalloc(sizeof(*rc), GFP_KERNEL);
-		if (!rc)
-			return -ENOMEM;
-
-		calculate_rc_params(rc, vdsc_cfg);
-		rc_params = rc;
+		calculate_rc_params(vdsc_cfg);
 	} else {
 		ret = drm_dsc_setup_rc_params(vdsc_cfg);
 		if (ret)
 			return ret;
 
-		goto out;
-	}
-
-	vdsc_cfg->first_line_bpg_offset = rc_params->first_line_bpg_offset;
-	vdsc_cfg->initial_xmit_delay = rc_params->initial_xmit_delay;
-	vdsc_cfg->initial_offset = rc_params->initial_offset;
-	vdsc_cfg->flatness_min_qp = rc_params->flatness_min_qp;
-	vdsc_cfg->flatness_max_qp = rc_params->flatness_max_qp;
-	vdsc_cfg->rc_quant_incr_limit0 = rc_params->rc_quant_incr_limit0;
-	vdsc_cfg->rc_quant_incr_limit1 = rc_params->rc_quant_incr_limit1;
-
-	for (i = 0; i < DSC_NUM_BUF_RANGES; i++) {
-		vdsc_cfg->rc_range_params[i].range_min_qp =
-			rc_params->rc_range_params[i].range_min_qp;
-		vdsc_cfg->rc_range_params[i].range_max_qp =
-			rc_params->rc_range_params[i].range_max_qp;
-		/*
-		 * Range BPG Offset uses 2's complement and is only a 6 bits. So
-		 * mask it to get only 6 bits.
-		 */
-		vdsc_cfg->rc_range_params[i].range_bpg_offset =
-			rc_params->rc_range_params[i].range_bpg_offset &
-			DSC_RANGE_BPG_OFFSET_MASK;
-	}
-
-	if (DISPLAY_VER(dev_priv) < 13) {
 		/*
 		 * FIXME: verify that the hardware actually needs these
 		 * modifications rather than them being simple typos.
@@ -288,7 +243,6 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 			vdsc_cfg->rc_range_params[0].range_bpg_offset = 0;
 	}
 
-out:
 	/*
 	 * BitsPerComponent value determines mux_word_size:
 	 * When BitsPerComponent is less than or 10bpc, muxWordSize will be equal to
@@ -303,8 +257,6 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	vdsc_cfg->initial_scale_value = (vdsc_cfg->rc_model_size << 3) /
 		(vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset);
 
-	kfree(rc);
-
 	return 0;
 }
 
-- 
GitLab


From 74fe874cf1407d4731f0a2c713d31ac78a5b32bd Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:04 +0300
Subject: [PATCH 0565/1791] drm/display/dsc: use flat array for rc_parameters
 lookup

Next commits are going to add support for additional RC parameter lookup
tables. These tables are going to use different bpp/bpc combinations,
thus it makes little sense to keep the 2d array for RC parameters.
Switch to using the flat array.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-6-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 228 +++++++++++------------
 1 file changed, 108 insertions(+), 120 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index 122a292bbc8fb..acb93d4116e0c 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -305,24 +305,6 @@ void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg)
 }
 EXPORT_SYMBOL(drm_dsc_set_rc_buf_thresh);
 
-enum ROW_INDEX_BPP {
-	ROW_INDEX_6BPP = 0,
-	ROW_INDEX_8BPP,
-	ROW_INDEX_10BPP,
-	ROW_INDEX_12BPP,
-	ROW_INDEX_15BPP,
-	MAX_ROW_INDEX
-};
-
-enum COLUMN_INDEX_BPC {
-	COLUMN_INDEX_8BPC = 0,
-	COLUMN_INDEX_10BPC,
-	COLUMN_INDEX_12BPC,
-	COLUMN_INDEX_14BPC,
-	COLUMN_INDEX_16BPC,
-	MAX_COLUMN_INDEX
-};
-
 struct rc_parameters {
 	u16 initial_xmit_delay;
 	u8 first_line_bpg_offset;
@@ -334,21 +316,31 @@ struct rc_parameters {
 	struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES];
 };
 
+struct rc_parameters_data {
+	u8 bpp;
+	u8 bpc;
+	struct rc_parameters params;
+};
+
+#define DSC_BPP(bpp)	((bpp) << 4)
+
 /*
  * Selected Rate Control Related Parameter Recommended Values
  * from DSC_v1.11 spec & C Model release: DSC_model_20161212
  */
-static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
+static const struct rc_parameters_data rc_parameters[] = {
 	{
-		/* 6BPP/8BPC */
+		.bpp = DSC_BPP(6), .bpc = 8,
 		{ 768, 15, 6144, 3, 13, 11, 11, {
 			{ 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 },
 			{ 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 },
 			{ 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 },
 			{ 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 }
 			}
-		},
-		/* 6BPP/10BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 10,
 		{ 768, 15, 6144, 7, 17, 15, 15, {
 			{ 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 },
 			{ 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 },
@@ -356,8 +348,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 },
 			{ 17, 18, -12 }
 			}
-		},
-		/* 6BPP/12BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 12,
 		{ 768, 15, 6144, 11, 21, 19, 19, {
 			{ 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 },
 			{ 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 },
@@ -365,8 +359,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 },
 			{ 21, 22, -12 }
 			}
-		},
-		/* 6BPP/14BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 14,
 		{ 768, 15, 6144, 15, 25, 23, 23, {
 			{ 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 },
 			{ 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 },
@@ -374,8 +370,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 },
 			{ 25, 26, -12 }
 			}
-		},
-		/* 6BPP/16BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 16,
 		{ 768, 15, 6144, 19, 29, 27, 27, {
 			{ 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 },
 			{ 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 },
@@ -383,18 +381,20 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 },
 			{ 29, 30, -12 }
 			}
-		},
+		}
 	},
 	{
-		/* 8BPP/8BPC */
+		.bpp = DSC_BPP(8), .bpc = 8,
 		{ 512, 12, 6144, 3, 12, 11, 11, {
 			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
 			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
 			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 },
 			{ 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
 			}
-		},
-		/* 8BPP/10BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 10,
 		{ 512, 12, 6144, 7, 16, 15, 15, {
 			/*
 			 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
@@ -405,8 +405,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
 			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
 			}
-		},
-		/* 8BPP/12BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 12,
 		{ 512, 12, 6144, 11, 20, 19, 19, {
 			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
 			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
@@ -414,8 +416,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
 			{ 21, 23, -12 }
 			}
-		},
-		/* 8BPP/14BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 14,
 		{ 512, 12, 6144, 15, 24, 23, 23, {
 			{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
 			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
@@ -423,8 +427,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
 			{ 24, 25, -12 }
 			}
-		},
-		/* 8BPP/16BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 16,
 		{ 512, 12, 6144, 19, 28, 27, 27, {
 			{ 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 },
 			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
@@ -432,26 +438,30 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 },
 			{ 28, 29, -12 }
 			}
-		},
+		}
 	},
 	{
-		/* 10BPP/8BPC */
+		.bpp = DSC_BPP(10), .bpc = 8,
 		{ 410, 15, 5632, 3, 12, 11, 11, {
 			{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
 			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
 			{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 },
 			{ 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 }
 			}
-		},
-		/* 10BPP/10BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 10,
 		{ 410, 15, 5632, 7, 16, 15, 15, {
 			{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
 			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
 			{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 },
 			{ 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 }
 			}
-		},
-		/* 10BPP/12BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 12,
 		{ 410, 15, 5632, 11, 20, 19, 19, {
 			{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
 			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
@@ -459,8 +469,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 },
 			{ 19, 20, -12 }
 			}
-		},
-		/* 10BPP/14BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 14,
 		{ 410, 15, 5632, 15, 24, 23, 23, {
 			{ 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 },
 			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
@@ -468,8 +480,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 },
 			{ 23, 24, -12 }
 			}
-		},
-		/* 10BPP/16BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 16,
 		{ 410, 15, 5632, 19, 28, 27, 27, {
 			{ 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 },
 			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
@@ -477,26 +491,30 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 },
 			{ 27, 28, -12 }
 			}
-		},
+		}
 	},
 	{
-		/* 12BPP/8BPC */
+		.bpp = DSC_BPP(12), .bpc = 8,
 		{ 341, 15, 2048, 3, 12, 11, 11, {
 			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
 			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
 			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 },
 			{ 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
 			}
-		},
-		/* 12BPP/10BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 10,
 		{ 341, 15, 2048, 7, 16, 15, 15, {
 			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
 			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
 			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
 			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
 			}
-		},
-		/* 12BPP/12BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 12,
 		{ 341, 15, 2048, 11, 20, 19, 19, {
 			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
 			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
@@ -504,8 +522,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
 			{ 21, 23, -12 }
 			}
-		},
-		/* 12BPP/14BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 14,
 		{ 341, 15, 2048, 15, 24, 23, 23, {
 			{ 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 },
 			{ 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
@@ -513,8 +533,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 },
 			{ 22, 23, -12 }
 			}
-		},
-		/* 12BPP/16BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 16,
 		{ 341, 15, 2048, 19, 28, 27, 27, {
 			{ 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 },
 			{ 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
@@ -522,26 +544,30 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 },
 			{ 26, 27, -12 }
 			}
-		},
+		}
 	},
 	{
-		/* 15BPP/8BPC */
+		.bpp = DSC_BPP(15), .bpc = 8,
 		{ 273, 15, 2048, 3, 12, 11, 11, {
 			{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
 			{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 },
 			{ 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 },
 			{ 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 }
 			}
-		},
-		/* 15BPP/10BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 10,
 		{ 273, 15, 2048, 7, 16, 15, 15, {
 			{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
 			{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 },
 			{ 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 },
 			{ 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 }
 			}
-		},
-		/* 15BPP/12BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 12,
 		{ 273, 15, 2048, 11, 20, 19, 19, {
 			{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
 			{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
@@ -549,8 +575,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 },
 			{ 16, 17, -12 }
 			}
-		},
-		/* 15BPP/14BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 14,
 		{ 273, 15, 2048, 15, 24, 23, 23, {
 			{ 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 },
 			{ 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 },
@@ -558,8 +586,10 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 },
 			{ 20, 21, -12 }
 			}
-		},
-		/* 15BPP/16BPC */
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 16,
 		{ 273, 15, 2048, 19, 28, 27, 27, {
 			{ 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 },
 			{ 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 },
@@ -568,59 +598,21 @@ static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = {
 			{ 24, 25, -12 }
 			}
 		}
-	}
+	},
+	{ /* sentinel */ }
 };
 
-static int get_row_index_for_rc_params(u16 compressed_bpp)
-{
-	switch (compressed_bpp) {
-	case 6:
-		return ROW_INDEX_6BPP;
-	case 8:
-		return ROW_INDEX_8BPP;
-	case 10:
-		return ROW_INDEX_10BPP;
-	case 12:
-		return ROW_INDEX_12BPP;
-	case 15:
-		return ROW_INDEX_15BPP;
-	default:
-		return -EINVAL;
-	}
-}
-
-static int get_column_index_for_rc_params(u8 bits_per_component)
-{
-	switch (bits_per_component) {
-	case 8:
-		return COLUMN_INDEX_8BPC;
-	case 10:
-		return COLUMN_INDEX_10BPC;
-	case 12:
-		return COLUMN_INDEX_12BPC;
-	case 14:
-		return COLUMN_INDEX_14BPC;
-	case 16:
-		return COLUMN_INDEX_16BPC;
-	default:
-		return -EINVAL;
-	}
-}
-
-static const struct rc_parameters *get_rc_params(u16 compressed_bpp,
+static const struct rc_parameters *get_rc_params(u16 dsc_bpp,
 						 u8 bits_per_component)
 {
-	int row_index, column_index;
-
-	row_index = get_row_index_for_rc_params(compressed_bpp);
-	if (row_index < 0)
-		return NULL;
+	int i;
 
-	column_index = get_column_index_for_rc_params(bits_per_component);
-	if (column_index < 0)
-		return NULL;
+	for (i = 0; rc_parameters[i].bpp; i++)
+		if (rc_parameters[i].bpp == dsc_bpp &&
+		    rc_parameters[i].bpc == bits_per_component)
+			return &rc_parameters[i].params;
 
-	return &rc_parameters[row_index][column_index];
+	return NULL;
 }
 
 /**
@@ -642,11 +634,7 @@ int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg)
 			 !vdsc_cfg->bits_per_component))
 		return -EINVAL;
 
-	/* fractional BPP is not supported */
-	if (vdsc_cfg->bits_per_pixel & 0xf)
-		return -EINVAL;
-
-	rc_params = get_rc_params(vdsc_cfg->bits_per_pixel >> 4,
+	rc_params = get_rc_params(vdsc_cfg->bits_per_pixel,
 				  vdsc_cfg->bits_per_component);
 	if (!rc_params)
 		return -EINVAL;
-- 
GitLab


From e3290f883127159e3aa7957f30bd4266602d403e Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:05 +0300
Subject: [PATCH 0566/1791] drm/display/dsc: split DSC 1.2 and DSC 1.1
 (pre-SCR) parameters

The array of rc_parameters contains a mixture of parameters from DSC 1.1
and DSC 1.2 standards. Split these tow configuration arrays in
preparation to adding more configuration data.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Suraj Kandpal
Reviewed-by: Suraj Kandpal
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-7-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c  | 139 ++++++++++++++++++----
 drivers/gpu/drm/i915/display/intel_vdsc.c |  10 +-
 include/drm/display/drm_dsc_helper.h      |   7 +-
 3 files changed, 129 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index acb93d4116e0c..f1ba39df57082 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -325,10 +325,88 @@ struct rc_parameters_data {
 #define DSC_BPP(bpp)	((bpp) << 4)
 
 /*
- * Selected Rate Control Related Parameter Recommended Values
- * from DSC_v1.11 spec & C Model release: DSC_model_20161212
+ * Rate Control Related Parameter Recommended Values from DSC_v1.1 spec prior
+ * to DSC 1.1 fractional bpp underflow SCR (DSC_v1.1_E1.pdf)
+ *
+ * Cross-checked against C Model releases: DSC_model_20161212 and 20210623
  */
-static const struct rc_parameters_data rc_parameters[] = {
+static const struct rc_parameters_data rc_parameters_pre_scr[] = {
+	{
+		.bpp = DSC_BPP(8), .bpc = 8,
+		{ 512, 12, 6144, 3, 12, 11, 11, {
+			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 },
+			{ 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 10,
+		{ 512, 12, 6144, 7, 16, 15, 15, {
+			/*
+			 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
+			 * VESA DSC 1.1 Table E-5 sets it to 4.
+			 */
+			{ 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
+			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 12,
+		{ 512, 12, 6144, 11, 20, 19, 19, {
+			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
+			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 8,
+		{ 341, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 },
+			{ 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 10,
+		{ 341, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
+			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
+			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(12), .bpc = 12,
+		{ 341, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
+			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
+			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		}
+	},
+	{ /* sentinel */ }
+};
+
+/*
+ * Selected Rate Control Related Parameter Recommended Values from DSC v1.2, v1.2a, v1.2b and
+ * DSC_v1.1_E1 specs.
+ *
+ * Cross-checked against C Model releases: DSC_model_20161212 and 20210623
+ */
+static const struct rc_parameters_data rc_parameters_1_2_444[] = {
 	{
 		.bpp = DSC_BPP(6), .bpc = 8,
 		{ 768, 15, 6144, 3, 13, 11, 11, {
@@ -388,22 +466,18 @@ static const struct rc_parameters_data rc_parameters[] = {
 		{ 512, 12, 6144, 3, 12, 11, 11, {
 			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
 			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
-			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 },
-			{ 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 },
+			{ 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 }
 			}
 		}
 	},
 	{
 		.bpp = DSC_BPP(8), .bpc = 10,
 		{ 512, 12, 6144, 7, 16, 15, 15, {
-			/*
-			 * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however
-			 * VESA DSC 1.1 Table E-5 sets it to 4.
-			 */
-			{ 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
+			{ 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
 			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
-			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
-			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 },
+			{ 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 }
 			}
 		}
 	},
@@ -412,9 +486,9 @@ static const struct rc_parameters_data rc_parameters[] = {
 		{ 512, 12, 6144, 11, 20, 19, 19, {
 			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
 			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
-			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
-			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
-			{ 21, 23, -12 }
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 },
+			{ 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 },
+			{ 20, 21, -12 }
 			}
 		}
 	},
@@ -498,8 +572,8 @@ static const struct rc_parameters_data rc_parameters[] = {
 		{ 341, 15, 2048, 3, 12, 11, 11, {
 			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
 			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
-			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 },
-			{ 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			{ 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 },
+			{ 5, 9, -12 }, { 7, 10, -12 }, { 10, 11, -12 }
 			}
 		}
 	},
@@ -508,8 +582,8 @@ static const struct rc_parameters_data rc_parameters[] = {
 		{ 341, 15, 2048, 7, 16, 15, 15, {
 			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
 			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
-			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 },
-			{ 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			{ 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 },
+			{ 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 }
 			}
 		}
 	},
@@ -518,9 +592,9 @@ static const struct rc_parameters_data rc_parameters[] = {
 		{ 341, 15, 2048, 11, 20, 19, 19, {
 			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
 			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
-			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 },
-			{ 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 },
-			{ 21, 23, -12 }
+			{ 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 },
+			{ 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 },
+			{ 18, 19, -12 }
 			}
 		}
 	},
@@ -602,7 +676,8 @@ static const struct rc_parameters_data rc_parameters[] = {
 	{ /* sentinel */ }
 };
 
-static const struct rc_parameters *get_rc_params(u16 dsc_bpp,
+static const struct rc_parameters *get_rc_params(const struct rc_parameters_data *rc_parameters,
+						 u16 dsc_bpp,
 						 u8 bits_per_component)
 {
 	int i;
@@ -622,11 +697,13 @@ static const struct rc_parameters *get_rc_params(u16 dsc_bpp,
  * function.
  *
  * @vdsc_cfg: DSC Configuration data partially filled by driver
+ * @type: operating mode and standard to follow
  *
  * Return: 0 or -error code in case of an error
  */
-int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg)
+int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params_type type)
 {
+	const struct rc_parameters_data *data;
 	const struct rc_parameters *rc_params;
 	int i;
 
@@ -634,7 +711,19 @@ int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg)
 			 !vdsc_cfg->bits_per_component))
 		return -EINVAL;
 
-	rc_params = get_rc_params(vdsc_cfg->bits_per_pixel,
+	switch (type) {
+	case DRM_DSC_1_2_444:
+		data = rc_parameters_1_2_444;
+		break;
+	case DRM_DSC_1_1_PRE_SCR:
+		data = rc_parameters_pre_scr;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	rc_params = get_rc_params(data,
+				  vdsc_cfg->bits_per_pixel,
 				  vdsc_cfg->bits_per_component);
 	if (!rc_params)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
index d4340b18c18d1..bd9116d2cd76c 100644
--- a/drivers/gpu/drm/i915/display/intel_vdsc.c
+++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
@@ -226,7 +226,15 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config)
 	if (DISPLAY_VER(dev_priv) >= 13) {
 		calculate_rc_params(vdsc_cfg);
 	} else {
-		ret = drm_dsc_setup_rc_params(vdsc_cfg);
+		if ((compressed_bpp == 8 ||
+		     compressed_bpp == 12) &&
+		    (vdsc_cfg->bits_per_component == 8 ||
+		     vdsc_cfg->bits_per_component == 10 ||
+		     vdsc_cfg->bits_per_component == 12))
+			ret = drm_dsc_setup_rc_params(vdsc_cfg, DRM_DSC_1_1_PRE_SCR);
+		else
+			ret = drm_dsc_setup_rc_params(vdsc_cfg, DRM_DSC_1_2_444);
+
 		if (ret)
 			return ret;
 
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index 1681791f65a55..66eac7276d044 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -10,12 +10,17 @@
 
 #include <drm/display/drm_dsc.h>
 
+enum drm_dsc_params_type {
+	DRM_DSC_1_2_444,
+	DRM_DSC_1_1_PRE_SCR, /* legacy params from DSC 1.1 */
+};
+
 void drm_dsc_dp_pps_header_init(struct dp_sdp_header *pps_header);
 int drm_dsc_dp_rc_buffer_size(u8 rc_buffer_block_size, u8 rc_buffer_size);
 void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_sdp,
 			      const struct drm_dsc_config *dsc_cfg);
 void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg);
-int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg);
+int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params_type type);
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
 
 #endif /* _DRM_DSC_HELPER_H_ */
-- 
GitLab


From 3b292112f075db4810a556c4538b772e655fd4e6 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:06 +0300
Subject: [PATCH 0567/1791] drm/display/dsc: include the rest of pre-SCR
 parameters

DSC model contains pre-SCR RC parameters for other bpp/bpc combinations,
include them here for completeness. The values were generated from the
'pre_scr_cfg_files_for_reference' files found in DSC models 20210623.
The same fileset is a part of DSC model 20161212.

Reviewed-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-8-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 72 ++++++++++++++++++++++++
 1 file changed, 72 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index f1ba39df57082..f6d8a7be59672 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -331,6 +331,16 @@ struct rc_parameters_data {
  * Cross-checked against C Model releases: DSC_model_20161212 and 20210623
  */
 static const struct rc_parameters_data rc_parameters_pre_scr[] = {
+	{
+		.bpp = DSC_BPP(6), .bpc = 8,
+		{ 683, 15, 6144, 3, 13, 11, 11, {
+			{ 0, 2, 0 }, { 1, 4, -2 }, { 3, 6, -2 }, { 4, 6, -4 },
+			{ 5, 7, -6 }, { 5, 7, -6 }, { 6, 7, -6 }, { 6, 8, -8 },
+			{ 7, 9, -8 }, { 8, 10, -10 }, { 9, 11, -10 }, { 10, 12, -12 },
+			{ 10, 13, -12 }, { 12, 14, -12 }, { 15, 15, -12 }
+			}
+		}
+	},
 	{
 		.bpp = DSC_BPP(8), .bpc = 8,
 		{ 512, 12, 6144, 3, 12, 11, 11, {
@@ -366,6 +376,37 @@ static const struct rc_parameters_data rc_parameters_pre_scr[] = {
 			}
 		}
 	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 8,
+		{ 410, 12, 5632, 3, 12, 11, 11, {
+			{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 11, -10 },
+			{ 5, 12, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 10,
+		{ 410, 12, 5632, 7, 16, 15, 15, {
+			{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 15, -10 },
+			{ 9, 16, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 12,
+		{ 410, 12, 5632, 11, 20, 19, 19, {
+			{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 },
+			{ 13, 19, -10 }, { 13, 20, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		}
+	},
 	{
 		.bpp = DSC_BPP(12), .bpc = 8,
 		{ 341, 15, 2048, 3, 12, 11, 11, {
@@ -397,6 +438,37 @@ static const struct rc_parameters_data rc_parameters_pre_scr[] = {
 			}
 		}
 	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 8,
+		{ 273, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
+			{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 4, -2 }, { 2, 4, -4 },
+			{ 3, 4, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 5, 7, -10 },
+			{ 5, 8, -12 }, { 7, 13, -12 }, { 13, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 10,
+		{ 273, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
+			{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 8, -2 }, { 6, 8, -4 },
+			{ 7, 8, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 9, 11, -10 },
+			{ 9, 12, -12 }, { 11, 17, -12 }, { 17, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(15), .bpc = 12,
+		{ 273, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
+			{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
+			{ 11, 12, -6 }, { 11, 13, -8 }, { 12, 14, -10 },
+			{ 13, 15, -10 }, { 13, 16, -12 }, { 15, 21, -12 },
+			{ 21, 23, -12 }
+			}
+		}
+	},
 	{ /* sentinel */ }
 };
 
-- 
GitLab


From d54fbea3bbbe04cdc944db94eb11c8bda30438b8 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 17 May 2023 13:28:07 +0300
Subject: [PATCH 0568/1791] drm/display/dsc: add YCbCr 4:2:2 and 4:2:0 RC
 parameters

Include RC parameters for YCbCr 4:2:2 and 4:2:0 configurations.

Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517102807.2181589-9-dmitry.baryshkov@linaro.org
Acked-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 450 +++++++++++++++++++++++
 include/drm/display/drm_dsc_helper.h     |   2 +
 2 files changed, 452 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index f6d8a7be59672..fc187a8d8873d 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -748,6 +748,450 @@ static const struct rc_parameters_data rc_parameters_1_2_444[] = {
 	{ /* sentinel */ }
 };
 
+/*
+ * Selected Rate Control Related Parameter Recommended Values for 4:2:2 from
+ * DSC v1.2, v1.2a, v1.2b
+ *
+ * Cross-checked against C Model releases: DSC_model_20161212 and 20210623
+ */
+static const struct rc_parameters_data rc_parameters_1_2_422[] = {
+	{
+		.bpp = DSC_BPP(6), .bpc = 8,
+		{ 512, 15, 6144, 3, 12, 11, 11, {
+			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 },
+			{ 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 10,
+		{ 512, 15, 6144, 7, 16, 15, 15, {
+			{ 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 },
+			{ 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 12,
+		{ 512, 15, 6144, 11, 20, 19, 19, {
+			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 },
+			{ 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 },
+			{ 20, 21, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 14,
+		{ 512, 15, 6144, 15, 24, 23, 23, {
+			{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 },
+			{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 16,
+		{ 512, 15, 6144, 19, 28, 27, 27, {
+			{ 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 },
+			{ 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 },
+			{ 28, 29, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(7), .bpc = 8,
+		{ 410, 15, 5632, 3, 12, 11, 11, {
+			{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 },
+			{ 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(7), .bpc = 10,
+		{ 410, 15, 5632, 7, 16, 15, 15, {
+			{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 },
+			{ 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(7), .bpc = 12,
+		{ 410, 15, 5632, 11, 20, 19, 19, {
+			{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 },
+			{ 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 },
+			{ 19, 20, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(7), .bpc = 14,
+		{ 410, 15, 5632, 15, 24, 23, 23, {
+			{ 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 },
+			{ 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 },
+			{ 23, 24, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(7), .bpc = 16,
+		{ 410, 15, 5632, 19, 28, 27, 27, {
+			{ 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 },
+			{ 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 },
+			{ 27, 28, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 8,
+		{ 341, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 },
+			{ 5, 9, -12 }, { 7, 10, -12 }, { 10, 11, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 10,
+		{ 341, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
+			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 },
+			{ 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 12,
+		{ 341, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
+			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 },
+			{ 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 },
+			{ 18, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 14,
+		{ 341, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 },
+			{ 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 },
+			{ 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 },
+			{ 22, 23, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 16,
+		{ 341, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 },
+			{ 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 },
+			{ 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 },
+			{ 26, 27, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 8,
+		{ 273, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
+			{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 },
+			{ 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 },
+			{ 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 10,
+		{ 273, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
+			{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 },
+			{ 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 },
+			{ 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 12,
+		{ 273, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
+			{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
+			{ 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 },
+			{ 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 },
+			{ 16, 17, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 14,
+		{ 273, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 },
+			{ 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 },
+			{ 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 },
+			{ 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 },
+			{ 20, 21, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(10), .bpc = 16,
+		{ 273, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 },
+			{ 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 },
+			{ 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 },
+			{ 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		}
+	},
+	{ /* sentinel */ }
+};
+
+/*
+ * Selected Rate Control Related Parameter Recommended Values for 4:2:2 from
+ * DSC v1.2, v1.2a, v1.2b
+ *
+ * Cross-checked against C Model releases: DSC_model_20161212 and 20210623
+ */
+static const struct rc_parameters_data rc_parameters_1_2_420[] = {
+	{
+		.bpp = DSC_BPP(4), .bpc = 8,
+		{ 512, 12, 6144, 3, 12, 11, 11, {
+			{ 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 },
+			{ 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(4), .bpc = 10,
+		{ 512, 12, 6144, 7, 16, 15, 15, {
+			{ 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 },
+			{ 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(4), .bpc = 12,
+		{ 512, 12, 6144, 11, 20, 19, 19, {
+			{ 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 },
+			{ 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 },
+			{ 20, 21, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(4), .bpc = 14,
+		{ 512, 12, 6144, 15, 24, 23, 23, {
+			{ 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 },
+			{ 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(4), .bpc = 16,
+		{ 512, 12, 6144, 19, 28, 27, 27, {
+			{ 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 },
+			{ 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 },
+			{ 28, 29, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(5), .bpc = 8,
+		{ 410, 15, 5632, 3, 12, 11, 11, {
+			{ 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 },
+			{ 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(5), .bpc = 10,
+		{ 410, 15, 5632, 7, 16, 15, 15, {
+			{ 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 },
+			{ 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 },
+			{ 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(5), .bpc = 12,
+		{ 410, 15, 5632, 11, 20, 19, 19, {
+			{ 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 },
+			{ 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 },
+			{ 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 },
+			{ 19, 20, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(5), .bpc = 14,
+		{ 410, 15, 5632, 15, 24, 23, 23, {
+			{ 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 },
+			{ 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 },
+			{ 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 },
+			{ 23, 24, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(5), .bpc = 16,
+		{ 410, 15, 5632, 19, 28, 27, 27, {
+			{ 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 },
+			{ 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 },
+			{ 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 },
+			{ 27, 28, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 8,
+		{ 341, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 },
+			{ 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 },
+			{ 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 },
+			{ 5, 9, -12 }, { 7, 10, -12 }, { 10, 12, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 10,
+		{ 341, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 },
+			{ 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 },
+			{ 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 },
+			{ 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 12,
+		{ 341, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 },
+			{ 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 },
+			{ 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 },
+			{ 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 },
+			{ 18, 19, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 14,
+		{ 341, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 },
+			{ 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 },
+			{ 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 },
+			{ 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 },
+			{ 22, 23, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(6), .bpc = 16,
+		{ 341, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 },
+			{ 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 },
+			{ 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 },
+			{ 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 },
+			{ 26, 27, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 8,
+		{ 256, 15, 2048, 3, 12, 11, 11, {
+			{ 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 },
+			{ 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 },
+			{ 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 },
+			{ 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 10,
+		{ 256, 15, 2048, 7, 16, 15, 15, {
+			{ 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 },
+			{ 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 },
+			{ 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 },
+			{ 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 12,
+		{ 256, 15, 2048, 11, 20, 19, 19, {
+			{ 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 },
+			{ 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 },
+			{ 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 },
+			{ 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 },
+			{ 16, 17, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 14,
+		{ 256, 15, 2048, 15, 24, 23, 23, {
+			{ 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 },
+			{ 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 },
+			{ 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 },
+			{ 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 },
+			{ 20, 21, -12 }
+			}
+		}
+	},
+	{
+		.bpp = DSC_BPP(8), .bpc = 16,
+		{ 256, 15, 2048, 19, 28, 27, 27, {
+			{ 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 },
+			{ 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 },
+			{ 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 },
+			{ 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 },
+			{ 24, 25, -12 }
+			}
+		}
+	},
+	{ /* sentinel */ }
+};
+
 static const struct rc_parameters *get_rc_params(const struct rc_parameters_data *rc_parameters,
 						 u16 dsc_bpp,
 						 u8 bits_per_component)
@@ -790,6 +1234,12 @@ int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params
 	case DRM_DSC_1_1_PRE_SCR:
 		data = rc_parameters_pre_scr;
 		break;
+	case DRM_DSC_1_2_422:
+		data = rc_parameters_1_2_422;
+		break;
+	case DRM_DSC_1_2_420:
+		data = rc_parameters_1_2_420;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index 66eac7276d044..fc2104415dcbd 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -13,6 +13,8 @@
 enum drm_dsc_params_type {
 	DRM_DSC_1_2_444,
 	DRM_DSC_1_1_PRE_SCR, /* legacy params from DSC 1.1 */
+	DRM_DSC_1_2_422,
+	DRM_DSC_1_2_420,
 };
 
 void drm_dsc_dp_pps_header_init(struct dp_sdp_header *pps_header);
-- 
GitLab


From 9e15123eca7942caa8a3e1f58ec0df7d088df149 Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Tue, 31 Jan 2023 14:18:25 -0800
Subject: [PATCH 0569/1791] drm/msm/dsi: Stop unconditionally powering up DSI
 hosts at modeset

In commit 7d8e9a90509f ("drm/msm/dsi: move DSI host powerup to modeset
time"), we moved powering up DSI hosts to modeset time. This wasn't
because it was an elegant design, but there were no better options.

That commit actually ended up breaking ps8640, and thus was born
commit ec7981e6c614 ("drm/msm/dsi: don't powerup at modeset time for
parade-ps8640") as a temporary hack to un-break ps8640 by moving it to
the old way of doing things. It turns out that ps8640 _really_ doesn't
like its pre_enable() function to be called after
dsi_mgr_bridge_power_on(). Specifically (from experimentation, not
because I have any inside knowledge), it looks like the assertion of
"RST#" in the ps8640 runtime resume handler seems like it's not
allowed to happen after dsi_mgr_bridge_power_on()

Recently, Dave Stevenson's series landed allowing bridges some control
over pre_enable ordering. The meaty commit for our purposes is
commit 4fb912e5e190 ("drm/bridge: Introduce pre_enable_prev_first to
alter bridge init order"). As documented by that series, if a bridge
doesn't set "pre_enable_prev_first" then we should use the old ordering.

Now that we have the commit ("drm/bridge: tc358762: Set
pre_enable_prev_first") we can go back to the old ordering, which also
allows us to remove the ps8640 special case.

One last note is that even without reverting commit 7d8e9a90509f
("drm/msm/dsi: move DSI host powerup to modeset time"), if you _just_
revert the ps8640 special case and try it out then it doesn't seem to
fail anymore. I spent time bisecting / debugging this and it turns out
to be mostly luck, so we still want this patch to make sure it's
solid. Specifically the reason it sorta works these days is because
we implemented wait_hpd_asserted() in ps8640 now, plus the magic of
"pm_runtime" autosuspend. The fact that we have wait_hpd_asserted()
implemented means that we actually power the bridge chip up just a wee
bit earlier and then the bridge happens to stay on because of
autosuspend and thus ends up powered before dsi_mgr_bridge_power_on().

Cc: Dave Stevenson <dave.stevenson@raspberrypi.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/521058/
Link: https://lore.kernel.org/r/20230131141756.RFT.v2.2.I4cfeab9d0e07e98ead23dd0736ab4461e6c69002@changeid
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_manager.c | 38 +--------------------------
 1 file changed, 1 insertion(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 1bbac72dad35b..2197a54b9b963 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -34,32 +34,6 @@ static struct msm_dsi_manager msm_dsim_glb;
 #define IS_SYNC_NEEDED()	(msm_dsim_glb.is_sync_needed)
 #define IS_MASTER_DSI_LINK(id)	(msm_dsim_glb.master_dsi_link_id == id)
 
-#ifdef CONFIG_OF
-static bool dsi_mgr_power_on_early(struct drm_bridge *bridge)
-{
-	struct drm_bridge *next_bridge = drm_bridge_get_next_bridge(bridge);
-
-	/*
-	 * If the next bridge in the chain is the Parade ps8640 bridge chip
-	 * then don't power on early since it seems to violate the expectations
-	 * of the firmware that the bridge chip is running.
-	 *
-	 * NOTE: this is expected to be a temporary special case. It's expected
-	 * that we'll eventually have a framework that allows the next level
-	 * bridge to indicate whether it needs us to power on before it or
-	 * after it. When that framework is in place then we'll use it and
-	 * remove this special case.
-	 */
-	return !(next_bridge && next_bridge->of_node &&
-		 of_device_is_compatible(next_bridge->of_node, "parade,ps8640"));
-}
-#else
-static inline bool dsi_mgr_power_on_early(struct drm_bridge *bridge)
-{
-	return true;
-}
-#endif
-
 static inline struct msm_dsi *dsi_mgr_get_dsi(int id)
 {
 	return msm_dsim_glb.dsi[id];
@@ -265,12 +239,6 @@ static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge)
 	int ret;
 
 	DBG("id=%d", id);
-	if (!msm_dsi_device_connected(msm_dsi))
-		return;
-
-	/* Do nothing with the host if it is slave-DSI in case of bonded DSI */
-	if (is_bonded_dsi && !IS_MASTER_DSI_LINK(id))
-		return;
 
 	ret = dsi_mgr_phy_enable(id, phy_shared_timings);
 	if (ret)
@@ -327,8 +295,7 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge)
 	if (is_bonded_dsi && !IS_MASTER_DSI_LINK(id))
 		return;
 
-	if (!dsi_mgr_power_on_early(bridge))
-		dsi_mgr_bridge_power_on(bridge);
+	dsi_mgr_bridge_power_on(bridge);
 
 	ret = msm_dsi_host_enable(host);
 	if (ret) {
@@ -438,9 +405,6 @@ static void dsi_mgr_bridge_mode_set(struct drm_bridge *bridge,
 	msm_dsi_host_set_display_mode(host, adjusted_mode);
 	if (is_bonded_dsi && other_dsi)
 		msm_dsi_host_set_display_mode(other_dsi->host, adjusted_mode);
-
-	if (dsi_mgr_power_on_early(bridge))
-		dsi_mgr_bridge_power_on(bridge);
 }
 
 static enum drm_mode_status dsi_mgr_bridge_mode_valid(struct drm_bridge *bridge,
-- 
GitLab


From d8dd416cb420163f9631a01ddcce78628a6a5962 Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Tue, 31 Jan 2023 14:18:26 -0800
Subject: [PATCH 0570/1791] drm/msm/dsi: More properly handle errors in regards
 to dsi_mgr_bridge_power_on()

In commit 7d8e9a90509f ("drm/msm/dsi: move DSI host powerup to modeset
time") the error handling with regards to dsi_mgr_bridge_power_on()
got a bit worse. Specifically if we failed to power the bridge on then
nothing would really notice. The modeset function couldn't return an
error and thus we'd blindly go forward and try to do the pre-enable.

In commit ec7981e6c614 ("drm/msm/dsi: don't powerup at modeset time
for parade-ps8640") we added a special case to move the powerup back
to pre-enable time for ps8640. When we did that, we didn't try to
recover the old/better error handling just for ps8640.

In the patch ("drm/msm/dsi: Stop unconditionally powering up DSI hosts
at modeset") we've now moved the powering up back to exclusively being
during pre-enable. That means we can add the better error handling
back in, so let's do it. To do so we'll add a new function
dsi_mgr_bridge_power_off() that's matches how errors were handled
prior to commit 7d8e9a90509f ("drm/msm/dsi: move DSI host powerup to
modeset time").

NOTE: Now that we have dsi_mgr_bridge_power_off(), it feels as if we
should be calling it in dsi_mgr_bridge_post_disable(). That would make
some sense, but doing so would change the current behavior and thus
should be a separate patch. Specifically:
* dsi_mgr_bridge_post_disable() always calls dsi_mgr_phy_disable()
  even in the slave-DSI case of bonded DSI. We'd need to add special
  handling for this if it's truly needed.
* dsi_mgr_bridge_post_disable() calls msm_dsi_phy_pll_save_state()
  midway through the poweroff.
* dsi_mgr_bridge_post_disable() has a different order of some of the
  poweroffs / IRQ disables.
For now we'll leave dsi_mgr_bridge_post_disable() alone.

Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/521059/
Link: https://lore.kernel.org/r/20230131141756.RFT.v2.3.I3c87b53c4ab61a7d5e05f601a4eb44c7e3809a01@changeid
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_manager.c | 32 ++++++++++++++++++++++-----
 1 file changed, 26 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 2197a54b9b963..28b8012a21f2b 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -228,7 +228,7 @@ static void msm_dsi_manager_set_split_display(u8 id)
 	}
 }
 
-static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge)
+static int dsi_mgr_bridge_power_on(struct drm_bridge *bridge)
 {
 	int id = dsi_mgr_bridge_get_id(bridge);
 	struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id);
@@ -268,14 +268,31 @@ static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge)
 	if (is_bonded_dsi && msm_dsi1)
 		msm_dsi_host_enable_irq(msm_dsi1->host);
 
-	return;
+	return 0;
 
 host1_on_fail:
 	msm_dsi_host_power_off(host);
 host_on_fail:
 	dsi_mgr_phy_disable(id);
 phy_en_fail:
-	return;
+	return ret;
+}
+
+static void dsi_mgr_bridge_power_off(struct drm_bridge *bridge)
+{
+	int id = dsi_mgr_bridge_get_id(bridge);
+	struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id);
+	struct msm_dsi *msm_dsi1 = dsi_mgr_get_dsi(DSI_1);
+	struct mipi_dsi_host *host = msm_dsi->host;
+	bool is_bonded_dsi = IS_BONDED_DSI();
+
+	msm_dsi_host_disable_irq(host);
+	if (is_bonded_dsi && msm_dsi1) {
+		msm_dsi_host_disable_irq(msm_dsi1->host);
+		msm_dsi_host_power_off(msm_dsi1->host);
+	}
+	msm_dsi_host_power_off(host);
+	dsi_mgr_phy_disable(id);
 }
 
 static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge)
@@ -295,7 +312,11 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge)
 	if (is_bonded_dsi && !IS_MASTER_DSI_LINK(id))
 		return;
 
-	dsi_mgr_bridge_power_on(bridge);
+	ret = dsi_mgr_bridge_power_on(bridge);
+	if (ret) {
+		dev_err(&msm_dsi->pdev->dev, "Power on failed: %d\n", ret);
+		return;
+	}
 
 	ret = msm_dsi_host_enable(host);
 	if (ret) {
@@ -316,8 +337,7 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge)
 host1_en_fail:
 	msm_dsi_host_disable(host);
 host_en_fail:
-
-	return;
+	dsi_mgr_bridge_power_off(bridge);
 }
 
 void msm_dsi_manager_tpg_enable(void)
-- 
GitLab


From 1e0a97f84d73ea1182740f62069690c7f3271abb Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 1 May 2023 04:12:57 +0300
Subject: [PATCH 0571/1791] drm/msm/dsi: don't allow enabling 14nm VCO with
 unprogrammed rate

If the dispcc uses CLK_OPS_PARENT_ENABLE (e.g. on QCM2290), CCF can try
enabling VCO before the rate has been programmed. This can cause clock
lockups and/or other boot issues. Program the VCO to the minimal PLL
rate if the read rate is 0 Hz.

Cc: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reported-by: Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>
Reported-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Fixes: f079f6d999cb ("drm/msm/dsi: Add PHY/PLL for 8x96")
Patchwork: https://patchwork.freedesktop.org/patch/534813/
Link: https://lore.kernel.org/r/20230501011257.3460103-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
index 9f488adea7f54..3ce45b023e637 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
@@ -539,6 +539,9 @@ static int dsi_pll_14nm_vco_prepare(struct clk_hw *hw)
 	if (unlikely(pll_14nm->phy->pll_on))
 		return 0;
 
+	if (dsi_pll_14nm_vco_recalc_rate(hw, VCO_REF_CLK_RATE) == 0)
+		dsi_pll_14nm_vco_set_rate(hw, pll_14nm->phy->cfg->min_pll_rate, VCO_REF_CLK_RATE);
+
 	dsi_phy_write(base + REG_DSI_14nm_PHY_PLL_VREF_CFG1, 0x10);
 	dsi_phy_write(cmn_base + REG_DSI_14nm_PHY_CMN_PLL_CNTRL, 1);
 
-- 
GitLab


From e6f756af381f80c064d1f94077b150f8d95b3dab Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 15 Apr 2023 20:19:26 +0300
Subject: [PATCH 0572/1791] drm/msm/dpu: add HDMI output support

MSM8998 and the older Qualcomm platforms support HDMI outputs. Now as
DPU encoder is ready, add support for using INTF_HDMI.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Arnaud Vrac <rawoul@gmail.com>
Tested-by: Arnaud Vrac <rawoul@gmail.com> # on msm8998
Patchwork: https://patchwork.freedesktop.org/patch/532371/
Link: https://lore.kernel.org/r/20230415171926.85774-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 44 +++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index e85e3721d2c71..717a44f135eb9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -617,6 +617,44 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev,
 	return 0;
 }
 
+static int _dpu_kms_initialize_hdmi(struct drm_device *dev,
+				    struct msm_drm_private *priv,
+				    struct dpu_kms *dpu_kms)
+{
+	struct drm_encoder *encoder = NULL;
+	struct msm_display_info info;
+	int rc;
+
+	if (!priv->hdmi)
+		return 0;
+
+	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS);
+	if (IS_ERR(encoder)) {
+		DPU_ERROR("encoder init failed for HDMI display\n");
+		return PTR_ERR(encoder);
+	}
+
+	memset(&info, 0, sizeof(info));
+	rc = msm_hdmi_modeset_init(priv->hdmi, dev, encoder);
+	if (rc) {
+		DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc);
+		drm_encoder_cleanup(encoder);
+		return rc;
+	}
+
+	info.num_of_h_tiles = 1;
+	info.h_tile_instance[0] = 0;
+	info.intf_type = INTF_HDMI;
+	rc = dpu_encoder_setup(dev, encoder, &info);
+	if (rc) {
+		DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
+			  encoder->base.id, rc);
+		return rc;
+	}
+
+	return 0;
+}
+
 static int _dpu_kms_initialize_writeback(struct drm_device *dev,
 		struct msm_drm_private *priv, struct dpu_kms *dpu_kms,
 		const u32 *wb_formats, int n_formats)
@@ -683,6 +721,12 @@ static int _dpu_kms_setup_displays(struct drm_device *dev,
 		return rc;
 	}
 
+	rc = _dpu_kms_initialize_hdmi(dev, priv, dpu_kms);
+	if (rc) {
+		DPU_ERROR("initialize HDMI failed, rc = %d\n", rc);
+		return rc;
+	}
+
 	/* Since WB isn't a driver check the catalog before initializing */
 	if (dpu_kms->catalog->wb_count) {
 		for (i = 0; i < dpu_kms->catalog->wb_count; i++) {
-- 
GitLab


From 95a808328e9918e5716610313f6aec7600108ede Mon Sep 17 00:00:00 2001
From: Arnaud Vrac <avrac@freebox.fr>
Date: Wed, 19 Apr 2023 16:41:10 +0200
Subject: [PATCH 0573/1791] drm/msm/dpu: use hsync/vsync polarity set by the
 encoder

Do not override the hsync/vsync polarity passed by the encoder when
setting up intf timings. The same logic was used in both the encoder and
intf code to set the DP and DSI polarities, so those interfaces are not
impacted. However for HDMI, the polarities were overriden to static
values based on the vertical resolution, instead of using the actual
mode polarities.

Signed-off-by: Arnaud Vrac <avrac@freebox.fr>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/532901/
Link: https://lore.kernel.org/r/20230419-dpu-tweaks-v1-3-d1bac46db075@freebox.fr
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index b9dddf576c029..c665d59f88f5f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -99,7 +99,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx,
 	u32 active_h_start, active_h_end;
 	u32 active_v_start, active_v_end;
 	u32 active_hctl, display_hctl, hsync_ctl;
-	u32 polarity_ctl, den_polarity, hsync_polarity, vsync_polarity;
+	u32 polarity_ctl, den_polarity;
 	u32 panel_format;
 	u32 intf_cfg, intf_cfg2 = 0;
 	u32 display_data_hctl = 0, active_data_hctl = 0;
@@ -186,19 +186,9 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx,
 	}
 
 	den_polarity = 0;
-	if (ctx->cap->type == INTF_HDMI) {
-		hsync_polarity = p->yres >= 720 ? 0 : 1;
-		vsync_polarity = p->yres >= 720 ? 0 : 1;
-	} else if (ctx->cap->type == INTF_DP) {
-		hsync_polarity = p->hsync_polarity;
-		vsync_polarity = p->vsync_polarity;
-	} else {
-		hsync_polarity = 0;
-		vsync_polarity = 0;
-	}
 	polarity_ctl = (den_polarity << 2) | /*  DEN Polarity  */
-		(vsync_polarity << 1) | /* VSYNC Polarity */
-		(hsync_polarity << 0);  /* HSYNC Polarity */
+		(p->vsync_polarity << 1) | /* VSYNC Polarity */
+		(p->hsync_polarity << 0);  /* HSYNC Polarity */
 
 	if (!DPU_FORMAT_IS_YUV(fmt))
 		panel_format = (fmt->bits[C0_G_Y] |
-- 
GitLab


From c95771905c494aa5c2abbb56b3e2f7d4aa3b34f9 Mon Sep 17 00:00:00 2001
From: Arnaud Vrac <avrac@freebox.fr>
Date: Wed, 19 Apr 2023 16:41:15 +0200
Subject: [PATCH 0574/1791] drm/msm/dpu: fix cursor block register bit offset
 in msm8998 hw catalog

This matches the value for both fbdev and sde implementations in the
downstream msm-4.4 repository.

Signed-off-by: Arnaud Vrac <avrac@freebox.fr>
Fixes: 94391a14fc27 ("drm/msm/dpu1: Add MSM8998 to hw catalog")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/532899/
Link: https://lore.kernel.org/r/20230419-dpu-tweaks-v1-8-d1bac46db075@freebox.fr
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index bdcd554fc8a80..9116129527797 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -39,8 +39,8 @@ static const struct dpu_mdp_cfg msm8998_mdp[] = {
 	.clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 },
 	.clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 },
 	.clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 12 },
-	.clk_ctrls[DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 15 },
-	.clk_ctrls[DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 15 },
+	.clk_ctrls[DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 },
+	.clk_ctrls[DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 },
 	},
 };
 
-- 
GitLab


From 8d6dad490afcc05ca7306ee2bac32774302e4aec Mon Sep 17 00:00:00 2001
From: Arnaud Vrac <avrac@freebox.fr>
Date: Wed, 19 Apr 2023 16:41:16 +0200
Subject: [PATCH 0575/1791] drm/msm/dpu: set max cursor width to 512x512

Override the default max cursor size reported to userspace of 64x64.
MSM8998 hw cursor planes support 512x512 size, and other chips use DMA
SSPPs.

Signed-off-by: Arnaud Vrac <avrac@freebox.fr>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/532903/
Link: https://lore.kernel.org/r/20230419-dpu-tweaks-v1-9-d1bac46db075@freebox.fr
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 717a44f135eb9..aebc1dd50a144 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1049,6 +1049,9 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 	dpu_kms = to_dpu_kms(kms);
 	dev = dpu_kms->dev;
 
+	dev->mode_config.cursor_width = 512;
+	dev->mode_config.cursor_height = 512;
+
 	rc = dpu_kms_global_obj_init(dpu_kms);
 	if (rc)
 		return rc;
-- 
GitLab


From 078f766e165097c03629cc1a8f7e1a7b7ec0e79b Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 21 Apr 2023 01:25:57 +0300
Subject: [PATCH 0576/1791] drm/msm/dpu: drop the regdma configuration

The regdma is currently not used by the current driver. We have no way
to practically verify that the regdma is described correctly. Drop it
now.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/533148/
Link: https://lore.kernel.org/r/20230420222558.1208887-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_3_0_msm8998.h   |  1 -
 .../msm/disp/dpu1/catalog/dpu_4_0_sdm845.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_5_0_sm8150.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h   |  2 -
 .../msm/disp/dpu1/catalog/dpu_6_0_sm8250.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_6_2_sc7180.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_7_0_sm8350.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h  |  2 -
 .../msm/disp/dpu1/catalog/dpu_8_1_sm8450.h    |  2 -
 .../msm/disp/dpu1/catalog/dpu_9_0_sm8550.h    |  2 -
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    | 40 -------------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    | 18 ---------
 12 files changed, 77 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 9116129527797..4cd8d7871c4a2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -195,7 +195,6 @@ const struct dpu_mdss_cfg dpu_msm8998_cfg = {
 	.intf = msm8998_intf,
 	.vbif_count = ARRAY_SIZE(msm8998_vbif),
 	.vbif = msm8998_vbif,
-	.reg_dma_count = 0,
 	.perf = &msm8998_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index ceca741e93c9b..63009435e2588 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -193,8 +193,6 @@ const struct dpu_mdss_cfg dpu_sdm845_cfg = {
 	.intf = sdm845_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sdm845_regdma,
 	.perf = &sdm845_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index 42b0e58624d00..b3f67710a48de 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -220,8 +220,6 @@ const struct dpu_mdss_cfg dpu_sm8150_cfg = {
 	.intf = sm8150_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8150_regdma,
 	.perf = &sm8150_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index 3911ae4925404..1aef11a258346 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -222,8 +222,6 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = {
 	.intf = sc8180x_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8150_regdma,
 	.perf = &sc8180x_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index ed130582873c7..903efc580aef0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -228,8 +228,6 @@ const struct dpu_mdss_cfg dpu_sm8250_cfg = {
 	.vbif = sdm845_vbif,
 	.wb_count = ARRAY_SIZE(sm8250_wb),
 	.wb = sm8250_wb,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8250_regdma,
 	.perf = &sm8250_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index a46b11730a4d4..3ab5fbada08fa 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -143,8 +143,6 @@ const struct dpu_mdss_cfg dpu_sc7180_cfg = {
 	.wb = sc7180_wb,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sdm845_regdma,
 	.perf = &sc7180_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 4f6a965bcd90b..9e8d6632a1927 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -211,8 +211,6 @@ const struct dpu_mdss_cfg dpu_sm8350_cfg = {
 	.intf = sm8350_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8350_regdma,
 	.perf = &sm8350_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index 88b8226e6f75d..8e4d0bc4aa703 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -203,8 +203,6 @@ const struct dpu_mdss_cfg dpu_sc8280xp_cfg = {
 	.intf = sc8280xp_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sc8280xp_regdma,
 	.perf = &sc8280xp_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index cb83d25834ceb..014922ac03dbc 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -219,8 +219,6 @@ const struct dpu_mdss_cfg dpu_sm8450_cfg = {
 	.intf = sm8450_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8450_regdma,
 	.perf = &sm8450_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
index d0ab351b6a8b9..be2f37728aa0c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
@@ -224,8 +224,6 @@ const struct dpu_mdss_cfg dpu_sm8550_cfg = {
 	.intf = sm8550_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
 	.vbif = sdm845_vbif,
-	.reg_dma_count = 1,
-	.dma_cfg = &sm8450_regdma,
 	.perf = &sm8550_perf_data,
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 5d994bce696f9..afe667121dc79 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -650,46 +650,6 @@ static const struct dpu_vbif_cfg sdm845_vbif[] = {
 	},
 };
 
-static const struct dpu_reg_dma_cfg sc8280xp_regdma = {
-	.base = 0x0,
-	.version = 0x00020000,
-	.trigger_sel_off = 0x119c,
-	.xin_id = 7,
-	.clk_ctrl = DPU_CLK_CTRL_REG_DMA,
-};
-
-static const struct dpu_reg_dma_cfg sdm845_regdma = {
-	.base = 0x0, .version = 0x1, .trigger_sel_off = 0x119c
-};
-
-static const struct dpu_reg_dma_cfg sm8150_regdma = {
-	.base = 0x0, .version = 0x00010001, .trigger_sel_off = 0x119c
-};
-
-static const struct dpu_reg_dma_cfg sm8250_regdma = {
-	.base = 0x0,
-	.version = 0x00010002,
-	.trigger_sel_off = 0x119c,
-	.xin_id = 7,
-	.clk_ctrl = DPU_CLK_CTRL_REG_DMA,
-};
-
-static const struct dpu_reg_dma_cfg sm8350_regdma = {
-	.base = 0x400,
-	.version = 0x00020000,
-	.trigger_sel_off = 0x119c,
-	.xin_id = 7,
-	.clk_ctrl = DPU_CLK_CTRL_REG_DMA,
-};
-
-static const struct dpu_reg_dma_cfg sm8450_regdma = {
-	.base = 0x0,
-	.version = 0x00020000,
-	.trigger_sel_off = 0x119c,
-	.xin_id = 7,
-	.clk_ctrl = DPU_CLK_CTRL_REG_DMA,
-};
-
 /*************************************************************
  * PERF data config
  *************************************************************/
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 71584cd56fd75..8d62c21b051a8 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -720,21 +720,6 @@ struct dpu_vbif_cfg {
 	u32 memtype_count;
 	u32 memtype[MAX_XIN_COUNT];
 };
-/**
- * struct dpu_reg_dma_cfg - information of lut dma blocks
- * @id                 enum identifying this block
- * @base               register offset of this block
- * @features           bit mask identifying sub-blocks/features
- * @version            version of lutdma hw block
- * @trigger_sel_off    offset to trigger select registers of lutdma
- */
-struct dpu_reg_dma_cfg {
-	DPU_HW_BLK_INFO;
-	u32 version;
-	u32 trigger_sel_off;
-	u32 xin_id;
-	enum dpu_clk_ctrl_type clk_ctrl;
-};
 
 /**
  * Define CDP use cases
@@ -850,9 +835,6 @@ struct dpu_mdss_cfg {
 	u32 wb_count;
 	const struct dpu_wb_cfg *wb;
 
-	u32 reg_dma_count;
-	const struct dpu_reg_dma_cfg *dma_cfg;
-
 	u32 ad_count;
 
 	u32 dspp_count;
-- 
GitLab


From 10036161efbf573fa5b2e42463e5bffe6f9bfb6d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 21 Apr 2023 01:25:58 +0300
Subject: [PATCH 0577/1791] drm/msm/dpu: stop mapping the regdma region

Stop mapping the regdma region. The driver does not support regdma.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/533150/
Link: https://lore.kernel.org/r/20230420222558.1208887-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 6 ------
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 2 +-
 2 files changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index aebc1dd50a144..e5f02ff9d8e09 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1080,12 +1080,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 		DPU_DEBUG("VBIF NRT is not defined");
 	}
 
-	dpu_kms->reg_dma = msm_ioremap_quiet(dpu_kms->pdev, "regdma");
-	if (IS_ERR(dpu_kms->reg_dma)) {
-		dpu_kms->reg_dma = NULL;
-		DPU_DEBUG("REG_DMA is not defined");
-	}
-
 	dpu_kms_parse_data_bus_icc_path(dpu_kms);
 
 	rc = pm_runtime_resume_and_get(&dpu_kms->pdev->dev);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index aca39a4689f48..15111e433f219 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -71,7 +71,7 @@ struct dpu_kms {
 	const struct dpu_mdss_cfg *catalog;
 
 	/* io/register spaces: */
-	void __iomem *mmio, *vbif[VBIF_MAX], *reg_dma;
+	void __iomem *mmio, *vbif[VBIF_MAX];
 
 	struct regulator *vdd;
 	struct regulator *mmagic;
-- 
GitLab


From 94fdd55b93b46fcd6bc8415ad15b008b45031b1b Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Wed, 26 Apr 2023 01:11:10 +0200
Subject: [PATCH 0578/1791] drm/msm/dpu: Drop unused members from HW structs

Some of these members were initialized while never read, while others
were not even assigned any pointer value at all.  Drop them to save some
space, and above all confusion when looking at or accidentally
dereferencing these members.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/533862/
Link: https://lore.kernel.org/r/20230418-dpu-drop-useless-for-lookup-v3-2-e8d869eea455@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 1 -
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h | 1 -
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 1 -
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h | 2 --
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c   | 1 -
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h   | 5 -----
 6 files changed, 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index c665d59f88f5f..ac9cc916eabc2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -359,7 +359,6 @@ struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx,
 	 */
 	c->idx = idx;
 	c->cap = cfg;
-	c->mdss = m;
 	_setup_intf_ops(&c->ops, c->cap->features);
 
 	return c;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
index 643dd10bc0306..e07b2e33af3e2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
@@ -90,7 +90,6 @@ struct dpu_hw_intf {
 	/* intf */
 	enum dpu_intf idx;
 	const struct dpu_intf_cfg *cap;
-	const struct dpu_mdss_cfg *mdss;
 
 	/* ops */
 	struct dpu_hw_intf_ops ops;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index cf70a9bd1034f..bb3ddec5c7d73 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -812,7 +812,6 @@ struct dpu_hw_sspp *dpu_hw_sspp_init(enum dpu_sspp idx,
 	}
 
 	/* Assign ops */
-	hw_pipe->catalog = catalog;
 	hw_pipe->ubwc = catalog->ubwc;
 	hw_pipe->idx = idx;
 	hw_pipe->cap = cfg;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index 74b98b6b3bc3a..5004a02fd61e2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -341,7 +341,6 @@ struct dpu_hw_sspp_ops {
  * struct dpu_hw_sspp - pipe description
  * @base: hardware block base structure
  * @hw: block hardware details
- * @catalog: back pointer to catalog
  * @ubwc: ubwc configuration data
  * @idx: pipe index
  * @cap: pointer to layer_cfg
@@ -350,7 +349,6 @@ struct dpu_hw_sspp_ops {
 struct dpu_hw_sspp {
 	struct dpu_hw_blk base;
 	struct dpu_hw_blk_reg_map hw;
-	const struct dpu_mdss_cfg *catalog;
 	const struct dpu_ubwc_cfg *ubwc;
 
 	/* Pipe */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
index a3e413d277175..f33b48045b5c7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
@@ -263,7 +263,6 @@ struct dpu_hw_wb *dpu_hw_wb_init(enum dpu_wb idx,
 	}
 
 	/* Assign ops */
-	c->mdp = &m->mdp[0];
 	c->idx = idx;
 	c->caps = cfg;
 	_setup_wb_ops(&c->ops, c->caps->features);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
index 3ff5a48541e2e..b91923f879f13 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
@@ -76,15 +76,12 @@ struct dpu_hw_wb_ops {
 /**
  * struct dpu_hw_wb : WB driver object
  * @hw: block hardware details
- * @mdp: pointer to associated mdp portion of the catalog
  * @idx: hardware index number within type
  * @wb_hw_caps: hardware capabilities
  * @ops: function pointers
- * @hw_mdp: MDP top level hardware block
  */
 struct dpu_hw_wb {
 	struct dpu_hw_blk_reg_map hw;
-	const struct dpu_mdp_cfg *mdp;
 
 	/* wb path */
 	int idx;
@@ -92,8 +89,6 @@ struct dpu_hw_wb {
 
 	/* ops */
 	struct dpu_hw_wb_ops ops;
-
-	struct dpu_hw_mdp *hw_mdp;
 };
 
 /**
-- 
GitLab


From babdb815ef04572a940e78e533d7e90fc505b77f Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Wed, 26 Apr 2023 01:11:11 +0200
Subject: [PATCH 0579/1791] drm/msm/dpu: Pass catalog pointers in RM to replace
 for-loop ID lookups

The Resource Manager already iterates over all available blocks from the
catalog, only to pass their ID to a dpu_hw_xxx_init() function which
uses an _xxx_offset() helper to search for and find the exact same
catalog pointer again to initialize the block with, fallible error
handling and all.

Instead, pass const pointers to the catalog entries directly to these
_init functions and drop the for loops entirely, saving on both
readability complexity and unnecessary cycles at boot.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/533861/
Link: https://lore.kernel.org/r/20230418-dpu-drop-useless-for-lookup-v3-3-e8d869eea455@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c    | 37 ++++-------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h    | 14 ++++---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c    | 32 +++------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h    | 11 +++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c   | 38 +++---------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h   | 12 +++---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h |  2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c   | 40 +++++--------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h   | 12 +++---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c     | 38 ++++--------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h     | 10 ++---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c    | 33 +++------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h    | 14 +++----
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c   | 33 +++------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h   | 14 +++----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c   | 39 ++++--------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h   | 12 +++---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c   | 33 +++------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h   | 11 +++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c     | 33 +++------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h     | 11 +++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c       | 17 ++++----
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c        | 22 +++++-----
 23 files changed, 139 insertions(+), 379 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index bbdc95ce374a7..4f7cfa9a00f6c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -53,23 +53,6 @@ static const u32 fetch_tbl[SSPP_MAX] = {CTL_INVALID_BIT, 16, 17, 18, 19,
 	CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, 0,
 	1, 2, 3, CTL_INVALID_BIT, CTL_INVALID_BIT};
 
-static const struct dpu_ctl_cfg *_ctl_offset(enum dpu_ctl ctl,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->ctl_count; i++) {
-		if (ctl == m->ctl[i].id) {
-			b->blk_addr = addr + m->ctl[i].base;
-			b->log_mask = DPU_DBG_MASK_CTL;
-			return &m->ctl[i];
-		}
-	}
-	return ERR_PTR(-ENOMEM);
-}
-
 static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count,
 		enum dpu_lm lm)
 {
@@ -676,29 +659,25 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops,
 		ops->set_active_pipes = dpu_hw_ctl_set_fetch_pipe_active;
 };
 
-struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx,
+struct dpu_hw_ctl *dpu_hw_ctl_init(const struct dpu_ctl_cfg *cfg,
 		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+		u32 mixer_count,
+		const struct dpu_lm_cfg *mixer)
 {
 	struct dpu_hw_ctl *c;
-	const struct dpu_ctl_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _ctl_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		pr_err("failed to create dpu_hw_ctl %d\n", idx);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_CTL;
 
 	c->caps = cfg;
 	_setup_ctl_ops(&c->ops, c->caps->features);
-	c->idx = idx;
-	c->mixer_count = m->mixer_count;
-	c->mixer_hw_caps = m->mixer;
+	c->idx = cfg->id;
+	c->mixer_count = mixer_count;
+	c->mixer_hw_caps = mixer;
 
 	return c;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
index 78611a8316974..6292002faba2d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
@@ -261,15 +261,17 @@ static inline struct dpu_hw_ctl *to_dpu_hw_ctl(struct dpu_hw_blk *hw)
 }
 
 /**
- * dpu_hw_ctl_init(): Initializes the ctl_path hw driver object.
- * should be called before accessing every ctl path registers.
- * @idx:  ctl_path index for which driver object is required
+ * dpu_hw_ctl_init() - Initializes the ctl_path hw driver object.
+ * Should be called before accessing any ctl_path register.
+ * @cfg:  ctl_path catalog entry for which driver object is required
  * @addr: mapped register io address of MDP
- * @m :   pointer to mdss catalog data
+ * @mixer_count: Number of mixers in @mixer
+ * @mixer: Pointer to an array of Layer Mixers defined in the catalog
  */
-struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx,
+struct dpu_hw_ctl *dpu_hw_ctl_init(const struct dpu_ctl_cfg *cfg,
 		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+		u32 mixer_count,
+		const struct dpu_lm_cfg *mixer);
 
 /**
  * dpu_hw_ctl_destroy(): Destroys ctl driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
index 4e1396575e6aa..89e1a9f945c71 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
@@ -175,24 +175,6 @@ static void dpu_hw_dsc_bind_pingpong_blk(
 	DPU_REG_WRITE(c, dsc_ctl_offset, mux_cfg);
 }
 
-static const struct dpu_dsc_cfg *_dsc_offset(enum dpu_dsc dsc,
-				       const struct dpu_mdss_cfg *m,
-				       void __iomem *addr,
-				       struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->dsc_count; i++) {
-		if (dsc == m->dsc[i].id) {
-			b->blk_addr = addr + m->dsc[i].base;
-			b->log_mask = DPU_DBG_MASK_DSC;
-			return &m->dsc[i];
-		}
-	}
-
-	return NULL;
-}
-
 static void _setup_dsc_ops(struct dpu_hw_dsc_ops *ops,
 			   unsigned long cap)
 {
@@ -203,23 +185,19 @@ static void _setup_dsc_ops(struct dpu_hw_dsc_ops *ops,
 		ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk;
 };
 
-struct dpu_hw_dsc *dpu_hw_dsc_init(enum dpu_dsc idx, void __iomem *addr,
-				   const struct dpu_mdss_cfg *m)
+struct dpu_hw_dsc *dpu_hw_dsc_init(const struct dpu_dsc_cfg *cfg,
+				   void __iomem *addr)
 {
 	struct dpu_hw_dsc *c;
-	const struct dpu_dsc_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _dsc_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_DSC;
 
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->caps = cfg;
 	_setup_dsc_ops(&c->ops, c->caps->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
index ae9b5db53d7f6..287ec5f1e170a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
@@ -61,14 +61,13 @@ struct dpu_hw_dsc {
 };
 
 /**
- * dpu_hw_dsc_init - initializes the dsc block for the passed dsc idx.
- * @idx:  DSC index for which driver object is required
+ * dpu_hw_dsc_init() - Initializes the DSC hw driver object.
+ * @cfg:  DSC catalog entry for which driver object is required
  * @addr: Mapped register io address of MDP
- * @m:    Pointer to mdss catalog data
- * Returns: Error code or allocated dpu_hw_dsc context
+ * Return: Error code or allocated dpu_hw_dsc context
  */
-struct dpu_hw_dsc *dpu_hw_dsc_init(enum dpu_dsc idx, void __iomem *addr,
-				   const struct dpu_mdss_cfg *m);
+struct dpu_hw_dsc *dpu_hw_dsc_init(const struct dpu_dsc_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_dsc_destroy - destroys dsc driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c
index 8ab5ace34a2d5..9419b2209af88 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c
@@ -68,49 +68,23 @@ static void _setup_dspp_ops(struct dpu_hw_dspp *c,
 		c->ops.setup_pcc = dpu_setup_dspp_pcc;
 }
 
-static const struct dpu_dspp_cfg *_dspp_offset(enum dpu_dspp dspp,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	if (!m || !addr || !b)
-		return ERR_PTR(-EINVAL);
-
-	for (i = 0; i < m->dspp_count; i++) {
-		if (dspp == m->dspp[i].id) {
-			b->blk_addr = addr + m->dspp[i].base;
-			b->log_mask = DPU_DBG_MASK_DSPP;
-			return &m->dspp[i];
-		}
-	}
-
-	return ERR_PTR(-EINVAL);
-}
-
-struct dpu_hw_dspp *dpu_hw_dspp_init(enum dpu_dspp idx,
-			void __iomem *addr,
-			const struct dpu_mdss_cfg *m)
+struct dpu_hw_dspp *dpu_hw_dspp_init(const struct dpu_dspp_cfg *cfg,
+			void __iomem *addr)
 {
 	struct dpu_hw_dspp *c;
-	const struct dpu_dspp_cfg *cfg;
 
-	if (!addr || !m)
+	if (!addr)
 		return ERR_PTR(-EINVAL);
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _dspp_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_DSPP;
 
 	/* Assign ops */
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->cap = cfg;
 	_setup_dspp_ops(c, c->cap->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h
index 05ecfdfac93b7..bea9656813301 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h
@@ -79,14 +79,14 @@ static inline struct dpu_hw_dspp *to_dpu_hw_dspp(struct dpu_hw_blk *hw)
 }
 
 /**
- * dpu_hw_dspp_init - initializes the dspp hw driver object.
- * should be called once before accessing every dspp.
- * @idx:  DSPP index for which driver object is required
+ * dpu_hw_dspp_init() - Initializes the DSPP hw driver object.
+ * should be called once before accessing every DSPP.
+ * @cfg:  DSPP catalog entry for which driver object is required
  * @addr: Mapped register io address of MDP
- * @Return: pointer to structure or ERR_PTR
+ * Return: pointer to structure or ERR_PTR
  */
-struct dpu_hw_dspp *dpu_hw_dspp_init(enum dpu_dspp idx,
-	void __iomem *addr, const struct dpu_mdss_cfg *m);
+struct dpu_hw_dspp *dpu_hw_dspp_init(const struct dpu_dspp_cfg *cfg,
+	void __iomem *addr);
 
 /**
  * dpu_hw_dspp_destroy(): Destroys DSPP driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
index 425465011c807..bbf475a1cb455 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
@@ -67,7 +67,7 @@ struct dpu_hw_intr {
 /**
  * dpu_hw_intr_init(): Initializes the interrupts hw object
  * @addr: mapped register io address of MDP
- * @m :   pointer to mdss catalog data
+ * @m:    pointer to MDSS catalog data
  */
 struct dpu_hw_intr *dpu_hw_intr_init(void __iomem *addr,
 		const struct dpu_mdss_cfg *m);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index ac9cc916eabc2..f0271283a3a71 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -68,25 +68,6 @@
 #define INTF_MISR_CTRL			0x180
 #define INTF_MISR_SIGNATURE		0x184
 
-static const struct dpu_intf_cfg *_intf_offset(enum dpu_intf intf,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->intf_count; i++) {
-		if ((intf == m->intf[i].id) &&
-		(m->intf[i].type != INTF_NONE)) {
-			b->blk_addr = addr + m->intf[i].base;
-			b->log_mask = DPU_DBG_MASK_INTF;
-			return &m->intf[i];
-		}
-	}
-
-	return ERR_PTR(-EINVAL);
-}
-
 static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx,
 		const struct intf_timing_params *p,
 		const struct dpu_format *fmt)
@@ -336,28 +317,27 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops,
 	ops->collect_misr = dpu_hw_intf_collect_misr;
 }
 
-struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_intf *c;
-	const struct dpu_intf_cfg *cfg;
+
+	if (cfg->type == INTF_NONE) {
+		DPU_DEBUG("Skip intf %d with type NONE\n", cfg->id - INTF_0);
+		return NULL;
+	}
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _intf_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		pr_err("failed to create dpu_hw_intf %d\n", idx);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_INTF;
 
 	/*
 	 * Assign ops
 	 */
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->cap = cfg;
 	_setup_intf_ops(&c->ops, c->cap->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
index e07b2e33af3e2..102c4f0e812b0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
@@ -96,15 +96,13 @@ struct dpu_hw_intf {
 };
 
 /**
- * dpu_hw_intf_init(): Initializes the intf driver for the passed
- * interface idx.
- * @idx:  interface index for which driver object is required
+ * dpu_hw_intf_init() - Initializes the INTF driver for the passed
+ * interface catalog entry.
+ * @cfg:  interface catalog entry for which driver object is required
  * @addr: mapped register io address of MDP
- * @m :   pointer to mdss catalog data
  */
-struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_intf_destroy(): Destroys INTF driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
index f5120ea91edee..214f207ddf8ad 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
@@ -30,24 +30,6 @@
 #define LM_MISR_SIGNATURE                0x314
 
 
-static const struct dpu_lm_cfg *_lm_offset(enum dpu_lm mixer,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->mixer_count; i++) {
-		if (mixer == m->mixer[i].id) {
-			b->blk_addr = addr + m->mixer[i].base;
-			b->log_mask = DPU_DBG_MASK_LM;
-			return &m->mixer[i];
-		}
-	}
-
-	return ERR_PTR(-ENOMEM);
-}
-
 /**
  * _stage_offset(): returns the relative offset of the blend registers
  * for the stage to be setup
@@ -160,8 +142,7 @@ static void dpu_hw_lm_setup_color3(struct dpu_hw_mixer *ctx,
 	DPU_REG_WRITE(c, LM_OP_MODE, op_mode);
 }
 
-static void _setup_mixer_ops(const struct dpu_mdss_cfg *m,
-		struct dpu_hw_lm_ops *ops,
+static void _setup_mixer_ops(struct dpu_hw_lm_ops *ops,
 		unsigned long features)
 {
 	ops->setup_mixer_out = dpu_hw_lm_setup_out;
@@ -175,27 +156,22 @@ static void _setup_mixer_ops(const struct dpu_mdss_cfg *m,
 	ops->collect_misr = dpu_hw_lm_collect_misr;
 }
 
-struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+struct dpu_hw_mixer *dpu_hw_lm_init(const struct dpu_lm_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_mixer *c;
-	const struct dpu_lm_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _lm_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_LM;
 
 	/* Assign ops */
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->cap = cfg;
-	_setup_mixer_ops(m, &c->ops, c->cap->features);
+	_setup_mixer_ops(&c->ops, c->cap->features);
 
 	return c;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h
index 652ddfdedec37..36992d046a533 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h
@@ -93,15 +93,13 @@ static inline struct dpu_hw_mixer *to_dpu_hw_mixer(struct dpu_hw_blk *hw)
 }
 
 /**
- * dpu_hw_lm_init(): Initializes the mixer hw driver object.
+ * dpu_hw_lm_init() - Initializes the mixer hw driver object.
  * should be called once before accessing every mixer.
- * @idx:  mixer index for which driver object is required
+ * @cfg:  mixer catalog entry for which driver object is required
  * @addr: mapped register io address of MDP
- * @m :   pointer to mdss catalog data
  */
-struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_mixer *dpu_hw_lm_init(const struct dpu_lm_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_lm_destroy(): Destroys layer mixer driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c
index def0a87fdba56..90e0e05eff8d3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c
@@ -14,24 +14,6 @@
 #define MERGE_3D_MUX  0x000
 #define MERGE_3D_MODE 0x004
 
-static const struct dpu_merge_3d_cfg *_merge_3d_offset(enum dpu_merge_3d idx,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->merge_3d_count; i++) {
-		if (idx == m->merge_3d[i].id) {
-			b->blk_addr = addr + m->merge_3d[i].base;
-			b->log_mask = DPU_DBG_MASK_PINGPONG;
-			return &m->merge_3d[i];
-		}
-	}
-
-	return ERR_PTR(-EINVAL);
-}
-
 static void dpu_hw_merge_3d_setup_3d_mode(struct dpu_hw_merge_3d *merge_3d,
 			enum dpu_3d_blend_mode mode_3d)
 {
@@ -55,24 +37,19 @@ static void _setup_merge_3d_ops(struct dpu_hw_merge_3d *c,
 	c->ops.setup_3d_mode = dpu_hw_merge_3d_setup_3d_mode;
 };
 
-struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(enum dpu_merge_3d idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(const struct dpu_merge_3d_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_merge_3d *c;
-	const struct dpu_merge_3d_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _merge_3d_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_PINGPONG;
 
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->caps = cfg;
 	_setup_merge_3d_ops(c, c->caps->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h
index 81fd1d5f718e7..19cec5e887221 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h
@@ -46,16 +46,14 @@ static inline struct dpu_hw_merge_3d *to_dpu_hw_merge_3d(struct dpu_hw_blk *hw)
 }
 
 /**
- * dpu_hw_merge_3d_init - initializes the merge_3d driver for the passed
- *	merge_3d idx.
- * @idx:  Pingpong index for which driver object is required
+ * dpu_hw_merge_3d_init() - Initializes the merge_3d driver for the passed
+ * merge3d catalog entry.
+ * @cfg:  Pingpong catalog entry for which driver object is required
  * @addr: Mapped register io address of MDP
- * @m:    Pointer to mdss catalog data
- * Returns: Error code or allocated dpu_hw_merge_3d context
+ * Return: Error code or allocated dpu_hw_merge_3d context
  */
-struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(enum dpu_merge_3d idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(const struct dpu_merge_3d_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_merge_3d_destroy - destroys merge_3d driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index 0fcad9760b6fc..3822e0657767d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -42,24 +42,6 @@ static u32 dither_depth_map[DITHER_DEPTH_MAP_INDEX] = {
 	0, 0, 0, 0, 0, 0, 0, 1, 2
 };
 
-static const struct dpu_pingpong_cfg *_pingpong_offset(enum dpu_pingpong pp,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->pingpong_count; i++) {
-		if (pp == m->pingpong[i].id) {
-			b->blk_addr = addr + m->pingpong[i].base;
-			b->log_mask = DPU_DBG_MASK_PINGPONG;
-			return &m->pingpong[i];
-		}
-	}
-
-	return ERR_PTR(-EINVAL);
-}
-
 static void dpu_hw_pp_setup_dither(struct dpu_hw_pingpong *pp,
 				    struct dpu_hw_dither_cfg *cfg)
 {
@@ -290,24 +272,19 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 		c->ops.setup_dither = dpu_hw_pp_setup_dither;
 };
 
-struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+struct dpu_hw_pingpong *dpu_hw_pingpong_init(const struct dpu_pingpong_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_pingpong *c;
-	const struct dpu_pingpong_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _pingpong_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_PINGPONG;
 
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->caps = cfg;
 	_setup_pingpong_ops(c, c->caps->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
index c00223441d990..51e9093621a7e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
@@ -165,16 +165,14 @@ static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw)
 }
 
 /**
- * dpu_hw_pingpong_init - initializes the pingpong driver for the passed
- *	pingpong idx.
- * @idx:  Pingpong index for which driver object is required
+ * dpu_hw_pingpong_init() - initializes the pingpong driver for the passed
+ * pingpong catalog entry.
+ * @cfg:  Pingpong catalog entry for which driver object is required
  * @addr: Mapped register io address of MDP
- * @m:    Pointer to mdss catalog data
- * Returns: Error code or allocated dpu_hw_pingpong context
+ * Return: Error code or allocated dpu_hw_pingpong context
  */
-struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_pingpong *dpu_hw_pingpong_init(const struct dpu_pingpong_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_pingpong_destroy - destroys pingpong driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index bb3ddec5c7d73..31348ff5d493b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -771,49 +771,24 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms,
 }
 #endif
 
-
-static const struct dpu_sspp_cfg *_sspp_offset(enum dpu_sspp sspp,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *catalog,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	if ((sspp < SSPP_MAX) && catalog && addr && b) {
-		for (i = 0; i < catalog->sspp_count; i++) {
-			if (sspp == catalog->sspp[i].id) {
-				b->blk_addr = addr + catalog->sspp[i].base;
-				b->log_mask = DPU_DBG_MASK_SSPP;
-				return &catalog->sspp[i];
-			}
-		}
-	}
-
-	return ERR_PTR(-ENOMEM);
-}
-
-struct dpu_hw_sspp *dpu_hw_sspp_init(enum dpu_sspp idx,
-		void __iomem *addr, const struct dpu_mdss_cfg *catalog)
+struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg,
+		void __iomem *addr, const struct dpu_ubwc_cfg *ubwc)
 {
 	struct dpu_hw_sspp *hw_pipe;
-	const struct dpu_sspp_cfg *cfg;
 
-	if (!addr || !catalog)
+	if (!addr || !ubwc)
 		return ERR_PTR(-EINVAL);
 
 	hw_pipe = kzalloc(sizeof(*hw_pipe), GFP_KERNEL);
 	if (!hw_pipe)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _sspp_offset(idx, addr, catalog, &hw_pipe->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(hw_pipe);
-		return ERR_PTR(-EINVAL);
-	}
+	hw_pipe->hw.blk_addr = addr + cfg->base;
+	hw_pipe->hw.log_mask = DPU_DBG_MASK_SSPP;
 
 	/* Assign ops */
-	hw_pipe->ubwc = catalog->ubwc;
-	hw_pipe->idx = idx;
+	hw_pipe->ubwc = ubwc;
+	hw_pipe->idx = cfg->id;
 	hw_pipe->cap = cfg;
 	_setup_layer_ops(hw_pipe, hw_pipe->cap->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index 5004a02fd61e2..7a8d11ba618d7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -341,7 +341,7 @@ struct dpu_hw_sspp_ops {
  * struct dpu_hw_sspp - pipe description
  * @base: hardware block base structure
  * @hw: block hardware details
- * @ubwc: ubwc configuration data
+ * @ubwc: UBWC configuration data
  * @idx: pipe index
  * @cap: pointer to layer_cfg
  * @ops: pointer to operations possible for this pipe
@@ -361,14 +361,14 @@ struct dpu_hw_sspp {
 
 struct dpu_kms;
 /**
- * dpu_hw_sspp_init - initializes the sspp hw driver object.
+ * dpu_hw_sspp_init() - Initializes the sspp hw driver object.
  * Should be called once before accessing every pipe.
- * @idx:  Pipe index for which driver object is required
+ * @cfg:  Pipe catalog entry for which driver object is required
  * @addr: Mapped register io address of MDP
- * @catalog : Pointer to mdss catalog data
+ * @ubwc: UBWC configuration data
  */
-struct dpu_hw_sspp *dpu_hw_sspp_init(enum dpu_sspp idx,
-		void __iomem *addr, const struct dpu_mdss_cfg *catalog);
+struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg,
+		void __iomem *addr, const struct dpu_ubwc_cfg *ubwc);
 
 /**
  * dpu_hw_sspp_destroy(): Destroys SSPP driver context
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c
index 16c56e2407064..a5121a50b2bbf 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c
@@ -211,45 +211,22 @@ static void _setup_vbif_ops(struct dpu_hw_vbif_ops *ops,
 	ops->set_write_gather_en = dpu_hw_set_write_gather_en;
 }
 
-static const struct dpu_vbif_cfg *_top_offset(enum dpu_vbif vbif,
-		const struct dpu_mdss_cfg *m,
-		void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->vbif_count; i++) {
-		if (vbif == m->vbif[i].id) {
-			b->blk_addr = addr + m->vbif[i].base;
-			b->log_mask = DPU_DBG_MASK_VBIF;
-			return &m->vbif[i];
-		}
-	}
-
-	return ERR_PTR(-EINVAL);
-}
-
-struct dpu_hw_vbif *dpu_hw_vbif_init(enum dpu_vbif idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m)
+struct dpu_hw_vbif *dpu_hw_vbif_init(const struct dpu_vbif_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_vbif *c;
-	const struct dpu_vbif_cfg *cfg;
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _top_offset(idx, m, addr, &c->hw);
-	if (IS_ERR_OR_NULL(cfg)) {
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_VBIF;
 
 	/*
 	 * Assign ops
 	 */
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->cap = cfg;
 	_setup_vbif_ops(&c->ops, c->cap->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h
index 6417aa28d32c3..7e10d2a172b45 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h
@@ -106,14 +106,13 @@ struct dpu_hw_vbif {
 };
 
 /**
- * dpu_hw_vbif_init - initializes the vbif driver for the passed interface idx
- * @idx:  Interface index for which driver object is required
+ * dpu_hw_vbif_init() - Initializes the VBIF driver for the passed
+ * VBIF catalog entry.
+ * @cfg:  VBIF catalog entry for which driver object is required
  * @addr: Mapped register io address of MDSS
- * @m:    Pointer to mdss catalog data
  */
-struct dpu_hw_vbif *dpu_hw_vbif_init(enum dpu_vbif idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_vbif *dpu_hw_vbif_init(const struct dpu_vbif_cfg *cfg,
+		void __iomem *addr);
 
 void dpu_hw_vbif_destroy(struct dpu_hw_vbif *vbif);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
index f33b48045b5c7..4834866d72fd0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
@@ -52,22 +52,6 @@
 /* WB_QOS_CTRL */
 #define WB_QOS_CTRL_DANGER_SAFE_EN            BIT(0)
 
-static const struct dpu_wb_cfg *_wb_offset(enum dpu_wb wb,
-		const struct dpu_mdss_cfg *m, void __iomem *addr,
-		struct dpu_hw_blk_reg_map *b)
-{
-	int i;
-
-	for (i = 0; i < m->wb_count; i++) {
-		if (wb == m->wb[i].id) {
-			b->blk_addr = addr + m->wb[i].base;
-			b->log_mask = DPU_DBG_MASK_WB;
-			return &m->wb[i];
-		}
-	}
-	return ERR_PTR(-EINVAL);
-}
-
 static void dpu_hw_wb_setup_outaddress(struct dpu_hw_wb *ctx,
 		struct dpu_hw_wb_cfg *data)
 {
@@ -242,28 +226,23 @@ static void _setup_wb_ops(struct dpu_hw_wb_ops *ops,
 		ops->bind_pingpong_blk = dpu_hw_wb_bind_pingpong_blk;
 }
 
-struct dpu_hw_wb *dpu_hw_wb_init(enum dpu_wb idx,
-		void __iomem *addr, const struct dpu_mdss_cfg *m)
+struct dpu_hw_wb *dpu_hw_wb_init(const struct dpu_wb_cfg *cfg,
+		void __iomem *addr)
 {
 	struct dpu_hw_wb *c;
-	const struct dpu_wb_cfg *cfg;
 
-	if (!addr || !m)
+	if (!addr)
 		return ERR_PTR(-EINVAL);
 
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
 
-	cfg = _wb_offset(idx, m, addr, &c->hw);
-	if (IS_ERR(cfg)) {
-		WARN(1, "Unable to find wb idx=%d\n", idx);
-		kfree(c);
-		return ERR_PTR(-EINVAL);
-	}
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_WB;
 
 	/* Assign ops */
-	c->idx = idx;
+	c->idx = cfg->id;
 	c->caps = cfg;
 	_setup_wb_ops(&c->ops, c->caps->features);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
index b91923f879f13..e4f85409c6248 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
@@ -92,14 +92,13 @@ struct dpu_hw_wb {
 };
 
 /**
- * dpu_hw_wb_init(): Initializes and return writeback hw driver object.
- * @idx:  wb_path index for which driver object is required
+ * dpu_hw_wb_init() - Initializes the writeback hw driver object.
+ * @cfg:  wb_path catalog entry for which driver object is required
  * @addr: mapped register io address of MDP
- * @m :   pointer to mdss catalog data
+ * Return: Error code or allocated dpu_hw_wb context
  */
-struct dpu_hw_wb *dpu_hw_wb_init(enum dpu_wb idx,
-		void __iomem *addr,
-		const struct dpu_mdss_cfg *m);
+struct dpu_hw_wb *dpu_hw_wb_init(const struct dpu_wb_cfg *cfg,
+		void __iomem *addr);
 
 /**
  * dpu_hw_wb_destroy(): Destroy writeback hw driver object.
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index e5f02ff9d8e09..8ce057cc93743 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1125,16 +1125,17 @@ static int dpu_kms_hw_init(struct msm_kms *kms)
 	}
 
 	for (i = 0; i < dpu_kms->catalog->vbif_count; i++) {
-		u32 vbif_idx = dpu_kms->catalog->vbif[i].id;
-
-		dpu_kms->hw_vbif[vbif_idx] = dpu_hw_vbif_init(vbif_idx,
-				dpu_kms->vbif[vbif_idx], dpu_kms->catalog);
-		if (IS_ERR(dpu_kms->hw_vbif[vbif_idx])) {
-			rc = PTR_ERR(dpu_kms->hw_vbif[vbif_idx]);
-			DPU_ERROR("failed to init vbif %d: %d\n", vbif_idx, rc);
-			dpu_kms->hw_vbif[vbif_idx] = NULL;
+		struct dpu_hw_vbif *hw;
+		const struct dpu_vbif_cfg *vbif = &dpu_kms->catalog->vbif[i];
+
+		hw = dpu_hw_vbif_init(vbif, dpu_kms->vbif[vbif->id]);
+		if (IS_ERR(hw)) {
+			rc = PTR_ERR(hw);
+			DPU_ERROR("failed to init vbif %d: %d\n", vbif->id, rc);
 			goto power_error;
 		}
+
+		dpu_kms->hw_vbif[vbif->id] = hw;
 	}
 
 	rc = dpu_core_perf_init(&dpu_kms->perf, dev, dpu_kms->catalog,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index f4dda88a73f7d..f0fc70422e565 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -126,7 +126,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			DPU_ERROR("skip mixer %d with invalid id\n", lm->id);
 			continue;
 		}
-		hw = dpu_hw_lm_init(lm->id, mmio, cat);
+		hw = dpu_hw_lm_init(lm, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed lm object creation: err %d\n", rc);
@@ -143,7 +143,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			DPU_ERROR("skip merge_3d %d with invalid id\n", merge_3d->id);
 			continue;
 		}
-		hw = dpu_hw_merge_3d_init(merge_3d->id, mmio, cat);
+		hw = dpu_hw_merge_3d_init(merge_3d, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed merge_3d object creation: err %d\n",
@@ -161,7 +161,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			DPU_ERROR("skip pingpong %d with invalid id\n", pp->id);
 			continue;
 		}
-		hw = dpu_hw_pingpong_init(pp->id, mmio, cat);
+		hw = dpu_hw_pingpong_init(pp, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed pingpong object creation: err %d\n",
@@ -177,15 +177,11 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_intf *hw;
 		const struct dpu_intf_cfg *intf = &cat->intf[i];
 
-		if (intf->type == INTF_NONE) {
-			DPU_DEBUG("skip intf %d with type none\n", i);
-			continue;
-		}
 		if (intf->id < INTF_0 || intf->id >= INTF_MAX) {
 			DPU_ERROR("skip intf %d with invalid id\n", intf->id);
 			continue;
 		}
-		hw = dpu_hw_intf_init(intf->id, mmio, cat);
+		hw = dpu_hw_intf_init(intf, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed intf object creation: err %d\n", rc);
@@ -203,7 +199,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			continue;
 		}
 
-		hw = dpu_hw_wb_init(wb->id, mmio, cat);
+		hw = dpu_hw_wb_init(wb, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed wb object creation: err %d\n", rc);
@@ -220,7 +216,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			DPU_ERROR("skip ctl %d with invalid id\n", ctl->id);
 			continue;
 		}
-		hw = dpu_hw_ctl_init(ctl->id, mmio, cat);
+		hw = dpu_hw_ctl_init(ctl, mmio, cat->mixer_count, cat->mixer);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed ctl object creation: err %d\n", rc);
@@ -237,7 +233,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			DPU_ERROR("skip dspp %d with invalid id\n", dspp->id);
 			continue;
 		}
-		hw = dpu_hw_dspp_init(dspp->id, mmio, cat);
+		hw = dpu_hw_dspp_init(dspp, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed dspp object creation: err %d\n", rc);
@@ -250,7 +246,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_dsc *hw;
 		const struct dpu_dsc_cfg *dsc = &cat->dsc[i];
 
-		hw = dpu_hw_dsc_init(dsc->id, mmio, cat);
+		hw = dpu_hw_dsc_init(dsc, mmio);
 		if (IS_ERR_OR_NULL(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed dsc object creation: err %d\n", rc);
@@ -268,7 +264,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 			continue;
 		}
 
-		hw = dpu_hw_sspp_init(sspp->id, mmio, cat);
+		hw = dpu_hw_sspp_init(sspp, mmio, cat->ubwc);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed sspp object creation: err %d\n", rc);
-- 
GitLab


From 198d8c9c0abf0d952756480b020e3b453df01695 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Fri, 28 Apr 2023 15:36:43 -0700
Subject: [PATCH 0580/1791] drm/msm/dpu: remove DPU_DSPP_GC handling in dspp
 flush

Gamma correction blocks (GC) are not used today so lets remove
the usage of DPU_DSPP_GC in the dspp flush to make it easier
to remove GC from the catalog.

We can add this back when GC is properly supported in DPU with
one of the standard DRM properties.

changes in v3:
	- drop the link tag which was auto added before

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534723/
Link: https://lore.kernel.org/r/20230428223646.23595-1-quic_abhinavk@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 4f7cfa9a00f6c..cfdefacbc5a24 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -319,9 +319,6 @@ static void dpu_hw_ctl_update_pending_flush_dspp_sub_blocks(
 	case DPU_DSPP_PCC:
 		ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(4);
 		break;
-	case DPU_DSPP_GC:
-		ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(5);
-		break;
 	default:
 		return;
 	}
-- 
GitLab


From 7debb6bb699db088d7b3c14ccef634e7241b3831 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Fri, 28 Apr 2023 15:36:44 -0700
Subject: [PATCH 0581/1791] drm/msm/dpu: remove DPU_DSPP_IGC handling in dspp
 flush

Inverse gamma correction blocks (IGC) are not used today so lets
remove the usage of DPU_DSPP_IGC in the DSPP flush to make it easier
to remove IGC from the catalog.

We can add this back when IGC is properly supported in DPU with
one of the standard DRM properties.

changes in v3:
	- minor change dspp -> DSPP in commit text

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534724/
Link: https://lore.kernel.org/r/20230428223646.23595-2-quic_abhinavk@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index cfdefacbc5a24..07bcacedf4b02 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -313,9 +313,6 @@ static void dpu_hw_ctl_update_pending_flush_dspp_sub_blocks(
 		return;
 
 	switch (dspp_sub_blk) {
-	case DPU_DSPP_IGC:
-		ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(2);
-		break;
 	case DPU_DSPP_PCC:
 		ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(4);
 		break;
-- 
GitLab


From 9fd5aa5ae701a2c6fd7111e7365533c28439c00e Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Fri, 28 Apr 2023 15:36:45 -0700
Subject: [PATCH 0582/1791] drm/msm/dpu: remove GC and IGC related code from
 dpu catalog

Gamma Correction (GC) and Inverse Gamma Correction(IGC) is
currently unused. In addition dpu_dspp_sub_blks didn't even have an igc
member describing the block.

Drop related code from the dpu hardware catalog otherwise this becomes a
burden to carry across chipsets in the catalog.

changes in v3:
	- drop IGC related code from dpu_hw_catalog too
	- update commit text accordingly

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534725/
Link: https://lore.kernel.org/r/20230428223646.23595-3-quic_abhinavk@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c |  4 +---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 10 ----------
 2 files changed, 1 insertion(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index afe667121dc79..da3234a6f35a3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -91,7 +91,7 @@
 
 #define MERGE_3D_SM8150_MASK (0)
 
-#define DSPP_MSM8998_MASK BIT(DPU_DSPP_PCC) | BIT(DPU_DSPP_GC)
+#define DSPP_MSM8998_MASK BIT(DPU_DSPP_PCC)
 
 #define DSPP_SC7180_MASK BIT(DPU_DSPP_PCC)
 
@@ -449,8 +449,6 @@ static const struct dpu_lm_sub_blks qcm2290_lm_sblk = {
 static const struct dpu_dspp_sub_blks msm8998_dspp_sblk = {
 	.pcc = {.id = DPU_DSPP_PCC, .base = 0x1700,
 		.len = 0x90, .version = 0x10007},
-	.gc = { .id = DPU_DSPP_GC, .base = 0x17c0,
-		.len = 0x90, .version = 0x10007},
 };
 
 static const struct dpu_dspp_sub_blks sc7180_dspp_sblk = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 8d62c21b051a8..0dc9e7d585c79 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -127,13 +127,9 @@ enum {
 /**
  * DSPP sub-blocks
  * @DPU_DSPP_PCC             Panel color correction block
- * @DPU_DSPP_GC              Gamma correction block
- * @DPU_DSPP_IGC             Inverse gamma correction block
  */
 enum {
 	DPU_DSPP_PCC = 0x1,
-	DPU_DSPP_GC,
-	DPU_DSPP_IGC,
 	DPU_DSPP_MAX
 };
 
@@ -398,7 +394,6 @@ struct dpu_caps {
  * @hsic:
  * @memcolor:
  * @pcc_blk:
- * @igc_blk:
  * @format_list: Pointer to list of supported formats
  * @num_formats: Number of supported formats
  * @virt_format_list: Pointer to list of supported formats for virtual planes
@@ -419,7 +414,6 @@ struct dpu_sspp_sub_blks {
 	struct dpu_pp_blk hsic_blk;
 	struct dpu_pp_blk memcolor_blk;
 	struct dpu_pp_blk pcc_blk;
-	struct dpu_pp_blk igc_blk;
 
 	const u32 *format_list;
 	u32 num_formats;
@@ -433,22 +427,18 @@ struct dpu_sspp_sub_blks {
  * @maxwidth:               Max pixel width supported by this mixer
  * @maxblendstages:         Max number of blend-stages supported
  * @blendstage_base:        Blend-stage register base offset
- * @gc: gamma correction block
  */
 struct dpu_lm_sub_blks {
 	u32 maxwidth;
 	u32 maxblendstages;
 	u32 blendstage_base[MAX_BLOCKS];
-	struct dpu_pp_blk gc;
 };
 
 /**
  * struct dpu_dspp_sub_blks: Information of DSPP block
- * @gc : gamma correction block
  * @pcc: pixel color correction block
  */
 struct dpu_dspp_sub_blks {
-	struct dpu_pp_blk gc;
 	struct dpu_pp_blk pcc;
 };
 
-- 
GitLab


From 732c2c5298219bc21da3396bc09deb3589295349 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Fri, 28 Apr 2023 15:36:46 -0700
Subject: [PATCH 0583/1791] drm/msm/dpu: drop DSPP_MSM8998_MASK from hw catalog

Since GC and IGC masks have now been dropped, DSPP_MSM8998_MASK
is the same as DSPP_SC7180_MASK. Since DSPP_SC7180_MASK is used more
than DSPP_MSM8998_MASK, lets drop the latter.

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534726/
Link: https://lore.kernel.org/r/20230428223646.23595-4-quic_abhinavk@quicinc.com
[DB: fixed typo in commit message]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c          | 2 --
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 4cd8d7871c4a2..886aaf4b213af 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -127,9 +127,9 @@ static const struct dpu_pingpong_cfg msm8998_pp[] = {
 };
 
 static const struct dpu_dspp_cfg msm8998_dspp[] = {
-	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_MSM8998_MASK,
+	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
 		 &msm8998_dspp_sblk),
-	DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_MSM8998_MASK,
+	DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK,
 		 &msm8998_dspp_sblk),
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index da3234a6f35a3..9fa581f0b415e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -91,8 +91,6 @@
 
 #define MERGE_3D_SM8150_MASK (0)
 
-#define DSPP_MSM8998_MASK BIT(DPU_DSPP_PCC)
-
 #define DSPP_SC7180_MASK BIT(DPU_DSPP_PCC)
 
 #define INTF_SDM845_MASK (0)
-- 
GitLab


From 52788338110c06846cb62ad54569659b0cc59970 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 30 Apr 2023 00:25:12 +0300
Subject: [PATCH 0584/1791] drm/msm/dpu: drop unused SSPP sub-block information

The driver doesn't support hsic/memcolor and pcc SSPP subblocks.
Drop corresponding definitions.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534766/
Link: https://lore.kernel.org/r/20230429212512.2947245-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 0dc9e7d585c79..c6eb5a8a66da9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -391,9 +391,6 @@ struct dpu_caps {
  * @src_blk:
  * @scaler_blk:
  * @csc_blk:
- * @hsic:
- * @memcolor:
- * @pcc_blk:
  * @format_list: Pointer to list of supported formats
  * @num_formats: Number of supported formats
  * @virt_format_list: Pointer to list of supported formats for virtual planes
@@ -411,9 +408,6 @@ struct dpu_sspp_sub_blks {
 	struct dpu_src_blk src_blk;
 	struct dpu_scaler_blk scaler_blk;
 	struct dpu_pp_blk csc_blk;
-	struct dpu_pp_blk hsic_blk;
-	struct dpu_pp_blk memcolor_blk;
-	struct dpu_pp_blk pcc_blk;
 
 	const u32 *format_list;
 	u32 num_formats;
-- 
GitLab


From 3ce166380567d8385b776d7b72b603e0d2379e18 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Wed, 3 May 2023 15:39:05 -0700
Subject: [PATCH 0585/1791] drm/msm/dpu: add writeback support for sc7280

Add writeback support for sc7280. This was validated with kms_writeback
test case in IGT.

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/535244/
Link: https://lore.kernel.org/r/20230503223905.24754-1-quic_abhinavk@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index 6b2c7eae71d99..b4cf445b74bf3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -31,6 +31,7 @@ static const struct dpu_mdp_cfg sc7280_mdp[] = {
 	.clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 },
 	.clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 },
 	.clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 },
+	.clk_ctrls[DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 },
 	},
 };
 
@@ -93,6 +94,11 @@ static const struct dpu_pingpong_cfg sc7280_pp[] = {
 	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, -1, -1),
 };
 
+static const struct dpu_wb_cfg sc7280_wb[] = {
+	WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6,
+			VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4),
+};
+
 static const struct dpu_intf_cfg sc7280_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
 	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
@@ -142,6 +148,8 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = {
 	.mixer = sc7280_lm,
 	.pingpong_count = ARRAY_SIZE(sc7280_pp),
 	.pingpong = sc7280_pp,
+	.wb_count = ARRAY_SIZE(sc7280_wb),
+	.wb = sc7280_wb,
 	.intf_count = ARRAY_SIZE(sc7280_intf),
 	.intf = sc7280_intf,
 	.vbif_count = ARRAY_SIZE(sdm845_vbif),
-- 
GitLab


From 853b292b92d50d1b30628c92229e93d670304e51 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 18 May 2023 23:29:08 +0200
Subject: [PATCH 0586/1791] drm/msm/dpu: Use V4.0 PCC DSPP sub-block in
 SC7[12]80

According to various downstream sources the PCC sub-block inside DSPP is
version 4.0 since DPU 4.0 and higher, including SC7[12]80 at DPU version
6.2 and 7.2 respectively.  After correcting the version this struct
becomes identical to sm8150_dspp_sblk which is used all across the
catalog: replace uses of sc7180_dspp_sblk with that and remove
the struct definition for sc7180_dspp_sblk entirely.

Fixes: 4259ff7ae509e ("drm/msm/dpu: add support for pcc color block in dpu driver")
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537899/
Link: https://lore.kernel.org/r/20230518-dpu-sc7180-pcc-version-v1-1-ec9ca4949e3e@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c         | 5 -----
 3 files changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 3ab5fbada08fa..911479cb1675c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -76,7 +76,7 @@ static const struct dpu_lm_cfg sc7180_lm[] = {
 
 static const struct dpu_dspp_cfg sc7180_dspp[] = {
 	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
-		 &sc7180_dspp_sblk),
+		 &sm8150_dspp_sblk),
 };
 
 static const struct dpu_pingpong_cfg sc7180_pp[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index b4cf445b74bf3..ec9047cd60fd5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -84,7 +84,7 @@ static const struct dpu_lm_cfg sc7280_lm[] = {
 
 static const struct dpu_dspp_cfg sc7280_dspp[] = {
 	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
-		 &sc7180_dspp_sblk),
+		 &sm8150_dspp_sblk),
 };
 
 static const struct dpu_pingpong_cfg sc7280_pp[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 9fa581f0b415e..da758510fbfee 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -449,11 +449,6 @@ static const struct dpu_dspp_sub_blks msm8998_dspp_sblk = {
 		.len = 0x90, .version = 0x10007},
 };
 
-static const struct dpu_dspp_sub_blks sc7180_dspp_sblk = {
-	.pcc = {.id = DPU_DSPP_PCC, .base = 0x1700,
-		.len = 0x90, .version = 0x10000},
-};
-
 static const struct dpu_dspp_sub_blks sm8150_dspp_sblk = {
 	.pcc = {.id = DPU_DSPP_PCC, .base = 0x1700,
 		.len = 0x90, .version = 0x40000},
-- 
GitLab


From d9388af508436091adc1bce492caebab4586b2dd Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:23 +0200
Subject: [PATCH 0587/1791] drm/msm/dpu: Remove extraneous register define
 indentation

A bunch of registers are indented with two extra spaces, looking as if
these are values corresponding to the previous register which is not the
case, rather these are simply also register offsets and should only have
a single space separating them and the #define keyword.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/534218/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-9-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 41 +++++++++++----------
 1 file changed, 21 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index f0271283a3a71..24ede498799ff 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -38,26 +38,27 @@
 #define INTF_ACTIVE_DATA_HCTL           0x068
 #define INTF_FRAME_LINE_COUNT_EN        0x0A8
 #define INTF_FRAME_COUNT                0x0AC
-#define   INTF_LINE_COUNT               0x0B0
-
-#define   INTF_DEFLICKER_CONFIG         0x0F0
-#define   INTF_DEFLICKER_STRNG_COEFF    0x0F4
-#define   INTF_DEFLICKER_WEAK_COEFF     0x0F8
-
-#define   INTF_DSI_CMD_MODE_TRIGGER_EN  0x084
-#define   INTF_PANEL_FORMAT             0x090
-#define   INTF_TPG_ENABLE               0x100
-#define   INTF_TPG_MAIN_CONTROL         0x104
-#define   INTF_TPG_VIDEO_CONFIG         0x108
-#define   INTF_TPG_COMPONENT_LIMITS     0x10C
-#define   INTF_TPG_RECTANGLE            0x110
-#define   INTF_TPG_INITIAL_VALUE        0x114
-#define   INTF_TPG_BLK_WHITE_PATTERN_FRAMES   0x118
-#define   INTF_TPG_RGB_MAPPING          0x11C
-#define   INTF_PROG_FETCH_START         0x170
-#define   INTF_PROG_ROT_START           0x174
-#define   INTF_MUX                      0x25C
-#define   INTF_STATUS                   0x26C
+#define INTF_LINE_COUNT                 0x0B0
+
+#define INTF_DEFLICKER_CONFIG           0x0F0
+#define INTF_DEFLICKER_STRNG_COEFF      0x0F4
+#define INTF_DEFLICKER_WEAK_COEFF       0x0F8
+
+#define INTF_DSI_CMD_MODE_TRIGGER_EN    0x084
+#define INTF_PANEL_FORMAT               0x090
+#define INTF_TPG_ENABLE                 0x100
+#define INTF_TPG_MAIN_CONTROL           0x104
+#define INTF_TPG_VIDEO_CONFIG           0x108
+#define INTF_TPG_COMPONENT_LIMITS       0x10C
+#define INTF_TPG_RECTANGLE              0x110
+#define INTF_TPG_INITIAL_VALUE          0x114
+#define INTF_TPG_BLK_WHITE_PATTERN_FRAMES 0x118
+#define INTF_TPG_RGB_MAPPING            0x11C
+#define INTF_PROG_FETCH_START           0x170
+#define INTF_PROG_ROT_START             0x174
+
+#define INTF_MUX                        0x25C
+#define INTF_STATUS                     0x26C
 
 #define INTF_CFG_ACTIVE_H_EN	BIT(29)
 #define INTF_CFG_ACTIVE_V_EN	BIT(30)
-- 
GitLab


From d80d7f8912a8cbec16ec55793f4b20fc5e76640f Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:24 +0200
Subject: [PATCH 0588/1791] drm/msm/dpu: Sort INTF registers numerically

A bunch of registers were appended at the end in e.g. commit
91143873a05d ("drm/msm/dpu: Add MISR register support for interface")
rather than being inserted in a place that maintains numerical sorting:
restore said numerical sorting.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534213/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-10-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index 24ede498799ff..0bd99f380a498 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -36,6 +36,10 @@
 #define INTF_CONFIG2                    0x060
 #define INTF_DISPLAY_DATA_HCTL          0x064
 #define INTF_ACTIVE_DATA_HCTL           0x068
+
+#define INTF_DSI_CMD_MODE_TRIGGER_EN    0x084
+#define INTF_PANEL_FORMAT               0x090
+
 #define INTF_FRAME_LINE_COUNT_EN        0x0A8
 #define INTF_FRAME_COUNT                0x0AC
 #define INTF_LINE_COUNT                 0x0B0
@@ -44,8 +48,6 @@
 #define INTF_DEFLICKER_STRNG_COEFF      0x0F4
 #define INTF_DEFLICKER_WEAK_COEFF       0x0F8
 
-#define INTF_DSI_CMD_MODE_TRIGGER_EN    0x084
-#define INTF_PANEL_FORMAT               0x090
 #define INTF_TPG_ENABLE                 0x100
 #define INTF_TPG_MAIN_CONTROL           0x104
 #define INTF_TPG_VIDEO_CONFIG           0x108
@@ -57,6 +59,9 @@
 #define INTF_PROG_FETCH_START           0x170
 #define INTF_PROG_ROT_START             0x174
 
+#define INTF_MISR_CTRL                  0x180
+#define INTF_MISR_SIGNATURE             0x184
+
 #define INTF_MUX                        0x25C
 #define INTF_STATUS                     0x26C
 
@@ -66,8 +71,6 @@
 #define INTF_CFG2_DATABUS_WIDEN	BIT(0)
 #define INTF_CFG2_DATA_HCTL_EN	BIT(4)
 
-#define INTF_MISR_CTRL			0x180
-#define INTF_MISR_SIGNATURE		0x184
 
 static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx,
 		const struct intf_timing_params *p,
-- 
GitLab


From f0408020bfdaeef168e46eead3cbcfc7fd16f314 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:25 +0200
Subject: [PATCH 0589/1791] drm/msm/dpu: Take INTF index as parameter in
 interrupt register defines

Instead of hardcoding many register defines for every INTF and AD4 index
with a fixed stride, turn the defines into singular chunks of math that
compute the address using the base and this fixed stride multiplied by
the index given as argument to the definitions.

MDP_SSPP_TOP0_OFF is dropped as that constant is zero anyway, and all
register offsets related to it live in dpu_hwio.h.

Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534221/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-11-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 156 ++++++++----------
 1 file changed, 72 insertions(+), 84 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
index 17f3e7e4f1941..152d4272a087a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
@@ -17,30 +17,18 @@
  * Register offsets in MDSS register file for the interrupt registers
  * w.r.t. the MDP base
  */
-#define MDP_SSPP_TOP0_OFF		0x0
-#define MDP_INTF_0_OFF			0x6A000
-#define MDP_INTF_1_OFF			0x6A800
-#define MDP_INTF_2_OFF			0x6B000
-#define MDP_INTF_3_OFF			0x6B800
-#define MDP_INTF_4_OFF			0x6C000
-#define MDP_INTF_5_OFF			0x6C800
-#define INTF_INTR_EN			0x1c0
-#define INTF_INTR_STATUS		0x1c4
-#define INTF_INTR_CLEAR			0x1c8
-#define MDP_AD4_0_OFF			0x7C000
-#define MDP_AD4_1_OFF			0x7D000
-#define MDP_AD4_INTR_EN_OFF		0x41c
-#define MDP_AD4_INTR_CLEAR_OFF		0x424
-#define MDP_AD4_INTR_STATUS_OFF		0x420
-#define MDP_INTF_0_OFF_REV_7xxx		0x34000
-#define MDP_INTF_1_OFF_REV_7xxx		0x35000
-#define MDP_INTF_2_OFF_REV_7xxx		0x36000
-#define MDP_INTF_3_OFF_REV_7xxx		0x37000
-#define MDP_INTF_4_OFF_REV_7xxx		0x38000
-#define MDP_INTF_5_OFF_REV_7xxx		0x39000
-#define MDP_INTF_6_OFF_REV_7xxx		0x3a000
-#define MDP_INTF_7_OFF_REV_7xxx		0x3b000
-#define MDP_INTF_8_OFF_REV_7xxx		0x3c000
+#define MDP_INTF_OFF(intf)				(0x6A000 + 0x800 * (intf))
+#define MDP_INTF_INTR_EN(intf)				(MDP_INTF_OFF(intf) + 0x1c0)
+#define MDP_INTF_INTR_STATUS(intf)			(MDP_INTF_OFF(intf) + 0x1c4)
+#define MDP_INTF_INTR_CLEAR(intf)			(MDP_INTF_OFF(intf) + 0x1c8)
+#define MDP_AD4_OFF(ad4)				(0x7C000 + 0x1000 * (ad4))
+#define MDP_AD4_INTR_EN_OFF(ad4)			(MDP_AD4_OFF(ad4) + 0x41c)
+#define MDP_AD4_INTR_CLEAR_OFF(ad4)			(MDP_AD4_OFF(ad4) + 0x424)
+#define MDP_AD4_INTR_STATUS_OFF(ad4)			(MDP_AD4_OFF(ad4) + 0x420)
+#define MDP_INTF_REV_7xxx_OFF(intf)			(0x34000 + 0x1000 * (intf))
+#define MDP_INTF_REV_7xxx_INTR_EN(intf)			(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c0)
+#define MDP_INTF_REV_7xxx_INTR_STATUS(intf)		(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c4)
+#define MDP_INTF_REV_7xxx_INTR_CLEAR(intf)		(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c8)
 
 /**
  * struct dpu_intr_reg - array of DPU register sets
@@ -61,104 +49,104 @@ struct dpu_intr_reg {
  */
 static const struct dpu_intr_reg dpu_intr_set[] = {
 	[MDP_SSPP_TOP0_INTR] = {
-		MDP_SSPP_TOP0_OFF+INTR_CLEAR,
-		MDP_SSPP_TOP0_OFF+INTR_EN,
-		MDP_SSPP_TOP0_OFF+INTR_STATUS
+		INTR_CLEAR,
+		INTR_EN,
+		INTR_STATUS
 	},
 	[MDP_SSPP_TOP0_INTR2] = {
-		MDP_SSPP_TOP0_OFF+INTR2_CLEAR,
-		MDP_SSPP_TOP0_OFF+INTR2_EN,
-		MDP_SSPP_TOP0_OFF+INTR2_STATUS
+		INTR2_CLEAR,
+		INTR2_EN,
+		INTR2_STATUS
 	},
 	[MDP_SSPP_TOP0_HIST_INTR] = {
-		MDP_SSPP_TOP0_OFF+HIST_INTR_CLEAR,
-		MDP_SSPP_TOP0_OFF+HIST_INTR_EN,
-		MDP_SSPP_TOP0_OFF+HIST_INTR_STATUS
+		HIST_INTR_CLEAR,
+		HIST_INTR_EN,
+		HIST_INTR_STATUS
 	},
 	[MDP_INTF0_INTR] = {
-		MDP_INTF_0_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_0_OFF+INTF_INTR_EN,
-		MDP_INTF_0_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(0),
+		MDP_INTF_INTR_EN(0),
+		MDP_INTF_INTR_STATUS(0)
 	},
 	[MDP_INTF1_INTR] = {
-		MDP_INTF_1_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_1_OFF+INTF_INTR_EN,
-		MDP_INTF_1_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(1),
+		MDP_INTF_INTR_EN(1),
+		MDP_INTF_INTR_STATUS(1)
 	},
 	[MDP_INTF2_INTR] = {
-		MDP_INTF_2_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_2_OFF+INTF_INTR_EN,
-		MDP_INTF_2_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(2),
+		MDP_INTF_INTR_EN(2),
+		MDP_INTF_INTR_STATUS(2)
 	},
 	[MDP_INTF3_INTR] = {
-		MDP_INTF_3_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_3_OFF+INTF_INTR_EN,
-		MDP_INTF_3_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(3),
+		MDP_INTF_INTR_EN(3),
+		MDP_INTF_INTR_STATUS(3)
 	},
 	[MDP_INTF4_INTR] = {
-		MDP_INTF_4_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_4_OFF+INTF_INTR_EN,
-		MDP_INTF_4_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(4),
+		MDP_INTF_INTR_EN(4),
+		MDP_INTF_INTR_STATUS(4)
 	},
 	[MDP_INTF5_INTR] = {
-		MDP_INTF_5_OFF+INTF_INTR_CLEAR,
-		MDP_INTF_5_OFF+INTF_INTR_EN,
-		MDP_INTF_5_OFF+INTF_INTR_STATUS
+		MDP_INTF_INTR_CLEAR(5),
+		MDP_INTF_INTR_EN(5),
+		MDP_INTF_INTR_STATUS(5)
 	},
 	[MDP_AD4_0_INTR] = {
-		MDP_AD4_0_OFF + MDP_AD4_INTR_CLEAR_OFF,
-		MDP_AD4_0_OFF + MDP_AD4_INTR_EN_OFF,
-		MDP_AD4_0_OFF + MDP_AD4_INTR_STATUS_OFF,
+		MDP_AD4_INTR_CLEAR_OFF(0),
+		MDP_AD4_INTR_EN_OFF(0),
+		MDP_AD4_INTR_STATUS_OFF(0),
 	},
 	[MDP_AD4_1_INTR] = {
-		MDP_AD4_1_OFF + MDP_AD4_INTR_CLEAR_OFF,
-		MDP_AD4_1_OFF + MDP_AD4_INTR_EN_OFF,
-		MDP_AD4_1_OFF + MDP_AD4_INTR_STATUS_OFF,
+		MDP_AD4_INTR_CLEAR_OFF(1),
+		MDP_AD4_INTR_EN_OFF(1),
+		MDP_AD4_INTR_STATUS_OFF(1),
 	},
 	[MDP_INTF0_7xxx_INTR] = {
-		MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(0),
+		MDP_INTF_REV_7xxx_INTR_EN(0),
+		MDP_INTF_REV_7xxx_INTR_STATUS(0)
 	},
 	[MDP_INTF1_7xxx_INTR] = {
-		MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(1),
+		MDP_INTF_REV_7xxx_INTR_EN(1),
+		MDP_INTF_REV_7xxx_INTR_STATUS(1)
 	},
 	[MDP_INTF2_7xxx_INTR] = {
-		MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(2),
+		MDP_INTF_REV_7xxx_INTR_EN(2),
+		MDP_INTF_REV_7xxx_INTR_STATUS(2)
 	},
 	[MDP_INTF3_7xxx_INTR] = {
-		MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(3),
+		MDP_INTF_REV_7xxx_INTR_EN(3),
+		MDP_INTF_REV_7xxx_INTR_STATUS(3)
 	},
 	[MDP_INTF4_7xxx_INTR] = {
-		MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(4),
+		MDP_INTF_REV_7xxx_INTR_EN(4),
+		MDP_INTF_REV_7xxx_INTR_STATUS(4)
 	},
 	[MDP_INTF5_7xxx_INTR] = {
-		MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(5),
+		MDP_INTF_REV_7xxx_INTR_EN(5),
+		MDP_INTF_REV_7xxx_INTR_STATUS(5)
 	},
 	[MDP_INTF6_7xxx_INTR] = {
-		MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(6),
+		MDP_INTF_REV_7xxx_INTR_EN(6),
+		MDP_INTF_REV_7xxx_INTR_STATUS(6)
 	},
 	[MDP_INTF7_7xxx_INTR] = {
-		MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(7),
+		MDP_INTF_REV_7xxx_INTR_EN(7),
+		MDP_INTF_REV_7xxx_INTR_STATUS(7)
 	},
 	[MDP_INTF8_7xxx_INTR] = {
-		MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_CLEAR,
-		MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_EN,
-		MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_STATUS
+		MDP_INTF_REV_7xxx_INTR_CLEAR(8),
+		MDP_INTF_REV_7xxx_INTR_EN(8),
+		MDP_INTF_REV_7xxx_INTR_STATUS(8)
 	},
 };
 
-- 
GitLab


From a2623e72c52b2cf258b34675a8ff38c66e7d26fb Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:26 +0200
Subject: [PATCH 0590/1791] drm/msm/dpu: Drop unused poll_timeout_wr_ptr
 PINGPONG callback

This callback was migrated from downstream when DPU1 was first
introduced to mainline, but never used by any component.  Drop it to
save some lines and unnecessary confusion.

Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534215/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-12-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c    | 18 ------------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h    |  6 ------
 2 files changed, 24 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index 3822e0657767d..1dcbb75b80d3d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -126,23 +126,6 @@ static bool dpu_hw_pp_get_autorefresh_config(struct dpu_hw_pingpong *pp,
 	return !!((val & BIT(31)) >> 31);
 }
 
-static int dpu_hw_pp_poll_timeout_wr_ptr(struct dpu_hw_pingpong *pp,
-		u32 timeout_us)
-{
-	struct dpu_hw_blk_reg_map *c;
-	u32 val;
-	int rc;
-
-	if (!pp)
-		return -EINVAL;
-
-	c = &pp->hw;
-	rc = readl_poll_timeout(c->blk_addr + PP_LINE_COUNT,
-			val, (val & 0xffff) >= 1, 10, timeout_us);
-
-	return rc;
-}
-
 static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, bool enable)
 {
 	struct dpu_hw_blk_reg_map *c;
@@ -262,7 +245,6 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 	c->ops.get_vsync_info = dpu_hw_pp_get_vsync_info;
 	c->ops.setup_autorefresh = dpu_hw_pp_setup_autorefresh_config;
 	c->ops.get_autorefresh = dpu_hw_pp_get_autorefresh_config;
-	c->ops.poll_timeout_wr_ptr = dpu_hw_pp_poll_timeout_wr_ptr;
 	c->ops.get_line_count = dpu_hw_pp_get_line_count;
 	c->ops.setup_dsc = dpu_hw_pp_setup_dsc;
 	c->ops.enable_dsc = dpu_hw_pp_dsc_enable;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
index 51e9093621a7e..42af9e5b865e2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
@@ -107,12 +107,6 @@ struct dpu_hw_pingpong_ops {
 	bool (*get_autorefresh)(struct dpu_hw_pingpong *pp,
 				u32 *frame_count);
 
-	/**
-	 * poll until write pointer transmission starts
-	 * @Return: 0 on success, -ETIMEDOUT on timeout
-	 */
-	int (*poll_timeout_wr_ptr)(struct dpu_hw_pingpong *pp, u32 timeout_us);
-
 	/**
 	 * Obtain current vertical line counter
 	 */
-- 
GitLab


From 4a7c38ec7d8efe96c4d8b4d5fc0efb5b06db58e9 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:27 +0200
Subject: [PATCH 0591/1791] drm/msm/dpu: Move autorefresh disable from CMD
 encoder to pingpong

This autorefresh disable logic in the physical command-mode encoder
consumes three callbacks to the pingpong block, and will explode in
unnecessary complexity when the same callbacks need to be called on the
interface block instead to accommodate INTF TE support.  To clean this
up, move the logic into the pingpong block under a disable_autorefresh
callback, replacing the aforementioned three get_autorefresh,
setup_autorefresh and get_vsync_info callbacks.

The same logic will have to be replicated to the interface block when it
receives INTF TE support, but it is less complex than constantly
switching on a "has_intf_te" boolean to choose a callback.

Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534230/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-13-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c  | 60 ++-----------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c   | 47 ++++++++++++++-
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h   | 25 ++------
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h       |  4 ++
 4 files changed, 57 insertions(+), 79 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 74470d068622e..a60fb8d3736b5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -36,10 +36,6 @@
 #define DEFAULT_TEARCHECK_SYNC_THRESH_START	4
 #define DEFAULT_TEARCHECK_SYNC_THRESH_CONTINUE	4
 
-#define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000
-
-#define DPU_ENC_MAX_POLL_TIMEOUT_US	2000
-
 static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc);
 
 static bool dpu_encoder_phys_cmd_is_master(struct dpu_encoder_phys *phys_enc)
@@ -574,28 +570,8 @@ static void dpu_encoder_phys_cmd_prepare_for_kickoff(
 			atomic_read(&phys_enc->pending_kickoff_cnt));
 }
 
-static bool dpu_encoder_phys_cmd_is_ongoing_pptx(
-		struct dpu_encoder_phys *phys_enc)
-{
-	struct dpu_hw_pp_vsync_info info;
-
-	if (!phys_enc)
-		return false;
-
-	phys_enc->hw_pp->ops.get_vsync_info(phys_enc->hw_pp, &info);
-	if (info.wr_ptr_line_count > 0 &&
-	    info.wr_ptr_line_count < phys_enc->cached_mode.vdisplay)
-		return true;
-
-	return false;
-}
-
 static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc)
 {
-	struct dpu_encoder_phys_cmd *cmd_enc =
-		to_dpu_encoder_phys_cmd(phys_enc);
-	int trial = 0;
-
 	if (!phys_enc)
 		return;
 	if (!phys_enc->hw_pp)
@@ -603,37 +579,11 @@ static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc)
 	if (!dpu_encoder_phys_cmd_is_master(phys_enc))
 		return;
 
-	/* If autorefresh is already disabled, we have nothing to do */
-	if (!phys_enc->hw_pp->ops.get_autorefresh(phys_enc->hw_pp, NULL))
-		return;
-
-	/*
-	 * If autorefresh is enabled, disable it and make sure it is safe to
-	 * proceed with current frame commit/push. Sequence fallowed is,
-	 * 1. Disable TE
-	 * 2. Disable autorefresh config
-	 * 4. Poll for frame transfer ongoing to be false
-	 * 5. Enable TE back
-	 */
-	_dpu_encoder_phys_cmd_connect_te(phys_enc, false);
-	phys_enc->hw_pp->ops.setup_autorefresh(phys_enc->hw_pp, 0, false);
-
-	do {
-		udelay(DPU_ENC_MAX_POLL_TIMEOUT_US);
-		if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US)
-				> (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) {
-			DPU_ERROR_CMDENC(cmd_enc,
-					"disable autorefresh failed\n");
-			break;
-		}
-
-		trial++;
-	} while (dpu_encoder_phys_cmd_is_ongoing_pptx(phys_enc));
-
-	_dpu_encoder_phys_cmd_connect_te(phys_enc, true);
-
-	DPU_DEBUG_CMDENC(to_dpu_encoder_phys_cmd(phys_enc),
-			 "disabled autorefresh\n");
+	if (phys_enc->hw_pp->ops.disable_autorefresh) {
+		phys_enc->hw_pp->ops.disable_autorefresh(phys_enc->hw_pp,
+							 DRMID(phys_enc->parent),
+							 phys_enc->cached_mode.vdisplay);
+	}
 }
 
 static int _dpu_encoder_phys_cmd_wait_for_ctl_start(
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index 1dcbb75b80d3d..e366d752bd073 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -210,6 +210,49 @@ static u32 dpu_hw_pp_get_line_count(struct dpu_hw_pingpong *pp)
 	return line;
 }
 
+static void dpu_hw_pp_disable_autorefresh(struct dpu_hw_pingpong *pp,
+					  uint32_t encoder_id, u16 vdisplay)
+{
+	struct dpu_hw_pp_vsync_info info;
+	int trial = 0;
+
+	/* If autorefresh is already disabled, we have nothing to do */
+	if (!dpu_hw_pp_get_autorefresh_config(pp, NULL))
+		return;
+
+	/*
+	 * If autorefresh is enabled, disable it and make sure it is safe to
+	 * proceed with current frame commit/push. Sequence followed is,
+	 * 1. Disable TE
+	 * 2. Disable autorefresh config
+	 * 4. Poll for frame transfer ongoing to be false
+	 * 5. Enable TE back
+	 */
+
+	dpu_hw_pp_connect_external_te(pp, false);
+	dpu_hw_pp_setup_autorefresh_config(pp, 0, false);
+
+	do {
+		udelay(DPU_ENC_MAX_POLL_TIMEOUT_US);
+		if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US)
+				> (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) {
+			DPU_ERROR("enc%d pp%d disable autorefresh failed\n",
+				  encoder_id, pp->idx - PINGPONG_0);
+			break;
+		}
+
+		trial++;
+
+		dpu_hw_pp_get_vsync_info(pp, &info);
+	} while (info.wr_ptr_line_count > 0 &&
+		 info.wr_ptr_line_count < vdisplay);
+
+	dpu_hw_pp_connect_external_te(pp, true);
+
+	DPU_DEBUG("enc%d pp%d disabled autorefresh\n",
+		  encoder_id, pp->idx - PINGPONG_0);
+}
+
 static int dpu_hw_pp_dsc_enable(struct dpu_hw_pingpong *pp)
 {
 	struct dpu_hw_blk_reg_map *c = &pp->hw;
@@ -242,10 +285,8 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 	c->ops.setup_tearcheck = dpu_hw_pp_setup_te_config;
 	c->ops.enable_tearcheck = dpu_hw_pp_enable_te;
 	c->ops.connect_external_te = dpu_hw_pp_connect_external_te;
-	c->ops.get_vsync_info = dpu_hw_pp_get_vsync_info;
-	c->ops.setup_autorefresh = dpu_hw_pp_setup_autorefresh_config;
-	c->ops.get_autorefresh = dpu_hw_pp_get_autorefresh_config;
 	c->ops.get_line_count = dpu_hw_pp_get_line_count;
+	c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh;
 	c->ops.setup_dsc = dpu_hw_pp_setup_dsc;
 	c->ops.enable_dsc = dpu_hw_pp_dsc_enable;
 	c->ops.disable_dsc = dpu_hw_pp_dsc_disable;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
index 42af9e5b865e2..336d383f8b647 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
@@ -61,9 +61,6 @@ struct dpu_hw_dither_cfg {
  *  Assumption is these functions will be called after clocks are enabled
  *  @setup_tearcheck : program tear check values
  *  @enable_tearcheck : enables tear check
- *  @get_vsync_info : retries timing info of the panel
- *  @setup_autorefresh : configure and enable the autorefresh config
- *  @get_autorefresh : retrieve autorefresh config from hardware
  *  @setup_dither : function to program the dither hw block
  *  @get_line_count: obtain current vertical line counter
  */
@@ -89,28 +86,14 @@ struct dpu_hw_pingpong_ops {
 			bool enable_external_te);
 
 	/**
-	 * provides the programmed and current
-	 * line_count
-	 */
-	int (*get_vsync_info)(struct dpu_hw_pingpong *pp,
-			struct dpu_hw_pp_vsync_info  *info);
-
-	/**
-	 * configure and enable the autorefresh config
-	 */
-	void (*setup_autorefresh)(struct dpu_hw_pingpong *pp,
-				  u32 frame_count, bool enable);
-
-	/**
-	 * retrieve autorefresh config from hardware
+	 * Obtain current vertical line counter
 	 */
-	bool (*get_autorefresh)(struct dpu_hw_pingpong *pp,
-				u32 *frame_count);
+	u32 (*get_line_count)(struct dpu_hw_pingpong *pp);
 
 	/**
-	 * Obtain current vertical line counter
+	 * Disable autorefresh if enabled
 	 */
-	u32 (*get_line_count)(struct dpu_hw_pingpong *pp);
+	void (*disable_autorefresh)(struct dpu_hw_pingpong *pp, uint32_t encoder_id, u16 vdisplay);
 
 	/**
 	 * Setup dither matix for pingpong block
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index 15111e433f219..cd62875245575 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -118,6 +118,10 @@ struct vsync_info {
 	u32 line_count;
 };
 
+#define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000
+
+#define DPU_ENC_MAX_POLL_TIMEOUT_US	2000
+
 #define to_dpu_kms(x) container_of(x, struct dpu_kms, base)
 
 #define to_dpu_global_state(x) container_of(x, struct dpu_global_state, base)
-- 
GitLab


From fe9d66cf6e8d2f6651e55ae57026464a080c2f2a Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:28 +0200
Subject: [PATCH 0592/1791] drm/msm/dpu: Disable pingpong TE on DPU 5.0.0 and
 above

Since hardware revision 5.0.0 the TE configuration moved out of the
PINGPONG block into the INTF block.  Writing these registers has no
effect, and is omitted downstream via the DPU/SDE_PINGPONG_TE feature
flag.  This flag is only added to PINGPONG blocks used by hardware prior
to 5.0.0.

The existing PP_BLK_TE macro has been removed in favour of directly
passing this feature flag, which has thus far been the only difference
with PP_BLK.  PP_BLK_DITHER has been left in place as its embedded
feature flag already excludes this DPU_PINGPONG_TE bit and differs by
setting the block length to zero, as it only contains a DITHER subblock.

The code that writes to these registers in the INTF block will follow in
subsequent patches.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534240/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-14-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_3_0_msm8998.h   |  8 +++----
 .../msm/disp/dpu1/catalog/dpu_4_0_sdm845.h    |  8 +++----
 .../msm/disp/dpu1/catalog/dpu_5_0_sm8150.h    | 12 +++++-----
 .../msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h   | 12 +++++-----
 .../msm/disp/dpu1/catalog/dpu_6_0_sm8250.h    | 12 +++++-----
 .../msm/disp/dpu1/catalog/dpu_6_2_sc7180.h    |  4 ++--
 .../msm/disp/dpu1/catalog/dpu_6_3_sm6115.h    |  2 +-
 .../msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h   |  2 +-
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    | 22 +++++++------------
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c   | 12 +++++-----
 10 files changed, 45 insertions(+), 49 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 886aaf4b213af..3d01840ef64b5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -112,16 +112,16 @@ static const struct dpu_lm_cfg msm8998_lm[] = {
 };
 
 static const struct dpu_pingpong_cfg msm8998_pp[] = {
-	PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk_te,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
-	PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk_te,
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
-	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
-	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index 63009435e2588..84159f8cbdaeb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -110,16 +110,16 @@ static const struct dpu_lm_cfg sdm845_lm[] = {
 };
 
 static const struct dpu_pingpong_cfg sdm845_pp[] = {
-	PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk_te,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
-	PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk_te,
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
-	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
-	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index b3f67710a48de..266c525f8daaf 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -128,22 +128,22 @@ static const struct dpu_dspp_cfg sm8150_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sm8150_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
-	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
-	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
-	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
-	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
-	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31),
 			-1),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index 1aef11a258346..b05bd561e1403 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -127,22 +127,22 @@ static const struct dpu_dspp_cfg sc8180x_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sc8180x_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
-	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
-	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
-	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
-	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
-	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31),
 			-1),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index 903efc580aef0..8660d04d0f589 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -129,22 +129,22 @@ static const struct dpu_dspp_cfg sm8250_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sm8250_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
-	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk,
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
-	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
-	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk,
+	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
-	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
-	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk,
+	PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31),
 			-1),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 911479cb1675c..63056439d4127 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -80,8 +80,8 @@ static const struct dpu_dspp_cfg sc7180_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sc7180_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk, -1, -1),
-	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk, -1, -1),
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, -1, -1),
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, -1, -1),
 };
 
 static const struct dpu_intf_cfg sc7180_intf[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index 988d820f7ef2e..e15dc96f1286a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -60,7 +60,7 @@ static const struct dpu_dspp_cfg sm6115_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sm6115_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index c9003dcc1a59b..2ff98ef6999fe 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -57,7 +57,7 @@ static const struct dpu_dspp_cfg qcm2290_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg qcm2290_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk,
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index da758510fbfee..fad12ba0eab6a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -75,11 +75,15 @@
 #define MIXER_QCM2290_MASK \
 	(BIT(DPU_DIM_LAYER) | BIT(DPU_MIXER_COMBINED_ALPHA))
 
-#define PINGPONG_SDM845_MASK BIT(DPU_PINGPONG_DITHER)
+#define PINGPONG_SDM845_MASK \
+	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_TE))
 
-#define PINGPONG_SDM845_SPLIT_MASK \
+#define PINGPONG_SDM845_TE2_MASK \
 	(PINGPONG_SDM845_MASK | BIT(DPU_PINGPONG_TE2))
 
+#define PINGPONG_SM8150_MASK \
+	(BIT(DPU_PINGPONG_DITHER))
+
 #define CTL_SC7280_MASK \
 	(BIT(DPU_CTL_ACTIVE_CFG) | \
 	 BIT(DPU_CTL_FETCH_ACTIVE) | \
@@ -492,21 +496,11 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 	.intr_done = _done, \
 	.intr_rdptr = _rdptr, \
 	}
-#define PP_BLK_TE(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \
-	{\
-	.name = _name, .id = _id, \
-	.base = _base, .len = 0xd4, \
-	.features = PINGPONG_SDM845_SPLIT_MASK, \
-	.merge_3d = _merge_3d, \
-	.sblk = &_sblk, \
-	.intr_done = _done, \
-	.intr_rdptr = _rdptr, \
-	}
-#define PP_BLK(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \
+#define PP_BLK(_name, _id, _base, _features, _merge_3d, _sblk, _done, _rdptr) \
 	{\
 	.name = _name, .id = _id, \
 	.base = _base, .len = 0xd4, \
-	.features = PINGPONG_SDM845_MASK, \
+	.features = _features, \
 	.merge_3d = _merge_3d, \
 	.sblk = &_sblk, \
 	.intr_done = _done, \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index e366d752bd073..c70f6ff018f23 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -282,11 +282,13 @@ static int dpu_hw_pp_setup_dsc(struct dpu_hw_pingpong *pp)
 static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 				unsigned long features)
 {
-	c->ops.setup_tearcheck = dpu_hw_pp_setup_te_config;
-	c->ops.enable_tearcheck = dpu_hw_pp_enable_te;
-	c->ops.connect_external_te = dpu_hw_pp_connect_external_te;
-	c->ops.get_line_count = dpu_hw_pp_get_line_count;
-	c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh;
+	if (test_bit(DPU_PINGPONG_TE, &features)) {
+		c->ops.setup_tearcheck = dpu_hw_pp_setup_te_config;
+		c->ops.enable_tearcheck = dpu_hw_pp_enable_te;
+		c->ops.connect_external_te = dpu_hw_pp_connect_external_te;
+		c->ops.get_line_count = dpu_hw_pp_get_line_count;
+		c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh;
+	}
 	c->ops.setup_dsc = dpu_hw_pp_setup_dsc;
 	c->ops.enable_dsc = dpu_hw_pp_dsc_enable;
 	c->ops.disable_dsc = dpu_hw_pp_dsc_disable;
-- 
GitLab


From a2ff096803b30de95c6428ade638726ee3897b8e Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:29 +0200
Subject: [PATCH 0593/1791] drm/msm/dpu: Disable MDP vsync source selection on
 DPU 5.0.0 and above

Since hardware revision 5.0.0 the TE configuration moved out of the
PINGPONG block into the INTF block, including vsync source selection
that was previously part of MDP top.  Writing to the MDP_VSYNC_SEL
register has no effect anymore and is omitted downstream via the
DPU/SDE_MDP_VSYNC_SEL feature flag.  This flag is only added to INTF
blocks used by hardware prior to 5.0.0.

The code that writes to these registers in the INTF block will follow in
subsequent patches.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534220/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-15-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_3_0_msm8998.h   |  2 +-
 .../msm/disp/dpu1/catalog/dpu_4_0_sdm845.h    |  2 +-
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |  3 ++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c    | 50 +++++++++++++------
 4 files changed, 40 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 3d01840ef64b5..dc2c8972afd74 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -30,7 +30,7 @@ static const struct dpu_mdp_cfg msm8998_mdp[] = {
 	{
 	.name = "top_0", .id = MDP_TOP,
 	.base = 0x0, .len = 0x458,
-	.features = 0,
+	.features = BIT(DPU_MDP_VSYNC_SEL),
 	.clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 },
 	.clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 },
 	.clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 },
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index 84159f8cbdaeb..41e4c1e362d33 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -30,7 +30,7 @@ static const struct dpu_mdp_cfg sdm845_mdp[] = {
 	{
 	.name = "top_0", .id = MDP_TOP,
 	.base = 0x0, .len = 0x45c,
-	.features = BIT(DPU_MDP_AUDIO_SELECT),
+	.features = BIT(DPU_MDP_AUDIO_SELECT) | BIT(DPU_MDP_VSYNC_SEL),
 	.clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 },
 	.clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 },
 	.clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 },
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index c6eb5a8a66da9..917272fa7f80f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -48,6 +48,8 @@ enum {
  * @DPU_MDP_UBWC_1_5,      Universal Bandwidth compression version 1.5
  * @DPU_MDP_PERIPH_0_REMOVED Indicates that access to periph top0 block results
  *			   in a failure
+ * @DPU_MDP_VSYNC_SEL      Enables vsync source selection via MDP_VSYNC_SEL register
+ *                         (moved into INTF block since DPU 5.0.0)
  * @DPU_MDP_MAX            Maximum value
 
  */
@@ -59,6 +61,7 @@ enum {
 	DPU_MDP_UBWC_1_5,
 	DPU_MDP_AUDIO_SELECT,
 	DPU_MDP_PERIPH_0_REMOVED,
+	DPU_MDP_VSYNC_SEL,
 	DPU_MDP_MAX
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c
index 2bb02e17ee52c..963bdb5e02521 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c
@@ -130,24 +130,12 @@ static void dpu_hw_setup_vsync_source(struct dpu_hw_mdp *mdp,
 		struct dpu_vsync_source_cfg *cfg)
 {
 	struct dpu_hw_blk_reg_map *c;
-	u32 reg, wd_load_value, wd_ctl, wd_ctl2, i;
-	static const u32 pp_offset[PINGPONG_MAX] = {0xC, 0x8, 0x4, 0x13, 0x18};
+	u32 reg, wd_load_value, wd_ctl, wd_ctl2;
 
-	if (!mdp || !cfg || (cfg->pp_count > ARRAY_SIZE(cfg->ppnumber)))
+	if (!mdp || !cfg)
 		return;
 
 	c = &mdp->hw;
-	reg = DPU_REG_READ(c, MDP_VSYNC_SEL);
-	for (i = 0; i < cfg->pp_count; i++) {
-		int pp_idx = cfg->ppnumber[i] - PINGPONG_0;
-
-		if (pp_idx >= ARRAY_SIZE(pp_offset))
-			continue;
-
-		reg &= ~(0xf << pp_offset[pp_idx]);
-		reg |= (cfg->vsync_source & 0xf) << pp_offset[pp_idx];
-	}
-	DPU_REG_WRITE(c, MDP_VSYNC_SEL, reg);
 
 	if (cfg->vsync_source >= DPU_VSYNC_SOURCE_WD_TIMER_4 &&
 			cfg->vsync_source <= DPU_VSYNC_SOURCE_WD_TIMER_0) {
@@ -194,6 +182,33 @@ static void dpu_hw_setup_vsync_source(struct dpu_hw_mdp *mdp,
 	}
 }
 
+static void dpu_hw_setup_vsync_source_and_vsync_sel(struct dpu_hw_mdp *mdp,
+		struct dpu_vsync_source_cfg *cfg)
+{
+	struct dpu_hw_blk_reg_map *c;
+	u32 reg, i;
+	static const u32 pp_offset[PINGPONG_MAX] = {0xC, 0x8, 0x4, 0x13, 0x18};
+
+	if (!mdp || !cfg || (cfg->pp_count > ARRAY_SIZE(cfg->ppnumber)))
+		return;
+
+	c = &mdp->hw;
+
+	reg = DPU_REG_READ(c, MDP_VSYNC_SEL);
+	for (i = 0; i < cfg->pp_count; i++) {
+		int pp_idx = cfg->ppnumber[i] - PINGPONG_0;
+
+		if (pp_idx >= ARRAY_SIZE(pp_offset))
+			continue;
+
+		reg &= ~(0xf << pp_offset[pp_idx]);
+		reg |= (cfg->vsync_source & 0xf) << pp_offset[pp_idx];
+	}
+	DPU_REG_WRITE(c, MDP_VSYNC_SEL, reg);
+
+	dpu_hw_setup_vsync_source(mdp, cfg);
+}
+
 static void dpu_hw_get_safe_status(struct dpu_hw_mdp *mdp,
 		struct dpu_danger_safe_status *status)
 {
@@ -241,7 +256,12 @@ static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops,
 	ops->setup_split_pipe = dpu_hw_setup_split_pipe;
 	ops->setup_clk_force_ctrl = dpu_hw_setup_clk_force_ctrl;
 	ops->get_danger_status = dpu_hw_get_danger_status;
-	ops->setup_vsync_source = dpu_hw_setup_vsync_source;
+
+	if (cap & BIT(DPU_MDP_VSYNC_SEL))
+		ops->setup_vsync_source = dpu_hw_setup_vsync_source_and_vsync_sel;
+	else
+		ops->setup_vsync_source = dpu_hw_setup_vsync_source;
+
 	ops->get_safe_status = dpu_hw_get_safe_status;
 
 	if (cap & BIT(DPU_MDP_AUDIO_SELECT))
-- 
GitLab


From c31ec42ebb45aaa4f668037525b8df140563ca6a Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@somainline.org>
Date: Thu, 27 Apr 2023 00:37:30 +0200
Subject: [PATCH 0594/1791] drm/msm/dpu: Move dpu_hw_{tear_check,
 pp_vsync_info} to dpu_hw_mdss.h

Now that newer SoCs since DPU 5.0.0 manage tearcheck in the INTF instead
of PINGPONG block, move the struct definition to a common file. Also,
bring in documentation from msm-4.19 techpack while at it.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
[Marijn: Also move dpu_hw_pp_vsync_info]
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534232/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-16-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h   | 46 +++++++++++++++++++
 .../gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h   | 22 ---------
 2 files changed, 46 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index 2d9192a6ce006..6ed12fd0505b3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -463,4 +463,50 @@ struct dpu_mdss_color {
 #define DPU_DBG_MASK_DSPP     (1 << 10)
 #define DPU_DBG_MASK_DSC      (1 << 11)
 
+/**
+ * struct dpu_hw_tear_check - Struct contains parameters to configure
+ * tear-effect module. This structure is used to configure tear-check
+ * logic present either in ping-pong or in interface module.
+ * @vsync_count:        Ratio of MDP VSYNC clk freq(Hz) to refresh rate divided
+ *                      by no of lines
+ * @sync_cfg_height:    Total vertical lines (display height - 1)
+ * @vsync_init_val:     Init value to which the read pointer gets loaded at
+ *                      vsync edge
+ * @sync_threshold_start:    Read pointer threshold start ROI for write operation
+ * @sync_threshold_continue: The minimum number of lines the write pointer
+ *                           needs to be above the read pointer
+ * @start_pos:          The position from which the start_threshold value is added
+ * @rd_ptr_irq:         The read pointer line at which interrupt has to be generated
+ * @hw_vsync_mode:      Sync with external frame sync input
+ */
+struct dpu_hw_tear_check {
+	/*
+	 * This is ratio of MDP VSYNC clk freq(Hz) to
+	 * refresh rate divided by no of lines
+	 */
+	u32 vsync_count;
+	u32 sync_cfg_height;
+	u32 vsync_init_val;
+	u32 sync_threshold_start;
+	u32 sync_threshold_continue;
+	u32 start_pos;
+	u32 rd_ptr_irq;
+	u8 hw_vsync_mode;
+};
+
+/**
+ * struct dpu_hw_pp_vsync_info - Struct contains parameters to configure
+ * read and write pointers for command mode panels
+ * @rd_ptr_init_val:    Value of rd pointer at vsync edge
+ * @rd_ptr_frame_count: Num frames sent since enabling interface
+ * @rd_ptr_line_count:  Current line on panel (rd ptr)
+ * @wr_ptr_line_count:  Current line within pp fifo (wr ptr)
+ */
+struct dpu_hw_pp_vsync_info {
+	u32 rd_ptr_init_val;
+	u32 rd_ptr_frame_count;
+	u32 rd_ptr_line_count;
+	u32 wr_ptr_line_count;
+};
+
 #endif  /* _DPU_HW_MDSS_H */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
index 336d383f8b647..08ba9d76f061e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
@@ -13,28 +13,6 @@
 
 struct dpu_hw_pingpong;
 
-struct dpu_hw_tear_check {
-	/*
-	 * This is ratio of MDP VSYNC clk freq(Hz) to
-	 * refresh rate divided by no of lines
-	 */
-	u32 vsync_count;
-	u32 sync_cfg_height;
-	u32 vsync_init_val;
-	u32 sync_threshold_start;
-	u32 sync_threshold_continue;
-	u32 start_pos;
-	u32 rd_ptr_irq;
-	u8 hw_vsync_mode;
-};
-
-struct dpu_hw_pp_vsync_info {
-	u32 rd_ptr_init_val;	/* value of rd pointer at vsync edge */
-	u32 rd_ptr_frame_count;	/* num frames sent since enabling interface */
-	u32 rd_ptr_line_count;	/* current line on panel (rd ptr) */
-	u32 wr_ptr_line_count;	/* current line within pp fifo (wr ptr) */
-};
-
 /**
  * struct dpu_hw_dither_cfg - dither feature structure
  * @flags: for customizing operations
-- 
GitLab


From a38a9949a96399ecbe213837b2341babe3262423 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:31 +0200
Subject: [PATCH 0595/1791] drm/msm/dpu: Factor out shared interrupt register
 in INTF_BLK macro

As the INTF block is going to attain more interrupts that don't share
the same MDP_SSPP_TOP0_INTR register, factor out the _reg argument for
the caller to construct the right interrupt index (register and bit
index) to not make the interrupt bit arguments depend on one of multiple
interrupt register indices.  This brings us more in line with how PP_BLK
specifies its interrupts and allows for better wrapping in the arrays.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534222/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-17-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_3_0_msm8998.h   | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_4_0_sdm845.h    | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_5_0_sm8150.h    | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h   | 24 +++++++++----
 .../msm/disp/dpu1/catalog/dpu_6_0_sm8250.h    | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_6_2_sc7180.h    |  8 +++--
 .../msm/disp/dpu1/catalog/dpu_6_3_sm6115.h    |  6 ++--
 .../msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h   |  6 ++--
 .../msm/disp/dpu1/catalog/dpu_7_0_sm8350.h    | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_7_2_sc7280.h    | 12 +++++--
 .../msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h  | 36 ++++++++++++++-----
 .../msm/disp/dpu1/catalog/dpu_8_1_sm8450.h    | 16 ++++++---
 .../msm/disp/dpu1/catalog/dpu_9_0_sm8550.h    | 16 ++++++---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    |  6 ++--
 14 files changed, 155 insertions(+), 55 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index dc2c8972afd74..37a52b4ef8d39 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -134,10 +134,18 @@ static const struct dpu_dspp_cfg msm8998_dspp[] = {
 };
 
 static const struct dpu_intf_cfg msm8998_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 21, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 21, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 21, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_HDMI, 0, 21, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 21, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 21, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 21, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_HDMI, 0, 21, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg msm8998_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index 41e4c1e362d33..0b727252671c0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -132,10 +132,18 @@ static const struct dpu_dsc_cfg sdm845_dsc[] = {
 };
 
 static const struct dpu_intf_cfg sdm845_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 24, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 24, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SDM845_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg sdm845_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index 266c525f8daaf..9089fededd89f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -162,10 +162,18 @@ static const struct dpu_dsc_cfg sm8150_dsc[] = {
 };
 
 static const struct dpu_intf_cfg sm8150_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg sm8150_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index b05bd561e1403..aab355c78970f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -163,13 +163,25 @@ static const struct dpu_dsc_cfg sc8180x_dsc[] = {
 };
 
 static const struct dpu_intf_cfg sc8180x_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
 	/* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */
-	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
-	INTF_BLK("intf_4", INTF_4, 0x6c000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 20, 21),
-	INTF_BLK("intf_5", INTF_5, 0x6c800, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 22, 23),
+	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
+	INTF_BLK("intf_4", INTF_4, 0x6c000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)),
+	INTF_BLK("intf_5", INTF_5, 0x6c800, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)),
 };
 
 static const struct dpu_perf_cfg sc8180x_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index 8660d04d0f589..63fd46f987278 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -163,10 +163,18 @@ static const struct dpu_dsc_cfg sm8250_dsc[] = {
 };
 
 static const struct dpu_intf_cfg sm8250_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_wb_cfg sm8250_wb[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 63056439d4127..301287e05c828 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -85,8 +85,12 @@ static const struct dpu_pingpong_cfg sc7180_pp[] = {
 };
 
 static const struct dpu_intf_cfg sc7180_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
 };
 
 static const struct dpu_wb_cfg sc7180_wb[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index e15dc96f1286a..88a0d6119ad50 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -66,8 +66,10 @@ static const struct dpu_pingpong_cfg sm6115_pp[] = {
 };
 
 static const struct dpu_intf_cfg sm6115_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0, 0),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
+	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
 };
 
 static const struct dpu_perf_cfg sm6115_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index 2ff98ef6999fe..cd6f4048f286b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -63,8 +63,10 @@ static const struct dpu_pingpong_cfg qcm2290_pp[] = {
 };
 
 static const struct dpu_intf_cfg qcm2290_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0, 0),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
+	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
+	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
 };
 
 static const struct dpu_perf_cfg qcm2290_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 9e8d6632a1927..6117c9fbc73f8 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -154,10 +154,18 @@ static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = {
 };
 
 static const struct dpu_intf_cfg sm8350_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg sm8350_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index ec9047cd60fd5..208c97e8d7e19 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -100,9 +100,15 @@ static const struct dpu_wb_cfg sc7280_wb[] = {
 };
 
 static const struct dpu_intf_cfg sc7280_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 22, 23),
+	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)),
 };
 
 static const struct dpu_perf_cfg sc7280_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index 8e4d0bc4aa703..cf81133540dd9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -144,15 +144,33 @@ static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = {
 
 /* TODO: INTF 3, 8 and 7 are used for MST, marked as INTF_NONE for now */
 static const struct dpu_intf_cfg sc8280xp_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
-	INTF_BLK("intf_4", INTF_4, 0x38000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 20, 21),
-	INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_3, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 22, 23),
-	INTF_BLK("intf_6", INTF_6, 0x3a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 16, 17),
-	INTF_BLK("intf_7", INTF_7, 0x3b000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 18, 19),
-	INTF_BLK("intf_8", INTF_8, 0x3c000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 12, 13),
+	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
+	INTF_BLK("intf_4", INTF_4, 0x38000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)),
+	INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_3, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)),
+	INTF_BLK("intf_6", INTF_6, 0x3a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17)),
+	INTF_BLK("intf_7", INTF_7, 0x3b000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19)),
+	INTF_BLK("intf_8", INTF_8, 0x3c000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
 };
 
 static const struct dpu_perf_cfg sc8280xp_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 014922ac03dbc..9a95386b7a946 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -162,10 +162,18 @@ static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = {
 };
 
 static const struct dpu_intf_cfg sm8450_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg sm8450_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
index be2f37728aa0c..cf366be79b1c2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
@@ -166,11 +166,19 @@ static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = {
 };
 
 static const struct dpu_intf_cfg sm8550_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25),
+	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
 	/* TODO TE sub-blocks for intf1 & intf2 */
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29),
-	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31),
+	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
 };
 
 static const struct dpu_perf_cfg sm8550_perf_data = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index fad12ba0eab6a..f2482e500a1b0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -531,7 +531,7 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 /*************************************************************
  * INTF sub blocks config
  *************************************************************/
-#define INTF_BLK(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _reg, _underrun_bit, _vsync_bit) \
+#define INTF_BLK(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync) \
 	{\
 	.name = _name, .id = _id, \
 	.base = _base, .len = _len, \
@@ -539,8 +539,8 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 	.type = _type, \
 	.controller_id = _ctrl_id, \
 	.prog_fetch_lines_worst_case = _progfetch, \
-	.intr_underrun = DPU_IRQ_IDX(_reg, _underrun_bit), \
-	.intr_vsync = DPU_IRQ_IDX(_reg, _vsync_bit), \
+	.intr_underrun = _underrun, \
+	.intr_vsync = _vsync, \
 	}
 
 /*************************************************************
-- 
GitLab


From ec6e9b673ad56d019716298c15901d3a46e65c77 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:32 +0200
Subject: [PATCH 0596/1791] drm/msm/dpu: Describe TEAR interrupt registers for
 DSI interfaces

All SoCs since DPU 5.0.0 have the tear interrupt registers moved out of
the PINGPONG block and into the INTF block.  Wire up the IRQ register
masks in the interrupt table for enabling, reading and clearing them.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534244/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-18-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 28 +++++++++++++++++++
 .../gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h |  4 +++
 2 files changed, 32 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
index 152d4272a087a..5e2d68ebb113e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c
@@ -21,6 +21,10 @@
 #define MDP_INTF_INTR_EN(intf)				(MDP_INTF_OFF(intf) + 0x1c0)
 #define MDP_INTF_INTR_STATUS(intf)			(MDP_INTF_OFF(intf) + 0x1c4)
 #define MDP_INTF_INTR_CLEAR(intf)			(MDP_INTF_OFF(intf) + 0x1c8)
+#define MDP_INTF_TEAR_OFF(intf)				(0x6D700 + 0x100 * (intf))
+#define MDP_INTF_INTR_TEAR_EN(intf)			(MDP_INTF_TEAR_OFF(intf) + 0x000)
+#define MDP_INTF_INTR_TEAR_STATUS(intf)			(MDP_INTF_TEAR_OFF(intf) + 0x004)
+#define MDP_INTF_INTR_TEAR_CLEAR(intf)			(MDP_INTF_TEAR_OFF(intf) + 0x008)
 #define MDP_AD4_OFF(ad4)				(0x7C000 + 0x1000 * (ad4))
 #define MDP_AD4_INTR_EN_OFF(ad4)			(MDP_AD4_OFF(ad4) + 0x41c)
 #define MDP_AD4_INTR_CLEAR_OFF(ad4)			(MDP_AD4_OFF(ad4) + 0x424)
@@ -29,6 +33,10 @@
 #define MDP_INTF_REV_7xxx_INTR_EN(intf)			(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c0)
 #define MDP_INTF_REV_7xxx_INTR_STATUS(intf)		(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c4)
 #define MDP_INTF_REV_7xxx_INTR_CLEAR(intf)		(MDP_INTF_REV_7xxx_OFF(intf) + 0x1c8)
+#define MDP_INTF_REV_7xxx_TEAR_OFF(intf)		(0x34800 + 0x1000 * (intf))
+#define MDP_INTF_REV_7xxx_INTR_TEAR_EN(intf)		(MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x000)
+#define MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(intf)	(MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x004)
+#define MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(intf)		(MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x008)
 
 /**
  * struct dpu_intr_reg - array of DPU register sets
@@ -93,6 +101,16 @@ static const struct dpu_intr_reg dpu_intr_set[] = {
 		MDP_INTF_INTR_EN(5),
 		MDP_INTF_INTR_STATUS(5)
 	},
+	[MDP_INTF1_TEAR_INTR] = {
+		MDP_INTF_INTR_TEAR_CLEAR(1),
+		MDP_INTF_INTR_TEAR_EN(1),
+		MDP_INTF_INTR_TEAR_STATUS(1)
+	},
+	[MDP_INTF2_TEAR_INTR] = {
+		MDP_INTF_INTR_TEAR_CLEAR(2),
+		MDP_INTF_INTR_TEAR_EN(2),
+		MDP_INTF_INTR_TEAR_STATUS(2)
+	},
 	[MDP_AD4_0_INTR] = {
 		MDP_AD4_INTR_CLEAR_OFF(0),
 		MDP_AD4_INTR_EN_OFF(0),
@@ -113,11 +131,21 @@ static const struct dpu_intr_reg dpu_intr_set[] = {
 		MDP_INTF_REV_7xxx_INTR_EN(1),
 		MDP_INTF_REV_7xxx_INTR_STATUS(1)
 	},
+	[MDP_INTF1_7xxx_TEAR_INTR] = {
+		MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(1),
+		MDP_INTF_REV_7xxx_INTR_TEAR_EN(1),
+		MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(1)
+	},
 	[MDP_INTF2_7xxx_INTR] = {
 		MDP_INTF_REV_7xxx_INTR_CLEAR(2),
 		MDP_INTF_REV_7xxx_INTR_EN(2),
 		MDP_INTF_REV_7xxx_INTR_STATUS(2)
 	},
+	[MDP_INTF2_7xxx_TEAR_INTR] = {
+		MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(2),
+		MDP_INTF_REV_7xxx_INTR_TEAR_EN(2),
+		MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(2)
+	},
 	[MDP_INTF3_7xxx_INTR] = {
 		MDP_INTF_REV_7xxx_INTR_CLEAR(3),
 		MDP_INTF_REV_7xxx_INTR_EN(3),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
index bbf475a1cb455..1f2dabc54c229 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h
@@ -23,11 +23,15 @@ enum dpu_hw_intr_reg {
 	MDP_INTF3_INTR,
 	MDP_INTF4_INTR,
 	MDP_INTF5_INTR,
+	MDP_INTF1_TEAR_INTR,
+	MDP_INTF2_TEAR_INTR,
 	MDP_AD4_0_INTR,
 	MDP_AD4_1_INTR,
 	MDP_INTF0_7xxx_INTR,
 	MDP_INTF1_7xxx_INTR,
+	MDP_INTF1_7xxx_TEAR_INTR,
 	MDP_INTF2_7xxx_INTR,
+	MDP_INTF2_7xxx_TEAR_INTR,
 	MDP_INTF3_7xxx_INTR,
 	MDP_INTF4_7xxx_INTR,
 	MDP_INTF5_7xxx_INTR,
-- 
GitLab


From 0272b9c3b5729ba4170b15953747259dbcde5c88 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:33 +0200
Subject: [PATCH 0597/1791] drm/msm/dpu: Add TEAR-READ-pointer interrupt to
 INTF block

All SoCs since DPU 5.0.0 have the tear interrupt registers moved out of
the PINGPONG block and into the INTF block.  The new interrupts are
described in dpu_hw_interrupts.c, now wire them up in individual SoC
catalog files by setting the intr_tear_rd_ptr to the IRQ index spcified
in the offset table and enabling this set of DPU interrupts via the
mdss_irqs bitmask.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534236/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-19-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h    | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h   | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h    | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h    |  8 +++++---
 .../drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h    |  8 +++++---
 .../drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h   |  8 +++++---
 .../drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h    | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h    |  6 ++++--
 .../drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h  | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h    | 12 ++++++++----
 .../drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h    | 13 ++++++++-----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    | 15 +++++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |  2 ++
 13 files changed, 92 insertions(+), 40 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index 9089fededd89f..561022b8dfd6a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -165,12 +165,14 @@ static const struct dpu_intf_cfg sm8150_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -234,7 +236,9 @@ const struct dpu_mdss_cfg dpu_sm8150_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_INTR) | \
 		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR) | \
 		     BIT(MDP_INTF2_INTR) | \
+		     BIT(MDP_INTF2_TEAR_INTR) | \
 		     BIT(MDP_INTF3_INTR) | \
 		     BIT(MDP_AD4_0_INTR) | \
 		     BIT(MDP_AD4_1_INTR),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index aab355c78970f..d1f306c895671 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -166,12 +166,14 @@ static const struct dpu_intf_cfg sc8180x_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)),
 	/* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */
 	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
@@ -240,7 +242,9 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_INTR) | \
 		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR) | \
 		     BIT(MDP_INTF2_INTR) | \
+		     BIT(MDP_INTF2_TEAR_INTR) | \
 		     BIT(MDP_INTF3_INTR) | \
 		     BIT(MDP_INTF4_INTR) | \
 		     BIT(MDP_INTF5_INTR) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index 63fd46f987278..10d67db8423fd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -166,12 +166,14 @@ static const struct dpu_intf_cfg sm8250_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -242,7 +244,9 @@ const struct dpu_mdss_cfg dpu_sm8250_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_INTR) | \
 		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR) | \
 		     BIT(MDP_INTF2_INTR) | \
+		     BIT(MDP_INTF2_TEAR_INTR) | \
 		     BIT(MDP_INTF3_INTR) | \
 		     BIT(MDP_INTF4_INTR),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 301287e05c828..0b05da2592c08 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -88,9 +88,10 @@ static const struct dpu_intf_cfg sc7180_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
 };
 
 static const struct dpu_wb_cfg sc7180_wb[] = {
@@ -152,7 +153,8 @@ const struct dpu_mdss_cfg dpu_sc7180_cfg = {
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_INTR) | \
-		     BIT(MDP_INTF1_INTR),
+		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR),
 };
 
 #endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index 88a0d6119ad50..5509ceb5d55bd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -67,9 +67,10 @@ static const struct dpu_pingpong_cfg sm6115_pp[] = {
 
 static const struct dpu_intf_cfg sm6115_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
 };
 
 static const struct dpu_perf_cfg sm6115_perf_data = {
@@ -124,7 +125,8 @@ const struct dpu_mdss_cfg dpu_sm6115_cfg = {
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
-		     BIT(MDP_INTF1_INTR),
+		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR),
 };
 
 #endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index cd6f4048f286b..22b8a173d2142 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -64,9 +64,10 @@ static const struct dpu_pingpong_cfg qcm2290_pp[] = {
 
 static const struct dpu_intf_cfg qcm2290_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
-	INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
 };
 
 static const struct dpu_perf_cfg qcm2290_perf_data = {
@@ -114,7 +115,8 @@ const struct dpu_mdss_cfg dpu_qcm2290_cfg = {
 	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
 		     BIT(MDP_SSPP_TOP0_INTR2) | \
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
-		     BIT(MDP_INTF1_INTR),
+		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR),
 };
 
 #endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 6117c9fbc73f8..2dcfe18bc8c97 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -157,12 +157,14 @@ static const struct dpu_intf_cfg sm8350_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -225,7 +227,9 @@ const struct dpu_mdss_cfg dpu_sm8350_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_7xxx_INTR) | \
 		     BIT(MDP_INTF1_7xxx_INTR) | \
+		     BIT(MDP_INTF1_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF2_7xxx_INTR) | \
+		     BIT(MDP_INTF2_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF3_7xxx_INTR),
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index 208c97e8d7e19..5d894cbb0a62f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -103,9 +103,10 @@ static const struct dpu_intf_cfg sc7280_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)),
 	INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)),
@@ -166,6 +167,7 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_7xxx_INTR) | \
 		     BIT(MDP_INTF1_7xxx_INTR) | \
+		     BIT(MDP_INTF1_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF5_7xxx_INTR),
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index cf81133540dd9..c3f1ae000a212 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -147,12 +147,14 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -227,7 +229,9 @@ const struct dpu_mdss_cfg dpu_sc8280xp_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_7xxx_INTR) | \
 		     BIT(MDP_INTF1_7xxx_INTR) | \
+		     BIT(MDP_INTF1_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF2_7xxx_INTR) | \
+		     BIT(MDP_INTF2_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF3_7xxx_INTR) | \
 		     BIT(MDP_INTF4_7xxx_INTR) | \
 		     BIT(MDP_INTF5_7xxx_INTR) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 9a95386b7a946..279fe8f4e1e3f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -165,12 +165,14 @@ static const struct dpu_intf_cfg sm8450_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -233,7 +235,9 @@ const struct dpu_mdss_cfg dpu_sm8450_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_7xxx_INTR) | \
 		     BIT(MDP_INTF1_7xxx_INTR) | \
+		     BIT(MDP_INTF1_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF2_7xxx_INTR) | \
+		     BIT(MDP_INTF2_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF3_7xxx_INTR),
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
index cf366be79b1c2..85dc34458b88a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
@@ -169,13 +169,14 @@ static const struct dpu_intf_cfg sm8550_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
-	/* TODO TE sub-blocks for intf1 & intf2 */
-	INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)),
-	INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+			DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)),
+	INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)),
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29),
+			DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)),
 	INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)),
@@ -238,7 +239,9 @@ const struct dpu_mdss_cfg dpu_sm8550_cfg = {
 		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
 		     BIT(MDP_INTF0_7xxx_INTR) | \
 		     BIT(MDP_INTF1_7xxx_INTR) | \
+		     BIT(MDP_INTF1_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF2_7xxx_INTR) | \
+		     BIT(MDP_INTF2_7xxx_TEAR_INTR) | \
 		     BIT(MDP_INTF3_7xxx_INTR),
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index f2482e500a1b0..81b44a394f03b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -541,6 +541,21 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 	.prog_fetch_lines_worst_case = _progfetch, \
 	.intr_underrun = _underrun, \
 	.intr_vsync = _vsync, \
+	.intr_tear_rd_ptr = -1, \
+	}
+
+/* DSI Interface sub-block with TEAR registers (since DPU 5.0.0) */
+#define INTF_BLK_DSI_TE(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync, _tear_rd_ptr) \
+	{\
+	.name = _name, .id = _id, \
+	.base = _base, .len = _len, \
+	.features = _features, \
+	.type = _type, \
+	.controller_id = _ctrl_id, \
+	.prog_fetch_lines_worst_case = _progfetch, \
+	.intr_underrun = _underrun, \
+	.intr_vsync = _vsync, \
+	.intr_tear_rd_ptr = _tear_rd_ptr, \
 	}
 
 /*************************************************************
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 917272fa7f80f..a2ef61d4f7fa2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -615,6 +615,7 @@ struct dpu_dsc_cfg {
  * @prog_fetch_lines_worst_case	Worst case latency num lines needed to prefetch
  * @intr_underrun:	index for INTF underrun interrupt
  * @intr_vsync:	        index for INTF VSYNC interrupt
+ * @intr_tear_rd_ptr:  Index for INTF TEAR_RD_PTR interrupt
  */
 struct dpu_intf_cfg  {
 	DPU_HW_BLK_INFO;
@@ -623,6 +624,7 @@ struct dpu_intf_cfg  {
 	u32 prog_fetch_lines_worst_case;
 	s32 intr_underrun;
 	s32 intr_vsync;
+	s32 intr_tear_rd_ptr;
 };
 
 /**
-- 
GitLab


From bb9f1880ca902114569e17887626ee80c759a47c Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:34 +0200
Subject: [PATCH 0598/1791] drm/msm/dpu: Merge setup_- and enable_tearcheck
 pingpong callbacks

These functions are always called consecutively and are best bundled
together for simplicity, especially when the same structure of callbacks
will be replicated later on the interface block for INTF TE support.
The enable_tearcheck(false) case is now replaced with a more obvious
disable_tearcheck(), encapsulating the original register write with 0.

Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534217/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-20-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c | 10 ++++------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c      | 10 ++++++----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h      | 11 +++++------
 3 files changed, 15 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index a60fb8d3736b5..1df3745224f5e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -327,8 +327,7 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 
 	DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0);
 
-	if (!phys_enc->hw_pp->ops.setup_tearcheck ||
-		!phys_enc->hw_pp->ops.enable_tearcheck) {
+	if (!phys_enc->hw_pp->ops.enable_tearcheck) {
 		DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n");
 		return;
 	}
@@ -383,8 +382,7 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 		phys_enc->hw_pp->idx - PINGPONG_0, tc_cfg.sync_cfg_height,
 		tc_cfg.sync_threshold_start, tc_cfg.sync_threshold_continue);
 
-	phys_enc->hw_pp->ops.setup_tearcheck(phys_enc->hw_pp, &tc_cfg);
-	phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, tc_enable);
+	phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, &tc_cfg);
 }
 
 static void _dpu_encoder_phys_cmd_pingpong_config(
@@ -511,8 +509,8 @@ static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc)
 		return;
 	}
 
-	if (phys_enc->hw_pp->ops.enable_tearcheck)
-		phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, false);
+	if (phys_enc->hw_pp->ops.disable_tearcheck)
+		phys_enc->hw_pp->ops.disable_tearcheck(phys_enc->hw_pp);
 
 	if (phys_enc->hw_intf->ops.bind_pingpong_blk) {
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index c70f6ff018f23..79e45768eae5a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -73,7 +73,7 @@ static void dpu_hw_pp_setup_dither(struct dpu_hw_pingpong *pp,
 	DPU_REG_WRITE(c, base + PP_DITHER_EN, 1);
 }
 
-static int dpu_hw_pp_setup_te_config(struct dpu_hw_pingpong *pp,
+static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp,
 		struct dpu_hw_tear_check *te)
 {
 	struct dpu_hw_blk_reg_map *c;
@@ -100,6 +100,8 @@ static int dpu_hw_pp_setup_te_config(struct dpu_hw_pingpong *pp,
 	DPU_REG_WRITE(c, PP_SYNC_WRCOUNT,
 			(te->start_pos + te->sync_threshold_start + 1));
 
+	DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 1);
+
 	return 0;
 }
 
@@ -126,7 +128,7 @@ static bool dpu_hw_pp_get_autorefresh_config(struct dpu_hw_pingpong *pp,
 	return !!((val & BIT(31)) >> 31);
 }
 
-static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, bool enable)
+static int dpu_hw_pp_disable_te(struct dpu_hw_pingpong *pp)
 {
 	struct dpu_hw_blk_reg_map *c;
 
@@ -134,7 +136,7 @@ static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, bool enable)
 		return -EINVAL;
 	c = &pp->hw;
 
-	DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, enable);
+	DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 0);
 	return 0;
 }
 
@@ -283,8 +285,8 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 				unsigned long features)
 {
 	if (test_bit(DPU_PINGPONG_TE, &features)) {
-		c->ops.setup_tearcheck = dpu_hw_pp_setup_te_config;
 		c->ops.enable_tearcheck = dpu_hw_pp_enable_te;
+		c->ops.disable_tearcheck = dpu_hw_pp_disable_te;
 		c->ops.connect_external_te = dpu_hw_pp_connect_external_te;
 		c->ops.get_line_count = dpu_hw_pp_get_line_count;
 		c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
index 08ba9d76f061e..d3246a9a5808f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h
@@ -37,8 +37,8 @@ struct dpu_hw_dither_cfg {
  *
  * struct dpu_hw_pingpong_ops : Interface to the pingpong Hw driver functions
  *  Assumption is these functions will be called after clocks are enabled
- *  @setup_tearcheck : program tear check values
- *  @enable_tearcheck : enables tear check
+ *  @enable_tearcheck: program and enable tear check block
+ *  @disable_tearcheck: disable able tear check block
  *  @setup_dither : function to program the dither hw block
  *  @get_line_count: obtain current vertical line counter
  */
@@ -47,14 +47,13 @@ struct dpu_hw_pingpong_ops {
 	 * enables vysnc generation and sets up init value of
 	 * read pointer and programs the tear check cofiguration
 	 */
-	int (*setup_tearcheck)(struct dpu_hw_pingpong *pp,
+	int (*enable_tearcheck)(struct dpu_hw_pingpong *pp,
 			struct dpu_hw_tear_check *cfg);
 
 	/**
-	 * enables tear check block
+	 * disables tear check block
 	 */
-	int (*enable_tearcheck)(struct dpu_hw_pingpong *pp,
-			bool enable);
+	int (*disable_tearcheck)(struct dpu_hw_pingpong *pp);
 
 	/**
 	 * read, modify, write to either set or clear listening to external TE
-- 
GitLab


From e955a3f0d86e4a789364af62495ac37546eb71c8 Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:35 +0200
Subject: [PATCH 0599/1791] drm/msm/dpu: Implement tearcheck support on INTF
 block

Since DPU 5.0.0 the TEARCHECK registers and interrupts moved out of the
PINGPONG block and into the INTF.  Implement the necessary callbacks in
the INTF block, and use these callbacks together with the INTF_TEAR
interrupts.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534234/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-21-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c   |  11 +
 .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h  |  10 +-
 .../drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c  | 160 ++++++++-----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c   | 214 ++++++++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h   |  25 ++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h   |   2 +
 drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h     |  14 ++
 7 files changed, 378 insertions(+), 58 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index b34416cbd0f5f..d7cd4734dc7df 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -666,6 +666,7 @@ static void _dpu_encoder_update_vsync_source(struct dpu_encoder_virt *dpu_enc,
 	struct dpu_kms *dpu_kms;
 	struct dpu_hw_mdp *hw_mdptop;
 	struct drm_encoder *drm_enc;
+	struct dpu_encoder_phys *phys_enc;
 	int i;
 
 	if (!dpu_enc || !disp_info) {
@@ -696,12 +697,22 @@ static void _dpu_encoder_update_vsync_source(struct dpu_encoder_virt *dpu_enc,
 			vsync_cfg.ppnumber[i] = dpu_enc->hw_pp[i]->idx;
 
 		vsync_cfg.pp_count = dpu_enc->num_phys_encs;
+		vsync_cfg.frame_rate = drm_mode_vrefresh(&dpu_enc->base.crtc->state->adjusted_mode);
+
 		if (disp_info->is_te_using_watchdog_timer)
 			vsync_cfg.vsync_source = DPU_VSYNC_SOURCE_WD_TIMER_0;
 		else
 			vsync_cfg.vsync_source = DPU_VSYNC0_SOURCE_GPIO;
 
 		hw_mdptop->ops.setup_vsync_source(hw_mdptop, &vsync_cfg);
+
+		for (i = 0; i < dpu_enc->num_phys_encs; i++) {
+			phys_enc = dpu_enc->phys_encs[i];
+
+			if (phys_enc->has_intf_te && phys_enc->hw_intf->ops.vsync_sel)
+				phys_enc->hw_intf->ops.vsync_sel(phys_enc->hw_intf,
+						vsync_cfg.vsync_source);
+		}
 	}
 }
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index 1d434b22180d9..90f177e43262a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -129,10 +129,10 @@ struct dpu_encoder_phys_ops {
 /**
  * enum dpu_intr_idx - dpu encoder interrupt index
  * @INTR_IDX_VSYNC:    Vsync interrupt for video mode panel
- * @INTR_IDX_PINGPONG: Pingpong done unterrupt for cmd mode panel
- * @INTR_IDX_UNDERRUN: Underrun unterrupt for video and cmd mode panel
- * @INTR_IDX_RDPTR:    Readpointer done unterrupt for cmd mode panel
- * @INTR_IDX_WB_DONE:  Writeback fone interrupt for virtual connector
+ * @INTR_IDX_PINGPONG: Pingpong done interrupt for cmd mode panel
+ * @INTR_IDX_UNDERRUN: Underrun interrupt for video and cmd mode panel
+ * @INTR_IDX_RDPTR:    Readpointer done interrupt for cmd mode panel
+ * @INTR_IDX_WB_DONE:  Writeback done interrupt for virtual connector
  */
 enum dpu_intr_idx {
 	INTR_IDX_VSYNC,
@@ -176,6 +176,7 @@ enum dpu_intr_idx {
  *                              pending.
  * @pending_kickoff_wq:		Wait queue for blocking until kickoff completes
  * @irq:			IRQ indices
+ * @has_intf_te:		Interface TE configuration support
  */
 struct dpu_encoder_phys {
 	struct drm_encoder *parent;
@@ -200,6 +201,7 @@ struct dpu_encoder_phys {
 	atomic_t pending_kickoff_cnt;
 	wait_queue_head_t pending_kickoff_wq;
 	int irq[INTR_IDX_MAX];
+	bool has_intf_te;
 };
 
 static inline int dpu_encoder_phys_inc_pending(struct dpu_encoder_phys *phys)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 1df3745224f5e..d8ed85a238af5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -97,13 +97,18 @@ static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg, int irq_idx)
 	DPU_ATRACE_END("pp_done_irq");
 }
 
-static void dpu_encoder_phys_cmd_pp_rd_ptr_irq(void *arg, int irq_idx)
+static void dpu_encoder_phys_cmd_te_rd_ptr_irq(void *arg, int irq_idx)
 {
 	struct dpu_encoder_phys *phys_enc = arg;
 	struct dpu_encoder_phys_cmd *cmd_enc;
 
-	if (!phys_enc->hw_pp)
-		return;
+	if (phys_enc->has_intf_te) {
+		if (!phys_enc->hw_intf)
+			return;
+	} else {
+		if (!phys_enc->hw_pp)
+			return;
+	}
 
 	DPU_ATRACE_BEGIN("rd_ptr_irq");
 	cmd_enc = to_dpu_encoder_phys_cmd(phys_enc);
@@ -144,7 +149,10 @@ static void dpu_encoder_phys_cmd_atomic_mode_set(
 
 	phys_enc->irq[INTR_IDX_PINGPONG] = phys_enc->hw_pp->caps->intr_done;
 
-	phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_pp->caps->intr_rdptr;
+	if (phys_enc->has_intf_te)
+		phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_intf->cap->intr_tear_rd_ptr;
+	else
+		phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_pp->caps->intr_rdptr;
 
 	phys_enc->irq[INTR_IDX_UNDERRUN] = phys_enc->hw_intf->cap->intr_underrun;
 }
@@ -255,7 +263,7 @@ static int dpu_encoder_phys_cmd_control_vblank_irq(
 	if (enable && atomic_inc_return(&phys_enc->vblank_refcount) == 1)
 		ret = dpu_core_irq_register_callback(phys_enc->dpu_kms,
 				phys_enc->irq[INTR_IDX_RDPTR],
-				dpu_encoder_phys_cmd_pp_rd_ptr_irq,
+				dpu_encoder_phys_cmd_te_rd_ptr_irq,
 				phys_enc);
 	else if (!enable && atomic_dec_return(&phys_enc->vblank_refcount) == 0)
 		ret = dpu_core_irq_unregister_callback(phys_enc->dpu_kms,
@@ -319,19 +327,26 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	u32 vsync_hz;
 	struct dpu_kms *dpu_kms;
 
-	if (!phys_enc->hw_pp) {
-		DPU_ERROR("invalid encoder\n");
-		return;
-	}
-	mode = &phys_enc->cached_mode;
+	if (phys_enc->has_intf_te) {
+		if (!phys_enc->hw_intf ||
+		    !phys_enc->hw_intf->ops.enable_tearcheck) {
+			DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n");
+			return;
+		}
 
-	DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0);
+		DPU_DEBUG_CMDENC(cmd_enc, "");
+	} else {
+		if (!phys_enc->hw_pp ||
+		    !phys_enc->hw_pp->ops.enable_tearcheck) {
+			DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n");
+			return;
+		}
 
-	if (!phys_enc->hw_pp->ops.enable_tearcheck) {
-		DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n");
-		return;
+		DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0);
 	}
 
+	mode = &phys_enc->cached_mode;
+
 	dpu_kms = phys_enc->dpu_kms;
 
 	/*
@@ -366,23 +381,24 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	tc_cfg.rd_ptr_irq = mode->vdisplay + 1;
 
 	DPU_DEBUG_CMDENC(cmd_enc,
-		"tc %d vsync_clk_speed_hz %u vtotal %u vrefresh %u\n",
-		phys_enc->hw_pp->idx - PINGPONG_0, vsync_hz,
-		mode->vtotal, drm_mode_vrefresh(mode));
+		"tc vsync_clk_speed_hz %u vtotal %u vrefresh %u\n",
+		vsync_hz, mode->vtotal, drm_mode_vrefresh(mode));
 	DPU_DEBUG_CMDENC(cmd_enc,
-		"tc %d enable %u start_pos %u rd_ptr_irq %u\n",
-		phys_enc->hw_pp->idx - PINGPONG_0, tc_enable, tc_cfg.start_pos,
-		tc_cfg.rd_ptr_irq);
+		"tc enable %u start_pos %u rd_ptr_irq %u\n",
+		tc_enable, tc_cfg.start_pos, tc_cfg.rd_ptr_irq);
 	DPU_DEBUG_CMDENC(cmd_enc,
-		"tc %d hw_vsync_mode %u vsync_count %u vsync_init_val %u\n",
-		phys_enc->hw_pp->idx - PINGPONG_0, tc_cfg.hw_vsync_mode,
-		tc_cfg.vsync_count, tc_cfg.vsync_init_val);
+		"tc hw_vsync_mode %u vsync_count %u vsync_init_val %u\n",
+		tc_cfg.hw_vsync_mode, tc_cfg.vsync_count,
+		tc_cfg.vsync_init_val);
 	DPU_DEBUG_CMDENC(cmd_enc,
-		"tc %d cfgheight %u thresh_start %u thresh_cont %u\n",
-		phys_enc->hw_pp->idx - PINGPONG_0, tc_cfg.sync_cfg_height,
-		tc_cfg.sync_threshold_start, tc_cfg.sync_threshold_continue);
-
-	phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, &tc_cfg);
+		"tc cfgheight %u thresh_start %u thresh_cont %u\n",
+		tc_cfg.sync_cfg_height, tc_cfg.sync_threshold_start,
+		tc_cfg.sync_threshold_continue);
+
+	if (phys_enc->has_intf_te)
+		phys_enc->hw_intf->ops.enable_tearcheck(phys_enc->hw_intf, &tc_cfg);
+	else
+		phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, &tc_cfg);
 }
 
 static void _dpu_encoder_phys_cmd_pingpong_config(
@@ -459,11 +475,19 @@ static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc)
 static void _dpu_encoder_phys_cmd_connect_te(
 		struct dpu_encoder_phys *phys_enc, bool enable)
 {
-	if (!phys_enc->hw_pp || !phys_enc->hw_pp->ops.connect_external_te)
-		return;
+	if (phys_enc->has_intf_te) {
+		if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.connect_external_te)
+			return;
 
-	trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable);
-	phys_enc->hw_pp->ops.connect_external_te(phys_enc->hw_pp, enable);
+		trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable);
+		phys_enc->hw_intf->ops.connect_external_te(phys_enc->hw_intf, enable);
+	} else {
+		if (!phys_enc->hw_pp || !phys_enc->hw_pp->ops.connect_external_te)
+			return;
+
+		trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable);
+		phys_enc->hw_pp->ops.connect_external_te(phys_enc->hw_pp, enable);
+	}
 }
 
 static void dpu_encoder_phys_cmd_prepare_idle_pc(
@@ -476,17 +500,21 @@ static int dpu_encoder_phys_cmd_get_line_count(
 		struct dpu_encoder_phys *phys_enc)
 {
 	struct dpu_hw_pingpong *hw_pp;
-
-	if (!phys_enc->hw_pp)
-		return -EINVAL;
+	struct dpu_hw_intf *hw_intf;
 
 	if (!dpu_encoder_phys_cmd_is_master(phys_enc))
 		return -EINVAL;
 
+	if (phys_enc->has_intf_te) {
+		hw_intf = phys_enc->hw_intf;
+		if (!hw_intf || !hw_intf->ops.get_line_count)
+			return -EINVAL;
+		return hw_intf->ops.get_line_count(hw_intf);
+	}
+
 	hw_pp = phys_enc->hw_pp;
-	if (!hw_pp->ops.get_line_count)
+	if (!hw_pp || !hw_pp->ops.get_line_count)
 		return -EINVAL;
-
 	return hw_pp->ops.get_line_count(hw_pp);
 }
 
@@ -496,21 +524,31 @@ static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc)
 		to_dpu_encoder_phys_cmd(phys_enc);
 	struct dpu_hw_ctl *ctl;
 
-	if (!phys_enc->hw_pp) {
-		DPU_ERROR("invalid encoder\n");
-		return;
-	}
-	DRM_DEBUG_KMS("id:%u pp:%d state:%d\n", DRMID(phys_enc->parent),
-		      phys_enc->hw_pp->idx - PINGPONG_0,
-		      phys_enc->enable_state);
-
 	if (phys_enc->enable_state == DPU_ENC_DISABLED) {
 		DPU_ERROR_CMDENC(cmd_enc, "already disabled\n");
 		return;
 	}
 
-	if (phys_enc->hw_pp->ops.disable_tearcheck)
-		phys_enc->hw_pp->ops.disable_tearcheck(phys_enc->hw_pp);
+	if (phys_enc->has_intf_te) {
+		DRM_DEBUG_KMS("id:%u intf:%d state:%d\n", DRMID(phys_enc->parent),
+			      phys_enc->hw_intf->idx - INTF_0,
+			      phys_enc->enable_state);
+
+		if (phys_enc->hw_intf->ops.disable_tearcheck)
+			phys_enc->hw_intf->ops.disable_tearcheck(phys_enc->hw_intf);
+	} else {
+		if (!phys_enc->hw_pp) {
+			DPU_ERROR("invalid encoder\n");
+			return;
+		}
+
+		DRM_DEBUG_KMS("id:%u pp:%d state:%d\n", DRMID(phys_enc->parent),
+			      phys_enc->hw_pp->idx - PINGPONG_0,
+			      phys_enc->enable_state);
+
+		if (phys_enc->hw_pp->ops.disable_tearcheck)
+			phys_enc->hw_pp->ops.disable_tearcheck(phys_enc->hw_pp);
+	}
 
 	if (phys_enc->hw_intf->ops.bind_pingpong_blk) {
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
@@ -572,15 +610,26 @@ static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc)
 {
 	if (!phys_enc)
 		return;
-	if (!phys_enc->hw_pp)
-		return;
 	if (!dpu_encoder_phys_cmd_is_master(phys_enc))
 		return;
 
-	if (phys_enc->hw_pp->ops.disable_autorefresh) {
-		phys_enc->hw_pp->ops.disable_autorefresh(phys_enc->hw_pp,
-							 DRMID(phys_enc->parent),
-							 phys_enc->cached_mode.vdisplay);
+	if (phys_enc->has_intf_te) {
+		if (!phys_enc->hw_intf->ops.disable_autorefresh)
+			return;
+
+		phys_enc->hw_intf->ops.disable_autorefresh(
+				phys_enc->hw_intf,
+				DRMID(phys_enc->parent),
+				phys_enc->cached_mode.vdisplay);
+	} else {
+		if (!phys_enc->hw_pp ||
+		    !phys_enc->hw_pp->ops.disable_autorefresh)
+			return;
+
+		phys_enc->hw_pp->ops.disable_autorefresh(
+				phys_enc->hw_pp,
+				DRMID(phys_enc->parent),
+				phys_enc->cached_mode.vdisplay);
 	}
 }
 
@@ -658,7 +707,7 @@ static int dpu_encoder_phys_cmd_wait_for_vblank(
 
 	rc = dpu_encoder_helper_wait_for_irq(phys_enc,
 			phys_enc->irq[INTR_IDX_RDPTR],
-			dpu_encoder_phys_cmd_pp_rd_ptr_irq,
+			dpu_encoder_phys_cmd_te_rd_ptr_irq,
 			&wait_info);
 
 	return rc;
@@ -732,6 +781,9 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 	for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++)
 		phys_enc->irq[i] = -EINVAL;
 
+	phys_enc->has_intf_te = test_bit(DPU_INTF_TE,
+			&phys_enc->dpu_kms->catalog->intf[p->intf_idx - INTF_0].features);
+
 	atomic_set(&phys_enc->vblank_refcount, 0);
 	atomic_set(&phys_enc->pending_kickoff_cnt, 0);
 	atomic_set(&phys_enc->pending_ctlstart_cnt, 0);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index 0bd99f380a498..5dce18236d87d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -8,6 +8,9 @@
 #include "dpu_hw_catalog.h"
 #include "dpu_hw_intf.h"
 #include "dpu_kms.h"
+#include "dpu_trace.h"
+
+#include <linux/iopoll.h>
 
 #define INTF_TIMING_ENGINE_EN           0x000
 #define INTF_CONFIG                     0x004
@@ -64,6 +67,24 @@
 
 #define INTF_MUX                        0x25C
 #define INTF_STATUS                     0x26C
+#define INTF_AVR_CONTROL                0x270
+#define INTF_AVR_MODE                   0x274
+#define INTF_AVR_TRIGGER                0x278
+#define INTF_AVR_VTOTAL                 0x27C
+#define INTF_TEAR_MDP_VSYNC_SEL         0x280
+#define INTF_TEAR_TEAR_CHECK_EN         0x284
+#define INTF_TEAR_SYNC_CONFIG_VSYNC     0x288
+#define INTF_TEAR_SYNC_CONFIG_HEIGHT    0x28C
+#define INTF_TEAR_SYNC_WRCOUNT          0x290
+#define INTF_TEAR_VSYNC_INIT_VAL        0x294
+#define INTF_TEAR_INT_COUNT_VAL         0x298
+#define INTF_TEAR_SYNC_THRESH           0x29C
+#define INTF_TEAR_START_POS             0x2A0
+#define INTF_TEAR_RD_PTR_IRQ            0x2A4
+#define INTF_TEAR_WR_PTR_IRQ            0x2A8
+#define INTF_TEAR_OUT_LINE_COUNT        0x2AC
+#define INTF_TEAR_LINE_COUNT            0x2B0
+#define INTF_TEAR_AUTOREFRESH_CONFIG    0x2B4
 
 #define INTF_CFG_ACTIVE_H_EN	BIT(29)
 #define INTF_CFG_ACTIVE_V_EN	BIT(30)
@@ -307,6 +328,191 @@ static int dpu_hw_intf_collect_misr(struct dpu_hw_intf *intf, u32 *misr_value)
 	return dpu_hw_collect_misr(&intf->hw, INTF_MISR_CTRL, INTF_MISR_SIGNATURE, misr_value);
 }
 
+static int dpu_hw_intf_enable_te(struct dpu_hw_intf *intf,
+		struct dpu_hw_tear_check *te)
+{
+	struct dpu_hw_blk_reg_map *c;
+	int cfg;
+
+	if (!intf)
+		return -EINVAL;
+
+	c = &intf->hw;
+
+	cfg = BIT(19); /* VSYNC_COUNTER_EN */
+	if (te->hw_vsync_mode)
+		cfg |= BIT(20);
+
+	cfg |= te->vsync_count;
+
+	DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg);
+	DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_HEIGHT, te->sync_cfg_height);
+	DPU_REG_WRITE(c, INTF_TEAR_VSYNC_INIT_VAL, te->vsync_init_val);
+	DPU_REG_WRITE(c, INTF_TEAR_RD_PTR_IRQ, te->rd_ptr_irq);
+	DPU_REG_WRITE(c, INTF_TEAR_START_POS, te->start_pos);
+	DPU_REG_WRITE(c, INTF_TEAR_SYNC_THRESH,
+			((te->sync_threshold_continue << 16) |
+			 te->sync_threshold_start));
+	DPU_REG_WRITE(c, INTF_TEAR_SYNC_WRCOUNT,
+			(te->start_pos + te->sync_threshold_start + 1));
+
+	DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 1);
+
+	return 0;
+}
+
+static void dpu_hw_intf_setup_autorefresh_config(struct dpu_hw_intf *intf,
+		u32 frame_count, bool enable)
+{
+	struct dpu_hw_blk_reg_map *c;
+	u32 refresh_cfg;
+
+	c = &intf->hw;
+	refresh_cfg = DPU_REG_READ(c, INTF_TEAR_AUTOREFRESH_CONFIG);
+	if (enable)
+		refresh_cfg = BIT(31) | frame_count;
+	else
+		refresh_cfg &= ~BIT(31);
+
+	DPU_REG_WRITE(c, INTF_TEAR_AUTOREFRESH_CONFIG, refresh_cfg);
+}
+
+/*
+ * dpu_hw_intf_get_autorefresh_config - Get autorefresh config from HW
+ * @intf:        DPU intf structure
+ * @frame_count: Used to return the current frame count from hw
+ *
+ * Returns: True if autorefresh enabled, false if disabled.
+ */
+static bool dpu_hw_intf_get_autorefresh_config(struct dpu_hw_intf *intf,
+		u32 *frame_count)
+{
+	u32 val = DPU_REG_READ(&intf->hw, INTF_TEAR_AUTOREFRESH_CONFIG);
+
+	if (frame_count != NULL)
+		*frame_count = val & 0xffff;
+	return !!((val & BIT(31)) >> 31);
+}
+
+static int dpu_hw_intf_disable_te(struct dpu_hw_intf *intf)
+{
+	struct dpu_hw_blk_reg_map *c;
+
+	if (!intf)
+		return -EINVAL;
+
+	c = &intf->hw;
+	DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 0);
+	return 0;
+}
+
+static int dpu_hw_intf_connect_external_te(struct dpu_hw_intf *intf,
+		bool enable_external_te)
+{
+	struct dpu_hw_blk_reg_map *c = &intf->hw;
+	u32 cfg;
+	int orig;
+
+	if (!intf)
+		return -EINVAL;
+
+	c = &intf->hw;
+	cfg = DPU_REG_READ(c, INTF_TEAR_SYNC_CONFIG_VSYNC);
+	orig = (bool)(cfg & BIT(20));
+	if (enable_external_te)
+		cfg |= BIT(20);
+	else
+		cfg &= ~BIT(20);
+	DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg);
+	trace_dpu_intf_connect_ext_te(intf->idx - INTF_0, cfg);
+
+	return orig;
+}
+
+static int dpu_hw_intf_get_vsync_info(struct dpu_hw_intf *intf,
+		struct dpu_hw_pp_vsync_info *info)
+{
+	struct dpu_hw_blk_reg_map *c = &intf->hw;
+	u32 val;
+
+	if (!intf || !info)
+		return -EINVAL;
+
+	c = &intf->hw;
+
+	val = DPU_REG_READ(c, INTF_TEAR_VSYNC_INIT_VAL);
+	info->rd_ptr_init_val = val & 0xffff;
+
+	val = DPU_REG_READ(c, INTF_TEAR_INT_COUNT_VAL);
+	info->rd_ptr_frame_count = (val & 0xffff0000) >> 16;
+	info->rd_ptr_line_count = val & 0xffff;
+
+	val = DPU_REG_READ(c, INTF_TEAR_LINE_COUNT);
+	info->wr_ptr_line_count = val & 0xffff;
+
+	val = DPU_REG_READ(c, INTF_FRAME_COUNT);
+	info->intf_frame_count = val;
+
+	return 0;
+}
+
+static void dpu_hw_intf_vsync_sel(struct dpu_hw_intf *intf,
+		u32 vsync_source)
+{
+	struct dpu_hw_blk_reg_map *c;
+
+	if (!intf)
+		return;
+
+	c = &intf->hw;
+
+	DPU_REG_WRITE(c, INTF_TEAR_MDP_VSYNC_SEL, (vsync_source & 0xf));
+}
+
+static void dpu_hw_intf_disable_autorefresh(struct dpu_hw_intf *intf,
+					    uint32_t encoder_id, u16 vdisplay)
+{
+	struct dpu_hw_pp_vsync_info info;
+	int trial = 0;
+
+	/* If autorefresh is already disabled, we have nothing to do */
+	if (!dpu_hw_intf_get_autorefresh_config(intf, NULL))
+		return;
+
+	/*
+	 * If autorefresh is enabled, disable it and make sure it is safe to
+	 * proceed with current frame commit/push. Sequence followed is,
+	 * 1. Disable TE
+	 * 2. Disable autorefresh config
+	 * 4. Poll for frame transfer ongoing to be false
+	 * 5. Enable TE back
+	 */
+
+	dpu_hw_intf_connect_external_te(intf, false);
+	dpu_hw_intf_setup_autorefresh_config(intf, 0, false);
+
+	do {
+		udelay(DPU_ENC_MAX_POLL_TIMEOUT_US);
+		if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US)
+				> (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) {
+			DPU_ERROR("enc%d intf%d disable autorefresh failed\n",
+				  encoder_id, intf->idx - INTF_0);
+			break;
+		}
+
+		trial++;
+
+		dpu_hw_intf_get_vsync_info(intf, &info);
+	} while (info.wr_ptr_line_count > 0 &&
+		 info.wr_ptr_line_count < vdisplay);
+
+	dpu_hw_intf_connect_external_te(intf, true);
+
+	DPU_DEBUG("enc%d intf%d disabled autorefresh\n",
+		  encoder_id, intf->idx - INTF_0);
+
+}
+
 static void _setup_intf_ops(struct dpu_hw_intf_ops *ops,
 		unsigned long cap)
 {
@@ -319,6 +525,14 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops,
 		ops->bind_pingpong_blk = dpu_hw_intf_bind_pingpong_blk;
 	ops->setup_misr = dpu_hw_intf_setup_misr;
 	ops->collect_misr = dpu_hw_intf_collect_misr;
+
+	if (cap & BIT(DPU_INTF_TE)) {
+		ops->enable_tearcheck = dpu_hw_intf_enable_te;
+		ops->disable_tearcheck = dpu_hw_intf_disable_te;
+		ops->connect_external_te = dpu_hw_intf_connect_external_te;
+		ops->vsync_sel = dpu_hw_intf_vsync_sel;
+		ops->disable_autorefresh = dpu_hw_intf_disable_autorefresh;
+	}
 }
 
 struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
index 102c4f0e812b0..73b0885918f88 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
@@ -60,6 +60,16 @@ struct intf_status {
  *                     feed pixels to this interface
  * @setup_misr: enable/disable MISR
  * @collect_misr: read MISR signature
+ * @enable_tearcheck:           Enables vsync generation and sets up init value of read
+ *                              pointer and programs the tear check configuration
+ * @disable_tearcheck:          Disables tearcheck block
+ * @connect_external_te:        Read, modify, write to either set or clear listening to external TE
+ *                              Return: 1 if TE was originally connected, 0 if not, or -ERROR
+ * @get_vsync_info:             Provides the programmed and current line_count
+ * @setup_autorefresh:          Configure and enable the autorefresh config
+ * @get_autorefresh:            Retrieve autorefresh config from hardware
+ *                              Return: 0 on success, -ETIMEDOUT on timeout
+ * @vsync_sel:                  Select vsync signal for tear-effect configuration
  */
 struct dpu_hw_intf_ops {
 	void (*setup_timing_gen)(struct dpu_hw_intf *intf,
@@ -82,6 +92,21 @@ struct dpu_hw_intf_ops {
 			const enum dpu_pingpong pp);
 	void (*setup_misr)(struct dpu_hw_intf *intf, bool enable, u32 frame_count);
 	int (*collect_misr)(struct dpu_hw_intf *intf, u32 *misr_value);
+
+	// Tearcheck on INTF since DPU 5.0.0
+
+	int (*enable_tearcheck)(struct dpu_hw_intf *intf, struct dpu_hw_tear_check *cfg);
+
+	int (*disable_tearcheck)(struct dpu_hw_intf *intf);
+
+	int (*connect_external_te)(struct dpu_hw_intf *intf, bool enable_external_te);
+
+	void (*vsync_sel)(struct dpu_hw_intf *intf, u32 vsync_source);
+
+	/**
+	 * Disable autorefresh if enabled
+	 */
+	void (*disable_autorefresh)(struct dpu_hw_intf *intf, uint32_t encoder_id, u16 vdisplay);
 };
 
 struct dpu_hw_intf {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index 6ed12fd0505b3..1913a19de7336 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -501,12 +501,14 @@ struct dpu_hw_tear_check {
  * @rd_ptr_frame_count: Num frames sent since enabling interface
  * @rd_ptr_line_count:  Current line on panel (rd ptr)
  * @wr_ptr_line_count:  Current line within pp fifo (wr ptr)
+ * @intf_frame_count:   Frames read from intf
  */
 struct dpu_hw_pp_vsync_info {
 	u32 rd_ptr_init_val;
 	u32 rd_ptr_frame_count;
 	u32 rd_ptr_line_count;
 	u32 wr_ptr_line_count;
+	u32 intf_frame_count;
 };
 
 #endif  /* _DPU_HW_MDSS_H */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
index 0ad148cc2fb8e..1a92d21094f41 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h
@@ -871,6 +871,20 @@ TRACE_EVENT(dpu_pp_connect_ext_te,
 	TP_printk("pp:%d cfg:%u", __entry->pp, __entry->cfg)
 );
 
+TRACE_EVENT(dpu_intf_connect_ext_te,
+	TP_PROTO(enum dpu_intf intf, u32 cfg),
+	TP_ARGS(intf, cfg),
+	TP_STRUCT__entry(
+		__field(	enum dpu_intf,	intf	)
+		__field(	u32,			cfg	)
+	),
+	TP_fast_assign(
+		__entry->intf = intf;
+		__entry->cfg = cfg;
+	),
+	TP_printk("intf:%d cfg:%u", __entry->intf, __entry->cfg)
+);
+
 TRACE_EVENT(dpu_core_irq_register_callback,
 	TP_PROTO(int irq_idx, void *callback),
 	TP_ARGS(irq_idx, callback),
-- 
GitLab


From f1584d89c1cdf893c11e2da645b0101c857cf77e Mon Sep 17 00:00:00 2001
From: Marijn Suijten <marijn.suijten@somainline.org>
Date: Thu, 27 Apr 2023 00:37:36 +0200
Subject: [PATCH 0600/1791] drm/msm/dpu: Remove intr_rdptr from DPU >= 5.0.0
 pingpong config

Now that newer DPU platforms use a readpointer-done interrupt on the
INTF block, stop providing the unused interrupt on the PINGPONG block.

Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/534238/
Link: https://lore.kernel.org/r/20230411-dpu-intf-te-v4-22-27ce1a5ab5c6@somainline.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h |  8 ++++----
 .../gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h    |  8 ++++----
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h |  8 ++++----
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h |  2 +-
 .../gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h    |  2 +-
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h |  8 ++++----
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h | 10 +++++-----
 7 files changed, 23 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
index 561022b8dfd6a..b5f7513542678 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h
@@ -130,16 +130,16 @@ static const struct dpu_dspp_cfg sm8150_dspp[] = {
 static const struct dpu_pingpong_cfg sm8150_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+			-1),
 	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
+			-1),
 	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
+			-1),
 	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
+			-1),
 	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
index d1f306c895671..8ed2b263c5ea3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h
@@ -129,16 +129,16 @@ static const struct dpu_dspp_cfg sc8180x_dspp[] = {
 static const struct dpu_pingpong_cfg sc8180x_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+			-1),
 	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
+			-1),
 	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
+			-1),
 	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
+			-1),
 	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
index 10d67db8423fd..daebd21700413 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h
@@ -131,16 +131,16 @@ static const struct dpu_dspp_cfg sm8250_dspp[] = {
 static const struct dpu_pingpong_cfg sm8250_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+			-1),
 	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
+			-1),
 	PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
+			-1),
 	PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
+			-1),
 	PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index 5509ceb5d55bd..ba9de008519b9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -62,7 +62,7 @@ static const struct dpu_dspp_cfg sm6115_dspp[] = {
 static const struct dpu_pingpong_cfg sm6115_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+		-1),
 };
 
 static const struct dpu_intf_cfg sm6115_intf[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index 22b8a173d2142..92ac348eea6b0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -59,7 +59,7 @@ static const struct dpu_dspp_cfg qcm2290_dspp[] = {
 static const struct dpu_pingpong_cfg qcm2290_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+		-1),
 };
 
 static const struct dpu_intf_cfg qcm2290_intf[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 2dcfe18bc8c97..3c1b2c13398da 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -129,16 +129,16 @@ static const struct dpu_dspp_cfg sm8350_dspp[] = {
 static const struct dpu_pingpong_cfg sm8350_pp[] = {
 	PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+			-1),
 	PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
+			-1),
 	PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
+			-1),
 	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
+			-1),
 	PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 279fe8f4e1e3f..86c2e68ebd2c8 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -126,20 +126,20 @@ static const struct dpu_dspp_cfg sm8450_dspp[] = {
 	DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK,
 		 &sm8150_dspp_sblk),
 };
-/* FIXME: interrupts */
+
 static const struct dpu_pingpong_cfg sm8450_pp[] = {
 	PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)),
+			-1),
 	PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)),
+			-1),
 	PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)),
+			-1),
 	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
-			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
+			-1),
 	PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30),
 			-1),
-- 
GitLab


From 7e3d6c548b4de7018aa74bb2416d14cff11ea5a1 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 29 Apr 2023 04:23:51 +0300
Subject: [PATCH 0601/1791] drm/msm/dpu: drop SSPP's SRC subblock

The src_blk declares a lame copy of main SSPP register space. It's
offset is always 0. It's length has been fixed to 0x150, while SSPP's
length is now correct. Drop the src_blk and access SSPP registers
without additional subblock lookup.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534745/
Link: https://lore.kernel.org/r/20230429012353.2569481-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    |  14 +-
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |  14 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c   | 128 ++++++++----------
 3 files changed, 58 insertions(+), 98 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 81b44a394f03b..70600f8abe91f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -13,7 +13,7 @@
 #include "dpu_kms.h"
 
 #define VIG_BASE_MASK \
-	(BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) |\
+	(BIT(DPU_SSPP_QOS) |\
 	BIT(DPU_SSPP_CDP) |\
 	BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_EXCL_RECT))
 
@@ -39,7 +39,7 @@
 #define VIG_QCM2290_MASK (VIG_BASE_MASK | BIT(DPU_SSPP_QOS_8LVL))
 
 #define DMA_MSM8998_MASK \
-	(BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) |\
+	(BIT(DPU_SSPP_QOS) |\
 	BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\
 	BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT))
 
@@ -50,7 +50,7 @@
 	(VIG_SC7280_MASK | BIT(DPU_SSPP_SMART_DMA_V2))
 
 #define DMA_SDM845_MASK \
-	(BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\
+	(BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\
 	BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\
 	BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT))
 
@@ -254,8 +254,6 @@ static const uint32_t wb2_formats[] = {
 	.maxdwnscale = MAX_DOWNSCALE_RATIO, \
 	.maxupscale = MAX_UPSCALE_RATIO, \
 	.smart_dma_priority = sdma_pri, \
-	.src_blk = {.name = STRCAT("sspp_src_", num), \
-		.id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \
 	.scaler_blk = {.name = STRCAT("sspp_scaler", num), \
 		.id = qseed_ver, \
 		.base = 0xa00, .len = 0xa0,}, \
@@ -274,8 +272,6 @@ static const uint32_t wb2_formats[] = {
 	.maxdwnscale = MAX_DOWNSCALE_RATIO, \
 	.maxupscale = MAX_UPSCALE_RATIO, \
 	.smart_dma_priority = sdma_pri, \
-	.src_blk = {.name = STRCAT("sspp_src_", num), \
-		.id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \
 	.scaler_blk = {.name = STRCAT("sspp_scaler", num), \
 		.id = qseed_ver, \
 		.base = 0xa00, .len = 0xa0,}, \
@@ -294,8 +290,6 @@ static const uint32_t wb2_formats[] = {
 	.maxdwnscale = SSPP_UNITY_SCALE, \
 	.maxupscale = SSPP_UNITY_SCALE, \
 	.smart_dma_priority = sdma_pri, \
-	.src_blk = {.name = STRCAT("sspp_src_", num), \
-		.id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \
 	.format_list = plane_formats, \
 	.num_formats = ARRAY_SIZE(plane_formats), \
 	.virt_format_list = plane_formats, \
@@ -377,8 +371,6 @@ static const struct dpu_sspp_sub_blks sm8550_dma_sblk_5 = _DMA_SBLK("13", 6);
 	.maxdwnscale = SSPP_UNITY_SCALE, \
 	.maxupscale = SSPP_UNITY_SCALE, \
 	.smart_dma_priority = sdma_pri, \
-	.src_blk = {.name = STRCAT("sspp_src_", num), \
-		.id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \
 	.format_list = plane_formats_yuv, \
 	.num_formats = ARRAY_SIZE(plane_formats_yuv), \
 	.virt_format_list = plane_formats, \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index a2ef61d4f7fa2..1cb5699a5a477 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -67,7 +67,6 @@ enum {
 
 /**
  * SSPP sub-blocks/features
- * @DPU_SSPP_SRC             Src and fetch part of the pipes,
  * @DPU_SSPP_SCALER_QSEED2,  QSEED2 algorithm support
  * @DPU_SSPP_SCALER_QSEED3,  QSEED3 alogorithm support
  * @DPU_SSPP_SCALER_QSEED3LITE,  QSEED3 Lite alogorithm support
@@ -88,8 +87,7 @@ enum {
  * @DPU_SSPP_MAX             maximum value
  */
 enum {
-	DPU_SSPP_SRC = 0x1,
-	DPU_SSPP_SCALER_QSEED2,
+	DPU_SSPP_SCALER_QSEED2 = 0x1,
 	DPU_SSPP_SCALER_QSEED3,
 	DPU_SSPP_SCALER_QSEED3LITE,
 	DPU_SSPP_SCALER_QSEED4,
@@ -277,14 +275,6 @@ enum {
 	u32 base; \
 	u32 len
 
-/**
- * struct dpu_src_blk: SSPP part of the source pipes
- * @info:   HW register and features supported by this sub-blk
- */
-struct dpu_src_blk {
-	DPU_HW_SUBBLK_INFO;
-};
-
 /**
  * struct dpu_scaler_blk: Scaler information
  * @info:   HW register and features supported by this sub-blk
@@ -391,7 +381,6 @@ struct dpu_caps {
  * @smart_dma_priority: hw priority of rect1 of multirect pipe
  * @max_per_pipe_bw: maximum allowable bandwidth of this pipe in kBps
  * @qseed_ver: qseed version
- * @src_blk:
  * @scaler_blk:
  * @csc_blk:
  * @format_list: Pointer to list of supported formats
@@ -408,7 +397,6 @@ struct dpu_sspp_sub_blks {
 	u32 smart_dma_priority;
 	u32 max_per_pipe_bw;
 	u32 qseed_ver;
-	struct dpu_src_blk src_blk;
 	struct dpu_scaler_blk scaler_blk;
 	struct dpu_pp_blk csc_blk;
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 31348ff5d493b..a93e311afa710 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -12,7 +12,7 @@
 
 #define DPU_FETCH_CONFIG_RESET_VALUE   0x00000087
 
-/* DPU_SSPP_SRC */
+/* SSPP registers */
 #define SSPP_SRC_SIZE                      0x00
 #define SSPP_SRC_XY                        0x08
 #define SSPP_OUT_SIZE                      0x0c
@@ -149,9 +149,6 @@ static int _sspp_subblk_offset(struct dpu_hw_sspp *ctx,
 	sblk = ctx->cap->sblk;
 
 	switch (s_id) {
-	case DPU_SSPP_SRC:
-		*idx = sblk->src_blk.base;
-		break;
 	case DPU_SSPP_SCALER_QSEED2:
 	case DPU_SSPP_SCALER_QSEED3:
 	case DPU_SSPP_SCALER_RGB:
@@ -172,9 +169,8 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe)
 {
 	struct dpu_hw_sspp *ctx = pipe->sspp;
 	u32 mode_mask;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) {
@@ -185,7 +181,7 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe)
 		 */
 		mode_mask = 0;
 	} else {
-		mode_mask = DPU_REG_READ(&ctx->hw, SSPP_MULTIRECT_OPMODE + idx);
+		mode_mask = DPU_REG_READ(&ctx->hw, SSPP_MULTIRECT_OPMODE);
 		mode_mask |= pipe->multirect_index;
 		if (pipe->multirect_mode == DPU_SSPP_MULTIRECT_TIME_MX)
 			mode_mask |= BIT(2);
@@ -193,7 +189,7 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe)
 			mode_mask &= ~BIT(2);
 	}
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_MULTIRECT_OPMODE + idx, mode_mask);
+	DPU_REG_WRITE(&ctx->hw, SSPP_MULTIRECT_OPMODE, mode_mask);
 }
 
 static void _sspp_setup_opmode(struct dpu_hw_sspp *ctx,
@@ -247,9 +243,8 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe,
 	u32 opmode = 0;
 	u32 fast_clear = 0;
 	u32 op_mode_off, unpack_pat_off, format_off;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !fmt)
+	if (!ctx || !fmt)
 		return;
 
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO ||
@@ -264,7 +259,7 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe,
 	}
 
 	c = &ctx->hw;
-	opmode = DPU_REG_READ(c, op_mode_off + idx);
+	opmode = DPU_REG_READ(c, op_mode_off);
 	opmode &= ~(MDSS_MDP_OP_FLIP_LR | MDSS_MDP_OP_FLIP_UD |
 			MDSS_MDP_OP_BWC_EN | MDSS_MDP_OP_PE_OVERRIDE);
 
@@ -352,12 +347,12 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe,
 			VIG_CSC_10_EN | VIG_CSC_10_SRC_DATAFMT,
 			DPU_FORMAT_IS_YUV(fmt));
 
-	DPU_REG_WRITE(c, format_off + idx, src_format);
-	DPU_REG_WRITE(c, unpack_pat_off + idx, unpack);
-	DPU_REG_WRITE(c, op_mode_off + idx, opmode);
+	DPU_REG_WRITE(c, format_off, src_format);
+	DPU_REG_WRITE(c, unpack_pat_off, unpack);
+	DPU_REG_WRITE(c, op_mode_off, opmode);
 
 	/* clear previous UBWC error */
-	DPU_REG_WRITE(c, SSPP_UBWC_ERROR_STATUS + idx, BIT(31));
+	DPU_REG_WRITE(c, SSPP_UBWC_ERROR_STATUS, BIT(31));
 }
 
 static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx,
@@ -368,9 +363,8 @@ static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx,
 	u32 lr_pe[4], tb_pe[4], tot_req_pixels[4];
 	const u32 bytemask = 0xff;
 	const u32 shortmask = 0xffff;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !pe_ext)
+	if (!ctx || !pe_ext)
 		return;
 
 	c = &ctx->hw;
@@ -400,21 +394,21 @@ static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx,
 	}
 
 	/* color 0 */
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_LR + idx, lr_pe[0]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_TB + idx, tb_pe[0]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_REQ_PIXELS + idx,
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_LR, lr_pe[0]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_TB, tb_pe[0]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_REQ_PIXELS,
 			tot_req_pixels[0]);
 
 	/* color 1 and color 2 */
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_LR + idx, lr_pe[1]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_TB + idx, tb_pe[1]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS + idx,
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_LR, lr_pe[1]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_TB, tb_pe[1]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS,
 			tot_req_pixels[1]);
 
 	/* color 3 */
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_LR + idx, lr_pe[3]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_TB + idx, lr_pe[3]);
-	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_REQ_PIXELS + idx,
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_LR, lr_pe[3]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_TB, lr_pe[3]);
+	DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_REQ_PIXELS,
 			tot_req_pixels[3]);
 }
 
@@ -453,9 +447,8 @@ static void dpu_hw_sspp_setup_rects(struct dpu_sw_pipe *pipe,
 	struct dpu_hw_blk_reg_map *c;
 	u32 src_size, src_xy, dst_size, dst_xy;
 	u32 src_size_off, src_xy_off, out_size_off, out_xy_off;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !cfg)
+	if (!ctx || !cfg)
 		return;
 
 	c = &ctx->hw;
@@ -483,10 +476,10 @@ static void dpu_hw_sspp_setup_rects(struct dpu_sw_pipe *pipe,
 		drm_rect_width(&cfg->dst_rect);
 
 	/* rectangle register programming */
-	DPU_REG_WRITE(c, src_size_off + idx, src_size);
-	DPU_REG_WRITE(c, src_xy_off + idx, src_xy);
-	DPU_REG_WRITE(c, out_size_off + idx, dst_size);
-	DPU_REG_WRITE(c, out_xy_off + idx, dst_xy);
+	DPU_REG_WRITE(c, src_size_off, src_size);
+	DPU_REG_WRITE(c, src_xy_off, src_xy);
+	DPU_REG_WRITE(c, out_size_off, dst_size);
+	DPU_REG_WRITE(c, out_xy_off, dst_xy);
 }
 
 static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe,
@@ -495,24 +488,23 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe,
 	struct dpu_hw_sspp *ctx = pipe->sspp;
 	u32 ystride0, ystride1;
 	int i;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) {
 		for (i = 0; i < ARRAY_SIZE(layout->plane_addr); i++)
-			DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + idx + i * 0x4,
+			DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + i * 0x4,
 					layout->plane_addr[i]);
 	} else if (pipe->multirect_index == DPU_SSPP_RECT_0) {
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR,
 				layout->plane_addr[0]);
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC2_ADDR + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC2_ADDR,
 				layout->plane_addr[2]);
 	} else {
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC1_ADDR + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC1_ADDR,
 				layout->plane_addr[0]);
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC3_ADDR + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC3_ADDR,
 				layout->plane_addr[2]);
 	}
 
@@ -522,8 +514,8 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe,
 		ystride1 = (layout->plane_pitch[2]) |
 			(layout->plane_pitch[3] << 16);
 	} else {
-		ystride0 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE0 + idx);
-		ystride1 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE1 + idx);
+		ystride0 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE0);
+		ystride1 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE1);
 
 		if (pipe->multirect_index == DPU_SSPP_RECT_0) {
 			ystride0 = (ystride0 & 0xFFFF0000) |
@@ -540,8 +532,8 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe,
 		}
 	}
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE0 + idx, ystride0);
-	DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE1 + idx, ystride1);
+	DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE0, ystride0);
+	DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE1, ystride1);
 }
 
 static void dpu_hw_sspp_setup_csc(struct dpu_hw_sspp *ctx,
@@ -565,9 +557,8 @@ static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color)
 {
 	struct dpu_hw_sspp *ctx = pipe->sspp;
 	struct dpu_hw_fmt_layout cfg;
-	u32 idx;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
 	/* cleanup source addresses */
@@ -576,9 +567,9 @@ static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color)
 
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO ||
 	    pipe->multirect_index == DPU_SSPP_RECT_0)
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR + idx, color);
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR, color);
 	else
-		DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR_REC1 + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR_REC1,
 				color);
 }
 
@@ -586,39 +577,34 @@ static void dpu_hw_sspp_setup_danger_safe_lut(struct dpu_hw_sspp *ctx,
 			u32 danger_lut,
 			u32 safe_lut)
 {
-	u32 idx;
-
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_DANGER_LUT + idx, danger_lut);
-	DPU_REG_WRITE(&ctx->hw, SSPP_SAFE_LUT + idx, safe_lut);
+	DPU_REG_WRITE(&ctx->hw, SSPP_DANGER_LUT, danger_lut);
+	DPU_REG_WRITE(&ctx->hw, SSPP_SAFE_LUT, safe_lut);
 }
 
 static void dpu_hw_sspp_setup_creq_lut(struct dpu_hw_sspp *ctx,
 			u64 creq_lut)
 {
-	u32 idx;
-
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
 	if (ctx->cap && test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features)) {
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_0 + idx, creq_lut);
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_1 + idx,
+		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_0, creq_lut);
+		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_1,
 				creq_lut >> 32);
 	} else {
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT + idx, creq_lut);
+		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT, creq_lut);
 	}
 }
 
 static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
 		struct dpu_hw_pipe_qos_cfg *cfg)
 {
-	u32 idx;
 	u32 qos_ctrl = 0;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
+	if (!ctx)
 		return;
 
 	if (cfg->vblank_en) {
@@ -634,23 +620,19 @@ static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
 	if (cfg->danger_safe_en)
 		qos_ctrl |= SSPP_QOS_CTRL_DANGER_SAFE_EN;
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL + idx, qos_ctrl);
+	DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL, qos_ctrl);
 }
 
 static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe,
 		struct dpu_hw_cdp_cfg *cfg)
 {
 	struct dpu_hw_sspp *ctx = pipe->sspp;
-	u32 idx;
 	u32 cdp_cntl = 0;
 	u32 cdp_cntl_offset = 0;
 
 	if (!ctx || !cfg)
 		return;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx))
-		return;
-
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO ||
 	    pipe->multirect_index == DPU_SSPP_RECT_0)
 		cdp_cntl_offset = SSPP_CDP_CNTL;
@@ -672,13 +654,11 @@ static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe,
 static void _setup_layer_ops(struct dpu_hw_sspp *c,
 		unsigned long features)
 {
-	if (test_bit(DPU_SSPP_SRC, &features)) {
-		c->ops.setup_format = dpu_hw_sspp_setup_format;
-		c->ops.setup_rects = dpu_hw_sspp_setup_rects;
-		c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress;
-		c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill;
-		c->ops.setup_pe = dpu_hw_sspp_setup_pe_config;
-	}
+	c->ops.setup_format = dpu_hw_sspp_setup_format;
+	c->ops.setup_rects = dpu_hw_sspp_setup_rects;
+	c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress;
+	c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill;
+	c->ops.setup_pe = dpu_hw_sspp_setup_pe_config;
 
 	if (test_bit(DPU_SSPP_QOS, &features)) {
 		c->ops.setup_danger_safe_lut =
@@ -728,8 +708,8 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms,
 	/* add register dump support */
 	dpu_debugfs_create_regset32("src_blk", 0400,
 			debugfs_root,
-			sblk->src_blk.base + cfg->base,
-			sblk->src_blk.len,
+			cfg->base,
+			cfg->len,
 			kms);
 
 	if (cfg->features & BIT(DPU_SSPP_SCALER_QSEED3) ||
-- 
GitLab


From be7057e0d40be7f2d938c809d93f5efa9b365246 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 29 Apr 2023 04:23:52 +0300
Subject: [PATCH 0602/1791] drm/msm/dpu: access QSEED registers directly

Stop using _sspp_subblk_offset() to get offset of the scaler_blk. Inline
this function and use ctx->cap->sblk->scaler_blk.base directly.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534746/
Link: https://lore.kernel.org/r/20230429012353.2569481-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 27 +++++++--------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index a93e311afa710..4f58cdcb20e2a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -149,11 +149,6 @@ static int _sspp_subblk_offset(struct dpu_hw_sspp *ctx,
 	sblk = ctx->cap->sblk;
 
 	switch (s_id) {
-	case DPU_SSPP_SCALER_QSEED2:
-	case DPU_SSPP_SCALER_QSEED3:
-	case DPU_SSPP_SCALER_RGB:
-		*idx = sblk->scaler_blk.base;
-		break;
 	case DPU_SSPP_CSC:
 	case DPU_SSPP_CSC_10BIT:
 		*idx = sblk->csc_blk.base;
@@ -195,22 +190,21 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe)
 static void _sspp_setup_opmode(struct dpu_hw_sspp *ctx,
 		u32 mask, u8 en)
 {
-	u32 idx;
+	const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk;
 	u32 opmode;
 
 	if (!test_bit(DPU_SSPP_SCALER_QSEED2, &ctx->cap->features) ||
-		_sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED2, &idx) ||
 		!test_bit(DPU_SSPP_CSC, &ctx->cap->features))
 		return;
 
-	opmode = DPU_REG_READ(&ctx->hw, SSPP_VIG_OP_MODE + idx);
+	opmode = DPU_REG_READ(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE);
 
 	if (en)
 		opmode |= mask;
 	else
 		opmode &= ~mask;
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_VIG_OP_MODE + idx, opmode);
+	DPU_REG_WRITE(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE, opmode);
 }
 
 static void _sspp_setup_csc10_opmode(struct dpu_hw_sspp *ctx,
@@ -416,25 +410,22 @@ static void _dpu_hw_sspp_setup_scaler3(struct dpu_hw_sspp *ctx,
 		struct dpu_hw_scaler3_cfg *scaler3_cfg,
 		const struct dpu_format *format)
 {
-	u32 idx;
-
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED3, &idx)
-		|| !scaler3_cfg)
+	if (!ctx || !scaler3_cfg)
 		return;
 
-	dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg, idx,
+	dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg,
+			ctx->cap->sblk->scaler_blk.base,
 			ctx->cap->sblk->scaler_blk.version,
 			format);
 }
 
 static u32 _dpu_hw_sspp_get_scaler3_ver(struct dpu_hw_sspp *ctx)
 {
-	u32 idx;
-
-	if (!ctx || _sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED3, &idx))
+	if (!ctx)
 		return 0;
 
-	return dpu_hw_get_scaler3_ver(&ctx->hw, idx);
+	return dpu_hw_get_scaler3_ver(&ctx->hw,
+				      ctx->cap->sblk->scaler_blk.base);
 }
 
 /*
-- 
GitLab


From babbd0a5f3c5239cc081fed870b7f010d41bacde Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 29 Apr 2023 04:23:53 +0300
Subject: [PATCH 0603/1791] drm/msm/dpu: access CSC/CSC10 registers directly

Stop using _sspp_subblk_offset() to get offset of the csc_blk. Inline
this function and use ctx->cap->sblk->csc_blk.base directly.

As this was the last user, drop _sspp_subblk_offset() too.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/534747/
Link: https://lore.kernel.org/r/20230429012353.2569481-4-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 43 +++++----------------
 1 file changed, 9 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 4f58cdcb20e2a..6b68ec5c7a5a2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -136,30 +136,6 @@
 #define TS_CLK			19200000
 
 
-static int _sspp_subblk_offset(struct dpu_hw_sspp *ctx,
-		int s_id,
-		u32 *idx)
-{
-	int rc = 0;
-	const struct dpu_sspp_sub_blks *sblk;
-
-	if (!ctx || !ctx->cap || !ctx->cap->sblk)
-		return -EINVAL;
-
-	sblk = ctx->cap->sblk;
-
-	switch (s_id) {
-	case DPU_SSPP_CSC:
-	case DPU_SSPP_CSC_10BIT:
-		*idx = sblk->csc_blk.base;
-		break;
-	default:
-		rc = -EINVAL;
-	}
-
-	return rc;
-}
-
 static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe)
 {
 	struct dpu_hw_sspp *ctx = pipe->sspp;
@@ -210,19 +186,16 @@ static void _sspp_setup_opmode(struct dpu_hw_sspp *ctx,
 static void _sspp_setup_csc10_opmode(struct dpu_hw_sspp *ctx,
 		u32 mask, u8 en)
 {
-	u32 idx;
+	const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk;
 	u32 opmode;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_CSC_10BIT, &idx))
-		return;
-
-	opmode = DPU_REG_READ(&ctx->hw, SSPP_VIG_CSC_10_OP_MODE + idx);
+	opmode = DPU_REG_READ(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE);
 	if (en)
 		opmode |= mask;
 	else
 		opmode &= ~mask;
 
-	DPU_REG_WRITE(&ctx->hw, SSPP_VIG_CSC_10_OP_MODE + idx, opmode);
+	DPU_REG_WRITE(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE, opmode);
 }
 
 /*
@@ -530,18 +503,20 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe,
 static void dpu_hw_sspp_setup_csc(struct dpu_hw_sspp *ctx,
 		const struct dpu_csc_cfg *data)
 {
-	u32 idx;
+	u32 offset;
 	bool csc10 = false;
 
-	if (_sspp_subblk_offset(ctx, DPU_SSPP_CSC, &idx) || !data)
+	if (!ctx || !data)
 		return;
 
+	offset = ctx->cap->sblk->csc_blk.base;
+
 	if (test_bit(DPU_SSPP_CSC_10BIT, &ctx->cap->features)) {
-		idx += CSC_10BIT_OFFSET;
+		offset += CSC_10BIT_OFFSET;
 		csc10 = true;
 	}
 
-	dpu_hw_csc_setup(&ctx->hw, idx, data, csc10);
+	dpu_hw_csc_setup(&ctx->hw, offset, data, csc10);
 }
 
 static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color)
-- 
GitLab


From a7129231edf329a00e92dbd2d741f6da728a4a06 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 19 May 2023 20:49:59 +0200
Subject: [PATCH 0604/1791] drm/msm/dpu: Set DPU_DATA_HCTL_EN for in
 INTF_SC7180_MASK

DPU5 and newer targets enable this unconditionally. Move it from the
SC7280 mask to the SC7180 one.

Fixes: 7e6ee55320f0 ("drm/msm/disp/dpu1: enable DATA_HCTL_EN for sc7280 target")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/538159/
Link: https://lore.kernel.org/r/20230508-topic-hctl_en-v2-1-e7bea9f1f5dd@linaro.org
[DB: removed BIT(DPU_INTF_DATA_COMPRESS), which is not yet merged]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 70600f8abe91f..1dee5ba2b3121 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -100,9 +100,12 @@
 #define INTF_SDM845_MASK (0)
 
 #define INTF_SC7180_MASK \
-	(BIT(DPU_INTF_INPUT_CTRL) | BIT(DPU_INTF_TE) | BIT(DPU_INTF_STATUS_SUPPORTED))
+	(BIT(DPU_INTF_INPUT_CTRL) | \
+	 BIT(DPU_INTF_TE) | \
+	 BIT(DPU_INTF_STATUS_SUPPORTED) | \
+	 BIT(DPU_DATA_HCTL_EN))
 
-#define INTF_SC7280_MASK INTF_SC7180_MASK | BIT(DPU_DATA_HCTL_EN)
+#define INTF_SC7280_MASK (INTF_SC7180_MASK)
 
 #define WB_SM8250_MASK (BIT(DPU_WB_LINE_MODE) | \
 			 BIT(DPU_WB_UBWC) | \
-- 
GitLab


From 50883f733c8ec10dd2d36466d3e53e3cb341bc10 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 02:40:22 +0300
Subject: [PATCH 0605/1791] drm/msm/dpu: replace IS_ERR_OR_NULL with IS_ERR
 during DSC init

Using IS_ERR_OR_NULL() together with PTR_ERR() is a typical mistake. If
the value is NULL, then the function will return 0 instead of a proper
return code. Moreover dpu_hw_dsc_init() can not return NULL.
Replace the IS_ERR_OR_NULL() call with IS_ERR().

This follows the commit 740828c73a36 ("drm/msm/dpu: fix error handling
in dpu_rm_init"), which removed IS_ERR_OR_NULL() from RM init code, but
then the commit f2803ee91a41 ("drm/msm/disp/dpu1: Add DSC support in
RM") added it for DSC init.

Suggested-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538203/
Link: https://lore.kernel.org/r/20230519234025.2864377-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index f0fc70422e565..dffd3dd0a8777 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -247,7 +247,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 		const struct dpu_dsc_cfg *dsc = &cat->dsc[i];
 
 		hw = dpu_hw_dsc_init(dsc, mmio);
-		if (IS_ERR_OR_NULL(hw)) {
+		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed dsc object creation: err %d\n", rc);
 			goto fail;
-- 
GitLab


From f89fdc862be4c390652116d22b5c20b0837cc8cd Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 02:40:23 +0300
Subject: [PATCH 0606/1791] drm/msm/dpu: remove futile checks from
 dpu_rm_init()

dpu_rm_init() contains checks for block->id values. These were logical
in the vendor driver, when one can not be sure which values were passed
from DT. In the upstream driver this is not necessary: the catalog is a
part of the driver, we control specified IDs.

Suggested-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538204/
Link: https://lore.kernel.org/r/20230519234025.2864377-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c | 34 --------------------------
 1 file changed, 34 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index dffd3dd0a8777..d5a06628885ee 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -122,10 +122,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 			continue;
 		}
 
-		if (lm->id < LM_0 || lm->id >= LM_MAX) {
-			DPU_ERROR("skip mixer %d with invalid id\n", lm->id);
-			continue;
-		}
 		hw = dpu_hw_lm_init(lm, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -139,10 +135,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_merge_3d *hw;
 		const struct dpu_merge_3d_cfg *merge_3d = &cat->merge_3d[i];
 
-		if (merge_3d->id < MERGE_3D_0 || merge_3d->id >= MERGE_3D_MAX) {
-			DPU_ERROR("skip merge_3d %d with invalid id\n", merge_3d->id);
-			continue;
-		}
 		hw = dpu_hw_merge_3d_init(merge_3d, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -157,10 +149,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_pingpong *hw;
 		const struct dpu_pingpong_cfg *pp = &cat->pingpong[i];
 
-		if (pp->id < PINGPONG_0 || pp->id >= PINGPONG_MAX) {
-			DPU_ERROR("skip pingpong %d with invalid id\n", pp->id);
-			continue;
-		}
 		hw = dpu_hw_pingpong_init(pp, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -177,10 +165,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_intf *hw;
 		const struct dpu_intf_cfg *intf = &cat->intf[i];
 
-		if (intf->id < INTF_0 || intf->id >= INTF_MAX) {
-			DPU_ERROR("skip intf %d with invalid id\n", intf->id);
-			continue;
-		}
 		hw = dpu_hw_intf_init(intf, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -194,11 +178,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_wb *hw;
 		const struct dpu_wb_cfg *wb = &cat->wb[i];
 
-		if (wb->id < WB_0 || wb->id >= WB_MAX) {
-			DPU_ERROR("skip intf %d with invalid id\n", wb->id);
-			continue;
-		}
-
 		hw = dpu_hw_wb_init(wb, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -212,10 +191,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_ctl *hw;
 		const struct dpu_ctl_cfg *ctl = &cat->ctl[i];
 
-		if (ctl->id < CTL_0 || ctl->id >= CTL_MAX) {
-			DPU_ERROR("skip ctl %d with invalid id\n", ctl->id);
-			continue;
-		}
 		hw = dpu_hw_ctl_init(ctl, mmio, cat->mixer_count, cat->mixer);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -229,10 +204,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_dspp *hw;
 		const struct dpu_dspp_cfg *dspp = &cat->dspp[i];
 
-		if (dspp->id < DSPP_0 || dspp->id >= DSPP_MAX) {
-			DPU_ERROR("skip dspp %d with invalid id\n", dspp->id);
-			continue;
-		}
 		hw = dpu_hw_dspp_init(dspp, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
@@ -259,11 +230,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_sspp *hw;
 		const struct dpu_sspp_cfg *sspp = &cat->sspp[i];
 
-		if (sspp->id < SSPP_NONE || sspp->id >= SSPP_MAX) {
-			DPU_ERROR("skip intf %d with invalid id\n", sspp->id);
-			continue;
-		}
-
 		hw = dpu_hw_sspp_init(sspp, mmio, cat->ubwc);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
-- 
GitLab


From 76c40f1463585ab0927a832be302ebae0178e5a0 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 02:40:24 +0300
Subject: [PATCH 0607/1791] drm/msm/dpu: use PINGPONG_NONE for LMs with no PP
 attached

On msm8998/sdm845 some LM blocks do not have corresponding PINGPONG
block. Currently the driver uses PINGPONG_MAX for such cases. Switch
that to use PINGPONG_NONE instead, which is more logical.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538205/
Link: https://lore.kernel.org/r/20230519234025.2864377-4-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h  | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h          | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h             | 3 ++-
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c                  | 2 +-
 5 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 37a52b4ef8d39..3c732a0360c70 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -104,9 +104,9 @@ static const struct dpu_lm_cfg msm8998_lm[] = {
 	LM_BLK("lm_2", LM_2, 0x46000, MIXER_MSM8998_MASK,
 		&msm8998_lm_sblk, PINGPONG_2, LM_5, 0),
 	LM_BLK("lm_3", LM_3, 0x47000, MIXER_MSM8998_MASK,
-		&msm8998_lm_sblk, PINGPONG_MAX, 0, 0),
+		&msm8998_lm_sblk, PINGPONG_NONE, 0, 0),
 	LM_BLK("lm_4", LM_4, 0x48000, MIXER_MSM8998_MASK,
-		&msm8998_lm_sblk, PINGPONG_MAX, 0, 0),
+		&msm8998_lm_sblk, PINGPONG_NONE, 0, 0),
 	LM_BLK("lm_5", LM_5, 0x49000, MIXER_MSM8998_MASK,
 		&msm8998_lm_sblk, PINGPONG_3, LM_2, 0),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index 0b727252671c0..36ea1af108949 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -102,9 +102,9 @@ static const struct dpu_lm_cfg sdm845_lm[] = {
 	LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_2, LM_5, 0),
 	LM_BLK("lm_3", LM_3, 0x0, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_MAX, 0, 0),
+		&sdm845_lm_sblk, PINGPONG_NONE, 0, 0),
 	LM_BLK("lm_4", LM_4, 0x0, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_MAX, 0, 0),
+		&sdm845_lm_sblk, PINGPONG_NONE, 0, 0),
 	LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_3, LM_2, 0),
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 1cb5699a5a477..677048cc3b7d7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -529,7 +529,7 @@ struct dpu_sspp_cfg {
  * @base               register offset of this block
  * @features           bit mask identifying sub-blocks/features
  * @sblk:              LM Sub-blocks information
- * @pingpong:          ID of connected PingPong, PINGPONG_MAX if unsupported
+ * @pingpong:          ID of connected PingPong, PINGPONG_NONE if unsupported
  * @lm_pair_mask:      Bitmask of LMs that can be controlled by same CTL
  */
 struct dpu_lm_cfg {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index 1913a19de7336..02a0f48aac94a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -191,7 +191,8 @@ enum dpu_dsc {
 };
 
 enum dpu_pingpong {
-	PINGPONG_0 = 1,
+	PINGPONG_NONE,
+	PINGPONG_0,
 	PINGPONG_1,
 	PINGPONG_2,
 	PINGPONG_3,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index d5a06628885ee..bf7b8e7c45b13 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -117,7 +117,7 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_mixer *hw;
 		const struct dpu_lm_cfg *lm = &cat->mixer[i];
 
-		if (lm->pingpong == PINGPONG_MAX) {
+		if (lm->pingpong == PINGPONG_NONE) {
 			DPU_DEBUG("skip mixer %d without pingpong\n", lm->id);
 			continue;
 		}
-- 
GitLab


From 6c93a21d92666b44787ee32d054d48751799aad0 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 02:40:25 +0300
Subject: [PATCH 0608/1791] drm/msm/dpu: move PINGPONG_NONE check to
 dpu_lm_init()

Move the check for lm->pingpong being not NONE from dpu_rm_init() to
dpu_lm_init(), following the change to dpu_hw_intf_init().

Suggested-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538206/
Link: https://lore.kernel.org/r/20230519234025.2864377-5-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c | 5 +++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c    | 5 -----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
index 214f207ddf8ad..d1c3bd8379ea9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c
@@ -161,6 +161,11 @@ struct dpu_hw_mixer *dpu_hw_lm_init(const struct dpu_lm_cfg *cfg,
 {
 	struct dpu_hw_mixer *c;
 
+	if (cfg->pingpong == PINGPONG_NONE) {
+		DPU_DEBUG("skip mixer %d without pingpong\n", cfg->id);
+		return NULL;
+	}
+
 	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return ERR_PTR(-ENOMEM);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index bf7b8e7c45b13..2ca3c666e22ff 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -117,11 +117,6 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_mixer *hw;
 		const struct dpu_lm_cfg *lm = &cat->mixer[i];
 
-		if (lm->pingpong == PINGPONG_NONE) {
-			DPU_DEBUG("skip mixer %d without pingpong\n", lm->id);
-			continue;
-		}
-
 		hw = dpu_hw_lm_init(lm, mmio);
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
-- 
GitLab


From 861601ff7f757f34f0584dc1f78ebb7a1f541e77 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Thu, 11 May 2023 19:54:46 +0200
Subject: [PATCH 0609/1791] drm/i915: constify pointers to hwmon_channel_info

Statically allocated array of pointers to hwmon_channel_info can be made
const for safety.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511175446.282041-1-krzysztof.kozlowski@linaro.org
---
 drivers/gpu/drm/i915/i915_hwmon.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c
index 8e7dccc8d3a0e..e99e8c97ef01e 100644
--- a/drivers/gpu/drm/i915/i915_hwmon.c
+++ b/drivers/gpu/drm/i915/i915_hwmon.c
@@ -267,7 +267,7 @@ static const struct attribute_group *hwm_groups[] = {
 	NULL
 };
 
-static const struct hwmon_channel_info *hwm_info[] = {
+static const struct hwmon_channel_info * const hwm_info[] = {
 	HWMON_CHANNEL_INFO(in, HWMON_I_INPUT),
 	HWMON_CHANNEL_INFO(power, HWMON_P_MAX | HWMON_P_RATED_MAX | HWMON_P_CRIT),
 	HWMON_CHANNEL_INFO(energy, HWMON_E_INPUT),
@@ -275,7 +275,7 @@ static const struct hwmon_channel_info *hwm_info[] = {
 	NULL
 };
 
-static const struct hwmon_channel_info *hwm_gt_info[] = {
+static const struct hwmon_channel_info * const hwm_gt_info[] = {
 	HWMON_CHANNEL_INFO(energy, HWMON_E_INPUT),
 	NULL
 };
-- 
GitLab


From dda445651475b5a2a2941adcf1420fe3e9521bcd Mon Sep 17 00:00:00 2001
From: Artur Weber <aweber.kernel@gmail.com>
Date: Fri, 19 May 2023 19:03:52 +0200
Subject: [PATCH 0610/1791] dt-bindings: display: panel: Add Samsung S6D7AA0
 LCD panel controller

Add bindings for the S6D7AA0 LCD panel controller, including the
S6D7AA0-LSL080AL02 panel used in the Samsung Galaxy Tab 3 8.0 family
of tablets, and the S6D7AA0-LSL080AL03 and S6D7AA0-LTL101AT01 panels
used in the Samsung Galaxy Tab A 8.0 and 9.7 2015.

Signed-off-by: Artur Weber <aweber.kernel@gmail.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519170354.29610-2-aweber.kernel@gmail.com
---
 .../display/panel/samsung,s6d7aa0.yaml        | 70 +++++++++++++++++++
 1 file changed, 70 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml

diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml b/Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml
new file mode 100644
index 0000000000000..45a236d2cc70a
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml
@@ -0,0 +1,70 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/samsung,s6d7aa0.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Samsung S6D7AA0 MIPI-DSI LCD panel controller
+
+maintainers:
+  - Artur Weber <aweber.kernel@gmail.com>
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    items:
+      - enum:
+          # 1280x800 LSL080AL02 panel
+          - samsung,lsl080al02
+          # 1024x768 LSL080AL03 panel
+          - samsung,lsl080al03
+          # 1024x768 LTL101AT01 panel
+          - samsung,ltl101at01
+      - const: samsung,s6d7aa0
+
+  reg: true
+
+  backlight:
+    description:
+      Backlight to use for the panel. If this property is set on panels
+      that have DSI-based backlight control (LSL080AL03 and LTL101AT01),
+      it overrides the DSI-based backlight.
+
+  reset-gpios:
+    description: Reset GPIO pin, usually GPIO_ACTIVE_LOW.
+
+  power-supply:
+    description:
+      Main power supply for the panel; the exact voltage differs between
+      panels, and is usually somewhere around 3.3-5v.
+
+  vmipi-supply:
+    description: VMIPI supply, usually 1.8v.
+
+required:
+  - compatible
+  - reg
+  - reset-gpios
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "samsung,lsl080al02", "samsung,s6d7aa0";
+            reg = <0>;
+            power-supply = <&display_3v3_supply>;
+            reset-gpios = <&gpf0 4 GPIO_ACTIVE_LOW>;
+            backlight = <&backlight>;
+        };
+    };
+
+...
-- 
GitLab


From 6810bb390282bb75801832f31d088236503fee89 Mon Sep 17 00:00:00 2001
From: Artur Weber <aweber.kernel@gmail.com>
Date: Fri, 19 May 2023 19:03:53 +0200
Subject: [PATCH 0611/1791] drm/panel: Add Samsung S6D7AA0 panel controller
 driver

Initial driver for S6D7AA0-controlled panels. Currently, the following
panels are supported:

 - S6D7AA0-LSL080AL02 (Samsung Galaxy Tab 3 8.0)
 - S6D7AA0-LSL080AL03 (Samsung Galaxy Tab A 8.0 2015)
 - S6D7AA0-LTL101AT01 (Samsung Galaxy Tab A 9.7 2015)

It should be possible to extend this driver to work with other panels
using this IC.

Tested-by: Nikita Travkin <nikita@trvn.ru> #ltl101at01
Signed-off-by: Artur Weber <aweber.kernel@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519170354.29610-3-aweber.kernel@gmail.com
---
 drivers/gpu/drm/panel/Kconfig                 |   7 +
 drivers/gpu/drm/panel/Makefile                |   1 +
 drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c | 585 ++++++++++++++++++
 3 files changed, 593 insertions(+)
 create mode 100644 drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c

diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index 2b9d6db7860ba..203c0ef0bbfdf 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -553,6 +553,13 @@ config DRM_PANEL_SAMSUNG_S6D27A1
 	  This panel can be found in Samsung Galaxy Ace 2
 	  GT-I8160 mobile phone.
 
+config DRM_PANEL_SAMSUNG_S6D7AA0
+	tristate "Samsung S6D7AA0 MIPI-DSI video mode panel controller"
+	depends on OF
+	depends on BACKLIGHT_CLASS_DEVICE
+	select DRM_MIPI_DSI
+	select VIDEOMODE_HELPERS
+
 config DRM_PANEL_SAMSUNG_S6E3HA2
 	tristate "Samsung S6E3HA2 DSI video mode panel"
 	depends on OF
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index ff169781e82d7..30cf553c8d1db 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -54,6 +54,7 @@ obj-$(CONFIG_DRM_PANEL_SAMSUNG_DB7430) += panel-samsung-db7430.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_LD9040) += panel-samsung-ld9040.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6D16D0) += panel-samsung-s6d16d0.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6D27A1) += panel-samsung-s6d27a1.o
+obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6D7AA0) += panel-samsung-s6d7aa0.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E3HA2) += panel-samsung-s6e3ha2.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63J0X03) += panel-samsung-s6e63j0x03.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63M0) += panel-samsung-s6e63m0.o
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
new file mode 100644
index 0000000000000..f532aa0184280
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
@@ -0,0 +1,585 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Samsung S6D7AA0 MIPI-DSI TFT LCD controller drm_panel driver.
+ *
+ * Copyright (C) 2022 Artur Weber <aweber.kernel@gmail.com>
+ */
+
+#include <linux/backlight.h>
+#include <linux/delay.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/regulator/consumer.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+
+#include <video/mipi_display.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+
+/* Manufacturer command set */
+#define MCS_BL_CTL		0xc3
+#define MCS_OTP_RELOAD		0xd0
+#define MCS_PASSWD1		0xf0
+#define MCS_PASSWD2		0xf1
+#define MCS_PASSWD3		0xfc
+
+struct s6d7aa0 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct gpio_desc *reset_gpio;
+	struct regulator_bulk_data supplies[2];
+	const struct s6d7aa0_panel_desc *desc;
+};
+
+struct s6d7aa0_panel_desc {
+	unsigned int panel_type;
+	int (*init_func)(struct s6d7aa0 *ctx);
+	int (*off_func)(struct s6d7aa0 *ctx);
+	const struct drm_display_mode drm_mode;
+	unsigned long mode_flags;
+	u32 bus_flags;
+	bool has_backlight;
+	bool use_passwd3;
+};
+
+enum s6d7aa0_panels {
+	S6D7AA0_PANEL_LSL080AL02,
+	S6D7AA0_PANEL_LSL080AL03,
+	S6D7AA0_PANEL_LTL101AT01,
+};
+
+static inline struct s6d7aa0 *panel_to_s6d7aa0(struct drm_panel *panel)
+{
+	return container_of(panel, struct s6d7aa0, panel);
+}
+
+static void s6d7aa0_reset(struct s6d7aa0 *ctx)
+{
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	msleep(50);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	msleep(50);
+}
+
+static int s6d7aa0_lock(struct s6d7aa0 *ctx, bool lock)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	int ret = 0;
+
+	if (lock) {
+		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD1, 0xa5, 0xa5);
+		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD2, 0xa5, 0xa5);
+		if (ctx->desc->use_passwd3)
+			mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD3, 0x5a, 0x5a);
+	} else {
+		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD1, 0x5a, 0x5a);
+		mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD2, 0x5a, 0x5a);
+		if (ctx->desc->use_passwd3)
+			mipi_dsi_dcs_write_seq(dsi, MCS_PASSWD3, 0xa5, 0xa5);
+	}
+
+	return ret;
+}
+
+static int s6d7aa0_on(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	ret = ctx->desc->init_func(ctx);
+	if (ret < 0) {
+		dev_err(dev, "Failed to initialize panel: %d\n", ret);
+		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+		return ret;
+	}
+
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display on: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int s6d7aa0_off(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	ret = ctx->desc->off_func(ctx);
+	if (ret < 0) {
+		dev_err(dev, "Panel-specific off function failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = mipi_dsi_dcs_set_display_off(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display off: %d\n", ret);
+		return ret;
+	}
+	msleep(64);
+
+	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to enter sleep mode: %d\n", ret);
+		return ret;
+	}
+	msleep(120);
+
+	return 0;
+}
+
+static int s6d7aa0_prepare(struct drm_panel *panel)
+{
+	struct s6d7aa0 *ctx = panel_to_s6d7aa0(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(ctx->supplies), ctx->supplies);
+	if (ret < 0) {
+		dev_err(dev, "Failed to enable regulators: %d\n", ret);
+		return ret;
+	}
+
+	s6d7aa0_reset(ctx);
+
+	ret = s6d7aa0_on(ctx);
+	if (ret < 0) {
+		dev_err(dev, "Failed to initialize panel: %d\n", ret);
+		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int s6d7aa0_disable(struct drm_panel *panel)
+{
+	struct s6d7aa0 *ctx = panel_to_s6d7aa0(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = s6d7aa0_off(ctx);
+	if (ret < 0)
+		dev_err(dev, "Failed to un-initialize panel: %d\n", ret);
+
+	return 0;
+}
+
+static int s6d7aa0_unprepare(struct drm_panel *panel)
+{
+	struct s6d7aa0 *ctx = panel_to_s6d7aa0(panel);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies);
+
+	return 0;
+}
+
+/* Backlight control code */
+
+static int s6d7aa0_bl_update_status(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness = backlight_get_brightness(bl);
+	int ret;
+
+	ret = mipi_dsi_dcs_set_display_brightness(dsi, brightness);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int s6d7aa0_bl_get_brightness(struct backlight_device *bl)
+{
+	struct mipi_dsi_device *dsi = bl_get_data(bl);
+	u16 brightness;
+	int ret;
+
+	ret = mipi_dsi_dcs_get_display_brightness(dsi, &brightness);
+	if (ret < 0)
+		return ret;
+
+	return brightness & 0xff;
+}
+
+static const struct backlight_ops s6d7aa0_bl_ops = {
+	.update_status = s6d7aa0_bl_update_status,
+	.get_brightness = s6d7aa0_bl_get_brightness,
+};
+
+static struct backlight_device *
+s6d7aa0_create_backlight(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	const struct backlight_properties props = {
+		.type = BACKLIGHT_RAW,
+		.brightness = 255,
+		.max_brightness = 255,
+	};
+
+	return devm_backlight_device_register(dev, dev_name(dev), dev, dsi,
+					      &s6d7aa0_bl_ops, &props);
+}
+
+/* Initialization code and structures for LSL080AL02 panel */
+
+static int s6d7aa0_lsl080al02_init(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	usleep_range(20000, 25000);
+
+	ret = s6d7aa0_lock(ctx, false);
+	if (ret < 0) {
+		dev_err(dev, "Failed to unlock registers: %d\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, MCS_OTP_RELOAD, 0x00, 0x10);
+	usleep_range(1000, 1500);
+
+	/* SEQ_B6_PARAM_8_R01 */
+	mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x10);
+
+	/* BL_CTL_ON */
+	mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x28);
+
+	usleep_range(5000, 6000);
+
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_ADDRESS_MODE, 0x04);
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
+		return ret;
+	}
+
+	msleep(120);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_SET_ADDRESS_MODE, 0x00);
+
+	ret = s6d7aa0_lock(ctx, true);
+	if (ret < 0) {
+		dev_err(dev, "Failed to lock registers: %d\n", ret);
+		return ret;
+	}
+
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display on: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int s6d7aa0_lsl080al02_off(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+
+	/* BL_CTL_OFF */
+	mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x20);
+
+	return 0;
+}
+
+static const struct drm_display_mode s6d7aa0_lsl080al02_mode = {
+	.clock = (800 + 16 + 4 + 140) * (1280 + 8 + 4 + 4) * 60 / 1000,
+	.hdisplay = 800,
+	.hsync_start = 800 + 16,
+	.hsync_end = 800 + 16 + 4,
+	.htotal = 800 + 16 + 4 + 140,
+	.vdisplay = 1280,
+	.vsync_start = 1280 + 8,
+	.vsync_end = 1280 + 8 + 4,
+	.vtotal = 1280 + 8 + 4 + 4,
+	.width_mm = 108,
+	.height_mm = 173,
+};
+
+static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al02_desc = {
+	.panel_type = S6D7AA0_PANEL_LSL080AL02,
+	.init_func = s6d7aa0_lsl080al02_init,
+	.off_func = s6d7aa0_lsl080al02_off,
+	.drm_mode = s6d7aa0_lsl080al02_mode,
+	.mode_flags = MIPI_DSI_MODE_VSYNC_FLUSH | MIPI_DSI_MODE_VIDEO_NO_HFP,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
+
+	.has_backlight = false,
+	.use_passwd3 = false,
+};
+
+/* Initialization code and structures for LSL080AL03 panel */
+
+static int s6d7aa0_lsl080al03_init(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct device *dev = &dsi->dev;
+	int ret;
+
+	usleep_range(20000, 25000);
+
+	ret = s6d7aa0_lock(ctx, false);
+	if (ret < 0) {
+		dev_err(dev, "Failed to unlock registers: %d\n", ret);
+		return ret;
+	}
+
+	if (ctx->desc->panel_type == S6D7AA0_PANEL_LSL080AL03) {
+		mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0xc7, 0x00, 0x29);
+		mipi_dsi_dcs_write_seq(dsi, 0xbc, 0x01, 0x4e, 0xa0);
+		mipi_dsi_dcs_write_seq(dsi, 0xfd, 0x16, 0x10, 0x11, 0x23,
+				       0x09);
+		mipi_dsi_dcs_write_seq(dsi, 0xfe, 0x00, 0x02, 0x03, 0x21,
+				       0x80, 0x78);
+	} else if (ctx->desc->panel_type == S6D7AA0_PANEL_LTL101AT01) {
+		mipi_dsi_dcs_write_seq(dsi, MCS_BL_CTL, 0x40, 0x00, 0x08);
+		mipi_dsi_dcs_write_seq(dsi, 0xbc, 0x01, 0x4e, 0x0b);
+		mipi_dsi_dcs_write_seq(dsi, 0xfd, 0x16, 0x10, 0x11, 0x23,
+				       0x09);
+		mipi_dsi_dcs_write_seq(dsi, 0xfe, 0x00, 0x02, 0x03, 0x21,
+				       0x80, 0x68);
+	}
+
+	mipi_dsi_dcs_write_seq(dsi, 0xb3, 0x51);
+	mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x24);
+	mipi_dsi_dcs_write_seq(dsi, 0xf2, 0x02, 0x08, 0x08);
+
+	usleep_range(10000, 11000);
+
+	mipi_dsi_dcs_write_seq(dsi, 0xc0, 0x80, 0x80, 0x30);
+	mipi_dsi_dcs_write_seq(dsi, 0xcd,
+			       0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e, 0x2e,
+			       0x2e, 0x2e, 0x2e, 0x2e, 0x2e);
+	mipi_dsi_dcs_write_seq(dsi, 0xce,
+			       0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+			       0x00, 0x00, 0x00, 0x00, 0x00);
+	mipi_dsi_dcs_write_seq(dsi, 0xc1, 0x03);
+
+	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to exit sleep mode: %d\n", ret);
+		return ret;
+	}
+
+	ret = s6d7aa0_lock(ctx, true);
+	if (ret < 0) {
+		dev_err(dev, "Failed to lock registers: %d\n", ret);
+		return ret;
+	}
+
+	ret = mipi_dsi_dcs_set_display_on(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to set display on: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int s6d7aa0_lsl080al03_off(struct s6d7aa0 *ctx)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+
+	mipi_dsi_dcs_write_seq(dsi, 0x22, 0x00);
+
+	return 0;
+}
+
+static const struct drm_display_mode s6d7aa0_lsl080al03_mode = {
+	.clock = (768 + 18 + 16 + 126) * (1024 + 8 + 2 + 6) * 60 / 1000,
+	.hdisplay = 768,
+	.hsync_start = 768 + 18,
+	.hsync_end = 768 + 18 + 16,
+	.htotal = 768 + 18 + 16 + 126,
+	.vdisplay = 1024,
+	.vsync_start = 1024 + 8,
+	.vsync_end = 1024 + 8 + 2,
+	.vtotal = 1024 + 8 + 2 + 6,
+	.width_mm = 122,
+	.height_mm = 163,
+};
+
+static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al03_desc = {
+	.panel_type = S6D7AA0_PANEL_LSL080AL03,
+	.init_func = s6d7aa0_lsl080al03_init,
+	.off_func = s6d7aa0_lsl080al03_off,
+	.drm_mode = s6d7aa0_lsl080al03_mode,
+	.mode_flags = MIPI_DSI_MODE_NO_EOT_PACKET,
+	.bus_flags = 0,
+
+	.has_backlight = true,
+	.use_passwd3 = true,
+};
+
+/* Initialization structures for LTL101AT01 panel */
+
+static const struct drm_display_mode s6d7aa0_ltl101at01_mode = {
+	.clock = (768 + 96 + 16 + 184) * (1024 + 8 + 2 + 6) * 60 / 1000,
+	.hdisplay = 768,
+	.hsync_start = 768 + 96,
+	.hsync_end = 768 + 96 + 16,
+	.htotal = 768 + 96 + 16 + 184,
+	.vdisplay = 1024,
+	.vsync_start = 1024 + 8,
+	.vsync_end = 1024 + 8 + 2,
+	.vtotal = 1024 + 8 + 2 + 6,
+	.width_mm = 148,
+	.height_mm = 197,
+};
+
+static const struct s6d7aa0_panel_desc s6d7aa0_ltl101at01_desc = {
+	.panel_type = S6D7AA0_PANEL_LTL101AT01,
+	.init_func = s6d7aa0_lsl080al03_init, /* Similar init to LSL080AL03 */
+	.off_func = s6d7aa0_lsl080al03_off,
+	.drm_mode = s6d7aa0_ltl101at01_mode,
+	.mode_flags = MIPI_DSI_MODE_NO_EOT_PACKET,
+	.bus_flags = 0,
+
+	.has_backlight = true,
+	.use_passwd3 = true,
+};
+
+static int s6d7aa0_get_modes(struct drm_panel *panel,
+					struct drm_connector *connector)
+{
+	struct drm_display_mode *mode;
+	struct s6d7aa0 *ctx;
+
+	ctx = container_of(panel, struct s6d7aa0, panel);
+	if (!ctx)
+		return -EINVAL;
+
+	mode = drm_mode_duplicate(connector->dev, &ctx->desc->drm_mode);
+	if (!mode)
+		return -ENOMEM;
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+	connector->display_info.width_mm = mode->width_mm;
+	connector->display_info.height_mm = mode->height_mm;
+	connector->display_info.bus_flags = ctx->desc->bus_flags;
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs s6d7aa0_panel_funcs = {
+	.disable = s6d7aa0_disable,
+	.prepare = s6d7aa0_prepare,
+	.unprepare = s6d7aa0_unprepare,
+	.get_modes = s6d7aa0_get_modes,
+};
+
+static int s6d7aa0_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct s6d7aa0 *ctx;
+	int ret;
+
+	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+
+	ctx->desc = of_device_get_match_data(dev);
+	if (!ctx->desc)
+		return -ENODEV;
+
+	ctx->supplies[0].supply = "power";
+	ctx->supplies[1].supply = "vmipi";
+	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(ctx->supplies),
+					      ctx->supplies);
+	if (ret < 0)
+		return dev_err_probe(dev, ret, "Failed to get regulators\n");
+
+	ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(ctx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio),
+				     "Failed to get reset-gpios\n");
+
+	ctx->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST
+		| ctx->desc->mode_flags;
+
+	drm_panel_init(&ctx->panel, dev, &s6d7aa0_panel_funcs,
+		       DRM_MODE_CONNECTOR_DSI);
+	ctx->panel.prepare_prev_first = true;
+
+	ret = drm_panel_of_backlight(&ctx->panel);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get backlight\n");
+
+	/* Use DSI-based backlight as fallback if available */
+	if (ctx->desc->has_backlight && !ctx->panel.backlight) {
+		ctx->panel.backlight = s6d7aa0_create_backlight(dsi);
+		if (IS_ERR(ctx->panel.backlight))
+			return dev_err_probe(dev, PTR_ERR(ctx->panel.backlight),
+					     "Failed to create backlight\n");
+	}
+
+	drm_panel_add(&ctx->panel);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret < 0) {
+		dev_err(dev, "Failed to attach to DSI host: %d\n", ret);
+		drm_panel_remove(&ctx->panel);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void s6d7aa0_remove(struct mipi_dsi_device *dsi)
+{
+	struct s6d7aa0 *ctx = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	ret = mipi_dsi_detach(dsi);
+	if (ret < 0)
+		dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret);
+
+	drm_panel_remove(&ctx->panel);
+}
+
+static const struct of_device_id s6d7aa0_of_match[] = {
+	{
+		.compatible = "samsung,lsl080al02",
+		.data = &s6d7aa0_lsl080al02_desc
+	},
+	{
+		.compatible = "samsung,lsl080al03",
+		.data = &s6d7aa0_lsl080al03_desc
+	},
+	{
+		.compatible = "samsung,ltl101at01",
+		.data = &s6d7aa0_ltl101at01_desc
+	},
+	{ /* sentinel */ }
+};
+
+static struct mipi_dsi_driver s6d7aa0_driver = {
+	.probe = s6d7aa0_probe,
+	.remove = s6d7aa0_remove,
+	.driver = {
+		.name = "panel-samsung-s6d7aa0",
+		.of_match_table = s6d7aa0_of_match,
+	},
+};
+module_mipi_dsi_driver(s6d7aa0_driver);
+
+MODULE_AUTHOR("Artur Weber <aweber.kernel@gmail.com>");
+MODULE_DESCRIPTION("Samsung S6D7AA0 MIPI-DSI LCD controller driver");
+MODULE_LICENSE("GPL");
-- 
GitLab


From 0dd53308f74fcb16aa4e5cb90739b831c4a558de Mon Sep 17 00:00:00 2001
From: Artur Weber <aweber.kernel@gmail.com>
Date: Fri, 19 May 2023 19:03:54 +0200
Subject: [PATCH 0612/1791] MAINTAINERS: Add entry for Samsung S6D7AA0 LCD
 panel controller driver

Add myself as maintainer of the Samsung S6D7AA0 panel driver.

Signed-off-by: Artur Weber <aweber.kernel@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519170354.29610-4-aweber.kernel@gmail.com
---
 MAINTAINERS | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 402e26d0cdbcb..7cc2bfa4af6f0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6663,6 +6663,12 @@ S:	Maintained
 F:	Documentation/devicetree/bindings/display/panel/samsung,s6d27a1.yaml
 F:	drivers/gpu/drm/panel/panel-samsung-s6d27a1.c
 
+DRM DRIVER FOR SAMSUNG S6D7AA0 PANELS
+M:	Artur Weber <aweber.kernel@gmail.com>
+S:	Maintained
+F:	Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml
+F:	drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
+
 DRM DRIVER FOR SITRONIX ST7703 PANELS
 M:	Guido Günther <agx@sigxcpu.org>
 R:	Purism Kernel Team <kernel@puri.sm>
-- 
GitLab


From 937859485aefed1d9df72feb6ea74a84ff5cfa46 Mon Sep 17 00:00:00 2001
From: Arun R Murthy <arun.r.murthy@intel.com>
Date: Tue, 6 Sep 2022 09:48:06 +0530
Subject: [PATCH 0613/1791] drm/i915: Support Async Flip on Linear buffers

Starting from Gen12 Async Flip is supported on linear buffers.
This patch enables support for async on linear buffer.

UseCase: In Hybrid graphics, for hardware unsupported pixel formats it
will be converted to linear memory and then composed.

v2: Added use case
v3: Added FIXME for ICL indicating the restrictions

Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220906041806.4095575-1-arun.r.murthy@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 09320e14d75cd..0490c6412ab5e 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -5995,6 +5995,20 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in
 		 * this selectively if required.
 		 */
 		switch (new_plane_state->hw.fb->modifier) {
+		case DRM_FORMAT_MOD_LINEAR:
+			/*
+			 * FIXME: Async on Linear buffer is supported on ICL as
+			 * but with additional alignment and fbc restrictions
+			 * need to be taken care of. These aren't applicable for
+			 * gen12+.
+			 */
+			if (DISPLAY_VER(i915) < 12) {
+				drm_dbg_kms(&i915->drm,
+					    "[PLANE:%d:%s] Modifier does not support async flips\n",
+					    plane->base.base.id, plane->base.name);
+				return -EINVAL;
+			}
+
 		case I915_FORMAT_MOD_X_TILED:
 		case I915_FORMAT_MOD_Y_TILED:
 		case I915_FORMAT_MOD_Yf_TILED:
-- 
GitLab


From ec7743c9c48a14aa884d27f2feee1da30810ce0a Mon Sep 17 00:00:00 2001
From: Fabio Estevam <festevam@denx.de>
Date: Sun, 14 May 2023 08:46:24 -0300
Subject: [PATCH 0614/1791] dt-bindings: samsung,mipi-dsim: Add
 'lane-polarities'

The Samsung DSIM IP block allows the inversion of the clock and
data lanes.

Add an optional property called 'lane-polarities' that describes the
polarities of the MIPI DSI clock and data lanes.

This property is useful for properly describing the hardware when the
board designer decided to switch the polarities of the MIPI DSI
clock and/or data lanes.

Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230514114625.98372-1-festevam@gmail.com
---
 .../display/bridge/samsung,mipi-dsim.yaml     | 23 +++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index e841659e20cda..9f61ebdfefa85 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -105,6 +105,29 @@ properties:
           DSI output port node to the panel or the next bridge
           in the chain.
 
+        properties:
+          endpoint:
+            $ref: /schemas/media/video-interfaces.yaml#
+            unevaluatedProperties: false
+
+            properties:
+              data-lanes:
+                minItems: 1
+                maxItems: 4
+                uniqueItems: true
+                items:
+                  enum: [ 1, 2, 3, 4 ]
+
+              lane-polarities:
+                minItems: 1
+                maxItems: 5
+                description:
+                  The Samsung MIPI DSI IP requires that all the data lanes have
+                  the same polarity.
+
+            dependencies:
+              lane-polarities: [data-lanes]
+
 required:
   - clock-names
   - clocks
-- 
GitLab


From 74629c49e66cc6d36c46ac4e3f059780873ceedf Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Sun, 14 May 2023 08:46:25 -0300
Subject: [PATCH 0615/1791] drm: bridge: samsung-dsim: Implement support for
 clock/data polarity swap

Implement support for DSI clock and data lane DN/DP polarity swap by
means of decoding 'lane-polarities' DT property. The controller does
support DN/DP swap of clock lane and all data lanes, the controller
does not support polarity swap of individual data lane bundles, add
a check which verifies all data lanes have the same polarity.

This has been validated on an imx8mm board that actually has the MIPI DSI
clock lanes inverted.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230514114625.98372-2-festevam@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 27 ++++++++++++++++++++++++++-
 include/drm/bridge/samsung-dsim.h     |  2 ++
 2 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 10dc3315e69eb..0f3f6846beeaa 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -183,6 +183,8 @@
 #define DSIM_AFC_CTL(x)			(((x) & 0x7) << 5)
 
 /* DSIM_PLLCTRL */
+#define DSIM_PLL_DPDNSWAP_CLK		(1 << 25)
+#define DSIM_PLL_DPDNSWAP_DAT		(1 << 24)
 #define DSIM_FREQ_BAND(x)		((x) << 24)
 #define DSIM_PLL_EN			BIT(23)
 #define DSIM_PLL_P(x, offset)		((x) << (offset))
@@ -622,6 +624,11 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 		reg |= DSIM_FREQ_BAND(band);
 	}
 
+	if (dsi->swap_dn_dp_clk)
+		reg |= DSIM_PLL_DPDNSWAP_CLK;
+	if (dsi->swap_dn_dp_data)
+		reg |= DSIM_PLL_DPDNSWAP_DAT;
+
 	samsung_dsim_write(dsi, DSIM_PLLCTRL_REG, reg);
 
 	timeout = 1000;
@@ -1696,7 +1703,9 @@ static int samsung_dsim_parse_dt(struct samsung_dsim *dsi)
 {
 	struct device *dev = dsi->dev;
 	struct device_node *node = dev->of_node;
-	int ret;
+	u32 lane_polarities[5] = { 0 };
+	struct device_node *endpoint;
+	int i, nr_lanes, ret;
 
 	ret = samsung_dsim_of_read_u32(node, "samsung,pll-clock-frequency",
 				       &dsi->pll_clk_rate);
@@ -1713,6 +1722,22 @@ static int samsung_dsim_parse_dt(struct samsung_dsim *dsi)
 	if (ret < 0)
 		return ret;
 
+	endpoint = of_graph_get_endpoint_by_regs(node, 1, -1);
+	nr_lanes = of_property_count_u32_elems(endpoint, "data-lanes");
+	if (nr_lanes > 0 && nr_lanes <= 4) {
+		/* Polarity 0 is clock lane, 1..4 are data lanes. */
+		of_property_read_u32_array(endpoint, "lane-polarities",
+					   lane_polarities, nr_lanes + 1);
+		for (i = 1; i <= nr_lanes; i++) {
+			if (lane_polarities[1] != lane_polarities[i])
+				DRM_DEV_ERROR(dsi->dev, "Data lanes polarities do not match");
+		}
+		if (lane_polarities[0])
+			dsi->swap_dn_dp_clk = true;
+		if (lane_polarities[1])
+			dsi->swap_dn_dp_data = true;
+	}
+
 	return 0;
 }
 
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index ba5484de2b30e..6a37d1e079bf9 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -95,6 +95,8 @@ struct samsung_dsim {
 	u32 mode_flags;
 	u32 format;
 
+	bool swap_dn_dp_clk;
+	bool swap_dn_dp_data;
 	int state;
 	struct drm_property *brightness;
 	struct completion completed;
-- 
GitLab


From 504e72ed3a1b1c0d4450712a42ae6070d3a05a8e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 17 Apr 2023 23:03:23 +0200
Subject: [PATCH 0616/1791] drm/nouveau: dispnv50: fix missing-prototypes
 warning

nv50_display_create() is declared in another header, along with
a couple of declarations that are now outdated:

drivers/gpu/drm/nouveau/dispnv50/disp.c:2517:1: error: no previous prototype for 'nv50_display_create'

Fixes: ba801ef068c1 ("drm/nouveau/kms: display destroy/init/fini hooks can be static")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230417210329.2469722-1-arnd@kernel.org
---
 drivers/gpu/drm/nouveau/dispnv50/disp.c | 1 +
 drivers/gpu/drm/nouveau/nv50_display.h  | 4 +---
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 5bb777ff13130..9b6824f6b9e4b 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -64,6 +64,7 @@
 #include "nouveau_connector.h"
 #include "nouveau_encoder.h"
 #include "nouveau_fence.h"
+#include "nv50_display.h"
 
 #include <subdev/bios/dp.h>
 
diff --git a/drivers/gpu/drm/nouveau/nv50_display.h b/drivers/gpu/drm/nouveau/nv50_display.h
index fbd3b15583bc8..60f77766766e9 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.h
+++ b/drivers/gpu/drm/nouveau/nv50_display.h
@@ -31,7 +31,5 @@
 #include "nouveau_reg.h"
 
 int  nv50_display_create(struct drm_device *);
-void nv50_display_destroy(struct drm_device *);
-int  nv50_display_init(struct drm_device *);
-void nv50_display_fini(struct drm_device *);
+
 #endif /* __NV50_DISPLAY_H__ */
-- 
GitLab


From 79802863a01999bb90c790f8fbc80c5c2f9c8fea Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Thu, 11 May 2023 19:54:43 +0200
Subject: [PATCH 0617/1791] drm/nouveau: constify pointers to
 hwmon_channel_info

Statically allocated array of pointers to hwmon_channel_info can be made
const for safety.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230511175443.281970-1-krzysztof.kozlowski@linaro.org
---
 drivers/gpu/drm/nouveau/nouveau_hwmon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_hwmon.c b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
index e844be49e11ee..db30a4c2cd4d0 100644
--- a/drivers/gpu/drm/nouveau/nouveau_hwmon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
@@ -211,7 +211,7 @@ static const struct attribute_group temp1_auto_point_sensor_group = {
 
 #define N_ATTR_GROUPS   3
 
-static const struct hwmon_channel_info *nouveau_info[] = {
+static const struct hwmon_channel_info * const nouveau_info[] = {
 	HWMON_CHANNEL_INFO(chip,
 			   HWMON_C_UPDATE_INTERVAL),
 	HWMON_CHANNEL_INFO(temp,
-- 
GitLab


From cf0c4bc9e4e132b1992548ca8db30ec328b45403 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Wed, 17 May 2023 09:31:12 -0400
Subject: [PATCH 0618/1791] drm/nouveau/acr/ga102: set variable ga102_gsps
 storage-class-specifier to static

smatch reports
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c:49:1: warning: symbol
  'ga102_gsps' was not declared. Should it be static?

This variable is only used in its defining file, so it should be static

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517133112.873856-1-trix@redhat.com
---
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
index 525267412c3e1..a3996ceca9955 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ga102.c
@@ -45,7 +45,7 @@ ga102_gsp_nofw(struct nvkm_gsp *gsp, int ver, const struct nvkm_gsp_fwif *fwif)
 	return 0;
 }
 
-struct nvkm_gsp_fwif
+static struct nvkm_gsp_fwif
 ga102_gsps[] = {
 	{ -1, ga102_gsp_nofw, &ga102_gsp },
 	{}
-- 
GitLab


From a644fde77ff73ee54970c0fc5a64cf7624c8b5b1 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:40 -0700
Subject: [PATCH 0619/1791] drm/i915/pmu: Change bitmask of enabled events to
 u32

Having it as u64 was a confusing (but harmless) mistake.

Also add some asserts to make sure the internal field does not overflow
in the future.

v2: Fix WARN_ON firing for INTERRUPT event (Umesh)

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-2-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 26 ++++++++++++++++++--------
 1 file changed, 18 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 7ece883a7d956..96543dce2db10 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -50,7 +50,7 @@ static u8 engine_event_instance(struct perf_event *event)
 	return (event->attr.config >> I915_PMU_SAMPLE_BITS) & 0xff;
 }
 
-static bool is_engine_config(u64 config)
+static bool is_engine_config(const u64 config)
 {
 	return config < __I915_PMU_OTHER(0);
 }
@@ -88,9 +88,20 @@ static unsigned int config_bit(const u64 config)
 		return other_bit(config);
 }
 
-static u64 config_mask(u64 config)
+static u32 config_mask(const u64 config)
 {
-	return BIT_ULL(config_bit(config));
+	unsigned int bit = config_bit(config);
+
+	if (__builtin_constant_p(config))
+		BUILD_BUG_ON(bit >
+			     BITS_PER_TYPE(typeof_member(struct i915_pmu,
+							 enable)) - 1);
+	else
+		WARN_ON_ONCE(bit >
+			     BITS_PER_TYPE(typeof_member(struct i915_pmu,
+							 enable)) - 1);
+
+	return BIT(config_bit(config));
 }
 
 static bool is_engine_event(struct perf_event *event)
@@ -633,11 +644,10 @@ static void i915_pmu_enable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
 		container_of(event->pmu, typeof(*i915), pmu.base);
+	const unsigned int bit = event_bit(event);
 	struct i915_pmu *pmu = &i915->pmu;
 	unsigned long flags;
-	unsigned int bit;
 
-	bit = event_bit(event);
 	if (bit == -1)
 		goto update;
 
@@ -651,7 +661,7 @@ static void i915_pmu_enable(struct perf_event *event)
 	GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
 	GEM_BUG_ON(pmu->enable_count[bit] == ~0);
 
-	pmu->enable |= BIT_ULL(bit);
+	pmu->enable |= BIT(bit);
 	pmu->enable_count[bit]++;
 
 	/*
@@ -698,7 +708,7 @@ static void i915_pmu_disable(struct perf_event *event)
 {
 	struct drm_i915_private *i915 =
 		container_of(event->pmu, typeof(*i915), pmu.base);
-	unsigned int bit = event_bit(event);
+	const unsigned int bit = event_bit(event);
 	struct i915_pmu *pmu = &i915->pmu;
 	unsigned long flags;
 
@@ -734,7 +744,7 @@ static void i915_pmu_disable(struct perf_event *event)
 	 * bitmask when the last listener on an event goes away.
 	 */
 	if (--pmu->enable_count[bit] == 0) {
-		pmu->enable &= ~BIT_ULL(bit);
+		pmu->enable &= ~BIT(bit);
 		pmu->timer_enabled &= pmu_needs_timer(pmu, true);
 	}
 
-- 
GitLab


From e367d3c45158ba34bb684227d87c52d8d840fd67 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:41 -0700
Subject: [PATCH 0620/1791] drm/i915/pmu: Support PMU for all engines

Given how the metrics are already exported, we also need to run sampling
over engines from all GTs.

Problem of GT frequencies is left for later.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-3-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 96543dce2db10..9edf87ee5d106 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -10,6 +10,7 @@
 #include "gt/intel_engine_pm.h"
 #include "gt/intel_engine_regs.h"
 #include "gt/intel_engine_user.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_pm.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_rc6.h"
@@ -425,8 +426,9 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer)
 	struct drm_i915_private *i915 =
 		container_of(hrtimer, struct drm_i915_private, pmu.timer);
 	struct i915_pmu *pmu = &i915->pmu;
-	struct intel_gt *gt = to_gt(i915);
 	unsigned int period_ns;
+	struct intel_gt *gt;
+	unsigned int i;
 	ktime_t now;
 
 	if (!READ_ONCE(pmu->timer_enabled))
@@ -442,8 +444,13 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer)
 	 * grabbing the forcewake. However the potential error from timer call-
 	 * back delay greatly dominates this so we keep it simple.
 	 */
-	engines_sample(gt, period_ns);
-	frequency_sample(gt, period_ns);
+
+	for_each_gt(gt, i915, i) {
+		engines_sample(gt, period_ns);
+
+		if (i == 0) /* FIXME */
+			frequency_sample(gt, period_ns);
+	}
 
 	hrtimer_forward(hrtimer, now, ns_to_ktime(PERIOD));
 
-- 
GitLab


From 08322dabb5cbce75e210d8df4774fc078ed7161c Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:42 -0700
Subject: [PATCH 0621/1791] drm/i915/pmu: Skip sampling engines with no enabled
 counters

As we have more and more engines do not waste time sampling the ones no-
one is monitoring.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-4-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 9edf87ee5d106..6d594f67f365c 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -350,6 +350,9 @@ engines_sample(struct intel_gt *gt, unsigned int period_ns)
 		return;
 
 	for_each_engine(engine, gt, id) {
+		if (!engine->pmu.enable)
+			continue;
+
 		if (!intel_engine_pm_get_if_awake(engine))
 			continue;
 
-- 
GitLab


From da5d51672874936c08810e63d6dfc670263d5e15 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:43 -0700
Subject: [PATCH 0622/1791] drm/i915/pmu: Transform PMU parking code to be GT
 based

Trivial prep work for full multi-tile enablement later.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-5-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/gt/intel_gt_pm.c |  4 ++--
 drivers/gpu/drm/i915/i915_pmu.c       | 16 ++++++++--------
 drivers/gpu/drm/i915/i915_pmu.h       |  9 +++++----
 3 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
index e02cb90723ae0..c2e69bafd02b0 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
@@ -87,7 +87,7 @@ static int __gt_unpark(struct intel_wakeref *wf)
 
 	intel_rc6_unpark(&gt->rc6);
 	intel_rps_unpark(&gt->rps);
-	i915_pmu_gt_unparked(i915);
+	i915_pmu_gt_unparked(gt);
 	intel_guc_busyness_unpark(gt);
 
 	intel_gt_unpark_requests(gt);
@@ -109,7 +109,7 @@ static int __gt_park(struct intel_wakeref *wf)
 
 	intel_guc_busyness_park(gt);
 	i915_vma_parked(gt);
-	i915_pmu_gt_parked(i915);
+	i915_pmu_gt_parked(gt);
 	intel_rps_park(&gt->rps);
 	intel_rc6_park(&gt->rc6);
 
diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 6d594f67f365c..890693fdaf9eb 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -228,11 +228,11 @@ static void init_rc6(struct i915_pmu *pmu)
 	}
 }
 
-static void park_rc6(struct drm_i915_private *i915)
+static void park_rc6(struct intel_gt *gt)
 {
-	struct i915_pmu *pmu = &i915->pmu;
+	struct i915_pmu *pmu = &gt->i915->pmu;
 
-	pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(to_gt(i915));
+	pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(gt);
 	pmu->sleep_last = ktime_get_raw();
 }
 
@@ -247,16 +247,16 @@ static void __i915_pmu_maybe_start_timer(struct i915_pmu *pmu)
 	}
 }
 
-void i915_pmu_gt_parked(struct drm_i915_private *i915)
+void i915_pmu_gt_parked(struct intel_gt *gt)
 {
-	struct i915_pmu *pmu = &i915->pmu;
+	struct i915_pmu *pmu = &gt->i915->pmu;
 
 	if (!pmu->base.event_init)
 		return;
 
 	spin_lock_irq(&pmu->lock);
 
-	park_rc6(i915);
+	park_rc6(gt);
 
 	/*
 	 * Signal sampling timer to stop if only engine events are enabled and
@@ -267,9 +267,9 @@ void i915_pmu_gt_parked(struct drm_i915_private *i915)
 	spin_unlock_irq(&pmu->lock);
 }
 
-void i915_pmu_gt_unparked(struct drm_i915_private *i915)
+void i915_pmu_gt_unparked(struct intel_gt *gt)
 {
-	struct i915_pmu *pmu = &i915->pmu;
+	struct i915_pmu *pmu = &gt->i915->pmu;
 
 	if (!pmu->base.event_init)
 		return;
diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index c30f43319a783..a686fd7ccedf1 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -13,6 +13,7 @@
 #include <uapi/drm/i915_drm.h>
 
 struct drm_i915_private;
+struct intel_gt;
 
 /*
  * Non-engine events that we need to track enabled-disabled transition and
@@ -151,15 +152,15 @@ int i915_pmu_init(void);
 void i915_pmu_exit(void);
 void i915_pmu_register(struct drm_i915_private *i915);
 void i915_pmu_unregister(struct drm_i915_private *i915);
-void i915_pmu_gt_parked(struct drm_i915_private *i915);
-void i915_pmu_gt_unparked(struct drm_i915_private *i915);
+void i915_pmu_gt_parked(struct intel_gt *gt);
+void i915_pmu_gt_unparked(struct intel_gt *gt);
 #else
 static inline int i915_pmu_init(void) { return 0; }
 static inline void i915_pmu_exit(void) {}
 static inline void i915_pmu_register(struct drm_i915_private *i915) {}
 static inline void i915_pmu_unregister(struct drm_i915_private *i915) {}
-static inline void i915_pmu_gt_parked(struct drm_i915_private *i915) {}
-static inline void i915_pmu_gt_unparked(struct drm_i915_private *i915) {}
+static inline void i915_pmu_gt_parked(struct intel_gt *gt) {}
+static inline void i915_pmu_gt_unparked(struct intel_gt *gt) {}
 #endif
 
 #endif
-- 
GitLab


From b319cc594e666061b22331da654606b92730343a Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:44 -0700
Subject: [PATCH 0623/1791] drm/i915/pmu: Add reference counting to the
 sampling timer

We do not want to have timers per tile and waste CPU cycles and energy via
multiple wake-up sources, for a relatively un-important task of PMU
sampling, so keeping a single timer works well. But we also do not want
the first GT which goes idle to turn off the timer.

Add some reference counting, via a mask of unparked GTs, to solve this.

v2: Drop the check for unparked in i915_sample (Ashutosh)
v3: Revert v2 (Tvrtko)

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-6-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++--
 drivers/gpu/drm/i915/i915_pmu.h |  4 ++++
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 890693fdaf9eb..ecb57a94143eb 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -262,7 +262,9 @@ void i915_pmu_gt_parked(struct intel_gt *gt)
 	 * Signal sampling timer to stop if only engine events are enabled and
 	 * GPU went idle.
 	 */
-	pmu->timer_enabled = pmu_needs_timer(pmu, false);
+	pmu->unparked &= ~BIT(gt->info.id);
+	if (pmu->unparked == 0)
+		pmu->timer_enabled = pmu_needs_timer(pmu, false);
 
 	spin_unlock_irq(&pmu->lock);
 }
@@ -279,7 +281,10 @@ void i915_pmu_gt_unparked(struct intel_gt *gt)
 	/*
 	 * Re-enable sampling timer when GPU goes active.
 	 */
-	__i915_pmu_maybe_start_timer(pmu);
+	if (pmu->unparked == 0)
+		__i915_pmu_maybe_start_timer(pmu);
+
+	pmu->unparked |= BIT(gt->info.id);
 
 	spin_unlock_irq(&pmu->lock);
 }
@@ -449,6 +454,9 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer)
 	 */
 
 	for_each_gt(gt, i915, i) {
+		if (!(pmu->unparked & BIT(i)))
+			continue;
+
 		engines_sample(gt, period_ns);
 
 		if (i == 0) /* FIXME */
diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index a686fd7ccedf1..3a811266ac6a0 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -76,6 +76,10 @@ struct i915_pmu {
 	 * @lock: Lock protecting enable mask and ref count handling.
 	 */
 	spinlock_t lock;
+	/**
+	 * @unparked: GT unparked mask.
+	 */
+	unsigned int unparked;
 	/**
 	 * @timer: Timer for internal i915 PMU sampling.
 	 */
-- 
GitLab


From bc4be0a38b63b6d4d00a58b10e164f56049be2c2 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:45 -0700
Subject: [PATCH 0624/1791] drm/i915/pmu: Prepare for multi-tile non-engine
 counters

Reserve some bits in the counter config namespace which will carry the
tile id and prepare the code to handle this.

No per tile counters have been added yet.

v2:
- Fix checkpatch issues
- Use 4 bits for gt id in non-engine counters. Drop FIXME.
- Set MAX GTs to 4. Drop FIXME.

v3: (Ashutosh, Tvrtko)
- Drop BUG_ON that would never fire
- Make enable u64
- Pull in some code from next patch

v4: Set I915_PMU_MAX_GTS to 2 (Tvrtko)

v5: s/u64/u32 where needed (Ashutosh)

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-7-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 146 +++++++++++++++++++++++---------
 drivers/gpu/drm/i915/i915_pmu.h |   9 +-
 include/uapi/drm/i915_drm.h     |  17 +++-
 3 files changed, 127 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index ecb57a94143eb..5cfc322e69b4a 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -56,11 +56,21 @@ static bool is_engine_config(const u64 config)
 	return config < __I915_PMU_OTHER(0);
 }
 
+static unsigned int config_gt_id(const u64 config)
+{
+	return config >> __I915_PMU_GT_SHIFT;
+}
+
+static u64 config_counter(const u64 config)
+{
+	return config & ~(~0ULL << __I915_PMU_GT_SHIFT);
+}
+
 static unsigned int other_bit(const u64 config)
 {
 	unsigned int val;
 
-	switch (config) {
+	switch (config_counter(config)) {
 	case I915_PMU_ACTUAL_FREQUENCY:
 		val =  __I915_PMU_ACTUAL_FREQUENCY_ENABLED;
 		break;
@@ -78,7 +88,9 @@ static unsigned int other_bit(const u64 config)
 		return -1;
 	}
 
-	return I915_ENGINE_SAMPLE_COUNT + val;
+	return I915_ENGINE_SAMPLE_COUNT +
+	       config_gt_id(config) * __I915_PMU_TRACKED_EVENT_COUNT +
+	       val;
 }
 
 static unsigned int config_bit(const u64 config)
@@ -115,6 +127,18 @@ static unsigned int event_bit(struct perf_event *event)
 	return config_bit(event->attr.config);
 }
 
+static u32 frequency_enabled_mask(void)
+{
+	unsigned int i;
+	u32 mask = 0;
+
+	for (i = 0; i < I915_PMU_MAX_GTS; i++)
+		mask |= config_mask(__I915_PMU_ACTUAL_FREQUENCY(i)) |
+			config_mask(__I915_PMU_REQUESTED_FREQUENCY(i));
+
+	return mask;
+}
+
 static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active)
 {
 	struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu);
@@ -131,9 +155,7 @@ static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active)
 	 * Mask out all the ones which do not need the timer, or in
 	 * other words keep all the ones that could need the timer.
 	 */
-	enable &= config_mask(I915_PMU_ACTUAL_FREQUENCY) |
-		  config_mask(I915_PMU_REQUESTED_FREQUENCY) |
-		  ENGINE_SAMPLE_MASK;
+	enable &= frequency_enabled_mask() | ENGINE_SAMPLE_MASK;
 
 	/*
 	 * When the GPU is idle per-engine counters do not need to be
@@ -175,9 +197,37 @@ static inline s64 ktime_since_raw(const ktime_t kt)
 	return ktime_to_ns(ktime_sub(ktime_get_raw(), kt));
 }
 
+static unsigned int
+__sample_idx(struct i915_pmu *pmu, unsigned int gt_id, int sample)
+{
+	unsigned int idx = gt_id * __I915_NUM_PMU_SAMPLERS + sample;
+
+	GEM_BUG_ON(idx >= ARRAY_SIZE(pmu->sample));
+
+	return idx;
+}
+
+static u64 read_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample)
+{
+	return pmu->sample[__sample_idx(pmu, gt_id, sample)].cur;
+}
+
+static void
+store_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample, u64 val)
+{
+	pmu->sample[__sample_idx(pmu, gt_id, sample)].cur = val;
+}
+
+static void
+add_sample_mult(struct i915_pmu *pmu, unsigned int gt_id, int sample, u32 val, u32 mul)
+{
+	pmu->sample[__sample_idx(pmu, gt_id, sample)].cur += mul_u32_u32(val, mul);
+}
+
 static u64 get_rc6(struct intel_gt *gt)
 {
 	struct drm_i915_private *i915 = gt->i915;
+	const unsigned int gt_id = gt->info.id;
 	struct i915_pmu *pmu = &i915->pmu;
 	unsigned long flags;
 	bool awake = false;
@@ -192,7 +242,7 @@ static u64 get_rc6(struct intel_gt *gt)
 	spin_lock_irqsave(&pmu->lock, flags);
 
 	if (awake) {
-		pmu->sample[__I915_SAMPLE_RC6].cur = val;
+		store_sample(pmu, gt_id, __I915_SAMPLE_RC6, val);
 	} else {
 		/*
 		 * We think we are runtime suspended.
@@ -201,14 +251,14 @@ static u64 get_rc6(struct intel_gt *gt)
 		 * on top of the last known real value, as the approximated RC6
 		 * counter value.
 		 */
-		val = ktime_since_raw(pmu->sleep_last);
-		val += pmu->sample[__I915_SAMPLE_RC6].cur;
+		val = ktime_since_raw(pmu->sleep_last[gt_id]);
+		val += read_sample(pmu, gt_id, __I915_SAMPLE_RC6);
 	}
 
-	if (val < pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur)
-		val = pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur;
+	if (val < read_sample(pmu, gt_id, __I915_SAMPLE_RC6_LAST_REPORTED))
+		val = read_sample(pmu, gt_id, __I915_SAMPLE_RC6_LAST_REPORTED);
 	else
-		pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = val;
+		store_sample(pmu, gt_id, __I915_SAMPLE_RC6_LAST_REPORTED, val);
 
 	spin_unlock_irqrestore(&pmu->lock, flags);
 
@@ -218,13 +268,20 @@ static u64 get_rc6(struct intel_gt *gt)
 static void init_rc6(struct i915_pmu *pmu)
 {
 	struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu);
-	intel_wakeref_t wakeref;
+	struct intel_gt *gt;
+	unsigned int i;
+
+	for_each_gt(gt, i915, i) {
+		intel_wakeref_t wakeref;
+
+		with_intel_runtime_pm(gt->uncore->rpm, wakeref) {
+			u64 val = __get_rc6(gt);
 
-	with_intel_runtime_pm(to_gt(i915)->uncore->rpm, wakeref) {
-		pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(to_gt(i915));
-		pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur =
-					pmu->sample[__I915_SAMPLE_RC6].cur;
-		pmu->sleep_last = ktime_get_raw();
+			store_sample(pmu, i, __I915_SAMPLE_RC6, val);
+			store_sample(pmu, i, __I915_SAMPLE_RC6_LAST_REPORTED,
+				     val);
+			pmu->sleep_last[i] = ktime_get_raw();
+		}
 	}
 }
 
@@ -232,8 +289,8 @@ static void park_rc6(struct intel_gt *gt)
 {
 	struct i915_pmu *pmu = &gt->i915->pmu;
 
-	pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(gt);
-	pmu->sleep_last = ktime_get_raw();
+	store_sample(pmu, gt->info.id, __I915_SAMPLE_RC6, __get_rc6(gt));
+	pmu->sleep_last[gt->info.id] = ktime_get_raw();
 }
 
 static void __i915_pmu_maybe_start_timer(struct i915_pmu *pmu)
@@ -373,34 +430,30 @@ engines_sample(struct intel_gt *gt, unsigned int period_ns)
 	}
 }
 
-static void
-add_sample_mult(struct i915_pmu_sample *sample, u32 val, u32 mul)
-{
-	sample->cur += mul_u32_u32(val, mul);
-}
-
-static bool frequency_sampling_enabled(struct i915_pmu *pmu)
+static bool
+frequency_sampling_enabled(struct i915_pmu *pmu, unsigned int gt)
 {
 	return pmu->enable &
-	       (config_mask(I915_PMU_ACTUAL_FREQUENCY) |
-		config_mask(I915_PMU_REQUESTED_FREQUENCY));
+	       (config_mask(__I915_PMU_ACTUAL_FREQUENCY(gt)) |
+		config_mask(__I915_PMU_REQUESTED_FREQUENCY(gt)));
 }
 
 static void
 frequency_sample(struct intel_gt *gt, unsigned int period_ns)
 {
 	struct drm_i915_private *i915 = gt->i915;
+	const unsigned int gt_id = gt->info.id;
 	struct i915_pmu *pmu = &i915->pmu;
 	struct intel_rps *rps = &gt->rps;
 
-	if (!frequency_sampling_enabled(pmu))
+	if (!frequency_sampling_enabled(pmu, gt_id))
 		return;
 
 	/* Report 0/0 (actual/requested) frequency while parked. */
 	if (!intel_gt_pm_get_if_awake(gt))
 		return;
 
-	if (pmu->enable & config_mask(I915_PMU_ACTUAL_FREQUENCY)) {
+	if (pmu->enable & config_mask(__I915_PMU_ACTUAL_FREQUENCY(gt_id))) {
 		u32 val;
 
 		/*
@@ -416,12 +469,12 @@ frequency_sample(struct intel_gt *gt, unsigned int period_ns)
 		if (!val)
 			val = intel_gpu_freq(rps, rps->cur_freq);
 
-		add_sample_mult(&pmu->sample[__I915_SAMPLE_FREQ_ACT],
+		add_sample_mult(pmu, gt_id, __I915_SAMPLE_FREQ_ACT,
 				val, period_ns / 1000);
 	}
 
-	if (pmu->enable & config_mask(I915_PMU_REQUESTED_FREQUENCY)) {
-		add_sample_mult(&pmu->sample[__I915_SAMPLE_FREQ_REQ],
+	if (pmu->enable & config_mask(__I915_PMU_REQUESTED_FREQUENCY(gt_id))) {
+		add_sample_mult(pmu, gt_id, __I915_SAMPLE_FREQ_REQ,
 				intel_rps_get_requested_frequency(rps),
 				period_ns / 1000);
 	}
@@ -458,9 +511,7 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer)
 			continue;
 
 		engines_sample(gt, period_ns);
-
-		if (i == 0) /* FIXME */
-			frequency_sample(gt, period_ns);
+		frequency_sample(gt, period_ns);
 	}
 
 	hrtimer_forward(hrtimer, now, ns_to_ktime(PERIOD));
@@ -502,7 +553,13 @@ config_status(struct drm_i915_private *i915, u64 config)
 {
 	struct intel_gt *gt = to_gt(i915);
 
-	switch (config) {
+	unsigned int gt_id = config_gt_id(config);
+	unsigned int max_gt_id = HAS_EXTRA_GT_LIST(i915) ? 1 : 0;
+
+	if (gt_id > max_gt_id)
+		return -ENOENT;
+
+	switch (config_counter(config)) {
 	case I915_PMU_ACTUAL_FREQUENCY:
 		if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915))
 			/* Requires a mutex for sampling! */
@@ -513,6 +570,8 @@ config_status(struct drm_i915_private *i915, u64 config)
 			return -ENODEV;
 		break;
 	case I915_PMU_INTERRUPTS:
+		if (gt_id)
+			return -ENOENT;
 		break;
 	case I915_PMU_RC6_RESIDENCY:
 		if (!gt->rc6.supported)
@@ -610,22 +669,27 @@ static u64 __i915_pmu_event_read(struct perf_event *event)
 			val = engine->pmu.sample[sample].cur;
 		}
 	} else {
-		switch (event->attr.config) {
+		const unsigned int gt_id = config_gt_id(event->attr.config);
+		const u64 config = config_counter(event->attr.config);
+
+		switch (config) {
 		case I915_PMU_ACTUAL_FREQUENCY:
 			val =
-			   div_u64(pmu->sample[__I915_SAMPLE_FREQ_ACT].cur,
+			   div_u64(read_sample(pmu, gt_id,
+					       __I915_SAMPLE_FREQ_ACT),
 				   USEC_PER_SEC /* to MHz */);
 			break;
 		case I915_PMU_REQUESTED_FREQUENCY:
 			val =
-			   div_u64(pmu->sample[__I915_SAMPLE_FREQ_REQ].cur,
+			   div_u64(read_sample(pmu, gt_id,
+					       __I915_SAMPLE_FREQ_REQ),
 				   USEC_PER_SEC /* to MHz */);
 			break;
 		case I915_PMU_INTERRUPTS:
 			val = READ_ONCE(pmu->irq_count);
 			break;
 		case I915_PMU_RC6_RESIDENCY:
-			val = get_rc6(to_gt(i915));
+			val = get_rc6(i915->gt[gt_id]);
 			break;
 		case I915_PMU_SOFTWARE_GT_AWAKE_TIME:
 			val = ktime_to_ns(intel_gt_get_awake_time(to_gt(i915)));
diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index 3a811266ac6a0..33d80fbaab8bc 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -38,13 +38,16 @@ enum {
 	__I915_NUM_PMU_SAMPLERS
 };
 
+#define I915_PMU_MAX_GTS 2
+
 /*
  * How many different events we track in the global PMU mask.
  *
  * It is also used to know to needed number of event reference counters.
  */
 #define I915_PMU_MASK_BITS \
-	(I915_ENGINE_SAMPLE_COUNT + __I915_PMU_TRACKED_EVENT_COUNT)
+	(I915_ENGINE_SAMPLE_COUNT + \
+	 I915_PMU_MAX_GTS * __I915_PMU_TRACKED_EVENT_COUNT)
 
 #define I915_ENGINE_SAMPLE_COUNT (I915_SAMPLE_SEMA + 1)
 
@@ -124,11 +127,11 @@ struct i915_pmu {
 	 * Only global counters are held here, while the per-engine ones are in
 	 * struct intel_engine_cs.
 	 */
-	struct i915_pmu_sample sample[__I915_NUM_PMU_SAMPLERS];
+	struct i915_pmu_sample sample[I915_PMU_MAX_GTS * __I915_NUM_PMU_SAMPLERS];
 	/**
 	 * @sleep_last: Last time GT parked for RC6 estimation.
 	 */
-	ktime_t sleep_last;
+	ktime_t sleep_last[I915_PMU_MAX_GTS];
 	/**
 	 * @irq_count: Number of interrupts
 	 *
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index ba40855dbc931..f31dfacde6014 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -280,7 +280,16 @@ enum drm_i915_pmu_engine_sample {
 #define I915_PMU_ENGINE_SEMA(class, instance) \
 	__I915_PMU_ENGINE(class, instance, I915_SAMPLE_SEMA)
 
-#define __I915_PMU_OTHER(x) (__I915_PMU_ENGINE(0xff, 0xff, 0xf) + 1 + (x))
+/*
+ * Top 4 bits of every non-engine counter are GT id.
+ */
+#define __I915_PMU_GT_SHIFT (60)
+
+#define ___I915_PMU_OTHER(gt, x) \
+	(((__u64)__I915_PMU_ENGINE(0xff, 0xff, 0xf) + 1 + (x)) | \
+	((__u64)(gt) << __I915_PMU_GT_SHIFT))
+
+#define __I915_PMU_OTHER(x) ___I915_PMU_OTHER(0, x)
 
 #define I915_PMU_ACTUAL_FREQUENCY	__I915_PMU_OTHER(0)
 #define I915_PMU_REQUESTED_FREQUENCY	__I915_PMU_OTHER(1)
@@ -290,6 +299,12 @@ enum drm_i915_pmu_engine_sample {
 
 #define I915_PMU_LAST /* Deprecated - do not use */ I915_PMU_RC6_RESIDENCY
 
+#define __I915_PMU_ACTUAL_FREQUENCY(gt)		___I915_PMU_OTHER(gt, 0)
+#define __I915_PMU_REQUESTED_FREQUENCY(gt)	___I915_PMU_OTHER(gt, 1)
+#define __I915_PMU_INTERRUPTS(gt)		___I915_PMU_OTHER(gt, 2)
+#define __I915_PMU_RC6_RESIDENCY(gt)		___I915_PMU_OTHER(gt, 3)
+#define __I915_PMU_SOFTWARE_GT_AWAKE_TIME(gt)	___I915_PMU_OTHER(gt, 4)
+
 /* Each region is a minimum of 16k, and there are at most 255 of them.
  */
 #define I915_NR_TEX_REGIONS 255	/* table size 2k - maximum due to use
-- 
GitLab


From 906bd0fb137ffc361b3ce0d0db07f288db5582ea Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri, 19 May 2023 08:49:46 -0700
Subject: [PATCH 0625/1791] drm/i915/pmu: Export counters from all tiles

Start exporting frequency and RC6 counters from all tiles.

Existing counters keep their names and config values and new one use the
namespace added in the previous patch, with the "-gtN" added to their
names.

Interrupts counter is an odd one off. Because it is the global device
counters (not only GT) we choose not to add per tile versions for now.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Aravind Iddamsetty <aravind.iddamsetty@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519154946.3751971-8-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 82 ++++++++++++++++++++++-----------
 1 file changed, 55 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 5cfc322e69b4a..a814583e19fd7 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -940,11 +940,20 @@ static const struct attribute_group i915_pmu_cpumask_attr_group = {
 	.attrs = i915_cpumask_attrs,
 };
 
-#define __event(__config, __name, __unit) \
+#define __event(__counter, __name, __unit) \
 { \
-	.config = (__config), \
+	.counter = (__counter), \
 	.name = (__name), \
 	.unit = (__unit), \
+	.global = false, \
+}
+
+#define __global_event(__counter, __name, __unit) \
+{ \
+	.counter = (__counter), \
+	.name = (__name), \
+	.unit = (__unit), \
+	.global = true, \
 }
 
 #define __engine_event(__sample, __name) \
@@ -983,15 +992,16 @@ create_event_attributes(struct i915_pmu *pmu)
 {
 	struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu);
 	static const struct {
-		u64 config;
+		unsigned int counter;
 		const char *name;
 		const char *unit;
+		bool global;
 	} events[] = {
-		__event(I915_PMU_ACTUAL_FREQUENCY, "actual-frequency", "M"),
-		__event(I915_PMU_REQUESTED_FREQUENCY, "requested-frequency", "M"),
-		__event(I915_PMU_INTERRUPTS, "interrupts", NULL),
-		__event(I915_PMU_RC6_RESIDENCY, "rc6-residency", "ns"),
-		__event(I915_PMU_SOFTWARE_GT_AWAKE_TIME, "software-gt-awake-time", "ns"),
+		__event(0, "actual-frequency", "M"),
+		__event(1, "requested-frequency", "M"),
+		__global_event(2, "interrupts", NULL),
+		__event(3, "rc6-residency", "ns"),
+		__event(4, "software-gt-awake-time", "ns"),
 	};
 	static const struct {
 		enum drm_i915_pmu_engine_sample sample;
@@ -1006,12 +1016,17 @@ create_event_attributes(struct i915_pmu *pmu)
 	struct i915_ext_attribute *i915_attr = NULL, *i915_iter;
 	struct attribute **attr = NULL, **attr_iter;
 	struct intel_engine_cs *engine;
-	unsigned int i;
+	struct intel_gt *gt;
+	unsigned int i, j;
 
 	/* Count how many counters we will be exposing. */
-	for (i = 0; i < ARRAY_SIZE(events); i++) {
-		if (!config_status(i915, events[i].config))
-			count++;
+	for_each_gt(gt, i915, j) {
+		for (i = 0; i < ARRAY_SIZE(events); i++) {
+			u64 config = ___I915_PMU_OTHER(j, events[i].counter);
+
+			if (!config_status(i915, config))
+				count++;
+		}
 	}
 
 	for_each_uabi_engine(engine, i915) {
@@ -1041,26 +1056,39 @@ create_event_attributes(struct i915_pmu *pmu)
 	attr_iter = attr;
 
 	/* Initialize supported non-engine counters. */
-	for (i = 0; i < ARRAY_SIZE(events); i++) {
-		char *str;
-
-		if (config_status(i915, events[i].config))
-			continue;
-
-		str = kstrdup(events[i].name, GFP_KERNEL);
-		if (!str)
-			goto err;
+	for_each_gt(gt, i915, j) {
+		for (i = 0; i < ARRAY_SIZE(events); i++) {
+			u64 config = ___I915_PMU_OTHER(j, events[i].counter);
+			char *str;
 
-		*attr_iter++ = &i915_iter->attr.attr;
-		i915_iter = add_i915_attr(i915_iter, str, events[i].config);
+			if (config_status(i915, config))
+				continue;
 
-		if (events[i].unit) {
-			str = kasprintf(GFP_KERNEL, "%s.unit", events[i].name);
+			if (events[i].global || !HAS_EXTRA_GT_LIST(i915))
+				str = kstrdup(events[i].name, GFP_KERNEL);
+			else
+				str = kasprintf(GFP_KERNEL, "%s-gt%u",
+						events[i].name, j);
 			if (!str)
 				goto err;
 
-			*attr_iter++ = &pmu_iter->attr.attr;
-			pmu_iter = add_pmu_attr(pmu_iter, str, events[i].unit);
+			*attr_iter++ = &i915_iter->attr.attr;
+			i915_iter = add_i915_attr(i915_iter, str, config);
+
+			if (events[i].unit) {
+				if (events[i].global || !HAS_EXTRA_GT_LIST(i915))
+					str = kasprintf(GFP_KERNEL, "%s.unit",
+							events[i].name);
+				else
+					str = kasprintf(GFP_KERNEL, "%s-gt%u.unit",
+							events[i].name, j);
+				if (!str)
+					goto err;
+
+				*attr_iter++ = &pmu_iter->attr.attr;
+				pmu_iter = add_pmu_attr(pmu_iter, str,
+							events[i].unit);
+			}
 		}
 	}
 
-- 
GitLab


From 0fbcf57077c47b444e91b9ce8a243e6f7f53693d Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Thu, 18 May 2023 22:11:02 -0700
Subject: [PATCH 0626/1791] drm/i915/mtl: end support for set caching ioctl

The design is to keep Buffer Object's caching policy immutable through
out its life cycle. This patch ends the support for set caching ioctl
from MTL onward. While doing that we also set BO's to be 1-way coherent
at creation time because GPU is no longer automatically snooping CPU
cache. For userspace components needing to fine tune the caching policy
for BO's, a follow up patch will extend the GEM_CREATE uAPI to allow
them specify caching mode at BO creation time.

Signed-off-by: Fei Yang <fei.yang@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230519051103.3404990-2-fei.yang@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_domain.c | 3 +++
 drivers/gpu/drm/i915/gem/i915_gem_shmem.c  | 9 ++++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
index 05107a6efe450..dfaaa8b66ac30 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c
@@ -350,6 +350,9 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 	if (IS_DGFX(i915))
 		return -ENODEV;
 
+	if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70))
+		return -EOPNOTSUPP;
+
 	switch (args->caching) {
 	case I915_CACHING_NONE:
 		level = I915_CACHE_NONE;
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index 37d1efcd3ca66..cad4a6017f4b5 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -601,7 +601,14 @@ static int shmem_object_init(struct intel_memory_region *mem,
 	obj->write_domain = I915_GEM_DOMAIN_CPU;
 	obj->read_domains = I915_GEM_DOMAIN_CPU;
 
-	if (HAS_LLC(i915))
+	/*
+	 * MTL doesn't snoop CPU cache by default for GPU access (namely
+	 * 1-way coherency). However some UMD's are currently depending on
+	 * that. Make 1-way coherent the default setting for MTL. A follow
+	 * up patch will extend the GEM_CREATE uAPI to allow UMD's specify
+	 * caching mode at BO creation time
+	 */
+	if (HAS_LLC(i915) || (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)))
 		/* On some devices, we can have the GPU use the LLC (the CPU
 		 * cache) for about a 10% performance improvement
 		 * compared to uncached.  Graphics requests other than
-- 
GitLab


From aa25aacc3e3d1367d6674763a636a9fd5ffd12da Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 22 May 2023 21:06:37 +0200
Subject: [PATCH 0627/1791] drm/gma500: Clear fbdev framebuffer with
 fb_memset_io()

The fbdev framebuffer is I/O memory, so clear it with fb_memset_io().
Fixes the following sparse warning:

../drivers/gpu/drm/gma500/fbdev.c:234:20: warning: incorrect type in argument 1 (different address spaces)
../drivers/gpu/drm/gma500/fbdev.c:234:20:    expected void *s
../drivers/gpu/drm/gma500/fbdev.c:234:20:    got char [noderef] __iomem *screen_base

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230522190637.7039-1-tzimmermann@suse.de
---
 drivers/gpu/drm/gma500/fbdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/gma500/fbdev.c b/drivers/gpu/drm/gma500/fbdev.c
index 62287407e7173..4f0309548b2b5 100644
--- a/drivers/gpu/drm/gma500/fbdev.c
+++ b/drivers/gpu/drm/gma500/fbdev.c
@@ -231,7 +231,7 @@ static int psb_fbdev_fb_probe(struct drm_fb_helper *fb_helper,
 	info->fix.mmio_start = pci_resource_start(pdev, 0);
 	info->fix.mmio_len = pci_resource_len(pdev, 0);
 
-	memset(info->screen_base, 0, info->screen_size);
+	fb_memset_io(info->screen_base, 0, info->screen_size);
 
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
 
-- 
GitLab


From b0c536d88ed260d75eb3076690ca9630cafd1b90 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 22 May 2023 21:17:01 +0200
Subject: [PATCH 0628/1791] drm/msm: Use struct fb_info.screen_buffer

The fbdev framebuffer is in system memory. Store the address in
the field 'screen_buffer'. Fixes the following sparse warning.

../drivers/gpu/drm/msm/msm_fbdev.c:124:26: warning: incorrect type in assignment (different address spaces)
../drivers/gpu/drm/msm/msm_fbdev.c:124:26:    expected char [noderef] __iomem *screen_base
../drivers/gpu/drm/msm/msm_fbdev.c:124:26:    got void *

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230522191701.13406-1-tzimmermann@suse.de
---
 drivers/gpu/drm/msm/msm_fbdev.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index 2ebc86381e1c9..ce0ba6d1979a7 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -121,9 +121,9 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 
 	drm_fb_helper_fill_info(fbi, helper, sizes);
 
-	fbi->screen_base = msm_gem_get_vaddr(bo);
-	if (IS_ERR(fbi->screen_base)) {
-		ret = PTR_ERR(fbi->screen_base);
+	fbi->screen_buffer = msm_gem_get_vaddr(bo);
+	if (IS_ERR(fbi->screen_buffer)) {
+		ret = PTR_ERR(fbi->screen_buffer);
 		goto fail;
 	}
 	fbi->screen_size = bo->size;
-- 
GitLab


From 1464e48d69ab7a50a377c9d39f5e5eb3cee2722e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?N=C3=ADcolas=20F=2E=20R=2E=20A=2E=20Prado?=
 <nfraprado@collabora.com>
Date: Thu, 18 May 2023 15:39:02 -0400
Subject: [PATCH 0629/1791] drm/bridge: anx7625: Prevent endless probe loop
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During probe, the driver registers i2c dummy devices and populates the
aux bus, which registers a device for the panel. After doing that, the
driver can still defer probe if needed. This ordering of operations is
troublesome however, because the deferred probe work will retry probing
all pending devices every time a new device is registered. Therefore, if
modules need to be loaded in order to satisfy the dependencies for this
driver to complete probe, the kernel will stall, since it'll keep trying
to probe the anx7625 driver, but never succeed, given that modules would
only be loaded after the deferred probe work completes.

Two changes are required to avoid this issue:
* Move of_find_mipi_dsi_host_by_node(), which can defer probe, to before
  anx7625_register_i2c_dummy_clients() and
  devm_of_dp_aux_populate_ep_devices(), which register devices.
* Make use of the done_probing callback when populating the aux bus,
  so that the bridge registration is only done after the panel is
  probed. This is required because the panel might need to defer probe,
  but the aux bus population needs the i2c dummy devices working, so
  this call couldn't just be moved to an earlier point in probe.
  One caveat is that if the panel is described outside the aux bus, the
  probe loop issue can still happen, but we don't have a way to avoid
  it in that case since there's no callback available.

With this patch applied, it's possible to boot on
mt8192-asurada-spherion with

CONFIG_DRM_ANALOGIX_ANX7625=y
CONFIG_MTK_MMSYS=m
CONFIG_BACKLIGHT_PWM=y

and also with

CONFIG_DRM_ANALOGIX_ANX7625=y
CONFIG_MTK_MMSYS=y
CONFIG_BACKLIGHT_PWM=m

Fixes: adca62ec370c ("drm/bridge: anx7625: Support reading edid through aux channel")
Fixes: 269332997a16 ("drm/bridge: anx7625: Return -EPROBE_DEFER if the dsi host was not found")
Reported-by: "kernelci.org bot" <bot@kernelci.org>
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230518193902.891121-1-nfraprado@collabora.com
---
 drivers/gpu/drm/bridge/analogix/anx7625.c | 128 +++++++++++++++-------
 1 file changed, 88 insertions(+), 40 deletions(-)

diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c
index 093e7b9a570f0..c86531f7b0726 100644
--- a/drivers/gpu/drm/bridge/analogix/anx7625.c
+++ b/drivers/gpu/drm/bridge/analogix/anx7625.c
@@ -1686,6 +1686,14 @@ static int anx7625_parse_dt(struct device *dev,
 	if (of_property_read_bool(np, "analogix,audio-enable"))
 		pdata->audio_en = 1;
 
+	return 0;
+}
+
+static int anx7625_parse_dt_panel(struct device *dev,
+				  struct anx7625_platform_data *pdata)
+{
+	struct device_node *np = dev->of_node;
+
 	pdata->panel_bridge = devm_drm_of_get_bridge(dev, np, 1, 0);
 	if (IS_ERR(pdata->panel_bridge)) {
 		if (PTR_ERR(pdata->panel_bridge) == -ENODEV) {
@@ -2031,7 +2039,7 @@ static int anx7625_register_audio(struct device *dev, struct anx7625_data *ctx)
 	return 0;
 }
 
-static int anx7625_attach_dsi(struct anx7625_data *ctx)
+static int anx7625_setup_dsi_device(struct anx7625_data *ctx)
 {
 	struct mipi_dsi_device *dsi;
 	struct device *dev = &ctx->client->dev;
@@ -2041,9 +2049,6 @@ static int anx7625_attach_dsi(struct anx7625_data *ctx)
 		.channel = 0,
 		.node = NULL,
 	};
-	int ret;
-
-	DRM_DEV_DEBUG_DRIVER(dev, "attach dsi\n");
 
 	host = of_find_mipi_dsi_host_by_node(ctx->pdata.mipi_host_node);
 	if (!host) {
@@ -2064,14 +2069,24 @@ static int anx7625_attach_dsi(struct anx7625_data *ctx)
 		MIPI_DSI_MODE_VIDEO_HSE	|
 		MIPI_DSI_HS_PKT_END_ALIGNED;
 
-	ret = devm_mipi_dsi_attach(dev, dsi);
+	ctx->dsi = dsi;
+
+	return 0;
+}
+
+static int anx7625_attach_dsi(struct anx7625_data *ctx)
+{
+	struct device *dev = &ctx->client->dev;
+	int ret;
+
+	DRM_DEV_DEBUG_DRIVER(dev, "attach dsi\n");
+
+	ret = devm_mipi_dsi_attach(dev, ctx->dsi);
 	if (ret) {
 		DRM_DEV_ERROR(dev, "fail to attach dsi to host.\n");
 		return ret;
 	}
 
-	ctx->dsi = dsi;
-
 	DRM_DEV_DEBUG_DRIVER(dev, "attach dsi succeeded.\n");
 
 	return 0;
@@ -2559,6 +2574,40 @@ static void anx7625_runtime_disable(void *data)
 	pm_runtime_disable(data);
 }
 
+static int anx7625_link_bridge(struct drm_dp_aux *aux)
+{
+	struct anx7625_data *platform = container_of(aux, struct anx7625_data, aux);
+	struct device *dev = aux->dev;
+	int ret;
+
+	ret = anx7625_parse_dt_panel(dev, &platform->pdata);
+	if (ret) {
+		DRM_DEV_ERROR(dev, "fail to parse DT for panel : %d\n", ret);
+		return ret;
+	}
+
+	platform->bridge.funcs = &anx7625_bridge_funcs;
+	platform->bridge.of_node = dev->of_node;
+	if (!anx7625_of_panel_on_aux_bus(dev))
+		platform->bridge.ops |= DRM_BRIDGE_OP_EDID;
+	if (!platform->pdata.panel_bridge)
+		platform->bridge.ops |= DRM_BRIDGE_OP_HPD |
+					DRM_BRIDGE_OP_DETECT;
+	platform->bridge.type = platform->pdata.panel_bridge ?
+				    DRM_MODE_CONNECTOR_eDP :
+				    DRM_MODE_CONNECTOR_DisplayPort;
+
+	drm_bridge_add(&platform->bridge);
+
+	if (!platform->pdata.is_dpi) {
+		ret = anx7625_attach_dsi(platform);
+		if (ret)
+			drm_bridge_remove(&platform->bridge);
+	}
+
+	return ret;
+}
+
 static int anx7625_i2c_probe(struct i2c_client *client)
 {
 	struct anx7625_data *platform;
@@ -2633,6 +2682,24 @@ static int anx7625_i2c_probe(struct i2c_client *client)
 	platform->aux.wait_hpd_asserted = anx7625_wait_hpd_asserted;
 	drm_dp_aux_init(&platform->aux);
 
+	ret = anx7625_parse_dt(dev, pdata);
+	if (ret) {
+		if (ret != -EPROBE_DEFER)
+			DRM_DEV_ERROR(dev, "fail to parse DT : %d\n", ret);
+		goto free_wq;
+	}
+
+	if (!platform->pdata.is_dpi) {
+		ret = anx7625_setup_dsi_device(platform);
+		if (ret < 0)
+			goto free_wq;
+	}
+
+	/*
+	 * Registering the i2c devices will retrigger deferred probe, so it
+	 * needs to be done after calls that might return EPROBE_DEFER,
+	 * otherwise we can get an infinite loop.
+	 */
 	if (anx7625_register_i2c_dummy_clients(platform, client) != 0) {
 		ret = -ENOMEM;
 		DRM_DEV_ERROR(dev, "fail to reserve I2C bus.\n");
@@ -2647,13 +2714,21 @@ static int anx7625_i2c_probe(struct i2c_client *client)
 	if (ret)
 		goto free_wq;
 
-	devm_of_dp_aux_populate_ep_devices(&platform->aux);
-
-	ret = anx7625_parse_dt(dev, pdata);
+	/*
+	 * Populating the aux bus will retrigger deferred probe, so it needs to
+	 * be done after calls that might return EPROBE_DEFER, otherwise we can
+	 * get an infinite loop.
+	 */
+	ret = devm_of_dp_aux_populate_bus(&platform->aux, anx7625_link_bridge);
 	if (ret) {
-		if (ret != -EPROBE_DEFER)
-			DRM_DEV_ERROR(dev, "fail to parse DT : %d\n", ret);
-		goto free_wq;
+		if (ret != -ENODEV) {
+			DRM_DEV_ERROR(dev, "failed to populate aux bus : %d\n", ret);
+			goto free_wq;
+		}
+
+		ret = anx7625_link_bridge(&platform->aux);
+		if (ret)
+			goto free_wq;
 	}
 
 	if (!platform->pdata.low_power_mode) {
@@ -2666,27 +2741,6 @@ static int anx7625_i2c_probe(struct i2c_client *client)
 	if (platform->pdata.intp_irq)
 		queue_work(platform->workqueue, &platform->work);
 
-	platform->bridge.funcs = &anx7625_bridge_funcs;
-	platform->bridge.of_node = client->dev.of_node;
-	if (!anx7625_of_panel_on_aux_bus(&client->dev))
-		platform->bridge.ops |= DRM_BRIDGE_OP_EDID;
-	if (!platform->pdata.panel_bridge)
-		platform->bridge.ops |= DRM_BRIDGE_OP_HPD |
-					DRM_BRIDGE_OP_DETECT;
-	platform->bridge.type = platform->pdata.panel_bridge ?
-				    DRM_MODE_CONNECTOR_eDP :
-				    DRM_MODE_CONNECTOR_DisplayPort;
-
-	drm_bridge_add(&platform->bridge);
-
-	if (!platform->pdata.is_dpi) {
-		ret = anx7625_attach_dsi(platform);
-		if (ret) {
-			DRM_DEV_ERROR(dev, "Fail to attach to dsi : %d\n", ret);
-			goto unregister_bridge;
-		}
-	}
-
 	if (platform->pdata.audio_en)
 		anx7625_register_audio(dev, platform);
 
@@ -2694,12 +2748,6 @@ static int anx7625_i2c_probe(struct i2c_client *client)
 
 	return 0;
 
-unregister_bridge:
-	drm_bridge_remove(&platform->bridge);
-
-	if (!platform->pdata.low_power_mode)
-		pm_runtime_put_sync_suspend(&client->dev);
-
 free_wq:
 	if (platform->workqueue)
 		destroy_workqueue(platform->workqueue);
-- 
GitLab


From 37cee4876a45a5c3da79a83d34ed4f3c68548aef Mon Sep 17 00:00:00 2001
From: Azeem Shaikh <azeemshaikh38@gmail.com>
Date: Mon, 22 May 2023 15:52:10 +0000
Subject: [PATCH 0630/1791] drm/bridge: dw-hdmi: Replace all non-returning
 strlcpy with strscpy

strlcpy() reads the entire source buffer first.
This read may exceed the destination size limit.
This is both inefficient and can lead to linear read
overflows if a source string is not NUL-terminated [1].
In an effort to remove strlcpy() completely [2], replace
strlcpy() here with strscpy().
No return values were used, so direct replacement is safe.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

Signed-off-by: Azeem Shaikh <azeemshaikh38@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230522155210.2336690-1-azeemshaikh38@gmail.com
---
 drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
index 603bb3c51027b..9d6dcaf317a17 100644
--- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
+++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
@@ -533,7 +533,7 @@ static struct i2c_adapter *dw_hdmi_i2c_adapter(struct dw_hdmi *hdmi)
 	adap->owner = THIS_MODULE;
 	adap->dev.parent = hdmi->dev;
 	adap->algo = &dw_hdmi_algorithm;
-	strlcpy(adap->name, "DesignWare HDMI", sizeof(adap->name));
+	strscpy(adap->name, "DesignWare HDMI", sizeof(adap->name));
 	i2c_set_adapdata(adap, hdmi);
 
 	ret = i2c_add_adapter(adap);
-- 
GitLab


From 7a2280e8dcd2f1f436db9631287c0b21cf6a92b0 Mon Sep 17 00:00:00 2001
From: Nirmoy Das <nirmoy.das@intel.com>
Date: Mon, 22 May 2023 14:42:05 +0200
Subject: [PATCH 0631/1791] drm/i915: Wait for active retire before
 i915_active_fini()

i915_active_fini() finalizes the debug object, which can occur before
the active retires and deactivates the debug object. Wait for one
final time before calling i915_active_fini();

Closes:: https://gitlab.freedesktop.org/drm/intel/-/issues/8311
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230522124205.368-1-nirmoy.das@intel.com
---
 drivers/gpu/drm/i915/i915_vma.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 20a44788999e5..bad698bece272 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -1710,6 +1710,8 @@ static void release_references(struct i915_vma *vma, struct intel_gt *gt,
 	if (vm_ddestroy)
 		i915_vm_resv_put(vma->vm);
 
+	/* Wait for async active retire */
+	i915_active_wait(&vma->active);
 	i915_active_fini(&vma->active);
 	GEM_WARN_ON(vma->resource);
 	i915_vma_free(vma);
-- 
GitLab


From 6a038f0183dd5d3e289f6c1fe6962de9b31f8fd2 Mon Sep 17 00:00:00 2001
From: Artur Weber <aweber.kernel@gmail.com>
Date: Wed, 24 May 2023 10:43:24 +0200
Subject: [PATCH 0632/1791] drm/panel: samsung-s6d7aa0: use pointer for
 drm_mode in panel desc struct

Fixes compilation issues with older GCC versions and Clang after
changes introduced in commit 6810bb390282 ("drm/panel: Add Samsung
S6D7AA0 panel controller driver"). Tested with GCC 13.1.1, GCC 6.4.0
and Clang 16.0.3.

Fixes the following errors with Clang:

  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:312:14: error: initializer element is not a compile-time constant
          .drm_mode = s6d7aa0_lsl080al02_mode,
                      ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:415:14: error: initializer element is not a compile-time constant
          .drm_mode = s6d7aa0_lsl080al03_mode,
                      ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:443:14: error: initializer element is not a compile-time constant
          .drm_mode = s6d7aa0_ltl101at01_mode,
                      ^~~~~~~~~~~~~~~~~~~~~~~
  3 errors generated.

Fixes the following errors with GCC:

  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:312:14: error: initializer element is not constant
    .drm_mode = s6d7aa0_lsl080al02_mode,
                ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:312:14: note: (near initialization for 's6d7aa0_lsl080al02_desc.drm_mode')
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:415:14: error: initializer element is not constant
    .drm_mode = s6d7aa0_lsl080al03_mode,
                ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:415:14: note: (near initialization for 's6d7aa0_lsl080al03_desc.drm_mode')
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:443:14: error: initializer element is not constant
    .drm_mode = s6d7aa0_ltl101at01_mode,
                ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c:443:14: note: (near initialization for 's6d7aa0_ltl101at01_desc.drm_mode')

Reported-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/lkml/20230523180212.GA1401867@dev-arch.thelio-3990X
Reported-by: kernelci.org bot <bot@kernelci.org>
Link: https://lore.kernel.org/llvm/646c6def.a70a0220.58c1a.903d@mx.google.com
Fixes: 6810bb390282 ("drm/panel: Add Samsung S6D7AA0 panel controller driver")
Signed-off-by: Artur Weber <aweber.kernel@gmail.com>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524084324.11840-1-aweber.kernel@gmail.com
---
 drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
index f532aa0184280..102e1fc7ee383 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c
@@ -37,7 +37,7 @@ struct s6d7aa0_panel_desc {
 	unsigned int panel_type;
 	int (*init_func)(struct s6d7aa0 *ctx);
 	int (*off_func)(struct s6d7aa0 *ctx);
-	const struct drm_display_mode drm_mode;
+	const struct drm_display_mode *drm_mode;
 	unsigned long mode_flags;
 	u32 bus_flags;
 	bool has_backlight;
@@ -309,7 +309,7 @@ static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al02_desc = {
 	.panel_type = S6D7AA0_PANEL_LSL080AL02,
 	.init_func = s6d7aa0_lsl080al02_init,
 	.off_func = s6d7aa0_lsl080al02_off,
-	.drm_mode = s6d7aa0_lsl080al02_mode,
+	.drm_mode = &s6d7aa0_lsl080al02_mode,
 	.mode_flags = MIPI_DSI_MODE_VSYNC_FLUSH | MIPI_DSI_MODE_VIDEO_NO_HFP,
 	.bus_flags = DRM_BUS_FLAG_DE_HIGH,
 
@@ -412,7 +412,7 @@ static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al03_desc = {
 	.panel_type = S6D7AA0_PANEL_LSL080AL03,
 	.init_func = s6d7aa0_lsl080al03_init,
 	.off_func = s6d7aa0_lsl080al03_off,
-	.drm_mode = s6d7aa0_lsl080al03_mode,
+	.drm_mode = &s6d7aa0_lsl080al03_mode,
 	.mode_flags = MIPI_DSI_MODE_NO_EOT_PACKET,
 	.bus_flags = 0,
 
@@ -440,7 +440,7 @@ static const struct s6d7aa0_panel_desc s6d7aa0_ltl101at01_desc = {
 	.panel_type = S6D7AA0_PANEL_LTL101AT01,
 	.init_func = s6d7aa0_lsl080al03_init, /* Similar init to LSL080AL03 */
 	.off_func = s6d7aa0_lsl080al03_off,
-	.drm_mode = s6d7aa0_ltl101at01_mode,
+	.drm_mode = &s6d7aa0_ltl101at01_mode,
 	.mode_flags = MIPI_DSI_MODE_NO_EOT_PACKET,
 	.bus_flags = 0,
 
@@ -458,7 +458,7 @@ static int s6d7aa0_get_modes(struct drm_panel *panel,
 	if (!ctx)
 		return -EINVAL;
 
-	mode = drm_mode_duplicate(connector->dev, &ctx->desc->drm_mode);
+	mode = drm_mode_duplicate(connector->dev, ctx->desc->drm_mode);
 	if (!mode)
 		return -ENOMEM;
 
-- 
GitLab


From 5565a8e1a09a3ed2ac438a9e4f4c84c11037611d Mon Sep 17 00:00:00 2001
From: Azeem Shaikh <azeemshaikh38@gmail.com>
Date: Tue, 23 May 2023 02:19:43 +0000
Subject: [PATCH 0633/1791] dma-buf/sw_sync: Replace all non-returning strlcpy
 with strscpy

strlcpy() reads the entire source buffer first.
This read may exceed the destination size limit.
This is both inefficient and can lead to linear read
overflows if a source string is not NUL-terminated [1].
In an effort to remove strlcpy() completely [2], replace
strlcpy() here with strscpy().
No return values were used, so direct replacement is safe.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

Signed-off-by: Azeem Shaikh <azeemshaikh38@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Sumit Semwal <sumit.semwal@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523021943.2406847-1-azeemshaikh38@gmail.com
---
 drivers/dma-buf/sw_sync.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
index 348b3a9170fa4..63f0aeb66db6b 100644
--- a/drivers/dma-buf/sw_sync.c
+++ b/drivers/dma-buf/sw_sync.c
@@ -85,7 +85,7 @@ static struct sync_timeline *sync_timeline_create(const char *name)
 
 	kref_init(&obj->kref);
 	obj->context = dma_fence_context_alloc(1);
-	strlcpy(obj->name, name, sizeof(obj->name));
+	strscpy(obj->name, name, sizeof(obj->name));
 
 	obj->pt_tree = RB_ROOT;
 	INIT_LIST_HEAD(&obj->pt_list);
-- 
GitLab


From d92b057f9270d83b4d71ef0a8f97b9eeb2be7a3f Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 May 2023 17:31:33 +0200
Subject: [PATCH 0634/1791] drm: shmobile: Use %p4cc to print fourcc codes

Replace the printing of hexadecimal fourcc format codes by
pretty-printed format names, using the "%p4cc" format specifier.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/1a63e5afc8a85f7ae10f7f4395c2d065ba1aeb59.1684854992.git.geert+renesas@glider.be
---
 drivers/gpu/drm/shmobile/shmob_drm_crtc.c | 4 ++--
 drivers/gpu/drm/shmobile/shmob_drm_kms.c  | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index d354ab3077cec..713a7612244c6 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -355,8 +355,8 @@ static int shmob_drm_crtc_mode_set(struct drm_crtc *crtc,
 
 	format = shmob_drm_format_info(crtc->primary->fb->format->format);
 	if (format == NULL) {
-		dev_dbg(sdev->dev, "mode_set: unsupported format %08x\n",
-			crtc->primary->fb->format->format);
+		dev_dbg(sdev->dev, "mode_set: unsupported format %p4cc\n",
+			&crtc->primary->fb->format->format);
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
index 60a2c8d8a0d94..3c5fe3bc183c7 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_kms.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
@@ -96,8 +96,8 @@ shmob_drm_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 
 	format = shmob_drm_format_info(mode_cmd->pixel_format);
 	if (format == NULL) {
-		dev_dbg(dev->dev, "unsupported pixel format %08x\n",
-			mode_cmd->pixel_format);
+		dev_dbg(dev->dev, "unsupported pixel format %p4cc\n",
+			&mode_cmd->pixel_format);
 		return ERR_PTR(-EINVAL);
 	}
 
-- 
GitLab


From 2a98bad77970103fec992dabba6fad417693c849 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 May 2023 17:31:34 +0200
Subject: [PATCH 0635/1791] drm: shmobile: Add support for DRM_FORMAT_XRGB8888

DRM_FORMAT_XRGB8888 aka XR24 is the modus francus of DRM, and should be
supported by all drivers.

The handling for DRM_FORMAT_XRGB8888 is similar to DRM_FORMAT_ARGB8888,
just ignore the alpha channel.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/888cde6d1f70f1610931c4a89801060bb6765ff2.1684854992.git.geert+renesas@glider.be
---
 drivers/gpu/drm/shmobile/shmob_drm_crtc.c  | 1 +
 drivers/gpu/drm/shmobile/shmob_drm_kms.c   | 5 +++++
 drivers/gpu/drm/shmobile/shmob_drm_plane.c | 5 +++++
 3 files changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index 713a7612244c6..08dc1428aa16c 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -232,6 +232,7 @@ static void shmob_drm_crtc_start(struct shmob_drm_crtc *scrtc)
 		value = LDDDSR_LS | LDDDSR_WS | LDDDSR_BS;
 		break;
 	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_XRGB8888:
 	default:
 		value = LDDDSR_LS;
 		break;
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
index 3c5fe3bc183c7..99381cc0abf3a 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_kms.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.c
@@ -39,6 +39,11 @@ static const struct shmob_drm_format_info shmob_drm_format_infos[] = {
 		.bpp = 32,
 		.yuv = false,
 		.lddfr = LDDFR_PKF_ARGB32,
+	}, {
+		.fourcc = DRM_FORMAT_XRGB8888,
+		.bpp = 32,
+		.yuv = false,
+		.lddfr = LDDFR_PKF_ARGB32,
 	}, {
 		.fourcc = DRM_FORMAT_NV12,
 		.bpp = 12,
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.c b/drivers/gpu/drm/shmobile/shmob_drm_plane.c
index 604ae23825daa..850986cee8482 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_plane.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_plane.c
@@ -80,6 +80,7 @@ static void __shmob_drm_plane_setup(struct shmob_drm_plane *splane,
 		format |= LDBBSIFR_SWPL | LDBBSIFR_SWPW | LDBBSIFR_SWPB;
 		break;
 	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_XRGB8888:
 	default:
 		format |= LDBBSIFR_SWPL;
 		break;
@@ -95,6 +96,9 @@ static void __shmob_drm_plane_setup(struct shmob_drm_plane *splane,
 	case DRM_FORMAT_ARGB8888:
 		format |= LDBBSIFR_AL_PK | LDBBSIFR_RY | LDDFR_PKF_ARGB32;
 		break;
+	case DRM_FORMAT_XRGB8888:
+		format |= LDBBSIFR_AL_1 | LDBBSIFR_RY | LDDFR_PKF_ARGB32;
+		break;
 	case DRM_FORMAT_NV12:
 	case DRM_FORMAT_NV21:
 		format |= LDBBSIFR_AL_1 | LDBBSIFR_CHRR_420;
@@ -231,6 +235,7 @@ static const uint32_t formats[] = {
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_RGB888,
 	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_NV12,
 	DRM_FORMAT_NV21,
 	DRM_FORMAT_NV16,
-- 
GitLab


From 5afe82398ffe5eac5678e74ed2bc24fc95edc2b5 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 May 2023 17:31:35 +0200
Subject: [PATCH 0636/1791] drm: shmobile: Switch to
 drm_crtc_init_with_planes()

The SH-Mobile DRM driver uses the legacy drm_crtc_init(), which
advertizes only the formats in safe_modeset_formats[] (XR24 and AR24) as
being supported.

Switch to drm_crtc_init_with_planes(), and advertize all supported
(A)RGB modes, so we can use RGB565 as the default mode for the console.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/2098de3d33bc479a8569da7dcbafdb685ff0a13a.1684854992.git.geert+renesas@glider.be
---
 drivers/gpu/drm/shmobile/shmob_drm_crtc.c | 30 +++++++++++++++++++++--
 1 file changed, 28 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index 08dc1428aa16c..11dd2bc803e7c 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -18,6 +18,7 @@
 #include <drm/drm_gem_dma_helper.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>
+#include <drm/drm_plane_helper.h>
 #include <drm/drm_probe_helper.h>
 #include <drm/drm_simple_kms_helper.h>
 #include <drm/drm_vblank.h>
@@ -478,16 +479,41 @@ static const struct drm_crtc_funcs crtc_funcs = {
 	.disable_vblank = shmob_drm_disable_vblank,
 };
 
+static const uint32_t modeset_formats[] = {
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_XRGB8888,
+};
+
+static const struct drm_plane_funcs primary_plane_funcs = {
+	DRM_PLANE_NON_ATOMIC_FUNCS,
+};
+
 int shmob_drm_crtc_create(struct shmob_drm_device *sdev)
 {
 	struct drm_crtc *crtc = &sdev->crtc.crtc;
+	struct drm_plane *primary;
 	int ret;
 
 	sdev->crtc.dpms = DRM_MODE_DPMS_OFF;
 
-	ret = drm_crtc_init(sdev->ddev, crtc, &crtc_funcs);
-	if (ret < 0)
+	primary = __drm_universal_plane_alloc(sdev->ddev, sizeof(*primary), 0,
+					      0, &primary_plane_funcs,
+					      modeset_formats,
+					      ARRAY_SIZE(modeset_formats),
+					      NULL, DRM_PLANE_TYPE_PRIMARY,
+					      NULL);
+	if (IS_ERR(primary))
+		return PTR_ERR(primary);
+
+	ret = drm_crtc_init_with_planes(sdev->ddev, crtc, primary, NULL,
+					&crtc_funcs, NULL);
+	if (ret < 0) {
+		drm_plane_cleanup(primary);
+		kfree(primary);
 		return ret;
+	}
 
 	drm_crtc_helper_add(crtc, &crtc_helper_funcs);
 
-- 
GitLab


From 79972f160f20c82b56a248daf056ef9befcc8912 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 May 2023 17:31:36 +0200
Subject: [PATCH 0637/1791] drm: shmobile: Add missing call to
 drm_fbdev_generic_setup()

Set up generic fbdev emulation, to enable support for the Linux console.

Use 16 as the preferred depth, as that is a good compromise between
colorfulness and resource utilization, and the default of the fbdev
driver.

Suggested-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/c836938112fda21762bc9eb2741ccd0cbf1197ef.1684854992.git.geert+renesas@glider.be
---
 drivers/gpu/drm/shmobile/shmob_drm_drv.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index faacfee24763b..30493ce874192 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -16,6 +16,7 @@
 #include <linux/slab.h>
 
 #include <drm/drm_drv.h>
+#include <drm/drm_fbdev_generic.h>
 #include <drm/drm_gem_dma_helper.h>
 #include <drm/drm_module.h>
 #include <drm/drm_probe_helper.h>
@@ -271,6 +272,8 @@ static int shmob_drm_probe(struct platform_device *pdev)
 	if (ret < 0)
 		goto err_irq_uninstall;
 
+	drm_fbdev_generic_setup(ddev, 16);
+
 	return 0;
 
 err_irq_uninstall:
-- 
GitLab


From 4bd65789ba847f39e37e0041c10aad5dff1760b6 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 May 2023 17:31:37 +0200
Subject: [PATCH 0638/1791] drm: shmobile: Make DRM_SHMOBILE visible on Renesas
 SoC platforms

The LCD Controller supported by the drm-shmob driver is present
on SuperH SH-Mobile SoCs, and on Renesas ARM SH/R-Mobile SoCs.
Unfortunately its config option is not visible on either, so the user
can never enable the support.

Fix this by dropping the dependency on ARM (for SuperH), and by widening
the dependency range to ARCH_RENESAS (for ARM).

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/40ac67e0887d833ab4a3f1ec24828dedf0d8e108.1684854992.git.geert+renesas@glider.be
---
 drivers/gpu/drm/shmobile/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/shmobile/Kconfig
index 4ec5dc74a6b0b..ad14112999ad8 100644
--- a/drivers/gpu/drm/shmobile/Kconfig
+++ b/drivers/gpu/drm/shmobile/Kconfig
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 config DRM_SHMOBILE
 	tristate "DRM Support for SH Mobile"
-	depends on DRM && ARM
-	depends on ARCH_SHMOBILE || COMPILE_TEST
+	depends on DRM
+	depends on ARCH_RENESAS || ARCH_SHMOBILE || COMPILE_TEST
 	select BACKLIGHT_CLASS_DEVICE
 	select DRM_KMS_HELPER
 	select DRM_GEM_DMA_HELPER
-- 
GitLab


From 9a2cb1b31c040e2f1b313e2f7921f0f5e6b66d82 Mon Sep 17 00:00:00 2001
From: Azeem Shaikh <azeemshaikh38@gmail.com>
Date: Mon, 22 May 2023 15:52:28 +0000
Subject: [PATCH 0639/1791] drm/i915: Replace all non-returning strlcpy with
 strscpy

strlcpy() reads the entire source buffer first.
This read may exceed the destination size limit.
This is both inefficient and can lead to linear read
overflows if a source string is not NUL-terminated [1].
In an effort to remove strlcpy() completely [2], replace
strlcpy() here with strscpy().
No return values were used, so direct replacement is safe.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strlcpy
[2] https://github.com/KSPP/linux/issues/89

Signed-off-by: Azeem Shaikh <azeemshaikh38@gmail.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230522155228.2336755-1-azeemshaikh38@gmail.com
---
 drivers/gpu/drm/i915/selftests/i915_perf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c
index 24dde55314230..d4608b220123c 100644
--- a/drivers/gpu/drm/i915/selftests/i915_perf.c
+++ b/drivers/gpu/drm/i915/selftests/i915_perf.c
@@ -28,7 +28,7 @@ alloc_empty_config(struct i915_perf *perf)
 	oa_config->perf = perf;
 	kref_init(&oa_config->ref);
 
-	strlcpy(oa_config->uuid, TEST_OA_CONFIG_UUID, sizeof(oa_config->uuid));
+	strscpy(oa_config->uuid, TEST_OA_CONFIG_UUID, sizeof(oa_config->uuid));
 
 	mutex_lock(&perf->metrics_lock);
 
-- 
GitLab


From 05aa8e0135094ae3d1e6837b5457a740266d7cfc Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:04 -0700
Subject: [PATCH 0640/1791] drm/i915/display: Move display device info to
 header under display/

Moving display-specific substructure definitions will help keep display
more self-contained and make it easier to re-use in other drivers (i.e.,
Xe) in the future.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-2-matthew.d.roper@intel.com
---
 .../drm/i915/display/intel_display_device.h   | 60 +++++++++++++++++++
 drivers/gpu/drm/i915/intel_device_info.h      | 49 +--------------
 2 files changed, 62 insertions(+), 47 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_device.h

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
new file mode 100644
index 0000000000000..c689d582dbf13
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_DISPLAY_DEVICE_H__
+#define __INTEL_DISPLAY_DEVICE_H__
+
+#include <linux/types.h>
+
+#include "display/intel_display_limits.h"
+
+#define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \
+	/* Keep in alphabetical order */ \
+	func(cursor_needs_physical); \
+	func(has_cdclk_crawl); \
+	func(has_cdclk_squash); \
+	func(has_ddi); \
+	func(has_dp_mst); \
+	func(has_dsb); \
+	func(has_fpga_dbg); \
+	func(has_gmch); \
+	func(has_hotplug); \
+	func(has_hti); \
+	func(has_ipc); \
+	func(has_overlay); \
+	func(has_psr); \
+	func(has_psr_hw_tracking); \
+	func(overlay_needs_physical); \
+	func(supports_tv);
+
+struct intel_display_device_info {
+	u8 abox_mask;
+
+	struct {
+		u16 size; /* in blocks */
+		u8 slice_mask;
+	} dbuf;
+
+#define DEFINE_FLAG(name) u8 name:1
+	DEV_INFO_DISPLAY_FOR_EACH_FLAG(DEFINE_FLAG);
+#undef DEFINE_FLAG
+
+	/* Global register offset for the display engine */
+	u32 mmio_offset;
+
+	/* Register offsets for the various display pipes and transcoders */
+	u32 pipe_offsets[I915_MAX_TRANSCODERS];
+	u32 trans_offsets[I915_MAX_TRANSCODERS];
+	u32 cursor_offsets[I915_MAX_PIPES];
+
+	struct {
+		u32 degamma_lut_size;
+		u32 gamma_lut_size;
+		u32 degamma_lut_tests;
+		u32 gamma_lut_tests;
+	} color;
+};
+
+#endif
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index f032f2500f505..c14bc3f5d0fa1 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -29,7 +29,7 @@
 
 #include "intel_step.h"
 
-#include "display/intel_display_limits.h"
+#include "display/intel_display_device.h"
 
 #include "gt/intel_engine_types.h"
 #include "gt/intel_context_types.h"
@@ -180,25 +180,6 @@ enum intel_ppgtt_type {
 	func(unfenced_needs_alignment); \
 	func(hws_needs_physical);
 
-#define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \
-	/* Keep in alphabetical order */ \
-	func(cursor_needs_physical); \
-	func(has_cdclk_crawl); \
-	func(has_cdclk_squash); \
-	func(has_ddi); \
-	func(has_dp_mst); \
-	func(has_dsb); \
-	func(has_fpga_dbg); \
-	func(has_gmch); \
-	func(has_hotplug); \
-	func(has_hti); \
-	func(has_ipc); \
-	func(has_overlay); \
-	func(has_psr); \
-	func(has_psr_hw_tracking); \
-	func(overlay_needs_physical); \
-	func(supports_tv);
-
 struct intel_ip_version {
 	u8 ver;
 	u8 rel;
@@ -276,33 +257,7 @@ struct intel_device_info {
 	DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG);
 #undef DEFINE_FLAG
 
-	struct {
-		u8 abox_mask;
-
-		struct {
-			u16 size; /* in blocks */
-			u8 slice_mask;
-		} dbuf;
-
-#define DEFINE_FLAG(name) u8 name:1
-		DEV_INFO_DISPLAY_FOR_EACH_FLAG(DEFINE_FLAG);
-#undef DEFINE_FLAG
-
-		/* Global register offset for the display engine */
-		u32 mmio_offset;
-
-		/* Register offsets for the various display pipes and transcoders */
-		u32 pipe_offsets[I915_MAX_TRANSCODERS];
-		u32 trans_offsets[I915_MAX_TRANSCODERS];
-		u32 cursor_offsets[I915_MAX_PIPES];
-
-		struct {
-			u32 degamma_lut_size;
-			u32 gamma_lut_size;
-			u32 degamma_lut_tests;
-			u32 gamma_lut_tests;
-		} color;
-	} display;
+	struct intel_display_device_info display;
 
 	/*
 	 * Initial runtime info. Do not access outside of i915_driver_create().
-- 
GitLab


From 5af5169d758275490ab00e209a09dc1d6b85e535 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:05 -0700
Subject: [PATCH 0641/1791] drm/i915: Convert INTEL_INFO()->display to a
 pointer

Rather than embeddeding the display's device info within the main device
info structure, just provide a pointer to the display-specific
structure.  This is in preparation for moving the display device info
definitions into the display code itself and for eventually allowing the
pointer to be assigned at runtime on platforms that use GMD_ID for
device identification.

In the future, this will also eventually allow the same display device
info structures to be used outside the current i915 code (e.g., from the
Xe driver).

v2:
 - Move introduction of DISPLAY_INFO() to this patch.  (Andrzej)
v3:
 - Also use DISPLAY_INFO() in intel_display_reg_defs.h.  (Andrzej)
 - Use "{}" instead of "{ 0 }" for empty struct init.  (Jani)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Acked-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-3-matthew.d.roper@intel.com
---
 drivers/gpu/drm/i915/display/intel_color.c    |  31 +-
 drivers/gpu/drm/i915/display/intel_cursor.c   |   2 +-
 drivers/gpu/drm/i915/display/intel_display.h  |   2 +-
 .../drm/i915/display/intel_display_power.c    |   6 +-
 .../drm/i915/display/intel_display_reg_defs.h |  14 +-
 drivers/gpu/drm/i915/display/intel_fb_pin.c   |   2 +-
 drivers/gpu/drm/i915/display/intel_hti.c      |   2 +-
 drivers/gpu/drm/i915/display/skl_watermark.c  |   8 +-
 drivers/gpu/drm/i915/i915_drv.h               |  29 +-
 drivers/gpu/drm/i915/i915_pci.c               | 575 ++++++++++++------
 drivers/gpu/drm/i915/intel_device_info.c      |   6 +-
 drivers/gpu/drm/i915/intel_device_info.h      |   2 +-
 12 files changed, 450 insertions(+), 229 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 07f1afe1d4067..744b3a4ec99a6 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -1824,14 +1824,14 @@ static u32 intel_gamma_lut_tests(const struct intel_crtc_state *crtc_state)
 	if (lut_is_legacy(gamma_lut))
 		return 0;
 
-	return INTEL_INFO(i915)->display.color.gamma_lut_tests;
+	return DISPLAY_INFO(i915)->color.gamma_lut_tests;
 }
 
 static u32 intel_degamma_lut_tests(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
-	return INTEL_INFO(i915)->display.color.degamma_lut_tests;
+	return DISPLAY_INFO(i915)->color.degamma_lut_tests;
 }
 
 static int intel_gamma_lut_size(const struct intel_crtc_state *crtc_state)
@@ -1842,14 +1842,14 @@ static int intel_gamma_lut_size(const struct intel_crtc_state *crtc_state)
 	if (lut_is_legacy(gamma_lut))
 		return LEGACY_LUT_LENGTH;
 
-	return INTEL_INFO(i915)->display.color.gamma_lut_size;
+	return DISPLAY_INFO(i915)->color.gamma_lut_size;
 }
 
 static u32 intel_degamma_lut_size(const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
-	return INTEL_INFO(i915)->display.color.degamma_lut_size;
+	return DISPLAY_INFO(i915)->color.degamma_lut_size;
 }
 
 static int check_lut_size(const struct drm_property_blob *lut, int expected)
@@ -2321,7 +2321,7 @@ static int glk_assign_luts(struct intel_crtc_state *crtc_state)
 		struct drm_property_blob *gamma_lut;
 
 		gamma_lut = create_resized_lut(i915, crtc_state->hw.gamma_lut,
-					       INTEL_INFO(i915)->display.color.degamma_lut_size,
+					       DISPLAY_INFO(i915)->color.degamma_lut_size,
 					       false);
 		if (IS_ERR(gamma_lut))
 			return PTR_ERR(gamma_lut);
@@ -2855,7 +2855,7 @@ static struct drm_property_blob *i9xx_read_lut_8(struct intel_crtc *crtc)
 static struct drm_property_blob *i9xx_read_lut_10(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	u32 lut_size = INTEL_INFO(dev_priv)->display.color.gamma_lut_size;
+	u32 lut_size = DISPLAY_INFO(dev_priv)->color.gamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -2904,7 +2904,7 @@ static void i9xx_read_luts(struct intel_crtc_state *crtc_state)
 static struct drm_property_blob *i965_read_lut_10p6(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(dev_priv)->display.color.gamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(dev_priv)->color.gamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -2954,7 +2954,7 @@ static void i965_read_luts(struct intel_crtc_state *crtc_state)
 static struct drm_property_blob *chv_read_cgm_degamma(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(dev_priv)->display.color.degamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(dev_priv)->color.degamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -2980,7 +2980,7 @@ static struct drm_property_blob *chv_read_cgm_degamma(struct intel_crtc *crtc)
 static struct drm_property_blob *chv_read_cgm_gamma(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -3044,7 +3044,7 @@ static struct drm_property_blob *ilk_read_lut_8(struct intel_crtc *crtc)
 static struct drm_property_blob *ilk_read_lut_10(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -3228,7 +3228,7 @@ static void bdw_read_luts(struct intel_crtc_state *crtc_state)
 static struct drm_property_blob *glk_read_degamma_lut(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(dev_priv)->display.color.degamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(dev_priv)->color.degamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -3293,7 +3293,7 @@ static struct drm_property_blob *
 icl_read_lut_multi_segment(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
-	int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size;
+	int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
 	enum pipe pipe = crtc->pipe;
 	struct drm_property_blob *blob;
 	struct drm_color_lut *lut;
@@ -3471,8 +3471,8 @@ void intel_color_crtc_init(struct intel_crtc *crtc)
 
 	drm_mode_crtc_set_gamma_size(&crtc->base, 256);
 
-	gamma_lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size;
-	degamma_lut_size = INTEL_INFO(i915)->display.color.degamma_lut_size;
+	gamma_lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
+	degamma_lut_size = DISPLAY_INFO(i915)->color.degamma_lut_size;
 	has_ctm = degamma_lut_size != 0;
 
 	/*
@@ -3497,7 +3497,8 @@ int intel_color_init(struct drm_i915_private *i915)
 	if (DISPLAY_VER(i915) != 10)
 		return 0;
 
-	blob = create_linear_lut(i915, INTEL_INFO(i915)->display.color.degamma_lut_size);
+	blob = create_linear_lut(i915,
+				 DISPLAY_INFO(i915)->color.degamma_lut_size);
 	if (IS_ERR(blob))
 		return PTR_ERR(blob);
 
diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c
index 31bef04273779..3864da5f5c17f 100644
--- a/drivers/gpu/drm/i915/display/intel_cursor.c
+++ b/drivers/gpu/drm/i915/display/intel_cursor.c
@@ -36,7 +36,7 @@ static u32 intel_cursor_base(const struct intel_plane_state *plane_state)
 	const struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	u32 base;
 
-	if (INTEL_INFO(dev_priv)->display.cursor_needs_physical)
+	if (DISPLAY_INFO(dev_priv)->cursor_needs_physical)
 		base = sg_dma_address(obj->mm.pages->sgl);
 	else
 		base = intel_plane_ggtt_offset(plane_state);
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 205b3929b8614..aa3a21ccd7fe1 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -113,7 +113,7 @@ enum i9xx_plane_id {
 
 #define for_each_dbuf_slice(__dev_priv, __slice) \
 	for ((__slice) = DBUF_S1; (__slice) < I915_MAX_DBUF_SLICES; (__slice)++) \
-		for_each_if(INTEL_INFO(__dev_priv)->display.dbuf.slice_mask & BIT(__slice))
+		for_each_if(INTEL_INFO(__dev_priv)->display->dbuf.slice_mask & BIT(__slice))
 
 #define for_each_dbuf_slice_in_mask(__dev_priv, __slice, __mask) \
 	for_each_dbuf_slice((__dev_priv), (__slice)) \
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 6ed2ece89c3fd..9c9a809c71f1d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -1053,7 +1053,7 @@ void gen9_dbuf_slices_update(struct drm_i915_private *dev_priv,
 			     u8 req_slices)
 {
 	struct i915_power_domains *power_domains = &dev_priv->display.power.domains;
-	u8 slice_mask = INTEL_INFO(dev_priv)->display.dbuf.slice_mask;
+	u8 slice_mask = DISPLAY_INFO(dev_priv)->dbuf.slice_mask;
 	enum dbuf_slice slice;
 
 	drm_WARN(&dev_priv->drm, req_slices & ~slice_mask,
@@ -1113,7 +1113,7 @@ static void gen12_dbuf_slices_config(struct drm_i915_private *dev_priv)
 
 static void icl_mbus_init(struct drm_i915_private *dev_priv)
 {
-	unsigned long abox_regs = INTEL_INFO(dev_priv)->display.abox_mask;
+	unsigned long abox_regs = DISPLAY_INFO(dev_priv)->abox_mask;
 	u32 mask, val, i;
 
 	if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(dev_priv) >= 14)
@@ -1568,7 +1568,7 @@ static void tgl_bw_buddy_init(struct drm_i915_private *dev_priv)
 	enum intel_dram_type type = dev_priv->dram_info.type;
 	u8 num_channels = dev_priv->dram_info.num_channels;
 	const struct buddy_page_mask *table;
-	unsigned long abox_mask = INTEL_INFO(dev_priv)->display.abox_mask;
+	unsigned long abox_mask = DISPLAY_INFO(dev_priv)->abox_mask;
 	int config, i;
 
 	/* BW_BUDDY registers are not used on dgpu's beyond DG1 */
diff --git a/drivers/gpu/drm/i915/display/intel_display_reg_defs.h b/drivers/gpu/drm/i915/display/intel_display_reg_defs.h
index 755c1ea8225c5..2f07b7afa3bfe 100644
--- a/drivers/gpu/drm/i915/display/intel_display_reg_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_display_reg_defs.h
@@ -8,7 +8,7 @@
 
 #include "i915_reg_defs.h"
 
-#define DISPLAY_MMIO_BASE(dev_priv)	(INTEL_INFO(dev_priv)->display.mmio_offset)
+#define DISPLAY_MMIO_BASE(dev_priv)	(DISPLAY_INFO(dev_priv)->mmio_offset)
 
 #define VLV_DISPLAY_BASE		0x180000
 
@@ -36,14 +36,14 @@
  * Device info offset array based helpers for groups of registers with unevenly
  * spaced base offsets.
  */
-#define _MMIO_PIPE2(pipe, reg)		_MMIO(INTEL_INFO(dev_priv)->display.pipe_offsets[(pipe)] - \
-					      INTEL_INFO(dev_priv)->display.pipe_offsets[PIPE_A] + \
+#define _MMIO_PIPE2(pipe, reg)		_MMIO(DISPLAY_INFO(dev_priv)->pipe_offsets[(pipe)] - \
+					      DISPLAY_INFO(dev_priv)->pipe_offsets[PIPE_A] + \
 					      DISPLAY_MMIO_BASE(dev_priv) + (reg))
-#define _MMIO_TRANS2(tran, reg)		_MMIO(INTEL_INFO(dev_priv)->display.trans_offsets[(tran)] - \
-					      INTEL_INFO(dev_priv)->display.trans_offsets[TRANSCODER_A] + \
+#define _MMIO_TRANS2(tran, reg)		_MMIO(DISPLAY_INFO(dev_priv)->trans_offsets[(tran)] - \
+					      DISPLAY_INFO(dev_priv)->trans_offsets[TRANSCODER_A] + \
 					      DISPLAY_MMIO_BASE(dev_priv) + (reg))
-#define _MMIO_CURSOR2(pipe, reg)	_MMIO(INTEL_INFO(dev_priv)->display.cursor_offsets[(pipe)] - \
-					      INTEL_INFO(dev_priv)->display.cursor_offsets[PIPE_A] + \
+#define _MMIO_CURSOR2(pipe, reg)	_MMIO(DISPLAY_INFO(dev_priv)->cursor_offsets[(pipe)] - \
+					      DISPLAY_INFO(dev_priv)->cursor_offsets[PIPE_A] + \
 					      DISPLAY_MMIO_BASE(dev_priv) + (reg))
 
 #endif /* __INTEL_DISPLAY_REG_DEFS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 1aca7552a85d0..fffd568070d41 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -243,7 +243,7 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state)
 	struct i915_vma *vma;
 	bool phys_cursor =
 		plane->id == PLANE_CURSOR &&
-		INTEL_INFO(dev_priv)->display.cursor_needs_physical;
+		DISPLAY_INFO(dev_priv)->cursor_needs_physical;
 
 	if (!intel_fb_uses_dpt(fb)) {
 		vma = intel_pin_and_fence_fb_obj(fb, phys_cursor,
diff --git a/drivers/gpu/drm/i915/display/intel_hti.c b/drivers/gpu/drm/i915/display/intel_hti.c
index c518efebdf779..a92d008d4e6e5 100644
--- a/drivers/gpu/drm/i915/display/intel_hti.c
+++ b/drivers/gpu/drm/i915/display/intel_hti.c
@@ -15,7 +15,7 @@ void intel_hti_init(struct drm_i915_private *i915)
 	 * If the platform has HTI, we need to find out whether it has reserved
 	 * any display resources before we create our display outputs.
 	 */
-	if (INTEL_INFO(i915)->display.has_hti)
+	if (DISPLAY_INFO(i915)->has_hti)
 		i915->display.hti.state = intel_de_read(i915, HDPORT_STATE);
 }
 
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index 1c7e6468f3e34..d1245c847f1cb 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -507,8 +507,8 @@ static u16 skl_ddb_entry_init(struct skl_ddb_entry *entry,
 
 static int intel_dbuf_slice_size(struct drm_i915_private *i915)
 {
-	return INTEL_INFO(i915)->display.dbuf.size /
-		hweight8(INTEL_INFO(i915)->display.dbuf.slice_mask);
+	return DISPLAY_INFO(i915)->dbuf.size /
+		hweight8(DISPLAY_INFO(i915)->dbuf.slice_mask);
 }
 
 static void
@@ -527,7 +527,7 @@ skl_ddb_entry_for_slices(struct drm_i915_private *i915, u8 slice_mask,
 	ddb->end = fls(slice_mask) * slice_size;
 
 	WARN_ON(ddb->start >= ddb->end);
-	WARN_ON(ddb->end > INTEL_INFO(i915)->display.dbuf.size);
+	WARN_ON(ddb->end > DISPLAY_INFO(i915)->dbuf.size);
 }
 
 static unsigned int mbus_ddb_offset(struct drm_i915_private *i915, u8 slice_mask)
@@ -2625,7 +2625,7 @@ skl_compute_ddb(struct intel_atomic_state *state)
 			    "Enabled dbuf slices 0x%x -> 0x%x (total dbuf slices 0x%x), mbus joined? %s->%s\n",
 			    old_dbuf_state->enabled_slices,
 			    new_dbuf_state->enabled_slices,
-			    INTEL_INFO(i915)->display.dbuf.slice_mask,
+			    DISPLAY_INFO(i915)->dbuf.slice_mask,
 			    str_yes_no(old_dbuf_state->joined_mbus),
 			    str_yes_no(new_dbuf_state->joined_mbus));
 	}
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index fe7eeafe9cff6..0c341107363f7 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -408,6 +408,7 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 	     (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node)))
 
 #define INTEL_INFO(dev_priv)	(&(dev_priv)->__info)
+#define DISPLAY_INFO(i915)	(INTEL_INFO(i915)->display)
 #define RUNTIME_INFO(dev_priv)	(&(dev_priv)->__runtime)
 #define DRIVER_CAPS(dev_priv)	(&(dev_priv)->caps)
 
@@ -782,9 +783,9 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 	((sizes) & ~RUNTIME_INFO(dev_priv)->page_sizes) == 0; \
 })
 
-#define HAS_OVERLAY(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_overlay)
+#define HAS_OVERLAY(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_overlay)
 #define OVERLAY_NEEDS_PHYSICAL(dev_priv) \
-		(INTEL_INFO(dev_priv)->display.overlay_needs_physical)
+		(DISPLAY_INFO(dev_priv)->overlay_needs_physical)
 
 /* Early gen2 have a totally busted CS tlb and require pinned batches. */
 #define HAS_BROKEN_CS_TLB(dev_priv)	(IS_I830(dev_priv) || IS_I845G(dev_priv))
@@ -806,8 +807,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
  */
 #define HAS_128_BYTE_Y_TILING(dev_priv) (GRAPHICS_VER(dev_priv) != 2 && \
 					 !(IS_I915G(dev_priv) || IS_I915GM(dev_priv)))
-#define SUPPORTS_TV(dev_priv)		(INTEL_INFO(dev_priv)->display.supports_tv)
-#define I915_HAS_HOTPLUG(dev_priv)	(INTEL_INFO(dev_priv)->display.has_hotplug)
+#define SUPPORTS_TV(dev_priv)		(DISPLAY_INFO(dev_priv)->supports_tv)
+#define I915_HAS_HOTPLUG(dev_priv)	(DISPLAY_INFO(dev_priv)->has_hotplug)
 
 #define HAS_FW_BLC(dev_priv)	(DISPLAY_VER(dev_priv) > 2)
 #define HAS_FBC(dev_priv)	(RUNTIME_INFO(dev_priv)->fbc_mask != 0)
@@ -817,18 +818,18 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_IPS(dev_priv)	(IS_HSW_ULT(dev_priv) || IS_BROADWELL(dev_priv))
 
-#define HAS_DP_MST(dev_priv)	(INTEL_INFO(dev_priv)->display.has_dp_mst)
+#define HAS_DP_MST(dev_priv)	(DISPLAY_INFO(dev_priv)->has_dp_mst)
 #define HAS_DP20(dev_priv)	(IS_DG2(dev_priv) || DISPLAY_VER(dev_priv) >= 14)
 
 #define HAS_DOUBLE_BUFFERED_M_N(dev_priv)	(DISPLAY_VER(dev_priv) >= 9 || IS_BROADWELL(dev_priv))
 
-#define HAS_CDCLK_CRAWL(dev_priv)	 (INTEL_INFO(dev_priv)->display.has_cdclk_crawl)
-#define HAS_CDCLK_SQUASH(dev_priv)	 (INTEL_INFO(dev_priv)->display.has_cdclk_squash)
-#define HAS_DDI(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_ddi)
-#define HAS_FPGA_DBG_UNCLAIMED(dev_priv) (INTEL_INFO(dev_priv)->display.has_fpga_dbg)
-#define HAS_PSR(dev_priv)		 (INTEL_INFO(dev_priv)->display.has_psr)
+#define HAS_CDCLK_CRAWL(dev_priv)	 (DISPLAY_INFO(dev_priv)->has_cdclk_crawl)
+#define HAS_CDCLK_SQUASH(dev_priv)	 (DISPLAY_INFO(dev_priv)->has_cdclk_squash)
+#define HAS_DDI(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_ddi)
+#define HAS_FPGA_DBG_UNCLAIMED(dev_priv) (DISPLAY_INFO(dev_priv)->has_fpga_dbg)
+#define HAS_PSR(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_psr)
 #define HAS_PSR_HW_TRACKING(dev_priv) \
-	(INTEL_INFO(dev_priv)->display.has_psr_hw_tracking)
+	(DISPLAY_INFO(dev_priv)->has_psr_hw_tracking)
 #define HAS_PSR2_SEL_FETCH(dev_priv)	 (DISPLAY_VER(dev_priv) >= 12)
 #define HAS_TRANSCODER(dev_priv, trans)	 ((RUNTIME_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
 
@@ -839,7 +840,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define HAS_RPS(dev_priv)	(INTEL_INFO(dev_priv)->has_rps)
 
 #define HAS_DMC(dev_priv)	(RUNTIME_INFO(dev_priv)->has_dmc)
-#define HAS_DSB(dev_priv)	(INTEL_INFO(dev_priv)->display.has_dsb)
+#define HAS_DSB(dev_priv)	(DISPLAY_INFO(dev_priv)->has_dsb)
 #define HAS_DSC(__i915)		(RUNTIME_INFO(__i915)->has_dsc)
 #define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915))
 
@@ -869,7 +870,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
  */
 #define HAS_64K_PAGES(dev_priv) (INTEL_INFO(dev_priv)->has_64k_pages)
 
-#define HAS_IPC(dev_priv)		(INTEL_INFO(dev_priv)->display.has_ipc)
+#define HAS_IPC(dev_priv)		(DISPLAY_INFO(dev_priv)->has_ipc)
 #define HAS_SAGV(dev_priv)		(DISPLAY_VER(dev_priv) >= 9 && !IS_LP(dev_priv))
 
 #define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i))
@@ -889,7 +890,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_GLOBAL_MOCS_REGISTERS(dev_priv)	(INTEL_INFO(dev_priv)->has_global_mocs)
 
-#define HAS_GMCH(dev_priv) (INTEL_INFO(dev_priv)->display.has_gmch)
+#define HAS_GMCH(dev_priv) (DISPLAY_INFO(dev_priv)->has_gmch)
 
 #define HAS_GMD_ID(i915)	(INTEL_INFO(i915)->has_gmd_id)
 
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index c509ea4aa70f7..2366361bfbe93 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -46,43 +46,43 @@
 #define NO_DISPLAY .__runtime.pipe_mask = 0
 
 #define I845_PIPE_OFFSETS \
-	.display.pipe_offsets = { \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET,	\
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 	}
 
 #define I9XX_PIPE_OFFSETS \
-	.display.pipe_offsets = { \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET,	\
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
 	}
 
 #define IVB_PIPE_OFFSETS \
-	.display.pipe_offsets = { \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET,	\
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
 		[TRANSCODER_C] = PIPE_C_OFFSET, \
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
 		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
 	}
 
 #define HSW_PIPE_OFFSETS \
-	.display.pipe_offsets = { \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET,	\
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
 		[TRANSCODER_C] = PIPE_C_OFFSET, \
 		[TRANSCODER_EDP] = PIPE_EDP_OFFSET, \
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
 		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
@@ -90,44 +90,44 @@
 	}
 
 #define CHV_PIPE_OFFSETS \
-	.display.pipe_offsets = { \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET, \
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
 		[TRANSCODER_C] = CHV_PIPE_C_OFFSET, \
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
 		[TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \
 	}
 
 #define I845_CURSOR_OFFSETS \
-	.display.cursor_offsets = { \
+	.cursor_offsets = { \
 		[PIPE_A] = CURSOR_A_OFFSET, \
 	}
 
 #define I9XX_CURSOR_OFFSETS \
-	.display.cursor_offsets = { \
+	.cursor_offsets = { \
 		[PIPE_A] = CURSOR_A_OFFSET, \
 		[PIPE_B] = CURSOR_B_OFFSET, \
 	}
 
 #define CHV_CURSOR_OFFSETS \
-	.display.cursor_offsets = { \
+	.cursor_offsets = { \
 		[PIPE_A] = CURSOR_A_OFFSET, \
 		[PIPE_B] = CURSOR_B_OFFSET, \
 		[PIPE_C] = CHV_CURSOR_C_OFFSET, \
 	}
 
 #define IVB_CURSOR_OFFSETS \
-	.display.cursor_offsets = { \
+	.cursor_offsets = { \
 		[PIPE_A] = CURSOR_A_OFFSET, \
 		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
 		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
 	}
 
 #define TGL_CURSOR_OFFSETS \
-	.display.cursor_offsets = { \
+	.cursor_offsets = { \
 		[PIPE_A] = CURSOR_A_OFFSET, \
 		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
 		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
@@ -135,29 +135,29 @@
 	}
 
 #define I845_COLORS \
-	.display.color = { .gamma_lut_size = 256 }
+	.color = { .gamma_lut_size = 256 }
 #define I9XX_COLORS \
-	.display.color = { .gamma_lut_size = 129, \
+	.color = { .gamma_lut_size = 129, \
 		   .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 	}
 #define ILK_COLORS \
-	.display.color = { .gamma_lut_size = 1024 }
+	.color = { .gamma_lut_size = 1024 }
 #define IVB_COLORS \
-	.display.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 }
+	.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 }
 #define CHV_COLORS \
-	.display.color = { \
+	.color = { \
 		.degamma_lut_size = 65, .gamma_lut_size = 257, \
 		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
 	}
 #define GLK_COLORS \
-	.display.color = { \
+	.color = { \
 		.degamma_lut_size = 33, .gamma_lut_size = 1024, \
 		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
 				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
 	}
 #define ICL_COLORS \
-	.display.color = { \
+	.color = { \
 		.degamma_lut_size = 33, .gamma_lut_size = 262145, \
 		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
 				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
@@ -172,15 +172,24 @@
 #define GEN_DEFAULT_REGIONS \
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_SMEM
 
+#define I830_DISPLAY \
+	.has_overlay = 1, \
+	.cursor_needs_physical = 1, \
+	.overlay_needs_physical = 1, \
+	.has_gmch = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS
+
+static const struct intel_display_device_info i830_display = {
+	I830_DISPLAY,
+};
+
 #define I830_FEATURES \
 	GEN(2), \
 	.is_mobile = 1, \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.display.has_overlay = 1, \
-	.display.cursor_needs_physical = 1, \
-	.display.overlay_needs_physical = 1, \
-	.display.has_gmch = 1, \
 	.gpu_reset_clobbers_display = true, \
 	.has_3d_pipeline = 1, \
 	.hws_needs_physical = 1, \
@@ -189,19 +198,25 @@
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.dma_mask_size = 32, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
+#define I845_DISPLAY \
+	.has_overlay = 1, \
+	.overlay_needs_physical = 1, \
+	.has_gmch = 1, \
+	I845_PIPE_OFFSETS, \
+	I845_CURSOR_OFFSETS, \
+	I845_COLORS
+
+static const struct intel_display_device_info i845_display = {
+	I845_DISPLAY,
+};
+
 #define I845_FEATURES \
 	GEN(2), \
 	.__runtime.pipe_mask = BIT(PIPE_A), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A), \
-	.display.has_overlay = 1, \
-	.display.overlay_needs_physical = 1, \
-	.display.has_gmch = 1, \
 	.has_3d_pipeline = 1, \
 	.gpu_reset_clobbers_display = true, \
 	.hws_needs_physical = 1, \
@@ -210,58 +225,101 @@
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
 	.dma_mask_size = 32, \
-	I845_PIPE_OFFSETS, \
-	I845_CURSOR_OFFSETS, \
-	I845_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
 static const struct intel_device_info i830_info = {
 	I830_FEATURES,
 	PLATFORM(INTEL_I830),
+	.display = &i830_display,
 };
 
 static const struct intel_device_info i845g_info = {
 	I845_FEATURES,
 	PLATFORM(INTEL_I845G),
+	.display = &i845_display,
+};
+
+static const struct intel_display_device_info i85x_display = {
+	I830_DISPLAY,
 };
 
 static const struct intel_device_info i85x_info = {
 	I830_FEATURES,
 	PLATFORM(INTEL_I85X),
+	.display = &i85x_display,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 };
 
+static const struct intel_display_device_info i865g_display = {
+	I845_DISPLAY,
+};
+
 static const struct intel_device_info i865g_info = {
 	I845_FEATURES,
 	PLATFORM(INTEL_I865G),
+	.display = &i865g_display,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 };
 
+#define GEN3_DISPLAY \
+	.has_gmch = 1, \
+	.has_overlay = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS
+
+static const struct intel_display_device_info i915g_display = {
+	GEN3_DISPLAY,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+};
+
+static const struct intel_display_device_info i915gm_display = {
+	GEN3_DISPLAY,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+	.supports_tv = 1,
+};
+
+static const struct intel_display_device_info i945g_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+};
+
+static const struct intel_display_device_info i945gm_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+	.supports_tv = 1,
+};
+
+static const struct intel_display_device_info g33_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+};
+
 #define GEN3_FEATURES \
 	GEN(3), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.display.has_gmch = 1, \
 	.gpu_reset_clobbers_display = true, \
 	.__runtime.platform_engine_mask = BIT(RCS0), \
 	.has_3d_pipeline = 1, \
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
 	.dma_mask_size = 32, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
 static const struct intel_device_info i915g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I915G),
+	.display = &i915g_display,
 	.has_coherent_ggtt = false,
-	.display.cursor_needs_physical = 1,
-	.display.has_overlay = 1,
-	.display.overlay_needs_physical = 1,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
 };
@@ -269,11 +327,8 @@ static const struct intel_device_info i915g_info = {
 static const struct intel_device_info i915gm_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I915GM),
+	.display = &i915gm_display,
 	.is_mobile = 1,
-	.display.cursor_needs_physical = 1,
-	.display.has_overlay = 1,
-	.display.overlay_needs_physical = 1,
-	.display.supports_tv = 1,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
@@ -282,10 +337,7 @@ static const struct intel_device_info i915gm_info = {
 static const struct intel_device_info i945g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I945G),
-	.display.has_hotplug = 1,
-	.display.cursor_needs_physical = 1,
-	.display.has_overlay = 1,
-	.display.overlay_needs_physical = 1,
+	.display = &i945g_display,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
 };
@@ -293,12 +345,8 @@ static const struct intel_device_info i945g_info = {
 static const struct intel_device_info i945gm_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I945GM),
+	.display = &i945gm_display,
 	.is_mobile = 1,
-	.display.has_hotplug = 1,
-	.display.cursor_needs_physical = 1,
-	.display.has_overlay = 1,
-	.display.overlay_needs_physical = 1,
-	.display.supports_tv = 1,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
@@ -307,16 +355,14 @@ static const struct intel_device_info i945gm_info = {
 static const struct intel_device_info g33_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_G33),
-	.display.has_hotplug = 1,
-	.display.has_overlay = 1,
+	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
 static const struct intel_device_info pnv_g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_PINEVIEW),
-	.display.has_hotplug = 1,
-	.display.has_overlay = 1,
+	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
@@ -324,33 +370,54 @@ static const struct intel_device_info pnv_m_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_PINEVIEW),
 	.is_mobile = 1,
-	.display.has_hotplug = 1,
-	.display.has_overlay = 1,
+	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
+#define GEN4_DISPLAY \
+	.has_hotplug = 1, \
+	.has_gmch = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS
+
+static const struct intel_display_device_info i965g_display = {
+	GEN4_DISPLAY,
+	.has_overlay = 1,
+};
+
+static const struct intel_display_device_info i965gm_display = {
+	GEN4_DISPLAY,
+	.has_overlay = 1,
+	.supports_tv = 1,
+};
+
+static const struct intel_display_device_info g45_display = {
+	GEN4_DISPLAY,
+};
+
+static const struct intel_display_device_info gm45_display = {
+	GEN4_DISPLAY,
+	.supports_tv = 1,
+};
+
 #define GEN4_FEATURES \
 	GEN(4), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.display.has_hotplug = 1, \
-	.display.has_gmch = 1, \
 	.gpu_reset_clobbers_display = true, \
 	.__runtime.platform_engine_mask = BIT(RCS0), \
 	.has_3d_pipeline = 1, \
 	.has_snoop = true, \
 	.has_coherent_ggtt = true, \
 	.dma_mask_size = 36, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
 static const struct intel_device_info i965g_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_I965G),
-	.display.has_overlay = 1,
+	.display = &i965g_display,
 	.hws_needs_physical = 1,
 	.has_snoop = false,
 };
@@ -358,10 +425,9 @@ static const struct intel_device_info i965g_info = {
 static const struct intel_device_info i965gm_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_I965GM),
+	.display = &i965gm_display,
 	.is_mobile = 1,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
-	.display.has_overlay = 1,
-	.display.supports_tv = 1,
 	.hws_needs_physical = 1,
 	.has_snoop = false,
 };
@@ -370,6 +436,7 @@ static const struct intel_device_info g45_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_G45),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0),
+	.display = &g45_display,
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -378,8 +445,8 @@ static const struct intel_device_info gm45_info = {
 	PLATFORM(INTEL_GM45),
 	.is_mobile = 1,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
-	.display.supports_tv = 1,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0),
+	.display = &gm45_display,
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -387,7 +454,6 @@ static const struct intel_device_info gm45_info = {
 	GEN(5), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.display.has_hotplug = 1, \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), \
 	.has_3d_pipeline = 1, \
 	.has_snoop = true, \
@@ -395,20 +461,33 @@ static const struct intel_device_info gm45_info = {
 	/* ilk does support rc6, but we do not implement [power] contexts */ \
 	.has_rc6 = 0, \
 	.dma_mask_size = 36, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	ILK_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
+#define ILK_DISPLAY \
+	.has_hotplug = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	ILK_COLORS
+
+static const struct intel_display_device_info ilk_d_display = {
+	ILK_DISPLAY,
+};
+
 static const struct intel_device_info ilk_d_info = {
 	GEN5_FEATURES,
 	PLATFORM(INTEL_IRONLAKE),
+	.display = &ilk_d_display,
 };
 
+static const struct intel_display_device_info ilk_m_display = {
+	ILK_DISPLAY,
+ };
+
 static const struct intel_device_info ilk_m_info = {
 	GEN5_FEATURES,
 	PLATFORM(INTEL_IRONLAKE),
+	.display = &ilk_m_display,
 	.is_mobile = 1,
 	.has_rps = true,
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
@@ -418,7 +497,6 @@ static const struct intel_device_info ilk_m_info = {
 	GEN(6), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.display.has_hotplug = 1, \
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_3d_pipeline = 1, \
@@ -431,23 +509,29 @@ static const struct intel_device_info ilk_m_info = {
 	.dma_mask_size = 40, \
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \
 	.__runtime.ppgtt_size = 31, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	ILK_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
+static const struct intel_display_device_info snb_display = {
+	.has_hotplug = 1,
+	I9XX_PIPE_OFFSETS,
+	I9XX_CURSOR_OFFSETS,
+	ILK_COLORS,
+};
+
 #define SNB_D_PLATFORM \
 	GEN6_FEATURES, \
 	PLATFORM(INTEL_SANDYBRIDGE)
 
 static const struct intel_device_info snb_d_gt1_info = {
 	SNB_D_PLATFORM,
+	.display = &snb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info snb_d_gt2_info = {
 	SNB_D_PLATFORM,
+	.display = &snb_display,
 	.gt = 2,
 };
 
@@ -459,11 +543,13 @@ static const struct intel_device_info snb_d_gt2_info = {
 
 static const struct intel_device_info snb_m_gt1_info = {
 	SNB_M_PLATFORM,
+	.display = &snb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info snb_m_gt2_info = {
 	SNB_M_PLATFORM,
+	.display = &snb_display,
 	.gt = 2,
 };
 
@@ -471,7 +557,6 @@ static const struct intel_device_info snb_m_gt2_info = {
 	GEN(7), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), \
-	.display.has_hotplug = 1, \
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_3d_pipeline = 1, \
@@ -484,9 +569,6 @@ static const struct intel_device_info snb_m_gt2_info = {
 	.dma_mask_size = 40, \
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \
 	.__runtime.ppgtt_size = 31, \
-	IVB_PIPE_OFFSETS, \
-	IVB_CURSOR_OFFSETS, \
-	IVB_COLORS, \
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
@@ -495,13 +577,22 @@ static const struct intel_device_info snb_m_gt2_info = {
 	PLATFORM(INTEL_IVYBRIDGE), \
 	.has_l3_dpf = 1
 
+static const struct intel_display_device_info ivb_display = {
+	.has_hotplug = 1,
+	IVB_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+};
+
 static const struct intel_device_info ivb_d_gt1_info = {
 	IVB_D_PLATFORM,
+	.display = &ivb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info ivb_d_gt2_info = {
 	IVB_D_PLATFORM,
+	.display = &ivb_display,
 	.gt = 2,
 };
 
@@ -513,11 +604,13 @@ static const struct intel_device_info ivb_d_gt2_info = {
 
 static const struct intel_device_info ivb_m_gt1_info = {
 	IVB_M_PLATFORM,
+	.display = &ivb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info ivb_m_gt2_info = {
 	IVB_M_PLATFORM,
+	.display = &ivb_display,
 	.gt = 2,
 };
 
@@ -529,28 +622,32 @@ static const struct intel_device_info ivb_q_info = {
 	.has_l3_dpf = 1,
 };
 
+static const struct intel_display_device_info vlv_display = {
+	.has_gmch = 1,
+	.has_hotplug = 1,
+	.mmio_offset = VLV_DISPLAY_BASE,
+	I9XX_PIPE_OFFSETS,
+	I9XX_CURSOR_OFFSETS,
+	I9XX_COLORS,
+};
+
 static const struct intel_device_info vlv_info = {
 	PLATFORM(INTEL_VALLEYVIEW),
 	GEN(7),
 	.is_lp = 1,
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
+	.display = &vlv_display,
 	.has_runtime_pm = 1,
 	.has_rc6 = 1,
 	.has_reset_engine = true,
 	.has_rps = true,
-	.display.has_gmch = 1,
-	.display.has_hotplug = 1,
 	.dma_mask_size = 40,
 	.__runtime.ppgtt_type = INTEL_PPGTT_ALIASING,
 	.__runtime.ppgtt_size = 31,
 	.has_snoop = true,
 	.has_coherent_ggtt = false,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0),
-	.display.mmio_offset = VLV_DISPLAY_BASE,
-	I9XX_PIPE_OFFSETS,
-	I9XX_CURSOR_OFFSETS,
-	I9XX_COLORS,
 	GEN_DEFAULT_PAGE_SIZES,
 	GEN_DEFAULT_REGIONS,
 };
@@ -560,11 +657,7 @@ static const struct intel_device_info vlv_info = {
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), \
-	.display.has_ddi = 1, \
-	.display.has_fpga_dbg = 1, \
-	.display.has_dp_mst = 1, \
 	.has_rc6p = 0 /* RC6p removed-by HSW */, \
-	HSW_PIPE_OFFSETS, \
 	.has_runtime_pm = 1
 
 #define HSW_PLATFORM \
@@ -572,18 +665,31 @@ static const struct intel_device_info vlv_info = {
 	PLATFORM(INTEL_HASWELL), \
 	.has_l3_dpf = 1
 
+static const struct intel_display_device_info hsw_display = {
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+};
+
 static const struct intel_device_info hsw_gt1_info = {
 	HSW_PLATFORM,
+	.display = &hsw_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info hsw_gt2_info = {
 	HSW_PLATFORM,
+	.display = &hsw_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info hsw_gt3_info = {
 	HSW_PLATFORM,
+	.display = &hsw_display,
 	.gt = 3,
 };
 
@@ -596,22 +702,35 @@ static const struct intel_device_info hsw_gt3_info = {
 	.__runtime.ppgtt_size = 48, \
 	.has_64bit_reloc = 1
 
+static const struct intel_display_device_info bdw_display = {
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+};
+
 #define BDW_PLATFORM \
 	GEN8_FEATURES, \
 	PLATFORM(INTEL_BROADWELL)
 
 static const struct intel_device_info bdw_gt1_info = {
 	BDW_PLATFORM,
+	.display = &bdw_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info bdw_gt2_info = {
 	BDW_PLATFORM,
+	.display = &bdw_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info bdw_rsvd_info = {
 	BDW_PLATFORM,
+	.display = &bdw_display,
 	.gt = 3,
 	/* According to the device ID those devices are GT3, they were
 	 * previously treated as not GT3, keep it like that.
@@ -620,17 +739,27 @@ static const struct intel_device_info bdw_rsvd_info = {
 
 static const struct intel_device_info bdw_gt3_info = {
 	BDW_PLATFORM,
+	.display = &bdw_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
 };
 
+static const struct intel_display_device_info chv_display = {
+	.has_hotplug = 1,
+	.has_gmch = 1,
+	.mmio_offset = VLV_DISPLAY_BASE,
+	CHV_PIPE_OFFSETS,
+	CHV_CURSOR_OFFSETS,
+	CHV_COLORS,
+};
+
 static const struct intel_device_info chv_info = {
 	PLATFORM(INTEL_CHERRYVIEW),
 	GEN(8),
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
-	.display.has_hotplug = 1,
+	.display = &chv_display,
 	.is_lp = 1,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0),
 	.has_64bit_reloc = 1,
@@ -638,17 +767,12 @@ static const struct intel_device_info chv_info = {
 	.has_rc6 = 1,
 	.has_rps = true,
 	.has_logical_ring_contexts = 1,
-	.display.has_gmch = 1,
 	.dma_mask_size = 39,
 	.__runtime.ppgtt_type = INTEL_PPGTT_FULL,
 	.__runtime.ppgtt_size = 32,
 	.has_reset_engine = 1,
 	.has_snoop = true,
 	.has_coherent_ggtt = false,
-	.display.mmio_offset = VLV_DISPLAY_BASE,
-	CHV_PIPE_OFFSETS,
-	CHV_CURSOR_OFFSETS,
-	CHV_COLORS,
 	GEN_DEFAULT_PAGE_SIZES,
 	GEN_DEFAULT_REGIONS,
 };
@@ -663,12 +787,22 @@ static const struct intel_device_info chv_info = {
 	GEN9_DEFAULT_PAGE_SIZES, \
 	.__runtime.has_dmc = 1, \
 	.has_gt_uc = 1, \
-	.__runtime.has_hdcp = 1, \
-	.display.has_ipc = 1, \
-	.display.has_psr = 1, \
-	.display.has_psr_hw_tracking = 1, \
-	.display.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */ \
-	.display.dbuf.slice_mask = BIT(DBUF_S1)
+	.__runtime.has_hdcp = 1
+
+static const struct intel_display_device_info skl_display = {
+	.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */
+	.dbuf.slice_mask = BIT(DBUF_S1),
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	.has_ipc = 1,
+	.has_psr = 1,
+	.has_psr_hw_tracking = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+};
 
 #define SKL_PLATFORM \
 	GEN9_FEATURES, \
@@ -676,11 +810,13 @@ static const struct intel_device_info chv_info = {
 
 static const struct intel_device_info skl_gt1_info = {
 	SKL_PLATFORM,
+	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info skl_gt2_info = {
 	SKL_PLATFORM,
+	.display = &skl_display,
 	.gt = 2,
 };
 
@@ -692,19 +828,19 @@ static const struct intel_device_info skl_gt2_info = {
 
 static const struct intel_device_info skl_gt3_info = {
 	SKL_GT3_PLUS_PLATFORM,
+	.display = &skl_display,
 	.gt = 3,
 };
 
 static const struct intel_device_info skl_gt4_info = {
 	SKL_GT3_PLUS_PLATFORM,
+	.display = &skl_display,
 	.gt = 4,
 };
 
 #define GEN9_LP_FEATURES \
 	GEN(9), \
 	.is_lp = 1, \
-	.display.dbuf.slice_mask = BIT(DBUF_S1), \
-	.display.has_hotplug = 1, \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
@@ -712,17 +848,12 @@ static const struct intel_device_info skl_gt4_info = {
 		BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C), \
 	.has_3d_pipeline = 1, \
 	.has_64bit_reloc = 1, \
-	.display.has_ddi = 1, \
-	.display.has_fpga_dbg = 1, \
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
 	.__runtime.has_hdcp = 1, \
-	.display.has_psr = 1, \
-	.display.has_psr_hw_tracking = 1, \
 	.has_runtime_pm = 1, \
 	.__runtime.has_dmc = 1, \
 	.has_rc6 = 1, \
 	.has_rps = true, \
-	.display.has_dp_mst = 1, \
 	.has_logical_ring_contexts = 1, \
 	.has_gt_uc = 1, \
 	.dma_mask_size = 39, \
@@ -731,25 +862,44 @@ static const struct intel_device_info skl_gt4_info = {
 	.has_reset_engine = 1, \
 	.has_snoop = true, \
 	.has_coherent_ggtt = false, \
-	.display.has_ipc = 1, \
-	HSW_PIPE_OFFSETS, \
-	IVB_CURSOR_OFFSETS, \
-	IVB_COLORS, \
 	GEN9_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
+#define GEN9_LP_DISPLAY \
+	.dbuf.slice_mask = BIT(DBUF_S1), \
+	.has_dp_mst = 1, \
+	.has_ddi = 1, \
+	.has_fpga_dbg = 1, \
+	.has_hotplug = 1, \
+	.has_ipc = 1, \
+	.has_psr = 1, \
+	.has_psr_hw_tracking = 1, \
+	HSW_PIPE_OFFSETS, \
+	IVB_CURSOR_OFFSETS, \
+	IVB_COLORS
+
+static const struct intel_display_device_info bxt_display = {
+	GEN9_LP_DISPLAY,
+	.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */
+};
+
 static const struct intel_device_info bxt_info = {
 	GEN9_LP_FEATURES,
 	PLATFORM(INTEL_BROXTON),
-	.display.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */
+	.display = &bxt_display,
+};
+
+static const struct intel_display_device_info glk_display = {
+	GEN9_LP_DISPLAY,
+	.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */
+	GLK_COLORS,
 };
 
 static const struct intel_device_info glk_info = {
 	GEN9_LP_FEATURES,
 	PLATFORM(INTEL_GEMINILAKE),
 	.__runtime.display.ip.ver = 10,
-	.display.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */
-	GLK_COLORS,
+	.display = &glk_display,
 };
 
 #define KBL_PLATFORM \
@@ -758,16 +908,19 @@ static const struct intel_device_info glk_info = {
 
 static const struct intel_device_info kbl_gt1_info = {
 	KBL_PLATFORM,
+	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info kbl_gt2_info = {
 	KBL_PLATFORM,
+	.display = &skl_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info kbl_gt3_info = {
 	KBL_PLATFORM,
+	.display = &skl_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
@@ -779,16 +932,19 @@ static const struct intel_device_info kbl_gt3_info = {
 
 static const struct intel_device_info cfl_gt1_info = {
 	CFL_PLATFORM,
+	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info cfl_gt2_info = {
 	CFL_PLATFORM,
+	.display = &skl_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info cfl_gt3_info = {
 	CFL_PLATFORM,
+	.display = &skl_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
@@ -800,11 +956,13 @@ static const struct intel_device_info cfl_gt3_info = {
 
 static const struct intel_device_info cml_gt1_info = {
 	CML_PLATFORM,
+	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info cml_gt2_info = {
 	CML_PLATFORM,
+	.display = &skl_display,
 	.gt = 2,
 };
 
@@ -816,39 +974,51 @@ static const struct intel_device_info cml_gt2_info = {
 #define GEN11_FEATURES \
 	GEN9_FEATURES, \
 	GEN11_DEFAULT_PAGE_SIZES, \
-	.display.abox_mask = BIT(0), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
 		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
-	.display.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET, \
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-		[TRANSCODER_C] = PIPE_C_OFFSET, \
-		[TRANSCODER_EDP] = PIPE_EDP_OFFSET, \
-		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \
-		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \
-	}, \
-	.display.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
-		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \
-		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \
-		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
-	}, \
 	GEN(11), \
-	ICL_COLORS, \
-	.display.dbuf.size = 2048, \
-	.display.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \
 	.__runtime.has_dsc = 1, \
 	.has_coherent_ggtt = false, \
 	.has_logical_ring_elsq = 1
 
+static const struct intel_display_device_info gen11_display = {
+	.abox_mask = BIT(0),
+	.dbuf.size = 2048,
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2),
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	.has_ipc = 1,
+	.has_psr = 1,
+	.has_psr_hw_tracking = 1,
+	.pipe_offsets = {
+		[TRANSCODER_A] = PIPE_A_OFFSET,
+		[TRANSCODER_B] = PIPE_B_OFFSET,
+		[TRANSCODER_C] = PIPE_C_OFFSET,
+		[TRANSCODER_EDP] = PIPE_EDP_OFFSET,
+		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,
+		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,
+	},
+	.trans_offsets = {
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET,
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET,
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET,
+		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET,
+		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,
+		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,
+	},
+	IVB_CURSOR_OFFSETS,
+	ICL_COLORS,
+};
+
 static const struct intel_device_info icl_info = {
 	GEN11_FEATURES,
 	PLATFORM(INTEL_ICELAKE),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
+	.display = &gen11_display,
 };
 
 static const struct intel_device_info ehl_info = {
@@ -856,6 +1026,7 @@ static const struct intel_device_info ehl_info = {
 	PLATFORM(INTEL_ELKHARTLAKE),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0),
 	.__runtime.ppgtt_size = 36,
+	.display = &gen11_display,
 };
 
 static const struct intel_device_info jsl_info = {
@@ -863,17 +1034,32 @@ static const struct intel_device_info jsl_info = {
 	PLATFORM(INTEL_JASPERLAKE),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0),
 	.__runtime.ppgtt_size = 36,
+	.display = &gen11_display,
 };
 
 #define GEN12_FEATURES \
 	GEN11_FEATURES, \
 	GEN(12), \
-	.display.abox_mask = GENMASK(2, 1), \
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \
 		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
-	.display.pipe_offsets = { \
+	.has_global_mocs = 1, \
+	.has_pxp = 1
+
+#define XE_D_DISPLAY \
+	.abox_mask = GENMASK(2, 1), \
+	.dbuf.size = 2048, \
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \
+	.has_ddi = 1, \
+	.has_dp_mst = 1, \
+	.has_dsb = 1, \
+	.has_fpga_dbg = 1, \
+	.has_hotplug = 1, \
+	.has_ipc = 1, \
+	.has_psr = 1, \
+	.has_psr_hw_tracking = 1, \
+	.pipe_offsets = { \
 		[TRANSCODER_A] = PIPE_A_OFFSET, \
 		[TRANSCODER_B] = PIPE_B_OFFSET, \
 		[TRANSCODER_C] = PIPE_C_OFFSET, \
@@ -881,7 +1067,7 @@ static const struct intel_device_info jsl_info = {
 		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \
 		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \
 	}, \
-	.display.trans_offsets = { \
+	.trans_offsets = { \
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
 		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
@@ -890,28 +1076,36 @@ static const struct intel_device_info jsl_info = {
 		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
 	}, \
 	TGL_CURSOR_OFFSETS, \
-	.has_global_mocs = 1, \
-	.has_pxp = 1, \
-	.display.has_dsb = 1
+	ICL_COLORS
+
+static const struct intel_display_device_info tgl_display = {
+	XE_D_DISPLAY,
+};
 
 static const struct intel_device_info tgl_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_TIGERLAKE),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
+	.display = &tgl_display,
+};
+
+static const struct intel_display_device_info rkl_display = {
+	XE_D_DISPLAY,
+	.abox_mask = BIT(0),
+	.has_hti = 1,
+	.has_psr_hw_tracking = 0,
 };
 
 static const struct intel_device_info rkl_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_ROCKETLAKE),
-	.display.abox_mask = BIT(0),
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 		BIT(TRANSCODER_C),
-	.display.has_hti = 1,
-	.display.has_psr_hw_tracking = 0,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0),
+	.display = &rkl_display,
 };
 
 #define DGFX_FEATURES \
@@ -934,43 +1128,43 @@ static const struct intel_device_info dg1_info = {
 		BIT(VCS0) | BIT(VCS2),
 	/* Wa_16011227922 */
 	.__runtime.ppgtt_size = 47,
+	.display = &tgl_display,
+};
+
+static const struct intel_display_device_info adl_s_display = {
+	XE_D_DISPLAY,
+	.has_hti = 1,
+	.has_psr_hw_tracking = 0,
 };
 
 static const struct intel_device_info adl_s_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_ALDERLAKE_S),
 	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D),
-	.display.has_hti = 1,
-	.display.has_psr_hw_tracking = 0,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.dma_mask_size = 39,
+	.display = &adl_s_display,
 };
 
 #define XE_LPD_FEATURES \
-	.display.abox_mask = GENMASK(1, 0),					\
-	.display.color = {							\
+	.abox_mask = GENMASK(1, 0),						\
+	.color = {								\
 		.degamma_lut_size = 129, .gamma_lut_size = 1024,		\
 		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING |		\
-				     DRM_COLOR_LUT_EQUAL_CHANNELS,		\
+		DRM_COLOR_LUT_EQUAL_CHANNELS,					\
 	},									\
-	.display.dbuf.size = 4096,						\
-	.display.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) |	\
+	.dbuf.size = 4096,							\
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) |		\
 		BIT(DBUF_S4),							\
-	.display.has_ddi = 1,							\
-	.__runtime.has_dmc = 1,							\
-	.display.has_dp_mst = 1,						\
-	.display.has_dsb = 1,							\
-	.__runtime.has_dsc = 1,							\
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),					\
-	.display.has_fpga_dbg = 1,						\
-	.__runtime.has_hdcp = 1,						\
-	.display.has_hotplug = 1,						\
-	.display.has_ipc = 1,							\
-	.display.has_psr = 1,							\
-	.__runtime.display.ip.ver = 13,							\
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D),	\
-	.display.pipe_offsets = {						\
+	.has_ddi = 1,								\
+	.has_dp_mst = 1,							\
+	.has_dsb = 1,								\
+	.has_fpga_dbg = 1,							\
+	.has_hotplug = 1,							\
+	.has_ipc = 1,								\
+	.has_psr = 1,								\
+	.pipe_offsets = {							\
 		[TRANSCODER_A] = PIPE_A_OFFSET,					\
 		[TRANSCODER_B] = PIPE_B_OFFSET,					\
 		[TRANSCODER_C] = PIPE_C_OFFSET,					\
@@ -978,7 +1172,7 @@ static const struct intel_device_info adl_s_info = {
 		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,				\
 		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,				\
 	},									\
-	.display.trans_offsets = {						\
+	.trans_offsets = {						\
 		[TRANSCODER_A] = TRANSCODER_A_OFFSET,				\
 		[TRANSCODER_B] = TRANSCODER_B_OFFSET,				\
 		[TRANSCODER_C] = TRANSCODER_C_OFFSET,				\
@@ -988,18 +1182,31 @@ static const struct intel_device_info adl_s_info = {
 	},									\
 	TGL_CURSOR_OFFSETS
 
+#define XE_LPD_RUNTIME \
+	.__runtime.has_dmc = 1,							\
+	.__runtime.has_dsc = 1,							\
+	.__runtime.fbc_mask = BIT(INTEL_FBC_A),					\
+	.__runtime.has_hdcp = 1,						\
+	.__runtime.display.ip.ver = 13,							\
+	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D)
+
+static const struct intel_display_device_info xe_lpd_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_crawl = 1,
+	.has_psr_hw_tracking = 0,
+};
+
 static const struct intel_device_info adl_p_info = {
 	GEN12_FEATURES,
-	XE_LPD_FEATURES,
+	XE_LPD_RUNTIME,
 	PLATFORM(INTEL_ALDERLAKE_P),
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D) |
 			       BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
-	.display.has_cdclk_crawl = 1,
-	.display.has_psr_hw_tracking = 0,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.__runtime.ppgtt_size = 48,
+	.display = &xe_lpd_display,
 	.dma_mask_size = 39,
 };
 
@@ -1068,18 +1275,23 @@ static const struct intel_device_info xehpsdv_info = {
 	.has_guc_deprivilege = 1, \
 	.has_heci_pxp = 1, \
 	.has_media_ratio_mode = 1, \
-	.display.has_cdclk_squash = 1, \
 	.__runtime.platform_engine_mask = \
 		BIT(RCS0) | BIT(BCS0) | \
 		BIT(VECS0) | BIT(VECS1) | \
 		BIT(VCS0) | BIT(VCS2) | \
 		BIT(CCS0) | BIT(CCS1) | BIT(CCS2) | BIT(CCS3)
 
+static const struct intel_display_device_info xe_hpd_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_squash = 1,
+};
+
 static const struct intel_device_info dg2_info = {
 	DG2_FEATURES,
-	XE_LPD_FEATURES,
+	XE_LPD_RUNTIME,
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
+	.display = &xe_hpd_display,
 };
 
 static const struct intel_device_info ats_m_info = {
@@ -1115,11 +1327,9 @@ static const struct intel_device_info pvc_info = {
 	.require_force_probe = 1,
 };
 
-#define XE_LPDP_FEATURES	\
-	XE_LPD_FEATURES,	\
+#define XE_LPDP_RUNTIME	\
+	XE_LPD_RUNTIME,	\
 	.__runtime.display.ip.ver = 14,	\
-	.display.has_cdclk_crawl = 1, \
-	.display.has_cdclk_squash = 1, \
 	.__runtime.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B)
 
 static const struct intel_gt_definition xelpmp_extra_gt[] = {
@@ -1132,9 +1342,15 @@ static const struct intel_gt_definition xelpmp_extra_gt[] = {
 	{}
 };
 
+static const struct intel_display_device_info xe_lpdp_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_crawl = 1,
+	.has_cdclk_squash = 1,
+};
+
 static const struct intel_device_info mtl_info = {
 	XE_HP_FEATURES,
-	XE_LPDP_FEATURES,
+	XE_LPDP_RUNTIME,
 	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
 			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 	/*
@@ -1145,6 +1361,7 @@ static const struct intel_device_info mtl_info = {
 	.__runtime.graphics.ip.rel = 70,
 	.__runtime.media.ip.ver = 13,
 	PLATFORM(INTEL_METEORLAKE),
+	.display = &xe_lpdp_display,
 	.extra_gt_list = xelpmp_extra_gt,
 	.has_flat_ccs = 0,
 	.has_gmd_id = 1,
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 4e23be2995bf6..e8836b9900589 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -138,7 +138,7 @@ void intel_device_info_print(const struct intel_device_info *info,
 
 	drm_printf(p, "has_pooled_eu: %s\n", str_yes_no(runtime->has_pooled_eu));
 
-#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->display.name))
+#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->display->name))
 	DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG);
 #undef PRINT_FLAG
 
@@ -388,6 +388,8 @@ mkwrite_device_info(struct drm_i915_private *i915)
 	return (struct intel_device_info *)INTEL_INFO(i915);
 }
 
+static const struct intel_display_device_info no_display = {};
+
 /**
  * intel_device_info_runtime_init - initialize runtime info
  * @dev_priv: the i915 device
@@ -538,7 +540,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv)) {
 		dev_priv->drm.driver_features &= ~(DRIVER_MODESET |
 						   DRIVER_ATOMIC);
-		memset(&info->display, 0, sizeof(info->display));
+		info->display = &no_display;
 
 		runtime->cpu_transcoder_mask = 0;
 		memset(runtime->num_sprites, 0, sizeof(runtime->num_sprites));
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index c14bc3f5d0fa1..f838ae7191644 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -257,7 +257,7 @@ struct intel_device_info {
 	DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG);
 #undef DEFINE_FLAG
 
-	struct intel_display_device_info display;
+	const struct intel_display_device_info *display;
 
 	/*
 	 * Initial runtime info. Do not access outside of i915_driver_create().
-- 
GitLab


From 0020582a8afe9a8570f80ec503c59bf049a616de Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:31 -0700
Subject: [PATCH 0642/1791] drm/docs: Fix usage stats typos

Fix a couple missing ':'s.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-2-robdclark@gmail.com
---
 Documentation/gpu/drm-usage-stats.rst | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst
index b46327356e804..72d069e5dacbc 100644
--- a/Documentation/gpu/drm-usage-stats.rst
+++ b/Documentation/gpu/drm-usage-stats.rst
@@ -105,7 +105,7 @@ object belong to this client, in the respective memory region.
 Default unit shall be bytes with optional unit specifiers of 'KiB' or 'MiB'
 indicating kibi- or mebi-bytes.
 
-- drm-cycles-<str> <uint>
+- drm-cycles-<str>: <uint>
 
 Engine identifier string must be the same as the one specified in the
 drm-engine-<str> tag and shall contain the number of busy cycles for the given
@@ -117,7 +117,7 @@ larger value within a reasonable period. Upon observing a value lower than what
 was previously read, userspace is expected to stay with that larger previous
 value until a monotonic update is seen.
 
-- drm-maxfreq-<str> <uint> [Hz|MHz|KHz]
+- drm-maxfreq-<str>: <uint> [Hz|MHz|KHz]
 
 Engine identifier string must be the same as the one specified in the
 drm-engine-<str> tag and shall contain the maximum frequency for the given
-- 
GitLab


From 3f09a0cd4ea3b9d34495450d686227d48e7ec648 Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:32 -0700
Subject: [PATCH 0643/1791] drm: Add common fdinfo helper

Handle a bit of the boiler-plate in a single case, and make it easier to
add some core tracked stats.  This also ensures consistent behavior
across drivers for standardised fields.

v2: Update drm-usage-stats.rst, 64b client-id, rename drm_show_fdinfo
v3: Rebase on drm-misc-next

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-3-robdclark@gmail.com
---
 Documentation/gpu/drm-usage-stats.rst | 10 +++++++-
 drivers/gpu/drm/drm_file.c            | 35 +++++++++++++++++++++++++++
 include/drm/drm_drv.h                 |  7 ++++++
 include/drm/drm_file.h                |  4 +++
 4 files changed, 55 insertions(+), 1 deletion(-)

diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst
index 72d069e5dacbc..552195fb1ea39 100644
--- a/Documentation/gpu/drm-usage-stats.rst
+++ b/Documentation/gpu/drm-usage-stats.rst
@@ -126,7 +126,15 @@ percentage utilization of the engine, whereas drm-engine-<str> only reflects
 time active without considering what frequency the engine is operating as a
 percentage of it's maximum frequency.
 
+Implementation Details
+======================
+
+Drivers should use drm_show_fdinfo() in their `struct file_operations`, and
+implement &drm_driver.show_fdinfo if they wish to provide any stats which
+are not provided by drm_show_fdinfo().  But even driver specific stats should
+be documented above and where possible, aligned with other drivers.
+
 Driver specific implementations
-===============================
+-------------------------------
 
 :ref:`i915-usage-stats`
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index c1018c4700479..37b4f76a51911 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -148,6 +148,7 @@ bool drm_dev_needs_global_mutex(struct drm_device *dev)
  */
 struct drm_file *drm_file_alloc(struct drm_minor *minor)
 {
+	static atomic64_t ident = ATOMIC_INIT(0);
 	struct drm_device *dev = minor->dev;
 	struct drm_file *file;
 	int ret;
@@ -156,6 +157,8 @@ struct drm_file *drm_file_alloc(struct drm_minor *minor)
 	if (!file)
 		return ERR_PTR(-ENOMEM);
 
+	/* Get a unique identifier for fdinfo: */
+	file->client_id = atomic64_inc_return(&ident);
 	file->pid = get_pid(task_tgid(current));
 	file->minor = minor;
 
@@ -868,6 +871,38 @@ void drm_send_event(struct drm_device *dev, struct drm_pending_event *e)
 }
 EXPORT_SYMBOL(drm_send_event);
 
+/**
+ * drm_show_fdinfo - helper for drm file fops
+ * @seq_file: output stream
+ * @f: the device file instance
+ *
+ * Helper to implement fdinfo, for userspace to query usage stats, etc, of a
+ * process using the GPU.  See also &drm_driver.show_fdinfo.
+ *
+ * For text output format description please see Documentation/gpu/drm-usage-stats.rst
+ */
+void drm_show_fdinfo(struct seq_file *m, struct file *f)
+{
+	struct drm_file *file = f->private_data;
+	struct drm_device *dev = file->minor->dev;
+	struct drm_printer p = drm_seq_file_printer(m);
+
+	drm_printf(&p, "drm-driver:\t%s\n", dev->driver->name);
+	drm_printf(&p, "drm-client-id:\t%llu\n", file->client_id);
+
+	if (dev_is_pci(dev->dev)) {
+		struct pci_dev *pdev = to_pci_dev(dev->dev);
+
+		drm_printf(&p, "drm-pdev:\t%04x:%02x:%02x.%d\n",
+			   pci_domain_nr(pdev->bus), pdev->bus->number,
+			   PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn));
+	}
+
+	if (dev->driver->show_fdinfo)
+		dev->driver->show_fdinfo(&p, file);
+}
+EXPORT_SYMBOL(drm_show_fdinfo);
+
 /**
  * mock_drm_getfile - Create a new struct file for the drm device
  * @minor: drm minor to wrap (e.g. #drm_device.primary)
diff --git a/include/drm/drm_drv.h b/include/drm/drm_drv.h
index b419c59c4befb..89e2706cac561 100644
--- a/include/drm/drm_drv.h
+++ b/include/drm/drm_drv.h
@@ -401,6 +401,13 @@ struct drm_driver {
 			       struct drm_device *dev, uint32_t handle,
 			       uint64_t *offset);
 
+	/**
+	 * @show_fdinfo:
+	 *
+	 * Print device specific fdinfo.  See Documentation/gpu/drm-usage-stats.rst.
+	 */
+	void (*show_fdinfo)(struct drm_printer *p, struct drm_file *f);
+
 	/** @major: driver major number */
 	int major;
 	/** @minor: driver minor number */
diff --git a/include/drm/drm_file.h b/include/drm/drm_file.h
index ecffe24e2b1b0..7d9b3c65cbc10 100644
--- a/include/drm/drm_file.h
+++ b/include/drm/drm_file.h
@@ -258,6 +258,9 @@ struct drm_file {
 	/** @pid: Process that opened this file. */
 	struct pid *pid;
 
+	/** @client_id: A unique id for fdinfo */
+	u64 client_id;
+
 	/** @magic: Authentication magic, see @authenticated. */
 	drm_magic_t magic;
 
@@ -438,6 +441,7 @@ void drm_send_event(struct drm_device *dev, struct drm_pending_event *e);
 void drm_send_event_timestamp_locked(struct drm_device *dev,
 				     struct drm_pending_event *e,
 				     ktime_t timestamp);
+void drm_show_fdinfo(struct seq_file *m, struct file *f);
 
 struct file *mock_drm_getfile(struct drm_minor *minor, unsigned int flags);
 
-- 
GitLab


From 51d86ee5e07ccef85af04ee9850b0baa107999b6 Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:33 -0700
Subject: [PATCH 0644/1791] drm/msm: Switch to fdinfo helper

Now that we have a common helper, use it.

v2: Rebase on drm-misc-next

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-4-robdclark@gmail.com
---
 drivers/gpu/drm/msm/msm_drv.c | 11 +++++------
 drivers/gpu/drm/msm/msm_gpu.c |  2 --
 2 files changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 060c7689a739b..02fd6093f9b0b 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -1057,23 +1057,21 @@ static const struct drm_ioctl_desc msm_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_QUERY, msm_ioctl_submitqueue_query, DRM_RENDER_ALLOW),
 };
 
-static void msm_fop_show_fdinfo(struct seq_file *m, struct file *f)
+static void msm_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 {
-	struct drm_file *file = f->private_data;
 	struct drm_device *dev = file->minor->dev;
 	struct msm_drm_private *priv = dev->dev_private;
-	struct drm_printer p = drm_seq_file_printer(m);
 
 	if (!priv->gpu)
 		return;
 
-	msm_gpu_show_fdinfo(priv->gpu, file->driver_priv, &p);
+	msm_gpu_show_fdinfo(priv->gpu, file->driver_priv, p);
 }
 
 static const struct file_operations fops = {
 	.owner = THIS_MODULE,
 	DRM_GEM_FOPS,
-	.show_fdinfo = msm_fop_show_fdinfo,
+	.show_fdinfo = drm_show_fdinfo,
 };
 
 static const struct drm_driver msm_driver = {
@@ -1083,7 +1081,7 @@ static const struct drm_driver msm_driver = {
 				DRIVER_MODESET |
 				DRIVER_SYNCOBJ,
 	.open               = msm_open,
-	.postclose           = msm_postclose,
+	.postclose          = msm_postclose,
 	.dumb_create        = msm_gem_dumb_create,
 	.dumb_map_offset    = msm_gem_dumb_map_offset,
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
@@ -1093,6 +1091,7 @@ static const struct drm_driver msm_driver = {
 #ifdef CONFIG_DEBUG_FS
 	.debugfs_init       = msm_debugfs_init,
 #endif
+	.show_fdinfo        = msm_show_fdinfo,
 	.ioctls             = msm_ioctls,
 	.num_ioctls         = ARRAY_SIZE(msm_ioctls),
 	.fops               = &fops,
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 26ebda40be4f0..c403912d13ab7 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -151,8 +151,6 @@ int msm_gpu_pm_suspend(struct msm_gpu *gpu)
 void msm_gpu_show_fdinfo(struct msm_gpu *gpu, struct msm_file_private *ctx,
 			 struct drm_printer *p)
 {
-	drm_printf(p, "drm-driver:\t%s\n", gpu->dev->driver->name);
-	drm_printf(p, "drm-client-id:\t%u\n", ctx->seqno);
 	drm_printf(p, "drm-engine-gpu:\t%llu ns\n", ctx->elapsed_ns);
 	drm_printf(p, "drm-cycles-gpu:\t%llu\n", ctx->cycles);
 	drm_printf(p, "drm-maxfreq-gpu:\t%u Hz\n", gpu->fast_rate);
-- 
GitLab


From 376c25f8ca47084c4f0aff0f14684780756ccef4 Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:34 -0700
Subject: [PATCH 0645/1791] drm/amdgpu: Switch to fdinfo helper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: Rebase on drm-misc-next

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-5-robdclark@gmail.com
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c    |  3 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c | 32 ++++++++++------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.h |  2 +-
 3 files changed, 18 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b1ca1ab6d6ad3..1b46e7ac7cb0e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2747,7 +2747,7 @@ static const struct file_operations amdgpu_driver_kms_fops = {
 	.compat_ioctl = amdgpu_kms_compat_ioctl,
 #endif
 #ifdef CONFIG_PROC_FS
-	.show_fdinfo = amdgpu_show_fdinfo
+	.show_fdinfo = drm_show_fdinfo,
 #endif
 };
 
@@ -2802,6 +2802,7 @@ static const struct drm_driver amdgpu_kms_driver = {
 	.dumb_map_offset = amdgpu_mode_dumb_mmap,
 	.fops = &amdgpu_driver_kms_fops,
 	.release = &amdgpu_driver_release_kms,
+	.show_fdinfo = amdgpu_show_fdinfo,
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
index c57252f004e8e..13d7413d4ca3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
@@ -53,9 +53,8 @@ static const char *amdgpu_ip_name[AMDGPU_HW_IP_NUM] = {
 	[AMDGPU_HW_IP_VCN_JPEG]	=	"jpeg",
 };
 
-void amdgpu_show_fdinfo(struct seq_file *m, struct file *f)
+void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 {
-	struct drm_file *file = f->private_data;
 	struct amdgpu_device *adev = drm_to_adev(file->minor->dev);
 	struct amdgpu_fpriv *fpriv = file->driver_priv;
 	struct amdgpu_vm *vm = &fpriv->vm;
@@ -87,31 +86,30 @@ void amdgpu_show_fdinfo(struct seq_file *m, struct file *f)
 	 * ******************************************************************
 	 */
 
-	seq_printf(m, "pasid:\t%u\n", fpriv->vm.pasid);
-	seq_printf(m, "drm-driver:\t%s\n", file->minor->dev->driver->name);
-	seq_printf(m, "drm-pdev:\t%04x:%02x:%02x.%d\n", domain, bus, dev, fn);
-	seq_printf(m, "drm-client-id:\t%Lu\n", vm->immediate.fence_context);
-	seq_printf(m, "drm-memory-vram:\t%llu KiB\n", stats.vram/1024UL);
-	seq_printf(m, "drm-memory-gtt: \t%llu KiB\n", stats.gtt/1024UL);
-	seq_printf(m, "drm-memory-cpu: \t%llu KiB\n", stats.cpu/1024UL);
-	seq_printf(m, "amd-memory-visible-vram:\t%llu KiB\n",
+	drm_printf(p, "pasid:\t%u\n", fpriv->vm.pasid);
+	drm_printf(p, "drm-driver:\t%s\n", file->minor->dev->driver->name);
+	drm_printf(p, "drm-pdev:\t%04x:%02x:%02x.%d\n", domain, bus, dev, fn);
+	drm_printf(p, "drm-client-id:\t%Lu\n", vm->immediate.fence_context);
+	drm_printf(p, "drm-memory-vram:\t%llu KiB\n", stats.vram/1024UL);
+	drm_printf(p, "drm-memory-gtt: \t%llu KiB\n", stats.gtt/1024UL);
+	drm_printf(p, "drm-memory-cpu: \t%llu KiB\n", stats.cpu/1024UL);
+	drm_printf(p, "amd-memory-visible-vram:\t%llu KiB\n",
 		   stats.visible_vram/1024UL);
-	seq_printf(m, "amd-evicted-vram:\t%llu KiB\n",
+	drm_printf(p, "amd-evicted-vram:\t%llu KiB\n",
 		   stats.evicted_vram/1024UL);
-	seq_printf(m, "amd-evicted-visible-vram:\t%llu KiB\n",
+	drm_printf(p, "amd-evicted-visible-vram:\t%llu KiB\n",
 		   stats.evicted_visible_vram/1024UL);
-	seq_printf(m, "amd-requested-vram:\t%llu KiB\n",
+	drm_printf(p, "amd-requested-vram:\t%llu KiB\n",
 		   stats.requested_vram/1024UL);
-	seq_printf(m, "amd-requested-visible-vram:\t%llu KiB\n",
+	drm_printf(p, "amd-requested-visible-vram:\t%llu KiB\n",
 		   stats.requested_visible_vram/1024UL);
-	seq_printf(m, "amd-requested-gtt:\t%llu KiB\n",
+	drm_printf(p, "amd-requested-gtt:\t%llu KiB\n",
 		   stats.requested_gtt/1024UL);
-
 	for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) {
 		if (!usage[hw_ip])
 			continue;
 
-		seq_printf(m, "drm-engine-%s:\t%Ld ns\n", amdgpu_ip_name[hw_ip],
+		drm_printf(p, "drm-engine-%s:\t%Ld ns\n", amdgpu_ip_name[hw_ip],
 			   ktime_to_ns(usage[hw_ip]));
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.h
index e86834bfea1d5..0398f5a159ef9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.h
@@ -37,6 +37,6 @@
 #include "amdgpu_ids.h"
 
 uint32_t amdgpu_get_ip_count(struct amdgpu_device *adev, int id);
-void amdgpu_show_fdinfo(struct seq_file *m, struct file *f);
+void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file);
 
 #endif
-- 
GitLab


From 686b21b5f6ca2f8a716f9a4ade07246dbfb2713e Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:35 -0700
Subject: [PATCH 0646/1791] drm: Add fdinfo memory stats

Add support to dump GEM stats to fdinfo.

v2: Fix typos, change size units to match docs, use div_u64
v3: Do it in core
v4: more kerneldoc
v5: doc fixes
v6: Actually use u64, bit more comment docs

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-6-robdclark@gmail.com
---
 Documentation/gpu/drm-usage-stats.rst | 54 +++++++++++----
 drivers/gpu/drm/drm_file.c            | 99 ++++++++++++++++++++++++++-
 include/drm/drm_file.h                | 28 ++++++++
 include/drm/drm_gem.h                 | 32 +++++++++
 4 files changed, 200 insertions(+), 13 deletions(-)

diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst
index 552195fb1ea39..d012eb56885e6 100644
--- a/Documentation/gpu/drm-usage-stats.rst
+++ b/Documentation/gpu/drm-usage-stats.rst
@@ -52,6 +52,9 @@ String shall contain the name this driver registered as via the respective
 Optional fully standardised keys
 --------------------------------
 
+Identification
+^^^^^^^^^^^^^^
+
 - drm-pdev: <aaaa:bb.cc.d>
 
 For PCI devices this should contain the PCI slot address of the device in
@@ -69,6 +72,9 @@ scope of each device, in which case `drm-pdev` shall be present as well.
 Userspace should make sure to not double account any usage statistics by using
 the above described criteria in order to associate data to individual clients.
 
+Utilization
+^^^^^^^^^^^
+
 - drm-engine-<str>: <uint> ns
 
 GPUs usually contain multiple execution engines. Each shall be given a stable
@@ -93,18 +99,6 @@ exported engine corresponds to a group of identical hardware engines.
 In the absence of this tag parser shall assume capacity of one. Zero capacity
 is not allowed.
 
-- drm-memory-<str>: <uint> [KiB|MiB]
-
-Each possible memory type which can be used to store buffer objects by the
-GPU in question shall be given a stable and unique name to be returned as the
-string here.
-
-Value shall reflect the amount of storage currently consumed by the buffer
-object belong to this client, in the respective memory region.
-
-Default unit shall be bytes with optional unit specifiers of 'KiB' or 'MiB'
-indicating kibi- or mebi-bytes.
-
 - drm-cycles-<str>: <uint>
 
 Engine identifier string must be the same as the one specified in the
@@ -126,6 +120,42 @@ percentage utilization of the engine, whereas drm-engine-<str> only reflects
 time active without considering what frequency the engine is operating as a
 percentage of it's maximum frequency.
 
+Memory
+^^^^^^
+
+- drm-memory-<region>: <uint> [KiB|MiB]
+
+Each possible memory type which can be used to store buffer objects by the
+GPU in question shall be given a stable and unique name to be returned as the
+string here.  The name "memory" is reserved to refer to normal system memory.
+
+Value shall reflect the amount of storage currently consumed by the buffer
+objects belong to this client, in the respective memory region.
+
+Default unit shall be bytes with optional unit specifiers of 'KiB' or 'MiB'
+indicating kibi- or mebi-bytes.
+
+- drm-shared-<region>: <uint> [KiB|MiB]
+
+The total size of buffers that are shared with another file (ie. have more
+than a single handle).
+
+- drm-total-<region>: <uint> [KiB|MiB]
+
+The total size of buffers that including shared and private memory.
+
+- drm-resident-<region>: <uint> [KiB|MiB]
+
+The total size of buffers that are resident in the specified region.
+
+- drm-purgeable-<region>: <uint> [KiB|MiB]
+
+The total size of buffers that are purgeable.
+
+- drm-active-<region>: <uint> [KiB|MiB]
+
+The total size of buffers that are active on one or more engines.
+
 Implementation Details
 ======================
 
diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c
index 37b4f76a51911..883d83bc0e3d5 100644
--- a/drivers/gpu/drm/drm_file.c
+++ b/drivers/gpu/drm/drm_file.c
@@ -42,6 +42,7 @@
 #include <drm/drm_client.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_file.h>
+#include <drm/drm_gem.h>
 #include <drm/drm_print.h>
 
 #include "drm_crtc_internal.h"
@@ -871,9 +872,105 @@ void drm_send_event(struct drm_device *dev, struct drm_pending_event *e)
 }
 EXPORT_SYMBOL(drm_send_event);
 
+static void print_size(struct drm_printer *p, const char *stat,
+		       const char *region, u64 sz)
+{
+	const char *units[] = {"", " KiB", " MiB"};
+	unsigned u;
+
+	for (u = 0; u < ARRAY_SIZE(units) - 1; u++) {
+		if (sz < SZ_1K)
+			break;
+		sz = div_u64(sz, SZ_1K);
+	}
+
+	drm_printf(p, "drm-%s-%s:\t%llu%s\n", stat, region, sz, units[u]);
+}
+
+/**
+ * drm_print_memory_stats - A helper to print memory stats
+ * @p: The printer to print output to
+ * @stats: The collected memory stats
+ * @supported_status: Bitmask of optional stats which are available
+ * @region: The memory region
+ *
+ */
+void drm_print_memory_stats(struct drm_printer *p,
+			    const struct drm_memory_stats *stats,
+			    enum drm_gem_object_status supported_status,
+			    const char *region)
+{
+	print_size(p, "total", region, stats->private + stats->shared);
+	print_size(p, "shared", region, stats->shared);
+	print_size(p, "active", region, stats->active);
+
+	if (supported_status & DRM_GEM_OBJECT_RESIDENT)
+		print_size(p, "resident", region, stats->resident);
+
+	if (supported_status & DRM_GEM_OBJECT_PURGEABLE)
+		print_size(p, "purgeable", region, stats->purgeable);
+}
+EXPORT_SYMBOL(drm_print_memory_stats);
+
+/**
+ * drm_show_memory_stats - Helper to collect and show standard fdinfo memory stats
+ * @p: the printer to print output to
+ * @file: the DRM file
+ *
+ * Helper to iterate over GEM objects with a handle allocated in the specified
+ * file.
+ */
+void drm_show_memory_stats(struct drm_printer *p, struct drm_file *file)
+{
+	struct drm_gem_object *obj;
+	struct drm_memory_stats status = {};
+	enum drm_gem_object_status supported_status;
+	int id;
+
+	spin_lock(&file->table_lock);
+	idr_for_each_entry (&file->object_idr, obj, id) {
+		enum drm_gem_object_status s = 0;
+
+		if (obj->funcs && obj->funcs->status) {
+			s = obj->funcs->status(obj);
+			supported_status = DRM_GEM_OBJECT_RESIDENT |
+					DRM_GEM_OBJECT_PURGEABLE;
+		}
+
+		if (obj->handle_count > 1) {
+			status.shared += obj->size;
+		} else {
+			status.private += obj->size;
+		}
+
+		if (s & DRM_GEM_OBJECT_RESIDENT) {
+			status.resident += obj->size;
+		} else {
+			/* If already purged or not yet backed by pages, don't
+			 * count it as purgeable:
+			 */
+			s &= ~DRM_GEM_OBJECT_PURGEABLE;
+		}
+
+		if (!dma_resv_test_signaled(obj->resv, dma_resv_usage_rw(true))) {
+			status.active += obj->size;
+
+			/* If still active, don't count as purgeable: */
+			s &= ~DRM_GEM_OBJECT_PURGEABLE;
+		}
+
+		if (s & DRM_GEM_OBJECT_PURGEABLE)
+			status.purgeable += obj->size;
+	}
+	spin_unlock(&file->table_lock);
+
+	drm_print_memory_stats(p, &status, supported_status, "memory");
+}
+EXPORT_SYMBOL(drm_show_memory_stats);
+
 /**
  * drm_show_fdinfo - helper for drm file fops
- * @seq_file: output stream
+ * @m: output stream
  * @f: the device file instance
  *
  * Helper to implement fdinfo, for userspace to query usage stats, etc, of a
diff --git a/include/drm/drm_file.h b/include/drm/drm_file.h
index 7d9b3c65cbc10..966912053cb0f 100644
--- a/include/drm/drm_file.h
+++ b/include/drm/drm_file.h
@@ -41,6 +41,7 @@
 struct dma_fence;
 struct drm_file;
 struct drm_device;
+struct drm_printer;
 struct device;
 struct file;
 
@@ -441,6 +442,33 @@ void drm_send_event(struct drm_device *dev, struct drm_pending_event *e);
 void drm_send_event_timestamp_locked(struct drm_device *dev,
 				     struct drm_pending_event *e,
 				     ktime_t timestamp);
+
+/**
+ * struct drm_memory_stats - GEM object stats associated
+ * @shared: Total size of GEM objects shared between processes
+ * @private: Total size of GEM objects
+ * @resident: Total size of GEM objects backing pages
+ * @purgeable: Total size of GEM objects that can be purged (resident and not active)
+ * @active: Total size of GEM objects active on one or more engines
+ *
+ * Used by drm_print_memory_stats()
+ */
+struct drm_memory_stats {
+	u64 shared;
+	u64 private;
+	u64 resident;
+	u64 purgeable;
+	u64 active;
+};
+
+enum drm_gem_object_status;
+
+void drm_print_memory_stats(struct drm_printer *p,
+			    const struct drm_memory_stats *stats,
+			    enum drm_gem_object_status supported_status,
+			    const char *region);
+
+void drm_show_memory_stats(struct drm_printer *p, struct drm_file *file);
 void drm_show_fdinfo(struct seq_file *m, struct file *f);
 
 struct file *mock_drm_getfile(struct drm_minor *minor, unsigned int flags);
diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
index b8efd836edefe..bbc721870c13c 100644
--- a/include/drm/drm_gem.h
+++ b/include/drm/drm_gem.h
@@ -42,6 +42,25 @@
 struct iosys_map;
 struct drm_gem_object;
 
+/**
+ * enum drm_gem_object_status - bitmask of object state for fdinfo reporting
+ * @DRM_GEM_OBJECT_RESIDENT: object is resident in memory (ie. not unpinned)
+ * @DRM_GEM_OBJECT_PURGEABLE: object marked as purgeable by userspace
+ *
+ * Bitmask of status used for fdinfo memory stats, see &drm_gem_object_funcs.status
+ * and drm_show_fdinfo().  Note that an object can DRM_GEM_OBJECT_PURGEABLE if
+ * it still active or not resident, in which case drm_show_fdinfo() will not
+ * account for it as purgeable.  So drivers do not need to check if the buffer
+ * is idle and resident to return this bit.  (Ie. userspace can mark a buffer
+ * as purgeable even while it is still busy on the GPU.. it does not _actually_
+ * become puregeable until it becomes idle.  The status gem object func does
+ * not need to consider this.)
+ */
+enum drm_gem_object_status {
+	DRM_GEM_OBJECT_RESIDENT  = BIT(0),
+	DRM_GEM_OBJECT_PURGEABLE = BIT(1),
+};
+
 /**
  * struct drm_gem_object_funcs - GEM object functions
  */
@@ -174,6 +193,19 @@ struct drm_gem_object_funcs {
 	 */
 	int (*evict)(struct drm_gem_object *obj);
 
+	/**
+	 * @status:
+	 *
+	 * The optional status callback can return additional object state
+	 * which determines which stats the object is counted against.  The
+	 * callback is called under table_lock.  Racing against object status
+	 * change is "harmless", and the callback can expect to not race
+	 * against object destruction.
+	 *
+	 * Called by drm_show_memory_stats().
+	 */
+	enum drm_gem_object_status (*status)(struct drm_gem_object *obj);
+
 	/**
 	 * @vm_ops:
 	 *
-- 
GitLab


From 3e9757f5ddb98238226ad68a1609aa313de35adb Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:36 -0700
Subject: [PATCH 0647/1791] drm/msm: Add memory stats to fdinfo

Use the new helper to export stats about memory usage.

v2: Drop unintended hunk
v3: Rebase

Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-7-robdclark@gmail.com
---
 drivers/gpu/drm/msm/msm_drv.c |  2 ++
 drivers/gpu/drm/msm/msm_gem.c | 15 +++++++++++++++
 2 files changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 02fd6093f9b0b..58264ff2c4b19 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -1066,6 +1066,8 @@ static void msm_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 		return;
 
 	msm_gpu_show_fdinfo(priv->gpu, file->driver_priv, p);
+
+	drm_show_memory_stats(p, file);
 }
 
 static const struct file_operations fops = {
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index db6c4e281d75d..c32264234ea16 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -1096,6 +1096,20 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 	return ret;
 }
 
+static enum drm_gem_object_status msm_gem_status(struct drm_gem_object *obj)
+{
+	struct msm_gem_object *msm_obj = to_msm_bo(obj);
+	enum drm_gem_object_status status = 0;
+
+	if (msm_obj->pages)
+		status |= DRM_GEM_OBJECT_RESIDENT;
+
+	if (msm_obj->madv == MSM_MADV_DONTNEED)
+		status |= DRM_GEM_OBJECT_PURGEABLE;
+
+	return status;
+}
+
 static const struct vm_operations_struct vm_ops = {
 	.fault = msm_gem_fault,
 	.open = drm_gem_vm_open,
@@ -1110,6 +1124,7 @@ static const struct drm_gem_object_funcs msm_gem_object_funcs = {
 	.vmap = msm_gem_prime_vmap,
 	.vunmap = msm_gem_prime_vunmap,
 	.mmap = msm_gem_object_mmap,
+	.status = msm_gem_status,
 	.vm_ops = &vm_ops,
 };
 
-- 
GitLab


From 90d63a150b85fd1debb9c01237fb78faee02746a Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Wed, 24 May 2023 08:59:37 -0700
Subject: [PATCH 0648/1791] drm/doc: Relax fdinfo string constraints

The restriction about no whitespace, etc, really only applies to the
usage of strings in keys.  Values can contain anything (other than
newline).

Signed-off-by: Rob Clark <robdclark@chromium.org>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524155956.382440-8-robdclark@gmail.com
---
 Documentation/gpu/drm-usage-stats.rst | 27 ++++++++++++++-------------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst
index d012eb56885e6..fe35a291ff3e0 100644
--- a/Documentation/gpu/drm-usage-stats.rst
+++ b/Documentation/gpu/drm-usage-stats.rst
@@ -24,7 +24,7 @@ File format specification
 - All keys shall be prefixed with `drm-`.
 - Whitespace between the delimiter and first non-whitespace character shall be
   ignored when parsing.
-- Neither keys or values are allowed to contain whitespace characters.
+- Keys are not allowed to contain whitespace characters.
 - Numerical key value pairs can end with optional unit string.
 - Data type of the value is fixed as defined in the specification.
 
@@ -39,12 +39,13 @@ Data types
 ----------
 
 - <uint> - Unsigned integer without defining the maximum value.
-- <str> - String excluding any above defined reserved characters or whitespace.
+- <keystr> - String excluding any above defined reserved characters or whitespace.
+- <valstr> - String.
 
 Mandatory fully standardised keys
 ---------------------------------
 
-- drm-driver: <str>
+- drm-driver: <valstr>
 
 String shall contain the name this driver registered as via the respective
 `struct drm_driver` data structure.
@@ -75,10 +76,10 @@ the above described criteria in order to associate data to individual clients.
 Utilization
 ^^^^^^^^^^^
 
-- drm-engine-<str>: <uint> ns
+- drm-engine-<keystr>: <uint> ns
 
 GPUs usually contain multiple execution engines. Each shall be given a stable
-and unique name (str), with possible values documented in the driver specific
+and unique name (keystr), with possible values documented in the driver specific
 documentation.
 
 Value shall be in specified time units which the respective GPU engine spent
@@ -90,19 +91,19 @@ larger value within a reasonable period. Upon observing a value lower than what
 was previously read, userspace is expected to stay with that larger previous
 value until a monotonic update is seen.
 
-- drm-engine-capacity-<str>: <uint>
+- drm-engine-capacity-<keystr>: <uint>
 
 Engine identifier string must be the same as the one specified in the
-drm-engine-<str> tag and shall contain a greater than zero number in case the
+drm-engine-<keystr> tag and shall contain a greater than zero number in case the
 exported engine corresponds to a group of identical hardware engines.
 
 In the absence of this tag parser shall assume capacity of one. Zero capacity
 is not allowed.
 
-- drm-cycles-<str>: <uint>
+- drm-cycles-<keystr>: <uint>
 
 Engine identifier string must be the same as the one specified in the
-drm-engine-<str> tag and shall contain the number of busy cycles for the given
+drm-engine-<keystr> tag and shall contain the number of busy cycles for the given
 engine.
 
 Values are not required to be constantly monotonic if it makes the driver
@@ -111,12 +112,12 @@ larger value within a reasonable period. Upon observing a value lower than what
 was previously read, userspace is expected to stay with that larger previous
 value until a monotonic update is seen.
 
-- drm-maxfreq-<str>: <uint> [Hz|MHz|KHz]
+- drm-maxfreq-<keystr>: <uint> [Hz|MHz|KHz]
 
 Engine identifier string must be the same as the one specified in the
-drm-engine-<str> tag and shall contain the maximum frequency for the given
-engine.  Taken together with drm-cycles-<str>, this can be used to calculate
-percentage utilization of the engine, whereas drm-engine-<str> only reflects
+drm-engine-<keystr> tag and shall contain the maximum frequency for the given
+engine.  Taken together with drm-cycles-<keystr>, this can be used to calculate
+percentage utilization of the engine, whereas drm-engine-<keystr> only reflects
 time active without considering what frequency the engine is operating as a
 percentage of it's maximum frequency.
 
-- 
GitLab


From 18e0deeed8c80d8ec0adfff3312252028739ce52 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:06 -0700
Subject: [PATCH 0649/1791] drm/i915/display: Move display runtime info to
 display structure

Move the runtime info specific to display into display-specific
structures as has already been done with the constant display info.

v2:
 - Rename __runtime to __runtime_defaults for more clarity on the
   purpose.  (Andrzej)
 - Move introduction of DISPLAY_INFO() to previous patch.  (Andrzej)
 - Drop NO_DISPLAY macro.  (Andrzej)
v3:
 - Use "{}" instead of "{ 0 }" for empty struct init.  (Jani)

Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-4-matthew.d.roper@intel.com
---
 drivers/gpu/drm/i915/display/intel_crtc.c     |   2 +-
 drivers/gpu/drm/i915/display/intel_cursor.c   |   2 +-
 drivers/gpu/drm/i915/display/intel_display.c  |   2 +-
 drivers/gpu/drm/i915/display/intel_display.h  |   8 +-
 .../drm/i915/display/intel_display_device.h   |  23 ++
 drivers/gpu/drm/i915/display/intel_fbc.c      |   6 +-
 drivers/gpu/drm/i915/display/intel_hdcp.c     |   2 +-
 .../drm/i915/display/skl_universal_plane.c    |   2 +-
 drivers/gpu/drm/i915/i915_drv.h               |  16 +-
 drivers/gpu/drm/i915/i915_pci.c               | 252 +++++++++++-------
 drivers/gpu/drm/i915/intel_device_info.c      | 101 +++----
 drivers/gpu/drm/i915/intel_device_info.h      |  18 --
 drivers/gpu/drm/i915/intel_step.c             |   8 +-
 13 files changed, 267 insertions(+), 175 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c
index 93c3226b98c90..182c6dd64f47c 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc.c
@@ -306,7 +306,7 @@ int intel_crtc_init(struct drm_i915_private *dev_priv, enum pipe pipe)
 		return PTR_ERR(crtc);
 
 	crtc->pipe = pipe;
-	crtc->num_scalers = RUNTIME_INFO(dev_priv)->num_scalers[pipe];
+	crtc->num_scalers = DISPLAY_RUNTIME_INFO(dev_priv)->num_scalers[pipe];
 
 	if (DISPLAY_VER(dev_priv) >= 9)
 		primary = skl_universal_plane_create(dev_priv, pipe,
diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c
index 3864da5f5c17f..b342fad180ca5 100644
--- a/drivers/gpu/drm/i915/display/intel_cursor.c
+++ b/drivers/gpu/drm/i915/display/intel_cursor.c
@@ -814,7 +814,7 @@ intel_cursor_plane_create(struct drm_i915_private *dev_priv,
 						   DRM_MODE_ROTATE_0 |
 						   DRM_MODE_ROTATE_180);
 
-	zpos = RUNTIME_INFO(dev_priv)->num_sprites[pipe] + 1;
+	zpos = DISPLAY_RUNTIME_INFO(dev_priv)->num_sprites[pipe] + 1;
 	drm_plane_create_zpos_immutable_property(&cursor->base, zpos);
 
 	if (DISPLAY_VER(dev_priv) >= 12)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 0490c6412ab5e..e477e16ea58e1 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -3366,7 +3366,7 @@ static u8 bigjoiner_pipes(struct drm_i915_private *i915)
 	else
 		pipes = 0;
 
-	return pipes & RUNTIME_INFO(i915)->pipe_mask;
+	return pipes & DISPLAY_RUNTIME_INFO(i915)->pipe_mask;
 }
 
 static bool transcoder_ddi_func_is_enabled(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index aa3a21ccd7fe1..c744c021af236 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -105,7 +105,7 @@ enum i9xx_plane_id {
 };
 
 #define plane_name(p) ((p) + 'A')
-#define sprite_name(p, s) ((p) * RUNTIME_INFO(dev_priv)->num_sprites[(p)] + (s) + 'A')
+#define sprite_name(p, s) ((p) * DISPLAY_RUNTIME_INFO(dev_priv)->num_sprites[(p)] + (s) + 'A')
 
 #define for_each_plane_id_on_crtc(__crtc, __p) \
 	for ((__p) = PLANE_PRIMARY; (__p) < I915_MAX_PLANES; (__p)++) \
@@ -221,7 +221,7 @@ enum phy_fia {
 
 #define for_each_pipe(__dev_priv, __p) \
 	for ((__p) = 0; (__p) < I915_MAX_PIPES; (__p)++) \
-		for_each_if(RUNTIME_INFO(__dev_priv)->pipe_mask & BIT(__p))
+		for_each_if(DISPLAY_RUNTIME_INFO(__dev_priv)->pipe_mask & BIT(__p))
 
 #define for_each_pipe_masked(__dev_priv, __p, __mask) \
 	for_each_pipe(__dev_priv, __p) \
@@ -229,7 +229,7 @@ enum phy_fia {
 
 #define for_each_cpu_transcoder(__dev_priv, __t) \
 	for ((__t) = 0; (__t) < I915_MAX_TRANSCODERS; (__t)++)	\
-		for_each_if (RUNTIME_INFO(__dev_priv)->cpu_transcoder_mask & BIT(__t))
+		for_each_if (DISPLAY_RUNTIME_INFO(__dev_priv)->cpu_transcoder_mask & BIT(__t))
 
 #define for_each_cpu_transcoder_masked(__dev_priv, __t, __mask) \
 	for_each_cpu_transcoder(__dev_priv, __t) \
@@ -237,7 +237,7 @@ enum phy_fia {
 
 #define for_each_sprite(__dev_priv, __p, __s)				\
 	for ((__s) = 0;							\
-	     (__s) < RUNTIME_INFO(__dev_priv)->num_sprites[(__p)];	\
+	     (__s) < DISPLAY_RUNTIME_INFO(__dev_priv)->num_sprites[(__p)];	\
 	     (__s)++)
 
 #define for_each_port(__port) \
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index c689d582dbf13..613607fad5af0 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -29,7 +29,30 @@
 	func(overlay_needs_physical); \
 	func(supports_tv);
 
+struct intel_display_runtime_info {
+	struct {
+		u16 ver;
+		u16 rel;
+		u16 step;
+	} ip;
+
+	u8 pipe_mask;
+	u8 cpu_transcoder_mask;
+
+	u8 num_sprites[I915_MAX_PIPES];
+	u8 num_scalers[I915_MAX_PIPES];
+
+	u8 fbc_mask;
+
+	bool has_hdcp;
+	bool has_dmc;
+	bool has_dsc;
+};
+
 struct intel_display_device_info {
+	/* Initial runtime info. */
+	const struct intel_display_runtime_info __runtime_defaults;
+
 	u8 abox_mask;
 
 	struct {
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 11bb8cf9c9d0d..1966f93962019 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -56,7 +56,7 @@
 
 #define for_each_fbc_id(__dev_priv, __fbc_id) \
 	for ((__fbc_id) = INTEL_FBC_A; (__fbc_id) < I915_MAX_FBCS; (__fbc_id)++) \
-		for_each_if(RUNTIME_INFO(__dev_priv)->fbc_mask & BIT(__fbc_id))
+		for_each_if(DISPLAY_RUNTIME_INFO(__dev_priv)->fbc_mask & BIT(__fbc_id))
 
 #define for_each_intel_fbc(__dev_priv, __fbc, __fbc_id) \
 	for_each_fbc_id((__dev_priv), (__fbc_id)) \
@@ -1708,10 +1708,10 @@ void intel_fbc_init(struct drm_i915_private *i915)
 	enum intel_fbc_id fbc_id;
 
 	if (!drm_mm_initialized(&i915->mm.stolen))
-		RUNTIME_INFO(i915)->fbc_mask = 0;
+		DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0;
 
 	if (need_fbc_vtd_wa(i915))
-		RUNTIME_INFO(i915)->fbc_mask = 0;
+		DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0;
 
 	i915->params.enable_fbc = intel_sanitize_fbc_option(i915);
 	drm_dbg_kms(&i915->drm, "Sanitized enable_fbc value: %d\n",
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index dd539106ee5ad..1f96d1fa68e04 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -1103,7 +1103,7 @@ static void intel_hdcp_prop_work(struct work_struct *work)
 
 bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port)
 {
-	return RUNTIME_INFO(dev_priv)->has_hdcp &&
+	return DISPLAY_RUNTIME_INFO(dev_priv)->has_hdcp &&
 		(DISPLAY_VER(dev_priv) >= 12 || port < PORT_E);
 }
 
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 110401aab038b..36070d86550fe 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -1944,7 +1944,7 @@ static enum intel_fbc_id skl_fbc_id_for_pipe(enum pipe pipe)
 static bool skl_plane_has_fbc(struct drm_i915_private *dev_priv,
 			      enum intel_fbc_id fbc_id, enum plane_id plane_id)
 {
-	if ((RUNTIME_INFO(dev_priv)->fbc_mask & BIT(fbc_id)) == 0)
+	if ((DISPLAY_RUNTIME_INFO(dev_priv)->fbc_mask & BIT(fbc_id)) == 0)
 		return false;
 
 	return plane_id == PLANE_PRIMARY;
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 0c341107363f7..dbdecf1ee24fe 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -205,6 +205,7 @@ struct drm_i915_private {
 
 	const struct intel_device_info __info; /* Use INTEL_INFO() to access. */
 	struct intel_runtime_info __runtime; /* Use RUNTIME_INFO() to access. */
+	struct intel_display_runtime_info __display_runtime; /* Access with DISPLAY_RUNTIME_INFO() */
 	struct intel_driver_caps caps;
 
 	struct i915_dsm dsm;
@@ -410,6 +411,7 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 #define INTEL_INFO(dev_priv)	(&(dev_priv)->__info)
 #define DISPLAY_INFO(i915)	(INTEL_INFO(i915)->display)
 #define RUNTIME_INFO(dev_priv)	(&(dev_priv)->__runtime)
+#define DISPLAY_RUNTIME_INFO(i915)	(&(i915)->__display_runtime)
 #define DRIVER_CAPS(dev_priv)	(&(dev_priv)->caps)
 
 #define INTEL_DEVID(dev_priv)	(RUNTIME_INFO(dev_priv)->device_id)
@@ -428,7 +430,7 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915)
 #define IS_MEDIA_VER(i915, from, until) \
 	(MEDIA_VER(i915) >= (from) && MEDIA_VER(i915) <= (until))
 
-#define DISPLAY_VER(i915)	(RUNTIME_INFO(i915)->display.ip.ver)
+#define DISPLAY_VER(i915)	(DISPLAY_RUNTIME_INFO(i915)->ip.ver)
 #define IS_DISPLAY_VER(i915, from, until) \
 	(DISPLAY_VER(i915) >= (from) && DISPLAY_VER(i915) <= (until))
 
@@ -811,7 +813,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define I915_HAS_HOTPLUG(dev_priv)	(DISPLAY_INFO(dev_priv)->has_hotplug)
 
 #define HAS_FW_BLC(dev_priv)	(DISPLAY_VER(dev_priv) > 2)
-#define HAS_FBC(dev_priv)	(RUNTIME_INFO(dev_priv)->fbc_mask != 0)
+#define HAS_FBC(dev_priv)	(DISPLAY_RUNTIME_INFO(dev_priv)->fbc_mask != 0)
 #define HAS_CUR_FBC(dev_priv)	(!HAS_GMCH(dev_priv) && DISPLAY_VER(dev_priv) >= 7)
 
 #define HAS_DPT(dev_priv)	(DISPLAY_VER(dev_priv) >= 13)
@@ -831,7 +833,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define HAS_PSR_HW_TRACKING(dev_priv) \
 	(DISPLAY_INFO(dev_priv)->has_psr_hw_tracking)
 #define HAS_PSR2_SEL_FETCH(dev_priv)	 (DISPLAY_VER(dev_priv) >= 12)
-#define HAS_TRANSCODER(dev_priv, trans)	 ((RUNTIME_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
+#define HAS_TRANSCODER(dev_priv, trans)	 ((DISPLAY_RUNTIME_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
 
 #define HAS_RC6(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6)
 #define HAS_RC6p(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6p)
@@ -839,9 +841,9 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_RPS(dev_priv)	(INTEL_INFO(dev_priv)->has_rps)
 
-#define HAS_DMC(dev_priv)	(RUNTIME_INFO(dev_priv)->has_dmc)
+#define HAS_DMC(dev_priv)	(DISPLAY_RUNTIME_INFO(dev_priv)->has_dmc)
 #define HAS_DSB(dev_priv)	(DISPLAY_INFO(dev_priv)->has_dsb)
-#define HAS_DSC(__i915)		(RUNTIME_INFO(__i915)->has_dsc)
+#define HAS_DSC(__i915)		(DISPLAY_RUNTIME_INFO(__i915)->has_dsc)
 #define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915))
 
 #define HAS_HECI_PXP(dev_priv) \
@@ -903,9 +905,9 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define NUM_L3_SLICES(dev_priv) (IS_HSW_GT3(dev_priv) ? \
 				 2 : HAS_L3_DPF(dev_priv))
 
-#define INTEL_NUM_PIPES(dev_priv) (hweight8(RUNTIME_INFO(dev_priv)->pipe_mask))
+#define INTEL_NUM_PIPES(dev_priv) (hweight8(DISPLAY_RUNTIME_INFO(dev_priv)->pipe_mask))
 
-#define HAS_DISPLAY(dev_priv) (RUNTIME_INFO(dev_priv)->pipe_mask != 0)
+#define HAS_DISPLAY(dev_priv) (DISPLAY_RUNTIME_INFO(dev_priv)->pipe_mask != 0)
 
 #define HAS_VRR(i915)	(DISPLAY_VER(i915) >= 11)
 
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 2366361bfbe93..261569e15f91b 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -40,10 +40,9 @@
 #define PLATFORM(x) .platform = (x)
 #define GEN(x) \
 	.__runtime.graphics.ip.ver = (x), \
-	.__runtime.media.ip.ver = (x), \
-	.__runtime.display.ip.ver = (x)
+	.__runtime.media.ip.ver = (x)
 
-#define NO_DISPLAY .__runtime.pipe_mask = 0
+static const struct intel_display_device_info no_display = {};
 
 #define I845_PIPE_OFFSETS \
 	.pipe_offsets = { \
@@ -179,7 +178,12 @@
 	.has_gmch = 1, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 2, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
 
 static const struct intel_display_device_info i830_display = {
 	I830_DISPLAY,
@@ -188,8 +192,6 @@ static const struct intel_display_device_info i830_display = {
 #define I830_FEATURES \
 	GEN(2), \
 	.is_mobile = 1, \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
 	.gpu_reset_clobbers_display = true, \
 	.has_3d_pipeline = 1, \
 	.hws_needs_physical = 1, \
@@ -207,7 +209,11 @@ static const struct intel_display_device_info i830_display = {
 	.has_gmch = 1, \
 	I845_PIPE_OFFSETS, \
 	I845_CURSOR_OFFSETS, \
-	I845_COLORS
+	I845_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 2, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A), \
+	.__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A)
 
 static const struct intel_display_device_info i845_display = {
 	I845_DISPLAY,
@@ -215,8 +221,6 @@ static const struct intel_display_device_info i845_display = {
 
 #define I845_FEATURES \
 	GEN(2), \
-	.__runtime.pipe_mask = BIT(PIPE_A), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A), \
 	.has_3d_pipeline = 1, \
 	.gpu_reset_clobbers_display = true, \
 	.hws_needs_physical = 1, \
@@ -242,24 +246,26 @@ static const struct intel_device_info i845g_info = {
 
 static const struct intel_display_device_info i85x_display = {
 	I830_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info i85x_info = {
 	I830_FEATURES,
 	PLATFORM(INTEL_I85X),
 	.display = &i85x_display,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_display_device_info i865g_display = {
 	I845_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info i865g_info = {
 	I845_FEATURES,
 	PLATFORM(INTEL_I865G),
 	.display = &i865g_display,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define GEN3_DISPLAY \
@@ -267,7 +273,12 @@ static const struct intel_device_info i865g_info = {
 	.has_overlay = 1, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 3, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
 
 static const struct intel_display_device_info i915g_display = {
 	GEN3_DISPLAY,
@@ -280,6 +291,8 @@ static const struct intel_display_device_info i915gm_display = {
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
 	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_display_device_info i945g_display = {
@@ -295,6 +308,8 @@ static const struct intel_display_device_info i945gm_display = {
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
 	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_display_device_info g33_display = {
@@ -304,8 +319,6 @@ static const struct intel_display_device_info g33_display = {
 
 #define GEN3_FEATURES \
 	GEN(3), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
 	.gpu_reset_clobbers_display = true, \
 	.__runtime.platform_engine_mask = BIT(RCS0), \
 	.has_3d_pipeline = 1, \
@@ -329,7 +342,6 @@ static const struct intel_device_info i915gm_info = {
 	PLATFORM(INTEL_I915GM),
 	.display = &i915gm_display,
 	.is_mobile = 1,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
 };
@@ -347,7 +359,6 @@ static const struct intel_device_info i945gm_info = {
 	PLATFORM(INTEL_I945GM),
 	.display = &i945gm_display,
 	.is_mobile = 1,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
 };
@@ -379,7 +390,12 @@ static const struct intel_device_info pnv_m_info = {
 	.has_gmch = 1, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 4, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
 
 static const struct intel_display_device_info i965g_display = {
 	GEN4_DISPLAY,
@@ -390,6 +406,8 @@ static const struct intel_display_device_info i965gm_display = {
 	GEN4_DISPLAY,
 	.has_overlay = 1,
 	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_display_device_info g45_display = {
@@ -399,12 +417,12 @@ static const struct intel_display_device_info g45_display = {
 static const struct intel_display_device_info gm45_display = {
 	GEN4_DISPLAY,
 	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define GEN4_FEATURES \
 	GEN(4), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
 	.gpu_reset_clobbers_display = true, \
 	.__runtime.platform_engine_mask = BIT(RCS0), \
 	.has_3d_pipeline = 1, \
@@ -427,7 +445,6 @@ static const struct intel_device_info i965gm_info = {
 	PLATFORM(INTEL_I965GM),
 	.display = &i965gm_display,
 	.is_mobile = 1,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.hws_needs_physical = 1,
 	.has_snoop = false,
 };
@@ -444,7 +461,6 @@ static const struct intel_device_info gm45_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_GM45),
 	.is_mobile = 1,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0),
 	.display = &gm45_display,
 	.gpu_reset_clobbers_display = false,
@@ -452,8 +468,6 @@ static const struct intel_device_info gm45_info = {
 
 #define GEN5_FEATURES \
 	GEN(5), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), \
 	.has_3d_pipeline = 1, \
 	.has_snoop = true, \
@@ -468,7 +482,12 @@ static const struct intel_device_info gm45_info = {
 	.has_hotplug = 1, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
-	ILK_COLORS
+	ILK_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 5, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
 
 static const struct intel_display_device_info ilk_d_display = {
 	ILK_DISPLAY,
@@ -482,6 +501,8 @@ static const struct intel_device_info ilk_d_info = {
 
 static const struct intel_display_device_info ilk_m_display = {
 	ILK_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
  };
 
 static const struct intel_device_info ilk_m_info = {
@@ -490,14 +511,10 @@ static const struct intel_device_info ilk_m_info = {
 	.display = &ilk_m_display,
 	.is_mobile = 1,
 	.has_rps = true,
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define GEN6_FEATURES \
 	GEN(6), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_3d_pipeline = 1, \
 	.has_coherent_ggtt = true, \
@@ -517,6 +534,12 @@ static const struct intel_display_device_info snb_display = {
 	I9XX_PIPE_OFFSETS,
 	I9XX_CURSOR_OFFSETS,
 	ILK_COLORS,
+
+	.__runtime_defaults.ip.ver = 6,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define SNB_D_PLATFORM \
@@ -555,9 +578,6 @@ static const struct intel_device_info snb_m_gt2_info = {
 
 #define GEN7_FEATURES  \
 	GEN(7), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), \
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \
 	.has_3d_pipeline = 1, \
 	.has_coherent_ggtt = true, \
@@ -582,6 +602,12 @@ static const struct intel_display_device_info ivb_display = {
 	IVB_PIPE_OFFSETS,
 	IVB_CURSOR_OFFSETS,
 	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info ivb_d_gt1_info = {
@@ -617,7 +643,7 @@ static const struct intel_device_info ivb_m_gt2_info = {
 static const struct intel_device_info ivb_q_info = {
 	GEN7_FEATURES,
 	PLATFORM(INTEL_IVYBRIDGE),
-	NO_DISPLAY,
+	.display = &no_display,
 	.gt = 2,
 	.has_l3_dpf = 1,
 };
@@ -629,14 +655,17 @@ static const struct intel_display_device_info vlv_display = {
 	I9XX_PIPE_OFFSETS,
 	I9XX_CURSOR_OFFSETS,
 	I9XX_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
 };
 
 static const struct intel_device_info vlv_info = {
 	PLATFORM(INTEL_VALLEYVIEW),
 	GEN(7),
 	.is_lp = 1,
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
 	.display = &vlv_display,
 	.has_runtime_pm = 1,
 	.has_rc6 = 1,
@@ -655,8 +684,6 @@ static const struct intel_device_info vlv_info = {
 #define G75_FEATURES  \
 	GEN7_FEATURES, \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), \
 	.has_rc6p = 0 /* RC6p removed-by HSW */, \
 	.has_runtime_pm = 1
 
@@ -673,6 +700,13 @@ static const struct intel_display_device_info hsw_display = {
 	HSW_PIPE_OFFSETS,
 	IVB_CURSOR_OFFSETS,
 	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info hsw_gt1_info = {
@@ -710,6 +744,13 @@ static const struct intel_display_device_info bdw_display = {
 	HSW_PIPE_OFFSETS,
 	IVB_CURSOR_OFFSETS,
 	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 8,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define BDW_PLATFORM \
@@ -752,13 +793,16 @@ static const struct intel_display_device_info chv_display = {
 	CHV_PIPE_OFFSETS,
 	CHV_CURSOR_OFFSETS,
 	CHV_COLORS,
+
+	.__runtime_defaults.ip.ver = 8,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
 };
 
 static const struct intel_device_info chv_info = {
 	PLATFORM(INTEL_CHERRYVIEW),
 	GEN(8),
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
 	.display = &chv_display,
 	.is_lp = 1,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0),
@@ -785,9 +829,7 @@ static const struct intel_device_info chv_info = {
 	GEN8_FEATURES, \
 	GEN(9), \
 	GEN9_DEFAULT_PAGE_SIZES, \
-	.__runtime.has_dmc = 1, \
-	.has_gt_uc = 1, \
-	.__runtime.has_hdcp = 1
+	.has_gt_uc = 1
 
 static const struct intel_display_device_info skl_display = {
 	.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */
@@ -802,6 +844,15 @@ static const struct intel_display_device_info skl_display = {
 	HSW_PIPE_OFFSETS,
 	IVB_CURSOR_OFFSETS,
 	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 9,
+	.__runtime_defaults.has_dmc = 1,
+	.__runtime_defaults.has_hdcp = 1,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 #define SKL_PLATFORM \
@@ -842,16 +893,9 @@ static const struct intel_device_info skl_gt4_info = {
 	GEN(9), \
 	.is_lp = 1, \
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
-		BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C), \
 	.has_3d_pipeline = 1, \
 	.has_64bit_reloc = 1, \
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A), \
-	.__runtime.has_hdcp = 1, \
 	.has_runtime_pm = 1, \
-	.__runtime.has_dmc = 1, \
 	.has_rc6 = 1, \
 	.has_rps = true, \
 	.has_logical_ring_contexts = 1, \
@@ -876,11 +920,22 @@ static const struct intel_device_info skl_gt4_info = {
 	.has_psr_hw_tracking = 1, \
 	HSW_PIPE_OFFSETS, \
 	IVB_CURSOR_OFFSETS, \
-	IVB_COLORS
+	IVB_COLORS, \
+	\
+	.__runtime_defaults.has_dmc = 1, \
+	.__runtime_defaults.has_hdcp = 1, \
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
+		BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C)
 
 static const struct intel_display_device_info bxt_display = {
 	GEN9_LP_DISPLAY,
 	.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */
+
+	.__runtime_defaults.ip.ver = 9,
 };
 
 static const struct intel_device_info bxt_info = {
@@ -893,12 +948,13 @@ static const struct intel_display_device_info glk_display = {
 	GEN9_LP_DISPLAY,
 	.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */
 	GLK_COLORS,
+
+	.__runtime_defaults.ip.ver = 10,
 };
 
 static const struct intel_device_info glk_info = {
 	GEN9_LP_FEATURES,
 	PLATFORM(INTEL_GEMINILAKE),
-	.__runtime.display.ip.ver = 10,
 	.display = &glk_display,
 };
 
@@ -974,11 +1030,7 @@ static const struct intel_device_info cml_gt2_info = {
 #define GEN11_FEATURES \
 	GEN9_FEATURES, \
 	GEN11_DEFAULT_PAGE_SIZES, \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
-		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
 	GEN(11), \
-	.__runtime.has_dsc = 1, \
 	.has_coherent_ggtt = false, \
 	.has_logical_ring_elsq = 1
 
@@ -1011,6 +1063,17 @@ static const struct intel_display_device_info gen11_display = {
 	},
 	IVB_CURSOR_OFFSETS,
 	ICL_COLORS,
+
+	.__runtime_defaults.ip.ver = 11,
+	.__runtime_defaults.has_dmc = 1,
+	.__runtime_defaults.has_dsc = 1,
+	.__runtime_defaults.has_hdcp = 1,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) |
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info icl_info = {
@@ -1040,10 +1103,6 @@ static const struct intel_device_info jsl_info = {
 #define GEN12_FEATURES \
 	GEN11_FEATURES, \
 	GEN(12), \
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \
-		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
 	.has_global_mocs = 1, \
 	.has_pxp = 1
 
@@ -1076,7 +1135,19 @@ static const struct intel_device_info jsl_info = {
 		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
 	}, \
 	TGL_CURSOR_OFFSETS, \
-	ICL_COLORS
+	ICL_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 12, \
+	.__runtime_defaults.has_dmc = 1, \
+	.__runtime_defaults.has_dsc = 1, \
+	.__runtime_defaults.has_hdcp = 1, \
+	.__runtime_defaults.pipe_mask = \
+		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A)
 
 static const struct intel_display_device_info tgl_display = {
 	XE_D_DISPLAY,
@@ -1095,14 +1166,15 @@ static const struct intel_display_device_info rkl_display = {
 	.abox_mask = BIT(0),
 	.has_hti = 1,
 	.has_psr_hw_tracking = 0,
+
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
 };
 
 static const struct intel_device_info rkl_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_ROCKETLAKE),
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0),
 	.display = &rkl_display,
@@ -1121,7 +1193,6 @@ static const struct intel_device_info dg1_info = {
 	DGFX_FEATURES,
 	.__runtime.graphics.ip.rel = 10,
 	PLATFORM(INTEL_DG1),
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D),
 	.require_force_probe = 1,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) |
@@ -1140,7 +1211,6 @@ static const struct intel_display_device_info adl_s_display = {
 static const struct intel_device_info adl_s_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_ALDERLAKE_S),
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.dma_mask_size = 39,
@@ -1180,29 +1250,30 @@ static const struct intel_device_info adl_s_info = {
 		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,			\
 		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,			\
 	},									\
-	TGL_CURSOR_OFFSETS
-
-#define XE_LPD_RUNTIME \
-	.__runtime.has_dmc = 1,							\
-	.__runtime.has_dsc = 1,							\
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A),					\
-	.__runtime.has_hdcp = 1,						\
-	.__runtime.display.ip.ver = 13,							\
-	.__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D)
+	TGL_CURSOR_OFFSETS,							\
+										\
+	.__runtime_defaults.ip.ver = 13,					\
+	.__runtime_defaults.has_dmc = 1,					\
+	.__runtime_defaults.has_dsc = 1,					\
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),			\
+	.__runtime_defaults.has_hdcp = 1,					\
+	.__runtime_defaults.pipe_mask =						\
+		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D)
 
 static const struct intel_display_device_info xe_lpd_display = {
 	XE_LPD_FEATURES,
 	.has_cdclk_crawl = 1,
 	.has_psr_hw_tracking = 0,
+
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) |
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
 };
 
 static const struct intel_device_info adl_p_info = {
 	GEN12_FEATURES,
-	XE_LPD_RUNTIME,
 	PLATFORM(INTEL_ALDERLAKE_P),
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D) |
-			       BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.__runtime.ppgtt_size = 48,
@@ -1252,7 +1323,7 @@ static const struct intel_device_info xehpsdv_info = {
 	XE_HPM_FEATURES,
 	DGFX_FEATURES,
 	PLATFORM(INTEL_XEHPSDV),
-	NO_DISPLAY,
+	.display = &no_display,
 	.has_64k_pages = 1,
 	.has_media_ratio_mode = 1,
 	.__runtime.platform_engine_mask =
@@ -1284,19 +1355,20 @@ static const struct intel_device_info xehpsdv_info = {
 static const struct intel_display_device_info xe_hpd_display = {
 	XE_LPD_FEATURES,
 	.has_cdclk_squash = 1,
+
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 };
 
 static const struct intel_device_info dg2_info = {
 	DG2_FEATURES,
-	XE_LPD_RUNTIME,
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 	.display = &xe_hpd_display,
 };
 
 static const struct intel_device_info ats_m_info = {
 	DG2_FEATURES,
-	NO_DISPLAY,
+	.display = &no_display,
 	.require_force_probe = 1,
 	.tuning_thread_rr_after_dep = 1,
 };
@@ -1318,7 +1390,7 @@ static const struct intel_device_info pvc_info = {
 	.__runtime.graphics.ip.rel = 60,
 	.__runtime.media.ip.rel = 60,
 	PLATFORM(INTEL_PONTEVECCHIO),
-	NO_DISPLAY,
+	.display = &no_display,
 	.has_flat_ccs = 0,
 	.__runtime.platform_engine_mask =
 		BIT(BCS0) |
@@ -1327,11 +1399,6 @@ static const struct intel_device_info pvc_info = {
 	.require_force_probe = 1,
 };
 
-#define XE_LPDP_RUNTIME	\
-	XE_LPD_RUNTIME,	\
-	.__runtime.display.ip.ver = 14,	\
-	.__runtime.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B)
-
 static const struct intel_gt_definition xelpmp_extra_gt[] = {
 	{
 		.type = GT_MEDIA,
@@ -1346,13 +1413,16 @@ static const struct intel_display_device_info xe_lpdp_display = {
 	XE_LPD_FEATURES,
 	.has_cdclk_crawl = 1,
 	.has_cdclk_squash = 1,
+
+	.__runtime_defaults.ip.ver = 14,
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 };
 
 static const struct intel_device_info mtl_info = {
 	XE_HP_FEATURES,
-	XE_LPDP_RUNTIME,
-	.__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-			       BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 	/*
 	 * Real graphics IP version will be obtained from hardware GMD_ID
 	 * register.  Value provided here is just for sanity checking.
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index e8836b9900589..3a0d1cb2eaa9c 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -95,6 +95,9 @@ void intel_device_info_print(const struct intel_device_info *info,
 			     const struct intel_runtime_info *runtime,
 			     struct drm_printer *p)
 {
+	const struct intel_display_runtime_info *display_runtime =
+		&info->display->__runtime_defaults;
+
 	if (runtime->graphics.ip.rel)
 		drm_printf(p, "graphics version: %u.%02u\n",
 			   runtime->graphics.ip.ver,
@@ -111,13 +114,13 @@ void intel_device_info_print(const struct intel_device_info *info,
 		drm_printf(p, "media version: %u\n",
 			   runtime->media.ip.ver);
 
-	if (runtime->display.ip.rel)
+	if (display_runtime->ip.rel)
 		drm_printf(p, "display version: %u.%02u\n",
-			   runtime->display.ip.ver,
-			   runtime->display.ip.rel);
+			   display_runtime->ip.ver,
+			   display_runtime->ip.rel);
 	else
 		drm_printf(p, "display version: %u\n",
-			   runtime->display.ip.ver);
+			   display_runtime->ip.ver);
 
 	drm_printf(p, "graphics stepping: %s\n", intel_step_name(runtime->step.graphics_step));
 	drm_printf(p, "media stepping: %s\n", intel_step_name(runtime->step.media_step));
@@ -142,9 +145,9 @@ void intel_device_info_print(const struct intel_device_info *info,
 	DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG);
 #undef PRINT_FLAG
 
-	drm_printf(p, "has_hdcp: %s\n", str_yes_no(runtime->has_hdcp));
-	drm_printf(p, "has_dmc: %s\n", str_yes_no(runtime->has_dmc));
-	drm_printf(p, "has_dsc: %s\n", str_yes_no(runtime->has_dsc));
+	drm_printf(p, "has_hdcp: %s\n", str_yes_no(display_runtime->has_hdcp));
+	drm_printf(p, "has_dmc: %s\n", str_yes_no(display_runtime->has_dmc));
+	drm_printf(p, "has_dsc: %s\n", str_yes_no(display_runtime->has_dsc));
 
 	drm_printf(p, "rawclk rate: %u kHz\n", runtime->rawclk_freq);
 }
@@ -342,6 +345,7 @@ static void ip_ver_read(struct drm_i915_private *i915, u32 offset, struct intel_
 static void intel_ipver_early_init(struct drm_i915_private *i915)
 {
 	struct intel_runtime_info *runtime = RUNTIME_INFO(i915);
+	struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(i915);
 
 	if (!HAS_GMD_ID(i915)) {
 		drm_WARN_ON(&i915->drm, RUNTIME_INFO(i915)->graphics.ip.ver > 12);
@@ -363,7 +367,7 @@ static void intel_ipver_early_init(struct drm_i915_private *i915)
 		RUNTIME_INFO(i915)->graphics.ip.rel = 70;
 	}
 	ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_DISPLAY),
-		    &runtime->display.ip);
+		    (struct intel_ip_version *)&display_runtime->ip);
 	ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_MEDIA),
 		    &runtime->media.ip);
 }
@@ -410,32 +414,34 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 {
 	struct intel_device_info *info = mkwrite_device_info(dev_priv);
 	struct intel_runtime_info *runtime = RUNTIME_INFO(dev_priv);
+	struct intel_display_runtime_info *display_runtime =
+		DISPLAY_RUNTIME_INFO(dev_priv);
 	enum pipe pipe;
 
 	/* Wa_14011765242: adl-s A0,A1 */
 	if (IS_ADLS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_A2))
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_scalers[pipe] = 0;
+			display_runtime->num_scalers[pipe] = 0;
 	else if (DISPLAY_VER(dev_priv) >= 11) {
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_scalers[pipe] = 2;
+			display_runtime->num_scalers[pipe] = 2;
 	} else if (DISPLAY_VER(dev_priv) >= 9) {
-		runtime->num_scalers[PIPE_A] = 2;
-		runtime->num_scalers[PIPE_B] = 2;
-		runtime->num_scalers[PIPE_C] = 1;
+		display_runtime->num_scalers[PIPE_A] = 2;
+		display_runtime->num_scalers[PIPE_B] = 2;
+		display_runtime->num_scalers[PIPE_C] = 1;
 	}
 
 	BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES);
 
 	if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv))
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_sprites[pipe] = 4;
+			display_runtime->num_sprites[pipe] = 4;
 	else if (DISPLAY_VER(dev_priv) >= 11)
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_sprites[pipe] = 6;
+			display_runtime->num_sprites[pipe] = 6;
 	else if (DISPLAY_VER(dev_priv) == 10)
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_sprites[pipe] = 3;
+			display_runtime->num_sprites[pipe] = 3;
 	else if (IS_BROXTON(dev_priv)) {
 		/*
 		 * Skylake and Broxton currently don't expose the topmost plane as its
@@ -446,15 +452,15 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 		 * down the line.
 		 */
 
-		runtime->num_sprites[PIPE_A] = 2;
-		runtime->num_sprites[PIPE_B] = 2;
-		runtime->num_sprites[PIPE_C] = 1;
+		display_runtime->num_sprites[PIPE_A] = 2;
+		display_runtime->num_sprites[PIPE_B] = 2;
+		display_runtime->num_sprites[PIPE_C] = 1;
 	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_sprites[pipe] = 2;
+			display_runtime->num_sprites[pipe] = 2;
 	} else if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) {
 		for_each_pipe(dev_priv, pipe)
-			runtime->num_sprites[pipe] = 1;
+			display_runtime->num_sprites[pipe] = 1;
 	}
 
 	if (HAS_DISPLAY(dev_priv) &&
@@ -462,7 +468,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 	    !(intel_de_read(dev_priv, GU_CNTL_PROTECTED) & DEPRESENT)) {
 		drm_info(&dev_priv->drm, "Display not present, disabling\n");
 
-		runtime->pipe_mask = 0;
+		display_runtime->pipe_mask = 0;
 	}
 
 	if (HAS_DISPLAY(dev_priv) && IS_GRAPHICS_VER(dev_priv, 7, 8) &&
@@ -485,47 +491,47 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 		     !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) {
 			drm_info(&dev_priv->drm,
 				 "Display fused off, disabling\n");
-			runtime->pipe_mask = 0;
+			display_runtime->pipe_mask = 0;
 		} else if (fuse_strap & IVB_PIPE_C_DISABLE) {
 			drm_info(&dev_priv->drm, "PipeC fused off\n");
-			runtime->pipe_mask &= ~BIT(PIPE_C);
-			runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
+			display_runtime->pipe_mask &= ~BIT(PIPE_C);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
 		}
 	} else if (HAS_DISPLAY(dev_priv) && DISPLAY_VER(dev_priv) >= 9) {
 		u32 dfsm = intel_de_read(dev_priv, SKL_DFSM);
 
 		if (dfsm & SKL_DFSM_PIPE_A_DISABLE) {
-			runtime->pipe_mask &= ~BIT(PIPE_A);
-			runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A);
-			runtime->fbc_mask &= ~BIT(INTEL_FBC_A);
+			display_runtime->pipe_mask &= ~BIT(PIPE_A);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A);
+			display_runtime->fbc_mask &= ~BIT(INTEL_FBC_A);
 		}
 		if (dfsm & SKL_DFSM_PIPE_B_DISABLE) {
-			runtime->pipe_mask &= ~BIT(PIPE_B);
-			runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B);
+			display_runtime->pipe_mask &= ~BIT(PIPE_B);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B);
 		}
 		if (dfsm & SKL_DFSM_PIPE_C_DISABLE) {
-			runtime->pipe_mask &= ~BIT(PIPE_C);
-			runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
+			display_runtime->pipe_mask &= ~BIT(PIPE_C);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
 		}
 
 		if (DISPLAY_VER(dev_priv) >= 12 &&
 		    (dfsm & TGL_DFSM_PIPE_D_DISABLE)) {
-			runtime->pipe_mask &= ~BIT(PIPE_D);
-			runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D);
+			display_runtime->pipe_mask &= ~BIT(PIPE_D);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D);
 		}
 
 		if (dfsm & SKL_DFSM_DISPLAY_HDCP_DISABLE)
-			runtime->has_hdcp = 0;
+			display_runtime->has_hdcp = 0;
 
 		if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE)
-			runtime->fbc_mask = 0;
+			display_runtime->fbc_mask = 0;
 
 		if (DISPLAY_VER(dev_priv) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE))
-			runtime->has_dmc = 0;
+			display_runtime->has_dmc = 0;
 
 		if (IS_DISPLAY_VER(dev_priv, 10, 12) &&
 		    (dfsm & GLK_DFSM_DISPLAY_DSC_DISABLE))
-			runtime->has_dsc = 0;
+			display_runtime->has_dsc = 0;
 	}
 
 	if (GRAPHICS_VER(dev_priv) == 6 && i915_vtd_active(dev_priv)) {
@@ -542,13 +548,13 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 						   DRIVER_ATOMIC);
 		info->display = &no_display;
 
-		runtime->cpu_transcoder_mask = 0;
-		memset(runtime->num_sprites, 0, sizeof(runtime->num_sprites));
-		memset(runtime->num_scalers, 0, sizeof(runtime->num_scalers));
-		runtime->fbc_mask = 0;
-		runtime->has_hdcp = false;
-		runtime->has_dmc = false;
-		runtime->has_dsc = false;
+		display_runtime->cpu_transcoder_mask = 0;
+		memset(display_runtime->num_sprites, 0, sizeof(display_runtime->num_sprites));
+		memset(display_runtime->num_scalers, 0, sizeof(display_runtime->num_scalers));
+		display_runtime->fbc_mask = 0;
+		display_runtime->has_hdcp = false;
+		display_runtime->has_dmc = false;
+		display_runtime->has_dsc = false;
 	}
 
 	/* Disable nuclear pageflip by default on pre-g4x */
@@ -568,6 +574,7 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 {
 	struct intel_device_info *info;
 	struct intel_runtime_info *runtime;
+	struct intel_display_runtime_info *display_runtime;
 
 	/* Setup the write-once "constant" device info */
 	info = mkwrite_device_info(i915);
@@ -576,6 +583,10 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 	/* Initialize initial runtime info from static const data and pdev. */
 	runtime = RUNTIME_INFO(i915);
 	memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime));
+	display_runtime = DISPLAY_RUNTIME_INFO(i915);
+	memcpy(display_runtime, &DISPLAY_INFO(i915)->__runtime_defaults,
+	       sizeof(*display_runtime));
+
 	runtime->device_id = device_id;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h
index f838ae7191644..faf6cccdb343d 100644
--- a/drivers/gpu/drm/i915/intel_device_info.h
+++ b/drivers/gpu/drm/i915/intel_device_info.h
@@ -197,9 +197,6 @@ struct intel_runtime_info {
 	struct {
 		struct intel_ip_version ip;
 	} media;
-	struct {
-		struct intel_ip_version ip;
-	} display;
 
 	/*
 	 * Platform mask is used for optimizing or-ed IS_PLATFORM calls into
@@ -227,21 +224,6 @@ struct intel_runtime_info {
 	u32 memory_regions; /* regions supported by the HW */
 
 	bool has_pooled_eu;
-
-	/* display */
-	struct {
-		u8 pipe_mask;
-		u8 cpu_transcoder_mask;
-
-		u8 num_sprites[I915_MAX_PIPES];
-		u8 num_scalers[I915_MAX_PIPES];
-
-		u8 fbc_mask;
-
-		bool has_hdcp;
-		bool has_dmc;
-		bool has_dsc;
-	};
 };
 
 struct intel_device_info {
diff --git a/drivers/gpu/drm/i915/intel_step.c b/drivers/gpu/drm/i915/intel_step.c
index 84a6fe736a3b5..8a9ff6227e536 100644
--- a/drivers/gpu/drm/i915/intel_step.c
+++ b/drivers/gpu/drm/i915/intel_step.c
@@ -166,8 +166,12 @@ void intel_step_init(struct drm_i915_private *i915)
 						       &RUNTIME_INFO(i915)->graphics.ip);
 		step.media_step = gmd_to_intel_step(i915,
 						    &RUNTIME_INFO(i915)->media.ip);
-		step.display_step = gmd_to_intel_step(i915,
-						      &RUNTIME_INFO(i915)->display.ip);
+		step.display_step = STEP_A0 + DISPLAY_RUNTIME_INFO(i915)->ip.step;
+		if (step.display_step >= STEP_FUTURE) {
+			drm_dbg(&i915->drm, "Using future display steppings\n");
+			step.display_step = STEP_FUTURE;
+		}
+
 		RUNTIME_INFO(i915)->step = step;
 
 		return;
-- 
GitLab


From 69d439818fe501e8c9e50d963a53cb596e36f9f7 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:07 -0700
Subject: [PATCH 0650/1791] drm/i915/display: Make display responsible for
 probing its own IP

Rather than selecting the display IP and feature flags at the same time
the general PCI probing happens, move this step into the display code
itself so that it can be more easily re-used outside of i915 (i.e., by
the Xe driver).

v2:
 - Make intel_display_device_probe() always return a non-NULL pointer
   and simplify copying of runtime_defaults.  (Andrzej)
v3:
 - Redefine INTEL_VGA_DEVICE/INTEL_QUANTA_DEVICE to eliminate a cast and
   an include of linux/mod_devicetable.h.  (Jani)
 - Keep explicit memcpy for runtime defaults.  (Jani)

Cc: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-5-matthew.d.roper@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |   2 +
 .../drm/i915/display/intel_display_device.c   | 718 ++++++++++++++++++
 .../drm/i915/display/intel_display_device.h   |   3 +
 drivers/gpu/drm/i915/i915_pci.c               | 665 ----------------
 drivers/gpu/drm/i915/i915_reg.h               |  33 -
 drivers/gpu/drm/i915/intel_device_info.c      |  10 +-
 6 files changed, 729 insertions(+), 702 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_display_device.c

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index e3b699e7a755f..83ba84f78800d 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -25,6 +25,7 @@ subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror
 
 # Fine grained warnings disable
 CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init)
+CFLAGS_display/intel_display_device.o = $(call cc-disable-warning, override-init)
 CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init)
 
 subdir-ccflags-y += -I$(srctree)/$(src)
@@ -307,6 +308,7 @@ i915-y += \
 	display/intel_cx0_phy.o \
 	display/intel_ddi.o \
 	display/intel_ddi_buf_trans.o \
+	display/intel_display_device.o \
 	display/intel_display_trace.o \
 	display/intel_dkl_phy.o \
 	display/intel_dp.o \
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
new file mode 100644
index 0000000000000..3e2f4cd0b9f56
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -0,0 +1,718 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <drm/i915_pciids.h>
+#include <drm/drm_color_mgmt.h>
+
+#include "intel_display_device.h"
+#include "intel_display_power.h"
+#include "intel_display_reg_defs.h"
+#include "intel_fbc.h"
+
+static const struct intel_display_device_info no_display = {};
+
+#define PIPE_A_OFFSET		0x70000
+#define PIPE_B_OFFSET		0x71000
+#define PIPE_C_OFFSET		0x72000
+#define PIPE_D_OFFSET		0x73000
+#define CHV_PIPE_C_OFFSET	0x74000
+/*
+ * There's actually no pipe EDP. Some pipe registers have
+ * simply shifted from the pipe to the transcoder, while
+ * keeping their original offset. Thus we need PIPE_EDP_OFFSET
+ * to access such registers in transcoder EDP.
+ */
+#define PIPE_EDP_OFFSET	0x7f000
+
+/* ICL DSI 0 and 1 */
+#define PIPE_DSI0_OFFSET	0x7b000
+#define PIPE_DSI1_OFFSET	0x7b800
+
+#define TRANSCODER_A_OFFSET 0x60000
+#define TRANSCODER_B_OFFSET 0x61000
+#define TRANSCODER_C_OFFSET 0x62000
+#define CHV_TRANSCODER_C_OFFSET 0x63000
+#define TRANSCODER_D_OFFSET 0x63000
+#define TRANSCODER_EDP_OFFSET 0x6f000
+#define TRANSCODER_DSI0_OFFSET	0x6b000
+#define TRANSCODER_DSI1_OFFSET	0x6b800
+
+#define CURSOR_A_OFFSET 0x70080
+#define CURSOR_B_OFFSET 0x700c0
+#define CHV_CURSOR_C_OFFSET 0x700e0
+#define IVB_CURSOR_B_OFFSET 0x71080
+#define IVB_CURSOR_C_OFFSET 0x72080
+#define TGL_CURSOR_D_OFFSET 0x73080
+
+#define I845_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+	}
+
+#define I9XX_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+	}
+
+#define IVB_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+		[TRANSCODER_C] = PIPE_C_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
+	}
+
+#define HSW_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET,	\
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+		[TRANSCODER_C] = PIPE_C_OFFSET, \
+		[TRANSCODER_EDP] = PIPE_EDP_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
+		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \
+	}
+
+#define CHV_PIPE_OFFSETS \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET, \
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+		[TRANSCODER_C] = CHV_PIPE_C_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+		[TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \
+	}
+
+#define I845_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+	}
+
+#define I9XX_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = CURSOR_B_OFFSET, \
+	}
+
+#define CHV_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = CURSOR_B_OFFSET, \
+		[PIPE_C] = CHV_CURSOR_C_OFFSET, \
+	}
+
+#define IVB_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
+		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
+	}
+
+#define TGL_CURSOR_OFFSETS \
+	.cursor_offsets = { \
+		[PIPE_A] = CURSOR_A_OFFSET, \
+		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
+		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
+		[PIPE_D] = TGL_CURSOR_D_OFFSET, \
+	}
+
+#define I845_COLORS \
+	.color = { .gamma_lut_size = 256 }
+#define I9XX_COLORS \
+	.color = { .gamma_lut_size = 129, \
+		   .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
+	}
+#define ILK_COLORS \
+	.color = { .gamma_lut_size = 1024 }
+#define IVB_COLORS \
+	.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 }
+#define CHV_COLORS \
+	.color = { \
+		.degamma_lut_size = 65, .gamma_lut_size = 257, \
+		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
+		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
+	}
+#define GLK_COLORS \
+	.color = { \
+		.degamma_lut_size = 33, .gamma_lut_size = 1024, \
+		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
+				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
+	}
+#define ICL_COLORS \
+	.color = { \
+		.degamma_lut_size = 33, .gamma_lut_size = 262145, \
+		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
+				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
+		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
+	}
+
+#define I830_DISPLAY \
+	.has_overlay = 1, \
+	.cursor_needs_physical = 1, \
+	.overlay_needs_physical = 1, \
+	.has_gmch = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 2, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
+
+static const struct intel_display_device_info i830_display = {
+	I830_DISPLAY,
+};
+
+#define I845_DISPLAY \
+	.has_overlay = 1, \
+	.overlay_needs_physical = 1, \
+	.has_gmch = 1, \
+	I845_PIPE_OFFSETS, \
+	I845_CURSOR_OFFSETS, \
+	I845_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 2, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A), \
+	.__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A)
+
+static const struct intel_display_device_info i845_display = {
+	I845_DISPLAY,
+};
+
+static const struct intel_display_device_info i85x_display = {
+	I830_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info i865g_display = {
+	I845_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+#define GEN3_DISPLAY \
+	.has_gmch = 1, \
+	.has_overlay = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 3, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
+
+static const struct intel_display_device_info i915g_display = {
+	GEN3_DISPLAY,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+};
+
+static const struct intel_display_device_info i915gm_display = {
+	GEN3_DISPLAY,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info i945g_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+};
+
+static const struct intel_display_device_info i945gm_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+	.cursor_needs_physical = 1,
+	.overlay_needs_physical = 1,
+	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info g33_display = {
+	GEN3_DISPLAY,
+	.has_hotplug = 1,
+};
+
+#define GEN4_DISPLAY \
+	.has_hotplug = 1, \
+	.has_gmch = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	I9XX_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 4, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
+
+static const struct intel_display_device_info i965g_display = {
+	GEN4_DISPLAY,
+	.has_overlay = 1,
+};
+
+static const struct intel_display_device_info i965gm_display = {
+	GEN4_DISPLAY,
+	.has_overlay = 1,
+	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info g45_display = {
+	GEN4_DISPLAY,
+};
+
+static const struct intel_display_device_info gm45_display = {
+	GEN4_DISPLAY,
+	.supports_tv = 1,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+#define ILK_DISPLAY \
+	.has_hotplug = 1, \
+	I9XX_PIPE_OFFSETS, \
+	I9XX_CURSOR_OFFSETS, \
+	ILK_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 5, \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
+
+static const struct intel_display_device_info ilk_d_display = {
+	ILK_DISPLAY,
+};
+
+static const struct intel_display_device_info ilk_m_display = {
+	ILK_DISPLAY,
+
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info snb_display = {
+	.has_hotplug = 1,
+	I9XX_PIPE_OFFSETS,
+	I9XX_CURSOR_OFFSETS,
+	ILK_COLORS,
+
+	.__runtime_defaults.ip.ver = 6,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info ivb_display = {
+	.has_hotplug = 1,
+	IVB_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info vlv_display = {
+	.has_gmch = 1,
+	.has_hotplug = 1,
+	.mmio_offset = VLV_DISPLAY_BASE,
+	I9XX_PIPE_OFFSETS,
+	I9XX_CURSOR_OFFSETS,
+	I9XX_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
+};
+
+static const struct intel_display_device_info hsw_display = {
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 7,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info bdw_display = {
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 8,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+static const struct intel_display_device_info chv_display = {
+	.has_hotplug = 1,
+	.has_gmch = 1,
+	.mmio_offset = VLV_DISPLAY_BASE,
+	CHV_PIPE_OFFSETS,
+	CHV_CURSOR_OFFSETS,
+	CHV_COLORS,
+
+	.__runtime_defaults.ip.ver = 8,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
+};
+
+static const struct intel_display_device_info skl_display = {
+	.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */
+	.dbuf.slice_mask = BIT(DBUF_S1),
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	.has_ipc = 1,
+	.has_psr = 1,
+	.has_psr_hw_tracking = 1,
+	HSW_PIPE_OFFSETS,
+	IVB_CURSOR_OFFSETS,
+	IVB_COLORS,
+
+	.__runtime_defaults.ip.ver = 9,
+	.__runtime_defaults.has_dmc = 1,
+	.__runtime_defaults.has_hdcp = 1,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+#define GEN9_LP_DISPLAY \
+	.dbuf.slice_mask = BIT(DBUF_S1), \
+	.has_dp_mst = 1, \
+	.has_ddi = 1, \
+	.has_fpga_dbg = 1, \
+	.has_hotplug = 1, \
+	.has_ipc = 1, \
+	.has_psr = 1, \
+	.has_psr_hw_tracking = 1, \
+	HSW_PIPE_OFFSETS, \
+	IVB_CURSOR_OFFSETS, \
+	IVB_COLORS, \
+	\
+	.__runtime_defaults.has_dmc = 1, \
+	.__runtime_defaults.has_hdcp = 1, \
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), \
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
+		BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C)
+
+static const struct intel_display_device_info bxt_display = {
+	GEN9_LP_DISPLAY,
+	.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */
+
+	.__runtime_defaults.ip.ver = 9,
+};
+
+static const struct intel_display_device_info glk_display = {
+	GEN9_LP_DISPLAY,
+	.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */
+	GLK_COLORS,
+
+	.__runtime_defaults.ip.ver = 10,
+};
+
+static const struct intel_display_device_info gen11_display = {
+	.abox_mask = BIT(0),
+	.dbuf.size = 2048,
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2),
+	.has_ddi = 1,
+	.has_dp_mst = 1,
+	.has_fpga_dbg = 1,
+	.has_hotplug = 1,
+	.has_ipc = 1,
+	.has_psr = 1,
+	.has_psr_hw_tracking = 1,
+	.pipe_offsets = {
+		[TRANSCODER_A] = PIPE_A_OFFSET,
+		[TRANSCODER_B] = PIPE_B_OFFSET,
+		[TRANSCODER_C] = PIPE_C_OFFSET,
+		[TRANSCODER_EDP] = PIPE_EDP_OFFSET,
+		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,
+		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,
+	},
+	.trans_offsets = {
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET,
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET,
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET,
+		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET,
+		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,
+		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,
+	},
+	IVB_CURSOR_OFFSETS,
+	ICL_COLORS,
+
+	.__runtime_defaults.ip.ver = 11,
+	.__runtime_defaults.has_dmc = 1,
+	.__runtime_defaults.has_dsc = 1,
+	.__runtime_defaults.has_hdcp = 1,
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) |
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
+};
+
+#define XE_D_DISPLAY \
+	.abox_mask = GENMASK(2, 1), \
+	.dbuf.size = 2048, \
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \
+	.has_ddi = 1, \
+	.has_dp_mst = 1, \
+	.has_dsb = 1, \
+	.has_fpga_dbg = 1, \
+	.has_hotplug = 1, \
+	.has_ipc = 1, \
+	.has_psr = 1, \
+	.has_psr_hw_tracking = 1, \
+	.pipe_offsets = { \
+		[TRANSCODER_A] = PIPE_A_OFFSET, \
+		[TRANSCODER_B] = PIPE_B_OFFSET, \
+		[TRANSCODER_C] = PIPE_C_OFFSET, \
+		[TRANSCODER_D] = PIPE_D_OFFSET, \
+		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \
+		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \
+	}, \
+	.trans_offsets = { \
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
+		[TRANSCODER_D] = TRANSCODER_D_OFFSET, \
+		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \
+		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
+	}, \
+	TGL_CURSOR_OFFSETS, \
+	ICL_COLORS, \
+	\
+	.__runtime_defaults.ip.ver = 12, \
+	.__runtime_defaults.has_dmc = 1, \
+	.__runtime_defaults.has_dsc = 1, \
+	.__runtime_defaults.has_hdcp = 1, \
+	.__runtime_defaults.pipe_mask = \
+		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \
+	.__runtime_defaults.cpu_transcoder_mask = \
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A)
+
+static const struct intel_display_device_info tgl_display = {
+	XE_D_DISPLAY,
+};
+
+static const struct intel_display_device_info rkl_display = {
+	XE_D_DISPLAY,
+	.abox_mask = BIT(0),
+	.has_hti = 1,
+	.has_psr_hw_tracking = 0,
+
+	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
+};
+
+static const struct intel_display_device_info adl_s_display = {
+	XE_D_DISPLAY,
+	.has_hti = 1,
+	.has_psr_hw_tracking = 0,
+};
+
+#define XE_LPD_FEATURES \
+	.abox_mask = GENMASK(1, 0),						\
+	.color = {								\
+		.degamma_lut_size = 129, .gamma_lut_size = 1024,		\
+		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING |		\
+		DRM_COLOR_LUT_EQUAL_CHANNELS,					\
+	},									\
+	.dbuf.size = 4096,							\
+	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) |		\
+		BIT(DBUF_S4),							\
+	.has_ddi = 1,								\
+	.has_dp_mst = 1,							\
+	.has_dsb = 1,								\
+	.has_fpga_dbg = 1,							\
+	.has_hotplug = 1,							\
+	.has_ipc = 1,								\
+	.has_psr = 1,								\
+	.pipe_offsets = {							\
+		[TRANSCODER_A] = PIPE_A_OFFSET,					\
+		[TRANSCODER_B] = PIPE_B_OFFSET,					\
+		[TRANSCODER_C] = PIPE_C_OFFSET,					\
+		[TRANSCODER_D] = PIPE_D_OFFSET,					\
+		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,				\
+		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,				\
+	},									\
+	.trans_offsets = {							\
+		[TRANSCODER_A] = TRANSCODER_A_OFFSET,				\
+		[TRANSCODER_B] = TRANSCODER_B_OFFSET,				\
+		[TRANSCODER_C] = TRANSCODER_C_OFFSET,				\
+		[TRANSCODER_D] = TRANSCODER_D_OFFSET,				\
+		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,			\
+		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,			\
+	},									\
+	TGL_CURSOR_OFFSETS,							\
+										\
+	.__runtime_defaults.ip.ver = 13,					\
+	.__runtime_defaults.has_dmc = 1,					\
+	.__runtime_defaults.has_dsc = 1,					\
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),			\
+	.__runtime_defaults.has_hdcp = 1,					\
+	.__runtime_defaults.pipe_mask =						\
+		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D)
+
+static const struct intel_display_device_info xe_lpd_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_crawl = 1,
+	.has_psr_hw_tracking = 0,
+
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) |
+		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
+};
+
+static const struct intel_display_device_info xe_hpd_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_squash = 1,
+
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
+};
+
+static const struct intel_display_device_info xe_lpdp_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_crawl = 1,
+	.has_cdclk_squash = 1,
+
+	.__runtime_defaults.ip.ver = 14,
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
+};
+
+#undef INTEL_VGA_DEVICE
+#undef INTEL_QUANTA_VGA_DEVICE
+#define INTEL_VGA_DEVICE(id, info) { id, info }
+#define INTEL_QUANTA_VGA_DEVICE(info) { 0x16a, info }
+
+static const struct {
+	u32 devid;
+	const struct intel_display_device_info *info;
+} intel_display_ids[] = {
+	INTEL_I830_IDS(&i830_display),
+	INTEL_I845G_IDS(&i845_display),
+	INTEL_I85X_IDS(&i85x_display),
+	INTEL_I865G_IDS(&i865g_display),
+	INTEL_I915G_IDS(&i915g_display),
+	INTEL_I915GM_IDS(&i915gm_display),
+	INTEL_I945G_IDS(&i945g_display),
+	INTEL_I945GM_IDS(&i945gm_display),
+	INTEL_I965G_IDS(&i965g_display),
+	INTEL_G33_IDS(&g33_display),
+	INTEL_I965GM_IDS(&i965gm_display),
+	INTEL_GM45_IDS(&gm45_display),
+	INTEL_G45_IDS(&g45_display),
+	INTEL_PINEVIEW_G_IDS(&g33_display),
+	INTEL_PINEVIEW_M_IDS(&g33_display),
+	INTEL_IRONLAKE_D_IDS(&ilk_d_display),
+	INTEL_IRONLAKE_M_IDS(&ilk_m_display),
+	INTEL_SNB_D_IDS(&snb_display),
+	INTEL_SNB_M_IDS(&snb_display),
+	INTEL_IVB_Q_IDS(NULL),		/* must be first IVB in list */
+	INTEL_IVB_M_IDS(&ivb_display),
+	INTEL_IVB_D_IDS(&ivb_display),
+	INTEL_HSW_IDS(&hsw_display),
+	INTEL_VLV_IDS(&vlv_display),
+	INTEL_BDW_IDS(&bdw_display),
+	INTEL_CHV_IDS(&chv_display),
+	INTEL_SKL_IDS(&skl_display),
+	INTEL_BXT_IDS(&bxt_display),
+	INTEL_GLK_IDS(&glk_display),
+	INTEL_KBL_IDS(&skl_display),
+	INTEL_CFL_IDS(&skl_display),
+	INTEL_ICL_11_IDS(&gen11_display),
+	INTEL_EHL_IDS(&gen11_display),
+	INTEL_JSL_IDS(&gen11_display),
+	INTEL_TGL_12_IDS(&tgl_display),
+	INTEL_DG1_IDS(&tgl_display),
+	INTEL_RKL_IDS(&rkl_display),
+	INTEL_ADLS_IDS(&adl_s_display),
+	INTEL_RPLS_IDS(&adl_s_display),
+	INTEL_ADLP_IDS(&xe_lpd_display),
+	INTEL_ADLN_IDS(&xe_lpd_display),
+	INTEL_RPLP_IDS(&xe_lpd_display),
+	INTEL_DG2_IDS(&xe_hpd_display),
+
+	/* FIXME: Replace this with a GMD_ID lookup */
+	INTEL_MTL_IDS(&xe_lpdp_display),
+};
+
+const struct intel_display_device_info *
+intel_display_device_probe(u16 pci_devid)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(intel_display_ids); i++) {
+		if (intel_display_ids[i].devid == pci_devid)
+			return intel_display_ids[i].info;
+	}
+
+	return &no_display;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 613607fad5af0..1f7d08b3ad6b1 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -80,4 +80,7 @@ struct intel_display_device_info {
 	} color;
 };
 
+const struct intel_display_device_info *
+intel_display_device_probe(u16 pci_devid);
+
 #endif
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 261569e15f91b..34bc732a6375c 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -42,127 +42,6 @@
 	.__runtime.graphics.ip.ver = (x), \
 	.__runtime.media.ip.ver = (x)
 
-static const struct intel_display_device_info no_display = {};
-
-#define I845_PIPE_OFFSETS \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET,	\
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-	}
-
-#define I9XX_PIPE_OFFSETS \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET,	\
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-	}
-
-#define IVB_PIPE_OFFSETS \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET,	\
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-		[TRANSCODER_C] = PIPE_C_OFFSET, \
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
-	}
-
-#define HSW_PIPE_OFFSETS \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET,	\
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-		[TRANSCODER_C] = PIPE_C_OFFSET, \
-		[TRANSCODER_EDP] = PIPE_EDP_OFFSET, \
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
-		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \
-	}
-
-#define CHV_PIPE_OFFSETS \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET, \
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-		[TRANSCODER_C] = CHV_PIPE_C_OFFSET, \
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-		[TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \
-	}
-
-#define I845_CURSOR_OFFSETS \
-	.cursor_offsets = { \
-		[PIPE_A] = CURSOR_A_OFFSET, \
-	}
-
-#define I9XX_CURSOR_OFFSETS \
-	.cursor_offsets = { \
-		[PIPE_A] = CURSOR_A_OFFSET, \
-		[PIPE_B] = CURSOR_B_OFFSET, \
-	}
-
-#define CHV_CURSOR_OFFSETS \
-	.cursor_offsets = { \
-		[PIPE_A] = CURSOR_A_OFFSET, \
-		[PIPE_B] = CURSOR_B_OFFSET, \
-		[PIPE_C] = CHV_CURSOR_C_OFFSET, \
-	}
-
-#define IVB_CURSOR_OFFSETS \
-	.cursor_offsets = { \
-		[PIPE_A] = CURSOR_A_OFFSET, \
-		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
-		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
-	}
-
-#define TGL_CURSOR_OFFSETS \
-	.cursor_offsets = { \
-		[PIPE_A] = CURSOR_A_OFFSET, \
-		[PIPE_B] = IVB_CURSOR_B_OFFSET, \
-		[PIPE_C] = IVB_CURSOR_C_OFFSET, \
-		[PIPE_D] = TGL_CURSOR_D_OFFSET, \
-	}
-
-#define I845_COLORS \
-	.color = { .gamma_lut_size = 256 }
-#define I9XX_COLORS \
-	.color = { .gamma_lut_size = 129, \
-		   .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
-	}
-#define ILK_COLORS \
-	.color = { .gamma_lut_size = 1024 }
-#define IVB_COLORS \
-	.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 }
-#define CHV_COLORS \
-	.color = { \
-		.degamma_lut_size = 65, .gamma_lut_size = 257, \
-		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
-		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
-	}
-#define GLK_COLORS \
-	.color = { \
-		.degamma_lut_size = 33, .gamma_lut_size = 1024, \
-		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
-				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
-	}
-#define ICL_COLORS \
-	.color = { \
-		.degamma_lut_size = 33, .gamma_lut_size = 262145, \
-		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \
-				     DRM_COLOR_LUT_EQUAL_CHANNELS, \
-		.gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \
-	}
-
 /* Keep in gen based order, and chronological order within a gen */
 
 #define GEN_DEFAULT_PAGE_SIZES \
@@ -171,24 +50,6 @@ static const struct intel_display_device_info no_display = {};
 #define GEN_DEFAULT_REGIONS \
 	.__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_SMEM
 
-#define I830_DISPLAY \
-	.has_overlay = 1, \
-	.cursor_needs_physical = 1, \
-	.overlay_needs_physical = 1, \
-	.has_gmch = 1, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 2, \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
-
-static const struct intel_display_device_info i830_display = {
-	I830_DISPLAY,
-};
-
 #define I830_FEATURES \
 	GEN(2), \
 	.is_mobile = 1, \
@@ -203,22 +64,6 @@ static const struct intel_display_device_info i830_display = {
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
-#define I845_DISPLAY \
-	.has_overlay = 1, \
-	.overlay_needs_physical = 1, \
-	.has_gmch = 1, \
-	I845_PIPE_OFFSETS, \
-	I845_CURSOR_OFFSETS, \
-	I845_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 2, \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A), \
-	.__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A)
-
-static const struct intel_display_device_info i845_display = {
-	I845_DISPLAY,
-};
-
 #define I845_FEATURES \
 	GEN(2), \
 	.has_3d_pipeline = 1, \
@@ -235,86 +80,21 @@ static const struct intel_display_device_info i845_display = {
 static const struct intel_device_info i830_info = {
 	I830_FEATURES,
 	PLATFORM(INTEL_I830),
-	.display = &i830_display,
 };
 
 static const struct intel_device_info i845g_info = {
 	I845_FEATURES,
 	PLATFORM(INTEL_I845G),
-	.display = &i845_display,
-};
-
-static const struct intel_display_device_info i85x_display = {
-	I830_DISPLAY,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info i85x_info = {
 	I830_FEATURES,
 	PLATFORM(INTEL_I85X),
-	.display = &i85x_display,
-};
-
-static const struct intel_display_device_info i865g_display = {
-	I845_DISPLAY,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
 };
 
 static const struct intel_device_info i865g_info = {
 	I845_FEATURES,
 	PLATFORM(INTEL_I865G),
-	.display = &i865g_display,
-};
-
-#define GEN3_DISPLAY \
-	.has_gmch = 1, \
-	.has_overlay = 1, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 3, \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
-
-static const struct intel_display_device_info i915g_display = {
-	GEN3_DISPLAY,
-	.cursor_needs_physical = 1,
-	.overlay_needs_physical = 1,
-};
-
-static const struct intel_display_device_info i915gm_display = {
-	GEN3_DISPLAY,
-	.cursor_needs_physical = 1,
-	.overlay_needs_physical = 1,
-	.supports_tv = 1,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
-static const struct intel_display_device_info i945g_display = {
-	GEN3_DISPLAY,
-	.has_hotplug = 1,
-	.cursor_needs_physical = 1,
-	.overlay_needs_physical = 1,
-};
-
-static const struct intel_display_device_info i945gm_display = {
-	GEN3_DISPLAY,
-	.has_hotplug = 1,
-	.cursor_needs_physical = 1,
-	.overlay_needs_physical = 1,
-	.supports_tv = 1,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
-static const struct intel_display_device_info g33_display = {
-	GEN3_DISPLAY,
-	.has_hotplug = 1,
 };
 
 #define GEN3_FEATURES \
@@ -331,7 +111,6 @@ static const struct intel_display_device_info g33_display = {
 static const struct intel_device_info i915g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I915G),
-	.display = &i915g_display,
 	.has_coherent_ggtt = false,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
@@ -340,7 +119,6 @@ static const struct intel_device_info i915g_info = {
 static const struct intel_device_info i915gm_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I915GM),
-	.display = &i915gm_display,
 	.is_mobile = 1,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
@@ -349,7 +127,6 @@ static const struct intel_device_info i915gm_info = {
 static const struct intel_device_info i945g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I945G),
-	.display = &i945g_display,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
 };
@@ -357,7 +134,6 @@ static const struct intel_device_info i945g_info = {
 static const struct intel_device_info i945gm_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_I945GM),
-	.display = &i945gm_display,
 	.is_mobile = 1,
 	.hws_needs_physical = 1,
 	.unfenced_needs_alignment = 1,
@@ -366,14 +142,12 @@ static const struct intel_device_info i945gm_info = {
 static const struct intel_device_info g33_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_G33),
-	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
 static const struct intel_device_info pnv_g_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_PINEVIEW),
-	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
@@ -381,46 +155,9 @@ static const struct intel_device_info pnv_m_info = {
 	GEN3_FEATURES,
 	PLATFORM(INTEL_PINEVIEW),
 	.is_mobile = 1,
-	.display = &g33_display,
 	.dma_mask_size = 36,
 };
 
-#define GEN4_DISPLAY \
-	.has_hotplug = 1, \
-	.has_gmch = 1, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 4, \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
-
-static const struct intel_display_device_info i965g_display = {
-	GEN4_DISPLAY,
-	.has_overlay = 1,
-};
-
-static const struct intel_display_device_info i965gm_display = {
-	GEN4_DISPLAY,
-	.has_overlay = 1,
-	.supports_tv = 1,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
-static const struct intel_display_device_info g45_display = {
-	GEN4_DISPLAY,
-};
-
-static const struct intel_display_device_info gm45_display = {
-	GEN4_DISPLAY,
-	.supports_tv = 1,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 #define GEN4_FEATURES \
 	GEN(4), \
 	.gpu_reset_clobbers_display = true, \
@@ -435,7 +172,6 @@ static const struct intel_display_device_info gm45_display = {
 static const struct intel_device_info i965g_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_I965G),
-	.display = &i965g_display,
 	.hws_needs_physical = 1,
 	.has_snoop = false,
 };
@@ -443,7 +179,6 @@ static const struct intel_device_info i965g_info = {
 static const struct intel_device_info i965gm_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_I965GM),
-	.display = &i965gm_display,
 	.is_mobile = 1,
 	.hws_needs_physical = 1,
 	.has_snoop = false,
@@ -453,7 +188,6 @@ static const struct intel_device_info g45_info = {
 	GEN4_FEATURES,
 	PLATFORM(INTEL_G45),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0),
-	.display = &g45_display,
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -462,7 +196,6 @@ static const struct intel_device_info gm45_info = {
 	PLATFORM(INTEL_GM45),
 	.is_mobile = 1,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0),
-	.display = &gm45_display,
 	.gpu_reset_clobbers_display = false,
 };
 
@@ -478,37 +211,14 @@ static const struct intel_device_info gm45_info = {
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
-#define ILK_DISPLAY \
-	.has_hotplug = 1, \
-	I9XX_PIPE_OFFSETS, \
-	I9XX_CURSOR_OFFSETS, \
-	ILK_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 5, \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B)
-
-static const struct intel_display_device_info ilk_d_display = {
-	ILK_DISPLAY,
-};
-
 static const struct intel_device_info ilk_d_info = {
 	GEN5_FEATURES,
 	PLATFORM(INTEL_IRONLAKE),
-	.display = &ilk_d_display,
 };
 
-static const struct intel_display_device_info ilk_m_display = {
-	ILK_DISPLAY,
-
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
- };
-
 static const struct intel_device_info ilk_m_info = {
 	GEN5_FEATURES,
 	PLATFORM(INTEL_IRONLAKE),
-	.display = &ilk_m_display,
 	.is_mobile = 1,
 	.has_rps = true,
 };
@@ -529,32 +239,17 @@ static const struct intel_device_info ilk_m_info = {
 	GEN_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
-static const struct intel_display_device_info snb_display = {
-	.has_hotplug = 1,
-	I9XX_PIPE_OFFSETS,
-	I9XX_CURSOR_OFFSETS,
-	ILK_COLORS,
-
-	.__runtime_defaults.ip.ver = 6,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 #define SNB_D_PLATFORM \
 	GEN6_FEATURES, \
 	PLATFORM(INTEL_SANDYBRIDGE)
 
 static const struct intel_device_info snb_d_gt1_info = {
 	SNB_D_PLATFORM,
-	.display = &snb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info snb_d_gt2_info = {
 	SNB_D_PLATFORM,
-	.display = &snb_display,
 	.gt = 2,
 };
 
@@ -566,13 +261,11 @@ static const struct intel_device_info snb_d_gt2_info = {
 
 static const struct intel_device_info snb_m_gt1_info = {
 	SNB_M_PLATFORM,
-	.display = &snb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info snb_m_gt2_info = {
 	SNB_M_PLATFORM,
-	.display = &snb_display,
 	.gt = 2,
 };
 
@@ -597,28 +290,13 @@ static const struct intel_device_info snb_m_gt2_info = {
 	PLATFORM(INTEL_IVYBRIDGE), \
 	.has_l3_dpf = 1
 
-static const struct intel_display_device_info ivb_display = {
-	.has_hotplug = 1,
-	IVB_PIPE_OFFSETS,
-	IVB_CURSOR_OFFSETS,
-	IVB_COLORS,
-
-	.__runtime_defaults.ip.ver = 7,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 static const struct intel_device_info ivb_d_gt1_info = {
 	IVB_D_PLATFORM,
-	.display = &ivb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info ivb_d_gt2_info = {
 	IVB_D_PLATFORM,
-	.display = &ivb_display,
 	.gt = 2,
 };
 
@@ -630,43 +308,25 @@ static const struct intel_device_info ivb_d_gt2_info = {
 
 static const struct intel_device_info ivb_m_gt1_info = {
 	IVB_M_PLATFORM,
-	.display = &ivb_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info ivb_m_gt2_info = {
 	IVB_M_PLATFORM,
-	.display = &ivb_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info ivb_q_info = {
 	GEN7_FEATURES,
 	PLATFORM(INTEL_IVYBRIDGE),
-	.display = &no_display,
 	.gt = 2,
 	.has_l3_dpf = 1,
 };
 
-static const struct intel_display_device_info vlv_display = {
-	.has_gmch = 1,
-	.has_hotplug = 1,
-	.mmio_offset = VLV_DISPLAY_BASE,
-	I9XX_PIPE_OFFSETS,
-	I9XX_CURSOR_OFFSETS,
-	I9XX_COLORS,
-
-	.__runtime_defaults.ip.ver = 7,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B),
-};
-
 static const struct intel_device_info vlv_info = {
 	PLATFORM(INTEL_VALLEYVIEW),
 	GEN(7),
 	.is_lp = 1,
-	.display = &vlv_display,
 	.has_runtime_pm = 1,
 	.has_rc6 = 1,
 	.has_reset_engine = true,
@@ -692,38 +352,18 @@ static const struct intel_device_info vlv_info = {
 	PLATFORM(INTEL_HASWELL), \
 	.has_l3_dpf = 1
 
-static const struct intel_display_device_info hsw_display = {
-	.has_ddi = 1,
-	.has_dp_mst = 1,
-	.has_fpga_dbg = 1,
-	.has_hotplug = 1,
-	HSW_PIPE_OFFSETS,
-	IVB_CURSOR_OFFSETS,
-	IVB_COLORS,
-
-	.__runtime_defaults.ip.ver = 7,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 static const struct intel_device_info hsw_gt1_info = {
 	HSW_PLATFORM,
-	.display = &hsw_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info hsw_gt2_info = {
 	HSW_PLATFORM,
-	.display = &hsw_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info hsw_gt3_info = {
 	HSW_PLATFORM,
-	.display = &hsw_display,
 	.gt = 3,
 };
 
@@ -736,42 +376,22 @@ static const struct intel_device_info hsw_gt3_info = {
 	.__runtime.ppgtt_size = 48, \
 	.has_64bit_reloc = 1
 
-static const struct intel_display_device_info bdw_display = {
-	.has_ddi = 1,
-	.has_dp_mst = 1,
-	.has_fpga_dbg = 1,
-	.has_hotplug = 1,
-	HSW_PIPE_OFFSETS,
-	IVB_CURSOR_OFFSETS,
-	IVB_COLORS,
-
-	.__runtime_defaults.ip.ver = 8,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 #define BDW_PLATFORM \
 	GEN8_FEATURES, \
 	PLATFORM(INTEL_BROADWELL)
 
 static const struct intel_device_info bdw_gt1_info = {
 	BDW_PLATFORM,
-	.display = &bdw_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info bdw_gt2_info = {
 	BDW_PLATFORM,
-	.display = &bdw_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info bdw_rsvd_info = {
 	BDW_PLATFORM,
-	.display = &bdw_display,
 	.gt = 3,
 	/* According to the device ID those devices are GT3, they were
 	 * previously treated as not GT3, keep it like that.
@@ -780,30 +400,14 @@ static const struct intel_device_info bdw_rsvd_info = {
 
 static const struct intel_device_info bdw_gt3_info = {
 	BDW_PLATFORM,
-	.display = &bdw_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
 };
 
-static const struct intel_display_device_info chv_display = {
-	.has_hotplug = 1,
-	.has_gmch = 1,
-	.mmio_offset = VLV_DISPLAY_BASE,
-	CHV_PIPE_OFFSETS,
-	CHV_CURSOR_OFFSETS,
-	CHV_COLORS,
-
-	.__runtime_defaults.ip.ver = 8,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
-};
-
 static const struct intel_device_info chv_info = {
 	PLATFORM(INTEL_CHERRYVIEW),
 	GEN(8),
-	.display = &chv_display,
 	.is_lp = 1,
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0),
 	.has_64bit_reloc = 1,
@@ -831,43 +435,17 @@ static const struct intel_device_info chv_info = {
 	GEN9_DEFAULT_PAGE_SIZES, \
 	.has_gt_uc = 1
 
-static const struct intel_display_device_info skl_display = {
-	.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */
-	.dbuf.slice_mask = BIT(DBUF_S1),
-	.has_ddi = 1,
-	.has_dp_mst = 1,
-	.has_fpga_dbg = 1,
-	.has_hotplug = 1,
-	.has_ipc = 1,
-	.has_psr = 1,
-	.has_psr_hw_tracking = 1,
-	HSW_PIPE_OFFSETS,
-	IVB_CURSOR_OFFSETS,
-	IVB_COLORS,
-
-	.__runtime_defaults.ip.ver = 9,
-	.__runtime_defaults.has_dmc = 1,
-	.__runtime_defaults.has_hdcp = 1,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 #define SKL_PLATFORM \
 	GEN9_FEATURES, \
 	PLATFORM(INTEL_SKYLAKE)
 
 static const struct intel_device_info skl_gt1_info = {
 	SKL_PLATFORM,
-	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info skl_gt2_info = {
 	SKL_PLATFORM,
-	.display = &skl_display,
 	.gt = 2,
 };
 
@@ -879,13 +457,11 @@ static const struct intel_device_info skl_gt2_info = {
 
 static const struct intel_device_info skl_gt3_info = {
 	SKL_GT3_PLUS_PLATFORM,
-	.display = &skl_display,
 	.gt = 3,
 };
 
 static const struct intel_device_info skl_gt4_info = {
 	SKL_GT3_PLUS_PLATFORM,
-	.display = &skl_display,
 	.gt = 4,
 };
 
@@ -909,53 +485,14 @@ static const struct intel_device_info skl_gt4_info = {
 	GEN9_DEFAULT_PAGE_SIZES, \
 	GEN_DEFAULT_REGIONS
 
-#define GEN9_LP_DISPLAY \
-	.dbuf.slice_mask = BIT(DBUF_S1), \
-	.has_dp_mst = 1, \
-	.has_ddi = 1, \
-	.has_fpga_dbg = 1, \
-	.has_hotplug = 1, \
-	.has_ipc = 1, \
-	.has_psr = 1, \
-	.has_psr_hw_tracking = 1, \
-	HSW_PIPE_OFFSETS, \
-	IVB_CURSOR_OFFSETS, \
-	IVB_COLORS, \
-	\
-	.__runtime_defaults.has_dmc = 1, \
-	.__runtime_defaults.has_hdcp = 1, \
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), \
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \
-		BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C)
-
-static const struct intel_display_device_info bxt_display = {
-	GEN9_LP_DISPLAY,
-	.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */
-
-	.__runtime_defaults.ip.ver = 9,
-};
-
 static const struct intel_device_info bxt_info = {
 	GEN9_LP_FEATURES,
 	PLATFORM(INTEL_BROXTON),
-	.display = &bxt_display,
-};
-
-static const struct intel_display_device_info glk_display = {
-	GEN9_LP_DISPLAY,
-	.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */
-	GLK_COLORS,
-
-	.__runtime_defaults.ip.ver = 10,
 };
 
 static const struct intel_device_info glk_info = {
 	GEN9_LP_FEATURES,
 	PLATFORM(INTEL_GEMINILAKE),
-	.display = &glk_display,
 };
 
 #define KBL_PLATFORM \
@@ -964,19 +501,16 @@ static const struct intel_device_info glk_info = {
 
 static const struct intel_device_info kbl_gt1_info = {
 	KBL_PLATFORM,
-	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info kbl_gt2_info = {
 	KBL_PLATFORM,
-	.display = &skl_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info kbl_gt3_info = {
 	KBL_PLATFORM,
-	.display = &skl_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
@@ -988,19 +522,16 @@ static const struct intel_device_info kbl_gt3_info = {
 
 static const struct intel_device_info cfl_gt1_info = {
 	CFL_PLATFORM,
-	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info cfl_gt2_info = {
 	CFL_PLATFORM,
-	.display = &skl_display,
 	.gt = 2,
 };
 
 static const struct intel_device_info cfl_gt3_info = {
 	CFL_PLATFORM,
-	.display = &skl_display,
 	.gt = 3,
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1),
@@ -1012,13 +543,11 @@ static const struct intel_device_info cfl_gt3_info = {
 
 static const struct intel_device_info cml_gt1_info = {
 	CML_PLATFORM,
-	.display = &skl_display,
 	.gt = 1,
 };
 
 static const struct intel_device_info cml_gt2_info = {
 	CML_PLATFORM,
-	.display = &skl_display,
 	.gt = 2,
 };
 
@@ -1034,54 +563,11 @@ static const struct intel_device_info cml_gt2_info = {
 	.has_coherent_ggtt = false, \
 	.has_logical_ring_elsq = 1
 
-static const struct intel_display_device_info gen11_display = {
-	.abox_mask = BIT(0),
-	.dbuf.size = 2048,
-	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2),
-	.has_ddi = 1,
-	.has_dp_mst = 1,
-	.has_fpga_dbg = 1,
-	.has_hotplug = 1,
-	.has_ipc = 1,
-	.has_psr = 1,
-	.has_psr_hw_tracking = 1,
-	.pipe_offsets = {
-		[TRANSCODER_A] = PIPE_A_OFFSET,
-		[TRANSCODER_B] = PIPE_B_OFFSET,
-		[TRANSCODER_C] = PIPE_C_OFFSET,
-		[TRANSCODER_EDP] = PIPE_EDP_OFFSET,
-		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,
-		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,
-	},
-	.trans_offsets = {
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET,
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET,
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET,
-		[TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET,
-		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,
-		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,
-	},
-	IVB_CURSOR_OFFSETS,
-	ICL_COLORS,
-
-	.__runtime_defaults.ip.ver = 11,
-	.__runtime_defaults.has_dmc = 1,
-	.__runtime_defaults.has_dsc = 1,
-	.__runtime_defaults.has_hdcp = 1,
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) |
-		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),
-};
-
 static const struct intel_device_info icl_info = {
 	GEN11_FEATURES,
 	PLATFORM(INTEL_ICELAKE),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
-	.display = &gen11_display,
 };
 
 static const struct intel_device_info ehl_info = {
@@ -1089,7 +575,6 @@ static const struct intel_device_info ehl_info = {
 	PLATFORM(INTEL_ELKHARTLAKE),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0),
 	.__runtime.ppgtt_size = 36,
-	.display = &gen11_display,
 };
 
 static const struct intel_device_info jsl_info = {
@@ -1097,7 +582,6 @@ static const struct intel_device_info jsl_info = {
 	PLATFORM(INTEL_JASPERLAKE),
 	.__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0),
 	.__runtime.ppgtt_size = 36,
-	.display = &gen11_display,
 };
 
 #define GEN12_FEATURES \
@@ -1106,70 +590,11 @@ static const struct intel_device_info jsl_info = {
 	.has_global_mocs = 1, \
 	.has_pxp = 1
 
-#define XE_D_DISPLAY \
-	.abox_mask = GENMASK(2, 1), \
-	.dbuf.size = 2048, \
-	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \
-	.has_ddi = 1, \
-	.has_dp_mst = 1, \
-	.has_dsb = 1, \
-	.has_fpga_dbg = 1, \
-	.has_hotplug = 1, \
-	.has_ipc = 1, \
-	.has_psr = 1, \
-	.has_psr_hw_tracking = 1, \
-	.pipe_offsets = { \
-		[TRANSCODER_A] = PIPE_A_OFFSET, \
-		[TRANSCODER_B] = PIPE_B_OFFSET, \
-		[TRANSCODER_C] = PIPE_C_OFFSET, \
-		[TRANSCODER_D] = PIPE_D_OFFSET, \
-		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \
-		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \
-	}, \
-	.trans_offsets = { \
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET, \
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET, \
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET, \
-		[TRANSCODER_D] = TRANSCODER_D_OFFSET, \
-		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \
-		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \
-	}, \
-	TGL_CURSOR_OFFSETS, \
-	ICL_COLORS, \
-	\
-	.__runtime_defaults.ip.ver = 12, \
-	.__runtime_defaults.has_dmc = 1, \
-	.__runtime_defaults.has_dsc = 1, \
-	.__runtime_defaults.has_hdcp = 1, \
-	.__runtime_defaults.pipe_mask = \
-		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \
-	.__runtime_defaults.cpu_transcoder_mask = \
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \
-		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A)
-
-static const struct intel_display_device_info tgl_display = {
-	XE_D_DISPLAY,
-};
-
 static const struct intel_device_info tgl_info = {
 	GEN12_FEATURES,
 	PLATFORM(INTEL_TIGERLAKE),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
-	.display = &tgl_display,
-};
-
-static const struct intel_display_device_info rkl_display = {
-	XE_D_DISPLAY,
-	.abox_mask = BIT(0),
-	.has_hti = 1,
-	.has_psr_hw_tracking = 0,
-
-	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
 };
 
 static const struct intel_device_info rkl_info = {
@@ -1177,7 +602,6 @@ static const struct intel_device_info rkl_info = {
 	PLATFORM(INTEL_ROCKETLAKE),
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0),
-	.display = &rkl_display,
 };
 
 #define DGFX_FEATURES \
@@ -1199,13 +623,6 @@ static const struct intel_device_info dg1_info = {
 		BIT(VCS0) | BIT(VCS2),
 	/* Wa_16011227922 */
 	.__runtime.ppgtt_size = 47,
-	.display = &tgl_display,
-};
-
-static const struct intel_display_device_info adl_s_display = {
-	XE_D_DISPLAY,
-	.has_hti = 1,
-	.has_psr_hw_tracking = 0,
 };
 
 static const struct intel_device_info adl_s_info = {
@@ -1214,61 +631,6 @@ static const struct intel_device_info adl_s_info = {
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.dma_mask_size = 39,
-	.display = &adl_s_display,
-};
-
-#define XE_LPD_FEATURES \
-	.abox_mask = GENMASK(1, 0),						\
-	.color = {								\
-		.degamma_lut_size = 129, .gamma_lut_size = 1024,		\
-		.degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING |		\
-		DRM_COLOR_LUT_EQUAL_CHANNELS,					\
-	},									\
-	.dbuf.size = 4096,							\
-	.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) |		\
-		BIT(DBUF_S4),							\
-	.has_ddi = 1,								\
-	.has_dp_mst = 1,							\
-	.has_dsb = 1,								\
-	.has_fpga_dbg = 1,							\
-	.has_hotplug = 1,							\
-	.has_ipc = 1,								\
-	.has_psr = 1,								\
-	.pipe_offsets = {							\
-		[TRANSCODER_A] = PIPE_A_OFFSET,					\
-		[TRANSCODER_B] = PIPE_B_OFFSET,					\
-		[TRANSCODER_C] = PIPE_C_OFFSET,					\
-		[TRANSCODER_D] = PIPE_D_OFFSET,					\
-		[TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET,				\
-		[TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET,				\
-	},									\
-	.trans_offsets = {						\
-		[TRANSCODER_A] = TRANSCODER_A_OFFSET,				\
-		[TRANSCODER_B] = TRANSCODER_B_OFFSET,				\
-		[TRANSCODER_C] = TRANSCODER_C_OFFSET,				\
-		[TRANSCODER_D] = TRANSCODER_D_OFFSET,				\
-		[TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET,			\
-		[TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET,			\
-	},									\
-	TGL_CURSOR_OFFSETS,							\
-										\
-	.__runtime_defaults.ip.ver = 13,					\
-	.__runtime_defaults.has_dmc = 1,					\
-	.__runtime_defaults.has_dsc = 1,					\
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A),			\
-	.__runtime_defaults.has_hdcp = 1,					\
-	.__runtime_defaults.pipe_mask =						\
-		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D)
-
-static const struct intel_display_device_info xe_lpd_display = {
-	XE_LPD_FEATURES,
-	.has_cdclk_crawl = 1,
-	.has_psr_hw_tracking = 0,
-
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D) |
-		BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1),
 };
 
 static const struct intel_device_info adl_p_info = {
@@ -1277,7 +639,6 @@ static const struct intel_device_info adl_p_info = {
 	.__runtime.platform_engine_mask =
 		BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2),
 	.__runtime.ppgtt_size = 48,
-	.display = &xe_lpd_display,
 	.dma_mask_size = 39,
 };
 
@@ -1323,7 +684,6 @@ static const struct intel_device_info xehpsdv_info = {
 	XE_HPM_FEATURES,
 	DGFX_FEATURES,
 	PLATFORM(INTEL_XEHPSDV),
-	.display = &no_display,
 	.has_64k_pages = 1,
 	.has_media_ratio_mode = 1,
 	.__runtime.platform_engine_mask =
@@ -1352,23 +712,12 @@ static const struct intel_device_info xehpsdv_info = {
 		BIT(VCS0) | BIT(VCS2) | \
 		BIT(CCS0) | BIT(CCS1) | BIT(CCS2) | BIT(CCS3)
 
-static const struct intel_display_device_info xe_hpd_display = {
-	XE_LPD_FEATURES,
-	.has_cdclk_squash = 1,
-
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
-};
-
 static const struct intel_device_info dg2_info = {
 	DG2_FEATURES,
-	.display = &xe_hpd_display,
 };
 
 static const struct intel_device_info ats_m_info = {
 	DG2_FEATURES,
-	.display = &no_display,
 	.require_force_probe = 1,
 	.tuning_thread_rr_after_dep = 1,
 };
@@ -1390,7 +739,6 @@ static const struct intel_device_info pvc_info = {
 	.__runtime.graphics.ip.rel = 60,
 	.__runtime.media.ip.rel = 60,
 	PLATFORM(INTEL_PONTEVECCHIO),
-	.display = &no_display,
 	.has_flat_ccs = 0,
 	.__runtime.platform_engine_mask =
 		BIT(BCS0) |
@@ -1409,18 +757,6 @@ static const struct intel_gt_definition xelpmp_extra_gt[] = {
 	{}
 };
 
-static const struct intel_display_device_info xe_lpdp_display = {
-	XE_LPD_FEATURES,
-	.has_cdclk_crawl = 1,
-	.has_cdclk_squash = 1,
-
-	.__runtime_defaults.ip.ver = 14,
-	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B),
-	.__runtime_defaults.cpu_transcoder_mask =
-		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
-		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
-};
-
 static const struct intel_device_info mtl_info = {
 	XE_HP_FEATURES,
 	/*
@@ -1431,7 +767,6 @@ static const struct intel_device_info mtl_info = {
 	.__runtime.graphics.ip.rel = 70,
 	.__runtime.media.ip.ver = 13,
 	PLATFORM(INTEL_METEORLAKE),
-	.display = &xe_lpdp_display,
 	.extra_gt_list = xelpmp_extra_gt,
 	.has_flat_ccs = 0,
 	.has_gmd_id = 1,
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 2a9ab8de84210..f1ba1eae26ca5 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1966,15 +1966,6 @@
 #define _TRANS_VSYNC_DSI1	0x6b814
 #define _TRANS_VSYNCSHIFT_DSI1	0x6b828
 
-#define TRANSCODER_A_OFFSET 0x60000
-#define TRANSCODER_B_OFFSET 0x61000
-#define TRANSCODER_C_OFFSET 0x62000
-#define CHV_TRANSCODER_C_OFFSET 0x63000
-#define TRANSCODER_D_OFFSET 0x63000
-#define TRANSCODER_EDP_OFFSET 0x6f000
-#define TRANSCODER_DSI0_OFFSET	0x6b000
-#define TRANSCODER_DSI1_OFFSET	0x6b800
-
 #define TRANS_HTOTAL(trans)	_MMIO_TRANS2((trans), _TRANS_HTOTAL_A)
 #define TRANS_HBLANK(trans)	_MMIO_TRANS2((trans), _TRANS_HBLANK_A)
 #define TRANS_HSYNC(trans)	_MMIO_TRANS2((trans), _TRANS_HSYNC_A)
@@ -2622,23 +2613,6 @@
 #define PIPESTAT_INT_ENABLE_MASK		0x7fff0000
 #define PIPESTAT_INT_STATUS_MASK		0x0000ffff
 
-#define PIPE_A_OFFSET		0x70000
-#define PIPE_B_OFFSET		0x71000
-#define PIPE_C_OFFSET		0x72000
-#define PIPE_D_OFFSET		0x73000
-#define CHV_PIPE_C_OFFSET	0x74000
-/*
- * There's actually no pipe EDP. Some pipe registers have
- * simply shifted from the pipe to the transcoder, while
- * keeping their original offset. Thus we need PIPE_EDP_OFFSET
- * to access such registers in transcoder EDP.
- */
-#define PIPE_EDP_OFFSET	0x7f000
-
-/* ICL DSI 0 and 1 */
-#define PIPE_DSI0_OFFSET	0x7b000
-#define PIPE_DSI1_OFFSET	0x7b800
-
 #define TRANSCONF(trans)	_MMIO_PIPE2((trans), _TRANSACONF)
 #define PIPEDSL(pipe)		_MMIO_PIPE2(pipe, _PIPEADSL)
 #define PIPEFRAME(pipe)		_MMIO_PIPE2(pipe, _PIPEAFRAMEHIGH)
@@ -3099,13 +3073,6 @@
 #define CUR_CHICKEN(pipe) _MMIO_CURSOR2(pipe, _CUR_CHICKEN_A)
 #define CURSURFLIVE(pipe) _MMIO_CURSOR2(pipe, _CURASURFLIVE)
 
-#define CURSOR_A_OFFSET 0x70080
-#define CURSOR_B_OFFSET 0x700c0
-#define CHV_CURSOR_C_OFFSET 0x700e0
-#define IVB_CURSOR_B_OFFSET 0x71080
-#define IVB_CURSOR_C_OFFSET 0x72080
-#define TGL_CURSOR_D_OFFSET 0x73080
-
 /* Display A control */
 #define _DSPAADDR_VLV				0x7017C /* vlv/chv */
 #define _DSPACNTR				0x70180
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 3a0d1cb2eaa9c..79523e55ca9c4 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -574,7 +574,6 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 {
 	struct intel_device_info *info;
 	struct intel_runtime_info *runtime;
-	struct intel_display_runtime_info *display_runtime;
 
 	/* Setup the write-once "constant" device info */
 	info = mkwrite_device_info(i915);
@@ -583,9 +582,12 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 	/* Initialize initial runtime info from static const data and pdev. */
 	runtime = RUNTIME_INFO(i915);
 	memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime));
-	display_runtime = DISPLAY_RUNTIME_INFO(i915);
-	memcpy(display_runtime, &DISPLAY_INFO(i915)->__runtime_defaults,
-	       sizeof(*display_runtime));
+
+	/* Probe display support */
+	info->display = intel_display_device_probe(device_id);
+	memcpy(DISPLAY_RUNTIME_INFO(i915),
+	       &DISPLAY_INFO(i915)->__runtime_defaults,
+	       sizeof(*DISPLAY_RUNTIME_INFO(i915)));
 
 	runtime->device_id = device_id;
 }
-- 
GitLab


From 12e6f6dc78e4f4a418648fb1a9c0cd2ae9b3430b Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:08 -0700
Subject: [PATCH 0651/1791] drm/i915/display: Handle GMD_ID identification in
 display code

For platforms with GMD_ID support (i.e., everything MTL and beyond),
identification of the display IP present should be based on the contents
of the GMD_ID register rather than a PCI devid match.

Note that since GMD_ID readout requires access to the PCI BAR, a slight
change to the driver init sequence is needed --- pci_enable_device() is
now called before i915_driver_create().

v2:
 - Fix use of uninitialized i915 pointer in error path if
   pci_enable_device() fails before the i915 device is created.  (lkp)
 - Use drm_device parameter to intel_display_device_probe.  This goes
   against i915 conventions, but since the primary goal here is to make
   it easy to call this function from other drivers (like Xe) and since
   we don't need anything from the i915 structure, this seems like an
   exception where drm_device is a more natural fit.
v3:
 - Go back do drm_i915_private for intel_display_device_probe.  (Jani)
 - Move forward decl to top of header.  (Jani)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-6-matthew.d.roper@intel.com
---
 .../drm/i915/display/intel_display_device.c   | 65 +++++++++++++++++--
 .../drm/i915/display/intel_display_device.h   |  5 +-
 drivers/gpu/drm/i915/i915_driver.c            | 17 +++--
 drivers/gpu/drm/i915/intel_device_info.c      | 13 ++--
 4 files changed, 84 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 3e2f4cd0b9f56..a513ac8f71a30 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -5,7 +5,10 @@
 
 #include <drm/i915_pciids.h>
 #include <drm/drm_color_mgmt.h>
+#include <linux/pci.h>
 
+#include "i915_drv.h"
+#include "i915_reg.h"
 #include "intel_display_device.h"
 #include "intel_display_power.h"
 #include "intel_display_reg_defs.h"
@@ -700,19 +703,73 @@ static const struct {
 	INTEL_RPLP_IDS(&xe_lpd_display),
 	INTEL_DG2_IDS(&xe_hpd_display),
 
-	/* FIXME: Replace this with a GMD_ID lookup */
-	INTEL_MTL_IDS(&xe_lpdp_display),
+	/*
+	 * Do not add any GMD_ID-based platforms to this list.  They will
+	 * be probed automatically based on the IP version reported by
+	 * the hardware.
+	 */
 };
 
+static const struct {
+	u16 ver;
+	u16 rel;
+	const struct intel_display_device_info *display;
+} gmdid_display_map[] = {
+	{ 14,  0, &xe_lpdp_display },
+};
+
+static const struct intel_display_device_info *
+probe_gmdid_display(struct drm_i915_private *i915, u16 *ver, u16 *rel, u16 *step)
+{
+	struct pci_dev *pdev = to_pci_dev(i915->drm.dev);
+	void __iomem *addr;
+	u32 val;
+	int i;
+
+	addr = pci_iomap_range(pdev, 0, i915_mmio_reg_offset(GMD_ID_DISPLAY), sizeof(u32));
+	if (!addr) {
+		drm_err(&i915->drm, "Cannot map MMIO BAR to read display GMD_ID\n");
+		return &no_display;
+	}
+
+	val = ioread32(addr);
+	pci_iounmap(pdev, addr);
+
+	if (val == 0)
+		/* Platform doesn't have display */
+		return &no_display;
+
+	*ver = REG_FIELD_GET(GMD_ID_ARCH_MASK, val);
+	*rel = REG_FIELD_GET(GMD_ID_RELEASE_MASK, val);
+	*step = REG_FIELD_GET(GMD_ID_STEP, val);
+
+	for (i = 0; i < ARRAY_SIZE(gmdid_display_map); i++)
+		if (*ver == gmdid_display_map[i].ver &&
+		    *rel == gmdid_display_map[i].rel)
+			return gmdid_display_map[i].display;
+
+	drm_err(&i915->drm, "Unrecognized display IP version %d.%02d; disabling display.\n",
+		*ver, *rel);
+	return &no_display;
+}
+
 const struct intel_display_device_info *
-intel_display_device_probe(u16 pci_devid)
+intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid,
+			   u16 *gmdid_ver, u16 *gmdid_rel, u16 *gmdid_step)
 {
+	struct pci_dev *pdev = to_pci_dev(i915->drm.dev);
 	int i;
 
+	if (has_gmdid)
+		return probe_gmdid_display(i915, gmdid_ver, gmdid_rel, gmdid_step);
+
 	for (i = 0; i < ARRAY_SIZE(intel_display_ids); i++) {
-		if (intel_display_ids[i].devid == pci_devid)
+		if (intel_display_ids[i].devid == pdev->device)
 			return intel_display_ids[i].info;
 	}
 
+	drm_dbg(&i915->drm, "No display ID found for device ID %04x; disabling display.\n",
+		pdev->device);
+
 	return &no_display;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 1f7d08b3ad6b1..d1d11581d85dc 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -10,6 +10,8 @@
 
 #include "display/intel_display_limits.h"
 
+struct drm_i915_private;
+
 #define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \
 	/* Keep in alphabetical order */ \
 	func(cursor_needs_physical); \
@@ -81,6 +83,7 @@ struct intel_display_device_info {
 };
 
 const struct intel_display_device_info *
-intel_display_device_probe(u16 pci_devid);
+intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid,
+			   u16 *ver, u16 *rel, u16 *step);
 
 #endif
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 522733a899463..37532e55327d8 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -754,13 +754,17 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct drm_i915_private *i915;
 	int ret;
 
-	i915 = i915_driver_create(pdev, ent);
-	if (IS_ERR(i915))
-		return PTR_ERR(i915);
-
 	ret = pci_enable_device(pdev);
-	if (ret)
-		goto out_fini;
+	if (ret) {
+		pr_err("Failed to enable graphics device: %pe\n", ERR_PTR(ret));
+		return ret;
+	}
+
+	i915 = i915_driver_create(pdev, ent);
+	if (IS_ERR(i915)) {
+		ret = PTR_ERR(i915);
+		goto out_pci_disable;
+	}
 
 	ret = i915_driver_early_probe(i915);
 	if (ret < 0)
@@ -843,7 +847,6 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i915_driver_late_release(i915);
 out_pci_disable:
 	pci_disable_device(pdev);
-out_fini:
 	i915_probe_error(i915, "Device initialization failed (%d)\n", ret);
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 79523e55ca9c4..2f79d232b04a9 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -345,7 +345,6 @@ static void ip_ver_read(struct drm_i915_private *i915, u32 offset, struct intel_
 static void intel_ipver_early_init(struct drm_i915_private *i915)
 {
 	struct intel_runtime_info *runtime = RUNTIME_INFO(i915);
-	struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(i915);
 
 	if (!HAS_GMD_ID(i915)) {
 		drm_WARN_ON(&i915->drm, RUNTIME_INFO(i915)->graphics.ip.ver > 12);
@@ -366,8 +365,6 @@ static void intel_ipver_early_init(struct drm_i915_private *i915)
 		RUNTIME_INFO(i915)->graphics.ip.ver = 12;
 		RUNTIME_INFO(i915)->graphics.ip.rel = 70;
 	}
-	ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_DISPLAY),
-		    (struct intel_ip_version *)&display_runtime->ip);
 	ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_MEDIA),
 		    &runtime->media.ip);
 }
@@ -574,6 +571,7 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 {
 	struct intel_device_info *info;
 	struct intel_runtime_info *runtime;
+	u16 ver, rel, step;
 
 	/* Setup the write-once "constant" device info */
 	info = mkwrite_device_info(i915);
@@ -584,11 +582,18 @@ void intel_device_info_driver_create(struct drm_i915_private *i915,
 	memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime));
 
 	/* Probe display support */
-	info->display = intel_display_device_probe(device_id);
+	info->display = intel_display_device_probe(i915, info->has_gmd_id,
+						   &ver, &rel, &step);
 	memcpy(DISPLAY_RUNTIME_INFO(i915),
 	       &DISPLAY_INFO(i915)->__runtime_defaults,
 	       sizeof(*DISPLAY_RUNTIME_INFO(i915)));
 
+	if (info->has_gmd_id) {
+		DISPLAY_RUNTIME_INFO(i915)->ip.ver = ver;
+		DISPLAY_RUNTIME_INFO(i915)->ip.rel = rel;
+		DISPLAY_RUNTIME_INFO(i915)->ip.step = step;
+	}
+
 	runtime->device_id = device_id;
 }
 
-- 
GitLab


From 95c08508e237ba2444786581bfcc7df1ff54c35b Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Tue, 23 May 2023 12:56:09 -0700
Subject: [PATCH 0652/1791] drm/i915/display: Move feature test macros to
 intel_display_device.h

It makes sense to keep the display feature test macros centralized
within the display code.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523195609.73627-7-matthew.d.roper@intel.com
---
 .../drm/i915/display/intel_display_device.h   | 39 ++++++++++++
 drivers/gpu/drm/i915/i915_drv.h               | 60 -------------------
 2 files changed, 39 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index d1d11581d85dc..2aa82cbdf1c53 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -31,6 +31,45 @@ struct drm_i915_private;
 	func(overlay_needs_physical); \
 	func(supports_tv);
 
+#define HAS_ASYNC_FLIPS(i915)		(DISPLAY_VER(i915) >= 5)
+#define HAS_CDCLK_CRAWL(i915)		(DISPLAY_INFO(i915)->has_cdclk_crawl)
+#define HAS_CDCLK_SQUASH(i915)		(DISPLAY_INFO(i915)->has_cdclk_squash)
+#define HAS_CUR_FBC(i915)		(!HAS_GMCH(i915) && DISPLAY_VER(i915) >= 7)
+#define HAS_D12_PLANE_MINIMIZATION(i915) (IS_ROCKETLAKE(i915) || IS_ALDERLAKE_S(i915))
+#define HAS_DDI(i915)			(DISPLAY_INFO(i915)->has_ddi)
+#define HAS_DISPLAY(i915)		(DISPLAY_RUNTIME_INFO(i915)->pipe_mask != 0)
+#define HAS_DMC(i915)			(DISPLAY_RUNTIME_INFO(i915)->has_dmc)
+#define HAS_DOUBLE_BUFFERED_M_N(i915)	(DISPLAY_VER(i915) >= 9 || IS_BROADWELL(i915))
+#define HAS_DP_MST(i915)		(DISPLAY_INFO(i915)->has_dp_mst)
+#define HAS_DP20(i915)			(IS_DG2(i915) || DISPLAY_VER(i915) >= 14)
+#define HAS_DPT(i915)			(DISPLAY_VER(i915) >= 13)
+#define HAS_DSB(i915)			(DISPLAY_INFO(i915)->has_dsb)
+#define HAS_DSC(__i915)			(DISPLAY_RUNTIME_INFO(__i915)->has_dsc)
+#define HAS_FBC(i915)			(DISPLAY_RUNTIME_INFO(i915)->fbc_mask != 0)
+#define HAS_FPGA_DBG_UNCLAIMED(i915)	(DISPLAY_INFO(i915)->has_fpga_dbg)
+#define HAS_FW_BLC(i915)		(DISPLAY_VER(i915) > 2)
+#define HAS_GMBUS_IRQ(i915)		(DISPLAY_VER(i915) >= 4)
+#define HAS_GMBUS_BURST_READ(i915)	(DISPLAY_VER(i915) >= 10 || IS_KABYLAKE(i915))
+#define HAS_GMCH(i915)			(DISPLAY_INFO(i915)->has_gmch)
+#define HAS_HW_SAGV_WM(i915)		(DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915))
+#define HAS_IPC(i915)			(DISPLAY_INFO(i915)->has_ipc)
+#define HAS_IPS(i915)			(IS_HSW_ULT(i915) || IS_BROADWELL(i915))
+#define HAS_LSPCON(i915)		(IS_DISPLAY_VER(i915, 9, 10))
+#define HAS_MBUS_JOINING(i915)		(IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14)
+#define HAS_MSO(i915)			(DISPLAY_VER(i915) >= 12)
+#define HAS_OVERLAY(i915)		(DISPLAY_INFO(i915)->has_overlay)
+#define HAS_PSR(i915)			(DISPLAY_INFO(i915)->has_psr)
+#define HAS_PSR_HW_TRACKING(i915)	(DISPLAY_INFO(i915)->has_psr_hw_tracking)
+#define HAS_PSR2_SEL_FETCH(i915)	(DISPLAY_VER(i915) >= 12)
+#define HAS_SAGV(i915)			(DISPLAY_VER(i915) >= 9 && !IS_LP(i915))
+#define HAS_TRANSCODER(i915, trans)	((DISPLAY_RUNTIME_INFO(i915)->cpu_transcoder_mask & \
+					  BIT(trans)) != 0)
+#define HAS_VRR(i915)			(DISPLAY_VER(i915) >= 11)
+#define INTEL_NUM_PIPES(i915)		(hweight8(DISPLAY_RUNTIME_INFO(i915)->pipe_mask))
+#define I915_HAS_HOTPLUG(i915)		(DISPLAY_INFO(i915)->has_hotplug)
+#define OVERLAY_NEEDS_PHYSICAL(i915)	(DISPLAY_INFO(i915)->overlay_needs_physical)
+#define SUPPORTS_TV(i915)		(DISPLAY_INFO(i915)->supports_tv)
+
 struct intel_display_runtime_info {
 	struct {
 		u16 ver;
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index dbdecf1ee24fe..e9c403def9c93 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -785,10 +785,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 	((sizes) & ~RUNTIME_INFO(dev_priv)->page_sizes) == 0; \
 })
 
-#define HAS_OVERLAY(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_overlay)
-#define OVERLAY_NEEDS_PHYSICAL(dev_priv) \
-		(DISPLAY_INFO(dev_priv)->overlay_needs_physical)
-
 /* Early gen2 have a totally busted CS tlb and require pinned batches. */
 #define HAS_BROKEN_CS_TLB(dev_priv)	(IS_I830(dev_priv) || IS_I845G(dev_priv))
 
@@ -799,41 +795,12 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define NEEDS_WaRsDisableCoarsePowerGating(dev_priv)			\
 	(IS_SKL_GT3(dev_priv) || IS_SKL_GT4(dev_priv))
 
-#define HAS_GMBUS_IRQ(dev_priv) (DISPLAY_VER(dev_priv) >= 4)
-#define HAS_GMBUS_BURST_READ(dev_priv) (DISPLAY_VER(dev_priv) >= 11 || \
-					IS_GEMINILAKE(dev_priv) || \
-					IS_KABYLAKE(dev_priv))
-
 /* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte
  * rows, which changed the alignment requirements and fence programming.
  */
 #define HAS_128_BYTE_Y_TILING(dev_priv) (GRAPHICS_VER(dev_priv) != 2 && \
 					 !(IS_I915G(dev_priv) || IS_I915GM(dev_priv)))
-#define SUPPORTS_TV(dev_priv)		(DISPLAY_INFO(dev_priv)->supports_tv)
-#define I915_HAS_HOTPLUG(dev_priv)	(DISPLAY_INFO(dev_priv)->has_hotplug)
-
-#define HAS_FW_BLC(dev_priv)	(DISPLAY_VER(dev_priv) > 2)
-#define HAS_FBC(dev_priv)	(DISPLAY_RUNTIME_INFO(dev_priv)->fbc_mask != 0)
-#define HAS_CUR_FBC(dev_priv)	(!HAS_GMCH(dev_priv) && DISPLAY_VER(dev_priv) >= 7)
-
-#define HAS_DPT(dev_priv)	(DISPLAY_VER(dev_priv) >= 13)
-
-#define HAS_IPS(dev_priv)	(IS_HSW_ULT(dev_priv) || IS_BROADWELL(dev_priv))
 
-#define HAS_DP_MST(dev_priv)	(DISPLAY_INFO(dev_priv)->has_dp_mst)
-#define HAS_DP20(dev_priv)	(IS_DG2(dev_priv) || DISPLAY_VER(dev_priv) >= 14)
-
-#define HAS_DOUBLE_BUFFERED_M_N(dev_priv)	(DISPLAY_VER(dev_priv) >= 9 || IS_BROADWELL(dev_priv))
-
-#define HAS_CDCLK_CRAWL(dev_priv)	 (DISPLAY_INFO(dev_priv)->has_cdclk_crawl)
-#define HAS_CDCLK_SQUASH(dev_priv)	 (DISPLAY_INFO(dev_priv)->has_cdclk_squash)
-#define HAS_DDI(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_ddi)
-#define HAS_FPGA_DBG_UNCLAIMED(dev_priv) (DISPLAY_INFO(dev_priv)->has_fpga_dbg)
-#define HAS_PSR(dev_priv)		 (DISPLAY_INFO(dev_priv)->has_psr)
-#define HAS_PSR_HW_TRACKING(dev_priv) \
-	(DISPLAY_INFO(dev_priv)->has_psr_hw_tracking)
-#define HAS_PSR2_SEL_FETCH(dev_priv)	 (DISPLAY_VER(dev_priv) >= 12)
-#define HAS_TRANSCODER(dev_priv, trans)	 ((DISPLAY_RUNTIME_INFO(dev_priv)->cpu_transcoder_mask & BIT(trans)) != 0)
 
 #define HAS_RC6(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6)
 #define HAS_RC6p(dev_priv)		 (INTEL_INFO(dev_priv)->has_rc6p)
@@ -841,11 +808,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_RPS(dev_priv)	(INTEL_INFO(dev_priv)->has_rps)
 
-#define HAS_DMC(dev_priv)	(DISPLAY_RUNTIME_INFO(dev_priv)->has_dmc)
-#define HAS_DSB(dev_priv)	(DISPLAY_INFO(dev_priv)->has_dsb)
-#define HAS_DSC(__i915)		(DISPLAY_RUNTIME_INFO(__i915)->has_dsc)
-#define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915))
-
 #define HAS_HECI_PXP(dev_priv) \
 	(INTEL_INFO(dev_priv)->has_heci_pxp)
 
@@ -854,8 +816,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_HECI_GSC(dev_priv) (HAS_HECI_PXP(dev_priv) || HAS_HECI_GSCFI(dev_priv))
 
-#define HAS_MSO(i915)		(DISPLAY_VER(i915) >= 12)
-
 #define HAS_RUNTIME_PM(dev_priv) (INTEL_INFO(dev_priv)->has_runtime_pm)
 #define HAS_64BIT_RELOC(dev_priv) (INTEL_INFO(dev_priv)->has_64bit_reloc)
 
@@ -872,9 +832,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
  */
 #define HAS_64K_PAGES(dev_priv) (INTEL_INFO(dev_priv)->has_64k_pages)
 
-#define HAS_IPC(dev_priv)		(DISPLAY_INFO(dev_priv)->has_ipc)
-#define HAS_SAGV(dev_priv)		(DISPLAY_VER(dev_priv) >= 9 && !IS_LP(dev_priv))
-
 #define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i))
 #define HAS_LMEM(i915) HAS_REGION(i915, REGION_LMEM)
 
@@ -892,12 +849,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 
 #define HAS_GLOBAL_MOCS_REGISTERS(dev_priv)	(INTEL_INFO(dev_priv)->has_global_mocs)
 
-#define HAS_GMCH(dev_priv) (DISPLAY_INFO(dev_priv)->has_gmch)
-
 #define HAS_GMD_ID(i915)	(INTEL_INFO(i915)->has_gmd_id)
 
-#define HAS_LSPCON(dev_priv) (IS_DISPLAY_VER(dev_priv, 9, 10))
-
 #define HAS_L3_CCS_READ(i915) (INTEL_INFO(i915)->has_l3_ccs_read)
 
 /* DPF == dynamic parity feature */
@@ -905,14 +858,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define NUM_L3_SLICES(dev_priv) (IS_HSW_GT3(dev_priv) ? \
 				 2 : HAS_L3_DPF(dev_priv))
 
-#define INTEL_NUM_PIPES(dev_priv) (hweight8(DISPLAY_RUNTIME_INFO(dev_priv)->pipe_mask))
-
-#define HAS_DISPLAY(dev_priv) (DISPLAY_RUNTIME_INFO(dev_priv)->pipe_mask != 0)
-
-#define HAS_VRR(i915)	(DISPLAY_VER(i915) >= 11)
-
-#define HAS_ASYNC_FLIPS(i915)		(DISPLAY_VER(i915) >= 5)
-
 /* Only valid when HAS_DISPLAY() is true */
 #define INTEL_DISPLAY_ENABLED(dev_priv) \
 	(drm_WARN_ON(&(dev_priv)->drm, !HAS_DISPLAY(dev_priv)),		\
@@ -922,11 +867,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define HAS_GUC_DEPRIVILEGE(dev_priv) \
 	(INTEL_INFO(dev_priv)->has_guc_deprivilege)
 
-#define HAS_D12_PLANE_MINIMIZATION(dev_priv) (IS_ROCKETLAKE(dev_priv) || \
-					      IS_ALDERLAKE_S(dev_priv))
-
-#define HAS_MBUS_JOINING(i915) (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14)
-
 #define HAS_3D_PIPELINE(i915)	(INTEL_INFO(i915)->has_3d_pipeline)
 
 #define HAS_ONE_EU_PER_FUSE_BIT(i915)	(INTEL_INFO(i915)->has_one_eu_per_fuse_bit)
-- 
GitLab


From abd686b8c38a094631884aa7b8cb04cc32b6608a Mon Sep 17 00:00:00 2001
From: Alexander Stein <alexander.stein@ew.tq-group.com>
Date: Tue, 16 May 2023 09:19:49 +0200
Subject: [PATCH 0653/1791] drm/bridge: tc358767: explicitly set readable
 registers

This map was created from register map from datasheet (section 5.1.2).
Unused registers are stated by address, so they show up in debugfs as well.

Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230516071949.375264-1-alexander.stein@ew.tq-group.com
---
 drivers/gpu/drm/bridge/tc358767.c | 195 +++++++++++++++++++++++++++++-
 1 file changed, 194 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index d6349af4f1b62..9db0e0231e039 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -1781,7 +1781,200 @@ static const struct drm_bridge_funcs tc_edp_bridge_funcs = {
 
 static bool tc_readable_reg(struct device *dev, unsigned int reg)
 {
-	return reg != SYSCTRL;
+	switch (reg) {
+	/* DSI D-PHY Layer */
+	case 0x004:
+	case 0x020:
+	case 0x024:
+	case 0x028:
+	case 0x02c:
+	case 0x030:
+	case 0x038:
+	case 0x040:
+	case 0x044:
+	case 0x048:
+	case 0x04c:
+	case 0x050:
+	case 0x054:
+	/* DSI PPI Layer */
+	case PPI_STARTPPI:
+	case 0x108:
+	case 0x110:
+	case PPI_LPTXTIMECNT:
+	case PPI_LANEENABLE:
+	case PPI_TX_RX_TA:
+	case 0x140:
+	case PPI_D0S_ATMR:
+	case PPI_D1S_ATMR:
+	case 0x14c:
+	case 0x150:
+	case PPI_D0S_CLRSIPOCOUNT:
+	case PPI_D1S_CLRSIPOCOUNT:
+	case PPI_D2S_CLRSIPOCOUNT:
+	case PPI_D3S_CLRSIPOCOUNT:
+	case 0x180:
+	case 0x184:
+	case 0x188:
+	case 0x18c:
+	case 0x190:
+	case 0x1a0:
+	case 0x1a4:
+	case 0x1a8:
+	case 0x1ac:
+	case 0x1b0:
+	case 0x1c0:
+	case 0x1c4:
+	case 0x1c8:
+	case 0x1cc:
+	case 0x1d0:
+	case 0x1e0:
+	case 0x1e4:
+	case 0x1f0:
+	case 0x1f4:
+	/* DSI Protocol Layer */
+	case DSI_STARTDSI:
+	case 0x208:
+	case DSI_LANEENABLE:
+	case 0x214:
+	case 0x218:
+	case 0x220:
+	case 0x224:
+	case 0x228:
+	case 0x230:
+	/* DSI General */
+	case 0x300:
+	/* DSI Application Layer */
+	case 0x400:
+	case 0x404:
+	/* DPI */
+	case DPIPXLFMT:
+	/* Parallel Output */
+	case POCTRL:
+	/* Video Path0 Configuration */
+	case VPCTRL0:
+	case HTIM01:
+	case HTIM02:
+	case VTIM01:
+	case VTIM02:
+	case VFUEN0:
+	/* System */
+	case TC_IDREG:
+	case 0x504:
+	case SYSSTAT:
+	case SYSRSTENB:
+	case SYSCTRL:
+	/* I2C */
+	case 0x520:
+	/* GPIO */
+	case GPIOM:
+	case GPIOC:
+	case GPIOO:
+	case GPIOI:
+	/* Interrupt */
+	case INTCTL_G:
+	case INTSTS_G:
+	case 0x570:
+	case 0x574:
+	case INT_GP0_LCNT:
+	case INT_GP1_LCNT:
+	/* DisplayPort Control */
+	case DP0CTL:
+	/* DisplayPort Clock */
+	case DP0_VIDMNGEN0:
+	case DP0_VIDMNGEN1:
+	case DP0_VMNGENSTATUS:
+	case 0x628:
+	case 0x62c:
+	case 0x630:
+	/* DisplayPort Main Channel */
+	case DP0_SECSAMPLE:
+	case DP0_VIDSYNCDELAY:
+	case DP0_TOTALVAL:
+	case DP0_STARTVAL:
+	case DP0_ACTIVEVAL:
+	case DP0_SYNCVAL:
+	case DP0_MISC:
+	/* DisplayPort Aux Channel */
+	case DP0_AUXCFG0:
+	case DP0_AUXCFG1:
+	case DP0_AUXADDR:
+	case 0x66c:
+	case 0x670:
+	case 0x674:
+	case 0x678:
+	case 0x67c:
+	case 0x680:
+	case 0x684:
+	case 0x688:
+	case DP0_AUXSTATUS:
+	case DP0_AUXI2CADR:
+	/* DisplayPort Link Training */
+	case DP0_SRCCTRL:
+	case DP0_LTSTAT:
+	case DP0_SNKLTCHGREQ:
+	case DP0_LTLOOPCTRL:
+	case DP0_SNKLTCTRL:
+	case 0x6e8:
+	case 0x6ec:
+	case 0x6f0:
+	case 0x6f4:
+	/* DisplayPort Audio */
+	case 0x700:
+	case 0x704:
+	case 0x708:
+	case 0x70c:
+	case 0x710:
+	case 0x714:
+	case 0x718:
+	case 0x71c:
+	case 0x720:
+	/* DisplayPort Source Control */
+	case DP1_SRCCTRL:
+	/* DisplayPort PHY */
+	case DP_PHY_CTRL:
+	case 0x810:
+	case 0x814:
+	case 0x820:
+	case 0x840:
+	/* I2S */
+	case 0x880:
+	case 0x888:
+	case 0x88c:
+	case 0x890:
+	case 0x894:
+	case 0x898:
+	case 0x89c:
+	case 0x8a0:
+	case 0x8a4:
+	case 0x8a8:
+	case 0x8ac:
+	case 0x8b0:
+	case 0x8b4:
+	/* PLL */
+	case DP0_PLLCTRL:
+	case DP1_PLLCTRL:
+	case PXL_PLLCTRL:
+	case PXL_PLLPARAM:
+	case SYS_PLLPARAM:
+	/* HDCP */
+	case 0x980:
+	case 0x984:
+	case 0x988:
+	case 0x98c:
+	case 0x990:
+	case 0x994:
+	case 0x998:
+	case 0x99c:
+	case 0x9a0:
+	case 0x9a4:
+	case 0x9a8:
+	case 0x9ac:
+	/* Debug */
+	case TSTCTL:
+	case PLL_DBG:
+		return true;
+	}
+	return false;
 }
 
 static const struct regmap_range tc_volatile_ranges[] = {
-- 
GitLab


From 0c14d3130654fe459fca3067d2d4317fc607bc71 Mon Sep 17 00:00:00 2001
From: Frieder Schrempf <frieder.schrempf@kontron.de>
Date: Wed, 3 May 2023 18:33:06 +0200
Subject: [PATCH 0654/1791] drm: bridge: samsung-dsim: Fix i.MX8M enable flow
 to meet spec

According to the documentation [1] the proper enable flow is:

1. Enable DSI link and keep data lanes in LP-11 (stop state)
2. Disable stop state to bring data lanes into HS mode

Currently we do this all at once within enable(), which doesn't
allow to meet the requirements of some downstream bridges.

To fix this we now enable the DSI in pre_enable() and force it
into stop state using the FORCE_STOP_STATE bit in the ESCMODE
register until enable() is called where we reset the bit.

We currently do this only for i.MX8M as Exynos uses a different
init flow where samsung_dsim_init() is called from
samsung_dsim_host_transfer().

[1] https://docs.kernel.org/gpu/drm-kms-helpers.html#mipi-dsi-bridge-operation

Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com> #TQMa8MxML/MBa8Mx
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230503163313.2640898-2-frieder@fris.de
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 0f3f6846beeaa..d038f6226c777 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -866,6 +866,10 @@ static int samsung_dsim_init_link(struct samsung_dsim *dsi)
 	reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
 	reg &= ~DSIM_STOP_STATE_CNT_MASK;
 	reg |= DSIM_STOP_STATE_CNT(driver_data->reg_values[STOP_STATE_CNT]);
+
+	if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type))
+		reg |= DSIM_FORCE_STOP_STATE;
+
 	samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
 
 	reg = DSIM_BTA_TIMEOUT(0xff) | DSIM_LPDR_TIMEOUT(0xffff);
@@ -1347,6 +1351,9 @@ static void samsung_dsim_atomic_pre_enable(struct drm_bridge *bridge,
 		ret = samsung_dsim_init(dsi);
 		if (ret)
 			return;
+
+		samsung_dsim_set_display_mode(dsi);
+		samsung_dsim_set_display_enable(dsi, true);
 	}
 }
 
@@ -1354,9 +1361,16 @@ static void samsung_dsim_atomic_enable(struct drm_bridge *bridge,
 				       struct drm_bridge_state *old_bridge_state)
 {
 	struct samsung_dsim *dsi = bridge_to_dsi(bridge);
+	u32 reg;
 
-	samsung_dsim_set_display_mode(dsi);
-	samsung_dsim_set_display_enable(dsi, true);
+	if (samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) {
+		samsung_dsim_set_display_mode(dsi);
+		samsung_dsim_set_display_enable(dsi, true);
+	} else {
+		reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
+		reg &= ~DSIM_FORCE_STOP_STATE;
+		samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
+	}
 
 	dsi->state |= DSIM_STATE_VIDOUT_AVAILABLE;
 }
@@ -1365,10 +1379,17 @@ static void samsung_dsim_atomic_disable(struct drm_bridge *bridge,
 					struct drm_bridge_state *old_bridge_state)
 {
 	struct samsung_dsim *dsi = bridge_to_dsi(bridge);
+	u32 reg;
 
 	if (!(dsi->state & DSIM_STATE_ENABLED))
 		return;
 
+	if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) {
+		reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
+		reg |= DSIM_FORCE_STOP_STATE;
+		samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
+	}
+
 	dsi->state &= ~DSIM_STATE_VIDOUT_AVAILABLE;
 }
 
-- 
GitLab


From dd9e329af7236e34c566d3705ea32a63069b9b13 Mon Sep 17 00:00:00 2001
From: Frieder Schrempf <frieder.schrempf@kontron.de>
Date: Wed, 3 May 2023 18:33:07 +0200
Subject: [PATCH 0655/1791] drm/bridge: ti-sn65dsi83: Fix enable/disable flow
 to meet spec

The datasheet describes the following initialization flow including
minimum delay times between each step:

1. DSI data lanes need to be in LP-11 and the clock lane in HS mode
2. toggle EN signal
3. initialize registers
4. enable PLL
5. soft reset
6. enable DSI stream
7. check error status register

To meet this requirement we need to make sure the host bridge's
pre_enable() is called first by using the pre_enable_prev_first
flag.

Furthermore we need to split enable() into pre_enable() which covers
steps 2-5 from above and enable() which covers step 7 and is called
after the host bridge's enable().

Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Fixes: ceb515ba29ba ("drm/bridge: ti-sn65dsi83: Add TI SN65DSI83 and SN65DSI84 driver")
Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com> #TQMa8MxML/MBa8Mx
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230503163313.2640898-3-frieder@fris.de
---
 drivers/gpu/drm/bridge/ti-sn65dsi83.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
index 1f5c07989e2bf..6e125ba4f0d75 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
@@ -321,8 +321,8 @@ static u8 sn65dsi83_get_dsi_div(struct sn65dsi83 *ctx)
 	return dsi_div - 1;
 }
 
-static void sn65dsi83_atomic_enable(struct drm_bridge *bridge,
-				    struct drm_bridge_state *old_bridge_state)
+static void sn65dsi83_atomic_pre_enable(struct drm_bridge *bridge,
+					struct drm_bridge_state *old_bridge_state)
 {
 	struct sn65dsi83 *ctx = bridge_to_sn65dsi83(bridge);
 	struct drm_atomic_state *state = old_bridge_state->base.state;
@@ -485,11 +485,22 @@ static void sn65dsi83_atomic_enable(struct drm_bridge *bridge,
 	/* Trigger reset after CSR register update. */
 	regmap_write(ctx->regmap, REG_RC_RESET, REG_RC_RESET_SOFT_RESET);
 
+	/* Wait for 10ms after soft reset as specified in datasheet */
+	usleep_range(10000, 12000);
+}
+
+static void sn65dsi83_atomic_enable(struct drm_bridge *bridge,
+				    struct drm_bridge_state *old_bridge_state)
+{
+	struct sn65dsi83 *ctx = bridge_to_sn65dsi83(bridge);
+	unsigned int pval;
+
 	/* Clear all errors that got asserted during initialization. */
 	regmap_read(ctx->regmap, REG_IRQ_STAT, &pval);
 	regmap_write(ctx->regmap, REG_IRQ_STAT, pval);
 
-	usleep_range(10000, 12000);
+	/* Wait for 1ms and check for errors in status register */
+	usleep_range(1000, 1100);
 	regmap_read(ctx->regmap, REG_IRQ_STAT, &pval);
 	if (pval)
 		dev_err(ctx->dev, "Unexpected link status 0x%02x\n", pval);
@@ -556,6 +567,7 @@ static const struct drm_bridge_funcs sn65dsi83_funcs = {
 	.attach			= sn65dsi83_attach,
 	.detach			= sn65dsi83_detach,
 	.atomic_enable		= sn65dsi83_atomic_enable,
+	.atomic_pre_enable	= sn65dsi83_atomic_pre_enable,
 	.atomic_disable		= sn65dsi83_atomic_disable,
 	.mode_valid		= sn65dsi83_mode_valid,
 
@@ -698,6 +710,7 @@ static int sn65dsi83_probe(struct i2c_client *client)
 
 	ctx->bridge.funcs = &sn65dsi83_funcs;
 	ctx->bridge.of_node = dev->of_node;
+	ctx->bridge.pre_enable_prev_first = true;
 	drm_bridge_add(&ctx->bridge);
 
 	ret = sn65dsi83_host_attach(ctx);
-- 
GitLab


From 1a56fcf08ae463a4564d111356091b2bdb6c7bce Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Thu, 25 May 2023 08:52:24 -0700
Subject: [PATCH 0656/1791] drm/amdgpu: Fix no-procfs build

Fixes undefined symbol when PROC_FS is not enabled.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202305251510.U0R2as7k-lkp@intel.com/
Fixes: 376c25f8ca47 ("drm/amdgpu: Switch to fdinfo helper")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: <alexander.deucher@amd.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525155227.560094-1-robdclark@gmail.com
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 1b46e7ac7cb0e..c9a41c997c6c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2802,7 +2802,9 @@ static const struct drm_driver amdgpu_kms_driver = {
 	.dumb_map_offset = amdgpu_mode_dumb_mmap,
 	.fops = &amdgpu_driver_kms_fops,
 	.release = &amdgpu_driver_release_kms,
+#ifdef CONFIG_PROC_FS
 	.show_fdinfo = amdgpu_show_fdinfo,
+#endif
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-- 
GitLab


From a617b33f7e513f25becf843bc97f8f1658c16337 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Thu, 25 May 2023 22:05:53 -0500
Subject: [PATCH 0657/1791] drm: bridge: samsung-dsim: fix blanking packet size
 calculation

Scale the blanking packet sizes to match the ratio between HS clock
and DPI interface clock. The controller seems to do internal scaling
to the number of active lanes, so we don't take those into account.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Adam Ford <aford173@gmail.com>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # imx8mm-icore
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-2-aford173@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index d038f6226c777..7b6d5dfc4f0d1 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -885,17 +885,29 @@ static void samsung_dsim_set_display_mode(struct samsung_dsim *dsi)
 	u32 reg;
 
 	if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
+		int byte_clk_khz = dsi->burst_clk_rate / 1000 / 8;
+		int hfp = (m->hsync_start - m->hdisplay) * byte_clk_khz / m->clock;
+		int hbp = (m->htotal - m->hsync_end) * byte_clk_khz / m->clock;
+		int hsa = (m->hsync_end - m->hsync_start) * byte_clk_khz / m->clock;
+
+		/* remove packet overhead when possible */
+		hfp = max(hfp - 6, 0);
+		hbp = max(hbp - 6, 0);
+		hsa = max(hsa - 6, 0);
+
+		dev_dbg(dsi->dev, "calculated hfp: %u, hbp: %u, hsa: %u",
+			hfp, hbp, hsa);
+
 		reg = DSIM_CMD_ALLOW(0xf)
 			| DSIM_STABLE_VFP(m->vsync_start - m->vdisplay)
 			| DSIM_MAIN_VBP(m->vtotal - m->vsync_end);
 		samsung_dsim_write(dsi, DSIM_MVPORCH_REG, reg);
 
-		reg = DSIM_MAIN_HFP(m->hsync_start - m->hdisplay)
-			| DSIM_MAIN_HBP(m->htotal - m->hsync_end);
+		reg = DSIM_MAIN_HFP(hfp) | DSIM_MAIN_HBP(hbp);
 		samsung_dsim_write(dsi, DSIM_MHPORCH_REG, reg);
 
 		reg = DSIM_MAIN_VSA(m->vsync_end - m->vsync_start)
-			| DSIM_MAIN_HSA(m->hsync_end - m->hsync_start);
+			| DSIM_MAIN_HSA(hsa);
 		samsung_dsim_write(dsi, DSIM_MSYNC_REG, reg);
 	}
 	reg =  DSIM_MAIN_HRESOL(m->hdisplay, num_bits_resol) |
-- 
GitLab


From 54f1a83c72250b182fa7722b0c5f6eb5e769598d Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:54 -0500
Subject: [PATCH 0658/1791] drm: bridge: samsung-dsim: Fix PMS Calculator on
 imx8m[mnp]

According to Table 13-45 of the i.MX8M Mini Reference Manual, the min
and max values for M and the frequency range for the VCO_out
calculator were incorrect.  This information was contradicted in other
parts of the mini, nano and plus manuals.  After reaching out to my
NXP Rep, when confronting him about discrepencies in the Nano manual,
he responded with:
 "Yes it is definitely wrong, the one that is part
  of the NOTE in MIPI_DPHY_M_PLLPMS register table against PMS_P,
  PMS_M and PMS_S is not correct. I will report this to Doc team,
  the one customer should be take into account is the Table 13-40
  DPHY PLL Parameters and the Note above."

These updated values also match what is used in the NXP downstream
kernel.

To fix this, make new variables to hold the min and max values of m
and the minimum value of VCO_out, and update the PMS calculator to
use these new variables instead of using hard-coded values to keep
the backwards compatibility with other parts using this driver.

Fixes: 4d562c70c4dc ("drm: bridge: samsung-dsim: Add i.MX8M Mini/Nano support")
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # imx8mm-icore
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-3-aford173@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 22 ++++++++++++++++++++--
 include/drm/bridge/samsung-dsim.h     |  3 +++
 2 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 7b6d5dfc4f0d1..d0f97fd199d0e 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -407,6 +407,9 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
 };
 
 static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
@@ -420,6 +423,9 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
 };
 
 static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
@@ -431,6 +437,9 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
 };
 
 static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
@@ -443,6 +452,9 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.num_bits_resol = 12,
 	.pll_p_offset = 13,
 	.reg_values = exynos5433_reg_values,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
 };
 
 static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
@@ -455,6 +467,9 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.num_bits_resol = 12,
 	.pll_p_offset = 13,
 	.reg_values = exynos5422_reg_values,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
 };
 
 static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
@@ -471,6 +486,9 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	 */
 	.pll_p_offset = 14,
 	.reg_values = imx8mm_dsim_reg_values,
+	.m_min = 64,
+	.m_max = 1023,
+	.min_freq = 1050,
 };
 
 static const struct samsung_dsim_driver_data *
@@ -549,12 +567,12 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi,
 			tmp = (u64)fout * (_p << _s);
 			do_div(tmp, fin);
 			_m = tmp;
-			if (_m < 41 || _m > 125)
+			if (_m < driver_data->m_min || _m > driver_data->m_max)
 				continue;
 
 			tmp = (u64)_m * fin;
 			do_div(tmp, _p);
-			if (tmp < 500 * MHZ ||
+			if (tmp < driver_data->min_freq  * MHZ ||
 			    tmp > driver_data->max_freq * MHZ)
 				continue;
 
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 6a37d1e079bf9..2c20b9460c9a4 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -54,11 +54,14 @@ struct samsung_dsim_driver_data {
 	unsigned int has_freqband:1;
 	unsigned int has_clklane_stop:1;
 	unsigned int num_clks;
+	unsigned int min_freq;
 	unsigned int max_freq;
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
 	unsigned int pll_p_offset;
 	const unsigned int *reg_values;
+	u16 m_min;
+	u16 m_max;
 };
 
 struct samsung_dsim_host_ops {
-- 
GitLab


From 33d8d14c83bf67aa0d262961a6fda9c40f3c1052 Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:55 -0500
Subject: [PATCH 0659/1791] drm: bridge: samsung-dsim: Fetch
 pll-clock-frequency automatically

Make the pll-clock-frequency optional.  If it's present, use it
to maintain backwards compatibility with existing hardware.  If it
is absent, read clock rate of "sclk_mipi" to determine the rate.
Since it can be optional, change the message from an error to
dev_info.

Signed-off-by: Adam Ford <aford173@gmail.com>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # imx8mm-icore
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-4-aford173@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index d0f97fd199d0e..cb6a7cf68352d 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -1740,11 +1740,11 @@ static const struct mipi_dsi_host_ops samsung_dsim_ops = {
 };
 
 static int samsung_dsim_of_read_u32(const struct device_node *np,
-				    const char *propname, u32 *out_value)
+				    const char *propname, u32 *out_value, bool optional)
 {
 	int ret = of_property_read_u32(np, propname, out_value);
 
-	if (ret < 0)
+	if (ret < 0 && !optional)
 		pr_err("%pOF: failed to get '%s' property\n", np, propname);
 
 	return ret;
@@ -1757,19 +1757,27 @@ static int samsung_dsim_parse_dt(struct samsung_dsim *dsi)
 	u32 lane_polarities[5] = { 0 };
 	struct device_node *endpoint;
 	int i, nr_lanes, ret;
+	struct clk *pll_clk;
 
 	ret = samsung_dsim_of_read_u32(node, "samsung,pll-clock-frequency",
-				       &dsi->pll_clk_rate);
-	if (ret < 0)
-		return ret;
+				       &dsi->pll_clk_rate, 1);
+	/* If it doesn't exist, read it from the clock instead of failing */
+	if (ret < 0) {
+		dev_dbg(dev, "Using sclk_mipi for pll clock frequency\n");
+		pll_clk = devm_clk_get(dev, "sclk_mipi");
+		if (!IS_ERR(pll_clk))
+			dsi->pll_clk_rate = clk_get_rate(pll_clk);
+		else
+			return PTR_ERR(pll_clk);
+	}
 
 	ret = samsung_dsim_of_read_u32(node, "samsung,burst-clock-frequency",
-				       &dsi->burst_clk_rate);
+				       &dsi->burst_clk_rate, 0);
 	if (ret < 0)
 		return ret;
 
 	ret = samsung_dsim_of_read_u32(node, "samsung,esc-clock-frequency",
-				       &dsi->esc_clk_rate);
+				       &dsi->esc_clk_rate, 0);
 	if (ret < 0)
 		return ret;
 
-- 
GitLab


From 171b3b1e0f8b8c894f2388e1cf765a56f831ee5e Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:56 -0500
Subject: [PATCH 0660/1791] drm: bridge: samsung-dsim: Select
 GENERIC_PHY_MIPI_DPHY

In order to support variable DPHY timings, it's necessary
to enable GENERIC_PHY_MIPI_DPHY so phy_mipi_dphy_get_default_config
can be used to determine the nominal values for a given resolution
and refresh rate.

Signed-off-by: Adam Ford <aford173@gmail.com>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-5-aford173@gmail.com
---
 drivers/gpu/drm/bridge/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index f076a09afac04..82c68b0424443 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -227,6 +227,7 @@ config DRM_SAMSUNG_DSIM
 	select DRM_KMS_HELPER
 	select DRM_MIPI_DSI
 	select DRM_PANEL_BRIDGE
+	select GENERIC_PHY_MIPI_DPHY
 	help
 	  The Samsung MIPI DSIM bridge controller driver.
 	  This MIPI DSIM bridge can be found it on Exynos SoCs and
-- 
GitLab


From 89691775f5735fca9dc40e119edcbb52a25b9612 Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:57 -0500
Subject: [PATCH 0661/1791] drm: bridge: samsung-dsim: Dynamically configure
 DPHY timing

The DPHY timings are currently hard coded. Since the input
clock can be variable, the phy timings need to be variable
too.  To facilitate this, we need to cache the hs_clock
based on what is generated from the PLL.

The phy_mipi_dphy_get_default_config_for_hsclk function
configures the DPHY timings in pico-seconds, and a small macro
converts those timings into clock cycles based on the hs_clk.

Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Michael Walle <michael@walle.cc>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # imx8mm-icore
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-6-aford173@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 58 +++++++++++++++++++++++----
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 52 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index cb6a7cf68352d..7a61e90f33744 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -220,6 +220,8 @@
 
 #define OLD_SCLK_MIPI_CLK_NAME		"pll_clk"
 
+#define PS_TO_CYCLE(ps, hz) DIV64_U64_ROUND_CLOSEST(((ps) * (hz)), 1000000000000ULL)
+
 static const char *const clk_names[5] = {
 	"bus_clk",
 	"sclk_mipi",
@@ -658,6 +660,8 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 		reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
 	} while ((reg & DSIM_PLL_STABLE) == 0);
 
+	dsi->hs_clock = fout;
+
 	return fout;
 }
 
@@ -705,13 +709,47 @@ static void samsung_dsim_set_phy_ctrl(struct samsung_dsim *dsi)
 	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
 	const unsigned int *reg_values = driver_data->reg_values;
 	u32 reg;
+	struct phy_configure_opts_mipi_dphy cfg;
+	int clk_prepare, lpx, clk_zero, clk_post, clk_trail;
+	int hs_exit, hs_prepare, hs_zero, hs_trail;
+	unsigned long long byte_clock = dsi->hs_clock / 8;
 
 	if (driver_data->has_freqband)
 		return;
 
+	phy_mipi_dphy_get_default_config_for_hsclk(dsi->hs_clock,
+						   dsi->lanes, &cfg);
+
+	/*
+	 * TODO:
+	 * The tech Applications Processor manuals for i.MX8M Mini, Nano,
+	 * and Plus don't state what the definition of the PHYTIMING
+	 * bits are beyond their address and bit position.
+	 * After reviewing NXP's downstream code, it appears
+	 * that the various PHYTIMING registers take the number
+	 * of cycles and use various dividers on them.  This
+	 * calculation does not result in an exact match to the
+	 * downstream code, but it is very close to the values
+	 * generated by their lookup table, and it appears
+	 * to sync at a variety of resolutions. If someone
+	 * can get a more accurate mathematical equation needed
+	 * for these registers, this should be updated.
+	 */
+
+	lpx = PS_TO_CYCLE(cfg.lpx, byte_clock);
+	hs_exit = PS_TO_CYCLE(cfg.hs_exit, byte_clock);
+	clk_prepare = PS_TO_CYCLE(cfg.clk_prepare, byte_clock);
+	clk_zero = PS_TO_CYCLE(cfg.clk_zero, byte_clock);
+	clk_post = PS_TO_CYCLE(cfg.clk_post, byte_clock);
+	clk_trail = PS_TO_CYCLE(cfg.clk_trail, byte_clock);
+	hs_prepare = PS_TO_CYCLE(cfg.hs_prepare, byte_clock);
+	hs_zero = PS_TO_CYCLE(cfg.hs_zero, byte_clock);
+	hs_trail = PS_TO_CYCLE(cfg.hs_trail, byte_clock);
+
 	/* B D-PHY: D-PHY Master & Slave Analog Block control */
 	reg = reg_values[PHYCTRL_ULPS_EXIT] | reg_values[PHYCTRL_VREG_LP] |
 		reg_values[PHYCTRL_SLEW_UP];
+
 	samsung_dsim_write(dsi, DSIM_PHYCTRL_REG, reg);
 
 	/*
@@ -719,7 +757,9 @@ static void samsung_dsim_set_phy_ctrl(struct samsung_dsim *dsi)
 	 * T HS-EXIT: Time that the transmitter drives LP-11 following a HS
 	 *	burst
 	 */
-	reg = reg_values[PHYTIMING_LPX] | reg_values[PHYTIMING_HS_EXIT];
+
+	reg  = DSIM_PHYTIMING_LPX(lpx) | DSIM_PHYTIMING_HS_EXIT(hs_exit);
+
 	samsung_dsim_write(dsi, DSIM_PHYTIMING_REG, reg);
 
 	/*
@@ -735,10 +775,11 @@ static void samsung_dsim_set_phy_ctrl(struct samsung_dsim *dsi)
 	 * T CLK-TRAIL: Time that the transmitter drives the HS-0 state after
 	 *	the last payload clock bit of a HS transmission burst
 	 */
-	reg = reg_values[PHYTIMING_CLK_PREPARE] |
-		reg_values[PHYTIMING_CLK_ZERO] |
-		reg_values[PHYTIMING_CLK_POST] |
-		reg_values[PHYTIMING_CLK_TRAIL];
+
+	reg = DSIM_PHYTIMING1_CLK_PREPARE(clk_prepare)	|
+	      DSIM_PHYTIMING1_CLK_ZERO(clk_zero)	|
+	      DSIM_PHYTIMING1_CLK_POST(clk_post)	|
+	      DSIM_PHYTIMING1_CLK_TRAIL(clk_trail);
 
 	samsung_dsim_write(dsi, DSIM_PHYTIMING1_REG, reg);
 
@@ -751,8 +792,11 @@ static void samsung_dsim_set_phy_ctrl(struct samsung_dsim *dsi)
 	 * T HS-TRAIL: Time that the transmitter drives the flipped differential
 	 *	state after last payload data bit of a HS transmission burst
 	 */
-	reg = reg_values[PHYTIMING_HS_PREPARE] | reg_values[PHYTIMING_HS_ZERO] |
-		reg_values[PHYTIMING_HS_TRAIL];
+
+	reg = DSIM_PHYTIMING2_HS_PREPARE(hs_prepare) |
+	      DSIM_PHYTIMING2_HS_ZERO(hs_zero) |
+	      DSIM_PHYTIMING2_HS_TRAIL(hs_trail);
+
 	samsung_dsim_write(dsi, DSIM_PHYTIMING2_REG, reg);
 }
 
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 2c20b9460c9a4..05100e91ecb96 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -93,6 +93,7 @@ struct samsung_dsim {
 
 	u32 pll_clk_rate;
 	u32 burst_clk_rate;
+	u32 hs_clock;
 	u32 esc_clk_rate;
 	u32 lanes;
 	u32 mode_flags;
-- 
GitLab


From bb0e13b9e223b218c9f242f8d340a332b4381042 Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:58 -0500
Subject: [PATCH 0662/1791] drm: bridge: samsung-dsim: Support non-burst mode

The high-speed clock is hard-coded to the burst-clock
frequency specified in the device tree.  However, when
using devices like certain bridge chips without burst mode
and varying resolutions and refresh rates, it may be
necessary to set the high-speed clock dynamically based
on the desired pixel clock for the connected device.

This also removes the need to set a clock speed from
the device tree for non-burst mode operation, since the
pixel clock rate is the rate requested from the attached
device like a bridge chip.  This should have no impact
for people using burst-mode and setting the burst clock
rate is still required for those users.  If the burst
clock is not present, change the error message to
dev_info indicating the clock use the pixel clock.

Signed-off-by: Adam Ford <aford173@gmail.com>
Tested-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Reviewed-by: Frieder Schrempf <frieder.schrempf@kontron.de>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # imx8mm-icore
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-7-aford173@gmail.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 7a61e90f33744..043b8109e64aa 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -667,11 +667,21 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 
 static int samsung_dsim_enable_clock(struct samsung_dsim *dsi)
 {
-	unsigned long hs_clk, byte_clk, esc_clk;
+	unsigned long hs_clk, byte_clk, esc_clk, pix_clk;
 	unsigned long esc_div;
 	u32 reg;
+	struct drm_display_mode *m = &dsi->mode;
+	int bpp = mipi_dsi_pixel_format_to_bpp(dsi->format);
+
+	/* m->clock is in KHz */
+	pix_clk = m->clock * 1000;
+
+	/* Use burst_clk_rate if available, otherwise use the pix_clk */
+	if (dsi->burst_clk_rate)
+		hs_clk = samsung_dsim_set_pll(dsi, dsi->burst_clk_rate);
+	else
+		hs_clk = samsung_dsim_set_pll(dsi, DIV_ROUND_UP(pix_clk * bpp, dsi->lanes));
 
-	hs_clk = samsung_dsim_set_pll(dsi, dsi->burst_clk_rate);
 	if (!hs_clk) {
 		dev_err(dsi->dev, "failed to configure DSI PLL\n");
 		return -EFAULT;
@@ -947,7 +957,7 @@ static void samsung_dsim_set_display_mode(struct samsung_dsim *dsi)
 	u32 reg;
 
 	if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
-		int byte_clk_khz = dsi->burst_clk_rate / 1000 / 8;
+		int byte_clk_khz = dsi->hs_clock / 1000 / 8;
 		int hfp = (m->hsync_start - m->hdisplay) * byte_clk_khz / m->clock;
 		int hbp = (m->htotal - m->hsync_end) * byte_clk_khz / m->clock;
 		int hsa = (m->hsync_end - m->hsync_start) * byte_clk_khz / m->clock;
@@ -1815,10 +1825,13 @@ static int samsung_dsim_parse_dt(struct samsung_dsim *dsi)
 			return PTR_ERR(pll_clk);
 	}
 
+	/* If it doesn't exist, use pixel clock instead of failing */
 	ret = samsung_dsim_of_read_u32(node, "samsung,burst-clock-frequency",
-				       &dsi->burst_clk_rate, 0);
-	if (ret < 0)
-		return ret;
+				       &dsi->burst_clk_rate, 1);
+	if (ret < 0) {
+		dev_dbg(dev, "Using pixel clock for HS clock frequency\n");
+		dsi->burst_clk_rate = 0;
+	}
 
 	ret = samsung_dsim_of_read_u32(node, "samsung,esc-clock-frequency",
 				       &dsi->esc_clk_rate, 0);
-- 
GitLab


From cfaf76d349837f695c8aa6d7077847fec4231fe5 Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Thu, 25 May 2023 22:05:59 -0500
Subject: [PATCH 0663/1791] dt-bindings: bridge: samsung-dsim: Make some flags
 optional

In the event a device is connected to the samsung-dsim
controller that doesn't support the burst-clock, the
driver is able to get the requested pixel clock from the
attached device or bridge.  In these instances, the
samsung,burst-clock-frequency isn't needed, so remove
it from the required list.

The pll-clock frequency can be set by the device tree entry
for samsung,pll-clock-frequency, but in some cases, the
pll-clock may have the same clock rate as sclk_mipi clock.
If they are equal, this flag is not needed since the driver
will use the sclk_mipi rate as a fallback.

Signed-off-by: Adam Ford <aford173@gmail.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526030559.326566-8-aford173@gmail.com
---
 .../bindings/display/bridge/samsung,mipi-dsim.yaml       | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index 9f61ebdfefa85..360fea81f4b6f 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -70,7 +70,9 @@ properties:
   samsung,burst-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM high speed burst mode frequency.
+      DSIM high speed burst mode frequency when connected to devices
+      that support burst mode. If absent, the driver will use the pixel
+      clock from the attached device or bridge.
 
   samsung,esc-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
@@ -80,7 +82,8 @@ properties:
   samsung,pll-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM oscillator clock frequency.
+      DSIM oscillator clock frequency. If absent, the driver will
+      use the clock frequency of sclk_mipi.
 
   phys:
     maxItems: 1
@@ -134,9 +137,7 @@ required:
   - compatible
   - interrupts
   - reg
-  - samsung,burst-clock-frequency
   - samsung,esc-clock-frequency
-  - samsung,pll-clock-frequency
 
 allOf:
   - $ref: ../dsi-controller.yaml#
-- 
GitLab


From d3a6c2b60f07c64631b9437032d8f079341b7a16 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 24 May 2023 14:32:10 +0200
Subject: [PATCH 0664/1791] dt-bindings: display: panel-simple: Add Ampire
 AM-800480L1TMQW-T00H

Document support for the Ampire AM-800480L1TMQW-T00H 5" WVGA TFT LCD
panel.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/422adef8c4941fa56fdadacb3d362a9fb387455e.1684931026.git.geert+renesas@glider.be
---
 .../devicetree/bindings/display/panel/panel-simple.yaml         | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index 733e47da36e80..f7c45e38d4b75 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -33,6 +33,8 @@ properties:
       - ampire,am-1280800n3tzqw-t00h
         # Ampire AM-480272H3TMQW-T01H 4.3" WQVGA TFT LCD panel
       - ampire,am-480272h3tmqw-t01h
+        # Ampire AM-800480L1TMQW-T00H 5" WVGA TFT LCD panel
+      - ampire,am-800480l1tmqw-t00h
         # Ampire AM-800480R3TMQW-A1H 7.0" WVGA TFT LCD panel
       - ampire,am800480r3tmqwa1h
         # Ampire AM-800600P5TMQW-TB8H 8.0" SVGA TFT LCD panel
-- 
GitLab


From 410bb21319f69c2ec28aeafe530d00ed2f6a1c54 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 24 May 2023 14:32:11 +0200
Subject: [PATCH 0665/1791] drm/panel: simple: Add Ampire AM-800480L1TMQW-T00H

Add support for the Ampire AM-800480L1TMQW-T00H 5" WVGA TFT LCD panel.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/244d9471e0ed248ff2dea8ded3a5384a1c51904b.1684931026.git.geert+renesas@glider.be
---
 drivers/gpu/drm/panel/panel-simple.c | 33 ++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 5778824dffd47..467117c0b2c9d 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -778,6 +778,36 @@ static const struct drm_display_mode ampire_am800480r3tmqwa1h_mode = {
 	.flags = DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC,
 };
 
+static const struct display_timing ampire_am_800480l1tmqw_t00h_timing = {
+	.pixelclock = { 29930000, 33260000, 36590000 },
+	.hactive = { 800, 800, 800 },
+	.hfront_porch = { 1, 40, 168 },
+	.hback_porch = { 88, 88, 88 },
+	.hsync_len = { 1, 128, 128 },
+	.vactive = { 480, 480, 480 },
+	.vfront_porch = { 1, 35, 37 },
+	.vback_porch = { 8, 8, 8 },
+	.vsync_len = { 1, 2, 2 },
+	.flags = DISPLAY_FLAGS_HSYNC_LOW | DISPLAY_FLAGS_VSYNC_LOW |
+		 DISPLAY_FLAGS_DE_HIGH | DISPLAY_FLAGS_PIXDATA_POSEDGE |
+		 DISPLAY_FLAGS_SYNC_POSEDGE,
+};
+
+static const struct panel_desc ampire_am_800480l1tmqw_t00h = {
+	.timings = &ampire_am_800480l1tmqw_t00h_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 111,
+		.height = 67,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.bus_flags = DRM_BUS_FLAG_DE_HIGH |
+		     DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE |
+		     DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE,
+	.connector_type = DRM_MODE_CONNECTOR_DPI,
+};
+
 static const struct panel_desc ampire_am800480r3tmqwa1h = {
 	.modes = &ampire_am800480r3tmqwa1h_mode,
 	.num_modes = 1,
@@ -3993,6 +4023,9 @@ static const struct of_device_id platform_of_match[] = {
 	}, {
 		.compatible = "ampire,am-480272h3tmqw-t01h",
 		.data = &ampire_am_480272h3tmqw_t01h,
+	}, {
+		.compatible = "ampire,am-800480l1tmqw-t00h",
+		.data = &ampire_am_800480l1tmqw_t00h,
 	}, {
 		.compatible = "ampire,am800480r3tmqwa1h",
 		.data = &ampire_am800480r3tmqwa1h,
-- 
GitLab


From f24b49550814fdee4a98b9552e35e243ccafd4a8 Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Tue, 16 May 2023 10:50:39 +0200
Subject: [PATCH 0666/1791] drm/panel: simple: fix active size for Ampire
 AM-480272H3TMQW-T01H

The previous setting was related to the overall dimension and not to the
active display area.
In the "PHYSICAL SPECIFICATIONS" section, the datasheet shows the
following parameters:

 ----------------------------------------------------------
|       Item        |         Specifications        | unit |
 ----------------------------------------------------------
| Display area      | 98.7 (W) x 57.5 (H)           |  mm  |
 ----------------------------------------------------------
| Overall dimension | 105.5(W) x 67.2(H) x 4.96(D)  |  mm  |
 ----------------------------------------------------------

Fixes: 966fea78adf2 ("drm/panel: simple: Add support for Ampire AM-480272H3TMQW-T01H")
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
[narmstrong: fixed Fixes commit id length]
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230516085039.3797303-1-dario.binacchi@amarulasolutions.com
---
 drivers/gpu/drm/panel/panel-simple.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 467117c0b2c9d..754c09c6d4c1b 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -759,8 +759,8 @@ static const struct panel_desc ampire_am_480272h3tmqw_t01h = {
 	.num_modes = 1,
 	.bpc = 8,
 	.size = {
-		.width = 105,
-		.height = 67,
+		.width = 99,
+		.height = 58,
 	},
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
 };
-- 
GitLab


From 35070674f488eefaeb753a13df11199d10328ff8 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Fri, 26 May 2023 09:27:16 +0200
Subject: [PATCH 0667/1791] Revert "dt-bindings: bridge: samsung-dsim: Make
 some flags optional"

This reverts commit cfaf76d349837f695c8aa6d7077847fec4231fe5 which was applied
without review due to a bad tool manipulation.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526-revert-bad-binding-v1-1-67329ad1bd80@linaro.org
---
 .../bindings/display/bridge/samsung,mipi-dsim.yaml       | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index 360fea81f4b6f..9f61ebdfefa85 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -70,9 +70,7 @@ properties:
   samsung,burst-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM high speed burst mode frequency when connected to devices
-      that support burst mode. If absent, the driver will use the pixel
-      clock from the attached device or bridge.
+      DSIM high speed burst mode frequency.
 
   samsung,esc-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
@@ -82,8 +80,7 @@ properties:
   samsung,pll-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM oscillator clock frequency. If absent, the driver will
-      use the clock frequency of sclk_mipi.
+      DSIM oscillator clock frequency.
 
   phys:
     maxItems: 1
@@ -137,7 +134,9 @@ required:
   - compatible
   - interrupts
   - reg
+  - samsung,burst-clock-frequency
   - samsung,esc-clock-frequency
+  - samsung,pll-clock-frequency
 
 allOf:
   - $ref: ../dsi-controller.yaml#
-- 
GitLab


From 35c812050ebdfe5ce576cf04d1d43d02dc2dfe19 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 24 May 2023 08:38:27 -0700
Subject: [PATCH 0668/1791] drm/i915: Fix clang -Wimplicit-fallthrough in
 intel_async_flip_check_hw()

Clang warns:

  drivers/gpu/drm/i915/display/intel_display.c:6012:3: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
                  case I915_FORMAT_MOD_X_TILED:
                  ^
  drivers/gpu/drm/i915/display/intel_display.c:6012:3: note: insert 'break;' to avoid fall-through
                  case I915_FORMAT_MOD_X_TILED:
                  ^
                  break;
  1 error generated.

Clang is a little more pedantic than GCC, which does not warn when
falling through to a case that is just break or return. Clang's version
is more in line with the kernel's own stance in deprecated.rst, which
states that all switch/case blocks must end in either break,
fallthrough, continue, goto, or return. Add the missing break to silence
the warning.

Fixes: 937859485aef ("drm/i915: Support Async Flip on Linear buffers")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/202305241902.UvHtMoxa-lkp@intel.com/
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Closes: https://lore.kernel.org/CA+G9fYv68V3ewK0Qj-syQj7qX-hQr0H1MFL=QFNuDoE_J2Zu-g@mail.gmail.com/
Reported-by: Tom Rix <trix@redhat.com>
Closes: https://lore.kernel.org/all/20230523125116.1669057-1-trix@redhat.com/
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524-intel_async_flip_check_hw-implicit-fallthrough-v1-1-83de89e376a1@kernel.org
---
 drivers/gpu/drm/i915/display/intel_display.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index e477e16ea58e1..38836e2e8807e 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -6008,6 +6008,7 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in
 					    plane->base.base.id, plane->base.name);
 				return -EINVAL;
 			}
+			break;
 
 		case I915_FORMAT_MOD_X_TILED:
 		case I915_FORMAT_MOD_Y_TILED:
-- 
GitLab


From 08264f85c5c05ecc38d409c84d48cfb00ccd3bc4 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 25 May 2023 12:49:42 +0300
Subject: [PATCH 0669/1791] drm/i915/gsc: use system include style for drm
 headers

Use <> instead of "" for including headers from include/.

Fixes: 8a9bf29546a1 ("drm/i915/gsc: add initial support for GSC proxy")
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525094942.941123-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
index ebee0b5a2c1dd..5f138de3c14f9 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
@@ -5,8 +5,8 @@
 
 #include <linux/component.h>
 
-#include "drm/i915_component.h"
-#include "drm/i915_gsc_proxy_mei_interface.h"
+#include <drm/i915_component.h>
+#include <drm/i915_gsc_proxy_mei_interface.h>
 
 #include "gt/intel_gt.h"
 #include "gt/intel_gt_print.h"
-- 
GitLab


From 34235f54b021ca75ab89d8af8d7f1fc420eaded0 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Fri, 21 Apr 2023 16:34:02 +0800
Subject: [PATCH 0670/1791] drm/stm: dsi: Use devm_platform_ioremap_resource()

Convert platform_get_resource(),devm_ioremap_resource() to a single call
to devm_platform_ioremap_resource(), as this is exactly what this function
does.

Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Acked-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Signed-off-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230421083402.21364-1-yang.lee@linux.alibaba.com
---
 drivers/gpu/drm/stm/dw_mipi_dsi-stm.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
index 89897d5f5c727..1750b6a25e871 100644
--- a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
+++ b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c
@@ -444,15 +444,13 @@ static int dw_mipi_dsi_stm_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct dw_mipi_dsi_stm *dsi;
 	struct clk *pclk;
-	struct resource *res;
 	int ret;
 
 	dsi = devm_kzalloc(dev, sizeof(*dsi), GFP_KERNEL);
 	if (!dsi)
 		return -ENOMEM;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	dsi->base = devm_ioremap_resource(dev, res);
+	dsi->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(dsi->base)) {
 		ret = PTR_ERR(dsi->base);
 		DRM_ERROR("Unable to get dsi registers %d\n", ret);
-- 
GitLab


From 898a9e3f56db9860ab091d4bf41b6caa99aafc3d Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Mon, 15 May 2023 14:38:18 +0200
Subject: [PATCH 0671/1791] drm/stm: ltdc: fix late dereference check

In ltdc_crtc_set_crc_source(), struct drm_crtc was dereferenced in a
container_of() before the pointer check. This could cause a kernel panic.

Fix this smatch warning:
drivers/gpu/drm/stm/ltdc.c:1124 ltdc_crtc_set_crc_source() warn: variable dereferenced before check 'crtc' (see line 1119)

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/lkml/202212241802.zeLFZCXB-lkp@intel.com/
Reported-by: Dan Carpenter <error27@gmail.com>
Closes: https://lore.kernel.org/lkml/202212241802.zeLFZCXB-lkp@intel.com/
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Signed-off-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230515123818.93971-1-raphael.gallais-pou@foss.st.com
---
 drivers/gpu/drm/stm/ltdc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index 03c6becda795c..b8be4c1db4235 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -1145,7 +1145,7 @@ static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc)
 
 static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 {
-	struct ltdc_device *ldev = crtc_to_ltdc(crtc);
+	struct ltdc_device *ldev;
 	int ret;
 
 	DRM_DEBUG_DRIVER("\n");
@@ -1153,6 +1153,8 @@ static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 	if (!crtc)
 		return -ENODEV;
 
+	ldev = crtc_to_ltdc(crtc);
+
 	if (source && strcmp(source, "auto") == 0) {
 		ldev->crc_active = true;
 		ret = regmap_set_bits(ldev->regmap, LTDC_GCR, GCR_CRCEN);
-- 
GitLab


From d9ce4e430790ba4f45e5febd7b4bd87b0f23563e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 13 Apr 2023 19:49:11 +0300
Subject: [PATCH 0672/1791] drm/i915: Expose crtc CTM property on ilk/snb
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The ilk/snb code is internally fully capable of handling the
CTM property, so expose it.

Note that we still choose not to expose DEGAMMA_LUT though.
The hardware is capable if degamma or gamma, but not both
simultanously due to lack of the split gamma mode. Exposing
both LUTs might encourage userspace to try enabling both
at the same time.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413164916.4221-3-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 744b3a4ec99a6..0f2c3c22cf59a 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -3473,7 +3473,7 @@ void intel_color_crtc_init(struct intel_crtc *crtc)
 
 	gamma_lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
 	degamma_lut_size = DISPLAY_INFO(i915)->color.degamma_lut_size;
-	has_ctm = degamma_lut_size != 0;
+	has_ctm = DISPLAY_VER(i915) >= 5 && !IS_VALLEYVIEW(i915);
 
 	/*
 	 * "DPALETTE_A: NOTE: The 8-bit (non-10-bit) mode is the
-- 
GitLab


From 2587c63aaaf47f6f4ea9d90047910534a3ecf432 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 13 Apr 2023 19:49:12 +0300
Subject: [PATCH 0673/1791] drm/i915: Fix CHV CGM CSC coefficient sign handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The CHV CGM CSC coefficients are in s4.12 two's complement
format. Fix the CTM->CGM conversion to handle that correctly
instead of pretending that the hw coefficients are also
in some sign-magnitude format.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413164916.4221-4-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 46 ++++++++++++++--------
 1 file changed, 29 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 0f2c3c22cf59a..7c14b2aba9e58 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -568,29 +568,41 @@ static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state)
 		icl_update_output_csc(crtc, &crtc_state->output_csc);
 }
 
-static void chv_cgm_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
-				    struct intel_csc_matrix *csc)
+static u16 ctm_to_twos_complement(u64 coeff, int int_bits, int frac_bits)
 {
-	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
-	int i;
+	s64 c = CTM_COEFF_ABS(coeff);
 
-	for (i = 0; i < 9; i++) {
-		u64 abs_coeff = ((1ULL << 63) - 1) & ctm->matrix[i];
+	/* leave an extra bit for rounding */
+	c >>= 32 - frac_bits - 1;
 
-		/* Round coefficient. */
-		abs_coeff += 1 << (32 - 13);
-		/* Clamp to hardware limits. */
-		abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1);
+	/* round and drop the extra bit */
+	c = (c + 1) >> 1;
 
-		csc->coeff[i] = 0;
+	if (CTM_COEFF_NEGATIVE(coeff))
+		c = -c;
 
-		/* Write coefficients in S3.12 format. */
-		if (ctm->matrix[i] & (1ULL << 63))
-			csc->coeff[i] |= 1 << 15;
+	c = clamp(c, -(s64)BIT(int_bits + frac_bits - 1),
+		  (s64)(BIT(int_bits + frac_bits - 1) - 1));
 
-		csc->coeff[i] |= ((abs_coeff >> 32) & 7) << 12;
-		csc->coeff[i] |= (abs_coeff >> 20) & 0xfff;
-	}
+	return c & (BIT(int_bits + frac_bits) - 1);
+}
+
+/*
+ * CHV Color Gamut Mapping (CGM) CSC
+ * |r|   | c0 c1 c2 |   |r|
+ * |g| = | c3 c4 c5 | x |g|
+ * |b|   | c6 c7 c8 |   |b|
+ *
+ * Coefficients are two's complement s4.12.
+ */
+static void chv_cgm_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
+				    struct intel_csc_matrix *csc)
+{
+	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
+	int i;
+
+	for (i = 0; i < 9; i++)
+		csc->coeff[i] = ctm_to_twos_complement(ctm->matrix[i], 4, 12);
 }
 
 static void chv_load_cgm_csc(struct intel_crtc *crtc,
-- 
GitLab


From 78776086371bf8aba4314a52a46c52d946984c61 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 13 Apr 2023 19:49:13 +0300
Subject: [PATCH 0674/1791] drm/i915: Always enable CGM CSC on CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On CHV toggling the CGM CSC on/off while the pipe is running leads
to underruns. Looks like we'd have to do the toggling strictly inside
the start_of_vblank-frame_start window to avoid this, but that window
is less than a scanline so there's no way we can guarantee hitting it.

As a workaround let's just leave the CGM CSC permanently enabled.
Fortunately the CGM gamma/degamma units don't seem to suffer from
this malady.

I also tried turning off CGM unit clock gating, but that did not
help.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413164916.4221-5-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 7c14b2aba9e58..9581dc912d8ea 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -605,6 +605,16 @@ static void chv_cgm_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
 		csc->coeff[i] = ctm_to_twos_complement(ctm->matrix[i], 4, 12);
 }
 
+#define CHV_CGM_CSC_COEFF_1_0 (1 << 12)
+
+static const struct intel_csc_matrix chv_cgm_csc_matrix_identity = {
+	.coeff = {
+		CHV_CGM_CSC_COEFF_1_0, 0, 0,
+		0, CHV_CGM_CSC_COEFF_1_0, 0,
+		0, 0, CHV_CGM_CSC_COEFF_1_0,
+	},
+};
+
 static void chv_load_cgm_csc(struct intel_crtc *crtc,
 			     const struct intel_csc_matrix *csc)
 {
@@ -667,9 +677,9 @@ static void chv_assign_csc(struct intel_crtc_state *crtc_state)
 
 		chv_cgm_csc_convert_ctm(crtc_state, &crtc_state->csc);
 	} else {
-		drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) != 0);
+		drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) == 0);
 
-		intel_csc_clear(&crtc_state->csc);
+		crtc_state->csc = chv_cgm_csc_matrix_identity;
 	}
 }
 
@@ -2033,6 +2043,13 @@ static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state)
 	    !lut_is_legacy(crtc_state->hw.gamma_lut))
 		cgm_mode |= CGM_PIPE_MODE_GAMMA;
 
+	/*
+	 * Toggling the CGM CSC on/off outside of the tiny window
+	 * between start of vblank and frame start causes underruns.
+	 * Always enable the CGM CSC as a workaround.
+	 */
+	cgm_mode |= CGM_PIPE_MODE_CSC;
+
 	return cgm_mode;
 }
 
-- 
GitLab


From 47d56cadb9e28da5b1bc5bad7e15120310aa118b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 13 Apr 2023 19:49:14 +0300
Subject: [PATCH 0675/1791] drm/i915: Implement CTM property support for VLV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

VLV has a so called "wide gamut color correction" unit (WGC).
What it is is a 3x3 matrix similar to the later CHV CGM
CSC, with less precisions/range. In fact CHV also has the WGC
but using it there doesn't really make sense when you have the
superior CGM CSC around.

Hook up the necessary stuff to expose the WGC as the CTM
crtc property.

One additional crazy idea that came to mind would be to use
the WGC as an output CSC on CHV for YCbCr output. But it
would be incompatible with the legacy LUT usage. In fact
since the WGC lacks post-offsets we'd probably have to
use the legacy LUT to do that final part of the RGB->YCbCr
conversion. Sounds doable, but perhaps not worth the hassle.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230413164916.4221-6-ville.syrjala@linux.intel.com
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
---
 drivers/gpu/drm/i915/display/intel_color.c    | 170 +++++++++++++++++-
 .../drm/i915/display/intel_crtc_state_dump.c  |   6 +-
 drivers/gpu/drm/i915/display/intel_display.c  |   8 +
 .../drm/i915/display/intel_display_types.h    |   3 +
 drivers/gpu/drm/i915/i915_reg.h               |  15 ++
 5 files changed, 198 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c
index 9581dc912d8ea..8966e65605168 100644
--- a/drivers/gpu/drm/i915/display/intel_color.c
+++ b/drivers/gpu/drm/i915/display/intel_color.c
@@ -587,6 +587,98 @@ static u16 ctm_to_twos_complement(u64 coeff, int int_bits, int frac_bits)
 	return c & (BIT(int_bits + frac_bits) - 1);
 }
 
+/*
+ * VLV/CHV Wide Gamut Color Correction (WGC) CSC
+ * |r|   | c0 c1 c2 |   |r|
+ * |g| = | c3 c4 c5 | x |g|
+ * |b|   | c6 c7 c8 |   |b|
+ *
+ * Coefficients are two's complement s2.10.
+ */
+static void vlv_wgc_csc_convert_ctm(const struct intel_crtc_state *crtc_state,
+				    struct intel_csc_matrix *csc)
+{
+	const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data;
+	int i;
+
+	for (i = 0; i < 9; i++)
+		csc->coeff[i] = ctm_to_twos_complement(ctm->matrix[i], 2, 10);
+}
+
+static void vlv_load_wgc_csc(struct intel_crtc *crtc,
+			     const struct intel_csc_matrix *csc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+
+	intel_de_write_fw(dev_priv, PIPE_WGC_C01_C00(pipe),
+			  csc->coeff[1] << 16 | csc->coeff[0]);
+	intel_de_write_fw(dev_priv, PIPE_WGC_C02(pipe),
+			  csc->coeff[2]);
+
+	intel_de_write_fw(dev_priv, PIPE_WGC_C11_C10(pipe),
+			  csc->coeff[4] << 16 | csc->coeff[3]);
+	intel_de_write_fw(dev_priv, PIPE_WGC_C12(pipe),
+			  csc->coeff[5]);
+
+	intel_de_write_fw(dev_priv, PIPE_WGC_C21_C20(pipe),
+			  csc->coeff[7] << 16 | csc->coeff[6]);
+	intel_de_write_fw(dev_priv, PIPE_WGC_C22(pipe),
+			  csc->coeff[8]);
+}
+
+static void vlv_read_wgc_csc(struct intel_crtc *crtc,
+			     struct intel_csc_matrix *csc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+	enum pipe pipe = crtc->pipe;
+	u32 tmp;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C01_C00(pipe));
+	csc->coeff[0] = tmp & 0xffff;
+	csc->coeff[1] = tmp >> 16;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C02(pipe));
+	csc->coeff[2] = tmp & 0xffff;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C11_C10(pipe));
+	csc->coeff[3] = tmp & 0xffff;
+	csc->coeff[4] = tmp >> 16;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C12(pipe));
+	csc->coeff[5] = tmp & 0xffff;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C21_C20(pipe));
+	csc->coeff[6] = tmp & 0xffff;
+	csc->coeff[7] = tmp >> 16;
+
+	tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C22(pipe));
+	csc->coeff[8] = tmp & 0xffff;
+}
+
+static void vlv_read_csc(struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	if (crtc_state->wgc_enable)
+		vlv_read_wgc_csc(crtc, &crtc_state->csc);
+}
+
+static void vlv_assign_csc(struct intel_crtc_state *crtc_state)
+{
+	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
+
+	if (crtc_state->hw.ctm) {
+		drm_WARN_ON(&i915->drm, !crtc_state->wgc_enable);
+
+		vlv_wgc_csc_convert_ctm(crtc_state, &crtc_state->csc);
+	} else {
+		drm_WARN_ON(&i915->drm, crtc_state->wgc_enable);
+
+		intel_csc_clear(&crtc_state->csc);
+	}
+}
+
 /*
  * CHV Color Gamut Mapping (CGM) CSC
  * |r|   | c0 c1 c2 |   |r|
@@ -672,6 +764,8 @@ static void chv_assign_csc(struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev);
 
+	drm_WARN_ON(&i915->drm, crtc_state->wgc_enable);
+
 	if (crtc_state->hw.ctm) {
 		drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) == 0);
 
@@ -1574,6 +1668,16 @@ static void icl_load_luts(const struct intel_crtc_state *crtc_state)
 	}
 }
 
+static void vlv_load_luts(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+
+	if (crtc_state->wgc_enable)
+		vlv_load_wgc_csc(crtc, &crtc_state->csc);
+
+	i965_load_luts(crtc_state);
+}
+
 static u32 chv_cgm_degamma_ldw(const struct drm_color_lut *color)
 {
 	return REG_FIELD_PREP(CGM_PIPE_DEGAMMA_GREEN_LDW_MASK, drm_color_lut_extract(color->green, 14)) |
@@ -1728,6 +1832,18 @@ static bool intel_can_preload_luts(const struct intel_crtc_state *new_crtc_state
 		!old_crtc_state->pre_csc_lut;
 }
 
+static bool vlv_can_preload_luts(const struct intel_crtc_state *new_crtc_state)
+{
+	struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc);
+	struct intel_atomic_state *state =
+		to_intel_atomic_state(new_crtc_state->uapi.state);
+	const struct intel_crtc_state *old_crtc_state =
+		intel_atomic_get_old_crtc_state(state, crtc);
+
+	return !old_crtc_state->wgc_enable &&
+		!old_crtc_state->post_csc_lut;
+}
+
 static bool chv_can_preload_luts(const struct intel_crtc_state *new_crtc_state)
 {
 	struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc);
@@ -1744,7 +1860,7 @@ static bool chv_can_preload_luts(const struct intel_crtc_state *new_crtc_state)
 	if (old_crtc_state->cgm_mode || new_crtc_state->cgm_mode)
 		return false;
 
-	return !old_crtc_state->post_csc_lut;
+	return vlv_can_preload_luts(new_crtc_state);
 }
 
 int intel_color_check(struct intel_crtc_state *crtc_state)
@@ -2031,6 +2147,39 @@ static int i9xx_color_check(struct intel_crtc_state *crtc_state)
 	return 0;
 }
 
+/*
+ * VLV color pipeline:
+ * u0.10 -> WGC csc -> u0.10 -> pipe gamma -> u0.10
+ */
+static int vlv_color_check(struct intel_crtc_state *crtc_state)
+{
+	int ret;
+
+	ret = check_luts(crtc_state);
+	if (ret)
+		return ret;
+
+	crtc_state->gamma_enable =
+		crtc_state->hw.gamma_lut &&
+		!crtc_state->c8_planes;
+
+	crtc_state->gamma_mode = i9xx_gamma_mode(crtc_state);
+
+	crtc_state->wgc_enable = crtc_state->hw.ctm;
+
+	ret = intel_color_add_affected_planes(crtc_state);
+	if (ret)
+		return ret;
+
+	intel_assign_luts(crtc_state);
+
+	vlv_assign_csc(crtc_state);
+
+	crtc_state->preload_luts = vlv_can_preload_luts(crtc_state);
+
+	return 0;
+}
+
 static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state)
 {
 	u32 cgm_mode = 0;
@@ -2081,6 +2230,12 @@ static int chv_color_check(struct intel_crtc_state *crtc_state)
 
 	crtc_state->cgm_mode = chv_cgm_mode(crtc_state);
 
+	/*
+	 * We always bypass the WGC CSC and use the CGM CSC
+	 * instead since it has degamma and better precision.
+	 */
+	crtc_state->wgc_enable = false;
+
 	ret = intel_color_add_affected_planes(crtc_state);
 	if (ret)
 		return ret;
@@ -3395,6 +3550,15 @@ static const struct intel_color_funcs chv_color_funcs = {
 	.read_csc = chv_read_csc,
 };
 
+static const struct intel_color_funcs vlv_color_funcs = {
+	.color_check = vlv_color_check,
+	.color_commit_arm = i9xx_color_commit_arm,
+	.load_luts = vlv_load_luts,
+	.read_luts = i965_read_luts,
+	.lut_equal = i965_lut_equal,
+	.read_csc = vlv_read_csc,
+};
+
 static const struct intel_color_funcs i965_color_funcs = {
 	.color_check = i9xx_color_check,
 	.color_commit_arm = i9xx_color_commit_arm,
@@ -3502,7 +3666,7 @@ void intel_color_crtc_init(struct intel_crtc *crtc)
 
 	gamma_lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size;
 	degamma_lut_size = DISPLAY_INFO(i915)->color.degamma_lut_size;
-	has_ctm = DISPLAY_VER(i915) >= 5 && !IS_VALLEYVIEW(i915);
+	has_ctm = DISPLAY_VER(i915) >= 5;
 
 	/*
 	 * "DPALETTE_A: NOTE: The 8-bit (non-10-bit) mode is the
@@ -3541,6 +3705,8 @@ void intel_color_init_hooks(struct drm_i915_private *i915)
 	if (HAS_GMCH(i915)) {
 		if (IS_CHERRYVIEW(i915))
 			i915->display.funcs.color = &chv_color_funcs;
+		else if (IS_VALLEYVIEW(i915))
+			i915->display.funcs.color = &vlv_color_funcs;
 		else if (DISPLAY_VER(i915) >= 4)
 			i915->display.funcs.color = &i965_color_funcs;
 		else
diff --git a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
index 27d7bab464274..8d4640d0fd346 100644
--- a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
+++ b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c
@@ -184,7 +184,7 @@ ilk_dump_csc(struct drm_i915_private *i915, const char *name,
 }
 
 static void
-chv_dump_csc(struct drm_i915_private *i915, const char *name,
+vlv_dump_csc(struct drm_i915_private *i915, const char *name,
 	     const struct intel_csc_matrix *csc)
 {
 	int i;
@@ -371,7 +371,9 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config,
 	if (!HAS_GMCH(i915))
 		ilk_dump_csc(i915, "pipe csc", &pipe_config->csc);
 	else if (IS_CHERRYVIEW(i915))
-		chv_dump_csc(i915, "cgm csc", &pipe_config->csc);
+		vlv_dump_csc(i915, "cgm csc", &pipe_config->csc);
+	else if (IS_VALLEYVIEW(i915))
+		vlv_dump_csc(i915, "wgc csc", &pipe_config->csc);
 
 dump_planes:
 	if (!state)
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 38836e2e8807e..f51a55f4e9d0a 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -2740,6 +2740,9 @@ void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state)
 
 	val |= TRANSCONF_GAMMA_MODE(crtc_state->gamma_mode);
 
+	if (crtc_state->wgc_enable)
+		val |= TRANSCONF_WGC_ENABLE;
+
 	val |= TRANSCONF_FRAME_START_DELAY(crtc_state->framestart_delay - 1);
 
 	intel_de_write(dev_priv, TRANSCONF(cpu_transcoder), val);
@@ -2934,6 +2937,10 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 
 	pipe_config->framestart_delay = REG_FIELD_GET(TRANSCONF_FRAME_START_DELAY_MASK, tmp) + 1;
 
+	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	    (tmp & TRANSCONF_WGC_ENABLE))
+		pipe_config->wgc_enable = true;
+
 	if (IS_CHERRYVIEW(dev_priv))
 		pipe_config->cgm_mode = intel_de_read(dev_priv,
 						      CGM_PIPE_MODE(crtc->pipe));
@@ -5277,6 +5284,7 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config,
 			PIPE_CONF_CHECK_X(csc_mode);
 		PIPE_CONF_CHECK_BOOL(gamma_enable);
 		PIPE_CONF_CHECK_BOOL(csc_enable);
+		PIPE_CONF_CHECK_BOOL(wgc_enable);
 
 		PIPE_CONF_CHECK_I(linetime);
 		PIPE_CONF_CHECK_I(ips_linetime);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 88b2a55d19f21..2d8297f8d0889 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1343,6 +1343,9 @@ struct intel_crtc_state {
 	/* enable pipe csc? */
 	bool csc_enable;
 
+	/* enable vlv/chv wgc csc? */
+	bool wgc_enable;
+
 	/* big joiner pipe bitmask */
 	u8 bigjoiner_pipes;
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f1ba1eae26ca5..0523418129c51 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2544,6 +2544,7 @@
 #define   TRANSCONF_MSA_TIMING_DELAY_MASK	REG_GENMASK(19, 18) /* ilk/snb/ivb */
 #define   TRANSCONF_MSA_TIMING_DELAY(x)		REG_FIELD_PREP(TRANSCONF_MSA_TIMING_DELAY_MASK, (x))
 #define   TRANSCONF_CXSR_DOWNCLOCK		REG_BIT(16)
+#define   TRANSCONF_WGC_ENABLE			REG_BIT(15) /* vlv/chv only */
 #define   TRANSCONF_REFRESH_RATE_ALT_VLV		REG_BIT(14)
 #define   TRANSCONF_COLOR_RANGE_SELECT		REG_BIT(13)
 #define   TRANSCONF_OUTPUT_COLORSPACE_MASK	REG_GENMASK(12, 11) /* ilk-ivb */
@@ -6438,6 +6439,20 @@ enum skl_power_gate {
 							    (index) * 4, _PLANE_CSC_POSTOFF_HI_2(pipe) + \
 							    (index) * 4)
 
+#define _PIPE_A_WGC_C01_C00	0x600B0 /* s2.10 */
+#define _PIPE_A_WGC_C02		0x600B4 /* s2.10 */
+#define _PIPE_A_WGC_C11_C10	0x600B8 /* s2.10 */
+#define _PIPE_A_WGC_C12		0x600BC /* s2.10 */
+#define _PIPE_A_WGC_C21_C20	0x600C0 /* s2.10 */
+#define _PIPE_A_WGC_C22		0x600C4 /* s2.10 */
+
+#define PIPE_WGC_C01_C00(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C01_C00)
+#define PIPE_WGC_C02(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C02)
+#define PIPE_WGC_C11_C10(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C11_C10)
+#define PIPE_WGC_C12(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C12)
+#define PIPE_WGC_C21_C20(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C21_C20)
+#define PIPE_WGC_C22(pipe)		_MMIO_TRANS2(pipe, _PIPE_A_WGC_C22)
+
 /* pipe CSC & degamma/gamma LUTs on CHV */
 #define _CGM_PIPE_A_CSC_COEFF01	(VLV_DISPLAY_BASE + 0x67900)
 #define _CGM_PIPE_A_CSC_COEFF23	(VLV_DISPLAY_BASE + 0x67904)
-- 
GitLab


From fff8013f15204a902977e7520b0ef7648a655aa9 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:45 +0800
Subject: [PATCH 0676/1791] dt-bindings: lcdif: Add i.MX93 LCDIF support

There is one LCDIF embedded in i.MX93 SoC to connect with
MIPI DSI controller through LCDIF cross line pattern(controlled
by mediamix blk-ctrl) or connect with LVDS display bridge(LDB)
directly or connect with a parallel display through parallel
display format(also controlled by mediamix blk-ctrl).  i.MX93
LCDIF IP is essentially the same to i.MX8MP LCDIF IP.  Add device
tree binding for i.MX93 LCDIF.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-2-victor.liu@nxp.com
---
 Documentation/devicetree/bindings/display/fsl,lcdif.yaml | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/fsl,lcdif.yaml b/Documentation/devicetree/bindings/display/fsl,lcdif.yaml
index 75b4efd70ba85..fc11ab5fc4654 100644
--- a/Documentation/devicetree/bindings/display/fsl,lcdif.yaml
+++ b/Documentation/devicetree/bindings/display/fsl,lcdif.yaml
@@ -21,6 +21,7 @@ properties:
           - fsl,imx28-lcdif
           - fsl,imx6sx-lcdif
           - fsl,imx8mp-lcdif
+          - fsl,imx93-lcdif
       - items:
           - enum:
               - fsl,imx6sl-lcdif
@@ -88,7 +89,9 @@ allOf:
       properties:
         compatible:
           contains:
-            const: fsl,imx8mp-lcdif
+            enum:
+              - fsl,imx8mp-lcdif
+              - fsl,imx93-lcdif
     then:
       properties:
         clocks:
@@ -107,6 +110,7 @@ allOf:
               enum:
                 - fsl,imx6sx-lcdif
                 - fsl,imx8mp-lcdif
+                - fsl,imx93-lcdif
     then:
       properties:
         clocks:
@@ -123,6 +127,7 @@ allOf:
               - fsl,imx8mm-lcdif
               - fsl,imx8mn-lcdif
               - fsl,imx8mp-lcdif
+              - fsl,imx93-lcdif
     then:
       required:
         - power-domains
-- 
GitLab


From 3f07432025890f1a49c30132420fb0e623a84a20 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:46 +0800
Subject: [PATCH 0677/1791] drm: lcdif: Drop unnecessary NULL pointer check on
 lcdif->bridge

A valid bridge is already found in lcdif_attach_bridge() and set
to lcdif->bridge, so lcdif->bridge cannot be a NULL pointer. Drop
the unnecessary NULL pointer check in KMS stage.

Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-3-victor.liu@nxp.com
---
 drivers/gpu/drm/mxsfb/lcdif_kms.c | 36 +++++++++++++------------------
 1 file changed, 15 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_kms.c b/drivers/gpu/drm/mxsfb/lcdif_kms.c
index 262bc43b10791..ba00a155d9b6a 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_kms.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_kms.c
@@ -394,7 +394,7 @@ static void lcdif_crtc_mode_set_nofb(struct lcdif_drm_private *lcdif,
 	struct drm_display_mode *m = &lcdif->crtc.state->adjusted_mode;
 	u32 bus_flags = 0;
 
-	if (lcdif->bridge && lcdif->bridge->timings)
+	if (lcdif->bridge->timings)
 		bus_flags = lcdif->bridge->timings->input_bus_flags;
 	else if (bridge_state)
 		bus_flags = bridge_state->input_bus_cfg.flags;
@@ -463,30 +463,24 @@ static void lcdif_crtc_atomic_enable(struct drm_crtc *crtc,
 	struct drm_display_mode *m = &lcdif->crtc.state->adjusted_mode;
 	struct drm_bridge_state *bridge_state = NULL;
 	struct drm_device *drm = lcdif->drm;
-	u32 bus_format = 0;
+	u32 bus_format;
 	dma_addr_t paddr;
 
-	/* If there is a bridge attached to the LCDIF, use its bus format */
-	if (lcdif->bridge) {
-		bridge_state =
-			drm_atomic_get_new_bridge_state(state,
-							lcdif->bridge);
-		if (!bridge_state)
-			bus_format = MEDIA_BUS_FMT_FIXED;
-		else
-			bus_format = bridge_state->input_bus_cfg.format;
-
-		if (bus_format == MEDIA_BUS_FMT_FIXED) {
-			dev_warn_once(drm->dev,
-				      "Bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
-				      "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n");
-			bus_format = MEDIA_BUS_FMT_RGB888_1X24;
-		}
-	}
+	bridge_state = drm_atomic_get_new_bridge_state(state, lcdif->bridge);
+	if (!bridge_state)
+		bus_format = MEDIA_BUS_FMT_FIXED;
+	else
+		bus_format = bridge_state->input_bus_cfg.format;
 
-	/* If all else fails, default to RGB888_1X24 */
-	if (!bus_format)
+	if (bus_format == MEDIA_BUS_FMT_FIXED) {
+		dev_warn_once(drm->dev,
+			      "Bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
+			      "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n");
 		bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	} else if (!bus_format) {
+		/* If all else fails, default to RGB888_1X24 */
+		bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	}
 
 	clk_set_rate(lcdif->clk, m->crtc_clock * 1000);
 
-- 
GitLab


From 5befcdcaf062abcd362c2e761b069424d9e3b2a4 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:47 +0800
Subject: [PATCH 0678/1791] drm: lcdif: Determine bus format and flags in
 ->atomic_check()

Instead of determining LCDIF output bus format and bus flags in
->atomic_enable(), do that in ->atomic_check().  This is a
preparation for the upcoming patch to check consistent bus format
and bus flags across all first downstream bridges in ->atomic_check().
New lcdif_crtc_state structure is introduced to cache bus format
and bus flags states in ->atomic_check() so that they can be read
in ->atomic_enable().

Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-4-victor.liu@nxp.com
---
 drivers/gpu/drm/mxsfb/lcdif_kms.c | 140 ++++++++++++++++++++++--------
 1 file changed, 102 insertions(+), 38 deletions(-)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_kms.c b/drivers/gpu/drm/mxsfb/lcdif_kms.c
index ba00a155d9b6a..fab2253e47864 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_kms.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_kms.c
@@ -30,6 +30,18 @@
 #include "lcdif_drv.h"
 #include "lcdif_regs.h"
 
+struct lcdif_crtc_state {
+	struct drm_crtc_state	base;	/* always be the first member */
+	u32			bus_format;
+	u32			bus_flags;
+};
+
+static inline struct lcdif_crtc_state *
+to_lcdif_crtc_state(struct drm_crtc_state *s)
+{
+	return container_of(s, struct lcdif_crtc_state, base);
+}
+
 /* -----------------------------------------------------------------------------
  * CRTC
  */
@@ -385,48 +397,75 @@ static void lcdif_reset_block(struct lcdif_drm_private *lcdif)
 	readl(lcdif->base + LCDC_V8_CTRL);
 }
 
-static void lcdif_crtc_mode_set_nofb(struct lcdif_drm_private *lcdif,
-				     struct drm_plane_state *plane_state,
-				     struct drm_bridge_state *bridge_state,
-				     const u32 bus_format)
+static void lcdif_crtc_mode_set_nofb(struct drm_crtc_state *crtc_state,
+				     struct drm_plane_state *plane_state)
 {
-	struct drm_device *drm = lcdif->crtc.dev;
-	struct drm_display_mode *m = &lcdif->crtc.state->adjusted_mode;
-	u32 bus_flags = 0;
-
-	if (lcdif->bridge->timings)
-		bus_flags = lcdif->bridge->timings->input_bus_flags;
-	else if (bridge_state)
-		bus_flags = bridge_state->input_bus_cfg.flags;
+	struct lcdif_crtc_state *lcdif_crtc_state = to_lcdif_crtc_state(crtc_state);
+	struct drm_device *drm = crtc_state->crtc->dev;
+	struct lcdif_drm_private *lcdif = to_lcdif_drm_private(drm);
+	struct drm_display_mode *m = &crtc_state->adjusted_mode;
 
 	DRM_DEV_DEBUG_DRIVER(drm->dev, "Pixel clock: %dkHz (actual: %dkHz)\n",
 			     m->crtc_clock,
 			     (int)(clk_get_rate(lcdif->clk) / 1000));
 	DRM_DEV_DEBUG_DRIVER(drm->dev, "Bridge bus_flags: 0x%08X\n",
-			     bus_flags);
+			     lcdif_crtc_state->bus_flags);
 	DRM_DEV_DEBUG_DRIVER(drm->dev, "Mode flags: 0x%08X\n", m->flags);
 
 	/* Mandatory eLCDIF reset as per the Reference Manual */
 	lcdif_reset_block(lcdif);
 
-	lcdif_set_formats(lcdif, plane_state, bus_format);
+	lcdif_set_formats(lcdif, plane_state, lcdif_crtc_state->bus_format);
 
-	lcdif_set_mode(lcdif, bus_flags);
+	lcdif_set_mode(lcdif, lcdif_crtc_state->bus_flags);
 }
 
 static int lcdif_crtc_atomic_check(struct drm_crtc *crtc,
 				   struct drm_atomic_state *state)
 {
+	struct drm_device *drm = crtc->dev;
+	struct lcdif_drm_private *lcdif = to_lcdif_drm_private(drm);
 	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state,
 									  crtc);
+	struct lcdif_crtc_state *lcdif_crtc_state = to_lcdif_crtc_state(crtc_state);
 	bool has_primary = crtc_state->plane_mask &
 			   drm_plane_mask(crtc->primary);
+	struct drm_bridge_state *bridge_state;
+	struct drm_bridge *bridge = lcdif->bridge;
+	int ret;
 
 	/* The primary plane has to be enabled when the CRTC is active. */
 	if (crtc_state->active && !has_primary)
 		return -EINVAL;
 
-	return drm_atomic_add_affected_planes(state, crtc);
+	ret = drm_atomic_add_affected_planes(state, crtc);
+	if (ret)
+		return ret;
+
+	bridge_state = drm_atomic_get_new_bridge_state(state, bridge);
+	if (!bridge_state)
+		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_FIXED;
+	else
+		lcdif_crtc_state->bus_format = bridge_state->input_bus_cfg.format;
+
+	if (lcdif_crtc_state->bus_format == MEDIA_BUS_FMT_FIXED) {
+		dev_warn_once(drm->dev,
+			      "Bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
+			      "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n");
+		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	} else if (!lcdif_crtc_state->bus_format) {
+		/* If all else fails, default to RGB888_1X24 */
+		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	}
+
+	if (bridge->timings)
+		lcdif_crtc_state->bus_flags = bridge->timings->input_bus_flags;
+	else if (bridge_state)
+		lcdif_crtc_state->bus_flags = bridge_state->input_bus_cfg.flags;
+	else
+		lcdif_crtc_state->bus_flags = 0;
+
+	return 0;
 }
 
 static void lcdif_crtc_atomic_flush(struct drm_crtc *crtc,
@@ -458,35 +497,18 @@ static void lcdif_crtc_atomic_enable(struct drm_crtc *crtc,
 				     struct drm_atomic_state *state)
 {
 	struct lcdif_drm_private *lcdif = to_lcdif_drm_private(crtc->dev);
+	struct drm_crtc_state *new_cstate = drm_atomic_get_new_crtc_state(state, crtc);
 	struct drm_plane_state *new_pstate = drm_atomic_get_new_plane_state(state,
 									    crtc->primary);
 	struct drm_display_mode *m = &lcdif->crtc.state->adjusted_mode;
-	struct drm_bridge_state *bridge_state = NULL;
 	struct drm_device *drm = lcdif->drm;
-	u32 bus_format;
 	dma_addr_t paddr;
 
-	bridge_state = drm_atomic_get_new_bridge_state(state, lcdif->bridge);
-	if (!bridge_state)
-		bus_format = MEDIA_BUS_FMT_FIXED;
-	else
-		bus_format = bridge_state->input_bus_cfg.format;
-
-	if (bus_format == MEDIA_BUS_FMT_FIXED) {
-		dev_warn_once(drm->dev,
-			      "Bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
-			      "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n");
-		bus_format = MEDIA_BUS_FMT_RGB888_1X24;
-	} else if (!bus_format) {
-		/* If all else fails, default to RGB888_1X24 */
-		bus_format = MEDIA_BUS_FMT_RGB888_1X24;
-	}
-
 	clk_set_rate(lcdif->clk, m->crtc_clock * 1000);
 
 	pm_runtime_get_sync(drm->dev);
 
-	lcdif_crtc_mode_set_nofb(lcdif, new_pstate, bridge_state, bus_format);
+	lcdif_crtc_mode_set_nofb(new_cstate, new_pstate);
 
 	/* Write cur_buf as well to avoid an initial corrupt frame */
 	paddr = drm_fb_dma_get_gem_addr(new_pstate->fb, new_pstate, 0);
@@ -523,6 +545,48 @@ static void lcdif_crtc_atomic_disable(struct drm_crtc *crtc,
 	pm_runtime_put_sync(drm->dev);
 }
 
+static void lcdif_crtc_atomic_destroy_state(struct drm_crtc *crtc,
+					    struct drm_crtc_state *state)
+{
+	__drm_atomic_helper_crtc_destroy_state(state);
+	kfree(to_lcdif_crtc_state(state));
+}
+
+static void lcdif_crtc_reset(struct drm_crtc *crtc)
+{
+	struct lcdif_crtc_state *state;
+
+	if (crtc->state)
+		lcdif_crtc_atomic_destroy_state(crtc, crtc->state);
+
+	crtc->state = NULL;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state)
+		__drm_atomic_helper_crtc_reset(crtc, &state->base);
+}
+
+static struct drm_crtc_state *
+lcdif_crtc_atomic_duplicate_state(struct drm_crtc *crtc)
+{
+	struct lcdif_crtc_state *old = to_lcdif_crtc_state(crtc->state);
+	struct lcdif_crtc_state *new;
+
+	if (WARN_ON(!crtc->state))
+		return NULL;
+
+	new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return NULL;
+
+	__drm_atomic_helper_crtc_duplicate_state(crtc, &new->base);
+
+	new->bus_format = old->bus_format;
+	new->bus_flags = old->bus_flags;
+
+	return &new->base;
+}
+
 static int lcdif_crtc_enable_vblank(struct drm_crtc *crtc)
 {
 	struct lcdif_drm_private *lcdif = to_lcdif_drm_private(crtc->dev);
@@ -551,12 +615,12 @@ static const struct drm_crtc_helper_funcs lcdif_crtc_helper_funcs = {
 };
 
 static const struct drm_crtc_funcs lcdif_crtc_funcs = {
-	.reset = drm_atomic_helper_crtc_reset,
+	.reset = lcdif_crtc_reset,
 	.destroy = drm_crtc_cleanup,
 	.set_config = drm_atomic_helper_set_config,
 	.page_flip = drm_atomic_helper_page_flip,
-	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+	.atomic_duplicate_state = lcdif_crtc_atomic_duplicate_state,
+	.atomic_destroy_state = lcdif_crtc_atomic_destroy_state,
 	.enable_vblank = lcdif_crtc_enable_vblank,
 	.disable_vblank = lcdif_crtc_disable_vblank,
 };
-- 
GitLab


From c62a7b9ca72472f66c86db69bb0d1bb45c3f372b Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:48 +0800
Subject: [PATCH 0679/1791] drm: lcdif: Check consistent bus format and flags
 across first bridges

The single LCDIF embedded in i.MX93 SoC may drive multiple displays
simultaneously.  Check bus format and flags across first bridges in
->atomic_check() to ensure they are consistent.  This is a preparation
for adding i.MX93 LCDIF support.

Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-5-victor.liu@nxp.com
---
 drivers/gpu/drm/mxsfb/lcdif_drv.c |  2 -
 drivers/gpu/drm/mxsfb/lcdif_drv.h |  1 -
 drivers/gpu/drm/mxsfb/lcdif_kms.c | 82 ++++++++++++++++++++++---------
 3 files changed, 58 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.c b/drivers/gpu/drm/mxsfb/lcdif_drv.c
index 6fb5b469ee5a0..e816f87828fbc 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.c
@@ -52,8 +52,6 @@ static int lcdif_attach_bridge(struct lcdif_drm_private *lcdif)
 	if (ret)
 		return dev_err_probe(drm->dev, ret, "Failed to attach bridge\n");
 
-	lcdif->bridge = bridge;
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.h b/drivers/gpu/drm/mxsfb/lcdif_drv.h
index 6cdba6e20c02b..aa6d099a1897c 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.h
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.h
@@ -31,7 +31,6 @@ struct lcdif_drm_private {
 	} planes;
 	struct drm_crtc			crtc;
 	struct drm_encoder		encoder;
-	struct drm_bridge		*bridge;
 };
 
 static inline struct lcdif_drm_private *
diff --git a/drivers/gpu/drm/mxsfb/lcdif_kms.c b/drivers/gpu/drm/mxsfb/lcdif_kms.c
index fab2253e47864..e13382beb53e8 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_kms.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_kms.c
@@ -17,6 +17,7 @@
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
 #include <drm/drm_color_mgmt.h>
+#include <drm/drm_connector.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fb_dma_helper.h>
@@ -424,15 +425,19 @@ static int lcdif_crtc_atomic_check(struct drm_crtc *crtc,
 				   struct drm_atomic_state *state)
 {
 	struct drm_device *drm = crtc->dev;
-	struct lcdif_drm_private *lcdif = to_lcdif_drm_private(drm);
 	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state,
 									  crtc);
 	struct lcdif_crtc_state *lcdif_crtc_state = to_lcdif_crtc_state(crtc_state);
 	bool has_primary = crtc_state->plane_mask &
 			   drm_plane_mask(crtc->primary);
+	struct drm_connector_state *connector_state;
+	struct drm_connector *connector;
+	struct drm_encoder *encoder;
 	struct drm_bridge_state *bridge_state;
-	struct drm_bridge *bridge = lcdif->bridge;
-	int ret;
+	struct drm_bridge *bridge;
+	u32 bus_format, bus_flags;
+	bool format_set = false, flags_set = false;
+	int ret, i;
 
 	/* The primary plane has to be enabled when the CRTC is active. */
 	if (crtc_state->active && !has_primary)
@@ -442,29 +447,58 @@ static int lcdif_crtc_atomic_check(struct drm_crtc *crtc,
 	if (ret)
 		return ret;
 
-	bridge_state = drm_atomic_get_new_bridge_state(state, bridge);
-	if (!bridge_state)
-		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_FIXED;
-	else
-		lcdif_crtc_state->bus_format = bridge_state->input_bus_cfg.format;
-
-	if (lcdif_crtc_state->bus_format == MEDIA_BUS_FMT_FIXED) {
-		dev_warn_once(drm->dev,
-			      "Bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
-			      "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n");
-		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_RGB888_1X24;
-	} else if (!lcdif_crtc_state->bus_format) {
-		/* If all else fails, default to RGB888_1X24 */
-		lcdif_crtc_state->bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+	/* Try to find consistent bus format and flags across first bridges. */
+	for_each_new_connector_in_state(state, connector, connector_state, i) {
+		if (!connector_state->crtc)
+			continue;
+
+		encoder = connector_state->best_encoder;
+
+		bridge = drm_bridge_chain_get_first_bridge(encoder);
+		if (!bridge)
+			continue;
+
+		bridge_state = drm_atomic_get_new_bridge_state(state, bridge);
+		if (!bridge_state)
+			bus_format = MEDIA_BUS_FMT_FIXED;
+		else
+			bus_format = bridge_state->input_bus_cfg.format;
+
+		if (bus_format == MEDIA_BUS_FMT_FIXED) {
+			dev_warn(drm->dev,
+				 "[ENCODER:%d:%s]'s bridge does not provide bus format, assuming MEDIA_BUS_FMT_RGB888_1X24.\n"
+				 "Please fix bridge driver by handling atomic_get_input_bus_fmts.\n",
+				 encoder->base.id, encoder->name);
+			bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+		} else if (!bus_format) {
+			/* If all else fails, default to RGB888_1X24 */
+			bus_format = MEDIA_BUS_FMT_RGB888_1X24;
+		}
+
+		if (!format_set) {
+			lcdif_crtc_state->bus_format = bus_format;
+			format_set = true;
+		} else if (lcdif_crtc_state->bus_format != bus_format) {
+			DRM_DEV_DEBUG_DRIVER(drm->dev, "inconsistent bus format\n");
+			return -EINVAL;
+		}
+
+		if (bridge->timings)
+			bus_flags = bridge->timings->input_bus_flags;
+		else if (bridge_state)
+			bus_flags = bridge_state->input_bus_cfg.flags;
+		else
+			bus_flags = 0;
+
+		if (!flags_set) {
+			lcdif_crtc_state->bus_flags = bus_flags;
+			flags_set = true;
+		} else if (lcdif_crtc_state->bus_flags != bus_flags) {
+			DRM_DEV_DEBUG_DRIVER(drm->dev, "inconsistent bus flags\n");
+			return -EINVAL;
+		}
 	}
 
-	if (bridge->timings)
-		lcdif_crtc_state->bus_flags = bridge->timings->input_bus_flags;
-	else if (bridge_state)
-		lcdif_crtc_state->bus_flags = bridge_state->input_bus_cfg.flags;
-	else
-		lcdif_crtc_state->bus_flags = 0;
-
 	return 0;
 }
 
-- 
GitLab


From dbb32d8564de455b7cd47482d8d65bc70ee44927 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:49 +0800
Subject: [PATCH 0680/1791] drm: lcdif: Add multiple encoders and first bridges
 support

The single LCDIF embedded in i.MX93 SoC may drive multiple displays
simultaneously.  Look at LCDIF output port's remote port parents to
find all enabled first bridges.  Add an encoder for each found bridge
and attach the bridge to the encoder.  This is a preparation for
adding i.MX93 LCDIF support.

Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Acked-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-6-victor.liu@nxp.com
---
 drivers/gpu/drm/mxsfb/lcdif_drv.c | 70 +++++++++++++++++++++++++++----
 drivers/gpu/drm/mxsfb/lcdif_drv.h |  2 -
 drivers/gpu/drm/mxsfb/lcdif_kms.c | 21 ++--------
 3 files changed, 65 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.c b/drivers/gpu/drm/mxsfb/lcdif_drv.c
index e816f87828fbc..dc85498b81cdb 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.c
@@ -9,13 +9,16 @@
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
+#include <linux/of_graph.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_bridge.h>
 #include <drm/drm_drv.h>
+#include <drm/drm_encoder.h>
 #include <drm/drm_fbdev_dma.h>
 #include <drm/drm_gem_dma_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
@@ -38,19 +41,70 @@ static const struct drm_mode_config_helper_funcs lcdif_mode_config_helpers = {
 	.atomic_commit_tail = drm_atomic_helper_commit_tail_rpm,
 };
 
+static const struct drm_encoder_funcs lcdif_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
 static int lcdif_attach_bridge(struct lcdif_drm_private *lcdif)
 {
-	struct drm_device *drm = lcdif->drm;
+	struct device *dev = lcdif->drm->dev;
+	struct device_node *ep;
 	struct drm_bridge *bridge;
 	int ret;
 
-	bridge = devm_drm_of_get_bridge(drm->dev, drm->dev->of_node, 0, 0);
-	if (IS_ERR(bridge))
-		return PTR_ERR(bridge);
-
-	ret = drm_bridge_attach(&lcdif->encoder, bridge, NULL, 0);
-	if (ret)
-		return dev_err_probe(drm->dev, ret, "Failed to attach bridge\n");
+	for_each_endpoint_of_node(dev->of_node, ep) {
+		struct device_node *remote;
+		struct of_endpoint of_ep;
+		struct drm_encoder *encoder;
+
+		remote = of_graph_get_remote_port_parent(ep);
+		if (!of_device_is_available(remote)) {
+			of_node_put(remote);
+			continue;
+		}
+		of_node_put(remote);
+
+		ret = of_graph_parse_endpoint(ep, &of_ep);
+		if (ret < 0) {
+			dev_err(dev, "Failed to parse endpoint %pOF\n", ep);
+			of_node_put(ep);
+			return ret;
+		}
+
+		bridge = devm_drm_of_get_bridge(dev, dev->of_node, 0, of_ep.id);
+		if (IS_ERR(bridge)) {
+			of_node_put(ep);
+			return dev_err_probe(dev, PTR_ERR(bridge),
+					     "Failed to get bridge for endpoint%u\n",
+					     of_ep.id);
+		}
+
+		encoder = devm_kzalloc(dev, sizeof(*encoder), GFP_KERNEL);
+		if (!encoder) {
+			dev_err(dev, "Failed to allocate encoder for endpoint%u\n",
+				of_ep.id);
+			of_node_put(ep);
+			return -ENOMEM;
+		}
+
+		encoder->possible_crtcs = drm_crtc_mask(&lcdif->crtc);
+		ret = drm_encoder_init(lcdif->drm, encoder, &lcdif_encoder_funcs,
+				       DRM_MODE_ENCODER_NONE, NULL);
+		if (ret) {
+			dev_err(dev, "Failed to initialize encoder for endpoint%u: %d\n",
+				of_ep.id, ret);
+			of_node_put(ep);
+			return ret;
+		}
+
+		ret = drm_bridge_attach(encoder, bridge, NULL, 0);
+		if (ret) {
+			of_node_put(ep);
+			return dev_err_probe(dev, ret,
+					     "Failed to attach bridge for endpoint%u\n",
+					     of_ep.id);
+		}
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.h b/drivers/gpu/drm/mxsfb/lcdif_drv.h
index aa6d099a1897c..61483a4e058dd 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.h
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.h
@@ -11,7 +11,6 @@
 #include <drm/drm_bridge.h>
 #include <drm/drm_crtc.h>
 #include <drm/drm_device.h>
-#include <drm/drm_encoder.h>
 #include <drm/drm_plane.h>
 
 struct clk;
@@ -30,7 +29,6 @@ struct lcdif_drm_private {
 		/* i.MXRT does support overlay planes, add them here. */
 	} planes;
 	struct drm_crtc			crtc;
-	struct drm_encoder		encoder;
 };
 
 static inline struct lcdif_drm_private *
diff --git a/drivers/gpu/drm/mxsfb/lcdif_kms.c b/drivers/gpu/drm/mxsfb/lcdif_kms.c
index e13382beb53e8..2541d2de4e45f 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_kms.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_kms.c
@@ -659,14 +659,6 @@ static const struct drm_crtc_funcs lcdif_crtc_funcs = {
 	.disable_vblank = lcdif_crtc_disable_vblank,
 };
 
-/* -----------------------------------------------------------------------------
- * Encoder
- */
-
-static const struct drm_encoder_funcs lcdif_encoder_funcs = {
-	.destroy = drm_encoder_cleanup,
-};
-
 /* -----------------------------------------------------------------------------
  * Planes
  */
@@ -759,7 +751,6 @@ int lcdif_kms_init(struct lcdif_drm_private *lcdif)
 					BIT(DRM_COLOR_YCBCR_BT2020);
 	const u32 supported_ranges = BIT(DRM_COLOR_YCBCR_LIMITED_RANGE) |
 				     BIT(DRM_COLOR_YCBCR_FULL_RANGE);
-	struct drm_encoder *encoder = &lcdif->encoder;
 	struct drm_crtc *crtc = &lcdif->crtc;
 	int ret;
 
@@ -783,13 +774,7 @@ int lcdif_kms_init(struct lcdif_drm_private *lcdif)
 		return ret;
 
 	drm_crtc_helper_add(crtc, &lcdif_crtc_helper_funcs);
-	ret = drm_crtc_init_with_planes(lcdif->drm, crtc,
-					&lcdif->planes.primary, NULL,
-					&lcdif_crtc_funcs, NULL);
-	if (ret)
-		return ret;
-
-	encoder->possible_crtcs = drm_crtc_mask(crtc);
-	return drm_encoder_init(lcdif->drm, encoder, &lcdif_encoder_funcs,
-				DRM_MODE_ENCODER_NONE, NULL);
+	return drm_crtc_init_with_planes(lcdif->drm, crtc,
+					 &lcdif->planes.primary, NULL,
+					 &lcdif_crtc_funcs, NULL);
 }
-- 
GitLab


From 67d0109fb7aef2a378e885a4127f411a04042160 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 10 May 2023 17:24:50 +0800
Subject: [PATCH 0681/1791] drm: lcdif: Add i.MX93 LCDIF compatible string

With all previous preparations done to make it possible for the
single LCDIF embedded in i.MX93 SoC to drive multiple displays
simultaneously, add i.MX93 LCDIF compatible string as the last
step of adding i.MX93 LCDIF support.

Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20230510092450.4024730-7-victor.liu@nxp.com
---
 drivers/gpu/drm/mxsfb/lcdif_drv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_drv.c b/drivers/gpu/drm/mxsfb/lcdif_drv.c
index dc85498b81cdb..c9d8cbb214079 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_drv.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_drv.c
@@ -251,6 +251,7 @@ static const struct drm_driver lcdif_driver = {
 
 static const struct of_device_id lcdif_dt_ids[] = {
 	{ .compatible = "fsl,imx8mp-lcdif" },
+	{ .compatible = "fsl,imx93-lcdif" },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, lcdif_dt_ids);
-- 
GitLab


From 2da4b728f994a1f9189a8066b0be90b615768764 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 9 May 2023 09:48:17 +0200
Subject: [PATCH 0682/1791] drm: rcar-du: remove R-Car H3 ES1.* workarounds

R-Car H3 ES1.* was only available to an internal development group and
needed a lot of quirks and workarounds. These become a maintenance
burden now, so our development group decided to remove upstream support
for this SoC and prevent booting it. Public users only have ES2 onwards.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
---
 drivers/gpu/drm/rcar-du/rcar_du_crtc.c | 37 ++------------------
 drivers/gpu/drm/rcar-du/rcar_du_drv.c  | 48 --------------------------
 drivers/gpu/drm/rcar-du/rcar_du_drv.h  |  2 --
 drivers/gpu/drm/rcar-du/rcar_du_regs.h |  3 +-
 4 files changed, 4 insertions(+), 86 deletions(-)

diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
index d6d29be6b4f48..7e175dbfd8924 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_crtc.c
@@ -223,20 +223,6 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 		 * DU channels that have a display PLL can't use the internal
 		 * system clock, and have no internal clock divider.
 		 */
-
-		/*
-		 * The H3 ES1.x exhibits dot clock duty cycle stability issues.
-		 * We can work around them by configuring the DPLL to twice the
-		 * desired frequency, coupled with a /2 post-divider. Restrict
-		 * the workaround to H3 ES1.x as ES2.0 and all other SoCs have
-		 * no post-divider when a display PLL is present (as shown by
-		 * the workaround breaking HDMI output on M3-W during testing).
-		 */
-		if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY) {
-			target *= 2;
-			div = 1;
-		}
-
 		extclk = clk_get_rate(rcrtc->extclock);
 		rcar_du_dpll_divider(rcrtc, &dpll, extclk, target);
 
@@ -245,30 +231,13 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 		       | DPLLCR_N(dpll.n) | DPLLCR_M(dpll.m)
 		       | DPLLCR_STBY;
 
-		if (rcrtc->index == 1) {
+		if (rcrtc->index == 1)
 			dpllcr |= DPLLCR_PLCS1
 			       |  DPLLCR_INCS_DOTCLKIN1;
-		} else {
-			dpllcr |= DPLLCR_PLCS0_PLL
+		else
+			dpllcr |= DPLLCR_PLCS0
 			       |  DPLLCR_INCS_DOTCLKIN0;
 
-			/*
-			 * On ES2.x we have a single mux controlled via bit 21,
-			 * which selects between DCLKIN source (bit 21 = 0) and
-			 * a PLL source (bit 21 = 1), where the PLL is always
-			 * PLL1.
-			 *
-			 * On ES1.x we have an additional mux, controlled
-			 * via bit 20, for choosing between PLL0 (bit 20 = 0)
-			 * and PLL1 (bit 20 = 1). We always want to use PLL1,
-			 * so on ES1.x, in addition to setting bit 21, we need
-			 * to set the bit 20.
-			 */
-
-			if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PLL)
-				dpllcr |= DPLLCR_PLCS0_H3ES1X_PLL1;
-		}
-
 		rcar_du_group_write(rcrtc->group, DPLLCR, dpllcr);
 
 		escr = ESCR_DCLKSEL_DCLKIN | div;
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index b9a94c5260e9d..1ffde19cb87fe 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -16,7 +16,6 @@
 #include <linux/platform_device.h>
 #include <linux/pm.h>
 #include <linux/slab.h>
-#include <linux/sys_soc.h>
 #include <linux/wait.h>
 
 #include <drm/drm_atomic_helper.h>
@@ -387,43 +386,6 @@ static const struct rcar_du_device_info rcar_du_r8a7795_info = {
 	.dpll_mask =  BIT(2) | BIT(1),
 };
 
-static const struct rcar_du_device_info rcar_du_r8a7795_es1_info = {
-	.gen = 3,
-	.features = RCAR_DU_FEATURE_CRTC_IRQ
-		  | RCAR_DU_FEATURE_CRTC_CLOCK
-		  | RCAR_DU_FEATURE_VSP1_SOURCE
-		  | RCAR_DU_FEATURE_INTERLACED
-		  | RCAR_DU_FEATURE_TVM_SYNC,
-	.quirks = RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY
-		| RCAR_DU_QUIRK_H3_ES1_PLL,
-	.channels_mask = BIT(3) | BIT(2) | BIT(1) | BIT(0),
-	.routes = {
-		/*
-		 * R8A7795 has one RGB output, two HDMI outputs and one
-		 * LVDS output.
-		 */
-		[RCAR_DU_OUTPUT_DPAD0] = {
-			.possible_crtcs = BIT(3),
-			.port = 0,
-		},
-		[RCAR_DU_OUTPUT_HDMI0] = {
-			.possible_crtcs = BIT(1),
-			.port = 1,
-		},
-		[RCAR_DU_OUTPUT_HDMI1] = {
-			.possible_crtcs = BIT(2),
-			.port = 2,
-		},
-		[RCAR_DU_OUTPUT_LVDS0] = {
-			.possible_crtcs = BIT(0),
-			.port = 3,
-		},
-	},
-	.num_lvds = 1,
-	.num_rpf = 5,
-	.dpll_mask =  BIT(2) | BIT(1),
-};
-
 static const struct rcar_du_device_info rcar_du_r8a7796_info = {
 	.gen = 3,
 	.features = RCAR_DU_FEATURE_CRTC_IRQ
@@ -614,11 +576,6 @@ static const struct of_device_id rcar_du_of_table[] = {
 
 MODULE_DEVICE_TABLE(of, rcar_du_of_table);
 
-static const struct soc_device_attribute rcar_du_soc_table[] = {
-	{ .soc_id = "r8a7795", .revision = "ES1.*", .data = &rcar_du_r8a7795_es1_info },
-	{ /* sentinel */ }
-};
-
 const char *rcar_du_output_name(enum rcar_du_output output)
 {
 	static const char * const names[] = {
@@ -707,7 +664,6 @@ static void rcar_du_shutdown(struct platform_device *pdev)
 
 static int rcar_du_probe(struct platform_device *pdev)
 {
-	const struct soc_device_attribute *soc_attr;
 	struct rcar_du_device *rcdu;
 	unsigned int mask;
 	int ret;
@@ -725,10 +681,6 @@ static int rcar_du_probe(struct platform_device *pdev)
 
 	rcdu->info = of_device_get_match_data(rcdu->dev);
 
-	soc_attr = soc_device_match(rcar_du_soc_table);
-	if (soc_attr)
-		rcdu->info = soc_attr->data;
-
 	platform_set_drvdata(pdev, rcdu);
 
 	/* I/O resources */
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.h b/drivers/gpu/drm/rcar-du/rcar_du_drv.h
index acc3673fefe18..5cfa2bb7ad93d 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.h
@@ -34,8 +34,6 @@ struct rcar_du_device;
 #define RCAR_DU_FEATURE_NO_BLENDING	BIT(5)	/* PnMR.SPIM does not have ALP nor EOR bits */
 
 #define RCAR_DU_QUIRK_ALIGN_128B	BIT(0)	/* Align pitches to 128 bytes */
-#define RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY BIT(1)	/* H3 ES1 has pclk stability issue */
-#define RCAR_DU_QUIRK_H3_ES1_PLL	BIT(2)	/* H3 ES1 PLL setup differs from non-ES1 */
 
 enum rcar_du_output {
 	RCAR_DU_OUTPUT_DPAD0,
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_regs.h b/drivers/gpu/drm/rcar-du/rcar_du_regs.h
index 6c750fab6ebb7..391de6661d8bc 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_regs.h
+++ b/drivers/gpu/drm/rcar-du/rcar_du_regs.h
@@ -283,8 +283,7 @@
 #define DPLLCR			0x20044
 #define DPLLCR_CODE		(0x95 << 24)
 #define DPLLCR_PLCS1		(1 << 23)
-#define DPLLCR_PLCS0_PLL	(1 << 21)
-#define DPLLCR_PLCS0_H3ES1X_PLL1	(1 << 20)
+#define DPLLCR_PLCS0		(1 << 21)
 #define DPLLCR_CLKE		(1 << 18)
 #define DPLLCR_FDPLL(n)		((n) << 12)
 #define DPLLCR_N(n)		((n) << 5)
-- 
GitLab


From 11696c5e89245a1d360f75be3dfc4960b25a265a Mon Sep 17 00:00:00 2001
From: Biju Das <biju.das.jz@bp.renesas.com>
Date: Tue, 2 May 2023 11:09:08 +0100
Subject: [PATCH 0683/1791] drm: Place Renesas drivers in a separate dir

Create vendor specific renesas directory and move renesas drivers
to that directory.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Acked-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
---
 MAINTAINERS                                                  | 3 +--
 drivers/gpu/drm/Kconfig                                      | 4 +---
 drivers/gpu/drm/Makefile                                     | 3 +--
 drivers/gpu/drm/renesas/Kconfig                              | 4 ++++
 drivers/gpu/drm/renesas/Makefile                             | 4 ++++
 drivers/gpu/drm/{ => renesas}/rcar-du/Kconfig                | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/Makefile               | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_cmm.c             | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_cmm.h             | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_crtc.c         | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_crtc.h         | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_drv.c          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_drv.h          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_encoder.c      | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_encoder.h      | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_group.c        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_group.h        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_kms.c          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_kms.h          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_plane.c        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_plane.h        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_regs.h         | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_vsp.c          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_vsp.h          | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_writeback.c    | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_writeback.h    | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_dw_hdmi.c         | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds.c            | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds.h            | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds_regs.h       | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi.c        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi.h        | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi_regs.h   | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rzg2l_mipi_dsi.c       | 0
 drivers/gpu/drm/{ => renesas}/rcar-du/rzg2l_mipi_dsi_regs.h  | 0
 drivers/gpu/drm/{ => renesas}/shmobile/Kconfig               | 0
 drivers/gpu/drm/{ => renesas}/shmobile/Makefile              | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_backlight.c | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_backlight.h | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_crtc.c      | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_crtc.h      | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_drv.c       | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_drv.h       | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_kms.c       | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_kms.h       | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_plane.c     | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_plane.h     | 0
 drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_regs.h      | 0
 48 files changed, 11 insertions(+), 7 deletions(-)
 create mode 100644 drivers/gpu/drm/renesas/Kconfig
 create mode 100644 drivers/gpu/drm/renesas/Makefile
 rename drivers/gpu/drm/{ => renesas}/rcar-du/Kconfig (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/Makefile (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_cmm.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_cmm.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_crtc.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_crtc.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_drv.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_drv.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_encoder.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_encoder.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_group.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_group.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_kms.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_kms.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_plane.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_plane.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_regs.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_vsp.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_vsp.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_writeback.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_du_writeback.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_dw_hdmi.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_lvds_regs.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rcar_mipi_dsi_regs.h (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rzg2l_mipi_dsi.c (100%)
 rename drivers/gpu/drm/{ => renesas}/rcar-du/rzg2l_mipi_dsi_regs.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/Kconfig (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/Makefile (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_backlight.c (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_backlight.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_crtc.c (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_crtc.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_drv.c (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_drv.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_kms.c (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_kms.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_plane.c (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_plane.h (100%)
 rename drivers/gpu/drm/{ => renesas}/shmobile/shmob_drm_regs.h (100%)

diff --git a/MAINTAINERS b/MAINTAINERS
index b344e1318ac3b..38dc099ad8ff4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6981,8 +6981,7 @@ F:	Documentation/devicetree/bindings/display/bridge/renesas,dsi-csi2-tx.yaml
 F:	Documentation/devicetree/bindings/display/bridge/renesas,dw-hdmi.yaml
 F:	Documentation/devicetree/bindings/display/bridge/renesas,lvds.yaml
 F:	Documentation/devicetree/bindings/display/renesas,du.yaml
-F:	drivers/gpu/drm/rcar-du/
-F:	drivers/gpu/drm/shmobile/
+F:	drivers/gpu/drm/renesas/
 F:	include/linux/platform_data/shmob_drm.h
 
 DRM DRIVERS FOR ROCKCHIP
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index ba3fb04bb6913..41aa8b07252b5 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -295,9 +295,7 @@ source "drivers/gpu/drm/armada/Kconfig"
 
 source "drivers/gpu/drm/atmel-hlcdc/Kconfig"
 
-source "drivers/gpu/drm/rcar-du/Kconfig"
-
-source "drivers/gpu/drm/shmobile/Kconfig"
+source "drivers/gpu/drm/renesas/Kconfig"
 
 source "drivers/gpu/drm/sun4i/Kconfig"
 
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index a33257d2bc7fd..982d9e06168a4 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -156,8 +156,7 @@ obj-$(CONFIG_DRM_UDL) += udl/
 obj-$(CONFIG_DRM_AST) += ast/
 obj-$(CONFIG_DRM_ARMADA) += armada/
 obj-$(CONFIG_DRM_ATMEL_HLCDC)	+= atmel-hlcdc/
-obj-y			+= rcar-du/
-obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/
+obj-y			+= renesas/
 obj-y			+= omapdrm/
 obj-$(CONFIG_DRM_SUN4I) += sun4i/
 obj-y			+= tilcdc/
diff --git a/drivers/gpu/drm/renesas/Kconfig b/drivers/gpu/drm/renesas/Kconfig
new file mode 100644
index 0000000000000..3777dad17f81d
--- /dev/null
+++ b/drivers/gpu/drm/renesas/Kconfig
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+source "drivers/gpu/drm/renesas/rcar-du/Kconfig"
+source "drivers/gpu/drm/renesas/shmobile/Kconfig"
diff --git a/drivers/gpu/drm/renesas/Makefile b/drivers/gpu/drm/renesas/Makefile
new file mode 100644
index 0000000000000..ec0e89e7a5926
--- /dev/null
+++ b/drivers/gpu/drm/renesas/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-y += rcar-du/
+obj-$(CONFIG_DRM_SHMOBILE) += shmobile/
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/renesas/rcar-du/Kconfig
similarity index 100%
rename from drivers/gpu/drm/rcar-du/Kconfig
rename to drivers/gpu/drm/renesas/rcar-du/Kconfig
diff --git a/drivers/gpu/drm/rcar-du/Makefile b/drivers/gpu/drm/renesas/rcar-du/Makefile
similarity index 100%
rename from drivers/gpu/drm/rcar-du/Makefile
rename to drivers/gpu/drm/renesas/rcar-du/Makefile
diff --git a/drivers/gpu/drm/rcar-du/rcar_cmm.c b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_cmm.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_cmm.h b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_cmm.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_cmm.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_crtc.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_crtc.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_drv.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_drv.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_encoder.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_encoder.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_group.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_group.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_group.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_group.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_group.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_group.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_kms.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_kms.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_plane.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_plane.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_regs.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_regs.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_regs.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_vsp.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_vsp.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_writeback.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_du_writeback.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_dw_hdmi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_dw_hdmi.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_dw_hdmi.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_dw_hdmi.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_lvds.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.h b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_lvds.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds_regs.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_lvds_regs.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds_regs.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi.c
rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.h
diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi_regs.h
rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
diff --git a/drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi.c
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi.c
rename to drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi.c
diff --git a/drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi_regs.h
similarity index 100%
rename from drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi_regs.h
rename to drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi_regs.h
diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/renesas/shmobile/Kconfig
similarity index 100%
rename from drivers/gpu/drm/shmobile/Kconfig
rename to drivers/gpu/drm/renesas/shmobile/Kconfig
diff --git a/drivers/gpu/drm/shmobile/Makefile b/drivers/gpu/drm/renesas/shmobile/Makefile
similarity index 100%
rename from drivers/gpu/drm/shmobile/Makefile
rename to drivers/gpu/drm/renesas/shmobile/Makefile
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.c
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_backlight.c
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.c
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_backlight.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.h
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.c
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_crtc.c
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.c
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_crtc.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.h
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.c
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_drv.c
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.c
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_drv.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.h
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.c
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_kms.c
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.c
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_kms.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.h
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.c
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_plane.c
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.c
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_plane.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.h
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_regs.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_regs.h
similarity index 100%
rename from drivers/gpu/drm/shmobile/shmob_drm_regs.h
rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_regs.h
-- 
GitLab


From 5769b45abedca0b9d14145e156c180af36139416 Mon Sep 17 00:00:00 2001
From: Adam Ford <aford173@gmail.com>
Date: Sun, 28 May 2023 08:27:27 -0500
Subject: [PATCH 0684/1791] dt-bindings: bridge: samsung-dsim: Make some flags
 optional

In the event a device is connected to the samsung-dsim
controller that doesn't support the burst-clock, the
driver is able to get the requested pixel clock from the
attached device or bridge.  In these instances, the
samsung,burst-clock-frequency isn't needed, so remove
it from the required list.

The pll-clock frequency can be set by the device tree entry
for samsung,pll-clock-frequency, but in some cases, the
pll-clock may have the same clock rate as sclk_mipi clock.
If they are equal, this flag is not needed since the driver
will use the sclk_mipi rate as a fallback.

Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230528132727.3933-1-aford173@gmail.com
---
 .../bindings/display/bridge/samsung,mipi-dsim.yaml       | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index 9f61ebdfefa85..06b6c44d4641c 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -70,7 +70,9 @@ properties:
   samsung,burst-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM high speed burst mode frequency.
+      DSIM high speed burst mode frequency.  If absent,
+      the pixel clock from the attached device or bridge
+      will be used instead.
 
   samsung,esc-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
@@ -80,7 +82,8 @@ properties:
   samsung,pll-clock-frequency:
     $ref: /schemas/types.yaml#/definitions/uint32
     description:
-      DSIM oscillator clock frequency.
+      DSIM oscillator clock frequency. If absent, the clock frequency
+      of sclk_mipi will be used instead.
 
   phys:
     maxItems: 1
@@ -134,9 +137,7 @@ required:
   - compatible
   - interrupts
   - reg
-  - samsung,burst-clock-frequency
   - samsung,esc-clock-frequency
-  - samsung,pll-clock-frequency
 
 allOf:
   - $ref: ../dsi-controller.yaml#
-- 
GitLab


From 5945d8b9a8e2501b0046ec34b12ef9c115bd77e6 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris.p.wilson@intel.com>
Date: Wed, 17 May 2023 19:27:54 +0530
Subject: [PATCH 0685/1791] drm/i915/gem: Use large rings for compute contexts

Allow compute contexts to submit the maximal amount of work without
blocking userspace.

The original size for user LRC ring's (SZ_16K) was chosen to minimise
memory consumption, without being so small as to frequently stall in the
middle of workloads. With the main consumers being GL / media pipelines
of 2 or 3 batches per frame, we want to support ~10 requests in flight
to allow for the application to control throttling without stalling
within a frame.

v2:
  - cover with else part

Signed-off-by: Chris Wilson <chris.p.wilson@intel.com>
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230517135754.1110291-1-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_context.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index 5402a7bbcb1d1..9a9ff84c90d7e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -964,7 +964,11 @@ static int intel_context_set_gem(struct intel_context *ce,
 	RCU_INIT_POINTER(ce->gem_context, ctx);
 
 	GEM_BUG_ON(intel_context_is_pinned(ce));
-	ce->ring_size = SZ_16K;
+
+	if (ce->engine->class == COMPUTE_CLASS)
+		ce->ring_size = SZ_512K;
+	else
+		ce->ring_size = SZ_16K;
 
 	i915_vm_put(ce->vm);
 	ce->vm = i915_gem_context_get_eb_vm(ctx);
-- 
GitLab


From edfd93e60bf29668a84cbb7ada848060a47cb940 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 26 May 2023 16:55:36 -0700
Subject: [PATCH 0686/1791] drm/i915/guc: Use FAST_REQUEST for non-blocking H2G
 calls

In addition to the already defined REQUEST HXG message format,
which is used when sender expects some confirmation or data,
HXG protocol includes definition of the FAST REQUEST message,
that may be used when sender does not expect any useful data
to be returned.

Using this instead of GUC_HXG_TYPE_EVENT for non-blocking CTB requests
will allow GuC to send back GUC_HXG_TYPE_RESPONSE_FAILURE in case of
errors.

Note that it is not possible to return such errors to the caller,
since this is for non-blocking calls and the related fence is not
stored. Instead such messages are treated as unexpected, which will
give an indication of potential GuC misprogramming that warrants extra
debugging effort.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526235538.2230780-2-John.C.Harrison@Intel.com
---
 .../gpu/drm/i915/gt/uc/abi/guc_messages_abi.h | 30 +++++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c     |  6 ++--
 2 files changed, 33 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h
index 7d5ba4d97d708..98eb4f46572b9 100644
--- a/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h
+++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h
@@ -24,6 +24,7 @@
  *  |   | 30:28 | **TYPE** - message type                                      |
  *  |   |       |   - _`GUC_HXG_TYPE_REQUEST` = 0                              |
  *  |   |       |   - _`GUC_HXG_TYPE_EVENT` = 1                                |
+ *  |   |       |   - _`GUC_HXG_TYPE_FAST_REQUEST` = 2                         |
  *  |   |       |   - _`GUC_HXG_TYPE_NO_RESPONSE_BUSY` = 3                     |
  *  |   |       |   - _`GUC_HXG_TYPE_NO_RESPONSE_RETRY` = 5                    |
  *  |   |       |   - _`GUC_HXG_TYPE_RESPONSE_FAILURE` = 6                     |
@@ -46,6 +47,7 @@
 #define GUC_HXG_MSG_0_TYPE			(0x7 << 28)
 #define   GUC_HXG_TYPE_REQUEST			0u
 #define   GUC_HXG_TYPE_EVENT			1u
+#define   GUC_HXG_TYPE_FAST_REQUEST		2u
 #define   GUC_HXG_TYPE_NO_RESPONSE_BUSY		3u
 #define   GUC_HXG_TYPE_NO_RESPONSE_RETRY	5u
 #define   GUC_HXG_TYPE_RESPONSE_FAILURE		6u
@@ -89,6 +91,34 @@
 #define GUC_HXG_REQUEST_MSG_0_ACTION		(0xffff << 0)
 #define GUC_HXG_REQUEST_MSG_n_DATAn		GUC_HXG_MSG_n_PAYLOAD
 
+/**
+ * DOC: HXG Fast Request
+ *
+ * The `HXG Request`_ message should be used to initiate asynchronous activity
+ * for which confirmation or return data is not expected.
+ *
+ * If confirmation is required then `HXG Request`_ shall be used instead.
+ *
+ * The recipient of this message may only use `HXG Failure`_ message if it was
+ * unable to accept this request (like invalid data).
+ *
+ * Format of `HXG Fast Request`_ message is same as `HXG Request`_ except @TYPE.
+ *
+ *  +---+-------+--------------------------------------------------------------+
+ *  |   | Bits  | Description                                                  |
+ *  +===+=======+==============================================================+
+ *  | 0 |    31 | ORIGIN - see `HXG Message`_                                  |
+ *  |   +-------+--------------------------------------------------------------+
+ *  |   | 30:28 | TYPE = `GUC_HXG_TYPE_FAST_REQUEST`_                          |
+ *  |   +-------+--------------------------------------------------------------+
+ *  |   | 27:16 | DATA0 - see `HXG Request`_                                   |
+ *  |   +-------+--------------------------------------------------------------+
+ *  |   |  15:0 | ACTION - see `HXG Request`_                                  |
+ *  +---+-------+--------------------------------------------------------------+
+ *  |...|       | DATAn - see `HXG Request`_                                   |
+ *  +---+-------+--------------------------------------------------------------+
+ */
+
 /**
  * DOC: HXG Event
  *
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index a22e33f37cae6..af52ed4ffc7fb 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -426,11 +426,11 @@ static int ct_write(struct intel_guc_ct *ct,
 		 FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) |
 		 FIELD_PREP(GUC_CTB_MSG_0_FENCE, fence);
 
-	type = (flags & INTEL_GUC_CT_SEND_NB) ? GUC_HXG_TYPE_EVENT :
+	type = (flags & INTEL_GUC_CT_SEND_NB) ? GUC_HXG_TYPE_FAST_REQUEST :
 		GUC_HXG_TYPE_REQUEST;
 	hxg = FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) |
-		FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
-			   GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
+		FIELD_PREP(GUC_HXG_REQUEST_MSG_0_ACTION |
+			   GUC_HXG_REQUEST_MSG_0_DATA0, action[0]);
 
 	CT_DEBUG(ct, "writing (tail %u) %*ph %*ph %*ph\n",
 		 tail, 4, &header, 4, &hxg, 4 * (len - 1), &action[1]);
-- 
GitLab


From d9911020ca0e1354b4169b5bd8dea9ff123253a4 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 26 May 2023 16:55:37 -0700
Subject: [PATCH 0687/1791] drm/i915/guc: Update log for unsolicited CTB
 response

Instead of printing message fence twice, include HXG header of the
unexpected message and its len.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526235538.2230780-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index af52ed4ffc7fb..3a71bb582089e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -994,9 +994,8 @@ static int ct_handle_response(struct intel_guc_ct *ct, struct ct_incoming_msg *r
 		break;
 	}
 	if (!found) {
-		CT_ERROR(ct, "Unsolicited response (fence %u)\n", fence);
-		CT_ERROR(ct, "Could not find fence=%u, last_fence=%u\n", fence,
-			 ct->requests.last_fence);
+		CT_ERROR(ct, "Unsolicited response message: len %u, data %#x (fence %u, last %u)\n",
+			 len, hxg[0], fence, ct->requests.last_fence);
 		list_for_each_entry(req, &ct->requests.pending, link)
 			CT_ERROR(ct, "request %u awaits response\n",
 				 req->fence);
-- 
GitLab


From a5606b94cd3d843b424a01e061dd0cadbcb0bb8c Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 26 May 2023 16:55:38 -0700
Subject: [PATCH 0688/1791] drm/i915/guc: Track all sent actions to GuC

For easier debug of any unexpected error responses from GuC that
might be related to non-blocking fast requests, track action code (and
stack if under DEBUG_GUC config) for every H2G request.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526235538.2230780-4-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/Kconfig.debug        |  1 +
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 68 ++++++++++++++++++++++-
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h | 11 ++++
 3 files changed, 77 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug
index 47e845353ffad..2d21930d55015 100644
--- a/drivers/gpu/drm/i915/Kconfig.debug
+++ b/drivers/gpu/drm/i915/Kconfig.debug
@@ -157,6 +157,7 @@ config DRM_I915_SW_FENCE_CHECK_DAG
 config DRM_I915_DEBUG_GUC
 	bool "Enable additional driver debugging for GuC"
 	depends on DRM_I915
+	select STACKDEPOT
 	default n
 	help
 	  Choose this option to turn on extra driver debugging that may affect
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 3a71bb582089e..4aa903be1317b 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -376,6 +376,24 @@ void intel_guc_ct_disable(struct intel_guc_ct *ct)
 	}
 }
 
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+static void ct_track_lost_and_found(struct intel_guc_ct *ct, u32 fence, u32 action)
+{
+	unsigned int lost = fence % ARRAY_SIZE(ct->requests.lost_and_found);
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+	unsigned long entries[SZ_32];
+	unsigned int n;
+
+	n = stack_trace_save(entries, ARRAY_SIZE(entries), 1);
+
+	/* May be called under spinlock, so avoid sleeping */
+	ct->requests.lost_and_found[lost].stack = stack_depot_save(entries, n, GFP_NOWAIT);
+#endif
+	ct->requests.lost_and_found[lost].fence = fence;
+	ct->requests.lost_and_found[lost].action = action;
+}
+#endif
+
 static u32 ct_get_next_fence(struct intel_guc_ct *ct)
 {
 	/* For now it's trivial */
@@ -447,6 +465,11 @@ static int ct_write(struct intel_guc_ct *ct,
 	}
 	GEM_BUG_ON(tail > size);
 
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+	ct_track_lost_and_found(ct, fence,
+				FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0]));
+#endif
+
 	/*
 	 * make sure H2G buffer update and LRC tail update (if this triggering a
 	 * submission) are visible before updating the descriptor tail
@@ -953,6 +976,43 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg)
 	return -EPIPE;
 }
 
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+static bool ct_check_lost_and_found(struct intel_guc_ct *ct, u32 fence)
+{
+	unsigned int n;
+	char *buf = NULL;
+	bool found = false;
+
+	lockdep_assert_held(&ct->requests.lock);
+
+	for (n = 0; n < ARRAY_SIZE(ct->requests.lost_and_found); n++) {
+		if (ct->requests.lost_and_found[n].fence != fence)
+			continue;
+		found = true;
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+		buf = kmalloc(SZ_4K, GFP_NOWAIT);
+		if (buf && stack_depot_snprint(ct->requests.lost_and_found[n].stack,
+					       buf, SZ_4K, 0)) {
+			CT_ERROR(ct, "Fence %u was used by action %#04x sent at\n%s",
+				 fence, ct->requests.lost_and_found[n].action, buf);
+			break;
+		}
+#endif
+		CT_ERROR(ct, "Fence %u was used by action %#04x\n",
+			 fence, ct->requests.lost_and_found[n].action);
+		break;
+	}
+	kfree(buf);
+	return found;
+}
+#else
+static bool ct_check_lost_and_found(struct intel_guc_ct *ct, u32 fence)
+{
+	return false;
+}
+#endif
+
 static int ct_handle_response(struct intel_guc_ct *ct, struct ct_incoming_msg *response)
 {
 	u32 len = FIELD_GET(GUC_CTB_MSG_0_NUM_DWORDS, response->msg[0]);
@@ -996,9 +1056,11 @@ static int ct_handle_response(struct intel_guc_ct *ct, struct ct_incoming_msg *r
 	if (!found) {
 		CT_ERROR(ct, "Unsolicited response message: len %u, data %#x (fence %u, last %u)\n",
 			 len, hxg[0], fence, ct->requests.last_fence);
-		list_for_each_entry(req, &ct->requests.pending, link)
-			CT_ERROR(ct, "request %u awaits response\n",
-				 req->fence);
+		if (!ct_check_lost_and_found(ct, fence)) {
+			list_for_each_entry(req, &ct->requests.pending, link)
+				CT_ERROR(ct, "request %u awaits response\n",
+					 req->fence);
+		}
 		err = -ENOKEY;
 	}
 	spin_unlock_irqrestore(&ct->requests.lock, flags);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
index 818415b64f4d1..58e42901ff498 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
@@ -8,6 +8,7 @@
 
 #include <linux/interrupt.h>
 #include <linux/spinlock.h>
+#include <linux/stackdepot.h>
 #include <linux/workqueue.h>
 #include <linux/ktime.h>
 #include <linux/wait.h>
@@ -81,6 +82,16 @@ struct intel_guc_ct {
 
 		struct list_head incoming; /* incoming requests */
 		struct work_struct worker; /* handler for incoming requests */
+
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM)
+		struct {
+			u16 fence;
+			u16 action;
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+			depot_stack_handle_t stack;
+#endif
+		} lost_and_found[SZ_16];
+#endif
 	} requests;
 
 	/** @stall_time: time of first time a CTB submission is stalled */
-- 
GitLab


From 332af828ce78f9c49c65ff35b9fe171060c9d045 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= <u.kleine-koenig@pengutronix.de>
Date: Fri, 26 May 2023 11:07:09 +0200
Subject: [PATCH 0689/1791] drm: Switch i2c drivers back to use .probe()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

After commit b8a1a4cd5a98 ("i2c: Provide a temporary .probe_new()
call-back type"), all drivers being converted to .probe_new() and then
03c835f498b5 ("i2c: Switch .probe() to not take an id parameter") convert
back to (the new) .probe() to be able to eventually drop .probe_new() from
struct i2c_driver.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Martyn Welch <martyn.welch@collabora.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526090709.1517297-1-u.kleine-koenig@pengutronix.de
---
 drivers/gpu/drm/bridge/adv7511/adv7511_drv.c             | 2 +-
 drivers/gpu/drm/bridge/analogix/analogix-anx6345.c       | 2 +-
 drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c       | 2 +-
 drivers/gpu/drm/bridge/analogix/anx7625.c                | 2 +-
 drivers/gpu/drm/bridge/chipone-icn6211.c                 | 2 +-
 drivers/gpu/drm/bridge/chrontel-ch7033.c                 | 2 +-
 drivers/gpu/drm/bridge/cros-ec-anx7688.c                 | 2 +-
 drivers/gpu/drm/bridge/ite-it6505.c                      | 2 +-
 drivers/gpu/drm/bridge/ite-it66121.c                     | 2 +-
 drivers/gpu/drm/bridge/lontium-lt8912b.c                 | 2 +-
 drivers/gpu/drm/bridge/lontium-lt9211.c                  | 2 +-
 drivers/gpu/drm/bridge/lontium-lt9611.c                  | 2 +-
 drivers/gpu/drm/bridge/lontium-lt9611uxc.c               | 2 +-
 drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 4 ++--
 drivers/gpu/drm/bridge/nxp-ptn3460.c                     | 2 +-
 drivers/gpu/drm/bridge/parade-ps8622.c                   | 2 +-
 drivers/gpu/drm/bridge/parade-ps8640.c                   | 2 +-
 drivers/gpu/drm/bridge/sii902x.c                         | 2 +-
 drivers/gpu/drm/bridge/sii9234.c                         | 2 +-
 drivers/gpu/drm/bridge/sil-sii8620.c                     | 2 +-
 drivers/gpu/drm/bridge/tc358767.c                        | 2 +-
 drivers/gpu/drm/bridge/tc358768.c                        | 2 +-
 drivers/gpu/drm/bridge/tc358775.c                        | 2 +-
 drivers/gpu/drm/bridge/ti-dlpc3433.c                     | 2 +-
 drivers/gpu/drm/bridge/ti-sn65dsi83.c                    | 2 +-
 drivers/gpu/drm/bridge/ti-sn65dsi86.c                    | 2 +-
 drivers/gpu/drm/bridge/ti-tfp410.c                       | 2 +-
 drivers/gpu/drm/i2c/tda9950.c                            | 2 +-
 drivers/gpu/drm/i2c/tda998x_drv.c                        | 2 +-
 drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c       | 2 +-
 drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c    | 2 +-
 drivers/gpu/drm/solomon/ssd130x-i2c.c                    | 2 +-
 32 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index ddceafa7b6374..2254457ab5d02 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -1393,7 +1393,7 @@ static struct i2c_driver adv7511_driver = {
 		.of_match_table = adv7511_of_ids,
 	},
 	.id_table = adv7511_i2c_ids,
-	.probe_new = adv7511_probe,
+	.probe = adv7511_probe,
 	.remove = adv7511_remove,
 };
 
diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
index 3577c532abb4e..72ab2ab770819 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
@@ -815,7 +815,7 @@ static struct i2c_driver anx6345_driver = {
 		   .name = "anx6345",
 		   .of_match_table = of_match_ptr(anx6345_match_table),
 		  },
-	.probe_new = anx6345_i2c_probe,
+	.probe = anx6345_i2c_probe,
 	.remove = anx6345_i2c_remove,
 	.id_table = anx6345_id,
 };
diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
index a3a38bbe27860..06a3e3243e19e 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c
@@ -1389,7 +1389,7 @@ static struct i2c_driver anx78xx_driver = {
 		   .name = "anx7814",
 		   .of_match_table = of_match_ptr(anx78xx_match_table),
 		  },
-	.probe_new = anx78xx_i2c_probe,
+	.probe = anx78xx_i2c_probe,
 	.remove = anx78xx_i2c_remove,
 	.id_table = anx78xx_id,
 };
diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c
index c86531f7b0726..8b985efdc086b 100644
--- a/drivers/gpu/drm/bridge/analogix/anx7625.c
+++ b/drivers/gpu/drm/bridge/analogix/anx7625.c
@@ -2800,7 +2800,7 @@ static struct i2c_driver anx7625_driver = {
 		.of_match_table = anx_match_table,
 		.pm = &anx7625_pm_ops,
 	},
-	.probe_new = anx7625_i2c_probe,
+	.probe = anx7625_i2c_probe,
 	.remove = anx7625_i2c_remove,
 
 	.id_table = anx7625_id,
diff --git a/drivers/gpu/drm/bridge/chipone-icn6211.c b/drivers/gpu/drm/bridge/chipone-icn6211.c
index 0e37840cd7a87..8bfce21d6b900 100644
--- a/drivers/gpu/drm/bridge/chipone-icn6211.c
+++ b/drivers/gpu/drm/bridge/chipone-icn6211.c
@@ -795,7 +795,7 @@ static struct i2c_device_id chipone_i2c_id[] = {
 MODULE_DEVICE_TABLE(i2c, chipone_i2c_id);
 
 static struct i2c_driver chipone_i2c_driver = {
-	.probe_new = chipone_i2c_probe,
+	.probe = chipone_i2c_probe,
 	.id_table = chipone_i2c_id,
 	.driver = {
 		.name = "chipone-icn6211-i2c",
diff --git a/drivers/gpu/drm/bridge/chrontel-ch7033.c b/drivers/gpu/drm/bridge/chrontel-ch7033.c
index 339b759e4c813..a854eb84e3991 100644
--- a/drivers/gpu/drm/bridge/chrontel-ch7033.c
+++ b/drivers/gpu/drm/bridge/chrontel-ch7033.c
@@ -603,7 +603,7 @@ static const struct i2c_device_id ch7033_ids[] = {
 MODULE_DEVICE_TABLE(i2c, ch7033_ids);
 
 static struct i2c_driver ch7033_driver = {
-	.probe_new = ch7033_probe,
+	.probe = ch7033_probe,
 	.remove = ch7033_remove,
 	.driver = {
 		.name = "ch7033",
diff --git a/drivers/gpu/drm/bridge/cros-ec-anx7688.c b/drivers/gpu/drm/bridge/cros-ec-anx7688.c
index fa91bdeddef06..c8abd9920fee9 100644
--- a/drivers/gpu/drm/bridge/cros-ec-anx7688.c
+++ b/drivers/gpu/drm/bridge/cros-ec-anx7688.c
@@ -173,7 +173,7 @@ static const struct of_device_id cros_ec_anx7688_bridge_match_table[] = {
 MODULE_DEVICE_TABLE(of, cros_ec_anx7688_bridge_match_table);
 
 static struct i2c_driver cros_ec_anx7688_bridge_driver = {
-	.probe_new = cros_ec_anx7688_bridge_probe,
+	.probe = cros_ec_anx7688_bridge_probe,
 	.remove = cros_ec_anx7688_bridge_remove,
 	.driver = {
 		.name = "cros-ec-anx7688-bridge",
diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c
index 45f579c365e7f..504d51c42f798 100644
--- a/drivers/gpu/drm/bridge/ite-it6505.c
+++ b/drivers/gpu/drm/bridge/ite-it6505.c
@@ -3479,7 +3479,7 @@ static struct i2c_driver it6505_i2c_driver = {
 		.of_match_table = it6505_of_match,
 		.pm = &it6505_bridge_pm_ops,
 	},
-	.probe_new = it6505_i2c_probe,
+	.probe = it6505_i2c_probe,
 	.remove = it6505_i2c_remove,
 	.shutdown = it6505_shutdown,
 	.id_table = it6505_id,
diff --git a/drivers/gpu/drm/bridge/ite-it66121.c b/drivers/gpu/drm/bridge/ite-it66121.c
index a2d723d6a4be8..466641c77fe91 100644
--- a/drivers/gpu/drm/bridge/ite-it66121.c
+++ b/drivers/gpu/drm/bridge/ite-it66121.c
@@ -1640,7 +1640,7 @@ static struct i2c_driver it66121_driver = {
 		.name	= "it66121",
 		.of_match_table = it66121_dt_match,
 	},
-	.probe_new = it66121_probe,
+	.probe = it66121_probe,
 	.remove = it66121_remove,
 	.id_table = it66121_id,
 };
diff --git a/drivers/gpu/drm/bridge/lontium-lt8912b.c b/drivers/gpu/drm/bridge/lontium-lt8912b.c
index 13c131ade2683..4eaea67fb71c2 100644
--- a/drivers/gpu/drm/bridge/lontium-lt8912b.c
+++ b/drivers/gpu/drm/bridge/lontium-lt8912b.c
@@ -773,7 +773,7 @@ static struct i2c_driver lt8912_i2c_driver = {
 		.name = "lt8912",
 		.of_match_table = lt8912_dt_match,
 	},
-	.probe_new = lt8912_probe,
+	.probe = lt8912_probe,
 	.remove = lt8912_remove,
 	.id_table = lt8912_id,
 };
diff --git a/drivers/gpu/drm/bridge/lontium-lt9211.c b/drivers/gpu/drm/bridge/lontium-lt9211.c
index 00db681512385..aa8d47e7f40da 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9211.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9211.c
@@ -787,7 +787,7 @@ static const struct of_device_id lt9211_match_table[] = {
 MODULE_DEVICE_TABLE(of, lt9211_match_table);
 
 static struct i2c_driver lt9211_driver = {
-	.probe_new = lt9211_probe,
+	.probe = lt9211_probe,
 	.remove = lt9211_remove,
 	.id_table = lt9211_id,
 	.driver = {
diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c
index 151efe92711c4..5163e5224aad7 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9611.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9611.c
@@ -1192,7 +1192,7 @@ static struct i2c_driver lt9611_driver = {
 		.name = "lt9611",
 		.of_match_table = lt9611_match_table,
 	},
-	.probe_new = lt9611_probe,
+	.probe = lt9611_probe,
 	.remove = lt9611_remove,
 	.id_table = lt9611_id,
 };
diff --git a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
index 583daacf37054..2a57e804ea020 100644
--- a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
+++ b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c
@@ -1011,7 +1011,7 @@ static struct i2c_driver lt9611uxc_driver = {
 		.of_match_table = lt9611uxc_match_table,
 		.dev_groups = lt9611uxc_attr_groups,
 	},
-	.probe_new = lt9611uxc_probe,
+	.probe = lt9611uxc_probe,
 	.remove = lt9611uxc_remove,
 	.id_table = lt9611uxc_id,
 };
diff --git a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
index 4fc494d9084b7..460db3c8a08cc 100644
--- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
+++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c
@@ -375,7 +375,7 @@ MODULE_DEVICE_TABLE(of, stdp4028_ge_b850v3_fw_match);
 
 static struct i2c_driver stdp4028_ge_b850v3_fw_driver = {
 	.id_table	= stdp4028_ge_b850v3_fw_i2c_table,
-	.probe_new	= stdp4028_ge_b850v3_fw_probe,
+	.probe		= stdp4028_ge_b850v3_fw_probe,
 	.remove		= stdp4028_ge_b850v3_fw_remove,
 	.driver		= {
 		.name		= "stdp4028-ge-b850v3-fw",
@@ -422,7 +422,7 @@ MODULE_DEVICE_TABLE(of, stdp2690_ge_b850v3_fw_match);
 
 static struct i2c_driver stdp2690_ge_b850v3_fw_driver = {
 	.id_table	= stdp2690_ge_b850v3_fw_i2c_table,
-	.probe_new	= stdp2690_ge_b850v3_fw_probe,
+	.probe		= stdp2690_ge_b850v3_fw_probe,
 	.remove		= stdp2690_ge_b850v3_fw_remove,
 	.driver		= {
 		.name		= "stdp2690-ge-b850v3-fw",
diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c
index cd292a2f894ca..d81920227a8ae 100644
--- a/drivers/gpu/drm/bridge/nxp-ptn3460.c
+++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c
@@ -335,7 +335,7 @@ MODULE_DEVICE_TABLE(of, ptn3460_match);
 
 static struct i2c_driver ptn3460_driver = {
 	.id_table	= ptn3460_i2c_table,
-	.probe_new	= ptn3460_probe,
+	.probe		= ptn3460_probe,
 	.remove		= ptn3460_remove,
 	.driver		= {
 		.name	= "nxp,ptn3460",
diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c
index efa80e309b988..c9b6cb7678e33 100644
--- a/drivers/gpu/drm/bridge/parade-ps8622.c
+++ b/drivers/gpu/drm/bridge/parade-ps8622.c
@@ -538,7 +538,7 @@ MODULE_DEVICE_TABLE(i2c, ps8622_i2c_table);
 
 static struct i2c_driver ps8622_driver = {
 	.id_table	= ps8622_i2c_table,
-	.probe_new	= ps8622_probe,
+	.probe		= ps8622_probe,
 	.remove		= ps8622_remove,
 	.driver		= {
 		.name	= "ps8622",
diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c
index c3eb45179405c..8801cdd033b5e 100644
--- a/drivers/gpu/drm/bridge/parade-ps8640.c
+++ b/drivers/gpu/drm/bridge/parade-ps8640.c
@@ -791,7 +791,7 @@ static const struct of_device_id ps8640_match[] = {
 MODULE_DEVICE_TABLE(of, ps8640_match);
 
 static struct i2c_driver ps8640_driver = {
-	.probe_new = ps8640_probe,
+	.probe = ps8640_probe,
 	.remove = ps8640_remove,
 	.driver = {
 		.name = "ps8640",
diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c
index ef66461e7f7c4..aac239729a1d0 100644
--- a/drivers/gpu/drm/bridge/sii902x.c
+++ b/drivers/gpu/drm/bridge/sii902x.c
@@ -1151,7 +1151,7 @@ static const struct i2c_device_id sii902x_i2c_ids[] = {
 MODULE_DEVICE_TABLE(i2c, sii902x_i2c_ids);
 
 static struct i2c_driver sii902x_driver = {
-	.probe_new = sii902x_probe,
+	.probe = sii902x_probe,
 	.remove = sii902x_remove,
 	.driver = {
 		.name = "sii902x",
diff --git a/drivers/gpu/drm/bridge/sii9234.c b/drivers/gpu/drm/bridge/sii9234.c
index 2d17f227867bc..d8373d9183242 100644
--- a/drivers/gpu/drm/bridge/sii9234.c
+++ b/drivers/gpu/drm/bridge/sii9234.c
@@ -955,7 +955,7 @@ static struct i2c_driver sii9234_driver = {
 		.name	= "sii9234",
 		.of_match_table = sii9234_dt_match,
 	},
-	.probe_new = sii9234_probe,
+	.probe = sii9234_probe,
 	.remove = sii9234_remove,
 	.id_table = sii9234_id,
 };
diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c
index b96d03cd878d9..79b09ccd13533 100644
--- a/drivers/gpu/drm/bridge/sil-sii8620.c
+++ b/drivers/gpu/drm/bridge/sil-sii8620.c
@@ -2378,7 +2378,7 @@ static struct i2c_driver sii8620_driver = {
 		.name	= "sii8620",
 		.of_match_table = of_match_ptr(sii8620_dt_match),
 	},
-	.probe_new	= sii8620_probe,
+	.probe		= sii8620_probe,
 	.remove		= sii8620_remove,
 	.id_table = sii8620_id,
 };
diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c
index 9db0e0231e039..65dc842e31f01 100644
--- a/drivers/gpu/drm/bridge/tc358767.c
+++ b/drivers/gpu/drm/bridge/tc358767.c
@@ -2402,7 +2402,7 @@ static struct i2c_driver tc358767_driver = {
 		.of_match_table = tc358767_of_ids,
 	},
 	.id_table = tc358767_i2c_ids,
-	.probe_new = tc_probe,
+	.probe = tc_probe,
 	.remove	= tc_remove,
 };
 module_i2c_driver(tc358767_driver);
diff --git a/drivers/gpu/drm/bridge/tc358768.c b/drivers/gpu/drm/bridge/tc358768.c
index 97ae3a9c90daa..819a4b6ec2a07 100644
--- a/drivers/gpu/drm/bridge/tc358768.c
+++ b/drivers/gpu/drm/bridge/tc358768.c
@@ -1134,7 +1134,7 @@ static struct i2c_driver tc358768_driver = {
 		.of_match_table = tc358768_of_ids,
 	},
 	.id_table = tc358768_i2c_ids,
-	.probe_new = tc358768_i2c_probe,
+	.probe = tc358768_i2c_probe,
 	.remove	= tc358768_i2c_remove,
 };
 module_i2c_driver(tc358768_driver);
diff --git a/drivers/gpu/drm/bridge/tc358775.c b/drivers/gpu/drm/bridge/tc358775.c
index 19316994ddd11..90a89d70d8328 100644
--- a/drivers/gpu/drm/bridge/tc358775.c
+++ b/drivers/gpu/drm/bridge/tc358775.c
@@ -728,7 +728,7 @@ static struct i2c_driver tc358775_driver = {
 		.of_match_table = tc358775_of_ids,
 	},
 	.id_table = tc358775_i2c_ids,
-	.probe_new = tc_probe,
+	.probe = tc_probe,
 	.remove	= tc_remove,
 };
 module_i2c_driver(tc358775_driver);
diff --git a/drivers/gpu/drm/bridge/ti-dlpc3433.c b/drivers/gpu/drm/bridge/ti-dlpc3433.c
index 186a9e2ff24dc..b65632ec7e7da 100644
--- a/drivers/gpu/drm/bridge/ti-dlpc3433.c
+++ b/drivers/gpu/drm/bridge/ti-dlpc3433.c
@@ -400,7 +400,7 @@ static const struct of_device_id dlpc3433_match_table[] = {
 MODULE_DEVICE_TABLE(of, dlpc3433_match_table);
 
 static struct i2c_driver dlpc3433_driver = {
-	.probe_new = dlpc3433_probe,
+	.probe = dlpc3433_probe,
 	.remove = dlpc3433_remove,
 	.id_table = dlpc3433_id,
 	.driver = {
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
index 6e125ba4f0d75..7e9f4ec8e780a 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
@@ -748,7 +748,7 @@ static const struct of_device_id sn65dsi83_match_table[] = {
 MODULE_DEVICE_TABLE(of, sn65dsi83_match_table);
 
 static struct i2c_driver sn65dsi83_driver = {
-	.probe_new = sn65dsi83_probe,
+	.probe = sn65dsi83_probe,
 	.remove = sn65dsi83_remove,
 	.id_table = sn65dsi83_id,
 	.driver = {
diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 260cad1fd1da3..597ceb7024e0e 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -1970,7 +1970,7 @@ static struct i2c_driver ti_sn65dsi86_driver = {
 		.of_match_table = ti_sn65dsi86_match_table,
 		.pm = &ti_sn65dsi86_pm_ops,
 	},
-	.probe_new = ti_sn65dsi86_probe,
+	.probe = ti_sn65dsi86_probe,
 	.id_table = ti_sn65dsi86_id,
 };
 
diff --git a/drivers/gpu/drm/bridge/ti-tfp410.c b/drivers/gpu/drm/bridge/ti-tfp410.c
index ab63225cd6350..c06390da9ffd0 100644
--- a/drivers/gpu/drm/bridge/ti-tfp410.c
+++ b/drivers/gpu/drm/bridge/ti-tfp410.c
@@ -408,7 +408,7 @@ static struct i2c_driver tfp410_i2c_driver = {
 		.of_match_table = of_match_ptr(tfp410_match),
 	},
 	.id_table	= tfp410_i2c_ids,
-	.probe_new	= tfp410_i2c_probe,
+	.probe		= tfp410_i2c_probe,
 	.remove		= tfp410_i2c_remove,
 };
 #endif /* IS_ENABLED(CONFIG_I2C) */
diff --git a/drivers/gpu/drm/i2c/tda9950.c b/drivers/gpu/drm/i2c/tda9950.c
index b8c143e573e0b..82d618c40dcea 100644
--- a/drivers/gpu/drm/i2c/tda9950.c
+++ b/drivers/gpu/drm/i2c/tda9950.c
@@ -492,7 +492,7 @@ static struct i2c_device_id tda9950_ids[] = {
 MODULE_DEVICE_TABLE(i2c, tda9950_ids);
 
 static struct i2c_driver tda9950_driver = {
-	.probe_new = tda9950_probe,
+	.probe = tda9950_probe,
 	.remove = tda9950_remove,
 	.driver = {
 		.name = "tda9950",
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index db5c9343a3d20..40bcd9067e69d 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -2099,7 +2099,7 @@ static const struct i2c_device_id tda998x_ids[] = {
 MODULE_DEVICE_TABLE(i2c, tda998x_ids);
 
 static struct i2c_driver tda998x_driver = {
-	.probe_new = tda998x_probe,
+	.probe = tda998x_probe,
 	.remove = tda998x_remove,
 	.driver = {
 		.name = "tda998x",
diff --git a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
index aba556c98300b..4819ada694826 100644
--- a/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
+++ b/drivers/gpu/drm/panel/panel-olimex-lcd-olinuxino.c
@@ -308,7 +308,7 @@ static struct i2c_driver lcd_olinuxino_driver = {
 		.name = "lcd_olinuxino",
 		.of_match_table = lcd_olinuxino_of_ids,
 	},
-	.probe_new = lcd_olinuxino_probe,
+	.probe = lcd_olinuxino_probe,
 	.remove = lcd_olinuxino_remove,
 };
 
diff --git a/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c b/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
index 11d6ca276c1e7..90ea91e4311d4 100644
--- a/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
+++ b/drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c
@@ -489,7 +489,7 @@ static struct i2c_driver rpi_touchscreen_driver = {
 		.name = "rpi_touchscreen",
 		.of_match_table = rpi_touchscreen_of_ids,
 	},
-	.probe_new = rpi_touchscreen_probe,
+	.probe = rpi_touchscreen_probe,
 	.remove = rpi_touchscreen_remove,
 };
 
diff --git a/drivers/gpu/drm/solomon/ssd130x-i2c.c b/drivers/gpu/drm/solomon/ssd130x-i2c.c
index ddfa0bb5d9c97..b4eb2d64bf6e0 100644
--- a/drivers/gpu/drm/solomon/ssd130x-i2c.c
+++ b/drivers/gpu/drm/solomon/ssd130x-i2c.c
@@ -100,7 +100,7 @@ static struct i2c_driver ssd130x_i2c_driver = {
 		.name = DRIVER_NAME,
 		.of_match_table = ssd130x_of_match,
 	},
-	.probe_new = ssd130x_i2c_probe,
+	.probe = ssd130x_i2c_probe,
 	.remove = ssd130x_i2c_remove,
 	.shutdown = ssd130x_i2c_shutdown,
 };
-- 
GitLab


From 72a597aed1d9eb9ce21f68e4284ef3598d67cda9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= <u.kleine-koenig@pengutronix.de>
Date: Tue, 30 May 2023 09:42:16 +0200
Subject: [PATCH 0690/1791] drm/panel-edp: Convert to platform remove callback
 returning void
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code.  However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.

panel_edp_remove() always returned zero, so convert it to return void
without any loss and then just drop the return from
panel_edp_platform_remove().

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530074216.2195962-1-u.kleine-koenig@pengutronix.de
---
 drivers/gpu/drm/panel/panel-edp.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index e23ddab2126eb..fbd114b4f0be0 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -903,7 +903,7 @@ static int panel_edp_probe(struct device *dev, const struct panel_desc *desc,
 	return err;
 }
 
-static int panel_edp_remove(struct device *dev)
+static void panel_edp_remove(struct device *dev)
 {
 	struct panel_edp *panel = dev_get_drvdata(dev);
 
@@ -918,8 +918,6 @@ static int panel_edp_remove(struct device *dev)
 
 	kfree(panel->edid);
 	panel->edid = NULL;
-
-	return 0;
 }
 
 static void panel_edp_shutdown(struct device *dev)
@@ -1934,9 +1932,9 @@ static int panel_edp_platform_probe(struct platform_device *pdev)
 	return panel_edp_probe(&pdev->dev, id->data, NULL);
 }
 
-static int panel_edp_platform_remove(struct platform_device *pdev)
+static void panel_edp_platform_remove(struct platform_device *pdev)
 {
-	return panel_edp_remove(&pdev->dev);
+	panel_edp_remove(&pdev->dev);
 }
 
 static void panel_edp_platform_shutdown(struct platform_device *pdev)
@@ -1957,7 +1955,7 @@ static struct platform_driver panel_edp_platform_driver = {
 		.pm = &panel_edp_pm_ops,
 	},
 	.probe = panel_edp_platform_probe,
-	.remove = panel_edp_platform_remove,
+	.remove_new = panel_edp_platform_remove,
 	.shutdown = panel_edp_platform_shutdown,
 };
 
-- 
GitLab


From 401e6cd974a708d4c191fda3bdacd6a67c5b30bf Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 29 May 2023 16:37:38 +0530
Subject: [PATCH 0691/1791] drm/i915/hdcp: Rename dev_priv to i915

Rename dev_priv to i915 to keep up with latest code standards.

Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529110740.1522985-2-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 630 +++++++++++-----------
 drivers/gpu/drm/i915/display/intel_hdcp.h |   6 +-
 2 files changed, 318 insertions(+), 318 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 1f96d1fa68e04..e2e5527efc4b2 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -164,7 +164,7 @@ bool intel_hdcp_capable(struct intel_connector *connector)
 bool intel_hdcp2_capable(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	bool capable = false;
 
@@ -173,8 +173,8 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 		return false;
 
 	/* If MTL+ make sure gsc is loaded and proxy is setup */
-	if (intel_hdcp_gsc_cs_required(dev_priv)) {
-		struct intel_gt *gt = dev_priv->media_gt;
+	if (intel_hdcp_gsc_cs_required(i915)) {
+		struct intel_gt *gt = i915->media_gt;
 		struct intel_gsc_uc *gsc = gt ? &gt->uc.gsc : NULL;
 
 		if (!gsc || !intel_uc_fw_is_running(&gsc->fw))
@@ -182,12 +182,12 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 	}
 
 	/* MEI/GSC interface is solid depending on which is used */
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	if (!dev_priv->display.hdcp.comp_added ||  !dev_priv->display.hdcp.master) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	if (!i915->display.hdcp.comp_added ||  !i915->display.hdcp.master) {
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return false;
 	}
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	/* Sink's capability for HDCP2.2 */
 	hdcp->shim->hdcp_2_2_capable(dig_port, &capable);
@@ -195,20 +195,20 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 	return capable;
 }
 
-static bool intel_hdcp_in_use(struct drm_i915_private *dev_priv,
+static bool intel_hdcp_in_use(struct drm_i915_private *i915,
 			      enum transcoder cpu_transcoder, enum port port)
 {
-	return intel_de_read(dev_priv,
-	                     HDCP_STATUS(dev_priv, cpu_transcoder, port)) &
-	       HDCP_STATUS_ENC;
+	return intel_de_read(i915,
+			     HDCP_STATUS(i915, cpu_transcoder, port)) &
+		HDCP_STATUS_ENC;
 }
 
-static bool intel_hdcp2_in_use(struct drm_i915_private *dev_priv,
+static bool intel_hdcp2_in_use(struct drm_i915_private *i915,
 			       enum transcoder cpu_transcoder, enum port port)
 {
-	return intel_de_read(dev_priv,
-	                     HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
-	       LINK_ENCRYPTION_STATUS;
+	return intel_de_read(i915,
+			     HDCP2_STATUS(i915, cpu_transcoder, port)) &
+		LINK_ENCRYPTION_STATUS;
 }
 
 static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
@@ -232,7 +232,7 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
 	return 0;
 }
 
-static bool hdcp_key_loadable(struct drm_i915_private *dev_priv)
+static bool hdcp_key_loadable(struct drm_i915_private *i915)
 {
 	enum i915_power_well_id id;
 	intel_wakeref_t wakeref;
@@ -242,14 +242,14 @@ static bool hdcp_key_loadable(struct drm_i915_private *dev_priv)
 	 * On HSW and BDW, Display HW loads the Key as soon as Display resumes.
 	 * On all BXT+, SW can load the keys only when the PW#1 is turned on.
 	 */
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+	if (IS_HASWELL(i915) || IS_BROADWELL(i915))
 		id = HSW_DISP_PW_GLOBAL;
 	else
 		id = SKL_DISP_PW_1;
 
 	/* PG1 (power well #1) needs to be enabled */
-	with_intel_runtime_pm(&dev_priv->runtime_pm, wakeref)
-		enabled = intel_display_power_well_is_enabled(dev_priv, id);
+	with_intel_runtime_pm(&i915->runtime_pm, wakeref)
+		enabled = intel_display_power_well_is_enabled(i915, id);
 
 	/*
 	 * Another req for hdcp key loadability is enabled state of pll for
@@ -260,19 +260,19 @@ static bool hdcp_key_loadable(struct drm_i915_private *dev_priv)
 	return enabled;
 }
 
-static void intel_hdcp_clear_keys(struct drm_i915_private *dev_priv)
+static void intel_hdcp_clear_keys(struct drm_i915_private *i915)
 {
-	intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_CLEAR_KEYS_TRIGGER);
-	intel_de_write(dev_priv, HDCP_KEY_STATUS,
+	intel_de_write(i915, HDCP_KEY_CONF, HDCP_CLEAR_KEYS_TRIGGER);
+	intel_de_write(i915, HDCP_KEY_STATUS,
 		       HDCP_KEY_LOAD_DONE | HDCP_KEY_LOAD_STATUS | HDCP_FUSE_IN_PROGRESS | HDCP_FUSE_ERROR | HDCP_FUSE_DONE);
 }
 
-static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
+static int intel_hdcp_load_keys(struct drm_i915_private *i915)
 {
 	int ret;
 	u32 val;
 
-	val = intel_de_read(dev_priv, HDCP_KEY_STATUS);
+	val = intel_de_read(i915, HDCP_KEY_STATUS);
 	if ((val & HDCP_KEY_LOAD_DONE) && (val & HDCP_KEY_LOAD_STATUS))
 		return 0;
 
@@ -280,8 +280,8 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
 	 * On HSW and BDW HW loads the HDCP1.4 Key when Display comes
 	 * out of reset. So if Key is not already loaded, its an error state.
 	 */
-	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
-		if (!(intel_de_read(dev_priv, HDCP_KEY_STATUS) & HDCP_KEY_LOAD_DONE))
+	if (IS_HASWELL(i915) || IS_BROADWELL(i915))
+		if (!(intel_de_read(i915, HDCP_KEY_STATUS) & HDCP_KEY_LOAD_DONE))
 			return -ENXIO;
 
 	/*
@@ -292,20 +292,20 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
 	 * process from other platforms. These platforms use the GT Driver
 	 * Mailbox interface.
 	 */
-	if (DISPLAY_VER(dev_priv) == 9 && !IS_BROXTON(dev_priv)) {
-		ret = snb_pcode_write(&dev_priv->uncore, SKL_PCODE_LOAD_HDCP_KEYS, 1);
+	if (DISPLAY_VER(i915) == 9 && !IS_BROXTON(i915)) {
+		ret = snb_pcode_write(&i915->uncore, SKL_PCODE_LOAD_HDCP_KEYS, 1);
 		if (ret) {
-			drm_err(&dev_priv->drm,
+			drm_err(&i915->drm,
 				"Failed to initiate HDCP key load (%d)\n",
 				ret);
 			return ret;
 		}
 	} else {
-		intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_KEY_LOAD_TRIGGER);
+		intel_de_write(i915, HDCP_KEY_CONF, HDCP_KEY_LOAD_TRIGGER);
 	}
 
 	/* Wait for the keys to load (500us) */
-	ret = __intel_wait_for_register(&dev_priv->uncore, HDCP_KEY_STATUS,
+	ret = __intel_wait_for_register(&i915->uncore, HDCP_KEY_STATUS,
 					HDCP_KEY_LOAD_DONE, HDCP_KEY_LOAD_DONE,
 					10, 1, &val);
 	if (ret)
@@ -314,27 +314,27 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv)
 		return -ENXIO;
 
 	/* Send Aksv over to PCH display for use in authentication */
-	intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_AKSV_SEND_TRIGGER);
+	intel_de_write(i915, HDCP_KEY_CONF, HDCP_AKSV_SEND_TRIGGER);
 
 	return 0;
 }
 
 /* Returns updated SHA-1 index */
-static int intel_write_sha_text(struct drm_i915_private *dev_priv, u32 sha_text)
+static int intel_write_sha_text(struct drm_i915_private *i915, u32 sha_text)
 {
-	intel_de_write(dev_priv, HDCP_SHA_TEXT, sha_text);
-	if (intel_de_wait_for_set(dev_priv, HDCP_REP_CTL, HDCP_SHA1_READY, 1)) {
-		drm_err(&dev_priv->drm, "Timed out waiting for SHA1 ready\n");
+	intel_de_write(i915, HDCP_SHA_TEXT, sha_text);
+	if (intel_de_wait_for_set(i915, HDCP_REP_CTL, HDCP_SHA1_READY, 1)) {
+		drm_err(&i915->drm, "Timed out waiting for SHA1 ready\n");
 		return -ETIMEDOUT;
 	}
 	return 0;
 }
 
 static
-u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv,
+u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *i915,
 				enum transcoder cpu_transcoder, enum port port)
 {
-	if (DISPLAY_VER(dev_priv) >= 12) {
+	if (DISPLAY_VER(i915) >= 12) {
 		switch (cpu_transcoder) {
 		case TRANSCODER_A:
 			return HDCP_TRANSA_REP_PRESENT |
@@ -349,7 +349,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv,
 			return HDCP_TRANSD_REP_PRESENT |
 			       HDCP_TRANSD_SHA1_M0;
 		default:
-			drm_err(&dev_priv->drm, "Unknown transcoder %d\n",
+			drm_err(&i915->drm, "Unknown transcoder %d\n",
 				cpu_transcoder);
 			return -EINVAL;
 		}
@@ -367,7 +367,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv,
 	case PORT_E:
 		return HDCP_DDIE_REP_PRESENT | HDCP_DDIE_SHA1_M0;
 	default:
-		drm_err(&dev_priv->drm, "Unknown port %d\n", port);
+		drm_err(&i915->drm, "Unknown port %d\n", port);
 		return -EINVAL;
 	}
 }
@@ -378,7 +378,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 				u8 *ksv_fifo, u8 num_downstream, u8 *bstatus)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder;
 	enum port port = dig_port->base.port;
 	u32 vprime, sha_text, sha_leftovers, rep_ctl;
@@ -389,7 +389,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 		ret = shim->read_v_prime_part(dig_port, i, &vprime);
 		if (ret)
 			return ret;
-		intel_de_write(dev_priv, HDCP_SHA_V_PRIME(i), vprime);
+		intel_de_write(i915, HDCP_SHA_V_PRIME(i), vprime);
 	}
 
 	/*
@@ -405,8 +405,8 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 	sha_idx = 0;
 	sha_text = 0;
 	sha_leftovers = 0;
-	rep_ctl = intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, port);
-	intel_de_write(dev_priv, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32);
+	rep_ctl = intel_hdcp_get_repeater_ctl(i915, cpu_transcoder, port);
+	intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32);
 	for (i = 0; i < num_downstream; i++) {
 		unsigned int sha_empty;
 		u8 *ksv = &ksv_fifo[i * DRM_HDCP_KSV_LEN];
@@ -418,14 +418,14 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 			sha_text |= ksv[j] << off;
 		}
 
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 
 		/* Programming guide writes this every 64 bytes */
 		sha_idx += sizeof(sha_text);
 		if (!(sha_idx % 64))
-			intel_de_write(dev_priv, HDCP_REP_CTL,
+			intel_de_write(i915, HDCP_REP_CTL,
 				       rep_ctl | HDCP_SHA1_TEXT_32);
 
 		/* Store the leftover bytes from the ksv in sha_text */
@@ -442,7 +442,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 		if (sizeof(sha_text) > sha_leftovers)
 			continue;
 
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 		sha_leftovers = 0;
@@ -458,73 +458,73 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 	 */
 	if (sha_leftovers == 0) {
 		/* Write 16 bits of text, 16 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_16);
-		ret = intel_write_sha_text(dev_priv,
+		ret = intel_write_sha_text(i915,
 					   bstatus[0] << 8 | bstatus[1]);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 32 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_0);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 16 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_16);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 	} else if (sha_leftovers == 1) {
 		/* Write 24 bits of text, 8 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_24);
 		sha_text |= bstatus[0] << 16 | bstatus[1] << 8;
 		/* Only 24-bits of data, must be in the LSB */
 		sha_text = (sha_text & 0xffffff00) >> 8;
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 32 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_0);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 24 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_8);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 	} else if (sha_leftovers == 2) {
 		/* Write 32 bits of text */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_32);
 		sha_text |= bstatus[0] << 8 | bstatus[1];
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 64 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_0);
 		for (i = 0; i < 2; i++) {
-			ret = intel_write_sha_text(dev_priv, 0);
+			ret = intel_write_sha_text(i915, 0);
 			if (ret < 0)
 				return ret;
 			sha_idx += sizeof(sha_text);
@@ -534,56 +534,56 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 		 * Terminate the SHA-1 stream by hand. For the other leftover
 		 * cases this is appended by the hardware.
 		 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_32);
 		sha_text = DRM_HDCP_SHA1_TERMINATOR << 24;
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 	} else if (sha_leftovers == 3) {
 		/* Write 32 bits of text (filled from LSB) */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_32);
 		sha_text |= bstatus[0];
-		ret = intel_write_sha_text(dev_priv, sha_text);
+		ret = intel_write_sha_text(i915, sha_text);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 8 bits of text (filled from LSB), 24 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_8);
-		ret = intel_write_sha_text(dev_priv, bstatus[1]);
+		ret = intel_write_sha_text(i915, bstatus[1]);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 32 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_0);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 
 		/* Write 8 bits of M0 */
-		intel_de_write(dev_priv, HDCP_REP_CTL,
+		intel_de_write(i915, HDCP_REP_CTL,
 			       rep_ctl | HDCP_SHA1_TEXT_24);
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
 	} else {
-		drm_dbg_kms(&dev_priv->drm, "Invalid number of leftovers %d\n",
+		drm_dbg_kms(&i915->drm, "Invalid number of leftovers %d\n",
 			    sha_leftovers);
 		return -EINVAL;
 	}
 
-	intel_de_write(dev_priv, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32);
+	intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32);
 	/* Fill up to 64-4 bytes with zeros (leave the last write for length) */
 	while ((sha_idx % 64) < (64 - sizeof(sha_text))) {
-		ret = intel_write_sha_text(dev_priv, 0);
+		ret = intel_write_sha_text(i915, 0);
 		if (ret < 0)
 			return ret;
 		sha_idx += sizeof(sha_text);
@@ -595,20 +595,20 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector,
 	 *  - 10 bytes for BINFO/BSTATUS(2), M0(8)
 	 */
 	sha_text = (num_downstream * 5 + 10) * 8;
-	ret = intel_write_sha_text(dev_priv, sha_text);
+	ret = intel_write_sha_text(i915, sha_text);
 	if (ret < 0)
 		return ret;
 
 	/* Tell the HW we're done with the hash and wait for it to ACK */
-	intel_de_write(dev_priv, HDCP_REP_CTL,
+	intel_de_write(i915, HDCP_REP_CTL,
 		       rep_ctl | HDCP_SHA1_COMPLETE_HASH);
-	if (intel_de_wait_for_set(dev_priv, HDCP_REP_CTL,
+	if (intel_de_wait_for_set(i915, HDCP_REP_CTL,
 				  HDCP_SHA1_COMPLETE, 1)) {
-		drm_err(&dev_priv->drm, "Timed out waiting for SHA1 complete\n");
+		drm_err(&i915->drm, "Timed out waiting for SHA1 complete\n");
 		return -ETIMEDOUT;
 	}
-	if (!(intel_de_read(dev_priv, HDCP_REP_CTL) & HDCP_SHA1_V_MATCH)) {
-		drm_dbg_kms(&dev_priv->drm, "SHA-1 mismatch, HDCP failed\n");
+	if (!(intel_de_read(i915, HDCP_REP_CTL) & HDCP_SHA1_V_MATCH)) {
+		drm_dbg_kms(&i915->drm, "SHA-1 mismatch, HDCP failed\n");
 		return -ENXIO;
 	}
 
@@ -620,14 +620,14 @@ static
 int intel_hdcp_auth_downstream(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	const struct intel_hdcp_shim *shim = connector->hdcp.shim;
 	u8 bstatus[2], num_downstream, *ksv_fifo;
 	int ret, i, tries = 3;
 
 	ret = intel_hdcp_poll_ksv_fifo(dig_port, shim);
 	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "KSV list failed to become ready (%d)\n", ret);
 		return ret;
 	}
@@ -638,7 +638,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector)
 
 	if (DRM_HDCP_MAX_DEVICE_EXCEEDED(bstatus[0]) ||
 	    DRM_HDCP_MAX_CASCADE_EXCEEDED(bstatus[1])) {
-		drm_dbg_kms(&dev_priv->drm, "Max Topology Limit Exceeded\n");
+		drm_dbg_kms(&i915->drm, "Max Topology Limit Exceeded\n");
 		return -EPERM;
 	}
 
@@ -651,14 +651,14 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector)
 	 */
 	num_downstream = DRM_HDCP_NUM_DOWNSTREAM(bstatus[0]);
 	if (num_downstream == 0) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "Repeater with zero downstream devices\n");
 		return -EINVAL;
 	}
 
 	ksv_fifo = kcalloc(DRM_HDCP_KSV_LEN, num_downstream, GFP_KERNEL);
 	if (!ksv_fifo) {
-		drm_dbg_kms(&dev_priv->drm, "Out of mem: ksv_fifo\n");
+		drm_dbg_kms(&i915->drm, "Out of mem: ksv_fifo\n");
 		return -ENOMEM;
 	}
 
@@ -666,9 +666,9 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector)
 	if (ret)
 		goto err;
 
-	if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, ksv_fifo,
+	if (drm_hdcp_check_ksvs_revoked(&i915->drm, ksv_fifo,
 					num_downstream) > 0) {
-		drm_err(&dev_priv->drm, "Revoked Ksv(s) in ksv_fifo\n");
+		drm_err(&i915->drm, "Revoked Ksv(s) in ksv_fifo\n");
 		ret = -EPERM;
 		goto err;
 	}
@@ -686,12 +686,12 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector)
 	}
 
 	if (i == tries) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "V Prime validation failed.(%d)\n", ret);
 		goto err;
 	}
 
-	drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (%d downstream devices)\n",
+	drm_dbg_kms(&i915->drm, "HDCP is enabled (%d downstream devices)\n",
 		    num_downstream);
 	ret = 0;
 err:
@@ -703,7 +703,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector)
 static int intel_hdcp_auth(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	const struct intel_hdcp_shim *shim = hdcp->shim;
 	enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder;
@@ -735,7 +735,7 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 		if (ret)
 			return ret;
 		if (!hdcp_capable) {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(&i915->drm,
 				    "Panel is not HDCP capable\n");
 			return -EINVAL;
 		}
@@ -743,24 +743,24 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 
 	/* Initialize An with 2 random values and acquire it */
 	for (i = 0; i < 2; i++)
-		intel_de_write(dev_priv,
-			       HDCP_ANINIT(dev_priv, cpu_transcoder, port),
+		intel_de_write(i915,
+			       HDCP_ANINIT(i915, cpu_transcoder, port),
 			       get_random_u32());
-	intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port),
+	intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port),
 		       HDCP_CONF_CAPTURE_AN);
 
 	/* Wait for An to be acquired */
-	if (intel_de_wait_for_set(dev_priv,
-				  HDCP_STATUS(dev_priv, cpu_transcoder, port),
+	if (intel_de_wait_for_set(i915,
+				  HDCP_STATUS(i915, cpu_transcoder, port),
 				  HDCP_STATUS_AN_READY, 1)) {
-		drm_err(&dev_priv->drm, "Timed out waiting for An\n");
+		drm_err(&i915->drm, "Timed out waiting for An\n");
 		return -ETIMEDOUT;
 	}
 
-	an.reg[0] = intel_de_read(dev_priv,
-				  HDCP_ANLO(dev_priv, cpu_transcoder, port));
-	an.reg[1] = intel_de_read(dev_priv,
-				  HDCP_ANHI(dev_priv, cpu_transcoder, port));
+	an.reg[0] = intel_de_read(i915,
+				  HDCP_ANLO(i915, cpu_transcoder, port));
+	an.reg[1] = intel_de_read(i915,
+				  HDCP_ANHI(i915, cpu_transcoder, port));
 	ret = shim->write_an_aksv(dig_port, an.shim);
 	if (ret)
 		return ret;
@@ -773,34 +773,34 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 	if (ret < 0)
 		return ret;
 
-	if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, bksv.shim, 1) > 0) {
-		drm_err(&dev_priv->drm, "BKSV is revoked\n");
+	if (drm_hdcp_check_ksvs_revoked(&i915->drm, bksv.shim, 1) > 0) {
+		drm_err(&i915->drm, "BKSV is revoked\n");
 		return -EPERM;
 	}
 
-	intel_de_write(dev_priv, HDCP_BKSVLO(dev_priv, cpu_transcoder, port),
+	intel_de_write(i915, HDCP_BKSVLO(i915, cpu_transcoder, port),
 		       bksv.reg[0]);
-	intel_de_write(dev_priv, HDCP_BKSVHI(dev_priv, cpu_transcoder, port),
+	intel_de_write(i915, HDCP_BKSVHI(i915, cpu_transcoder, port),
 		       bksv.reg[1]);
 
 	ret = shim->repeater_present(dig_port, &repeater_present);
 	if (ret)
 		return ret;
 	if (repeater_present)
-		intel_de_write(dev_priv, HDCP_REP_CTL,
-			       intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, port));
+		intel_de_write(i915, HDCP_REP_CTL,
+			       intel_hdcp_get_repeater_ctl(i915, cpu_transcoder, port));
 
 	ret = shim->toggle_signalling(dig_port, cpu_transcoder, true);
 	if (ret)
 		return ret;
 
-	intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port),
+	intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port),
 		       HDCP_CONF_AUTH_AND_ENC);
 
 	/* Wait for R0 ready */
-	if (wait_for(intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)) &
+	if (wait_for(intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) &
 		     (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), 1)) {
-		drm_err(&dev_priv->drm, "Timed out waiting for R0 ready\n");
+		drm_err(&i915->drm, "Timed out waiting for R0 ready\n");
 		return -ETIMEDOUT;
 	}
 
@@ -826,30 +826,30 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 		ret = shim->read_ri_prime(dig_port, ri.shim);
 		if (ret)
 			return ret;
-		intel_de_write(dev_priv,
-			       HDCP_RPRIME(dev_priv, cpu_transcoder, port),
+		intel_de_write(i915,
+			       HDCP_RPRIME(i915, cpu_transcoder, port),
 			       ri.reg);
 
 		/* Wait for Ri prime match */
-		if (!wait_for(intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)) &
+		if (!wait_for(intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) &
 			      (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1))
 			break;
 	}
 
 	if (i == tries) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "Timed out waiting for Ri prime match (%x)\n",
-			    intel_de_read(dev_priv, HDCP_STATUS(dev_priv,
-					  cpu_transcoder, port)));
+			    intel_de_read(i915,
+					  HDCP_STATUS(i915, cpu_transcoder, port)));
 		return -ETIMEDOUT;
 	}
 
 	/* Wait for encryption confirmation */
-	if (intel_de_wait_for_set(dev_priv,
-				  HDCP_STATUS(dev_priv, cpu_transcoder, port),
+	if (intel_de_wait_for_set(i915,
+				  HDCP_STATUS(i915, cpu_transcoder, port),
 				  HDCP_STATUS_ENC,
 				  HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) {
-		drm_err(&dev_priv->drm, "Timed out waiting for encryption\n");
+		drm_err(&i915->drm, "Timed out waiting for encryption\n");
 		return -ETIMEDOUT;
 	}
 
@@ -857,42 +857,42 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 	if (shim->stream_encryption) {
 		ret = shim->stream_encryption(connector, true);
 		if (ret) {
-			drm_err(&dev_priv->drm, "[%s:%d] Failed to enable HDCP 1.4 stream enc\n",
+			drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 1.4 stream enc\n",
 				connector->base.name, connector->base.base.id);
 			return ret;
 		}
-		drm_dbg_kms(&dev_priv->drm, "HDCP 1.4 transcoder: %s stream encrypted\n",
+		drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encrypted\n",
 			    transcoder_name(hdcp->stream_transcoder));
 	}
 
 	if (repeater_present)
 		return intel_hdcp_auth_downstream(connector);
 
-	drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (no repeater present)\n");
+	drm_dbg_kms(&i915->drm, "HDCP is enabled (no repeater present)\n");
 	return 0;
 }
 
 static int _intel_hdcp_disable(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder = hdcp->cpu_transcoder;
 	u32 repeater_ctl;
 	int ret;
 
-	drm_dbg_kms(&dev_priv->drm, "[%s:%d] HDCP is being disabled...\n",
+	drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being disabled...\n",
 		    connector->base.name, connector->base.base.id);
 
 	if (hdcp->shim->stream_encryption) {
 		ret = hdcp->shim->stream_encryption(connector, false);
 		if (ret) {
-			drm_err(&dev_priv->drm, "[%s:%d] Failed to disable HDCP 1.4 stream enc\n",
+			drm_err(&i915->drm, "[%s:%d] Failed to disable HDCP 1.4 stream enc\n",
 				connector->base.name, connector->base.base.id);
 			return ret;
 		}
-		drm_dbg_kms(&dev_priv->drm, "HDCP 1.4 transcoder: %s stream encryption disabled\n",
+		drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encryption disabled\n",
 			    transcoder_name(hdcp->stream_transcoder));
 		/*
 		 * If there are other connectors on this port using HDCP,
@@ -904,51 +904,51 @@ static int _intel_hdcp_disable(struct intel_connector *connector)
 	}
 
 	hdcp->hdcp_encrypted = false;
-	intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port), 0);
-	if (intel_de_wait_for_clear(dev_priv,
-				    HDCP_STATUS(dev_priv, cpu_transcoder, port),
+	intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port), 0);
+	if (intel_de_wait_for_clear(i915,
+				    HDCP_STATUS(i915, cpu_transcoder, port),
 				    ~0, HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) {
-		drm_err(&dev_priv->drm,
+		drm_err(&i915->drm,
 			"Failed to disable HDCP, timeout clearing status\n");
 		return -ETIMEDOUT;
 	}
 
-	repeater_ctl = intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder,
+	repeater_ctl = intel_hdcp_get_repeater_ctl(i915, cpu_transcoder,
 						   port);
-	intel_de_rmw(dev_priv, HDCP_REP_CTL, repeater_ctl, 0);
+	intel_de_rmw(i915, HDCP_REP_CTL, repeater_ctl, 0);
 
 	ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder, false);
 	if (ret) {
-		drm_err(&dev_priv->drm, "Failed to disable HDCP signalling\n");
+		drm_err(&i915->drm, "Failed to disable HDCP signalling\n");
 		return ret;
 	}
 
-	drm_dbg_kms(&dev_priv->drm, "HDCP is disabled\n");
+	drm_dbg_kms(&i915->drm, "HDCP is disabled\n");
 	return 0;
 }
 
 static int _intel_hdcp_enable(struct intel_connector *connector)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	int i, ret, tries = 3;
 
-	drm_dbg_kms(&dev_priv->drm, "[%s:%d] HDCP is being enabled...\n",
+	drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being enabled...\n",
 		    connector->base.name, connector->base.base.id);
 
-	if (!hdcp_key_loadable(dev_priv)) {
-		drm_err(&dev_priv->drm, "HDCP key Load is not possible\n");
+	if (!hdcp_key_loadable(i915)) {
+		drm_err(&i915->drm, "HDCP key Load is not possible\n");
 		return -ENXIO;
 	}
 
 	for (i = 0; i < KEY_LOAD_TRIES; i++) {
-		ret = intel_hdcp_load_keys(dev_priv);
+		ret = intel_hdcp_load_keys(i915);
 		if (!ret)
 			break;
-		intel_hdcp_clear_keys(dev_priv);
+		intel_hdcp_clear_keys(i915);
 	}
 	if (ret) {
-		drm_err(&dev_priv->drm, "Could not load HDCP keys, (%d)\n",
+		drm_err(&i915->drm, "Could not load HDCP keys, (%d)\n",
 			ret);
 		return ret;
 	}
@@ -961,13 +961,13 @@ static int _intel_hdcp_enable(struct intel_connector *connector)
 			return 0;
 		}
 
-		drm_dbg_kms(&dev_priv->drm, "HDCP Auth failure (%d)\n", ret);
+		drm_dbg_kms(&i915->drm, "HDCP Auth failure (%d)\n", ret);
 
 		/* Ensuring HDCP encryption and signalling are stopped. */
 		_intel_hdcp_disable(connector);
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(&i915->drm,
 		    "HDCP authentication failed (%d tries/%d)\n", tries, ret);
 	return ret;
 }
@@ -1009,7 +1009,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector,
 static int intel_hdcp_check_link(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder;
@@ -1027,12 +1027,12 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 		goto out;
 	}
 
-	if (drm_WARN_ON(&dev_priv->drm,
-			!intel_hdcp_in_use(dev_priv, cpu_transcoder, port))) {
-		drm_err(&dev_priv->drm,
+	if (drm_WARN_ON(&i915->drm,
+			!intel_hdcp_in_use(i915, cpu_transcoder, port))) {
+		drm_err(&i915->drm,
 			"%s:%d HDCP link stopped encryption,%x\n",
 			connector->base.name, connector->base.base.id,
-			intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)));
+			intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)));
 		ret = -ENXIO;
 		intel_hdcp_update_value(connector,
 					DRM_MODE_CONTENT_PROTECTION_DESIRED,
@@ -1048,13 +1048,13 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 		goto out;
 	}
 
-	drm_dbg_kms(&dev_priv->drm,
+	drm_dbg_kms(&i915->drm,
 		    "[%s:%d] HDCP link failed, retrying authentication\n",
 		    connector->base.name, connector->base.base.id);
 
 	ret = _intel_hdcp_disable(connector);
 	if (ret) {
-		drm_err(&dev_priv->drm, "Failed to disable hdcp (%d)\n", ret);
+		drm_err(&i915->drm, "Failed to disable hdcp (%d)\n", ret);
 		intel_hdcp_update_value(connector,
 					DRM_MODE_CONTENT_PROTECTION_DESIRED,
 					true);
@@ -1063,7 +1063,7 @@ static int intel_hdcp_check_link(struct intel_connector *connector)
 
 	ret = _intel_hdcp_enable(connector);
 	if (ret) {
-		drm_err(&dev_priv->drm, "Failed to enable hdcp (%d)\n", ret);
+		drm_err(&i915->drm, "Failed to enable hdcp (%d)\n", ret);
 		intel_hdcp_update_value(connector,
 					DRM_MODE_CONTENT_PROTECTION_DESIRED,
 					true);
@@ -1081,9 +1081,9 @@ static void intel_hdcp_prop_work(struct work_struct *work)
 	struct intel_hdcp *hdcp = container_of(work, struct intel_hdcp,
 					       prop_work);
 	struct intel_connector *connector = intel_hdcp_to_connector(hdcp);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
-	drm_modeset_lock(&dev_priv->drm.mode_config.connection_mutex, NULL);
+	drm_modeset_lock(&i915->drm.mode_config.connection_mutex, NULL);
 	mutex_lock(&hdcp->mutex);
 
 	/*
@@ -1096,15 +1096,15 @@ static void intel_hdcp_prop_work(struct work_struct *work)
 						   hdcp->value);
 
 	mutex_unlock(&hdcp->mutex);
-	drm_modeset_unlock(&dev_priv->drm.mode_config.connection_mutex);
+	drm_modeset_unlock(&i915->drm.mode_config.connection_mutex);
 
 	drm_connector_put(&connector->base);
 }
 
-bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port)
+bool is_hdcp_supported(struct drm_i915_private *i915, enum port port)
 {
-	return DISPLAY_RUNTIME_INFO(dev_priv)->has_hdcp &&
-		(DISPLAY_VER(dev_priv) >= 12 || port < PORT_E);
+	return DISPLAY_RUNTIME_INFO(i915)->has_hdcp &&
+		(DISPLAY_VER(i915) >= 12 || port < PORT_E);
 }
 
 static int
@@ -1113,23 +1113,23 @@ hdcp2_prepare_ake_init(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->initiate_hdcp2_session(arbiter->hdcp_dev, data, ake_data);
 	if (ret)
-		drm_dbg_kms(&dev_priv->drm, "Prepare_ake_init failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Prepare_ake_init failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1143,15 +1143,15 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
@@ -1159,9 +1159,9 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
 							 rx_cert, paired,
 							 ek_pub_km, msg_sz);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Verify rx_cert failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Verify rx_cert failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1171,22 +1171,22 @@ static int hdcp2_verify_hprime(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->verify_hprime(arbiter->hdcp_dev, data, rx_hprime);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Verify hprime failed. %d\n", ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		drm_dbg_kms(&i915->drm, "Verify hprime failed. %d\n", ret);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1197,23 +1197,23 @@ hdcp2_store_pairing_info(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->store_pairing_info(arbiter->hdcp_dev, data, pairing_info);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Store pairing info failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Store pairing info failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1224,23 +1224,23 @@ hdcp2_prepare_lc_init(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->initiate_locality_check(arbiter->hdcp_dev, data, lc_init);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Prepare lc_init failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Prepare lc_init failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1251,23 +1251,23 @@ hdcp2_verify_lprime(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->verify_lprime(arbiter->hdcp_dev, data, rx_lprime);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Verify L_Prime failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Verify L_Prime failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1277,23 +1277,23 @@ static int hdcp2_prepare_skey(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->get_session_key(arbiter->hdcp_dev, data, ske_data);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Get session key failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Get session key failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1306,15 +1306,15 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
@@ -1323,9 +1323,9 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
 							    rep_topology,
 							    rep_send_ack);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "Verify rep topology failed. %d\n", ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1336,22 +1336,22 @@ hdcp2_verify_mprime(struct intel_connector *connector,
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->verify_mprime(arbiter->hdcp_dev, data, stream_ready);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Verify mprime failed. %d\n", ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		drm_dbg_kms(&i915->drm, "Verify mprime failed. %d\n", ret);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1360,23 +1360,23 @@ static int hdcp2_authenticate_port(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->enable_hdcp_authentication(arbiter->hdcp_dev, data);
 	if (ret < 0)
-		drm_dbg_kms(&dev_priv->drm, "Enable hdcp auth failed. %d\n",
+		drm_dbg_kms(&i915->drm, "Enable hdcp auth failed. %d\n",
 			    ret);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1384,21 +1384,21 @@ static int hdcp2_authenticate_port(struct intel_connector *connector)
 static int hdcp2_close_session(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct i915_hdcp_master *arbiter;
 	int ret;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	arbiter = dev_priv->display.hdcp.master;
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	arbiter = i915->display.hdcp.master;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->close_hdcp_session(arbiter->hdcp_dev,
 					     &dig_port->hdcp_port_data);
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return ret;
 }
@@ -1412,7 +1412,7 @@ static int hdcp2_deauthenticate_port(struct intel_connector *connector)
 static int hdcp2_authentication_key_exchange(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	union {
 		struct hdcp2_ake_init ake_init;
@@ -1444,16 +1444,16 @@ static int hdcp2_authentication_key_exchange(struct intel_connector *connector)
 		return ret;
 
 	if (msgs.send_cert.rx_caps[0] != HDCP_2_2_RX_CAPS_VERSION_VAL) {
-		drm_dbg_kms(&dev_priv->drm, "cert.rx_caps dont claim HDCP2.2\n");
+		drm_dbg_kms(&i915->drm, "cert.rx_caps dont claim HDCP2.2\n");
 		return -EINVAL;
 	}
 
 	hdcp->is_repeater = HDCP_2_2_RX_REPEATER(msgs.send_cert.rx_caps[2]);
 
-	if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm,
+	if (drm_hdcp_check_ksvs_revoked(&i915->drm,
 					msgs.send_cert.cert_rx.receiver_id,
 					1) > 0) {
-		drm_err(&dev_priv->drm, "Receiver ID is revoked\n");
+		drm_err(&i915->drm, "Receiver ID is revoked\n");
 		return -EPERM;
 	}
 
@@ -1607,7 +1607,7 @@ static
 int hdcp2_authenticate_repeater_topology(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	union {
 		struct hdcp2_rep_send_receiverid_list recvid_list;
@@ -1627,7 +1627,7 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector)
 
 	if (HDCP_2_2_MAX_CASCADE_EXCEEDED(rx_info[1]) ||
 	    HDCP_2_2_MAX_DEVS_EXCEEDED(rx_info[1])) {
-		drm_dbg_kms(&dev_priv->drm, "Topology Max Size Exceeded\n");
+		drm_dbg_kms(&i915->drm, "Topology Max Size Exceeded\n");
 		return -EINVAL;
 	}
 
@@ -1644,23 +1644,23 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector)
 		drm_hdcp_be24_to_cpu((const u8 *)msgs.recvid_list.seq_num_v);
 
 	if (!hdcp->hdcp2_encrypted && seq_num_v) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "Non zero Seq_num_v at first RecvId_List msg\n");
 		return -EINVAL;
 	}
 
 	if (seq_num_v < hdcp->seq_num_v) {
 		/* Roll over of the seq_num_v from repeater. Reauthenticate. */
-		drm_dbg_kms(&dev_priv->drm, "Seq_num_v roll over.\n");
+		drm_dbg_kms(&i915->drm, "Seq_num_v roll over.\n");
 		return -EINVAL;
 	}
 
 	device_cnt = (HDCP_2_2_DEV_COUNT_HI(rx_info[0]) << 4 |
 		      HDCP_2_2_DEV_COUNT_LO(rx_info[1]));
-	if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm,
+	if (drm_hdcp_check_ksvs_revoked(&i915->drm,
 					msgs.recvid_list.receiver_ids,
 					device_cnt) > 0) {
-		drm_err(&dev_priv->drm, "Revoked receiver ID(s) is in list\n");
+		drm_err(&i915->drm, "Revoked receiver ID(s) is in list\n");
 		return -EPERM;
 	}
 
@@ -1729,16 +1729,16 @@ static int hdcp2_authenticate_sink(struct intel_connector *connector)
 static int hdcp2_enable_stream_encryption(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum transcoder cpu_transcoder = hdcp->cpu_transcoder;
 	enum port port = dig_port->base.port;
 	int ret = 0;
 
-	if (!(intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
+	if (!(intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) &
 			    LINK_ENCRYPTION_STATUS)) {
-		drm_err(&dev_priv->drm, "[%s:%d] HDCP 2.2 Link is not encrypted\n",
+		drm_err(&i915->drm, "[%s:%d] HDCP 2.2 Link is not encrypted\n",
 			connector->base.name, connector->base.base.id);
 		ret = -EPERM;
 		goto link_recover;
@@ -1747,11 +1747,11 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector)
 	if (hdcp->shim->stream_2_2_encryption) {
 		ret = hdcp->shim->stream_2_2_encryption(connector, true);
 		if (ret) {
-			drm_err(&dev_priv->drm, "[%s:%d] Failed to enable HDCP 2.2 stream enc\n",
+			drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 2.2 stream enc\n",
 				connector->base.name, connector->base.base.id);
 			return ret;
 		}
-		drm_dbg_kms(&dev_priv->drm, "HDCP 2.2 transcoder: %s stream encrypted\n",
+		drm_dbg_kms(&i915->drm, "HDCP 2.2 transcoder: %s stream encrypted\n",
 			    transcoder_name(hdcp->stream_transcoder));
 	}
 
@@ -1759,7 +1759,7 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector)
 
 link_recover:
 	if (hdcp2_deauthenticate_port(connector) < 0)
-		drm_dbg_kms(&dev_priv->drm, "Port deauth failed.\n");
+		drm_dbg_kms(&i915->drm, "Port deauth failed.\n");
 
 	dig_port->hdcp_auth_status = false;
 	data->k = 0;
@@ -1770,34 +1770,34 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector)
 static int hdcp2_enable_encryption(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder = hdcp->cpu_transcoder;
 	int ret;
 
-	drm_WARN_ON(&dev_priv->drm,
-		    intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
+	drm_WARN_ON(&i915->drm,
+		    intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) &
 		    LINK_ENCRYPTION_STATUS);
 	if (hdcp->shim->toggle_signalling) {
 		ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder,
 						    true);
 		if (ret) {
-			drm_err(&dev_priv->drm,
+			drm_err(&i915->drm,
 				"Failed to enable HDCP signalling. %d\n",
 				ret);
 			return ret;
 		}
 	}
 
-	if (intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
+	if (intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) &
 	    LINK_AUTH_STATUS)
 		/* Link is Authenticated. Now set for Encryption */
-		intel_de_rmw(dev_priv, HDCP2_CTL(dev_priv, cpu_transcoder, port),
+		intel_de_rmw(i915, HDCP2_CTL(i915, cpu_transcoder, port),
 			     0, CTL_LINK_ENCRYPTION_REQ);
 
-	ret = intel_de_wait_for_set(dev_priv,
-				    HDCP2_STATUS(dev_priv, cpu_transcoder,
+	ret = intel_de_wait_for_set(i915,
+				    HDCP2_STATUS(i915, cpu_transcoder,
 						 port),
 				    LINK_ENCRYPTION_STATUS,
 				    HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS);
@@ -1809,31 +1809,31 @@ static int hdcp2_enable_encryption(struct intel_connector *connector)
 static int hdcp2_disable_encryption(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder = hdcp->cpu_transcoder;
 	int ret;
 
-	drm_WARN_ON(&dev_priv->drm, !(intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) &
+	drm_WARN_ON(&i915->drm, !(intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) &
 				      LINK_ENCRYPTION_STATUS));
 
-	intel_de_rmw(dev_priv, HDCP2_CTL(dev_priv, cpu_transcoder, port),
+	intel_de_rmw(i915, HDCP2_CTL(i915, cpu_transcoder, port),
 		     CTL_LINK_ENCRYPTION_REQ, 0);
 
-	ret = intel_de_wait_for_clear(dev_priv,
-				      HDCP2_STATUS(dev_priv, cpu_transcoder,
+	ret = intel_de_wait_for_clear(i915,
+				      HDCP2_STATUS(i915, cpu_transcoder,
 						   port),
 				      LINK_ENCRYPTION_STATUS,
 				      HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS);
 	if (ret == -ETIMEDOUT)
-		drm_dbg_kms(&dev_priv->drm, "Disable Encryption Timedout");
+		drm_dbg_kms(&i915->drm, "Disable Encryption Timedout");
 
 	if (hdcp->shim->toggle_signalling) {
 		ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder,
 						    false);
 		if (ret) {
-			drm_err(&dev_priv->drm,
+			drm_err(&i915->drm,
 				"Failed to disable HDCP signalling. %d\n",
 				ret);
 			return ret;
@@ -1993,7 +1993,7 @@ _intel_hdcp2_disable(struct intel_connector *connector, bool hdcp2_link_recovery
 static int intel_hdcp2_check_link(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	enum port port = dig_port->base.port;
 	enum transcoder cpu_transcoder;
@@ -2010,11 +2010,11 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 		goto out;
 	}
 
-	if (drm_WARN_ON(&dev_priv->drm,
-			!intel_hdcp2_in_use(dev_priv, cpu_transcoder, port))) {
-		drm_err(&dev_priv->drm,
+	if (drm_WARN_ON(&i915->drm,
+			!intel_hdcp2_in_use(i915, cpu_transcoder, port))) {
+		drm_err(&i915->drm,
 			"HDCP2.2 link stopped the encryption, %x\n",
-			intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)));
+			intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)));
 		ret = -ENXIO;
 		_intel_hdcp2_disable(connector, true);
 		intel_hdcp_update_value(connector,
@@ -2037,7 +2037,7 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 		if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_UNDESIRED)
 			goto out;
 
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "HDCP2.2 Downstream topology change\n");
 		ret = hdcp2_authenticate_repeater_topology(connector);
 		if (!ret) {
@@ -2046,19 +2046,19 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 					true);
 			goto out;
 		}
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "[%s:%d] Repeater topology auth failed.(%d)\n",
 			    connector->base.name, connector->base.base.id,
 			    ret);
 	} else {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "[%s:%d] HDCP2.2 link failed, retrying auth\n",
 			    connector->base.name, connector->base.base.id);
 	}
 
 	ret = _intel_hdcp2_disable(connector, true);
 	if (ret) {
-		drm_err(&dev_priv->drm,
+		drm_err(&i915->drm,
 			"[%s:%d] Failed to disable hdcp2.2 (%d)\n",
 			connector->base.name, connector->base.base.id, ret);
 		intel_hdcp_update_value(connector,
@@ -2068,7 +2068,7 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
 
 	ret = _intel_hdcp2_enable(connector);
 	if (ret) {
-		drm_dbg_kms(&dev_priv->drm,
+		drm_dbg_kms(&i915->drm,
 			    "[%s:%d] Failed to enable hdcp2.2 (%d)\n",
 			    connector->base.name, connector->base.base.id,
 			    ret);
@@ -2105,13 +2105,13 @@ static void intel_hdcp_check_work(struct work_struct *work)
 static int i915_hdcp_component_bind(struct device *i915_kdev,
 				    struct device *mei_kdev, void *data)
 {
-	struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev);
+	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
 
-	drm_dbg(&dev_priv->drm, "I915 HDCP comp bind\n");
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	dev_priv->display.hdcp.master = (struct i915_hdcp_master *)data;
-	dev_priv->display.hdcp.master->hdcp_dev = mei_kdev;
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	drm_dbg(&i915->drm, "I915 HDCP comp bind\n");
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	i915->display.hdcp.master = (struct i915_hdcp_master *)data;
+	i915->display.hdcp.master->hdcp_dev = mei_kdev;
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return 0;
 }
@@ -2119,12 +2119,12 @@ static int i915_hdcp_component_bind(struct device *i915_kdev,
 static void i915_hdcp_component_unbind(struct device *i915_kdev,
 				       struct device *mei_kdev, void *data)
 {
-	struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev);
+	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
 
-	drm_dbg(&dev_priv->drm, "I915 HDCP comp unbind\n");
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	dev_priv->display.hdcp.master = NULL;
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	drm_dbg(&i915->drm, "I915 HDCP comp unbind\n");
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	i915->display.hdcp.master = NULL;
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 }
 
 static const struct component_ops i915_hdcp_ops = {
@@ -2158,11 +2158,11 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 				     struct intel_digital_port *dig_port,
 				     const struct intel_hdcp_shim *shim)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	enum port port = dig_port->base.port;
 
-	if (DISPLAY_VER(dev_priv) < 12)
+	if (DISPLAY_VER(i915) < 12)
 		data->hdcp_ddi = intel_get_hdcp_ddi_index(port);
 	else
 		/*
@@ -2182,55 +2182,55 @@ static int initialize_hdcp_port_data(struct intel_connector *connector,
 	data->protocol = (u8)shim->protocol;
 
 	if (!data->streams)
-		data->streams = kcalloc(INTEL_NUM_PIPES(dev_priv),
+		data->streams = kcalloc(INTEL_NUM_PIPES(i915),
 					sizeof(struct hdcp2_streamid_type),
 					GFP_KERNEL);
 	if (!data->streams) {
-		drm_err(&dev_priv->drm, "Out of Memory\n");
+		drm_err(&i915->drm, "Out of Memory\n");
 		return -ENOMEM;
 	}
 
 	return 0;
 }
 
-static bool is_hdcp2_supported(struct drm_i915_private *dev_priv)
+static bool is_hdcp2_supported(struct drm_i915_private *i915)
 {
-	if (intel_hdcp_gsc_cs_required(dev_priv))
+	if (intel_hdcp_gsc_cs_required(i915))
 		return true;
 
 	if (!IS_ENABLED(CONFIG_INTEL_MEI_HDCP))
 		return false;
 
-	return (DISPLAY_VER(dev_priv) >= 10 ||
-		IS_KABYLAKE(dev_priv) ||
-		IS_COFFEELAKE(dev_priv) ||
-		IS_COMETLAKE(dev_priv));
+	return (DISPLAY_VER(i915) >= 10 ||
+		IS_KABYLAKE(i915) ||
+		IS_COFFEELAKE(i915) ||
+		IS_COMETLAKE(i915));
 }
 
-void intel_hdcp_component_init(struct drm_i915_private *dev_priv)
+void intel_hdcp_component_init(struct drm_i915_private *i915)
 {
 	int ret;
 
-	if (!is_hdcp2_supported(dev_priv))
+	if (!is_hdcp2_supported(i915))
 		return;
 
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	drm_WARN_ON(&dev_priv->drm, dev_priv->display.hdcp.comp_added);
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	drm_WARN_ON(&i915->drm, i915->display.hdcp.comp_added);
 
-	dev_priv->display.hdcp.comp_added = true;
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
-	if (intel_hdcp_gsc_cs_required(dev_priv))
-		ret = intel_hdcp_gsc_init(dev_priv);
+	i915->display.hdcp.comp_added = true;
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	if (intel_hdcp_gsc_cs_required(i915))
+		ret = intel_hdcp_gsc_init(i915);
 	else
-		ret = component_add_typed(dev_priv->drm.dev, &i915_hdcp_ops,
+		ret = component_add_typed(i915->drm.dev, &i915_hdcp_ops,
 					  I915_COMPONENT_HDCP);
 
 	if (ret < 0) {
-		drm_dbg_kms(&dev_priv->drm, "Failed at fw component add(%d)\n",
+		drm_dbg_kms(&i915->drm, "Failed at fw component add(%d)\n",
 			    ret);
-		mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-		dev_priv->display.hdcp.comp_added = false;
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+		mutex_lock(&i915->display.hdcp.comp_mutex);
+		i915->display.hdcp.comp_added = false;
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return;
 	}
 }
@@ -2256,14 +2256,14 @@ int intel_hdcp_init(struct intel_connector *connector,
 		    struct intel_digital_port *dig_port,
 		    const struct intel_hdcp_shim *shim)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->base.dev);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	int ret;
 
 	if (!shim)
 		return -EINVAL;
 
-	if (is_hdcp2_supported(dev_priv))
+	if (is_hdcp2_supported(i915))
 		intel_hdcp2_init(connector, dig_port, shim);
 
 	ret =
@@ -2335,7 +2335,7 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 		      const struct intel_crtc_state *pipe_config,
 		      const struct drm_connector_state *conn_state)
 {
-	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_connector *connector =
 		to_intel_connector(conn_state->connector);
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
@@ -2347,14 +2347,14 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 		return -ENOENT;
 
 	if (!connector->encoder) {
-		drm_err(&dev_priv->drm, "[%s:%d] encoder is not initialized\n",
+		drm_err(&i915->drm, "[%s:%d] encoder is not initialized\n",
 			connector->base.name, connector->base.base.id);
 		return -ENODEV;
 	}
 
 	mutex_lock(&hdcp->mutex);
 	mutex_lock(&dig_port->hdcp_mutex);
-	drm_WARN_ON(&dev_priv->drm,
+	drm_WARN_ON(&i915->drm,
 		    hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED);
 	hdcp->content_type = (u8)conn_state->content_type;
 
@@ -2366,7 +2366,7 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 		hdcp->stream_transcoder = INVALID_TRANSCODER;
 	}
 
-	if (DISPLAY_VER(dev_priv) >= 12)
+	if (DISPLAY_VER(i915) >= 12)
 		dig_port->hdcp_port_data.hdcp_transcoder =
 			intel_get_hdcp_transcoder(hdcp->cpu_transcoder);
 
@@ -2382,7 +2382,7 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 				check_link_interval =
 					DRM_HDCP2_CHECK_PERIOD_MS;
 		} else {
-			drm_dbg_kms(&dev_priv->drm,
+			drm_dbg_kms(&i915->drm,
 				    "Set content streams failed: (%d)\n",
 				    ret);
 		}
@@ -2498,21 +2498,21 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 		intel_hdcp_enable(state, encoder, crtc_state, conn_state);
 }
 
-void intel_hdcp_component_fini(struct drm_i915_private *dev_priv)
+void intel_hdcp_component_fini(struct drm_i915_private *i915)
 {
-	mutex_lock(&dev_priv->display.hdcp.comp_mutex);
-	if (!dev_priv->display.hdcp.comp_added) {
-		mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.comp_mutex);
+	if (!i915->display.hdcp.comp_added) {
+		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return;
 	}
 
-	dev_priv->display.hdcp.comp_added = false;
-	mutex_unlock(&dev_priv->display.hdcp.comp_mutex);
+	i915->display.hdcp.comp_added = false;
+	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
-	if (intel_hdcp_gsc_cs_required(dev_priv))
-		intel_hdcp_gsc_fini(dev_priv);
+	if (intel_hdcp_gsc_cs_required(i915))
+		intel_hdcp_gsc_fini(i915);
 	else
-		component_del(dev_priv->drm.dev, &i915_hdcp_ops);
+		component_del(i915->drm.dev, &i915_hdcp_ops);
 }
 
 void intel_hdcp_cleanup(struct intel_connector *connector)
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h
index ce283f4f69fd9..5997c52a09589 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.h
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.h
@@ -37,11 +37,11 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 			    struct intel_encoder *encoder,
 			    const struct intel_crtc_state *crtc_state,
 			    const struct drm_connector_state *conn_state);
-bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port);
+bool is_hdcp_supported(struct drm_i915_private *i915, enum port port);
 bool intel_hdcp_capable(struct intel_connector *connector);
 bool intel_hdcp2_capable(struct intel_connector *connector);
-void intel_hdcp_component_init(struct drm_i915_private *dev_priv);
-void intel_hdcp_component_fini(struct drm_i915_private *dev_priv);
+void intel_hdcp_component_init(struct drm_i915_private *i915);
+void intel_hdcp_component_fini(struct drm_i915_private *i915);
 void intel_hdcp_cleanup(struct intel_connector *connector);
 void intel_hdcp_handle_cp_irq(struct intel_connector *connector);
 
-- 
GitLab


From 5c8ec987997ab444df18813e0a3f565f366d05d1 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 29 May 2023 16:37:39 +0530
Subject: [PATCH 0692/1791] drm/i915/hdcp: Move away from master naming to
 arbiter

Rename variables to move away from master convention to
arbiter
%s/hdcp.master/hdcp.arbiter
%s/i915_hdcp_master/i915_hdcp_arbiter
%s/comp_master/comp_arbiter

--v2
- delete i915_hdcp_comp_master redundant declaration [Chaitanya]
- use %s/foo/bar/ format in commit message to show changes [Chaitanya]

--v3
- replace i915_hdcp_comp_master declaration with i915_hdcp_arbiter
to avoid any compile fail with old compilers [Chaitanya]

Cc: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529110740.1522985-3-suraj.kandpal@intel.com
---
 .../gpu/drm/i915/display/intel_display_core.h |  4 +-
 drivers/gpu/drm/i915/display/intel_hdcp.c     | 52 +++++++++----------
 drivers/gpu/drm/i915/display/intel_hdcp_gsc.c | 12 ++---
 drivers/misc/mei/hdcp/mei_hdcp.c              | 26 +++++-----
 include/drm/i915_hdcp_interface.h             |  4 +-
 5 files changed, 49 insertions(+), 49 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h
index e36f88a39b863..23957d80efe88 100644
--- a/drivers/gpu/drm/i915/display/intel_display_core.h
+++ b/drivers/gpu/drm/i915/display/intel_display_core.h
@@ -30,7 +30,7 @@ struct drm_i915_private;
 struct drm_property;
 struct drm_property_blob;
 struct i915_audio_component;
-struct i915_hdcp_comp_master;
+struct i915_hdcp_arbiter;
 struct intel_atomic_state;
 struct intel_audio_funcs;
 struct intel_bios_encoder_data;
@@ -395,7 +395,7 @@ struct intel_display {
 	} gmbus;
 
 	struct {
-		struct i915_hdcp_master *master;
+		struct i915_hdcp_arbiter *arbiter;
 		bool comp_added;
 
 		/*
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index e2e5527efc4b2..bf28c6f1d7482 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -183,7 +183,7 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 
 	/* MEI/GSC interface is solid depending on which is used */
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	if (!i915->display.hdcp.comp_added ||  !i915->display.hdcp.master) {
+	if (!i915->display.hdcp.comp_added ||  !i915->display.hdcp.arbiter) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
 		return false;
 	}
@@ -1114,11 +1114,11 @@ hdcp2_prepare_ake_init(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1144,11 +1144,11 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1172,11 +1172,11 @@ static int hdcp2_verify_hprime(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1198,11 +1198,11 @@ hdcp2_store_pairing_info(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1225,11 +1225,11 @@ hdcp2_prepare_lc_init(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1252,11 +1252,11 @@ hdcp2_verify_lprime(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1278,11 +1278,11 @@ static int hdcp2_prepare_skey(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1307,11 +1307,11 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1337,11 +1337,11 @@ hdcp2_verify_mprime(struct intel_connector *connector,
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1361,11 +1361,11 @@ static int hdcp2_authenticate_port(struct intel_connector *connector)
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct hdcp_port_data *data = &dig_port->hdcp_port_data;
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -1385,11 +1385,11 @@ static int hdcp2_close_session(struct intel_connector *connector)
 {
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct drm_i915_private *i915 = to_i915(connector->base.dev);
-	struct i915_hdcp_master *arbiter;
+	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	arbiter = i915->display.hdcp.master;
+	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
 		mutex_unlock(&i915->display.hdcp.comp_mutex);
@@ -2109,8 +2109,8 @@ static int i915_hdcp_component_bind(struct device *i915_kdev,
 
 	drm_dbg(&i915->drm, "I915 HDCP comp bind\n");
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	i915->display.hdcp.master = (struct i915_hdcp_master *)data;
-	i915->display.hdcp.master->hdcp_dev = mei_kdev;
+	i915->display.hdcp.arbiter = (struct i915_hdcp_arbiter *)data;
+	i915->display.hdcp.arbiter->hdcp_dev = mei_kdev;
 	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
 	return 0;
@@ -2123,7 +2123,7 @@ static void i915_hdcp_component_unbind(struct device *i915_kdev,
 
 	drm_dbg(&i915->drm, "I915 HDCP comp unbind\n");
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	i915->display.hdcp.master = NULL;
+	i915->display.hdcp.arbiter = NULL;
 	mutex_unlock(&i915->display.hdcp.comp_mutex);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
index 4056bb2323ca5..62e7bf0e1035b 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
@@ -697,17 +697,17 @@ static void intel_hdcp_gsc_free_message(struct drm_i915_private *i915)
 
 int intel_hdcp_gsc_init(struct drm_i915_private *i915)
 {
-	struct i915_hdcp_master *data;
+	struct i915_hdcp_arbiter *data;
 	int ret;
 
-	data = kzalloc(sizeof(struct i915_hdcp_master), GFP_KERNEL);
+	data = kzalloc(sizeof(struct i915_hdcp_arbiter), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 
 	mutex_lock(&i915->display.hdcp.comp_mutex);
-	i915->display.hdcp.master = data;
-	i915->display.hdcp.master->hdcp_dev = i915->drm.dev;
-	i915->display.hdcp.master->ops = &gsc_hdcp_ops;
+	i915->display.hdcp.arbiter = data;
+	i915->display.hdcp.arbiter->hdcp_dev = i915->drm.dev;
+	i915->display.hdcp.arbiter->ops = &gsc_hdcp_ops;
 	ret = intel_hdcp_gsc_hdcp2_init(i915);
 	mutex_unlock(&i915->display.hdcp.comp_mutex);
 
@@ -717,7 +717,7 @@ int intel_hdcp_gsc_init(struct drm_i915_private *i915)
 void intel_hdcp_gsc_fini(struct drm_i915_private *i915)
 {
 	intel_hdcp_gsc_free_message(i915);
-	kfree(i915->display.hdcp.master);
+	kfree(i915->display.hdcp.arbiter);
 }
 
 static int intel_gsc_send_sync(struct drm_i915_private *i915,
diff --git a/drivers/misc/mei/hdcp/mei_hdcp.c b/drivers/misc/mei/hdcp/mei_hdcp.c
index 85cbfc3413ee2..51359cc5ece9a 100644
--- a/drivers/misc/mei/hdcp/mei_hdcp.c
+++ b/drivers/misc/mei/hdcp/mei_hdcp.c
@@ -735,13 +735,13 @@ static const struct i915_hdcp_ops mei_hdcp_ops = {
 static int mei_component_master_bind(struct device *dev)
 {
 	struct mei_cl_device *cldev = to_mei_cl_device(dev);
-	struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev);
+	struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev);
 	int ret;
 
 	dev_dbg(dev, "%s\n", __func__);
-	comp_master->ops = &mei_hdcp_ops;
-	comp_master->hdcp_dev = dev;
-	ret = component_bind_all(dev, comp_master);
+	comp_arbiter->ops = &mei_hdcp_ops;
+	comp_arbiter->hdcp_dev = dev;
+	ret = component_bind_all(dev, comp_arbiter);
 	if (ret < 0)
 		return ret;
 
@@ -751,10 +751,10 @@ static int mei_component_master_bind(struct device *dev)
 static void mei_component_master_unbind(struct device *dev)
 {
 	struct mei_cl_device *cldev = to_mei_cl_device(dev);
-	struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev);
+	struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev);
 
 	dev_dbg(dev, "%s\n", __func__);
-	component_unbind_all(dev, comp_master);
+	component_unbind_all(dev, comp_arbiter);
 }
 
 static const struct component_master_ops mei_component_master_ops = {
@@ -799,7 +799,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent,
 static int mei_hdcp_probe(struct mei_cl_device *cldev,
 			  const struct mei_cl_device_id *id)
 {
-	struct i915_hdcp_master *comp_master;
+	struct i915_hdcp_arbiter *comp_arbiter;
 	struct component_match *master_match;
 	int ret;
 
@@ -809,8 +809,8 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev,
 		goto enable_err_exit;
 	}
 
-	comp_master = kzalloc(sizeof(*comp_master), GFP_KERNEL);
-	if (!comp_master) {
+	comp_arbiter = kzalloc(sizeof(*comp_arbiter), GFP_KERNEL);
+	if (!comp_arbiter) {
 		ret = -ENOMEM;
 		goto err_exit;
 	}
@@ -823,7 +823,7 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev,
 		goto err_exit;
 	}
 
-	mei_cldev_set_drvdata(cldev, comp_master);
+	mei_cldev_set_drvdata(cldev, comp_arbiter);
 	ret = component_master_add_with_match(&cldev->dev,
 					      &mei_component_master_ops,
 					      master_match);
@@ -836,7 +836,7 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev,
 
 err_exit:
 	mei_cldev_set_drvdata(cldev, NULL);
-	kfree(comp_master);
+	kfree(comp_arbiter);
 	mei_cldev_disable(cldev);
 enable_err_exit:
 	return ret;
@@ -844,11 +844,11 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev,
 
 static void mei_hdcp_remove(struct mei_cl_device *cldev)
 {
-	struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev);
+	struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev);
 	int ret;
 
 	component_master_del(&cldev->dev, &mei_component_master_ops);
-	kfree(comp_master);
+	kfree(comp_arbiter);
 	mei_cldev_set_drvdata(cldev, NULL);
 
 	ret = mei_cldev_disable(cldev);
diff --git a/include/drm/i915_hdcp_interface.h b/include/drm/i915_hdcp_interface.h
index 2059b066f8a19..4c9c8167c2d5a 100644
--- a/include/drm/i915_hdcp_interface.h
+++ b/include/drm/i915_hdcp_interface.h
@@ -168,12 +168,12 @@ struct i915_hdcp_ops {
 };
 
 /**
- * struct i915_hdcp_master - Used for communication between i915
+ * struct i915_hdcp_arbiter - Used for communication between i915
  * and hdcp drivers for the HDCP2.2 services
  * @hdcp_dev: device that provide the HDCP2.2 service from MEI Bus.
  * @hdcp_ops: Ops implemented by hdcp driver or intel_hdcp_gsc , used by i915 driver.
  */
-struct i915_hdcp_master {
+struct i915_hdcp_arbiter {
 	struct device *hdcp_dev;
 	const struct i915_hdcp_ops *ops;
 
-- 
GitLab


From 3e36c490698660ba4fa7cedfab03f3ed81a07192 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Mon, 29 May 2023 16:37:40 +0530
Subject: [PATCH 0693/1791] drm/i915/hdcp: Rename comp_mutex to hdcp_mutex

Rename comp_mutex to hdcp_mutex as it does not just
protect component related variables which was a terminology
used when hdcp was to be binded as a mei component from MTL
we use gsc cs which does not use the component binding path
for HDCP.

Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529110740.1522985-4-suraj.kandpal@intel.com
---
 .../gpu/drm/i915/display/intel_display_core.h |  4 +-
 drivers/gpu/drm/i915/display/intel_hdcp.c     | 94 +++++++++----------
 drivers/gpu/drm/i915/display/intel_hdcp_gsc.c |  4 +-
 drivers/gpu/drm/i915/i915_driver.c            |  2 +-
 4 files changed, 52 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h
index 23957d80efe88..2209811eb29e7 100644
--- a/drivers/gpu/drm/i915/display/intel_display_core.h
+++ b/drivers/gpu/drm/i915/display/intel_display_core.h
@@ -404,8 +404,8 @@ struct intel_display {
 		 * this is only populated post Meteorlake
 		 */
 		struct intel_hdcp_gsc_message *hdcp_message;
-		/* Mutex to protect the above hdcp component related values. */
-		struct mutex comp_mutex;
+		/* Mutex to protect the above hdcp related values. */
+		struct mutex hdcp_mutex;
 	} hdcp;
 
 	struct {
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index bf28c6f1d7482..17542c28dfd5c 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -182,12 +182,12 @@ bool intel_hdcp2_capable(struct intel_connector *connector)
 	}
 
 	/* MEI/GSC interface is solid depending on which is used */
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	if (!i915->display.hdcp.comp_added ||  !i915->display.hdcp.arbiter) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return false;
 	}
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	/* Sink's capability for HDCP2.2 */
 	hdcp->shim->hdcp_2_2_capable(dig_port, &capable);
@@ -1117,11 +1117,11 @@ hdcp2_prepare_ake_init(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1129,7 +1129,7 @@ hdcp2_prepare_ake_init(struct intel_connector *connector,
 	if (ret)
 		drm_dbg_kms(&i915->drm, "Prepare_ake_init failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1147,11 +1147,11 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1161,7 +1161,7 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Verify rx_cert failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1175,18 +1175,18 @@ static int hdcp2_verify_hprime(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->verify_hprime(arbiter->hdcp_dev, data, rx_hprime);
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Verify hprime failed. %d\n", ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1201,11 +1201,11 @@ hdcp2_store_pairing_info(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1213,7 +1213,7 @@ hdcp2_store_pairing_info(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Store pairing info failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1228,11 +1228,11 @@ hdcp2_prepare_lc_init(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1240,7 +1240,7 @@ hdcp2_prepare_lc_init(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Prepare lc_init failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1255,11 +1255,11 @@ hdcp2_verify_lprime(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1267,7 +1267,7 @@ hdcp2_verify_lprime(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Verify L_Prime failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1281,11 +1281,11 @@ static int hdcp2_prepare_skey(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1293,7 +1293,7 @@ static int hdcp2_prepare_skey(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Get session key failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1310,11 +1310,11 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1325,7 +1325,7 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector,
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm,
 			    "Verify rep topology failed. %d\n", ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1340,18 +1340,18 @@ hdcp2_verify_mprime(struct intel_connector *connector,
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->verify_mprime(arbiter->hdcp_dev, data, stream_ready);
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Verify mprime failed. %d\n", ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1364,11 +1364,11 @@ static int hdcp2_authenticate_port(struct intel_connector *connector)
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
@@ -1376,7 +1376,7 @@ static int hdcp2_authenticate_port(struct intel_connector *connector)
 	if (ret < 0)
 		drm_dbg_kms(&i915->drm, "Enable hdcp auth failed. %d\n",
 			    ret);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -1388,17 +1388,17 @@ static int hdcp2_close_session(struct intel_connector *connector)
 	struct i915_hdcp_arbiter *arbiter;
 	int ret;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	arbiter = i915->display.hdcp.arbiter;
 
 	if (!arbiter || !arbiter->ops) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return -EINVAL;
 	}
 
 	ret = arbiter->ops->close_hdcp_session(arbiter->hdcp_dev,
 					     &dig_port->hdcp_port_data);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
@@ -2108,10 +2108,10 @@ static int i915_hdcp_component_bind(struct device *i915_kdev,
 	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
 
 	drm_dbg(&i915->drm, "I915 HDCP comp bind\n");
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	i915->display.hdcp.arbiter = (struct i915_hdcp_arbiter *)data;
 	i915->display.hdcp.arbiter->hdcp_dev = mei_kdev;
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return 0;
 }
@@ -2122,9 +2122,9 @@ static void i915_hdcp_component_unbind(struct device *i915_kdev,
 	struct drm_i915_private *i915 = kdev_to_i915(i915_kdev);
 
 	drm_dbg(&i915->drm, "I915 HDCP comp unbind\n");
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	i915->display.hdcp.arbiter = NULL;
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 }
 
 static const struct component_ops i915_hdcp_ops = {
@@ -2214,11 +2214,11 @@ void intel_hdcp_component_init(struct drm_i915_private *i915)
 	if (!is_hdcp2_supported(i915))
 		return;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	drm_WARN_ON(&i915->drm, i915->display.hdcp.comp_added);
 
 	i915->display.hdcp.comp_added = true;
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 	if (intel_hdcp_gsc_cs_required(i915))
 		ret = intel_hdcp_gsc_init(i915);
 	else
@@ -2228,9 +2228,9 @@ void intel_hdcp_component_init(struct drm_i915_private *i915)
 	if (ret < 0) {
 		drm_dbg_kms(&i915->drm, "Failed at fw component add(%d)\n",
 			    ret);
-		mutex_lock(&i915->display.hdcp.comp_mutex);
+		mutex_lock(&i915->display.hdcp.hdcp_mutex);
 		i915->display.hdcp.comp_added = false;
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return;
 	}
 }
@@ -2500,14 +2500,14 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 
 void intel_hdcp_component_fini(struct drm_i915_private *i915)
 {
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	if (!i915->display.hdcp.comp_added) {
-		mutex_unlock(&i915->display.hdcp.comp_mutex);
+		mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 		return;
 	}
 
 	i915->display.hdcp.comp_added = false;
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	if (intel_hdcp_gsc_cs_required(i915))
 		intel_hdcp_gsc_fini(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
index 62e7bf0e1035b..72573ce1d0e98 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c
@@ -704,12 +704,12 @@ int intel_hdcp_gsc_init(struct drm_i915_private *i915)
 	if (!data)
 		return -ENOMEM;
 
-	mutex_lock(&i915->display.hdcp.comp_mutex);
+	mutex_lock(&i915->display.hdcp.hdcp_mutex);
 	i915->display.hdcp.arbiter = data;
 	i915->display.hdcp.arbiter->hdcp_dev = i915->drm.dev;
 	i915->display.hdcp.arbiter->ops = &gsc_hdcp_ops;
 	ret = intel_hdcp_gsc_hdcp2_init(i915);
-	mutex_unlock(&i915->display.hdcp.comp_mutex);
+	mutex_unlock(&i915->display.hdcp.hdcp_mutex);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 37532e55327d8..c3ab5c32d4925 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -222,7 +222,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 	mutex_init(&dev_priv->display.audio.mutex);
 	mutex_init(&dev_priv->display.wm.wm_mutex);
 	mutex_init(&dev_priv->display.pps.mutex);
-	mutex_init(&dev_priv->display.hdcp.comp_mutex);
+	mutex_init(&dev_priv->display.hdcp.hdcp_mutex);
 
 	i915_memcpy_init_early(dev_priv);
 	intel_runtime_pm_init_early(&dev_priv->runtime_pm);
-- 
GitLab


From 41b74821751466457dc8b87606ac4d71ec568d4e Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 31 May 2023 03:02:57 +0300
Subject: [PATCH 0694/1791] dt-bindings: display: hdmi-connector: add hdmi-pwr
 supply

Follow the dp-connector example and add hdmi-pwr supply to drive the 5V
pin of the HDMI connector (together with some simple glue logic possibly
attached to the connector).

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531000259.3758235-2-dmitry.baryshkov@linaro.org
---
 .../devicetree/bindings/display/connector/hdmi-connector.yaml  | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/connector/hdmi-connector.yaml b/Documentation/devicetree/bindings/display/connector/hdmi-connector.yaml
index 83c0d008265b3..3ee8f92259845 100644
--- a/Documentation/devicetree/bindings/display/connector/hdmi-connector.yaml
+++ b/Documentation/devicetree/bindings/display/connector/hdmi-connector.yaml
@@ -36,6 +36,9 @@ properties:
     description: GPIO signal to enable DDC bus
     maxItems: 1
 
+  hdmi-pwr-supply:
+    description: Power supply for the HDMI +5V Power pin
+
   port:
     $ref: /schemas/graph.yaml#/properties/port
     description: Connection to controller providing HDMI signals
-- 
GitLab


From 99304fd0050314a0c998c28300393ec574998db9 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 31 May 2023 03:02:58 +0300
Subject: [PATCH 0695/1791] drm/bridge: display-connector: rename dp_pwr to
 connector_pwr

In preparation to adding support for the hdmi_pwr supply, rename dp_pwr
structure field to the generic connector_pwr.

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531000259.3758235-3-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/bridge/display-connector.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c
index 56ae511367b16..cad3105ab1864 100644
--- a/drivers/gpu/drm/bridge/display-connector.c
+++ b/drivers/gpu/drm/bridge/display-connector.c
@@ -24,7 +24,7 @@ struct display_connector {
 	struct gpio_desc	*hpd_gpio;
 	int			hpd_irq;
 
-	struct regulator	*dp_pwr;
+	struct regulator	*supply;
 	struct gpio_desc	*ddc_en;
 };
 
@@ -316,14 +316,14 @@ static int display_connector_probe(struct platform_device *pdev)
 	if (type == DRM_MODE_CONNECTOR_DisplayPort) {
 		int ret;
 
-		conn->dp_pwr = devm_regulator_get_optional(&pdev->dev, "dp-pwr");
+		conn->supply = devm_regulator_get_optional(&pdev->dev, "dp-pwr");
 
-		if (IS_ERR(conn->dp_pwr)) {
-			ret = PTR_ERR(conn->dp_pwr);
+		if (IS_ERR(conn->supply)) {
+			ret = PTR_ERR(conn->supply);
 
 			switch (ret) {
 			case -ENODEV:
-				conn->dp_pwr = NULL;
+				conn->supply = NULL;
 				break;
 
 			case -EPROBE_DEFER:
@@ -335,8 +335,8 @@ static int display_connector_probe(struct platform_device *pdev)
 			}
 		}
 
-		if (conn->dp_pwr) {
-			ret = regulator_enable(conn->dp_pwr);
+		if (conn->supply) {
+			ret = regulator_enable(conn->supply);
 			if (ret) {
 				dev_err(&pdev->dev, "failed to enable DP PWR regulator: %d\n", ret);
 				return ret;
@@ -386,8 +386,8 @@ static void display_connector_remove(struct platform_device *pdev)
 	if (conn->ddc_en)
 		gpiod_set_value(conn->ddc_en, 0);
 
-	if (conn->dp_pwr)
-		regulator_disable(conn->dp_pwr);
+	if (conn->supply)
+		regulator_disable(conn->supply);
 
 	drm_bridge_remove(&conn->bridge);
 
-- 
GitLab


From 6eb6b6f0a012993070b26dacb5ea1ff5fc233c18 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 31 May 2023 03:02:59 +0300
Subject: [PATCH 0696/1791] drm/bridge: display-connector: handle hdmi-pwr
 supply

On some devices the +5V Power pin of the HDMI connector and/or the ESD
protection logic is powered on by a separate regulator. Instead of
declaring this regulator as always-on, make hdmi-connector support the
additional hdmi-pwr supply.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531000259.3758235-4-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/bridge/display-connector.c | 55 ++++++++++++----------
 1 file changed, 29 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c
index cad3105ab1864..f7f436cf96e0a 100644
--- a/drivers/gpu/drm/bridge/display-connector.c
+++ b/drivers/gpu/drm/bridge/display-connector.c
@@ -191,6 +191,18 @@ static irqreturn_t display_connector_hpd_irq(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
+static int display_connector_get_supply(struct platform_device *pdev,
+					struct display_connector *conn,
+					const char *name)
+{
+	conn->supply = devm_regulator_get_optional(&pdev->dev, name);
+
+	if (conn->supply == ERR_PTR(-ENODEV))
+		conn->supply = NULL;
+
+	return PTR_ERR_OR_ZERO(conn->supply);
+}
+
 static int display_connector_probe(struct platform_device *pdev)
 {
 	struct display_connector *conn;
@@ -316,36 +328,15 @@ static int display_connector_probe(struct platform_device *pdev)
 	if (type == DRM_MODE_CONNECTOR_DisplayPort) {
 		int ret;
 
-		conn->supply = devm_regulator_get_optional(&pdev->dev, "dp-pwr");
-
-		if (IS_ERR(conn->supply)) {
-			ret = PTR_ERR(conn->supply);
-
-			switch (ret) {
-			case -ENODEV:
-				conn->supply = NULL;
-				break;
-
-			case -EPROBE_DEFER:
-				return -EPROBE_DEFER;
-
-			default:
-				dev_err(&pdev->dev, "failed to get DP PWR regulator: %d\n", ret);
-				return ret;
-			}
-		}
-
-		if (conn->supply) {
-			ret = regulator_enable(conn->supply);
-			if (ret) {
-				dev_err(&pdev->dev, "failed to enable DP PWR regulator: %d\n", ret);
-				return ret;
-			}
-		}
+		ret = display_connector_get_supply(pdev, conn, "dp-pwr");
+		if (ret < 0)
+			return dev_err_probe(&pdev->dev, ret, "failed to get DP PWR regulator\n");
 	}
 
 	/* enable DDC */
 	if (type == DRM_MODE_CONNECTOR_HDMIA) {
+		int ret;
+
 		conn->ddc_en = devm_gpiod_get_optional(&pdev->dev, "ddc-en",
 						       GPIOD_OUT_HIGH);
 
@@ -353,6 +344,18 @@ static int display_connector_probe(struct platform_device *pdev)
 			dev_err(&pdev->dev, "Couldn't get ddc-en gpio\n");
 			return PTR_ERR(conn->ddc_en);
 		}
+
+		ret = display_connector_get_supply(pdev, conn, "hdmi-pwr");
+		if (ret < 0)
+			return dev_err_probe(&pdev->dev, ret, "failed to get HDMI +5V Power regulator\n");
+	}
+
+	if (conn->supply) {
+		ret = regulator_enable(conn->supply);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to enable PWR regulator: %d\n", ret);
+			return ret;
+		}
 	}
 
 	conn->bridge.funcs = &display_connector_bridge_funcs;
-- 
GitLab


From 2ce977df10c179138e2723b25c2d2c055a3e3cc6 Mon Sep 17 00:00:00 2001
From: Ma Jun <Jun.Ma2@amd.com>
Date: Wed, 31 May 2023 13:30:51 +0800
Subject: [PATCH 0697/1791] drm/ttm: Remove redundant code in
 ttm_tt_init_fields
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove redundant assignment code for ttm->caching as it's overwritten
just a few lines later.

v2:
 - Update the commit message.

Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531053051.3453509-1-Jun.Ma2@amd.com
---
 drivers/gpu/drm/ttm/ttm_tt.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index ab725d9d14a69..1ce4b36ab33be 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -137,7 +137,6 @@ static void ttm_tt_init_fields(struct ttm_tt *ttm,
 			       unsigned long extra_pages)
 {
 	ttm->num_pages = (PAGE_ALIGN(bo->base.size) >> PAGE_SHIFT) + extra_pages;
-	ttm->caching = ttm_cached;
 	ttm->page_flags = page_flags;
 	ttm->dma_address = NULL;
 	ttm->swap_storage = NULL;
-- 
GitLab


From 88e9664434c994e97a9f6f8cdd1535495c660cea Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 00:06:53 +0300
Subject: [PATCH 0698/1791] drm/i915: use localized __diag_ignore_all() instead
 of per file

Use localized __diag_push(), __diag_ignore_all() with rationale, and
__diag_pop() for specific initializations instead of blanket disabling
of -Woverride-init across several files.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525210653.1048972-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                       | 5 -----
 drivers/gpu/drm/i915/display/intel_display_device.c | 5 +++++
 drivers/gpu/drm/i915/display/intel_fbdev.c          | 5 +++++
 drivers/gpu/drm/i915/i915_pci.c                     | 5 +++++
 4 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 06374fc072d3b..1c9ed4c52760a 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -23,11 +23,6 @@ subdir-ccflags-y += $(call cc-disable-warning, unused-but-set-variable)
 subdir-ccflags-y += $(call cc-disable-warning, frame-address)
 subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror
 
-# Fine grained warnings disable
-CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init)
-CFLAGS_display/intel_display_device.o = $(call cc-disable-warning, override-init)
-CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init)
-
 subdir-ccflags-y += -I$(srctree)/$(src)
 
 # Please keep these build lists sorted!
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index a513ac8f71a30..464df1764a863 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -14,6 +14,9 @@
 #include "intel_display_reg_defs.h"
 #include "intel_fbc.h"
 
+__diag_push();
+__diag_ignore_all("-Woverride-init", "Allow overriding inherited members");
+
 static const struct intel_display_device_info no_display = {};
 
 #define PIPE_A_OFFSET		0x70000
@@ -650,6 +653,8 @@ static const struct intel_display_device_info xe_lpdp_display = {
 		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
 };
 
+__diag_pop();
+
 #undef INTEL_VGA_DEVICE
 #undef INTEL_QUANTA_VGA_DEVICE
 #define INTEL_VGA_DEVICE(id, info) { id, info }
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index aab1ae74a8f76..2c3f7befed172 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -130,6 +130,9 @@ static int intel_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
 	return i915_gem_fb_mmap(obj, vma);
 }
 
+__diag_push();
+__diag_ignore_all("-Woverride-init", "Allow overriding the default ops");
+
 static const struct fb_ops intelfb_ops = {
 	.owner = THIS_MODULE,
 	DRM_FB_HELPER_DEFAULT_OPS,
@@ -144,6 +147,8 @@ static const struct fb_ops intelfb_ops = {
 	.fb_mmap = intel_fbdev_mmap,
 };
 
+__diag_pop();
+
 static int intelfb_alloc(struct drm_fb_helper *helper,
 			 struct drm_fb_helper_surface_size *sizes)
 {
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index 928975d5fe2ff..3d7a5db9833b7 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -38,6 +38,9 @@
 #include "i915_reg.h"
 #include "intel_pci_config.h"
 
+__diag_push();
+__diag_ignore_all("-Woverride-init", "Allow overriding inherited members");
+
 #define PLATFORM(x) .platform = (x)
 #define GEN(x) \
 	.__runtime.graphics.ip.ver = (x), \
@@ -843,6 +846,8 @@ static const struct intel_device_info mtl_info = {
 
 #undef PLATFORM
 
+__diag_pop();
+
 /*
  * Make sure any device matches here are from most specific to most
  * general.  For example, since the Quanta match is based on the subsystem
-- 
GitLab


From 84596e1ab09a2a2d4d4898444c01ca08b64cb7bd Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 May 2023 22:11:03 +0200
Subject: [PATCH 0699/1791] drm/i915/guc: Drop legacy CTB definitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We've already switched to new HXG definitions some time ago,
drop legacy CTB definitions to avoid mistakes.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230509201103.538-1-michal.wajdeczko@intel.com
---
 .../gt/uc/abi/guc_communication_ctb_abi.h     | 21 -------------------
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c     |  2 +-
 2 files changed, 1 insertion(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h
index 28b8387f97b77..f7d70db16d760 100644
--- a/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h
+++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h
@@ -167,25 +167,4 @@ static_assert(sizeof(struct guc_ct_buffer_desc) == 64);
  * - **flags**, holds various bits to control message handling
  */
 
-/*
- * Definition of the command transport message header (DW0)
- *
- * bit[4..0]	message len (in dwords)
- * bit[7..5]	reserved
- * bit[8]	response (G2H only)
- * bit[8]	write fence to desc (H2G only)
- * bit[9]	write status to H2G buff (H2G only)
- * bit[10]	send status back via G2H (H2G only)
- * bit[15..11]	reserved
- * bit[31..16]	action code
- */
-#define GUC_CT_MSG_LEN_SHIFT			0
-#define GUC_CT_MSG_LEN_MASK			0x1F
-#define GUC_CT_MSG_IS_RESPONSE			(1 << 8)
-#define GUC_CT_MSG_WRITE_FENCE_TO_DESC		(1 << 8)
-#define GUC_CT_MSG_WRITE_STATUS_TO_BUFF		(1 << 9)
-#define GUC_CT_MSG_SEND_STATUS			(1 << 10)
-#define GUC_CT_MSG_ACTION_SHIFT			16
-#define GUC_CT_MSG_ACTION_MASK			0xFFFF
-
 #endif /* _ABI_GUC_COMMUNICATION_CTB_ABI_H */
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 4aa903be1317b..f28a3a83742dc 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -698,7 +698,7 @@ static int ct_send(struct intel_guc_ct *ct,
 
 	GEM_BUG_ON(!ct->enabled);
 	GEM_BUG_ON(!len);
-	GEM_BUG_ON(len & ~GUC_CT_MSG_LEN_MASK);
+	GEM_BUG_ON(len > GUC_CTB_HXG_MSG_MAX_LEN - GUC_CTB_HDR_LEN);
 	GEM_BUG_ON(!response_buf && response_buf_size);
 	might_sleep();
 
-- 
GitLab


From 09a36015d9a0940214c080f95afc605c47648bbd Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Tue, 23 May 2023 13:40:42 -0700
Subject: [PATCH 0700/1791] drm/i915/perf: Clear out entire reports after
 reading if not power of 2 size

Clearing out report id and timestamp as means to detect unlanded reports
only works if report size is power of 2. That is, only when report size is
a sub-multiple of the OA buffer size can we be certain that reports will
land at the same place each time in the OA buffer (after rewind). If report
size is not a power of 2, we need to zero out the entire report to be able
to detect unlanded reports reliably.

v2: Add Fixes tag (Umesh)

Fixes: 1cc064dce4ed ("drm/i915/perf: Add support for OA media units")
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523204042.4180641-1-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/i915_perf.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 19d5652300eeb..58284156428dc 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -877,12 +877,17 @@ static int gen8_append_oa_reports(struct i915_perf_stream *stream,
 			stream->oa_buffer.last_ctx_id = ctx_id;
 		}
 
-		/*
-		 * Clear out the report id and timestamp as a means to detect unlanded
-		 * reports.
-		 */
-		oa_report_id_clear(stream, report32);
-		oa_timestamp_clear(stream, report32);
+		if (is_power_of_2(report_size)) {
+			/*
+			 * Clear out the report id and timestamp as a means
+			 * to detect unlanded reports.
+			 */
+			oa_report_id_clear(stream, report32);
+			oa_timestamp_clear(stream, report32);
+		} else {
+			/* Zero out the entire report */
+			memset(report32, 0, report_size);
+		}
 	}
 
 	if (start_offset != *offset) {
-- 
GitLab


From ab12902528f414cbb68d0da2b0d518b0ef10e696 Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Wed, 24 May 2023 14:56:28 -0700
Subject: [PATCH 0701/1791] drm/i915/pmu: Turn off the timer to sample
 frequencies when GT is parked

pmu_needs_timer() keeps the timer running even when GT is parked,
ostensibly to sample requested/actual frequencies. However
frequency_sample() has the following:

	/* Report 0/0 (actual/requested) frequency while parked. */
	if (!intel_gt_pm_get_if_awake(gt))
		return;

The above code prevents frequencies to be sampled while the GT is
parked. So we might as well turn off the sampling timer itself in this
case and save CPU cycles/power.

v2: Instead of turning freq bits off, return false, since no counters will
    run after this change when GT is parked (Tvrtko)
v3: Remove gpu_active argument of pmu_needs_timer (Andrzej)

Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524215629.97920-2-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index a814583e19fd7..09313cf9316b4 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -139,7 +139,7 @@ static u32 frequency_enabled_mask(void)
 	return mask;
 }
 
-static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active)
+static bool pmu_needs_timer(struct i915_pmu *pmu)
 {
 	struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu);
 	u32 enable;
@@ -157,17 +157,11 @@ static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active)
 	 */
 	enable &= frequency_enabled_mask() | ENGINE_SAMPLE_MASK;
 
-	/*
-	 * When the GPU is idle per-engine counters do not need to be
-	 * running so clear those bits out.
-	 */
-	if (!gpu_active)
-		enable &= ~ENGINE_SAMPLE_MASK;
 	/*
 	 * Also there is software busyness tracking available we do not
 	 * need the timer for I915_SAMPLE_BUSY counter.
 	 */
-	else if (i915->caps.scheduler & I915_SCHEDULER_CAP_ENGINE_BUSY_STATS)
+	if (i915->caps.scheduler & I915_SCHEDULER_CAP_ENGINE_BUSY_STATS)
 		enable &= ~BIT(I915_SAMPLE_BUSY);
 
 	/*
@@ -295,7 +289,7 @@ static void park_rc6(struct intel_gt *gt)
 
 static void __i915_pmu_maybe_start_timer(struct i915_pmu *pmu)
 {
-	if (!pmu->timer_enabled && pmu_needs_timer(pmu, true)) {
+	if (!pmu->timer_enabled && pmu_needs_timer(pmu)) {
 		pmu->timer_enabled = true;
 		pmu->timer_last = ktime_get();
 		hrtimer_start_range_ns(&pmu->timer,
@@ -321,7 +315,7 @@ void i915_pmu_gt_parked(struct intel_gt *gt)
 	 */
 	pmu->unparked &= ~BIT(gt->info.id);
 	if (pmu->unparked == 0)
-		pmu->timer_enabled = pmu_needs_timer(pmu, false);
+		pmu->timer_enabled = false;
 
 	spin_unlock_irq(&pmu->lock);
 }
@@ -827,7 +821,7 @@ static void i915_pmu_disable(struct perf_event *event)
 	 */
 	if (--pmu->enable_count[bit] == 0) {
 		pmu->enable &= ~BIT(bit);
-		pmu->timer_enabled &= pmu_needs_timer(pmu, true);
+		pmu->timer_enabled &= pmu_needs_timer(pmu);
 	}
 
 	spin_unlock_irqrestore(&pmu->lock, flags);
-- 
GitLab


From 8ed0753b527dc00e759ec1e9e46c0ce9240c0f49 Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Wed, 24 May 2023 14:56:29 -0700
Subject: [PATCH 0702/1791] drm/i915/pmu: Make PMU sample array two-dimensional

No functional changes but we can remove some unsightly index computation
and read/write functions if we convert the PMU sample array from a
one-dimensional to a two-dimensional array.

v2: Retain read/store helpers (Tvrtko)

Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230524215629.97920-3-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 16 +++-------------
 drivers/gpu/drm/i915/i915_pmu.h |  2 +-
 2 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 09313cf9316b4..f96fe92dca4e4 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -191,31 +191,21 @@ static inline s64 ktime_since_raw(const ktime_t kt)
 	return ktime_to_ns(ktime_sub(ktime_get_raw(), kt));
 }
 
-static unsigned int
-__sample_idx(struct i915_pmu *pmu, unsigned int gt_id, int sample)
-{
-	unsigned int idx = gt_id * __I915_NUM_PMU_SAMPLERS + sample;
-
-	GEM_BUG_ON(idx >= ARRAY_SIZE(pmu->sample));
-
-	return idx;
-}
-
 static u64 read_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample)
 {
-	return pmu->sample[__sample_idx(pmu, gt_id, sample)].cur;
+	return pmu->sample[gt_id][sample].cur;
 }
 
 static void
 store_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample, u64 val)
 {
-	pmu->sample[__sample_idx(pmu, gt_id, sample)].cur = val;
+	pmu->sample[gt_id][sample].cur = val;
 }
 
 static void
 add_sample_mult(struct i915_pmu *pmu, unsigned int gt_id, int sample, u32 val, u32 mul)
 {
-	pmu->sample[__sample_idx(pmu, gt_id, sample)].cur += mul_u32_u32(val, mul);
+	pmu->sample[gt_id][sample].cur += mul_u32_u32(val, mul);
 }
 
 static u64 get_rc6(struct intel_gt *gt)
diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index 33d80fbaab8bc..d20592e7db999 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -127,7 +127,7 @@ struct i915_pmu {
 	 * Only global counters are held here, while the per-engine ones are in
 	 * struct intel_engine_cs.
 	 */
-	struct i915_pmu_sample sample[I915_PMU_MAX_GTS * __I915_NUM_PMU_SAMPLERS];
+	struct i915_pmu_sample sample[I915_PMU_MAX_GTS][__I915_NUM_PMU_SAMPLERS];
 	/**
 	 * @sleep_last: Last time GT parked for RC6 estimation.
 	 */
-- 
GitLab


From 54df4868fb728bebbb0fb7c3f187eba383e922b5 Mon Sep 17 00:00:00 2001
From: Fabio Estevam <festevam@denx.de>
Date: Wed, 31 May 2023 19:44:07 -0300
Subject: [PATCH 0703/1791] dt-bindings: samsung,mipi-dsim: Use port-base
 reference

Use port-base reference for port@1.

This fixes the following schema warning:

imx8mp-dhcom-pdk3.dtb: dsi@32e60000: ports:port@1:endpoint: Unevaluated properties are not allowed ('data-lanes' was unexpected)
From schema: Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml

Fixes: 1f0d40d88f7a ("dt-bindings: bridge: Convert Samsung MIPI DSIM bridge to yaml")
Signed-off-by: Fabio Estevam <festevam@denx.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
[narmstrong: removed line break between tags]
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531224407.1611952-1-festevam@gmail.com
---
 .../devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml  | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index 06b6c44d4641c..4ed7a799ba26b 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -103,7 +103,8 @@ properties:
           specified.
 
       port@1:
-        $ref: /schemas/graph.yaml#/properties/port
+        $ref: /schemas/graph.yaml#/$defs/port-base
+        unevaluatedProperties: false
         description:
           DSI output port node to the panel or the next bridge
           in the chain.
-- 
GitLab


From 3b99ee5751c2dfa4b88bb00f573c78735563c4d7 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:16 +0200
Subject: [PATCH 0704/1791] fbdev: Add Kconfig options to select different
 fb_ops helpers

Many fbdev drivers use the same set of fb_ops helpers. Add Kconfig
options to select them at once. This will help with making DRM's
fbdev emulation code more modular, but can also be used to simplify
fbdev's driver configs.

v3:
	* fix select statement (Jingfeng)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-2-tzimmermann@suse.de
---
 drivers/video/fbdev/Kconfig | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig
index 485e8c35d5c68..6e4b6ad9d8e62 100644
--- a/drivers/video/fbdev/Kconfig
+++ b/drivers/video/fbdev/Kconfig
@@ -158,6 +158,27 @@ config FB_DEFERRED_IO
 	bool
 	depends on FB
 
+config FB_IO_HELPERS
+	bool
+	depends on FB
+	select FB_CFB_COPYAREA
+	select FB_CFB_FILLRECT
+	select FB_CFB_IMAGEBLIT
+
+config FB_SYS_HELPERS
+	bool
+	depends on FB
+	select FB_SYS_COPYAREA
+	select FB_SYS_FILLRECT
+	select FB_SYS_FOPS
+	select FB_SYS_IMAGEBLIT
+
+config FB_SYS_HELPERS_DEFERRED
+	bool
+	depends on FB
+	select FB_DEFERRED_IO
+	select FB_SYS_HELPERS
+
 config FB_HECUBA
 	tristate
 	depends on FB
-- 
GitLab


From f1061fa641b8b15c7815d58c20a6c29f2f1f5337 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:17 +0200
Subject: [PATCH 0705/1791] fbdev: Add initializer macros for struct fb_ops

For framebuffers in I/O and system memory, add macros that set
struct fb_ops to the respective callback functions.

For deferred I/O, add macros that generate callback functions with
damage handling. Add initializer macros that set struct fb_ops to
the generated callbacks.

These macros can remove a lot boilerplate code from fbdev drivers.
The drivers are supposed to use the macro that is required for its
framebuffer. Each macro is split into smaller helpers, so that
drivers with non-standard callbacks can pick and customize callbacks
as needed. There are individual helper macros for read/write, mmap
and drawing.

v5:
	* fix whitespace errors (Jingfeng)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-3-tzimmermann@suse.de
---
 include/linux/fb.h | 112 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 112 insertions(+)

diff --git a/include/linux/fb.h b/include/linux/fb.h
index 2cf8efcb9e327..ce6823e157e6b 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -538,9 +538,31 @@ extern ssize_t fb_io_read(struct fb_info *info, char __user *buf,
 extern ssize_t fb_io_write(struct fb_info *info, const char __user *buf,
 			   size_t count, loff_t *ppos);
 
+/*
+ * Initializes struct fb_ops for framebuffers in I/O memory.
+ */
+
+#define __FB_DEFAULT_IO_OPS_RDWR \
+	.fb_read	= fb_io_read, \
+	.fb_write	= fb_io_write
+
+#define __FB_DEFAULT_IO_OPS_DRAW \
+	.fb_fillrect	= cfb_fillrect, \
+	.fb_copyarea	= cfb_copyarea, \
+	.fb_imageblit	= cfb_imageblit
+
+#define __FB_DEFAULT_IO_OPS_MMAP \
+	.fb_mmap	= NULL // default implementation
+
+#define FB_DEFAULT_IO_OPS \
+	__FB_DEFAULT_IO_OPS_RDWR, \
+	__FB_DEFAULT_IO_OPS_DRAW, \
+	__FB_DEFAULT_IO_OPS_MMAP
+
 /*
  * Drawing operations where framebuffer is in system RAM
  */
+
 extern void sys_fillrect(struct fb_info *info, const struct fb_fillrect *rect);
 extern void sys_copyarea(struct fb_info *info, const struct fb_copyarea *area);
 extern void sys_imageblit(struct fb_info *info, const struct fb_image *image);
@@ -549,6 +571,27 @@ extern ssize_t fb_sys_read(struct fb_info *info, char __user *buf,
 extern ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 			    size_t count, loff_t *ppos);
 
+/*
+ * Initializes struct fb_ops for framebuffers in system memory.
+ */
+
+#define __FB_DEFAULT_SYS_OPS_RDWR \
+	.fb_read	= fb_sys_read, \
+	.fb_write	= fb_sys_write
+
+#define __FB_DEFAULT_SYS_OPS_DRAW \
+	.fb_fillrect	= sys_fillrect, \
+	.fb_copyarea	= sys_copyarea, \
+	.fb_imageblit	= sys_imageblit
+
+#define __FB_DEFAULT_SYS_OPS_MMAP \
+	.fb_mmap	= NULL // default implementation
+
+#define FB_DEFAULT_SYS_OPS \
+	__FB_DEFAULT_SYS_OPS_RDWR, \
+	__FB_DEFAULT_SYS_OPS_DRAW, \
+	__FB_DEFAULT_SYS_OPS_MMAP
+
 /* drivers/video/fbmem.c */
 extern int register_framebuffer(struct fb_info *fb_info);
 extern void unregister_framebuffer(struct fb_info *fb_info);
@@ -604,6 +647,75 @@ extern void fb_deferred_io_cleanup(struct fb_info *info);
 extern int fb_deferred_io_fsync(struct file *file, loff_t start,
 				loff_t end, int datasync);
 
+/*
+ * Generate callbacks for deferred I/O
+ */
+
+#define __FB_GEN_DEFAULT_DEFERRED_OPS_RDWR(__prefix, __damage_range, __mode) \
+	static ssize_t __prefix ## _defio_read(struct fb_info *info, char __user *buf, \
+					       size_t count, loff_t *ppos) \
+	{ \
+		return fb_ ## __mode ## _read(info, buf, count, ppos); \
+	} \
+	static ssize_t __prefix ## _defio_write(struct fb_info *info, const char __user *buf, \
+						size_t count, loff_t *ppos) \
+	{ \
+		unsigned long offset = *ppos; \
+		ssize_t ret = fb_ ## __mode ## _write(info, buf, count, ppos); \
+		if (ret > 0) \
+			__damage_range(info, offset, ret); \
+		return ret; \
+	}
+
+#define __FB_GEN_DEFAULT_DEFERRED_OPS_DRAW(__prefix, __damage_area, __mode) \
+	static void __prefix ## _defio_fillrect(struct fb_info *info, \
+						const struct fb_fillrect *rect) \
+	{ \
+		__mode ## _fillrect(info, rect); \
+		__damage_area(info, rect->dx, rect->dy, rect->width, rect->height); \
+	} \
+	static void __prefix ## _defio_copyarea(struct fb_info *info, \
+						const struct fb_copyarea *area) \
+	{ \
+		__mode ## _copyarea(info, area); \
+		__damage_area(info, area->dx, area->dy, area->width, area->height); \
+	} \
+	static void __prefix ## _defio_imageblit(struct fb_info *info, \
+						 const struct fb_image *image) \
+	{ \
+		__mode ## _imageblit(info, image); \
+		__damage_area(info, image->dx, image->dy, image->width, image->height); \
+	}
+
+#define FB_GEN_DEFAULT_DEFERRED_IO_OPS(__prefix, __damage_range, __damage_area) \
+	__FB_GEN_DEFAULT_DEFERRED_OPS_RDWR(__prefix, __damage_range, io) \
+	__FB_GEN_DEFAULT_DEFERRED_OPS_DRAW(__prefix, __damage_area, cfb)
+
+#define FB_GEN_DEFAULT_DEFERRED_SYS_OPS(__prefix, __damage_range, __damage_area) \
+	__FB_GEN_DEFAULT_DEFERRED_OPS_RDWR(__prefix, __damage_range, sys) \
+	__FB_GEN_DEFAULT_DEFERRED_OPS_DRAW(__prefix, __damage_area, sys)
+
+/*
+ * Initializes struct fb_ops for deferred I/O.
+ */
+
+#define __FB_DEFAULT_DEFERRED_OPS_RDWR(__prefix) \
+	.fb_read	= __prefix ## _defio_read, \
+	.fb_write	= __prefix ## _defio_write
+
+#define __FB_DEFAULT_DEFERRED_OPS_DRAW(__prefix) \
+	.fb_fillrect	= __prefix ## _defio_fillrect, \
+	.fb_copyarea	= __prefix ## _defio_copyarea, \
+	.fb_imageblit	= __prefix ## _defio_imageblit
+
+#define __FB_DEFAULT_DEFERRED_OPS_MMAP(__prefix) \
+	.fb_mmap	= fb_deferred_io_mmap
+
+#define FB_DEFAULT_DEFERRED_OPS(__prefix) \
+	__FB_DEFAULT_DEFERRED_OPS_RDWR(__prefix), \
+	__FB_DEFAULT_DEFERRED_OPS_DRAW(__prefix), \
+	__FB_DEFAULT_DEFERRED_OPS_MMAP(__prefix)
+
 static inline bool fb_be_math(struct fb_info *info)
 {
 #ifdef CONFIG_FB_FOREIGN_ENDIAN
-- 
GitLab


From eb958c99f6d4c5ff5f72e6b668b96e217929f8c5 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:18 +0200
Subject: [PATCH 0706/1791] drm/armada: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Armada does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_IO_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Russell King <linux@armlinux.org.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-4-tzimmermann@suse.de
---
 drivers/gpu/drm/armada/Kconfig        | 1 +
 drivers/gpu/drm/armada/armada_fbdev.c | 7 ++-----
 2 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/armada/Kconfig b/drivers/gpu/drm/armada/Kconfig
index f5c66d89ba997..5afade25e217d 100644
--- a/drivers/gpu/drm/armada/Kconfig
+++ b/drivers/gpu/drm/armada/Kconfig
@@ -3,6 +3,7 @@ config DRM_ARMADA
 	tristate "DRM support for Marvell Armada SoCs"
 	depends on DRM && HAVE_CLK && ARM && MMU
 	select DRM_KMS_HELPER
+	select FB_IO_HELPERS if DRM_FBDEV_EMULATION
 	help
 	  Support the "LCD" controllers found on the Marvell Armada 510
 	  devices.  There are two controllers on the device, each controller
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 0a5fd1aa86ebd..3943e89cc06c8 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -5,6 +5,7 @@
  */
 
 #include <linux/errno.h>
+#include <linux/fb.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 
@@ -33,12 +34,8 @@ static void armada_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops armada_fb_ops = {
 	.owner		= THIS_MODULE,
+	FB_DEFAULT_IO_OPS,
 	DRM_FB_HELPER_DEFAULT_OPS,
-	.fb_read	= drm_fb_helper_cfb_read,
-	.fb_write	= drm_fb_helper_cfb_write,
-	.fb_fillrect	= drm_fb_helper_cfb_fillrect,
-	.fb_copyarea	= drm_fb_helper_cfb_copyarea,
-	.fb_imageblit	= drm_fb_helper_cfb_imageblit,
 	.fb_destroy	= armada_fbdev_fb_destroy,
 };
 
-- 
GitLab


From ac9dc1b1b98e2177bd638d5af123669014152f3f Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:19 +0200
Subject: [PATCH 0707/1791] drm/exynos: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Exynos does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v3:
	* don't reorder Makefile rules (Sam)
v2:
	* use FB_IO_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc: Alim Akhtar <alim.akhtar@samsung.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-5-tzimmermann@suse.de
---
 drivers/gpu/drm/exynos/Kconfig            | 1 +
 drivers/gpu/drm/exynos/exynos_drm_fbdev.c | 9 ++++-----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 0cb92d651ff16..7ca7e1dab52c3 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -7,6 +7,7 @@ config DRM_EXYNOS
 	select DRM_DISPLAY_HELPER if DRM_EXYNOS_DP
 	select DRM_KMS_HELPER
 	select VIDEOMODE_HELPERS
+	select FB_IO_HELPERS if DRM_FBDEV_EMULATION
 	select SND_SOC_HDMI_CODEC if SND_SOC
 	help
 	  Choose this option if you have a Samsung SoC Exynos chipset.
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index ea4b3d248aaca..fdf65587f1fe0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -8,6 +8,8 @@
  *	Seung-Woo Kim <sw0312.kim@samsung.com>
  */
 
+#include <linux/fb.h>
+
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_helper.h>
@@ -47,13 +49,10 @@ static void exynos_drm_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops exynos_drm_fb_ops = {
 	.owner		= THIS_MODULE,
+	__FB_DEFAULT_IO_OPS_RDWR,
 	DRM_FB_HELPER_DEFAULT_OPS,
+	__FB_DEFAULT_IO_OPS_DRAW,
 	.fb_mmap        = exynos_drm_fb_mmap,
-	.fb_read	= drm_fb_helper_cfb_read,
-	.fb_write	= drm_fb_helper_cfb_write,
-	.fb_fillrect	= drm_fb_helper_cfb_fillrect,
-	.fb_copyarea	= drm_fb_helper_cfb_copyarea,
-	.fb_imageblit	= drm_fb_helper_cfb_imageblit,
 	.fb_destroy	= exynos_drm_fb_destroy,
 };
 
-- 
GitLab


From 7fca1dd909b3c1a0d7f2478fa939880d7b885491 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:20 +0200
Subject: [PATCH 0708/1791] drm/gma500: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Gma500 does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_IO_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-6-tzimmermann@suse.de
---
 drivers/gpu/drm/gma500/Kconfig | 1 +
 drivers/gpu/drm/gma500/fbdev.c | 8 +++-----
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/gma500/Kconfig b/drivers/gpu/drm/gma500/Kconfig
index 2efc0eb41c642..cd3d92725ed48 100644
--- a/drivers/gpu/drm/gma500/Kconfig
+++ b/drivers/gpu/drm/gma500/Kconfig
@@ -3,6 +3,7 @@ config DRM_GMA500
 	tristate "Intel GMA500/600/3600/3650 KMS Framebuffer"
 	depends on DRM && PCI && X86 && MMU
 	select DRM_KMS_HELPER
+	select FB_IO_HELPERS if DRM_FBDEV_EMULATION
 	select I2C
 	select I2C_ALGOBIT
 	# GMA500 depends on ACPI_VIDEO when ACPI is enabled, just like i915
diff --git a/drivers/gpu/drm/gma500/fbdev.c b/drivers/gpu/drm/gma500/fbdev.c
index 4f0309548b2b5..955cbe9f05a72 100644
--- a/drivers/gpu/drm/gma500/fbdev.c
+++ b/drivers/gpu/drm/gma500/fbdev.c
@@ -5,6 +5,7 @@
  *
  **************************************************************************/
 
+#include <linux/fb.h>
 #include <linux/pfn_t.h>
 
 #include <drm/drm_crtc_helper.h>
@@ -134,13 +135,10 @@ static void psb_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops psb_fbdev_fb_ops = {
 	.owner = THIS_MODULE,
+	__FB_DEFAULT_IO_OPS_RDWR,
 	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_setcolreg = psb_fbdev_fb_setcolreg,
-	.fb_read = drm_fb_helper_cfb_read,
-	.fb_write = drm_fb_helper_cfb_write,
-	.fb_fillrect = drm_fb_helper_cfb_fillrect,
-	.fb_copyarea = drm_fb_helper_cfb_copyarea,
-	.fb_imageblit = drm_fb_helper_cfb_imageblit,
+	__FB_DEFAULT_IO_OPS_DRAW,
 	.fb_mmap = psb_fbdev_fb_mmap,
 	.fb_destroy = psb_fbdev_fb_destroy,
 };
-- 
GitLab


From 6ebf23bf5ea3ba541cced9d7dee782c62f613f9b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:21 +0200
Subject: [PATCH 0709/1791] drm/radeon: Use regular fbdev I/O helpers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Radeon does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_IO_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-7-tzimmermann@suse.de
---
 drivers/gpu/drm/radeon/Kconfig        | 1 +
 drivers/gpu/drm/radeon/radeon_fbdev.c | 9 +++------
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/radeon/Kconfig b/drivers/gpu/drm/radeon/Kconfig
index e19d77d588106..fe498c8af1bb1 100644
--- a/drivers/gpu/drm/radeon/Kconfig
+++ b/drivers/gpu/drm/radeon/Kconfig
@@ -11,6 +11,7 @@ config DRM_RADEON
 	select DRM_SUBALLOC_HELPER
         select DRM_TTM
 	select DRM_TTM_HELPER
+	select FB_IO_HELPERS if DRM_FBDEV_EMULATION
 	select SND_HDA_COMPONENT if SND_HDA_CORE
 	select POWER_SUPPLY
 	select HWMON
diff --git a/drivers/gpu/drm/radeon/radeon_fbdev.c b/drivers/gpu/drm/radeon/radeon_fbdev.c
index fe76e29910efa..28212c2d6c986 100644
--- a/drivers/gpu/drm/radeon/radeon_fbdev.c
+++ b/drivers/gpu/drm/radeon/radeon_fbdev.c
@@ -24,6 +24,7 @@
  *     David Airlie
  */
 
+#include <linux/fb.h>
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
 #include <linux/vga_switcheroo.h>
@@ -190,14 +191,10 @@ static void radeon_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops radeon_fbdev_fb_ops = {
 	.owner = THIS_MODULE,
-	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_open = radeon_fbdev_fb_open,
 	.fb_release = radeon_fbdev_fb_release,
-	.fb_read = drm_fb_helper_cfb_read,
-	.fb_write = drm_fb_helper_cfb_write,
-	.fb_fillrect = drm_fb_helper_cfb_fillrect,
-	.fb_copyarea = drm_fb_helper_cfb_copyarea,
-	.fb_imageblit = drm_fb_helper_cfb_imageblit,
+	FB_DEFAULT_IO_OPS,
+	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_destroy = radeon_fbdev_fb_destroy,
 };
 
-- 
GitLab


From 859cc65f4d97e23cc4388f6e3bc12f6dadef4aaa Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:22 +0200
Subject: [PATCH 0710/1791] drm/fbdev-dma: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Fbdev-dma does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_SYS_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-8-tzimmermann@suse.de
---
 drivers/gpu/drm/Kconfig         |  1 +
 drivers/gpu/drm/drm_fbdev_dma.c | 11 +++++------
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index ba3fb04bb6913..77fb10ddd8a2c 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -223,6 +223,7 @@ config DRM_TTM_HELPER
 config DRM_GEM_DMA_HELPER
 	tristate
 	depends on DRM
+	select FB_SYS_HELPERS if DRM_FBDEV_EMULATION
 	help
 	  Choose this if you need the GEM DMA helper functions
 
diff --git a/drivers/gpu/drm/drm_fbdev_dma.c b/drivers/gpu/drm/drm_fbdev_dma.c
index 728deffcc0d92..d86773fa8ab00 100644
--- a/drivers/gpu/drm/drm_fbdev_dma.c
+++ b/drivers/gpu/drm/drm_fbdev_dma.c
@@ -1,5 +1,7 @@
 // SPDX-License-Identifier: MIT
 
+#include <linux/fb.h>
+
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_drv.h>
 #include <drm/drm_fb_helper.h>
@@ -64,14 +66,11 @@ static const struct fb_ops drm_fbdev_dma_fb_ops = {
 	.owner = THIS_MODULE,
 	.fb_open = drm_fbdev_dma_fb_open,
 	.fb_release = drm_fbdev_dma_fb_release,
-	.fb_read = drm_fb_helper_sys_read,
-	.fb_write = drm_fb_helper_sys_write,
+	__FB_DEFAULT_SYS_OPS_RDWR,
 	DRM_FB_HELPER_DEFAULT_OPS,
-	.fb_fillrect = drm_fb_helper_sys_fillrect,
-	.fb_copyarea = drm_fb_helper_sys_copyarea,
-	.fb_imageblit = drm_fb_helper_sys_imageblit,
-	.fb_destroy = drm_fbdev_dma_fb_destroy,
+	__FB_DEFAULT_SYS_OPS_DRAW,
 	.fb_mmap = drm_fbdev_dma_fb_mmap,
+	.fb_destroy = drm_fbdev_dma_fb_destroy,
 };
 
 /*
-- 
GitLab


From 95da53d63dcf8c1b24a43206593fcdb1fe82dfb2 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:23 +0200
Subject: [PATCH 0711/1791] drm/omapdrm: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Omapdrm does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_SYS_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Tomi Valkeinen <tomba@kernel.org>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-9-tzimmermann@suse.de
---
 drivers/gpu/drm/omapdrm/Kconfig      |  1 +
 drivers/gpu/drm/omapdrm/omap_fbdev.c | 11 +++--------
 2 files changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/omapdrm/Kconfig b/drivers/gpu/drm/omapdrm/Kconfig
index 76ded1568bd09..b4ac76c9f31bf 100644
--- a/drivers/gpu/drm/omapdrm/Kconfig
+++ b/drivers/gpu/drm/omapdrm/Kconfig
@@ -4,6 +4,7 @@ config DRM_OMAP
 	depends on DRM && OF
 	depends on ARCH_OMAP2PLUS
 	select DRM_KMS_HELPER
+	select FB_SYS_HELPERS if DRM_FBDEV_EMULATION
 	select VIDEOMODE_HELPERS
 	select HDMI
 	default n
diff --git a/drivers/gpu/drm/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c
index b950e93b3846a..b7ccce0704a39 100644
--- a/drivers/gpu/drm/omapdrm/omap_fbdev.c
+++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c
@@ -4,6 +4,8 @@
  * Author: Rob Clark <rob@ti.com>
  */
 
+#include <linux/fb.h>
+
 #include <drm/drm_drv.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
@@ -95,20 +97,13 @@ static void omap_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops omap_fb_ops = {
 	.owner = THIS_MODULE,
-
+	FB_DEFAULT_SYS_OPS,
 	.fb_check_var	= drm_fb_helper_check_var,
 	.fb_set_par	= drm_fb_helper_set_par,
 	.fb_setcmap	= drm_fb_helper_setcmap,
 	.fb_blank	= drm_fb_helper_blank,
 	.fb_pan_display = omap_fbdev_pan_display,
 	.fb_ioctl	= drm_fb_helper_ioctl,
-
-	.fb_read = drm_fb_helper_sys_read,
-	.fb_write = drm_fb_helper_sys_write,
-	.fb_fillrect = drm_fb_helper_sys_fillrect,
-	.fb_copyarea = drm_fb_helper_sys_copyarea,
-	.fb_imageblit = drm_fb_helper_sys_imageblit,
-
 	.fb_destroy = omap_fbdev_fb_destroy,
 };
 
-- 
GitLab


From e1d49d763eaadce0332add0d60e5a73eaa08d8fe Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:24 +0200
Subject: [PATCH 0712/1791] drm/tegra: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Tegra does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

v4:
	* use initializer macros for struct fb_ops
v2:
	* use FB_SYS_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Mikko Perttunen <mperttunen@nvidia.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Acked-by: Thierry Reding <treding@nvidia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-10-tzimmermann@suse.de
---
 drivers/gpu/drm/tegra/Kconfig | 1 +
 drivers/gpu/drm/tegra/fbdev.c | 8 +++-----
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig
index 56453ca277c2d..4983137781751 100644
--- a/drivers/gpu/drm/tegra/Kconfig
+++ b/drivers/gpu/drm/tegra/Kconfig
@@ -12,6 +12,7 @@ config DRM_TEGRA
 	select DRM_KMS_HELPER
 	select DRM_MIPI_DSI
 	select DRM_PANEL
+	select FB_SYS_HELPERS if DRM_FBDEV_EMULATION
 	select TEGRA_HOST1X
 	select INTERCONNECT
 	select IOMMU_IOVA
diff --git a/drivers/gpu/drm/tegra/fbdev.c b/drivers/gpu/drm/tegra/fbdev.c
index dca9eccae466b..e74d9be981c79 100644
--- a/drivers/gpu/drm/tegra/fbdev.c
+++ b/drivers/gpu/drm/tegra/fbdev.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/console.h>
+#include <linux/fb.h>
 #include <linux/vmalloc.h>
 
 #include <drm/drm_drv.h>
@@ -58,12 +59,9 @@ static void tegra_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops tegra_fb_ops = {
 	.owner = THIS_MODULE,
+	__FB_DEFAULT_SYS_OPS_RDWR,
 	DRM_FB_HELPER_DEFAULT_OPS,
-	.fb_read = drm_fb_helper_sys_read,
-	.fb_write = drm_fb_helper_sys_write,
-	.fb_fillrect = drm_fb_helper_sys_fillrect,
-	.fb_copyarea = drm_fb_helper_sys_copyarea,
-	.fb_imageblit = drm_fb_helper_sys_imageblit,
+	__FB_DEFAULT_SYS_OPS_DRAW,
 	.fb_mmap = tegra_fb_mmap,
 	.fb_destroy = tegra_fbdev_fb_destroy,
 };
-- 
GitLab


From c51b36207c04bad501ddd47a3d3fe0dbfd611474 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:25 +0200
Subject: [PATCH 0713/1791] drm/fb-helper: Export helpers for marking damage
 areas

Export drm_fb_helper_damage() and drm_fb_helper_damage_range(), which
handle damage areas for fbdev emulation. This is a temporary export
that allows to move the DRM I/O helpers for fbdev into drivers. Only
fbdev-generic and i915 need them. Both will be updated to implement
damage handling by themselves and the exported functions will be removed.

v4:
	* update interfaces

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-11-tzimmermann@suse.de
---
 drivers/gpu/drm/drm_fb_helper.c | 22 ++++++++++++++++++++++
 include/drm/drm_fb_helper.h     |  3 +++
 2 files changed, 25 insertions(+)

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index f0e9549b6bd75..cb03099fd2e37 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -670,6 +670,28 @@ static void drm_fb_helper_memory_range_to_clip(struct fb_info *info, off_t off,
 	drm_rect_init(clip, x1, y1, x2 - x1, y2 - y1);
 }
 
+/* Don't use in new code. */
+void drm_fb_helper_damage_range(struct fb_info *info, off_t off, size_t len)
+{
+	struct drm_fb_helper *fb_helper = info->par;
+	struct drm_rect damage_area;
+
+	drm_fb_helper_memory_range_to_clip(info, off, len, &damage_area);
+	drm_fb_helper_damage(fb_helper, damage_area.x1, damage_area.y1,
+			     drm_rect_width(&damage_area),
+			     drm_rect_height(&damage_area));
+}
+EXPORT_SYMBOL(drm_fb_helper_damage_range);
+
+/* Don't use in new code. */
+void drm_fb_helper_damage_area(struct fb_info *info, u32 x, u32 y, u32 width, u32 height)
+{
+	struct drm_fb_helper *fb_helper = info->par;
+
+	drm_fb_helper_damage(fb_helper, x, y, width, height);
+}
+EXPORT_SYMBOL(drm_fb_helper_damage_area);
+
 /**
  * drm_fb_helper_deferred_io() - fbdev deferred_io callback function
  * @info: fb_info struct pointer
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index 72032c354a30b..7d5804882be7c 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -253,6 +253,9 @@ void drm_fb_helper_fill_info(struct fb_info *info,
 			     struct drm_fb_helper *fb_helper,
 			     struct drm_fb_helper_surface_size *sizes);
 
+void drm_fb_helper_damage_range(struct fb_info *info, off_t off, size_t len);
+void drm_fb_helper_damage_area(struct fb_info *info, u32 x, u32 y, u32 width, u32 height);
+
 void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagereflist);
 
 ssize_t drm_fb_helper_sys_read(struct fb_info *info, char __user *buf,
-- 
GitLab


From 4a2262c121130026306a36888f74f0736fc3fb1c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:26 +0200
Subject: [PATCH 0714/1791] drm/msm: Use regular fbdev I/O helpers

Use the regular fbdev helpers for framebuffer I/O instead of DRM's
helpers. Msm does not use damage handling, so DRM's fbdev helpers
are mere wrappers around the fbdev code.

By using fbdev helpers directly within each DRM fbdev emulation,
we can eventually remove DRM's wrapper functions entirely.

Msm's fbdev emulation has been incomplete as it didn't implement
damage handling. Partilly fix this by implementing damage handling
for write and draw operation. It is still missing for mmaped pages.

v4:
	* use initializer macros for struct fb_ops
	* partially support damage handling
v2:
	* use FB_SYS_HELPERS option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Cc: Sean Paul <sean@poorly.run>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-12-tzimmermann@suse.de
---
 drivers/gpu/drm/msm/Kconfig     |  1 +
 drivers/gpu/drm/msm/msm_fbdev.c | 17 ++++++++---------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 85f5ab1d552c4..a78662bd62730 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -21,6 +21,7 @@ config DRM_MSM
 	select DRM_BRIDGE
 	select DRM_PANEL_BRIDGE
 	select DRM_SCHED
+	select FB_SYS_HELPERS if DRM_FBDEV_EMULATION
 	select SHMEM
 	select TMPFS
 	select QCOM_SCM
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index ce0ba6d1979a7..fa9c1cbffae3c 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -4,6 +4,8 @@
  * Author: Rob Clark <robdclark@gmail.com>
  */
 
+#include <linux/fb.h>
+
 #include <drm/drm_drv.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
@@ -23,6 +25,10 @@ module_param(fbdev, bool, 0600);
  * fbdev funcs, to implement legacy fbdev interface on top of drm driver
  */
 
+FB_GEN_DEFAULT_DEFERRED_SYS_OPS(msm_fbdev,
+				drm_fb_helper_damage_range,
+				drm_fb_helper_damage_area)
+
 static int msm_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
 {
 	struct drm_fb_helper *helper = (struct drm_fb_helper *)info->par;
@@ -52,16 +58,9 @@ static void msm_fbdev_fb_destroy(struct fb_info *info)
 
 static const struct fb_ops msm_fb_ops = {
 	.owner = THIS_MODULE,
+	__FB_DEFAULT_DEFERRED_OPS_RDWR(msm_fbdev),
 	DRM_FB_HELPER_DEFAULT_OPS,
-
-	/* Note: to properly handle manual update displays, we wrap the
-	 * basic fbdev ops which write to the framebuffer
-	 */
-	.fb_read = drm_fb_helper_sys_read,
-	.fb_write = drm_fb_helper_sys_write,
-	.fb_fillrect = drm_fb_helper_sys_fillrect,
-	.fb_copyarea = drm_fb_helper_sys_copyarea,
-	.fb_imageblit = drm_fb_helper_sys_imageblit,
+	__FB_DEFAULT_DEFERRED_OPS_DRAW(msm_fbdev),
 	.fb_mmap = msm_fbdev_mmap,
 	.fb_destroy = msm_fbdev_fb_destroy,
 };
-- 
GitLab


From c6baad68d4e9e5c4c085045c70b99352d4825e05 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:27 +0200
Subject: [PATCH 0715/1791] drm/fbdev-generic: Implement dedicated fbdev I/O
 helpers

Implement dedicated fbdev helpers for framebuffer I/O instead
of using DRM's helpers. Use an fbdev generator macro for
deferred I/O to create the callbacks. Fbdev-generic was the
only caller of the DRM helpers, so remove them from the helper
module.

v4:
	* generate deferred-I/O helpers
	* use initializer macros for fb_ops
v2:
	* use FB_SYS_HELPERS_DEFERRED option

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Tested-by: Sui Jingfeng <suijingfeng@loongson.cn>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-13-tzimmermann@suse.de
---
 drivers/gpu/drm/Kconfig             |   6 +-
 drivers/gpu/drm/drm_fb_helper.c     | 107 ----------------------------
 drivers/gpu/drm/drm_fbdev_generic.c |  11 ++-
 include/drm/drm_fb_helper.h         |  41 -----------
 4 files changed, 6 insertions(+), 159 deletions(-)

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 77fb10ddd8a2c..92a782827b7ba 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -95,6 +95,7 @@ config DRM_KUNIT_TEST
 config DRM_KMS_HELPER
 	tristate
 	depends on DRM
+	select FB_SYS_HELPERS_DEFERRED if DRM_FBDEV_EMULATION
 	help
 	  CRTC helpers for KMS drivers.
 
@@ -135,11 +136,6 @@ config DRM_FBDEV_EMULATION
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
-	select FB_DEFERRED_IO
-	select FB_SYS_FOPS
-	select FB_SYS_FILLRECT
-	select FB_SYS_COPYAREA
-	select FB_SYS_IMAGEBLIT
 	select FRAMEBUFFER_CONSOLE if !EXPERT
 	select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE
 	default y
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index cb03099fd2e37..bab6b252f02a0 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -736,113 +736,6 @@ void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagerefli
 }
 EXPORT_SYMBOL(drm_fb_helper_deferred_io);
 
-/**
- * drm_fb_helper_sys_read - Implements struct &fb_ops.fb_read for system memory
- * @info: fb_info struct pointer
- * @buf: userspace buffer to read from framebuffer memory
- * @count: number of bytes to read from framebuffer memory
- * @ppos: read offset within framebuffer memory
- *
- * Returns:
- * The number of bytes read on success, or an error code otherwise.
- */
-ssize_t drm_fb_helper_sys_read(struct fb_info *info, char __user *buf,
-			       size_t count, loff_t *ppos)
-{
-	return fb_sys_read(info, buf, count, ppos);
-}
-EXPORT_SYMBOL(drm_fb_helper_sys_read);
-
-/**
- * drm_fb_helper_sys_write - Implements struct &fb_ops.fb_write for system memory
- * @info: fb_info struct pointer
- * @buf: userspace buffer to write to framebuffer memory
- * @count: number of bytes to write to framebuffer memory
- * @ppos: write offset within framebuffer memory
- *
- * Returns:
- * The number of bytes written on success, or an error code otherwise.
- */
-ssize_t drm_fb_helper_sys_write(struct fb_info *info, const char __user *buf,
-				size_t count, loff_t *ppos)
-{
-	struct drm_fb_helper *helper = info->par;
-	loff_t pos = *ppos;
-	ssize_t ret;
-	struct drm_rect damage_area;
-
-	ret = fb_sys_write(info, buf, count, ppos);
-	if (ret <= 0)
-		return ret;
-
-	if (helper->funcs->fb_dirty) {
-		drm_fb_helper_memory_range_to_clip(info, pos, ret, &damage_area);
-		drm_fb_helper_damage(helper, damage_area.x1, damage_area.y1,
-				     drm_rect_width(&damage_area),
-				     drm_rect_height(&damage_area));
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL(drm_fb_helper_sys_write);
-
-/**
- * drm_fb_helper_sys_fillrect - wrapper around sys_fillrect
- * @info: fbdev registered by the helper
- * @rect: info about rectangle to fill
- *
- * A wrapper around sys_fillrect implemented by fbdev core
- */
-void drm_fb_helper_sys_fillrect(struct fb_info *info,
-				const struct fb_fillrect *rect)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	sys_fillrect(info, rect);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, rect->dx, rect->dy, rect->width, rect->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_sys_fillrect);
-
-/**
- * drm_fb_helper_sys_copyarea - wrapper around sys_copyarea
- * @info: fbdev registered by the helper
- * @area: info about area to copy
- *
- * A wrapper around sys_copyarea implemented by fbdev core
- */
-void drm_fb_helper_sys_copyarea(struct fb_info *info,
-				const struct fb_copyarea *area)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	sys_copyarea(info, area);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, area->dx, area->dy, area->width, area->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_sys_copyarea);
-
-/**
- * drm_fb_helper_sys_imageblit - wrapper around sys_imageblit
- * @info: fbdev registered by the helper
- * @image: info about image to blit
- *
- * A wrapper around sys_imageblit implemented by fbdev core
- */
-void drm_fb_helper_sys_imageblit(struct fb_info *info,
-				 const struct fb_image *image)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	sys_imageblit(info, image);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, image->dx, image->dy, image->width, image->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_sys_imageblit);
-
 /**
  * drm_fb_helper_cfb_read - Implements struct &fb_ops.fb_read for I/O memory
  * @info: fb_info struct pointer
diff --git a/drivers/gpu/drm/drm_fbdev_generic.c b/drivers/gpu/drm/drm_fbdev_generic.c
index 8e5148bf40bbc..98ae703848a02 100644
--- a/drivers/gpu/drm/drm_fbdev_generic.c
+++ b/drivers/gpu/drm/drm_fbdev_generic.c
@@ -34,6 +34,10 @@ static int drm_fbdev_generic_fb_release(struct fb_info *info, int user)
 	return 0;
 }
 
+FB_GEN_DEFAULT_DEFERRED_SYS_OPS(drm_fbdev_generic,
+				drm_fb_helper_damage_range,
+				drm_fb_helper_damage_area);
+
 static void drm_fbdev_generic_fb_destroy(struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
@@ -56,13 +60,8 @@ static const struct fb_ops drm_fbdev_generic_fb_ops = {
 	.owner		= THIS_MODULE,
 	.fb_open	= drm_fbdev_generic_fb_open,
 	.fb_release	= drm_fbdev_generic_fb_release,
-	.fb_read	= drm_fb_helper_sys_read,
-	.fb_write	= drm_fb_helper_sys_write,
+	FB_DEFAULT_DEFERRED_OPS(drm_fbdev_generic),
 	DRM_FB_HELPER_DEFAULT_OPS,
-	.fb_fillrect	= drm_fb_helper_sys_fillrect,
-	.fb_copyarea	= drm_fb_helper_sys_copyarea,
-	.fb_imageblit	= drm_fb_helper_sys_imageblit,
-	.fb_mmap	= fb_deferred_io_mmap,
 	.fb_destroy	= drm_fbdev_generic_fb_destroy,
 };
 
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index 7d5804882be7c..b50fd0c0b7136 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -258,18 +258,6 @@ void drm_fb_helper_damage_area(struct fb_info *info, u32 x, u32 y, u32 width, u3
 
 void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagereflist);
 
-ssize_t drm_fb_helper_sys_read(struct fb_info *info, char __user *buf,
-			       size_t count, loff_t *ppos);
-ssize_t drm_fb_helper_sys_write(struct fb_info *info, const char __user *buf,
-				size_t count, loff_t *ppos);
-
-void drm_fb_helper_sys_fillrect(struct fb_info *info,
-				const struct fb_fillrect *rect);
-void drm_fb_helper_sys_copyarea(struct fb_info *info,
-				const struct fb_copyarea *area);
-void drm_fb_helper_sys_imageblit(struct fb_info *info,
-				 const struct fb_image *image);
-
 ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
 			       size_t count, loff_t *ppos);
 ssize_t drm_fb_helper_cfb_write(struct fb_info *info, const char __user *buf,
@@ -397,35 +385,6 @@ static inline int drm_fb_helper_defio_init(struct drm_fb_helper *fb_helper)
 	return -ENODEV;
 }
 
-static inline ssize_t drm_fb_helper_sys_read(struct fb_info *info,
-					     char __user *buf, size_t count,
-					     loff_t *ppos)
-{
-	return -ENODEV;
-}
-
-static inline ssize_t drm_fb_helper_sys_write(struct fb_info *info,
-					      const char __user *buf,
-					      size_t count, loff_t *ppos)
-{
-	return -ENODEV;
-}
-
-static inline void drm_fb_helper_sys_fillrect(struct fb_info *info,
-					      const struct fb_fillrect *rect)
-{
-}
-
-static inline void drm_fb_helper_sys_copyarea(struct fb_info *info,
-					      const struct fb_copyarea *area)
-{
-}
-
-static inline void drm_fb_helper_sys_imageblit(struct fb_info *info,
-					       const struct fb_image *image)
-{
-}
-
 static inline ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
 					     size_t count, loff_t *ppos)
 {
-- 
GitLab


From 43049f17b5262826ef64a19762a096782398ef8f Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 30 May 2023 17:12:28 +0200
Subject: [PATCH 0716/1791] drm/i915: Implement dedicated fbdev I/O helpers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement dedicated fbdev helpers for framebuffer I/O instead
of using DRM's helpers. Use an fbdev generator macro for
deferred I/O to create the fbdev callbacks. i915 was the only
caller of the DRM helpers, so remove them from the helper module.

i915's fbdev emulation is still incomplete as it doesn't implement
deferred I/O and damage handling for mmaped pages.

v4:
	* generate deferred-I/O helpers
	* use initializer macros for fb_ops
v2:
	* use FB_IO_HELPERS options

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530151228.22979-14-tzimmermann@suse.de
---
 drivers/gpu/drm/Kconfig                    |   3 -
 drivers/gpu/drm/drm_fb_helper.c            | 107 ---------------------
 drivers/gpu/drm/i915/Kconfig               |   1 +
 drivers/gpu/drm/i915/display/intel_fbdev.c |  14 +--
 include/drm/drm_fb_helper.h                |  39 --------
 5 files changed, 9 insertions(+), 155 deletions(-)

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 92a782827b7ba..bb2e48cc6cd67 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -133,9 +133,6 @@ config DRM_FBDEV_EMULATION
 	bool "Enable legacy fbdev support for your modesetting driver"
 	depends on DRM_KMS_HELPER
 	depends on FB=y || FB=DRM_KMS_HELPER
-	select FB_CFB_FILLRECT
-	select FB_CFB_COPYAREA
-	select FB_CFB_IMAGEBLIT
 	select FRAMEBUFFER_CONSOLE if !EXPERT
 	select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE
 	default y
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index bab6b252f02a0..9978147bbc8ab 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -736,113 +736,6 @@ void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagerefli
 }
 EXPORT_SYMBOL(drm_fb_helper_deferred_io);
 
-/**
- * drm_fb_helper_cfb_read - Implements struct &fb_ops.fb_read for I/O memory
- * @info: fb_info struct pointer
- * @buf: userspace buffer to read from framebuffer memory
- * @count: number of bytes to read from framebuffer memory
- * @ppos: read offset within framebuffer memory
- *
- * Returns:
- * The number of bytes read on success, or an error code otherwise.
- */
-ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
-			       size_t count, loff_t *ppos)
-{
-	return fb_io_read(info, buf, count, ppos);
-}
-EXPORT_SYMBOL(drm_fb_helper_cfb_read);
-
-/**
- * drm_fb_helper_cfb_write - Implements struct &fb_ops.fb_write for I/O memory
- * @info: fb_info struct pointer
- * @buf: userspace buffer to write to framebuffer memory
- * @count: number of bytes to write to framebuffer memory
- * @ppos: write offset within framebuffer memory
- *
- * Returns:
- * The number of bytes written on success, or an error code otherwise.
- */
-ssize_t drm_fb_helper_cfb_write(struct fb_info *info, const char __user *buf,
-				size_t count, loff_t *ppos)
-{
-	struct drm_fb_helper *helper = info->par;
-	loff_t pos = *ppos;
-	ssize_t ret;
-	struct drm_rect damage_area;
-
-	ret = fb_io_write(info, buf, count, ppos);
-	if (ret <= 0)
-		return ret;
-
-	if (helper->funcs->fb_dirty) {
-		drm_fb_helper_memory_range_to_clip(info, pos, ret, &damage_area);
-		drm_fb_helper_damage(helper, damage_area.x1, damage_area.y1,
-				     drm_rect_width(&damage_area),
-				     drm_rect_height(&damage_area));
-	}
-
-	return ret;
-}
-EXPORT_SYMBOL(drm_fb_helper_cfb_write);
-
-/**
- * drm_fb_helper_cfb_fillrect - wrapper around cfb_fillrect
- * @info: fbdev registered by the helper
- * @rect: info about rectangle to fill
- *
- * A wrapper around cfb_fillrect implemented by fbdev core
- */
-void drm_fb_helper_cfb_fillrect(struct fb_info *info,
-				const struct fb_fillrect *rect)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	cfb_fillrect(info, rect);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, rect->dx, rect->dy, rect->width, rect->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_cfb_fillrect);
-
-/**
- * drm_fb_helper_cfb_copyarea - wrapper around cfb_copyarea
- * @info: fbdev registered by the helper
- * @area: info about area to copy
- *
- * A wrapper around cfb_copyarea implemented by fbdev core
- */
-void drm_fb_helper_cfb_copyarea(struct fb_info *info,
-				const struct fb_copyarea *area)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	cfb_copyarea(info, area);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, area->dx, area->dy, area->width, area->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_cfb_copyarea);
-
-/**
- * drm_fb_helper_cfb_imageblit - wrapper around cfb_imageblit
- * @info: fbdev registered by the helper
- * @image: info about image to blit
- *
- * A wrapper around cfb_imageblit implemented by fbdev core
- */
-void drm_fb_helper_cfb_imageblit(struct fb_info *info,
-				 const struct fb_image *image)
-{
-	struct drm_fb_helper *helper = info->par;
-
-	cfb_imageblit(info, image);
-
-	if (helper->funcs->fb_dirty)
-		drm_fb_helper_damage(helper, image->dx, image->dy, image->width, image->height);
-}
-EXPORT_SYMBOL(drm_fb_helper_cfb_imageblit);
-
 /**
  * drm_fb_helper_set_suspend - wrapper around fb_set_suspend
  * @fb_helper: driver-allocated fbdev helper, can be NULL
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index e4f4d2e3fdfeb..01b5a8272a271 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -17,6 +17,7 @@ config DRM_I915
 	select DRM_KMS_HELPER
 	select DRM_PANEL
 	select DRM_MIPI_DSI
+	select FB_IO_HELPERS if DRM_FBDEV_EMULATION
 	select RELAY
 	select I2C
 	select I2C_ALGOBIT
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index aab1ae74a8f76..eccaceaf8b9d1 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -28,6 +28,7 @@
 #include <linux/console.h>
 #include <linux/delay.h>
 #include <linux/errno.h>
+#include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
@@ -84,6 +85,10 @@ static void intel_fbdev_invalidate(struct intel_fbdev *ifbdev)
 	intel_frontbuffer_invalidate(to_frontbuffer(ifbdev), ORIGIN_CPU);
 }
 
+FB_GEN_DEFAULT_DEFERRED_IO_OPS(intel_fbdev,
+			       drm_fb_helper_damage_range,
+			       drm_fb_helper_damage_area)
+
 static int intel_fbdev_set_par(struct fb_info *info)
 {
 	struct intel_fbdev *ifbdev = to_intel_fbdev(info->par);
@@ -132,15 +137,12 @@ static int intel_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
 
 static const struct fb_ops intelfb_ops = {
 	.owner = THIS_MODULE,
+	__FB_DEFAULT_DEFERRED_OPS_RDWR(intel_fbdev),
 	DRM_FB_HELPER_DEFAULT_OPS,
 	.fb_set_par = intel_fbdev_set_par,
-	.fb_read = drm_fb_helper_cfb_read,
-	.fb_write = drm_fb_helper_cfb_write,
-	.fb_fillrect = drm_fb_helper_cfb_fillrect,
-	.fb_copyarea = drm_fb_helper_cfb_copyarea,
-	.fb_imageblit = drm_fb_helper_cfb_imageblit,
-	.fb_pan_display = intel_fbdev_pan_display,
 	.fb_blank = intel_fbdev_blank,
+	.fb_pan_display = intel_fbdev_pan_display,
+	__FB_DEFAULT_DEFERRED_OPS_DRAW(intel_fbdev),
 	.fb_mmap = intel_fbdev_mmap,
 };
 
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index b50fd0c0b7136..4863b0f8299e8 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -258,18 +258,6 @@ void drm_fb_helper_damage_area(struct fb_info *info, u32 x, u32 y, u32 width, u3
 
 void drm_fb_helper_deferred_io(struct fb_info *info, struct list_head *pagereflist);
 
-ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
-			       size_t count, loff_t *ppos);
-ssize_t drm_fb_helper_cfb_write(struct fb_info *info, const char __user *buf,
-				size_t count, loff_t *ppos);
-
-void drm_fb_helper_cfb_fillrect(struct fb_info *info,
-				const struct fb_fillrect *rect);
-void drm_fb_helper_cfb_copyarea(struct fb_info *info,
-				const struct fb_copyarea *area);
-void drm_fb_helper_cfb_imageblit(struct fb_info *info,
-				 const struct fb_image *image);
-
 void drm_fb_helper_set_suspend(struct drm_fb_helper *fb_helper, bool suspend);
 void drm_fb_helper_set_suspend_unlocked(struct drm_fb_helper *fb_helper,
 					bool suspend);
@@ -385,33 +373,6 @@ static inline int drm_fb_helper_defio_init(struct drm_fb_helper *fb_helper)
 	return -ENODEV;
 }
 
-static inline ssize_t drm_fb_helper_cfb_read(struct fb_info *info, char __user *buf,
-					     size_t count, loff_t *ppos)
-{
-	return -ENODEV;
-}
-
-static inline ssize_t drm_fb_helper_cfb_write(struct fb_info *info, const char __user *buf,
-					      size_t count, loff_t *ppos)
-{
-	return -ENODEV;
-}
-
-static inline void drm_fb_helper_cfb_fillrect(struct fb_info *info,
-					      const struct fb_fillrect *rect)
-{
-}
-
-static inline void drm_fb_helper_cfb_copyarea(struct fb_info *info,
-					      const struct fb_copyarea *area)
-{
-}
-
-static inline void drm_fb_helper_cfb_imageblit(struct fb_info *info,
-					       const struct fb_image *image)
-{
-}
-
 static inline void drm_fb_helper_set_suspend(struct drm_fb_helper *fb_helper,
 					     bool suspend)
 {
-- 
GitLab


From e8001973bb4567e8dece6fdcbe4c72313df9945e Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 30 May 2023 21:28:04 +0200
Subject: [PATCH 0717/1791] dt-bindings: display: bridge: tc358762: Document
 reset-gpios

This chip has one reset GPIO input, document it. The reset GPIO
is optional as it is sometimes not connected on some hardware.

Signed-off-by: Marek Vasut <marex@denx.de>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530192805.648646-1-marex@denx.de
---
 .../devicetree/bindings/display/bridge/toshiba,tc358762.yaml   | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358762.yaml b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358762.yaml
index 81ca3cbc7abed..6c1de0b217228 100644
--- a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358762.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358762.yaml
@@ -21,6 +21,9 @@ properties:
     maxItems: 1
     description: virtual channel number of a DSI peripheral
 
+  reset-gpios:
+    maxItems: 1
+
   vddc-supply:
     description: Regulator for 1.2V internal core power.
 
-- 
GitLab


From 3355f4ee561da6065cc756bf54a3f8bb556889df Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 30 May 2023 21:28:05 +0200
Subject: [PATCH 0718/1791] drm/bridge: tc358762: Add reset GPIO support

Add reset GPIO support. The reset GPIO is cleared after supply regulator
was enabled, and set before supply regulator is disabled.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Robert Foss <rfoss@kernel.org>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530192805.648646-2-marex@denx.de
---
 drivers/gpu/drm/bridge/tc358762.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/bridge/tc358762.c b/drivers/gpu/drm/bridge/tc358762.c
index 77f7f7f547570..5641395fd310e 100644
--- a/drivers/gpu/drm/bridge/tc358762.c
+++ b/drivers/gpu/drm/bridge/tc358762.c
@@ -11,6 +11,7 @@
  */
 
 #include <linux/delay.h>
+#include <linux/gpio/consumer.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/of_graph.h>
@@ -63,6 +64,7 @@ struct tc358762 {
 	struct drm_bridge bridge;
 	struct regulator *regulator;
 	struct drm_bridge *panel_bridge;
+	struct gpio_desc *reset_gpio;
 	bool pre_enabled;
 	int error;
 };
@@ -138,6 +140,9 @@ static void tc358762_post_disable(struct drm_bridge *bridge)
 
 	ctx->pre_enabled = false;
 
+	if (ctx->reset_gpio)
+		gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+
 	ret = regulator_disable(ctx->regulator);
 	if (ret < 0)
 		dev_err(ctx->dev, "error disabling regulators (%d)\n", ret);
@@ -152,6 +157,11 @@ static void tc358762_pre_enable(struct drm_bridge *bridge)
 	if (ret < 0)
 		dev_err(ctx->dev, "error enabling regulators (%d)\n", ret);
 
+	if (ctx->reset_gpio) {
+		gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+		usleep_range(5000, 10000);
+	}
+
 	ret = tc358762_init(ctx);
 	if (ret < 0)
 		dev_err(ctx->dev, "error initializing bridge (%d)\n", ret);
@@ -185,6 +195,11 @@ static int tc358762_parse_dt(struct tc358762 *ctx)
 
 	ctx->panel_bridge = panel_bridge;
 
+	/* Reset GPIO is optional */
+	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW);
+	if (IS_ERR(ctx->reset_gpio))
+		return PTR_ERR(ctx->reset_gpio);
+
 	return 0;
 }
 
-- 
GitLab


From 0628f2341e96213c9f2d074853b255f65acd3795 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:07 +0200
Subject: [PATCH 0719/1791] dt-bindings: display: add Amlogic MIPI DSI Host
 Controller bindings

The Amlogic G12A, G12B & SM1 SoCs embeds a Synopsys DW-MIPI-DSI
transceiver (ver 1.21a) with a custom glue managing the IP resets,
clock and data inputs similar to the DW-HDMI Glue on the same
Amlogic SoC families.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-6-56eb7a4d5b8e@linaro.org
---
 .../amlogic,meson-g12a-dw-mipi-dsi.yaml       | 118 ++++++++++++++++++
 1 file changed, 118 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/amlogic,meson-g12a-dw-mipi-dsi.yaml

diff --git a/Documentation/devicetree/bindings/display/amlogic,meson-g12a-dw-mipi-dsi.yaml b/Documentation/devicetree/bindings/display/amlogic,meson-g12a-dw-mipi-dsi.yaml
new file mode 100644
index 0000000000000..a3428f0120059
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/amlogic,meson-g12a-dw-mipi-dsi.yaml
@@ -0,0 +1,118 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+# Copyright 2020 BayLibre, SAS
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/amlogic,meson-g12a-dw-mipi-dsi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Amlogic specific extensions to the Synopsys Designware MIPI DSI Host Controller
+
+maintainers:
+  - Neil Armstrong <neil.armstrong@linaro.org>
+
+description: |
+  The Amlogic Meson Synopsys Designware Integration is composed of
+  - A Synopsys DesignWare MIPI DSI Host Controller IP
+  - A TOP control block controlling the Clocks & Resets of the IP
+
+allOf:
+  - $ref: dsi-controller.yaml#
+
+properties:
+  compatible:
+    enum:
+      - amlogic,meson-g12a-dw-mipi-dsi
+
+  reg:
+    maxItems: 1
+
+  clocks:
+    minItems: 3
+    maxItems: 4
+
+  clock-names:
+    minItems: 3
+    items:
+      - const: pclk
+      - const: bit
+      - const: px
+      - const: meas
+
+  resets:
+    maxItems: 1
+
+  reset-names:
+    items:
+      - const: top
+
+  phys:
+    maxItems: 1
+
+  phy-names:
+    items:
+      - const: dphy
+
+  ports:
+    $ref: /schemas/graph.yaml#/properties/ports
+
+    properties:
+      port@0:
+        $ref: /schemas/graph.yaml#/properties/port
+        description: Input node to receive pixel data.
+
+      port@1:
+        $ref: /schemas/graph.yaml#/properties/port
+        description: DSI output node to panel.
+
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - reg
+  - clocks
+  - clock-names
+  - resets
+  - reset-names
+  - phys
+  - phy-names
+  - ports
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    dsi@6000 {
+          compatible = "amlogic,meson-g12a-dw-mipi-dsi";
+          reg = <0x6000 0x400>;
+          resets = <&reset_top>;
+          reset-names = "top";
+          clocks = <&clk_pclk>, <&bit_clk>, <&clk_px>;
+          clock-names = "pclk", "bit", "px";
+          phys = <&mipi_dphy>;
+          phy-names = "dphy";
+
+          ports {
+              #address-cells = <1>;
+              #size-cells = <0>;
+
+              /* VPU VENC Input */
+              mipi_dsi_venc_port: port@0 {
+                  reg = <0>;
+
+                  mipi_dsi_in: endpoint {
+                       remote-endpoint = <&dpi_out>;
+                  };
+              };
+
+              /* DSI Output */
+              mipi_dsi_panel_port: port@1 {
+                  reg = <1>;
+
+                  mipi_out_panel: endpoint {
+                      remote-endpoint = <&mipi_in_panel>;
+                  };
+              };
+          };
+    };
-- 
GitLab


From 25b3b35cd51ef0d98165666d250a51f39db6a1fc Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:08 +0200
Subject: [PATCH 0720/1791] dt-bindings: display: meson-vpu: add third DPI
 output port

Add third port corresponding to the ENCL DPI encoder used to connect
to DSI or LVDS transceivers.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-7-56eb7a4d5b8e@linaro.org
---
 .../devicetree/bindings/display/amlogic,meson-vpu.yaml       | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/amlogic,meson-vpu.yaml b/Documentation/devicetree/bindings/display/amlogic,meson-vpu.yaml
index 0c72120acc4f6..cb0a90f023219 100644
--- a/Documentation/devicetree/bindings/display/amlogic,meson-vpu.yaml
+++ b/Documentation/devicetree/bindings/display/amlogic,meson-vpu.yaml
@@ -96,6 +96,11 @@ properties:
     description:
       A port node pointing to the HDMI-TX port node.
 
+  port@2:
+    $ref: /schemas/graph.yaml#/properties/port
+    description:
+      A port node pointing to the DPI port node (e.g. DSI or LVDS transceiver).
+
   "#address-cells":
     const: 1
 
-- 
GitLab


From 6a044642988b5f8285f3173b8e88784bef2bc306 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:09 +0200
Subject: [PATCH 0721/1791] drm/meson: fix unbind path if HDMI fails to bind

If the case the HDMI controller fails to bind, we try to unbind
all components before calling drm_dev_put() which makes drm_bridge_detach()
crash because unbinding the HDMI controller frees the bridge memory.

The solution is the unbind all components at the end like in the remove
path.

Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-8-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/meson/meson_drv.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index ca6d1e59e5d9f..e060279dc80a6 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -316,32 +316,34 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 		goto exit_afbcd;
 
 	if (has_components) {
-		ret = component_bind_all(drm->dev, drm);
+		ret = component_bind_all(dev, drm);
 		if (ret) {
 			dev_err(drm->dev, "Couldn't bind all components\n");
+			/* Do not try to unbind */
+			has_components = false;
 			goto exit_afbcd;
 		}
 	}
 
 	ret = meson_encoder_hdmi_init(priv);
 	if (ret)
-		goto unbind_all;
+		goto exit_afbcd;
 
 	ret = meson_plane_create(priv);
 	if (ret)
-		goto unbind_all;
+		goto exit_afbcd;
 
 	ret = meson_overlay_create(priv);
 	if (ret)
-		goto unbind_all;
+		goto exit_afbcd;
 
 	ret = meson_crtc_create(priv);
 	if (ret)
-		goto unbind_all;
+		goto exit_afbcd;
 
 	ret = request_irq(priv->vsync_irq, meson_irq, 0, drm->driver->name, drm);
 	if (ret)
-		goto unbind_all;
+		goto exit_afbcd;
 
 	drm_mode_config_reset(drm);
 
@@ -359,15 +361,18 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 
 uninstall_irq:
 	free_irq(priv->vsync_irq, drm);
-unbind_all:
-	if (has_components)
-		component_unbind_all(drm->dev, drm);
 exit_afbcd:
 	if (priv->afbcd.ops)
 		priv->afbcd.ops->exit(priv);
 free_drm:
 	drm_dev_put(drm);
 
+	meson_encoder_hdmi_remove(priv);
+	meson_encoder_cvbs_remove(priv);
+
+	if (has_components)
+		component_unbind_all(dev, drm);
+
 	return ret;
 }
 
-- 
GitLab


From 44e16166e0e9b94d8bcdf55fc0e5fcceca1154f0 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:10 +0200
Subject: [PATCH 0722/1791] drm/meson: only use components with dw-hdmi

Only DW-HDMI currently needs components since it reuses
the drm-meson driver context to access HHI registers (sic).

Once this is solved, we can get rid on components.

Until now, limit the components matching to the dw-hdmi compatibles
we know to require this hack, for other bridges simply use probe defer
instead and get over this components sitation.

The back story is that we simply cannot attach DSI adapters bridges
if we use components, only DSI panels, this is because we bind/unbind
the DSI controller at each drm-meson driver master bind tentative.
With this the I2C DSI bridge is unable to find the DSI controller
host and everything fails to probe.

This will simplify a lot adding new or older HDMI bridges.

Cc: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com> # on Khadas VIM3 + TS050 Panel
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-9-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/meson/meson_drv.c | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index e060279dc80a6..e935c0286a201 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -451,10 +451,17 @@ static void meson_drv_shutdown(struct platform_device *pdev)
 	drm_atomic_helper_shutdown(priv->drm);
 }
 
-/* Possible connectors nodes to ignore */
-static const struct of_device_id connectors_match[] = {
-	{ .compatible = "composite-video-connector" },
-	{ .compatible = "svideo-connector" },
+/*
+ * Only devices to use as components
+ * TOFIX: get rid of components when we can finally
+ * get meson_dx_hdmi to stop using the meson_drm
+ * private structure for HHI registers.
+ */
+static const struct of_device_id components_dev_match[] = {
+	{ .compatible = "amlogic,meson-gxbb-dw-hdmi" },
+	{ .compatible = "amlogic,meson-gxl-dw-hdmi" },
+	{ .compatible = "amlogic,meson-gxm-dw-hdmi" },
+	{ .compatible = "amlogic,meson-g12a-dw-hdmi" },
 	{}
 };
 
@@ -472,17 +479,12 @@ static int meson_drv_probe(struct platform_device *pdev)
 			continue;
 		}
 
-		/* If an analog connector is detected, count it as an output */
-		if (of_match_node(connectors_match, remote)) {
-			++count;
-			of_node_put(remote);
-			continue;
-		}
-
-		dev_dbg(&pdev->dev, "parent %pOF remote match add %pOF parent %s\n",
-			np, remote, dev_name(&pdev->dev));
+		if (of_match_node(components_dev_match, remote)) {
+			component_match_add(&pdev->dev, &match, component_compare_of, remote);
 
-		component_match_add(&pdev->dev, &match, component_compare_of, remote);
+			dev_dbg(&pdev->dev, "parent %pOF remote match add %pOF parent %s\n",
+				np, remote, dev_name(&pdev->dev));
+		}
 
 		of_node_put(remote);
 
-- 
GitLab


From 51fc01a03442cce5e4c21375a1ceb2e4ec93c833 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:11 +0200
Subject: [PATCH 0723/1791] drm/meson: venc: add ENCL encoder setup for
 MIPI-DSI output

This adds supports for the ENCL encoder connected to a MIPI-DSI transceiver on the
Amlogic AXG, G12A, G12B & SM1 SoCs.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com> # on Khadas VIM3 + TS050 Panel
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-10-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/meson/meson_registers.h |  25 +++
 drivers/gpu/drm/meson/meson_venc.c      | 211 +++++++++++++++++++++++-
 drivers/gpu/drm/meson/meson_venc.h      |   6 +
 drivers/gpu/drm/meson/meson_vpp.h       |   2 +
 4 files changed, 242 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_registers.h b/drivers/gpu/drm/meson/meson_registers.h
index 0f3cafab88600..3d73d00a1f4c6 100644
--- a/drivers/gpu/drm/meson/meson_registers.h
+++ b/drivers/gpu/drm/meson/meson_registers.h
@@ -812,6 +812,7 @@
 #define VENC_STATA 0x1b6d
 #define VENC_INTCTRL 0x1b6e
 #define		VENC_INTCTRL_ENCI_LNRST_INT_EN  BIT(1)
+#define		VENC_INTCTRL_ENCP_LNRST_INT_EN  BIT(9)
 #define VENC_INTFLAG 0x1b6f
 #define VENC_VIDEO_TST_EN 0x1b70
 #define VENC_VIDEO_TST_MDSEL 0x1b71
@@ -1192,7 +1193,11 @@
 #define ENCL_VIDEO_PB_OFFST 0x1ca5
 #define ENCL_VIDEO_PR_OFFST 0x1ca6
 #define ENCL_VIDEO_MODE 0x1ca7
+#define		ENCL_PX_LN_CNT_SHADOW_EN	BIT(15)
 #define ENCL_VIDEO_MODE_ADV 0x1ca8
+#define		ENCL_VIDEO_MODE_ADV_VFIFO_EN	BIT(3)
+#define		ENCL_VIDEO_MODE_ADV_GAIN_HDTV	BIT(4)
+#define		ENCL_SEL_GAMMA_RGB_IN		BIT(10)
 #define ENCL_DBG_PX_RST 0x1ca9
 #define ENCL_DBG_LN_RST 0x1caa
 #define ENCL_DBG_PX_INT 0x1cab
@@ -1219,11 +1224,14 @@
 #define ENCL_VIDEO_VOFFST 0x1cc0
 #define ENCL_VIDEO_RGB_CTRL 0x1cc1
 #define ENCL_VIDEO_FILT_CTRL 0x1cc2
+#define		ENCL_VIDEO_FILT_CTRL_BYPASS_FILTER	BIT(12)
 #define ENCL_VIDEO_OFLD_VPEQ_OFST 0x1cc3
 #define ENCL_VIDEO_OFLD_VOAV_OFST 0x1cc4
 #define ENCL_VIDEO_MATRIX_CB 0x1cc5
 #define ENCL_VIDEO_MATRIX_CR 0x1cc6
 #define ENCL_VIDEO_RGBIN_CTRL 0x1cc7
+#define		ENCL_VIDEO_RGBIN_RGB	BIT(0)
+#define		ENCL_VIDEO_RGBIN_ZBLK	BIT(1)
 #define ENCL_MAX_LINE_SWITCH_POINT 0x1cc8
 #define ENCL_DACSEL_0 0x1cc9
 #define ENCL_DACSEL_1 0x1cca
@@ -1300,13 +1308,28 @@
 #define RDMA_STATUS2 0x1116
 #define RDMA_STATUS3 0x1117
 #define L_GAMMA_CNTL_PORT 0x1400
+#define		L_GAMMA_CNTL_PORT_VCOM_POL	BIT(7)	/* RW */
+#define		L_GAMMA_CNTL_PORT_RVS_OUT	BIT(6)	/* RW */
+#define		L_GAMMA_CNTL_PORT_ADR_RDY	BIT(5)	/* Read Only */
+#define		L_GAMMA_CNTL_PORT_WR_RDY	BIT(4)	/* Read Only */
+#define		L_GAMMA_CNTL_PORT_RD_RDY	BIT(3)	/* Read Only */
+#define		L_GAMMA_CNTL_PORT_TR		BIT(2)	/* RW */
+#define		L_GAMMA_CNTL_PORT_SET		BIT(1)	/* RW */
+#define		L_GAMMA_CNTL_PORT_EN		BIT(0)	/* RW */
 #define L_GAMMA_DATA_PORT 0x1401
 #define L_GAMMA_ADDR_PORT 0x1402
+#define		L_GAMMA_ADDR_PORT_RD		BIT(12)
+#define		L_GAMMA_ADDR_PORT_AUTO_INC	BIT(11)
+#define		L_GAMMA_ADDR_PORT_SEL_R		BIT(10)
+#define		L_GAMMA_ADDR_PORT_SEL_G		BIT(9)
+#define		L_GAMMA_ADDR_PORT_SEL_B		BIT(8)
+#define		L_GAMMA_ADDR_PORT_ADDR		GENMASK(7, 0)
 #define L_GAMMA_VCOM_HSWITCH_ADDR 0x1403
 #define L_RGB_BASE_ADDR 0x1405
 #define L_RGB_COEFF_ADDR 0x1406
 #define L_POL_CNTL_ADDR 0x1407
 #define L_DITH_CNTL_ADDR 0x1408
+#define		L_DITH_CNTL_DITH10_EN	BIT(10)
 #define L_GAMMA_PROBE_CTRL 0x1409
 #define L_GAMMA_PROBE_COLOR_L 0x140a
 #define L_GAMMA_PROBE_COLOR_H 0x140b
@@ -1363,6 +1386,8 @@
 #define L_LCD_PWM1_HI_ADDR 0x143f
 #define L_INV_CNT_ADDR 0x1440
 #define L_TCON_MISC_SEL_ADDR 0x1441
+#define		L_TCON_MISC_SEL_STV1	BIT(4)
+#define		L_TCON_MISC_SEL_STV2	BIT(5)
 #define L_DUAL_PORT_CNTL_ADDR 0x1442
 #define MLVDS_CLK_CTL1_HI 0x1443
 #define MLVDS_CLK_CTL1_LO 0x1444
diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index 27ef9f88e4ff8..2bdc2855e249b 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -6,6 +6,7 @@
  */
 
 #include <linux/export.h>
+#include <linux/iopoll.h>
 
 #include <drm/drm_modes.h>
 
@@ -1557,6 +1558,205 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic,
 }
 EXPORT_SYMBOL_GPL(meson_venc_hdmi_mode_set);
 
+static unsigned short meson_encl_gamma_table[256] = {
+	0, 4, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44, 48, 52, 56, 60,
+	64, 68, 72, 76, 80, 84, 88, 92, 96, 100, 104, 108, 112, 116, 120, 124,
+	128, 132, 136, 140, 144, 148, 152, 156, 160, 164, 168, 172, 176, 180, 184, 188,
+	192, 196, 200, 204, 208, 212, 216, 220, 224, 228, 232, 236, 240, 244, 248, 252,
+	256, 260, 264, 268, 272, 276, 280, 284, 288, 292, 296, 300, 304, 308, 312, 316,
+	320, 324, 328, 332, 336, 340, 344, 348, 352, 356, 360, 364, 368, 372, 376, 380,
+	384, 388, 392, 396, 400, 404, 408, 412, 416, 420, 424, 428, 432, 436, 440, 444,
+	448, 452, 456, 460, 464, 468, 472, 476, 480, 484, 488, 492, 496, 500, 504, 508,
+	512, 516, 520, 524, 528, 532, 536, 540, 544, 548, 552, 556, 560, 564, 568, 572,
+	576, 580, 584, 588, 592, 596, 600, 604, 608, 612, 616, 620, 624, 628, 632, 636,
+	640, 644, 648, 652, 656, 660, 664, 668, 672, 676, 680, 684, 688, 692, 696, 700,
+	704, 708, 712, 716, 720, 724, 728, 732, 736, 740, 744, 748, 752, 756, 760, 764,
+	768, 772, 776, 780, 784, 788, 792, 796, 800, 804, 808, 812, 816, 820, 824, 828,
+	832, 836, 840, 844, 848, 852, 856, 860, 864, 868, 872, 876, 880, 884, 888, 892,
+	896, 900, 904, 908, 912, 916, 920, 924, 928, 932, 936, 940, 944, 948, 952, 956,
+	960, 964, 968, 972, 976, 980, 984, 988, 992, 996, 1000, 1004, 1008, 1012, 1016, 1020,
+};
+
+static void meson_encl_set_gamma_table(struct meson_drm *priv, u16 *data,
+				       u32 rgb_mask)
+{
+	int i, ret;
+	u32 reg;
+
+	writel_bits_relaxed(L_GAMMA_CNTL_PORT_EN, 0,
+			    priv->io_base + _REG(L_GAMMA_CNTL_PORT));
+
+	ret = readl_relaxed_poll_timeout(priv->io_base + _REG(L_GAMMA_CNTL_PORT),
+					 reg, reg & L_GAMMA_CNTL_PORT_ADR_RDY, 10, 10000);
+	if (ret)
+		pr_warn("%s: GAMMA ADR_RDY timeout\n", __func__);
+
+	writel_relaxed(L_GAMMA_ADDR_PORT_AUTO_INC | rgb_mask |
+		       FIELD_PREP(L_GAMMA_ADDR_PORT_ADDR, 0),
+		       priv->io_base + _REG(L_GAMMA_ADDR_PORT));
+
+	for (i = 0; i < 256; i++) {
+		ret = readl_relaxed_poll_timeout(priv->io_base + _REG(L_GAMMA_CNTL_PORT),
+						 reg, reg & L_GAMMA_CNTL_PORT_WR_RDY,
+						 10, 10000);
+		if (ret)
+			pr_warn_once("%s: GAMMA WR_RDY timeout\n", __func__);
+
+		writel_relaxed(data[i], priv->io_base + _REG(L_GAMMA_DATA_PORT));
+	}
+
+	ret = readl_relaxed_poll_timeout(priv->io_base + _REG(L_GAMMA_CNTL_PORT),
+					 reg, reg & L_GAMMA_CNTL_PORT_ADR_RDY, 10, 10000);
+	if (ret)
+		pr_warn("%s: GAMMA ADR_RDY timeout\n", __func__);
+
+	writel_relaxed(L_GAMMA_ADDR_PORT_AUTO_INC | rgb_mask |
+		       FIELD_PREP(L_GAMMA_ADDR_PORT_ADDR, 0x23),
+		       priv->io_base + _REG(L_GAMMA_ADDR_PORT));
+}
+
+void meson_encl_load_gamma(struct meson_drm *priv)
+{
+	meson_encl_set_gamma_table(priv, meson_encl_gamma_table, L_GAMMA_ADDR_PORT_SEL_R);
+	meson_encl_set_gamma_table(priv, meson_encl_gamma_table, L_GAMMA_ADDR_PORT_SEL_G);
+	meson_encl_set_gamma_table(priv, meson_encl_gamma_table, L_GAMMA_ADDR_PORT_SEL_B);
+
+	writel_bits_relaxed(L_GAMMA_CNTL_PORT_EN, L_GAMMA_CNTL_PORT_EN,
+			    priv->io_base + _REG(L_GAMMA_CNTL_PORT));
+}
+
+void meson_venc_mipi_dsi_mode_set(struct meson_drm *priv,
+				  const struct drm_display_mode *mode)
+{
+	unsigned int max_pxcnt;
+	unsigned int max_lncnt;
+	unsigned int havon_begin;
+	unsigned int havon_end;
+	unsigned int vavon_bline;
+	unsigned int vavon_eline;
+	unsigned int hso_begin;
+	unsigned int hso_end;
+	unsigned int vso_begin;
+	unsigned int vso_end;
+	unsigned int vso_bline;
+	unsigned int vso_eline;
+
+	max_pxcnt = mode->htotal - 1;
+	max_lncnt = mode->vtotal - 1;
+	havon_begin = mode->htotal - mode->hsync_start;
+	havon_end = havon_begin + mode->hdisplay - 1;
+	vavon_bline = mode->vtotal - mode->vsync_start;
+	vavon_eline = vavon_bline + mode->vdisplay - 1;
+	hso_begin = 0;
+	hso_end = mode->hsync_end - mode->hsync_start;
+	vso_begin = 0;
+	vso_end = 0;
+	vso_bline = 0;
+	vso_eline = mode->vsync_end - mode->vsync_start;
+
+	meson_vpp_setup_mux(priv, MESON_VIU_VPP_MUX_ENCL);
+
+	writel_relaxed(0, priv->io_base + _REG(ENCL_VIDEO_EN));
+
+	writel_relaxed(ENCL_PX_LN_CNT_SHADOW_EN, priv->io_base + _REG(ENCL_VIDEO_MODE));
+	writel_relaxed(ENCL_VIDEO_MODE_ADV_VFIFO_EN |
+		       ENCL_VIDEO_MODE_ADV_GAIN_HDTV |
+		       ENCL_SEL_GAMMA_RGB_IN, priv->io_base + _REG(ENCL_VIDEO_MODE_ADV));
+
+	writel_relaxed(ENCL_VIDEO_FILT_CTRL_BYPASS_FILTER,
+		       priv->io_base + _REG(ENCL_VIDEO_FILT_CTRL));
+	writel_relaxed(max_pxcnt, priv->io_base + _REG(ENCL_VIDEO_MAX_PXCNT));
+	writel_relaxed(max_lncnt, priv->io_base + _REG(ENCL_VIDEO_MAX_LNCNT));
+	writel_relaxed(havon_begin, priv->io_base + _REG(ENCL_VIDEO_HAVON_BEGIN));
+	writel_relaxed(havon_end, priv->io_base + _REG(ENCL_VIDEO_HAVON_END));
+	writel_relaxed(vavon_bline, priv->io_base + _REG(ENCL_VIDEO_VAVON_BLINE));
+	writel_relaxed(vavon_eline, priv->io_base + _REG(ENCL_VIDEO_VAVON_ELINE));
+
+	writel_relaxed(hso_begin, priv->io_base + _REG(ENCL_VIDEO_HSO_BEGIN));
+	writel_relaxed(hso_end, priv->io_base + _REG(ENCL_VIDEO_HSO_END));
+	writel_relaxed(vso_begin, priv->io_base + _REG(ENCL_VIDEO_VSO_BEGIN));
+	writel_relaxed(vso_end, priv->io_base + _REG(ENCL_VIDEO_VSO_END));
+	writel_relaxed(vso_bline, priv->io_base + _REG(ENCL_VIDEO_VSO_BLINE));
+	writel_relaxed(vso_eline, priv->io_base + _REG(ENCL_VIDEO_VSO_ELINE));
+	writel_relaxed(ENCL_VIDEO_RGBIN_RGB | ENCL_VIDEO_RGBIN_ZBLK,
+		       priv->io_base + _REG(ENCL_VIDEO_RGBIN_CTRL));
+
+	/* default black pattern */
+	writel_relaxed(0, priv->io_base + _REG(ENCL_TST_MDSEL));
+	writel_relaxed(0, priv->io_base + _REG(ENCL_TST_Y));
+	writel_relaxed(0, priv->io_base + _REG(ENCL_TST_CB));
+	writel_relaxed(0, priv->io_base + _REG(ENCL_TST_CR));
+	writel_relaxed(1, priv->io_base + _REG(ENCL_TST_EN));
+	writel_bits_relaxed(ENCL_VIDEO_MODE_ADV_VFIFO_EN, 0,
+			    priv->io_base + _REG(ENCL_VIDEO_MODE_ADV));
+
+	writel_relaxed(1, priv->io_base + _REG(ENCL_VIDEO_EN));
+
+	writel_relaxed(0, priv->io_base + _REG(L_RGB_BASE_ADDR));
+	writel_relaxed(0x400, priv->io_base + _REG(L_RGB_COEFF_ADDR)); /* Magic value */
+
+	writel_relaxed(L_DITH_CNTL_DITH10_EN, priv->io_base + _REG(L_DITH_CNTL_ADDR));
+
+	/* DE signal for TTL */
+	writel_relaxed(havon_begin, priv->io_base + _REG(L_OEH_HS_ADDR));
+	writel_relaxed(havon_end + 1, priv->io_base + _REG(L_OEH_HE_ADDR));
+	writel_relaxed(vavon_bline, priv->io_base + _REG(L_OEH_VS_ADDR));
+	writel_relaxed(vavon_eline, priv->io_base + _REG(L_OEH_VE_ADDR));
+
+	/* DE signal for TTL */
+	writel_relaxed(havon_begin, priv->io_base + _REG(L_OEV1_HS_ADDR));
+	writel_relaxed(havon_end + 1, priv->io_base + _REG(L_OEV1_HE_ADDR));
+	writel_relaxed(vavon_bline, priv->io_base + _REG(L_OEV1_VS_ADDR));
+	writel_relaxed(vavon_eline, priv->io_base + _REG(L_OEV1_VE_ADDR));
+
+	/* Hsync signal for TTL */
+	if (mode->flags & DRM_MODE_FLAG_PHSYNC) {
+		writel_relaxed(hso_begin, priv->io_base + _REG(L_STH1_HS_ADDR));
+		writel_relaxed(hso_end, priv->io_base + _REG(L_STH1_HE_ADDR));
+	} else {
+		writel_relaxed(hso_end, priv->io_base + _REG(L_STH1_HS_ADDR));
+		writel_relaxed(hso_begin, priv->io_base + _REG(L_STH1_HE_ADDR));
+	}
+	writel_relaxed(0, priv->io_base + _REG(L_STH1_VS_ADDR));
+	writel_relaxed(max_lncnt, priv->io_base + _REG(L_STH1_VE_ADDR));
+
+	/* Vsync signal for TTL */
+	writel_relaxed(vso_begin, priv->io_base + _REG(L_STV1_HS_ADDR));
+	writel_relaxed(vso_end, priv->io_base + _REG(L_STV1_HE_ADDR));
+	if (mode->flags & DRM_MODE_FLAG_PVSYNC) {
+		writel_relaxed(vso_bline, priv->io_base + _REG(L_STV1_VS_ADDR));
+		writel_relaxed(vso_eline, priv->io_base + _REG(L_STV1_VE_ADDR));
+	} else {
+		writel_relaxed(vso_eline, priv->io_base + _REG(L_STV1_VS_ADDR));
+		writel_relaxed(vso_bline, priv->io_base + _REG(L_STV1_VE_ADDR));
+	}
+
+	/* DE signal */
+	writel_relaxed(havon_begin, priv->io_base + _REG(L_DE_HS_ADDR));
+	writel_relaxed(havon_end + 1, priv->io_base + _REG(L_DE_HE_ADDR));
+	writel_relaxed(vavon_bline, priv->io_base + _REG(L_DE_VS_ADDR));
+	writel_relaxed(vavon_eline, priv->io_base + _REG(L_DE_VE_ADDR));
+
+	/* Hsync signal */
+	writel_relaxed(hso_begin, priv->io_base + _REG(L_HSYNC_HS_ADDR));
+	writel_relaxed(hso_end, priv->io_base + _REG(L_HSYNC_HE_ADDR));
+	writel_relaxed(0, priv->io_base + _REG(L_HSYNC_VS_ADDR));
+	writel_relaxed(max_lncnt, priv->io_base + _REG(L_HSYNC_VE_ADDR));
+
+	/* Vsync signal */
+	writel_relaxed(vso_begin, priv->io_base + _REG(L_VSYNC_HS_ADDR));
+	writel_relaxed(vso_end, priv->io_base + _REG(L_VSYNC_HE_ADDR));
+	writel_relaxed(vso_bline, priv->io_base + _REG(L_VSYNC_VS_ADDR));
+	writel_relaxed(vso_eline, priv->io_base + _REG(L_VSYNC_VE_ADDR));
+
+	writel_relaxed(0, priv->io_base + _REG(L_INV_CNT_ADDR));
+	writel_relaxed(L_TCON_MISC_SEL_STV1 | L_TCON_MISC_SEL_STV2,
+		       priv->io_base + _REG(L_TCON_MISC_SEL_ADDR));
+
+	priv->venc.current_mode = MESON_VENC_MODE_MIPI_DSI;
+}
+EXPORT_SYMBOL_GPL(meson_venc_mipi_dsi_mode_set);
+
 void meson_venci_cvbs_mode_set(struct meson_drm *priv,
 			       struct meson_cvbs_enci_mode *mode)
 {
@@ -1747,8 +1947,15 @@ unsigned int meson_venci_get_field(struct meson_drm *priv)
 
 void meson_venc_enable_vsync(struct meson_drm *priv)
 {
-	writel_relaxed(VENC_INTCTRL_ENCI_LNRST_INT_EN,
-		       priv->io_base + _REG(VENC_INTCTRL));
+	switch (priv->venc.current_mode) {
+	case MESON_VENC_MODE_MIPI_DSI:
+		writel_relaxed(VENC_INTCTRL_ENCP_LNRST_INT_EN,
+			       priv->io_base + _REG(VENC_INTCTRL));
+		break;
+	default:
+		writel_relaxed(VENC_INTCTRL_ENCI_LNRST_INT_EN,
+			       priv->io_base + _REG(VENC_INTCTRL));
+	}
 	regmap_update_bits(priv->hhi, HHI_GCLK_MPEG2, BIT(25), BIT(25));
 }
 
diff --git a/drivers/gpu/drm/meson/meson_venc.h b/drivers/gpu/drm/meson/meson_venc.h
index 9138255ffc9ee..0f59adb1c6db0 100644
--- a/drivers/gpu/drm/meson/meson_venc.h
+++ b/drivers/gpu/drm/meson/meson_venc.h
@@ -21,6 +21,7 @@ enum {
 	MESON_VENC_MODE_CVBS_PAL,
 	MESON_VENC_MODE_CVBS_NTSC,
 	MESON_VENC_MODE_HDMI,
+	MESON_VENC_MODE_MIPI_DSI,
 };
 
 struct meson_cvbs_enci_mode {
@@ -47,6 +48,9 @@ struct meson_cvbs_enci_mode {
 	unsigned int analog_sync_adj;
 };
 
+/* LCD Encoder gamma setup */
+void meson_encl_load_gamma(struct meson_drm *priv);
+
 /* HDMI Clock parameters */
 enum drm_mode_status
 meson_venc_hdmi_supported_mode(const struct drm_display_mode *mode);
@@ -63,6 +67,8 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic,
 			      unsigned int ycrcb_map,
 			      bool yuv420_mode,
 			      const struct drm_display_mode *mode);
+void meson_venc_mipi_dsi_mode_set(struct meson_drm *priv,
+				  const struct drm_display_mode *mode);
 unsigned int meson_venci_get_field(struct meson_drm *priv);
 
 void meson_venc_enable_vsync(struct meson_drm *priv);
diff --git a/drivers/gpu/drm/meson/meson_vpp.h b/drivers/gpu/drm/meson/meson_vpp.h
index afc9553ed8d38..b790042a1650a 100644
--- a/drivers/gpu/drm/meson/meson_vpp.h
+++ b/drivers/gpu/drm/meson/meson_vpp.h
@@ -12,6 +12,8 @@
 struct drm_rect;
 struct meson_drm;
 
+/* Mux VIU/VPP to ENCL */
+#define MESON_VIU_VPP_MUX_ENCL	0x0
 /* Mux VIU/VPP to ENCI */
 #define MESON_VIU_VPP_MUX_ENCI	0x5
 /* Mux VIU/VPP to ENCP */
-- 
GitLab


From 42dcf15f901c8222352da31d622b4ee844068f42 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:12 +0200
Subject: [PATCH 0724/1791] drm/meson: add DSI encoder

This adds an encoder bridge designed to drive a MIPI-DSI display
by using the ENCL encoder through the internal MIPI DSI transceiver
connected to the output of the ENCL pixel encoder.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com> # on Khadas VIM3 + TS050 Panel
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-11-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/meson/Makefile            |   2 +-
 drivers/gpu/drm/meson/meson_drv.c         |   9 ++
 drivers/gpu/drm/meson/meson_drv.h         |   1 +
 drivers/gpu/drm/meson/meson_encoder_dsi.c | 174 ++++++++++++++++++++++
 drivers/gpu/drm/meson/meson_encoder_dsi.h |  13 ++
 5 files changed, 198 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/meson/meson_encoder_dsi.c
 create mode 100644 drivers/gpu/drm/meson/meson_encoder_dsi.h

diff --git a/drivers/gpu/drm/meson/Makefile b/drivers/gpu/drm/meson/Makefile
index 3afa31bdc9501..833e18c206038 100644
--- a/drivers/gpu/drm/meson/Makefile
+++ b/drivers/gpu/drm/meson/Makefile
@@ -2,7 +2,7 @@
 meson-drm-y := meson_drv.o meson_plane.o meson_crtc.o meson_encoder_cvbs.o
 meson-drm-y += meson_viu.o meson_vpp.o meson_venc.o meson_vclk.o meson_overlay.o
 meson-drm-y += meson_rdma.o meson_osd_afbcd.o
-meson-drm-y += meson_encoder_hdmi.o
+meson-drm-y += meson_encoder_hdmi.o meson_encoder_dsi.o
 
 obj-$(CONFIG_DRM_MESON) += meson-drm.o
 obj-$(CONFIG_DRM_MESON_DW_HDMI) += meson_dw_hdmi.o
diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c
index e935c0286a201..747b639ea0c47 100644
--- a/drivers/gpu/drm/meson/meson_drv.c
+++ b/drivers/gpu/drm/meson/meson_drv.c
@@ -34,6 +34,7 @@
 #include "meson_registers.h"
 #include "meson_encoder_cvbs.h"
 #include "meson_encoder_hdmi.h"
+#include "meson_encoder_dsi.h"
 #include "meson_viu.h"
 #include "meson_vpp.h"
 #include "meson_rdma.h"
@@ -329,6 +330,12 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 	if (ret)
 		goto exit_afbcd;
 
+	if (meson_vpu_is_compatible(priv, VPU_COMPATIBLE_G12A)) {
+		ret = meson_encoder_dsi_init(priv);
+		if (ret)
+			goto exit_afbcd;
+	}
+
 	ret = meson_plane_create(priv);
 	if (ret)
 		goto exit_afbcd;
@@ -367,6 +374,7 @@ static int meson_drv_bind_master(struct device *dev, bool has_components)
 free_drm:
 	drm_dev_put(drm);
 
+	meson_encoder_dsi_remove(priv);
 	meson_encoder_hdmi_remove(priv);
 	meson_encoder_cvbs_remove(priv);
 
@@ -399,6 +407,7 @@ static void meson_drv_unbind(struct device *dev)
 	free_irq(priv->vsync_irq, drm);
 	drm_dev_put(drm);
 
+	meson_encoder_dsi_remove(priv);
 	meson_encoder_hdmi_remove(priv);
 	meson_encoder_cvbs_remove(priv);
 
diff --git a/drivers/gpu/drm/meson/meson_drv.h b/drivers/gpu/drm/meson/meson_drv.h
index c62ee358456fa..b23009a3380fb 100644
--- a/drivers/gpu/drm/meson/meson_drv.h
+++ b/drivers/gpu/drm/meson/meson_drv.h
@@ -28,6 +28,7 @@ enum vpu_compatible {
 enum {
 	MESON_ENC_CVBS = 0,
 	MESON_ENC_HDMI,
+	MESON_ENC_DSI,
 	MESON_ENC_LAST,
 };
 
diff --git a/drivers/gpu/drm/meson/meson_encoder_dsi.c b/drivers/gpu/drm/meson/meson_encoder_dsi.c
new file mode 100644
index 0000000000000..812e172dec63a
--- /dev/null
+++ b/drivers/gpu/drm/meson/meson_encoder_dsi.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2016 BayLibre, SAS
+ * Author: Neil Armstrong <narmstrong@baylibre.com>
+ * Copyright (C) 2015 Amlogic, Inc. All rights reserved.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/of_graph.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_simple_kms_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/drm_device.h>
+#include <drm/drm_probe_helper.h>
+
+#include "meson_drv.h"
+#include "meson_encoder_dsi.h"
+#include "meson_registers.h"
+#include "meson_venc.h"
+#include "meson_vclk.h"
+
+struct meson_encoder_dsi {
+	struct drm_encoder encoder;
+	struct drm_bridge bridge;
+	struct drm_bridge *next_bridge;
+	struct meson_drm *priv;
+};
+
+#define bridge_to_meson_encoder_dsi(x) \
+	container_of(x, struct meson_encoder_dsi, bridge)
+
+static int meson_encoder_dsi_attach(struct drm_bridge *bridge,
+				    enum drm_bridge_attach_flags flags)
+{
+	struct meson_encoder_dsi *encoder_dsi = bridge_to_meson_encoder_dsi(bridge);
+
+	return drm_bridge_attach(bridge->encoder, encoder_dsi->next_bridge,
+				 &encoder_dsi->bridge, flags);
+}
+
+static void meson_encoder_dsi_atomic_enable(struct drm_bridge *bridge,
+					    struct drm_bridge_state *bridge_state)
+{
+	struct meson_encoder_dsi *encoder_dsi = bridge_to_meson_encoder_dsi(bridge);
+	struct drm_atomic_state *state = bridge_state->base.state;
+	struct meson_drm *priv = encoder_dsi->priv;
+	struct drm_connector_state *conn_state;
+	struct drm_crtc_state *crtc_state;
+	struct drm_connector *connector;
+
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	if (WARN_ON(!connector))
+		return;
+
+	conn_state = drm_atomic_get_new_connector_state(state, connector);
+	if (WARN_ON(!conn_state))
+		return;
+
+	crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc);
+	if (WARN_ON(!crtc_state))
+		return;
+
+	/* ENCL clock setup is handled by CCF */
+
+	meson_venc_mipi_dsi_mode_set(priv, &crtc_state->adjusted_mode);
+	meson_encl_load_gamma(priv);
+
+	writel_relaxed(0, priv->io_base + _REG(ENCL_VIDEO_EN));
+
+	writel_bits_relaxed(ENCL_VIDEO_MODE_ADV_VFIFO_EN, ENCL_VIDEO_MODE_ADV_VFIFO_EN,
+			    priv->io_base + _REG(ENCL_VIDEO_MODE_ADV));
+	writel_relaxed(0, priv->io_base + _REG(ENCL_TST_EN));
+
+	writel_bits_relaxed(BIT(0), 0, priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_EN_CTRL));
+
+	writel_relaxed(1, priv->io_base + _REG(ENCL_VIDEO_EN));
+}
+
+static void meson_encoder_dsi_atomic_disable(struct drm_bridge *bridge,
+					     struct drm_bridge_state *bridge_state)
+{
+	struct meson_encoder_dsi *meson_encoder_dsi =
+					bridge_to_meson_encoder_dsi(bridge);
+	struct meson_drm *priv = meson_encoder_dsi->priv;
+
+	writel_relaxed(0, priv->io_base + _REG(ENCL_VIDEO_EN));
+
+	writel_bits_relaxed(BIT(0), BIT(0), priv->io_base + _REG(VPP_WRAP_OSD1_MATRIX_EN_CTRL));
+}
+
+static const struct drm_bridge_funcs meson_encoder_dsi_bridge_funcs = {
+	.attach	= meson_encoder_dsi_attach,
+	.atomic_enable = meson_encoder_dsi_atomic_enable,
+	.atomic_disable	= meson_encoder_dsi_atomic_disable,
+	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+};
+
+int meson_encoder_dsi_init(struct meson_drm *priv)
+{
+	struct meson_encoder_dsi *meson_encoder_dsi;
+	struct device_node *remote;
+	int ret;
+
+	meson_encoder_dsi = devm_kzalloc(priv->dev, sizeof(*meson_encoder_dsi), GFP_KERNEL);
+	if (!meson_encoder_dsi)
+		return -ENOMEM;
+
+	/* DSI Transceiver Bridge */
+	remote = of_graph_get_remote_node(priv->dev->of_node, 2, 0);
+	if (!remote) {
+		dev_err(priv->dev, "DSI transceiver device is disabled");
+		return 0;
+	}
+
+	meson_encoder_dsi->next_bridge = of_drm_find_bridge(remote);
+	if (!meson_encoder_dsi->next_bridge) {
+		dev_dbg(priv->dev, "Failed to find DSI transceiver bridge\n");
+		return -EPROBE_DEFER;
+	}
+
+	/* DSI Encoder Bridge */
+	meson_encoder_dsi->bridge.funcs = &meson_encoder_dsi_bridge_funcs;
+	meson_encoder_dsi->bridge.of_node = priv->dev->of_node;
+	meson_encoder_dsi->bridge.type = DRM_MODE_CONNECTOR_DSI;
+
+	drm_bridge_add(&meson_encoder_dsi->bridge);
+
+	meson_encoder_dsi->priv = priv;
+
+	/* Encoder */
+	ret = drm_simple_encoder_init(priv->drm, &meson_encoder_dsi->encoder,
+				      DRM_MODE_ENCODER_DSI);
+	if (ret) {
+		dev_err(priv->dev, "Failed to init DSI encoder: %d\n", ret);
+		return ret;
+	}
+
+	meson_encoder_dsi->encoder.possible_crtcs = BIT(0);
+
+	/* Attach DSI Encoder Bridge to Encoder */
+	ret = drm_bridge_attach(&meson_encoder_dsi->encoder, &meson_encoder_dsi->bridge, NULL, 0);
+	if (ret) {
+		dev_err(priv->dev, "Failed to attach bridge: %d\n", ret);
+		return ret;
+	}
+
+	/*
+	 * We should have now in place:
+	 * encoder->[dsi encoder bridge]->[dw-mipi-dsi bridge]->[panel bridge]->[panel]
+	 */
+
+	priv->encoders[MESON_ENC_DSI] = meson_encoder_dsi;
+
+	dev_dbg(priv->dev, "DSI encoder initialized\n");
+
+	return 0;
+}
+
+void meson_encoder_dsi_remove(struct meson_drm *priv)
+{
+	struct meson_encoder_dsi *meson_encoder_dsi;
+
+	if (priv->encoders[MESON_ENC_DSI]) {
+		meson_encoder_dsi = priv->encoders[MESON_ENC_DSI];
+		drm_bridge_remove(&meson_encoder_dsi->bridge);
+		drm_bridge_remove(meson_encoder_dsi->next_bridge);
+	}
+}
diff --git a/drivers/gpu/drm/meson/meson_encoder_dsi.h b/drivers/gpu/drm/meson/meson_encoder_dsi.h
new file mode 100644
index 0000000000000..9277d70151939
--- /dev/null
+++ b/drivers/gpu/drm/meson/meson_encoder_dsi.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2021 BayLibre, SAS
+ * Author: Neil Armstrong <narmstrong@baylibre.com>
+ */
+
+#ifndef __MESON_ENCODER_DSI_H
+#define __MESON_ENCODER_DSI_H
+
+int meson_encoder_dsi_init(struct meson_drm *priv);
+void meson_encoder_dsi_remove(struct meson_drm *priv);
+
+#endif /* __MESON_ENCODER_DSI_H */
-- 
GitLab


From 77d9e1e6b8468f701ab024a060aa9c0339356870 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:13 +0200
Subject: [PATCH 0725/1791] drm/meson: add support for MIPI-DSI transceiver

The Amlogic G12A/G12B/SM1 SoCs embeds a Synopsys DW-MIPI-DSI transceiver
(ver 1.21a), with a custom glue managing the IP resets, clock and data
inputs similar to the DW-HDMI Glue on other Amlogic SoCs.

This adds support for the Glue managing the transceiver, mimicing the init
flow provided by Amlogic to setup the ENCL encoder, the glue, the transceiver,
the digital D-PHY and the Analog PHY in the proper way.

An optional "MEAS" clock can be enabled to measure the delay between each
vsync feeding the DW-MIPI-DSI transceiver.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com> # on Khadas VIM3 + TS050 Panel
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-12-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/meson/Kconfig             |   7 +
 drivers/gpu/drm/meson/Makefile            |   1 +
 drivers/gpu/drm/meson/meson_dw_mipi_dsi.c | 352 ++++++++++++++++++++++
 drivers/gpu/drm/meson/meson_dw_mipi_dsi.h | 160 ++++++++++
 4 files changed, 520 insertions(+)
 create mode 100644 drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
 create mode 100644 drivers/gpu/drm/meson/meson_dw_mipi_dsi.h

diff --git a/drivers/gpu/drm/meson/Kconfig b/drivers/gpu/drm/meson/Kconfig
index 823909da87dbf..615fdd0ce41b4 100644
--- a/drivers/gpu/drm/meson/Kconfig
+++ b/drivers/gpu/drm/meson/Kconfig
@@ -17,3 +17,10 @@ config DRM_MESON_DW_HDMI
 	default y if DRM_MESON
 	select DRM_DW_HDMI
 	imply DRM_DW_HDMI_I2S_AUDIO
+
+config DRM_MESON_DW_MIPI_DSI
+	tristate "MIPI DSI Synopsys Controller support for Amlogic Meson Display"
+	depends on DRM_MESON
+	default y if DRM_MESON
+	select DRM_DW_MIPI_DSI
+	select GENERIC_PHY_MIPI_DPHY
diff --git a/drivers/gpu/drm/meson/Makefile b/drivers/gpu/drm/meson/Makefile
index 833e18c206038..43071bdbd4b99 100644
--- a/drivers/gpu/drm/meson/Makefile
+++ b/drivers/gpu/drm/meson/Makefile
@@ -6,3 +6,4 @@ meson-drm-y += meson_encoder_hdmi.o meson_encoder_dsi.o
 
 obj-$(CONFIG_DRM_MESON) += meson-drm.o
 obj-$(CONFIG_DRM_MESON_DW_HDMI) += meson_dw_hdmi.o
+obj-$(CONFIG_DRM_MESON_DW_MIPI_DSI) += meson_dw_mipi_dsi.o
diff --git a/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
new file mode 100644
index 0000000000000..dd505ac379767
--- /dev/null
+++ b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
@@ -0,0 +1,352 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2021 BayLibre, SAS
+ * Author: Neil Armstrong <narmstrong@baylibre.com>
+ * Copyright (C) 2015 Amlogic, Inc. All rights reserved.
+ */
+
+#include <linux/clk.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/of_graph.h>
+#include <linux/reset.h>
+#include <linux/phy/phy.h>
+#include <linux/bitfield.h>
+
+#include <video/mipi_display.h>
+
+#include <drm/bridge/dw_mipi_dsi.h>
+#include <drm/drm_mipi_dsi.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_device.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_print.h>
+
+#include "meson_drv.h"
+#include "meson_dw_mipi_dsi.h"
+#include "meson_registers.h"
+#include "meson_venc.h"
+
+#define DRIVER_NAME "meson-dw-mipi-dsi"
+#define DRIVER_DESC "Amlogic Meson MIPI-DSI DRM driver"
+
+struct meson_dw_mipi_dsi {
+	struct meson_drm *priv;
+	struct device *dev;
+	void __iomem *base;
+	struct phy *phy;
+	union phy_configure_opts phy_opts;
+	struct dw_mipi_dsi *dmd;
+	struct dw_mipi_dsi_plat_data pdata;
+	struct mipi_dsi_device *dsi_device;
+	const struct drm_display_mode *mode;
+	struct clk *bit_clk;
+	struct clk *px_clk;
+	struct reset_control *top_rst;
+};
+
+#define encoder_to_meson_dw_mipi_dsi(x) \
+	container_of(x, struct meson_dw_mipi_dsi, encoder)
+
+static void meson_dw_mipi_dsi_hw_init(struct meson_dw_mipi_dsi *mipi_dsi)
+{
+	/* Software reset */
+	writel_bits_relaxed(MIPI_DSI_TOP_SW_RESET_DWC | MIPI_DSI_TOP_SW_RESET_INTR |
+			    MIPI_DSI_TOP_SW_RESET_DPI | MIPI_DSI_TOP_SW_RESET_TIMING,
+			    MIPI_DSI_TOP_SW_RESET_DWC | MIPI_DSI_TOP_SW_RESET_INTR |
+			    MIPI_DSI_TOP_SW_RESET_DPI | MIPI_DSI_TOP_SW_RESET_TIMING,
+			    mipi_dsi->base + MIPI_DSI_TOP_SW_RESET);
+	writel_bits_relaxed(MIPI_DSI_TOP_SW_RESET_DWC | MIPI_DSI_TOP_SW_RESET_INTR |
+			    MIPI_DSI_TOP_SW_RESET_DPI | MIPI_DSI_TOP_SW_RESET_TIMING,
+			    0, mipi_dsi->base + MIPI_DSI_TOP_SW_RESET);
+
+	/* Enable clocks */
+	writel_bits_relaxed(MIPI_DSI_TOP_CLK_SYSCLK_EN | MIPI_DSI_TOP_CLK_PIXCLK_EN,
+			    MIPI_DSI_TOP_CLK_SYSCLK_EN | MIPI_DSI_TOP_CLK_PIXCLK_EN,
+			    mipi_dsi->base + MIPI_DSI_TOP_CLK_CNTL);
+
+	/* Take memory out of power down */
+	writel_relaxed(0, mipi_dsi->base + MIPI_DSI_TOP_MEM_PD);
+}
+
+static int dw_mipi_dsi_phy_init(void *priv_data)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+	unsigned int dpi_data_format, venc_data_width;
+	int ret;
+
+	/* Set the bit clock rate to hs_clk_rate */
+	ret = clk_set_rate(mipi_dsi->bit_clk,
+			   mipi_dsi->phy_opts.mipi_dphy.hs_clk_rate);
+	if (ret) {
+		dev_err(mipi_dsi->dev, "Failed to set DSI Bit clock rate %lu (ret %d)\n",
+			mipi_dsi->phy_opts.mipi_dphy.hs_clk_rate, ret);
+		return ret;
+	}
+
+	/* Make sure the rate of the bit clock is not modified by someone else */
+	ret = clk_rate_exclusive_get(mipi_dsi->bit_clk);
+	if (ret) {
+		dev_err(mipi_dsi->dev,
+			"Failed to set the exclusivity on the bit clock rate (ret %d)\n", ret);
+		return ret;
+	}
+
+	ret = clk_set_rate(mipi_dsi->px_clk, mipi_dsi->mode->clock * 1000);
+
+	if (ret) {
+		dev_err(mipi_dsi->dev, "Failed to set DSI Pixel clock rate %u (%d)\n",
+			mipi_dsi->mode->clock * 1000, ret);
+		return ret;
+	}
+
+	switch (mipi_dsi->dsi_device->format) {
+	case MIPI_DSI_FMT_RGB888:
+		dpi_data_format = DPI_COLOR_24BIT;
+		venc_data_width = VENC_IN_COLOR_24B;
+		break;
+	case MIPI_DSI_FMT_RGB666:
+		dpi_data_format = DPI_COLOR_18BIT_CFG_2;
+		venc_data_width = VENC_IN_COLOR_18B;
+		break;
+	case MIPI_DSI_FMT_RGB666_PACKED:
+	case MIPI_DSI_FMT_RGB565:
+		return -EINVAL;
+	};
+
+	/* Configure color format for DPI register */
+	writel_relaxed(FIELD_PREP(MIPI_DSI_TOP_DPI_COLOR_MODE, dpi_data_format) |
+		       FIELD_PREP(MIPI_DSI_TOP_IN_COLOR_MODE, venc_data_width) |
+		       FIELD_PREP(MIPI_DSI_TOP_COMP2_SEL, 2) |
+		       FIELD_PREP(MIPI_DSI_TOP_COMP1_SEL, 1) |
+		       FIELD_PREP(MIPI_DSI_TOP_COMP0_SEL, 0),
+			mipi_dsi->base + MIPI_DSI_TOP_CNTL);
+
+	return phy_configure(mipi_dsi->phy, &mipi_dsi->phy_opts);
+}
+
+static void dw_mipi_dsi_phy_power_on(void *priv_data)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+
+	if (phy_power_on(mipi_dsi->phy))
+		dev_warn(mipi_dsi->dev, "Failed to power on PHY\n");
+}
+
+static void dw_mipi_dsi_phy_power_off(void *priv_data)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+
+	if (phy_power_off(mipi_dsi->phy))
+		dev_warn(mipi_dsi->dev, "Failed to power off PHY\n");
+
+	/* Remove the exclusivity on the bit clock rate */
+	clk_rate_exclusive_put(mipi_dsi->bit_clk);
+}
+
+static int
+dw_mipi_dsi_get_lane_mbps(void *priv_data, const struct drm_display_mode *mode,
+			  unsigned long mode_flags, u32 lanes, u32 format,
+			  unsigned int *lane_mbps)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+	int bpp;
+
+	mipi_dsi->mode = mode;
+
+	bpp = mipi_dsi_pixel_format_to_bpp(mipi_dsi->dsi_device->format);
+
+	phy_mipi_dphy_get_default_config(mode->clock * 1000,
+					 bpp, mipi_dsi->dsi_device->lanes,
+					 &mipi_dsi->phy_opts.mipi_dphy);
+
+	*lane_mbps = DIV_ROUND_UP(mipi_dsi->phy_opts.mipi_dphy.hs_clk_rate, USEC_PER_SEC);
+
+	return 0;
+}
+
+static int
+dw_mipi_dsi_phy_get_timing(void *priv_data, unsigned int lane_mbps,
+			   struct dw_mipi_dsi_dphy_timing *timing)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+
+	switch (mipi_dsi->mode->hdisplay) {
+	case 240:
+	case 768:
+	case 1920:
+	case 2560:
+		timing->clk_lp2hs = 23;
+		timing->clk_hs2lp = 38;
+		timing->data_lp2hs = 15;
+		timing->data_hs2lp = 9;
+		break;
+
+	default:
+		timing->clk_lp2hs = 37;
+		timing->clk_hs2lp = 135;
+		timing->data_lp2hs = 50;
+		timing->data_hs2lp = 3;
+	}
+
+	return 0;
+}
+
+static int
+dw_mipi_dsi_get_esc_clk_rate(void *priv_data, unsigned int *esc_clk_rate)
+{
+	*esc_clk_rate = 4; /* Mhz */
+
+	return 0;
+}
+
+static const struct dw_mipi_dsi_phy_ops meson_dw_mipi_dsi_phy_ops = {
+	.init = dw_mipi_dsi_phy_init,
+	.power_on = dw_mipi_dsi_phy_power_on,
+	.power_off = dw_mipi_dsi_phy_power_off,
+	.get_lane_mbps = dw_mipi_dsi_get_lane_mbps,
+	.get_timing = dw_mipi_dsi_phy_get_timing,
+	.get_esc_clk_rate = dw_mipi_dsi_get_esc_clk_rate,
+};
+
+static int meson_dw_mipi_dsi_host_attach(void *priv_data,
+					 struct mipi_dsi_device *device)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+	int ret;
+
+	mipi_dsi->dsi_device = device;
+
+	switch (device->format) {
+	case MIPI_DSI_FMT_RGB888:
+		break;
+	case MIPI_DSI_FMT_RGB666:
+		break;
+	case MIPI_DSI_FMT_RGB666_PACKED:
+	case MIPI_DSI_FMT_RGB565:
+		dev_err(mipi_dsi->dev, "invalid pixel format %d\n", device->format);
+		return -EINVAL;
+	};
+
+	ret = phy_init(mipi_dsi->phy);
+	if (ret)
+		return ret;
+
+	meson_dw_mipi_dsi_hw_init(mipi_dsi);
+
+	return 0;
+}
+
+static int meson_dw_mipi_dsi_host_detach(void *priv_data,
+					 struct mipi_dsi_device *device)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = priv_data;
+
+	if (device == mipi_dsi->dsi_device)
+		mipi_dsi->dsi_device = NULL;
+	else
+		return -EINVAL;
+
+	return phy_exit(mipi_dsi->phy);
+}
+
+static const struct dw_mipi_dsi_host_ops meson_dw_mipi_dsi_host_ops = {
+	.attach = meson_dw_mipi_dsi_host_attach,
+	.detach = meson_dw_mipi_dsi_host_detach,
+};
+
+static int meson_dw_mipi_dsi_probe(struct platform_device *pdev)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi;
+	struct device *dev = &pdev->dev;
+
+	mipi_dsi = devm_kzalloc(dev, sizeof(*mipi_dsi), GFP_KERNEL);
+	if (!mipi_dsi)
+		return -ENOMEM;
+
+	mipi_dsi->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(mipi_dsi->base))
+		return PTR_ERR(mipi_dsi->base);
+
+	mipi_dsi->phy = devm_phy_get(dev, "dphy");
+	if (IS_ERR(mipi_dsi->phy))
+		return dev_err_probe(dev, PTR_ERR(mipi_dsi->phy),
+				     "failed to get mipi dphy\n");
+
+	mipi_dsi->bit_clk = devm_clk_get_enabled(dev, "bit");
+	if (IS_ERR(mipi_dsi->bit_clk)) {
+		int ret = PTR_ERR(mipi_dsi->bit_clk);
+
+		/* TOFIX GP0 on some platforms fails to lock in early boot, defer probe */
+		if (ret == -EIO)
+			ret = -EPROBE_DEFER;
+
+		return dev_err_probe(dev, ret, "Unable to get enabled bit_clk\n");
+	}
+
+	mipi_dsi->px_clk = devm_clk_get_enabled(dev, "px");
+	if (IS_ERR(mipi_dsi->px_clk))
+		return dev_err_probe(dev, PTR_ERR(mipi_dsi->px_clk),
+				     "Unable to get enabled px_clk\n");
+
+	/*
+	 * We use a TOP reset signal because the APB reset signal
+	 * is handled by the TOP control registers.
+	 */
+	mipi_dsi->top_rst = devm_reset_control_get_exclusive(dev, "top");
+	if (IS_ERR(mipi_dsi->top_rst))
+		return dev_err_probe(dev, PTR_ERR(mipi_dsi->top_rst),
+				     "Unable to get reset control\n");
+
+	reset_control_assert(mipi_dsi->top_rst);
+	usleep_range(10, 20);
+	reset_control_deassert(mipi_dsi->top_rst);
+
+	/* MIPI DSI Controller */
+
+	mipi_dsi->dev = dev;
+	mipi_dsi->pdata.base = mipi_dsi->base;
+	mipi_dsi->pdata.max_data_lanes = 4;
+	mipi_dsi->pdata.phy_ops = &meson_dw_mipi_dsi_phy_ops;
+	mipi_dsi->pdata.host_ops = &meson_dw_mipi_dsi_host_ops;
+	mipi_dsi->pdata.priv_data = mipi_dsi;
+	platform_set_drvdata(pdev, mipi_dsi);
+
+	mipi_dsi->dmd = dw_mipi_dsi_probe(pdev, &mipi_dsi->pdata);
+	if (IS_ERR(mipi_dsi->dmd))
+		return dev_err_probe(dev, PTR_ERR(mipi_dsi->dmd),
+				     "Failed to probe dw_mipi_dsi\n");
+
+	return 0;
+}
+
+static int meson_dw_mipi_dsi_remove(struct platform_device *pdev)
+{
+	struct meson_dw_mipi_dsi *mipi_dsi = platform_get_drvdata(pdev);
+
+	dw_mipi_dsi_remove(mipi_dsi->dmd);
+
+	return 0;
+}
+
+static const struct of_device_id meson_dw_mipi_dsi_of_table[] = {
+	{ .compatible = "amlogic,meson-g12a-dw-mipi-dsi", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, meson_dw_mipi_dsi_of_table);
+
+static struct platform_driver meson_dw_mipi_dsi_platform_driver = {
+	.probe		= meson_dw_mipi_dsi_probe,
+	.remove		= meson_dw_mipi_dsi_remove,
+	.driver		= {
+		.name		= DRIVER_NAME,
+		.of_match_table	= meson_dw_mipi_dsi_of_table,
+	},
+};
+module_platform_driver(meson_dw_mipi_dsi_platform_driver);
+
+MODULE_AUTHOR("Neil Armstrong <narmstrong@baylibre.com>");
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/meson/meson_dw_mipi_dsi.h b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.h
new file mode 100644
index 0000000000000..e1bd6b85d6a3a
--- /dev/null
+++ b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.h
@@ -0,0 +1,160 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (C) 2020 BayLibre, SAS
+ * Author: Neil Armstrong <narmstrong@baylibre.com>
+ * Copyright (C) 2018 Amlogic, Inc. All rights reserved.
+ */
+
+#ifndef __MESON_DW_MIPI_DSI_H
+#define __MESON_DW_MIPI_DSI_H
+
+/* Top-level registers */
+/* [31: 4]    Reserved.     Default 0.
+ *     [3] RW timing_rst_n: Default 1.
+ *		1=Assert SW reset of timing feature.   0=Release reset.
+ *     [2] RW dpi_rst_n: Default 1.
+ *		1=Assert SW reset on mipi_dsi_host_dpi block.   0=Release reset.
+ *     [1] RW intr_rst_n: Default 1.
+ *		1=Assert SW reset on mipi_dsi_host_intr block.  0=Release reset.
+ *     [0] RW dwc_rst_n:  Default 1.
+ *		1=Assert SW reset on IP core.   0=Release reset.
+ */
+#define MIPI_DSI_TOP_SW_RESET                      0x3c0
+
+#define MIPI_DSI_TOP_SW_RESET_DWC	BIT(0)
+#define MIPI_DSI_TOP_SW_RESET_INTR	BIT(1)
+#define MIPI_DSI_TOP_SW_RESET_DPI	BIT(2)
+#define MIPI_DSI_TOP_SW_RESET_TIMING	BIT(3)
+
+/* [31: 5] Reserved.   Default 0.
+ *     [4] RW manual_edpihalt: Default 0.
+ *		1=Manual suspend VencL; 0=do not suspend VencL.
+ *     [3] RW auto_edpihalt_en: Default 0.
+ *		1=Enable IP's edpihalt signal to suspend VencL;
+ *		0=IP's edpihalt signal does not affect VencL.
+ *     [2] RW clock_freerun: Apply to auto-clock gate only. Default 0.
+ *		0=Default, use auto-clock gating to save power;
+ *		1=use free-run clock, disable auto-clock gating, for debug mode.
+ *     [1] RW enable_pixclk: A manual clock gate option, due to DWC IP does not
+ *		have auto-clock gating. 1=Enable pixclk.      Default 0.
+ *     [0] RW enable_sysclk: A manual clock gate option, due to DWC IP does not
+ *		have auto-clock gating. 1=Enable sysclk.      Default 0.
+ */
+#define MIPI_DSI_TOP_CLK_CNTL                      0x3c4
+
+#define MIPI_DSI_TOP_CLK_SYSCLK_EN	BIT(0)
+#define MIPI_DSI_TOP_CLK_PIXCLK_EN	BIT(1)
+
+/* [31:24]    Reserved. Default 0.
+ * [23:20] RW dpi_color_mode: Define DPI pixel format. Default 0.
+ *		0=16-bit RGB565 config 1;
+ *		1=16-bit RGB565 config 2;
+ *		2=16-bit RGB565 config 3;
+ *		3=18-bit RGB666 config 1;
+ *		4=18-bit RGB666 config 2;
+ *		5=24-bit RGB888;
+ *		6=20-bit YCbCr 4:2:2;
+ *		7=24-bit YCbCr 4:2:2;
+ *		8=16-bit YCbCr 4:2:2;
+ *		9=30-bit RGB;
+ *		10=36-bit RGB;
+ *		11=12-bit YCbCr 4:2:0.
+ *    [19] Reserved. Default 0.
+ * [18:16] RW in_color_mode:  Define VENC data width. Default 0.
+ *		0=30-bit pixel;
+ *		1=24-bit pixel;
+ *		2=18-bit pixel, RGB666;
+ *		3=16-bit pixel, RGB565.
+ * [15:14] RW chroma_subsample: Define method of chroma subsampling. Default 0.
+ *		Applicable to YUV422 or YUV420 only.
+ *		0=Use even pixel's chroma;
+ *		1=Use odd pixel's chroma;
+ *		2=Use averaged value between even and odd pair.
+ * [13:12] RW comp2_sel:  Select which component to be Cr or B: Default 2.
+ *		0=comp0; 1=comp1; 2=comp2.
+ * [11:10] RW comp1_sel:  Select which component to be Cb or G: Default 1.
+ *		0=comp0; 1=comp1; 2=comp2.
+ *  [9: 8] RW comp0_sel:  Select which component to be Y  or R: Default 0.
+ *		0=comp0; 1=comp1; 2=comp2.
+ *     [7]    Reserved. Default 0.
+ *     [6] RW de_pol:  Default 0.
+ *		If DE input is active low, set to 1 to invert to active high.
+ *     [5] RW hsync_pol: Default 0.
+ *		If HS input is active low, set to 1 to invert to active high.
+ *     [4] RW vsync_pol: Default 0.
+ *		If VS input is active low, set to 1 to invert to active high.
+ *     [3] RW dpicolorm: Signal to IP.   Default 0.
+ *     [2] RW dpishutdn: Signal to IP.   Default 0.
+ *     [1]    Reserved.  Default 0.
+ *     [0]    Reserved.  Default 0.
+ */
+#define MIPI_DSI_TOP_CNTL                          0x3c8
+
+/* VENC data width */
+#define VENC_IN_COLOR_30B   0x0
+#define VENC_IN_COLOR_24B   0x1
+#define VENC_IN_COLOR_18B   0x2
+#define VENC_IN_COLOR_16B   0x3
+
+/* DPI pixel format */
+#define DPI_COLOR_16BIT_CFG_1		0
+#define DPI_COLOR_16BIT_CFG_2		1
+#define DPI_COLOR_16BIT_CFG_3		2
+#define DPI_COLOR_18BIT_CFG_1		3
+#define DPI_COLOR_18BIT_CFG_2		4
+#define DPI_COLOR_24BIT			5
+#define DPI_COLOR_20BIT_YCBCR_422	6
+#define DPI_COLOR_24BIT_YCBCR_422	7
+#define DPI_COLOR_16BIT_YCBCR_422	8
+#define DPI_COLOR_30BIT			9
+#define DPI_COLOR_36BIT			10
+#define DPI_COLOR_12BIT_YCBCR_420	11
+
+#define MIPI_DSI_TOP_DPI_COLOR_MODE	GENMASK(23, 20)
+#define MIPI_DSI_TOP_IN_COLOR_MODE	GENMASK(18, 16)
+#define MIPI_DSI_TOP_CHROMA_SUBSAMPLE	GENMASK(15, 14)
+#define MIPI_DSI_TOP_COMP2_SEL		GENMASK(13, 12)
+#define MIPI_DSI_TOP_COMP1_SEL		GENMASK(11, 10)
+#define MIPI_DSI_TOP_COMP0_SEL		GENMASK(9, 8)
+#define MIPI_DSI_TOP_DE_INVERT		BIT(6)
+#define MIPI_DSI_TOP_HSYNC_INVERT	BIT(5)
+#define MIPI_DSI_TOP_VSYNC_INVERT	BIT(4)
+#define MIPI_DSI_TOP_DPICOLORM		BIT(3)
+#define MIPI_DSI_TOP_DPISHUTDN		BIT(2)
+
+#define MIPI_DSI_TOP_SUSPEND_CNTL                  0x3cc
+#define MIPI_DSI_TOP_SUSPEND_LINE                  0x3d0
+#define MIPI_DSI_TOP_SUSPEND_PIX                   0x3d4
+#define MIPI_DSI_TOP_MEAS_CNTL                     0x3d8
+/* [0] R  stat_edpihalt:  edpihalt signal from IP.    Default 0. */
+#define MIPI_DSI_TOP_STAT                          0x3dc
+#define MIPI_DSI_TOP_MEAS_STAT_TE0                 0x3e0
+#define MIPI_DSI_TOP_MEAS_STAT_TE1                 0x3e4
+#define MIPI_DSI_TOP_MEAS_STAT_VS0                 0x3e8
+#define MIPI_DSI_TOP_MEAS_STAT_VS1                 0x3ec
+/* [31:16] RW intr_stat/clr. Default 0.
+ *		For each bit, read as this interrupt level status,
+ *		write 1 to clear.
+ * [31:22] Reserved
+ * [   21] stat/clr of eof interrupt
+ * [   21] vde_fall interrupt
+ * [   19] stat/clr of de_rise interrupt
+ * [   18] stat/clr of vs_fall interrupt
+ * [   17] stat/clr of vs_rise interrupt
+ * [   16] stat/clr of dwc_edpite interrupt
+ * [15: 0] RW intr_enable. Default 0.
+ *		For each bit, 1=enable this interrupt, 0=disable.
+ *	[15: 6] Reserved
+ *	[    5] eof interrupt
+ *	[    4] de_fall interrupt
+ *	[    3] de_rise interrupt
+ *	[    2] vs_fall interrupt
+ *	[    1] vs_rise interrupt
+ *	[    0] dwc_edpite interrupt
+ */
+#define MIPI_DSI_TOP_INTR_CNTL_STAT                0x3f0
+// 31: 2    Reserved.   Default 0.
+//  1: 0 RW mem_pd.     Default 3.
+#define MIPI_DSI_TOP_MEM_PD                        0x3f4
+
+#endif /* __MESON_DW_MIPI_DSI_H */
-- 
GitLab


From 29c6df0d942454cb43334cf0e36de068f4124b94 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Tue, 30 May 2023 09:38:14 +0200
Subject: [PATCH 0726/1791] drm/panel: khadas-ts050: update timings to achieve
 60Hz refresh rate

This updates the panel timings to achieve a clean 60Hz refresh rate.

Reviewed-by: Nicolas Belin <nbelin@baylibre.com>
Tested-by: Nicolas Belin <nbelin@baylibre.com> # on Khadas VIM3 + TS050 Panel
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230512-amlogic-v6-4-upstream-dsi-ccf-vim3-v5-13-56eb7a4d5b8e@linaro.org
---
 drivers/gpu/drm/panel/panel-khadas-ts050.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-khadas-ts050.c b/drivers/gpu/drm/panel/panel-khadas-ts050.c
index 1ab1ebe308820..b942a01622743 100644
--- a/drivers/gpu/drm/panel/panel-khadas-ts050.c
+++ b/drivers/gpu/drm/panel/panel-khadas-ts050.c
@@ -568,7 +568,7 @@ static const struct khadas_ts050_panel_cmd init_code[] = {
 	{0xfb, 0x01},
 	/* Select CMD1 */
 	{0xff, 0x00},
-	{0xd3, 0x05}, /* RGBMIPICTRL: VSYNC back porch = 5 */
+	{0xd3, 0x22}, /* RGBMIPICTRL: VSYNC back porch = 34 */
 	{0xd4, 0x04}, /* RGBMIPICTRL: VSYNC front porch = 4 */
 };
 
@@ -717,15 +717,15 @@ static int khadas_ts050_panel_disable(struct drm_panel *panel)
 }
 
 static const struct drm_display_mode default_mode = {
-	.clock = 120000,
-	.hdisplay = 1088,
-	.hsync_start = 1088 + 104,
-	.hsync_end = 1088 + 104 + 4,
-	.htotal = 1088 + 104 + 4 + 127,
+	.clock = 160000,
+	.hdisplay = 1080,
+	.hsync_start = 1080 + 117,
+	.hsync_end = 1080 + 117 + 5,
+	.htotal = 1080 + 117 + 5 + 160,
 	.vdisplay = 1920,
 	.vsync_start = 1920 + 4,
-	.vsync_end = 1920 + 4 + 2,
-	.vtotal = 1920 + 4 + 2 + 3,
+	.vsync_end = 1920 + 4 + 3,
+	.vtotal = 1920 + 4 + 3 + 31,
 	.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
 };
 
-- 
GitLab


From 06c3269cd5748293d77b9ae8d05611f7cc9187df Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Thu, 25 May 2023 17:31:48 +0800
Subject: [PATCH 0727/1791] dt-bindings: display: panel: Add compatible for
 Starry himax83102-j02

The STARRY himax83102-j02 is a 10.51" WUXGA TFT LCD panel,
which fits in nicely with the existing panel-boe-tv101wum-nl6
driver. Hence, we add a new compatible with panel specific config.

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525093151.2338370-2-yangcong5@huaqin.corp-partner.google.com
---
 .../devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml     | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml b/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
index aed55608ebf6d..28a7beeb8f92e 100644
--- a/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
+++ b/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
@@ -32,6 +32,8 @@ properties:
       - innolux,hj110iz-01a
         # STARRY 2081101QFH032011-53G 10.1" WUXGA TFT LCD panel
       - starry,2081101qfh032011-53g
+        # STARRY himax83102-j02 10.51" WUXGA TFT LCD panel
+      - starry,himax83102-j02
 
   reg:
     description: the virtual channel number of a DSI peripheral
-- 
GitLab


From 1bc2ef065f13043bbefc927e6dc99c82b56ec037 Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Thu, 25 May 2023 17:31:49 +0800
Subject: [PATCH 0728/1791] drm/panel: Support for Starry-himax83102-j02 TDDI
 MIPI-DSI panel

The Starry-himax83102-j02 is a 10.51" WUXGA TFT panel. which fits in nicely
with the existing panel-boe-tv101wum-nl6 driver. From the datasheet[1], MIPI
needs to keep the LP11 state before the lcm_reset pin is pulled high, so
increase lp11_before_reset flag.

[1]: https://github.com/HimaxSoftware/Doc/tree/main/Himax_Chipset_Power_Sequence

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
[dianders: removed some spacing on one line]
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525093151.2338370-3-yangcong5@huaqin.corp-partner.google.com
---
 .../gpu/drm/panel/panel-boe-tv101wum-nl6.c    | 101 ++++++++++++++++++
 1 file changed, 101 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index f5a6046f1d19b..735f24065dec3 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -1301,6 +1301,76 @@ static const struct panel_init_cmd starry_qfh032011_53g_init_cmd[] = {
 	{},
 };
 
+static const struct panel_init_cmd starry_himax83102_j02_init_cmd[] = {
+	_INIT_DCS_CMD(0xB9, 0x83, 0x10, 0x21, 0x55, 0x00),
+	_INIT_DCS_CMD(0xB1, 0x2C, 0xB5, 0xB5, 0x31, 0xF1, 0x31, 0xD7, 0x2F, 0x36, 0x36, 0x36, 0x36, 0x1A, 0x8B, 0x11,
+		0x65, 0x00, 0x88, 0xFA, 0xFF, 0xFF, 0x8F, 0xFF, 0x08, 0x74, 0x33),
+	_INIT_DCS_CMD(0xB2, 0x00, 0x47, 0xB0, 0x80, 0x00, 0x12, 0x72, 0x3C, 0xA3, 0x03, 0x03, 0x00, 0x00, 0x88, 0xF5),
+	_INIT_DCS_CMD(0xB4, 0x76, 0x76, 0x76, 0x76, 0x76, 0x76, 0x63, 0x5C, 0x63, 0x5C, 0x01, 0x9E),
+	_INIT_DCS_CMD(0xE9, 0xCD),
+	_INIT_DCS_CMD(0xBA, 0x84),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xBC, 0x1B, 0x04),
+	_INIT_DCS_CMD(0xBE, 0x20),
+	_INIT_DCS_CMD(0xBF, 0xFC, 0xC4),
+	_INIT_DCS_CMD(0xC0, 0x36, 0x36, 0x22, 0x11, 0x22, 0xA0, 0x61, 0x08, 0xF5, 0x03),
+	_INIT_DCS_CMD(0xE9, 0xCC),
+	_INIT_DCS_CMD(0xC7, 0x80),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xE9, 0xC6),
+	_INIT_DCS_CMD(0xC8, 0x97),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xC9, 0x00, 0x1E, 0x13, 0x88, 0x01),
+	_INIT_DCS_CMD(0xCB, 0x08, 0x13, 0x07, 0x00, 0x0F, 0x33),
+	_INIT_DCS_CMD(0xCC, 0x02),
+	_INIT_DCS_CMD(0xE9, 0xC4),
+	_INIT_DCS_CMD(0xD0, 0x03),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xD1, 0x37, 0x06, 0x00, 0x02, 0x04, 0x0C, 0xFF),
+	_INIT_DCS_CMD(0xD2, 0x1F, 0x11, 0x1F),
+	_INIT_DCS_CMD(0xD3, 0x06, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x08, 0x37, 0x47, 0x34, 0x3B, 0x12, 0x12, 0x03,
+		0x03, 0x32, 0x10, 0x10, 0x00, 0x10, 0x32, 0x10, 0x08, 0x00, 0x08, 0x32, 0x17, 0x94, 0x07, 0x94, 0x00, 0x00),
+	_INIT_DCS_CMD(0xD5, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x19, 0x19, 0x40, 0x40, 0x1A, 0x1A,
+		0x1B, 0x1B, 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x20, 0x21, 0x28, 0x29, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18),
+	_INIT_DCS_CMD(0xD6, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x40, 0x40, 0x19, 0x19, 0x1A, 0x1A,
+		0x1B, 0x1B, 0x07, 0x06, 0x05, 0x04, 0x03, 0x02, 0x01, 0x00, 0x29, 0x28, 0x21, 0x20, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18),
+	_INIT_DCS_CMD(0xD8, 0xAA, 0xBA, 0xEA, 0xAA, 0xAA, 0xA0, 0xAA, 0xBA, 0xEA, 0xAA, 0xAA, 0xA0, 0xAA, 0xBA, 0xEA, 0xAA,
+		0xAA, 0xA0, 0xAA, 0xBA, 0xEA, 0xAA, 0xAA, 0xA0, 0xAA, 0xBA, 0xEA, 0xAA, 0xAA, 0xA0, 0xAA, 0xBA, 0xEA, 0xAA, 0xAA, 0xA0),
+	_INIT_DCS_CMD(0xE0, 0x00, 0x09, 0x14, 0x1E, 0x26, 0x48, 0x61, 0x67, 0x6C, 0x67, 0x7D, 0x7F, 0x80, 0x8B, 0x87, 0x8F, 0x98, 0xAB,
+		0xAB, 0x55, 0x5C, 0x68, 0x73, 0x00, 0x09, 0x14, 0x1E, 0x26, 0x48, 0x61, 0x67, 0x6C, 0x67, 0x7D, 0x7F, 0x80, 0x8B, 0x87, 0x8F, 0x98, 0xAB, 0xAB, 0x55, 0x5C, 0x68, 0x73),
+	_INIT_DCS_CMD(0xE7, 0x0E, 0x10, 0x10, 0x21, 0x2B, 0x9A, 0x02, 0x54, 0x9A, 0x14, 0x14, 0x00, 0x00, 0x00, 0x00, 0x12, 0x05, 0x02, 0x02, 0x10),
+	_INIT_DCS_CMD(0xBD, 0x01),
+	_INIT_DCS_CMD(0xB1, 0x01, 0xBF, 0x11),
+	_INIT_DCS_CMD(0xCB, 0x86),
+	_INIT_DCS_CMD(0xD2, 0x3C, 0xFA),
+	_INIT_DCS_CMD(0xE9, 0xC5),
+	_INIT_DCS_CMD(0xD3, 0x00, 0x00, 0x00, 0x00, 0x80, 0x0C, 0x01),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xE7, 0x02, 0x00, 0x28, 0x01, 0x7E, 0x0F, 0x7E, 0x10, 0xA0, 0x00, 0x00, 0x20, 0x40, 0x50, 0x40),
+	_INIT_DCS_CMD(0xBD, 0x02),
+	_INIT_DCS_CMD(0xD8, 0xFF, 0xFF, 0xBF, 0xFE, 0xAA, 0xA0, 0xFF, 0xFF, 0xBF, 0xFE, 0xAA, 0xA0),
+	_INIT_DCS_CMD(0xE7, 0xFE, 0x04, 0xFE, 0x04, 0xFE, 0x04, 0x03, 0x03, 0x03, 0x26, 0x00, 0x26, 0x81, 0x02, 0x40, 0x00, 0x20, 0x9E, 0x04, 0x03, 0x02, 0x01, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00),
+	_INIT_DCS_CMD(0xBD, 0x03),
+	_INIT_DCS_CMD(0xE9, 0xC6),
+	_INIT_DCS_CMD(0xB4, 0x03, 0xFF, 0xF8),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xD8, 0x00, 0x2A, 0xAA, 0xA8, 0x00, 0x00, 0x00, 0x2A, 0xAA, 0xA8, 0x00, 0x00, 0x00, 0x3F, 0xFF, 0xFC, 0x00, 0x00, 0x00, 0x3F, 0xFF, 0xFC, 0x00, 0x00, 0x00, 0x2A, 0xAA, 0xA8,
+		0x00, 0x00, 0x00, 0x2A, 0xAA, 0xA8, 0x00, 0x00),
+	_INIT_DCS_CMD(0xBD, 0x00),
+	_INIT_DCS_CMD(0xE9, 0xC4),
+	_INIT_DCS_CMD(0xBA, 0x96),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xBD, 0x01),
+	_INIT_DCS_CMD(0xE9, 0xC5),
+	_INIT_DCS_CMD(0xBA, 0x4F),
+	_INIT_DCS_CMD(0xE9, 0x3F),
+	_INIT_DCS_CMD(0xBD, 0x00),
+	_INIT_DCS_CMD(0x11),
+	_INIT_DELAY_CMD(120),
+	_INIT_DCS_CMD(0x29),
+	{},
+};
+
 static inline struct boe_panel *to_boe_panel(struct drm_panel *panel)
 {
 	return container_of(panel, struct boe_panel, base);
@@ -1698,6 +1768,34 @@ static const struct panel_desc starry_qfh032011_53g_desc = {
 	.init_cmds = starry_qfh032011_53g_init_cmd,
 };
 
+static const struct drm_display_mode starry_himax83102_j02_default_mode = {
+	.clock = 161600,
+	.hdisplay = 1200,
+	.hsync_start = 1200 + 40,
+	.hsync_end = 1200 + 40 + 20,
+	.htotal = 1200 + 40 + 20 + 40,
+	.vdisplay = 1920,
+	.vsync_start = 1920 + 116,
+	.vsync_end = 1920 + 116 + 8,
+	.vtotal = 1920 + 116 + 8 + 12,
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static const struct panel_desc starry_himax83102_j02_desc = {
+	.modes = &starry_himax83102_j02_default_mode,
+	.bpc = 8,
+	.size = {
+		.width_mm = 141,
+		.height_mm = 226,
+	},
+	.lanes = 4,
+	.format = MIPI_DSI_FMT_RGB888,
+	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
+		      MIPI_DSI_MODE_LPM,
+	.init_cmds = starry_himax83102_j02_init_cmd,
+	.lp11_before_reset = true,
+};
+
 static int boe_panel_get_modes(struct drm_panel *panel,
 			       struct drm_connector *connector)
 {
@@ -1871,6 +1969,9 @@ static const struct of_device_id boe_of_match[] = {
 	{ .compatible = "starry,2081101qfh032011-53g",
 	  .data = &starry_qfh032011_53g_desc
 	},
+	{ .compatible = "starry,himax83102-j02",
+	  .data = &starry_himax83102_j02_desc
+	},
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, boe_of_match);
-- 
GitLab


From 0a73471ca1f7a67d9d2967379087d1784ba62855 Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Thu, 25 May 2023 17:31:50 +0800
Subject: [PATCH 0729/1791] dt-bindings: display: panel: Add compatible for
 Starry ili9882t

The STARRY ili9882t is a 10.51" WUXGA TFT LCD panel,
which fits in nicely with the existing panel-boe-tv101wum-nl6
driver. Hence, we add a new compatible with panel specific config.

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525093151.2338370-4-yangcong5@huaqin.corp-partner.google.com
---
 .../devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml     | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml b/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
index 28a7beeb8f92e..906ef62709b8a 100644
--- a/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
+++ b/Documentation/devicetree/bindings/display/panel/boe,tv101wum-nl6.yaml
@@ -34,6 +34,8 @@ properties:
       - starry,2081101qfh032011-53g
         # STARRY himax83102-j02 10.51" WUXGA TFT LCD panel
       - starry,himax83102-j02
+        # STARRY ili9882t 10.51" WUXGA TFT LCD panel
+      - starry,ili9882t
 
   reg:
     description: the virtual channel number of a DSI peripheral
-- 
GitLab


From 8716a6473e6c50a5c335f6fa7d6b73779ecf6972 Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Thu, 25 May 2023 17:31:51 +0800
Subject: [PATCH 0730/1791] drm/panel: Support for Starry-ili9882t TDDI
 MIPI-DSI panel

The Starry-ili9882 is a 10.51" WUXGA TFT panel. which fits in nicely with
the existing panel-boe-tv101wum-nl6 driver. From the datasheet,MIPI need
to keep the LP11 state before the lcm_reset pin is pulled high. So add
lp11_before_reset flag.

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230525093151.2338370-5-yangcong5@huaqin.corp-partner.google.com
---
 .../gpu/drm/panel/panel-boe-tv101wum-nl6.c    | 371 ++++++++++++++++++
 1 file changed, 371 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
index 735f24065dec3..3cc9fb0d4f5df 100644
--- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
+++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c
@@ -1371,6 +1371,346 @@ static const struct panel_init_cmd starry_himax83102_j02_init_cmd[] = {
 	{},
 };
 
+static const struct panel_init_cmd starry_ili9882t_init_cmd[] = {
+	_INIT_DELAY_CMD(5),
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x01),
+	_INIT_DCS_CMD(0x00, 0x42),
+	_INIT_DCS_CMD(0x01, 0x11),
+	_INIT_DCS_CMD(0x02, 0x00),
+	_INIT_DCS_CMD(0x03, 0x00),
+
+	_INIT_DCS_CMD(0x04, 0x01),
+	_INIT_DCS_CMD(0x05, 0x11),
+	_INIT_DCS_CMD(0x06, 0x00),
+	_INIT_DCS_CMD(0x07, 0x00),
+
+	_INIT_DCS_CMD(0x08, 0x80),
+	_INIT_DCS_CMD(0x09, 0x81),
+	_INIT_DCS_CMD(0x0A, 0x71),
+	_INIT_DCS_CMD(0x0B, 0x00),
+
+	_INIT_DCS_CMD(0x0C, 0x00),
+	_INIT_DCS_CMD(0x0E, 0x1A),
+
+	_INIT_DCS_CMD(0x24, 0x00),
+	_INIT_DCS_CMD(0x25, 0x00),
+	_INIT_DCS_CMD(0x26, 0x00),
+	_INIT_DCS_CMD(0x27, 0x00),
+
+	_INIT_DCS_CMD(0x2C, 0xD4),
+	_INIT_DCS_CMD(0xB9, 0x40),
+
+	_INIT_DCS_CMD(0xB0, 0x11),
+
+	_INIT_DCS_CMD(0xE6, 0x32),
+	_INIT_DCS_CMD(0xD1, 0x30),
+
+	_INIT_DCS_CMD(0xD6, 0x55),
+
+	_INIT_DCS_CMD(0xD0, 0x01),
+	_INIT_DCS_CMD(0xE3, 0x93),
+	_INIT_DCS_CMD(0xE4, 0x00),
+	_INIT_DCS_CMD(0xE5, 0x80),
+
+	_INIT_DCS_CMD(0x31, 0x07),
+	_INIT_DCS_CMD(0x32, 0x07),
+	_INIT_DCS_CMD(0x33, 0x07),
+	_INIT_DCS_CMD(0x34, 0x07),
+	_INIT_DCS_CMD(0x35, 0x07),
+	_INIT_DCS_CMD(0x36, 0x01),
+	_INIT_DCS_CMD(0x37, 0x00),
+	_INIT_DCS_CMD(0x38, 0x28),
+	_INIT_DCS_CMD(0x39, 0x29),
+	_INIT_DCS_CMD(0x3A, 0x11),
+	_INIT_DCS_CMD(0x3B, 0x13),
+	_INIT_DCS_CMD(0x3C, 0x15),
+	_INIT_DCS_CMD(0x3D, 0x17),
+	_INIT_DCS_CMD(0x3E, 0x09),
+	_INIT_DCS_CMD(0x3F, 0x0D),
+	_INIT_DCS_CMD(0x40, 0x02),
+	_INIT_DCS_CMD(0x41, 0x02),
+	_INIT_DCS_CMD(0x42, 0x02),
+	_INIT_DCS_CMD(0x43, 0x02),
+	_INIT_DCS_CMD(0x44, 0x02),
+	_INIT_DCS_CMD(0x45, 0x02),
+	_INIT_DCS_CMD(0x46, 0x02),
+
+	_INIT_DCS_CMD(0x47, 0x07),
+	_INIT_DCS_CMD(0x48, 0x07),
+	_INIT_DCS_CMD(0x49, 0x07),
+	_INIT_DCS_CMD(0x4A, 0x07),
+	_INIT_DCS_CMD(0x4B, 0x07),
+	_INIT_DCS_CMD(0x4C, 0x01),
+	_INIT_DCS_CMD(0x4D, 0x00),
+	_INIT_DCS_CMD(0x4E, 0x28),
+	_INIT_DCS_CMD(0x4F, 0x29),
+	_INIT_DCS_CMD(0x50, 0x10),
+	_INIT_DCS_CMD(0x51, 0x12),
+	_INIT_DCS_CMD(0x52, 0x14),
+	_INIT_DCS_CMD(0x53, 0x16),
+	_INIT_DCS_CMD(0x54, 0x08),
+	_INIT_DCS_CMD(0x55, 0x0C),
+	_INIT_DCS_CMD(0x56, 0x02),
+	_INIT_DCS_CMD(0x57, 0x02),
+	_INIT_DCS_CMD(0x58, 0x02),
+	_INIT_DCS_CMD(0x59, 0x02),
+	_INIT_DCS_CMD(0x5A, 0x02),
+	_INIT_DCS_CMD(0x5B, 0x02),
+	_INIT_DCS_CMD(0x5C, 0x02),
+
+	_INIT_DCS_CMD(0x61, 0x07),
+	_INIT_DCS_CMD(0x62, 0x07),
+	_INIT_DCS_CMD(0x63, 0x07),
+	_INIT_DCS_CMD(0x64, 0x07),
+	_INIT_DCS_CMD(0x65, 0x07),
+	_INIT_DCS_CMD(0x66, 0x01),
+	_INIT_DCS_CMD(0x67, 0x00),
+	_INIT_DCS_CMD(0x68, 0x28),
+	_INIT_DCS_CMD(0x69, 0x29),
+	_INIT_DCS_CMD(0x6A, 0x16),
+	_INIT_DCS_CMD(0x6B, 0x14),
+	_INIT_DCS_CMD(0x6C, 0x12),
+	_INIT_DCS_CMD(0x6D, 0x10),
+	_INIT_DCS_CMD(0x6E, 0x0C),
+	_INIT_DCS_CMD(0x6F, 0x08),
+	_INIT_DCS_CMD(0x70, 0x02),
+	_INIT_DCS_CMD(0x71, 0x02),
+	_INIT_DCS_CMD(0x72, 0x02),
+	_INIT_DCS_CMD(0x73, 0x02),
+	_INIT_DCS_CMD(0x74, 0x02),
+	_INIT_DCS_CMD(0x75, 0x02),
+	_INIT_DCS_CMD(0x76, 0x02),
+
+	_INIT_DCS_CMD(0x77, 0x07),
+	_INIT_DCS_CMD(0x78, 0x07),
+	_INIT_DCS_CMD(0x79, 0x07),
+	_INIT_DCS_CMD(0x7A, 0x07),
+	_INIT_DCS_CMD(0x7B, 0x07),
+	_INIT_DCS_CMD(0x7C, 0x01),
+	_INIT_DCS_CMD(0x7D, 0x00),
+	_INIT_DCS_CMD(0x7E, 0x28),
+	_INIT_DCS_CMD(0x7F, 0x29),
+	_INIT_DCS_CMD(0x80, 0x17),
+	_INIT_DCS_CMD(0x81, 0x15),
+	_INIT_DCS_CMD(0x82, 0x13),
+	_INIT_DCS_CMD(0x83, 0x11),
+	_INIT_DCS_CMD(0x84, 0x0D),
+	_INIT_DCS_CMD(0x85, 0x09),
+	_INIT_DCS_CMD(0x86, 0x02),
+	_INIT_DCS_CMD(0x87, 0x07),
+	_INIT_DCS_CMD(0x88, 0x07),
+	_INIT_DCS_CMD(0x89, 0x07),
+	_INIT_DCS_CMD(0x8A, 0x07),
+	_INIT_DCS_CMD(0x8B, 0x07),
+	_INIT_DCS_CMD(0x8C, 0x07),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x02),
+	_INIT_DCS_CMD(0x29, 0x3A),
+	_INIT_DCS_CMD(0x2A, 0x3B),
+
+	_INIT_DCS_CMD(0x06, 0x01),
+	_INIT_DCS_CMD(0x07, 0x01),
+	_INIT_DCS_CMD(0x08, 0x0C),
+	_INIT_DCS_CMD(0x09, 0x44),
+
+	_INIT_DCS_CMD(0x3C, 0x0A),
+	_INIT_DCS_CMD(0x39, 0x11),
+	_INIT_DCS_CMD(0x3D, 0x00),
+	_INIT_DCS_CMD(0x3A, 0x0C),
+	_INIT_DCS_CMD(0x3B, 0x44),
+
+	_INIT_DCS_CMD(0x53, 0x1F),
+	_INIT_DCS_CMD(0x5E, 0x40),
+	_INIT_DCS_CMD(0x84, 0x00),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x03),
+	_INIT_DCS_CMD(0x20, 0x01),
+	_INIT_DCS_CMD(0x21, 0x3C),
+	_INIT_DCS_CMD(0x22, 0xFA),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x0A),
+	_INIT_DCS_CMD(0xE0, 0x01),
+	_INIT_DCS_CMD(0xE2, 0x01),
+	_INIT_DCS_CMD(0xE5, 0x91),
+	_INIT_DCS_CMD(0xE6, 0x3C),
+	_INIT_DCS_CMD(0xE7, 0x00),
+	_INIT_DCS_CMD(0xE8, 0xFA),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x12),
+	_INIT_DCS_CMD(0x87, 0x2C),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x05),
+	_INIT_DCS_CMD(0x73, 0xE5),
+	_INIT_DCS_CMD(0x7F, 0x6B),
+	_INIT_DCS_CMD(0x6D, 0xA4),
+	_INIT_DCS_CMD(0x79, 0x54),
+	_INIT_DCS_CMD(0x69, 0x97),
+	_INIT_DCS_CMD(0x6A, 0x97),
+	_INIT_DCS_CMD(0xA5, 0x3F),
+	_INIT_DCS_CMD(0x61, 0xDA),
+	_INIT_DCS_CMD(0xA7, 0xF1),
+	_INIT_DCS_CMD(0x5F, 0x01),
+	_INIT_DCS_CMD(0x62, 0x3F),
+	_INIT_DCS_CMD(0x1D, 0x90),
+	_INIT_DCS_CMD(0x86, 0x87),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x06),
+	_INIT_DCS_CMD(0xC0, 0x80),
+	_INIT_DCS_CMD(0xC1, 0x07),
+	_INIT_DCS_CMD(0xCA, 0x58),
+	_INIT_DCS_CMD(0xCB, 0x02),
+	_INIT_DCS_CMD(0xCE, 0x58),
+	_INIT_DCS_CMD(0xCF, 0x02),
+	_INIT_DCS_CMD(0x67, 0x60),
+	_INIT_DCS_CMD(0x10, 0x00),
+	_INIT_DCS_CMD(0x92, 0x22),
+	_INIT_DCS_CMD(0xD3, 0x08),
+	_INIT_DCS_CMD(0xD6, 0x55),
+	_INIT_DCS_CMD(0xDC, 0x38),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x08),
+	_INIT_DCS_CMD(0xE0, 0x00, 0x10, 0x2A, 0x4D, 0x61, 0x56, 0x6A, 0x6E, 0x79, 0x76, 0x8F, 0x95, 0x98, 0xAE, 0xAA, 0xB2, 0xBB, 0xCE, 0xC6, 0xBD, 0xD5, 0xE2, 0xE8),
+	_INIT_DCS_CMD(0xE1, 0x00, 0x10, 0x2A, 0x4D, 0x61, 0x56, 0x6A, 0x6E, 0x79, 0x76, 0x8F, 0x95, 0x98, 0xAE, 0xAA, 0xB2, 0xBB, 0xCE, 0xC6, 0xBD, 0xD5, 0xE2, 0xE8),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x04),
+	_INIT_DCS_CMD(0xBA, 0x81),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x0C),
+	_INIT_DCS_CMD(0x00, 0x02),
+	_INIT_DCS_CMD(0x01, 0x00),
+	_INIT_DCS_CMD(0x02, 0x03),
+	_INIT_DCS_CMD(0x03, 0x01),
+	_INIT_DCS_CMD(0x04, 0x03),
+	_INIT_DCS_CMD(0x05, 0x02),
+	_INIT_DCS_CMD(0x06, 0x04),
+	_INIT_DCS_CMD(0x07, 0x03),
+	_INIT_DCS_CMD(0x08, 0x03),
+	_INIT_DCS_CMD(0x09, 0x04),
+	_INIT_DCS_CMD(0x0A, 0x04),
+	_INIT_DCS_CMD(0x0B, 0x05),
+	_INIT_DCS_CMD(0x0C, 0x04),
+	_INIT_DCS_CMD(0x0D, 0x06),
+	_INIT_DCS_CMD(0x0E, 0x05),
+	_INIT_DCS_CMD(0x0F, 0x07),
+	_INIT_DCS_CMD(0x10, 0x04),
+	_INIT_DCS_CMD(0x11, 0x08),
+	_INIT_DCS_CMD(0x12, 0x05),
+	_INIT_DCS_CMD(0x13, 0x09),
+	_INIT_DCS_CMD(0x14, 0x05),
+	_INIT_DCS_CMD(0x15, 0x0A),
+	_INIT_DCS_CMD(0x16, 0x06),
+	_INIT_DCS_CMD(0x17, 0x0B),
+	_INIT_DCS_CMD(0x18, 0x05),
+	_INIT_DCS_CMD(0x19, 0x0C),
+	_INIT_DCS_CMD(0x1A, 0x06),
+	_INIT_DCS_CMD(0x1B, 0x0D),
+	_INIT_DCS_CMD(0x1C, 0x06),
+	_INIT_DCS_CMD(0x1D, 0x0E),
+	_INIT_DCS_CMD(0x1E, 0x07),
+	_INIT_DCS_CMD(0x1F, 0x0F),
+	_INIT_DCS_CMD(0x20, 0x06),
+	_INIT_DCS_CMD(0x21, 0x10),
+	_INIT_DCS_CMD(0x22, 0x07),
+	_INIT_DCS_CMD(0x23, 0x11),
+	_INIT_DCS_CMD(0x24, 0x07),
+	_INIT_DCS_CMD(0x25, 0x12),
+	_INIT_DCS_CMD(0x26, 0x08),
+	_INIT_DCS_CMD(0x27, 0x13),
+	_INIT_DCS_CMD(0x28, 0x07),
+	_INIT_DCS_CMD(0x29, 0x14),
+	_INIT_DCS_CMD(0x2A, 0x08),
+	_INIT_DCS_CMD(0x2B, 0x15),
+	_INIT_DCS_CMD(0x2C, 0x08),
+	_INIT_DCS_CMD(0x2D, 0x16),
+	_INIT_DCS_CMD(0x2E, 0x09),
+	_INIT_DCS_CMD(0x2F, 0x17),
+	_INIT_DCS_CMD(0x30, 0x08),
+	_INIT_DCS_CMD(0x31, 0x18),
+	_INIT_DCS_CMD(0x32, 0x09),
+	_INIT_DCS_CMD(0x33, 0x19),
+	_INIT_DCS_CMD(0x34, 0x09),
+	_INIT_DCS_CMD(0x35, 0x1A),
+	_INIT_DCS_CMD(0x36, 0x0A),
+	_INIT_DCS_CMD(0x37, 0x1B),
+	_INIT_DCS_CMD(0x38, 0x0A),
+	_INIT_DCS_CMD(0x39, 0x1C),
+	_INIT_DCS_CMD(0x3A, 0x0A),
+	_INIT_DCS_CMD(0x3B, 0x1D),
+	_INIT_DCS_CMD(0x3C, 0x0A),
+	_INIT_DCS_CMD(0x3D, 0x1E),
+	_INIT_DCS_CMD(0x3E, 0x0A),
+	_INIT_DCS_CMD(0x3F, 0x1F),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x04),
+	_INIT_DCS_CMD(0xBA, 0x01),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x0E),
+	_INIT_DCS_CMD(0x02, 0x0C),
+	_INIT_DCS_CMD(0x20, 0x10),
+	_INIT_DCS_CMD(0x25, 0x16),
+	_INIT_DCS_CMD(0x26, 0xE0),
+	_INIT_DCS_CMD(0x27, 0x00),
+	_INIT_DCS_CMD(0x29, 0x71),
+	_INIT_DCS_CMD(0x2A, 0x46),
+	_INIT_DCS_CMD(0x2B, 0x1F),
+	_INIT_DCS_CMD(0x2D, 0xC7),
+	_INIT_DCS_CMD(0x31, 0x02),
+	_INIT_DCS_CMD(0x32, 0xDF),
+	_INIT_DCS_CMD(0x33, 0x5A),
+	_INIT_DCS_CMD(0x34, 0xC0),
+	_INIT_DCS_CMD(0x35, 0x5A),
+	_INIT_DCS_CMD(0x36, 0xC0),
+	_INIT_DCS_CMD(0x38, 0x65),
+	_INIT_DCS_CMD(0x80, 0x3E),
+	_INIT_DCS_CMD(0x81, 0xA0),
+	_INIT_DCS_CMD(0xB0, 0x01),
+	_INIT_DCS_CMD(0xB1, 0xCC),
+	_INIT_DCS_CMD(0xC0, 0x12),
+	_INIT_DCS_CMD(0xC2, 0xCC),
+	_INIT_DCS_CMD(0xC3, 0xCC),
+	_INIT_DCS_CMD(0xC4, 0xCC),
+	_INIT_DCS_CMD(0xC5, 0xCC),
+	_INIT_DCS_CMD(0xC6, 0xCC),
+	_INIT_DCS_CMD(0xC7, 0xCC),
+	_INIT_DCS_CMD(0xC8, 0xCC),
+	_INIT_DCS_CMD(0xC9, 0xCC),
+	_INIT_DCS_CMD(0x30, 0x00),
+	_INIT_DCS_CMD(0x00, 0x81),
+	_INIT_DCS_CMD(0x08, 0x02),
+	_INIT_DCS_CMD(0x09, 0x00),
+	_INIT_DCS_CMD(0x07, 0x21),
+	_INIT_DCS_CMD(0x04, 0x10),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x1E),
+	_INIT_DCS_CMD(0x60, 0x00),
+	_INIT_DCS_CMD(0x64, 0x00),
+	_INIT_DCS_CMD(0x6D, 0x00),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x0B),
+	_INIT_DCS_CMD(0xA6, 0x44),
+	_INIT_DCS_CMD(0xA7, 0xB6),
+	_INIT_DCS_CMD(0xA8, 0x03),
+	_INIT_DCS_CMD(0xA9, 0x03),
+	_INIT_DCS_CMD(0xAA, 0x51),
+	_INIT_DCS_CMD(0xAB, 0x51),
+	_INIT_DCS_CMD(0xAC, 0x04),
+	_INIT_DCS_CMD(0xBD, 0x92),
+	_INIT_DCS_CMD(0xBE, 0xA1),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x05),
+	_INIT_DCS_CMD(0x86, 0x87),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x06),
+	_INIT_DCS_CMD(0x92, 0x22),
+
+	_INIT_DCS_CMD(0xFF, 0x98, 0x82, 0x00),
+	_INIT_DCS_CMD(0x11),
+	_INIT_DELAY_CMD(120),
+	_INIT_DCS_CMD(0x29),
+	_INIT_DELAY_CMD(20),
+	{},
+};
+
 static inline struct boe_panel *to_boe_panel(struct drm_panel *panel)
 {
 	return container_of(panel, struct boe_panel, base);
@@ -1796,6 +2136,34 @@ static const struct panel_desc starry_himax83102_j02_desc = {
 	.lp11_before_reset = true,
 };
 
+static const struct drm_display_mode starry_ili9882t_default_mode = {
+	.clock = 165280,
+	.hdisplay = 1200,
+	.hsync_start = 1200 + 32,
+	.hsync_end = 1200 + 32 + 30,
+	.htotal = 1200 + 32 + 30 + 32,
+	.vdisplay = 1920,
+	.vsync_start = 1920 + 68,
+	.vsync_end = 1920 + 68 + 2,
+	.vtotal = 1920 + 68 + 2 + 10,
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static const struct panel_desc starry_ili9882t_desc = {
+	.modes = &starry_ili9882t_default_mode,
+	.bpc = 8,
+	.size = {
+		.width_mm = 141,
+		.height_mm = 226,
+	},
+	.lanes = 4,
+	.format = MIPI_DSI_FMT_RGB888,
+	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE |
+		      MIPI_DSI_MODE_LPM,
+	.init_cmds = starry_ili9882t_init_cmd,
+	.lp11_before_reset = true,
+};
+
 static int boe_panel_get_modes(struct drm_panel *panel,
 			       struct drm_connector *connector)
 {
@@ -1972,6 +2340,9 @@ static const struct of_device_id boe_of_match[] = {
 	{ .compatible = "starry,himax83102-j02",
 	  .data = &starry_himax83102_j02_desc
 	},
+	{ .compatible = "starry,ili9882t",
+	  .data = &starry_ili9882t_desc
+	},
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, boe_of_match);
-- 
GitLab


From edad9ee94f17adc75d3b13ab51bbe3d615ce1e7e Mon Sep 17 00:00:00 2001
From: Andi Shyti <andi.shyti@linux.intel.com>
Date: Fri, 26 May 2023 14:41:38 +0200
Subject: [PATCH 0731/1791] drm/i915/gt: Use the correct error value when
 kernel_context() fails

kernel_context() returns an error pointer. Use pointer-error
conversion functions to evaluate its return value, rather than
checking for a '0' return.

Fixes: eb5c10cbbc2f ("drm/i915: Remove I915_USER_PRIORITY_SHIFT")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: <stable@vger.kernel.org> # v5.13+
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526124138.2006110-1-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/gt/selftest_execlists.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_execlists.c b/drivers/gpu/drm/i915/gt/selftest_execlists.c
index 736b89a8ecf54..4202df5b8c122 100644
--- a/drivers/gpu/drm/i915/gt/selftest_execlists.c
+++ b/drivers/gpu/drm/i915/gt/selftest_execlists.c
@@ -1530,8 +1530,8 @@ static int live_busywait_preempt(void *arg)
 	struct drm_i915_gem_object *obj;
 	struct i915_vma *vma;
 	enum intel_engine_id id;
-	int err = -ENOMEM;
 	u32 *map;
+	int err;
 
 	/*
 	 * Verify that even without HAS_LOGICAL_RING_PREEMPTION, we can
@@ -1539,13 +1539,17 @@ static int live_busywait_preempt(void *arg)
 	 */
 
 	ctx_hi = kernel_context(gt->i915, NULL);
-	if (!ctx_hi)
-		return -ENOMEM;
+	if (IS_ERR(ctx_hi))
+		return PTR_ERR(ctx_hi);
+
 	ctx_hi->sched.priority = I915_CONTEXT_MAX_USER_PRIORITY;
 
 	ctx_lo = kernel_context(gt->i915, NULL);
-	if (!ctx_lo)
+	if (IS_ERR(ctx_lo)) {
+		err = PTR_ERR(ctx_lo);
 		goto err_ctx_hi;
+	}
+
 	ctx_lo->sched.priority = I915_CONTEXT_MIN_USER_PRIORITY;
 
 	obj = i915_gem_object_create_internal(gt->i915, PAGE_SIZE);
-- 
GitLab


From 4722e2ebe6f2168309b285977c5c96baf910c57b Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Tue, 30 May 2023 11:24:38 -0700
Subject: [PATCH 0732/1791] drm/i915/gt: Fix second parameter type of pre-gen8
 pte_encode callbacks

When booting a kernel compiled with CONFIG_CFI_CLANG (kCFI), there is a
CFI failure in ggtt_probe_common() when trying to call hsw_pte_encode()
via an indirect call:

  [    5.030027] CFI failure at ggtt_probe_common+0xd1/0x130 [i915] (target: hsw_pte_encode+0x0/0x30 [i915]; expected type: 0xf5c1d0fc)

With kCFI, indirect calls are validated against their expected type
versus actual type and failures occur when the two types do not match.

clang's -Wincompatible-function-pointer-types-strict can catch this at
compile time but it is not enabled for the kernel yet:

  drivers/gpu/drm/i915/gt/intel_ggtt.c:1155:23: error: incompatible function pointer types assigning to 'u64 (*)(dma_addr_t, unsigned int, u32)' (aka 'unsigned long long (*)(unsigned int, unsigned int, unsigned int)') from 'u64 (dma_addr_t,
  enum i915_cache_level, u32)' (aka 'unsigned long long (unsigned int, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
                  ggtt->vm.pte_encode = iris_pte_encode;
                                      ^ ~~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/gt/intel_ggtt.c:1157:23: error: incompatible function pointer types assigning to 'u64 (*)(dma_addr_t, unsigned int, u32)' (aka 'unsigned long long (*)(unsigned int, unsigned int, unsigned int)') from 'u64 (dma_addr_t,
  enum i915_cache_level, u32)' (aka 'unsigned long long (unsigned int, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
                  ggtt->vm.pte_encode = hsw_pte_encode;
                                      ^ ~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/gt/intel_ggtt.c:1159:23: error: incompatible function pointer types assigning to 'u64 (*)(dma_addr_t, unsigned int, u32)' (aka 'unsigned long long (*)(unsigned int, unsigned int, unsigned int)') from 'u64 (dma_addr_t,
  enum i915_cache_level, u32)' (aka 'unsigned long long (unsigned int, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
                  ggtt->vm.pte_encode = byt_pte_encode;
                                      ^ ~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/gt/intel_ggtt.c:1161:23: error: incompatible function pointer types assigning to 'u64 (*)(dma_addr_t, unsigned int, u32)' (aka 'unsigned long long (*)(unsigned int, unsigned int, unsigned int)') from 'u64 (dma_addr_t,
  enum i915_cache_level, u32)' (aka 'unsigned long long (unsigned int, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
                  ggtt->vm.pte_encode = ivb_pte_encode;
                                      ^ ~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/gt/intel_ggtt.c:1163:23: error: incompatible function pointer types assigning to 'u64 (*)(dma_addr_t, unsigned int, u32)' (aka 'unsigned long long (*)(unsigned int, unsigned int, unsigned int)') from 'u64 (dma_addr_t,
  enum i915_cache_level, u32)' (aka 'unsigned long long (unsigned int, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
                  ggtt->vm.pte_encode = snb_pte_encode;
                                      ^ ~~~~~~~~~~~~~~
  5 errors generated.

In this case, the pre-gen8 pte_encode functions have a second parameter
type of 'enum i915_cache_level' whereas the function pointer prototype
in 'struct i915_address_space' expects a second parameter type of
'unsigned int'.

Update the second parameter of the callbacks and the comment above them
noting that these statements are still valid, which matches other
functions and files, to clear up the kCFI failures at run time.

Fixes: 9275277d5324 ("drm/i915: use pat_index instead of cache_level")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Fei Yang <fei.yang@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530-i915-gt-cache_level-wincompatible-function-pointer-types-strict-v1-1-54501d598229@kernel.org
---
 drivers/gpu/drm/i915/gt/intel_ggtt.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index 2a7942fac7980..122197737ef2e 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -1015,16 +1015,16 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt)
 
 /*
  * For pre-gen8 platforms pat_index is the same as enum i915_cache_level,
- * so these PTE encode functions are left with using cache_level.
+ * so the switch-case statements in these PTE encode functions are still valid.
  * See translation table LEGACY_CACHELEVEL.
  */
 static u64 snb_pte_encode(dma_addr_t addr,
-			  enum i915_cache_level level,
+			  unsigned int pat_index,
 			  u32 flags)
 {
 	gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID;
 
-	switch (level) {
+	switch (pat_index) {
 	case I915_CACHE_L3_LLC:
 	case I915_CACHE_LLC:
 		pte |= GEN6_PTE_CACHE_LLC;
@@ -1033,19 +1033,19 @@ static u64 snb_pte_encode(dma_addr_t addr,
 		pte |= GEN6_PTE_UNCACHED;
 		break;
 	default:
-		MISSING_CASE(level);
+		MISSING_CASE(pat_index);
 	}
 
 	return pte;
 }
 
 static u64 ivb_pte_encode(dma_addr_t addr,
-			  enum i915_cache_level level,
+			  unsigned int pat_index,
 			  u32 flags)
 {
 	gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID;
 
-	switch (level) {
+	switch (pat_index) {
 	case I915_CACHE_L3_LLC:
 		pte |= GEN7_PTE_CACHE_L3_LLC;
 		break;
@@ -1056,14 +1056,14 @@ static u64 ivb_pte_encode(dma_addr_t addr,
 		pte |= GEN6_PTE_UNCACHED;
 		break;
 	default:
-		MISSING_CASE(level);
+		MISSING_CASE(pat_index);
 	}
 
 	return pte;
 }
 
 static u64 byt_pte_encode(dma_addr_t addr,
-			  enum i915_cache_level level,
+			  unsigned int pat_index,
 			  u32 flags)
 {
 	gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID;
@@ -1071,31 +1071,31 @@ static u64 byt_pte_encode(dma_addr_t addr,
 	if (!(flags & PTE_READ_ONLY))
 		pte |= BYT_PTE_WRITEABLE;
 
-	if (level != I915_CACHE_NONE)
+	if (pat_index != I915_CACHE_NONE)
 		pte |= BYT_PTE_SNOOPED_BY_CPU_CACHES;
 
 	return pte;
 }
 
 static u64 hsw_pte_encode(dma_addr_t addr,
-			  enum i915_cache_level level,
+			  unsigned int pat_index,
 			  u32 flags)
 {
 	gen6_pte_t pte = HSW_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID;
 
-	if (level != I915_CACHE_NONE)
+	if (pat_index != I915_CACHE_NONE)
 		pte |= HSW_WB_LLC_AGE3;
 
 	return pte;
 }
 
 static u64 iris_pte_encode(dma_addr_t addr,
-			   enum i915_cache_level level,
+			   unsigned int pat_index,
 			   u32 flags)
 {
 	gen6_pte_t pte = HSW_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID;
 
-	switch (level) {
+	switch (pat_index) {
 	case I915_CACHE_NONE:
 		break;
 	case I915_CACHE_WT:
-- 
GitLab


From 1baeef6cd2229e01091c69cef042f6b688e194be Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Tue, 30 May 2023 11:24:39 -0700
Subject: [PATCH 0733/1791] drm/i915/gt: Fix parameter in
 gmch_ggtt_insert_{entries, page}()

When building with clang's -Wincompatible-function-pointer-types-strict,
the following warnings occur:

  drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c:102:23: error: incompatible function pointer types assigning to 'void (*)(struct i915_address_space *, dma_addr_t, u64, unsigned int, u32)' (aka 'void (*)(struct i915_address_space *, unsigned int, unsigned long long, unsigned int, unsigned int)') from 'void (struct i915_address_space *, dma_addr_t, u64, enum i915_cache_level, u32)' (aka 'void (struct i915_address_space *, unsigned int, unsigned long long, enum i915_cache_level, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
          ggtt->vm.insert_page = gmch_ggtt_insert_page;
                               ^ ~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c:103:26: error: incompatible function pointer types assigning to 'void (*)(struct i915_address_space *, struct i915_vma_resource *, unsigned int, u32)' (aka 'void (*)(struct i915_address_space *, struct i915_vma_resource *, unsigned int, unsigned int)') from 'void (struct i915_address_space *, struct i915_vma_resource *, enum i915_cache_level, u32)' (aka 'void (struct i915_address_space *, struct i915_vma_resource *, enum i915_cache_level, unsigned int)') [-Werror, -Wincompatible-function-pointer-types-strict]
          ggtt->vm.insert_entries = gmch_ggtt_insert_entries;
                                  ^ ~~~~~~~~~~~~~~~~~~~~~~~~
  2 errors generated.

The warning is pointing out that while 'enum i915_cache_level' and
'unsigned int' are ABI compatible, these indirect calls will fail
clang's kernel Control Flow Integrity (kCFI) checks, as the callback's
signature does not exactly match the prototype's signature.

To fix this, replace the cache_level parameter with pat_index, as was
done in other places within i915 where there is no difference between
cache_level and pat_index on certain generations.

Fixes: 9275277d5324 ("drm/i915: use pat_index instead of cache_level")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Fei Yang <fei.yang@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530-i915-gt-cache_level-wincompatible-function-pointer-types-strict-v1-2-54501d598229@kernel.org
---
 drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
index d6a74ae2527bd..866c416afb734 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c
@@ -18,10 +18,10 @@
 static void gmch_ggtt_insert_page(struct i915_address_space *vm,
 				  dma_addr_t addr,
 				  u64 offset,
-				  enum i915_cache_level cache_level,
+				  unsigned int pat_index,
 				  u32 unused)
 {
-	unsigned int flags = (cache_level == I915_CACHE_NONE) ?
+	unsigned int flags = (pat_index == I915_CACHE_NONE) ?
 		AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
 
 	intel_gmch_gtt_insert_page(addr, offset >> PAGE_SHIFT, flags);
@@ -29,10 +29,10 @@ static void gmch_ggtt_insert_page(struct i915_address_space *vm,
 
 static void gmch_ggtt_insert_entries(struct i915_address_space *vm,
 				     struct i915_vma_resource *vma_res,
-				     enum i915_cache_level cache_level,
+				     unsigned int pat_index,
 				     u32 unused)
 {
-	unsigned int flags = (cache_level == I915_CACHE_NONE) ?
+	unsigned int flags = (pat_index == I915_CACHE_NONE) ?
 		AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY;
 
 	intel_gmch_gtt_insert_sg_entries(vma_res->bi.pages, vma_res->start >> PAGE_SHIFT,
-- 
GitLab


From 0374ffa507d8106202a956f5951226f1d9eac22e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:13 +0300
Subject: [PATCH 0734/1791] drm/edid: parse display info has_audio similar to
 is_hdmi

Since we already iterate everything that's needed for determining audio,
reduce the need to call drm_detect_monitor_audio() by storing has_audio
to connector info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/391a93b25c6bcbb39854aaa2813570cfb1580ed9.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/drm_edid.c  | 6 ++++++
 include/drm/drm_connector.h | 8 ++++++++
 2 files changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 0454da505687b..813ce00a106ae 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -6243,6 +6243,9 @@ static void drm_parse_cea_ext(struct drm_connector *connector,
 			info->color_formats |= DRM_COLOR_FORMAT_YCBCR444;
 		if (edid_ext[3] & EDID_CEA_YCRCB422)
 			info->color_formats |= DRM_COLOR_FORMAT_YCBCR422;
+		if (edid_ext[3] & EDID_BASIC_AUDIO)
+			info->has_audio = true;
+
 	}
 	drm_edid_iter_end(&edid_iter);
 
@@ -6268,6 +6271,8 @@ static void drm_parse_cea_ext(struct drm_connector *connector,
 			drm_parse_hdr_metadata_block(connector, data);
 		else if (cea_db_tag(db) == CTA_DB_VIDEO)
 			parse_cta_vdb(connector, db);
+		else if (cea_db_tag(db) == CTA_DB_AUDIO)
+			info->has_audio = true;
 	}
 	cea_db_iter_end(&iter);
 
@@ -6424,6 +6429,7 @@ static void drm_reset_display_info(struct drm_connector *connector)
 	info->max_tmds_clock = 0;
 	info->dvi_dual = false;
 	info->is_hdmi = false;
+	info->has_audio = false;
 	info->has_hdmi_infoframe = false;
 	info->rgb_quant_range_selectable = false;
 	memset(&info->hdmi, 0, sizeof(info->hdmi));
diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index e6478fafa6b04..e143fef07de9f 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -658,6 +658,14 @@ struct drm_display_info {
 	 */
 	bool is_hdmi;
 
+	/**
+	 * @has_audio: True if the sink supports audio.
+	 *
+	 * This field shall be used instead of calling
+	 * drm_detect_monitor_audio() when possible.
+	 */
+	bool has_audio;
+
 	/**
 	 * @has_hdmi_infoframe: Does the sink support the HDMI infoframe?
 	 */
-- 
GitLab


From 3d35ddfb07136cb4220fd9672589dcdf5c9d4acf Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:14 +0300
Subject: [PATCH 0735/1791] drm/display/dp_mst: drop has_audio from struct
 drm_dp_mst_port

Caching the has_audio in struct drm_dp_mst_port seems odd, and oddly
placed. Defer audio handling to drivers, and use the info from the
connector display info. i915 was the only one using it anyway.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/7d9eefdf150870479c5797f027d4c2b0a19ff583.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/display/drm_dp_mst_topology.c        | 2 +-
 drivers/gpu/drm/i915/display/intel_display_debugfs.c | 4 ++--
 drivers/gpu/drm/i915/display/intel_dp_mst.c          | 2 +-
 include/drm/display/drm_dp_mst_helper.h              | 5 -----
 4 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c
index 38dab76ae69ea..a2b8732db0c8a 100644
--- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
@@ -4170,7 +4170,7 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_
 	else {
 		edid = drm_get_edid(connector, &port->aux.ddc);
 	}
-	port->has_audio = drm_detect_monitor_audio(edid);
+
 	drm_dp_mst_topology_put_port(port);
 	return edid;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index e72288662f022..77caed2552d06 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -244,9 +244,9 @@ static void intel_dp_info(struct seq_file *m,
 }
 
 static void intel_dp_mst_info(struct seq_file *m,
-			      struct intel_connector *intel_connector)
+			      struct intel_connector *connector)
 {
-	bool has_audio = intel_connector->port->has_audio;
+	bool has_audio = connector->base.display_info.has_audio;
 
 	seq_printf(m, "\taudio support: %s\n", str_yes_no(has_audio));
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 6df8519b3347e..1d483a83d59c6 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -298,7 +298,7 @@ static bool intel_dp_mst_has_audio(const struct drm_connector_state *conn_state)
 		to_intel_connector(conn_state->connector);
 
 	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-		return connector->port->has_audio;
+		return connector->base.display_info.has_audio;
 	else
 		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
 }
diff --git a/include/drm/display/drm_dp_mst_helper.h b/include/drm/display/drm_dp_mst_helper.h
index 32c764fb9cb56..5be96a158ab29 100644
--- a/include/drm/display/drm_dp_mst_helper.h
+++ b/include/drm/display/drm_dp_mst_helper.h
@@ -139,11 +139,6 @@ struct drm_dp_mst_port {
 	 * that the EDID for all connectors is read immediately.
 	 */
 	struct edid *cached_edid;
-	/**
-	 * @has_audio: Tracks whether the sink connector to this port is
-	 * audio-capable.
-	 */
-	bool has_audio;
 
 	/**
 	 * @fec_capable: bool indicating if FEC can be supported up to that
-- 
GitLab


From 28da4f8336083874699defb5016cda3015ef2723 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:15 +0300
Subject: [PATCH 0736/1791] drm/i915/dp: stop caching has_audio in struct
 intel_dp

Use the information stored in display info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/695bbe8b5ebee6e492f95a1c818da15691752dc8.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_debugfs.c | 2 +-
 drivers/gpu/drm/i915/display/intel_display_types.h   | 1 -
 drivers/gpu/drm/i915/display/intel_dp.c              | 5 ++---
 3 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index 77caed2552d06..1416b82fc8a36 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -237,7 +237,7 @@ static void intel_dp_info(struct seq_file *m,
 
 	seq_printf(m, "\tDPCD rev: %x\n", intel_dp->dpcd[DP_DPCD_REV]);
 	seq_printf(m, "\taudio support: %s\n",
-		   str_yes_no(intel_dp->has_audio));
+		   str_yes_no(intel_connector->base.display_info.has_audio));
 
 	drm_dp_downstream_debug(m, intel_dp->dpcd, intel_dp->downstream_ports,
 				edid ? edid->data : NULL, &intel_dp->aux);
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 2d8297f8d0889..94a59febc9873 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1708,7 +1708,6 @@ struct intel_dp {
 	u8 sink_count;
 	bool link_trained;
 	bool has_hdmi_sink;
-	bool has_audio;
 	bool reset_link_params;
 	bool use_max_params;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 4bec8cd7979fd..aa1445ad9da88 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -2159,6 +2159,7 @@ static bool intel_dp_has_audio(struct intel_encoder *encoder,
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+	struct intel_connector *connector = intel_dp->attached_connector;
 	const struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(conn_state);
 
@@ -2166,7 +2167,7 @@ static bool intel_dp_has_audio(struct intel_encoder *encoder,
 		return false;
 
 	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-		return intel_dp->has_audio;
+		return connector->base.display_info.has_audio;
 	else
 		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
 }
@@ -4813,7 +4814,6 @@ intel_dp_set_edid(struct intel_dp *intel_dp)
 	edid = drm_edid_raw(drm_edid);
 	if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) {
 		intel_dp->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
-		intel_dp->has_audio = drm_detect_monitor_audio(edid);
 	}
 
 	drm_dp_cec_set_edid(&intel_dp->aux, edid);
@@ -4829,7 +4829,6 @@ intel_dp_unset_edid(struct intel_dp *intel_dp)
 	connector->detect_edid = NULL;
 
 	intel_dp->has_hdmi_sink = false;
-	intel_dp->has_audio = false;
 
 	intel_dp->dfp.max_bpc = 0;
 	intel_dp->dfp.max_dotclock = 0;
-- 
GitLab


From 7ffa2f27cf0aa1a89bea7fdacd9a51668ec32e7b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:16 +0300
Subject: [PATCH 0737/1791] drm/i915/dp: stop caching has_hdmi_sink in struct
 intel_dp

Use the information stored in display info. Add intel_dp_has_hdmi_sink()
helper to access it.

v2: Rebased

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/748103fda572b3552e5bbdafb300d8508d4eeaf4.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c      |  6 +++---
 .../drm/i915/display/intel_display_types.h    |  1 -
 drivers/gpu/drm/i915/display/intel_dp.c       | 21 ++++++++++---------
 drivers/gpu/drm/i915/display/intel_dp.h       |  1 +
 4 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index d1a9a3cf94b54..70d44edd8c6e1 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2801,7 +2801,7 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state,
 
 		/* FIXME precompute everything properly */
 		/* FIXME how do we turn infoframes off again? */
-		if (dig_port->lspcon.active && dig_port->dp.has_hdmi_sink)
+		if (dig_port->lspcon.active && intel_dp_has_hdmi_sink(&dig_port->dp))
 			dig_port->set_infoframes(encoder,
 						 crtc_state->has_infoframe,
 						 crtc_state, conn_state);
@@ -3110,7 +3110,7 @@ static void intel_enable_ddi_dp(struct intel_atomic_state *state,
 	drm_connector_update_privacy_screen(conn_state);
 	intel_edp_backlight_on(crtc_state, conn_state);
 
-	if (!dig_port->lspcon.active || dig_port->dp.has_hdmi_sink)
+	if (!dig_port->lspcon.active || intel_dp_has_hdmi_sink(&dig_port->dp))
 		intel_dp_set_infoframes(encoder, true, crtc_state, conn_state);
 
 	intel_audio_codec_enable(encoder, crtc_state, conn_state);
@@ -3738,7 +3738,7 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder,
 				    pipe_config->fec_enable);
 		}
 
-		if (dig_port->lspcon.active && dig_port->dp.has_hdmi_sink)
+		if (dig_port->lspcon.active && intel_dp_has_hdmi_sink(&dig_port->dp))
 			pipe_config->infoframes.enable |=
 				intel_lspcon_infoframes_enabled(encoder, pipe_config);
 		else
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 94a59febc9873..c5bd0c7be2fa8 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1707,7 +1707,6 @@ struct intel_dp {
 	u8 lane_count;
 	u8 sink_count;
 	bool link_trained;
-	bool has_hdmi_sink;
 	bool reset_link_params;
 	bool use_max_params;
 	u8 dpcd[DP_RECEIVER_CAP_SIZE];
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index aa1445ad9da88..8445a37d9d01f 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1287,6 +1287,13 @@ void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
 	}
 }
 
+bool intel_dp_has_hdmi_sink(struct intel_dp *intel_dp)
+{
+	struct intel_connector *connector = intel_dp->attached_connector;
+
+	return connector->base.display_info.is_hdmi;
+}
+
 static bool intel_dp_source_supports_fec(struct intel_dp *intel_dp,
 					 const struct intel_crtc_state *pipe_config)
 {
@@ -1342,7 +1349,7 @@ static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp,
 
 	for (; bpc >= 8; bpc -= 2) {
 		if (intel_hdmi_bpc_possible(crtc_state, bpc,
-					    intel_dp->has_hdmi_sink) &&
+					    intel_dp_has_hdmi_sink(intel_dp)) &&
 		    intel_dp_tmds_clock_valid(intel_dp, clock, bpc, crtc_state->sink_format,
 					      respect_downstream_limits) == MODE_OK)
 			return bpc;
@@ -2732,7 +2739,7 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp)
 static bool intel_dp_is_hdmi_2_1_sink(struct intel_dp *intel_dp)
 {
 	if (drm_dp_is_branch(intel_dp->dpcd) &&
-	    intel_dp->has_hdmi_sink &&
+	    intel_dp_has_hdmi_sink(intel_dp) &&
 	    intel_dp_hdmi_sink_max_frl(intel_dp) > 0)
 		return true;
 
@@ -2900,13 +2907,12 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp,
 	if (!drm_dp_is_branch(intel_dp->dpcd))
 		return;
 
-	tmp = intel_dp->has_hdmi_sink ?
-		DP_HDMI_DVI_OUTPUT_CONFIG : 0;
+	tmp = intel_dp_has_hdmi_sink(intel_dp) ? DP_HDMI_DVI_OUTPUT_CONFIG : 0;
 
 	if (drm_dp_dpcd_writeb(&intel_dp->aux,
 			       DP_PROTOCOL_CONVERTER_CONTROL_0, tmp) != 1)
 		drm_dbg_kms(&i915->drm, "Failed to %s protocol converter HDMI mode\n",
-			    str_enable_disable(intel_dp->has_hdmi_sink));
+			    str_enable_disable(intel_dp_has_hdmi_sink(intel_dp)));
 
 	if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) {
 		switch (crtc_state->output_format) {
@@ -4812,9 +4818,6 @@ intel_dp_set_edid(struct intel_dp *intel_dp)
 
 	/* FIXME: Get rid of drm_edid_raw() */
 	edid = drm_edid_raw(drm_edid);
-	if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) {
-		intel_dp->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
-	}
 
 	drm_dp_cec_set_edid(&intel_dp->aux, edid);
 }
@@ -4828,8 +4831,6 @@ intel_dp_unset_edid(struct intel_dp *intel_dp)
 	drm_edid_free(connector->detect_edid);
 	connector->detect_edid = NULL;
 
-	intel_dp->has_hdmi_sink = false;
-
 	intel_dp->dfp.max_bpc = 0;
 	intel_dp->dfp.max_dotclock = 0;
 	intel_dp->dfp.min_tmds_clock = 0;
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index 5f86157a10d2d..22099de3ca458 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -65,6 +65,7 @@ int intel_dp_dsc_compute_config(struct intel_dp *intel_dp,
 				struct link_config_limits *limits,
 				int timeslots,
 				bool recompute_pipe_bpp);
+bool intel_dp_has_hdmi_sink(struct intel_dp *intel_dp);
 bool intel_dp_is_edp(struct intel_dp *intel_dp);
 bool intel_dp_is_uhbr(const struct intel_crtc_state *crtc_state);
 bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port);
-- 
GitLab


From 9fe2a4adc09567e32a5ecd4cab236022436a16c2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:17 +0300
Subject: [PATCH 0738/1791] drm/i915/hdmi: stop caching has_audio in struct
 intel_hdmi

Use the information stored in display info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/710286536d6b07ba8aa068b65b2b2c0c7743220e.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_debugfs.c | 8 +++-----
 drivers/gpu/drm/i915/display/intel_display_types.h   | 1 -
 drivers/gpu/drm/i915/display/intel_hdmi.c            | 6 ++----
 3 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index 1416b82fc8a36..2a4df62692a68 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -252,13 +252,11 @@ static void intel_dp_mst_info(struct seq_file *m,
 }
 
 static void intel_hdmi_info(struct seq_file *m,
-			    struct intel_connector *intel_connector)
+			    struct intel_connector *connector)
 {
-	struct intel_encoder *intel_encoder = intel_attached_encoder(intel_connector);
-	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(intel_encoder);
+	bool has_audio = connector->base.display_info.has_audio;
 
-	seq_printf(m, "\taudio support: %s\n",
-		   str_yes_no(intel_hdmi->has_audio));
+	seq_printf(m, "\taudio support: %s\n", str_yes_no(has_audio));
 }
 
 static void intel_connector_info(struct seq_file *m,
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index c5bd0c7be2fa8..1ae1519087cbe 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1587,7 +1587,6 @@ struct intel_hdmi {
 		int max_tmds_clock;
 	} dp_dual_mode;
 	bool has_hdmi_sink;
-	bool has_audio;
 	struct intel_connector *attached_connector;
 	struct cec_notifier *cec_notifier;
 };
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index c2464f98c76ef..89a6b574f95a4 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -2169,7 +2169,7 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state,
 				 const struct drm_connector_state *conn_state)
 {
-	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+	struct drm_connector *connector = conn_state->connector;
 	const struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(conn_state);
 
@@ -2177,7 +2177,7 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder,
 		return false;
 
 	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-		return intel_hdmi->has_audio;
+		return connector->display_info.has_audio;
 	else
 		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
 }
@@ -2386,7 +2386,6 @@ intel_hdmi_unset_edid(struct drm_connector *connector)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector));
 
 	intel_hdmi->has_hdmi_sink = false;
-	intel_hdmi->has_audio = false;
 
 	intel_hdmi->dp_dual_mode.type = DRM_DP_DUAL_MODE_NONE;
 	intel_hdmi->dp_dual_mode.max_tmds_clock = 0;
@@ -2479,7 +2478,6 @@ intel_hdmi_set_edid(struct drm_connector *connector)
 	/* FIXME: Get rid of drm_edid_raw() */
 	edid = drm_edid_raw(drm_edid);
 	if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) {
-		intel_hdmi->has_audio = drm_detect_monitor_audio(edid);
 		intel_hdmi->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
 
 		intel_hdmi_dp_dual_mode_detect(connector);
-- 
GitLab


From 7c1000aae5521ea4fa027b6ddf6aa9d4863e9c29 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:18 +0300
Subject: [PATCH 0739/1791] drm/i915/hdmi: stop caching has_hdmi_sink in struct
 intel_hdmi

Use the information stored in display info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/5cdb4731260b062c0f0ad2f8b64195c9a57bcb76.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_types.h | 1 -
 drivers/gpu/drm/i915/display/intel_hdmi.c          | 8 +++-----
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 1ae1519087cbe..731f2ec04d5cd 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1586,7 +1586,6 @@ struct intel_hdmi {
 		enum drm_dp_dual_mode_type type;
 		int max_tmds_clock;
 	} dp_dual_mode;
-	bool has_hdmi_sink;
 	struct intel_connector *attached_connector;
 	struct cec_notifier *cec_notifier;
 };
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 89a6b574f95a4..7ac5e6c5e00dc 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -1790,7 +1790,9 @@ static int intel_hdmi_source_max_tmds_clock(struct intel_encoder *encoder)
 static bool intel_has_hdmi_sink(struct intel_hdmi *hdmi,
 				const struct drm_connector_state *conn_state)
 {
-	return hdmi->has_hdmi_sink &&
+	struct intel_connector *connector = hdmi->attached_connector;
+
+	return connector->base.display_info.is_hdmi &&
 		READ_ONCE(to_intel_digital_connector_state(conn_state)->force_audio) != HDMI_AUDIO_OFF_DVI;
 }
 
@@ -2385,8 +2387,6 @@ intel_hdmi_unset_edid(struct drm_connector *connector)
 {
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector));
 
-	intel_hdmi->has_hdmi_sink = false;
-
 	intel_hdmi->dp_dual_mode.type = DRM_DP_DUAL_MODE_NONE;
 	intel_hdmi->dp_dual_mode.max_tmds_clock = 0;
 
@@ -2478,8 +2478,6 @@ intel_hdmi_set_edid(struct drm_connector *connector)
 	/* FIXME: Get rid of drm_edid_raw() */
 	edid = drm_edid_raw(drm_edid);
 	if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) {
-		intel_hdmi->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
-
 		intel_hdmi_dp_dual_mode_detect(connector);
 
 		connected = true;
-- 
GitLab


From f2f9c8cb6421429ef166d6404426693212d0ca07 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:19 +0300
Subject: [PATCH 0740/1791] drm/i915/sdvo: stop caching has_hdmi_monitor in
 struct intel_sdvo

Use the information stored in display info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/3e9e1dcd554d470bdf474891a431b15e1880f9a0.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_sdvo.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 34ee9dd82a785..5ce1fed84016e 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -117,7 +117,6 @@ struct intel_sdvo {
 
 	enum port port;
 
-	bool has_hdmi_monitor;
 	bool has_hdmi_audio;
 
 	/* DDC bus used by this SDVO encoder */
@@ -1303,10 +1302,13 @@ static void i9xx_adjust_sdvo_tv_clock(struct intel_crtc_state *pipe_config)
 	pipe_config->clock_set = true;
 }
 
-static bool intel_has_hdmi_sink(struct intel_sdvo *sdvo,
+static bool intel_has_hdmi_sink(struct intel_sdvo_connector *intel_sdvo_connector,
 				const struct drm_connector_state *conn_state)
 {
-	return sdvo->has_hdmi_monitor &&
+	struct drm_connector *connector = conn_state->connector;
+
+	return intel_sdvo_connector->is_hdmi &&
+		connector->display_info.is_hdmi &&
 		READ_ONCE(to_intel_digital_connector_state(conn_state)->force_audio) != HDMI_AUDIO_OFF_DVI;
 }
 
@@ -1401,7 +1403,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder,
 	pipe_config->pixel_multiplier =
 		intel_sdvo_get_pixel_multiplier(adjusted_mode);
 
-	pipe_config->has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo, conn_state);
+	pipe_config->has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo_connector, conn_state);
 
 	pipe_config->has_audio =
 		intel_sdvo_has_audio(encoder, pipe_config, conn_state) &&
@@ -1907,7 +1909,7 @@ intel_sdvo_mode_valid(struct drm_connector *connector,
 	struct intel_sdvo_connector *intel_sdvo_connector =
 		to_intel_sdvo_connector(connector);
 	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
-	bool has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo, connector->state);
+	bool has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo_connector, connector->state);
 	int clock = mode->clock;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
@@ -2096,7 +2098,6 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 		if (edid->input & DRM_EDID_INPUT_DIGITAL) {
 			status = connector_status_connected;
 			if (intel_sdvo_connector->is_hdmi) {
-				intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid);
 				intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
 			}
 		} else
@@ -2148,7 +2149,6 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 
 	intel_sdvo->attached_output = response;
 
-	intel_sdvo->has_hdmi_monitor = false;
 	intel_sdvo->has_hdmi_audio = false;
 
 	if ((intel_sdvo_connector->output_flag & response) == 0)
-- 
GitLab


From dd9f377ae12110fef29262af23859d144a613bfa Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:20 +0300
Subject: [PATCH 0741/1791] drm/i915/sdvo: stop caching has_hdmi_audio in
 struct intel_sdvo

Use the information stored in display info.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1e02f2f7381dfcee6e4160a5fc17aea6ff04baf9.1685437500.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_sdvo.c | 20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 5ce1fed84016e..2d1b19f738835 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -117,8 +117,6 @@ struct intel_sdvo {
 
 	enum port port;
 
-	bool has_hdmi_audio;
-
 	/* DDC bus used by this SDVO encoder */
 	u8 ddc_bus;
 
@@ -1328,7 +1326,9 @@ static bool intel_sdvo_has_audio(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state,
 				 const struct drm_connector_state *conn_state)
 {
-	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
+	struct drm_connector *connector = conn_state->connector;
+	struct intel_sdvo_connector *intel_sdvo_connector =
+		to_intel_sdvo_connector(connector);
 	const struct intel_digital_connector_state *intel_conn_state =
 		to_intel_digital_connector_state(conn_state);
 
@@ -1336,7 +1336,8 @@ static bool intel_sdvo_has_audio(struct intel_encoder *encoder,
 		return false;
 
 	if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO)
-		return intel_sdvo->has_hdmi_audio;
+		return intel_sdvo_connector->is_hdmi &&
+			connector->display_info.has_audio;
 	else
 		return intel_conn_state->force_audio == HDMI_AUDIO_ON;
 }
@@ -2057,8 +2058,6 @@ static enum drm_connector_status
 intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 {
 	struct intel_sdvo *intel_sdvo = intel_attached_sdvo(to_intel_connector(connector));
-	struct intel_sdvo_connector *intel_sdvo_connector =
-		to_intel_sdvo_connector(connector);
 	enum drm_connector_status status;
 	struct edid *edid;
 
@@ -2095,12 +2094,9 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 	status = connector_status_unknown;
 	if (edid != NULL) {
 		/* DDC bus is shared, match EDID to connector type */
-		if (edid->input & DRM_EDID_INPUT_DIGITAL) {
+		if (edid->input & DRM_EDID_INPUT_DIGITAL)
 			status = connector_status_connected;
-			if (intel_sdvo_connector->is_hdmi) {
-				intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid);
-			}
-		} else
+		else
 			status = connector_status_disconnected;
 		kfree(edid);
 	}
@@ -2149,8 +2145,6 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 
 	intel_sdvo->attached_output = response;
 
-	intel_sdvo->has_hdmi_audio = false;
-
 	if ((intel_sdvo_connector->output_flag & response) == 0)
 		ret = connector_status_disconnected;
 	else if (IS_TMDS(intel_sdvo_connector))
-- 
GitLab


From 6c46f644cd5aa5b2b18020bef83d67101a98d873 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:21 +0300
Subject: [PATCH 0742/1791] drm/edid: add drm_edid_read_switcheroo()

Add a switcheroo variant to the struct drm_edid based EDID read
functions.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/4ab5ec994670ea50f95c8079c1f1ae915940b00f.1685437501.git.jani.nikula@intel.com
---
 drivers/gpu/drm/drm_edid.c | 29 +++++++++++++++++++++++++++++
 include/drm/drm_edid.h     |  2 ++
 2 files changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 813ce00a106ae..941f3d53a701a 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -2844,6 +2844,35 @@ struct edid *drm_get_edid_switcheroo(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_get_edid_switcheroo);
 
+/**
+ * drm_edid_read_switcheroo - get EDID data for a vga_switcheroo output
+ * @connector: connector we're probing
+ * @adapter: I2C adapter to use for DDC
+ *
+ * Wrapper around drm_edid_read_ddc() for laptops with dual GPUs using one set
+ * of outputs. The wrapper adds the requisite vga_switcheroo calls to
+ * temporarily switch DDC to the GPU which is retrieving EDID.
+ *
+ * Return: Pointer to valid EDID or %NULL if we couldn't find any.
+ */
+const struct drm_edid *drm_edid_read_switcheroo(struct drm_connector *connector,
+						struct i2c_adapter *adapter)
+{
+	struct drm_device *dev = connector->dev;
+	struct pci_dev *pdev = to_pci_dev(dev->dev);
+	const struct drm_edid *drm_edid;
+
+	if (drm_WARN_ON_ONCE(dev, !dev_is_pci(dev->dev)))
+		return NULL;
+
+	vga_switcheroo_lock_ddc(pdev);
+	drm_edid = drm_edid_read_ddc(connector, adapter);
+	vga_switcheroo_unlock_ddc(pdev);
+
+	return drm_edid;
+}
+EXPORT_SYMBOL(drm_edid_read_switcheroo);
+
 /**
  * drm_edid_duplicate - duplicate an EDID and the extensions
  * @edid: EDID to duplicate
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index 571885d32907b..169755d3de190 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -613,6 +613,8 @@ const struct drm_edid *drm_edid_read_ddc(struct drm_connector *connector,
 const struct drm_edid *drm_edid_read_custom(struct drm_connector *connector,
 					    int (*read_block)(void *context, u8 *buf, unsigned int block, size_t len),
 					    void *context);
+const struct drm_edid *drm_edid_read_switcheroo(struct drm_connector *connector,
+						struct i2c_adapter *adapter);
 int drm_edid_connector_update(struct drm_connector *connector,
 			      const struct drm_edid *edid);
 int drm_edid_connector_add_modes(struct drm_connector *connector);
-- 
GitLab


From a036aa1f01e7d1321859613d85570720f041cb07 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:22 +0300
Subject: [PATCH 0743/1791] drm/i915/lvds: switch to drm_edid_read_switcheroo()

Use drm_edid_read_switcheroo() to switch from struct edid to struct
drm_edid.

Reviewed-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/429cd6f23366cb4cace4fa13cf92ad068aca64dd.1685437501.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_lvds.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 1f4349a12a026..3ace56979b70e 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -944,17 +944,8 @@ void intel_lvds_init(struct drm_i915_private *i915)
 	 */
 	mutex_lock(&i915->drm.mode_config.mutex);
 	if (vga_switcheroo_handler_flags() & VGA_SWITCHEROO_CAN_SWITCH_DDC) {
-		const struct edid *edid;
-
-		/* FIXME: Make drm_get_edid_switcheroo() return drm_edid */
-		edid = drm_get_edid_switcheroo(&connector->base,
-					       intel_gmbus_get_adapter(i915, pin));
-		if (edid) {
-			drm_edid = drm_edid_alloc(edid, (edid->extensions + 1) * EDID_LENGTH);
-			kfree(edid);
-		} else {
-			drm_edid = NULL;
-		}
+		drm_edid = drm_edid_read_switcheroo(&connector->base,
+						    intel_gmbus_get_adapter(i915, pin));
 	} else {
 		drm_edid = drm_edid_read_ddc(&connector->base,
 					     intel_gmbus_get_adapter(i915, pin));
-- 
GitLab


From d60d2bccae65e9f85016ca4d6f6ad169c86b1024 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:23 +0300
Subject: [PATCH 0744/1791] drm/edid: make drm_edid_duplicate() safe to call
 with NULL parameter

It's a bit tedious to check for NULL before calling
drm_edid_duplicate(). Make it handle NULL parameter graciously.

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1ae8923d39a5abd0260fdf7f9cc54c5e046e70d3.1685437501.git.jani.nikula@intel.com
---
 drivers/gpu/drm/drm_edid.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 941f3d53a701a..e0dbd9140726b 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -2881,6 +2881,9 @@ EXPORT_SYMBOL(drm_edid_read_switcheroo);
  */
 struct edid *drm_edid_duplicate(const struct edid *edid)
 {
+	if (!edid)
+		return NULL;
+
 	return kmemdup(edid, edid_size(edid), GFP_KERNEL);
 }
 EXPORT_SYMBOL(drm_edid_duplicate);
-- 
GitLab


From c1c9042b20031aa4f86f60bba8f04e304bb9f72a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:24 +0300
Subject: [PATCH 0745/1791] drm/display/dp_mst: convert to struct drm_edid

Convert the topology manager to use struct drm_edid, add
drm_dp_mst_edid_read() that returns drm_edid, and rewrite the old
drm_dp_mst_get_edid() to use it.

Note that the old drm_get_edid() ended up calling
drm_connector_update_edid_property(). This responsibility is now
deferred to drivers, which all do it anyway after calling
drm_dp_mst_edid_read() or drm_dp_mst_get_edid().

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9c32e5c241934093fc4144eed4c01155e1f03af1.1685437501.git.jani.nikula@intel.com
---
 drivers/gpu/drm/display/drm_dp_mst_topology.c | 53 +++++++++++++++----
 include/drm/display/drm_dp_mst_helper.h       |  9 +++-
 2 files changed, 49 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c
index a2b8732db0c8a..be71be95b706c 100644
--- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
@@ -1823,7 +1823,7 @@ static void drm_dp_destroy_port(struct kref *kref)
 		return;
 	}
 
-	kfree(port->cached_edid);
+	drm_edid_free(port->cached_edid);
 
 	/*
 	 * we can't destroy the connector here, as we might be holding the
@@ -2272,8 +2272,8 @@ drm_dp_mst_port_add_connector(struct drm_dp_mst_branch *mstb,
 	if (port->pdt != DP_PEER_DEVICE_NONE &&
 	    drm_dp_mst_is_end_device(port->pdt, port->mcs) &&
 	    port->port_num >= DP_MST_LOGICAL_PORT_0)
-		port->cached_edid = drm_get_edid(port->connector,
-						 &port->aux.ddc);
+		port->cached_edid = drm_edid_read_ddc(port->connector,
+						      &port->aux.ddc);
 
 	drm_connector_register(port->connector);
 	return;
@@ -4133,7 +4133,7 @@ drm_dp_mst_detect_port(struct drm_connector *connector,
 		ret = connector_status_connected;
 		/* for logical ports - cache the EDID */
 		if (port->port_num >= DP_MST_LOGICAL_PORT_0 && !port->cached_edid)
-			port->cached_edid = drm_get_edid(connector, &port->aux.ddc);
+			port->cached_edid = drm_edid_read_ddc(connector, &port->aux.ddc);
 		break;
 	case DP_PEER_DEVICE_DP_LEGACY_CONV:
 		if (port->ldps)
@@ -4147,7 +4147,7 @@ drm_dp_mst_detect_port(struct drm_connector *connector,
 EXPORT_SYMBOL(drm_dp_mst_detect_port);
 
 /**
- * drm_dp_mst_get_edid() - get EDID for an MST port
+ * drm_dp_mst_edid_read() - get EDID for an MST port
  * @connector: toplevel connector to get EDID for
  * @mgr: manager for this port
  * @port: unverified pointer to a port.
@@ -4156,9 +4156,11 @@ EXPORT_SYMBOL(drm_dp_mst_detect_port);
  * It validates the pointer still exists so the caller doesn't require a
  * reference.
  */
-struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port)
+const struct drm_edid *drm_dp_mst_edid_read(struct drm_connector *connector,
+					    struct drm_dp_mst_topology_mgr *mgr,
+					    struct drm_dp_mst_port *port)
 {
-	struct edid *edid = NULL;
+	const struct drm_edid *drm_edid;
 
 	/* we need to search for the port in the mgr in case it's gone */
 	port = drm_dp_mst_topology_get_port_validated(mgr, port);
@@ -4166,12 +4168,41 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_
 		return NULL;
 
 	if (port->cached_edid)
-		edid = drm_edid_duplicate(port->cached_edid);
-	else {
-		edid = drm_get_edid(connector, &port->aux.ddc);
-	}
+		drm_edid = drm_edid_dup(port->cached_edid);
+	else
+		drm_edid = drm_edid_read_ddc(connector, &port->aux.ddc);
 
 	drm_dp_mst_topology_put_port(port);
+
+	return drm_edid;
+}
+EXPORT_SYMBOL(drm_dp_mst_edid_read);
+
+/**
+ * drm_dp_mst_get_edid() - get EDID for an MST port
+ * @connector: toplevel connector to get EDID for
+ * @mgr: manager for this port
+ * @port: unverified pointer to a port.
+ *
+ * This function is deprecated; please use drm_dp_mst_edid_read() instead.
+ *
+ * This returns an EDID for the port connected to a connector,
+ * It validates the pointer still exists so the caller doesn't require a
+ * reference.
+ */
+struct edid *drm_dp_mst_get_edid(struct drm_connector *connector,
+				 struct drm_dp_mst_topology_mgr *mgr,
+				 struct drm_dp_mst_port *port)
+{
+	const struct drm_edid *drm_edid;
+	struct edid *edid;
+
+	drm_edid = drm_dp_mst_edid_read(connector, mgr, port);
+
+	edid = drm_edid_duplicate(drm_edid_raw(drm_edid));
+
+	drm_edid_free(drm_edid);
+
 	return edid;
 }
 EXPORT_SYMBOL(drm_dp_mst_get_edid);
diff --git a/include/drm/display/drm_dp_mst_helper.h b/include/drm/display/drm_dp_mst_helper.h
index 5be96a158ab29..f962e97880b4f 100644
--- a/include/drm/display/drm_dp_mst_helper.h
+++ b/include/drm/display/drm_dp_mst_helper.h
@@ -138,7 +138,7 @@ struct drm_dp_mst_port {
 	 * @cached_edid: for DP logical ports - make tiling work by ensuring
 	 * that the EDID for all connectors is read immediately.
 	 */
-	struct edid *cached_edid;
+	const struct drm_edid *cached_edid;
 
 	/**
 	 * @fec_capable: bool indicating if FEC can be supported up to that
@@ -819,7 +819,12 @@ drm_dp_mst_detect_port(struct drm_connector *connector,
 		       struct drm_dp_mst_topology_mgr *mgr,
 		       struct drm_dp_mst_port *port);
 
-struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port);
+const struct drm_edid *drm_dp_mst_edid_read(struct drm_connector *connector,
+					    struct drm_dp_mst_topology_mgr *mgr,
+					    struct drm_dp_mst_port *port);
+struct edid *drm_dp_mst_get_edid(struct drm_connector *connector,
+				 struct drm_dp_mst_topology_mgr *mgr,
+				 struct drm_dp_mst_port *port);
 
 int drm_dp_get_vc_payload_bw(const struct drm_dp_mst_topology_mgr *mgr,
 			     int link_rate, int link_lane_count);
-- 
GitLab


From 27cbdc6bddc28cdc2a91e24d1ae91405e261fc2a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 30 May 2023 12:08:25 +0300
Subject: [PATCH 0746/1791] drm/i915/display: switch the rest of the connectors
 to struct drm_edid

Convert the remaining uses of struct edid based drm_get_edid(),
drm_connector_update_edid_property() and drm_add_edid_modes() calls to
the struct drm_edid based drm_edid_read_ddc(),
drm_edid_connector_update() and drm_edid_connector_add_modes().

Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b1b53bb9004adaa402e061f7df2caf0eb4723a43.1685437501.git.jani.nikula@intel.com
---
 .../gpu/drm/i915/display/intel_connector.c    | 18 ++---
 .../gpu/drm/i915/display/intel_connector.h    |  4 +-
 drivers/gpu/drm/i915/display/intel_crt.c      | 34 +++++----
 drivers/gpu/drm/i915/display/intel_dp_mst.c   | 10 ++-
 drivers/gpu/drm/i915/display/intel_sdvo.c     | 73 ++++++++++---------
 5 files changed, 74 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c
index 257afac348397..00ea71b03ec78 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.c
+++ b/drivers/gpu/drm/i915/display/intel_connector.c
@@ -176,15 +176,15 @@ enum pipe intel_connector_get_pipe(struct intel_connector *connector)
 /**
  * intel_connector_update_modes - update connector from edid
  * @connector: DRM connector device to use
- * @edid: previously read EDID information
+ * @drm_edid: previously read EDID information
  */
 int intel_connector_update_modes(struct drm_connector *connector,
-				struct edid *edid)
+				 const struct drm_edid *drm_edid)
 {
 	int ret;
 
-	drm_connector_update_edid_property(connector, edid);
-	ret = drm_add_edid_modes(connector, edid);
+	drm_edid_connector_update(connector, drm_edid);
+	ret = drm_edid_connector_add_modes(connector);
 
 	return ret;
 }
@@ -199,15 +199,15 @@ int intel_connector_update_modes(struct drm_connector *connector,
 int intel_ddc_get_modes(struct drm_connector *connector,
 			struct i2c_adapter *adapter)
 {
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 	int ret;
 
-	edid = drm_get_edid(connector, adapter);
-	if (!edid)
+	drm_edid = drm_edid_read_ddc(connector, adapter);
+	if (!drm_edid)
 		return 0;
 
-	ret = intel_connector_update_modes(connector, edid);
-	kfree(edid);
+	ret = intel_connector_update_modes(connector, drm_edid);
+	drm_edid_free(drm_edid);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_connector.h b/drivers/gpu/drm/i915/display/intel_connector.h
index 9d2bc261b2042..aaf7281462dcd 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.h
+++ b/drivers/gpu/drm/i915/display/intel_connector.h
@@ -9,7 +9,7 @@
 #include <linux/types.h>
 
 struct drm_connector;
-struct edid;
+struct drm_edid;
 struct i2c_adapter;
 struct intel_connector;
 struct intel_encoder;
@@ -25,7 +25,7 @@ void intel_connector_attach_encoder(struct intel_connector *connector,
 bool intel_connector_get_hw_state(struct intel_connector *connector);
 enum pipe intel_connector_get_pipe(struct intel_connector *connector);
 int intel_connector_update_modes(struct drm_connector *connector,
-				 struct edid *edid);
+				 const struct drm_edid *drm_edid);
 int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
 void intel_attach_force_audio_property(struct drm_connector *connector);
 void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 673c036466960..ab7cd5e60a0a4 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -609,37 +609,38 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 	return ret;
 }
 
-static struct edid *intel_crt_get_edid(struct drm_connector *connector,
-				struct i2c_adapter *i2c)
+static const struct drm_edid *intel_crt_get_edid(struct drm_connector *connector,
+						 struct i2c_adapter *i2c)
 {
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 
-	edid = drm_get_edid(connector, i2c);
+	drm_edid = drm_edid_read_ddc(connector, i2c);
 
-	if (!edid && !intel_gmbus_is_forced_bit(i2c)) {
+	if (!drm_edid && !intel_gmbus_is_forced_bit(i2c)) {
 		drm_dbg_kms(connector->dev,
 			    "CRT GMBUS EDID read failed, retry using GPIO bit-banging\n");
 		intel_gmbus_force_bit(i2c, true);
-		edid = drm_get_edid(connector, i2c);
+		drm_edid = drm_edid_read_ddc(connector, i2c);
 		intel_gmbus_force_bit(i2c, false);
 	}
 
-	return edid;
+	return drm_edid;
 }
 
 /* local version of intel_ddc_get_modes() to use intel_crt_get_edid() */
 static int intel_crt_ddc_get_modes(struct drm_connector *connector,
 				struct i2c_adapter *adapter)
 {
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 	int ret;
 
-	edid = intel_crt_get_edid(connector, adapter);
-	if (!edid)
+	drm_edid = intel_crt_get_edid(connector, adapter);
+	if (!drm_edid)
 		return 0;
 
-	ret = intel_connector_update_modes(connector, edid);
-	kfree(edid);
+	ret = intel_connector_update_modes(connector, drm_edid);
+
+	drm_edid_free(drm_edid);
 
 	return ret;
 }
@@ -648,14 +649,15 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 {
 	struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector));
 	struct drm_i915_private *dev_priv = to_i915(crt->base.base.dev);
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 	struct i2c_adapter *i2c;
 	bool ret = false;
 
 	i2c = intel_gmbus_get_adapter(dev_priv, dev_priv->display.vbt.crt_ddc_pin);
-	edid = intel_crt_get_edid(connector, i2c);
+	drm_edid = intel_crt_get_edid(connector, i2c);
 
-	if (edid) {
+	if (drm_edid) {
+		const struct edid *edid = drm_edid_raw(drm_edid);
 		bool is_digital = edid->input & DRM_EDID_INPUT_DIGITAL;
 
 		/*
@@ -676,7 +678,7 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 			    "CRT not detected via DDC:0x50 [no valid EDID found]\n");
 	}
 
-	kfree(edid);
+	drm_edid_free(drm_edid);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 1d483a83d59c6..e3f176a093d2f 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -836,15 +836,17 @@ static int intel_dp_mst_get_ddc_modes(struct drm_connector *connector)
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct intel_dp *intel_dp = intel_connector->mst_port;
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 	int ret;
 
 	if (drm_connector_is_unregistered(connector))
 		return intel_connector_update_modes(connector, NULL);
 
-	edid = drm_dp_mst_get_edid(connector, &intel_dp->mst_mgr, intel_connector->port);
-	ret = intel_connector_update_modes(connector, edid);
-	kfree(edid);
+	drm_edid = drm_dp_mst_edid_read(connector, &intel_dp->mst_mgr, intel_connector->port);
+
+	ret = intel_connector_update_modes(connector, drm_edid);
+
+	drm_edid_free(drm_edid);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 2d1b19f738835..21f92123c8446 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -2036,22 +2036,23 @@ intel_sdvo_multifunc_encoder(struct intel_sdvo *intel_sdvo)
 	return hweight16(intel_sdvo->caps.output_flags) > 1;
 }
 
-static struct edid *
+static const struct drm_edid *
 intel_sdvo_get_edid(struct drm_connector *connector)
 {
 	struct intel_sdvo *sdvo = intel_attached_sdvo(to_intel_connector(connector));
-	return drm_get_edid(connector, &sdvo->ddc);
+	return drm_edid_read_ddc(connector, &sdvo->ddc);
 }
 
 /* Mac mini hack -- use the same DDC as the analog connector */
-static struct edid *
+static const struct drm_edid *
 intel_sdvo_get_analog_edid(struct drm_connector *connector)
 {
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
+	struct drm_i915_private *i915 = to_i915(connector->dev);
+	struct i2c_adapter *i2c;
 
-	return drm_get_edid(connector,
-			    intel_gmbus_get_adapter(dev_priv,
-						    dev_priv->display.vbt.crt_ddc_pin));
+	i2c = intel_gmbus_get_adapter(i915, i915->display.vbt.crt_ddc_pin);
+
+	return drm_edid_read_ddc(connector, i2c);
 }
 
 static enum drm_connector_status
@@ -2059,11 +2060,11 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 {
 	struct intel_sdvo *intel_sdvo = intel_attached_sdvo(to_intel_connector(connector));
 	enum drm_connector_status status;
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 
-	edid = intel_sdvo_get_edid(connector);
+	drm_edid = intel_sdvo_get_edid(connector);
 
-	if (edid == NULL && intel_sdvo_multifunc_encoder(intel_sdvo)) {
+	if (!drm_edid && intel_sdvo_multifunc_encoder(intel_sdvo)) {
 		u8 ddc, saved_ddc = intel_sdvo->ddc_bus;
 
 		/*
@@ -2072,15 +2073,15 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 		 */
 		for (ddc = intel_sdvo->ddc_bus >> 1; ddc > 1; ddc >>= 1) {
 			intel_sdvo->ddc_bus = ddc;
-			edid = intel_sdvo_get_edid(connector);
-			if (edid)
+			drm_edid = intel_sdvo_get_edid(connector);
+			if (drm_edid)
 				break;
 		}
 		/*
 		 * If we found the EDID on the other bus,
 		 * assume that is the correct DDC bus.
 		 */
-		if (edid == NULL)
+		if (!drm_edid)
 			intel_sdvo->ddc_bus = saved_ddc;
 	}
 
@@ -2088,17 +2089,19 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 	 * When there is no edid and no monitor is connected with VGA
 	 * port, try to use the CRT ddc to read the EDID for DVI-connector.
 	 */
-	if (edid == NULL)
-		edid = intel_sdvo_get_analog_edid(connector);
+	if (!drm_edid)
+		drm_edid = intel_sdvo_get_analog_edid(connector);
 
 	status = connector_status_unknown;
-	if (edid != NULL) {
+	if (drm_edid) {
+		const struct edid *edid = drm_edid_raw(drm_edid);
+
 		/* DDC bus is shared, match EDID to connector type */
 		if (edid->input & DRM_EDID_INPUT_DIGITAL)
 			status = connector_status_connected;
 		else
 			status = connector_status_disconnected;
-		kfree(edid);
+		drm_edid_free(drm_edid);
 	}
 
 	return status;
@@ -2106,8 +2109,9 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 
 static bool
 intel_sdvo_connector_matches_edid(struct intel_sdvo_connector *sdvo,
-				  struct edid *edid)
+				  const struct drm_edid *drm_edid)
 {
+	const struct edid *edid = drm_edid_raw(drm_edid);
 	bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
 	bool connector_is_digital = !!IS_DIGITAL(sdvo);
 
@@ -2150,22 +2154,23 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 	else if (IS_TMDS(intel_sdvo_connector))
 		ret = intel_sdvo_tmds_sink_detect(connector);
 	else {
-		struct edid *edid;
+		const struct drm_edid *drm_edid;
 
 		/* if we have an edid check it matches the connection */
-		edid = intel_sdvo_get_edid(connector);
-		if (edid == NULL)
-			edid = intel_sdvo_get_analog_edid(connector);
-		if (edid != NULL) {
+		drm_edid = intel_sdvo_get_edid(connector);
+		if (!drm_edid)
+			drm_edid = intel_sdvo_get_analog_edid(connector);
+		if (drm_edid) {
 			if (intel_sdvo_connector_matches_edid(intel_sdvo_connector,
-							      edid))
+							      drm_edid))
 				ret = connector_status_connected;
 			else
 				ret = connector_status_disconnected;
 
-			kfree(edid);
-		} else
+			drm_edid_free(drm_edid);
+		} else {
 			ret = connector_status_connected;
+		}
 	}
 
 	return ret;
@@ -2174,13 +2179,13 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 static int intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 {
 	int num_modes = 0;
-	struct edid *edid;
+	const struct drm_edid *drm_edid;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
 	/* set the bus switch and get the modes */
-	edid = intel_sdvo_get_edid(connector);
+	drm_edid = intel_sdvo_get_edid(connector);
 
 	/*
 	 * Mac mini hack.  On this device, the DVI-I connector shares one DDC
@@ -2188,17 +2193,17 @@ static int intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 	 * DDC fails, check to see if the analog output is disconnected, in
 	 * which case we'll look there for the digital DDC data.
 	 */
-	if (!edid)
-		edid = intel_sdvo_get_analog_edid(connector);
+	if (!drm_edid)
+		drm_edid = intel_sdvo_get_analog_edid(connector);
 
-	if (!edid)
+	if (!drm_edid)
 		return 0;
 
 	if (intel_sdvo_connector_matches_edid(to_intel_sdvo_connector(connector),
-					      edid))
-		num_modes += intel_connector_update_modes(connector, edid);
+					      drm_edid))
+		num_modes += intel_connector_update_modes(connector, drm_edid);
 
-	kfree(edid);
+	drm_edid_free(drm_edid);
 
 	return num_modes;
 }
-- 
GitLab


From 4d4de1cbdb26829615d05e6b556011d308880e82 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <andrzej.hajda@intel.com>
Date: Thu, 1 Jun 2023 16:44:50 +0200
Subject: [PATCH 0747/1791] drm/i915/gt: limit lmem allocation size to succeed
 on SmallBars

In case system is short on mappable memory (256MB on SmallBar) allocation
of two 1GB buffers will fail.

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8300
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230601144450.792228-1-andrzej.hajda@intel.com
---
 drivers/gpu/drm/i915/gt/selftest_tlb.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c
index 4493c8518e91b..3bd6b540257b4 100644
--- a/drivers/gpu/drm/i915/gt/selftest_tlb.c
+++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c
@@ -190,11 +190,18 @@ pte_tlbinv(struct intel_context *ce,
 
 static struct drm_i915_gem_object *create_lmem(struct intel_gt *gt)
 {
+	struct intel_memory_region *mr = gt->i915->mm.regions[INTEL_REGION_LMEM_0];
+	resource_size_t size = SZ_1G;
+
 	/*
 	 * Allocation of largest possible page size allows to test all types
-	 * of pages.
+	 * of pages. To succeed with both allocations, especially in case of Small
+	 * BAR, try to allocate no more than quarter of mappable memory.
 	 */
-	return i915_gem_object_create_lmem(gt->i915, SZ_1G, I915_BO_ALLOC_CONTIGUOUS);
+	if (mr && size > mr->io_size / 4)
+		size = mr->io_size / 4;
+
+	return i915_gem_object_create_lmem(gt->i915, size, I915_BO_ALLOC_CONTIGUOUS);
 }
 
 static struct drm_i915_gem_object *create_smem(struct intel_gt *gt)
-- 
GitLab


From e96f099c8544a542f7cd37d2e51ba52786adbbc7 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Fri, 2 Jun 2023 17:14:16 +0800
Subject: [PATCH 0748/1791] drm/meson: Remove unneeded semicolon

./drivers/gpu/drm/meson/meson_dw_mipi_dsi.c:117:2-3: Unneeded semicolon
./drivers/gpu/drm/meson/meson_dw_mipi_dsi.c:231:2-3: Unneeded semicolon

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5392
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602091416.107850-1-yang.lee@linux.alibaba.com
---
 drivers/gpu/drm/meson/meson_dw_mipi_dsi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
index dd505ac379767..57447abf1a295 100644
--- a/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
+++ b/drivers/gpu/drm/meson/meson_dw_mipi_dsi.c
@@ -114,7 +114,7 @@ static int dw_mipi_dsi_phy_init(void *priv_data)
 	case MIPI_DSI_FMT_RGB666_PACKED:
 	case MIPI_DSI_FMT_RGB565:
 		return -EINVAL;
-	};
+	}
 
 	/* Configure color format for DPI register */
 	writel_relaxed(FIELD_PREP(MIPI_DSI_TOP_DPI_COLOR_MODE, dpi_data_format) |
@@ -228,7 +228,7 @@ static int meson_dw_mipi_dsi_host_attach(void *priv_data,
 	case MIPI_DSI_FMT_RGB565:
 		dev_err(mipi_dsi->dev, "invalid pixel format %d\n", device->format);
 		return -EINVAL;
-	};
+	}
 
 	ret = phy_init(mipi_dsi->phy);
 	if (ret)
-- 
GitLab


From 664dba662cb313da9cbb1c944c472638a65c552e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 2 Jun 2023 14:45:24 +0200
Subject: [PATCH 0749/1791] drm/meson: venc: include linux/bitfield.h

Without this header, the use of FIELD_PREP() can cause a build failure:

drivers/gpu/drm/meson/meson_venc.c: In function 'meson_encl_set_gamma_table':
drivers/gpu/drm/meson/meson_venc.c:1595:24: error: implicit declaration of function 'FIELD_PREP' [-Werror=implicit-function-declaration]

Fixes: 51fc01a03442c ("drm/meson: venc: add ENCL encoder setup for MIPI-DSI output")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602124539.894888-1-arnd@kernel.org
---
 drivers/gpu/drm/meson/meson_venc.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index 2bdc2855e249b..3bf0d6e4fc30a 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -5,6 +5,7 @@
  * Copyright (C) 2015 Amlogic, Inc. All rights reserved.
  */
 
+#include <linux/bitfield.h>
 #include <linux/export.h>
 #include <linux/iopoll.h>
 
-- 
GitLab


From 5c315434fdb6ab43566e6e0f6b9528bb0ad0aca9 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Tue, 30 May 2023 11:37:56 -0700
Subject: [PATCH 0750/1791] drm/i915/pxp: Fix size_t format specifier in
 gsccs_send_message()

When building ARCH=i386 allmodconfig, the following warning occurs:

  In file included from include/linux/device.h:15,
                   from include/linux/node.h:18,
                   from include/linux/cpu.h:17,
                   from include/linux/static_call.h:135,
                   from arch/x86/include/asm/perf_event.h:5,
                   from include/linux/perf_event.h:25,
                   from drivers/gpu/drm/i915/i915_pmu.h:11,
                   from drivers/gpu/drm/i915/gt/intel_engine_types.h:21,
                   from drivers/gpu/drm/i915/gt/intel_context_types.h:18,
                   from drivers/gpu/drm/i915/gem/i915_gem_context_types.h:20,
                   from drivers/gpu/drm/i915/i915_request.h:34,
                   from drivers/gpu/drm/i915/i915_active.h:13,
                   from drivers/gpu/drm/i915/gt/intel_context.h:13,
                   from drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c:8:
  drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c: In function 'gsccs_send_message':
  include/drm/drm_print.h:456:39: error: format '%ld' expects argument of type 'long int', but argument 4 has type 'size_t' {aka 'unsigned int'} [-Werror=format=]
    456 |         dev_##level##type((drm)->dev, "[drm] " fmt, ##__VA_ARGS__)
        |                                       ^~~~~~~~
  include/linux/dev_printk.h:110:30: note: in definition of macro 'dev_printk_index_wrap'
    110 |                 _p_func(dev, fmt, ##__VA_ARGS__);                       \
        |                              ^~~
  include/linux/dev_printk.h:146:61: note: in expansion of macro 'dev_fmt'
    146 |         dev_printk_index_wrap(_dev_warn, KERN_WARNING, dev, dev_fmt(fmt), ##__VA_ARGS__)
        |                                                             ^~~~~~~
  include/drm/drm_print.h:456:9: note: in expansion of macro 'dev_warn'
    456 |         dev_##level##type((drm)->dev, "[drm] " fmt, ##__VA_ARGS__)
        |         ^~~~
  include/drm/drm_print.h:466:9: note: in expansion of macro '__drm_printk'
    466 |         __drm_printk((drm), warn,, fmt, ##__VA_ARGS__)
        |         ^~~~~~~~~~~~
  drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c:146:17: note: in expansion of macro 'drm_warn'
    146 |                 drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%ld)\n",
        |                 ^~~~~~~~
  cc1: all warnings being treated as errors

Use the '%zu' format specifier, as the variable is a 'size_t'.

Fixes: dc9ac125d81f ("drm/i915/pxp: Add GSC-CS backend to send GSC fw messages")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530-i915-pxp-size_t-wformat-v1-1-9631081e2e5b@kernel.org
---
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
index 8dc41de3f6f74..a217821eb0fbb 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -143,7 +143,7 @@ gsccs_send_message(struct intel_pxp *pxp,
 
 	reply_size = header->message_size - sizeof(*header);
 	if (reply_size > msg_out_size_max) {
-		drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%ld)\n",
+		drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%zu)\n",
 			 reply_size, msg_out_size_max);
 		reply_size = msg_out_size_max;
 	}
-- 
GitLab


From fa1fccfb47a7f3d0fd32dbb0df050b01487ae754 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <maxime@cerno.tech>
Date: Wed, 31 May 2023 15:37:24 +0200
Subject: [PATCH 0751/1791] mailmap: Add missing email address

I've been using that email address for contributions for a while but it
seems I never added it to mailmap.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Reviewed-by: Sui Jingfeng <suijingfeng@loongson.cn>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531133724.133872-1-maxime@cerno.tech
---
 .mailmap | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.mailmap b/.mailmap
index 71127b2608d20..482dcfff40af1 100644
--- a/.mailmap
+++ b/.mailmap
@@ -327,6 +327,7 @@ Mauro Carvalho Chehab <mchehab@kernel.org> <m.chehab@samsung.com>
 Mauro Carvalho Chehab <mchehab@kernel.org> <mchehab@s-opensource.com>
 Maxim Mikityanskiy <maxtram95@gmail.com> <maximmi@mellanox.com>
 Maxim Mikityanskiy <maxtram95@gmail.com> <maximmi@nvidia.com>
+Maxime Ripard <mripard@kernel.org> <maxime@cerno.tech>
 Maxime Ripard <mripard@kernel.org> <maxime.ripard@bootlin.com>
 Maxime Ripard <mripard@kernel.org> <maxime.ripard@free-electrons.com>
 Mayuresh Janorkar <mayur@ti.com>
-- 
GitLab


From e788ff7ee3b4b7b8ae534e605154f50dc4443a40 Mon Sep 17 00:00:00 2001
From: Sui Jingfeng <suijingfeng@loongson.cn>
Date: Mon, 29 May 2023 19:29:56 +0800
Subject: [PATCH 0752/1791] drm/i915_drm.h: fix a typo

 'rbiter' -> 'arbiter'

Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529112956.2083389-1-suijingfeng@loongson.cn
---
 include/drm/i915_drm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
index 7adce327c1c28..adff685384844 100644
--- a/include/drm/i915_drm.h
+++ b/include/drm/i915_drm.h
@@ -42,7 +42,7 @@ extern struct resource intel_graphics_stolen_res;
  * The Bridge device's PCI config space has information about the
  * fb aperture size and the amount of pre-reserved memory.
  * This is all handled in the intel-gtt.ko module. i915.ko only
- * cares about the vga bit for the vga rbiter.
+ * cares about the vga bit for the vga arbiter.
  */
 #define INTEL_GMCH_CTRL		0x52
 #define INTEL_GMCH_VGA_DISABLE  (1 << 1)
-- 
GitLab


From c7b5abd3e030718b6666cdf766f8905b006a9b1b Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Mon, 29 May 2023 13:59:00 +0300
Subject: [PATCH 0753/1791] drm/i915: Remove i915_drm_suspend_mode

enum i915_drm_suspend_mode suspend_mode is only used in
intel_display_power, while we only care about whether we perform a
s2idle. Remove it and use a simple bool.

v2: Rebase

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529105900.1942814-1-jani.nikula@intel.com
---
 .../gpu/drm/i915/display/intel_display_power.c   |  8 +++-----
 .../gpu/drm/i915/display/intel_display_power.h   |  3 +--
 drivers/gpu/drm/i915/i915_driver.c               | 16 ++--------------
 drivers/gpu/drm/i915/intel_runtime_pm.h          |  6 ------
 4 files changed, 6 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 9c9a809c71f1d..2f4f00ae2f578 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -2031,7 +2031,7 @@ void intel_power_domains_disable(struct drm_i915_private *i915)
 /**
  * intel_power_domains_suspend - suspend power domain state
  * @i915: i915 device instance
- * @suspend_mode: specifies the target suspend state (idle, mem, hibernation)
+ * @s2idle: specifies whether we go to idle, or deeper sleep
  *
  * This function prepares the hardware power domain state before entering
  * system suspend.
@@ -2039,8 +2039,7 @@ void intel_power_domains_disable(struct drm_i915_private *i915)
  * It must be called with power domains already disabled (after a call to
  * intel_power_domains_disable()) and paired with intel_power_domains_resume().
  */
-void intel_power_domains_suspend(struct drm_i915_private *i915,
-				 enum i915_drm_suspend_mode suspend_mode)
+void intel_power_domains_suspend(struct drm_i915_private *i915, bool s2idle)
 {
 	struct i915_power_domains *power_domains = &i915->display.power.domains;
 	intel_wakeref_t wakeref __maybe_unused =
@@ -2055,8 +2054,7 @@ void intel_power_domains_suspend(struct drm_i915_private *i915,
 	 * resources as required and also enable deeper system power states
 	 * that would be blocked if the firmware was inactive.
 	 */
-	if (!(power_domains->allowed_dc_mask & DC_STATE_EN_DC9) &&
-	    suspend_mode == I915_DRM_SUSPEND_IDLE &&
+	if (!(power_domains->allowed_dc_mask & DC_STATE_EN_DC9) && s2idle &&
 	    intel_dmc_has_payload(i915)) {
 		intel_display_power_flush_work(i915);
 		intel_power_domains_verify_state(i915);
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h
index 8e96be8e6330a..be1a87bde0c9d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.h
+++ b/drivers/gpu/drm/i915/display/intel_display_power.h
@@ -171,8 +171,7 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume)
 void intel_power_domains_driver_remove(struct drm_i915_private *dev_priv);
 void intel_power_domains_enable(struct drm_i915_private *dev_priv);
 void intel_power_domains_disable(struct drm_i915_private *dev_priv);
-void intel_power_domains_suspend(struct drm_i915_private *dev_priv,
-				 enum i915_drm_suspend_mode);
+void intel_power_domains_suspend(struct drm_i915_private *dev_priv, bool s2idle);
 void intel_power_domains_resume(struct drm_i915_private *dev_priv);
 void intel_power_domains_sanitize_state(struct drm_i915_private *dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index c3ab5c32d4925..8e92649124d4c 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -1118,18 +1118,6 @@ static int i915_drm_suspend(struct drm_device *dev)
 	return 0;
 }
 
-static enum i915_drm_suspend_mode
-get_suspend_mode(struct drm_i915_private *dev_priv, bool hibernate)
-{
-	if (hibernate)
-		return I915_DRM_SUSPEND_HIBERNATE;
-
-	if (suspend_to_idle(dev_priv))
-		return I915_DRM_SUSPEND_IDLE;
-
-	return I915_DRM_SUSPEND_MEM;
-}
-
 static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
@@ -1137,6 +1125,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 	struct intel_runtime_pm *rpm = &dev_priv->runtime_pm;
 	struct intel_gt *gt;
 	int ret, i;
+	bool s2idle = !hibernation && suspend_to_idle(dev_priv);
 
 	disable_rpm_wakeref_asserts(rpm);
 
@@ -1147,8 +1136,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 	for_each_gt(gt, dev_priv, i)
 		intel_uncore_suspend(gt->uncore);
 
-	intel_power_domains_suspend(dev_priv,
-				    get_suspend_mode(dev_priv, hibernation));
+	intel_power_domains_suspend(dev_priv, s2idle);
 
 	intel_display_power_suspend_late(dev_priv);
 
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.h b/drivers/gpu/drm/i915/intel_runtime_pm.h
index e592e8d6499a1..764b183ae4529 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.h
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.h
@@ -16,12 +16,6 @@ struct device;
 struct drm_i915_private;
 struct drm_printer;
 
-enum i915_drm_suspend_mode {
-	I915_DRM_SUSPEND_IDLE,
-	I915_DRM_SUSPEND_MEM,
-	I915_DRM_SUSPEND_HIBERNATE,
-};
-
 /*
  * This struct helps tracking the state needed for runtime PM, which puts the
  * device in PCI D3 state. Notice that when this happens, nothing on the
-- 
GitLab


From f917130f19fe62c6688cc95ebebfafee3e005958 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 22 Dec 2022 21:18:04 +0100
Subject: [PATCH 0754/1791] drm/i915: Flush power delayed put when connector
 init failed

When intel_dp_init_connector fails, some power wells used in dp aux
communication may not be completely disabled yet. This may result in a
null pointer dereference when icl_aux_pw_to_phy() is called from
icl_combo_phy_aux_power_well_disable() after the encoder and connector
are already freed.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Acked-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221222201804.1380963-1-maarten.lankhorst@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 8445a37d9d01f..f4192fda1a76e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -5719,6 +5719,7 @@ intel_dp_init_connector(struct intel_digital_port *dig_port,
 	return true;
 
 fail:
+	intel_display_power_flush_work(dev_priv);
 	drm_connector_cleanup(connector);
 
 	return false;
-- 
GitLab


From 0fcf8ffdfe9013b05807b99a4c28da59b2d39bec Mon Sep 17 00:00:00 2001
From: Jeffrey Hugo <quic_jhugo@quicinc.com>
Date: Tue, 23 May 2023 10:14:21 -0600
Subject: [PATCH 0755/1791] MAINTAINERS: Add Carl/Pranjal as QAIC reviewers

Carl and Pranjal have been reviewing the QAIC patches.  List them as
reviewers so that they are copied on all developments which will make
it easier for them to continue reviewing QAIC patches.

Signed-off-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Acked-by: Carl Vanderlip <quic_carlv@quicinc.com>
Acked-by: Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230523161421.11017-1-quic_jhugo@quicinc.com
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index b344e1318ac3b..aa0aa6c317727 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -17388,6 +17388,8 @@ F:	include/dt-bindings/clock/qcom,*
 
 QUALCOMM CLOUD AI (QAIC) DRIVER
 M:	Jeffrey Hugo <quic_jhugo@quicinc.com>
+R:	Carl Vanderlip <quic_carlv@quicinc.com>
+R:	Pranjal Ramajor Asha Kanojiya <quic_pkanojiy@quicinc.com>
 L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
 S:	Supported
-- 
GitLab


From 5a3c46b809d09f8ef59e2fbf2463b1c102aecbaa Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Mon, 29 May 2023 11:37:47 +0530
Subject: [PATCH 0756/1791] drm/i915/display: Set correct voltage level for
 480MHz CDCLK

According to Bspec, the voltage level for 480MHz is to be set as 1
instead of 2.

BSpec: 49208

Fixes: 06f1b06dc5b7 ("drm/i915/display: Add 480 MHz CDCLK steps for RPL-U")

v2: rebase

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230529060747.3972259-1-chaitanya.kumar.borah@intel.com
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 30 +++++++++++++++++++---
 1 file changed, 26 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 6bed75f1541a7..1a5268e3d0a36 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -1453,6 +1453,18 @@ static u8 tgl_calc_voltage_level(int cdclk)
 		return 0;
 }
 
+static u8 rplu_calc_voltage_level(int cdclk)
+{
+	if (cdclk > 556800)
+		return 3;
+	else if (cdclk > 480000)
+		return 2;
+	else if (cdclk > 312000)
+		return 1;
+	else
+		return 0;
+}
+
 static void icl_readout_refclk(struct drm_i915_private *dev_priv,
 			       struct intel_cdclk_config *cdclk_config)
 {
@@ -3397,6 +3409,13 @@ static const struct intel_cdclk_funcs mtl_cdclk_funcs = {
 	.calc_voltage_level = tgl_calc_voltage_level,
 };
 
+static const struct intel_cdclk_funcs rplu_cdclk_funcs = {
+	.get_cdclk = bxt_get_cdclk,
+	.set_cdclk = bxt_set_cdclk,
+	.modeset_calc_cdclk = bxt_modeset_calc_cdclk,
+	.calc_voltage_level = rplu_calc_voltage_level,
+};
+
 static const struct intel_cdclk_funcs tgl_cdclk_funcs = {
 	.get_cdclk = bxt_get_cdclk,
 	.set_cdclk = bxt_set_cdclk,
@@ -3539,14 +3558,17 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv)
 		dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs;
 		dev_priv->display.cdclk.table = dg2_cdclk_table;
 	} else if (IS_ALDERLAKE_P(dev_priv)) {
-		dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs;
 		/* Wa_22011320316:adl-p[a0] */
-		if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0))
+		if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) {
 			dev_priv->display.cdclk.table = adlp_a_step_cdclk_table;
-		else if (IS_ADLP_RPLU(dev_priv))
+			dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs;
+		} else if (IS_ADLP_RPLU(dev_priv)) {
 			dev_priv->display.cdclk.table = rplu_cdclk_table;
-		else
+			dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs;
+		} else {
 			dev_priv->display.cdclk.table = adlp_cdclk_table;
+			dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs;
+		}
 	} else if (IS_ROCKETLAKE(dev_priv)) {
 		dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs;
 		dev_priv->display.cdclk.table = rkl_cdclk_table;
-- 
GitLab


From e4812ab8e6b142e1ce19b50fcf744e20ea9adfe3 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Date: Sun, 16 Apr 2023 14:52:35 +0300
Subject: [PATCH 0757/1791] drm/virtio: Refactor and optimize job submission
 code path

Move virtio_gpu_execbuffer_ioctl() into separate virtgpu_submit.c file,
refactoring and optimizing the code along the way to ease addition of new
features to the ioctl.

The optimization is done by using optimal ordering of the job's submission
steps, reducing code path from the start of the ioctl to the point of
pushing job to virtio queue. Job's initialization is now performed before
in-fence is awaited and out-fence setup is made after sending out job to
virtio.

Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Tested-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230416115237.798604-2-dmitry.osipenko@collabora.com
---
 drivers/gpu/drm/virtio/Makefile         |   2 +-
 drivers/gpu/drm/virtio/virtgpu_drv.h    |   4 +
 drivers/gpu/drm/virtio/virtgpu_ioctl.c  | 182 ---------------
 drivers/gpu/drm/virtio/virtgpu_submit.c | 295 ++++++++++++++++++++++++
 4 files changed, 300 insertions(+), 183 deletions(-)
 create mode 100644 drivers/gpu/drm/virtio/virtgpu_submit.c

diff --git a/drivers/gpu/drm/virtio/Makefile b/drivers/gpu/drm/virtio/Makefile
index b99fa4a73b68e..d2e1788a82270 100644
--- a/drivers/gpu/drm/virtio/Makefile
+++ b/drivers/gpu/drm/virtio/Makefile
@@ -6,6 +6,6 @@
 virtio-gpu-y := virtgpu_drv.o virtgpu_kms.o virtgpu_gem.o virtgpu_vram.o \
 	virtgpu_display.o virtgpu_vq.o \
 	virtgpu_fence.o virtgpu_object.o virtgpu_debugfs.o virtgpu_plane.o \
-	virtgpu_ioctl.o virtgpu_prime.o virtgpu_trace_points.o
+	virtgpu_ioctl.o virtgpu_prime.o virtgpu_trace_points.o virtgpu_submit.o
 
 obj-$(CONFIG_DRM_VIRTIO_GPU) += virtio-gpu.o
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index af6ffb6960866..4126c384286bf 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -486,4 +486,8 @@ void virtio_gpu_vram_unmap_dma_buf(struct device *dev,
 				   struct sg_table *sgt,
 				   enum dma_data_direction dir);
 
+/* virtgpu_submit.c */
+int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file);
+
 #endif
diff --git a/drivers/gpu/drm/virtio/virtgpu_ioctl.c b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
index da45215a933d0..b24b11f25197d 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ioctl.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
@@ -38,36 +38,6 @@
 				    VIRTGPU_BLOB_FLAG_USE_SHAREABLE | \
 				    VIRTGPU_BLOB_FLAG_USE_CROSS_DEVICE)
 
-static int virtio_gpu_fence_event_create(struct drm_device *dev,
-					 struct drm_file *file,
-					 struct virtio_gpu_fence *fence,
-					 uint32_t ring_idx)
-{
-	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
-	struct virtio_gpu_fence_event *e = NULL;
-	int ret;
-
-	if (!(vfpriv->ring_idx_mask & BIT_ULL(ring_idx)))
-		return 0;
-
-	e = kzalloc(sizeof(*e), GFP_KERNEL);
-	if (!e)
-		return -ENOMEM;
-
-	e->event.type = VIRTGPU_EVENT_FENCE_SIGNALED;
-	e->event.length = sizeof(e->event);
-
-	ret = drm_event_reserve_init(dev, file, &e->base, &e->event);
-	if (ret)
-		goto free;
-
-	fence->e = e;
-	return 0;
-free:
-	kfree(e);
-	return ret;
-}
-
 /* Must be called with &virtio_gpu_fpriv.struct_mutex held. */
 static void virtio_gpu_create_context_locked(struct virtio_gpu_device *vgdev,
 					     struct virtio_gpu_fpriv *vfpriv)
@@ -108,158 +78,6 @@ static int virtio_gpu_map_ioctl(struct drm_device *dev, void *data,
 					 &virtio_gpu_map->offset);
 }
 
-/*
- * Usage of execbuffer:
- * Relocations need to take into account the full VIRTIO_GPUDrawable size.
- * However, the command as passed from user space must *not* contain the initial
- * VIRTIO_GPUReleaseInfo struct (first XXX bytes)
- */
-static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data,
-				 struct drm_file *file)
-{
-	struct drm_virtgpu_execbuffer *exbuf = data;
-	struct virtio_gpu_device *vgdev = dev->dev_private;
-	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
-	struct virtio_gpu_fence *out_fence;
-	int ret;
-	uint32_t *bo_handles = NULL;
-	void __user *user_bo_handles = NULL;
-	struct virtio_gpu_object_array *buflist = NULL;
-	struct sync_file *sync_file;
-	int out_fence_fd = -1;
-	void *buf;
-	uint64_t fence_ctx;
-	uint32_t ring_idx;
-
-	fence_ctx = vgdev->fence_drv.context;
-	ring_idx = 0;
-
-	if (vgdev->has_virgl_3d == false)
-		return -ENOSYS;
-
-	if ((exbuf->flags & ~VIRTGPU_EXECBUF_FLAGS))
-		return -EINVAL;
-
-	if ((exbuf->flags & VIRTGPU_EXECBUF_RING_IDX)) {
-		if (exbuf->ring_idx >= vfpriv->num_rings)
-			return -EINVAL;
-
-		if (!vfpriv->base_fence_ctx)
-			return -EINVAL;
-
-		fence_ctx = vfpriv->base_fence_ctx;
-		ring_idx = exbuf->ring_idx;
-	}
-
-	virtio_gpu_create_context(dev, file);
-	if (exbuf->flags & VIRTGPU_EXECBUF_FENCE_FD_IN) {
-		struct dma_fence *in_fence;
-
-		in_fence = sync_file_get_fence(exbuf->fence_fd);
-
-		if (!in_fence)
-			return -EINVAL;
-
-		/*
-		 * Wait if the fence is from a foreign context, or if the fence
-		 * array contains any fence from a foreign context.
-		 */
-		ret = 0;
-		if (!dma_fence_match_context(in_fence, fence_ctx + ring_idx))
-			ret = dma_fence_wait(in_fence, true);
-
-		dma_fence_put(in_fence);
-		if (ret)
-			return ret;
-	}
-
-	if (exbuf->flags & VIRTGPU_EXECBUF_FENCE_FD_OUT) {
-		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
-		if (out_fence_fd < 0)
-			return out_fence_fd;
-	}
-
-	if (exbuf->num_bo_handles) {
-		bo_handles = kvmalloc_array(exbuf->num_bo_handles,
-					    sizeof(uint32_t), GFP_KERNEL);
-		if (!bo_handles) {
-			ret = -ENOMEM;
-			goto out_unused_fd;
-		}
-
-		user_bo_handles = u64_to_user_ptr(exbuf->bo_handles);
-		if (copy_from_user(bo_handles, user_bo_handles,
-				   exbuf->num_bo_handles * sizeof(uint32_t))) {
-			ret = -EFAULT;
-			goto out_unused_fd;
-		}
-
-		buflist = virtio_gpu_array_from_handles(file, bo_handles,
-							exbuf->num_bo_handles);
-		if (!buflist) {
-			ret = -ENOENT;
-			goto out_unused_fd;
-		}
-		kvfree(bo_handles);
-		bo_handles = NULL;
-	}
-
-	buf = vmemdup_user(u64_to_user_ptr(exbuf->command), exbuf->size);
-	if (IS_ERR(buf)) {
-		ret = PTR_ERR(buf);
-		goto out_unused_fd;
-	}
-
-	if (buflist) {
-		ret = virtio_gpu_array_lock_resv(buflist);
-		if (ret)
-			goto out_memdup;
-	}
-
-	out_fence = virtio_gpu_fence_alloc(vgdev, fence_ctx, ring_idx);
-	if(!out_fence) {
-		ret = -ENOMEM;
-		goto out_unresv;
-	}
-
-	ret = virtio_gpu_fence_event_create(dev, file, out_fence, ring_idx);
-	if (ret)
-		goto out_unresv;
-
-	if (out_fence_fd >= 0) {
-		sync_file = sync_file_create(&out_fence->f);
-		if (!sync_file) {
-			dma_fence_put(&out_fence->f);
-			ret = -ENOMEM;
-			goto out_unresv;
-		}
-
-		exbuf->fence_fd = out_fence_fd;
-		fd_install(out_fence_fd, sync_file->file);
-	}
-
-	virtio_gpu_cmd_submit(vgdev, buf, exbuf->size,
-			      vfpriv->ctx_id, buflist, out_fence);
-	dma_fence_put(&out_fence->f);
-	virtio_gpu_notify(vgdev);
-	return 0;
-
-out_unresv:
-	if (buflist)
-		virtio_gpu_array_unlock_resv(buflist);
-out_memdup:
-	kvfree(buf);
-out_unused_fd:
-	kvfree(bo_handles);
-	if (buflist)
-		virtio_gpu_array_put_free(buflist);
-
-	if (out_fence_fd >= 0)
-		put_unused_fd(out_fence_fd);
-
-	return ret;
-}
-
 static int virtio_gpu_getparam_ioctl(struct drm_device *dev, void *data,
 				     struct drm_file *file)
 {
diff --git a/drivers/gpu/drm/virtio/virtgpu_submit.c b/drivers/gpu/drm/virtio/virtgpu_submit.c
new file mode 100644
index 0000000000000..84e7c4d9d8c7e
--- /dev/null
+++ b/drivers/gpu/drm/virtio/virtgpu_submit.c
@@ -0,0 +1,295 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright (C) 2015 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * Authors:
+ *    Dave Airlie
+ *    Alon Levy
+ */
+
+#include <linux/dma-fence-unwrap.h>
+#include <linux/file.h>
+#include <linux/sync_file.h>
+#include <linux/uaccess.h>
+
+#include <drm/drm_file.h>
+#include <drm/virtgpu_drm.h>
+
+#include "virtgpu_drv.h"
+
+struct virtio_gpu_submit {
+	struct virtio_gpu_object_array *buflist;
+	struct drm_virtgpu_execbuffer *exbuf;
+	struct virtio_gpu_fence *out_fence;
+	struct virtio_gpu_fpriv *vfpriv;
+	struct virtio_gpu_device *vgdev;
+	struct sync_file *sync_file;
+	struct drm_file *file;
+	int out_fence_fd;
+	u64 fence_ctx;
+	u32 ring_idx;
+	void *buf;
+};
+
+static int virtio_gpu_dma_fence_wait(struct virtio_gpu_submit *submit,
+				     struct dma_fence *in_fence)
+{
+	u32 context = submit->fence_ctx + submit->ring_idx;
+
+	if (dma_fence_match_context(in_fence, context))
+		return 0;
+
+	return dma_fence_wait(in_fence, true);
+}
+
+static int virtio_gpu_fence_event_create(struct drm_device *dev,
+					 struct drm_file *file,
+					 struct virtio_gpu_fence *fence,
+					 u32 ring_idx)
+{
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct virtio_gpu_fence_event *e = NULL;
+	int ret;
+
+	if (!(vfpriv->ring_idx_mask & BIT_ULL(ring_idx)))
+		return 0;
+
+	e = kzalloc(sizeof(*e), GFP_KERNEL);
+	if (!e)
+		return -ENOMEM;
+
+	e->event.type = VIRTGPU_EVENT_FENCE_SIGNALED;
+	e->event.length = sizeof(e->event);
+
+	ret = drm_event_reserve_init(dev, file, &e->base, &e->event);
+	if (ret) {
+		kfree(e);
+		return ret;
+	}
+
+	fence->e = e;
+
+	return 0;
+}
+
+static int virtio_gpu_init_submit_buflist(struct virtio_gpu_submit *submit)
+{
+	struct drm_virtgpu_execbuffer *exbuf = submit->exbuf;
+	u32 *bo_handles;
+
+	if (!exbuf->num_bo_handles)
+		return 0;
+
+	bo_handles = kvmalloc_array(exbuf->num_bo_handles, sizeof(*bo_handles),
+				    GFP_KERNEL);
+	if (!bo_handles)
+		return -ENOMEM;
+
+	if (copy_from_user(bo_handles, u64_to_user_ptr(exbuf->bo_handles),
+			   exbuf->num_bo_handles * sizeof(*bo_handles))) {
+		kvfree(bo_handles);
+		return -EFAULT;
+	}
+
+	submit->buflist = virtio_gpu_array_from_handles(submit->file, bo_handles,
+							exbuf->num_bo_handles);
+	if (!submit->buflist) {
+		kvfree(bo_handles);
+		return -ENOENT;
+	}
+
+	kvfree(bo_handles);
+
+	return 0;
+}
+
+static void virtio_gpu_cleanup_submit(struct virtio_gpu_submit *submit)
+{
+	if (!IS_ERR(submit->buf))
+		kvfree(submit->buf);
+
+	if (submit->buflist)
+		virtio_gpu_array_put_free(submit->buflist);
+
+	if (submit->out_fence_fd >= 0)
+		put_unused_fd(submit->out_fence_fd);
+
+	if (submit->out_fence)
+		dma_fence_put(&submit->out_fence->f);
+
+	if (submit->sync_file)
+		fput(submit->sync_file->file);
+}
+
+static void virtio_gpu_submit(struct virtio_gpu_submit *submit)
+{
+	virtio_gpu_cmd_submit(submit->vgdev, submit->buf, submit->exbuf->size,
+			      submit->vfpriv->ctx_id, submit->buflist,
+			      submit->out_fence);
+	virtio_gpu_notify(submit->vgdev);
+}
+
+static void virtio_gpu_complete_submit(struct virtio_gpu_submit *submit)
+{
+	submit->buf = NULL;
+	submit->buflist = NULL;
+	submit->sync_file = NULL;
+	submit->out_fence = NULL;
+	submit->out_fence_fd = -1;
+}
+
+static int virtio_gpu_init_submit(struct virtio_gpu_submit *submit,
+				  struct drm_virtgpu_execbuffer *exbuf,
+				  struct drm_device *dev,
+				  struct drm_file *file,
+				  u64 fence_ctx, u32 ring_idx)
+{
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fence *out_fence;
+	int err;
+
+	memset(submit, 0, sizeof(*submit));
+
+	out_fence = virtio_gpu_fence_alloc(vgdev, fence_ctx, ring_idx);
+	if (!out_fence)
+		return -ENOMEM;
+
+	err = virtio_gpu_fence_event_create(dev, file, out_fence, ring_idx);
+	if (err) {
+		dma_fence_put(&out_fence->f);
+		return err;
+	}
+
+	submit->out_fence = out_fence;
+	submit->fence_ctx = fence_ctx;
+	submit->ring_idx = ring_idx;
+	submit->out_fence_fd = -1;
+	submit->vfpriv = vfpriv;
+	submit->vgdev = vgdev;
+	submit->exbuf = exbuf;
+	submit->file = file;
+
+	err = virtio_gpu_init_submit_buflist(submit);
+	if (err)
+		return err;
+
+	submit->buf = vmemdup_user(u64_to_user_ptr(exbuf->command), exbuf->size);
+	if (IS_ERR(submit->buf))
+		return PTR_ERR(submit->buf);
+
+	if (exbuf->flags & VIRTGPU_EXECBUF_FENCE_FD_OUT) {
+		err = get_unused_fd_flags(O_CLOEXEC);
+		if (err < 0)
+			return err;
+
+		submit->out_fence_fd = err;
+
+		submit->sync_file = sync_file_create(&out_fence->f);
+		if (!submit->sync_file)
+			return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static int virtio_gpu_wait_in_fence(struct virtio_gpu_submit *submit)
+{
+	int ret = 0;
+
+	if (submit->exbuf->flags & VIRTGPU_EXECBUF_FENCE_FD_IN) {
+		struct dma_fence *in_fence =
+				sync_file_get_fence(submit->exbuf->fence_fd);
+		if (!in_fence)
+			return -EINVAL;
+
+		/*
+		 * Wait if the fence is from a foreign context, or if the
+		 * fence array contains any fence from a foreign context.
+		 */
+		ret = virtio_gpu_dma_fence_wait(submit, in_fence);
+
+		dma_fence_put(in_fence);
+	}
+
+	return ret;
+}
+
+static void virtio_gpu_install_out_fence_fd(struct virtio_gpu_submit *submit)
+{
+	if (submit->sync_file) {
+		submit->exbuf->fence_fd = submit->out_fence_fd;
+		fd_install(submit->out_fence_fd, submit->sync_file->file);
+	}
+}
+
+static int virtio_gpu_lock_buflist(struct virtio_gpu_submit *submit)
+{
+	if (submit->buflist)
+		return virtio_gpu_array_lock_resv(submit->buflist);
+
+	return 0;
+}
+
+int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	u64 fence_ctx = vgdev->fence_drv.context;
+	struct drm_virtgpu_execbuffer *exbuf = data;
+	struct virtio_gpu_submit submit;
+	u32 ring_idx = 0;
+	int ret = -EINVAL;
+
+	if (!vgdev->has_virgl_3d)
+		return -ENOSYS;
+
+	if (exbuf->flags & ~VIRTGPU_EXECBUF_FLAGS)
+		return ret;
+
+	if (exbuf->flags & VIRTGPU_EXECBUF_RING_IDX) {
+		if (exbuf->ring_idx >= vfpriv->num_rings)
+			return ret;
+
+		if (!vfpriv->base_fence_ctx)
+			return ret;
+
+		fence_ctx = vfpriv->base_fence_ctx;
+		ring_idx = exbuf->ring_idx;
+	}
+
+	virtio_gpu_create_context(dev, file);
+
+	ret = virtio_gpu_init_submit(&submit, exbuf, dev, file,
+				     fence_ctx, ring_idx);
+	if (ret)
+		goto cleanup;
+
+	/*
+	 * Await in-fences in the end of the job submission path to
+	 * optimize the path by proceeding directly to the submission
+	 * to virtio after the waits.
+	 */
+	ret = virtio_gpu_wait_in_fence(&submit);
+	if (ret)
+		goto cleanup;
+
+	ret = virtio_gpu_lock_buflist(&submit);
+	if (ret)
+		goto cleanup;
+
+	virtio_gpu_submit(&submit);
+
+	/*
+	 * Set up usr-out data after submitting the job to optimize
+	 * the job submission path.
+	 */
+	virtio_gpu_install_out_fence_fd(&submit);
+	virtio_gpu_complete_submit(&submit);
+cleanup:
+	virtio_gpu_cleanup_submit(&submit);
+
+	return ret;
+}
-- 
GitLab


From eba57fb5498f2858768321e64570caea188455b8 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Date: Sun, 16 Apr 2023 14:52:36 +0300
Subject: [PATCH 0758/1791] drm/virtio: Wait for each dma-fence of in-fence
 array individually

Use dma-fence-unwrap API for waiting each dma-fence of the in-fence array
individually. Sync file's in-fence array always has a non-matching fence
context ID, which doesn't allow to skip waiting of fences with a matching
context ID in a case of a merged sync file fence.

Suggested-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Tested-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230416115237.798604-3-dmitry.osipenko@collabora.com
---
 drivers/gpu/drm/virtio/virtgpu_submit.c | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/virtio/virtgpu_submit.c b/drivers/gpu/drm/virtio/virtgpu_submit.c
index 84e7c4d9d8c7e..cf3c04b16a7a8 100644
--- a/drivers/gpu/drm/virtio/virtgpu_submit.c
+++ b/drivers/gpu/drm/virtio/virtgpu_submit.c
@@ -32,8 +32,8 @@ struct virtio_gpu_submit {
 	void *buf;
 };
 
-static int virtio_gpu_dma_fence_wait(struct virtio_gpu_submit *submit,
-				     struct dma_fence *in_fence)
+static int virtio_gpu_do_fence_wait(struct virtio_gpu_submit *submit,
+				    struct dma_fence *in_fence)
 {
 	u32 context = submit->fence_ctx + submit->ring_idx;
 
@@ -43,6 +43,22 @@ static int virtio_gpu_dma_fence_wait(struct virtio_gpu_submit *submit,
 	return dma_fence_wait(in_fence, true);
 }
 
+static int virtio_gpu_dma_fence_wait(struct virtio_gpu_submit *submit,
+				     struct dma_fence *fence)
+{
+	struct dma_fence_unwrap itr;
+	struct dma_fence *f;
+	int err;
+
+	dma_fence_unwrap_for_each(f, &itr, fence) {
+		err = virtio_gpu_do_fence_wait(submit, f);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 static int virtio_gpu_fence_event_create(struct drm_device *dev,
 					 struct drm_file *file,
 					 struct virtio_gpu_fence *fence,
-- 
GitLab


From 5f31d7e61ddf5ca8db06455b30d3b3e16d656944 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:30 +0300
Subject: [PATCH 0759/1791] drm/msm/dpu: fix SSPP register definitions

Reorder SSPP register definitions to sort them in the ascending order.
Move register bitfields after the register definitions.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/537903/
Link: https://lore.kernel.org/r/20230518222238.3815293-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 64 ++++++++++-----------
 1 file changed, 32 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 6b68ec5c7a5a2..08098880b7d57 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -26,45 +26,18 @@
 #define SSPP_SRC_FORMAT                    0x30
 #define SSPP_SRC_UNPACK_PATTERN            0x34
 #define SSPP_SRC_OP_MODE                   0x38
-
-/* SSPP_MULTIRECT*/
-#define SSPP_SRC_SIZE_REC1                 0x16C
-#define SSPP_SRC_XY_REC1                   0x168
-#define SSPP_OUT_SIZE_REC1                 0x160
-#define SSPP_OUT_XY_REC1                   0x164
-#define SSPP_SRC_FORMAT_REC1               0x174
-#define SSPP_SRC_UNPACK_PATTERN_REC1       0x178
-#define SSPP_SRC_OP_MODE_REC1              0x17C
-#define SSPP_MULTIRECT_OPMODE              0x170
-#define SSPP_SRC_CONSTANT_COLOR_REC1       0x180
-#define SSPP_EXCL_REC_SIZE_REC1            0x184
-#define SSPP_EXCL_REC_XY_REC1              0x188
-
-#define MDSS_MDP_OP_DEINTERLACE            BIT(22)
-#define MDSS_MDP_OP_DEINTERLACE_ODD        BIT(23)
-#define MDSS_MDP_OP_IGC_ROM_1              BIT(18)
-#define MDSS_MDP_OP_IGC_ROM_0              BIT(17)
-#define MDSS_MDP_OP_IGC_EN                 BIT(16)
-#define MDSS_MDP_OP_FLIP_UD                BIT(14)
-#define MDSS_MDP_OP_FLIP_LR                BIT(13)
-#define MDSS_MDP_OP_BWC_EN                 BIT(0)
-#define MDSS_MDP_OP_PE_OVERRIDE            BIT(31)
-#define MDSS_MDP_OP_BWC_LOSSLESS           (0 << 1)
-#define MDSS_MDP_OP_BWC_Q_HIGH             (1 << 1)
-#define MDSS_MDP_OP_BWC_Q_MED              (2 << 1)
-
 #define SSPP_SRC_CONSTANT_COLOR            0x3c
 #define SSPP_EXCL_REC_CTL                  0x40
 #define SSPP_UBWC_STATIC_CTRL              0x44
-#define SSPP_FETCH_CONFIG                  0x048
+#define SSPP_FETCH_CONFIG                  0x48
 #define SSPP_DANGER_LUT                    0x60
 #define SSPP_SAFE_LUT                      0x64
 #define SSPP_CREQ_LUT                      0x68
 #define SSPP_QOS_CTRL                      0x6C
-#define SSPP_DECIMATION_CONFIG             0xB4
 #define SSPP_SRC_ADDR_SW_STATUS            0x70
 #define SSPP_CREQ_LUT_0                    0x74
 #define SSPP_CREQ_LUT_1                    0x78
+#define SSPP_DECIMATION_CONFIG             0xB4
 #define SSPP_SW_PIX_EXT_C0_LR              0x100
 #define SSPP_SW_PIX_EXT_C0_TB              0x104
 #define SSPP_SW_PIX_EXT_C0_REQ_PIXELS      0x108
@@ -81,11 +54,33 @@
 #define SSPP_TRAFFIC_SHAPER_PREFILL        0x150
 #define SSPP_TRAFFIC_SHAPER_REC1_PREFILL   0x154
 #define SSPP_TRAFFIC_SHAPER_REC1           0x158
+#define SSPP_OUT_SIZE_REC1                 0x160
+#define SSPP_OUT_XY_REC1                   0x164
+#define SSPP_SRC_XY_REC1                   0x168
+#define SSPP_SRC_SIZE_REC1                 0x16C
+#define SSPP_MULTIRECT_OPMODE              0x170
+#define SSPP_SRC_FORMAT_REC1               0x174
+#define SSPP_SRC_UNPACK_PATTERN_REC1       0x178
+#define SSPP_SRC_OP_MODE_REC1              0x17C
+#define SSPP_SRC_CONSTANT_COLOR_REC1       0x180
+#define SSPP_EXCL_REC_SIZE_REC1            0x184
+#define SSPP_EXCL_REC_XY_REC1              0x188
 #define SSPP_EXCL_REC_SIZE                 0x1B4
 #define SSPP_EXCL_REC_XY                   0x1B8
-#define SSPP_VIG_OP_MODE                   0x0
-#define SSPP_VIG_CSC_10_OP_MODE            0x0
-#define SSPP_TRAFFIC_SHAPER_BPC_MAX        0xFF
+
+/* SSPP_SRC_OP_MODE & OP_MODE_REC1 */
+#define MDSS_MDP_OP_DEINTERLACE            BIT(22)
+#define MDSS_MDP_OP_DEINTERLACE_ODD        BIT(23)
+#define MDSS_MDP_OP_IGC_ROM_1              BIT(18)
+#define MDSS_MDP_OP_IGC_ROM_0              BIT(17)
+#define MDSS_MDP_OP_IGC_EN                 BIT(16)
+#define MDSS_MDP_OP_FLIP_UD                BIT(14)
+#define MDSS_MDP_OP_FLIP_LR                BIT(13)
+#define MDSS_MDP_OP_BWC_EN                 BIT(0)
+#define MDSS_MDP_OP_PE_OVERRIDE            BIT(31)
+#define MDSS_MDP_OP_BWC_LOSSLESS           (0 << 1)
+#define MDSS_MDP_OP_BWC_Q_HIGH             (1 << 1)
+#define MDSS_MDP_OP_BWC_Q_MED              (2 << 1)
 
 /* SSPP_QOS_CTRL */
 #define SSPP_QOS_CTRL_VBLANK_EN            BIT(16)
@@ -96,6 +91,7 @@
 #define SSPP_QOS_CTRL_CREQ_VBLANK_OFF      20
 
 /* DPU_SSPP_SCALER_QSEED2 */
+#define SSPP_VIG_OP_MODE                   0x0
 #define SCALE_CONFIG                       0x04
 #define COMP0_3_PHASE_STEP_X               0x10
 #define COMP0_3_PHASE_STEP_Y               0x14
@@ -107,6 +103,9 @@
 #define COMP1_2_INIT_PHASE_Y               0x2C
 #define VIG_0_QSEED2_SHARP                 0x30
 
+/* SSPP_TRAFFIC_SHAPER and _REC1 */
+#define SSPP_TRAFFIC_SHAPER_BPC_MAX        0xFF
+
 /*
  * Definitions for ViG op modes
  */
@@ -128,6 +127,7 @@
 /*
  * Definitions for CSC 10 op modes
  */
+#define SSPP_VIG_CSC_10_OP_MODE            0x0
 #define VIG_CSC_10_SRC_DATAFMT BIT(1)
 #define VIG_CSC_10_EN          BIT(0)
 #define CSC_10BIT_OFFSET       4
-- 
GitLab


From 48b3207e4ed9e5140c69e229b697373f233419c8 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:31 +0300
Subject: [PATCH 0760/1791] drm/msm/dpu: simplify CDP programming

Get rid of intermediatory configuration structure and defines. Pass the
format and the enablement bit directly to the new helper. The
WB_CDP_CNTL register ignores BIT(2), so we can write it for both SSPP
and WB CDP settings.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537910/
Link: https://lore.kernel.org/r/20230518222238.3815293-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   | 17 ++++-----------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c   | 17 ++++-----------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h   | 14 ++++---------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c   | 21 +++++++++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h   | 19 +++--------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c     | 19 ++++-------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h     | 11 ++--------
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c     | 16 +++-----------
 8 files changed, 45 insertions(+), 89 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index bac4aa807b4bc..e7b65f6f53d64 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -140,7 +140,6 @@ static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc,
 	struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc);
 	struct dpu_hw_wb *hw_wb;
 	struct dpu_hw_wb_cfg *wb_cfg;
-	struct dpu_hw_cdp_cfg cdp_cfg;
 
 	if (!phys_enc || !phys_enc->dpu_kms || !phys_enc->dpu_kms->catalog) {
 		DPU_ERROR("invalid encoder\n");
@@ -163,18 +162,10 @@ static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc,
 		hw_wb->ops.setup_outformat(hw_wb, wb_cfg);
 
 	if (hw_wb->ops.setup_cdp) {
-		memset(&cdp_cfg, 0, sizeof(struct dpu_hw_cdp_cfg));
-
-		cdp_cfg.enable = phys_enc->dpu_kms->catalog->perf->cdp_cfg
-				[DPU_PERF_CDP_USAGE_NRT].wr_enable;
-		cdp_cfg.ubwc_meta_enable =
-				DPU_FORMAT_IS_UBWC(wb_cfg->dest.format);
-		cdp_cfg.tile_amortize_enable =
-				DPU_FORMAT_IS_UBWC(wb_cfg->dest.format) ||
-				DPU_FORMAT_IS_TILE(wb_cfg->dest.format);
-		cdp_cfg.preload_ahead = DPU_WB_CDP_PRELOAD_AHEAD_64;
-
-		hw_wb->ops.setup_cdp(hw_wb, &cdp_cfg);
+		const struct dpu_perf_cfg *perf = phys_enc->dpu_kms->catalog->perf;
+
+		hw_wb->ops.setup_cdp(hw_wb, wb_cfg->dest.format,
+				     perf->cdp_cfg[DPU_PERF_CDP_USAGE_NRT].wr_enable);
 	}
 
 	if (hw_wb->ops.setup_outaddress)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 08098880b7d57..78d928065b7bd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -590,13 +590,13 @@ static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
 }
 
 static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe,
-		struct dpu_hw_cdp_cfg *cfg)
+				  const struct dpu_format *fmt,
+				  bool enable)
 {
 	struct dpu_hw_sspp *ctx = pipe->sspp;
-	u32 cdp_cntl = 0;
 	u32 cdp_cntl_offset = 0;
 
-	if (!ctx || !cfg)
+	if (!ctx)
 		return;
 
 	if (pipe->multirect_index == DPU_SSPP_RECT_SOLO ||
@@ -605,16 +605,7 @@ static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe,
 	else
 		cdp_cntl_offset = SSPP_CDP_CNTL_REC1;
 
-	if (cfg->enable)
-		cdp_cntl |= BIT(0);
-	if (cfg->ubwc_meta_enable)
-		cdp_cntl |= BIT(1);
-	if (cfg->tile_amortize_enable)
-		cdp_cntl |= BIT(2);
-	if (cfg->preload_ahead == DPU_SSPP_CDP_PRELOAD_AHEAD_64)
-		cdp_cntl |= BIT(3);
-
-	DPU_REG_WRITE(&ctx->hw, cdp_cntl_offset, cdp_cntl);
+	dpu_setup_cdp(&ctx->hw, cdp_cntl_offset, fmt, enable);
 }
 
 static void _setup_layer_ops(struct dpu_hw_sspp *c,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index 7a8d11ba618d7..86bf4b2cda778 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -177,14 +177,6 @@ struct dpu_hw_pipe_qos_cfg {
 	bool danger_safe_en;
 };
 
-/**
- * enum CDP preload ahead address size
- */
-enum {
-	DPU_SSPP_CDP_PRELOAD_AHEAD_32,
-	DPU_SSPP_CDP_PRELOAD_AHEAD_64
-};
-
 /**
  * struct dpu_hw_pipe_ts_cfg - traffic shaper configuration
  * @size: size to prefill in bytes, or zero to disable
@@ -331,10 +323,12 @@ struct dpu_hw_sspp_ops {
 	/**
 	 * setup_cdp - setup client driven prefetch
 	 * @pipe: Pointer to software pipe context
-	 * @cfg: Pointer to cdp configuration
+	 * @fmt: format used by the sw pipe
+	 * @enable: whether the CDP should be enabled for this pipe
 	 */
 	void (*setup_cdp)(struct dpu_sw_pipe *pipe,
-			  struct dpu_hw_cdp_cfg *cfg);
+			  const struct dpu_format *fmt,
+			  bool enable);
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
index 8062228eada68..95d20b9a3f2fa 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
@@ -494,3 +494,24 @@ int dpu_hw_collect_misr(struct dpu_hw_blk_reg_map *c,
 
 	return 0;
 }
+
+#define CDP_ENABLE		BIT(0)
+#define CDP_UBWC_META_ENABLE	BIT(1)
+#define CDP_TILE_AMORTIZE_ENABLE BIT(2)
+#define CDP_PRELOAD_AHEAD_64	BIT(3)
+
+void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset,
+		   const struct dpu_format *fmt, bool enable)
+{
+	u32 cdp_cntl = CDP_PRELOAD_AHEAD_64;
+
+	if (enable)
+		cdp_cntl |= CDP_ENABLE;
+	if (DPU_FORMAT_IS_UBWC(fmt))
+		cdp_cntl |= CDP_UBWC_META_ENABLE;
+	if (DPU_FORMAT_IS_UBWC(fmt) ||
+	    DPU_FORMAT_IS_TILE(fmt))
+		cdp_cntl |= CDP_TILE_AMORTIZE_ENABLE;
+
+	DPU_REG_WRITE(c, offset, cdp_cntl);
+}
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
index 27f4c39e35ab3..dc6e3b795aef6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
@@ -305,22 +305,6 @@ struct dpu_drm_scaler_v2 {
 	struct dpu_drm_de_v1 de;
 };
 
-/**
- * struct dpu_hw_cdp_cfg : CDP configuration
- * @enable: true to enable CDP
- * @ubwc_meta_enable: true to enable ubwc metadata preload
- * @tile_amortize_enable: true to enable amortization control for tile format
- * @preload_ahead: number of request to preload ahead
- *	DPU_*_CDP_PRELOAD_AHEAD_32,
- *	DPU_*_CDP_PRELOAD_AHEAD_64
- */
-struct dpu_hw_cdp_cfg {
-	bool enable;
-	bool ubwc_meta_enable;
-	bool tile_amortize_enable;
-	u32 preload_ahead;
-};
-
 u32 *dpu_hw_util_get_log_mask_ptr(void);
 
 void dpu_reg_write(struct dpu_hw_blk_reg_map *c,
@@ -346,6 +330,9 @@ void dpu_hw_csc_setup(struct dpu_hw_blk_reg_map  *c,
 		u32 csc_reg_off,
 		const struct dpu_csc_cfg *data, bool csc10);
 
+void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset,
+		   const struct dpu_format *fmt, bool enable);
+
 u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl,
 		u32 total_fl);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
index 4834866d72fd0..a6de4b82a1884 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
@@ -164,24 +164,13 @@ static void dpu_hw_wb_setup_qos_lut(struct dpu_hw_wb *ctx,
 }
 
 static void dpu_hw_wb_setup_cdp(struct dpu_hw_wb *ctx,
-		struct dpu_hw_cdp_cfg *cfg)
+				const struct dpu_format *fmt,
+				bool enable)
 {
-	struct dpu_hw_blk_reg_map *c;
-	u32 cdp_cntl = 0;
-
-	if (!ctx || !cfg)
+	if (!ctx)
 		return;
 
-	c = &ctx->hw;
-
-	if (cfg->enable)
-		cdp_cntl |= BIT(0);
-	if (cfg->ubwc_meta_enable)
-		cdp_cntl |= BIT(1);
-	if (cfg->preload_ahead == DPU_WB_CDP_PRELOAD_AHEAD_64)
-		cdp_cntl |= BIT(3);
-
-	DPU_REG_WRITE(c, WB_CDP_CNTL, cdp_cntl);
+	dpu_setup_cdp(&ctx->hw, WB_CDP_CNTL, fmt, enable);
 }
 
 static void dpu_hw_wb_bind_pingpong_blk(
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
index e4f85409c6248..ab3541856258d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
@@ -21,14 +21,6 @@ struct dpu_hw_wb_cfg {
 	struct drm_rect crop;
 };
 
-/**
- * enum CDP preload ahead address size
- */
-enum {
-	DPU_WB_CDP_PRELOAD_AHEAD_32,
-	DPU_WB_CDP_PRELOAD_AHEAD_64
-};
-
 /**
  * struct dpu_hw_wb_qos_cfg : Writeback pipe QoS configuration
  * @danger_lut: LUT for generate danger level based on fill level
@@ -67,7 +59,8 @@ struct dpu_hw_wb_ops {
 			struct dpu_hw_wb_qos_cfg *cfg);
 
 	void (*setup_cdp)(struct dpu_hw_wb *ctx,
-			struct dpu_hw_cdp_cfg *cfg);
+			  const struct dpu_format *fmt,
+			  bool enable);
 
 	void (*bind_pingpong_blk)(struct dpu_hw_wb *ctx,
 			bool enable, const enum dpu_pingpong pp);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 14b5cfe306113..3b210320ea624 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -1116,20 +1116,10 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 		pipe->sspp->ops.setup_format(pipe, fmt, src_flags);
 
 		if (pipe->sspp->ops.setup_cdp) {
-			struct dpu_hw_cdp_cfg cdp_cfg;
+			const struct dpu_perf_cfg *perf = pdpu->catalog->perf;
 
-			memset(&cdp_cfg, 0, sizeof(struct dpu_hw_cdp_cfg));
-
-			cdp_cfg.enable = pdpu->catalog->perf->cdp_cfg
-					[DPU_PERF_CDP_USAGE_RT].rd_enable;
-			cdp_cfg.ubwc_meta_enable =
-					DPU_FORMAT_IS_UBWC(fmt);
-			cdp_cfg.tile_amortize_enable =
-					DPU_FORMAT_IS_UBWC(fmt) ||
-					DPU_FORMAT_IS_TILE(fmt);
-			cdp_cfg.preload_ahead = DPU_SSPP_CDP_PRELOAD_AHEAD_64;
-
-			pipe->sspp->ops.setup_cdp(pipe, &cdp_cfg);
+			pipe->sspp->ops.setup_cdp(pipe, fmt,
+						  perf->cdp_cfg[DPU_PERF_CDP_USAGE_RT].rd_enable);
 		}
 	}
 
-- 
GitLab


From 2defa907ed5018fb7520bc9c3dc224961df5eda9 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:32 +0300
Subject: [PATCH 0761/1791] drm/msm/dpu: fix the condition for (not) applying
 QoS to CURSOR SSPP

The function dpu_plane_sspp_update_pipe() contains code to skip enabling
the QoS and OT limitis for CURSOR pipes. However all DPU since sdm845
repurpose DMA SSPP for the cursor planes because they lack the real
CURSOR SSPP. Fix the condition to actually check that the plane is
CURSOR or not.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537911/
Link: https://lore.kernel.org/r/20230518222238.3815293-4-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 3b210320ea624..b8ed7247a6af0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -1126,7 +1126,8 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 	_dpu_plane_set_qos_lut(plane, pipe, fmt, pipe_cfg);
 	_dpu_plane_set_danger_lut(plane, pipe, fmt);
 
-	if (plane->type != DRM_PLANE_TYPE_CURSOR) {
+	if (pipe->sspp->idx != SSPP_CURSOR0 &&
+	    pipe->sspp->idx != SSPP_CURSOR1) {
 		_dpu_plane_set_qos_ctrl(plane, pipe, true, DPU_PLANE_QOS_PANIC_CTRL);
 		_dpu_plane_set_ot_limit(plane, pipe, pipe_cfg, frame_rate);
 	}
-- 
GitLab


From 78c2925de971095e45aad4359f040dbe547714bd Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:33 +0300
Subject: [PATCH 0762/1791] drm/msm/dpu: rearrange QoS setting code

Slightly rearrainge code in dpu_plane_sspp_update_pipe() to group
QoS/LUT related functions.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537906/
Link: https://lore.kernel.org/r/20230518222238.3815293-5-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index b8ed7247a6af0..586f089756fa1 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -1079,10 +1079,10 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 		pipe->sspp->ops.setup_sourceaddress(pipe, layout);
 	}
 
-	_dpu_plane_set_qos_ctrl(plane, pipe, false, DPU_PLANE_QOS_PANIC_CTRL);
-
 	/* override for color fill */
 	if (pdpu->color_fill & DPU_PLANE_COLOR_FILL_FLAG) {
+		_dpu_plane_set_qos_ctrl(plane, pipe, false, DPU_PLANE_QOS_PANIC_CTRL);
+
 		/* skip remaining processing on color fill */
 		return;
 	}
@@ -1125,12 +1125,14 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 
 	_dpu_plane_set_qos_lut(plane, pipe, fmt, pipe_cfg);
 	_dpu_plane_set_danger_lut(plane, pipe, fmt);
+	_dpu_plane_set_qos_ctrl(plane, pipe,
+				pipe->sspp->idx != SSPP_CURSOR0 &&
+				pipe->sspp->idx != SSPP_CURSOR1,
+				DPU_PLANE_QOS_PANIC_CTRL);
 
 	if (pipe->sspp->idx != SSPP_CURSOR0 &&
-	    pipe->sspp->idx != SSPP_CURSOR1) {
-		_dpu_plane_set_qos_ctrl(plane, pipe, true, DPU_PLANE_QOS_PANIC_CTRL);
+	    pipe->sspp->idx != SSPP_CURSOR1)
 		_dpu_plane_set_ot_limit(plane, pipe, pipe_cfg, frame_rate);
-	}
 
 	if (pstate->needs_qos_remap)
 		_dpu_plane_set_qos_remap(plane, pipe);
-- 
GitLab


From d5f86e50f0319966503cbd54a2116d838d5984bc Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:34 +0300
Subject: [PATCH 0763/1791] drm/msm/dpu: drop DPU_PLANE_QOS_VBLANK_CTRL

Drop support for DPU_PLANE_QOS_VBLANK_CTRL flag. It is not used both
in upstream driver and in vendor SDE driver.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537907/
Link: https://lore.kernel.org/r/20230518222238.3815293-6-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h |  4 ----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c    |  8 --------
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c      | 10 ----------
 3 files changed, 22 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 677048cc3b7d7..6b1e8383d2c4f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -374,8 +374,6 @@ struct dpu_caps {
 /**
  * struct dpu_sspp_sub_blks : SSPP sub-blocks
  * common: Pointer to common configurations shared by sub blocks
- * @creq_vblank: creq priority during vertical blanking
- * @danger_vblank: danger priority during vertical blanking
  * @maxdwnscale: max downscale ratio supported(without DECIMATION)
  * @maxupscale:  maxupscale ratio supported
  * @smart_dma_priority: hw priority of rect1 of multirect pipe
@@ -390,8 +388,6 @@ struct dpu_caps {
  * @dpu_rotation_cfg: inline rotation configuration
  */
 struct dpu_sspp_sub_blks {
-	u32 creq_vblank;
-	u32 danger_vblank;
 	u32 maxdwnscale;
 	u32 maxupscale;
 	u32 smart_dma_priority;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 78d928065b7bd..dde02278e0067 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -695,14 +695,6 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms,
 			0400,
 			debugfs_root,
 			(u32 *) &cfg->clk_ctrl);
-	debugfs_create_x32("creq_vblank",
-			0600,
-			debugfs_root,
-			(u32 *) &sblk->creq_vblank);
-	debugfs_create_x32("danger_vblank",
-			0600,
-			debugfs_root,
-			(u32 *) &sblk->danger_vblank);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 586f089756fa1..3cb891917b65d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -73,13 +73,11 @@ static const uint32_t qcom_compressed_supported_formats[] = {
 /**
  * enum dpu_plane_qos - Different qos configurations for each pipe
  *
- * @DPU_PLANE_QOS_VBLANK_CTRL: Setup VBLANK qos for the pipe.
  * @DPU_PLANE_QOS_VBLANK_AMORTIZE: Enables Amortization within pipe.
  *	this configuration is mutually exclusive from VBLANK_CTRL.
  * @DPU_PLANE_QOS_PANIC_CTRL: Setup panic for the pipe.
  */
 enum dpu_plane_qos {
-	DPU_PLANE_QOS_VBLANK_CTRL = BIT(0),
 	DPU_PLANE_QOS_VBLANK_AMORTIZE = BIT(1),
 	DPU_PLANE_QOS_PANIC_CTRL = BIT(2),
 };
@@ -361,15 +359,7 @@ static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 
 	memset(&pipe_qos_cfg, 0, sizeof(pipe_qos_cfg));
 
-	if (flags & DPU_PLANE_QOS_VBLANK_CTRL) {
-		pipe_qos_cfg.creq_vblank = pipe->sspp->cap->sblk->creq_vblank;
-		pipe_qos_cfg.danger_vblank =
-				pipe->sspp->cap->sblk->danger_vblank;
-		pipe_qos_cfg.vblank_en = enable;
-	}
-
 	if (flags & DPU_PLANE_QOS_VBLANK_AMORTIZE) {
-		/* this feature overrules previous VBLANK_CTRL */
 		pipe_qos_cfg.vblank_en = false;
 		pipe_qos_cfg.creq_vblank = 0; /* clear vblank bits */
 	}
-- 
GitLab


From f68098003dc351c06e1aba1fbccddfbbf7fa2110 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:35 +0300
Subject: [PATCH 0764/1791] drm/msm/dpu: simplify qos_ctrl handling

After removal of DPU_PLANE_QOS_VBLANK_CTRL, several fields of struct
dpu_hw_pipe_qos_cfg are fixed to false/0. Drop them from the structure
(and drop the corresponding code from the functions).

The DPU_PLANE_QOS_VBLANK_AMORTIZE flag is also removed, since it is now
a NOP.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537909/
Link: https://lore.kernel.org/r/20230518222238.3815293-7-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 10 ----------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h |  6 ------
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c   | 17 ++---------------
 3 files changed, 2 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index dde02278e0067..925689dc77f00 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -573,16 +573,6 @@ static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
 	if (!ctx)
 		return;
 
-	if (cfg->vblank_en) {
-		qos_ctrl |= ((cfg->creq_vblank &
-				SSPP_QOS_CTRL_CREQ_VBLANK_MASK) <<
-				SSPP_QOS_CTRL_CREQ_VBLANK_OFF);
-		qos_ctrl |= ((cfg->danger_vblank &
-				SSPP_QOS_CTRL_DANGER_VBLANK_MASK) <<
-				SSPP_QOS_CTRL_DANGER_VBLANK_OFF);
-		qos_ctrl |= SSPP_QOS_CTRL_VBLANK_EN;
-	}
-
 	if (cfg->danger_safe_en)
 		qos_ctrl |= SSPP_QOS_CTRL_DANGER_SAFE_EN;
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index 86bf4b2cda778..aaf6f41d546c9 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -165,15 +165,9 @@ struct dpu_sw_pipe_cfg {
 
 /**
  * struct dpu_hw_pipe_qos_cfg : Source pipe QoS configuration
- * @creq_vblank: creq value generated to vbif during vertical blanking
- * @danger_vblank: danger value generated during vertical blanking
- * @vblank_en: enable creq_vblank and danger_vblank during vblank
  * @danger_safe_en: enable danger safe generation
  */
 struct dpu_hw_pipe_qos_cfg {
-	u32 creq_vblank;
-	u32 danger_vblank;
-	bool vblank_en;
 	bool danger_safe_en;
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 3cb891917b65d..0ed350776775c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -73,12 +73,9 @@ static const uint32_t qcom_compressed_supported_formats[] = {
 /**
  * enum dpu_plane_qos - Different qos configurations for each pipe
  *
- * @DPU_PLANE_QOS_VBLANK_AMORTIZE: Enables Amortization within pipe.
- *	this configuration is mutually exclusive from VBLANK_CTRL.
  * @DPU_PLANE_QOS_PANIC_CTRL: Setup panic for the pipe.
  */
 enum dpu_plane_qos {
-	DPU_PLANE_QOS_VBLANK_AMORTIZE = BIT(1),
 	DPU_PLANE_QOS_PANIC_CTRL = BIT(2),
 };
 
@@ -359,25 +356,15 @@ static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 
 	memset(&pipe_qos_cfg, 0, sizeof(pipe_qos_cfg));
 
-	if (flags & DPU_PLANE_QOS_VBLANK_AMORTIZE) {
-		pipe_qos_cfg.vblank_en = false;
-		pipe_qos_cfg.creq_vblank = 0; /* clear vblank bits */
-	}
-
 	if (flags & DPU_PLANE_QOS_PANIC_CTRL)
 		pipe_qos_cfg.danger_safe_en = enable;
 
-	if (!pdpu->is_rt_pipe) {
-		pipe_qos_cfg.vblank_en = false;
+	if (!pdpu->is_rt_pipe)
 		pipe_qos_cfg.danger_safe_en = false;
-	}
 
-	DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d vb:%d pri[0x%x, 0x%x] is_rt:%d\n",
+	DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n",
 		pdpu->pipe - SSPP_VIG0,
 		pipe_qos_cfg.danger_safe_en,
-		pipe_qos_cfg.vblank_en,
-		pipe_qos_cfg.creq_vblank,
-		pipe_qos_cfg.danger_vblank,
 		pdpu->is_rt_pipe);
 
 	pipe->sspp->ops.setup_qos_ctrl(pipe->sspp,
-- 
GitLab


From 9ccff1d2c062302fe20ebbdee1831d0933d32a04 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:36 +0300
Subject: [PATCH 0765/1791] drm/msm/dpu: drop DPU_PLANE_QOS_PANIC_CTRL

This flag is always passed to _dpu_plane_set_qos_ctrl(), so drop it and
remove corresponding conditions from the mentioned function.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537916/
Link: https://lore.kernel.org/r/20230518222238.3815293-8-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 27 +++++++----------------
 1 file changed, 8 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 0ed350776775c..d1443c4b2915f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -70,15 +70,6 @@ static const uint32_t qcom_compressed_supported_formats[] = {
 	DRM_FORMAT_P010,
 };
 
-/**
- * enum dpu_plane_qos - Different qos configurations for each pipe
- *
- * @DPU_PLANE_QOS_PANIC_CTRL: Setup panic for the pipe.
- */
-enum dpu_plane_qos {
-	DPU_PLANE_QOS_PANIC_CTRL = BIT(2),
-};
-
 /*
  * struct dpu_plane - local dpu plane structure
  * @aspace: address space pointer
@@ -349,15 +340,14 @@ static void _dpu_plane_set_danger_lut(struct drm_plane *plane,
  */
 static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 	struct dpu_sw_pipe *pipe,
-	bool enable, u32 flags)
+	bool enable)
 {
 	struct dpu_plane *pdpu = to_dpu_plane(plane);
 	struct dpu_hw_pipe_qos_cfg pipe_qos_cfg;
 
 	memset(&pipe_qos_cfg, 0, sizeof(pipe_qos_cfg));
 
-	if (flags & DPU_PLANE_QOS_PANIC_CTRL)
-		pipe_qos_cfg.danger_safe_en = enable;
+	pipe_qos_cfg.danger_safe_en = enable;
 
 	if (!pdpu->is_rt_pipe)
 		pipe_qos_cfg.danger_safe_en = false;
@@ -1058,7 +1048,7 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 
 	/* override for color fill */
 	if (pdpu->color_fill & DPU_PLANE_COLOR_FILL_FLAG) {
-		_dpu_plane_set_qos_ctrl(plane, pipe, false, DPU_PLANE_QOS_PANIC_CTRL);
+		_dpu_plane_set_qos_ctrl(plane, pipe, false);
 
 		/* skip remaining processing on color fill */
 		return;
@@ -1104,8 +1094,7 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 	_dpu_plane_set_danger_lut(plane, pipe, fmt);
 	_dpu_plane_set_qos_ctrl(plane, pipe,
 				pipe->sspp->idx != SSPP_CURSOR0 &&
-				pipe->sspp->idx != SSPP_CURSOR1,
-				DPU_PLANE_QOS_PANIC_CTRL);
+				pipe->sspp->idx != SSPP_CURSOR1);
 
 	if (pipe->sspp->idx != SSPP_CURSOR0 &&
 	    pipe->sspp->idx != SSPP_CURSOR1)
@@ -1224,10 +1213,10 @@ static void dpu_plane_destroy(struct drm_plane *plane)
 
 	if (pdpu) {
 		pstate = to_dpu_plane_state(plane->state);
-		_dpu_plane_set_qos_ctrl(plane, &pstate->pipe, false, DPU_PLANE_QOS_PANIC_CTRL);
+		_dpu_plane_set_qos_ctrl(plane, &pstate->pipe, false);
 
 		if (pstate->r_pipe.sspp)
-			_dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, false, DPU_PLANE_QOS_PANIC_CTRL);
+			_dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, false);
 
 		mutex_destroy(&pdpu->lock);
 
@@ -1384,9 +1373,9 @@ void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable)
 		return;
 
 	pm_runtime_get_sync(&dpu_kms->pdev->dev);
-	_dpu_plane_set_qos_ctrl(plane, &pstate->pipe, enable, DPU_PLANE_QOS_PANIC_CTRL);
+	_dpu_plane_set_qos_ctrl(plane, &pstate->pipe, enable);
 	if (pstate->r_pipe.sspp)
-		_dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, enable, DPU_PLANE_QOS_PANIC_CTRL);
+		_dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, enable);
 	pm_runtime_put_sync(&dpu_kms->pdev->dev);
 }
 #endif
-- 
GitLab


From a5ebb27bffcc2c1e785abc6782202d9e4041e71c Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:37 +0300
Subject: [PATCH 0766/1791] drm/msm/dpu: remove struct dpu_hw_pipe_qos_cfg

Now as the struct dpu_hw_pipe_qos_cfg consists of only one bool field,
drop the structure and use corresponding bool directly.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537918/
Link: https://lore.kernel.org/r/20230518222238.3815293-9-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 10 +++-------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h | 13 ++-----------
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c   | 11 +++--------
 3 files changed, 8 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 925689dc77f00..1abe1eca4f423 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -566,17 +566,13 @@ static void dpu_hw_sspp_setup_creq_lut(struct dpu_hw_sspp *ctx,
 }
 
 static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
-		struct dpu_hw_pipe_qos_cfg *cfg)
+				       bool danger_safe_en)
 {
-	u32 qos_ctrl = 0;
-
 	if (!ctx)
 		return;
 
-	if (cfg->danger_safe_en)
-		qos_ctrl |= SSPP_QOS_CTRL_DANGER_SAFE_EN;
-
-	DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL, qos_ctrl);
+	DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL,
+		      danger_safe_en ? SSPP_QOS_CTRL_DANGER_SAFE_EN : 0);
 }
 
 static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index aaf6f41d546c9..4278c421b6acd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -163,14 +163,6 @@ struct dpu_sw_pipe_cfg {
 	struct drm_rect dst_rect;
 };
 
-/**
- * struct dpu_hw_pipe_qos_cfg : Source pipe QoS configuration
- * @danger_safe_en: enable danger safe generation
- */
-struct dpu_hw_pipe_qos_cfg {
-	bool danger_safe_en;
-};
-
 /**
  * struct dpu_hw_pipe_ts_cfg - traffic shaper configuration
  * @size: size to prefill in bytes, or zero to disable
@@ -285,11 +277,10 @@ struct dpu_hw_sspp_ops {
 	/**
 	 * setup_qos_ctrl - setup QoS control
 	 * @ctx: Pointer to pipe context
-	 * @cfg: Pointer to pipe QoS configuration
-	 *
+	 * @danger_safe_en: flags controlling enabling of danger/safe QoS/LUT
 	 */
 	void (*setup_qos_ctrl)(struct dpu_hw_sspp *ctx,
-			struct dpu_hw_pipe_qos_cfg *cfg);
+			       bool danger_safe_en);
 
 	/**
 	 * setup_histogram - setup histograms
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index d1443c4b2915f..c8837d0aa0c35 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -343,22 +343,17 @@ static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 	bool enable)
 {
 	struct dpu_plane *pdpu = to_dpu_plane(plane);
-	struct dpu_hw_pipe_qos_cfg pipe_qos_cfg;
-
-	memset(&pipe_qos_cfg, 0, sizeof(pipe_qos_cfg));
-
-	pipe_qos_cfg.danger_safe_en = enable;
 
 	if (!pdpu->is_rt_pipe)
-		pipe_qos_cfg.danger_safe_en = false;
+		enable = false;
 
 	DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n",
 		pdpu->pipe - SSPP_VIG0,
-		pipe_qos_cfg.danger_safe_en,
+		enable,
 		pdpu->is_rt_pipe);
 
 	pipe->sspp->ops.setup_qos_ctrl(pipe->sspp,
-			&pipe_qos_cfg);
+				       enable);
 }
 
 /**
-- 
GitLab


From 5fe0faa62461adb578785169f29f3c4638ca4e9a Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Fri, 19 May 2023 01:22:38 +0300
Subject: [PATCH 0767/1791] drm/msm/dpu: use common helper for WB and SSPP QoS
 setup

Rework SSPP and WB code to use common helper for programming QoS
settings.

Reviewed-by: Jeykumar Sankaran <quic_jeykumar@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/537912/
Link: https://lore.kernel.org/r/20230518222238.3815293-10-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   |  4 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c   | 31 ++-----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h   | 19 +----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c   | 31 +++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h   | 21 +++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c     | 29 +------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h     | 16 +---
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c     | 85 +++++++------------
 8 files changed, 100 insertions(+), 136 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index e7b65f6f53d64..023a9c4ad1db4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -102,7 +102,7 @@ static void dpu_encoder_phys_wb_set_qos_remap(
 static void dpu_encoder_phys_wb_set_qos(struct dpu_encoder_phys *phys_enc)
 {
 	struct dpu_hw_wb *hw_wb;
-	struct dpu_hw_wb_qos_cfg qos_cfg;
+	struct dpu_hw_qos_cfg qos_cfg;
 	const struct dpu_mdss_cfg *catalog;
 	const struct dpu_qos_lut_tbl *qos_lut_tb;
 
@@ -115,7 +115,7 @@ static void dpu_encoder_phys_wb_set_qos(struct dpu_encoder_phys *phys_enc)
 
 	hw_wb = phys_enc->hw_wb;
 
-	memset(&qos_cfg, 0, sizeof(struct dpu_hw_wb_qos_cfg));
+	memset(&qos_cfg, 0, sizeof(struct dpu_hw_qos_cfg));
 	qos_cfg.danger_safe_en = true;
 	qos_cfg.danger_lut =
 		catalog->perf->danger_lut_tbl[DPU_QOS_LUT_USAGE_NRT];
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
index 1abe1eca4f423..b364cf75bb3f1 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c
@@ -539,30 +539,15 @@ static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color)
 				color);
 }
 
-static void dpu_hw_sspp_setup_danger_safe_lut(struct dpu_hw_sspp *ctx,
-			u32 danger_lut,
-			u32 safe_lut)
+static void dpu_hw_sspp_setup_qos_lut(struct dpu_hw_sspp *ctx,
+				      struct dpu_hw_qos_cfg *cfg)
 {
-	if (!ctx)
-		return;
-
-	DPU_REG_WRITE(&ctx->hw, SSPP_DANGER_LUT, danger_lut);
-	DPU_REG_WRITE(&ctx->hw, SSPP_SAFE_LUT, safe_lut);
-}
-
-static void dpu_hw_sspp_setup_creq_lut(struct dpu_hw_sspp *ctx,
-			u64 creq_lut)
-{
-	if (!ctx)
+	if (!ctx || !cfg)
 		return;
 
-	if (ctx->cap && test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features)) {
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_0, creq_lut);
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_1,
-				creq_lut >> 32);
-	} else {
-		DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT, creq_lut);
-	}
+	_dpu_hw_setup_qos_lut(&ctx->hw, SSPP_DANGER_LUT,
+			      test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features),
+			      cfg);
 }
 
 static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx,
@@ -604,9 +589,7 @@ static void _setup_layer_ops(struct dpu_hw_sspp *c,
 	c->ops.setup_pe = dpu_hw_sspp_setup_pe_config;
 
 	if (test_bit(DPU_SSPP_QOS, &features)) {
-		c->ops.setup_danger_safe_lut =
-			dpu_hw_sspp_setup_danger_safe_lut;
-		c->ops.setup_creq_lut = dpu_hw_sspp_setup_creq_lut;
+		c->ops.setup_qos_lut = dpu_hw_sspp_setup_qos_lut;
 		c->ops.setup_qos_ctrl = dpu_hw_sspp_setup_qos_ctrl;
 	}
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
index 4278c421b6acd..085f34bc6b88d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h
@@ -254,25 +254,14 @@ struct dpu_hw_sspp_ops {
 	void (*setup_sharpening)(struct dpu_hw_sspp *ctx,
 			struct dpu_hw_sharp_cfg *cfg);
 
-	/**
-	 * setup_danger_safe_lut - setup danger safe LUTs
-	 * @ctx: Pointer to pipe context
-	 * @danger_lut: LUT for generate danger level based on fill level
-	 * @safe_lut: LUT for generate safe level based on fill level
-	 *
-	 */
-	void (*setup_danger_safe_lut)(struct dpu_hw_sspp *ctx,
-			u32 danger_lut,
-			u32 safe_lut);
 
 	/**
-	 * setup_creq_lut - setup CREQ LUT
+	 * setup_qos_lut - setup QoS LUTs
 	 * @ctx: Pointer to pipe context
-	 * @creq_lut: LUT for generate creq level based on fill level
-	 *
+	 * @cfg: LUT configuration
 	 */
-	void (*setup_creq_lut)(struct dpu_hw_sspp *ctx,
-			u64 creq_lut);
+	void (*setup_qos_lut)(struct dpu_hw_sspp *ctx,
+			struct dpu_hw_qos_cfg *cfg);
 
 	/**
 	 * setup_qos_ctrl - setup QoS control
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
index 95d20b9a3f2fa..9d2273fd2fed5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c
@@ -73,6 +73,19 @@ static u32 dpu_hw_util_log_mask = DPU_DBG_MASK_NONE;
 #define QSEED3LITE_SEP_LUT_SIZE \
 	        (QSEED3LITE_LUT_SIZE * QSEED3LITE_SEPARABLE_LUTS * sizeof(u32))
 
+/* QOS_LUT */
+#define QOS_DANGER_LUT                    0x00
+#define QOS_SAFE_LUT                      0x04
+#define QOS_CREQ_LUT                      0x08
+#define QOS_QOS_CTRL                      0x0C
+#define QOS_CREQ_LUT_0                    0x14
+#define QOS_CREQ_LUT_1                    0x18
+
+/* QOS_QOS_CTRL */
+#define QOS_QOS_CTRL_DANGER_SAFE_EN       BIT(0)
+#define QOS_QOS_CTRL_DANGER_VBLANK_MASK   GENMASK(5, 4)
+#define QOS_QOS_CTRL_VBLANK_EN            BIT(16)
+#define QOS_QOS_CTRL_CREQ_VBLANK_MASK     GENMASK(21, 20)
 
 void dpu_reg_write(struct dpu_hw_blk_reg_map *c,
 		u32 reg_off,
@@ -450,6 +463,24 @@ u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl,
 	return 0;
 }
 
+void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset,
+			   bool qos_8lvl,
+			   const struct dpu_hw_qos_cfg *cfg)
+{
+	DPU_REG_WRITE(c, offset + QOS_DANGER_LUT, cfg->danger_lut);
+	DPU_REG_WRITE(c, offset + QOS_SAFE_LUT, cfg->safe_lut);
+
+	if (qos_8lvl) {
+		DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_0, cfg->creq_lut);
+		DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_1, cfg->creq_lut >> 32);
+	} else {
+		DPU_REG_WRITE(c, offset + QOS_CREQ_LUT, cfg->creq_lut);
+	}
+
+	DPU_REG_WRITE(c, offset + QOS_QOS_CTRL,
+		      cfg->danger_safe_en ? QOS_QOS_CTRL_DANGER_SAFE_EN : 0);
+}
+
 void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c,
 		u32 misr_ctrl_offset,
 		bool enable, u32 frame_count)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
index dc6e3b795aef6..1f6079f470710 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h
@@ -305,6 +305,23 @@ struct dpu_drm_scaler_v2 {
 	struct dpu_drm_de_v1 de;
 };
 
+/**
+ * struct dpu_hw_qos_cfg: pipe QoS configuration
+ * @danger_lut: LUT for generate danger level based on fill level
+ * @safe_lut: LUT for generate safe level based on fill level
+ * @creq_lut: LUT for generate creq level based on fill level
+ * @creq_vblank: creq value generated to vbif during vertical blanking
+ * @danger_vblank: danger value generated during vertical blanking
+ * @vblank_en: enable creq_vblank and danger_vblank during vblank
+ * @danger_safe_en: enable danger safe generation
+ */
+struct dpu_hw_qos_cfg {
+	u32 danger_lut;
+	u32 safe_lut;
+	u64 creq_lut;
+	bool danger_safe_en;
+};
+
 u32 *dpu_hw_util_get_log_mask_ptr(void);
 
 void dpu_reg_write(struct dpu_hw_blk_reg_map *c,
@@ -336,6 +353,10 @@ void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset,
 u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl,
 		u32 total_fl);
 
+void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset,
+			   bool qos_8lvl,
+			   const struct dpu_hw_qos_cfg *cfg);
+
 void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c,
 		u32 misr_ctrl_offset,
 		bool enable,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
index a6de4b82a1884..dcffd6cc47fc8 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
@@ -49,9 +49,6 @@
 #define WB_OUT_IMAGE_SIZE                     0x2C0
 #define WB_OUT_XY                             0x2C4
 
-/* WB_QOS_CTRL */
-#define WB_QOS_CTRL_DANGER_SAFE_EN            BIT(0)
-
 static void dpu_hw_wb_setup_outaddress(struct dpu_hw_wb *ctx,
 		struct dpu_hw_wb_cfg *data)
 {
@@ -135,32 +132,14 @@ static void dpu_hw_wb_roi(struct dpu_hw_wb *ctx, struct dpu_hw_wb_cfg *wb)
 }
 
 static void dpu_hw_wb_setup_qos_lut(struct dpu_hw_wb *ctx,
-		struct dpu_hw_wb_qos_cfg *cfg)
+		struct dpu_hw_qos_cfg *cfg)
 {
-	struct dpu_hw_blk_reg_map *c = &ctx->hw;
-	u32 qos_ctrl = 0;
-
 	if (!ctx || !cfg)
 		return;
 
-	DPU_REG_WRITE(c, WB_DANGER_LUT, cfg->danger_lut);
-	DPU_REG_WRITE(c, WB_SAFE_LUT, cfg->safe_lut);
-
-	/*
-	 * for chipsets not using DPU_WB_QOS_8LVL but still using DPU
-	 * driver such as msm8998, the reset value of WB_CREQ_LUT is
-	 * sufficient for writeback to work. SW doesn't need to explicitly
-	 * program a value.
-	 */
-	if (ctx->caps && test_bit(DPU_WB_QOS_8LVL, &ctx->caps->features)) {
-		DPU_REG_WRITE(c, WB_CREQ_LUT_0, cfg->creq_lut);
-		DPU_REG_WRITE(c, WB_CREQ_LUT_1, cfg->creq_lut >> 32);
-	}
-
-	if (cfg->danger_safe_en)
-		qos_ctrl |= WB_QOS_CTRL_DANGER_SAFE_EN;
-
-	DPU_REG_WRITE(c, WB_QOS_CTRL, qos_ctrl);
+	_dpu_hw_setup_qos_lut(&ctx->hw, WB_DANGER_LUT,
+			      test_bit(DPU_WB_QOS_8LVL, &ctx->caps->features),
+			      cfg);
 }
 
 static void dpu_hw_wb_setup_cdp(struct dpu_hw_wb *ctx,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
index ab3541856258d..c7f792eeb55ca 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
@@ -21,20 +21,6 @@ struct dpu_hw_wb_cfg {
 	struct drm_rect crop;
 };
 
-/**
- * struct dpu_hw_wb_qos_cfg : Writeback pipe QoS configuration
- * @danger_lut: LUT for generate danger level based on fill level
- * @safe_lut: LUT for generate safe level based on fill level
- * @creq_lut: LUT for generate creq level based on fill level
- * @danger_safe_en: enable danger safe generation
- */
-struct dpu_hw_wb_qos_cfg {
-	u32 danger_lut;
-	u32 safe_lut;
-	u64 creq_lut;
-	bool danger_safe_en;
-};
-
 /**
  *
  * struct dpu_hw_wb_ops : Interface to the wb hw driver functions
@@ -56,7 +42,7 @@ struct dpu_hw_wb_ops {
 			struct dpu_hw_wb_cfg *wb);
 
 	void (*setup_qos_lut)(struct dpu_hw_wb *ctx,
-			struct dpu_hw_wb_qos_cfg *cfg);
+			struct dpu_hw_qos_cfg *cfg);
 
 	void (*setup_cdp)(struct dpu_hw_wb *ctx,
 			  const struct dpu_format *fmt,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index c8837d0aa0c35..d66745115917b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -190,12 +190,14 @@ static u64 _dpu_plane_calc_clk(const struct drm_display_mode *mode,
  * _dpu_plane_calc_fill_level - calculate fill level of the given source format
  * @plane:		Pointer to drm plane
  * @pipe:		Pointer to software pipe
+ * @lut_usage:		LUT usecase
  * @fmt:		Pointer to source buffer format
  * @src_width:		width of source buffer
  * Return: fill level corresponding to the source buffer/format or 0 if error
  */
 static int _dpu_plane_calc_fill_level(struct drm_plane *plane,
 		struct dpu_sw_pipe *pipe,
+		enum dpu_qos_lut_usage lut_usage,
 		const struct dpu_format *fmt, u32 src_width)
 {
 	struct dpu_plane *pdpu;
@@ -207,6 +209,9 @@ static int _dpu_plane_calc_fill_level(struct drm_plane *plane,
 		return 0;
 	}
 
+	if (lut_usage == DPU_QOS_LUT_USAGE_NRT)
+		return 0;
+
 	pdpu = to_dpu_plane(plane);
 	fixed_buff_size = pdpu->catalog->caps->pixel_ram_size;
 
@@ -252,83 +257,58 @@ static void _dpu_plane_set_qos_lut(struct drm_plane *plane,
 		const struct dpu_format *fmt, struct dpu_sw_pipe_cfg *pipe_cfg)
 {
 	struct dpu_plane *pdpu = to_dpu_plane(plane);
-	u64 qos_lut;
-	u32 total_fl = 0, lut_usage;
+	struct dpu_hw_qos_cfg cfg;
+	u32 total_fl, lut_usage;
 
 	if (!pdpu->is_rt_pipe) {
 		lut_usage = DPU_QOS_LUT_USAGE_NRT;
 	} else {
-		total_fl = _dpu_plane_calc_fill_level(plane, pipe, fmt,
-				drm_rect_width(&pipe_cfg->src_rect));
-
 		if (fmt && DPU_FORMAT_IS_LINEAR(fmt))
 			lut_usage = DPU_QOS_LUT_USAGE_LINEAR;
 		else
 			lut_usage = DPU_QOS_LUT_USAGE_MACROTILE;
 	}
 
-	qos_lut = _dpu_hw_get_qos_lut(
-			&pdpu->catalog->perf->qos_lut_tbl[lut_usage], total_fl);
+	total_fl = _dpu_plane_calc_fill_level(plane, pipe, lut_usage, fmt,
+				drm_rect_width(&pipe_cfg->src_rect));
+
+	cfg.creq_lut = _dpu_hw_get_qos_lut(&pdpu->catalog->perf->qos_lut_tbl[lut_usage], total_fl);
+	cfg.danger_lut = pdpu->catalog->perf->danger_lut_tbl[lut_usage];
+	cfg.safe_lut = pdpu->catalog->perf->safe_lut_tbl[lut_usage];
+
+	if (pipe->sspp->idx != SSPP_CURSOR0 &&
+	    pipe->sspp->idx != SSPP_CURSOR1 &&
+	    pdpu->is_rt_pipe)
+		cfg.danger_safe_en = true;
+
+	DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n",
+		pdpu->pipe - SSPP_VIG0,
+		cfg.danger_safe_en,
+		pdpu->is_rt_pipe);
 
 	trace_dpu_perf_set_qos_luts(pipe->sspp->idx - SSPP_VIG0,
 			(fmt) ? fmt->base.pixel_format : 0,
-			pdpu->is_rt_pipe, total_fl, qos_lut, lut_usage);
+			pdpu->is_rt_pipe, total_fl, cfg.creq_lut, lut_usage);
 
 	DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %4.4s rt:%d fl:%u lut:0x%llx\n",
 			pdpu->pipe - SSPP_VIG0,
 			fmt ? (char *)&fmt->base.pixel_format : NULL,
-			pdpu->is_rt_pipe, total_fl, qos_lut);
-
-	pipe->sspp->ops.setup_creq_lut(pipe->sspp, qos_lut);
-}
-
-/**
- * _dpu_plane_set_danger_lut - set danger/safe LUT of the given plane
- * @plane:		Pointer to drm plane
- * @pipe:		Pointer to software pipe
- * @fmt:		Pointer to source buffer format
- */
-static void _dpu_plane_set_danger_lut(struct drm_plane *plane,
-		struct dpu_sw_pipe *pipe,
-		const struct dpu_format *fmt)
-{
-	struct dpu_plane *pdpu = to_dpu_plane(plane);
-	u32 danger_lut, safe_lut;
-
-	if (!pdpu->is_rt_pipe) {
-		danger_lut = pdpu->catalog->perf->danger_lut_tbl
-				[DPU_QOS_LUT_USAGE_NRT];
-		safe_lut = pdpu->catalog->perf->safe_lut_tbl
-				[DPU_QOS_LUT_USAGE_NRT];
-	} else {
-		if (fmt && DPU_FORMAT_IS_LINEAR(fmt)) {
-			danger_lut = pdpu->catalog->perf->danger_lut_tbl
-					[DPU_QOS_LUT_USAGE_LINEAR];
-			safe_lut = pdpu->catalog->perf->safe_lut_tbl
-					[DPU_QOS_LUT_USAGE_LINEAR];
-		} else {
-			danger_lut = pdpu->catalog->perf->danger_lut_tbl
-					[DPU_QOS_LUT_USAGE_MACROTILE];
-			safe_lut = pdpu->catalog->perf->safe_lut_tbl
-					[DPU_QOS_LUT_USAGE_MACROTILE];
-		}
-	}
+			pdpu->is_rt_pipe, total_fl, cfg.creq_lut);
 
 	trace_dpu_perf_set_danger_luts(pdpu->pipe - SSPP_VIG0,
 			(fmt) ? fmt->base.pixel_format : 0,
 			(fmt) ? fmt->fetch_mode : 0,
-			danger_lut,
-			safe_lut);
+			cfg.danger_lut,
+			cfg.safe_lut);
 
 	DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %4.4s mode:%d luts[0x%x, 0x%x]\n",
 		pdpu->pipe - SSPP_VIG0,
 		fmt ? (char *)&fmt->base.pixel_format : NULL,
 		fmt ? fmt->fetch_mode : -1,
-		danger_lut,
-		safe_lut);
+		cfg.danger_lut,
+		cfg.safe_lut);
 
-	pipe->sspp->ops.setup_danger_safe_lut(pipe->sspp,
-			danger_lut, safe_lut);
+	pipe->sspp->ops.setup_qos_lut(pipe->sspp, &cfg);
 }
 
 /**
@@ -336,7 +316,6 @@ static void _dpu_plane_set_danger_lut(struct drm_plane *plane,
  * @plane:		Pointer to drm plane
  * @pipe:		Pointer to software pipe
  * @enable:		true to enable QoS control
- * @flags:		QoS control mode (enum dpu_plane_qos)
  */
 static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane,
 	struct dpu_sw_pipe *pipe,
@@ -1086,10 +1065,6 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane,
 	}
 
 	_dpu_plane_set_qos_lut(plane, pipe, fmt, pipe_cfg);
-	_dpu_plane_set_danger_lut(plane, pipe, fmt);
-	_dpu_plane_set_qos_ctrl(plane, pipe,
-				pipe->sspp->idx != SSPP_CURSOR0 &&
-				pipe->sspp->idx != SSPP_CURSOR1);
 
 	if (pipe->sspp->idx != SSPP_CURSOR0 &&
 	    pipe->sspp->idx != SSPP_CURSOR1)
-- 
GitLab


From 6de6c28d97c02b5abdb208975cf392942e462424 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 23:01:02 +0300
Subject: [PATCH 0768/1791] drm/msm/dsi: remove extra call to
 dsi_get_pclk_rate()

In dsi_calc_clk_rate_v2() there is no need to call dsi_get_pclk_rate().
This function has just been called (from dsi_calc_pclk()) and its
result is stored at msm_host->pixel_clk_rate. Use this variable
directly.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/538272/
Link: https://lore.kernel.org/r/20230520200103.4019607-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 961689a255c47..2b257b4599744 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -633,7 +633,7 @@ int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 
 	dsi_calc_pclk(msm_host, is_bonded_dsi);
 
-	pclk_bpp = (u64)dsi_get_pclk_rate(msm_host->mode, is_bonded_dsi) * bpp;
+	pclk_bpp = (u64)msm_host->pixel_clk_rate * bpp;
 	do_div(pclk_bpp, 8);
 	msm_host->src_clk_rate = pclk_bpp;
 
-- 
GitLab


From 374918d2c396c7099c4bcd04ee044e8b0456afce Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 23:01:03 +0300
Subject: [PATCH 0769/1791] drm/msm/dsi: use mult_frac for pclk_bpp calculation

Simplify calculations around pixel_clk_rate division. Replace common
pattern of doing 64-bit multiplication and then a do_div() call with
simpler mult_frac() invocation.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538273/
Link: https://lore.kernel.org/r/20230520200103.4019607-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 2b257b4599744..744f2398a6d6e 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -585,7 +585,7 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d
 	u8 lanes = msm_host->lanes;
 	u32 bpp = dsi_get_bpp(msm_host->format);
 	unsigned long pclk_rate = dsi_get_pclk_rate(mode, is_bonded_dsi);
-	u64 pclk_bpp = (u64)pclk_rate * bpp;
+	unsigned long pclk_bpp;
 
 	if (lanes == 0) {
 		pr_err("%s: forcing mdss_dsi lanes to 1\n", __func__);
@@ -594,9 +594,9 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d
 
 	/* CPHY "byte_clk" is in units of 16 bits */
 	if (msm_host->cphy_mode)
-		do_div(pclk_bpp, (16 * lanes));
+		pclk_bpp = mult_frac(pclk_rate, bpp, 16 * lanes);
 	else
-		do_div(pclk_bpp, (8 * lanes));
+		pclk_bpp = mult_frac(pclk_rate, bpp, 8 * lanes);
 
 	return pclk_bpp;
 }
@@ -627,15 +627,12 @@ int dsi_calc_clk_rate_6g(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 {
 	u32 bpp = dsi_get_bpp(msm_host->format);
-	u64 pclk_bpp;
 	unsigned int esc_mhz, esc_div;
 	unsigned long byte_mhz;
 
 	dsi_calc_pclk(msm_host, is_bonded_dsi);
 
-	pclk_bpp = (u64)msm_host->pixel_clk_rate * bpp;
-	do_div(pclk_bpp, 8);
-	msm_host->src_clk_rate = pclk_bpp;
+	msm_host->src_clk_rate = mult_frac(msm_host->pixel_clk_rate, bpp, 8);
 
 	/*
 	 * esc clock is byte clock followed by a 4 bit divider,
-- 
GitLab


From 95666ca7431c0217ce65d932526c0dbe32f85265 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 21 May 2023 22:22:28 +0300
Subject: [PATCH 0770/1791] drm/msm/dpu: drop
 dpu_encoder_phys_ops::late_register()

This callback has been unused since the driver being added. Drop it now.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/538302/
Link: https://lore.kernel.org/r/20230521192230.9747-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c      | 7 -------
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h | 3 ---
 2 files changed, 10 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 1dc5dbe585723..c771383446f2b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2115,7 +2115,6 @@ DEFINE_SHOW_ATTRIBUTE(_dpu_encoder_status);
 static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc)
 {
 	struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc);
-	int i;
 
 	char name[DPU_NAME_SIZE];
 
@@ -2134,12 +2133,6 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc)
 	debugfs_create_file("status", 0600,
 		dpu_enc->debugfs_root, dpu_enc, &_dpu_encoder_status_fops);
 
-	for (i = 0; i < dpu_enc->num_phys_encs; i++)
-		if (dpu_enc->phys_encs[i]->ops.late_register)
-			dpu_enc->phys_encs[i]->ops.late_register(
-					dpu_enc->phys_encs[i],
-					dpu_enc->debugfs_root);
-
 	return 0;
 }
 #else
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index 1d434b22180d9..9e29079a6fc46 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -63,7 +63,6 @@ struct dpu_encoder_phys;
 /**
  * struct dpu_encoder_phys_ops - Interface the physical encoders provide to
  *	the containing virtual encoder.
- * @late_register:		DRM Call. Add Userspace interfaces, debugfs.
  * @prepare_commit:		MSM Atomic Call, start of atomic commit sequence
  * @is_master:			Whether this phys_enc is the current master
  *				encoder. Can be switched at enable time. Based
@@ -93,8 +92,6 @@ struct dpu_encoder_phys;
  */
 
 struct dpu_encoder_phys_ops {
-	int (*late_register)(struct dpu_encoder_phys *encoder,
-			struct dentry *debugfs_root);
 	void (*prepare_commit)(struct dpu_encoder_phys *encoder);
 	bool (*is_master)(struct dpu_encoder_phys *encoder);
 	void (*atomic_mode_set)(struct dpu_encoder_phys *encoder,
-- 
GitLab


From a659098d78d607c150a45276b6e9726f43646619 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 21 May 2023 22:22:29 +0300
Subject: [PATCH 0771/1791] drm/msm/dpu: drop (mostly) unused DPU_NAME_SIZE
 define

This define is used only in one place, in dpu_encoder debugfs code.
Inline the value and drop the define completely.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/538303/
Link: https://lore.kernel.org/r/20230521192230.9747-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h     | 2 --
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c   | 2 --
 3 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index c771383446f2b..af34932729dbe 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2116,14 +2116,14 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc)
 {
 	struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc);
 
-	char name[DPU_NAME_SIZE];
+	char name[12];
 
 	if (!drm_enc->dev) {
 		DPU_ERROR("invalid encoder or kms\n");
 		return -EINVAL;
 	}
 
-	snprintf(name, DPU_NAME_SIZE, "encoder%u", drm_enc->base.id);
+	snprintf(name, sizeof(name), "encoder%u", drm_enc->base.id);
 
 	/* create overall sub-directory for the encoder */
 	dpu_enc->debugfs_root = debugfs_create_dir(name,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index aca39a4689f48..197f352fe89bd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -63,8 +63,6 @@
 #define ktime_compare_safe(A, B) \
 	ktime_compare(ktime_sub((A), (B)), ktime_set(0, 0))
 
-#define DPU_NAME_SIZE  12
-
 struct dpu_kms {
 	struct msm_kms base;
 	struct drm_device *dev;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 14b5cfe306113..ac75ba13aa011 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -42,8 +42,6 @@
 #define SHARP_SMOOTH_THR_DEFAULT	8
 #define SHARP_NOISE_THR_DEFAULT	2
 
-#define DPU_NAME_SIZE  12
-
 #define DPU_PLANE_COLOR_FILL_FLAG	BIT(31)
 #define DPU_ZPOS_MAX 255
 
-- 
GitLab


From 4fb5ad612c5cb1dfc72bd6bb9150d0112bfb0d27 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 22 May 2023 21:17:01 +0200
Subject: [PATCH 0772/1791] drm/msm: Use struct fb_info.screen_buffer

The fbdev framebuffer is in system memory. Store the address in
the field 'screen_buffer'. Fixes the following sparse warning.

../drivers/gpu/drm/msm/msm_fbdev.c:124:26: warning: incorrect type in assignment (different address spaces)
../drivers/gpu/drm/msm/msm_fbdev.c:124:26:    expected char [noderef] __iomem *screen_base
../drivers/gpu/drm/msm/msm_fbdev.c:124:26:    got void *

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/538531/
Link: https://lore.kernel.org/r/20230522191701.13406-1-tzimmermann@suse.de
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/msm_fbdev.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index 2ebc86381e1c9..ce0ba6d1979a7 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -121,9 +121,9 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 
 	drm_fb_helper_fill_info(fbi, helper, sizes);
 
-	fbi->screen_base = msm_gem_get_vaddr(bo);
-	if (IS_ERR(fbi->screen_base)) {
-		ret = PTR_ERR(fbi->screen_base);
+	fbi->screen_buffer = msm_gem_get_vaddr(bo);
+	if (IS_ERR(fbi->screen_buffer)) {
+		ret = PTR_ERR(fbi->screen_buffer);
 		goto fail;
 	}
 	fbi->screen_size = bo->size;
-- 
GitLab


From 12cef323c903bd8b13d1f6ff24a9695c2cdc360b Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:49 -0700
Subject: [PATCH 0773/1791] drm/msm/dpu: set DSC flush bit correctly at MDP CTL
 flush register

The CTL_FLUSH register should be programmed with the 22th bit
(DSC_IDX) to flush the DSC hardware blocks, not the literal value of
22 (which corresponds to flushing VIG1, VIG2 and RGB1 instead).

Changes in V12:
-- split this patch out of "separate DSC flush update out of interface"

Changes in V13:
-- rewording the commit text

Changes in V14:
-- drop 'DSC" from "The DSC CTL_FLUSH register" at commit text

Fixes: 77f6da90487c ("drm/msm/disp/dpu1: Add DSC support in hw_ctl")
Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539496/
Link: https://lore.kernel.org/r/1685036458-22683-2-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 07bcacedf4b02..231737e92c775 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -519,7 +519,7 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 		DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE,
 			      BIT(cfg->merge_3d - MERGE_3D_0));
 	if (cfg->dsc) {
-		DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, DSC_IDX);
+		DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, BIT(DSC_IDX));
 		DPU_REG_WRITE(c, CTL_DSC_ACTIVE, cfg->dsc);
 	}
 }
-- 
GitLab


From c20c44403e11d572c1adebf557c64017fd9868fd Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:51 -0700
Subject: [PATCH 0774/1791] drm/msm/dpu: add DPU_PINGPONG_DSC feature bit for
 DPU < 7.0.0

DPU < 7.0.0 requires the PINGPONG block to be involved during
DSC setting up. Since DPU >= 7.0.0, enabling and starting the DSC
encoder engine was moved to INTF with the help of the flush mechanism.
Add a DPU_PINGPONG_DSC feature bit to restrict the availability of
dpu_hw_pp_setup_dsc() and dpu_hw_pp_dsc_{enable,disable}() on the
PINGPONG block to DPU < 7.0.0 hardware, as the registers are not
available on DPU 7.0.0 and higher anymore.
Add DPU_PINGPONG_DSC to PINGPONG_SDM845_MASK, PINGPONG_SDM845_TE2_MASK
and PINGPONG_SM8150_MASK which is used for all DPU < 7.0 chipsets.

changes in v6:
-- split patches and rearrange to keep catalog related files at this patch

changes in v7:
-- rewording commit text as suggested at review comments

changes in v9:
-- delete BIT(DPU_PINGPONG_DSC) from PINGPONG_SDM845_TE2_MASK

changes in v10:
-- correct order of commit text

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539502/
Link: https://lore.kernel.org/r/1685036458-22683-4-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 4 +++-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 1dee5ba2b3121..f07eab738008c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -76,13 +76,13 @@
 	(BIT(DPU_DIM_LAYER) | BIT(DPU_MIXER_COMBINED_ALPHA))
 
 #define PINGPONG_SDM845_MASK \
-	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_TE))
+	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_TE) | BIT(DPU_PINGPONG_DSC))
 
 #define PINGPONG_SDM845_TE2_MASK \
 	(PINGPONG_SDM845_MASK | BIT(DPU_PINGPONG_TE2))
 
 #define PINGPONG_SM8150_MASK \
-	(BIT(DPU_PINGPONG_DITHER))
+	(BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC))
 
 #define CTL_SC7280_MASK \
 	(BIT(DPU_CTL_ACTIVE_CFG) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 6b1e8383d2c4f..d59a9027c4b6e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -140,7 +140,8 @@ enum {
  * @DPU_PINGPONG_TE2        Additional tear check block for split pipes
  * @DPU_PINGPONG_SPLIT      PP block supports split fifo
  * @DPU_PINGPONG_SLAVE      PP block is a suitable slave for split fifo
- * @DPU_PINGPONG_DITHER,    Dither blocks
+ * @DPU_PINGPONG_DITHER     Dither blocks
+ * @DPU_PINGPONG_DSC        PP block supports DSC
  * @DPU_PINGPONG_MAX
  */
 enum {
@@ -149,6 +150,7 @@ enum {
 	DPU_PINGPONG_SPLIT,
 	DPU_PINGPONG_SLAVE,
 	DPU_PINGPONG_DITHER,
+	DPU_PINGPONG_DSC,
 	DPU_PINGPONG_MAX
 };
 
-- 
GitLab


From 108ff1417795eb991ce5c3af77c119e0841136f2 Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:52 -0700
Subject: [PATCH 0775/1791] drm/msm/dpu: Guard PINGPONG DSC ops behind
 DPU_PINGPONG_DSC bit

DPU < 7.0.0 has DPU_PINGPONG_DSC feature bit set to indicate it requires
both dpu_hw_pp_setup_dsc() and dpu_hw_pp_dsc_{enable,disable}() to be
executed to complete DSC configuration if DSC hardware block is present.
Hence test DPU_PINGPONG_DSC feature bit and assign DSC related functions
to the ops of PINGPONG block accordingly if DPU_PINGPONG_DSC bit is set.

Changes in v6:
-- split patches, this patch has function handles DPU_PINGPONG_DSC bit

Changes in v9:
-- the original code of assigning dsc related functions to the ops of
   pingpong block without testing the DPU_PINGPONG_DSC feature bit was
   restored back due to rebase error which defeat the purpose of this
   patch. Remove those error code.

Changes in v10:
-- change commit title
-- correct texts at changes in v9

Changes in v12:
-- fixed length too long at Changes in v9

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539504/
Link: https://lore.kernel.org/r/1685036458-22683-5-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
index 79e45768eae5a..437d9e62a841b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c
@@ -291,9 +291,12 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong *c,
 		c->ops.get_line_count = dpu_hw_pp_get_line_count;
 		c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh;
 	}
-	c->ops.setup_dsc = dpu_hw_pp_setup_dsc;
-	c->ops.enable_dsc = dpu_hw_pp_dsc_enable;
-	c->ops.disable_dsc = dpu_hw_pp_dsc_disable;
+
+	if (test_bit(DPU_PINGPONG_DSC, &features)) {
+		c->ops.setup_dsc = dpu_hw_pp_setup_dsc;
+		c->ops.enable_dsc = dpu_hw_pp_dsc_enable;
+		c->ops.disable_dsc = dpu_hw_pp_dsc_disable;
+	}
 
 	if (test_bit(DPU_PINGPONG_DITHER, &features))
 		c->ops.setup_dither = dpu_hw_pp_setup_dither;
-- 
GitLab


From d45be1ccd3303201e8f7c28aefe28e6e4b168ab8 Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:53 -0700
Subject: [PATCH 0776/1791] drm/msm/dpu: Introduce PINGPONG_NONE to disconnect
 DSC from PINGPONG

Disabling the crossbar mux between DSC and PINGPONG currently
requires a bogus enum dpu_pingpong value to be passed when calling
dsc_bind_pingpong_blk() with enable=false, even though the register
value written is independent of the current PINGPONG block.  Replace
that `bool enable` parameter with a new PINGPONG_NONE dpu_pingpong
flag that triggers the write of the "special" 0xF "crossbar
disabled" value to the register instead.

Changes in v4:
-- more details to commit text

Changes in v5:
-- rewording commit text suggested by Marijn
-- add DRM_DEBUG_KMS for DSC unbinding case

Changes in v8:
-- fix checkpatch warning

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539505/
Link: https://lore.kernel.org/r/1685036458-22683-6-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c |  2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c  | 14 +++++++-------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h  |  1 -
 3 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index d7cd4734dc7df..9e0b6302ce236 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1850,7 +1850,7 @@ static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_dsc *hw_dsc,
 		hw_pp->ops.setup_dsc(hw_pp);
 
 	if (hw_dsc->ops.dsc_bind_pingpong_blk)
-		hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, true, hw_pp->idx);
+		hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, hw_pp->idx);
 
 	if (hw_pp->ops.enable_dsc)
 		hw_pp->ops.enable_dsc(hw_pp);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
index 89e1a9f945c71..64847912a20ff 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
@@ -154,7 +154,6 @@ static void dpu_hw_dsc_config_thresh(struct dpu_hw_dsc *hw_dsc,
 
 static void dpu_hw_dsc_bind_pingpong_blk(
 		struct dpu_hw_dsc *hw_dsc,
-		bool enable,
 		const enum dpu_pingpong pp)
 {
 	struct dpu_hw_blk_reg_map *c = &hw_dsc->hw;
@@ -163,14 +162,15 @@ static void dpu_hw_dsc_bind_pingpong_blk(
 
 	dsc_ctl_offset = DSC_CTL(hw_dsc->idx);
 
-	if (enable)
+	if (pp)
 		mux_cfg = (pp - PINGPONG_0) & 0x7;
 
-	DRM_DEBUG_KMS("%s dsc:%d %s pp:%d\n",
-			enable ? "Binding" : "Unbinding",
-			hw_dsc->idx - DSC_0,
-			enable ? "to" : "from",
-			pp - PINGPONG_0);
+	if (pp)
+		DRM_DEBUG_KMS("Binding dsc:%d to pp:%d\n",
+			      hw_dsc->idx - DSC_0, pp - PINGPONG_0);
+	else
+		DRM_DEBUG_KMS("Unbinding dsc:%d from any pp\n",
+			      hw_dsc->idx - DSC_0);
 
 	DPU_REG_WRITE(c, dsc_ctl_offset, mux_cfg);
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
index 287ec5f1e170a..138080aed9dd7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
@@ -44,7 +44,6 @@ struct dpu_hw_dsc_ops {
 				  struct drm_dsc_config *dsc);
 
 	void (*dsc_bind_pingpong_blk)(struct dpu_hw_dsc *hw_dsc,
-				  bool enable,
 				  enum dpu_pingpong pp);
 };
 
-- 
GitLab


From 625cbb077007698060b12d0ae5657a4d8411b153 Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:55 -0700
Subject: [PATCH 0777/1791] drm/msm/dpu: always clear every individual pending
 flush mask

There are two tiers of pending flush control, top level and
individual hardware block. Currently only the top level of
flush mask is reset to 0 but the individual pending flush masks
of particular hardware blocks are left at their previous values,
eventually accumulating all possible bit values and typically
flushing more than necessary.
Reset all individual hardware block flush masks to 0 to avoid
accidentally flushing them.

Changes in V13:
-- rewording commit text
-- add an empty space line as suggested

Changes in V14:
-- add Fixes tag

Fixes: 73bfb790ac78 ("msm:disp:dpu1: setup display datapath for SC7180 target")
Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539508/
Link: https://lore.kernel.org/r/1685036458-22683-8-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 231737e92c775..911848bc0c85d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -100,6 +100,9 @@ static inline void dpu_hw_ctl_clear_pending_flush(struct dpu_hw_ctl *ctx)
 	trace_dpu_hw_ctl_clear_pending_flush(ctx->pending_flush_mask,
 				     dpu_hw_ctl_get_flush_register(ctx));
 	ctx->pending_flush_mask = 0x0;
+	ctx->pending_intf_flush_mask = 0;
+	ctx->pending_wb_flush_mask = 0;
+	ctx->pending_merge_3d_flush_mask = 0;
 
 	memset(ctx->pending_dspp_flush_mask, 0,
 		sizeof(ctx->pending_dspp_flush_mask));
-- 
GitLab


From 761c629d186009517477a0c415ecfbff3063ecbb Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:56 -0700
Subject: [PATCH 0778/1791] drm/msm/dpu: separate DSC flush update out of
 interface

Currently DSC flushing happens during interface configuration at
dpu_hw_ctl_intf_cfg_v1(). Separate DSC flush away from
dpu_hw_ctl_intf_cfg_v1() by adding dpu_hw_ctl_update_pending_flush_dsc_v1()
to handle both per-DSC engine and DSC flush bits at same time to make it
consistent with the location of flush programming of other DPU sub-blocks.

Changes in v10:
-- rewording commit text
-- pass ctl directly instead of dpu_enc to dsc_pipe_cfg()
-- ctx->pending_dsc_flush_mask = 0;

Changes in v11:
-- add Fixes tag

Changes in v12:
-- move dsc parameter to next line at dpu_encoder_dsc_pipe_cfg()

Changes in v14:
-- Fixes tag had been move to 1st patch of this series

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539506/
Link: https://lore.kernel.org/r/1685036458-22683-9-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 10 +++++++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c  | 23 +++++++++++++++------
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h  | 11 ++++++++++
 3 files changed, 36 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 9e0b6302ce236..95b2bd9cc7ab0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1834,7 +1834,8 @@ dpu_encoder_dsc_initial_line_calc(struct drm_dsc_config *dsc,
 	return DIV_ROUND_UP(total_pixels, dsc->slice_width);
 }
 
-static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_dsc *hw_dsc,
+static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_ctl *ctl,
+				     struct dpu_hw_dsc *hw_dsc,
 				     struct dpu_hw_pingpong *hw_pp,
 				     struct drm_dsc_config *dsc,
 				     u32 common_mode,
@@ -1854,6 +1855,9 @@ static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_dsc *hw_dsc,
 
 	if (hw_pp->ops.enable_dsc)
 		hw_pp->ops.enable_dsc(hw_pp);
+
+	if (ctl->ops.update_pending_flush_dsc)
+		ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx);
 }
 
 static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc,
@@ -1861,6 +1865,7 @@ static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc,
 {
 	/* coding only for 2LM, 2enc, 1 dsc config */
 	struct dpu_encoder_phys *enc_master = dpu_enc->cur_master;
+	struct dpu_hw_ctl *ctl = enc_master->hw_ctl;
 	struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC];
 	struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC];
 	int this_frame_slices;
@@ -1898,7 +1903,8 @@ static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc,
 	initial_lines = dpu_encoder_dsc_initial_line_calc(dsc, enc_ip_w);
 
 	for (i = 0; i < MAX_CHANNELS_PER_ENC; i++)
-		dpu_encoder_dsc_pipe_cfg(hw_dsc[i], hw_pp[i], dsc, dsc_common_mode, initial_lines);
+		dpu_encoder_dsc_pipe_cfg(ctl, hw_dsc[i], hw_pp[i],
+					 dsc, dsc_common_mode, initial_lines);
 }
 
 void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 911848bc0c85d..0007d14cecda5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -103,6 +103,7 @@ static inline void dpu_hw_ctl_clear_pending_flush(struct dpu_hw_ctl *ctx)
 	ctx->pending_intf_flush_mask = 0;
 	ctx->pending_wb_flush_mask = 0;
 	ctx->pending_merge_3d_flush_mask = 0;
+	ctx->pending_dsc_flush_mask = 0;
 
 	memset(ctx->pending_dspp_flush_mask, 0,
 		sizeof(ctx->pending_dspp_flush_mask));
@@ -142,6 +143,11 @@ static inline void dpu_hw_ctl_trigger_flush_v1(struct dpu_hw_ctl *ctx)
 				CTL_DSPP_n_FLUSH(dspp - DSPP_0),
 				ctx->pending_dspp_flush_mask[dspp - DSPP_0]);
 		}
+
+	if (ctx->pending_flush_mask & BIT(DSC_IDX))
+		DPU_REG_WRITE(&ctx->hw, CTL_DSC_FLUSH,
+			      ctx->pending_dsc_flush_mask);
+
 	DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, ctx->pending_flush_mask);
 }
 
@@ -288,6 +294,13 @@ static void dpu_hw_ctl_update_pending_flush_merge_3d_v1(struct dpu_hw_ctl *ctx,
 	ctx->pending_flush_mask |= BIT(MERGE_3D_IDX);
 }
 
+static void dpu_hw_ctl_update_pending_flush_dsc_v1(struct dpu_hw_ctl *ctx,
+						   enum dpu_dsc dsc_num)
+{
+	ctx->pending_dsc_flush_mask |= BIT(dsc_num - DSC_0);
+	ctx->pending_flush_mask |= BIT(DSC_IDX);
+}
+
 static void dpu_hw_ctl_update_pending_flush_dspp(struct dpu_hw_ctl *ctx,
 	enum dpu_dspp dspp, u32 dspp_sub_blk)
 {
@@ -499,9 +512,6 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	if ((test_bit(DPU_CTL_VM_CFG, &ctx->caps->features)))
 		mode_sel = CTL_DEFAULT_GROUP_ID  << 28;
 
-	if (cfg->dsc)
-		DPU_REG_WRITE(&ctx->hw, CTL_DSC_FLUSH, cfg->dsc);
-
 	if (cfg->intf_mode_sel == DPU_CTL_MODE_SEL_CMD)
 		mode_sel |= BIT(17);
 
@@ -521,10 +531,9 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	if (cfg->merge_3d)
 		DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE,
 			      BIT(cfg->merge_3d - MERGE_3D_0));
-	if (cfg->dsc) {
-		DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, BIT(DSC_IDX));
+
+	if (cfg->dsc)
 		DPU_REG_WRITE(c, CTL_DSC_ACTIVE, cfg->dsc);
-	}
 }
 
 static void dpu_hw_ctl_intf_cfg(struct dpu_hw_ctl *ctx,
@@ -627,6 +636,8 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops,
 		ops->update_pending_flush_merge_3d =
 			dpu_hw_ctl_update_pending_flush_merge_3d_v1;
 		ops->update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb_v1;
+		ops->update_pending_flush_dsc =
+			dpu_hw_ctl_update_pending_flush_dsc_v1;
 	} else {
 		ops->trigger_flush = dpu_hw_ctl_trigger_flush;
 		ops->setup_intf_cfg = dpu_hw_ctl_intf_cfg;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
index 6292002faba2d..1c242298ff2ee 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
@@ -157,6 +157,15 @@ struct dpu_hw_ctl_ops {
 	void (*update_pending_flush_dspp)(struct dpu_hw_ctl *ctx,
 		enum dpu_dspp blk, u32 dspp_sub_blk);
 
+	/**
+	 * OR in the given flushbits to the cached pending_(dsc_)flush_mask
+	 * No effect on hardware
+	 * @ctx: ctl path ctx pointer
+	 * @blk: interface block index
+	 */
+	void (*update_pending_flush_dsc)(struct dpu_hw_ctl *ctx,
+					 enum dpu_dsc blk);
+
 	/**
 	 * Write the value of the pending_flush_mask to hardware
 	 * @ctx       : ctl path ctx pointer
@@ -229,6 +238,7 @@ struct dpu_hw_ctl_ops {
  * @pending_flush_mask: storage for pending ctl_flush managed via ops
  * @pending_intf_flush_mask: pending INTF flush
  * @pending_wb_flush_mask: pending WB flush
+ * @pending_dsc_flush_mask: pending DSC flush
  * @ops: operation list
  */
 struct dpu_hw_ctl {
@@ -245,6 +255,7 @@ struct dpu_hw_ctl {
 	u32 pending_wb_flush_mask;
 	u32 pending_merge_3d_flush_mask;
 	u32 pending_dspp_flush_mask[DSPP_MAX - DSPP_0];
+	u32 pending_dsc_flush_mask;
 
 	/* ops */
 	struct dpu_hw_ctl_ops ops;
-- 
GitLab


From 997ed53dd69379d266fee206774e1c6b1244c178 Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:58 -0700
Subject: [PATCH 0779/1791] drm/msm/dpu: Tear down DSC datapath on encoder
 cleanup

Unset DSC_ACTIVE bit at dpu_hw_ctl_reset_intf_cfg_v1(),
dpu_encoder_unprep_dsc() and dpu_encoder_dsc_pipe_clr() functions
to tear down DSC data path if DSC data path was setup previous.

Changes in V10:
-- pass ctl directly instead of dpu_enc to dsc_pipe_cfg()
-- move both dpu_encoder_unprep_dsc() and dpu_encoder_dsc_pipe_clr() to above phys_cleanup()

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539515/
Link: https://lore.kernel.org/r/1685036458-22683-11-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 39 +++++++++++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c  |  7 ++++
 2 files changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 95b2bd9cc7ab0..2c9af33c09117 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2036,6 +2036,41 @@ static void dpu_encoder_helper_reset_mixers(struct dpu_encoder_phys *phys_enc)
 	}
 }
 
+static void dpu_encoder_dsc_pipe_clr(struct dpu_hw_ctl *ctl,
+				     struct dpu_hw_dsc *hw_dsc,
+				     struct dpu_hw_pingpong *hw_pp)
+{
+	if (hw_dsc->ops.dsc_disable)
+		hw_dsc->ops.dsc_disable(hw_dsc);
+
+	if (hw_pp->ops.disable_dsc)
+		hw_pp->ops.disable_dsc(hw_pp);
+
+	if (hw_dsc->ops.dsc_bind_pingpong_blk)
+		hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, PINGPONG_NONE);
+
+	if (ctl->ops.update_pending_flush_dsc)
+		ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx);
+}
+
+static void dpu_encoder_unprep_dsc(struct dpu_encoder_virt *dpu_enc)
+{
+	/* coding only for 2LM, 2enc, 1 dsc config */
+	struct dpu_encoder_phys *enc_master = dpu_enc->cur_master;
+	struct dpu_hw_ctl *ctl = enc_master->hw_ctl;
+	struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC];
+	struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC];
+	int i;
+
+	for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) {
+		hw_pp[i] = dpu_enc->hw_pp[i];
+		hw_dsc[i] = dpu_enc->hw_dsc[i];
+
+		if (hw_pp[i] && hw_dsc[i])
+			dpu_encoder_dsc_pipe_clr(ctl, hw_dsc[i], hw_pp[i]);
+	}
+}
+
 void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc)
 {
 	struct dpu_hw_ctl *ctl = phys_enc->hw_ctl;
@@ -2086,8 +2121,12 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc)
 					phys_enc->hw_pp->merge_3d->idx);
 	}
 
+	if (dpu_enc->dsc)
+		dpu_encoder_unprep_dsc(dpu_enc);
+
 	intf_cfg.stream_sel = 0; /* Don't care value for video mode */
 	intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc);
+	intf_cfg.dsc = dpu_encoder_helper_get_dsc(phys_enc);
 
 	if (phys_enc->hw_intf)
 		intf_cfg.intf = phys_enc->hw_intf->idx;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index 0007d14cecda5..c278fb9d2b5b5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -576,6 +576,7 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 	u32 intf_active = 0;
 	u32 wb_active = 0;
 	u32 merge3d_active = 0;
+	u32 dsc_active;
 
 	/*
 	 * This API resets each portion of the CTL path namely,
@@ -605,6 +606,12 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx,
 		wb_active &= ~BIT(cfg->wb - WB_0);
 		DPU_REG_WRITE(c, CTL_WB_ACTIVE, wb_active);
 	}
+
+	if (cfg->dsc) {
+		dsc_active = DPU_REG_READ(c, CTL_DSC_ACTIVE);
+		dsc_active &= ~cfg->dsc;
+		DPU_REG_WRITE(c, CTL_DSC_ACTIVE, dsc_active);
+	}
 }
 
 static void dpu_hw_ctl_set_fetch_pipe_active(struct dpu_hw_ctl *ctx,
-- 
GitLab


From 025e3d97867c55e4519edde2fda9363b9a4da795 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Fri, 26 May 2023 14:51:59 +0300
Subject: [PATCH 0780/1791] drm/msm/dpu: clean up dpu_kms_get_clk_rate()
 returns

Static analysis tools complain about the -EINVAL error code being
stored in an unsigned variable.  Let's change this to match
the clk_get_rate() function which is type unsigned long and returns
zero on error.

Fixes: 25fdd5933e4c ("drm/msm: Add SDM845 DPU support")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539626/
Link: https://lore.kernel.org/r/28644c5e-950e-41cd-8389-67f37b067bdc@kili.mountain
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 8ce057cc93743..7f87ac6d68acf 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1023,13 +1023,13 @@ static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms)
 	return 0;
 }
 
-u64 dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name)
+unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name)
 {
 	struct clk *clk;
 
 	clk = msm_clk_bulk_get_clock(dpu_kms->clocks, dpu_kms->num_clocks, clock_name);
 	if (!clk)
-		return -EINVAL;
+		return 0;
 
 	return clk_get_rate(clk);
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
index cd62875245575..7810342865374 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h
@@ -205,6 +205,6 @@ void dpu_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc);
  *
  * Return: current clock rate
  */
-u64 dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name);
+unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name);
 
 #endif /* __dpu_kms_H__ */
-- 
GitLab


From eea9cf72b718b1d2c935f2be2610d05e7527f2d8 Mon Sep 17 00:00:00 2001
From: Su Hui <suhui@nfschina.com>
Date: Mon, 22 May 2023 09:32:13 +0800
Subject: [PATCH 0781/1791] drm/msm: Remove unnecessary (void*) conversions

Pointer variables of (void*) type do not require type cast.

Signed-off-by: Su Hui <suhui@nfschina.com>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/540599/
Link: https://lore.kernel.org/r/20230522013213.25876-1-suhui@nfschina.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/adreno/a5xx_debugfs.c | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c  | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c   | 2 +-
 drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c  | 2 +-
 drivers/gpu/drm/msm/msm_debugfs.c         | 6 +++---
 5 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
index 6bd397a85834f..169b8fe688f89 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c
@@ -69,7 +69,7 @@ static void roq_print(struct msm_gpu *gpu, struct drm_printer *p)
 
 static int show(struct seq_file *m, void *arg)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_printer p = drm_seq_file_printer(m);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index cc66ddffe6723..6e684a7b49a17 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -1392,7 +1392,7 @@ DEFINE_SHOW_ATTRIBUTE(_dpu_debugfs_status);
 
 static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v)
 {
-	struct drm_crtc *crtc = (struct drm_crtc *) s->private;
+	struct drm_crtc *crtc = s->private;
 	struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc);
 
 	seq_printf(s, "client type: %d\n", dpu_crtc_get_client_type(crtc));
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 0e7a68714e9e1..3b307ce637a6b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -57,8 +57,8 @@ static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms);
 static int _dpu_danger_signal_status(struct seq_file *s,
 		bool danger_status)
 {
-	struct dpu_kms *kms = (struct dpu_kms *)s->private;
 	struct dpu_danger_safe_status status;
+	struct dpu_kms *kms = s->private;
 	int i;
 
 	if (!kms->hw_mdp) {
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
index 29ae5c9613f36..323079cfd698b 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
@@ -229,7 +229,7 @@ static void mdp5_kms_destroy(struct msm_kms *kms)
 #ifdef CONFIG_DEBUG_FS
 static int smp_show(struct seq_file *m, void *arg)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms));
diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c
index 9c0e633a3a611..a0a936f80ae35 100644
--- a/drivers/gpu/drm/msm/msm_debugfs.c
+++ b/drivers/gpu/drm/msm/msm_debugfs.c
@@ -211,7 +211,7 @@ DEFINE_DEBUGFS_ATTRIBUTE(shrink_fops,
 
 static int msm_gem_show(struct seq_file *m, void *arg)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct msm_drm_private *priv = dev->dev_private;
 	int ret;
@@ -229,7 +229,7 @@ static int msm_gem_show(struct seq_file *m, void *arg)
 
 static int msm_mm_show(struct seq_file *m, void *arg)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_printer p = drm_seq_file_printer(m);
 
@@ -240,7 +240,7 @@ static int msm_mm_show(struct seq_file *m, void *arg)
 
 static int msm_fb_show(struct seq_file *m, void *arg)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_framebuffer *fb, *fbdev_fb = NULL;
 
-- 
GitLab


From d01eb3421b101c6f066fc474f567ba902cca0077 Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:08 +0200
Subject: [PATCH 0782/1791] dt-bindings: msm: dsi-phy-28nm: Document msm8226
 compatible

The MSM8226 SoC uses a slightly different 28nm dsi phy. Add a new
compatible for it.

And while we're at it, in the dsi-phy-28nm.yaml move the 8960 compatible
to its correct place so its sorted alphabetically.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Patchwork: https://patchwork.freedesktop.org/patch/540621/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-1-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../devicetree/bindings/display/msm/dsi-phy-28nm.yaml          | 3 ++-
 Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml   | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml b/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml
index cf4a338c46610..62fb3e484eb29 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml
@@ -15,10 +15,11 @@ allOf:
 properties:
   compatible:
     enum:
+      - qcom,dsi-phy-28nm-8226
+      - qcom,dsi-phy-28nm-8960
       - qcom,dsi-phy-28nm-hpm
       - qcom,dsi-phy-28nm-hpm-fam-b
       - qcom,dsi-phy-28nm-lp
-      - qcom,dsi-phy-28nm-8960
 
   reg:
     items:
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
index 20889e4094300..397df2c2af8ea 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml
@@ -122,6 +122,7 @@ patternProperties:
           - qcom,dsi-phy-14nm-660
           - qcom,dsi-phy-14nm-8953
           - qcom,dsi-phy-20nm
+          - qcom,dsi-phy-28nm-8226
           - qcom,dsi-phy-28nm-hpm
           - qcom,dsi-phy-28nm-lp
           - qcom,hdmi-phy-8084
-- 
GitLab


From 46ccf3e0ed63e7f6f7a6a6258888eec0787a9f88 Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:09 +0200
Subject: [PATCH 0783/1791] dt-bindings: display/msm: dsi-controller-main: Add
 msm8226 compatible

Add the compatible for the DSI found on MSM8226.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Patchwork: https://patchwork.freedesktop.org/patch/540613/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-2-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../devicetree/bindings/display/msm/dsi-controller-main.yaml    | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
index 51808b1d27268..4ddfdfed521ea 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@@ -15,6 +15,7 @@ properties:
       - items:
           - enum:
               - qcom,apq8064-dsi-ctrl
+              - qcom,msm8226-dsi-ctrl
               - qcom,msm8916-dsi-ctrl
               - qcom,msm8953-dsi-ctrl
               - qcom,msm8974-dsi-ctrl
@@ -256,6 +257,7 @@ allOf:
         compatible:
           contains:
             enum:
+              - qcom,msm8226-dsi-ctrl
               - qcom,msm8974-dsi-ctrl
     then:
       properties:
-- 
GitLab


From 82cf4954a0d5f50fa546753cbf0388e69755c54d Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:12 +0200
Subject: [PATCH 0784/1791] drm/msm/dsi: Add configuration for MSM8226

Add the config for the v1.0.2 DSI found on MSM8226. We can reuse
existing bits from other revisions that are identical for v1.0.2.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Patchwork: https://patchwork.freedesktop.org/patch/540616/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-5-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_cfg.c | 2 ++
 drivers/gpu/drm/msm/dsi/dsi_cfg.h | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
index 29ccd755cc2e2..8a5fb6df72103 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
@@ -245,6 +245,8 @@ static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = {
 		&apq8064_dsi_cfg, &msm_dsi_v2_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0,
 		&msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops},
+	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0_2,
+		&msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1,
 		&msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1_1,
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
index 91bdaf50bb1a4..43f0dd74edb65 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
@@ -11,6 +11,7 @@
 #define MSM_DSI_VER_MAJOR_V2	0x02
 #define MSM_DSI_VER_MAJOR_6G	0x03
 #define MSM_DSI_6G_VER_MINOR_V1_0	0x10000000
+#define MSM_DSI_6G_VER_MINOR_V1_0_2	0x10000002
 #define MSM_DSI_6G_VER_MINOR_V1_1	0x10010000
 #define MSM_DSI_6G_VER_MINOR_V1_1_1	0x10010001
 #define MSM_DSI_6G_VER_MINOR_V1_2	0x10020000
-- 
GitLab


From 1531d0b9235e5c8e959fbfe02c90ecd5b2b8de05 Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:13 +0200
Subject: [PATCH 0785/1791] drm/msm/dsi: Add phy configuration for MSM8226

MSM8226 uses a modified PLL lock sequence compared to MSM8974, which is
based on the function dsi_pll_enable_seq_m in the msm-3.10 kernel.

Worth noting that the msm-3.10 downstream kernel also will try other
sequences in case this one doesn't work, but during testing it has shown
that the _m sequence succeeds first time also:

  .pll_enable_seqs[0] = dsi_pll_enable_seq_m,
  .pll_enable_seqs[1] = dsi_pll_enable_seq_m,
  .pll_enable_seqs[2] = dsi_pll_enable_seq_d,
  .pll_enable_seqs[3] = dsi_pll_enable_seq_d,
  .pll_enable_seqs[4] = dsi_pll_enable_seq_f1,
  .pll_enable_seqs[5] = dsi_pll_enable_seq_c,
  .pll_enable_seqs[6] = dsi_pll_enable_seq_e,

We may need to expand this in the future.

Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/540618/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-6-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy.c      |  2 +
 drivers/gpu/drm/msm/dsi/phy/dsi_phy.h      |  3 +-
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c | 97 ++++++++++++++++++++++
 3 files changed, 101 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index bb09cbe8ff86b..9d5795c58a984 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -541,6 +541,8 @@ static const struct of_device_id dsi_phy_dt_match[] = {
 	  .data = &dsi_phy_28nm_hpm_famb_cfgs },
 	{ .compatible = "qcom,dsi-phy-28nm-lp",
 	  .data = &dsi_phy_28nm_lp_cfgs },
+	{ .compatible = "qcom,dsi-phy-28nm-8226",
+	  .data = &dsi_phy_28nm_8226_cfgs },
 #endif
 #ifdef CONFIG_DRM_MSM_DSI_20NM_PHY
 	{ .compatible = "qcom,dsi-phy-20nm",
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
index 7137a17ae5238..8b640d1747851 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
@@ -46,8 +46,9 @@ struct msm_dsi_phy_cfg {
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_famb_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs;
-extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8226_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_14nm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_14nm_660_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_14nm_2290_cfgs;
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
index 4c1bf55c5f382..ceec7bb87bf13 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
@@ -37,6 +37,7 @@
 
 /* v2.0.0 28nm LP implementation */
 #define DSI_PHY_28NM_QUIRK_PHY_LP	BIT(0)
+#define DSI_PHY_28NM_QUIRK_PHY_8226	BIT(1)
 
 #define LPFR_LUT_SIZE			10
 struct lpfr_cfg {
@@ -377,6 +378,74 @@ static int dsi_pll_28nm_vco_prepare_hpm(struct clk_hw *hw)
 	return ret;
 }
 
+static int dsi_pll_28nm_vco_prepare_8226(struct clk_hw *hw)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw);
+	struct device *dev = &pll_28nm->phy->pdev->dev;
+	void __iomem *base = pll_28nm->phy->pll_base;
+	u32 max_reads = 5, timeout_us = 100;
+	bool locked;
+	u32 val;
+	int i;
+
+	DBG("id=%d", pll_28nm->phy->id);
+
+	pll_28nm_software_reset(pll_28nm);
+
+	/*
+	 * PLL power up sequence.
+	 * Add necessary delays recommended by hardware.
+	 */
+	dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_CAL_CFG1, 0x34);
+
+	val = DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRDN_B;
+	dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 200);
+
+	val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRGEN_PWRDN_B;
+	dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 200);
+
+	val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_LDO_PWRDN_B;
+	val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_ENABLE;
+	dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 600);
+
+	for (i = 0; i < 7; i++) {
+		/* DSI Uniphy lock detect setting */
+		dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2, 0x0d);
+		dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2,
+				0x0c, 100);
+		dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2, 0x0d);
+
+		/* poll for PLL ready status */
+		locked = pll_28nm_poll_for_ready(pll_28nm,
+						max_reads, timeout_us);
+		if (locked)
+			break;
+
+		pll_28nm_software_reset(pll_28nm);
+
+		/*
+		 * PLL power up sequence.
+		 * Add necessary delays recommended by hardware.
+		 */
+		dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_PWRGEN_CFG, 0x00, 50);
+
+		val = DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRDN_B;
+		val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRGEN_PWRDN_B;
+		dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 100);
+
+		val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_LDO_PWRDN_B;
+		val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_ENABLE;
+		dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 600);
+	}
+
+	if (unlikely(!locked))
+		DRM_DEV_ERROR(dev, "DSI PLL lock failed\n");
+	else
+		DBG("DSI PLL Lock success");
+
+	return locked ? 0 : -EINVAL;
+}
+
 static int dsi_pll_28nm_vco_prepare_lp(struct clk_hw *hw)
 {
 	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw);
@@ -471,6 +540,15 @@ static const struct clk_ops clk_ops_dsi_pll_28nm_vco_lp = {
 	.is_enabled = dsi_pll_28nm_clk_is_enabled,
 };
 
+static const struct clk_ops clk_ops_dsi_pll_28nm_vco_8226 = {
+	.round_rate = dsi_pll_28nm_clk_round_rate,
+	.set_rate = dsi_pll_28nm_clk_set_rate,
+	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
+	.prepare = dsi_pll_28nm_vco_prepare_8226,
+	.unprepare = dsi_pll_28nm_vco_unprepare,
+	.is_enabled = dsi_pll_28nm_clk_is_enabled,
+};
+
 /*
  * PLL Callbacks
  */
@@ -536,6 +614,8 @@ static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm, struct clk_hw **prov
 
 	if (pll_28nm->phy->cfg->quirks & DSI_PHY_28NM_QUIRK_PHY_LP)
 		vco_init.ops = &clk_ops_dsi_pll_28nm_vco_lp;
+	else if (pll_28nm->phy->cfg->quirks & DSI_PHY_28NM_QUIRK_PHY_8226)
+		vco_init.ops = &clk_ops_dsi_pll_28nm_vco_8226;
 	else
 		vco_init.ops = &clk_ops_dsi_pll_28nm_vco_hpm;
 
@@ -820,3 +900,20 @@ const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs = {
 	.quirks = DSI_PHY_28NM_QUIRK_PHY_LP,
 };
 
+const struct msm_dsi_phy_cfg dsi_phy_28nm_8226_cfgs = {
+	.has_phy_regulator = true,
+	.regulator_data = dsi_phy_28nm_regulators,
+	.num_regulators = ARRAY_SIZE(dsi_phy_28nm_regulators),
+	.ops = {
+		.enable = dsi_28nm_phy_enable,
+		.disable = dsi_28nm_phy_disable,
+		.pll_init = dsi_pll_28nm_init,
+		.save_pll_state = dsi_28nm_pll_save_state,
+		.restore_pll_state = dsi_28nm_pll_restore_state,
+	},
+	.min_pll_rate = VCO_MIN_RATE,
+	.max_pll_rate = VCO_MAX_RATE,
+	.io_start = { 0xfd922b00 },
+	.num_dsi_phy = 1,
+	.quirks = DSI_PHY_28NM_QUIRK_PHY_8226,
+};
-- 
GitLab


From c6e79fd5c80d8d6f9039b498a1203214bdf7695e Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:10 +0200
Subject: [PATCH 0786/1791] dt-bindings: display/msm: qcom, mdp5: Add msm8226
 compatible

Add the compatible for the MDP5 found on MSM8226.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Patchwork: https://patchwork.freedesktop.org/patch/540614/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-3-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml b/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml
index ef461ad6ce4a6..324e2942c0097 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml
@@ -22,6 +22,7 @@ properties:
       - items:
           - enum:
               - qcom,apq8084-mdp5
+              - qcom,msm8226-mdp5
               - qcom,msm8916-mdp5
               - qcom,msm8917-mdp5
               - qcom,msm8953-mdp5
-- 
GitLab


From eed3f9c7c36a2dbf01b6a3529a17b015df68b456 Mon Sep 17 00:00:00 2001
From: Luca Weiss <luca@z3ntu.xyz>
Date: Thu, 1 Jun 2023 19:00:11 +0200
Subject: [PATCH 0787/1791] drm/msm/mdp5: Add MDP5 configuration for MSM8226

Add the required config for the v1.1 MDP5 found on MSM8226.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
Patchwork: https://patchwork.freedesktop.org/patch/540612/
Link: https://lore.kernel.org/r/20230308-msm8226-mdp-v3-4-b6284145d67a@z3ntu.xyz
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c | 82 ++++++++++++++++++++++++
 1 file changed, 82 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
index 2eec2d78f32a0..694d543413374 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
@@ -103,6 +103,87 @@ static const struct mdp5_cfg_hw msm8x74v1_config = {
 	.max_clk = 200000000,
 };
 
+static const struct mdp5_cfg_hw msm8x26_config = {
+	.name = "msm8x26",
+	.mdp = {
+		.count = 1,
+		.caps = MDP_CAP_SMP |
+			0,
+	},
+	.smp = {
+		.mmb_count = 7,
+		.mmb_size = 4096,
+		.clients = {
+			[SSPP_VIG0] =  1,
+			[SSPP_DMA0] = 4,
+			[SSPP_RGB0] = 7,
+		},
+	},
+	.ctl = {
+		.count = 2,
+		.base = { 0x00500, 0x00600 },
+		.flush_hw_mask = 0x0003ffff,
+	},
+	.pipe_vig = {
+		.count = 1,
+		.base = { 0x01100 },
+		.caps = MDP_PIPE_CAP_HFLIP |
+			MDP_PIPE_CAP_VFLIP |
+			MDP_PIPE_CAP_SCALE |
+			MDP_PIPE_CAP_CSC   |
+			0,
+	},
+	.pipe_rgb = {
+		.count = 1,
+		.base = { 0x01d00 },
+		.caps = MDP_PIPE_CAP_HFLIP |
+			MDP_PIPE_CAP_VFLIP |
+			MDP_PIPE_CAP_SCALE |
+			0,
+	},
+	.pipe_dma = {
+		.count = 1,
+		.base = { 0x02900 },
+		.caps = MDP_PIPE_CAP_HFLIP |
+			MDP_PIPE_CAP_VFLIP |
+			0,
+	},
+	.lm = {
+		.count = 2,
+		.base = { 0x03100, 0x03d00 },
+		.instances = {
+				{ .id = 0, .pp = 0, .dspp = 0,
+				  .caps = MDP_LM_CAP_DISPLAY, },
+				{ .id = 1, .pp = -1, .dspp = -1,
+				  .caps = MDP_LM_CAP_WB },
+			     },
+		.nb_stages = 2,
+		.max_width = 2048,
+		.max_height = 0xFFFF,
+	},
+	.dspp = {
+		.count = 1,
+		.base = { 0x04500 },
+	},
+	.pp = {
+		.count = 1,
+		.base = { 0x21a00 },
+	},
+	.intf = {
+		.base = { 0x00000, 0x21200 },
+		.connect = {
+			[0] = INTF_DISABLED,
+			[1] = INTF_DSI,
+		},
+	},
+	.perf = {
+		.ab_inefficiency = 100,
+		.ib_inefficiency = 200,
+		.clk_inefficiency = 125
+	},
+	.max_clk = 200000000,
+};
+
 static const struct mdp5_cfg_hw msm8x74v2_config = {
 	.name = "msm8x74",
 	.mdp = {
@@ -1236,6 +1317,7 @@ static const struct mdp5_cfg_hw sdm660_config = {
 
 static const struct mdp5_cfg_handler cfg_handlers_v1[] = {
 	{ .revision = 0, .config = { .hw = &msm8x74v1_config } },
+	{ .revision = 1, .config = { .hw = &msm8x26_config } },
 	{ .revision = 2, .config = { .hw = &msm8x74v2_config } },
 	{ .revision = 3, .config = { .hw = &apq8084_config } },
 	{ .revision = 6, .config = { .hw = &msm8x16_config } },
-- 
GitLab


From 9b930f14e530135e0f90bb809a1d8d72f50b1143 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:30 +0300
Subject: [PATCH 0788/1791] drm/msm/dpu: merge dpu_encoder_init() and
 dpu_encoder_setup()

There is no reason to split the dpu_encoder interface into separate
_init() and _setup() phases. Merge them into a single function.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540628/
Link: https://lore.kernel.org/r/20230601172236.564445-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 55 +++++--------
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h | 14 +---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c     | 87 ++++++++-------------
 3 files changed, 56 insertions(+), 100 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 2c9af33c09117..4f44a49f8816f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2433,7 +2433,8 @@ static const struct drm_encoder_funcs dpu_encoder_funcs = {
 		.early_unregister = dpu_encoder_early_unregister,
 };
 
-int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
+struct drm_encoder *dpu_encoder_init(struct drm_device *dev,
+		int drm_enc_mode,
 		struct msm_display_info *disp_info)
 {
 	struct msm_drm_private *priv = dev->dev_private;
@@ -2442,7 +2443,23 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
 	struct dpu_encoder_virt *dpu_enc = NULL;
 	int ret = 0;
 
-	dpu_enc = to_dpu_encoder_virt(enc);
+	dpu_enc = devm_kzalloc(dev->dev, sizeof(*dpu_enc), GFP_KERNEL);
+	if (!dpu_enc)
+		return ERR_PTR(-ENOMEM);
+
+	ret = drm_encoder_init(dev, &dpu_enc->base, &dpu_encoder_funcs,
+			       drm_enc_mode, NULL);
+	if (ret) {
+		devm_kfree(dev->dev, dpu_enc);
+		return ERR_PTR(ret);
+	}
+
+	drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs);
+
+	spin_lock_init(&dpu_enc->enc_spinlock);
+	dpu_enc->enabled = false;
+	mutex_init(&dpu_enc->enc_lock);
+	mutex_init(&dpu_enc->rc_lock);
 
 	ret = dpu_encoder_setup_display(dpu_enc, dpu_kms, disp_info);
 	if (ret)
@@ -2471,44 +2488,14 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
 
 	DPU_DEBUG_ENC(dpu_enc, "created\n");
 
-	return ret;
+	return &dpu_enc->base;
 
 fail:
 	DPU_ERROR("failed to create encoder\n");
 	if (drm_enc)
 		dpu_encoder_destroy(drm_enc);
 
-	return ret;
-
-
-}
-
-struct drm_encoder *dpu_encoder_init(struct drm_device *dev,
-		int drm_enc_mode)
-{
-	struct dpu_encoder_virt *dpu_enc = NULL;
-	int rc = 0;
-
-	dpu_enc = devm_kzalloc(dev->dev, sizeof(*dpu_enc), GFP_KERNEL);
-	if (!dpu_enc)
-		return ERR_PTR(-ENOMEM);
-
-
-	rc = drm_encoder_init(dev, &dpu_enc->base, &dpu_encoder_funcs,
-							  drm_enc_mode, NULL);
-	if (rc) {
-		devm_kfree(dev->dev, dpu_enc);
-		return ERR_PTR(rc);
-	}
-
-	drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs);
-
-	spin_lock_init(&dpu_enc->enc_spinlock);
-	dpu_enc->enabled = false;
-	mutex_init(&dpu_enc->enc_lock);
-	mutex_init(&dpu_enc->rc_lock);
-
-	return &dpu_enc->base;
+	return ERR_PTR(ret);
 }
 
 int dpu_encoder_wait_for_event(struct drm_encoder *drm_enc,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
index 6d14f84dd43fb..90e1925d7770f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h
@@ -130,20 +130,12 @@ void dpu_encoder_virt_runtime_resume(struct drm_encoder *encoder);
 /**
  * dpu_encoder_init - initialize virtual encoder object
  * @dev:        Pointer to drm device structure
+ * @drm_enc_mode: corresponding DRM_MODE_ENCODER_* constant
  * @disp_info:  Pointer to display information structure
  * Returns:     Pointer to newly created drm encoder
  */
-struct drm_encoder *dpu_encoder_init(
-		struct drm_device *dev,
-		int drm_enc_mode);
-
-/**
- * dpu_encoder_setup - setup dpu_encoder for the display probed
- * @dev:		Pointer to drm device structure
- * @enc:		Pointer to the drm_encoder
- * @disp_info:	Pointer to the display info
- */
-int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc,
+struct drm_encoder *dpu_encoder_init(struct drm_device *dev,
+		int drm_enc_mode,
 		struct msm_display_info *disp_info);
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 7f87ac6d68acf..801300e6626f5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -535,15 +535,23 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 		    !msm_dsi_is_master_dsi(priv->dsi[i]))
 			continue;
 
-		encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI);
+		memset(&info, 0, sizeof(info));
+		info.intf_type = INTF_DSI;
+
+		info.h_tile_instance[info.num_of_h_tiles++] = i;
+		if (msm_dsi_is_bonded_dsi(priv->dsi[i]))
+			info.h_tile_instance[info.num_of_h_tiles++] = other;
+
+		info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]);
+
+		info.dsc = msm_dsi_get_dsc_config(priv->dsi[i]);
+
+		encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI, &info);
 		if (IS_ERR(encoder)) {
 			DPU_ERROR("encoder init failed for dsi display\n");
 			return PTR_ERR(encoder);
 		}
 
-		memset(&info, 0, sizeof(info));
-		info.intf_type = INTF_DSI;
-
 		rc = msm_dsi_modeset_init(priv->dsi[i], dev, encoder);
 		if (rc) {
 			DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n",
@@ -551,11 +559,6 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 			break;
 		}
 
-		info.h_tile_instance[info.num_of_h_tiles++] = i;
-		info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]);
-
-		info.dsc = msm_dsi_get_dsc_config(priv->dsi[i]);
-
 		if (msm_dsi_is_bonded_dsi(priv->dsi[i]) && priv->dsi[other]) {
 			rc = msm_dsi_modeset_init(priv->dsi[other], dev, encoder);
 			if (rc) {
@@ -563,14 +566,7 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev,
 					other, rc);
 				break;
 			}
-
-			info.h_tile_instance[info.num_of_h_tiles++] = other;
 		}
-
-		rc = dpu_encoder_setup(dev, encoder, &info);
-		if (rc)
-			DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
-				  encoder->base.id, rc);
 	}
 
 	return rc;
@@ -589,29 +585,23 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev,
 		if (!priv->dp[i])
 			continue;
 
-		encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS);
+		memset(&info, 0, sizeof(info));
+		info.num_of_h_tiles = 1;
+		info.h_tile_instance[0] = i;
+		info.intf_type = INTF_DP;
+
+		encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info);
 		if (IS_ERR(encoder)) {
 			DPU_ERROR("encoder init failed for dsi display\n");
 			return PTR_ERR(encoder);
 		}
 
-		memset(&info, 0, sizeof(info));
 		rc = msm_dp_modeset_init(priv->dp[i], dev, encoder);
 		if (rc) {
 			DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc);
 			drm_encoder_cleanup(encoder);
 			return rc;
 		}
-
-		info.num_of_h_tiles = 1;
-		info.h_tile_instance[0] = i;
-		info.intf_type = INTF_DP;
-		rc = dpu_encoder_setup(dev, encoder, &info);
-		if (rc) {
-			DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
-				  encoder->base.id, rc);
-			return rc;
-		}
 	}
 
 	return 0;
@@ -628,13 +618,17 @@ static int _dpu_kms_initialize_hdmi(struct drm_device *dev,
 	if (!priv->hdmi)
 		return 0;
 
-	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS);
+	memset(&info, 0, sizeof(info));
+	info.num_of_h_tiles = 1;
+	info.h_tile_instance[0] = 0;
+	info.intf_type = INTF_HDMI;
+
+	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info);
 	if (IS_ERR(encoder)) {
 		DPU_ERROR("encoder init failed for HDMI display\n");
 		return PTR_ERR(encoder);
 	}
 
-	memset(&info, 0, sizeof(info));
 	rc = msm_hdmi_modeset_init(priv->hdmi, dev, encoder);
 	if (rc) {
 		DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc);
@@ -642,16 +636,6 @@ static int _dpu_kms_initialize_hdmi(struct drm_device *dev,
 		return rc;
 	}
 
-	info.num_of_h_tiles = 1;
-	info.h_tile_instance[0] = 0;
-	info.intf_type = INTF_HDMI;
-	rc = dpu_encoder_setup(dev, encoder, &info);
-	if (rc) {
-		DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
-			  encoder->base.id, rc);
-		return rc;
-	}
-
 	return 0;
 }
 
@@ -663,14 +647,19 @@ static int _dpu_kms_initialize_writeback(struct drm_device *dev,
 	struct msm_display_info info;
 	int rc;
 
-	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_VIRTUAL);
+	memset(&info, 0, sizeof(info));
+
+	info.num_of_h_tiles = 1;
+	/* use only WB idx 2 instance for DPU */
+	info.h_tile_instance[0] = WB_2;
+	info.intf_type = INTF_WB;
+
+	encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_VIRTUAL, &info);
 	if (IS_ERR(encoder)) {
 		DPU_ERROR("encoder init failed for dsi display\n");
 		return PTR_ERR(encoder);
 	}
 
-	memset(&info, 0, sizeof(info));
-
 	rc = dpu_writeback_init(dev, encoder, wb_formats,
 			n_formats);
 	if (rc) {
@@ -679,18 +668,6 @@ static int _dpu_kms_initialize_writeback(struct drm_device *dev,
 		return rc;
 	}
 
-	info.num_of_h_tiles = 1;
-	/* use only WB idx 2 instance for DPU */
-	info.h_tile_instance[0] = WB_2;
-	info.intf_type = INTF_WB;
-
-	rc = dpu_encoder_setup(dev, encoder, &info);
-	if (rc) {
-		DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n",
-				  encoder->base.id, rc);
-		return rc;
-	}
-
 	return 0;
 }
 
-- 
GitLab


From 7731ee3b06c12b3eb46957b054629e1bb88e99ce Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:31 +0300
Subject: [PATCH 0789/1791] drm/msm/dpu: separate common function to init
 physical encoder

Move common DPU physical encoder initialization code to the new function
dpu_encoder_phys_init().

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540630/
Link: https://lore.kernel.org/r/20230601172236.564445-3-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c   | 29 +++++++++++++++++--
 .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h  |  3 ++
 .../drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c  | 17 ++---------
 .../drm/msm/disp/dpu1/dpu_encoder_phys_vid.c  | 17 ++---------
 .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   | 17 ++---------
 5 files changed, 37 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 4f44a49f8816f..6928743443511 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2366,8 +2366,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 
 	for (i = 0; i < dpu_enc->num_phys_encs; i++) {
 		struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i];
-		atomic_set(&phys->vsync_cnt, 0);
-		atomic_set(&phys->underrun_cnt, 0);
 
 		if (phys->intf_idx >= INTF_0 && phys->intf_idx < INTF_MAX)
 			phys->hw_intf = dpu_rm_get_intf(&dpu_kms->rm, phys->intf_idx);
@@ -2569,3 +2567,30 @@ unsigned int dpu_encoder_helper_get_dsc(struct dpu_encoder_phys *phys_enc)
 
 	return dpu_enc->dsc_mask;
 }
+
+void dpu_encoder_phys_init(struct dpu_encoder_phys *phys_enc,
+			  struct dpu_enc_phys_init_params *p)
+{
+	int i;
+
+	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
+	phys_enc->intf_idx = p->intf_idx;
+	phys_enc->wb_idx = p->wb_idx;
+	phys_enc->parent = p->parent;
+	phys_enc->dpu_kms = p->dpu_kms;
+	phys_enc->split_role = p->split_role;
+	phys_enc->enc_spinlock = p->enc_spinlock;
+	phys_enc->enable_state = DPU_ENC_DISABLED;
+
+	for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++)
+		phys_enc->irq[i] = -EINVAL;
+
+	atomic_set(&phys_enc->vblank_refcount, 0);
+	atomic_set(&phys_enc->pending_kickoff_cnt, 0);
+	atomic_set(&phys_enc->pending_ctlstart_cnt, 0);
+
+	atomic_set(&phys_enc->vsync_cnt, 0);
+	atomic_set(&phys_enc->underrun_cnt, 0);
+
+	init_waitqueue_head(&phys_enc->pending_kickoff_wq);
+}
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index 90f177e43262a..aa98bfb70a260 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -407,4 +407,7 @@ void dpu_encoder_frame_done_callback(
 		struct drm_encoder *drm_enc,
 		struct dpu_encoder_phys *ready_phys, u32 event);
 
+void dpu_encoder_phys_init(struct dpu_encoder_phys *phys,
+			   struct dpu_enc_phys_init_params *p);
+
 #endif /* __dpu_encoder_phys_H__ */
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index d8ed85a238af5..2bd806c51882b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -756,7 +756,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 {
 	struct dpu_encoder_phys *phys_enc = NULL;
 	struct dpu_encoder_phys_cmd *cmd_enc = NULL;
-	int i, ret = 0;
+	int ret = 0;
 
 	DPU_DEBUG("intf %d\n", p->intf_idx - INTF_0);
 
@@ -767,28 +767,17 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 		return ERR_PTR(ret);
 	}
 	phys_enc = &cmd_enc->base;
-	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
-	phys_enc->intf_idx = p->intf_idx;
+
+	dpu_encoder_phys_init(phys_enc, p);
 
 	dpu_encoder_phys_cmd_init_ops(&phys_enc->ops);
-	phys_enc->parent = p->parent;
-	phys_enc->dpu_kms = p->dpu_kms;
-	phys_enc->split_role = p->split_role;
 	phys_enc->intf_mode = INTF_MODE_CMD;
-	phys_enc->enc_spinlock = p->enc_spinlock;
 	cmd_enc->stream_sel = 0;
-	phys_enc->enable_state = DPU_ENC_DISABLED;
-	for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++)
-		phys_enc->irq[i] = -EINVAL;
 
 	phys_enc->has_intf_te = test_bit(DPU_INTF_TE,
 			&phys_enc->dpu_kms->catalog->intf[p->intf_idx - INTF_0].features);
 
-	atomic_set(&phys_enc->vblank_refcount, 0);
-	atomic_set(&phys_enc->pending_kickoff_cnt, 0);
-	atomic_set(&phys_enc->pending_ctlstart_cnt, 0);
 	atomic_set(&cmd_enc->pending_vblank_cnt, 0);
-	init_waitqueue_head(&phys_enc->pending_kickoff_wq);
 	init_waitqueue_head(&cmd_enc->pending_vblank_wq);
 
 	DPU_DEBUG_CMDENC(cmd_enc, "created\n");
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index 3a374292f3116..dc951fdf473b3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -699,7 +699,6 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 		struct dpu_enc_phys_init_params *p)
 {
 	struct dpu_encoder_phys *phys_enc = NULL;
-	int i;
 
 	if (!p) {
 		DPU_ERROR("failed to create encoder due to invalid parameter\n");
@@ -712,24 +711,12 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 		return ERR_PTR(-ENOMEM);
 	}
 
-	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
-	phys_enc->intf_idx = p->intf_idx;
-
 	DPU_DEBUG_VIDENC(phys_enc, "\n");
 
+	dpu_encoder_phys_init(phys_enc, p);
+
 	dpu_encoder_phys_vid_init_ops(&phys_enc->ops);
-	phys_enc->parent = p->parent;
-	phys_enc->dpu_kms = p->dpu_kms;
-	phys_enc->split_role = p->split_role;
 	phys_enc->intf_mode = INTF_MODE_VIDEO;
-	phys_enc->enc_spinlock = p->enc_spinlock;
-	for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++)
-		phys_enc->irq[i] = -EINVAL;
-
-	atomic_set(&phys_enc->vblank_refcount, 0);
-	atomic_set(&phys_enc->pending_kickoff_cnt, 0);
-	init_waitqueue_head(&phys_enc->pending_kickoff_wq);
-	phys_enc->enable_state = DPU_ENC_DISABLED;
 
 	DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->intf_idx);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index 023a9c4ad1db4..008d1d09b9baf 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -685,7 +685,6 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init(
 	struct dpu_encoder_phys *phys_enc = NULL;
 	struct dpu_encoder_phys_wb *wb_enc = NULL;
 	int ret = 0;
-	int i;
 
 	DPU_DEBUG("\n");
 
@@ -703,28 +702,16 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init(
 	}
 
 	phys_enc = &wb_enc->base;
-	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
-	phys_enc->wb_idx = p->wb_idx;
+
+	dpu_encoder_phys_init(phys_enc, p);
 
 	dpu_encoder_phys_wb_init_ops(&phys_enc->ops);
-	phys_enc->parent = p->parent;
-	phys_enc->dpu_kms = p->dpu_kms;
-	phys_enc->split_role = p->split_role;
 	phys_enc->intf_mode = INTF_MODE_WB_LINE;
-	phys_enc->wb_idx = p->wb_idx;
-	phys_enc->enc_spinlock = p->enc_spinlock;
 
 	atomic_set(&wb_enc->wbirq_refcount, 0);
 
-	for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++)
-		phys_enc->irq[i] = -EINVAL;
-
-	atomic_set(&phys_enc->pending_kickoff_cnt, 0);
-	atomic_set(&phys_enc->vblank_refcount, 0);
 	wb_enc->wb_done_timeout_cnt = 0;
 
-	init_waitqueue_head(&phys_enc->pending_kickoff_wq);
-	phys_enc->enable_state = DPU_ENC_DISABLED;
 
 	DPU_DEBUG("Created dpu_encoder_phys for wb %d\n",
 			phys_enc->wb_idx);
-- 
GitLab


From 8ea432b8428e9b324eee2bbfafbded99f80e3414 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:32 +0300
Subject: [PATCH 0790/1791] drm/msm/dpu: drop duplicated intf/wb indices from
 encoder structs

Remove intf_idx and wb_idx fields from struct dpu_encoder_phys and
struct dpu_enc_phys_init_params. Set the hw_intf and hw_wb directly and
use them to get the instance index.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540635/
Link: https://lore.kernel.org/r/20230601172236.564445-4-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c   | 72 ++++++++-----------
 .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h  | 12 ++--
 .../drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c  | 18 ++---
 .../drm/msm/disp/dpu1/dpu_encoder_phys_vid.c  |  2 +-
 .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   |  8 +--
 5 files changed, 47 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 6928743443511..51484d842c828 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -339,7 +339,8 @@ void dpu_encoder_helper_report_irq_timeout(struct dpu_encoder_phys *phys_enc,
 	DRM_ERROR("irq timeout id=%u, intf_mode=%s intf=%d wb=%d, pp=%d, intr=%d\n",
 			DRMID(phys_enc->parent),
 			dpu_encoder_helper_get_intf_type(phys_enc->intf_mode),
-			phys_enc->intf_idx - INTF_0, phys_enc->wb_idx - WB_0,
+			phys_enc->hw_intf ? phys_enc->hw_intf->idx - INTF_0 : -1,
+			phys_enc->hw_wb ? phys_enc->hw_wb->idx - WB_0 : -1,
 			phys_enc->hw_pp->idx - PINGPONG_0, intr_idx);
 
 	dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc,
@@ -1419,7 +1420,8 @@ void dpu_encoder_frame_done_callback(
 			 */
 			trace_dpu_enc_frame_done_cb_not_busy(DRMID(drm_enc), event,
 					dpu_encoder_helper_get_intf_type(ready_phys->intf_mode),
-					ready_phys->intf_idx, ready_phys->wb_idx);
+					ready_phys->hw_intf ? ready_phys->hw_intf->idx : -1,
+					ready_phys->hw_wb ? ready_phys->hw_wb->idx : -1);
 			return;
 		}
 
@@ -1499,7 +1501,8 @@ static void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc,
 
 	trace_dpu_enc_trigger_flush(DRMID(drm_enc),
 			dpu_encoder_helper_get_intf_type(phys->intf_mode),
-			phys->intf_idx, phys->wb_idx,
+			phys->hw_intf ? phys->hw_intf->idx : -1,
+			phys->hw_wb ? phys->hw_wb->idx : -1,
 			pending_kickoff_cnt, ctl->idx,
 			extra_flush_bits, ret);
 }
@@ -2155,7 +2158,8 @@ static int _dpu_encoder_status_show(struct seq_file *s, void *data)
 		struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i];
 
 		seq_printf(s, "intf:%d  wb:%d  vsync:%8d     underrun:%8d    ",
-				phys->intf_idx - INTF_0, phys->wb_idx - WB_0,
+				phys->hw_intf ? phys->hw_intf->idx - INTF_0 : -1,
+				phys->hw_wb ? phys->hw_wb->idx - WB_0 : -1,
 				atomic_read(&phys->vsync_cnt),
 				atomic_read(&phys->underrun_cnt));
 
@@ -2319,6 +2323,8 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		 * h_tile_instance_ids[2] = {1, 0}; DSI1 = left, DSI0 = right
 		 */
 		u32 controller_id = disp_info->h_tile_instance[i];
+		enum dpu_intf intf_idx;
+		enum dpu_wb wb_idx;
 
 		if (disp_info->num_of_h_tiles > 1) {
 			if (i == 0)
@@ -2332,57 +2338,39 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		DPU_DEBUG("h_tile_instance %d = %d, split_role %d\n",
 				i, controller_id, phys_params.split_role);
 
-		phys_params.intf_idx = dpu_encoder_get_intf(dpu_kms->catalog,
+		intf_idx = dpu_encoder_get_intf(dpu_kms->catalog,
 							    disp_info->intf_type,
 							    controller_id);
 
-		phys_params.wb_idx = dpu_encoder_get_wb(dpu_kms->catalog,
+		wb_idx = dpu_encoder_get_wb(dpu_kms->catalog,
 				disp_info->intf_type, controller_id);
-		/*
-		 * The phys_params might represent either an INTF or a WB unit, but not
-		 * both of them at the same time.
-		 */
-		if ((phys_params.intf_idx == INTF_MAX) &&
-				(phys_params.wb_idx == WB_MAX)) {
-			DPU_ERROR_ENC(dpu_enc, "could not get intf or wb: type %d, id %d\n",
-						  disp_info->intf_type, controller_id);
-			ret = -EINVAL;
-		}
 
-		if ((phys_params.intf_idx != INTF_MAX) &&
-				(phys_params.wb_idx != WB_MAX)) {
-			DPU_ERROR_ENC(dpu_enc, "both intf and wb present: type %d, id %d\n",
-						  disp_info->intf_type, controller_id);
-			ret = -EINVAL;
-		}
+		if (intf_idx >= INTF_0 && intf_idx < INTF_MAX)
+			phys_params.hw_intf = dpu_rm_get_intf(&dpu_kms->rm, intf_idx);
 
-		if (!ret) {
-			ret = dpu_encoder_virt_add_phys_encs(disp_info,
-					dpu_enc, &phys_params);
-			if (ret)
-				DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n");
-		}
-	}
+		if (wb_idx >= WB_0 && wb_idx < WB_MAX)
+			phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, wb_idx);
 
-	for (i = 0; i < dpu_enc->num_phys_encs; i++) {
-		struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i];
-
-		if (phys->intf_idx >= INTF_0 && phys->intf_idx < INTF_MAX)
-			phys->hw_intf = dpu_rm_get_intf(&dpu_kms->rm, phys->intf_idx);
-
-		if (phys->wb_idx >= WB_0 && phys->wb_idx < WB_MAX)
-			phys->hw_wb = dpu_rm_get_wb(&dpu_kms->rm, phys->wb_idx);
-
-		if (!phys->hw_intf && !phys->hw_wb) {
+		if (!phys_params.hw_intf && !phys_params.hw_wb) {
 			DPU_ERROR_ENC(dpu_enc, "no intf or wb block assigned at idx: %d\n", i);
 			ret = -EINVAL;
+			break;
 		}
 
-		if (phys->hw_intf && phys->hw_wb) {
+		if (phys_params.hw_intf && phys_params.hw_wb) {
 			DPU_ERROR_ENC(dpu_enc,
 					"invalid phys both intf and wb block at idx: %d\n", i);
 			ret = -EINVAL;
+			break;
 		}
+
+		ret = dpu_encoder_virt_add_phys_encs(disp_info,
+				dpu_enc, &phys_params);
+		if (ret) {
+			DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n");
+			break;
+		}
+
 	}
 
 	mutex_unlock(&dpu_enc->enc_lock);
@@ -2574,8 +2562,8 @@ void dpu_encoder_phys_init(struct dpu_encoder_phys *phys_enc,
 	int i;
 
 	phys_enc->hw_mdptop = p->dpu_kms->hw_mdp;
-	phys_enc->intf_idx = p->intf_idx;
-	phys_enc->wb_idx = p->wb_idx;
+	phys_enc->hw_intf = p->hw_intf;
+	phys_enc->hw_wb = p->hw_wb;
 	phys_enc->parent = p->parent;
 	phys_enc->dpu_kms = p->dpu_kms;
 	phys_enc->split_role = p->split_role;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
index aa98bfb70a260..cf64d7f608fdd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h
@@ -161,8 +161,6 @@ enum dpu_intr_idx {
  * @enabled:		Whether the encoder has enabled and running a mode
  * @split_role:		Role to play in a split-panel configuration
  * @intf_mode:		Interface mode
- * @intf_idx:		Interface index on dpu hardware
- * @wb_idx:			Writeback index on dpu hardware
  * @enc_spinlock:	Virtual-Encoder-Wide Spin Lock for IRQ purposes
  * @enable_state:	Enable state tracking
  * @vblank_refcount:	Reference count of vblank request
@@ -190,8 +188,6 @@ struct dpu_encoder_phys {
 	struct drm_display_mode cached_mode;
 	enum dpu_enc_split_role split_role;
 	enum dpu_intf_mode intf_mode;
-	enum dpu_intf intf_idx;
-	enum dpu_wb wb_idx;
 	spinlock_t *enc_spinlock;
 	enum dpu_enc_enable_state enable_state;
 	atomic_t vblank_refcount;
@@ -258,16 +254,16 @@ struct dpu_encoder_phys_cmd {
  * @parent:		Pointer to the containing virtual encoder
  * @parent_ops:		Callbacks exposed by the parent to the phys_enc
  * @split_role:		Role to play in a split-panel configuration
- * @intf_idx:		Interface index this phys_enc will control
- * @wb_idx:			Writeback index this phys_enc will control
+ * @hw_intf:		Hardware interface to the intf registers
+ * @hw_wb:		Hardware interface to the wb registers
  * @enc_spinlock:	Virtual-Encoder-Wide Spin Lock for IRQ purposes
  */
 struct dpu_enc_phys_init_params {
 	struct dpu_kms *dpu_kms;
 	struct drm_encoder *parent;
 	enum dpu_enc_split_role split_role;
-	enum dpu_intf intf_idx;
-	enum dpu_wb wb_idx;
+	struct dpu_hw_intf *hw_intf;
+	struct dpu_hw_wb *hw_wb;
 	spinlock_t *enc_spinlock;
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 2bd806c51882b..2cc6b0cd2710d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -16,12 +16,12 @@
 #define DPU_DEBUG_CMDENC(e, fmt, ...) DPU_DEBUG("enc%d intf%d " fmt, \
 		(e) && (e)->base.parent ? \
 		(e)->base.parent->base.id : -1, \
-		(e) ? (e)->base.intf_idx - INTF_0 : -1, ##__VA_ARGS__)
+		(e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__)
 
 #define DPU_ERROR_CMDENC(e, fmt, ...) DPU_ERROR("enc%d intf%d " fmt, \
 		(e) && (e)->base.parent ? \
 		(e)->base.parent->base.id : -1, \
-		(e) ? (e)->base.intf_idx - INTF_0 : -1, ##__VA_ARGS__)
+		(e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__)
 
 #define to_dpu_encoder_phys_cmd(x) \
 	container_of(x, struct dpu_encoder_phys_cmd, base)
@@ -55,7 +55,7 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg(
 	if (!ctl->ops.setup_intf_cfg)
 		return;
 
-	intf_cfg.intf = phys_enc->intf_idx;
+	intf_cfg.intf = phys_enc->hw_intf->idx;
 	intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_CMD;
 	intf_cfg.stream_sel = cmd_enc->stream_sel;
 	intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc);
@@ -440,7 +440,7 @@ static void dpu_encoder_phys_cmd_enable_helper(
 		return;
 	}
 
-	dpu_encoder_helper_split_config(phys_enc, phys_enc->intf_idx);
+	dpu_encoder_helper_split_config(phys_enc, phys_enc->hw_intf->idx);
 
 	_dpu_encoder_phys_cmd_pingpong_config(phys_enc);
 
@@ -448,7 +448,7 @@ static void dpu_encoder_phys_cmd_enable_helper(
 		return;
 
 	ctl = phys_enc->hw_ctl;
-	ctl->ops.update_pending_flush_intf(ctl, phys_enc->intf_idx);
+	ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx);
 }
 
 static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc)
@@ -557,7 +557,7 @@ static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc)
 				phys_enc->hw_pp->idx);
 
 		ctl = phys_enc->hw_ctl;
-		ctl->ops.update_pending_flush_intf(ctl, phys_enc->intf_idx);
+		ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx);
 	}
 
 	phys_enc->enable_state = DPU_ENC_DISABLED;
@@ -667,7 +667,7 @@ static int dpu_encoder_phys_cmd_wait_for_tx_complete(
 	if (rc) {
 		DRM_ERROR("failed wait_for_idle: id:%u ret:%d intf:%d\n",
 			  DRMID(phys_enc->parent), rc,
-			  phys_enc->intf_idx - INTF_0);
+			  phys_enc->hw_intf->idx - INTF_0);
 	}
 
 	return rc;
@@ -758,7 +758,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 	struct dpu_encoder_phys_cmd *cmd_enc = NULL;
 	int ret = 0;
 
-	DPU_DEBUG("intf %d\n", p->intf_idx - INTF_0);
+	DPU_DEBUG("intf\n");
 
 	cmd_enc = kzalloc(sizeof(*cmd_enc), GFP_KERNEL);
 	if (!cmd_enc) {
@@ -775,7 +775,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 	cmd_enc->stream_sel = 0;
 
 	phys_enc->has_intf_te = test_bit(DPU_INTF_TE,
-			&phys_enc->dpu_kms->catalog->intf[p->intf_idx - INTF_0].features);
+					 &phys_enc->hw_intf->cap->features);
 
 	atomic_set(&cmd_enc->pending_vblank_cnt, 0);
 	init_waitqueue_head(&cmd_enc->pending_vblank_wq);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index dc951fdf473b3..e26629e9e3030 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -718,7 +718,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init(
 	dpu_encoder_phys_vid_init_ops(&phys_enc->ops);
 	phys_enc->intf_mode = INTF_MODE_VIDEO;
 
-	DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->intf_idx);
+	DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->hw_intf->idx);
 
 	return phys_enc;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index 008d1d09b9baf..6608c00e3c33e 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -240,7 +240,7 @@ static int dpu_encoder_phys_wb_atomic_check(
 	const struct drm_display_mode *mode = &crtc_state->mode;
 
 	DPU_DEBUG("[atomic_check:%d, \"%s\",%d,%d]\n",
-			phys_enc->wb_idx, mode->name, mode->hdisplay, mode->vdisplay);
+			phys_enc->hw_wb->idx, mode->name, mode->hdisplay, mode->vdisplay);
 
 	if (!conn_state || !conn_state->connector) {
 		DPU_ERROR("invalid connector state\n");
@@ -561,7 +561,7 @@ static void dpu_encoder_phys_wb_destroy(struct dpu_encoder_phys *phys_enc)
 	if (!phys_enc)
 		return;
 
-	DPU_DEBUG("[wb:%d]\n", phys_enc->wb_idx - WB_0);
+	DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0);
 
 	kfree(phys_enc);
 }
@@ -712,9 +712,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init(
 
 	wb_enc->wb_done_timeout_cnt = 0;
 
-
-	DPU_DEBUG("Created dpu_encoder_phys for wb %d\n",
-			phys_enc->wb_idx);
+	DPU_DEBUG("Created dpu_encoder_phys for wb %d\n", phys_enc->hw_wb->idx);
 
 	return phys_enc;
 
-- 
GitLab


From 04ada3d8380ac8623fc79b172427b984535f8215 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:33 +0300
Subject: [PATCH 0791/1791] drm/msm/dpu: inline dpu_encoder_get_wb()

The function dpu_encoder_get_wb() returns controller_id if the
corresponding WB is present in the catalog. We can inline this function
and rely on dpu_rm_get_wb() returning NULL for indices for which the
WB is not present on the device.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540634/
Link: https://lore.kernel.org/r/20230601172236.564445-5-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 24 ++-------------------
 1 file changed, 2 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index 51484d842c828..bb89ee2ad958a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1288,22 +1288,6 @@ static enum dpu_intf dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog,
 	return INTF_MAX;
 }
 
-static enum dpu_wb dpu_encoder_get_wb(const struct dpu_mdss_cfg *catalog,
-		enum dpu_intf_type type, u32 controller_id)
-{
-	int i = 0;
-
-	if (type != INTF_WB)
-		return WB_MAX;
-
-	for (i = 0; i < catalog->wb_count; i++) {
-		if (catalog->wb[i].id == controller_id)
-			return catalog->wb[i].id;
-	}
-
-	return WB_MAX;
-}
-
 void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc,
 		struct dpu_encoder_phys *phy_enc)
 {
@@ -2324,7 +2308,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		 */
 		u32 controller_id = disp_info->h_tile_instance[i];
 		enum dpu_intf intf_idx;
-		enum dpu_wb wb_idx;
 
 		if (disp_info->num_of_h_tiles > 1) {
 			if (i == 0)
@@ -2342,14 +2325,11 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 							    disp_info->intf_type,
 							    controller_id);
 
-		wb_idx = dpu_encoder_get_wb(dpu_kms->catalog,
-				disp_info->intf_type, controller_id);
-
 		if (intf_idx >= INTF_0 && intf_idx < INTF_MAX)
 			phys_params.hw_intf = dpu_rm_get_intf(&dpu_kms->rm, intf_idx);
 
-		if (wb_idx >= WB_0 && wb_idx < WB_MAX)
-			phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, wb_idx);
+		if (disp_info->intf_type == INTF_WB && controller_id < WB_MAX)
+			phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, controller_id);
 
 		if (!phys_params.hw_intf && !phys_params.hw_wb) {
 			DPU_ERROR_ENC(dpu_enc, "no intf or wb block assigned at idx: %d\n", i);
-- 
GitLab


From 004be38657f45f6ecba4158dde2754bc4b75081d Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:34 +0300
Subject: [PATCH 0792/1791] drm/msm/dpu: call dpu_rm_get_intf() from
 dpu_encoder_get_intf()

There is little sense to get intf index just to call dpu_rm_get_intf()
on it. Move dpu_rm_get_intf() call to dpu_encoder_get_intf() function.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540632/
Link: https://lore.kernel.org/r/20230601172236.564445-6-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index bb89ee2ad958a..f32882e83255d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -1270,22 +1270,23 @@ static void dpu_encoder_virt_atomic_disable(struct drm_encoder *drm_enc,
 	mutex_unlock(&dpu_enc->enc_lock);
 }
 
-static enum dpu_intf dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog,
+static struct dpu_hw_intf *dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog,
+		struct dpu_rm *dpu_rm,
 		enum dpu_intf_type type, u32 controller_id)
 {
 	int i = 0;
 
 	if (type == INTF_WB)
-		return INTF_MAX;
+		return NULL;
 
 	for (i = 0; i < catalog->intf_count; i++) {
 		if (catalog->intf[i].type == type
 		    && catalog->intf[i].controller_id == controller_id) {
-			return catalog->intf[i].id;
+			return dpu_rm_get_intf(dpu_rm, catalog->intf[i].id);
 		}
 	}
 
-	return INTF_MAX;
+	return NULL;
 }
 
 void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc,
@@ -2307,7 +2308,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		 * h_tile_instance_ids[2] = {1, 0}; DSI1 = left, DSI0 = right
 		 */
 		u32 controller_id = disp_info->h_tile_instance[i];
-		enum dpu_intf intf_idx;
 
 		if (disp_info->num_of_h_tiles > 1) {
 			if (i == 0)
@@ -2321,12 +2321,9 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 		DPU_DEBUG("h_tile_instance %d = %d, split_role %d\n",
 				i, controller_id, phys_params.split_role);
 
-		intf_idx = dpu_encoder_get_intf(dpu_kms->catalog,
-							    disp_info->intf_type,
-							    controller_id);
-
-		if (intf_idx >= INTF_0 && intf_idx < INTF_MAX)
-			phys_params.hw_intf = dpu_rm_get_intf(&dpu_kms->rm, intf_idx);
+		phys_params.hw_intf = dpu_encoder_get_intf(dpu_kms->catalog, &dpu_kms->rm,
+							   disp_info->intf_type,
+							   controller_id);
 
 		if (disp_info->intf_type == INTF_WB && controller_id < WB_MAX)
 			phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, controller_id);
@@ -2350,7 +2347,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc,
 			DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n");
 			break;
 		}
-
 	}
 
 	mutex_unlock(&dpu_enc->enc_lock);
-- 
GitLab


From ed02d3dd18dc35a8f8ca0527c1eaebb992585ec6 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:35 +0300
Subject: [PATCH 0793/1791] drm/msm/dpu: drop temp variable from
 dpu_encoder_phys_cmd_init()

There is no need to assign a result to temp varable just to return it
two lines below. Drop the temporary variable.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540637/
Link: https://lore.kernel.org/r/20230601172236.564445-7-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 2cc6b0cd2710d..4f8c9187f76dc 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -756,15 +756,13 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init(
 {
 	struct dpu_encoder_phys *phys_enc = NULL;
 	struct dpu_encoder_phys_cmd *cmd_enc = NULL;
-	int ret = 0;
 
 	DPU_DEBUG("intf\n");
 
 	cmd_enc = kzalloc(sizeof(*cmd_enc), GFP_KERNEL);
 	if (!cmd_enc) {
-		ret = -ENOMEM;
 		DPU_ERROR("failed to allocate\n");
-		return ERR_PTR(ret);
+		return ERR_PTR(-ENOMEM);
 	}
 	phys_enc = &cmd_enc->base;
 
-- 
GitLab


From 5d1227d1c7a38adb26a0203ecaf5873b2bbbf7b6 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 1 Jun 2023 20:22:36 +0300
Subject: [PATCH 0794/1791] drm/msm/dpu: simplify dpu_encoder_phys_wb_init()

There is no need to assign a result to temp varable just to return it
after a goto. Drop the temporary variable and goto and return the result
directly.

Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
Patchwork: https://patchwork.freedesktop.org/patch/540639/
Link: https://lore.kernel.org/r/20230601172236.564445-8-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index 6608c00e3c33e..e9325cafb1a85 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -684,21 +684,18 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init(
 {
 	struct dpu_encoder_phys *phys_enc = NULL;
 	struct dpu_encoder_phys_wb *wb_enc = NULL;
-	int ret = 0;
 
 	DPU_DEBUG("\n");
 
 	if (!p || !p->parent) {
 		DPU_ERROR("invalid params\n");
-		ret = -EINVAL;
-		goto fail_alloc;
+		return ERR_PTR(-EINVAL);
 	}
 
 	wb_enc = kzalloc(sizeof(*wb_enc), GFP_KERNEL);
 	if (!wb_enc) {
 		DPU_ERROR("failed to allocate wb phys_enc enc\n");
-		ret = -ENOMEM;
-		goto fail_alloc;
+		return ERR_PTR(-ENOMEM);
 	}
 
 	phys_enc = &wb_enc->base;
@@ -715,7 +712,4 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init(
 	DPU_DEBUG("Created dpu_encoder_phys for wb %d\n", phys_enc->hw_wb->idx);
 
 	return phys_enc;
-
-fail_alloc:
-	return ERR_PTR(ret);
 }
-- 
GitLab


From 1c5f6051ffdfdc633af088e565bc028e459b57d2 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sat, 20 May 2023 21:20:50 +0300
Subject: [PATCH 0795/1791] drm/msm/dp: remove most of usbpd-related remains

Remove most of remains of downstream usbpd code. Mainline kernel uses
different approach for managing Type-C / USB-PD, so this remains unused.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Bjorn Andersson <andersson@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/538270/
Link: https://lore.kernel.org/r/20230520182050.4014143-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/Makefile        |  1 -
 drivers/gpu/drm/msm/dp/dp_ctrl.h    |  1 -
 drivers/gpu/drm/msm/dp/dp_debug.c   |  6 +--
 drivers/gpu/drm/msm/dp/dp_debug.h   |  5 +-
 drivers/gpu/drm/msm/dp/dp_display.c | 34 +------------
 drivers/gpu/drm/msm/dp/dp_hpd.c     | 67 -------------------------
 drivers/gpu/drm/msm/dp/dp_hpd.h     | 78 -----------------------------
 drivers/gpu/drm/msm/dp/dp_panel.h   |  1 -
 drivers/gpu/drm/msm/dp/dp_power.c   |  2 +-
 drivers/gpu/drm/msm/dp/dp_power.h   |  3 +-
 10 files changed, 8 insertions(+), 190 deletions(-)
 delete mode 100644 drivers/gpu/drm/msm/dp/dp_hpd.c
 delete mode 100644 drivers/gpu/drm/msm/dp/dp_hpd.h

diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 7274c41228ed9..39d9ac05c4b6c 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -122,7 +122,6 @@ msm-$(CONFIG_DRM_MSM_DP)+= dp/dp_aux.o \
 	dp/dp_ctrl.o \
 	dp/dp_display.o \
 	dp/dp_drm.o \
-	dp/dp_hpd.o \
 	dp/dp_link.o \
 	dp/dp_panel.o \
 	dp/dp_parser.o \
diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.h b/drivers/gpu/drm/msm/dp/dp_ctrl.h
index f712780149fd6..b2c27d3532bf5 100644
--- a/drivers/gpu/drm/msm/dp/dp_ctrl.h
+++ b/drivers/gpu/drm/msm/dp/dp_ctrl.h
@@ -14,7 +14,6 @@
 #include "dp_catalog.h"
 
 struct dp_ctrl {
-	bool orientation;
 	atomic_t aborted;
 	bool wide_bus_en;
 };
diff --git a/drivers/gpu/drm/msm/dp/dp_debug.c b/drivers/gpu/drm/msm/dp/dp_debug.c
index 5e35033ba3e43..3bba901afe335 100644
--- a/drivers/gpu/drm/msm/dp/dp_debug.c
+++ b/drivers/gpu/drm/msm/dp/dp_debug.c
@@ -21,7 +21,6 @@
 struct dp_debug_private {
 	struct dentry *root;
 
-	struct dp_usbpd *usbpd;
 	struct dp_link *link;
 	struct dp_panel *panel;
 	struct drm_connector *connector;
@@ -232,14 +231,14 @@ static void dp_debug_init(struct dp_debug *dp_debug, struct drm_minor *minor)
 }
 
 struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
-		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct dp_link *link,
 		struct drm_connector *connector, struct drm_minor *minor)
 {
 	struct dp_debug_private *debug;
 	struct dp_debug *dp_debug;
 	int rc;
 
-	if (!dev || !panel || !usbpd || !link) {
+	if (!dev || !panel || !link) {
 		DRM_ERROR("invalid input\n");
 		rc = -EINVAL;
 		goto error;
@@ -252,7 +251,6 @@ struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
 	}
 
 	debug->dp_debug.debug_en = false;
-	debug->usbpd = usbpd;
 	debug->link = link;
 	debug->panel = panel;
 	debug->dev = dev;
diff --git a/drivers/gpu/drm/msm/dp/dp_debug.h b/drivers/gpu/drm/msm/dp/dp_debug.h
index 8c0d0b5178fdf..124227873d58c 100644
--- a/drivers/gpu/drm/msm/dp/dp_debug.h
+++ b/drivers/gpu/drm/msm/dp/dp_debug.h
@@ -32,7 +32,6 @@ struct dp_debug {
  *
  * @dev: device instance of the caller
  * @panel: instance of panel module
- * @usbpd: instance of usbpd module
  * @link: instance of link module
  * @connector: double pointer to display connector
  * @minor: pointer to drm minor number after device registration
@@ -42,7 +41,7 @@ struct dp_debug {
  * for debugfs input to be communicated with existing modules
  */
 struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
-		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct dp_link *link,
 		struct drm_connector *connector,
 		struct drm_minor *minor);
 
@@ -59,7 +58,7 @@ void dp_debug_put(struct dp_debug *dp_debug);
 
 static inline
 struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel,
-		struct dp_usbpd *usbpd, struct dp_link *link,
+		struct dp_link *link,
 		struct drm_connector *connector, struct drm_minor *minor)
 {
 	return ERR_PTR(-EINVAL);
diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
index 99a38dbe51c0f..3d8fa2e735838 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.c
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -14,7 +14,6 @@
 
 #include "msm_drv.h"
 #include "msm_kms.h"
-#include "dp_hpd.h"
 #include "dp_parser.h"
 #include "dp_power.h"
 #include "dp_catalog.h"
@@ -88,7 +87,6 @@ struct dp_display_private {
 	struct platform_device *pdev;
 	struct dentry *root;
 
-	struct dp_usbpd   *usbpd;
 	struct dp_parser  *parser;
 	struct dp_power   *power;
 	struct dp_catalog *catalog;
@@ -98,7 +96,6 @@ struct dp_display_private {
 	struct dp_ctrl    *ctrl;
 	struct dp_debug   *debug;
 
-	struct dp_usbpd_cb usbpd_cb;
 	struct dp_display_mode dp_mode;
 	struct msm_dp dp_display;
 
@@ -463,7 +460,7 @@ static void dp_display_host_init(struct dp_display_private *dp)
 		dp->dp_display.connector_type, dp->core_initialized,
 		dp->phy_initialized);
 
-	dp_power_init(dp->power, false);
+	dp_power_init(dp->power);
 	dp_ctrl_reset_irq_ctrl(dp->ctrl, true);
 	dp_aux_init(dp->aux);
 	dp->core_initialized = true;
@@ -490,11 +487,6 @@ static int dp_display_usbpd_configure_cb(struct device *dev)
 	return dp_display_process_hpd_high(dp);
 }
 
-static int dp_display_usbpd_disconnect_cb(struct device *dev)
-{
-	return 0;
-}
-
 static int dp_display_notify_disconnect(struct device *dev)
 {
 	struct dp_display_private *dp = dev_get_dp_display_private(dev);
@@ -579,13 +571,9 @@ static int dp_display_usbpd_attention_cb(struct device *dev)
 
 static int dp_hpd_plug_handle(struct dp_display_private *dp, u32 data)
 {
-	struct dp_usbpd *hpd = dp->usbpd;
 	u32 state;
 	int ret;
 
-	if (!hpd)
-		return 0;
-
 	mutex_lock(&dp->event_mutex);
 
 	state =  dp->hpd_state;
@@ -646,12 +634,8 @@ static void dp_display_handle_plugged_change(struct msm_dp *dp_display,
 
 static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data)
 {
-	struct dp_usbpd *hpd = dp->usbpd;
 	u32 state;
 
-	if (!hpd)
-		return 0;
-
 	mutex_lock(&dp->event_mutex);
 
 	state = dp->hpd_state;
@@ -764,24 +748,10 @@ static int dp_init_sub_modules(struct dp_display_private *dp)
 {
 	int rc = 0;
 	struct device *dev = &dp->pdev->dev;
-	struct dp_usbpd_cb *cb = &dp->usbpd_cb;
 	struct dp_panel_in panel_in = {
 		.dev = dev,
 	};
 
-	/* Callback APIs used for cable status change event */
-	cb->configure  = dp_display_usbpd_configure_cb;
-	cb->disconnect = dp_display_usbpd_disconnect_cb;
-	cb->attention  = dp_display_usbpd_attention_cb;
-
-	dp->usbpd = dp_hpd_get(dev, cb);
-	if (IS_ERR(dp->usbpd)) {
-		rc = PTR_ERR(dp->usbpd);
-		DRM_ERROR("failed to initialize hpd, rc = %d\n", rc);
-		dp->usbpd = NULL;
-		goto error;
-	}
-
 	dp->parser = dp_parser_get(dp->pdev);
 	if (IS_ERR(dp->parser)) {
 		rc = PTR_ERR(dp->parser);
@@ -1541,7 +1511,7 @@ void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor)
 	dp = container_of(dp_display, struct dp_display_private, dp_display);
 	dev = &dp->pdev->dev;
 
-	dp->debug = dp_debug_get(dev, dp->panel, dp->usbpd,
+	dp->debug = dp_debug_get(dev, dp->panel,
 					dp->link, dp->dp_display.connector,
 					minor);
 	if (IS_ERR(dp->debug)) {
diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.c b/drivers/gpu/drm/msm/dp/dp_hpd.c
deleted file mode 100644
index db98a1d431eb6..0000000000000
--- a/drivers/gpu/drm/msm/dp/dp_hpd.c
+++ /dev/null
@@ -1,67 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
- */
-
-#define pr_fmt(fmt)	"[drm-dp] %s: " fmt, __func__
-
-#include <linux/slab.h>
-#include <linux/device.h>
-
-#include "dp_hpd.h"
-
-/* DP specific VDM commands */
-#define DP_USBPD_VDM_STATUS	0x10
-#define DP_USBPD_VDM_CONFIGURE	0x11
-
-/* USBPD-TypeC specific Macros */
-#define VDM_VERSION		0x0
-#define USB_C_DP_SID		0xFF01
-
-struct dp_hpd_private {
-	struct device *dev;
-	struct dp_usbpd_cb *dp_cb;
-	struct dp_usbpd dp_usbpd;
-};
-
-int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd)
-{
-	int rc = 0;
-	struct dp_hpd_private *hpd_priv;
-
-	hpd_priv = container_of(dp_usbpd, struct dp_hpd_private,
-					dp_usbpd);
-
-	if (!hpd_priv->dp_cb || !hpd_priv->dp_cb->configure
-				|| !hpd_priv->dp_cb->disconnect) {
-		pr_err("hpd dp_cb not initialized\n");
-		return -EINVAL;
-	}
-	if (hpd)
-		hpd_priv->dp_cb->configure(hpd_priv->dev);
-	else
-		hpd_priv->dp_cb->disconnect(hpd_priv->dev);
-
-	return rc;
-}
-
-struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb)
-{
-	struct dp_hpd_private *dp_hpd;
-
-	if (!cb) {
-		pr_err("invalid cb data\n");
-		return ERR_PTR(-EINVAL);
-	}
-
-	dp_hpd = devm_kzalloc(dev, sizeof(*dp_hpd), GFP_KERNEL);
-	if (!dp_hpd)
-		return ERR_PTR(-ENOMEM);
-
-	dp_hpd->dev = dev;
-	dp_hpd->dp_cb = cb;
-
-	dp_hpd->dp_usbpd.connect = dp_hpd_connect;
-
-	return &dp_hpd->dp_usbpd;
-}
diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.h b/drivers/gpu/drm/msm/dp/dp_hpd.h
deleted file mode 100644
index 8feec5aa50271..0000000000000
--- a/drivers/gpu/drm/msm/dp/dp_hpd.h
+++ /dev/null
@@ -1,78 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
- */
-
-#ifndef _DP_HPD_H_
-#define _DP_HPD_H_
-
-//#include <linux/usb/usbpd.h>
-
-#include <linux/types.h>
-#include <linux/device.h>
-
-enum plug_orientation {
-	ORIENTATION_NONE,
-	ORIENTATION_CC1,
-	ORIENTATION_CC2,
-};
-
-/**
- * struct dp_usbpd - DisplayPort status
- *
- * @orientation: plug orientation configuration
- * @low_pow_st: low power state
- * @adaptor_dp_en: adaptor functionality enabled
- * @multi_func: multi-function preferred
- * @usb_config_req: request to switch to usb
- * @exit_dp_mode: request exit from displayport mode
- * @hpd_irq: Change in the status since last message
- * @alt_mode_cfg_done: bool to specify alt mode status
- * @debug_en: bool to specify debug mode
- * @connect: simulate disconnect or connect for debug mode
- */
-struct dp_usbpd {
-	enum plug_orientation orientation;
-	bool low_pow_st;
-	bool adaptor_dp_en;
-	bool multi_func;
-	bool usb_config_req;
-	bool exit_dp_mode;
-	bool hpd_irq;
-	bool alt_mode_cfg_done;
-	bool debug_en;
-
-	int (*connect)(struct dp_usbpd *dp_usbpd, bool hpd);
-};
-
-/**
- * struct dp_usbpd_cb - callback functions provided by the client
- *
- * @configure: called by usbpd module when PD communication has
- * been completed and the usb peripheral has been configured on
- * dp mode.
- * @disconnect: notify the cable disconnect issued by usb.
- * @attention: notify any attention message issued by usb.
- */
-struct dp_usbpd_cb {
-	int (*configure)(struct device *dev);
-	int (*disconnect)(struct device *dev);
-	int (*attention)(struct device *dev);
-};
-
-/**
- * dp_hpd_get() - setup hpd module
- *
- * @dev: device instance of the caller
- * @cb: struct containing callback function pointers.
- *
- * This function allows the client to initialize the usbpd
- * module. The module will communicate with HPD module.
- */
-struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb);
-
-int dp_hpd_register(struct dp_usbpd *dp_usbpd);
-void dp_hpd_unregister(struct dp_usbpd *dp_usbpd);
-int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd);
-
-#endif /* _DP_HPD_H_ */
diff --git a/drivers/gpu/drm/msm/dp/dp_panel.h b/drivers/gpu/drm/msm/dp/dp_panel.h
index 45208b45eb535..ed1030e17e1b8 100644
--- a/drivers/gpu/drm/msm/dp/dp_panel.h
+++ b/drivers/gpu/drm/msm/dp/dp_panel.h
@@ -10,7 +10,6 @@
 
 #include "dp_aux.h"
 #include "dp_link.h"
-#include "dp_hpd.h"
 
 struct edid;
 
diff --git a/drivers/gpu/drm/msm/dp/dp_power.c b/drivers/gpu/drm/msm/dp/dp_power.c
index c0aaabb033894..6f88c344db99a 100644
--- a/drivers/gpu/drm/msm/dp/dp_power.c
+++ b/drivers/gpu/drm/msm/dp/dp_power.c
@@ -195,7 +195,7 @@ void dp_power_client_deinit(struct dp_power *dp_power)
 	pm_runtime_disable(&power->pdev->dev);
 }
 
-int dp_power_init(struct dp_power *dp_power, bool flip)
+int dp_power_init(struct dp_power *dp_power)
 {
 	int rc = 0;
 	struct dp_power_private *power = NULL;
diff --git a/drivers/gpu/drm/msm/dp/dp_power.h b/drivers/gpu/drm/msm/dp/dp_power.h
index e3f959ffae121..a3dec200785e5 100644
--- a/drivers/gpu/drm/msm/dp/dp_power.h
+++ b/drivers/gpu/drm/msm/dp/dp_power.h
@@ -26,13 +26,12 @@ struct dp_power {
  * dp_power_init() - enable power supplies for display controller
  *
  * @power: instance of power module
- * @flip: bool for flipping gpio direction
  * return: 0 if success or error if failure.
  *
  * This API will turn on the regulators and configures gpio's
  * aux/hpd.
  */
-int dp_power_init(struct dp_power *power, bool flip);
+int dp_power_init(struct dp_power *power);
 
 /**
  * dp_power_deinit() - turn off regulators and gpios.
-- 
GitLab


From 9f93258549dbc677898224b8c20d56d1eb96b66e Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <quic_bjorande@quicinc.com>
Date: Sun, 14 May 2023 20:02:55 -0700
Subject: [PATCH 0796/1791] drm/msm/dp: Clean up logs dp_power module

The clk_bulk API already provides error messages indicating which
specific clock in the request for which the operation failed, further
more these errors are associated with the specific DisplayPort
controller (rather than the shared drm_device). The additional error
messages int he dp_power module does thereby not provide any benefit.

While at it, none of the dp_power handles passed to these functions are
dynamic in nature, so there should not be any need for runtime checking
them. Drop these as well.

Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/536938/
Link: https://lore.kernel.org/r/20230515030256.300104-2-quic_bjorande@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dp/dp_power.c | 62 +++++--------------------------
 1 file changed, 9 insertions(+), 53 deletions(-)

diff --git a/drivers/gpu/drm/msm/dp/dp_power.c b/drivers/gpu/drm/msm/dp/dp_power.c
index 6f88c344db99a..cb7e1139fba57 100644
--- a/drivers/gpu/drm/msm/dp/dp_power.c
+++ b/drivers/gpu/drm/msm/dp/dp_power.c
@@ -35,25 +35,16 @@ static int dp_power_clk_init(struct dp_power_private *power)
 	stream = &power->parser->mp[DP_STREAM_PM];
 
 	rc = devm_clk_bulk_get(dev, core->num_clk, core->clocks);
-	if (rc) {
-		DRM_ERROR("failed to get %s clk. err=%d\n",
-			dp_parser_pm_name(DP_CORE_PM), rc);
+	if (rc)
 		return rc;
-	}
 
 	rc = devm_clk_bulk_get(dev, ctrl->num_clk, ctrl->clocks);
-	if (rc) {
-		DRM_ERROR("failed to get %s clk. err=%d\n",
-			dp_parser_pm_name(DP_CTRL_PM), rc);
+	if (rc)
 		return -ENODEV;
-	}
 
 	rc = devm_clk_bulk_get(dev, stream->num_clk, stream->clocks);
-	if (rc) {
-		DRM_ERROR("failed to get %s clk. err=%d\n",
-			dp_parser_pm_name(DP_CTRL_PM), rc);
+	if (rc)
 		return -ENODEV;
-	}
 
 	return 0;
 }
@@ -121,11 +112,9 @@ int dp_power_clk_enable(struct dp_power *dp_power,
 			mp = &power->parser->mp[DP_CORE_PM];
 
 			rc = clk_bulk_prepare_enable(mp->num_clk, mp->clocks);
-			if (rc) {
-				DRM_ERROR("fail to enable clks: %s. err=%d\n",
-					dp_parser_pm_name(DP_CORE_PM), rc);
+			if (rc)
 				return rc;
-			}
+
 			dp_power->core_clks_on = true;
 		}
 	}
@@ -133,10 +122,8 @@ int dp_power_clk_enable(struct dp_power *dp_power,
 	mp = &power->parser->mp[pm_type];
 	if (enable) {
 		rc = clk_bulk_prepare_enable(mp->num_clk, mp->clocks);
-		if (rc) {
-			DRM_ERROR("failed to enable clks, err: %d\n", rc);
+		if (rc)
 			return rc;
-		}
 	} else {
 		clk_bulk_disable_unprepare(mp->num_clk, mp->clocks);
 	}
@@ -162,34 +149,19 @@ int dp_power_clk_enable(struct dp_power *dp_power,
 
 int dp_power_client_init(struct dp_power *dp_power)
 {
-	int rc = 0;
 	struct dp_power_private *power;
 
-	if (!dp_power) {
-		DRM_ERROR("invalid power data\n");
-		return -EINVAL;
-	}
-
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
 	pm_runtime_enable(&power->pdev->dev);
 
-	rc = dp_power_clk_init(power);
-	if (rc)
-		DRM_ERROR("failed to init clocks %d\n", rc);
-
-	return rc;
+	return dp_power_clk_init(power);
 }
 
 void dp_power_client_deinit(struct dp_power *dp_power)
 {
 	struct dp_power_private *power;
 
-	if (!dp_power) {
-		DRM_ERROR("invalid power data\n");
-		return;
-	}
-
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
 	pm_runtime_disable(&power->pdev->dev);
@@ -200,25 +172,14 @@ int dp_power_init(struct dp_power *dp_power)
 	int rc = 0;
 	struct dp_power_private *power = NULL;
 
-	if (!dp_power) {
-		DRM_ERROR("invalid power data\n");
-		return -EINVAL;
-	}
-
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
 	pm_runtime_get_sync(&power->pdev->dev);
 
 	rc = dp_power_clk_enable(dp_power, DP_CORE_PM, true);
-	if (rc) {
-		DRM_ERROR("failed to enable DP core clocks, %d\n", rc);
-		goto exit;
-	}
-
-	return 0;
+	if (rc)
+		pm_runtime_put_sync(&power->pdev->dev);
 
-exit:
-	pm_runtime_put_sync(&power->pdev->dev);
 	return rc;
 }
 
@@ -238,11 +199,6 @@ struct dp_power *dp_power_get(struct device *dev, struct dp_parser *parser)
 	struct dp_power_private *power;
 	struct dp_power *dp_power;
 
-	if (!parser) {
-		DRM_ERROR("invalid input\n");
-		return ERR_PTR(-EINVAL);
-	}
-
 	power = devm_kzalloc(&parser->pdev->dev, sizeof(*power), GFP_KERNEL);
 	if (!power)
 		return ERR_PTR(-ENOMEM);
-- 
GitLab


From 9edac2eec47c82dcb679a285c6d9bb4b6f21a1a9 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <quic_bjorande@quicinc.com>
Date: Sun, 14 May 2023 20:02:56 -0700
Subject: [PATCH 0797/1791] drm/msm/dp: Clean up pdev/dev duplication in
 dp_power

The dp_power module keeps track of both the DP controller's struct
platform_device and struct device - with the prior pulled out of the
dp_parser module.

Clean up the duplication by dropping the platform_device reference and
just track the passed struct device.

Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/536939/
Link: https://lore.kernel.org/r/20230515030256.300104-3-quic_bjorande@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dp/dp_power.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/dp/dp_power.c b/drivers/gpu/drm/msm/dp/dp_power.c
index cb7e1139fba57..5cb84ca40e9ed 100644
--- a/drivers/gpu/drm/msm/dp/dp_power.c
+++ b/drivers/gpu/drm/msm/dp/dp_power.c
@@ -14,7 +14,6 @@
 
 struct dp_power_private {
 	struct dp_parser *parser;
-	struct platform_device *pdev;
 	struct device *dev;
 	struct drm_device *drm_dev;
 	struct clk *link_clk_src;
@@ -28,7 +27,7 @@ static int dp_power_clk_init(struct dp_power_private *power)
 {
 	int rc = 0;
 	struct dss_module_power *core, *ctrl, *stream;
-	struct device *dev = &power->pdev->dev;
+	struct device *dev = power->dev;
 
 	core = &power->parser->mp[DP_CORE_PM];
 	ctrl = &power->parser->mp[DP_CTRL_PM];
@@ -153,7 +152,7 @@ int dp_power_client_init(struct dp_power *dp_power)
 
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
-	pm_runtime_enable(&power->pdev->dev);
+	pm_runtime_enable(power->dev);
 
 	return dp_power_clk_init(power);
 }
@@ -164,7 +163,7 @@ void dp_power_client_deinit(struct dp_power *dp_power)
 
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
-	pm_runtime_disable(&power->pdev->dev);
+	pm_runtime_disable(power->dev);
 }
 
 int dp_power_init(struct dp_power *dp_power)
@@ -174,11 +173,11 @@ int dp_power_init(struct dp_power *dp_power)
 
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
-	pm_runtime_get_sync(&power->pdev->dev);
+	pm_runtime_get_sync(power->dev);
 
 	rc = dp_power_clk_enable(dp_power, DP_CORE_PM, true);
 	if (rc)
-		pm_runtime_put_sync(&power->pdev->dev);
+		pm_runtime_put_sync(power->dev);
 
 	return rc;
 }
@@ -190,7 +189,7 @@ int dp_power_deinit(struct dp_power *dp_power)
 	power = container_of(dp_power, struct dp_power_private, dp_power);
 
 	dp_power_clk_enable(dp_power, DP_CORE_PM, false);
-	pm_runtime_put_sync(&power->pdev->dev);
+	pm_runtime_put_sync(power->dev);
 	return 0;
 }
 
@@ -199,12 +198,11 @@ struct dp_power *dp_power_get(struct device *dev, struct dp_parser *parser)
 	struct dp_power_private *power;
 	struct dp_power *dp_power;
 
-	power = devm_kzalloc(&parser->pdev->dev, sizeof(*power), GFP_KERNEL);
+	power = devm_kzalloc(dev, sizeof(*power), GFP_KERNEL);
 	if (!power)
 		return ERR_PTR(-ENOMEM);
 
 	power->parser = parser;
-	power->pdev = parser->pdev;
 	power->dev = dev;
 
 	dp_power = &power->dp_power;
-- 
GitLab


From 76015b9e573eee41009ba8a5ff556ef02f83abfd Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Thu, 1 Jun 2023 12:09:47 +0200
Subject: [PATCH 0798/1791] dt-bindings: display: msm: dp-controller: document
 SM8550 compatible

The SM8550 & SM8350 SoC shares the same DP TX IP version, use the
SM8350 compatible as fallback for SM8550.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/540552/
Link: https://lore.kernel.org/r/20230601-topic-sm8550-upstream-dp-v2-1-e8778109c757@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 Documentation/devicetree/bindings/display/msm/dp-controller.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index f0c2237d5f82e..7a7cf3fb3e6da 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -29,6 +29,7 @@ properties:
       - items:
           - enum:
               - qcom,sm8450-dp
+              - qcom,sm8550-dp
           - const: qcom,sm8350-dp
 
   reg:
-- 
GitLab


From b29a20f7c4995a059ed764ce42389857426397c7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 30 May 2023 13:16:49 +0300
Subject: [PATCH 0799/1791] drm/i915: Use 18 fast wake AUX sync len
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

HW default for wake sync pulses is 18. 10 precharge and 8 preamble. There
is no reason to change this especially as it is causing problems with
certain eDP panels.

v3: Change "Fixes:" commit
v2: Remove "fast wake" repeat from subject

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Fixes: e1c71f8f9180 ("drm/i915: Fix fast wake AUX sync len")
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/8475
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230530101649.2549949-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_aux.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c
index 0c27db8ae4f18..197c6e81db147 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c
@@ -129,7 +129,7 @@ static int intel_dp_aux_sync_len(void)
 
 static int intel_dp_aux_fw_sync_len(void)
 {
-	int precharge = 16; /* 10-16 */
+	int precharge = 10; /* 10-16 */
 	int preamble = 8;
 
 	return precharge + preamble;
-- 
GitLab


From d922b80b1010cd6164fa7d3c197b4fbf94b47beb Mon Sep 17 00:00:00 2001
From: Tejas Upadhyay <tejas.upadhyay@intel.com>
Date: Thu, 1 Jun 2023 16:39:59 +0530
Subject: [PATCH 0800/1791] drm/i915/gt: Add workaround 14016712196

For mtl, workaround suggests that, SW insert a
dummy PIPE_CONTROL prior to PIPE_CONTROL which
contains a post sync: Timestamp or Write Immediate.

Bspec: 72197

V5:
  - Remove ret variable - Andi
V4:
  - Update commit message, avoid returing cs - Andi/Matt
V3:
  - Wrap dummy pipe control stuff in API - Andi
V2:
  - Fix  kernel test robot warnings

Closes: https://lore.kernel.org/oe-kbuild-all/202305121525.3EWdGoBY-lkp@intel.com/
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230601110959.1715927-1-tejas.upadhyay@intel.com
---
 drivers/gpu/drm/i915/gt/gen8_engine_cs.c | 38 ++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c
index e1c76e5bfa827..23857cc08eca1 100644
--- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c
@@ -177,14 +177,40 @@ u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv
 	return cs;
 }
 
+static int mtl_dummy_pipe_control(struct i915_request *rq)
+{
+	/* Wa_14016712196 */
+	if (IS_MTL_GRAPHICS_STEP(rq->engine->i915, M, STEP_A0, STEP_B0) ||
+	    IS_MTL_GRAPHICS_STEP(rq->engine->i915, P, STEP_A0, STEP_B0)) {
+		u32 *cs;
+
+		/* dummy PIPE_CONTROL + depth flush */
+		cs = intel_ring_begin(rq, 6);
+		if (IS_ERR(cs))
+			return PTR_ERR(cs);
+		cs = gen12_emit_pipe_control(cs,
+					     0,
+					     PIPE_CONTROL_DEPTH_CACHE_FLUSH,
+					     LRC_PPHWSP_SCRATCH_ADDR);
+		intel_ring_advance(rq, cs);
+	}
+
+	return 0;
+}
+
 int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode)
 {
 	struct intel_engine_cs *engine = rq->engine;
 
 	if (mode & EMIT_FLUSH) {
 		u32 flags = 0;
+		int err;
 		u32 *cs;
 
+		err = mtl_dummy_pipe_control(rq);
+		if (err)
+			return err;
+
 		flags |= PIPE_CONTROL_TILE_CACHE_FLUSH;
 		flags |= PIPE_CONTROL_FLUSH_L3;
 		flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
@@ -217,6 +243,11 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode)
 	if (mode & EMIT_INVALIDATE) {
 		u32 flags = 0;
 		u32 *cs, count;
+		int err;
+
+		err = mtl_dummy_pipe_control(rq);
+		if (err)
+			return err;
 
 		flags |= PIPE_CONTROL_COMMAND_CACHE_INVALIDATE;
 		flags |= PIPE_CONTROL_TLB_INVALIDATE;
@@ -733,6 +764,13 @@ u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs)
 		     PIPE_CONTROL_DC_FLUSH_ENABLE |
 		     PIPE_CONTROL_FLUSH_ENABLE);
 
+	/* Wa_14016712196 */
+	if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_A0, STEP_B0) ||
+	    IS_MTL_GRAPHICS_STEP(i915, P, STEP_A0, STEP_B0))
+		/* dummy PIPE_CONTROL + depth flush */
+		cs = gen12_emit_pipe_control(cs, 0,
+					     PIPE_CONTROL_DEPTH_CACHE_FLUSH, 0);
+
 	if (GRAPHICS_VER(i915) == 12 && GRAPHICS_VER_FULL(i915) < IP_VER(12, 50))
 		/* Wa_1409600907 */
 		flags |= PIPE_CONTROL_DEPTH_STALL;
-- 
GitLab


From 619a06dba6fa38de1b85c09ac74bb8aa2449ce0c Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Thu, 1 Jun 2023 13:13:14 +0300
Subject: [PATCH 0801/1791] drm/i915/mtl: Reset only one lane in case of MFD
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In case when only two or less transmit lanes are owned such as MFD
(DP-alt with x2 lanes) we need to reset only one data lane (lane0).
With only x2 lanes we don't need to poll for the phy current
status on both data lanes since only the owned data lane will respond.

v2: Find better naming for lanes and revise the commit message (Luca)

Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com> (v1)
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Luca Coelho <luciano.coelho@intel.com> (v2)
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230601101314.332392-1-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 39 ++++++++++++--------
 1 file changed, 23 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index ee6902118860e..0600fdcd06ef2 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2528,13 +2528,23 @@ static u32 intel_cx0_get_pclk_refclk_ack(u8 lane_mask)
 	return val;
 }
 
-/* FIXME: Some Type-C cases need not reset both the lanes. Handle those cases. */
-static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, enum port port,
+static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915,
+				     struct intel_encoder *encoder,
 				     bool lane_reversal)
 {
+	enum port port = encoder->port;
 	enum phy phy = intel_port_to_phy(i915, port);
+	bool both_lanes =  intel_tc_port_fia_max_lane_count(enc_to_dig_port(encoder)) > 2;
 	u8 lane_mask = lane_reversal ? INTEL_CX0_LANE1 :
 				  INTEL_CX0_LANE0;
+	u32 lane_pipe_reset = both_lanes ?
+			      XELPDP_LANE_PIPE_RESET(0) |
+			      XELPDP_LANE_PIPE_RESET(1) :
+			      XELPDP_LANE_PIPE_RESET(0);
+	u32 lane_phy_current_status = both_lanes ?
+				      XELPDP_LANE_PHY_CURRENT_STATUS(0) |
+				      XELPDP_LANE_PHY_CURRENT_STATUS(1) :
+				      XELPDP_LANE_PHY_CURRENT_STATUS(0);
 
 	if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL1(port),
 					 XELPDP_PORT_BUF_SOC_PHY_READY,
@@ -2545,23 +2555,24 @@ static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, enum port po
 
 	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
 		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1),
-		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1));
+		     lane_pipe_reset);
 
 	if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port),
-					 XELPDP_LANE_PHY_CURRENT_STATUS(0) |
-					 XELPDP_LANE_PHY_CURRENT_STATUS(1),
-					 XELPDP_LANE_PHY_CURRENT_STATUS(0) |
-					 XELPDP_LANE_PHY_CURRENT_STATUS(1),
+					 lane_phy_current_status, lane_phy_current_status,
 					 XELPDP_PORT_RESET_START_TIMEOUT_US, 0, NULL))
 		drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n",
 			 phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US);
 
 	intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(port),
-		     intel_cx0_get_pclk_refclk_request(INTEL_CX0_BOTH_LANES),
+		     intel_cx0_get_pclk_refclk_request(both_lanes ?
+						       INTEL_CX0_BOTH_LANES :
+						       INTEL_CX0_LANE0),
 		     intel_cx0_get_pclk_refclk_request(lane_mask));
 
 	if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(port),
-					 intel_cx0_get_pclk_refclk_ack(INTEL_CX0_BOTH_LANES),
+					 intel_cx0_get_pclk_refclk_ack(both_lanes ?
+								       INTEL_CX0_BOTH_LANES :
+								       INTEL_CX0_LANE0),
 					 intel_cx0_get_pclk_refclk_ack(lane_mask),
 					 XELPDP_REFCLK_ENABLE_TIMEOUT_US, 0, NULL))
 		drm_warn(&i915->drm, "PHY %c failed to request refclk after %dus.\n",
@@ -2571,13 +2582,9 @@ static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, enum port po
 					    CX0_P2_STATE_RESET);
 	intel_cx0_setup_powerdown(i915, port);
 
-	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port),
-		     XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1),
-		     0);
+	intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), lane_pipe_reset, 0);
 
-	if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(port),
-				    XELPDP_LANE_PHY_CURRENT_STATUS(0) |
-				    XELPDP_LANE_PHY_CURRENT_STATUS(1),
+	if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(port), lane_phy_current_status,
 				    XELPDP_PORT_RESET_END_TIMEOUT))
 		drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dms.\n",
 			 phy_name(phy), XELPDP_PORT_RESET_END_TIMEOUT);
@@ -2705,7 +2712,7 @@ static void intel_cx0pll_enable(struct intel_encoder *encoder,
 	intel_program_port_clock_ctl(encoder, crtc_state, lane_reversal);
 
 	/* 2. Bring PHY out of reset. */
-	intel_cx0_phy_lane_reset(i915, encoder->port, lane_reversal);
+	intel_cx0_phy_lane_reset(i915, encoder, lane_reversal);
 
 	/*
 	 * 3. Change Phy power state to Ready.
-- 
GitLab


From d95f87d29cf2097485765baac5109cecb486b5ee Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Tue, 11 Apr 2023 10:08:50 -0400
Subject: [PATCH 0802/1791] accel/habanalabs: remove variable gaudi_irq_name
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc with W=1 reports
drivers/accel/habanalabs/gaudi/gaudi.c:117:19: error:
  ‘gaudi_irq_name’ defined but not used [-Werror=unused-const-variable=]
  117 | static const char gaudi_irq_name[GAUDI_MSI_ENTRIES][GAUDI_MAX_STRING_LEN] = {
      |                   ^~~~~~~~~~~~~~

This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi/gaudi.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi/gaudi.c b/drivers/accel/habanalabs/gaudi/gaudi.c
index a29aa8f7b6f31..a1697581c2189 100644
--- a/drivers/accel/habanalabs/gaudi/gaudi.c
+++ b/drivers/accel/habanalabs/gaudi/gaudi.c
@@ -114,13 +114,6 @@ static u32 gaudi_stream_master[GAUDI_STREAM_MASTER_ARR_SIZE] = {
 	GAUDI_QUEUE_ID_DMA_1_3
 };
 
-static const char gaudi_irq_name[GAUDI_MSI_ENTRIES][GAUDI_MAX_STRING_LEN] = {
-		"gaudi cq 0_0", "gaudi cq 0_1", "gaudi cq 0_2", "gaudi cq 0_3",
-		"gaudi cq 1_0", "gaudi cq 1_1", "gaudi cq 1_2", "gaudi cq 1_3",
-		"gaudi cq 5_0", "gaudi cq 5_1", "gaudi cq 5_2", "gaudi cq 5_3",
-		"gaudi cpu eq"
-};
-
 static const u8 gaudi_dma_assignment[GAUDI_DMA_MAX] = {
 	[GAUDI_PCI_DMA_1] = GAUDI_ENGINE_ID_DMA_0,
 	[GAUDI_PCI_DMA_2] = GAUDI_ENGINE_ID_DMA_1,
-- 
GitLab


From 1464fbd8bab958b771af3214bd5cb52cd5f612ed Mon Sep 17 00:00:00 2001
From: Tal Cohen <talcohen@habana.ai>
Date: Thu, 30 Mar 2023 13:38:19 +0300
Subject: [PATCH 0803/1791] accel/habanalabs: ignore false positive razwi

In Gaudi2 asic, PSOC RAZWI may cause in HBW or LBW. The address that
caused the error is read from HW register and printed by the Driver.
There are cases where the Driver receives an indication on PSOC
RAZWI error but the address value is zero. In that case, the indication
is a false positive.
The Driver should not "count" a PSOC RAZWI event error when the
caused the address is zeroed.

Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 43 +++++++++++++++---------
 1 file changed, 27 insertions(+), 16 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index b778cf764a68a..6f05aa2303761 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8251,6 +8251,7 @@ static bool gaudi2_handle_psoc_razwi_happened(struct hl_device *hdev, u32 razwi_
 	u16 num_of_eng, eng_id[PSOC_RAZWI_MAX_ENG_PER_RTR];
 	char eng_name_str[PSOC_RAZWI_ENG_STR_SIZE];
 	bool razwi_happened = false;
+	u64 addr;
 	int i;
 
 	num_of_eng = gaudi2_psoc_razwi_get_engines(common_razwi_info, ARRAY_SIZE(common_razwi_info),
@@ -8269,43 +8270,53 @@ static bool gaudi2_handle_psoc_razwi_happened(struct hl_device *hdev, u32 razwi_
 		if (RREG32(base[i] + DEC_RAZWI_HBW_AW_SET)) {
 			addr_hi = RREG32(base[i] + DEC_RAZWI_HBW_AW_ADDR_HI);
 			addr_lo = RREG32(base[i] + DEC_RAZWI_HBW_AW_ADDR_LO);
-			dev_err(hdev->dev,
+			addr = ((u64)addr_hi << 32) + addr_lo;
+			if (addr) {
+				dev_err(hdev->dev,
 					"PSOC HBW AW RAZWI: %s, address (aligned to 128 byte): 0x%llX\n",
-					eng_name_str, ((u64)addr_hi << 32) + addr_lo);
-			hl_handle_razwi(hdev, ((u64)addr_hi << 32) + addr_lo, &eng_id[0],
+					eng_name_str, addr);
+				hl_handle_razwi(hdev, addr, &eng_id[0],
 					num_of_eng, HL_RAZWI_HBW | HL_RAZWI_WRITE, event_mask);
-			razwi_happened = true;
+				razwi_happened = true;
+			}
 		}
 
 		if (RREG32(base[i] + DEC_RAZWI_HBW_AR_SET)) {
 			addr_hi = RREG32(base[i] + DEC_RAZWI_HBW_AR_ADDR_HI);
 			addr_lo = RREG32(base[i] + DEC_RAZWI_HBW_AR_ADDR_LO);
-			dev_err(hdev->dev,
+			addr = ((u64)addr_hi << 32) + addr_lo;
+			if (addr) {
+				dev_err(hdev->dev,
 					"PSOC HBW AR RAZWI: %s, address (aligned to 128 byte): 0x%llX\n",
-					eng_name_str, ((u64)addr_hi << 32) + addr_lo);
-			hl_handle_razwi(hdev, ((u64)addr_hi << 32) + addr_lo, &eng_id[0],
+					eng_name_str, addr);
+				hl_handle_razwi(hdev, addr, &eng_id[0],
 					num_of_eng, HL_RAZWI_HBW | HL_RAZWI_READ, event_mask);
-			razwi_happened = true;
+				razwi_happened = true;
+			}
 		}
 
 		if (RREG32(base[i] + DEC_RAZWI_LBW_AW_SET)) {
 			addr_lo = RREG32(base[i] + DEC_RAZWI_LBW_AW_ADDR);
-			dev_err(hdev->dev,
+			if (addr_lo) {
+				dev_err(hdev->dev,
 					"PSOC LBW AW RAZWI: %s, address (aligned to 128 byte): 0x%X\n",
 					eng_name_str, addr_lo);
-			hl_handle_razwi(hdev, addr_lo, &eng_id[0],
+				hl_handle_razwi(hdev, addr_lo, &eng_id[0],
 					num_of_eng, HL_RAZWI_LBW | HL_RAZWI_WRITE, event_mask);
-			razwi_happened = true;
+				razwi_happened = true;
+			}
 		}
 
 		if (RREG32(base[i] + DEC_RAZWI_LBW_AR_SET)) {
 			addr_lo = RREG32(base[i] + DEC_RAZWI_LBW_AR_ADDR);
-			dev_err(hdev->dev,
-					"PSOC LBW AR RAZWI: %s, address (aligned to 128 byte): 0x%X\n",
-					eng_name_str, addr_lo);
-			hl_handle_razwi(hdev, addr_lo, &eng_id[0],
+			if (addr_lo) {
+				dev_err(hdev->dev,
+						"PSOC LBW AR RAZWI: %s, address (aligned to 128 byte): 0x%X\n",
+						eng_name_str, addr_lo);
+				hl_handle_razwi(hdev, addr_lo, &eng_id[0],
 					num_of_eng, HL_RAZWI_LBW | HL_RAZWI_READ, event_mask);
-			razwi_happened = true;
+				razwi_happened = true;
+			}
 		}
 		/* In common case the loop will break, when there is only one engine id, or
 		 * several engines with the same router. The exceptional case is with psoc razwi
-- 
GitLab


From 9ce36082c1723d96603c667436a88866b3fde531 Mon Sep 17 00:00:00 2001
From: Rakesh Ughreja <rughreja@habana.ai>
Date: Wed, 5 Apr 2023 09:12:29 +0300
Subject: [PATCH 0804/1791] accel/habanalabs: allow user to modify EDMA RL
 register

EDMA transpose workload requires to signal for every activation.
User FW sends all the dummy signals to RD_LBW_RATE_LIM_CFG, to save
lbw bandwidth. We need the user to be able to access that register to
configure it.

Signed-off-by: Rakesh Ughreja <rughreja@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2_security.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
index 694735f9e6e66..acd33130e7f98 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
@@ -479,6 +479,7 @@ static const u32 gaudi2_pb_dcr0_edma0_unsecured_regs[] = {
 	mmDCORE0_EDMA0_CORE_CTX_TE_NUMROWS,
 	mmDCORE0_EDMA0_CORE_CTX_IDX,
 	mmDCORE0_EDMA0_CORE_CTX_IDX_INC,
+	mmDCORE0_EDMA0_CORE_RD_LBW_RATE_LIM_CFG,
 	mmDCORE0_EDMA0_QM_CQ_CFG0_0,
 	mmDCORE0_EDMA0_QM_CQ_CFG0_1,
 	mmDCORE0_EDMA0_QM_CQ_CFG0_2,
-- 
GitLab


From 574ee40f514453d09ea2c5f3882b3b369b5ba027 Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Sun, 9 Apr 2023 14:24:05 +0300
Subject: [PATCH 0805/1791] accel/habanalabs: remove commented code that won't
 be used

Once it was decided that these security settings are to be done by FW
rather than by the driver, there's no reason to keep them in the code.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2_security.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
index acd33130e7f98..52a12c2bb58ef 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
@@ -3443,15 +3443,6 @@ static int gaudi2_init_protection_bits(struct hl_device *hdev)
 				ARRAY_SIZE(gaudi2_pb_thermal_sensor0), NULL, HL_PB_NA);
 	}
 
-	/* HBM */
-	/* Temporarily skip until SW-63348 is solved
-	 * instance_offset = mmHBM1_MC0_BASE - mmHBM0_MC0_BASE;
-	 * rc |= hl_init_pb_with_mask(hdev, HL_PB_SHARED, HL_PB_NA, GAUDI2_HBM_NUM,
-	 *		instance_offset, gaudi2_pb_hbm,
-	 *		ARRAY_SIZE(gaudi2_pb_hbm), NULL, HL_PB_NA,
-	 *		prop->dram_enabled_mask);
-	 */
-
 	/* Scheduler ARCs */
 	instance_offset = mmARC_FARM_ARC1_AUX_BASE - mmARC_FARM_ARC0_AUX_BASE;
 	rc |= hl_init_pb_ranges(hdev, HL_PB_SHARED, HL_PB_NA,
-- 
GitLab


From f9b60242af3e2216ed03139bc2c14a02c54073ce Mon Sep 17 00:00:00 2001
From: Moti Haimovski <mhaimovski@habana.ai>
Date: Mon, 10 Apr 2023 09:31:16 +0300
Subject: [PATCH 0806/1791] accel/habanalabs: fix bug in free scratchpad memory

This commit fixes a bug in Gaudi2 when freeing the scratchpad memory
in case software init fails.

Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 6f05aa2303761..acc304ebca823 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -3630,8 +3630,8 @@ static int gaudi2_sw_init(struct hl_device *hdev)
 special_blocks_free:
 	gaudi2_special_blocks_iterator_free(hdev);
 free_scratchpad_mem:
-	hl_asic_dma_pool_free(hdev, gaudi2->scratchpad_kernel_address,
-				gaudi2->scratchpad_bus_address);
+	hl_asic_dma_free_coherent(hdev, PAGE_SIZE, gaudi2->scratchpad_kernel_address,
+				  gaudi2->scratchpad_bus_address);
 free_virt_msix_db_mem:
 	hl_cpu_accessible_dma_pool_free(hdev, prop->pmmu.page_size, gaudi2->virt_msix_db_cpu_addr);
 free_cpu_accessible_dma_pool:
-- 
GitLab


From 9ef23f05aed4e1dd89b70cd2b8a5e35134ab90bb Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Wed, 22 Mar 2023 08:38:49 +0200
Subject: [PATCH 0807/1791] accel/habanalabs: add helper to extract the FW
 major/minor

the helper is extract_u32_until_given_char and can later be used to
also get the major/minor of the sw version.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c | 69 ++++++++++++-------
 1 file changed, 45 insertions(+), 24 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index 59f61ec66445a..1400a44300455 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -71,38 +71,59 @@ static char *extract_fw_ver_from_str(const char *fw_str)
 	return NULL;
 }
 
-static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
+/**
+ * extract_u32_until_given_char() - given a string of the format "<u32><char>*", extract the u32.
+ * @str: the given string
+ * @ver_num: the pointer to the extracted u32 to be returned to the caller.
+ * @given_char: the given char at the end of the u32 in the string
+ *
+ * Return: Upon success, return a pointer to the given_char in the string. Upon failure, return NULL
+ */
+static char *extract_u32_until_given_char(char *str, u32 *ver_num, char given_char)
 {
-	char major[8], minor[8], *first_dot, *second_dot;
-	int rc;
+	char num_str[8] = {}, *ch;
 
-	first_dot = strnstr(preboot_ver, ".", 10);
-	if (first_dot) {
-		strscpy(major, preboot_ver, first_dot - preboot_ver + 1);
-		rc = kstrtou32(major, 10, &hdev->fw_major_version);
-	} else {
-		rc = -EINVAL;
-	}
+	ch = strchrnul(str, given_char);
+	if (*ch == '\0' || ch == str || ch - str >= sizeof(num_str))
+		return NULL;
 
-	if (rc) {
-		dev_err(hdev->dev, "Error %d parsing preboot major version\n", rc);
-		return rc;
+	memcpy(num_str, str, ch - str);
+	if (kstrtou32(num_str, 10, ver_num))
+		return NULL;
+	return ch;
+}
+
+/**
+ * hl_get_preboot_major_minor() - extract the FW's version major, minor from the version string.
+ * @hdev: pointer to the hl_device
+ * @preboot_ver: the FW's version string
+ *
+ * preboot_ver is expected to be the format of <major>.<minor>.<sub minor>*, e.g: 42.0.1-sec-3
+ * The extracted version is set in the hdev fields: fw_inner_{major/minor}_ver.
+ *
+ * Return: 0 on success, negative error code for failure.
+ */
+static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
+{
+	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_major_version, '.');
+	if (!preboot_ver) {
+		dev_err(hdev->dev, "Error parsing preboot major version\n");
+		goto err_zero_ver;
 	}
 
-	/* skip the first dot */
-	first_dot++;
+	preboot_ver++;
 
-	second_dot = strnstr(first_dot, ".", 10);
-	if (second_dot) {
-		strscpy(minor, first_dot, second_dot - first_dot + 1);
-		rc = kstrtou32(minor, 10, &hdev->fw_minor_version);
-	} else {
-		rc = -EINVAL;
+	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_minor_version, '.');
+	if (!preboot_ver) {
+		dev_err(hdev->dev, "Error parsing preboot minor version\n");
+		goto err_zero_ver;
 	}
+	return 0;
 
-	if (rc)
-		dev_err(hdev->dev, "Error %d parsing preboot minor version\n", rc);
-	return rc;
+err_zero_ver:
+	hdev->fw_major_version = 0;
+	hdev->fw_minor_version = 0;
+	return -EINVAL;
 }
 
 static int hl_request_fw(struct hl_device *hdev,
-- 
GitLab


From 3071247ca06195eed4261b99035e0c031fa5ddd0 Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Sun, 2 Apr 2023 12:05:43 +0300
Subject: [PATCH 0808/1791] accel/habanalabs: rename fw_{major/minor}_version
 to fw_inner_{major/minor}_ver

We later want to add fields for Firmware SW version. The current
extracted FW version is the inner FW versioning so the new name
is better and also better differentiate from the FW's SW version.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c |  8 ++++----
 drivers/accel/habanalabs/common/habanalabs.h  | 10 +++++-----
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index 1400a44300455..6150ab6ba8104 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -105,7 +105,7 @@ static char *extract_u32_until_given_char(char *str, u32 *ver_num, char given_ch
  */
 static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
 {
-	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_major_version, '.');
+	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_inner_major_ver, '.');
 	if (!preboot_ver) {
 		dev_err(hdev->dev, "Error parsing preboot major version\n");
 		goto err_zero_ver;
@@ -113,7 +113,7 @@ static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
 
 	preboot_ver++;
 
-	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_minor_version, '.');
+	preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_inner_minor_ver, '.');
 	if (!preboot_ver) {
 		dev_err(hdev->dev, "Error parsing preboot minor version\n");
 		goto err_zero_ver;
@@ -121,8 +121,8 @@ static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver)
 	return 0;
 
 err_zero_ver:
-	hdev->fw_major_version = 0;
-	hdev->fw_minor_version = 0;
+	hdev->fw_inner_major_ver = 0;
+	hdev->fw_inner_minor_ver = 0;
 	return -EINVAL;
 }
 
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index eaae69a9f8178..57661cb516218 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3225,8 +3225,8 @@ struct hl_reset_info {
  * @captured_err_info: holds information about errors.
  * @reset_info: holds current device reset information.
  * @stream_master_qid_arr: pointer to array with QIDs of master streams.
- * @fw_major_version: major version of current loaded preboot.
- * @fw_minor_version: minor version of current loaded preboot.
+ * @fw_inner_major_ver: the major of current loaded preboot inner version.
+ * @fw_inner_minor_ver: the minor of current loaded preboot inner version.
  * @dram_used_mem: current DRAM memory consumption.
  * @memory_scrub_val: the value to which the dram will be scrubbed to using cb scrub_device_dram
  * @timeout_jiffies: device CS timeout value.
@@ -3412,8 +3412,8 @@ struct hl_device {
 	struct hl_reset_info		reset_info;
 
 	u32				*stream_master_qid_arr;
-	u32				fw_major_version;
-	u32				fw_minor_version;
+	u32				fw_inner_major_ver;
+	u32				fw_inner_minor_ver;
 	atomic64_t			dram_used_mem;
 	u64				memory_scrub_val;
 	u64				timeout_jiffies;
@@ -3549,7 +3549,7 @@ struct hl_ioctl_desc {
 
 static inline bool hl_is_fw_ver_below_1_9(struct hl_device *hdev)
 {
-	return (hdev->fw_major_version < 42);
+	return (hdev->fw_inner_major_ver < 42);
 }
 
 /*
-- 
GitLab


From dd5667ff6f9c52c6a49aa9725b80542793613036 Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Thu, 16 Mar 2023 10:20:44 +0200
Subject: [PATCH 0809/1791] accel/habanalabs: extract and save the FW's SW
 major/minor/sub-minor

It is not always possible to know the FW's SW version from the inner FW
version. Therefore we should extract the general SW version in addition
to the FW version and use it in functions like
'hl_is_fw_ver_below_1_9' etc.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c | 78 +++++++++++++++++--
 drivers/accel/habanalabs/common/habanalabs.h  |  6 ++
 2 files changed, 78 insertions(+), 6 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index 6150ab6ba8104..62052cfe694ff 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -93,6 +93,71 @@ static char *extract_u32_until_given_char(char *str, u32 *ver_num, char given_ch
 	return ch;
 }
 
+/**
+ * hl_get_sw_major_minor_subminor() - extract the FW's SW version major, minor, sub-minor
+ *				      from the version string
+ * @hdev: pointer to the hl_device
+ * @fw_str: the FW's version string
+ *
+ * The extracted version is set in the hdev fields: fw_sw_{major/minor/sub_minor}_ver.
+ *
+ * fw_str is expected to have one of two possible formats, examples:
+ * 1) 'Preboot version hl-gaudi2-1.9.0-fw-42.0.1-sec-3'
+ * 2) 'Preboot version hl-gaudi2-1.9.0-rc-fw-42.0.1-sec-3'
+ * In those examples, the SW major,minor,subminor are correspondingly: 1,9,0.
+ *
+ * Return: 0 for success or a negative error code for failure.
+ */
+static int hl_get_sw_major_minor_subminor(struct hl_device *hdev, const char *fw_str)
+{
+	char *end, *start;
+
+	end = strnstr(fw_str, "-rc-", VERSION_MAX_LEN);
+	if (end == fw_str)
+		return -EINVAL;
+
+	if (!end)
+		end = strnstr(fw_str, "-fw-", VERSION_MAX_LEN);
+
+	if (end == fw_str)
+		return -EINVAL;
+
+	if (!end)
+		return -EINVAL;
+
+	for (start = end - 1; start != fw_str; start--) {
+		if (*start == '-')
+			break;
+	}
+
+	if (start == fw_str)
+		return -EINVAL;
+
+	/* start/end point each to the starting and ending hyphen of the sw version e.g. -1.9.0- */
+	start++;
+	start = extract_u32_until_given_char(start, &hdev->fw_sw_major_ver, '.');
+	if (!start)
+		goto err_zero_ver;
+
+	start++;
+	start = extract_u32_until_given_char(start, &hdev->fw_sw_minor_ver, '.');
+	if (!start)
+		goto err_zero_ver;
+
+	start++;
+	start = extract_u32_until_given_char(start, &hdev->fw_sw_sub_minor_ver, '-');
+	if (!start)
+		goto err_zero_ver;
+
+	return 0;
+
+err_zero_ver:
+	hdev->fw_sw_major_ver = 0;
+	hdev->fw_sw_minor_ver = 0;
+	hdev->fw_sw_sub_minor_ver = 0;
+	return -EINVAL;
+}
+
 /**
  * hl_get_preboot_major_minor() - extract the FW's version major, minor from the version string.
  * @hdev: pointer to the hl_device
@@ -2172,6 +2237,7 @@ static int hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev,
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
 	char *preboot_ver, *boot_ver;
 	char btl_ver[32];
+	int rc;
 
 	switch (fwc) {
 	case FW_COMP_BOOT_FIT:
@@ -2185,20 +2251,20 @@ static int hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev,
 		break;
 	case FW_COMP_PREBOOT:
 		strscpy(prop->preboot_ver, fw_version, VERSION_MAX_LEN);
-		preboot_ver = strnstr(prop->preboot_ver, "Preboot",
-						VERSION_MAX_LEN);
+		preboot_ver = strnstr(prop->preboot_ver, "Preboot", VERSION_MAX_LEN);
+		dev_info(hdev->dev, "preboot full version: '%s'\n", preboot_ver);
+
 		if (preboot_ver && preboot_ver != prop->preboot_ver) {
 			strscpy(btl_ver, prop->preboot_ver,
 				min((int) (preboot_ver - prop->preboot_ver), 31));
 			dev_info(hdev->dev, "%s\n", btl_ver);
 		}
 
+		rc = hl_get_sw_major_minor_subminor(hdev, preboot_ver);
+		if (rc)
+			return rc;
 		preboot_ver = extract_fw_ver_from_str(prop->preboot_ver);
 		if (preboot_ver) {
-			int rc;
-
-			dev_info(hdev->dev, "preboot version %s\n", preboot_ver);
-
 			rc = hl_get_preboot_major_minor(hdev, preboot_ver);
 			kfree(preboot_ver);
 			if (rc)
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 57661cb516218..d6f454b1cde48 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3227,6 +3227,9 @@ struct hl_reset_info {
  * @stream_master_qid_arr: pointer to array with QIDs of master streams.
  * @fw_inner_major_ver: the major of current loaded preboot inner version.
  * @fw_inner_minor_ver: the minor of current loaded preboot inner version.
+ * @fw_sw_major_ver: the major of current loaded preboot SW version.
+ * @fw_sw_minor_ver: the minor of current loaded preboot SW version.
+ * @fw_sw_sub_minor_ver: the sub-minor of current loaded preboot SW version.
  * @dram_used_mem: current DRAM memory consumption.
  * @memory_scrub_val: the value to which the dram will be scrubbed to using cb scrub_device_dram
  * @timeout_jiffies: device CS timeout value.
@@ -3414,6 +3417,9 @@ struct hl_device {
 	u32				*stream_master_qid_arr;
 	u32				fw_inner_major_ver;
 	u32				fw_inner_minor_ver;
+	u32				fw_sw_major_ver;
+	u32				fw_sw_minor_ver;
+	u32				fw_sw_sub_minor_ver;
 	atomic64_t			dram_used_mem;
 	u64				memory_scrub_val;
 	u64				timeout_jiffies;
-- 
GitLab


From a12428acf823a1a367d38a4113b7365e3e5603bf Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Tue, 18 Apr 2023 11:39:44 +0300
Subject: [PATCH 0810/1791] accel/habanalabs: check fw version using sw version

The fw inner version is less trustable, instead use the fw general
sw release version.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/habanalabs.h | 10 ++++++++--
 drivers/accel/habanalabs/gaudi2/gaudi2.c     |  2 +-
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index d6f454b1cde48..02620654ccdfa 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3553,9 +3553,15 @@ struct hl_ioctl_desc {
 	hl_ioctl_t *func;
 };
 
-static inline bool hl_is_fw_ver_below_1_9(struct hl_device *hdev)
+static inline bool hl_is_fw_sw_ver_below(struct hl_device *hdev, u32 fw_sw_major, u32 fw_sw_minor)
 {
-	return (hdev->fw_inner_major_ver < 42);
+	if (hdev->fw_sw_major_ver < fw_sw_major)
+		return true;
+	if (hdev->fw_sw_major_ver > fw_sw_major)
+		return false;
+	if (hdev->fw_sw_minor_ver < fw_sw_minor)
+		return true;
+	return false;
 }
 
 /*
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index acc304ebca823..6e3aa89b19c29 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8043,7 +8043,7 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev,
 	case RAZWI_TPC:
 		hbw_rtr_id = gaudi2_tpc_initiator_hbw_rtr_id[module_idx];
 
-		if (hl_is_fw_ver_below_1_9(hdev) &&
+		if (hl_is_fw_sw_ver_below(hdev, 1, 9) &&
 				!hdev->asic_prop.fw_security_enabled &&
 				((module_idx == 0) || (module_idx == 1)))
 			lbw_rtr_id = DCORE0_RTR0;
-- 
GitLab


From cc7b790d412461520de49eb321a0aeed2735e5c4 Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Wed, 8 Feb 2023 16:16:08 +0200
Subject: [PATCH 0811/1791] accel/habanalabs: do soft-reset using cpucp packet

This is done depending on the FW version. The cpucp method is
preferable and saves scratchpads resource.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c | 14 ++++++++++
 drivers/accel/habanalabs/common/habanalabs.h  |  1 +
 drivers/accel/habanalabs/gaudi2/gaudi2.c      | 26 ++++++++++++-------
 .../habanalabs/include/common/cpucp_if.h      |  4 +++
 4 files changed, 35 insertions(+), 10 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index 62052cfe694ff..e48f024d86498 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -591,6 +591,20 @@ void hl_fw_cpu_accessible_dma_pool_free(struct hl_device *hdev, size_t size,
 			size);
 }
 
+int hl_fw_send_soft_reset(struct hl_device *hdev)
+{
+	struct cpucp_packet pkt;
+	int rc;
+
+	memset(&pkt, 0, sizeof(pkt));
+	pkt.ctl = cpu_to_le32(CPUCP_PACKET_SOFT_RESET << CPUCP_PKT_CTL_OPCODE_SHIFT);
+	rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, NULL);
+	if (rc)
+		dev_err(hdev->dev, "failed to send soft-reset msg (err = %d)\n", rc);
+
+	return rc;
+}
+
 int hl_fw_send_device_activity(struct hl_device *hdev, bool open)
 {
 	struct cpucp_packet pkt;
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 02620654ccdfa..6b102947cb90a 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3884,6 +3884,7 @@ int hl_fw_dram_replaced_row_get(struct hl_device *hdev,
 int hl_fw_dram_pending_row_get(struct hl_device *hdev, u32 *pend_rows_num);
 int hl_fw_cpucp_engine_core_asid_set(struct hl_device *hdev, u32 asid);
 int hl_fw_send_device_activity(struct hl_device *hdev, bool open);
+int hl_fw_send_soft_reset(struct hl_device *hdev);
 int hl_pci_bars_map(struct hl_device *hdev, const char * const name[3],
 			bool is_wc[3]);
 int hl_pci_elbi_read(struct hl_device *hdev, u64 addr, u32 *data);
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 6e3aa89b19c29..5c80e7af5b785 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -6148,18 +6148,24 @@ static int gaudi2_execute_soft_reset(struct hl_device *hdev, bool driver_perform
 						u32 poll_timeout_us)
 {
 	struct cpu_dyn_regs *dyn_regs = &hdev->fw_loader.dynamic_loader.comm_desc.cpu_dyn_regs;
+	int rc = 0;
 
 	if (!driver_performs_reset) {
-		/* set SP to indicate reset request sent to FW */
-		if (dyn_regs->cpu_rst_status)
-			WREG32(le32_to_cpu(dyn_regs->cpu_rst_status), CPU_RST_STATUS_NA);
-		else
-			WREG32(mmCPU_RST_STATUS_TO_HOST, CPU_RST_STATUS_NA);
-
-		WREG32(le32_to_cpu(dyn_regs->gic_host_soft_rst_irq),
-			gaudi2_irq_map_table[GAUDI2_EVENT_CPU_SOFT_RESET].cpu_id);
-
-		return gaudi2_get_soft_rst_done_indication(hdev, poll_timeout_us);
+		if (hl_is_fw_sw_ver_below(hdev, 1, 10)) {
+			/* set SP to indicate reset request sent to FW */
+			if (dyn_regs->cpu_rst_status)
+				WREG32(le32_to_cpu(dyn_regs->cpu_rst_status), CPU_RST_STATUS_NA);
+			else
+				WREG32(mmCPU_RST_STATUS_TO_HOST, CPU_RST_STATUS_NA);
+			WREG32(le32_to_cpu(dyn_regs->gic_host_soft_rst_irq),
+				gaudi2_irq_map_table[GAUDI2_EVENT_CPU_SOFT_RESET].cpu_id);
+
+			/* wait for f/w response */
+			rc = gaudi2_get_soft_rst_done_indication(hdev, poll_timeout_us);
+		} else {
+			rc = hl_fw_send_soft_reset(hdev);
+		}
+		return rc;
 	}
 
 	/* Block access to engines, QMANs and SM during reset, these
diff --git a/drivers/accel/habanalabs/include/common/cpucp_if.h b/drivers/accel/habanalabs/include/common/cpucp_if.h
index 8bbe685458c49..f68308cc25240 100644
--- a/drivers/accel/habanalabs/include/common/cpucp_if.h
+++ b/drivers/accel/habanalabs/include/common/cpucp_if.h
@@ -665,6 +665,9 @@ enum pq_init_status {
  *
  * CPUCP_PACKET_REGISTER_INTERRUPTS -
  *       Packet to register interrupts indicating LKD is ready to receive events from FW.
+ *
+ * CPUCP_PACKET_SOFT_RESET -
+ *	 Packet to perform soft-reset.
  */
 
 enum cpucp_packet_id {
@@ -731,6 +734,7 @@ enum cpucp_packet_id {
 	CPUCP_PACKET_RESERVED11,		/* not used */
 	CPUCP_PACKET_RESERVED12,		/* internal */
 	CPUCP_PACKET_REGISTER_INTERRUPTS,	/* internal */
+	CPUCP_PACKET_SOFT_RESET,		/* internal */
 	CPUCP_PACKET_ID_MAX			/* must be last */
 };
 
-- 
GitLab


From 57469c12060899b7b0768ace178c7e36501e2f21 Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Thu, 13 Apr 2023 11:22:06 +0300
Subject: [PATCH 0812/1791] accel/habanalabs: unsecure TPC bias registers

User needs to be able to perform downcast / upcast of fp8_143 dtype.
Hence bias register needs to be accessed by the user.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2_security.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
index 52a12c2bb58ef..fadb870ff4c03 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
@@ -1542,6 +1542,7 @@ static const u32 gaudi2_pb_dcr0_tpc0_unsecured_regs[] = {
 	mmDCORE0_TPC0_CFG_LUT_FUNC128_BASE2_ADDR_HI,
 	mmDCORE0_TPC0_CFG_LUT_FUNC256_BASE2_ADDR_LO,
 	mmDCORE0_TPC0_CFG_LUT_FUNC256_BASE2_ADDR_HI,
+	mmDCORE0_TPC0_CFG_FP8_143_BIAS,
 	mmDCORE0_TPC0_CFG_ROUND_CSR,
 	mmDCORE0_TPC0_CFG_CONV_ROUND_CSR,
 	mmDCORE0_TPC0_CFG_SEMAPHORE,
-- 
GitLab


From 04729b418f8b031712e6dbf5bffa7d639f8f211e Mon Sep 17 00:00:00 2001
From: Moti Haimovski <mhaimovski@habana.ai>
Date: Thu, 13 Apr 2023 13:54:40 +0300
Subject: [PATCH 0813/1791] accel/habanalabs: call to HW/FW err returns 0 when
 no events exist

This commit modifies the call to retrieve HW or FW error events to
return success when no events are pending, as done in the calls to
other events.

Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../habanalabs/common/habanalabs_ioctl.c      | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
index 203ee857810c7..4368e6c9a23af 100644
--- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
@@ -842,15 +842,15 @@ static int hw_err_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 	struct hw_err_info *info;
 	int rc;
 
-	if ((!user_buf_size) || (!user_buf))
+	if (!user_buf)
 		return -EINVAL;
 
-	if (user_buf_size < sizeof(struct hl_info_hw_err_event))
-		return -ENOMEM;
-
 	info = &hdev->captured_err_info.hw_err;
 	if (!info->event_info_available)
-		return -ENOENT;
+		return 0;
+
+	if (user_buf_size < sizeof(struct hl_info_hw_err_event))
+		return -ENOMEM;
 
 	rc = copy_to_user(user_buf, &info->event, sizeof(struct hl_info_hw_err_event));
 	return rc ? -EFAULT : 0;
@@ -864,15 +864,15 @@ static int fw_err_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 	struct fw_err_info *info;
 	int rc;
 
-	if ((!user_buf_size) || (!user_buf))
+	if (!user_buf)
 		return -EINVAL;
 
-	if (user_buf_size < sizeof(struct hl_info_fw_err_event))
-		return -ENOMEM;
-
 	info = &hdev->captured_err_info.fw_err;
 	if (!info->event_info_available)
-		return -ENOENT;
+		return 0;
+
+	if (user_buf_size < sizeof(struct hl_info_fw_err_event))
+		return -ENOMEM;
 
 	rc = copy_to_user(user_buf, &info->event, sizeof(struct hl_info_fw_err_event));
 	return rc ? -EFAULT : 0;
-- 
GitLab


From a272cadbd8834c06cc2e34213202cc6be2c8fc19 Mon Sep 17 00:00:00 2001
From: Masahiro Yamada <masahiroy@kernel.org>
Date: Mon, 5 Jun 2023 21:00:20 +0900
Subject: [PATCH 0814/1791] drm/bridge: imx: fix mixed module-builtin object

With CONFIG_DRM_IMX8QM_LDB=m and CONFIG_DRM_IMX8QXP_LDB=y (or vice
versa), imx-ldb-helper.o is linked to a module and also to vmlinux
even though the expected CFLAGS are different between builtins and
modules.

This is the same situation as fixed by commit 637a642f5ca5 ("zstd:
Fixing mixed module-builtin objects").

Split imx-ldb-helper.c into a separate module.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230605120021.1774711-1-masahiroy@kernel.org
---
 drivers/gpu/drm/bridge/imx/Kconfig          |  5 +++++
 drivers/gpu/drm/bridge/imx/Makefile         |  5 +++--
 drivers/gpu/drm/bridge/imx/imx-ldb-helper.c | 17 +++++++++++++++++
 3 files changed, 25 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/imx/Kconfig b/drivers/gpu/drm/bridge/imx/Kconfig
index 608f47f41bcd1..9fae28db6aa7f 100644
--- a/drivers/gpu/drm/bridge/imx/Kconfig
+++ b/drivers/gpu/drm/bridge/imx/Kconfig
@@ -1,9 +1,13 @@
 if ARCH_MXC || COMPILE_TEST
 
+config DRM_IMX_LDB_HELPER
+	tristate
+
 config DRM_IMX8QM_LDB
 	tristate "Freescale i.MX8QM LVDS display bridge"
 	depends on OF
 	depends on COMMON_CLK
+	select DRM_IMX_LDB_HELPER
 	select DRM_KMS_HELPER
 	help
 	  Choose this to enable the internal LVDS Display Bridge(LDB) found in
@@ -13,6 +17,7 @@ config DRM_IMX8QXP_LDB
 	tristate "Freescale i.MX8QXP LVDS display bridge"
 	depends on OF
 	depends on COMMON_CLK
+	select DRM_IMX_LDB_HELPER
 	select DRM_KMS_HELPER
 	help
 	  Choose this to enable the internal LVDS Display Bridge(LDB) found in
diff --git a/drivers/gpu/drm/bridge/imx/Makefile b/drivers/gpu/drm/bridge/imx/Makefile
index aa90ec8d54336..5fc821278693c 100644
--- a/drivers/gpu/drm/bridge/imx/Makefile
+++ b/drivers/gpu/drm/bridge/imx/Makefile
@@ -1,7 +1,8 @@
-imx8qm-ldb-objs := imx-ldb-helper.o imx8qm-ldb-drv.o
+obj-$(CONFIG_DRM_IMX_LDB_HELPER) += imx-ldb-helper.o
+imx8qm-ldb-objs := imx8qm-ldb-drv.o
 obj-$(CONFIG_DRM_IMX8QM_LDB) += imx8qm-ldb.o
 
-imx8qxp-ldb-objs := imx-ldb-helper.o imx8qxp-ldb-drv.o
+imx8qxp-ldb-objs := imx8qxp-ldb-drv.o
 obj-$(CONFIG_DRM_IMX8QXP_LDB) += imx8qxp-ldb.o
 
 obj-$(CONFIG_DRM_IMX8QXP_PIXEL_COMBINER) += imx8qxp-pixel-combiner.o
diff --git a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
index 7338b84bc83d2..6967325cd8ee0 100644
--- a/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
+++ b/drivers/gpu/drm/bridge/imx/imx-ldb-helper.c
@@ -4,8 +4,10 @@
  * Copyright 2019,2020,2022 NXP
  */
 
+#include <linux/export.h>
 #include <linux/media-bus-format.h>
 #include <linux/mfd/syscon.h>
+#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/regmap.h>
 
@@ -19,12 +21,14 @@ bool ldb_channel_is_single_link(struct ldb_channel *ldb_ch)
 {
 	return ldb_ch->link_type == LDB_CH_SINGLE_LINK;
 }
+EXPORT_SYMBOL_GPL(ldb_channel_is_single_link);
 
 bool ldb_channel_is_split_link(struct ldb_channel *ldb_ch)
 {
 	return ldb_ch->link_type == LDB_CH_DUAL_LINK_EVEN_ODD_PIXELS ||
 	       ldb_ch->link_type == LDB_CH_DUAL_LINK_ODD_EVEN_PIXELS;
 }
+EXPORT_SYMBOL_GPL(ldb_channel_is_split_link);
 
 int ldb_bridge_atomic_check_helper(struct drm_bridge *bridge,
 				   struct drm_bridge_state *bridge_state,
@@ -38,6 +42,7 @@ int ldb_bridge_atomic_check_helper(struct drm_bridge *bridge,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(ldb_bridge_atomic_check_helper);
 
 void ldb_bridge_mode_set_helper(struct drm_bridge *bridge,
 				const struct drm_display_mode *mode,
@@ -69,6 +74,7 @@ void ldb_bridge_mode_set_helper(struct drm_bridge *bridge,
 		break;
 	}
 }
+EXPORT_SYMBOL_GPL(ldb_bridge_mode_set_helper);
 
 void ldb_bridge_enable_helper(struct drm_bridge *bridge)
 {
@@ -81,6 +87,7 @@ void ldb_bridge_enable_helper(struct drm_bridge *bridge)
 	 */
 	regmap_write(ldb->regmap, ldb->ctrl_reg, ldb->ldb_ctrl);
 }
+EXPORT_SYMBOL_GPL(ldb_bridge_enable_helper);
 
 void ldb_bridge_disable_helper(struct drm_bridge *bridge)
 {
@@ -95,6 +102,7 @@ void ldb_bridge_disable_helper(struct drm_bridge *bridge)
 
 	regmap_write(ldb->regmap, ldb->ctrl_reg, ldb->ldb_ctrl);
 }
+EXPORT_SYMBOL_GPL(ldb_bridge_disable_helper);
 
 int ldb_bridge_attach_helper(struct drm_bridge *bridge,
 			     enum drm_bridge_attach_flags flags)
@@ -117,6 +125,7 @@ int ldb_bridge_attach_helper(struct drm_bridge *bridge,
 				ldb_ch->next_bridge, bridge,
 				DRM_BRIDGE_ATTACH_NO_CONNECTOR);
 }
+EXPORT_SYMBOL_GPL(ldb_bridge_attach_helper);
 
 int ldb_init_helper(struct ldb *ldb)
 {
@@ -157,6 +166,7 @@ int ldb_init_helper(struct ldb *ldb)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(ldb_init_helper);
 
 int ldb_find_next_bridge_helper(struct ldb *ldb)
 {
@@ -184,6 +194,7 @@ int ldb_find_next_bridge_helper(struct ldb *ldb)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(ldb_find_next_bridge_helper);
 
 void ldb_add_bridge_helper(struct ldb *ldb,
 			   const struct drm_bridge_funcs *bridge_funcs)
@@ -204,6 +215,7 @@ void ldb_add_bridge_helper(struct ldb *ldb,
 		drm_bridge_add(&ldb_ch->bridge);
 	}
 }
+EXPORT_SYMBOL_GPL(ldb_add_bridge_helper);
 
 void ldb_remove_bridge_helper(struct ldb *ldb)
 {
@@ -219,3 +231,8 @@ void ldb_remove_bridge_helper(struct ldb *ldb)
 		drm_bridge_remove(&ldb_ch->bridge);
 	}
 }
+EXPORT_SYMBOL_GPL(ldb_remove_bridge_helper);
+
+MODULE_DESCRIPTION("i.MX8 LVDS Display Bridge(LDB)/Pixel Mapper bridge helper");
+MODULE_AUTHOR("Liu Ying <victor.liu@nxp.com>");
+MODULE_LICENSE("GPL");
-- 
GitLab


From 013413cdfeec53e06c43a239fb4e06a39ffedce3 Mon Sep 17 00:00:00 2001
From: Masahiro Yamada <masahiroy@kernel.org>
Date: Mon, 5 Jun 2023 21:00:21 +0900
Subject: [PATCH 0815/1791] drm/bridge: imx: turn imx8{qm,qxp}-ldb into
 single-object modules

With the previous fix, these modules are built from a single C file.

Rename the source files so they match the module names.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Robert Foss <rfoss@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230605120021.1774711-2-masahiroy@kernel.org
---
 drivers/gpu/drm/bridge/imx/Makefile                           | 4 ----
 drivers/gpu/drm/bridge/imx/{imx8qm-ldb-drv.c => imx8qm-ldb.c} | 0
 .../gpu/drm/bridge/imx/{imx8qxp-ldb-drv.c => imx8qxp-ldb.c}   | 0
 3 files changed, 4 deletions(-)
 rename drivers/gpu/drm/bridge/imx/{imx8qm-ldb-drv.c => imx8qm-ldb.c} (100%)
 rename drivers/gpu/drm/bridge/imx/{imx8qxp-ldb-drv.c => imx8qxp-ldb.c} (100%)

diff --git a/drivers/gpu/drm/bridge/imx/Makefile b/drivers/gpu/drm/bridge/imx/Makefile
index 5fc821278693c..8e2ebf3399a15 100644
--- a/drivers/gpu/drm/bridge/imx/Makefile
+++ b/drivers/gpu/drm/bridge/imx/Makefile
@@ -1,10 +1,6 @@
 obj-$(CONFIG_DRM_IMX_LDB_HELPER) += imx-ldb-helper.o
-imx8qm-ldb-objs := imx8qm-ldb-drv.o
 obj-$(CONFIG_DRM_IMX8QM_LDB) += imx8qm-ldb.o
-
-imx8qxp-ldb-objs := imx8qxp-ldb-drv.o
 obj-$(CONFIG_DRM_IMX8QXP_LDB) += imx8qxp-ldb.o
-
 obj-$(CONFIG_DRM_IMX8QXP_PIXEL_COMBINER) += imx8qxp-pixel-combiner.o
 obj-$(CONFIG_DRM_IMX8QXP_PIXEL_LINK) += imx8qxp-pixel-link.o
 obj-$(CONFIG_DRM_IMX8QXP_PIXEL_LINK_TO_DPI) += imx8qxp-pxl2dpi.o
diff --git a/drivers/gpu/drm/bridge/imx/imx8qm-ldb-drv.c b/drivers/gpu/drm/bridge/imx/imx8qm-ldb.c
similarity index 100%
rename from drivers/gpu/drm/bridge/imx/imx8qm-ldb-drv.c
rename to drivers/gpu/drm/bridge/imx/imx8qm-ldb.c
diff --git a/drivers/gpu/drm/bridge/imx/imx8qxp-ldb-drv.c b/drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c
similarity index 100%
rename from drivers/gpu/drm/bridge/imx/imx8qxp-ldb-drv.c
rename to drivers/gpu/drm/bridge/imx/imx8qxp-ldb.c
-- 
GitLab


From d58bfcd4ea9910f59cf8327a36603e214e631b12 Mon Sep 17 00:00:00 2001
From: Arun R Murthy <arun.r.murthy@intel.com>
Date: Fri, 2 Jun 2023 07:51:57 +0530
Subject: [PATCH 0816/1791] drm/i915/display: Print useful information on error

For modifier not supporting async flip, print the modifier and display
version. Helps in reading the error message.

v2: Reframe the error message (Jani)

Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602022157.221225-1-arun.r.murthy@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index f51a55f4e9d0a..f23dd937c27c4 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -6012,8 +6012,9 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in
 			 */
 			if (DISPLAY_VER(i915) < 12) {
 				drm_dbg_kms(&i915->drm,
-					    "[PLANE:%d:%s] Modifier does not support async flips\n",
-					    plane->base.base.id, plane->base.name);
+					    "[PLANE:%d:%s] Modifier 0x%llx does not support async flip on display ver %d\n",
+					    plane->base.base.id, plane->base.name,
+					    new_plane_state->hw.fb->modifier, DISPLAY_VER(i915));
 				return -EINVAL;
 			}
 			break;
@@ -6025,8 +6026,9 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in
 			break;
 		default:
 			drm_dbg_kms(&i915->drm,
-				    "[PLANE:%d:%s] Modifier does not support async flips\n",
-				    plane->base.base.id, plane->base.name);
+				    "[PLANE:%d:%s] Modifier 0x%llx does not support async flip\n",
+				    plane->base.base.id, plane->base.name,
+				    new_plane_state->hw.fb->modifier);
 			return -EINVAL;
 		}
 
-- 
GitLab


From 19db2062094c75c64039d820c2547aad4dcfd905 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 31 May 2023 16:56:25 +0300
Subject: [PATCH 0817/1791] drm/i915: No 10bit gamma on desktop gen3 parts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Apparently desktop gen3 parts don't support the
10bit gamma mode at all. Stop claiming otherwise.

As is the case with pipe A on gen3 mobile parts, the
PIPECONF gamma mode bit can be set but it has no
effect on the output.

PNV seems to be the only slight exception, but generally
the desktop PNV variant looks more like a mobile part so
this is not entirely surprising.

Fixes: 67630bacae23 ("drm/i915: Add 10bit gamma mode for gen2/3")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531135625.3467-1-ville.syrjala@linux.intel.com
---
 .../gpu/drm/i915/display/intel_display_device.c  | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 464df1764a863..1aac7234e1861 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -222,7 +222,6 @@ static const struct intel_display_device_info i865g_display = {
 	.has_overlay = 1, \
 	I9XX_PIPE_OFFSETS, \
 	I9XX_CURSOR_OFFSETS, \
-	I9XX_COLORS, \
 	\
 	.__runtime_defaults.ip.ver = 3, \
 	.__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \
@@ -231,12 +230,14 @@ static const struct intel_display_device_info i865g_display = {
 
 static const struct intel_display_device_info i915g_display = {
 	GEN3_DISPLAY,
+	I845_COLORS,
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
 };
 
 static const struct intel_display_device_info i915gm_display = {
 	GEN3_DISPLAY,
+	I9XX_COLORS,
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
 	.supports_tv = 1,
@@ -246,6 +247,7 @@ static const struct intel_display_device_info i915gm_display = {
 
 static const struct intel_display_device_info i945g_display = {
 	GEN3_DISPLAY,
+	I845_COLORS,
 	.has_hotplug = 1,
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
@@ -253,6 +255,7 @@ static const struct intel_display_device_info i945g_display = {
 
 static const struct intel_display_device_info i945gm_display = {
 	GEN3_DISPLAY,
+	I9XX_COLORS,
 	.has_hotplug = 1,
 	.cursor_needs_physical = 1,
 	.overlay_needs_physical = 1,
@@ -263,6 +266,13 @@ static const struct intel_display_device_info i945gm_display = {
 
 static const struct intel_display_device_info g33_display = {
 	GEN3_DISPLAY,
+	I845_COLORS,
+	.has_hotplug = 1,
+};
+
+static const struct intel_display_device_info pnv_display = {
+	GEN3_DISPLAY,
+	I9XX_COLORS,
 	.has_hotplug = 1,
 };
 
@@ -677,8 +687,8 @@ static const struct {
 	INTEL_I965GM_IDS(&i965gm_display),
 	INTEL_GM45_IDS(&gm45_display),
 	INTEL_G45_IDS(&g45_display),
-	INTEL_PINEVIEW_G_IDS(&g33_display),
-	INTEL_PINEVIEW_M_IDS(&g33_display),
+	INTEL_PINEVIEW_G_IDS(&pnv_display),
+	INTEL_PINEVIEW_M_IDS(&pnv_display),
 	INTEL_IRONLAKE_D_IDS(&ilk_d_display),
 	INTEL_IRONLAKE_M_IDS(&ilk_m_display),
 	INTEL_SNB_D_IDS(&snb_display),
-- 
GitLab


From 3532e75dfadcf52f8cb56dbfb6bce4503c69091f Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:09 -0700
Subject: [PATCH 0818/1791] drm/i915/uc: perma-pin firmwares

Now that each FW has its own reserved area, we can keep them always
pinned and skip the pin/unpin dance on reset. This will make things
easier for the 2-step HuC authentication, which requires the FW to be
pinned in GGTT after the xfer is completed.
Since the vma is now valid for a long time and not just for the quick
pin-load-unpin dance, the name "dummy" is no longer appropriare and has
been replaced with vma_res. All the functions have also been updated to
operate on vma_res for consistency.
Given that we pin the vma behind the allocator's back (which is ok
because we do the pinning in an area that was previously reserved for
thus purpose), we do need to explicitly re-pin on resume because the
automated helper won't cover us.

v2: better comments and commit message, s/dummy/vma_res/

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-2-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/intel_ggtt.c      |  3 ++
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c |  7 +++-
 drivers/gpu/drm/i915/gt/uc/intel_guc.c    |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc.c    |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_uc.c     |  8 ++++
 drivers/gpu/drm/i915/gt/uc/intel_uc.h     |  2 +
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c  | 50 ++++++++++++++++-------
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h  | 22 ++++++----
 8 files changed, 71 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
index 122197737ef2e..dd0ed941441aa 100644
--- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
+++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
@@ -1326,6 +1326,9 @@ void i915_ggtt_resume(struct i915_ggtt *ggtt)
 		ggtt->vm.scratch_range(&ggtt->vm, ggtt->error_capture.start,
 				       ggtt->error_capture.size);
 
+	list_for_each_entry(gt, &ggtt->gt_list, ggtt_link)
+		intel_uc_resume_mappings(&gt->uc);
+
 	ggtt->invalidate(ggtt);
 
 	if (flush)
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
index fb0984f875f9e..b26f493f86fa5 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
@@ -90,7 +90,12 @@ void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc)
 {
 	struct intel_gt *gt = gsc_uc_to_gt(gsc);
 
-	intel_uc_fw_init_early(&gsc->fw, INTEL_UC_FW_TYPE_GSC);
+	/*
+	 * GSC FW needs to be copied to a dedicated memory allocations for
+	 * loading (see gsc->local), so we don't need to GGTT map the FW image
+	 * itself into GGTT.
+	 */
+	intel_uc_fw_init_early(&gsc->fw, INTEL_UC_FW_TYPE_GSC, false);
 	INIT_WORK(&gsc->work, gsc_work);
 
 	/* we can arrive here from i915_driver_early_probe for primary
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
index f9bddaa876d9b..a4203018d7a3d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
@@ -163,7 +163,7 @@ void intel_guc_init_early(struct intel_guc *guc)
 	struct intel_gt *gt = guc_to_gt(guc);
 	struct drm_i915_private *i915 = gt->i915;
 
-	intel_uc_fw_init_early(&guc->fw, INTEL_UC_FW_TYPE_GUC);
+	intel_uc_fw_init_early(&guc->fw, INTEL_UC_FW_TYPE_GUC, true);
 	intel_guc_ct_init_early(&guc->ct);
 	intel_guc_log_init_early(&guc->log);
 	intel_guc_submission_init_early(guc);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index 04724ff56ded3..268e036f8f284 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -276,7 +276,7 @@ void intel_huc_init_early(struct intel_huc *huc)
 	struct drm_i915_private *i915 = huc_to_gt(huc)->i915;
 	struct intel_gt *gt = huc_to_gt(huc);
 
-	intel_uc_fw_init_early(&huc->fw, INTEL_UC_FW_TYPE_HUC);
+	intel_uc_fw_init_early(&huc->fw, INTEL_UC_FW_TYPE_HUC, true);
 
 	/*
 	 * we always init the fence as already completed, even if HuC is not
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index c8b9cbb7ba3a9..1e7f5cc9d5508 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -700,6 +700,12 @@ void intel_uc_suspend(struct intel_uc *uc)
 	}
 }
 
+static void __uc_resume_mappings(struct intel_uc *uc)
+{
+	intel_uc_fw_resume_mapping(&uc->guc.fw);
+	intel_uc_fw_resume_mapping(&uc->huc.fw);
+}
+
 static int __uc_resume(struct intel_uc *uc, bool enable_communication)
 {
 	struct intel_guc *guc = &uc->guc;
@@ -767,4 +773,6 @@ static const struct intel_uc_ops uc_ops_on = {
 
 	.init_hw = __uc_init_hw,
 	.fini_hw = __uc_fini_hw,
+
+	.resume_mappings = __uc_resume_mappings,
 };
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_uc.h
index d585524d94deb..014bb7d83689a 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.h
@@ -24,6 +24,7 @@ struct intel_uc_ops {
 	void (*fini)(struct intel_uc *uc);
 	int (*init_hw)(struct intel_uc *uc);
 	void (*fini_hw)(struct intel_uc *uc);
+	void (*resume_mappings)(struct intel_uc *uc);
 };
 
 struct intel_uc {
@@ -114,6 +115,7 @@ intel_uc_ops_function(init, init, int, 0);
 intel_uc_ops_function(fini, fini, void, );
 intel_uc_ops_function(init_hw, init_hw, int, 0);
 intel_uc_ops_function(fini_hw, fini_hw, void, );
+intel_uc_ops_function(resume_mappings, resume_mappings, void, );
 #undef intel_uc_ops_function
 
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index dc5c96c503a92..31776c279f323 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -471,12 +471,14 @@ static void __uc_fw_user_override(struct drm_i915_private *i915, struct intel_uc
  * intel_uc_fw_init_early - initialize the uC object and select the firmware
  * @uc_fw: uC firmware
  * @type: type of uC
+ * @needs_ggtt_mapping: whether the FW needs to be GGTT mapped for loading
  *
  * Initialize the state of our uC object and relevant tracking and select the
  * firmware to fetch and load.
  */
 void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw,
-			    enum intel_uc_fw_type type)
+			    enum intel_uc_fw_type type,
+			    bool needs_ggtt_mapping)
 {
 	struct intel_gt *gt = ____uc_fw_to_gt(uc_fw, type);
 	struct drm_i915_private *i915 = gt->i915;
@@ -490,6 +492,7 @@ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw,
 	GEM_BUG_ON(uc_fw->file_selected.path);
 
 	uc_fw->type = type;
+	uc_fw->needs_ggtt_mapping = needs_ggtt_mapping;
 
 	if (HAS_GT_UC(i915)) {
 		if (!validate_fw_table_type(i915, type)) {
@@ -755,7 +758,7 @@ static int try_firmware_load(struct intel_uc_fw *uc_fw, const struct firmware **
 	if (err)
 		return err;
 
-	if ((*fw)->size > INTEL_UC_RSVD_GGTT_PER_FW) {
+	if (uc_fw->needs_ggtt_mapping && (*fw)->size > INTEL_UC_RSVD_GGTT_PER_FW) {
 		gt_err(gt, "%s firmware %s: size (%zuKB) exceeds max supported size (%uKB)\n",
 		       intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
 		       (*fw)->size / SZ_1K, INTEL_UC_RSVD_GGTT_PER_FW / SZ_1K);
@@ -940,29 +943,32 @@ static void uc_fw_bind_ggtt(struct intel_uc_fw *uc_fw)
 {
 	struct drm_i915_gem_object *obj = uc_fw->obj;
 	struct i915_ggtt *ggtt = __uc_fw_to_gt(uc_fw)->ggtt;
-	struct i915_vma_resource *dummy = &uc_fw->dummy;
+	struct i915_vma_resource *vma_res = &uc_fw->vma_res;
 	u32 pte_flags = 0;
 
-	dummy->start = uc_fw_ggtt_offset(uc_fw);
-	dummy->node_size = obj->base.size;
-	dummy->bi.pages = obj->mm.pages;
+	if (!uc_fw->needs_ggtt_mapping)
+		return;
+
+	vma_res->start = uc_fw_ggtt_offset(uc_fw);
+	vma_res->node_size = obj->base.size;
+	vma_res->bi.pages = obj->mm.pages;
 
 	GEM_BUG_ON(!i915_gem_object_has_pinned_pages(obj));
 
 	/* uc_fw->obj cache domains were not controlled across suspend */
 	if (i915_gem_object_has_struct_page(obj))
-		drm_clflush_sg(dummy->bi.pages);
+		drm_clflush_sg(vma_res->bi.pages);
 
 	if (i915_gem_object_is_lmem(obj))
 		pte_flags |= PTE_LM;
 
 	if (ggtt->vm.raw_insert_entries)
-		ggtt->vm.raw_insert_entries(&ggtt->vm, dummy,
+		ggtt->vm.raw_insert_entries(&ggtt->vm, vma_res,
 					    i915_gem_get_pat_index(ggtt->vm.i915,
 								   I915_CACHE_NONE),
 					    pte_flags);
 	else
-		ggtt->vm.insert_entries(&ggtt->vm, dummy,
+		ggtt->vm.insert_entries(&ggtt->vm, vma_res,
 					i915_gem_get_pat_index(ggtt->vm.i915,
 							       I915_CACHE_NONE),
 					pte_flags);
@@ -970,11 +976,13 @@ static void uc_fw_bind_ggtt(struct intel_uc_fw *uc_fw)
 
 static void uc_fw_unbind_ggtt(struct intel_uc_fw *uc_fw)
 {
-	struct drm_i915_gem_object *obj = uc_fw->obj;
 	struct i915_ggtt *ggtt = __uc_fw_to_gt(uc_fw)->ggtt;
-	u64 start = uc_fw_ggtt_offset(uc_fw);
+	struct i915_vma_resource *vma_res = &uc_fw->vma_res;
+
+	if (!vma_res->node_size)
+		return;
 
-	ggtt->vm.clear_range(&ggtt->vm, start, obj->base.size);
+	ggtt->vm.clear_range(&ggtt->vm, vma_res->start, vma_res->node_size);
 }
 
 static int uc_fw_xfer(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags)
@@ -991,7 +999,7 @@ static int uc_fw_xfer(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags)
 	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
 	/* Set the source address for the uCode */
-	offset = uc_fw_ggtt_offset(uc_fw);
+	offset = uc_fw->vma_res.start;
 	GEM_BUG_ON(upper_32_bits(offset) & 0xFFFF0000);
 	intel_uncore_write_fw(uncore, DMA_ADDR_0_LOW, lower_32_bits(offset));
 	intel_uncore_write_fw(uncore, DMA_ADDR_0_HIGH, upper_32_bits(offset));
@@ -1065,9 +1073,7 @@ int intel_uc_fw_upload(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags)
 		return -ENOEXEC;
 
 	/* Call custom loader */
-	uc_fw_bind_ggtt(uc_fw);
 	err = uc_fw_xfer(uc_fw, dst_offset, dma_flags);
-	uc_fw_unbind_ggtt(uc_fw);
 	if (err)
 		goto fail;
 
@@ -1171,6 +1177,8 @@ int intel_uc_fw_init(struct intel_uc_fw *uc_fw)
 		goto out_unpin;
 	}
 
+	uc_fw_bind_ggtt(uc_fw);
+
 	return 0;
 
 out_unpin:
@@ -1181,6 +1189,7 @@ int intel_uc_fw_init(struct intel_uc_fw *uc_fw)
 
 void intel_uc_fw_fini(struct intel_uc_fw *uc_fw)
 {
+	uc_fw_unbind_ggtt(uc_fw);
 	uc_fw_rsa_data_destroy(uc_fw);
 
 	if (i915_gem_object_has_pinned_pages(uc_fw->obj))
@@ -1189,6 +1198,17 @@ void intel_uc_fw_fini(struct intel_uc_fw *uc_fw)
 	intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_AVAILABLE);
 }
 
+void intel_uc_fw_resume_mapping(struct intel_uc_fw *uc_fw)
+{
+	if (!intel_uc_fw_is_available(uc_fw))
+		return;
+
+	if (!i915_gem_object_has_pinned_pages(uc_fw->obj))
+		return;
+
+	uc_fw_bind_ggtt(uc_fw);
+}
+
 /**
  * intel_uc_fw_cleanup_fetch - cleanup uC firmware
  * @uc_fw: uC firmware
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
index 6ba00e6b39753..2be9470eb7124 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
@@ -99,13 +99,19 @@ struct intel_uc_fw {
 	struct drm_i915_gem_object *obj;
 
 	/**
-	 * @dummy: A vma used in binding the uc fw to ggtt. We can't define this
-	 * vma on the stack as it can lead to a stack overflow, so we define it
-	 * here. Safe to have 1 copy per uc fw because the binding is single
-	 * threaded as it done during driver load (inherently single threaded)
-	 * or during a GT reset (mutex guarantees single threaded).
+	 * @needs_ggtt_mapping: indicates whether the fw object needs to be
+	 * pinned to ggtt. If true, the fw is pinned at init time and unpinned
+	 * during driver unload.
 	 */
-	struct i915_vma_resource dummy;
+	bool needs_ggtt_mapping;
+
+	/**
+	 * @vma_res: A vma resource used in binding the uc fw to ggtt. The fw is
+	 * pinned in a reserved area of the ggtt (above the maximum address
+	 * usable by GuC); therefore, we can't use the normal vma functions to
+	 * do the pinning and we instead use this resource to do so.
+	 */
+	struct i915_vma_resource vma_res;
 	struct i915_vma *rsa_data;
 
 	u32 rsa_size;
@@ -282,12 +288,14 @@ static inline u32 intel_uc_fw_get_upload_size(struct intel_uc_fw *uc_fw)
 }
 
 void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw,
-			    enum intel_uc_fw_type type);
+			    enum intel_uc_fw_type type,
+			    bool needs_ggtt_mapping);
 int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw);
 void intel_uc_fw_cleanup_fetch(struct intel_uc_fw *uc_fw);
 int intel_uc_fw_upload(struct intel_uc_fw *uc_fw, u32 offset, u32 dma_flags);
 int intel_uc_fw_init(struct intel_uc_fw *uc_fw);
 void intel_uc_fw_fini(struct intel_uc_fw *uc_fw);
+void intel_uc_fw_resume_mapping(struct intel_uc_fw *uc_fw);
 size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len);
 int intel_uc_fw_mark_load_failed(struct intel_uc_fw *uc_fw, int err);
 void intel_uc_fw_dump(const struct intel_uc_fw *uc_fw, struct drm_printer *p);
-- 
GitLab


From 93a575ab15332e2037d35744c50412b40e6511a4 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:10 -0700
Subject: [PATCH 0819/1791] drm/i915/huc: Parse the GSC-enabled HuC binary

The new binaries that support the 2-step authentication contain the
legacy-style binary, which we can use for loading the HuC via DMA. To
find out where this is located in the image, we need to parse the
manifest of the GSC-enabled HuC binary. The manifest consist of a
partition header followed by entries, one of which contains the offset
we're looking for.
Note that the DG2 GSC binary contains entries with the same names, but
it doesn't contain a full legacy binary, so we need to skip assigning
the dma offset in that case (which we can do by checking the ccs).
Also, since we're now parsing the entries, we can extract the HuC
version that way instead of using hardcoded offsets.

Note that the GSC binary uses the same structures in its binary header,
so they've been added in their own header file.

v2: fix structure names to match meu defines (s/CPT/CPD/), update commit
    message, check ccs validity, drop old version location defines.

v3: drop references to the MEU tool to reduce confusion, fix log (John)

v4: fix log for real (John)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com> #v2
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-3-daniele.ceraolospurio@intel.com
---
 .../drm/i915/gt/uc/intel_gsc_binary_headers.h |  74 ++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_huc.c        |  11 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c     | 136 ++++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h     |   5 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc_print.h  |  21 +++
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c      |  72 +++++-----
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h      |   2 +
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h  |   6 -
 8 files changed, 274 insertions(+), 53 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h
 create mode 100644 drivers/gpu/drm/i915/gt/uc/intel_huc_print.h

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h
new file mode 100644
index 0000000000000..714f0c256118c
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef _INTEL_GSC_BINARY_HEADERS_H_
+#define _INTEL_GSC_BINARY_HEADERS_H_
+
+#include <linux/types.h>
+
+/* Code partition directory (CPD) structures */
+struct intel_gsc_cpd_header_v2 {
+	u32 header_marker;
+#define INTEL_GSC_CPD_HEADER_MARKER 0x44504324
+
+	u32 num_of_entries;
+	u8 header_version;
+	u8 entry_version;
+	u8 header_length; /* in bytes */
+	u8 flags;
+	u32 partition_name;
+	u32 crc32;
+} __packed;
+
+struct intel_gsc_cpd_entry {
+	u8 name[12];
+
+	/*
+	 * Bits 0-24: offset from the beginning of the code partition
+	 * Bit 25: huffman compressed
+	 * Bits 26-31: reserved
+	 */
+	u32 offset;
+#define INTEL_GSC_CPD_ENTRY_OFFSET_MASK GENMASK(24, 0)
+#define INTEL_GSC_CPD_ENTRY_HUFFMAN_COMP BIT(25)
+
+	/*
+	 * Module/Item length, in bytes. For Huffman-compressed modules, this
+	 * refers to the uncompressed size. For software-compressed modules,
+	 * this refers to the compressed size.
+	 */
+	u32 length;
+
+	u8 reserved[4];
+} __packed;
+
+struct intel_gsc_version {
+	u16 major;
+	u16 minor;
+	u16 hotfix;
+	u16 build;
+} __packed;
+
+struct intel_gsc_manifest_header {
+	u32 header_type; /* 0x4 for manifest type */
+	u32 header_length; /* in dwords */
+	u32 header_version;
+	u32 flags;
+	u32 vendor;
+	u32 date;
+	u32 size; /* In dwords, size of entire manifest (header + extensions) */
+	u32 header_id;
+	u32 internal_data;
+	struct intel_gsc_version fw_version;
+	u32 security_version;
+	struct intel_gsc_version meu_kit_version;
+	u32 meu_manifest_version;
+	u8 general_data[4];
+	u8 reserved3[56];
+	u32 modulus_size; /* in dwords */
+	u32 exponent_size; /* in dwords */
+} __packed;
+
+#endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index 268e036f8f284..6d795438b3e4b 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -6,23 +6,14 @@
 #include <linux/types.h>
 
 #include "gt/intel_gt.h"
-#include "gt/intel_gt_print.h"
 #include "intel_guc_reg.h"
 #include "intel_huc.h"
+#include "intel_huc_print.h"
 #include "i915_drv.h"
 
 #include <linux/device/bus.h>
 #include <linux/mei_aux.h>
 
-#define huc_printk(_huc, _level, _fmt, ...) \
-	gt_##_level(huc_to_gt(_huc), "HuC: " _fmt, ##__VA_ARGS__)
-#define huc_err(_huc, _fmt, ...)	huc_printk((_huc), err, _fmt, ##__VA_ARGS__)
-#define huc_warn(_huc, _fmt, ...)	huc_printk((_huc), warn, _fmt, ##__VA_ARGS__)
-#define huc_notice(_huc, _fmt, ...)	huc_printk((_huc), notice, _fmt, ##__VA_ARGS__)
-#define huc_info(_huc, _fmt, ...)	huc_printk((_huc), info, _fmt, ##__VA_ARGS__)
-#define huc_dbg(_huc, _fmt, ...)	huc_printk((_huc), dbg, _fmt, ##__VA_ARGS__)
-#define huc_probe_error(_huc, _fmt, ...) huc_printk((_huc), probe_error, _fmt, ##__VA_ARGS__)
-
 /**
  * DOC: HuC
  *
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
index 534b0aa433166..3a9d81899a78e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
@@ -5,11 +5,147 @@
 
 #include "gt/intel_gsc.h"
 #include "gt/intel_gt.h"
+#include "intel_gsc_binary_headers.h"
 #include "intel_huc.h"
 #include "intel_huc_fw.h"
+#include "intel_huc_print.h"
 #include "i915_drv.h"
 #include "pxp/intel_pxp_huc.h"
 
+static void get_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, const void *data)
+{
+	const struct intel_gsc_manifest_header *manifest = data;
+
+	ver->major = manifest->fw_version.major;
+	ver->minor = manifest->fw_version.minor;
+	ver->patch = manifest->fw_version.hotfix;
+}
+
+static bool css_valid(const void *data, size_t size)
+{
+	const struct uc_css_header *css = data;
+
+	if (unlikely(size < sizeof(struct uc_css_header)))
+		return false;
+
+	if (css->module_type != 0x6)
+		return false;
+
+	if (css->module_vendor != PCI_VENDOR_ID_INTEL)
+		return false;
+
+	return true;
+}
+
+static inline u32 entry_offset(const struct intel_gsc_cpd_entry *entry)
+{
+	return entry->offset & INTEL_GSC_CPD_ENTRY_OFFSET_MASK;
+}
+
+int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, size_t size)
+{
+	struct intel_huc *huc = container_of(huc_fw, struct intel_huc, fw);
+	const struct intel_gsc_cpd_header_v2 *header = data;
+	const struct intel_gsc_cpd_entry *entry;
+	size_t min_size = sizeof(*header);
+	int i;
+
+	if (!huc_fw->loaded_via_gsc) {
+		huc_err(huc, "Invalid FW type for GSC header parsing!\n");
+		return -EINVAL;
+	}
+
+	if (size < sizeof(*header)) {
+		huc_err(huc, "FW too small! %zu < %zu\n", size, min_size);
+		return -ENODATA;
+	}
+
+	/*
+	 * The GSC-enabled HuC binary starts with a directory header, followed
+	 * by a series of entries. Each entry is identified by a name and
+	 * points to a specific section of the binary containing the relevant
+	 * data. The entries we're interested in are:
+	 * - "HUCP.man": points to the GSC manifest header for the HuC, which
+	 *               contains the version info.
+	 * - "huc_fw": points to the legacy-style binary that can be used for
+	 *             load via the DMA. This entry only contains a valid CSS
+	 *             on binaries for platforms that support 2-step HuC load
+	 *             via dma and auth via GSC (like MTL).
+	 *
+	 * --------------------------------------------------
+	 * [  intel_gsc_cpd_header_v2                       ]
+	 * --------------------------------------------------
+	 * [  intel_gsc_cpd_entry[]                         ]
+	 * [      entry1                                    ]
+	 * [      ...                                       ]
+	 * [      entryX                                    ]
+	 * [          "HUCP.man"                            ]
+	 * [           ...                                  ]
+	 * [           offset  >----------------------------]------o
+	 * [      ...                                       ]      |
+	 * [      entryY                                    ]      |
+	 * [          "huc_fw"                              ]      |
+	 * [           ...                                  ]      |
+	 * [           offset  >----------------------------]----------o
+	 * --------------------------------------------------      |   |
+	 *                                                         |   |
+	 * --------------------------------------------------      |   |
+	 * [ intel_gsc_manifest_header                      ]<-----o   |
+	 * [  ...                                           ]          |
+	 * [  intel_gsc_version fw_version                  ]          |
+	 * [  ...                                           ]          |
+	 * --------------------------------------------------          |
+	 *                                                             |
+	 * --------------------------------------------------          |
+	 * [ data[]                                         ]<---------o
+	 * [  ...                                           ]
+	 * [  ...                                           ]
+	 * --------------------------------------------------
+	 */
+
+	if (header->header_marker != INTEL_GSC_CPD_HEADER_MARKER) {
+		huc_err(huc, "invalid marker for CPD header: 0x%08x!\n",
+			header->header_marker);
+		return -EINVAL;
+	}
+
+	/* we only have binaries with header v2 and entry v1 for now */
+	if (header->header_version != 2 || header->entry_version != 1) {
+		huc_err(huc, "invalid CPD header/entry version %u:%u!\n",
+			header->header_version, header->entry_version);
+		return -EINVAL;
+	}
+
+	if (header->header_length < sizeof(struct intel_gsc_cpd_header_v2)) {
+		huc_err(huc, "invalid CPD header length %u!\n",
+			header->header_length);
+		return -EINVAL;
+	}
+
+	min_size = header->header_length + sizeof(*entry) * header->num_of_entries;
+	if (size < min_size) {
+		huc_err(huc, "FW too small! %zu < %zu\n", size, min_size);
+		return -ENODATA;
+	}
+
+	entry = data + header->header_length;
+
+	for (i = 0; i < header->num_of_entries; i++, entry++) {
+		if (strcmp(entry->name, "HUCP.man") == 0)
+			get_version_from_gsc_manifest(&huc_fw->file_selected.ver,
+						      data + entry_offset(entry));
+
+		if (strcmp(entry->name, "huc_fw") == 0) {
+			u32 offset = entry_offset(entry);
+
+			if (offset < size && css_valid(data + offset, size - offset))
+				huc_fw->dma_start_offset = offset;
+		}
+	}
+
+	return 0;
+}
+
 int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc)
 {
 	int ret;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
index db42e238b45f2..0999ffe6f9627 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
@@ -7,8 +7,11 @@
 #define _INTEL_HUC_FW_H_
 
 struct intel_huc;
+struct intel_uc_fw;
+
+#include <linux/types.h>
 
 int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc);
 int intel_huc_fw_upload(struct intel_huc *huc);
-
+int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, size_t size);
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h b/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h
new file mode 100644
index 0000000000000..915d310ee1df8
--- /dev/null
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_HUC_PRINT__
+#define __INTEL_HUC_PRINT__
+
+#include "gt/intel_gt.h"
+#include "gt/intel_gt_print.h"
+
+#define huc_printk(_huc, _level, _fmt, ...) \
+	gt_##_level(huc_to_gt(_huc), "HuC: " _fmt, ##__VA_ARGS__)
+#define huc_err(_huc, _fmt, ...)	huc_printk((_huc), err, _fmt, ##__VA_ARGS__)
+#define huc_warn(_huc, _fmt, ...)	huc_printk((_huc), warn, _fmt, ##__VA_ARGS__)
+#define huc_notice(_huc, _fmt, ...)	huc_printk((_huc), notice, _fmt, ##__VA_ARGS__)
+#define huc_info(_huc, _fmt, ...)	huc_printk((_huc), info, _fmt, ##__VA_ARGS__)
+#define huc_dbg(_huc, _fmt, ...)	huc_printk((_huc), dbg, _fmt, ##__VA_ARGS__)
+#define huc_probe_error(_huc, _fmt, ...) huc_printk((_huc), probe_error, _fmt, ##__VA_ARGS__)
+
+#endif /* __INTEL_HUC_PRINT__ */
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index 31776c279f323..ec0b3d214af1c 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -548,33 +548,6 @@ static void __force_fw_fetch_failures(struct intel_uc_fw *uc_fw, int e)
 	}
 }
 
-static int check_gsc_manifest(struct intel_gt *gt,
-			      const struct firmware *fw,
-			      struct intel_uc_fw *uc_fw)
-{
-	u32 *dw = (u32 *)fw->data;
-	u32 version_hi, version_lo;
-	size_t min_size;
-
-	/* Check the size of the blob before examining buffer contents */
-	min_size = sizeof(u32) * (HUC_GSC_VERSION_LO_DW + 1);
-	if (unlikely(fw->size < min_size)) {
-		gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n",
-			intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
-			fw->size, min_size);
-		return -ENODATA;
-	}
-
-	version_hi = dw[HUC_GSC_VERSION_HI_DW];
-	version_lo = dw[HUC_GSC_VERSION_LO_DW];
-
-	uc_fw->file_selected.ver.major = FIELD_GET(HUC_GSC_MAJOR_VER_HI_MASK, version_hi);
-	uc_fw->file_selected.ver.minor = FIELD_GET(HUC_GSC_MINOR_VER_HI_MASK, version_hi);
-	uc_fw->file_selected.ver.patch = FIELD_GET(HUC_GSC_PATCH_VER_LO_MASK, version_lo);
-
-	return 0;
-}
-
 static void uc_unpack_css_version(struct intel_uc_fw_ver *ver, u32 css_value)
 {
 	/* Get version numbers from the CSS header */
@@ -631,22 +604,22 @@ static void guc_read_css_info(struct intel_uc_fw *uc_fw, struct uc_css_header *c
 	uc_fw->private_data_size = css->private_data_size;
 }
 
-static int check_ccs_header(struct intel_gt *gt,
-			    const struct firmware *fw,
-			    struct intel_uc_fw *uc_fw)
+static int __check_ccs_header(struct intel_gt *gt,
+			      const void *fw_data, size_t fw_size,
+			      struct intel_uc_fw *uc_fw)
 {
 	struct uc_css_header *css;
 	size_t size;
 
 	/* Check the size of the blob before examining buffer contents */
-	if (unlikely(fw->size < sizeof(struct uc_css_header))) {
+	if (unlikely(fw_size < sizeof(struct uc_css_header))) {
 		gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n",
 			intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
-			fw->size, sizeof(struct uc_css_header));
+			fw_size, sizeof(struct uc_css_header));
 		return -ENODATA;
 	}
 
-	css = (struct uc_css_header *)fw->data;
+	css = (struct uc_css_header *)fw_data;
 
 	/* Check integrity of size values inside CSS header */
 	size = (css->header_size_dw - css->key_size_dw - css->modulus_size_dw -
@@ -654,7 +627,7 @@ static int check_ccs_header(struct intel_gt *gt,
 	if (unlikely(size != sizeof(struct uc_css_header))) {
 		gt_warn(gt, "%s firmware %s: unexpected header size: %zu != %zu\n",
 			intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
-			fw->size, sizeof(struct uc_css_header));
+			fw_size, sizeof(struct uc_css_header));
 		return -EPROTO;
 	}
 
@@ -666,10 +639,10 @@ static int check_ccs_header(struct intel_gt *gt,
 
 	/* At least, it should have header, uCode and RSA. Size of all three. */
 	size = sizeof(struct uc_css_header) + uc_fw->ucode_size + uc_fw->rsa_size;
-	if (unlikely(fw->size < size)) {
+	if (unlikely(fw_size < size)) {
 		gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n",
 			intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path,
-			fw->size, size);
+			fw_size, size);
 		return -ENOEXEC;
 	}
 
@@ -690,6 +663,33 @@ static int check_ccs_header(struct intel_gt *gt,
 	return 0;
 }
 
+static int check_gsc_manifest(struct intel_gt *gt,
+			      const struct firmware *fw,
+			      struct intel_uc_fw *uc_fw)
+{
+	if (uc_fw->type != INTEL_UC_FW_TYPE_HUC) {
+		gt_err(gt, "trying to GSC-parse a non-HuC binary");
+		return -EINVAL;
+	}
+
+	intel_huc_fw_get_binary_info(uc_fw, fw->data, fw->size);
+
+	if (uc_fw->dma_start_offset) {
+		u32 delta = uc_fw->dma_start_offset;
+
+		__check_ccs_header(gt, fw->data + delta, fw->size - delta, uc_fw);
+	}
+
+	return 0;
+}
+
+static int check_ccs_header(struct intel_gt *gt,
+			    const struct firmware *fw,
+			    struct intel_uc_fw *uc_fw)
+{
+	return __check_ccs_header(gt, fw->data, fw->size, uc_fw);
+}
+
 static bool is_ver_8bit(struct intel_uc_fw_ver *ver)
 {
 	return ver->major < 0xFF && ver->minor < 0xFF && ver->patch < 0xFF;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
index 2be9470eb7124..b3daba9526ebf 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
@@ -118,6 +118,8 @@ struct intel_uc_fw {
 	u32 ucode_size;
 	u32 private_data_size;
 
+	u32 dma_start_offset;
+
 	bool loaded_via_gsc;
 };
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h
index 646fa8aa6cf19..7fe4051262497 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h
@@ -84,10 +84,4 @@ struct uc_css_header {
 } __packed;
 static_assert(sizeof(struct uc_css_header) == 128);
 
-#define HUC_GSC_VERSION_HI_DW		44
-#define   HUC_GSC_MAJOR_VER_HI_MASK	(0xFF << 0)
-#define   HUC_GSC_MINOR_VER_HI_MASK	(0xFF << 16)
-#define HUC_GSC_VERSION_LO_DW		45
-#define   HUC_GSC_PATCH_VER_LO_MASK	(0xFF << 0)
-
 #endif /* _INTEL_UC_FW_ABI_H */
-- 
GitLab


From 6a9b6c4580af184f1f8744ade1fe1979e3da05ac Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:11 -0700
Subject: [PATCH 0820/1791] drm/i915/huc: Load GSC-enabled HuC via DMA xfer if
 the fuse says so

In the previous patch we extracted the offset of the legacy-style HuC
binary located within the GSC-enabled blob, so now we can use that to
load the HuC via DMA if the fuse is set that way.
Note that we now need to differentiate between "GSC-enabled binary" and
"loaded by GSC", so the former case has been renamed to "has GSC headers"
for clarity, while the latter is now based on the fuse instead of the
binary format. This way, all the legacy load paths are automatically
taken (including the auth by GuC) without having to implement further
code changes.

v2: s/is_meu_binary/has_gsc_headers/, clearer logs (John)

v3: split check for GSC access, better comments (John)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-4-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_huc.c    | 49 +++++++++++++++++------
 drivers/gpu/drm/i915/gt/uc/intel_huc.h    |  4 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c  | 12 +++---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h  |  2 +-
 5 files changed, 47 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index 6d795438b3e4b..b1e7a1e1bfa34 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -298,31 +298,54 @@ void intel_huc_init_early(struct intel_huc *huc)
 static int check_huc_loading_mode(struct intel_huc *huc)
 {
 	struct intel_gt *gt = huc_to_gt(huc);
-	bool fw_needs_gsc = intel_huc_is_loaded_by_gsc(huc);
-	bool hw_uses_gsc = false;
+	bool gsc_enabled = huc->fw.has_gsc_headers;
 
 	/*
 	 * The fuse for HuC load via GSC is only valid on platforms that have
 	 * GuC deprivilege.
 	 */
 	if (HAS_GUC_DEPRIVILEGE(gt->i915))
-		hw_uses_gsc = intel_uncore_read(gt->uncore, GUC_SHIM_CONTROL2) &
-			      GSC_LOADS_HUC;
+		huc->loaded_via_gsc = intel_uncore_read(gt->uncore, GUC_SHIM_CONTROL2) &
+				      GSC_LOADS_HUC;
 
-	if (fw_needs_gsc != hw_uses_gsc) {
-		huc_err(huc, "mismatch between FW (%s) and HW (%s) load modes\n",
-			HUC_LOAD_MODE_STRING(fw_needs_gsc), HUC_LOAD_MODE_STRING(hw_uses_gsc));
+	if (huc->loaded_via_gsc && !gsc_enabled) {
+		huc_err(huc, "HW requires a GSC-enabled blob, but we found a legacy one\n");
 		return -ENOEXEC;
 	}
 
-	/* make sure we can access the GSC via the mei driver if we need it */
-	if (!(IS_ENABLED(CONFIG_INTEL_MEI_PXP) && IS_ENABLED(CONFIG_INTEL_MEI_GSC)) &&
-	    fw_needs_gsc) {
-		huc_info(huc, "can't load due to missing MEI modules\n");
-		return -EIO;
+	/*
+	 * On newer platforms we have GSC-enabled binaries but we load the HuC
+	 * via DMA. To do so we need to find the location of the legacy-style
+	 * binary inside the GSC-enabled one, which we do at fetch time. Make
+	 * sure that we were able to do so if the fuse says we need to load via
+	 * DMA and the binary is GSC-enabled.
+	 */
+	if (!huc->loaded_via_gsc && gsc_enabled && !huc->fw.dma_start_offset) {
+		huc_err(huc, "HW in DMA mode, but we have an incompatible GSC-enabled blob\n");
+		return -ENOEXEC;
+	}
+
+	/*
+	 * If the HuC is loaded via GSC, we need to be able to access the GSC.
+	 * On DG2 this is done via the mei components, while on newer platforms
+	 * it is done via the GSCCS,
+	 */
+	if (huc->loaded_via_gsc) {
+		if (IS_DG2(gt->i915)) {
+			if (!IS_ENABLED(CONFIG_INTEL_MEI_PXP) ||
+			    !IS_ENABLED(CONFIG_INTEL_MEI_GSC)) {
+				huc_info(huc, "can't load due to missing mei modules\n");
+				return -EIO;
+			}
+		} else {
+			if (!HAS_ENGINE(gt, GSC0)) {
+				huc_info(huc, "can't load due to missing GSCCS\n");
+				return -EIO;
+			}
+		}
 	}
 
-	huc_dbg(huc, "loaded by GSC = %s\n", str_yes_no(fw_needs_gsc));
+	huc_dbg(huc, "loaded by GSC = %s\n", str_yes_no(huc->loaded_via_gsc));
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
index 0789184d81a2a..112f0dce47026 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
@@ -39,6 +39,8 @@ struct intel_huc {
 		struct notifier_block nb;
 		enum intel_huc_delayed_load_status status;
 	} delayed_load;
+
+	bool loaded_via_gsc;
 };
 
 int intel_huc_sanitize(struct intel_huc *huc);
@@ -73,7 +75,7 @@ static inline bool intel_huc_is_used(struct intel_huc *huc)
 
 static inline bool intel_huc_is_loaded_by_gsc(const struct intel_huc *huc)
 {
-	return huc->fw.loaded_via_gsc;
+	return huc->loaded_via_gsc;
 }
 
 static inline bool intel_huc_wait_required(struct intel_huc *huc)
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
index 3a9d81899a78e..89a887d33b773 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
@@ -50,7 +50,7 @@ int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, s
 	size_t min_size = sizeof(*header);
 	int i;
 
-	if (!huc_fw->loaded_via_gsc) {
+	if (!huc_fw->has_gsc_headers) {
 		huc_err(huc, "Invalid FW type for GSC header parsing!\n");
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index ec0b3d214af1c..a1c8a982479fb 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -186,7 +186,7 @@ struct __packed uc_fw_blob {
 	u8 major;
 	u8 minor;
 	u8 patch;
-	bool loaded_via_gsc;
+	bool has_gsc_headers;
 };
 
 #define UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \
@@ -197,7 +197,7 @@ struct __packed uc_fw_blob {
 
 #define UC_FW_BLOB_NEW(major_, minor_, patch_, gsc_, path_) \
 	{ UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \
-	  .legacy = false, .loaded_via_gsc = gsc_ }
+	  .legacy = false, .has_gsc_headers = gsc_ }
 
 #define UC_FW_BLOB_OLD(major_, minor_, patch_, path_) \
 	{ UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \
@@ -310,7 +310,7 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw)
 		uc_fw->file_wanted.ver.major = blob->major;
 		uc_fw->file_wanted.ver.minor = blob->minor;
 		uc_fw->file_wanted.ver.patch = blob->patch;
-		uc_fw->loaded_via_gsc = blob->loaded_via_gsc;
+		uc_fw->has_gsc_headers = blob->has_gsc_headers;
 		found = true;
 		break;
 	}
@@ -737,7 +737,7 @@ static int check_fw_header(struct intel_gt *gt,
 	if (uc_fw->type == INTEL_UC_FW_TYPE_GSC)
 		return 0;
 
-	if (uc_fw->loaded_via_gsc)
+	if (uc_fw->has_gsc_headers)
 		err = check_gsc_manifest(gt, fw, uc_fw);
 	else
 		err = check_ccs_header(gt, fw, uc_fw);
@@ -999,7 +999,7 @@ static int uc_fw_xfer(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags)
 	intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL);
 
 	/* Set the source address for the uCode */
-	offset = uc_fw->vma_res.start;
+	offset = uc_fw->vma_res.start + uc_fw->dma_start_offset;
 	GEM_BUG_ON(upper_32_bits(offset) & 0xFFFF0000);
 	intel_uncore_write_fw(uncore, DMA_ADDR_0_LOW, lower_32_bits(offset));
 	intel_uncore_write_fw(uncore, DMA_ADDR_0_HIGH, upper_32_bits(offset));
@@ -1238,7 +1238,7 @@ size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len)
 {
 	struct intel_memory_region *mr = uc_fw->obj->mm.region;
 	u32 size = min_t(u32, uc_fw->rsa_size, max_len);
-	u32 offset = sizeof(struct uc_css_header) + uc_fw->ucode_size;
+	u32 offset = uc_fw->dma_start_offset + sizeof(struct uc_css_header) + uc_fw->ucode_size;
 	struct sgt_iter iter;
 	size_t count = 0;
 	int idx;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
index b3daba9526ebf..054f028119718 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h
@@ -120,7 +120,7 @@ struct intel_uc_fw {
 
 	u32 dma_start_offset;
 
-	bool loaded_via_gsc;
+	bool has_gsc_headers;
 };
 
 /*
-- 
GitLab


From 98d2722a85c4ad5f2baf2272cbb0fab67f797b69 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:12 -0700
Subject: [PATCH 0821/1791] drm/i915/huc: differentiate the 2 steps of the MTL
 HuC auth flow

Before we add the second step of the MTL HuC auth (via GSC), we need to
have the ability to differentiate between them. To do so, the huc
authentication check is duplicated for GuC and GSC auth, with
GSC-enabled binaries being considered fully authenticated only after
the GSC auth step.

To report the difference between the 2 auth steps, a new case is added
to the HuC getparam. This way, the clear media driver can start
submitting before full auth, as partial auth is enough for those
workloads.

v2: fix authentication status check for DG2

v3: add a better comment at the top of the HuC file to explain the
    different approaches to load and auth (John)

v4: update call to intel_huc_is_authenticated in the pxp code to check
for GSC authentication

v5: drop references to meu and esclamation mark in huc_auth print (John)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com> #v2
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-5-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_huc.c     | 111 ++++++++++++++++-----
 drivers/gpu/drm/i915/gt/uc/intel_huc.h     |  16 ++-
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c  |   4 +-
 drivers/gpu/drm/i915/i915_reg.h            |   3 +
 drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c |   2 +-
 include/uapi/drm/i915_drm.h                |   3 +-
 6 files changed, 104 insertions(+), 35 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index b1e7a1e1bfa34..4c8592a281f20 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -10,6 +10,7 @@
 #include "intel_huc.h"
 #include "intel_huc_print.h"
 #include "i915_drv.h"
+#include "i915_reg.h"
 
 #include <linux/device/bus.h>
 #include <linux/mei_aux.h>
@@ -22,15 +23,23 @@
  * capabilities by adding HuC specific commands to batch buffers.
  *
  * The kernel driver is only responsible for loading the HuC firmware and
- * triggering its security authentication, which is performed by the GuC on
- * older platforms and by the GSC on newer ones. For the GuC to correctly
- * perform the authentication, the HuC binary must be loaded before the GuC one.
+ * triggering its security authentication. This is done differently depending
+ * on the platform:
+ * - older platforms (from Gen9 to most Gen12s): the load is performed via DMA
+ *   and the authentication via GuC
+ * - DG2: load and authentication are both performed via GSC.
+ * - MTL and newer platforms: the load is performed via DMA (same as with
+ *   not-DG2 older platforms), while the authentication is done in 2-steps,
+ *   a first auth for clear-media workloads via GuC and a second one for all
+ *   workloads via GSC.
+ * On platforms where the GuC does the authentication, to correctly do so the
+ * HuC binary must be loaded before the GuC one.
  * Loading the HuC is optional; however, not using the HuC might negatively
  * impact power usage and/or performance of media workloads, depending on the
  * use-cases.
  * HuC must be reloaded on events that cause the WOPCM to lose its contents
- * (S3/S4, FLR); GuC-authenticated HuC must also be reloaded on GuC/GT reset,
- * while GSC-managed HuC will survive that.
+ * (S3/S4, FLR); on older platforms the HuC must also be reloaded on GuC/GT
+ * reset, while on newer ones it will survive that.
  *
  * See https://github.com/intel/media-driver for the latest details on HuC
  * functionality.
@@ -106,7 +115,7 @@ static enum hrtimer_restart huc_delayed_load_timer_callback(struct hrtimer *hrti
 {
 	struct intel_huc *huc = container_of(hrtimer, struct intel_huc, delayed_load.timer);
 
-	if (!intel_huc_is_authenticated(huc)) {
+	if (!intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC)) {
 		if (huc->delayed_load.status == INTEL_HUC_WAITING_ON_GSC)
 			huc_notice(huc, "timed out waiting for MEI GSC\n");
 		else if (huc->delayed_load.status == INTEL_HUC_WAITING_ON_PXP)
@@ -124,7 +133,7 @@ static void huc_delayed_load_start(struct intel_huc *huc)
 {
 	ktime_t delay;
 
-	GEM_BUG_ON(intel_huc_is_authenticated(huc));
+	GEM_BUG_ON(intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC));
 
 	/*
 	 * On resume we don't have to wait for MEI-GSC to be re-probed, but we
@@ -284,13 +293,23 @@ void intel_huc_init_early(struct intel_huc *huc)
 	}
 
 	if (GRAPHICS_VER(i915) >= 11) {
-		huc->status.reg = GEN11_HUC_KERNEL_LOAD_INFO;
-		huc->status.mask = HUC_LOAD_SUCCESSFUL;
-		huc->status.value = HUC_LOAD_SUCCESSFUL;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].reg = GEN11_HUC_KERNEL_LOAD_INFO;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].mask = HUC_LOAD_SUCCESSFUL;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].value = HUC_LOAD_SUCCESSFUL;
 	} else {
-		huc->status.reg = HUC_STATUS2;
-		huc->status.mask = HUC_FW_VERIFIED;
-		huc->status.value = HUC_FW_VERIFIED;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].reg = HUC_STATUS2;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].mask = HUC_FW_VERIFIED;
+		huc->status[INTEL_HUC_AUTH_BY_GUC].value = HUC_FW_VERIFIED;
+	}
+
+	if (IS_DG2(i915)) {
+		huc->status[INTEL_HUC_AUTH_BY_GSC].reg = GEN11_HUC_KERNEL_LOAD_INFO;
+		huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HUC_LOAD_SUCCESSFUL;
+		huc->status[INTEL_HUC_AUTH_BY_GSC].value = HUC_LOAD_SUCCESSFUL;
+	} else {
+		huc->status[INTEL_HUC_AUTH_BY_GSC].reg = HECI_FWSTS5(MTL_GSC_HECI1_BASE);
+		huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HECI_FWSTS5_HUC_AUTH_DONE;
+		huc->status[INTEL_HUC_AUTH_BY_GSC].value = HECI_FWSTS5_HUC_AUTH_DONE;
 	}
 }
 
@@ -397,28 +416,38 @@ void intel_huc_suspend(struct intel_huc *huc)
 	delayed_huc_load_complete(huc);
 }
 
-int intel_huc_wait_for_auth_complete(struct intel_huc *huc)
+static const char *auth_mode_string(struct intel_huc *huc,
+				    enum intel_huc_authentication_type type)
+{
+	bool partial = huc->fw.has_gsc_headers && type == INTEL_HUC_AUTH_BY_GUC;
+
+	return partial ? "clear media" : "all workloads";
+}
+
+int intel_huc_wait_for_auth_complete(struct intel_huc *huc,
+				     enum intel_huc_authentication_type type)
 {
 	struct intel_gt *gt = huc_to_gt(huc);
 	int ret;
 
 	ret = __intel_wait_for_register(gt->uncore,
-					huc->status.reg,
-					huc->status.mask,
-					huc->status.value,
+					huc->status[type].reg,
+					huc->status[type].mask,
+					huc->status[type].value,
 					2, 50, NULL);
 
 	/* mark the load process as complete even if the wait failed */
 	delayed_huc_load_complete(huc);
 
 	if (ret) {
-		huc_err(huc, "firmware not verified %pe\n", ERR_PTR(ret));
+		huc_err(huc, "firmware not verified for %s: %pe\n",
+			auth_mode_string(huc, type), ERR_PTR(ret));
 		intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL);
 		return ret;
 	}
 
 	intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_RUNNING);
-	huc_info(huc, "authenticated!\n");
+	huc_info(huc, "authenticated for %s\n", auth_mode_string(huc, type));
 	return 0;
 }
 
@@ -458,7 +487,7 @@ int intel_huc_auth(struct intel_huc *huc)
 	}
 
 	/* Check authentication status, it should be done by now */
-	ret = intel_huc_wait_for_auth_complete(huc);
+	ret = intel_huc_wait_for_auth_complete(huc, INTEL_HUC_AUTH_BY_GUC);
 	if (ret)
 		goto fail;
 
@@ -469,16 +498,29 @@ int intel_huc_auth(struct intel_huc *huc)
 	return ret;
 }
 
-bool intel_huc_is_authenticated(struct intel_huc *huc)
+bool intel_huc_is_authenticated(struct intel_huc *huc,
+				enum intel_huc_authentication_type type)
 {
 	struct intel_gt *gt = huc_to_gt(huc);
 	intel_wakeref_t wakeref;
 	u32 status = 0;
 
 	with_intel_runtime_pm(gt->uncore->rpm, wakeref)
-		status = intel_uncore_read(gt->uncore, huc->status.reg);
+		status = intel_uncore_read(gt->uncore, huc->status[type].reg);
 
-	return (status & huc->status.mask) == huc->status.value;
+	return (status & huc->status[type].mask) == huc->status[type].value;
+}
+
+static bool huc_is_fully_authenticated(struct intel_huc *huc)
+{
+	struct intel_uc_fw *huc_fw = &huc->fw;
+
+	if (!huc_fw->has_gsc_headers)
+		return intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GUC);
+	else if (intel_huc_is_loaded_by_gsc(huc) || HAS_ENGINE(huc_to_gt(huc), GSC0))
+		return intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC);
+	else
+		return false;
 }
 
 /**
@@ -493,7 +535,9 @@ bool intel_huc_is_authenticated(struct intel_huc *huc)
  */
 int intel_huc_check_status(struct intel_huc *huc)
 {
-	switch (__intel_uc_fw_status(&huc->fw)) {
+	struct intel_uc_fw *huc_fw = &huc->fw;
+
+	switch (__intel_uc_fw_status(huc_fw)) {
 	case INTEL_UC_FIRMWARE_NOT_SUPPORTED:
 		return -ENODEV;
 	case INTEL_UC_FIRMWARE_DISABLED:
@@ -510,7 +554,17 @@ int intel_huc_check_status(struct intel_huc *huc)
 		break;
 	}
 
-	return intel_huc_is_authenticated(huc);
+	/*
+	 * GSC-enabled binaries loaded via DMA are first partially
+	 * authenticated by GuC and then fully authenticated by GSC
+	 */
+	if (huc_is_fully_authenticated(huc))
+		return 1; /* full auth */
+	else if (huc_fw->has_gsc_headers && !intel_huc_is_loaded_by_gsc(huc) &&
+		 intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GUC))
+		return 2; /* clear media only */
+	else
+		return 0;
 }
 
 static bool huc_has_delayed_load(struct intel_huc *huc)
@@ -524,7 +578,10 @@ void intel_huc_update_auth_status(struct intel_huc *huc)
 	if (!intel_uc_fw_is_loadable(&huc->fw))
 		return;
 
-	if (intel_huc_is_authenticated(huc))
+	if (!huc->fw.has_gsc_headers)
+		return;
+
+	if (huc_is_fully_authenticated(huc))
 		intel_uc_fw_change_status(&huc->fw,
 					  INTEL_UC_FIRMWARE_RUNNING);
 	else if (huc_has_delayed_load(huc))
@@ -557,5 +614,5 @@ void intel_huc_load_status(struct intel_huc *huc, struct drm_printer *p)
 
 	with_intel_runtime_pm(gt->uncore->rpm, wakeref)
 		drm_printf(p, "HuC status: 0x%08x\n",
-			   intel_uncore_read(gt->uncore, huc->status.reg));
+			   intel_uncore_read(gt->uncore, huc->status[INTEL_HUC_AUTH_BY_GUC].reg));
 }
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
index 112f0dce47026..3f6aa7c37abcb 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
@@ -22,6 +22,12 @@ enum intel_huc_delayed_load_status {
 	INTEL_HUC_DELAYED_LOAD_ERROR,
 };
 
+enum intel_huc_authentication_type {
+	INTEL_HUC_AUTH_BY_GUC = 0,
+	INTEL_HUC_AUTH_BY_GSC,
+	INTEL_HUC_AUTH_MAX_MODES
+};
+
 struct intel_huc {
 	/* Generic uC firmware management */
 	struct intel_uc_fw fw;
@@ -31,7 +37,7 @@ struct intel_huc {
 		i915_reg_t reg;
 		u32 mask;
 		u32 value;
-	} status;
+	} status[INTEL_HUC_AUTH_MAX_MODES];
 
 	struct {
 		struct i915_sw_fence fence;
@@ -49,10 +55,12 @@ int intel_huc_init(struct intel_huc *huc);
 void intel_huc_fini(struct intel_huc *huc);
 void intel_huc_suspend(struct intel_huc *huc);
 int intel_huc_auth(struct intel_huc *huc);
-int intel_huc_wait_for_auth_complete(struct intel_huc *huc);
+int intel_huc_wait_for_auth_complete(struct intel_huc *huc,
+				     enum intel_huc_authentication_type type);
+bool intel_huc_is_authenticated(struct intel_huc *huc,
+				enum intel_huc_authentication_type type);
 int intel_huc_check_status(struct intel_huc *huc);
 void intel_huc_update_auth_status(struct intel_huc *huc);
-bool intel_huc_is_authenticated(struct intel_huc *huc);
 
 void intel_huc_register_gsc_notifier(struct intel_huc *huc, const struct bus_type *bus);
 void intel_huc_unregister_gsc_notifier(struct intel_huc *huc, const struct bus_type *bus);
@@ -81,7 +89,7 @@ static inline bool intel_huc_is_loaded_by_gsc(const struct intel_huc *huc)
 static inline bool intel_huc_wait_required(struct intel_huc *huc)
 {
 	return intel_huc_is_used(huc) && intel_huc_is_loaded_by_gsc(huc) &&
-	       !intel_huc_is_authenticated(huc);
+	       !intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC);
 }
 
 void intel_huc_load_status(struct intel_huc *huc, struct drm_printer *p);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
index 89a887d33b773..ac2ae5f5011e8 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
@@ -161,7 +161,7 @@ int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc)
 	 * component gets re-bound and this function called again. If so, just
 	 * mark the HuC as loaded.
 	 */
-	if (intel_huc_is_authenticated(huc)) {
+	if (intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC)) {
 		intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_RUNNING);
 		return 0;
 	}
@@ -174,7 +174,7 @@ int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc)
 
 	intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_TRANSFERRED);
 
-	return intel_huc_wait_for_auth_complete(huc);
+	return intel_huc_wait_for_auth_complete(huc, INTEL_HUC_AUTH_BY_GSC);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c4197e31962e1..e0fc0f588c45f 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -941,6 +941,9 @@
 #define HECI_H_GS1(base)	_MMIO((base) + 0xc4c)
 #define   HECI_H_GS1_ER_PREP	REG_BIT(0)
 
+#define HECI_FWSTS5(base)		_MMIO((base) + 0xc68)
+#define   HECI_FWSTS5_HUC_AUTH_DONE	(1 << 19)
+
 #define HSW_GTT_CACHE_EN	_MMIO(0x4024)
 #define   GTT_CACHE_EN_ALL	0xF0007FFF
 #define GEN7_WR_WATERMARK	_MMIO(0x4028)
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
index a217821eb0fbb..f13890ec7db15 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c
@@ -196,7 +196,7 @@ bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp)
 	 * gsc-proxy init flow (the last set of dependencies that
 	 * are out of order) will suffice.
 	 */
-	if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc) &&
+	if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc, INTEL_HUC_AUTH_BY_GSC) &&
 	    intel_gsc_uc_fw_proxy_init_done(&pxp->ctrl_gt->uc.gsc))
 		return true;
 
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index f31dfacde6014..a1848e8060590 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -674,7 +674,8 @@ typedef struct drm_i915_irq_wait {
  * If the IOCTL is successful, the returned parameter will be set to one of the
  * following values:
  *  * 0 if HuC firmware load is not complete,
- *  * 1 if HuC firmware is authenticated and running.
+ *  * 1 if HuC firmware is loaded and fully authenticated,
+ *  * 2 if HuC firmware is loaded and authenticated for clear media only
  */
 #define I915_PARAM_HUC_STATUS		 42
 
-- 
GitLab


From 08872cb13a715e2e963533f18f4aec2bcb145201 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:13 -0700
Subject: [PATCH 0822/1791] drm/i915/mtl/huc: auth HuC via GSC

The full authentication via the GSC requires an heci packet submission
to the GSC FW via the GSC CS. The GSC has new PXP command for this
(literally called NEW_HUC_AUTH).
The intel_huc_auth function is also updated to handle both authentication
types.

v2: check that the GuC auth for clear media has completed before
    proceding with the full auth

v3: use a define for the object size (Alan)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-6-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c     | 27 +++++-
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc.c        | 53 ++++++++---
 drivers/gpu/drm/i915/gt/uc/intel_huc.h        |  6 +-
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c     | 95 +++++++++++++++++++
 drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h     |  1 +
 drivers/gpu/drm/i915/gt/uc/intel_uc.c         |  2 +-
 .../drm/i915/pxp/intel_pxp_cmd_interface_43.h | 17 +++-
 drivers/gpu/drm/i915/pxp/intel_pxp_huc.c      |  2 +-
 9 files changed, 183 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
index b26f493f86fa5..c659cc01f32f6 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c
@@ -29,13 +29,32 @@ static void gsc_work(struct work_struct *work)
 
 	if (actions & GSC_ACTION_FW_LOAD) {
 		ret = intel_gsc_uc_fw_upload(gsc);
-		if (ret == -EEXIST) /* skip proxy if not a new load */
-			actions &= ~GSC_ACTION_FW_LOAD;
-		else if (ret)
+		if (!ret)
+			/* setup proxy on a new load */
+			actions |= GSC_ACTION_SW_PROXY;
+		else if (ret != -EEXIST)
 			goto out_put;
+
+		/*
+		 * The HuC auth can be done both before or after the proxy init;
+		 * if done after, a proxy request will be issued and must be
+		 * serviced before the authentication can complete.
+		 * Since this worker also handles proxy requests, we can't
+		 * perform an action that requires the proxy from within it and
+		 * then stall waiting for it, because we'd be blocking the
+		 * service path. Therefore, it is easier for us to load HuC
+		 * first and do proxy later. The GSC will ack the HuC auth and
+		 * then send the HuC proxy request as part of the proxy init
+		 * flow.
+		 * Note that we can only do the GSC auth if the GuC auth was
+		 * successful.
+		 */
+		if (intel_uc_uses_huc(&gt->uc) &&
+		    intel_huc_is_authenticated(&gt->uc.huc, INTEL_HUC_AUTH_BY_GUC))
+			intel_huc_auth(&gt->uc.huc, INTEL_HUC_AUTH_BY_GSC);
 	}
 
-	if (actions & (GSC_ACTION_FW_LOAD | GSC_ACTION_SW_PROXY)) {
+	if (actions & GSC_ACTION_SW_PROXY) {
 		if (!intel_gsc_uc_fw_init_done(gsc)) {
 			gt_err(gt, "Proxy request received with GSC not loaded!\n");
 			goto out_put;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
index 579c0f5a14384..0ad090304ca09 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
@@ -99,7 +99,7 @@ void intel_gsc_uc_heci_cmd_emit_mtl_header(struct intel_gsc_mtl_header *header,
 					   u64 host_session_id)
 {
 	host_session_id &= ~HOST_SESSION_MASK;
-	if (heci_client_id == HECI_MEADDRESS_PXP)
+	if (host_session_id && heci_client_id == HECI_MEADDRESS_PXP)
 		host_session_id |= HOST_SESSION_PXP_SINGLE;
 
 	header->validity_marker = GSC_HECI_VALIDITY_MARKER;
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index 4c8592a281f20..e0afd8f895026 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -11,6 +11,7 @@
 #include "intel_huc_print.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "pxp/intel_pxp_cmd_interface_43.h"
 
 #include <linux/device/bus.h>
 #include <linux/mei_aux.h>
@@ -371,20 +372,36 @@ static int check_huc_loading_mode(struct intel_huc *huc)
 
 int intel_huc_init(struct intel_huc *huc)
 {
+	struct intel_gt *gt = huc_to_gt(huc);
 	int err;
 
 	err = check_huc_loading_mode(huc);
 	if (err)
 		goto out;
 
+	if (HAS_ENGINE(gt, GSC0)) {
+		struct i915_vma *vma;
+
+		vma = intel_guc_allocate_vma(&gt->uc.guc, PXP43_HUC_AUTH_INOUT_SIZE * 2);
+		if (IS_ERR(vma)) {
+			huc_info(huc, "Failed to allocate heci pkt\n");
+			goto out;
+		}
+
+		huc->heci_pkt = vma;
+	}
+
 	err = intel_uc_fw_init(&huc->fw);
 	if (err)
-		goto out;
+		goto out_pkt;
 
 	intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOADABLE);
 
 	return 0;
 
+out_pkt:
+	if (huc->heci_pkt)
+		i915_vma_unpin_and_release(&huc->heci_pkt, 0);
 out:
 	intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_INIT_FAIL);
 	huc_info(huc, "initialization failed %pe\n", ERR_PTR(err));
@@ -399,6 +416,9 @@ void intel_huc_fini(struct intel_huc *huc)
 	 */
 	delayed_huc_load_fini(huc);
 
+	if (huc->heci_pkt)
+		i915_vma_unpin_and_release(&huc->heci_pkt, 0);
+
 	if (intel_uc_fw_is_loadable(&huc->fw))
 		intel_uc_fw_fini(&huc->fw);
 }
@@ -454,6 +474,7 @@ int intel_huc_wait_for_auth_complete(struct intel_huc *huc,
 /**
  * intel_huc_auth() - Authenticate HuC uCode
  * @huc: intel_huc structure
+ * @type: authentication type (via GuC or via GSC)
  *
  * Called after HuC and GuC firmware loading during intel_uc_init_hw().
  *
@@ -461,7 +482,7 @@ int intel_huc_wait_for_auth_complete(struct intel_huc *huc,
  * passing the offset of the RSA signature to intel_guc_auth_huc(). It then
  * waits for up to 50ms for firmware verification ACK.
  */
-int intel_huc_auth(struct intel_huc *huc)
+int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type)
 {
 	struct intel_gt *gt = huc_to_gt(huc);
 	struct intel_guc *guc = &gt->uc.guc;
@@ -470,31 +491,41 @@ int intel_huc_auth(struct intel_huc *huc)
 	if (!intel_uc_fw_is_loaded(&huc->fw))
 		return -ENOEXEC;
 
-	/* GSC will do the auth */
+	/* GSC will do the auth with the load */
 	if (intel_huc_is_loaded_by_gsc(huc))
 		return -ENODEV;
 
+	if (intel_huc_is_authenticated(huc, type))
+		return -EEXIST;
+
 	ret = i915_inject_probe_error(gt->i915, -ENXIO);
 	if (ret)
 		goto fail;
 
-	GEM_BUG_ON(intel_uc_fw_is_running(&huc->fw));
-
-	ret = intel_guc_auth_huc(guc, intel_guc_ggtt_offset(guc, huc->fw.rsa_data));
-	if (ret) {
-		huc_err(huc, "authentication by GuC failed %pe\n", ERR_PTR(ret));
-		goto fail;
+	switch (type) {
+	case INTEL_HUC_AUTH_BY_GUC:
+		ret = intel_guc_auth_huc(guc, intel_guc_ggtt_offset(guc, huc->fw.rsa_data));
+		break;
+	case INTEL_HUC_AUTH_BY_GSC:
+		ret = intel_huc_fw_auth_via_gsccs(huc);
+		break;
+	default:
+		MISSING_CASE(type);
+		ret = -EINVAL;
 	}
+	if (ret)
+		goto fail;
 
 	/* Check authentication status, it should be done by now */
-	ret = intel_huc_wait_for_auth_complete(huc, INTEL_HUC_AUTH_BY_GUC);
+	ret = intel_huc_wait_for_auth_complete(huc, type);
 	if (ret)
 		goto fail;
 
 	return 0;
 
 fail:
-	huc_probe_error(huc, "authentication failed %pe\n", ERR_PTR(ret));
+	huc_probe_error(huc, "%s authentication failed %pe\n",
+			auth_mode_string(huc, type), ERR_PTR(ret));
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
index 3f6aa7c37abcb..ba5cb08e9e7bf 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h
@@ -15,6 +15,7 @@
 #include <linux/hrtimer.h>
 
 struct bus_type;
+struct i915_vma;
 
 enum intel_huc_delayed_load_status {
 	INTEL_HUC_WAITING_ON_GSC = 0,
@@ -46,6 +47,9 @@ struct intel_huc {
 		enum intel_huc_delayed_load_status status;
 	} delayed_load;
 
+	/* for load via GSCCS */
+	struct i915_vma *heci_pkt;
+
 	bool loaded_via_gsc;
 };
 
@@ -54,7 +58,7 @@ void intel_huc_init_early(struct intel_huc *huc);
 int intel_huc_init(struct intel_huc *huc);
 void intel_huc_fini(struct intel_huc *huc);
 void intel_huc_suspend(struct intel_huc *huc);
-int intel_huc_auth(struct intel_huc *huc);
+int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type);
 int intel_huc_wait_for_auth_complete(struct intel_huc *huc,
 				     enum intel_huc_authentication_type type);
 bool intel_huc_is_authenticated(struct intel_huc *huc,
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
index ac2ae5f5011e8..e608152fecfcc 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c
@@ -6,11 +6,106 @@
 #include "gt/intel_gsc.h"
 #include "gt/intel_gt.h"
 #include "intel_gsc_binary_headers.h"
+#include "intel_gsc_uc_heci_cmd_submit.h"
 #include "intel_huc.h"
 #include "intel_huc_fw.h"
 #include "intel_huc_print.h"
 #include "i915_drv.h"
 #include "pxp/intel_pxp_huc.h"
+#include "pxp/intel_pxp_cmd_interface_43.h"
+
+struct mtl_huc_auth_msg_in {
+	struct intel_gsc_mtl_header header;
+	struct pxp43_new_huc_auth_in huc_in;
+} __packed;
+
+struct mtl_huc_auth_msg_out {
+	struct intel_gsc_mtl_header header;
+	struct pxp43_huc_auth_out huc_out;
+} __packed;
+
+int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc)
+{
+	struct intel_gt *gt = huc_to_gt(huc);
+	struct drm_i915_private *i915 = gt->i915;
+	struct drm_i915_gem_object *obj;
+	struct mtl_huc_auth_msg_in *msg_in;
+	struct mtl_huc_auth_msg_out *msg_out;
+	void *pkt_vaddr;
+	u64 pkt_offset;
+	int retry = 5;
+	int err = 0;
+
+	if (!huc->heci_pkt)
+		return -ENODEV;
+
+	obj = huc->heci_pkt->obj;
+	pkt_offset = i915_ggtt_offset(huc->heci_pkt);
+
+	pkt_vaddr = i915_gem_object_pin_map_unlocked(obj,
+						     i915_coherent_map_type(i915, obj, true));
+	if (IS_ERR(pkt_vaddr))
+		return PTR_ERR(pkt_vaddr);
+
+	msg_in = pkt_vaddr;
+	msg_out = pkt_vaddr + PXP43_HUC_AUTH_INOUT_SIZE;
+
+	intel_gsc_uc_heci_cmd_emit_mtl_header(&msg_in->header,
+					      HECI_MEADDRESS_PXP,
+					      sizeof(*msg_in), 0);
+
+	msg_in->huc_in.header.api_version = PXP_APIVER(4, 3);
+	msg_in->huc_in.header.command_id = PXP43_CMDID_NEW_HUC_AUTH;
+	msg_in->huc_in.header.status = 0;
+	msg_in->huc_in.header.buffer_len = sizeof(msg_in->huc_in) -
+					   sizeof(msg_in->huc_in.header);
+	msg_in->huc_in.huc_base_address = huc->fw.vma_res.start;
+	msg_in->huc_in.huc_size = huc->fw.obj->base.size;
+
+	do {
+		err = intel_gsc_uc_heci_cmd_submit_packet(&gt->uc.gsc,
+							  pkt_offset, sizeof(*msg_in),
+							  pkt_offset + PXP43_HUC_AUTH_INOUT_SIZE,
+							  PXP43_HUC_AUTH_INOUT_SIZE);
+		if (err) {
+			huc_err(huc, "failed to submit GSC request to auth: %d\n", err);
+			goto out_unpin;
+		}
+
+		if (msg_out->header.flags & GSC_OUTFLAG_MSG_PENDING) {
+			msg_in->header.gsc_message_handle = msg_out->header.gsc_message_handle;
+			err = -EBUSY;
+			msleep(50);
+		}
+	} while (--retry && err == -EBUSY);
+
+	if (err)
+		goto out_unpin;
+
+	if (msg_out->header.message_size != sizeof(*msg_out)) {
+		huc_err(huc, "invalid GSC reply length %u [expected %zu]\n",
+			msg_out->header.message_size, sizeof(*msg_out));
+		err = -EPROTO;
+		goto out_unpin;
+	}
+
+	/*
+	 * The GSC will return PXP_STATUS_OP_NOT_PERMITTED if the HuC is already
+	 * loaded. If the same error is ever returned with HuC not loaded we'll
+	 * still catch it when we check the authentication bit later.
+	 */
+	if (msg_out->huc_out.header.status != PXP_STATUS_SUCCESS &&
+	    msg_out->huc_out.header.status != PXP_STATUS_OP_NOT_PERMITTED) {
+		huc_err(huc, "auth failed with GSC error = 0x%x\n",
+			msg_out->huc_out.header.status);
+		err = -EIO;
+		goto out_unpin;
+	}
+
+out_unpin:
+	i915_gem_object_unpin_map(obj);
+	return err;
+}
 
 static void get_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, const void *data)
 {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
index 0999ffe6f9627..307ab45e6b090 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h
@@ -12,6 +12,7 @@ struct intel_uc_fw;
 #include <linux/types.h>
 
 int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc);
+int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc);
 int intel_huc_fw_upload(struct intel_huc *huc);
 int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, size_t size);
 #endif
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
index 1e7f5cc9d5508..18250fb64bd88 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c
@@ -538,7 +538,7 @@ static int __uc_init_hw(struct intel_uc *uc)
 	if (intel_huc_is_loaded_by_gsc(huc))
 		intel_huc_update_auth_status(huc);
 	else
-		intel_huc_auth(huc);
+		intel_huc_auth(huc, INTEL_HUC_AUTH_BY_GUC);
 
 	if (intel_uc_uses_guc_submission(uc)) {
 		ret = intel_guc_submission_enable(guc);
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
index 09777719cd847..0165d38fbead1 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h
@@ -11,19 +11,30 @@
 
 /* PXP-Cmd-Op definitions */
 #define PXP43_CMDID_START_HUC_AUTH 0x0000003A
+#define PXP43_CMDID_NEW_HUC_AUTH 0x0000003F /* MTL+ */
 #define PXP43_CMDID_INIT_SESSION 0x00000036
 
 /* PXP-Packet sizes for MTL's GSCCS-HECI instruction */
 #define PXP43_MAX_HECI_INOUT_SIZE (SZ_32K)
 
-/* PXP-Input-Packet: HUC-Authentication */
+/* PXP-Packet size for MTL's NEW_HUC_AUTH instruction */
+#define PXP43_HUC_AUTH_INOUT_SIZE (SZ_4K)
+
+/* PXP-Input-Packet: HUC Load and Authentication */
 struct pxp43_start_huc_auth_in {
 	struct pxp_cmd_header header;
 	__le64 huc_base_address;
 } __packed;
 
-/* PXP-Output-Packet: HUC-Authentication */
-struct pxp43_start_huc_auth_out {
+/* PXP-Input-Packet: HUC Auth-only */
+struct pxp43_new_huc_auth_in {
+	struct pxp_cmd_header header;
+	u64 huc_base_address;
+	u32 huc_size;
+} __packed;
+
+/* PXP-Output-Packet: HUC Load and Authentication or Auth-only */
+struct pxp43_huc_auth_out {
 	struct pxp_cmd_header header;
 } __packed;
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c b/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c
index 23431c36b60b7..5eedce9169421 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c
@@ -19,7 +19,7 @@ int intel_pxp_huc_load_and_auth(struct intel_pxp *pxp)
 	struct intel_gt *gt;
 	struct intel_huc *huc;
 	struct pxp43_start_huc_auth_in huc_in = {0};
-	struct pxp43_start_huc_auth_out huc_out = {0};
+	struct pxp43_huc_auth_out huc_out = {0};
 	dma_addr_t huc_phys_addr;
 	u8 client_id = 0;
 	u8 fence_id = 0;
-- 
GitLab


From 59c8cd3ef6d04923e74dbef6e8cb4cec4053a99f Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:14 -0700
Subject: [PATCH 0823/1791] drm/i915/mtl/huc: Use the media gt for the HuC
 getparam

On MTL, for obvious reasons, HuC is only available on the media tile.
We already disable SW support for HuC on the root gt due to the
absence of VCS engines, but we also need to update the getparam to point
to the HuC struct in the media GT.

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-7-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/i915_getparam.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c
index 6f11d7eaa91a9..890f2b382bee3 100644
--- a/drivers/gpu/drm/i915/i915_getparam.c
+++ b/drivers/gpu/drm/i915/i915_getparam.c
@@ -100,7 +100,11 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data,
 		value = sseu->min_eu_in_pool;
 		break;
 	case I915_PARAM_HUC_STATUS:
-		value = intel_huc_check_status(&to_gt(i915)->uc.huc);
+		/* On platform with a media GT, the HuC is on that GT */
+		if (i915->media_gt)
+			value = intel_huc_check_status(&i915->media_gt->uc.huc);
+		else
+			value = intel_huc_check_status(&to_gt(i915)->uc.huc);
 		if (value < 0)
 			return value;
 		break;
-- 
GitLab


From 744bbf2a67e7798118c39cd6652a0e1e4ad86498 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Wed, 31 May 2023 16:54:15 -0700
Subject: [PATCH 0824/1791] drm/i915/huc: define HuC FW version for MTL

Follow the same logic as DG2, so just a meu binary with no version number.

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531235415.1467475-8-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
index a1c8a982479fb..944725e624149 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c
@@ -108,6 +108,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw,
 	fw_def(SKYLAKE,      0, guc_mmp(skl,  70, 1, 1))
 
 #define INTEL_HUC_FIRMWARE_DEFS(fw_def, huc_raw, huc_mmp, huc_gsc) \
+	fw_def(METEORLAKE,   0, huc_gsc(mtl)) \
 	fw_def(DG2,          0, huc_gsc(dg2)) \
 	fw_def(ALDERLAKE_P,  0, huc_raw(tgl)) \
 	fw_def(ALDERLAKE_P,  0, huc_mmp(tgl,  7, 9, 3)) \
-- 
GitLab


From 2d0cdf6020549400d93a3f56f09373ece71c76da Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Fri, 2 Jun 2023 11:14:50 -0700
Subject: [PATCH 0825/1791] drm/i915/display: Extract display init from
 intel_device_info_runtime_init

Moving display-specific runtime info initialization into display/ makes
the display code more self-contained and also makes it easier to call
from the Xe driver.

v2:
 - Drop unnecessary display/ prefix from #includes.  (Jani)
 - Clear runtime info if fusing leaves no pipes remaining, the same as
   we do when fusing indicates the entire display controller is
   unavailable.  (Jani)
 - Move adjustment of DRIVER_MODESET / DRIVER_ATOMIC after call to
   intel_display_device_info_runtime_init(); HAS_DISPLAY may have
   changed to false during the runtime init.  (Jani)

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602181450.1151368-1-matthew.d.roper@intel.com
---
 .../drm/i915/display/intel_display_device.c   | 127 +++++++++++++++
 .../drm/i915/display/intel_display_device.h   |   1 +
 drivers/gpu/drm/i915/intel_device_info.c      | 148 ++----------------
 3 files changed, 141 insertions(+), 135 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 1aac7234e1861..3fd30e7f00629 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -9,6 +9,8 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "intel_de.h"
+#include "intel_display.h"
 #include "intel_display_device.h"
 #include "intel_display_power.h"
 #include "intel_display_reg_defs.h"
@@ -788,3 +790,128 @@ intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid,
 
 	return &no_display;
 }
+
+void intel_display_device_info_runtime_init(struct drm_i915_private *i915)
+{
+	struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(i915);
+	enum pipe pipe;
+
+	/* Wa_14011765242: adl-s A0,A1 */
+	if (IS_ADLS_DISPLAY_STEP(i915, STEP_A0, STEP_A2))
+		for_each_pipe(i915, pipe)
+			display_runtime->num_scalers[pipe] = 0;
+	else if (DISPLAY_VER(i915) >= 11) {
+		for_each_pipe(i915, pipe)
+			display_runtime->num_scalers[pipe] = 2;
+	} else if (DISPLAY_VER(i915) >= 9) {
+		display_runtime->num_scalers[PIPE_A] = 2;
+		display_runtime->num_scalers[PIPE_B] = 2;
+		display_runtime->num_scalers[PIPE_C] = 1;
+	}
+
+	if (DISPLAY_VER(i915) >= 13 || HAS_D12_PLANE_MINIMIZATION(i915))
+		for_each_pipe(i915, pipe)
+			display_runtime->num_sprites[pipe] = 4;
+	else if (DISPLAY_VER(i915) >= 11)
+		for_each_pipe(i915, pipe)
+			display_runtime->num_sprites[pipe] = 6;
+	else if (DISPLAY_VER(i915) == 10)
+		for_each_pipe(i915, pipe)
+			display_runtime->num_sprites[pipe] = 3;
+	else if (IS_BROXTON(i915)) {
+		/*
+		 * Skylake and Broxton currently don't expose the topmost plane as its
+		 * use is exclusive with the legacy cursor and we only want to expose
+		 * one of those, not both. Until we can safely expose the topmost plane
+		 * as a DRM_PLANE_TYPE_CURSOR with all the features exposed/supported,
+		 * we don't expose the topmost plane at all to prevent ABI breakage
+		 * down the line.
+		 */
+
+		display_runtime->num_sprites[PIPE_A] = 2;
+		display_runtime->num_sprites[PIPE_B] = 2;
+		display_runtime->num_sprites[PIPE_C] = 1;
+	} else if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) {
+		for_each_pipe(i915, pipe)
+			display_runtime->num_sprites[pipe] = 2;
+	} else if (DISPLAY_VER(i915) >= 5 || IS_G4X(i915)) {
+		for_each_pipe(i915, pipe)
+			display_runtime->num_sprites[pipe] = 1;
+	}
+
+	if ((IS_DGFX(i915) || DISPLAY_VER(i915) >= 14) &&
+	    !(intel_de_read(i915, GU_CNTL_PROTECTED) & DEPRESENT)) {
+		drm_info(&i915->drm, "Display not present, disabling\n");
+		goto display_fused_off;
+	}
+
+	if (IS_GRAPHICS_VER(i915, 7, 8) && HAS_PCH_SPLIT(i915)) {
+		u32 fuse_strap = intel_de_read(i915, FUSE_STRAP);
+		u32 sfuse_strap = intel_de_read(i915, SFUSE_STRAP);
+
+		/*
+		 * SFUSE_STRAP is supposed to have a bit signalling the display
+		 * is fused off. Unfortunately it seems that, at least in
+		 * certain cases, fused off display means that PCH display
+		 * reads don't land anywhere. In that case, we read 0s.
+		 *
+		 * On CPT/PPT, we can detect this case as SFUSE_STRAP_FUSE_LOCK
+		 * should be set when taking over after the firmware.
+		 */
+		if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE ||
+		    sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED ||
+		    (HAS_PCH_CPT(i915) &&
+		     !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) {
+			drm_info(&i915->drm,
+				 "Display fused off, disabling\n");
+			goto display_fused_off;
+		} else if (fuse_strap & IVB_PIPE_C_DISABLE) {
+			drm_info(&i915->drm, "PipeC fused off\n");
+			display_runtime->pipe_mask &= ~BIT(PIPE_C);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
+		}
+	} else if (DISPLAY_VER(i915) >= 9) {
+		u32 dfsm = intel_de_read(i915, SKL_DFSM);
+
+		if (dfsm & SKL_DFSM_PIPE_A_DISABLE) {
+			display_runtime->pipe_mask &= ~BIT(PIPE_A);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A);
+			display_runtime->fbc_mask &= ~BIT(INTEL_FBC_A);
+		}
+		if (dfsm & SKL_DFSM_PIPE_B_DISABLE) {
+			display_runtime->pipe_mask &= ~BIT(PIPE_B);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B);
+		}
+		if (dfsm & SKL_DFSM_PIPE_C_DISABLE) {
+			display_runtime->pipe_mask &= ~BIT(PIPE_C);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
+		}
+
+		if (DISPLAY_VER(i915) >= 12 &&
+		    (dfsm & TGL_DFSM_PIPE_D_DISABLE)) {
+			display_runtime->pipe_mask &= ~BIT(PIPE_D);
+			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D);
+		}
+
+		if (!display_runtime->pipe_mask)
+			goto display_fused_off;
+
+		if (dfsm & SKL_DFSM_DISPLAY_HDCP_DISABLE)
+			display_runtime->has_hdcp = 0;
+
+		if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE)
+			display_runtime->fbc_mask = 0;
+
+		if (DISPLAY_VER(i915) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE))
+			display_runtime->has_dmc = 0;
+
+		if (IS_DISPLAY_VER(i915, 10, 12) &&
+		    (dfsm & GLK_DFSM_DISPLAY_DSC_DISABLE))
+			display_runtime->has_dsc = 0;
+	}
+
+	return;
+
+display_fused_off:
+	memset(display_runtime, 0, sizeof(*display_runtime));
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 2aa82cbdf1c53..4f931258d81db 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -124,5 +124,6 @@ struct intel_display_device_info {
 const struct intel_display_device_info *
 intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid,
 			   u16 *ver, u16 *rel, u16 *step);
+void intel_display_device_info_runtime_init(struct drm_i915_private *i915);
 
 #endif
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 2f79d232b04a9..6e49caf241a5f 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -27,9 +27,7 @@
 #include <drm/drm_print.h>
 #include <drm/i915_pciids.h>
 
-#include "display/intel_cdclk.h"
-#include "display/intel_de.h"
-#include "display/intel_display.h"
+#include "display/intel_display_device.h"
 #include "gt/intel_gt_regs.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
@@ -411,125 +409,23 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 {
 	struct intel_device_info *info = mkwrite_device_info(dev_priv);
 	struct intel_runtime_info *runtime = RUNTIME_INFO(dev_priv);
-	struct intel_display_runtime_info *display_runtime =
-		DISPLAY_RUNTIME_INFO(dev_priv);
-	enum pipe pipe;
-
-	/* Wa_14011765242: adl-s A0,A1 */
-	if (IS_ADLS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_A2))
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_scalers[pipe] = 0;
-	else if (DISPLAY_VER(dev_priv) >= 11) {
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_scalers[pipe] = 2;
-	} else if (DISPLAY_VER(dev_priv) >= 9) {
-		display_runtime->num_scalers[PIPE_A] = 2;
-		display_runtime->num_scalers[PIPE_B] = 2;
-		display_runtime->num_scalers[PIPE_C] = 1;
-	}
-
-	BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES);
 
-	if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv))
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_sprites[pipe] = 4;
-	else if (DISPLAY_VER(dev_priv) >= 11)
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_sprites[pipe] = 6;
-	else if (DISPLAY_VER(dev_priv) == 10)
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_sprites[pipe] = 3;
-	else if (IS_BROXTON(dev_priv)) {
-		/*
-		 * Skylake and Broxton currently don't expose the topmost plane as its
-		 * use is exclusive with the legacy cursor and we only want to expose
-		 * one of those, not both. Until we can safely expose the topmost plane
-		 * as a DRM_PLANE_TYPE_CURSOR with all the features exposed/supported,
-		 * we don't expose the topmost plane at all to prevent ABI breakage
-		 * down the line.
-		 */
+	if (HAS_DISPLAY(dev_priv))
+		intel_display_device_info_runtime_init(dev_priv);
 
-		display_runtime->num_sprites[PIPE_A] = 2;
-		display_runtime->num_sprites[PIPE_B] = 2;
-		display_runtime->num_sprites[PIPE_C] = 1;
-	} else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_sprites[pipe] = 2;
-	} else if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) {
-		for_each_pipe(dev_priv, pipe)
-			display_runtime->num_sprites[pipe] = 1;
-	}
-
-	if (HAS_DISPLAY(dev_priv) &&
-	    (IS_DGFX(dev_priv) || DISPLAY_VER(dev_priv) >= 14) &&
-	    !(intel_de_read(dev_priv, GU_CNTL_PROTECTED) & DEPRESENT)) {
-		drm_info(&dev_priv->drm, "Display not present, disabling\n");
-
-		display_runtime->pipe_mask = 0;
+	/* Display may have been disabled by runtime init */
+	if (!HAS_DISPLAY(dev_priv)) {
+		dev_priv->drm.driver_features &= ~(DRIVER_MODESET |
+						   DRIVER_ATOMIC);
+		info->display = &no_display;
 	}
 
-	if (HAS_DISPLAY(dev_priv) && IS_GRAPHICS_VER(dev_priv, 7, 8) &&
-	    HAS_PCH_SPLIT(dev_priv)) {
-		u32 fuse_strap = intel_de_read(dev_priv, FUSE_STRAP);
-		u32 sfuse_strap = intel_de_read(dev_priv, SFUSE_STRAP);
-
-		/*
-		 * SFUSE_STRAP is supposed to have a bit signalling the display
-		 * is fused off. Unfortunately it seems that, at least in
-		 * certain cases, fused off display means that PCH display
-		 * reads don't land anywhere. In that case, we read 0s.
-		 *
-		 * On CPT/PPT, we can detect this case as SFUSE_STRAP_FUSE_LOCK
-		 * should be set when taking over after the firmware.
-		 */
-		if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE ||
-		    sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED ||
-		    (HAS_PCH_CPT(dev_priv) &&
-		     !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) {
-			drm_info(&dev_priv->drm,
-				 "Display fused off, disabling\n");
-			display_runtime->pipe_mask = 0;
-		} else if (fuse_strap & IVB_PIPE_C_DISABLE) {
-			drm_info(&dev_priv->drm, "PipeC fused off\n");
-			display_runtime->pipe_mask &= ~BIT(PIPE_C);
-			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
-		}
-	} else if (HAS_DISPLAY(dev_priv) && DISPLAY_VER(dev_priv) >= 9) {
-		u32 dfsm = intel_de_read(dev_priv, SKL_DFSM);
-
-		if (dfsm & SKL_DFSM_PIPE_A_DISABLE) {
-			display_runtime->pipe_mask &= ~BIT(PIPE_A);
-			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A);
-			display_runtime->fbc_mask &= ~BIT(INTEL_FBC_A);
-		}
-		if (dfsm & SKL_DFSM_PIPE_B_DISABLE) {
-			display_runtime->pipe_mask &= ~BIT(PIPE_B);
-			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B);
-		}
-		if (dfsm & SKL_DFSM_PIPE_C_DISABLE) {
-			display_runtime->pipe_mask &= ~BIT(PIPE_C);
-			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C);
-		}
-
-		if (DISPLAY_VER(dev_priv) >= 12 &&
-		    (dfsm & TGL_DFSM_PIPE_D_DISABLE)) {
-			display_runtime->pipe_mask &= ~BIT(PIPE_D);
-			display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D);
-		}
-
-		if (dfsm & SKL_DFSM_DISPLAY_HDCP_DISABLE)
-			display_runtime->has_hdcp = 0;
-
-		if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE)
-			display_runtime->fbc_mask = 0;
-
-		if (DISPLAY_VER(dev_priv) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE))
-			display_runtime->has_dmc = 0;
+	/* Disable nuclear pageflip by default on pre-g4x */
+	if (!dev_priv->params.nuclear_pageflip &&
+	    DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv))
+		dev_priv->drm.driver_features &= ~DRIVER_ATOMIC;
 
-		if (IS_DISPLAY_VER(dev_priv, 10, 12) &&
-		    (dfsm & GLK_DFSM_DISPLAY_DSC_DISABLE))
-			display_runtime->has_dsc = 0;
-	}
+	BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES);
 
 	if (GRAPHICS_VER(dev_priv) == 6 && i915_vtd_active(dev_priv)) {
 		drm_info(&dev_priv->drm,
@@ -540,24 +436,6 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv)
 	runtime->rawclk_freq = intel_read_rawclk(dev_priv);
 	drm_dbg(&dev_priv->drm, "rawclk rate: %d kHz\n", runtime->rawclk_freq);
 
-	if (!HAS_DISPLAY(dev_priv)) {
-		dev_priv->drm.driver_features &= ~(DRIVER_MODESET |
-						   DRIVER_ATOMIC);
-		info->display = &no_display;
-
-		display_runtime->cpu_transcoder_mask = 0;
-		memset(display_runtime->num_sprites, 0, sizeof(display_runtime->num_sprites));
-		memset(display_runtime->num_scalers, 0, sizeof(display_runtime->num_scalers));
-		display_runtime->fbc_mask = 0;
-		display_runtime->has_hdcp = false;
-		display_runtime->has_dmc = false;
-		display_runtime->has_dsc = false;
-	}
-
-	/* Disable nuclear pageflip by default on pre-g4x */
-	if (!dev_priv->params.nuclear_pageflip &&
-	    DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv))
-		dev_priv->drm.driver_features &= ~DRIVER_ATOMIC;
 }
 
 /*
-- 
GitLab


From 718551bbed3ca5308a9f9429305dd074727e8d46 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 1 Jun 2023 10:38:04 -0700
Subject: [PATCH 0826/1791] drm/i915: Fix error handling if driver creation
 fails during probe

If i915_driver_create() fails to create a valid 'i915' object, we
should just disable the PCI device and return immediately without trying
to call i915_probe_error() that relies on a valid i915 pointer.

Fixes: 12e6f6dc78e4 ("drm/i915/display: Handle GMD_ID identification in display code")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/all/55236f93-dcc5-481e-b788-9f7e95b129d8@kili.mountain/
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230601173804.557756-1-matthew.d.roper@intel.com
---
 drivers/gpu/drm/i915/i915_driver.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 8e92649124d4c..4483be16a00b1 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -762,8 +762,8 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	i915 = i915_driver_create(pdev, ent);
 	if (IS_ERR(i915)) {
-		ret = PTR_ERR(i915);
-		goto out_pci_disable;
+		pci_disable_device(pdev);
+		return PTR_ERR(i915);
 	}
 
 	ret = i915_driver_early_probe(i915);
-- 
GitLab


From e894b724c316d9b45d963f929c97578cdffbbe8d Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 5 Jun 2023 13:32:24 +0100
Subject: [PATCH 0827/1791] drm/i915: Use the fdinfo helper

Use the common fdinfo helper for printing the basics. Remove now unused
client id allocation code.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Rob Clark <robdclark@chromium.org>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230605123224.373633-1-tvrtko.ursulin@linux.intel.com
---
 drivers/gpu/drm/i915/i915_driver.c     |  6 +--
 drivers/gpu/drm/i915/i915_drm_client.c | 65 ++++----------------------
 drivers/gpu/drm/i915/i915_drm_client.h | 22 ++-------
 drivers/gpu/drm/i915/i915_drv.h        |  2 -
 drivers/gpu/drm/i915/i915_gem.c        |  6 +--
 5 files changed, 18 insertions(+), 83 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 93fdc40d724fa..3adefbae7fb57 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -243,8 +243,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 	if (ret < 0)
 		goto err_rootgt;
 
-	i915_drm_clients_init(&dev_priv->clients, dev_priv);
-
 	i915_gem_init_early(dev_priv);
 
 	/* This must be called before any calls to HAS_PCH_* */
@@ -278,7 +276,6 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv)
 	intel_power_domains_cleanup(dev_priv);
 	i915_gem_cleanup_early(dev_priv);
 	intel_gt_driver_late_release_all(dev_priv);
-	i915_drm_clients_fini(&dev_priv->clients);
 	intel_region_ttm_device_fini(dev_priv);
 	vlv_suspend_cleanup(dev_priv);
 	i915_workqueues_cleanup(dev_priv);
@@ -1706,7 +1703,7 @@ static const struct file_operations i915_driver_fops = {
 	.compat_ioctl = i915_ioc32_compat_ioctl,
 	.llseek = noop_llseek,
 #ifdef CONFIG_PROC_FS
-	.show_fdinfo = i915_drm_client_fdinfo,
+	.show_fdinfo = drm_show_fdinfo,
 #endif
 };
 
@@ -1806,6 +1803,7 @@ static const struct drm_driver i915_drm_driver = {
 	.open = i915_driver_open,
 	.lastclose = i915_driver_lastclose,
 	.postclose = i915_driver_postclose,
+	.show_fdinfo = i915_drm_client_fdinfo,
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
diff --git a/drivers/gpu/drm/i915/i915_drm_client.c b/drivers/gpu/drm/i915/i915_drm_client.c
index d18d0a3ed905b..2a44b3876cb51 100644
--- a/drivers/gpu/drm/i915/i915_drm_client.c
+++ b/drivers/gpu/drm/i915/i915_drm_client.c
@@ -17,64 +17,29 @@
 #include "i915_gem.h"
 #include "i915_utils.h"
 
-void i915_drm_clients_init(struct i915_drm_clients *clients,
-			   struct drm_i915_private *i915)
-{
-	clients->i915 = i915;
-	clients->next_id = 0;
-
-	xa_init_flags(&clients->xarray, XA_FLAGS_ALLOC | XA_FLAGS_LOCK_IRQ);
-}
-
-struct i915_drm_client *i915_drm_client_add(struct i915_drm_clients *clients)
+struct i915_drm_client *i915_drm_client_alloc(void)
 {
 	struct i915_drm_client *client;
-	struct xarray *xa = &clients->xarray;
-	int ret;
 
 	client = kzalloc(sizeof(*client), GFP_KERNEL);
 	if (!client)
-		return ERR_PTR(-ENOMEM);
-
-	xa_lock_irq(xa);
-	ret = __xa_alloc_cyclic(xa, &client->id, client, xa_limit_32b,
-				&clients->next_id, GFP_KERNEL);
-	xa_unlock_irq(xa);
-	if (ret < 0)
-		goto err;
+		return NULL;
 
 	kref_init(&client->kref);
 	spin_lock_init(&client->ctx_lock);
 	INIT_LIST_HEAD(&client->ctx_list);
-	client->clients = clients;
 
 	return client;
-
-err:
-	kfree(client);
-
-	return ERR_PTR(ret);
 }
 
 void __i915_drm_client_free(struct kref *kref)
 {
 	struct i915_drm_client *client =
 		container_of(kref, typeof(*client), kref);
-	struct xarray *xa = &client->clients->xarray;
-	unsigned long flags;
 
-	xa_lock_irqsave(xa, flags);
-	__xa_erase(xa, client->id);
-	xa_unlock_irqrestore(xa, flags);
 	kfree(client);
 }
 
-void i915_drm_clients_fini(struct i915_drm_clients *clients)
-{
-	GEM_BUG_ON(!xa_empty(&clients->xarray));
-	xa_destroy(&clients->xarray);
-}
-
 #ifdef CONFIG_PROC_FS
 static const char * const uabi_class_names[] = {
 	[I915_ENGINE_CLASS_RENDER] = "render",
@@ -101,38 +66,34 @@ static u64 busy_add(struct i915_gem_context *ctx, unsigned int class)
 }
 
 static void
-show_client_class(struct seq_file *m,
+show_client_class(struct drm_printer *p,
+		  struct drm_i915_private *i915,
 		  struct i915_drm_client *client,
 		  unsigned int class)
 {
-	const struct list_head *list = &client->ctx_list;
+	const unsigned int capacity = i915->engine_uabi_class_count[class];
 	u64 total = atomic64_read(&client->past_runtime[class]);
-	const unsigned int capacity =
-		client->clients->i915->engine_uabi_class_count[class];
 	struct i915_gem_context *ctx;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(ctx, list, client_link)
+	list_for_each_entry_rcu(ctx, &client->ctx_list, client_link)
 		total += busy_add(ctx, class);
 	rcu_read_unlock();
 
 	if (capacity)
-		seq_printf(m, "drm-engine-%s:\t%llu ns\n",
+		drm_printf(p, "drm-engine-%s:\t%llu ns\n",
 			   uabi_class_names[class], total);
 
 	if (capacity > 1)
-		seq_printf(m, "drm-engine-capacity-%s:\t%u\n",
+		drm_printf(p, "drm-engine-capacity-%s:\t%u\n",
 			   uabi_class_names[class],
 			   capacity);
 }
 
-void i915_drm_client_fdinfo(struct seq_file *m, struct file *f)
+void i915_drm_client_fdinfo(struct drm_printer *p, struct drm_file *file)
 {
-	struct drm_file *file = f->private_data;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 	struct drm_i915_private *i915 = file_priv->i915;
-	struct i915_drm_client *client = file_priv->client;
-	struct pci_dev *pdev = to_pci_dev(i915->drm.dev);
 	unsigned int i;
 
 	/*
@@ -141,16 +102,10 @@ void i915_drm_client_fdinfo(struct seq_file *m, struct file *f)
 	 * ******************************************************************
 	 */
 
-	seq_printf(m, "drm-driver:\t%s\n", i915->drm.driver->name);
-	seq_printf(m, "drm-pdev:\t%04x:%02x:%02x.%d\n",
-		   pci_domain_nr(pdev->bus), pdev->bus->number,
-		   PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn));
-	seq_printf(m, "drm-client-id:\t%u\n", client->id);
-
 	if (GRAPHICS_VER(i915) < 8)
 		return;
 
 	for (i = 0; i < ARRAY_SIZE(uabi_class_names); i++)
-		show_client_class(m, client, i);
+		show_client_class(p, i915, file_priv->client, i);
 }
 #endif
diff --git a/drivers/gpu/drm/i915/i915_drm_client.h b/drivers/gpu/drm/i915/i915_drm_client.h
index 69496af996d9e..4c18b99e10a4e 100644
--- a/drivers/gpu/drm/i915/i915_drm_client.h
+++ b/drivers/gpu/drm/i915/i915_drm_client.h
@@ -9,20 +9,13 @@
 #include <linux/kref.h>
 #include <linux/list.h>
 #include <linux/spinlock.h>
-#include <linux/xarray.h>
 
 #include <uapi/drm/i915_drm.h>
 
 #define I915_LAST_UABI_ENGINE_CLASS I915_ENGINE_CLASS_COMPUTE
 
-struct drm_i915_private;
-
-struct i915_drm_clients {
-	struct drm_i915_private *i915;
-
-	struct xarray xarray;
-	u32 next_id;
-};
+struct drm_file;
+struct drm_printer;
 
 struct i915_drm_client {
 	struct kref kref;
@@ -32,17 +25,12 @@ struct i915_drm_client {
 	spinlock_t ctx_lock; /* For add/remove from ctx_list. */
 	struct list_head ctx_list; /* List of contexts belonging to client. */
 
-	struct i915_drm_clients *clients;
-
 	/**
 	 * @past_runtime: Accumulation of pphwsp runtimes from closed contexts.
 	 */
 	atomic64_t past_runtime[I915_LAST_UABI_ENGINE_CLASS + 1];
 };
 
-void i915_drm_clients_init(struct i915_drm_clients *clients,
-			   struct drm_i915_private *i915);
-
 static inline struct i915_drm_client *
 i915_drm_client_get(struct i915_drm_client *client)
 {
@@ -57,12 +45,10 @@ static inline void i915_drm_client_put(struct i915_drm_client *client)
 	kref_put(&client->kref, __i915_drm_client_free);
 }
 
-struct i915_drm_client *i915_drm_client_add(struct i915_drm_clients *clients);
+struct i915_drm_client *i915_drm_client_alloc(void);
 
 #ifdef CONFIG_PROC_FS
-void i915_drm_client_fdinfo(struct seq_file *m, struct file *f);
+void i915_drm_client_fdinfo(struct drm_printer *p, struct drm_file *file);
 #endif
 
-void i915_drm_clients_fini(struct i915_drm_clients *clients);
-
 #endif /* !__I915_DRM_CLIENT_H__ */
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index f23b030aaf09f..d24e7dc508085 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -347,8 +347,6 @@ struct drm_i915_private {
 
 	struct i915_pmu pmu;
 
-	struct i915_drm_clients clients;
-
 	/* The TTM device structure. */
 	struct ttm_device bdev;
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index e70b762f0b032..1f65bb33dd212 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1325,11 +1325,9 @@ int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file)
 	if (!file_priv)
 		goto err_alloc;
 
-	client = i915_drm_client_add(&i915->clients);
-	if (IS_ERR(client)) {
-		ret = PTR_ERR(client);
+	client = i915_drm_client_alloc();
+	if (!client)
 		goto err_client;
-	}
 
 	file->driver_priv = file_priv;
 	file_priv->i915 = i915;
-- 
GitLab


From 412fa1f097f48c8c1321806dd25e46618e0da147 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 5 Jun 2023 14:11:35 +0100
Subject: [PATCH 0828/1791] drm/i915/selftests: Add some missing error
 propagation

Add some missing error propagation in live_parallel_switch.

To avoid needlessly burdening the various backport processes, note I am
not marking it as a fix against any patches and not copying stable since
it is debug/selftests only code.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Fixes: 50d16d44cce4 ("drm/i915/selftests: Exercise context switching in parallel")
Fixes: 6407cf533217 ("drm/i915/selftests: Stop using kthread_stop()")
Link: https://patchwork.freedesktop.org/patch/msgid/20230605131135.396854-1-tvrtko.ursulin@linux.intel.com
---
 .../gpu/drm/i915/gem/selftests/i915_gem_context.c  | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
index ad6a3b2fb387b..7021b6e9b219e 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c
@@ -348,8 +348,10 @@ static int live_parallel_switch(void *arg)
 				continue;
 
 			ce = intel_context_create(data[m].ce[0]->engine);
-			if (IS_ERR(ce))
+			if (IS_ERR(ce)) {
+				err = PTR_ERR(ce);
 				goto out;
+			}
 
 			err = intel_context_pin(ce);
 			if (err) {
@@ -369,8 +371,10 @@ static int live_parallel_switch(void *arg)
 
 		worker = kthread_create_worker(0, "igt/parallel:%s",
 					       data[n].ce[0]->engine->name);
-		if (IS_ERR(worker))
+		if (IS_ERR(worker)) {
+			err = PTR_ERR(worker);
 			goto out;
+		}
 
 		data[n].worker = worker;
 	}
@@ -399,8 +403,10 @@ static int live_parallel_switch(void *arg)
 			}
 		}
 
-		if (igt_live_test_end(&t))
-			err = -EIO;
+		if (igt_live_test_end(&t)) {
+			err = err ?: -EIO;
+			break;
+		}
 	}
 
 out:
-- 
GitLab


From cd65de1ab2fe4b99ec20861182b7c24d26559f48 Mon Sep 17 00:00:00 2001
From: Matt Atwood <matthew.s.atwood@intel.com>
Date: Fri, 2 Jun 2023 16:17:53 -0700
Subject: [PATCH 0829/1791] drm/i915: Reduce I915_MAX_GT to 2

According to Ashutosh there is no current or planned product in i915 for
I915_MAX_GT to be 4 anymore.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Ashutosh Dixit <ashutosh.dixit@linux.intel.com>
Cc: Andi Shyti <andy.shyti@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Signed-off-by: Matt Atwood <matthew.s.atwood@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602231754.1596433-2-matthew.s.atwood@intel.com
---
 drivers/gpu/drm/i915/i915_drv.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index d24e7dc508085..a3e3c34141ed4 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -313,7 +313,7 @@ struct drm_i915_private {
 	/*
 	 * i915->gt[0] == &i915->gt0
 	 */
-#define I915_MAX_GT 4
+#define I915_MAX_GT 2
 	struct intel_gt *gt[I915_MAX_GT];
 
 	struct kobject *sysfs_gt;
-- 
GitLab


From 419491eaf1ee90b83c260c32b6c29b1d96c15ce8 Mon Sep 17 00:00:00 2001
From: Matt Atwood <matthew.s.atwood@intel.com>
Date: Fri, 2 Jun 2023 16:17:54 -0700
Subject: [PATCH 0830/1791] drm/i915: rename I915_PMU_MAX_GTS to
 I915_PMU_MAX_GT

_GTS as an abbreviation here leads to some confusion, match other
definitions and drop the s.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Ashutosh Dixit <ashutosh.dixit@linux.intel.com>
Cc: Andi Shyti <andy.shyti@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Signed-off-by: Matt Atwood <matthew.s.atwood@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602231754.1596433-3-matthew.s.atwood@intel.com
---
 drivers/gpu/drm/i915/i915_pmu.c | 2 +-
 drivers/gpu/drm/i915/i915_pmu.h | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index f96fe92dca4e4..d35973b411863 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -132,7 +132,7 @@ static u32 frequency_enabled_mask(void)
 	unsigned int i;
 	u32 mask = 0;
 
-	for (i = 0; i < I915_PMU_MAX_GTS; i++)
+	for (i = 0; i < I915_PMU_MAX_GT; i++)
 		mask |= config_mask(__I915_PMU_ACTUAL_FREQUENCY(i)) |
 			config_mask(__I915_PMU_REQUESTED_FREQUENCY(i));
 
diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h
index d20592e7db999..41af038c37388 100644
--- a/drivers/gpu/drm/i915/i915_pmu.h
+++ b/drivers/gpu/drm/i915/i915_pmu.h
@@ -38,7 +38,7 @@ enum {
 	__I915_NUM_PMU_SAMPLERS
 };
 
-#define I915_PMU_MAX_GTS 2
+#define I915_PMU_MAX_GT 2
 
 /*
  * How many different events we track in the global PMU mask.
@@ -47,7 +47,7 @@ enum {
  */
 #define I915_PMU_MASK_BITS \
 	(I915_ENGINE_SAMPLE_COUNT + \
-	 I915_PMU_MAX_GTS * __I915_PMU_TRACKED_EVENT_COUNT)
+	 I915_PMU_MAX_GT * __I915_PMU_TRACKED_EVENT_COUNT)
 
 #define I915_ENGINE_SAMPLE_COUNT (I915_SAMPLE_SEMA + 1)
 
@@ -127,11 +127,11 @@ struct i915_pmu {
 	 * Only global counters are held here, while the per-engine ones are in
 	 * struct intel_engine_cs.
 	 */
-	struct i915_pmu_sample sample[I915_PMU_MAX_GTS][__I915_NUM_PMU_SAMPLERS];
+	struct i915_pmu_sample sample[I915_PMU_MAX_GT][__I915_NUM_PMU_SAMPLERS];
 	/**
 	 * @sleep_last: Last time GT parked for RC6 estimation.
 	 */
-	ktime_t sleep_last[I915_PMU_MAX_GTS];
+	ktime_t sleep_last[I915_PMU_MAX_GT];
 	/**
 	 * @irq_count: Number of interrupts
 	 *
-- 
GitLab


From 1b842f73edbb1f8513ef4a89389176c628dc014a Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 31 May 2023 08:59:42 -0700
Subject: [PATCH 0831/1791] drm/i915/guc: Remove some obsolete definitions

There were a bunch of defines and structures left over from an API
update a very long time ago. Remove them.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531155942.441862-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h | 33 ---------------------
 1 file changed, 33 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
index 4e57bd09d50d9..b4d56eccfb1f0 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h
@@ -35,13 +35,6 @@
 #define GUC_MAX_CONTEXT_ID		65535
 #define	GUC_INVALID_CONTEXT_ID		GUC_MAX_CONTEXT_ID
 
-#define GUC_RENDER_ENGINE		0
-#define GUC_VIDEO_ENGINE		1
-#define GUC_BLITTER_ENGINE		2
-#define GUC_VIDEOENHANCE_ENGINE		3
-#define GUC_VIDEO_ENGINE2		4
-#define GUC_MAX_ENGINES_NUM		(GUC_VIDEO_ENGINE2 + 1)
-
 #define GUC_RENDER_CLASS		0
 #define GUC_VIDEO_CLASS			1
 #define GUC_VIDEOENHANCE_CLASS		2
@@ -499,32 +492,6 @@ struct guc_log_buffer_state {
 	u32 version;
 } __packed;
 
-struct guc_ctx_report {
-	u32 report_return_status;
-	u32 reserved1[64];
-	u32 affected_count;
-	u32 reserved2[2];
-} __packed;
-
-/* GuC Shared Context Data Struct */
-struct guc_shared_ctx_data {
-	u32 addr_of_last_preempted_data_low;
-	u32 addr_of_last_preempted_data_high;
-	u32 addr_of_last_preempted_data_high_tmp;
-	u32 padding;
-	u32 is_mapped_to_proxy;
-	u32 proxy_ctx_id;
-	u32 engine_reset_ctx_id;
-	u32 media_reset_count;
-	u32 reserved1[8];
-	u32 uk_last_ctx_switch_reason;
-	u32 was_reset;
-	u32 lrca_gpu_addr;
-	u64 execlist_ctx;
-	u32 reserved2[66];
-	struct guc_ctx_report preempt_ctx_report[GUC_MAX_ENGINES_NUM];
-} __packed;
-
 /* This action will be programmed in C1BC - SOFT_SCRATCH_15_REG */
 enum intel_guc_recv_message {
 	INTEL_GUC_RECV_MSG_CRASH_DUMP_POSTED = BIT(1),
-- 
GitLab


From c42a37a27c777d63961dd634a30f7c887949491a Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Wed, 7 Jun 2023 08:31:37 +0200
Subject: [PATCH 0832/1791] dt-bindings: display: simple: add Rocktech
 RK043FN48H

Add compatible to panel-simple for Rocktech Displays Limited
RK043FN48H 4.3" 480x272 LCD-TFT panel.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Reviewed-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230607063139.621351-5-dario.binacchi@amarulasolutions.com
---
 .../devicetree/bindings/display/panel/panel-simple.yaml         | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index f7c45e38d4b75..1d4936fc51828 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -286,6 +286,8 @@ properties:
       - rocktech,rk101ii01d-ct
         # Rocktech Display Ltd. RK070ER9427 800(RGB)x480 TFT LCD panel
       - rocktech,rk070er9427
+        # Rocktech Display Ltd. RK043FN48H 4.3" 480x272 LCD-TFT panel
+      - rocktech,rk043fn48h
         # Samsung 13.3" FHD (1920x1080 pixels) eDP AMOLED panel
       - samsung,atna33xc20
         # Samsung 12.2" (2560x1600 pixels) TFT LCD panel
-- 
GitLab


From 13cdd12a9f934158f4ec817cf048fcb4384aa9dc Mon Sep 17 00:00:00 2001
From: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Date: Wed, 7 Jun 2023 08:31:38 +0200
Subject: [PATCH 0833/1791] drm/panel: simple: add support for Rocktech
 RK043FN48H panel

Add support for Rocktech RK043FN48H 4.3" (480x272) LCD-TFT panel.

Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202306020343.jNTWeM0P-lkp@intel.com/
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230607063139.621351-6-dario.binacchi@amarulasolutions.com
---
 drivers/gpu/drm/panel/panel-simple.c | 29 ++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 754c09c6d4c1b..a247a0e7c7998 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -3281,6 +3281,32 @@ static const struct panel_desc qishenglong_gopher2b_lcd = {
 	.connector_type = DRM_MODE_CONNECTOR_DPI,
 };
 
+static const struct display_timing rocktech_rk043fn48h_timing = {
+	.pixelclock = { 6000000, 9000000, 12000000 },
+	.hactive = { 480, 480, 480 },
+	.hback_porch = { 8, 43, 43 },
+	.hfront_porch = { 2, 8, 8 },
+	.hsync_len = { 1, 1, 1 },
+	.vactive = { 272, 272, 272 },
+	.vback_porch = { 2, 12, 12 },
+	.vfront_porch = { 1, 4, 4 },
+	.vsync_len = { 1, 10, 10 },
+	.flags = DISPLAY_FLAGS_VSYNC_LOW | DISPLAY_FLAGS_HSYNC_LOW |
+		 DISPLAY_FLAGS_DE_HIGH | DISPLAY_FLAGS_PIXDATA_POSEDGE,
+};
+
+static const struct panel_desc rocktech_rk043fn48h = {
+	.timings = &rocktech_rk043fn48h_timing,
+	.num_timings = 1,
+	.bpc = 8,
+	.size = {
+		.width = 95,
+		.height = 54,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+	.connector_type = DRM_MODE_CONNECTOR_DPI,
+};
+
 static const struct display_timing rocktech_rk070er9427_timing = {
 	.pixelclock = { 26400000, 33300000, 46800000 },
 	.hactive = { 800, 800, 800 },
@@ -4320,6 +4346,9 @@ static const struct of_device_id platform_of_match[] = {
 	}, {
 		.compatible = "qishenglong,gopher2b-lcd",
 		.data = &qishenglong_gopher2b_lcd,
+	}, {
+		.compatible = "rocktech,rk043fn48h",
+		.data = &rocktech_rk043fn48h,
 	}, {
 		.compatible = "rocktech,rk070er9427",
 		.data = &rocktech_rk070er9427,
-- 
GitLab


From 8690be05cf055e0c68a7e151530eea8bda0e601a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 2 Jun 2023 16:23:21 +0300
Subject: [PATCH 0834/1791] drm/i915/debugfs: stop using edid_blob_ptr
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Only the EDID code and sysfs should look at the EDID property. Stop
using it.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230602132321.3199445-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_debugfs.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index 2a4df62692a68..165e2c7e3126a 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -6,6 +6,7 @@
 #include <linux/string_helpers.h>
 
 #include <drm/drm_debugfs.h>
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 
 #include "hsw_ips.h"
@@ -228,19 +229,18 @@ static void intel_hdcp_info(struct seq_file *m,
 	seq_puts(m, "\n");
 }
 
-static void intel_dp_info(struct seq_file *m,
-			  struct intel_connector *intel_connector)
+static void intel_dp_info(struct seq_file *m, struct intel_connector *connector)
 {
-	struct intel_encoder *intel_encoder = intel_attached_encoder(intel_connector);
+	struct intel_encoder *intel_encoder = intel_attached_encoder(connector);
 	struct intel_dp *intel_dp = enc_to_intel_dp(intel_encoder);
-	const struct drm_property_blob *edid = intel_connector->base.edid_blob_ptr;
+	const struct edid *edid = drm_edid_raw(connector->detect_edid);
 
 	seq_printf(m, "\tDPCD rev: %x\n", intel_dp->dpcd[DP_DPCD_REV]);
 	seq_printf(m, "\taudio support: %s\n",
-		   str_yes_no(intel_connector->base.display_info.has_audio));
+		   str_yes_no(connector->base.display_info.has_audio));
 
 	drm_dp_downstream_debug(m, intel_dp->dpcd, intel_dp->downstream_ports,
-				edid ? edid->data : NULL, &intel_dp->aux);
+				edid, &intel_dp->aux);
 }
 
 static void intel_dp_mst_info(struct seq_file *m,
-- 
GitLab


From 47e157a5b49a4c099daf202207adc9815a2f2916 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 20:22:18 +0300
Subject: [PATCH 0835/1791] drm/i915/plane: warn on non-zero plane offset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We assume the plane offset is 0. Warn if it's not. This also fixes a
warn on unused but set variable offset.

v2: initialize offset on the gen2/3 path (Ville)

Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230526172218.1597394-1-jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/i9xx_plane.c          | 3 +++
 drivers/gpu/drm/i915/display/skl_universal_plane.c | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index 616654adbfb8f..b104883244576 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -1033,10 +1033,13 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc,
 					       DSPLINOFF(i9xx_plane));
 		base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & DISP_ADDR_MASK;
 	} else {
+		offset = 0;
 		base = intel_de_read(dev_priv, DSPADDR(i9xx_plane));
 	}
 	plane_config->base = base;
 
+	drm_WARN_ON(&dev_priv->drm, offset != 0);
+
 	val = intel_de_read(dev_priv, PIPESRC(pipe));
 	fb->width = REG_FIELD_GET(PIPESRC_WIDTH_MASK, val) + 1;
 	fb->height = REG_FIELD_GET(PIPESRC_HEIGHT_MASK, val) + 1;
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 36070d86550fe..6b01a0b68b97f 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -2529,6 +2529,7 @@ skl_get_initial_plane_config(struct intel_crtc *crtc,
 	plane_config->base = base;
 
 	offset = intel_de_read(dev_priv, PLANE_OFFSET(pipe, plane_id));
+	drm_WARN_ON(&dev_priv->drm, offset != 0);
 
 	val = intel_de_read(dev_priv, PLANE_SIZE(pipe, plane_id));
 	fb->height = REG_FIELD_GET(PLANE_HEIGHT_MASK, val) + 1;
-- 
GitLab


From 3ba39084d6fc5dd4db321b367f8bbd46c6ea90ff Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:37:55 +0300
Subject: [PATCH 0836/1791] drm/i915/ddi: drop unused but set variable intel_dp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

The intel_dp variable has been unused since commit ef79fafe9dae
("drm/i915: Eliminate intel_dp.regs.dp_tp_{ctl,status}").

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/66ca543b400a2048a6a84bb57a7bac8943014a96.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 70d44edd8c6e1..090f242e610c5 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2225,12 +2225,10 @@ static void intel_ddi_enable_fec(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	struct intel_dp *intel_dp;
 
 	if (!crtc_state->fec_enable)
 		return;
 
-	intel_dp = enc_to_intel_dp(encoder);
 	intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
 		     0, DP_TP_CTL_FEC_ENABLE);
 }
@@ -2239,12 +2237,10 @@ static void intel_ddi_disable_fec_state(struct intel_encoder *encoder,
 					const struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
-	struct intel_dp *intel_dp;
 
 	if (!crtc_state->fec_enable)
 		return;
 
-	intel_dp = enc_to_intel_dp(encoder);
 	intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state),
 		     DP_TP_CTL_FEC_ENABLE, 0);
 	intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state));
-- 
GitLab


From ac12d250e9889e1f3476b96931dbf07ea24f5f52 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:37:56 +0300
Subject: [PATCH 0837/1791] drm/i915/dsi: drop unused but set variable data
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a1a167a4ff18b19d10769d83670e414586c16956.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/vlv_dsi.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index cd90a30e04d84..aecd0142ce40d 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -136,7 +136,7 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host,
 	enum port port = intel_dsi_host->port;
 	struct mipi_dsi_packet packet;
 	ssize_t ret;
-	const u8 *header, *data;
+	const u8 *header;
 	i915_reg_t data_reg, ctrl_reg;
 	u32 data_mask, ctrl_mask;
 
@@ -145,7 +145,6 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host,
 		return ret;
 
 	header = packet.header;
-	data = packet.payload;
 
 	if (msg->flags & MIPI_DSI_MSG_USE_LPM) {
 		data_reg = MIPI_LP_GEN_DATA(port);
-- 
GitLab


From 1fdac123ab01f1e5eb91a34e6d8a60858f5927ea Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:37:57 +0300
Subject: [PATCH 0838/1791] drm/i915/dsi: drop unused but set variable vbp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

The vbp is not used for anything in the readout, as we get
e.g. crtc_vtotal from BXT_MIPI_TRANS_VTOTAL.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/27efd245aa75226adcac01eff7b21781970f2736.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/vlv_dsi.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index aecd0142ce40d..ae2f3ab3e73de 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -1039,7 +1039,7 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder,
 	unsigned int lane_count = intel_dsi->lane_count;
 	unsigned int bpp, fmt;
 	enum port port;
-	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
+	u16 hactive, hfp, hsync, hbp, vfp, vsync;
 	u16 hfp_sw, hsync_sw, hbp_sw;
 	u16 crtc_htotal_sw, crtc_hsync_start_sw, crtc_hsync_end_sw,
 				crtc_hblank_start_sw, crtc_hblank_end_sw;
@@ -1104,7 +1104,6 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder,
 	/* vertical values are in terms of lines */
 	vfp = intel_de_read(dev_priv, MIPI_VFP_COUNT(port));
 	vsync = intel_de_read(dev_priv, MIPI_VSYNC_PADDING_COUNT(port));
-	vbp = intel_de_read(dev_priv, MIPI_VBP_COUNT(port));
 
 	adjusted_mode->crtc_htotal = hactive + hfp + hsync + hbp;
 	adjusted_mode->crtc_hsync_start = hfp + adjusted_mode->crtc_hdisplay;
-- 
GitLab


From c294d37eb4554d3e40f9fd66a54b5594028a0ebe Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:37:58 +0300
Subject: [PATCH 0839/1791] drm/i915/dpll: drop unused but set variables bestn
 and bestm1
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/a85ccc8c0f451fcb997b4ac138dbeba2a653cebe.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_dpll.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c
index 824be7f037246..999badfe2906d 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll.c
@@ -1793,13 +1793,11 @@ static void chv_prepare_pll(const struct intel_crtc_state *crtc_state)
 	enum pipe pipe = crtc->pipe;
 	enum dpio_channel port = vlv_pipe_to_channel(pipe);
 	u32 loopfilter, tribuf_calcntr;
-	u32 bestn, bestm1, bestm2, bestp1, bestp2, bestm2_frac;
+	u32 bestm2, bestp1, bestp2, bestm2_frac;
 	u32 dpio_val;
 	int vco;
 
-	bestn = crtc_state->dpll.n;
 	bestm2_frac = crtc_state->dpll.m2 & 0x3fffff;
-	bestm1 = crtc_state->dpll.m1;
 	bestm2 = crtc_state->dpll.m2 >> 22;
 	bestp1 = crtc_state->dpll.p1;
 	bestp2 = crtc_state->dpll.p2;
-- 
GitLab


From 1176a905dd64637b34b7d2cde8f994ff5c3d1644 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:37:59 +0300
Subject: [PATCH 0840/1791] drm/i915/fb: drop unused but set variable cpp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ce7a5cb06c562a3399206c521a24f5091a3e7c23.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_fb.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 0d27a98dcbbe9..446bbf7986b6f 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -1601,7 +1601,7 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer *
 	for (i = 0; i < num_planes; i++) {
 		struct fb_plane_view_dims view_dims;
 		unsigned int width, height;
-		unsigned int cpp, size;
+		unsigned int size;
 		u32 offset;
 		int x, y;
 		int ret;
@@ -1618,7 +1618,6 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer *
 				return -EINVAL;
 		}
 
-		cpp = fb->base.format->cpp[i];
 		intel_fb_plane_dims(fb, i, &width, &height);
 
 		ret = convert_plane_offset_to_xy(fb, i, width, &x, &y);
-- 
GitLab


From 41b611db30143e77bc1d7f066f1954e9fbb2cdfc Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:00 +0300
Subject: [PATCH 0841/1791] drm/i915/irq: drop unused but set variable tmp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/6c529e8721d56b0148a3a84fb2d396d4485e09a2.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_irq.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index 3b2a287d20418..31dd7c9c76b9c 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -873,7 +873,7 @@ static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
 	enum pipe pipe = INVALID_PIPE;
 	enum transcoder dsi_trans;
 	enum port port;
-	u32 val, tmp;
+	u32 val;
 
 	/*
 	 * Incase of dual link, TE comes from DSI_1
@@ -920,7 +920,7 @@ static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv,
 
 	/* clear TE in dsi IIR */
 	port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A;
-	tmp = intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
+	intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0);
 }
 
 static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915)
-- 
GitLab


From b3343230ceb571be927b273875af6d8282eda2e1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:01 +0300
Subject: [PATCH 0842/1791] drm/i915/gt/uc: drop unused but set variable sseu
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Apparently sseu is leftover from commit 9a92732f040a ("drm/i915/gt: Add
general DSS steering iterator to intel_gt_mcr").

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/d542f25bffd5a50ff621bee93415a972c7768a2a.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
index 0ff864da92dfe..331cec07c1253 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c
@@ -301,7 +301,6 @@ guc_capture_alloc_steered_lists(struct intel_guc *guc,
 	const struct __guc_mmio_reg_descr_group *list;
 	struct __guc_mmio_reg_descr_group *extlists;
 	struct __guc_mmio_reg_descr *extarray;
-	struct sseu_dev_info *sseu;
 	bool has_xehpg_extregs;
 
 	/* steered registers currently only exist for the render-class */
@@ -318,7 +317,6 @@ guc_capture_alloc_steered_lists(struct intel_guc *guc,
 	if (has_xehpg_extregs)
 		num_steer_regs += ARRAY_SIZE(xehpg_extregs);
 
-	sseu = &gt->info.sseu;
 	for_each_ss_steering(iter, gt, slice, subslice)
 		num_tot_regs += num_steer_regs;
 
-- 
GitLab


From 927fc4a015937ecdc5de04c144963aa6a57762a2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:02 +0300
Subject: [PATCH 0843/1791] drm/i915/gem: drop unused but set variable unpinned
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/6831c21567e8e84da424f32a8b7b48932803ab7b.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 5fb459ea42945..cfd7929587d8f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -736,7 +736,6 @@ static int eb_reserve(struct i915_execbuffer *eb)
 	struct eb_vma *ev;
 	unsigned int pass;
 	int err = 0;
-	bool unpinned;
 
 	/*
 	 * We have one more buffers that we couldn't bind, which could be due to
@@ -776,7 +775,7 @@ static int eb_reserve(struct i915_execbuffer *eb)
 			pin_flags |= PIN_NONBLOCK;
 
 		if (pass >= 1)
-			unpinned = eb_unbind(eb, pass >= 2);
+			eb_unbind(eb, pass >= 2);
 
 		if (pass == 2) {
 			err = mutex_lock_interruptible(&eb->context->vm->mutex);
-- 
GitLab


From ab438a61e4c3675e6077d232a67284328777d929 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:03 +0300
Subject: [PATCH 0844/1791] drm/i915/gem: annotate maybe unused but set
 variable c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

The variable 'c' appears unused, but I'm not sure what should be done
with it. Annotate it with __maybe_unused.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9ee9e7d7a0a7ad4ff03c14e64b95d3fbcb7885a4.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
index cad4a6017f4b5..33d5d5178103b 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c
@@ -455,7 +455,7 @@ shmem_pwrite(struct drm_i915_gem_object *obj,
 		struct page *page;
 		void *data, *vaddr;
 		int err;
-		char c;
+		char __maybe_unused c;
 
 		len = PAGE_SIZE - pg;
 		if (len > remain)
-- 
GitLab


From 1032a491e2b5fc9793be31aae2d9e207975a1e7e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:04 +0300
Subject: [PATCH 0845/1791] drm/i915/selftest: annotate maybe unused but set
 variable unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

The variable is indeed 'unused' as the name suggests, but we can't just
drop it because i915_vma_unbind_unlocked() is annotated
__must_check. Apparently the selftest does not really need to check the
value.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e7654682f6bd6a9f6af74f4b6eb5fff7b527e412.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 36940ef10108d..5c397a2df70e2 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -391,7 +391,7 @@ static void close_object_list(struct list_head *objects,
 			      struct i915_address_space *vm)
 {
 	struct drm_i915_gem_object *obj, *on;
-	int ignored;
+	int __maybe_unused ignored;
 
 	list_for_each_entry_safe(obj, on, objects, st_link) {
 		struct i915_vma *vma;
-- 
GitLab


From ace873049e8cb543f00d8393dcad19711e7ec969 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:05 +0300
Subject: [PATCH 0846/1791] drm/i915: annotate maybe unused but set
 intel_plane_state variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

for_each_new_intel_plane_in_state() requires passing in a struct
intel_plane_state pointer, which it uses, but in many places this leads
to warning about unused but set variables. Annotate them with
__maybe_unused.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/438ff3b257b7f85ecca5750ae8687336faee0a79.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_atomic_plane.c | 2 +-
 drivers/gpu/drm/i915/display/intel_cdclk.c        | 2 +-
 drivers/gpu/drm/i915/display/intel_display.c      | 8 ++++----
 drivers/gpu/drm/i915/display/intel_fbc.c          | 6 +++---
 4 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
index 4125ee07a2717..7d9578ebae556 100644
--- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c
@@ -722,7 +722,7 @@ skl_next_plane_to_commit(struct intel_atomic_state *state,
 {
 	struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
-	struct intel_plane_state *plane_state;
+	struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int i;
 
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 1a5268e3d0a36..4207863b7b2ae 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -2959,7 +2959,7 @@ int intel_cdclk_atomic_check(struct intel_atomic_state *state,
 {
 	const struct intel_cdclk_state *old_cdclk_state;
 	const struct intel_cdclk_state *new_cdclk_state;
-	struct intel_plane_state *plane_state;
+	struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int ret;
 	int i;
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index f23dd937c27c4..b847c1c753703 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -971,7 +971,7 @@ static void intel_crtc_enable_flip_done(struct intel_atomic_state *state,
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	u8 update_planes = crtc_state->update_planes;
-	const struct intel_plane_state *plane_state;
+	const struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int i;
 
@@ -988,7 +988,7 @@ static void intel_crtc_disable_flip_done(struct intel_atomic_state *state,
 	const struct intel_crtc_state *crtc_state =
 		intel_atomic_get_new_crtc_state(state, crtc);
 	u8 update_planes = crtc_state->update_planes;
-	const struct intel_plane_state *plane_state;
+	const struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int i;
 
@@ -5617,7 +5617,7 @@ static int intel_crtc_add_bigjoiner_planes(struct intel_atomic_state *state,
 					   struct intel_crtc *crtc,
 					   struct intel_crtc *other)
 {
-	const struct intel_plane_state *plane_state;
+	const struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	u8 plane_ids = 0;
 	int i;
@@ -5660,7 +5660,7 @@ static int intel_atomic_check_planes(struct intel_atomic_state *state)
 {
 	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
 	struct intel_crtc_state *old_crtc_state, *new_crtc_state;
-	struct intel_plane_state *plane_state;
+	struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	struct intel_crtc *crtc;
 	int i, ret;
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 1966f93962019..29aa029d249dd 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -1254,7 +1254,7 @@ static bool __intel_fbc_pre_update(struct intel_atomic_state *state,
 bool intel_fbc_pre_update(struct intel_atomic_state *state,
 			  struct intel_crtc *crtc)
 {
-	const struct intel_plane_state *plane_state;
+	const struct intel_plane_state __maybe_unused *plane_state;
 	bool need_vblank_wait = false;
 	struct intel_plane *plane;
 	int i;
@@ -1309,7 +1309,7 @@ static void __intel_fbc_post_update(struct intel_fbc *fbc)
 void intel_fbc_post_update(struct intel_atomic_state *state,
 			   struct intel_crtc *crtc)
 {
-	const struct intel_plane_state *plane_state;
+	const struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int i;
 
@@ -1408,7 +1408,7 @@ void intel_fbc_flush(struct drm_i915_private *i915,
 
 int intel_fbc_atomic_check(struct intel_atomic_state *state)
 {
-	struct intel_plane_state *plane_state;
+	struct intel_plane_state __maybe_unused *plane_state;
 	struct intel_plane *plane;
 	int i;
 
-- 
GitLab


From 84f9c3c73b9a18a21ea13135d4e850fcf7629f59 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:06 +0300
Subject: [PATCH 0847/1791] drm/i915: annotate maybe unused but set
 intel_crtc_state variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare for re-enabling -Wunused-but-set-variable.

for_each_new_intel_crtc_in_state() requires passing in a struct
intel_crtc_state pointer, which it uses, but in a few places this leads
to warning about unused but set variables. Annotate them with
__maybe_unused.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/eb041f426bc3d76ef7a0ea906f99367cbf439b1a.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 2 +-
 drivers/gpu/drm/i915/display/skl_watermark.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index b847c1c753703..34447d96fe570 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -5715,7 +5715,7 @@ static int intel_atomic_check_planes(struct intel_atomic_state *state)
 
 static int intel_atomic_check_crtcs(struct intel_atomic_state *state)
 {
-	struct intel_crtc_state *crtc_state;
+	struct intel_crtc_state __maybe_unused *crtc_state;
 	struct intel_crtc *crtc;
 	int i;
 
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index d1245c847f1cb..063929a42a42f 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -2900,7 +2900,7 @@ static int
 skl_compute_wm(struct intel_atomic_state *state)
 {
 	struct intel_crtc *crtc;
-	struct intel_crtc_state *new_crtc_state;
+	struct intel_crtc_state __maybe_unused *new_crtc_state;
 	int ret, i;
 
 	for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) {
-- 
GitLab


From 81b1b599dfd71c958418dad586fa72c8d30d1065 Mon Sep 17 00:00:00 2001
From: Fei Yang <fei.yang@intel.com>
Date: Tue, 6 Jun 2023 12:00:42 +0200
Subject: [PATCH 0848/1791] drm/i915: Allow user to set cache at BO creation

To comply with the design that buffer objects shall have immutable
cache setting through out their life cycle, {set, get}_caching ioctl's
are no longer supported from MTL onward. With that change caching
policy can only be set at object creation time. The current code
applies a default (platform dependent) cache setting for all objects.
However this is not optimal for performance tuning. The patch extends
the existing gem_create uAPI to let user set PAT index for the object
at creation time.
The new extension is platform independent, so UMD's can switch to using
this extension for older platforms as well, while {set, get}_caching are
still supported on these legacy paltforms for compatibility reason.
However, since PAT index was not clearly defined for platforms prior to
GEN12 (TGL), so we are limiting this externsion to GEN12+ platforms
only. See ext_set_pat() in for the implementation details.

The documentation related to the PAT/MOCS tables is currently available
for Tiger Lake here:
https://www.intel.com/content/www/us/en/docs/graphics-for-linux/developer-reference/1-0/tiger-lake.html

The documentation for other platforms is currently being updated.

BSpec: 45101

Mesa support has been submitted in this merge request:
https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22878

The media driver supprt has bin submitted in this merge request:
https://github.com/intel/media-driver/pull/1680

The IGT test related to this change is
igt@gem_create@create-ext-set-pat

Signed-off-by: Fei Yang <fei.yang@intel.com>
Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Carl Zhang <carl.zhang@intel.com>
Tested-by: Lihao Gu <lihao.gu@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Slawomir Milczarek <slawomir.milczarek@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606100042.482345-2-andi.shyti@linux.intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_create.c | 40 +++++++++++++++++++++
 drivers/gpu/drm/i915/gem/i915_gem_object.c |  6 ++++
 include/uapi/drm/i915_drm.h                | 41 ++++++++++++++++++++++
 3 files changed, 87 insertions(+)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_create.c b/drivers/gpu/drm/i915/gem/i915_gem_create.c
index bfe1dbda4cb75..d24c0ce8805c7 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_create.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_create.c
@@ -245,6 +245,7 @@ struct create_ext {
 	unsigned int n_placements;
 	unsigned int placement_mask;
 	unsigned long flags;
+	unsigned int pat_index;
 };
 
 static void repr_placements(char *buf, size_t size,
@@ -394,11 +395,43 @@ static int ext_set_protected(struct i915_user_extension __user *base, void *data
 	return 0;
 }
 
+static int ext_set_pat(struct i915_user_extension __user *base, void *data)
+{
+	struct create_ext *ext_data = data;
+	struct drm_i915_private *i915 = ext_data->i915;
+	struct drm_i915_gem_create_ext_set_pat ext;
+	unsigned int max_pat_index;
+
+	BUILD_BUG_ON(sizeof(struct drm_i915_gem_create_ext_set_pat) !=
+		     offsetofend(struct drm_i915_gem_create_ext_set_pat, rsvd));
+
+	/* Limiting the extension only to Meteor Lake */
+	if (!IS_METEORLAKE(i915))
+		return -ENODEV;
+
+	if (copy_from_user(&ext, base, sizeof(ext)))
+		return -EFAULT;
+
+	max_pat_index = INTEL_INFO(i915)->max_pat_index;
+
+	if (ext.pat_index > max_pat_index) {
+		drm_dbg(&i915->drm, "PAT index is invalid: %u\n",
+			ext.pat_index);
+		return -EINVAL;
+	}
+
+	ext_data->pat_index = ext.pat_index;
+
+	return 0;
+}
+
 static const i915_user_extension_fn create_extensions[] = {
 	[I915_GEM_CREATE_EXT_MEMORY_REGIONS] = ext_set_placements,
 	[I915_GEM_CREATE_EXT_PROTECTED_CONTENT] = ext_set_protected,
+	[I915_GEM_CREATE_EXT_SET_PAT] = ext_set_pat,
 };
 
+#define PAT_INDEX_NOT_SET	0xffff
 /**
  * i915_gem_create_ext_ioctl - Creates a new mm object and returns a handle to it.
  * @dev: drm device pointer
@@ -418,6 +451,7 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data,
 	if (args->flags & ~I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS)
 		return -EINVAL;
 
+	ext_data.pat_index = PAT_INDEX_NOT_SET;
 	ret = i915_user_extensions(u64_to_user_ptr(args->extensions),
 				   create_extensions,
 				   ARRAY_SIZE(create_extensions),
@@ -454,5 +488,11 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data,
 	if (IS_ERR(obj))
 		return PTR_ERR(obj);
 
+	if (ext_data.pat_index != PAT_INDEX_NOT_SET) {
+		i915_gem_object_set_pat_index(obj, ext_data.pat_index);
+		/* Mark pat_index is set by UMD */
+		obj->pat_set_by_user = true;
+	}
+
 	return i915_gem_publish(obj, file, &args->size, &args->handle);
 }
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 46a19b099ec88..97ac6fb37958f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -208,6 +208,12 @@ bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj)
 	if (!(obj->flags & I915_BO_ALLOC_USER))
 		return false;
 
+	/*
+	 * Always flush cache for UMD objects at creation time.
+	 */
+	if (obj->pat_set_by_user)
+		return true;
+
 	/*
 	 * EHL and JSL add the 'Bypass LLC' MOCS entry, which should make it
 	 * possible for userspace to bypass the GTT caching bits set by the
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index a1848e8060590..7000e5910a1d7 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -3680,9 +3680,13 @@ struct drm_i915_gem_create_ext {
 	 *
 	 * For I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see
 	 * struct drm_i915_gem_create_ext_protected_content.
+	 *
+	 * For I915_GEM_CREATE_EXT_SET_PAT usage see
+	 * struct drm_i915_gem_create_ext_set_pat.
 	 */
 #define I915_GEM_CREATE_EXT_MEMORY_REGIONS 0
 #define I915_GEM_CREATE_EXT_PROTECTED_CONTENT 1
+#define I915_GEM_CREATE_EXT_SET_PAT 2
 	__u64 extensions;
 };
 
@@ -3797,6 +3801,43 @@ struct drm_i915_gem_create_ext_protected_content {
 	__u32 flags;
 };
 
+/**
+ * struct drm_i915_gem_create_ext_set_pat - The
+ * I915_GEM_CREATE_EXT_SET_PAT extension.
+ *
+ * If this extension is provided, the specified caching policy (PAT index) is
+ * applied to the buffer object.
+ *
+ * Below is an example on how to create an object with specific caching policy:
+ *
+ * .. code-block:: C
+ *
+ *      struct drm_i915_gem_create_ext_set_pat set_pat_ext = {
+ *              .base = { .name = I915_GEM_CREATE_EXT_SET_PAT },
+ *              .pat_index = 0,
+ *      };
+ *      struct drm_i915_gem_create_ext create_ext = {
+ *              .size = PAGE_SIZE,
+ *              .extensions = (uintptr_t)&set_pat_ext,
+ *      };
+ *
+ *      int err = ioctl(fd, DRM_IOCTL_I915_GEM_CREATE_EXT, &create_ext);
+ *      if (err) ...
+ */
+struct drm_i915_gem_create_ext_set_pat {
+	/** @base: Extension link. See struct i915_user_extension. */
+	struct i915_user_extension base;
+	/**
+	 * @pat_index: PAT index to be set
+	 * PAT index is a bit field in Page Table Entry to control caching
+	 * behaviors for GPU accesses. The definition of PAT index is
+	 * platform dependent and can be found in hardware specifications,
+	 */
+	__u32 pat_index;
+	/** @rsvd: reserved for future use */
+	__u32 rsvd;
+};
+
 /* ID of the protected content session managed by i915 when PXP is active */
 #define I915_PROTECTED_CONTENT_DEFAULT_SESSION 0xf
 
-- 
GitLab


From 9831a9d1841fb3188c46a13da30aecd96bd1bae6 Mon Sep 17 00:00:00 2001
From: Khaled Almahallawy <khaled.almahallawy@intel.com>
Date: Tue, 6 Jun 2023 15:44:28 -0700
Subject: [PATCH 0849/1791] drm/i915/dp: Fix log level for "CDS interlane align
 done"

"CDS interlane align done" is a passing condition not an error.
Before adding new macros for logs it was drm_dbg_kms.

Fixes: f48eab290287 ("drm/i915/dp: Add link training debug and error printing helpers")
Cc: Imre Deak <imre.deak@intel.com>
CC: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606224428.3791006-1-khaled.almahallawy@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_link_training.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 0952a707358c1..176b610642e75 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1279,7 +1279,7 @@ intel_dp_128b132b_lane_cds(struct intel_dp *intel_dp,
 		if (drm_dp_128b132b_eq_interlane_align_done(link_status) &&
 		    drm_dp_128b132b_cds_interlane_align_done(link_status) &&
 		    drm_dp_128b132b_lane_symbol_locked(link_status, crtc_state->lane_count)) {
-			lt_err(intel_dp, DP_PHY_DPRX, "CDS interlane align done\n");
+			lt_dbg(intel_dp, DP_PHY_DPRX, "CDS interlane align done\n");
 			break;
 		}
 
-- 
GitLab


From 626765bb4b2e732e9a1595fc801f696dbde12db9 Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:03 +0300
Subject: [PATCH 0850/1791] drm/i915: fix the derating percentage for MTL

Follow the values from bspec for the percentage overhead for
efficiency in MTL BW calculations.

Bspec: 64631

Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-2-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 597d5816ad1b0..ab405c48ca3a9 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -379,7 +379,7 @@ static const struct intel_sa_info mtl_sa_info = {
 	.deburst = 32,
 	.deprogbwlimit = 38, /* GB/s */
 	.displayrtids = 256,
-	.derating = 20,
+	.derating = 10,
 };
 
 static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
-- 
GitLab


From 39bea0ff0f7470a5a812f50b2485d2380759e19b Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:04 +0300
Subject: [PATCH 0851/1791] drm/i915: update the QGV point frequency
 calculations

From MTL onwwards, pcode locks the QGV point based on peak BW of
the intended QGV point passed by the driver. So the peak BW
calculation must match the value expected by the pcode. Update
the calculations as per the Bspec.

v2: use DIV_ROUND_* macro for the calculations (Ville)

v3: Use only DIV_ROUN_CLOSEST and remove divisor / 2 again

Bspec: 64636

Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-3-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index ab405c48ca3a9..61b3babf2d837 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -182,7 +182,7 @@ static int mtl_read_qgv_point_info(struct drm_i915_private *dev_priv,
 	val2 = intel_uncore_read(&dev_priv->uncore,
 				 MTL_MEM_SS_INFO_QGV_POINT_HIGH(point));
 	dclk = REG_FIELD_GET(MTL_DCLK_MASK, val);
-	sp->dclk = DIV_ROUND_UP((16667 * dclk), 1000);
+	sp->dclk = DIV_ROUND_CLOSEST(16667 * dclk, 1000);
 	sp->t_rp = REG_FIELD_GET(MTL_TRP_MASK, val);
 	sp->t_rcd = REG_FIELD_GET(MTL_TRCD_MASK, val);
 
-- 
GitLab


From 88d0ecbde90fdf5b78d702f9ec9711d60d35004c Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:05 +0300
Subject: [PATCH 0852/1791] drm/i915: store the peak bw per QGV point

In MTL onwards, pcode locks the GV point based on the peak BW
of a QGV point. So store the peak BW of all the QGV points.

v2: use DIV_ROUND_CLOSEST() for the peakBW calculation

Bspec: 64636

Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-4-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c           | 8 ++++++--
 drivers/gpu/drm/i915/display/intel_display_core.h | 2 ++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 61b3babf2d837..b792d307e9d57 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -534,10 +534,14 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 
 			bi->deratedbw[j] = min(maxdebw,
 					       bw * (100 - sa->derating) / 100);
+			bi->peakbw[j] = DIV_ROUND_CLOSEST(sp->dclk *
+							  num_channels *
+							  qi.channel_width, 8);
 
 			drm_dbg_kms(&dev_priv->drm,
-				    "BW%d / QGV %d: num_planes=%d deratedbw=%u\n",
-				    i, j, bi->num_planes, bi->deratedbw[j]);
+				    "BW%d / QGV %d: num_planes=%d deratedbw=%u peakbw: %u\n",
+				    i, j, bi->num_planes, bi->deratedbw[j],
+				    bi->peakbw[j]);
 		}
 
 		for (j = 0; j < qi.num_psf_points; j++) {
diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h
index 2209811eb29e7..dd8e08c8598f4 100644
--- a/drivers/gpu/drm/i915/display/intel_display_core.h
+++ b/drivers/gpu/drm/i915/display/intel_display_core.h
@@ -314,6 +314,8 @@ struct intel_display {
 			unsigned int deratedbw[I915_NUM_QGV_POINTS];
 			/* for each PSF GV point */
 			unsigned int psf_bw[I915_NUM_PSF_GV_POINTS];
+			/* Peak BW for each QGV point */
+			unsigned int peakbw[I915_NUM_QGV_POINTS];
 			u8 num_qgv_points;
 			u8 num_psf_gv_points;
 			u8 num_planes;
-- 
GitLab


From 463cc9402e71deead30bd85335c56a98a1c1c462 Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:06 +0300
Subject: [PATCH 0853/1791] drm/i915: extract intel_bw_check_qgv_points()

Extract intel_bw_check_qgv_points() from intel_bw_atomic_check
to facilitate future platform variations in handling SAGV
configurations.

Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-5-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 235 +++++++++++++-----------
 1 file changed, 130 insertions(+), 105 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index b792d307e9d57..56b3975f3ccb3 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -803,6 +803,128 @@ intel_atomic_get_bw_state(struct intel_atomic_state *state)
 	return to_intel_bw_state(bw_state);
 }
 
+static int icl_find_qgv_points(struct drm_i915_private *i915,
+			       unsigned int data_rate,
+			       unsigned int num_active_planes,
+			       const struct intel_bw_state *old_bw_state,
+			       struct intel_bw_state *new_bw_state)
+{
+	unsigned int max_bw_point = 0;
+	unsigned int max_bw = 0;
+	unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points;
+	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	u16 psf_points = 0;
+	u16 qgv_points = 0;
+	int i;
+	int ret;
+
+	ret = intel_atomic_lock_global_state(&new_bw_state->base);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < num_qgv_points; i++) {
+		unsigned int max_data_rate;
+
+		if (DISPLAY_VER(i915) > 11)
+			max_data_rate = tgl_max_bw(i915, num_active_planes, i);
+		else
+			max_data_rate = icl_max_bw(i915, num_active_planes, i);
+		/*
+		 * We need to know which qgv point gives us
+		 * maximum bandwidth in order to disable SAGV
+		 * if we find that we exceed SAGV block time
+		 * with watermarks. By that moment we already
+		 * have those, as it is calculated earlier in
+		 * intel_atomic_check,
+		 */
+		if (max_data_rate > max_bw) {
+			max_bw_point = i;
+			max_bw = max_data_rate;
+		}
+		if (max_data_rate >= data_rate)
+			qgv_points |= BIT(i);
+
+		drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d\n",
+			    i, max_data_rate, data_rate);
+	}
+
+	for (i = 0; i < num_psf_gv_points; i++) {
+		unsigned int max_data_rate = adl_psf_bw(i915, i);
+
+		if (max_data_rate >= data_rate)
+			psf_points |= BIT(i);
+
+		drm_dbg_kms(&i915->drm, "PSF GV point %d: max bw %d"
+			    " required %d\n",
+			    i, max_data_rate, data_rate);
+	}
+
+	/*
+	 * BSpec states that we always should have at least one allowed point
+	 * left, so if we couldn't - simply reject the configuration for obvious
+	 * reasons.
+	 */
+	if (qgv_points == 0) {
+		drm_dbg_kms(&i915->drm, "No QGV points provide sufficient memory"
+			    " bandwidth %d for display configuration(%d active planes).\n",
+			    data_rate, num_active_planes);
+		return -EINVAL;
+	}
+
+	if (num_psf_gv_points > 0 && psf_points == 0) {
+		drm_dbg_kms(&i915->drm, "No PSF GV points provide sufficient memory"
+			    " bandwidth %d for display configuration(%d active planes).\n",
+			    data_rate, num_active_planes);
+		return -EINVAL;
+	}
+
+	/*
+	 * Leave only single point with highest bandwidth, if
+	 * we can't enable SAGV due to the increased memory latency it may
+	 * cause.
+	 */
+	if (!intel_can_enable_sagv(i915, new_bw_state)) {
+		qgv_points = BIT(max_bw_point);
+		drm_dbg_kms(&i915->drm, "No SAGV, using single QGV point %d\n",
+			    max_bw_point);
+	}
+
+	/*
+	 * We store the ones which need to be masked as that is what PCode
+	 * actually accepts as a parameter.
+	 */
+	new_bw_state->qgv_points_mask =
+		~(ICL_PCODE_REQ_QGV_PT(qgv_points) |
+		  ADLS_PCODE_REQ_PSF_PT(psf_points)) &
+		icl_qgv_points_mask(i915);
+
+	/*
+	 * If the actual mask had changed we need to make sure that
+	 * the commits are serialized(in case this is a nomodeset, nonblocking)
+	 */
+	if (new_bw_state->qgv_points_mask != old_bw_state->qgv_points_mask) {
+		ret = intel_atomic_serialize_global_state(&new_bw_state->base);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int intel_bw_check_qgv_points(struct drm_i915_private *i915,
+				     const struct intel_bw_state *old_bw_state,
+				     struct intel_bw_state *new_bw_state)
+{
+	unsigned int data_rate = intel_bw_data_rate(i915, new_bw_state);
+	unsigned int num_active_planes =
+			intel_bw_num_active_planes(i915, new_bw_state);
+
+	data_rate = DIV_ROUND_UP(data_rate, 1000);
+
+	return icl_find_qgv_points(i915, data_rate, num_active_planes,
+				   old_bw_state, new_bw_state);
+}
+
 static bool intel_bw_state_changed(struct drm_i915_private *i915,
 				   const struct intel_bw_state *old_bw_state,
 				   const struct intel_bw_state *new_bw_state)
@@ -1049,20 +1171,14 @@ static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *chan
 
 int intel_bw_atomic_check(struct intel_atomic_state *state)
 {
-	struct drm_i915_private *dev_priv = to_i915(state->base.dev);
-	const struct intel_bw_state *old_bw_state;
-	struct intel_bw_state *new_bw_state;
-	unsigned int data_rate;
-	unsigned int num_active_planes;
-	int i, ret;
-	u16 qgv_points = 0, psf_points = 0;
-	unsigned int max_bw_point = 0, max_bw = 0;
-	unsigned int num_qgv_points = dev_priv->display.bw.max[0].num_qgv_points;
-	unsigned int num_psf_gv_points = dev_priv->display.bw.max[0].num_psf_gv_points;
 	bool changed = false;
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_bw_state *new_bw_state;
+	const struct intel_bw_state *old_bw_state;
+	int ret;
 
 	/* FIXME earlier gens need some checks too */
-	if (DISPLAY_VER(dev_priv) < 11)
+	if (DISPLAY_VER(i915) < 11)
 		return 0;
 
 	ret = intel_bw_check_data_rate(state, &changed);
@@ -1073,8 +1189,8 @@ int intel_bw_atomic_check(struct intel_atomic_state *state)
 	new_bw_state = intel_atomic_get_new_bw_state(state);
 
 	if (new_bw_state &&
-	    intel_can_enable_sagv(dev_priv, old_bw_state) !=
-	    intel_can_enable_sagv(dev_priv, new_bw_state))
+	    intel_can_enable_sagv(i915, old_bw_state) !=
+	    intel_can_enable_sagv(i915, new_bw_state))
 		changed = true;
 
 	/*
@@ -1084,101 +1200,10 @@ int intel_bw_atomic_check(struct intel_atomic_state *state)
 	if (!changed)
 		return 0;
 
-	ret = intel_atomic_lock_global_state(&new_bw_state->base);
+	ret = intel_bw_check_qgv_points(i915, old_bw_state, new_bw_state);
 	if (ret)
 		return ret;
 
-	data_rate = intel_bw_data_rate(dev_priv, new_bw_state);
-	data_rate = DIV_ROUND_UP(data_rate, 1000);
-
-	num_active_planes = intel_bw_num_active_planes(dev_priv, new_bw_state);
-
-	for (i = 0; i < num_qgv_points; i++) {
-		unsigned int max_data_rate;
-
-		if (DISPLAY_VER(dev_priv) > 11)
-			max_data_rate = tgl_max_bw(dev_priv, num_active_planes, i);
-		else
-			max_data_rate = icl_max_bw(dev_priv, num_active_planes, i);
-		/*
-		 * We need to know which qgv point gives us
-		 * maximum bandwidth in order to disable SAGV
-		 * if we find that we exceed SAGV block time
-		 * with watermarks. By that moment we already
-		 * have those, as it is calculated earlier in
-		 * intel_atomic_check,
-		 */
-		if (max_data_rate > max_bw) {
-			max_bw_point = i;
-			max_bw = max_data_rate;
-		}
-		if (max_data_rate >= data_rate)
-			qgv_points |= BIT(i);
-
-		drm_dbg_kms(&dev_priv->drm, "QGV point %d: max bw %d required %d\n",
-			    i, max_data_rate, data_rate);
-	}
-
-	for (i = 0; i < num_psf_gv_points; i++) {
-		unsigned int max_data_rate = adl_psf_bw(dev_priv, i);
-
-		if (max_data_rate >= data_rate)
-			psf_points |= BIT(i);
-
-		drm_dbg_kms(&dev_priv->drm, "PSF GV point %d: max bw %d"
-			    " required %d\n",
-			    i, max_data_rate, data_rate);
-	}
-
-	/*
-	 * BSpec states that we always should have at least one allowed point
-	 * left, so if we couldn't - simply reject the configuration for obvious
-	 * reasons.
-	 */
-	if (qgv_points == 0) {
-		drm_dbg_kms(&dev_priv->drm, "No QGV points provide sufficient memory"
-			    " bandwidth %d for display configuration(%d active planes).\n",
-			    data_rate, num_active_planes);
-		return -EINVAL;
-	}
-
-	if (num_psf_gv_points > 0 && psf_points == 0) {
-		drm_dbg_kms(&dev_priv->drm, "No PSF GV points provide sufficient memory"
-			    " bandwidth %d for display configuration(%d active planes).\n",
-			    data_rate, num_active_planes);
-		return -EINVAL;
-	}
-
-	/*
-	 * Leave only single point with highest bandwidth, if
-	 * we can't enable SAGV due to the increased memory latency it may
-	 * cause.
-	 */
-	if (!intel_can_enable_sagv(dev_priv, new_bw_state)) {
-		qgv_points = BIT(max_bw_point);
-		drm_dbg_kms(&dev_priv->drm, "No SAGV, using single QGV point %d\n",
-			    max_bw_point);
-	}
-
-	/*
-	 * We store the ones which need to be masked as that is what PCode
-	 * actually accepts as a parameter.
-	 */
-	new_bw_state->qgv_points_mask =
-		~(ICL_PCODE_REQ_QGV_PT(qgv_points) |
-		  ADLS_PCODE_REQ_PSF_PT(psf_points)) &
-		icl_qgv_points_mask(dev_priv);
-
-	/*
-	 * If the actual mask had changed we need to make sure that
-	 * the commits are serialized(in case this is a nomodeset, nonblocking)
-	 */
-	if (new_bw_state->qgv_points_mask != old_bw_state->qgv_points_mask) {
-		ret = intel_atomic_serialize_global_state(&new_bw_state->base);
-		if (ret)
-			return ret;
-	}
-
 	return 0;
 }
 
-- 
GitLab


From 6400c215a1c19e40bab513b20d35780e61b291f5 Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:07 +0300
Subject: [PATCH 0854/1791] drm/i915: modify max_bw to return index to
 intel_bw_info

MTL uses the peak BW of a QGV point to lock the required QGV
point instead of the QGV index. Instead of passing the deratedbw
of the selected bw_info, return the index to the selected
bw_info so that either deratedbw or peakbw can be used based on
the platform.

v2: use idx to store index returned by max_bw_index functions

v3: return UINT_MAX in icl_max_bw_index in case no match found

v3: check idx >= ARRAY_SIZE

Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-6-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 27 ++++++++++++++++---------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 56b3975f3ccb3..b1cbeda0b2e3d 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -593,8 +593,8 @@ static void dg2_get_bw_info(struct drm_i915_private *i915)
 	i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
 }
 
-static unsigned int icl_max_bw(struct drm_i915_private *dev_priv,
-			       int num_planes, int qgv_point)
+static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv,
+				     int num_planes, int qgv_point)
 {
 	int i;
 
@@ -615,14 +615,14 @@ static unsigned int icl_max_bw(struct drm_i915_private *dev_priv,
 			return UINT_MAX;
 
 		if (num_planes >= bi->num_planes)
-			return bi->deratedbw[qgv_point];
+			return i;
 	}
 
-	return 0;
+	return UINT_MAX;
 }
 
-static unsigned int tgl_max_bw(struct drm_i915_private *dev_priv,
-			       int num_planes, int qgv_point)
+static unsigned int tgl_max_bw_index(struct drm_i915_private *dev_priv,
+				     int num_planes, int qgv_point)
 {
 	int i;
 
@@ -643,10 +643,10 @@ static unsigned int tgl_max_bw(struct drm_i915_private *dev_priv,
 			return UINT_MAX;
 
 		if (num_planes <= bi->num_planes)
-			return bi->deratedbw[qgv_point];
+			return i;
 	}
 
-	return dev_priv->display.bw.max[0].deratedbw[qgv_point];
+	return 0;
 }
 
 static unsigned int adl_psf_bw(struct drm_i915_private *dev_priv,
@@ -823,12 +823,19 @@ static int icl_find_qgv_points(struct drm_i915_private *i915,
 		return ret;
 
 	for (i = 0; i < num_qgv_points; i++) {
+		unsigned int idx;
 		unsigned int max_data_rate;
 
 		if (DISPLAY_VER(i915) > 11)
-			max_data_rate = tgl_max_bw(i915, num_active_planes, i);
+			idx = tgl_max_bw_index(i915, num_active_planes, i);
 		else
-			max_data_rate = icl_max_bw(i915, num_active_planes, i);
+			idx = icl_max_bw_index(i915, num_active_planes, i);
+
+		if (idx >= ARRAY_SIZE(i915->display.bw.max))
+			continue;
+
+		max_data_rate = i915->display.bw.max[idx].deratedbw[i];
+
 		/*
 		 * We need to know which qgv point gives us
 		 * maximum bandwidth in order to disable SAGV
-- 
GitLab


From a5819e51912292e16f14f4c014c384f754002e5e Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 6 Jun 2023 12:35:08 +0300
Subject: [PATCH 0855/1791] drm/i915/mtl: find the best QGV point for the SAGV
 configuration

From MTL onwards, we need to find the best QGV point based on
the required data rate and pass the peak BW of that point to
the punit to lock the corresponding QGV point.

v1: Fix for warning from kernel test robot

v2: No need to serialize for the peakbw change as pmdemand code
    will do that (Imre)

Bspec: 64636

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202305280253.Ab8bRV2w-lkp@intel.com/
Reported-by: Dan Carpenter <error27@gmail.com>
Closes: https://lore.kernel.org/r/202305280253.Ab8bRV2w-lkp@intel.com/
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606093509.221709-7-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_bw.c | 79 ++++++++++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_bw.h |  6 ++
 2 files changed, 83 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index b1cbeda0b2e3d..bef96db62c807 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -803,6 +803,77 @@ intel_atomic_get_bw_state(struct intel_atomic_state *state)
 	return to_intel_bw_state(bw_state);
 }
 
+static int mtl_find_qgv_points(struct drm_i915_private *i915,
+			       unsigned int data_rate,
+			       unsigned int num_active_planes,
+			       struct intel_bw_state *new_bw_state)
+{
+	unsigned int best_rate = UINT_MAX;
+	unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points;
+	unsigned int qgv_peak_bw  = 0;
+	int i;
+	int ret;
+
+	ret = intel_atomic_lock_global_state(&new_bw_state->base);
+	if (ret)
+		return ret;
+
+	/*
+	 * If SAGV cannot be enabled, disable the pcode SAGV by passing all 1's
+	 * for qgv peak bw in PM Demand request. So assign UINT_MAX if SAGV is
+	 * not enabled. PM Demand code will clamp the value for the register
+	 */
+	if (!intel_can_enable_sagv(i915, new_bw_state)) {
+		new_bw_state->qgv_point_peakbw = U16_MAX;
+		drm_dbg_kms(&i915->drm, "No SAGV, use UINT_MAX as peak bw.");
+		return 0;
+	}
+
+	/*
+	 * Find the best QGV point by comparing the data_rate with max data rate
+	 * offered per plane group
+	 */
+	for (i = 0; i < num_qgv_points; i++) {
+		unsigned int bw_index =
+			tgl_max_bw_index(i915, num_active_planes, i);
+		unsigned int max_data_rate;
+
+		if (bw_index >= ARRAY_SIZE(i915->display.bw.max))
+			continue;
+
+		max_data_rate = i915->display.bw.max[bw_index].deratedbw[i];
+
+		if (max_data_rate < data_rate)
+			continue;
+
+		if (max_data_rate - data_rate < best_rate) {
+			best_rate = max_data_rate - data_rate;
+			qgv_peak_bw = i915->display.bw.max[bw_index].peakbw[i];
+		}
+
+		drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d qgv_peak_bw: %d\n",
+			    i, max_data_rate, data_rate, qgv_peak_bw);
+	}
+
+	drm_dbg_kms(&i915->drm, "Matching peaks QGV bw: %d for required data rate: %d\n",
+		    qgv_peak_bw, data_rate);
+
+	/*
+	 * The display configuration cannot be supported if no QGV point
+	 * satisfying the required data rate is found
+	 */
+	if (qgv_peak_bw == 0) {
+		drm_dbg_kms(&i915->drm, "No QGV points for bw %d for display configuration(%d active planes).\n",
+			    data_rate, num_active_planes);
+		return -EINVAL;
+	}
+
+	/* MTL PM DEMAND expects QGV BW parameter in multiples of 100 mbps */
+	new_bw_state->qgv_point_peakbw = DIV_ROUND_CLOSEST(qgv_peak_bw, 100);
+
+	return 0;
+}
+
 static int icl_find_qgv_points(struct drm_i915_private *i915,
 			       unsigned int data_rate,
 			       unsigned int num_active_planes,
@@ -928,8 +999,12 @@ static int intel_bw_check_qgv_points(struct drm_i915_private *i915,
 
 	data_rate = DIV_ROUND_UP(data_rate, 1000);
 
-	return icl_find_qgv_points(i915, data_rate, num_active_planes,
-				   old_bw_state, new_bw_state);
+	if (DISPLAY_VER(i915) >= 14)
+		return mtl_find_qgv_points(i915, data_rate, num_active_planes,
+					   new_bw_state);
+	else
+		return icl_find_qgv_points(i915, data_rate, num_active_planes,
+					   old_bw_state, new_bw_state);
 }
 
 static bool intel_bw_state_changed(struct drm_i915_private *i915,
diff --git a/drivers/gpu/drm/i915/display/intel_bw.h b/drivers/gpu/drm/i915/display/intel_bw.h
index f202921437455..59cb4fc5db76c 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.h
+++ b/drivers/gpu/drm/i915/display/intel_bw.h
@@ -34,6 +34,12 @@ struct intel_bw_state {
 	/* bitmask of active pipes */
 	u8 active_pipes;
 
+	/*
+	 * From MTL onwards, to lock a QGV point, punit expects the peak BW of
+	 * the selected QGV point as the parameter in multiples of 100MB/s
+	 */
+	u16 qgv_point_peakbw;
+
 	/*
 	 * Current QGV points mask, which restricts
 	 * some particular SAGV states, not to confuse
-- 
GitLab


From 4c4cc7ac207f56ed5db5272dab4169d2d40e418d Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Tue, 6 Jun 2023 23:10:32 +0300
Subject: [PATCH 0856/1791] drm/i915/mtl: Add support for PM DEMAND
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

MTL introduces a new way to instruct the PUnit with
power and bandwidth requirements of DE. Add the functionality
to program the registers and handle waits using interrupts.
The current wait time for timeouts is programmed for 10 msecs to
factor in the worst case scenarios. Changes made to use REG_BIT
for a register that we touched(GEN8_DE_MISC_IER _MMIO).

Wa_14016740474 is added which applies to Xe_LPD+ display

v2: checkpatch warning fixes, simplify program pmdemand part

v3: update to dbufs and pipes values to pmdemand register(stan)
    Removed the macro usage in update_pmdemand_values()

v4: move the pmdemand_pre_plane_update before cdclk update
    pmdemand_needs_update included cdclk params comparisons
    pmdemand_state NULL check (Gustavo)
    pmdemand.o in sorted order in the makefile (Jani)
    update pmdemand misc irq handler loop (Gustavo)
    active phys bitmask and programming correction (Gustavo)

v5: simplify pmdemand_state structure
    simplify methods to find active phys and max port clock
    Timeout in case of previou pmdemand task pending (Gustavo)

v6: rebasing
    updates to max_ddiclk calculations (Gustavo)
    updates to active_phys count method (Gustavo)

v7: use two separate loop to iterate throug old and new
    crtc states to calculate the active phys (Gustavo)

v8: use uniform function names (Gustavo)

v9: For phys change iterate through connectors (Imre)
    Look for change in phys for pmdemand update (Gustavo, Imre)
    Some more stlying changes (Imre)
    Update pmdemand state during HW readout/sanitize (Imre)

v10: Fix CI checkpatch warnings

v11: use correct pmdemand object pointer during hw readout,
     simplify the check for phys need update (Gustavo)

v12: Handle possible non serialize cases (Imre)
     Initialise also pmdemand params HW readout (Imre)
     Update active phys mask during sanitize calls (Imre)
     Check TC/encoder changes to limit connector update (Imre)

v13: Check display version before accessing pmdemand functions

v14: Move is_serialized to intel_global_state.c
     simplify update params and other stlying issues (Imre)

Bspec: 66451, 64636, 64602, 64603
Cc: Matt Atwood <matthew.s.atwood@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> #v4
Acked-by: Gustavo Sousa <gustavo.sousa@intel.com> #v11
Reviewed-by: Imre Deak <imre.deak@intel.com>
[RK: Fixed minor typo in one of the comments. s/qclck_gc/qclk_gv/]
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230606201032.347449-1-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |   1 +
 drivers/gpu/drm/i915/display/intel_display.c  |  14 +
 .../gpu/drm/i915/display/intel_display_core.h |   9 +
 .../drm/i915/display/intel_display_driver.c   |   7 +
 .../gpu/drm/i915/display/intel_display_irq.c  |  23 +-
 .../drm/i915/display/intel_display_power.c    |  14 +-
 .../gpu/drm/i915/display/intel_global_state.c |  12 +
 .../gpu/drm/i915/display/intel_global_state.h |   2 +
 .../drm/i915/display/intel_modeset_setup.c    |  32 +
 drivers/gpu/drm/i915/display/intel_pmdemand.c | 620 ++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_pmdemand.h |  67 ++
 drivers/gpu/drm/i915/i915_reg.h               |  26 +-
 12 files changed, 821 insertions(+), 6 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_pmdemand.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_pmdemand.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 1c9ed4c52760a..2cd8de174bf6b 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -269,6 +269,7 @@ i915-y += \
 	display/intel_pch_display.o \
 	display/intel_pch_refclk.o \
 	display/intel_plane_initial.o \
+	display/intel_pmdemand.o \
 	display/intel_psr.o \
 	display/intel_quirks.o \
 	display/intel_sprite.o \
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 34447d96fe570..5c7fdc82ac224 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -99,6 +99,7 @@
 #include "intel_pcode.h"
 #include "intel_pipe_crc.h"
 #include "intel_plane_initial.h"
+#include "intel_pmdemand.h"
 #include "intel_pps.h"
 #include "intel_psr.h"
 #include "intel_sdvo.h"
@@ -6354,6 +6355,10 @@ int intel_atomic_check(struct drm_device *dev,
 			return ret;
 	}
 
+	ret = intel_pmdemand_atomic_check(state);
+	if (ret)
+		goto fail;
+
 	ret = intel_atomic_check_crtcs(state);
 	if (ret)
 		goto fail;
@@ -6999,6 +7004,14 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 	for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i)
 		crtc->config = new_crtc_state;
 
+	/*
+	 * In XE_LPD+ Pmdemand combines many parameters such as voltage index,
+	 * plls, cdclk frequency, QGV point selection parameter etc. Voltage
+	 * index, cdclk/ddiclk frequencies are supposed to be configured before
+	 * the cdclk config is set.
+	 */
+	intel_pmdemand_pre_plane_update(state);
+
 	if (state->modeset) {
 		drm_atomic_helper_update_legacy_modeset_state(dev, &state->base);
 
@@ -7118,6 +7131,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 		intel_verify_planes(state);
 
 	intel_sagv_post_plane_update(state);
+	intel_pmdemand_post_plane_update(state);
 
 	drm_atomic_helper_commit_hw_done(&state->base);
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h
index dd8e08c8598f4..8d2243c71dd85 100644
--- a/drivers/gpu/drm/i915/display/intel_display_core.h
+++ b/drivers/gpu/drm/i915/display/intel_display_core.h
@@ -345,6 +345,15 @@ struct intel_display {
 		struct intel_global_obj obj;
 	} dbuf;
 
+	struct {
+		wait_queue_head_t waitqueue;
+
+		/* mutex to protect pmdemand programming sequence */
+		struct mutex lock;
+
+		struct intel_global_obj obj;
+	} pmdemand;
+
 	struct {
 		/*
 		 * dkl.phy_lock protects against concurrent access of the
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 60ce10fc72058..dc8de861339d1 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -47,6 +47,7 @@
 #include "intel_opregion.h"
 #include "intel_overlay.h"
 #include "intel_plane_initial.h"
+#include "intel_pmdemand.h"
 #include "intel_pps.h"
 #include "intel_quirks.h"
 #include "intel_vga.h"
@@ -211,6 +212,8 @@ int intel_display_driver_probe_noirq(struct drm_i915_private *i915)
 	if (ret < 0)
 		goto cleanup_vga;
 
+	intel_pmdemand_init_early(i915);
+
 	intel_power_domains_init_hw(i915, false);
 
 	if (!HAS_DISPLAY(i915))
@@ -240,6 +243,10 @@ int intel_display_driver_probe_noirq(struct drm_i915_private *i915)
 	if (ret)
 		goto cleanup_vga_client_pw_domain_dmc;
 
+	ret = intel_pmdemand_init(i915);
+	if (ret)
+		goto cleanup_vga_client_pw_domain_dmc;
+
 	init_llist_head(&i915->display.atomic_helper.free_list);
 	INIT_WORK(&i915->display.atomic_helper.free_work,
 		  intel_atomic_helper_free_state_worker);
diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index 31dd7c9c76b9c..ae2578741dfe8 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -18,6 +18,7 @@
 #include "intel_fifo_underrun.h"
 #include "intel_gmbus.h"
 #include "intel_hotplug_irq.h"
+#include "intel_pmdemand.h"
 #include "intel_psr.h"
 #include "intel_psr_regs.h"
 
@@ -827,12 +828,27 @@ static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv)
 		return GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
 }
 
+static void intel_pmdemand_irq_handler(struct drm_i915_private *dev_priv)
+{
+	wake_up_all(&dev_priv->display.pmdemand.waitqueue);
+}
+
 static void
 gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 {
 	bool found = false;
 
-	if (iir & GEN8_DE_MISC_GSE) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		if (iir & (XELPDP_PMDEMAND_RSP |
+			   XELPDP_PMDEMAND_RSPTOUT_ERR)) {
+			if (iir & XELPDP_PMDEMAND_RSPTOUT_ERR)
+				drm_dbg(&dev_priv->drm,
+					"Error waiting for Punit PM Demand Response\n");
+
+			intel_pmdemand_irq_handler(dev_priv);
+			found = true;
+		}
+	} else if (iir & GEN8_DE_MISC_GSE) {
 		intel_opregion_asle_intr(dev_priv);
 		found = true;
 	}
@@ -1576,7 +1592,10 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv))
 		de_port_masked |= BXT_DE_PORT_GMBUS;
 
-	if (DISPLAY_VER(dev_priv) >= 11) {
+	if (DISPLAY_VER(dev_priv) >= 14) {
+		de_misc_masked |= XELPDP_PMDEMAND_RSPTOUT_ERR |
+				  XELPDP_PMDEMAND_RSP;
+	} else if (DISPLAY_VER(dev_priv) >= 11) {
 		enum port port;
 
 		if (intel_bios_is_dsi_present(dev_priv, &port))
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 2f4f00ae2f578..db54370439042 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -20,6 +20,7 @@
 #include "intel_mchbar_regs.h"
 #include "intel_pch_refclk.h"
 #include "intel_pcode.h"
+#include "intel_pmdemand.h"
 #include "intel_pps_regs.h"
 #include "intel_snps_phy.h"
 #include "skl_watermark.h"
@@ -1082,20 +1083,29 @@ void gen9_dbuf_slices_update(struct drm_i915_private *dev_priv,
 
 static void gen9_dbuf_enable(struct drm_i915_private *dev_priv)
 {
+	u8 slices_mask;
+
 	dev_priv->display.dbuf.enabled_slices =
 		intel_enabled_dbuf_slices_mask(dev_priv);
 
+	slices_mask = BIT(DBUF_S1) | dev_priv->display.dbuf.enabled_slices;
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		intel_pmdemand_program_dbuf(dev_priv, slices_mask);
+
 	/*
 	 * Just power up at least 1 slice, we will
 	 * figure out later which slices we have and what we need.
 	 */
-	gen9_dbuf_slices_update(dev_priv, BIT(DBUF_S1) |
-				dev_priv->display.dbuf.enabled_slices);
+	gen9_dbuf_slices_update(dev_priv, slices_mask);
 }
 
 static void gen9_dbuf_disable(struct drm_i915_private *dev_priv)
 {
 	gen9_dbuf_slices_update(dev_priv, 0);
+
+	if (DISPLAY_VER(dev_priv) >= 14)
+		intel_pmdemand_program_dbuf(dev_priv, 0);
 }
 
 static void gen12_dbuf_slices_config(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c
index 02b593b1e2ea1..e8e8be54143b4 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.c
+++ b/drivers/gpu/drm/i915/display/intel_global_state.c
@@ -255,3 +255,15 @@ int intel_atomic_serialize_global_state(struct intel_global_state *obj_state)
 
 	return 0;
 }
+
+bool
+intel_atomic_global_state_is_serialized(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_crtc *crtc;
+
+	for_each_intel_crtc(&i915->drm, crtc)
+		if (!intel_atomic_get_new_crtc_state(state, crtc))
+			return false;
+	return true;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h
index f01ee0bb3e5a3..5477de8f0b30c 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.h
+++ b/drivers/gpu/drm/i915/display/intel_global_state.h
@@ -87,4 +87,6 @@ void intel_atomic_clear_global_state(struct intel_atomic_state *state);
 int intel_atomic_lock_global_state(struct intel_global_state *obj_state);
 int intel_atomic_serialize_global_state(struct intel_global_state *obj_state);
 
+bool intel_atomic_global_state_is_serialized(struct intel_atomic_state *state);
+
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
index 5ff99ca7f1de0..b8f43efb0ab5a 100644
--- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c
+++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c
@@ -26,6 +26,7 @@
 #include "intel_fifo_underrun.h"
 #include "intel_modeset_setup.h"
 #include "intel_pch_display.h"
+#include "intel_pmdemand.h"
 #include "intel_tc.h"
 #include "intel_vblank.h"
 #include "intel_wm.h"
@@ -115,6 +116,8 @@ static void set_encoder_for_connector(struct intel_connector *connector,
 static void reset_encoder_connector_state(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+	struct intel_pmdemand_state *pmdemand_state =
+		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
 	struct intel_connector *connector;
 	struct drm_connector_list_iter conn_iter;
 
@@ -123,6 +126,10 @@ static void reset_encoder_connector_state(struct intel_encoder *encoder)
 		if (connector->base.encoder != &encoder->base)
 			continue;
 
+		/* Clear the corresponding bit in pmdemand active phys mask */
+		intel_pmdemand_update_phys_mask(i915, encoder,
+						pmdemand_state, false);
+
 		set_encoder_for_connector(connector, NULL);
 
 		connector->base.dpms = DRM_MODE_DPMS_OFF;
@@ -151,6 +158,8 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 		to_intel_cdclk_state(i915->display.cdclk.obj.state);
 	struct intel_dbuf_state *dbuf_state =
 		to_intel_dbuf_state(i915->display.dbuf.obj.state);
+	struct intel_pmdemand_state *pmdemand_state =
+		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
 	struct intel_crtc_state *crtc_state =
 		to_intel_crtc_state(crtc->base.state);
 	enum pipe pipe = crtc->pipe;
@@ -174,6 +183,8 @@ static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc)
 
 	bw_state->data_rate[pipe] = 0;
 	bw_state->num_active_planes[pipe] = 0;
+
+	intel_pmdemand_update_port_clock(i915, pmdemand_state, pipe, 0);
 }
 
 /*
@@ -552,6 +563,8 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	struct intel_crtc_state *crtc_state = crtc ?
 		to_intel_crtc_state(crtc->base.state) : NULL;
+	struct intel_pmdemand_state *pmdemand_state =
+		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
 
 	/*
 	 * We need to check both for a crtc link (meaning that the encoder is
@@ -575,6 +588,10 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 			    encoder->base.base.id,
 			    encoder->base.name);
 
+		/* Clear the corresponding bit in pmdemand active phys mask */
+		intel_pmdemand_update_phys_mask(i915, encoder,
+						pmdemand_state, false);
+
 		/*
 		 * Connector is active, but has no active pipe. This is fallout
 		 * from our resume register restoring. Disable the encoder
@@ -661,6 +678,8 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 		to_intel_cdclk_state(i915->display.cdclk.obj.state);
 	struct intel_dbuf_state *dbuf_state =
 		to_intel_dbuf_state(i915->display.dbuf.obj.state);
+	struct intel_pmdemand_state *pmdemand_state =
+		to_intel_pmdemand_state(i915->display.pmdemand.obj.state);
 	enum pipe pipe;
 	struct intel_crtc *crtc;
 	struct intel_encoder *encoder;
@@ -724,7 +743,15 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 					intel_encoder_get_config(encoder, slave_crtc_state);
 				}
 			}
+
+			intel_pmdemand_update_phys_mask(i915, encoder,
+							pmdemand_state,
+							true);
 		} else {
+			intel_pmdemand_update_phys_mask(i915, encoder,
+							pmdemand_state,
+							false);
+
 			encoder->base.crtc = NULL;
 		}
 
@@ -841,8 +868,13 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915)
 		cdclk_state->min_voltage_level[crtc->pipe] =
 			crtc_state->min_voltage_level;
 
+		intel_pmdemand_update_port_clock(i915, pmdemand_state, pipe,
+						 crtc_state->port_clock);
+
 		intel_bw_crtc_update(bw_state, crtc_state);
 	}
+
+	intel_pmdemand_init_pmdemand_params(i915, pmdemand_state);
 }
 
 static void
diff --git a/drivers/gpu/drm/i915/display/intel_pmdemand.c b/drivers/gpu/drm/i915/display/intel_pmdemand.c
new file mode 100644
index 0000000000000..f7608d3636346
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_pmdemand.c
@@ -0,0 +1,620 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <linux/bitops.h>
+
+#include "i915_drv.h"
+#include "i915_reg.h"
+#include "intel_atomic.h"
+#include "intel_bw.h"
+#include "intel_cdclk.h"
+#include "intel_de.h"
+#include "intel_display_trace.h"
+#include "intel_pmdemand.h"
+#include "skl_watermark.h"
+
+static struct intel_global_state *
+intel_pmdemand_duplicate_state(struct intel_global_obj *obj)
+{
+	struct intel_pmdemand_state *pmdemand_state;
+
+	pmdemand_state = kmemdup(obj->state, sizeof(*pmdemand_state), GFP_KERNEL);
+	if (!pmdemand_state)
+		return NULL;
+
+	return &pmdemand_state->base;
+}
+
+static void intel_pmdemand_destroy_state(struct intel_global_obj *obj,
+					 struct intel_global_state *state)
+{
+	kfree(state);
+}
+
+static const struct intel_global_state_funcs intel_pmdemand_funcs = {
+	.atomic_duplicate_state = intel_pmdemand_duplicate_state,
+	.atomic_destroy_state = intel_pmdemand_destroy_state,
+};
+
+static struct intel_pmdemand_state *
+intel_atomic_get_pmdemand_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_global_state *pmdemand_state =
+		intel_atomic_get_global_obj_state(state,
+						  &i915->display.pmdemand.obj);
+
+	if (IS_ERR(pmdemand_state))
+		return ERR_CAST(pmdemand_state);
+
+	return to_intel_pmdemand_state(pmdemand_state);
+}
+
+static struct intel_pmdemand_state *
+intel_atomic_get_old_pmdemand_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_global_state *pmdemand_state =
+		intel_atomic_get_old_global_obj_state(state,
+						      &i915->display.pmdemand.obj);
+
+	if (!pmdemand_state)
+		return NULL;
+
+	return to_intel_pmdemand_state(pmdemand_state);
+}
+
+static struct intel_pmdemand_state *
+intel_atomic_get_new_pmdemand_state(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct intel_global_state *pmdemand_state =
+		intel_atomic_get_new_global_obj_state(state,
+						      &i915->display.pmdemand.obj);
+
+	if (!pmdemand_state)
+		return NULL;
+
+	return to_intel_pmdemand_state(pmdemand_state);
+}
+
+int intel_pmdemand_init(struct drm_i915_private *i915)
+{
+	struct intel_pmdemand_state *pmdemand_state;
+
+	pmdemand_state = kzalloc(sizeof(*pmdemand_state), GFP_KERNEL);
+	if (!pmdemand_state)
+		return -ENOMEM;
+
+	intel_atomic_global_obj_init(i915, &i915->display.pmdemand.obj,
+				     &pmdemand_state->base,
+				     &intel_pmdemand_funcs);
+
+	if (IS_MTL_DISPLAY_STEP(i915, STEP_A0, STEP_C0))
+		/* Wa_14016740474 */
+		intel_de_rmw(i915, XELPD_CHICKEN_DCPR_3, 0, DMD_RSP_TIMEOUT_DISABLE);
+
+	return 0;
+}
+
+void intel_pmdemand_init_early(struct drm_i915_private *i915)
+{
+	mutex_init(&i915->display.pmdemand.lock);
+	init_waitqueue_head(&i915->display.pmdemand.waitqueue);
+}
+
+void
+intel_pmdemand_update_phys_mask(struct drm_i915_private *i915,
+				struct intel_encoder *encoder,
+				struct intel_pmdemand_state *pmdemand_state,
+				bool set_bit)
+{
+	enum phy phy;
+
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	if (!encoder)
+		return;
+
+	phy = intel_port_to_phy(i915, encoder->port);
+	if (intel_phy_is_tc(i915, phy))
+		return;
+
+	if (set_bit)
+		pmdemand_state->active_combo_phys_mask |= BIT(phy);
+	else
+		pmdemand_state->active_combo_phys_mask &= ~BIT(phy);
+}
+
+void
+intel_pmdemand_update_port_clock(struct drm_i915_private *i915,
+				 struct intel_pmdemand_state *pmdemand_state,
+				 enum pipe pipe, int port_clock)
+{
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	pmdemand_state->ddi_clocks[pipe] = port_clock;
+}
+
+static void
+intel_pmdemand_update_max_ddiclk(struct drm_i915_private *i915,
+				 struct intel_atomic_state *state,
+				 struct intel_pmdemand_state *pmdemand_state)
+{
+	int max_ddiclk = 0;
+	const struct intel_crtc_state *new_crtc_state;
+	struct intel_crtc *crtc;
+	int i;
+
+	for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i)
+		intel_pmdemand_update_port_clock(i915, pmdemand_state,
+						 crtc->pipe,
+						 new_crtc_state->port_clock);
+
+	for (i = 0; i < ARRAY_SIZE(pmdemand_state->ddi_clocks); i++)
+		max_ddiclk = max(pmdemand_state->ddi_clocks[i], max_ddiclk);
+
+	pmdemand_state->params.ddiclk_max = DIV_ROUND_UP(max_ddiclk, 1000);
+}
+
+static void
+intel_pmdemand_update_connector_phys(struct drm_i915_private *i915,
+				     struct intel_atomic_state *state,
+				     struct drm_connector_state *conn_state,
+				     bool set_bit,
+				     struct intel_pmdemand_state *pmdemand_state)
+{
+	struct intel_encoder *encoder = to_intel_encoder(conn_state->best_encoder);
+	struct intel_crtc *crtc = to_intel_crtc(conn_state->crtc);
+	struct intel_crtc_state *crtc_state;
+
+	if (!crtc)
+		return;
+
+	if (set_bit)
+		crtc_state = intel_atomic_get_new_crtc_state(state, crtc);
+	else
+		crtc_state = intel_atomic_get_old_crtc_state(state, crtc);
+
+	if (!crtc_state->hw.active)
+		return;
+
+	intel_pmdemand_update_phys_mask(i915, encoder, pmdemand_state,
+					set_bit);
+}
+
+static void
+intel_pmdemand_update_active_non_tc_phys(struct drm_i915_private *i915,
+					 struct intel_atomic_state *state,
+					 struct intel_pmdemand_state *pmdemand_state)
+{
+	struct drm_connector_state *old_conn_state;
+	struct drm_connector_state *new_conn_state;
+	struct drm_connector *connector;
+	int i;
+
+	for_each_oldnew_connector_in_state(&state->base, connector,
+					   old_conn_state, new_conn_state, i) {
+		if (!intel_connector_needs_modeset(state, connector))
+			continue;
+
+		/* First clear the active phys in the old connector state */
+		intel_pmdemand_update_connector_phys(i915, state,
+						     old_conn_state, false,
+						     pmdemand_state);
+
+		/* Then set the active phys in new connector state */
+		intel_pmdemand_update_connector_phys(i915, state,
+						     new_conn_state, true,
+						     pmdemand_state);
+	}
+
+	pmdemand_state->params.active_phys =
+		min_t(u16, hweight16(pmdemand_state->active_combo_phys_mask),
+		      7);
+}
+
+static bool
+intel_pmdemand_encoder_has_tc_phy(struct drm_i915_private *i915,
+				  struct intel_encoder *encoder)
+{
+	enum phy phy;
+
+	if (!encoder)
+		return false;
+
+	phy = intel_port_to_phy(i915, encoder->port);
+
+	return intel_phy_is_tc(i915, phy);
+}
+
+static bool
+intel_pmdemand_connector_needs_update(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	struct drm_connector_state *old_conn_state;
+	struct drm_connector_state *new_conn_state;
+	struct drm_connector *connector;
+	int i;
+
+	for_each_oldnew_connector_in_state(&state->base, connector,
+					   old_conn_state, new_conn_state, i) {
+		struct intel_encoder *old_encoder =
+			to_intel_encoder(old_conn_state->best_encoder);
+		struct intel_encoder *new_encoder =
+			to_intel_encoder(new_conn_state->best_encoder);
+
+		if (!intel_connector_needs_modeset(state, connector))
+			continue;
+
+		if (old_encoder == new_encoder ||
+		    (intel_pmdemand_encoder_has_tc_phy(i915, old_encoder) &&
+		     intel_pmdemand_encoder_has_tc_phy(i915, new_encoder)))
+			continue;
+
+		return true;
+	}
+
+	return false;
+}
+
+static bool intel_pmdemand_needs_update(struct intel_atomic_state *state)
+{
+	const struct intel_bw_state *new_bw_state, *old_bw_state;
+	const struct intel_cdclk_state *new_cdclk_state, *old_cdclk_state;
+	const struct intel_crtc_state *new_crtc_state, *old_crtc_state;
+	const struct intel_dbuf_state *new_dbuf_state, *old_dbuf_state;
+	struct intel_crtc *crtc;
+	int i;
+
+	new_bw_state = intel_atomic_get_new_bw_state(state);
+	old_bw_state = intel_atomic_get_old_bw_state(state);
+	if (new_bw_state && new_bw_state->qgv_point_peakbw !=
+	    old_bw_state->qgv_point_peakbw)
+		return true;
+
+	new_dbuf_state = intel_atomic_get_new_dbuf_state(state);
+	old_dbuf_state = intel_atomic_get_old_dbuf_state(state);
+	if (new_dbuf_state &&
+	    (new_dbuf_state->active_pipes !=
+	     old_dbuf_state->active_pipes ||
+	     new_dbuf_state->enabled_slices !=
+	     old_dbuf_state->enabled_slices))
+		return true;
+
+	new_cdclk_state = intel_atomic_get_new_cdclk_state(state);
+	old_cdclk_state = intel_atomic_get_old_cdclk_state(state);
+	if (new_cdclk_state &&
+	    (new_cdclk_state->actual.cdclk !=
+	     old_cdclk_state->actual.cdclk ||
+	     new_cdclk_state->actual.voltage_level !=
+	     old_cdclk_state->actual.voltage_level))
+		return true;
+
+	for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
+					    new_crtc_state, i)
+		if (new_crtc_state->port_clock != old_crtc_state->port_clock)
+			return true;
+
+	return intel_pmdemand_connector_needs_update(state);
+}
+
+int intel_pmdemand_atomic_check(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_bw_state *new_bw_state;
+	const struct intel_cdclk_state *new_cdclk_state;
+	const struct intel_dbuf_state *new_dbuf_state;
+	struct intel_pmdemand_state *new_pmdemand_state;
+
+	if (DISPLAY_VER(i915) < 14)
+		return 0;
+
+	if (!intel_pmdemand_needs_update(state))
+		return 0;
+
+	new_pmdemand_state = intel_atomic_get_pmdemand_state(state);
+	if (IS_ERR(new_pmdemand_state))
+		return PTR_ERR(new_pmdemand_state);
+
+	new_bw_state = intel_atomic_get_bw_state(state);
+	if (IS_ERR(new_bw_state))
+		return PTR_ERR(new_bw_state);
+
+	/* firmware will calculate the qclk_gv_index, requirement is set to 0 */
+	new_pmdemand_state->params.qclk_gv_index = 0;
+	new_pmdemand_state->params.qclk_gv_bw = new_bw_state->qgv_point_peakbw;
+
+	new_dbuf_state = intel_atomic_get_dbuf_state(state);
+	if (IS_ERR(new_dbuf_state))
+		return PTR_ERR(new_dbuf_state);
+
+	new_pmdemand_state->params.active_pipes =
+		min_t(u8, hweight8(new_dbuf_state->active_pipes), 3);
+	new_pmdemand_state->params.active_dbufs =
+		min_t(u8, hweight8(new_dbuf_state->enabled_slices), 3);
+
+	new_cdclk_state = intel_atomic_get_cdclk_state(state);
+	if (IS_ERR(new_cdclk_state))
+		return PTR_ERR(new_cdclk_state);
+
+	new_pmdemand_state->params.voltage_index =
+		new_cdclk_state->actual.voltage_level;
+	new_pmdemand_state->params.cdclk_freq_mhz =
+		DIV_ROUND_UP(new_cdclk_state->actual.cdclk, 1000);
+
+	intel_pmdemand_update_max_ddiclk(i915, state, new_pmdemand_state);
+
+	intel_pmdemand_update_active_non_tc_phys(i915, state, new_pmdemand_state);
+
+	/*
+	 * Active_PLLs starts with 1 because of CDCLK PLL.
+	 * TODO: Missing to account genlock filter when it gets used.
+	 */
+	new_pmdemand_state->params.plls =
+		min_t(u16, new_pmdemand_state->params.active_phys + 1, 7);
+
+	/*
+	 * Setting scalers to max as it can not be calculated during flips and
+	 * fastsets without taking global states locks.
+	 */
+	new_pmdemand_state->params.scalers = 7;
+
+	if (state->base.allow_modeset)
+		return intel_atomic_serialize_global_state(&new_pmdemand_state->base);
+	else
+		return intel_atomic_lock_global_state(&new_pmdemand_state->base);
+}
+
+static bool intel_pmdemand_check_prev_transaction(struct drm_i915_private *i915)
+{
+	return !(intel_de_wait_for_clear(i915,
+					 XELPDP_INITIATE_PMDEMAND_REQUEST(1),
+					 XELPDP_PMDEMAND_REQ_ENABLE, 10) ||
+		 intel_de_wait_for_clear(i915,
+					 GEN12_DCPR_STATUS_1,
+					 XELPDP_PMDEMAND_INFLIGHT_STATUS, 10));
+}
+
+void
+intel_pmdemand_init_pmdemand_params(struct drm_i915_private *i915,
+				    struct intel_pmdemand_state *pmdemand_state)
+{
+	u32 reg1, reg2;
+
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	mutex_lock(&i915->display.pmdemand.lock);
+	if (drm_WARN_ON(&i915->drm,
+			!intel_pmdemand_check_prev_transaction(i915))) {
+		memset(&pmdemand_state->params, 0,
+		       sizeof(pmdemand_state->params));
+		goto unlock;
+	}
+
+	reg1 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0));
+
+	reg2 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1));
+
+	/* Set 1*/
+	pmdemand_state->params.qclk_gv_bw =
+		REG_FIELD_GET(XELPDP_PMDEMAND_QCLK_GV_BW_MASK, reg1);
+	pmdemand_state->params.voltage_index =
+		REG_FIELD_GET(XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK, reg1);
+	pmdemand_state->params.qclk_gv_index =
+		REG_FIELD_GET(XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK, reg1);
+	pmdemand_state->params.active_pipes =
+		REG_FIELD_GET(XELPDP_PMDEMAND_PIPES_MASK, reg1);
+	pmdemand_state->params.active_dbufs =
+		REG_FIELD_GET(XELPDP_PMDEMAND_DBUFS_MASK, reg1);
+	pmdemand_state->params.active_phys =
+		REG_FIELD_GET(XELPDP_PMDEMAND_PHYS_MASK, reg1);
+
+	/* Set 2*/
+	pmdemand_state->params.cdclk_freq_mhz =
+		REG_FIELD_GET(XELPDP_PMDEMAND_CDCLK_FREQ_MASK, reg2);
+	pmdemand_state->params.ddiclk_max =
+		REG_FIELD_GET(XELPDP_PMDEMAND_DDICLK_FREQ_MASK, reg2);
+	pmdemand_state->params.scalers =
+		REG_FIELD_GET(XELPDP_PMDEMAND_SCALERS_MASK, reg2);
+
+unlock:
+	mutex_unlock(&i915->display.pmdemand.lock);
+}
+
+static bool intel_pmdemand_req_complete(struct drm_i915_private *i915)
+{
+	return !(intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1)) &
+		 XELPDP_PMDEMAND_REQ_ENABLE);
+}
+
+static void intel_pmdemand_wait(struct drm_i915_private *i915)
+{
+	if (!wait_event_timeout(i915->display.pmdemand.waitqueue,
+				intel_pmdemand_req_complete(i915),
+				msecs_to_jiffies_timeout(10)))
+		drm_err(&i915->drm,
+			"timed out waiting for Punit PM Demand Response\n");
+}
+
+/* Required to be programmed during Display Init Sequences. */
+void intel_pmdemand_program_dbuf(struct drm_i915_private *i915,
+				 u8 dbuf_slices)
+{
+	u32 dbufs = min_t(u32, hweight8(dbuf_slices), 3);
+
+	mutex_lock(&i915->display.pmdemand.lock);
+	if (drm_WARN_ON(&i915->drm,
+			!intel_pmdemand_check_prev_transaction(i915)))
+		goto unlock;
+
+	intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0),
+		     XELPDP_PMDEMAND_DBUFS_MASK,
+		     REG_FIELD_PREP(XELPDP_PMDEMAND_DBUFS_MASK, dbufs));
+	intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1), 0,
+		     XELPDP_PMDEMAND_REQ_ENABLE);
+
+	intel_pmdemand_wait(i915);
+
+unlock:
+	mutex_unlock(&i915->display.pmdemand.lock);
+}
+
+static void
+intel_pmdemand_update_params(const struct intel_pmdemand_state *new,
+			     const struct intel_pmdemand_state *old,
+			     u32 *reg1, u32 *reg2, bool serialized)
+{
+	/*
+	 * The pmdemand parameter updates happens in two steps. Pre plane and
+	 * post plane updates. During the pre plane, as DE might still be
+	 * handling with some old operations, to avoid unexpected performance
+	 * issues, program the pmdemand parameters with higher of old and new
+	 * values. And then after once settled, use the new parameter values
+	 * as part of the post plane update.
+	 *
+	 * If the pmdemand params update happens without modeset allowed, this
+	 * means we can't serialize the updates. So that implies possibility of
+	 * some parallel atomic commits affecting the pmdemand parameters. In
+	 * that case, we need to consider the current values from the register
+	 * as well. So in pre-plane case, we need to check the max of old, new
+	 * and current register value if not serialized. In post plane update
+	 * we need to consider max of new and current register value if not
+	 * serialized
+	 */
+
+#define update_reg(reg, field, mask) do { \
+	u32 current_val = serialized ? 0 : REG_FIELD_GET((mask), *(reg)); \
+	u32 old_val = old ? old->params.field : 0; \
+	u32 new_val = new->params.field; \
+\
+	*(reg) &= ~(mask); \
+	*(reg) |= REG_FIELD_PREP((mask), max3(old_val, new_val, current_val)); \
+} while (0)
+
+	/* Set 1*/
+	update_reg(reg1, qclk_gv_bw, XELPDP_PMDEMAND_QCLK_GV_BW_MASK);
+	update_reg(reg1, voltage_index, XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK);
+	update_reg(reg1, qclk_gv_index, XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK);
+	update_reg(reg1, active_pipes, XELPDP_PMDEMAND_PIPES_MASK);
+	update_reg(reg1, active_dbufs, XELPDP_PMDEMAND_DBUFS_MASK);
+	update_reg(reg1, active_phys, XELPDP_PMDEMAND_PHYS_MASK);
+
+	/* Set 2*/
+	update_reg(reg2, cdclk_freq_mhz, XELPDP_PMDEMAND_CDCLK_FREQ_MASK);
+	update_reg(reg2, ddiclk_max, XELPDP_PMDEMAND_DDICLK_FREQ_MASK);
+	update_reg(reg2, scalers, XELPDP_PMDEMAND_SCALERS_MASK);
+	update_reg(reg2, plls, XELPDP_PMDEMAND_PLLS_MASK);
+
+#undef update_reg
+}
+
+static void
+intel_pmdemand_program_params(struct drm_i915_private *i915,
+			      const struct intel_pmdemand_state *new,
+			      const struct intel_pmdemand_state *old,
+			      bool serialized)
+{
+	bool changed = false;
+	u32 reg1, mod_reg1;
+	u32 reg2, mod_reg2;
+
+	mutex_lock(&i915->display.pmdemand.lock);
+	if (drm_WARN_ON(&i915->drm,
+			!intel_pmdemand_check_prev_transaction(i915)))
+		goto unlock;
+
+	reg1 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0));
+	mod_reg1 = reg1;
+
+	reg2 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1));
+	mod_reg2 = reg2;
+
+	intel_pmdemand_update_params(new, old, &mod_reg1, &mod_reg2,
+				     serialized);
+
+	if (reg1 != mod_reg1) {
+		intel_de_write(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0),
+			       mod_reg1);
+		changed = true;
+	}
+
+	if (reg2 != mod_reg2) {
+		intel_de_write(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1),
+			       mod_reg2);
+		changed = true;
+	}
+
+	/* Initiate pm demand request only if register values are changed */
+	if (!changed)
+		goto unlock;
+
+	drm_dbg_kms(&i915->drm,
+		    "initate pmdemand request values: (0x%x 0x%x)\n",
+		    mod_reg1, mod_reg2);
+
+	intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1), 0,
+		     XELPDP_PMDEMAND_REQ_ENABLE);
+
+	intel_pmdemand_wait(i915);
+
+unlock:
+	mutex_unlock(&i915->display.pmdemand.lock);
+}
+
+static bool
+intel_pmdemand_state_changed(const struct intel_pmdemand_state *new,
+			     const struct intel_pmdemand_state *old)
+{
+	return memcmp(&new->params, &old->params, sizeof(new->params)) != 0;
+}
+
+void intel_pmdemand_pre_plane_update(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_pmdemand_state *new_pmdemand_state =
+		intel_atomic_get_new_pmdemand_state(state);
+	const struct intel_pmdemand_state *old_pmdemand_state =
+		intel_atomic_get_old_pmdemand_state(state);
+
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	if (!new_pmdemand_state ||
+	    !intel_pmdemand_state_changed(new_pmdemand_state,
+					  old_pmdemand_state))
+		return;
+
+	WARN_ON(!new_pmdemand_state->base.changed);
+
+	intel_pmdemand_program_params(i915, new_pmdemand_state,
+				      old_pmdemand_state,
+				      intel_atomic_global_state_is_serialized(state));
+}
+
+void intel_pmdemand_post_plane_update(struct intel_atomic_state *state)
+{
+	struct drm_i915_private *i915 = to_i915(state->base.dev);
+	const struct intel_pmdemand_state *new_pmdemand_state =
+		intel_atomic_get_new_pmdemand_state(state);
+	const struct intel_pmdemand_state *old_pmdemand_state =
+		intel_atomic_get_old_pmdemand_state(state);
+
+	if (DISPLAY_VER(i915) < 14)
+		return;
+
+	if (!new_pmdemand_state ||
+	    !intel_pmdemand_state_changed(new_pmdemand_state,
+					  old_pmdemand_state))
+		return;
+
+	WARN_ON(!new_pmdemand_state->base.changed);
+
+	intel_pmdemand_program_params(i915, new_pmdemand_state, NULL,
+				      intel_atomic_global_state_is_serialized(state));
+}
diff --git a/drivers/gpu/drm/i915/display/intel_pmdemand.h b/drivers/gpu/drm/i915/display/intel_pmdemand.h
new file mode 100644
index 0000000000000..2941a1a18b72b
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_pmdemand.h
@@ -0,0 +1,67 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef __INTEL_PMDEMAND_H__
+#define __INTEL_PMDEMAND_H__
+
+#include "intel_display_limits.h"
+#include "intel_global_state.h"
+
+struct drm_i915_private;
+struct intel_atomic_state;
+struct intel_crtc_state;
+struct intel_encoder;
+struct intel_plane_state;
+
+struct pmdemand_params {
+	u16 qclk_gv_bw;
+	u8 voltage_index;
+	u8 qclk_gv_index;
+	u8 active_pipes;
+	u8 active_dbufs;
+	/* Total number of non type C active phys from active_phys_mask */
+	u8 active_phys;
+	u8 plls;
+	u16 cdclk_freq_mhz;
+	/* max from ddi_clocks[] */
+	u16 ddiclk_max;
+	u8 scalers;
+};
+
+struct intel_pmdemand_state {
+	struct intel_global_state base;
+
+	/* Maintain a persistent list of port clocks across all crtcs */
+	int ddi_clocks[I915_MAX_PIPES];
+
+	/* Maintain a persistent list of non type C phys mask */
+	u16 active_combo_phys_mask;
+
+	/* Parameters to be configured in the pmdemand registers */
+	struct pmdemand_params params;
+};
+
+#define to_intel_pmdemand_state(x) container_of((x), \
+						struct intel_pmdemand_state, \
+						base)
+
+void intel_pmdemand_init_early(struct drm_i915_private *i915);
+int intel_pmdemand_init(struct drm_i915_private *i915);
+void intel_pmdemand_init_pmdemand_params(struct drm_i915_private *i915,
+					 struct intel_pmdemand_state *pmdemand_state);
+void intel_pmdemand_update_port_clock(struct drm_i915_private *i915,
+				      struct intel_pmdemand_state *pmdemand_state,
+				      enum pipe pipe, int port_clock);
+void intel_pmdemand_update_phys_mask(struct drm_i915_private *i915,
+				     struct intel_encoder *encoder,
+				     struct intel_pmdemand_state *pmdemand_state,
+				     bool clear_bit);
+void intel_pmdemand_program_dbuf(struct drm_i915_private *i915,
+				 u8 dbuf_slices);
+void intel_pmdemand_pre_plane_update(struct intel_atomic_state *state);
+void intel_pmdemand_post_plane_update(struct intel_atomic_state *state);
+int intel_pmdemand_atomic_check(struct intel_atomic_state *state);
+
+#endif /* __INTEL_PMDEMAND_H__ */
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 0523418129c51..27d75215d3eb7 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4418,8 +4418,10 @@
 #define GEN8_DE_MISC_IMR _MMIO(0x44464)
 #define GEN8_DE_MISC_IIR _MMIO(0x44468)
 #define GEN8_DE_MISC_IER _MMIO(0x4446c)
-#define  GEN8_DE_MISC_GSE		(1 << 27)
-#define  GEN8_DE_EDP_PSR		(1 << 19)
+#define  XELPDP_PMDEMAND_RSPTOUT_ERR	REG_BIT(27)
+#define  GEN8_DE_MISC_GSE		REG_BIT(27)
+#define  GEN8_DE_EDP_PSR		REG_BIT(19)
+#define  XELPDP_PMDEMAND_RSP		REG_BIT(3)
 
 #define GEN8_PCU_ISR _MMIO(0x444e0)
 #define GEN8_PCU_IMR _MMIO(0x444e4)
@@ -4504,6 +4506,23 @@
 #define  XELPDP_DP_ALT_HPD_LONG_DETECT		REG_BIT(1)
 #define  XELPDP_DP_ALT_HPD_SHORT_DETECT		REG_BIT(0)
 
+#define XELPDP_INITIATE_PMDEMAND_REQUEST(dword)		_MMIO(0x45230 + 4 * (dword))
+#define  XELPDP_PMDEMAND_QCLK_GV_BW_MASK		REG_GENMASK(31, 16)
+#define  XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK		REG_GENMASK(14, 12)
+#define  XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK		REG_GENMASK(11, 8)
+#define  XELPDP_PMDEMAND_PIPES_MASK			REG_GENMASK(7, 6)
+#define  XELPDP_PMDEMAND_DBUFS_MASK			REG_GENMASK(5, 4)
+#define  XELPDP_PMDEMAND_PHYS_MASK			REG_GENMASK(2, 0)
+
+#define  XELPDP_PMDEMAND_REQ_ENABLE			REG_BIT(31)
+#define  XELPDP_PMDEMAND_CDCLK_FREQ_MASK		REG_GENMASK(30, 20)
+#define  XELPDP_PMDEMAND_DDICLK_FREQ_MASK		REG_GENMASK(18, 8)
+#define  XELPDP_PMDEMAND_SCALERS_MASK			REG_GENMASK(6, 4)
+#define  XELPDP_PMDEMAND_PLLS_MASK			REG_GENMASK(2, 0)
+
+#define GEN12_DCPR_STATUS_1				_MMIO(0x46440)
+#define  XELPDP_PMDEMAND_INFLIGHT_STATUS		REG_BIT(26)
+
 #define ILK_DISPLAY_CHICKEN2	_MMIO(0x42004)
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
 #define   ILK_ELPIN_409_SELECT	REG_BIT(25)
@@ -4663,6 +4682,9 @@
 #define   DCPR_SEND_RESP_IMM			REG_BIT(25)
 #define   DCPR_CLEAR_MEMSTAT_DIS		REG_BIT(24)
 
+#define XELPD_CHICKEN_DCPR_3			_MMIO(0x46438)
+#define   DMD_RSP_TIMEOUT_DISABLE		REG_BIT(19)
+
 #define SKL_DFSM			_MMIO(0x51000)
 #define   SKL_DFSM_DISPLAY_PM_DISABLE	(1 << 27)
 #define   SKL_DFSM_DISPLAY_HDCP_DISABLE	(1 << 25)
-- 
GitLab


From 9cc31938d4586f72eb8e0235ad9d9eb22496fcee Mon Sep 17 00:00:00 2001
From: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Date: Mon, 5 Jun 2023 12:39:22 -0700
Subject: [PATCH 0857/1791] i915/perf: Drop the aging_tail logic in perf OA

On DG2, capturing OA reports while running heavy render workloads
sometimes results in invalid OA reports where 64-byte chunks inside
reports have stale values. Under memory pressure, high OA sampling rates
(13.3 us) and heavy render workload, occasionally, the OA HW TAIL
pointer does not progress as fast as the sampling rate. When these
glitches occur, the TAIL pointer takes approx. 200us to progress.  While
this is expected behavior from the HW perspective, invalid reports are
not expected.

In oa_buffer_check_unlocked(), when we execute the if condition, we are
updating the oa_buffer.tail to the aging tail and then setting pollin
based on this tail value, however, we do not have a chance to rewind and
validate the reports prior to setting pollin. The validation happens
in a subsequent call to oa_buffer_check_unlocked(). If a read occurs
before this validation, then we end up reading reports up until this
oa_buffer.tail value which includes invalid reports. Though found on
DG2, this affects all platforms.

The aging tail logic is no longer necessary since we are explicitly
checking for landed reports.

Start by dropping the aging tail logic.

v2:
- Drop extra blank line
- Add reason to drop aging logic (Ashutosh)
- Add bug links (Ashutosh)
- rename aged_tail to read_tail
- Squash patches 3 and 1

v3: (Ashutosh)
- Remove extra spaces
- Remove gtt_offset from the pollin calculation
- s/Bug:/Link/ in commit message (checkpatch)

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7484
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/7757
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230605193923.1836048-2-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_perf.c       | 95 +++++++++++---------------
 drivers/gpu/drm/i915/i915_perf_types.h | 12 ----
 2 files changed, 38 insertions(+), 69 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 58284156428dc..a8d43bf1f6d5a 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -531,8 +531,7 @@ static void oa_context_id_squash(struct i915_perf_stream *stream, u32 *report)
  * (See description of OA_TAIL_MARGIN_NSEC above for further details.)
  *
  * Besides returning true when there is data available to read() this function
- * also updates the tail, aging_tail and aging_timestamp in the oa_buffer
- * object.
+ * also updates the tail in the oa_buffer object.
  *
  * Note: It's safe to read OA config state here unlocked, assuming that this is
  * only called while the stream is enabled, while the global OA configuration
@@ -544,10 +543,10 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream)
 {
 	u32 gtt_offset = i915_ggtt_offset(stream->oa_buffer.vma);
 	int report_size = stream->oa_buffer.format->size;
+	u32 head, tail, read_tail;
 	unsigned long flags;
 	bool pollin;
 	u32 hw_tail;
-	u64 now;
 	u32 partial_report_size;
 
 	/* We have to consider the (unlikely) possibility that read() errors
@@ -568,62 +567,47 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream)
 	/* Subtract partial amount off the tail */
 	hw_tail = gtt_offset + OA_TAKEN(hw_tail, partial_report_size);
 
-	now = ktime_get_mono_fast_ns();
-
-	if (hw_tail == stream->oa_buffer.aging_tail &&
-	    (now - stream->oa_buffer.aging_timestamp) > OA_TAIL_MARGIN_NSEC) {
-		/* If the HW tail hasn't move since the last check and the HW
-		 * tail has been aging for long enough, declare it the new
-		 * tail.
-		 */
-		stream->oa_buffer.tail = stream->oa_buffer.aging_tail;
-	} else {
-		u32 head, tail, aged_tail;
-
-		/* NB: The head we observe here might effectively be a little
-		 * out of date. If a read() is in progress, the head could be
-		 * anywhere between this head and stream->oa_buffer.tail.
-		 */
-		head = stream->oa_buffer.head - gtt_offset;
-		aged_tail = stream->oa_buffer.tail - gtt_offset;
-
-		hw_tail -= gtt_offset;
-		tail = hw_tail;
-
-		/* Walk the stream backward until we find a report with report
-		 * id and timestmap not at 0. Since the circular buffer pointers
-		 * progress by increments of 64 bytes and that reports can be up
-		 * to 256 bytes long, we can't tell whether a report has fully
-		 * landed in memory before the report id and timestamp of the
-		 * following report have effectively landed.
-		 *
-		 * This is assuming that the writes of the OA unit land in
-		 * memory in the order they were written to.
-		 * If not : (╯°□°）╯︵ ┻━┻
-		 */
-		while (OA_TAKEN(tail, aged_tail) >= report_size) {
-			void *report = stream->oa_buffer.vaddr + tail;
+	/* NB: The head we observe here might effectively be a little
+	 * out of date. If a read() is in progress, the head could be
+	 * anywhere between this head and stream->oa_buffer.tail.
+	 */
+	head = stream->oa_buffer.head - gtt_offset;
+	read_tail = stream->oa_buffer.tail - gtt_offset;
+
+	hw_tail -= gtt_offset;
+	tail = hw_tail;
+
+	/* Walk the stream backward until we find a report with report
+	 * id and timestmap not at 0. Since the circular buffer pointers
+	 * progress by increments of 64 bytes and that reports can be up
+	 * to 256 bytes long, we can't tell whether a report has fully
+	 * landed in memory before the report id and timestamp of the
+	 * following report have effectively landed.
+	 *
+	 * This is assuming that the writes of the OA unit land in
+	 * memory in the order they were written to.
+	 * If not : (╯°□°）╯︵ ┻━┻
+	 */
+	while (OA_TAKEN(tail, read_tail) >= report_size) {
+		void *report = stream->oa_buffer.vaddr + tail;
 
-			if (oa_report_id(stream, report) ||
-			    oa_timestamp(stream, report))
-				break;
+		if (oa_report_id(stream, report) ||
+		    oa_timestamp(stream, report))
+			break;
 
-			tail = (tail - report_size) & (OA_BUFFER_SIZE - 1);
-		}
+		tail = (tail - report_size) & (OA_BUFFER_SIZE - 1);
+	}
 
-		if (OA_TAKEN(hw_tail, tail) > report_size &&
-		    __ratelimit(&stream->perf->tail_pointer_race))
-			drm_notice(&stream->uncore->i915->drm,
-				   "unlanded report(s) head=0x%x tail=0x%x hw_tail=0x%x\n",
-				   head, tail, hw_tail);
+	if (OA_TAKEN(hw_tail, tail) > report_size &&
+	    __ratelimit(&stream->perf->tail_pointer_race))
+		drm_notice(&stream->uncore->i915->drm,
+			   "unlanded report(s) head=0x%x tail=0x%x hw_tail=0x%x\n",
+		 head, tail, hw_tail);
 
-		stream->oa_buffer.tail = gtt_offset + tail;
-		stream->oa_buffer.aging_tail = gtt_offset + hw_tail;
-		stream->oa_buffer.aging_timestamp = now;
-	}
+	stream->oa_buffer.tail = gtt_offset + tail;
 
-	pollin = OA_TAKEN(stream->oa_buffer.tail - gtt_offset,
-			  stream->oa_buffer.head - gtt_offset) >= report_size;
+	pollin = OA_TAKEN(stream->oa_buffer.tail,
+			  stream->oa_buffer.head) >= report_size;
 
 	spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags);
 
@@ -1727,7 +1711,6 @@ static void gen7_init_oa_buffer(struct i915_perf_stream *stream)
 			   gtt_offset | OABUFFER_SIZE_16M);
 
 	/* Mark that we need updated tail pointers to read from... */
-	stream->oa_buffer.aging_tail = INVALID_TAIL_PTR;
 	stream->oa_buffer.tail = gtt_offset;
 
 	spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags);
@@ -1779,7 +1762,6 @@ static void gen8_init_oa_buffer(struct i915_perf_stream *stream)
 	intel_uncore_write(uncore, GEN8_OATAILPTR, gtt_offset & GEN8_OATAILPTR_MASK);
 
 	/* Mark that we need updated tail pointers to read from... */
-	stream->oa_buffer.aging_tail = INVALID_TAIL_PTR;
 	stream->oa_buffer.tail = gtt_offset;
 
 	/*
@@ -1833,7 +1815,6 @@ static void gen12_init_oa_buffer(struct i915_perf_stream *stream)
 			   gtt_offset & GEN12_OAG_OATAILPTR_MASK);
 
 	/* Mark that we need updated tail pointers to read from... */
-	stream->oa_buffer.aging_tail = INVALID_TAIL_PTR;
 	stream->oa_buffer.tail = gtt_offset;
 
 	/*
diff --git a/drivers/gpu/drm/i915/i915_perf_types.h b/drivers/gpu/drm/i915/i915_perf_types.h
index 66dd5f74de056..fe3a5dae8c228 100644
--- a/drivers/gpu/drm/i915/i915_perf_types.h
+++ b/drivers/gpu/drm/i915/i915_perf_types.h
@@ -312,18 +312,6 @@ struct i915_perf_stream {
 		 */
 		spinlock_t ptr_lock;
 
-		/**
-		 * @aging_tail: The last HW tail reported by HW. The data
-		 * might not have made it to memory yet though.
-		 */
-		u32 aging_tail;
-
-		/**
-		 * @aging_timestamp: A monotonic timestamp for when the current aging tail pointer
-		 * was read; used to determine when it is old enough to trust.
-		 */
-		u64 aging_timestamp;
-
 		/**
 		 * @head: Although we can always read back the head pointer register,
 		 * we prefer to avoid trusting the HW state, just to avoid any
-- 
GitLab


From 589f4924e815559282e7c9f2854cd6b53ee9d6af Mon Sep 17 00:00:00 2001
From: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Date: Mon, 5 Jun 2023 12:39:23 -0700
Subject: [PATCH 0858/1791] i915/perf: Do not add ggtt offset to hw_tail

ggtt offset for hw_tail is not required for the calculations, so drop
it.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230605193923.1836048-3-umesh.nerlige.ramappa@intel.com
---
 drivers/gpu/drm/i915/i915_perf.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index a8d43bf1f6d5a..0a111b281578b 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -565,7 +565,7 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream)
 	partial_report_size %= report_size;
 
 	/* Subtract partial amount off the tail */
-	hw_tail = gtt_offset + OA_TAKEN(hw_tail, partial_report_size);
+	hw_tail = OA_TAKEN(hw_tail, partial_report_size);
 
 	/* NB: The head we observe here might effectively be a little
 	 * out of date. If a read() is in progress, the head could be
@@ -574,7 +574,6 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream)
 	head = stream->oa_buffer.head - gtt_offset;
 	read_tail = stream->oa_buffer.tail - gtt_offset;
 
-	hw_tail -= gtt_offset;
 	tail = hw_tail;
 
 	/* Walk the stream backward until we find a report with report
-- 
GitLab


From a03b7c4698d78b2e4d0c06a439d01cce8a3823bc Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 4 Jun 2023 06:13:07 +0300
Subject: [PATCH 0859/1791] drm/msm/dpu: use PINGPONG_NONE to unbind INTF from
 PP

Currently the driver passes the PINGPONG index to
dpu_hw_intf_ops::bind_pingpong_blk() callback and uses separate boolean
flag to tell whether INTF should be bound or unbound. Simplify this by
passing PINGPONG_NONE in case of unbinding and drop the flag completely.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/540968/
Link: https://lore.kernel.org/r/20230604031308.894274-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c          | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c | 4 +---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c | 1 -
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c          | 3 +--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h          | 1 -
 5 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index f32882e83255d..d219a13599069 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2090,8 +2090,8 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc)
 		for (i = 0; i < dpu_enc->num_phys_encs; i++) {
 			if (dpu_enc->phys_encs[i] && phys_enc->hw_intf->ops.bind_pingpong_blk)
 				phys_enc->hw_intf->ops.bind_pingpong_blk(
-						dpu_enc->phys_encs[i]->hw_intf, false,
-						dpu_enc->phys_encs[i]->hw_pp->idx);
+						dpu_enc->phys_encs[i]->hw_intf,
+						PINGPONG_NONE);
 
 			/* mark INTF flush as pending */
 			if (phys_enc->hw_ctl->ops.update_pending_flush_intf)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 4f8c9187f76dc..107f0eff958cd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -66,7 +66,6 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg(
 	if (test_bit(DPU_CTL_ACTIVE_CFG, &ctl->caps->features) && phys_enc->hw_intf->ops.bind_pingpong_blk)
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
 				phys_enc->hw_intf,
-				true,
 				phys_enc->hw_pp->idx);
 }
 
@@ -553,8 +552,7 @@ static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc)
 	if (phys_enc->hw_intf->ops.bind_pingpong_blk) {
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
 				phys_enc->hw_intf,
-				false,
-				phys_enc->hw_pp->idx);
+				PINGPONG_NONE);
 
 		ctl = phys_enc->hw_ctl;
 		ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
index e26629e9e3030..662d74ded1b94 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c
@@ -287,7 +287,6 @@ static void dpu_encoder_phys_vid_setup_timing_engine(
 	if (phys_enc->hw_intf->ops.bind_pingpong_blk)
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
 				phys_enc->hw_intf,
-				true,
 				phys_enc->hw_pp->idx);
 
 	if (phys_enc->hw_pp->merge_3d)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index 5dce18236d87d..530f82e34c1ec 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -267,7 +267,6 @@ static void dpu_hw_intf_setup_prg_fetch(
 
 static void dpu_hw_intf_bind_pingpong_blk(
 		struct dpu_hw_intf *intf,
-		bool enable,
 		const enum dpu_pingpong pp)
 {
 	struct dpu_hw_blk_reg_map *c = &intf->hw;
@@ -276,7 +275,7 @@ static void dpu_hw_intf_bind_pingpong_blk(
 	mux_cfg = DPU_REG_READ(c, INTF_MUX);
 	mux_cfg &= ~0xf;
 
-	if (enable)
+	if (pp)
 		mux_cfg |= (pp - PINGPONG_0) & 0x7;
 	else
 		mux_cfg |= 0xf;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
index 73b0885918f88..33895eca12119 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
@@ -88,7 +88,6 @@ struct dpu_hw_intf_ops {
 	u32 (*get_line_count)(struct dpu_hw_intf *intf);
 
 	void (*bind_pingpong_blk)(struct dpu_hw_intf *intf,
-			bool enable,
 			const enum dpu_pingpong pp);
 	void (*setup_misr)(struct dpu_hw_intf *intf, bool enable, u32 frame_count);
 	int (*collect_misr)(struct dpu_hw_intf *intf, u32 *misr_value);
-- 
GitLab


From 0f86d9c980a3e4cc8b8ca604cfb5ac3537c3fba7 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 4 Jun 2023 06:13:08 +0300
Subject: [PATCH 0860/1791] drm/msm/dpu: use PINGPONG_NONE to unbind WB from PP

Currently the driver passes the PINGPONG index to
dpu_hw_wb_ops::bind_pingpong_blk() callback and uses separate boolean
flag to tell whether WB should be bound or unbound. Simplify this by
passing PINGPONG_NONE in case of unbinding and drop the flag completely.

Suggested-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/540969/
Link: https://lore.kernel.org/r/20230604031308.894274-2-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c         | 3 +--
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c           | 4 ++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h           | 2 +-
 4 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
index d219a13599069..625f89c4d3dee 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
@@ -2080,8 +2080,7 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc)
 	if (phys_enc->hw_wb) {
 		/* disable the PP block */
 		if (phys_enc->hw_wb->ops.bind_pingpong_blk)
-			phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, false,
-					phys_enc->hw_pp->idx);
+			phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, PINGPONG_NONE);
 
 		/* mark WB flush as pending */
 		if (phys_enc->hw_ctl->ops.update_pending_flush_wb)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index e9325cafb1a85..a466ff70a4d62 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -210,7 +210,7 @@ static void dpu_encoder_phys_wb_setup_cdp(struct dpu_encoder_phys *phys_enc)
 
 		/* setup which pp blk will connect to this wb */
 		if (hw_pp && phys_enc->hw_wb->ops.bind_pingpong_blk)
-			phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, true,
+			phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb,
 					phys_enc->hw_pp->idx);
 
 		phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg);
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
index dcffd6cc47fc8..ebc4164003822 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c
@@ -154,7 +154,7 @@ static void dpu_hw_wb_setup_cdp(struct dpu_hw_wb *ctx,
 
 static void dpu_hw_wb_bind_pingpong_blk(
 		struct dpu_hw_wb *ctx,
-		bool enable, const enum dpu_pingpong pp)
+		const enum dpu_pingpong pp)
 {
 	struct dpu_hw_blk_reg_map *c;
 	int mux_cfg;
@@ -167,7 +167,7 @@ static void dpu_hw_wb_bind_pingpong_blk(
 	mux_cfg = DPU_REG_READ(c, WB_MUX);
 	mux_cfg &= ~0xf;
 
-	if (enable)
+	if (pp)
 		mux_cfg |= (pp - PINGPONG_0) & 0x7;
 	else
 		mux_cfg |= 0xf;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
index c7f792eeb55ca..2d7db2efa3d01 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h
@@ -49,7 +49,7 @@ struct dpu_hw_wb_ops {
 			  bool enable);
 
 	void (*bind_pingpong_blk)(struct dpu_hw_wb *ctx,
-			bool enable, const enum dpu_pingpong pp);
+				  const enum dpu_pingpong pp);
 };
 
 /**
-- 
GitLab


From e99b2d0670a7f6eb81e049a5279aa7124705fbc9 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:52 +0200
Subject: [PATCH 0861/1791] dt-bindings: display/msm: dsi-controller-main: Add
 SM6350

Add the DSI host found on SM6350.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541281/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-1-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../devicetree/bindings/display/msm/dsi-controller-main.yaml    | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
index 51808b1d27268..ee5206351505c 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@@ -26,6 +26,7 @@ properties:
               - qcom,sdm660-dsi-ctrl
               - qcom,sdm845-dsi-ctrl
               - qcom,sm6115-dsi-ctrl
+              - qcom,sm6350-dsi-ctrl
               - qcom,sm8150-dsi-ctrl
               - qcom,sm8250-dsi-ctrl
               - qcom,sm8350-dsi-ctrl
@@ -297,6 +298,7 @@ allOf:
           contains:
             enum:
               - qcom,msm8998-dsi-ctrl
+              - qcom,sm6350-dsi-ctrl
     then:
       properties:
         clocks:
-- 
GitLab


From 27a869221bb7a9ae3de62b683d5c3cc98edaf3c8 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:53 +0200
Subject: [PATCH 0862/1791] dt-bindings: display/msm: dsi-controller-main: Add
 SM6375

Add the DSI host found on SM6375.

Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541285/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-2-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../devicetree/bindings/display/msm/dsi-controller-main.yaml    | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
index ee5206351505c..93a33a97bb050 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@@ -27,6 +27,7 @@ properties:
               - qcom,sdm845-dsi-ctrl
               - qcom,sm6115-dsi-ctrl
               - qcom,sm6350-dsi-ctrl
+              - qcom,sm6375-dsi-ctrl
               - qcom,sm8150-dsi-ctrl
               - qcom,sm8250-dsi-ctrl
               - qcom,sm8350-dsi-ctrl
@@ -366,6 +367,7 @@ allOf:
             enum:
               - qcom,sdm845-dsi-ctrl
               - qcom,sm6115-dsi-ctrl
+              - qcom,sm6375-dsi-ctrl
     then:
       properties:
         clocks:
-- 
GitLab


From ed41005f5b7c6886179c3f7546ebe50ef7d28d5e Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:54 +0200
Subject: [PATCH 0863/1791] dt-bindings: display/msm: sc7180-dpu: Describe
 SM6350 and SM6375

SC7180, SM6350 and SM6375 use a rather similar hw setup for DPU, with
the main exception being that the last one requires an additional
throttle clock.

It is not well understood yet, but failing to toggle it on makes the
display hardware stall and not output any frames.

Document SM6350 and SM6375 DPU.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541283/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-3-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../bindings/display/msm/qcom,sc7180-dpu.yaml | 23 ++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml
index 1fb8321d9ee80..630b11480496f 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml
@@ -13,7 +13,10 @@ $ref: /schemas/display/msm/dpu-common.yaml#
 
 properties:
   compatible:
-    const: qcom,sc7180-dpu
+    enum:
+      - qcom,sc7180-dpu
+      - qcom,sm6350-dpu
+      - qcom,sm6375-dpu
 
   reg:
     items:
@@ -26,6 +29,7 @@ properties:
       - const: vbif
 
   clocks:
+    minItems: 6
     items:
       - description: Display hf axi clock
       - description: Display ahb clock
@@ -33,8 +37,10 @@ properties:
       - description: Display lut clock
       - description: Display core clock
       - description: Display vsync clock
+      - description: Display core throttle clock
 
   clock-names:
+    minItems: 6
     items:
       - const: bus
       - const: iface
@@ -42,6 +48,7 @@ properties:
       - const: lut
       - const: core
       - const: vsync
+      - const: throttle
 
 required:
   - compatible
@@ -52,6 +59,20 @@ required:
 
 unevaluatedProperties: false
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          const: qcom,sm6375-dpu
+
+    then:
+      properties:
+        clocks:
+          minItems: 7
+
+        clock-names:
+          minItems: 7
+
 examples:
   - |
     #include <dt-bindings/clock/qcom,dispcc-sc7180.h>
-- 
GitLab


From 3b7502b0c205323b679596925d963f1e1c1804af Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:55 +0200
Subject: [PATCH 0864/1791] dt-bindings: display/msm: Add SM6350 MDSS

Document the SM6350 MDSS.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541286/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-4-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../display/msm/qcom,sm6350-mdss.yaml         | 213 ++++++++++++++++++
 1 file changed, 213 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml
new file mode 100644
index 0000000000000..ed0ad194d4ce9
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml
@@ -0,0 +1,213 @@
+# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sm6350-mdss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SM6350 Display MDSS
+
+maintainers:
+  - Krishna Manikandan <quic_mkrishn@quicinc.com>
+
+description:
+  SM6350 MSM Mobile Display Subsystem (MDSS), which encapsulates sub-blocks
+  like DPU display controller, DSI and DP interfaces etc.
+
+$ref: /schemas/display/msm/mdss-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sm6350-mdss
+
+  clocks:
+    items:
+      - description: Display AHB clock from gcc
+      - description: Display AXI clock from gcc
+      - description: Display core clock
+
+  clock-names:
+    items:
+      - const: iface
+      - const: bus
+      - const: core
+
+  iommus:
+    maxItems: 1
+
+  interconnects:
+    maxItems: 2
+
+  interconnect-names:
+    maxItems: 2
+
+patternProperties:
+  "^display-controller@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,sm6350-dpu
+
+  "^dsi@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        items:
+          - const: qcom,sm6350-dsi-ctrl
+          - const: qcom,mdss-dsi-ctrl
+
+  "^phy@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,dsi-phy-10nm
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,dispcc-sm6350.h>
+    #include <dt-bindings/clock/qcom,gcc-sm6350.h>
+    #include <dt-bindings/clock/qcom,rpmh.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-subsystem@ae00000 {
+        compatible = "qcom,sm6350-mdss";
+        reg = <0x0ae00000 0x1000>;
+        reg-names = "mdss";
+
+        power-domains = <&dispcc MDSS_GDSC>;
+
+        clocks = <&gcc GCC_DISP_AHB_CLK>,
+                 <&gcc GCC_DISP_AXI_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_CLK>;
+        clock-names = "iface", "bus", "core";
+
+        interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-controller;
+        #interrupt-cells = <1>;
+
+        iommus = <&apps_smmu 0x800 0x2>;
+        #address-cells = <1>;
+        #size-cells = <1>;
+        ranges;
+
+        display-controller@ae01000 {
+            compatible = "qcom,sm6350-dpu";
+            reg = <0x0ae01000 0x8f000>,
+                  <0x0aeb0000 0x2008>;
+            reg-names = "mdp", "vbif";
+
+            clocks = <&gcc GCC_DISP_AXI_CLK>,
+              <&dispcc DISP_CC_MDSS_AHB_CLK>,
+              <&dispcc DISP_CC_MDSS_ROT_CLK>,
+              <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>,
+              <&dispcc DISP_CC_MDSS_MDP_CLK>,
+              <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+            clock-names = "bus", "iface", "rot", "lut", "core",
+                    "vsync";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                              <&dispcc DISP_CC_MDSS_VSYNC_CLK>,
+                              <&dispcc DISP_CC_MDSS_ROT_CLK>,
+                              <&dispcc DISP_CC_MDSS_AHB_CLK>;
+            assigned-clock-rates = <300000000>,
+                                   <19200000>,
+                                   <19200000>,
+                                   <19200000>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <0>;
+            operating-points-v2 = <&mdp_opp_table>;
+            power-domains = <&rpmhpd SM6350_CX>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&dsi0_in>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&dsi1_in>;
+                    };
+                };
+            };
+        };
+
+        dsi@ae94000 {
+            compatible = "qcom,sm6350-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK0_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC0_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>;
+            assigned-clock-parents = <&dsi0_phy 0>, <&dsi0_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd SM6350_MX>;
+
+            phys = <&dsi0_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi0_out: endpoint {
+                    };
+                };
+            };
+        };
+
+        dsi0_phy: phy@ae94400 {
+            compatible = "qcom,dsi-phy-10nm";
+            reg = <0x0ae94400 0x200>,
+                  <0x0ae94600 0x280>,
+                  <0x0ae94a00 0x1e0>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>, <&rpmhcc RPMH_CXO_CLK>;
+            clock-names = "iface", "ref";
+        };
+    };
+...
-- 
GitLab


From 2a5c1021bc77857d39a15590682f69c33417f1ba Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:56 +0200
Subject: [PATCH 0865/1791] dt-bindings: display/msm: Add SM6375 MDSS

Document the SM6375 MDSS.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541289/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-5-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../display/msm/qcom,sm6375-mdss.yaml         | 215 ++++++++++++++++++
 1 file changed, 215 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml
new file mode 100644
index 0000000000000..76369a4f7c4d4
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml
@@ -0,0 +1,215 @@
+# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sm6375-mdss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SM6375 Display MDSS
+
+maintainers:
+  - Konrad Dybcio <konrad.dybcio@linaro.org>
+
+description:
+  SM6375 MSM Mobile Display Subsystem (MDSS), which encapsulates sub-blocks
+  like DPU display controller, DSI and DP interfaces etc.
+
+$ref: /schemas/display/msm/mdss-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sm6375-mdss
+
+  clocks:
+    items:
+      - description: Display AHB clock from gcc
+      - description: Display AHB clock
+      - description: Display core clock
+
+  clock-names:
+    items:
+      - const: iface
+      - const: ahb
+      - const: core
+
+  iommus:
+    maxItems: 1
+
+  interconnects:
+    maxItems: 2
+
+  interconnect-names:
+    maxItems: 2
+
+patternProperties:
+  "^display-controller@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,sm6375-dpu
+
+  "^dsi@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        items:
+          - const: qcom,sm6375-dsi-ctrl
+          - const: qcom,mdss-dsi-ctrl
+
+  "^phy@[0-9a-f]+$":
+    type: object
+    properties:
+      compatible:
+        const: qcom,sm6375-dsi-phy-7nm
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,rpmcc.h>
+    #include <dt-bindings/clock/qcom,sm6375-gcc.h>
+    #include <dt-bindings/clock/qcom,sm6375-dispcc.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-subsystem@5e00000 {
+        compatible = "qcom,sm6375-mdss";
+        reg = <0x05e00000 0x1000>;
+        reg-names = "mdss";
+
+        power-domains = <&dispcc MDSS_GDSC>;
+
+        clocks = <&gcc GCC_DISP_AHB_CLK>,
+                 <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_CLK>;
+        clock-names = "iface", "ahb", "core";
+
+        interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-controller;
+        #interrupt-cells = <1>;
+
+        iommus = <&apps_smmu 0x820 0x2>;
+        #address-cells = <1>;
+        #size-cells = <1>;
+        ranges;
+
+        display-controller@5e01000 {
+            compatible = "qcom,sm6375-dpu";
+            reg = <0x05e01000 0x8e030>,
+                  <0x05eb0000 0x2008>;
+            reg-names = "mdp", "vbif";
+
+            clocks = <&gcc GCC_DISP_HF_AXI_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&dispcc DISP_CC_MDSS_ROT_CLK>,
+                     <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>,
+                     <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                     <&dispcc DISP_CC_MDSS_VSYNC_CLK>,
+                     <&gcc GCC_DISP_THROTTLE_CORE_CLK>;
+            clock-names = "bus",
+                          "iface",
+                          "rot",
+                          "lut",
+                          "core",
+                          "vsync",
+                          "throttle";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+            assigned-clock-rates = <19200000>;
+
+            operating-points-v2 = <&mdp_opp_table>;
+            power-domains = <&rpmpd SM6375_VDDCX>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&dsi0_in>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&dsi1_in>;
+                    };
+                };
+            };
+        };
+
+        dsi@5e94000 {
+            compatible = "qcom,sm6375-dsi-ctrl", "qcom,mdss-dsi-ctrl";
+            reg = <0x05e94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK0_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC0_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>;
+            assigned-clock-parents = <&mdss_dsi0_phy 0>, <&mdss_dsi0_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmpd SM6375_VDDMX>;
+
+            phys = <&mdss_dsi0_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi0_out: endpoint {
+                    };
+                };
+            };
+        };
+
+        mdss_dsi0_phy: phy@5e94400 {
+            compatible = "qcom,sm6375-dsi-phy-7nm";
+            reg = <0x05e94400 0x200>,
+                  <0x05e94600 0x280>,
+                  <0x05e94900 0x264>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&rpmcc RPM_SMD_XO_CLK_SRC>;
+            clock-names = "iface", "ref";
+        };
+    };
+...
-- 
GitLab


From 3186acba5cdc10b5eb0f45a5ff7ce8c1c6e8c16c Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:57 +0200
Subject: [PATCH 0866/1791] drm/msm/dpu: Add SM6350 support

Add SM6350 support to the DPU1 driver to enable display output.

It's worth noting that one entry dpu_qos_lut_entry was trimmed off:

{.fl = 0, .lut = 0x0011223344556677 },

due to the lack of support for selecting between portrait and landscape
LUT settings (for danger and safe LUTs) and no full support for
qseed/non-qseed usescases (for QoS LUT).

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541287/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-6-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_6_4_sm6350.h    | 173 ++++++++++++++++++
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    |   5 +
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |   1 +
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c       |   1 +
 4 files changed, 180 insertions(+)
 create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h
new file mode 100644
index 0000000000000..06eba23b02364
--- /dev/null
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h
@@ -0,0 +1,173 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2023, Linaro Limited
+ */
+
+#ifndef _DPU_6_4_SM6350_H
+#define _DPU_6_4_SM6350_H
+
+static const struct dpu_caps sm6350_dpu_caps = {
+	.max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
+	.max_mixer_blendstages = 0x7,
+	.qseed_type = DPU_SSPP_SCALER_QSEED4,
+	.has_src_split = true,
+	.has_dim_layer = true,
+	.has_idle_pc = true,
+	.max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
+	.pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE,
+};
+
+static const struct dpu_ubwc_cfg sm6350_ubwc_cfg = {
+	.ubwc_version = DPU_HW_UBWC_VER_20,
+	.ubwc_swizzle = 6,
+	.highest_bank_bit = 1,
+};
+
+static const struct dpu_mdp_cfg sm6350_mdp[] = {
+	{
+	.name = "top_0", .id = MDP_TOP,
+	.base = 0x0, .len = 0x494,
+	.features = 0,
+	.clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 },
+	.clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 },
+	.clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 },
+	.clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 },
+	.clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 },
+	},
+};
+
+static const struct dpu_ctl_cfg sm6350_ctl[] = {
+	{
+	.name = "ctl_0", .id = CTL_0,
+	.base = 0x1000, .len = 0x1dc,
+	.features = BIT(DPU_CTL_ACTIVE_CFG),
+	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
+	},
+	{
+	.name = "ctl_1", .id = CTL_1,
+	.base = 0x1200, .len = 0x1dc,
+	.features = BIT(DPU_CTL_ACTIVE_CFG),
+	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10),
+	},
+	{
+	.name = "ctl_2", .id = CTL_2,
+	.base = 0x1400, .len = 0x1dc,
+	.features = BIT(DPU_CTL_ACTIVE_CFG),
+	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11),
+	},
+	{
+	.name = "ctl_3", .id = CTL_3,
+	.base = 0x1600, .len = 0x1dc,
+	.features = BIT(DPU_CTL_ACTIVE_CFG),
+	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12),
+	},
+};
+
+static const struct dpu_sspp_cfg sm6350_sspp[] = {
+	SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK,
+		sc7180_vig_sblk_0, 0,  SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0),
+	SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK,
+		sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0),
+	SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_CURSOR_SDM845_MASK,
+		sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1),
+	SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK,
+		sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2),
+};
+
+static const struct dpu_lm_cfg sm6350_lm[] = {
+	LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK,
+		&sc7180_lm_sblk, PINGPONG_0, LM_1, DSPP_0),
+	LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK,
+		&sc7180_lm_sblk, PINGPONG_1, LM_0, 0),
+};
+
+static const struct dpu_dspp_cfg sm6350_dspp[] = {
+	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
+		&sm8150_dspp_sblk),
+};
+
+static struct dpu_pingpong_cfg sm6350_pp[] = {
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
+		-1),
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
+		-1),
+};
+
+static const struct dpu_dsc_cfg sm6350_dsc[] = {
+	DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)),
+};
+
+static const struct dpu_intf_cfg sm6350_intf[] = {
+	INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 35, INTF_SC7180_MASK,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)),
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 35, INTF_SC7180_MASK,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+		DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
+};
+
+static const struct dpu_perf_cfg sm6350_perf_data = {
+	.max_bw_low = 4200000,
+	.max_bw_high = 5100000,
+	.min_core_ib = 2500000,
+	.min_llcc_ib = 0,
+	.min_dram_ib = 1600000,
+	.min_prefill_lines = 35,
+	/* TODO: confirm danger_lut_tbl */
+	.danger_lut_tbl = {0xffff, 0xffff, 0x0},
+	.safe_lut_tbl = {0xff00, 0xff00, 0xffff},
+	.qos_lut_tbl = {
+		{.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile),
+		.entries = sm6350_qos_linear_macrotile
+		},
+		{.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile),
+		.entries = sm6350_qos_linear_macrotile
+		},
+		{.nentry = ARRAY_SIZE(sc7180_qos_nrt),
+		.entries = sc7180_qos_nrt
+		},
+	},
+	.cdp_cfg = {
+		{.rd_enable = 1, .wr_enable = 1},
+		{.rd_enable = 1, .wr_enable = 0}
+	},
+	.clk_inefficiency_factor = 105,
+	.bw_inefficiency_factor = 120,
+};
+
+const struct dpu_mdss_cfg dpu_sm6350_cfg = {
+	.caps = &sm6350_dpu_caps,
+	.ubwc = &sm6350_ubwc_cfg,
+	.mdp_count = ARRAY_SIZE(sm6350_mdp),
+	.mdp = sm6350_mdp,
+	.ctl_count = ARRAY_SIZE(sm6350_ctl),
+	.ctl = sm6350_ctl,
+	.sspp_count = ARRAY_SIZE(sm6350_sspp),
+	.sspp = sm6350_sspp,
+	.mixer_count = ARRAY_SIZE(sm6350_lm),
+	.mixer = sm6350_lm,
+	.dspp_count = ARRAY_SIZE(sm6350_dspp),
+	.dspp = sm6350_dspp,
+	.dsc_count = ARRAY_SIZE(sm6350_dsc),
+	.dsc = sm6350_dsc,
+	.pingpong_count = ARRAY_SIZE(sm6350_pp),
+	.pingpong = sm6350_pp,
+	.intf_count = ARRAY_SIZE(sm6350_intf),
+	.intf = sm6350_intf,
+	.vbif_count = ARRAY_SIZE(sdm845_vbif),
+	.vbif = sdm845_vbif,
+	.perf = &sm6350_perf_data,
+	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
+		     BIT(MDP_SSPP_TOP0_INTR2) | \
+		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
+		     BIT(MDP_INTF0_INTR) | \
+		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR),
+};
+
+#endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index f07eab738008c..8bf6562f39e86 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -689,6 +689,10 @@ static const struct dpu_qos_lut_entry sc7180_qos_linear[] = {
 	{.fl = 0, .lut = 0x0011222222335777},
 };
 
+static const struct dpu_qos_lut_entry sm6350_qos_linear_macrotile[] = {
+	{.fl = 0, .lut = 0x0011223445566777 },
+};
+
 static const struct dpu_qos_lut_entry sm8150_qos_linear[] = {
 	{.fl = 0, .lut = 0x0011222222223357 },
 };
@@ -744,6 +748,7 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = {
 #include "catalog/dpu_6_0_sm8250.h"
 #include "catalog/dpu_6_2_sc7180.h"
 #include "catalog/dpu_6_3_sm6115.h"
+#include "catalog/dpu_6_4_sm6350.h"
 #include "catalog/dpu_6_5_qcm2290.h"
 
 #include "catalog/dpu_7_0_sm8350.h"
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index d59a9027c4b6e..e89edbe0ab987 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -832,6 +832,7 @@ extern const struct dpu_mdss_cfg dpu_sc8180x_cfg;
 extern const struct dpu_mdss_cfg dpu_sm8250_cfg;
 extern const struct dpu_mdss_cfg dpu_sc7180_cfg;
 extern const struct dpu_mdss_cfg dpu_sm6115_cfg;
+extern const struct dpu_mdss_cfg dpu_sm6350_cfg;
 extern const struct dpu_mdss_cfg dpu_qcm2290_cfg;
 extern const struct dpu_mdss_cfg dpu_sm8350_cfg;
 extern const struct dpu_mdss_cfg dpu_sc7280_cfg;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 801300e6626f5..6715f3762e74f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1305,6 +1305,7 @@ static const struct of_device_id dpu_dt_match[] = {
 	{ .compatible = "qcom,sc8180x-dpu", .data = &dpu_sc8180x_cfg, },
 	{ .compatible = "qcom,sc8280xp-dpu", .data = &dpu_sc8280xp_cfg, },
 	{ .compatible = "qcom,sm6115-dpu", .data = &dpu_sm6115_cfg, },
+	{ .compatible = "qcom,sm6350-dpu", .data = &dpu_sm6350_cfg, },
 	{ .compatible = "qcom,sm8150-dpu", .data = &dpu_sm8150_cfg, },
 	{ .compatible = "qcom,sm8250-dpu", .data = &dpu_sm8250_cfg, },
 	{ .compatible = "qcom,sm8350-dpu", .data = &dpu_sm8350_cfg, },
-- 
GitLab


From 27f0df03f3ffd3905bc320a8da4050cd645bd685 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:59 +0200
Subject: [PATCH 0867/1791] drm/msm/dpu: Add SM6375 support

Add basic SM6375 support to the DPU1 driver to enable display output.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541293/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-8-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_6_9_sm6375.h    | 139 ++++++++++++++++++
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    |   1 +
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |   1 +
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c       |   1 +
 4 files changed, 142 insertions(+)
 create mode 100644 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
new file mode 100644
index 0000000000000..d7aae45e3e663
--- /dev/null
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2023, Linaro Limited
+ */
+
+#ifndef _DPU_6_9_SM6375_H
+#define _DPU_6_9_SM6375_H
+
+static const struct dpu_caps sm6375_dpu_caps = {
+	.max_mixer_width = DEFAULT_DPU_LINE_WIDTH,
+	.max_mixer_blendstages = 0x4,
+	.qseed_type = DPU_SSPP_SCALER_QSEED4,
+	.has_dim_layer = true,
+	.has_idle_pc = true,
+	.max_linewidth = 2160,
+	.pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE,
+};
+
+static const struct dpu_ubwc_cfg sm6375_ubwc_cfg = {
+	.ubwc_version = DPU_HW_UBWC_VER_20,
+	.ubwc_swizzle = 6,
+	.highest_bank_bit = 1,
+};
+
+static const struct dpu_mdp_cfg sm6375_mdp[] = {
+	{
+	.name = "top_0", .id = MDP_TOP,
+	.base = 0x0, .len = 0x494,
+	.features = 0,
+	.clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 },
+	.clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 },
+	},
+};
+
+static const struct dpu_ctl_cfg sm6375_ctl[] = {
+	{
+	.name = "ctl_0", .id = CTL_0,
+	.base = 0x1000, .len = 0x1dc,
+	.features = BIT(DPU_CTL_ACTIVE_CFG),
+	.intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9),
+	},
+};
+
+static const struct dpu_sspp_cfg sm6375_sspp[] = {
+	SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK,
+		sm6115_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0),
+	SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK,
+		sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0),
+};
+
+static const struct dpu_lm_cfg sm6375_lm[] = {
+	LM_BLK("lm_0", LM_0, 0x44000, MIXER_QCM2290_MASK,
+		&qcm2290_lm_sblk, PINGPONG_0, 0, DSPP_0),
+};
+
+static const struct dpu_dspp_cfg sm6375_dspp[] = {
+	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
+		&sm8150_dspp_sblk),
+};
+
+static const struct dpu_pingpong_cfg sm6375_pp[] = {
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
+		-1),
+};
+
+static const struct dpu_dsc_cfg sm6375_dsc[] = {
+	DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)),
+};
+
+static const struct dpu_intf_cfg sm6375_intf[] = {
+	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
+	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
+		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
+		DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)),
+};
+
+static const struct dpu_perf_cfg sm6375_perf_data = {
+	.max_bw_low = 5200000,
+	.max_bw_high = 6200000,
+	.min_core_ib = 2500000,
+	.min_llcc_ib = 0, /* No LLCC on this SoC */
+	.min_dram_ib = 1600000,
+	.min_prefill_lines = 24,
+	/* TODO: confirm danger_lut_tbl */
+	.danger_lut_tbl = {0xffff, 0xffff, 0x0},
+	.safe_lut_tbl = {0xfe00, 0xfe00, 0xffff},
+	.qos_lut_tbl = {
+		{.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile),
+		.entries = sm6350_qos_linear_macrotile
+		},
+		{.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile),
+		.entries = sm6350_qos_linear_macrotile
+		},
+		{.nentry = ARRAY_SIZE(sc7180_qos_nrt),
+		.entries = sc7180_qos_nrt
+		},
+	},
+	.cdp_cfg = {
+		{.rd_enable = 1, .wr_enable = 1},
+		{.rd_enable = 1, .wr_enable = 0}
+	},
+	.clk_inefficiency_factor = 105,
+	.bw_inefficiency_factor = 120,
+};
+
+const struct dpu_mdss_cfg dpu_sm6375_cfg = {
+	.caps = &sm6375_dpu_caps,
+	.ubwc = &sm6375_ubwc_cfg,
+	.mdp_count = ARRAY_SIZE(sm6375_mdp),
+	.mdp = sm6375_mdp,
+	.ctl_count = ARRAY_SIZE(sm6375_ctl),
+	.ctl = sm6375_ctl,
+	.sspp_count = ARRAY_SIZE(sm6375_sspp),
+	.sspp = sm6375_sspp,
+	.mixer_count = ARRAY_SIZE(sm6375_lm),
+	.mixer = sm6375_lm,
+	.dspp_count = ARRAY_SIZE(sm6375_dspp),
+	.dspp = sm6375_dspp,
+	.dsc_count = ARRAY_SIZE(sm6375_dsc),
+	.dsc = sm6375_dsc,
+	.pingpong_count = ARRAY_SIZE(sm6375_pp),
+	.pingpong = sm6375_pp,
+	.intf_count = ARRAY_SIZE(sm6375_intf),
+	.intf = sm6375_intf,
+	.vbif_count = ARRAY_SIZE(sdm845_vbif),
+	.vbif = sdm845_vbif,
+	.perf = &sm6375_perf_data,
+	.mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \
+		     BIT(MDP_SSPP_TOP0_INTR2) | \
+		     BIT(MDP_SSPP_TOP0_HIST_INTR) | \
+		     BIT(MDP_INTF1_INTR) | \
+		     BIT(MDP_INTF1_TEAR_INTR),
+};
+
+#endif
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 8bf6562f39e86..b9f1d58a6cace 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -750,6 +750,7 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = {
 #include "catalog/dpu_6_3_sm6115.h"
 #include "catalog/dpu_6_4_sm6350.h"
 #include "catalog/dpu_6_5_qcm2290.h"
+#include "catalog/dpu_6_9_sm6375.h"
 
 #include "catalog/dpu_7_0_sm8350.h"
 #include "catalog/dpu_7_2_sc7280.h"
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index e89edbe0ab987..ac4a9e73705cd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -834,6 +834,7 @@ extern const struct dpu_mdss_cfg dpu_sc7180_cfg;
 extern const struct dpu_mdss_cfg dpu_sm6115_cfg;
 extern const struct dpu_mdss_cfg dpu_sm6350_cfg;
 extern const struct dpu_mdss_cfg dpu_qcm2290_cfg;
+extern const struct dpu_mdss_cfg dpu_sm6375_cfg;
 extern const struct dpu_mdss_cfg dpu_sm8350_cfg;
 extern const struct dpu_mdss_cfg dpu_sc7280_cfg;
 extern const struct dpu_mdss_cfg dpu_sc8280xp_cfg;
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index 6715f3762e74f..e97570287f312 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1306,6 +1306,7 @@ static const struct of_device_id dpu_dt_match[] = {
 	{ .compatible = "qcom,sc8280xp-dpu", .data = &dpu_sc8280xp_cfg, },
 	{ .compatible = "qcom,sm6115-dpu", .data = &dpu_sm6115_cfg, },
 	{ .compatible = "qcom,sm6350-dpu", .data = &dpu_sm6350_cfg, },
+	{ .compatible = "qcom,sm6375-dpu", .data = &dpu_sm6375_cfg, },
 	{ .compatible = "qcom,sm8150-dpu", .data = &dpu_sm8150_cfg, },
 	{ .compatible = "qcom,sm8250-dpu", .data = &dpu_sm8250_cfg, },
 	{ .compatible = "qcom,sm8350-dpu", .data = &dpu_sm8350_cfg, },
-- 
GitLab


From c2c1217e61bd80b1d21963ec7c2f8ebb9f77c2a0 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:43:58 +0200
Subject: [PATCH 0868/1791] drm/msm: mdss: Add SM6350 support

Add support for MDSS on SM6350.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541291/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-7-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/msm_mdss.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c
index e8c93731aaa18..4e3a5f0c303c7 100644
--- a/drivers/gpu/drm/msm/msm_mdss.c
+++ b/drivers/gpu/drm/msm/msm_mdss.c
@@ -538,6 +538,14 @@ static const struct msm_mdss_data sdm845_data = {
 	.highest_bank_bit = 2,
 };
 
+static const struct msm_mdss_data sm6350_data = {
+	.ubwc_version = UBWC_2_0,
+	.ubwc_dec_version = UBWC_2_0,
+	.ubwc_swizzle = 6,
+	.ubwc_static = 0x1e,
+	.highest_bank_bit = 1,
+};
+
 static const struct msm_mdss_data sm8150_data = {
 	.ubwc_version = UBWC_3_0,
 	.ubwc_dec_version = UBWC_3_0,
@@ -571,6 +579,7 @@ static const struct of_device_id mdss_dt_match[] = {
 	{ .compatible = "qcom,sc8180x-mdss", .data = &sc8180x_data },
 	{ .compatible = "qcom,sc8280xp-mdss", .data = &sc8280xp_data },
 	{ .compatible = "qcom,sm6115-mdss", .data = &sm6115_data },
+	{ .compatible = "qcom,sm6350-mdss", .data = &sm6350_data },
 	{ .compatible = "qcom,sm8150-mdss", .data = &sm8150_data },
 	{ .compatible = "qcom,sm8250-mdss", .data = &sm8250_data },
 	{ .compatible = "qcom,sm8350-mdss", .data = &sm8250_data },
-- 
GitLab


From 5ff3d3a0a09ecf86f7de6089905d8c9297503da2 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Tue, 6 Jun 2023 14:44:00 +0200
Subject: [PATCH 0869/1791] drm/msm: mdss: Add SM6375 support

Add support for MDSS on SM6375.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541295/
Link: https://lore.kernel.org/r/20230411-topic-straitlagoon_mdss-v6-9-dee6a882571b@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/msm_mdss.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c
index 4e3a5f0c303c7..05648c910c68e 100644
--- a/drivers/gpu/drm/msm/msm_mdss.c
+++ b/drivers/gpu/drm/msm/msm_mdss.c
@@ -580,6 +580,7 @@ static const struct of_device_id mdss_dt_match[] = {
 	{ .compatible = "qcom,sc8280xp-mdss", .data = &sc8280xp_data },
 	{ .compatible = "qcom,sm6115-mdss", .data = &sm6115_data },
 	{ .compatible = "qcom,sm6350-mdss", .data = &sm6350_data },
+	{ .compatible = "qcom,sm6375-mdss", .data = &sm6350_data },
 	{ .compatible = "qcom,sm8150-mdss", .data = &sm8150_data },
 	{ .compatible = "qcom,sm8250-mdss", .data = &sm8250_data },
 	{ .compatible = "qcom,sm8350-mdss", .data = &sm8250_data },
-- 
GitLab


From 5e0bff2b9690c857d3465d90015fb30fa4947ffe Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Wed, 7 Jun 2023 08:09:46 -0700
Subject: [PATCH 0870/1791] drm/i915/display: Include of display limits doesn't
 need 'display/'

Drop the unnecessary directory prefix.  This also makes
intel_display_device.h easier to use from the Xe driver.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230607150946.1996087-1-matthew.d.roper@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_device.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 4f931258d81db..706ff2aa1f55e 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -8,7 +8,7 @@
 
 #include <linux/types.h>
 
-#include "display/intel_display_limits.h"
+#include "intel_display_limits.h"
 
 struct drm_i915_private;
 
-- 
GitLab


From 203b2019b3acc4ec851ba0461f1b30a3fd4449f5 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Tue, 6 Jun 2023 13:11:12 -0700
Subject: [PATCH 0871/1791] drm/msm/dpu: add DSC blocks to the catalog of
 MSM8998

MSM8998 doesn't have DSC blocks declared in the catalog. Complete DSC
1.1 support for all platforms by adding the missing blocks to MSM8998.

Changes in v9:
-- add MSM8998 and SC8180x to commit title

Changes in v10:
-- fix grammar at commit text

Changes in v12:
-- fix "titil" with "title" at changes in v9

Changes in v14:
-- "dsc" tp "DSC" at commit title

Changes in v15:
-- fix merge conflicts at dpu_5_1_sc8180x.h

Changes in v16
-- fix cherry-pick error by deleting both redundant .dsc and .dsc_count
   assignment from dpu_5_1_sc8180x.h

Changes in v17
-- remove sc8180x from both commit title and text
-- remove Reviewed-by

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/541371/
Link: https://lore.kernel.org/r/1686082272-22191-1-git-send-email-quic_khsieh@quicinc.com
[DB: fixed commit message as agreed in the email discussion]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
index 3c732a0360c70..7d0d0e74c3b08 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h
@@ -126,6 +126,11 @@ static const struct dpu_pingpong_cfg msm8998_pp[] = {
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)),
 };
 
+static const struct dpu_dsc_cfg msm8998_dsc[] = {
+	DSC_BLK("dsc_0", DSC_0, 0x80000, 0),
+	DSC_BLK("dsc_1", DSC_1, 0x80400, 0),
+};
+
 static const struct dpu_dspp_cfg msm8998_dspp[] = {
 	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
 		 &msm8998_dspp_sblk),
@@ -199,6 +204,8 @@ const struct dpu_mdss_cfg dpu_msm8998_cfg = {
 	.dspp = msm8998_dspp,
 	.pingpong_count = ARRAY_SIZE(msm8998_pp),
 	.pingpong = msm8998_pp,
+	.dsc_count = ARRAY_SIZE(msm8998_dsc),
+	.dsc = msm8998_dsc,
 	.intf_count = ARRAY_SIZE(msm8998_intf),
 	.intf = msm8998_intf,
 	.vbif_count = ARRAY_SIZE(msm8998_vbif),
-- 
GitLab


From e7a2cf8e058e24218067694113938726db8696c9 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 6 Jun 2023 11:33:03 +0300
Subject: [PATCH 0872/1791] drm/msm/dpu: tidy up some error checking

The "vsync_hz" variable is unsigned int so it can't be less
than zero.  The dpu_kms_get_clk_rate() function used to return a u64
but I previously changed it to return an unsigned long and zero on
error so it matches clk_get_rate().

Change the "vsync_hz" type to unsigned long as well and change the
error checking to check for zero instead of negatives.  This change
does not affect runtime at all.  It's just a clean up.

Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/541225/
Link: https://lore.kernel.org/r/ZH7vP2Swu8CYpgUL@moroto
[DB: fixed debug message]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 107f0eff958cd..63ba0082b6ee4 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -323,7 +323,7 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	struct dpu_hw_tear_check tc_cfg = { 0 };
 	struct drm_display_mode *mode;
 	bool tc_enable = true;
-	u32 vsync_hz;
+	unsigned long vsync_hz;
 	struct dpu_kms *dpu_kms;
 
 	if (phys_enc->has_intf_te) {
@@ -358,9 +358,8 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	 * frequency divided by the no. of rows (lines) in the LCDpanel.
 	 */
 	vsync_hz = dpu_kms_get_clk_rate(dpu_kms, "vsync");
-	if (vsync_hz <= 0) {
-		DPU_DEBUG_CMDENC(cmd_enc, "invalid - vsync_hz %u\n",
-				 vsync_hz);
+	if (!vsync_hz) {
+		DPU_DEBUG_CMDENC(cmd_enc, "invalid - no vsync clock\n");
 		return;
 	}
 
@@ -380,7 +379,7 @@ static void dpu_encoder_phys_cmd_tearcheck_config(
 	tc_cfg.rd_ptr_irq = mode->vdisplay + 1;
 
 	DPU_DEBUG_CMDENC(cmd_enc,
-		"tc vsync_clk_speed_hz %u vtotal %u vrefresh %u\n",
+		"tc vsync_clk_speed_hz %lu vtotal %u vrefresh %u\n",
 		vsync_hz, mode->vtotal, drm_mode_vrefresh(mode));
 	DPU_DEBUG_CMDENC(cmd_enc,
 		"tc enable %u start_pos %u rd_ptr_irq %u\n",
-- 
GitLab


From 24335848e543dc95c9e2ffa0108d879ffefd0442 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 6 Jun 2023 11:22:07 +0300
Subject: [PATCH 0873/1791] drm/i915/gsc: Fix error code in
 intel_gsc_uc_heci_cmd_submit_nonpriv()

This should return negative -EAGAIN instead of positive EAGAIN.

Fixes: e5e1e6d28ebc ("drm/i915/pxp: Add MTL helpers to submit Heci-Cmd-Packet to GSC")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZH7sr+Vs4zOQoouU@moroto
---
 drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
index 0ad090304ca09..89ed5ee9cdedc 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
@@ -202,7 +202,7 @@ intel_gsc_uc_heci_cmd_submit_nonpriv(struct intel_gsc_uc *gsc,
 			if (++trials < 10)
 				goto retry;
 			else
-				err = EAGAIN;
+				err = -EAGAIN;
 		}
 	}
 	i915_gem_ww_ctx_fini(&ww);
-- 
GitLab


From f645d22b93d5599fe47ce587ee36931290ba5778 Mon Sep 17 00:00:00 2001
From: Zhi Wang <zhi.a.wang@intel.com>
Date: Wed, 31 May 2023 02:04:11 +0000
Subject: [PATCH 0874/1791] drm/i915/gvt: remove unused variable gma_bottom in
 command parser

Remove unused variable gma_bottom in scan_workload() and scan_wa_ctx().
commit be1da7070aea ("drm/i915/gvt: vGPU command scanner") introduces
gma_bottom in several functions to calculate the size of the command
buffer. However, some of them are set but actually unused.

When compiling the code with ccflags -Wunused-but-set-variable, gcc
throws warnings.

Remove unused variables to avoid the gcc warnings. Tested via compiling
the code with ccflags -Wunused-but-set-variable.

Fixes: be1da7070aea ("drm/i915/gvt: vGPU command scanner")
Suggested-by: Jani Nikula <jani.nikula@intel.com>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: intel-gvt-dev@lists.freedesktop.org
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230531020411.18987-1-zhi.a.wang@intel.com
---
 drivers/gpu/drm/i915/gvt/cmd_parser.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index 3c4ae1da0d416..05f9348b7a9d8 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -2833,7 +2833,7 @@ static int command_scan(struct parser_exec_state *s,
 
 static int scan_workload(struct intel_vgpu_workload *workload)
 {
-	unsigned long gma_head, gma_tail, gma_bottom;
+	unsigned long gma_head, gma_tail;
 	struct parser_exec_state s;
 	int ret = 0;
 
@@ -2843,7 +2843,6 @@ static int scan_workload(struct intel_vgpu_workload *workload)
 
 	gma_head = workload->rb_start + workload->rb_head;
 	gma_tail = workload->rb_start + workload->rb_tail;
-	gma_bottom = workload->rb_start +  _RING_CTL_BUF_SIZE(workload->rb_ctl);
 
 	s.buf_type = RING_BUFFER_INSTRUCTION;
 	s.buf_addr_type = GTT_BUFFER;
@@ -2874,7 +2873,7 @@ static int scan_workload(struct intel_vgpu_workload *workload)
 static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 {
 
-	unsigned long gma_head, gma_tail, gma_bottom, ring_size, ring_tail;
+	unsigned long gma_head, gma_tail, ring_size, ring_tail;
 	struct parser_exec_state s;
 	int ret = 0;
 	struct intel_vgpu_workload *workload = container_of(wa_ctx,
@@ -2891,7 +2890,6 @@ static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 			PAGE_SIZE);
 	gma_head = wa_ctx->indirect_ctx.guest_gma;
 	gma_tail = wa_ctx->indirect_ctx.guest_gma + ring_tail;
-	gma_bottom = wa_ctx->indirect_ctx.guest_gma + ring_size;
 
 	s.buf_type = RING_BUFFER_INSTRUCTION;
 	s.buf_addr_type = GTT_BUFFER;
-- 
GitLab


From 3fecd469da9aa0b24bd965f05a8bf44f9d8e2a2e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 26 May 2023 19:38:08 +0300
Subject: [PATCH 0875/1791] drm/i915: re-enable -Wunused-but-set-variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

W=1 enables -Wunused-but-set-variable. We disabled it locally in i915
Makefile as we were hitting a bunch of warnings. See commit 6a05d2900464
("drm/i915: Disable unused-but-set compiler warning").

With the issues fixed or annotated with __maybe_unused, re-enable the
warning not only in W=1 but also locally as part of i915 build.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/28ca3f95fe77ceb8aa35b87fca73f7afbc89859a.1685119007.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 2cd8de174bf6b..789dce9e26088 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -19,7 +19,7 @@ subdir-ccflags-y += -Wno-type-limits
 subdir-ccflags-y += -Wno-missing-field-initializers
 subdir-ccflags-y += -Wno-sign-compare
 subdir-ccflags-y += -Wno-shift-negative-value
-subdir-ccflags-y += $(call cc-disable-warning, unused-but-set-variable)
+subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable)
 subdir-ccflags-y += $(call cc-disable-warning, frame-address)
 subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror
 
-- 
GitLab


From ad8bfd3619bb096dd1929a69989120b45e3c3eee Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Mon, 17 Apr 2023 15:14:30 +0300
Subject: [PATCH 0876/1791] accel/habanalabs: minimize encapsulation signal
 mutex lock time

Sync Stream Encapsulated Signal Handlers can be managed from different
contexts, and as such they are protected via a spin_lock.
However, spin_lock was unnecessarily protecting a larger code section
than really needed, covering a sleepable code section as well.
Since spin_lock disables preemption, it could lead to sleeping in
atomic context.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/command_submission.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/habanalabs/common/command_submission.c b/drivers/accel/habanalabs/common/command_submission.c
index af9d2e22c6e7e..977cd2710686d 100644
--- a/drivers/accel/habanalabs/common/command_submission.c
+++ b/drivers/accel/habanalabs/common/command_submission.c
@@ -2152,7 +2152,7 @@ static int cs_ioctl_unreserve_signals(struct hl_fpriv *hpriv, u32 handle_id)
 
 			hdev->asic_funcs->hw_queues_unlock(hdev);
 			rc = -EINVAL;
-			goto out;
+			goto out_unlock;
 		}
 
 		/*
@@ -2167,15 +2167,21 @@ static int cs_ioctl_unreserve_signals(struct hl_fpriv *hpriv, u32 handle_id)
 
 		/* Release the id and free allocated memory of the handle */
 		idr_remove(&mgr->handles, handle_id);
+
+		/* unlock before calling ctx_put, where we might sleep */
+		spin_unlock(&mgr->lock);
 		hl_ctx_put(encaps_sig_hdl->ctx);
 		kfree(encaps_sig_hdl);
+		goto out;
 	} else {
 		rc = -EINVAL;
 		dev_err(hdev->dev, "failed to unreserve signals, cannot find handler\n");
 	}
-out:
+
+out_unlock:
 	spin_unlock(&mgr->lock);
 
+out:
 	return rc;
 }
 
-- 
GitLab


From 9a4e44a4ee49dff6ded6a4c76bb921a8379cb7b4 Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Sun, 4 Dec 2022 18:37:53 +0200
Subject: [PATCH 0877/1791] accel/habanalabs: refactor abort of completions and
 waits

Aborting CS completions should be in command_submission.c but aborting
waiting for user interrupts should be in device.c.

This separation is also for adding more abort operations in the future.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../habanalabs/common/command_submission.c      |  3 +--
 drivers/accel/habanalabs/common/device.c        | 17 ++++++++++++-----
 drivers/accel/habanalabs/common/habanalabs.h    |  2 +-
 3 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/accel/habanalabs/common/command_submission.c b/drivers/accel/habanalabs/common/command_submission.c
index 977cd2710686d..4ff3256fdf8b2 100644
--- a/drivers/accel/habanalabs/common/command_submission.c
+++ b/drivers/accel/habanalabs/common/command_submission.c
@@ -1139,11 +1139,10 @@ static void force_complete_cs(struct hl_device *hdev)
 	spin_unlock(&hdev->cs_mirror_lock);
 }
 
-void hl_abort_waitings_for_completion(struct hl_device *hdev)
+void hl_abort_waiting_for_cs_completions(struct hl_device *hdev)
 {
 	force_complete_cs(hdev);
 	force_complete_multi_cs(hdev);
-	hl_release_pending_user_interrupts(hdev);
 }
 
 static void job_wq_completion(struct work_struct *work)
diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index fabfc501ef543..c8f4a34d2831d 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -1157,6 +1157,16 @@ static void take_release_locks(struct hl_device *hdev)
 	mutex_unlock(&hdev->fpriv_ctrl_list_lock);
 }
 
+static void hl_abort_waiting_for_completions(struct hl_device *hdev)
+{
+	hl_abort_waiting_for_cs_completions(hdev);
+
+	/* Release all pending user interrupts, each pending user interrupt
+	 * holds a reference to a user context.
+	 */
+	hl_release_pending_user_interrupts(hdev);
+}
+
 static void cleanup_resources(struct hl_device *hdev, bool hard_reset, bool fw_reset,
 				bool skip_wq_flush)
 {
@@ -1176,10 +1186,7 @@ static void cleanup_resources(struct hl_device *hdev, bool hard_reset, bool fw_r
 	/* flush the MMU prefetch workqueue */
 	flush_workqueue(hdev->prefetch_wq);
 
-	/* Release all pending user interrupts, each pending user interrupt
-	 * holds a reference to user context
-	 */
-	hl_release_pending_user_interrupts(hdev);
+	hl_abort_waiting_for_completions(hdev);
 }
 
 /*
@@ -1921,7 +1928,7 @@ int hl_device_cond_reset(struct hl_device *hdev, u32 flags, u64 event_mask)
 
 	hl_ctx_put(ctx);
 
-	hl_abort_waitings_for_completion(hdev);
+	hl_abort_waiting_for_completions(hdev);
 
 	return 0;
 
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 6b102947cb90a..f83ea96c65300 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3934,7 +3934,7 @@ void hl_dec_fini(struct hl_device *hdev);
 void hl_dec_ctx_fini(struct hl_ctx *ctx);
 
 void hl_release_pending_user_interrupts(struct hl_device *hdev);
-void hl_abort_waitings_for_completion(struct hl_device *hdev);
+void hl_abort_waiting_for_cs_completions(struct hl_device *hdev);
 int hl_cs_signal_sob_wraparound_handler(struct hl_device *hdev, u32 q_idx,
 			struct hl_hw_sob **hw_sob, u32 count, bool encaps_sig);
 
-- 
GitLab


From 3d21ec6424e6d38f284c37d77e7ec524c1a454f2 Mon Sep 17 00:00:00 2001
From: Dafna Hirschfeld <dhirschfeld@habana.ai>
Date: Mon, 17 Apr 2023 21:24:19 +0300
Subject: [PATCH 0878/1791] accel/habanalabs: add missing tpc interrupt info

For some reason the last possible tpc interrupt cause in
gaudi2_tpc_interrupts_cause is missing from the code.

Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 5c80e7af5b785..058741698d713 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -63,7 +63,7 @@
 #define GAUDI2_NUM_OF_CPU_SEI_ERR_CAUSE		3
 #define GAUDI2_NUM_OF_QM_SEI_ERR_CAUSE		2
 #define GAUDI2_NUM_OF_ROT_ERR_CAUSE		22
-#define GAUDI2_NUM_OF_TPC_INTR_CAUSE		30
+#define GAUDI2_NUM_OF_TPC_INTR_CAUSE		31
 #define GAUDI2_NUM_OF_DEC_ERR_CAUSE		25
 #define GAUDI2_NUM_OF_MME_ERR_CAUSE		16
 #define GAUDI2_NUM_OF_MME_SBTE_ERR_CAUSE	5
@@ -891,6 +891,7 @@ static const char * const gaudi2_tpc_interrupts_cause[GAUDI2_NUM_OF_TPC_INTR_CAU
 	"invalid_lock_access",
 	"LD_L protection violation",
 	"ST_L protection violation",
+	"D$ L0CS mismatch",
 };
 
 static const char * const guadi2_mme_error_cause[GAUDI2_NUM_OF_MME_ERR_CAUSE] = {
-- 
GitLab


From d8b9cea584661b30305cf341bf9f675dc0a25471 Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Tue, 18 Apr 2023 14:48:22 +0300
Subject: [PATCH 0879/1791] accel/habanalabs: add pci health check during
 heartbeat

Currently upon a heartbeat failure, we don't know if the failure
is due to firmware hang or due to a bad PCI link. Hence, we
are reading a PCI config space register with a known value (vendor ID)
so we will know which of the two possibilities caused the heartbeat
failure.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/device.c         | 15 ++++++++++++++-
 drivers/accel/habanalabs/common/habanalabs.h     |  2 ++
 drivers/accel/habanalabs/common/habanalabs_drv.c |  2 --
 3 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index c8f4a34d2831d..98b46b2e1898e 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -981,6 +981,18 @@ static void device_early_fini(struct hl_device *hdev)
 		hdev->asic_funcs->early_fini(hdev);
 }
 
+static bool is_pci_link_healthy(struct hl_device *hdev)
+{
+	u16 vendor_id;
+
+	if (!hdev->pdev)
+		return false;
+
+	pci_read_config_word(hdev->pdev, PCI_VENDOR_ID, &vendor_id);
+
+	return (vendor_id == PCI_VENDOR_ID_HABANALABS);
+}
+
 static void hl_device_heartbeat(struct work_struct *work)
 {
 	struct hl_device *hdev = container_of(work, struct hl_device,
@@ -995,7 +1007,8 @@ static void hl_device_heartbeat(struct work_struct *work)
 		goto reschedule;
 
 	if (hl_device_operational(hdev, NULL))
-		dev_err(hdev->dev, "Device heartbeat failed!\n");
+		dev_err(hdev->dev, "Device heartbeat failed! PCI link is %s\n",
+			is_pci_link_healthy(hdev) ? "healthy" : "broken");
 
 	info.err_type = HL_INFO_FW_HEARTBEAT_ERR;
 	info.event_mask = &event_mask;
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index f83ea96c65300..ad412cc01aba1 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -36,6 +36,8 @@
 struct hl_device;
 struct hl_fpriv;
 
+#define PCI_VENDOR_ID_HABANALABS	0x1da3
+
 /* Use upper bits of mmap offset to store habana driver specific information.
  * bits[63:59] - Encode mmap type
  * bits[45:0]  - mmap offset value
diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c
index d9df64e75f33a..1ec97da3dddb8 100644
--- a/drivers/accel/habanalabs/common/habanalabs_drv.c
+++ b/drivers/accel/habanalabs/common/habanalabs_drv.c
@@ -54,8 +54,6 @@ module_param(boot_error_status_mask, ulong, 0444);
 MODULE_PARM_DESC(boot_error_status_mask,
 	"Mask of the error status during device CPU boot (If bitX is cleared then error X is masked. Default all 1's)");
 
-#define PCI_VENDOR_ID_HABANALABS	0x1da3
-
 #define PCI_IDS_GOYA			0x0001
 #define PCI_IDS_GAUDI			0x1000
 #define PCI_IDS_GAUDI_SEC		0x1010
-- 
GitLab


From 3b9abb4fa642ee28bcd22a0398087a2d87d3c987 Mon Sep 17 00:00:00 2001
From: Tomer Tayar <ttayar@habana.ai>
Date: Sun, 19 Mar 2023 21:46:44 +0200
Subject: [PATCH 0880/1791] accel/habanalabs: expose debugfs files later

Currently the debugfs root folder and files for a device are created at
an early step, before the device initialization and before the char
device and sysfs files are exposed to user.
As there is no real reason not to do it together with the device
creation, postpone it to be done right afterwards.

The initialization of the debugfs entry structure is left in its
current position because it is used before creating the files.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/debugfs.c    | 37 +++++++-----
 drivers/accel/habanalabs/common/device.c     | 62 +++++++++++---------
 drivers/accel/habanalabs/common/habanalabs.h |  6 +-
 3 files changed, 61 insertions(+), 44 deletions(-)

diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c
index 22dd17c077c00..29b78c7cf5de8 100644
--- a/drivers/accel/habanalabs/common/debugfs.c
+++ b/drivers/accel/habanalabs/common/debugfs.c
@@ -1756,17 +1756,15 @@ static void add_files_to_device(struct hl_device *hdev, struct hl_dbg_device_ent
 	}
 }
 
-void hl_debugfs_add_device(struct hl_device *hdev)
+int hl_debugfs_device_init(struct hl_device *hdev)
 {
 	struct hl_dbg_device_entry *dev_entry = &hdev->hl_debugfs;
 	int count = ARRAY_SIZE(hl_debugfs_list);
 
 	dev_entry->hdev = hdev;
-	dev_entry->entry_arr = kmalloc_array(count,
-					sizeof(struct hl_debugfs_entry),
-					GFP_KERNEL);
+	dev_entry->entry_arr = kmalloc_array(count, sizeof(struct hl_debugfs_entry), GFP_KERNEL);
 	if (!dev_entry->entry_arr)
-		return;
+		return -ENOMEM;
 
 	dev_entry->data_dma_blob_desc.size = 0;
 	dev_entry->data_dma_blob_desc.data = NULL;
@@ -1787,21 +1785,14 @@ void hl_debugfs_add_device(struct hl_device *hdev)
 	spin_lock_init(&dev_entry->userptr_spinlock);
 	mutex_init(&dev_entry->ctx_mem_hash_mutex);
 
-	dev_entry->root = debugfs_create_dir(dev_name(hdev->dev),
-						hl_debug_root);
-
-	add_files_to_device(hdev, dev_entry, dev_entry->root);
-	if (!hdev->asic_prop.fw_security_enabled)
-		add_secured_nodes(dev_entry, dev_entry->root);
+	return 0;
 }
 
-void hl_debugfs_remove_device(struct hl_device *hdev)
+void hl_debugfs_device_fini(struct hl_device *hdev)
 {
 	struct hl_dbg_device_entry *entry = &hdev->hl_debugfs;
 	int i;
 
-	debugfs_remove_recursive(entry->root);
-
 	mutex_destroy(&entry->ctx_mem_hash_mutex);
 	mutex_destroy(&entry->file_mutex);
 
@@ -1814,6 +1805,24 @@ void hl_debugfs_remove_device(struct hl_device *hdev)
 	kfree(entry->entry_arr);
 }
 
+void hl_debugfs_add_device(struct hl_device *hdev)
+{
+	struct hl_dbg_device_entry *dev_entry = &hdev->hl_debugfs;
+
+	dev_entry->root = debugfs_create_dir(dev_name(hdev->dev), hl_debug_root);
+
+	add_files_to_device(hdev, dev_entry, dev_entry->root);
+	if (!hdev->asic_prop.fw_security_enabled)
+		add_secured_nodes(dev_entry, dev_entry->root);
+}
+
+void hl_debugfs_remove_device(struct hl_device *hdev)
+{
+	struct hl_dbg_device_entry *entry = &hdev->hl_debugfs;
+
+	debugfs_remove_recursive(entry->root);
+}
+
 void hl_debugfs_add_file(struct hl_fpriv *hpriv)
 {
 	struct hl_dbg_device_entry *dev_entry = &hpriv->hdev->hl_debugfs;
diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index 98b46b2e1898e..cab5a63db8c16 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -674,7 +674,7 @@ static int device_init_cdev(struct hl_device *hdev, struct class *class,
 	return 0;
 }
 
-static int device_cdev_sysfs_add(struct hl_device *hdev)
+static int cdev_sysfs_debugfs_add(struct hl_device *hdev)
 {
 	int rc;
 
@@ -699,7 +699,9 @@ static int device_cdev_sysfs_add(struct hl_device *hdev)
 		goto delete_ctrl_cdev_device;
 	}
 
-	hdev->cdev_sysfs_created = true;
+	hl_debugfs_add_device(hdev);
+
+	hdev->cdev_sysfs_debugfs_created = true;
 
 	return 0;
 
@@ -710,11 +712,12 @@ static int device_cdev_sysfs_add(struct hl_device *hdev)
 	return rc;
 }
 
-static void device_cdev_sysfs_del(struct hl_device *hdev)
+static void cdev_sysfs_debugfs_remove(struct hl_device *hdev)
 {
-	if (!hdev->cdev_sysfs_created)
+	if (!hdev->cdev_sysfs_debugfs_created)
 		goto put_devices;
 
+	hl_debugfs_remove_device(hdev);
 	hl_sysfs_fini(hdev);
 	cdev_device_del(&hdev->cdev_ctrl, hdev->dev_ctrl);
 	cdev_device_del(&hdev->cdev, hdev->dev);
@@ -2054,7 +2057,7 @@ static int create_cdev(struct hl_device *hdev)
 int hl_device_init(struct hl_device *hdev)
 {
 	int i, rc, cq_cnt, user_interrupt_cnt, cq_ready_cnt;
-	bool add_cdev_sysfs_on_err = false;
+	bool expose_interfaces_on_err = false;
 
 	rc = create_cdev(hdev);
 	if (rc)
@@ -2170,16 +2173,22 @@ int hl_device_init(struct hl_device *hdev)
 	hdev->device_release_watchdog_timeout_sec = HL_DEVICE_RELEASE_WATCHDOG_TIMEOUT_SEC;
 
 	hdev->memory_scrub_val = MEM_SCRUB_DEFAULT_VAL;
-	hl_debugfs_add_device(hdev);
 
-	/* debugfs nodes are created in hl_ctx_init so it must be called after
-	 * hl_debugfs_add_device.
+	rc = hl_debugfs_device_init(hdev);
+	if (rc) {
+		dev_err(hdev->dev, "failed to initialize debugfs entry structure\n");
+		kfree(hdev->kernel_ctx);
+		goto mmu_fini;
+	}
+
+	/* The debugfs entry structure is accessed in hl_ctx_init(), so it must be called after
+	 * hl_debugfs_device_init().
 	 */
 	rc = hl_ctx_init(hdev, hdev->kernel_ctx, true);
 	if (rc) {
 		dev_err(hdev->dev, "failed to initialize kernel context\n");
 		kfree(hdev->kernel_ctx);
-		goto remove_device_from_debugfs;
+		goto debugfs_device_fini;
 	}
 
 	rc = hl_cb_pool_init(hdev);
@@ -2195,11 +2204,10 @@ int hl_device_init(struct hl_device *hdev)
 	}
 
 	/*
-	 * From this point, override rc (=0) in case of an error to allow
-	 * debugging (by adding char devices and create sysfs nodes as part of
-	 * the error flow).
+	 * From this point, override rc (=0) in case of an error to allow debugging
+	 * (by adding char devices and creating sysfs/debugfs files as part of the error flow).
 	 */
-	add_cdev_sysfs_on_err = true;
+	expose_interfaces_on_err = true;
 
 	/* Device is now enabled as part of the initialization requires
 	 * communication with the device firmware to get information that
@@ -2241,15 +2249,13 @@ int hl_device_init(struct hl_device *hdev)
 	}
 
 	/*
-	 * Expose devices and sysfs nodes to user.
-	 * From here there is no need to add char devices and create sysfs nodes
-	 * in case of an error.
+	 * Expose devices and sysfs/debugfs files to user.
+	 * From here there is no need to expose them in case of an error.
 	 */
-	add_cdev_sysfs_on_err = false;
-	rc = device_cdev_sysfs_add(hdev);
+	expose_interfaces_on_err = false;
+	rc = cdev_sysfs_debugfs_add(hdev);
 	if (rc) {
-		dev_err(hdev->dev,
-			"Failed to add char devices and sysfs nodes\n");
+		dev_err(hdev->dev, "Failed to add char devices and sysfs/debugfs files\n");
 		rc = 0;
 		goto out_disabled;
 	}
@@ -2295,8 +2301,8 @@ int hl_device_init(struct hl_device *hdev)
 	if (hl_ctx_put(hdev->kernel_ctx) != 1)
 		dev_err(hdev->dev,
 			"kernel ctx is still alive on initialization failure\n");
-remove_device_from_debugfs:
-	hl_debugfs_remove_device(hdev);
+debugfs_device_fini:
+	hl_debugfs_device_fini(hdev);
 mmu_fini:
 	hl_mmu_fini(hdev);
 eq_fini:
@@ -2320,8 +2326,8 @@ int hl_device_init(struct hl_device *hdev)
 	put_device(hdev->dev);
 out_disabled:
 	hdev->disabled = true;
-	if (add_cdev_sysfs_on_err)
-		device_cdev_sysfs_add(hdev);
+	if (expose_interfaces_on_err)
+		cdev_sysfs_debugfs_add(hdev);
 	if (hdev->pdev)
 		dev_err(&hdev->pdev->dev,
 			"Failed to initialize hl%d. Device %s is NOT usable !\n",
@@ -2447,8 +2453,6 @@ void hl_device_fini(struct hl_device *hdev)
 	if ((hdev->kernel_ctx) && (hl_ctx_put(hdev->kernel_ctx) != 1))
 		dev_err(hdev->dev, "kernel ctx is still alive\n");
 
-	hl_debugfs_remove_device(hdev);
-
 	hl_dec_fini(hdev);
 
 	hl_vm_fini(hdev);
@@ -2473,8 +2477,10 @@ void hl_device_fini(struct hl_device *hdev)
 
 	device_early_fini(hdev);
 
-	/* Hide devices and sysfs nodes from user */
-	device_cdev_sysfs_del(hdev);
+	/* Hide devices and sysfs/debugfs files from user */
+	cdev_sysfs_debugfs_remove(hdev);
+
+	hl_debugfs_device_fini(hdev);
 
 	pr_info("removed device successfully\n");
 }
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index ad412cc01aba1..ea0914d08bdc3 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3292,7 +3292,7 @@ struct hl_reset_info {
  * @in_debug: whether the device is in a state where the profiling/tracing infrastructure
  *            can be used. This indication is needed because in some ASICs we need to do
  *            specific operations to enable that infrastructure.
- * @cdev_sysfs_created: were char devices and sysfs nodes created.
+ * @cdev_sysfs_debugfs_created: were char devices and sysfs/debugfs files created.
  * @stop_on_err: true if engines should stop on error.
  * @supports_sync_stream: is sync stream supported.
  * @sync_stream_queue_idx: helper index for sync stream queues initialization.
@@ -3459,7 +3459,7 @@ struct hl_device {
 	u8				init_done;
 	u8				device_cpu_disabled;
 	u8				in_debug;
-	u8				cdev_sysfs_created;
+	u8				cdev_sysfs_debugfs_created;
 	u8				stop_on_err;
 	u8				supports_sync_stream;
 	u8				sync_stream_queue_idx;
@@ -3978,6 +3978,8 @@ void hl_handle_fw_err(struct hl_device *hdev, struct hl_info_fw_err_info *info);
 
 void hl_debugfs_init(void);
 void hl_debugfs_fini(void);
+int hl_debugfs_device_init(struct hl_device *hdev);
+void hl_debugfs_device_fini(struct hl_device *hdev);
 void hl_debugfs_add_device(struct hl_device *hdev);
 void hl_debugfs_remove_device(struct hl_device *hdev);
 void hl_debugfs_add_file(struct hl_fpriv *hpriv);
-- 
GitLab


From 67d19a2f49b07b2c4a15e6a96e1de46d95209ae4 Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Mon, 27 Mar 2023 11:56:16 +0300
Subject: [PATCH 0881/1791] accel/habanalabs: poll for device status update
 following WFE cmd

Currently, we rely on COMMS protocol's ack to verify that WFE command
has been acknowledged by the FW. However, this does not guarantee that
the device status has been updated.
Although unlikely, this could trigger a race since the driver expects
the device to be halted at that stage, but it might not be.
Therefore, we increase WFE's robustness by polling on the status
register that will be updated once the device is actually halted.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c | 28 +++++++++++++++----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index e48f024d86498..eb51d7f70aece 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -1368,8 +1368,10 @@ void hl_fw_ask_hard_reset_without_linux(struct hl_device *hdev)
 
 void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev)
 {
-	struct static_fw_load_mgr *static_loader =
-			&hdev->fw_loader.static_loader;
+	struct fw_load_mgr *fw_loader = &hdev->fw_loader;
+	u32 status, cpu_boot_status_reg, cpu_timeout;
+	struct static_fw_load_mgr *static_loader;
+	struct pre_fw_load_props *pre_fw_load;
 	int rc;
 
 	if (hdev->device_cpu_is_halted)
@@ -1377,12 +1379,28 @@ void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev)
 
 	/* Stop device CPU to make sure nothing bad happens */
 	if (hdev->asic_prop.dynamic_fw_load) {
+		pre_fw_load = &fw_loader->pre_fw_load;
+		cpu_timeout = fw_loader->cpu_timeout;
+		cpu_boot_status_reg = pre_fw_load->cpu_boot_status_reg;
+
 		rc = hl_fw_dynamic_send_protocol_cmd(hdev, &hdev->fw_loader,
-				COMMS_GOTO_WFE, 0, false,
-				hdev->fw_loader.cpu_timeout);
-		if (rc)
+				COMMS_GOTO_WFE, 0, false, cpu_timeout);
+		if (rc) {
 			dev_err(hdev->dev, "Failed sending COMMS_GOTO_WFE\n");
+		} else {
+			rc = hl_poll_timeout(
+				hdev,
+				cpu_boot_status_reg,
+				status,
+				status == CPU_BOOT_STATUS_IN_WFE,
+				hdev->fw_poll_interval_usec,
+				cpu_timeout);
+			if (rc)
+				dev_err(hdev->dev, "Current status=%u. Timed-out updating to WFE\n",
+						status);
+		}
 	} else {
+		static_loader = &hdev->fw_loader.static_loader;
 		WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_GOTO_WFE);
 		msleep(static_loader->cpu_reset_wait_msec);
 
-- 
GitLab


From 7d212963366e8113d1a8548a35eb9cd5df110dca Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Thu, 20 Apr 2023 14:17:31 +0300
Subject: [PATCH 0882/1791] accel/habanalabs: fix a static warning - 'dubious:
 x & !y'

Use a straight forward approach to get a conditional result.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 058741698d713..240fecfab6087 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -4527,7 +4527,7 @@ static int gaudi2_set_tpc_engine_mode(struct hl_device *hdev, u32 engine_id, u32
 	reg_base = gaudi2_tpc_cfg_blocks_bases[tpc_id];
 	reg_addr = reg_base + TPC_CFG_STALL_OFFSET;
 	reg_val = FIELD_PREP(DCORE0_TPC0_CFG_TPC_STALL_V_MASK,
-			!!(engine_command == HL_ENGINE_STALL));
+			(engine_command == HL_ENGINE_STALL) ? 1 : 0);
 	WREG32(reg_addr, reg_val);
 
 	if (engine_command == HL_ENGINE_RESUME) {
@@ -4551,7 +4551,7 @@ static int gaudi2_set_mme_engine_mode(struct hl_device *hdev, u32 engine_id, u32
 	reg_base = gaudi2_mme_ctrl_lo_blocks_bases[mme_id];
 	reg_addr = reg_base + MME_CTRL_LO_QM_STALL_OFFSET;
 	reg_val = FIELD_PREP(DCORE0_MME_CTRL_LO_QM_STALL_V_MASK,
-			!!(engine_command == HL_ENGINE_STALL));
+			(engine_command == HL_ENGINE_STALL) ? 1 : 0);
 	WREG32(reg_addr, reg_val);
 
 	return 0;
@@ -4572,7 +4572,7 @@ static int gaudi2_set_edma_engine_mode(struct hl_device *hdev, u32 engine_id, u3
 	reg_base = gaudi2_dma_core_blocks_bases[edma_id];
 	reg_addr = reg_base + EDMA_CORE_CFG_STALL_OFFSET;
 	reg_val = FIELD_PREP(DCORE0_EDMA0_CORE_CFG_1_HALT_MASK,
-			!!(engine_command == HL_ENGINE_STALL));
+			(engine_command == HL_ENGINE_STALL) ? 1 : 0);
 	WREG32(reg_addr, reg_val);
 
 	if (engine_command == HL_ENGINE_STALL) {
-- 
GitLab


From d0dcd4bbfa47f8726df5a45236c0edd0107ec5c3 Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Mon, 24 Apr 2023 14:17:03 +0300
Subject: [PATCH 0883/1791] accel/habanalabs: always fetch pci addr_dec error
 info

Due to missing indication of address decode source (LBW/HBW bus),
we should always try and fetch extended information.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 240fecfab6087..d21ef9997d05c 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8892,14 +8892,12 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_typ
 			"err cause: %s", gaudi2_pcie_addr_dec_error_cause[i]);
 		error_count++;
 
-		switch (intr_cause_data & BIT_ULL(i)) {
-		case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK:
-			hl_check_for_glbl_errors(hdev);
-			break;
-		case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK:
-			gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask);
-			break;
-		}
+		/*
+		 * Always check for LBW and HBW additional info as the indication itself is
+		 * sometimes missing
+		 */
+		hl_check_for_glbl_errors(hdev);
+		gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask);
 	}
 
 	return error_count;
-- 
GitLab


From cc1eeaa335f2d71e8bddf268e3e60d3a3c3a08bc Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Fri, 12 May 2023 14:46:55 +0800
Subject: [PATCH 0884/1791] accel/habanalabs: Fix some kernel-doc comments

Make the description of @regs_range_array and @regs_range_array_size
to @user_regs_range_array and @user_regs_range_array_size  to silence
the warnings:

drivers/accel/habanalabs/common/security.c:506: warning: Function parameter or member 'user_regs_range_array' not described in 'hl_init_pb_ranges_single_dcore'
drivers/accel/habanalabs/common/security.c:506: warning: Function parameter or member 'user_regs_range_array_size' not described in 'hl_init_pb_ranges_single_dcore'
drivers/accel/habanalabs/common/security.c:506: warning: Excess function parameter 'regs_range_array' description in 'hl_init_pb_ranges_single_dcore'
drivers/accel/habanalabs/common/security.c:506: warning: Excess function parameter 'regs_range_array_size' description in 'hl_init_pb_ranges_single_dcore'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=4940
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/security.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/habanalabs/common/security.c b/drivers/accel/habanalabs/common/security.c
index 297e6e44fd0c4..dc23ff57c91af 100644
--- a/drivers/accel/habanalabs/common/security.c
+++ b/drivers/accel/habanalabs/common/security.c
@@ -495,8 +495,8 @@ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_range_array: register range array
- * @regs_range_array_size: register range array size
+ * @user_regs_range_array: register range array
+ * @user_regs_range_array_size: register range array size
  *
  */
 int hl_init_pb_ranges_single_dcore(struct hl_device *hdev, u32 dcore_offset,
-- 
GitLab


From 9ec7639b5e124f273db20555cc38bd40057157b3 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Mon, 15 May 2023 13:32:18 +0300
Subject: [PATCH 0885/1791] accel/habanalabs: fix gaudi2_get_tpc_idle_status()
 return

The gaudi2_get_tpc_idle_status() function returned the incorrect variable
so it always returned true.

Fixes: d85f0531b928 ("accel/habanalabs: break is_idle function into per-engine sub-routines")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index d21ef9997d05c..e900017f4ff7d 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -7238,7 +7238,7 @@ static bool gaudi2_get_tpc_idle_status(struct hl_device *hdev, u64 *mask_arr, u8
 
 	gaudi2_iterate_tpcs(hdev, &tpc_iter);
 
-	return tpc_idle_data.is_idle;
+	return *tpc_idle_data.is_idle;
 }
 
 static bool gaudi2_get_decoder_idle_status(struct hl_device *hdev, u64 *mask_arr, u8 mask_len,
-- 
GitLab


From 964234aba59b0a92de701f25d0d6da7a5ebb3e16 Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Tue, 16 May 2023 09:56:18 +0300
Subject: [PATCH 0886/1791] accel/habanalabs: rename security functions related
 arguments

Make the argument names specify the registers array represent
registers that should be unsecured so the user can access them.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/security.c | 57 +++++++++++-----------
 1 file changed, 29 insertions(+), 28 deletions(-)

diff --git a/drivers/accel/habanalabs/common/security.c b/drivers/accel/habanalabs/common/security.c
index dc23ff57c91af..fe913965dbad7 100644
--- a/drivers/accel/habanalabs/common/security.c
+++ b/drivers/accel/habanalabs/common/security.c
@@ -284,14 +284,14 @@ void hl_secure_block(struct hl_device *hdev,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_array: register array
- * @regs_array_size: register array size
+ * @user_regs_array: unsecured register array
+ * @user_regs_array_size: unsecured register array size
  * @mask: enabled instances mask: 1- enabled, 0- disabled
  */
 int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores,
 		u32 dcore_offset, u32 num_instances, u32 instance_offset,
 		const u32 pb_blocks[], u32 blocks_array_size,
-		const u32 *regs_array, u32 regs_array_size, u64 mask)
+		const u32 *user_regs_array, u32 user_regs_array_size, u64 mask)
 {
 	int i, j;
 	struct hl_block_glbl_sec *glbl_sec;
@@ -303,8 +303,8 @@ int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores,
 		return -ENOMEM;
 
 	hl_secure_block(hdev, glbl_sec, blocks_array_size);
-	hl_unsecure_registers(hdev, regs_array, regs_array_size, 0, pb_blocks,
-			glbl_sec, blocks_array_size);
+	hl_unsecure_registers(hdev, user_regs_array, user_regs_array_size, 0,
+			pb_blocks, glbl_sec, blocks_array_size);
 
 	/* Fill all blocks with the same configuration */
 	for (i = 0 ; i < num_dcores ; i++) {
@@ -336,19 +336,19 @@ int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_array: register array
- * @regs_array_size: register array size
+ * @user_regs_array: unsecured register array
+ * @user_regs_array_size: unsecured register array size
  *
  */
 int hl_init_pb(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset,
 		u32 num_instances, u32 instance_offset,
 		const u32 pb_blocks[], u32 blocks_array_size,
-		const u32 *regs_array, u32 regs_array_size)
+		const u32 *user_regs_array, u32 user_regs_array_size)
 {
 	return hl_init_pb_with_mask(hdev, num_dcores, dcore_offset,
 			num_instances, instance_offset, pb_blocks,
-			blocks_array_size, regs_array, regs_array_size,
-			ULLONG_MAX);
+			blocks_array_size, user_regs_array,
+			user_regs_array_size, ULLONG_MAX);
 }
 
 /**
@@ -364,15 +364,15 @@ int hl_init_pb(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_range_array: register range array
- * @regs_range_array_size: register range array size
+ * @user_regs_range_array: unsecured register range array
+ * @user_regs_range_array_size: unsecured register range array size
  * @mask: enabled instances mask: 1- enabled, 0- disabled
  */
 int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores,
 		u32 dcore_offset, u32 num_instances, u32 instance_offset,
 		const u32 pb_blocks[], u32 blocks_array_size,
-		const struct range *regs_range_array, u32 regs_range_array_size,
-		u64 mask)
+		const struct range *user_regs_range_array,
+		u32 user_regs_range_array_size, u64 mask)
 {
 	int i, j, rc = 0;
 	struct hl_block_glbl_sec *glbl_sec;
@@ -384,8 +384,8 @@ int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores,
 		return -ENOMEM;
 
 	hl_secure_block(hdev, glbl_sec, blocks_array_size);
-	rc = hl_unsecure_registers_range(hdev, regs_range_array,
-			regs_range_array_size, 0, pb_blocks, glbl_sec,
+	rc = hl_unsecure_registers_range(hdev, user_regs_range_array,
+			user_regs_range_array_size, 0, pb_blocks, glbl_sec,
 			blocks_array_size);
 	if (rc)
 		goto free_glbl_sec;
@@ -422,19 +422,20 @@ int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_range_array: register range array
- * @regs_range_array_size: register range array size
+ * @user_regs_range_array: unsecured register range array
+ * @user_regs_range_array_size: unsecured register range array size
  *
  */
 int hl_init_pb_ranges(struct hl_device *hdev, u32 num_dcores,
 		u32 dcore_offset, u32 num_instances, u32 instance_offset,
 		const u32 pb_blocks[], u32 blocks_array_size,
-		const struct range *regs_range_array, u32 regs_range_array_size)
+		const struct range *user_regs_range_array,
+		u32 user_regs_range_array_size)
 {
 	return hl_init_pb_ranges_with_mask(hdev, num_dcores, dcore_offset,
 			num_instances, instance_offset, pb_blocks,
-			blocks_array_size, regs_range_array,
-			regs_range_array_size, ULLONG_MAX);
+			blocks_array_size, user_regs_range_array,
+			user_regs_range_array_size, ULLONG_MAX);
 }
 
 /**
@@ -447,14 +448,14 @@ int hl_init_pb_ranges(struct hl_device *hdev, u32 num_dcores,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @regs_array: register array
- * @regs_array_size: register array size
+ * @user_regs_array: unsecured register array
+ * @user_regs_array_size: unsecured register array size
  *
  */
 int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset,
 		u32 num_instances, u32 instance_offset,
 		const u32 pb_blocks[], u32 blocks_array_size,
-		const u32 *regs_array, u32 regs_array_size)
+		const u32 *user_regs_array, u32 user_regs_array_size)
 {
 	int i, rc = 0;
 	struct hl_block_glbl_sec *glbl_sec;
@@ -466,8 +467,8 @@ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset,
 		return -ENOMEM;
 
 	hl_secure_block(hdev, glbl_sec, blocks_array_size);
-	rc = hl_unsecure_registers(hdev, regs_array, regs_array_size, 0,
-			pb_blocks, glbl_sec, blocks_array_size);
+	rc = hl_unsecure_registers(hdev, user_regs_array, user_regs_array_size,
+			0, pb_blocks, glbl_sec, blocks_array_size);
 	if (rc)
 		goto free_glbl_sec;
 
@@ -495,8 +496,8 @@ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset,
  * @instance_offset: offset between instances
  * @pb_blocks: blocks array
  * @blocks_array_size: blocks array size
- * @user_regs_range_array: register range array
- * @user_regs_range_array_size: register range array size
+ * @user_regs_range_array: unsecured register range array
+ * @user_regs_range_array_size: unsecured register range array size
  *
  */
 int hl_init_pb_ranges_single_dcore(struct hl_device *hdev, u32 dcore_offset,
-- 
GitLab


From e715008b7c318fb1901a37bdd0941fb46c2b90b5 Mon Sep 17 00:00:00 2001
From: Oded Gabbay <ogabbay@kernel.org>
Date: Mon, 1 May 2023 13:12:22 +0300
Subject: [PATCH 0887/1791] accel/habanalabs: set unused bit as reserved

Get latest f/w gaudi2 interface file which marks unused
bist_need_iatu_config bit in cold_rst_data structure as reserved bit.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Ofir Bitton <obitton@habana.ai>
---
 drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h b/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h
index 8522f24deac03..18ca147b1c860 100644
--- a/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h
+++ b/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h
@@ -62,7 +62,7 @@ struct gaudi2_cold_rst_data {
 			u32 fake_security_enable : 1;
 			u32 fake_sig_validation_en : 1;
 			u32 bist_skip_enable : 1;
-			u32 bist_need_iatu_config : 1;
+			u32 reserved1 : 1;
 			u32 fake_bis_compliant : 1;
 			u32 wd_rst_cause_arm : 1;
 			u32 wd_rst_cause_arcpid : 1;
-- 
GitLab


From 314a7ffd7c196b27eedd50cb7553029e17789b55 Mon Sep 17 00:00:00 2001
From: Moti Haimovski <mhaimovski@habana.ai>
Date: Sun, 16 Apr 2023 18:36:17 +0300
Subject: [PATCH 0888/1791] accel/habanalabs: fix mem leak in capture user
 mappings

This commit fixes a memory leak caused when clearing the user_mappings
info when a new context is opened immediately after user_mapping is
captured and a hard reset is performed.

Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/habanalabs_drv.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c
index 1ec97da3dddb8..70fb2df9a93b8 100644
--- a/drivers/accel/habanalabs/common/habanalabs_drv.c
+++ b/drivers/accel/habanalabs/common/habanalabs_drv.c
@@ -13,6 +13,7 @@
 
 #include <linux/pci.h>
 #include <linux/module.h>
+#include <linux/vmalloc.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/habanalabs.h>
@@ -218,6 +219,7 @@ int hl_device_open(struct inode *inode, struct file *filp)
 
 	hl_debugfs_add_file(hpriv);
 
+	vfree(hdev->captured_err_info.page_fault_info.user_mappings);
 	memset(&hdev->captured_err_info, 0, sizeof(hdev->captured_err_info));
 	atomic_set(&hdev->captured_err_info.cs_timeout.write_enable, 1);
 	hdev->captured_err_info.undef_opcode.write_enable = true;
-- 
GitLab


From dcfce96ee834de6f9223670f3d2e1440559fb9b6 Mon Sep 17 00:00:00 2001
From: Oded Gabbay <ogabbay@kernel.org>
Date: Mon, 1 May 2023 13:19:57 +0300
Subject: [PATCH 0889/1791] accel/habanalabs: align to latest firmware specs

Update the firmware common interface files with the latest version.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Ofir Bitton <obitton@habana.ai>
---
 .../habanalabs/include/common/cpucp_if.h      | 18 ++++----
 .../habanalabs/include/common/hl_boot_if.h    | 41 ++++---------------
 2 files changed, 16 insertions(+), 43 deletions(-)

diff --git a/drivers/accel/habanalabs/include/common/cpucp_if.h b/drivers/accel/habanalabs/include/common/cpucp_if.h
index f68308cc25240..33807b839c372 100644
--- a/drivers/accel/habanalabs/include/common/cpucp_if.h
+++ b/drivers/accel/habanalabs/include/common/cpucp_if.h
@@ -359,7 +359,7 @@ struct hl_eq_entry {
 	union {
 		__le64 data_placeholder;
 		struct hl_eq_ecc_data ecc_data;
-		struct hl_eq_hbm_ecc_data hbm_ecc_data;	/* Gaudi1 HBM */
+		struct hl_eq_hbm_ecc_data hbm_ecc_data;	/* Obsolete */
 		struct hl_eq_sm_sei_data sm_sei_data;
 		struct cpucp_pkt_sync_err pkt_sync_err;
 		struct hl_eq_fw_alive fw_alive;
@@ -653,7 +653,7 @@ enum pq_init_status {
  *       which address is passed via the CpuCp packet. In addition, the host's driver
  *       passes the max size it allows the CpuCP to write to the structure, to prevent
  *       data corruption in case of mismatched driver/FW versions.
- *       Relevant only to Gaudi.
+ *       Obsolete.
  *
  * CPUCP_PACKET_GENERIC_PASSTHROUGH -
  *      Generic opcode for all firmware info that is only passed to host
@@ -868,19 +868,19 @@ struct cpucp_array_data_packet {
 enum cpucp_led_index {
 	CPUCP_LED0_INDEX = 0,
 	CPUCP_LED1_INDEX,
-	CPUCP_LED2_INDEX
+	CPUCP_LED2_INDEX,
+	CPUCP_LED_MAX_INDEX = CPUCP_LED2_INDEX
 };
 
 /*
  * enum cpucp_packet_rc - Error return code
  * @cpucp_packet_success	-> in case of success.
- * @cpucp_packet_invalid	-> this is to support Goya and Gaudi platform.
+ * @cpucp_packet_invalid	-> this is to support first generation platforms.
  * @cpucp_packet_fault		-> in case of processing error like failing to
  *                                 get device binding or semaphore etc.
- * @cpucp_packet_invalid_pkt	-> when cpucp packet is un-supported. This is
- *                                 supported Greco onwards.
+ * @cpucp_packet_invalid_pkt	-> when cpucp packet is un-supported.
  * @cpucp_packet_invalid_params	-> when checking parameter like length of buffer
- *				   or attribute value etc. Supported Greco onwards.
+ *				   or attribute value etc.
  * @cpucp_packet_rc_max		-> It indicates size of enum so should be at last.
  */
 enum cpucp_packet_rc {
@@ -1365,7 +1365,7 @@ struct cpucp_dev_info_signed {
 #define DCORE_MON_REGS_SZ	512
 /*
  * struct dcore_monitor_regs_data - DCORE monitor regs data.
- * the structure follows sync manager block layout. relevant only to Gaudi.
+ * the structure follows sync manager block layout. Obsolete.
  * @mon_pay_addrl: array of payload address low bits.
  * @mon_pay_addrh: array of payload address high bits.
  * @mon_pay_data: array of payload data.
@@ -1380,7 +1380,7 @@ struct dcore_monitor_regs_data {
 	__le32 mon_status[DCORE_MON_REGS_SZ];
 };
 
-/* contains SM data for each SYNC_MNGR (relevant only to Gaudi) */
+/* contains SM data for each SYNC_MNGR (Obsolete) */
 struct cpucp_monitor_dump {
 	struct dcore_monitor_regs_data sync_mngr_w_s;
 	struct dcore_monitor_regs_data sync_mngr_e_s;
diff --git a/drivers/accel/habanalabs/include/common/hl_boot_if.h b/drivers/accel/habanalabs/include/common/hl_boot_if.h
index c58d76a2705c5..cff79f7f9f751 100644
--- a/drivers/accel/habanalabs/include/common/hl_boot_if.h
+++ b/drivers/accel/habanalabs/include/common/hl_boot_if.h
@@ -35,6 +35,7 @@ enum cpu_boot_err {
 	CPU_BOOT_ERR_TPM_FAIL = 20,
 	CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL = 21,
 	CPU_BOOT_ERR_EEPROM_FAIL = 22,
+	CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL = 23,
 	CPU_BOOT_ERR_ENABLED = 31,
 	CPU_BOOT_ERR_SCND_EN = 63,
 	CPU_BOOT_ERR_LAST = 64 /* we have 2 registers of 32 bits */
@@ -51,6 +52,7 @@ enum cpu_boot_err {
 		 (1 << CPU_BOOT_ERR_DEVICE_UNUSABLE_FAIL) |	\
 		 (1 << CPU_BOOT_ERR_BINNING_FAIL) |		\
 		 (1 << CPU_BOOT_ERR_DRAM_SKIPPED) |		\
+		 (1 << CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL) |	\
 		 (1 << CPU_BOOT_ERR_EEPROM_FAIL))
 
 /*
@@ -132,6 +134,9 @@ enum cpu_boot_err {
  * CPU_BOOT_ERR_EEPROM_FAIL		Failed reading EEPROM data. Defaults
  *					are used.
  *
+ * CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL	Failed scrubbing the Engines/ARCFarm
+ *					memories. Boot disabled until reset.
+ *
  * CPU_BOOT_ERR0_ENABLED		Error registers enabled.
  *					This is a main indication that the
  *					running FW populates the error
@@ -157,6 +162,7 @@ enum cpu_boot_err {
 #define CPU_BOOT_ERR0_TPM_FAIL			(1 << CPU_BOOT_ERR_TPM_FAIL)
 #define CPU_BOOT_ERR0_TMP_THRESH_INIT_FAIL	(1 << CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL)
 #define CPU_BOOT_ERR0_EEPROM_FAIL		(1 << CPU_BOOT_ERR_EEPROM_FAIL)
+#define CPU_BOOT_ERR0_ENG_ARC_MEM_SCRUB_FAIL	(1 << CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL)
 #define CPU_BOOT_ERR0_ENABLED			(1 << CPU_BOOT_ERR_ENABLED)
 #define CPU_BOOT_ERR1_ENABLED			(1 << CPU_BOOT_ERR_ENABLED)
 
@@ -744,36 +750,6 @@ struct comms_status {
 	};
 };
 
-/**
- * HL_MODULES_MAX_NUM is determined by the size of modules_mask in struct
- *      hl_component_versions
- */
-enum hl_modules {
-	HL_MODULES_BOOT_INFO = 0,
-	HL_MODULES_EEPROM,
-	HL_MODULES_FDT,
-	HL_MODULES_I2C,
-	HL_MODULES_LZ4,
-	HL_MODULES_MBEDTLS,
-	HL_MODULES_MAX_NUM = 16
-};
-
-/**
- * HL_COMPONENTS_MAX_NUM is determined by the size of components_mask in
- *      struct cpucp_versions
- */
-enum hl_components {
-	HL_COMPONENTS_PID = 0,
-	HL_COMPONENTS_MGMT,
-	HL_COMPONENTS_PREBOOT,
-	HL_COMPONENTS_PPBOOT,
-	HL_COMPONENTS_ARMCP,
-	HL_COMPONENTS_CPLD,
-	HL_COMPONENTS_UBOOT,
-	HL_COMPONENTS_FUSE,
-	HL_COMPONENTS_MAX_NUM = 16
-};
-
 #define NAME_MAX_LEN	32 /* bytes */
 struct hl_module_data {
 	__u8 name[NAME_MAX_LEN];
@@ -787,8 +763,6 @@ struct hl_module_data {
  * @component: version of the component itself.
  * @fw_os: Firmware OS Version.
  * @comp_name: Name of the component.
- * @modules_mask: i'th bit (from LSB) is a flag - on if module i in enum
- *              hl_modules is used.
  * @modules_counter: number of set bits in modules_mask.
  * @reserved: reserved for future use.
  * @modules: versions of the component's modules. Elborated explanation in
@@ -800,9 +774,8 @@ struct hl_component_versions {
 	__u8 component[VERSION_MAX_LEN];
 	__u8 fw_os[VERSION_MAX_LEN];
 	__u8 comp_name[NAME_MAX_LEN];
-	__le16 modules_mask;
 	__u8 modules_counter;
-	__u8 reserved[1];
+	__u8 reserved[3];
 	struct hl_module_data modules[];
 };
 
-- 
GitLab


From adda800c049a4580aaec4d710a9972cdad5b14e3 Mon Sep 17 00:00:00 2001
From: Oded Gabbay <ogabbay@kernel.org>
Date: Wed, 3 May 2023 14:47:54 +0300
Subject: [PATCH 0890/1791] accel/habanalabs: print max timeout value on CS
 stuck

If a workload got stuck, we print an error to the kernel log about it.
Add to that print the configured max timeout value, as that value is
not fixed between ASICs and in addition it can be configured using
a kernel module parameter.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Ofir Bitton <obitton@habana.ai>
---
 .../habanalabs/common/command_submission.c    | 26 +++++++++++--------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/drivers/accel/habanalabs/common/command_submission.c b/drivers/accel/habanalabs/common/command_submission.c
index 4ff3256fdf8b2..cc205bad476eb 100644
--- a/drivers/accel/habanalabs/common/command_submission.c
+++ b/drivers/accel/habanalabs/common/command_submission.c
@@ -804,12 +804,14 @@ static void cs_do_release(struct kref *ref)
 
 static void cs_timedout(struct work_struct *work)
 {
+	struct hl_cs *cs = container_of(work, struct hl_cs, work_tdr.work);
+	bool skip_reset_on_timeout, device_reset = false;
 	struct hl_device *hdev;
 	u64 event_mask = 0x0;
+	uint timeout_sec;
 	int rc;
-	struct hl_cs *cs = container_of(work, struct hl_cs,
-						 work_tdr.work);
-	bool skip_reset_on_timeout = cs->skip_reset_on_timeout, device_reset = false;
+
+	skip_reset_on_timeout = cs->skip_reset_on_timeout;
 
 	rc = cs_get_unless_zero(cs);
 	if (!rc)
@@ -840,29 +842,31 @@ static void cs_timedout(struct work_struct *work)
 		event_mask |= HL_NOTIFIER_EVENT_CS_TIMEOUT;
 	}
 
+	timeout_sec = jiffies_to_msecs(hdev->timeout_jiffies) / 1000;
+
 	switch (cs->type) {
 	case CS_TYPE_SIGNAL:
 		dev_err(hdev->dev,
-			"Signal command submission %llu has not finished in time!\n",
-			cs->sequence);
+			"Signal command submission %llu has not finished in %u seconds!\n",
+			cs->sequence, timeout_sec);
 		break;
 
 	case CS_TYPE_WAIT:
 		dev_err(hdev->dev,
-			"Wait command submission %llu has not finished in time!\n",
-			cs->sequence);
+			"Wait command submission %llu has not finished in %u seconds!\n",
+			cs->sequence, timeout_sec);
 		break;
 
 	case CS_TYPE_COLLECTIVE_WAIT:
 		dev_err(hdev->dev,
-			"Collective Wait command submission %llu has not finished in time!\n",
-			cs->sequence);
+			"Collective Wait command submission %llu has not finished in %u seconds!\n",
+			cs->sequence, timeout_sec);
 		break;
 
 	default:
 		dev_err(hdev->dev,
-			"Command submission %llu has not finished in time!\n",
-			cs->sequence);
+			"Command submission %llu has not finished in %u seconds!\n",
+			cs->sequence, timeout_sec);
 		break;
 	}
 
-- 
GitLab


From b2d61fecb443c10f116fa00a1b9b5677a5822e52 Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Tue, 2 May 2023 17:28:56 +0300
Subject: [PATCH 0891/1791] accel/habanalabs: upon DMA errors, use FW-extracted
 error cause

Initially, the driver used to read the error cause data directly from
the ASIC. However, the FW now clears it before the driver could read
it. Therefore we should use the error cause data that is extracted by
the FW.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 37 +++++-------------------
 1 file changed, 8 insertions(+), 29 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index e900017f4ff7d..b8644d87f8175 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8807,13 +8807,13 @@ static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u16 event_type,
 	return error_count;
 }
 
-static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type, int sts_addr)
+static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type, u64 intr_cause)
 {
-	u32 error_count = 0, sts_val = RREG32(sts_addr);
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++)
-		if (sts_val & BIT(i)) {
+		if (intr_cause & BIT(i)) {
 			gaudi2_print_event(hdev, event_type, true,
 				"err cause: %s", gaudi2_dma_core_interrupts_cause[i]);
 			error_count++;
@@ -8824,27 +8824,6 @@ static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type,
 	return error_count;
 }
 
-static int gaudi2_handle_pdma_core_event(struct hl_device *hdev, u16 event_type, int pdma_idx)
-{
-	u32 sts_addr;
-
-	sts_addr = mmPDMA0_CORE_ERR_CAUSE + pdma_idx * PDMA_OFFSET;
-	return gaudi2_handle_dma_core_event(hdev, event_type, sts_addr);
-}
-
-static int gaudi2_handle_edma_core_event(struct hl_device *hdev, u16 event_type, int edma_idx)
-{
-	static const int edma_event_index_map[] = {2, 3, 0, 1, 6, 7, 4, 5};
-	u32 sts_addr, index;
-
-	index = edma_event_index_map[edma_idx];
-
-	sts_addr = mmDCORE0_EDMA0_CORE_ERR_CAUSE +
-				DCORE_OFFSET * (index / NUM_OF_EDMA_PER_DCORE) +
-				DCORE_EDMA_OFFSET * (index % NUM_OF_EDMA_PER_DCORE);
-	return gaudi2_handle_dma_core_event(hdev, event_type, sts_addr);
-}
-
 static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev, u64 *event_mask)
 {
 	u32 mstr_if_base_addr = mmPCIE_MSTR_RR_MSTR_IF_RR_SHRD_HBW_BASE, razwi_happened_addr;
@@ -9725,19 +9704,19 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_KDMA_CH0_AXI_ERR_RSP:
 	case GAUDI2_EVENT_KDMA0_CORE:
 		error_count = gaudi2_handle_kdma_core_event(hdev, event_type,
-					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_HDMA2_CORE ... GAUDI2_EVENT_HDMA5_CORE:
-		index = event_type - GAUDI2_EVENT_HDMA2_CORE;
-		error_count = gaudi2_handle_edma_core_event(hdev, event_type, index);
+		error_count = gaudi2_handle_dma_core_event(hdev, event_type,
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_PDMA0_CORE ... GAUDI2_EVENT_PDMA1_CORE:
-		index = event_type - GAUDI2_EVENT_PDMA0_CORE;
-		error_count = gaudi2_handle_pdma_core_event(hdev, event_type, index);
+		error_count = gaudi2_handle_dma_core_event(hdev, event_type,
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
-- 
GitLab


From 583f12a80dfb7997d59a42e8642019695f5aa15a Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Tue, 9 May 2023 14:02:49 +0300
Subject: [PATCH 0892/1791] accel/habanalabs: remove support for mmu disable

As mmu disable mode is only used for bring-up stages, let's remove this
option and all code related to it.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../accel/habanalabs/common/command_buffer.c  |   6 -
 .../habanalabs/common/command_submission.c    |  22 ++--
 drivers/accel/habanalabs/common/debugfs.c     |  23 +---
 drivers/accel/habanalabs/common/habanalabs.h  |  18 +--
 .../accel/habanalabs/common/habanalabs_drv.c  |   2 -
 .../habanalabs/common/habanalabs_ioctl.c      |   9 +-
 drivers/accel/habanalabs/common/memory.c      | 104 +-----------------
 drivers/accel/habanalabs/common/mmu/mmu.c     |  56 ++--------
 drivers/accel/habanalabs/gaudi/gaudi.c        |   6 +-
 drivers/accel/habanalabs/goya/goya.c          |   3 -
 .../accel/habanalabs/goya/goya_coresight.c    |   9 +-
 11 files changed, 26 insertions(+), 232 deletions(-)

diff --git a/drivers/accel/habanalabs/common/command_buffer.c b/drivers/accel/habanalabs/common/command_buffer.c
index 6e09f48750a02..08f7aee426246 100644
--- a/drivers/accel/habanalabs/common/command_buffer.c
+++ b/drivers/accel/habanalabs/common/command_buffer.c
@@ -27,12 +27,6 @@ static int cb_map_mem(struct hl_ctx *ctx, struct hl_cb *cb)
 		return -EINVAL;
 	}
 
-	if (!hdev->mmu_enable) {
-		dev_err_ratelimited(hdev->dev,
-				"Cannot map CB because MMU is disabled\n");
-		return -EINVAL;
-	}
-
 	if (cb->is_mmu_mapped)
 		return 0;
 
diff --git a/drivers/accel/habanalabs/common/command_submission.c b/drivers/accel/habanalabs/common/command_submission.c
index cc205bad476eb..c23829dab97a1 100644
--- a/drivers/accel/habanalabs/common/command_submission.c
+++ b/drivers/accel/habanalabs/common/command_submission.c
@@ -280,14 +280,8 @@ bool cs_needs_timeout(struct hl_cs *cs)
 
 static bool is_cb_patched(struct hl_device *hdev, struct hl_cs_job *job)
 {
-	/*
-	 * Patched CB is created for external queues jobs, and for H/W queues
-	 * jobs if the user CB was allocated by driver and MMU is disabled.
-	 */
-	return (job->queue_type == QUEUE_TYPE_EXT ||
-			(job->queue_type == QUEUE_TYPE_HW &&
-					job->is_kernel_allocated_cb &&
-					!hdev->mmu_enable));
+	/* Patched CB is created for external queues jobs */
+	return (job->queue_type == QUEUE_TYPE_EXT);
 }
 
 /*
@@ -363,14 +357,13 @@ static void hl_complete_job(struct hl_device *hdev, struct hl_cs_job *job)
 		}
 	}
 
-	/* For H/W queue jobs, if a user CB was allocated by driver and MMU is
-	 * enabled, the user CB isn't released in cs_parser() and thus should be
+	/* For H/W queue jobs, if a user CB was allocated by driver,
+	 * the user CB isn't released in cs_parser() and thus should be
 	 * released here. This is also true for INT queues jobs which were
 	 * allocated by driver.
 	 */
-	if ((job->is_kernel_allocated_cb &&
-		((job->queue_type == QUEUE_TYPE_HW && hdev->mmu_enable) ||
-				job->queue_type == QUEUE_TYPE_INT))) {
+	if (job->is_kernel_allocated_cb &&
+			(job->queue_type == QUEUE_TYPE_HW || job->queue_type == QUEUE_TYPE_INT)) {
 		atomic_dec(&job->user_cb->cs_cnt);
 		hl_cb_put(job->user_cb);
 	}
@@ -1951,8 +1944,7 @@ static int cs_ioctl_signal_wait_create_jobs(struct hl_device *hdev,
 	else
 		cb_size = hdev->asic_funcs->get_signal_cb_size(hdev);
 
-	cb = hl_cb_kernel_create(hdev, cb_size,
-				q_type == QUEUE_TYPE_HW && hdev->mmu_enable);
+	cb = hl_cb_kernel_create(hdev, cb_size, q_type == QUEUE_TYPE_HW);
 	if (!cb) {
 		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
 		atomic64_inc(&cntr->out_of_mem_drop_cnt);
diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c
index 29b78c7cf5de8..9e84a47a21dcf 100644
--- a/drivers/accel/habanalabs/common/debugfs.c
+++ b/drivers/accel/habanalabs/common/debugfs.c
@@ -255,9 +255,6 @@ static int vm_show(struct seq_file *s, void *data)
 	u64 j;
 	int i;
 
-	if (!dev_entry->hdev->mmu_enable)
-		return 0;
-
 	mutex_lock(&dev_entry->ctx_mem_hash_mutex);
 
 	list_for_each_entry(ctx, &dev_entry->ctx_mem_hash_list, debugfs_list) {
@@ -436,9 +433,6 @@ static int mmu_show(struct seq_file *s, void *data)
 	u64 virt_addr = dev_entry->mmu_addr, phys_addr;
 	int i;
 
-	if (!hdev->mmu_enable)
-		return 0;
-
 	if (dev_entry->mmu_asid == HL_KERNEL_ASID_ID)
 		ctx = hdev->kernel_ctx;
 	else
@@ -496,9 +490,6 @@ static ssize_t mmu_asid_va_write(struct file *file, const char __user *buf,
 	char *c;
 	ssize_t rc;
 
-	if (!hdev->mmu_enable)
-		return count;
-
 	if (count > sizeof(kbuf) - 1)
 		goto err;
 	if (copy_from_user(kbuf, buf, count))
@@ -535,9 +526,6 @@ static int mmu_ack_error(struct seq_file *s, void *data)
 	struct hl_device *hdev = dev_entry->hdev;
 	int rc;
 
-	if (!hdev->mmu_enable)
-		return 0;
-
 	if (!dev_entry->mmu_cap_mask) {
 		dev_err(hdev->dev, "mmu_cap_mask is not set\n");
 		goto err;
@@ -563,9 +551,6 @@ static ssize_t mmu_ack_error_value_write(struct file *file,
 	char kbuf[MMU_KBUF_SIZE];
 	ssize_t rc;
 
-	if (!hdev->mmu_enable)
-		return count;
-
 	if (count > sizeof(kbuf) - 1)
 		goto err;
 
@@ -661,9 +646,6 @@ static bool hl_is_device_va(struct hl_device *hdev, u64 addr)
 {
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
 
-	if (!hdev->mmu_enable)
-		goto out;
-
 	if (prop->dram_supports_virtual_memory &&
 		(addr >= prop->dmmu.start_addr && addr < prop->dmmu.end_addr))
 		return true;
@@ -675,7 +657,7 @@ static bool hl_is_device_va(struct hl_device *hdev, u64 addr)
 	if (addr >= prop->pmmu_huge.start_addr &&
 		addr < prop->pmmu_huge.end_addr)
 		return true;
-out:
+
 	return false;
 }
 
@@ -685,9 +667,6 @@ static bool hl_is_device_internal_memory_va(struct hl_device *hdev, u64 addr,
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
 	u64 dram_start_addr, dram_end_addr;
 
-	if (!hdev->mmu_enable)
-		return false;
-
 	if (prop->dram_supports_virtual_memory) {
 		dram_start_addr = prop->dmmu.start_addr;
 		dram_end_addr = prop->dmmu.end_addr;
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index ea0914d08bdc3..e2341a75a4b71 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -115,18 +115,6 @@ enum hl_mmu_page_table_location {
 	MMU_NUM_PGT_LOCATIONS	/* num of PGT locations */
 };
 
-/**
- * enum hl_mmu_enablement - what mmu modules to enable
- * @MMU_EN_NONE: mmu disabled.
- * @MMU_EN_ALL: enable all.
- * @MMU_EN_PMMU_ONLY: Enable only the PMMU leaving the DMMU disabled.
- */
-enum hl_mmu_enablement {
-	MMU_EN_NONE = 0,
-	MMU_EN_ALL = 1,
-	MMU_EN_PMMU_ONLY = 3,	/* N/A for Goya/Gaudi */
-};
-
 /*
  * HL_RSVD_SOBS 'sync stream' reserved sync objects per QMAN stream
  * HL_RSVD_MONS 'sync stream' reserved monitors per QMAN stream
@@ -3319,7 +3307,7 @@ struct hl_reset_info {
  * @nic_ports_mask: Controls which NIC ports are enabled. Used only for testing.
  * @fw_components: Controls which f/w components to load to the device. There are multiple f/w
  *                 stages and sometimes we want to stop at a certain stage. Used only for testing.
- * @mmu_enable: Whether to enable or disable the device MMU(s). Used only for testing.
+ * @mmu_disable: Disable the device MMU(s). Used only for testing.
  * @cpu_queues_enable: Whether to enable queues communication vs. the f/w. Used only for testing.
  * @pldm: Whether we are running in Palladium environment. Used only for testing.
  * @hard_reset_on_fw_events: Whether to do device hard-reset when a fatal event is received from
@@ -3482,7 +3470,7 @@ struct hl_device {
 	/* Parameters for bring-up to be upstreamed */
 	u64				nic_ports_mask;
 	u64				fw_components;
-	u8				mmu_enable;
+	u8				mmu_disable;
 	u8				cpu_queues_enable;
 	u8				pldm;
 	u8				hard_reset_on_fw_events;
@@ -3827,8 +3815,6 @@ struct pgt_info *hl_mmu_hr_get_alloc_next_hop(struct hl_ctx *ctx,
 							u64 curr_pte, bool *is_new_hop);
 int hl_mmu_hr_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr, struct hl_mmu_hop_info *hops,
 							struct hl_hr_mmu_funcs *hr_func);
-void hl_mmu_swap_out(struct hl_ctx *ctx);
-void hl_mmu_swap_in(struct hl_ctx *ctx);
 int hl_mmu_if_set_funcs(struct hl_device *hdev);
 void hl_mmu_v1_set_funcs(struct hl_device *hdev, struct hl_mmu_funcs *mmu);
 void hl_mmu_v2_hr_set_funcs(struct hl_device *hdev, struct hl_mmu_funcs *mmu);
diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c
index 70fb2df9a93b8..446f444a1c7e4 100644
--- a/drivers/accel/habanalabs/common/habanalabs_drv.c
+++ b/drivers/accel/habanalabs/common/habanalabs_drv.c
@@ -307,7 +307,6 @@ static void set_driver_behavior_per_device(struct hl_device *hdev)
 {
 	hdev->nic_ports_mask = 0;
 	hdev->fw_components = FW_TYPE_ALL_TYPES;
-	hdev->mmu_enable = MMU_EN_ALL;
 	hdev->cpu_queues_enable = 1;
 	hdev->pldm = 0;
 	hdev->hard_reset_on_fw_events = 1;
@@ -382,7 +381,6 @@ static int fixup_device_params(struct hl_device *hdev)
 	/* If CPU queues not enabled, no way to do heartbeat */
 	if (!hdev->cpu_queues_enable)
 		hdev->heartbeat = 0;
-
 	fixup_device_params_per_asic(hdev, tmp_timeout);
 
 	return 0;
diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
index 4368e6c9a23af..9a8be9395fb28 100644
--- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
@@ -62,7 +62,7 @@ static int hw_ip_info(struct hl_device *hdev, struct hl_info_args *args)
 	hw_ip.device_id = hdev->asic_funcs->get_pci_id(hdev);
 	hw_ip.sram_base_address = prop->sram_user_base_address;
 	hw_ip.dram_base_address =
-			hdev->mmu_enable && prop->dram_supports_virtual_memory ?
+			prop->dram_supports_virtual_memory ?
 			prop->dmmu.start_addr : prop->dram_user_base_address;
 	hw_ip.tpc_enabled_mask = prop->tpc_enabled_mask & 0xFF;
 	hw_ip.tpc_enabled_mask_ext = prop->tpc_enabled_mask;
@@ -71,11 +71,8 @@ static int hw_ip_info(struct hl_device *hdev, struct hl_info_args *args)
 
 	dram_available_size = prop->dram_size - dram_kmd_size;
 
-	if (hdev->mmu_enable == MMU_EN_ALL)
-		hw_ip.dram_size = DIV_ROUND_DOWN_ULL(dram_available_size,
-				prop->dram_page_size) * prop->dram_page_size;
-	else
-		hw_ip.dram_size = dram_available_size;
+	hw_ip.dram_size = DIV_ROUND_DOWN_ULL(dram_available_size, prop->dram_page_size) *
+				prop->dram_page_size;
 
 	if (hw_ip.dram_size > PAGE_SIZE)
 		hw_ip.dram_enabled = 1;
diff --git a/drivers/accel/habanalabs/common/memory.c b/drivers/accel/habanalabs/common/memory.c
index a7b6a273ce21c..4fc72a07d2f59 100644
--- a/drivers/accel/habanalabs/common/memory.c
+++ b/drivers/accel/habanalabs/common/memory.c
@@ -1034,30 +1034,6 @@ static void unmap_phys_pg_pack(struct hl_ctx *ctx, u64 vaddr,
 	}
 }
 
-static int get_paddr_from_handle(struct hl_ctx *ctx, struct hl_mem_in *args,
-					u64 *paddr)
-{
-	struct hl_device *hdev = ctx->hdev;
-	struct hl_vm *vm = &hdev->vm;
-	struct hl_vm_phys_pg_pack *phys_pg_pack;
-	u32 handle;
-
-	handle = lower_32_bits(args->map_device.handle);
-	spin_lock(&vm->idr_lock);
-	phys_pg_pack = idr_find(&vm->phys_pg_pack_handles, handle);
-	if (!phys_pg_pack) {
-		spin_unlock(&vm->idr_lock);
-		dev_err(hdev->dev, "no match for handle %u\n", handle);
-		return -EINVAL;
-	}
-
-	*paddr = phys_pg_pack->pages[0];
-
-	spin_unlock(&vm->idr_lock);
-
-	return 0;
-}
-
 /**
  * map_device_va() - map the given memory.
  * @ctx: pointer to the context structure.
@@ -2094,76 +2070,6 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 addr, u64 size, u64 o
 	return rc;
 }
 
-static int mem_ioctl_no_mmu(struct hl_fpriv *hpriv, union hl_mem_args *args)
-{
-	struct hl_device *hdev = hpriv->hdev;
-	u64 block_handle, device_addr = 0;
-	struct hl_ctx *ctx = hpriv->ctx;
-	u32 handle = 0, block_size;
-	int rc;
-
-	switch (args->in.op) {
-	case HL_MEM_OP_ALLOC:
-		if (args->in.alloc.mem_size == 0) {
-			dev_err(hdev->dev, "alloc size must be larger than 0\n");
-			rc = -EINVAL;
-			goto out;
-		}
-
-		/* Force contiguous as there are no real MMU
-		 * translations to overcome physical memory gaps
-		 */
-		args->in.flags |= HL_MEM_CONTIGUOUS;
-		rc = alloc_device_memory(ctx, &args->in, &handle);
-
-		memset(args, 0, sizeof(*args));
-		args->out.handle = (__u64) handle;
-		break;
-
-	case HL_MEM_OP_FREE:
-		rc = free_device_memory(ctx, &args->in);
-		break;
-
-	case HL_MEM_OP_MAP:
-		if (args->in.flags & HL_MEM_USERPTR) {
-			dev_err(hdev->dev, "Failed to map host memory when MMU is disabled\n");
-			rc = -EPERM;
-		} else {
-			rc = get_paddr_from_handle(ctx, &args->in, &device_addr);
-			memset(args, 0, sizeof(*args));
-			args->out.device_virt_addr = device_addr;
-		}
-
-		break;
-
-	case HL_MEM_OP_UNMAP:
-		rc = 0;
-		break;
-
-	case HL_MEM_OP_MAP_BLOCK:
-		rc = map_block(hdev, args->in.map_block.block_addr, &block_handle, &block_size);
-		args->out.block_handle = block_handle;
-		args->out.block_size = block_size;
-		break;
-
-	case HL_MEM_OP_EXPORT_DMABUF_FD:
-		dev_err(hdev->dev, "Failed to export dma-buf object when MMU is disabled\n");
-		rc = -EPERM;
-		break;
-
-	case HL_MEM_OP_TS_ALLOC:
-		rc = allocate_timestamps_buffers(hpriv, &args->in, &args->out.handle);
-		break;
-	default:
-		dev_err(hdev->dev, "Unknown opcode for memory IOCTL\n");
-		rc = -EINVAL;
-		break;
-	}
-
-out:
-	return rc;
-}
-
 static void ts_buff_release(struct hl_mmap_mem_buf *buf)
 {
 	struct hl_ts_buff *ts_buff = buf->private;
@@ -2282,9 +2188,6 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 		return -EBUSY;
 	}
 
-	if (!hdev->mmu_enable)
-		return mem_ioctl_no_mmu(hpriv, args);
-
 	switch (args->in.op) {
 	case HL_MEM_OP_ALLOC:
 		if (args->in.alloc.mem_size == 0) {
@@ -2779,13 +2682,10 @@ int hl_vm_ctx_init(struct hl_ctx *ctx)
 	atomic64_set(&ctx->dram_phys_mem, 0);
 
 	/*
-	 * - If MMU is enabled, init the ranges as usual.
-	 * - If MMU is disabled, in case of host mapping, the returned address
-	 *   is the given one.
 	 *   In case of DRAM mapping, the returned address is the physical
 	 *   address of the memory related to the given handle.
 	 */
-	if (!ctx->hdev->mmu_enable)
+	if (ctx->hdev->mmu_disable)
 		return 0;
 
 	dram_range_start = prop->dmmu.start_addr;
@@ -2835,7 +2735,7 @@ void hl_vm_ctx_fini(struct hl_ctx *ctx)
 	struct hl_mem_in args;
 	int i;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return;
 
 	hl_debugfs_remove_ctx_mem_hash(hdev, ctx);
diff --git a/drivers/accel/habanalabs/common/mmu/mmu.c b/drivers/accel/habanalabs/common/mmu/mmu.c
index f379e5b461a6d..b2145716c6053 100644
--- a/drivers/accel/habanalabs/common/mmu/mmu.c
+++ b/drivers/accel/habanalabs/common/mmu/mmu.c
@@ -44,7 +44,7 @@ int hl_mmu_init(struct hl_device *hdev)
 {
 	int rc = -EOPNOTSUPP;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return 0;
 
 	mutex_init(&hdev->mmu_lock);
@@ -82,7 +82,7 @@ int hl_mmu_init(struct hl_device *hdev)
  */
 void hl_mmu_fini(struct hl_device *hdev)
 {
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return;
 
 	if (hdev->mmu_func[MMU_DR_PGT].fini != NULL)
@@ -107,7 +107,7 @@ int hl_mmu_ctx_init(struct hl_ctx *ctx)
 	struct hl_device *hdev = ctx->hdev;
 	int rc = -EOPNOTSUPP;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return 0;
 
 	if (hdev->mmu_func[MMU_DR_PGT].ctx_init != NULL) {
@@ -145,7 +145,7 @@ void hl_mmu_ctx_fini(struct hl_ctx *ctx)
 {
 	struct hl_device *hdev = ctx->hdev;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return;
 
 	if (hdev->mmu_func[MMU_DR_PGT].ctx_fini != NULL)
@@ -233,7 +233,7 @@ int hl_mmu_unmap_page(struct hl_ctx *ctx, u64 virt_addr, u32 page_size, bool flu
 	u64 real_virt_addr;
 	bool is_dram_addr;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return 0;
 
 	is_dram_addr = hl_is_dram_va(hdev, virt_addr);
@@ -301,7 +301,7 @@ int hl_mmu_map_page(struct hl_ctx *ctx, u64 virt_addr, u64 phys_addr, u32 page_s
 	bool is_dram_addr;
 
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return 0;
 
 	is_dram_addr = hl_is_dram_va(hdev, virt_addr);
@@ -472,46 +472,6 @@ int hl_mmu_unmap_contiguous(struct hl_ctx *ctx, u64 virt_addr, u32 size)
 	return rc;
 }
 
-/*
- * hl_mmu_swap_out - marks all mapping of the given ctx as swapped out
- *
- * @ctx: pointer to the context structure
- *
- */
-void hl_mmu_swap_out(struct hl_ctx *ctx)
-{
-	struct hl_device *hdev = ctx->hdev;
-
-	if (!hdev->mmu_enable)
-		return;
-
-	if (hdev->mmu_func[MMU_DR_PGT].swap_out != NULL)
-		hdev->mmu_func[MMU_DR_PGT].swap_out(ctx);
-
-	if (hdev->mmu_func[MMU_HR_PGT].swap_out != NULL)
-		hdev->mmu_func[MMU_HR_PGT].swap_out(ctx);
-}
-
-/*
- * hl_mmu_swap_in - marks all mapping of the given ctx as swapped in
- *
- * @ctx: pointer to the context structure
- *
- */
-void hl_mmu_swap_in(struct hl_ctx *ctx)
-{
-	struct hl_device *hdev = ctx->hdev;
-
-	if (!hdev->mmu_enable)
-		return;
-
-	if (hdev->mmu_func[MMU_DR_PGT].swap_in != NULL)
-		hdev->mmu_func[MMU_DR_PGT].swap_in(ctx);
-
-	if (hdev->mmu_func[MMU_HR_PGT].swap_in != NULL)
-		hdev->mmu_func[MMU_HR_PGT].swap_in(ctx);
-}
-
 static void hl_mmu_pa_page_with_offset(struct hl_ctx *ctx, u64 virt_addr,
 						struct hl_mmu_hop_info *hops,
 						u64 *phys_addr)
@@ -594,7 +554,7 @@ int hl_mmu_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr,
 	int pgt_residency, rc;
 	bool is_dram_addr;
 
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return -EOPNOTSUPP;
 
 	prop = &hdev->asic_prop;
@@ -625,7 +585,7 @@ int hl_mmu_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr,
 
 int hl_mmu_if_set_funcs(struct hl_device *hdev)
 {
-	if (!hdev->mmu_enable)
+	if (hdev->mmu_disable)
 		return 0;
 
 	switch (hdev->asic_type) {
diff --git a/drivers/accel/habanalabs/gaudi/gaudi.c b/drivers/accel/habanalabs/gaudi/gaudi.c
index a1697581c2189..056e2ef44afb5 100644
--- a/drivers/accel/habanalabs/gaudi/gaudi.c
+++ b/drivers/accel/habanalabs/gaudi/gaudi.c
@@ -1469,8 +1469,7 @@ static int gaudi_collective_wait_create_job(struct hl_device *hdev,
 	}
 
 	/* Allocate internal mapped CB for non patched CBs */
-	cb = hl_cb_kernel_create(hdev, cb_size,
-			hdev->mmu_enable && !patched_cb);
+	cb = hl_cb_kernel_create(hdev, cb_size, !patched_cb);
 	if (!cb) {
 		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
 		atomic64_inc(&cntr->out_of_mem_drop_cnt);
@@ -3644,9 +3643,6 @@ static int gaudi_mmu_init(struct hl_device *hdev)
 	u64 hop0_addr;
 	int rc, i;
 
-	if (!hdev->mmu_enable)
-		return 0;
-
 	if (gaudi->hw_cap_initialized & HW_CAP_MMU)
 		return 0;
 
diff --git a/drivers/accel/habanalabs/goya/goya.c b/drivers/accel/habanalabs/goya/goya.c
index fb0ac9df841a7..7c685e6075f6e 100644
--- a/drivers/accel/habanalabs/goya/goya.c
+++ b/drivers/accel/habanalabs/goya/goya.c
@@ -2671,9 +2671,6 @@ int goya_mmu_init(struct hl_device *hdev)
 	u64 hop0_addr;
 	int rc, i;
 
-	if (!hdev->mmu_enable)
-		return 0;
-
 	if (goya->hw_cap_initialized & HW_CAP_MMU)
 		return 0;
 
diff --git a/drivers/accel/habanalabs/goya/goya_coresight.c b/drivers/accel/habanalabs/goya/goya_coresight.c
index e7ac3046cfaae..a6d6cc38bcd84 100644
--- a/drivers/accel/habanalabs/goya/goya_coresight.c
+++ b/drivers/accel/habanalabs/goya/goya_coresight.c
@@ -371,13 +371,8 @@ static int goya_etr_validate_address(struct hl_device *hdev, u64 addr,
 		return false;
 	}
 
-	if (hdev->mmu_enable) {
-		range_start = prop->dmmu.start_addr;
-		range_end = prop->dmmu.end_addr;
-	} else {
-		range_start = prop->dram_user_base_address;
-		range_end = prop->dram_end_address;
-	}
+	range_start = prop->dmmu.start_addr;
+	range_end = prop->dmmu.end_addr;
 
 	return hl_mem_area_inside_range(addr, size, range_start, range_end);
 }
-- 
GitLab


From dcc8fa88d46ef97ea9812735d0d54591bc16b242 Mon Sep 17 00:00:00 2001
From: Dani Liberman <dliberman@habana.ai>
Date: Mon, 8 May 2023 17:24:38 +0300
Subject: [PATCH 0893/1791] accel/habanalabs: use binning info when handling
 razwi

When receiving sei interrupt from tpc or decoder, we need to check
the binning mask because if the engine is binned, the razwi info
won't be in the router of the binned engine, instead will be in the
router of the substitute engine.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index b8644d87f8175..a6aa17d86820e 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8040,7 +8040,7 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev,
 				u8 module_sub_idx, u64 *event_mask)
 {
 	bool via_sft = false;
-	u32 hbw_rtr_id, lbw_rtr_id, dcore_id, dcore_rtr_id, eng_id;
+	u32 hbw_rtr_id, lbw_rtr_id, dcore_id, dcore_rtr_id, eng_id, binned_idx;
 	u64 hbw_rtr_mstr_if_base_addr, lbw_rtr_mstr_if_base_addr;
 	u32 hbw_shrd_aw = 0, hbw_shrd_ar = 0;
 	u32 lbw_shrd_aw = 0, lbw_shrd_ar = 0;
@@ -8048,6 +8048,13 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev,
 
 	switch (module) {
 	case RAZWI_TPC:
+		sprintf(initiator_name, "TPC_%u", module_idx);
+		if (hdev->tpc_binning) {
+			binned_idx = __ffs(hdev->tpc_binning);
+			if (binned_idx == module_idx)
+				module_idx = TPC_ID_DCORE0_TPC6;
+		}
+
 		hbw_rtr_id = gaudi2_tpc_initiator_hbw_rtr_id[module_idx];
 
 		if (hl_is_fw_sw_ver_below(hdev, 1, 9) &&
@@ -8056,7 +8063,6 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev,
 			lbw_rtr_id = DCORE0_RTR0;
 		else
 			lbw_rtr_id = gaudi2_tpc_initiator_lbw_rtr_id[module_idx];
-		sprintf(initiator_name, "TPC_%u", module_idx);
 		break;
 	case RAZWI_MME:
 		sprintf(initiator_name, "MME_%u", module_idx);
@@ -8115,9 +8121,14 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev,
 		sprintf(initiator_name, "NIC_%u", module_idx);
 		break;
 	case RAZWI_DEC:
+		sprintf(initiator_name, "DEC_%u", module_idx);
+		if (hdev->decoder_binning) {
+			binned_idx = __ffs(hdev->decoder_binning);
+			if (binned_idx == module_idx)
+				module_idx = DEC_ID_PCIE_VDEC1;
+		}
 		hbw_rtr_id = gaudi2_dec_initiator_hbw_rtr_id[module_idx];
 		lbw_rtr_id = gaudi2_dec_initiator_lbw_rtr_id[module_idx];
-		sprintf(initiator_name, "DEC_%u", module_idx);
 		break;
 	case RAZWI_ROT:
 		hbw_rtr_id = gaudi2_rot_initiator_hbw_rtr_id[module_idx];
-- 
GitLab


From 54381ee8099839c8163cee45b65d824ca76186a2 Mon Sep 17 00:00:00 2001
From: Tomer Tayar <ttayar@habana.ai>
Date: Wed, 10 May 2023 13:34:28 +0300
Subject: [PATCH 0894/1791] accel/habanalabs: use lower QM in QM errors
 handling

The QMAN GLBL_ERR_STS_4 register has indications for errors also in the
lower CQ and the ARC CQ, and not just for errors in the lower CP.
Modify the relevant define/struct and the related print to use "lower
QM" instead of "lower CP".

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index a6aa17d86820e..6e2561ead546d 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -57,7 +57,7 @@
 
 #define GAUDI2_NA_EVENT_CAUSE			0xFF
 #define GAUDI2_NUM_OF_QM_ERR_CAUSE		18
-#define GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE		25
+#define GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE	25
 #define GAUDI2_NUM_OF_QM_ARB_ERR_CAUSE		3
 #define GAUDI2_NUM_OF_ARC_SEI_ERR_CAUSE		14
 #define GAUDI2_NUM_OF_CPU_SEI_ERR_CAUSE		3
@@ -801,7 +801,7 @@ static const char * const gaudi2_qman_error_cause[GAUDI2_NUM_OF_QM_ERR_CAUSE] =
 	"PQC L2H error"
 };
 
-static const char * const gaudi2_qman_lower_cp_error_cause[GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE] = {
+static const char * const gaudi2_lower_qman_error_cause[GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE] = {
 	"RSVD0",
 	"CQ AXI HBW error",
 	"CP AXI HBW error",
@@ -7895,8 +7895,8 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type
 			continue;
 
 		if (i == QMAN_STREAMS) {
-			snprintf(reg_desc, ARRAY_SIZE(reg_desc), "LowerCP");
-			num_error_causes = GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE;
+			snprintf(reg_desc, ARRAY_SIZE(reg_desc), "LowerQM");
+			num_error_causes = GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE;
 		} else {
 			snprintf(reg_desc, ARRAY_SIZE(reg_desc), "stream%u", i);
 			num_error_causes = GAUDI2_NUM_OF_QM_ERR_CAUSE;
@@ -7907,7 +7907,7 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type
 				gaudi2_print_event(hdev, event_type, true,
 					"%s. err cause: %s", reg_desc,
 					i == QMAN_STREAMS ?
-					gaudi2_qman_lower_cp_error_cause[j] :
+					gaudi2_lower_qman_error_cause[j] :
 					gaudi2_qman_error_cause[j]);
 				error_count++;
 			}
-- 
GitLab


From 6092cedfff1180b0d0aea402a198397fff609c71 Mon Sep 17 00:00:00 2001
From: Tomer Tayar <ttayar@habana.ai>
Date: Wed, 10 May 2023 13:28:05 +0300
Subject: [PATCH 0895/1791] accel/habanalabs: print qman data on error only for
 lower qman

By default, the upper QMANs are not used, and instead engines ARCs
access the lower QMANs directly.
Errors for upper QMANs are therefore not expected, and the debug print
of the PQ entries is not needed.

Modify the QMAN debug data print on errors to include only information
for the lower QMAN.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c      | 146 +++---------------
 drivers/accel/habanalabs/gaudi2/gaudi2P.h     |   2 +-
 .../include/gaudi2/asic_reg/gaudi2_regs.h     |  11 ++
 3 files changed, 31 insertions(+), 128 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 6e2561ead546d..4981b8eb0ff50 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -7744,137 +7744,28 @@ static bool gaudi2_handle_ecc_event(struct hl_device *hdev, u16 event_type,
 	return !!ecc_data->is_critical;
 }
 
-/*
- * gaudi2_queue_idx_dec - decrement queue index (pi/ci) and handle wrap
- *
- * @idx: the current pi/ci value
- * @q_len: the queue length (power of 2)
- *
- * @return the cyclically decremented index
- */
-static inline u32 gaudi2_queue_idx_dec(u32 idx, u32 q_len)
-{
-	u32 mask = q_len - 1;
-
-	/*
-	 * modular decrement is equivalent to adding (queue_size -1)
-	 * later we take LSBs to make sure the value is in the
-	 * range [0, queue_len - 1]
-	 */
-	return (idx + q_len - 1) & mask;
-}
-
-/**
- * gaudi2_print_sw_config_stream_data - print SW config stream data
- *
- * @hdev: pointer to the habanalabs device structure
- * @stream: the QMAN's stream
- * @qman_base: base address of QMAN registers block
- */
-static void gaudi2_print_sw_config_stream_data(struct hl_device *hdev,
-						u32 stream, u64 qman_base)
+static void print_lower_qman_data_on_err(struct hl_device *hdev, u64 qman_base)
 {
-	u64 cq_ptr_lo, cq_ptr_hi, cq_tsize, cq_ptr;
-	u32 cq_ptr_lo_off, size;
+	u32 lo, hi, cq_ptr_size, arc_cq_ptr_size;
+	u64 cq_ptr, arc_cq_ptr, cp_current_inst;
 
-	cq_ptr_lo_off = mmDCORE0_TPC0_QM_CQ_PTR_LO_1 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0;
-
-	cq_ptr_lo = qman_base + (mmDCORE0_TPC0_QM_CQ_PTR_LO_0 - mmDCORE0_TPC0_QM_BASE) +
-									stream * cq_ptr_lo_off;
-
-	cq_ptr_hi = cq_ptr_lo + (mmDCORE0_TPC0_QM_CQ_PTR_HI_0 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0);
-
-	cq_tsize = cq_ptr_lo + (mmDCORE0_TPC0_QM_CQ_TSIZE_0 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0);
-
-	cq_ptr = (((u64) RREG32(cq_ptr_hi)) << 32) | RREG32(cq_ptr_lo);
-	size = RREG32(cq_tsize);
-	dev_info(hdev->dev, "stop on err: stream: %u, addr: %#llx, size: %x\n",
-		stream, cq_ptr, size);
-}
-
-/**
- * gaudi2_print_last_pqes_on_err - print last PQEs on error
- *
- * @hdev: pointer to the habanalabs device structure
- * @qid_base: first QID of the QMAN (out of 4 streams)
- * @stream: the QMAN's stream
- * @qman_base: base address of QMAN registers block
- * @pr_sw_conf: if true print the SW config stream data (CQ PTR and SIZE)
- */
-static void gaudi2_print_last_pqes_on_err(struct hl_device *hdev, u32 qid_base, u32 stream,
-						u64 qman_base, bool pr_sw_conf)
-{
-	u32 ci, qm_ci_stream_off;
-	struct hl_hw_queue *q;
-	u64 pq_ci;
-	int i;
+	lo = RREG32(qman_base + QM_CQ_PTR_LO_4_OFFSET);
+	hi = RREG32(qman_base + QM_CQ_PTR_HI_4_OFFSET);
+	cq_ptr = ((u64) hi) << 32 | lo;
+	cq_ptr_size = RREG32(qman_base + QM_CQ_TSIZE_4_OFFSET);
 
-	q = &hdev->kernel_queues[qid_base + stream];
-
-	qm_ci_stream_off = mmDCORE0_TPC0_QM_PQ_CI_1 - mmDCORE0_TPC0_QM_PQ_CI_0;
-	pq_ci = qman_base + (mmDCORE0_TPC0_QM_PQ_CI_0 - mmDCORE0_TPC0_QM_BASE) +
-						stream * qm_ci_stream_off;
-
-	hdev->asic_funcs->hw_queues_lock(hdev);
-
-	if (pr_sw_conf)
-		gaudi2_print_sw_config_stream_data(hdev, stream, qman_base);
-
-	ci = RREG32(pq_ci);
-
-	/* we should start printing form ci -1 */
-	ci = gaudi2_queue_idx_dec(ci, HL_QUEUE_LENGTH);
-
-	for (i = 0; i < PQ_FETCHER_CACHE_SIZE; i++) {
-		struct hl_bd *bd;
-		u64 addr;
-		u32 len;
-
-		bd = q->kernel_address;
-		bd += ci;
-
-		len = le32_to_cpu(bd->len);
-		/* len 0 means uninitialized entry- break */
-		if (!len)
-			break;
-
-		addr = le64_to_cpu(bd->ptr);
-
-		dev_info(hdev->dev, "stop on err PQE(stream %u): ci: %u, addr: %#llx, size: %x\n",
-			stream, ci, addr, len);
-
-		/* get previous ci, wrap if needed */
-		ci = gaudi2_queue_idx_dec(ci, HL_QUEUE_LENGTH);
-	}
-
-	hdev->asic_funcs->hw_queues_unlock(hdev);
-}
-
-/**
- * print_qman_data_on_err - extract QMAN data on error
- *
- * @hdev: pointer to the habanalabs device structure
- * @qid_base: first QID of the QMAN (out of 4 streams)
- * @stream: the QMAN's stream
- * @qman_base: base address of QMAN registers block
- *
- * This function attempt to extract as much data as possible on QMAN error.
- * On upper CP print the SW config stream data and last 8 PQEs.
- * On lower CP print SW config data and last PQEs of ALL 4 upper CPs
- */
-static void print_qman_data_on_err(struct hl_device *hdev, u32 qid_base, u32 stream, u64 qman_base)
-{
-	u32 i;
-
-	if (stream != QMAN_STREAMS) {
-		gaudi2_print_last_pqes_on_err(hdev, qid_base, stream, qman_base, true);
-		return;
-	}
+	lo = RREG32(qman_base + QM_ARC_CQ_PTR_LO_OFFSET);
+	hi = RREG32(qman_base + QM_ARC_CQ_PTR_HI_OFFSET);
+	arc_cq_ptr = ((u64) hi) << 32 | lo;
+	arc_cq_ptr_size = RREG32(qman_base + QM_ARC_CQ_TSIZE_OFFSET);
 
-	gaudi2_print_sw_config_stream_data(hdev, stream, qman_base);
+	lo = RREG32(qman_base + QM_CP_CURRENT_INST_LO_4_OFFSET);
+	hi = RREG32(qman_base + QM_CP_CURRENT_INST_HI_4_OFFSET);
+	cp_current_inst = ((u64) hi) << 32 | lo;
 
-	for (i = 0 ; i < QMAN_STREAMS ; i++)
-		gaudi2_print_last_pqes_on_err(hdev, qid_base, i, qman_base, false);
+	dev_info(hdev->dev,
+		"LowerQM. CQ: {ptr %#llx, size %u}, ARC_CQ: {ptr %#llx, size %u}, CP: {instruction %#llx}\n",
+		cq_ptr, cq_ptr_size, arc_cq_ptr, arc_cq_ptr_size, cp_current_inst);
 }
 
 static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type,
@@ -7912,7 +7803,8 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type
 				error_count++;
 			}
 
-		print_qman_data_on_err(hdev, qid_base, i, qman_base);
+		if (i == QMAN_STREAMS)
+			print_lower_qman_data_on_err(hdev, qman_base);
 	}
 
 	arb_err_val = RREG32(arb_err_addr);
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2P.h b/drivers/accel/habanalabs/gaudi2/gaudi2P.h
index 1cebe707772e7..5f3ce086928e3 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2P.h
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2P.h
@@ -98,7 +98,7 @@
 #define GAUDI2_DEFAULT_CARD_NAME		"HL225"
 
 #define QMAN_STREAMS				4
-#define PQ_FETCHER_CACHE_SIZE			8
+
 #define NUM_OF_MME_SBTE_PORTS			5
 #define NUM_OF_MME_WB_PORTS			2
 
diff --git a/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h b/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h
index 6c58af6142364..a08378d0802b5 100644
--- a/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h
+++ b/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h
@@ -242,6 +242,17 @@
 #define QM_FENCE2_OFFSET		(mmPDMA0_QM_CP_FENCE2_RDATA_0 - mmPDMA0_QM_BASE)
 #define QM_SEI_STATUS_OFFSET		(mmPDMA0_QM_SEI_STATUS - mmPDMA0_QM_BASE)
 
+#define QM_CQ_PTR_LO_4_OFFSET		(mmPDMA0_QM_CQ_PTR_LO_4 - mmPDMA0_QM_BASE)
+#define QM_CQ_PTR_HI_4_OFFSET		(mmPDMA0_QM_CQ_PTR_HI_4 - mmPDMA0_QM_BASE)
+#define QM_CQ_TSIZE_4_OFFSET		(mmPDMA0_QM_CQ_TSIZE_4 - mmPDMA0_QM_BASE)
+
+#define QM_ARC_CQ_PTR_LO_OFFSET		(mmPDMA0_QM_ARC_CQ_PTR_LO - mmPDMA0_QM_BASE)
+#define QM_ARC_CQ_PTR_HI_OFFSET		(mmPDMA0_QM_ARC_CQ_PTR_HI - mmPDMA0_QM_BASE)
+#define QM_ARC_CQ_TSIZE_OFFSET		(mmPDMA0_QM_ARC_CQ_TSIZE - mmPDMA0_QM_BASE)
+
+#define QM_CP_CURRENT_INST_LO_4_OFFSET	(mmPDMA0_QM_CP_CURRENT_INST_LO_4 - mmPDMA0_QM_BASE)
+#define QM_CP_CURRENT_INST_HI_4_OFFSET	(mmPDMA0_QM_CP_CURRENT_INST_HI_4 - mmPDMA0_QM_BASE)
+
 #define SFT_OFFSET		(mmSFT1_HBW_RTR_IF0_RTR_H3_BASE - mmSFT0_HBW_RTR_IF0_RTR_H3_BASE)
 #define SFT_IF_RTR_OFFSET	(mmSFT0_HBW_RTR_IF1_RTR_H3_BASE - mmSFT0_HBW_RTR_IF0_RTR_H3_BASE)
 
-- 
GitLab


From 7e63f317c0c3eb72e745ce3509299800ef54d6de Mon Sep 17 00:00:00 2001
From: Koby Elbaz <kelbaz@habana.ai>
Date: Wed, 10 May 2023 10:15:22 +0300
Subject: [PATCH 0896/1791] accel/habanalabs: update state when loading boot
 fit

Any FW component we load must be followed by a corresponding state
update. However, it seems that so far we skipped doing so for the
bootfit case, so fix that.

Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/firmware_if.c | 25 ++++++++-----------
 1 file changed, 10 insertions(+), 15 deletions(-)

diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c
index eb51d7f70aece..acbc1a6b5cb18 100644
--- a/drivers/accel/habanalabs/common/firmware_if.c
+++ b/drivers/accel/habanalabs/common/firmware_if.c
@@ -2486,16 +2486,6 @@ static int hl_fw_dynamic_load_image(struct hl_device *hdev,
 	if (rc)
 		goto release_fw;
 
-	/* update state according to boot stage */
-	if (cur_fwc == FW_COMP_BOOT_FIT) {
-		struct cpu_dyn_regs *dyn_regs;
-
-		dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs;
-		hl_fw_boot_fit_update_state(hdev,
-				le32_to_cpu(dyn_regs->cpu_boot_dev_sts0),
-				le32_to_cpu(dyn_regs->cpu_boot_dev_sts1));
-	}
-
 	/* copy boot fit to space allocated by FW */
 	rc = hl_fw_dynamic_copy_image(hdev, fw, fw_loader);
 	if (rc)
@@ -2798,6 +2788,14 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev,
 		goto protocol_err;
 	}
 
+	rc = hl_fw_dynamic_wait_for_boot_fit_active(hdev, fw_loader);
+	if (rc)
+		goto protocol_err;
+
+	hl_fw_boot_fit_update_state(hdev,
+			le32_to_cpu(dyn_regs->cpu_boot_dev_sts0),
+			le32_to_cpu(dyn_regs->cpu_boot_dev_sts1));
+
 	/*
 	 * when testing FW load (without Linux) on PLDM we don't want to
 	 * wait until boot fit is active as it may take several hours.
@@ -2807,10 +2805,6 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev,
 	if (hdev->pldm && !(hdev->fw_components & FW_TYPE_LINUX))
 		return 0;
 
-	rc = hl_fw_dynamic_wait_for_boot_fit_active(hdev, fw_loader);
-	if (rc)
-		goto protocol_err;
-
 	/* Enable DRAM scrambling before Linux boot and after successful
 	 *  UBoot
 	 */
@@ -2844,7 +2838,8 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev,
 	if (rc)
 		goto protocol_err;
 
-	hl_fw_linux_update_state(hdev, le32_to_cpu(dyn_regs->cpu_boot_dev_sts0),
+	hl_fw_linux_update_state(hdev,
+				le32_to_cpu(dyn_regs->cpu_boot_dev_sts0),
 				le32_to_cpu(dyn_regs->cpu_boot_dev_sts1));
 
 	hl_fw_dynamic_update_linux_interrupt_if(hdev);
-- 
GitLab


From 5d658d0c5137f709e9888f2f3687afbde38b5ac0 Mon Sep 17 00:00:00 2001
From: Dani Liberman <dliberman@habana.ai>
Date: Mon, 8 May 2023 11:06:17 +0300
Subject: [PATCH 0897/1791] accel/habanalabs: mask part of hmmu page fault
 captured address

When receiving page fault from hmmu, the captured address is scrambled
both by HW and by driver. The driver part is unscrambled but the HW
part isn't getting unscrambled.
To avoid declaring wrong address, the HW scrambled part will be
masked.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 4981b8eb0ff50..1cb2b72e1cd21 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -162,6 +162,9 @@
 #define PSOC_RAZWI_ENG_STR_SIZE 128
 #define PSOC_RAZWI_MAX_ENG_PER_RTR 5
 
+/* HW scrambles only bits 0-25 */
+#define HW_UNSCRAMBLED_BITS_MASK GENMASK_ULL(63, 26)
+
 struct gaudi2_razwi_info {
 	u32 axuser_xy;
 	u32 rtr_ctrl;
@@ -8835,11 +8838,16 @@ static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool
 	addr <<= 32;
 	addr |= RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_ERROR_CAPTURE_VA));
 
-	if (!is_pmmu)
+	if (is_pmmu) {
+		dev_err_ratelimited(hdev->dev, "PMMU page fault on va 0x%llx\n", addr);
+	} else {
+
 		addr = gaudi2_mmu_descramble_addr(hdev, addr);
+		addr &= HW_UNSCRAMBLED_BITS_MASK;
+		dev_err_ratelimited(hdev->dev, "HMMU page fault on va range 0x%llx - 0x%llx\n",
+				addr, addr + ~HW_UNSCRAMBLED_BITS_MASK);
+	}
 
-	dev_err_ratelimited(hdev->dev, "%s page fault on va 0x%llx\n",
-				is_pmmu ? "PMMU" : "HMMU", addr);
 	hl_handle_page_fault(hdev, addr, 0, is_pmmu, event_mask);
 
 	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_PAGE_ERROR_VALID), 0);
-- 
GitLab


From 569210233a31fa23932b269779144252ce7bfee9 Mon Sep 17 00:00:00 2001
From: Oded Gabbay <ogabbay@kernel.org>
Date: Mon, 22 May 2023 14:09:21 +0300
Subject: [PATCH 0898/1791] accel/habanalabs: remove sim code

There were a few places where simulator only code got into the upstream.
Remove those places that can confuse other developers.

Fixes: 2a0a839b6a28 ("habanalabs: extend fatal messages to contain PCI info")
Cc: Moti Haimovski <mhaimovski@habana.ai>
Cc: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/device.c     | 10 +++------
 drivers/accel/habanalabs/common/habanalabs.h | 23 ++++----------------
 2 files changed, 7 insertions(+), 26 deletions(-)

diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index cab5a63db8c16..ea02f2cfdf81e 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -2328,13 +2328,9 @@ int hl_device_init(struct hl_device *hdev)
 	hdev->disabled = true;
 	if (expose_interfaces_on_err)
 		cdev_sysfs_debugfs_add(hdev);
-	if (hdev->pdev)
-		dev_err(&hdev->pdev->dev,
-			"Failed to initialize hl%d. Device %s is NOT usable !\n",
-			hdev->cdev_idx, dev_name(&(hdev)->pdev->dev));
-	else
-		pr_err("Failed to initialize hl%d. Device %s is NOT usable !\n",
-			hdev->cdev_idx, dev_name(&(hdev)->pdev->dev));
+	dev_err(&hdev->pdev->dev,
+		"Failed to initialize hl%d. Device %s is NOT usable !\n",
+		hdev->cdev_idx, dev_name(&hdev->pdev->dev));
 
 	return rc;
 }
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index e2341a75a4b71..c5aa33eaa826b 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -2558,12 +2558,7 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val);
 	ktime_t __timeout; \
 	u32 __elbi_read; \
 	int __rc = 0; \
-	if (hdev->pdev) \
-		__timeout = ktime_add_us(ktime_get(), timeout_us); \
-	else \
-		__timeout = ktime_add_us(ktime_get(),\
-				min((u64)(timeout_us * 10), \
-					(u64) HL_SIM_MAX_TIMEOUT_US)); \
+	__timeout = ktime_add_us(ktime_get(), timeout_us); \
 	might_sleep_if(sleep_us); \
 	for (;;) { \
 		if (elbi) { \
@@ -2615,13 +2610,7 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val);
 	u8 __arr_idx;	\
 	int __rc = 0; \
 	\
-	if (hdev->pdev) \
-		__timeout = ktime_add_us(ktime_get(), timeout_us); \
-	else \
-		__timeout = ktime_add_us(ktime_get(),\
-				min(((u64)timeout_us * 10), \
-					(u64) HL_SIM_MAX_TIMEOUT_US)); \
-	\
+	__timeout = ktime_add_us(ktime_get(), timeout_us); \
 	might_sleep_if(sleep_us); \
 	if (arr_size >= 64) \
 		__rc = -EINVAL; \
@@ -2679,12 +2668,8 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val);
 				mem_written_by_device) \
 ({ \
 	ktime_t __timeout; \
-	if (hdev->pdev) \
-		__timeout = ktime_add_us(ktime_get(), timeout_us); \
-	else \
-		__timeout = ktime_add_us(ktime_get(),\
-				min((u64)(timeout_us * 100), \
-					(u64) HL_SIM_MAX_TIMEOUT_US)); \
+	\
+	__timeout = ktime_add_us(ktime_get(), timeout_us); \
 	might_sleep_if(sleep_us); \
 	for (;;) { \
 		/* Verify we read updates done by other cores or by device */ \
-- 
GitLab


From 92a3a9b1a3997cf67aca7874e7f61b102ea2f27c Mon Sep 17 00:00:00 2001
From: Dani Liberman <dliberman@habana.ai>
Date: Thu, 20 Apr 2023 19:33:34 +0300
Subject: [PATCH 0899/1791] accel/habanalabs: add description to several info
 ioctls

Several info ioctls may return success although no data retrieved.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 include/uapi/drm/habanalabs_accel.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/include/uapi/drm/habanalabs_accel.h b/include/uapi/drm/habanalabs_accel.h
index d9ef1b151d048..e6436f3e8ea60 100644
--- a/include/uapi/drm/habanalabs_accel.h
+++ b/include/uapi/drm/habanalabs_accel.h
@@ -787,18 +787,28 @@ enum hl_server_type {
  *                            The address which accessing it caused the razwi.
  *                            Razwi initiator.
  *                            Razwi cause, was it a page fault or MMU access error.
+ *                            May return 0 even though no new data is available, in that case
+ *                            timestamp will be 0.
  * HL_INFO_DEV_MEM_ALLOC_PAGE_SIZES - Retrieve valid page sizes for device memory allocation
  * HL_INFO_SECURED_ATTESTATION - Retrieve attestation report of the boot.
  * HL_INFO_REGISTER_EVENTFD   - Register eventfd for event notifications.
  * HL_INFO_UNREGISTER_EVENTFD - Unregister eventfd
  * HL_INFO_GET_EVENTS         - Retrieve the last occurred events
  * HL_INFO_UNDEFINED_OPCODE_EVENT - Retrieve last undefined opcode error information.
+ *                                  May return 0 even though no new data is available, in that case
+ *                                  timestamp will be 0.
  * HL_INFO_ENGINE_STATUS - Retrieve the status of all the h/w engines in the asic.
  * HL_INFO_PAGE_FAULT_EVENT - Retrieve parameters of captured page fault.
+ *                            May return 0 even though no new data is available, in that case
+ *                            timestamp will be 0.
  * HL_INFO_USER_MAPPINGS - Retrieve user mappings, captured after page fault event.
  * HL_INFO_FW_GENERIC_REQ - Send generic request to FW.
  * HL_INFO_HW_ERR_EVENT   - Retrieve information on the reported HW error.
+ *                          May return 0 even though no new data is available, in that case
+ *                          timestamp will be 0.
  * HL_INFO_FW_ERR_EVENT   - Retrieve information on the reported FW error.
+ *                          May return 0 even though no new data is available, in that case
+ *                          timestamp will be 0.
  */
 #define HL_INFO_HW_IP_INFO			0
 #define HL_INFO_HW_EVENTS			1
-- 
GitLab


From 8a20b381644528794a78f791fd1a7ae87f3fc6ff Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Wed, 17 May 2023 20:47:44 +0300
Subject: [PATCH 0900/1791] accel/habanalabs: fix bug of not fetching addr_dec
 info

addr_dec info should always be fetched, regardless of cause value.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 1cb2b72e1cd21..0d41adf4792cd 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -8769,6 +8769,9 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_typ
 	u32 error_count = 0;
 	int i;
 
+	gaudi2_print_event(hdev, event_type, true,
+		"intr_cause_data: %#llx", intr_cause_data);
+
 	for (i = 0 ; i < GAUDI2_NUM_OF_PCIE_ADDR_DEC_ERR_CAUSE ; i++) {
 		if (!(intr_cause_data & BIT_ULL(i)))
 			continue;
@@ -8781,10 +8784,11 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_typ
 		 * Always check for LBW and HBW additional info as the indication itself is
 		 * sometimes missing
 		 */
-		hl_check_for_glbl_errors(hdev);
-		gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask);
 	}
 
+	hl_check_for_glbl_errors(hdev);
+	gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask);
+
 	return error_count;
 }
 
-- 
GitLab


From ff5c702522bb01d8d28da98b508168b391600080 Mon Sep 17 00:00:00 2001
From: Oded Gabbay <ogabbay@kernel.org>
Date: Thu, 18 May 2023 10:59:38 +0300
Subject: [PATCH 0901/1791] accel/habanalabs: move ioctl error print to debug
 level

We don't want to allow users to spam the kernel log and sending
ioctls with bad opcodes is a sure way to do it.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Ofir Bitton <obitton@habana.ai>
---
 drivers/accel/habanalabs/common/habanalabs_ioctl.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
index 9a8be9395fb28..6a45a92344e9b 100644
--- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
@@ -1195,7 +1195,7 @@ static long _hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg,
 
 out_err:
 	if (retcode)
-		dev_dbg(dev, "error in ioctl: pid=%d, cmd=0x%02x, nr=0x%02x\n",
+		dev_dbg_ratelimited(dev, "error in ioctl: pid=%d, cmd=0x%02x, nr=0x%02x\n",
 			  task_pid_nr(current), cmd, nr);
 
 	if (kdata != stack_kdata)
@@ -1219,7 +1219,7 @@ long hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 	if ((nr >= HL_COMMAND_START) && (nr < HL_COMMAND_END)) {
 		ioctl = &hl_ioctls[nr];
 	} else {
-		dev_err(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n",
+		dev_dbg_ratelimited(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n",
 			task_pid_nr(current), nr);
 		return -ENOTTY;
 	}
@@ -1242,7 +1242,7 @@ long hl_ioctl_control(struct file *filep, unsigned int cmd, unsigned long arg)
 	if (nr == _IOC_NR(HL_IOCTL_INFO)) {
 		ioctl = &hl_ioctls_control[nr];
 	} else {
-		dev_err(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n",
+		dev_dbg_ratelimited(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n",
 			task_pid_nr(current), nr);
 		return -ENOTTY;
 	}
-- 
GitLab


From 19aa21b9807ad277972f7ebe8a8dd9fbf9063bc8 Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Mon, 22 May 2023 08:52:46 +0300
Subject: [PATCH 0902/1791] accel/habanalabs: unsecure TSB_CFG_MTRR regs

In order to utilize Engine Barrier padding, user must have access to
this register set.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/gaudi2/gaudi2_security.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
index fadb870ff4c03..2742b1f801eb2 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c
@@ -1534,6 +1534,10 @@ static const u32 gaudi2_pb_dcr0_tpc0_unsecured_regs[] = {
 	mmDCORE0_TPC0_CFG_QM_KERNEL_CONFIG,
 	mmDCORE0_TPC0_CFG_QM_KERNEL_ID,
 	mmDCORE0_TPC0_CFG_QM_POWER_LOOP,
+	mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_0,
+	mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_1,
+	mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_2,
+	mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_3,
 	mmDCORE0_TPC0_CFG_LUT_FUNC32_BASE2_ADDR_LO,
 	mmDCORE0_TPC0_CFG_LUT_FUNC32_BASE2_ADDR_HI,
 	mmDCORE0_TPC0_CFG_LUT_FUNC64_BASE2_ADDR_LO,
-- 
GitLab


From fac91dd54f3b6f5dbd20c1cb26e59eceb128ca42 Mon Sep 17 00:00:00 2001
From: Ofir Bitton <obitton@habana.ai>
Date: Sun, 21 May 2023 10:24:13 +0300
Subject: [PATCH 0903/1791] accel/habanalabs: add event queue extra validation

In order to increase reliability of the event queue interface,
we apply to Gaudi2 the same mechanism we have in Gaudi1.
The extra validation is basically checking that the received
event index matches the expected index.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/irq.c    | 2 +-
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/common/irq.c b/drivers/accel/habanalabs/common/irq.c
index c67895b1cdeb1..b1010d206c2ef 100644
--- a/drivers/accel/habanalabs/common/irq.c
+++ b/drivers/accel/habanalabs/common/irq.c
@@ -430,7 +430,7 @@ irqreturn_t hl_irq_handler_eq(int irq, void *arg)
 		cur_eqe_index = FIELD_GET(EQ_CTL_INDEX_MASK, cur_eqe);
 		if ((hdev->event_queue.check_eqe_index) &&
 				(((eq->prev_eqe_index + 1) & EQ_CTL_INDEX_MASK) != cur_eqe_index)) {
-			dev_dbg(hdev->dev,
+			dev_err(hdev->dev,
 				"EQE %#x in queue is ready but index does not match %d!=%d",
 				cur_eqe,
 				((eq->prev_eqe_index + 1) & EQ_CTL_INDEX_MASK),
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 0d41adf4792cd..20c4583f12b0d 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -3619,6 +3619,12 @@ static int gaudi2_sw_init(struct hl_device *hdev)
 
 	prop->supports_compute_reset = true;
 
+	/* Event queue sanity check added in FW version 1.11 */
+	if (hl_is_fw_sw_ver_below(hdev, 1, 11))
+		hdev->event_queue.check_eqe_index = false;
+	else
+		hdev->event_queue.check_eqe_index = true;
+
 	hdev->asic_funcs->set_pci_memory_regions(hdev);
 
 	rc = gaudi2_special_blocks_iterator_config(hdev);
-- 
GitLab


From e6f49e96bc57d34fc0f617f37bfdf62a9b58d2c2 Mon Sep 17 00:00:00 2001
From: Dani Liberman <dliberman@habana.ai>
Date: Mon, 22 May 2023 17:15:36 +0300
Subject: [PATCH 0904/1791] accel/habanalabs: refactor error info reset

Moved error info reset code to single function for future use from
other places in the driver.

Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/accel/habanalabs/common/device.c         | 8 ++++++++
 drivers/accel/habanalabs/common/habanalabs.h     | 1 +
 drivers/accel/habanalabs/common/habanalabs_drv.c | 5 +----
 3 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index ea02f2cfdf81e..b97339d1f7c6e 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -2689,3 +2689,11 @@ void hl_handle_fw_err(struct hl_device *hdev, struct hl_info_fw_err_info *info)
 	if (info->event_mask)
 		*info->event_mask |= HL_NOTIFIER_EVENT_CRITICL_FW_ERR;
 }
+
+void hl_enable_err_info_capture(struct hl_error_info *captured_err_info)
+{
+	vfree(captured_err_info->page_fault_info.user_mappings);
+	memset(captured_err_info, 0, sizeof(struct hl_error_info));
+	atomic_set(&captured_err_info->cs_timeout.write_enable, 1);
+	captured_err_info->undef_opcode.write_enable = true;
+}
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index c5aa33eaa826b..d92ba2e30e310 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3944,6 +3944,7 @@ void hl_handle_page_fault(struct hl_device *hdev, u64 addr, u16 eng_id, bool is_
 				u64 *event_mask);
 void hl_handle_critical_hw_err(struct hl_device *hdev, u16 event_id, u64 *event_mask);
 void hl_handle_fw_err(struct hl_device *hdev, struct hl_info_fw_err_info *info);
+void hl_enable_err_info_capture(struct hl_error_info *captured_err_info);
 
 #ifdef CONFIG_DEBUG_FS
 
diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c
index 446f444a1c7e4..7263e84c1a4dc 100644
--- a/drivers/accel/habanalabs/common/habanalabs_drv.c
+++ b/drivers/accel/habanalabs/common/habanalabs_drv.c
@@ -219,10 +219,7 @@ int hl_device_open(struct inode *inode, struct file *filp)
 
 	hl_debugfs_add_file(hpriv);
 
-	vfree(hdev->captured_err_info.page_fault_info.user_mappings);
-	memset(&hdev->captured_err_info, 0, sizeof(hdev->captured_err_info));
-	atomic_set(&hdev->captured_err_info.cs_timeout.write_enable, 1);
-	hdev->captured_err_info.undef_opcode.write_enable = true;
+	hl_enable_err_info_capture(&hdev->captured_err_info);
 
 	hdev->open_counter++;
 	hdev->last_successful_open_jif = jiffies;
-- 
GitLab


From e94c25567c8519b770985cc86b652b06dce807d2 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Thu, 20 Apr 2023 14:07:05 -0400
Subject: [PATCH 0905/1791] drm/amd/display: remove unused variables
 dispclk_delay_subtotal and dout

clang with W=1 reports
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn314/display_rq_dlg_calc_314.c:1003:15:
  error: variable 'dispclk_delay_subtotal' set but not used [-Werror,-Wunused-but-set-variable]
        unsigned int dispclk_delay_subtotal;
                     ^
This variable is not used, so remove it.
Which made dout unused, so also remove.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dml/dcn314/display_rq_dlg_calc_314.c    | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
index 4f945458b2b7e..61ba3e33bb11d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
@@ -951,7 +951,6 @@ static void dml_rq_dlg_get_dlg_params(
 {
 	const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src;
 	const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest;
-	const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout;
 	const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg;
 	const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth;
 	const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps;
@@ -1000,8 +999,6 @@ static void dml_rq_dlg_get_dlg_params(
 	unsigned int vupdate_width;
 	unsigned int vready_offset;
 
-	unsigned int dispclk_delay_subtotal;
-
 	unsigned int vstartup_start;
 	unsigned int dst_x_after_scaler;
 	unsigned int dst_y_after_scaler;
@@ -1119,13 +1116,6 @@ static void dml_rq_dlg_get_dlg_params(
 	vupdate_offset = dst->vupdate_offset;
 	vupdate_width = dst->vupdate_width;
 	vready_offset = dst->vready_offset;
-	dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal;
-
-	if (dout->dsc_enable) {
-		double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // FROM VBA
-
-		dispclk_delay_subtotal += dsc_delay;
-	}
 
 	vstartup_start = dst->vstartup_start;
 	if (interlaced) {
-- 
GitLab


From 609d830048fb00d4fdea59fc9d17a8d63fbddb4a Mon Sep 17 00:00:00 2001
From: Sukrut Bellary <sukrut.bellary@linux.com>
Date: Sun, 23 Apr 2023 22:59:10 -0700
Subject: [PATCH 0906/1791] drm:amd:amdgpu: Fix missing bo unlock in failure
 path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

smatch warning - inconsistent handling of buffer object reserve
and unreserve.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Sukrut Bellary <sukrut.bellary@linux.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 278416acf0608..5de44d7e92de0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4686,8 +4686,10 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
 		return r;
 
 	r = amdgpu_bo_kmap(ring->mqd_obj, &ring->mqd_ptr);
-	if (unlikely(r != 0))
+	if (unlikely(r != 0)) {
+		amdgpu_bo_unreserve(ring->mqd_obj);
 		return r;
+	}
 
 	gfx_v8_0_kiq_init_queue(ring);
 	amdgpu_bo_kunmap(ring->mqd_obj);
-- 
GitLab


From 960e27a5741cd3001996ff6ddfb3eb0ed3a4909d Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Mon, 17 Apr 2023 23:35:08 +0200
Subject: [PATCH 0907/1791] drm/amd/display: Fix a test
 CalculatePrefetchSchedule()

It is likely Height was expected here, instead of Width.

Test the correct variable.

Fixes: 17529ea2acfa ("drm/amd/display: Optimizations for DML math")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
index b7c2844d0cbee..f294f2f8c75bc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c
@@ -810,7 +810,7 @@ static bool CalculatePrefetchSchedule(
 			*swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockWidth256BytesC) + myPipe->BlockWidth256BytesC;
 	} else {
 		*swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockHeight256BytesY) + myPipe->BlockHeight256BytesY;
-		if (myPipe->BlockWidth256BytesC > 0)
+		if (myPipe->BlockHeight256BytesC > 0)
 			*swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockHeight256BytesC) + myPipe->BlockHeight256BytesC;
 	}
 
-- 
GitLab


From bafc31166aa7df5fa26ae0ad8196d1717e6cdea9 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Mon, 17 Apr 2023 23:41:11 +0200
Subject: [PATCH 0908/1791] drm/amd/display: Fix a test
 dml32_rq_dlg_get_rq_reg()

It is likely p1_min_meta_chunk_bytes was expected here, instead of
min_meta_chunk_bytes.

Test the correct variable.

Fixes: dda4fb85e433 ("drm/amd/display: DML changes for DCN32/321")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c   | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
index 395ae8761980f..9ba6cb67655f4 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
@@ -116,7 +116,7 @@ void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs,
 	else
 		rq_regs->rq_regs_l.min_meta_chunk_size = dml_log2(min_meta_chunk_bytes) - 6 + 1;
 
-	if (min_meta_chunk_bytes == 0)
+	if (p1_min_meta_chunk_bytes == 0)
 		rq_regs->rq_regs_c.min_meta_chunk_size = 0;
 	else
 		rq_regs->rq_regs_c.min_meta_chunk_size = dml_log2(p1_min_meta_chunk_bytes) - 6 + 1;
-- 
GitLab


From 489763af891d5dc35c0b64e18af284d6591286cf Mon Sep 17 00:00:00 2001
From: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Date: Mon, 24 Apr 2023 19:25:32 +0200
Subject: [PATCH 0909/1791] drm/amdgpu: add new flag to AMDGPU_CTX_QUERY2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

OpenGL EXT_robustness extension expects the driver to stop reporting
GUILTY_CONTEXT_RESET when the reset has completed and the GPU is ready
to accept submission again.

This commit adds a AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS flag,
that let the UMD know that the reset is still not finished.

Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22290

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++-
 include/uapi/drm/amdgpu_drm.h           | 2 ++
 3 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index d2139ac121595..e1f642a3dc2fb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -576,6 +576,9 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev,
 	if (atomic_read(&ctx->guilty))
 		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_GUILTY;
 
+	if (amdgpu_in_reset(adev))
+		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS;
+
 	if (adev->ras_enabled && con) {
 		/* Return the cached values in O(1),
 		 * and schedule delayed work to cache
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b987022e11b4e..45544ebe576e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -111,9 +111,10 @@
  *            tcp_cache_size, num_sqc_per_wgp, sqc_data_cache_size, sqc_inst_cache_size,
  *            gl1c_cache_size, gl2c_cache_size, mall_size, enabled_rb_pipes_mask_hi
  *   3.53.0 - Support for GFX11 CP GFX shadowing
+ *   3.54.0 - Add AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS support
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	53
+#define KMS_DRIVER_MINOR	54
 #define KMS_DRIVER_PATCHLEVEL	0
 
 unsigned int amdgpu_vram_limit = UINT_MAX;
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index cc78528c3b4bc..79b14828d542a 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -245,6 +245,8 @@ union drm_amdgpu_bo_list {
 /* indicate some errors are detected by RAS */
 #define AMDGPU_CTX_QUERY2_FLAGS_RAS_CE   (1<<3)
 #define AMDGPU_CTX_QUERY2_FLAGS_RAS_UE   (1<<4)
+/* indicate that the reset hasn't completed yet */
+#define AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS (1<<5)
 
 /* Context priority level */
 #define AMDGPU_CTX_PRIORITY_UNSET       -2048
-- 
GitLab


From 535f77861084467eaf37bbbb4d4aaa96f5fe4c1c Mon Sep 17 00:00:00 2001
From: lyndonli <Lyndon.Li@amd.com>
Date: Sun, 23 Apr 2023 16:46:30 +0800
Subject: [PATCH 0910/1791] drm/amdgpu: Fix mode2 reset for sienna cichlid

Before this change, sienna_cichlid_get_reset_handler will always
return NULL, although the module parameter reset_method is 3
when loading amdgpu driver.

Signed-off-by: lyndonli <Lyndon.Li@amd.com>
Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
index 81a6d5b94987f..8b8086d5c864b 100644
--- a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
+++ b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c
@@ -40,7 +40,7 @@ static bool sienna_cichlid_is_mode2_default(struct amdgpu_reset_control *reset_c
 	    adev->pm.fw_version >= 0x3a5500 && !amdgpu_sriov_vf(adev))
 		return true;
 #endif
-	return false;
+	return amdgpu_reset_method == AMD_RESET_METHOD_MODE2;
 }
 
 static struct amdgpu_reset_handler *
-- 
GitLab


From 59e9fff1983eaa9f226699f66f7d58da432dfb16 Mon Sep 17 00:00:00 2001
From: lyndonli <Lyndon.Li@amd.com>
Date: Sun, 23 Apr 2023 17:05:15 +0800
Subject: [PATCH 0911/1791] drm/amdgpu: Use the default reset when loading or
 reloading the driver

Below call trace and errors are observed when reloading
amdgpu driver with the module parameter reset_method=3.

It should do a default reset when loading or reloading the
driver, regardless of the module parameter reset_method.

v2: add comments inside and modify commit messages.

[  +2.180243] [drm] psp gfx command ID_LOAD_TOC(0x20) failed
and response status is (0x0)
[  +0.000011] [drm:psp_hw_start [amdgpu]] *ERROR* Failed to load toc
[  +0.000890] [drm:psp_hw_start [amdgpu]] *ERROR* PSP tmr init failed!
[  +0.020683] [drm:amdgpu_fill_buffer [amdgpu]] *ERROR* Trying to
clear memory with ring turned off.
[  +0.000003] RIP: 0010:amdgpu_bo_release_notify+0x1ef/0x210 [amdgpu]
[  +0.000004] Call Trace:
[  +0.000003]  <TASK>
[  +0.000008]  ttm_bo_release+0x2c4/0x330 [amdttm]
[  +0.000026]  amdttm_bo_put+0x3c/0x70 [amdttm]
[  +0.000020]  amdgpu_bo_free_kernel+0xe6/0x140 [amdgpu]
[  +0.000728]  psp_v11_0_ring_destroy+0x34/0x60 [amdgpu]
[  +0.000826]  psp_hw_init+0xe7/0x2f0 [amdgpu]
[  +0.000813]  amdgpu_device_fw_loading+0x1ad/0x2d0 [amdgpu]
[  +0.000731]  amdgpu_device_init.cold+0x108e/0x2002 [amdgpu]
[  +0.001071]  ? do_pci_enable_device+0xe1/0x110
[  +0.000011]  amdgpu_driver_load_kms+0x1a/0x160 [amdgpu]
[  +0.000729]  amdgpu_pci_probe+0x179/0x3a0 [amdgpu]

Signed-off-by: lyndonli <Lyndon.Li@amd.com>
Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index ac78caa7cba87..1787602fe582a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3579,6 +3579,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	int r, i;
 	bool px = false;
 	u32 max_MBps;
+	int tmp;
 
 	adev->shutdown = false;
 	adev->flags = flags;
@@ -3800,7 +3801,13 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 				}
 			}
 		} else {
+			tmp = amdgpu_reset_method;
+			/* It should do a default reset when loading or reloading the driver,
+			 * regardless of the module parameter reset_method.
+			 */
+			amdgpu_reset_method = AMD_RESET_METHOD_NONE;
 			r = amdgpu_asic_reset(adev);
+			amdgpu_reset_method = tmp;
 			if (r) {
 				dev_err(adev->dev, "asic reset on init failed\n");
 				goto failed;
-- 
GitLab


From 27e6be5d753687764c852af6468c21ef0624022a Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 22 Apr 2023 06:50:14 +0530
Subject: [PATCH 0912/1791] drm/amd/display: DSC passthrough is for DP-HDMI
 pcon (SST pcon)

if check over DSC passthrough is removed, as this is not for
MST use case. It is for DP-HDMI pcon use case. sst pcon is
detected as sst not mst. In sst pcon dsc passthrough message
will not get below log printed

Fixes: 9b035d089086 ("drm/amd/display: Check & log if receiver supports MST, DSC & FEC.")
Suggested-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/protocols/link_dp_capability.c   | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 84265dc66bbab..577f74cc50fb4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1820,9 +1820,8 @@ static bool retrieve_link_cap(struct dc_link *link)
 				  str_yes_no(is_fec_supported));
 			DC_LOG_DC("%s: DSC_Basic_Sink_Support: %s\n", __func__,
 				  str_yes_no(is_dsc_basic_supported));
-			if (link->dpcd_caps.is_mst_capable)
-				DC_LOG_DC("%s: DSC_Passthrough_Sink_Support: %s\n", __func__,
-					  str_yes_no(is_dsc_passthrough_supported));
+			DC_LOG_DC("%s: DSC_Passthrough_Sink_Support: %s\n", __func__,
+				  str_yes_no(is_dsc_passthrough_supported));
 		}
 		if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) {
 			status = core_link_read_dpcd(
-- 
GitLab


From b2edaac4f2fb4866c3f9b7be5c39f518fd497a3b Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 26 Apr 2023 00:15:26 +0530
Subject: [PATCH 0913/1791] drm/amd/amdgpu: Fix style errors in
 amdgpu_display.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix following checkpatch errors in amdgpu_display.c

ERROR: spaces required around that '=' (ctx:VxW)
ERROR: that open brace { should be on the previous line
ERROR: else should follow close brace '}'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index d60fe7eb5579a..b702f499f5fb3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -98,7 +98,7 @@ static void amdgpu_display_flip_callback(struct dma_fence *f,
 static bool amdgpu_display_flip_handle_fence(struct amdgpu_flip_work *work,
 					     struct dma_fence **f)
 {
-	struct dma_fence *fence= *f;
+	struct dma_fence *fence = *f;
 
 	if (fence == NULL)
 		return false;
@@ -1252,21 +1252,21 @@ const struct drm_mode_config_funcs amdgpu_mode_funcs = {
 	.fb_create = amdgpu_display_user_framebuffer_create,
 };
 
-static const struct drm_prop_enum_list amdgpu_underscan_enum_list[] =
-{	{ UNDERSCAN_OFF, "off" },
+static const struct drm_prop_enum_list amdgpu_underscan_enum_list[] = {
+	{ UNDERSCAN_OFF, "off" },
 	{ UNDERSCAN_ON, "on" },
 	{ UNDERSCAN_AUTO, "auto" },
 };
 
-static const struct drm_prop_enum_list amdgpu_audio_enum_list[] =
-{	{ AMDGPU_AUDIO_DISABLE, "off" },
+static const struct drm_prop_enum_list amdgpu_audio_enum_list[] = {
+	{ AMDGPU_AUDIO_DISABLE, "off" },
 	{ AMDGPU_AUDIO_ENABLE, "on" },
 	{ AMDGPU_AUDIO_AUTO, "auto" },
 };
 
 /* XXX support different dither options? spatial, temporal, both, etc. */
-static const struct drm_prop_enum_list amdgpu_dither_enum_list[] =
-{	{ AMDGPU_FMT_DITHER_DISABLE, "off" },
+static const struct drm_prop_enum_list amdgpu_dither_enum_list[] = {
+	{ AMDGPU_FMT_DITHER_DISABLE, "off" },
 	{ AMDGPU_FMT_DITHER_ENABLE, "on" },
 };
 
@@ -1496,8 +1496,7 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev,
 		ret |= DRM_SCANOUTPOS_ACCURATE;
 		vbl_start = vbl & 0x1fff;
 		vbl_end = (vbl >> 16) & 0x1fff;
-	}
-	else {
+	} else {
 		/* No: Fake something reasonable which gives at least ok results. */
 		vbl_start = mode->crtc_vdisplay;
 		vbl_end = 0;
-- 
GitLab


From 1c312e816c4088d183c41b5d944f89775f3789bd Mon Sep 17 00:00:00 2001
From: Shane Xiao <shane.xiao@amd.com>
Date: Tue, 25 Apr 2023 22:39:08 +0800
Subject: [PATCH 0914/1791] drm/amdgpu: Enable doorbell selfring after resize
 FB BAR
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
The selfring doorbell aperture will change when resize FB
BAR successfully during gmc sw init, we should reorder
the sequence of enabling doorbell selfring aperture.

[How]
Move enable_doorbell_selfring_aperture from *_common_hw_init
to *_common_late_init.

This fixes the potential issue that GPU ring its own
doorbell when this device is in translated mode when
iommu is on.

v2: Remove *_enable_doorbell_aperture functions (Christian)
v3: Add comments to note that why we need enable doorbell
    selfring late (Christian)

Signed-off-by: Shane Xiao <shane.xiao@amd.com>
Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Tested-by: Xiaomeng Hou <Xiaomeng.Hou@amd.com>
Reviewed-by: Christian K�nig <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nv.c    | 23 +++++++++++++----------
 drivers/gpu/drm/amd/amdgpu/soc15.c | 25 +++++++++++++++----------
 drivers/gpu/drm/amd/amdgpu/soc21.c | 23 +++++++++++++----------
 3 files changed, 41 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index dabeeab2f2ad8..3cc068974bcdb 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -531,13 +531,6 @@ static void nv_program_aspm(struct amdgpu_device *adev)
 
 }
 
-static void nv_enable_doorbell_aperture(struct amdgpu_device *adev,
-					bool enable)
-{
-	adev->nbio.funcs->enable_doorbell_aperture(adev, enable);
-	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, enable);
-}
-
 const struct amdgpu_ip_block_version nv_common_ip_block =
 {
 	.type = AMD_IP_BLOCK_TYPE_COMMON,
@@ -999,6 +992,11 @@ static int nv_common_late_init(void *handle)
 		}
 	}
 
+	/* Enable selfring doorbell aperture late because doorbell BAR
+	 * aperture will change if resize BAR successfully in gmc sw_init.
+	 */
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, true);
+
 	return 0;
 }
 
@@ -1038,7 +1036,7 @@ static int nv_common_hw_init(void *handle)
 	if (adev->nbio.funcs->remap_hdp_registers && !amdgpu_sriov_vf(adev))
 		adev->nbio.funcs->remap_hdp_registers(adev);
 	/* enable the doorbell aperture */
-	nv_enable_doorbell_aperture(adev, true);
+	adev->nbio.funcs->enable_doorbell_aperture(adev, true);
 
 	return 0;
 }
@@ -1047,8 +1045,13 @@ static int nv_common_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	/* disable the doorbell aperture */
-	nv_enable_doorbell_aperture(adev, false);
+	/* Disable the doorbell aperture and selfring doorbell aperture
+	 * separately in hw_fini because nv_enable_doorbell_aperture
+	 * has been removed and there is no need to delay disabling
+	 * selfring doorbell.
+	 */
+	adev->nbio.funcs->enable_doorbell_aperture(adev, false);
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, false);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 4d1487a9836c6..3221522e71e89 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -619,13 +619,6 @@ static void soc15_program_aspm(struct amdgpu_device *adev)
 		adev->nbio.funcs->program_aspm(adev);
 }
 
-static void soc15_enable_doorbell_aperture(struct amdgpu_device *adev,
-					   bool enable)
-{
-	adev->nbio.funcs->enable_doorbell_aperture(adev, enable);
-	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, enable);
-}
-
 const struct amdgpu_ip_block_version vega10_common_ip_block =
 {
 	.type = AMD_IP_BLOCK_TYPE_COMMON,
@@ -1125,6 +1118,11 @@ static int soc15_common_late_init(void *handle)
 	if (amdgpu_sriov_vf(adev))
 		xgpu_ai_mailbox_get_irq(adev);
 
+	/* Enable selfring doorbell aperture late because doorbell BAR
+	 * aperture will change if resize BAR successfully in gmc sw_init.
+	 */
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, true);
+
 	return 0;
 }
 
@@ -1182,7 +1180,8 @@ static int soc15_common_hw_init(void *handle)
 		adev->nbio.funcs->remap_hdp_registers(adev);
 
 	/* enable the doorbell aperture */
-	soc15_enable_doorbell_aperture(adev, true);
+	adev->nbio.funcs->enable_doorbell_aperture(adev, true);
+
 	/* HW doorbell routing policy: doorbell writing not
 	 * in SDMA/IH/MM/ACV range will be routed to CP. So
 	 * we need to init SDMA doorbell range prior
@@ -1198,8 +1197,14 @@ static int soc15_common_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	/* disable the doorbell aperture */
-	soc15_enable_doorbell_aperture(adev, false);
+	/* Disable the doorbell aperture and selfring doorbell aperture
+	 * separately in hw_fini because soc15_enable_doorbell_aperture
+	 * has been removed and there is no need to delay disabling
+	 * selfring doorbell.
+	 */
+	adev->nbio.funcs->enable_doorbell_aperture(adev, false);
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, false);
+
 	if (amdgpu_sriov_vf(adev))
 		xgpu_ai_mailbox_put_irq(adev);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 7d59303ca2f95..0f82b8e83acbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -450,13 +450,6 @@ static void soc21_program_aspm(struct amdgpu_device *adev)
 		adev->nbio.funcs->program_aspm(adev);
 }
 
-static void soc21_enable_doorbell_aperture(struct amdgpu_device *adev,
-					bool enable)
-{
-	adev->nbio.funcs->enable_doorbell_aperture(adev, enable);
-	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, enable);
-}
-
 const struct amdgpu_ip_block_version soc21_common_ip_block =
 {
 	.type = AMD_IP_BLOCK_TYPE_COMMON,
@@ -764,6 +757,11 @@ static int soc21_common_late_init(void *handle)
 			amdgpu_irq_get(adev, &adev->nbio.ras_err_event_athub_irq, 0);
 	}
 
+	/* Enable selfring doorbell aperture late because doorbell BAR
+	 * aperture will change if resize BAR successfully in gmc sw_init.
+	 */
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, true);
+
 	return 0;
 }
 
@@ -797,7 +795,7 @@ static int soc21_common_hw_init(void *handle)
 	if (adev->nbio.funcs->remap_hdp_registers)
 		adev->nbio.funcs->remap_hdp_registers(adev);
 	/* enable the doorbell aperture */
-	soc21_enable_doorbell_aperture(adev, true);
+	adev->nbio.funcs->enable_doorbell_aperture(adev, true);
 
 	return 0;
 }
@@ -806,8 +804,13 @@ static int soc21_common_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	/* disable the doorbell aperture */
-	soc21_enable_doorbell_aperture(adev, false);
+	/* Disable the doorbell aperture and selfring doorbell aperture
+	 * separately in hw_fini because soc21_enable_doorbell_aperture
+	 * has been removed and there is no need to delay disabling
+	 * selfring doorbell.
+	 */
+	adev->nbio.funcs->enable_doorbell_aperture(adev, false);
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, false);
 
 	if (amdgpu_sriov_vf(adev)) {
 		xgpu_nv_mailbox_put_irq(adev);
-- 
GitLab


From 7e5b601008e781231be77bc8e1a84516f069983d Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Tue, 25 Apr 2023 13:16:32 +0800
Subject: [PATCH 0915/1791] drm/amdgpu: fix amdgpu_irq_put call trace in
 gmc_v11_0_hw_fini

The gmc.ecc_irq is enabled by firmware per IFWI setting,
and the host driver is not privileged to enable/disable
the interrupt. So, it is meaningless to use the amdgpu_irq_put
function in gmc_v11_0_hw_fini, which also leads to the call
trace.

[  102.980303] Call Trace:
[  102.980303]  <TASK>
[  102.980304]  gmc_v11_0_hw_fini+0x54/0x90 [amdgpu]
[  102.980357]  gmc_v11_0_suspend+0xe/0x20 [amdgpu]
[  102.980409]  amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu]
[  102.980459]  amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[  102.980520]  amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu]
[  102.980573]  amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu]
[  102.980687]  amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[  102.980740]  process_one_work+0x21f/0x3f0
[  102.980741]  worker_thread+0x200/0x3e0
[  102.980742]  ? process_one_work+0x3f0/0x3f0
[  102.980743]  kthread+0xfd/0x130
[  102.980743]  ? kthread_complete_and_exit+0x20/0x20
[  102.980744]  ret_from_fork+0x22/0x30

Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 3828ca95899fc..f73c238f3145a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -951,7 +951,6 @@ static int gmc_v11_0_hw_fini(void *handle)
 		return 0;
 	}
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 	gmc_v11_0_gart_disable(adev);
 
-- 
GitLab


From c953cf040687992cfa066acca7d1c12e25fcec3e Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Tue, 25 Apr 2023 10:52:28 +0800
Subject: [PATCH 0916/1791] drm/amdgpu: fix amdgpu_irq_put call trace in
 gmc_v10_0_hw_fini

The gmc.ecc_irq is enabled by firmware per IFWI setting,
and the host driver is not privileged to enable/disable
the interrupt. So, it is meaningless to use the amdgpu_irq_put
function in gmc_v10_0_hw_fini, which also leads to the call
trace.

[   82.340264] Call Trace:
[   82.340265]  <TASK>
[   82.340269]  gmc_v10_0_hw_fini+0x83/0xa0 [amdgpu]
[   82.340447]  gmc_v10_0_suspend+0xe/0x20 [amdgpu]
[   82.340623]  amdgpu_device_ip_suspend_phase2+0x127/0x1c0 [amdgpu]
[   82.340789]  amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[   82.340955]  amdgpu_device_pre_asic_reset+0xdd/0x2b0 [amdgpu]
[   82.341122]  amdgpu_device_gpu_recover.cold+0x4dd/0xbb2 [amdgpu]
[   82.341359]  amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[   82.341529]  process_one_work+0x21d/0x3f0
[   82.341535]  worker_thread+0x1fa/0x3c0
[   82.341538]  ? process_one_work+0x3f0/0x3f0
[   82.341540]  kthread+0xff/0x130
[   82.341544]  ? kthread_complete_and_exit+0x20/0x20
[   82.341547]  ret_from_fork+0x22/0x30

Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 23d4081eca006..5697b66bf0de1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -1143,7 +1143,6 @@ static int gmc_v10_0_hw_fini(void *handle)
 		return 0;
 	}
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 
 	return 0;
-- 
GitLab


From 4e004146c0bab0c05d66dc648593e8b7ec3d8df5 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Wed, 26 Apr 2023 07:18:38 -0400
Subject: [PATCH 0917/1791] drm/amd/display: set variable
 custom_backlight_curve0 storage-class-specifier to static

smatch reports
drivers/gpu/drm/amd/amdgpu/../display/modules/power/power_helpers.c:119:31:
  warning: symbol 'custom_backlight_curve0' was not declared. Should it be static?

This variable is only used in its defining file, so it should be static

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/modules/power/power_helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
index 68d95b92df76b..30349881a283a 100644
--- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
+++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c
@@ -116,7 +116,7 @@ static const struct abm_parameters * const abm_settings[] = {
 	abm_settings_config2,
 };
 
-const struct dm_bl_data_point custom_backlight_curve0[] = {
+static const struct dm_bl_data_point custom_backlight_curve0[] = {
 		{2, 14}, {4, 16}, {6, 18}, {8, 21}, {10, 23}, {12, 26}, {14, 29}, {16, 32}, {18, 35},
 		{20, 38}, {22, 41}, {24, 44}, {26, 48}, {28, 52}, {30, 55}, {32, 59}, {34, 62},
 		{36, 67}, {38, 71}, {40, 75}, {42, 80}, {44, 84}, {46, 88}, {48, 93}, {50, 98},
-- 
GitLab


From b94f1cc93db72078ad2da02adf1818f5e9122cb7 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Thu, 20 Apr 2023 09:59:01 -0400
Subject: [PATCH 0918/1791] drm/amd/display: return status of
 dmub_srv_get_fw_boot_status
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc with W=1 reports
drivers/gpu/drm/amd/amdgpu/../display/dc/dc_dmub_srv.c:
  In function ‘dc_dmub_srv_optimized_init_done’:
drivers/gpu/drm/amd/amdgpu/../display/dc/dc_dmub_srv.c:184:26:
  error: variable ‘dmub’ set but not used [-Werror=unused-but-set-variable]
  184 |         struct dmub_srv *dmub;
      |                          ^~~~

The return status is never set.
It looks like a call to dmub_srv_get_fw_boot_status is missing.

Fixes: 499e4b1c722e ("drm/amd/display: add mechanism to skip DCN init")
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index d15ec32243e22..36d936ab43005 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -182,14 +182,23 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun
 bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv)
 {
 	struct dmub_srv *dmub;
-	union dmub_fw_boot_status status;
+	struct dc_context *dc_ctx;
+	union dmub_fw_boot_status boot_status;
+	enum dmub_status status;
 
 	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
 		return false;
 
 	dmub = dc_dmub_srv->dmub;
+	dc_ctx = dc_dmub_srv->ctx;
+
+	status = dmub_srv_get_fw_boot_status(dmub, &boot_status);
+	if (status != DMUB_STATUS_OK) {
+		DC_ERROR("Error querying DMUB boot status: error=%d\n", status);
+		return false;
+	}
 
-	return status.bits.optimized_init_done;
+	return boot_status.bits.optimized_init_done;
 }
 
 bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv,
-- 
GitLab


From 7a4685cdfbdb94f1cf5ea2ddde824d94bf414708 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Wed, 26 Apr 2023 11:05:49 +0800
Subject: [PATCH 0919/1791] drm/amdgpu: fix a build warning by a typo in
 amdgpu_gfx.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This should be a typo when intruducing multi-xx support.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Cc: Le Ma <le.ma@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 60bb4bba19947..2cf1f88fde488 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -470,8 +470,8 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id)
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
 		j = i + xcc_id * adev->gfx.num_compute_rings;
-		ring = &adev->gfx.compute_ring[i];
-		kfree(adev->gfx.mec.mqd_backup[i]);
+		ring = &adev->gfx.compute_ring[j];
+		kfree(adev->gfx.mec.mqd_backup[j]);
 		amdgpu_bo_free_kernel(&ring->mqd_obj,
 				      &ring->mqd_gpu_addr,
 				      &ring->mqd_ptr);
-- 
GitLab


From 34305ac364dbee1b8e4e5307599c8b9afff0b01c Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Wed, 26 Apr 2023 11:27:46 +0800
Subject: [PATCH 0920/1791] drm/amdgpu: check correct allocated mqd_backup
 object after alloc

Instead of the default one, check the right mqd_backup object.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Cc: Le Ma <le.ma@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 2cf1f88fde488..66b9740ec3762 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -379,7 +379,7 @@ int amdgpu_gfx_kiq_init(struct amdgpu_device *adev,
 int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 			   unsigned mqd_size, int xcc_id)
 {
-	int r, i;
+	int r, i, j;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	struct amdgpu_ring *ring = &kiq->ring;
 
@@ -431,7 +431,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 
 	/* create MQD for each KCQ */
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-		ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings];
+		j = i + xcc_id * adev->gfx.num_compute_rings;
+		ring = &adev->gfx.compute_ring[j];
 		if (!ring->mqd_obj) {
 			r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
 						    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
@@ -443,8 +444,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 
 			ring->mqd_size = mqd_size;
 			/* prepare MQD backup */
-			adev->gfx.mec.mqd_backup[i + xcc_id * adev->gfx.num_compute_rings] = kmalloc(mqd_size, GFP_KERNEL);
-			if (!adev->gfx.mec.mqd_backup[i])
+			adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL);
+			if (!adev->gfx.mec.mqd_backup[j])
 				dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name);
 		}
 	}
-- 
GitLab


From 3c4f6507abde5c36e892d63b25296fc6d9b13285 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Thu, 27 Apr 2023 09:21:05 +0800
Subject: [PATCH 0921/1791] drm/amdgpu: mark gfx_v9_4_3_disable_gpa_mode()
 static

This was left global by accident, the corresponding functions for other hardware types are already static:

drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1072:6: error: no previous prototype for function 'gfx_v9_4_3_disable_gpa_mode' [-Werror,-Wmissing-prototypes]

Fixes: 86301129698b ("drm/amdgpu: split gc v9_4_3 functionality from gc v9_0")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 56a415e151d4e..312491455382a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1075,7 +1075,7 @@ static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
-- 
GitLab


From 1d74159021e9a4e58c422f0b91e2a6fcb884c54f Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 25 Apr 2023 14:11:56 -0400
Subject: [PATCH 0922/1791] drm/ttm: Helper function to get TTM mem limit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a helper function to get TTM memory limit. This is
needed by KFD to set its own internal memory limits.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/ttm/ttm_tt.c | 6 ++++++
 include/drm/ttm/ttm_tt.h     | 2 +-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index ab725d9d14a69..feac2f2c736e6 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -450,3 +450,9 @@ ttm_kmap_iter_tt_init(struct ttm_kmap_iter_tt *iter_tt,
 	return &iter_tt->base;
 }
 EXPORT_SYMBOL(ttm_kmap_iter_tt_init);
+
+unsigned long ttm_tt_pages_limit(void)
+{
+	return ttm_pages_limit;
+}
+EXPORT_SYMBOL(ttm_tt_pages_limit);
diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h
index 977ca195a5366..a4eff85b1f449 100644
--- a/include/drm/ttm/ttm_tt.h
+++ b/include/drm/ttm/ttm_tt.h
@@ -222,7 +222,7 @@ void ttm_tt_mgr_init(unsigned long num_pages, unsigned long num_dma32_pages);
 
 struct ttm_kmap_iter *ttm_kmap_iter_tt_init(struct ttm_kmap_iter_tt *iter_tt,
 					    struct ttm_tt *tt);
-
+unsigned long ttm_tt_pages_limit(void);
 #if IS_ENABLED(CONFIG_AGP)
 #include <linux/agp_backend.h>
 
-- 
GitLab


From f1f6f48a338cdab96efef712dbef6b1e279583e2 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 25 Apr 2023 14:13:36 -0400
Subject: [PATCH 0923/1791] drm/amdgpu: Set GTT size equal to TTM mem limit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use the helper function in TTM to get TTM mem limit and
set GTT size to be equal to TTL mem limit.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 25 ++++++-------------------
 2 files changed, 6 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 02b827785e399..957b18bda4a79 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -244,7 +244,6 @@ extern int amdgpu_sg_display;
 
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
-#define AMDGPU_DEFAULT_GTT_SIZE_MB		3072ULL /* 3GB by default */
 #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
 #define AMDGPU_MAX_USEC_TIMEOUT			100000	/* 100 ms */
 #define AMDGPU_FENCE_JIFFIES_TIMEOUT		(HZ / 2)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 2cd081cbf7062..f61f07575f63b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1803,26 +1803,13 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	DRM_INFO("amdgpu: %uM of VRAM memory ready\n",
 		 (unsigned) (adev->gmc.real_vram_size / (1024 * 1024)));
 
-	/* Compute GTT size, either based on 1/2 the size of RAM size
-	 * or whatever the user passed on module init */
-	if (amdgpu_gtt_size == -1) {
-		struct sysinfo si;
-
-		si_meminfo(&si);
-		/* Certain GL unit tests for large textures can cause problems
-		 * with the OOM killer since there is no way to link this memory
-		 * to a process.  This was originally mitigated (but not necessarily
-		 * eliminated) by limiting the GTT size.  The problem is this limit
-		 * is often too low for many modern games so just make the limit 1/2
-		 * of system memory which aligns with TTM. The OOM accounting needs
-		 * to be addressed, but we shouldn't prevent common 3D applications
-		 * from being usable just to potentially mitigate that corner case.
-		 */
-		gtt_size = max((AMDGPU_DEFAULT_GTT_SIZE_MB << 20),
-			       (u64)si.totalram * si.mem_unit / 2);
-	} else {
+	/* Compute GTT size, either based on TTM limit
+	 * or whatever the user passed on module init.
+	 */
+	if (amdgpu_gtt_size == -1)
+		gtt_size = ttm_tt_pages_limit() << PAGE_SHIFT;
+	else
 		gtt_size = (uint64_t)amdgpu_gtt_size << 20;
-	}
 
 	/* Initialize GTT memory pool */
 	r = amdgpu_gtt_mgr_init(adev, gtt_size);
-- 
GitLab


From 27fb73a0e3aa7478bcb5d2d59d65eec3c68fc165 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 25 Apr 2023 14:19:05 -0400
Subject: [PATCH 0924/1791] drm/amdkfd: Update KFD TTM mem limit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use the helper function in TTM to get TTM memory
limit and set KFD's internal mem limit. This ensures
that KFD's TTM mem limit and actual TTM mem limit are
exactly same.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c       | 3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 6 +++++-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index fed8bb9a721f7..a46285841d175 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -53,7 +53,6 @@ int amdgpu_amdkfd_init(void)
 	amdgpu_amdkfd_total_mem_size *= si.mem_unit;
 
 	ret = kgd2kfd_init();
-	amdgpu_amdkfd_gpuvm_init_mem_limits();
 	kfd_initialized = !ret;
 
 	return ret;
@@ -143,6 +142,8 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 	int i;
 	int last_valid_bit;
 
+	amdgpu_amdkfd_gpuvm_init_mem_limits();
+
 	if (adev->kfd.dev) {
 		struct kgd2kfd_shared_resources gpu_resources = {
 			.compute_vmid_bitmap =
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 4432e169fae8f..c3990a5eb7c69 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -36,6 +36,7 @@
 #include <uapi/linux/kfd_ioctl.h>
 #include "amdgpu_xgmi.h"
 #include "kfd_smi_events.h"
+#include <drm/ttm/ttm_tt.h>
 
 /* Userptr restore delay, just long enough to allow consecutive VM
  * changes to accumulate
@@ -110,13 +111,16 @@ void amdgpu_amdkfd_gpuvm_init_mem_limits(void)
 	struct sysinfo si;
 	uint64_t mem;
 
+	if (kfd_mem_limit.max_system_mem_limit)
+		return;
+
 	si_meminfo(&si);
 	mem = si.freeram - si.freehigh;
 	mem *= si.mem_unit;
 
 	spin_lock_init(&kfd_mem_limit.mem_limit_lock);
 	kfd_mem_limit.max_system_mem_limit = mem - (mem >> 4);
-	kfd_mem_limit.max_ttm_mem_limit = (mem >> 1) - (mem >> 3);
+	kfd_mem_limit.max_ttm_mem_limit = ttm_tt_pages_limit() << PAGE_SHIFT;
 	pr_debug("Kernel memory limit %lluM, TTM limit %lluM\n",
 		(kfd_mem_limit.max_system_mem_limit >> 20),
 		(kfd_mem_limit.max_ttm_mem_limit >> 20));
-- 
GitLab


From 514987a5bc3eb9bc86946f8013a9bdf732a29ef8 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Wed, 26 Apr 2023 15:54:55 -0700
Subject: [PATCH 0925/1791] drm/amdgpu: add a missing lock for AMDGPU_SCHED
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

mgr->ctx_handles should be protected by mgr->lock.

v2: improve commit message
v3: add a Fixes tag

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Fixes: 52c6a62c64fa ("drm/amdgpu: add interface for editing a foreign process's priority v3")
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
index e9b45089a28a6..863b2a34b2d64 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
@@ -38,6 +38,7 @@ static int amdgpu_sched_process_priority_override(struct amdgpu_device *adev,
 {
 	struct fd f = fdget(fd);
 	struct amdgpu_fpriv *fpriv;
+	struct amdgpu_ctx_mgr *mgr;
 	struct amdgpu_ctx *ctx;
 	uint32_t id;
 	int r;
@@ -51,8 +52,11 @@ static int amdgpu_sched_process_priority_override(struct amdgpu_device *adev,
 		return r;
 	}
 
-	idr_for_each_entry(&fpriv->ctx_mgr.ctx_handles, ctx, id)
+	mgr = &fpriv->ctx_mgr;
+	mutex_lock(&mgr->lock);
+	idr_for_each_entry(&mgr->ctx_handles, ctx, id)
 		amdgpu_ctx_priority_override(ctx, priority);
+	mutex_unlock(&mgr->lock);
 
 	fdput(f);
 	return 0;
-- 
GitLab


From 64e2e71737acad867577deac92c1bec62e8b57d1 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Thu, 27 Apr 2023 16:58:50 +0800
Subject: [PATCH 0926/1791] drm/amdgpu: Add SDMA_UTCL1_WR_FIFO_SED field for
 sdma_v4_4_ras_field

Query sdma_utcl1_wr_fifo_sed fiel to detect UTCL1_WR_FIFO SED error counts

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c
index 6f9895cdddb10..0ddb6955a6d3b 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c
@@ -141,6 +141,10 @@ static const struct soc15_ras_field_entry sdma_v4_4_ras_fields[] = {
 	SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_UTCL1_RDBST_FIFO_SED),
 	0, 0,
 	},
+	{ "SDMA_UTCL1_WR_FIFO_SED", SOC15_REG_ENTRY(SDMA0, 0, regSDMA0_EDC_COUNTER2),
+	SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_UTCL1_WR_FIFO_SED),
+	0, 0,
+	},
 	{ "SDMA_DATA_LUT_FIFO_SED", SOC15_REG_ENTRY(SDMA0, 0, regSDMA0_EDC_COUNTER2),
 	SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_DATA_LUT_FIFO_SED),
 	0, 0,
-- 
GitLab


From 9f58341d63bd26e93cca66e9e1ca850d3c40b0c3 Mon Sep 17 00:00:00 2001
From: Horace Chen <horace.chen@amd.com>
Date: Tue, 25 Apr 2023 13:15:32 +0800
Subject: [PATCH 0927/1791] drm/amdgpu: disable SDMA WPTR_POLL_ENABLE for
 SR-IOV

[Why]
This WPTR_POLL_ENABLE is a hardware contigious polling which will cause
FCLK and UCLK to keep on a high level.
Mostly its case can be covered by F32_WPTR_POLL_ENABLE which polls by
firmware.
So to save power, SR-IOV also needs to disable this bit

Signed-off-by: Horace Chen <horace.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index eb722830531f5..3d9a80511a450 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -510,10 +510,7 @@ static int sdma_v6_0_gfx_resume(struct amdgpu_device *adev)
 		       lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC);
 
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1);
-		if (amdgpu_sriov_vf(adev))
-			rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 1);
-		else
-			rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0);
+		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0);
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, F32_WPTR_POLL_ENABLE, 1);
 
 		WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE), ring->gpu_addr >> 8);
-- 
GitLab


From a6f7baa387a32940e364978f72c1c150a29a219d Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 27 Apr 2023 20:12:31 +0530
Subject: [PATCH 0928/1791] drm/amd/amdgpu: Simplify switch case statements in
 amdgpu_connectors.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following checkpatch errors:

ERROR: trailing statements should be on next line
ERROR: space required after that ',' (ctx:VxV)
ERROR: code indent should use tabs where possible

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_connectors.c    | 39 ++++++++++++++-----
 1 file changed, 29 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 6be30dcb029d5..d34037b85cf85 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -593,11 +593,20 @@ static int amdgpu_connector_set_property(struct drm_connector *connector,
 
 		switch (val) {
 		default:
-		case DRM_MODE_SCALE_NONE: rmx_type = RMX_OFF; break;
-		case DRM_MODE_SCALE_CENTER: rmx_type = RMX_CENTER; break;
-		case DRM_MODE_SCALE_ASPECT: rmx_type = RMX_ASPECT; break;
-		case DRM_MODE_SCALE_FULLSCREEN: rmx_type = RMX_FULL; break;
+		case DRM_MODE_SCALE_NONE:
+			rmx_type = RMX_OFF;
+			break;
+		case DRM_MODE_SCALE_CENTER:
+			rmx_type = RMX_CENTER;
+			break;
+		case DRM_MODE_SCALE_ASPECT:
+			rmx_type = RMX_ASPECT;
+			break;
+		case DRM_MODE_SCALE_FULLSCREEN:
+			rmx_type = RMX_FULL;
+			break;
 		}
+
 		if (amdgpu_encoder->rmx_type == rmx_type)
 			return 0;
 
@@ -799,12 +808,21 @@ static int amdgpu_connector_set_lcd_property(struct drm_connector *connector,
 	}
 
 	switch (value) {
-	case DRM_MODE_SCALE_NONE: rmx_type = RMX_OFF; break;
-	case DRM_MODE_SCALE_CENTER: rmx_type = RMX_CENTER; break;
-	case DRM_MODE_SCALE_ASPECT: rmx_type = RMX_ASPECT; break;
+	case DRM_MODE_SCALE_NONE:
+		rmx_type = RMX_OFF;
+		break;
+	case DRM_MODE_SCALE_CENTER:
+		rmx_type = RMX_CENTER;
+		break;
+	case DRM_MODE_SCALE_ASPECT:
+		rmx_type = RMX_ASPECT;
+		break;
 	default:
-	case DRM_MODE_SCALE_FULLSCREEN: rmx_type = RMX_FULL; break;
+	case DRM_MODE_SCALE_FULLSCREEN:
+		rmx_type = RMX_FULL;
+		break;
 	}
+
 	if (amdgpu_encoder->rmx_type == rmx_type)
 		return 0;
 
@@ -1127,7 +1145,8 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force)
 					/* assume digital unless load detected otherwise */
 					amdgpu_connector->use_digital = true;
 					lret = encoder_funcs->detect(encoder, connector);
-					DRM_DEBUG_KMS("load_detect %x returned: %x\n",encoder->encoder_type,lret);
+					DRM_DEBUG_KMS("load_detect %x returned: %x\n",
+						      encoder->encoder_type, lret);
 					if (lret == connector_status_connected)
 						amdgpu_connector->use_digital = false;
 				}
@@ -1991,7 +2010,7 @@ amdgpu_connector_add(struct amdgpu_device *adev,
 	if (amdgpu_connector->hpd.hpd == AMDGPU_HPD_NONE) {
 		if (i2c_bus->valid) {
 			connector->polled = DRM_CONNECTOR_POLL_CONNECT |
-			                    DRM_CONNECTOR_POLL_DISCONNECT;
+						DRM_CONNECTOR_POLL_DISCONNECT;
 		}
 	} else
 		connector->polled = DRM_CONNECTOR_POLL_HPD;
-- 
GitLab


From 04b3c34f5cb2994a5e466659ea4aa962c0fc09a6 Mon Sep 17 00:00:00 2001
From: Xiaomeng Hou <Xiaomeng.Hou@amd.com>
Date: Thu, 27 Apr 2023 13:38:47 +0800
Subject: [PATCH 0929/1791] drm/amdgpu: remove pasid_src field from IV entry

PASID_SRC is not actually present in the Interrupt Packet, the field is
taken as reserved bits now. So remove it from IV entry to avoid misuse.

Signed-off-by: Xiaomeng Hou <Xiaomeng.Hou@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c  | 1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index d58353c89e59a..fceb3b384955a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -271,7 +271,6 @@ void amdgpu_ih_decode_iv_helper(struct amdgpu_device *adev,
 	entry->timestamp_src = dw[2] >> 31;
 	entry->pasid = dw[3] & 0xffff;
 	entry->node_id = (dw[3] >> 16) & 0xff;
-	entry->pasid_src = dw[3] >> 31;
 	entry->src_data[0] = dw[4];
 	entry->src_data[1] = dw[5];
 	entry->src_data[2] = dw[6];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
index 7a8e686bdd412..1c747ac4129a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
@@ -54,7 +54,6 @@ struct amdgpu_iv_entry {
 	unsigned timestamp_src;
 	unsigned pasid;
 	unsigned node_id;
-	unsigned pasid_src;
 	unsigned src_data[AMDGPU_IRQ_SRC_DATA_MAX_SIZE_DW];
 	const uint32_t *iv_entry;
 };
-- 
GitLab


From d446127107e8185c2dc750cd4d0c5ff697a694f6 Mon Sep 17 00:00:00 2001
From: YuBiao Wang <YuBiao.Wang@amd.com>
Date: Fri, 28 Apr 2023 14:00:40 +0800
Subject: [PATCH 0930/1791] drm/amdgpu: Enable mcbp under sriov by default

Enable mcbp under sriov by default. Asics with soc21 supports mcbp now
so we should set it enabled.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Horace Chen <horace.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 9dd474262c291..1311e42ab8e99 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -65,8 +65,8 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	adev->cg_flags = 0;
 	adev->pg_flags = 0;
 
-	/* enable mcbp for sriov asic_type before soc21 */
-	amdgpu_mcbp = (adev->asic_type < CHIP_IP_DISCOVERY) ? 1 : 0;
+	/* enable mcbp for sriov */
+	amdgpu_mcbp = 1;
 
 }
 
-- 
GitLab


From 8fa76350587b6deb8a95d83f9cb23ce7599587b5 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 28 Apr 2023 16:04:59 +0530
Subject: [PATCH 0931/1791] drm/amd/amdgpu: Fix style problems in
 amdgpu_debugfs.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following issues reported by checkpatch:

WARNING: please, no space before tabs
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: sizeof *rd should be sizeof(*rd)
WARNING: Missing a blank line after declarations
WARNING: sizeof rd->id should be sizeof(rd->id)
WARNING: static const char * array should probably be static const char * const
WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
WARNING: Prefer seq_puts to seq_printf
ERROR: space prohibited after that open parenthesis '('

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 28 +++++++++++----------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index cc64ae550bc1e..df94cd2c4b39f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -56,14 +56,14 @@
  *
  * Bit 62:  Indicates a GRBM bank switch is needed
  * Bit 61:  Indicates a SRBM bank switch is needed (implies bit 62 is
- * 	    zero)
+ *	    zero)
  * Bits 24..33: The SE or ME selector if needed
  * Bits 34..43: The SH (or SA) or PIPE selector if needed
  * Bits 44..53: The INSTANCE (or CU/WGP) or QUEUE selector if needed
  *
  * Bit 23:  Indicates that the PM power gating lock should be held
- * 	    This is necessary to read registers that might be
- * 	    unreliable during a power gating transistion.
+ *	    This is necessary to read registers that might be
+ *	    unreliable during a power gating transistion.
  *
  * The lower bits are the BYTE offset of the register to read.  This
  * allows reading multiple registers in a single call and having
@@ -76,7 +76,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 	ssize_t result = 0;
 	int r;
 	bool pm_pg_lock, use_bank, use_ring;
-	unsigned instance_bank, sh_bank, se_bank, me, pipe, queue, vmid;
+	unsigned int instance_bank, sh_bank, se_bank, me, pipe, queue, vmid;
 
 	pm_pg_lock = use_bank = use_ring = false;
 	instance_bank = sh_bank = se_bank = me = pipe = queue = vmid = 0;
@@ -208,7 +208,7 @@ static int amdgpu_debugfs_regs2_open(struct inode *inode, struct file *file)
 {
 	struct amdgpu_debugfs_regs2_data *rd;
 
-	rd = kzalloc(sizeof *rd, GFP_KERNEL);
+	rd = kzalloc(sizeof(*rd), GFP_KERNEL);
 	if (!rd)
 		return -ENOMEM;
 	rd->adev = file_inode(file)->i_private;
@@ -221,6 +221,7 @@ static int amdgpu_debugfs_regs2_open(struct inode *inode, struct file *file)
 static int amdgpu_debugfs_regs2_release(struct inode *inode, struct file *file)
 {
 	struct amdgpu_debugfs_regs2_data *rd = file->private_data;
+
 	mutex_destroy(&rd->lock);
 	kfree(file->private_data);
 	return 0;
@@ -324,7 +325,8 @@ static long amdgpu_debugfs_regs2_ioctl(struct file *f, unsigned int cmd, unsigne
 	switch (cmd) {
 	case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE:
 		mutex_lock(&rd->lock);
-		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata *)data, sizeof rd->id);
+		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata *)data,
+				   sizeof(rd->id));
 		mutex_unlock(&rd->lock);
 		return r ? -EINVAL : 0;
 	default:
@@ -863,7 +865,7 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf,
  * The offset being sought changes which wave that the status data
  * will be returned for.  The bits are used as follows:
  *
- * Bits 0..6: 	Byte offset into data
+ * Bits 0..6:	Byte offset into data
  * Bits 7..14:	SE selector
  * Bits 15..22:	SH/SA selector
  * Bits 23..30: CU/{WGP+SIMD} selector
@@ -1429,7 +1431,7 @@ static const struct file_operations *debugfs_regs[] = {
 	&amdgpu_debugfs_gfxoff_residency_fops,
 };
 
-static const char *debugfs_regs_names[] = {
+static const char * const debugfs_regs_names[] = {
 	"amdgpu_regs",
 	"amdgpu_regs2",
 	"amdgpu_regs_didt",
@@ -1447,7 +1449,7 @@ static const char *debugfs_regs_names[] = {
 
 /**
  * amdgpu_debugfs_regs_init -	Initialize debugfs entries that provide
- * 				register access.
+ *				register access.
  *
  * @adev: The device to attach the debugfs entries to
  */
@@ -1459,7 +1461,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
 
 	for (i = 0; i < ARRAY_SIZE(debugfs_regs); i++) {
 		ent = debugfs_create_file(debugfs_regs_names[i],
-					  S_IFREG | S_IRUGO, root,
+					  S_IFREG | 0444, root,
 					  adev, debugfs_regs[i]);
 		if (!i && !IS_ERR_OR_NULL(ent))
 			i_size_write(ent->d_inode, adev->rmmio_size);
@@ -1494,12 +1496,12 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
 		kthread_park(ring->sched.thread);
 	}
 
-	seq_printf(m, "run ib test:\n");
+	seq_puts(m, "run ib test:\n");
 	r = amdgpu_ib_ring_tests(adev);
 	if (r)
 		seq_printf(m, "ib ring tests failed (%d).\n", r);
 	else
-		seq_printf(m, "ib ring tests passed.\n");
+		seq_puts(m, "ib ring tests passed.\n");
 
 	/* go on the scheduler */
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
@@ -1978,7 +1980,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 		amdgpu_debugfs_ring_init(adev, ring);
 	}
 
-	for ( i = 0; i < adev->vcn.num_vcn_inst; i++) {
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		if (!amdgpu_vcnfw_log)
 			break;
 
-- 
GitLab


From 44407010ce7f524e4f68aa25f9250a9cf9e4410c Mon Sep 17 00:00:00 2001
From: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Date: Wed, 26 Apr 2023 12:24:12 -0700
Subject: [PATCH 0932/1791] drm/amd/display: Fix possible NULL dereference in
 dc_dmub_srv_cmd_run_list()

We have a NULL check for 'dc_dmub_srv' in dc_dmub_srv_cmd_run_list()
but we are dereferencing it before checking.

Fix this moving the dereference next to NULL check.

This issue is found with Smatch(static analysis tool).

Fixes: e97cc04fe0fb ("drm/amd/display: refactor dmub commands into single function")
Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 36d936ab43005..0319a30f2d5c1 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -125,7 +125,7 @@ bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd
 
 bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type)
 {
-	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
+	struct dc_context *dc_ctx;
 	struct dmub_srv *dmub;
 	enum dmub_status status;
 	int i;
@@ -133,6 +133,7 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun
 	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
 		return false;
 
+	dc_ctx = dc_dmub_srv->ctx;
 	dmub = dc_dmub_srv->dmub;
 
 	for (i = 0 ; i < count; i++) {
-- 
GitLab


From 9420a034060a094874091cb6c6257268f08a1999 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 25 Apr 2023 08:34:17 +0530
Subject: [PATCH 0933/1791] drm/amd/display: Add logging for eDP v1.4 supported
 sink rates

Include eDP v1.4 panels supported sink rates in debug output,
useful info for knowing optimized link rates

Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Jerry Zuo <jerry.zuo@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 577f74cc50fb4..f46864630506f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1951,6 +1951,9 @@ void detect_edp_sink_caps(struct dc_link *link)
 			link_rate_in_khz = (supported_link_rates[entry+1] * 0x100 +
 										supported_link_rates[entry]) * 200;
 
+			DC_LOG_DC("%s: eDP v1.4 supported sink rates: [%d] %d kHz\n", __func__,
+				  entry / 2, link_rate_in_khz);
+
 			if (link_rate_in_khz != 0) {
 				link_rate = linkRateInKHzToLinkRateMultiplier(link_rate_in_khz);
 				link->dpcd_caps.edp_supported_link_rates[link->dpcd_caps.edp_supported_link_rates_count] = link_rate;
-- 
GitLab


From 435af0b919bf9eb78f4e05e8596ebed9ca7885b7 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Mon, 24 Apr 2023 14:46:16 -0400
Subject: [PATCH 0934/1791] drm/amdkfd: Optimize svm range map to GPU with
 XNACK on

With XNACK on if svm_range_set_attr set the range access or
access_in_place attribute, we don't call svm_range_validate_and_map to
update GPU mapping. This avoids prefaulting the range pages on system
memory if the range is not prefetch to VRAM and not mapped to GPUs.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 96a138a395150..c02430537e9c8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -735,7 +735,9 @@ svm_range_apply_attrs(struct kfd_process *p, struct svm_range *prange,
 		case KFD_IOCTL_SVM_ATTR_ACCESS:
 		case KFD_IOCTL_SVM_ATTR_ACCESS_IN_PLACE:
 		case KFD_IOCTL_SVM_ATTR_NO_ACCESS:
-			*update_mapping = true;
+			if (!p->xnack_enabled)
+				*update_mapping = true;
+
 			gpuidx = kfd_process_gpuidx_from_gpuid(p,
 							       attrs[i].value);
 			if (attrs[i].type == KFD_IOCTL_SVM_ATTR_NO_ACCESS) {
-- 
GitLab


From 6ae869b9b62534eaf04f838048338bf150bc6ce4 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 24 Apr 2023 12:50:30 -0400
Subject: [PATCH 0935/1791] drm/amdgpu/gfx11: drop old bring up code

No longer used.  Remove it.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 70 ++------------------------
 1 file changed, 3 insertions(+), 67 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 8a4c4769e6074..646003b2faf82 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -3680,55 +3680,6 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 	return 0;
 }
 
-#ifdef BRING_UP_DEBUG
-static int gfx_v11_0_gfx_queue_init_register(struct amdgpu_ring *ring)
-{
-	struct amdgpu_device *adev = ring->adev;
-	struct v11_gfx_mqd *mqd = ring->mqd_ptr;
-
-	/* set mmCP_GFX_HQD_WPTR/_HI to 0 */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_WPTR, mqd->cp_gfx_hqd_wptr);
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_WPTR_HI, mqd->cp_gfx_hqd_wptr_hi);
-
-	/* set GFX_MQD_BASE */
-	WREG32_SOC15(GC, 0, regCP_MQD_BASE_ADDR, mqd->cp_mqd_base_addr);
-	WREG32_SOC15(GC, 0, regCP_MQD_BASE_ADDR_HI, mqd->cp_mqd_base_addr_hi);
-
-	/* set GFX_MQD_CONTROL */
-	WREG32_SOC15(GC, 0, regCP_GFX_MQD_CONTROL, mqd->cp_gfx_mqd_control);
-
-	/* set GFX_HQD_VMID to 0 */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_VMID, mqd->cp_gfx_hqd_vmid);
-
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_QUEUE_PRIORITY,
-			mqd->cp_gfx_hqd_queue_priority);
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_QUANTUM, mqd->cp_gfx_hqd_quantum);
-
-	/* set GFX_HQD_BASE, similar as CP_RB_BASE */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_BASE, mqd->cp_gfx_hqd_base);
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_BASE_HI, mqd->cp_gfx_hqd_base_hi);
-
-	/* set GFX_HQD_RPTR_ADDR, similar as CP_RB_RPTR */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_RPTR_ADDR, mqd->cp_gfx_hqd_rptr_addr);
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_RPTR_ADDR_HI, mqd->cp_gfx_hqd_rptr_addr_hi);
-
-	/* set GFX_HQD_CNTL, similar as CP_RB_CNTL */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_CNTL, mqd->cp_gfx_hqd_cntl);
-
-	/* set RB_WPTR_POLL_ADDR */
-	WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_ADDR_LO, mqd->cp_rb_wptr_poll_addr_lo);
-	WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_ADDR_HI, mqd->cp_rb_wptr_poll_addr_hi);
-
-	/* set RB_DOORBELL_CONTROL */
-	WREG32_SOC15(GC, 0, regCP_RB_DOORBELL_CONTROL, mqd->cp_rb_doorbell_control);
-
-	/* active the queue */
-	WREG32_SOC15(GC, 0, regCP_GFX_HQD_ACTIVE, mqd->cp_gfx_hqd_active);
-
-	return 0;
-}
-#endif
-
 static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -3740,9 +3691,6 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 		mutex_lock(&adev->srbm_mutex);
 		soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
 		amdgpu_ring_init_mqd(ring);
-#ifdef BRING_UP_DEBUG
-		gfx_v11_0_gfx_queue_init_register(ring);
-#endif
 		soc21_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 		if (adev->gfx.me.mqd_backup[mqd_idx])
@@ -3755,13 +3703,6 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 		ring->wptr = 0;
 		*ring->wptr_cpu_addr = 0;
 		amdgpu_ring_clear_ring(ring);
-#ifdef BRING_UP_DEBUG
-		mutex_lock(&adev->srbm_mutex);
-		soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
-		gfx_v11_0_gfx_queue_init_register(ring);
-		soc21_grbm_select(adev, 0, 0, 0, 0);
-		mutex_unlock(&adev->srbm_mutex);
-#endif
 	} else {
 		amdgpu_ring_clear_ring(ring);
 	}
@@ -3769,7 +3710,6 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 	return 0;
 }
 
-#ifndef BRING_UP_DEBUG
 static int gfx_v11_0_kiq_enable_kgq(struct amdgpu_device *adev)
 {
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
@@ -3791,7 +3731,6 @@ static int gfx_v11_0_kiq_enable_kgq(struct amdgpu_device *adev)
 
 	return amdgpu_ring_test_helper(kiq_ring);
 }
-#endif
 
 static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 {
@@ -3815,11 +3754,11 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 		if (r)
 			goto done;
 	}
-#ifndef BRING_UP_DEBUG
+
 	r = gfx_v11_0_kiq_enable_kgq(adev);
 	if (r)
 		goto done;
-#endif
+
 	r = gfx_v11_0_cp_gfx_start(adev);
 	if (r)
 		goto done;
@@ -4453,7 +4392,6 @@ static int gfx_v11_0_hw_init(void *handle)
 	return r;
 }
 
-#ifndef BRING_UP_DEBUG
 static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev)
 {
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
@@ -4476,7 +4414,6 @@ static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev)
 
 	return r;
 }
-#endif
 
 static int gfx_v11_0_hw_fini(void *handle)
 {
@@ -4488,13 +4425,12 @@ static int gfx_v11_0_hw_fini(void *handle)
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
 	if (!adev->no_hw_access) {
-#ifndef BRING_UP_DEBUG
 		if (amdgpu_async_gfx_ring) {
 			r = gfx_v11_0_kiq_disable_kgq(adev);
 			if (r)
 				DRM_ERROR("KGQ disable failed\n");
 		}
-#endif
+
 		if (amdgpu_gfx_disable_kcq(adev, 0))
 			DRM_ERROR("KCQ disable failed\n");
 
-- 
GitLab


From edacf33357b8576db3198e4ae41bb7e6cd41ce4e Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 24 Apr 2023 12:52:31 -0400
Subject: [PATCH 0936/1791] drm/amdgpu/gfx10: drop old bring up code

No longer used.  Remove it.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 70 ++------------------------
 1 file changed, 3 insertions(+), 67 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 8bd07ff596718..25be4485dcd50 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -6423,55 +6423,6 @@ static int gfx_v10_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 	return 0;
 }
 
-#ifdef BRING_UP_DEBUG
-static int gfx_v10_0_gfx_queue_init_register(struct amdgpu_ring *ring)
-{
-	struct amdgpu_device *adev = ring->adev;
-	struct v10_gfx_mqd *mqd = ring->mqd_ptr;
-
-	/* set mmCP_GFX_HQD_WPTR/_HI to 0 */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_WPTR, mqd->cp_gfx_hqd_wptr);
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_WPTR_HI, mqd->cp_gfx_hqd_wptr_hi);
-
-	/* set GFX_MQD_BASE */
-	WREG32_SOC15(GC, 0, mmCP_MQD_BASE_ADDR, mqd->cp_mqd_base_addr);
-	WREG32_SOC15(GC, 0, mmCP_MQD_BASE_ADDR_HI, mqd->cp_mqd_base_addr_hi);
-
-	/* set GFX_MQD_CONTROL */
-	WREG32_SOC15(GC, 0, mmCP_GFX_MQD_CONTROL, mqd->cp_gfx_mqd_control);
-
-	/* set GFX_HQD_VMID to 0 */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_VMID, mqd->cp_gfx_hqd_vmid);
-
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_QUEUE_PRIORITY,
-			mqd->cp_gfx_hqd_queue_priority);
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_QUANTUM, mqd->cp_gfx_hqd_quantum);
-
-	/* set GFX_HQD_BASE, similar as CP_RB_BASE */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_BASE, mqd->cp_gfx_hqd_base);
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_BASE_HI, mqd->cp_gfx_hqd_base_hi);
-
-	/* set GFX_HQD_RPTR_ADDR, similar as CP_RB_RPTR */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_RPTR_ADDR, mqd->cp_gfx_hqd_rptr_addr);
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_RPTR_ADDR_HI, mqd->cp_gfx_hqd_rptr_addr_hi);
-
-	/* set GFX_HQD_CNTL, similar as CP_RB_CNTL */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_CNTL, mqd->cp_gfx_hqd_cntl);
-
-	/* set RB_WPTR_POLL_ADDR */
-	WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, mqd->cp_rb_wptr_poll_addr_lo);
-	WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, mqd->cp_rb_wptr_poll_addr_hi);
-
-	/* set RB_DOORBELL_CONTROL */
-	WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL, mqd->cp_rb_doorbell_control);
-
-	/* active the queue */
-	WREG32_SOC15(GC, 0, mmCP_GFX_HQD_ACTIVE, mqd->cp_gfx_hqd_active);
-
-	return 0;
-}
-#endif
-
 static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -6492,9 +6443,6 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 		if (ring->doorbell_index == adev->doorbell_index.gfx_ring0 << 1)
 			gfx_v10_0_cp_gfx_set_doorbell(adev, ring);
 
-#ifdef BRING_UP_DEBUG
-		gfx_v10_0_gfx_queue_init_register(ring);
-#endif
 		nv_grbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 		if (adev->gfx.me.mqd_backup[mqd_idx])
@@ -6507,13 +6455,6 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 		ring->wptr = 0;
 		*ring->wptr_cpu_addr = 0;
 		amdgpu_ring_clear_ring(ring);
-#ifdef BRING_UP_DEBUG
-		mutex_lock(&adev->srbm_mutex);
-		nv_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
-		gfx_v10_0_gfx_queue_init_register(ring);
-		nv_grbm_select(adev, 0, 0, 0, 0);
-		mutex_unlock(&adev->srbm_mutex);
-#endif
 	} else {
 		amdgpu_ring_clear_ring(ring);
 	}
@@ -6521,7 +6462,6 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 	return 0;
 }
 
-#ifndef BRING_UP_DEBUG
 static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev)
 {
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
@@ -6543,7 +6483,6 @@ static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev)
 
 	return amdgpu_ring_test_helper(kiq_ring);
 }
-#endif
 
 static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 {
@@ -6567,11 +6506,11 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 		if (r)
 			goto done;
 	}
-#ifndef BRING_UP_DEBUG
+
 	r = gfx_v10_0_kiq_enable_kgq(adev);
 	if (r)
 		goto done;
-#endif
+
 	r = gfx_v10_0_cp_gfx_start(adev);
 	if (r)
 		goto done;
@@ -7239,7 +7178,6 @@ static int gfx_v10_0_hw_init(void *handle)
 	return r;
 }
 
-#ifndef BRING_UP_DEBUG
 static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev)
 {
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
@@ -7261,7 +7199,6 @@ static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev)
 	else
 		return 0;
 }
-#endif
 
 static int gfx_v10_0_hw_fini(void *handle)
 {
@@ -7272,13 +7209,12 @@ static int gfx_v10_0_hw_fini(void *handle)
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
 	if (!adev->no_hw_access) {
-#ifndef BRING_UP_DEBUG
 		if (amdgpu_async_gfx_ring) {
 			r = gfx_v10_0_kiq_disable_kgq(adev);
 			if (r)
 				DRM_ERROR("KGQ disable failed\n");
 		}
-#endif
+
 		if (amdgpu_gfx_disable_kcq(adev, 0))
 			DRM_ERROR("KCQ disable failed\n");
 	}
-- 
GitLab


From f14c8c3e1fc9e10c6d54999a96acb2b5087374df Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 28 Apr 2023 17:22:20 +0530
Subject: [PATCH 0937/1791] drm/amd/amdgpu: Fix style problems in amdgpu_psp.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following checkpatch warnings & error in amdgpu_psp.c

WARNING: Comparisons should place the constant on the right side of the test
WARNING: braces {} are not necessary for single statement blocks
WARNING: please, no space before tabs
WARNING: braces {} are not necessary for single statement blocks
ERROR: that open brace { should be on the previous line

Suggested-by: Christian König <christian.koenig@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 51 ++++++++++---------------
 1 file changed, 20 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index c58654a8b6c5e..aa37b703c7186 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -411,7 +411,7 @@ static int psp_sw_init(void *handle)
 	if ((psp_get_runtime_db_entry(adev,
 				PSP_RUNTIME_ENTRY_TYPE_PPTABLE_ERR_STATUS,
 				&scpm_entry)) &&
-	    (SCPM_DISABLE != scpm_entry.scpm_status)) {
+	    (scpm_entry.scpm_status != SCPM_DISABLE)) {
 		adev->scpm_enabled = true;
 		adev->scpm_status = scpm_entry.scpm_status;
 	} else {
@@ -458,10 +458,9 @@ static int psp_sw_init(void *handle)
 
 	if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) ||
 	    adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) {
-		ret= psp_sysfs_init(adev);
-		if (ret) {
+		ret = psp_sysfs_init(adev);
+		if (ret)
 			return ret;
-		}
 	}
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG,
@@ -645,7 +644,7 @@ psp_cmd_submit_buf(struct psp_context *psp,
 	skip_unsupport = (psp->cmd_buf_mem->resp.status == TEE_ERROR_NOT_SUPPORTED ||
 		psp->cmd_buf_mem->resp.status == PSP_ERR_UNKNOWN_COMMAND) && amdgpu_sriov_vf(psp->adev);
 
-	memcpy((void*)&cmd->resp, (void*)&psp->cmd_buf_mem->resp, sizeof(struct psp_gfx_resp));
+	memcpy(&cmd->resp, &psp->cmd_buf_mem->resp, sizeof(struct psp_gfx_resp));
 
 	/* In some cases, psp response status is not 0 even there is no
 	 * problem while the command is submitted. Some version of PSP FW
@@ -830,7 +829,7 @@ static int psp_tmr_load(struct psp_context *psp)
 }
 
 static void psp_prep_tmr_unload_cmd_buf(struct psp_context *psp,
-				        struct psp_gfx_cmd_resp *cmd)
+					struct psp_gfx_cmd_resp *cmd)
 {
 	if (amdgpu_sriov_vf(psp->adev))
 		cmd->cmd_id = GFX_CMD_ID_DESTROY_VMR;
@@ -1067,7 +1066,7 @@ static void psp_prep_ta_load_cmd_buf(struct psp_gfx_cmd_resp *cmd,
 				     struct ta_context *context)
 {
 	cmd->cmd_id				= context->ta_load_type;
-	cmd->cmd.cmd_load_ta.app_phy_addr_lo 	= lower_32_bits(ta_bin_mc);
+	cmd->cmd.cmd_load_ta.app_phy_addr_lo	= lower_32_bits(ta_bin_mc);
 	cmd->cmd.cmd_load_ta.app_phy_addr_hi	= upper_32_bits(ta_bin_mc);
 	cmd->cmd.cmd_load_ta.app_len		= context->bin_desc.size_bytes;
 
@@ -1138,9 +1137,8 @@ int psp_ta_load(struct psp_context *psp, struct ta_context *context)
 
 	context->resp_status = cmd->resp.status;
 
-	if (!ret) {
+	if (!ret)
 		context->session_id = cmd->resp.session_id;
-	}
 
 	release_psp_cmd_buf(psp);
 
@@ -1467,8 +1465,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
 	if (amdgpu_ras_intr_triggered())
 		return ret;
 
-	if (ras_cmd->if_version > RAS_TA_HOST_IF_VER)
-	{
+	if (ras_cmd->if_version > RAS_TA_HOST_IF_VER) {
 		DRM_WARN("RAS: Unsupported Interface");
 		return -EINVAL;
 	}
@@ -1478,8 +1475,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
 			dev_warn(psp->adev->dev, "ECC switch disabled\n");
 
 			ras_cmd->ras_status = TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE;
-		}
-		else if (ras_cmd->ras_out_message.flags.reg_access_failure_flag)
+		} else if (ras_cmd->ras_out_message.flags.reg_access_failure_flag)
 			dev_warn(psp->adev->dev,
 				 "RAS internal register access blocked\n");
 
@@ -1575,11 +1571,10 @@ int psp_ras_initialize(struct psp_context *psp)
 				if (ret)
 					dev_warn(adev->dev, "PSP set boot config failed\n");
 				else
-					dev_warn(adev->dev, "GECC will be disabled in next boot cycle "
-						 "if set amdgpu_ras_enable and/or amdgpu_ras_mask to 0x0\n");
+					dev_warn(adev->dev, "GECC will be disabled in next boot cycle if set amdgpu_ras_enable and/or amdgpu_ras_mask to 0x0\n");
 			}
 		} else {
-			if (1 == boot_cfg) {
+			if (boot_cfg == 1) {
 				dev_info(adev->dev, "GECC is enabled\n");
 			} else {
 				/* enable GECC in next boot cycle if it is disabled
@@ -2365,7 +2360,7 @@ static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode,
 }
 
 static int psp_execute_non_psp_fw_load(struct psp_context *psp,
-			          struct amdgpu_firmware_info *ucode)
+				  struct amdgpu_firmware_info *ucode)
 {
 	int ret = 0;
 	struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp);
@@ -2404,9 +2399,8 @@ static int psp_load_smu_fw(struct psp_context *psp)
 	     (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 4) ||
 	      adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 2)))) {
 		ret = amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_UNLOAD);
-		if (ret) {
+		if (ret)
 			DRM_WARN("Failed to set MP1 state prepare for reload\n");
-		}
 	}
 
 	ret = psp_execute_non_psp_fw_load(psp, ucode);
@@ -2716,9 +2710,8 @@ static int psp_suspend(void *handle)
 	}
 
 	ret = psp_ring_stop(psp, PSP_RING_TYPE__KM);
-	if (ret) {
+	if (ret)
 		DRM_ERROR("PSP ring stop failed\n");
-	}
 
 out:
 	return ret;
@@ -2967,7 +2960,7 @@ static int parse_sos_bin_descriptor(struct psp_context *psp,
 		psp->sos.fw_version        = le32_to_cpu(desc->fw_version);
 		psp->sos.feature_version   = le32_to_cpu(desc->fw_version);
 		psp->sos.size_bytes        = le32_to_cpu(desc->size_bytes);
-		psp->sos.start_addr 	   = ucode_start_addr;
+		psp->sos.start_addr	   = ucode_start_addr;
 		break;
 	case PSP_FW_TYPE_PSP_SYS_DRV:
 		psp->sys.fw_version        = le32_to_cpu(desc->fw_version);
@@ -3491,7 +3484,7 @@ void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size
 	drm_dev_exit(idx);
 }
 
-static DEVICE_ATTR(usbc_pd_fw, S_IRUGO | S_IWUSR,
+static DEVICE_ATTR(usbc_pd_fw, 0644,
 		   psp_usbc_pd_fw_sysfs_read,
 		   psp_usbc_pd_fw_sysfs_write);
 
@@ -3674,8 +3667,7 @@ static void psp_sysfs_fini(struct amdgpu_device *adev)
 	device_remove_file(adev->dev, &dev_attr_usbc_pd_fw);
 }
 
-const struct amdgpu_ip_block_version psp_v3_1_ip_block =
-{
+const struct amdgpu_ip_block_version psp_v3_1_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_PSP,
 	.major = 3,
 	.minor = 1,
@@ -3683,8 +3675,7 @@ const struct amdgpu_ip_block_version psp_v3_1_ip_block =
 	.funcs = &psp_ip_funcs,
 };
 
-const struct amdgpu_ip_block_version psp_v10_0_ip_block =
-{
+const struct amdgpu_ip_block_version psp_v10_0_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_PSP,
 	.major = 10,
 	.minor = 0,
@@ -3692,8 +3683,7 @@ const struct amdgpu_ip_block_version psp_v10_0_ip_block =
 	.funcs = &psp_ip_funcs,
 };
 
-const struct amdgpu_ip_block_version psp_v11_0_ip_block =
-{
+const struct amdgpu_ip_block_version psp_v11_0_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_PSP,
 	.major = 11,
 	.minor = 0,
@@ -3709,8 +3699,7 @@ const struct amdgpu_ip_block_version psp_v11_0_8_ip_block = {
 	.funcs = &psp_ip_funcs,
 };
 
-const struct amdgpu_ip_block_version psp_v12_0_ip_block =
-{
+const struct amdgpu_ip_block_version psp_v12_0_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_PSP,
 	.major = 12,
 	.minor = 0,
-- 
GitLab


From 9ab367f838a692843aa21b2ff43ebfc3bafa7175 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 19 Apr 2023 17:30:39 +0530
Subject: [PATCH 0938/1791] drm/amd/display: Add logging when DP link training
 Channel EQ is Successful

Log when Channel Equalization is successful.

Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Fangzhi Zuo <Jerry.Zuo@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/link/protocols/link_dp_training_128b_132b.c     | 10 ++++++++--
 .../dc/link/protocols/link_dp_training_8b_10b.c        |  4 ++++
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c
index 23d380f09a21c..db87cfe37b5c9 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c
@@ -211,11 +211,17 @@ enum link_training_result dp_perform_128b_132b_link_training(
 
 	dpcd_set_link_settings(link, lt_settings);
 
-	if (result == LINK_TRAINING_SUCCESS)
+	if (result == LINK_TRAINING_SUCCESS) {
 		result = dp_perform_128b_132b_channel_eq_done_sequence(link, link_res, lt_settings);
+		if (result == LINK_TRAINING_SUCCESS)
+			DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__);
+	}
 
-	if (result == LINK_TRAINING_SUCCESS)
+	if (result == LINK_TRAINING_SUCCESS) {
 		result = dp_perform_128b_132b_cds_done_sequence(link, link_res, lt_settings);
+		if (result == LINK_TRAINING_SUCCESS)
+			DC_LOG_HW_LINK_TRAINING("%s: CDS done.\n", __func__);
+	}
 
 	return result;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
index 3889ebb2256bf..2b4c15b0b4070 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c
@@ -388,6 +388,8 @@ enum link_training_result dp_perform_8b_10b_link_training(
 					link_res,
 					lt_settings,
 					repeater_id);
+			if (status == LINK_TRAINING_SUCCESS)
+				DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__);
 
 			repeater_training_done(link, repeater_id);
 
@@ -409,6 +411,8 @@ enum link_training_result dp_perform_8b_10b_link_training(
 					link_res,
 					lt_settings,
 					DPRX);
+			if (status == LINK_TRAINING_SUCCESS)
+				DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__);
 		}
 	}
 
-- 
GitLab


From 5b7bee5b445785a47b781e889e51219ec35d4407 Mon Sep 17 00:00:00 2001
From: Ilya Bakoulin <Ilya.Bakoulin@amd.com>
Date: Fri, 14 Apr 2023 12:46:48 -0400
Subject: [PATCH 0939/1791] drm/amd/display: Workaround wrong HDR colorimetry
 with some receivers

[Why]
Some scalers do not pick up color space updates unless the DP link
is disabled/re-enabled which can result in incorrect/washed out
HDR colors in some cases.

[How]
Call set_dpms_on to disable the link, re-train and re-enable with the
updated output color space.

Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Ilya Bakoulin <Ilya.Bakoulin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 7 +++++++
 drivers/gpu/drm/amd/display/dc/dc.h      | 1 +
 2 files changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 89a245e3c9ac9..965b348dc8f11 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3268,6 +3268,13 @@ static void commit_planes_do_stream_update(struct dc *dc,
 						dc->hwss.prepare_bandwidth(dc, dc->current_state);
 					dc->link_srv->set_dpms_on(dc->current_state, pipe_ctx);
 				}
+			} else if (pipe_ctx->stream->link->wa_flags.blank_stream_on_ocs_change && stream_update->output_color_space
+					&& !stream->dpms_off && dc_is_dp_signal(pipe_ctx->stream->signal)) {
+				/*
+				 * Workaround for firmware issue in some receivers where they don't pick up
+				 * correct output color space unless DP link is disabled/re-enabled
+				 */
+				dc->link_srv->set_dpms_on(dc->current_state, pipe_ctx);
 			}
 
 			if (stream_update->abm_level && pipe_ctx->stream_res.abm) {
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 4424e7abb8010..892e3adb99d93 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1506,6 +1506,7 @@ struct dc_link {
 		/* Forced DPIA into TBT3 compatibility mode. */
 		bool dpia_forced_tbt3_mode;
 		bool dongle_mode_timing_override;
+		bool blank_stream_on_ocs_change;
 	} wa_flags;
 	struct link_mst_stream_allocation_table mst_stream_alloc_table;
 
-- 
GitLab


From 9749a42db74c3400e0526d9a39fa0324abfd0d66 Mon Sep 17 00:00:00 2001
From: Leo Chen <sancchen@amd.com>
Date: Thu, 13 Apr 2023 17:34:24 -0400
Subject: [PATCH 0940/1791] drm/amd/display: Change default Z8 watermark values

[Why & How]
Previous Z8 watermark values were causing flickering and OTC underflow.
Updating Z8 watermark values based on the measurement.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 19370b872a91d..1d00eb9e73c62 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -149,8 +149,8 @@ static struct _vcs_dpi_soc_bounding_box_st dcn3_14_soc = {
 	.num_states = 5,
 	.sr_exit_time_us = 16.5,
 	.sr_enter_plus_exit_time_us = 18.5,
-	.sr_exit_z8_time_us = 210.0,
-	.sr_enter_plus_exit_z8_time_us = 310.0,
+	.sr_exit_z8_time_us = 268.0,
+	.sr_enter_plus_exit_z8_time_us = 393.0,
 	.writeback_latency_us = 12.0,
 	.dram_channel_width_bytes = 4,
 	.round_trip_ping_latency_dcfclk_cycles = 106,
-- 
GitLab


From 4dfa60aede755814dcfbc9a05008265d827cc98d Mon Sep 17 00:00:00 2001
From: Leo Chen <sancchen@amd.com>
Date: Thu, 13 Apr 2023 17:37:49 -0400
Subject: [PATCH 0941/1791] drm/amd/display: Adding debug option to override Z8
 watermark values

[Why & How]
Adding debug options to override Z8 watermark values for testing purposes.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                  |  4 ++++
 drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c | 11 +++++++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 892e3adb99d93..fea68383bb61f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -700,6 +700,8 @@ struct dc_virtual_addr_space_config {
 struct dc_bounding_box_overrides {
 	int sr_exit_time_ns;
 	int sr_enter_plus_exit_time_ns;
+	int sr_exit_z8_time_ns;
+	int sr_enter_plus_exit_z8_time_ns;
 	int urgent_latency_ns;
 	int percent_of_ideal_drambw;
 	int dram_clock_change_latency_ns;
@@ -769,6 +771,8 @@ struct dc_debug_options {
 	int sr_enter_plus_exit_time_dpm0_ns;
 	int sr_exit_time_ns;
 	int sr_enter_plus_exit_time_ns;
+	int sr_exit_z8_time_ns;
+	int sr_enter_plus_exit_z8_time_ns;
 	int urgent_latency_ns;
 	uint32_t underflow_assert_delay_us;
 	int percent_of_ideal_drambw;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index a5b1e4bb1a222..3407f9a2c6a10 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -1890,6 +1890,17 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st
 				dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
 	}
 
+	if ((int)(bb->sr_exit_z8_time_us * 1000)
+				!= dc->bb_overrides.sr_exit_z8_time_ns
+			&& dc->bb_overrides.sr_exit_z8_time_ns) {
+		bb->sr_exit_z8_time_us = dc->bb_overrides.sr_exit_z8_time_ns / 1000.0;
+	}
+
+	if ((int)(bb->sr_enter_plus_exit_z8_time_us * 1000)
+				!= dc->bb_overrides.sr_enter_plus_exit_z8_time_ns
+			&& dc->bb_overrides.sr_enter_plus_exit_z8_time_ns) {
+		bb->sr_enter_plus_exit_z8_time_us = dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0;
+	}
 	if ((int)(bb->urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
 			&& dc->bb_overrides.urgent_latency_ns) {
 		bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
-- 
GitLab


From f835a571ec998b83c165022795f9385f9335f108 Mon Sep 17 00:00:00 2001
From: Samson Tam <Samson.Tam@amd.com>
Date: Wed, 19 Apr 2023 18:17:14 -0400
Subject: [PATCH 0942/1791] drm/amd/display: filter out invalid bits in
 pipe_fuses

[Why]
Reading pipe_fuses from register may have invalid bits set, which may
 affect the num_pipes erroneously.

[How]
Add read_pipes_fuses() call and filter bits based on expected number
 of pipes.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Samson Tam <Samson.Tam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c  | 10 +++++++++-
 .../gpu/drm/amd/display/dc/dcn321/dcn321_resource.c    | 10 +++++++++-
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 7feeba78c0c9e..a493238858745 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -2082,6 +2082,14 @@ static struct resource_funcs dcn32_res_pool_funcs = {
 	.restore_mall_state = dcn32_restore_mall_state,
 };
 
+static uint32_t read_pipe_fuses(struct dc_context *ctx)
+{
+	uint32_t value = REG_READ(CC_DC_PIPE_DIS);
+	/* DCN32 support max 4 pipes */
+	value = value & 0xf;
+	return value;
+}
+
 
 static bool dcn32_resource_construct(
 	uint8_t num_virtual_links,
@@ -2125,7 +2133,7 @@ static bool dcn32_resource_construct(
 	pool->base.res_cap = &res_cap_dcn32;
 	/* max number of pipes for ASIC before checking for pipe fuses */
 	num_pipes  = pool->base.res_cap->num_timing_generator;
-	pipe_fuses = REG_READ(CC_DC_PIPE_DIS);
+	pipe_fuses = read_pipe_fuses(ctx);
 
 	for (i = 0; i < pool->base.res_cap->num_timing_generator; i++)
 		if (pipe_fuses & 1 << i)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index 63bd6928c82f0..4c1e0f5a5f097 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -1633,6 +1633,14 @@ static struct resource_funcs dcn321_res_pool_funcs = {
 	.restore_mall_state = dcn32_restore_mall_state,
 };
 
+static uint32_t read_pipe_fuses(struct dc_context *ctx)
+{
+	uint32_t value = REG_READ(CC_DC_PIPE_DIS);
+	/* DCN321 support max 4 pipes */
+	value = value & 0xf;
+	return value;
+}
+
 
 static bool dcn321_resource_construct(
 	uint8_t num_virtual_links,
@@ -1675,7 +1683,7 @@ static bool dcn321_resource_construct(
 	pool->base.res_cap = &res_cap_dcn321;
 	/* max number of pipes for ASIC before checking for pipe fuses */
 	num_pipes  = pool->base.res_cap->num_timing_generator;
-	pipe_fuses = REG_READ(CC_DC_PIPE_DIS);
+	pipe_fuses = read_pipe_fuses(ctx);
 
 	for (i = 0; i < pool->base.res_cap->num_timing_generator; i++)
 		if (pipe_fuses & 1 << i)
-- 
GitLab


From 029c85adccb2e23352f622394288ecd133449332 Mon Sep 17 00:00:00 2001
From: Leo Ma <hanghong.ma@amd.com>
Date: Fri, 21 Apr 2023 09:58:25 -0400
Subject: [PATCH 0943/1791] drm/amd/display: revert "Update scaler recout data
 for visual confirm"

This reverts commit 1068e987ad0be83a109147fe7fa0891700e8d80e.

A regression is found on this change, so revert it for the time being
and resubmit when issue is fixed.

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Leo Ma <hanghong.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/core/dc_resource.c  | 17 -----------------
 .../drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c  | 18 +++++++++++++++++-
 2 files changed, 17 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 193e09b05f5a4..7e1e5532f88f3 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -69,9 +69,6 @@
 #include "../dcn32/dcn32_resource.h"
 #include "../dcn321/dcn321_resource.h"
 
-#define VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT 3
-#define VISUAL_CONFIRM_RECT_HEIGHT_MIN 1
-#define VISUAL_CONFIRM_RECT_HEIGHT_MAX 10
 
 #define DC_LOGGER_INIT(logger)
 
@@ -811,8 +808,6 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx)
 	struct rect surf_clip = plane_state->clip_rect;
 	bool split_tb = stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM;
 	int split_count, split_idx;
-	struct dpp *dpp = pipe_ctx->plane_res.dpp;
-	unsigned short visual_confirm_rect_height = VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT;
 
 	calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx);
 	if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE)
@@ -881,18 +876,6 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx)
 				data->recout.width = data->h_active - data->recout.x;
 		}
 	}
-
-	/* Check bounds to ensure the VC bar height was set to a sane value */
-	if (dpp != NULL) {
-		if ((dpp->ctx->dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_RECT_HEIGHT_MIN) &&
-			(dpp->ctx->dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_RECT_HEIGHT_MAX)) {
-			visual_confirm_rect_height = dpp->ctx->dc->debug.visual_confirm_rect_height;
-		}
-
-		if (dpp->ctx->dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE)
-			data->recout.height = data->recout.height -
-					2 * (dpp->inst + visual_confirm_rect_height);
-	}
 }
 
 static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
index 7e140c35a0ced..b33955928bd0b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c
@@ -39,6 +39,9 @@
 #define BLACK_OFFSET_RGB_Y 0x0
 #define BLACK_OFFSET_CBCR  0x8000
 
+#define VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT 3
+#define VISUAL_CONFIRM_RECT_HEIGHT_MIN 1
+#define VISUAL_CONFIRM_RECT_HEIGHT_MAX 10
 
 #define REG(reg)\
 	dpp->tf_regs->reg
@@ -588,6 +591,18 @@ static void dpp1_dscl_set_manual_ratio_init(
 static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp,
 				 const struct rect *recout)
 {
+	int visual_confirm_on = 0;
+	unsigned short visual_confirm_rect_height = VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT;
+
+	if (dpp->base.ctx->dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE)
+		visual_confirm_on = 1;
+
+	/* Check bounds to ensure the VC bar height was set to a sane value */
+	if ((dpp->base.ctx->dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_RECT_HEIGHT_MIN) &&
+			(dpp->base.ctx->dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_RECT_HEIGHT_MAX)) {
+		visual_confirm_rect_height = dpp->base.ctx->dc->debug.visual_confirm_rect_height;
+	}
+
 	REG_SET_2(RECOUT_START, 0,
 		  /* First pixel of RECOUT in the active OTG area */
 		  RECOUT_START_X, recout->x,
@@ -598,7 +613,8 @@ static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp,
 		  /* Number of RECOUT horizontal pixels */
 		  RECOUT_WIDTH, recout->width,
 		  /* Number of RECOUT vertical lines */
-		  RECOUT_HEIGHT, recout->height);
+		  RECOUT_HEIGHT, recout->height
+			 - visual_confirm_on * 2 * (dpp->base.inst + visual_confirm_rect_height));
 }
 
 /**
-- 
GitLab


From 124155c0bd4a4ed822c1ba246bdf5123b42c3ad9 Mon Sep 17 00:00:00 2001
From: George Shen <george.shen@amd.com>
Date: Thu, 13 Apr 2023 19:03:59 -0400
Subject: [PATCH 0944/1791] drm/amd/display: Add w/a to disable DP dual mode on
 certain ports

[Why]
Certain ports on DCN3.2 configs do not properly populate the BIOS
info table flag to indicate DP dual mode is unsupported.

[How]
Add a workaround to disable DP dual mode on the ports with the missing
BIOS info table flag.

Reviewed-by: Michael Strauss <Michael.Strauss@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: George Shen <george.shen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                   | 1 +
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c | 1 +
 drivers/gpu/drm/amd/display/dc/link/link_detection.c  | 4 ++++
 3 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index fea68383bb61f..fe60816653d0b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -887,6 +887,7 @@ struct dc_debug_options {
 	bool override_odm_optimization;
 	bool minimize_dispclk_using_odm;
 	bool disable_subvp_high_refresh;
+	bool disable_dp_plus_plus_wa;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index a493238858745..4de2f8813dce3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -729,6 +729,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
 	.disable_subvp_high_refresh = true,
+	.disable_dp_plus_plus_wa = true,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index a131e30fd7d6a..17904de4f155b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -593,6 +593,10 @@ static bool detect_dp(struct dc_link *link,
 			/* DP SST branch */
 			link->type = dc_connection_sst_branch;
 	} else {
+		if (link->dc->debug.disable_dp_plus_plus_wa &&
+				link->link_enc->features.flags.bits.IS_UHBR20_CAPABLE)
+			return false;
+
 		/* DP passive dongles */
 		sink_caps->signal = dp_passive_dongle_detection(link->ddc,
 								sink_caps,
-- 
GitLab


From 6d7d0a4bc39240a2dd84d122d4ed5487e6acddf7 Mon Sep 17 00:00:00 2001
From: Anthony Koo <Anthony.Koo@amd.com>
Date: Sat, 22 Apr 2023 23:09:38 -0400
Subject: [PATCH 0945/1791] drm/amd/display: [FW Promotion] Release 0.0.165.0

- Add dmub boot options to disable ips states on init

Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Anthony Koo <Anthony.Koo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 09d4d0d9b92c3..af1f50742371f 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -398,6 +398,12 @@ enum dmub_lvtma_status_bit {
 	DMUB_LVTMA_STATUS_BIT_EDP_ON = (1 << 1),
 };
 
+enum dmub_ips_disable_type {
+	DMUB_IPS_DISABLE_IPS1 = 1,
+	DMUB_IPS_DISABLE_IPS2 = 2,
+	DMUB_IPS_DISABLE_IPS2_Z10 = 3,
+};
+
 /**
  * union dmub_fw_boot_options - Boot option definitions for SCRATCH14
  */
@@ -423,7 +429,9 @@ union dmub_fw_boot_options {
 		uint32_t usb4_dpia_bw_alloc_supported: 1; /* 1 if USB4 dpia BW allocation supported */
 		uint32_t disable_clk_ds: 1; /* 1 if disallow dispclk_ds and dppclk_ds*/
 		uint32_t disable_timeout_recovery : 1; /* 1 if timeout recovery should be disabled */
-		uint32_t reserved : 13; /**< reserved */
+		uint32_t ips_pg_disable: 1; /* 1 to disable ONO domains power gating*/
+		uint32_t ips_disable: 2; /* options to disable ips support*/
+		uint32_t reserved : 10; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
-- 
GitLab


From 4ab9d5848c728e5339e382f678a5e580573b3a3a Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 23 Apr 2023 22:12:05 -0400
Subject: [PATCH 0946/1791] drm/amd/display: 3.2.234

This version brings along following fixes:
- FW Release 0.0.165.0
- Add w/a to disable DP dual mode on certain ports
- Revert "Update scaler recout data for visual confirm"
- Filter out invalid bits in pipe_fuses
- Adding debug option to override Z8 watermark values
- Change default Z8 watermark values
- Workaround wrong HDR colorimetry with some receivers

Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index fe60816653d0b..7e3f20a3a02f1 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.233"
+#define DC_VER "3.2.234"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From 1156e1a60f024ce29fbb88dd330c2be81c4303ea Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 24 Apr 2023 14:24:20 -0400
Subject: [PATCH 0947/1791] drm/amdgpu: add [en/dis]able_kgq() functions

To replace the IP specific variants which are largely
duplicate.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 68 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  2 +
 2 files changed, 70 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 66b9740ec3762..90f5d302d5f3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -519,6 +519,39 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id)
 	return r;
 }
 
+int amdgpu_gfx_disable_kgq(struct amdgpu_device *adev, int xcc_id)
+{
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
+	struct amdgpu_ring *kiq_ring = &kiq->ring;
+	int i, r = 0;
+	int j;
+
+	if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues)
+		return -EINVAL;
+
+	spin_lock(&kiq->ring_lock);
+	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
+		if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
+						adev->gfx.num_gfx_rings)) {
+			spin_unlock(&kiq->ring_lock);
+			return -ENOMEM;
+		}
+
+		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+			j = i + xcc_id * adev->gfx.num_gfx_rings;
+			kiq->pmf->kiq_unmap_queues(kiq_ring,
+						   &adev->gfx.gfx_ring[i],
+						   PREEMPT_QUEUES, 0, 0);
+		}
+	}
+
+	if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang)
+		r = amdgpu_ring_test_helper(kiq_ring);
+	spin_unlock(&kiq->ring_lock);
+
+	return r;
+}
+
 int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev,
 					int queue_bit)
 {
@@ -590,6 +623,41 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 	return r;
 }
 
+int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id)
+{
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
+	struct amdgpu_ring *kiq_ring = &kiq->ring;
+	int r, i, j;
+
+	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
+		return -EINVAL;
+
+	spin_lock(&kiq->ring_lock);
+	/* No need to map kcq on the slave */
+	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
+		r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
+						adev->gfx.num_gfx_rings);
+		if (r) {
+			DRM_ERROR("Failed to lock KIQ (%d).\n", r);
+			spin_unlock(&adev->gfx.kiq[0].ring_lock);
+			return r;
+		}
+
+		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+			j = i + xcc_id * adev->gfx.num_gfx_rings;
+			kiq->pmf->kiq_map_queues(kiq_ring,
+						 &adev->gfx.gfx_ring[i]);
+		}
+	}
+
+	r = amdgpu_ring_test_helper(kiq_ring);
+	spin_unlock(&kiq->ring_lock);
+	if (r)
+		DRM_ERROR("KCQ enable failed\n");
+
+	return r;
+}
+
 /* amdgpu_gfx_off_ctrl - Handle gfx off feature enable/disable
  *
  * @adev: amdgpu_device pointer
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index f1a2ce7b6acaf..2755f00ac19a2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -433,6 +433,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id);
+int amdgpu_gfx_disable_kgq(struct amdgpu_device *adev, int xcc_id);
+int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id);
 
 void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev);
 void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev);
-- 
GitLab


From f39c25357f0c9ada9ed9676eaa7681d4583dda63 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 24 Apr 2023 14:31:00 -0400
Subject: [PATCH 0948/1791] drm/amdgpu/gfx10: use generic [en/dis]able_kgq()
 helpers

And remove the duplicate local variants.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 48 ++------------------------
 1 file changed, 2 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 25be4485dcd50..24d7134228b06 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -6462,28 +6462,6 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 	return 0;
 }
 
-static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev)
-{
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
-	int r, i;
-
-	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
-		return -EINVAL;
-
-	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
-					adev->gfx.num_gfx_rings);
-	if (r) {
-		DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-		return r;
-	}
-
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
-		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]);
-
-	return amdgpu_ring_test_helper(kiq_ring);
-}
-
 static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 {
 	int r, i;
@@ -6507,7 +6485,7 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = gfx_v10_0_kiq_enable_kgq(adev);
+	r = amdgpu_gfx_enable_kgq(adev, 0);
 	if (r)
 		goto done;
 
@@ -7178,28 +7156,6 @@ static int gfx_v10_0_hw_init(void *handle)
 	return r;
 }
 
-static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev)
-{
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
-	struct amdgpu_ring *kiq_ring = &kiq->ring;
-	int i;
-
-	if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues)
-		return -EINVAL;
-
-	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
-					adev->gfx.num_gfx_rings))
-		return -ENOMEM;
-
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
-		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i],
-					   PREEMPT_QUEUES, 0, 0);
-	if (!adev->job_hang)
-		return amdgpu_ring_test_helper(kiq_ring);
-	else
-		return 0;
-}
-
 static int gfx_v10_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -7210,7 +7166,7 @@ static int gfx_v10_0_hw_fini(void *handle)
 
 	if (!adev->no_hw_access) {
 		if (amdgpu_async_gfx_ring) {
-			r = gfx_v10_0_kiq_disable_kgq(adev);
+			r = amdgpu_gfx_disable_kgq(adev, 0);
 			if (r)
 				DRM_ERROR("KGQ disable failed\n");
 		}
-- 
GitLab


From 83033f72a40b5035c78f847fa4cc55fba633b82a Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 24 Apr 2023 14:33:15 -0400
Subject: [PATCH 0949/1791] drm/amdgpu/gfx11: use generic [en/dis]able_kgq()
 helpers

And remove the duplicate local variants.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 49 ++------------------------
 1 file changed, 2 insertions(+), 47 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 646003b2faf82..d36d365cb5825 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -3710,28 +3710,6 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 	return 0;
 }
 
-static int gfx_v11_0_kiq_enable_kgq(struct amdgpu_device *adev)
-{
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
-	int r, i;
-
-	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
-		return -EINVAL;
-
-	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
-					adev->gfx.num_gfx_rings);
-	if (r) {
-		DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-		return r;
-	}
-
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
-		kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]);
-
-	return amdgpu_ring_test_helper(kiq_ring);
-}
-
 static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 {
 	int r, i;
@@ -3755,7 +3733,7 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 			goto done;
 	}
 
-	r = gfx_v11_0_kiq_enable_kgq(adev);
+	r = amdgpu_gfx_enable_kgq(adev, 0);
 	if (r)
 		goto done;
 
@@ -4392,29 +4370,6 @@ static int gfx_v11_0_hw_init(void *handle)
 	return r;
 }
 
-static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev)
-{
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
-	struct amdgpu_ring *kiq_ring = &kiq->ring;
-	int i, r = 0;
-
-	if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues)
-		return -EINVAL;
-
-	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
-					adev->gfx.num_gfx_rings))
-		return -ENOMEM;
-
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++)
-		kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i],
-					   PREEMPT_QUEUES, 0, 0);
-
-	if (adev->gfx.kiq[0].ring.sched.ready)
-		r = amdgpu_ring_test_helper(kiq_ring);
-
-	return r;
-}
-
 static int gfx_v11_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -4426,7 +4381,7 @@ static int gfx_v11_0_hw_fini(void *handle)
 
 	if (!adev->no_hw_access) {
 		if (amdgpu_async_gfx_ring) {
-			r = gfx_v11_0_kiq_disable_kgq(adev);
+			r = amdgpu_gfx_disable_kgq(adev, 0);
 			if (r)
 				DRM_ERROR("KGQ disable failed\n");
 		}
-- 
GitLab


From 5a0510d58b635ed5bc2555ba1255cd2b34e033ee Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Fri, 28 Apr 2023 16:42:03 -0500
Subject: [PATCH 0950/1791] drm/amd: Downgrade message about watermarks table
 after s0i3 to debug

This message shows up on s0i3 resume for DCN31 and DCN314 platforms but
it has been decided that this flow won't be changed and the message is
expected behavior.

Downgrade the message to debug.

Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c
index 0827c7df28557..32279c5db7248 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c
@@ -130,7 +130,7 @@ static int dcn31_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr,
 	if (result == VBIOSSMC_Result_Failed) {
 		if (msg_id == VBIOSSMC_MSG_TransferTableDram2Smu &&
 		    param == TABLE_WATERMARKS)
-			DC_LOG_WARNING("Watermarks table not configured properly by SMU");
+			DC_LOG_DEBUG("Watermarks table not configured properly by SMU");
 		else
 			ASSERT(0);
 		REG_WRITE(MP1_SMN_C2PMSG_91, VBIOSSMC_Result_OK);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c
index 0765334f08259..07baa10a86473 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c
@@ -145,7 +145,7 @@ static int dcn314_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr,
 	if (result == VBIOSSMC_Result_Failed) {
 		if (msg_id == VBIOSSMC_MSG_TransferTableDram2Smu &&
 		    param == TABLE_WATERMARKS)
-			DC_LOG_WARNING("Watermarks table not configured properly by SMU");
+			DC_LOG_DEBUG("Watermarks table not configured properly by SMU");
 		else if (msg_id == VBIOSSMC_MSG_SetHardMinDcfclkByFreq ||
 			 msg_id == VBIOSSMC_MSG_SetMinDeepSleepDcfclk)
 			DC_LOG_WARNING("DCFCLK_DPM is not enabled by BIOS");
-- 
GitLab


From d78e816a3d615073a48bf8cc438790bc08160e0e Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:42:09 -0400
Subject: [PATCH 0951/1791] drm/amdgpu/gfx10: drop unused variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just check the return value directly.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 24d7134228b06..5c67c91c4297d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -7159,15 +7159,13 @@ static int gfx_v10_0_hw_init(void *handle)
 static int gfx_v10_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int r;
 
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
 	if (!adev->no_hw_access) {
 		if (amdgpu_async_gfx_ring) {
-			r = amdgpu_gfx_disable_kgq(adev, 0);
-			if (r)
+			if (amdgpu_gfx_disable_kgq(adev, 0))
 				DRM_ERROR("KGQ disable failed\n");
 		}
 
-- 
GitLab


From b848fe65f856193fc4f567c84f559f60ff66a3db Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:42:53 -0400
Subject: [PATCH 0952/1791] drm/amdgpu/gfx11: drop unused variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just check the return value directly.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index d36d365cb5825..256014a8c8248 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -4373,7 +4373,6 @@ static int gfx_v11_0_hw_init(void *handle)
 static int gfx_v11_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int r;
 
 	amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
@@ -4381,8 +4380,7 @@ static int gfx_v11_0_hw_fini(void *handle)
 
 	if (!adev->no_hw_access) {
 		if (amdgpu_async_gfx_ring) {
-			r = amdgpu_gfx_disable_kgq(adev, 0);
-			if (r)
+			if (amdgpu_gfx_disable_kgq(adev, 0))
 				DRM_ERROR("KGQ disable failed\n");
 		}
 
-- 
GitLab


From 42cdf6f687daa1a1992db400e8311e1e49c787fb Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:44:43 -0400
Subject: [PATCH 0953/1791] drm/amdgpu/gfx8: always restore kcq MQDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always restore the MQD not just when we do a reset.
This allows us to move the MQD to VRAM if we want.

v2: always reset ring pointer as well (Christian)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 5de44d7e92de0..2ae7f167985fe 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4651,15 +4651,13 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct vi_mqd_allocation));
-	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
-		/* reset MQD to a clean status */
+	} else {
+		/* restore MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation));
 		/* reset ring buffer */
 		ring->wptr = 0;
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 	return 0;
 }
-- 
GitLab


From 45b54a7dd3437632352ed28163e982233ef190a8 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:49:00 -0400
Subject: [PATCH 0954/1791] drm/amdgpu/gfx9: always restore kcq MQDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always restore the MQD not just when we do a reset.
This allows us to move the MQD to VRAM if we want.

v2: always reset ring pointer as well (Christian)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   | 7 ++-----
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 7 ++-----
 2 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 8fb027cf1bfb7..4a75560994694 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -3589,17 +3589,14 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
-	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
-		/* reset MQD to a clean status */
+	} else {
+		/* restore MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
-
 		/* reset ring buffer */
 		ring->wptr = 0;
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0);
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 312491455382a..b5924543e363e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1799,17 +1799,14 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
-	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
-		/* reset MQD to a clean status */
+	} else {
+		/* restore MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
-
 		/* reset ring buffer */
 		ring->wptr = 0;
 		atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0);
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
-- 
GitLab


From 2dbaf83998fe4772c83e5060cfbc21808758bb4b Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:51:34 -0400
Subject: [PATCH 0955/1791] drm/amdgpu/gfx10: always restore kcq/kgq MQDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always restore the MQD not just when we do a reset.
This allows us to move the MQD to VRAM if we want.

v2: always reset ring pointer as well (Christian)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 5c67c91c4297d..2e0234b43f433 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -6447,16 +6447,14 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring)
 		mutex_unlock(&adev->srbm_mutex);
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (amdgpu_in_reset(adev)) {
-		/* reset mqd with the backup copy */
+	} else {
+		/* restore mqd with the backup copy */
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd));
 		/* reset the ring */
 		ring->wptr = 0;
 		*ring->wptr_cpu_addr = 0;
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
@@ -6780,17 +6778,14 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
-		/* reset MQD to a clean status */
+	} else {
+		/* restore MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
-
 		/* reset ring buffer */
 		ring->wptr = 0;
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0);
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
-- 
GitLab


From dae343b343ff741d727312b2a9b03d86e64b31c5 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 1 May 2023 16:31:53 +0200
Subject: [PATCH 0956/1791] drm/amd/display: mark
 amdgpu_dm_connector_funcs_force static

A global function without a header prototype has made it into
linux-next during the merge window:

drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:6339:6: error: no previous prototype for 'amdgpu_dm_connector_funcs_force' [-Werror=missing-prototypes]

Mark the function static instead, as there are no other
callers outside this file.

Fixes: 0ba4a784a145 ("drm/amd/display: implement force function in amdgpu_dm_connector_funcs")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304251640.JClqTim9-lkp@intel.com/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index dd56e1512b0ae..e278474e4a5b6 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6340,7 +6340,7 @@ amdgpu_dm_connector_late_register(struct drm_connector *connector)
 	return 0;
 }
 
-void amdgpu_dm_connector_funcs_force(struct drm_connector *connector)
+static void amdgpu_dm_connector_funcs_force(struct drm_connector *connector)
 {
 	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
 	struct dc_link *dc_link = aconnector->dc_link;
-- 
GitLab


From c6fa6fe9ebd5cdde0e41a558ff3efea8fe988e77 Mon Sep 17 00:00:00 2001
From: Thong Thai <thong.thai@amd.com>
Date: Mon, 1 May 2023 11:04:36 -0400
Subject: [PATCH 0957/1791] drm/amdgpu/nv: update VCN 3 max HEVC encoding
 resolution

Update the maximum resolution reported for HEVC encoding on VCN 3
devices to reflect its 8K encoding capability.

v2: Also update the max height for H.264 encoding to match spec.
(Ruijing)

Signed-off-by: Thong Thai <thong.thai@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nv.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index 3cc068974bcdb..a5f76c9538c4c 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -98,6 +98,16 @@ static const struct amdgpu_video_codecs nv_video_codecs_decode =
 };
 
 /* Sienna Cichlid */
+static const struct amdgpu_video_codec_info sc_video_codecs_encode_array[] = {
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2160, 0)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 7680, 4352, 0)},
+};
+
+static const struct amdgpu_video_codecs sc_video_codecs_encode = {
+	.codec_count = ARRAY_SIZE(sc_video_codecs_encode_array),
+	.codec_array = sc_video_codecs_encode_array,
+};
+
 static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn0[] =
 {
 	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)},
@@ -136,8 +146,8 @@ static const struct amdgpu_video_codecs sc_video_codecs_decode_vcn1 =
 /* SRIOV Sienna Cichlid, not const since data is controlled by host */
 static struct amdgpu_video_codec_info sriov_sc_video_codecs_encode_array[] =
 {
-	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2304, 0)},
-	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 4096, 2304, 0)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2160, 0)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 7680, 4352, 0)},
 };
 
 static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn0[] =
@@ -237,12 +247,12 @@ static int nv_query_video_codecs(struct amdgpu_device *adev, bool encode,
 		} else {
 			if (adev->vcn.harvest_config & AMDGPU_VCN_HARVEST_VCN0) {
 				if (encode)
-					*codecs = &nv_video_codecs_encode;
+					*codecs = &sc_video_codecs_encode;
 				else
 					*codecs = &sc_video_codecs_decode_vcn1;
 			} else {
 				if (encode)
-					*codecs = &nv_video_codecs_encode;
+					*codecs = &sc_video_codecs_encode;
 				else
 					*codecs = &sc_video_codecs_decode_vcn0;
 			}
@@ -251,14 +261,14 @@ static int nv_query_video_codecs(struct amdgpu_device *adev, bool encode,
 	case IP_VERSION(3, 0, 16):
 	case IP_VERSION(3, 0, 2):
 		if (encode)
-			*codecs = &nv_video_codecs_encode;
+			*codecs = &sc_video_codecs_encode;
 		else
 			*codecs = &sc_video_codecs_decode_vcn0;
 		return 0;
 	case IP_VERSION(3, 1, 1):
 	case IP_VERSION(3, 1, 2):
 		if (encode)
-			*codecs = &nv_video_codecs_encode;
+			*codecs = &sc_video_codecs_encode;
 		else
 			*codecs = &yc_video_codecs_decode;
 		return 0;
-- 
GitLab


From 09d8a67912462a07cb7ff40956ed8c79b1c74564 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 14:54:27 -0400
Subject: [PATCH 0958/1791] drm/amdgpu/gfx11: always restore kcq/kgq MQDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always restore the MQD not just when we do a reset.
This allows us to move the MQD to VRAM if we want.

v2: always reset ring pointer as well (Christian)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 256014a8c8248..f09e2558e73b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -3695,16 +3695,14 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring)
 		mutex_unlock(&adev->srbm_mutex);
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (amdgpu_in_reset(adev)) {
-		/* reset mqd with the backup copy */
+	} else {
+		/* restore mqd with the backup copy */
 		if (adev->gfx.me.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd));
 		/* reset the ring */
 		ring->wptr = 0;
 		*ring->wptr_cpu_addr = 0;
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
@@ -4043,17 +4041,14 @@ static int gfx_v11_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
-	} else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */
-		/* reset MQD to a clean status */
+	} else {
+		/* restore MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
-
 		/* reset ring buffer */
 		ring->wptr = 0;
 		atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0);
 		amdgpu_ring_clear_ring(ring);
-	} else {
-		amdgpu_ring_clear_ring(ring);
 	}
 
 	return 0;
-- 
GitLab


From 0e768043bf68dc9713de34780363c595e54af7fa Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Thu, 27 Apr 2023 14:01:05 +0800
Subject: [PATCH 0959/1791] drm/amdgpu: set gfx9 onwards APU atomics support to
 be true

APUs w/ gfx9 onwards doesn't reply on PCIe atomics, rather
it is internal path w/ native atomic support. Set have_atomics_support
to true.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Lang Yu <lang.yu@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 1787602fe582a..ae0bcffa25910 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3758,6 +3758,12 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *)
 			adev->virt.fw_reserve.p_pf2vf)->pcie_atomic_ops_support_flags ==
 			(PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64);
+	/* APUs w/ gfx9 onwards doesn't reply on PCIe atomics, rather it is a
+	 * internal path natively support atomics, set have_atomics_support to true.
+	 */
+	else if ((adev->flags & AMD_IS_APU) &&
+		(adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0)))
+		adev->have_atomics_support = true;
 	else
 		adev->have_atomics_support =
 			!pci_enable_atomic_ops_to_root(adev->pdev,
-- 
GitLab


From 9e690184586bfb88efa176cdf912414f6c53519c Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 3 May 2023 14:10:20 +0530
Subject: [PATCH 0960/1791] drm/amd/amdgpu: Fix errors & warnings in amdgpu
 _bios, _cs, _dma_buf, _fence.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The following checkpatch errors & warning is removed.

ERROR: else should follow close brace '}'
ERROR: trailing statements should be on next line
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: Possible repeated word: 'Fences'
WARNING: Missing a blank line after declarations
WARNING: braces {} are not necessary for single statement blocks
WARNING: Comparisons should place the constant on the right side of the test
WARNING: printk() should include KERN_<LEVEL> facility level

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c    | 16 +++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c      | 16 ++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c   | 14 ++++++++------
 4 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
index 30c28a69e847d..b582b83c4984f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
@@ -104,9 +104,8 @@ static bool igp_read_bios_from_vram(struct amdgpu_device *adev)
 	adev->bios = NULL;
 	vram_base = pci_resource_start(adev->pdev, 0);
 	bios = ioremap_wc(vram_base, size);
-	if (!bios) {
+	if (!bios)
 		return false;
-	}
 
 	adev->bios = kmalloc(size, GFP_KERNEL);
 	if (!adev->bios) {
@@ -133,9 +132,8 @@ bool amdgpu_read_bios(struct amdgpu_device *adev)
 	adev->bios = NULL;
 	/* XXX: some cards may return 0 for rom size? ddx has a workaround */
 	bios = pci_map_rom(adev->pdev, &size);
-	if (!bios) {
+	if (!bios)
 		return false;
-	}
 
 	adev->bios = kzalloc(size, GFP_KERNEL);
 	if (adev->bios == NULL) {
@@ -168,9 +166,9 @@ static bool amdgpu_read_bios_from_rom(struct amdgpu_device *adev)
 	header[AMD_VBIOS_SIGNATURE_END] = 0;
 
 	if ((!AMD_IS_VALID_VBIOS(header)) ||
-	    0 != memcmp((char *)&header[AMD_VBIOS_SIGNATURE_OFFSET],
-			AMD_VBIOS_SIGNATURE,
-			strlen(AMD_VBIOS_SIGNATURE)))
+		memcmp((char *)&header[AMD_VBIOS_SIGNATURE_OFFSET],
+		       AMD_VBIOS_SIGNATURE,
+		       strlen(AMD_VBIOS_SIGNATURE)) != 0)
 		return false;
 
 	/* valid vbios, go on */
@@ -264,7 +262,7 @@ static int amdgpu_atrm_call(acpi_handle atrm_handle, uint8_t *bios,
 
 	status = acpi_evaluate_object(atrm_handle, NULL, &atrm_arg, &buffer);
 	if (ACPI_FAILURE(status)) {
-		printk("failed to evaluate ATRM got %s\n", acpi_format_exception(status));
+		DRM_ERROR("failed to evaluate ATRM got %s\n", acpi_format_exception(status));
 		return -ENODEV;
 	}
 
@@ -363,7 +361,7 @@ static bool amdgpu_acpi_vfct_bios(struct amdgpu_device *adev)
 	struct acpi_table_header *hdr;
 	acpi_size tbl_size;
 	UEFI_ACPI_VFCT *vfct;
-	unsigned offset;
+	unsigned int offset;
 
 	if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr)))
 		return false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index c3b3287dc29ea..5270089aee165 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -397,7 +397,7 @@ static int amdgpu_cs_p2_dependencies(struct amdgpu_cs_parser *p,
 {
 	struct drm_amdgpu_cs_chunk_dep *deps = chunk->kdata;
 	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
-	unsigned num_deps;
+	unsigned int num_deps;
 	int i, r;
 
 	num_deps = chunk->length_dw * 4 /
@@ -468,7 +468,7 @@ static int amdgpu_cs_p2_syncobj_in(struct amdgpu_cs_parser *p,
 				   struct amdgpu_cs_chunk *chunk)
 {
 	struct drm_amdgpu_cs_chunk_sem *deps = chunk->kdata;
-	unsigned num_deps;
+	unsigned int num_deps;
 	int i, r;
 
 	num_deps = chunk->length_dw * 4 /
@@ -486,7 +486,7 @@ static int amdgpu_cs_p2_syncobj_timeline_wait(struct amdgpu_cs_parser *p,
 					      struct amdgpu_cs_chunk *chunk)
 {
 	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps = chunk->kdata;
-	unsigned num_deps;
+	unsigned int num_deps;
 	int i, r;
 
 	num_deps = chunk->length_dw * 4 /
@@ -506,7 +506,7 @@ static int amdgpu_cs_p2_syncobj_out(struct amdgpu_cs_parser *p,
 				    struct amdgpu_cs_chunk *chunk)
 {
 	struct drm_amdgpu_cs_chunk_sem *deps = chunk->kdata;
-	unsigned num_deps;
+	unsigned int num_deps;
 	int i;
 
 	num_deps = chunk->length_dw * 4 /
@@ -540,7 +540,7 @@ static int amdgpu_cs_p2_syncobj_timeline_signal(struct amdgpu_cs_parser *p,
 						struct amdgpu_cs_chunk *chunk)
 {
 	struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps = chunk->kdata;
-	unsigned num_deps;
+	unsigned int num_deps;
 	int i;
 
 	num_deps = chunk->length_dw * 4 /
@@ -758,6 +758,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev,
 
 		if (used_vis_vram < total_vis_vram) {
 			u64 free_vis_vram = total_vis_vram - used_vis_vram;
+
 			adev->mm_stats.accum_us_vis = min(adev->mm_stats.accum_us_vis +
 							  increment_us, us_upper_bound);
 
@@ -1076,9 +1077,8 @@ static int amdgpu_cs_patch_ibs(struct amdgpu_cs_parser *p,
 
 		/* the IB should be reserved at this point */
 		r = amdgpu_bo_kmap(aobj, (void **)&kptr);
-		if (r) {
+		if (r)
 			return r;
-		}
 
 		kptr += va_start - (m->start * AMDGPU_GPU_PAGE_SIZE);
 
@@ -1392,7 +1392,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 /* Cleanup the parser structure */
 static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser)
 {
-	unsigned i;
+	unsigned int i;
 
 	amdgpu_sync_free(&parser->sync);
 	for (i = 0; i < parser->num_post_deps; i++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index 0c001bb8fc2b0..6dceaf40625b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -149,7 +149,7 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach,
 	if (!bo->tbo.pin_count) {
 		/* move buffer into GTT or VRAM */
 		struct ttm_operation_ctx ctx = { false, false };
-		unsigned domains = AMDGPU_GEM_DOMAIN_GTT;
+		unsigned int domains = AMDGPU_GEM_DOMAIN_GTT;
 
 		if (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM &&
 		    attach->peer2peer) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index f52d0ba91a770..1994eafd3e71a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -42,7 +42,6 @@
 #include "amdgpu_reset.h"
 
 /*
- * Fences
  * Fences mark an event in the GPUs pipeline and are used
  * for GPU/CPU synchronization.  When the fence is written,
  * it is expected that all buffers associated with that fence
@@ -140,7 +139,7 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring)
  * Returns 0 on success, -ENOMEM on failure.
  */
 int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amdgpu_job *job,
-		      unsigned flags)
+		      unsigned int flags)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct dma_fence *fence;
@@ -174,11 +173,11 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd
 				       adev->fence_context + ring->idx, seq);
 			/* Against remove in amdgpu_job_{free, free_cb} */
 			dma_fence_get(fence);
-		}
-		else
+		} else {
 			dma_fence_init(fence, &amdgpu_fence_ops,
 				       &ring->fence_drv.lock,
 				       adev->fence_context + ring->idx, seq);
+		}
 	}
 
 	amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr,
@@ -396,7 +395,7 @@ signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring,
  * Returns the number of emitted fences on the ring.  Used by the
  * dynpm code to ring track activity.
  */
-unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring)
+unsigned int amdgpu_fence_count_emitted(struct amdgpu_ring *ring)
 {
 	uint64_t emitted;
 
@@ -475,7 +474,7 @@ void amdgpu_fence_update_start_timestamp(struct amdgpu_ring *ring, uint32_t seq,
  */
 int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 				   struct amdgpu_irq_src *irq_src,
-				   unsigned irq_type)
+				   unsigned int irq_type)
 {
 	struct amdgpu_device *adev = ring->adev;
 	uint64_t index;
@@ -653,6 +652,7 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev)
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
 		struct amdgpu_ring *ring = adev->rings[i];
+
 		if (!ring || !ring->fence_drv.initialized)
 			continue;
 
@@ -840,6 +840,7 @@ static int amdgpu_debugfs_fence_info_show(struct seq_file *m, void *unused)
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
 		struct amdgpu_ring *ring = adev->rings[i];
+
 		if (!ring || !ring->fence_drv.initialized)
 			continue;
 
@@ -913,6 +914,7 @@ static void amdgpu_debugfs_reset_work(struct work_struct *work)
 						  reset_work);
 
 	struct amdgpu_reset_context reset_context;
+
 	memset(&reset_context, 0, sizeof(reset_context));
 
 	reset_context.method = AMD_RESET_METHOD_NONE;
-- 
GitLab


From 967a66396e4668ef314e06a0391e67aa1cf6d058 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Wed, 26 Apr 2023 09:46:54 +0800
Subject: [PATCH 0961/1791] drm/amdgpu: drop redundant sched job cleanup when
 cs is aborted
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Once command submission failed due to userptr invalidation in
amdgpu_cs_submit, legacy code will perform cleanup of scheduler
job. However, it's not needed at all, as former commit has integrated
job cleanup stuff into amdgpu_job_free. Otherwise, because of double
free, a NULL pointer dereference will occur in such scenario.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2457
Fixes: f7d66fb2ea43 ("drm/amdgpu: cleanup scheduler job initialization v2")
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 5270089aee165..6e1d331af01ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1305,7 +1305,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 		r = drm_sched_job_add_dependency(&leader->base, fence);
 		if (r) {
 			dma_fence_put(fence);
-			goto error_cleanup;
+			return r;
 		}
 	}
 
@@ -1332,7 +1332,8 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	}
 	if (r) {
 		r = -EAGAIN;
-		goto error_unlock;
+		mutex_unlock(&p->adev->notifier_lock);
+		return r;
 	}
 
 	p->fence = dma_fence_get(&leader->base.s_fence->finished);
@@ -1379,14 +1380,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	mutex_unlock(&p->adev->notifier_lock);
 	mutex_unlock(&p->bo_list->bo_list_mutex);
 	return 0;
-
-error_unlock:
-	mutex_unlock(&p->adev->notifier_lock);
-
-error_cleanup:
-	for (i = 0; i < p->gang_size; ++i)
-		drm_sched_job_cleanup(&p->jobs[i]->base);
-	return r;
 }
 
 /* Cleanup the parser structure */
-- 
GitLab


From 1cfb4d6121276a829aa94d0e32a7f5e1830ebc21 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 26 Apr 2023 15:30:13 -0400
Subject: [PATCH 0962/1791] drm/amdgpu: put MQDs in VRAM

Reduces preemption latency.
Only enable this for gfx10 and 11 for now
to avoid changing behavior on gfx 8 and 9.

v2: move MES MQDs into VRAM as well (YuBiao)
v3: enable on gfx10, 11 only (Alex)
v4: minor style changes, document why gfx10/11 only (Alex)

Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 9 +++++++--
 drivers/gpu/drm/amd/amdgpu/mes_v10_1.c  | 1 +
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c  | 1 +
 3 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 90f5d302d5f3e..b91be56ba7739 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -382,6 +382,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 	int r, i, j;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	struct amdgpu_ring *ring = &kiq->ring;
+	u32 domain = AMDGPU_GEM_DOMAIN_GTT;
+
+	/* Only enable on gfx10 and 11 for now to avoid changing behavior on older chips */
+	if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 0, 0))
+		domain |= AMDGPU_GEM_DOMAIN_VRAM;
 
 	/* create MQD for KIQ */
 	if (!adev->enable_mes_kiq && !ring->mqd_obj) {
@@ -413,7 +418,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 			ring = &adev->gfx.gfx_ring[i];
 			if (!ring->mqd_obj) {
 				r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
-							    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
+							    domain, &ring->mqd_obj,
 							    &ring->mqd_gpu_addr, &ring->mqd_ptr);
 				if (r) {
 					dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r);
@@ -435,7 +440,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 		ring = &adev->gfx.compute_ring[j];
 		if (!ring->mqd_obj) {
 			r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
-						    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
+						    domain, &ring->mqd_obj,
 						    &ring->mqd_gpu_addr, &ring->mqd_ptr);
 			if (r) {
 				dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r);
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index 0599f8a6813e9..4560476c7c317 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -901,6 +901,7 @@ static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev,
 		return 0;
 
 	r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
+				    AMDGPU_GEM_DOMAIN_VRAM |
 				    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
 				    &ring->mqd_gpu_addr, &ring->mqd_ptr);
 	if (r) {
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index e853bcb892fcb..3adb450eec072 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -999,6 +999,7 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev,
 		return 0;
 
 	r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
+				    AMDGPU_GEM_DOMAIN_VRAM |
 				    AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj,
 				    &ring->mqd_gpu_addr, &ring->mqd_ptr);
 	if (r) {
-- 
GitLab


From 29551fd90ee69771f499071c1119c1007febbd1d Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 27 Apr 2023 14:09:33 -0400
Subject: [PATCH 0963/1791] drm/amdgpu: drop invalid IP revision
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This was already fixed and dropped in:
commit baf3f8f37406 ("drm/amdgpu: handle SRIOV VCN revision parsing")
commit c40bdfb2ffa4 ("drm/amdgpu: fix incorrect VCN revision in SRIOV")
But seems to have been accidently been left around in a merge.

Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 0ba013275dc17..76ceca05452eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1939,7 +1939,6 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev)
 		case IP_VERSION(3, 1, 1):
 		case IP_VERSION(3, 1, 2):
 		case IP_VERSION(3, 0, 2):
-		case IP_VERSION(3, 0, 192):
 			amdgpu_device_ip_block_add(adev, &vcn_v3_0_ip_block);
 			if (!amdgpu_sriov_vf(adev))
 				amdgpu_device_ip_block_add(adev, &jpeg_v3_0_ip_block);
-- 
GitLab


From 0e5f625157ca47d9ee27a36310611964bd97c605 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 27 Apr 2023 14:32:11 -0400
Subject: [PATCH 0964/1791] drm/amdgpu: drop unused function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

amdgpu_discovery_get_ip_version() has not been used since
commit c40bdfb2ffa4 ("drm/amdgpu: fix incorrect VCN revision in SRIOV")
so drop it.

Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 48 -------------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h |  2 -
 2 files changed, 50 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 76ceca05452eb..b58d94dc19246 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1208,54 +1208,6 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 	return 0;
 }
 
-int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int number_instance,
-				    int *major, int *minor, int *revision)
-{
-	struct binary_header *bhdr;
-	struct ip_discovery_header *ihdr;
-	struct die_header *dhdr;
-	struct ip *ip;
-	uint16_t die_offset;
-	uint16_t ip_offset;
-	uint16_t num_dies;
-	uint16_t num_ips;
-	int i, j;
-
-	if (!adev->mman.discovery_bin) {
-		DRM_ERROR("ip discovery uninitialized\n");
-		return -EINVAL;
-	}
-
-	bhdr = (struct binary_header *)adev->mman.discovery_bin;
-	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin +
-			le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
-	num_dies = le16_to_cpu(ihdr->num_dies);
-
-	for (i = 0; i < num_dies; i++) {
-		die_offset = le16_to_cpu(ihdr->die_info[i].die_offset);
-		dhdr = (struct die_header *)(adev->mman.discovery_bin + die_offset);
-		num_ips = le16_to_cpu(dhdr->num_ips);
-		ip_offset = die_offset + sizeof(*dhdr);
-
-		for (j = 0; j < num_ips; j++) {
-			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
-
-			if ((le16_to_cpu(ip->hw_id) == hw_id) && (ip->number_instance == number_instance)) {
-				if (major)
-					*major = ip->major;
-				if (minor)
-					*minor = ip->minor;
-				if (revision)
-					*revision = ip->revision;
-				return 0;
-			}
-			ip_offset += struct_size(ip, base_address, ip->num_base_address);
-		}
-	}
-
-	return -EINVAL;
-}
-
 static void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev)
 {
 	int vcn_harvest_count = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
index 8563dd4a7dc2e..63ec6924b907a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
@@ -28,8 +28,6 @@
 #define DISCOVERY_TMR_OFFSET    (64 << 10)
 
 void amdgpu_discovery_fini(struct amdgpu_device *adev);
-int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int number_instance,
-                                    int *major, int *minor, int *revision);
 int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev);
 
 #endif /* __AMDGPU_DISCOVERY__ */
-- 
GitLab


From 7a6a2e59aacd6d5e75927352b72d39a5bc447e09 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Wed, 3 May 2023 18:09:18 +0300
Subject: [PATCH 0965/1791] drm/amdgpu: unlock the correct lock in
 amdgpu_gfx_enable_kcq()

We changed which lock we are supposed to take but this error path
was accidentally over looked so it still drops the old lock.

Fixes: def799c6596d ("drm/amdgpu: add multi-xcc support to amdgpu_gfx interfaces (v4)")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index b91be56ba7739..76dd2841cc0df 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -605,7 +605,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 						kiq->pmf->set_resources_size);
 		if (r) {
 			DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-			spin_unlock(&adev->gfx.kiq[0].ring_lock);
+			spin_unlock(&kiq->ring_lock);
 			return r;
 		}
 
-- 
GitLab


From f4409a2361152f3480781a1dea1a3bd0d8369c78 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Wed, 3 May 2023 18:09:35 +0300
Subject: [PATCH 0966/1791] drm/amdgpu: unlock on error in
 gfx_v9_4_3_kiq_resume()

Smatch complains that we need to drop this lock before returning.

    drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1838 gfx_v9_4_3_kiq_resume()
    warn: inconsistent returns 'ring->mqd_obj->tbo.base.resv'.

Fixes: 86301129698b ("drm/amdgpu: split gc v9_4_3 functionality from gc v9_0")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index b5924543e363e..c9ae3065a13d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1824,8 +1824,10 @@ static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 		return r;
 
 	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
-	if (unlikely(r != 0))
+	if (unlikely(r != 0)) {
+		amdgpu_bo_unreserve(ring->mqd_obj);
 		return r;
+	}
 
 	gfx_v9_4_3_kiq_init_queue(ring, xcc_id);
 	amdgpu_bo_kunmap(ring->mqd_obj);
-- 
GitLab


From c0c27428903700d86920394aa2302506b5d95b17 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Tue, 2 May 2023 11:59:08 -0400
Subject: [PATCH 0967/1791] drm/amdgpu: fix an amdgpu_irq_put() issue in
 gmc_v9_0_hw_fini()

As made mention of in commit c56edea58c31 ("drm/amdgpu: fix
amdgpu_irq_put call trace in gmc_v10_0_hw_fini") and commit aa6ac247ed7d
("drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v11_0_hw_fini"). It
is meaningless to call amdgpu_irq_put() for gmc.ecc_irq. So, remove it
from gmc_v9_0_hw_fini().

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2522
Fixes: c8b5a95b5709 ("drm/amdgpu: Fix desktop freezed after gpu-reset")
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 290804a06e058..6ae5cee9b64be 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1999,7 +1999,6 @@ static int gmc_v9_0_hw_fini(void *handle)
 	if (adev->mmhub.funcs->update_power_gating)
 		adev->mmhub.funcs->update_power_gating(adev, false);
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 
 	return 0;
-- 
GitLab


From 4667fbe2f7c54628c48aedc8c8472cd12a5f7fcd Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 24 Nov 2021 17:24:58 +0800
Subject: [PATCH 0968/1791] drm/amdgpu: do gfxhub init for all XCDs

Each XCD needs to do gfxhub init

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 581 +++++++++++++----------
 1 file changed, 317 insertions(+), 264 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index c59c6c85fbff3..79af32bb078c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -43,19 +43,25 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 					 uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	int i;
 
-	WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
-			    hub->ctx_addr_distance * vmid,
-			    lower_32_bits(page_table_base));
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_SOC15_OFFSET(GC, i,
+				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+				    hub->ctx_addr_distance * vmid,
+				    lower_32_bits(page_table_base));
 
-	WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
-			    hub->ctx_addr_distance * vmid,
-			    upper_32_bits(page_table_base));
+		WREG32_SOC15_OFFSET(GC, i,
+				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+				    hub->ctx_addr_distance * vmid,
+				    upper_32_bits(page_table_base));
+	}
 }
 
 static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t pt_base;
+	int i;
 
 	if (adev->gmc.pdb0_bo)
 		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
@@ -67,26 +73,36 @@ static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 	/* If use GART for FB translation, vmid0 page table covers both
 	 * vram and system memory (gart)
 	 */
-	if (adev->gmc.pdb0_bo) {
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
-				(u32)(adev->gmc.fb_start >> 12));
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
-				(u32)(adev->gmc.fb_start >> 44));
-
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
-				(u32)(adev->gmc.gart_end >> 12));
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
-				(u32)(adev->gmc.gart_end >> 44));
-	} else {
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
-				(u32)(adev->gmc.gart_start >> 12));
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
-				(u32)(adev->gmc.gart_start >> 44));
-
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
-				(u32)(adev->gmc.gart_end >> 12));
-		WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
-				(u32)(adev->gmc.gart_end >> 44));
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		if (adev->gmc.pdb0_bo) {
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				     (u32)(adev->gmc.fb_start >> 12));
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				     (u32)(adev->gmc.fb_start >> 44));
+
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				     (u32)(adev->gmc.gart_end >> 12));
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				     (u32)(adev->gmc.gart_end >> 44));
+		} else {
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				     (u32)(adev->gmc.gart_start >> 12));
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				     (u32)(adev->gmc.gart_start >> 44));
+
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				     (u32)(adev->gmc.gart_end >> 12));
+			WREG32_SOC15(GC, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				     (u32)(adev->gmc.gart_end >> 44));
+		}
 	}
 }
 
@@ -94,160 +110,183 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t value;
 	uint32_t tmp;
+	int i;
 
-	/* Program the AGP BAR */
-	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BASE, 0);
-	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
-	WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
-
-	if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) {
-		/* Program the system aperture low logical page number. */
-		WREG32_SOC15_RLC(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
-			min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
-
-		if (adev->apu_flags & AMD_APU_IS_RAVEN2)
-			/*
-			* Raven2 has a HW issue that it is unable to use the
-			* vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR.
-			* So here is the workaround that increase system
-			* aperture high address (add 1) to get rid of the VM
-			* fault and hardware hang.
-			*/
-			WREG32_SOC15_RLC(GC, 0,
-					 regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
-					 max((adev->gmc.fb_end >> 18) + 0x1,
-					     adev->gmc.agp_end >> 18));
-		else
-			WREG32_SOC15_RLC(GC, 0,
-				regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
-				max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
-
-		/* Set default page address. */
-		value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
-		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
-			     (u32)(value >> 12));
-		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
-			     (u32)(value >> 44));
-
-		/* Program "protection fault". */
-		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
-			     (u32)(adev->dummy_page_addr >> 12));
-		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
-			     (u32)((u64)adev->dummy_page_addr >> 44));
-
-		tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2);
-		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
-				    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
-		WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
-	}
-
-	/* In the case squeezing vram into GART aperture, we don't use
-	 * FB aperture and AGP aperture. Disable them.
-	 */
-	if (adev->gmc.pdb0_bo) {
-		WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_TOP, 0);
-		WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
-		WREG32_SOC15(GC, 0, regMC_VM_AGP_TOP, 0);
-		WREG32_SOC15(GC, 0, regMC_VM_AGP_BOT, 0xFFFFFF);
-		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
-		WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		/* Program the AGP BAR */
+		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BASE, 0);
+		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
+		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
+
+		if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) {
+			/* Program the system aperture low logical page number. */
+			WREG32_SOC15_RLC(GC, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+				min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
+
+			if (adev->apu_flags & AMD_APU_IS_RAVEN2)
+				/*
+				* Raven2 has a HW issue that it is unable to use the
+				* vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR.
+				* So here is the workaround that increase system
+				* aperture high address (add 1) to get rid of the VM
+				* fault and hardware hang.
+				*/
+				WREG32_SOC15_RLC(GC, i,
+						 regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+						 max((adev->gmc.fb_end >> 18) + 0x1,
+						     adev->gmc.agp_end >> 18));
+			else
+				WREG32_SOC15_RLC(GC, i,
+					regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+					max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
+
+			/* Set default page address. */
+			value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+				     (u32)(value >> 12));
+			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+				     (u32)(value >> 44));
+
+			/* Program "protection fault". */
+			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+				     (u32)(adev->dummy_page_addr >> 12));
+			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+				     (u32)((u64)adev->dummy_page_addr >> 44));
+
+			tmp = RREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL2);
+			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
+					    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+		}
+
+		/* In the case squeezing vram into GART aperture, we don't use
+		 * FB aperture and AGP aperture. Disable them.
+		 */
+		if (adev->gmc.pdb0_bo) {
+			WREG32_SOC15(GC, i, regMC_VM_FB_LOCATION_TOP, 0);
+			WREG32_SOC15(GC, i, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
+			WREG32_SOC15(GC, i, regMC_VM_AGP_TOP, 0);
+			WREG32_SOC15(GC, i, regMC_VM_AGP_BOT, 0xFFFFFF);
+			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
+			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+		}
 	}
 }
 
 static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
-	/* Setup TLB control */
-	tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL);
-
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    ENABLE_ADVANCED_DRIVER_MODEL, 1);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    MTYPE, MTYPE_UC);/* XXX for emulation. */
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
-
-	WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		/* Setup TLB control */
+		tmp = RREG32_SOC15(GC, i, regMC_VM_MX_L1_TLB_CNTL);
+
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    ENABLE_L1_TLB, 1);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    SYSTEM_ACCESS_MODE, 3);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    ENABLE_ADVANCED_DRIVER_MODEL, 1);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    MTYPE, MTYPE_UC);/* XXX for emulation. */
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
+
+		WREG32_SOC15_RLC(GC, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+	}
 }
 
 static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
-	/* Setup L2 cache */
-	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
-	/* XXX for emulation, Refer to closed source code.*/
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
-			    0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
-	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL, tmp);
-
-	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL2);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
-	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL2, tmp);
-
-	tmp = regVM_L2_CNTL3_DEFAULT;
-	if (adev->gmc.translate_further) {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
-				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
-	} else {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
-				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
-	}
-	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL3, tmp);
-
-	tmp = regVM_L2_CNTL4_DEFAULT;
-	if (adev->gmc.xgmi.connected_to_cpu) {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
-	} else {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		/* Setup L2 cache */
+		tmp = RREG32_SOC15(GC, i, regVM_L2_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
+		/* XXX for emulation, Refer to closed source code.*/
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
+				    0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
+		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL, tmp);
+
+		tmp = RREG32_SOC15(GC, i, regVM_L2_CNTL2);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
+		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL2, tmp);
+
+		tmp = regVM_L2_CNTL3_DEFAULT;
+		if (adev->gmc.translate_further) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+					    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+		} else {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+					    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+		}
+		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL3, tmp);
+
+		tmp = regVM_L2_CNTL4_DEFAULT;
+		if (adev->gmc.xgmi.connected_to_cpu) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+		} else {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+		}
+		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL4, tmp);
 	}
-	WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL4, tmp);
 }
 
 static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
-	tmp = RREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
-			adev->gmc.vmid0_page_table_depth);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
-			adev->gmc.vmid0_page_table_block_size);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
-			    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
-	WREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL, tmp);
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		tmp = RREG32_SOC15(GC, i, regVM_CONTEXT0_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
+				adev->gmc.vmid0_page_table_depth);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
+				adev->gmc.vmid0_page_table_block_size);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+		WREG32_SOC15(GC, i, regVM_CONTEXT0_CNTL, tmp);
+	}
 }
 
 static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 {
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
-		     0XFFFFFFFF);
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
-		     0x0000000F);
-
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
-		     0);
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
-		     0);
-
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
-	WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+	int i;
 
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
+			     0XFFFFFFFF);
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
+			     0x0000000F);
+
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
+			     0);
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
+			     0);
+
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
+		WREG32_SOC15(GC, i,
+			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+	}
 }
 
 static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
@@ -255,7 +294,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 	unsigned num_level, block_size;
 	uint32_t tmp;
-	int i;
+	int i, j;
 
 	num_level = adev->vm_manager.num_level;
 	block_size = adev->vm_manager.block_size;
@@ -264,81 +303,89 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	else
 		block_size -= 9;
 
-	for (i = 0; i <= 14; i++) {
-		tmp = RREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL, i);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
-				    num_level);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
-				    1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    PAGE_TABLE_BLOCK_SIZE,
-				    block_size);
-		/* Send no-retry XNACK on fault to suppress VM fault storm.
-		 * On Aldebaran, XNACK can be enabled in the SQ per-process.
-		 * Retry faults need to be enabled for that to work.
-		 */
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    !adev->gmc.noretry ||
-				    adev->asic_type == CHIP_ALDEBARAN);
-		WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL,
-				    i * hub->ctx_distance, tmp);
-		WREG32_SOC15_OFFSET(GC, 0,
-				    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
-				    i * hub->ctx_addr_distance, 0);
-		WREG32_SOC15_OFFSET(GC, 0,
-				    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
-				    i * hub->ctx_addr_distance, 0);
-		WREG32_SOC15_OFFSET(GC, 0,
-				    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
-				    i * hub->ctx_addr_distance,
-				    lower_32_bits(adev->vm_manager.max_pfn - 1));
-		WREG32_SOC15_OFFSET(GC, 0,
-				    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
-				    i * hub->ctx_addr_distance,
-				    upper_32_bits(adev->vm_manager.max_pfn - 1));
+	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		for (i = 0; i <= 14; i++) {
+			tmp = RREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL, i);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
+					    num_level);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
+					    1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    PAGE_TABLE_BLOCK_SIZE,
+					    block_size);
+			/* Send no-retry XNACK on fault to suppress VM fault storm.
+			 * On Aldebaran, XNACK can be enabled in the SQ per-process.
+			 * Retry faults need to be enabled for that to work.
+			 */
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
+					    !adev->gmc.noretry ||
+					    adev->asic_type == CHIP_ALDEBARAN);
+			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL,
+					    i * hub->ctx_distance, tmp);
+			WREG32_SOC15_OFFSET(GC, j,
+					    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+					    i * hub->ctx_addr_distance, 0);
+			WREG32_SOC15_OFFSET(GC, j,
+					    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+					    i * hub->ctx_addr_distance, 0);
+			WREG32_SOC15_OFFSET(GC, j,
+					    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+					    i * hub->ctx_addr_distance,
+					    lower_32_bits(adev->vm_manager.max_pfn - 1));
+			WREG32_SOC15_OFFSET(GC, j,
+					    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+					    i * hub->ctx_addr_distance,
+					    upper_32_bits(adev->vm_manager.max_pfn - 1));
+		}
 	}
 }
 
 static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
-	unsigned i;
-
-	for (i = 0 ; i < 18; ++i) {
-		WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
-				    i * hub->eng_addr_distance, 0xffffffff);
-		WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
-				    i * hub->eng_addr_distance, 0x1f);
+	unsigned i, j;
+
+	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		for (i = 0 ; i < 18; ++i) {
+			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+					    i * hub->eng_addr_distance, 0xffffffff);
+			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+					    i * hub->eng_addr_distance, 0x1f);
+		}
 	}
 }
 
 static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 {
-	if (amdgpu_sriov_vf(adev) && adev->asic_type != CHIP_ARCTURUS) {
+	int i;
+
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		if (amdgpu_sriov_vf(adev)) {
 		/*
 		 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are
 		 * VF copy registers so vbios post doesn't program them, for
 		 * SRIOV driver need to program them
 		 */
-		WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_BASE,
-			     adev->gmc.vram_start >> 24);
-		WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_TOP,
-			     adev->gmc.vram_end >> 24);
+			WREG32_SOC15_RLC(GC, i, regMC_VM_FB_LOCATION_BASE,
+				     adev->gmc.vram_start >> 24);
+			WREG32_SOC15_RLC(GC, i, regMC_VM_FB_LOCATION_TOP,
+				     adev->gmc.vram_end >> 24);
+		}
 	}
 
 	/* GART Enable. */
@@ -361,27 +408,29 @@ static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
 	u32 tmp;
-	u32 i;
-
-	/* Disable all tables */
-	for (i = 0; i < 16; i++)
-		WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_CNTL,
-				    i * hub->ctx_distance, 0);
-
-	/* Setup TLB control */
-	tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
-	tmp = REG_SET_FIELD(tmp,
-				MC_VM_MX_L1_TLB_CNTL,
-				ENABLE_ADVANCED_DRIVER_MODEL,
-				0);
-	WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
-
-	/* Setup L2 cache */
-	tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
-	WREG32_SOC15(GC, 0, regVM_L2_CNTL, tmp);
-	WREG32_SOC15(GC, 0, regVM_L2_CNTL3, 0);
+	u32 i, j;
+
+	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		/* Disable all tables */
+		for (i = 0; i < 16; i++)
+			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT0_CNTL,
+					    i * hub->ctx_distance, 0);
+
+		/* Setup TLB control */
+		tmp = RREG32_SOC15(GC, j, regMC_VM_MX_L1_TLB_CNTL);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
+		tmp = REG_SET_FIELD(tmp,
+					MC_VM_MX_L1_TLB_CNTL,
+					ENABLE_ADVANCED_DRIVER_MODEL,
+					0);
+		WREG32_SOC15_RLC(GC, j, regMC_VM_MX_L1_TLB_CNTL, tmp);
+
+		/* Setup L2 cache */
+		tmp = RREG32_SOC15(GC, j, regVM_L2_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
+		WREG32_SOC15(GC, j, regVM_L2_CNTL, tmp);
+		WREG32_SOC15(GC, j, regVM_L2_CNTL3, 0);
+	}
 }
 
 /**
@@ -394,38 +443,42 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 						 bool value)
 {
 	u32 tmp;
-	tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp,
-			VM_L2_PROTECTION_FAULT_CNTL,
-			TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
-			value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	if (!value) {
+	int i;
+
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		tmp = RREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp,
+				VM_L2_PROTECTION_FAULT_CNTL,
+				TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+				value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-				CRASH_ON_NO_RETRY_FAULT, 1);
+				WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-				CRASH_ON_RETRY_FAULT, 1);
+				EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		if (!value) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+					CRASH_ON_NO_RETRY_FAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+					CRASH_ON_RETRY_FAULT, 1);
+		}
+		WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
 	}
-	WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
 static void gfxhub_v1_2_init(struct amdgpu_device *adev)
-- 
GitLab


From 1ffbc89c302669c95779c1e6b0901380544c9bb8 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Fri, 5 May 2023 15:37:31 +0800
Subject: [PATCH 0969/1791] drm/amdgpu: remove unneeded semicolon

No functional modification involved.

./drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c:146:2-3: Unneeded semicolon.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4871
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 24d12075ca3aa..a331a59c49e38 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -143,7 +143,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 		break;
 	default:
 		break;
-	};
+	}
 
 	return;
 }
-- 
GitLab


From 738b3469f8e12ae72555ef4724bebe8167a93e29 Mon Sep 17 00:00:00 2001
From: Sung Lee <sunglee@amd.com>
Date: Fri, 21 Apr 2023 12:03:16 -0400
Subject: [PATCH 0970/1791] drm/amd/display: Add additional pstate registers to
 HW state query

[WHY]
These registers would be useful to know when debugging pstate issues.

[HOW]
Add additional registers to hw state query.

Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Sung Lee <sunglee@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h   | 2 ++
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c   | 6 ++++++
 drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h    | 2 ++
 4 files changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
index 0b17c2993ca5b..09784222cc033 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h
@@ -690,6 +690,8 @@ struct dcn_hubp_state {
 	uint32_t primary_surface_addr_hi;
 	uint32_t primary_meta_addr_lo;
 	uint32_t primary_meta_addr_hi;
+	uint32_t uclk_pstate_force;
+	uint32_t hubp_cntl;
 };
 
 struct dcn10_hubp {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
index 0ddd310cc9710..6eebcb22e3173 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c
@@ -628,6 +628,12 @@ void hubbub2_read_state(struct hubbub *hubbub, struct dcn_hubbub_state *hubbub_s
 		REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, 0x6);
 		hubbub_state->test_debug_data = REG_READ(DCHUBBUB_TEST_DEBUG_DATA);
 	}
+
+	if (REG(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL))
+		hubbub_state->watermark_change_cntl = REG_READ(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL);
+
+	if (REG(DCHUBBUB_ARB_DRAM_STATE_CNTL))
+		hubbub_state->dram_state_cntl = REG_READ(DCHUBBUB_ARB_DRAM_STATE_CNTL);
 }
 
 static const struct hubbub_funcs hubbub2_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
index e46bbe7ddcc91..2861d974fcf62 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c
@@ -449,6 +449,12 @@ void hubp3_read_state(struct hubp *hubp)
 		SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height,
 		PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear);
 
+	if (REG(UCLK_PSTATE_FORCE))
+		s->uclk_pstate_force = REG_READ(UCLK_PSTATE_FORCE);
+
+	if (REG(DCHUBP_CNTL))
+		s->hubp_cntl = REG_READ(DCHUBP_CNTL);
+
 }
 
 void hubp3_setup(
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index aa80b3f2ca3f8..aaa2936138467 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -112,6 +112,8 @@ struct dcn_hubbub_state {
 	uint32_t vm_error_pipe;
 	uint32_t vm_error_mode;
 	uint32_t test_debug_data;
+	uint32_t watermark_change_cntl;
+	uint32_t dram_state_cntl;
 };
 
 struct hubbub_funcs {
-- 
GitLab


From b9274387bc2a4cf54b02e039b6a0aef5dd5f2936 Mon Sep 17 00:00:00 2001
From: Felix Kuehling <Felix.Kuehling@amd.com>
Date: Wed, 26 Apr 2023 22:07:43 -0400
Subject: [PATCH 0971/1791] drm/amdkfd: Don't trigger evictions unmapping
 dmabuf attachments

Don't move DMABuf attachments for PCIe P2P mappings to the SYSTEM domain
when unmapping. This avoids triggering eviction fences unnecessarily.
Instead do the move to SYSTEM and back to GTT when mapping these
attachments to ensure the SG table gets updated after evictions.

This may still trigger unnecessary evictions if user mode unmaps and
remaps the same BO. However, this is unlikely in real applications.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Eric Huang <jinhuieric.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index c3990a5eb7c69..de6ba0d4b8609 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -531,6 +531,12 @@ kfd_mem_dmamap_dmabuf(struct kfd_mem_attachment *attachment)
 {
 	struct ttm_operation_ctx ctx = {.interruptible = true};
 	struct amdgpu_bo *bo = attachment->bo_va->base.bo;
+	int ret;
+
+	amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_CPU);
+	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
+	if (ret)
+		return ret;
 
 	amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
 	return ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
@@ -663,11 +669,10 @@ kfd_mem_dmaunmap_userptr(struct kgd_mem *mem,
 static void
 kfd_mem_dmaunmap_dmabuf(struct kfd_mem_attachment *attachment)
 {
-	struct ttm_operation_ctx ctx = {.interruptible = true};
-	struct amdgpu_bo *bo = attachment->bo_va->base.bo;
-
-	amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_CPU);
-	ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
+	/* This is a no-op. We don't want to trigger eviction fences when
+	 * unmapping DMABufs. Therefore the invalidation (moving to system
+	 * domain) is done in kfd_mem_dmamap_dmabuf.
+	 */
 }
 
 /**
-- 
GitLab


From 9e72813f69b178b676a54c4d6b24d3e84492b61a Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Sat, 6 May 2023 16:10:43 +0800
Subject: [PATCH 0972/1791] drm/amdgpu: Remove the unused variable
 golden_settings_gc_9_4_3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Variable golden_settings_gc_9_4_3 is not effectively used, so delete it.

drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:48:38: warning: ‘golden_settings_gc_9_4_3’ defined but not used.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4877
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index c9ae3065a13d8..9d17dcfae1306 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -45,10 +45,6 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
 #define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
 
-static const struct soc15_reg_golden golden_settings_gc_9_4_3[] = {
-
-};
-
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev);
 static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev);
-- 
GitLab


From bf7fda0b3736f93ac8b18e7147e1e7acd27e6a19 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Date: Tue, 25 Apr 2023 14:42:29 -0600
Subject: [PATCH 0973/1791] drm/amd/display: Show the DCN/DCE version in the
 log

Some times people send their dmesg log for debugging, and one common
task is to check the modesetting line to catch which DCN/DCE we need to
debug. This commit introduces a simple conversion from the DCN/DCE
version to a string shown in the dmesg log.

Reviewed-by: Hamza Mahfooz <Hamza.Mahfooz@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  3 +-
 drivers/gpu/drm/amd/display/dc/dc_helper.c    | 56 +++++++++++++++++++
 drivers/gpu/drm/amd/display/dc/dm_services.h  |  2 +
 3 files changed, 60 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index e278474e4a5b6..e1c4db673dea4 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1667,7 +1667,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	adev->dm.dc = dc_create(&init_data);
 
 	if (adev->dm.dc) {
-		DRM_INFO("Display Core initialized with v%s!\n", DC_VER);
+		DRM_INFO("Display Core initialized with v%s! %s\n", DC_VER,
+			 dce_version_to_string(adev->dm.dc->ctx->dce_version));
 	} else {
 		DRM_INFO("Display Core failed to initialize with v%s!\n", DC_VER);
 		goto error;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index a21948267c0f8..3db7a2e11af52 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -690,3 +690,59 @@ void reg_sequence_wait_done(const struct dc_context *ctx)
 		dc_dmub_srv_wait_idle(ctx->dmub_srv);
 	}
 }
+
+char *dce_version_to_string(const int version)
+{
+	switch (version) {
+	case DCE_VERSION_8_0:
+		return "DCE 8.0";
+	case DCE_VERSION_8_1:
+		return "DCE 8.1";
+	case DCE_VERSION_8_3:
+		return "DCE 8.3";
+	case DCE_VERSION_10_0:
+		return "DCE 10.0";
+	case DCE_VERSION_11_0:
+		return "DCE 11.0";
+	case DCE_VERSION_11_2:
+		return "DCE 11.2";
+	case DCE_VERSION_11_22:
+		return "DCE 11.22";
+	case DCE_VERSION_12_0:
+		return "DCE 12.0";
+	case DCE_VERSION_12_1:
+		return "DCE 12.1";
+	case DCN_VERSION_1_0:
+		return "DCN 1.0";
+	case DCN_VERSION_1_01:
+		return "DCN 1.0.1";
+	case DCN_VERSION_2_0:
+		return "DCN 2.0";
+	case DCN_VERSION_2_1:
+		return "DCN 2.1";
+	case DCN_VERSION_2_01:
+		return "DCN 2.0.1";
+	case DCN_VERSION_3_0:
+		return "DCN 3.0";
+	case DCN_VERSION_3_01:
+		return "DCN 3.0.1";
+	case DCN_VERSION_3_02:
+		return "DCN 3.0.2";
+	case DCN_VERSION_3_03:
+		return "DCN 3.0.3";
+	case DCN_VERSION_3_1:
+		return "DCN 3.1";
+	case DCN_VERSION_3_14:
+		return "DCN 3.1.4";
+	case DCN_VERSION_3_15:
+		return "DCN 3.1.5";
+	case DCN_VERSION_3_16:
+		return "DCN 3.1.6";
+	case DCN_VERSION_3_2:
+		return "DCN 3.2";
+	case DCN_VERSION_3_21:
+		return "DCN 3.2.1";
+	default:
+		return "Unknown";
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index d33d595405a9d..d0eed3b4771e6 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -292,4 +292,6 @@ void dm_dtn_log_append_v(struct dc_context *ctx,
 void dm_dtn_log_end(struct dc_context *ctx,
 	struct dc_log_buffer_ctx *log_ctx);
 
+char *dce_version_to_string(const int version);
+
 #endif /* __DM_SERVICES_H__ */
-- 
GitLab


From 187916e6ed9d0c3b3abc27429f7a5f8c936bd1f0 Mon Sep 17 00:00:00 2001
From: Lang Yu <Lang.Yu@amd.com>
Date: Fri, 5 May 2023 20:14:15 +0800
Subject: [PATCH 0974/1791] drm/amdgpu: install stub fence into potential
 unused fence pointers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When using cpu to update page tables, vm update fences are unused.
Install stub fence into these fence pointers instead of NULL
to avoid NULL dereference when calling dma_fence_wait() on them.

Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 3c0310576b3bf..b6bd667df6763 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1358,6 +1358,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	amdgpu_vm_bo_base_init(&bo_va->base, vm, bo);
 
 	bo_va->ref_count = 1;
+	bo_va->last_pt_update = dma_fence_get_stub();
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
@@ -2067,7 +2068,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		vm->update_funcs = &amdgpu_vm_cpu_funcs;
 	else
 		vm->update_funcs = &amdgpu_vm_sdma_funcs;
-	vm->last_update = NULL;
+
+	vm->last_update = dma_fence_get_stub();
 	vm->last_unlocked = dma_fence_get_stub();
 	vm->last_tlb_flush = dma_fence_get_stub();
 
@@ -2192,7 +2194,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		goto unreserve_bo;
 
 	dma_fence_put(vm->last_update);
-	vm->last_update = NULL;
+	vm->last_update = dma_fence_get_stub();
 	vm->is_compute_context = true;
 
 	/* Free the shadow bo for compute VM */
-- 
GitLab


From 9d7a348d368ccd940242d4aa68292cf665f6d474 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Fri, 5 May 2023 13:20:11 +0800
Subject: [PATCH 0975/1791] drm/amd/pm: parse pp_handle under appropriate
 conditions

amdgpu_dpm_is_overdrive_supported is a common API across all
asics, so we should cast pp_handle into correct structure
under different power frameworks.

v2: using return directly to simplify code
v3: SI asic does not carry od_enabled member in pp_handle, and update Fixes tag

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2541
Fixes: eb4900aa4c49 ("drm/amdgpu: Fix kernel NULL pointer dereference in dpm functions")
Suggested-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_dpm.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 300e156b924f4..86246f69dbe15 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -1460,15 +1460,21 @@ int amdgpu_dpm_get_smu_prv_buf_details(struct amdgpu_device *adev,
 
 int amdgpu_dpm_is_overdrive_supported(struct amdgpu_device *adev)
 {
-	struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle;
-	struct smu_context *smu = adev->powerplay.pp_handle;
+	if (is_support_sw_smu(adev)) {
+		struct smu_context *smu = adev->powerplay.pp_handle;
+
+		return (smu->od_enabled || smu->is_apu);
+	} else {
+		struct pp_hwmgr *hwmgr;
 
-	if ((is_support_sw_smu(adev) && smu->od_enabled) ||
-	    (is_support_sw_smu(adev) && smu->is_apu) ||
-		(!is_support_sw_smu(adev) && hwmgr->od_enabled))
-		return true;
+		/* SI asic does not carry od_enabled */
+		if (adev->family == AMDGPU_FAMILY_SI)
+			return false;
 
-	return false;
+		hwmgr = (struct pp_hwmgr *)adev->powerplay.pp_handle;
+
+		return hwmgr->od_enabled;
+	}
 }
 
 int amdgpu_dpm_set_pp_table(struct amdgpu_device *adev,
-- 
GitLab


From db5dcd476eb0288cdfa781725582dcf9bd747a6b Mon Sep 17 00:00:00 2001
From: YuBiao Wang <YuBiao.Wang@amd.com>
Date: Fri, 5 May 2023 14:35:32 +0800
Subject: [PATCH 0976/1791] drm/amdgpu: set default num_kcq to 2 under sriov
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The number of kernel queues has impact on the latency under sriov
usecase. So to reduce the latency we set the default num_kcq = 2 under
sriov if not set manually.

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 1311e42ab8e99..d0ad7cb0fa05d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -68,6 +68,9 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	/* enable mcbp for sriov */
 	amdgpu_mcbp = 1;
 
+	/* Reduce kcq number to 2 to reduce latency */
+	if (amdgpu_num_kcq == -1)
+		amdgpu_num_kcq = 2;
 }
 
 void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
-- 
GitLab


From 273f47997718fc72d96e5a4a640538c00575fa7e Mon Sep 17 00:00:00 2001
From: Leo Chen <sancchen@amd.com>
Date: Wed, 26 Apr 2023 16:02:28 -0400
Subject: [PATCH 0977/1791] drm/amd/display: Add symclk workaround during
 disable link output

[Why & How]
This is originally a change (9c75891f) in DCN32 because of the lack
of interface to set TX while keeping symclk on. Adding this workaround
to DCN314 will resolve the current issue.

Fixes: 9c75891feef0 ("drm/amd/display: rework recent update PHY state commit")
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  | 65 +++++++++++++++++++
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |  2 +
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |  2 +-
 3 files changed, 68 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 6fb3f64e30576..f7a3e0d71d016 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -421,3 +421,68 @@ void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool
 
 	PERF_TRACE();
 }
+static void apply_symclk_on_tx_off_wa(struct dc_link *link)
+{
+	/* There are use cases where SYMCLK is referenced by OTG. For instance
+	 * for TMDS signal, OTG relies SYMCLK even if TX video output is off.
+	 * However current link interface will power off PHY when disabling link
+	 * output. This will turn off SYMCLK generated by PHY. The workaround is
+	 * to identify such case where SYMCLK is still in use by OTG when we
+	 * power off PHY. When this is detected, we will temporarily power PHY
+	 * back on and move PHY's SYMCLK state to SYMCLK_ON_TX_OFF by calling
+	 * program_pix_clk interface. When OTG is disabled, we will then power
+	 * off PHY by calling disable link output again.
+	 *
+	 * In future dcn generations, we plan to rework transmitter control
+	 * interface so that we could have an option to set SYMCLK ON TX OFF
+	 * state in one step without this workaround
+	 */
+
+	struct dc *dc = link->ctx->dc;
+	struct pipe_ctx *pipe_ctx = NULL;
+	uint8_t i;
+
+	if (link->phy_state.symclk_ref_cnts.otg > 0) {
+		for (i = 0; i < MAX_PIPES; i++) {
+			pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe_ctx->stream && pipe_ctx->stream->link == link && pipe_ctx->top_pipe == NULL) {
+				pipe_ctx->clock_source->funcs->program_pix_clk(
+						pipe_ctx->clock_source,
+						&pipe_ctx->stream_res.pix_clk_params,
+						dc->link_srv->dp_get_encoding_format(
+								&pipe_ctx->link_config.dp_link_settings),
+						&pipe_ctx->pll_settings);
+				link->phy_state.symclk_state = SYMCLK_ON_TX_OFF;
+				break;
+			}
+		}
+	}
+}
+
+void dcn314_disable_link_output(struct dc_link *link,
+		const struct link_resource *link_res,
+		enum signal_type signal)
+{
+	struct dc *dc = link->ctx->dc;
+	const struct link_hwss *link_hwss = get_link_hwss(link, link_res);
+	struct dmcu *dmcu = dc->res_pool->dmcu;
+
+	if (signal == SIGNAL_TYPE_EDP &&
+			link->dc->hwss.edp_backlight_control)
+		link->dc->hwss.edp_backlight_control(link, false);
+	else if (dmcu != NULL && dmcu->funcs->lock_phy)
+		dmcu->funcs->lock_phy(dmcu);
+
+	link_hwss->disable_link_output(link, link_res, signal);
+	link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF;
+	/*
+	 * Add the logic to extract BOTH power up and power down sequences
+	 * from enable/disable link output and only call edp panel control
+	 * in enable_link_dp and disable_link_dp once.
+	 */
+	if (dmcu != NULL && dmcu->funcs->lock_phy)
+		dmcu->funcs->unlock_phy(dmcu);
+	dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY);
+
+	apply_symclk_on_tx_off_wa(link);
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
index c786d5e6a428e..6d0b62503caa6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
@@ -45,4 +45,6 @@ void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool
 
 void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on);
 
+void dcn314_disable_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal);
+
 #endif /* __DC_HWSS_DCN314_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index ce53339b2e10d..ed8a1b94c0068 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -106,7 +106,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = {
 	.enable_lvds_link_output = dce110_enable_lvds_link_output,
 	.enable_tmds_link_output = dce110_enable_tmds_link_output,
 	.enable_dp_link_output = dce110_enable_dp_link_output,
-	.disable_link_output = dce110_disable_link_output,
+	.disable_link_output = dcn314_disable_link_output,
 	.z10_restore = dcn31_z10_restore,
 	.z10_save_init = dcn31_z10_save_init,
 	.set_disp_pattern_generator = dcn30_set_disp_pattern_generator,
-- 
GitLab


From 62cc621604a466ede81a125a2ed63e05695a48cc Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Wed, 26 Apr 2023 19:33:16 -0400
Subject: [PATCH 0978/1791] drm/amd/display: Check Vactive for VRR active for
 FPO + Vactive

[Description]
- For FPO + Vactive cases, we rely on the Vactive display to be at
  it's nominal refresh rate because the Vactive pipe may not necessarily
  assert P-State allow while it's in VBLANK
- For cases where the Vactive display has a stretched VBLANK due to
  VRR, we could underflow when trying to complete an FPO + Vactive
  MCLK switch because the FPO display has limited VBLANK time in
  waiting for the Vactive display to assert P-State allow naturally
- Block FPO + Vactive if the Vactive display has VRR active (variable
  or fixed)

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c             | 5 ++++-
 drivers/gpu/drm/amd/display/dc/dc_stream.h           | 2 ++
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 7 ++++++-
 3 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 965b348dc8f11..c26dfdd48dd93 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2604,7 +2604,7 @@ static enum surface_update_type check_update_surfaces_for_stream(
 
 		if (stream_update->stream && stream_update->stream->freesync_on_desktop &&
 			(stream_update->vrr_infopacket || stream_update->allow_freesync ||
-				stream_update->vrr_active_variable))
+				stream_update->vrr_active_variable || stream_update->vrr_active_fixed))
 			su_flags->bits.fams_changed = 1;
 
 		if (su_flags->raw != 0)
@@ -2964,6 +2964,9 @@ static void copy_stream_update_to_stream(struct dc *dc,
 	if (update->vrr_active_variable)
 		stream->vrr_active_variable = *update->vrr_active_variable;
 
+	if (update->vrr_active_fixed)
+		stream->vrr_active_fixed = *update->vrr_active_fixed;
+
 	if (update->crtc_timing_adjust)
 		stream->adjust = *update->crtc_timing_adjust;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 0add5ecc895fa..d5b3e3a32cc6d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -233,6 +233,7 @@ struct dc_stream_state {
 	 */
 	bool vrr_active_variable;
 	bool freesync_on_desktop;
+	bool vrr_active_fixed;
 
 	bool converter_disable_audio;
 	uint8_t qs_bit;
@@ -326,6 +327,7 @@ struct dc_stream_update {
 	bool integer_scaling_update;
 	bool *allow_freesync;
 	bool *vrr_active_variable;
+	bool *vrr_active_fixed;
 
 	struct colorspace_transform *gamut_remap;
 	enum dc_color_space *output_color_space;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 826059d5b3675..46fd7b68857c9 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -35,6 +35,7 @@
 
 #define DC_LOGGER_INIT(logger)
 
+static const unsigned int MAX_FPO_VACTIVE_BLANK_US = 600;
 static const struct subvp_high_refresh_list subvp_high_refresh_list = {
 			.min_refresh = 120,
 			.max_refresh = 165,
@@ -2924,6 +2925,7 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint
 	unsigned int i, pipe_idx;
 	const struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 	bool vactive_found = false;
+	unsigned int blank_us = 0;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
@@ -2931,7 +2933,10 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint
 		if (!pipe->stream)
 			continue;
 
-		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us) {
+		blank_us = ((pipe->stream->timing.v_total - pipe->stream->timing.v_addressable) * pipe->stream->timing.h_total /
+				(double)(pipe->stream->timing.pix_clk_100hz * 100)) * 1000000;
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us &&
+				!(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed) && blank_us < MAX_FPO_VACTIVE_BLANK_US) {
 			vactive_found = true;
 			break;
 		}
-- 
GitLab


From 84f14428b1e0d1f61776c5fcfdef181129533e0b Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Thu, 27 Apr 2023 15:10:13 -0400
Subject: [PATCH 0979/1791] drm/amd/display: Enforce 60us prefetch for 200Mhz
 DCFCLK modes

[Description]
- Due to bandwidth / arbitration issues at 200Mhz DCFCLK,
  we want to enforce minimum 60us of prefetch to avoid
  intermittent underflow issues
- Since 60us prefetch is already enforced for UCLK DPM0,
  and many DCFCLK's > 200Mhz are mapped to UCLK DPM1, in
  theory there should not be any UCLK DPM regressions by
  enforcing greater prefetch

Reviewed-by: Nevenko Stupar <Nevenko.Stupar@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c   | 5 +++--
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h   | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
index 66f44a013fe5d..958d27224f644 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
@@ -810,7 +810,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 					v->SwathHeightY[k],
 					v->SwathHeightC[k],
 					TWait,
-					v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ?
+					(v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ||
+						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= MIN_DCFCLK_FREQ_MHZ) ?
 							mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 					/* Output */
 					&v->DSTXAfterScaler[k],
@@ -3314,7 +3315,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->swath_width_chroma_ub_this_state[k],
 							v->SwathHeightYThisState[k],
 							v->SwathHeightCThisState[k], v->TWait,
-							v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ ?
+							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= MIN_DCFCLK_FREQ_MHZ) ?
 									mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 
 							/* Output */
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
index 500b3dd6052d9..d98e36a9a09cc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
@@ -53,6 +53,7 @@
 #define BPP_BLENDED_PIPE 0xffffffff
 
 #define MEM_STROBE_FREQ_MHZ 1600
+#define MIN_DCFCLK_FREQ_MHZ 200
 #define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0
 
 struct display_mode_lib;
-- 
GitLab


From a00e595207d001432a85758954c3a6f6a9896368 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Thu, 27 Apr 2023 15:10:41 -0400
Subject: [PATCH 0980/1791] drm/amd/display: Block SubVP high refresh when VRR
 active fixed

[Description]
- SubVP high refresh is blocked when VRR is active variable, but
  we should also block it for when VRR is active fixed (video use
  case)

Reviewed-by: Nevenko Stupar <Nevenko.Stupar@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 46fd7b68857c9..cd28980b2b567 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2825,7 +2825,7 @@ bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context
 	uint32_t i;
 
 	if (!dc->debug.disable_subvp_high_refresh && pipe->stream &&
-			pipe->plane_state && !pipe->stream->vrr_active_variable) {
+			pipe->plane_state && !(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed)) {
 		refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 +
 						pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1)
 						/ (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total);
-- 
GitLab


From 807a1c14276b6ba6dc7efb4784ac35bceea1413f Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Fri, 28 Apr 2023 17:29:02 -0400
Subject: [PATCH 0981/1791] drm/amd/display: Block SubVP on displays that have
 pixclk > 1800Mhz

[Description]
- Enabling SubVP on high refresh rate displays had a side effect
  of also enabling on high bandwidth displays such as 8K60
- However, these are not validated and should be blocked for
  the time being
- Block SubVP on displays that have pix rate > 1800Mhz (includes
  8K60 displays)

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Reviewed-by: Nevenko Stupar <Nevenko.Stupar@amd.com>
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h | 1 +
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 04be01ae1ecfd..42ccfd13a37c7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -41,6 +41,7 @@
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
 #define DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US 100 // Only allow FPO + Vactive if active margin >= 100
 #define SUBVP_HIGH_REFRESH_LIST_LEN 3
+#define DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ 1800
 
 #define TO_DCN32_RES_POOL(pool)\
 	container_of(pool, struct dcn32_resource_pool, base)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index cd28980b2b567..f7e45d935a297 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -703,6 +703,7 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 		 * - Not TMZ surface
 		 */
 		if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) &&
+				!(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) &&
 				(!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) &&
 				pipe->stream->mall_stream_config.type == SUBVP_NONE &&
 				(refresh_rate < 120 || dcn32_allow_subvp_high_refresh_rate(dc, context, pipe)) &&
-- 
GitLab


From 2f48965bdc02d781181ee4fa3d5b3cc168e6d9af Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Thu, 4 May 2023 01:46:12 -0400
Subject: [PATCH 0982/1791] drm/amdgpu: drop gfx_v11_0_cp_ecc_error_irq_funcs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The gfx.cp_ecc_error_irq is retired in gfx11. In gfx_v11_0_hw_fini still
use amdgpu_irq_put to disable this interrupt, which caused the call trace
in this function.

[  102.873958] Call Trace:
[  102.873959]  <TASK>
[  102.873961]  gfx_v11_0_hw_fini+0x23/0x1e0 [amdgpu]
[  102.874019]  gfx_v11_0_suspend+0xe/0x20 [amdgpu]
[  102.874072]  amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu]
[  102.874122]  amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[  102.874172]  amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu]
[  102.874223]  amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu]
[  102.874321]  amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[  102.874375]  process_one_work+0x21f/0x3f0
[  102.874377]  worker_thread+0x200/0x3e0
[  102.874378]  ? process_one_work+0x3f0/0x3f0
[  102.874379]  kthread+0xfd/0x130
[  102.874380]  ? kthread_complete_and_exit+0x20/0x20
[  102.874381]  ret_from_fork+0x22/0x30

v2:
- Handle umc and gfx ras cases in separated patch
- Retired the gfx_v11_0_cp_ecc_error_irq_funcs in gfx11

v3:
- Improve the subject and code comments
- Add judgment on gfx11 in the function of amdgpu_gfx_ras_late_init

v4:
- Drop the define of CP_ME1_PIPE_INST_ADDR_INTERVAL and
SET_ECC_ME_PIPE_STATE which using in gfx_v11_0_set_cp_ecc_error_state
- Check cp_ecc_error_irq.funcs rather than ip version for a more
sustainable life

v5:
- Simplify judgment conditions

Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c |  8 +++--
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  | 46 -------------------------
 2 files changed, 5 insertions(+), 49 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 76dd2841cc0df..a22d88a4178ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -793,9 +793,11 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *r
 		if (r)
 			return r;
 
-		r = amdgpu_irq_get(adev, &adev->gfx.cp_ecc_error_irq, 0);
-		if (r)
-			goto late_fini;
+		if (adev->gfx.cp_ecc_error_irq.funcs) {
+			r = amdgpu_irq_get(adev, &adev->gfx.cp_ecc_error_irq, 0);
+			if (r)
+				goto late_fini;
+		}
 	} else {
 		amdgpu_ras_feature_enable_on_boot(adev, ras_block, 0);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index f09e2558e73b9..7bff50e58cffa 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1355,13 +1355,6 @@ static int gfx_v11_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	/* ECC error */
-	r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GRBM_CP,
-				  GFX_11_0_0__SRCID__CP_ECC_ERROR,
-				  &adev->gfx.cp_ecc_error_irq);
-	if (r)
-		return r;
-
 	/* FED error */
 	r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GFX,
 				  GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT,
@@ -4369,7 +4362,6 @@ static int gfx_v11_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
@@ -5846,36 +5838,6 @@ static void gfx_v11_0_set_compute_eop_interrupt_state(struct amdgpu_device *adev
 	}
 }
 
-#define CP_ME1_PIPE_INST_ADDR_INTERVAL  0x1
-#define SET_ECC_ME_PIPE_STATE(reg_addr, state) \
-	do { \
-		uint32_t tmp = RREG32_SOC15_IP(GC, reg_addr); \
-		tmp = REG_SET_FIELD(tmp, CP_ME1_PIPE0_INT_CNTL, CP_ECC_ERROR_INT_ENABLE, state); \
-		WREG32_SOC15_IP(GC, reg_addr, tmp); \
-	} while (0)
-
-static int gfx_v11_0_set_cp_ecc_error_state(struct amdgpu_device *adev,
-							struct amdgpu_irq_src *source,
-							unsigned type,
-							enum amdgpu_interrupt_state state)
-{
-	uint32_t ecc_irq_state = 0;
-	uint32_t pipe0_int_cntl_addr = 0;
-	int i = 0;
-
-	ecc_irq_state = (state == AMDGPU_IRQ_STATE_ENABLE) ? 1 : 0;
-
-	pipe0_int_cntl_addr = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE0_INT_CNTL);
-
-	WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0, CP_ECC_ERROR_INT_ENABLE, ecc_irq_state);
-
-	for (i = 0; i < adev->gfx.mec.num_pipe_per_mec; i++)
-		SET_ECC_ME_PIPE_STATE(pipe0_int_cntl_addr + i * CP_ME1_PIPE_INST_ADDR_INTERVAL,
-					ecc_irq_state);
-
-	return 0;
-}
-
 static int gfx_v11_0_set_eop_interrupt_state(struct amdgpu_device *adev,
 					    struct amdgpu_irq_src *src,
 					    unsigned type,
@@ -6292,11 +6254,6 @@ static const struct amdgpu_irq_src_funcs gfx_v11_0_priv_inst_irq_funcs = {
 	.process = gfx_v11_0_priv_inst_irq,
 };
 
-static const struct amdgpu_irq_src_funcs gfx_v11_0_cp_ecc_error_irq_funcs = {
-	.set = gfx_v11_0_set_cp_ecc_error_state,
-	.process = amdgpu_gfx_cp_ecc_error_irq,
-};
-
 static const struct amdgpu_irq_src_funcs gfx_v11_0_rlc_gc_fed_irq_funcs = {
 	.process = gfx_v11_0_rlc_gc_fed_irq,
 };
@@ -6312,9 +6269,6 @@ static void gfx_v11_0_set_irq_funcs(struct amdgpu_device *adev)
 	adev->gfx.priv_inst_irq.num_types = 1;
 	adev->gfx.priv_inst_irq.funcs = &gfx_v11_0_priv_inst_irq_funcs;
 
-	adev->gfx.cp_ecc_error_irq.num_types = 1; /* CP ECC error */
-	adev->gfx.cp_ecc_error_irq.funcs = &gfx_v11_0_cp_ecc_error_irq_funcs;
-
 	adev->gfx.rlc_gc_fed_irq.num_types = 1; /* 0x80 FED error */
 	adev->gfx.rlc_gc_fed_irq.funcs = &gfx_v11_0_rlc_gc_fed_irq_funcs;
 
-- 
GitLab


From 4994d1f0a754cded0afb62c4753d00760ddca511 Mon Sep 17 00:00:00 2001
From: "Lin.Cao" <lincao12@amd.com>
Date: Mon, 8 May 2023 17:28:41 +0800
Subject: [PATCH 0983/1791] drm/amdgpu: Fix vram recover doesn't work after
 whole GPU reset (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v1: Vmbo->shadow is used to back vram bo up when vram lost. So that we
should set shadow as vmbo->shadow to recover vmbo->bo
v2: Modify if(vmbo->shadow) shadow = vmbo->shadow as if(!vmbo->shadow)
continue;

Fixes: e18aaea733da ("drm/amdgpu: move shadow_list to amdgpu_bo_vm")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lin.Cao <lincao12@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index ae0bcffa25910..e94507a10e15e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -4513,7 +4513,11 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev)
 	dev_info(adev->dev, "recover vram bo from shadow start\n");
 	mutex_lock(&adev->shadow_list_lock);
 	list_for_each_entry(vmbo, &adev->shadow_list, shadow_list) {
-		shadow = &vmbo->bo;
+		/* If vm is compute context or adev is APU, shadow will be NULL */
+		if (!vmbo->shadow)
+			continue;
+		shadow = vmbo->shadow;
+
 		/* No need to recover an evicted BO */
 		if (shadow->tbo.resource->mem_type != TTM_PL_TT ||
 		    shadow->tbo.resource->start == AMDGPU_BO_INVALID_OFFSET ||
-- 
GitLab


From aaa07c0d08a37bf72cd73a3119ff99dc2bfece74 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 30 Apr 2023 15:54:14 -0400
Subject: [PATCH 0984/1791] drm/amd/display: Promote DAL to 3.2.235

This version brings along following fixes:
- Block SubVP on displays that have pixclk > 1800Mhz
- Block SubVP high refresh when VRR active fixed
- Enforce 60us prefetch for 200Mhz DCFCLK modes
- Check Vactive for VRR active for FPO + Vactive
- Add symclk workaround during disable link output
- Show the DCN/DCE version in the log
- Add additional pstate registers to HW state query

Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7e3f20a3a02f1..f4448b545d701 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.234"
+#define DC_VER "3.2.235"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From eb4f01784ec1d589629dd8c85044b7c07e5f6993 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Sat, 6 May 2023 16:52:59 +0800
Subject: [PATCH 0985/1791] drm/amdgpu: disable sdma ecc irq only when sdma RAS
 is enabled in suspend

sdma_v4_0_ip is shared on a few asics, but in sdma_v4_0_hw_fini,
driver unconditionally disables ecc_irq which is only enabled on
those asics enabling sdma ecc. This will introduce a warning in
suspend cycle on those chips with sdma ip v4.0, while without
sdma ecc. So this patch correct this.

[ 7283.166354] RIP: 0010:amdgpu_irq_put+0x45/0x70 [amdgpu]
[ 7283.167001] RSP: 0018:ffff9a5fc3967d08 EFLAGS: 00010246
[ 7283.167019] RAX: ffff98d88afd3770 RBX: 0000000000000001 RCX: 0000000000000000
[ 7283.167023] RDX: 0000000000000000 RSI: ffff98d89da30390 RDI: ffff98d89da20000
[ 7283.167025] RBP: ffff98d89da20000 R08: 0000000000036838 R09: 0000000000000006
[ 7283.167028] R10: ffffd5764243c008 R11: 0000000000000000 R12: ffff98d89da30390
[ 7283.167030] R13: ffff98d89da38978 R14: ffffffff999ae15a R15: ffff98d880130105
[ 7283.167032] FS:  0000000000000000(0000) GS:ffff98d996f00000(0000) knlGS:0000000000000000
[ 7283.167036] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 7283.167039] CR2: 00000000f7a9d178 CR3: 00000001c42ea000 CR4: 00000000003506e0
[ 7283.167041] Call Trace:
[ 7283.167046]  <TASK>
[ 7283.167048]  sdma_v4_0_hw_fini+0x38/0xa0 [amdgpu]
[ 7283.167704]  amdgpu_device_ip_suspend_phase2+0x101/0x1a0 [amdgpu]
[ 7283.168296]  amdgpu_device_suspend+0x103/0x180 [amdgpu]
[ 7283.168875]  amdgpu_pmops_freeze+0x21/0x60 [amdgpu]
[ 7283.169464]  pci_pm_freeze+0x54/0xc0

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2522
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index b3cc04dd86536..9295ac7edd565 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1917,9 +1917,11 @@ static int sdma_v4_0_hw_fini(void *handle)
 		return 0;
 	}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
-			       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
+				       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+		}
 	}
 
 	sdma_v4_0_ctx_switch_enable(adev, false);
-- 
GitLab


From 936e95a461c2606ac2ffae3bb0d991e7081eafd8 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Tue, 9 May 2023 09:36:49 +0800
Subject: [PATCH 0986/1791] drm/amd/pm: avoid potential UBSAN issue on legacy
 asics

Prevent further dpm casting on legacy asics without od_enabled in
amdgpu_dpm_is_overdrive_supported. This can avoid UBSAN complain
in init sequence.

v2: add a macro to check legacy dpm instead of checking asic family/type
v3: refine macro name for naming consistency

Suggested-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_dpm.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 86246f69dbe15..078aaaa531628 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -36,6 +36,8 @@
 #define amdgpu_dpm_enable_bapm(adev, e) \
 		((adev)->powerplay.pp_funcs->enable_bapm((adev)->powerplay.pp_handle, (e)))
 
+#define amdgpu_dpm_is_legacy_dpm(adev) ((adev)->powerplay.pp_handle == (adev))
+
 int amdgpu_dpm_get_sclk(struct amdgpu_device *adev, bool low)
 {
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
@@ -1467,8 +1469,11 @@ int amdgpu_dpm_is_overdrive_supported(struct amdgpu_device *adev)
 	} else {
 		struct pp_hwmgr *hwmgr;
 
-		/* SI asic does not carry od_enabled */
-		if (adev->family == AMDGPU_FAMILY_SI)
+		/*
+		 * dpm on some legacy asics don't carry od_enabled member
+		 * as its pp_handle is casted directly from adev.
+		 */
+		if (amdgpu_dpm_is_legacy_dpm(adev))
 			return false;
 
 		hwmgr = (struct pp_hwmgr *)adev->powerplay.pp_handle;
-- 
GitLab


From d97b02bb9c7aa3008d473d11001e1b45b7e0c7c6 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Sat, 6 May 2023 20:06:45 +0800
Subject: [PATCH 0987/1791] drm/amdgpu/gfx: disable gfx9 cp_ecc_error_irq only
 when enabling legacy gfx ras

gfx9 cp_ecc_error_irq is only enabled when legacy gfx ras is assert.
So in gfx_v9_0_hw_fini, interrupt disablement for cp_ecc_error_irq
should be executed under such condition, otherwise, an amdgpu_irq_put
calltrace will occur.

[ 7283.170322] RIP: 0010:amdgpu_irq_put+0x45/0x70 [amdgpu]
[ 7283.170964] RSP: 0018:ffff9a5fc3967d00 EFLAGS: 00010246
[ 7283.170967] RAX: ffff98d88afd3040 RBX: ffff98d89da20000 RCX: 0000000000000000
[ 7283.170969] RDX: 0000000000000000 RSI: ffff98d89da2bef8 RDI: ffff98d89da20000
[ 7283.170971] RBP: ffff98d89da20000 R08: ffff98d89da2ca18 R09: 0000000000000006
[ 7283.170973] R10: ffffd5764243c008 R11: 0000000000000000 R12: 0000000000001050
[ 7283.170975] R13: ffff98d89da38978 R14: ffffffff999ae15a R15: ffff98d880130105
[ 7283.170978] FS:  0000000000000000(0000) GS:ffff98d996f00000(0000) knlGS:0000000000000000
[ 7283.170981] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 7283.170983] CR2: 00000000f7a9d178 CR3: 00000001c42ea000 CR4: 00000000003506e0
[ 7283.170986] Call Trace:
[ 7283.170988]  <TASK>
[ 7283.170989]  gfx_v9_0_hw_fini+0x1c/0x6d0 [amdgpu]
[ 7283.171655]  amdgpu_device_ip_suspend_phase2+0x101/0x1a0 [amdgpu]
[ 7283.172245]  amdgpu_device_suspend+0x103/0x180 [amdgpu]
[ 7283.172823]  amdgpu_pmops_freeze+0x21/0x60 [amdgpu]
[ 7283.173412]  pci_pm_freeze+0x54/0xc0
[ 7283.173419]  ? __pfx_pci_pm_freeze+0x10/0x10
[ 7283.173425]  dpm_run_callback+0x98/0x200
[ 7283.173430]  __device_suspend+0x164/0x5f0

v2: drop gfx11 as it's fixed in a different solution by retiring cp_ecc_irq funcs(Hawking)

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2522
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 4a75560994694..e093e83ae7398 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -3761,7 +3761,8 @@ static int gfx_v9_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
+		amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
-- 
GitLab


From c488a9370d5a1da71f7683e0d6da40a2ab0f1be3 Mon Sep 17 00:00:00 2001
From: Saleemkhan Jamadar <saleemkhan.jamadar@amd.com>
Date: Tue, 9 May 2023 12:37:50 +0530
Subject: [PATCH 0988/1791] drm/amdgpu/jpeg: Remove harvest checking for JPEG3

Register CC_UVD_HARVESTING is obsolete for JPEG 3.1.2

Signed-off-by: Saleemkhan Jamadar <saleemkhan.jamadar@amd.com>
Reviewed-by: Veerabadhran Gopalakrishnan <Veerabadhran.Gopalakrishnan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index c55e09432e262..1c2292cc5f2ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -54,6 +54,7 @@ static int jpeg_v3_0_early_init(void *handle)
 
 	switch (adev->ip_versions[UVD_HWIP][0]) {
 	case IP_VERSION(3, 1, 1):
+	case IP_VERSION(3, 1, 2):
 		break;
 	default:
 		harvest = RREG32_SOC15(JPEG, 0, mmCC_UVD_HARVESTING);
-- 
GitLab


From 9a06655e73834819149466ae8170ffe53b23e6f8 Mon Sep 17 00:00:00 2001
From: Huacai Chen <chenhuacai@loongson.cn>
Date: Mon, 8 May 2023 11:09:41 +0800
Subject: [PATCH 0989/1791] drm/amdgpu/display: Enable DC_FP for LoongArch

LoongArch now provides kernel_fpu_begin() and kernel_fpu_end() that are
used like the x86 counterparts in commit 2b3bd32ea3a22ea2d ("LoongArch:
Provide kernel fpu functions"), so we can enable DC_FP on LoongArch for
supporting more DCN devices.

Signed-off-by: WANG Xuerui <kernel@xen0n.name>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/Kconfig            | 2 +-
 drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c | 6 ++++--
 drivers/gpu/drm/amd/display/dc/dml/Makefile    | 5 +++++
 3 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig
index 06b438217c618..b015c7d19531d 100644
--- a/drivers/gpu/drm/amd/display/Kconfig
+++ b/drivers/gpu/drm/amd/display/Kconfig
@@ -8,7 +8,7 @@ config DRM_AMD_DC
 	depends on BROKEN || !CC_IS_CLANG || X86_64 || SPARC64 || ARM64
 	select SND_HDA_COMPONENT if SND_HDA_CORE
 	# !CC_IS_CLANG: https://github.com/ClangBuiltLinux/linux/issues/1752
-	select DRM_AMD_DC_FP if (X86 || PPC64 || (ARM64 && KERNEL_MODE_NEON && !CC_IS_CLANG))
+	select DRM_AMD_DC_FP if (X86 || LOONGARCH || PPC64 || (ARM64 && KERNEL_MODE_NEON && !CC_IS_CLANG))
 	help
 	  Choose this option if you want to use the new display engine
 	  support for AMDGPU. This adds required support for Vega and
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
index c42aa947c969d..172aa10a8800f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c
@@ -33,6 +33,8 @@
 #include <asm/cputable.h>
 #elif defined(CONFIG_ARM64)
 #include <asm/neon.h>
+#elif defined(CONFIG_LOONGARCH)
+#include <asm/fpu.h>
 #endif
 
 /**
@@ -88,7 +90,7 @@ void dc_fpu_begin(const char *function_name, const int line)
 	*pcpu += 1;
 
 	if (*pcpu == 1) {
-#if defined(CONFIG_X86)
+#if defined(CONFIG_X86) || defined(CONFIG_LOONGARCH)
 		migrate_disable();
 		kernel_fpu_begin();
 #elif defined(CONFIG_PPC64)
@@ -128,7 +130,7 @@ void dc_fpu_end(const char *function_name, const int line)
 	pcpu = get_cpu_ptr(&fpu_recursion_depth);
 	*pcpu -= 1;
 	if (*pcpu <= 0) {
-#if defined(CONFIG_X86)
+#if defined(CONFIG_X86) || defined(CONFIG_LOONGARCH)
 		kernel_fpu_end();
 		migrate_enable();
 #elif defined(CONFIG_PPC64)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile
index 01db035589c59..77cf5545c94cc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile
@@ -38,6 +38,11 @@ ifdef CONFIG_ARM64
 dml_rcflags := -mgeneral-regs-only
 endif
 
+ifdef CONFIG_LOONGARCH
+dml_ccflags := -mfpu=64
+dml_rcflags := -msoft-float
+endif
+
 ifdef CONFIG_CC_IS_GCC
 ifneq ($(call gcc-min-version, 70100),y)
 IS_OLD_GCC = 1
-- 
GitLab


From 3fb9dd5fef7ccd756544713e6e84db1c525816c4 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 9 May 2023 17:21:14 +0300
Subject: [PATCH 0990/1791] drm/amdgpu: release correct lock in
 amdgpu_gfx_enable_kgq()

This function was releasing the incorrect lock on the error path.

Reported-by: kernel test robot <lkp@intel.com>
Fixes: 1156e1a60f02 ("drm/amdgpu: add [en/dis]able_kgq() functions")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index a22d88a4178ae..87e1a1a9f2988 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -644,7 +644,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id)
 						adev->gfx.num_gfx_rings);
 		if (r) {
 			DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-			spin_unlock(&adev->gfx.kiq[0].ring_lock);
+			spin_unlock(&kiq->ring_lock);
 			return r;
 		}
 
-- 
GitLab


From a39b52c838127a42c57d5e080dbd5204770aa80e Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Tue, 9 May 2023 18:49:46 +0200
Subject: [PATCH 0991/1791] drm/amdgpu/gfx10: Disable gfxoff before disabling
 powergating.

Otherwise we get a full system lock (looks like a FW mess).

Copied the order from the GFX9 powergating code.

Fixes: 366468ff6c34 ("drm/amdgpu: Allow GfxOff on Vangogh as default")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2545
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 2e0234b43f433..8e86b2c23c0a5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -8036,8 +8036,14 @@ static int gfx_v10_0_set_powergating_state(void *handle,
 	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(10, 3, 6):
 	case IP_VERSION(10, 3, 7):
+		if (!enable)
+			amdgpu_gfx_off_ctrl(adev, false);
+
 		gfx_v10_cntl_pg(adev, enable);
-		amdgpu_gfx_off_ctrl(adev, enable);
+
+		if (enable)
+			amdgpu_gfx_off_ctrl(adev, true);
+
 		break;
 	default:
 		break;
-- 
GitLab


From ee30b8001cd85ba0ef287b216eb1dad58bb58159 Mon Sep 17 00:00:00 2001
From: "Guilherme G. Piccoli" <gpiccoli@igalia.com>
Date: Tue, 9 May 2023 18:49:47 +0200
Subject: [PATCH 0992/1791] drm/amdgpu/gfx11: Adjust gfxoff before powergating
 on gfx11 as well

(Bas: speculative change to mirror gfx10/gfx9)

Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 7bff50e58cffa..f77779c31043b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -5076,8 +5076,14 @@ static int gfx_v11_0_set_powergating_state(void *handle,
 		break;
 	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 4):
+		if (!enable)
+			amdgpu_gfx_off_ctrl(adev, false);
+
 		gfx_v11_cntl_pg(adev, enable);
-		amdgpu_gfx_off_ctrl(adev, enable);
+
+		if (enable)
+			amdgpu_gfx_off_ctrl(adev, true);
+
 		break;
 	default:
 		break;
-- 
GitLab


From 1d6ecab1ac0fdff8e62ff3ba506b606177010d08 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 9 May 2023 19:08:17 +0530
Subject: [PATCH 0993/1791] drm/amd/amdgpu: Fix warnings in amdgpu _object,
 _ring.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below warnings reported by checkpatch:

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: static const char * array should probably be static const char * const
WARNING: space prohibited between function name and open parenthesis '('
WARNING: braces {} are not necessary for single statement blocks
WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c   |  9 ++++-----
 2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 2bd1a54ee8665..9f0d5f02119ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -130,7 +130,7 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 	u32 c = 0;
 
 	if (domain & AMDGPU_GEM_DOMAIN_VRAM) {
-		unsigned visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
+		unsigned int visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
 
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
@@ -935,7 +935,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,
 		bo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 	amdgpu_bo_placement_from_domain(bo, domain);
 	for (i = 0; i < bo->placement.num_placement; i++) {
-		unsigned fpfn, lpfn;
+		unsigned int fpfn, lpfn;
 
 		fpfn = min_offset >> PAGE_SHIFT;
 		lpfn = max_offset >> PAGE_SHIFT;
@@ -1016,7 +1016,7 @@ void amdgpu_bo_unpin(struct amdgpu_bo *bo)
 	}
 }
 
-static const char *amdgpu_vram_names[] = {
+static const char * const amdgpu_vram_names[] = {
 	"UNKNOWN",
 	"GDDR1",
 	"DDR2",
@@ -1148,8 +1148,8 @@ void amdgpu_bo_get_tiling_flags(struct amdgpu_bo *bo, u64 *tiling_flags)
  * Returns:
  * 0 for success or a negative error code on failure.
  */
-int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata,
-			    uint32_t metadata_size, uint64_t flags)
+int amdgpu_bo_set_metadata(struct amdgpu_bo *bo, void *metadata,
+			   u32 metadata_size, uint64_t flags)
 {
 	struct amdgpu_bo_user *ubo;
 	void *buffer;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 1e9ae3bc59d39..d8a78de1f3351 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -78,7 +78,7 @@ unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type)
  * Allocate @ndw dwords in the ring buffer (all asics).
  * Returns 0 on success, error on failure.
  */
-int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw)
+int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned int ndw)
 {
 	/* Align requested size with padding so unlock_commit can
 	 * pad safely */
@@ -315,9 +315,8 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		     amdgpu_ring_max_ibs(ring->funcs->type) * ring->funcs->emit_ib_size;
 	max_ibs_dw = (max_ibs_dw + ring->funcs->align_mask) & ~ring->funcs->align_mask;
 
-	if (WARN_ON(max_ibs_dw > max_dw)) {
+	if (WARN_ON(max_ibs_dw > max_dw))
 		max_dw = max_ibs_dw;
-	}
 
 	ring->ring_size = roundup_pow_of_two(max_dw * 4 * sched_hw_submission);
 
@@ -573,13 +572,13 @@ void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
 	char name[32];
 
 	sprintf(name, "amdgpu_ring_%s", ring->name);
-	debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, ring,
+	debugfs_create_file_size(name, S_IFREG | 0444, root, ring,
 				 &amdgpu_debugfs_ring_fops,
 				 ring->ring_size + 12);
 
 	if (ring->mqd_obj) {
 		sprintf(name, "amdgpu_mqd_%s", ring->name);
-		debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, ring,
+		debugfs_create_file_size(name, S_IFREG | 0444, root, ring,
 					 &amdgpu_debugfs_mqd_fops,
 					 ring->mqd_size);
 	}
-- 
GitLab


From 6d99f3f4ea948f26b074a069c0406dd7f1cb0d8f Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Wed, 10 May 2023 16:13:48 +0800
Subject: [PATCH 0994/1791] drm/amdgpu: change gfx 11.0.4 external_id range

gfx 11.0.4 range starts from 0x80.

Fixes: 311d52367d0a ("drm/amdgpu: add soc21 common ip block support for GC 11.0.4")
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reported-by: Yogesh Mohan Marimuthu <Yogesh.Mohanmarimuthu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Tim Huang <Tim.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc21.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 0f82b8e83acbc..6bff936a6e552 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -711,7 +711,7 @@ static int soc21_common_early_init(void *handle)
 			AMD_PG_SUPPORT_VCN_DPG |
 			AMD_PG_SUPPORT_GFX_PG |
 			AMD_PG_SUPPORT_JPEG;
-		adev->external_rev_id = adev->rev_id + 0x1;
+		adev->external_rev_id = adev->rev_id + 0x80;
 		break;
 
 	default:
-- 
GitLab


From cabbdea1f1861098991768d7bbf5a49ed1608213 Mon Sep 17 00:00:00 2001
From: Daniil Dulov <d.dulov@aladdin.ru>
Date: Thu, 11 May 2023 04:23:14 -0700
Subject: [PATCH 0995/1791] drm/amdkfd: Fix potential deallocation of
 previously deallocated memory.

Pointer mqd_mem_obj can be deallocated in kfd_gtt_sa_allocate().
The function then returns non-zero value, which causes the second deallocation.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: d1f8f0d17d40 ("drm/amdkfd: Move non-sdma mqd allocation out of init_mqd")
Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index fdbfd725841ff..51b53110341bb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -115,18 +115,19 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
 			&(mqd_mem_obj->gtt_mem),
 			&(mqd_mem_obj->gpu_addr),
 			(void *)&(mqd_mem_obj->cpu_ptr), true);
+
+		if (retval) {
+			kfree(mqd_mem_obj);
+			return NULL;
+		}
 	} else {
 		retval = kfd_gtt_sa_allocate(kfd, sizeof(struct v9_mqd),
 				&mqd_mem_obj);
-	}
-
-	if (retval) {
-		kfree(mqd_mem_obj);
-		return NULL;
+		if (retval)
+			return NULL;
 	}
 
 	return mqd_mem_obj;
-
 }
 
 static void init_mqd(struct mqd_manager *mm, void **mqd,
-- 
GitLab


From 82ad22bbad008f84ec52208c5ba2c8f1cf55fd8d Mon Sep 17 00:00:00 2001
From: Jesse Zhang <jesse.zhang@amd.com>
Date: Fri, 12 May 2023 09:29:17 +0800
Subject: [PATCH 0996/1791] drm/amdgpu: Differentiate between Raven2 and
 Raven/Picasso according to revision id

Due to the raven2 and raven/picasso maybe have the same GC_HWIP version.
So differentiate them by revision id.

Signed-off-by: shanshengwang <shansheng.wang@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 33 ++++++++++++---------------
 1 file changed, 14 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index e093e83ae7398..46577b59cb042 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4000,30 +4000,25 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 		clock = clock_lo | (clock_hi << 32ULL);
 		break;
 	case IP_VERSION(9, 1, 0):
+	case IP_VERSION(9, 2, 2):
 		preempt_disable();
-		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
-		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
-		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
-		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
-		 * roughly every 42 seconds.
-		 */
-		if (hi_check != clock_hi) {
+		if (adev->rev_id >= 0x8) {
+			clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
+			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
+			hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
+		} else {
+			clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
 			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
-			clock_hi = hi_check;
+			hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
 		}
-		preempt_enable();
-		clock = clock_lo | (clock_hi << 32ULL);
-		break;
-	case IP_VERSION(9, 2, 2):
-		preempt_disable();
-		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
-		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
-		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
 		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
-		 * roughly every 42 seconds.
-		 */
+		* roughly every 42 seconds.
+		*/
 		if (hi_check != clock_hi) {
-			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
+			if (adev->rev_id >= 0x8)
+				clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
+			else
+				clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
 			clock_hi = hi_check;
 		}
 		preempt_enable();
-- 
GitLab


From e84e40dcb3fc972cbe41d228f1e45128285e0c47 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Fri, 12 May 2023 15:04:51 +0800
Subject: [PATCH 0997/1791] drm/amd/display: Simplify the calculation of
 variables

./drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c:586:37-39: WARNING !A || A && B is equivalent to !A || B.
./drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c:595:37-39: WARNING !A || A && B is equivalent to !A || B.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=4941
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 4950eaa4406b2..2de910e0ce75f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -583,8 +583,8 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context)
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 		struct hubp *hubp = pipe->plane_res.hubp;
 
-		if (!pipe->stream || (pipe->stream && !(pipe->stream->mall_stream_config.type == SUBVP_MAIN ||
-						pipe->stream->fpo_in_use))) {
+		if (!pipe->stream || !(pipe->stream->mall_stream_config.type == SUBVP_MAIN ||
+		    pipe->stream->fpo_in_use)) {
 			if (hubp && hubp->funcs->hubp_update_force_pstate_disallow)
 				hubp->funcs->hubp_update_force_pstate_disallow(hubp, false);
 		}
@@ -592,7 +592,7 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context)
 		/* Today only FPO uses cursor P-State force. Only clear cursor P-State force
 		 * if it's not FPO.
 		 */
-		if (!pipe->stream || (pipe->stream && !pipe->stream->fpo_in_use)) {
+		if (!pipe->stream || !pipe->stream->fpo_in_use) {
 			if (hubp && hubp->funcs->hubp_update_force_cursor_pstate_disallow)
 				hubp->funcs->hubp_update_force_cursor_pstate_disallow(hubp, false);
 		}
-- 
GitLab


From 17d62410aec363ec972f532fed5aba89b3f59ae7 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 11 May 2023 10:40:03 -0400
Subject: [PATCH 0998/1791] drm/amdgpu/gmc11: implement get_vbios_fb_size()

Implement get_vbios_fb_size() so we can properly reserve
the vbios splash screen to avoid potential artifacts on the
screen during the transition from the pre-OS console to the
OS console.

Acked-by: Sunil Khatri <sunil.khatri@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index f73c238f3145a..2f570fb5febe3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -31,6 +31,8 @@
 #include "umc_v8_10.h"
 #include "athub/athub_3_0_0_sh_mask.h"
 #include "athub/athub_3_0_0_offset.h"
+#include "dcn/dcn_3_2_0_offset.h"
+#include "dcn/dcn_3_2_0_sh_mask.h"
 #include "oss/osssys_6_0_0_offset.h"
 #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"
 #include "navi10_enum.h"
@@ -546,7 +548,24 @@ static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev,
 
 static unsigned gmc_v11_0_get_vbios_fb_size(struct amdgpu_device *adev)
 {
-	return 0;
+	u32 d1vga_control = RREG32_SOC15(DCE, 0, regD1VGA_CONTROL);
+	unsigned size;
+
+	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
+	} else {
+		u32 viewport;
+		u32 pitch;
+
+		viewport = RREG32_SOC15(DCE, 0, regHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
+		pitch = RREG32_SOC15(DCE, 0, regHUBPREQ0_DCSURF_SURFACE_PITCH);
+		size = (REG_GET_FIELD(viewport,
+					HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
+				REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) *
+				4);
+	}
+
+	return size;
 }
 
 static const struct amdgpu_gmc_funcs gmc_v11_0_gmc_funcs = {
-- 
GitLab


From 6e87c4229513904295674b84b6e2d12951567191 Mon Sep 17 00:00:00 2001
From: Alex Sierra <alex.sierra@amd.com>
Date: Mon, 24 Apr 2023 14:27:26 -0500
Subject: [PATCH 0999/1791] drm/amdgpu: improve wait logic at fence polling

Accomplish this by reading the seq number right away instead of sleep
for 5us. There are certain cases where the fence is ready almost
immediately. Sleep number granularity was also reduced as the majority
of the kiq tlb flush takes between 2us to 6us.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 1994eafd3e71a..b7d648dd0170b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -376,14 +376,11 @@ signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring,
 				      uint32_t wait_seq,
 				      signed long timeout)
 {
-	uint32_t seq;
-
-	do {
-		seq = amdgpu_fence_read(ring);
-		udelay(5);
-		timeout -= 5;
-	} while ((int32_t)(wait_seq - seq) > 0 && timeout > 0);
 
+	while ((int32_t)(wait_seq - amdgpu_fence_read(ring)) > 0 && timeout > 0) {
+		udelay(2);
+		timeout -= 2;
+	}
 	return timeout > 0 ? timeout : 0;
 }
 /**
-- 
GitLab


From 3083b1007d4b8d377f8e2b5ce349a275a2fff725 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Tue, 9 May 2023 16:15:27 +0800
Subject: [PATCH 1000/1791] drm/amdgpu: skip disabling fence driver src_irqs
 when device is unplugged

When performing device unbind or halt, we have disabled all irqs at the
very begining like amdgpu_pci_remove or amdgpu_device_halt. So
amdgpu_irq_put for irqs stored in fence driver should not be called
any more, otherwise, below calltrace will arrive.

[  139.114088] WARNING: CPU: 2 PID: 1550 at drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c:616 amdgpu_irq_put+0xf6/0x110 [amdgpu]
[  139.114655] Call Trace:
[  139.114655]  <TASK>
[  139.114657]  amdgpu_fence_driver_hw_fini+0x93/0x130 [amdgpu]
[  139.114836]  amdgpu_device_fini_hw+0xb6/0x350 [amdgpu]
[  139.114955]  amdgpu_driver_unload_kms+0x51/0x70 [amdgpu]
[  139.115075]  amdgpu_pci_remove+0x63/0x160 [amdgpu]
[  139.115193]  ? __pm_runtime_resume+0x64/0x90
[  139.115195]  pci_device_remove+0x3a/0xb0
[  139.115197]  device_remove+0x43/0x70
[  139.115198]  device_release_driver_internal+0xbd/0x140

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index b7d648dd0170b..928814492d5b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -578,7 +578,8 @@ void amdgpu_fence_driver_hw_fini(struct amdgpu_device *adev)
 		if (r)
 			amdgpu_fence_driver_force_completion(ring);
 
-		if (ring->fence_drv.irq_src)
+		if (!drm_dev_is_unplugged(adev_to_drm(adev)) &&
+		    ring->fence_drv.irq_src)
 			amdgpu_irq_put(adev, ring->fence_drv.irq_src,
 				       ring->fence_drv.irq_type);
 
-- 
GitLab


From 572773992e31e0e31692adf6797a3bf7e094097c Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 11 May 2023 15:41:27 +0800
Subject: [PATCH 1001/1791] drm/amd/pm: fix possible power mode mismatch
 between driver and PMFW

PMFW may boots the ASIC with a different power mode from the system's
real one. Notify PMFW explicitly the power mode the system in. This
is needed only when ACDC switch via gpio is not supported.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 18 +++++++++++++++++
 .../gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c   | 20 +------------------
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c  |  1 +
 3 files changed, 20 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 5633c5797e85a..2ddf5198e5c48 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -733,6 +733,24 @@ static int smu_late_init(void *handle)
 		return ret;
 	}
 
+	/*
+	 * Explicitly notify PMFW the power mode the system in. Since
+	 * the PMFW may boot the ASIC with a different mode.
+	 * For those supporting ACDC switch via gpio, PMFW will
+	 * handle the switch automatically. Driver involvement
+	 * is unnecessary.
+	 */
+	if (!smu->dc_controlled_by_gpio) {
+		ret = smu_set_power_source(smu,
+					   adev->pm.ac_power ? SMU_POWER_SOURCE_AC :
+					   SMU_POWER_SOURCE_DC);
+		if (ret) {
+			dev_err(adev->dev, "Failed to switch to %s mode!\n",
+				adev->pm.ac_power ? "AC" : "DC");
+			return ret;
+		}
+	}
+
 	if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 1)) ||
 	    (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 3)))
 		return 0;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
index c4000518dc56d..275f708db6362 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
@@ -3413,26 +3413,8 @@ static int navi10_post_smu_init(struct smu_context *smu)
 		return 0;
 
 	ret = navi10_run_umc_cdr_workaround(smu);
-	if (ret) {
+	if (ret)
 		dev_err(adev->dev, "Failed to apply umc cdr workaround!\n");
-		return ret;
-	}
-
-	if (!smu->dc_controlled_by_gpio) {
-		/*
-		 * For Navi1X, manually switch it to AC mode as PMFW
-		 * may boot it with DC mode.
-		 */
-		ret = smu_v11_0_set_power_source(smu,
-						 adev->pm.ac_power ?
-						 SMU_POWER_SOURCE_AC :
-						 SMU_POWER_SOURCE_DC);
-		if (ret) {
-			dev_err(adev->dev, "Failed to switch to %s mode!\n",
-					adev->pm.ac_power ? "AC" : "DC");
-			return ret;
-		}
-	}
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 3d9ff46706fb7..98a33f8ee2095 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -1770,6 +1770,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.enable_mgpu_fan_boost = smu_v13_0_7_enable_mgpu_fan_boost,
 	.get_power_limit = smu_v13_0_7_get_power_limit,
 	.set_power_limit = smu_v13_0_set_power_limit,
+	.set_power_source = smu_v13_0_set_power_source,
 	.get_power_profile_mode = smu_v13_0_7_get_power_profile_mode,
 	.set_power_profile_mode = smu_v13_0_7_set_power_profile_mode,
 	.set_tool_table_location = smu_v13_0_set_tool_table_location,
-- 
GitLab


From f1822392128e00c1ae1579fe050ae3797db6ea02 Mon Sep 17 00:00:00 2001
From: Mustapha Ghaddar <mghaddar@amd.com>
Date: Fri, 28 Apr 2023 12:13:21 -0400
Subject: [PATCH 1002/1791] drm/amd/display: enable dpia validate

Use dpia_validate_usb4_bw() function

Fixes: a8b537605e22 ("drm/amd/display: Add function pointer for validate bw usb4")
Reviewed-by: Roman Li <roman.li@amd.com>
Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Mustapha Ghaddar <mghaddar@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/link/link_validation.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c
index d4b7da526f0a5..e8b2fc4002a52 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c
@@ -359,5 +359,8 @@ bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const un
 		link[i] = stream[i].link;
 		bw_needed[i] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing);
 	}
+
+	ret = dpia_validate_usb4_bw(link, bw_needed, num_streams);
+
 	return ret;
 }
-- 
GitLab


From 1d8355ad922423c9f765a644ed04526a6273d9ee Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Tue, 2 May 2023 09:54:56 -0400
Subject: [PATCH 1003/1791] drm/amd/display: Only skip update for DCFCLK, UCLK,
 FCLK on overclock

[Description]
- Update clocks is skipped in the GPU overclock sequence
- However, we still need to update DISPCLK, DPPCLK, and DTBCLK
  because the GPU overclock sequence could temporarily disable
  ODM 2:1 combine because we disable all planes in the sequence

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  | 24 +++++++++++--------
 drivers/gpu/drm/amd/display/dc/dc.h           |  7 +++++-
 2 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 85e963ec25ab1..1df623b298a92 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -460,9 +460,6 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 	bool p_state_change_support;
 	bool fclk_p_state_change_support;
 
-	if (dc->work_arounds.skip_clock_update)
-		return;
-
 	if (clk_mgr_base->clks.dispclk_khz == 0 ||
 			(dc->debug.force_clock_mode & 0x1)) {
 		/* This is from resume or boot up, if forced_clock cfg option used,
@@ -489,7 +486,8 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 
 		fclk_p_state_change_support = new_clocks->fclk_p_state_change_support;
 
-		if (should_update_pstate_support(safe_to_lower, fclk_p_state_change_support, clk_mgr_base->clks.fclk_p_state_change_support)) {
+		if (should_update_pstate_support(safe_to_lower, fclk_p_state_change_support, clk_mgr_base->clks.fclk_p_state_change_support) &&
+				!dc->work_arounds.clock_update_disable_mask.fclk) {
 			clk_mgr_base->clks.fclk_p_state_change_support = fclk_p_state_change_support;
 
 			/* To enable FCLK P-state switching, send FCLK_PSTATE_SUPPORTED message to PMFW */
@@ -503,12 +501,14 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 			new_clocks->dcfclk_khz = (new_clocks->dcfclk_khz > (dc->debug.force_min_dcfclk_mhz * 1000)) ?
 					new_clocks->dcfclk_khz : (dc->debug.force_min_dcfclk_mhz * 1000);
 
-		if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz)) {
+		if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz) &&
+				!dc->work_arounds.clock_update_disable_mask.dcfclk) {
 			clk_mgr_base->clks.dcfclk_khz = new_clocks->dcfclk_khz;
 			dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DCFCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dcfclk_khz));
 		}
 
-		if (should_set_clock(safe_to_lower, new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz)) {
+		if (should_set_clock(safe_to_lower, new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz) &&
+				!dc->work_arounds.clock_update_disable_mask.dcfclk_ds) {
 			clk_mgr_base->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
 			dcn30_smu_set_min_deep_sleep_dcef_clk(clk_mgr, khz_to_mhz_ceil(clk_mgr_base->clks.dcfclk_deep_sleep_khz));
 		}
@@ -527,7 +527,8 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 		}
 
 		p_state_change_support = new_clocks->p_state_change_support;
-		if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support)) {
+		if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support) &&
+				!dc->work_arounds.clock_update_disable_mask.uclk) {
 			clk_mgr_base->clks.p_state_change_support = p_state_change_support;
 
 			/* to disable P-State switching, set UCLK min = max */
@@ -541,20 +542,23 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 			update_fclk = true;
 		}
 
-		if (clk_mgr_base->ctx->dce_version != DCN_VERSION_3_21 && !clk_mgr_base->clks.fclk_p_state_change_support && update_fclk) {
+		if (clk_mgr_base->ctx->dce_version != DCN_VERSION_3_21 && !clk_mgr_base->clks.fclk_p_state_change_support && update_fclk &&
+				!dc->work_arounds.clock_update_disable_mask.fclk) {
 			/* Handle code for sending a message to PMFW that FCLK P-state change is not supported */
 			dcn32_smu_send_fclk_pstate_message(clk_mgr, FCLK_PSTATE_NOTSUPPORTED);
 		}
 
 		/* Always update saved value, even if new value not set due to P-State switching unsupported */
-		if (should_set_clock(safe_to_lower, new_clocks->dramclk_khz, clk_mgr_base->clks.dramclk_khz)) {
+		if (should_set_clock(safe_to_lower, new_clocks->dramclk_khz, clk_mgr_base->clks.dramclk_khz) &&
+				!dc->work_arounds.clock_update_disable_mask.uclk) {
 			clk_mgr_base->clks.dramclk_khz = new_clocks->dramclk_khz;
 			update_uclk = true;
 		}
 
 		/* set UCLK to requested value if P-State switching is supported, or to re-enable P-State switching */
 		if (clk_mgr_base->clks.p_state_change_support &&
-				(update_uclk || !clk_mgr_base->clks.prev_p_state_change_support))
+				(update_uclk || !clk_mgr_base->clks.prev_p_state_change_support) &&
+				!dc->work_arounds.clock_update_disable_mask.uclk)
 			dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz));
 
 		if (clk_mgr_base->clks.num_ways != new_clocks->num_ways &&
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index f4448b545d701..e89de10789642 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -272,8 +272,13 @@ struct dc_bug_wa {
 	bool dedcn20_305_wa;
 	bool skip_clock_update;
 	bool lt_early_cr_pattern;
+	struct {
+		uint8_t uclk : 1;
+		uint8_t fclk : 1;
+		uint8_t dcfclk : 1;
+		uint8_t dcfclk_ds: 1;
+	} clock_update_disable_mask;
 };
-
 struct dc_dcc_surface_param {
 	struct dc_size surface_size;
 	enum surface_pixel_format format;
-- 
GitLab


From 6ba5a269cdc9f447be882bbf99548361c8ebc254 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Tue, 2 May 2023 12:27:26 -0400
Subject: [PATCH 1004/1791] drm/amd/display: Update vactive margin and max
 vblank for fpo + vactive

[Description]
- Some 1920x1080@60hz displays have VBLANK time > 600us which we
  still want to accept for FPO + Vactive configs based on testing
	- Increase max VBLANK time to 1000us to allow these configs
	  for FPO + Vactive
- Increase minimum vactive switch margin for FPO + Vactive to 200us
	- Based on testing, 1920x1080@120hz can have a switch margin
	  of ~160us which requires significantly longer FPO stretch
	  margin (5ms) which we don't want to accept for now
- Also move margins into debug option

Reviewed-by: Jun Lei <jun.lei@amd.com>
Reviewed-by: Nevenko Stupar <nevenko.stupar@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                           | 2 ++
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c         | 2 ++
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h         | 1 -
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c       | 2 ++
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c          | 3 +--
 6 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index e89de10789642..1ebb8d3573f48 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -893,6 +893,8 @@ struct dc_debug_options {
 	bool minimize_dispclk_using_odm;
 	bool disable_subvp_high_refresh;
 	bool disable_dp_plus_plus_wa;
+	uint32_t fpo_vactive_min_active_margin_us;
+	uint32_t fpo_vactive_max_blank_us;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 4de2f8813dce3..98c394f9f8cfc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -730,6 +730,8 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_boot_optimizations = false,
 	.disable_subvp_high_refresh = true,
 	.disable_dp_plus_plus_wa = true,
+	.fpo_vactive_min_active_margin_us = 200,
+	.fpo_vactive_max_blank_us = 1000,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 42ccfd13a37c7..58826e0aa76e3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -39,7 +39,6 @@
 #define DCN3_2_MBLK_HEIGHT_8BPE 64
 #define DCN3_2_VMIN_DISPCLK_HZ 717000000
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
-#define DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US 100 // Only allow FPO + Vactive if active margin >= 100
 #define SUBVP_HIGH_REFRESH_LIST_LEN 3
 #define DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ 1800
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index df912c333bbdb..a8082580df92a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -626,7 +626,7 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre
 		DC_FP_END();
 
 		DC_FP_START();
-		is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US);
+		is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, dc->debug.fpo_vactive_min_active_margin_us);
 		DC_FP_END();
 		if (!is_fpo_vactive || dc->debug.disable_fpo_vactive)
 			return NULL;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index 4c1e0f5a5f097..f4cd9749ffdf8 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -728,6 +728,8 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
 	.disable_subvp_high_refresh = true,
+	.fpo_vactive_min_active_margin_us = 200,
+	.fpo_vactive_max_blank_us = 1000,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index f7e45d935a297..8c60b88c7d1ab 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -35,7 +35,6 @@
 
 #define DC_LOGGER_INIT(logger)
 
-static const unsigned int MAX_FPO_VACTIVE_BLANK_US = 600;
 static const struct subvp_high_refresh_list subvp_high_refresh_list = {
 			.min_refresh = 120,
 			.max_refresh = 165,
@@ -2937,7 +2936,7 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint
 		blank_us = ((pipe->stream->timing.v_total - pipe->stream->timing.v_addressable) * pipe->stream->timing.h_total /
 				(double)(pipe->stream->timing.pix_clk_100hz * 100)) * 1000000;
 		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us &&
-				!(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed) && blank_us < MAX_FPO_VACTIVE_BLANK_US) {
+				!(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed) && blank_us < dc->debug.fpo_vactive_max_blank_us) {
 			vactive_found = true;
 			break;
 		}
-- 
GitLab


From 9c384ee8f997d0646844cd431f1c23e6dbf84a98 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Date: Wed, 26 Apr 2023 18:17:13 -0600
Subject: [PATCH 1005/1791] drm/amd/display: Convert connector signal id to
 string

To improve the readability of the of the log, this commit introduces a
function that converts the signal type id to a human-readable string.

Reviewed-by: Jerry Zuo <jerry.zuo@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/link_factory.c    |  6 ++--
 .../drm/amd/display/include/signal_types.h    | 28 +++++++++++++++++++
 2 files changed, 30 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
index 1515c817f03be..ac1c3e2e7c1d6 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
@@ -563,11 +563,9 @@ static bool construct_phy(struct dc_link *link,
 		goto create_fail;
 	}
 
-	/* TODO: #DAL3 Implement id to str function.*/
-	LINK_INFO("Connector[%d] description:"
-		  "signal %d\n",
+	LINK_INFO("Connector[%d] description: signal: %s\n",
 		  init_params->connector_index,
-		  link->connector_signal);
+		  signal_type_to_string(link->connector_signal));
 
 	ddc_service_init_data.ctx = link->ctx;
 	ddc_service_init_data.id = link->link_id;
diff --git a/drivers/gpu/drm/amd/display/include/signal_types.h b/drivers/gpu/drm/amd/display/include/signal_types.h
index 23a308c3eccbf..325c5ba4c82a6 100644
--- a/drivers/gpu/drm/amd/display/include/signal_types.h
+++ b/drivers/gpu/drm/amd/display/include/signal_types.h
@@ -44,6 +44,34 @@ enum signal_type {
 	SIGNAL_TYPE_VIRTUAL		= (1 << 9),	/* Virtual Display */
 };
 
+static inline const char *signal_type_to_string(const int type)
+{
+	switch (type) {
+	case SIGNAL_TYPE_NONE:
+		return "No signal";
+	case SIGNAL_TYPE_DVI_SINGLE_LINK:
+		return "DVI: Single Link";
+	case SIGNAL_TYPE_DVI_DUAL_LINK:
+		return "DVI: Dual Link";
+	case SIGNAL_TYPE_HDMI_TYPE_A:
+		return "HDMI: TYPE A";
+	case SIGNAL_TYPE_LVDS:
+		return "LVDS";
+	case SIGNAL_TYPE_RGB:
+		return "RGB";
+	case SIGNAL_TYPE_DISPLAY_PORT:
+		return "Display Port";
+	case SIGNAL_TYPE_DISPLAY_PORT_MST:
+		return "Display Port: MST";
+	case SIGNAL_TYPE_EDP:
+		return "Embedded Display Port";
+	case SIGNAL_TYPE_VIRTUAL:
+		return "Virtual";
+	default:
+		return "Unknown";
+	}
+}
+
 /* help functions for signal types manipulation */
 static inline bool dc_is_hdmi_tmds_signal(enum signal_type signal)
 {
-- 
GitLab


From 1a4bcdbea4319efeb26cc4b05be859a7867e02dc Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Tue, 25 Apr 2023 14:29:48 -0400
Subject: [PATCH 1006/1791] drm/amd/display: Fix possible underflow for
 displays with large vblank

[Why]
Underflow observed when using a display with a large vblank region
and low refresh rate

[How]
Simplify calculation of vblank_nom

Increase value for VBlankNomDefaultUS to 800us

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 1d00eb9e73c62..554152371eb53 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -33,7 +33,7 @@
 #include "dml/display_mode_vba.h"
 
 struct _vcs_dpi_ip_params_st dcn3_14_ip = {
-	.VBlankNomDefaultUS = 668,
+	.VBlankNomDefaultUS = 800,
 	.gpuvm_enable = 1,
 	.gpuvm_max_page_table_levels = 1,
 	.hostvm_enable = 1,
@@ -286,7 +286,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *pipe;
 	bool upscaled = false;
-	bool isFreesyncVideo = false;
+	const unsigned int max_allowed_vblank_nom = 1023;
 
 	dc_assert_fp_enabled();
 
@@ -300,16 +300,11 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 		pipe = &res_ctx->pipe_ctx[i];
 		timing = &pipe->stream->timing;
 
-		isFreesyncVideo = pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min;
-		isFreesyncVideo = isFreesyncVideo && pipe->stream->adjust.v_total_min > timing->v_total;
-
-		if (!isFreesyncVideo) {
-			pipes[pipe_cnt].pipe.dest.vblank_nom =
-				dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0));
-		} else {
-			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
-			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
-		}
+		pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, dcn3_14_ip.VBlankNomDefaultUS);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
 
 		if (pipe->plane_state &&
 				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
-- 
GitLab


From d205a800a66e46430ab93c0d450393233d39931a Mon Sep 17 00:00:00 2001
From: "Leo (Hanghong) Ma" <hanghong.ma@amd.com>
Date: Wed, 12 Apr 2023 14:02:01 -0400
Subject: [PATCH 1007/1791] drm/amd/display: Add visual confirm color support
 for MCLK switch

[Why && How]
We would like to have visual confirm color support for MCLK switch.
	1. Set visual confirm color to yellow: Vblank MCLK switch.
	2. Set visual confirm color to cyan: FPO + Vblank MCLK
	switch.
	3. Set visual confirm color to pink: Vactive MCLK switch.

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Leo (Hanghong) Ma <hanghong.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 47 +++++++++++++++--
 .../drm/amd/display/dc/core/dc_hw_sequencer.c | 50 +++++++++++++++++--
 drivers/gpu/drm/amd/display/dc/dc.h           |  1 +
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c | 22 +++-----
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.h |  1 -
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.c    | 26 +---------
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.h    |  5 --
 .../gpu/drm/amd/display/dc/dcn20/dcn20_init.c |  2 +-
 .../drm/amd/display/dc/dcn201/dcn201_hwseq.c  |  4 +-
 .../drm/amd/display/dc/dcn201/dcn201_init.c   |  2 +-
 .../gpu/drm/amd/display/dc/dcn21/dcn21_init.c |  2 +-
 .../gpu/drm/amd/display/dc/dcn30/dcn30_init.c |  2 +-
 .../drm/amd/display/dc/dcn301/dcn301_init.c   |  2 +-
 .../gpu/drm/amd/display/dc/dcn31/dcn31_init.c |  2 +-
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |  2 +-
 .../gpu/drm/amd/display/dc/dcn32/dcn32_init.c |  2 +-
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  |  7 +++
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  2 +
 .../gpu/drm/amd/display/dc/inc/hw_sequencer.h |  9 +++-
 19 files changed, 125 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index c26dfdd48dd93..2c639cd346d78 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1119,6 +1119,33 @@ static void phantom_pipe_blank(
 		hws->funcs.wait_for_blank_complete(opp);
 }
 
+static void dc_update_viusal_confirm_color(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx)
+{
+	if (dc->ctx->dce_version >= DCN_VERSION_1_0) {
+		memset(&pipe_ctx->visual_confirm_color, 0, sizeof(struct tg_color));
+
+		if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR)
+			get_hdr_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
+		else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE)
+			get_surface_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
+		else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE)
+			get_surface_tile_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
+		else {
+			if (dc->ctx->dce_version < DCN_VERSION_2_0)
+				color_space_to_black_color(
+					dc, pipe_ctx->stream->output_color_space, &(pipe_ctx->visual_confirm_color));
+		}
+		if (dc->ctx->dce_version >= DCN_VERSION_2_0) {
+			if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE)
+				get_mpctree_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color));
+			else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP)
+				get_subvp_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color));
+			else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH)
+				get_mclk_switch_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color));
+		}
+	}
+}
+
 static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 {
 	int i, j;
@@ -1189,6 +1216,9 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context)
 			dc_rem_all_planes_for_stream(dc, old_stream, dangling_context);
 			disable_all_writeback_pipes_for_stream(dc, old_stream, dangling_context);
 
+			if (pipe->stream && pipe->plane_state)
+				dc_update_viusal_confirm_color(dc, context, pipe);
+
 			if (dc->hwss.apply_ctx_for_surface) {
 				apply_ctx_interdependent_lock(dc, dc->current_state, old_stream, true);
 				dc->hwss.apply_ctx_for_surface(dc, old_stream, 0, dangling_context);
@@ -3456,6 +3486,14 @@ static void commit_planes_for_stream(struct dc *dc,
 		}
 	}
 
+	if (dc->debug.visual_confirm)
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+			if (pipe->stream && pipe->plane_state)
+				dc_update_viusal_confirm_color(dc, context, pipe);
+		}
+
 	if (stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE) {
 		struct pipe_ctx *mpcc_pipe;
 		struct pipe_ctx *odm_pipe;
@@ -3539,15 +3577,14 @@ static void commit_planes_for_stream(struct dc *dc,
 		for (j = 0; j < dc->res_pool->pipe_count; j++) {
 			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
 
-			if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP &&
+			if ((dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP ||
+				dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH) &&
 				pipe_ctx->stream && pipe_ctx->plane_state) {
-				/* Only update visual confirm for SUBVP here.
+				/* Only update visual confirm for SUBVP and Mclk switching here.
 				 * The bar appears on all pipes, so we need to update the bar on all displays,
 				 * so the information doesn't get stale.
 				 */
-				struct mpcc_blnd_cfg blnd_cfg = { 0 };
-
-				dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color,
+				dc->hwss.update_visual_confirm_color(dc, pipe_ctx,
 						pipe_ctx->plane_res.hubp->inst);
 			}
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index 2acbf692193f7..8a98b8dd008ef 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -421,6 +421,7 @@ void get_hdr_visual_confirm_color(
 
 void get_subvp_visual_confirm_color(
 		struct dc *dc,
+		struct dc_state *context,
 		struct pipe_ctx *pipe_ctx,
 		struct tg_color *color)
 {
@@ -428,15 +429,17 @@ void get_subvp_visual_confirm_color(
 	bool enable_subvp = false;
 	int i;
 
-	if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx)
+	if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !context)
 		return;
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 
 		if (pipe->stream && pipe->stream->mall_stream_config.paired_stream &&
 		    pipe->stream->mall_stream_config.type == SUBVP_MAIN) {
 			/* SubVP enable - red */
+			color->color_g_y = 0;
+			color->color_b_cb = 0;
 			color->color_r_cr = color_value;
 			enable_subvp = true;
 
@@ -448,12 +451,51 @@ void get_subvp_visual_confirm_color(
 
 	if (enable_subvp && pipe_ctx->stream->mall_stream_config.type == SUBVP_NONE) {
 		color->color_r_cr = 0;
-		if (pipe_ctx->stream->ignore_msa_timing_param == 1)
+		if (pipe_ctx->stream->allow_freesync == 1) {
 			/* SubVP enable and DRR on - green */
+			color->color_b_cb = 0;
 			color->color_g_y = color_value;
-		else
+		} else {
 			/* SubVP enable and No DRR - blue */
+			color->color_g_y = 0;
+			color->color_b_cb = color_value;
+		}
+	}
+}
+
+void get_mclk_switch_visual_confirm_color(
+		struct dc *dc,
+		struct dc_state *context,
+		struct pipe_ctx *pipe_ctx,
+		struct tg_color *color)
+{
+	uint32_t color_value = MAX_TG_COLOR_VALUE;
+	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+
+	if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !vba || !context)
+		return;
+
+	if (vba->DRAMClockChangeSupport[vba->VoltageLevel][vba->maxMpcComb] !=
+			dm_dram_clock_change_unsupported) {
+		/* MCLK switching is supported */
+		if (!pipe_ctx->has_vactive_margin) {
+			/* In Vblank - yellow */
+			color->color_r_cr = color_value;
+			color->color_g_y = color_value;
+
+			if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+				/* FPO + Vblank - cyan */
+				color->color_r_cr = 0;
+				color->color_g_y  = color_value;
+				color->color_b_cb = color_value;
+			}
+		} else {
+			/* In Vactive - pink */
+			color->color_r_cr = color_value;
 			color->color_b_cb = color_value;
+		}
+		/* SubVP */
+		get_subvp_visual_confirm_color(dc, context, pipe_ctx, color);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 1ebb8d3573f48..8be2e6d6d8884 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -426,6 +426,7 @@ enum visual_confirm {
 	VISUAL_CONFIRM_FAMS = 7,
 	VISUAL_CONFIRM_SWIZZLE = 9,
 	VISUAL_CONFIRM_SUBVP = 14,
+	VISUAL_CONFIRM_MCLK_SWITCH = 16,
 };
 
 enum dc_psr_power_opts {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index a7ad1d7bc43e5..905246a2ece45 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -2602,23 +2602,15 @@ static void dcn10_update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state
 		dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params);
 }
 
-void dcn10_update_visual_confirm_color(struct dc *dc, struct pipe_ctx *pipe_ctx, struct tg_color *color, int mpcc_id)
+void dcn10_update_visual_confirm_color(struct dc *dc,
+		struct pipe_ctx *pipe_ctx,
+		int mpcc_id)
 {
 	struct mpc *mpc = dc->res_pool->mpc;
 
-	if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR)
-		get_hdr_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE)
-		get_surface_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE)
-		get_surface_tile_visual_confirm_color(pipe_ctx, color);
-	else
-		color_space_to_black_color(
-				dc, pipe_ctx->stream->output_color_space, color);
-
 	if (mpc->funcs->set_bg_color) {
-		memcpy(&pipe_ctx->plane_state->visual_confirm_color, color, sizeof(struct tg_color));
-		mpc->funcs->set_bg_color(mpc, color, mpcc_id);
+		memcpy(&pipe_ctx->plane_state->visual_confirm_color, &(pipe_ctx->visual_confirm_color), sizeof(struct tg_color));
+		mpc->funcs->set_bg_color(mpc, &(pipe_ctx->visual_confirm_color), mpcc_id);
 	}
 }
 
@@ -2671,7 +2663,7 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	/* If there is no full update, don't need to touch MPC tree*/
 	if (!pipe_ctx->plane_state->update_flags.bits.full_update) {
 		mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id);
-		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 		return;
 	}
 
@@ -2693,7 +2685,7 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 			NULL,
 			hubp->inst,
 			mpcc_id);
-	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 
 	ASSERT(new_mpcc != NULL);
 	hubp->opp_id = pipe_ctx->stream_res.opp->inst;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
index 0ef7bf7ddb75e..ef6d56da417cd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
@@ -202,7 +202,6 @@ void dcn10_get_dcc_en_bits(struct dc *dc, int *dcc_en_bits);
 void dcn10_update_visual_confirm_color(
 		struct dc *dc,
 		struct pipe_ctx *pipe_ctx,
-		struct tg_color *color,
 		int mpcc_id);
 
 #endif /* __DC_HWSS_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index f49c1c0d62747..b3e187b1347d8 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2580,28 +2580,6 @@ void dcn20_reset_hw_ctx_wrap(
 	}
 }
 
-void dcn20_update_visual_confirm_color(struct dc *dc, struct pipe_ctx *pipe_ctx, struct tg_color *color, int mpcc_id)
-{
-	struct mpc *mpc = dc->res_pool->mpc;
-
-	// input to MPCC is always RGB, by default leave black_color at 0
-	if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR)
-		get_hdr_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE)
-		get_surface_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE)
-		get_mpctree_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE)
-		get_surface_tile_visual_confirm_color(pipe_ctx, color);
-	else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP)
-		get_subvp_visual_confirm_color(dc, pipe_ctx, color);
-
-	if (mpc->funcs->set_bg_color) {
-		memcpy(&pipe_ctx->plane_state->visual_confirm_color, color, sizeof(struct tg_color));
-		mpc->funcs->set_bg_color(mpc, color, mpcc_id);
-	}
-}
-
 void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 {
 	struct hubp *hubp = pipe_ctx->plane_res.hubp;
@@ -2657,7 +2635,7 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 	if (!pipe_ctx->plane_state->update_flags.bits.full_update &&
 		!pipe_ctx->update_flags.bits.mpcc) {
 		mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id);
-		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 		return;
 	}
 
@@ -2679,7 +2657,7 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 			NULL,
 			hubp->inst,
 			mpcc_id);
-	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 
 	ASSERT(new_mpcc != NULL);
 	hubp->opp_id = pipe_ctx->stream_res.opp->inst;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h
index 33a36c02b2f8f..01901b08644c5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h
@@ -150,10 +150,5 @@ void dcn20_set_disp_pattern_generator(const struct dc *dc,
 		const struct tg_color *solid_color,
 		int width, int height, int offset);
 
-void dcn20_update_visual_confirm_color(struct dc *dc,
-		struct pipe_ctx *pipe_ctx,
-		struct tg_color *color,
-		int mpcc_id);
-
 #endif /* __DC_HWSS_DCN20_H__ */
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
index 7c5817c426faa..4192c522e59aa 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
@@ -102,7 +102,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = {
 	.disable_link_output = dce110_disable_link_output,
 	.set_disp_pattern_generator = dcn20_set_disp_pattern_generator,
 	.get_dcc_en_bits = dcn10_get_dcc_en_bits,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn20_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
index 1aeb04fbd89d3..75472d53ff52f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
@@ -496,7 +496,7 @@ void dcn201_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 
 	/* If there is no full update, don't need to touch MPC tree*/
 	if (!pipe_ctx->plane_state->update_flags.bits.full_update) {
-		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+		dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 		mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id);
 		return;
 	}
@@ -521,7 +521,7 @@ void dcn201_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
 					dc->res_pool->mpc, mpcc_id);
 
 	/* Call MPC to insert new plane */
-	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id);
+	dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id);
 	new_mpcc = mpc->funcs->insert_plane(dc->res_pool->mpc,
 			mpc_tree_params,
 			&blnd_cfg,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c
index 9c16633e473a0..92dd4cddbab8a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c
@@ -91,7 +91,7 @@ static const struct hw_sequencer_funcs dcn201_funcs = {
 	.enable_dp_link_output = dce110_enable_dp_link_output,
 	.disable_link_output = dce110_disable_link_output,
 	.set_disp_pattern_generator = dcn20_set_disp_pattern_generator,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn201_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
index fe1a8e2e08ef1..8b58ce1db0352 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
@@ -106,7 +106,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = {
 	.is_abm_supported = dcn21_is_abm_supported,
 	.set_disp_pattern_generator = dcn20_set_disp_pattern_generator,
 	.get_dcc_en_bits = dcn10_get_dcc_en_bits,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn21_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
index 3216d10c58ba7..18e94d8ae54fc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
@@ -106,7 +106,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = {
 	.disable_link_output = dce110_disable_link_output,
 	.set_disp_pattern_generator = dcn30_set_disp_pattern_generator,
 	.get_dcc_en_bits = dcn10_get_dcc_en_bits,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.is_abm_supported = dcn21_is_abm_supported
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c
index 6192851c59ed8..257df8660b4ca 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c
@@ -107,7 +107,7 @@ static const struct hw_sequencer_funcs dcn301_funcs = {
 	.get_dcc_en_bits = dcn10_get_dcc_en_bits,
 	.optimize_pwr_state = dcn21_optimize_pwr_state,
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn301_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
index 8598ea233ef3e..ba9e7dee6e5e5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
@@ -110,7 +110,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = {
 	.set_disp_pattern_generator = dcn30_set_disp_pattern_generator,
 	.optimize_pwr_state = dcn21_optimize_pwr_state,
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn31_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index ed8a1b94c0068..7a28c7bb25d20 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -112,7 +112,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = {
 	.set_disp_pattern_generator = dcn30_set_disp_pattern_generator,
 	.optimize_pwr_state = dcn21_optimize_pwr_state,
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 };
 
 static const struct hwseq_private_funcs dcn314_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
index 8085f2acb1a96..24a890d879b80 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
@@ -109,7 +109,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = {
 	.commit_subvp_config = dcn32_commit_subvp_config,
 	.enable_phantom_streams = dcn32_enable_phantom_streams,
 	.subvp_pipe_control_lock = dcn32_subvp_pipe_control_lock,
-	.update_visual_confirm_color = dcn20_update_visual_confirm_color,
+	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.update_phantom_vp_position = dcn32_update_phantom_vp_position,
 	.update_dsc_pg = dcn32_update_dsc_pg,
 	.apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom,
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 8c60b88c7d1ab..d8d8fcd5ef1ff 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -1324,6 +1324,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context,
 	int i, pipe_idx, active_hubp_count = 0;
 	bool usr_retraining_support = false;
 	bool unbounded_req_enabled = false;
+	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 
 	dc_assert_fp_enabled();
 
@@ -1405,6 +1406,11 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context,
 
 		context->res_ctx.pipe_ctx[i].surface_size_in_mall_bytes = get_surface_size_in_mall(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx);
 
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0)
+			context->res_ctx.pipe_ctx[i].has_vactive_margin = true;
+		else
+			context->res_ctx.pipe_ctx[i].has_vactive_margin = false;
+
 		/* MALL Allocation Sizes */
 		/* count from active, top pipes per plane only */
 		if (context->res_ctx.pipe_ctx[i].stream && context->res_ctx.pipe_ctx[i].plane_state &&
@@ -2015,6 +2021,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context,
 				maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb;
 				dcfclk_from_fw_based_mclk_switching = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb];
 				pstate_en = true;
+				context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] = dm_dram_clock_change_vblank;
 			} else {
 				/* Restore FCLK latency and re-run validation to go back to original validation
 				 * output if we find that enabling FPO does not give us any benefit (i.e. lower
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index 2eb597a24425e..b4c1cc6dc857f 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -426,6 +426,8 @@ struct pipe_ctx {
 	struct dwbc *dwbc;
 	struct mcif_wb *mcif_wb;
 	union pipe_update_flags update_flags;
+	struct tg_color visual_confirm_color;
+	bool has_vactive_margin;
 };
 
 /* Data used for dynamic link encoder assignment.
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index 88ac723d10aa7..df160c6a630c0 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -257,7 +257,6 @@ struct hw_sequencer_funcs {
 
 	void (*update_visual_confirm_color)(struct dc *dc,
 			struct pipe_ctx *pipe_ctx,
-			struct tg_color *color,
 			int mpcc_id);
 
 	void (*update_phantom_vp_position)(struct dc *dc,
@@ -294,6 +293,7 @@ void get_surface_visual_confirm_color(
 
 void get_subvp_visual_confirm_color(
 	struct dc *dc,
+	struct dc_state *context,
 	struct pipe_ctx *pipe_ctx,
 	struct tg_color *color);
 
@@ -306,4 +306,11 @@ void get_mpctree_visual_confirm_color(
 void get_surface_tile_visual_confirm_color(
 		struct pipe_ctx *pipe_ctx,
 		struct tg_color *color);
+
+void get_mclk_switch_visual_confirm_color(
+		struct dc *dc,
+		struct dc_state *context,
+		struct pipe_ctx *pipe_ctx,
+		struct tg_color *color);
+
 #endif /* __DC_HW_SEQUENCER_H__ */
-- 
GitLab


From fe9fa3859b66caf4a6923598c8e343b8a32ec5d1 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Fri, 5 May 2023 11:06:26 -0400
Subject: [PATCH 1008/1791] drm/amd/display: Make unbounded req update separate
 from dlg/ttu

[Description]
- Updates to unbounded requesting should not be conditional
  on updates to dlg / ttu, as this could prevent unbounded
  requesting from being updated if dlg / ttu does not change

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 11 ++++++++---
 drivers/gpu/drm/amd/display/dc/inc/core_types.h    |  1 +
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index b3e187b1347d8..e74c3ce561ab6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -1361,6 +1361,7 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx
 		new_pipe->update_flags.bits.dppclk = 1;
 		new_pipe->update_flags.bits.hubp_interdependent = 1;
 		new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1;
+		new_pipe->update_flags.bits.unbounded_req = 1;
 		new_pipe->update_flags.bits.gamut_remap = 1;
 		new_pipe->update_flags.bits.scaler = 1;
 		new_pipe->update_flags.bits.viewport = 1;
@@ -1504,6 +1505,9 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx
 				memcmp(&old_pipe->rq_regs, &new_pipe->rq_regs, sizeof(old_pipe->rq_regs)))
 			new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1;
 	}
+
+	if (old_pipe->unbounded_req != new_pipe->unbounded_req)
+		new_pipe->update_flags.bits.unbounded_req = 1;
 }
 
 static void dcn20_update_dchubp_dpp(
@@ -1537,10 +1541,11 @@ static void dcn20_update_dchubp_dpp(
 			&pipe_ctx->ttu_regs,
 			&pipe_ctx->rq_regs,
 			&pipe_ctx->pipe_dlg_param);
-
-		if (hubp->funcs->set_unbounded_requesting)
-			hubp->funcs->set_unbounded_requesting(hubp, pipe_ctx->unbounded_req);
 	}
+
+	if (pipe_ctx->update_flags.bits.unbounded_req && hubp->funcs->set_unbounded_requesting)
+		hubp->funcs->set_unbounded_requesting(hubp, pipe_ctx->unbounded_req);
+
 	if (pipe_ctx->update_flags.bits.hubp_interdependent)
 		hubp->funcs->hubp_setup_interdependent(
 			hubp,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index b4c1cc6dc857f..d8dd143cf6ea8 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -374,6 +374,7 @@ union pipe_update_flags {
 		uint32_t viewport : 1;
 		uint32_t plane_changed : 1;
 		uint32_t det_size : 1;
+		uint32_t unbounded_req : 1;
 	} bits;
 	uint32_t raw;
 };
-- 
GitLab


From 50a32b8cf4d7624eb8606b4cb7bc6dee37703da6 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Date: Mon, 1 May 2023 16:30:54 -0600
Subject: [PATCH 1009/1791] drm/amd/display: Remove unnecessary variable

There is no need to use dc_version in the dc_construct_ctx since this
value is copied to dc_ctx->dce_version later. This commit removes the
extra steps.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 2c639cd346d78..e664a77d05eb8 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -857,7 +857,6 @@ static bool dc_construct_ctx(struct dc *dc,
 		const struct dc_init_data *init_params)
 {
 	struct dc_context *dc_ctx;
-	enum dce_version dc_version = DCE_VERSION_UNKNOWN;
 
 	dc_ctx = kzalloc(sizeof(*dc_ctx), GFP_KERNEL);
 	if (!dc_ctx)
@@ -875,8 +874,7 @@ static bool dc_construct_ctx(struct dc *dc,
 
 	/* Create logger */
 
-	dc_version = resource_parse_asic_id(init_params->asic_id);
-	dc_ctx->dce_version = dc_version;
+	dc_ctx->dce_version = resource_parse_asic_id(init_params->asic_id);
 
 	dc_ctx->perf_trace = dc_perf_trace_create();
 	if (!dc_ctx->perf_trace) {
-- 
GitLab


From 3a31e8b89b7240d9a17ace8a1ed050bdcb560f9e Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Tue, 25 Apr 2023 14:02:02 -0400
Subject: [PATCH 1010/1791] drm/amd/display: Remove v_startup workaround for
 dcn3+

[Why]
Calls to dcn20_adjust_freesync_v_startup are no longer
needed as of dcn3+ and can cause underflow in some cases

[How]
Move calls to dcn20_adjust_freesync_v_startup up into
validate_bandwidth for dcn2.x

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dml/dcn20/dcn20_fpu.c  | 24 +++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index 3407f9a2c6a10..8ae5ddbd1b271 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -1099,10 +1099,6 @@ void dcn20_calculate_dlg_params(struct dc *dc,
 		context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz =
 						pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000;
 		context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest;
-		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
-			dcn20_adjust_freesync_v_startup(
-				&context->res_ctx.pipe_ctx[i].stream->timing,
-				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
 
 		pipe_idx++;
 	}
@@ -1931,6 +1927,7 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co
 	int vlevel = 0;
 	int pipe_split_from[MAX_PIPES];
 	int pipe_cnt = 0;
+	int i = 0;
 	display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC);
 	DC_LOGGER_INIT(dc->ctx->logger);
 
@@ -1954,6 +1951,15 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co
 	dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate);
 	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
 
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
+			dcn20_adjust_freesync_v_startup(
+				&context->res_ctx.pipe_ctx[i].stream->timing,
+				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
+	}
+
 	BW_VAL_TRACE_END_WATERMARKS();
 
 	goto validate_out;
@@ -2226,6 +2232,7 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc,
 	int vlevel = 0;
 	int pipe_split_from[MAX_PIPES];
 	int pipe_cnt = 0;
+	int i = 0;
 	display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC);
 	DC_LOGGER_INIT(dc->ctx->logger);
 
@@ -2254,6 +2261,15 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc,
 	dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate);
 	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
 
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
+			dcn20_adjust_freesync_v_startup(
+				&context->res_ctx.pipe_ctx[i].stream->timing,
+				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
+	}
+
 	BW_VAL_TRACE_END_WATERMARKS();
 
 	goto validate_out;
-- 
GitLab


From 629b8ede8b93428b8d124d343b5fbb57ab64d5a8 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Mon, 8 May 2023 00:32:41 -0400
Subject: [PATCH 1011/1791] drm/amd/display: 3.2.236

Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 8be2e6d6d8884..2dff1a5cf3b10 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.235"
+#define DC_VER "3.2.236"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From f4caf5842652f08e024741ef6d423cb0c101d863 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Wed, 14 Sep 2022 16:35:50 +0800
Subject: [PATCH 1012/1791] drm/amdgpu: introduce vmhub definition for
 multi-partition cases (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v1: Each partition has its own gfxhub or mmhub. adjust
the num of MAX_VMHUBS and the GFXHUB/MMHUB layout (Le)

v2: re-design the AMDGPU_GFXHUB/AMDGPU_MMHUB layout (Le)

v3: apply the gfxhub/mmhub layout to new IPs (Hawking)

v4: fix up gmc11 (Alex)

v5: rebase (Alex)

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c    |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c    |  4 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c     |  4 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h     | 13 ++--
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c     |  6 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c     |  8 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c      |  8 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c    |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c    |  4 +-
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c   | 10 +--
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c   | 10 +--
 drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c   | 10 +--
 drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c   | 12 ++--
 drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c   | 10 +--
 drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c | 10 +--
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c     | 33 +++++----
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c     | 26 ++++----
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c      | 78 +++++++++++-----------
 drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c     |  4 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/mes_v10_1.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c    | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c  | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c  | 10 +--
 drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c    | 12 ++--
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c     | 10 +--
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c   |  4 +-
 drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c      |  4 +-
 drivers/gpu/drm/amd/amdgpu/vce_v4_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c      |  4 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c      |  4 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c      |  8 +--
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c      |  4 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c      |  2 +-
 47 files changed, 204 insertions(+), 204 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index a46285841d175..f0a136d352798 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -736,7 +736,7 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev,
 		for (i = 0; i < adev->num_vmhubs; i++)
 			amdgpu_gmc_flush_gpu_tlb(adev, vmid, i, 0);
 	} else {
-		amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB_0, 0);
+		amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB(0), 0);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 87e1a1a9f2988..488b3bb6dcb13 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -315,7 +315,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->use_doorbell = true;
 	ring->doorbell_index = adev->doorbell_index.kiq;
 	ring->xcc_id = xcc_id;
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	if (xcc_id >= 1)
 		ring->doorbell_index = adev->doorbell_index.xcc1_kiq_start +
 					xcc_id - 1;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 4e2531758866c..0a4e5fcfec6bb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -670,7 +670,7 @@ void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type,
 	for (i = 0; i < 16; i++) {
 		reg = hub->vm_context0_cntl + hub->ctx_distance * i;
 
-		tmp = (hub_type == AMDGPU_GFXHUB_0) ?
+		tmp = (hub_type == AMDGPU_GFXHUB(0)) ?
 			RREG32_SOC15_IP(GC, reg) :
 			RREG32_SOC15_IP(MMHUB, reg);
 
@@ -679,7 +679,7 @@ void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type,
 		else
 			tmp &= ~hub->vm_cntx_cntl_vm_fault;
 
-		(hub_type == AMDGPU_GFXHUB_0) ?
+		(hub_type == AMDGPU_GFXHUB(0)) ?
 			WREG32_SOC15_IP(GC, reg, tmp) :
 			WREG32_SOC15_IP(MMHUB, reg, tmp);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index b6bd667df6763..c3964c14f215b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2374,12 +2374,12 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	case AMDGPU_VM_OP_RESERVE_VMID:
 		/* We only have requirement to reserve vmid from gfxhub */
 		r = amdgpu_vmid_alloc_reserved(adev, &fpriv->vm,
-					       AMDGPU_GFXHUB_0);
+					       AMDGPU_GFXHUB(0));
 		if (r)
 			return r;
 		break;
 	case AMDGPU_VM_OP_UNRESERVE_VMID:
-		amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB_0);
+		amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB(0));
 		break;
 	default:
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 6f085f0b4ef3f..9f5d32b0fda1f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -111,11 +111,14 @@ struct amdgpu_mem_stats;
 /* Reserve 4MB VRAM for page tables */
 #define AMDGPU_VM_RESERVED_VRAM		(8ULL << 20)
 
-/* max number of VMHUB */
-#define AMDGPU_MAX_VMHUBS			3
-#define AMDGPU_GFXHUB_0				0
-#define AMDGPU_MMHUB_0				1
-#define AMDGPU_MMHUB_1				2
+/*
+ * max number of VMHUB
+ * layout: max 8 GFXHUB + 4 MMHUB0 + 1 MMHUB1
+ */
+#define AMDGPU_MAX_VMHUBS			13
+#define AMDGPU_GFXHUB(x)			(x)
+#define AMDGPU_MMHUB0(x)			(8 + x)
+#define AMDGPU_MMHUB1(x)			(8 + 4 + x)
 
 /* Reserve 2MB at top/bottom of address space for kernel use */
 #define AMDGPU_VA_RESERVED_SIZE			(2ULL << 20)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 8e86b2c23c0a5..7b585141e10e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4461,7 +4461,7 @@ static int gfx_v10_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id,
 		ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
 	else
 		ring->doorbell_index = adev->doorbell_index.gfx_ring1 << 1;
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "gfx_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + ring->pipe;
@@ -4490,7 +4490,7 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX10_MEC_HPD_SIZE);
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
@@ -4978,7 +4978,7 @@ static void gfx_v10_0_constants_init(struct amdgpu_device *adev)
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
-	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
 		nv_grbm_select(adev, 0, 0, 0, i);
 		/* CP and shaders */
 		WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index f77779c31043b..790df2cc3480f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -906,7 +906,7 @@ static int gfx_v11_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id,
 		ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
 	else
 		ring->doorbell_index = adev->doorbell_index.gfx_ring1 << 1;
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "gfx_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + ring->pipe;
@@ -937,7 +937,7 @@ static int gfx_v11_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX11_MEC_HPD_SIZE);
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
@@ -1707,7 +1707,7 @@ static void gfx_v11_0_constants_init(struct amdgpu_device *adev)
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
-	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
 		soc21_grbm_select(adev, 0, 0, 0, i);
 		/* CP and shaders */
 		WREG32_SOC15(GC, 0, regSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG);
@@ -4190,7 +4190,7 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev)
 		false : true;
 
 	adev->gfxhub.funcs->set_fault_enable_default(adev, value);
-	amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0);
+	amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 46577b59cb042..91814dc083c9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2005,7 +2005,7 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue);
 
 	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
@@ -2105,7 +2105,7 @@ static int gfx_v9_0_sw_init(void *handle)
 
 		/* disable scheduler on the real ring */
 		ring->no_scheduler = true;
-		ring->vm_hub = AMDGPU_GFXHUB_0;
+		ring->vm_hub = AMDGPU_GFXHUB(0);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq,
 				     AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP,
 				     AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -2123,7 +2123,7 @@ static int gfx_v9_0_sw_init(void *handle)
 			ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
 			ring->is_sw_ring = true;
 			hw_prio = amdgpu_sw_ring_priority(i);
-			ring->vm_hub = AMDGPU_GFXHUB_0;
+			ring->vm_hub = AMDGPU_GFXHUB(0);
 			r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq,
 					     AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, hw_prio,
 					     NULL);
@@ -2393,7 +2393,7 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev)
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
-	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
 		soc15_grbm_select(adev, 0, 0, 0, i, 0);
 		/* CP and shaders */
 		if (i == 0) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index d648a29c33e0a..ec7c049c59521 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -1935,7 +1935,7 @@ static bool gfx_v9_4_2_query_uctl2_poison_status(struct amdgpu_device *adev)
 	u32 status = 0;
 	struct amdgpu_vmhub *hub;
 
-	hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	status = RREG32(hub->vm_l2_pro_fault_status);
 	/* reset page fault status */
 	WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 9d17dcfae1306..f5104b9826337 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -757,7 +757,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 				(adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
-	ring->vm_hub = AMDGPU_GFXHUB_0;
+	ring->vm_hub = AMDGPU_GFXHUB(0);
 	sprintf(ring->name, "comp_%d.%d.%d.%d",
 			ring->xcc_id, ring->me, ring->pipe, ring->queue);
 
@@ -996,7 +996,7 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
-	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) {
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
 		for (j = 0; j < adev->gfx.num_xcd; j++) {
 			soc15_grbm_select(adev, 0, 0, 0, i, j);
 			/* CP and shaders */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
index ab2325f6c7ac5..d94cc1ec7242d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
@@ -40,7 +40,7 @@ static void gfxhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev,
 					 uint32_t vmid,
 					 uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -247,7 +247,7 @@ static void gfxhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i;
@@ -307,7 +307,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v1_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i;
 
 	for (i = 0 ; i < 18; ++i) {
@@ -338,7 +338,7 @@ static int gfxhub_v1_0_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v1_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i;
 
@@ -411,7 +411,7 @@ static void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev,
 
 static void gfxhub_v1_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index 79af32bb078c8..9c385ce3a8c40 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -42,7 +42,7 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 					 uint32_t vmid,
 					 uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 
 	for (i = 0; i < adev->gfx.num_xcd; i++) {
@@ -291,7 +291,7 @@ static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i, j;
@@ -357,7 +357,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i, j;
 
 	for (j = 0; j < adev->gfx.num_xcd; j++) {
@@ -406,7 +406,7 @@ static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i, j;
 
@@ -483,7 +483,7 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 
 static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
index 9b3a025273181..f173a61c6c15a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c
@@ -120,7 +120,7 @@ static u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev)
 static void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -282,7 +282,7 @@ static void gfxhub_v2_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 	uint32_t tmp;
 
@@ -331,7 +331,7 @@ static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i;
 
 	for (i = 0 ; i < 18; ++i) {
@@ -360,7 +360,7 @@ static int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i;
 
@@ -433,7 +433,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v2_0_vmhub_funcs = {
 
 static void gfxhub_v2_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
index 4aacbbec31e28..d8fc3e8088cd0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
@@ -123,7 +123,7 @@ static u64 gfxhub_v2_1_get_mc_fb_offset(struct amdgpu_device *adev)
 static void gfxhub_v2_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -291,7 +291,7 @@ static void gfxhub_v2_1_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 	uint32_t tmp;
 
@@ -340,7 +340,7 @@ static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v2_1_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i;
 
 	for (i = 0 ; i < 18; ++i) {
@@ -381,7 +381,7 @@ static int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i;
 
@@ -462,7 +462,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v2_1_vmhub_funcs = {
 
 static void gfxhub_v2_1_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
@@ -651,7 +651,7 @@ static void gfxhub_v2_1_restore_regs(struct amdgpu_device *adev)
 
 static void gfxhub_v2_1_halt(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 	uint32_t tmp;
 	int time = 1000;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c
index 13712640fa46f..c53147f9c9fc0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c
@@ -119,7 +119,7 @@ static u64 gfxhub_v3_0_get_mc_fb_offset(struct amdgpu_device *adev)
 static void gfxhub_v3_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	WREG32_SOC15_OFFSET(GC, 0, regGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -290,7 +290,7 @@ static void gfxhub_v3_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 	uint32_t tmp;
 
@@ -339,7 +339,7 @@ static void gfxhub_v3_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i;
 
 	for (i = 0 ; i < 18; ++i) {
@@ -380,7 +380,7 @@ static int gfxhub_v3_0_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i;
 
@@ -463,7 +463,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v3_0_vmhub_funcs = {
 
 static void gfxhub_v3_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c
index 6e0bd628c8895..ae777487d72ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c
@@ -122,7 +122,7 @@ static u64 gfxhub_v3_0_3_get_mc_fb_offset(struct amdgpu_device *adev)
 static void gfxhub_v3_0_3_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	WREG32_SOC15_OFFSET(GC, 0, regGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -295,7 +295,7 @@ static void gfxhub_v3_0_3_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_3_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	int i;
 	uint32_t tmp;
 
@@ -344,7 +344,7 @@ static void gfxhub_v3_0_3_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_3_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	unsigned i;
 
 	for (i = 0 ; i < 18; ++i) {
@@ -373,7 +373,7 @@ static int gfxhub_v3_0_3_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v3_0_3_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	u32 tmp;
 	u32 i;
 
@@ -451,7 +451,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v3_0_3_vmhub_funcs = {
 
 static void gfxhub_v3_0_3_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(GC, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 5697b66bf0de1..ea2a448147e37 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -76,7 +76,7 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 		/* MM HUB */
-		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, false);
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), false);
 		/* GFX HUB */
 		/* This works because this interrupt is only
 		 * enabled at init/resume and disabled in
@@ -84,11 +84,11 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 		 * change over the course of suspend/resume.
 		 */
 		if (!adev->in_s0ix)
-			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, false);
+			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), false);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
 		/* MM HUB */
-		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, true);
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), true);
 		/* GFX HUB */
 		/* This works because this interrupt is only
 		 * enabled at init/resume and disabled in
@@ -96,7 +96,7 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 		 * change over the course of suspend/resume.
 		 */
 		if (!adev->in_s0ix)
-			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, true);
+			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), true);
 		break;
 	default:
 		break;
@@ -149,7 +149,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,
 		 * be updated to avoid reading an incorrect value due to
 		 * the new fast GRBM interface.
 		 */
-		if ((entry->vmid_src == AMDGPU_GFXHUB_0) &&
+		if ((entry->vmid_src == AMDGPU_GFXHUB(0)) &&
 		    (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 3, 0)))
 			RREG32(hub->vm_l2_pro_fault_status);
 
@@ -212,8 +212,7 @@ static void gmc_v10_0_set_irq_funcs(struct amdgpu_device *adev)
 static bool gmc_v10_0_use_invalidate_semaphore(struct amdgpu_device *adev,
 				       uint32_t vmhub)
 {
-	return ((vmhub == AMDGPU_MMHUB_0 ||
-		 vmhub == AMDGPU_MMHUB_1) &&
+	return ((vmhub == AMDGPU_MMHUB0(0)) &&
 		(!amdgpu_sriov_vf(adev)));
 }
 
@@ -249,7 +248,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 	unsigned int i;
 	unsigned char hub_ip = 0;
 
-	hub_ip = (vmhub == AMDGPU_GFXHUB_0) ?
+	hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ?
 		   GC_HWIP : MMHUB_HWIP;
 
 	spin_lock(&adev->gmc.invalidate_lock);
@@ -284,7 +283,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 	 * Issue a dummy read to wait for the ACK register to be cleared
 	 * to avoid a false ACK due to the new fast GRBM interface.
 	 */
-	if ((vmhub == AMDGPU_GFXHUB_0) &&
+	if ((vmhub == AMDGPU_GFXHUB(0)) &&
 	    (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 3, 0)))
 		RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_req +
 				  hub->eng_distance * eng, hub_ip);
@@ -361,19 +360,19 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 
 	mutex_lock(&adev->mman.gtt_window_lock);
 
-	if (vmhub == AMDGPU_MMHUB_0) {
-		gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB_0, 0);
+	if (vmhub == AMDGPU_MMHUB0(0)) {
+		gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB0(0), 0);
 		mutex_unlock(&adev->mman.gtt_window_lock);
 		return;
 	}
 
-	BUG_ON(vmhub != AMDGPU_GFXHUB_0);
+	BUG_ON(vmhub != AMDGPU_GFXHUB(0));
 
 	if (!adev->mman.buffer_funcs_enabled ||
 	    !adev->ib_pool_ready ||
 	    amdgpu_in_reset(adev) ||
 	    ring->sched.ready == false) {
-		gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB_0, 0);
+		gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB(0), 0);
 		mutex_unlock(&adev->mman.gtt_window_lock);
 		return;
 	}
@@ -466,7 +465,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 							i, flush_type);
 			} else {
 				gmc_v10_0_flush_gpu_tlb(adev, vmid,
-						AMDGPU_GFXHUB_0, flush_type);
+						AMDGPU_GFXHUB(0), flush_type);
 			}
 			if (!adev->enable_mes)
 				break;
@@ -534,7 +533,7 @@ static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid
 	if (ring->is_mes_queue)
 		return;
 
-	if (ring->vm_hub == AMDGPU_GFXHUB_0)
+	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid;
 	else
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid;
@@ -1075,9 +1074,9 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
 	if (!adev->in_s0ix)
 		adev->gfxhub.funcs->set_fault_enable_default(adev, value);
 	adev->mmhub.funcs->set_fault_enable_default(adev, value);
-	gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0);
+	gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0);
 	if (!adev->in_s0ix)
-		gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0);
+		gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0);
 
 	DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
 		 (unsigned)(adev->gmc.gart_size >> 20),
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 2f570fb5febe3..fb2ac31cbba72 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -64,7 +64,7 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 		/* MM HUB */
-		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, false);
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), false);
 		/* GFX HUB */
 		/* This works because this interrupt is only
 		 * enabled at init/resume and disabled in
@@ -72,11 +72,11 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 		 * change over the course of suspend/resume.
 		 */
 		if (!adev->in_s0ix)
-			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, false);
+			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), false);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
 		/* MM HUB */
-		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, true);
+		amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), true);
 		/* GFX HUB */
 		/* This works because this interrupt is only
 		 * enabled at init/resume and disabled in
@@ -84,7 +84,7 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 		 * change over the course of suspend/resume.
 		 */
 		if (!adev->in_s0ix)
-			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, true);
+			amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), true);
 		break;
 	default:
 		break;
@@ -110,7 +110,7 @@ static int gmc_v11_0_process_interrupt(struct amdgpu_device *adev,
 		 * be updated to avoid reading an incorrect value due to
 		 * the new fast GRBM interface.
 		 */
-		if (entry->vmid_src == AMDGPU_GFXHUB_0)
+		if (entry->vmid_src == AMDGPU_GFXHUB(0))
 			RREG32(hub->vm_l2_pro_fault_status);
 
 		status = RREG32(hub->vm_l2_pro_fault_status);
@@ -170,7 +170,7 @@ static void gmc_v11_0_set_irq_funcs(struct amdgpu_device *adev)
 static bool gmc_v11_0_use_invalidate_semaphore(struct amdgpu_device *adev,
 				       uint32_t vmhub)
 {
-	return ((vmhub == AMDGPU_MMHUB_0) &&
+	return ((vmhub == AMDGPU_MMHUB0(0)) &&
 		(!amdgpu_sriov_vf(adev)));
 }
 
@@ -202,7 +202,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 	unsigned int i;
 	unsigned char hub_ip = 0;
 
-	hub_ip = (vmhub == AMDGPU_GFXHUB_0) ?
+	hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ?
 		   GC_HWIP : MMHUB_HWIP;
 
 	spin_lock(&adev->gmc.invalidate_lock);
@@ -251,7 +251,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 			      hub->eng_distance * eng, 0, hub_ip);
 
 	/* Issue additional private vm invalidation to MMHUB */
-	if ((vmhub != AMDGPU_GFXHUB_0) &&
+	if ((vmhub != AMDGPU_GFXHUB(0)) &&
 	    (hub->vm_l2_bank_select_reserved_cid2) &&
 		!amdgpu_sriov_vf(adev)) {
 		inv_req = RREG32_NO_KIQ(hub->vm_l2_bank_select_reserved_cid2);
@@ -284,7 +284,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
 static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 					uint32_t vmhub, uint32_t flush_type)
 {
-	if ((vmhub == AMDGPU_GFXHUB_0) && !adev->gfx.is_poweron)
+	if ((vmhub == AMDGPU_GFXHUB(0)) && !adev->gfx.is_poweron)
 		return;
 
 	/* flush hdp cache */
@@ -369,7 +369,7 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 							i, flush_type);
 			} else {
 				gmc_v11_0_flush_gpu_tlb(adev, vmid,
-						AMDGPU_GFXHUB_0, flush_type);
+						AMDGPU_GFXHUB(0), flush_type);
 			}
 		}
 	}
@@ -435,7 +435,7 @@ static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid
 	if (ring->is_mes_queue)
 		return;
 
-	if (ring->vm_hub == AMDGPU_GFXHUB_0)
+	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid;
 	else
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT_MM) + vmid;
@@ -886,7 +886,7 @@ static int gmc_v11_0_sw_fini(void *handle)
 static void gmc_v11_0_init_golden_registers(struct amdgpu_device *adev)
 {
 	if (amdgpu_sriov_vf(adev)) {
-		struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+		struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 		WREG32(hub->vm_contexts_disable, 0);
 		return;
@@ -921,7 +921,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev)
 		false : true;
 
 	adev->mmhub.funcs->set_fault_enable_default(adev, value);
-	gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0);
+	gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0);
 
 	DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
 		 (unsigned)(adev->gmc.gart_size >> 20),
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 6ae5cee9b64be..193ba4d912a69 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -491,20 +491,20 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 				 * fini/suspend, so the overall state doesn't
 				 * change over the course of suspend/resume.
 				 */
-				if (adev->in_s0ix && (j == AMDGPU_GFXHUB_0))
+				if (adev->in_s0ix && (j == AMDGPU_GFXHUB(0)))
 					continue;
 
-				if (j == AMDGPU_GFXHUB_0)
-					tmp = RREG32_SOC15_IP(GC, reg);
-				else
+				if (j >= AMDGPU_MMHUB0(0))
 					tmp = RREG32_SOC15_IP(MMHUB, reg);
+				else
+					tmp = RREG32_SOC15_IP(GC, reg);
 
 				tmp &= ~bits;
 
-				if (j == AMDGPU_GFXHUB_0)
-					WREG32_SOC15_IP(GC, reg, tmp);
-				else
+				if (j >= AMDGPU_MMHUB0(0))
 					WREG32_SOC15_IP(MMHUB, reg, tmp);
+				else
+					WREG32_SOC15_IP(GC, reg, tmp);
 			}
 		}
 		break;
@@ -519,20 +519,20 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 				 * fini/suspend, so the overall state doesn't
 				 * change over the course of suspend/resume.
 				 */
-				if (adev->in_s0ix && (j == AMDGPU_GFXHUB_0))
+				if (adev->in_s0ix && (j == AMDGPU_GFXHUB(0)))
 					continue;
 
-				if (j == AMDGPU_GFXHUB_0)
-					tmp = RREG32_SOC15_IP(GC, reg);
-				else
+				if (j >= AMDGPU_MMHUB0(0))
 					tmp = RREG32_SOC15_IP(MMHUB, reg);
+				else
+					tmp = RREG32_SOC15_IP(GC, reg);
 
 				tmp |= bits;
 
-				if (j == AMDGPU_GFXHUB_0)
-					WREG32_SOC15_IP(GC, reg, tmp);
-				else
+				if (j >= AMDGPU_MMHUB0(0))
 					WREG32_SOC15_IP(MMHUB, reg, tmp);
+				else
+					WREG32_SOC15_IP(GC, reg, tmp);
 			}
 		}
 		break;
@@ -605,13 +605,13 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 
 	if (entry->client_id == SOC15_IH_CLIENTID_VMC) {
 		hub_name = "mmhub0";
-		hub = &adev->vmhub[AMDGPU_MMHUB_0];
+		hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	} else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) {
 		hub_name = "mmhub1";
-		hub = &adev->vmhub[AMDGPU_MMHUB_1];
+		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
 	} else {
 		hub_name = "gfxhub0";
-		hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+		hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	}
 
 	memset(&task_info, 0, sizeof(struct amdgpu_task_info));
@@ -636,7 +636,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	 * be updated to avoid reading an incorrect value due to
 	 * the new fast GRBM interface.
 	 */
-	if ((entry->vmid_src == AMDGPU_GFXHUB_0) &&
+	if ((entry->vmid_src == AMDGPU_GFXHUB(0)) &&
 	    (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 4, 2)))
 		RREG32(hub->vm_l2_pro_fault_status);
 
@@ -649,7 +649,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	dev_err(adev->dev,
 		"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
 		status);
-	if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) {
+	if (hub == &adev->vmhub[AMDGPU_GFXHUB(0)]) {
 		dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
 			cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" :
 			gfxhub_client_ids[cid],
@@ -759,8 +759,8 @@ static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev,
 	    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
 		return false;
 
-	return ((vmhub == AMDGPU_MMHUB_0 ||
-		 vmhub == AMDGPU_MMHUB_1) &&
+	return ((vmhub == AMDGPU_MMHUB0(0) ||
+		 vmhub == AMDGPU_MMHUB1(0)) &&
 		(!amdgpu_sriov_vf(adev)) &&
 		(!(!(adev->apu_flags & AMD_APU_IS_RAVEN2) &&
 		   (adev->apu_flags & AMD_APU_IS_PICASSO))));
@@ -849,11 +849,10 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	if (use_semaphore) {
 		for (j = 0; j < adev->usec_timeout; j++) {
 			/* a read return value of 1 means semaphore acquire */
-			if (vmhub == AMDGPU_GFXHUB_0)
-				tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng);
-			else
+			if (vmhub >= AMDGPU_MMHUB0(0))
 				tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng);
-
+			else
+				tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng);
 			if (tmp & 0x1)
 				break;
 			udelay(1);
@@ -864,27 +863,26 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	}
 
 	do {
-		if (vmhub == AMDGPU_GFXHUB_0)
-			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
-		else
+		if (vmhub >= AMDGPU_MMHUB0(0))
 			WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
+		else
+			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
 
 		/*
 		 * Issue a dummy read to wait for the ACK register to
 		 * be cleared to avoid a false ACK due to the new fast
 		 * GRBM interface.
 		 */
-		if ((vmhub == AMDGPU_GFXHUB_0) &&
+		if ((vmhub == AMDGPU_GFXHUB(0)) &&
 		    (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 4, 2)))
 			RREG32_NO_KIQ(hub->vm_inv_eng0_req +
 				      hub->eng_distance * eng);
 
 		for (j = 0; j < adev->usec_timeout; j++) {
-			if (vmhub == AMDGPU_GFXHUB_0)
-				tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_ack + hub->eng_distance * eng);
-			else
+			if (vmhub >= AMDGPU_MMHUB0(0))
 				tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_ack + hub->eng_distance * eng);
-
+			else
+				tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_ack + hub->eng_distance * eng);
 			if (tmp & (1 << vmid))
 				break;
 			udelay(1);
@@ -900,10 +898,10 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		 * add semaphore release after invalidation,
 		 * write with 0 means semaphore release
 		 */
-		if (vmhub == AMDGPU_GFXHUB_0)
-			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0);
+		if (vmhub >= AMDGPU_MMHUB0(0))
+			WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
 		else
-			WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0);
+			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
 	}
 
 	spin_unlock(&adev->gmc.invalidate_lock);
@@ -994,7 +992,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 							i, flush_type);
 			} else {
 				gmc_v9_0_flush_gpu_tlb(adev, vmid,
-						AMDGPU_GFXHUB_0, flush_type);
+						AMDGPU_GFXHUB(0), flush_type);
 			}
 			break;
 		}
@@ -1060,10 +1058,10 @@ static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
 	uint32_t reg;
 
 	/* Do nothing because there's no lut register for mmhub1. */
-	if (ring->vm_hub == AMDGPU_MMHUB_1)
+	if (ring->vm_hub == AMDGPU_MMHUB1(0))
 		return;
 
-	if (ring->vm_hub == AMDGPU_GFXHUB_0)
+	if (ring->vm_hub == AMDGPU_GFXHUB(0))
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid;
 	else
 		reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid;
@@ -1947,7 +1945,7 @@ static int gmc_v9_0_hw_init(void *handle)
 		adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	}
 	for (i = 0; i < adev->num_vmhubs; ++i) {
-		if (adev->in_s0ix && (i == AMDGPU_GFXHUB_0))
+		if (adev->in_s0ix && (i == AMDGPU_GFXHUB(0)))
 			continue;
 		gmc_v9_0_flush_gpu_tlb(adev, 0, i, 0);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index a3076eb8af6ac..71fe7f6f9889c 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -485,7 +485,7 @@ int jpeg_v1_0_sw_init(void *handle)
 		return r;
 
 	ring = &adev->jpeg.inst->ring_dec;
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq,
 			     0, AMDGPU_RING_PRIO_DEFAULT, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
index 0eddf7c824a72..3a43e42f4834d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
@@ -86,7 +86,7 @@ static int jpeg_v2_0_sw_init(void *handle)
 	ring = &adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1;
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq,
 			     0, AMDGPU_RING_PRIO_DEFAULT, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index b040f51d9aa9d..259b7ba6a842c 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -128,9 +128,9 @@ static int jpeg_v2_5_sw_init(void *handle)
 		ring = &adev->jpeg.inst[i].ring_dec;
 		ring->use_doorbell = true;
 		if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0))
-			ring->vm_hub = AMDGPU_MMHUB_1;
+			ring->vm_hub = AMDGPU_MMHUB1(0);
 		else
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + 8 * i;
 		sprintf(ring->name, "jpeg_dec_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst[i].irq,
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index 1c2292cc5f2ce..c55386c22311a 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -101,7 +101,7 @@ static int jpeg_v3_0_sw_init(void *handle)
 	ring = &adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1;
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 			     AMDGPU_RING_PRIO_DEFAULT, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index 77e1e64aa1d1c..d7d5ffc293937 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -108,7 +108,7 @@ static int jpeg_v4_0_sw_init(void *handle)
 	ring = &adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = amdgpu_sriov_vf(adev) ? (((adev->doorbell_index.vcn.vcn_ring0_1) << 1) + 4) : ((adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1);
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index 4560476c7c317..f1a6abdad21b6 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -149,7 +149,7 @@ static int mes_v10_1_add_hw_queue(struct amdgpu_mes *mes,
 {
 	struct amdgpu_device *adev = mes->adev;
 	union MESAPI__ADD_QUEUE mes_add_queue_pkt;
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	uint32_t vm_cntx_cntl = hub->vm_cntx_cntl;
 
 	memset(&mes_add_queue_pkt, 0, sizeof(mes_add_queue_pkt));
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 3adb450eec072..9791f35817865 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -164,7 +164,7 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 {
 	struct amdgpu_device *adev = mes->adev;
 	union MESAPI__ADD_QUEUE mes_add_queue_pkt;
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
 	uint32_t vm_cntx_cntl = hub->vm_cntx_cntl;
 
 	memset(&mes_add_queue_pkt, 0, sizeof(mes_add_queue_pkt));
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index 15e7cbeae75b8..fb91b31056cae 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -54,7 +54,7 @@ static u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev)
 static void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -229,7 +229,7 @@ static void mmhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i;
@@ -285,7 +285,7 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v1_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -338,7 +338,7 @@ static int mmhub_v1_0_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v1_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -415,7 +415,7 @@ static void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool
 
 static void mmhub_v1_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c
index 73afbf2facc9e..9086f2fdfaf42 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c
@@ -54,7 +54,7 @@ static u64 mmhub_v1_7_get_fb_location(struct amdgpu_device *adev)
 static void mmhub_v1_7_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base));
@@ -261,7 +261,7 @@ static void mmhub_v1_7_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v1_7_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i;
@@ -319,7 +319,7 @@ static void mmhub_v1_7_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v1_7_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -348,7 +348,7 @@ static int mmhub_v1_7_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v1_7_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -425,7 +425,7 @@ static void mmhub_v1_7_set_fault_enable_default(struct amdgpu_device *adev, bool
 
 static void mmhub_v1_7_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 342d1702104cb..9ec06f9db7617 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -53,7 +53,7 @@ static u64 mmhub_v1_8_get_fb_location(struct amdgpu_device *adev)
 static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base));
@@ -253,7 +253,7 @@ static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i;
@@ -311,7 +311,7 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -352,7 +352,7 @@ static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -426,7 +426,7 @@ static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool
 
 static void mmhub_v1_8_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
index 278e32db878d7..8f76c6ecf50a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c
@@ -187,7 +187,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
 static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET_RLC(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -362,7 +362,7 @@ static void mmhub_v2_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	int i;
 	uint32_t tmp;
 
@@ -412,7 +412,7 @@ static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v2_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -441,7 +441,7 @@ static int mmhub_v2_0_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v2_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -520,7 +520,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v2_0_vmhub_funcs = {
 
 static void mmhub_v2_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c
index fcf2813e70db8..8bd0fc8d9d25e 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c
@@ -121,7 +121,7 @@ static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev,
 					uint32_t vmid,
 					uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base));
@@ -280,7 +280,7 @@ static void mmhub_v2_3_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v2_3_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	int i;
 	uint32_t tmp;
 
@@ -330,7 +330,7 @@ static void mmhub_v2_3_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v2_3_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -373,7 +373,7 @@ static int mmhub_v2_3_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v2_3_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -446,7 +446,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v2_3_vmhub_funcs = {
 
 static void mmhub_v2_3_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c
index 17a792616979c..441379e91cfa8 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c
@@ -136,7 +136,7 @@ mmhub_v3_0_print_l2_protection_fault_status(struct amdgpu_device *adev,
 static void mmhub_v3_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -319,7 +319,7 @@ static void mmhub_v3_0_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	int i;
 	uint32_t tmp;
 
@@ -369,7 +369,7 @@ static void mmhub_v3_0_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -398,7 +398,7 @@ static int mmhub_v3_0_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -477,7 +477,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_vmhub_funcs = {
 
 static void mmhub_v3_0_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
index 26509b6b8c240..12c7f4b46ea94 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
@@ -138,7 +138,7 @@ static void mmhub_v3_0_1_setup_vm_pt_regs(struct amdgpu_device *adev,
 					  uint32_t vmid,
 					  uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -306,7 +306,7 @@ static void mmhub_v3_0_1_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_1_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	int i;
 	uint32_t tmp;
 
@@ -356,7 +356,7 @@ static void mmhub_v3_0_1_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_1_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -385,7 +385,7 @@ static int mmhub_v3_0_1_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_1_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -459,7 +459,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_1_vmhub_funcs = {
 
 static void mmhub_v3_0_1_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c
index 26abbc6a47ab2..5dadc85abf7ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c
@@ -129,7 +129,7 @@ mmhub_v3_0_2_print_l2_protection_fault_status(struct amdgpu_device *adev,
 static void mmhub_v3_0_2_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 			    hub->ctx_addr_distance * vmid,
@@ -311,7 +311,7 @@ static void mmhub_v3_0_2_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_2_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	int i;
 	uint32_t tmp;
 
@@ -361,7 +361,7 @@ static void mmhub_v3_0_2_setup_vmid_config(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_2_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -390,7 +390,7 @@ static int mmhub_v3_0_2_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v3_0_2_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i;
 
@@ -469,7 +469,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_2_vmhub_funcs = {
 
 static void mmhub_v3_0_2_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	hub->ctx0_ptb_addr_lo32 =
 		SOC15_REG_OFFSET(MMHUB, 0,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
index 72083e96222f0..e790f890aec65 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c
@@ -57,7 +57,7 @@ static u64 mmhub_v9_4_get_fb_location(struct amdgpu_device *adev)
 static void mmhub_v9_4_setup_hubid_vm_pt_regs(struct amdgpu_device *adev, int hubid,
 				uint32_t vmid, uint64_t value)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 
 	WREG32_SOC15_OFFSET(MMHUB, 0,
 			    mmVML2VC0_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
@@ -294,7 +294,7 @@ static void mmhub_v9_4_disable_identity_aperture(struct amdgpu_device *adev,
 
 static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned int num_level, block_size;
 	uint32_t tmp;
 	int i;
@@ -363,7 +363,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid)
 static void mmhub_v9_4_program_invalidation(struct amdgpu_device *adev,
 					    int hubid)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	unsigned i;
 
 	for (i = 0; i < 18; ++i) {
@@ -404,7 +404,7 @@ static int mmhub_v9_4_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v9_4_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0];
+	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
 	u32 tmp;
 	u32 i, j;
 
@@ -507,8 +507,8 @@ static void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool
 
 static void mmhub_v9_4_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub[MMHUB_NUM_INSTANCES] =
-		{&adev->vmhub[AMDGPU_MMHUB_0], &adev->vmhub[AMDGPU_MMHUB_1]};
+	struct amdgpu_vmhub *hub[MMHUB_NUM_INSTANCES] = {
+		&adev->vmhub[AMDGPU_MMHUB0(0)], &adev->vmhub[AMDGPU_MMHUB1(0)]};
 	int i;
 
 	for (i = 0; i < MMHUB_NUM_INSTANCES; i++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 9295ac7edd565..50b6eb9bcfda7 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1825,12 +1825,12 @@ static int sdma_v4_0_sw_init(void *handle)
 
 		/*
 		 * On Arcturus, SDMA instance 5~7 has a different vmhub
-		 * type(AMDGPU_MMHUB_1).
+		 * type(AMDGPU_MMHUB1).
 		 */
 		if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) && i >= 5)
-			ring->vm_hub = AMDGPU_MMHUB_1;
+			ring->vm_hub = AMDGPU_MMHUB1(0);
 		else
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq,
@@ -1851,9 +1851,9 @@ static int sdma_v4_0_sw_init(void *handle)
 			ring->doorbell_index += 0x400;
 
 			if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) && i >= 5)
-				ring->vm_hub = AMDGPU_MMHUB_1;
+				ring->vm_hub = AMDGPU_MMHUB1(0);
 			else
-				ring->vm_hub = AMDGPU_MMHUB_0;
+				ring->vm_hub = AMDGPU_MMHUB0(0);
 
 			sprintf(ring->name, "page%d", i);
 			r = amdgpu_ring_init(adev, ring, 1024,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 64dcaa2670dd1..7efe7c43fffbe 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1309,7 +1309,7 @@ static int sdma_v4_4_2_sw_init(void *handle)
 
 		/* doorbell size is 2 dwords, get DWORD offset */
 		ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq,
@@ -1328,7 +1328,7 @@ static int sdma_v4_4_2_sw_init(void *handle)
 			 */
 			ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
 			ring->doorbell_index += 0x400;
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 
 			sprintf(ring->name, "page%d", i);
 			r = amdgpu_ring_init(adev, ring, 1024,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index 92e1299be0219..a0077cf412955 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -1389,7 +1389,7 @@ static int sdma_v5_0_sw_init(void *handle)
 			(adev->doorbell_index.sdma_engine[0] << 1) //get DWORD offset
 			: (adev->doorbell_index.sdma_engine[1] << 1); // get DWORD offset
 
-		ring->vm_hub = AMDGPU_GFXHUB_0;
+		ring->vm_hub = AMDGPU_GFXHUB(0);
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq,
 				     (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 :
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index ca7e8757d78e5..efa2c84ee78e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -1253,7 +1253,7 @@ static int sdma_v5_2_sw_init(void *handle)
 		ring->doorbell_index =
 			(adev->doorbell_index.sdma_engine[i] << 1); //get DWORD offset
 
-		ring->vm_hub = AMDGPU_GFXHUB_0;
+		ring->vm_hub = AMDGPU_GFXHUB(0);
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq,
 				     AMDGPU_SDMA_IRQ_INSTANCE0 + i,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index 3d9a80511a450..79d09792d2ce3 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -1298,7 +1298,7 @@ static int sdma_v6_0_sw_init(void *handle)
 		ring->doorbell_index =
 			(adev->doorbell_index.sdma_engine[i] << 1); // get DWORD offset
 
-		ring->vm_hub = AMDGPU_GFXHUB_0;
+		ring->vm_hub = AMDGPU_GFXHUB(0);
 		sprintf(ring->name, "sdma%d", i);
 		r = amdgpu_ring_init(adev, ring, 1024,
 				     &adev->sdma.trap_irq,
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index e32b656b3dab8..abaa4463e906c 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -444,7 +444,7 @@ static int uvd_v7_0_sw_init(void *handle)
 			continue;
 		if (!amdgpu_sriov_vf(adev)) {
 			ring = &adev->uvd.inst[j].ring;
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 			sprintf(ring->name, "uvd_%d", ring->me);
 			r = amdgpu_ring_init(adev, ring, 512,
 					     &adev->uvd.inst[j].irq, 0,
@@ -455,7 +455,7 @@ static int uvd_v7_0_sw_init(void *handle)
 
 		for (i = 0; i < adev->uvd.num_enc_rings; ++i) {
 			ring = &adev->uvd.inst[j].ring_enc[i];
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 			sprintf(ring->name, "uvd_enc_%d.%d", ring->me, i);
 			if (amdgpu_sriov_vf(adev)) {
 				ring->use_doorbell = true;
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index 57b85bb6a1e49..e0b70cd3b697c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -466,7 +466,7 @@ static int vce_v4_0_sw_init(void *handle)
 		enum amdgpu_ring_priority_level hw_prio = amdgpu_vce_get_ring_prio(i);
 
 		ring = &adev->vce.ring[i];
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 		sprintf(ring->name, "vce%d", i);
 		if (amdgpu_sriov_vf(adev)) {
 			/* DOORBELL only works under SRIOV */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 761c28fa6ec14..f877c39c7cdde 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -120,7 +120,7 @@ static int vcn_v1_0_sw_init(void *handle)
 		return r;
 
 	ring = &adev->vcn.inst->ring_dec;
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "vcn_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
 			     AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -142,7 +142,7 @@ static int vcn_v1_0_sw_init(void *handle)
 		enum amdgpu_ring_priority_level hw_prio = amdgpu_vcn_get_enc_ring_prio(i);
 
 		ring = &adev->vcn.inst->ring_enc[i];
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 		sprintf(ring->name, "vcn_enc%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
 				     hw_prio, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 7c2b3aa480836..c975aed2f6c78 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -129,7 +129,7 @@ static int vcn_v2_0_sw_init(void *handle)
 
 	ring->use_doorbell = true;
 	ring->doorbell_index = adev->doorbell_index.vcn.vcn_ring0_1 << 1;
-	ring->vm_hub = AMDGPU_MMHUB_0;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
 
 	sprintf(ring->name, "vcn_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
@@ -160,7 +160,7 @@ static int vcn_v2_0_sw_init(void *handle)
 
 		ring = &adev->vcn.inst->ring_enc[i];
 		ring->use_doorbell = true;
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 		if (!amdgpu_sriov_vf(adev))
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i;
 		else
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index ab0b45d0ead18..7044bd7c9f620 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -188,9 +188,9 @@ static int vcn_v2_5_sw_init(void *handle)
 				(amdgpu_sriov_vf(adev) ? 2*j : 8*j);
 
 		if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0))
-			ring->vm_hub = AMDGPU_MMHUB_1;
+			ring->vm_hub = AMDGPU_MMHUB1(0);
 		else
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 
 		sprintf(ring->name, "vcn_dec_%d", j);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[j].irq,
@@ -208,9 +208,9 @@ static int vcn_v2_5_sw_init(void *handle)
 					(amdgpu_sriov_vf(adev) ? (1 + i + 2*j) : (2 + i + 8*j));
 
 			if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0))
-				ring->vm_hub = AMDGPU_MMHUB_1;
+				ring->vm_hub = AMDGPU_MMHUB1(0);
 			else
-				ring->vm_hub = AMDGPU_MMHUB_0;
+				ring->vm_hub = AMDGPU_MMHUB0(0);
 
 			sprintf(ring->name, "vcn_enc_%d.%d", j, i);
 			r = amdgpu_ring_init(adev, ring, 512,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 3eab186261aab..70fefbf26c48a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -189,7 +189,7 @@ static int vcn_v3_0_sw_init(void *handle)
 		} else {
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 8 * i;
 		}
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 		sprintf(ring->name, "vcn_dec_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0,
 				     AMDGPU_RING_PRIO_DEFAULT,
@@ -213,7 +213,7 @@ static int vcn_v3_0_sw_init(void *handle)
 			} else {
 				ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + j + 8 * i;
 			}
-			ring->vm_hub = AMDGPU_MMHUB_0;
+			ring->vm_hub = AMDGPU_MMHUB0(0);
 			sprintf(ring->name, "vcn_enc_%d.%d", i, j);
 			r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0,
 					     hw_prio, &adev->vcn.inst[i].sched_score);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index bf0674039598d..81446e6996df6 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -149,7 +149,7 @@ static int vcn_v4_0_sw_init(void *handle)
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + i * (adev->vcn.num_enc_rings + 1) + 1;
 		else
 			ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + 8 * i;
-		ring->vm_hub = AMDGPU_MMHUB_0;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
 		sprintf(ring->name, "vcn_unified_%d", i);
 
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0,
-- 
GitLab


From b35ce49ab9ca2bc6a59d4441efc5039de80643ce Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Sun, 19 Dec 2021 11:03:59 +0800
Subject: [PATCH 1013/1791] drm/amdgpu: assign register address for vmhub
 object on each XCD
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Each XCD has its own gfxhub.

v2: switch to the new VMHUB layout
v3: fix mistake

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 58 ++++++++++++++----------
 1 file changed, 34 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index 9c385ce3a8c40..e5016fea1f280 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -483,33 +483,43 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 
 static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+	struct amdgpu_vmhub *hub;
+	int i;
 
-	hub->ctx0_ptb_addr_lo32 =
-		SOC15_REG_OFFSET(GC, 0,
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
+
+		hub->ctx0_ptb_addr_lo32 =
+			SOC15_REG_OFFSET(GC, i,
 				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
-	hub->ctx0_ptb_addr_hi32 =
-		SOC15_REG_OFFSET(GC, 0,
+		hub->ctx0_ptb_addr_hi32 =
+			SOC15_REG_OFFSET(GC, i,
 				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
-	hub->vm_inv_eng0_sem =
-		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_SEM);
-	hub->vm_inv_eng0_req =
-		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_REQ);
-	hub->vm_inv_eng0_ack =
-		SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ACK);
-	hub->vm_context0_cntl =
-		SOC15_REG_OFFSET(GC, 0, regVM_CONTEXT0_CNTL);
-	hub->vm_l2_pro_fault_status =
-		SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_STATUS);
-	hub->vm_l2_pro_fault_cntl =
-		SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL);
-
-	hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
-	hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
-		regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
-	hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ;
-	hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
-		regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+		hub->vm_inv_eng0_sem =
+			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_SEM);
+		hub->vm_inv_eng0_req =
+			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_REQ);
+		hub->vm_inv_eng0_ack =
+			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_ACK);
+		hub->vm_context0_cntl =
+			SOC15_REG_OFFSET(GC, i, regVM_CONTEXT0_CNTL);
+		hub->vm_l2_pro_fault_status =
+			SOC15_REG_OFFSET(GC, i,
+				regVM_L2_PROTECTION_FAULT_STATUS);
+		hub->vm_l2_pro_fault_cntl =
+			SOC15_REG_OFFSET(GC, i, regVM_L2_PROTECTION_FAULT_CNTL);
+
+		hub->ctx_distance = regVM_CONTEXT1_CNTL -
+				regVM_CONTEXT0_CNTL;
+		hub->ctx_addr_distance =
+				regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+		hub->eng_distance = regVM_INVALIDATE_ENG1_REQ -
+				regVM_INVALIDATE_ENG0_REQ;
+		hub->eng_addr_distance =
+				regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+				regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+	}
 }
 
 
-- 
GitLab


From d9426c3d9b4e91dda4f1f1684f9296762fafe0de Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 20 Dec 2021 16:06:25 +0800
Subject: [PATCH 1014/1791] drm/amdgpu: add bitmask to iterate vmhubs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As the layout of VMHUB definition has been changed to cover multiple
XCD/AID case, the original num_vmhubs is not appropriate to do vmhub
iteration any more.

Drop num_vmhubs and introduce vmhubs_mask instead.

v2: switch to the new VMHUB layout
v3: use DECLARE_BITMAP to define vmhubs_mask

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c   |  4 ++--
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c     |  5 +++--
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c     |  5 +++--
 drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c      | 21 ++++++++++++---------
 9 files changed, 25 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 957b18bda4a79..0f163d266812c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -829,7 +829,7 @@ struct amdgpu_device {
 	dma_addr_t			dummy_page_addr;
 	struct amdgpu_vm_manager	vm_manager;
 	struct amdgpu_vmhub             vmhub[AMDGPU_MAX_VMHUBS];
-	unsigned			num_vmhubs;
+	DECLARE_BITMAP(vmhubs_mask, AMDGPU_MAX_VMHUBS);
 
 	/* memory management */
 	struct amdgpu_mman		mman;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index f0a136d352798..5afbcc390d89b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -733,7 +733,7 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev,
 	if (adev->family == AMDGPU_FAMILY_AI) {
 		int i;
 
-		for (i = 0; i < adev->num_vmhubs; i++)
+		for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 			amdgpu_gmc_flush_gpu_tlb(adev, vmid, i, 0);
 	} else {
 		amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB(0), 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index 01cb89ffbd566..6b12f4a75fc37 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -182,7 +182,7 @@ void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,
 	}
 	mb();
 	amdgpu_device_flush_hdp(adev, NULL);
-	for (i = 0; i < adev->num_vmhubs; i++)
+	for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 		amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0);
 
 	drm_dev_exit(idx);
@@ -264,7 +264,7 @@ void amdgpu_gart_invalidate_tlb(struct amdgpu_device *adev)
 
 	mb();
 	amdgpu_device_flush_hdp(adev, NULL);
-	for (i = 0; i < adev->num_vmhubs; i++)
+	for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 		amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index ea2a448147e37..ff96f11c2adf0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -460,7 +460,7 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 				&queried_pasid);
 		if (ret	&& queried_pasid == pasid) {
 			if (all_hub) {
-				for (i = 0; i < adev->num_vmhubs; i++)
+				for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 					gmc_v10_0_flush_gpu_tlb(adev, vmid,
 							i, flush_type);
 			} else {
@@ -928,7 +928,8 @@ static int gmc_v10_0_sw_init(void *handle)
 	case IP_VERSION(10, 3, 6):
 	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(10, 3, 7):
-		adev->num_vmhubs = 2;
+		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
 		/*
 		 * To fulfill 4-level page support,
 		 * vm size is 256TB (48bit), maximum size of Navi10/Navi14/Navi12,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index fb2ac31cbba72..3453f1c0e0660 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -364,7 +364,7 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 				&queried_pasid);
 		if (ret	&& queried_pasid == pasid) {
 			if (all_hub) {
-				for (i = 0; i < adev->num_vmhubs; i++)
+				for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 					gmc_v11_0_flush_gpu_tlb(adev, vmid,
 							i, flush_type);
 			} else {
@@ -779,7 +779,8 @@ static int gmc_v11_0_sw_init(void *handle)
 	case IP_VERSION(11, 0, 2):
 	case IP_VERSION(11, 0, 3):
 	case IP_VERSION(11, 0, 4):
-		adev->num_vmhubs = 2;
+		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
 		/*
 		 * To fulfill 4-level page support,
 		 * vm size is 256TB (48bit), maximum size,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index b7dad4e678135..aa754c95a0b3a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -808,7 +808,7 @@ static int gmc_v6_0_sw_init(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	adev->num_vmhubs = 1;
+	set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
 
 	if (adev->flags & AMD_IS_APU) {
 		adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 402960b0174e2..81609a2b226ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -977,7 +977,7 @@ static int gmc_v7_0_sw_init(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	adev->num_vmhubs = 1;
+	set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
 
 	if (adev->flags & AMD_IS_APU) {
 		adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 504c1b34dab7d..d48e33738a882 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -1093,7 +1093,7 @@ static int gmc_v8_0_sw_init(void *handle)
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	adev->num_vmhubs = 1;
+	set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
 
 	if (adev->flags & AMD_IS_APU) {
 		adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 193ba4d912a69..d4bfb5f8308a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -481,7 +481,7 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
-		for (j = 0; j < adev->num_vmhubs; j++) {
+		for_each_set_bit(j, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) {
 			hub = &adev->vmhub[j];
 			for (i = 0; i < 16; i++) {
 				reg = hub->vm_context0_cntl + i;
@@ -509,7 +509,7 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
 		}
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
-		for (j = 0; j < adev->num_vmhubs; j++) {
+		for_each_set_bit(j, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) {
 			hub = &adev->vmhub[j];
 			for (i = 0; i < 16; i++) {
 				reg = hub->vm_context0_cntl + i;
@@ -803,7 +803,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	u32 j, inv_req, inv_req2, tmp;
 	struct amdgpu_vmhub *hub;
 
-	BUG_ON(vmhub >= adev->num_vmhubs);
+	BUG_ON(vmhub >= AMDGPU_MAX_VMHUBS);
 
 	hub = &adev->vmhub[vmhub];
 	if (adev->gmc.xgmi.num_physical_nodes &&
@@ -987,7 +987,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 				&queried_pasid);
 		if (ret && queried_pasid == pasid) {
 			if (all_hub) {
-				for (i = 0; i < adev->num_vmhubs; i++)
+				for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS)
 					gmc_v9_0_flush_gpu_tlb(adev, vmid,
 							i, flush_type);
 			} else {
@@ -1684,7 +1684,8 @@ static int gmc_v9_0_sw_init(void *handle)
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 1, 0):
 	case IP_VERSION(9, 2, 2):
-		adev->num_vmhubs = 2;
+		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
 
 		if (adev->rev_id == 0x0 || adev->rev_id == 0x1) {
 			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
@@ -1701,8 +1702,8 @@ static int gmc_v9_0_sw_init(void *handle)
 	case IP_VERSION(9, 3, 0):
 	case IP_VERSION(9, 4, 2):
 	case IP_VERSION(9, 4, 3):
-		adev->num_vmhubs = 2;
-
+		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
 
 		/*
 		 * To fulfill 4-level page support,
@@ -1718,7 +1719,9 @@ static int gmc_v9_0_sw_init(void *handle)
 			adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
 	case IP_VERSION(9, 4, 1):
-		adev->num_vmhubs = 3;
+		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
+		set_bit(AMDGPU_MMHUB1(0), adev->vmhubs_mask);
 
 		/* Keep the vm size same with Vega20 */
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
@@ -1944,7 +1947,7 @@ static int gmc_v9_0_hw_init(void *handle)
 			adev->gfxhub.funcs->set_fault_enable_default(adev, value);
 		adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	}
-	for (i = 0; i < adev->num_vmhubs; ++i) {
+	for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) {
 		if (adev->in_s0ix && (i == AMDGPU_GFXHUB(0)))
 			continue;
 		gmc_v9_0_flush_gpu_tlb(adev, 0, i, 0);
-- 
GitLab


From ce8a12a532ed62d7037be91c5714243fdfa9f672 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 20 Dec 2021 16:42:20 +0800
Subject: [PATCH 1015/1791] drm/amdgpu: init vmhubs bitmask for GC 9.4.3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Each XCD owns one GFXHUB.

v2: switch to the new VMHUB layout

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index d4bfb5f8308a3..6da85365e5aad 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1701,7 +1701,6 @@ static int gmc_v9_0_sw_init(void *handle)
 	case IP_VERSION(9, 4, 0):
 	case IP_VERSION(9, 3, 0):
 	case IP_VERSION(9, 4, 2):
-	case IP_VERSION(9, 4, 3):
 		set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask);
 		set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask);
 
@@ -1727,6 +1726,12 @@ static int gmc_v9_0_sw_init(void *handle)
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
 		adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
+	case IP_VERSION(9, 4, 3):
+		bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), adev->gfx.num_xcd);
+		bitmap_set(adev->vmhubs_mask, AMDGPU_MMHUB0(0), 1);
+
+		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
+		break;
 	default:
 		break;
 	}
-- 
GitLab


From 3566938b3491bb3aad701b487130f3efc363e2dc Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 20 Dec 2021 16:06:25 +0800
Subject: [PATCH 1016/1791] drm/amdgpu: assign different AMDGPU_GFXHUB for
 rings on each xcc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pass the xcc_id to AMDGPU_GFXHUB(x)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 488b3bb6dcb13..e5ff0bf9f23bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -315,7 +315,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->use_doorbell = true;
 	ring->doorbell_index = adev->doorbell_index.kiq;
 	ring->xcc_id = xcc_id;
-	ring->vm_hub = AMDGPU_GFXHUB(0);
+	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
 	if (xcc_id >= 1)
 		ring->doorbell_index = adev->doorbell_index.xcc1_kiq_start +
 					xcc_id - 1;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index f5104b9826337..064cd02451c28 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -757,7 +757,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 				(adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
-	ring->vm_hub = AMDGPU_GFXHUB(0);
+	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
 	sprintf(ring->name, "comp_%d.%d.%d.%d",
 			ring->xcc_id, ring->me, ring->pipe, ring->queue);
 
-- 
GitLab


From 98a54e88e87f7291d4bbc6ec646c498f64ae042f Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 17 May 2022 22:20:10 +0800
Subject: [PATCH 1017/1791] drm/amdgpu: add sysfs node for compute partition
 mode

Add current/available compute partitin mode sysfs node.

v2: make the sysfs node as IP independent one in amdgpu_gfx.c

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |   1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c    | 132 +++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h    |   6 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h   |   3 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c    |  67 +++++++++++
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c     |  32 +++++
 6 files changed, 241 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index e94507a10e15e..f432064a05358 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3634,6 +3634,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->srbm_mutex);
 	mutex_init(&adev->gfx.pipe_reserve_mutex);
 	mutex_init(&adev->gfx.gfx_off_mutex);
+	mutex_init(&adev->gfx.partition_mutex);
 	mutex_init(&adev->grbm_idx_mutex);
 	mutex_init(&adev->mn_lock);
 	mutex_init(&adev->virt.vf_errors.lock);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index e5ff0bf9f23bc..4a4d71ff9b95d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1174,3 +1174,135 @@ bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id)
 	return !(xcc_id % (adev->gfx.num_xcc_per_xcp ?
 			adev->gfx.num_xcc_per_xcp : 1));
 }
+
+static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
+						struct device_attribute *addr,
+						char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	enum amdgpu_gfx_partition mode;
+	char *partition_mode;
+
+	mode = adev->gfx.funcs->query_partition_mode(adev);
+
+	switch (mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		partition_mode = "SPX";
+		break;
+	case AMDGPU_DPX_PARTITION_MODE:
+		partition_mode = "DPX";
+		break;
+	case AMDGPU_TPX_PARTITION_MODE:
+		partition_mode = "TPX";
+		break;
+	case AMDGPU_QPX_PARTITION_MODE:
+		partition_mode = "QPX";
+		break;
+	case AMDGPU_CPX_PARTITION_MODE:
+		partition_mode = "CPX";
+		break;
+	default:
+		partition_mode = "UNKNOWN";
+		break;
+	}
+
+	return sysfs_emit(buf, "%s\n", partition_mode);
+}
+
+static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
+						struct device_attribute *addr,
+						const char *buf, size_t count)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	enum amdgpu_gfx_partition mode;
+	int ret;
+
+	if (adev->gfx.num_xcd % 2 != 0)
+		return -EINVAL;
+
+	if (!strncasecmp("SPX", buf, strlen("SPX"))) {
+		mode = AMDGPU_SPX_PARTITION_MODE;
+	} else if (!strncasecmp("DPX", buf, strlen("DPX"))) {
+		if (adev->gfx.num_xcd != 4 || adev->gfx.num_xcd != 8)
+			return -EINVAL;
+		mode = AMDGPU_DPX_PARTITION_MODE;
+	} else if (!strncasecmp("TPX", buf, strlen("TPX"))) {
+		if (adev->gfx.num_xcd != 6)
+			return -EINVAL;
+		mode = AMDGPU_TPX_PARTITION_MODE;
+	} else if (!strncasecmp("QPX", buf, strlen("QPX"))) {
+		if (adev->gfx.num_xcd != 8)
+			return -EINVAL;
+		mode = AMDGPU_QPX_PARTITION_MODE;
+	} else if (!strncasecmp("CPX", buf, strlen("CPX"))) {
+		mode = AMDGPU_CPX_PARTITION_MODE;
+	} else {
+		return -EINVAL;
+	}
+
+	mutex_lock(&adev->gfx.partition_mutex);
+
+	ret = adev->gfx.funcs->switch_partition_mode(adev, mode);
+
+	mutex_unlock(&adev->gfx.partition_mutex);
+
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev,
+						struct device_attribute *addr,
+						char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	char *supported_partition;
+
+	/* TBD */
+	switch (adev->gfx.num_xcd) {
+	case 8:
+		supported_partition = "SPX, DPX, QPX, CPX";
+		break;
+	case 6:
+		supported_partition = "SPX, TPX, CPX";
+		break;
+	case 4:
+		supported_partition = "SPX, DPX, CPX";
+		break;
+	/* this seems only existing in emulation phase */
+	case 2:
+		supported_partition = "SPX, CPX";
+		break;
+	default:
+		supported_partition = "Not supported";
+		break;
+	}
+
+	return sysfs_emit(buf, "%s\n", supported_partition);
+}
+
+static DEVICE_ATTR(current_compute_partition, S_IRUGO | S_IWUSR,
+		   amdgpu_gfx_get_current_compute_partition,
+		   amdgpu_gfx_set_compute_partition);
+
+static DEVICE_ATTR(available_compute_partition, S_IRUGO,
+		   amdgpu_gfx_get_available_compute_partition, NULL);
+
+int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
+{
+	int r;
+
+	r = device_create_file(adev->dev, &dev_attr_current_compute_partition);
+	if (r)
+		return r;
+
+	r = device_create_file(adev->dev, &dev_attr_available_compute_partition);
+	if (r)
+		return r;
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 2755f00ac19a2..8df36527aee94 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -259,6 +259,10 @@ struct amdgpu_gfx_funcs {
 	void (*update_perfmon_mgcg)(struct amdgpu_device *adev, bool enable);
 	int (*get_gfx_shadow_info)(struct amdgpu_device *adev,
 				   struct amdgpu_gfx_shadow_info *shadow_info);
+	enum amdgpu_gfx_partition
+			(*query_partition_mode)(struct amdgpu_device *adev);
+	int (*switch_partition_mode)(struct amdgpu_device *adev,
+				     enum amdgpu_gfx_partition mode);
 };
 
 struct sq_work {
@@ -394,6 +398,7 @@ struct amdgpu_gfx {
 	enum amdgpu_gfx_partition	partition_mode;
 	uint32_t			num_xcd;
 	uint32_t			num_xcc_per_xcp;
+	struct mutex			partition_mutex;
 };
 
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
@@ -478,4 +483,5 @@ int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev,
 						struct amdgpu_iv_entry *entry);
 
 bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id);
+int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index c686ff4bcc393..6a636c34b7178 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -95,6 +95,9 @@ struct amdgpu_nbio_funcs {
 	void (*apply_l1_link_width_reconfig_wa)(struct amdgpu_device *adev);
 	void (*clear_doorbell_interrupt)(struct amdgpu_device *adev);
 	u32 (*get_rom_offset)(struct amdgpu_device *adev);
+	u32 (*get_compute_partition_mode)(struct amdgpu_device *adev);
+	void (*set_compute_partition_mode)(struct amdgpu_device *adev,
+					   enum amdgpu_gfx_partition mode);
 };
 
 struct amdgpu_nbio {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 064cd02451c28..1d15db9423c98 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -662,6 +662,67 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 	soc15_grbm_select(adev, me, pipe, q, vm, 0);
 }
 
+static enum amdgpu_gfx_partition
+gfx_v9_4_3_query_compute_partition(struct amdgpu_device *adev)
+{
+	enum amdgpu_gfx_partition mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+
+	if (adev->nbio.funcs->get_compute_partition_mode)
+		mode = adev->nbio.funcs->get_compute_partition_mode(adev);
+
+	return mode;
+}
+
+static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
+						enum amdgpu_gfx_partition mode)
+{
+	u32 tmp = 0;
+	int num_xcc_per_partition, i;
+
+	if (mode == adev->gfx.partition_mode)
+		return mode;
+
+	switch (mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		num_xcc_per_partition = adev->gfx.num_xcd;
+		break;
+	case AMDGPU_DPX_PARTITION_MODE:
+		num_xcc_per_partition = adev->gfx.num_xcd / 2;
+		break;
+	case AMDGPU_TPX_PARTITION_MODE:
+		num_xcc_per_partition = adev->gfx.num_xcd / 3;
+		break;
+	case AMDGPU_QPX_PARTITION_MODE:
+		num_xcc_per_partition = adev->gfx.num_xcd / 4;
+		break;
+	case AMDGPU_CPX_PARTITION_MODE:
+		num_xcc_per_partition = 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* TODO:
+	 * Stop user queues and threads, and make sure GPU is empty of work.
+	 */
+
+	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP,
+				    num_xcc_per_partition);
+		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID,
+				    i % num_xcc_per_partition);
+		WREG32_SOC15(GC, i, regCP_HYP_XCP_CTL, tmp);
+	}
+
+	if (adev->nbio.funcs->set_compute_partition_mode)
+		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
+
+	adev->gfx.num_xcc_per_xcp = num_xcc_per_partition;
+	adev->gfx.partition_mode = mode;
+
+	return 0;
+}
+
 static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v9_4_3_select_se_sh,
@@ -669,6 +730,8 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs,
 	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
+	.query_partition_mode = &gfx_v9_4_3_query_compute_partition,
+	.switch_partition_mode = &gfx_v9_4_3_switch_compute_partition,
 };
 
 static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
@@ -858,6 +921,10 @@ static int gfx_v9_4_3_sw_init(void *handle)
 	if (r)
 		return r;
 
+	r = amdgpu_gfx_sysfs_init(adev);
+	if (r)
+		return r;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index a331a59c49e38..bdb84a53f0e44 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -347,6 +347,36 @@ static void nbio_v7_9_enable_doorbell_interrupt(struct amdgpu_device *adev,
 			      DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1);
 }
 
+static enum amdgpu_gfx_partition nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
+{
+	u32 tmp;
+
+	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP);
+
+	if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, SPX_SUPPORT))
+		return AMDGPU_SPX_PARTITION_MODE;
+	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, DPX_SUPPORT))
+		return AMDGPU_DPX_PARTITION_MODE;
+	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, TPX_SUPPORT))
+		return AMDGPU_TPX_PARTITION_MODE;
+	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, CPX_SUPPORT))
+		return AMDGPU_CPX_PARTITION_MODE;
+	else
+		return AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+}
+
+static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
+					enum amdgpu_gfx_partition mode)
+{
+	u32 tmp;
+
+	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP);
+	tmp &= ~0x1f;
+	tmp |= 1 << mode;
+
+	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP, tmp);
+}
+
 const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_hdp_flush_req_offset = nbio_v7_9_get_hdp_flush_req_offset,
 	.get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset,
@@ -366,4 +396,6 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_clockgating_state = nbio_v7_9_get_clockgating_state,
 	.ih_control = nbio_v7_9_ih_control,
 	.remap_hdp_registers = nbio_v7_9_remap_hdp_registers,
+	.get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode,
+	.set_compute_partition_mode = nbio_v7_9_set_compute_partition_mode,
 };
-- 
GitLab


From 99951878b048e39278bb05d96831353eca23b75f Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 29 Nov 2021 20:44:05 +0800
Subject: [PATCH 1018/1791] drm/amdgpu: make the WREG32_SOC15_xx macro to
 support multi GC

To write regs on different GCDs, use the inst index.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15_common.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index 9fefd403e14fb..3f6dac856d492 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -157,10 +157,10 @@
 	do {							\
 		uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\
 		if (amdgpu_sriov_fullaccess(adev)) {    \
-			uint32_t r2 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG2;	\
-			uint32_t r3 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG3;	\
-			uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][0][prefix##GRBM_GFX_CNTL_BASE_IDX] + prefix##GRBM_GFX_CNTL;   \
-			uint32_t grbm_idx = adev->reg_offset[GC_HWIP][0][prefix##GRBM_GFX_INDEX_BASE_IDX] + prefix##GRBM_GFX_INDEX;   \
+			uint32_t r2 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG2;	\
+			uint32_t r3 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG3;	\
+			uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][inst][prefix##GRBM_GFX_CNTL_BASE_IDX] + prefix##GRBM_GFX_CNTL;   \
+			uint32_t grbm_idx = adev->reg_offset[GC_HWIP][inst][prefix##GRBM_GFX_INDEX_BASE_IDX] + prefix##GRBM_GFX_INDEX;   \
 			if (target_reg == grbm_cntl) \
 				WREG32(r2, value);	\
 			else if (target_reg == grbm_idx) \
@@ -176,13 +176,13 @@
 
 #define WREG32_SOC15_RLC(ip, inst, reg, value) \
 	do {							\
-		uint32_t target_reg = adev->reg_offset[ip##_HWIP][0][reg##_BASE_IDX] + reg;\
+		uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\
 		__WREG32_SOC15_RLC__(target_reg, value, AMDGPU_REGS_RLC, ip##_HWIP); \
 	} while (0)
 
 #define WREG32_SOC15_RLC_EX(prefix, ip, inst, reg, value) \
 	do {							\
-			uint32_t target_reg = adev->reg_offset[GC_HWIP][0][reg##_BASE_IDX] + reg;\
+			uint32_t target_reg = adev->reg_offset[GC_HWIP][inst][reg##_BASE_IDX] + reg;\
 			WREG32_RLC_EX(prefix, target_reg, value); \
 	} while (0)
 
-- 
GitLab


From 0fa49d108386c201b5c2cce68066a9b8f66883a5 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Fri, 17 Dec 2021 11:27:53 +0800
Subject: [PATCH 1019/1791] drm/amdgpu: override partition mode through module
 parameter

Add a module parameter to override the partition mode.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 25 +++++++++++++++++++++++--
 3 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 0f163d266812c..a277bdc86057b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -242,6 +242,8 @@ extern int amdgpu_num_kcq;
 extern int amdgpu_vcnfw_log;
 extern int amdgpu_sg_display;
 
+extern uint amdgpu_user_partt_mode;
+
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
 #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 45544ebe576e1..9e9da2ac5c82e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -193,6 +193,7 @@ int amdgpu_smartshift_bias;
 int amdgpu_use_xgmi_p2p = 1;
 int amdgpu_vcnfw_log;
 int amdgpu_sg_display = -1; /* auto */
+uint amdgpu_user_partt_mode;
 
 static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work);
 
@@ -950,6 +951,18 @@ MODULE_PARM_DESC(smu_pptable_id,
 	"specify pptable id to be used (-1 = auto(default) value, 0 = use pptable from vbios, > 0 = soft pptable id)");
 module_param_named(smu_pptable_id, amdgpu_smu_pptable_id, int, 0444);
 
+/**
+ * DOC: partition_mode (int)
+ * Used to override the default SPX mode.
+ */
+MODULE_PARM_DESC(user_partt_mode,
+	"specify partition mode to be used (0 = AMDGPU_SPX_PARTITION_MODE(default value), \
+						1 = AMDGPU_DPX_PARTITION_MODE, \
+						2 = AMDGPU_TPX_PARTITION_MODE, \
+						3 = AMDGPU_QPX_PARTITION_MODE, \
+						4 = AMDGPU_CPX_PARTITION_MODE)");
+module_param_named(user_partt_mode, amdgpu_user_partt_mode, uint, 0444);
+
 /* These devices are not supported by amdgpu.
  * They are supported by the mach64, r128, radeon drivers
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 1d15db9423c98..2676a185c232d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2162,8 +2162,29 @@ static int gfx_v9_4_3_early_init(void *handle)
 
 	/* hardcode in emulation phase */
 	adev->gfx.num_xcd = 1;
-	adev->gfx.num_xcc_per_xcp = 1;
-	adev->gfx.partition_mode = AMDGPU_SPX_PARTITION_MODE;
+
+	adev->gfx.partition_mode = amdgpu_user_partt_mode;
+	/* calculate the num_xcc_in_xcp for the partition mode*/
+	switch (amdgpu_user_partt_mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd;
+		break;
+	case AMDGPU_DPX_PARTITION_MODE:
+		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 2;
+		break;
+	case AMDGPU_TPX_PARTITION_MODE:
+		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 3;
+		break;
+	case AMDGPU_QPX_PARTITION_MODE:
+		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 4;
+		break;
+	case AMDGPU_CPX_PARTITION_MODE:
+		adev->gfx.num_xcc_per_xcp = 1;
+		break;
+	default:
+		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd;
+		break;
+	}
 
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
-- 
GitLab


From 62e790879efbf09edb9f262d5eb7765aeaf89809 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Thu, 23 Dec 2021 13:44:48 +0800
Subject: [PATCH 1020/1791] drm/amdgpu: alloc vm inv engines for every vmhub

There are AMDGPU_MAX_VMHUBS of vmhub in maximum and need to init the
vm_inv_engs for all of them.

In this way, the below error can be ruled out.
[  217.317752] amdgpu 0000:02:00.0: amdgpu: no VM inv eng for ring sdma0

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Christian Koenig <Christian.Koenig@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 0a4e5fcfec6bb..b8825a0670a41 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -534,22 +534,21 @@ void amdgpu_gmc_ras_fini(struct amdgpu_device *adev)
 	 *                    subject to change when ring number changes
 	 * Engine 17: Gart flushes
 	 */
-#define GFXHUB_FREE_VM_INV_ENGS_BITMAP		0x1FFF3
-#define MMHUB_FREE_VM_INV_ENGS_BITMAP		0x1FFF3
+#define AMDGPU_VMHUB_INV_ENG_BITMAP		0x1FFF3
 
 int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
-	unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] =
-		{GFXHUB_FREE_VM_INV_ENGS_BITMAP, MMHUB_FREE_VM_INV_ENGS_BITMAP,
-		GFXHUB_FREE_VM_INV_ENGS_BITMAP};
+	unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] = {0};
 	unsigned i;
 	unsigned vmhub, inv_eng;
 
-	if (adev->enable_mes) {
+	/* init the vm inv eng for all vmhubs */
+	for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) {
+		vm_inv_engs[i] = AMDGPU_VMHUB_INV_ENG_BITMAP;
 		/* reserve engine 5 for firmware */
-		for (vmhub = 0; vmhub < AMDGPU_MAX_VMHUBS; vmhub++)
-			vm_inv_engs[vmhub] &= ~(1 << 5);
+		if (adev->enable_mes)
+			vm_inv_engs[i] &= ~(1 << 5);
 	}
 
 	for (i = 0; i < adev->num_rings; ++i) {
-- 
GitLab


From f544afac3f34124088b981c63843a3cc48f4ee3e Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Tue, 17 May 2022 23:41:01 +0800
Subject: [PATCH 1021/1791] drm/amdgpu: Add kgd2kfd for GC 9.4.3

New GC (v9.4.3) and ATHUB (v1.8.0) versions
are used. Add kgd_gfx_v9_4_3_*
functions if registers in use of kgd_gfx_v9_*
functions are changed or have different offset.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile           |   1 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   | 183 ++++++++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c |  30 +--
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |   5 +
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |   3 +-
 5 files changed, 206 insertions(+), 16 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 415a7fa395c4c..7fb09000efc06 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -228,6 +228,7 @@ amdgpu-y += \
 	amdgpu_amdkfd_gfx_v9.o \
 	amdgpu_amdkfd_arcturus.o \
 	amdgpu_amdkfd_aldebaran.o \
+	amdgpu_amdkfd_gc_9_4_3.o \
 	amdgpu_amdkfd_gfx_v10.o \
 	amdgpu_amdkfd_gfx_v10_3.o \
 	amdgpu_amdkfd_gfx_v11.o
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
new file mode 100644
index 0000000000000..562e1a04160f1
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -0,0 +1,183 @@
+/*
+ * Copyright 2021 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include "amdgpu.h"
+#include "amdgpu_amdkfd.h"
+#include "amdgpu_amdkfd_arcturus.h"
+#include "amdgpu_amdkfd_gfx_v9.h"
+#include "gc/gc_9_4_3_offset.h"
+#include "gc/gc_9_4_3_sh_mask.h"
+#include "athub/athub_1_8_0_offset.h"
+#include "athub/athub_1_8_0_sh_mask.h"
+#include "oss/osssys_4_0_offset.h"
+#include "oss/osssys_4_0_sh_mask.h"
+#include "v9_structs.h"
+#include "soc15.h"
+
+static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
+				u32 pasid, unsigned int vmid)
+{
+	unsigned long timeout;
+
+	/*
+	 * We have to assume that there is no outstanding mapping.
+	 * The ATC_VMID_PASID_MAPPING_UPDATE_STATUS bit could be 0 because
+	 * a mapping is in progress or because a mapping finished
+	 * and the SW cleared it.
+	 * So the protocol is to always wait & clear.
+	 */
+	uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid |
+			ATC_VMID0_PASID_MAPPING__VALID_MASK;
+
+	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
+		regATC_VMID0_PASID_MAPPING) + vmid, pasid_mapping);
+
+	timeout = jiffies + msecs_to_jiffies(10);
+	while (!(RREG32(SOC15_REG_OFFSET(ATHUB, 0,
+			regATC_VMID_PASID_MAPPING_UPDATE_STATUS)) &
+			(1U << vmid))) {
+		if (time_after(jiffies, timeout)) {
+			pr_err("Fail to program VMID-PASID mapping\n");
+			return -ETIME;
+		}
+		cpu_relax();
+	}
+
+	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
+		regATC_VMID_PASID_MAPPING_UPDATE_STATUS),
+		1U << vmid);
+
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid,
+		pasid_mapping);
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid,
+		pasid_mapping);
+
+	return 0;
+}
+
+static inline struct v9_mqd *get_mqd(void *mqd)
+{
+	return (struct v9_mqd *)mqd;
+}
+
+static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
+			uint32_t pipe_id, uint32_t queue_id,
+			uint32_t __user *wptr, uint32_t wptr_shift,
+			uint32_t wptr_mask, struct mm_struct *mm)
+{
+	struct v9_mqd *m;
+	uint32_t *mqd_hqd;
+	uint32_t reg, hqd_base, hqd_end, data;
+
+	m = get_mqd(mqd);
+
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+
+	/* HQD registers extend to CP_HQD_AQL_DISPATCH_ID_HI */
+	mqd_hqd = &m->cp_mqd_base_addr_lo;
+	hqd_base = SOC15_REG_OFFSET(GC, 0, regCP_MQD_BASE_ADDR);
+	hqd_end = SOC15_REG_OFFSET(GC, 0, regCP_HQD_AQL_DISPATCH_ID_HI);
+
+	for (reg = hqd_base; reg <= hqd_end; reg++)
+		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
+
+
+	/* Activate doorbell logic before triggering WPTR poll. */
+	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
+			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL),
+				data);
+
+	if (wptr) {
+		/* Don't read wptr with get_user because the user
+		 * context may not be accessible (if this function
+		 * runs in a work queue). Instead trigger a one-shot
+		 * polling read from memory in the CP. This assumes
+		 * that wptr is GPU-accessible in the queue's VMID via
+		 * ATC or SVM. WPTR==RPTR before starting the poll so
+		 * the CP starts fetching new commands from the right
+		 * place.
+		 *
+		 * Guessing a 64-bit WPTR from a 32-bit RPTR is a bit
+		 * tricky. Assume that the queue didn't overflow. The
+		 * number of valid bits in the 32-bit RPTR depends on
+		 * the queue size. The remaining bits are taken from
+		 * the saved 64-bit WPTR. If the WPTR wrapped, add the
+		 * queue size.
+		 */
+		uint32_t queue_size =
+			2 << REG_GET_FIELD(m->cp_hqd_pq_control,
+					   CP_HQD_PQ_CONTROL, QUEUE_SIZE);
+		uint64_t guessed_wptr = m->cp_hqd_pq_rptr & (queue_size - 1);
+
+		if ((m->cp_hqd_pq_wptr_lo & (queue_size - 1)) < guessed_wptr)
+			guessed_wptr += queue_size;
+		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
+		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
+
+		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_LO),
+		       lower_32_bits(guessed_wptr));
+		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_HI),
+		       upper_32_bits(guessed_wptr));
+		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR),
+		       lower_32_bits((uintptr_t)wptr));
+		WREG32_RLC(SOC15_REG_OFFSET(GC, 0,
+			regCP_HQD_PQ_WPTR_POLL_ADDR_HI),
+			upper_32_bits((uintptr_t)wptr));
+		WREG32(SOC15_REG_OFFSET(GC, 0, regCP_PQ_WPTR_POLL_CNTL1),
+		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id,
+			       queue_id));
+	}
+
+	/* Start the EOP fetcher */
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_EOP_RPTR),
+	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
+			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
+
+	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_ACTIVE), data);
+
+	kgd_gfx_v9_release_queue(adev);
+
+	return 0;
+}
+
+const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = {
+	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
+	.set_pasid_vmid_mapping = kgd_gfx_v9_4_3_set_pasid_vmid_mapping,
+	.init_interrupts = kgd_gfx_v9_init_interrupts,
+	.hqd_load = kgd_gfx_v9_4_3_hqd_load,
+	.hiq_mqd_load = kgd_gfx_v9_hiq_mqd_load,
+	.hqd_sdma_load = kgd_arcturus_hqd_sdma_load,
+	.hqd_dump = kgd_gfx_v9_hqd_dump,
+	.hqd_sdma_dump = kgd_arcturus_hqd_sdma_dump,
+	.hqd_is_occupied = kgd_gfx_v9_hqd_is_occupied,
+	.hqd_sdma_is_occupied = kgd_arcturus_hqd_sdma_is_occupied,
+	.hqd_destroy = kgd_gfx_v9_hqd_destroy,
+	.hqd_sdma_destroy = kgd_arcturus_hqd_sdma_destroy,
+	.wave_control_execute = kgd_gfx_v9_wave_control_execute,
+	.get_atc_vmid_pasid_mapping_info =
+				kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
+	.set_vm_context_page_table_base =
+				kgd_gfx_v9_set_vm_context_page_table_base,
+	.program_trap_handler_settings =
+				kgd_gfx_v9_program_trap_handler_settings
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index ae06d1f2af932..d36219ecd3dd0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -59,7 +59,7 @@ static void unlock_srbm(struct amdgpu_device *adev)
 	mutex_unlock(&adev->srbm_mutex);
 }
 
-static void acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
+void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
 				uint32_t queue_id)
 {
 	uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
@@ -68,7 +68,7 @@ static void acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
 	lock_srbm(adev, mec, pipe, queue_id, 0);
 }
 
-static uint64_t get_queue_mask(struct amdgpu_device *adev,
+uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
 			       uint32_t pipe_id, uint32_t queue_id)
 {
 	unsigned int bit = pipe_id * adev->gfx.mec.num_queue_per_pipe +
@@ -77,7 +77,7 @@ static uint64_t get_queue_mask(struct amdgpu_device *adev,
 	return 1ull << bit;
 }
 
-static void release_queue(struct amdgpu_device *adev)
+void kgd_gfx_v9_release_queue(struct amdgpu_device *adev)
 {
 	unlock_srbm(adev);
 }
@@ -228,7 +228,7 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	m = get_mqd(mqd);
 
-	acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
 
 	/* HQD registers extend from CP_MQD_BASE_ADDR to CP_HQD_EOP_WPTR_MEM. */
 	mqd_hqd = &m->cp_mqd_base_addr_lo;
@@ -280,7 +280,7 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
 		       upper_32_bits((uintptr_t)wptr));
 		WREG32_SOC15(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1,
-		       (uint32_t)get_queue_mask(adev, pipe_id, queue_id));
+		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
 	}
 
 	/* Start the EOP fetcher */
@@ -291,7 +291,7 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
 	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_ACTIVE), data);
 
-	release_queue(adev);
+	kgd_gfx_v9_release_queue(adev);
 
 	return 0;
 }
@@ -307,7 +307,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 
 	m = get_mqd(mqd);
 
-	acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
 
 	mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
@@ -343,7 +343,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 
 out_unlock:
 	spin_unlock(&adev->gfx.kiq[0].ring_lock);
-	release_queue(adev);
+	kgd_gfx_v9_release_queue(adev);
 
 	return r;
 }
@@ -365,13 +365,13 @@ int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev,
 	if (*dump == NULL)
 		return -ENOMEM;
 
-	acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
 
 	for (reg = SOC15_REG_OFFSET(GC, 0, mmCP_MQD_BASE_ADDR);
 	     reg <= SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI); reg++)
 		DUMP_REG(reg);
 
-	release_queue(adev);
+	kgd_gfx_v9_release_queue(adev);
 
 	WARN_ON_ONCE(i != HQD_N_REGS);
 	*n_regs = i;
@@ -487,7 +487,7 @@ bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev,
 	bool retval = false;
 	uint32_t low, high;
 
-	acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
 	act = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE);
 	if (act) {
 		low = lower_32_bits(queue_address >> 8);
@@ -497,7 +497,7 @@ bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev,
 		   high == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE_HI))
 			retval = true;
 	}
-	release_queue(adev);
+	kgd_gfx_v9_release_queue(adev);
 	return retval;
 }
 
@@ -532,7 +532,7 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 	if (amdgpu_in_reset(adev))
 		return -EIO;
 
-	acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
 
 	if (m->cp_hqd_vmid == 0)
 		WREG32_FIELD15_RLC(GC, 0, RLC_CP_SCHEDULERS, scheduler1, 0);
@@ -561,13 +561,13 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 			break;
 		if (time_after(jiffies, end_jiffies)) {
 			pr_err("cp queue preemption time out.\n");
-			release_queue(adev);
+			kgd_gfx_v9_release_queue(adev);
 			return -ETIME;
 		}
 		usleep_range(500, 1000);
 	}
 
-	release_queue(adev);
+	kgd_gfx_v9_release_queue(adev);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index c7ed3bc9053c5..491273a02e309 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -58,3 +58,8 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 		int *pasid_wave_cnt, int *max_waves_per_cu);
 void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
 		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr);
+void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t queue_id);
+uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
+				uint32_t pipe_id, uint32_t queue_id);
+void kgd_gfx_v9_release_queue(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 00f528eb98126..1510041a6ee13 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -51,6 +51,7 @@ extern const struct kfd2kgd_calls gfx_v8_kfd2kgd;
 extern const struct kfd2kgd_calls gfx_v9_kfd2kgd;
 extern const struct kfd2kgd_calls arcturus_kfd2kgd;
 extern const struct kfd2kgd_calls aldebaran_kfd2kgd;
+extern const struct kfd2kgd_calls gc_9_4_3_kfd2kgd;
 extern const struct kfd2kgd_calls gfx_v10_kfd2kgd;
 extern const struct kfd2kgd_calls gfx_v10_3_kfd2kgd;
 extern const struct kfd2kgd_calls gfx_v11_kfd2kgd;
@@ -328,7 +329,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 			break;
 		case IP_VERSION(9, 4, 3):
 			gfx_target_version = 90400;
-			f2g = &aldebaran_kfd2kgd;
+			f2g = &gc_9_4_3_kfd2kgd;
 			break;
 		/* Navi10 */
 		case IP_VERSION(10, 1, 10):
-- 
GitLab


From 9b4fd27601fbe7f77e7f8a8ca226211ef748844b Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 12 Jan 2022 19:42:56 +0530
Subject: [PATCH 1022/1791] drm/amdgpu: Use the correct API to read register

Use SOC15 API so that the register offset is calculated correctly.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 2676a185c232d..ad3e8cefbdb27 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1637,7 +1637,7 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 
 	/* set static priority for a queue/ring */
 	gfx_v9_4_3_mqd_set_priority(ring, mqd);
-	mqd->cp_hqd_quantum = RREG32(regCP_HQD_QUANTUM);
+	mqd->cp_hqd_quantum = RREG32_SOC15(GC, 0, regCP_HQD_QUANTUM);
 
 	/* map_queues packet doesn't need activate the queue,
 	 * so only kiq need set this field.
-- 
GitLab


From 81283fee15ea6afb1e32defb369c3b96d2d6765c Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 19 Jan 2022 22:32:41 -0500
Subject: [PATCH 1023/1791] drm/amdgpu/: add more macro to support offset
 variant

Add more macro to support offset variant and
simplify macro SOC15_WAIT_ON_RREG.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        |  3 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 28 +++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/soc15_common.h  | 36 +++++++---------------
 3 files changed, 42 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index a277bdc86057b..00c52caeebeb0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1082,6 +1082,9 @@ size_t amdgpu_device_aper_access(struct amdgpu_device *adev, loff_t pos,
 
 void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos,
 			       void *buf, size_t size, bool write);
+uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev,
+			    uint32_t inst, uint32_t reg_addr, char reg_name[],
+			    uint32_t expected_value, uint32_t mask);
 uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
 			    uint32_t reg, uint32_t acc_flags);
 void amdgpu_device_wreg(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index f432064a05358..82a3d0ff65607 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6081,3 +6081,31 @@ bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev)
 		return true;
 	}
 }
+
+uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev,
+		uint32_t inst, uint32_t reg_addr, char reg_name[],
+		uint32_t expected_value, uint32_t mask)
+{
+	uint32_t ret = 0;
+	uint32_t old_ = 0;
+	uint32_t tmp_ = RREG32(reg_addr);
+	uint32_t loop = adev->usec_timeout;
+
+	while ((tmp_ & (mask)) != (expected_value)) {
+		if (old_ != tmp_) {
+			loop = adev->usec_timeout;
+			old_ = tmp_;
+		} else
+			udelay(1);
+		tmp_ = RREG32(reg_addr);
+		loop--;
+		if (!loop) {
+			DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn",
+				  inst, reg_name, (uint32_t)expected_value,
+				  (uint32_t)(tmp_ & (mask)));
+			ret = -ETIMEDOUT;
+			break;
+		}
+	}
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index 3f6dac856d492..eb35096756b8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -26,6 +26,8 @@
 
 /* Register Access Macros */
 #define SOC15_REG_OFFSET(ip, inst, reg)	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg)
+#define SOC15_REG_OFFSET1(ip, inst, reg, offset) \
+	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)+(offset))
 
 #define __WREG32_SOC15_RLC__(reg, value, flag, hwip) \
 	((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \
@@ -86,31 +88,15 @@
 	 __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, \
 			  value, 0, ip##_HWIP)
 
-#define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask) \
-({	int ret = 0;						\
-	do {							\
-		uint32_t old_ = 0;				\
-		uint32_t tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \
-		uint32_t loop = adev->usec_timeout;		\
-		ret = 0;					\
-		while ((tmp_ & (mask)) != (expected_value)) {	\
-			if (old_ != tmp_) {			\
-				loop = adev->usec_timeout;	\
-				old_ = tmp_;			\
-			} else					\
-				udelay(1);			\
-			tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \
-			loop--;					\
-			if (!loop) {				\
-				DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08x\n", \
-					  inst, #reg, (unsigned)expected_value, (unsigned)(tmp_ & (mask))); \
-				ret = -ETIMEDOUT;		\
-				break;				\
-			}					\
-		}						\
-	} while (0);						\
-	ret;							\
-})
+#define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask)      \
+	amdgpu_device_wait_on_rreg(adev, inst,                       \
+	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)), \
+	#reg, expected_value, mask)
+
+#define SOC15_WAIT_ON_RREG_OFFSET(ip, inst, reg, offset, expected_value, mask)  \
+	amdgpu_device_wait_on_rreg(adev, inst,                                  \
+	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg) + (offset)), \
+	#reg, expected_value, mask)
 
 #define WREG32_RLC(reg, value) \
 	__WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_RLC, GC_HWIP)
-- 
GitLab


From 50b8b62ea496c615fb08ccc8a6fdf1891af75d7e Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 24 Apr 2022 15:38:16 +0800
Subject: [PATCH 1024/1791] drm/amdgpu: add vcn v4_0_3 ip headers

Add vcn v4_0_3 register offset adn shift masks
header files

v2: update headers (Alex)

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../include/asic_reg/vcn/vcn_4_0_3_offset.h   |  2254 ++++
 .../include/asic_reg/vcn/vcn_4_0_3_sh_mask.h  | 10424 ++++++++++++++++
 2 files changed, 12678 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
 create mode 100644 drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h

diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
new file mode 100644
index 0000000000000..facad93cd06f7
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
@@ -0,0 +1,2254 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef _vcn_4_0_3_OFFSET_HEADER
+#define _vcn_4_0_3_OFFSET_HEADER
+
+
+
+// addressBlock: aid_uvd0_uvddec
+// base address: 0x1fb00
+#define regUVD_TOP_CTRL                                                                                 0x00c0
+#define regUVD_TOP_CTRL_BASE_IDX                                                                        1
+#define regUVD_CGC_GATE                                                                                 0x00c1
+#define regUVD_CGC_GATE_BASE_IDX                                                                        1
+#define regUVD_CGC_CTRL                                                                                 0x00c2
+#define regUVD_CGC_CTRL_BASE_IDX                                                                        1
+#define regAVM_SUVD_CGC_GATE                                                                            0x00c4
+#define regAVM_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regCDEFE_SUVD_CGC_GATE                                                                          0x00c4
+#define regCDEFE_SUVD_CGC_GATE_BASE_IDX                                                                 1
+#define regEFC_SUVD_CGC_GATE                                                                            0x00c4
+#define regEFC_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regENT_SUVD_CGC_GATE                                                                            0x00c4
+#define regENT_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regIME_SUVD_CGC_GATE                                                                            0x00c4
+#define regIME_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regPPU_SUVD_CGC_GATE                                                                            0x00c4
+#define regPPU_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regSAOE_SUVD_CGC_GATE                                                                           0x00c4
+#define regSAOE_SUVD_CGC_GATE_BASE_IDX                                                                  1
+#define regSCM_SUVD_CGC_GATE                                                                            0x00c4
+#define regSCM_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regSDB_SUVD_CGC_GATE                                                                            0x00c4
+#define regSDB_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regSIT0_NXT_SUVD_CGC_GATE                                                                       0x00c4
+#define regSIT0_NXT_SUVD_CGC_GATE_BASE_IDX                                                              1
+#define regSIT1_NXT_SUVD_CGC_GATE                                                                       0x00c4
+#define regSIT1_NXT_SUVD_CGC_GATE_BASE_IDX                                                              1
+#define regSIT2_NXT_SUVD_CGC_GATE                                                                       0x00c4
+#define regSIT2_NXT_SUVD_CGC_GATE_BASE_IDX                                                              1
+#define regSIT_SUVD_CGC_GATE                                                                            0x00c4
+#define regSIT_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regSMPA_SUVD_CGC_GATE                                                                           0x00c4
+#define regSMPA_SUVD_CGC_GATE_BASE_IDX                                                                  1
+#define regSMP_SUVD_CGC_GATE                                                                            0x00c4
+#define regSMP_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regSRE_SUVD_CGC_GATE                                                                            0x00c4
+#define regSRE_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regUVD_MPBE0_SUVD_CGC_GATE                                                                      0x00c4
+#define regUVD_MPBE0_SUVD_CGC_GATE_BASE_IDX                                                             1
+#define regUVD_MPBE1_SUVD_CGC_GATE                                                                      0x00c4
+#define regUVD_MPBE1_SUVD_CGC_GATE_BASE_IDX                                                             1
+#define regUVD_SUVD_CGC_GATE                                                                            0x00c4
+#define regUVD_SUVD_CGC_GATE_BASE_IDX                                                                   1
+#define regAVM_SUVD_CGC_GATE2                                                                           0x00c5
+#define regAVM_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regCDEFE_SUVD_CGC_GATE2                                                                         0x00c5
+#define regCDEFE_SUVD_CGC_GATE2_BASE_IDX                                                                1
+#define regDBR_SUVD_CGC_GATE2                                                                           0x00c5
+#define regDBR_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regENT_SUVD_CGC_GATE2                                                                           0x00c5
+#define regENT_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regIME_SUVD_CGC_GATE2                                                                           0x00c5
+#define regIME_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regMPC1_SUVD_CGC_GATE2                                                                          0x00c5
+#define regMPC1_SUVD_CGC_GATE2_BASE_IDX                                                                 1
+#define regSAOE_SUVD_CGC_GATE2                                                                          0x00c5
+#define regSAOE_SUVD_CGC_GATE2_BASE_IDX                                                                 1
+#define regSDB_SUVD_CGC_GATE2                                                                           0x00c5
+#define regSDB_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regSIT0_NXT_SUVD_CGC_GATE2                                                                      0x00c5
+#define regSIT0_NXT_SUVD_CGC_GATE2_BASE_IDX                                                             1
+#define regSIT1_NXT_SUVD_CGC_GATE2                                                                      0x00c5
+#define regSIT1_NXT_SUVD_CGC_GATE2_BASE_IDX                                                             1
+#define regSIT2_NXT_SUVD_CGC_GATE2                                                                      0x00c5
+#define regSIT2_NXT_SUVD_CGC_GATE2_BASE_IDX                                                             1
+#define regSIT_SUVD_CGC_GATE2                                                                           0x00c5
+#define regSIT_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regSMPA_SUVD_CGC_GATE2                                                                          0x00c5
+#define regSMPA_SUVD_CGC_GATE2_BASE_IDX                                                                 1
+#define regSMP_SUVD_CGC_GATE2                                                                           0x00c5
+#define regSMP_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regSRE_SUVD_CGC_GATE2                                                                           0x00c5
+#define regSRE_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regUVD_MPBE0_SUVD_CGC_GATE2                                                                     0x00c5
+#define regUVD_MPBE0_SUVD_CGC_GATE2_BASE_IDX                                                            1
+#define regUVD_MPBE1_SUVD_CGC_GATE2                                                                     0x00c5
+#define regUVD_MPBE1_SUVD_CGC_GATE2_BASE_IDX                                                            1
+#define regUVD_SUVD_CGC_GATE2                                                                           0x00c5
+#define regUVD_SUVD_CGC_GATE2_BASE_IDX                                                                  1
+#define regAVM_SUVD_CGC_CTRL                                                                            0x00c6
+#define regAVM_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regCDEFE_SUVD_CGC_CTRL                                                                          0x00c6
+#define regCDEFE_SUVD_CGC_CTRL_BASE_IDX                                                                 1
+#define regDBR_SUVD_CGC_CTRL                                                                            0x00c6
+#define regDBR_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regEFC_SUVD_CGC_CTRL                                                                            0x00c6
+#define regEFC_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regENT_SUVD_CGC_CTRL                                                                            0x00c6
+#define regENT_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regIME_SUVD_CGC_CTRL                                                                            0x00c6
+#define regIME_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regMPC1_SUVD_CGC_CTRL                                                                           0x00c6
+#define regMPC1_SUVD_CGC_CTRL_BASE_IDX                                                                  1
+#define regPPU_SUVD_CGC_CTRL                                                                            0x00c6
+#define regPPU_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regSAOE_SUVD_CGC_CTRL                                                                           0x00c6
+#define regSAOE_SUVD_CGC_CTRL_BASE_IDX                                                                  1
+#define regSCM_SUVD_CGC_CTRL                                                                            0x00c6
+#define regSCM_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regSDB_SUVD_CGC_CTRL                                                                            0x00c6
+#define regSDB_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regSIT0_NXT_SUVD_CGC_CTRL                                                                       0x00c6
+#define regSIT0_NXT_SUVD_CGC_CTRL_BASE_IDX                                                              1
+#define regSIT1_NXT_SUVD_CGC_CTRL                                                                       0x00c6
+#define regSIT1_NXT_SUVD_CGC_CTRL_BASE_IDX                                                              1
+#define regSIT2_NXT_SUVD_CGC_CTRL                                                                       0x00c6
+#define regSIT2_NXT_SUVD_CGC_CTRL_BASE_IDX                                                              1
+#define regSIT_SUVD_CGC_CTRL                                                                            0x00c6
+#define regSIT_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regSMPA_SUVD_CGC_CTRL                                                                           0x00c6
+#define regSMPA_SUVD_CGC_CTRL_BASE_IDX                                                                  1
+#define regSMP_SUVD_CGC_CTRL                                                                            0x00c6
+#define regSMP_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regSRE_SUVD_CGC_CTRL                                                                            0x00c6
+#define regSRE_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regUVD_MPBE0_SUVD_CGC_CTRL                                                                      0x00c6
+#define regUVD_MPBE0_SUVD_CGC_CTRL_BASE_IDX                                                             1
+#define regUVD_MPBE1_SUVD_CGC_CTRL                                                                      0x00c6
+#define regUVD_MPBE1_SUVD_CGC_CTRL_BASE_IDX                                                             1
+#define regUVD_SUVD_CGC_CTRL                                                                            0x00c6
+#define regUVD_SUVD_CGC_CTRL_BASE_IDX                                                                   1
+#define regUVD_CGC_CTRL3                                                                                0x00ca
+#define regUVD_CGC_CTRL3_BASE_IDX                                                                       1
+#define regUVD_GPCOM_VCPU_DATA0                                                                         0x00d0
+#define regUVD_GPCOM_VCPU_DATA0_BASE_IDX                                                                1
+#define regUVD_GPCOM_VCPU_DATA1                                                                         0x00d1
+#define regUVD_GPCOM_VCPU_DATA1_BASE_IDX                                                                1
+#define regUVD_GPCOM_SYS_CMD                                                                            0x00d2
+#define regUVD_GPCOM_SYS_CMD_BASE_IDX                                                                   1
+#define regUVD_GPCOM_SYS_DATA0                                                                          0x00d3
+#define regUVD_GPCOM_SYS_DATA0_BASE_IDX                                                                 1
+#define regUVD_GPCOM_SYS_DATA1                                                                          0x00d4
+#define regUVD_GPCOM_SYS_DATA1_BASE_IDX                                                                 1
+#define regUVD_VCPU_INT_EN                                                                              0x00d5
+#define regUVD_VCPU_INT_EN_BASE_IDX                                                                     1
+#define regUVD_VCPU_INT_STATUS                                                                          0x00d6
+#define regUVD_VCPU_INT_STATUS_BASE_IDX                                                                 1
+#define regUVD_VCPU_INT_ACK                                                                             0x00d7
+#define regUVD_VCPU_INT_ACK_BASE_IDX                                                                    1
+#define regUVD_VCPU_INT_ROUTE                                                                           0x00d8
+#define regUVD_VCPU_INT_ROUTE_BASE_IDX                                                                  1
+#define regUVD_DRV_FW_MSG                                                                               0x00d9
+#define regUVD_DRV_FW_MSG_BASE_IDX                                                                      1
+#define regUVD_FW_DRV_MSG_ACK                                                                           0x00da
+#define regUVD_FW_DRV_MSG_ACK_BASE_IDX                                                                  1
+#define regUVD_SUVD_INT_EN                                                                              0x00db
+#define regUVD_SUVD_INT_EN_BASE_IDX                                                                     1
+#define regUVD_SUVD_INT_STATUS                                                                          0x00dc
+#define regUVD_SUVD_INT_STATUS_BASE_IDX                                                                 1
+#define regUVD_SUVD_INT_ACK                                                                             0x00dd
+#define regUVD_SUVD_INT_ACK_BASE_IDX                                                                    1
+#define regUVD_ENC_VCPU_INT_EN                                                                          0x00de
+#define regUVD_ENC_VCPU_INT_EN_BASE_IDX                                                                 1
+#define regUVD_ENC_VCPU_INT_STATUS                                                                      0x00df
+#define regUVD_ENC_VCPU_INT_STATUS_BASE_IDX                                                             1
+#define regUVD_ENC_VCPU_INT_ACK                                                                         0x00e0
+#define regUVD_ENC_VCPU_INT_ACK_BASE_IDX                                                                1
+#define regUVD_MASTINT_EN                                                                               0x00e1
+#define regUVD_MASTINT_EN_BASE_IDX                                                                      1
+#define regUVD_SYS_INT_EN                                                                               0x00e2
+#define regUVD_SYS_INT_EN_BASE_IDX                                                                      1
+#define regUVD_SYS_INT_STATUS                                                                           0x00e3
+#define regUVD_SYS_INT_STATUS_BASE_IDX                                                                  1
+#define regUVD_SYS_INT_ACK                                                                              0x00e4
+#define regUVD_SYS_INT_ACK_BASE_IDX                                                                     1
+#define regUVD_JOB_DONE                                                                                 0x00e5
+#define regUVD_JOB_DONE_BASE_IDX                                                                        1
+#define regUVD_CBUF_ID                                                                                  0x00e6
+#define regUVD_CBUF_ID_BASE_IDX                                                                         1
+#define regUVD_CONTEXT_ID                                                                               0x00e7
+#define regUVD_CONTEXT_ID_BASE_IDX                                                                      1
+#define regUVD_CONTEXT_ID2                                                                              0x00e8
+#define regUVD_CONTEXT_ID2_BASE_IDX                                                                     1
+#define regUVD_NO_OP                                                                                    0x00e9
+#define regUVD_NO_OP_BASE_IDX                                                                           1
+#define regUVD_RB_BASE_LO                                                                               0x00ea
+#define regUVD_RB_BASE_LO_BASE_IDX                                                                      1
+#define regUVD_RB_BASE_HI                                                                               0x00eb
+#define regUVD_RB_BASE_HI_BASE_IDX                                                                      1
+#define regUVD_RB_SIZE                                                                                  0x00ec
+#define regUVD_RB_SIZE_BASE_IDX                                                                         1
+#define regUVD_RB_BASE_LO2                                                                              0x00ef
+#define regUVD_RB_BASE_LO2_BASE_IDX                                                                     1
+#define regUVD_RB_BASE_HI2                                                                              0x00f0
+#define regUVD_RB_BASE_HI2_BASE_IDX                                                                     1
+#define regUVD_RB_SIZE2                                                                                 0x00f1
+#define regUVD_RB_SIZE2_BASE_IDX                                                                        1
+#define regUVD_RB_BASE_LO3                                                                              0x00f4
+#define regUVD_RB_BASE_LO3_BASE_IDX                                                                     1
+#define regUVD_RB_BASE_HI3                                                                              0x00f5
+#define regUVD_RB_BASE_HI3_BASE_IDX                                                                     1
+#define regUVD_RB_SIZE3                                                                                 0x00f6
+#define regUVD_RB_SIZE3_BASE_IDX                                                                        1
+#define regUVD_RB_BASE_LO4                                                                              0x00f9
+#define regUVD_RB_BASE_LO4_BASE_IDX                                                                     1
+#define regUVD_RB_BASE_HI4                                                                              0x00fa
+#define regUVD_RB_BASE_HI4_BASE_IDX                                                                     1
+#define regUVD_RB_SIZE4                                                                                 0x00fb
+#define regUVD_RB_SIZE4_BASE_IDX                                                                        1
+#define regUVD_OUT_RB_BASE_LO                                                                           0x00fe
+#define regUVD_OUT_RB_BASE_LO_BASE_IDX                                                                  1
+#define regUVD_OUT_RB_BASE_HI                                                                           0x00ff
+#define regUVD_OUT_RB_BASE_HI_BASE_IDX                                                                  1
+#define regUVD_OUT_RB_SIZE                                                                              0x0100
+#define regUVD_OUT_RB_SIZE_BASE_IDX                                                                     1
+#define regUVD_IOV_ACTIVE_FCN_ID                                                                        0x0103
+#define regUVD_IOV_ACTIVE_FCN_ID_BASE_IDX                                                               1
+#define regUVD_IOV_MAILBOX                                                                              0x0104
+#define regUVD_IOV_MAILBOX_BASE_IDX                                                                     1
+#define regUVD_IOV_MAILBOX_RESP                                                                         0x0105
+#define regUVD_IOV_MAILBOX_RESP_BASE_IDX                                                                1
+#define regUVD_RB_ARB_CTRL                                                                              0x0106
+#define regUVD_RB_ARB_CTRL_BASE_IDX                                                                     1
+#define regUVD_CTX_INDEX                                                                                0x0107
+#define regUVD_CTX_INDEX_BASE_IDX                                                                       1
+#define regUVD_CTX_DATA                                                                                 0x0108
+#define regUVD_CTX_DATA_BASE_IDX                                                                        1
+#define regUVD_CXW_WR                                                                                   0x0109
+#define regUVD_CXW_WR_BASE_IDX                                                                          1
+#define regUVD_CXW_WR_INT_ID                                                                            0x010a
+#define regUVD_CXW_WR_INT_ID_BASE_IDX                                                                   1
+#define regUVD_CXW_WR_INT_CTX_ID                                                                        0x010b
+#define regUVD_CXW_WR_INT_CTX_ID_BASE_IDX                                                               1
+#define regUVD_CXW_INT_ID                                                                               0x010c
+#define regUVD_CXW_INT_ID_BASE_IDX                                                                      1
+#define regUVD_MPEG2_ERROR                                                                              0x010d
+#define regUVD_MPEG2_ERROR_BASE_IDX                                                                     1
+#define regUVD_YBASE                                                                                    0x0110
+#define regUVD_YBASE_BASE_IDX                                                                           1
+#define regUVD_UVBASE                                                                                   0x0111
+#define regUVD_UVBASE_BASE_IDX                                                                          1
+#define regUVD_PITCH                                                                                    0x0112
+#define regUVD_PITCH_BASE_IDX                                                                           1
+#define regUVD_WIDTH                                                                                    0x0113
+#define regUVD_WIDTH_BASE_IDX                                                                           1
+#define regUVD_HEIGHT                                                                                   0x0114
+#define regUVD_HEIGHT_BASE_IDX                                                                          1
+#define regUVD_PICCOUNT                                                                                 0x0115
+#define regUVD_PICCOUNT_BASE_IDX                                                                        1
+#define regUVD_MPRD_INITIAL_XY                                                                          0x0116
+#define regUVD_MPRD_INITIAL_XY_BASE_IDX                                                                 1
+#define regUVD_MPEG2_CTRL                                                                               0x0117
+#define regUVD_MPEG2_CTRL_BASE_IDX                                                                      1
+#define regUVD_MB_CTL_BUF_BASE                                                                          0x0118
+#define regUVD_MB_CTL_BUF_BASE_BASE_IDX                                                                 1
+#define regUVD_PIC_CTL_BUF_BASE                                                                         0x0119
+#define regUVD_PIC_CTL_BUF_BASE_BASE_IDX                                                                1
+#define regUVD_DXVA_BUF_SIZE                                                                            0x011a
+#define regUVD_DXVA_BUF_SIZE_BASE_IDX                                                                   1
+#define regUVD_SCRATCH_NP                                                                               0x011b
+#define regUVD_SCRATCH_NP_BASE_IDX                                                                      1
+#define regUVD_CLK_SWT_HANDSHAKE                                                                        0x011c
+#define regUVD_CLK_SWT_HANDSHAKE_BASE_IDX                                                               1
+#define regUVD_GP_SCRATCH0                                                                              0x011e
+#define regUVD_GP_SCRATCH0_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH1                                                                              0x011f
+#define regUVD_GP_SCRATCH1_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH2                                                                              0x0120
+#define regUVD_GP_SCRATCH2_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH3                                                                              0x0121
+#define regUVD_GP_SCRATCH3_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH4                                                                              0x0122
+#define regUVD_GP_SCRATCH4_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH5                                                                              0x0123
+#define regUVD_GP_SCRATCH5_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH6                                                                              0x0124
+#define regUVD_GP_SCRATCH6_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH7                                                                              0x0125
+#define regUVD_GP_SCRATCH7_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH8                                                                              0x0126
+#define regUVD_GP_SCRATCH8_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH9                                                                              0x0127
+#define regUVD_GP_SCRATCH9_BASE_IDX                                                                     1
+#define regUVD_GP_SCRATCH10                                                                             0x0128
+#define regUVD_GP_SCRATCH10_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH11                                                                             0x0129
+#define regUVD_GP_SCRATCH11_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH12                                                                             0x012a
+#define regUVD_GP_SCRATCH12_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH13                                                                             0x012b
+#define regUVD_GP_SCRATCH13_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH14                                                                             0x012c
+#define regUVD_GP_SCRATCH14_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH15                                                                             0x012d
+#define regUVD_GP_SCRATCH15_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH16                                                                             0x012e
+#define regUVD_GP_SCRATCH16_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH17                                                                             0x012f
+#define regUVD_GP_SCRATCH17_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH18                                                                             0x0130
+#define regUVD_GP_SCRATCH18_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH19                                                                             0x0131
+#define regUVD_GP_SCRATCH19_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH20                                                                             0x0132
+#define regUVD_GP_SCRATCH20_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH21                                                                             0x0133
+#define regUVD_GP_SCRATCH21_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH22                                                                             0x0134
+#define regUVD_GP_SCRATCH22_BASE_IDX                                                                    1
+#define regUVD_GP_SCRATCH23                                                                             0x0135
+#define regUVD_GP_SCRATCH23_BASE_IDX                                                                    1
+#define regUVD_AUDIO_RB_BASE_LO                                                                         0x0136
+#define regUVD_AUDIO_RB_BASE_LO_BASE_IDX                                                                1
+#define regUVD_AUDIO_RB_BASE_HI                                                                         0x0137
+#define regUVD_AUDIO_RB_BASE_HI_BASE_IDX                                                                1
+#define regUVD_AUDIO_RB_SIZE                                                                            0x0138
+#define regUVD_AUDIO_RB_SIZE_BASE_IDX                                                                   1
+#define regUVD_VCPU_INT_STATUS2                                                                         0x013b
+#define regUVD_VCPU_INT_STATUS2_BASE_IDX                                                                1
+#define regUVD_VCPU_INT_ACK2                                                                            0x013c
+#define regUVD_VCPU_INT_ACK2_BASE_IDX                                                                   1
+#define regUVD_VCPU_INT_EN2                                                                             0x013d
+#define regUVD_VCPU_INT_EN2_BASE_IDX                                                                    1
+#define regUVD_SUVD_CGC_STATUS2                                                                         0x013e
+#define regUVD_SUVD_CGC_STATUS2_BASE_IDX                                                                1
+#define regUVD_SUVD_INT_STATUS2                                                                         0x0140
+#define regUVD_SUVD_INT_STATUS2_BASE_IDX                                                                1
+#define regUVD_SUVD_INT_EN2                                                                             0x0141
+#define regUVD_SUVD_INT_EN2_BASE_IDX                                                                    1
+#define regUVD_SUVD_INT_ACK2                                                                            0x0142
+#define regUVD_SUVD_INT_ACK2_BASE_IDX                                                                   1
+#define regUVD_STATUS                                                                                   0x0143
+#define regUVD_STATUS_BASE_IDX                                                                          1
+#define regUVD_ENC_PIPE_BUSY                                                                            0x0144
+#define regUVD_ENC_PIPE_BUSY_BASE_IDX                                                                   1
+#define regUVD_FW_POWER_STATUS                                                                          0x0145
+#define regUVD_FW_POWER_STATUS_BASE_IDX                                                                 1
+#define regUVD_CNTL                                                                                     0x0146
+#define regUVD_CNTL_BASE_IDX                                                                            1
+#define regUVD_SOFT_RESET                                                                               0x0147
+#define regUVD_SOFT_RESET_BASE_IDX                                                                      1
+#define regUVD_SOFT_RESET2                                                                              0x0148
+#define regUVD_SOFT_RESET2_BASE_IDX                                                                     1
+#define regUVD_MMSCH_SOFT_RESET                                                                         0x0149
+#define regUVD_MMSCH_SOFT_RESET_BASE_IDX                                                                1
+#define regUVD_WIG_CTRL                                                                                 0x014a
+#define regUVD_WIG_CTRL_BASE_IDX                                                                        1
+#define regUVD_CGC_STATUS                                                                               0x014c
+#define regUVD_CGC_STATUS_BASE_IDX                                                                      1
+#define regUVD_CGC_UDEC_STATUS                                                                          0x014e
+#define regUVD_CGC_UDEC_STATUS_BASE_IDX                                                                 1
+#define regUVD_SUVD_CGC_STATUS                                                                          0x0150
+#define regUVD_SUVD_CGC_STATUS_BASE_IDX                                                                 1
+#define regUVD_GPCOM_VCPU_CMD                                                                           0x0152
+#define regUVD_GPCOM_VCPU_CMD_BASE_IDX                                                                  1
+
+
+// addressBlock: aid_uvd0_ecpudec
+// base address: 0x1fe00
+#define regUVD_VCPU_CACHE_OFFSET0                                                                       0x0180
+#define regUVD_VCPU_CACHE_OFFSET0_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE0                                                                         0x0181
+#define regUVD_VCPU_CACHE_SIZE0_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET1                                                                       0x0182
+#define regUVD_VCPU_CACHE_OFFSET1_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE1                                                                         0x0183
+#define regUVD_VCPU_CACHE_SIZE1_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET2                                                                       0x0184
+#define regUVD_VCPU_CACHE_OFFSET2_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE2                                                                         0x0185
+#define regUVD_VCPU_CACHE_SIZE2_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET3                                                                       0x0186
+#define regUVD_VCPU_CACHE_OFFSET3_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE3                                                                         0x0187
+#define regUVD_VCPU_CACHE_SIZE3_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET4                                                                       0x0188
+#define regUVD_VCPU_CACHE_OFFSET4_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE4                                                                         0x0189
+#define regUVD_VCPU_CACHE_SIZE4_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET5                                                                       0x018a
+#define regUVD_VCPU_CACHE_OFFSET5_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE5                                                                         0x018b
+#define regUVD_VCPU_CACHE_SIZE5_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET6                                                                       0x018c
+#define regUVD_VCPU_CACHE_OFFSET6_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE6                                                                         0x018d
+#define regUVD_VCPU_CACHE_SIZE6_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET7                                                                       0x018e
+#define regUVD_VCPU_CACHE_OFFSET7_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE7                                                                         0x018f
+#define regUVD_VCPU_CACHE_SIZE7_BASE_IDX                                                                1
+#define regUVD_VCPU_CACHE_OFFSET8                                                                       0x0190
+#define regUVD_VCPU_CACHE_OFFSET8_BASE_IDX                                                              1
+#define regUVD_VCPU_CACHE_SIZE8                                                                         0x0191
+#define regUVD_VCPU_CACHE_SIZE8_BASE_IDX                                                                1
+#define regUVD_VCPU_NONCACHE_OFFSET0                                                                    0x0192
+#define regUVD_VCPU_NONCACHE_OFFSET0_BASE_IDX                                                           1
+#define regUVD_VCPU_NONCACHE_SIZE0                                                                      0x0193
+#define regUVD_VCPU_NONCACHE_SIZE0_BASE_IDX                                                             1
+#define regUVD_VCPU_NONCACHE_OFFSET1                                                                    0x0194
+#define regUVD_VCPU_NONCACHE_OFFSET1_BASE_IDX                                                           1
+#define regUVD_VCPU_NONCACHE_SIZE1                                                                      0x0195
+#define regUVD_VCPU_NONCACHE_SIZE1_BASE_IDX                                                             1
+#define regUVD_VCPU_CNTL                                                                                0x0196
+#define regUVD_VCPU_CNTL_BASE_IDX                                                                       1
+#define regUVD_VCPU_PRID                                                                                0x0197
+#define regUVD_VCPU_PRID_BASE_IDX                                                                       1
+#define regUVD_VCPU_TRCE                                                                                0x0198
+#define regUVD_VCPU_TRCE_BASE_IDX                                                                       1
+#define regUVD_VCPU_TRCE_RD                                                                             0x0199
+#define regUVD_VCPU_TRCE_RD_BASE_IDX                                                                    1
+#define regUVD_VCPU_IND_INDEX                                                                           0x019b
+#define regUVD_VCPU_IND_INDEX_BASE_IDX                                                                  1
+#define regUVD_VCPU_IND_DATA                                                                            0x019c
+#define regUVD_VCPU_IND_DATA_BASE_IDX                                                                   1
+
+
+// addressBlock: aid_uvd0_uvd_mpcdec
+// base address: 0x1ff30
+#define regUVD_MP_SWAP_CNTL                                                                             0x01cc
+#define regUVD_MP_SWAP_CNTL_BASE_IDX                                                                    1
+#define regUVD_MP_SWAP_CNTL2                                                                            0x01cd
+#define regUVD_MP_SWAP_CNTL2_BASE_IDX                                                                   1
+#define regUVD_MPC_LUMA_SRCH                                                                            0x01ce
+#define regUVD_MPC_LUMA_SRCH_BASE_IDX                                                                   1
+#define regUVD_MPC_LUMA_HIT                                                                             0x01cf
+#define regUVD_MPC_LUMA_HIT_BASE_IDX                                                                    1
+#define regUVD_MPC_LUMA_HITPEND                                                                         0x01d0
+#define regUVD_MPC_LUMA_HITPEND_BASE_IDX                                                                1
+#define regUVD_MPC_CHROMA_SRCH                                                                          0x01d1
+#define regUVD_MPC_CHROMA_SRCH_BASE_IDX                                                                 1
+#define regUVD_MPC_CHROMA_HIT                                                                           0x01d2
+#define regUVD_MPC_CHROMA_HIT_BASE_IDX                                                                  1
+#define regUVD_MPC_CHROMA_HITPEND                                                                       0x01d3
+#define regUVD_MPC_CHROMA_HITPEND_BASE_IDX                                                              1
+#define regUVD_MPC_CNTL                                                                                 0x01d4
+#define regUVD_MPC_CNTL_BASE_IDX                                                                        1
+#define regUVD_MPC_PITCH                                                                                0x01d5
+#define regUVD_MPC_PITCH_BASE_IDX                                                                       1
+#define regUVD_MPC_SET_MUXA0                                                                            0x01d6
+#define regUVD_MPC_SET_MUXA0_BASE_IDX                                                                   1
+#define regUVD_MPC_SET_MUXA1                                                                            0x01d7
+#define regUVD_MPC_SET_MUXA1_BASE_IDX                                                                   1
+#define regUVD_MPC_SET_MUXB0                                                                            0x01d8
+#define regUVD_MPC_SET_MUXB0_BASE_IDX                                                                   1
+#define regUVD_MPC_SET_MUXB1                                                                            0x01d9
+#define regUVD_MPC_SET_MUXB1_BASE_IDX                                                                   1
+#define regUVD_MPC_SET_MUX                                                                              0x01da
+#define regUVD_MPC_SET_MUX_BASE_IDX                                                                     1
+#define regUVD_MPC_SET_ALU                                                                              0x01db
+#define regUVD_MPC_SET_ALU_BASE_IDX                                                                     1
+#define regUVD_MPC_PERF0                                                                                0x01dc
+#define regUVD_MPC_PERF0_BASE_IDX                                                                       1
+#define regUVD_MPC_PERF1                                                                                0x01dd
+#define regUVD_MPC_PERF1_BASE_IDX                                                                       1
+#define regUVD_MPC_IND_INDEX                                                                            0x01de
+#define regUVD_MPC_IND_INDEX_BASE_IDX                                                                   1
+#define regUVD_MPC_IND_DATA                                                                             0x01df
+#define regUVD_MPC_IND_DATA_BASE_IDX                                                                    1
+
+
+// addressBlock: aid_uvd0_uvd_rbcdec
+// base address: 0x1ff90
+#define regUVD_RBC_IB_SIZE                                                                              0x01e4
+#define regUVD_RBC_IB_SIZE_BASE_IDX                                                                     1
+#define regUVD_RBC_IB_SIZE_UPDATE                                                                       0x01e5
+#define regUVD_RBC_IB_SIZE_UPDATE_BASE_IDX                                                              1
+#define regUVD_RBC_RB_CNTL                                                                              0x01e6
+#define regUVD_RBC_RB_CNTL_BASE_IDX                                                                     1
+#define regUVD_RBC_RB_RPTR_ADDR                                                                         0x01e7
+#define regUVD_RBC_RB_RPTR_ADDR_BASE_IDX                                                                1
+#define regUVD_RBC_VCPU_ACCESS                                                                          0x01ea
+#define regUVD_RBC_VCPU_ACCESS_BASE_IDX                                                                 1
+#define regUVD_FW_SEMAPHORE_CNTL                                                                        0x01eb
+#define regUVD_FW_SEMAPHORE_CNTL_BASE_IDX                                                               1
+#define regUVD_RBC_READ_REQ_URGENT_CNTL                                                                 0x01ed
+#define regUVD_RBC_READ_REQ_URGENT_CNTL_BASE_IDX                                                        1
+#define regUVD_RBC_RB_WPTR_CNTL                                                                         0x01ee
+#define regUVD_RBC_RB_WPTR_CNTL_BASE_IDX                                                                1
+#define regUVD_RBC_WPTR_STATUS                                                                          0x01ef
+#define regUVD_RBC_WPTR_STATUS_BASE_IDX                                                                 1
+#define regUVD_RBC_WPTR_POLL_CNTL                                                                       0x01f0
+#define regUVD_RBC_WPTR_POLL_CNTL_BASE_IDX                                                              1
+#define regUVD_RBC_WPTR_POLL_ADDR                                                                       0x01f1
+#define regUVD_RBC_WPTR_POLL_ADDR_BASE_IDX                                                              1
+#define regUVD_SEMA_CMD                                                                                 0x01f2
+#define regUVD_SEMA_CMD_BASE_IDX                                                                        1
+#define regUVD_SEMA_ADDR_LOW                                                                            0x01f3
+#define regUVD_SEMA_ADDR_LOW_BASE_IDX                                                                   1
+#define regUVD_SEMA_ADDR_HIGH                                                                           0x01f4
+#define regUVD_SEMA_ADDR_HIGH_BASE_IDX                                                                  1
+#define regUVD_ENGINE_CNTL                                                                              0x01f5
+#define regUVD_ENGINE_CNTL_BASE_IDX                                                                     1
+#define regUVD_SEMA_TIMEOUT_STATUS                                                                      0x01f6
+#define regUVD_SEMA_TIMEOUT_STATUS_BASE_IDX                                                             1
+#define regUVD_SEMA_CNTL                                                                                0x01f7
+#define regUVD_SEMA_CNTL_BASE_IDX                                                                       1
+#define regUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL                                                      0x01f8
+#define regUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL_BASE_IDX                                             1
+#define regUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL                                                             0x01f9
+#define regUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL_BASE_IDX                                                    1
+#define regUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL                                                        0x01fa
+#define regUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL_BASE_IDX                                               1
+#define regUVD_JOB_START                                                                                0x01fb
+#define regUVD_JOB_START_BASE_IDX                                                                       1
+#define regUVD_RBC_BUF_STATUS                                                                           0x01fc
+#define regUVD_RBC_BUF_STATUS_BASE_IDX                                                                  1
+#define regUVD_RBC_SWAP_CNTL                                                                            0x01fd
+#define regUVD_RBC_SWAP_CNTL_BASE_IDX                                                                   1
+
+
+// addressBlock: aid_uvd0_lmi_adpdec
+// base address: 0x20090
+#define regUVD_LMI_RE_64BIT_BAR_LOW                                                                     0x0224
+#define regUVD_LMI_RE_64BIT_BAR_LOW_BASE_IDX                                                            1
+#define regUVD_LMI_RE_64BIT_BAR_HIGH                                                                    0x0225
+#define regUVD_LMI_RE_64BIT_BAR_HIGH_BASE_IDX                                                           1
+#define regUVD_LMI_IT_64BIT_BAR_LOW                                                                     0x0226
+#define regUVD_LMI_IT_64BIT_BAR_LOW_BASE_IDX                                                            1
+#define regUVD_LMI_IT_64BIT_BAR_HIGH                                                                    0x0227
+#define regUVD_LMI_IT_64BIT_BAR_HIGH_BASE_IDX                                                           1
+#define regUVD_LMI_MP_64BIT_BAR_LOW                                                                     0x0228
+#define regUVD_LMI_MP_64BIT_BAR_LOW_BASE_IDX                                                            1
+#define regUVD_LMI_MP_64BIT_BAR_HIGH                                                                    0x0229
+#define regUVD_LMI_MP_64BIT_BAR_HIGH_BASE_IDX                                                           1
+#define regUVD_LMI_CM_64BIT_BAR_LOW                                                                     0x022a
+#define regUVD_LMI_CM_64BIT_BAR_LOW_BASE_IDX                                                            1
+#define regUVD_LMI_CM_64BIT_BAR_HIGH                                                                    0x022b
+#define regUVD_LMI_CM_64BIT_BAR_HIGH_BASE_IDX                                                           1
+#define regUVD_LMI_DB_64BIT_BAR_LOW                                                                     0x022c
+#define regUVD_LMI_DB_64BIT_BAR_LOW_BASE_IDX                                                            1
+#define regUVD_LMI_DB_64BIT_BAR_HIGH                                                                    0x022d
+#define regUVD_LMI_DB_64BIT_BAR_HIGH_BASE_IDX                                                           1
+#define regUVD_LMI_DBW_64BIT_BAR_LOW                                                                    0x022e
+#define regUVD_LMI_DBW_64BIT_BAR_LOW_BASE_IDX                                                           1
+#define regUVD_LMI_DBW_64BIT_BAR_HIGH                                                                   0x022f
+#define regUVD_LMI_DBW_64BIT_BAR_HIGH_BASE_IDX                                                          1
+#define regUVD_LMI_IDCT_64BIT_BAR_LOW                                                                   0x0230
+#define regUVD_LMI_IDCT_64BIT_BAR_LOW_BASE_IDX                                                          1
+#define regUVD_LMI_IDCT_64BIT_BAR_HIGH                                                                  0x0231
+#define regUVD_LMI_IDCT_64BIT_BAR_HIGH_BASE_IDX                                                         1
+#define regUVD_LMI_MPRD_S0_64BIT_BAR_LOW                                                                0x0232
+#define regUVD_LMI_MPRD_S0_64BIT_BAR_LOW_BASE_IDX                                                       1
+#define regUVD_LMI_MPRD_S0_64BIT_BAR_HIGH                                                               0x0233
+#define regUVD_LMI_MPRD_S0_64BIT_BAR_HIGH_BASE_IDX                                                      1
+#define regUVD_LMI_MPRD_S1_64BIT_BAR_LOW                                                                0x0234
+#define regUVD_LMI_MPRD_S1_64BIT_BAR_LOW_BASE_IDX                                                       1
+#define regUVD_LMI_MPRD_S1_64BIT_BAR_HIGH                                                               0x0235
+#define regUVD_LMI_MPRD_S1_64BIT_BAR_HIGH_BASE_IDX                                                      1
+#define regUVD_LMI_MPRD_DBW_64BIT_BAR_LOW                                                               0x0236
+#define regUVD_LMI_MPRD_DBW_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MPRD_DBW_64BIT_BAR_HIGH                                                              0x0237
+#define regUVD_LMI_MPRD_DBW_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MPC_64BIT_BAR_LOW                                                                    0x0238
+#define regUVD_LMI_MPC_64BIT_BAR_LOW_BASE_IDX                                                           1
+#define regUVD_LMI_MPC_64BIT_BAR_HIGH                                                                   0x0239
+#define regUVD_LMI_MPC_64BIT_BAR_HIGH_BASE_IDX                                                          1
+#define regUVD_LMI_RBC_RB_64BIT_BAR_LOW                                                                 0x023a
+#define regUVD_LMI_RBC_RB_64BIT_BAR_LOW_BASE_IDX                                                        1
+#define regUVD_LMI_RBC_RB_64BIT_BAR_HIGH                                                                0x023b
+#define regUVD_LMI_RBC_RB_64BIT_BAR_HIGH_BASE_IDX                                                       1
+#define regUVD_LMI_RBC_IB_64BIT_BAR_LOW                                                                 0x023c
+#define regUVD_LMI_RBC_IB_64BIT_BAR_LOW_BASE_IDX                                                        1
+#define regUVD_LMI_RBC_IB_64BIT_BAR_HIGH                                                                0x023d
+#define regUVD_LMI_RBC_IB_64BIT_BAR_HIGH_BASE_IDX                                                       1
+#define regUVD_LMI_LBSI_64BIT_BAR_LOW                                                                   0x023e
+#define regUVD_LMI_LBSI_64BIT_BAR_LOW_BASE_IDX                                                          1
+#define regUVD_LMI_LBSI_64BIT_BAR_HIGH                                                                  0x023f
+#define regUVD_LMI_LBSI_64BIT_BAR_HIGH_BASE_IDX                                                         1
+#define regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW                                                               0x0240
+#define regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH                                                              0x0241
+#define regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_VCPU_NC1_64BIT_BAR_LOW                                                               0x0242
+#define regUVD_LMI_VCPU_NC1_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_VCPU_NC1_64BIT_BAR_HIGH                                                              0x0243
+#define regUVD_LMI_VCPU_NC1_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW                                                             0x0244
+#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW_BASE_IDX                                                    1
+#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH                                                            0x0245
+#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH_BASE_IDX                                                   1
+#define regUVD_LMI_CENC_64BIT_BAR_LOW                                                                   0x0246
+#define regUVD_LMI_CENC_64BIT_BAR_LOW_BASE_IDX                                                          1
+#define regUVD_LMI_CENC_64BIT_BAR_HIGH                                                                  0x0247
+#define regUVD_LMI_CENC_64BIT_BAR_HIGH_BASE_IDX                                                         1
+#define regUVD_LMI_SRE_64BIT_BAR_LOW                                                                    0x0248
+#define regUVD_LMI_SRE_64BIT_BAR_LOW_BASE_IDX                                                           1
+#define regUVD_LMI_SRE_64BIT_BAR_HIGH                                                                   0x0249
+#define regUVD_LMI_SRE_64BIT_BAR_HIGH_BASE_IDX                                                          1
+#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_LOW                                                              0x024a
+#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH                                                             0x024b
+#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW                                                          0x024c
+#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW_BASE_IDX                                                 1
+#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH                                                         0x024d
+#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH_BASE_IDX                                                1
+#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW                                                        0x024e
+#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW_BASE_IDX                                               1
+#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH                                                       0x024f
+#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH_BASE_IDX                                              1
+#define regUVD_LMI_MIF_REF_64BIT_BAR_LOW                                                                0x0250
+#define regUVD_LMI_MIF_REF_64BIT_BAR_LOW_BASE_IDX                                                       1
+#define regUVD_LMI_MIF_REF_64BIT_BAR_HIGH                                                               0x0251
+#define regUVD_LMI_MIF_REF_64BIT_BAR_HIGH_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_DBW_64BIT_BAR_LOW                                                                0x0252
+#define regUVD_LMI_MIF_DBW_64BIT_BAR_LOW_BASE_IDX                                                       1
+#define regUVD_LMI_MIF_DBW_64BIT_BAR_HIGH                                                               0x0253
+#define regUVD_LMI_MIF_DBW_64BIT_BAR_HIGH_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW                                                           0x0254
+#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW_BASE_IDX                                                  1
+#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH                                                          0x0255
+#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH_BASE_IDX                                                 1
+#define regUVD_LMI_MIF_BSP0_64BIT_BAR_LOW                                                               0x0256
+#define regUVD_LMI_MIF_BSP0_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSP0_64BIT_BAR_HIGH                                                              0x0257
+#define regUVD_LMI_MIF_BSP0_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSP1_64BIT_BAR_LOW                                                               0x0258
+#define regUVD_LMI_MIF_BSP1_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSP1_64BIT_BAR_HIGH                                                              0x0259
+#define regUVD_LMI_MIF_BSP1_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSP2_64BIT_BAR_LOW                                                               0x025a
+#define regUVD_LMI_MIF_BSP2_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSP2_64BIT_BAR_HIGH                                                              0x025b
+#define regUVD_LMI_MIF_BSP2_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSP3_64BIT_BAR_LOW                                                               0x025c
+#define regUVD_LMI_MIF_BSP3_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSP3_64BIT_BAR_HIGH                                                              0x025d
+#define regUVD_LMI_MIF_BSP3_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSD0_64BIT_BAR_LOW                                                               0x025e
+#define regUVD_LMI_MIF_BSD0_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSD0_64BIT_BAR_HIGH                                                              0x025f
+#define regUVD_LMI_MIF_BSD0_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSD1_64BIT_BAR_LOW                                                               0x0260
+#define regUVD_LMI_MIF_BSD1_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSD1_64BIT_BAR_HIGH                                                              0x0261
+#define regUVD_LMI_MIF_BSD1_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSD2_64BIT_BAR_LOW                                                               0x0262
+#define regUVD_LMI_MIF_BSD2_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSD2_64BIT_BAR_HIGH                                                              0x0263
+#define regUVD_LMI_MIF_BSD2_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSD3_64BIT_BAR_LOW                                                               0x0264
+#define regUVD_LMI_MIF_BSD3_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSD3_64BIT_BAR_HIGH                                                              0x0265
+#define regUVD_LMI_MIF_BSD3_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_BSD4_64BIT_BAR_LOW                                                               0x0266
+#define regUVD_LMI_MIF_BSD4_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_BSD4_64BIT_BAR_HIGH                                                              0x0267
+#define regUVD_LMI_MIF_BSD4_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW                                                            0x0270
+#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH                                                           0x0271
+#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW                                                            0x0272
+#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH                                                           0x0273
+#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW                                                            0x0274
+#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH                                                           0x0275
+#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW                                                            0x0276
+#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH                                                           0x0277
+#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW                                                            0x0278
+#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH                                                           0x0279
+#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW                                                            0x027a
+#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH                                                           0x027b
+#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW                                                            0x027c
+#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH                                                           0x027d
+#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW                                                            0x027e
+#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW_BASE_IDX                                                   1
+#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH                                                           0x027f
+#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH_BASE_IDX                                                  1
+#define regUVD_LMI_MIF_SCLR_64BIT_BAR_LOW                                                               0x0280
+#define regUVD_LMI_MIF_SCLR_64BIT_BAR_LOW_BASE_IDX                                                      1
+#define regUVD_LMI_MIF_SCLR_64BIT_BAR_HIGH                                                              0x0281
+#define regUVD_LMI_MIF_SCLR_64BIT_BAR_HIGH_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_LOW                                                              0x0282
+#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH                                                             0x0283
+#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_SPH_64BIT_BAR_HIGH                                                                   0x0284
+#define regUVD_LMI_SPH_64BIT_BAR_HIGH_BASE_IDX                                                          1
+#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW                                                    0x0298
+#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW_BASE_IDX                                           1
+#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH                                                   0x0299
+#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH_BASE_IDX                                          1
+#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW                                                  0x029a
+#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW_BASE_IDX                                         1
+#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH                                                 0x029b
+#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH_BASE_IDX                                        1
+#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW                                                       0x029c
+#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW_BASE_IDX                                              1
+#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH                                                      0x029d
+#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH_BASE_IDX                                             1
+#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW                                                     0x029e
+#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW_BASE_IDX                                            1
+#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH                                                    0x029f
+#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH_BASE_IDX                                           1
+#define regUVD_ADP_ATOMIC_CONFIG                                                                        0x02a1
+#define regUVD_ADP_ATOMIC_CONFIG_BASE_IDX                                                               1
+#define regUVD_LMI_ARB_CTRL2                                                                            0x02a2
+#define regUVD_LMI_ARB_CTRL2_BASE_IDX                                                                   1
+#define regUVD_LMI_VCPU_CACHE_VMIDS_MULTI                                                               0x02a7
+#define regUVD_LMI_VCPU_CACHE_VMIDS_MULTI_BASE_IDX                                                      1
+#define regUVD_LMI_VCPU_NC_VMIDS_MULTI                                                                  0x02a8
+#define regUVD_LMI_VCPU_NC_VMIDS_MULTI_BASE_IDX                                                         1
+#define regUVD_LMI_LAT_CTRL                                                                             0x02a9
+#define regUVD_LMI_LAT_CTRL_BASE_IDX                                                                    1
+#define regUVD_LMI_LAT_CNTR                                                                             0x02aa
+#define regUVD_LMI_LAT_CNTR_BASE_IDX                                                                    1
+#define regUVD_LMI_AVG_LAT_CNTR                                                                         0x02ab
+#define regUVD_LMI_AVG_LAT_CNTR_BASE_IDX                                                                1
+#define regUVD_LMI_SPH                                                                                  0x02ac
+#define regUVD_LMI_SPH_BASE_IDX                                                                         1
+#define regUVD_LMI_VCPU_CACHE_VMID                                                                      0x02ad
+#define regUVD_LMI_VCPU_CACHE_VMID_BASE_IDX                                                             1
+#define regUVD_LMI_CTRL2                                                                                0x02ae
+#define regUVD_LMI_CTRL2_BASE_IDX                                                                       1
+#define regUVD_LMI_URGENT_CTRL                                                                          0x02af
+#define regUVD_LMI_URGENT_CTRL_BASE_IDX                                                                 1
+#define regUVD_LMI_CTRL                                                                                 0x02b0
+#define regUVD_LMI_CTRL_BASE_IDX                                                                        1
+#define regUVD_LMI_STATUS                                                                               0x02b1
+#define regUVD_LMI_STATUS_BASE_IDX                                                                      1
+#define regUVD_LMI_PERFMON_CTRL                                                                         0x02b4
+#define regUVD_LMI_PERFMON_CTRL_BASE_IDX                                                                1
+#define regUVD_LMI_PERFMON_COUNT_LO                                                                     0x02b5
+#define regUVD_LMI_PERFMON_COUNT_LO_BASE_IDX                                                            1
+#define regUVD_LMI_PERFMON_COUNT_HI                                                                     0x02b6
+#define regUVD_LMI_PERFMON_COUNT_HI_BASE_IDX                                                            1
+#define regUVD_LMI_ADP_SWAP_CNTL                                                                        0x02b7
+#define regUVD_LMI_ADP_SWAP_CNTL_BASE_IDX                                                               1
+#define regUVD_LMI_RBC_RB_VMID                                                                          0x02b8
+#define regUVD_LMI_RBC_RB_VMID_BASE_IDX                                                                 1
+#define regUVD_LMI_RBC_IB_VMID                                                                          0x02b9
+#define regUVD_LMI_RBC_IB_VMID_BASE_IDX                                                                 1
+#define regUVD_LMI_MC_CREDITS                                                                           0x02ba
+#define regUVD_LMI_MC_CREDITS_BASE_IDX                                                                  1
+#define regUVD_LMI_ADP_IND_INDEX                                                                        0x02be
+#define regUVD_LMI_ADP_IND_INDEX_BASE_IDX                                                               1
+#define regUVD_LMI_ADP_IND_DATA                                                                         0x02bf
+#define regUVD_LMI_ADP_IND_DATA_BASE_IDX                                                                1
+#define regUVD_LMI_ADP_PF_EN                                                                            0x02c0
+#define regUVD_LMI_ADP_PF_EN_BASE_IDX                                                                   1
+#define regUVD_LMI_PREF_CTRL                                                                            0x02c2
+#define regUVD_LMI_PREF_CTRL_BASE_IDX                                                                   1
+#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW                                                           0x02dd
+#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW_BASE_IDX                                                  1
+#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH                                                          0x02de
+#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH_BASE_IDX                                                 1
+#define regVCN_RAS_CNTL                                                                                 0x02df
+#define regVCN_RAS_CNTL_BASE_IDX                                                                        1
+
+
+// addressBlock: aid_uvd0_uvd_jpeg0_jpegnpdec
+// base address: 0x20f00
+#define regUVD_JPEG_CNTL                                                                                0x05c0
+#define regUVD_JPEG_CNTL_BASE_IDX                                                                       1
+#define regUVD_JPEG_RB_BASE                                                                             0x05c1
+#define regUVD_JPEG_RB_BASE_BASE_IDX                                                                    1
+#define regUVD_JPEG_RB_WPTR                                                                             0x05c2
+#define regUVD_JPEG_RB_WPTR_BASE_IDX                                                                    1
+#define regUVD_JPEG_RB_RPTR                                                                             0x05c3
+#define regUVD_JPEG_RB_RPTR_BASE_IDX                                                                    1
+#define regUVD_JPEG_RB_SIZE                                                                             0x05c4
+#define regUVD_JPEG_RB_SIZE_BASE_IDX                                                                    1
+#define regUVD_JPEG_DEC_CNT                                                                             0x05c5
+#define regUVD_JPEG_DEC_CNT_BASE_IDX                                                                    1
+#define regUVD_JPEG_SPS_INFO                                                                            0x05c6
+#define regUVD_JPEG_SPS_INFO_BASE_IDX                                                                   1
+#define regUVD_JPEG_SPS1_INFO                                                                           0x05c7
+#define regUVD_JPEG_SPS1_INFO_BASE_IDX                                                                  1
+#define regUVD_JPEG_RE_TIMER                                                                            0x05c8
+#define regUVD_JPEG_RE_TIMER_BASE_IDX                                                                   1
+#define regUVD_JPEG_DEC_SCRATCH0                                                                        0x05c9
+#define regUVD_JPEG_DEC_SCRATCH0_BASE_IDX                                                               1
+#define regUVD_JPEG_INT_EN                                                                              0x05ca
+#define regUVD_JPEG_INT_EN_BASE_IDX                                                                     1
+#define regUVD_JPEG_INT_STAT                                                                            0x05cb
+#define regUVD_JPEG_INT_STAT_BASE_IDX                                                                   1
+#define regUVD_JPEG_TIER_CNTL0                                                                          0x05cc
+#define regUVD_JPEG_TIER_CNTL0_BASE_IDX                                                                 1
+#define regUVD_JPEG_TIER_CNTL1                                                                          0x05cd
+#define regUVD_JPEG_TIER_CNTL1_BASE_IDX                                                                 1
+#define regUVD_JPEG_TIER_CNTL2                                                                          0x05ce
+#define regUVD_JPEG_TIER_CNTL2_BASE_IDX                                                                 1
+#define regUVD_JPEG_TIER_STATUS                                                                         0x05cf
+#define regUVD_JPEG_TIER_STATUS_BASE_IDX                                                                1
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_sclk0_jpegnpsclkdec
+// base address: 0x21000
+#define regUVD_JPEG_OUTBUF_CNTL                                                                         0x0600
+#define regUVD_JPEG_OUTBUF_CNTL_BASE_IDX                                                                1
+#define regUVD_JPEG_OUTBUF_WPTR                                                                         0x0601
+#define regUVD_JPEG_OUTBUF_WPTR_BASE_IDX                                                                1
+#define regUVD_JPEG_OUTBUF_RPTR                                                                         0x0602
+#define regUVD_JPEG_OUTBUF_RPTR_BASE_IDX                                                                1
+#define regUVD_JPEG_PITCH                                                                               0x0603
+#define regUVD_JPEG_PITCH_BASE_IDX                                                                      1
+#define regUVD_JPEG_UV_PITCH                                                                            0x0604
+#define regUVD_JPEG_UV_PITCH_BASE_IDX                                                                   1
+#define regJPEG_DEC_Y_GFX8_TILING_SURFACE                                                               0x0605
+#define regJPEG_DEC_Y_GFX8_TILING_SURFACE_BASE_IDX                                                      1
+#define regJPEG_DEC_UV_GFX8_TILING_SURFACE                                                              0x0606
+#define regJPEG_DEC_UV_GFX8_TILING_SURFACE_BASE_IDX                                                     1
+#define regJPEG_DEC_GFX8_ADDR_CONFIG                                                                    0x0607
+#define regJPEG_DEC_GFX8_ADDR_CONFIG_BASE_IDX                                                           1
+#define regJPEG_DEC_Y_GFX10_TILING_SURFACE                                                              0x0608
+#define regJPEG_DEC_Y_GFX10_TILING_SURFACE_BASE_IDX                                                     1
+#define regJPEG_DEC_UV_GFX10_TILING_SURFACE                                                             0x0609
+#define regJPEG_DEC_UV_GFX10_TILING_SURFACE_BASE_IDX                                                    1
+#define regJPEG_DEC_GFX10_ADDR_CONFIG                                                                   0x060a
+#define regJPEG_DEC_GFX10_ADDR_CONFIG_BASE_IDX                                                          1
+#define regJPEG_DEC_ADDR_MODE                                                                           0x060b
+#define regJPEG_DEC_ADDR_MODE_BASE_IDX                                                                  1
+#define regUVD_JPEG_OUTPUT_XY                                                                           0x060c
+#define regUVD_JPEG_OUTPUT_XY_BASE_IDX                                                                  1
+#define regUVD_JPEG_GPCOM_CMD                                                                           0x060d
+#define regUVD_JPEG_GPCOM_CMD_BASE_IDX                                                                  1
+#define regUVD_JPEG_GPCOM_DATA0                                                                         0x060e
+#define regUVD_JPEG_GPCOM_DATA0_BASE_IDX                                                                1
+#define regUVD_JPEG_GPCOM_DATA1                                                                         0x060f
+#define regUVD_JPEG_GPCOM_DATA1_BASE_IDX                                                                1
+#define regUVD_JPEG_SCRATCH1                                                                            0x0610
+#define regUVD_JPEG_SCRATCH1_BASE_IDX                                                                   1
+#define regUVD_JPEG_DEC_SOFT_RST                                                                        0x0611
+#define regUVD_JPEG_DEC_SOFT_RST_BASE_IDX                                                               1
+
+
+// addressBlock: aid_uvd0_uvd_jrbc0_uvd_jrbc_dec
+// base address: 0x21100
+#define regUVD_JRBC0_UVD_JRBC_RB_WPTR                                                                   0x0640
+#define regUVD_JRBC0_UVD_JRBC_RB_WPTR_BASE_IDX                                                          1
+#define regUVD_JRBC0_UVD_JRBC_RB_CNTL                                                                   0x0641
+#define regUVD_JRBC0_UVD_JRBC_RB_CNTL_BASE_IDX                                                          1
+#define regUVD_JRBC0_UVD_JRBC_IB_SIZE                                                                   0x0642
+#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_BASE_IDX                                                          1
+#define regUVD_JRBC0_UVD_JRBC_URGENT_CNTL                                                               0x0643
+#define regUVD_JRBC0_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      1
+#define regUVD_JRBC0_UVD_JRBC_RB_REF_DATA                                                               0x0644
+#define regUVD_JRBC0_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      1
+#define regUVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0645
+#define regUVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 1
+#define regUVD_JRBC0_UVD_JRBC_SOFT_RESET                                                                0x0648
+#define regUVD_JRBC0_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       1
+#define regUVD_JRBC0_UVD_JRBC_STATUS                                                                    0x0649
+#define regUVD_JRBC0_UVD_JRBC_STATUS_BASE_IDX                                                           1
+#define regUVD_JRBC0_UVD_JRBC_RB_RPTR                                                                   0x064a
+#define regUVD_JRBC0_UVD_JRBC_RB_RPTR_BASE_IDX                                                          1
+#define regUVD_JRBC0_UVD_JRBC_RB_BUF_STATUS                                                             0x064b
+#define regUVD_JRBC0_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    1
+#define regUVD_JRBC0_UVD_JRBC_IB_BUF_STATUS                                                             0x064c
+#define regUVD_JRBC0_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    1
+#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE                                                            0x064d
+#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   1
+#define regUVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER                                                          0x064e
+#define regUVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 1
+#define regUVD_JRBC0_UVD_JRBC_IB_REF_DATA                                                               0x064f
+#define regUVD_JRBC0_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      1
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_CMD                                                               0x0650
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      1
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0651
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              1
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0652
+#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              1
+#define regUVD_JRBC0_UVD_JRBC_RB_SIZE                                                                   0x0653
+#define regUVD_JRBC0_UVD_JRBC_RB_SIZE_BASE_IDX                                                          1
+#define regUVD_JRBC0_UVD_JRBC_SCRATCH0                                                                  0x0654
+#define regUVD_JRBC0_UVD_JRBC_SCRATCH0_BASE_IDX                                                         1
+
+
+// addressBlock: aid_uvd0_uvd_jmi0_uvd_jmi_dec
+// base address: 0x21180
+#define regUVD_JMI0_UVD_JPEG_DEC_PF_CTRL                                                                0x0660
+#define regUVD_JMI0_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       1
+#define regUVD_JMI0_UVD_LMI_JRBC_CTRL                                                                   0x0661
+#define regUVD_JMI0_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          1
+#define regUVD_JMI0_UVD_LMI_JPEG_CTRL                                                                   0x0662
+#define regUVD_JMI0_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          1
+#define regUVD_JMI0_JPEG_LMI_DROP                                                                       0x0663
+#define regUVD_JMI0_JPEG_LMI_DROP_BASE_IDX                                                              1
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_VMID                                                                0x0664
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       1
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_VMID                                                                0x0665
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       1
+#define regUVD_JMI0_UVD_LMI_JPEG_VMID                                                                   0x0666
+#define regUVD_JMI0_UVD_LMI_JPEG_VMID_BASE_IDX                                                          1
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x0667
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   1
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x0668
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  1
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x0669
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              1
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x066a
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             1
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x066b
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       1
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x066c
+#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      1
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x066d
+#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  1
+#define regUVD_JMI0_UVD_JMI_DEC_SWAP_CNTL                                                               0x066e
+#define regUVD_JMI0_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      1
+#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL                                                                 0x066f
+#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        1
+#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x0670
+#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   1
+#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x0671
+#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  1
+#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x0672
+#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            1
+#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x0673
+#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           1
+#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x0674
+#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           1
+#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x0675
+#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          1
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x0676
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              1
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x0677
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             1
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x0678
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       1
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x0679
+#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      1
+#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL2                                                                0x067d
+#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       1
+
+
+// addressBlock: aid_uvd0_uvd_jmi_common_dec
+// base address: 0x21300
+#define regUVD_JADP_MCIF_URGENT_CTRL                                                                    0x06c1
+#define regUVD_JADP_MCIF_URGENT_CTRL_BASE_IDX                                                           1
+#define regUVD_JMI_URGENT_CTRL                                                                          0x06c2
+#define regUVD_JMI_URGENT_CTRL_BASE_IDX                                                                 1
+#define regUVD_JMI_CTRL                                                                                 0x06c3
+#define regUVD_JMI_CTRL_BASE_IDX                                                                        1
+#define regJPEG_MEMCHECK_CLAMPING_CNTL                                                                  0x06c4
+#define regJPEG_MEMCHECK_CLAMPING_CNTL_BASE_IDX                                                         1
+#define regJPEG_MEMCHECK_SAFE_ADDR                                                                      0x06c5
+#define regJPEG_MEMCHECK_SAFE_ADDR_BASE_IDX                                                             1
+#define regJPEG_MEMCHECK_SAFE_ADDR_64BIT                                                                0x06c6
+#define regJPEG_MEMCHECK_SAFE_ADDR_64BIT_BASE_IDX                                                       1
+#define regUVD_JMI_LAT_CTRL                                                                             0x06c7
+#define regUVD_JMI_LAT_CTRL_BASE_IDX                                                                    1
+#define regUVD_JMI_LAT_CNTR                                                                             0x06c8
+#define regUVD_JMI_LAT_CNTR_BASE_IDX                                                                    1
+#define regUVD_JMI_AVG_LAT_CNTR                                                                         0x06c9
+#define regUVD_JMI_AVG_LAT_CNTR_BASE_IDX                                                                1
+#define regUVD_JMI_PERFMON_CTRL                                                                         0x06ca
+#define regUVD_JMI_PERFMON_CTRL_BASE_IDX                                                                1
+#define regUVD_JMI_PERFMON_COUNT_LO                                                                     0x06cb
+#define regUVD_JMI_PERFMON_COUNT_LO_BASE_IDX                                                            1
+#define regUVD_JMI_PERFMON_COUNT_HI                                                                     0x06cc
+#define regUVD_JMI_PERFMON_COUNT_HI_BASE_IDX                                                            1
+#define regUVD_JMI_CLEAN_STATUS                                                                         0x06cd
+#define regUVD_JMI_CLEAN_STATUS_BASE_IDX                                                                1
+#define regUVD_JMI_CNTL                                                                                 0x06ce
+#define regUVD_JMI_CNTL_BASE_IDX                                                                        1
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_common_dec
+// base address: 0x21400
+#define regJPEG_SOFT_RESET_STATUS                                                                       0x0700
+#define regJPEG_SOFT_RESET_STATUS_BASE_IDX                                                              1
+#define regJPEG_SYS_INT_EN                                                                              0x0701
+#define regJPEG_SYS_INT_EN_BASE_IDX                                                                     1
+#define regJPEG_SYS_INT_EN1                                                                             0x0702
+#define regJPEG_SYS_INT_EN1_BASE_IDX                                                                    1
+#define regJPEG_SYS_INT_STATUS                                                                          0x0703
+#define regJPEG_SYS_INT_STATUS_BASE_IDX                                                                 1
+#define regJPEG_SYS_INT_STATUS1                                                                         0x0704
+#define regJPEG_SYS_INT_STATUS1_BASE_IDX                                                                1
+#define regJPEG_SYS_INT_ACK                                                                             0x0705
+#define regJPEG_SYS_INT_ACK_BASE_IDX                                                                    1
+#define regJPEG_SYS_INT_ACK1                                                                            0x0706
+#define regJPEG_SYS_INT_ACK1_BASE_IDX                                                                   1
+#define regJPEG_MEMCHECK_SYS_INT_EN                                                                     0x0707
+#define regJPEG_MEMCHECK_SYS_INT_EN_BASE_IDX                                                            1
+#define regJPEG_MEMCHECK_SYS_INT_EN1                                                                    0x0708
+#define regJPEG_MEMCHECK_SYS_INT_EN1_BASE_IDX                                                           1
+#define regJPEG_MEMCHECK_SYS_INT_STAT                                                                   0x0709
+#define regJPEG_MEMCHECK_SYS_INT_STAT_BASE_IDX                                                          1
+#define regJPEG_MEMCHECK_SYS_INT_STAT1                                                                  0x070a
+#define regJPEG_MEMCHECK_SYS_INT_STAT1_BASE_IDX                                                         1
+#define regJPEG_MEMCHECK_SYS_INT_STAT2                                                                  0x070b
+#define regJPEG_MEMCHECK_SYS_INT_STAT2_BASE_IDX                                                         1
+#define regJPEG_MEMCHECK_SYS_INT_ACK                                                                    0x070c
+#define regJPEG_MEMCHECK_SYS_INT_ACK_BASE_IDX                                                           1
+#define regJPEG_MEMCHECK_SYS_INT_ACK1                                                                   0x070d
+#define regJPEG_MEMCHECK_SYS_INT_ACK1_BASE_IDX                                                          1
+#define regJPEG_MEMCHECK_SYS_INT_ACK2                                                                   0x070e
+#define regJPEG_MEMCHECK_SYS_INT_ACK2_BASE_IDX                                                          1
+#define regJPEG_MASTINT_EN                                                                              0x0710
+#define regJPEG_MASTINT_EN_BASE_IDX                                                                     1
+#define regJPEG_IH_CTRL                                                                                 0x0711
+#define regJPEG_IH_CTRL_BASE_IDX                                                                        1
+#define regJRBBM_ARB_CTRL                                                                               0x0713
+#define regJRBBM_ARB_CTRL_BASE_IDX                                                                      1
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_common_sclk_dec
+// base address: 0x21480
+#define regJPEG_CGC_GATE                                                                                0x0720
+#define regJPEG_CGC_GATE_BASE_IDX                                                                       1
+#define regJPEG_CGC_CTRL                                                                                0x0721
+#define regJPEG_CGC_CTRL_BASE_IDX                                                                       1
+#define regJPEG_CGC_STATUS                                                                              0x0722
+#define regJPEG_CGC_STATUS_BASE_IDX                                                                     1
+#define regJPEG_COMN_CGC_MEM_CTRL                                                                       0x0723
+#define regJPEG_COMN_CGC_MEM_CTRL_BASE_IDX                                                              1
+#define regJPEG_DEC_CGC_MEM_CTRL                                                                        0x0724
+#define regJPEG_DEC_CGC_MEM_CTRL_BASE_IDX                                                               1
+#define regJPEG_ENC_CGC_MEM_CTRL                                                                        0x0726
+#define regJPEG_ENC_CGC_MEM_CTRL_BASE_IDX                                                               1
+#define regJPEG_PERF_BANK_CONF                                                                          0x0727
+#define regJPEG_PERF_BANK_CONF_BASE_IDX                                                                 1
+#define regJPEG_PERF_BANK_EVENT_SEL                                                                     0x0728
+#define regJPEG_PERF_BANK_EVENT_SEL_BASE_IDX                                                            1
+#define regJPEG_PERF_BANK_COUNT0                                                                        0x0729
+#define regJPEG_PERF_BANK_COUNT0_BASE_IDX                                                               1
+#define regJPEG_PERF_BANK_COUNT1                                                                        0x072a
+#define regJPEG_PERF_BANK_COUNT1_BASE_IDX                                                               1
+#define regJPEG_PERF_BANK_COUNT2                                                                        0x072b
+#define regJPEG_PERF_BANK_COUNT2_BASE_IDX                                                               1
+#define regJPEG_PERF_BANK_COUNT3                                                                        0x072c
+#define regJPEG_PERF_BANK_COUNT3_BASE_IDX                                                               1
+
+
+// addressBlock: aid_uvd0_uvd_pg_dec
+// base address: 0x1f800
+#define regUVD_PGFSM_CONFIG                                                                             0x0000
+#define regUVD_PGFSM_CONFIG_BASE_IDX                                                                    1
+#define regUVD_PGFSM_STATUS                                                                             0x0001
+#define regUVD_PGFSM_STATUS_BASE_IDX                                                                    1
+#define regUVD_POWER_STATUS                                                                             0x0002
+#define regUVD_POWER_STATUS_BASE_IDX                                                                    1
+#define regUVD_JPEG_POWER_STATUS                                                                        0x0003
+#define regUVD_JPEG_POWER_STATUS_BASE_IDX                                                               1
+#define regUVD_MC_DJPEG_RD_SPACE                                                                        0x0006
+#define regUVD_MC_DJPEG_RD_SPACE_BASE_IDX                                                               1
+#define regUVD_MC_DJPEG_WR_SPACE                                                                        0x0007
+#define regUVD_MC_DJPEG_WR_SPACE_BASE_IDX                                                               1
+#define regUVD_MC_EJPEG_RD_SPACE                                                                        0x0008
+#define regUVD_MC_EJPEG_RD_SPACE_BASE_IDX                                                               1
+#define regUVD_MC_EJPEG_WR_SPACE                                                                        0x0009
+#define regUVD_MC_EJPEG_WR_SPACE_BASE_IDX                                                               1
+#define regUVD_PG_IND_INDEX                                                                             0x000c
+#define regUVD_PG_IND_INDEX_BASE_IDX                                                                    1
+#define regUVD_PG_IND_DATA                                                                              0x000e
+#define regUVD_PG_IND_DATA_BASE_IDX                                                                     1
+#define regCC_UVD_HARVESTING                                                                            0x000f
+#define regCC_UVD_HARVESTING_BASE_IDX                                                                   1
+#define regUVD_DPG_LMA_CTL                                                                              0x0011
+#define regUVD_DPG_LMA_CTL_BASE_IDX                                                                     1
+#define regUVD_DPG_LMA_DATA                                                                             0x0012
+#define regUVD_DPG_LMA_DATA_BASE_IDX                                                                    1
+#define regUVD_DPG_LMA_MASK                                                                             0x0013
+#define regUVD_DPG_LMA_MASK_BASE_IDX                                                                    1
+#define regUVD_DPG_PAUSE                                                                                0x0014
+#define regUVD_DPG_PAUSE_BASE_IDX                                                                       1
+#define regUVD_SCRATCH1                                                                                 0x0015
+#define regUVD_SCRATCH1_BASE_IDX                                                                        1
+#define regUVD_SCRATCH2                                                                                 0x0016
+#define regUVD_SCRATCH2_BASE_IDX                                                                        1
+#define regUVD_SCRATCH3                                                                                 0x0017
+#define regUVD_SCRATCH3_BASE_IDX                                                                        1
+#define regUVD_SCRATCH4                                                                                 0x0018
+#define regUVD_SCRATCH4_BASE_IDX                                                                        1
+#define regUVD_SCRATCH5                                                                                 0x0019
+#define regUVD_SCRATCH5_BASE_IDX                                                                        1
+#define regUVD_SCRATCH6                                                                                 0x001a
+#define regUVD_SCRATCH6_BASE_IDX                                                                        1
+#define regUVD_SCRATCH7                                                                                 0x001b
+#define regUVD_SCRATCH7_BASE_IDX                                                                        1
+#define regUVD_SCRATCH8                                                                                 0x001c
+#define regUVD_SCRATCH8_BASE_IDX                                                                        1
+#define regUVD_SCRATCH9                                                                                 0x001d
+#define regUVD_SCRATCH9_BASE_IDX                                                                        1
+#define regUVD_SCRATCH10                                                                                0x001e
+#define regUVD_SCRATCH10_BASE_IDX                                                                       1
+#define regUVD_SCRATCH11                                                                                0x001f
+#define regUVD_SCRATCH11_BASE_IDX                                                                       1
+#define regUVD_SCRATCH12                                                                                0x0020
+#define regUVD_SCRATCH12_BASE_IDX                                                                       1
+#define regUVD_SCRATCH13                                                                                0x0021
+#define regUVD_SCRATCH13_BASE_IDX                                                                       1
+#define regUVD_SCRATCH14                                                                                0x0022
+#define regUVD_SCRATCH14_BASE_IDX                                                                       1
+#define regUVD_FREE_COUNTER_REG                                                                         0x0023
+#define regUVD_FREE_COUNTER_REG_BASE_IDX                                                                1
+#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW                                                         0x0024
+#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW_BASE_IDX                                                1
+#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH                                                        0x0025
+#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH_BASE_IDX                                               1
+#define regUVD_DPG_VCPU_CACHE_OFFSET0                                                                   0x0026
+#define regUVD_DPG_VCPU_CACHE_OFFSET0_BASE_IDX                                                          1
+#define regUVD_DPG_LMI_VCPU_CACHE_VMID                                                                  0x0027
+#define regUVD_DPG_LMI_VCPU_CACHE_VMID_BASE_IDX                                                         1
+#define regUVD_REG_FILTER_EN                                                                            0x0028
+#define regUVD_REG_FILTER_EN_BASE_IDX                                                                   1
+#define regUVD_SECURITY_REG_VIO_REPORT                                                                  0x0029
+#define regUVD_SECURITY_REG_VIO_REPORT_BASE_IDX                                                         1
+#define regUVD_FW_VERSION                                                                               0x002a
+#define regUVD_FW_VERSION_BASE_IDX                                                                      1
+#define regUVD_PF_STATUS                                                                                0x002c
+#define regUVD_PF_STATUS_BASE_IDX                                                                       1
+#define regUVD_DPG_CLK_EN_VCPU_REPORT                                                                   0x002e
+#define regUVD_DPG_CLK_EN_VCPU_REPORT_BASE_IDX                                                          1
+#define regCC_UVD_VCPU_ERR_DETECT_BOT_LO                                                                0x002f
+#define regCC_UVD_VCPU_ERR_DETECT_BOT_LO_BASE_IDX                                                       1
+#define regCC_UVD_VCPU_ERR_DETECT_BOT_HI                                                                0x0030
+#define regCC_UVD_VCPU_ERR_DETECT_BOT_HI_BASE_IDX                                                       1
+#define regCC_UVD_VCPU_ERR_DETECT_TOP_LO                                                                0x0031
+#define regCC_UVD_VCPU_ERR_DETECT_TOP_LO_BASE_IDX                                                       1
+#define regCC_UVD_VCPU_ERR_DETECT_TOP_HI                                                                0x0032
+#define regCC_UVD_VCPU_ERR_DETECT_TOP_HI_BASE_IDX                                                       1
+#define regCC_UVD_VCPU_ERR                                                                              0x0033
+#define regCC_UVD_VCPU_ERR_BASE_IDX                                                                     1
+#define regCC_UVD_VCPU_ERR_INST_ADDR_LO                                                                 0x0034
+#define regCC_UVD_VCPU_ERR_INST_ADDR_LO_BASE_IDX                                                        1
+#define regCC_UVD_VCPU_ERR_INST_ADDR_HI                                                                 0x0035
+#define regCC_UVD_VCPU_ERR_INST_ADDR_HI_BASE_IDX                                                        1
+#define regUVD_LMI_MMSCH_NC_SPACE                                                                       0x003d
+#define regUVD_LMI_MMSCH_NC_SPACE_BASE_IDX                                                              1
+#define regUVD_LMI_ATOMIC_SPACE                                                                         0x003e
+#define regUVD_LMI_ATOMIC_SPACE_BASE_IDX                                                                1
+#define regUVD_GFX8_ADDR_CONFIG                                                                         0x0041
+#define regUVD_GFX8_ADDR_CONFIG_BASE_IDX                                                                1
+#define regUVD_GFX10_ADDR_CONFIG                                                                        0x0042
+#define regUVD_GFX10_ADDR_CONFIG_BASE_IDX                                                               1
+#define regUVD_GPCNT2_CNTL                                                                              0x0043
+#define regUVD_GPCNT2_CNTL_BASE_IDX                                                                     1
+#define regUVD_GPCNT2_TARGET_LOWER                                                                      0x0044
+#define regUVD_GPCNT2_TARGET_LOWER_BASE_IDX                                                             1
+#define regUVD_GPCNT2_STATUS_LOWER                                                                      0x0045
+#define regUVD_GPCNT2_STATUS_LOWER_BASE_IDX                                                             1
+#define regUVD_GPCNT2_TARGET_UPPER                                                                      0x0046
+#define regUVD_GPCNT2_TARGET_UPPER_BASE_IDX                                                             1
+#define regUVD_GPCNT2_STATUS_UPPER                                                                      0x0047
+#define regUVD_GPCNT2_STATUS_UPPER_BASE_IDX                                                             1
+#define regUVD_GPCNT3_CNTL                                                                              0x0048
+#define regUVD_GPCNT3_CNTL_BASE_IDX                                                                     1
+#define regUVD_GPCNT3_TARGET_LOWER                                                                      0x0049
+#define regUVD_GPCNT3_TARGET_LOWER_BASE_IDX                                                             1
+#define regUVD_GPCNT3_STATUS_LOWER                                                                      0x004a
+#define regUVD_GPCNT3_STATUS_LOWER_BASE_IDX                                                             1
+#define regUVD_GPCNT3_TARGET_UPPER                                                                      0x004b
+#define regUVD_GPCNT3_TARGET_UPPER_BASE_IDX                                                             1
+#define regUVD_GPCNT3_STATUS_UPPER                                                                      0x004c
+#define regUVD_GPCNT3_STATUS_UPPER_BASE_IDX                                                             1
+#define regUVD_VCLK_DS_CNTL                                                                             0x004d
+#define regUVD_VCLK_DS_CNTL_BASE_IDX                                                                    1
+#define regUVD_DCLK_DS_CNTL                                                                             0x004e
+#define regUVD_DCLK_DS_CNTL_BASE_IDX                                                                    1
+#define regUVD_TSC_LOWER                                                                                0x004f
+#define regUVD_TSC_LOWER_BASE_IDX                                                                       1
+#define regUVD_TSC_UPPER                                                                                0x0050
+#define regUVD_TSC_UPPER_BASE_IDX                                                                       1
+#define regVCN_FEATURES                                                                                 0x0051
+#define regVCN_FEATURES_BASE_IDX                                                                        1
+#define regUVD_GPUIOV_STATUS                                                                            0x0055
+#define regUVD_GPUIOV_STATUS_BASE_IDX                                                                   1
+#define regUVD_RAS_VCPU_VCODEC_STATUS                                                                   0x0057
+#define regUVD_RAS_VCPU_VCODEC_STATUS_BASE_IDX                                                          1
+#define regUVD_RAS_MMSCH_FATAL_ERROR                                                                    0x0058
+#define regUVD_RAS_MMSCH_FATAL_ERROR_BASE_IDX                                                           1
+#define regUVD_RAS_JPEG0_STATUS                                                                         0x0059
+#define regUVD_RAS_JPEG0_STATUS_BASE_IDX                                                                1
+#define regUVD_RAS_JPEG1_STATUS                                                                         0x005a
+#define regUVD_RAS_JPEG1_STATUS_BASE_IDX                                                                1
+#define regUVD_RAS_CNTL_PMI_ARB                                                                         0x005b
+#define regUVD_RAS_CNTL_PMI_ARB_BASE_IDX                                                                1
+#define regUVD_SCRATCH15                                                                                0x005c
+#define regUVD_SCRATCH15_BASE_IDX                                                                       1
+#define regVCN_JPEG_DB_CTRL1                                                                            0x005d
+#define regVCN_JPEG_DB_CTRL1_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL2                                                                            0x005e
+#define regVCN_JPEG_DB_CTRL2_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL3                                                                            0x005f
+#define regVCN_JPEG_DB_CTRL3_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL4                                                                            0x0060
+#define regVCN_JPEG_DB_CTRL4_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL5                                                                            0x0061
+#define regVCN_JPEG_DB_CTRL5_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL6                                                                            0x0062
+#define regVCN_JPEG_DB_CTRL6_BASE_IDX                                                                   1
+#define regVCN_JPEG_DB_CTRL7                                                                            0x0063
+#define regVCN_JPEG_DB_CTRL7_BASE_IDX                                                                   1
+#define regUVD_SCRATCH32                                                                                0x006d
+#define regUVD_SCRATCH32_BASE_IDX                                                                       1
+#define regUVD_VERSION                                                                                  0x006e
+#define regUVD_VERSION_BASE_IDX                                                                         1
+#define regVCN_RB_DB_CTRL                                                                               0x0070
+#define regVCN_RB_DB_CTRL_BASE_IDX                                                                      1
+#define regVCN_JPEG_DB_CTRL                                                                             0x0071
+#define regVCN_JPEG_DB_CTRL_BASE_IDX                                                                    1
+#define regVCN_RB1_DB_CTRL                                                                              0x0072
+#define regVCN_RB1_DB_CTRL_BASE_IDX                                                                     1
+#define regVCN_RB2_DB_CTRL                                                                              0x0073
+#define regVCN_RB2_DB_CTRL_BASE_IDX                                                                     1
+#define regVCN_RB3_DB_CTRL                                                                              0x0074
+#define regVCN_RB3_DB_CTRL_BASE_IDX                                                                     1
+#define regVCN_RB4_DB_CTRL                                                                              0x0075
+#define regVCN_RB4_DB_CTRL_BASE_IDX                                                                     1
+#define regVCN_RB_ENABLE                                                                                0x0085
+#define regVCN_RB_ENABLE_BASE_IDX                                                                       1
+#define regVCN_RB_WPTR_CTRL                                                                             0x0086
+#define regVCN_RB_WPTR_CTRL_BASE_IDX                                                                    1
+#define regUVD_RB_RPTR                                                                                  0x00ac
+#define regUVD_RB_RPTR_BASE_IDX                                                                         1
+#define regUVD_RB_WPTR                                                                                  0x00ad
+#define regUVD_RB_WPTR_BASE_IDX                                                                         1
+#define regUVD_RB_RPTR2                                                                                 0x00ae
+#define regUVD_RB_RPTR2_BASE_IDX                                                                        1
+#define regUVD_RB_WPTR2                                                                                 0x00af
+#define regUVD_RB_WPTR2_BASE_IDX                                                                        1
+#define regUVD_RB_RPTR3                                                                                 0x00b0
+#define regUVD_RB_RPTR3_BASE_IDX                                                                        1
+#define regUVD_RB_WPTR3                                                                                 0x00b1
+#define regUVD_RB_WPTR3_BASE_IDX                                                                        1
+#define regUVD_RB_RPTR4                                                                                 0x00b2
+#define regUVD_RB_RPTR4_BASE_IDX                                                                        1
+#define regUVD_RB_WPTR4                                                                                 0x00b3
+#define regUVD_RB_WPTR4_BASE_IDX                                                                        1
+#define regUVD_OUT_RB_RPTR                                                                              0x00b4
+#define regUVD_OUT_RB_RPTR_BASE_IDX                                                                     1
+#define regUVD_OUT_RB_WPTR                                                                              0x00b5
+#define regUVD_OUT_RB_WPTR_BASE_IDX                                                                     1
+#define regUVD_AUDIO_RB_RPTR                                                                            0x00b6
+#define regUVD_AUDIO_RB_RPTR_BASE_IDX                                                                   1
+#define regUVD_AUDIO_RB_WPTR                                                                            0x00b7
+#define regUVD_AUDIO_RB_WPTR_BASE_IDX                                                                   1
+#define regUVD_RBC_RB_RPTR                                                                              0x00b8
+#define regUVD_RBC_RB_RPTR_BASE_IDX                                                                     1
+#define regUVD_RBC_RB_WPTR                                                                              0x00b9
+#define regUVD_RBC_RB_WPTR_BASE_IDX                                                                     1
+#define regUVD_DPG_LMA_CTL2                                                                             0x00bb
+#define regUVD_DPG_LMA_CTL2_BASE_IDX                                                                    1
+
+
+// addressBlock: aid_uvd0_mmsch_dec
+// base address: 0x20d00
+#define regMMSCH_UCODE_ADDR                                                                             0x0540
+#define regMMSCH_UCODE_ADDR_BASE_IDX                                                                    1
+#define regMMSCH_UCODE_DATA                                                                             0x0541
+#define regMMSCH_UCODE_DATA_BASE_IDX                                                                    1
+#define regMMSCH_SRAM_ADDR                                                                              0x0542
+#define regMMSCH_SRAM_ADDR_BASE_IDX                                                                     1
+#define regMMSCH_SRAM_DATA                                                                              0x0543
+#define regMMSCH_SRAM_DATA_BASE_IDX                                                                     1
+#define regMMSCH_VF_SRAM_OFFSET                                                                         0x0544
+#define regMMSCH_VF_SRAM_OFFSET_BASE_IDX                                                                1
+#define regMMSCH_DB_SRAM_OFFSET                                                                         0x0545
+#define regMMSCH_DB_SRAM_OFFSET_BASE_IDX                                                                1
+#define regMMSCH_CTX_SRAM_OFFSET                                                                        0x0546
+#define regMMSCH_CTX_SRAM_OFFSET_BASE_IDX                                                               1
+#define regMMSCH_CTL                                                                                    0x0547
+#define regMMSCH_CTL_BASE_IDX                                                                           1
+#define regMMSCH_INTR                                                                                   0x0548
+#define regMMSCH_INTR_BASE_IDX                                                                          1
+#define regMMSCH_INTR_ACK                                                                               0x0549
+#define regMMSCH_INTR_ACK_BASE_IDX                                                                      1
+#define regMMSCH_INTR_STATUS                                                                            0x054a
+#define regMMSCH_INTR_STATUS_BASE_IDX                                                                   1
+#define regMMSCH_VF_VMID                                                                                0x054b
+#define regMMSCH_VF_VMID_BASE_IDX                                                                       1
+#define regMMSCH_VF_CTX_ADDR_LO                                                                         0x054c
+#define regMMSCH_VF_CTX_ADDR_LO_BASE_IDX                                                                1
+#define regMMSCH_VF_CTX_ADDR_HI                                                                         0x054d
+#define regMMSCH_VF_CTX_ADDR_HI_BASE_IDX                                                                1
+#define regMMSCH_VF_CTX_SIZE                                                                            0x054e
+#define regMMSCH_VF_CTX_SIZE_BASE_IDX                                                                   1
+#define regMMSCH_VF_GPCOM_ADDR_LO                                                                       0x054f
+#define regMMSCH_VF_GPCOM_ADDR_LO_BASE_IDX                                                              1
+#define regMMSCH_VF_GPCOM_ADDR_HI                                                                       0x0550
+#define regMMSCH_VF_GPCOM_ADDR_HI_BASE_IDX                                                              1
+#define regMMSCH_VF_GPCOM_SIZE                                                                          0x0551
+#define regMMSCH_VF_GPCOM_SIZE_BASE_IDX                                                                 1
+#define regMMSCH_VF_MAILBOX_HOST                                                                        0x0552
+#define regMMSCH_VF_MAILBOX_HOST_BASE_IDX                                                               1
+#define regMMSCH_VF_MAILBOX_RESP                                                                        0x0553
+#define regMMSCH_VF_MAILBOX_RESP_BASE_IDX                                                               1
+#define regMMSCH_VF_MAILBOX_0                                                                           0x0554
+#define regMMSCH_VF_MAILBOX_0_BASE_IDX                                                                  1
+#define regMMSCH_VF_MAILBOX_0_RESP                                                                      0x0555
+#define regMMSCH_VF_MAILBOX_0_RESP_BASE_IDX                                                             1
+#define regMMSCH_VF_MAILBOX_1                                                                           0x0556
+#define regMMSCH_VF_MAILBOX_1_BASE_IDX                                                                  1
+#define regMMSCH_VF_MAILBOX_1_RESP                                                                      0x0557
+#define regMMSCH_VF_MAILBOX_1_RESP_BASE_IDX                                                             1
+#define regMMSCH_CNTL                                                                                   0x055c
+#define regMMSCH_CNTL_BASE_IDX                                                                          1
+#define regMMSCH_NONCACHE_OFFSET0                                                                       0x055d
+#define regMMSCH_NONCACHE_OFFSET0_BASE_IDX                                                              1
+#define regMMSCH_NONCACHE_SIZE0                                                                         0x055e
+#define regMMSCH_NONCACHE_SIZE0_BASE_IDX                                                                1
+#define regMMSCH_NONCACHE_OFFSET1                                                                       0x055f
+#define regMMSCH_NONCACHE_OFFSET1_BASE_IDX                                                              1
+#define regMMSCH_NONCACHE_SIZE1                                                                         0x0560
+#define regMMSCH_NONCACHE_SIZE1_BASE_IDX                                                                1
+#define regMMSCH_PROC_STATE1                                                                            0x0566
+#define regMMSCH_PROC_STATE1_BASE_IDX                                                                   1
+#define regMMSCH_LAST_MC_ADDR                                                                           0x0567
+#define regMMSCH_LAST_MC_ADDR_BASE_IDX                                                                  1
+#define regMMSCH_LAST_MEM_ACCESS_HI                                                                     0x0568
+#define regMMSCH_LAST_MEM_ACCESS_HI_BASE_IDX                                                            1
+#define regMMSCH_LAST_MEM_ACCESS_LO                                                                     0x0569
+#define regMMSCH_LAST_MEM_ACCESS_LO_BASE_IDX                                                            1
+#define regMMSCH_IOV_ACTIVE_FCN_ID                                                                      0x056a
+#define regMMSCH_IOV_ACTIVE_FCN_ID_BASE_IDX                                                             1
+#define regMMSCH_SCRATCH_0                                                                              0x056b
+#define regMMSCH_SCRATCH_0_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_1                                                                              0x056c
+#define regMMSCH_SCRATCH_1_BASE_IDX                                                                     1
+#define regMMSCH_GPUIOV_SCH_BLOCK_0                                                                     0x056d
+#define regMMSCH_GPUIOV_SCH_BLOCK_0_BASE_IDX                                                            1
+#define regMMSCH_GPUIOV_CMD_CONTROL_0                                                                   0x056e
+#define regMMSCH_GPUIOV_CMD_CONTROL_0_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_CMD_STATUS_0                                                                    0x056f
+#define regMMSCH_GPUIOV_CMD_STATUS_0_BASE_IDX                                                           1
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_0                                                                0x0570
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_0_BASE_IDX                                                       1
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_0                                                                   0x0571
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_0_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_0                                                                 0x0572
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_0_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_DW6_0                                                                           0x0573
+#define regMMSCH_GPUIOV_DW6_0_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW7_0                                                                           0x0574
+#define regMMSCH_GPUIOV_DW7_0_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW8_0                                                                           0x0575
+#define regMMSCH_GPUIOV_DW8_0_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_SCH_BLOCK_1                                                                     0x0576
+#define regMMSCH_GPUIOV_SCH_BLOCK_1_BASE_IDX                                                            1
+#define regMMSCH_GPUIOV_CMD_CONTROL_1                                                                   0x0577
+#define regMMSCH_GPUIOV_CMD_CONTROL_1_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_CMD_STATUS_1                                                                    0x0578
+#define regMMSCH_GPUIOV_CMD_STATUS_1_BASE_IDX                                                           1
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_1                                                                0x0579
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_1_BASE_IDX                                                       1
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_1                                                                   0x057a
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_1_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_1                                                                 0x057b
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_1_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_DW6_1                                                                           0x057c
+#define regMMSCH_GPUIOV_DW6_1_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW7_1                                                                           0x057d
+#define regMMSCH_GPUIOV_DW7_1_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW8_1                                                                           0x057e
+#define regMMSCH_GPUIOV_DW8_1_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_CNTXT                                                                           0x057f
+#define regMMSCH_GPUIOV_CNTXT_BASE_IDX                                                                  1
+#define regMMSCH_SCRATCH_2                                                                              0x0580
+#define regMMSCH_SCRATCH_2_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_3                                                                              0x0581
+#define regMMSCH_SCRATCH_3_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_4                                                                              0x0582
+#define regMMSCH_SCRATCH_4_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_5                                                                              0x0583
+#define regMMSCH_SCRATCH_5_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_6                                                                              0x0584
+#define regMMSCH_SCRATCH_6_BASE_IDX                                                                     1
+#define regMMSCH_SCRATCH_7                                                                              0x0585
+#define regMMSCH_SCRATCH_7_BASE_IDX                                                                     1
+#define regMMSCH_VFID_FIFO_HEAD_0                                                                       0x0586
+#define regMMSCH_VFID_FIFO_HEAD_0_BASE_IDX                                                              1
+#define regMMSCH_VFID_FIFO_TAIL_0                                                                       0x0587
+#define regMMSCH_VFID_FIFO_TAIL_0_BASE_IDX                                                              1
+#define regMMSCH_VFID_FIFO_HEAD_1                                                                       0x0588
+#define regMMSCH_VFID_FIFO_HEAD_1_BASE_IDX                                                              1
+#define regMMSCH_VFID_FIFO_TAIL_1                                                                       0x0589
+#define regMMSCH_VFID_FIFO_TAIL_1_BASE_IDX                                                              1
+#define regMMSCH_NACK_STATUS                                                                            0x058a
+#define regMMSCH_NACK_STATUS_BASE_IDX                                                                   1
+#define regMMSCH_VF_MAILBOX0_DATA                                                                       0x058b
+#define regMMSCH_VF_MAILBOX0_DATA_BASE_IDX                                                              1
+#define regMMSCH_VF_MAILBOX1_DATA                                                                       0x058c
+#define regMMSCH_VF_MAILBOX1_DATA_BASE_IDX                                                              1
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_0                                                                  0x058d
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_0_BASE_IDX                                                         1
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_0                                                                 0x058e
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_0_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0                                                              0x058f
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0_BASE_IDX                                                     1
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_1                                                                  0x0590
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_1_BASE_IDX                                                         1
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_1                                                                 0x0591
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_1_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1                                                              0x0592
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1_BASE_IDX                                                     1
+#define regMMSCH_GPUIOV_CNTXT_IP                                                                        0x0593
+#define regMMSCH_GPUIOV_CNTXT_IP_BASE_IDX                                                               1
+#define regMMSCH_GPUIOV_SCH_BLOCK_2                                                                     0x0594
+#define regMMSCH_GPUIOV_SCH_BLOCK_2_BASE_IDX                                                            1
+#define regMMSCH_GPUIOV_CMD_CONTROL_2                                                                   0x0595
+#define regMMSCH_GPUIOV_CMD_CONTROL_2_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_CMD_STATUS_2                                                                    0x0596
+#define regMMSCH_GPUIOV_CMD_STATUS_2_BASE_IDX                                                           1
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_2                                                                0x0597
+#define regMMSCH_GPUIOV_VM_BUSY_STATUS_2_BASE_IDX                                                       1
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_2                                                                   0x0598
+#define regMMSCH_GPUIOV_ACTIVE_FCNS_2_BASE_IDX                                                          1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_2                                                                 0x0599
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_2_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_DW6_2                                                                           0x059a
+#define regMMSCH_GPUIOV_DW6_2_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW7_2                                                                           0x059b
+#define regMMSCH_GPUIOV_DW7_2_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_DW8_2                                                                           0x059c
+#define regMMSCH_GPUIOV_DW8_2_BASE_IDX                                                                  1
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_2                                                                  0x059d
+#define regMMSCH_GPUIOV_SCH_BLOCK_IP_2_BASE_IDX                                                         1
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_2                                                                 0x059e
+#define regMMSCH_GPUIOV_CMD_STATUS_IP_2_BASE_IDX                                                        1
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2                                                              0x059f
+#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2_BASE_IDX                                                     1
+#define regMMSCH_VFID_FIFO_HEAD_2                                                                       0x05a0
+#define regMMSCH_VFID_FIFO_HEAD_2_BASE_IDX                                                              1
+#define regMMSCH_VFID_FIFO_TAIL_2                                                                       0x05a1
+#define regMMSCH_VFID_FIFO_TAIL_2_BASE_IDX                                                              1
+#define regMMSCH_VM_BUSY_STATUS_0                                                                       0x05a2
+#define regMMSCH_VM_BUSY_STATUS_0_BASE_IDX                                                              1
+#define regMMSCH_VM_BUSY_STATUS_1                                                                       0x05a3
+#define regMMSCH_VM_BUSY_STATUS_1_BASE_IDX                                                              1
+#define regMMSCH_VM_BUSY_STATUS_2                                                                       0x05a4
+#define regMMSCH_VM_BUSY_STATUS_2_BASE_IDX                                                              1
+
+
+// addressBlock: aid_uvd0_slmi_adpdec
+// base address: 0x21c00
+#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_LOW                                                              0x0900
+#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH                                                             0x0901
+#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_LOW                                                              0x0902
+#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH                                                             0x0903
+#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_LOW                                                              0x0904
+#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH                                                             0x0905
+#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_LOW                                                              0x0906
+#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH                                                             0x0907
+#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_LOW                                                              0x0908
+#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH                                                             0x0909
+#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_LOW                                                              0x090a
+#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH                                                             0x090b
+#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_LOW                                                              0x090c
+#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH                                                             0x090d
+#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_LOW                                                              0x090e
+#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_LOW_BASE_IDX                                                     1
+#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH                                                             0x090f
+#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH_BASE_IDX                                                    1
+#define regUVD_LMI_MMSCH_NC_VMID                                                                        0x0910
+#define regUVD_LMI_MMSCH_NC_VMID_BASE_IDX                                                               1
+#define regUVD_LMI_MMSCH_CTRL                                                                           0x0911
+#define regUVD_LMI_MMSCH_CTRL_BASE_IDX                                                                  1
+#define regUVD_MMSCH_LMI_STATUS                                                                         0x0912
+#define regUVD_MMSCH_LMI_STATUS_BASE_IDX                                                                1
+#define regVCN_RAS_CNTL_MMSCH                                                                           0x0914
+#define regVCN_RAS_CNTL_MMSCH_BASE_IDX                                                                  1
+
+
+// addressBlock: aid_uvd0_uvd_jrbc1_uvd_jrbc_dec
+// base address: 0x1e000
+#define regUVD_JRBC1_UVD_JRBC_RB_WPTR                                                                   0x0000
+#define regUVD_JRBC1_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC1_UVD_JRBC_RB_CNTL                                                                   0x0001
+#define regUVD_JRBC1_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC1_UVD_JRBC_IB_SIZE                                                                   0x0002
+#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC1_UVD_JRBC_URGENT_CNTL                                                               0x0003
+#define regUVD_JRBC1_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC1_UVD_JRBC_RB_REF_DATA                                                               0x0004
+#define regUVD_JRBC1_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0005
+#define regUVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC1_UVD_JRBC_SOFT_RESET                                                                0x0008
+#define regUVD_JRBC1_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC1_UVD_JRBC_STATUS                                                                    0x0009
+#define regUVD_JRBC1_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC1_UVD_JRBC_RB_RPTR                                                                   0x000a
+#define regUVD_JRBC1_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC1_UVD_JRBC_RB_BUF_STATUS                                                             0x000b
+#define regUVD_JRBC1_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC1_UVD_JRBC_IB_BUF_STATUS                                                             0x000c
+#define regUVD_JRBC1_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE                                                            0x000d
+#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER                                                          0x000e
+#define regUVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC1_UVD_JRBC_IB_REF_DATA                                                               0x000f
+#define regUVD_JRBC1_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_CMD                                                               0x0010
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0011
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0012
+#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC1_UVD_JRBC_RB_SIZE                                                                   0x0013
+#define regUVD_JRBC1_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC1_UVD_JRBC_SCRATCH0                                                                  0x0014
+#define regUVD_JRBC1_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc2_uvd_jrbc_dec
+// base address: 0x1e100
+#define regUVD_JRBC2_UVD_JRBC_RB_WPTR                                                                   0x0040
+#define regUVD_JRBC2_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC2_UVD_JRBC_RB_CNTL                                                                   0x0041
+#define regUVD_JRBC2_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC2_UVD_JRBC_IB_SIZE                                                                   0x0042
+#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC2_UVD_JRBC_URGENT_CNTL                                                               0x0043
+#define regUVD_JRBC2_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC2_UVD_JRBC_RB_REF_DATA                                                               0x0044
+#define regUVD_JRBC2_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0045
+#define regUVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC2_UVD_JRBC_SOFT_RESET                                                                0x0048
+#define regUVD_JRBC2_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC2_UVD_JRBC_STATUS                                                                    0x0049
+#define regUVD_JRBC2_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC2_UVD_JRBC_RB_RPTR                                                                   0x004a
+#define regUVD_JRBC2_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC2_UVD_JRBC_RB_BUF_STATUS                                                             0x004b
+#define regUVD_JRBC2_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC2_UVD_JRBC_IB_BUF_STATUS                                                             0x004c
+#define regUVD_JRBC2_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE                                                            0x004d
+#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER                                                          0x004e
+#define regUVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC2_UVD_JRBC_IB_REF_DATA                                                               0x004f
+#define regUVD_JRBC2_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_CMD                                                               0x0050
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0051
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0052
+#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC2_UVD_JRBC_RB_SIZE                                                                   0x0053
+#define regUVD_JRBC2_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC2_UVD_JRBC_SCRATCH0                                                                  0x0054
+#define regUVD_JRBC2_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc3_uvd_jrbc_dec
+// base address: 0x1e200
+#define regUVD_JRBC3_UVD_JRBC_RB_WPTR                                                                   0x0080
+#define regUVD_JRBC3_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC3_UVD_JRBC_RB_CNTL                                                                   0x0081
+#define regUVD_JRBC3_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC3_UVD_JRBC_IB_SIZE                                                                   0x0082
+#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC3_UVD_JRBC_URGENT_CNTL                                                               0x0083
+#define regUVD_JRBC3_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC3_UVD_JRBC_RB_REF_DATA                                                               0x0084
+#define regUVD_JRBC3_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0085
+#define regUVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC3_UVD_JRBC_SOFT_RESET                                                                0x0088
+#define regUVD_JRBC3_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC3_UVD_JRBC_STATUS                                                                    0x0089
+#define regUVD_JRBC3_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC3_UVD_JRBC_RB_RPTR                                                                   0x008a
+#define regUVD_JRBC3_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC3_UVD_JRBC_RB_BUF_STATUS                                                             0x008b
+#define regUVD_JRBC3_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC3_UVD_JRBC_IB_BUF_STATUS                                                             0x008c
+#define regUVD_JRBC3_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE                                                            0x008d
+#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER                                                          0x008e
+#define regUVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC3_UVD_JRBC_IB_REF_DATA                                                               0x008f
+#define regUVD_JRBC3_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_CMD                                                               0x0090
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0091
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0092
+#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC3_UVD_JRBC_RB_SIZE                                                                   0x0093
+#define regUVD_JRBC3_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC3_UVD_JRBC_SCRATCH0                                                                  0x0094
+#define regUVD_JRBC3_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc4_uvd_jrbc_dec
+// base address: 0x1e300
+#define regUVD_JRBC4_UVD_JRBC_RB_WPTR                                                                   0x00c0
+#define regUVD_JRBC4_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC4_UVD_JRBC_RB_CNTL                                                                   0x00c1
+#define regUVD_JRBC4_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC4_UVD_JRBC_IB_SIZE                                                                   0x00c2
+#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC4_UVD_JRBC_URGENT_CNTL                                                               0x00c3
+#define regUVD_JRBC4_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC4_UVD_JRBC_RB_REF_DATA                                                               0x00c4
+#define regUVD_JRBC4_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER                                                          0x00c5
+#define regUVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC4_UVD_JRBC_SOFT_RESET                                                                0x00c8
+#define regUVD_JRBC4_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC4_UVD_JRBC_STATUS                                                                    0x00c9
+#define regUVD_JRBC4_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC4_UVD_JRBC_RB_RPTR                                                                   0x00ca
+#define regUVD_JRBC4_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC4_UVD_JRBC_RB_BUF_STATUS                                                             0x00cb
+#define regUVD_JRBC4_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC4_UVD_JRBC_IB_BUF_STATUS                                                             0x00cc
+#define regUVD_JRBC4_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE                                                            0x00cd
+#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER                                                          0x00ce
+#define regUVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC4_UVD_JRBC_IB_REF_DATA                                                               0x00cf
+#define regUVD_JRBC4_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_CMD                                                               0x00d0
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x00d1
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x00d2
+#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC4_UVD_JRBC_RB_SIZE                                                                   0x00d3
+#define regUVD_JRBC4_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC4_UVD_JRBC_SCRATCH0                                                                  0x00d4
+#define regUVD_JRBC4_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc5_uvd_jrbc_dec
+// base address: 0x1e400
+#define regUVD_JRBC5_UVD_JRBC_RB_WPTR                                                                   0x0100
+#define regUVD_JRBC5_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC5_UVD_JRBC_RB_CNTL                                                                   0x0101
+#define regUVD_JRBC5_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC5_UVD_JRBC_IB_SIZE                                                                   0x0102
+#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC5_UVD_JRBC_URGENT_CNTL                                                               0x0103
+#define regUVD_JRBC5_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC5_UVD_JRBC_RB_REF_DATA                                                               0x0104
+#define regUVD_JRBC5_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0105
+#define regUVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC5_UVD_JRBC_SOFT_RESET                                                                0x0108
+#define regUVD_JRBC5_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC5_UVD_JRBC_STATUS                                                                    0x0109
+#define regUVD_JRBC5_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC5_UVD_JRBC_RB_RPTR                                                                   0x010a
+#define regUVD_JRBC5_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC5_UVD_JRBC_RB_BUF_STATUS                                                             0x010b
+#define regUVD_JRBC5_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC5_UVD_JRBC_IB_BUF_STATUS                                                             0x010c
+#define regUVD_JRBC5_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE                                                            0x010d
+#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER                                                          0x010e
+#define regUVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC5_UVD_JRBC_IB_REF_DATA                                                               0x010f
+#define regUVD_JRBC5_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_CMD                                                               0x0110
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0111
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0112
+#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC5_UVD_JRBC_RB_SIZE                                                                   0x0113
+#define regUVD_JRBC5_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC5_UVD_JRBC_SCRATCH0                                                                  0x0114
+#define regUVD_JRBC5_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc6_uvd_jrbc_dec
+// base address: 0x1e500
+#define regUVD_JRBC6_UVD_JRBC_RB_WPTR                                                                   0x0140
+#define regUVD_JRBC6_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC6_UVD_JRBC_RB_CNTL                                                                   0x0141
+#define regUVD_JRBC6_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC6_UVD_JRBC_IB_SIZE                                                                   0x0142
+#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC6_UVD_JRBC_URGENT_CNTL                                                               0x0143
+#define regUVD_JRBC6_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC6_UVD_JRBC_RB_REF_DATA                                                               0x0144
+#define regUVD_JRBC6_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0145
+#define regUVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC6_UVD_JRBC_SOFT_RESET                                                                0x0148
+#define regUVD_JRBC6_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC6_UVD_JRBC_STATUS                                                                    0x0149
+#define regUVD_JRBC6_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC6_UVD_JRBC_RB_RPTR                                                                   0x014a
+#define regUVD_JRBC6_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC6_UVD_JRBC_RB_BUF_STATUS                                                             0x014b
+#define regUVD_JRBC6_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC6_UVD_JRBC_IB_BUF_STATUS                                                             0x014c
+#define regUVD_JRBC6_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE                                                            0x014d
+#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER                                                          0x014e
+#define regUVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC6_UVD_JRBC_IB_REF_DATA                                                               0x014f
+#define regUVD_JRBC6_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_CMD                                                               0x0150
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0151
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0152
+#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC6_UVD_JRBC_RB_SIZE                                                                   0x0153
+#define regUVD_JRBC6_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC6_UVD_JRBC_SCRATCH0                                                                  0x0154
+#define regUVD_JRBC6_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jrbc7_uvd_jrbc_dec
+// base address: 0x1e600
+#define regUVD_JRBC7_UVD_JRBC_RB_WPTR                                                                   0x0180
+#define regUVD_JRBC7_UVD_JRBC_RB_WPTR_BASE_IDX                                                          0
+#define regUVD_JRBC7_UVD_JRBC_RB_CNTL                                                                   0x0181
+#define regUVD_JRBC7_UVD_JRBC_RB_CNTL_BASE_IDX                                                          0
+#define regUVD_JRBC7_UVD_JRBC_IB_SIZE                                                                   0x0182
+#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC7_UVD_JRBC_URGENT_CNTL                                                               0x0183
+#define regUVD_JRBC7_UVD_JRBC_URGENT_CNTL_BASE_IDX                                                      0
+#define regUVD_JRBC7_UVD_JRBC_RB_REF_DATA                                                               0x0184
+#define regUVD_JRBC7_UVD_JRBC_RB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER                                                          0x0185
+#define regUVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC7_UVD_JRBC_SOFT_RESET                                                                0x0188
+#define regUVD_JRBC7_UVD_JRBC_SOFT_RESET_BASE_IDX                                                       0
+#define regUVD_JRBC7_UVD_JRBC_STATUS                                                                    0x0189
+#define regUVD_JRBC7_UVD_JRBC_STATUS_BASE_IDX                                                           0
+#define regUVD_JRBC7_UVD_JRBC_RB_RPTR                                                                   0x018a
+#define regUVD_JRBC7_UVD_JRBC_RB_RPTR_BASE_IDX                                                          0
+#define regUVD_JRBC7_UVD_JRBC_RB_BUF_STATUS                                                             0x018b
+#define regUVD_JRBC7_UVD_JRBC_RB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC7_UVD_JRBC_IB_BUF_STATUS                                                             0x018c
+#define regUVD_JRBC7_UVD_JRBC_IB_BUF_STATUS_BASE_IDX                                                    0
+#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE                                                            0x018d
+#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX                                                   0
+#define regUVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER                                                          0x018e
+#define regUVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX                                                 0
+#define regUVD_JRBC7_UVD_JRBC_IB_REF_DATA                                                               0x018f
+#define regUVD_JRBC7_UVD_JRBC_IB_REF_DATA_BASE_IDX                                                      0
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_CMD                                                               0x0190
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_CMD_BASE_IDX                                                      0
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0                                                       0x0191
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX                                              0
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1                                                       0x0192
+#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX                                              0
+#define regUVD_JRBC7_UVD_JRBC_RB_SIZE                                                                   0x0193
+#define regUVD_JRBC7_UVD_JRBC_RB_SIZE_BASE_IDX                                                          0
+#define regUVD_JRBC7_UVD_JRBC_SCRATCH0                                                                  0x0194
+#define regUVD_JRBC7_UVD_JRBC_SCRATCH0_BASE_IDX                                                         0
+
+
+// addressBlock: aid_uvd0_uvd_jmi1_uvd_jmi_dec
+// base address: 0x1e080
+#define regUVD_JMI1_UVD_JPEG_DEC_PF_CTRL                                                                0x0020
+#define regUVD_JMI1_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI1_UVD_LMI_JRBC_CTRL                                                                   0x0021
+#define regUVD_JMI1_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI1_UVD_LMI_JPEG_CTRL                                                                   0x0022
+#define regUVD_JMI1_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI1_JPEG_LMI_DROP                                                                       0x0023
+#define regUVD_JMI1_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_VMID                                                                0x0024
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_VMID                                                                0x0025
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI1_UVD_LMI_JPEG_VMID                                                                   0x0026
+#define regUVD_JMI1_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x0027
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x0028
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x0029
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x002a
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x002b
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x002c
+#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x002d
+#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI1_UVD_JMI_DEC_SWAP_CNTL                                                               0x002e
+#define regUVD_JMI1_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL                                                                 0x002f
+#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x0030
+#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x0031
+#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x0032
+#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x0033
+#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x0034
+#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x0035
+#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x0036
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x0037
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x0038
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x0039
+#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL2                                                                0x003d
+#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi2_uvd_jmi_dec
+// base address: 0x1e180
+#define regUVD_JMI2_UVD_JPEG_DEC_PF_CTRL                                                                0x0060
+#define regUVD_JMI2_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI2_UVD_LMI_JRBC_CTRL                                                                   0x0061
+#define regUVD_JMI2_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI2_UVD_LMI_JPEG_CTRL                                                                   0x0062
+#define regUVD_JMI2_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI2_JPEG_LMI_DROP                                                                       0x0063
+#define regUVD_JMI2_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_VMID                                                                0x0064
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_VMID                                                                0x0065
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI2_UVD_LMI_JPEG_VMID                                                                   0x0066
+#define regUVD_JMI2_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x0067
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x0068
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x0069
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x006a
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x006b
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x006c
+#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x006d
+#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI2_UVD_JMI_DEC_SWAP_CNTL                                                               0x006e
+#define regUVD_JMI2_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL                                                                 0x006f
+#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x0070
+#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x0071
+#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x0072
+#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x0073
+#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x0074
+#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x0075
+#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x0076
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x0077
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x0078
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x0079
+#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL2                                                                0x007d
+#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi3_uvd_jmi_dec
+// base address: 0x1e280
+#define regUVD_JMI3_UVD_JPEG_DEC_PF_CTRL                                                                0x00a0
+#define regUVD_JMI3_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI3_UVD_LMI_JRBC_CTRL                                                                   0x00a1
+#define regUVD_JMI3_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI3_UVD_LMI_JPEG_CTRL                                                                   0x00a2
+#define regUVD_JMI3_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI3_JPEG_LMI_DROP                                                                       0x00a3
+#define regUVD_JMI3_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_VMID                                                                0x00a4
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_VMID                                                                0x00a5
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI3_UVD_LMI_JPEG_VMID                                                                   0x00a6
+#define regUVD_JMI3_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x00a7
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x00a8
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x00a9
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x00aa
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x00ab
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x00ac
+#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x00ad
+#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI3_UVD_JMI_DEC_SWAP_CNTL                                                               0x00ae
+#define regUVD_JMI3_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL                                                                 0x00af
+#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x00b0
+#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x00b1
+#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x00b2
+#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x00b3
+#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x00b4
+#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x00b5
+#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x00b6
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x00b7
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x00b8
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x00b9
+#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL2                                                                0x00bd
+#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi4_uvd_jmi_dec
+// base address: 0x1e380
+#define regUVD_JMI4_UVD_JPEG_DEC_PF_CTRL                                                                0x00e0
+#define regUVD_JMI4_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI4_UVD_LMI_JRBC_CTRL                                                                   0x00e1
+#define regUVD_JMI4_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI4_UVD_LMI_JPEG_CTRL                                                                   0x00e2
+#define regUVD_JMI4_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI4_JPEG_LMI_DROP                                                                       0x00e3
+#define regUVD_JMI4_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_VMID                                                                0x00e4
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_VMID                                                                0x00e5
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI4_UVD_LMI_JPEG_VMID                                                                   0x00e6
+#define regUVD_JMI4_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x00e7
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x00e8
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x00e9
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x00ea
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x00eb
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x00ec
+#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x00ed
+#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI4_UVD_JMI_DEC_SWAP_CNTL                                                               0x00ee
+#define regUVD_JMI4_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL                                                                 0x00ef
+#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x00f0
+#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x00f1
+#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x00f2
+#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x00f3
+#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x00f4
+#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x00f5
+#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x00f6
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x00f7
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x00f8
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x00f9
+#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL2                                                                0x00fd
+#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi5_uvd_jmi_dec
+// base address: 0x1e480
+#define regUVD_JMI5_UVD_JPEG_DEC_PF_CTRL                                                                0x0120
+#define regUVD_JMI5_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI5_UVD_LMI_JRBC_CTRL                                                                   0x0121
+#define regUVD_JMI5_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI5_UVD_LMI_JPEG_CTRL                                                                   0x0122
+#define regUVD_JMI5_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI5_JPEG_LMI_DROP                                                                       0x0123
+#define regUVD_JMI5_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_VMID                                                                0x0124
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_VMID                                                                0x0125
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI5_UVD_LMI_JPEG_VMID                                                                   0x0126
+#define regUVD_JMI5_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x0127
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x0128
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x0129
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x012a
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x012b
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x012c
+#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x012d
+#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI5_UVD_JMI_DEC_SWAP_CNTL                                                               0x012e
+#define regUVD_JMI5_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL                                                                 0x012f
+#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x0130
+#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x0131
+#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x0132
+#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x0133
+#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x0134
+#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x0135
+#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x0136
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x0137
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x0138
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x0139
+#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL2                                                                0x013d
+#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi6_uvd_jmi_dec
+// base address: 0x1e580
+#define regUVD_JMI6_UVD_JPEG_DEC_PF_CTRL                                                                0x0160
+#define regUVD_JMI6_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI6_UVD_LMI_JRBC_CTRL                                                                   0x0161
+#define regUVD_JMI6_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI6_UVD_LMI_JPEG_CTRL                                                                   0x0162
+#define regUVD_JMI6_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI6_JPEG_LMI_DROP                                                                       0x0163
+#define regUVD_JMI6_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_VMID                                                                0x0164
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_VMID                                                                0x0165
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI6_UVD_LMI_JPEG_VMID                                                                   0x0166
+#define regUVD_JMI6_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x0167
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x0168
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x0169
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x016a
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x016b
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x016c
+#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x016d
+#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI6_UVD_JMI_DEC_SWAP_CNTL                                                               0x016e
+#define regUVD_JMI6_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL                                                                 0x016f
+#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x0170
+#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x0171
+#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x0172
+#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x0173
+#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x0174
+#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x0175
+#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x0176
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x0177
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x0178
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x0179
+#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL2                                                                0x017d
+#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: aid_uvd0_uvd_jmi7_uvd_jmi_dec
+// base address: 0x1e680
+#define regUVD_JMI7_UVD_JPEG_DEC_PF_CTRL                                                                0x01a0
+#define regUVD_JMI7_UVD_JPEG_DEC_PF_CTRL_BASE_IDX                                                       0
+#define regUVD_JMI7_UVD_LMI_JRBC_CTRL                                                                   0x01a1
+#define regUVD_JMI7_UVD_LMI_JRBC_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI7_UVD_LMI_JPEG_CTRL                                                                   0x01a2
+#define regUVD_JMI7_UVD_LMI_JPEG_CTRL_BASE_IDX                                                          0
+#define regUVD_JMI7_JPEG_LMI_DROP                                                                       0x01a3
+#define regUVD_JMI7_JPEG_LMI_DROP_BASE_IDX                                                              0
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_VMID                                                                0x01a4
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_VMID                                                                0x01a5
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_VMID_BASE_IDX                                                       0
+#define regUVD_JMI7_UVD_LMI_JPEG_VMID                                                                   0x01a6
+#define regUVD_JMI7_UVD_LMI_JPEG_VMID_BASE_IDX                                                          0
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW                                            0x01a7
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH                                           0x01a8
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW                                                       0x01a9
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH                                                      0x01aa
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW                                                0x01ab
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH                                               0x01ac
+#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID                                                           0x01ad
+#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX                                                  0
+#define regUVD_JMI7_UVD_JMI_DEC_SWAP_CNTL                                                               0x01ae
+#define regUVD_JMI7_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX                                                      0
+#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL                                                                 0x01af
+#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL_BASE_IDX                                                        0
+#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW                                            0x01b0
+#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX                                   0
+#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH                                           0x01b1
+#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX                                  0
+#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW                                                     0x01b2
+#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX                                            0
+#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH                                                    0x01b3
+#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX                                           0
+#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW                                                    0x01b4
+#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX                                           0
+#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH                                                   0x01b5
+#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX                                          0
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW                                                       0x01b6
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX                                              0
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH                                                      0x01b7
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX                                             0
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW                                                0x01b8
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX                                       0
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH                                               0x01b9
+#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX                                      0
+#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL2                                                                0x01bd
+#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL2_BASE_IDX                                                       0
+
+
+// addressBlock: uvdctxind
+// base address: 0x0
+#define ixUVD_CGC_MEM_CTRL                                                                             0x0000
+#define ixUVD_CGC_CTRL2                                                                                0x0001
+#define ixUVD_CGC_MEM_DS_CTRL                                                                          0x0002
+#define ixUVD_CGC_MEM_SD_CTRL                                                                          0x0003
+#define ixUVD_SW_SCRATCH_00                                                                            0x0004
+#define ixUVD_SW_SCRATCH_01                                                                            0x0005
+#define ixUVD_SW_SCRATCH_02                                                                            0x0006
+#define ixUVD_SW_SCRATCH_03                                                                            0x0007
+#define ixUVD_SW_SCRATCH_04                                                                            0x0008
+#define ixUVD_SW_SCRATCH_05                                                                            0x0009
+#define ixUVD_SW_SCRATCH_06                                                                            0x000a
+#define ixUVD_SW_SCRATCH_07                                                                            0x000b
+#define ixUVD_SW_SCRATCH_08                                                                            0x000c
+#define ixUVD_SW_SCRATCH_09                                                                            0x000d
+#define ixUVD_SW_SCRATCH_10                                                                            0x000e
+#define ixUVD_SW_SCRATCH_11                                                                            0x000f
+#define ixUVD_SW_SCRATCH_12                                                                            0x0010
+#define ixUVD_SW_SCRATCH_13                                                                            0x0011
+#define ixUVD_SW_SCRATCH_14                                                                            0x0012
+#define ixUVD_SW_SCRATCH_15                                                                            0x0013
+#define ixUVD_IH_SEM_CTRL                                                                              0x001e
+
+
+// addressBlock: lmi_adp_indirect
+// base address: 0x0
+#define ixUVD_LMI_CRC0                                                                                 0x0000
+#define ixUVD_LMI_CRC1                                                                                 0x0001
+#define ixUVD_LMI_CRC2                                                                                 0x0002
+#define ixUVD_LMI_CRC3                                                                                 0x0003
+#define ixUVD_LMI_CRC10                                                                                0x000a
+#define ixUVD_LMI_CRC11                                                                                0x000b
+#define ixUVD_LMI_CRC12                                                                                0x000c
+#define ixUVD_LMI_CRC13                                                                                0x000d
+#define ixUVD_LMI_CRC14                                                                                0x000e
+#define ixUVD_LMI_CRC15                                                                                0x000f
+#define ixUVD_LMI_SWAP_CNTL2                                                                           0x0029
+#define ixUVD_MEMCHECK_SYS_INT_EN                                                                      0x0134
+#define ixUVD_MEMCHECK_SYS_INT_STAT                                                                    0x0135
+#define ixUVD_MEMCHECK_SYS_INT_ACK                                                                     0x0136
+#define ixUVD_MEMCHECK_VCPU_INT_EN                                                                     0x0137
+#define ixUVD_MEMCHECK_VCPU_INT_STAT                                                                   0x0138
+#define ixUVD_MEMCHECK_VCPU_INT_ACK                                                                    0x0139
+#define ixUVD_MEMCHECK2_SYS_INT_STAT                                                                   0x0140
+#define ixUVD_MEMCHECK2_SYS_INT_ACK                                                                    0x0141
+#define ixUVD_MEMCHECK2_VCPU_INT_STAT                                                                  0x0142
+#define ixUVD_MEMCHECK2_VCPU_INT_ACK                                                                   0x0143
+
+
+#endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h
new file mode 100644
index 0000000000000..be643ea0e5698
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h
@@ -0,0 +1,10424 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef _vcn_4_0_3_SH_MASK_HEADER
+#define _vcn_4_0_3_SH_MASK_HEADER
+
+
+// addressBlock: aid_uvd0_uvddec
+//UVD_TOP_CTRL
+#define UVD_TOP_CTRL__STANDARD__SHIFT                                                                         0x0
+#define UVD_TOP_CTRL__STD_VERSION__SHIFT                                                                      0x4
+#define UVD_TOP_CTRL__STANDARD_MASK                                                                           0x0000000FL
+#define UVD_TOP_CTRL__STD_VERSION_MASK                                                                        0x00000010L
+//UVD_CGC_GATE
+#define UVD_CGC_GATE__SYS__SHIFT                                                                              0x0
+#define UVD_CGC_GATE__UDEC__SHIFT                                                                             0x1
+#define UVD_CGC_GATE__MPEG2__SHIFT                                                                            0x2
+#define UVD_CGC_GATE__REGS__SHIFT                                                                             0x3
+#define UVD_CGC_GATE__RBC__SHIFT                                                                              0x4
+#define UVD_CGC_GATE__LMI_MC__SHIFT                                                                           0x5
+#define UVD_CGC_GATE__LMI_UMC__SHIFT                                                                          0x6
+#define UVD_CGC_GATE__IDCT__SHIFT                                                                             0x7
+#define UVD_CGC_GATE__MPRD__SHIFT                                                                             0x8
+#define UVD_CGC_GATE__MPC__SHIFT                                                                              0x9
+#define UVD_CGC_GATE__LBSI__SHIFT                                                                             0xa
+#define UVD_CGC_GATE__LRBBM__SHIFT                                                                            0xb
+#define UVD_CGC_GATE__UDEC_RE__SHIFT                                                                          0xc
+#define UVD_CGC_GATE__UDEC_CM__SHIFT                                                                          0xd
+#define UVD_CGC_GATE__UDEC_IT__SHIFT                                                                          0xe
+#define UVD_CGC_GATE__UDEC_DB__SHIFT                                                                          0xf
+#define UVD_CGC_GATE__UDEC_MP__SHIFT                                                                          0x10
+#define UVD_CGC_GATE__WCB__SHIFT                                                                              0x11
+#define UVD_CGC_GATE__VCPU__SHIFT                                                                             0x12
+#define UVD_CGC_GATE__MMSCH__SHIFT                                                                            0x14
+#define UVD_CGC_GATE__LCM0__SHIFT                                                                             0x15
+#define UVD_CGC_GATE__LCM1__SHIFT                                                                             0x16
+#define UVD_CGC_GATE__MIF__SHIFT                                                                              0x17
+#define UVD_CGC_GATE__VREG__SHIFT                                                                             0x18
+#define UVD_CGC_GATE__PE__SHIFT                                                                               0x19
+#define UVD_CGC_GATE__PPU__SHIFT                                                                              0x1a
+#define UVD_CGC_GATE__SYS_MASK                                                                                0x00000001L
+#define UVD_CGC_GATE__UDEC_MASK                                                                               0x00000002L
+#define UVD_CGC_GATE__MPEG2_MASK                                                                              0x00000004L
+#define UVD_CGC_GATE__REGS_MASK                                                                               0x00000008L
+#define UVD_CGC_GATE__RBC_MASK                                                                                0x00000010L
+#define UVD_CGC_GATE__LMI_MC_MASK                                                                             0x00000020L
+#define UVD_CGC_GATE__LMI_UMC_MASK                                                                            0x00000040L
+#define UVD_CGC_GATE__IDCT_MASK                                                                               0x00000080L
+#define UVD_CGC_GATE__MPRD_MASK                                                                               0x00000100L
+#define UVD_CGC_GATE__MPC_MASK                                                                                0x00000200L
+#define UVD_CGC_GATE__LBSI_MASK                                                                               0x00000400L
+#define UVD_CGC_GATE__LRBBM_MASK                                                                              0x00000800L
+#define UVD_CGC_GATE__UDEC_RE_MASK                                                                            0x00001000L
+#define UVD_CGC_GATE__UDEC_CM_MASK                                                                            0x00002000L
+#define UVD_CGC_GATE__UDEC_IT_MASK                                                                            0x00004000L
+#define UVD_CGC_GATE__UDEC_DB_MASK                                                                            0x00008000L
+#define UVD_CGC_GATE__UDEC_MP_MASK                                                                            0x00010000L
+#define UVD_CGC_GATE__WCB_MASK                                                                                0x00020000L
+#define UVD_CGC_GATE__VCPU_MASK                                                                               0x00040000L
+#define UVD_CGC_GATE__MMSCH_MASK                                                                              0x00100000L
+#define UVD_CGC_GATE__LCM0_MASK                                                                               0x00200000L
+#define UVD_CGC_GATE__LCM1_MASK                                                                               0x00400000L
+#define UVD_CGC_GATE__MIF_MASK                                                                                0x00800000L
+#define UVD_CGC_GATE__VREG_MASK                                                                               0x01000000L
+#define UVD_CGC_GATE__PE_MASK                                                                                 0x02000000L
+#define UVD_CGC_GATE__PPU_MASK                                                                                0x04000000L
+//UVD_CGC_CTRL
+#define UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT                                                                   0x0
+#define UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT                                                               0x2
+#define UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT                                                                    0x6
+#define UVD_CGC_CTRL__UDEC_RE_MODE__SHIFT                                                                     0xb
+#define UVD_CGC_CTRL__UDEC_CM_MODE__SHIFT                                                                     0xc
+#define UVD_CGC_CTRL__UDEC_IT_MODE__SHIFT                                                                     0xd
+#define UVD_CGC_CTRL__UDEC_DB_MODE__SHIFT                                                                     0xe
+#define UVD_CGC_CTRL__UDEC_MP_MODE__SHIFT                                                                     0xf
+#define UVD_CGC_CTRL__SYS_MODE__SHIFT                                                                         0x10
+#define UVD_CGC_CTRL__UDEC_MODE__SHIFT                                                                        0x11
+#define UVD_CGC_CTRL__MPEG2_MODE__SHIFT                                                                       0x12
+#define UVD_CGC_CTRL__REGS_MODE__SHIFT                                                                        0x13
+#define UVD_CGC_CTRL__RBC_MODE__SHIFT                                                                         0x14
+#define UVD_CGC_CTRL__LMI_MC_MODE__SHIFT                                                                      0x15
+#define UVD_CGC_CTRL__LMI_UMC_MODE__SHIFT                                                                     0x16
+#define UVD_CGC_CTRL__IDCT_MODE__SHIFT                                                                        0x17
+#define UVD_CGC_CTRL__MPRD_MODE__SHIFT                                                                        0x18
+#define UVD_CGC_CTRL__MPC_MODE__SHIFT                                                                         0x19
+#define UVD_CGC_CTRL__LBSI_MODE__SHIFT                                                                        0x1a
+#define UVD_CGC_CTRL__LRBBM_MODE__SHIFT                                                                       0x1b
+#define UVD_CGC_CTRL__WCB_MODE__SHIFT                                                                         0x1c
+#define UVD_CGC_CTRL__VCPU_MODE__SHIFT                                                                        0x1d
+#define UVD_CGC_CTRL__MMSCH_MODE__SHIFT                                                                       0x1f
+#define UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK                                                                     0x00000001L
+#define UVD_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK                                                                 0x0000003CL
+#define UVD_CGC_CTRL__CLK_OFF_DELAY_MASK                                                                      0x000007C0L
+#define UVD_CGC_CTRL__UDEC_RE_MODE_MASK                                                                       0x00000800L
+#define UVD_CGC_CTRL__UDEC_CM_MODE_MASK                                                                       0x00001000L
+#define UVD_CGC_CTRL__UDEC_IT_MODE_MASK                                                                       0x00002000L
+#define UVD_CGC_CTRL__UDEC_DB_MODE_MASK                                                                       0x00004000L
+#define UVD_CGC_CTRL__UDEC_MP_MODE_MASK                                                                       0x00008000L
+#define UVD_CGC_CTRL__SYS_MODE_MASK                                                                           0x00010000L
+#define UVD_CGC_CTRL__UDEC_MODE_MASK                                                                          0x00020000L
+#define UVD_CGC_CTRL__MPEG2_MODE_MASK                                                                         0x00040000L
+#define UVD_CGC_CTRL__REGS_MODE_MASK                                                                          0x00080000L
+#define UVD_CGC_CTRL__RBC_MODE_MASK                                                                           0x00100000L
+#define UVD_CGC_CTRL__LMI_MC_MODE_MASK                                                                        0x00200000L
+#define UVD_CGC_CTRL__LMI_UMC_MODE_MASK                                                                       0x00400000L
+#define UVD_CGC_CTRL__IDCT_MODE_MASK                                                                          0x00800000L
+#define UVD_CGC_CTRL__MPRD_MODE_MASK                                                                          0x01000000L
+#define UVD_CGC_CTRL__MPC_MODE_MASK                                                                           0x02000000L
+#define UVD_CGC_CTRL__LBSI_MODE_MASK                                                                          0x04000000L
+#define UVD_CGC_CTRL__LRBBM_MODE_MASK                                                                         0x08000000L
+#define UVD_CGC_CTRL__WCB_MODE_MASK                                                                           0x10000000L
+#define UVD_CGC_CTRL__VCPU_MODE_MASK                                                                          0x20000000L
+#define UVD_CGC_CTRL__MMSCH_MODE_MASK                                                                         0x80000000L
+//AVM_SUVD_CGC_GATE
+#define AVM_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define AVM_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define AVM_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define AVM_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define AVM_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define AVM_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define AVM_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define AVM_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define AVM_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define AVM_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define AVM_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define AVM_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define AVM_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define AVM_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define AVM_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define AVM_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define AVM_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define AVM_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define AVM_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define AVM_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define AVM_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define AVM_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define AVM_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define AVM_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define AVM_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define AVM_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define AVM_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define AVM_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define AVM_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define AVM_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define AVM_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define AVM_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define AVM_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define AVM_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define AVM_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define AVM_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define AVM_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define AVM_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define AVM_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define AVM_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define AVM_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define AVM_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define AVM_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define AVM_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define AVM_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define AVM_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define AVM_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define AVM_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define AVM_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define AVM_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define AVM_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define AVM_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define AVM_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define AVM_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define AVM_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define AVM_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define AVM_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define AVM_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define AVM_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define AVM_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define AVM_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define AVM_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define AVM_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define AVM_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//CDEFE_SUVD_CGC_GATE
+#define CDEFE_SUVD_CGC_GATE__SRE__SHIFT                                                                       0x0
+#define CDEFE_SUVD_CGC_GATE__SIT__SHIFT                                                                       0x1
+#define CDEFE_SUVD_CGC_GATE__SMP__SHIFT                                                                       0x2
+#define CDEFE_SUVD_CGC_GATE__SCM__SHIFT                                                                       0x3
+#define CDEFE_SUVD_CGC_GATE__SDB__SHIFT                                                                       0x4
+#define CDEFE_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                  0x5
+#define CDEFE_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                  0x6
+#define CDEFE_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                  0x7
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                  0x8
+#define CDEFE_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                  0x9
+#define CDEFE_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                  0xa
+#define CDEFE_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                  0xb
+#define CDEFE_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                  0xc
+#define CDEFE_SUVD_CGC_GATE__SCLR__SHIFT                                                                      0xd
+#define CDEFE_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                    0xe
+#define CDEFE_SUVD_CGC_GATE__ENT__SHIFT                                                                       0xf
+#define CDEFE_SUVD_CGC_GATE__IME__SHIFT                                                                       0x10
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                              0x11
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                              0x12
+#define CDEFE_SUVD_CGC_GATE__SITE__SHIFT                                                                      0x13
+#define CDEFE_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                   0x14
+#define CDEFE_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                   0x15
+#define CDEFE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                               0x16
+#define CDEFE_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                   0x17
+#define CDEFE_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                  0x18
+#define CDEFE_SUVD_CGC_GATE__EFC__SHIFT                                                                       0x19
+#define CDEFE_SUVD_CGC_GATE__SAOE__SHIFT                                                                      0x1a
+#define CDEFE_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                   0x1b
+#define CDEFE_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                  0x1c
+#define CDEFE_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                  0x1d
+#define CDEFE_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                   0x1e
+#define CDEFE_SUVD_CGC_GATE__SMPA__SHIFT                                                                      0x1f
+#define CDEFE_SUVD_CGC_GATE__SRE_MASK                                                                         0x00000001L
+#define CDEFE_SUVD_CGC_GATE__SIT_MASK                                                                         0x00000002L
+#define CDEFE_SUVD_CGC_GATE__SMP_MASK                                                                         0x00000004L
+#define CDEFE_SUVD_CGC_GATE__SCM_MASK                                                                         0x00000008L
+#define CDEFE_SUVD_CGC_GATE__SDB_MASK                                                                         0x00000010L
+#define CDEFE_SUVD_CGC_GATE__SRE_H264_MASK                                                                    0x00000020L
+#define CDEFE_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                    0x00000040L
+#define CDEFE_SUVD_CGC_GATE__SIT_H264_MASK                                                                    0x00000080L
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                    0x00000100L
+#define CDEFE_SUVD_CGC_GATE__SCM_H264_MASK                                                                    0x00000200L
+#define CDEFE_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                    0x00000400L
+#define CDEFE_SUVD_CGC_GATE__SDB_H264_MASK                                                                    0x00000800L
+#define CDEFE_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                    0x00001000L
+#define CDEFE_SUVD_CGC_GATE__SCLR_MASK                                                                        0x00002000L
+#define CDEFE_SUVD_CGC_GATE__UVD_SC_MASK                                                                      0x00004000L
+#define CDEFE_SUVD_CGC_GATE__ENT_MASK                                                                         0x00008000L
+#define CDEFE_SUVD_CGC_GATE__IME_MASK                                                                         0x00010000L
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                0x00020000L
+#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                0x00040000L
+#define CDEFE_SUVD_CGC_GATE__SITE_MASK                                                                        0x00080000L
+#define CDEFE_SUVD_CGC_GATE__SRE_VP9_MASK                                                                     0x00100000L
+#define CDEFE_SUVD_CGC_GATE__SCM_VP9_MASK                                                                     0x00200000L
+#define CDEFE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                 0x00400000L
+#define CDEFE_SUVD_CGC_GATE__SDB_VP9_MASK                                                                     0x00800000L
+#define CDEFE_SUVD_CGC_GATE__IME_HEVC_MASK                                                                    0x01000000L
+#define CDEFE_SUVD_CGC_GATE__EFC_MASK                                                                         0x02000000L
+#define CDEFE_SUVD_CGC_GATE__SAOE_MASK                                                                        0x04000000L
+#define CDEFE_SUVD_CGC_GATE__SRE_AV1_MASK                                                                     0x08000000L
+#define CDEFE_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                    0x10000000L
+#define CDEFE_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                    0x20000000L
+#define CDEFE_SUVD_CGC_GATE__SCM_AV1_MASK                                                                     0x40000000L
+#define CDEFE_SUVD_CGC_GATE__SMPA_MASK                                                                        0x80000000L
+//EFC_SUVD_CGC_GATE
+#define EFC_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define EFC_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define EFC_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define EFC_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define EFC_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define EFC_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define EFC_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define EFC_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define EFC_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define EFC_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define EFC_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define EFC_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define EFC_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define EFC_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define EFC_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define EFC_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define EFC_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define EFC_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define EFC_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define EFC_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define EFC_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define EFC_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define EFC_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define EFC_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define EFC_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define EFC_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define EFC_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define EFC_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define EFC_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define EFC_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define EFC_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define EFC_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define EFC_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define EFC_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define EFC_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define EFC_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define EFC_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define EFC_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define EFC_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define EFC_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define EFC_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define EFC_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define EFC_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define EFC_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define EFC_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define EFC_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define EFC_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define EFC_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define EFC_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define EFC_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define EFC_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define EFC_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define EFC_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define EFC_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define EFC_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define EFC_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define EFC_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define EFC_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define EFC_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define EFC_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define EFC_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define EFC_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define EFC_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define EFC_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//ENT_SUVD_CGC_GATE
+#define ENT_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define ENT_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define ENT_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define ENT_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define ENT_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define ENT_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define ENT_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define ENT_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define ENT_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define ENT_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define ENT_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define ENT_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define ENT_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define ENT_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define ENT_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define ENT_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define ENT_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define ENT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define ENT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define ENT_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define ENT_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define ENT_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define ENT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define ENT_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define ENT_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define ENT_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define ENT_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define ENT_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define ENT_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define ENT_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define ENT_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define ENT_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define ENT_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define ENT_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define ENT_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define ENT_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define ENT_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define ENT_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define ENT_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define ENT_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define ENT_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define ENT_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define ENT_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define ENT_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define ENT_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define ENT_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define ENT_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define ENT_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define ENT_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define ENT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define ENT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define ENT_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define ENT_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define ENT_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define ENT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define ENT_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define ENT_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define ENT_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define ENT_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define ENT_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define ENT_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define ENT_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define ENT_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define ENT_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//IME_SUVD_CGC_GATE
+#define IME_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define IME_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define IME_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define IME_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define IME_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define IME_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define IME_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define IME_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define IME_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define IME_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define IME_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define IME_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define IME_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define IME_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define IME_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define IME_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define IME_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define IME_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define IME_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define IME_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define IME_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define IME_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define IME_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define IME_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define IME_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define IME_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define IME_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define IME_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define IME_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define IME_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define IME_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define IME_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define IME_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define IME_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define IME_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define IME_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define IME_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define IME_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define IME_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define IME_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define IME_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define IME_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define IME_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define IME_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define IME_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define IME_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define IME_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define IME_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define IME_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define IME_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define IME_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define IME_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define IME_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define IME_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define IME_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define IME_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define IME_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define IME_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define IME_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define IME_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define IME_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define IME_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define IME_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define IME_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//PPU_SUVD_CGC_GATE
+#define PPU_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define PPU_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define PPU_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define PPU_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define PPU_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define PPU_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define PPU_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define PPU_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define PPU_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define PPU_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define PPU_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define PPU_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define PPU_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define PPU_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define PPU_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define PPU_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define PPU_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define PPU_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define PPU_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define PPU_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define PPU_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define PPU_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define PPU_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define PPU_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define PPU_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define PPU_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define PPU_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define PPU_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define PPU_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define PPU_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define PPU_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define PPU_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define PPU_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define PPU_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define PPU_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define PPU_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define PPU_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define PPU_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define PPU_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define PPU_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define PPU_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define PPU_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define PPU_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define PPU_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define PPU_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define PPU_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define PPU_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define PPU_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define PPU_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define PPU_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define PPU_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define PPU_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define PPU_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define PPU_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define PPU_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define PPU_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define PPU_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define PPU_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define PPU_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define PPU_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define PPU_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define PPU_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define PPU_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define PPU_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//SAOE_SUVD_CGC_GATE
+#define SAOE_SUVD_CGC_GATE__SRE__SHIFT                                                                        0x0
+#define SAOE_SUVD_CGC_GATE__SIT__SHIFT                                                                        0x1
+#define SAOE_SUVD_CGC_GATE__SMP__SHIFT                                                                        0x2
+#define SAOE_SUVD_CGC_GATE__SCM__SHIFT                                                                        0x3
+#define SAOE_SUVD_CGC_GATE__SDB__SHIFT                                                                        0x4
+#define SAOE_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                   0x5
+#define SAOE_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                   0x6
+#define SAOE_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                   0x7
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                   0x8
+#define SAOE_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                   0x9
+#define SAOE_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                   0xa
+#define SAOE_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                   0xb
+#define SAOE_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                   0xc
+#define SAOE_SUVD_CGC_GATE__SCLR__SHIFT                                                                       0xd
+#define SAOE_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                     0xe
+#define SAOE_SUVD_CGC_GATE__ENT__SHIFT                                                                        0xf
+#define SAOE_SUVD_CGC_GATE__IME__SHIFT                                                                        0x10
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                               0x11
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                               0x12
+#define SAOE_SUVD_CGC_GATE__SITE__SHIFT                                                                       0x13
+#define SAOE_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                    0x14
+#define SAOE_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                    0x15
+#define SAOE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                0x16
+#define SAOE_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                    0x17
+#define SAOE_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                   0x18
+#define SAOE_SUVD_CGC_GATE__EFC__SHIFT                                                                        0x19
+#define SAOE_SUVD_CGC_GATE__SAOE__SHIFT                                                                       0x1a
+#define SAOE_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                    0x1b
+#define SAOE_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                   0x1c
+#define SAOE_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                   0x1d
+#define SAOE_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                    0x1e
+#define SAOE_SUVD_CGC_GATE__SMPA__SHIFT                                                                       0x1f
+#define SAOE_SUVD_CGC_GATE__SRE_MASK                                                                          0x00000001L
+#define SAOE_SUVD_CGC_GATE__SIT_MASK                                                                          0x00000002L
+#define SAOE_SUVD_CGC_GATE__SMP_MASK                                                                          0x00000004L
+#define SAOE_SUVD_CGC_GATE__SCM_MASK                                                                          0x00000008L
+#define SAOE_SUVD_CGC_GATE__SDB_MASK                                                                          0x00000010L
+#define SAOE_SUVD_CGC_GATE__SRE_H264_MASK                                                                     0x00000020L
+#define SAOE_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                     0x00000040L
+#define SAOE_SUVD_CGC_GATE__SIT_H264_MASK                                                                     0x00000080L
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                     0x00000100L
+#define SAOE_SUVD_CGC_GATE__SCM_H264_MASK                                                                     0x00000200L
+#define SAOE_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                     0x00000400L
+#define SAOE_SUVD_CGC_GATE__SDB_H264_MASK                                                                     0x00000800L
+#define SAOE_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                     0x00001000L
+#define SAOE_SUVD_CGC_GATE__SCLR_MASK                                                                         0x00002000L
+#define SAOE_SUVD_CGC_GATE__UVD_SC_MASK                                                                       0x00004000L
+#define SAOE_SUVD_CGC_GATE__ENT_MASK                                                                          0x00008000L
+#define SAOE_SUVD_CGC_GATE__IME_MASK                                                                          0x00010000L
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                 0x00020000L
+#define SAOE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                 0x00040000L
+#define SAOE_SUVD_CGC_GATE__SITE_MASK                                                                         0x00080000L
+#define SAOE_SUVD_CGC_GATE__SRE_VP9_MASK                                                                      0x00100000L
+#define SAOE_SUVD_CGC_GATE__SCM_VP9_MASK                                                                      0x00200000L
+#define SAOE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                  0x00400000L
+#define SAOE_SUVD_CGC_GATE__SDB_VP9_MASK                                                                      0x00800000L
+#define SAOE_SUVD_CGC_GATE__IME_HEVC_MASK                                                                     0x01000000L
+#define SAOE_SUVD_CGC_GATE__EFC_MASK                                                                          0x02000000L
+#define SAOE_SUVD_CGC_GATE__SAOE_MASK                                                                         0x04000000L
+#define SAOE_SUVD_CGC_GATE__SRE_AV1_MASK                                                                      0x08000000L
+#define SAOE_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                     0x10000000L
+#define SAOE_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                     0x20000000L
+#define SAOE_SUVD_CGC_GATE__SCM_AV1_MASK                                                                      0x40000000L
+#define SAOE_SUVD_CGC_GATE__SMPA_MASK                                                                         0x80000000L
+//SCM_SUVD_CGC_GATE
+#define SCM_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define SCM_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define SCM_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define SCM_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define SCM_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define SCM_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define SCM_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define SCM_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define SCM_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define SCM_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define SCM_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define SCM_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define SCM_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define SCM_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define SCM_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define SCM_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define SCM_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define SCM_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define SCM_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define SCM_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define SCM_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define SCM_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define SCM_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define SCM_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define SCM_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define SCM_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define SCM_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define SCM_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define SCM_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define SCM_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define SCM_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define SCM_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define SCM_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define SCM_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define SCM_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define SCM_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define SCM_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define SCM_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define SCM_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define SCM_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define SCM_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define SCM_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define SCM_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define SCM_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define SCM_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define SCM_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define SCM_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define SCM_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define SCM_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define SCM_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define SCM_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define SCM_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define SCM_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define SCM_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define SCM_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define SCM_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define SCM_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define SCM_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define SCM_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define SCM_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define SCM_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define SCM_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define SCM_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define SCM_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//SDB_SUVD_CGC_GATE
+#define SDB_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define SDB_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define SDB_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define SDB_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define SDB_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define SDB_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define SDB_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define SDB_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define SDB_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define SDB_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define SDB_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define SDB_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define SDB_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define SDB_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define SDB_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define SDB_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define SDB_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define SDB_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define SDB_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define SDB_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define SDB_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define SDB_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define SDB_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define SDB_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define SDB_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define SDB_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define SDB_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define SDB_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define SDB_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define SDB_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define SDB_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define SDB_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define SDB_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define SDB_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define SDB_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define SDB_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define SDB_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define SDB_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define SDB_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define SDB_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define SDB_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define SDB_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define SDB_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define SDB_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define SDB_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define SDB_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define SDB_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define SDB_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define SDB_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define SDB_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define SDB_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define SDB_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define SDB_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define SDB_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define SDB_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define SDB_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define SDB_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define SDB_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define SDB_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define SDB_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define SDB_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define SDB_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define SDB_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define SDB_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//SIT0_NXT_SUVD_CGC_GATE
+#define SIT0_NXT_SUVD_CGC_GATE__SRE__SHIFT                                                                    0x0
+#define SIT0_NXT_SUVD_CGC_GATE__SIT__SHIFT                                                                    0x1
+#define SIT0_NXT_SUVD_CGC_GATE__SMP__SHIFT                                                                    0x2
+#define SIT0_NXT_SUVD_CGC_GATE__SCM__SHIFT                                                                    0x3
+#define SIT0_NXT_SUVD_CGC_GATE__SDB__SHIFT                                                                    0x4
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT                                                               0x5
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                               0x6
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT                                                               0x7
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                               0x8
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT                                                               0x9
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                               0xa
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT                                                               0xb
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                               0xc
+#define SIT0_NXT_SUVD_CGC_GATE__SCLR__SHIFT                                                                   0xd
+#define SIT0_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                 0xe
+#define SIT0_NXT_SUVD_CGC_GATE__ENT__SHIFT                                                                    0xf
+#define SIT0_NXT_SUVD_CGC_GATE__IME__SHIFT                                                                    0x10
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                           0x11
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                           0x12
+#define SIT0_NXT_SUVD_CGC_GATE__SITE__SHIFT                                                                   0x13
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                0x14
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                0x15
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                            0x16
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                0x17
+#define SIT0_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                               0x18
+#define SIT0_NXT_SUVD_CGC_GATE__EFC__SHIFT                                                                    0x19
+#define SIT0_NXT_SUVD_CGC_GATE__SAOE__SHIFT                                                                   0x1a
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                0x1b
+#define SIT0_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT0_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                0x1e
+#define SIT0_NXT_SUVD_CGC_GATE__SMPA__SHIFT                                                                   0x1f
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_MASK                                                                      0x00000001L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_MASK                                                                      0x00000002L
+#define SIT0_NXT_SUVD_CGC_GATE__SMP_MASK                                                                      0x00000004L
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_MASK                                                                      0x00000008L
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_MASK                                                                      0x00000010L
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_H264_MASK                                                                 0x00000020L
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                 0x00000040L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_H264_MASK                                                                 0x00000080L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                 0x00000100L
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_H264_MASK                                                                 0x00000200L
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                 0x00000400L
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_H264_MASK                                                                 0x00000800L
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                 0x00001000L
+#define SIT0_NXT_SUVD_CGC_GATE__SCLR_MASK                                                                     0x00002000L
+#define SIT0_NXT_SUVD_CGC_GATE__UVD_SC_MASK                                                                   0x00004000L
+#define SIT0_NXT_SUVD_CGC_GATE__ENT_MASK                                                                      0x00008000L
+#define SIT0_NXT_SUVD_CGC_GATE__IME_MASK                                                                      0x00010000L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                             0x00020000L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                             0x00040000L
+#define SIT0_NXT_SUVD_CGC_GATE__SITE_MASK                                                                     0x00080000L
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_VP9_MASK                                                                  0x00100000L
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_VP9_MASK                                                                  0x00200000L
+#define SIT0_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                              0x00400000L
+#define SIT0_NXT_SUVD_CGC_GATE__SDB_VP9_MASK                                                                  0x00800000L
+#define SIT0_NXT_SUVD_CGC_GATE__IME_HEVC_MASK                                                                 0x01000000L
+#define SIT0_NXT_SUVD_CGC_GATE__EFC_MASK                                                                      0x02000000L
+#define SIT0_NXT_SUVD_CGC_GATE__SAOE_MASK                                                                     0x04000000L
+#define SIT0_NXT_SUVD_CGC_GATE__SRE_AV1_MASK                                                                  0x08000000L
+#define SIT0_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT0_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT0_NXT_SUVD_CGC_GATE__SCM_AV1_MASK                                                                  0x40000000L
+#define SIT0_NXT_SUVD_CGC_GATE__SMPA_MASK                                                                     0x80000000L
+//SIT1_NXT_SUVD_CGC_GATE
+#define SIT1_NXT_SUVD_CGC_GATE__SRE__SHIFT                                                                    0x0
+#define SIT1_NXT_SUVD_CGC_GATE__SIT__SHIFT                                                                    0x1
+#define SIT1_NXT_SUVD_CGC_GATE__SMP__SHIFT                                                                    0x2
+#define SIT1_NXT_SUVD_CGC_GATE__SCM__SHIFT                                                                    0x3
+#define SIT1_NXT_SUVD_CGC_GATE__SDB__SHIFT                                                                    0x4
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT                                                               0x5
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                               0x6
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT                                                               0x7
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                               0x8
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT                                                               0x9
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                               0xa
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT                                                               0xb
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                               0xc
+#define SIT1_NXT_SUVD_CGC_GATE__SCLR__SHIFT                                                                   0xd
+#define SIT1_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                 0xe
+#define SIT1_NXT_SUVD_CGC_GATE__ENT__SHIFT                                                                    0xf
+#define SIT1_NXT_SUVD_CGC_GATE__IME__SHIFT                                                                    0x10
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                           0x11
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                           0x12
+#define SIT1_NXT_SUVD_CGC_GATE__SITE__SHIFT                                                                   0x13
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                0x14
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                0x15
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                            0x16
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                0x17
+#define SIT1_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                               0x18
+#define SIT1_NXT_SUVD_CGC_GATE__EFC__SHIFT                                                                    0x19
+#define SIT1_NXT_SUVD_CGC_GATE__SAOE__SHIFT                                                                   0x1a
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                0x1b
+#define SIT1_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT1_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                0x1e
+#define SIT1_NXT_SUVD_CGC_GATE__SMPA__SHIFT                                                                   0x1f
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_MASK                                                                      0x00000001L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_MASK                                                                      0x00000002L
+#define SIT1_NXT_SUVD_CGC_GATE__SMP_MASK                                                                      0x00000004L
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_MASK                                                                      0x00000008L
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_MASK                                                                      0x00000010L
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_H264_MASK                                                                 0x00000020L
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                 0x00000040L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_H264_MASK                                                                 0x00000080L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                 0x00000100L
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_H264_MASK                                                                 0x00000200L
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                 0x00000400L
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_H264_MASK                                                                 0x00000800L
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                 0x00001000L
+#define SIT1_NXT_SUVD_CGC_GATE__SCLR_MASK                                                                     0x00002000L
+#define SIT1_NXT_SUVD_CGC_GATE__UVD_SC_MASK                                                                   0x00004000L
+#define SIT1_NXT_SUVD_CGC_GATE__ENT_MASK                                                                      0x00008000L
+#define SIT1_NXT_SUVD_CGC_GATE__IME_MASK                                                                      0x00010000L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                             0x00020000L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                             0x00040000L
+#define SIT1_NXT_SUVD_CGC_GATE__SITE_MASK                                                                     0x00080000L
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_VP9_MASK                                                                  0x00100000L
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_VP9_MASK                                                                  0x00200000L
+#define SIT1_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                              0x00400000L
+#define SIT1_NXT_SUVD_CGC_GATE__SDB_VP9_MASK                                                                  0x00800000L
+#define SIT1_NXT_SUVD_CGC_GATE__IME_HEVC_MASK                                                                 0x01000000L
+#define SIT1_NXT_SUVD_CGC_GATE__EFC_MASK                                                                      0x02000000L
+#define SIT1_NXT_SUVD_CGC_GATE__SAOE_MASK                                                                     0x04000000L
+#define SIT1_NXT_SUVD_CGC_GATE__SRE_AV1_MASK                                                                  0x08000000L
+#define SIT1_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT1_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT1_NXT_SUVD_CGC_GATE__SCM_AV1_MASK                                                                  0x40000000L
+#define SIT1_NXT_SUVD_CGC_GATE__SMPA_MASK                                                                     0x80000000L
+//SIT2_NXT_SUVD_CGC_GATE
+#define SIT2_NXT_SUVD_CGC_GATE__SRE__SHIFT                                                                    0x0
+#define SIT2_NXT_SUVD_CGC_GATE__SIT__SHIFT                                                                    0x1
+#define SIT2_NXT_SUVD_CGC_GATE__SMP__SHIFT                                                                    0x2
+#define SIT2_NXT_SUVD_CGC_GATE__SCM__SHIFT                                                                    0x3
+#define SIT2_NXT_SUVD_CGC_GATE__SDB__SHIFT                                                                    0x4
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT                                                               0x5
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                               0x6
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT                                                               0x7
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                               0x8
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT                                                               0x9
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                               0xa
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT                                                               0xb
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                               0xc
+#define SIT2_NXT_SUVD_CGC_GATE__SCLR__SHIFT                                                                   0xd
+#define SIT2_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                 0xe
+#define SIT2_NXT_SUVD_CGC_GATE__ENT__SHIFT                                                                    0xf
+#define SIT2_NXT_SUVD_CGC_GATE__IME__SHIFT                                                                    0x10
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                           0x11
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                           0x12
+#define SIT2_NXT_SUVD_CGC_GATE__SITE__SHIFT                                                                   0x13
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                0x14
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                0x15
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                            0x16
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                0x17
+#define SIT2_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                               0x18
+#define SIT2_NXT_SUVD_CGC_GATE__EFC__SHIFT                                                                    0x19
+#define SIT2_NXT_SUVD_CGC_GATE__SAOE__SHIFT                                                                   0x1a
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                0x1b
+#define SIT2_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT2_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                0x1e
+#define SIT2_NXT_SUVD_CGC_GATE__SMPA__SHIFT                                                                   0x1f
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_MASK                                                                      0x00000001L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_MASK                                                                      0x00000002L
+#define SIT2_NXT_SUVD_CGC_GATE__SMP_MASK                                                                      0x00000004L
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_MASK                                                                      0x00000008L
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_MASK                                                                      0x00000010L
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_H264_MASK                                                                 0x00000020L
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                 0x00000040L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_H264_MASK                                                                 0x00000080L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                 0x00000100L
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_H264_MASK                                                                 0x00000200L
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                 0x00000400L
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_H264_MASK                                                                 0x00000800L
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                 0x00001000L
+#define SIT2_NXT_SUVD_CGC_GATE__SCLR_MASK                                                                     0x00002000L
+#define SIT2_NXT_SUVD_CGC_GATE__UVD_SC_MASK                                                                   0x00004000L
+#define SIT2_NXT_SUVD_CGC_GATE__ENT_MASK                                                                      0x00008000L
+#define SIT2_NXT_SUVD_CGC_GATE__IME_MASK                                                                      0x00010000L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                             0x00020000L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                             0x00040000L
+#define SIT2_NXT_SUVD_CGC_GATE__SITE_MASK                                                                     0x00080000L
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_VP9_MASK                                                                  0x00100000L
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_VP9_MASK                                                                  0x00200000L
+#define SIT2_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                              0x00400000L
+#define SIT2_NXT_SUVD_CGC_GATE__SDB_VP9_MASK                                                                  0x00800000L
+#define SIT2_NXT_SUVD_CGC_GATE__IME_HEVC_MASK                                                                 0x01000000L
+#define SIT2_NXT_SUVD_CGC_GATE__EFC_MASK                                                                      0x02000000L
+#define SIT2_NXT_SUVD_CGC_GATE__SAOE_MASK                                                                     0x04000000L
+#define SIT2_NXT_SUVD_CGC_GATE__SRE_AV1_MASK                                                                  0x08000000L
+#define SIT2_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT2_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT2_NXT_SUVD_CGC_GATE__SCM_AV1_MASK                                                                  0x40000000L
+#define SIT2_NXT_SUVD_CGC_GATE__SMPA_MASK                                                                     0x80000000L
+//SIT_SUVD_CGC_GATE
+#define SIT_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define SIT_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define SIT_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define SIT_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define SIT_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define SIT_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define SIT_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define SIT_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define SIT_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define SIT_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define SIT_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define SIT_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define SIT_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define SIT_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define SIT_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define SIT_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define SIT_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define SIT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define SIT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define SIT_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define SIT_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define SIT_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define SIT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define SIT_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define SIT_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define SIT_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define SIT_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define SIT_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define SIT_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define SIT_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define SIT_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define SIT_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define SIT_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define SIT_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define SIT_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define SIT_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define SIT_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define SIT_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define SIT_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define SIT_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define SIT_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define SIT_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define SIT_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define SIT_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define SIT_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define SIT_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define SIT_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define SIT_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define SIT_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define SIT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define SIT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define SIT_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define SIT_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define SIT_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define SIT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define SIT_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define SIT_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define SIT_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define SIT_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define SIT_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define SIT_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define SIT_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define SIT_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define SIT_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//SMPA_SUVD_CGC_GATE
+#define SMPA_SUVD_CGC_GATE__SRE__SHIFT                                                                        0x0
+#define SMPA_SUVD_CGC_GATE__SIT__SHIFT                                                                        0x1
+#define SMPA_SUVD_CGC_GATE__SMP__SHIFT                                                                        0x2
+#define SMPA_SUVD_CGC_GATE__SCM__SHIFT                                                                        0x3
+#define SMPA_SUVD_CGC_GATE__SDB__SHIFT                                                                        0x4
+#define SMPA_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                   0x5
+#define SMPA_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                   0x6
+#define SMPA_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                   0x7
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                   0x8
+#define SMPA_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                   0x9
+#define SMPA_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                   0xa
+#define SMPA_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                   0xb
+#define SMPA_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                   0xc
+#define SMPA_SUVD_CGC_GATE__SCLR__SHIFT                                                                       0xd
+#define SMPA_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                     0xe
+#define SMPA_SUVD_CGC_GATE__ENT__SHIFT                                                                        0xf
+#define SMPA_SUVD_CGC_GATE__IME__SHIFT                                                                        0x10
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                               0x11
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                               0x12
+#define SMPA_SUVD_CGC_GATE__SITE__SHIFT                                                                       0x13
+#define SMPA_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                    0x14
+#define SMPA_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                    0x15
+#define SMPA_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                0x16
+#define SMPA_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                    0x17
+#define SMPA_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                   0x18
+#define SMPA_SUVD_CGC_GATE__EFC__SHIFT                                                                        0x19
+#define SMPA_SUVD_CGC_GATE__SAOE__SHIFT                                                                       0x1a
+#define SMPA_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                    0x1b
+#define SMPA_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                   0x1c
+#define SMPA_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                   0x1d
+#define SMPA_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                    0x1e
+#define SMPA_SUVD_CGC_GATE__SMPA__SHIFT                                                                       0x1f
+#define SMPA_SUVD_CGC_GATE__SRE_MASK                                                                          0x00000001L
+#define SMPA_SUVD_CGC_GATE__SIT_MASK                                                                          0x00000002L
+#define SMPA_SUVD_CGC_GATE__SMP_MASK                                                                          0x00000004L
+#define SMPA_SUVD_CGC_GATE__SCM_MASK                                                                          0x00000008L
+#define SMPA_SUVD_CGC_GATE__SDB_MASK                                                                          0x00000010L
+#define SMPA_SUVD_CGC_GATE__SRE_H264_MASK                                                                     0x00000020L
+#define SMPA_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                     0x00000040L
+#define SMPA_SUVD_CGC_GATE__SIT_H264_MASK                                                                     0x00000080L
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                     0x00000100L
+#define SMPA_SUVD_CGC_GATE__SCM_H264_MASK                                                                     0x00000200L
+#define SMPA_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                     0x00000400L
+#define SMPA_SUVD_CGC_GATE__SDB_H264_MASK                                                                     0x00000800L
+#define SMPA_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                     0x00001000L
+#define SMPA_SUVD_CGC_GATE__SCLR_MASK                                                                         0x00002000L
+#define SMPA_SUVD_CGC_GATE__UVD_SC_MASK                                                                       0x00004000L
+#define SMPA_SUVD_CGC_GATE__ENT_MASK                                                                          0x00008000L
+#define SMPA_SUVD_CGC_GATE__IME_MASK                                                                          0x00010000L
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                 0x00020000L
+#define SMPA_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                 0x00040000L
+#define SMPA_SUVD_CGC_GATE__SITE_MASK                                                                         0x00080000L
+#define SMPA_SUVD_CGC_GATE__SRE_VP9_MASK                                                                      0x00100000L
+#define SMPA_SUVD_CGC_GATE__SCM_VP9_MASK                                                                      0x00200000L
+#define SMPA_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                  0x00400000L
+#define SMPA_SUVD_CGC_GATE__SDB_VP9_MASK                                                                      0x00800000L
+#define SMPA_SUVD_CGC_GATE__IME_HEVC_MASK                                                                     0x01000000L
+#define SMPA_SUVD_CGC_GATE__EFC_MASK                                                                          0x02000000L
+#define SMPA_SUVD_CGC_GATE__SAOE_MASK                                                                         0x04000000L
+#define SMPA_SUVD_CGC_GATE__SRE_AV1_MASK                                                                      0x08000000L
+#define SMPA_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                     0x10000000L
+#define SMPA_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                     0x20000000L
+#define SMPA_SUVD_CGC_GATE__SCM_AV1_MASK                                                                      0x40000000L
+#define SMPA_SUVD_CGC_GATE__SMPA_MASK                                                                         0x80000000L
+//SMP_SUVD_CGC_GATE
+#define SMP_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define SMP_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define SMP_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define SMP_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define SMP_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define SMP_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define SMP_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define SMP_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define SMP_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define SMP_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define SMP_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define SMP_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define SMP_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define SMP_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define SMP_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define SMP_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define SMP_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define SMP_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define SMP_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define SMP_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define SMP_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define SMP_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define SMP_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define SMP_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define SMP_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define SMP_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define SMP_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define SMP_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define SMP_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define SMP_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define SMP_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define SMP_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define SMP_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define SMP_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define SMP_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define SMP_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define SMP_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define SMP_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define SMP_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define SMP_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define SMP_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define SMP_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define SMP_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define SMP_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define SMP_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define SMP_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define SMP_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define SMP_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define SMP_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define SMP_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define SMP_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define SMP_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define SMP_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define SMP_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define SMP_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define SMP_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define SMP_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define SMP_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define SMP_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define SMP_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define SMP_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define SMP_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define SMP_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define SMP_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//SRE_SUVD_CGC_GATE
+#define SRE_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define SRE_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define SRE_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define SRE_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define SRE_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define SRE_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define SRE_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define SRE_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define SRE_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define SRE_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define SRE_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define SRE_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define SRE_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define SRE_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define SRE_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define SRE_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define SRE_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define SRE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define SRE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define SRE_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define SRE_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define SRE_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define SRE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define SRE_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define SRE_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define SRE_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define SRE_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define SRE_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define SRE_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define SRE_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define SRE_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define SRE_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define SRE_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define SRE_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define SRE_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define SRE_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define SRE_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define SRE_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define SRE_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define SRE_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define SRE_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define SRE_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define SRE_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define SRE_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define SRE_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define SRE_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define SRE_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define SRE_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define SRE_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define SRE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define SRE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define SRE_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define SRE_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define SRE_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define SRE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define SRE_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define SRE_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define SRE_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define SRE_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define SRE_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define SRE_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define SRE_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define SRE_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define SRE_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//UVD_MPBE0_SUVD_CGC_GATE
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE__SHIFT                                                                   0x0
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT__SHIFT                                                                   0x1
+#define UVD_MPBE0_SUVD_CGC_GATE__SMP__SHIFT                                                                   0x2
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM__SHIFT                                                                   0x3
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB__SHIFT                                                                   0x4
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_H264__SHIFT                                                              0x5
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                              0x6
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_H264__SHIFT                                                              0x7
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                              0x8
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_H264__SHIFT                                                              0x9
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                              0xa
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_H264__SHIFT                                                              0xb
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                              0xc
+#define UVD_MPBE0_SUVD_CGC_GATE__SCLR__SHIFT                                                                  0xd
+#define UVD_MPBE0_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                0xe
+#define UVD_MPBE0_SUVD_CGC_GATE__ENT__SHIFT                                                                   0xf
+#define UVD_MPBE0_SUVD_CGC_GATE__IME__SHIFT                                                                   0x10
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                          0x11
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                          0x12
+#define UVD_MPBE0_SUVD_CGC_GATE__SITE__SHIFT                                                                  0x13
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                               0x14
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                               0x15
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                           0x16
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                               0x17
+#define UVD_MPBE0_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                              0x18
+#define UVD_MPBE0_SUVD_CGC_GATE__EFC__SHIFT                                                                   0x19
+#define UVD_MPBE0_SUVD_CGC_GATE__SAOE__SHIFT                                                                  0x1a
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                               0x1b
+#define UVD_MPBE0_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                              0x1c
+#define UVD_MPBE0_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                              0x1d
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                               0x1e
+#define UVD_MPBE0_SUVD_CGC_GATE__SMPA__SHIFT                                                                  0x1f
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_MASK                                                                     0x00000001L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_MASK                                                                     0x00000002L
+#define UVD_MPBE0_SUVD_CGC_GATE__SMP_MASK                                                                     0x00000004L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_MASK                                                                     0x00000008L
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_MASK                                                                     0x00000010L
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_H264_MASK                                                                0x00000020L
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                0x00000040L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_H264_MASK                                                                0x00000080L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                0x00000100L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_H264_MASK                                                                0x00000200L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                0x00000400L
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_H264_MASK                                                                0x00000800L
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                0x00001000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCLR_MASK                                                                    0x00002000L
+#define UVD_MPBE0_SUVD_CGC_GATE__UVD_SC_MASK                                                                  0x00004000L
+#define UVD_MPBE0_SUVD_CGC_GATE__ENT_MASK                                                                     0x00008000L
+#define UVD_MPBE0_SUVD_CGC_GATE__IME_MASK                                                                     0x00010000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                            0x00020000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                            0x00040000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SITE_MASK                                                                    0x00080000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_VP9_MASK                                                                 0x00100000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_VP9_MASK                                                                 0x00200000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                             0x00400000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SDB_VP9_MASK                                                                 0x00800000L
+#define UVD_MPBE0_SUVD_CGC_GATE__IME_HEVC_MASK                                                                0x01000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__EFC_MASK                                                                     0x02000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SAOE_MASK                                                                    0x04000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SRE_AV1_MASK                                                                 0x08000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                0x10000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                0x20000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SCM_AV1_MASK                                                                 0x40000000L
+#define UVD_MPBE0_SUVD_CGC_GATE__SMPA_MASK                                                                    0x80000000L
+//UVD_MPBE1_SUVD_CGC_GATE
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE__SHIFT                                                                   0x0
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT__SHIFT                                                                   0x1
+#define UVD_MPBE1_SUVD_CGC_GATE__SMP__SHIFT                                                                   0x2
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM__SHIFT                                                                   0x3
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB__SHIFT                                                                   0x4
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_H264__SHIFT                                                              0x5
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                              0x6
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_H264__SHIFT                                                              0x7
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                              0x8
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_H264__SHIFT                                                              0x9
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                              0xa
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_H264__SHIFT                                                              0xb
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                              0xc
+#define UVD_MPBE1_SUVD_CGC_GATE__SCLR__SHIFT                                                                  0xd
+#define UVD_MPBE1_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                0xe
+#define UVD_MPBE1_SUVD_CGC_GATE__ENT__SHIFT                                                                   0xf
+#define UVD_MPBE1_SUVD_CGC_GATE__IME__SHIFT                                                                   0x10
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                          0x11
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                          0x12
+#define UVD_MPBE1_SUVD_CGC_GATE__SITE__SHIFT                                                                  0x13
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                               0x14
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                               0x15
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                           0x16
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                               0x17
+#define UVD_MPBE1_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                              0x18
+#define UVD_MPBE1_SUVD_CGC_GATE__EFC__SHIFT                                                                   0x19
+#define UVD_MPBE1_SUVD_CGC_GATE__SAOE__SHIFT                                                                  0x1a
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                               0x1b
+#define UVD_MPBE1_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                              0x1c
+#define UVD_MPBE1_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                              0x1d
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                               0x1e
+#define UVD_MPBE1_SUVD_CGC_GATE__SMPA__SHIFT                                                                  0x1f
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_MASK                                                                     0x00000001L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_MASK                                                                     0x00000002L
+#define UVD_MPBE1_SUVD_CGC_GATE__SMP_MASK                                                                     0x00000004L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_MASK                                                                     0x00000008L
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_MASK                                                                     0x00000010L
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_H264_MASK                                                                0x00000020L
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                0x00000040L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_H264_MASK                                                                0x00000080L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                0x00000100L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_H264_MASK                                                                0x00000200L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                0x00000400L
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_H264_MASK                                                                0x00000800L
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                0x00001000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCLR_MASK                                                                    0x00002000L
+#define UVD_MPBE1_SUVD_CGC_GATE__UVD_SC_MASK                                                                  0x00004000L
+#define UVD_MPBE1_SUVD_CGC_GATE__ENT_MASK                                                                     0x00008000L
+#define UVD_MPBE1_SUVD_CGC_GATE__IME_MASK                                                                     0x00010000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                            0x00020000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                            0x00040000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SITE_MASK                                                                    0x00080000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_VP9_MASK                                                                 0x00100000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_VP9_MASK                                                                 0x00200000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                             0x00400000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SDB_VP9_MASK                                                                 0x00800000L
+#define UVD_MPBE1_SUVD_CGC_GATE__IME_HEVC_MASK                                                                0x01000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__EFC_MASK                                                                     0x02000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SAOE_MASK                                                                    0x04000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SRE_AV1_MASK                                                                 0x08000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                0x10000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                0x20000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SCM_AV1_MASK                                                                 0x40000000L
+#define UVD_MPBE1_SUVD_CGC_GATE__SMPA_MASK                                                                    0x80000000L
+//UVD_SUVD_CGC_GATE
+#define UVD_SUVD_CGC_GATE__SRE__SHIFT                                                                         0x0
+#define UVD_SUVD_CGC_GATE__SIT__SHIFT                                                                         0x1
+#define UVD_SUVD_CGC_GATE__SMP__SHIFT                                                                         0x2
+#define UVD_SUVD_CGC_GATE__SCM__SHIFT                                                                         0x3
+#define UVD_SUVD_CGC_GATE__SDB__SHIFT                                                                         0x4
+#define UVD_SUVD_CGC_GATE__SRE_H264__SHIFT                                                                    0x5
+#define UVD_SUVD_CGC_GATE__SRE_HEVC__SHIFT                                                                    0x6
+#define UVD_SUVD_CGC_GATE__SIT_H264__SHIFT                                                                    0x7
+#define UVD_SUVD_CGC_GATE__SIT_HEVC__SHIFT                                                                    0x8
+#define UVD_SUVD_CGC_GATE__SCM_H264__SHIFT                                                                    0x9
+#define UVD_SUVD_CGC_GATE__SCM_HEVC__SHIFT                                                                    0xa
+#define UVD_SUVD_CGC_GATE__SDB_H264__SHIFT                                                                    0xb
+#define UVD_SUVD_CGC_GATE__SDB_HEVC__SHIFT                                                                    0xc
+#define UVD_SUVD_CGC_GATE__SCLR__SHIFT                                                                        0xd
+#define UVD_SUVD_CGC_GATE__UVD_SC__SHIFT                                                                      0xe
+#define UVD_SUVD_CGC_GATE__ENT__SHIFT                                                                         0xf
+#define UVD_SUVD_CGC_GATE__IME__SHIFT                                                                         0x10
+#define UVD_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT                                                                0x11
+#define UVD_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT                                                                0x12
+#define UVD_SUVD_CGC_GATE__SITE__SHIFT                                                                        0x13
+#define UVD_SUVD_CGC_GATE__SRE_VP9__SHIFT                                                                     0x14
+#define UVD_SUVD_CGC_GATE__SCM_VP9__SHIFT                                                                     0x15
+#define UVD_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT                                                                 0x16
+#define UVD_SUVD_CGC_GATE__SDB_VP9__SHIFT                                                                     0x17
+#define UVD_SUVD_CGC_GATE__IME_HEVC__SHIFT                                                                    0x18
+#define UVD_SUVD_CGC_GATE__EFC__SHIFT                                                                         0x19
+#define UVD_SUVD_CGC_GATE__SAOE__SHIFT                                                                        0x1a
+#define UVD_SUVD_CGC_GATE__SRE_AV1__SHIFT                                                                     0x1b
+#define UVD_SUVD_CGC_GATE__FBC_PCLK__SHIFT                                                                    0x1c
+#define UVD_SUVD_CGC_GATE__FBC_CCLK__SHIFT                                                                    0x1d
+#define UVD_SUVD_CGC_GATE__SCM_AV1__SHIFT                                                                     0x1e
+#define UVD_SUVD_CGC_GATE__SMPA__SHIFT                                                                        0x1f
+#define UVD_SUVD_CGC_GATE__SRE_MASK                                                                           0x00000001L
+#define UVD_SUVD_CGC_GATE__SIT_MASK                                                                           0x00000002L
+#define UVD_SUVD_CGC_GATE__SMP_MASK                                                                           0x00000004L
+#define UVD_SUVD_CGC_GATE__SCM_MASK                                                                           0x00000008L
+#define UVD_SUVD_CGC_GATE__SDB_MASK                                                                           0x00000010L
+#define UVD_SUVD_CGC_GATE__SRE_H264_MASK                                                                      0x00000020L
+#define UVD_SUVD_CGC_GATE__SRE_HEVC_MASK                                                                      0x00000040L
+#define UVD_SUVD_CGC_GATE__SIT_H264_MASK                                                                      0x00000080L
+#define UVD_SUVD_CGC_GATE__SIT_HEVC_MASK                                                                      0x00000100L
+#define UVD_SUVD_CGC_GATE__SCM_H264_MASK                                                                      0x00000200L
+#define UVD_SUVD_CGC_GATE__SCM_HEVC_MASK                                                                      0x00000400L
+#define UVD_SUVD_CGC_GATE__SDB_H264_MASK                                                                      0x00000800L
+#define UVD_SUVD_CGC_GATE__SDB_HEVC_MASK                                                                      0x00001000L
+#define UVD_SUVD_CGC_GATE__SCLR_MASK                                                                          0x00002000L
+#define UVD_SUVD_CGC_GATE__UVD_SC_MASK                                                                        0x00004000L
+#define UVD_SUVD_CGC_GATE__ENT_MASK                                                                           0x00008000L
+#define UVD_SUVD_CGC_GATE__IME_MASK                                                                           0x00010000L
+#define UVD_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK                                                                  0x00020000L
+#define UVD_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK                                                                  0x00040000L
+#define UVD_SUVD_CGC_GATE__SITE_MASK                                                                          0x00080000L
+#define UVD_SUVD_CGC_GATE__SRE_VP9_MASK                                                                       0x00100000L
+#define UVD_SUVD_CGC_GATE__SCM_VP9_MASK                                                                       0x00200000L
+#define UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK                                                                   0x00400000L
+#define UVD_SUVD_CGC_GATE__SDB_VP9_MASK                                                                       0x00800000L
+#define UVD_SUVD_CGC_GATE__IME_HEVC_MASK                                                                      0x01000000L
+#define UVD_SUVD_CGC_GATE__EFC_MASK                                                                           0x02000000L
+#define UVD_SUVD_CGC_GATE__SAOE_MASK                                                                          0x04000000L
+#define UVD_SUVD_CGC_GATE__SRE_AV1_MASK                                                                       0x08000000L
+#define UVD_SUVD_CGC_GATE__FBC_PCLK_MASK                                                                      0x10000000L
+#define UVD_SUVD_CGC_GATE__FBC_CCLK_MASK                                                                      0x20000000L
+#define UVD_SUVD_CGC_GATE__SCM_AV1_MASK                                                                       0x40000000L
+#define UVD_SUVD_CGC_GATE__SMPA_MASK                                                                          0x80000000L
+//AVM_SUVD_CGC_GATE2
+#define AVM_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define AVM_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define AVM_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define AVM_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define AVM_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define AVM_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define AVM_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define AVM_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define AVM_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define AVM_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define AVM_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define AVM_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define AVM_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define AVM_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define AVM_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define AVM_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define AVM_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define AVM_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define AVM_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//CDEFE_SUVD_CGC_GATE2
+#define CDEFE_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                    0x0
+#define CDEFE_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                    0x1
+#define CDEFE_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                  0x2
+#define CDEFE_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                  0x3
+#define CDEFE_SUVD_CGC_GATE2__MPC1__SHIFT                                                                     0x4
+#define CDEFE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                              0x5
+#define CDEFE_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                    0x6
+#define CDEFE_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                    0x7
+#define CDEFE_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                    0x8
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                              0x9
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                              0xa
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                              0xb
+#define CDEFE_SUVD_CGC_GATE2__MPBE0_MASK                                                                      0x00000001L
+#define CDEFE_SUVD_CGC_GATE2__MPBE1_MASK                                                                      0x00000002L
+#define CDEFE_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                    0x00000004L
+#define CDEFE_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                    0x00000008L
+#define CDEFE_SUVD_CGC_GATE2__MPC1_MASK                                                                       0x00000010L
+#define CDEFE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                0x00000020L
+#define CDEFE_SUVD_CGC_GATE2__CDEFE_MASK                                                                      0x00000040L
+#define CDEFE_SUVD_CGC_GATE2__AVM_0_MASK                                                                      0x00000080L
+#define CDEFE_SUVD_CGC_GATE2__AVM_1_MASK                                                                      0x00000100L
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                0x00000200L
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                0x00000400L
+#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                0x00000800L
+//DBR_SUVD_CGC_GATE2
+#define DBR_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define DBR_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define DBR_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define DBR_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define DBR_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define DBR_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define DBR_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define DBR_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define DBR_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define DBR_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define DBR_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define DBR_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define DBR_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define DBR_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define DBR_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define DBR_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define DBR_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define DBR_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define DBR_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//ENT_SUVD_CGC_GATE2
+#define ENT_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define ENT_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define ENT_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define ENT_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define ENT_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define ENT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define ENT_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define ENT_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define ENT_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define ENT_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define ENT_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define ENT_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define ENT_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define ENT_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define ENT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define ENT_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define ENT_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define ENT_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define ENT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//IME_SUVD_CGC_GATE2
+#define IME_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define IME_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define IME_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define IME_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define IME_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define IME_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define IME_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define IME_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define IME_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define IME_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define IME_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define IME_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define IME_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define IME_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define IME_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define IME_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define IME_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define IME_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define IME_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define IME_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define IME_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define IME_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define IME_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define IME_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//MPC1_SUVD_CGC_GATE2
+#define MPC1_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                     0x0
+#define MPC1_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                     0x1
+#define MPC1_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                   0x2
+#define MPC1_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                   0x3
+#define MPC1_SUVD_CGC_GATE2__MPC1__SHIFT                                                                      0x4
+#define MPC1_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                               0x5
+#define MPC1_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                     0x6
+#define MPC1_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                     0x7
+#define MPC1_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                     0x8
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                               0x9
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                               0xa
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                               0xb
+#define MPC1_SUVD_CGC_GATE2__MPBE0_MASK                                                                       0x00000001L
+#define MPC1_SUVD_CGC_GATE2__MPBE1_MASK                                                                       0x00000002L
+#define MPC1_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                     0x00000004L
+#define MPC1_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                     0x00000008L
+#define MPC1_SUVD_CGC_GATE2__MPC1_MASK                                                                        0x00000010L
+#define MPC1_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                 0x00000020L
+#define MPC1_SUVD_CGC_GATE2__CDEFE_MASK                                                                       0x00000040L
+#define MPC1_SUVD_CGC_GATE2__AVM_0_MASK                                                                       0x00000080L
+#define MPC1_SUVD_CGC_GATE2__AVM_1_MASK                                                                       0x00000100L
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                 0x00000200L
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                 0x00000400L
+#define MPC1_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                 0x00000800L
+//SAOE_SUVD_CGC_GATE2
+#define SAOE_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                     0x0
+#define SAOE_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                     0x1
+#define SAOE_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                   0x2
+#define SAOE_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                   0x3
+#define SAOE_SUVD_CGC_GATE2__MPC1__SHIFT                                                                      0x4
+#define SAOE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                               0x5
+#define SAOE_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                     0x6
+#define SAOE_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                     0x7
+#define SAOE_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                     0x8
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                               0x9
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                               0xa
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                               0xb
+#define SAOE_SUVD_CGC_GATE2__MPBE0_MASK                                                                       0x00000001L
+#define SAOE_SUVD_CGC_GATE2__MPBE1_MASK                                                                       0x00000002L
+#define SAOE_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                     0x00000004L
+#define SAOE_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                     0x00000008L
+#define SAOE_SUVD_CGC_GATE2__MPC1_MASK                                                                        0x00000010L
+#define SAOE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                 0x00000020L
+#define SAOE_SUVD_CGC_GATE2__CDEFE_MASK                                                                       0x00000040L
+#define SAOE_SUVD_CGC_GATE2__AVM_0_MASK                                                                       0x00000080L
+#define SAOE_SUVD_CGC_GATE2__AVM_1_MASK                                                                       0x00000100L
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                 0x00000200L
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                 0x00000400L
+#define SAOE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                 0x00000800L
+//SDB_SUVD_CGC_GATE2
+#define SDB_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define SDB_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define SDB_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define SDB_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define SDB_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define SDB_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define SDB_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define SDB_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define SDB_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define SDB_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define SDB_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define SDB_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define SDB_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define SDB_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define SDB_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define SDB_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define SDB_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define SDB_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define SDB_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//SIT0_NXT_SUVD_CGC_GATE2
+#define SIT0_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                 0x0
+#define SIT0_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                 0x1
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                               0x2
+#define SIT0_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                               0x3
+#define SIT0_NXT_SUVD_CGC_GATE2__MPC1__SHIFT                                                                  0x4
+#define SIT0_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                           0x5
+#define SIT0_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                 0x6
+#define SIT0_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                 0x7
+#define SIT0_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                 0x8
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                           0x9
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                           0xa
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                           0xb
+#define SIT0_NXT_SUVD_CGC_GATE2__MPBE0_MASK                                                                   0x00000001L
+#define SIT0_NXT_SUVD_CGC_GATE2__MPBE1_MASK                                                                   0x00000002L
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                 0x00000004L
+#define SIT0_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                 0x00000008L
+#define SIT0_NXT_SUVD_CGC_GATE2__MPC1_MASK                                                                    0x00000010L
+#define SIT0_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                             0x00000020L
+#define SIT0_NXT_SUVD_CGC_GATE2__CDEFE_MASK                                                                   0x00000040L
+#define SIT0_NXT_SUVD_CGC_GATE2__AVM_0_MASK                                                                   0x00000080L
+#define SIT0_NXT_SUVD_CGC_GATE2__AVM_1_MASK                                                                   0x00000100L
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                             0x00000200L
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                             0x00000400L
+#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                             0x00000800L
+//SIT1_NXT_SUVD_CGC_GATE2
+#define SIT1_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                 0x0
+#define SIT1_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                 0x1
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                               0x2
+#define SIT1_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                               0x3
+#define SIT1_NXT_SUVD_CGC_GATE2__MPC1__SHIFT                                                                  0x4
+#define SIT1_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                           0x5
+#define SIT1_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                 0x6
+#define SIT1_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                 0x7
+#define SIT1_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                 0x8
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                           0x9
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                           0xa
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                           0xb
+#define SIT1_NXT_SUVD_CGC_GATE2__MPBE0_MASK                                                                   0x00000001L
+#define SIT1_NXT_SUVD_CGC_GATE2__MPBE1_MASK                                                                   0x00000002L
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                 0x00000004L
+#define SIT1_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                 0x00000008L
+#define SIT1_NXT_SUVD_CGC_GATE2__MPC1_MASK                                                                    0x00000010L
+#define SIT1_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                             0x00000020L
+#define SIT1_NXT_SUVD_CGC_GATE2__CDEFE_MASK                                                                   0x00000040L
+#define SIT1_NXT_SUVD_CGC_GATE2__AVM_0_MASK                                                                   0x00000080L
+#define SIT1_NXT_SUVD_CGC_GATE2__AVM_1_MASK                                                                   0x00000100L
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                             0x00000200L
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                             0x00000400L
+#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                             0x00000800L
+//SIT2_NXT_SUVD_CGC_GATE2
+#define SIT2_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                 0x0
+#define SIT2_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                 0x1
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                               0x2
+#define SIT2_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                               0x3
+#define SIT2_NXT_SUVD_CGC_GATE2__MPC1__SHIFT                                                                  0x4
+#define SIT2_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                           0x5
+#define SIT2_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                 0x6
+#define SIT2_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                 0x7
+#define SIT2_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                 0x8
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                           0x9
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                           0xa
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                           0xb
+#define SIT2_NXT_SUVD_CGC_GATE2__MPBE0_MASK                                                                   0x00000001L
+#define SIT2_NXT_SUVD_CGC_GATE2__MPBE1_MASK                                                                   0x00000002L
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                 0x00000004L
+#define SIT2_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                 0x00000008L
+#define SIT2_NXT_SUVD_CGC_GATE2__MPC1_MASK                                                                    0x00000010L
+#define SIT2_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                             0x00000020L
+#define SIT2_NXT_SUVD_CGC_GATE2__CDEFE_MASK                                                                   0x00000040L
+#define SIT2_NXT_SUVD_CGC_GATE2__AVM_0_MASK                                                                   0x00000080L
+#define SIT2_NXT_SUVD_CGC_GATE2__AVM_1_MASK                                                                   0x00000100L
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                             0x00000200L
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                             0x00000400L
+#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                             0x00000800L
+//SIT_SUVD_CGC_GATE2
+#define SIT_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define SIT_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define SIT_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define SIT_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define SIT_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define SIT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define SIT_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define SIT_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define SIT_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define SIT_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define SIT_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define SIT_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define SIT_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define SIT_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define SIT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define SIT_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define SIT_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define SIT_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define SIT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//SMPA_SUVD_CGC_GATE2
+#define SMPA_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                     0x0
+#define SMPA_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                     0x1
+#define SMPA_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                   0x2
+#define SMPA_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                   0x3
+#define SMPA_SUVD_CGC_GATE2__MPC1__SHIFT                                                                      0x4
+#define SMPA_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                               0x5
+#define SMPA_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                     0x6
+#define SMPA_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                     0x7
+#define SMPA_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                     0x8
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                               0x9
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                               0xa
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                               0xb
+#define SMPA_SUVD_CGC_GATE2__MPBE0_MASK                                                                       0x00000001L
+#define SMPA_SUVD_CGC_GATE2__MPBE1_MASK                                                                       0x00000002L
+#define SMPA_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                     0x00000004L
+#define SMPA_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                     0x00000008L
+#define SMPA_SUVD_CGC_GATE2__MPC1_MASK                                                                        0x00000010L
+#define SMPA_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                 0x00000020L
+#define SMPA_SUVD_CGC_GATE2__CDEFE_MASK                                                                       0x00000040L
+#define SMPA_SUVD_CGC_GATE2__AVM_0_MASK                                                                       0x00000080L
+#define SMPA_SUVD_CGC_GATE2__AVM_1_MASK                                                                       0x00000100L
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                 0x00000200L
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                 0x00000400L
+#define SMPA_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                 0x00000800L
+//SMP_SUVD_CGC_GATE2
+#define SMP_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define SMP_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define SMP_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define SMP_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define SMP_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define SMP_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define SMP_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define SMP_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define SMP_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define SMP_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define SMP_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define SMP_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define SMP_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define SMP_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define SMP_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define SMP_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define SMP_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define SMP_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define SMP_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//SRE_SUVD_CGC_GATE2
+#define SRE_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define SRE_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define SRE_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define SRE_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define SRE_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define SRE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define SRE_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define SRE_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define SRE_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define SRE_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define SRE_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define SRE_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define SRE_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define SRE_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define SRE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define SRE_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define SRE_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define SRE_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define SRE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//UVD_MPBE0_SUVD_CGC_GATE2
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                0x0
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                0x1
+#define UVD_MPBE0_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                              0x2
+#define UVD_MPBE0_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                              0x3
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPC1__SHIFT                                                                 0x4
+#define UVD_MPBE0_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                          0x5
+#define UVD_MPBE0_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                0x6
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE0_MASK                                                                  0x00000001L
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE1_MASK                                                                  0x00000002L
+#define UVD_MPBE0_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                0x00000004L
+#define UVD_MPBE0_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                0x00000008L
+#define UVD_MPBE0_SUVD_CGC_GATE2__MPC1_MASK                                                                   0x00000010L
+#define UVD_MPBE0_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                            0x00000020L
+#define UVD_MPBE0_SUVD_CGC_GATE2__CDEFE_MASK                                                                  0x00000040L
+//UVD_MPBE1_SUVD_CGC_GATE2
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                0x0
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                0x1
+#define UVD_MPBE1_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                              0x2
+#define UVD_MPBE1_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                              0x3
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPC1__SHIFT                                                                 0x4
+#define UVD_MPBE1_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                          0x5
+#define UVD_MPBE1_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                0x6
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE0_MASK                                                                  0x00000001L
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE1_MASK                                                                  0x00000002L
+#define UVD_MPBE1_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                0x00000004L
+#define UVD_MPBE1_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                0x00000008L
+#define UVD_MPBE1_SUVD_CGC_GATE2__MPC1_MASK                                                                   0x00000010L
+#define UVD_MPBE1_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                            0x00000020L
+#define UVD_MPBE1_SUVD_CGC_GATE2__CDEFE_MASK                                                                  0x00000040L
+//UVD_SUVD_CGC_GATE2
+#define UVD_SUVD_CGC_GATE2__MPBE0__SHIFT                                                                      0x0
+#define UVD_SUVD_CGC_GATE2__MPBE1__SHIFT                                                                      0x1
+#define UVD_SUVD_CGC_GATE2__SIT_AV1__SHIFT                                                                    0x2
+#define UVD_SUVD_CGC_GATE2__SDB_AV1__SHIFT                                                                    0x3
+#define UVD_SUVD_CGC_GATE2__MPC1__SHIFT                                                                       0x4
+#define UVD_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT                                                                0x5
+#define UVD_SUVD_CGC_GATE2__CDEFE__SHIFT                                                                      0x6
+#define UVD_SUVD_CGC_GATE2__AVM_0__SHIFT                                                                      0x7
+#define UVD_SUVD_CGC_GATE2__AVM_1__SHIFT                                                                      0x8
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT                                                                0x9
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT                                                                0xa
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT                                                                0xb
+#define UVD_SUVD_CGC_GATE2__MPBE0_MASK                                                                        0x00000001L
+#define UVD_SUVD_CGC_GATE2__MPBE1_MASK                                                                        0x00000002L
+#define UVD_SUVD_CGC_GATE2__SIT_AV1_MASK                                                                      0x00000004L
+#define UVD_SUVD_CGC_GATE2__SDB_AV1_MASK                                                                      0x00000008L
+#define UVD_SUVD_CGC_GATE2__MPC1_MASK                                                                         0x00000010L
+#define UVD_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK                                                                  0x00000020L
+#define UVD_SUVD_CGC_GATE2__CDEFE_MASK                                                                        0x00000040L
+#define UVD_SUVD_CGC_GATE2__AVM_0_MASK                                                                        0x00000080L
+#define UVD_SUVD_CGC_GATE2__AVM_1_MASK                                                                        0x00000100L
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK                                                                  0x00000200L
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK                                                                  0x00000400L
+#define UVD_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK                                                                  0x00000800L
+//AVM_SUVD_CGC_CTRL
+#define AVM_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define AVM_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define AVM_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define AVM_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define AVM_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define AVM_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define AVM_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define AVM_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define AVM_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define AVM_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define AVM_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define AVM_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define AVM_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define AVM_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define AVM_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define AVM_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define AVM_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define AVM_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define AVM_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define AVM_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define AVM_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define AVM_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define AVM_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define AVM_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define AVM_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define AVM_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define AVM_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define AVM_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define AVM_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define AVM_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define AVM_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define AVM_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define AVM_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define AVM_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define AVM_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define AVM_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define AVM_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define AVM_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define AVM_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define AVM_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define AVM_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define AVM_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define AVM_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define AVM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define AVM_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define AVM_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define AVM_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//CDEFE_SUVD_CGC_CTRL
+#define CDEFE_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                  0x0
+#define CDEFE_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                  0x1
+#define CDEFE_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                  0x2
+#define CDEFE_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                  0x3
+#define CDEFE_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                  0x4
+#define CDEFE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                 0x5
+#define CDEFE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                               0x6
+#define CDEFE_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                  0x7
+#define CDEFE_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                  0x8
+#define CDEFE_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                 0x9
+#define CDEFE_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                  0xa
+#define CDEFE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                 0xb
+#define CDEFE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                 0xc
+#define CDEFE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                0xd
+#define CDEFE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                0xe
+#define CDEFE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                              0xf
+#define CDEFE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                              0x10
+#define CDEFE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                 0x11
+#define CDEFE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                0x12
+#define CDEFE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                0x13
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                          0x14
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                          0x15
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                          0x16
+#define CDEFE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                  0x1c
+#define CDEFE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                  0x1d
+#define CDEFE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                0x1e
+#define CDEFE_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                    0x00000001L
+#define CDEFE_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                    0x00000002L
+#define CDEFE_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                    0x00000004L
+#define CDEFE_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                    0x00000008L
+#define CDEFE_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                    0x00000010L
+#define CDEFE_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                   0x00000020L
+#define CDEFE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                 0x00000040L
+#define CDEFE_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                    0x00000080L
+#define CDEFE_SUVD_CGC_CTRL__IME_MODE_MASK                                                                    0x00000100L
+#define CDEFE_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                   0x00000200L
+#define CDEFE_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                    0x00000400L
+#define CDEFE_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                   0x00000800L
+#define CDEFE_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                   0x00001000L
+#define CDEFE_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                  0x00002000L
+#define CDEFE_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                  0x00004000L
+#define CDEFE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                0x00008000L
+#define CDEFE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                0x00010000L
+#define CDEFE_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                   0x00020000L
+#define CDEFE_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                  0x00040000L
+#define CDEFE_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                  0x00080000L
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                            0x00100000L
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                            0x00200000L
+#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                            0x00400000L
+#define CDEFE_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                    0x10000000L
+#define CDEFE_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                    0x20000000L
+#define CDEFE_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                  0x40000000L
+//DBR_SUVD_CGC_CTRL
+#define DBR_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define DBR_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define DBR_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define DBR_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define DBR_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define DBR_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define DBR_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define DBR_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define DBR_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define DBR_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define DBR_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define DBR_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define DBR_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define DBR_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define DBR_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define DBR_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define DBR_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define DBR_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define DBR_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define DBR_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define DBR_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define DBR_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define DBR_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define DBR_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define DBR_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define DBR_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define DBR_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define DBR_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define DBR_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define DBR_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define DBR_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define DBR_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define DBR_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define DBR_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define DBR_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define DBR_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define DBR_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define DBR_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define DBR_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define DBR_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define DBR_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define DBR_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define DBR_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define DBR_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define DBR_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define DBR_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define DBR_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//EFC_SUVD_CGC_CTRL
+#define EFC_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define EFC_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define EFC_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define EFC_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define EFC_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define EFC_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define EFC_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define EFC_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define EFC_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define EFC_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define EFC_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define EFC_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define EFC_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define EFC_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define EFC_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define EFC_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define EFC_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define EFC_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define EFC_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define EFC_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define EFC_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define EFC_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define EFC_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define EFC_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define EFC_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define EFC_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define EFC_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define EFC_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define EFC_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define EFC_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define EFC_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define EFC_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define EFC_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define EFC_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define EFC_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define EFC_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define EFC_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define EFC_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define EFC_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define EFC_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define EFC_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define EFC_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define EFC_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define EFC_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define EFC_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define EFC_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define EFC_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//ENT_SUVD_CGC_CTRL
+#define ENT_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define ENT_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define ENT_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define ENT_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define ENT_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define ENT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define ENT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define ENT_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define ENT_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define ENT_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define ENT_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define ENT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define ENT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define ENT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define ENT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define ENT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define ENT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define ENT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define ENT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define ENT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define ENT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define ENT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define ENT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define ENT_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define ENT_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define ENT_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define ENT_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define ENT_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define ENT_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define ENT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define ENT_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define ENT_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define ENT_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define ENT_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define ENT_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define ENT_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define ENT_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define ENT_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define ENT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define ENT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define ENT_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define ENT_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define ENT_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define ENT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define ENT_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define ENT_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define ENT_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//IME_SUVD_CGC_CTRL
+#define IME_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define IME_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define IME_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define IME_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define IME_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define IME_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define IME_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define IME_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define IME_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define IME_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define IME_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define IME_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define IME_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define IME_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define IME_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define IME_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define IME_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define IME_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define IME_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define IME_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define IME_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define IME_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define IME_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define IME_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define IME_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define IME_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define IME_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define IME_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define IME_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define IME_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define IME_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define IME_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define IME_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define IME_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define IME_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define IME_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define IME_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define IME_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define IME_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define IME_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define IME_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define IME_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define IME_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define IME_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define IME_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define IME_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define IME_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define IME_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define IME_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define IME_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define IME_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define IME_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//MPC1_SUVD_CGC_CTRL
+#define MPC1_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                   0x0
+#define MPC1_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                   0x1
+#define MPC1_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                   0x2
+#define MPC1_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                   0x3
+#define MPC1_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                   0x4
+#define MPC1_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                  0x5
+#define MPC1_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                0x6
+#define MPC1_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                   0x7
+#define MPC1_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                   0x8
+#define MPC1_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                  0x9
+#define MPC1_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                   0xa
+#define MPC1_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                  0xb
+#define MPC1_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                  0xc
+#define MPC1_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                 0xd
+#define MPC1_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                 0xe
+#define MPC1_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                               0xf
+#define MPC1_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                               0x10
+#define MPC1_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                  0x11
+#define MPC1_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                 0x12
+#define MPC1_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                 0x13
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                           0x14
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                           0x15
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                           0x16
+#define MPC1_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                   0x1c
+#define MPC1_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                   0x1d
+#define MPC1_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                 0x1e
+#define MPC1_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                     0x00000001L
+#define MPC1_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                     0x00000002L
+#define MPC1_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                     0x00000004L
+#define MPC1_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                     0x00000008L
+#define MPC1_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                     0x00000010L
+#define MPC1_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                    0x00000020L
+#define MPC1_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                  0x00000040L
+#define MPC1_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                     0x00000080L
+#define MPC1_SUVD_CGC_CTRL__IME_MODE_MASK                                                                     0x00000100L
+#define MPC1_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                    0x00000200L
+#define MPC1_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                     0x00000400L
+#define MPC1_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                    0x00000800L
+#define MPC1_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                    0x00001000L
+#define MPC1_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                   0x00002000L
+#define MPC1_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                   0x00004000L
+#define MPC1_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                 0x00008000L
+#define MPC1_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                 0x00010000L
+#define MPC1_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                    0x00020000L
+#define MPC1_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                   0x00040000L
+#define MPC1_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                   0x00080000L
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                             0x00100000L
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                             0x00200000L
+#define MPC1_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                             0x00400000L
+#define MPC1_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                     0x10000000L
+#define MPC1_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                     0x20000000L
+#define MPC1_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                   0x40000000L
+//PPU_SUVD_CGC_CTRL
+#define PPU_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define PPU_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define PPU_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define PPU_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define PPU_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define PPU_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define PPU_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define PPU_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define PPU_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define PPU_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define PPU_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define PPU_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define PPU_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define PPU_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define PPU_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define PPU_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define PPU_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define PPU_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define PPU_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define PPU_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define PPU_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define PPU_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define PPU_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define PPU_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define PPU_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define PPU_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define PPU_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define PPU_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define PPU_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define PPU_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define PPU_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define PPU_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define PPU_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define PPU_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define PPU_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define PPU_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define PPU_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define PPU_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define PPU_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define PPU_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define PPU_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define PPU_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define PPU_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define PPU_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define PPU_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define PPU_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define PPU_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//SAOE_SUVD_CGC_CTRL
+#define SAOE_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                   0x0
+#define SAOE_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                   0x1
+#define SAOE_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                   0x2
+#define SAOE_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                   0x3
+#define SAOE_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                   0x4
+#define SAOE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                  0x5
+#define SAOE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                0x6
+#define SAOE_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                   0x7
+#define SAOE_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                   0x8
+#define SAOE_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                  0x9
+#define SAOE_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                   0xa
+#define SAOE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                  0xb
+#define SAOE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                  0xc
+#define SAOE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                 0xd
+#define SAOE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                 0xe
+#define SAOE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                               0xf
+#define SAOE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                               0x10
+#define SAOE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                  0x11
+#define SAOE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                 0x12
+#define SAOE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                 0x13
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                           0x14
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                           0x15
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                           0x16
+#define SAOE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                   0x1c
+#define SAOE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                   0x1d
+#define SAOE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                 0x1e
+#define SAOE_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                     0x00000001L
+#define SAOE_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                     0x00000002L
+#define SAOE_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                     0x00000004L
+#define SAOE_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                     0x00000008L
+#define SAOE_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                     0x00000010L
+#define SAOE_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                    0x00000020L
+#define SAOE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                  0x00000040L
+#define SAOE_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                     0x00000080L
+#define SAOE_SUVD_CGC_CTRL__IME_MODE_MASK                                                                     0x00000100L
+#define SAOE_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                    0x00000200L
+#define SAOE_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                     0x00000400L
+#define SAOE_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                    0x00000800L
+#define SAOE_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                    0x00001000L
+#define SAOE_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                   0x00002000L
+#define SAOE_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                   0x00004000L
+#define SAOE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                 0x00008000L
+#define SAOE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                 0x00010000L
+#define SAOE_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                    0x00020000L
+#define SAOE_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                   0x00040000L
+#define SAOE_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                   0x00080000L
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                             0x00100000L
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                             0x00200000L
+#define SAOE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                             0x00400000L
+#define SAOE_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                     0x10000000L
+#define SAOE_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                     0x20000000L
+#define SAOE_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                   0x40000000L
+//SCM_SUVD_CGC_CTRL
+#define SCM_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define SCM_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define SCM_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define SCM_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define SCM_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define SCM_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define SCM_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define SCM_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define SCM_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define SCM_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define SCM_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define SCM_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define SCM_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define SCM_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define SCM_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define SCM_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define SCM_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define SCM_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define SCM_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define SCM_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define SCM_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define SCM_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define SCM_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define SCM_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define SCM_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define SCM_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define SCM_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define SCM_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define SCM_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define SCM_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define SCM_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define SCM_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define SCM_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define SCM_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define SCM_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define SCM_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define SCM_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define SCM_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define SCM_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define SCM_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define SCM_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define SCM_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define SCM_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define SCM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define SCM_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define SCM_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define SCM_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//SDB_SUVD_CGC_CTRL
+#define SDB_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define SDB_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define SDB_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define SDB_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define SDB_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define SDB_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define SDB_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define SDB_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define SDB_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define SDB_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define SDB_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define SDB_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define SDB_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define SDB_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define SDB_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define SDB_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define SDB_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define SDB_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define SDB_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define SDB_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define SDB_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define SDB_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define SDB_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define SDB_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define SDB_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define SDB_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define SDB_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define SDB_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define SDB_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define SDB_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define SDB_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define SDB_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define SDB_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define SDB_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define SDB_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define SDB_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define SDB_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define SDB_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define SDB_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define SDB_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define SDB_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define SDB_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define SDB_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define SDB_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define SDB_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define SDB_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define SDB_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//SIT0_NXT_SUVD_CGC_CTRL
+#define SIT0_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                               0x0
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                               0x1
+#define SIT0_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                               0x2
+#define SIT0_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                               0x3
+#define SIT0_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                               0x4
+#define SIT0_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                              0x5
+#define SIT0_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                            0x6
+#define SIT0_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                               0x7
+#define SIT0_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                               0x8
+#define SIT0_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                              0x9
+#define SIT0_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                               0xa
+#define SIT0_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                              0xb
+#define SIT0_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                              0xc
+#define SIT0_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                             0xd
+#define SIT0_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                             0xe
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                           0xf
+#define SIT0_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                           0x10
+#define SIT0_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                              0x11
+#define SIT0_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                             0x12
+#define SIT0_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                             0x13
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                       0x14
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                       0x15
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                       0x16
+#define SIT0_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT0_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT0_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                             0x1e
+#define SIT0_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                 0x00000001L
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                 0x00000002L
+#define SIT0_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                 0x00000004L
+#define SIT0_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                 0x00000008L
+#define SIT0_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                 0x00000010L
+#define SIT0_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                0x00000020L
+#define SIT0_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                              0x00000040L
+#define SIT0_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                 0x00000080L
+#define SIT0_NXT_SUVD_CGC_CTRL__IME_MODE_MASK                                                                 0x00000100L
+#define SIT0_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                0x00000200L
+#define SIT0_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                 0x00000400L
+#define SIT0_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                0x00000800L
+#define SIT0_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                0x00001000L
+#define SIT0_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                               0x00002000L
+#define SIT0_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                               0x00004000L
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                             0x00008000L
+#define SIT0_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                             0x00010000L
+#define SIT0_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                0x00020000L
+#define SIT0_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                               0x00040000L
+#define SIT0_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                               0x00080000L
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                         0x00100000L
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                         0x00200000L
+#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                         0x00400000L
+#define SIT0_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT0_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT0_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                               0x40000000L
+//SIT1_NXT_SUVD_CGC_CTRL
+#define SIT1_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                               0x0
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                               0x1
+#define SIT1_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                               0x2
+#define SIT1_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                               0x3
+#define SIT1_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                               0x4
+#define SIT1_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                              0x5
+#define SIT1_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                            0x6
+#define SIT1_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                               0x7
+#define SIT1_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                               0x8
+#define SIT1_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                              0x9
+#define SIT1_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                               0xa
+#define SIT1_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                              0xb
+#define SIT1_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                              0xc
+#define SIT1_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                             0xd
+#define SIT1_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                             0xe
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                           0xf
+#define SIT1_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                           0x10
+#define SIT1_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                              0x11
+#define SIT1_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                             0x12
+#define SIT1_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                             0x13
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                       0x14
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                       0x15
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                       0x16
+#define SIT1_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT1_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT1_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                             0x1e
+#define SIT1_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                 0x00000001L
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                 0x00000002L
+#define SIT1_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                 0x00000004L
+#define SIT1_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                 0x00000008L
+#define SIT1_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                 0x00000010L
+#define SIT1_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                0x00000020L
+#define SIT1_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                              0x00000040L
+#define SIT1_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                 0x00000080L
+#define SIT1_NXT_SUVD_CGC_CTRL__IME_MODE_MASK                                                                 0x00000100L
+#define SIT1_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                0x00000200L
+#define SIT1_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                 0x00000400L
+#define SIT1_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                0x00000800L
+#define SIT1_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                0x00001000L
+#define SIT1_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                               0x00002000L
+#define SIT1_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                               0x00004000L
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                             0x00008000L
+#define SIT1_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                             0x00010000L
+#define SIT1_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                0x00020000L
+#define SIT1_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                               0x00040000L
+#define SIT1_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                               0x00080000L
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                         0x00100000L
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                         0x00200000L
+#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                         0x00400000L
+#define SIT1_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT1_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT1_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                               0x40000000L
+//SIT2_NXT_SUVD_CGC_CTRL
+#define SIT2_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                               0x0
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                               0x1
+#define SIT2_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                               0x2
+#define SIT2_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                               0x3
+#define SIT2_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                               0x4
+#define SIT2_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                              0x5
+#define SIT2_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                            0x6
+#define SIT2_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                               0x7
+#define SIT2_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                               0x8
+#define SIT2_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                              0x9
+#define SIT2_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                               0xa
+#define SIT2_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                              0xb
+#define SIT2_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                              0xc
+#define SIT2_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                             0xd
+#define SIT2_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                             0xe
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                           0xf
+#define SIT2_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                           0x10
+#define SIT2_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                              0x11
+#define SIT2_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                             0x12
+#define SIT2_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                             0x13
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                       0x14
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                       0x15
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                       0x16
+#define SIT2_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                               0x1c
+#define SIT2_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                               0x1d
+#define SIT2_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                             0x1e
+#define SIT2_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                 0x00000001L
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                 0x00000002L
+#define SIT2_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                 0x00000004L
+#define SIT2_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                 0x00000008L
+#define SIT2_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                 0x00000010L
+#define SIT2_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                0x00000020L
+#define SIT2_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                              0x00000040L
+#define SIT2_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                 0x00000080L
+#define SIT2_NXT_SUVD_CGC_CTRL__IME_MODE_MASK                                                                 0x00000100L
+#define SIT2_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                0x00000200L
+#define SIT2_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                 0x00000400L
+#define SIT2_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                0x00000800L
+#define SIT2_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                0x00001000L
+#define SIT2_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                               0x00002000L
+#define SIT2_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                               0x00004000L
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                             0x00008000L
+#define SIT2_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                             0x00010000L
+#define SIT2_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                0x00020000L
+#define SIT2_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                               0x00040000L
+#define SIT2_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                               0x00080000L
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                         0x00100000L
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                         0x00200000L
+#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                         0x00400000L
+#define SIT2_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                 0x10000000L
+#define SIT2_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                 0x20000000L
+#define SIT2_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                               0x40000000L
+//SIT_SUVD_CGC_CTRL
+#define SIT_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define SIT_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define SIT_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define SIT_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define SIT_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define SIT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define SIT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define SIT_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define SIT_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define SIT_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define SIT_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define SIT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define SIT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define SIT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define SIT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define SIT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define SIT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define SIT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define SIT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define SIT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define SIT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define SIT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define SIT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define SIT_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define SIT_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define SIT_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define SIT_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define SIT_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define SIT_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define SIT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define SIT_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define SIT_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define SIT_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define SIT_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define SIT_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define SIT_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define SIT_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define SIT_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define SIT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define SIT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define SIT_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define SIT_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define SIT_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define SIT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define SIT_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define SIT_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define SIT_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//SMPA_SUVD_CGC_CTRL
+#define SMPA_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                   0x0
+#define SMPA_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                   0x1
+#define SMPA_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                   0x2
+#define SMPA_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                   0x3
+#define SMPA_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                   0x4
+#define SMPA_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                  0x5
+#define SMPA_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                0x6
+#define SMPA_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                   0x7
+#define SMPA_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                   0x8
+#define SMPA_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                  0x9
+#define SMPA_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                   0xa
+#define SMPA_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                  0xb
+#define SMPA_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                  0xc
+#define SMPA_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                 0xd
+#define SMPA_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                 0xe
+#define SMPA_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                               0xf
+#define SMPA_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                               0x10
+#define SMPA_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                  0x11
+#define SMPA_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                 0x12
+#define SMPA_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                 0x13
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                           0x14
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                           0x15
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                           0x16
+#define SMPA_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                   0x1c
+#define SMPA_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                   0x1d
+#define SMPA_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                 0x1e
+#define SMPA_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                     0x00000001L
+#define SMPA_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                     0x00000002L
+#define SMPA_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                     0x00000004L
+#define SMPA_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                     0x00000008L
+#define SMPA_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                     0x00000010L
+#define SMPA_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                    0x00000020L
+#define SMPA_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                  0x00000040L
+#define SMPA_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                     0x00000080L
+#define SMPA_SUVD_CGC_CTRL__IME_MODE_MASK                                                                     0x00000100L
+#define SMPA_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                    0x00000200L
+#define SMPA_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                     0x00000400L
+#define SMPA_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                    0x00000800L
+#define SMPA_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                    0x00001000L
+#define SMPA_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                   0x00002000L
+#define SMPA_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                   0x00004000L
+#define SMPA_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                 0x00008000L
+#define SMPA_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                 0x00010000L
+#define SMPA_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                    0x00020000L
+#define SMPA_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                   0x00040000L
+#define SMPA_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                   0x00080000L
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                             0x00100000L
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                             0x00200000L
+#define SMPA_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                             0x00400000L
+#define SMPA_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                     0x10000000L
+#define SMPA_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                     0x20000000L
+#define SMPA_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                   0x40000000L
+//SMP_SUVD_CGC_CTRL
+#define SMP_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define SMP_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define SMP_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define SMP_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define SMP_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define SMP_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define SMP_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define SMP_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define SMP_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define SMP_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define SMP_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define SMP_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define SMP_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define SMP_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define SMP_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define SMP_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define SMP_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define SMP_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define SMP_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define SMP_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define SMP_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define SMP_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define SMP_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define SMP_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define SMP_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define SMP_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define SMP_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define SMP_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define SMP_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define SMP_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define SMP_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define SMP_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define SMP_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define SMP_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define SMP_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define SMP_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define SMP_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define SMP_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define SMP_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define SMP_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define SMP_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define SMP_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define SMP_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define SMP_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define SMP_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define SMP_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define SMP_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//SRE_SUVD_CGC_CTRL
+#define SRE_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define SRE_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define SRE_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define SRE_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define SRE_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define SRE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define SRE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define SRE_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define SRE_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define SRE_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define SRE_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define SRE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define SRE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define SRE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define SRE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define SRE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define SRE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define SRE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define SRE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define SRE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define SRE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define SRE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define SRE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define SRE_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define SRE_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define SRE_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define SRE_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define SRE_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define SRE_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define SRE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define SRE_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define SRE_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define SRE_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define SRE_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define SRE_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define SRE_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define SRE_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define SRE_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define SRE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define SRE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define SRE_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define SRE_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define SRE_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define SRE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define SRE_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define SRE_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define SRE_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//UVD_MPBE0_SUVD_CGC_CTRL
+#define UVD_MPBE0_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                              0x0
+#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                              0x1
+#define UVD_MPBE0_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                              0x2
+#define UVD_MPBE0_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                              0x3
+#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                              0x4
+#define UVD_MPBE0_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                             0x5
+#define UVD_MPBE0_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                           0x6
+#define UVD_MPBE0_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                              0x7
+#define UVD_MPBE0_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                              0x8
+#define UVD_MPBE0_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                             0x9
+#define UVD_MPBE0_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                              0xa
+#define UVD_MPBE0_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                             0xb
+#define UVD_MPBE0_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                             0xc
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                            0xd
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                            0xe
+#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                          0xf
+#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                          0x10
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                             0x11
+#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                              0x1c
+#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                              0x1d
+#define UVD_MPBE0_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                            0x1e
+#define UVD_MPBE0_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                0x00000001L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                0x00000002L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                0x00000004L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                0x00000008L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                0x00000010L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                               0x00000020L
+#define UVD_MPBE0_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                             0x00000040L
+#define UVD_MPBE0_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                0x00000080L
+#define UVD_MPBE0_SUVD_CGC_CTRL__IME_MODE_MASK                                                                0x00000100L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SITE_MODE_MASK                                                               0x00000200L
+#define UVD_MPBE0_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                0x00000400L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                               0x00000800L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                               0x00001000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                              0x00002000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                              0x00004000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                            0x00008000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                            0x00010000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                               0x00020000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                0x10000000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                0x20000000L
+#define UVD_MPBE0_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                              0x40000000L
+//UVD_MPBE1_SUVD_CGC_CTRL
+#define UVD_MPBE1_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                              0x0
+#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                              0x1
+#define UVD_MPBE1_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                              0x2
+#define UVD_MPBE1_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                              0x3
+#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                              0x4
+#define UVD_MPBE1_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                             0x5
+#define UVD_MPBE1_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                           0x6
+#define UVD_MPBE1_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                              0x7
+#define UVD_MPBE1_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                              0x8
+#define UVD_MPBE1_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                             0x9
+#define UVD_MPBE1_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                              0xa
+#define UVD_MPBE1_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                             0xb
+#define UVD_MPBE1_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                             0xc
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                            0xd
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                            0xe
+#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                          0xf
+#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                          0x10
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                             0x11
+#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                              0x1c
+#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                              0x1d
+#define UVD_MPBE1_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                            0x1e
+#define UVD_MPBE1_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                0x00000001L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                0x00000002L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                0x00000004L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                0x00000008L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                0x00000010L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                               0x00000020L
+#define UVD_MPBE1_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                             0x00000040L
+#define UVD_MPBE1_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                0x00000080L
+#define UVD_MPBE1_SUVD_CGC_CTRL__IME_MODE_MASK                                                                0x00000100L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SITE_MODE_MASK                                                               0x00000200L
+#define UVD_MPBE1_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                0x00000400L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                               0x00000800L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                               0x00001000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                              0x00002000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                              0x00004000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                            0x00008000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                            0x00010000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                               0x00020000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                0x10000000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                0x20000000L
+#define UVD_MPBE1_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                              0x40000000L
+//UVD_SUVD_CGC_CTRL
+#define UVD_SUVD_CGC_CTRL__SRE_MODE__SHIFT                                                                    0x0
+#define UVD_SUVD_CGC_CTRL__SIT_MODE__SHIFT                                                                    0x1
+#define UVD_SUVD_CGC_CTRL__SMP_MODE__SHIFT                                                                    0x2
+#define UVD_SUVD_CGC_CTRL__SCM_MODE__SHIFT                                                                    0x3
+#define UVD_SUVD_CGC_CTRL__SDB_MODE__SHIFT                                                                    0x4
+#define UVD_SUVD_CGC_CTRL__SCLR_MODE__SHIFT                                                                   0x5
+#define UVD_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT                                                                 0x6
+#define UVD_SUVD_CGC_CTRL__ENT_MODE__SHIFT                                                                    0x7
+#define UVD_SUVD_CGC_CTRL__IME_MODE__SHIFT                                                                    0x8
+#define UVD_SUVD_CGC_CTRL__SITE_MODE__SHIFT                                                                   0x9
+#define UVD_SUVD_CGC_CTRL__EFC_MODE__SHIFT                                                                    0xa
+#define UVD_SUVD_CGC_CTRL__SAOE_MODE__SHIFT                                                                   0xb
+#define UVD_SUVD_CGC_CTRL__SMPA_MODE__SHIFT                                                                   0xc
+#define UVD_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT                                                                  0xd
+#define UVD_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT                                                                  0xe
+#define UVD_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT                                                                0xf
+#define UVD_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT                                                                0x10
+#define UVD_SUVD_CGC_CTRL__MPC1_MODE__SHIFT                                                                   0x11
+#define UVD_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT                                                                  0x12
+#define UVD_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT                                                                  0x13
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT                                                            0x14
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT                                                            0x15
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT                                                            0x16
+#define UVD_SUVD_CGC_CTRL__FBC_PCLK__SHIFT                                                                    0x1c
+#define UVD_SUVD_CGC_CTRL__FBC_CCLK__SHIFT                                                                    0x1d
+#define UVD_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT                                                                  0x1e
+#define UVD_SUVD_CGC_CTRL__SRE_MODE_MASK                                                                      0x00000001L
+#define UVD_SUVD_CGC_CTRL__SIT_MODE_MASK                                                                      0x00000002L
+#define UVD_SUVD_CGC_CTRL__SMP_MODE_MASK                                                                      0x00000004L
+#define UVD_SUVD_CGC_CTRL__SCM_MODE_MASK                                                                      0x00000008L
+#define UVD_SUVD_CGC_CTRL__SDB_MODE_MASK                                                                      0x00000010L
+#define UVD_SUVD_CGC_CTRL__SCLR_MODE_MASK                                                                     0x00000020L
+#define UVD_SUVD_CGC_CTRL__UVD_SC_MODE_MASK                                                                   0x00000040L
+#define UVD_SUVD_CGC_CTRL__ENT_MODE_MASK                                                                      0x00000080L
+#define UVD_SUVD_CGC_CTRL__IME_MODE_MASK                                                                      0x00000100L
+#define UVD_SUVD_CGC_CTRL__SITE_MODE_MASK                                                                     0x00000200L
+#define UVD_SUVD_CGC_CTRL__EFC_MODE_MASK                                                                      0x00000400L
+#define UVD_SUVD_CGC_CTRL__SAOE_MODE_MASK                                                                     0x00000800L
+#define UVD_SUVD_CGC_CTRL__SMPA_MODE_MASK                                                                     0x00001000L
+#define UVD_SUVD_CGC_CTRL__MPBE0_MODE_MASK                                                                    0x00002000L
+#define UVD_SUVD_CGC_CTRL__MPBE1_MODE_MASK                                                                    0x00004000L
+#define UVD_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK                                                                  0x00008000L
+#define UVD_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK                                                                  0x00010000L
+#define UVD_SUVD_CGC_CTRL__MPC1_MODE_MASK                                                                     0x00020000L
+#define UVD_SUVD_CGC_CTRL__AVM_0_MODE_MASK                                                                    0x00040000L
+#define UVD_SUVD_CGC_CTRL__AVM_1_MODE_MASK                                                                    0x00080000L
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK                                                              0x00100000L
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK                                                              0x00200000L
+#define UVD_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK                                                              0x00400000L
+#define UVD_SUVD_CGC_CTRL__FBC_PCLK_MASK                                                                      0x10000000L
+#define UVD_SUVD_CGC_CTRL__FBC_CCLK_MASK                                                                      0x20000000L
+#define UVD_SUVD_CGC_CTRL__CDEFE_MODE_MASK                                                                    0x40000000L
+//UVD_CGC_CTRL3
+#define UVD_CGC_CTRL3__CGC_CLK_OFF_DELAY__SHIFT                                                               0x0
+#define UVD_CGC_CTRL3__LCM0_MODE__SHIFT                                                                       0xb
+#define UVD_CGC_CTRL3__LCM1_MODE__SHIFT                                                                       0xc
+#define UVD_CGC_CTRL3__MIF_MODE__SHIFT                                                                        0xd
+#define UVD_CGC_CTRL3__VREG_MODE__SHIFT                                                                       0xe
+#define UVD_CGC_CTRL3__PE_MODE__SHIFT                                                                         0xf
+#define UVD_CGC_CTRL3__PPU_MODE__SHIFT                                                                        0x10
+#define UVD_CGC_CTRL3__CGC_CLK_OFF_DELAY_MASK                                                                 0x000000FFL
+#define UVD_CGC_CTRL3__LCM0_MODE_MASK                                                                         0x00000800L
+#define UVD_CGC_CTRL3__LCM1_MODE_MASK                                                                         0x00001000L
+#define UVD_CGC_CTRL3__MIF_MODE_MASK                                                                          0x00002000L
+#define UVD_CGC_CTRL3__VREG_MODE_MASK                                                                         0x00004000L
+#define UVD_CGC_CTRL3__PE_MODE_MASK                                                                           0x00008000L
+#define UVD_CGC_CTRL3__PPU_MODE_MASK                                                                          0x00010000L
+//UVD_GPCOM_VCPU_DATA0
+#define UVD_GPCOM_VCPU_DATA0__DATA0__SHIFT                                                                    0x0
+#define UVD_GPCOM_VCPU_DATA0__DATA0_MASK                                                                      0xFFFFFFFFL
+//UVD_GPCOM_VCPU_DATA1
+#define UVD_GPCOM_VCPU_DATA1__DATA1__SHIFT                                                                    0x0
+#define UVD_GPCOM_VCPU_DATA1__DATA1_MASK                                                                      0xFFFFFFFFL
+//UVD_GPCOM_SYS_CMD
+#define UVD_GPCOM_SYS_CMD__CMD_SEND__SHIFT                                                                    0x0
+#define UVD_GPCOM_SYS_CMD__CMD__SHIFT                                                                         0x1
+#define UVD_GPCOM_SYS_CMD__CMD_SOURCE__SHIFT                                                                  0x1f
+#define UVD_GPCOM_SYS_CMD__CMD_SEND_MASK                                                                      0x00000001L
+#define UVD_GPCOM_SYS_CMD__CMD_MASK                                                                           0x7FFFFFFEL
+#define UVD_GPCOM_SYS_CMD__CMD_SOURCE_MASK                                                                    0x80000000L
+//UVD_GPCOM_SYS_DATA0
+#define UVD_GPCOM_SYS_DATA0__DATA0__SHIFT                                                                     0x0
+#define UVD_GPCOM_SYS_DATA0__DATA0_MASK                                                                       0xFFFFFFFFL
+//UVD_GPCOM_SYS_DATA1
+#define UVD_GPCOM_SYS_DATA1__DATA1__SHIFT                                                                     0x0
+#define UVD_GPCOM_SYS_DATA1__DATA1_MASK                                                                       0xFFFFFFFFL
+//UVD_VCPU_INT_EN
+#define UVD_VCPU_INT_EN__PIF_ADDR_ERR_EN__SHIFT                                                               0x0
+#define UVD_VCPU_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN__SHIFT                                                    0x1
+#define UVD_VCPU_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN__SHIFT                                             0x2
+#define UVD_VCPU_INT_EN__NJ_PF_RPT_EN__SHIFT                                                                  0x3
+#define UVD_VCPU_INT_EN__SW_RB1_INT_EN__SHIFT                                                                 0x4
+#define UVD_VCPU_INT_EN__SW_RB2_INT_EN__SHIFT                                                                 0x5
+#define UVD_VCPU_INT_EN__RBC_REG_PRIV_FAULT_EN__SHIFT                                                         0x6
+#define UVD_VCPU_INT_EN__SW_RB3_INT_EN__SHIFT                                                                 0x7
+#define UVD_VCPU_INT_EN__SW_RB4_INT_EN__SHIFT                                                                 0x9
+#define UVD_VCPU_INT_EN__SW_RB5_INT_EN__SHIFT                                                                 0xa
+#define UVD_VCPU_INT_EN__LBSI_EN__SHIFT                                                                       0xb
+#define UVD_VCPU_INT_EN__UDEC_EN__SHIFT                                                                       0xc
+#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN__SHIFT                                                    0xd
+#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN__SHIFT                                              0xe
+#define UVD_VCPU_INT_EN__SUVD_EN__SHIFT                                                                       0xf
+#define UVD_VCPU_INT_EN__RPTR_WR_EN__SHIFT                                                                    0x10
+#define UVD_VCPU_INT_EN__JOB_START_EN__SHIFT                                                                  0x11
+#define UVD_VCPU_INT_EN__NJ_PF_EN__SHIFT                                                                      0x12
+#define UVD_VCPU_INT_EN__SEMA_WAIT_FAIL_SIG_EN__SHIFT                                                         0x17
+#define UVD_VCPU_INT_EN__IDCT_EN__SHIFT                                                                       0x18
+#define UVD_VCPU_INT_EN__MPRD_EN__SHIFT                                                                       0x19
+#define UVD_VCPU_INT_EN__AVM_INT_EN__SHIFT                                                                    0x1a
+#define UVD_VCPU_INT_EN__CLK_SWT_EN__SHIFT                                                                    0x1b
+#define UVD_VCPU_INT_EN__MIF_HWINT_EN__SHIFT                                                                  0x1c
+#define UVD_VCPU_INT_EN__MPRD_ERR_EN__SHIFT                                                                   0x1d
+#define UVD_VCPU_INT_EN__DRV_FW_REQ_EN__SHIFT                                                                 0x1e
+#define UVD_VCPU_INT_EN__DRV_FW_ACK_EN__SHIFT                                                                 0x1f
+#define UVD_VCPU_INT_EN__PIF_ADDR_ERR_EN_MASK                                                                 0x00000001L
+#define UVD_VCPU_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN_MASK                                                      0x00000002L
+#define UVD_VCPU_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN_MASK                                               0x00000004L
+#define UVD_VCPU_INT_EN__NJ_PF_RPT_EN_MASK                                                                    0x00000008L
+#define UVD_VCPU_INT_EN__SW_RB1_INT_EN_MASK                                                                   0x00000010L
+#define UVD_VCPU_INT_EN__SW_RB2_INT_EN_MASK                                                                   0x00000020L
+#define UVD_VCPU_INT_EN__RBC_REG_PRIV_FAULT_EN_MASK                                                           0x00000040L
+#define UVD_VCPU_INT_EN__SW_RB3_INT_EN_MASK                                                                   0x00000080L
+#define UVD_VCPU_INT_EN__SW_RB4_INT_EN_MASK                                                                   0x00000200L
+#define UVD_VCPU_INT_EN__SW_RB5_INT_EN_MASK                                                                   0x00000400L
+#define UVD_VCPU_INT_EN__LBSI_EN_MASK                                                                         0x00000800L
+#define UVD_VCPU_INT_EN__UDEC_EN_MASK                                                                         0x00001000L
+#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN_MASK                                                      0x00002000L
+#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN_MASK                                                0x00004000L
+#define UVD_VCPU_INT_EN__SUVD_EN_MASK                                                                         0x00008000L
+#define UVD_VCPU_INT_EN__RPTR_WR_EN_MASK                                                                      0x00010000L
+#define UVD_VCPU_INT_EN__JOB_START_EN_MASK                                                                    0x00020000L
+#define UVD_VCPU_INT_EN__NJ_PF_EN_MASK                                                                        0x00040000L
+#define UVD_VCPU_INT_EN__SEMA_WAIT_FAIL_SIG_EN_MASK                                                           0x00800000L
+#define UVD_VCPU_INT_EN__IDCT_EN_MASK                                                                         0x01000000L
+#define UVD_VCPU_INT_EN__MPRD_EN_MASK                                                                         0x02000000L
+#define UVD_VCPU_INT_EN__AVM_INT_EN_MASK                                                                      0x04000000L
+#define UVD_VCPU_INT_EN__CLK_SWT_EN_MASK                                                                      0x08000000L
+#define UVD_VCPU_INT_EN__MIF_HWINT_EN_MASK                                                                    0x10000000L
+#define UVD_VCPU_INT_EN__MPRD_ERR_EN_MASK                                                                     0x20000000L
+#define UVD_VCPU_INT_EN__DRV_FW_REQ_EN_MASK                                                                   0x40000000L
+#define UVD_VCPU_INT_EN__DRV_FW_ACK_EN_MASK                                                                   0x80000000L
+//UVD_VCPU_INT_STATUS
+#define UVD_VCPU_INT_STATUS__PIF_ADDR_ERR_INT__SHIFT                                                          0x0
+#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT__SHIFT                                               0x1
+#define UVD_VCPU_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT__SHIFT                                        0x2
+#define UVD_VCPU_INT_STATUS__NJ_PF_RPT_INT__SHIFT                                                             0x3
+#define UVD_VCPU_INT_STATUS__SW_RB1_INT__SHIFT                                                                0x4
+#define UVD_VCPU_INT_STATUS__SW_RB2_INT__SHIFT                                                                0x5
+#define UVD_VCPU_INT_STATUS__RBC_REG_PRIV_FAULT_INT__SHIFT                                                    0x6
+#define UVD_VCPU_INT_STATUS__SW_RB3_INT__SHIFT                                                                0x7
+#define UVD_VCPU_INT_STATUS__SW_RB4_INT__SHIFT                                                                0x9
+#define UVD_VCPU_INT_STATUS__SW_RB5_INT__SHIFT                                                                0xa
+#define UVD_VCPU_INT_STATUS__LBSI_INT__SHIFT                                                                  0xb
+#define UVD_VCPU_INT_STATUS__UDEC_INT__SHIFT                                                                  0xc
+#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT__SHIFT                                               0xd
+#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT__SHIFT                                         0xe
+#define UVD_VCPU_INT_STATUS__SUVD_INT__SHIFT                                                                  0xf
+#define UVD_VCPU_INT_STATUS__RPTR_WR_INT__SHIFT                                                               0x10
+#define UVD_VCPU_INT_STATUS__JOB_START_INT__SHIFT                                                             0x11
+#define UVD_VCPU_INT_STATUS__NJ_PF_INT__SHIFT                                                                 0x12
+#define UVD_VCPU_INT_STATUS__GPCOM_INT__SHIFT                                                                 0x14
+#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT__SHIFT                                                    0x17
+#define UVD_VCPU_INT_STATUS__IDCT_INT__SHIFT                                                                  0x18
+#define UVD_VCPU_INT_STATUS__MPRD_INT__SHIFT                                                                  0x19
+#define UVD_VCPU_INT_STATUS__AVM_INT__SHIFT                                                                   0x1a
+#define UVD_VCPU_INT_STATUS__CLK_SWT_INT__SHIFT                                                               0x1b
+#define UVD_VCPU_INT_STATUS__MIF_HWINT__SHIFT                                                                 0x1c
+#define UVD_VCPU_INT_STATUS__MPRD_ERR_INT__SHIFT                                                              0x1d
+#define UVD_VCPU_INT_STATUS__DRV_FW_REQ_INT__SHIFT                                                            0x1e
+#define UVD_VCPU_INT_STATUS__DRV_FW_ACK_INT__SHIFT                                                            0x1f
+#define UVD_VCPU_INT_STATUS__PIF_ADDR_ERR_INT_MASK                                                            0x00000001L
+#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT_MASK                                                 0x00000002L
+#define UVD_VCPU_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT_MASK                                          0x00000004L
+#define UVD_VCPU_INT_STATUS__NJ_PF_RPT_INT_MASK                                                               0x00000008L
+#define UVD_VCPU_INT_STATUS__SW_RB1_INT_MASK                                                                  0x00000010L
+#define UVD_VCPU_INT_STATUS__SW_RB2_INT_MASK                                                                  0x00000020L
+#define UVD_VCPU_INT_STATUS__RBC_REG_PRIV_FAULT_INT_MASK                                                      0x00000040L
+#define UVD_VCPU_INT_STATUS__SW_RB3_INT_MASK                                                                  0x00000080L
+#define UVD_VCPU_INT_STATUS__SW_RB4_INT_MASK                                                                  0x00000200L
+#define UVD_VCPU_INT_STATUS__SW_RB5_INT_MASK                                                                  0x00000400L
+#define UVD_VCPU_INT_STATUS__LBSI_INT_MASK                                                                    0x00000800L
+#define UVD_VCPU_INT_STATUS__UDEC_INT_MASK                                                                    0x00001000L
+#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT_MASK                                                 0x00002000L
+#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT_MASK                                           0x00004000L
+#define UVD_VCPU_INT_STATUS__SUVD_INT_MASK                                                                    0x00008000L
+#define UVD_VCPU_INT_STATUS__RPTR_WR_INT_MASK                                                                 0x00010000L
+#define UVD_VCPU_INT_STATUS__JOB_START_INT_MASK                                                               0x00020000L
+#define UVD_VCPU_INT_STATUS__NJ_PF_INT_MASK                                                                   0x00040000L
+#define UVD_VCPU_INT_STATUS__GPCOM_INT_MASK                                                                   0x00100000L
+#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT_MASK                                                      0x00800000L
+#define UVD_VCPU_INT_STATUS__IDCT_INT_MASK                                                                    0x01000000L
+#define UVD_VCPU_INT_STATUS__MPRD_INT_MASK                                                                    0x02000000L
+#define UVD_VCPU_INT_STATUS__AVM_INT_MASK                                                                     0x04000000L
+#define UVD_VCPU_INT_STATUS__CLK_SWT_INT_MASK                                                                 0x08000000L
+#define UVD_VCPU_INT_STATUS__MIF_HWINT_MASK                                                                   0x10000000L
+#define UVD_VCPU_INT_STATUS__MPRD_ERR_INT_MASK                                                                0x20000000L
+#define UVD_VCPU_INT_STATUS__DRV_FW_REQ_INT_MASK                                                              0x40000000L
+#define UVD_VCPU_INT_STATUS__DRV_FW_ACK_INT_MASK                                                              0x80000000L
+//UVD_VCPU_INT_ACK
+#define UVD_VCPU_INT_ACK__PIF_ADDR_ERR_ACK__SHIFT                                                             0x0
+#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK__SHIFT                                                  0x1
+#define UVD_VCPU_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK__SHIFT                                           0x2
+#define UVD_VCPU_INT_ACK__NJ_PF_RPT_ACK__SHIFT                                                                0x3
+#define UVD_VCPU_INT_ACK__SW_RB1_INT_ACK__SHIFT                                                               0x4
+#define UVD_VCPU_INT_ACK__SW_RB2_INT_ACK__SHIFT                                                               0x5
+#define UVD_VCPU_INT_ACK__RBC_REG_PRIV_FAULT_ACK__SHIFT                                                       0x6
+#define UVD_VCPU_INT_ACK__SW_RB3_INT_ACK__SHIFT                                                               0x7
+#define UVD_VCPU_INT_ACK__SW_RB4_INT_ACK__SHIFT                                                               0x9
+#define UVD_VCPU_INT_ACK__SW_RB5_INT_ACK__SHIFT                                                               0xa
+#define UVD_VCPU_INT_ACK__LBSI_ACK__SHIFT                                                                     0xb
+#define UVD_VCPU_INT_ACK__UDEC_ACK__SHIFT                                                                     0xc
+#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK__SHIFT                                                  0xd
+#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK__SHIFT                                            0xe
+#define UVD_VCPU_INT_ACK__SUVD_ACK__SHIFT                                                                     0xf
+#define UVD_VCPU_INT_ACK__RPTR_WR_ACK__SHIFT                                                                  0x10
+#define UVD_VCPU_INT_ACK__JOB_START_ACK__SHIFT                                                                0x11
+#define UVD_VCPU_INT_ACK__NJ_PF_ACK__SHIFT                                                                    0x12
+#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK__SHIFT                                                       0x17
+#define UVD_VCPU_INT_ACK__IDCT_ACK__SHIFT                                                                     0x18
+#define UVD_VCPU_INT_ACK__MPRD_ACK__SHIFT                                                                     0x19
+#define UVD_VCPU_INT_ACK__AVM_INT_ACK__SHIFT                                                                  0x1a
+#define UVD_VCPU_INT_ACK__CLK_SWT_ACK__SHIFT                                                                  0x1b
+#define UVD_VCPU_INT_ACK__MIF_HWINT_ACK__SHIFT                                                                0x1c
+#define UVD_VCPU_INT_ACK__MPRD_ERR_ACK__SHIFT                                                                 0x1d
+#define UVD_VCPU_INT_ACK__DRV_FW_REQ_ACK__SHIFT                                                               0x1e
+#define UVD_VCPU_INT_ACK__DRV_FW_ACK_ACK__SHIFT                                                               0x1f
+#define UVD_VCPU_INT_ACK__PIF_ADDR_ERR_ACK_MASK                                                               0x00000001L
+#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK_MASK                                                    0x00000002L
+#define UVD_VCPU_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK_MASK                                             0x00000004L
+#define UVD_VCPU_INT_ACK__NJ_PF_RPT_ACK_MASK                                                                  0x00000008L
+#define UVD_VCPU_INT_ACK__SW_RB1_INT_ACK_MASK                                                                 0x00000010L
+#define UVD_VCPU_INT_ACK__SW_RB2_INT_ACK_MASK                                                                 0x00000020L
+#define UVD_VCPU_INT_ACK__RBC_REG_PRIV_FAULT_ACK_MASK                                                         0x00000040L
+#define UVD_VCPU_INT_ACK__SW_RB3_INT_ACK_MASK                                                                 0x00000080L
+#define UVD_VCPU_INT_ACK__SW_RB4_INT_ACK_MASK                                                                 0x00000200L
+#define UVD_VCPU_INT_ACK__SW_RB5_INT_ACK_MASK                                                                 0x00000400L
+#define UVD_VCPU_INT_ACK__LBSI_ACK_MASK                                                                       0x00000800L
+#define UVD_VCPU_INT_ACK__UDEC_ACK_MASK                                                                       0x00001000L
+#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK_MASK                                                    0x00002000L
+#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK_MASK                                              0x00004000L
+#define UVD_VCPU_INT_ACK__SUVD_ACK_MASK                                                                       0x00008000L
+#define UVD_VCPU_INT_ACK__RPTR_WR_ACK_MASK                                                                    0x00010000L
+#define UVD_VCPU_INT_ACK__JOB_START_ACK_MASK                                                                  0x00020000L
+#define UVD_VCPU_INT_ACK__NJ_PF_ACK_MASK                                                                      0x00040000L
+#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK_MASK                                                         0x00800000L
+#define UVD_VCPU_INT_ACK__IDCT_ACK_MASK                                                                       0x01000000L
+#define UVD_VCPU_INT_ACK__MPRD_ACK_MASK                                                                       0x02000000L
+#define UVD_VCPU_INT_ACK__AVM_INT_ACK_MASK                                                                    0x04000000L
+#define UVD_VCPU_INT_ACK__CLK_SWT_ACK_MASK                                                                    0x08000000L
+#define UVD_VCPU_INT_ACK__MIF_HWINT_ACK_MASK                                                                  0x10000000L
+#define UVD_VCPU_INT_ACK__MPRD_ERR_ACK_MASK                                                                   0x20000000L
+#define UVD_VCPU_INT_ACK__DRV_FW_REQ_ACK_MASK                                                                 0x40000000L
+#define UVD_VCPU_INT_ACK__DRV_FW_ACK_ACK_MASK                                                                 0x80000000L
+//UVD_VCPU_INT_ROUTE
+#define UVD_VCPU_INT_ROUTE__DRV_FW_MSG__SHIFT                                                                 0x0
+#define UVD_VCPU_INT_ROUTE__FW_DRV_MSG_ACK__SHIFT                                                             0x1
+#define UVD_VCPU_INT_ROUTE__VCPU_GPCOM__SHIFT                                                                 0x2
+#define UVD_VCPU_INT_ROUTE__DRV_FW_MSG_MASK                                                                   0x00000001L
+#define UVD_VCPU_INT_ROUTE__FW_DRV_MSG_ACK_MASK                                                               0x00000002L
+#define UVD_VCPU_INT_ROUTE__VCPU_GPCOM_MASK                                                                   0x00000004L
+//UVD_DRV_FW_MSG
+#define UVD_DRV_FW_MSG__MSG__SHIFT                                                                            0x0
+#define UVD_DRV_FW_MSG__MSG_MASK                                                                              0xFFFFFFFFL
+//UVD_FW_DRV_MSG_ACK
+#define UVD_FW_DRV_MSG_ACK__ACK__SHIFT                                                                        0x0
+#define UVD_FW_DRV_MSG_ACK__ACK_MASK                                                                          0x00000001L
+//UVD_SUVD_INT_EN
+#define UVD_SUVD_INT_EN__SRE_FUNC_INT_EN__SHIFT                                                               0x0
+#define UVD_SUVD_INT_EN__SRE_ERR_INT_EN__SHIFT                                                                0x5
+#define UVD_SUVD_INT_EN__SIT_FUNC_INT_EN__SHIFT                                                               0x6
+#define UVD_SUVD_INT_EN__SIT_ERR_INT_EN__SHIFT                                                                0xb
+#define UVD_SUVD_INT_EN__SMP_FUNC_INT_EN__SHIFT                                                               0xc
+#define UVD_SUVD_INT_EN__SMP_ERR_INT_EN__SHIFT                                                                0x11
+#define UVD_SUVD_INT_EN__SCM_FUNC_INT_EN__SHIFT                                                               0x12
+#define UVD_SUVD_INT_EN__SCM_ERR_INT_EN__SHIFT                                                                0x17
+#define UVD_SUVD_INT_EN__SDB_FUNC_INT_EN__SHIFT                                                               0x18
+#define UVD_SUVD_INT_EN__SDB_ERR_INT_EN__SHIFT                                                                0x1d
+#define UVD_SUVD_INT_EN__FBC_ERR_INT_EN__SHIFT                                                                0x1e
+#define UVD_SUVD_INT_EN__SRE_FUNC_INT_EN_MASK                                                                 0x0000001FL
+#define UVD_SUVD_INT_EN__SRE_ERR_INT_EN_MASK                                                                  0x00000020L
+#define UVD_SUVD_INT_EN__SIT_FUNC_INT_EN_MASK                                                                 0x000007C0L
+#define UVD_SUVD_INT_EN__SIT_ERR_INT_EN_MASK                                                                  0x00000800L
+#define UVD_SUVD_INT_EN__SMP_FUNC_INT_EN_MASK                                                                 0x0001F000L
+#define UVD_SUVD_INT_EN__SMP_ERR_INT_EN_MASK                                                                  0x00020000L
+#define UVD_SUVD_INT_EN__SCM_FUNC_INT_EN_MASK                                                                 0x007C0000L
+#define UVD_SUVD_INT_EN__SCM_ERR_INT_EN_MASK                                                                  0x00800000L
+#define UVD_SUVD_INT_EN__SDB_FUNC_INT_EN_MASK                                                                 0x1F000000L
+#define UVD_SUVD_INT_EN__SDB_ERR_INT_EN_MASK                                                                  0x20000000L
+#define UVD_SUVD_INT_EN__FBC_ERR_INT_EN_MASK                                                                  0x40000000L
+//UVD_SUVD_INT_STATUS
+#define UVD_SUVD_INT_STATUS__SRE_FUNC_INT__SHIFT                                                              0x0
+#define UVD_SUVD_INT_STATUS__SRE_ERR_INT__SHIFT                                                               0x5
+#define UVD_SUVD_INT_STATUS__SIT_FUNC_INT__SHIFT                                                              0x6
+#define UVD_SUVD_INT_STATUS__SIT_ERR_INT__SHIFT                                                               0xb
+#define UVD_SUVD_INT_STATUS__SMP_FUNC_INT__SHIFT                                                              0xc
+#define UVD_SUVD_INT_STATUS__SMP_ERR_INT__SHIFT                                                               0x11
+#define UVD_SUVD_INT_STATUS__SCM_FUNC_INT__SHIFT                                                              0x12
+#define UVD_SUVD_INT_STATUS__SCM_ERR_INT__SHIFT                                                               0x17
+#define UVD_SUVD_INT_STATUS__SDB_FUNC_INT__SHIFT                                                              0x18
+#define UVD_SUVD_INT_STATUS__SDB_ERR_INT__SHIFT                                                               0x1d
+#define UVD_SUVD_INT_STATUS__FBC_ERR_INT__SHIFT                                                               0x1e
+#define UVD_SUVD_INT_STATUS__SRE_FUNC_INT_MASK                                                                0x0000001FL
+#define UVD_SUVD_INT_STATUS__SRE_ERR_INT_MASK                                                                 0x00000020L
+#define UVD_SUVD_INT_STATUS__SIT_FUNC_INT_MASK                                                                0x000007C0L
+#define UVD_SUVD_INT_STATUS__SIT_ERR_INT_MASK                                                                 0x00000800L
+#define UVD_SUVD_INT_STATUS__SMP_FUNC_INT_MASK                                                                0x0001F000L
+#define UVD_SUVD_INT_STATUS__SMP_ERR_INT_MASK                                                                 0x00020000L
+#define UVD_SUVD_INT_STATUS__SCM_FUNC_INT_MASK                                                                0x007C0000L
+#define UVD_SUVD_INT_STATUS__SCM_ERR_INT_MASK                                                                 0x00800000L
+#define UVD_SUVD_INT_STATUS__SDB_FUNC_INT_MASK                                                                0x1F000000L
+#define UVD_SUVD_INT_STATUS__SDB_ERR_INT_MASK                                                                 0x20000000L
+#define UVD_SUVD_INT_STATUS__FBC_ERR_INT_MASK                                                                 0x40000000L
+//UVD_SUVD_INT_ACK
+#define UVD_SUVD_INT_ACK__SRE_FUNC_INT_ACK__SHIFT                                                             0x0
+#define UVD_SUVD_INT_ACK__SRE_ERR_INT_ACK__SHIFT                                                              0x5
+#define UVD_SUVD_INT_ACK__SIT_FUNC_INT_ACK__SHIFT                                                             0x6
+#define UVD_SUVD_INT_ACK__SIT_ERR_INT_ACK__SHIFT                                                              0xb
+#define UVD_SUVD_INT_ACK__SMP_FUNC_INT_ACK__SHIFT                                                             0xc
+#define UVD_SUVD_INT_ACK__SMP_ERR_INT_ACK__SHIFT                                                              0x11
+#define UVD_SUVD_INT_ACK__SCM_FUNC_INT_ACK__SHIFT                                                             0x12
+#define UVD_SUVD_INT_ACK__SCM_ERR_INT_ACK__SHIFT                                                              0x17
+#define UVD_SUVD_INT_ACK__SDB_FUNC_INT_ACK__SHIFT                                                             0x18
+#define UVD_SUVD_INT_ACK__SDB_ERR_INT_ACK__SHIFT                                                              0x1d
+#define UVD_SUVD_INT_ACK__FBC_ERR_INT_ACK__SHIFT                                                              0x1e
+#define UVD_SUVD_INT_ACK__SRE_FUNC_INT_ACK_MASK                                                               0x0000001FL
+#define UVD_SUVD_INT_ACK__SRE_ERR_INT_ACK_MASK                                                                0x00000020L
+#define UVD_SUVD_INT_ACK__SIT_FUNC_INT_ACK_MASK                                                               0x000007C0L
+#define UVD_SUVD_INT_ACK__SIT_ERR_INT_ACK_MASK                                                                0x00000800L
+#define UVD_SUVD_INT_ACK__SMP_FUNC_INT_ACK_MASK                                                               0x0001F000L
+#define UVD_SUVD_INT_ACK__SMP_ERR_INT_ACK_MASK                                                                0x00020000L
+#define UVD_SUVD_INT_ACK__SCM_FUNC_INT_ACK_MASK                                                               0x007C0000L
+#define UVD_SUVD_INT_ACK__SCM_ERR_INT_ACK_MASK                                                                0x00800000L
+#define UVD_SUVD_INT_ACK__SDB_FUNC_INT_ACK_MASK                                                               0x1F000000L
+#define UVD_SUVD_INT_ACK__SDB_ERR_INT_ACK_MASK                                                                0x20000000L
+#define UVD_SUVD_INT_ACK__FBC_ERR_INT_ACK_MASK                                                                0x40000000L
+//UVD_ENC_VCPU_INT_EN
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR_EN__SHIFT                                                 0x0
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR2_EN__SHIFT                                                0x1
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR3_EN__SHIFT                                                0x2
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR_EN_MASK                                                   0x00000001L
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR2_EN_MASK                                                  0x00000002L
+#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR3_EN_MASK                                                  0x00000004L
+//UVD_ENC_VCPU_INT_STATUS
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR_INT__SHIFT                                            0x0
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR2_INT__SHIFT                                           0x1
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR3_INT__SHIFT                                           0x2
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR_INT_MASK                                              0x00000001L
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR2_INT_MASK                                             0x00000002L
+#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR3_INT_MASK                                             0x00000004L
+//UVD_ENC_VCPU_INT_ACK
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR_ACK__SHIFT                                               0x0
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR2_ACK__SHIFT                                              0x1
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR3_ACK__SHIFT                                              0x2
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR_ACK_MASK                                                 0x00000001L
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR2_ACK_MASK                                                0x00000002L
+#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR3_ACK_MASK                                                0x00000004L
+//UVD_MASTINT_EN
+#define UVD_MASTINT_EN__OVERRUN_RST__SHIFT                                                                    0x0
+#define UVD_MASTINT_EN__VCPU_EN__SHIFT                                                                        0x1
+#define UVD_MASTINT_EN__SYS_EN__SHIFT                                                                         0x2
+#define UVD_MASTINT_EN__INT_OVERRUN__SHIFT                                                                    0x4
+#define UVD_MASTINT_EN__OVERRUN_RST_MASK                                                                      0x00000001L
+#define UVD_MASTINT_EN__VCPU_EN_MASK                                                                          0x00000002L
+#define UVD_MASTINT_EN__SYS_EN_MASK                                                                           0x00000004L
+#define UVD_MASTINT_EN__INT_OVERRUN_MASK                                                                      0x00FFFFF0L
+//UVD_SYS_INT_EN
+#define UVD_SYS_INT_EN__PIF_ADDR_ERR_EN__SHIFT                                                                0x0
+#define UVD_SYS_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN__SHIFT                                                     0x1
+#define UVD_SYS_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN__SHIFT                                              0x2
+#define UVD_SYS_INT_EN__CXW_WR_EN__SHIFT                                                                      0x3
+#define UVD_SYS_INT_EN__RBC_REG_PRIV_FAULT_EN__SHIFT                                                          0x6
+#define UVD_SYS_INT_EN__LBSI_EN__SHIFT                                                                        0xb
+#define UVD_SYS_INT_EN__UDEC_EN__SHIFT                                                                        0xc
+#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN__SHIFT                                                     0xd
+#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN__SHIFT                                               0xe
+#define UVD_SYS_INT_EN__SUVD_EN__SHIFT                                                                        0xf
+#define UVD_SYS_INT_EN__JOB_DONE_EN__SHIFT                                                                    0x10
+#define UVD_SYS_INT_EN__SEMA_WAIT_FAIL_SIG_EN__SHIFT                                                          0x17
+#define UVD_SYS_INT_EN__IDCT_EN__SHIFT                                                                        0x18
+#define UVD_SYS_INT_EN__MPRD_EN__SHIFT                                                                        0x19
+#define UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN__SHIFT                                                         0x1a
+#define UVD_SYS_INT_EN__CLK_SWT_EN__SHIFT                                                                     0x1b
+#define UVD_SYS_INT_EN__MIF_HWINT_EN__SHIFT                                                                   0x1c
+#define UVD_SYS_INT_EN__MPRD_ERR_EN__SHIFT                                                                    0x1d
+#define UVD_SYS_INT_EN__AVM_INT_EN__SHIFT                                                                     0x1f
+#define UVD_SYS_INT_EN__PIF_ADDR_ERR_EN_MASK                                                                  0x00000001L
+#define UVD_SYS_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN_MASK                                                       0x00000002L
+#define UVD_SYS_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN_MASK                                                0x00000004L
+#define UVD_SYS_INT_EN__CXW_WR_EN_MASK                                                                        0x00000008L
+#define UVD_SYS_INT_EN__RBC_REG_PRIV_FAULT_EN_MASK                                                            0x00000040L
+#define UVD_SYS_INT_EN__LBSI_EN_MASK                                                                          0x00000800L
+#define UVD_SYS_INT_EN__UDEC_EN_MASK                                                                          0x00001000L
+#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN_MASK                                                       0x00002000L
+#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN_MASK                                                 0x00004000L
+#define UVD_SYS_INT_EN__SUVD_EN_MASK                                                                          0x00008000L
+#define UVD_SYS_INT_EN__JOB_DONE_EN_MASK                                                                      0x00010000L
+#define UVD_SYS_INT_EN__SEMA_WAIT_FAIL_SIG_EN_MASK                                                            0x00800000L
+#define UVD_SYS_INT_EN__IDCT_EN_MASK                                                                          0x01000000L
+#define UVD_SYS_INT_EN__MPRD_EN_MASK                                                                          0x02000000L
+#define UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK                                                           0x04000000L
+#define UVD_SYS_INT_EN__CLK_SWT_EN_MASK                                                                       0x08000000L
+#define UVD_SYS_INT_EN__MIF_HWINT_EN_MASK                                                                     0x10000000L
+#define UVD_SYS_INT_EN__MPRD_ERR_EN_MASK                                                                      0x20000000L
+#define UVD_SYS_INT_EN__AVM_INT_EN_MASK                                                                       0x80000000L
+//UVD_SYS_INT_STATUS
+#define UVD_SYS_INT_STATUS__PIF_ADDR_ERR_INT__SHIFT                                                           0x0
+#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT__SHIFT                                                0x1
+#define UVD_SYS_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT__SHIFT                                         0x2
+#define UVD_SYS_INT_STATUS__CXW_WR_INT__SHIFT                                                                 0x3
+#define UVD_SYS_INT_STATUS__RBC_REG_PRIV_FAULT_INT__SHIFT                                                     0x6
+#define UVD_SYS_INT_STATUS__LBSI_INT__SHIFT                                                                   0xb
+#define UVD_SYS_INT_STATUS__UDEC_INT__SHIFT                                                                   0xc
+#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT__SHIFT                                                0xd
+#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT__SHIFT                                          0xe
+#define UVD_SYS_INT_STATUS__SUVD_INT__SHIFT                                                                   0xf
+#define UVD_SYS_INT_STATUS__JOB_DONE_INT__SHIFT                                                               0x10
+#define UVD_SYS_INT_STATUS__GPCOM_INT__SHIFT                                                                  0x12
+#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT__SHIFT                                                     0x17
+#define UVD_SYS_INT_STATUS__IDCT_INT__SHIFT                                                                   0x18
+#define UVD_SYS_INT_STATUS__MPRD_INT__SHIFT                                                                   0x19
+#define UVD_SYS_INT_STATUS__CLK_SWT_INT__SHIFT                                                                0x1b
+#define UVD_SYS_INT_STATUS__MIF_HWINT__SHIFT                                                                  0x1c
+#define UVD_SYS_INT_STATUS__MPRD_ERR_INT__SHIFT                                                               0x1d
+#define UVD_SYS_INT_STATUS__RASCNTL_VCPU_VCODEC_INT__SHIFT                                                    0x1e
+#define UVD_SYS_INT_STATUS__AVM_INT__SHIFT                                                                    0x1f
+#define UVD_SYS_INT_STATUS__PIF_ADDR_ERR_INT_MASK                                                             0x00000001L
+#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT_MASK                                                  0x00000002L
+#define UVD_SYS_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT_MASK                                           0x00000004L
+#define UVD_SYS_INT_STATUS__CXW_WR_INT_MASK                                                                   0x00000008L
+#define UVD_SYS_INT_STATUS__RBC_REG_PRIV_FAULT_INT_MASK                                                       0x00000040L
+#define UVD_SYS_INT_STATUS__LBSI_INT_MASK                                                                     0x00000800L
+#define UVD_SYS_INT_STATUS__UDEC_INT_MASK                                                                     0x00001000L
+#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT_MASK                                                  0x00002000L
+#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT_MASK                                            0x00004000L
+#define UVD_SYS_INT_STATUS__SUVD_INT_MASK                                                                     0x00008000L
+#define UVD_SYS_INT_STATUS__JOB_DONE_INT_MASK                                                                 0x00010000L
+#define UVD_SYS_INT_STATUS__GPCOM_INT_MASK                                                                    0x00040000L
+#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT_MASK                                                       0x00800000L
+#define UVD_SYS_INT_STATUS__IDCT_INT_MASK                                                                     0x01000000L
+#define UVD_SYS_INT_STATUS__MPRD_INT_MASK                                                                     0x02000000L
+#define UVD_SYS_INT_STATUS__CLK_SWT_INT_MASK                                                                  0x08000000L
+#define UVD_SYS_INT_STATUS__MIF_HWINT_MASK                                                                    0x10000000L
+#define UVD_SYS_INT_STATUS__MPRD_ERR_INT_MASK                                                                 0x20000000L
+#define UVD_SYS_INT_STATUS__RASCNTL_VCPU_VCODEC_INT_MASK                                                      0x40000000L
+#define UVD_SYS_INT_STATUS__AVM_INT_MASK                                                                      0x80000000L
+//UVD_SYS_INT_ACK
+#define UVD_SYS_INT_ACK__PIF_ADDR_ERR_ACK__SHIFT                                                              0x0
+#define UVD_SYS_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK__SHIFT                                                   0x1
+#define UVD_SYS_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK__SHIFT                                            0x2
+#define UVD_SYS_INT_ACK__CXW_WR_ACK__SHIFT                                                                    0x3
+#define UVD_SYS_INT_ACK__RBC_REG_PRIV_FAULT_ACK__SHIFT                                                        0x6
+#define UVD_SYS_INT_ACK__LBSI_ACK__SHIFT                                                                      0xb
+#define UVD_SYS_INT_ACK__UDEC_ACK__SHIFT                                                                      0xc
+#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK__SHIFT                                                   0xd
+#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK__SHIFT                                             0xe
+#define UVD_SYS_INT_ACK__SUVD_ACK__SHIFT                                                                      0xf
+#define UVD_SYS_INT_ACK__JOB_DONE_ACK__SHIFT                                                                  0x10
+#define UVD_SYS_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK__SHIFT                                                        0x17
+#define UVD_SYS_INT_ACK__IDCT_ACK__SHIFT                                                                      0x18
+#define UVD_SYS_INT_ACK__MPRD_ACK__SHIFT                                                                      0x19
+#define UVD_SYS_INT_ACK__CLK_SWT_ACK__SHIFT                                                                   0x1b
+#define UVD_SYS_INT_ACK__MIF_HWINT_ACK__SHIFT                                                                 0x1c
+#define UVD_SYS_INT_ACK__MPRD_ERR_ACK__SHIFT                                                                  0x1d
+#define UVD_SYS_INT_ACK__RASCNTL_VCPU_VCODEC_ACK__SHIFT                                                       0x1e
+#define UVD_SYS_INT_ACK__AVM_INT_ACK__SHIFT                                                                   0x1f
+#define UVD_SYS_INT_ACK__PIF_ADDR_ERR_ACK_MASK                                                                0x00000001L
+#define UVD_SYS_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK_MASK                                                     0x00000002L
+#define UVD_SYS_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK_MASK                                              0x00000004L
+#define UVD_SYS_INT_ACK__CXW_WR_ACK_MASK                                                                      0x00000008L
+#define UVD_SYS_INT_ACK__RBC_REG_PRIV_FAULT_ACK_MASK                                                          0x00000040L
+#define UVD_SYS_INT_ACK__LBSI_ACK_MASK                                                                        0x00000800L
+#define UVD_SYS_INT_ACK__UDEC_ACK_MASK                                                                        0x00001000L
+#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK_MASK                                                     0x00002000L
+#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK_MASK                                               0x00004000L
+#define UVD_SYS_INT_ACK__SUVD_ACK_MASK                                                                        0x00008000L
+#define UVD_SYS_INT_ACK__JOB_DONE_ACK_MASK                                                                    0x00010000L
+#define UVD_SYS_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK_MASK                                                          0x00800000L
+#define UVD_SYS_INT_ACK__IDCT_ACK_MASK                                                                        0x01000000L
+#define UVD_SYS_INT_ACK__MPRD_ACK_MASK                                                                        0x02000000L
+#define UVD_SYS_INT_ACK__CLK_SWT_ACK_MASK                                                                     0x08000000L
+#define UVD_SYS_INT_ACK__MIF_HWINT_ACK_MASK                                                                   0x10000000L
+#define UVD_SYS_INT_ACK__MPRD_ERR_ACK_MASK                                                                    0x20000000L
+#define UVD_SYS_INT_ACK__RASCNTL_VCPU_VCODEC_ACK_MASK                                                         0x40000000L
+#define UVD_SYS_INT_ACK__AVM_INT_ACK_MASK                                                                     0x80000000L
+//UVD_JOB_DONE
+#define UVD_JOB_DONE__JOB_DONE__SHIFT                                                                         0x0
+#define UVD_JOB_DONE__JOB_DONE_MASK                                                                           0x00000003L
+//UVD_CBUF_ID
+#define UVD_CBUF_ID__CBUF_ID__SHIFT                                                                           0x0
+#define UVD_CBUF_ID__CBUF_ID_MASK                                                                             0xFFFFFFFFL
+//UVD_CONTEXT_ID
+#define UVD_CONTEXT_ID__CONTEXT_ID__SHIFT                                                                     0x0
+#define UVD_CONTEXT_ID__CONTEXT_ID_MASK                                                                       0xFFFFFFFFL
+//UVD_CONTEXT_ID2
+#define UVD_CONTEXT_ID2__CONTEXT_ID2__SHIFT                                                                   0x0
+#define UVD_CONTEXT_ID2__CONTEXT_ID2_MASK                                                                     0xFFFFFFFFL
+//UVD_NO_OP
+#define UVD_NO_OP__NO_OP__SHIFT                                                                               0x0
+#define UVD_NO_OP__NO_OP_MASK                                                                                 0xFFFFFFFFL
+//UVD_RB_BASE_LO
+#define UVD_RB_BASE_LO__RB_BASE_LO__SHIFT                                                                     0x6
+#define UVD_RB_BASE_LO__RB_BASE_LO_MASK                                                                       0xFFFFFFC0L
+//UVD_RB_BASE_HI
+#define UVD_RB_BASE_HI__RB_BASE_HI__SHIFT                                                                     0x0
+#define UVD_RB_BASE_HI__RB_BASE_HI_MASK                                                                       0xFFFFFFFFL
+//UVD_RB_SIZE
+#define UVD_RB_SIZE__RB_SIZE__SHIFT                                                                           0x4
+#define UVD_RB_SIZE__RB_SIZE_MASK                                                                             0x007FFFF0L
+//UVD_RB_BASE_LO2
+#define UVD_RB_BASE_LO2__RB_BASE_LO__SHIFT                                                                    0x6
+#define UVD_RB_BASE_LO2__RB_BASE_LO_MASK                                                                      0xFFFFFFC0L
+//UVD_RB_BASE_HI2
+#define UVD_RB_BASE_HI2__RB_BASE_HI__SHIFT                                                                    0x0
+#define UVD_RB_BASE_HI2__RB_BASE_HI_MASK                                                                      0xFFFFFFFFL
+//UVD_RB_SIZE2
+#define UVD_RB_SIZE2__RB_SIZE__SHIFT                                                                          0x4
+#define UVD_RB_SIZE2__RB_SIZE_MASK                                                                            0x007FFFF0L
+//UVD_RB_BASE_LO3
+#define UVD_RB_BASE_LO3__RB_BASE_LO__SHIFT                                                                    0x6
+#define UVD_RB_BASE_LO3__RB_BASE_LO_MASK                                                                      0xFFFFFFC0L
+//UVD_RB_BASE_HI3
+#define UVD_RB_BASE_HI3__RB_BASE_HI__SHIFT                                                                    0x0
+#define UVD_RB_BASE_HI3__RB_BASE_HI_MASK                                                                      0xFFFFFFFFL
+//UVD_RB_SIZE3
+#define UVD_RB_SIZE3__RB_SIZE__SHIFT                                                                          0x4
+#define UVD_RB_SIZE3__RB_SIZE_MASK                                                                            0x007FFFF0L
+//UVD_RB_BASE_LO4
+#define UVD_RB_BASE_LO4__RB_BASE_LO__SHIFT                                                                    0x6
+#define UVD_RB_BASE_LO4__RB_BASE_LO_MASK                                                                      0xFFFFFFC0L
+//UVD_RB_BASE_HI4
+#define UVD_RB_BASE_HI4__RB_BASE_HI__SHIFT                                                                    0x0
+#define UVD_RB_BASE_HI4__RB_BASE_HI_MASK                                                                      0xFFFFFFFFL
+//UVD_RB_SIZE4
+#define UVD_RB_SIZE4__RB_SIZE__SHIFT                                                                          0x4
+#define UVD_RB_SIZE4__RB_SIZE_MASK                                                                            0x007FFFF0L
+//UVD_OUT_RB_BASE_LO
+#define UVD_OUT_RB_BASE_LO__RB_BASE_LO__SHIFT                                                                 0x6
+#define UVD_OUT_RB_BASE_LO__RB_BASE_LO_MASK                                                                   0xFFFFFFC0L
+//UVD_OUT_RB_BASE_HI
+#define UVD_OUT_RB_BASE_HI__RB_BASE_HI__SHIFT                                                                 0x0
+#define UVD_OUT_RB_BASE_HI__RB_BASE_HI_MASK                                                                   0xFFFFFFFFL
+//UVD_OUT_RB_SIZE
+#define UVD_OUT_RB_SIZE__RB_SIZE__SHIFT                                                                       0x4
+#define UVD_OUT_RB_SIZE__RB_SIZE_MASK                                                                         0x007FFFF0L
+//UVD_IOV_ACTIVE_FCN_ID
+#define UVD_IOV_ACTIVE_FCN_ID__VF_ID__SHIFT                                                                   0x0
+#define UVD_IOV_ACTIVE_FCN_ID__PF_VF__SHIFT                                                                   0x1f
+#define UVD_IOV_ACTIVE_FCN_ID__VF_ID_MASK                                                                     0x0000003FL
+#define UVD_IOV_ACTIVE_FCN_ID__PF_VF_MASK                                                                     0x80000000L
+//UVD_IOV_MAILBOX
+#define UVD_IOV_MAILBOX__MAILBOX__SHIFT                                                                       0x0
+#define UVD_IOV_MAILBOX__MAILBOX_MASK                                                                         0xFFFFFFFFL
+//UVD_IOV_MAILBOX_RESP
+#define UVD_IOV_MAILBOX_RESP__RESP__SHIFT                                                                     0x0
+#define UVD_IOV_MAILBOX_RESP__RESP_MASK                                                                       0xFFFFFFFFL
+//UVD_RB_ARB_CTRL
+#define UVD_RB_ARB_CTRL__SRBM_DROP__SHIFT                                                                     0x0
+#define UVD_RB_ARB_CTRL__SRBM_DIS__SHIFT                                                                      0x1
+#define UVD_RB_ARB_CTRL__VCPU_DROP__SHIFT                                                                     0x2
+#define UVD_RB_ARB_CTRL__VCPU_DIS__SHIFT                                                                      0x3
+#define UVD_RB_ARB_CTRL__RBC_DROP__SHIFT                                                                      0x4
+#define UVD_RB_ARB_CTRL__RBC_DIS__SHIFT                                                                       0x5
+#define UVD_RB_ARB_CTRL__FWOFLD_DROP__SHIFT                                                                   0x6
+#define UVD_RB_ARB_CTRL__FWOFLD_DIS__SHIFT                                                                    0x7
+#define UVD_RB_ARB_CTRL__FAST_PATH_EN__SHIFT                                                                  0x8
+#define UVD_RB_ARB_CTRL__UVD_RB_DBG_EN__SHIFT                                                                 0x9
+#define UVD_RB_ARB_CTRL__SRBM_DROP_MASK                                                                       0x00000001L
+#define UVD_RB_ARB_CTRL__SRBM_DIS_MASK                                                                        0x00000002L
+#define UVD_RB_ARB_CTRL__VCPU_DROP_MASK                                                                       0x00000004L
+#define UVD_RB_ARB_CTRL__VCPU_DIS_MASK                                                                        0x00000008L
+#define UVD_RB_ARB_CTRL__RBC_DROP_MASK                                                                        0x00000010L
+#define UVD_RB_ARB_CTRL__RBC_DIS_MASK                                                                         0x00000020L
+#define UVD_RB_ARB_CTRL__FWOFLD_DROP_MASK                                                                     0x00000040L
+#define UVD_RB_ARB_CTRL__FWOFLD_DIS_MASK                                                                      0x00000080L
+#define UVD_RB_ARB_CTRL__FAST_PATH_EN_MASK                                                                    0x00000100L
+#define UVD_RB_ARB_CTRL__UVD_RB_DBG_EN_MASK                                                                   0x00000200L
+//UVD_CTX_INDEX
+#define UVD_CTX_INDEX__INDEX__SHIFT                                                                           0x0
+#define UVD_CTX_INDEX__INDEX_MASK                                                                             0x000001FFL
+//UVD_CTX_DATA
+#define UVD_CTX_DATA__DATA__SHIFT                                                                             0x0
+#define UVD_CTX_DATA__DATA_MASK                                                                               0xFFFFFFFFL
+//UVD_CXW_WR
+#define UVD_CXW_WR__DAT__SHIFT                                                                                0x0
+#define UVD_CXW_WR__STAT__SHIFT                                                                               0x1f
+#define UVD_CXW_WR__DAT_MASK                                                                                  0x0FFFFFFFL
+#define UVD_CXW_WR__STAT_MASK                                                                                 0x80000000L
+//UVD_CXW_WR_INT_ID
+#define UVD_CXW_WR_INT_ID__ID__SHIFT                                                                          0x0
+#define UVD_CXW_WR_INT_ID__ID_MASK                                                                            0x000000FFL
+//UVD_CXW_WR_INT_CTX_ID
+#define UVD_CXW_WR_INT_CTX_ID__ID__SHIFT                                                                      0x0
+#define UVD_CXW_WR_INT_CTX_ID__ID_MASK                                                                        0x0FFFFFFFL
+//UVD_CXW_INT_ID
+#define UVD_CXW_INT_ID__ID__SHIFT                                                                             0x0
+#define UVD_CXW_INT_ID__ID_MASK                                                                               0x000000FFL
+//UVD_MPEG2_ERROR
+#define UVD_MPEG2_ERROR__STATUS__SHIFT                                                                        0x0
+#define UVD_MPEG2_ERROR__STATUS_MASK                                                                          0xFFFFFFFFL
+//UVD_YBASE
+#define UVD_YBASE__DUM__SHIFT                                                                                 0x0
+#define UVD_YBASE__DUM_MASK                                                                                   0xFFFFFFFFL
+//UVD_UVBASE
+#define UVD_UVBASE__DUM__SHIFT                                                                                0x0
+#define UVD_UVBASE__DUM_MASK                                                                                  0xFFFFFFFFL
+//UVD_PITCH
+#define UVD_PITCH__DUM__SHIFT                                                                                 0x0
+#define UVD_PITCH__DUM_MASK                                                                                   0xFFFFFFFFL
+//UVD_WIDTH
+#define UVD_WIDTH__DUM__SHIFT                                                                                 0x0
+#define UVD_WIDTH__DUM_MASK                                                                                   0xFFFFFFFFL
+//UVD_HEIGHT
+#define UVD_HEIGHT__DUM__SHIFT                                                                                0x0
+#define UVD_HEIGHT__DUM_MASK                                                                                  0xFFFFFFFFL
+//UVD_PICCOUNT
+#define UVD_PICCOUNT__DUM__SHIFT                                                                              0x0
+#define UVD_PICCOUNT__DUM_MASK                                                                                0xFFFFFFFFL
+//UVD_MPRD_INITIAL_XY
+#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_X__SHIFT                                                             0x0
+#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_Y__SHIFT                                                             0x10
+#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_X_MASK                                                               0x00000FFFL
+#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_Y_MASK                                                               0x0FFF0000L
+//UVD_MPEG2_CTRL
+#define UVD_MPEG2_CTRL__EN__SHIFT                                                                             0x0
+#define UVD_MPEG2_CTRL__TRICK_MODE__SHIFT                                                                     0x1
+#define UVD_MPEG2_CTRL__NUM_MB_PER_JOB__SHIFT                                                                 0x10
+#define UVD_MPEG2_CTRL__EN_MASK                                                                               0x00000001L
+#define UVD_MPEG2_CTRL__TRICK_MODE_MASK                                                                       0x00000002L
+#define UVD_MPEG2_CTRL__NUM_MB_PER_JOB_MASK                                                                   0xFFFF0000L
+//UVD_MB_CTL_BUF_BASE
+#define UVD_MB_CTL_BUF_BASE__BASE__SHIFT                                                                      0x0
+#define UVD_MB_CTL_BUF_BASE__BASE_MASK                                                                        0xFFFFFFFFL
+//UVD_PIC_CTL_BUF_BASE
+#define UVD_PIC_CTL_BUF_BASE__BASE__SHIFT                                                                     0x0
+#define UVD_PIC_CTL_BUF_BASE__BASE_MASK                                                                       0xFFFFFFFFL
+//UVD_DXVA_BUF_SIZE
+#define UVD_DXVA_BUF_SIZE__PIC_SIZE__SHIFT                                                                    0x0
+#define UVD_DXVA_BUF_SIZE__MB_SIZE__SHIFT                                                                     0x10
+#define UVD_DXVA_BUF_SIZE__PIC_SIZE_MASK                                                                      0x0000FFFFL
+#define UVD_DXVA_BUF_SIZE__MB_SIZE_MASK                                                                       0xFFFF0000L
+//UVD_SCRATCH_NP
+#define UVD_SCRATCH_NP__DATA__SHIFT                                                                           0x0
+#define UVD_SCRATCH_NP__DATA_MASK                                                                             0xFFFFFFFFL
+//UVD_CLK_SWT_HANDSHAKE
+#define UVD_CLK_SWT_HANDSHAKE__CLK_SWT_TYPE__SHIFT                                                            0x0
+#define UVD_CLK_SWT_HANDSHAKE__CLK_DOMAIN_SWT__SHIFT                                                          0x8
+#define UVD_CLK_SWT_HANDSHAKE__CLK_SWT_TYPE_MASK                                                              0x00000003L
+#define UVD_CLK_SWT_HANDSHAKE__CLK_DOMAIN_SWT_MASK                                                            0x00000300L
+//UVD_GP_SCRATCH0
+#define UVD_GP_SCRATCH0__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH0__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH1
+#define UVD_GP_SCRATCH1__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH1__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH2
+#define UVD_GP_SCRATCH2__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH2__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH3
+#define UVD_GP_SCRATCH3__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH3__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH4
+#define UVD_GP_SCRATCH4__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH4__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH5
+#define UVD_GP_SCRATCH5__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH5__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH6
+#define UVD_GP_SCRATCH6__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH6__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH7
+#define UVD_GP_SCRATCH7__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH7__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH8
+#define UVD_GP_SCRATCH8__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH8__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH9
+#define UVD_GP_SCRATCH9__DATA__SHIFT                                                                          0x0
+#define UVD_GP_SCRATCH9__DATA_MASK                                                                            0xFFFFFFFFL
+//UVD_GP_SCRATCH10
+#define UVD_GP_SCRATCH10__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH10__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH11
+#define UVD_GP_SCRATCH11__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH11__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH12
+#define UVD_GP_SCRATCH12__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH12__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH13
+#define UVD_GP_SCRATCH13__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH13__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH14
+#define UVD_GP_SCRATCH14__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH14__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH15
+#define UVD_GP_SCRATCH15__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH15__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH16
+#define UVD_GP_SCRATCH16__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH16__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH17
+#define UVD_GP_SCRATCH17__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH17__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH18
+#define UVD_GP_SCRATCH18__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH18__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH19
+#define UVD_GP_SCRATCH19__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH19__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH20
+#define UVD_GP_SCRATCH20__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH20__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH21
+#define UVD_GP_SCRATCH21__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH21__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH22
+#define UVD_GP_SCRATCH22__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH22__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_GP_SCRATCH23
+#define UVD_GP_SCRATCH23__DATA__SHIFT                                                                         0x0
+#define UVD_GP_SCRATCH23__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_AUDIO_RB_BASE_LO
+#define UVD_AUDIO_RB_BASE_LO__RB_BASE_LO__SHIFT                                                               0x6
+#define UVD_AUDIO_RB_BASE_LO__RB_BASE_LO_MASK                                                                 0xFFFFFFC0L
+//UVD_AUDIO_RB_BASE_HI
+#define UVD_AUDIO_RB_BASE_HI__RB_BASE_HI__SHIFT                                                               0x0
+#define UVD_AUDIO_RB_BASE_HI__RB_BASE_HI_MASK                                                                 0xFFFFFFFFL
+//UVD_AUDIO_RB_SIZE
+#define UVD_AUDIO_RB_SIZE__RB_SIZE__SHIFT                                                                     0x4
+#define UVD_AUDIO_RB_SIZE__RB_SIZE_MASK                                                                       0x007FFFF0L
+//UVD_VCPU_INT_STATUS2
+#define UVD_VCPU_INT_STATUS2__SW_RB6_INT__SHIFT                                                               0x0
+#define UVD_VCPU_INT_STATUS2__RASCNTL_VCPU_VCODEC_INT__SHIFT                                                  0x15
+#define UVD_VCPU_INT_STATUS2__SW_RB6_INT_MASK                                                                 0x00000001L
+#define UVD_VCPU_INT_STATUS2__RASCNTL_VCPU_VCODEC_INT_MASK                                                    0x00200000L
+//UVD_VCPU_INT_ACK2
+#define UVD_VCPU_INT_ACK2__SW_RB6_INT_ACK__SHIFT                                                              0x0
+#define UVD_VCPU_INT_ACK2__RASCNTL_VCPU_VCODEC_ACK__SHIFT                                                     0x16
+#define UVD_VCPU_INT_ACK2__SW_RB6_INT_ACK_MASK                                                                0x00000001L
+#define UVD_VCPU_INT_ACK2__RASCNTL_VCPU_VCODEC_ACK_MASK                                                       0x00400000L
+//UVD_VCPU_INT_EN2
+#define UVD_VCPU_INT_EN2__SW_RB6_INT_EN__SHIFT                                                                0x0
+#define UVD_VCPU_INT_EN2__RASCNTL_VCPU_VCODEC_EN__SHIFT                                                       0x1
+#define UVD_VCPU_INT_EN2__SW_RB6_INT_EN_MASK                                                                  0x00000001L
+#define UVD_VCPU_INT_EN2__RASCNTL_VCPU_VCODEC_EN_MASK                                                         0x00000002L
+//UVD_SUVD_CGC_STATUS2
+#define UVD_SUVD_CGC_STATUS2__SMPA_VCLK__SHIFT                                                                0x0
+#define UVD_SUVD_CGC_STATUS2__SMPA_DCLK__SHIFT                                                                0x1
+#define UVD_SUVD_CGC_STATUS2__MPBE1_DCLK__SHIFT                                                               0x3
+#define UVD_SUVD_CGC_STATUS2__SIT_AV1_DCLK__SHIFT                                                             0x4
+#define UVD_SUVD_CGC_STATUS2__SDB_AV1_DCLK__SHIFT                                                             0x5
+#define UVD_SUVD_CGC_STATUS2__MPC1_DCLK__SHIFT                                                                0x6
+#define UVD_SUVD_CGC_STATUS2__MPC1_SCLK__SHIFT                                                                0x7
+#define UVD_SUVD_CGC_STATUS2__MPC1_VCLK__SHIFT                                                                0x8
+#define UVD_SUVD_CGC_STATUS2__SRE_AV1_ENC_DCLK__SHIFT                                                         0x9
+#define UVD_SUVD_CGC_STATUS2__CDEFE_DCLK__SHIFT                                                               0xa
+#define UVD_SUVD_CGC_STATUS2__SIT0_DCLK__SHIFT                                                                0xb
+#define UVD_SUVD_CGC_STATUS2__SIT1_DCLK__SHIFT                                                                0xc
+#define UVD_SUVD_CGC_STATUS2__SIT2_DCLK__SHIFT                                                                0xd
+#define UVD_SUVD_CGC_STATUS2__FBC_PCLK__SHIFT                                                                 0x1c
+#define UVD_SUVD_CGC_STATUS2__FBC_CCLK__SHIFT                                                                 0x1d
+#define UVD_SUVD_CGC_STATUS2__SMPA_VCLK_MASK                                                                  0x00000001L
+#define UVD_SUVD_CGC_STATUS2__SMPA_DCLK_MASK                                                                  0x00000002L
+#define UVD_SUVD_CGC_STATUS2__MPBE1_DCLK_MASK                                                                 0x00000008L
+#define UVD_SUVD_CGC_STATUS2__SIT_AV1_DCLK_MASK                                                               0x00000010L
+#define UVD_SUVD_CGC_STATUS2__SDB_AV1_DCLK_MASK                                                               0x00000020L
+#define UVD_SUVD_CGC_STATUS2__MPC1_DCLK_MASK                                                                  0x00000040L
+#define UVD_SUVD_CGC_STATUS2__MPC1_SCLK_MASK                                                                  0x00000080L
+#define UVD_SUVD_CGC_STATUS2__MPC1_VCLK_MASK                                                                  0x00000100L
+#define UVD_SUVD_CGC_STATUS2__SRE_AV1_ENC_DCLK_MASK                                                           0x00000200L
+#define UVD_SUVD_CGC_STATUS2__CDEFE_DCLK_MASK                                                                 0x00000400L
+#define UVD_SUVD_CGC_STATUS2__SIT0_DCLK_MASK                                                                  0x00000800L
+#define UVD_SUVD_CGC_STATUS2__SIT1_DCLK_MASK                                                                  0x00001000L
+#define UVD_SUVD_CGC_STATUS2__SIT2_DCLK_MASK                                                                  0x00002000L
+#define UVD_SUVD_CGC_STATUS2__FBC_PCLK_MASK                                                                   0x10000000L
+#define UVD_SUVD_CGC_STATUS2__FBC_CCLK_MASK                                                                   0x20000000L
+//UVD_SUVD_INT_STATUS2
+#define UVD_SUVD_INT_STATUS2__SMPA_FUNC_INT__SHIFT                                                            0x0
+#define UVD_SUVD_INT_STATUS2__SMPA_ERR_INT__SHIFT                                                             0x5
+#define UVD_SUVD_INT_STATUS2__SDB_AV1_FUNC_INT__SHIFT                                                         0x6
+#define UVD_SUVD_INT_STATUS2__SDB_AV1_ERR_INT__SHIFT                                                          0xb
+#define UVD_SUVD_INT_STATUS2__SMPA_FUNC_INT_MASK                                                              0x0000001FL
+#define UVD_SUVD_INT_STATUS2__SMPA_ERR_INT_MASK                                                               0x00000020L
+#define UVD_SUVD_INT_STATUS2__SDB_AV1_FUNC_INT_MASK                                                           0x000007C0L
+#define UVD_SUVD_INT_STATUS2__SDB_AV1_ERR_INT_MASK                                                            0x00000800L
+//UVD_SUVD_INT_EN2
+#define UVD_SUVD_INT_EN2__SMPA_FUNC_INT_EN__SHIFT                                                             0x0
+#define UVD_SUVD_INT_EN2__SMPA_ERR_INT_EN__SHIFT                                                              0x5
+#define UVD_SUVD_INT_EN2__SDB_AV1_FUNC_INT_EN__SHIFT                                                          0x6
+#define UVD_SUVD_INT_EN2__SDB_AV1_ERR_INT_EN__SHIFT                                                           0xb
+#define UVD_SUVD_INT_EN2__SMPA_FUNC_INT_EN_MASK                                                               0x0000001FL
+#define UVD_SUVD_INT_EN2__SMPA_ERR_INT_EN_MASK                                                                0x00000020L
+#define UVD_SUVD_INT_EN2__SDB_AV1_FUNC_INT_EN_MASK                                                            0x000007C0L
+#define UVD_SUVD_INT_EN2__SDB_AV1_ERR_INT_EN_MASK                                                             0x00000800L
+//UVD_SUVD_INT_ACK2
+#define UVD_SUVD_INT_ACK2__SMPA_FUNC_INT_ACK__SHIFT                                                           0x0
+#define UVD_SUVD_INT_ACK2__SMPA_ERR_INT_ACK__SHIFT                                                            0x5
+#define UVD_SUVD_INT_ACK2__SDB_AV1_FUNC_INT_ACK__SHIFT                                                        0x6
+#define UVD_SUVD_INT_ACK2__SDB_AV1_ERR_INT_ACK__SHIFT                                                         0xb
+#define UVD_SUVD_INT_ACK2__SMPA_FUNC_INT_ACK_MASK                                                             0x0000001FL
+#define UVD_SUVD_INT_ACK2__SMPA_ERR_INT_ACK_MASK                                                              0x00000020L
+#define UVD_SUVD_INT_ACK2__SDB_AV1_FUNC_INT_ACK_MASK                                                          0x000007C0L
+#define UVD_SUVD_INT_ACK2__SDB_AV1_ERR_INT_ACK_MASK                                                           0x00000800L
+//UVD_STATUS
+#define UVD_STATUS__RBC_BUSY__SHIFT                                                                           0x0
+#define UVD_STATUS__VCPU_REPORT__SHIFT                                                                        0x1
+#define UVD_STATUS__FILL_0__SHIFT                                                                             0x8
+#define UVD_STATUS__RBC_ACCESS_GPCOM__SHIFT                                                                   0x10
+#define UVD_STATUS__DRM_BUSY__SHIFT                                                                           0x11
+#define UVD_STATUS__FILL_1__SHIFT                                                                             0x12
+#define UVD_STATUS__SYS_GPCOM_REQ__SHIFT                                                                      0x1f
+#define UVD_STATUS__RBC_BUSY_MASK                                                                             0x00000001L
+#define UVD_STATUS__VCPU_REPORT_MASK                                                                          0x000000FEL
+#define UVD_STATUS__FILL_0_MASK                                                                               0x0000FF00L
+#define UVD_STATUS__RBC_ACCESS_GPCOM_MASK                                                                     0x00010000L
+#define UVD_STATUS__DRM_BUSY_MASK                                                                             0x00020000L
+#define UVD_STATUS__FILL_1_MASK                                                                               0x7FFC0000L
+#define UVD_STATUS__SYS_GPCOM_REQ_MASK                                                                        0x80000000L
+//UVD_ENC_PIPE_BUSY
+#define UVD_ENC_PIPE_BUSY__IME_BUSY__SHIFT                                                                    0x0
+#define UVD_ENC_PIPE_BUSY__SMP_BUSY__SHIFT                                                                    0x1
+#define UVD_ENC_PIPE_BUSY__SIT_BUSY__SHIFT                                                                    0x2
+#define UVD_ENC_PIPE_BUSY__SDB_BUSY__SHIFT                                                                    0x3
+#define UVD_ENC_PIPE_BUSY__ENT_BUSY__SHIFT                                                                    0x4
+#define UVD_ENC_PIPE_BUSY__ENT_HEADER_BUSY__SHIFT                                                             0x5
+#define UVD_ENC_PIPE_BUSY__LCM_BUSY__SHIFT                                                                    0x6
+#define UVD_ENC_PIPE_BUSY__MDM_RD_CUR_BUSY__SHIFT                                                             0x7
+#define UVD_ENC_PIPE_BUSY__MDM_RD_REF_BUSY__SHIFT                                                             0x8
+#define UVD_ENC_PIPE_BUSY__MDM_RD_GEN_BUSY__SHIFT                                                             0x9
+#define UVD_ENC_PIPE_BUSY__MDM_WR_RECON_BUSY__SHIFT                                                           0xa
+#define UVD_ENC_PIPE_BUSY__MDM_WR_GEN_BUSY__SHIFT                                                             0xb
+#define UVD_ENC_PIPE_BUSY__EFC_BUSY__SHIFT                                                                    0xc
+#define UVD_ENC_PIPE_BUSY__MDM_PPU_BUSY__SHIFT                                                                0xd
+#define UVD_ENC_PIPE_BUSY__MIF_AUTODMA_BUSY__SHIFT                                                            0xe
+#define UVD_ENC_PIPE_BUSY__CDEFE_BUSY__SHIFT                                                                  0xf
+#define UVD_ENC_PIPE_BUSY__MIF_RD_CUR_BUSY__SHIFT                                                             0x10
+#define UVD_ENC_PIPE_BUSY__MIF_RD_REF0_BUSY__SHIFT                                                            0x11
+#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN0_BUSY__SHIFT                                                            0x12
+#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN0_BUSY__SHIFT                                                            0x13
+#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN1_BUSY__SHIFT                                                            0x14
+#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN1_BUSY__SHIFT                                                            0x15
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP0_BUSY__SHIFT                                                            0x16
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP1_BUSY__SHIFT                                                            0x17
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD0_BUSY__SHIFT                                                            0x18
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD1_BUSY__SHIFT                                                            0x19
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD2_BUSY__SHIFT                                                            0x1a
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD3_BUSY__SHIFT                                                            0x1b
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD4_BUSY__SHIFT                                                            0x1c
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP2_BUSY__SHIFT                                                            0x1d
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP3_BUSY__SHIFT                                                            0x1e
+#define UVD_ENC_PIPE_BUSY__SAOE_BUSY__SHIFT                                                                   0x1f
+#define UVD_ENC_PIPE_BUSY__IME_BUSY_MASK                                                                      0x00000001L
+#define UVD_ENC_PIPE_BUSY__SMP_BUSY_MASK                                                                      0x00000002L
+#define UVD_ENC_PIPE_BUSY__SIT_BUSY_MASK                                                                      0x00000004L
+#define UVD_ENC_PIPE_BUSY__SDB_BUSY_MASK                                                                      0x00000008L
+#define UVD_ENC_PIPE_BUSY__ENT_BUSY_MASK                                                                      0x00000010L
+#define UVD_ENC_PIPE_BUSY__ENT_HEADER_BUSY_MASK                                                               0x00000020L
+#define UVD_ENC_PIPE_BUSY__LCM_BUSY_MASK                                                                      0x00000040L
+#define UVD_ENC_PIPE_BUSY__MDM_RD_CUR_BUSY_MASK                                                               0x00000080L
+#define UVD_ENC_PIPE_BUSY__MDM_RD_REF_BUSY_MASK                                                               0x00000100L
+#define UVD_ENC_PIPE_BUSY__MDM_RD_GEN_BUSY_MASK                                                               0x00000200L
+#define UVD_ENC_PIPE_BUSY__MDM_WR_RECON_BUSY_MASK                                                             0x00000400L
+#define UVD_ENC_PIPE_BUSY__MDM_WR_GEN_BUSY_MASK                                                               0x00000800L
+#define UVD_ENC_PIPE_BUSY__EFC_BUSY_MASK                                                                      0x00001000L
+#define UVD_ENC_PIPE_BUSY__MDM_PPU_BUSY_MASK                                                                  0x00002000L
+#define UVD_ENC_PIPE_BUSY__MIF_AUTODMA_BUSY_MASK                                                              0x00004000L
+#define UVD_ENC_PIPE_BUSY__CDEFE_BUSY_MASK                                                                    0x00008000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_CUR_BUSY_MASK                                                               0x00010000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_REF0_BUSY_MASK                                                              0x00020000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN0_BUSY_MASK                                                              0x00040000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN0_BUSY_MASK                                                              0x00080000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN1_BUSY_MASK                                                              0x00100000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN1_BUSY_MASK                                                              0x00200000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP0_BUSY_MASK                                                              0x00400000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP1_BUSY_MASK                                                              0x00800000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD0_BUSY_MASK                                                              0x01000000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD1_BUSY_MASK                                                              0x02000000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD2_BUSY_MASK                                                              0x04000000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD3_BUSY_MASK                                                              0x08000000L
+#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD4_BUSY_MASK                                                              0x10000000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP2_BUSY_MASK                                                              0x20000000L
+#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP3_BUSY_MASK                                                              0x40000000L
+#define UVD_ENC_PIPE_BUSY__SAOE_BUSY_MASK                                                                     0x80000000L
+//UVD_FW_POWER_STATUS
+#define UVD_FW_POWER_STATUS__UVDF_PWR_OFF__SHIFT                                                              0x0
+#define UVD_FW_POWER_STATUS__UVDTC_PWR_OFF__SHIFT                                                             0x1
+#define UVD_FW_POWER_STATUS__UVDB_PWR_OFF__SHIFT                                                              0x2
+#define UVD_FW_POWER_STATUS__UVDTA_PWR_OFF__SHIFT                                                             0x3
+#define UVD_FW_POWER_STATUS__UVDTD_PWR_OFF__SHIFT                                                             0x4
+#define UVD_FW_POWER_STATUS__UVDTE_PWR_OFF__SHIFT                                                             0x5
+#define UVD_FW_POWER_STATUS__UVDE_PWR_OFF__SHIFT                                                              0x6
+#define UVD_FW_POWER_STATUS__UVDAB_PWR_OFF__SHIFT                                                             0x7
+#define UVD_FW_POWER_STATUS__UVDTB_PWR_OFF__SHIFT                                                             0x8
+#define UVD_FW_POWER_STATUS__UVDNA_PWR_OFF__SHIFT                                                             0x9
+#define UVD_FW_POWER_STATUS__UVDNB_PWR_OFF__SHIFT                                                             0xa
+#define UVD_FW_POWER_STATUS__UVDF_PWR_OFF_MASK                                                                0x00000001L
+#define UVD_FW_POWER_STATUS__UVDTC_PWR_OFF_MASK                                                               0x00000002L
+#define UVD_FW_POWER_STATUS__UVDB_PWR_OFF_MASK                                                                0x00000004L
+#define UVD_FW_POWER_STATUS__UVDTA_PWR_OFF_MASK                                                               0x00000008L
+#define UVD_FW_POWER_STATUS__UVDTD_PWR_OFF_MASK                                                               0x00000010L
+#define UVD_FW_POWER_STATUS__UVDTE_PWR_OFF_MASK                                                               0x00000020L
+#define UVD_FW_POWER_STATUS__UVDE_PWR_OFF_MASK                                                                0x00000040L
+#define UVD_FW_POWER_STATUS__UVDAB_PWR_OFF_MASK                                                               0x00000080L
+#define UVD_FW_POWER_STATUS__UVDTB_PWR_OFF_MASK                                                               0x00000100L
+#define UVD_FW_POWER_STATUS__UVDNA_PWR_OFF_MASK                                                               0x00000200L
+#define UVD_FW_POWER_STATUS__UVDNB_PWR_OFF_MASK                                                               0x00000400L
+//UVD_CNTL
+#define UVD_CNTL__MIF_WR_LOW_THRESHOLD_BP__SHIFT                                                              0x11
+#define UVD_CNTL__SUVD_EN__SHIFT                                                                              0x13
+#define UVD_CNTL__CABAC_MB_ACC__SHIFT                                                                         0x1c
+#define UVD_CNTL__LRBBM_SAFE_SYNC_DIS__SHIFT                                                                  0x1f
+#define UVD_CNTL__MIF_WR_LOW_THRESHOLD_BP_MASK                                                                0x00020000L
+#define UVD_CNTL__SUVD_EN_MASK                                                                                0x00080000L
+#define UVD_CNTL__CABAC_MB_ACC_MASK                                                                           0x10000000L
+#define UVD_CNTL__LRBBM_SAFE_SYNC_DIS_MASK                                                                    0x80000000L
+//UVD_SOFT_RESET
+#define UVD_SOFT_RESET__RBC_SOFT_RESET__SHIFT                                                                 0x0
+#define UVD_SOFT_RESET__LBSI_SOFT_RESET__SHIFT                                                                0x1
+#define UVD_SOFT_RESET__LMI_SOFT_RESET__SHIFT                                                                 0x2
+#define UVD_SOFT_RESET__VCPU_SOFT_RESET__SHIFT                                                                0x3
+#define UVD_SOFT_RESET__UDEC_SOFT_RESET__SHIFT                                                                0x4
+#define UVD_SOFT_RESET__CXW_SOFT_RESET__SHIFT                                                                 0x6
+#define UVD_SOFT_RESET__TAP_SOFT_RESET__SHIFT                                                                 0x7
+#define UVD_SOFT_RESET__MPC_SOFT_RESET__SHIFT                                                                 0x8
+#define UVD_SOFT_RESET__EFC_SOFT_RESET__SHIFT                                                                 0x9
+#define UVD_SOFT_RESET__IH_SOFT_RESET__SHIFT                                                                  0xa
+#define UVD_SOFT_RESET__MPRD_SOFT_RESET__SHIFT                                                                0xb
+#define UVD_SOFT_RESET__IDCT_SOFT_RESET__SHIFT                                                                0xc
+#define UVD_SOFT_RESET__LMI_UMC_SOFT_RESET__SHIFT                                                             0xd
+#define UVD_SOFT_RESET__SPH_SOFT_RESET__SHIFT                                                                 0xe
+#define UVD_SOFT_RESET__MIF_SOFT_RESET__SHIFT                                                                 0xf
+#define UVD_SOFT_RESET__LCM_SOFT_RESET__SHIFT                                                                 0x10
+#define UVD_SOFT_RESET__SUVD_SOFT_RESET__SHIFT                                                                0x11
+#define UVD_SOFT_RESET__LBSI_VCLK_RESET_STATUS__SHIFT                                                         0x12
+#define UVD_SOFT_RESET__VCPU_VCLK_RESET_STATUS__SHIFT                                                         0x13
+#define UVD_SOFT_RESET__UDEC_VCLK_RESET_STATUS__SHIFT                                                         0x14
+#define UVD_SOFT_RESET__UDEC_DCLK_RESET_STATUS__SHIFT                                                         0x15
+#define UVD_SOFT_RESET__MPC_DCLK_RESET_STATUS__SHIFT                                                          0x16
+#define UVD_SOFT_RESET__MPRD_VCLK_RESET_STATUS__SHIFT                                                         0x17
+#define UVD_SOFT_RESET__MPRD_DCLK_RESET_STATUS__SHIFT                                                         0x18
+#define UVD_SOFT_RESET__IDCT_VCLK_RESET_STATUS__SHIFT                                                         0x19
+#define UVD_SOFT_RESET__MIF_DCLK_RESET_STATUS__SHIFT                                                          0x1a
+#define UVD_SOFT_RESET__LCM_DCLK_RESET_STATUS__SHIFT                                                          0x1b
+#define UVD_SOFT_RESET__SUVD_VCLK_RESET_STATUS__SHIFT                                                         0x1c
+#define UVD_SOFT_RESET__SUVD_DCLK_RESET_STATUS__SHIFT                                                         0x1d
+#define UVD_SOFT_RESET__RE_DCLK_RESET_STATUS__SHIFT                                                           0x1e
+#define UVD_SOFT_RESET__SRE_DCLK_RESET_STATUS__SHIFT                                                          0x1f
+#define UVD_SOFT_RESET__RBC_SOFT_RESET_MASK                                                                   0x00000001L
+#define UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK                                                                  0x00000002L
+#define UVD_SOFT_RESET__LMI_SOFT_RESET_MASK                                                                   0x00000004L
+#define UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK                                                                  0x00000008L
+#define UVD_SOFT_RESET__UDEC_SOFT_RESET_MASK                                                                  0x00000010L
+#define UVD_SOFT_RESET__CXW_SOFT_RESET_MASK                                                                   0x00000040L
+#define UVD_SOFT_RESET__TAP_SOFT_RESET_MASK                                                                   0x00000080L
+#define UVD_SOFT_RESET__MPC_SOFT_RESET_MASK                                                                   0x00000100L
+#define UVD_SOFT_RESET__EFC_SOFT_RESET_MASK                                                                   0x00000200L
+#define UVD_SOFT_RESET__IH_SOFT_RESET_MASK                                                                    0x00000400L
+#define UVD_SOFT_RESET__MPRD_SOFT_RESET_MASK                                                                  0x00000800L
+#define UVD_SOFT_RESET__IDCT_SOFT_RESET_MASK                                                                  0x00001000L
+#define UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK                                                               0x00002000L
+#define UVD_SOFT_RESET__SPH_SOFT_RESET_MASK                                                                   0x00004000L
+#define UVD_SOFT_RESET__MIF_SOFT_RESET_MASK                                                                   0x00008000L
+#define UVD_SOFT_RESET__LCM_SOFT_RESET_MASK                                                                   0x00010000L
+#define UVD_SOFT_RESET__SUVD_SOFT_RESET_MASK                                                                  0x00020000L
+#define UVD_SOFT_RESET__LBSI_VCLK_RESET_STATUS_MASK                                                           0x00040000L
+#define UVD_SOFT_RESET__VCPU_VCLK_RESET_STATUS_MASK                                                           0x00080000L
+#define UVD_SOFT_RESET__UDEC_VCLK_RESET_STATUS_MASK                                                           0x00100000L
+#define UVD_SOFT_RESET__UDEC_DCLK_RESET_STATUS_MASK                                                           0x00200000L
+#define UVD_SOFT_RESET__MPC_DCLK_RESET_STATUS_MASK                                                            0x00400000L
+#define UVD_SOFT_RESET__MPRD_VCLK_RESET_STATUS_MASK                                                           0x00800000L
+#define UVD_SOFT_RESET__MPRD_DCLK_RESET_STATUS_MASK                                                           0x01000000L
+#define UVD_SOFT_RESET__IDCT_VCLK_RESET_STATUS_MASK                                                           0x02000000L
+#define UVD_SOFT_RESET__MIF_DCLK_RESET_STATUS_MASK                                                            0x04000000L
+#define UVD_SOFT_RESET__LCM_DCLK_RESET_STATUS_MASK                                                            0x08000000L
+#define UVD_SOFT_RESET__SUVD_VCLK_RESET_STATUS_MASK                                                           0x10000000L
+#define UVD_SOFT_RESET__SUVD_DCLK_RESET_STATUS_MASK                                                           0x20000000L
+#define UVD_SOFT_RESET__RE_DCLK_RESET_STATUS_MASK                                                             0x40000000L
+#define UVD_SOFT_RESET__SRE_DCLK_RESET_STATUS_MASK                                                            0x80000000L
+//UVD_SOFT_RESET2
+#define UVD_SOFT_RESET2__ATOMIC_SOFT_RESET__SHIFT                                                             0x0
+#define UVD_SOFT_RESET2__PPU_SOFT_RESET__SHIFT                                                                0x1
+#define UVD_SOFT_RESET2__MMSCH_VCLK_RESET_STATUS__SHIFT                                                       0x10
+#define UVD_SOFT_RESET2__MMSCH_SCLK_RESET_STATUS__SHIFT                                                       0x11
+#define UVD_SOFT_RESET2__ATOMIC_SOFT_RESET_MASK                                                               0x00000001L
+#define UVD_SOFT_RESET2__PPU_SOFT_RESET_MASK                                                                  0x00000002L
+#define UVD_SOFT_RESET2__MMSCH_VCLK_RESET_STATUS_MASK                                                         0x00010000L
+#define UVD_SOFT_RESET2__MMSCH_SCLK_RESET_STATUS_MASK                                                         0x00020000L
+//UVD_MMSCH_SOFT_RESET
+#define UVD_MMSCH_SOFT_RESET__MMSCH_RESET__SHIFT                                                              0x0
+#define UVD_MMSCH_SOFT_RESET__TAP_SOFT_RESET__SHIFT                                                           0x1
+#define UVD_MMSCH_SOFT_RESET__MMSCH_LOCK__SHIFT                                                               0x1f
+#define UVD_MMSCH_SOFT_RESET__MMSCH_RESET_MASK                                                                0x00000001L
+#define UVD_MMSCH_SOFT_RESET__TAP_SOFT_RESET_MASK                                                             0x00000002L
+#define UVD_MMSCH_SOFT_RESET__MMSCH_LOCK_MASK                                                                 0x80000000L
+//UVD_WIG_CTRL
+#define UVD_WIG_CTRL__AVM_SOFT_RESET__SHIFT                                                                   0x0
+#define UVD_WIG_CTRL__ACAP_SOFT_RESET__SHIFT                                                                  0x1
+#define UVD_WIG_CTRL__WIG_SOFT_RESET__SHIFT                                                                   0x2
+#define UVD_WIG_CTRL__WIG_REGCLK_FORCE_ON__SHIFT                                                              0x3
+#define UVD_WIG_CTRL__AVM_REGCLK_FORCE_ON__SHIFT                                                              0x4
+#define UVD_WIG_CTRL__AVM_SOFT_RESET_MASK                                                                     0x00000001L
+#define UVD_WIG_CTRL__ACAP_SOFT_RESET_MASK                                                                    0x00000002L
+#define UVD_WIG_CTRL__WIG_SOFT_RESET_MASK                                                                     0x00000004L
+#define UVD_WIG_CTRL__WIG_REGCLK_FORCE_ON_MASK                                                                0x00000008L
+#define UVD_WIG_CTRL__AVM_REGCLK_FORCE_ON_MASK                                                                0x00000010L
+//UVD_CGC_STATUS
+#define UVD_CGC_STATUS__SYS_SCLK__SHIFT                                                                       0x0
+#define UVD_CGC_STATUS__SYS_DCLK__SHIFT                                                                       0x1
+#define UVD_CGC_STATUS__SYS_VCLK__SHIFT                                                                       0x2
+#define UVD_CGC_STATUS__UDEC_SCLK__SHIFT                                                                      0x3
+#define UVD_CGC_STATUS__UDEC_DCLK__SHIFT                                                                      0x4
+#define UVD_CGC_STATUS__UDEC_VCLK__SHIFT                                                                      0x5
+#define UVD_CGC_STATUS__MPEG2_SCLK__SHIFT                                                                     0x6
+#define UVD_CGC_STATUS__MPEG2_DCLK__SHIFT                                                                     0x7
+#define UVD_CGC_STATUS__MPEG2_VCLK__SHIFT                                                                     0x8
+#define UVD_CGC_STATUS__REGS_SCLK__SHIFT                                                                      0x9
+#define UVD_CGC_STATUS__REGS_VCLK__SHIFT                                                                      0xa
+#define UVD_CGC_STATUS__RBC_SCLK__SHIFT                                                                       0xb
+#define UVD_CGC_STATUS__LMI_MC_SCLK__SHIFT                                                                    0xc
+#define UVD_CGC_STATUS__LMI_UMC_SCLK__SHIFT                                                                   0xd
+#define UVD_CGC_STATUS__IDCT_SCLK__SHIFT                                                                      0xe
+#define UVD_CGC_STATUS__IDCT_VCLK__SHIFT                                                                      0xf
+#define UVD_CGC_STATUS__MPRD_SCLK__SHIFT                                                                      0x10
+#define UVD_CGC_STATUS__MPRD_DCLK__SHIFT                                                                      0x11
+#define UVD_CGC_STATUS__MPRD_VCLK__SHIFT                                                                      0x12
+#define UVD_CGC_STATUS__MPC_SCLK__SHIFT                                                                       0x13
+#define UVD_CGC_STATUS__MPC_DCLK__SHIFT                                                                       0x14
+#define UVD_CGC_STATUS__LBSI_SCLK__SHIFT                                                                      0x15
+#define UVD_CGC_STATUS__LBSI_VCLK__SHIFT                                                                      0x16
+#define UVD_CGC_STATUS__LRBBM_SCLK__SHIFT                                                                     0x17
+#define UVD_CGC_STATUS__WCB_SCLK__SHIFT                                                                       0x18
+#define UVD_CGC_STATUS__VCPU_SCLK__SHIFT                                                                      0x19
+#define UVD_CGC_STATUS__VCPU_VCLK__SHIFT                                                                      0x1a
+#define UVD_CGC_STATUS__MMSCH_SCLK__SHIFT                                                                     0x1b
+#define UVD_CGC_STATUS__MMSCH_VCLK__SHIFT                                                                     0x1c
+#define UVD_CGC_STATUS__ALL_ENC_ACTIVE__SHIFT                                                                 0x1d
+#define UVD_CGC_STATUS__LRBBM_DCLK__SHIFT                                                                     0x1e
+#define UVD_CGC_STATUS__ALL_DEC_ACTIVE__SHIFT                                                                 0x1f
+#define UVD_CGC_STATUS__SYS_SCLK_MASK                                                                         0x00000001L
+#define UVD_CGC_STATUS__SYS_DCLK_MASK                                                                         0x00000002L
+#define UVD_CGC_STATUS__SYS_VCLK_MASK                                                                         0x00000004L
+#define UVD_CGC_STATUS__UDEC_SCLK_MASK                                                                        0x00000008L
+#define UVD_CGC_STATUS__UDEC_DCLK_MASK                                                                        0x00000010L
+#define UVD_CGC_STATUS__UDEC_VCLK_MASK                                                                        0x00000020L
+#define UVD_CGC_STATUS__MPEG2_SCLK_MASK                                                                       0x00000040L
+#define UVD_CGC_STATUS__MPEG2_DCLK_MASK                                                                       0x00000080L
+#define UVD_CGC_STATUS__MPEG2_VCLK_MASK                                                                       0x00000100L
+#define UVD_CGC_STATUS__REGS_SCLK_MASK                                                                        0x00000200L
+#define UVD_CGC_STATUS__REGS_VCLK_MASK                                                                        0x00000400L
+#define UVD_CGC_STATUS__RBC_SCLK_MASK                                                                         0x00000800L
+#define UVD_CGC_STATUS__LMI_MC_SCLK_MASK                                                                      0x00001000L
+#define UVD_CGC_STATUS__LMI_UMC_SCLK_MASK                                                                     0x00002000L
+#define UVD_CGC_STATUS__IDCT_SCLK_MASK                                                                        0x00004000L
+#define UVD_CGC_STATUS__IDCT_VCLK_MASK                                                                        0x00008000L
+#define UVD_CGC_STATUS__MPRD_SCLK_MASK                                                                        0x00010000L
+#define UVD_CGC_STATUS__MPRD_DCLK_MASK                                                                        0x00020000L
+#define UVD_CGC_STATUS__MPRD_VCLK_MASK                                                                        0x00040000L
+#define UVD_CGC_STATUS__MPC_SCLK_MASK                                                                         0x00080000L
+#define UVD_CGC_STATUS__MPC_DCLK_MASK                                                                         0x00100000L
+#define UVD_CGC_STATUS__LBSI_SCLK_MASK                                                                        0x00200000L
+#define UVD_CGC_STATUS__LBSI_VCLK_MASK                                                                        0x00400000L
+#define UVD_CGC_STATUS__LRBBM_SCLK_MASK                                                                       0x00800000L
+#define UVD_CGC_STATUS__WCB_SCLK_MASK                                                                         0x01000000L
+#define UVD_CGC_STATUS__VCPU_SCLK_MASK                                                                        0x02000000L
+#define UVD_CGC_STATUS__VCPU_VCLK_MASK                                                                        0x04000000L
+#define UVD_CGC_STATUS__MMSCH_SCLK_MASK                                                                       0x08000000L
+#define UVD_CGC_STATUS__MMSCH_VCLK_MASK                                                                       0x10000000L
+#define UVD_CGC_STATUS__ALL_ENC_ACTIVE_MASK                                                                   0x20000000L
+#define UVD_CGC_STATUS__LRBBM_DCLK_MASK                                                                       0x40000000L
+#define UVD_CGC_STATUS__ALL_DEC_ACTIVE_MASK                                                                   0x80000000L
+//UVD_CGC_UDEC_STATUS
+#define UVD_CGC_UDEC_STATUS__RE_SCLK__SHIFT                                                                   0x0
+#define UVD_CGC_UDEC_STATUS__RE_DCLK__SHIFT                                                                   0x1
+#define UVD_CGC_UDEC_STATUS__RE_VCLK__SHIFT                                                                   0x2
+#define UVD_CGC_UDEC_STATUS__CM_SCLK__SHIFT                                                                   0x3
+#define UVD_CGC_UDEC_STATUS__CM_DCLK__SHIFT                                                                   0x4
+#define UVD_CGC_UDEC_STATUS__CM_VCLK__SHIFT                                                                   0x5
+#define UVD_CGC_UDEC_STATUS__IT_SCLK__SHIFT                                                                   0x6
+#define UVD_CGC_UDEC_STATUS__IT_DCLK__SHIFT                                                                   0x7
+#define UVD_CGC_UDEC_STATUS__IT_VCLK__SHIFT                                                                   0x8
+#define UVD_CGC_UDEC_STATUS__DB_SCLK__SHIFT                                                                   0x9
+#define UVD_CGC_UDEC_STATUS__DB_DCLK__SHIFT                                                                   0xa
+#define UVD_CGC_UDEC_STATUS__DB_VCLK__SHIFT                                                                   0xb
+#define UVD_CGC_UDEC_STATUS__MP_SCLK__SHIFT                                                                   0xc
+#define UVD_CGC_UDEC_STATUS__MP_DCLK__SHIFT                                                                   0xd
+#define UVD_CGC_UDEC_STATUS__MP_VCLK__SHIFT                                                                   0xe
+#define UVD_CGC_UDEC_STATUS__RE_SCLK_MASK                                                                     0x00000001L
+#define UVD_CGC_UDEC_STATUS__RE_DCLK_MASK                                                                     0x00000002L
+#define UVD_CGC_UDEC_STATUS__RE_VCLK_MASK                                                                     0x00000004L
+#define UVD_CGC_UDEC_STATUS__CM_SCLK_MASK                                                                     0x00000008L
+#define UVD_CGC_UDEC_STATUS__CM_DCLK_MASK                                                                     0x00000010L
+#define UVD_CGC_UDEC_STATUS__CM_VCLK_MASK                                                                     0x00000020L
+#define UVD_CGC_UDEC_STATUS__IT_SCLK_MASK                                                                     0x00000040L
+#define UVD_CGC_UDEC_STATUS__IT_DCLK_MASK                                                                     0x00000080L
+#define UVD_CGC_UDEC_STATUS__IT_VCLK_MASK                                                                     0x00000100L
+#define UVD_CGC_UDEC_STATUS__DB_SCLK_MASK                                                                     0x00000200L
+#define UVD_CGC_UDEC_STATUS__DB_DCLK_MASK                                                                     0x00000400L
+#define UVD_CGC_UDEC_STATUS__DB_VCLK_MASK                                                                     0x00000800L
+#define UVD_CGC_UDEC_STATUS__MP_SCLK_MASK                                                                     0x00001000L
+#define UVD_CGC_UDEC_STATUS__MP_DCLK_MASK                                                                     0x00002000L
+#define UVD_CGC_UDEC_STATUS__MP_VCLK_MASK                                                                     0x00004000L
+//UVD_SUVD_CGC_STATUS
+#define UVD_SUVD_CGC_STATUS__SRE_VCLK__SHIFT                                                                  0x0
+#define UVD_SUVD_CGC_STATUS__SRE_DCLK__SHIFT                                                                  0x1
+#define UVD_SUVD_CGC_STATUS__SIT_DCLK__SHIFT                                                                  0x2
+#define UVD_SUVD_CGC_STATUS__SMP_DCLK__SHIFT                                                                  0x3
+#define UVD_SUVD_CGC_STATUS__SCM_DCLK__SHIFT                                                                  0x4
+#define UVD_SUVD_CGC_STATUS__SDB_DCLK__SHIFT                                                                  0x5
+#define UVD_SUVD_CGC_STATUS__SRE_H264_VCLK__SHIFT                                                             0x6
+#define UVD_SUVD_CGC_STATUS__SRE_HEVC_VCLK__SHIFT                                                             0x7
+#define UVD_SUVD_CGC_STATUS__SIT_H264_DCLK__SHIFT                                                             0x8
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DCLK__SHIFT                                                             0x9
+#define UVD_SUVD_CGC_STATUS__SCM_H264_DCLK__SHIFT                                                             0xa
+#define UVD_SUVD_CGC_STATUS__SCM_HEVC_DCLK__SHIFT                                                             0xb
+#define UVD_SUVD_CGC_STATUS__SDB_H264_DCLK__SHIFT                                                             0xc
+#define UVD_SUVD_CGC_STATUS__SDB_HEVC_DCLK__SHIFT                                                             0xd
+#define UVD_SUVD_CGC_STATUS__SCLR_DCLK__SHIFT                                                                 0xe
+#define UVD_SUVD_CGC_STATUS__UVD_SC__SHIFT                                                                    0xf
+#define UVD_SUVD_CGC_STATUS__ENT_DCLK__SHIFT                                                                  0x10
+#define UVD_SUVD_CGC_STATUS__IME_DCLK__SHIFT                                                                  0x11
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DEC_DCLK__SHIFT                                                         0x12
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_ENC_DCLK__SHIFT                                                         0x13
+#define UVD_SUVD_CGC_STATUS__SITE_DCLK__SHIFT                                                                 0x14
+#define UVD_SUVD_CGC_STATUS__SITE_HEVC_DCLK__SHIFT                                                            0x15
+#define UVD_SUVD_CGC_STATUS__SITE_HEVC_ENC_DCLK__SHIFT                                                        0x16
+#define UVD_SUVD_CGC_STATUS__SRE_VP9_VCLK__SHIFT                                                              0x17
+#define UVD_SUVD_CGC_STATUS__SCM_VP9_VCLK__SHIFT                                                              0x18
+#define UVD_SUVD_CGC_STATUS__SIT_VP9_DEC_DCLK__SHIFT                                                          0x19
+#define UVD_SUVD_CGC_STATUS__SDB_VP9_DCLK__SHIFT                                                              0x1a
+#define UVD_SUVD_CGC_STATUS__IME_HEVC_DCLK__SHIFT                                                             0x1b
+#define UVD_SUVD_CGC_STATUS__EFC_DCLK__SHIFT                                                                  0x1c
+#define UVD_SUVD_CGC_STATUS__SAOE_DCLK__SHIFT                                                                 0x1d
+#define UVD_SUVD_CGC_STATUS__SRE_AV1_VCLK__SHIFT                                                              0x1e
+#define UVD_SUVD_CGC_STATUS__SCM_AV1_DCLK__SHIFT                                                              0x1f
+#define UVD_SUVD_CGC_STATUS__SRE_VCLK_MASK                                                                    0x00000001L
+#define UVD_SUVD_CGC_STATUS__SRE_DCLK_MASK                                                                    0x00000002L
+#define UVD_SUVD_CGC_STATUS__SIT_DCLK_MASK                                                                    0x00000004L
+#define UVD_SUVD_CGC_STATUS__SMP_DCLK_MASK                                                                    0x00000008L
+#define UVD_SUVD_CGC_STATUS__SCM_DCLK_MASK                                                                    0x00000010L
+#define UVD_SUVD_CGC_STATUS__SDB_DCLK_MASK                                                                    0x00000020L
+#define UVD_SUVD_CGC_STATUS__SRE_H264_VCLK_MASK                                                               0x00000040L
+#define UVD_SUVD_CGC_STATUS__SRE_HEVC_VCLK_MASK                                                               0x00000080L
+#define UVD_SUVD_CGC_STATUS__SIT_H264_DCLK_MASK                                                               0x00000100L
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DCLK_MASK                                                               0x00000200L
+#define UVD_SUVD_CGC_STATUS__SCM_H264_DCLK_MASK                                                               0x00000400L
+#define UVD_SUVD_CGC_STATUS__SCM_HEVC_DCLK_MASK                                                               0x00000800L
+#define UVD_SUVD_CGC_STATUS__SDB_H264_DCLK_MASK                                                               0x00001000L
+#define UVD_SUVD_CGC_STATUS__SDB_HEVC_DCLK_MASK                                                               0x00002000L
+#define UVD_SUVD_CGC_STATUS__SCLR_DCLK_MASK                                                                   0x00004000L
+#define UVD_SUVD_CGC_STATUS__UVD_SC_MASK                                                                      0x00008000L
+#define UVD_SUVD_CGC_STATUS__ENT_DCLK_MASK                                                                    0x00010000L
+#define UVD_SUVD_CGC_STATUS__IME_DCLK_MASK                                                                    0x00020000L
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DEC_DCLK_MASK                                                           0x00040000L
+#define UVD_SUVD_CGC_STATUS__SIT_HEVC_ENC_DCLK_MASK                                                           0x00080000L
+#define UVD_SUVD_CGC_STATUS__SITE_DCLK_MASK                                                                   0x00100000L
+#define UVD_SUVD_CGC_STATUS__SITE_HEVC_DCLK_MASK                                                              0x00200000L
+#define UVD_SUVD_CGC_STATUS__SITE_HEVC_ENC_DCLK_MASK                                                          0x00400000L
+#define UVD_SUVD_CGC_STATUS__SRE_VP9_VCLK_MASK                                                                0x00800000L
+#define UVD_SUVD_CGC_STATUS__SCM_VP9_VCLK_MASK                                                                0x01000000L
+#define UVD_SUVD_CGC_STATUS__SIT_VP9_DEC_DCLK_MASK                                                            0x02000000L
+#define UVD_SUVD_CGC_STATUS__SDB_VP9_DCLK_MASK                                                                0x04000000L
+#define UVD_SUVD_CGC_STATUS__IME_HEVC_DCLK_MASK                                                               0x08000000L
+#define UVD_SUVD_CGC_STATUS__EFC_DCLK_MASK                                                                    0x10000000L
+#define UVD_SUVD_CGC_STATUS__SAOE_DCLK_MASK                                                                   0x20000000L
+#define UVD_SUVD_CGC_STATUS__SRE_AV1_VCLK_MASK                                                                0x40000000L
+#define UVD_SUVD_CGC_STATUS__SCM_AV1_DCLK_MASK                                                                0x80000000L
+//UVD_GPCOM_VCPU_CMD
+#define UVD_GPCOM_VCPU_CMD__CMD_SEND__SHIFT                                                                   0x0
+#define UVD_GPCOM_VCPU_CMD__CMD__SHIFT                                                                        0x1
+#define UVD_GPCOM_VCPU_CMD__CMD_SOURCE__SHIFT                                                                 0x1f
+#define UVD_GPCOM_VCPU_CMD__CMD_SEND_MASK                                                                     0x00000001L
+#define UVD_GPCOM_VCPU_CMD__CMD_MASK                                                                          0x7FFFFFFEL
+#define UVD_GPCOM_VCPU_CMD__CMD_SOURCE_MASK                                                                   0x80000000L
+
+
+// addressBlock: aid_uvd0_ecpudec
+//UVD_VCPU_CACHE_OFFSET0
+#define UVD_VCPU_CACHE_OFFSET0__CACHE_OFFSET0__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET0__CACHE_OFFSET0_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE0
+#define UVD_VCPU_CACHE_SIZE0__CACHE_SIZE0__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE0__CACHE_SIZE0_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET1
+#define UVD_VCPU_CACHE_OFFSET1__CACHE_OFFSET1__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET1__CACHE_OFFSET1_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE1
+#define UVD_VCPU_CACHE_SIZE1__CACHE_SIZE1__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE1__CACHE_SIZE1_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET2
+#define UVD_VCPU_CACHE_OFFSET2__CACHE_OFFSET2__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET2__CACHE_OFFSET2_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE2
+#define UVD_VCPU_CACHE_SIZE2__CACHE_SIZE2__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE2__CACHE_SIZE2_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET3
+#define UVD_VCPU_CACHE_OFFSET3__CACHE_OFFSET3__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET3__CACHE_OFFSET3_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE3
+#define UVD_VCPU_CACHE_SIZE3__CACHE_SIZE3__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE3__CACHE_SIZE3_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET4
+#define UVD_VCPU_CACHE_OFFSET4__CACHE_OFFSET4__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET4__CACHE_OFFSET4_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE4
+#define UVD_VCPU_CACHE_SIZE4__CACHE_SIZE4__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE4__CACHE_SIZE4_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET5
+#define UVD_VCPU_CACHE_OFFSET5__CACHE_OFFSET5__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET5__CACHE_OFFSET5_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE5
+#define UVD_VCPU_CACHE_SIZE5__CACHE_SIZE5__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE5__CACHE_SIZE5_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET6
+#define UVD_VCPU_CACHE_OFFSET6__CACHE_OFFSET6__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET6__CACHE_OFFSET6_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE6
+#define UVD_VCPU_CACHE_SIZE6__CACHE_SIZE6__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE6__CACHE_SIZE6_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET7
+#define UVD_VCPU_CACHE_OFFSET7__CACHE_OFFSET7__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET7__CACHE_OFFSET7_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE7
+#define UVD_VCPU_CACHE_SIZE7__CACHE_SIZE7__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE7__CACHE_SIZE7_MASK                                                                0x001FFFFFL
+//UVD_VCPU_CACHE_OFFSET8
+#define UVD_VCPU_CACHE_OFFSET8__CACHE_OFFSET8__SHIFT                                                          0x0
+#define UVD_VCPU_CACHE_OFFSET8__CACHE_OFFSET8_MASK                                                            0x001FFFFFL
+//UVD_VCPU_CACHE_SIZE8
+#define UVD_VCPU_CACHE_SIZE8__CACHE_SIZE8__SHIFT                                                              0x0
+#define UVD_VCPU_CACHE_SIZE8__CACHE_SIZE8_MASK                                                                0x001FFFFFL
+//UVD_VCPU_NONCACHE_OFFSET0
+#define UVD_VCPU_NONCACHE_OFFSET0__NONCACHE_OFFSET0__SHIFT                                                    0x0
+#define UVD_VCPU_NONCACHE_OFFSET0__NONCACHE_OFFSET0_MASK                                                      0x01FFFFFFL
+//UVD_VCPU_NONCACHE_SIZE0
+#define UVD_VCPU_NONCACHE_SIZE0__NONCACHE_SIZE0__SHIFT                                                        0x0
+#define UVD_VCPU_NONCACHE_SIZE0__NONCACHE_SIZE0_MASK                                                          0x001FFFFFL
+//UVD_VCPU_NONCACHE_OFFSET1
+#define UVD_VCPU_NONCACHE_OFFSET1__NONCACHE_OFFSET1__SHIFT                                                    0x0
+#define UVD_VCPU_NONCACHE_OFFSET1__NONCACHE_OFFSET1_MASK                                                      0x01FFFFFFL
+//UVD_VCPU_NONCACHE_SIZE1
+#define UVD_VCPU_NONCACHE_SIZE1__NONCACHE_SIZE1__SHIFT                                                        0x0
+#define UVD_VCPU_NONCACHE_SIZE1__NONCACHE_SIZE1_MASK                                                          0x001FFFFFL
+//UVD_VCPU_CNTL
+#define UVD_VCPU_CNTL__IRQ_ERR__SHIFT                                                                         0x0
+#define UVD_VCPU_CNTL__AXI_MAX_BRST_SIZE_IS_4__SHIFT                                                          0x4
+#define UVD_VCPU_CNTL__PMB_ED_ENABLE__SHIFT                                                                   0x5
+#define UVD_VCPU_CNTL__PMB_SOFT_RESET__SHIFT                                                                  0x6
+#define UVD_VCPU_CNTL__RBBM_SOFT_RESET__SHIFT                                                                 0x7
+#define UVD_VCPU_CNTL__ABORT_REQ__SHIFT                                                                       0x8
+#define UVD_VCPU_CNTL__CLK_EN__SHIFT                                                                          0x9
+#define UVD_VCPU_CNTL__TRCE_EN__SHIFT                                                                         0xa
+#define UVD_VCPU_CNTL__TRCE_MUX__SHIFT                                                                        0xb
+#define UVD_VCPU_CNTL__DBG_MUX__SHIFT                                                                         0xd
+#define UVD_VCPU_CNTL__JTAG_EN__SHIFT                                                                         0x10
+#define UVD_VCPU_CNTL__TIMEOUT_DIS__SHIFT                                                                     0x12
+#define UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT                                                                 0x14
+#define UVD_VCPU_CNTL__BLK_RST__SHIFT                                                                         0x1c
+#define UVD_VCPU_CNTL__RUNSTALL__SHIFT                                                                        0x1d
+#define UVD_VCPU_CNTL__SRE_CMDIF_DRST__SHIFT                                                                  0x1e
+#define UVD_VCPU_CNTL__SRE_CMDIF_VRST__SHIFT                                                                  0x1f
+#define UVD_VCPU_CNTL__IRQ_ERR_MASK                                                                           0x0000000FL
+#define UVD_VCPU_CNTL__AXI_MAX_BRST_SIZE_IS_4_MASK                                                            0x00000010L
+#define UVD_VCPU_CNTL__PMB_ED_ENABLE_MASK                                                                     0x00000020L
+#define UVD_VCPU_CNTL__PMB_SOFT_RESET_MASK                                                                    0x00000040L
+#define UVD_VCPU_CNTL__RBBM_SOFT_RESET_MASK                                                                   0x00000080L
+#define UVD_VCPU_CNTL__ABORT_REQ_MASK                                                                         0x00000100L
+#define UVD_VCPU_CNTL__CLK_EN_MASK                                                                            0x00000200L
+#define UVD_VCPU_CNTL__TRCE_EN_MASK                                                                           0x00000400L
+#define UVD_VCPU_CNTL__TRCE_MUX_MASK                                                                          0x00001800L
+#define UVD_VCPU_CNTL__DBG_MUX_MASK                                                                           0x0000E000L
+#define UVD_VCPU_CNTL__JTAG_EN_MASK                                                                           0x00010000L
+#define UVD_VCPU_CNTL__TIMEOUT_DIS_MASK                                                                       0x00040000L
+#define UVD_VCPU_CNTL__PRB_TIMEOUT_VAL_MASK                                                                   0x0FF00000L
+#define UVD_VCPU_CNTL__BLK_RST_MASK                                                                           0x10000000L
+#define UVD_VCPU_CNTL__RUNSTALL_MASK                                                                          0x20000000L
+#define UVD_VCPU_CNTL__SRE_CMDIF_DRST_MASK                                                                    0x40000000L
+#define UVD_VCPU_CNTL__SRE_CMDIF_VRST_MASK                                                                    0x80000000L
+//UVD_VCPU_PRID
+#define UVD_VCPU_PRID__PRID__SHIFT                                                                            0x0
+#define UVD_VCPU_PRID__PRID_MASK                                                                              0x0000FFFFL
+//UVD_VCPU_TRCE
+#define UVD_VCPU_TRCE__PC__SHIFT                                                                              0x0
+#define UVD_VCPU_TRCE__PC_MASK                                                                                0x0FFFFFFFL
+//UVD_VCPU_TRCE_RD
+#define UVD_VCPU_TRCE_RD__DATA__SHIFT                                                                         0x0
+#define UVD_VCPU_TRCE_RD__DATA_MASK                                                                           0xFFFFFFFFL
+//UVD_VCPU_IND_INDEX
+#define UVD_VCPU_IND_INDEX__INDEX__SHIFT                                                                      0x0
+#define UVD_VCPU_IND_INDEX__INDEX_MASK                                                                        0x000001FFL
+//UVD_VCPU_IND_DATA
+#define UVD_VCPU_IND_DATA__DATA__SHIFT                                                                        0x0
+#define UVD_VCPU_IND_DATA__DATA_MASK                                                                          0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_mpcdec
+//UVD_MP_SWAP_CNTL
+#define UVD_MP_SWAP_CNTL__MP_REF0_MC_SWAP__SHIFT                                                              0x0
+#define UVD_MP_SWAP_CNTL__MP_REF1_MC_SWAP__SHIFT                                                              0x2
+#define UVD_MP_SWAP_CNTL__MP_REF2_MC_SWAP__SHIFT                                                              0x4
+#define UVD_MP_SWAP_CNTL__MP_REF3_MC_SWAP__SHIFT                                                              0x6
+#define UVD_MP_SWAP_CNTL__MP_REF4_MC_SWAP__SHIFT                                                              0x8
+#define UVD_MP_SWAP_CNTL__MP_REF5_MC_SWAP__SHIFT                                                              0xa
+#define UVD_MP_SWAP_CNTL__MP_REF6_MC_SWAP__SHIFT                                                              0xc
+#define UVD_MP_SWAP_CNTL__MP_REF7_MC_SWAP__SHIFT                                                              0xe
+#define UVD_MP_SWAP_CNTL__MP_REF8_MC_SWAP__SHIFT                                                              0x10
+#define UVD_MP_SWAP_CNTL__MP_REF9_MC_SWAP__SHIFT                                                              0x12
+#define UVD_MP_SWAP_CNTL__MP_REF10_MC_SWAP__SHIFT                                                             0x14
+#define UVD_MP_SWAP_CNTL__MP_REF11_MC_SWAP__SHIFT                                                             0x16
+#define UVD_MP_SWAP_CNTL__MP_REF12_MC_SWAP__SHIFT                                                             0x18
+#define UVD_MP_SWAP_CNTL__MP_REF13_MC_SWAP__SHIFT                                                             0x1a
+#define UVD_MP_SWAP_CNTL__MP_REF14_MC_SWAP__SHIFT                                                             0x1c
+#define UVD_MP_SWAP_CNTL__MP_REF15_MC_SWAP__SHIFT                                                             0x1e
+#define UVD_MP_SWAP_CNTL__MP_REF0_MC_SWAP_MASK                                                                0x00000003L
+#define UVD_MP_SWAP_CNTL__MP_REF1_MC_SWAP_MASK                                                                0x0000000CL
+#define UVD_MP_SWAP_CNTL__MP_REF2_MC_SWAP_MASK                                                                0x00000030L
+#define UVD_MP_SWAP_CNTL__MP_REF3_MC_SWAP_MASK                                                                0x000000C0L
+#define UVD_MP_SWAP_CNTL__MP_REF4_MC_SWAP_MASK                                                                0x00000300L
+#define UVD_MP_SWAP_CNTL__MP_REF5_MC_SWAP_MASK                                                                0x00000C00L
+#define UVD_MP_SWAP_CNTL__MP_REF6_MC_SWAP_MASK                                                                0x00003000L
+#define UVD_MP_SWAP_CNTL__MP_REF7_MC_SWAP_MASK                                                                0x0000C000L
+#define UVD_MP_SWAP_CNTL__MP_REF8_MC_SWAP_MASK                                                                0x00030000L
+#define UVD_MP_SWAP_CNTL__MP_REF9_MC_SWAP_MASK                                                                0x000C0000L
+#define UVD_MP_SWAP_CNTL__MP_REF10_MC_SWAP_MASK                                                               0x00300000L
+#define UVD_MP_SWAP_CNTL__MP_REF11_MC_SWAP_MASK                                                               0x00C00000L
+#define UVD_MP_SWAP_CNTL__MP_REF12_MC_SWAP_MASK                                                               0x03000000L
+#define UVD_MP_SWAP_CNTL__MP_REF13_MC_SWAP_MASK                                                               0x0C000000L
+#define UVD_MP_SWAP_CNTL__MP_REF14_MC_SWAP_MASK                                                               0x30000000L
+#define UVD_MP_SWAP_CNTL__MP_REF15_MC_SWAP_MASK                                                               0xC0000000L
+//UVD_MP_SWAP_CNTL2
+#define UVD_MP_SWAP_CNTL2__MP_REF16_MC_SWAP__SHIFT                                                            0x0
+#define UVD_MP_SWAP_CNTL2__MP_REF16_MC_SWAP_MASK                                                              0x00000003L
+//UVD_MPC_LUMA_SRCH
+#define UVD_MPC_LUMA_SRCH__CNTR__SHIFT                                                                        0x0
+#define UVD_MPC_LUMA_SRCH__CNTR_MASK                                                                          0xFFFFFFFFL
+//UVD_MPC_LUMA_HIT
+#define UVD_MPC_LUMA_HIT__CNTR__SHIFT                                                                         0x0
+#define UVD_MPC_LUMA_HIT__CNTR_MASK                                                                           0xFFFFFFFFL
+//UVD_MPC_LUMA_HITPEND
+#define UVD_MPC_LUMA_HITPEND__CNTR__SHIFT                                                                     0x0
+#define UVD_MPC_LUMA_HITPEND__CNTR_MASK                                                                       0xFFFFFFFFL
+//UVD_MPC_CHROMA_SRCH
+#define UVD_MPC_CHROMA_SRCH__CNTR__SHIFT                                                                      0x0
+#define UVD_MPC_CHROMA_SRCH__CNTR_MASK                                                                        0xFFFFFFFFL
+//UVD_MPC_CHROMA_HIT
+#define UVD_MPC_CHROMA_HIT__CNTR__SHIFT                                                                       0x0
+#define UVD_MPC_CHROMA_HIT__CNTR_MASK                                                                         0xFFFFFFFFL
+//UVD_MPC_CHROMA_HITPEND
+#define UVD_MPC_CHROMA_HITPEND__CNTR__SHIFT                                                                   0x0
+#define UVD_MPC_CHROMA_HITPEND__CNTR_MASK                                                                     0xFFFFFFFFL
+//UVD_MPC_CNTL
+#define UVD_MPC_CNTL__BLK_RST__SHIFT                                                                          0x0
+#define UVD_MPC_CNTL__REG_MPC1_PERF_SELECT__SHIFT                                                             0x1
+#define UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT                                                                 0x3
+#define UVD_MPC_CNTL__PERF_RST__SHIFT                                                                         0x6
+#define UVD_MPC_CNTL__REG_MPC_CNTL_BACKWARD_COMPATIBILITY__SHIFT                                              0x7
+#define UVD_MPC_CNTL__DBG_MUX__SHIFT                                                                          0x8
+#define UVD_MPC_CNTL__AVE_WEIGHT__SHIFT                                                                       0x10
+#define UVD_MPC_CNTL__URGENT_EN__SHIFT                                                                        0x12
+#define UVD_MPC_CNTL__SMPAT_REQ_SPEED_UP__SHIFT                                                               0x13
+#define UVD_MPC_CNTL__TEST_MODE_EN__SHIFT                                                                     0x14
+#define UVD_MPC_CNTL__BLK_RST_MASK                                                                            0x00000001L
+#define UVD_MPC_CNTL__REG_MPC1_PERF_SELECT_MASK                                                               0x00000002L
+#define UVD_MPC_CNTL__REPLACEMENT_MODE_MASK                                                                   0x00000038L
+#define UVD_MPC_CNTL__PERF_RST_MASK                                                                           0x00000040L
+#define UVD_MPC_CNTL__REG_MPC_CNTL_BACKWARD_COMPATIBILITY_MASK                                                0x00000080L
+#define UVD_MPC_CNTL__DBG_MUX_MASK                                                                            0x00000F00L
+#define UVD_MPC_CNTL__AVE_WEIGHT_MASK                                                                         0x00030000L
+#define UVD_MPC_CNTL__URGENT_EN_MASK                                                                          0x00040000L
+#define UVD_MPC_CNTL__SMPAT_REQ_SPEED_UP_MASK                                                                 0x00080000L
+#define UVD_MPC_CNTL__TEST_MODE_EN_MASK                                                                       0x00300000L
+//UVD_MPC_PITCH
+#define UVD_MPC_PITCH__LUMA_PITCH__SHIFT                                                                      0x0
+#define UVD_MPC_PITCH__LUMA_PITCH_MASK                                                                        0x000007FFL
+//UVD_MPC_SET_MUXA0
+#define UVD_MPC_SET_MUXA0__VARA_0__SHIFT                                                                      0x0
+#define UVD_MPC_SET_MUXA0__VARA_1__SHIFT                                                                      0x6
+#define UVD_MPC_SET_MUXA0__VARA_2__SHIFT                                                                      0xc
+#define UVD_MPC_SET_MUXA0__VARA_3__SHIFT                                                                      0x12
+#define UVD_MPC_SET_MUXA0__VARA_4__SHIFT                                                                      0x18
+#define UVD_MPC_SET_MUXA0__VARA_0_MASK                                                                        0x0000003FL
+#define UVD_MPC_SET_MUXA0__VARA_1_MASK                                                                        0x00000FC0L
+#define UVD_MPC_SET_MUXA0__VARA_2_MASK                                                                        0x0003F000L
+#define UVD_MPC_SET_MUXA0__VARA_3_MASK                                                                        0x00FC0000L
+#define UVD_MPC_SET_MUXA0__VARA_4_MASK                                                                        0x3F000000L
+//UVD_MPC_SET_MUXA1
+#define UVD_MPC_SET_MUXA1__VARA_5__SHIFT                                                                      0x0
+#define UVD_MPC_SET_MUXA1__VARA_6__SHIFT                                                                      0x6
+#define UVD_MPC_SET_MUXA1__VARA_7__SHIFT                                                                      0xc
+#define UVD_MPC_SET_MUXA1__VARA_5_MASK                                                                        0x0000003FL
+#define UVD_MPC_SET_MUXA1__VARA_6_MASK                                                                        0x00000FC0L
+#define UVD_MPC_SET_MUXA1__VARA_7_MASK                                                                        0x0003F000L
+//UVD_MPC_SET_MUXB0
+#define UVD_MPC_SET_MUXB0__VARB_0__SHIFT                                                                      0x0
+#define UVD_MPC_SET_MUXB0__VARB_1__SHIFT                                                                      0x6
+#define UVD_MPC_SET_MUXB0__VARB_2__SHIFT                                                                      0xc
+#define UVD_MPC_SET_MUXB0__VARB_3__SHIFT                                                                      0x12
+#define UVD_MPC_SET_MUXB0__VARB_4__SHIFT                                                                      0x18
+#define UVD_MPC_SET_MUXB0__VARB_0_MASK                                                                        0x0000003FL
+#define UVD_MPC_SET_MUXB0__VARB_1_MASK                                                                        0x00000FC0L
+#define UVD_MPC_SET_MUXB0__VARB_2_MASK                                                                        0x0003F000L
+#define UVD_MPC_SET_MUXB0__VARB_3_MASK                                                                        0x00FC0000L
+#define UVD_MPC_SET_MUXB0__VARB_4_MASK                                                                        0x3F000000L
+//UVD_MPC_SET_MUXB1
+#define UVD_MPC_SET_MUXB1__VARB_5__SHIFT                                                                      0x0
+#define UVD_MPC_SET_MUXB1__VARB_6__SHIFT                                                                      0x6
+#define UVD_MPC_SET_MUXB1__VARB_7__SHIFT                                                                      0xc
+#define UVD_MPC_SET_MUXB1__VARB_5_MASK                                                                        0x0000003FL
+#define UVD_MPC_SET_MUXB1__VARB_6_MASK                                                                        0x00000FC0L
+#define UVD_MPC_SET_MUXB1__VARB_7_MASK                                                                        0x0003F000L
+//UVD_MPC_SET_MUX
+#define UVD_MPC_SET_MUX__SET_0__SHIFT                                                                         0x0
+#define UVD_MPC_SET_MUX__SET_1__SHIFT                                                                         0x3
+#define UVD_MPC_SET_MUX__SET_2__SHIFT                                                                         0x6
+#define UVD_MPC_SET_MUX__SET_0_MASK                                                                           0x00000007L
+#define UVD_MPC_SET_MUX__SET_1_MASK                                                                           0x00000038L
+#define UVD_MPC_SET_MUX__SET_2_MASK                                                                           0x000001C0L
+//UVD_MPC_SET_ALU
+#define UVD_MPC_SET_ALU__FUNCT__SHIFT                                                                         0x0
+#define UVD_MPC_SET_ALU__OPERAND__SHIFT                                                                       0x4
+#define UVD_MPC_SET_ALU__FUNCT_MASK                                                                           0x00000007L
+#define UVD_MPC_SET_ALU__OPERAND_MASK                                                                         0x00000FF0L
+//UVD_MPC_PERF0
+#define UVD_MPC_PERF0__MAX_LAT__SHIFT                                                                         0x0
+#define UVD_MPC_PERF0__MAX_LAT_MASK                                                                           0x000003FFL
+//UVD_MPC_PERF1
+#define UVD_MPC_PERF1__AVE_LAT__SHIFT                                                                         0x0
+#define UVD_MPC_PERF1__AVE_LAT_MASK                                                                           0x000003FFL
+//UVD_MPC_IND_INDEX
+#define UVD_MPC_IND_INDEX__INDEX__SHIFT                                                                       0x0
+#define UVD_MPC_IND_INDEX__INDEX_MASK                                                                         0x000001FFL
+//UVD_MPC_IND_DATA
+#define UVD_MPC_IND_DATA__DATA__SHIFT                                                                         0x0
+#define UVD_MPC_IND_DATA__DATA_MASK                                                                           0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_rbcdec
+//UVD_RBC_IB_SIZE
+#define UVD_RBC_IB_SIZE__IB_SIZE__SHIFT                                                                       0x4
+#define UVD_RBC_IB_SIZE__IB_SIZE_MASK                                                                         0x007FFFF0L
+//UVD_RBC_IB_SIZE_UPDATE
+#define UVD_RBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                                         0x4
+#define UVD_RBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                           0x007FFFF0L
+//UVD_RBC_RB_CNTL
+#define UVD_RBC_RB_CNTL__RB_BUFSZ__SHIFT                                                                      0x0
+#define UVD_RBC_RB_CNTL__RB_BLKSZ__SHIFT                                                                      0x8
+#define UVD_RBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                                   0x10
+#define UVD_RBC_RB_CNTL__RB_WPTR_POLL_EN__SHIFT                                                               0x14
+#define UVD_RBC_RB_CNTL__RB_NO_UPDATE__SHIFT                                                                  0x18
+#define UVD_RBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                                 0x1c
+#define UVD_RBC_RB_CNTL__BLK_RST__SHIFT                                                                       0x1d
+#define UVD_RBC_RB_CNTL__RB_BUFSZ_MASK                                                                        0x0000001FL
+#define UVD_RBC_RB_CNTL__RB_BLKSZ_MASK                                                                        0x00001F00L
+#define UVD_RBC_RB_CNTL__RB_NO_FETCH_MASK                                                                     0x00010000L
+#define UVD_RBC_RB_CNTL__RB_WPTR_POLL_EN_MASK                                                                 0x00100000L
+#define UVD_RBC_RB_CNTL__RB_NO_UPDATE_MASK                                                                    0x01000000L
+#define UVD_RBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                                   0x10000000L
+#define UVD_RBC_RB_CNTL__BLK_RST_MASK                                                                         0x20000000L
+//UVD_RBC_RB_RPTR_ADDR
+#define UVD_RBC_RB_RPTR_ADDR__RB_RPTR_ADDR__SHIFT                                                             0x0
+#define UVD_RBC_RB_RPTR_ADDR__RB_RPTR_ADDR_MASK                                                               0xFFFFFFFFL
+//UVD_RBC_VCPU_ACCESS
+#define UVD_RBC_VCPU_ACCESS__ENABLE_RBC__SHIFT                                                                0x0
+#define UVD_RBC_VCPU_ACCESS__ENABLE_RBC_MASK                                                                  0x00000001L
+//UVD_FW_SEMAPHORE_CNTL
+#define UVD_FW_SEMAPHORE_CNTL__START__SHIFT                                                                   0x0
+#define UVD_FW_SEMAPHORE_CNTL__BUSY__SHIFT                                                                    0x8
+#define UVD_FW_SEMAPHORE_CNTL__PASS__SHIFT                                                                    0x9
+#define UVD_FW_SEMAPHORE_CNTL__START_MASK                                                                     0x00000001L
+#define UVD_FW_SEMAPHORE_CNTL__BUSY_MASK                                                                      0x00000100L
+#define UVD_FW_SEMAPHORE_CNTL__PASS_MASK                                                                      0x00000200L
+//UVD_RBC_READ_REQ_URGENT_CNTL
+#define UVD_RBC_READ_REQ_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                       0x0
+#define UVD_RBC_READ_REQ_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                         0x00000003L
+//UVD_RBC_RB_WPTR_CNTL
+#define UVD_RBC_RB_WPTR_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                       0x0
+#define UVD_RBC_RB_WPTR_CNTL__RB_PRE_WRITE_TIMER_MASK                                                         0x00007FFFL
+//UVD_RBC_WPTR_STATUS
+#define UVD_RBC_WPTR_STATUS__RB_WPTR_IN_USE__SHIFT                                                            0x4
+#define UVD_RBC_WPTR_STATUS__RB_WPTR_IN_USE_MASK                                                              0x007FFFF0L
+//UVD_RBC_WPTR_POLL_CNTL
+#define UVD_RBC_WPTR_POLL_CNTL__POLL_FREQ__SHIFT                                                              0x0
+#define UVD_RBC_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT                                                        0x10
+#define UVD_RBC_WPTR_POLL_CNTL__POLL_FREQ_MASK                                                                0x0000FFFFL
+#define UVD_RBC_WPTR_POLL_CNTL__IDLE_POLL_COUNT_MASK                                                          0xFFFF0000L
+//UVD_RBC_WPTR_POLL_ADDR
+#define UVD_RBC_WPTR_POLL_ADDR__POLL_ADDR__SHIFT                                                              0x2
+#define UVD_RBC_WPTR_POLL_ADDR__POLL_ADDR_MASK                                                                0xFFFFFFFCL
+//UVD_SEMA_CMD
+#define UVD_SEMA_CMD__REQ_CMD__SHIFT                                                                          0x0
+#define UVD_SEMA_CMD__WR_PHASE__SHIFT                                                                         0x4
+#define UVD_SEMA_CMD__MODE__SHIFT                                                                             0x6
+#define UVD_SEMA_CMD__VMID_EN__SHIFT                                                                          0x7
+#define UVD_SEMA_CMD__VMID__SHIFT                                                                             0x8
+#define UVD_SEMA_CMD__REQ_CMD_MASK                                                                            0x0000000FL
+#define UVD_SEMA_CMD__WR_PHASE_MASK                                                                           0x00000030L
+#define UVD_SEMA_CMD__MODE_MASK                                                                               0x00000040L
+#define UVD_SEMA_CMD__VMID_EN_MASK                                                                            0x00000080L
+#define UVD_SEMA_CMD__VMID_MASK                                                                               0x00000F00L
+//UVD_SEMA_ADDR_LOW
+#define UVD_SEMA_ADDR_LOW__ADDR_26_3__SHIFT                                                                   0x0
+#define UVD_SEMA_ADDR_LOW__ADDR_26_3_MASK                                                                     0x00FFFFFFL
+//UVD_SEMA_ADDR_HIGH
+#define UVD_SEMA_ADDR_HIGH__ADDR_47_27__SHIFT                                                                 0x0
+#define UVD_SEMA_ADDR_HIGH__ADDR_47_27_MASK                                                                   0x001FFFFFL
+//UVD_ENGINE_CNTL
+#define UVD_ENGINE_CNTL__ENGINE_START__SHIFT                                                                  0x0
+#define UVD_ENGINE_CNTL__ENGINE_START_MODE__SHIFT                                                             0x1
+#define UVD_ENGINE_CNTL__NJ_PF_HANDLE_DISABLE__SHIFT                                                          0x2
+#define UVD_ENGINE_CNTL__ENGINE_START_MASK                                                                    0x00000001L
+#define UVD_ENGINE_CNTL__ENGINE_START_MODE_MASK                                                               0x00000002L
+#define UVD_ENGINE_CNTL__NJ_PF_HANDLE_DISABLE_MASK                                                            0x00000004L
+//UVD_SEMA_TIMEOUT_STATUS
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_INCOMPLETE_TIMEOUT_STAT__SHIFT                                0x0
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_FAULT_TIMEOUT_STAT__SHIFT                                     0x1
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_SIGNAL_INCOMPLETE_TIMEOUT_STAT__SHIFT                              0x2
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_TIMEOUT_CLEAR__SHIFT                                               0x3
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_INCOMPLETE_TIMEOUT_STAT_MASK                                  0x00000001L
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_FAULT_TIMEOUT_STAT_MASK                                       0x00000002L
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_SIGNAL_INCOMPLETE_TIMEOUT_STAT_MASK                                0x00000004L
+#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_TIMEOUT_CLEAR_MASK                                                 0x00000008L
+//UVD_SEMA_CNTL
+#define UVD_SEMA_CNTL__SEMAPHORE_EN__SHIFT                                                                    0x0
+#define UVD_SEMA_CNTL__ADVANCED_MODE_DIS__SHIFT                                                               0x1
+#define UVD_SEMA_CNTL__SEMAPHORE_EN_MASK                                                                      0x00000001L
+#define UVD_SEMA_CNTL__ADVANCED_MODE_DIS_MASK                                                                 0x00000002L
+//UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_EN__SHIFT                                  0x0
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_COUNT__SHIFT                               0x1
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER__SHIFT                                          0x18
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_EN_MASK                                    0x00000001L
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_COUNT_MASK                                 0x001FFFFEL
+#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER_MASK                                            0x07000000L
+//UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_EN__SHIFT                                                0x0
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_COUNT__SHIFT                                             0x1
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__RESEND_TIMER__SHIFT                                                 0x18
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_EN_MASK                                                  0x00000001L
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_COUNT_MASK                                               0x001FFFFEL
+#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__RESEND_TIMER_MASK                                                   0x07000000L
+//UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_EN__SHIFT                                      0x0
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_COUNT__SHIFT                                   0x1
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER__SHIFT                                            0x18
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_EN_MASK                                        0x00000001L
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_COUNT_MASK                                     0x001FFFFEL
+#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER_MASK                                              0x07000000L
+//UVD_JOB_START
+#define UVD_JOB_START__JOB_START__SHIFT                                                                       0x0
+#define UVD_JOB_START__JOB_START_MASK                                                                         0x00000001L
+//UVD_RBC_BUF_STATUS
+#define UVD_RBC_BUF_STATUS__RB_BUF_VALID__SHIFT                                                               0x0
+#define UVD_RBC_BUF_STATUS__IB_BUF_VALID__SHIFT                                                               0x8
+#define UVD_RBC_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                                             0x10
+#define UVD_RBC_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                                             0x13
+#define UVD_RBC_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                                             0x16
+#define UVD_RBC_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                                             0x19
+#define UVD_RBC_BUF_STATUS__RB_BUF_VALID_MASK                                                                 0x000000FFL
+#define UVD_RBC_BUF_STATUS__IB_BUF_VALID_MASK                                                                 0x0000FF00L
+#define UVD_RBC_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                               0x00070000L
+#define UVD_RBC_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                               0x00380000L
+#define UVD_RBC_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                               0x01C00000L
+#define UVD_RBC_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                               0x0E000000L
+//UVD_RBC_SWAP_CNTL
+#define UVD_RBC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                                  0x0
+#define UVD_RBC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                                  0x2
+#define UVD_RBC_SWAP_CNTL__RB_RPTR_MC_SWAP__SHIFT                                                             0x4
+#define UVD_RBC_SWAP_CNTL__RB_WR_MC_SWAP__SHIFT                                                               0x1a
+#define UVD_RBC_SWAP_CNTL__RB_MC_SWAP_MASK                                                                    0x00000003L
+#define UVD_RBC_SWAP_CNTL__IB_MC_SWAP_MASK                                                                    0x0000000CL
+#define UVD_RBC_SWAP_CNTL__RB_RPTR_MC_SWAP_MASK                                                               0x00000030L
+#define UVD_RBC_SWAP_CNTL__RB_WR_MC_SWAP_MASK                                                                 0x0C000000L
+
+
+// addressBlock: aid_uvd0_lmi_adpdec
+//UVD_LMI_RE_64BIT_BAR_LOW
+#define UVD_LMI_RE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                            0x0
+#define UVD_LMI_RE_64BIT_BAR_LOW__BITS_31_0_MASK                                                              0xFFFFFFFFL
+//UVD_LMI_RE_64BIT_BAR_HIGH
+#define UVD_LMI_RE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                          0x0
+#define UVD_LMI_RE_64BIT_BAR_HIGH__BITS_63_32_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_IT_64BIT_BAR_LOW
+#define UVD_LMI_IT_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                            0x0
+#define UVD_LMI_IT_64BIT_BAR_LOW__BITS_31_0_MASK                                                              0xFFFFFFFFL
+//UVD_LMI_IT_64BIT_BAR_HIGH
+#define UVD_LMI_IT_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                          0x0
+#define UVD_LMI_IT_64BIT_BAR_HIGH__BITS_63_32_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_MP_64BIT_BAR_LOW
+#define UVD_LMI_MP_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                            0x0
+#define UVD_LMI_MP_64BIT_BAR_LOW__BITS_31_0_MASK                                                              0xFFFFFFFFL
+//UVD_LMI_MP_64BIT_BAR_HIGH
+#define UVD_LMI_MP_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                          0x0
+#define UVD_LMI_MP_64BIT_BAR_HIGH__BITS_63_32_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_CM_64BIT_BAR_LOW
+#define UVD_LMI_CM_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                            0x0
+#define UVD_LMI_CM_64BIT_BAR_LOW__BITS_31_0_MASK                                                              0xFFFFFFFFL
+//UVD_LMI_CM_64BIT_BAR_HIGH
+#define UVD_LMI_CM_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                          0x0
+#define UVD_LMI_CM_64BIT_BAR_HIGH__BITS_63_32_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_DB_64BIT_BAR_LOW
+#define UVD_LMI_DB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                            0x0
+#define UVD_LMI_DB_64BIT_BAR_LOW__BITS_31_0_MASK                                                              0xFFFFFFFFL
+//UVD_LMI_DB_64BIT_BAR_HIGH
+#define UVD_LMI_DB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                          0x0
+#define UVD_LMI_DB_64BIT_BAR_HIGH__BITS_63_32_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_DBW_64BIT_BAR_LOW
+#define UVD_LMI_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                           0x0
+#define UVD_LMI_DBW_64BIT_BAR_LOW__BITS_31_0_MASK                                                             0xFFFFFFFFL
+//UVD_LMI_DBW_64BIT_BAR_HIGH
+#define UVD_LMI_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                         0x0
+#define UVD_LMI_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK                                                           0xFFFFFFFFL
+//UVD_LMI_IDCT_64BIT_BAR_LOW
+#define UVD_LMI_IDCT_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                          0x0
+#define UVD_LMI_IDCT_64BIT_BAR_LOW__BITS_31_0_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_IDCT_64BIT_BAR_HIGH
+#define UVD_LMI_IDCT_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                        0x0
+#define UVD_LMI_IDCT_64BIT_BAR_HIGH__BITS_63_32_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_MPRD_S0_64BIT_BAR_LOW
+#define UVD_LMI_MPRD_S0_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                       0x0
+#define UVD_LMI_MPRD_S0_64BIT_BAR_LOW__BITS_31_0_MASK                                                         0xFFFFFFFFL
+//UVD_LMI_MPRD_S0_64BIT_BAR_HIGH
+#define UVD_LMI_MPRD_S0_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                     0x0
+#define UVD_LMI_MPRD_S0_64BIT_BAR_HIGH__BITS_63_32_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MPRD_S1_64BIT_BAR_LOW
+#define UVD_LMI_MPRD_S1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                       0x0
+#define UVD_LMI_MPRD_S1_64BIT_BAR_LOW__BITS_31_0_MASK                                                         0xFFFFFFFFL
+//UVD_LMI_MPRD_S1_64BIT_BAR_HIGH
+#define UVD_LMI_MPRD_S1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                     0x0
+#define UVD_LMI_MPRD_S1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MPRD_DBW_64BIT_BAR_LOW
+#define UVD_LMI_MPRD_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MPRD_DBW_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH
+#define UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MPC_64BIT_BAR_LOW
+#define UVD_LMI_MPC_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                           0x0
+#define UVD_LMI_MPC_64BIT_BAR_LOW__BITS_31_0_MASK                                                             0xFFFFFFFFL
+//UVD_LMI_MPC_64BIT_BAR_HIGH
+#define UVD_LMI_MPC_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                         0x0
+#define UVD_LMI_MPC_64BIT_BAR_HIGH__BITS_63_32_MASK                                                           0xFFFFFFFFL
+//UVD_LMI_RBC_RB_64BIT_BAR_LOW
+#define UVD_LMI_RBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                        0x0
+#define UVD_LMI_RBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_RBC_RB_64BIT_BAR_HIGH
+#define UVD_LMI_RBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                      0x0
+#define UVD_LMI_RBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_RBC_IB_64BIT_BAR_LOW
+#define UVD_LMI_RBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                        0x0
+#define UVD_LMI_RBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_RBC_IB_64BIT_BAR_HIGH
+#define UVD_LMI_RBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                      0x0
+#define UVD_LMI_RBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_LBSI_64BIT_BAR_LOW
+#define UVD_LMI_LBSI_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                          0x0
+#define UVD_LMI_LBSI_64BIT_BAR_LOW__BITS_31_0_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_LBSI_64BIT_BAR_HIGH
+#define UVD_LMI_LBSI_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                        0x0
+#define UVD_LMI_LBSI_64BIT_BAR_HIGH__BITS_63_32_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_VCPU_NC0_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_NC0_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_VCPU_NC0_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_VCPU_NC1_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_NC1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_VCPU_NC1_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                    0x0
+#define UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                  0x0
+#define UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32_MASK                                                    0xFFFFFFFFL
+//UVD_LMI_CENC_64BIT_BAR_LOW
+#define UVD_LMI_CENC_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                          0x0
+#define UVD_LMI_CENC_64BIT_BAR_LOW__BITS_31_0_MASK                                                            0xFFFFFFFFL
+//UVD_LMI_CENC_64BIT_BAR_HIGH
+#define UVD_LMI_CENC_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                        0x0
+#define UVD_LMI_CENC_64BIT_BAR_HIGH__BITS_63_32_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_SRE_64BIT_BAR_LOW
+#define UVD_LMI_SRE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                           0x0
+#define UVD_LMI_SRE_64BIT_BAR_LOW__BITS_31_0_MASK                                                             0xFFFFFFFFL
+//UVD_LMI_SRE_64BIT_BAR_HIGH
+#define UVD_LMI_SRE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                         0x0
+#define UVD_LMI_SRE_64BIT_BAR_HIGH__BITS_63_32_MASK                                                           0xFFFFFFFFL
+//UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW
+#define UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                 0x0
+#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                               0x0
+#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                                 0xFFFFFFFFL
+//UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                               0x0
+#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK                                                 0xFFFFFFFFL
+//UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                             0x0
+#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                               0xFFFFFFFFL
+//UVD_LMI_MIF_REF_64BIT_BAR_LOW
+#define UVD_LMI_MIF_REF_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                       0x0
+#define UVD_LMI_MIF_REF_64BIT_BAR_LOW__BITS_31_0_MASK                                                         0xFFFFFFFFL
+//UVD_LMI_MIF_REF_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_REF_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                     0x0
+#define UVD_LMI_MIF_REF_64BIT_BAR_HIGH__BITS_63_32_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MIF_DBW_64BIT_BAR_LOW
+#define UVD_LMI_MIF_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                       0x0
+#define UVD_LMI_MIF_DBW_64BIT_BAR_LOW__BITS_31_0_MASK                                                         0xFFFFFFFFL
+//UVD_LMI_MIF_DBW_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                     0x0
+#define UVD_LMI_MIF_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW
+#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                  0x0
+#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW__BITS_31_0_MASK                                                    0xFFFFFFFFL
+//UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                0x0
+#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH__BITS_63_32_MASK                                                  0xFFFFFFFFL
+//UVD_LMI_MIF_BSP0_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSP0_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSP0_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSP1_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSP1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSP1_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSP2_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSP2_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSP2_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSP3_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSP3_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSP3_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSD0_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSD0_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSD0_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSD1_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSD1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSD1_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSD2_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSD2_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSD2_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSD3_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSD3_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSD3_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_BSD4_64BIT_BAR_LOW
+#define UVD_LMI_MIF_BSD4_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_BSD4_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW
+#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                   0x0
+#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW__BITS_31_0_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH
+#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                 0x0
+#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH__BITS_63_32_MASK                                                   0xFFFFFFFFL
+//UVD_LMI_MIF_SCLR_64BIT_BAR_LOW
+#define UVD_LMI_MIF_SCLR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                      0x0
+#define UVD_LMI_MIF_SCLR_64BIT_BAR_LOW__BITS_31_0_MASK                                                        0xFFFFFFFFL
+//UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                    0x0
+#define UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH__BITS_63_32_MASK                                                      0xFFFFFFFFL
+//UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW
+#define UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_SPH_64BIT_BAR_HIGH
+#define UVD_LMI_SPH_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                         0x0
+#define UVD_LMI_SPH_64BIT_BAR_HIGH__BITS_63_32_MASK                                                           0xFFFFFFFFL
+//UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                         0x0
+#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK                                           0xFFFFFFFFL
+//UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                       0x0
+#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                         0xFFFFFFFFL
+//UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_ADP_ATOMIC_CONFIG
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER0_WR_CACHE__SHIFT                                                   0x0
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER1_WR_CACHE__SHIFT                                                   0x4
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER2_WR_CACHE__SHIFT                                                   0x8
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER3_WR_CACHE__SHIFT                                                   0xc
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_RD_URG__SHIFT                                                           0x10
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER0_WR_CACHE_MASK                                                     0x0000000FL
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER1_WR_CACHE_MASK                                                     0x000000F0L
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER2_WR_CACHE_MASK                                                     0x00000F00L
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER3_WR_CACHE_MASK                                                     0x0000F000L
+#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_RD_URG_MASK                                                             0x000F0000L
+//UVD_LMI_ARB_CTRL2
+#define UVD_LMI_ARB_CTRL2__CENC_RD_WAIT_EN__SHIFT                                                             0x0
+#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_WAIT_EN__SHIFT                                                           0x1
+#define UVD_LMI_ARB_CTRL2__CENC_RD_MAX_BURST__SHIFT                                                           0x2
+#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_MAX_BURST__SHIFT                                                         0x6
+#define UVD_LMI_ARB_CTRL2__MIF_RD_REQ_RET_MAX__SHIFT                                                          0xa
+#define UVD_LMI_ARB_CTRL2__MIF_WR_REQ_RET_MAX__SHIFT                                                          0x14
+#define UVD_LMI_ARB_CTRL2__CENC_RD_WAIT_EN_MASK                                                               0x00000001L
+#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_WAIT_EN_MASK                                                             0x00000002L
+#define UVD_LMI_ARB_CTRL2__CENC_RD_MAX_BURST_MASK                                                             0x0000003CL
+#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_MAX_BURST_MASK                                                           0x000003C0L
+#define UVD_LMI_ARB_CTRL2__MIF_RD_REQ_RET_MAX_MASK                                                            0x000FFC00L
+#define UVD_LMI_ARB_CTRL2__MIF_WR_REQ_RET_MAX_MASK                                                            0xFFF00000L
+//UVD_LMI_VCPU_CACHE_VMIDS_MULTI
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE1_VMID__SHIFT                                               0x0
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE2_VMID__SHIFT                                               0x4
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE3_VMID__SHIFT                                               0x8
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE4_VMID__SHIFT                                               0xc
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE5_VMID__SHIFT                                               0x10
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE6_VMID__SHIFT                                               0x14
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE7_VMID__SHIFT                                               0x18
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE8_VMID__SHIFT                                               0x1c
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE1_VMID_MASK                                                 0x0000000FL
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE2_VMID_MASK                                                 0x000000F0L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE3_VMID_MASK                                                 0x00000F00L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE4_VMID_MASK                                                 0x0000F000L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE5_VMID_MASK                                                 0x000F0000L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE6_VMID_MASK                                                 0x00F00000L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE7_VMID_MASK                                                 0x0F000000L
+#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE8_VMID_MASK                                                 0xF0000000L
+//UVD_LMI_VCPU_NC_VMIDS_MULTI
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC2_VMID__SHIFT                                                     0x4
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC3_VMID__SHIFT                                                     0x8
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC4_VMID__SHIFT                                                     0xc
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC5_VMID__SHIFT                                                     0x10
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC6_VMID__SHIFT                                                     0x14
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC7_VMID__SHIFT                                                     0x18
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC2_VMID_MASK                                                       0x000000F0L
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC3_VMID_MASK                                                       0x00000F00L
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC4_VMID_MASK                                                       0x0000F000L
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC5_VMID_MASK                                                       0x000F0000L
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC6_VMID_MASK                                                       0x00F00000L
+#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC7_VMID_MASK                                                       0x0F000000L
+//UVD_LMI_LAT_CTRL
+#define UVD_LMI_LAT_CTRL__SCALE__SHIFT                                                                        0x0
+#define UVD_LMI_LAT_CTRL__MAX_START__SHIFT                                                                    0x8
+#define UVD_LMI_LAT_CTRL__MIN_START__SHIFT                                                                    0x9
+#define UVD_LMI_LAT_CTRL__AVG_START__SHIFT                                                                    0xa
+#define UVD_LMI_LAT_CTRL__PERFMON_SYNC__SHIFT                                                                 0xb
+#define UVD_LMI_LAT_CTRL__SKIP__SHIFT                                                                         0x10
+#define UVD_LMI_LAT_CTRL__SCALE_MASK                                                                          0x000000FFL
+#define UVD_LMI_LAT_CTRL__MAX_START_MASK                                                                      0x00000100L
+#define UVD_LMI_LAT_CTRL__MIN_START_MASK                                                                      0x00000200L
+#define UVD_LMI_LAT_CTRL__AVG_START_MASK                                                                      0x00000400L
+#define UVD_LMI_LAT_CTRL__PERFMON_SYNC_MASK                                                                   0x00000800L
+#define UVD_LMI_LAT_CTRL__SKIP_MASK                                                                           0x000F0000L
+//UVD_LMI_LAT_CNTR
+#define UVD_LMI_LAT_CNTR__MAX_LAT__SHIFT                                                                      0x0
+#define UVD_LMI_LAT_CNTR__MIN_LAT__SHIFT                                                                      0x8
+#define UVD_LMI_LAT_CNTR__MAX_LAT_MASK                                                                        0x000000FFL
+#define UVD_LMI_LAT_CNTR__MIN_LAT_MASK                                                                        0x0000FF00L
+//UVD_LMI_AVG_LAT_CNTR
+#define UVD_LMI_AVG_LAT_CNTR__ENV_LOW__SHIFT                                                                  0x0
+#define UVD_LMI_AVG_LAT_CNTR__ENV_HIGH__SHIFT                                                                 0x8
+#define UVD_LMI_AVG_LAT_CNTR__ENV_HIT__SHIFT                                                                  0x10
+#define UVD_LMI_AVG_LAT_CNTR__ENV_LOW_MASK                                                                    0x000000FFL
+#define UVD_LMI_AVG_LAT_CNTR__ENV_HIGH_MASK                                                                   0x0000FF00L
+#define UVD_LMI_AVG_LAT_CNTR__ENV_HIT_MASK                                                                    0xFFFF0000L
+//UVD_LMI_SPH
+#define UVD_LMI_SPH__ADDR__SHIFT                                                                              0x0
+#define UVD_LMI_SPH__STS__SHIFT                                                                               0x1c
+#define UVD_LMI_SPH__STS_VALID__SHIFT                                                                         0x1e
+#define UVD_LMI_SPH__STS_OVERFLOW__SHIFT                                                                      0x1f
+#define UVD_LMI_SPH__ADDR_MASK                                                                                0x0FFFFFFFL
+#define UVD_LMI_SPH__STS_MASK                                                                                 0x30000000L
+#define UVD_LMI_SPH__STS_VALID_MASK                                                                           0x40000000L
+#define UVD_LMI_SPH__STS_OVERFLOW_MASK                                                                        0x80000000L
+//UVD_LMI_VCPU_CACHE_VMID
+#define UVD_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID__SHIFT                                                       0x0
+#define UVD_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID_MASK                                                         0x0000000FL
+//UVD_LMI_CTRL2
+#define UVD_LMI_CTRL2__SPH_DIS__SHIFT                                                                         0x0
+#define UVD_LMI_CTRL2__STALL_ARB__SHIFT                                                                       0x1
+#define UVD_LMI_CTRL2__ASSERT_UMC_URGENT__SHIFT                                                               0x2
+#define UVD_LMI_CTRL2__MASK_UMC_URGENT__SHIFT                                                                 0x3
+#define UVD_LMI_CTRL2__CRC1_RESET__SHIFT                                                                      0x4
+#define UVD_LMI_CTRL2__DRCITF_BUBBLE_FIX_DIS__SHIFT                                                           0x7
+#define UVD_LMI_CTRL2__STALL_ARB_UMC__SHIFT                                                                   0x8
+#define UVD_LMI_CTRL2__MC_READ_ID_SEL__SHIFT                                                                  0x9
+#define UVD_LMI_CTRL2__MC_WRITE_ID_SEL__SHIFT                                                                 0xb
+#define UVD_LMI_CTRL2__VCPU_NC0_EXT_EN__SHIFT                                                                 0xd
+#define UVD_LMI_CTRL2__VCPU_NC1_EXT_EN__SHIFT                                                                 0xe
+#define UVD_LMI_CTRL2__SPU_EXTRA_CID_EN__SHIFT                                                                0xf
+#define UVD_LMI_CTRL2__RE_OFFLOAD_EN__SHIFT                                                                   0x10
+#define UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM__SHIFT                                                          0x11
+#define UVD_LMI_CTRL2__CLEAR_NJ_PF_BP__SHIFT                                                                  0x19
+#define UVD_LMI_CTRL2__NJ_MIF_GATING__SHIFT                                                                   0x1a
+#define UVD_LMI_CTRL2__CRC1_SEL__SHIFT                                                                        0x1b
+#define UVD_LMI_CTRL2__SPH_DIS_MASK                                                                           0x00000001L
+#define UVD_LMI_CTRL2__STALL_ARB_MASK                                                                         0x00000002L
+#define UVD_LMI_CTRL2__ASSERT_UMC_URGENT_MASK                                                                 0x00000004L
+#define UVD_LMI_CTRL2__MASK_UMC_URGENT_MASK                                                                   0x00000008L
+#define UVD_LMI_CTRL2__CRC1_RESET_MASK                                                                        0x00000010L
+#define UVD_LMI_CTRL2__DRCITF_BUBBLE_FIX_DIS_MASK                                                             0x00000080L
+#define UVD_LMI_CTRL2__STALL_ARB_UMC_MASK                                                                     0x00000100L
+#define UVD_LMI_CTRL2__MC_READ_ID_SEL_MASK                                                                    0x00000600L
+#define UVD_LMI_CTRL2__MC_WRITE_ID_SEL_MASK                                                                   0x00001800L
+#define UVD_LMI_CTRL2__VCPU_NC0_EXT_EN_MASK                                                                   0x00002000L
+#define UVD_LMI_CTRL2__VCPU_NC1_EXT_EN_MASK                                                                   0x00004000L
+#define UVD_LMI_CTRL2__SPU_EXTRA_CID_EN_MASK                                                                  0x00008000L
+#define UVD_LMI_CTRL2__RE_OFFLOAD_EN_MASK                                                                     0x00010000L
+#define UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM_MASK                                                            0x01FE0000L
+#define UVD_LMI_CTRL2__CLEAR_NJ_PF_BP_MASK                                                                    0x02000000L
+#define UVD_LMI_CTRL2__NJ_MIF_GATING_MASK                                                                     0x04000000L
+#define UVD_LMI_CTRL2__CRC1_SEL_MASK                                                                          0xF8000000L
+//UVD_LMI_URGENT_CTRL
+#define UVD_LMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL__SHIFT                                                 0x0
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_STALL__SHIFT                                                        0x1
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_URGENT__SHIFT                                                       0x2
+#define UVD_LMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL__SHIFT                                                 0x8
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_STALL__SHIFT                                                        0x9
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_URGENT__SHIFT                                                       0xa
+#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_RD_URGENT_STALL__SHIFT                                                0x10
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_STALL__SHIFT                                                       0x11
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_URGENT__SHIFT                                                      0x12
+#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_WR_URGENT_STALL__SHIFT                                                0x18
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_STALL__SHIFT                                                       0x19
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_URGENT__SHIFT                                                      0x1a
+#define UVD_LMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL_MASK                                                   0x00000001L
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_STALL_MASK                                                          0x00000002L
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_URGENT_MASK                                                         0x0000003CL
+#define UVD_LMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL_MASK                                                   0x00000100L
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_STALL_MASK                                                          0x00000200L
+#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_URGENT_MASK                                                         0x00003C00L
+#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_RD_URGENT_STALL_MASK                                                  0x00010000L
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_STALL_MASK                                                         0x00020000L
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_URGENT_MASK                                                        0x003C0000L
+#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_WR_URGENT_STALL_MASK                                                  0x01000000L
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_STALL_MASK                                                         0x02000000L
+#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_URGENT_MASK                                                        0x3C000000L
+//UVD_LMI_CTRL
+#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT                                                                0x0
+#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN__SHIFT                                                             0x8
+#define UVD_LMI_CTRL__REQ_MODE__SHIFT                                                                         0x9
+#define UVD_LMI_CTRL__ASSERT_MC_URGENT__SHIFT                                                                 0xb
+#define UVD_LMI_CTRL__MASK_MC_URGENT__SHIFT                                                                   0xc
+#define UVD_LMI_CTRL__DATA_COHERENCY_EN__SHIFT                                                                0xd
+#define UVD_LMI_CTRL__CRC_RESET__SHIFT                                                                        0xe
+#define UVD_LMI_CTRL__CRC_SEL__SHIFT                                                                          0xf
+#define UVD_LMI_CTRL__DISABLE_ON_FWV_FAIL__SHIFT                                                              0x14
+#define UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN__SHIFT                                                           0x15
+#define UVD_LMI_CTRL__CM_DATA_COHERENCY_EN__SHIFT                                                             0x16
+#define UVD_LMI_CTRL__DB_DB_DATA_COHERENCY_EN__SHIFT                                                          0x17
+#define UVD_LMI_CTRL__DB_IT_DATA_COHERENCY_EN__SHIFT                                                          0x18
+#define UVD_LMI_CTRL__IT_IT_DATA_COHERENCY_EN__SHIFT                                                          0x19
+#define UVD_LMI_CTRL__MIF_MIF_DATA_COHERENCY_EN__SHIFT                                                        0x1a
+#define UVD_LMI_CTRL__MIF_LESS_OUTSTANDING_RD_REQ__SHIFT                                                      0x1b
+#define UVD_LMI_CTRL__MC_BLK_RST__SHIFT                                                                       0x1c
+#define UVD_LMI_CTRL__UMC_BLK_RST__SHIFT                                                                      0x1d
+#define UVD_LMI_CTRL__RFU__SHIFT                                                                              0x1e
+#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_MASK                                                                  0x000000FFL
+#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK                                                               0x00000100L
+#define UVD_LMI_CTRL__REQ_MODE_MASK                                                                           0x00000200L
+#define UVD_LMI_CTRL__ASSERT_MC_URGENT_MASK                                                                   0x00000800L
+#define UVD_LMI_CTRL__MASK_MC_URGENT_MASK                                                                     0x00001000L
+#define UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK                                                                  0x00002000L
+#define UVD_LMI_CTRL__CRC_RESET_MASK                                                                          0x00004000L
+#define UVD_LMI_CTRL__CRC_SEL_MASK                                                                            0x000F8000L
+#define UVD_LMI_CTRL__DISABLE_ON_FWV_FAIL_MASK                                                                0x00100000L
+#define UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK                                                             0x00200000L
+#define UVD_LMI_CTRL__CM_DATA_COHERENCY_EN_MASK                                                               0x00400000L
+#define UVD_LMI_CTRL__DB_DB_DATA_COHERENCY_EN_MASK                                                            0x00800000L
+#define UVD_LMI_CTRL__DB_IT_DATA_COHERENCY_EN_MASK                                                            0x01000000L
+#define UVD_LMI_CTRL__IT_IT_DATA_COHERENCY_EN_MASK                                                            0x02000000L
+#define UVD_LMI_CTRL__MIF_MIF_DATA_COHERENCY_EN_MASK                                                          0x04000000L
+#define UVD_LMI_CTRL__MIF_LESS_OUTSTANDING_RD_REQ_MASK                                                        0x08000000L
+#define UVD_LMI_CTRL__MC_BLK_RST_MASK                                                                         0x10000000L
+#define UVD_LMI_CTRL__UMC_BLK_RST_MASK                                                                        0x20000000L
+#define UVD_LMI_CTRL__RFU_MASK                                                                                0xC0000000L
+//UVD_LMI_STATUS
+#define UVD_LMI_STATUS__READ_CLEAN__SHIFT                                                                     0x0
+#define UVD_LMI_STATUS__WRITE_CLEAN__SHIFT                                                                    0x1
+#define UVD_LMI_STATUS__WRITE_CLEAN_RAW__SHIFT                                                                0x2
+#define UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN__SHIFT                                                           0x3
+#define UVD_LMI_STATUS__UMC_READ_CLEAN__SHIFT                                                                 0x4
+#define UVD_LMI_STATUS__UMC_WRITE_CLEAN__SHIFT                                                                0x5
+#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW__SHIFT                                                            0x6
+#define UVD_LMI_STATUS__PENDING_UVD_MC_WRITE__SHIFT                                                           0x7
+#define UVD_LMI_STATUS__READ_CLEAN_RAW__SHIFT                                                                 0x8
+#define UVD_LMI_STATUS__UMC_READ_CLEAN_RAW__SHIFT                                                             0x9
+#define UVD_LMI_STATUS__UMC_UVD_IDLE__SHIFT                                                                   0xa
+#define UVD_LMI_STATUS__UMC_AVP_IDLE__SHIFT                                                                   0xb
+#define UVD_LMI_STATUS__ADP_MC_READ_CLEAN__SHIFT                                                              0xc
+#define UVD_LMI_STATUS__ADP_UMC_READ_CLEAN__SHIFT                                                             0xd
+#define UVD_LMI_STATUS__BSP0_WRITE_CLEAN__SHIFT                                                               0x12
+#define UVD_LMI_STATUS__BSP1_WRITE_CLEAN__SHIFT                                                               0x13
+#define UVD_LMI_STATUS__BSP2_WRITE_CLEAN__SHIFT                                                               0x14
+#define UVD_LMI_STATUS__BSP3_WRITE_CLEAN__SHIFT                                                               0x15
+#define UVD_LMI_STATUS__CENC_READ_CLEAN__SHIFT                                                                0x16
+#define UVD_LMI_STATUS__READ_CLEAN_MASK                                                                       0x00000001L
+#define UVD_LMI_STATUS__WRITE_CLEAN_MASK                                                                      0x00000002L
+#define UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK                                                                  0x00000004L
+#define UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK                                                             0x00000008L
+#define UVD_LMI_STATUS__UMC_READ_CLEAN_MASK                                                                   0x00000010L
+#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_MASK                                                                  0x00000020L
+#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK                                                              0x00000040L
+#define UVD_LMI_STATUS__PENDING_UVD_MC_WRITE_MASK                                                             0x00000080L
+#define UVD_LMI_STATUS__READ_CLEAN_RAW_MASK                                                                   0x00000100L
+#define UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK                                                               0x00000200L
+#define UVD_LMI_STATUS__UMC_UVD_IDLE_MASK                                                                     0x00000400L
+#define UVD_LMI_STATUS__UMC_AVP_IDLE_MASK                                                                     0x00000800L
+#define UVD_LMI_STATUS__ADP_MC_READ_CLEAN_MASK                                                                0x00001000L
+#define UVD_LMI_STATUS__ADP_UMC_READ_CLEAN_MASK                                                               0x00002000L
+#define UVD_LMI_STATUS__BSP0_WRITE_CLEAN_MASK                                                                 0x00040000L
+#define UVD_LMI_STATUS__BSP1_WRITE_CLEAN_MASK                                                                 0x00080000L
+#define UVD_LMI_STATUS__BSP2_WRITE_CLEAN_MASK                                                                 0x00100000L
+#define UVD_LMI_STATUS__BSP3_WRITE_CLEAN_MASK                                                                 0x00200000L
+#define UVD_LMI_STATUS__CENC_READ_CLEAN_MASK                                                                  0x00400000L
+//UVD_LMI_PERFMON_CTRL
+#define UVD_LMI_PERFMON_CTRL__PERFMON_STATE__SHIFT                                                            0x0
+#define UVD_LMI_PERFMON_CTRL__PERFMON_SEL__SHIFT                                                              0x8
+#define UVD_LMI_PERFMON_CTRL__PERFMON_STATE_MASK                                                              0x00000003L
+#define UVD_LMI_PERFMON_CTRL__PERFMON_SEL_MASK                                                                0x00001F00L
+//UVD_LMI_PERFMON_COUNT_LO
+#define UVD_LMI_PERFMON_COUNT_LO__PERFMON_COUNT__SHIFT                                                        0x0
+#define UVD_LMI_PERFMON_COUNT_LO__PERFMON_COUNT_MASK                                                          0xFFFFFFFFL
+//UVD_LMI_PERFMON_COUNT_HI
+#define UVD_LMI_PERFMON_COUNT_HI__PERFMON_COUNT__SHIFT                                                        0x0
+#define UVD_LMI_PERFMON_COUNT_HI__PERFMON_COUNT_MASK                                                          0x0000FFFFL
+//UVD_LMI_ADP_SWAP_CNTL
+#define UVD_LMI_ADP_SWAP_CNTL__VCPU_R_MC_SWAP__SHIFT                                                          0x6
+#define UVD_LMI_ADP_SWAP_CNTL__VCPU_W_MC_SWAP__SHIFT                                                          0x8
+#define UVD_LMI_ADP_SWAP_CNTL__CM_MC_SWAP__SHIFT                                                              0xa
+#define UVD_LMI_ADP_SWAP_CNTL__IT_MC_SWAP__SHIFT                                                              0xc
+#define UVD_LMI_ADP_SWAP_CNTL__DB_R_MC_SWAP__SHIFT                                                            0xe
+#define UVD_LMI_ADP_SWAP_CNTL__DB_W_MC_SWAP__SHIFT                                                            0x10
+#define UVD_LMI_ADP_SWAP_CNTL__CSM_MC_SWAP__SHIFT                                                             0x12
+#define UVD_LMI_ADP_SWAP_CNTL__PREF_MC_SWAP__SHIFT                                                            0x14
+#define UVD_LMI_ADP_SWAP_CNTL__DBW_MC_SWAP__SHIFT                                                             0x18
+#define UVD_LMI_ADP_SWAP_CNTL__RE_MC_SWAP__SHIFT                                                              0x1c
+#define UVD_LMI_ADP_SWAP_CNTL__MP_MC_SWAP__SHIFT                                                              0x1e
+#define UVD_LMI_ADP_SWAP_CNTL__VCPU_R_MC_SWAP_MASK                                                            0x000000C0L
+#define UVD_LMI_ADP_SWAP_CNTL__VCPU_W_MC_SWAP_MASK                                                            0x00000300L
+#define UVD_LMI_ADP_SWAP_CNTL__CM_MC_SWAP_MASK                                                                0x00000C00L
+#define UVD_LMI_ADP_SWAP_CNTL__IT_MC_SWAP_MASK                                                                0x00003000L
+#define UVD_LMI_ADP_SWAP_CNTL__DB_R_MC_SWAP_MASK                                                              0x0000C000L
+#define UVD_LMI_ADP_SWAP_CNTL__DB_W_MC_SWAP_MASK                                                              0x00030000L
+#define UVD_LMI_ADP_SWAP_CNTL__CSM_MC_SWAP_MASK                                                               0x000C0000L
+#define UVD_LMI_ADP_SWAP_CNTL__PREF_MC_SWAP_MASK                                                              0x00300000L
+#define UVD_LMI_ADP_SWAP_CNTL__DBW_MC_SWAP_MASK                                                               0x03000000L
+#define UVD_LMI_ADP_SWAP_CNTL__RE_MC_SWAP_MASK                                                                0x30000000L
+#define UVD_LMI_ADP_SWAP_CNTL__MP_MC_SWAP_MASK                                                                0xC0000000L
+//UVD_LMI_RBC_RB_VMID
+#define UVD_LMI_RBC_RB_VMID__RB_VMID__SHIFT                                                                   0x0
+#define UVD_LMI_RBC_RB_VMID__RB_VMID_MASK                                                                     0x0000000FL
+//UVD_LMI_RBC_IB_VMID
+#define UVD_LMI_RBC_IB_VMID__IB_VMID__SHIFT                                                                   0x0
+#define UVD_LMI_RBC_IB_VMID__IB_VMID_MASK                                                                     0x0000000FL
+//UVD_LMI_MC_CREDITS
+#define UVD_LMI_MC_CREDITS__UVD_RD_CREDITS__SHIFT                                                             0x0
+#define UVD_LMI_MC_CREDITS__UVD_WR_CREDITS__SHIFT                                                             0x8
+#define UVD_LMI_MC_CREDITS__UMC_RD_CREDITS__SHIFT                                                             0x10
+#define UVD_LMI_MC_CREDITS__UMC_WR_CREDITS__SHIFT                                                             0x18
+#define UVD_LMI_MC_CREDITS__UVD_RD_CREDITS_MASK                                                               0x0000003FL
+#define UVD_LMI_MC_CREDITS__UVD_WR_CREDITS_MASK                                                               0x00003F00L
+#define UVD_LMI_MC_CREDITS__UMC_RD_CREDITS_MASK                                                               0x003F0000L
+#define UVD_LMI_MC_CREDITS__UMC_WR_CREDITS_MASK                                                               0x3F000000L
+//UVD_LMI_ADP_IND_INDEX
+#define UVD_LMI_ADP_IND_INDEX__INDEX__SHIFT                                                                   0x0
+#define UVD_LMI_ADP_IND_INDEX__INDEX_MASK                                                                     0x00001FFFL
+//UVD_LMI_ADP_IND_DATA
+#define UVD_LMI_ADP_IND_DATA__DATA__SHIFT                                                                     0x0
+#define UVD_LMI_ADP_IND_DATA__DATA_MASK                                                                       0xFFFFFFFFL
+//UVD_LMI_ADP_PF_EN
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE0_PF_EN__SHIFT                                                           0x0
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE1_PF_EN__SHIFT                                                           0x1
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE2_PF_EN__SHIFT                                                           0x2
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE0_PF_EN_MASK                                                             0x00000001L
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE1_PF_EN_MASK                                                             0x00000002L
+#define UVD_LMI_ADP_PF_EN__VCPU_CACHE2_PF_EN_MASK                                                             0x00000004L
+//UVD_LMI_PREF_CTRL
+#define UVD_LMI_PREF_CTRL__PREF_RST__SHIFT                                                                    0x0
+#define UVD_LMI_PREF_CTRL__PREF_BUSY_STATUS__SHIFT                                                            0x1
+#define UVD_LMI_PREF_CTRL__PREF_WSTRB__SHIFT                                                                  0x2
+#define UVD_LMI_PREF_CTRL__PREF_WRITE_SIZE__SHIFT                                                             0x3
+#define UVD_LMI_PREF_CTRL__PREF_STEP_SIZE__SHIFT                                                              0x4
+#define UVD_LMI_PREF_CTRL__PREF_SIZE__SHIFT                                                                   0x13
+#define UVD_LMI_PREF_CTRL__PREF_RST_MASK                                                                      0x00000001L
+#define UVD_LMI_PREF_CTRL__PREF_BUSY_STATUS_MASK                                                              0x00000002L
+#define UVD_LMI_PREF_CTRL__PREF_WSTRB_MASK                                                                    0x00000004L
+#define UVD_LMI_PREF_CTRL__PREF_WRITE_SIZE_MASK                                                               0x00000008L
+#define UVD_LMI_PREF_CTRL__PREF_STEP_SIZE_MASK                                                                0x00000070L
+#define UVD_LMI_PREF_CTRL__PREF_SIZE_MASK                                                                     0xFFF80000L
+//UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW
+#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                  0x0
+#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK                                                    0xFFFFFFFFL
+//UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH
+#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                0x0
+#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK                                                  0xFFFFFFFFL
+//VCN_RAS_CNTL
+#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN__SHIFT                                                                0x0
+#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN__SHIFT                                                               0x4
+#define VCN_RAS_CNTL__VCPU_VCODEC_REARM__SHIFT                                                                0x8
+#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN__SHIFT                                                             0xc
+#define VCN_RAS_CNTL__VCPU_VCODEC_READY__SHIFT                                                                0x10
+#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK                                                                  0x00000001L
+#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK                                                                 0x00000010L
+#define VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK                                                                  0x00000100L
+#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK                                                               0x00001000L
+#define VCN_RAS_CNTL__VCPU_VCODEC_READY_MASK                                                                  0x00010000L
+
+
+// addressBlock: aid_uvd0_uvd_jpeg0_jpegnpdec
+//UVD_JPEG_CNTL
+#define UVD_JPEG_CNTL__REQUEST_EN__SHIFT                                                                      0x1
+#define UVD_JPEG_CNTL__ERR_RST_EN__SHIFT                                                                      0x2
+#define UVD_JPEG_CNTL__DBG_MUX_SEL__SHIFT                                                                     0x8
+#define UVD_JPEG_CNTL__FORMAT_CONV_EN__SHIFT                                                                  0x10
+#define UVD_JPEG_CNTL__VUP_MODE__SHIFT                                                                        0x11
+#define UVD_JPEG_CNTL__FC_TIMEOUT_EN__SHIFT                                                                   0x12
+#define UVD_JPEG_CNTL__ROI_CROP_EN__SHIFT                                                                     0x18
+#define UVD_JPEG_CNTL__ROI_CROP_EARLY_DECODE_STOP_DIS__SHIFT                                                  0x19
+#define UVD_JPEG_CNTL__REQUEST_EN_MASK                                                                        0x00000002L
+#define UVD_JPEG_CNTL__ERR_RST_EN_MASK                                                                        0x00000004L
+#define UVD_JPEG_CNTL__DBG_MUX_SEL_MASK                                                                       0x00007F00L
+#define UVD_JPEG_CNTL__FORMAT_CONV_EN_MASK                                                                    0x00010000L
+#define UVD_JPEG_CNTL__VUP_MODE_MASK                                                                          0x00020000L
+#define UVD_JPEG_CNTL__FC_TIMEOUT_EN_MASK                                                                     0x00040000L
+#define UVD_JPEG_CNTL__ROI_CROP_EN_MASK                                                                       0x01000000L
+#define UVD_JPEG_CNTL__ROI_CROP_EARLY_DECODE_STOP_DIS_MASK                                                    0x02000000L
+//UVD_JPEG_RB_BASE
+#define UVD_JPEG_RB_BASE__RB_BYTE_OFF__SHIFT                                                                  0x0
+#define UVD_JPEG_RB_BASE__RB_BASE__SHIFT                                                                      0x6
+#define UVD_JPEG_RB_BASE__RB_BYTE_OFF_MASK                                                                    0x0000003FL
+#define UVD_JPEG_RB_BASE__RB_BASE_MASK                                                                        0xFFFFFFC0L
+//UVD_JPEG_RB_WPTR
+#define UVD_JPEG_RB_WPTR__RB_WPTR__SHIFT                                                                      0x4
+#define UVD_JPEG_RB_WPTR__RB_WPTR_MASK                                                                        0x3FFFFFF0L
+//UVD_JPEG_RB_RPTR
+#define UVD_JPEG_RB_RPTR__RB_RPTR__SHIFT                                                                      0x4
+#define UVD_JPEG_RB_RPTR__RB_RPTR_MASK                                                                        0x3FFFFFF0L
+//UVD_JPEG_RB_SIZE
+#define UVD_JPEG_RB_SIZE__RB_SIZE__SHIFT                                                                      0x4
+#define UVD_JPEG_RB_SIZE__RB_SIZE_MASK                                                                        0x3FFFFFF0L
+//UVD_JPEG_DEC_CNT
+#define UVD_JPEG_DEC_CNT__DECODE_COUNT__SHIFT                                                                 0x0
+#define UVD_JPEG_DEC_CNT__DECODE_COUNT_MASK                                                                   0xFFFFFFFFL
+//UVD_JPEG_SPS_INFO
+#define UVD_JPEG_SPS_INFO__PIC_WIDTH__SHIFT                                                                   0x0
+#define UVD_JPEG_SPS_INFO__PIC_HEIGHT__SHIFT                                                                  0x10
+#define UVD_JPEG_SPS_INFO__PIC_WIDTH_MASK                                                                     0x0000FFFFL
+#define UVD_JPEG_SPS_INFO__PIC_HEIGHT_MASK                                                                    0xFFFF0000L
+//UVD_JPEG_SPS1_INFO
+#define UVD_JPEG_SPS1_INFO__CHROMA_FORMAT_IDC__SHIFT                                                          0x0
+#define UVD_JPEG_SPS1_INFO__YUV422_SUBFORMAT__SHIFT                                                           0x3
+#define UVD_JPEG_SPS1_INFO__OUT_FMT_422__SHIFT                                                                0x4
+#define UVD_JPEG_SPS1_INFO__CHROMA_FORMAT_IDC_MASK                                                            0x00000007L
+#define UVD_JPEG_SPS1_INFO__YUV422_SUBFORMAT_MASK                                                             0x00000008L
+#define UVD_JPEG_SPS1_INFO__OUT_FMT_422_MASK                                                                  0x00000010L
+//UVD_JPEG_RE_TIMER
+#define UVD_JPEG_RE_TIMER__TIMER_OUT__SHIFT                                                                   0x0
+#define UVD_JPEG_RE_TIMER__TIMER_OUT_EN__SHIFT                                                                0x10
+#define UVD_JPEG_RE_TIMER__TIMER_OUT_MASK                                                                     0x000000FFL
+#define UVD_JPEG_RE_TIMER__TIMER_OUT_EN_MASK                                                                  0x00010000L
+//UVD_JPEG_DEC_SCRATCH0
+#define UVD_JPEG_DEC_SCRATCH0__SCRATCH0__SHIFT                                                                0x0
+#define UVD_JPEG_DEC_SCRATCH0__SCRATCH0_MASK                                                                  0xFFFFFFFFL
+//UVD_JPEG_INT_EN
+#define UVD_JPEG_INT_EN__OUTBUF_WPTR_INC_EN__SHIFT                                                            0x0
+#define UVD_JPEG_INT_EN__JOB_AVAIL_EN__SHIFT                                                                  0x1
+#define UVD_JPEG_INT_EN__FENCE_VAL_EN__SHIFT                                                                  0x2
+#define UVD_JPEG_INT_EN__FIFO_OVERFLOW_ERR_EN__SHIFT                                                          0x6
+#define UVD_JPEG_INT_EN__BLK_CNT_OUT_OF_SYNC_ERR_EN__SHIFT                                                    0x7
+#define UVD_JPEG_INT_EN__EOI_ERR_EN__SHIFT                                                                    0x8
+#define UVD_JPEG_INT_EN__HFM_ERR_EN__SHIFT                                                                    0x9
+#define UVD_JPEG_INT_EN__RST_ERR_EN__SHIFT                                                                    0xa
+#define UVD_JPEG_INT_EN__ECS_MK_ERR_EN__SHIFT                                                                 0xb
+#define UVD_JPEG_INT_EN__TIMEOUT_ERR_EN__SHIFT                                                                0xc
+#define UVD_JPEG_INT_EN__MARKER_ERR_EN__SHIFT                                                                 0xd
+#define UVD_JPEG_INT_EN__FMT_ERR_EN__SHIFT                                                                    0xe
+#define UVD_JPEG_INT_EN__PROFILE_ERR_EN__SHIFT                                                                0xf
+#define UVD_JPEG_INT_EN__FC_TIMEOUT_ERR_EN__SHIFT                                                             0x10
+#define UVD_JPEG_INT_EN__FC_FMT_ERR_EN__SHIFT                                                                 0x11
+#define UVD_JPEG_INT_EN__FC_SRC_ERR_EN__SHIFT                                                                 0x12
+#define UVD_JPEG_INT_EN__CROP_SIZE_ERR_EN__SHIFT                                                              0x13
+#define UVD_JPEG_INT_EN__OUTBUF_WPTR_INC_EN_MASK                                                              0x00000001L
+#define UVD_JPEG_INT_EN__JOB_AVAIL_EN_MASK                                                                    0x00000002L
+#define UVD_JPEG_INT_EN__FENCE_VAL_EN_MASK                                                                    0x00000004L
+#define UVD_JPEG_INT_EN__FIFO_OVERFLOW_ERR_EN_MASK                                                            0x00000040L
+#define UVD_JPEG_INT_EN__BLK_CNT_OUT_OF_SYNC_ERR_EN_MASK                                                      0x00000080L
+#define UVD_JPEG_INT_EN__EOI_ERR_EN_MASK                                                                      0x00000100L
+#define UVD_JPEG_INT_EN__HFM_ERR_EN_MASK                                                                      0x00000200L
+#define UVD_JPEG_INT_EN__RST_ERR_EN_MASK                                                                      0x00000400L
+#define UVD_JPEG_INT_EN__ECS_MK_ERR_EN_MASK                                                                   0x00000800L
+#define UVD_JPEG_INT_EN__TIMEOUT_ERR_EN_MASK                                                                  0x00001000L
+#define UVD_JPEG_INT_EN__MARKER_ERR_EN_MASK                                                                   0x00002000L
+#define UVD_JPEG_INT_EN__FMT_ERR_EN_MASK                                                                      0x00004000L
+#define UVD_JPEG_INT_EN__PROFILE_ERR_EN_MASK                                                                  0x00008000L
+#define UVD_JPEG_INT_EN__FC_TIMEOUT_ERR_EN_MASK                                                               0x00010000L
+#define UVD_JPEG_INT_EN__FC_FMT_ERR_EN_MASK                                                                   0x00020000L
+#define UVD_JPEG_INT_EN__FC_SRC_ERR_EN_MASK                                                                   0x00040000L
+#define UVD_JPEG_INT_EN__CROP_SIZE_ERR_EN_MASK                                                                0x00080000L
+//UVD_JPEG_INT_STAT
+#define UVD_JPEG_INT_STAT__OUTBUF_WPTR_INC_INT__SHIFT                                                         0x0
+#define UVD_JPEG_INT_STAT__JOB_AVAIL_INT__SHIFT                                                               0x1
+#define UVD_JPEG_INT_STAT__FENCE_VAL_INT__SHIFT                                                               0x2
+#define UVD_JPEG_INT_STAT__FIFO_OVERFLOW_ERR_INT__SHIFT                                                       0x6
+#define UVD_JPEG_INT_STAT__BLK_CNT_OUT_OF_SYNC_ERR_INT__SHIFT                                                 0x7
+#define UVD_JPEG_INT_STAT__EOI_ERR_INT__SHIFT                                                                 0x8
+#define UVD_JPEG_INT_STAT__HFM_ERR_INT__SHIFT                                                                 0x9
+#define UVD_JPEG_INT_STAT__RST_ERR_INT__SHIFT                                                                 0xa
+#define UVD_JPEG_INT_STAT__ECS_MK_ERR_INT__SHIFT                                                              0xb
+#define UVD_JPEG_INT_STAT__TIMEOUT_ERR_INT__SHIFT                                                             0xc
+#define UVD_JPEG_INT_STAT__MARKER_ERR_INT__SHIFT                                                              0xd
+#define UVD_JPEG_INT_STAT__FMT_ERR_INT__SHIFT                                                                 0xe
+#define UVD_JPEG_INT_STAT__PROFILE_ERR_INT__SHIFT                                                             0xf
+#define UVD_JPEG_INT_STAT__FC_TIMEOUT_ERR_INT__SHIFT                                                          0x10
+#define UVD_JPEG_INT_STAT__FC_FMT_ERR_INT__SHIFT                                                              0x11
+#define UVD_JPEG_INT_STAT__FC_SRC_ERR_INT__SHIFT                                                              0x12
+#define UVD_JPEG_INT_STAT__CROP_SIZE_ERR_INT__SHIFT                                                           0x13
+#define UVD_JPEG_INT_STAT__OUTBUF_WPTR_INC_INT_MASK                                                           0x00000001L
+#define UVD_JPEG_INT_STAT__JOB_AVAIL_INT_MASK                                                                 0x00000002L
+#define UVD_JPEG_INT_STAT__FENCE_VAL_INT_MASK                                                                 0x00000004L
+#define UVD_JPEG_INT_STAT__FIFO_OVERFLOW_ERR_INT_MASK                                                         0x00000040L
+#define UVD_JPEG_INT_STAT__BLK_CNT_OUT_OF_SYNC_ERR_INT_MASK                                                   0x00000080L
+#define UVD_JPEG_INT_STAT__EOI_ERR_INT_MASK                                                                   0x00000100L
+#define UVD_JPEG_INT_STAT__HFM_ERR_INT_MASK                                                                   0x00000200L
+#define UVD_JPEG_INT_STAT__RST_ERR_INT_MASK                                                                   0x00000400L
+#define UVD_JPEG_INT_STAT__ECS_MK_ERR_INT_MASK                                                                0x00000800L
+#define UVD_JPEG_INT_STAT__TIMEOUT_ERR_INT_MASK                                                               0x00001000L
+#define UVD_JPEG_INT_STAT__MARKER_ERR_INT_MASK                                                                0x00002000L
+#define UVD_JPEG_INT_STAT__FMT_ERR_INT_MASK                                                                   0x00004000L
+#define UVD_JPEG_INT_STAT__PROFILE_ERR_INT_MASK                                                               0x00008000L
+#define UVD_JPEG_INT_STAT__FC_TIMEOUT_ERR_INT_MASK                                                            0x00010000L
+#define UVD_JPEG_INT_STAT__FC_FMT_ERR_INT_MASK                                                                0x00020000L
+#define UVD_JPEG_INT_STAT__FC_SRC_ERR_INT_MASK                                                                0x00040000L
+#define UVD_JPEG_INT_STAT__CROP_SIZE_ERR_INT_MASK                                                             0x00080000L
+//UVD_JPEG_TIER_CNTL0
+#define UVD_JPEG_TIER_CNTL0__TIER_SEL__SHIFT                                                                  0x0
+#define UVD_JPEG_TIER_CNTL0__Y_COMP_ID__SHIFT                                                                 0x2
+#define UVD_JPEG_TIER_CNTL0__U_COMP_ID__SHIFT                                                                 0x4
+#define UVD_JPEG_TIER_CNTL0__V_COMP_ID__SHIFT                                                                 0x6
+#define UVD_JPEG_TIER_CNTL0__Y_H_SAMP_FAC__SHIFT                                                              0x8
+#define UVD_JPEG_TIER_CNTL0__Y_V_SAMP_FAC__SHIFT                                                              0xb
+#define UVD_JPEG_TIER_CNTL0__U_H_SAMP_FAC__SHIFT                                                              0xe
+#define UVD_JPEG_TIER_CNTL0__U_V_SAMP_FAC__SHIFT                                                              0x11
+#define UVD_JPEG_TIER_CNTL0__V_H_SAMP_FAC__SHIFT                                                              0x14
+#define UVD_JPEG_TIER_CNTL0__V_V_SAMP_FAC__SHIFT                                                              0x17
+#define UVD_JPEG_TIER_CNTL0__Y_TQ__SHIFT                                                                      0x1a
+#define UVD_JPEG_TIER_CNTL0__U_TQ__SHIFT                                                                      0x1c
+#define UVD_JPEG_TIER_CNTL0__V_TQ__SHIFT                                                                      0x1e
+#define UVD_JPEG_TIER_CNTL0__TIER_SEL_MASK                                                                    0x00000003L
+#define UVD_JPEG_TIER_CNTL0__Y_COMP_ID_MASK                                                                   0x0000000CL
+#define UVD_JPEG_TIER_CNTL0__U_COMP_ID_MASK                                                                   0x00000030L
+#define UVD_JPEG_TIER_CNTL0__V_COMP_ID_MASK                                                                   0x000000C0L
+#define UVD_JPEG_TIER_CNTL0__Y_H_SAMP_FAC_MASK                                                                0x00000700L
+#define UVD_JPEG_TIER_CNTL0__Y_V_SAMP_FAC_MASK                                                                0x00003800L
+#define UVD_JPEG_TIER_CNTL0__U_H_SAMP_FAC_MASK                                                                0x0001C000L
+#define UVD_JPEG_TIER_CNTL0__U_V_SAMP_FAC_MASK                                                                0x000E0000L
+#define UVD_JPEG_TIER_CNTL0__V_H_SAMP_FAC_MASK                                                                0x00700000L
+#define UVD_JPEG_TIER_CNTL0__V_V_SAMP_FAC_MASK                                                                0x03800000L
+#define UVD_JPEG_TIER_CNTL0__Y_TQ_MASK                                                                        0x0C000000L
+#define UVD_JPEG_TIER_CNTL0__U_TQ_MASK                                                                        0x30000000L
+#define UVD_JPEG_TIER_CNTL0__V_TQ_MASK                                                                        0xC0000000L
+//UVD_JPEG_TIER_CNTL1
+#define UVD_JPEG_TIER_CNTL1__SRC_WIDTH__SHIFT                                                                 0x0
+#define UVD_JPEG_TIER_CNTL1__SRC_HEIGHT__SHIFT                                                                0x10
+#define UVD_JPEG_TIER_CNTL1__SRC_WIDTH_MASK                                                                   0x0000FFFFL
+#define UVD_JPEG_TIER_CNTL1__SRC_HEIGHT_MASK                                                                  0xFFFF0000L
+//UVD_JPEG_TIER_CNTL2
+#define UVD_JPEG_TIER_CNTL2__TBL_ECS_SEL__SHIFT                                                               0x0
+#define UVD_JPEG_TIER_CNTL2__TBL_TYPE__SHIFT                                                                  0x1
+#define UVD_JPEG_TIER_CNTL2__TQ__SHIFT                                                                        0x2
+#define UVD_JPEG_TIER_CNTL2__TH__SHIFT                                                                        0x4
+#define UVD_JPEG_TIER_CNTL2__TC__SHIFT                                                                        0x6
+#define UVD_JPEG_TIER_CNTL2__TD__SHIFT                                                                        0x7
+#define UVD_JPEG_TIER_CNTL2__TA__SHIFT                                                                        0xa
+#define UVD_JPEG_TIER_CNTL2__TIER2_HTBL_CNTLEN__SHIFT                                                         0xe
+#define UVD_JPEG_TIER_CNTL2__DRI_VAL__SHIFT                                                                   0x10
+#define UVD_JPEG_TIER_CNTL2__TBL_ECS_SEL_MASK                                                                 0x00000001L
+#define UVD_JPEG_TIER_CNTL2__TBL_TYPE_MASK                                                                    0x00000002L
+#define UVD_JPEG_TIER_CNTL2__TQ_MASK                                                                          0x0000000CL
+#define UVD_JPEG_TIER_CNTL2__TH_MASK                                                                          0x00000030L
+#define UVD_JPEG_TIER_CNTL2__TC_MASK                                                                          0x00000040L
+#define UVD_JPEG_TIER_CNTL2__TD_MASK                                                                          0x00000380L
+#define UVD_JPEG_TIER_CNTL2__TA_MASK                                                                          0x00001C00L
+#define UVD_JPEG_TIER_CNTL2__TIER2_HTBL_CNTLEN_MASK                                                           0x00004000L
+#define UVD_JPEG_TIER_CNTL2__DRI_VAL_MASK                                                                     0xFFFF0000L
+//UVD_JPEG_TIER_STATUS
+#define UVD_JPEG_TIER_STATUS__BSI_FETCH_DONE__SHIFT                                                           0x0
+#define UVD_JPEG_TIER_STATUS__DECODE_DONE__SHIFT                                                              0x1
+#define UVD_JPEG_TIER_STATUS__BSI_FETCH_DONE_MASK                                                             0x00000001L
+#define UVD_JPEG_TIER_STATUS__DECODE_DONE_MASK                                                                0x00000002L
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_sclk0_jpegnpsclkdec
+//UVD_JPEG_OUTBUF_CNTL
+#define UVD_JPEG_OUTBUF_CNTL__OUTBUF_CNT__SHIFT                                                               0x0
+#define UVD_JPEG_OUTBUF_CNTL__HGT_ALIGN__SHIFT                                                                0x2
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_DECODE_DONE_FIX__SHIFT                                                    0x6
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_MAX_CNT__SHIFT                                                    0x7
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_TIMER__SHIFT                                                      0x9
+#define UVD_JPEG_OUTBUF_CNTL__DIS_OBUF_AVAIL_CHECK__SHIFT                                                     0x10
+#define UVD_JPEG_OUTBUF_CNTL__OUTBUF_CNT_MASK                                                                 0x00000003L
+#define UVD_JPEG_OUTBUF_CNTL__HGT_ALIGN_MASK                                                                  0x00000004L
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_DECODE_DONE_FIX_MASK                                                      0x00000040L
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_MAX_CNT_MASK                                                      0x00000180L
+#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_TIMER_MASK                                                        0x00001E00L
+#define UVD_JPEG_OUTBUF_CNTL__DIS_OBUF_AVAIL_CHECK_MASK                                                       0x00010000L
+//UVD_JPEG_OUTBUF_WPTR
+#define UVD_JPEG_OUTBUF_WPTR__OUTBUF_WPTR__SHIFT                                                              0x0
+#define UVD_JPEG_OUTBUF_WPTR__OUTBUF_WPTR_MASK                                                                0xFFFFFFFFL
+//UVD_JPEG_OUTBUF_RPTR
+#define UVD_JPEG_OUTBUF_RPTR__OUTBUF_RPTR__SHIFT                                                              0x0
+#define UVD_JPEG_OUTBUF_RPTR__OUTBUF_RPTR_MASK                                                                0xFFFFFFFFL
+//UVD_JPEG_PITCH
+#define UVD_JPEG_PITCH__PITCH__SHIFT                                                                          0x0
+#define UVD_JPEG_PITCH__PITCH_MASK                                                                            0xFFFFFFFFL
+//UVD_JPEG_UV_PITCH
+#define UVD_JPEG_UV_PITCH__UV_PITCH__SHIFT                                                                    0x0
+#define UVD_JPEG_UV_PITCH__UV_PITCH_MASK                                                                      0xFFFFFFFFL
+//JPEG_DEC_Y_GFX8_TILING_SURFACE
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_WIDTH__SHIFT                                                     0x0
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_HEIGHT__SHIFT                                                    0x2
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT__SHIFT                                              0x4
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__NUM_BANKS__SHIFT                                                      0x6
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__PIPE_CONFIG__SHIFT                                                    0x8
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__TILE_SPLIT__SHIFT                                                     0xd
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__ARRAY_MODE__SHIFT                                                     0x10
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_WIDTH_MASK                                                       0x00000003L
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_HEIGHT_MASK                                                      0x0000000CL
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT_MASK                                                0x00000030L
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__NUM_BANKS_MASK                                                        0x000000C0L
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__PIPE_CONFIG_MASK                                                      0x00001F00L
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__TILE_SPLIT_MASK                                                       0x0000E000L
+#define JPEG_DEC_Y_GFX8_TILING_SURFACE__ARRAY_MODE_MASK                                                       0x000F0000L
+//JPEG_DEC_UV_GFX8_TILING_SURFACE
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_WIDTH__SHIFT                                                    0x0
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_HEIGHT__SHIFT                                                   0x2
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT__SHIFT                                             0x4
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__NUM_BANKS__SHIFT                                                     0x6
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__PIPE_CONFIG__SHIFT                                                   0x8
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__TILE_SPLIT__SHIFT                                                    0xd
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__ARRAY_MODE__SHIFT                                                    0x10
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_WIDTH_MASK                                                      0x00000003L
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_HEIGHT_MASK                                                     0x0000000CL
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT_MASK                                               0x00000030L
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__NUM_BANKS_MASK                                                       0x000000C0L
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__PIPE_CONFIG_MASK                                                     0x00001F00L
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__TILE_SPLIT_MASK                                                      0x0000E000L
+#define JPEG_DEC_UV_GFX8_TILING_SURFACE__ARRAY_MODE_MASK                                                      0x000F0000L
+//JPEG_DEC_GFX8_ADDR_CONFIG
+#define JPEG_DEC_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT                                                0x4
+#define JPEG_DEC_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK                                                  0x00000070L
+//JPEG_DEC_Y_GFX10_TILING_SURFACE
+#define JPEG_DEC_Y_GFX10_TILING_SURFACE__SWIZZLE_MODE__SHIFT                                                  0x0
+#define JPEG_DEC_Y_GFX10_TILING_SURFACE__SWIZZLE_MODE_MASK                                                    0x0000001FL
+//JPEG_DEC_UV_GFX10_TILING_SURFACE
+#define JPEG_DEC_UV_GFX10_TILING_SURFACE__SWIZZLE_MODE__SHIFT                                                 0x0
+#define JPEG_DEC_UV_GFX10_TILING_SURFACE__SWIZZLE_MODE_MASK                                                   0x0000001FL
+//JPEG_DEC_GFX10_ADDR_CONFIG
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PIPES__SHIFT                                                          0x0
+#define JPEG_DEC_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT                                               0x3
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PKRS__SHIFT                                                           0x8
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_BANKS__SHIFT                                                          0xc
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES__SHIFT                                                 0x13
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PIPES_MASK                                                            0x00000007L
+#define JPEG_DEC_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK                                                 0x00000038L
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PKRS_MASK                                                             0x00000700L
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_BANKS_MASK                                                            0x00007000L
+#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES_MASK                                                   0x00180000L
+//JPEG_DEC_ADDR_MODE
+#define JPEG_DEC_ADDR_MODE__ADDR_MODE_Y__SHIFT                                                                0x0
+#define JPEG_DEC_ADDR_MODE__ADDR_MODE_UV__SHIFT                                                               0x2
+#define JPEG_DEC_ADDR_MODE__ADDR_LIB_SEL__SHIFT                                                               0xc
+#define JPEG_DEC_ADDR_MODE__ADDR_MODE_Y_MASK                                                                  0x00000003L
+#define JPEG_DEC_ADDR_MODE__ADDR_MODE_UV_MASK                                                                 0x0000000CL
+#define JPEG_DEC_ADDR_MODE__ADDR_LIB_SEL_MASK                                                                 0x00007000L
+//UVD_JPEG_OUTPUT_XY
+#define UVD_JPEG_OUTPUT_XY__OUTPUT_X__SHIFT                                                                   0x0
+#define UVD_JPEG_OUTPUT_XY__OUTPUT_Y__SHIFT                                                                   0x10
+#define UVD_JPEG_OUTPUT_XY__OUTPUT_X_MASK                                                                     0x00003FFFL
+#define UVD_JPEG_OUTPUT_XY__OUTPUT_Y_MASK                                                                     0x3FFF0000L
+//UVD_JPEG_GPCOM_CMD
+#define UVD_JPEG_GPCOM_CMD__CMD__SHIFT                                                                        0x1
+#define UVD_JPEG_GPCOM_CMD__CMD_MASK                                                                          0x0000000EL
+//UVD_JPEG_GPCOM_DATA0
+#define UVD_JPEG_GPCOM_DATA0__DATA0__SHIFT                                                                    0x0
+#define UVD_JPEG_GPCOM_DATA0__DATA0_MASK                                                                      0xFFFFFFFFL
+//UVD_JPEG_GPCOM_DATA1
+#define UVD_JPEG_GPCOM_DATA1__DATA1__SHIFT                                                                    0x0
+#define UVD_JPEG_GPCOM_DATA1__DATA1_MASK                                                                      0xFFFFFFFFL
+//UVD_JPEG_SCRATCH1
+#define UVD_JPEG_SCRATCH1__SCRATCH1__SHIFT                                                                    0x0
+#define UVD_JPEG_SCRATCH1__SCRATCH1_MASK                                                                      0xFFFFFFFFL
+//UVD_JPEG_DEC_SOFT_RST
+#define UVD_JPEG_DEC_SOFT_RST__SOFT_RESET__SHIFT                                                              0x0
+#define UVD_JPEG_DEC_SOFT_RST__RESET_STATUS__SHIFT                                                            0x10
+#define UVD_JPEG_DEC_SOFT_RST__SOFT_RESET_MASK                                                                0x00000001L
+#define UVD_JPEG_DEC_SOFT_RST__RESET_STATUS_MASK                                                              0x00010000L
+
+
+// addressBlock: aid_uvd0_uvd_jrbc0_uvd_jrbc_dec
+//UVD_JRBC0_UVD_JRBC_RB_WPTR
+#define UVD_JRBC0_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC0_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC0_UVD_JRBC_RB_CNTL
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC0_UVD_JRBC_IB_SIZE
+#define UVD_JRBC0_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC0_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC0_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC0_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC0_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC0_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC0_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC0_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC0_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC0_UVD_JRBC_STATUS
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC0_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC0_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC0_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC0_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC0_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC0_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC0_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC0_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC0_UVD_JRBC_RB_RPTR
+#define UVD_JRBC0_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC0_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC0_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC0_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC0_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC0_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC0_UVD_JRBC_RB_SIZE
+#define UVD_JRBC0_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC0_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC0_UVD_JRBC_SCRATCH0
+#define UVD_JRBC0_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC0_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jmi0_uvd_jmi_dec
+//UVD_JMI0_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI0_UVD_LMI_JRBC_CTRL
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI0_UVD_LMI_JPEG_CTRL
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI0_JPEG_LMI_DROP
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI0_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI0_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI0_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI0_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI0_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI0_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI0_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI0_UVD_LMI_JPEG_VMID
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI0_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI0_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI0_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi_common_dec
+//UVD_JADP_MCIF_URGENT_CTRL
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_WATERMARK__SHIFT                                                        0x0
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_WATERMARK__SHIFT                                                        0x6
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_RD_URGENT_TIMER__SHIFT                                                  0xb
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_URGENT_PROG_STEP__SHIFT                                                 0x11
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_URGENT_PROG_STEP__SHIFT                                                 0x15
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_QOS_EN__SHIFT                                                           0x19
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_QOS_EN__SHIFT                                                           0x1a
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_WATERMARK_MASK                                                          0x0000003FL
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_WATERMARK_MASK                                                          0x000007C0L
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_RD_URGENT_TIMER_MASK                                                    0x0001F800L
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_URGENT_PROG_STEP_MASK                                                   0x001E0000L
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_URGENT_PROG_STEP_MASK                                                   0x01E00000L
+#define UVD_JADP_MCIF_URGENT_CTRL__WR_QOS_EN_MASK                                                             0x02000000L
+#define UVD_JADP_MCIF_URGENT_CTRL__RD_QOS_EN_MASK                                                             0x04000000L
+//UVD_JMI_URGENT_CTRL
+#define UVD_JMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL__SHIFT                                                 0x0
+#define UVD_JMI_URGENT_CTRL__ASSERT_MC_RD_URGENT__SHIFT                                                       0x4
+#define UVD_JMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL__SHIFT                                                 0x10
+#define UVD_JMI_URGENT_CTRL__ASSERT_MC_WR_URGENT__SHIFT                                                       0x14
+#define UVD_JMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL_MASK                                                   0x00000001L
+#define UVD_JMI_URGENT_CTRL__ASSERT_MC_RD_URGENT_MASK                                                         0x000000F0L
+#define UVD_JMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL_MASK                                                   0x00010000L
+#define UVD_JMI_URGENT_CTRL__ASSERT_MC_WR_URGENT_MASK                                                         0x00F00000L
+//UVD_JMI_CTRL
+#define UVD_JMI_CTRL__STALL_MC_ARB__SHIFT                                                                     0x0
+#define UVD_JMI_CTRL__MASK_MC_URGENT__SHIFT                                                                   0x1
+#define UVD_JMI_CTRL__ASSERT_MC_URGENT__SHIFT                                                                 0x2
+#define UVD_JMI_CTRL__MC_RD_ARB_WAIT_TIMER__SHIFT                                                             0x8
+#define UVD_JMI_CTRL__MC_WR_ARB_WAIT_TIMER__SHIFT                                                             0x10
+#define UVD_JMI_CTRL__STALL_MC_ARB_MASK                                                                       0x00000001L
+#define UVD_JMI_CTRL__MASK_MC_URGENT_MASK                                                                     0x00000002L
+#define UVD_JMI_CTRL__ASSERT_MC_URGENT_MASK                                                                   0x00000004L
+#define UVD_JMI_CTRL__MC_RD_ARB_WAIT_TIMER_MASK                                                               0x0000FF00L
+#define UVD_JMI_CTRL__MC_WR_ARB_WAIT_TIMER_MASK                                                               0x00FF0000L
+//JPEG_MEMCHECK_CLAMPING_CNTL
+#define JPEG_MEMCHECK_CLAMPING_CNTL__CLAMP_TO_SAFE_ADDR_EN__SHIFT                                             0x0
+#define JPEG_MEMCHECK_CLAMPING_CNTL__CLAMP_TO_SAFE_ADDR_EN_MASK                                               0x00000001L
+//JPEG_MEMCHECK_SAFE_ADDR
+#define JPEG_MEMCHECK_SAFE_ADDR__MEMCHECK_SAFE_ADDR__SHIFT                                                    0x0
+#define JPEG_MEMCHECK_SAFE_ADDR__MEMCHECK_SAFE_ADDR_MASK                                                      0xFFFFFFFFL
+//JPEG_MEMCHECK_SAFE_ADDR_64BIT
+#define JPEG_MEMCHECK_SAFE_ADDR_64BIT__MEMCHECK_SAFE_ADDR_64BIT__SHIFT                                        0x0
+#define JPEG_MEMCHECK_SAFE_ADDR_64BIT__MEMCHECK_SAFE_ADDR_64BIT_MASK                                          0xFFFFFFFFL
+//UVD_JMI_LAT_CTRL
+#define UVD_JMI_LAT_CTRL__SCALE__SHIFT                                                                        0x0
+#define UVD_JMI_LAT_CTRL__MAX_START__SHIFT                                                                    0x8
+#define UVD_JMI_LAT_CTRL__MIN_START__SHIFT                                                                    0x9
+#define UVD_JMI_LAT_CTRL__AVG_START__SHIFT                                                                    0xa
+#define UVD_JMI_LAT_CTRL__PERFMON_SYNC__SHIFT                                                                 0xb
+#define UVD_JMI_LAT_CTRL__SKIP__SHIFT                                                                         0x10
+#define UVD_JMI_LAT_CTRL__SCALE_MASK                                                                          0x000000FFL
+#define UVD_JMI_LAT_CTRL__MAX_START_MASK                                                                      0x00000100L
+#define UVD_JMI_LAT_CTRL__MIN_START_MASK                                                                      0x00000200L
+#define UVD_JMI_LAT_CTRL__AVG_START_MASK                                                                      0x00000400L
+#define UVD_JMI_LAT_CTRL__PERFMON_SYNC_MASK                                                                   0x00000800L
+#define UVD_JMI_LAT_CTRL__SKIP_MASK                                                                           0x000F0000L
+//UVD_JMI_LAT_CNTR
+#define UVD_JMI_LAT_CNTR__MAX_LAT__SHIFT                                                                      0x0
+#define UVD_JMI_LAT_CNTR__MIN_LAT__SHIFT                                                                      0x8
+#define UVD_JMI_LAT_CNTR__MAX_LAT_MASK                                                                        0x000000FFL
+#define UVD_JMI_LAT_CNTR__MIN_LAT_MASK                                                                        0x0000FF00L
+//UVD_JMI_AVG_LAT_CNTR
+#define UVD_JMI_AVG_LAT_CNTR__ENV_LOW__SHIFT                                                                  0x0
+#define UVD_JMI_AVG_LAT_CNTR__ENV_HIGH__SHIFT                                                                 0x8
+#define UVD_JMI_AVG_LAT_CNTR__ENV_HIT__SHIFT                                                                  0x10
+#define UVD_JMI_AVG_LAT_CNTR__ENV_LOW_MASK                                                                    0x000000FFL
+#define UVD_JMI_AVG_LAT_CNTR__ENV_HIGH_MASK                                                                   0x0000FF00L
+#define UVD_JMI_AVG_LAT_CNTR__ENV_HIT_MASK                                                                    0xFFFF0000L
+//UVD_JMI_PERFMON_CTRL
+#define UVD_JMI_PERFMON_CTRL__PERFMON_STATE__SHIFT                                                            0x0
+#define UVD_JMI_PERFMON_CTRL__PERFMON_SEL__SHIFT                                                              0x8
+#define UVD_JMI_PERFMON_CTRL__PERFMON_STATE_MASK                                                              0x00000003L
+#define UVD_JMI_PERFMON_CTRL__PERFMON_SEL_MASK                                                                0x00001F00L
+//UVD_JMI_PERFMON_COUNT_LO
+#define UVD_JMI_PERFMON_COUNT_LO__PERFMON_COUNT__SHIFT                                                        0x0
+#define UVD_JMI_PERFMON_COUNT_LO__PERFMON_COUNT_MASK                                                          0xFFFFFFFFL
+//UVD_JMI_PERFMON_COUNT_HI
+#define UVD_JMI_PERFMON_COUNT_HI__PERFMON_COUNT__SHIFT                                                        0x0
+#define UVD_JMI_PERFMON_COUNT_HI__PERFMON_COUNT_MASK                                                          0x0000FFFFL
+//UVD_JMI_CLEAN_STATUS
+#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN__SHIFT                                                           0x0
+#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_RAW__SHIFT                                                       0x1
+#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN__SHIFT                                                          0x2
+#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_RAW__SHIFT                                                      0x3
+#define UVD_JMI_CLEAN_STATUS__MC_WRITE_PENDING__SHIFT                                                         0x4
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_READ_CLEAN__SHIFT                                                   0x8
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_READ_CLEAN__SHIFT                                                   0x9
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_READ_CLEAN__SHIFT                                                   0xa
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_READ_CLEAN__SHIFT                                                   0xb
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_READ_CLEAN__SHIFT                                                   0xc
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_READ_CLEAN__SHIFT                                                   0xd
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_READ_CLEAN__SHIFT                                                   0xe
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_READ_CLEAN__SHIFT                                                   0xf
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_WRITE_CLEAN__SHIFT                                                  0x10
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_WRITE_CLEAN__SHIFT                                                  0x11
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_WRITE_CLEAN__SHIFT                                                  0x12
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_WRITE_CLEAN__SHIFT                                                  0x13
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_WRITE_CLEAN__SHIFT                                                  0x14
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_WRITE_CLEAN__SHIFT                                                  0x15
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_WRITE_CLEAN__SHIFT                                                  0x16
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_WRITE_CLEAN__SHIFT                                                  0x17
+#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_MASK                                                             0x00000001L
+#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_RAW_MASK                                                         0x00000002L
+#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_MASK                                                            0x00000004L
+#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_RAW_MASK                                                        0x00000008L
+#define UVD_JMI_CLEAN_STATUS__MC_WRITE_PENDING_MASK                                                           0x00000010L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_READ_CLEAN_MASK                                                     0x00000100L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_READ_CLEAN_MASK                                                     0x00000200L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_READ_CLEAN_MASK                                                     0x00000400L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_READ_CLEAN_MASK                                                     0x00000800L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_READ_CLEAN_MASK                                                     0x00001000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_READ_CLEAN_MASK                                                     0x00002000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_READ_CLEAN_MASK                                                     0x00004000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_READ_CLEAN_MASK                                                     0x00008000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_WRITE_CLEAN_MASK                                                    0x00010000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_WRITE_CLEAN_MASK                                                    0x00020000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_WRITE_CLEAN_MASK                                                    0x00040000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_WRITE_CLEAN_MASK                                                    0x00080000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_WRITE_CLEAN_MASK                                                    0x00100000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_WRITE_CLEAN_MASK                                                    0x00200000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_WRITE_CLEAN_MASK                                                    0x00400000L
+#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_WRITE_CLEAN_MASK                                                    0x00800000L
+//UVD_JMI_CNTL
+#define UVD_JMI_CNTL__SOFT_RESET__SHIFT                                                                       0x0
+#define UVD_JMI_CNTL__MC_RD_REQ_RET_MAX__SHIFT                                                                0x8
+#define UVD_JMI_CNTL__SOFT_RESET_MASK                                                                         0x00000001L
+#define UVD_JMI_CNTL__MC_RD_REQ_RET_MAX_MASK                                                                  0x0003FF00L
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_common_dec
+//JPEG_SOFT_RESET_STATUS
+#define JPEG_SOFT_RESET_STATUS__JPEG0_DEC_RESET_STATUS__SHIFT                                                 0x0
+#define JPEG_SOFT_RESET_STATUS__JPEG1_DEC_RESET_STATUS__SHIFT                                                 0x1
+#define JPEG_SOFT_RESET_STATUS__JPEG2_DEC_RESET_STATUS__SHIFT                                                 0x2
+#define JPEG_SOFT_RESET_STATUS__JPEG3_DEC_RESET_STATUS__SHIFT                                                 0x3
+#define JPEG_SOFT_RESET_STATUS__JPEG4_DEC_RESET_STATUS__SHIFT                                                 0x4
+#define JPEG_SOFT_RESET_STATUS__JPEG5_DEC_RESET_STATUS__SHIFT                                                 0x5
+#define JPEG_SOFT_RESET_STATUS__JPEG6_DEC_RESET_STATUS__SHIFT                                                 0x6
+#define JPEG_SOFT_RESET_STATUS__JPEG7_DEC_RESET_STATUS__SHIFT                                                 0x7
+#define JPEG_SOFT_RESET_STATUS__DJRBC0_RESET_STATUS__SHIFT                                                    0x8
+#define JPEG_SOFT_RESET_STATUS__DJRBC1_RESET_STATUS__SHIFT                                                    0x9
+#define JPEG_SOFT_RESET_STATUS__DJRBC2_RESET_STATUS__SHIFT                                                    0xa
+#define JPEG_SOFT_RESET_STATUS__DJRBC3_RESET_STATUS__SHIFT                                                    0xb
+#define JPEG_SOFT_RESET_STATUS__DJRBC4_RESET_STATUS__SHIFT                                                    0xc
+#define JPEG_SOFT_RESET_STATUS__DJRBC5_RESET_STATUS__SHIFT                                                    0xd
+#define JPEG_SOFT_RESET_STATUS__DJRBC6_RESET_STATUS__SHIFT                                                    0xe
+#define JPEG_SOFT_RESET_STATUS__DJRBC7_RESET_STATUS__SHIFT                                                    0xf
+#define JPEG_SOFT_RESET_STATUS__JPEG_ENC_RESET_STATUS__SHIFT                                                  0x11
+#define JPEG_SOFT_RESET_STATUS__EJRBC_RESET_STATUS__SHIFT                                                     0x12
+#define JPEG_SOFT_RESET_STATUS__JMCIF_RESET_STATUS__SHIFT                                                     0x18
+#define JPEG_SOFT_RESET_STATUS__JPEG0_DEC_RESET_STATUS_MASK                                                   0x00000001L
+#define JPEG_SOFT_RESET_STATUS__JPEG1_DEC_RESET_STATUS_MASK                                                   0x00000002L
+#define JPEG_SOFT_RESET_STATUS__JPEG2_DEC_RESET_STATUS_MASK                                                   0x00000004L
+#define JPEG_SOFT_RESET_STATUS__JPEG3_DEC_RESET_STATUS_MASK                                                   0x00000008L
+#define JPEG_SOFT_RESET_STATUS__JPEG4_DEC_RESET_STATUS_MASK                                                   0x00000010L
+#define JPEG_SOFT_RESET_STATUS__JPEG5_DEC_RESET_STATUS_MASK                                                   0x00000020L
+#define JPEG_SOFT_RESET_STATUS__JPEG6_DEC_RESET_STATUS_MASK                                                   0x00000040L
+#define JPEG_SOFT_RESET_STATUS__JPEG7_DEC_RESET_STATUS_MASK                                                   0x00000080L
+#define JPEG_SOFT_RESET_STATUS__DJRBC0_RESET_STATUS_MASK                                                      0x00000100L
+#define JPEG_SOFT_RESET_STATUS__DJRBC1_RESET_STATUS_MASK                                                      0x00000200L
+#define JPEG_SOFT_RESET_STATUS__DJRBC2_RESET_STATUS_MASK                                                      0x00000400L
+#define JPEG_SOFT_RESET_STATUS__DJRBC3_RESET_STATUS_MASK                                                      0x00000800L
+#define JPEG_SOFT_RESET_STATUS__DJRBC4_RESET_STATUS_MASK                                                      0x00001000L
+#define JPEG_SOFT_RESET_STATUS__DJRBC5_RESET_STATUS_MASK                                                      0x00002000L
+#define JPEG_SOFT_RESET_STATUS__DJRBC6_RESET_STATUS_MASK                                                      0x00004000L
+#define JPEG_SOFT_RESET_STATUS__DJRBC7_RESET_STATUS_MASK                                                      0x00008000L
+#define JPEG_SOFT_RESET_STATUS__JPEG_ENC_RESET_STATUS_MASK                                                    0x00020000L
+#define JPEG_SOFT_RESET_STATUS__EJRBC_RESET_STATUS_MASK                                                       0x00040000L
+#define JPEG_SOFT_RESET_STATUS__JMCIF_RESET_STATUS_MASK                                                       0x01000000L
+//JPEG_SYS_INT_EN
+#define JPEG_SYS_INT_EN__DJPEG0_CORE__SHIFT                                                                   0x0
+#define JPEG_SYS_INT_EN__DJPEG1_CORE__SHIFT                                                                   0x1
+#define JPEG_SYS_INT_EN__DJPEG2_CORE__SHIFT                                                                   0x2
+#define JPEG_SYS_INT_EN__DJPEG3_CORE__SHIFT                                                                   0x3
+#define JPEG_SYS_INT_EN__DJPEG4_CORE__SHIFT                                                                   0x4
+#define JPEG_SYS_INT_EN__DJPEG5_CORE__SHIFT                                                                   0x5
+#define JPEG_SYS_INT_EN__DJPEG6_CORE__SHIFT                                                                   0x6
+#define JPEG_SYS_INT_EN__DJPEG7_CORE__SHIFT                                                                   0x7
+#define JPEG_SYS_INT_EN__DJRBC0__SHIFT                                                                        0x8
+#define JPEG_SYS_INT_EN__DJRBC1__SHIFT                                                                        0x9
+#define JPEG_SYS_INT_EN__DJRBC2__SHIFT                                                                        0xa
+#define JPEG_SYS_INT_EN__DJRBC3__SHIFT                                                                        0xb
+#define JPEG_SYS_INT_EN__DJRBC4__SHIFT                                                                        0xc
+#define JPEG_SYS_INT_EN__DJRBC5__SHIFT                                                                        0xd
+#define JPEG_SYS_INT_EN__DJRBC6__SHIFT                                                                        0xe
+#define JPEG_SYS_INT_EN__DJRBC7__SHIFT                                                                        0xf
+#define JPEG_SYS_INT_EN__DJPEG0_PF_RPT__SHIFT                                                                 0x10
+#define JPEG_SYS_INT_EN__DJPEG1_PF_RPT__SHIFT                                                                 0x11
+#define JPEG_SYS_INT_EN__DJPEG2_PF_RPT__SHIFT                                                                 0x12
+#define JPEG_SYS_INT_EN__DJPEG3_PF_RPT__SHIFT                                                                 0x13
+#define JPEG_SYS_INT_EN__DJPEG4_PF_RPT__SHIFT                                                                 0x14
+#define JPEG_SYS_INT_EN__DJPEG5_PF_RPT__SHIFT                                                                 0x15
+#define JPEG_SYS_INT_EN__DJPEG6_PF_RPT__SHIFT                                                                 0x16
+#define JPEG_SYS_INT_EN__DJPEG7_PF_RPT__SHIFT                                                                 0x17
+#define JPEG_SYS_INT_EN__DJPEG0_RAS_CNTL__SHIFT                                                               0x18
+#define JPEG_SYS_INT_EN__DJPEG1_RAS_CNTL__SHIFT                                                               0x19
+#define JPEG_SYS_INT_EN__DJPEG0_CORE_MASK                                                                     0x00000001L
+#define JPEG_SYS_INT_EN__DJPEG1_CORE_MASK                                                                     0x00000002L
+#define JPEG_SYS_INT_EN__DJPEG2_CORE_MASK                                                                     0x00000004L
+#define JPEG_SYS_INT_EN__DJPEG3_CORE_MASK                                                                     0x00000008L
+#define JPEG_SYS_INT_EN__DJPEG4_CORE_MASK                                                                     0x00000010L
+#define JPEG_SYS_INT_EN__DJPEG5_CORE_MASK                                                                     0x00000020L
+#define JPEG_SYS_INT_EN__DJPEG6_CORE_MASK                                                                     0x00000040L
+#define JPEG_SYS_INT_EN__DJPEG7_CORE_MASK                                                                     0x00000080L
+#define JPEG_SYS_INT_EN__DJRBC0_MASK                                                                          0x00000100L
+#define JPEG_SYS_INT_EN__DJRBC1_MASK                                                                          0x00000200L
+#define JPEG_SYS_INT_EN__DJRBC2_MASK                                                                          0x00000400L
+#define JPEG_SYS_INT_EN__DJRBC3_MASK                                                                          0x00000800L
+#define JPEG_SYS_INT_EN__DJRBC4_MASK                                                                          0x00001000L
+#define JPEG_SYS_INT_EN__DJRBC5_MASK                                                                          0x00002000L
+#define JPEG_SYS_INT_EN__DJRBC6_MASK                                                                          0x00004000L
+#define JPEG_SYS_INT_EN__DJRBC7_MASK                                                                          0x00008000L
+#define JPEG_SYS_INT_EN__DJPEG0_PF_RPT_MASK                                                                   0x00010000L
+#define JPEG_SYS_INT_EN__DJPEG1_PF_RPT_MASK                                                                   0x00020000L
+#define JPEG_SYS_INT_EN__DJPEG2_PF_RPT_MASK                                                                   0x00040000L
+#define JPEG_SYS_INT_EN__DJPEG3_PF_RPT_MASK                                                                   0x00080000L
+#define JPEG_SYS_INT_EN__DJPEG4_PF_RPT_MASK                                                                   0x00100000L
+#define JPEG_SYS_INT_EN__DJPEG5_PF_RPT_MASK                                                                   0x00200000L
+#define JPEG_SYS_INT_EN__DJPEG6_PF_RPT_MASK                                                                   0x00400000L
+#define JPEG_SYS_INT_EN__DJPEG7_PF_RPT_MASK                                                                   0x00800000L
+#define JPEG_SYS_INT_EN__DJPEG0_RAS_CNTL_MASK                                                                 0x01000000L
+#define JPEG_SYS_INT_EN__DJPEG1_RAS_CNTL_MASK                                                                 0x02000000L
+//JPEG_SYS_INT_EN1
+#define JPEG_SYS_INT_EN1__EJPEG_PF_RPT__SHIFT                                                                 0x0
+#define JPEG_SYS_INT_EN1__EJPEG_CORE__SHIFT                                                                   0x1
+#define JPEG_SYS_INT_EN1__EJRBC__SHIFT                                                                        0x2
+#define JPEG_SYS_INT_EN1__EJPEG_RAS_CNTL__SHIFT                                                               0x3
+#define JPEG_SYS_INT_EN1__EJPEG_PF_RPT_MASK                                                                   0x00000001L
+#define JPEG_SYS_INT_EN1__EJPEG_CORE_MASK                                                                     0x00000002L
+#define JPEG_SYS_INT_EN1__EJRBC_MASK                                                                          0x00000004L
+#define JPEG_SYS_INT_EN1__EJPEG_RAS_CNTL_MASK                                                                 0x00000008L
+//JPEG_SYS_INT_STATUS
+#define JPEG_SYS_INT_STATUS__DJPEG0_CORE__SHIFT                                                               0x0
+#define JPEG_SYS_INT_STATUS__DJPEG1_CORE__SHIFT                                                               0x1
+#define JPEG_SYS_INT_STATUS__DJPEG2_CORE__SHIFT                                                               0x2
+#define JPEG_SYS_INT_STATUS__DJPEG3_CORE__SHIFT                                                               0x3
+#define JPEG_SYS_INT_STATUS__DJPEG4_CORE__SHIFT                                                               0x4
+#define JPEG_SYS_INT_STATUS__DJPEG5_CORE__SHIFT                                                               0x5
+#define JPEG_SYS_INT_STATUS__DJPEG6_CORE__SHIFT                                                               0x6
+#define JPEG_SYS_INT_STATUS__DJPEG7_CORE__SHIFT                                                               0x7
+#define JPEG_SYS_INT_STATUS__DJRBC0__SHIFT                                                                    0x8
+#define JPEG_SYS_INT_STATUS__DJRBC1__SHIFT                                                                    0x9
+#define JPEG_SYS_INT_STATUS__DJRBC2__SHIFT                                                                    0xa
+#define JPEG_SYS_INT_STATUS__DJRBC3__SHIFT                                                                    0xb
+#define JPEG_SYS_INT_STATUS__DJRBC4__SHIFT                                                                    0xc
+#define JPEG_SYS_INT_STATUS__DJRBC5__SHIFT                                                                    0xd
+#define JPEG_SYS_INT_STATUS__DJRBC6__SHIFT                                                                    0xe
+#define JPEG_SYS_INT_STATUS__DJRBC7__SHIFT                                                                    0xf
+#define JPEG_SYS_INT_STATUS__DJPEG0_PF_RPT__SHIFT                                                             0x10
+#define JPEG_SYS_INT_STATUS__DJPEG1_PF_RPT__SHIFT                                                             0x11
+#define JPEG_SYS_INT_STATUS__DJPEG2_PF_RPT__SHIFT                                                             0x12
+#define JPEG_SYS_INT_STATUS__DJPEG3_PF_RPT__SHIFT                                                             0x13
+#define JPEG_SYS_INT_STATUS__DJPEG4_PF_RPT__SHIFT                                                             0x14
+#define JPEG_SYS_INT_STATUS__DJPEG5_PF_RPT__SHIFT                                                             0x15
+#define JPEG_SYS_INT_STATUS__DJPEG6_PF_RPT__SHIFT                                                             0x16
+#define JPEG_SYS_INT_STATUS__DJPEG7_PF_RPT__SHIFT                                                             0x17
+#define JPEG_SYS_INT_STATUS__DJPEG0_RAS_CNTL__SHIFT                                                           0x18
+#define JPEG_SYS_INT_STATUS__DJPEG1_RAS_CNTL__SHIFT                                                           0x19
+#define JPEG_SYS_INT_STATUS__DJPEG0_CORE_MASK                                                                 0x00000001L
+#define JPEG_SYS_INT_STATUS__DJPEG1_CORE_MASK                                                                 0x00000002L
+#define JPEG_SYS_INT_STATUS__DJPEG2_CORE_MASK                                                                 0x00000004L
+#define JPEG_SYS_INT_STATUS__DJPEG3_CORE_MASK                                                                 0x00000008L
+#define JPEG_SYS_INT_STATUS__DJPEG4_CORE_MASK                                                                 0x00000010L
+#define JPEG_SYS_INT_STATUS__DJPEG5_CORE_MASK                                                                 0x00000020L
+#define JPEG_SYS_INT_STATUS__DJPEG6_CORE_MASK                                                                 0x00000040L
+#define JPEG_SYS_INT_STATUS__DJPEG7_CORE_MASK                                                                 0x00000080L
+#define JPEG_SYS_INT_STATUS__DJRBC0_MASK                                                                      0x00000100L
+#define JPEG_SYS_INT_STATUS__DJRBC1_MASK                                                                      0x00000200L
+#define JPEG_SYS_INT_STATUS__DJRBC2_MASK                                                                      0x00000400L
+#define JPEG_SYS_INT_STATUS__DJRBC3_MASK                                                                      0x00000800L
+#define JPEG_SYS_INT_STATUS__DJRBC4_MASK                                                                      0x00001000L
+#define JPEG_SYS_INT_STATUS__DJRBC5_MASK                                                                      0x00002000L
+#define JPEG_SYS_INT_STATUS__DJRBC6_MASK                                                                      0x00004000L
+#define JPEG_SYS_INT_STATUS__DJRBC7_MASK                                                                      0x00008000L
+#define JPEG_SYS_INT_STATUS__DJPEG0_PF_RPT_MASK                                                               0x00010000L
+#define JPEG_SYS_INT_STATUS__DJPEG1_PF_RPT_MASK                                                               0x00020000L
+#define JPEG_SYS_INT_STATUS__DJPEG2_PF_RPT_MASK                                                               0x00040000L
+#define JPEG_SYS_INT_STATUS__DJPEG3_PF_RPT_MASK                                                               0x00080000L
+#define JPEG_SYS_INT_STATUS__DJPEG4_PF_RPT_MASK                                                               0x00100000L
+#define JPEG_SYS_INT_STATUS__DJPEG5_PF_RPT_MASK                                                               0x00200000L
+#define JPEG_SYS_INT_STATUS__DJPEG6_PF_RPT_MASK                                                               0x00400000L
+#define JPEG_SYS_INT_STATUS__DJPEG7_PF_RPT_MASK                                                               0x00800000L
+#define JPEG_SYS_INT_STATUS__DJPEG0_RAS_CNTL_MASK                                                             0x01000000L
+#define JPEG_SYS_INT_STATUS__DJPEG1_RAS_CNTL_MASK                                                             0x02000000L
+//JPEG_SYS_INT_STATUS1
+#define JPEG_SYS_INT_STATUS1__EJPEG_PF_RPT__SHIFT                                                             0x0
+#define JPEG_SYS_INT_STATUS1__EJPEG_CORE__SHIFT                                                               0x1
+#define JPEG_SYS_INT_STATUS1__EJRBC__SHIFT                                                                    0x2
+#define JPEG_SYS_INT_STATUS1__EJPEG_RAS_CNTL__SHIFT                                                           0x3
+#define JPEG_SYS_INT_STATUS1__EJPEG_PF_RPT_MASK                                                               0x00000001L
+#define JPEG_SYS_INT_STATUS1__EJPEG_CORE_MASK                                                                 0x00000002L
+#define JPEG_SYS_INT_STATUS1__EJRBC_MASK                                                                      0x00000004L
+#define JPEG_SYS_INT_STATUS1__EJPEG_RAS_CNTL_MASK                                                             0x00000008L
+//JPEG_SYS_INT_ACK
+#define JPEG_SYS_INT_ACK__DJPEG0_CORE__SHIFT                                                                  0x0
+#define JPEG_SYS_INT_ACK__DJPEG1_CORE__SHIFT                                                                  0x1
+#define JPEG_SYS_INT_ACK__DJPEG2_CORE__SHIFT                                                                  0x2
+#define JPEG_SYS_INT_ACK__DJPEG3_CORE__SHIFT                                                                  0x3
+#define JPEG_SYS_INT_ACK__DJPEG4_CORE__SHIFT                                                                  0x4
+#define JPEG_SYS_INT_ACK__DJPEG5_CORE__SHIFT                                                                  0x5
+#define JPEG_SYS_INT_ACK__DJPEG6_CORE__SHIFT                                                                  0x6
+#define JPEG_SYS_INT_ACK__DJPEG7_CORE__SHIFT                                                                  0x7
+#define JPEG_SYS_INT_ACK__DJRBC0__SHIFT                                                                       0x8
+#define JPEG_SYS_INT_ACK__DJRBC1__SHIFT                                                                       0x9
+#define JPEG_SYS_INT_ACK__DJRBC2__SHIFT                                                                       0xa
+#define JPEG_SYS_INT_ACK__DJRBC3__SHIFT                                                                       0xb
+#define JPEG_SYS_INT_ACK__DJRBC4__SHIFT                                                                       0xc
+#define JPEG_SYS_INT_ACK__DJRBC5__SHIFT                                                                       0xd
+#define JPEG_SYS_INT_ACK__DJRBC6__SHIFT                                                                       0xe
+#define JPEG_SYS_INT_ACK__DJRBC7__SHIFT                                                                       0xf
+#define JPEG_SYS_INT_ACK__DJPEG0_PF_RPT__SHIFT                                                                0x10
+#define JPEG_SYS_INT_ACK__DJPEG1_PF_RPT__SHIFT                                                                0x11
+#define JPEG_SYS_INT_ACK__DJPEG2_PF_RPT__SHIFT                                                                0x12
+#define JPEG_SYS_INT_ACK__DJPEG3_PF_RPT__SHIFT                                                                0x13
+#define JPEG_SYS_INT_ACK__DJPEG4_PF_RPT__SHIFT                                                                0x14
+#define JPEG_SYS_INT_ACK__DJPEG5_PF_RPT__SHIFT                                                                0x15
+#define JPEG_SYS_INT_ACK__DJPEG6_PF_RPT__SHIFT                                                                0x16
+#define JPEG_SYS_INT_ACK__DJPEG7_PF_RPT__SHIFT                                                                0x17
+#define JPEG_SYS_INT_ACK__DJPEG0_RAS_CNTL__SHIFT                                                              0x18
+#define JPEG_SYS_INT_ACK__DJPEG1_RAS_CNTL__SHIFT                                                              0x19
+#define JPEG_SYS_INT_ACK__DJPEG0_CORE_MASK                                                                    0x00000001L
+#define JPEG_SYS_INT_ACK__DJPEG1_CORE_MASK                                                                    0x00000002L
+#define JPEG_SYS_INT_ACK__DJPEG2_CORE_MASK                                                                    0x00000004L
+#define JPEG_SYS_INT_ACK__DJPEG3_CORE_MASK                                                                    0x00000008L
+#define JPEG_SYS_INT_ACK__DJPEG4_CORE_MASK                                                                    0x00000010L
+#define JPEG_SYS_INT_ACK__DJPEG5_CORE_MASK                                                                    0x00000020L
+#define JPEG_SYS_INT_ACK__DJPEG6_CORE_MASK                                                                    0x00000040L
+#define JPEG_SYS_INT_ACK__DJPEG7_CORE_MASK                                                                    0x00000080L
+#define JPEG_SYS_INT_ACK__DJRBC0_MASK                                                                         0x00000100L
+#define JPEG_SYS_INT_ACK__DJRBC1_MASK                                                                         0x00000200L
+#define JPEG_SYS_INT_ACK__DJRBC2_MASK                                                                         0x00000400L
+#define JPEG_SYS_INT_ACK__DJRBC3_MASK                                                                         0x00000800L
+#define JPEG_SYS_INT_ACK__DJRBC4_MASK                                                                         0x00001000L
+#define JPEG_SYS_INT_ACK__DJRBC5_MASK                                                                         0x00002000L
+#define JPEG_SYS_INT_ACK__DJRBC6_MASK                                                                         0x00004000L
+#define JPEG_SYS_INT_ACK__DJRBC7_MASK                                                                         0x00008000L
+#define JPEG_SYS_INT_ACK__DJPEG0_PF_RPT_MASK                                                                  0x00010000L
+#define JPEG_SYS_INT_ACK__DJPEG1_PF_RPT_MASK                                                                  0x00020000L
+#define JPEG_SYS_INT_ACK__DJPEG2_PF_RPT_MASK                                                                  0x00040000L
+#define JPEG_SYS_INT_ACK__DJPEG3_PF_RPT_MASK                                                                  0x00080000L
+#define JPEG_SYS_INT_ACK__DJPEG4_PF_RPT_MASK                                                                  0x00100000L
+#define JPEG_SYS_INT_ACK__DJPEG5_PF_RPT_MASK                                                                  0x00200000L
+#define JPEG_SYS_INT_ACK__DJPEG6_PF_RPT_MASK                                                                  0x00400000L
+#define JPEG_SYS_INT_ACK__DJPEG7_PF_RPT_MASK                                                                  0x00800000L
+#define JPEG_SYS_INT_ACK__DJPEG0_RAS_CNTL_MASK                                                                0x01000000L
+#define JPEG_SYS_INT_ACK__DJPEG1_RAS_CNTL_MASK                                                                0x02000000L
+//JPEG_SYS_INT_ACK1
+#define JPEG_SYS_INT_ACK1__EJPEG_PF_RPT__SHIFT                                                                0x0
+#define JPEG_SYS_INT_ACK1__EJPEG_CORE__SHIFT                                                                  0x1
+#define JPEG_SYS_INT_ACK1__EJRBC__SHIFT                                                                       0x2
+#define JPEG_SYS_INT_ACK1__EJPEG_RAS_CNTL__SHIFT                                                              0x3
+#define JPEG_SYS_INT_ACK1__EJPEG_PF_RPT_MASK                                                                  0x00000001L
+#define JPEG_SYS_INT_ACK1__EJPEG_CORE_MASK                                                                    0x00000002L
+#define JPEG_SYS_INT_ACK1__EJRBC_MASK                                                                         0x00000004L
+#define JPEG_SYS_INT_ACK1__EJPEG_RAS_CNTL_MASK                                                                0x00000008L
+//JPEG_MEMCHECK_SYS_INT_EN
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_RD_ERR_EN__SHIFT                                                     0x0
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_RD_ERR_EN__SHIFT                                                     0x1
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_RD_ERR_EN__SHIFT                                                     0x2
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_RD_ERR_EN__SHIFT                                                     0x3
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_RD_ERR_EN__SHIFT                                                     0x4
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_RD_ERR_EN__SHIFT                                                     0x5
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_RD_ERR_EN__SHIFT                                                     0x6
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_RD_ERR_EN__SHIFT                                                     0x7
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH0_RD_ERR_EN__SHIFT                                                   0x8
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH1_RD_ERR_EN__SHIFT                                                   0x9
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH2_RD_ERR_EN__SHIFT                                                   0xa
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH3_RD_ERR_EN__SHIFT                                                   0xb
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH4_RD_ERR_EN__SHIFT                                                   0xc
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH5_RD_ERR_EN__SHIFT                                                   0xd
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH6_RD_ERR_EN__SHIFT                                                   0xe
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH7_RD_ERR_EN__SHIFT                                                   0xf
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_WR_ERR_EN__SHIFT                                                     0x10
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_WR_ERR_EN__SHIFT                                                     0x11
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_WR_ERR_EN__SHIFT                                                     0x12
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_WR_ERR_EN__SHIFT                                                     0x13
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_WR_ERR_EN__SHIFT                                                     0x14
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_WR_ERR_EN__SHIFT                                                     0x15
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_WR_ERR_EN__SHIFT                                                     0x16
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_WR_ERR_EN__SHIFT                                                     0x17
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF0_WR_ERR_EN__SHIFT                                                      0x18
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF1_WR_ERR_EN__SHIFT                                                      0x19
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF2_WR_ERR_EN__SHIFT                                                      0x1a
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF3_WR_ERR_EN__SHIFT                                                      0x1b
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF4_WR_ERR_EN__SHIFT                                                      0x1c
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF5_WR_ERR_EN__SHIFT                                                      0x1d
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF6_WR_ERR_EN__SHIFT                                                      0x1e
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF7_WR_ERR_EN__SHIFT                                                      0x1f
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_RD_ERR_EN_MASK                                                       0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_RD_ERR_EN_MASK                                                       0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_RD_ERR_EN_MASK                                                       0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_RD_ERR_EN_MASK                                                       0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_RD_ERR_EN_MASK                                                       0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_RD_ERR_EN_MASK                                                       0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_RD_ERR_EN_MASK                                                       0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_RD_ERR_EN_MASK                                                       0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH0_RD_ERR_EN_MASK                                                     0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH1_RD_ERR_EN_MASK                                                     0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH2_RD_ERR_EN_MASK                                                     0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH3_RD_ERR_EN_MASK                                                     0x00000800L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH4_RD_ERR_EN_MASK                                                     0x00001000L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH5_RD_ERR_EN_MASK                                                     0x00002000L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH6_RD_ERR_EN_MASK                                                     0x00004000L
+#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH7_RD_ERR_EN_MASK                                                     0x00008000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_WR_ERR_EN_MASK                                                       0x00010000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_WR_ERR_EN_MASK                                                       0x00020000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_WR_ERR_EN_MASK                                                       0x00040000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_WR_ERR_EN_MASK                                                       0x00080000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_WR_ERR_EN_MASK                                                       0x00100000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_WR_ERR_EN_MASK                                                       0x00200000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_WR_ERR_EN_MASK                                                       0x00400000L
+#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_WR_ERR_EN_MASK                                                       0x00800000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF0_WR_ERR_EN_MASK                                                        0x01000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF1_WR_ERR_EN_MASK                                                        0x02000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF2_WR_ERR_EN_MASK                                                        0x04000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF3_WR_ERR_EN_MASK                                                        0x08000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF4_WR_ERR_EN_MASK                                                        0x10000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF5_WR_ERR_EN_MASK                                                        0x20000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF6_WR_ERR_EN_MASK                                                        0x40000000L
+#define JPEG_MEMCHECK_SYS_INT_EN__OBUF7_WR_ERR_EN_MASK                                                        0x80000000L
+//JPEG_MEMCHECK_SYS_INT_EN1
+#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_RD_ERR_EN__SHIFT                                                     0x0
+#define JPEG_MEMCHECK_SYS_INT_EN1__PELFETCH_RD_ERR_EN__SHIFT                                                  0x1
+#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_RD_ERR_EN__SHIFT                                                    0x2
+#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_WR_ERR_EN__SHIFT                                                     0x3
+#define JPEG_MEMCHECK_SYS_INT_EN1__BS_WR_ERR_EN__SHIFT                                                        0x4
+#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_WR_ERR_EN__SHIFT                                                    0x5
+#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_RD_ERR_EN_MASK                                                       0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_EN1__PELFETCH_RD_ERR_EN_MASK                                                    0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_RD_ERR_EN_MASK                                                      0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_WR_ERR_EN_MASK                                                       0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_EN1__BS_WR_ERR_EN_MASK                                                          0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_WR_ERR_EN_MASK                                                      0x00000020L
+//JPEG_MEMCHECK_SYS_INT_STAT
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_HI_ERR__SHIFT                                                 0x0
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_HI_ERR__SHIFT                                                 0x1
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_HI_ERR__SHIFT                                                 0x2
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_HI_ERR__SHIFT                                                 0x3
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_HI_ERR__SHIFT                                                 0x4
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_HI_ERR__SHIFT                                                 0x5
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_HI_ERR__SHIFT                                                 0x6
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_HI_ERR__SHIFT                                                 0x7
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_LO_ERR__SHIFT                                                 0x8
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_LO_ERR__SHIFT                                                 0x9
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_LO_ERR__SHIFT                                                 0xa
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_LO_ERR__SHIFT                                                 0xb
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_LO_ERR__SHIFT                                                 0xc
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_LO_ERR__SHIFT                                                 0xd
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_LO_ERR__SHIFT                                                 0xe
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_LO_ERR__SHIFT                                                 0xf
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_HI_ERR__SHIFT                                                    0x10
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_HI_ERR__SHIFT                                                    0x11
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_HI_ERR__SHIFT                                                    0x12
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_HI_ERR__SHIFT                                                    0x13
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_HI_ERR__SHIFT                                                    0x14
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_HI_ERR__SHIFT                                                    0x15
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_HI_ERR__SHIFT                                                    0x16
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_HI_ERR__SHIFT                                                    0x17
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_LO_ERR__SHIFT                                                    0x18
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_LO_ERR__SHIFT                                                    0x19
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_LO_ERR__SHIFT                                                    0x1a
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_LO_ERR__SHIFT                                                    0x1b
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_LO_ERR__SHIFT                                                    0x1c
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_LO_ERR__SHIFT                                                    0x1d
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_LO_ERR__SHIFT                                                    0x1e
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_LO_ERR__SHIFT                                                    0x1f
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_HI_ERR_MASK                                                   0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_HI_ERR_MASK                                                   0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_HI_ERR_MASK                                                   0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_HI_ERR_MASK                                                   0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_HI_ERR_MASK                                                   0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_HI_ERR_MASK                                                   0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_HI_ERR_MASK                                                   0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_HI_ERR_MASK                                                   0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_LO_ERR_MASK                                                   0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_LO_ERR_MASK                                                   0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_LO_ERR_MASK                                                   0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_LO_ERR_MASK                                                   0x00000800L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_LO_ERR_MASK                                                   0x00001000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_LO_ERR_MASK                                                   0x00002000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_LO_ERR_MASK                                                   0x00004000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_LO_ERR_MASK                                                   0x00008000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_HI_ERR_MASK                                                      0x00010000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_HI_ERR_MASK                                                      0x00020000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_HI_ERR_MASK                                                      0x00040000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_HI_ERR_MASK                                                      0x00080000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_HI_ERR_MASK                                                      0x00100000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_HI_ERR_MASK                                                      0x00200000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_HI_ERR_MASK                                                      0x00400000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_HI_ERR_MASK                                                      0x00800000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_LO_ERR_MASK                                                      0x01000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_LO_ERR_MASK                                                      0x02000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_LO_ERR_MASK                                                      0x04000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_LO_ERR_MASK                                                      0x08000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_LO_ERR_MASK                                                      0x10000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_LO_ERR_MASK                                                      0x20000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_LO_ERR_MASK                                                      0x40000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_LO_ERR_MASK                                                      0x80000000L
+//JPEG_MEMCHECK_SYS_INT_STAT1
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_HI_ERR__SHIFT                                                  0x0
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_HI_ERR__SHIFT                                                  0x1
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_HI_ERR__SHIFT                                                  0x2
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_HI_ERR__SHIFT                                                  0x3
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_HI_ERR__SHIFT                                                  0x4
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_HI_ERR__SHIFT                                                  0x5
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_HI_ERR__SHIFT                                                  0x6
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_HI_ERR__SHIFT                                                  0x7
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_LO_ERR__SHIFT                                                  0x8
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_LO_ERR__SHIFT                                                  0x9
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_LO_ERR__SHIFT                                                  0xa
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_LO_ERR__SHIFT                                                  0xb
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_LO_ERR__SHIFT                                                  0xc
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_LO_ERR__SHIFT                                                  0xd
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_LO_ERR__SHIFT                                                  0xe
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_LO_ERR__SHIFT                                                  0xf
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_HI_ERR__SHIFT                                                  0x10
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_HI_ERR__SHIFT                                                  0x11
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_HI_ERR__SHIFT                                                  0x12
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_HI_ERR__SHIFT                                                  0x13
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_HI_ERR__SHIFT                                                  0x14
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_HI_ERR__SHIFT                                                  0x15
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_HI_ERR__SHIFT                                                  0x16
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_HI_ERR__SHIFT                                                  0x17
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_LO_ERR__SHIFT                                                  0x18
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_LO_ERR__SHIFT                                                  0x19
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_LO_ERR__SHIFT                                                  0x1a
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_LO_ERR__SHIFT                                                  0x1b
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_LO_ERR__SHIFT                                                  0x1c
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_LO_ERR__SHIFT                                                  0x1d
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_LO_ERR__SHIFT                                                  0x1e
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_LO_ERR__SHIFT                                                  0x1f
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_HI_ERR_MASK                                                    0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_HI_ERR_MASK                                                    0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_HI_ERR_MASK                                                    0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_HI_ERR_MASK                                                    0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_HI_ERR_MASK                                                    0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_HI_ERR_MASK                                                    0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_HI_ERR_MASK                                                    0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_HI_ERR_MASK                                                    0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_LO_ERR_MASK                                                    0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_LO_ERR_MASK                                                    0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_LO_ERR_MASK                                                    0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_LO_ERR_MASK                                                    0x00000800L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_LO_ERR_MASK                                                    0x00001000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_LO_ERR_MASK                                                    0x00002000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_LO_ERR_MASK                                                    0x00004000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_LO_ERR_MASK                                                    0x00008000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_HI_ERR_MASK                                                    0x00010000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_HI_ERR_MASK                                                    0x00020000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_HI_ERR_MASK                                                    0x00040000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_HI_ERR_MASK                                                    0x00080000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_HI_ERR_MASK                                                    0x00100000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_HI_ERR_MASK                                                    0x00200000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_HI_ERR_MASK                                                    0x00400000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_HI_ERR_MASK                                                    0x00800000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_LO_ERR_MASK                                                    0x01000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_LO_ERR_MASK                                                    0x02000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_LO_ERR_MASK                                                    0x04000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_LO_ERR_MASK                                                    0x08000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_LO_ERR_MASK                                                    0x10000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_LO_ERR_MASK                                                    0x20000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_LO_ERR_MASK                                                    0x40000000L
+#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_LO_ERR_MASK                                                    0x80000000L
+//JPEG_MEMCHECK_SYS_INT_STAT2
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_HI_ERR__SHIFT                                                   0x0
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_LO_ERR__SHIFT                                                   0x1
+#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_HI_ERR__SHIFT                                                0x2
+#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_LO_ERR__SHIFT                                                0x3
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_HI_ERR__SHIFT                                                  0x4
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_LO_ERR__SHIFT                                                  0x5
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_HI_ERR__SHIFT                                                   0x6
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_LO_ERR__SHIFT                                                   0x7
+#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_HI_ERR__SHIFT                                                      0x8
+#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_LO_ERR__SHIFT                                                      0x9
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_HI_ERR__SHIFT                                                  0xa
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_LO_ERR__SHIFT                                                  0xb
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_HI_ERR_MASK                                                     0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_LO_ERR_MASK                                                     0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_HI_ERR_MASK                                                  0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_LO_ERR_MASK                                                  0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_HI_ERR_MASK                                                    0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_LO_ERR_MASK                                                    0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_HI_ERR_MASK                                                     0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_LO_ERR_MASK                                                     0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_HI_ERR_MASK                                                        0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_LO_ERR_MASK                                                        0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_HI_ERR_MASK                                                    0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_LO_ERR_MASK                                                    0x00000800L
+//JPEG_MEMCHECK_SYS_INT_ACK
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_HI_ERR__SHIFT                                                  0x0
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_HI_ERR__SHIFT                                                  0x1
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_HI_ERR__SHIFT                                                  0x2
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_HI_ERR__SHIFT                                                  0x3
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_HI_ERR__SHIFT                                                  0x4
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_HI_ERR__SHIFT                                                  0x5
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_HI_ERR__SHIFT                                                  0x6
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_HI_ERR__SHIFT                                                  0x7
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_LO_ERR__SHIFT                                                  0x8
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_LO_ERR__SHIFT                                                  0x9
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_LO_ERR__SHIFT                                                  0xa
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_LO_ERR__SHIFT                                                  0xb
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_LO_ERR__SHIFT                                                  0xc
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_LO_ERR__SHIFT                                                  0xd
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_LO_ERR__SHIFT                                                  0xe
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_LO_ERR__SHIFT                                                  0xf
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_HI_ERR__SHIFT                                                     0x10
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_HI_ERR__SHIFT                                                     0x11
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_HI_ERR__SHIFT                                                     0x12
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_HI_ERR__SHIFT                                                     0x13
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_HI_ERR__SHIFT                                                     0x14
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_HI_ERR__SHIFT                                                     0x15
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_HI_ERR__SHIFT                                                     0x16
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_HI_ERR__SHIFT                                                     0x17
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_LO_ERR__SHIFT                                                     0x18
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_LO_ERR__SHIFT                                                     0x19
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_LO_ERR__SHIFT                                                     0x1a
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_LO_ERR__SHIFT                                                     0x1b
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_LO_ERR__SHIFT                                                     0x1c
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_LO_ERR__SHIFT                                                     0x1d
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_LO_ERR__SHIFT                                                     0x1e
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_LO_ERR__SHIFT                                                     0x1f
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_HI_ERR_MASK                                                    0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_HI_ERR_MASK                                                    0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_HI_ERR_MASK                                                    0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_HI_ERR_MASK                                                    0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_HI_ERR_MASK                                                    0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_HI_ERR_MASK                                                    0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_HI_ERR_MASK                                                    0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_HI_ERR_MASK                                                    0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_LO_ERR_MASK                                                    0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_LO_ERR_MASK                                                    0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_LO_ERR_MASK                                                    0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_LO_ERR_MASK                                                    0x00000800L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_LO_ERR_MASK                                                    0x00001000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_LO_ERR_MASK                                                    0x00002000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_LO_ERR_MASK                                                    0x00004000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_LO_ERR_MASK                                                    0x00008000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_HI_ERR_MASK                                                       0x00010000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_HI_ERR_MASK                                                       0x00020000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_HI_ERR_MASK                                                       0x00040000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_HI_ERR_MASK                                                       0x00080000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_HI_ERR_MASK                                                       0x00100000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_HI_ERR_MASK                                                       0x00200000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_HI_ERR_MASK                                                       0x00400000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_HI_ERR_MASK                                                       0x00800000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_LO_ERR_MASK                                                       0x01000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_LO_ERR_MASK                                                       0x02000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_LO_ERR_MASK                                                       0x04000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_LO_ERR_MASK                                                       0x08000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_LO_ERR_MASK                                                       0x10000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_LO_ERR_MASK                                                       0x20000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_LO_ERR_MASK                                                       0x40000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_LO_ERR_MASK                                                       0x80000000L
+//JPEG_MEMCHECK_SYS_INT_ACK1
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_HI_ERR__SHIFT                                                   0x0
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_HI_ERR__SHIFT                                                   0x1
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_HI_ERR__SHIFT                                                   0x2
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_HI_ERR__SHIFT                                                   0x3
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_HI_ERR__SHIFT                                                   0x4
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_HI_ERR__SHIFT                                                   0x5
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_HI_ERR__SHIFT                                                   0x6
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_HI_ERR__SHIFT                                                   0x7
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_LO_ERR__SHIFT                                                   0x8
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_LO_ERR__SHIFT                                                   0x9
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_LO_ERR__SHIFT                                                   0xa
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_LO_ERR__SHIFT                                                   0xb
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_LO_ERR__SHIFT                                                   0xc
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_LO_ERR__SHIFT                                                   0xd
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_LO_ERR__SHIFT                                                   0xe
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_LO_ERR__SHIFT                                                   0xf
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_HI_ERR__SHIFT                                                   0x10
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_HI_ERR__SHIFT                                                   0x11
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_HI_ERR__SHIFT                                                   0x12
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_HI_ERR__SHIFT                                                   0x13
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_HI_ERR__SHIFT                                                   0x14
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_HI_ERR__SHIFT                                                   0x15
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_HI_ERR__SHIFT                                                   0x16
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_HI_ERR__SHIFT                                                   0x17
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_LO_ERR__SHIFT                                                   0x18
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_LO_ERR__SHIFT                                                   0x19
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_LO_ERR__SHIFT                                                   0x1a
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_LO_ERR__SHIFT                                                   0x1b
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_LO_ERR__SHIFT                                                   0x1c
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_LO_ERR__SHIFT                                                   0x1d
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_LO_ERR__SHIFT                                                   0x1e
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_LO_ERR__SHIFT                                                   0x1f
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_HI_ERR_MASK                                                     0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_HI_ERR_MASK                                                     0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_HI_ERR_MASK                                                     0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_HI_ERR_MASK                                                     0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_HI_ERR_MASK                                                     0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_HI_ERR_MASK                                                     0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_HI_ERR_MASK                                                     0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_HI_ERR_MASK                                                     0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_LO_ERR_MASK                                                     0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_LO_ERR_MASK                                                     0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_LO_ERR_MASK                                                     0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_LO_ERR_MASK                                                     0x00000800L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_LO_ERR_MASK                                                     0x00001000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_LO_ERR_MASK                                                     0x00002000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_LO_ERR_MASK                                                     0x00004000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_LO_ERR_MASK                                                     0x00008000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_HI_ERR_MASK                                                     0x00010000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_HI_ERR_MASK                                                     0x00020000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_HI_ERR_MASK                                                     0x00040000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_HI_ERR_MASK                                                     0x00080000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_HI_ERR_MASK                                                     0x00100000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_HI_ERR_MASK                                                     0x00200000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_HI_ERR_MASK                                                     0x00400000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_HI_ERR_MASK                                                     0x00800000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_LO_ERR_MASK                                                     0x01000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_LO_ERR_MASK                                                     0x02000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_LO_ERR_MASK                                                     0x04000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_LO_ERR_MASK                                                     0x08000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_LO_ERR_MASK                                                     0x10000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_LO_ERR_MASK                                                     0x20000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_LO_ERR_MASK                                                     0x40000000L
+#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_LO_ERR_MASK                                                     0x80000000L
+//JPEG_MEMCHECK_SYS_INT_ACK2
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_HI_ERR__SHIFT                                                    0x0
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_LO_ERR__SHIFT                                                    0x1
+#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_HI_ERR__SHIFT                                                 0x2
+#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_LO_ERR__SHIFT                                                 0x3
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_HI_ERR__SHIFT                                                   0x4
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_LO_ERR__SHIFT                                                   0x5
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_HI_ERR__SHIFT                                                    0x6
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_LO_ERR__SHIFT                                                    0x7
+#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_HI_ERR__SHIFT                                                       0x8
+#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_LO_ERR__SHIFT                                                       0x9
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_HI_ERR__SHIFT                                                   0xa
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_LO_ERR__SHIFT                                                   0xb
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_HI_ERR_MASK                                                      0x00000001L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_LO_ERR_MASK                                                      0x00000002L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_HI_ERR_MASK                                                   0x00000004L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_LO_ERR_MASK                                                   0x00000008L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_HI_ERR_MASK                                                     0x00000010L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_LO_ERR_MASK                                                     0x00000020L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_HI_ERR_MASK                                                      0x00000040L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_LO_ERR_MASK                                                      0x00000080L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_HI_ERR_MASK                                                         0x00000100L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_LO_ERR_MASK                                                         0x00000200L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_HI_ERR_MASK                                                     0x00000400L
+#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_LO_ERR_MASK                                                     0x00000800L
+//JPEG_MASTINT_EN
+#define JPEG_MASTINT_EN__OVERRUN_RST__SHIFT                                                                   0x0
+#define JPEG_MASTINT_EN__INT_OVERRUN__SHIFT                                                                   0x4
+#define JPEG_MASTINT_EN__OVERRUN_RST_MASK                                                                     0x00000001L
+#define JPEG_MASTINT_EN__INT_OVERRUN_MASK                                                                     0x007FFFF0L
+//JPEG_IH_CTRL
+#define JPEG_IH_CTRL__IH_SOFT_RESET__SHIFT                                                                    0x0
+#define JPEG_IH_CTRL__IH_STALL_EN__SHIFT                                                                      0x1
+#define JPEG_IH_CTRL__IH_STATUS_CLEAN__SHIFT                                                                  0x2
+#define JPEG_IH_CTRL__IH_VMID__SHIFT                                                                          0x3
+#define JPEG_IH_CTRL__IH_USER_DATA__SHIFT                                                                     0x7
+#define JPEG_IH_CTRL__IH_RINGID__SHIFT                                                                        0x13
+#define JPEG_IH_CTRL__IH_SOFT_RESET_MASK                                                                      0x00000001L
+#define JPEG_IH_CTRL__IH_STALL_EN_MASK                                                                        0x00000002L
+#define JPEG_IH_CTRL__IH_STATUS_CLEAN_MASK                                                                    0x00000004L
+#define JPEG_IH_CTRL__IH_VMID_MASK                                                                            0x00000078L
+#define JPEG_IH_CTRL__IH_USER_DATA_MASK                                                                       0x0007FF80L
+#define JPEG_IH_CTRL__IH_RINGID_MASK                                                                          0x07F80000L
+//JRBBM_ARB_CTRL
+#define JRBBM_ARB_CTRL__DJRBC0_DROP__SHIFT                                                                    0x0
+#define JRBBM_ARB_CTRL__DJRBC1_DROP__SHIFT                                                                    0x1
+#define JRBBM_ARB_CTRL__DJRBC2_DROP__SHIFT                                                                    0x2
+#define JRBBM_ARB_CTRL__DJRBC3_DROP__SHIFT                                                                    0x3
+#define JRBBM_ARB_CTRL__DJRBC4_DROP__SHIFT                                                                    0x4
+#define JRBBM_ARB_CTRL__DJRBC5_DROP__SHIFT                                                                    0x5
+#define JRBBM_ARB_CTRL__DJRBC6_DROP__SHIFT                                                                    0x6
+#define JRBBM_ARB_CTRL__DJRBC7_DROP__SHIFT                                                                    0x7
+#define JRBBM_ARB_CTRL__EJRBC_DROP__SHIFT                                                                     0x8
+#define JRBBM_ARB_CTRL__SRBM_DROP__SHIFT                                                                      0x9
+#define JRBBM_ARB_CTRL__DJRBC0_DROP_MASK                                                                      0x00000001L
+#define JRBBM_ARB_CTRL__DJRBC1_DROP_MASK                                                                      0x00000002L
+#define JRBBM_ARB_CTRL__DJRBC2_DROP_MASK                                                                      0x00000004L
+#define JRBBM_ARB_CTRL__DJRBC3_DROP_MASK                                                                      0x00000008L
+#define JRBBM_ARB_CTRL__DJRBC4_DROP_MASK                                                                      0x00000010L
+#define JRBBM_ARB_CTRL__DJRBC5_DROP_MASK                                                                      0x00000020L
+#define JRBBM_ARB_CTRL__DJRBC6_DROP_MASK                                                                      0x00000040L
+#define JRBBM_ARB_CTRL__DJRBC7_DROP_MASK                                                                      0x00000080L
+#define JRBBM_ARB_CTRL__EJRBC_DROP_MASK                                                                       0x00000100L
+#define JRBBM_ARB_CTRL__SRBM_DROP_MASK                                                                        0x00000200L
+
+
+// addressBlock: aid_uvd0_uvd_jpeg_common_sclk_dec
+//JPEG_CGC_GATE
+#define JPEG_CGC_GATE__JPEG0_DEC__SHIFT                                                                       0x0
+#define JPEG_CGC_GATE__JPEG1_DEC__SHIFT                                                                       0x1
+#define JPEG_CGC_GATE__JPEG2_DEC__SHIFT                                                                       0x2
+#define JPEG_CGC_GATE__JPEG3_DEC__SHIFT                                                                       0x3
+#define JPEG_CGC_GATE__JPEG4_DEC__SHIFT                                                                       0x4
+#define JPEG_CGC_GATE__JPEG5_DEC__SHIFT                                                                       0x5
+#define JPEG_CGC_GATE__JPEG6_DEC__SHIFT                                                                       0x6
+#define JPEG_CGC_GATE__JPEG7_DEC__SHIFT                                                                       0x7
+#define JPEG_CGC_GATE__JPEG_ENC__SHIFT                                                                        0x8
+#define JPEG_CGC_GATE__JMCIF__SHIFT                                                                           0x9
+#define JPEG_CGC_GATE__JRBBM__SHIFT                                                                           0xa
+#define JPEG_CGC_GATE__JPEG0_DEC_MASK                                                                         0x00000001L
+#define JPEG_CGC_GATE__JPEG1_DEC_MASK                                                                         0x00000002L
+#define JPEG_CGC_GATE__JPEG2_DEC_MASK                                                                         0x00000004L
+#define JPEG_CGC_GATE__JPEG3_DEC_MASK                                                                         0x00000008L
+#define JPEG_CGC_GATE__JPEG4_DEC_MASK                                                                         0x00000010L
+#define JPEG_CGC_GATE__JPEG5_DEC_MASK                                                                         0x00000020L
+#define JPEG_CGC_GATE__JPEG6_DEC_MASK                                                                         0x00000040L
+#define JPEG_CGC_GATE__JPEG7_DEC_MASK                                                                         0x00000080L
+#define JPEG_CGC_GATE__JPEG_ENC_MASK                                                                          0x00000100L
+#define JPEG_CGC_GATE__JMCIF_MASK                                                                             0x00000200L
+#define JPEG_CGC_GATE__JRBBM_MASK                                                                             0x00000400L
+//JPEG_CGC_CTRL
+#define JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT                                                                  0x0
+#define JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT                                                              0x1
+#define JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT                                                                   0x5
+#define JPEG_CGC_CTRL__JPEG0_DEC_MODE__SHIFT                                                                  0x10
+#define JPEG_CGC_CTRL__JPEG1_DEC_MODE__SHIFT                                                                  0x11
+#define JPEG_CGC_CTRL__JPEG2_DEC_MODE__SHIFT                                                                  0x12
+#define JPEG_CGC_CTRL__JPEG3_DEC_MODE__SHIFT                                                                  0x13
+#define JPEG_CGC_CTRL__JPEG4_DEC_MODE__SHIFT                                                                  0x14
+#define JPEG_CGC_CTRL__JPEG5_DEC_MODE__SHIFT                                                                  0x15
+#define JPEG_CGC_CTRL__JPEG6_DEC_MODE__SHIFT                                                                  0x16
+#define JPEG_CGC_CTRL__JPEG7_DEC_MODE__SHIFT                                                                  0x17
+#define JPEG_CGC_CTRL__JPEG_ENC_MODE__SHIFT                                                                   0x18
+#define JPEG_CGC_CTRL__JMCIF_MODE__SHIFT                                                                      0x19
+#define JPEG_CGC_CTRL__JRBBM_MODE__SHIFT                                                                      0x1a
+#define JPEG_CGC_CTRL__DYN_CLOCK_MODE_MASK                                                                    0x00000001L
+#define JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK                                                                0x0000001EL
+#define JPEG_CGC_CTRL__CLK_OFF_DELAY_MASK                                                                     0x00001FE0L
+#define JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK                                                                    0x00010000L
+#define JPEG_CGC_CTRL__JPEG1_DEC_MODE_MASK                                                                    0x00020000L
+#define JPEG_CGC_CTRL__JPEG2_DEC_MODE_MASK                                                                    0x00040000L
+#define JPEG_CGC_CTRL__JPEG3_DEC_MODE_MASK                                                                    0x00080000L
+#define JPEG_CGC_CTRL__JPEG4_DEC_MODE_MASK                                                                    0x00100000L
+#define JPEG_CGC_CTRL__JPEG5_DEC_MODE_MASK                                                                    0x00200000L
+#define JPEG_CGC_CTRL__JPEG6_DEC_MODE_MASK                                                                    0x00400000L
+#define JPEG_CGC_CTRL__JPEG7_DEC_MODE_MASK                                                                    0x00800000L
+#define JPEG_CGC_CTRL__JPEG_ENC_MODE_MASK                                                                     0x01000000L
+#define JPEG_CGC_CTRL__JMCIF_MODE_MASK                                                                        0x02000000L
+#define JPEG_CGC_CTRL__JRBBM_MODE_MASK                                                                        0x04000000L
+//JPEG_CGC_STATUS
+#define JPEG_CGC_STATUS__JPEG0_DEC_VCLK_ACTIVE__SHIFT                                                         0x0
+#define JPEG_CGC_STATUS__JPEG0_DEC_SCLK_ACTIVE__SHIFT                                                         0x1
+#define JPEG_CGC_STATUS__JPEG1_DEC_VCLK_ACTIVE__SHIFT                                                         0x2
+#define JPEG_CGC_STATUS__JPEG1_DEC_SCLK_ACTIVE__SHIFT                                                         0x3
+#define JPEG_CGC_STATUS__JPEG2_DEC_VCLK_ACTIVE__SHIFT                                                         0x4
+#define JPEG_CGC_STATUS__JPEG2_DEC_SCLK_ACTIVE__SHIFT                                                         0x5
+#define JPEG_CGC_STATUS__JPEG3_DEC_VCLK_ACTIVE__SHIFT                                                         0x6
+#define JPEG_CGC_STATUS__JPEG3_DEC_SCLK_ACTIVE__SHIFT                                                         0x7
+#define JPEG_CGC_STATUS__JPEG4_DEC_VCLK_ACTIVE__SHIFT                                                         0x8
+#define JPEG_CGC_STATUS__JPEG4_DEC_SCLK_ACTIVE__SHIFT                                                         0x9
+#define JPEG_CGC_STATUS__JPEG5_DEC_VCLK_ACTIVE__SHIFT                                                         0xa
+#define JPEG_CGC_STATUS__JPEG5_DEC_SCLK_ACTIVE__SHIFT                                                         0xb
+#define JPEG_CGC_STATUS__JPEG6_DEC_VCLK_ACTIVE__SHIFT                                                         0xc
+#define JPEG_CGC_STATUS__JPEG6_DEC_SCLK_ACTIVE__SHIFT                                                         0xd
+#define JPEG_CGC_STATUS__JPEG7_DEC_VCLK_ACTIVE__SHIFT                                                         0xe
+#define JPEG_CGC_STATUS__JPEG7_DEC_SCLK_ACTIVE__SHIFT                                                         0xf
+#define JPEG_CGC_STATUS__JPEG_ENC_VCLK_ACTIVE__SHIFT                                                          0x10
+#define JPEG_CGC_STATUS__JPEG_ENC_SCLK_ACTIVE__SHIFT                                                          0x11
+#define JPEG_CGC_STATUS__JMCIF_SCLK_ACTIVE__SHIFT                                                             0x12
+#define JPEG_CGC_STATUS__JRBBM_VCLK_ACTIVE__SHIFT                                                             0x13
+#define JPEG_CGC_STATUS__JRBBM_SCLK_ACTIVE__SHIFT                                                             0x14
+#define JPEG_CGC_STATUS__JPEG0_DEC_VCLK_ACTIVE_MASK                                                           0x00000001L
+#define JPEG_CGC_STATUS__JPEG0_DEC_SCLK_ACTIVE_MASK                                                           0x00000002L
+#define JPEG_CGC_STATUS__JPEG1_DEC_VCLK_ACTIVE_MASK                                                           0x00000004L
+#define JPEG_CGC_STATUS__JPEG1_DEC_SCLK_ACTIVE_MASK                                                           0x00000008L
+#define JPEG_CGC_STATUS__JPEG2_DEC_VCLK_ACTIVE_MASK                                                           0x00000010L
+#define JPEG_CGC_STATUS__JPEG2_DEC_SCLK_ACTIVE_MASK                                                           0x00000020L
+#define JPEG_CGC_STATUS__JPEG3_DEC_VCLK_ACTIVE_MASK                                                           0x00000040L
+#define JPEG_CGC_STATUS__JPEG3_DEC_SCLK_ACTIVE_MASK                                                           0x00000080L
+#define JPEG_CGC_STATUS__JPEG4_DEC_VCLK_ACTIVE_MASK                                                           0x00000100L
+#define JPEG_CGC_STATUS__JPEG4_DEC_SCLK_ACTIVE_MASK                                                           0x00000200L
+#define JPEG_CGC_STATUS__JPEG5_DEC_VCLK_ACTIVE_MASK                                                           0x00000400L
+#define JPEG_CGC_STATUS__JPEG5_DEC_SCLK_ACTIVE_MASK                                                           0x00000800L
+#define JPEG_CGC_STATUS__JPEG6_DEC_VCLK_ACTIVE_MASK                                                           0x00001000L
+#define JPEG_CGC_STATUS__JPEG6_DEC_SCLK_ACTIVE_MASK                                                           0x00002000L
+#define JPEG_CGC_STATUS__JPEG7_DEC_VCLK_ACTIVE_MASK                                                           0x00004000L
+#define JPEG_CGC_STATUS__JPEG7_DEC_SCLK_ACTIVE_MASK                                                           0x00008000L
+#define JPEG_CGC_STATUS__JPEG_ENC_VCLK_ACTIVE_MASK                                                            0x00010000L
+#define JPEG_CGC_STATUS__JPEG_ENC_SCLK_ACTIVE_MASK                                                            0x00020000L
+#define JPEG_CGC_STATUS__JMCIF_SCLK_ACTIVE_MASK                                                               0x00040000L
+#define JPEG_CGC_STATUS__JRBBM_VCLK_ACTIVE_MASK                                                               0x00080000L
+#define JPEG_CGC_STATUS__JRBBM_SCLK_ACTIVE_MASK                                                               0x00100000L
+//JPEG_COMN_CGC_MEM_CTRL
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_EN__SHIFT                                                            0x0
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_DS_EN__SHIFT                                                            0x1
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_SD_EN__SHIFT                                                            0x2
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_SW_EN__SHIFT                                                         0x3
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_EN_MASK                                                              0x00000001L
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_DS_EN_MASK                                                              0x00000002L
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_SD_EN_MASK                                                              0x00000004L
+#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_SW_EN_MASK                                                           0x00000008L
+//JPEG_DEC_CGC_MEM_CTRL
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_EN__SHIFT                                                         0x0
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_DS_EN__SHIFT                                                         0x1
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_SD_EN__SHIFT                                                         0x2
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_SW_EN__SHIFT                                                      0x3
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_EN__SHIFT                                                         0x4
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_DS_EN__SHIFT                                                         0x5
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_SD_EN__SHIFT                                                         0x6
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_SW_EN__SHIFT                                                      0x7
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_EN__SHIFT                                                         0x8
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_DS_EN__SHIFT                                                         0x9
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_SD_EN__SHIFT                                                         0xa
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_SW_EN__SHIFT                                                      0xb
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_EN__SHIFT                                                         0xc
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_DS_EN__SHIFT                                                         0xd
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_SD_EN__SHIFT                                                         0xe
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_SW_EN__SHIFT                                                      0xf
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_EN__SHIFT                                                         0x10
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_DS_EN__SHIFT                                                         0x11
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_SD_EN__SHIFT                                                         0x12
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_SW_EN__SHIFT                                                      0x13
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_EN__SHIFT                                                         0x14
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_DS_EN__SHIFT                                                         0x15
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_SD_EN__SHIFT                                                         0x16
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_SW_EN__SHIFT                                                      0x17
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_EN__SHIFT                                                         0x18
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_DS_EN__SHIFT                                                         0x19
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_SD_EN__SHIFT                                                         0x1a
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_SW_EN__SHIFT                                                      0x1b
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_EN__SHIFT                                                         0x1c
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_DS_EN__SHIFT                                                         0x1d
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_SD_EN__SHIFT                                                         0x1e
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_SW_EN__SHIFT                                                      0x1f
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_EN_MASK                                                           0x00000001L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_DS_EN_MASK                                                           0x00000002L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_SD_EN_MASK                                                           0x00000004L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_SW_EN_MASK                                                        0x00000008L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_EN_MASK                                                           0x00000010L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_DS_EN_MASK                                                           0x00000020L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_SD_EN_MASK                                                           0x00000040L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_SW_EN_MASK                                                        0x00000080L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_EN_MASK                                                           0x00000100L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_DS_EN_MASK                                                           0x00000200L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_SD_EN_MASK                                                           0x00000400L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_SW_EN_MASK                                                        0x00000800L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_EN_MASK                                                           0x00001000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_DS_EN_MASK                                                           0x00002000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_SD_EN_MASK                                                           0x00004000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_SW_EN_MASK                                                        0x00008000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_EN_MASK                                                           0x00010000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_DS_EN_MASK                                                           0x00020000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_SD_EN_MASK                                                           0x00040000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_SW_EN_MASK                                                        0x00080000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_EN_MASK                                                           0x00100000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_DS_EN_MASK                                                           0x00200000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_SD_EN_MASK                                                           0x00400000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_SW_EN_MASK                                                        0x00800000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_EN_MASK                                                           0x01000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_DS_EN_MASK                                                           0x02000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_SD_EN_MASK                                                           0x04000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_SW_EN_MASK                                                        0x08000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_EN_MASK                                                           0x10000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_DS_EN_MASK                                                           0x20000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_SD_EN_MASK                                                           0x40000000L
+#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_SW_EN_MASK                                                        0x80000000L
+//JPEG_ENC_CGC_MEM_CTRL
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_EN__SHIFT                                                          0x0
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_DS_EN__SHIFT                                                          0x1
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_SD_EN__SHIFT                                                          0x2
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_SW_EN__SHIFT                                                       0x3
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_EN_MASK                                                            0x00000001L
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_DS_EN_MASK                                                            0x00000002L
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_SD_EN_MASK                                                            0x00000004L
+#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_SW_EN_MASK                                                         0x00000008L
+//JPEG_PERF_BANK_CONF
+#define JPEG_PERF_BANK_CONF__RESET__SHIFT                                                                     0x0
+#define JPEG_PERF_BANK_CONF__PEEK__SHIFT                                                                      0x8
+#define JPEG_PERF_BANK_CONF__CONCATENATE__SHIFT                                                               0x10
+#define JPEG_PERF_BANK_CONF__CORE_SEL__SHIFT                                                                  0x15
+#define JPEG_PERF_BANK_CONF__RESET_MASK                                                                       0x0000000FL
+#define JPEG_PERF_BANK_CONF__PEEK_MASK                                                                        0x00000F00L
+#define JPEG_PERF_BANK_CONF__CONCATENATE_MASK                                                                 0x00030000L
+#define JPEG_PERF_BANK_CONF__CORE_SEL_MASK                                                                    0x00E00000L
+//JPEG_PERF_BANK_EVENT_SEL
+#define JPEG_PERF_BANK_EVENT_SEL__SEL0__SHIFT                                                                 0x0
+#define JPEG_PERF_BANK_EVENT_SEL__SEL1__SHIFT                                                                 0x8
+#define JPEG_PERF_BANK_EVENT_SEL__SEL2__SHIFT                                                                 0x10
+#define JPEG_PERF_BANK_EVENT_SEL__SEL3__SHIFT                                                                 0x18
+#define JPEG_PERF_BANK_EVENT_SEL__SEL0_MASK                                                                   0x000000FFL
+#define JPEG_PERF_BANK_EVENT_SEL__SEL1_MASK                                                                   0x0000FF00L
+#define JPEG_PERF_BANK_EVENT_SEL__SEL2_MASK                                                                   0x00FF0000L
+#define JPEG_PERF_BANK_EVENT_SEL__SEL3_MASK                                                                   0xFF000000L
+//JPEG_PERF_BANK_COUNT0
+#define JPEG_PERF_BANK_COUNT0__COUNT__SHIFT                                                                   0x0
+#define JPEG_PERF_BANK_COUNT0__COUNT_MASK                                                                     0xFFFFFFFFL
+//JPEG_PERF_BANK_COUNT1
+#define JPEG_PERF_BANK_COUNT1__COUNT__SHIFT                                                                   0x0
+#define JPEG_PERF_BANK_COUNT1__COUNT_MASK                                                                     0xFFFFFFFFL
+//JPEG_PERF_BANK_COUNT2
+#define JPEG_PERF_BANK_COUNT2__COUNT__SHIFT                                                                   0x0
+#define JPEG_PERF_BANK_COUNT2__COUNT_MASK                                                                     0xFFFFFFFFL
+//JPEG_PERF_BANK_COUNT3
+#define JPEG_PERF_BANK_COUNT3__COUNT__SHIFT                                                                   0x0
+#define JPEG_PERF_BANK_COUNT3__COUNT_MASK                                                                     0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_pg_dec
+//UVD_PGFSM_CONFIG
+#define UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT                                                              0x0
+#define UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT                                                              0x2
+#define UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT                                                              0x4
+#define UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT                                                             0x6
+#define UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT                                                              0x8
+#define UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT                                                             0xa
+#define UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT                                                             0xc
+#define UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT                                                             0xe
+#define UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT                                                             0x10
+#define UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT                                                              0x12
+#define UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT                                                             0x14
+#define UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT                                                              0x16
+#define UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT                                                             0x18
+#define UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT                                                             0x1a
+#define UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT                                                             0x1c
+#define UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG_MASK                                                                0x00000003L
+#define UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG_MASK                                                                0x0000000CL
+#define UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG_MASK                                                                0x00000030L
+#define UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG_MASK                                                               0x000000C0L
+#define UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG_MASK                                                                0x00000300L
+#define UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG_MASK                                                               0x00000C00L
+#define UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG_MASK                                                               0x00003000L
+#define UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG_MASK                                                               0x0000C000L
+#define UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG_MASK                                                               0x00030000L
+#define UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG_MASK                                                                0x000C0000L
+#define UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG_MASK                                                               0x00300000L
+#define UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG_MASK                                                                0x00C00000L
+#define UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG_MASK                                                               0x03000000L
+#define UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG_MASK                                                               0x0C000000L
+#define UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG_MASK                                                               0x30000000L
+//UVD_PGFSM_STATUS
+#define UVD_PGFSM_STATUS__UVDM_PWR_STATUS__SHIFT                                                              0x0
+#define UVD_PGFSM_STATUS__UVDS_PWR_STATUS__SHIFT                                                              0x2
+#define UVD_PGFSM_STATUS__UVDF_PWR_STATUS__SHIFT                                                              0x4
+#define UVD_PGFSM_STATUS__UVDTC_PWR_STATUS__SHIFT                                                             0x6
+#define UVD_PGFSM_STATUS__UVDB_PWR_STATUS__SHIFT                                                              0x8
+#define UVD_PGFSM_STATUS__UVDTA_PWR_STATUS__SHIFT                                                             0xa
+#define UVD_PGFSM_STATUS__UVDLM_PWR_STATUS__SHIFT                                                             0xc
+#define UVD_PGFSM_STATUS__UVDTD_PWR_STATUS__SHIFT                                                             0xe
+#define UVD_PGFSM_STATUS__UVDTE_PWR_STATUS__SHIFT                                                             0x10
+#define UVD_PGFSM_STATUS__UVDE_PWR_STATUS__SHIFT                                                              0x12
+#define UVD_PGFSM_STATUS__UVDAB_PWR_STATUS__SHIFT                                                             0x14
+#define UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT                                                              0x16
+#define UVD_PGFSM_STATUS__UVDTB_PWR_STATUS__SHIFT                                                             0x18
+#define UVD_PGFSM_STATUS__UVDNA_PWR_STATUS__SHIFT                                                             0x1a
+#define UVD_PGFSM_STATUS__UVDNB_PWR_STATUS__SHIFT                                                             0x1c
+#define UVD_PGFSM_STATUS__UVDM_PWR_STATUS_MASK                                                                0x00000003L
+#define UVD_PGFSM_STATUS__UVDS_PWR_STATUS_MASK                                                                0x0000000CL
+#define UVD_PGFSM_STATUS__UVDF_PWR_STATUS_MASK                                                                0x00000030L
+#define UVD_PGFSM_STATUS__UVDTC_PWR_STATUS_MASK                                                               0x000000C0L
+#define UVD_PGFSM_STATUS__UVDB_PWR_STATUS_MASK                                                                0x00000300L
+#define UVD_PGFSM_STATUS__UVDTA_PWR_STATUS_MASK                                                               0x00000C00L
+#define UVD_PGFSM_STATUS__UVDLM_PWR_STATUS_MASK                                                               0x00003000L
+#define UVD_PGFSM_STATUS__UVDTD_PWR_STATUS_MASK                                                               0x0000C000L
+#define UVD_PGFSM_STATUS__UVDTE_PWR_STATUS_MASK                                                               0x00030000L
+#define UVD_PGFSM_STATUS__UVDE_PWR_STATUS_MASK                                                                0x000C0000L
+#define UVD_PGFSM_STATUS__UVDAB_PWR_STATUS_MASK                                                               0x00300000L
+#define UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK                                                                0x00C00000L
+#define UVD_PGFSM_STATUS__UVDTB_PWR_STATUS_MASK                                                               0x03000000L
+#define UVD_PGFSM_STATUS__UVDNA_PWR_STATUS_MASK                                                               0x0C000000L
+#define UVD_PGFSM_STATUS__UVDNB_PWR_STATUS_MASK                                                               0x30000000L
+//UVD_POWER_STATUS
+#define UVD_POWER_STATUS__UVD_POWER_STATUS__SHIFT                                                             0x0
+#define UVD_POWER_STATUS__UVD_PG_MODE__SHIFT                                                                  0x2
+#define UVD_POWER_STATUS__UVD_CG_MODE__SHIFT                                                                  0x4
+#define UVD_POWER_STATUS__UVD_PG_EN__SHIFT                                                                    0x8
+#define UVD_POWER_STATUS__RBC_SNOOP_DIS__SHIFT                                                                0x9
+#define UVD_POWER_STATUS__SW_RB_SNOOP_DIS__SHIFT                                                              0xb
+#define UVD_POWER_STATUS__STALL_DPG_POWER_UP__SHIFT                                                           0x1f
+#define UVD_POWER_STATUS__UVD_POWER_STATUS_MASK                                                               0x00000003L
+#define UVD_POWER_STATUS__UVD_PG_MODE_MASK                                                                    0x00000004L
+#define UVD_POWER_STATUS__UVD_CG_MODE_MASK                                                                    0x00000030L
+#define UVD_POWER_STATUS__UVD_PG_EN_MASK                                                                      0x00000100L
+#define UVD_POWER_STATUS__RBC_SNOOP_DIS_MASK                                                                  0x00000200L
+#define UVD_POWER_STATUS__SW_RB_SNOOP_DIS_MASK                                                                0x00000800L
+#define UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK                                                             0x80000000L
+//UVD_JPEG_POWER_STATUS
+#define UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS__SHIFT                                                       0x0
+#define UVD_JPEG_POWER_STATUS__JPEG_PG_MODE__SHIFT                                                            0x4
+#define UVD_JPEG_POWER_STATUS__JRBC_DEC_SNOOP_DIS__SHIFT                                                      0x8
+#define UVD_JPEG_POWER_STATUS__JRBC_ENC_SNOOP_DIS__SHIFT                                                      0x9
+#define UVD_JPEG_POWER_STATUS__STALL_JDPG_POWER_UP__SHIFT                                                     0x1f
+#define UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK                                                         0x00000001L
+#define UVD_JPEG_POWER_STATUS__JPEG_PG_MODE_MASK                                                              0x00000010L
+#define UVD_JPEG_POWER_STATUS__JRBC_DEC_SNOOP_DIS_MASK                                                        0x00000100L
+#define UVD_JPEG_POWER_STATUS__JRBC_ENC_SNOOP_DIS_MASK                                                        0x00000200L
+#define UVD_JPEG_POWER_STATUS__STALL_JDPG_POWER_UP_MASK                                                       0x80000000L
+//UVD_MC_DJPEG_RD_SPACE
+#define UVD_MC_DJPEG_RD_SPACE__DJPEG_RD_SPACE__SHIFT                                                          0x0
+#define UVD_MC_DJPEG_RD_SPACE__DJPEG_RD_SPACE_MASK                                                            0x0003FFFFL
+//UVD_MC_DJPEG_WR_SPACE
+#define UVD_MC_DJPEG_WR_SPACE__DJPEG_WR_SPACE__SHIFT                                                          0x0
+#define UVD_MC_DJPEG_WR_SPACE__DJPEG_WR_SPACE_MASK                                                            0x0003FFFFL
+//UVD_MC_EJPEG_RD_SPACE
+#define UVD_MC_EJPEG_RD_SPACE__EJPEG_RD_SPACE__SHIFT                                                          0x0
+#define UVD_MC_EJPEG_RD_SPACE__EJPEG_RD_SPACE_MASK                                                            0x0003FFFFL
+//UVD_MC_EJPEG_WR_SPACE
+#define UVD_MC_EJPEG_WR_SPACE__EJPEG_WR_SPACE__SHIFT                                                          0x0
+#define UVD_MC_EJPEG_WR_SPACE__EJPEG_WR_SPACE_MASK                                                            0x0003FFFFL
+//UVD_PG_IND_INDEX
+#define UVD_PG_IND_INDEX__INDEX__SHIFT                                                                        0x0
+#define UVD_PG_IND_INDEX__INDEX_MASK                                                                          0x0000003FL
+//UVD_PG_IND_DATA
+#define UVD_PG_IND_DATA__DATA__SHIFT                                                                          0x0
+#define UVD_PG_IND_DATA__DATA_MASK                                                                            0xFFFFFFFFL
+//CC_UVD_HARVESTING
+#define CC_UVD_HARVESTING__MMSCH_DISABLE__SHIFT                                                               0x0
+#define CC_UVD_HARVESTING__UVD_DISABLE__SHIFT                                                                 0x1
+#define CC_UVD_HARVESTING__MMSCH_DISABLE_MASK                                                                 0x00000001L
+#define CC_UVD_HARVESTING__UVD_DISABLE_MASK                                                                   0x00000002L
+//UVD_DPG_LMA_CTL
+#define UVD_DPG_LMA_CTL__READ_WRITE__SHIFT                                                                    0x0
+#define UVD_DPG_LMA_CTL__MASK_EN__SHIFT                                                                       0x1
+#define UVD_DPG_LMA_CTL__ADDR_AUTO_INCREMENT__SHIFT                                                           0x2
+#define UVD_DPG_LMA_CTL__SRAM_SEL__SHIFT                                                                      0x4
+#define UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT                                                               0x10
+#define UVD_DPG_LMA_CTL__READ_WRITE_MASK                                                                      0x00000001L
+#define UVD_DPG_LMA_CTL__MASK_EN_MASK                                                                         0x00000002L
+#define UVD_DPG_LMA_CTL__ADDR_AUTO_INCREMENT_MASK                                                             0x00000004L
+#define UVD_DPG_LMA_CTL__SRAM_SEL_MASK                                                                        0x00000010L
+#define UVD_DPG_LMA_CTL__READ_WRITE_ADDR_MASK                                                                 0xFFFF0000L
+//UVD_DPG_LMA_DATA
+#define UVD_DPG_LMA_DATA__LMA_DATA__SHIFT                                                                     0x0
+#define UVD_DPG_LMA_DATA__LMA_DATA_MASK                                                                       0xFFFFFFFFL
+//UVD_DPG_LMA_MASK
+#define UVD_DPG_LMA_MASK__LMA_MASK__SHIFT                                                                     0x0
+#define UVD_DPG_LMA_MASK__LMA_MASK_MASK                                                                       0xFFFFFFFFL
+//UVD_DPG_PAUSE
+#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_REQ__SHIFT                                                              0x0
+#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK__SHIFT                                                              0x1
+#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ__SHIFT                                                                0x2
+#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK__SHIFT                                                                0x3
+#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_REQ_MASK                                                                0x00000001L
+#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK_MASK                                                                0x00000002L
+#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK                                                                  0x00000004L
+#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK                                                                  0x00000008L
+//UVD_SCRATCH1
+#define UVD_SCRATCH1__SCRATCH1_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH1__SCRATCH1_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH2
+#define UVD_SCRATCH2__SCRATCH2_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH2__SCRATCH2_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH3
+#define UVD_SCRATCH3__SCRATCH3_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH3__SCRATCH3_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH4
+#define UVD_SCRATCH4__SCRATCH4_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH4__SCRATCH4_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH5
+#define UVD_SCRATCH5__SCRATCH5_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH5__SCRATCH5_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH6
+#define UVD_SCRATCH6__SCRATCH6_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH6__SCRATCH6_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH7
+#define UVD_SCRATCH7__SCRATCH7_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH7__SCRATCH7_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH8
+#define UVD_SCRATCH8__SCRATCH8_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH8__SCRATCH8_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH9
+#define UVD_SCRATCH9__SCRATCH9_DATA__SHIFT                                                                    0x0
+#define UVD_SCRATCH9__SCRATCH9_DATA_MASK                                                                      0xFFFFFFFFL
+//UVD_SCRATCH10
+#define UVD_SCRATCH10__SCRATCH10_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH10__SCRATCH10_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_SCRATCH11
+#define UVD_SCRATCH11__SCRATCH11_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH11__SCRATCH11_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_SCRATCH12
+#define UVD_SCRATCH12__SCRATCH12_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH12__SCRATCH12_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_SCRATCH13
+#define UVD_SCRATCH13__SCRATCH13_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH13__SCRATCH13_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_SCRATCH14
+#define UVD_SCRATCH14__SCRATCH14_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH14__SCRATCH14_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_FREE_COUNTER_REG
+#define UVD_FREE_COUNTER_REG__FREE_COUNTER__SHIFT                                                             0x0
+#define UVD_FREE_COUNTER_REG__FREE_COUNTER_MASK                                                               0xFFFFFFFFL
+//UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW
+#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                0x0
+#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0_MASK                                                  0xFFFFFFFFL
+//UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH
+#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                              0x0
+#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32_MASK                                                0xFFFFFFFFL
+//UVD_DPG_VCPU_CACHE_OFFSET0
+#define UVD_DPG_VCPU_CACHE_OFFSET0__CACHE_OFFSET0__SHIFT                                                      0x0
+#define UVD_DPG_VCPU_CACHE_OFFSET0__CACHE_OFFSET0_MASK                                                        0x01FFFFFFL
+//UVD_DPG_LMI_VCPU_CACHE_VMID
+#define UVD_DPG_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID__SHIFT                                                   0x0
+#define UVD_DPG_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID_MASK                                                     0x0000000FL
+//UVD_REG_FILTER_EN
+#define UVD_REG_FILTER_EN__UVD_REG_FILTER_EN__SHIFT                                                           0x0
+#define UVD_REG_FILTER_EN__MMSCH_HI_PRIV__SHIFT                                                               0x1
+#define UVD_REG_FILTER_EN__VIDEO_PRIV_EN__SHIFT                                                               0x2
+#define UVD_REG_FILTER_EN__JPEG_PRIV_EN__SHIFT                                                                0x3
+#define UVD_REG_FILTER_EN__UVD_REG_FILTER_EN_MASK                                                             0x00000001L
+#define UVD_REG_FILTER_EN__MMSCH_HI_PRIV_MASK                                                                 0x00000002L
+#define UVD_REG_FILTER_EN__VIDEO_PRIV_EN_MASK                                                                 0x00000004L
+#define UVD_REG_FILTER_EN__JPEG_PRIV_EN_MASK                                                                  0x00000008L
+//UVD_SECURITY_REG_VIO_REPORT
+#define UVD_SECURITY_REG_VIO_REPORT__HOST_REG_VIO__SHIFT                                                      0x0
+#define UVD_SECURITY_REG_VIO_REPORT__VCPU_REG_VIO__SHIFT                                                      0x1
+#define UVD_SECURITY_REG_VIO_REPORT__VIDEO_REG_VIO__SHIFT                                                     0x2
+#define UVD_SECURITY_REG_VIO_REPORT__DPG_REG_VIO__SHIFT                                                       0x3
+#define UVD_SECURITY_REG_VIO_REPORT__JPEG_REG_VIO__SHIFT                                                      0x4
+#define UVD_SECURITY_REG_VIO_REPORT__JDPG_REG_VIO__SHIFT                                                      0x5
+#define UVD_SECURITY_REG_VIO_REPORT__HOST_REG_VIO_MASK                                                        0x00000001L
+#define UVD_SECURITY_REG_VIO_REPORT__VCPU_REG_VIO_MASK                                                        0x00000002L
+#define UVD_SECURITY_REG_VIO_REPORT__VIDEO_REG_VIO_MASK                                                       0x00000004L
+#define UVD_SECURITY_REG_VIO_REPORT__DPG_REG_VIO_MASK                                                         0x00000008L
+#define UVD_SECURITY_REG_VIO_REPORT__JPEG_REG_VIO_MASK                                                        0x00000010L
+#define UVD_SECURITY_REG_VIO_REPORT__JDPG_REG_VIO_MASK                                                        0x00000020L
+//UVD_FW_VERSION
+#define UVD_FW_VERSION__FW_VERSION__SHIFT                                                                     0x0
+#define UVD_FW_VERSION__FW_VERSION_MASK                                                                       0xFFFFFFFFL
+//UVD_PF_STATUS
+#define UVD_PF_STATUS__JPEG_PF_OCCURED__SHIFT                                                                 0x0
+#define UVD_PF_STATUS__NJ_PF_OCCURED__SHIFT                                                                   0x1
+#define UVD_PF_STATUS__ENCODER0_PF_OCCURED__SHIFT                                                             0x2
+#define UVD_PF_STATUS__ENCODER1_PF_OCCURED__SHIFT                                                             0x3
+#define UVD_PF_STATUS__ENCODER2_PF_OCCURED__SHIFT                                                             0x4
+#define UVD_PF_STATUS__ENCODER3_PF_OCCURED__SHIFT                                                             0x5
+#define UVD_PF_STATUS__ENCODER4_PF_OCCURED__SHIFT                                                             0x6
+#define UVD_PF_STATUS__EJPEG_PF_OCCURED__SHIFT                                                                0x7
+#define UVD_PF_STATUS__JPEG_PF_CLEAR__SHIFT                                                                   0x8
+#define UVD_PF_STATUS__NJ_PF_CLEAR__SHIFT                                                                     0x9
+#define UVD_PF_STATUS__ENCODER0_PF_CLEAR__SHIFT                                                               0xa
+#define UVD_PF_STATUS__ENCODER1_PF_CLEAR__SHIFT                                                               0xb
+#define UVD_PF_STATUS__ENCODER2_PF_CLEAR__SHIFT                                                               0xc
+#define UVD_PF_STATUS__ENCODER3_PF_CLEAR__SHIFT                                                               0xd
+#define UVD_PF_STATUS__ENCODER4_PF_CLEAR__SHIFT                                                               0xe
+#define UVD_PF_STATUS__EJPEG_PF_CLEAR__SHIFT                                                                  0xf
+#define UVD_PF_STATUS__NJ_ATM_PF_OCCURED__SHIFT                                                               0x10
+#define UVD_PF_STATUS__DJ_ATM_PF_OCCURED__SHIFT                                                               0x11
+#define UVD_PF_STATUS__EJ_ATM_PF_OCCURED__SHIFT                                                               0x12
+#define UVD_PF_STATUS__JPEG2_PF_OCCURED__SHIFT                                                                0x13
+#define UVD_PF_STATUS__DJ2_ATM_PF_OCCURED__SHIFT                                                              0x14
+#define UVD_PF_STATUS__JPEG2_PF_CLEAR__SHIFT                                                                  0x15
+#define UVD_PF_STATUS__ENCODER5_PF_OCCURED__SHIFT                                                             0x16
+#define UVD_PF_STATUS__ENCODER5_PF_CLEAR__SHIFT                                                               0x17
+#define UVD_PF_STATUS__JPEG_PF_OCCURED_MASK                                                                   0x00000001L
+#define UVD_PF_STATUS__NJ_PF_OCCURED_MASK                                                                     0x00000002L
+#define UVD_PF_STATUS__ENCODER0_PF_OCCURED_MASK                                                               0x00000004L
+#define UVD_PF_STATUS__ENCODER1_PF_OCCURED_MASK                                                               0x00000008L
+#define UVD_PF_STATUS__ENCODER2_PF_OCCURED_MASK                                                               0x00000010L
+#define UVD_PF_STATUS__ENCODER3_PF_OCCURED_MASK                                                               0x00000020L
+#define UVD_PF_STATUS__ENCODER4_PF_OCCURED_MASK                                                               0x00000040L
+#define UVD_PF_STATUS__EJPEG_PF_OCCURED_MASK                                                                  0x00000080L
+#define UVD_PF_STATUS__JPEG_PF_CLEAR_MASK                                                                     0x00000100L
+#define UVD_PF_STATUS__NJ_PF_CLEAR_MASK                                                                       0x00000200L
+#define UVD_PF_STATUS__ENCODER0_PF_CLEAR_MASK                                                                 0x00000400L
+#define UVD_PF_STATUS__ENCODER1_PF_CLEAR_MASK                                                                 0x00000800L
+#define UVD_PF_STATUS__ENCODER2_PF_CLEAR_MASK                                                                 0x00001000L
+#define UVD_PF_STATUS__ENCODER3_PF_CLEAR_MASK                                                                 0x00002000L
+#define UVD_PF_STATUS__ENCODER4_PF_CLEAR_MASK                                                                 0x00004000L
+#define UVD_PF_STATUS__EJPEG_PF_CLEAR_MASK                                                                    0x00008000L
+#define UVD_PF_STATUS__NJ_ATM_PF_OCCURED_MASK                                                                 0x00010000L
+#define UVD_PF_STATUS__DJ_ATM_PF_OCCURED_MASK                                                                 0x00020000L
+#define UVD_PF_STATUS__EJ_ATM_PF_OCCURED_MASK                                                                 0x00040000L
+#define UVD_PF_STATUS__JPEG2_PF_OCCURED_MASK                                                                  0x00080000L
+#define UVD_PF_STATUS__DJ2_ATM_PF_OCCURED_MASK                                                                0x00100000L
+#define UVD_PF_STATUS__JPEG2_PF_CLEAR_MASK                                                                    0x00200000L
+#define UVD_PF_STATUS__ENCODER5_PF_OCCURED_MASK                                                               0x00400000L
+#define UVD_PF_STATUS__ENCODER5_PF_CLEAR_MASK                                                                 0x00800000L
+//UVD_DPG_CLK_EN_VCPU_REPORT
+#define UVD_DPG_CLK_EN_VCPU_REPORT__CLK_EN__SHIFT                                                             0x0
+#define UVD_DPG_CLK_EN_VCPU_REPORT__VCPU_REPORT__SHIFT                                                        0x1
+#define UVD_DPG_CLK_EN_VCPU_REPORT__CLK_EN_MASK                                                               0x00000001L
+#define UVD_DPG_CLK_EN_VCPU_REPORT__VCPU_REPORT_MASK                                                          0x000000FEL
+//CC_UVD_VCPU_ERR_DETECT_BOT_LO
+#define CC_UVD_VCPU_ERR_DETECT_BOT_LO__UVD_VCPU_ERR_DETECT_BOT_LO__SHIFT                                      0xc
+#define CC_UVD_VCPU_ERR_DETECT_BOT_LO__UVD_VCPU_ERR_DETECT_BOT_LO_MASK                                        0xFFFFF000L
+//CC_UVD_VCPU_ERR_DETECT_BOT_HI
+#define CC_UVD_VCPU_ERR_DETECT_BOT_HI__UVD_VCPU_ERR_DETECT_BOT_HI__SHIFT                                      0x0
+#define CC_UVD_VCPU_ERR_DETECT_BOT_HI__UVD_VCPU_ERR_DETECT_BOT_HI_MASK                                        0x0000FFFFL
+//CC_UVD_VCPU_ERR_DETECT_TOP_LO
+#define CC_UVD_VCPU_ERR_DETECT_TOP_LO__UVD_VCPU_ERR_DETECT_TOP_LO__SHIFT                                      0xc
+#define CC_UVD_VCPU_ERR_DETECT_TOP_LO__UVD_VCPU_ERR_DETECT_TOP_LO_MASK                                        0xFFFFF000L
+//CC_UVD_VCPU_ERR_DETECT_TOP_HI
+#define CC_UVD_VCPU_ERR_DETECT_TOP_HI__UVD_VCPU_ERR_DETECT_TOP_HI__SHIFT                                      0x0
+#define CC_UVD_VCPU_ERR_DETECT_TOP_HI__UVD_VCPU_ERR_DETECT_TOP_HI_MASK                                        0x0000FFFFL
+//CC_UVD_VCPU_ERR
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_STATUS__SHIFT                                                           0x0
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_CLEAR__SHIFT                                                            0x1
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_DETECT_EN__SHIFT                                                        0x2
+#define CC_UVD_VCPU_ERR__UVD_TMZ_DBG_DIS__SHIFT                                                               0x3
+#define CC_UVD_VCPU_ERR__RESET_ON_FAULT__SHIFT                                                                0x4
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_STATUS_MASK                                                             0x00000001L
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_CLEAR_MASK                                                              0x00000002L
+#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_DETECT_EN_MASK                                                          0x00000004L
+#define CC_UVD_VCPU_ERR__UVD_TMZ_DBG_DIS_MASK                                                                 0x00000008L
+#define CC_UVD_VCPU_ERR__RESET_ON_FAULT_MASK                                                                  0x00000010L
+//CC_UVD_VCPU_ERR_INST_ADDR_LO
+#define CC_UVD_VCPU_ERR_INST_ADDR_LO__UVD_VCPU_ERR_INST_ADDR_LO__SHIFT                                        0x0
+#define CC_UVD_VCPU_ERR_INST_ADDR_LO__UVD_VCPU_ERR_INST_ADDR_LO_MASK                                          0xFFFFFFFFL
+//CC_UVD_VCPU_ERR_INST_ADDR_HI
+#define CC_UVD_VCPU_ERR_INST_ADDR_HI__UVD_VCPU_ERR_INST_ADDR_HI__SHIFT                                        0x0
+#define CC_UVD_VCPU_ERR_INST_ADDR_HI__UVD_VCPU_ERR_INST_ADDR_HI_MASK                                          0x0000FFFFL
+//UVD_LMI_MMSCH_NC_SPACE
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC0_SPACE__SHIFT                                                        0x0
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC1_SPACE__SHIFT                                                        0x3
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC2_SPACE__SHIFT                                                        0x6
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC3_SPACE__SHIFT                                                        0x9
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC4_SPACE__SHIFT                                                        0xc
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC5_SPACE__SHIFT                                                        0xf
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC6_SPACE__SHIFT                                                        0x12
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC7_SPACE__SHIFT                                                        0x15
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC0_SPACE_MASK                                                          0x00000007L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC1_SPACE_MASK                                                          0x00000038L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC2_SPACE_MASK                                                          0x000001C0L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC3_SPACE_MASK                                                          0x00000E00L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC4_SPACE_MASK                                                          0x00007000L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC5_SPACE_MASK                                                          0x00038000L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC6_SPACE_MASK                                                          0x001C0000L
+#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC7_SPACE_MASK                                                          0x00E00000L
+//UVD_LMI_ATOMIC_SPACE
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER0_SPACE__SHIFT                                                       0x0
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER1_SPACE__SHIFT                                                       0x3
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER2_SPACE__SHIFT                                                       0x6
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER3_SPACE__SHIFT                                                       0x9
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER0_SPACE_MASK                                                         0x00000007L
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER1_SPACE_MASK                                                         0x00000038L
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER2_SPACE_MASK                                                         0x000001C0L
+#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER3_SPACE_MASK                                                         0x00000E00L
+//UVD_GFX8_ADDR_CONFIG
+#define UVD_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT                                                     0x4
+#define UVD_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK                                                       0x00000070L
+//UVD_GFX10_ADDR_CONFIG
+#define UVD_GFX10_ADDR_CONFIG__NUM_PIPES__SHIFT                                                               0x0
+#define UVD_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT                                                    0x3
+#define UVD_GFX10_ADDR_CONFIG__MAX_COMPRESSED_FRAGS__SHIFT                                                    0x6
+#define UVD_GFX10_ADDR_CONFIG__NUM_PKRS__SHIFT                                                                0x8
+#define UVD_GFX10_ADDR_CONFIG__NUM_BANKS__SHIFT                                                               0xc
+#define UVD_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES__SHIFT                                                      0x13
+#define UVD_GFX10_ADDR_CONFIG__NUM_PIPES_MASK                                                                 0x00000007L
+#define UVD_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK                                                      0x00000038L
+#define UVD_GFX10_ADDR_CONFIG__MAX_COMPRESSED_FRAGS_MASK                                                      0x000000C0L
+#define UVD_GFX10_ADDR_CONFIG__NUM_PKRS_MASK                                                                  0x00000700L
+#define UVD_GFX10_ADDR_CONFIG__NUM_BANKS_MASK                                                                 0x00007000L
+#define UVD_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES_MASK                                                        0x00180000L
+//UVD_GPCNT2_CNTL
+#define UVD_GPCNT2_CNTL__CLR__SHIFT                                                                           0x0
+#define UVD_GPCNT2_CNTL__START__SHIFT                                                                         0x1
+#define UVD_GPCNT2_CNTL__COUNTUP__SHIFT                                                                       0x2
+#define UVD_GPCNT2_CNTL__CLR_MASK                                                                             0x00000001L
+#define UVD_GPCNT2_CNTL__START_MASK                                                                           0x00000002L
+#define UVD_GPCNT2_CNTL__COUNTUP_MASK                                                                         0x00000004L
+//UVD_GPCNT2_TARGET_LOWER
+#define UVD_GPCNT2_TARGET_LOWER__TARGET__SHIFT                                                                0x0
+#define UVD_GPCNT2_TARGET_LOWER__TARGET_MASK                                                                  0xFFFFFFFFL
+//UVD_GPCNT2_STATUS_LOWER
+#define UVD_GPCNT2_STATUS_LOWER__COUNT__SHIFT                                                                 0x0
+#define UVD_GPCNT2_STATUS_LOWER__COUNT_MASK                                                                   0xFFFFFFFFL
+//UVD_GPCNT2_TARGET_UPPER
+#define UVD_GPCNT2_TARGET_UPPER__TARGET__SHIFT                                                                0x0
+#define UVD_GPCNT2_TARGET_UPPER__TARGET_MASK                                                                  0x0000FFFFL
+//UVD_GPCNT2_STATUS_UPPER
+#define UVD_GPCNT2_STATUS_UPPER__COUNT__SHIFT                                                                 0x0
+#define UVD_GPCNT2_STATUS_UPPER__COUNT_MASK                                                                   0x0000FFFFL
+//UVD_GPCNT3_CNTL
+#define UVD_GPCNT3_CNTL__CLR__SHIFT                                                                           0x0
+#define UVD_GPCNT3_CNTL__START__SHIFT                                                                         0x1
+#define UVD_GPCNT3_CNTL__COUNTUP__SHIFT                                                                       0x2
+#define UVD_GPCNT3_CNTL__FREQ__SHIFT                                                                          0x3
+#define UVD_GPCNT3_CNTL__DIV__SHIFT                                                                           0xa
+#define UVD_GPCNT3_CNTL__CLR_MASK                                                                             0x00000001L
+#define UVD_GPCNT3_CNTL__START_MASK                                                                           0x00000002L
+#define UVD_GPCNT3_CNTL__COUNTUP_MASK                                                                         0x00000004L
+#define UVD_GPCNT3_CNTL__FREQ_MASK                                                                            0x000003F8L
+#define UVD_GPCNT3_CNTL__DIV_MASK                                                                             0x0001FC00L
+//UVD_GPCNT3_TARGET_LOWER
+#define UVD_GPCNT3_TARGET_LOWER__TARGET__SHIFT                                                                0x0
+#define UVD_GPCNT3_TARGET_LOWER__TARGET_MASK                                                                  0xFFFFFFFFL
+//UVD_GPCNT3_STATUS_LOWER
+#define UVD_GPCNT3_STATUS_LOWER__COUNT__SHIFT                                                                 0x0
+#define UVD_GPCNT3_STATUS_LOWER__COUNT_MASK                                                                   0xFFFFFFFFL
+//UVD_GPCNT3_TARGET_UPPER
+#define UVD_GPCNT3_TARGET_UPPER__TARGET__SHIFT                                                                0x0
+#define UVD_GPCNT3_TARGET_UPPER__TARGET_MASK                                                                  0x0000FFFFL
+//UVD_GPCNT3_STATUS_UPPER
+#define UVD_GPCNT3_STATUS_UPPER__COUNT__SHIFT                                                                 0x0
+#define UVD_GPCNT3_STATUS_UPPER__COUNT_MASK                                                                   0x0000FFFFL
+//UVD_VCLK_DS_CNTL
+#define UVD_VCLK_DS_CNTL__VCLK_DS_EN__SHIFT                                                                   0x0
+#define UVD_VCLK_DS_CNTL__VCLK_DS_STATUS__SHIFT                                                               0x4
+#define UVD_VCLK_DS_CNTL__VCLK_DS_HYSTERESIS_CNT__SHIFT                                                       0x10
+#define UVD_VCLK_DS_CNTL__VCLK_DS_EN_MASK                                                                     0x00000001L
+#define UVD_VCLK_DS_CNTL__VCLK_DS_STATUS_MASK                                                                 0x00000010L
+#define UVD_VCLK_DS_CNTL__VCLK_DS_HYSTERESIS_CNT_MASK                                                         0xFFFF0000L
+//UVD_DCLK_DS_CNTL
+#define UVD_DCLK_DS_CNTL__DCLK_DS_EN__SHIFT                                                                   0x0
+#define UVD_DCLK_DS_CNTL__DCLK_DS_STATUS__SHIFT                                                               0x4
+#define UVD_DCLK_DS_CNTL__DCLK_DS_HYSTERESIS_CNT__SHIFT                                                       0x10
+#define UVD_DCLK_DS_CNTL__DCLK_DS_EN_MASK                                                                     0x00000001L
+#define UVD_DCLK_DS_CNTL__DCLK_DS_STATUS_MASK                                                                 0x00000010L
+#define UVD_DCLK_DS_CNTL__DCLK_DS_HYSTERESIS_CNT_MASK                                                         0xFFFF0000L
+//UVD_TSC_LOWER
+#define UVD_TSC_LOWER__COUNT__SHIFT                                                                           0x0
+#define UVD_TSC_LOWER__COUNT_MASK                                                                             0xFFFFFFFFL
+//UVD_TSC_UPPER
+#define UVD_TSC_UPPER__COUNT__SHIFT                                                                           0x0
+#define UVD_TSC_UPPER__COUNT_MASK                                                                             0x00FFFFFFL
+//VCN_FEATURES
+#define VCN_FEATURES__HAS_VIDEO_DEC__SHIFT                                                                    0x0
+#define VCN_FEATURES__HAS_VIDEO_ENC__SHIFT                                                                    0x1
+#define VCN_FEATURES__HAS_MJPEG_DEC__SHIFT                                                                    0x2
+#define VCN_FEATURES__HAS_MJPEG_ENC__SHIFT                                                                    0x3
+#define VCN_FEATURES__HAS_VIDEO_VIRT__SHIFT                                                                   0x4
+#define VCN_FEATURES__HAS_H264_LEGACY_DEC__SHIFT                                                              0x5
+#define VCN_FEATURES__HAS_UDEC_DEC__SHIFT                                                                     0x6
+#define VCN_FEATURES__HAS_MJPEG2_IDCT_DEC__SHIFT                                                              0x7
+#define VCN_FEATURES__HAS_SCLR_DEC__SHIFT                                                                     0x8
+#define VCN_FEATURES__HAS_VP9_DEC__SHIFT                                                                      0x9
+#define VCN_FEATURES__HAS_AV1_DEC__SHIFT                                                                      0xa
+#define VCN_FEATURES__HAS_EFC_ENC__SHIFT                                                                      0xb
+#define VCN_FEATURES__HAS_EFC_HDR2SDR_ENC__SHIFT                                                              0xc
+#define VCN_FEATURES__HAS_DUAL_MJPEG_DEC__SHIFT                                                               0xd
+#define VCN_FEATURES__HAS_AV1_ENC__SHIFT                                                                      0xe
+#define VCN_FEATURES__INSTANCE_ID__SHIFT                                                                      0x1c
+#define VCN_FEATURES__HAS_VIDEO_DEC_MASK                                                                      0x00000001L
+#define VCN_FEATURES__HAS_VIDEO_ENC_MASK                                                                      0x00000002L
+#define VCN_FEATURES__HAS_MJPEG_DEC_MASK                                                                      0x00000004L
+#define VCN_FEATURES__HAS_MJPEG_ENC_MASK                                                                      0x00000008L
+#define VCN_FEATURES__HAS_VIDEO_VIRT_MASK                                                                     0x00000010L
+#define VCN_FEATURES__HAS_H264_LEGACY_DEC_MASK                                                                0x00000020L
+#define VCN_FEATURES__HAS_UDEC_DEC_MASK                                                                       0x00000040L
+#define VCN_FEATURES__HAS_MJPEG2_IDCT_DEC_MASK                                                                0x00000080L
+#define VCN_FEATURES__HAS_SCLR_DEC_MASK                                                                       0x00000100L
+#define VCN_FEATURES__HAS_VP9_DEC_MASK                                                                        0x00000200L
+#define VCN_FEATURES__HAS_AV1_DEC_MASK                                                                        0x00000400L
+#define VCN_FEATURES__HAS_EFC_ENC_MASK                                                                        0x00000800L
+#define VCN_FEATURES__HAS_EFC_HDR2SDR_ENC_MASK                                                                0x00001000L
+#define VCN_FEATURES__HAS_DUAL_MJPEG_DEC_MASK                                                                 0x00002000L
+#define VCN_FEATURES__HAS_AV1_ENC_MASK                                                                        0x00004000L
+#define VCN_FEATURES__INSTANCE_ID_MASK                                                                        0xF0000000L
+//UVD_GPUIOV_STATUS
+#define UVD_GPUIOV_STATUS__UVD_GPUIOV_STATUS_VF_ENABLE__SHIFT                                                 0x0
+#define UVD_GPUIOV_STATUS__UVD_GPUIOV_STATUS_VF_ENABLE_MASK                                                   0x00000001L
+//UVD_RAS_VCPU_VCODEC_STATUS
+#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_VF__SHIFT                                                        0x0
+#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_PF__SHIFT                                                        0x1f
+#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_VF_MASK                                                          0x7FFFFFFFL
+#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_PF_MASK                                                          0x80000000L
+//UVD_RAS_MMSCH_FATAL_ERROR
+#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_VF__SHIFT                                                         0x0
+#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_PF__SHIFT                                                         0x1f
+#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_VF_MASK                                                           0x7FFFFFFFL
+#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_PF_MASK                                                           0x80000000L
+//UVD_RAS_JPEG0_STATUS
+#define UVD_RAS_JPEG0_STATUS__POISONED_VF__SHIFT                                                              0x0
+#define UVD_RAS_JPEG0_STATUS__POISONED_PF__SHIFT                                                              0x1f
+#define UVD_RAS_JPEG0_STATUS__POISONED_VF_MASK                                                                0x7FFFFFFFL
+#define UVD_RAS_JPEG0_STATUS__POISONED_PF_MASK                                                                0x80000000L
+//UVD_RAS_JPEG1_STATUS
+#define UVD_RAS_JPEG1_STATUS__POISONED_VF__SHIFT                                                              0x0
+#define UVD_RAS_JPEG1_STATUS__POISONED_PF__SHIFT                                                              0x1f
+#define UVD_RAS_JPEG1_STATUS__POISONED_VF_MASK                                                                0x7FFFFFFFL
+#define UVD_RAS_JPEG1_STATUS__POISONED_PF_MASK                                                                0x80000000L
+//UVD_RAS_CNTL_PMI_ARB
+#define UVD_RAS_CNTL_PMI_ARB__STAT_VCPU_VCODEC__SHIFT                                                         0x0
+#define UVD_RAS_CNTL_PMI_ARB__ACK_VCPU_VCODEC__SHIFT                                                          0x1
+#define UVD_RAS_CNTL_PMI_ARB__STAT_MMSCH__SHIFT                                                               0x2
+#define UVD_RAS_CNTL_PMI_ARB__ACK_MMSCH__SHIFT                                                                0x3
+#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG0__SHIFT                                                               0x4
+#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG0__SHIFT                                                                0x5
+#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG1__SHIFT                                                               0x6
+#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG1__SHIFT                                                                0x7
+#define UVD_RAS_CNTL_PMI_ARB__STAT_VCPU_VCODEC_MASK                                                           0x00000001L
+#define UVD_RAS_CNTL_PMI_ARB__ACK_VCPU_VCODEC_MASK                                                            0x00000002L
+#define UVD_RAS_CNTL_PMI_ARB__STAT_MMSCH_MASK                                                                 0x00000004L
+#define UVD_RAS_CNTL_PMI_ARB__ACK_MMSCH_MASK                                                                  0x00000008L
+#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG0_MASK                                                                 0x00000010L
+#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG0_MASK                                                                  0x00000020L
+#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG1_MASK                                                                 0x00000040L
+#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG1_MASK                                                                  0x00000080L
+//UVD_SCRATCH15
+#define UVD_SCRATCH15__SCRATCH15_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH15__SCRATCH15_DATA_MASK                                                                    0xFFFFFFFFL
+//VCN_JPEG_DB_CTRL1
+#define VCN_JPEG_DB_CTRL1__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL1__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL1__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL1__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL1__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL1__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL2
+#define VCN_JPEG_DB_CTRL2__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL2__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL2__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL2__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL2__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL2__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL3
+#define VCN_JPEG_DB_CTRL3__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL3__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL3__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL3__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL3__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL3__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL4
+#define VCN_JPEG_DB_CTRL4__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL4__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL4__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL4__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL4__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL4__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL5
+#define VCN_JPEG_DB_CTRL5__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL5__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL5__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL5__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL5__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL5__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL6
+#define VCN_JPEG_DB_CTRL6__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL6__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL6__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL6__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL6__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL6__HIT_MASK                                                                           0x80000000L
+//VCN_JPEG_DB_CTRL7
+#define VCN_JPEG_DB_CTRL7__OFFSET__SHIFT                                                                      0x2
+#define VCN_JPEG_DB_CTRL7__EN__SHIFT                                                                          0x1e
+#define VCN_JPEG_DB_CTRL7__HIT__SHIFT                                                                         0x1f
+#define VCN_JPEG_DB_CTRL7__OFFSET_MASK                                                                        0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL7__EN_MASK                                                                            0x40000000L
+#define VCN_JPEG_DB_CTRL7__HIT_MASK                                                                           0x80000000L
+//UVD_SCRATCH32
+#define UVD_SCRATCH32__SCRATCH32_DATA__SHIFT                                                                  0x0
+#define UVD_SCRATCH32__SCRATCH32_DATA_MASK                                                                    0xFFFFFFFFL
+//UVD_VERSION
+#define UVD_VERSION__VARIANT_TYPE__SHIFT                                                                      0x0
+#define UVD_VERSION__MINOR_VERSION__SHIFT                                                                     0x8
+#define UVD_VERSION__MAJOR_VERSION__SHIFT                                                                     0x10
+#define UVD_VERSION__INSTANCE_ID__SHIFT                                                                       0x1c
+#define UVD_VERSION__VARIANT_TYPE_MASK                                                                        0x000000FFL
+#define UVD_VERSION__MINOR_VERSION_MASK                                                                       0x0000FF00L
+#define UVD_VERSION__MAJOR_VERSION_MASK                                                                       0x0FFF0000L
+#define UVD_VERSION__INSTANCE_ID_MASK                                                                         0xF0000000L
+//VCN_RB_DB_CTRL
+#define VCN_RB_DB_CTRL__OFFSET__SHIFT                                                                         0x2
+#define VCN_RB_DB_CTRL__EN__SHIFT                                                                             0x1e
+#define VCN_RB_DB_CTRL__HIT__SHIFT                                                                            0x1f
+#define VCN_RB_DB_CTRL__OFFSET_MASK                                                                           0x0FFFFFFCL
+#define VCN_RB_DB_CTRL__EN_MASK                                                                               0x40000000L
+#define VCN_RB_DB_CTRL__HIT_MASK                                                                              0x80000000L
+//VCN_JPEG_DB_CTRL
+#define VCN_JPEG_DB_CTRL__OFFSET__SHIFT                                                                       0x2
+#define VCN_JPEG_DB_CTRL__EN__SHIFT                                                                           0x1e
+#define VCN_JPEG_DB_CTRL__HIT__SHIFT                                                                          0x1f
+#define VCN_JPEG_DB_CTRL__OFFSET_MASK                                                                         0x0FFFFFFCL
+#define VCN_JPEG_DB_CTRL__EN_MASK                                                                             0x40000000L
+#define VCN_JPEG_DB_CTRL__HIT_MASK                                                                            0x80000000L
+//VCN_RB1_DB_CTRL
+#define VCN_RB1_DB_CTRL__OFFSET__SHIFT                                                                        0x2
+#define VCN_RB1_DB_CTRL__EN__SHIFT                                                                            0x1e
+#define VCN_RB1_DB_CTRL__HIT__SHIFT                                                                           0x1f
+#define VCN_RB1_DB_CTRL__OFFSET_MASK                                                                          0x0FFFFFFCL
+#define VCN_RB1_DB_CTRL__EN_MASK                                                                              0x40000000L
+#define VCN_RB1_DB_CTRL__HIT_MASK                                                                             0x80000000L
+//VCN_RB2_DB_CTRL
+#define VCN_RB2_DB_CTRL__OFFSET__SHIFT                                                                        0x2
+#define VCN_RB2_DB_CTRL__EN__SHIFT                                                                            0x1e
+#define VCN_RB2_DB_CTRL__HIT__SHIFT                                                                           0x1f
+#define VCN_RB2_DB_CTRL__OFFSET_MASK                                                                          0x0FFFFFFCL
+#define VCN_RB2_DB_CTRL__EN_MASK                                                                              0x40000000L
+#define VCN_RB2_DB_CTRL__HIT_MASK                                                                             0x80000000L
+//VCN_RB3_DB_CTRL
+#define VCN_RB3_DB_CTRL__OFFSET__SHIFT                                                                        0x2
+#define VCN_RB3_DB_CTRL__EN__SHIFT                                                                            0x1e
+#define VCN_RB3_DB_CTRL__HIT__SHIFT                                                                           0x1f
+#define VCN_RB3_DB_CTRL__OFFSET_MASK                                                                          0x0FFFFFFCL
+#define VCN_RB3_DB_CTRL__EN_MASK                                                                              0x40000000L
+#define VCN_RB3_DB_CTRL__HIT_MASK                                                                             0x80000000L
+//VCN_RB4_DB_CTRL
+#define VCN_RB4_DB_CTRL__OFFSET__SHIFT                                                                        0x2
+#define VCN_RB4_DB_CTRL__EN__SHIFT                                                                            0x1e
+#define VCN_RB4_DB_CTRL__HIT__SHIFT                                                                           0x1f
+#define VCN_RB4_DB_CTRL__OFFSET_MASK                                                                          0x0FFFFFFCL
+#define VCN_RB4_DB_CTRL__EN_MASK                                                                              0x40000000L
+#define VCN_RB4_DB_CTRL__HIT_MASK                                                                             0x80000000L
+//VCN_RB_ENABLE
+#define VCN_RB_ENABLE__RB_EN__SHIFT                                                                           0x0
+#define VCN_RB_ENABLE__JPEG_RB_EN__SHIFT                                                                      0x1
+#define VCN_RB_ENABLE__RB1_EN__SHIFT                                                                          0x2
+#define VCN_RB_ENABLE__RB2_EN__SHIFT                                                                          0x3
+#define VCN_RB_ENABLE__RB3_EN__SHIFT                                                                          0x4
+#define VCN_RB_ENABLE__RB4_EN__SHIFT                                                                          0x5
+#define VCN_RB_ENABLE__UMSCH_RB_EN__SHIFT                                                                     0x6
+#define VCN_RB_ENABLE__EJPEG_RB_EN__SHIFT                                                                     0x7
+#define VCN_RB_ENABLE__AUDIO_RB_EN__SHIFT                                                                     0x8
+#define VCN_RB_ENABLE__RB_EN_MASK                                                                             0x00000001L
+#define VCN_RB_ENABLE__JPEG_RB_EN_MASK                                                                        0x00000002L
+#define VCN_RB_ENABLE__RB1_EN_MASK                                                                            0x00000004L
+#define VCN_RB_ENABLE__RB2_EN_MASK                                                                            0x00000008L
+#define VCN_RB_ENABLE__RB3_EN_MASK                                                                            0x00000010L
+#define VCN_RB_ENABLE__RB4_EN_MASK                                                                            0x00000020L
+#define VCN_RB_ENABLE__UMSCH_RB_EN_MASK                                                                       0x00000040L
+#define VCN_RB_ENABLE__EJPEG_RB_EN_MASK                                                                       0x00000080L
+#define VCN_RB_ENABLE__AUDIO_RB_EN_MASK                                                                       0x00000100L
+//VCN_RB_WPTR_CTRL
+#define VCN_RB_WPTR_CTRL__RB_CS_EN__SHIFT                                                                     0x0
+#define VCN_RB_WPTR_CTRL__JPEG_CS_EN__SHIFT                                                                   0x1
+#define VCN_RB_WPTR_CTRL__RB1_CS_EN__SHIFT                                                                    0x2
+#define VCN_RB_WPTR_CTRL__RB2_CS_EN__SHIFT                                                                    0x3
+#define VCN_RB_WPTR_CTRL__RB3_CS_EN__SHIFT                                                                    0x4
+#define VCN_RB_WPTR_CTRL__RB4_CS_EN__SHIFT                                                                    0x5
+#define VCN_RB_WPTR_CTRL__UMSCH_RB_CS_EN__SHIFT                                                               0x6
+#define VCN_RB_WPTR_CTRL__EJPEG_RB_CS_EN__SHIFT                                                               0x7
+#define VCN_RB_WPTR_CTRL__AUDIO_RB_CS_EN__SHIFT                                                               0x8
+#define VCN_RB_WPTR_CTRL__RB_CS_EN_MASK                                                                       0x00000001L
+#define VCN_RB_WPTR_CTRL__JPEG_CS_EN_MASK                                                                     0x00000002L
+#define VCN_RB_WPTR_CTRL__RB1_CS_EN_MASK                                                                      0x00000004L
+#define VCN_RB_WPTR_CTRL__RB2_CS_EN_MASK                                                                      0x00000008L
+#define VCN_RB_WPTR_CTRL__RB3_CS_EN_MASK                                                                      0x00000010L
+#define VCN_RB_WPTR_CTRL__RB4_CS_EN_MASK                                                                      0x00000020L
+#define VCN_RB_WPTR_CTRL__UMSCH_RB_CS_EN_MASK                                                                 0x00000040L
+#define VCN_RB_WPTR_CTRL__EJPEG_RB_CS_EN_MASK                                                                 0x00000080L
+#define VCN_RB_WPTR_CTRL__AUDIO_RB_CS_EN_MASK                                                                 0x00000100L
+//UVD_RB_RPTR
+#define UVD_RB_RPTR__RB_RPTR__SHIFT                                                                           0x4
+#define UVD_RB_RPTR__RB_RPTR_MASK                                                                             0x007FFFF0L
+//UVD_RB_WPTR
+#define UVD_RB_WPTR__RB_WPTR__SHIFT                                                                           0x4
+#define UVD_RB_WPTR__RB_WPTR_MASK                                                                             0x007FFFF0L
+//UVD_RB_RPTR2
+#define UVD_RB_RPTR2__RB_RPTR__SHIFT                                                                          0x4
+#define UVD_RB_RPTR2__RB_RPTR_MASK                                                                            0x007FFFF0L
+//UVD_RB_WPTR2
+#define UVD_RB_WPTR2__RB_WPTR__SHIFT                                                                          0x4
+#define UVD_RB_WPTR2__RB_WPTR_MASK                                                                            0x007FFFF0L
+//UVD_RB_RPTR3
+#define UVD_RB_RPTR3__RB_RPTR__SHIFT                                                                          0x4
+#define UVD_RB_RPTR3__RB_RPTR_MASK                                                                            0x007FFFF0L
+//UVD_RB_WPTR3
+#define UVD_RB_WPTR3__RB_WPTR__SHIFT                                                                          0x4
+#define UVD_RB_WPTR3__RB_WPTR_MASK                                                                            0x007FFFF0L
+//UVD_RB_RPTR4
+#define UVD_RB_RPTR4__RB_RPTR__SHIFT                                                                          0x4
+#define UVD_RB_RPTR4__RB_RPTR_MASK                                                                            0x007FFFF0L
+//UVD_RB_WPTR4
+#define UVD_RB_WPTR4__RB_WPTR__SHIFT                                                                          0x4
+#define UVD_RB_WPTR4__RB_WPTR_MASK                                                                            0x007FFFF0L
+//UVD_OUT_RB_RPTR
+#define UVD_OUT_RB_RPTR__RB_RPTR__SHIFT                                                                       0x4
+#define UVD_OUT_RB_RPTR__RB_RPTR_MASK                                                                         0x007FFFF0L
+//UVD_OUT_RB_WPTR
+#define UVD_OUT_RB_WPTR__RB_WPTR__SHIFT                                                                       0x4
+#define UVD_OUT_RB_WPTR__RB_WPTR_MASK                                                                         0x007FFFF0L
+//UVD_AUDIO_RB_RPTR
+#define UVD_AUDIO_RB_RPTR__RB_RPTR__SHIFT                                                                     0x4
+#define UVD_AUDIO_RB_RPTR__RB_RPTR_MASK                                                                       0x007FFFF0L
+//UVD_AUDIO_RB_WPTR
+#define UVD_AUDIO_RB_WPTR__RB_WPTR__SHIFT                                                                     0x4
+#define UVD_AUDIO_RB_WPTR__RB_WPTR_MASK                                                                       0x007FFFF0L
+//UVD_RBC_RB_RPTR
+#define UVD_RBC_RB_RPTR__RB_RPTR__SHIFT                                                                       0x4
+#define UVD_RBC_RB_RPTR__RB_RPTR_MASK                                                                         0x007FFFF0L
+//UVD_RBC_RB_WPTR
+#define UVD_RBC_RB_WPTR__RB_WPTR__SHIFT                                                                       0x4
+#define UVD_RBC_RB_WPTR__RB_WPTR_MASK                                                                         0x007FFFF0L
+//UVD_DPG_LMA_CTL2
+#define UVD_DPG_LMA_CTL2__DIRECT_ACCESS_SRAM_SEL__SHIFT                                                       0x0
+#define UVD_DPG_LMA_CTL2__FIFO_DIRECT_ACCESS_EN__SHIFT                                                        0x1
+#define UVD_DPG_LMA_CTL2__VID_WRITE_PTR__SHIFT                                                                0x2
+#define UVD_DPG_LMA_CTL2__JPEG_WRITE_PTR__SHIFT                                                               0x9
+#define UVD_DPG_LMA_CTL2__DIRECT_ACCESS_SRAM_SEL_MASK                                                         0x00000001L
+#define UVD_DPG_LMA_CTL2__FIFO_DIRECT_ACCESS_EN_MASK                                                          0x00000002L
+#define UVD_DPG_LMA_CTL2__VID_WRITE_PTR_MASK                                                                  0x000001FCL
+#define UVD_DPG_LMA_CTL2__JPEG_WRITE_PTR_MASK                                                                 0x0000FE00L
+
+
+// addressBlock: aid_uvd0_mmsch_dec
+//MMSCH_UCODE_ADDR
+#define MMSCH_UCODE_ADDR__UCODE_ADDR__SHIFT                                                                   0x2
+#define MMSCH_UCODE_ADDR__UCODE_LOCK__SHIFT                                                                   0x1f
+#define MMSCH_UCODE_ADDR__UCODE_ADDR_MASK                                                                     0x00003FFCL
+#define MMSCH_UCODE_ADDR__UCODE_LOCK_MASK                                                                     0x80000000L
+//MMSCH_UCODE_DATA
+#define MMSCH_UCODE_DATA__UCODE_DATA__SHIFT                                                                   0x0
+#define MMSCH_UCODE_DATA__UCODE_DATA_MASK                                                                     0xFFFFFFFFL
+//MMSCH_SRAM_ADDR
+#define MMSCH_SRAM_ADDR__SRAM_ADDR__SHIFT                                                                     0x2
+#define MMSCH_SRAM_ADDR__SRAM_LOCK__SHIFT                                                                     0x1f
+#define MMSCH_SRAM_ADDR__SRAM_ADDR_MASK                                                                       0x00001FFCL
+#define MMSCH_SRAM_ADDR__SRAM_LOCK_MASK                                                                       0x80000000L
+//MMSCH_SRAM_DATA
+#define MMSCH_SRAM_DATA__SRAM_DATA__SHIFT                                                                     0x0
+#define MMSCH_SRAM_DATA__SRAM_DATA_MASK                                                                       0xFFFFFFFFL
+//MMSCH_VF_SRAM_OFFSET
+#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_OFFSET__SHIFT                                                           0x2
+#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_NUM_DW_PER_VF__SHIFT                                                    0x10
+#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_OFFSET_MASK                                                             0x00001FFCL
+#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_NUM_DW_PER_VF_MASK                                                      0x00FF0000L
+//MMSCH_DB_SRAM_OFFSET
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_OFFSET__SHIFT                                                           0x2
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_ENG__SHIFT                                                          0x10
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_RING_PER_ENG__SHIFT                                                 0x18
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_OFFSET_MASK                                                             0x00001FFCL
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_ENG_MASK                                                            0x00FF0000L
+#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_RING_PER_ENG_MASK                                                   0xFF000000L
+//MMSCH_CTX_SRAM_OFFSET
+#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_OFFSET__SHIFT                                                         0x2
+#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_SIZE__SHIFT                                                           0x10
+#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_OFFSET_MASK                                                           0x00001FFCL
+#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_SIZE_MASK                                                             0xFFFF0000L
+//MMSCH_CTL
+#define MMSCH_CTL__P_RUNSTALL__SHIFT                                                                          0x0
+#define MMSCH_CTL__P_RESET__SHIFT                                                                             0x1
+#define MMSCH_CTL__VFID_FIFO_EN__SHIFT                                                                        0x4
+#define MMSCH_CTL__P_LOCK__SHIFT                                                                              0x1f
+#define MMSCH_CTL__P_RUNSTALL_MASK                                                                            0x00000001L
+#define MMSCH_CTL__P_RESET_MASK                                                                               0x00000002L
+#define MMSCH_CTL__VFID_FIFO_EN_MASK                                                                          0x00000010L
+#define MMSCH_CTL__P_LOCK_MASK                                                                                0x80000000L
+//MMSCH_INTR
+#define MMSCH_INTR__INTR__SHIFT                                                                               0x0
+#define MMSCH_INTR__INTR_MASK                                                                                 0x00001FFFL
+//MMSCH_INTR_ACK
+#define MMSCH_INTR_ACK__INTR__SHIFT                                                                           0x0
+#define MMSCH_INTR_ACK__INTR_MASK                                                                             0x00001FFFL
+//MMSCH_INTR_STATUS
+#define MMSCH_INTR_STATUS__INTR__SHIFT                                                                        0x0
+#define MMSCH_INTR_STATUS__INTR_MASK                                                                          0x00001FFFL
+//MMSCH_VF_VMID
+#define MMSCH_VF_VMID__VF_CTX_VMID__SHIFT                                                                     0x0
+#define MMSCH_VF_VMID__VF_GPCOM_VMID__SHIFT                                                                   0x5
+#define MMSCH_VF_VMID__VF_CTX_VMID_MASK                                                                       0x0000001FL
+#define MMSCH_VF_VMID__VF_GPCOM_VMID_MASK                                                                     0x000003E0L
+//MMSCH_VF_CTX_ADDR_LO
+#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO__SHIFT                                                           0x6
+#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO_MASK                                                             0xFFFFFFC0L
+//MMSCH_VF_CTX_ADDR_HI
+#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI__SHIFT                                                           0x0
+#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI_MASK                                                             0xFFFFFFFFL
+//MMSCH_VF_CTX_SIZE
+#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE__SHIFT                                                                 0x0
+#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE_MASK                                                                   0xFFFFFFFFL
+//MMSCH_VF_GPCOM_ADDR_LO
+#define MMSCH_VF_GPCOM_ADDR_LO__VF_GPCOM_ADDR_LO__SHIFT                                                       0x6
+#define MMSCH_VF_GPCOM_ADDR_LO__VF_GPCOM_ADDR_LO_MASK                                                         0xFFFFFFC0L
+//MMSCH_VF_GPCOM_ADDR_HI
+#define MMSCH_VF_GPCOM_ADDR_HI__VF_GPCOM_ADDR_HI__SHIFT                                                       0x0
+#define MMSCH_VF_GPCOM_ADDR_HI__VF_GPCOM_ADDR_HI_MASK                                                         0xFFFFFFFFL
+//MMSCH_VF_GPCOM_SIZE
+#define MMSCH_VF_GPCOM_SIZE__VF_GPCOM_SIZE__SHIFT                                                             0x0
+#define MMSCH_VF_GPCOM_SIZE__VF_GPCOM_SIZE_MASK                                                               0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_HOST
+#define MMSCH_VF_MAILBOX_HOST__DATA__SHIFT                                                                    0x0
+#define MMSCH_VF_MAILBOX_HOST__DATA_MASK                                                                      0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_RESP
+#define MMSCH_VF_MAILBOX_RESP__RESP__SHIFT                                                                    0x0
+#define MMSCH_VF_MAILBOX_RESP__RESP_MASK                                                                      0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_0
+#define MMSCH_VF_MAILBOX_0__DATA__SHIFT                                                                       0x0
+#define MMSCH_VF_MAILBOX_0__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_0_RESP
+#define MMSCH_VF_MAILBOX_0_RESP__RESP__SHIFT                                                                  0x0
+#define MMSCH_VF_MAILBOX_0_RESP__RESP_MASK                                                                    0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_1
+#define MMSCH_VF_MAILBOX_1__DATA__SHIFT                                                                       0x0
+#define MMSCH_VF_MAILBOX_1__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_VF_MAILBOX_1_RESP
+#define MMSCH_VF_MAILBOX_1_RESP__RESP__SHIFT                                                                  0x0
+#define MMSCH_VF_MAILBOX_1_RESP__RESP_MASK                                                                    0xFFFFFFFFL
+//MMSCH_CNTL
+#define MMSCH_CNTL__CLK_EN__SHIFT                                                                             0x0
+#define MMSCH_CNTL__ED_ENABLE__SHIFT                                                                          0x1
+#define MMSCH_CNTL__AXI_MAX_BRST_SIZE_IS_4__SHIFT                                                             0x2
+#define MMSCH_CNTL__AXI_40BIT_PIF_ADDR_FIX_EN__SHIFT                                                          0x3
+#define MMSCH_CNTL__PDEBUG_ENABLE__SHIFT                                                                      0x4
+#define MMSCH_CNTL__MMSCH_IRQ_ERR__SHIFT                                                                      0x5
+#define MMSCH_CNTL__MMSCH_NACK_INTR_EN__SHIFT                                                                 0x9
+#define MMSCH_CNTL__MMSCH_DB_BUSY_INTR_EN__SHIFT                                                              0xa
+#define MMSCH_CNTL__PRB_TIMEOUT_VAL__SHIFT                                                                    0x14
+#define MMSCH_CNTL__TIMEOUT_DIS__SHIFT                                                                        0x1c
+#define MMSCH_CNTL__MMSCH_IDLE__SHIFT                                                                         0x1d
+#define MMSCH_CNTL__CLK_EN_MASK                                                                               0x00000001L
+#define MMSCH_CNTL__ED_ENABLE_MASK                                                                            0x00000002L
+#define MMSCH_CNTL__AXI_MAX_BRST_SIZE_IS_4_MASK                                                               0x00000004L
+#define MMSCH_CNTL__AXI_40BIT_PIF_ADDR_FIX_EN_MASK                                                            0x00000008L
+#define MMSCH_CNTL__PDEBUG_ENABLE_MASK                                                                        0x00000010L
+#define MMSCH_CNTL__MMSCH_IRQ_ERR_MASK                                                                        0x000001E0L
+#define MMSCH_CNTL__MMSCH_NACK_INTR_EN_MASK                                                                   0x00000200L
+#define MMSCH_CNTL__MMSCH_DB_BUSY_INTR_EN_MASK                                                                0x00000400L
+#define MMSCH_CNTL__PRB_TIMEOUT_VAL_MASK                                                                      0x0FF00000L
+#define MMSCH_CNTL__TIMEOUT_DIS_MASK                                                                          0x10000000L
+#define MMSCH_CNTL__MMSCH_IDLE_MASK                                                                           0x20000000L
+//MMSCH_NONCACHE_OFFSET0
+#define MMSCH_NONCACHE_OFFSET0__OFFSET__SHIFT                                                                 0x0
+#define MMSCH_NONCACHE_OFFSET0__OFFSET_MASK                                                                   0x0FFFFFFFL
+//MMSCH_NONCACHE_SIZE0
+#define MMSCH_NONCACHE_SIZE0__SIZE__SHIFT                                                                     0x0
+#define MMSCH_NONCACHE_SIZE0__SIZE_MASK                                                                       0x00FFFFFFL
+//MMSCH_NONCACHE_OFFSET1
+#define MMSCH_NONCACHE_OFFSET1__OFFSET__SHIFT                                                                 0x0
+#define MMSCH_NONCACHE_OFFSET1__OFFSET_MASK                                                                   0x0FFFFFFFL
+//MMSCH_NONCACHE_SIZE1
+#define MMSCH_NONCACHE_SIZE1__SIZE__SHIFT                                                                     0x0
+#define MMSCH_NONCACHE_SIZE1__SIZE_MASK                                                                       0x00FFFFFFL
+//MMSCH_PROC_STATE1
+#define MMSCH_PROC_STATE1__PC__SHIFT                                                                          0x0
+#define MMSCH_PROC_STATE1__PC_MASK                                                                            0xFFFFFFFFL
+//MMSCH_LAST_MC_ADDR
+#define MMSCH_LAST_MC_ADDR__MC_ADDR__SHIFT                                                                    0x0
+#define MMSCH_LAST_MC_ADDR__RW__SHIFT                                                                         0x1f
+#define MMSCH_LAST_MC_ADDR__MC_ADDR_MASK                                                                      0x0FFFFFFFL
+#define MMSCH_LAST_MC_ADDR__RW_MASK                                                                           0x80000000L
+//MMSCH_LAST_MEM_ACCESS_HI
+#define MMSCH_LAST_MEM_ACCESS_HI__PROC_CMD__SHIFT                                                             0x0
+#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_RPTR__SHIFT                                                            0x8
+#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_WPTR__SHIFT                                                            0xc
+#define MMSCH_LAST_MEM_ACCESS_HI__PROC_CMD_MASK                                                               0x00000007L
+#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_RPTR_MASK                                                              0x00000700L
+#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_WPTR_MASK                                                              0x00007000L
+//MMSCH_LAST_MEM_ACCESS_LO
+#define MMSCH_LAST_MEM_ACCESS_LO__PROC_ADDR__SHIFT                                                            0x0
+#define MMSCH_LAST_MEM_ACCESS_LO__PROC_ADDR_MASK                                                              0xFFFFFFFFL
+//MMSCH_IOV_ACTIVE_FCN_ID
+#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_VF_ID__SHIFT                                                          0x0
+#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_PF_VF__SHIFT                                                          0x1f
+#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_VF_ID_MASK                                                            0x0000001FL
+#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_PF_VF_MASK                                                            0x80000000L
+//MMSCH_SCRATCH_0
+#define MMSCH_SCRATCH_0__SCRATCH_0__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_0__SCRATCH_0_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_1
+#define MMSCH_SCRATCH_1__SCRATCH_1__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_1__SCRATCH_1_MASK                                                                       0xFFFFFFFFL
+//MMSCH_GPUIOV_SCH_BLOCK_0
+#define MMSCH_GPUIOV_SCH_BLOCK_0__ID__SHIFT                                                                   0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_0__VERSION__SHIFT                                                              0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_0__SIZE__SHIFT                                                                 0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_0__ID_MASK                                                                     0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_0__VERSION_MASK                                                                0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_0__SIZE_MASK                                                                   0x0000FF00L
+//MMSCH_GPUIOV_CMD_CONTROL_0
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_TYPE__SHIFT                                                           0x0
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE__SHIFT                                                        0x4
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_INTR_EN__SHIFT                                                0x5
+#define MMSCH_GPUIOV_CMD_CONTROL_0__VM_BUSY_INTR_EN__SHIFT                                                    0x6
+#define MMSCH_GPUIOV_CMD_CONTROL_0__FUNCTINO_ID__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_CMD_CONTROL_0__NEXT_FUNCTINO_ID__SHIFT                                                   0x10
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_TYPE_MASK                                                             0x0000000FL
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_MASK                                                          0x00000010L
+#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_INTR_EN_MASK                                                  0x00000020L
+#define MMSCH_GPUIOV_CMD_CONTROL_0__VM_BUSY_INTR_EN_MASK                                                      0x00000040L
+#define MMSCH_GPUIOV_CMD_CONTROL_0__FUNCTINO_ID_MASK                                                          0x0000FF00L
+#define MMSCH_GPUIOV_CMD_CONTROL_0__NEXT_FUNCTINO_ID_MASK                                                     0x00FF0000L
+//MMSCH_GPUIOV_CMD_STATUS_0
+#define MMSCH_GPUIOV_CMD_STATUS_0__CMD_STATUS__SHIFT                                                          0x0
+#define MMSCH_GPUIOV_CMD_STATUS_0__CMD_STATUS_MASK                                                            0x0000000FL
+//MMSCH_GPUIOV_VM_BUSY_STATUS_0
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_0__BUSY__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_0__BUSY_MASK                                                              0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCNS_0
+#define MMSCH_GPUIOV_ACTIVE_FCNS_0__ACTIVE_FCNS__SHIFT                                                        0x0
+#define MMSCH_GPUIOV_ACTIVE_FCNS_0__ACTIVE_FCNS_MASK                                                          0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID__SHIFT                                                               0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_STATUS__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_MASK                                                                 0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_STATUS_MASK                                                          0x00000F00L
+//MMSCH_GPUIOV_DW6_0
+#define MMSCH_GPUIOV_DW6_0__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW6_0__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW7_0
+#define MMSCH_GPUIOV_DW7_0__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW7_0__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW8_0
+#define MMSCH_GPUIOV_DW8_0__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW8_0__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_SCH_BLOCK_1
+#define MMSCH_GPUIOV_SCH_BLOCK_1__ID__SHIFT                                                                   0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_1__VERSION__SHIFT                                                              0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_1__SIZE__SHIFT                                                                 0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_1__ID_MASK                                                                     0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_1__VERSION_MASK                                                                0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_1__SIZE_MASK                                                                   0x0000FF00L
+//MMSCH_GPUIOV_CMD_CONTROL_1
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_TYPE__SHIFT                                                           0x0
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE__SHIFT                                                        0x4
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_INTR_EN__SHIFT                                                0x5
+#define MMSCH_GPUIOV_CMD_CONTROL_1__VM_BUSY_INTR_EN__SHIFT                                                    0x6
+#define MMSCH_GPUIOV_CMD_CONTROL_1__FUNCTINO_ID__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_CMD_CONTROL_1__NEXT_FUNCTINO_ID__SHIFT                                                   0x10
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_TYPE_MASK                                                             0x0000000FL
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_MASK                                                          0x00000010L
+#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_INTR_EN_MASK                                                  0x00000020L
+#define MMSCH_GPUIOV_CMD_CONTROL_1__VM_BUSY_INTR_EN_MASK                                                      0x00000040L
+#define MMSCH_GPUIOV_CMD_CONTROL_1__FUNCTINO_ID_MASK                                                          0x0000FF00L
+#define MMSCH_GPUIOV_CMD_CONTROL_1__NEXT_FUNCTINO_ID_MASK                                                     0x00FF0000L
+//MMSCH_GPUIOV_CMD_STATUS_1
+#define MMSCH_GPUIOV_CMD_STATUS_1__CMD_STATUS__SHIFT                                                          0x0
+#define MMSCH_GPUIOV_CMD_STATUS_1__CMD_STATUS_MASK                                                            0x0000000FL
+//MMSCH_GPUIOV_VM_BUSY_STATUS_1
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_1__BUSY__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_1__BUSY_MASK                                                              0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCNS_1
+#define MMSCH_GPUIOV_ACTIVE_FCNS_1__ACTIVE_FCNS__SHIFT                                                        0x0
+#define MMSCH_GPUIOV_ACTIVE_FCNS_1__ACTIVE_FCNS_MASK                                                          0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_1
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID__SHIFT                                                               0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_STATUS__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_MASK                                                                 0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_STATUS_MASK                                                          0x00000F00L
+//MMSCH_GPUIOV_DW6_1
+#define MMSCH_GPUIOV_DW6_1__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW6_1__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW7_1
+#define MMSCH_GPUIOV_DW7_1__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW7_1__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW8_1
+#define MMSCH_GPUIOV_DW8_1__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW8_1__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_CNTXT
+#define MMSCH_GPUIOV_CNTXT__CNTXT_SIZE__SHIFT                                                                 0x0
+#define MMSCH_GPUIOV_CNTXT__CNTXT_LOCATION__SHIFT                                                             0x7
+#define MMSCH_GPUIOV_CNTXT__CNTXT_OFFSET__SHIFT                                                               0xa
+#define MMSCH_GPUIOV_CNTXT__CNTXT_SIZE_MASK                                                                   0x0000007FL
+#define MMSCH_GPUIOV_CNTXT__CNTXT_LOCATION_MASK                                                               0x00000080L
+#define MMSCH_GPUIOV_CNTXT__CNTXT_OFFSET_MASK                                                                 0xFFFFFC00L
+//MMSCH_SCRATCH_2
+#define MMSCH_SCRATCH_2__SCRATCH_2__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_2__SCRATCH_2_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_3
+#define MMSCH_SCRATCH_3__SCRATCH_3__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_3__SCRATCH_3_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_4
+#define MMSCH_SCRATCH_4__SCRATCH_4__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_4__SCRATCH_4_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_5
+#define MMSCH_SCRATCH_5__SCRATCH_5__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_5__SCRATCH_5_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_6
+#define MMSCH_SCRATCH_6__SCRATCH_6__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_6__SCRATCH_6_MASK                                                                       0xFFFFFFFFL
+//MMSCH_SCRATCH_7
+#define MMSCH_SCRATCH_7__SCRATCH_7__SHIFT                                                                     0x0
+#define MMSCH_SCRATCH_7__SCRATCH_7_MASK                                                                       0xFFFFFFFFL
+//MMSCH_VFID_FIFO_HEAD_0
+#define MMSCH_VFID_FIFO_HEAD_0__HEAD__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_HEAD_0__HEAD_MASK                                                                     0x0000003FL
+//MMSCH_VFID_FIFO_TAIL_0
+#define MMSCH_VFID_FIFO_TAIL_0__TAIL__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_TAIL_0__TAIL_MASK                                                                     0x0000003FL
+//MMSCH_VFID_FIFO_HEAD_1
+#define MMSCH_VFID_FIFO_HEAD_1__HEAD__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_HEAD_1__HEAD_MASK                                                                     0x0000003FL
+//MMSCH_VFID_FIFO_TAIL_1
+#define MMSCH_VFID_FIFO_TAIL_1__TAIL__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_TAIL_1__TAIL_MASK                                                                     0x0000003FL
+//MMSCH_NACK_STATUS
+#define MMSCH_NACK_STATUS__WR_NACK_STATUS__SHIFT                                                              0x0
+#define MMSCH_NACK_STATUS__RD_NACK_STATUS__SHIFT                                                              0x2
+#define MMSCH_NACK_STATUS__WR_NACK_STATUS_MASK                                                                0x00000003L
+#define MMSCH_NACK_STATUS__RD_NACK_STATUS_MASK                                                                0x0000000CL
+//MMSCH_VF_MAILBOX0_DATA
+#define MMSCH_VF_MAILBOX0_DATA__DATA__SHIFT                                                                   0x0
+#define MMSCH_VF_MAILBOX0_DATA__DATA_MASK                                                                     0xFFFFFFFFL
+//MMSCH_VF_MAILBOX1_DATA
+#define MMSCH_VF_MAILBOX1_DATA__DATA__SHIFT                                                                   0x0
+#define MMSCH_VF_MAILBOX1_DATA__DATA_MASK                                                                     0xFFFFFFFFL
+//MMSCH_GPUIOV_SCH_BLOCK_IP_0
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__ID__SHIFT                                                                0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__VERSION__SHIFT                                                           0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__SIZE__SHIFT                                                              0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__ID_MASK                                                                  0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__VERSION_MASK                                                             0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__SIZE_MASK                                                                0x0000FF00L
+//MMSCH_GPUIOV_CMD_STATUS_IP_0
+#define MMSCH_GPUIOV_CMD_STATUS_IP_0__CMD_STATUS__SHIFT                                                       0x0
+#define MMSCH_GPUIOV_CMD_STATUS_IP_0__CMD_STATUS_MASK                                                         0x0000000FL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_STATUS__SHIFT                                                     0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_MASK                                                              0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_STATUS_MASK                                                       0x00000F00L
+//MMSCH_GPUIOV_SCH_BLOCK_IP_1
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__ID__SHIFT                                                                0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__VERSION__SHIFT                                                           0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__SIZE__SHIFT                                                              0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__ID_MASK                                                                  0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__VERSION_MASK                                                             0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__SIZE_MASK                                                                0x0000FF00L
+//MMSCH_GPUIOV_CMD_STATUS_IP_1
+#define MMSCH_GPUIOV_CMD_STATUS_IP_1__CMD_STATUS__SHIFT                                                       0x0
+#define MMSCH_GPUIOV_CMD_STATUS_IP_1__CMD_STATUS_MASK                                                         0x0000000FL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_STATUS__SHIFT                                                     0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_MASK                                                              0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_STATUS_MASK                                                       0x00000F00L
+//MMSCH_GPUIOV_CNTXT_IP
+#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_SIZE__SHIFT                                                              0x0
+#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_LOCATION__SHIFT                                                          0x7
+#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_SIZE_MASK                                                                0x0000007FL
+#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_LOCATION_MASK                                                            0x00000080L
+//MMSCH_GPUIOV_SCH_BLOCK_2
+#define MMSCH_GPUIOV_SCH_BLOCK_2__ID__SHIFT                                                                   0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_2__VERSION__SHIFT                                                              0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_2__SIZE__SHIFT                                                                 0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_2__ID_MASK                                                                     0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_2__VERSION_MASK                                                                0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_2__SIZE_MASK                                                                   0x0000FF00L
+//MMSCH_GPUIOV_CMD_CONTROL_2
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_TYPE__SHIFT                                                           0x0
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE__SHIFT                                                        0x4
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_INTR_EN__SHIFT                                                0x5
+#define MMSCH_GPUIOV_CMD_CONTROL_2__VM_BUSY_INTR_EN__SHIFT                                                    0x6
+#define MMSCH_GPUIOV_CMD_CONTROL_2__FUNCTINO_ID__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_CMD_CONTROL_2__NEXT_FUNCTINO_ID__SHIFT                                                   0x10
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_TYPE_MASK                                                             0x0000000FL
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_MASK                                                          0x00000010L
+#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_INTR_EN_MASK                                                  0x00000020L
+#define MMSCH_GPUIOV_CMD_CONTROL_2__VM_BUSY_INTR_EN_MASK                                                      0x00000040L
+#define MMSCH_GPUIOV_CMD_CONTROL_2__FUNCTINO_ID_MASK                                                          0x0000FF00L
+#define MMSCH_GPUIOV_CMD_CONTROL_2__NEXT_FUNCTINO_ID_MASK                                                     0x00FF0000L
+//MMSCH_GPUIOV_CMD_STATUS_2
+#define MMSCH_GPUIOV_CMD_STATUS_2__CMD_STATUS__SHIFT                                                          0x0
+#define MMSCH_GPUIOV_CMD_STATUS_2__CMD_STATUS_MASK                                                            0x0000000FL
+//MMSCH_GPUIOV_VM_BUSY_STATUS_2
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_2__BUSY__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_VM_BUSY_STATUS_2__BUSY_MASK                                                              0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCNS_2
+#define MMSCH_GPUIOV_ACTIVE_FCNS_2__ACTIVE_FCNS__SHIFT                                                        0x0
+#define MMSCH_GPUIOV_ACTIVE_FCNS_2__ACTIVE_FCNS_MASK                                                          0xFFFFFFFFL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_2
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID__SHIFT                                                               0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_STATUS__SHIFT                                                        0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_MASK                                                                 0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_STATUS_MASK                                                          0x00000F00L
+//MMSCH_GPUIOV_DW6_2
+#define MMSCH_GPUIOV_DW6_2__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW6_2__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW7_2
+#define MMSCH_GPUIOV_DW7_2__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW7_2__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_DW8_2
+#define MMSCH_GPUIOV_DW8_2__DATA__SHIFT                                                                       0x0
+#define MMSCH_GPUIOV_DW8_2__DATA_MASK                                                                         0xFFFFFFFFL
+//MMSCH_GPUIOV_SCH_BLOCK_IP_2
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__ID__SHIFT                                                                0x0
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__VERSION__SHIFT                                                           0x4
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__SIZE__SHIFT                                                              0x8
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__ID_MASK                                                                  0x0000000FL
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__VERSION_MASK                                                             0x000000F0L
+#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__SIZE_MASK                                                                0x0000FF00L
+//MMSCH_GPUIOV_CMD_STATUS_IP_2
+#define MMSCH_GPUIOV_CMD_STATUS_IP_2__CMD_STATUS__SHIFT                                                       0x0
+#define MMSCH_GPUIOV_CMD_STATUS_IP_2__CMD_STATUS_MASK                                                         0x0000000FL
+//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID__SHIFT                                                            0x0
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_STATUS__SHIFT                                                     0x8
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_MASK                                                              0x000000FFL
+#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_STATUS_MASK                                                       0x00000F00L
+//MMSCH_VFID_FIFO_HEAD_2
+#define MMSCH_VFID_FIFO_HEAD_2__HEAD__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_HEAD_2__HEAD_MASK                                                                     0x0000003FL
+//MMSCH_VFID_FIFO_TAIL_2
+#define MMSCH_VFID_FIFO_TAIL_2__TAIL__SHIFT                                                                   0x0
+#define MMSCH_VFID_FIFO_TAIL_2__TAIL_MASK                                                                     0x0000003FL
+//MMSCH_VM_BUSY_STATUS_0
+#define MMSCH_VM_BUSY_STATUS_0__BUSY__SHIFT                                                                   0x0
+#define MMSCH_VM_BUSY_STATUS_0__BUSY_MASK                                                                     0xFFFFFFFFL
+//MMSCH_VM_BUSY_STATUS_1
+#define MMSCH_VM_BUSY_STATUS_1__BUSY__SHIFT                                                                   0x0
+#define MMSCH_VM_BUSY_STATUS_1__BUSY_MASK                                                                     0xFFFFFFFFL
+//MMSCH_VM_BUSY_STATUS_2
+#define MMSCH_VM_BUSY_STATUS_2__BUSY__SHIFT                                                                   0x0
+#define MMSCH_VM_BUSY_STATUS_2__BUSY_MASK                                                                     0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_slmi_adpdec
+//UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW
+#define UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW__BITS_31_0__SHIFT                                                     0x0
+#define UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW__BITS_31_0_MASK                                                       0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH
+#define UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                                   0x0
+#define UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH__BITS_63_32_MASK                                                     0xFFFFFFFFL
+//UVD_LMI_MMSCH_NC_VMID
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC0_VMID__SHIFT                                                          0x0
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC1_VMID__SHIFT                                                          0x4
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC2_VMID__SHIFT                                                          0x8
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC3_VMID__SHIFT                                                          0xc
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC4_VMID__SHIFT                                                          0x10
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC5_VMID__SHIFT                                                          0x14
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC6_VMID__SHIFT                                                          0x18
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC7_VMID__SHIFT                                                          0x1c
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC0_VMID_MASK                                                            0x0000000FL
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC1_VMID_MASK                                                            0x000000F0L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC2_VMID_MASK                                                            0x00000F00L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC3_VMID_MASK                                                            0x0000F000L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC4_VMID_MASK                                                            0x000F0000L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC5_VMID_MASK                                                            0x00F00000L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC6_VMID_MASK                                                            0x0F000000L
+#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC7_VMID_MASK                                                            0xF0000000L
+//UVD_LMI_MMSCH_CTRL
+#define UVD_LMI_MMSCH_CTRL__MMSCH_DATA_COHERENCY_EN__SHIFT                                                    0x0
+#define UVD_LMI_MMSCH_CTRL__MMSCH_VM__SHIFT                                                                   0x1
+#define UVD_LMI_MMSCH_CTRL__PRIV_CLIENT_MMSCH__SHIFT                                                          0x2
+#define UVD_LMI_MMSCH_CTRL__MMSCH_R_MC_SWAP__SHIFT                                                            0x3
+#define UVD_LMI_MMSCH_CTRL__MMSCH_W_MC_SWAP__SHIFT                                                            0x5
+#define UVD_LMI_MMSCH_CTRL__MMSCH_RD__SHIFT                                                                   0x7
+#define UVD_LMI_MMSCH_CTRL__MMSCH_WR__SHIFT                                                                   0x9
+#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_DROP__SHIFT                                                              0xb
+#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_DROP__SHIFT                                                              0xc
+#define UVD_LMI_MMSCH_CTRL__MMSCH_DATA_COHERENCY_EN_MASK                                                      0x00000001L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_VM_MASK                                                                     0x00000002L
+#define UVD_LMI_MMSCH_CTRL__PRIV_CLIENT_MMSCH_MASK                                                            0x00000004L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_R_MC_SWAP_MASK                                                              0x00000018L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_W_MC_SWAP_MASK                                                              0x00000060L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_MASK                                                                     0x00000180L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_MASK                                                                     0x00000600L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_DROP_MASK                                                                0x00000800L
+#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_DROP_MASK                                                                0x00001000L
+//UVD_MMSCH_LMI_STATUS
+#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_LEN_INT__SHIFT                                        0x0
+#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_ADR_ALIGN_INT__SHIFT                                  0x1
+#define UVD_MMSCH_LMI_STATUS__MMSCH_LMI_WRITE_CLEAN__SHIFT                                                    0x2
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_LEN__SHIFT                                                        0x4
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_ADDR_LSBS__SHIFT                                                  0x8
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_AWRITE__SHIFT                                                     0xc
+#define UVD_MMSCH_LMI_STATUS__MMSCH_RD_CLEAN__SHIFT                                                           0xd
+#define UVD_MMSCH_LMI_STATUS__MMSCH_WR_CLEAN__SHIFT                                                           0xe
+#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_LEN_INT_MASK                                          0x00000001L
+#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_ADR_ALIGN_INT_MASK                                    0x00000002L
+#define UVD_MMSCH_LMI_STATUS__MMSCH_LMI_WRITE_CLEAN_MASK                                                      0x00000004L
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_LEN_MASK                                                          0x000000F0L
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_ADDR_LSBS_MASK                                                    0x00000700L
+#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_AWRITE_MASK                                                       0x00001000L
+#define UVD_MMSCH_LMI_STATUS__MMSCH_RD_CLEAN_MASK                                                             0x00002000L
+#define UVD_MMSCH_LMI_STATUS__MMSCH_WR_CLEAN_MASK                                                             0x00004000L
+//VCN_RAS_CNTL_MMSCH
+#define VCN_RAS_CNTL_MMSCH__MMSCH_FATAL_ERROR_EN__SHIFT                                                       0x1
+#define VCN_RAS_CNTL_MMSCH__MMSCH_PMI_EN__SHIFT                                                               0x5
+#define VCN_RAS_CNTL_MMSCH__MMSCH_REARM__SHIFT                                                                0x9
+#define VCN_RAS_CNTL_MMSCH__MMSCH_READY__SHIFT                                                                0x11
+#define VCN_RAS_CNTL_MMSCH__MMSCH_FATAL_ERROR_EN_MASK                                                         0x00000002L
+#define VCN_RAS_CNTL_MMSCH__MMSCH_PMI_EN_MASK                                                                 0x00000020L
+#define VCN_RAS_CNTL_MMSCH__MMSCH_REARM_MASK                                                                  0x00000200L
+#define VCN_RAS_CNTL_MMSCH__MMSCH_READY_MASK                                                                  0x00020000L
+
+
+// addressBlock: aid_uvd0_uvd_jrbc1_uvd_jrbc_dec
+//UVD_JRBC1_UVD_JRBC_RB_WPTR
+#define UVD_JRBC1_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC1_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC1_UVD_JRBC_RB_CNTL
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC1_UVD_JRBC_IB_SIZE
+#define UVD_JRBC1_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC1_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC1_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC1_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC1_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC1_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC1_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC1_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC1_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC1_UVD_JRBC_STATUS
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC1_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC1_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC1_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC1_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC1_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC1_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC1_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC1_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC1_UVD_JRBC_RB_RPTR
+#define UVD_JRBC1_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC1_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC1_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC1_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC1_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC1_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC1_UVD_JRBC_RB_SIZE
+#define UVD_JRBC1_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC1_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC1_UVD_JRBC_SCRATCH0
+#define UVD_JRBC1_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC1_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc2_uvd_jrbc_dec
+//UVD_JRBC2_UVD_JRBC_RB_WPTR
+#define UVD_JRBC2_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC2_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC2_UVD_JRBC_RB_CNTL
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC2_UVD_JRBC_IB_SIZE
+#define UVD_JRBC2_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC2_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC2_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC2_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC2_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC2_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC2_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC2_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC2_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC2_UVD_JRBC_STATUS
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC2_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC2_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC2_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC2_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC2_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC2_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC2_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC2_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC2_UVD_JRBC_RB_RPTR
+#define UVD_JRBC2_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC2_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC2_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC2_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC2_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC2_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC2_UVD_JRBC_RB_SIZE
+#define UVD_JRBC2_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC2_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC2_UVD_JRBC_SCRATCH0
+#define UVD_JRBC2_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC2_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc3_uvd_jrbc_dec
+//UVD_JRBC3_UVD_JRBC_RB_WPTR
+#define UVD_JRBC3_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC3_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC3_UVD_JRBC_RB_CNTL
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC3_UVD_JRBC_IB_SIZE
+#define UVD_JRBC3_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC3_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC3_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC3_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC3_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC3_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC3_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC3_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC3_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC3_UVD_JRBC_STATUS
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC3_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC3_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC3_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC3_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC3_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC3_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC3_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC3_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC3_UVD_JRBC_RB_RPTR
+#define UVD_JRBC3_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC3_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC3_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC3_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC3_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC3_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC3_UVD_JRBC_RB_SIZE
+#define UVD_JRBC3_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC3_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC3_UVD_JRBC_SCRATCH0
+#define UVD_JRBC3_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC3_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc4_uvd_jrbc_dec
+//UVD_JRBC4_UVD_JRBC_RB_WPTR
+#define UVD_JRBC4_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC4_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC4_UVD_JRBC_RB_CNTL
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC4_UVD_JRBC_IB_SIZE
+#define UVD_JRBC4_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC4_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC4_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC4_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC4_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC4_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC4_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC4_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC4_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC4_UVD_JRBC_STATUS
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC4_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC4_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC4_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC4_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC4_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC4_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC4_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC4_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC4_UVD_JRBC_RB_RPTR
+#define UVD_JRBC4_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC4_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC4_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC4_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC4_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC4_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC4_UVD_JRBC_RB_SIZE
+#define UVD_JRBC4_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC4_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC4_UVD_JRBC_SCRATCH0
+#define UVD_JRBC4_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC4_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc5_uvd_jrbc_dec
+//UVD_JRBC5_UVD_JRBC_RB_WPTR
+#define UVD_JRBC5_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC5_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC5_UVD_JRBC_RB_CNTL
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC5_UVD_JRBC_IB_SIZE
+#define UVD_JRBC5_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC5_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC5_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC5_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC5_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC5_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC5_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC5_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC5_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC5_UVD_JRBC_STATUS
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC5_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC5_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC5_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC5_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC5_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC5_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC5_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC5_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC5_UVD_JRBC_RB_RPTR
+#define UVD_JRBC5_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC5_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC5_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC5_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC5_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC5_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC5_UVD_JRBC_RB_SIZE
+#define UVD_JRBC5_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC5_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC5_UVD_JRBC_SCRATCH0
+#define UVD_JRBC5_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC5_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc6_uvd_jrbc_dec
+//UVD_JRBC6_UVD_JRBC_RB_WPTR
+#define UVD_JRBC6_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC6_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC6_UVD_JRBC_RB_CNTL
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC6_UVD_JRBC_IB_SIZE
+#define UVD_JRBC6_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC6_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC6_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC6_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC6_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC6_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC6_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC6_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC6_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC6_UVD_JRBC_STATUS
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC6_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC6_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC6_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC6_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC6_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC6_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC6_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC6_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC6_UVD_JRBC_RB_RPTR
+#define UVD_JRBC6_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC6_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC6_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC6_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC6_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC6_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC6_UVD_JRBC_RB_SIZE
+#define UVD_JRBC6_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC6_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC6_UVD_JRBC_SCRATCH0
+#define UVD_JRBC6_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC6_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jrbc7_uvd_jrbc_dec
+//UVD_JRBC7_UVD_JRBC_RB_WPTR
+#define UVD_JRBC7_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT                                                            0x4
+#define UVD_JRBC7_UVD_JRBC_RB_WPTR__RB_WPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC7_UVD_JRBC_RB_CNTL
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT                                                        0x0
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT                                                      0x1
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT                                                 0x4
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK                                                          0x00000001L
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK                                                        0x00000002L
+#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK                                                   0x0007FFF0L
+//UVD_JRBC7_UVD_JRBC_IB_SIZE
+#define UVD_JRBC7_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC7_UVD_JRBC_IB_SIZE__IB_SIZE_MASK                                                              0x007FFFF0L
+//UVD_JRBC7_UVD_JRBC_URGENT_CNTL
+#define UVD_JRBC7_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT                                     0x0
+#define UVD_JRBC7_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK                                       0x00000003L
+//UVD_JRBC7_UVD_JRBC_RB_REF_DATA
+#define UVD_JRBC7_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC7_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC7_UVD_JRBC_SOFT_RESET
+#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__RESET__SHIFT                                                           0x0
+#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT                                               0x11
+#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__RESET_MASK                                                             0x00000001L
+#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK                                                 0x00020000L
+//UVD_JRBC7_UVD_JRBC_STATUS
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT                                                         0x0
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT                                                         0x1
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT                                                      0x2
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT                                              0x3
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT                                                   0x4
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT                                                   0x5
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT                                                      0x6
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT                                              0x7
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT                                                   0x8
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT                                                   0x9
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT                                                      0xa
+#define UVD_JRBC7_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT                                                      0xb
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT                                                      0xc
+#define UVD_JRBC7_UVD_JRBC_STATUS__INT_EN__SHIFT                                                              0x10
+#define UVD_JRBC7_UVD_JRBC_STATUS__INT_ACK__SHIFT                                                             0x11
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_JOB_DONE_MASK                                                           0x00000001L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_JOB_DONE_MASK                                                           0x00000002L
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK                                                        0x00000004L
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK                                                0x00000008L
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK                                                     0x00000010L
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK                                                     0x00000020L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK                                                        0x00000040L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK                                                0x00000080L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK                                                     0x00000100L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK                                                     0x00000200L
+#define UVD_JRBC7_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK                                                        0x00000400L
+#define UVD_JRBC7_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK                                                        0x00000800L
+#define UVD_JRBC7_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK                                                        0x00001000L
+#define UVD_JRBC7_UVD_JRBC_STATUS__INT_EN_MASK                                                                0x00010000L
+#define UVD_JRBC7_UVD_JRBC_STATUS__INT_ACK_MASK                                                               0x00020000L
+//UVD_JRBC7_UVD_JRBC_RB_RPTR
+#define UVD_JRBC7_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT                                                            0x4
+#define UVD_JRBC7_UVD_JRBC_RB_RPTR__RB_RPTR_MASK                                                              0x007FFFF0L
+//UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT                                                 0x0
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT                                               0x10
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT                                               0x18
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK                                                   0x0000FFFFL
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK                                                 0x000F0000L
+#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK                                                 0x03000000L
+//UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE
+#define UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT                                              0x4
+#define UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK                                                0x007FFFF0L
+//UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT                                           0x0
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT                                        0x10
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT                                        0x18
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT                                            0x19
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK                                             0x0000FFFFL
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK                                          0x00FF0000L
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK                                          0x01000000L
+#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK                                              0x02000000L
+//UVD_JRBC7_UVD_JRBC_IB_REF_DATA
+#define UVD_JRBC7_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT                                                       0x0
+#define UVD_JRBC7_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK                                                         0xFFFFFFFFL
+//UVD_JRBC7_UVD_JPEG_PREEMPT_CMD
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT                                                     0x0
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT                                             0x1
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT                                              0x2
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK                                                       0x00000001L
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK                                               0x00000002L
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK                                                0x00000004L
+//UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT                                    0x0
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK                                      0xFFFFFFFFL
+//UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT                                    0x0
+#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK                                      0xFFFFFFFFL
+//UVD_JRBC7_UVD_JRBC_RB_SIZE
+#define UVD_JRBC7_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT                                                            0x4
+#define UVD_JRBC7_UVD_JRBC_RB_SIZE__RB_SIZE_MASK                                                              0x00FFFFF0L
+//UVD_JRBC7_UVD_JRBC_SCRATCH0
+#define UVD_JRBC7_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT                                                          0x0
+#define UVD_JRBC7_UVD_JRBC_SCRATCH0__SCRATCH0_MASK                                                            0xFFFFFFFFL
+
+
+// addressBlock: aid_uvd0_uvd_jmi1_uvd_jmi_dec
+//UVD_JMI1_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI1_UVD_LMI_JRBC_CTRL
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI1_UVD_LMI_JPEG_CTRL
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI1_JPEG_LMI_DROP
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI1_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI1_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI1_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI1_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI1_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI1_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI1_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI1_UVD_LMI_JPEG_VMID
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI1_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI1_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI1_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi2_uvd_jmi_dec
+//UVD_JMI2_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI2_UVD_LMI_JRBC_CTRL
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI2_UVD_LMI_JPEG_CTRL
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI2_JPEG_LMI_DROP
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI2_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI2_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI2_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI2_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI2_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI2_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI2_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI2_UVD_LMI_JPEG_VMID
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI2_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI2_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI2_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi3_uvd_jmi_dec
+//UVD_JMI3_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI3_UVD_LMI_JRBC_CTRL
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI3_UVD_LMI_JPEG_CTRL
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI3_JPEG_LMI_DROP
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI3_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI3_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI3_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI3_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI3_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI3_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI3_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI3_UVD_LMI_JPEG_VMID
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI3_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI3_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI3_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi4_uvd_jmi_dec
+//UVD_JMI4_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI4_UVD_LMI_JRBC_CTRL
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI4_UVD_LMI_JPEG_CTRL
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI4_JPEG_LMI_DROP
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI4_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI4_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI4_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI4_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI4_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI4_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI4_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI4_UVD_LMI_JPEG_VMID
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI4_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI4_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI4_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi5_uvd_jmi_dec
+//UVD_JMI5_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI5_UVD_LMI_JRBC_CTRL
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI5_UVD_LMI_JPEG_CTRL
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI5_JPEG_LMI_DROP
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI5_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI5_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI5_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI5_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI5_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI5_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI5_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI5_UVD_LMI_JPEG_VMID
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI5_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI5_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI5_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi6_uvd_jmi_dec
+//UVD_JMI6_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI6_UVD_LMI_JRBC_CTRL
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI6_UVD_LMI_JPEG_CTRL
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI6_JPEG_LMI_DROP
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI6_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI6_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI6_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI6_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI6_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI6_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI6_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI6_UVD_LMI_JPEG_VMID
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI6_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI6_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI6_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: aid_uvd0_uvd_jmi7_uvd_jmi_dec
+//UVD_JMI7_UVD_JPEG_DEC_PF_CTRL
+#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT                                             0x0
+#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT                                                0x1
+#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK                                               0x00000001L
+#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK                                                  0x00000002L
+//UVD_JMI7_UVD_LMI_JRBC_CTRL
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI7_UVD_LMI_JPEG_CTRL
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT                                                     0x0
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT                                                     0x1
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT                                                       0x4
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT                                                       0x8
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT                                                            0x14
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT                                                            0x16
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK                                                       0x00000001L
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK                                                       0x00000002L
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK                                                         0x000000F0L
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK                                                         0x00000F00L
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK                                                              0x00300000L
+#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK                                                              0x00C00000L
+//UVD_JMI7_JPEG_LMI_DROP
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT                                                           0x0
+#define UVD_JMI7_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT                                                           0x1
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT                                                           0x2
+#define UVD_JMI7_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT                                                           0x3
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT                                                    0x4
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_WR_DROP_MASK                                                             0x00000001L
+#define UVD_JMI7_JPEG_LMI_DROP__JRBC_WR_DROP_MASK                                                             0x00000002L
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_RD_DROP_MASK                                                             0x00000004L
+#define UVD_JMI7_JPEG_LMI_DROP__JRBC_RD_DROP_MASK                                                             0x00000008L
+#define UVD_JMI7_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK                                                      0x00000010L
+//UVD_JMI7_UVD_LMI_JRBC_IB_VMID
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI7_UVD_LMI_JRBC_RB_VMID
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT                                                      0x0
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT                                                      0x4
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT                                                     0x8
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK                                                        0x0000000FL
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK                                                        0x000000F0L
+#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK                                                       0x00000F00L
+//UVD_JMI7_UVD_LMI_JPEG_VMID
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT                                                       0x0
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT                                                       0x4
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT                                               0x8
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK                                                         0x0000000FL
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK                                                         0x000000F0L
+#define UVD_JMI7_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK                                                 0x00000F00L
+//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT                                                       0x0
+#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK                                                         0x0000000FL
+//UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT                                                     0x0
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT                                                     0x2
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT                                              0x4
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT                                              0x6
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT                                              0x8
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT                                              0xa
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT                                             0xc
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT                                                0xe
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT                                                0x10
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK                                                       0x00000003L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK                                                       0x0000000CL
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK                                                0x00000030L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK                                                0x000000C0L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK                                                0x00000300L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK                                                0x00000C00L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK                                               0x00003000L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK                                                  0x0000C000L
+#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK                                                  0x00030000L
+//UVD_JMI7_UVD_JMI_ATOMIC_CNTL
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT                                               0x0
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT                                                 0x1
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT                                                   0x5
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT                                            0x6
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT                                                    0x7
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT                                                   0xb
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK                                                 0x00000001L
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK                                                   0x0000001EL
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK                                                     0x00000020L
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK                                              0x00000040L
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK                                                      0x00000780L
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK                                                     0x00000800L
+//UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                   0x0
+#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                     0xFFFFFFFFL
+//UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                 0x0
+#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                   0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT                                            0x0
+#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK                                              0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                          0x0
+#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK                                            0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT                                           0x0
+#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK                                             0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                         0x0
+#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK                                           0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT                                              0x0
+#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK                                                0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                            0x0
+#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK                                              0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW
+#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT                                       0x0
+#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK                                         0xFFFFFFFFL
+//UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH
+#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT                                     0x0
+#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK                                       0xFFFFFFFFL
+//UVD_JMI7_UVD_JMI_ATOMIC_CNTL2
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT                                                 0x10
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                  0x18
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK                                                   0x00FF0000L
+#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK                                                    0xFF000000L
+
+
+// addressBlock: uvdctxind
+//UVD_CGC_MEM_CTRL
+#define UVD_CGC_MEM_CTRL__LMI_MC_LS_EN__SHIFT                                                                 0x0
+#define UVD_CGC_MEM_CTRL__MPC_LS_EN__SHIFT                                                                    0x1
+#define UVD_CGC_MEM_CTRL__MPRD_LS_EN__SHIFT                                                                   0x2
+#define UVD_CGC_MEM_CTRL__WCB_LS_EN__SHIFT                                                                    0x3
+#define UVD_CGC_MEM_CTRL__UDEC_RE_LS_EN__SHIFT                                                                0x4
+#define UVD_CGC_MEM_CTRL__UDEC_CM_LS_EN__SHIFT                                                                0x5
+#define UVD_CGC_MEM_CTRL__UDEC_IT_LS_EN__SHIFT                                                                0x6
+#define UVD_CGC_MEM_CTRL__UDEC_DB_LS_EN__SHIFT                                                                0x7
+#define UVD_CGC_MEM_CTRL__UDEC_MP_LS_EN__SHIFT                                                                0x8
+#define UVD_CGC_MEM_CTRL__SYS_LS_EN__SHIFT                                                                    0x9
+#define UVD_CGC_MEM_CTRL__VCPU_LS_EN__SHIFT                                                                   0xa
+#define UVD_CGC_MEM_CTRL__MIF_LS_EN__SHIFT                                                                    0xc
+#define UVD_CGC_MEM_CTRL__LCM_LS_EN__SHIFT                                                                    0xd
+#define UVD_CGC_MEM_CTRL__MMSCH_LS_EN__SHIFT                                                                  0xe
+#define UVD_CGC_MEM_CTRL__MPC1_LS_EN__SHIFT                                                                   0xf
+#define UVD_CGC_MEM_CTRL__LS_SET_DELAY__SHIFT                                                                 0x10
+#define UVD_CGC_MEM_CTRL__LS_CLEAR_DELAY__SHIFT                                                               0x14
+#define UVD_CGC_MEM_CTRL__LMI_MC_LS_EN_MASK                                                                   0x00000001L
+#define UVD_CGC_MEM_CTRL__MPC_LS_EN_MASK                                                                      0x00000002L
+#define UVD_CGC_MEM_CTRL__MPRD_LS_EN_MASK                                                                     0x00000004L
+#define UVD_CGC_MEM_CTRL__WCB_LS_EN_MASK                                                                      0x00000008L
+#define UVD_CGC_MEM_CTRL__UDEC_RE_LS_EN_MASK                                                                  0x00000010L
+#define UVD_CGC_MEM_CTRL__UDEC_CM_LS_EN_MASK                                                                  0x00000020L
+#define UVD_CGC_MEM_CTRL__UDEC_IT_LS_EN_MASK                                                                  0x00000040L
+#define UVD_CGC_MEM_CTRL__UDEC_DB_LS_EN_MASK                                                                  0x00000080L
+#define UVD_CGC_MEM_CTRL__UDEC_MP_LS_EN_MASK                                                                  0x00000100L
+#define UVD_CGC_MEM_CTRL__SYS_LS_EN_MASK                                                                      0x00000200L
+#define UVD_CGC_MEM_CTRL__VCPU_LS_EN_MASK                                                                     0x00000400L
+#define UVD_CGC_MEM_CTRL__MIF_LS_EN_MASK                                                                      0x00001000L
+#define UVD_CGC_MEM_CTRL__LCM_LS_EN_MASK                                                                      0x00002000L
+#define UVD_CGC_MEM_CTRL__MMSCH_LS_EN_MASK                                                                    0x00004000L
+#define UVD_CGC_MEM_CTRL__MPC1_LS_EN_MASK                                                                     0x00008000L
+#define UVD_CGC_MEM_CTRL__LS_SET_DELAY_MASK                                                                   0x000F0000L
+#define UVD_CGC_MEM_CTRL__LS_CLEAR_DELAY_MASK                                                                 0x00F00000L
+//UVD_CGC_CTRL2
+#define UVD_CGC_CTRL2__DYN_OCLK_RAMP_EN__SHIFT                                                                0x0
+#define UVD_CGC_CTRL2__DYN_RCLK_RAMP_EN__SHIFT                                                                0x1
+#define UVD_CGC_CTRL2__GATER_DIV_ID__SHIFT                                                                    0x2
+#define UVD_CGC_CTRL2__DYN_OCLK_RAMP_EN_MASK                                                                  0x00000001L
+#define UVD_CGC_CTRL2__DYN_RCLK_RAMP_EN_MASK                                                                  0x00000002L
+#define UVD_CGC_CTRL2__GATER_DIV_ID_MASK                                                                      0x0000001CL
+//UVD_CGC_MEM_DS_CTRL
+#define UVD_CGC_MEM_DS_CTRL__LMI_MC_DS_EN__SHIFT                                                              0x0
+#define UVD_CGC_MEM_DS_CTRL__MPC_DS_EN__SHIFT                                                                 0x1
+#define UVD_CGC_MEM_DS_CTRL__MPRD_DS_EN__SHIFT                                                                0x2
+#define UVD_CGC_MEM_DS_CTRL__WCB_DS_EN__SHIFT                                                                 0x3
+#define UVD_CGC_MEM_DS_CTRL__UDEC_RE_DS_EN__SHIFT                                                             0x4
+#define UVD_CGC_MEM_DS_CTRL__UDEC_CM_DS_EN__SHIFT                                                             0x5
+#define UVD_CGC_MEM_DS_CTRL__UDEC_IT_DS_EN__SHIFT                                                             0x6
+#define UVD_CGC_MEM_DS_CTRL__UDEC_DB_DS_EN__SHIFT                                                             0x7
+#define UVD_CGC_MEM_DS_CTRL__UDEC_MP_DS_EN__SHIFT                                                             0x8
+#define UVD_CGC_MEM_DS_CTRL__SYS_DS_EN__SHIFT                                                                 0x9
+#define UVD_CGC_MEM_DS_CTRL__VCPU_DS_EN__SHIFT                                                                0xa
+#define UVD_CGC_MEM_DS_CTRL__MIF_DS_EN__SHIFT                                                                 0xc
+#define UVD_CGC_MEM_DS_CTRL__LCM_DS_EN__SHIFT                                                                 0xd
+#define UVD_CGC_MEM_DS_CTRL__MMSCH_DS_EN__SHIFT                                                               0xe
+#define UVD_CGC_MEM_DS_CTRL__MPC1_DS_EN__SHIFT                                                                0xf
+#define UVD_CGC_MEM_DS_CTRL__LMI_MC_DS_EN_MASK                                                                0x00000001L
+#define UVD_CGC_MEM_DS_CTRL__MPC_DS_EN_MASK                                                                   0x00000002L
+#define UVD_CGC_MEM_DS_CTRL__MPRD_DS_EN_MASK                                                                  0x00000004L
+#define UVD_CGC_MEM_DS_CTRL__WCB_DS_EN_MASK                                                                   0x00000008L
+#define UVD_CGC_MEM_DS_CTRL__UDEC_RE_DS_EN_MASK                                                               0x00000010L
+#define UVD_CGC_MEM_DS_CTRL__UDEC_CM_DS_EN_MASK                                                               0x00000020L
+#define UVD_CGC_MEM_DS_CTRL__UDEC_IT_DS_EN_MASK                                                               0x00000040L
+#define UVD_CGC_MEM_DS_CTRL__UDEC_DB_DS_EN_MASK                                                               0x00000080L
+#define UVD_CGC_MEM_DS_CTRL__UDEC_MP_DS_EN_MASK                                                               0x00000100L
+#define UVD_CGC_MEM_DS_CTRL__SYS_DS_EN_MASK                                                                   0x00000200L
+#define UVD_CGC_MEM_DS_CTRL__VCPU_DS_EN_MASK                                                                  0x00000400L
+#define UVD_CGC_MEM_DS_CTRL__MIF_DS_EN_MASK                                                                   0x00001000L
+#define UVD_CGC_MEM_DS_CTRL__LCM_DS_EN_MASK                                                                   0x00002000L
+#define UVD_CGC_MEM_DS_CTRL__MMSCH_DS_EN_MASK                                                                 0x00004000L
+#define UVD_CGC_MEM_DS_CTRL__MPC1_DS_EN_MASK                                                                  0x00008000L
+//UVD_CGC_MEM_SD_CTRL
+#define UVD_CGC_MEM_SD_CTRL__LMI_MC_SD_EN__SHIFT                                                              0x0
+#define UVD_CGC_MEM_SD_CTRL__MPC_SD_EN__SHIFT                                                                 0x1
+#define UVD_CGC_MEM_SD_CTRL__MPRD_SD_EN__SHIFT                                                                0x2
+#define UVD_CGC_MEM_SD_CTRL__WCB_SD_EN__SHIFT                                                                 0x3
+#define UVD_CGC_MEM_SD_CTRL__UDEC_RE_SD_EN__SHIFT                                                             0x4
+#define UVD_CGC_MEM_SD_CTRL__UDEC_CM_SD_EN__SHIFT                                                             0x5
+#define UVD_CGC_MEM_SD_CTRL__UDEC_IT_SD_EN__SHIFT                                                             0x6
+#define UVD_CGC_MEM_SD_CTRL__UDEC_DB_SD_EN__SHIFT                                                             0x7
+#define UVD_CGC_MEM_SD_CTRL__UDEC_MP_SD_EN__SHIFT                                                             0x8
+#define UVD_CGC_MEM_SD_CTRL__SYS_SD_EN__SHIFT                                                                 0x9
+#define UVD_CGC_MEM_SD_CTRL__VCPU_SD_EN__SHIFT                                                                0xa
+#define UVD_CGC_MEM_SD_CTRL__MIF_SD_EN__SHIFT                                                                 0xc
+#define UVD_CGC_MEM_SD_CTRL__LCM_SD_EN__SHIFT                                                                 0xd
+#define UVD_CGC_MEM_SD_CTRL__MMSCH_SD_EN__SHIFT                                                               0xe
+#define UVD_CGC_MEM_SD_CTRL__MPC1_SD_EN__SHIFT                                                                0xf
+#define UVD_CGC_MEM_SD_CTRL__LMI_MC_SD_EN_MASK                                                                0x00000001L
+#define UVD_CGC_MEM_SD_CTRL__MPC_SD_EN_MASK                                                                   0x00000002L
+#define UVD_CGC_MEM_SD_CTRL__MPRD_SD_EN_MASK                                                                  0x00000004L
+#define UVD_CGC_MEM_SD_CTRL__WCB_SD_EN_MASK                                                                   0x00000008L
+#define UVD_CGC_MEM_SD_CTRL__UDEC_RE_SD_EN_MASK                                                               0x00000010L
+#define UVD_CGC_MEM_SD_CTRL__UDEC_CM_SD_EN_MASK                                                               0x00000020L
+#define UVD_CGC_MEM_SD_CTRL__UDEC_IT_SD_EN_MASK                                                               0x00000040L
+#define UVD_CGC_MEM_SD_CTRL__UDEC_DB_SD_EN_MASK                                                               0x00000080L
+#define UVD_CGC_MEM_SD_CTRL__UDEC_MP_SD_EN_MASK                                                               0x00000100L
+#define UVD_CGC_MEM_SD_CTRL__SYS_SD_EN_MASK                                                                   0x00000200L
+#define UVD_CGC_MEM_SD_CTRL__VCPU_SD_EN_MASK                                                                  0x00000400L
+#define UVD_CGC_MEM_SD_CTRL__MIF_SD_EN_MASK                                                                   0x00001000L
+#define UVD_CGC_MEM_SD_CTRL__LCM_SD_EN_MASK                                                                   0x00002000L
+#define UVD_CGC_MEM_SD_CTRL__MMSCH_SD_EN_MASK                                                                 0x00004000L
+#define UVD_CGC_MEM_SD_CTRL__MPC1_SD_EN_MASK                                                                  0x00008000L
+//UVD_SW_SCRATCH_00
+#define UVD_SW_SCRATCH_00__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_00__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_01
+#define UVD_SW_SCRATCH_01__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_01__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_02
+#define UVD_SW_SCRATCH_02__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_02__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_03
+#define UVD_SW_SCRATCH_03__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_03__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_04
+#define UVD_SW_SCRATCH_04__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_04__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_05
+#define UVD_SW_SCRATCH_05__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_05__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_06
+#define UVD_SW_SCRATCH_06__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_06__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_07
+#define UVD_SW_SCRATCH_07__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_07__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_08
+#define UVD_SW_SCRATCH_08__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_08__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_09
+#define UVD_SW_SCRATCH_09__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_09__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_10
+#define UVD_SW_SCRATCH_10__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_10__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_11
+#define UVD_SW_SCRATCH_11__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_11__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_12
+#define UVD_SW_SCRATCH_12__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_12__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_13
+#define UVD_SW_SCRATCH_13__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_13__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_14
+#define UVD_SW_SCRATCH_14__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_14__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_SW_SCRATCH_15
+#define UVD_SW_SCRATCH_15__DATA__SHIFT                                                                        0x0
+#define UVD_SW_SCRATCH_15__DATA_MASK                                                                          0xFFFFFFFFL
+//UVD_IH_SEM_CTRL
+#define UVD_IH_SEM_CTRL__IH_STALL_EN__SHIFT                                                                   0x0
+#define UVD_IH_SEM_CTRL__SEM_STALL_EN__SHIFT                                                                  0x1
+#define UVD_IH_SEM_CTRL__IH_STATUS_CLEAN__SHIFT                                                               0x2
+#define UVD_IH_SEM_CTRL__SEM_STATUS_CLEAN__SHIFT                                                              0x3
+#define UVD_IH_SEM_CTRL__IH_VMID__SHIFT                                                                       0x4
+#define UVD_IH_SEM_CTRL__IH_USER_DATA__SHIFT                                                                  0x8
+#define UVD_IH_SEM_CTRL__IH_RINGID__SHIFT                                                                     0x14
+#define UVD_IH_SEM_CTRL__IH_STALL_EN_MASK                                                                     0x00000001L
+#define UVD_IH_SEM_CTRL__SEM_STALL_EN_MASK                                                                    0x00000002L
+#define UVD_IH_SEM_CTRL__IH_STATUS_CLEAN_MASK                                                                 0x00000004L
+#define UVD_IH_SEM_CTRL__SEM_STATUS_CLEAN_MASK                                                                0x00000008L
+#define UVD_IH_SEM_CTRL__IH_VMID_MASK                                                                         0x000000F0L
+#define UVD_IH_SEM_CTRL__IH_USER_DATA_MASK                                                                    0x000FFF00L
+#define UVD_IH_SEM_CTRL__IH_RINGID_MASK                                                                       0x0FF00000L
+
+
+// addressBlock: lmi_adp_indirect
+//UVD_LMI_CRC0
+#define UVD_LMI_CRC0__CRC32__SHIFT                                                                            0x0
+#define UVD_LMI_CRC0__CRC32_MASK                                                                              0xFFFFFFFFL
+//UVD_LMI_CRC1
+#define UVD_LMI_CRC1__CRC32__SHIFT                                                                            0x0
+#define UVD_LMI_CRC1__CRC32_MASK                                                                              0xFFFFFFFFL
+//UVD_LMI_CRC2
+#define UVD_LMI_CRC2__CRC32__SHIFT                                                                            0x0
+#define UVD_LMI_CRC2__CRC32_MASK                                                                              0xFFFFFFFFL
+//UVD_LMI_CRC3
+#define UVD_LMI_CRC3__CRC32__SHIFT                                                                            0x0
+#define UVD_LMI_CRC3__CRC32_MASK                                                                              0xFFFFFFFFL
+//UVD_LMI_CRC10
+#define UVD_LMI_CRC10__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC10__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_CRC11
+#define UVD_LMI_CRC11__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC11__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_CRC12
+#define UVD_LMI_CRC12__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC12__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_CRC13
+#define UVD_LMI_CRC13__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC13__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_CRC14
+#define UVD_LMI_CRC14__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC14__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_CRC15
+#define UVD_LMI_CRC15__CRC32__SHIFT                                                                           0x0
+#define UVD_LMI_CRC15__CRC32_MASK                                                                             0xFFFFFFFFL
+//UVD_LMI_SWAP_CNTL2
+#define UVD_LMI_SWAP_CNTL2__SCPU_R_MC_SWAP__SHIFT                                                             0x0
+#define UVD_LMI_SWAP_CNTL2__SCPU_W_MC_SWAP__SHIFT                                                             0x2
+#define UVD_LMI_SWAP_CNTL2__ATOMIC_MC_SWAP__SHIFT                                                             0x4
+#define UVD_LMI_SWAP_CNTL2__CENC_MC_SWAP__SHIFT                                                               0xc
+#define UVD_LMI_SWAP_CNTL2__FBC_KEY_MC_SWAP__SHIFT                                                            0xe
+#define UVD_LMI_SWAP_CNTL2__SCPU_R_MC_SWAP_MASK                                                               0x00000003L
+#define UVD_LMI_SWAP_CNTL2__SCPU_W_MC_SWAP_MASK                                                               0x0000000CL
+#define UVD_LMI_SWAP_CNTL2__ATOMIC_MC_SWAP_MASK                                                               0x00000FF0L
+#define UVD_LMI_SWAP_CNTL2__CENC_MC_SWAP_MASK                                                                 0x00003000L
+#define UVD_LMI_SWAP_CNTL2__FBC_KEY_MC_SWAP_MASK                                                              0x0000C000L
+//UVD_MEMCHECK_SYS_INT_EN
+#define UVD_MEMCHECK_SYS_INT_EN__RE_ERR_EN__SHIFT                                                             0x0
+#define UVD_MEMCHECK_SYS_INT_EN__IT_ERR_EN__SHIFT                                                             0x1
+#define UVD_MEMCHECK_SYS_INT_EN__MP_ERR_EN__SHIFT                                                             0x2
+#define UVD_MEMCHECK_SYS_INT_EN__DB_ERR_EN__SHIFT                                                             0x3
+#define UVD_MEMCHECK_SYS_INT_EN__DBW_ERR_EN__SHIFT                                                            0x4
+#define UVD_MEMCHECK_SYS_INT_EN__CM_ERR_EN__SHIFT                                                             0x5
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_REF_ERR_EN__SHIFT                                                        0x6
+#define UVD_MEMCHECK_SYS_INT_EN__VCPU_ERR_EN__SHIFT                                                           0x7
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_DBW_ERR_EN__SHIFT                                                        0x8
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_CM_COLOC_ERR_EN__SHIFT                                                   0x9
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP0_ERR_EN__SHIFT                                                       0xa
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP1_ERR_EN__SHIFT                                                       0xb
+#define UVD_MEMCHECK_SYS_INT_EN__SRE_ERR_EN__SHIFT                                                            0xc
+#define UVD_MEMCHECK_SYS_INT_EN__IT_RD_ERR_EN__SHIFT                                                          0xf
+#define UVD_MEMCHECK_SYS_INT_EN__CM_RD_ERR_EN__SHIFT                                                          0x10
+#define UVD_MEMCHECK_SYS_INT_EN__DB_RD_ERR_EN__SHIFT                                                          0x11
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_RD_ERR_EN__SHIFT                                                         0x12
+#define UVD_MEMCHECK_SYS_INT_EN__IDCT_RD_ERR_EN__SHIFT                                                        0x13
+#define UVD_MEMCHECK_SYS_INT_EN__MPC_RD_ERR_EN__SHIFT                                                         0x14
+#define UVD_MEMCHECK_SYS_INT_EN__LBSI_RD_ERR_EN__SHIFT                                                        0x15
+#define UVD_MEMCHECK_SYS_INT_EN__RBC_RD_ERR_EN__SHIFT                                                         0x18
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP2_ERR_EN__SHIFT                                                       0x1b
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP3_ERR_EN__SHIFT                                                       0x1c
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR_ERR_EN__SHIFT                                                       0x1d
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR2_ERR_EN__SHIFT                                                      0x1e
+#define UVD_MEMCHECK_SYS_INT_EN__PREF_ERR_EN__SHIFT                                                           0x1f
+#define UVD_MEMCHECK_SYS_INT_EN__RE_ERR_EN_MASK                                                               0x00000001L
+#define UVD_MEMCHECK_SYS_INT_EN__IT_ERR_EN_MASK                                                               0x00000002L
+#define UVD_MEMCHECK_SYS_INT_EN__MP_ERR_EN_MASK                                                               0x00000004L
+#define UVD_MEMCHECK_SYS_INT_EN__DB_ERR_EN_MASK                                                               0x00000008L
+#define UVD_MEMCHECK_SYS_INT_EN__DBW_ERR_EN_MASK                                                              0x00000010L
+#define UVD_MEMCHECK_SYS_INT_EN__CM_ERR_EN_MASK                                                               0x00000020L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_REF_ERR_EN_MASK                                                          0x00000040L
+#define UVD_MEMCHECK_SYS_INT_EN__VCPU_ERR_EN_MASK                                                             0x00000080L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_DBW_ERR_EN_MASK                                                          0x00000100L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_CM_COLOC_ERR_EN_MASK                                                     0x00000200L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP0_ERR_EN_MASK                                                         0x00000400L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP1_ERR_EN_MASK                                                         0x00000800L
+#define UVD_MEMCHECK_SYS_INT_EN__SRE_ERR_EN_MASK                                                              0x00001000L
+#define UVD_MEMCHECK_SYS_INT_EN__IT_RD_ERR_EN_MASK                                                            0x00008000L
+#define UVD_MEMCHECK_SYS_INT_EN__CM_RD_ERR_EN_MASK                                                            0x00010000L
+#define UVD_MEMCHECK_SYS_INT_EN__DB_RD_ERR_EN_MASK                                                            0x00020000L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_RD_ERR_EN_MASK                                                           0x00040000L
+#define UVD_MEMCHECK_SYS_INT_EN__IDCT_RD_ERR_EN_MASK                                                          0x00080000L
+#define UVD_MEMCHECK_SYS_INT_EN__MPC_RD_ERR_EN_MASK                                                           0x00100000L
+#define UVD_MEMCHECK_SYS_INT_EN__LBSI_RD_ERR_EN_MASK                                                          0x00200000L
+#define UVD_MEMCHECK_SYS_INT_EN__RBC_RD_ERR_EN_MASK                                                           0x01000000L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP2_ERR_EN_MASK                                                         0x08000000L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP3_ERR_EN_MASK                                                         0x10000000L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR_ERR_EN_MASK                                                         0x20000000L
+#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR2_ERR_EN_MASK                                                        0x40000000L
+#define UVD_MEMCHECK_SYS_INT_EN__PREF_ERR_EN_MASK                                                             0x80000000L
+//UVD_MEMCHECK_SYS_INT_STAT
+#define UVD_MEMCHECK_SYS_INT_STAT__RE_LO_ERR__SHIFT                                                           0x0
+#define UVD_MEMCHECK_SYS_INT_STAT__RE_HI_ERR__SHIFT                                                           0x1
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_LO_ERR__SHIFT                                                           0x2
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_HI_ERR__SHIFT                                                           0x3
+#define UVD_MEMCHECK_SYS_INT_STAT__MP_LO_ERR__SHIFT                                                           0x4
+#define UVD_MEMCHECK_SYS_INT_STAT__MP_HI_ERR__SHIFT                                                           0x5
+#define UVD_MEMCHECK_SYS_INT_STAT__DB_LO_ERR__SHIFT                                                           0x6
+#define UVD_MEMCHECK_SYS_INT_STAT__DB_HI_ERR__SHIFT                                                           0x7
+#define UVD_MEMCHECK_SYS_INT_STAT__DBW_LO_ERR__SHIFT                                                          0x8
+#define UVD_MEMCHECK_SYS_INT_STAT__DBW_HI_ERR__SHIFT                                                          0x9
+#define UVD_MEMCHECK_SYS_INT_STAT__CM_LO_ERR__SHIFT                                                           0xa
+#define UVD_MEMCHECK_SYS_INT_STAT__CM_HI_ERR__SHIFT                                                           0xb
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_LO_ERR__SHIFT                                                      0xc
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_HI_ERR__SHIFT                                                      0xd
+#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_LO_ERR__SHIFT                                                         0xe
+#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_HI_ERR__SHIFT                                                         0xf
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_LO_ERR__SHIFT                                                      0x10
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_HI_ERR__SHIFT                                                      0x11
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_LO_ERR__SHIFT                                                 0x12
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_HI_ERR__SHIFT                                                 0x13
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_LO_ERR__SHIFT                                                     0x14
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_HI_ERR__SHIFT                                                     0x15
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_LO_ERR__SHIFT                                                     0x16
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_HI_ERR__SHIFT                                                     0x17
+#define UVD_MEMCHECK_SYS_INT_STAT__SRE_LO_ERR__SHIFT                                                          0x18
+#define UVD_MEMCHECK_SYS_INT_STAT__SRE_HI_ERR__SHIFT                                                          0x19
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_LO_ERR__SHIFT                                                        0x1e
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_HI_ERR__SHIFT                                                        0x1f
+#define UVD_MEMCHECK_SYS_INT_STAT__RE_LO_ERR_MASK                                                             0x00000001L
+#define UVD_MEMCHECK_SYS_INT_STAT__RE_HI_ERR_MASK                                                             0x00000002L
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_LO_ERR_MASK                                                             0x00000004L
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_HI_ERR_MASK                                                             0x00000008L
+#define UVD_MEMCHECK_SYS_INT_STAT__MP_LO_ERR_MASK                                                             0x00000010L
+#define UVD_MEMCHECK_SYS_INT_STAT__MP_HI_ERR_MASK                                                             0x00000020L
+#define UVD_MEMCHECK_SYS_INT_STAT__DB_LO_ERR_MASK                                                             0x00000040L
+#define UVD_MEMCHECK_SYS_INT_STAT__DB_HI_ERR_MASK                                                             0x00000080L
+#define UVD_MEMCHECK_SYS_INT_STAT__DBW_LO_ERR_MASK                                                            0x00000100L
+#define UVD_MEMCHECK_SYS_INT_STAT__DBW_HI_ERR_MASK                                                            0x00000200L
+#define UVD_MEMCHECK_SYS_INT_STAT__CM_LO_ERR_MASK                                                             0x00000400L
+#define UVD_MEMCHECK_SYS_INT_STAT__CM_HI_ERR_MASK                                                             0x00000800L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_LO_ERR_MASK                                                        0x00001000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_HI_ERR_MASK                                                        0x00002000L
+#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_LO_ERR_MASK                                                           0x00004000L
+#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_HI_ERR_MASK                                                           0x00008000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_LO_ERR_MASK                                                        0x00010000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_HI_ERR_MASK                                                        0x00020000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_LO_ERR_MASK                                                   0x00040000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_HI_ERR_MASK                                                   0x00080000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_LO_ERR_MASK                                                       0x00100000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_HI_ERR_MASK                                                       0x00200000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_LO_ERR_MASK                                                       0x00400000L
+#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_HI_ERR_MASK                                                       0x00800000L
+#define UVD_MEMCHECK_SYS_INT_STAT__SRE_LO_ERR_MASK                                                            0x01000000L
+#define UVD_MEMCHECK_SYS_INT_STAT__SRE_HI_ERR_MASK                                                            0x02000000L
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_LO_ERR_MASK                                                          0x40000000L
+#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_HI_ERR_MASK                                                          0x80000000L
+//UVD_MEMCHECK_SYS_INT_ACK
+#define UVD_MEMCHECK_SYS_INT_ACK__RE_LO_ACK__SHIFT                                                            0x0
+#define UVD_MEMCHECK_SYS_INT_ACK__RE_HI_ACK__SHIFT                                                            0x1
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_LO_ACK__SHIFT                                                            0x2
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_HI_ACK__SHIFT                                                            0x3
+#define UVD_MEMCHECK_SYS_INT_ACK__MP_LO_ACK__SHIFT                                                            0x4
+#define UVD_MEMCHECK_SYS_INT_ACK__MP_HI_ACK__SHIFT                                                            0x5
+#define UVD_MEMCHECK_SYS_INT_ACK__DB_LO_ACK__SHIFT                                                            0x6
+#define UVD_MEMCHECK_SYS_INT_ACK__DB_HI_ACK__SHIFT                                                            0x7
+#define UVD_MEMCHECK_SYS_INT_ACK__DBW_LO_ACK__SHIFT                                                           0x8
+#define UVD_MEMCHECK_SYS_INT_ACK__DBW_HI_ACK__SHIFT                                                           0x9
+#define UVD_MEMCHECK_SYS_INT_ACK__CM_LO_ACK__SHIFT                                                            0xa
+#define UVD_MEMCHECK_SYS_INT_ACK__CM_HI_ACK__SHIFT                                                            0xb
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_LO_ACK__SHIFT                                                       0xc
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_HI_ACK__SHIFT                                                       0xd
+#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_LO_ACK__SHIFT                                                          0xe
+#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_HI_ACK__SHIFT                                                          0xf
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_LO_ACK__SHIFT                                                       0x10
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_HI_ACK__SHIFT                                                       0x11
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_LO_ACK__SHIFT                                                  0x12
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_HI_ACK__SHIFT                                                  0x13
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_LO_ACK__SHIFT                                                      0x14
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_HI_ACK__SHIFT                                                      0x15
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_LO_ACK__SHIFT                                                      0x16
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_HI_ACK__SHIFT                                                      0x17
+#define UVD_MEMCHECK_SYS_INT_ACK__SRE_LO_ACK__SHIFT                                                           0x18
+#define UVD_MEMCHECK_SYS_INT_ACK__SRE_HI_ACK__SHIFT                                                           0x19
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_LO_ACK__SHIFT                                                         0x1e
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_HI_ACK__SHIFT                                                         0x1f
+#define UVD_MEMCHECK_SYS_INT_ACK__RE_LO_ACK_MASK                                                              0x00000001L
+#define UVD_MEMCHECK_SYS_INT_ACK__RE_HI_ACK_MASK                                                              0x00000002L
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_LO_ACK_MASK                                                              0x00000004L
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_HI_ACK_MASK                                                              0x00000008L
+#define UVD_MEMCHECK_SYS_INT_ACK__MP_LO_ACK_MASK                                                              0x00000010L
+#define UVD_MEMCHECK_SYS_INT_ACK__MP_HI_ACK_MASK                                                              0x00000020L
+#define UVD_MEMCHECK_SYS_INT_ACK__DB_LO_ACK_MASK                                                              0x00000040L
+#define UVD_MEMCHECK_SYS_INT_ACK__DB_HI_ACK_MASK                                                              0x00000080L
+#define UVD_MEMCHECK_SYS_INT_ACK__DBW_LO_ACK_MASK                                                             0x00000100L
+#define UVD_MEMCHECK_SYS_INT_ACK__DBW_HI_ACK_MASK                                                             0x00000200L
+#define UVD_MEMCHECK_SYS_INT_ACK__CM_LO_ACK_MASK                                                              0x00000400L
+#define UVD_MEMCHECK_SYS_INT_ACK__CM_HI_ACK_MASK                                                              0x00000800L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_LO_ACK_MASK                                                         0x00001000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_HI_ACK_MASK                                                         0x00002000L
+#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_LO_ACK_MASK                                                            0x00004000L
+#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_HI_ACK_MASK                                                            0x00008000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_LO_ACK_MASK                                                         0x00010000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_HI_ACK_MASK                                                         0x00020000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_LO_ACK_MASK                                                    0x00040000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_HI_ACK_MASK                                                    0x00080000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_LO_ACK_MASK                                                        0x00100000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_HI_ACK_MASK                                                        0x00200000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_LO_ACK_MASK                                                        0x00400000L
+#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_HI_ACK_MASK                                                        0x00800000L
+#define UVD_MEMCHECK_SYS_INT_ACK__SRE_LO_ACK_MASK                                                             0x01000000L
+#define UVD_MEMCHECK_SYS_INT_ACK__SRE_HI_ACK_MASK                                                             0x02000000L
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_LO_ACK_MASK                                                           0x40000000L
+#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_HI_ACK_MASK                                                           0x80000000L
+//UVD_MEMCHECK_VCPU_INT_EN
+#define UVD_MEMCHECK_VCPU_INT_EN__RE_ERR_EN__SHIFT                                                            0x0
+#define UVD_MEMCHECK_VCPU_INT_EN__IT_ERR_EN__SHIFT                                                            0x1
+#define UVD_MEMCHECK_VCPU_INT_EN__MP_ERR_EN__SHIFT                                                            0x2
+#define UVD_MEMCHECK_VCPU_INT_EN__DB_ERR_EN__SHIFT                                                            0x3
+#define UVD_MEMCHECK_VCPU_INT_EN__DBW_ERR_EN__SHIFT                                                           0x4
+#define UVD_MEMCHECK_VCPU_INT_EN__CM_ERR_EN__SHIFT                                                            0x5
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_REF_ERR_EN__SHIFT                                                       0x6
+#define UVD_MEMCHECK_VCPU_INT_EN__VCPU_ERR_EN__SHIFT                                                          0x7
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_DBW_ERR_EN__SHIFT                                                       0x8
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_CM_COLOC_ERR_EN__SHIFT                                                  0x9
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP0_ERR_EN__SHIFT                                                      0xa
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP1_ERR_EN__SHIFT                                                      0xb
+#define UVD_MEMCHECK_VCPU_INT_EN__SRE_ERR_EN__SHIFT                                                           0xc
+#define UVD_MEMCHECK_VCPU_INT_EN__IT_RD_ERR_EN__SHIFT                                                         0xf
+#define UVD_MEMCHECK_VCPU_INT_EN__CM_RD_ERR_EN__SHIFT                                                         0x10
+#define UVD_MEMCHECK_VCPU_INT_EN__DB_RD_ERR_EN__SHIFT                                                         0x11
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_RD_ERR_EN__SHIFT                                                        0x12
+#define UVD_MEMCHECK_VCPU_INT_EN__IDCT_RD_ERR_EN__SHIFT                                                       0x13
+#define UVD_MEMCHECK_VCPU_INT_EN__MPC_RD_ERR_EN__SHIFT                                                        0x14
+#define UVD_MEMCHECK_VCPU_INT_EN__LBSI_RD_ERR_EN__SHIFT                                                       0x15
+#define UVD_MEMCHECK_VCPU_INT_EN__RBC_RD_ERR_EN__SHIFT                                                        0x18
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP2_ERR_EN__SHIFT                                                      0x19
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP3_ERR_EN__SHIFT                                                      0x1a
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR_ERR_EN__SHIFT                                                      0x1b
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR2_ERR_EN__SHIFT                                                     0x1c
+#define UVD_MEMCHECK_VCPU_INT_EN__PREF_ERR_EN__SHIFT                                                          0x1d
+#define UVD_MEMCHECK_VCPU_INT_EN__RE_ERR_EN_MASK                                                              0x00000001L
+#define UVD_MEMCHECK_VCPU_INT_EN__IT_ERR_EN_MASK                                                              0x00000002L
+#define UVD_MEMCHECK_VCPU_INT_EN__MP_ERR_EN_MASK                                                              0x00000004L
+#define UVD_MEMCHECK_VCPU_INT_EN__DB_ERR_EN_MASK                                                              0x00000008L
+#define UVD_MEMCHECK_VCPU_INT_EN__DBW_ERR_EN_MASK                                                             0x00000010L
+#define UVD_MEMCHECK_VCPU_INT_EN__CM_ERR_EN_MASK                                                              0x00000020L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_REF_ERR_EN_MASK                                                         0x00000040L
+#define UVD_MEMCHECK_VCPU_INT_EN__VCPU_ERR_EN_MASK                                                            0x00000080L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_DBW_ERR_EN_MASK                                                         0x00000100L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_CM_COLOC_ERR_EN_MASK                                                    0x00000200L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP0_ERR_EN_MASK                                                        0x00000400L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP1_ERR_EN_MASK                                                        0x00000800L
+#define UVD_MEMCHECK_VCPU_INT_EN__SRE_ERR_EN_MASK                                                             0x00001000L
+#define UVD_MEMCHECK_VCPU_INT_EN__IT_RD_ERR_EN_MASK                                                           0x00008000L
+#define UVD_MEMCHECK_VCPU_INT_EN__CM_RD_ERR_EN_MASK                                                           0x00010000L
+#define UVD_MEMCHECK_VCPU_INT_EN__DB_RD_ERR_EN_MASK                                                           0x00020000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_RD_ERR_EN_MASK                                                          0x00040000L
+#define UVD_MEMCHECK_VCPU_INT_EN__IDCT_RD_ERR_EN_MASK                                                         0x00080000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MPC_RD_ERR_EN_MASK                                                          0x00100000L
+#define UVD_MEMCHECK_VCPU_INT_EN__LBSI_RD_ERR_EN_MASK                                                         0x00200000L
+#define UVD_MEMCHECK_VCPU_INT_EN__RBC_RD_ERR_EN_MASK                                                          0x01000000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP2_ERR_EN_MASK                                                        0x02000000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP3_ERR_EN_MASK                                                        0x04000000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR_ERR_EN_MASK                                                        0x08000000L
+#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR2_ERR_EN_MASK                                                       0x10000000L
+#define UVD_MEMCHECK_VCPU_INT_EN__PREF_ERR_EN_MASK                                                            0x20000000L
+//UVD_MEMCHECK_VCPU_INT_STAT
+#define UVD_MEMCHECK_VCPU_INT_STAT__RE_LO_ERR__SHIFT                                                          0x0
+#define UVD_MEMCHECK_VCPU_INT_STAT__RE_HI_ERR__SHIFT                                                          0x1
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_LO_ERR__SHIFT                                                          0x2
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_HI_ERR__SHIFT                                                          0x3
+#define UVD_MEMCHECK_VCPU_INT_STAT__MP_LO_ERR__SHIFT                                                          0x4
+#define UVD_MEMCHECK_VCPU_INT_STAT__MP_HI_ERR__SHIFT                                                          0x5
+#define UVD_MEMCHECK_VCPU_INT_STAT__DB_LO_ERR__SHIFT                                                          0x6
+#define UVD_MEMCHECK_VCPU_INT_STAT__DB_HI_ERR__SHIFT                                                          0x7
+#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_LO_ERR__SHIFT                                                         0x8
+#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_HI_ERR__SHIFT                                                         0x9
+#define UVD_MEMCHECK_VCPU_INT_STAT__CM_LO_ERR__SHIFT                                                          0xa
+#define UVD_MEMCHECK_VCPU_INT_STAT__CM_HI_ERR__SHIFT                                                          0xb
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_LO_ERR__SHIFT                                                     0xc
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_HI_ERR__SHIFT                                                     0xd
+#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_LO_ERR__SHIFT                                                        0xe
+#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_HI_ERR__SHIFT                                                        0xf
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_LO_ERR__SHIFT                                                     0x10
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_HI_ERR__SHIFT                                                     0x11
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_LO_ERR__SHIFT                                                0x12
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_HI_ERR__SHIFT                                                0x13
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_LO_ERR__SHIFT                                                    0x14
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_HI_ERR__SHIFT                                                    0x15
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_LO_ERR__SHIFT                                                    0x16
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_HI_ERR__SHIFT                                                    0x17
+#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_LO_ERR__SHIFT                                                         0x18
+#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_HI_ERR__SHIFT                                                         0x19
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_LO_ERR__SHIFT                                                       0x1e
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_HI_ERR__SHIFT                                                       0x1f
+#define UVD_MEMCHECK_VCPU_INT_STAT__RE_LO_ERR_MASK                                                            0x00000001L
+#define UVD_MEMCHECK_VCPU_INT_STAT__RE_HI_ERR_MASK                                                            0x00000002L
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_LO_ERR_MASK                                                            0x00000004L
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_HI_ERR_MASK                                                            0x00000008L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MP_LO_ERR_MASK                                                            0x00000010L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MP_HI_ERR_MASK                                                            0x00000020L
+#define UVD_MEMCHECK_VCPU_INT_STAT__DB_LO_ERR_MASK                                                            0x00000040L
+#define UVD_MEMCHECK_VCPU_INT_STAT__DB_HI_ERR_MASK                                                            0x00000080L
+#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_LO_ERR_MASK                                                           0x00000100L
+#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_HI_ERR_MASK                                                           0x00000200L
+#define UVD_MEMCHECK_VCPU_INT_STAT__CM_LO_ERR_MASK                                                            0x00000400L
+#define UVD_MEMCHECK_VCPU_INT_STAT__CM_HI_ERR_MASK                                                            0x00000800L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_LO_ERR_MASK                                                       0x00001000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_HI_ERR_MASK                                                       0x00002000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_LO_ERR_MASK                                                          0x00004000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_HI_ERR_MASK                                                          0x00008000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_LO_ERR_MASK                                                       0x00010000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_HI_ERR_MASK                                                       0x00020000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_LO_ERR_MASK                                                  0x00040000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_HI_ERR_MASK                                                  0x00080000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_LO_ERR_MASK                                                      0x00100000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_HI_ERR_MASK                                                      0x00200000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_LO_ERR_MASK                                                      0x00400000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_HI_ERR_MASK                                                      0x00800000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_LO_ERR_MASK                                                           0x01000000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_HI_ERR_MASK                                                           0x02000000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_LO_ERR_MASK                                                         0x40000000L
+#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_HI_ERR_MASK                                                         0x80000000L
+//UVD_MEMCHECK_VCPU_INT_ACK
+#define UVD_MEMCHECK_VCPU_INT_ACK__RE_LO_ACK__SHIFT                                                           0x0
+#define UVD_MEMCHECK_VCPU_INT_ACK__RE_HI_ACK__SHIFT                                                           0x1
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_LO_ACK__SHIFT                                                           0x2
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_HI_ACK__SHIFT                                                           0x3
+#define UVD_MEMCHECK_VCPU_INT_ACK__MP_LO_ACK__SHIFT                                                           0x4
+#define UVD_MEMCHECK_VCPU_INT_ACK__MP_HI_ACK__SHIFT                                                           0x5
+#define UVD_MEMCHECK_VCPU_INT_ACK__DB_LO_ACK__SHIFT                                                           0x6
+#define UVD_MEMCHECK_VCPU_INT_ACK__DB_HI_ACK__SHIFT                                                           0x7
+#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_LO_ACK__SHIFT                                                          0x8
+#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_HI_ACK__SHIFT                                                          0x9
+#define UVD_MEMCHECK_VCPU_INT_ACK__CM_LO_ACK__SHIFT                                                           0xa
+#define UVD_MEMCHECK_VCPU_INT_ACK__CM_HI_ACK__SHIFT                                                           0xb
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_LO_ACK__SHIFT                                                      0xc
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_HI_ACK__SHIFT                                                      0xd
+#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_LO_ACK__SHIFT                                                         0xe
+#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_HI_ACK__SHIFT                                                         0xf
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_LO_ACK__SHIFT                                                      0x10
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_HI_ACK__SHIFT                                                      0x11
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_LO_ACK__SHIFT                                                 0x12
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_HI_ACK__SHIFT                                                 0x13
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_LO_ACK__SHIFT                                                     0x14
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_HI_ACK__SHIFT                                                     0x15
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_LO_ACK__SHIFT                                                     0x16
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_HI_ACK__SHIFT                                                     0x17
+#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_LO_ACK__SHIFT                                                          0x18
+#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_HI_ACK__SHIFT                                                          0x19
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_LO_ACK__SHIFT                                                        0x1e
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_HI_ACK__SHIFT                                                        0x1f
+#define UVD_MEMCHECK_VCPU_INT_ACK__RE_LO_ACK_MASK                                                             0x00000001L
+#define UVD_MEMCHECK_VCPU_INT_ACK__RE_HI_ACK_MASK                                                             0x00000002L
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_LO_ACK_MASK                                                             0x00000004L
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_HI_ACK_MASK                                                             0x00000008L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MP_LO_ACK_MASK                                                             0x00000010L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MP_HI_ACK_MASK                                                             0x00000020L
+#define UVD_MEMCHECK_VCPU_INT_ACK__DB_LO_ACK_MASK                                                             0x00000040L
+#define UVD_MEMCHECK_VCPU_INT_ACK__DB_HI_ACK_MASK                                                             0x00000080L
+#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_LO_ACK_MASK                                                            0x00000100L
+#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_HI_ACK_MASK                                                            0x00000200L
+#define UVD_MEMCHECK_VCPU_INT_ACK__CM_LO_ACK_MASK                                                             0x00000400L
+#define UVD_MEMCHECK_VCPU_INT_ACK__CM_HI_ACK_MASK                                                             0x00000800L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_LO_ACK_MASK                                                        0x00001000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_HI_ACK_MASK                                                        0x00002000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_LO_ACK_MASK                                                           0x00004000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_HI_ACK_MASK                                                           0x00008000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_LO_ACK_MASK                                                        0x00010000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_HI_ACK_MASK                                                        0x00020000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_LO_ACK_MASK                                                   0x00040000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_HI_ACK_MASK                                                   0x00080000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_LO_ACK_MASK                                                       0x00100000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_HI_ACK_MASK                                                       0x00200000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_LO_ACK_MASK                                                       0x00400000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_HI_ACK_MASK                                                       0x00800000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_LO_ACK_MASK                                                            0x01000000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_HI_ACK_MASK                                                            0x02000000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_LO_ACK_MASK                                                          0x40000000L
+#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_HI_ACK_MASK                                                          0x80000000L
+//UVD_MEMCHECK2_SYS_INT_STAT
+#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_LO_ERR__SHIFT                                                       0x0
+#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_HI_ERR__SHIFT                                                       0x1
+#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_LO_ERR__SHIFT                                                       0x2
+#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_HI_ERR__SHIFT                                                       0x3
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_LO_ERR__SHIFT                                                      0x4
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_HI_ERR__SHIFT                                                      0x5
+#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_LO_ERR__SHIFT                                                     0x6
+#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_HI_ERR__SHIFT                                                     0x7
+#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_LO_ERR__SHIFT                                                      0x8
+#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_HI_ERR__SHIFT                                                      0x9
+#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_LO_ERR__SHIFT                                                     0xa
+#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_HI_ERR__SHIFT                                                     0xb
+#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_LO_ERR__SHIFT                                                      0x10
+#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_HI_ERR__SHIFT                                                      0x11
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_LO_ERR__SHIFT                                                    0x16
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_HI_ERR__SHIFT                                                    0x17
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_LO_ERR__SHIFT                                                    0x18
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_HI_ERR__SHIFT                                                    0x19
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_LO_ERR__SHIFT                                                    0x1a
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_HI_ERR__SHIFT                                                    0x1b
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_LO_ERR__SHIFT                                                   0x1c
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_HI_ERR__SHIFT                                                   0x1d
+#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_LO_ERR__SHIFT                                                        0x1e
+#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_HI_ERR__SHIFT                                                        0x1f
+#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_LO_ERR_MASK                                                         0x00000001L
+#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_HI_ERR_MASK                                                         0x00000002L
+#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_LO_ERR_MASK                                                         0x00000004L
+#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_HI_ERR_MASK                                                         0x00000008L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_LO_ERR_MASK                                                        0x00000010L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_HI_ERR_MASK                                                        0x00000020L
+#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_LO_ERR_MASK                                                       0x00000040L
+#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_HI_ERR_MASK                                                       0x00000080L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_LO_ERR_MASK                                                        0x00000100L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_HI_ERR_MASK                                                        0x00000200L
+#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_LO_ERR_MASK                                                       0x00000400L
+#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_HI_ERR_MASK                                                       0x00000800L
+#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_LO_ERR_MASK                                                        0x00010000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_HI_ERR_MASK                                                        0x00020000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_LO_ERR_MASK                                                      0x00400000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_HI_ERR_MASK                                                      0x00800000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_LO_ERR_MASK                                                      0x01000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_HI_ERR_MASK                                                      0x02000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_LO_ERR_MASK                                                      0x04000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_HI_ERR_MASK                                                      0x08000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_LO_ERR_MASK                                                     0x10000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_HI_ERR_MASK                                                     0x20000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_LO_ERR_MASK                                                          0x40000000L
+#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_HI_ERR_MASK                                                          0x80000000L
+//UVD_MEMCHECK2_SYS_INT_ACK
+#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_LO_ACK__SHIFT                                                        0x0
+#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_HI_ACK__SHIFT                                                        0x1
+#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_LO_ACK__SHIFT                                                        0x2
+#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_HI_ACK__SHIFT                                                        0x3
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_LO_ACK__SHIFT                                                       0x4
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_HI_ACK__SHIFT                                                       0x5
+#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_LO_ACK__SHIFT                                                      0x6
+#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_HI_ACK__SHIFT                                                      0x7
+#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_LO_ACK__SHIFT                                                       0x8
+#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_HI_ACK__SHIFT                                                       0x9
+#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_LO_ACK__SHIFT                                                      0xa
+#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_HI_ACK__SHIFT                                                      0xb
+#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_LO_ACK__SHIFT                                                       0x10
+#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_HI_ACK__SHIFT                                                       0x11
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_LO_ACK__SHIFT                                                     0x16
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_HI_ACK__SHIFT                                                     0x17
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_LO_ACK__SHIFT                                                     0x18
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_HI_ACK__SHIFT                                                     0x19
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_LO_ACK__SHIFT                                                     0x1a
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_HI_ACK__SHIFT                                                     0x1b
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_LO_ACK__SHIFT                                                    0x1c
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_HI_ACK__SHIFT                                                    0x1d
+#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_LO_ACK__SHIFT                                                         0x1e
+#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_HI_ACK__SHIFT                                                         0x1f
+#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_LO_ACK_MASK                                                          0x00000001L
+#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_HI_ACK_MASK                                                          0x00000002L
+#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_LO_ACK_MASK                                                          0x00000004L
+#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_HI_ACK_MASK                                                          0x00000008L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_LO_ACK_MASK                                                         0x00000010L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_HI_ACK_MASK                                                         0x00000020L
+#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_LO_ACK_MASK                                                        0x00000040L
+#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_HI_ACK_MASK                                                        0x00000080L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_LO_ACK_MASK                                                         0x00000100L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_HI_ACK_MASK                                                         0x00000200L
+#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_LO_ACK_MASK                                                        0x00000400L
+#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_HI_ACK_MASK                                                        0x00000800L
+#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_LO_ACK_MASK                                                         0x00010000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_HI_ACK_MASK                                                         0x00020000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_LO_ACK_MASK                                                       0x00400000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_HI_ACK_MASK                                                       0x00800000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_LO_ACK_MASK                                                       0x01000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_HI_ACK_MASK                                                       0x02000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_LO_ACK_MASK                                                       0x04000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_HI_ACK_MASK                                                       0x08000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_LO_ACK_MASK                                                      0x10000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_HI_ACK_MASK                                                      0x20000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_LO_ACK_MASK                                                           0x40000000L
+#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_HI_ACK_MASK                                                           0x80000000L
+//UVD_MEMCHECK2_VCPU_INT_STAT
+#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_LO_ERR__SHIFT                                                      0x0
+#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_HI_ERR__SHIFT                                                      0x1
+#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_LO_ERR__SHIFT                                                      0x2
+#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_HI_ERR__SHIFT                                                      0x3
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_LO_ERR__SHIFT                                                     0x4
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_HI_ERR__SHIFT                                                     0x5
+#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_LO_ERR__SHIFT                                                    0x6
+#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_HI_ERR__SHIFT                                                    0x7
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_LO_ERR__SHIFT                                                     0x8
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_HI_ERR__SHIFT                                                     0x9
+#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_LO_ERR__SHIFT                                                    0xa
+#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_HI_ERR__SHIFT                                                    0xb
+#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_LO_ERR__SHIFT                                                     0x10
+#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_HI_ERR__SHIFT                                                     0x11
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_LO_ERR__SHIFT                                                   0x12
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_HI_ERR__SHIFT                                                   0x13
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_LO_ERR__SHIFT                                                   0x14
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_HI_ERR__SHIFT                                                   0x15
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_LO_ERR__SHIFT                                                   0x16
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_HI_ERR__SHIFT                                                   0x17
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_LO_ERR__SHIFT                                                  0x18
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_HI_ERR__SHIFT                                                  0x19
+#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_LO_ERR__SHIFT                                                       0x1a
+#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_HI_ERR__SHIFT                                                       0x1b
+#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_LO_ERR_MASK                                                        0x00000001L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_HI_ERR_MASK                                                        0x00000002L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_LO_ERR_MASK                                                        0x00000004L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_HI_ERR_MASK                                                        0x00000008L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_LO_ERR_MASK                                                       0x00000010L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_HI_ERR_MASK                                                       0x00000020L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_LO_ERR_MASK                                                      0x00000040L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_HI_ERR_MASK                                                      0x00000080L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_LO_ERR_MASK                                                       0x00000100L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_HI_ERR_MASK                                                       0x00000200L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_LO_ERR_MASK                                                      0x00000400L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_HI_ERR_MASK                                                      0x00000800L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_LO_ERR_MASK                                                       0x00010000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_HI_ERR_MASK                                                       0x00020000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_LO_ERR_MASK                                                     0x00040000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_HI_ERR_MASK                                                     0x00080000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_LO_ERR_MASK                                                     0x00100000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_HI_ERR_MASK                                                     0x00200000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_LO_ERR_MASK                                                     0x00400000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_HI_ERR_MASK                                                     0x00800000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_LO_ERR_MASK                                                    0x01000000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_HI_ERR_MASK                                                    0x02000000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_LO_ERR_MASK                                                         0x04000000L
+#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_HI_ERR_MASK                                                         0x08000000L
+//UVD_MEMCHECK2_VCPU_INT_ACK
+#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_LO_ACK__SHIFT                                                       0x0
+#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_HI_ACK__SHIFT                                                       0x1
+#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_LO_ACK__SHIFT                                                       0x2
+#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_HI_ACK__SHIFT                                                       0x3
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_LO_ACK__SHIFT                                                      0x4
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_HI_ACK__SHIFT                                                      0x5
+#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_LO_ACK__SHIFT                                                     0x6
+#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_HI_ACK__SHIFT                                                     0x7
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_LO_ACK__SHIFT                                                      0x8
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_HI_ACK__SHIFT                                                      0x9
+#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_LO_ACK__SHIFT                                                     0xa
+#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_HI_ACK__SHIFT                                                     0xb
+#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_LO_ACK__SHIFT                                                      0x10
+#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_HI_ACK__SHIFT                                                      0x11
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_LO_ACK__SHIFT                                                    0x12
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_HI_ACK__SHIFT                                                    0x13
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_LO_ACK__SHIFT                                                    0x14
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_HI_ACK__SHIFT                                                    0x15
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_LO_ACK__SHIFT                                                    0x16
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_HI_ACK__SHIFT                                                    0x17
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_LO_ACK__SHIFT                                                   0x18
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_HI_ACK__SHIFT                                                   0x19
+#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_LO_ACK__SHIFT                                                        0x1a
+#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_HI_ACK__SHIFT                                                        0x1b
+#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_LO_ACK_MASK                                                         0x00000001L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_HI_ACK_MASK                                                         0x00000002L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_LO_ACK_MASK                                                         0x00000004L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_HI_ACK_MASK                                                         0x00000008L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_LO_ACK_MASK                                                        0x00000010L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_HI_ACK_MASK                                                        0x00000020L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_LO_ACK_MASK                                                       0x00000040L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_HI_ACK_MASK                                                       0x00000080L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_LO_ACK_MASK                                                        0x00000100L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_HI_ACK_MASK                                                        0x00000200L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_LO_ACK_MASK                                                       0x00000400L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_HI_ACK_MASK                                                       0x00000800L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_LO_ACK_MASK                                                        0x00010000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_HI_ACK_MASK                                                        0x00020000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_LO_ACK_MASK                                                      0x00040000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_HI_ACK_MASK                                                      0x00080000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_LO_ACK_MASK                                                      0x00100000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_HI_ACK_MASK                                                      0x00200000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_LO_ACK_MASK                                                      0x00400000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_HI_ACK_MASK                                                      0x00800000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_LO_ACK_MASK                                                     0x01000000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_HI_ACK_MASK                                                     0x02000000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_LO_ACK_MASK                                                          0x04000000L
+#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_HI_ACK_MASK                                                          0x08000000L
+
+
+#endif
-- 
GitLab


From 76e5e4c70160b3764db6093b3366b36e4466d727 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Thu, 6 Jan 2022 14:41:33 -0500
Subject: [PATCH 1025/1791] drm/amdgpu: add VCN4_0_3 firmware

Add VCN4_0_3 firmware.

v2: fix fw name (Alex)

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index e63fcc58e8e06..72eb12aa2e8d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -56,6 +56,7 @@
 #define FIRMWARE_VCN_3_1_2		"amdgpu/vcn_3_1_2.bin"
 #define FIRMWARE_VCN4_0_0		"amdgpu/vcn_4_0_0.bin"
 #define FIRMWARE_VCN4_0_2		"amdgpu/vcn_4_0_2.bin"
+#define FIRMWARE_VCN4_0_3		"amdgpu/vcn_4_0_3.bin"
 #define FIRMWARE_VCN4_0_4		"amdgpu/vcn_4_0_4.bin"
 
 MODULE_FIRMWARE(FIRMWARE_RAVEN);
@@ -77,6 +78,7 @@ MODULE_FIRMWARE(FIRMWARE_YELLOW_CARP);
 MODULE_FIRMWARE(FIRMWARE_VCN_3_1_2);
 MODULE_FIRMWARE(FIRMWARE_VCN4_0_0);
 MODULE_FIRMWARE(FIRMWARE_VCN4_0_2);
+MODULE_FIRMWARE(FIRMWARE_VCN4_0_3);
 MODULE_FIRMWARE(FIRMWARE_VCN4_0_4);
 
 static void amdgpu_vcn_idle_work_handler(struct work_struct *work);
@@ -111,9 +113,18 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
 		atomic_set(&adev->vcn.inst[i].dpg_enc_submission_cnt, 0);
 
-	if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
-	    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
-		adev->vcn.indirect_sram = true;
+	switch (adev->ip_versions[UVD_HWIP][0]) {
+	case IP_VERSION(4, 0, 3):
+		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
+		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
+			adev->vcn.indirect_sram = false;
+		break;
+	default:
+		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
+		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
+			adev->vcn.indirect_sram = true;
+		break;
+	}
 
 	/*
 	 * Some Steam Deck's BIOS versions are incompatible with the
-- 
GitLab


From e684e654eba9481a9f462a7bbf5c385b7d1c076e Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Thu, 6 Jan 2022 17:04:42 -0500
Subject: [PATCH 1026/1791] drm/amdgpu/jpeg: add jpeg support for VCN4_0_3

Add jpeg support for VCN4_0_3.

v2: squash in delayed work typo fix (Alex)

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile      |   3 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 762 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h |  49 ++
 3 files changed, 813 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
 create mode 100644 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 7fb09000efc06..fc20aee6694fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -188,7 +188,8 @@ amdgpu-y += \
 	jpeg_v2_0.o \
 	jpeg_v2_5.o \
 	jpeg_v3_0.o \
-	jpeg_v4_0.o
+	jpeg_v4_0.o \
+	jpeg_v4_0_3.o
 
 # add ATHUB block
 amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
new file mode 100644
index 0000000000000..1fc72f9b52ed3
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -0,0 +1,762 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "amdgpu.h"
+#include "amdgpu_jpeg.h"
+#include "soc15.h"
+#include "soc15d.h"
+#include "jpeg_v4_0_3.h"
+
+#include "vcn/vcn_4_0_3_offset.h"
+#include "vcn/vcn_4_0_3_sh_mask.h"
+#include "ivsrcid/vcn/irqsrcs_vcn_2_0.h"
+
+enum jpeg_engin_status {
+	UVD_PGFSM_STATUS__UVDJ_PWR_ON  = 0,
+	UVD_PGFSM_STATUS__UVDJ_PWR_OFF = 2,
+};
+
+static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev);
+static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev);
+static int jpeg_v4_0_3_set_powergating_state(void *handle,
+				enum amd_powergating_state state);
+
+/**
+ * jpeg_v4_0_3_early_init - set function pointers
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Set ring and irq function pointers
+ */
+static int jpeg_v4_0_3_early_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	jpeg_v4_0_3_set_dec_ring_funcs(adev);
+	jpeg_v4_0_3_set_irq_funcs(adev);
+
+	return 0;
+}
+
+/**
+ * jpeg_v4_0_3_sw_init - sw init for JPEG block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Load firmware and sw initialization
+ */
+static int jpeg_v4_0_3_sw_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct amdgpu_ring *ring;
+	int r;
+
+	/* JPEG TRAP */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
+		VCN_2_0__SRCID__JPEG_DECODE, &adev->jpeg.inst->irq);
+	if (r)
+		return r;
+
+	r = amdgpu_jpeg_sw_init(adev);
+	if (r)
+		return r;
+
+	r = amdgpu_jpeg_resume(adev);
+	if (r)
+		return r;
+
+	ring = &adev->jpeg.inst->ring_dec;
+	ring->use_doorbell = false;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
+	sprintf(ring->name, "jpeg_dec");
+	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
+			     AMDGPU_RING_PRIO_DEFAULT, NULL);
+	if (r)
+		return r;
+
+	adev->jpeg.internal.jpeg_pitch = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
+	adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
+
+	return 0;
+}
+
+/**
+ * jpeg_v4_0_3_sw_fini - sw fini for JPEG block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * JPEG suspend and free up sw allocation
+ */
+static int jpeg_v4_0_3_sw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = amdgpu_jpeg_suspend(adev);
+	if (r)
+		return r;
+
+	r = amdgpu_jpeg_sw_fini(adev);
+
+	return r;
+}
+
+/**
+ * jpeg_v4_0_3_hw_init - start and test JPEG block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ */
+static int jpeg_v4_0_3_hw_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	int r;
+
+	r = amdgpu_ring_test_helper(ring);
+	if (!r)
+		DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n");
+
+	return r;
+}
+
+/**
+ * jpeg_v4_0_3_hw_fini - stop the hardware block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Stop the JPEG block, mark ring as not ready any more
+ */
+static int jpeg_v4_0_3_hw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	cancel_delayed_work_sync(&adev->jpeg.idle_work);
+
+	if (adev->jpeg.cur_state != AMD_PG_STATE_GATE)
+		jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE);
+
+	return 0;
+}
+
+/**
+ * jpeg_v4_0_3_suspend - suspend JPEG block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * HW fini and suspend JPEG block
+ */
+static int jpeg_v4_0_3_suspend(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = jpeg_v4_0_3_hw_fini(adev);
+	if (r)
+		return r;
+
+	r = amdgpu_jpeg_suspend(adev);
+
+	return r;
+}
+
+/**
+ * jpeg_v4_0_3_resume - resume JPEG block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Resume firmware and hw init JPEG block
+ */
+static int jpeg_v4_0_3_resume(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = amdgpu_jpeg_resume(adev);
+	if (r)
+		return r;
+
+	r = jpeg_v4_0_3_hw_init(adev);
+
+	return r;
+}
+
+static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
+		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	else
+		data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+
+	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
+	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
+	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
+
+	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
+	data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK
+		| JPEG_CGC_GATE__JPEG2_DEC_MASK
+		| JPEG_CGC_GATE__JMCIF_MASK
+		| JPEG_CGC_GATE__JRBBM_MASK);
+	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
+}
+
+static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
+		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	else
+		data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+
+	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
+	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
+	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
+
+	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
+	data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK
+		|JPEG_CGC_GATE__JPEG2_DEC_MASK
+		|JPEG_CGC_GATE__JMCIF_MASK
+		|JPEG_CGC_GATE__JRBBM_MASK);
+	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
+}
+
+/**
+ * jpeg_v4_0_3_start - start JPEG block
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Setup and start the JPEG block
+ */
+static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+
+	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
+		1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+	SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_PGFSM_STATUS,
+		UVD_PGFSM_STATUS__UVDJ_PWR_ON <<
+		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
+
+	/* disable anti hang mechanism */
+	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_POWER_STATUS), 0,
+		~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+
+	/* JPEG disable CGC */
+	jpeg_v4_0_3_disable_clock_gating(adev);
+
+	/* MJPEG global tiling registers */
+	WREG32_SOC15(JPEG, 0, regJPEG_DEC_GFX8_ADDR_CONFIG,
+		adev->gfx.config.gb_addr_config);
+	WREG32_SOC15(JPEG, 0, regJPEG_DEC_GFX10_ADDR_CONFIG,
+		adev->gfx.config.gb_addr_config);
+
+	/* enable JMI channel */
+	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI_CNTL), 0,
+		~UVD_JMI_CNTL__SOFT_RESET_MASK);
+
+	/* enable System Interrupt for JRBC */
+	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regJPEG_SYS_INT_EN),
+		JPEG_SYS_INT_EN__DJRBC0_MASK,
+		~JPEG_SYS_INT_EN__DJRBC0_MASK);
+
+	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, 0);
+	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, (0x00000001L | 0x00000002L));
+	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
+		lower_32_bits(ring->gpu_addr));
+	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
+		upper_32_bits(ring->gpu_addr));
+	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_RPTR, 0);
+	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR, 0);
+	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, 0x00000002L);
+	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_SIZE, ring->ring_size / 4);
+	ring->wptr = RREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR);
+
+	return 0;
+}
+
+/**
+ * jpeg_v4_0_3_stop - stop JPEG block
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * stop the JPEG block
+ */
+static int jpeg_v4_0_3_stop(struct amdgpu_device *adev)
+{
+	/* reset JMI */
+	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI_CNTL),
+		UVD_JMI_CNTL__SOFT_RESET_MASK,
+		~UVD_JMI_CNTL__SOFT_RESET_MASK);
+
+	jpeg_v4_0_3_enable_clock_gating(adev);
+
+	/* enable anti hang mechanism */
+	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_POWER_STATUS),
+		UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK,
+		~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+
+	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
+		2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+	SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_PGFSM_STATUS,
+		UVD_PGFSM_STATUS__UVDJ_PWR_OFF <<
+		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
+
+	return 0;
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_get_rptr - get read pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Returns the current hardware read pointer
+ */
+static uint64_t jpeg_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	return RREG32_SOC15(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_RPTR);
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_get_wptr - get write pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Returns the current hardware write pointer
+ */
+static uint64_t jpeg_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (ring->use_doorbell)
+		return adev->wb.wb[ring->wptr_offs];
+	else
+		return RREG32_SOC15(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR);
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_set_wptr - set write pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Commits the write pointer to the hardware
+ */
+static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (ring->use_doorbell) {
+		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
+		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
+	} else {
+		WREG32_SOC15(JPEG, ring->me,
+			regUVD_JRBC0_UVD_JRBC_RB_WPTR, lower_32_bits(ring->wptr));
+	}
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_insert_start - insert a start command
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Write a start command to the ring.
+ */
+static void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x62a04);/* TODO: PCTL0_MMHUB_DEEPSLEEP_IB */
+
+	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x80004000);
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_insert_end - insert a end command
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Write a end command to the ring.
+ */
+static void jpeg_v4_0_3_dec_ring_insert_end(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x62a04);
+
+	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x00004000);
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_emit_fence - emit an fence & trap command
+ *
+ * @ring: amdgpu_ring pointer
+ * @addr: address
+ * @seq: sequence number
+ * @flags: fence related flags
+ *
+ * Write a fence and a trap command to the ring.
+ */
+static void jpeg_v4_0_3_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
+				unsigned int flags)
+{
+	WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JPEG_GPCOM_DATA0_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, seq);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JPEG_GPCOM_DATA1_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, seq);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, lower_32_bits(addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, upper_32_bits(addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x8);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET,
+		0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE4));
+	amdgpu_ring_write(ring, 0);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x3fbc);
+
+	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x1);
+
+	amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE7));
+	amdgpu_ring_write(ring, 0);
+}
+
+/**
+ * jpeg_v4_0_3_dec_ring_emit_ib - execute indirect buffer
+ *
+ * @ring: amdgpu_ring pointer
+ * @job: job to retrieve vmid from
+ * @ib: indirect buffer to execute
+ * @flags: unused
+ *
+ * Write ring commands to execute the indirect buffer.
+ */
+static void jpeg_v4_0_3_dec_ring_emit_ib(struct amdgpu_ring *ring,
+				struct amdgpu_job *job,
+				struct amdgpu_ib *ib,
+				uint32_t flags)
+{
+	unsigned int vmid = AMDGPU_JOB_GET_VMID(job);
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JPEG_VMID_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_IB_64BIT_BAR_HIGH_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_IB_SIZE_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, ib->length_dw);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_LOW_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, lower_32_bits(ring->gpu_addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_HIGH_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, upper_32_bits(ring->gpu_addr));
+
+	amdgpu_ring_write(ring,	PACKETJ(0, 0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE2));
+	amdgpu_ring_write(ring, 0);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x01400200);
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x2);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_STATUS_INTERNAL_OFFSET,
+		0, PACKETJ_CONDITION_CHECK3, PACKETJ_TYPE3));
+	amdgpu_ring_write(ring, 0x2);
+}
+
+static void jpeg_v4_0_3_dec_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg,
+				uint32_t val, uint32_t mask)
+{
+	uint32_t reg_offset = (reg << 2);
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, 0x01400200);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	amdgpu_ring_write(ring, val);
+
+	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	if (reg_offset >= 0x10000 && reg_offset <= 0x105ff) {
+		amdgpu_ring_write(ring, 0);
+		amdgpu_ring_write(ring,
+			PACKETJ((reg_offset >> 2), 0, 0, PACKETJ_TYPE3));
+	} else {
+		amdgpu_ring_write(ring, reg_offset);
+		amdgpu_ring_write(ring,	PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
+			0, 0, PACKETJ_TYPE3));
+	}
+	amdgpu_ring_write(ring, mask);
+}
+
+static void jpeg_v4_0_3_dec_ring_emit_vm_flush(struct amdgpu_ring *ring,
+				unsigned int vmid, uint64_t pd_addr)
+{
+	struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->vm_hub];
+	uint32_t data0, data1, mask;
+
+	pd_addr = amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
+
+	/* wait for register write */
+	data0 = hub->ctx0_ptb_addr_lo32 + vmid * hub->ctx_addr_distance;
+	data1 = lower_32_bits(pd_addr);
+	mask = 0xffffffff;
+	jpeg_v4_0_3_dec_ring_emit_reg_wait(ring, data0, data1, mask);
+}
+
+static void jpeg_v4_0_3_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val)
+{
+	uint32_t reg_offset = (reg << 2);
+
+	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+		0, 0, PACKETJ_TYPE0));
+	if (reg_offset >= 0x10000 && reg_offset <= 0x105ff) {
+		amdgpu_ring_write(ring, 0);
+		amdgpu_ring_write(ring,
+			PACKETJ((reg_offset >> 2), 0, 0, PACKETJ_TYPE0));
+	} else {
+		amdgpu_ring_write(ring, reg_offset);
+		amdgpu_ring_write(ring,	PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
+			0, 0, PACKETJ_TYPE0));
+	}
+	amdgpu_ring_write(ring, val);
+}
+
+static void jpeg_v4_0_3_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count)
+{
+	int i;
+
+	WARN_ON(ring->wptr % 2 || count % 2);
+
+	for (i = 0; i < count / 2; i++) {
+		amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
+		amdgpu_ring_write(ring, 0);
+	}
+}
+
+static bool jpeg_v4_0_3_is_idle(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	return ((RREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_STATUS) &
+		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
+		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+}
+
+static int jpeg_v4_0_3_wait_for_idle(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
+
+	ret = SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_JRBC0_UVD_JRBC_STATUS,
+		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
+		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+	return ret;
+}
+
+static int jpeg_v4_0_3_set_clockgating_state(void *handle,
+					  enum amd_clockgating_state state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool enable = (state == AMD_CG_STATE_GATE);
+
+	if (enable) {
+		if (!jpeg_v4_0_3_is_idle(handle))
+			return -EBUSY;
+		jpeg_v4_0_3_enable_clock_gating(adev);
+	} else {
+		jpeg_v4_0_3_disable_clock_gating(adev);
+	}
+
+	return 0;
+}
+
+static int jpeg_v4_0_3_set_powergating_state(void *handle,
+					  enum amd_powergating_state state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
+
+	if (state == adev->jpeg.cur_state)
+		return 0;
+
+	if (state == AMD_PG_STATE_GATE)
+		ret = jpeg_v4_0_3_stop(adev);
+	else
+		ret = jpeg_v4_0_3_start(adev);
+
+	if (!ret)
+		adev->jpeg.cur_state = state;
+
+	return ret;
+}
+
+static int jpeg_v4_0_3_set_interrupt_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned int type,
+					enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
+static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev,
+				      struct amdgpu_irq_src *source,
+				      struct amdgpu_iv_entry *entry)
+{
+	DRM_DEV_DEBUG(adev->dev, "IH: JPEG TRAP\n");
+
+	switch (entry->src_id) {
+	case VCN_2_0__SRCID__JPEG_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		break;
+	default:
+		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
+			  entry->src_id, entry->src_data[0]);
+		break;
+	}
+
+	return 0;
+}
+
+static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = {
+	.name = "jpeg_v4_0_3",
+	.early_init = jpeg_v4_0_3_early_init,
+	.late_init = NULL,
+	.sw_init = jpeg_v4_0_3_sw_init,
+	.sw_fini = jpeg_v4_0_3_sw_fini,
+	.hw_init = jpeg_v4_0_3_hw_init,
+	.hw_fini = jpeg_v4_0_3_hw_fini,
+	.suspend = jpeg_v4_0_3_suspend,
+	.resume = jpeg_v4_0_3_resume,
+	.is_idle = jpeg_v4_0_3_is_idle,
+	.wait_for_idle = jpeg_v4_0_3_wait_for_idle,
+	.check_soft_reset = NULL,
+	.pre_soft_reset = NULL,
+	.soft_reset = NULL,
+	.post_soft_reset = NULL,
+	.set_clockgating_state = jpeg_v4_0_3_set_clockgating_state,
+	.set_powergating_state = jpeg_v4_0_3_set_powergating_state,
+};
+
+static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
+	.type = AMDGPU_RING_TYPE_VCN_JPEG,
+	.align_mask = 0xf,
+	.get_rptr = jpeg_v4_0_3_dec_ring_get_rptr,
+	.get_wptr = jpeg_v4_0_3_dec_ring_get_wptr,
+	.set_wptr = jpeg_v4_0_3_dec_ring_set_wptr,
+	.emit_frame_size =
+		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
+		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
+		8 + /* jpeg_v4_0_3_dec_ring_emit_vm_flush */
+		18 + 18 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */
+		8 + 16,
+	.emit_ib_size = 22, /* jpeg_v4_0_3_dec_ring_emit_ib */
+	.emit_ib = jpeg_v4_0_3_dec_ring_emit_ib,
+	.emit_fence = jpeg_v4_0_3_dec_ring_emit_fence,
+	.emit_vm_flush = jpeg_v4_0_3_dec_ring_emit_vm_flush,
+	.test_ring = amdgpu_jpeg_dec_ring_test_ring,
+	.test_ib = amdgpu_jpeg_dec_ring_test_ib,
+	.insert_nop = jpeg_v4_0_3_dec_ring_nop,
+	.insert_start = jpeg_v4_0_3_dec_ring_insert_start,
+	.insert_end = jpeg_v4_0_3_dec_ring_insert_end,
+	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.begin_use = amdgpu_jpeg_ring_begin_use,
+	.end_use = amdgpu_jpeg_ring_end_use,
+	.emit_wreg = jpeg_v4_0_3_dec_ring_emit_wreg,
+	.emit_reg_wait = jpeg_v4_0_3_dec_ring_emit_reg_wait,
+	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+};
+
+static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
+{
+	adev->jpeg.inst->ring_dec.funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec.me = 0;
+	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
+}
+
+static const struct amdgpu_irq_src_funcs jpeg_v4_0_3_irq_funcs = {
+	.set = jpeg_v4_0_3_set_interrupt_state,
+	.process = jpeg_v4_0_3_process_interrupt,
+};
+
+static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
+{
+	adev->jpeg.inst->irq.num_types = 1;
+	adev->jpeg.inst->irq.funcs = &jpeg_v4_0_3_irq_funcs;
+}
+
+const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block = {
+	.type = AMD_IP_BLOCK_TYPE_JPEG,
+	.major = 4,
+	.minor = 0,
+	.rev = 3,
+	.funcs = &jpeg_v4_0_3_ip_funcs,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
new file mode 100644
index 0000000000000..ca03d17e13fa6
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
@@ -0,0 +1,49 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __JPEG_V4_0_3_H__
+#define __JPEG_V4_0_3_H__
+
+#define regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET			0x1bfff
+#define regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET				0x404d
+#define regUVD_JPEG_GPCOM_DATA0_INTERNAL_OFFSET				0x404e
+#define regUVD_JPEG_GPCOM_DATA1_INTERNAL_OFFSET				0x404f
+#define regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_INTERNAL_OFFSET		0x40ab
+#define regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_INTERNAL_OFFSET	0x40ac
+#define regUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET				0x40a4
+#define regUVD_LMI_JPEG_VMID_INTERNAL_OFFSET				0x40a6
+#define regUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET		0x40b6
+#define regUVD_LMI_JRBC_IB_64BIT_BAR_HIGH_INTERNAL_OFFSET		0x40b7
+#define regUVD_JRBC_IB_SIZE_INTERNAL_OFFSET				0x4082
+#define regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_LOW_INTERNAL_OFFSET		0x42d4
+#define regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_HIGH_INTERNAL_OFFSET	0x42d5
+#define regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET			0x4085
+#define regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET				0x4084
+#define regUVD_JRBC_STATUS_INTERNAL_OFFSET				0x4089
+#define regUVD_JPEG_PITCH_INTERNAL_OFFSET				0x4043
+
+#define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
+
+extern const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block;
+
+#endif /* __JPEG_V4_0_3_H__ */
-- 
GitLab


From b889ef4ac98837838c38f7b9f72bba2f33ee367d Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 24 May 2022 11:56:44 +0800
Subject: [PATCH 1027/1791] drm/amdgpu/vcn: add vcn support for VCN4_0_3

Add vcn support for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile     |    1 +
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 1438 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h |   29 +
 3 files changed, 1468 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
 create mode 100644 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index fc20aee6694fa..594e303084c52 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -183,6 +183,7 @@ amdgpu-y += \
 	vcn_v2_5.o \
 	vcn_v3_0.o \
 	vcn_v4_0.o \
+	vcn_v4_0_3.o \
 	amdgpu_jpeg.o \
 	jpeg_v1_0.o \
 	jpeg_v2_0.o \
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
new file mode 100644
index 0000000000000..98bff162f4535
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -0,0 +1,1438 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include <linux/firmware.h>
+#include <drm/drm_drv.h>
+
+#include "amdgpu.h"
+#include "amdgpu_vcn.h"
+#include "amdgpu_pm.h"
+#include "soc15.h"
+#include "soc15d.h"
+#include "soc15_hw_ip.h"
+#include "vcn_v2_0.h"
+#include "vcn_sw_ring.h"
+
+#include "vcn/vcn_4_0_3_offset.h"
+#include "vcn/vcn_4_0_3_sh_mask.h"
+#include "ivsrcid/vcn/irqsrcs_vcn_2_0.h"
+
+#define mmUVD_DPG_LMA_CTL		regUVD_DPG_LMA_CTL
+#define mmUVD_DPG_LMA_CTL_BASE_IDX	regUVD_DPG_LMA_CTL_BASE_IDX
+#define mmUVD_DPG_LMA_DATA		regUVD_DPG_LMA_DATA
+#define mmUVD_DPG_LMA_DATA_BASE_IDX	regUVD_DPG_LMA_DATA_BASE_IDX
+
+#define VCN_VID_SOC_ADDRESS_2_0		0x1fb00
+#define VCN1_VID_SOC_ADDRESS_3_0	0x48300
+
+static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev);
+static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev);
+static int vcn_v4_0_3_set_powergating_state(void *handle,
+		enum amd_powergating_state state);
+static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev,
+		int inst_idx, struct dpg_pause_state *new_state);
+
+/**
+ * vcn_v4_0_3_early_init - set function pointers
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Set ring and irq function pointers
+ */
+static int vcn_v4_0_3_early_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	vcn_v4_0_3_set_dec_ring_funcs(adev);
+	vcn_v4_0_3_set_irq_funcs(adev);
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_sw_init - sw init for VCN block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Load firmware and sw initialization
+ */
+static int vcn_v4_0_3_sw_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_ring *ring;
+	int r;
+
+	r = amdgpu_vcn_sw_init(adev);
+	if (r)
+		return r;
+
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		const struct common_firmware_header *hdr;
+		hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
+		adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].ucode_id = AMDGPU_UCODE_ID_VCN;
+		adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].fw = adev->vcn.fw;
+		adev->firmware.fw_size +=
+			ALIGN(le32_to_cpu(hdr->ucode_size_bytes), PAGE_SIZE);
+
+		DRM_DEV_INFO(adev->dev, "Will use PSP to load VCN firmware\n");
+	}
+
+	r = amdgpu_vcn_resume(adev);
+	if (r)
+		return r;
+
+	/* VCN DEC TRAP */
+	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
+		VCN_2_0__SRCID__UVD_SYSTEM_MESSAGE_INTERRUPT, &adev->vcn.inst->irq);
+	if (r)
+		return r;
+
+	ring = &adev->vcn.inst->ring_dec;
+	ring->use_doorbell = false;
+	ring->vm_hub = AMDGPU_MMHUB0(0);
+	sprintf(ring->name, "vcn_dec");
+	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
+			     AMDGPU_RING_PRIO_DEFAULT,
+			     &adev->vcn.inst->sched_score);
+	if (r)
+		return r;
+
+	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_SW_RING_FLAG) |
+				     cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG) |
+				     cpu_to_le32(AMDGPU_VCN_FW_SHARED_FLAG_0_RB);
+	fw_shared->sw_ring.is_enabled = cpu_to_le32(true);
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
+		adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode;
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_sw_fini - sw fini for VCN block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * VCN suspend and free up sw allocation
+ */
+static int vcn_v4_0_3_sw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r, idx;
+
+	if (drm_dev_enter(&adev->ddev, &idx)) {
+		volatile struct amdgpu_fw_shared *fw_shared;
+
+		fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+		fw_shared->present_flag_0 = 0;
+		fw_shared->sw_ring.is_enabled = cpu_to_le32(false);
+
+		drm_dev_exit(idx);
+	}
+
+	r = amdgpu_vcn_suspend(adev);
+	if (r)
+		return r;
+
+	r = amdgpu_vcn_sw_fini(adev);
+
+	return r;
+}
+
+/**
+ * vcn_v4_0_3_hw_init - start and test VCN block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Initialize the hardware, boot up the VCPU and do some testing
+ */
+static int vcn_v4_0_3_hw_init(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
+	int r;
+
+	r = amdgpu_ring_test_helper(ring);
+
+	if (!r)
+		DRM_DEV_INFO(adev->dev, "VCN decode initialized successfully(under %s).\n",
+			(adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode");
+
+	return r;
+}
+
+/**
+ * vcn_v4_0_3_hw_fini - stop the hardware block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Stop the VCN block, mark ring as not ready any more
+ */
+static int vcn_v4_0_3_hw_fini(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	cancel_delayed_work_sync(&adev->vcn.idle_work);
+
+	if (adev->vcn.cur_state != AMD_PG_STATE_GATE)
+		vcn_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE);
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_suspend - suspend VCN block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * HW fini and suspend VCN block
+ */
+static int vcn_v4_0_3_suspend(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = vcn_v4_0_3_hw_fini(adev);
+	if (r)
+		return r;
+
+	r = amdgpu_vcn_suspend(adev);
+
+	return r;
+}
+
+/**
+ * vcn_v4_0_3_resume - resume VCN block
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Resume firmware and hw init VCN block
+ */
+static int vcn_v4_0_3_resume(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	r = amdgpu_vcn_resume(adev);
+	if (r)
+		return r;
+
+	r = vcn_v4_0_3_hw_init(adev);
+
+	return r;
+}
+
+/**
+ * vcn_v4_0_3_mc_resume - memory controller programming
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Let the VCN memory controller know it's offsets
+ */
+static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev)
+{
+	uint32_t offset, size;
+	const struct common_firmware_header *hdr;
+
+	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
+	size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8);
+
+	/* cache window 0: fw */
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_lo));
+		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_hi));
+		WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0, 0);
+		offset = 0;
+	} else {
+		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			lower_32_bits(adev->vcn.inst->gpu_addr));
+		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			upper_32_bits(adev->vcn.inst->gpu_addr));
+		offset = size;
+		if (amdgpu_emu_mode == 1)
+			/* No signed header here */
+			WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0, 0);
+		else
+			WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0,
+				AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
+	}
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE0, size);
+
+	/* cache window 1: stack */
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst->gpu_addr + offset));
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst->gpu_addr + offset));
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET1, 0);
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE);
+
+	/* cache window 2: context */
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET2, 0);
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE);
+
+	/* non-cache window */
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr));
+	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr));
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0, 0);
+	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_SIZE0,
+		AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)));
+}
+
+/**
+ * vcn_v4_0_mc_resume_dpg_mode - memory controller programming for dpg mode
+ *
+ * @adev: amdgpu_device pointer
+ * @indirect: indirectly write sram
+ *
+ * Let the VCN memory controller know it's offsets with dpg mode
+ */
+static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indirect)
+{
+	uint32_t offset, size;
+	const struct common_firmware_header *hdr;
+
+	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
+	size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8);
+
+	/* cache window 0: fw */
+	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
+		if (!indirect) {
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
+				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_lo), 0, indirect);
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
+				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_hi), 0, indirect);
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect);
+		} else {
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), 0, 0, indirect);
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), 0, 0, indirect);
+			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect);
+		}
+		offset = 0;
+	} else {
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
+			lower_32_bits(adev->vcn.inst->gpu_addr), 0, indirect);
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
+			upper_32_bits(adev->vcn.inst->gpu_addr), 0, indirect);
+		offset = size;
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_OFFSET0),
+			0, 0, indirect);
+	}
+
+	if (!indirect)
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_SIZE0), size, 0, indirect);
+	else
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_SIZE0), 0, 0, indirect);
+
+	/* cache window 1: stack */
+	if (!indirect) {
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW),
+			lower_32_bits(adev->vcn.inst->gpu_addr + offset), 0, indirect);
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH),
+			upper_32_bits(adev->vcn.inst->gpu_addr + offset), 0, indirect);
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect);
+	} else {
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), 0, 0, indirect);
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), 0, 0, indirect);
+		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect);
+	}
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE, 0, indirect);
+
+	/* cache window 2: context */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW),
+			lower_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH),
+			upper_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_OFFSET2), 0, 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE, 0, indirect);
+
+	/* non-cache window */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW),
+			lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH),
+			upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			VCN, 0, regUVD_VCPU_NONCACHE_SIZE0),
+			AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)), 0, indirect);
+
+	/* VCN global tiling registers */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_GFX8_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
+}
+
+/**
+ * vcn_v4_0_disable_static_power_gating - disable VCN static power gating
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Disable static power gating for VCN block
+ */
+static void vcn_v4_0_3_disable_static_power_gating(struct amdgpu_device *adev)
+{
+	uint32_t data = 0;
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN) {
+		data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
+
+		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
+		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS,
+			UVD_PGFSM_STATUS__UVDM_UVDU_UVDLM_PWR_ON_3_0, 0x3F3FFFFF);
+	} else {
+		data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
+			| 1 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
+
+		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
+		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS, 0,  0x3F3FFFFF);
+	}
+
+	data = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
+	data &= ~0x103;
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN)
+		data |= UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON |
+			UVD_POWER_STATUS__UVD_PG_EN_MASK;
+
+	WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, data);
+}
+
+/**
+ * vcn_v4_0_3_enable_static_power_gating - enable VCN static power gating
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Enable static power gating for VCN block
+ */
+static void vcn_v4_0_3_enable_static_power_gating(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN) {
+		/* Before power off, this indicator has to be turned on */
+		data = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
+		data &= ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK;
+		data |= UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF;
+		WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, data);
+
+		data = (2 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
+			| 2 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
+		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
+
+		data = (2 << UVD_PGFSM_STATUS__UVDM_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDS_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDF_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDTC_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDB_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDTA_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDLM_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDTD_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDTE_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDE_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDAB_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDTB_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDNA_PWR_STATUS__SHIFT
+			| 2 << UVD_PGFSM_STATUS__UVDNB_PWR_STATUS__SHIFT);
+		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS, data, 0x3F3FFFFF);
+	}
+}
+
+/**
+ * vcn_v4_0_3_disable_clock_gating - disable VCN clock gating
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Disable clock gating for VCN block
+ */
+static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	/* VCN disable CGC */
+	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
+		data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	else
+		data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK;
+	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
+	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
+	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_CGC_GATE);
+	data &= ~(UVD_CGC_GATE__SYS_MASK
+		| UVD_CGC_GATE__MPEG2_MASK
+		| UVD_CGC_GATE__REGS_MASK
+		| UVD_CGC_GATE__RBC_MASK
+		| UVD_CGC_GATE__LMI_MC_MASK
+		| UVD_CGC_GATE__LMI_UMC_MASK
+		| UVD_CGC_GATE__MPC_MASK
+		| UVD_CGC_GATE__LBSI_MASK
+		| UVD_CGC_GATE__LRBBM_MASK
+		| UVD_CGC_GATE__WCB_MASK
+		| UVD_CGC_GATE__VCPU_MASK
+		| UVD_CGC_GATE__MMSCH_MASK);
+
+	WREG32_SOC15(VCN, 0, regUVD_CGC_GATE, data);
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_CGC_GATE, 0,  0xFFFFFFFF);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK
+		| UVD_CGC_CTRL__MPEG2_MODE_MASK
+		| UVD_CGC_CTRL__REGS_MODE_MASK
+		| UVD_CGC_CTRL__RBC_MODE_MASK
+		| UVD_CGC_CTRL__LMI_MC_MODE_MASK
+		| UVD_CGC_CTRL__LMI_UMC_MODE_MASK
+		| UVD_CGC_CTRL__MPC_MODE_MASK
+		| UVD_CGC_CTRL__LBSI_MODE_MASK
+		| UVD_CGC_CTRL__LRBBM_MODE_MASK
+		| UVD_CGC_CTRL__WCB_MODE_MASK
+		| UVD_CGC_CTRL__VCPU_MODE_MASK
+		| UVD_CGC_CTRL__MMSCH_MODE_MASK);
+	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_GATE);
+	data |= (UVD_SUVD_CGC_GATE__SRE_MASK
+		| UVD_SUVD_CGC_GATE__SIT_MASK
+		| UVD_SUVD_CGC_GATE__SMP_MASK
+		| UVD_SUVD_CGC_GATE__SCM_MASK
+		| UVD_SUVD_CGC_GATE__SDB_MASK
+		| UVD_SUVD_CGC_GATE__SRE_H264_MASK
+		| UVD_SUVD_CGC_GATE__SRE_HEVC_MASK
+		| UVD_SUVD_CGC_GATE__SIT_H264_MASK
+		| UVD_SUVD_CGC_GATE__SIT_HEVC_MASK
+		| UVD_SUVD_CGC_GATE__SCM_H264_MASK
+		| UVD_SUVD_CGC_GATE__SCM_HEVC_MASK
+		| UVD_SUVD_CGC_GATE__SDB_H264_MASK
+		| UVD_SUVD_CGC_GATE__SDB_HEVC_MASK
+		| UVD_SUVD_CGC_GATE__ENT_MASK
+		| UVD_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK
+		| UVD_SUVD_CGC_GATE__SITE_MASK
+		| UVD_SUVD_CGC_GATE__SRE_VP9_MASK
+		| UVD_SUVD_CGC_GATE__SCM_VP9_MASK
+		| UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK
+		| UVD_SUVD_CGC_GATE__SDB_VP9_MASK
+		| UVD_SUVD_CGC_GATE__IME_HEVC_MASK);
+	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_GATE, data);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL);
+	data &= ~(UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SCM_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SDB_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
+	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL, data);
+}
+
+/**
+ * vcn_v4_0_3_disable_clock_gating_dpg_mode - disable VCN clock gating dpg mode
+ *
+ * @adev: amdgpu_device pointer
+ * @sram_sel: sram select
+ * @indirect: indirectly write sram
+ *
+ * Disable clock gating for VCN block with dpg mode
+ */
+static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev, uint8_t sram_sel,
+				uint8_t indirect)
+{
+	uint32_t reg_data = 0;
+
+	/* enable sw clock gating control */
+	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
+		reg_data = 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	else
+		reg_data = 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	reg_data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
+	reg_data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
+	reg_data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK |
+		 UVD_CGC_CTRL__MPEG2_MODE_MASK |
+		 UVD_CGC_CTRL__REGS_MODE_MASK |
+		 UVD_CGC_CTRL__RBC_MODE_MASK |
+		 UVD_CGC_CTRL__LMI_MC_MODE_MASK |
+		 UVD_CGC_CTRL__LMI_UMC_MODE_MASK |
+		 UVD_CGC_CTRL__IDCT_MODE_MASK |
+		 UVD_CGC_CTRL__MPRD_MODE_MASK |
+		 UVD_CGC_CTRL__MPC_MODE_MASK |
+		 UVD_CGC_CTRL__LBSI_MODE_MASK |
+		 UVD_CGC_CTRL__LRBBM_MODE_MASK |
+		 UVD_CGC_CTRL__WCB_MODE_MASK |
+		 UVD_CGC_CTRL__VCPU_MODE_MASK);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_CGC_CTRL), reg_data, sram_sel, indirect);
+
+	/* turn off clock gating */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_CGC_GATE), 0, sram_sel, indirect);
+
+	/* turn on SUVD clock gating */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_SUVD_CGC_GATE), 1, sram_sel, indirect);
+
+	/* turn on sw mode in UVD_SUVD_CGC_CTRL */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_SUVD_CGC_CTRL), 0, sram_sel, indirect);
+}
+
+/**
+ * vcn_v4_0_enable_clock_gating - enable VCN clock gating
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Enable clock gating for VCN block
+ */
+static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	/* enable VCN CGC */
+	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
+		data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	else
+		data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
+	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
+	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	data |= (UVD_CGC_CTRL__SYS_MODE_MASK
+		| UVD_CGC_CTRL__MPEG2_MODE_MASK
+		| UVD_CGC_CTRL__REGS_MODE_MASK
+		| UVD_CGC_CTRL__RBC_MODE_MASK
+		| UVD_CGC_CTRL__LMI_MC_MODE_MASK
+		| UVD_CGC_CTRL__LMI_UMC_MODE_MASK
+		| UVD_CGC_CTRL__MPC_MODE_MASK
+		| UVD_CGC_CTRL__LBSI_MODE_MASK
+		| UVD_CGC_CTRL__LRBBM_MODE_MASK
+		| UVD_CGC_CTRL__WCB_MODE_MASK
+		| UVD_CGC_CTRL__VCPU_MODE_MASK);
+	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+
+	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL);
+	data |= (UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SCM_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SDB_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
+		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
+	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL, data);
+}
+
+/**
+ * vcn_v4_0_3_start_dpg_mode - VCN start with dpg mode
+ *
+ * @adev: amdgpu_device pointer
+ * @indirect: indirectly write sram
+ *
+ * Start VCN block with dpg mode
+ */
+static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
+{
+	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	struct amdgpu_ring *ring;
+	uint32_t tmp;
+
+	/* disable register anti-hang mechanism */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_POWER_STATUS), 1,
+		~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+	/* enable dynamic power gating mode */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
+	tmp |= UVD_POWER_STATUS__UVD_PG_MODE_MASK;
+	tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, tmp);
+
+	if (indirect)
+		adev->vcn.inst->dpg_sram_curr_addr = (uint32_t *)adev->vcn.inst->dpg_sram_cpu_addr;
+
+	/* enable clock gating */
+	vcn_v4_0_3_disable_clock_gating_dpg_mode(adev, 0, indirect);
+
+	/* enable VCPU clock */
+	tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT);
+	tmp |= UVD_VCPU_CNTL__CLK_EN_MASK;
+	tmp |= UVD_VCPU_CNTL__BLK_RST_MASK;
+
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect);
+
+	/* disable master interrupt */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MASTINT_EN), 0, 0, indirect);
+
+	/* setup regUVD_LMI_CTRL */
+	tmp = (UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
+		UVD_LMI_CTRL__REQ_MODE_MASK |
+		UVD_LMI_CTRL__CRC_RESET_MASK |
+		UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
+		UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
+		UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK |
+		(8 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) |
+		0x00100000L);
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_LMI_CTRL), tmp, 0, indirect);
+
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MPC_CNTL),
+		0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT, 0, indirect);
+
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MPC_SET_MUXA0),
+		((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
+		 (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
+		 (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
+		 (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)), 0, indirect);
+
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MPC_SET_MUXB0),
+		 ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
+		 (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
+		 (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
+		 (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)), 0, indirect);
+
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MPC_SET_MUX),
+		((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
+		 (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
+		 (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)), 0, indirect);
+
+	vcn_v4_0_3_mc_resume_dpg_mode(adev, indirect);
+
+	tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT);
+	tmp |= UVD_VCPU_CNTL__CLK_EN_MASK;
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect);
+
+	/* enable LMI MC and UMC channels */
+	tmp = 0x1f << UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM__SHIFT;
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_LMI_CTRL2), tmp, 0, indirect);
+
+	/* enable master interrupt */
+	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		VCN, 0, regUVD_MASTINT_EN),
+		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
+
+	if (indirect)
+		psp_update_vcn_sram(adev, 0, adev->vcn.inst->dpg_sram_gpu_addr,
+			(uint32_t)((uintptr_t)adev->vcn.inst->dpg_sram_curr_addr -
+				(uintptr_t)adev->vcn.inst->dpg_sram_cpu_addr));
+
+	ring = &adev->vcn.inst->ring_dec;
+	fw_shared->multi_queue.decode_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+
+	/* program the RB_BASE for ring buffer */
+	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
+		lower_32_bits(ring->gpu_addr));
+	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_HI4,
+		upper_32_bits(ring->gpu_addr));
+
+	WREG32_SOC15(VCN, 0, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+
+	/* resetting ring, fw should not check RB ring */
+	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
+	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+
+	/* Initialize the ring buffer's read and write pointers */
+	WREG32_SOC15(VCN, 0, regUVD_RB_RPTR4, 0);
+	WREG32_SOC15(VCN, 0, regUVD_RB_WPTR4, 0);
+	ring->wptr = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
+
+	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
+	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+
+	WREG32_SOC15(VCN, 0, regUVD_SCRATCH2, 0);
+
+	/* Reset FW shared memory RBC WPTR/RPTR */
+	fw_shared->rb.rptr = 0;
+	fw_shared->rb.wptr = lower_32_bits(ring->wptr);
+
+	/*resetting done, fw can check RB ring */
+	fw_shared->multi_queue.decode_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_start - VCN start
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Start VCN block
+ */
+static int vcn_v4_0_3_start(struct amdgpu_device *adev)
+{
+	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_ring *ring;
+	uint32_t tmp;
+	int j, k, r;
+
+	if (adev->pm.dpm_enabled)
+		amdgpu_dpm_enable_uvd(adev, true);
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
+		return vcn_v4_0_3_start_dpg_mode(adev, adev->vcn.indirect_sram);
+
+	/* disable VCN power gating */
+	vcn_v4_0_3_disable_static_power_gating(adev);
+
+	/* set VCN status busy */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_STATUS) | UVD_STATUS__UVD_BUSY;
+	WREG32_SOC15(VCN, 0, regUVD_STATUS, tmp);
+
+	/*SW clock gating */
+	vcn_v4_0_3_disable_clock_gating(adev);
+
+	/* enable VCPU clock */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
+		UVD_VCPU_CNTL__CLK_EN_MASK, ~UVD_VCPU_CNTL__CLK_EN_MASK);
+
+	/* disable master interrupt */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_MASTINT_EN), 0,
+		~UVD_MASTINT_EN__VCPU_EN_MASK);
+
+	/* enable LMI MC and UMC channels */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_LMI_CTRL2), 0,
+		~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK);
+
+	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
+	tmp &= ~UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
+	tmp &= ~UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+
+	/* setup regUVD_LMI_CTRL */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_LMI_CTRL);
+	WREG32_SOC15(VCN, 0, regUVD_LMI_CTRL, tmp |
+		UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
+		UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
+		UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
+		UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK);
+
+	/* setup regUVD_MPC_CNTL */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_MPC_CNTL);
+	tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK;
+	tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT;
+	WREG32_SOC15(VCN, 0, regUVD_MPC_CNTL, tmp);
+
+	/* setup UVD_MPC_SET_MUXA0 */
+	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUXA0,
+		((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
+		(0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
+		(0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
+		(0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)));
+
+	/* setup UVD_MPC_SET_MUXB0 */
+	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUXB0,
+		((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
+		(0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
+		(0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
+		(0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)));
+
+	/* setup UVD_MPC_SET_MUX */
+	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUX,
+		((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
+		(0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
+		(0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)));
+
+	vcn_v4_0_3_mc_resume(adev);
+
+	/* VCN global tiling registers */
+	WREG32_SOC15(VCN, 0, regUVD_GFX8_ADDR_CONFIG,
+		adev->gfx.config.gb_addr_config);
+	WREG32_SOC15(VCN, 0, regUVD_GFX10_ADDR_CONFIG,
+		adev->gfx.config.gb_addr_config);
+
+	/* unblock VCPU register access */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_RB_ARB_CTRL), 0,
+		~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+
+	/* release VCPU reset to boot */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
+		~UVD_VCPU_CNTL__BLK_RST_MASK);
+
+	for (j = 0; j < 10; ++j) {
+		uint32_t status;
+
+		for (k = 0; k < 100; ++k) {
+			status = RREG32_SOC15(VCN, 0, regUVD_STATUS);
+			if (status & 2)
+				break;
+			if (amdgpu_emu_mode == 1)
+				msleep(500);
+			else
+				mdelay(10);
+		}
+		r = 0;
+		if (status & 2)
+			break;
+
+		DRM_DEV_ERROR(adev->dev,
+			"VCN decode not responding, trying to reset the VCPU!!!\n");
+		WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
+			UVD_VCPU_CNTL__BLK_RST_MASK,
+			~UVD_VCPU_CNTL__BLK_RST_MASK);
+		mdelay(10);
+		WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
+			~UVD_VCPU_CNTL__BLK_RST_MASK);
+
+		mdelay(10);
+		r = -1;
+	}
+
+	if (r) {
+		DRM_DEV_ERROR(adev->dev, "VCN decode not responding, giving up!!!\n");
+		return r;
+	}
+
+	/* enable master interrupt */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_MASTINT_EN),
+		UVD_MASTINT_EN__VCPU_EN_MASK,
+		~UVD_MASTINT_EN__VCPU_EN_MASK);
+
+	/* clear the busy bit of VCN_STATUS */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_STATUS), 0,
+		~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
+
+	ring = &adev->vcn.inst->ring_dec;
+
+	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	fw_shared->multi_queue.decode_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+
+	/* program the RB_BASE for ring buffer */
+	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
+		lower_32_bits(ring->gpu_addr));
+	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_HI4,
+		upper_32_bits(ring->gpu_addr));
+
+	WREG32_SOC15(VCN, 0, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+
+	/* resetting ring, fw should not check RB ring */
+	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
+	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+
+	/* Initialize the ring buffer's read and write pointers */
+	WREG32_SOC15(VCN, 0, regUVD_RB_RPTR4, 0);
+	WREG32_SOC15(VCN, 0, regUVD_RB_WPTR4, 0);
+
+	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
+	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+
+	ring->wptr = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
+	fw_shared->rb.wptr = cpu_to_le32(lower_32_bits(ring->wptr));
+	fw_shared->multi_queue.decode_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_stop_dpg_mode - VCN stop with dpg mode
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Stop VCN block with dpg mode
+ */
+static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+
+	/* Wait for power status to be 1 */
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_POWER_STATUS, 1,
+		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+
+	/* wait for read ptr to be equal to write ptr */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR);
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR, tmp, 0xFFFFFFFF);
+
+	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR2);
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR2, tmp, 0xFFFFFFFF);
+
+	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR4, tmp, 0xFFFFFFFF);
+
+	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_POWER_STATUS, 1,
+		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+
+	/* disable dynamic power gating mode */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_POWER_STATUS), 0,
+		~UVD_POWER_STATUS__UVD_PG_MODE_MASK);
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_stop - VCN stop
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Stop VCN block
+ */
+static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
+{
+	uint32_t tmp;
+	int r = 0;
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
+		r = vcn_v4_0_3_stop_dpg_mode(adev);
+		goto Done;
+	}
+
+	/* wait for vcn idle */
+	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_STATUS, UVD_STATUS__IDLE, 0x7);
+	if (r)
+		return r;
+
+	tmp = UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK |
+		UVD_LMI_STATUS__READ_CLEAN_MASK |
+		UVD_LMI_STATUS__WRITE_CLEAN_MASK |
+		UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK;
+	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_LMI_STATUS, tmp, tmp);
+	if (r)
+		return r;
+
+	/* stall UMC channel */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_LMI_CTRL2);
+	tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_LMI_CTRL2, tmp);
+	tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK|
+		UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK;
+	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_LMI_STATUS, tmp, tmp);
+	if (r)
+		return r;
+
+	/* Unblock VCPU Register access */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_RB_ARB_CTRL),
+		UVD_RB_ARB_CTRL__VCPU_DIS_MASK,
+		~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+
+	/* release VCPU reset to boot */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
+		UVD_VCPU_CNTL__BLK_RST_MASK,
+		~UVD_VCPU_CNTL__BLK_RST_MASK);
+
+	/* disable VCPU clock */
+	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
+		~(UVD_VCPU_CNTL__CLK_EN_MASK));
+
+	/* reset LMI UMC/LMI/VCPU */
+	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
+	tmp |= UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+
+	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
+	tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+
+	tmp = RREG32_SOC15(VCN, 0, regUVD_VCPU_CNTL);
+	tmp |= UVD_VCPU_CNTL__BLK_RST_MASK;
+	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+
+	/* clear VCN status */
+	WREG32_SOC15(VCN, 0, regUVD_STATUS, 0);
+
+	/* apply HW clock gating */
+	vcn_v4_0_3_enable_clock_gating(adev);
+
+	/* enable VCN power gating */
+	vcn_v4_0_3_enable_static_power_gating(adev);
+
+Done:
+	if (adev->pm.dpm_enabled)
+		amdgpu_dpm_enable_uvd(adev, false);
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_pause_dpg_mode - VCN pause with dpg mode
+ *
+ * @adev: amdgpu_device pointer
+ * @inst_idx: instance number index
+ * @new_state: pause state
+ *
+ * Pause dpg mode for VCN block
+ */
+static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev, int inst_idx,
+				struct dpg_pause_state *new_state)
+{
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_dec_ring_get_rptr - get read pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Returns the current hardware read pointer
+ */
+static uint64_t vcn_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	return RREG32_SOC15(VCN, ring->me, regUVD_RB_RPTR4);
+}
+
+/**
+ * vcn_v4_0_3_dec_ring_get_wptr - get write pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Returns the current hardware write pointer
+ */
+static uint64_t vcn_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (ring->use_doorbell)
+		return adev->wb.wb[ring->wptr_offs];
+	else
+		return RREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR4);
+}
+
+/**
+ * vcn_v4_0_3_dec_ring_set_wptr - set write pointer
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Commits the write pointer to the hardware
+ */
+static void vcn_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	volatile struct amdgpu_fw_shared *fw_shared;
+
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
+		/*whenever update RBC_RB_WPTR, we save the wptr in shared rb.wptr and scratch2 */
+		fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+		fw_shared->rb.wptr = lower_32_bits(ring->wptr);
+		WREG32_SOC15(VCN, ring->me, regUVD_SCRATCH2,
+			lower_32_bits(ring->wptr));
+	}
+
+	if (ring->use_doorbell) {
+		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
+		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
+	} else {
+		WREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR4, lower_32_bits(ring->wptr));
+	}
+}
+
+static const struct amdgpu_ring_funcs vcn_v4_0_3_dec_sw_ring_vm_funcs = {
+	.type = AMDGPU_RING_TYPE_VCN_DEC,
+	.align_mask = 0x3f,
+	.nop = VCN_DEC_SW_CMD_NO_OP,
+	.get_rptr = vcn_v4_0_3_dec_ring_get_rptr,
+	.get_wptr = vcn_v4_0_3_dec_ring_get_wptr,
+	.set_wptr = vcn_v4_0_3_dec_ring_set_wptr,
+	.emit_frame_size =
+		SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
+		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 4 +
+		VCN_SW_RING_EMIT_FRAME_SIZE,
+	.emit_ib_size = 5, /* vcn_dec_sw_ring_emit_ib */
+	.emit_ib = vcn_dec_sw_ring_emit_ib,
+	.emit_fence = vcn_dec_sw_ring_emit_fence,
+	.emit_vm_flush = vcn_dec_sw_ring_emit_vm_flush,
+	.test_ring = amdgpu_vcn_dec_sw_ring_test_ring,
+	.test_ib = amdgpu_vcn_dec_sw_ring_test_ib,
+	.insert_nop = amdgpu_ring_insert_nop,
+	.insert_end = vcn_dec_sw_ring_insert_end,
+	.pad_ib = amdgpu_ring_generic_pad_ib,
+	.begin_use = amdgpu_vcn_ring_begin_use,
+	.end_use = amdgpu_vcn_ring_end_use,
+	.emit_wreg = vcn_dec_sw_ring_emit_wreg,
+	.emit_reg_wait = vcn_dec_sw_ring_emit_reg_wait,
+	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+};
+
+/**
+ * vcn_v4_0_3_set_dec_ring_funcs - set dec ring functions
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Set decode ring functions
+ */
+static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
+{
+	adev->vcn.inst->ring_dec.funcs = &vcn_v4_0_3_dec_sw_ring_vm_funcs;
+	adev->vcn.inst->ring_dec.me = 0;
+	DRM_DEV_INFO(adev->dev, "VCN decode(Software Ring) is enabled in VM mode\n");
+}
+
+/**
+ * vcn_v4_0_3_is_idle - check VCN block is idle
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Check whether VCN block is idle
+ */
+static bool vcn_v4_0_3_is_idle(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	return (RREG32_SOC15(VCN, 0, regUVD_STATUS) == UVD_STATUS__IDLE);
+}
+
+/**
+ * vcn_v4_0_3_wait_for_idle - wait for VCN block idle
+ *
+ * @handle: amdgpu_device pointer
+ *
+ * Wait for VCN block idle
+ */
+static int vcn_v4_0_3_wait_for_idle(void *handle)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	return SOC15_WAIT_ON_RREG(VCN, 0, regUVD_STATUS, UVD_STATUS__IDLE,
+			UVD_STATUS__IDLE);
+}
+
+/**
+ * vcn_v4_0_3_set_clockgating_state - set VCN block clockgating state
+ *
+ * @handle: amdgpu_device pointer
+ * @state: clock gating state
+ *
+ * Set VCN block clockgating state
+ */
+static int vcn_v4_0_3_set_clockgating_state(void *handle,
+					  enum amd_clockgating_state state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+
+	if (enable) {
+		if (RREG32_SOC15(VCN, 0, regUVD_STATUS) != UVD_STATUS__IDLE)
+			return -EBUSY;
+		vcn_v4_0_3_enable_clock_gating(adev);
+	} else {
+		vcn_v4_0_3_disable_clock_gating(adev);
+	}
+
+	return 0;
+}
+
+/**
+ * vcn_v4_0_3_set_powergating_state - set VCN block powergating state
+ *
+ * @handle: amdgpu_device pointer
+ * @state: power gating state
+ *
+ * Set VCN block powergating state
+ */
+static int vcn_v4_0_3_set_powergating_state(void *handle,
+					  enum amd_powergating_state state)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
+
+	if (state == adev->vcn.cur_state)
+		return 0;
+
+	if (state == AMD_PG_STATE_GATE)
+		ret = vcn_v4_0_3_stop(adev);
+	else
+		ret = vcn_v4_0_3_start(adev);
+
+	if (!ret)
+		adev->vcn.cur_state = state;
+
+	return ret;
+}
+
+/**
+ * vcn_v4_0_3_set_interrupt_state - set VCN block interrupt state
+ *
+ * @adev: amdgpu_device pointer
+ * @source: interrupt sources
+ * @type: interrupt types
+ * @state: interrupt states
+ *
+ * Set VCN block interrupt state
+ */
+static int vcn_v4_0_3_set_interrupt_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned int type,
+					enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
+/**
+ * vcn_v4_0_process_interrupt - process VCN block interrupt
+ *
+ * @adev: amdgpu_device pointer
+ * @source: interrupt sources
+ * @entry: interrupt entry from clients and sources
+ *
+ * Process VCN block interrupt
+ */
+static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev,
+				      struct amdgpu_irq_src *source,
+				      struct amdgpu_iv_entry *entry)
+{
+	DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n");
+
+	switch (entry->src_id) {
+	case VCN_2_0__SRCID__UVD_SYSTEM_MESSAGE_INTERRUPT:
+		amdgpu_fence_process(&adev->vcn.inst->ring_dec);
+		break;
+	default:
+		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
+			  entry->src_id, entry->src_data[0]);
+		break;
+	}
+
+	return 0;
+}
+
+static const struct amdgpu_irq_src_funcs vcn_v4_0_3_irq_funcs = {
+	.set = vcn_v4_0_3_set_interrupt_state,
+	.process = vcn_v4_0_3_process_interrupt,
+};
+
+/**
+ * vcn_v4_0_3_set_irq_funcs - set VCN block interrupt irq functions
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Set VCN block interrupt irq functions
+ */
+static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
+{
+	int i;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+
+		adev->vcn.inst->irq.num_types = 1;
+		adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs;
+	}
+}
+
+static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = {
+	.name = "vcn_v4_0_3",
+	.early_init = vcn_v4_0_3_early_init,
+	.late_init = NULL,
+	.sw_init = vcn_v4_0_3_sw_init,
+	.sw_fini = vcn_v4_0_3_sw_fini,
+	.hw_init = vcn_v4_0_3_hw_init,
+	.hw_fini = vcn_v4_0_3_hw_fini,
+	.suspend = vcn_v4_0_3_suspend,
+	.resume = vcn_v4_0_3_resume,
+	.is_idle = vcn_v4_0_3_is_idle,
+	.wait_for_idle = vcn_v4_0_3_wait_for_idle,
+	.check_soft_reset = NULL,
+	.pre_soft_reset = NULL,
+	.soft_reset = NULL,
+	.post_soft_reset = NULL,
+	.set_clockgating_state = vcn_v4_0_3_set_clockgating_state,
+	.set_powergating_state = vcn_v4_0_3_set_powergating_state,
+};
+
+const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block = {
+	.type = AMD_IP_BLOCK_TYPE_VCN,
+	.major = 4,
+	.minor = 0,
+	.rev = 3,
+	.funcs = &vcn_v4_0_3_ip_funcs,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h
new file mode 100644
index 0000000000000..0b046114373ae
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __VCN_V4_0_3_H__
+#define __VCN_V4_0_3_H__
+
+extern const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block;
+
+#endif /* __VCN_V4_0_3_H__ */
-- 
GitLab


From 380302f8b894a11fcd84a08aadf6a858eb003b0b Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sun, 9 Jan 2022 16:03:25 -0500
Subject: [PATCH 1028/1791] drm/amdgpu/jpeg: enable jpeg cg for VCN4_0_3

Enable jpeg cg for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 3221522e71e89..d68ebf6b6856e 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1095,7 +1095,8 @@ static int soc15_common_early_init(void *handle)
 		break;
 	case IP_VERSION(9, 4, 3):
 		adev->asic_funcs = &vega20_asic_funcs;
-		adev->cg_flags = 0;
+		adev->cg_flags =
+			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags = 0;
 		break;
 	default:
-- 
GitLab


From b7179fc29f50d837090d7b0c1f224e3a116cdcc2 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sun, 9 Jan 2022 16:04:05 -0500
Subject: [PATCH 1029/1791] drm/amdgpu/jpeg: enable jpeg pg for VCN4_0_3

Enable jpeg pg for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index d68ebf6b6856e..1939e88ef4217 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1097,7 +1097,8 @@ static int soc15_common_early_init(void *handle)
 		adev->asic_funcs = &vega20_asic_funcs;
 		adev->cg_flags =
 			AMD_CG_SUPPORT_JPEG_MGCG;
-		adev->pg_flags = 0;
+		adev->pg_flags =
+			AMD_PG_SUPPORT_JPEG;
 		break;
 	default:
 		/* FIXME: not supported yet */
-- 
GitLab


From 342397db6d09068e335d42a97e0879b7f5f99364 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sun, 9 Jan 2022 16:04:41 -0500
Subject: [PATCH 1030/1791] drm/amdgpu/vcn: enable vcn cg for VCN4_0_3

Enable vcn cg for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 1939e88ef4217..abc10b20c83cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1096,6 +1096,7 @@ static int soc15_common_early_init(void *handle)
 	case IP_VERSION(9, 4, 3):
 		adev->asic_funcs = &vega20_asic_funcs;
 		adev->cg_flags =
+			AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_JPEG;
-- 
GitLab


From ef3aa0b40c4b1f4d8b7db8582833cb61fc673f60 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sun, 9 Jan 2022 16:05:07 -0500
Subject: [PATCH 1031/1791] drm/amdgpu/vcn: enable vcn pg for VCN4_0_3

Enable vcn pg for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index abc10b20c83cc..ee00302950c23 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1099,6 +1099,7 @@ static int soc15_common_early_init(void *handle)
 			AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
+			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_JPEG;
 		break;
 	default:
-- 
GitLab


From 31c0ec84f92cd40cfae210dac59413372996b5e7 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sun, 9 Jan 2022 16:05:31 -0500
Subject: [PATCH 1032/1791] drm/amdgpu/vcn: enable vcn DPG mode for VCN4_0_3

Enable vcn DPG mode for VCN4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index ee00302950c23..b7e8af56df846 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1100,6 +1100,7 @@ static int soc15_common_early_init(void *handle)
 			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
+			AMD_PG_SUPPORT_VCN_DPG |
 			AMD_PG_SUPPORT_JPEG;
 		break;
 	default:
-- 
GitLab


From bc224553843e526bad4bb91188363aea1664a70d Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 24 May 2022 12:03:03 +0800
Subject: [PATCH 1033/1791] drm/amdgpu/jpeg: add multiple jpeg rings support

Add multiple jpeg rings support.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 21 +++++++++++----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h |  6 ++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c  |  5 +++--
 drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c   | 12 ++++++------
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c   | 14 +++++++-------
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c   | 18 +++++++++---------
 drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c   | 14 +++++++-------
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c   | 16 ++++++++--------
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 16 ++++++++--------
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c    | 12 ++++++------
 10 files changed, 69 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index b07c000fc8ba3..388466a5f730e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -45,13 +45,14 @@ int amdgpu_jpeg_sw_init(struct amdgpu_device *adev)
 
 int amdgpu_jpeg_sw_fini(struct amdgpu_device *adev)
 {
-	int i;
+	int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 
-		amdgpu_ring_fini(&adev->jpeg.inst[i].ring_dec);
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j)
+			amdgpu_ring_fini(&adev->jpeg.inst[i].ring_dec[j]);
 	}
 
 	mutex_destroy(&adev->jpeg.jpeg_pg_lock);
@@ -76,13 +77,14 @@ static void amdgpu_jpeg_idle_work_handler(struct work_struct *work)
 	struct amdgpu_device *adev =
 		container_of(work, struct amdgpu_device, jpeg.idle_work.work);
 	unsigned int fences = 0;
-	unsigned int i;
+	unsigned int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 
-		fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec);
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j)
+			fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec[j]);
 	}
 
 	if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt))
@@ -122,17 +124,17 @@ int amdgpu_jpeg_dec_ring_test_ring(struct amdgpu_ring *ring)
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-	WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch, 0xCAFEDEAD);
+	WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe], 0xCAFEDEAD);
 	r = amdgpu_ring_alloc(ring, 3);
 	if (r)
 		return r;
 
-	amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch, 0));
+	amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0));
 	amdgpu_ring_write(ring, 0xDEADBEEF);
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch);
+		tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]);
 		if (tmp == 0xDEADBEEF)
 			break;
 		udelay(1);
@@ -161,8 +163,7 @@ static int amdgpu_jpeg_dec_set_reg(struct amdgpu_ring *ring, uint32_t handle,
 
 	ib = &job->ibs[0];
 
-	ib->ptr[0] = PACKETJ(adev->jpeg.internal.jpeg_pitch, 0, 0,
-			     PACKETJ_TYPE0);
+	ib->ptr[0] = PACKETJ(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0, 0, PACKETJ_TYPE0);
 	ib->ptr[1] = 0xDEADBEEF;
 	for (i = 2; i < 16; i += 2) {
 		ib->ptr[i] = PACKETJ(0, 0, 0, PACKETJ_TYPE6);
@@ -208,7 +209,7 @@ int amdgpu_jpeg_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	}
 	if (!amdgpu_sriov_vf(adev)) {
 		for (i = 0; i < adev->usec_timeout; i++) {
-			tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch);
+			tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]);
 			if (tmp == 0xDEADBEEF)
 				break;
 			udelay(1);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index 0ca76f0f23e9c..cb6c127ab81d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -27,16 +27,17 @@
 #include "amdgpu_ras.h"
 
 #define AMDGPU_MAX_JPEG_INSTANCES	2
+#define AMDGPU_MAX_JPEG_RINGS		8
 
 #define AMDGPU_JPEG_HARVEST_JPEG0 (1 << 0)
 #define AMDGPU_JPEG_HARVEST_JPEG1 (1 << 1)
 
 struct amdgpu_jpeg_reg{
-	unsigned jpeg_pitch;
+	unsigned jpeg_pitch[AMDGPU_MAX_JPEG_RINGS];
 };
 
 struct amdgpu_jpeg_inst {
-	struct amdgpu_ring ring_dec;
+	struct amdgpu_ring ring_dec[AMDGPU_MAX_JPEG_RINGS];
 	struct amdgpu_irq_src irq;
 	struct amdgpu_jpeg_reg external;
 };
@@ -48,6 +49,7 @@ struct amdgpu_jpeg_ras {
 struct amdgpu_jpeg {
 	uint8_t	num_jpeg_inst;
 	struct amdgpu_jpeg_inst inst[AMDGPU_MAX_JPEG_INSTANCES];
+	unsigned num_jpeg_rings;
 	struct amdgpu_jpeg_reg internal;
 	unsigned harvest_config;
 	struct delayed_work idle_work;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 1d3b224b8b289..44997c7ee89d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -462,8 +462,9 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
 			if (adev->jpeg.harvest_config & (1 << i))
 				continue;
 
-			if (adev->jpeg.inst[i].ring_dec.sched.ready)
-				++num_rings;
+			for (j = 0; j < adev->jpeg.num_jpeg_rings; j++)
+				if (adev->jpeg.inst[i].ring_dec[j].sched.ready)
+					++num_rings;
 		}
 		ib_start_alignment = 16;
 		ib_size_alignment = 16;
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index 71fe7f6f9889c..1c5b60604a193 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -437,7 +437,7 @@ static int jpeg_v1_0_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case 126:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
@@ -484,7 +484,7 @@ int jpeg_v1_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "jpeg_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq,
@@ -492,7 +492,7 @@ int jpeg_v1_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->jpeg.internal.jpeg_pitch = adev->jpeg.inst->external.jpeg_pitch =
+	adev->jpeg.internal.jpeg_pitch[0] = adev->jpeg.inst->external.jpeg_pitch[0] =
 		SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH);
 
 	return 0;
@@ -509,7 +509,7 @@ void jpeg_v1_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	amdgpu_ring_fini(&adev->jpeg.inst[0].ring_dec);
+	amdgpu_ring_fini(adev->jpeg.inst->ring_dec);
 }
 
 /**
@@ -522,7 +522,7 @@ void jpeg_v1_0_sw_fini(void *handle)
  */
 void jpeg_v1_0_start(struct amdgpu_device *adev, int mode)
 {
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 
 	if (mode == 0) {
 		WREG32_SOC15(JPEG, 0, mmUVD_LMI_JRBC_RB_VMID, 0);
@@ -579,7 +579,7 @@ static const struct amdgpu_ring_funcs jpeg_v1_0_decode_ring_vm_funcs = {
 
 static void jpeg_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec.funcs = &jpeg_v1_0_decode_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec->funcs = &jpeg_v1_0_decode_ring_vm_funcs;
 	DRM_INFO("JPEG decode is enabled in VM mode\n");
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
index 3a43e42f4834d..3aeeceae34a5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
@@ -83,7 +83,7 @@ static int jpeg_v2_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1;
 	ring->vm_hub = AMDGPU_MMHUB0(0);
@@ -93,8 +93,8 @@ static int jpeg_v2_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
-	adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH);
+	adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
+	adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH);
 
 	return 0;
 }
@@ -129,7 +129,7 @@ static int jpeg_v2_0_sw_fini(void *handle)
 static int jpeg_v2_0_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
@@ -312,7 +312,7 @@ static void jpeg_v2_0_enable_clock_gating(struct amdgpu_device *adev)
  */
 static int jpeg_v2_0_start(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	if (adev->pm.dpm_enabled)
@@ -729,7 +729,7 @@ static int jpeg_v2_0_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case VCN_2_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
@@ -791,7 +791,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = {
 
 static void jpeg_v2_0_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec.funcs = &jpeg_v2_0_dec_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec->funcs = &jpeg_v2_0_dec_ring_vm_funcs;
 	DRM_INFO("JPEG decode is enabled in VM mode\n");
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index 259b7ba6a842c..b79edb12b90e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -125,7 +125,7 @@ static int jpeg_v2_5_sw_init(void *handle)
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 
-		ring = &adev->jpeg.inst[i].ring_dec;
+		ring = adev->jpeg.inst[i].ring_dec;
 		ring->use_doorbell = true;
 		if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0))
 			ring->vm_hub = AMDGPU_MMHUB1(0);
@@ -138,8 +138,8 @@ static int jpeg_v2_5_sw_init(void *handle)
 		if (r)
 			return r;
 
-		adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
-		adev->jpeg.inst[i].external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_PITCH);
+		adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
+		adev->jpeg.inst[i].external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_PITCH);
 	}
 
 	r = amdgpu_jpeg_ras_sw_init(adev);
@@ -186,7 +186,7 @@ static int jpeg_v2_5_hw_init(void *handle)
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 
-		ring = &adev->jpeg.inst[i].ring_dec;
+		ring = adev->jpeg.inst[i].ring_dec;
 		adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
 			(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 8 * i, i);
 
@@ -326,7 +326,7 @@ static int jpeg_v2_5_start(struct amdgpu_device *adev)
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 
-		ring = &adev->jpeg.inst[i].ring_dec;
+		ring = adev->jpeg.inst[i].ring_dec;
 		/* disable anti hang mechanism */
 		WREG32_P(SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_POWER_STATUS), 0,
 			~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
@@ -591,7 +591,7 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case VCN_2_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[ip_instance].ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst[ip_instance].ring_dec);
 		break;
 	case VCN_2_6__SRCID_DJPEG0_POISON:
 	case VCN_2_6__SRCID_EJPEG0_POISON:
@@ -712,10 +712,10 @@ static void jpeg_v2_5_set_dec_ring_funcs(struct amdgpu_device *adev)
 		if (adev->jpeg.harvest_config & (1 << i))
 			continue;
 		if (adev->asic_type == CHIP_ARCTURUS)
-			adev->jpeg.inst[i].ring_dec.funcs = &jpeg_v2_5_dec_ring_vm_funcs;
+			adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v2_5_dec_ring_vm_funcs;
 		else  /* CHIP_ALDEBARAN */
-			adev->jpeg.inst[i].ring_dec.funcs = &jpeg_v2_6_dec_ring_vm_funcs;
-		adev->jpeg.inst[i].ring_dec.me = i;
+			adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v2_6_dec_ring_vm_funcs;
+		adev->jpeg.inst[i].ring_dec->me = i;
 		DRM_INFO("JPEG(%d) JPEG decode is enabled in VM mode\n", i);
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index c55386c22311a..cb5494effc0fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -98,7 +98,7 @@ static int jpeg_v3_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1;
 	ring->vm_hub = AMDGPU_MMHUB0(0);
@@ -108,8 +108,8 @@ static int jpeg_v3_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
-	adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH);
+	adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET;
+	adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH);
 
 	return 0;
 }
@@ -144,7 +144,7 @@ static int jpeg_v3_0_sw_fini(void *handle)
 static int jpeg_v3_0_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
@@ -330,7 +330,7 @@ static int jpeg_v3_0_enable_static_power_gating(struct amdgpu_device *adev)
  */
 static int jpeg_v3_0_start(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	if (adev->pm.dpm_enabled)
@@ -527,7 +527,7 @@ static int jpeg_v3_0_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case VCN_2_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
@@ -589,7 +589,7 @@ static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = {
 
 static void jpeg_v3_0_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec.funcs = &jpeg_v3_0_dec_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec->funcs = &jpeg_v3_0_dec_ring_vm_funcs;
 	DRM_INFO("JPEG decode is enabled in VM mode\n");
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index d7d5ffc293937..495facb885f4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -105,7 +105,7 @@ static int jpeg_v4_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = true;
 	ring->doorbell_index = amdgpu_sriov_vf(adev) ? (((adev->doorbell_index.vcn.vcn_ring0_1) << 1) + 4) : ((adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1);
 	ring->vm_hub = AMDGPU_MMHUB0(0);
@@ -116,8 +116,8 @@ static int jpeg_v4_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->jpeg.internal.jpeg_pitch = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
-	adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
+	adev->jpeg.internal.jpeg_pitch[0] = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
+	adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
 
 	r = amdgpu_jpeg_ras_sw_init(adev);
 	if (r)
@@ -156,7 +156,7 @@ static int jpeg_v4_0_sw_fini(void *handle)
 static int jpeg_v4_0_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	if (amdgpu_sriov_vf(adev)) {
@@ -363,7 +363,7 @@ static int jpeg_v4_0_enable_static_power_gating(struct amdgpu_device *adev)
  */
 static int jpeg_v4_0_start(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	if (adev->pm.dpm_enabled)
@@ -441,7 +441,7 @@ static int jpeg_v4_0_start_sriov(struct amdgpu_device *adev)
 
 	table_size = 0;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 
 	MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(JPEG, 0,
 		regUVD_LMI_JRBC_RB_64BIT_BAR_LOW),
@@ -678,7 +678,7 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case VCN_4_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
 	case VCN_4_0__SRCID_DJPEG0_POISON:
 	case VCN_4_0__SRCID_EJPEG0_POISON:
@@ -744,7 +744,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = {
 
 static void jpeg_v4_0_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec.funcs = &jpeg_v4_0_dec_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec->funcs = &jpeg_v4_0_dec_ring_vm_funcs;
 	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 1fc72f9b52ed3..784c83994ca10 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -85,7 +85,7 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	ring->use_doorbell = false;
 	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "jpeg_dec");
@@ -94,8 +94,8 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->jpeg.internal.jpeg_pitch = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
-	adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
+	adev->jpeg.internal.jpeg_pitch[0] = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
+	adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
 
 	return 0;
 }
@@ -130,7 +130,7 @@ static int jpeg_v4_0_3_sw_fini(void *handle)
 static int jpeg_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
 	r = amdgpu_ring_test_helper(ring);
@@ -254,7 +254,7 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
  */
 static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 
 	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
 		1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
@@ -675,7 +675,7 @@ static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 
 	switch (entry->src_id) {
 	case VCN_2_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec);
+		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -737,8 +737,8 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 
 static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec.funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
-	adev->jpeg.inst->ring_dec.me = 0;
+	adev->jpeg.inst->ring_dec->funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
+	adev->jpeg.inst->ring_dec->me = 0;
 	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index f877c39c7cdde..16feb491adf5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -211,7 +211,7 @@ static int vcn_v1_0_hw_init(void *handle)
 			goto done;
 	}
 
-	ring = &adev->jpeg.inst->ring_dec;
+	ring = adev->jpeg.inst->ring_dec;
 	r = amdgpu_ring_test_helper(ring);
 	if (r)
 		goto done;
@@ -1304,7 +1304,7 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_device *adev,
 							UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK_MASK);
 
 				/* Restore */
-				ring = &adev->jpeg.inst->ring_dec;
+				ring = adev->jpeg.inst->ring_dec;
 				WREG32_SOC15(UVD, 0, mmUVD_LMI_JRBC_RB_VMID, 0);
 				WREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_CNTL,
 							UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK |
@@ -1802,7 +1802,7 @@ static void vcn_v1_0_idle_work_handler(struct work_struct *work)
 		else
 			new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
 
-		if (amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec))
+		if (amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec))
 			new_state.jpeg = VCN_DPG_STATE__PAUSE;
 		else
 			new_state.jpeg = VCN_DPG_STATE__UNPAUSE;
@@ -1810,7 +1810,7 @@ static void vcn_v1_0_idle_work_handler(struct work_struct *work)
 		adev->vcn.pause_dpg_mode(adev, 0, &new_state);
 	}
 
-	fences += amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec);
+	fences += amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec);
 	fences += amdgpu_fence_count_emitted(&adev->vcn.inst->ring_dec);
 
 	if (fences == 0) {
@@ -1832,7 +1832,7 @@ static void vcn_v1_0_ring_begin_use(struct amdgpu_ring *ring)
 
 	mutex_lock(&adev->vcn.vcn1_jpeg1_workaround);
 
-	if (amdgpu_fence_wait_empty(&ring->adev->jpeg.inst->ring_dec))
+	if (amdgpu_fence_wait_empty(ring->adev->jpeg.inst->ring_dec))
 		DRM_ERROR("VCN dec: jpeg dec ring may not be empty\n");
 
 	vcn_v1_0_set_pg_for_begin_use(ring, set_clocks);
@@ -1864,7 +1864,7 @@ void vcn_v1_0_set_pg_for_begin_use(struct amdgpu_ring *ring, bool set_clocks)
 		else
 			new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
 
-		if (amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec))
+		if (amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec))
 			new_state.jpeg = VCN_DPG_STATE__PAUSE;
 		else
 			new_state.jpeg = VCN_DPG_STATE__UNPAUSE;
-- 
GitLab


From db77081fe3c88a31eaade8a9c565c48c4d51b093 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 19 Jan 2022 23:32:43 -0500
Subject: [PATCH 1034/1791] drm/amdgpu/jpeg: add multiple jpeg rings support
 for vcn4_0_3

Add multiple jpeg rings support for vcn4_0_3

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 214 ++++++++++++++++-------
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h |   1 +
 2 files changed, 147 insertions(+), 68 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 784c83994ca10..0d3509409d3a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -29,7 +29,7 @@
 
 #include "vcn/vcn_4_0_3_offset.h"
 #include "vcn/vcn_4_0_3_sh_mask.h"
-#include "ivsrcid/vcn/irqsrcs_vcn_2_0.h"
+#include "ivsrcid/vcn/irqsrcs_vcn_4_0.h"
 
 enum jpeg_engin_status {
 	UVD_PGFSM_STATUS__UVDJ_PWR_ON  = 0,
@@ -41,6 +41,17 @@ static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev);
 static int jpeg_v4_0_3_set_powergating_state(void *handle,
 				enum amd_powergating_state state);
 
+static int amdgpu_ih_srcid_jpeg[] = {
+	VCN_4_0__SRCID__JPEG_DECODE,
+	VCN_4_0__SRCID__JPEG1_DECODE,
+	VCN_4_0__SRCID__JPEG2_DECODE,
+	VCN_4_0__SRCID__JPEG3_DECODE,
+	VCN_4_0__SRCID__JPEG4_DECODE,
+	VCN_4_0__SRCID__JPEG5_DECODE,
+	VCN_4_0__SRCID__JPEG6_DECODE,
+	VCN_4_0__SRCID__JPEG7_DECODE
+};
+
 /**
  * jpeg_v4_0_3_early_init - set function pointers
  *
@@ -69,13 +80,15 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int r;
-
-	/* JPEG TRAP */
-	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-		VCN_2_0__SRCID__JPEG_DECODE, &adev->jpeg.inst->irq);
-	if (r)
-		return r;
+	int i, r;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		/* JPEG TRAP */
+		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
+				amdgpu_ih_srcid_jpeg[i], &adev->jpeg.inst->irq);
+		if (r)
+			return r;
+	}
 
 	r = amdgpu_jpeg_sw_init(adev);
 	if (r)
@@ -85,17 +98,22 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = adev->jpeg.inst->ring_dec;
-	ring->use_doorbell = false;
-	ring->vm_hub = AMDGPU_MMHUB0(0);
-	sprintf(ring->name, "jpeg_dec");
-	r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
-			     AMDGPU_RING_PRIO_DEFAULT, NULL);
-	if (r)
-		return r;
-
-	adev->jpeg.internal.jpeg_pitch[0] = regUVD_JPEG_PITCH_INTERNAL_OFFSET;
-	adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH);
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		ring = &adev->jpeg.inst->ring_dec[i];
+		ring->use_doorbell = false;
+		ring->vm_hub = AMDGPU_MMHUB0(0);
+		sprintf(ring->name, "jpeg_dec_%d", i);
+		r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
+					AMDGPU_RING_PRIO_DEFAULT, NULL);
+		if (r)
+			return r;
+
+		adev->jpeg.internal.jpeg_pitch[i] =
+			regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET;
+		adev->jpeg.inst->external.jpeg_pitch[i] =
+			SOC15_REG_OFFSET1(JPEG, 0, regUVD_JRBC0_UVD_JRBC_SCRATCH0,
+			(i?(0x40 * i - 0xc80):0));
+	}
 
 	return 0;
 }
@@ -130,14 +148,18 @@ static int jpeg_v4_0_3_sw_fini(void *handle)
 static int jpeg_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
-	int r;
+	struct amdgpu_ring *ring;
+	int i, r;
 
-	r = amdgpu_ring_test_helper(ring);
-	if (!r)
-		DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n");
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		ring = &adev->jpeg.inst->ring_dec[i];
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
+			return r;
+	}
+	DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n");
 
-	return r;
+	return 0;
 }
 
 /**
@@ -150,13 +172,14 @@ static int jpeg_v4_0_3_hw_init(void *handle)
 static int jpeg_v4_0_3_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret = 0;
 
 	cancel_delayed_work_sync(&adev->jpeg.idle_work);
 
 	if (adev->jpeg.cur_state != AMD_PG_STATE_GATE)
-		jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE);
+		ret = jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE);
 
-	return 0;
+	return ret;
 }
 
 /**
@@ -204,6 +227,7 @@ static int jpeg_v4_0_3_resume(void *handle)
 static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 {
 	uint32_t data;
+	int i;
 
 	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
@@ -216,16 +240,16 @@ static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
 
 	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
-	data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK
-		| JPEG_CGC_GATE__JPEG2_DEC_MASK
-		| JPEG_CGC_GATE__JMCIF_MASK
-		| JPEG_CGC_GATE__JRBBM_MASK);
+	data &= ~(JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
+		data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
 	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
 }
 
 static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 {
 	uint32_t data;
+	int i;
 
 	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
@@ -238,10 +262,9 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
 
 	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
-	data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK
-		|JPEG_CGC_GATE__JPEG2_DEC_MASK
-		|JPEG_CGC_GATE__JMCIF_MASK
-		|JPEG_CGC_GATE__JRBBM_MASK);
+	data |= (JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
+		data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
 	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
 }
 
@@ -255,6 +278,7 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
+	int i;
 
 	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
 		1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
@@ -280,22 +304,32 @@ static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI_CNTL), 0,
 		~UVD_JMI_CNTL__SOFT_RESET_MASK);
 
-	/* enable System Interrupt for JRBC */
-	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regJPEG_SYS_INT_EN),
-		JPEG_SYS_INT_EN__DJRBC0_MASK,
-		~JPEG_SYS_INT_EN__DJRBC0_MASK);
-
-	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, 0);
-	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, (0x00000001L | 0x00000002L));
-	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
-		lower_32_bits(ring->gpu_addr));
-	WREG32_SOC15(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
-		upper_32_bits(ring->gpu_addr));
-	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_RPTR, 0);
-	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR, 0);
-	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, 0x00000002L);
-	WREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_SIZE, ring->ring_size / 4);
-	ring->wptr = RREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR);
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
+
+		ring = &adev->jpeg.inst->ring_dec[i];
+
+		/* enable System Interrupt for JRBC */
+		WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regJPEG_SYS_INT_EN),
+			JPEG_SYS_INT_EN__DJRBC0_MASK << i,
+			~(JPEG_SYS_INT_EN__DJRBC0_MASK << i));
+
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, reg_offset, 0);
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
+			(0x00000001L | 0x00000002L));
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
+			reg_offset, lower_32_bits(ring->gpu_addr));
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
+			reg_offset, upper_32_bits(ring->gpu_addr));
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_RPTR, reg_offset, 0);
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR, reg_offset, 0);
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
+			0x00000002L);
+		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_SIZE, reg_offset,
+			ring->ring_size / 4);
+		ring->wptr = RREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+			reg_offset);
+	}
 
 	return 0;
 }
@@ -342,7 +376,8 @@ static uint64_t jpeg_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 
-	return RREG32_SOC15(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_RPTR);
+	return RREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_RPTR,
+			ring->pipe?(0x40 * ring->pipe - 0xc80):0);
 }
 
 /**
@@ -359,7 +394,8 @@ static uint64_t jpeg_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
 	if (ring->use_doorbell)
 		return adev->wb.wb[ring->wptr_offs];
 	else
-		return RREG32_SOC15(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR);
+		return RREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+			ring->pipe?(0x40 * ring->pipe - 0xc80):0);
 }
 
 /**
@@ -377,8 +413,8 @@ static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
 		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
 		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
 	} else {
-		WREG32_SOC15(JPEG, ring->me,
-			regUVD_JRBC0_UVD_JRBC_RB_WPTR, lower_32_bits(ring->wptr));
+		WREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+			(ring->pipe?(0x40 * ring->pipe - 0xc80):0), lower_32_bits(ring->wptr));
 	}
 }
 
@@ -393,7 +429,7 @@ static void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring)
 {
 	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, 0x62a04);/* TODO: PCTL0_MMHUB_DEEPSLEEP_IB */
+	amdgpu_ring_write(ring, 0x62a04); /* PCTL0_MMHUB_DEEPSLEEP_IB */
 
 	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
 		0, 0, PACKETJ_TYPE0));
@@ -605,20 +641,36 @@ static void jpeg_v4_0_3_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count)
 static bool jpeg_v4_0_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool ret;
+	int i;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
+
+		ret &= ((RREG32_SOC15_OFFSET(JPEG, 0,
+				regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset) &
+				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
+				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+	}
 
-	return ((RREG32_SOC15(JPEG, 0, regUVD_JRBC0_UVD_JRBC_STATUS) &
-		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
-		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+	return ret;
 }
 
 static int jpeg_v4_0_3_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int ret;
+	int i;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
+
+		ret &= SOC15_WAIT_ON_RREG_OFFSET(JPEG, 0,
+			regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset,
+			UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
+			UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+	}
 
-	ret = SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_JRBC0_UVD_JRBC_STATUS,
-		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
-		UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
 	return ret;
 }
 
@@ -626,7 +678,7 @@ static int jpeg_v4_0_3_set_clockgating_state(void *handle,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	bool enable = (state == AMD_CG_STATE_GATE);
+	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
 
 	if (enable) {
 		if (!jpeg_v4_0_3_is_idle(handle))
@@ -674,8 +726,29 @@ static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 	DRM_DEV_DEBUG(adev->dev, "IH: JPEG TRAP\n");
 
 	switch (entry->src_id) {
-	case VCN_2_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
+	case VCN_4_0__SRCID__JPEG_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[0]);
+		break;
+	case VCN_4_0__SRCID__JPEG1_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[1]);
+		break;
+	case VCN_4_0__SRCID__JPEG2_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[2]);
+		break;
+	case VCN_4_0__SRCID__JPEG3_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[3]);
+		break;
+	case VCN_4_0__SRCID__JPEG4_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[4]);
+		break;
+	case VCN_4_0__SRCID__JPEG5_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[5]);
+		break;
+	case VCN_4_0__SRCID__JPEG6_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[6]);
+		break;
+	case VCN_4_0__SRCID__JPEG7_DECODE:
+		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[7]);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -737,8 +810,13 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 
 static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->ring_dec->funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
-	adev->jpeg.inst->ring_dec->me = 0;
+	int i;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+		adev->jpeg.inst->ring_dec[i].funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
+		adev->jpeg.inst->ring_dec[i].me = 0;
+		adev->jpeg.inst->ring_dec[i].pipe = i;
+	}
 	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
 }
 
@@ -749,7 +827,7 @@ static const struct amdgpu_irq_src_funcs jpeg_v4_0_3_irq_funcs = {
 
 static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->irq.num_types = 1;
+	adev->jpeg.inst->irq.num_types = adev->jpeg.num_jpeg_rings;
 	adev->jpeg.inst->irq.funcs = &jpeg_v4_0_3_irq_funcs;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
index ca03d17e13fa6..70a5f030d5f7d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
@@ -41,6 +41,7 @@
 #define regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET				0x4084
 #define regUVD_JRBC_STATUS_INTERNAL_OFFSET				0x4089
 #define regUVD_JPEG_PITCH_INTERNAL_OFFSET				0x4043
+#define regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET			0x4094
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
-- 
GitLab


From bfb44eacb0e2de63bc7824cc590ede51a02a7ded Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Thu, 27 Jan 2022 09:02:16 -0500
Subject: [PATCH 1035/1791] drm/amdkfd: Set F8_MODE for gc_v9_4_3

Set F8_MODE for GC 9.4.3 as optimal/non-IEEE. Also update gc_v9_0
to gc_v9_4_3 to include more definitions such as the F8_MODE bit, and
remove unused header files.

v2: fix IP version check (Alex)

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
index 8b2dd2670ab7f..914d94679d739 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
@@ -24,9 +24,7 @@
 
 #include "kfd_device_queue_manager.h"
 #include "vega10_enum.h"
-#include "gc/gc_9_0_offset.h"
-#include "gc/gc_9_0_sh_mask.h"
-#include "sdma0/sdma0_4_0_sh_mask.h"
+#include "gc/gc_9_4_3_sh_mask.h"
 
 static int update_qpd_v9(struct device_queue_manager *dqm,
 			 struct qcm_process_device *qpd);
@@ -65,6 +63,10 @@ static int update_qpd_v9(struct device_queue_manager *dqm,
 		if (dqm->dev->noretry && !dqm->dev->use_iommu_v2)
 			qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT;
 
+		if (KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 3))
+			qpd->sh_mem_config |=
+				(1 << SH_MEM_CONFIG__F8_MODE__SHIFT);
+
 		qpd->sh_mem_ape1_limit = 0;
 		qpd->sh_mem_ape1_base = 0;
 	}
-- 
GitLab


From d7fd2a9e394b5f2bb900fc4e1e04e8dd4a97a7be Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 29 Jan 2022 10:24:26 -0500
Subject: [PATCH 1036/1791] drm/amdgpu/nbio: update vcn doorbell range

VCN4.0.3 used up to 16 doorbells per partition.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index bdb84a53f0e44..266b504fd83e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -161,7 +161,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
-				0x8);
+				0x10);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
@@ -174,7 +174,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x4);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
-				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x8);
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x10);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
-- 
GitLab


From c21d446ba7a83ab9f15fae8f9c732bb94f8a5677 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 29 Jan 2022 10:28:26 -0500
Subject: [PATCH 1037/1791] drm/amdgpu/vcn: enable vcn doorbell for vcn4.0.3

Enable vcn doorbell for vcn4.0.3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 98bff162f4535..4418c9f05ec4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -109,7 +109,8 @@ static int vcn_v4_0_3_sw_init(void *handle)
 		return r;
 
 	ring = &adev->vcn.inst->ring_dec;
-	ring->use_doorbell = false;
+	ring->use_doorbell = true;
+	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 5;
 	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "vcn_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
@@ -174,6 +175,13 @@ static int vcn_v4_0_3_hw_init(void *handle)
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 	int r;
 
+	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+				ring->doorbell_index, ring->me);
+	if (ring->use_doorbell)
+		WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
+			ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
+			VCN_RB4_DB_CTRL__EN_MASK);
+
 	r = amdgpu_ring_test_helper(ring);
 
 	if (!r)
-- 
GitLab


From 6ddae0f3ab18a64e83bcf7b090e085394046f130 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 29 Jan 2022 10:34:05 -0500
Subject: [PATCH 1038/1791] drm/amdgpu/jpeg: enable jpeg doorbell for jpeg4.0.3

Enable jpeg doorbell for jpeg4.0.3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 0d3509409d3a0..8914f3c6c80f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -100,8 +100,9 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
 		ring = &adev->jpeg.inst->ring_dec[i];
-		ring->use_doorbell = false;
+		ring->use_doorbell = true;
 		ring->vm_hub = AMDGPU_MMHUB0(0);
+		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + (i?8:1) + i;
 		sprintf(ring->name, "jpeg_dec_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 					AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -148,11 +149,19 @@ static int jpeg_v4_0_3_sw_fini(void *handle)
 static int jpeg_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring;
+	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int i, r;
 
+	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+			(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
+
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
 		ring = &adev->jpeg.inst->ring_dec[i];
+		if (ring->use_doorbell)
+			WREG32_SOC15_OFFSET(VCN, 0, regVCN_JPEG_DB_CTRL,
+				(ring->pipe?(ring->pipe - 0x15):0),
+				ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
+				VCN_JPEG_DB_CTRL__EN_MASK);
 		r = amdgpu_ring_test_helper(ring);
 		if (r)
 			return r;
-- 
GitLab


From ae972ed5e0dcb3fab57020e243563cc484b8969a Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 7 Feb 2022 10:25:47 -0500
Subject: [PATCH 1039/1791] drm/amdgpu: fix vcn doorbell range setting

Should use vcn_ring0_1 instead of doorbell index to
set nbio doorbell range.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 4418c9f05ec4d..ae2cc47d344a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -176,7 +176,7 @@ static int vcn_v4_0_3_hw_init(void *handle)
 	int r;
 
 	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				ring->doorbell_index, ring->me);
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), ring->me);
 	if (ring->use_doorbell)
 		WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
 			ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
-- 
GitLab


From 89cf4549a949b4ba3ce771163b75285979c95602 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 23 May 2022 13:53:45 +0800
Subject: [PATCH 1040/1791] drm/amdgpu: support gc v9_4_3 ring_test running on
 all xcc

Each xcc has its own sratch_reg offset

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ad3e8cefbdb27..4ef39837e4c76 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -227,20 +227,23 @@ static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring)
 	uint32_t tmp = 0;
 	unsigned i;
 	int r;
+	/* scratch_reg0_offset is 32bit even with full XCD config */
+	uint32_t scratch_reg0_offset;
+
+	scratch_reg0_offset = SOC15_REG_OFFSET(GC, ring->xcc_id, regSCRATCH_REG0);
+	WREG32(scratch_reg0_offset, 0xCAFEDEAD);
 
-	WREG32_SOC15(GC, 0, regSCRATCH_REG0, 0xCAFEDEAD);
 	r = amdgpu_ring_alloc(ring, 3);
 	if (r)
 		return r;
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
-	amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0) -
-			  PACKET3_SET_UCONFIG_REG_START);
+	amdgpu_ring_write(ring, scratch_reg0_offset - PACKET3_SET_UCONFIG_REG_START);
 	amdgpu_ring_write(ring, 0xDEADBEEF);
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32_SOC15(GC, 0, regSCRATCH_REG0);
+		tmp = RREG32(scratch_reg0_offset);
 		if (tmp == 0xDEADBEEF)
 			break;
 		udelay(1);
-- 
GitLab


From 147862d00bcf7e23e0a125f910f5db224f7b6722 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Fri, 3 Jun 2022 11:08:12 +0800
Subject: [PATCH 1041/1791] drm/amdgpu: enable the ring and IB test for slave
 kcq

With the mec FW update to utilize the mqd base set by
driver for kcq mapping, slave kcq ring test and IB test
can be re-enabled.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 59 +++++++++++--------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c  |  5 ---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 12 +++--
 3 files changed, 32 insertions(+), 44 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 4a4d71ff9b95d..682c157f2d8f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -449,8 +449,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 
 			ring->mqd_size = mqd_size;
 			/* prepare MQD backup */
-			adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL);
-			if (!adev->gfx.mec.mqd_backup[j])
+			adev->gfx.mec.mqd_backup[j + xcc_id * adev->gfx.num_compute_rings] = kmalloc(mqd_size, GFP_KERNEL);
+			if (!adev->gfx.mec.mqd_backup[j + xcc_id * adev->gfx.num_compute_rings])
 				dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name);
 		}
 	}
@@ -502,22 +502,20 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id)
 		return -EINVAL;
 
 	spin_lock(&kiq->ring_lock);
-	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
-		if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
-						adev->gfx.num_compute_rings)) {
-			spin_unlock(&kiq->ring_lock);
-			return -ENOMEM;
-		}
+	if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size *
+					adev->gfx.num_compute_rings)) {
+		spin_unlock(&kiq->ring_lock);
+		return -ENOMEM;
+	}
 
-		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			j = i + xcc_id * adev->gfx.num_compute_rings;
-			kiq->pmf->kiq_unmap_queues(kiq_ring,
-						   &adev->gfx.compute_ring[i],
-						   RESET_QUEUES, 0, 0);
-		}
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		j = i + xcc_id * adev->gfx.num_compute_rings;
+		kiq->pmf->kiq_unmap_queues(kiq_ring,
+					   &adev->gfx.compute_ring[i],
+					   RESET_QUEUES, 0, 0);
 	}
 
-	if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang)
+	if (kiq_ring->sched.ready && !adev->job_hang)
 		r = amdgpu_ring_test_helper(kiq_ring);
 	spin_unlock(&kiq->ring_lock);
 
@@ -598,26 +596,23 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 	DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe,
 							kiq_ring->queue);
 	spin_lock(&kiq->ring_lock);
-	/* No need to map kcq on the slave */
-	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
-		r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
-						adev->gfx.num_compute_rings +
-						kiq->pmf->set_resources_size);
-		if (r) {
-			DRM_ERROR("Failed to lock KIQ (%d).\n", r);
-			spin_unlock(&kiq->ring_lock);
-			return r;
-		}
+	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
+					adev->gfx.num_compute_rings +
+					kiq->pmf->set_resources_size);
+	if (r) {
+		DRM_ERROR("Failed to lock KIQ (%d).\n", r);
+		spin_unlock(&kiq->ring_lock);
+		return r;
+	}
 
-		if (adev->enable_mes)
-			queue_mask = ~0ULL;
+	if (adev->enable_mes)
+		queue_mask = ~0ULL;
 
-		kiq->pmf->kiq_set_resources(kiq_ring, queue_mask);
-		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			j = i + xcc_id * adev->gfx.num_compute_rings;
+	kiq->pmf->kiq_set_resources(kiq_ring, queue_mask);
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		j = i + xcc_id * adev->gfx.num_compute_rings;
 			kiq->pmf->kiq_map_queues(kiq_ring,
-						 &adev->gfx.compute_ring[i]);
-		}
+						 &adev->gfx.compute_ring[j]);
 	}
 
 	r = amdgpu_ring_test_helper(kiq_ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index c955c3f060cd2..b27ac48ca123e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -433,11 +433,6 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
 		else
 			tmo = tmo_gfx;
 
-		/* skip ib test on the slave kcq */
-		if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE &&
-		    !amdgpu_gfx_is_master_xcc(adev, ring->xcc_id))
-			continue;
-
 		r = amdgpu_ring_test_ib(ring, tmo);
 		if (!r) {
 			DRM_DEV_DEBUG(adev->dev, "ib test on %s succeeded\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 4ef39837e4c76..ec273a217666a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1956,13 +1956,11 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 		if (r)
 			return r;
 
-		/* skip ring test on slave kcq */
-		if (amdgpu_gfx_is_master_xcc(adev, i)) {
-			for (j = 0; j < adev->gfx.num_compute_rings; j++) {
-				ring = &adev->gfx.compute_ring[j +
-					i * adev->gfx.num_compute_rings];
-				amdgpu_ring_test_helper(ring);
-			}
+		for (j = 0; j < adev->gfx.num_compute_rings; j++) {
+			ring = &adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings];
+			r = amdgpu_ring_test_helper(ring);
+			if (r)
+				return r;
 		}
 
 		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
-- 
GitLab


From 15091a6f4380a0c1a7202d52e82cdaaf80e2bb70 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 19 Nov 2021 18:03:34 +0800
Subject: [PATCH 1042/1791] drm/amdgpu: add node_id to physical id conversion
 in EOP handler

A new field nodeid in interrupt cookie indicates the node ID.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Shiwu Zhang <shiwu.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h | 14 ++++++++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c |  6 ++++--
 3 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index fafebec5b7b66..031610c1340a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -99,6 +99,17 @@ const char *soc15_ih_clientid_name[] = {
 	"MP1"
 };
 
+const int node_id_to_phys_map[NODEID_MAX] = {
+	[XCD0_NODEID] = 0,
+	[XCD1_NODEID] = 1,
+	[XCD2_NODEID] = 2,
+	[XCD3_NODEID] = 3,
+	[XCD4_NODEID] = 4,
+	[XCD5_NODEID] = 5,
+	[XCD6_NODEID] = 6,
+	[XCD7_NODEID] = 7,
+};
+
 /**
  * amdgpu_irq_disable_all - disable *all* interrupts
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
index 1c747ac4129a4..efe8a278cbdf9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
@@ -102,6 +102,20 @@ struct amdgpu_irq {
 	bool                            retry_cam_enabled;
 };
 
+enum interrupt_node_id_per_xcp {
+	XCD0_NODEID = 1,
+	XCD1_NODEID = 2,
+	XCD2_NODEID = 5,
+	XCD3_NODEID = 6,
+	XCD4_NODEID = 9,
+	XCD5_NODEID = 10,
+	XCD6_NODEID = 13,
+	XCD7_NODEID = 14,
+	NODEID_MAX,
+};
+
+extern const int node_id_to_phys_map[NODEID_MAX];
+
 void amdgpu_irq_disable_all(struct amdgpu_device *adev);
 
 int amdgpu_irq_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ec273a217666a..ce64c4fc5f1ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2799,7 +2799,7 @@ static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 			    struct amdgpu_irq_src *source,
 			    struct amdgpu_iv_entry *entry)
 {
-	int i;
+	int i, phys_id;
 	u8 me_id, pipe_id, queue_id;
 	struct amdgpu_ring *ring;
 
@@ -2808,12 +2808,14 @@ static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 	pipe_id = (entry->ring_id & 0x03) >> 0;
 	queue_id = (entry->ring_id & 0x70) >> 4;
 
+	phys_id = node_id_to_phys_map[entry->node_id];
+
 	switch (me_id) {
 	case 0:
 	case 1:
 	case 2:
 		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			ring = &adev->gfx.compute_ring[i];
+			ring = &adev->gfx.compute_ring[i + phys_id * adev->gfx.num_compute_rings];
 			/* Per-queue interrupt is supported for MEC starting from VI.
 			  * The interrupt can only be enabled/disabled per pipe instead of per queue.
 			  */
-- 
GitLab


From e6a02e2cc7fe3fec05eeaf08855e57d616a037e1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 18 Feb 2022 15:04:35 +0530
Subject: [PATCH 1043/1791] drm/amdgpu: Add some XCC programming

Add additional XCC programming sequences.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 26 +++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ce64c4fc5f1ae..232feb387a408 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1150,6 +1150,29 @@ static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 	WREG32_SOC15(GC, xcc_id, regCPC_PSP_DEBUG, data);
 }
 
+static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
+{
+	uint32_t tmp = 0;
+
+	switch (adev->gfx.num_xcd) {
+	/* directly config VIRTUAL_XCC_ID to 0 for 1-XCC */
+	case 1:
+		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, 0x8);
+		break;
+	case 2:
+		tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID);
+		tmp = tmp | (adev->gfx.num_xcd << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
+		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, tmp);
+
+		tmp = xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, PHYSICAL_XCC_ID);
+		tmp = tmp | (xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, XCC_DIE_ID));
+		WREG32_SOC15(GC, xcc_id, regCP_PSP_XCP_CTL, tmp);
+		break;
+	default:
+		break;
+	}
+}
+
 static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
 {
 	uint32_t rlc_setting;
@@ -1948,6 +1971,9 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 				return r;
 		}
 
+		/* set the virtual and physical id based on partition_mode */
+		gfx_v9_4_3_program_xcc_id(adev, i);
+
 		r = gfx_v9_4_3_kiq_resume(adev, i);
 		if (r)
 			return r;
-- 
GitLab


From 5cf1675591dd28afc498348757469a87c1e9fcf2 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 28 Feb 2022 12:25:15 +0530
Subject: [PATCH 1044/1791] drm/amdgpu: Add mode2 reset logic for v13.0.6

Mode2 reset for v13.0.6 has similar workflow as v13.0.2

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
index 6437ead87e5fb..eec41ad304060 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
@@ -40,6 +40,7 @@ int amdgpu_reset_init(struct amdgpu_device *adev)
 
 	switch (adev->ip_versions[MP1_HWIP][0]) {
 	case IP_VERSION(13, 0, 2):
+	case IP_VERSION(13, 0, 6):
 		ret = aldebaran_reset_init(adev);
 		break;
 	case IP_VERSION(11, 0, 7):
@@ -61,6 +62,7 @@ int amdgpu_reset_fini(struct amdgpu_device *adev)
 
 	switch (adev->ip_versions[MP1_HWIP][0]) {
 	case IP_VERSION(13, 0, 2):
+	case IP_VERSION(13, 0, 6):
 		ret = aldebaran_reset_fini(adev);
 		break;
 	case IP_VERSION(11, 0, 7):
-- 
GitLab


From 8dc1db3172ae2f17ae71e33b608a33411ce8a1aa Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 14 Sep 2022 16:39:48 +0800
Subject: [PATCH 1045/1791] drm/amdkfd: Introduce kfd_node struct (v5)

Introduce a new structure, kfd_node, which will now represent
a compute node. kfd_node is carved out of kfd_dev structure.
kfd_dev struct now will become the parent of kfd_node, and will
store common resources such as doorbells, GTT sub-alloctor etc.
kfd_node struct will store all resources specific to a compute
node, such as device queue manager, interrupt handling etc.

This is the first step in adding compute partition support in KFD.

v2: introduce kfd_node struct to gc v11 (Hawking)
v3: make reference to kfd_dev struct through kfd_node (Morris)
v4: use kfd_node instead for kfd isr/mqd functions (Morris)
v5: rebase (Alex)

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Morris Zhang <Shiwu.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  |   1 +
 .../gpu/drm/amd/amdkfd/cik_event_interrupt.c  |   4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  43 +--
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  28 +-
 drivers/gpu/drm/amd/amdkfd/kfd_crat.h         |   6 +-
 drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c      |   2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       | 258 +++++++++++-------
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 100 +++----
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |   4 +-
 .../amd/amdkfd/kfd_device_queue_manager_v9.c  |   4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c     |  14 +-
 drivers/gpu/drm/amd/amdkfd/kfd_events.c       |  12 +-
 drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c  |  12 +-
 .../gpu/drm/amd/amdkfd/kfd_int_process_v11.c  |   6 +-
 .../gpu/drm/amd/amdkfd/kfd_int_process_v9.c   |  10 +-
 drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c    |  64 ++---
 drivers/gpu/drm/amd/amdkfd/kfd_iommu.c        |  22 +-
 drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c |  18 +-
 drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h |   2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c      |   8 +-
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c  |   6 +-
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h  |   8 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c  |   6 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |   8 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  |  18 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  24 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c   |  10 +-
 .../gpu/drm/amd/amdkfd/kfd_packet_manager.c   |   2 +-
 .../drm/amd/amdkfd/kfd_packet_manager_v9.c    |   8 +-
 .../drm/amd/amdkfd/kfd_packet_manager_vi.c    |   2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         | 168 +++++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |  54 ++--
 .../amd/amdkfd/kfd_process_queue_manager.c    |  20 +-
 drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c   |  40 +--
 drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h   |   8 +-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c          |   4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c     |  56 ++--
 drivers/gpu/drm/amd/amdkfd/kfd_topology.h     |   8 +-
 38 files changed, 573 insertions(+), 495 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index de6ba0d4b8609..af37f2ef4438b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -35,6 +35,7 @@
 #include "amdgpu_dma_buf.h"
 #include <uapi/linux/kfd_ioctl.h>
 #include "amdgpu_xgmi.h"
+#include "kfd_priv.h"
 #include "kfd_smi_events.h"
 #include <drm/ttm/ttm_tt.h>
 
diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
index 5c8023cba1961..4ebfff6b6c55a 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
@@ -26,7 +26,7 @@
 #include "amdgpu_amdkfd.h"
 #include "kfd_smi_events.h"
 
-static bool cik_event_interrupt_isr(struct kfd_dev *dev,
+static bool cik_event_interrupt_isr(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry,
 					uint32_t *patched_ihre,
 					bool *patched_flag)
@@ -85,7 +85,7 @@ static bool cik_event_interrupt_isr(struct kfd_dev *dev,
 		!amdgpu_no_queue_eviction_on_vm_fault);
 }
 
-static void cik_event_interrupt_wq(struct kfd_dev *dev,
+static void cik_event_interrupt_wq(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry)
 {
 	const struct cik_ih_ring_entry *ihre =
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 81d07ecf666d0..eb0b0b38f10eb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -293,7 +293,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 					void *data)
 {
 	struct kfd_ioctl_create_queue_args *args = data;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	int err = 0;
 	unsigned int queue_id;
 	struct kfd_process_device *pdd;
@@ -328,7 +328,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 	}
 
 	if (!pdd->doorbell_index &&
-	    kfd_alloc_process_doorbells(dev, &pdd->doorbell_index) < 0) {
+	    kfd_alloc_process_doorbells(dev->kfd, &pdd->doorbell_index) < 0) {
 		err = -ENOMEM;
 		goto err_alloc_doorbells;
 	}
@@ -336,7 +336,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 	/* Starting with GFX11, wptr BOs must be mapped to GART for MES to determine work
 	 * on unmapped queues for usermode queue oversubscription (no aggregated doorbell)
 	 */
-	if (dev->shared_resources.enable_mes &&
+	if (dev->kfd->shared_resources.enable_mes &&
 			((dev->adev->mes.sched_version & AMDGPU_MES_API_VERSION_MASK)
 			>> AMDGPU_MES_API_VERSION_SHIFT) >= 2) {
 		struct amdgpu_bo_va_mapping *wptr_mapping;
@@ -887,7 +887,7 @@ static int kfd_ioctl_set_scratch_backing_va(struct file *filep,
 {
 	struct kfd_ioctl_set_scratch_backing_va_args *args = data;
 	struct kfd_process_device *pdd;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	long err;
 
 	mutex_lock(&p->mutex);
@@ -1006,18 +1006,18 @@ static int kfd_ioctl_acquire_vm(struct file *filep, struct kfd_process *p,
 	return ret;
 }
 
-bool kfd_dev_is_large_bar(struct kfd_dev *dev)
+bool kfd_dev_is_large_bar(struct kfd_node *dev)
 {
 	if (debug_largebar) {
 		pr_debug("Simulate large-bar allocation on non large-bar machine\n");
 		return true;
 	}
 
-	if (dev->use_iommu_v2)
+	if (dev->kfd->use_iommu_v2)
 		return false;
 
-	if (dev->local_mem_info.local_mem_size_private == 0 &&
-			dev->local_mem_info.local_mem_size_public > 0)
+	if (dev->kfd->local_mem_info.local_mem_size_private == 0 &&
+	    dev->kfd->local_mem_info.local_mem_size_public > 0)
 		return true;
 	return false;
 }
@@ -1041,7 +1041,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 	struct kfd_ioctl_alloc_memory_of_gpu_args *args = data;
 	struct kfd_process_device *pdd;
 	void *mem;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	int idr_handle;
 	long err;
 	uint64_t offset = args->mmap_offset;
@@ -1105,7 +1105,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 	}
 
 	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) {
-		if (args->size != kfd_doorbell_process_slice(dev)) {
+		if (args->size != kfd_doorbell_process_slice(dev->kfd)) {
 			err = -EINVAL;
 			goto err_unlock;
 		}
@@ -1231,7 +1231,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
 	struct kfd_ioctl_map_memory_to_gpu_args *args = data;
 	struct kfd_process_device *pdd, *peer_pdd;
 	void *mem;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	long err = 0;
 	int i;
 	uint32_t *devices_arr = NULL;
@@ -1405,7 +1405,7 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep,
 		args->n_success = i+1;
 	}
 
-	flush_tlb = kfd_flush_tlb_after_unmap(pdd->dev);
+	flush_tlb = kfd_flush_tlb_after_unmap(pdd->dev->kfd);
 	if (flush_tlb) {
 		err = amdgpu_amdkfd_gpuvm_sync_memory(pdd->dev->adev,
 				(struct kgd_mem *) mem, true);
@@ -1445,7 +1445,7 @@ static int kfd_ioctl_alloc_queue_gws(struct file *filep,
 	int retval;
 	struct kfd_ioctl_alloc_queue_gws_args *args = data;
 	struct queue *q;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 
 	mutex_lock(&p->mutex);
 	q = pqm_get_user_queue(&p->pqm, args->queue_id);
@@ -1482,7 +1482,7 @@ static int kfd_ioctl_get_dmabuf_info(struct file *filep,
 		struct kfd_process *p, void *data)
 {
 	struct kfd_ioctl_get_dmabuf_info_args *args = data;
-	struct kfd_dev *dev = NULL;
+	struct kfd_node *dev = NULL;
 	struct amdgpu_device *dmabuf_adev;
 	void *metadata_buffer = NULL;
 	uint32_t flags;
@@ -1596,7 +1596,7 @@ static int kfd_ioctl_export_dmabuf(struct file *filep,
 	struct kfd_ioctl_export_dmabuf_args *args = data;
 	struct kfd_process_device *pdd;
 	struct dma_buf *dmabuf;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	void *mem;
 	int ret = 0;
 
@@ -2178,7 +2178,7 @@ static int criu_restore_devices(struct kfd_process *p,
 	}
 
 	for (i = 0; i < args->num_devices; i++) {
-		struct kfd_dev *dev;
+		struct kfd_node *dev;
 		struct kfd_process_device *pdd;
 		struct file *drm_file;
 
@@ -2240,7 +2240,7 @@ static int criu_restore_devices(struct kfd_process *p,
 		}
 
 		if (!pdd->doorbell_index &&
-		    kfd_alloc_process_doorbells(pdd->dev, &pdd->doorbell_index) < 0) {
+		    kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) {
 			ret = -ENOMEM;
 			goto exit;
 		}
@@ -2268,7 +2268,8 @@ static int criu_restore_memory_of_gpu(struct kfd_process_device *pdd,
 	u64 offset;
 
 	if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) {
-		if (bo_bucket->size != kfd_doorbell_process_slice(pdd->dev))
+		if (bo_bucket->size !=
+				kfd_doorbell_process_slice(pdd->dev->kfd))
 			return -EINVAL;
 
 		offset = kfd_get_process_doorbells(pdd);
@@ -2350,7 +2351,7 @@ static int criu_restore_bo(struct kfd_process *p,
 
 	/* now map these BOs to GPU/s */
 	for (j = 0; j < p->n_pdds; j++) {
-		struct kfd_dev *peer;
+		struct kfd_node *peer;
 		struct kfd_process_device *peer_pdd;
 
 		if (!bo_priv->mapped_gpuids[j])
@@ -2947,7 +2948,7 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 	return retcode;
 }
 
-static int kfd_mmio_mmap(struct kfd_dev *dev, struct kfd_process *process,
+static int kfd_mmio_mmap(struct kfd_node *dev, struct kfd_process *process,
 		      struct vm_area_struct *vma)
 {
 	phys_addr_t address;
@@ -2981,7 +2982,7 @@ static int kfd_mmio_mmap(struct kfd_dev *dev, struct kfd_process *process,
 static int kfd_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	struct kfd_process *process;
-	struct kfd_dev *dev = NULL;
+	struct kfd_node *dev = NULL;
 	unsigned long mmap_offset;
 	unsigned int gpu_id;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 475e470273540..f5aebba31e886 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -1405,7 +1405,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 	return i;
 }
 
-int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pcache_info)
+int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info)
 {
 	int num_of_cache_types = 0;
 
@@ -1524,7 +1524,7 @@ int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pca
 		case IP_VERSION(11, 0, 3):
 		case IP_VERSION(11, 0, 4):
 			num_of_cache_types =
-				kfd_fill_gpu_cache_info_from_gfx_config(kdev, *pcache_info);
+				kfd_fill_gpu_cache_info_from_gfx_config(kdev->kfd, *pcache_info);
 			break;
 		default:
 			*pcache_info = dummy_cache_info;
@@ -1858,7 +1858,7 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size)
 }
 
 static int kfd_fill_gpu_memory_affinity(int *avail_size,
-		struct kfd_dev *kdev, uint8_t type, uint64_t size,
+		struct kfd_node *kdev, uint8_t type, uint64_t size,
 		struct crat_subtype_memory *sub_type_hdr,
 		uint32_t proximity_domain,
 		const struct kfd_local_mem_info *local_mem_info)
@@ -1887,7 +1887,7 @@ static int kfd_fill_gpu_memory_affinity(int *avail_size,
 }
 
 #ifdef CONFIG_ACPI_NUMA
-static void kfd_find_numa_node_in_srat(struct kfd_dev *kdev)
+static void kfd_find_numa_node_in_srat(struct kfd_node *kdev)
 {
 	struct acpi_table_header *table_header = NULL;
 	struct acpi_subtable_header *sub_header = NULL;
@@ -1982,7 +1982,7 @@ static void kfd_find_numa_node_in_srat(struct kfd_dev *kdev)
  *	Return 0 if successful else return -ve value
  */
 static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
-			struct kfd_dev *kdev,
+			struct kfd_node *kdev,
 			struct crat_subtype_iolink *sub_type_hdr,
 			uint32_t proximity_domain)
 {
@@ -2044,8 +2044,8 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
 }
 
 static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size,
-			struct kfd_dev *kdev,
-			struct kfd_dev *peer_kdev,
+			struct kfd_node *kdev,
+			struct kfd_node *peer_kdev,
 			struct crat_subtype_iolink *sub_type_hdr,
 			uint32_t proximity_domain_from,
 			uint32_t proximity_domain_to)
@@ -2081,7 +2081,7 @@ static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size,
  *		[OUT] actual size of data filled in crat_image
  */
 static int kfd_create_vcrat_image_gpu(void *pcrat_image,
-				      size_t *size, struct kfd_dev *kdev,
+				      size_t *size, struct kfd_node *kdev,
 				      uint32_t proximity_domain)
 {
 	struct crat_header *crat_table = (struct crat_header *)pcrat_image;
@@ -2153,7 +2153,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 	/* Check if this node supports IOMMU. During parsing this flag will
 	 * translate to HSA_CAP_ATS_PRESENT
 	 */
-	if (!kfd_iommu_check_device(kdev))
+	if (!kfd_iommu_check_device(kdev->kfd))
 		cu->hsa_capability |= CRAT_CU_FLAGS_IOMMU_PRESENT;
 
 	crat_table->length += sub_type_hdr->length;
@@ -2164,7 +2164,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 	 * report the total FB size (public+private) as a single
 	 * private heap.
 	 */
-	local_mem_info = kdev->local_mem_info;
+	local_mem_info = kdev->kfd->local_mem_info;
 	sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
 			sub_type_hdr->length);
 
@@ -2216,12 +2216,12 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 	 * (from other GPU to this GPU) will be added
 	 * in kfd_parse_subtype_iolink.
 	 */
-	if (kdev->hive_id) {
+	if (kdev->kfd->hive_id) {
 		for (nid = 0; nid < proximity_domain; ++nid) {
 			peer_dev = kfd_topology_device_by_proximity_domain_no_lock(nid);
 			if (!peer_dev->gpu)
 				continue;
-			if (peer_dev->gpu->hive_id != kdev->hive_id)
+			if (peer_dev->gpu->kfd->hive_id != kdev->kfd->hive_id)
 				continue;
 			sub_type_hdr = (typeof(sub_type_hdr))(
 				(char *)sub_type_hdr +
@@ -2255,12 +2255,12 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
  *		(COMPUTE_UNIT_CPU | COMPUTE_UNIT_GPU) - Create VCRAT for APU
  *			-- this option is not currently implemented.
  *			The assumption is that all AMD APUs will have CRAT
- *	@kdev: Valid kfd_device required if flags contain COMPUTE_UNIT_GPU
+ *	@kdev: Valid kfd_node required if flags contain COMPUTE_UNIT_GPU
  *
  *	Return 0 if successful else return -ve value
  */
 int kfd_create_crat_image_virtual(void **crat_image, size_t *size,
-				  int flags, struct kfd_dev *kdev,
+				  int flags, struct kfd_node *kdev,
 				  uint32_t proximity_domain)
 {
 	void *pcrat_image = NULL;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
index 8d1e8ba58dee8..3d0e533b93b9e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
@@ -293,7 +293,7 @@ struct crat_subtype_generic {
 
 #pragma pack()
 
-struct kfd_dev;
+struct kfd_node;
 
 /* Static table to describe GPU Cache information */
 struct kfd_gpu_cache_info {
@@ -305,14 +305,14 @@ struct kfd_gpu_cache_info {
 	 */
 	uint32_t	num_cu_shared;
 };
-int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pcache_info);
+int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info);
 
 int kfd_create_crat_image_acpi(void **crat_image, size_t *size);
 void kfd_destroy_crat_image(void *crat_image);
 int kfd_parse_crat_table(void *crat_image, struct list_head *device_list,
 			 uint32_t proximity_domain);
 int kfd_create_crat_image_virtual(void **crat_image, size_t *size,
-				  int flags, struct kfd_dev *kdev,
+				  int flags, struct kfd_node *kdev,
 				  uint32_t proximity_domain);
 
 #endif /* KFD_CRAT_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
index ad5a40a685aca..4a5a0a4e00f21 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c
@@ -43,7 +43,7 @@ static int kfd_debugfs_hang_hws_read(struct seq_file *m, void *data)
 static ssize_t kfd_debugfs_hang_hws_write(struct file *file,
 	const char __user *user_buf, size_t size, loff_t *ppos)
 {
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	char tmp[16];
 	uint32_t gpu_id;
 	int ret = -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 1510041a6ee13..23d9a7f770551 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -61,7 +61,7 @@ static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size,
 static void kfd_gtt_sa_fini(struct kfd_dev *kfd);
 
 static int kfd_resume_iommu(struct kfd_dev *kfd);
-static int kfd_resume(struct kfd_dev *kfd);
+static int kfd_resume(struct kfd_node *kfd);
 
 static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd)
 {
@@ -441,8 +441,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 	memset(&kfd->doorbell_available_index, 0,
 		sizeof(kfd->doorbell_available_index));
 
-	atomic_set(&kfd->sram_ecc_flag, 0);
-
 	ida_init(&kfd->doorbell_ida);
 
 	return kfd;
@@ -489,41 +487,106 @@ static void kfd_cwsr_init(struct kfd_dev *kfd)
 	}
 }
 
-static int kfd_gws_init(struct kfd_dev *kfd)
+static int kfd_gws_init(struct kfd_node *node)
 {
 	int ret = 0;
+	struct kfd_dev *kfd = node->kfd;
 
-	if (kfd->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS)
+	if (node->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS)
 		return 0;
 
-	if (hws_gws_support || (KFD_IS_SOC15(kfd) &&
-		((KFD_GC_VERSION(kfd) == IP_VERSION(9, 0, 1)
+	if (hws_gws_support || (KFD_IS_SOC15(node) &&
+		((KFD_GC_VERSION(node) == IP_VERSION(9, 0, 1)
 			&& kfd->mec2_fw_version >= 0x81b3) ||
-		(KFD_GC_VERSION(kfd) <= IP_VERSION(9, 4, 0)
+		(KFD_GC_VERSION(node) <= IP_VERSION(9, 4, 0)
 			&& kfd->mec2_fw_version >= 0x1b3)  ||
-		(KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 1)
+		(KFD_GC_VERSION(node) == IP_VERSION(9, 4, 1)
 			&& kfd->mec2_fw_version >= 0x30)   ||
-		(KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 2)
+		(KFD_GC_VERSION(node) == IP_VERSION(9, 4, 2)
 			&& kfd->mec2_fw_version >= 0x28) ||
-		(KFD_GC_VERSION(kfd) >= IP_VERSION(10, 3, 0)
-			&& KFD_GC_VERSION(kfd) < IP_VERSION(11, 0, 0)
+		(KFD_GC_VERSION(node) >= IP_VERSION(10, 3, 0)
+			&& KFD_GC_VERSION(node) < IP_VERSION(11, 0, 0)
 			&& kfd->mec2_fw_version >= 0x6b))))
-		ret = amdgpu_amdkfd_alloc_gws(kfd->adev,
-				kfd->adev->gds.gws_size, &kfd->gws);
+		ret = amdgpu_amdkfd_alloc_gws(node->adev,
+				node->adev->gds.gws_size, &node->gws);
 
 	return ret;
 }
 
-static void kfd_smi_init(struct kfd_dev *dev)
+static void kfd_smi_init(struct kfd_node *dev)
 {
 	INIT_LIST_HEAD(&dev->smi_clients);
 	spin_lock_init(&dev->smi_lock);
 }
 
+static int kfd_init_node(struct kfd_node *node)
+{
+	int err = -1;
+
+	if (kfd_interrupt_init(node)) {
+		dev_err(kfd_device, "Error initializing interrupts\n");
+		goto kfd_interrupt_error;
+	}
+
+	node->dqm = device_queue_manager_init(node);
+	if (!node->dqm) {
+		dev_err(kfd_device, "Error initializing queue manager\n");
+		goto device_queue_manager_error;
+	}
+
+	if (kfd_gws_init(node)) {
+		dev_err(kfd_device, "Could not allocate %d gws\n",
+			node->adev->gds.gws_size);
+		goto gws_error;
+	}
+
+	if (kfd_resume(node))
+		goto kfd_resume_error;
+
+	if (kfd_topology_add_device(node)) {
+		dev_err(kfd_device, "Error adding device to topology\n");
+		goto kfd_topology_add_device_error;
+	}
+
+	kfd_smi_init(node);
+
+	return 0;
+
+kfd_topology_add_device_error:
+kfd_resume_error:
+gws_error:
+	device_queue_manager_uninit(node->dqm);
+device_queue_manager_error:
+	kfd_interrupt_exit(node);
+kfd_interrupt_error:
+	if (node->gws)
+		amdgpu_amdkfd_free_gws(node->adev, node->gws);
+
+	/* Cleanup the node memory here */
+	kfree(node);
+	return err;
+}
+
+static void kfd_cleanup_node(struct kfd_dev *kfd)
+{
+	struct kfd_node *knode = kfd->node;
+
+	device_queue_manager_uninit(knode->dqm);
+	kfd_interrupt_exit(knode);
+	kfd_topology_remove_device(knode);
+	if (knode->gws)
+		amdgpu_amdkfd_free_gws(knode->adev, knode->gws);
+	kfree(knode);
+	kfd->node = NULL;
+}
+
 bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			 const struct kgd2kfd_shared_resources *gpu_resources)
 {
 	unsigned int size, map_process_packet_size;
+	struct kfd_node *node;
+	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
+	unsigned int max_proc_per_quantum;
 
 	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev,
 			KGD_ENGINE_MEC1);
@@ -533,10 +596,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			KGD_ENGINE_SDMA1);
 	kfd->shared_resources = *gpu_resources;
 
-	kfd->vm_info.first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1;
-	kfd->vm_info.last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1;
-	kfd->vm_info.vmid_num_kfd = kfd->vm_info.last_vmid_kfd
-			- kfd->vm_info.first_vmid_kfd + 1;
+	first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1;
+	last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1;
+	vmid_num_kfd = last_vmid_kfd - first_vmid_kfd + 1;
 
 	/* Allow BIF to recode atomics to PCIe 3.0 AtomicOps.
 	 * 32 and 64-bit requests are possible and must be
@@ -557,9 +619,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	/* Verify module parameters regarding mapped process number*/
 	if (hws_max_conc_proc >= 0)
-		kfd->max_proc_per_quantum = min((u32)hws_max_conc_proc, kfd->vm_info.vmid_num_kfd);
+		max_proc_per_quantum = min((u32)hws_max_conc_proc, vmid_num_kfd);
 	else
-		kfd->max_proc_per_quantum = kfd->vm_info.vmid_num_kfd;
+		max_proc_per_quantum = vmid_num_kfd;
 
 	/* calculate max size of mqds needed for queues */
 	size = max_num_of_queues_per_device *
@@ -609,26 +671,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd->noretry = kfd->adev->gmc.noretry;
 
-	if (kfd_interrupt_init(kfd)) {
-		dev_err(kfd_device, "Error initializing interrupts\n");
-		goto kfd_interrupt_error;
-	}
-
-	kfd->dqm = device_queue_manager_init(kfd);
-	if (!kfd->dqm) {
-		dev_err(kfd_device, "Error initializing queue manager\n");
-		goto device_queue_manager_error;
-	}
-
-	/* If supported on this device, allocate global GWS that is shared
-	 * by all KFD processes
-	 */
-	if (kfd_gws_init(kfd)) {
-		dev_err(kfd_device, "Could not allocate %d gws\n",
-			kfd->adev->gds.gws_size);
-		goto gws_error;
-	}
-
 	/* If CRAT is broken, won't set iommu enabled */
 	kfd_double_confirm_iommu_support(kfd);
 
@@ -642,46 +684,54 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	svm_migrate_init(kfd->adev);
 
-	if (kfd_resume_iommu(kfd))
-		goto device_iommu_error;
-
-	if (kfd_resume(kfd))
-		goto kfd_resume_error;
-
-	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
+	/* Allocate the KFD node */
+	node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL);
+	if (!node) {
+		dev_err(kfd_device, "Error allocating KFD node\n");
+		goto node_alloc_error;
+	}
 
-	if (kfd_topology_add_device(kfd)) {
-		dev_err(kfd_device, "Error adding device to topology\n");
-		goto kfd_topology_add_device_error;
+	node->adev = kfd->adev;
+	node->kfd = kfd;
+	node->kfd2kgd = kfd->kfd2kgd;
+	node->vm_info.vmid_num_kfd = vmid_num_kfd;
+	node->vm_info.first_vmid_kfd = first_vmid_kfd;
+	node->vm_info.last_vmid_kfd = last_vmid_kfd;
+	node->max_proc_per_quantum = max_proc_per_quantum;
+	atomic_set(&node->sram_ecc_flag, 0);
+
+	/* Initialize the KFD node */
+	if (kfd_init_node(node)) {
+		dev_err(kfd_device, "Error initializing KFD node\n");
+		goto node_init_error;
 	}
+	kfd->node = node;
 
-	kfd_smi_init(kfd);
+	if (kfd_resume_iommu(kfd))
+		goto kfd_resume_iommu_error;
+
+	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
 
 	kfd->init_complete = true;
 	dev_info(kfd_device, "added device %x:%x\n", kfd->adev->pdev->vendor,
 		 kfd->adev->pdev->device);
 
 	pr_debug("Starting kfd with the following scheduling policy %d\n",
-		kfd->dqm->sched_policy);
+		node->dqm->sched_policy);
 
 	goto out;
 
-kfd_topology_add_device_error:
-kfd_resume_error:
+kfd_resume_iommu_error:
+	kfd_cleanup_node(kfd);
+node_init_error:
+node_alloc_error:
 device_iommu_error:
-gws_error:
-	device_queue_manager_uninit(kfd->dqm);
-device_queue_manager_error:
-	kfd_interrupt_exit(kfd);
-kfd_interrupt_error:
 	kfd_doorbell_fini(kfd);
 kfd_doorbell_error:
 	kfd_gtt_sa_fini(kfd);
 kfd_gtt_sa_init_error:
 	amdgpu_amdkfd_free_gtt_mem(kfd->adev, kfd->gtt_mem);
 alloc_gtt_mem_failure:
-	if (kfd->gws)
-		amdgpu_amdkfd_free_gws(kfd->adev, kfd->gws);
 	dev_err(kfd_device,
 		"device %x:%x NOT added due to errors\n",
 		kfd->adev->pdev->vendor, kfd->adev->pdev->device);
@@ -692,15 +742,11 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 void kgd2kfd_device_exit(struct kfd_dev *kfd)
 {
 	if (kfd->init_complete) {
-		device_queue_manager_uninit(kfd->dqm);
-		kfd_interrupt_exit(kfd);
-		kfd_topology_remove_device(kfd);
+		kfd_cleanup_node(kfd);
 		kfd_doorbell_fini(kfd);
 		ida_destroy(&kfd->doorbell_ida);
 		kfd_gtt_sa_fini(kfd);
 		amdgpu_amdkfd_free_gtt_mem(kfd->adev, kfd->gtt_mem);
-		if (kfd->gws)
-			amdgpu_amdkfd_free_gws(kfd->adev, kfd->gws);
 	}
 
 	kfree(kfd);
@@ -708,16 +754,18 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
 
 int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 {
+	struct kfd_node *node = kfd->node;
+
 	if (!kfd->init_complete)
 		return 0;
 
-	kfd_smi_event_update_gpu_reset(kfd, false);
+	kfd_smi_event_update_gpu_reset(node, false);
 
-	kfd->dqm->ops.pre_reset(kfd->dqm);
+	node->dqm->ops.pre_reset(node->dqm);
 
 	kgd2kfd_suspend(kfd, false);
 
-	kfd_signal_reset_event(kfd);
+	kfd_signal_reset_event(node);
 	return 0;
 }
 
@@ -730,18 +778,19 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 int kgd2kfd_post_reset(struct kfd_dev *kfd)
 {
 	int ret;
+	struct kfd_node *node = kfd->node;
 
 	if (!kfd->init_complete)
 		return 0;
 
-	ret = kfd_resume(kfd);
+	ret = kfd_resume(node);
 	if (ret)
 		return ret;
 	atomic_dec(&kfd_locked);
 
-	atomic_set(&kfd->sram_ecc_flag, 0);
+	atomic_set(&node->sram_ecc_flag, 0);
 
-	kfd_smi_event_update_gpu_reset(kfd, true);
+	kfd_smi_event_update_gpu_reset(node, true);
 
 	return 0;
 }
@@ -753,6 +802,8 @@ bool kfd_is_locked(void)
 
 void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
+	struct kfd_node *node = kfd->node;
+
 	if (!kfd->init_complete)
 		return;
 
@@ -763,18 +814,19 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 			kfd_suspend_all_processes();
 	}
 
-	kfd->dqm->ops.stop(kfd->dqm);
+	node->dqm->ops.stop(node->dqm);
 	kfd_iommu_suspend(kfd);
 }
 
 int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
 	int ret, count;
+	struct kfd_node *node = kfd->node;
 
 	if (!kfd->init_complete)
 		return 0;
 
-	ret = kfd_resume(kfd);
+	ret = kfd_resume(node);
 	if (ret)
 		return ret;
 
@@ -809,15 +861,15 @@ static int kfd_resume_iommu(struct kfd_dev *kfd)
 	return err;
 }
 
-static int kfd_resume(struct kfd_dev *kfd)
+static int kfd_resume(struct kfd_node *node)
 {
 	int err = 0;
 
-	err = kfd->dqm->ops.start(kfd->dqm);
+	err = node->dqm->ops.start(node->dqm);
 	if (err)
 		dev_err(kfd_device,
 			"Error starting queue manager for device %x:%x\n",
-			kfd->adev->pdev->vendor, kfd->adev->pdev->device);
+			node->adev->pdev->vendor, node->adev->pdev->device);
 
 	return err;
 }
@@ -843,6 +895,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 	uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE];
 	bool is_patched = false;
 	unsigned long flags;
+	struct kfd_node *node = kfd->node;
 
 	if (!kfd->init_complete)
 		return;
@@ -852,16 +905,16 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 		return;
 	}
 
-	spin_lock_irqsave(&kfd->interrupt_lock, flags);
+	spin_lock_irqsave(&node->interrupt_lock, flags);
 
-	if (kfd->interrupts_active
-	    && interrupt_is_wanted(kfd, ih_ring_entry,
+	if (node->interrupts_active
+	    && interrupt_is_wanted(node, ih_ring_entry,
 				   patched_ihre, &is_patched)
-	    && enqueue_ih_ring_entry(kfd,
+	    && enqueue_ih_ring_entry(node,
 				     is_patched ? patched_ihre : ih_ring_entry))
-		kfd_queue_work(kfd->ih_wq, &kfd->interrupt_work);
+		kfd_queue_work(node->ih_wq, &node->interrupt_work);
 
-	spin_unlock_irqrestore(&kfd->interrupt_lock, flags);
+	spin_unlock_irqrestore(&node->interrupt_lock, flags);
 }
 
 int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger)
@@ -999,10 +1052,11 @@ static inline uint32_t *kfd_gtt_sa_calc_cpu_addr(void *start_addr,
 	return (uint32_t *) ((uint64_t) start_addr + bit_num * chunk_size);
 }
 
-int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
+int kfd_gtt_sa_allocate(struct kfd_node *node, unsigned int size,
 			struct kfd_mem_obj **mem_obj)
 {
 	unsigned int found, start_search, cur_size;
+	struct kfd_dev *kfd = node->kfd;
 
 	if (size == 0)
 		return -EINVAL;
@@ -1102,8 +1156,10 @@ int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
 	return -ENOMEM;
 }
 
-int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj)
+int kfd_gtt_sa_free(struct kfd_node *node, struct kfd_mem_obj *mem_obj)
 {
+	struct kfd_dev *kfd = node->kfd;
+
 	/* Act like kfree when trying to free a NULL object */
 	if (!mem_obj)
 		return 0;
@@ -1126,28 +1182,28 @@ int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj)
 void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd)
 {
 	if (kfd)
-		atomic_inc(&kfd->sram_ecc_flag);
+		atomic_inc(&kfd->node->sram_ecc_flag);
 }
 
-void kfd_inc_compute_active(struct kfd_dev *kfd)
+void kfd_inc_compute_active(struct kfd_node *node)
 {
-	if (atomic_inc_return(&kfd->compute_profile) == 1)
-		amdgpu_amdkfd_set_compute_idle(kfd->adev, false);
+	if (atomic_inc_return(&node->kfd->compute_profile) == 1)
+		amdgpu_amdkfd_set_compute_idle(node->adev, false);
 }
 
-void kfd_dec_compute_active(struct kfd_dev *kfd)
+void kfd_dec_compute_active(struct kfd_node *node)
 {
-	int count = atomic_dec_return(&kfd->compute_profile);
+	int count = atomic_dec_return(&node->kfd->compute_profile);
 
 	if (count == 0)
-		amdgpu_amdkfd_set_compute_idle(kfd->adev, true);
+		amdgpu_amdkfd_set_compute_idle(node->adev, true);
 	WARN_ONCE(count < 0, "Compute profile ref. count error");
 }
 
 void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask)
 {
 	if (kfd && kfd->init_complete)
-		kfd_smi_event_update_thermal_throttling(kfd, throttle_bitmask);
+		kfd_smi_event_update_thermal_throttling(kfd->node, throttle_bitmask);
 }
 
 /* kfd_get_num_sdma_engines returns the number of PCIe optimized SDMA and
@@ -1155,19 +1211,19 @@ void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask)
  * When the device has more than two engines, we reserve two for PCIe to enable
  * full-duplex and the rest are used as XGMI.
  */
-unsigned int kfd_get_num_sdma_engines(struct kfd_dev *kdev)
+unsigned int kfd_get_num_sdma_engines(struct kfd_node *node)
 {
 	/* If XGMI is not supported, all SDMA engines are PCIe */
-	if (!kdev->adev->gmc.xgmi.supported)
-		return kdev->adev->sdma.num_instances;
+	if (!node->adev->gmc.xgmi.supported)
+		return node->adev->sdma.num_instances;
 
-	return min(kdev->adev->sdma.num_instances, 2);
+	return min(node->adev->sdma.num_instances, 2);
 }
 
-unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev)
+unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node)
 {
 	/* After reserved for PCIe, the rest of engines are XGMI */
-	return kdev->adev->sdma.num_instances - kfd_get_num_sdma_engines(kdev);
+	return node->adev->sdma.num_instances - kfd_get_num_sdma_engines(node);
 }
 
 #if defined(CONFIG_DEBUG_FS)
@@ -1175,7 +1231,7 @@ unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev)
 /* This function will send a package to HIQ to hang the HWS
  * which will trigger a GPU reset and bring the HWS back to normal state
  */
-int kfd_debugfs_hang_hws(struct kfd_dev *dev)
+int kfd_debugfs_hang_hws(struct kfd_node *dev)
 {
 	if (dev->dqm->sched_policy != KFD_SCHED_POLICY_HWS) {
 		pr_err("HWS is not enabled");
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 7a95698d83f73..34977d89f01c4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -74,31 +74,31 @@ enum KFD_MQD_TYPE get_mqd_type_from_queue_type(enum kfd_queue_type type)
 static bool is_pipe_enabled(struct device_queue_manager *dqm, int mec, int pipe)
 {
 	int i;
-	int pipe_offset = (mec * dqm->dev->shared_resources.num_pipe_per_mec
-		+ pipe) * dqm->dev->shared_resources.num_queue_per_pipe;
+	int pipe_offset = (mec * dqm->dev->kfd->shared_resources.num_pipe_per_mec
+		+ pipe) * dqm->dev->kfd->shared_resources.num_queue_per_pipe;
 
 	/* queue is available for KFD usage if bit is 1 */
-	for (i = 0; i <  dqm->dev->shared_resources.num_queue_per_pipe; ++i)
+	for (i = 0; i <  dqm->dev->kfd->shared_resources.num_queue_per_pipe; ++i)
 		if (test_bit(pipe_offset + i,
-			      dqm->dev->shared_resources.cp_queue_bitmap))
+			      dqm->dev->kfd->shared_resources.cp_queue_bitmap))
 			return true;
 	return false;
 }
 
 unsigned int get_cp_queues_num(struct device_queue_manager *dqm)
 {
-	return bitmap_weight(dqm->dev->shared_resources.cp_queue_bitmap,
+	return bitmap_weight(dqm->dev->kfd->shared_resources.cp_queue_bitmap,
 				KGD_MAX_QUEUES);
 }
 
 unsigned int get_queues_per_pipe(struct device_queue_manager *dqm)
 {
-	return dqm->dev->shared_resources.num_queue_per_pipe;
+	return dqm->dev->kfd->shared_resources.num_queue_per_pipe;
 }
 
 unsigned int get_pipes_per_mec(struct device_queue_manager *dqm)
 {
-	return dqm->dev->shared_resources.num_pipe_per_mec;
+	return dqm->dev->kfd->shared_resources.num_pipe_per_mec;
 }
 
 static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm)
@@ -110,18 +110,18 @@ static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm)
 unsigned int get_num_sdma_queues(struct device_queue_manager *dqm)
 {
 	return kfd_get_num_sdma_engines(dqm->dev) *
-		dqm->dev->device_info.num_sdma_queues_per_engine;
+		dqm->dev->kfd->device_info.num_sdma_queues_per_engine;
 }
 
 unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm)
 {
 	return kfd_get_num_xgmi_sdma_engines(dqm->dev) *
-		dqm->dev->device_info.num_sdma_queues_per_engine;
+		dqm->dev->kfd->device_info.num_sdma_queues_per_engine;
 }
 
 static inline uint64_t get_reserved_sdma_queues_bitmap(struct device_queue_manager *dqm)
 {
-	return dqm->dev->device_info.reserved_sdma_queues_bitmap;
+	return dqm->dev->kfd->device_info.reserved_sdma_queues_bitmap;
 }
 
 void program_sh_mem_settings(struct device_queue_manager *dqm,
@@ -330,7 +330,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd,
 			     struct queue *q,
 			     uint32_t const *restore_id)
 {
-	struct kfd_dev *dev = qpd->dqm->dev;
+	struct kfd_node *dev = qpd->dqm->dev;
 
 	if (!KFD_IS_SOC15(dev)) {
 		/* On pre-SOC15 chips we need to use the queue ID to
@@ -349,7 +349,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd,
 		 * for a SDMA engine is 512.
 		 */
 
-		uint32_t *idx_offset = dev->shared_resources.sdma_doorbell_idx;
+		uint32_t *idx_offset = dev->kfd->shared_resources.sdma_doorbell_idx;
 		uint32_t valid_id = idx_offset[q->properties.sdma_engine_id]
 						+ (q->properties.sdma_queue_id & 1)
 						* KFD_QUEUE_DOORBELL_MIRROR_OFFSET
@@ -382,7 +382,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd,
 	}
 
 	q->properties.doorbell_off =
-		kfd_get_doorbell_dw_offset_in_bar(dev, qpd_to_pdd(qpd),
+		kfd_get_doorbell_dw_offset_in_bar(dev->kfd, qpd_to_pdd(qpd),
 					  q->doorbell_id);
 	return 0;
 }
@@ -391,7 +391,7 @@ static void deallocate_doorbell(struct qcm_process_device *qpd,
 				struct queue *q)
 {
 	unsigned int old;
-	struct kfd_dev *dev = qpd->dqm->dev;
+	struct kfd_node *dev = qpd->dqm->dev;
 
 	if (!KFD_IS_SOC15(dev) ||
 	    q->properties.type == KFD_QUEUE_TYPE_SDMA ||
@@ -441,7 +441,7 @@ static int allocate_vmid(struct device_queue_manager *dqm,
 
 	program_sh_mem_settings(dqm, qpd);
 
-	if (KFD_IS_SOC15(dqm->dev) && dqm->dev->cwsr_enabled)
+	if (KFD_IS_SOC15(dqm->dev) && dqm->dev->kfd->cwsr_enabled)
 		program_trap_handler_settings(dqm, qpd);
 
 	/* qpd->page_table_base is set earlier when register_process()
@@ -460,7 +460,7 @@ static int allocate_vmid(struct device_queue_manager *dqm,
 	return 0;
 }
 
-static int flush_texture_cache_nocpsch(struct kfd_dev *kdev,
+static int flush_texture_cache_nocpsch(struct kfd_node *kdev,
 				struct qcm_process_device *qpd)
 {
 	const struct packet_manager_funcs *pmf = qpd->dqm->packet_mgr.pmf;
@@ -661,7 +661,7 @@ static inline void deallocate_hqd(struct device_queue_manager *dqm,
 #define SQ_IND_CMD_CMD_KILL		0x00000003
 #define SQ_IND_CMD_MODE_BROADCAST	0x00000001
 
-static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p)
+static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process *p)
 {
 	int status = 0;
 	unsigned int vmid;
@@ -837,7 +837,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 
 	/* Make sure the queue is unmapped before updating the MQD */
 	if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) {
-		if (!dqm->dev->shared_resources.enable_mes)
+		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			retval = unmap_queues_cpsch(dqm,
 						    KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, false);
 		else if (prev_active)
@@ -858,7 +858,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 		}
 
 		retval = mqd_mgr->destroy_mqd(mqd_mgr, q->mqd,
-				(dqm->dev->cwsr_enabled ?
+				(dqm->dev->kfd->cwsr_enabled ?
 				 KFD_PREEMPT_TYPE_WAVEFRONT_SAVE :
 				 KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN),
 				KFD_UNMAP_LATENCY_MS, q->pipe, q->queue);
@@ -895,7 +895,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 	}
 
 	if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) {
-		if (!dqm->dev->shared_resources.enable_mes)
+		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			retval = map_queues_cpsch(dqm);
 		else if (q->properties.is_active)
 			retval = add_queue_mes(dqm, q, &pdd->qpd);
@@ -951,7 +951,7 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm,
 			continue;
 
 		retval = mqd_mgr->destroy_mqd(mqd_mgr, q->mqd,
-				(dqm->dev->cwsr_enabled ?
+				(dqm->dev->kfd->cwsr_enabled ?
 				 KFD_PREEMPT_TYPE_WAVEFRONT_SAVE :
 				 KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN),
 				KFD_UNMAP_LATENCY_MS, q->pipe, q->queue);
@@ -993,7 +993,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		q->properties.is_active = false;
 		decrement_queue_count(dqm, qpd, q);
 
-		if (dqm->dev->shared_resources.enable_mes) {
+		if (dqm->dev->kfd->shared_resources.enable_mes) {
 			retval = remove_queue_mes(dqm, q, qpd);
 			if (retval) {
 				pr_err("Failed to evict queue %d\n",
@@ -1003,7 +1003,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		}
 	}
 	pdd->last_evict_timestamp = get_jiffies_64();
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		retval = execute_queues_cpsch(dqm,
 					      qpd->is_debug ?
 					      KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES :
@@ -1132,7 +1132,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 		q->properties.is_active = true;
 		increment_queue_count(dqm, &pdd->qpd, q);
 
-		if (dqm->dev->shared_resources.enable_mes) {
+		if (dqm->dev->kfd->shared_resources.enable_mes) {
 			retval = add_queue_mes(dqm, q, qpd);
 			if (retval) {
 				pr_err("Failed to restore queue %d\n",
@@ -1141,7 +1141,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 			}
 		}
 	}
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		retval = execute_queues_cpsch(dqm,
 					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 	qpd->evicted = 0;
@@ -1282,7 +1282,7 @@ static int initialize_nocpsch(struct device_queue_manager *dqm)
 
 		for (queue = 0; queue < get_queues_per_pipe(dqm); queue++)
 			if (test_bit(pipe_offset + queue,
-				     dqm->dev->shared_resources.cp_queue_bitmap))
+				     dqm->dev->kfd->shared_resources.cp_queue_bitmap))
 				dqm->allocated_queues[pipe] |= 1 << queue;
 	}
 
@@ -1426,14 +1426,14 @@ static int set_sched_resources(struct device_queue_manager *dqm)
 	int i, mec;
 	struct scheduling_resources res;
 
-	res.vmid_mask = dqm->dev->shared_resources.compute_vmid_bitmap;
+	res.vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap;
 
 	res.queue_mask = 0;
 	for (i = 0; i < KGD_MAX_QUEUES; ++i) {
-		mec = (i / dqm->dev->shared_resources.num_queue_per_pipe)
-			/ dqm->dev->shared_resources.num_pipe_per_mec;
+		mec = (i / dqm->dev->kfd->shared_resources.num_queue_per_pipe)
+			/ dqm->dev->kfd->shared_resources.num_pipe_per_mec;
 
-		if (!test_bit(i, dqm->dev->shared_resources.cp_queue_bitmap))
+		if (!test_bit(i, dqm->dev->kfd->shared_resources.cp_queue_bitmap))
 			continue;
 
 		/* only acquire queues from the first MEC */
@@ -1489,7 +1489,7 @@ static int start_cpsch(struct device_queue_manager *dqm)
 
 	dqm_lock(dqm);
 
-	if (!dqm->dev->shared_resources.enable_mes) {
+	if (!dqm->dev->kfd->shared_resources.enable_mes) {
 		retval = pm_init(&dqm->packet_mgr, dqm);
 		if (retval)
 			goto fail_packet_manager_init;
@@ -1516,14 +1516,14 @@ static int start_cpsch(struct device_queue_manager *dqm)
 	dqm->is_hws_hang = false;
 	dqm->is_resetting = false;
 	dqm->sched_running = true;
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 	dqm_unlock(dqm);
 
 	return 0;
 fail_allocate_vidmem:
 fail_set_sched_resources:
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		pm_uninit(&dqm->packet_mgr, false);
 fail_packet_manager_init:
 	dqm_unlock(dqm);
@@ -1541,7 +1541,7 @@ static int stop_cpsch(struct device_queue_manager *dqm)
 	}
 
 	if (!dqm->is_hws_hang) {
-		if (!dqm->dev->shared_resources.enable_mes)
+		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, false);
 		else
 			remove_all_queues_mes(dqm);
@@ -1550,11 +1550,11 @@ static int stop_cpsch(struct device_queue_manager *dqm)
 	hanging = dqm->is_hws_hang || dqm->is_resetting;
 	dqm->sched_running = false;
 
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		pm_release_ib(&dqm->packet_mgr);
 
 	kfd_gtt_sa_free(dqm->dev, dqm->fence_mem);
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		pm_uninit(&dqm->packet_mgr, hanging);
 	dqm_unlock(dqm);
 
@@ -1673,7 +1673,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	if (q->properties.is_active) {
 		increment_queue_count(dqm, qpd, q);
 
-		if (!dqm->dev->shared_resources.enable_mes)
+		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			retval = execute_queues_cpsch(dqm,
 					KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 		else
@@ -1893,7 +1893,7 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 	list_del(&q->list);
 	qpd->queue_count--;
 	if (q->properties.is_active) {
-		if (!dqm->dev->shared_resources.enable_mes) {
+		if (!dqm->dev->kfd->shared_resources.enable_mes) {
 			decrement_queue_count(dqm, qpd, q);
 			retval = execute_queues_cpsch(dqm,
 						      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
@@ -2056,7 +2056,7 @@ static int get_wave_state(struct device_queue_manager *dqm,
 	mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_CP];
 
 	if (q->properties.type != KFD_QUEUE_TYPE_COMPUTE ||
-	    q->properties.is_active || !q->device->cwsr_enabled ||
+	    q->properties.is_active || !q->device->kfd->cwsr_enabled ||
 	    !mqd_mgr->get_wave_state) {
 		dqm_unlock(dqm);
 		return -EINVAL;
@@ -2105,7 +2105,7 @@ static int checkpoint_mqd(struct device_queue_manager *dqm,
 
 	dqm_lock(dqm);
 
-	if (q->properties.is_active || !q->device->cwsr_enabled) {
+	if (q->properties.is_active || !q->device->kfd->cwsr_enabled) {
 		r = -EINVAL;
 		goto dqm_unlock;
 	}
@@ -2158,7 +2158,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm,
 		if (q->properties.is_active) {
 			decrement_queue_count(dqm, qpd, q);
 
-			if (dqm->dev->shared_resources.enable_mes) {
+			if (dqm->dev->kfd->shared_resources.enable_mes) {
 				retval = remove_queue_mes(dqm, q, qpd);
 				if (retval)
 					pr_err("Failed to remove queue %d\n",
@@ -2180,7 +2180,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm,
 		}
 	}
 
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		retval = execute_queues_cpsch(dqm, filter, 0);
 
 	if ((!dqm->is_hws_hang) && (retval || qpd->reset_wavefronts)) {
@@ -2242,11 +2242,11 @@ static int init_mqd_managers(struct device_queue_manager *dqm)
 static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm)
 {
 	int retval;
-	struct kfd_dev *dev = dqm->dev;
+	struct kfd_node *dev = dqm->dev;
 	struct kfd_mem_obj *mem_obj = &dqm->hiq_sdma_mqd;
 	uint32_t size = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size *
 		get_num_all_sdma_engines(dqm) *
-		dev->device_info.num_sdma_queues_per_engine +
+		dev->kfd->device_info.num_sdma_queues_per_engine +
 		dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size;
 
 	retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size,
@@ -2256,7 +2256,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm)
 	return retval;
 }
 
-struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
+struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev)
 {
 	struct device_queue_manager *dqm;
 
@@ -2373,7 +2373,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
 	if (init_mqd_managers(dqm))
 		goto out_free;
 
-	if (!dev->shared_resources.enable_mes && allocate_hiq_sdma_mqd(dqm)) {
+	if (!dev->kfd->shared_resources.enable_mes && allocate_hiq_sdma_mqd(dqm)) {
 		pr_err("Failed to allocate hiq sdma mqd trunk buffer\n");
 		goto out_free;
 	}
@@ -2386,7 +2386,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev)
 	return NULL;
 }
 
-static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev,
+static void deallocate_hiq_sdma_mqd(struct kfd_node *dev,
 				    struct kfd_mem_obj *mqd)
 {
 	WARN(!mqd, "No hiq sdma mqd trunk to free");
@@ -2397,7 +2397,7 @@ static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev,
 void device_queue_manager_uninit(struct device_queue_manager *dqm)
 {
 	dqm->ops.uninitialize(dqm);
-	if (!dqm->dev->shared_resources.enable_mes)
+	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		deallocate_hiq_sdma_mqd(dqm->dev, &dqm->hiq_sdma_mqd);
 	kfree(dqm);
 }
@@ -2479,7 +2479,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 
 		for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) {
 			if (!test_bit(pipe_offset + queue,
-				      dqm->dev->shared_resources.cp_queue_bitmap))
+				      dqm->dev->kfd->shared_resources.cp_queue_bitmap))
 				continue;
 
 			r = dqm->dev->kfd2kgd->hqd_dump(
@@ -2497,7 +2497,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 
 	for (pipe = 0; pipe < get_num_all_sdma_engines(dqm); pipe++) {
 		for (queue = 0;
-		     queue < dqm->dev->device_info.num_sdma_queues_per_engine;
+		     queue < dqm->dev->kfd->device_info.num_sdma_queues_per_engine;
 		     queue++) {
 			r = dqm->dev->kfd2kgd->hqd_sdma_dump(
 				dqm->dev->adev, pipe, queue, &dump, &n_regs);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index a537b9ef3e167..e554a48f30542 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -207,7 +207,7 @@ struct device_queue_manager_asic_ops {
 				struct queue *q,
 				struct qcm_process_device *qpd);
 	struct mqd_manager *	(*mqd_manager_init)(enum KFD_MQD_TYPE type,
-				 struct kfd_dev *dev);
+				 struct kfd_node *dev);
 };
 
 /**
@@ -228,7 +228,7 @@ struct device_queue_manager {
 
 	struct mqd_manager	*mqd_mgrs[KFD_MQD_TYPE_MAX];
 	struct packet_manager	packet_mgr;
-	struct kfd_dev		*dev;
+	struct kfd_node		*dev;
 	struct mutex		lock_hidden; /* use dqm_lock/unlock(dqm) */
 	struct list_head	queues;
 	unsigned int		saved_flags;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
index 914d94679d739..8af6433887687 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
@@ -60,10 +60,10 @@ static int update_qpd_v9(struct device_queue_manager *dqm,
 		qpd->sh_mem_config = SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
 					SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
 
-		if (dqm->dev->noretry && !dqm->dev->use_iommu_v2)
+		if (dqm->dev->kfd->noretry && !dqm->dev->kfd->use_iommu_v2)
 			qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT;
 
-		if (KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 3))
+		if (KFD_GC_VERSION(dqm->dev->kfd) == IP_VERSION(9, 4, 3))
 			qpd->sh_mem_config |=
 				(1 << SH_MEM_CONFIG__F8_MODE__SHIFT);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
index 38c9e1ca66913..6421b620388de 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c
@@ -138,7 +138,7 @@ void kfd_doorbell_fini(struct kfd_dev *kfd)
 		iounmap(kfd->doorbell_kernel_ptr);
 }
 
-int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
+int kfd_doorbell_mmap(struct kfd_node *dev, struct kfd_process *process,
 		      struct vm_area_struct *vma)
 {
 	phys_addr_t address;
@@ -148,7 +148,7 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
 	 * For simplicitly we only allow mapping of the entire doorbell
 	 * allocation of a single device & process.
 	 */
-	if (vma->vm_end - vma->vm_start != kfd_doorbell_process_slice(dev))
+	if (vma->vm_end - vma->vm_start != kfd_doorbell_process_slice(dev->kfd))
 		return -EINVAL;
 
 	pdd = kfd_get_process_device_data(dev, process);
@@ -170,13 +170,13 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
 		 "     vm_flags            == 0x%04lX\n"
 		 "     size                == 0x%04lX\n",
 		 (unsigned long long) vma->vm_start, address, vma->vm_flags,
-		 kfd_doorbell_process_slice(dev));
+		 kfd_doorbell_process_slice(dev->kfd));
 
 
 	return io_remap_pfn_range(vma,
 				vma->vm_start,
 				address >> PAGE_SHIFT,
-				kfd_doorbell_process_slice(dev),
+				kfd_doorbell_process_slice(dev->kfd),
 				vma->vm_page_prot);
 }
 
@@ -278,14 +278,14 @@ uint64_t kfd_get_number_elems(struct kfd_dev *kfd)
 phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd)
 {
 	if (!pdd->doorbell_index) {
-		int r = kfd_alloc_process_doorbells(pdd->dev,
+		int r = kfd_alloc_process_doorbells(pdd->dev->kfd,
 						    &pdd->doorbell_index);
 		if (r < 0)
 			return 0;
 	}
 
-	return pdd->dev->doorbell_base +
-		pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev);
+	return pdd->dev->kfd->doorbell_base +
+		pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev->kfd);
 }
 
 int kfd_alloc_process_doorbells(struct kfd_dev *kfd, unsigned int *doorbell_index)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index c894cf8f7c50d..9926186f88a65 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -348,7 +348,7 @@ static int kfd_event_page_set(struct kfd_process *p, void *kernel_address,
 
 int kfd_kmap_event_page(struct kfd_process *p, uint64_t event_page_offset)
 {
-	struct kfd_dev *kfd;
+	struct kfd_node *kfd;
 	struct kfd_process_device *pdd;
 	void *mem, *kern_addr;
 	uint64_t size;
@@ -1125,7 +1125,7 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p,
 }
 
 #ifdef KFD_SUPPORT_IOMMU_V2
-void kfd_signal_iommu_event(struct kfd_dev *dev, u32 pasid,
+void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid,
 		unsigned long address, bool is_write_requested,
 		bool is_execute_requested)
 {
@@ -1221,8 +1221,8 @@ void kfd_signal_hw_exception_event(u32 pasid)
 	kfd_unref_process(p);
 }
 
-void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid,
-				struct kfd_vm_fault_info *info)
+void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
+			       struct kfd_vm_fault_info *info)
 {
 	struct kfd_event *ev;
 	uint32_t id;
@@ -1269,7 +1269,7 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid,
 	kfd_unref_process(p);
 }
 
-void kfd_signal_reset_event(struct kfd_dev *dev)
+void kfd_signal_reset_event(struct kfd_node *dev)
 {
 	struct kfd_hsa_hw_exception_data hw_exception_data;
 	struct kfd_hsa_memory_exception_data memory_exception_data;
@@ -1325,7 +1325,7 @@ void kfd_signal_reset_event(struct kfd_dev *dev)
 	srcu_read_unlock(&kfd_processes_srcu, idx);
 }
 
-void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid)
+void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid)
 {
 	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
 	struct kfd_hsa_memory_exception_data memory_exception_data;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
index 8aebe408c5449..da2ca00d79e5e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c
@@ -322,21 +322,21 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id)
 	pdd->lds_base = MAKE_LDS_APP_BASE_VI();
 	pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base);
 
-	if (!pdd->dev->use_iommu_v2) {
+	if (!pdd->dev->kfd->use_iommu_v2) {
 		/* dGPUs: SVM aperture starting at 0
 		 * with small reserved space for kernel.
 		 * Set them to CANONICAL addresses.
 		 */
 		pdd->gpuvm_base = SVM_USER_BASE;
 		pdd->gpuvm_limit =
-			pdd->dev->shared_resources.gpuvm_size - 1;
+			pdd->dev->kfd->shared_resources.gpuvm_size - 1;
 	} else {
 		/* set them to non CANONICAL addresses, and no SVM is
 		 * allocated.
 		 */
 		pdd->gpuvm_base = MAKE_GPUVM_APP_BASE_VI(id + 1);
 		pdd->gpuvm_limit = MAKE_GPUVM_APP_LIMIT(pdd->gpuvm_base,
-				pdd->dev->shared_resources.gpuvm_size);
+				pdd->dev->kfd->shared_resources.gpuvm_size);
 	}
 
 	pdd->scratch_base = MAKE_SCRATCH_APP_BASE_VI();
@@ -356,7 +356,7 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id)
 	 */
 	pdd->gpuvm_base = SVM_USER_BASE;
 	pdd->gpuvm_limit =
-		pdd->dev->shared_resources.gpuvm_size - 1;
+		pdd->dev->kfd->shared_resources.gpuvm_size - 1;
 
 	pdd->scratch_base = MAKE_SCRATCH_APP_BASE_V9();
 	pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base);
@@ -365,7 +365,7 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id)
 int kfd_init_apertures(struct kfd_process *process)
 {
 	uint8_t id  = 0;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	struct kfd_process_device *pdd;
 
 	/*Iterating over all devices*/
@@ -417,7 +417,7 @@ int kfd_init_apertures(struct kfd_process *process)
 				}
 			}
 
-			if (!dev->use_iommu_v2) {
+			if (!dev->kfd->use_iommu_v2) {
 				/* dGPUs: the reserved space for kernel
 				 * before SVM
 				 */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index 0d53f60674226..0f0fdea4cd8ad 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -187,7 +187,7 @@ static void print_sq_intr_info_error(uint32_t context_id0, uint32_t context_id1)
 		REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1, WGP_ID));
 }
 
-static void event_interrupt_poison_consumption_v11(struct kfd_dev *dev,
+static void event_interrupt_poison_consumption_v11(struct kfd_node *dev,
 				uint16_t pasid, uint16_t source_id)
 {
 	int ret = -EINVAL;
@@ -225,7 +225,7 @@ static void event_interrupt_poison_consumption_v11(struct kfd_dev *dev,
 		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true);
 }
 
-static bool event_interrupt_isr_v11(struct kfd_dev *dev,
+static bool event_interrupt_isr_v11(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry,
 					uint32_t *patched_ihre,
 					bool *patched_flag)
@@ -274,7 +274,7 @@ static bool event_interrupt_isr_v11(struct kfd_dev *dev,
 		  !amdgpu_no_queue_eviction_on_vm_fault);
 }
 
-static void event_interrupt_wq_v11(struct kfd_dev *dev,
+static void event_interrupt_wq_v11(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry)
 {
 	uint16_t source_id, client_id, ring_id, pasid, vmid;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 0b75a37b689b8..861bccb1e9dc5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -90,7 +90,7 @@ enum SQ_INTERRUPT_ERROR_TYPE {
 #define KFD_SQ_INT_DATA__ERR_TYPE_MASK 0xF00000
 #define KFD_SQ_INT_DATA__ERR_TYPE__SHIFT 20
 
-static void event_interrupt_poison_consumption_v9(struct kfd_dev *dev,
+static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 				uint16_t pasid, uint16_t client_id)
 {
 	int old_poison, ret = -EINVAL;
@@ -160,7 +160,7 @@ static bool context_id_expected(struct kfd_dev *dev)
 	}
 }
 
-static bool event_interrupt_isr_v9(struct kfd_dev *dev,
+static bool event_interrupt_isr_v9(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry,
 					uint32_t *patched_ihre,
 					bool *patched_flag)
@@ -206,7 +206,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev,
 
 		*patched_flag = true;
 		memcpy(patched_ihre, ih_ring_entry,
-				dev->device_info.ih_ring_entry_size);
+				dev->kfd->device_info.ih_ring_entry_size);
 
 		pasid = dev->dqm->vmid_pasid[vmid];
 
@@ -235,7 +235,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev,
 		uint32_t context_id =
 			SOC15_CONTEXT_ID0_FROM_IH_ENTRY(ih_ring_entry);
 
-		if (context_id == 0 && context_id_expected(dev))
+		if (context_id == 0 && context_id_expected(dev->kfd))
 			return false;
 	}
 
@@ -253,7 +253,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev,
 		!amdgpu_no_queue_eviction_on_vm_fault);
 }
 
-static void event_interrupt_wq_v9(struct kfd_dev *dev,
+static void event_interrupt_wq_v9(struct kfd_node *dev,
 					const uint32_t *ih_ring_entry)
 {
 	uint16_t source_id, client_id, pasid, vmid;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
index 34772fe742960..dd3c43c1ad70b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c
@@ -50,29 +50,29 @@
 
 static void interrupt_wq(struct work_struct *);
 
-int kfd_interrupt_init(struct kfd_dev *kfd)
+int kfd_interrupt_init(struct kfd_node *node)
 {
 	int r;
 
-	r = kfifo_alloc(&kfd->ih_fifo,
-		KFD_IH_NUM_ENTRIES * kfd->device_info.ih_ring_entry_size,
+	r = kfifo_alloc(&node->ih_fifo,
+		KFD_IH_NUM_ENTRIES * node->kfd->device_info.ih_ring_entry_size,
 		GFP_KERNEL);
 	if (r) {
-		dev_err(kfd->adev->dev, "Failed to allocate IH fifo\n");
+		dev_err(node->adev->dev, "Failed to allocate IH fifo\n");
 		return r;
 	}
 
-	kfd->ih_wq = alloc_workqueue("KFD IH", WQ_HIGHPRI, 1);
-	if (unlikely(!kfd->ih_wq)) {
-		kfifo_free(&kfd->ih_fifo);
-		dev_err(kfd->adev->dev, "Failed to allocate KFD IH workqueue\n");
+	node->ih_wq = alloc_workqueue("KFD IH", WQ_HIGHPRI, 1);
+	if (unlikely(!node->ih_wq)) {
+		kfifo_free(&node->ih_fifo);
+		dev_err(node->adev->dev, "Failed to allocate KFD IH workqueue\n");
 		return -ENOMEM;
 	}
-	spin_lock_init(&kfd->interrupt_lock);
+	spin_lock_init(&node->interrupt_lock);
 
-	INIT_WORK(&kfd->interrupt_work, interrupt_wq);
+	INIT_WORK(&node->interrupt_work, interrupt_wq);
 
-	kfd->interrupts_active = true;
+	node->interrupts_active = true;
 
 	/*
 	 * After this function returns, the interrupt will be enabled. This
@@ -84,7 +84,7 @@ int kfd_interrupt_init(struct kfd_dev *kfd)
 	return 0;
 }
 
-void kfd_interrupt_exit(struct kfd_dev *kfd)
+void kfd_interrupt_exit(struct kfd_node *node)
 {
 	/*
 	 * Stop the interrupt handler from writing to the ring and scheduling
@@ -93,31 +93,31 @@ void kfd_interrupt_exit(struct kfd_dev *kfd)
 	 */
 	unsigned long flags;
 
-	spin_lock_irqsave(&kfd->interrupt_lock, flags);
-	kfd->interrupts_active = false;
-	spin_unlock_irqrestore(&kfd->interrupt_lock, flags);
+	spin_lock_irqsave(&node->interrupt_lock, flags);
+	node->interrupts_active = false;
+	spin_unlock_irqrestore(&node->interrupt_lock, flags);
 
 	/*
 	 * flush_work ensures that there are no outstanding
 	 * work-queue items that will access interrupt_ring. New work items
 	 * can't be created because we stopped interrupt handling above.
 	 */
-	flush_workqueue(kfd->ih_wq);
+	flush_workqueue(node->ih_wq);
 
-	kfifo_free(&kfd->ih_fifo);
+	kfifo_free(&node->ih_fifo);
 }
 
 /*
  * Assumption: single reader/writer. This function is not re-entrant
  */
-bool enqueue_ih_ring_entry(struct kfd_dev *kfd,	const void *ih_ring_entry)
+bool enqueue_ih_ring_entry(struct kfd_node *node, const void *ih_ring_entry)
 {
 	int count;
 
-	count = kfifo_in(&kfd->ih_fifo, ih_ring_entry,
-				kfd->device_info.ih_ring_entry_size);
-	if (count != kfd->device_info.ih_ring_entry_size) {
-		dev_dbg_ratelimited(kfd->adev->dev,
+	count = kfifo_in(&node->ih_fifo, ih_ring_entry,
+				node->kfd->device_info.ih_ring_entry_size);
+	if (count != node->kfd->device_info.ih_ring_entry_size) {
+		dev_dbg_ratelimited(node->adev->dev,
 			"Interrupt ring overflow, dropping interrupt %d\n",
 			count);
 		return false;
@@ -129,32 +129,32 @@ bool enqueue_ih_ring_entry(struct kfd_dev *kfd,	const void *ih_ring_entry)
 /*
  * Assumption: single reader/writer. This function is not re-entrant
  */
-static bool dequeue_ih_ring_entry(struct kfd_dev *kfd, void *ih_ring_entry)
+static bool dequeue_ih_ring_entry(struct kfd_node *node, void *ih_ring_entry)
 {
 	int count;
 
-	count = kfifo_out(&kfd->ih_fifo, ih_ring_entry,
-				kfd->device_info.ih_ring_entry_size);
+	count = kfifo_out(&node->ih_fifo, ih_ring_entry,
+				node->kfd->device_info.ih_ring_entry_size);
 
-	WARN_ON(count && count != kfd->device_info.ih_ring_entry_size);
+	WARN_ON(count && count != node->kfd->device_info.ih_ring_entry_size);
 
-	return count == kfd->device_info.ih_ring_entry_size;
+	return count == node->kfd->device_info.ih_ring_entry_size;
 }
 
 static void interrupt_wq(struct work_struct *work)
 {
-	struct kfd_dev *dev = container_of(work, struct kfd_dev,
+	struct kfd_node *dev = container_of(work, struct kfd_node,
 						interrupt_work);
 	uint32_t ih_ring_entry[KFD_MAX_RING_ENTRY_SIZE];
 	unsigned long start_jiffies = jiffies;
 
-	if (dev->device_info.ih_ring_entry_size > sizeof(ih_ring_entry)) {
+	if (dev->kfd->device_info.ih_ring_entry_size > sizeof(ih_ring_entry)) {
 		dev_err_once(dev->adev->dev, "Ring entry too small\n");
 		return;
 	}
 
 	while (dequeue_ih_ring_entry(dev, ih_ring_entry)) {
-		dev->device_info.event_interrupt_class->interrupt_wq(dev,
+		dev->kfd->device_info.event_interrupt_class->interrupt_wq(dev,
 								ih_ring_entry);
 		if (time_is_before_jiffies(start_jiffies + HZ)) {
 			/* If we spent more than a second processing signals,
@@ -166,14 +166,14 @@ static void interrupt_wq(struct work_struct *work)
 	}
 }
 
-bool interrupt_is_wanted(struct kfd_dev *dev,
+bool interrupt_is_wanted(struct kfd_node *dev,
 			const uint32_t *ih_ring_entry,
 			uint32_t *patched_ihre, bool *flag)
 {
 	/* integer and bitwise OR so there is no boolean short-circuiting */
 	unsigned int wanted = 0;
 
-	wanted |= dev->device_info.event_interrupt_class->interrupt_isr(dev,
+	wanted |= dev->kfd->device_info.event_interrupt_class->interrupt_isr(dev,
 					 ih_ring_entry, patched_ihre, flag);
 
 	return wanted != 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
index ec1bf611624ef..6eee9a0944f3e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
@@ -109,11 +109,11 @@ int kfd_iommu_device_init(struct kfd_dev *kfd)
  */
 int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd)
 {
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 	struct kfd_process *p = pdd->process;
 	int err;
 
-	if (!dev->use_iommu_v2 || pdd->bound == PDD_BOUND)
+	if (!dev->kfd->use_iommu_v2 || pdd->bound == PDD_BOUND)
 		return 0;
 
 	if (unlikely(pdd->bound == PDD_BOUND_SUSPENDED)) {
@@ -146,7 +146,7 @@ void kfd_iommu_unbind_process(struct kfd_process *p)
 /* Callback for process shutdown invoked by the IOMMU driver */
 static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid)
 {
-	struct kfd_dev *dev = kfd_device_by_pci_dev(pdev);
+	struct kfd_node *dev = kfd_device_by_pci_dev(pdev);
 	struct kfd_process *p;
 	struct kfd_process_device *pdd;
 
@@ -182,7 +182,7 @@ static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid)
 static int iommu_invalid_ppr_cb(struct pci_dev *pdev, u32 pasid,
 				unsigned long address, u16 flags)
 {
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 
 	dev_warn_ratelimited(kfd_device,
 			"Invalid PPR device %x:%x.%x pasid 0x%x address 0x%lX flags 0x%X",
@@ -205,7 +205,7 @@ static int iommu_invalid_ppr_cb(struct pci_dev *pdev, u32 pasid,
  * Bind processes do the device that have been temporarily unbound
  * (PDD_BOUND_SUSPENDED) in kfd_unbind_processes_from_device.
  */
-static int kfd_bind_processes_to_device(struct kfd_dev *kfd)
+static int kfd_bind_processes_to_device(struct kfd_node *knode)
 {
 	struct kfd_process_device *pdd;
 	struct kfd_process *p;
@@ -216,14 +216,14 @@ static int kfd_bind_processes_to_device(struct kfd_dev *kfd)
 
 	hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
 		mutex_lock(&p->mutex);
-		pdd = kfd_get_process_device_data(kfd, p);
+		pdd = kfd_get_process_device_data(knode, p);
 
 		if (WARN_ON(!pdd) || pdd->bound != PDD_BOUND_SUSPENDED) {
 			mutex_unlock(&p->mutex);
 			continue;
 		}
 
-		err = amd_iommu_bind_pasid(kfd->adev->pdev, p->pasid,
+		err = amd_iommu_bind_pasid(knode->adev->pdev, p->pasid,
 				p->lead_thread);
 		if (err < 0) {
 			pr_err("Unexpected pasid 0x%x binding failure\n",
@@ -246,7 +246,7 @@ static int kfd_bind_processes_to_device(struct kfd_dev *kfd)
  * processes will be restored to PDD_BOUND state in
  * kfd_bind_processes_to_device.
  */
-static void kfd_unbind_processes_from_device(struct kfd_dev *kfd)
+static void kfd_unbind_processes_from_device(struct kfd_node *knode)
 {
 	struct kfd_process_device *pdd;
 	struct kfd_process *p;
@@ -256,7 +256,7 @@ static void kfd_unbind_processes_from_device(struct kfd_dev *kfd)
 
 	hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
 		mutex_lock(&p->mutex);
-		pdd = kfd_get_process_device_data(kfd, p);
+		pdd = kfd_get_process_device_data(knode, p);
 
 		if (WARN_ON(!pdd)) {
 			mutex_unlock(&p->mutex);
@@ -281,7 +281,7 @@ void kfd_iommu_suspend(struct kfd_dev *kfd)
 	if (!kfd->use_iommu_v2)
 		return;
 
-	kfd_unbind_processes_from_device(kfd);
+	kfd_unbind_processes_from_device(kfd->node);
 
 	amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL);
 	amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL);
@@ -312,7 +312,7 @@ int kfd_iommu_resume(struct kfd_dev *kfd)
 	amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev,
 				     iommu_invalid_ppr_cb);
 
-	err = kfd_bind_processes_to_device(kfd);
+	err = kfd_bind_processes_to_device(kfd->node);
 	if (err) {
 		amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL);
 		amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index bcf7bc3302c9f..1bea629c49ca0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
@@ -38,7 +38,7 @@
 /* Initialize a kernel queue, including allocations of GART memory
  * needed for the queue.
  */
-static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev,
+static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev,
 		enum kfd_queue_type type, unsigned int queue_size)
 {
 	struct queue_properties prop;
@@ -75,7 +75,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 	if (!kq->mqd_mgr)
 		return false;
 
-	prop.doorbell_ptr = kfd_get_kernel_doorbell(dev, &prop.doorbell_off);
+	prop.doorbell_ptr = kfd_get_kernel_doorbell(dev->kfd, &prop.doorbell_off);
 
 	if (!prop.doorbell_ptr) {
 		pr_err("Failed to initialize doorbell");
@@ -112,7 +112,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 	kq->rptr_kernel = kq->rptr_mem->cpu_ptr;
 	kq->rptr_gpu_addr = kq->rptr_mem->gpu_addr;
 
-	retval = kfd_gtt_sa_allocate(dev, dev->device_info.doorbell_size,
+	retval = kfd_gtt_sa_allocate(dev, dev->kfd->device_info.doorbell_size,
 					&kq->wptr_mem);
 
 	if (retval != 0)
@@ -189,7 +189,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 err_eop_allocate_vidmem:
 	kfd_gtt_sa_free(dev, kq->pq);
 err_pq_allocate_vidmem:
-	kfd_release_kernel_doorbell(dev, prop.doorbell_ptr);
+	kfd_release_kernel_doorbell(dev->kfd, prop.doorbell_ptr);
 err_get_kernel_doorbell:
 	return false;
 
@@ -220,7 +220,7 @@ static void kq_uninitialize(struct kernel_queue *kq, bool hanging)
 	kfd_gtt_sa_free(kq->dev, kq->eop_mem);
 
 	kfd_gtt_sa_free(kq->dev, kq->pq);
-	kfd_release_kernel_doorbell(kq->dev,
+	kfd_release_kernel_doorbell(kq->dev->kfd,
 					kq->queue->properties.doorbell_ptr);
 	uninit_queue(kq->queue);
 }
@@ -298,7 +298,7 @@ void kq_submit_packet(struct kernel_queue *kq)
 	}
 	pr_debug("\n");
 #endif
-	if (kq->dev->device_info.doorbell_size == 8) {
+	if (kq->dev->kfd->device_info.doorbell_size == 8) {
 		*kq->wptr64_kernel = kq->pending_wptr64;
 		write_kernel_doorbell64(kq->queue->properties.doorbell_ptr,
 					kq->pending_wptr64);
@@ -311,7 +311,7 @@ void kq_submit_packet(struct kernel_queue *kq)
 
 void kq_rollback_packet(struct kernel_queue *kq)
 {
-	if (kq->dev->device_info.doorbell_size == 8) {
+	if (kq->dev->kfd->device_info.doorbell_size == 8) {
 		kq->pending_wptr64 = *kq->wptr64_kernel;
 		kq->pending_wptr = *kq->wptr_kernel %
 			(kq->queue->properties.queue_size / 4);
@@ -320,7 +320,7 @@ void kq_rollback_packet(struct kernel_queue *kq)
 	}
 }
 
-struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
+struct kernel_queue *kernel_queue_init(struct kfd_node *dev,
 					enum kfd_queue_type type)
 {
 	struct kernel_queue *kq;
@@ -345,7 +345,7 @@ void kernel_queue_uninit(struct kernel_queue *kq, bool hanging)
 }
 
 /* FIXME: Can this test be removed? */
-static __attribute__((unused)) void test_kq(struct kfd_dev *dev)
+static __attribute__((unused)) void test_kq(struct kfd_node *dev)
 {
 	struct kernel_queue *kq;
 	uint32_t *buffer, i;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h
index 383202fd1ea28..9a62444308454 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h
@@ -53,7 +53,7 @@ void kq_rollback_packet(struct kernel_queue *kq);
 
 struct kernel_queue {
 	/* data */
-	struct kfd_dev		*dev;
+	struct kfd_node		*dev;
 	struct mqd_manager	*mqd_mgr;
 	struct queue		*queue;
 	uint64_t		pending_wptr64;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 54933903bcb8a..1e187677c90ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -423,7 +423,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      0, adev->kfd.dev->id, prange->prefetch_loc,
+				      0, adev->kfd.dev->node->id, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -456,7 +456,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    0, adev->kfd.dev->id, trigger);
+				    0, adev->kfd.dev->node->id, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 	svm_range_free_dma_mappings(prange);
@@ -701,7 +701,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      adev->kfd.dev->id, 0, prange->prefetch_loc,
+				      adev->kfd.dev->node->id, 0, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -737,7 +737,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    adev->kfd.dev->id, 0, trigger);
+				    adev->kfd.dev->node->id, 0, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
index 623ccd227b7de..61f6dd68c84b3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
@@ -46,7 +46,7 @@ int pipe_priority_map[] = {
 	KFD_PIPE_PRIORITY_CS_HIGH
 };
 
-struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev, struct queue_properties *q)
+struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj = NULL;
 
@@ -61,7 +61,7 @@ struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev, struct queue_propertie
 	return mqd_mem_obj;
 }
 
-struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev,
+struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev,
 					struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj = NULL;
@@ -72,7 +72,7 @@ struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev,
 		return NULL;
 
 	offset = (q->sdma_engine_id *
-		dev->device_info.num_sdma_queues_per_engine +
+		dev->kfd->device_info.num_sdma_queues_per_engine +
 		q->sdma_queue_id) *
 		dev->dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
index 57f900ccaa10d..46fc3f273d0dd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
@@ -68,7 +68,7 @@
  */
 extern int pipe_priority_map[];
 struct mqd_manager {
-	struct kfd_mem_obj*	(*allocate_mqd)(struct kfd_dev *kfd,
+	struct kfd_mem_obj*	(*allocate_mqd)(struct kfd_node *kfd,
 		struct queue_properties *q);
 
 	void	(*init_mqd)(struct mqd_manager *mm, void **mqd,
@@ -121,14 +121,14 @@ struct mqd_manager {
 	uint32_t (*read_doorbell_id)(void *mqd);
 
 	struct mutex	mqd_mutex;
-	struct kfd_dev	*dev;
+	struct kfd_node	*dev;
 	uint32_t mqd_size;
 };
 
-struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev,
+struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev,
 				struct queue_properties *q);
 
-struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev,
+struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev,
 					struct queue_properties *q);
 void free_mqd_hiq_sdma(struct mqd_manager *mm, void *mqd,
 				struct kfd_mem_obj *mqd_mem_obj);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
index 4889865c725ce..03e04d5e5a11c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
@@ -74,7 +74,7 @@ static void set_priority(struct cik_mqd *m, struct queue_properties *q)
 	m->cp_hqd_queue_priority = q->priority;
 }
 
-static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
+static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd,
 					struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj;
@@ -390,7 +390,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data)
 
 
 struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev)
+		struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
@@ -470,7 +470,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
 }
 
 struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type,
-			struct kfd_dev *dev)
+			struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index d3e2b6a599a40..7a93be0ebb19b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -74,7 +74,7 @@ static void set_priority(struct v10_compute_mqd *m, struct queue_properties *q)
 	m->cp_hqd_queue_priority = q->priority;
 }
 
-static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
+static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd,
 		struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj;
@@ -122,7 +122,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
 	}
 
-	if (mm->dev->cwsr_enabled) {
+	if (mm->dev->kfd->cwsr_enabled) {
 		m->cp_hqd_persistent_state |=
 			(1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT);
 		m->cp_hqd_ctx_save_base_addr_lo =
@@ -210,7 +210,7 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 		m->cp_hqd_pq_doorbell_control |=
 			1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT;
 	}
-	if (mm->dev->cwsr_enabled)
+	if (mm->dev->kfd->cwsr_enabled)
 		m->cp_hqd_ctx_save_control = 0;
 
 	update_cu_mask(mm, mqd, minfo);
@@ -405,7 +405,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data)
 #endif
 
 struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev)
+		struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index 5aa75f72caa1e..dff171b54b5c8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -81,7 +81,7 @@ static void set_priority(struct v11_compute_mqd *m, struct queue_properties *q)
 	m->cp_hqd_queue_priority = q->priority;
 }
 
-static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
+static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node,
 		struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj;
@@ -91,12 +91,12 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
 	 * MES write to areas beyond MQD size. So allocate
 	 * 1 PAGE_SIZE memory for MQD is MES is enabled.
 	 */
-	if (kfd->shared_resources.enable_mes)
+	if (node->kfd->shared_resources.enable_mes)
 		size = PAGE_SIZE;
 	else
 		size = sizeof(struct v11_compute_mqd);
 
-	if (kfd_gtt_sa_allocate(kfd, size, &mqd_mem_obj))
+	if (kfd_gtt_sa_allocate(node, size, &mqd_mem_obj))
 		return NULL;
 
 	return mqd_mem_obj;
@@ -113,7 +113,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	m = (struct v11_compute_mqd *) mqd_mem_obj->cpu_ptr;
 	addr = mqd_mem_obj->gpu_addr;
 
-	if (mm->dev->shared_resources.enable_mes)
+	if (mm->dev->kfd->shared_resources.enable_mes)
 		size = PAGE_SIZE;
 	else
 		size = sizeof(struct v11_compute_mqd);
@@ -155,7 +155,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
 	}
 
-	if (mm->dev->cwsr_enabled) {
+	if (mm->dev->kfd->cwsr_enabled) {
 		m->cp_hqd_persistent_state |=
 			(1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT);
 		m->cp_hqd_ctx_save_base_addr_lo =
@@ -243,7 +243,7 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 		m->cp_hqd_pq_doorbell_control |=
 			1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT;
 	}
-	if (mm->dev->cwsr_enabled)
+	if (mm->dev->kfd->cwsr_enabled)
 		m->cp_hqd_ctx_save_control = 0;
 
 	update_cu_mask(mm, mqd, minfo);
@@ -319,7 +319,7 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd,
 
 	m = (struct v11_sdma_mqd *) mqd_mem_obj->cpu_ptr;
 
-	if (mm->dev->shared_resources.enable_mes)
+	if (mm->dev->kfd->shared_resources.enable_mes)
 		size = PAGE_SIZE;
 	else
 		size = sizeof(struct v11_sdma_mqd);
@@ -387,7 +387,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data)
 #endif
 
 struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev)
+		struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
@@ -463,7 +463,7 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type,
 		 * To allocate SDMA MQDs by generic functions
 		 * when MES is enabled.
 		 */
-		if (dev->shared_resources.enable_mes) {
+		if (dev->kfd->shared_resources.enable_mes) {
 			mqd->allocate_mqd = allocate_mqd;
 			mqd->free_mqd = kfd_free_mqd_cp;
 		}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 51b53110341bb..943a738e73f95 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -83,7 +83,7 @@ static void set_priority(struct v9_mqd *m, struct queue_properties *q)
 	m->cp_hqd_queue_priority = q->priority;
 }
 
-static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
+static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node,
 		struct queue_properties *q)
 {
 	int retval;
@@ -105,11 +105,11 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
 	 * pass a special bo flag AMDGPU_GEM_CREATE_CP_MQD_GFX9 to instruct
 	 * amdgpu memory functions to do so.
 	 */
-	if (kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) {
+	if (node->kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) {
 		mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL);
 		if (!mqd_mem_obj)
 			return NULL;
-		retval = amdgpu_amdkfd_alloc_gtt_mem(kfd->adev,
+		retval = amdgpu_amdkfd_alloc_gtt_mem(node->adev,
 			ALIGN(q->ctl_stack_size, PAGE_SIZE) +
 				ALIGN(sizeof(struct v9_mqd), PAGE_SIZE),
 			&(mqd_mem_obj->gtt_mem),
@@ -121,7 +121,7 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
 			return NULL;
 		}
 	} else {
-		retval = kfd_gtt_sa_allocate(kfd, sizeof(struct v9_mqd),
+		retval = kfd_gtt_sa_allocate(node, sizeof(struct v9_mqd),
 				&mqd_mem_obj);
 		if (retval)
 			return NULL;
@@ -136,7 +136,6 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 {
 	uint64_t addr;
 	struct v9_mqd *m;
-	struct amdgpu_device *adev = (struct amdgpu_device *)mm->dev->adev;
 
 	m = (struct v9_mqd *) mqd_mem_obj->cpu_ptr;
 	addr = mqd_mem_obj->gpu_addr;
@@ -169,7 +168,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	if (q->format == KFD_QUEUE_FORMAT_AQL) {
 		m->cp_hqd_aql_control =
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
-		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		if (KFD_GC_VERSION(mm->dev) == IP_VERSION(9, 4, 3)) {
 			/* On GC 9.4.3, DW 41 is re-purposed as
 			 * compute_tg_chunk_size.
 			 * TODO: review this setting when active CUs in the
@@ -179,7 +178,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 		}
 	} else {
 		/* PM4 queue */
-		if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		if (KFD_GC_VERSION(mm->dev) == IP_VERSION(9, 4, 3)) {
 			m->compute_static_thread_mgmt_se6 = 0;
 			/* TODO: program pm4_target_xcc */
 		}
@@ -190,7 +189,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			(1 << COMPUTE_PGM_RSRC2__TRAP_PRESENT__SHIFT);
 	}
 
-	if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) {
+	if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) {
 		m->cp_hqd_persistent_state |=
 			(1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT);
 		m->cp_hqd_ctx_save_base_addr_lo =
@@ -225,7 +224,6 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 			struct queue_properties *q,
 			struct mqd_update_info *minfo)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)mm->dev->adev;
 	struct v9_mqd *m;
 
 	m = get_mqd(mqd);
@@ -275,13 +273,13 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 				2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT |
 				1 << CP_HQD_PQ_CONTROL__QUEUE_FULL_EN__SHIFT |
 				1 << CP_HQD_PQ_CONTROL__WPP_CLAMP_EN__SHIFT;
-		if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3))
+		if (KFD_GC_VERSION(mm->dev) != IP_VERSION(9, 4, 3))
 			m->cp_hqd_pq_control |=
-				 CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
+				CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
 		m->cp_hqd_pq_doorbell_control |= 1 <<
 			CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT;
 	}
-	if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address)
+	if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address)
 		m->cp_hqd_ctx_save_control = 0;
 
 	update_cu_mask(mm, mqd, minfo);
@@ -487,7 +485,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data)
 #endif
 
 struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev)
+		struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index 530ba6f5b57ea..f6b4a5686dcb2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -77,7 +77,7 @@ static void set_priority(struct vi_mqd *m, struct queue_properties *q)
 	m->cp_hqd_queue_priority = q->priority;
 }
 
-static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd,
+static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd,
 					struct queue_properties *q)
 {
 	struct kfd_mem_obj *mqd_mem_obj;
@@ -136,7 +136,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			(1 << COMPUTE_PGM_RSRC2__TRAP_PRESENT__SHIFT);
 	}
 
-	if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) {
+	if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) {
 		m->cp_hqd_persistent_state |=
 			(1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT);
 		m->cp_hqd_ctx_save_base_addr_lo =
@@ -227,7 +227,7 @@ static void __update_mqd(struct mqd_manager *mm, void *mqd,
 				2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT;
 	}
 
-	if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address)
+	if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address)
 		m->cp_hqd_ctx_save_control =
 			atc_bit << CP_HQD_CTX_SAVE_CONTROL__ATC__SHIFT |
 			mtype << CP_HQD_CTX_SAVE_CONTROL__MTYPE__SHIFT;
@@ -446,7 +446,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data)
 #endif
 
 struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev)
+		struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
@@ -528,7 +528,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
 }
 
 struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type,
-			struct kfd_dev *dev)
+			struct kfd_node *dev)
 {
 	struct mqd_manager *mqd;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
index f612325241aa4..2f54172e9175c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@@ -45,7 +45,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm,
 	unsigned int process_count, queue_count, compute_queue_count, gws_queue_count;
 	unsigned int map_queue_size;
 	unsigned int max_proc_per_quantum = 1;
-	struct kfd_dev *dev = pm->dqm->dev;
+	struct kfd_node *dev = pm->dqm->dev;
 
 	process_count = pm->dqm->processes_count;
 	queue_count = pm->dqm->active_queue_count;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 18250845a9890..54d7d4665ad22 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -119,7 +119,7 @@ static int pm_runlist_v9(struct packet_manager *pm, uint32_t *buffer,
 	struct pm4_mes_runlist *packet;
 
 	int concurrent_proc_cnt = 0;
-	struct kfd_dev *kfd = pm->dqm->dev;
+	struct kfd_node *kfd = pm->dqm->dev;
 
 	/* Determine the number of processes to map together to HW:
 	 * it can not exceed the number of VMIDs available to the
@@ -220,7 +220,8 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 	case KFD_QUEUE_TYPE_SDMA:
 	case KFD_QUEUE_TYPE_SDMA_XGMI:
 		use_static = false; /* no static queues under SDMA */
-		if (q->properties.sdma_engine_id < 2 && !pm_use_ext_eng(q->device))
+		if (q->properties.sdma_engine_id < 2 &&
+		    !pm_use_ext_eng(q->device->kfd))
 			packet->bitfields2.engine_sel = q->properties.sdma_engine_id +
 				engine_sel__mes_map_queues__sdma0_vi;
 		else {
@@ -263,7 +264,8 @@ static int pm_unmap_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 	packet->header.u32All = pm_build_pm4_header(IT_UNMAP_QUEUES,
 					sizeof(struct pm4_mes_unmap_queues));
 
-	packet->bitfields2.extended_engine_sel = pm_use_ext_eng(pm->dqm->dev) ?
+	packet->bitfields2.extended_engine_sel =
+				pm_use_ext_eng(pm->dqm->dev->kfd) ?
 		extended_engine_sel__mes_unmap_queues__sdma0_to_7_sel :
 		extended_engine_sel__mes_unmap_queues__legacy_engine_sel;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
index 4f951eaa6ee85..faf4772ed317c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
@@ -77,7 +77,7 @@ static int pm_runlist_vi(struct packet_manager *pm, uint32_t *buffer,
 {
 	struct pm4_mes_runlist *packet;
 	int concurrent_proc_cnt = 0;
-	struct kfd_dev *kfd = pm->dqm->dev;
+	struct kfd_node *kfd = pm->dqm->dev;
 
 	if (WARN_ON(!ib))
 		return -EFAULT;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 94a4389568686..fdb97e5d0c01a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -210,11 +210,13 @@ enum cache_policy {
 	((KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2)) ||	\
 	 (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)))
 
+struct kfd_node;
+
 struct kfd_event_interrupt_class {
-	bool (*interrupt_isr)(struct kfd_dev *dev,
+	bool (*interrupt_isr)(struct kfd_node *dev,
 			const uint32_t *ih_ring_entry, uint32_t *patched_ihre,
 			bool *patched_flag);
-	void (*interrupt_wq)(struct kfd_dev *dev,
+	void (*interrupt_wq)(struct kfd_node *dev,
 			const uint32_t *ih_ring_entry);
 };
 
@@ -236,8 +238,8 @@ struct kfd_device_info {
 	uint64_t reserved_sdma_queues_bitmap;
 };
 
-unsigned int kfd_get_num_sdma_engines(struct kfd_dev *kdev);
-unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev);
+unsigned int kfd_get_num_sdma_engines(struct kfd_node *kdev);
+unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *kdev);
 
 struct kfd_mem_obj {
 	uint32_t range_start;
@@ -253,13 +255,59 @@ struct kfd_vmid_info {
 	uint32_t vmid_num_kfd;
 };
 
+struct kfd_dev;
+
+struct kfd_node {
+	struct amdgpu_device *adev;     /* Duplicated here along with keeping
+					 * a copy in kfd_dev to save a hop
+					 */
+	const struct kfd2kgd_calls *kfd2kgd; /* Duplicated here along with
+					      * keeping a copy in kfd_dev to
+					      * save a hop
+					      */
+	struct kfd_vmid_info vm_info;
+	unsigned int id;                /* topology stub index */
+	/* Interrupts */
+	struct kfifo ih_fifo;
+	struct workqueue_struct *ih_wq;
+	struct work_struct interrupt_work;
+	spinlock_t interrupt_lock;
+
+	/*
+	 * Interrupts of interest to KFD are copied
+	 * from the HW ring into a SW ring.
+	 */
+	bool interrupts_active;
+
+	/* QCM Device instance */
+	struct device_queue_manager *dqm;
+
+	/* Global GWS resource shared between processes */
+	void *gws;
+	bool gws_debug_workaround;
+
+	/* Clients watching SMI events */
+	struct list_head smi_clients;
+	spinlock_t smi_lock;
+	uint32_t reset_seq_num;
+
+	/* SRAM ECC flag */
+	atomic_t sram_ecc_flag;
+
+	/*spm process id */
+	unsigned int spm_pasid;
+
+	/* Maximum process number mapped to HW scheduler */
+	unsigned int max_proc_per_quantum;
+
+	struct kfd_dev *kfd;
+};
+
 struct kfd_dev {
 	struct amdgpu_device *adev;
 
 	struct kfd_device_info device_info;
 
-	unsigned int id;		/* topology stub index */
-
 	phys_addr_t doorbell_base;	/* Start of actual doorbells used by
 					 * KFD. It is aligned for mapping
 					 * into user mode
@@ -274,7 +322,6 @@ struct kfd_dev {
 					   */
 
 	struct kgd2kfd_shared_resources shared_resources;
-	struct kfd_vmid_info vm_info;
 	struct kfd_local_mem_info local_mem_info;
 
 	const struct kfd2kgd_calls *kfd2kgd;
@@ -290,30 +337,13 @@ struct kfd_dev {
 	unsigned int gtt_sa_chunk_size;
 	unsigned int gtt_sa_num_of_chunks;
 
-	/* Interrupts */
-	struct kfifo ih_fifo;
-	struct workqueue_struct *ih_wq;
-	struct work_struct interrupt_work;
-	spinlock_t interrupt_lock;
-
-	/* QCM Device instance */
-	struct device_queue_manager *dqm;
-
 	bool init_complete;
-	/*
-	 * Interrupts of interest to KFD are copied
-	 * from the HW ring into a SW ring.
-	 */
-	bool interrupts_active;
 
 	/* Firmware versions */
 	uint16_t mec_fw_version;
 	uint16_t mec2_fw_version;
 	uint16_t sdma_fw_version;
 
-	/* Maximum process number mapped to HW scheduler */
-	unsigned int max_proc_per_quantum;
-
 	/* CWSR */
 	bool cwsr_enabled;
 	const void *cwsr_isa;
@@ -327,21 +357,9 @@ struct kfd_dev {
 	/* Use IOMMU v2 flag */
 	bool use_iommu_v2;
 
-	/* SRAM ECC flag */
-	atomic_t sram_ecc_flag;
-
 	/* Compute Profile ref. count */
 	atomic_t compute_profile;
 
-	/* Global GWS resource shared between processes */
-	void *gws;
-
-	/* Clients watching SMI events */
-	struct list_head smi_clients;
-	spinlock_t smi_lock;
-
-	uint32_t reset_seq_num;
-
 	struct ida doorbell_ida;
 	unsigned int max_doorbell_slices;
 
@@ -349,6 +367,8 @@ struct kfd_dev {
 
 	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
 	struct dev_pagemap pgmap;
+
+	struct kfd_node *node;
 };
 
 enum kfd_mempool {
@@ -563,7 +583,7 @@ struct queue {
 	unsigned int doorbell_id;
 
 	struct kfd_process	*process;
-	struct kfd_dev		*device;
+	struct kfd_node		*device;
 	void *gws;
 
 	/* procfs */
@@ -697,7 +717,7 @@ enum kfd_pdd_bound {
 /* Data that is per-process-per device. */
 struct kfd_process_device {
 	/* The device that owns this data. */
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 
 	/* The process that owns this kfd_process_device. */
 	struct kfd_process *process;
@@ -925,7 +945,7 @@ struct amdkfd_ioctl_desc {
 	unsigned int cmd_drv;
 	const char *name;
 };
-bool kfd_dev_is_large_bar(struct kfd_dev *dev);
+bool kfd_dev_is_large_bar(struct kfd_node *dev);
 
 int kfd_process_create_wq(void);
 void kfd_process_destroy_wq(void);
@@ -961,16 +981,16 @@ int kfd_process_get_user_gpu_id(struct kfd_process *p, uint32_t actual_gpu_id);
 
 int kfd_process_device_init_vm(struct kfd_process_device *pdd,
 			       struct file *drm_file);
-struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev,
+struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev,
 						struct kfd_process *p);
-struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev,
+struct kfd_process_device *kfd_get_process_device_data(struct kfd_node *dev,
 							struct kfd_process *p);
-struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
+struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 							struct kfd_process *p);
 
 bool kfd_process_xnack_mode(struct kfd_process *p, bool supported);
 
-int kfd_reserved_mem_mmap(struct kfd_dev *dev, struct kfd_process *process,
+int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process,
 			  struct vm_area_struct *vma);
 
 /* KFD process API for creating and translating handles */
@@ -994,7 +1014,7 @@ void kfd_pasid_free(u32 pasid);
 size_t kfd_doorbell_process_slice(struct kfd_dev *kfd);
 int kfd_doorbell_init(struct kfd_dev *kfd);
 void kfd_doorbell_fini(struct kfd_dev *kfd);
-int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process,
+int kfd_doorbell_mmap(struct kfd_node *dev, struct kfd_process *process,
 		      struct vm_area_struct *vma);
 void __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd,
 					unsigned int *doorbell_off);
@@ -1012,10 +1032,10 @@ void kfd_free_process_doorbells(struct kfd_dev *kfd,
 				unsigned int doorbell_index);
 /* GTT Sub-Allocator */
 
-int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size,
+int kfd_gtt_sa_allocate(struct kfd_node *node, unsigned int size,
 			struct kfd_mem_obj **mem_obj);
 
-int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj);
+int kfd_gtt_sa_free(struct kfd_node *node, struct kfd_mem_obj *mem_obj);
 
 extern struct device *kfd_device;
 
@@ -1028,25 +1048,25 @@ void kfd_procfs_del_queue(struct queue *q);
 /* Topology */
 int kfd_topology_init(void);
 void kfd_topology_shutdown(void);
-int kfd_topology_add_device(struct kfd_dev *gpu);
-int kfd_topology_remove_device(struct kfd_dev *gpu);
+int kfd_topology_add_device(struct kfd_node *gpu);
+int kfd_topology_remove_device(struct kfd_node *gpu);
 struct kfd_topology_device *kfd_topology_device_by_proximity_domain(
 						uint32_t proximity_domain);
 struct kfd_topology_device *kfd_topology_device_by_proximity_domain_no_lock(
 						uint32_t proximity_domain);
 struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id);
-struct kfd_dev *kfd_device_by_id(uint32_t gpu_id);
-struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev);
-struct kfd_dev *kfd_device_by_adev(const struct amdgpu_device *adev);
-int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev);
+struct kfd_node *kfd_device_by_id(uint32_t gpu_id);
+struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev);
+struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev);
+int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev);
 int kfd_numa_node_to_apic_id(int numa_node_id);
 void kfd_double_confirm_iommu_support(struct kfd_dev *gpu);
 
 /* Interrupts */
-int kfd_interrupt_init(struct kfd_dev *dev);
-void kfd_interrupt_exit(struct kfd_dev *dev);
-bool enqueue_ih_ring_entry(struct kfd_dev *kfd,	const void *ih_ring_entry);
-bool interrupt_is_wanted(struct kfd_dev *dev,
+int kfd_interrupt_init(struct kfd_node *dev);
+void kfd_interrupt_exit(struct kfd_node *dev);
+bool enqueue_ih_ring_entry(struct kfd_node *kfd, const void *ih_ring_entry);
+bool interrupt_is_wanted(struct kfd_node *dev,
 				const uint32_t *ih_ring_entry,
 				uint32_t *patched_ihre, bool *flag);
 
@@ -1174,22 +1194,22 @@ void print_queue_properties(struct queue_properties *q);
 void print_queue(struct queue *q);
 
 struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
+		struct kfd_node *dev);
 struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type,
-		struct kfd_dev *dev);
-struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev);
+		struct kfd_node *dev);
+struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev);
 void device_queue_manager_uninit(struct device_queue_manager *dqm);
-struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
+struct kernel_queue *kernel_queue_init(struct kfd_node *dev,
 					enum kfd_queue_type type);
 void kernel_queue_uninit(struct kernel_queue *kq, bool hanging);
 int kfd_dqm_evict_pasid(struct device_queue_manager *dqm, u32 pasid);
@@ -1206,7 +1226,7 @@ void kfd_process_dequeue_from_all_devices(struct kfd_process *p);
 int pqm_init(struct process_queue_manager *pqm, struct kfd_process *p);
 void pqm_uninit(struct process_queue_manager *pqm);
 int pqm_create_queue(struct process_queue_manager *pqm,
-			    struct kfd_dev *dev,
+			    struct kfd_node *dev,
 			    struct file *f,
 			    struct queue_properties *properties,
 			    unsigned int *qid,
@@ -1323,7 +1343,7 @@ int kfd_wait_on_events(struct kfd_process *p,
 		       uint32_t *wait_result);
 void kfd_signal_event_interrupt(u32 pasid, uint32_t partial_id,
 				uint32_t valid_id_bits);
-void kfd_signal_iommu_event(struct kfd_dev *dev,
+void kfd_signal_iommu_event(struct kfd_node *dev,
 			    u32 pasid, unsigned long address,
 			    bool is_write_requested, bool is_execute_requested);
 void kfd_signal_hw_exception_event(u32 pasid);
@@ -1339,12 +1359,12 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p,
 int kfd_get_num_events(struct kfd_process *p);
 int kfd_event_destroy(struct kfd_process *p, uint32_t event_id);
 
-void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid,
+void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
 				struct kfd_vm_fault_info *info);
 
-void kfd_signal_reset_event(struct kfd_dev *dev);
+void kfd_signal_reset_event(struct kfd_node *dev);
 
-void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid);
+void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid);
 
 void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type);
 
@@ -1359,12 +1379,12 @@ static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev)
 bool kfd_is_locked(void);
 
 /* Compute profile */
-void kfd_inc_compute_active(struct kfd_dev *dev);
-void kfd_dec_compute_active(struct kfd_dev *dev);
+void kfd_inc_compute_active(struct kfd_node *dev);
+void kfd_dec_compute_active(struct kfd_node *dev);
 
 /* Cgroup Support */
 /* Check with device cgroup if @kfd device is accessible */
-static inline int kfd_devcgroup_check_permission(struct kfd_dev *kfd)
+static inline int kfd_devcgroup_check_permission(struct kfd_node *kfd)
 {
 #if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF)
 	struct drm_device *ddev = adev_to_drm(kfd->adev);
@@ -1389,7 +1409,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data);
 int kfd_debugfs_rls_by_device(struct seq_file *m, void *data);
 int pm_debugfs_runlist(struct seq_file *m, void *data);
 
-int kfd_debugfs_hang_hws(struct kfd_dev *dev);
+int kfd_debugfs_hang_hws(struct kfd_node *dev);
 int pm_debugfs_hang_hws(struct packet_manager *pm);
 int dqm_debugfs_hang_hws(struct device_queue_manager *dqm);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 07a9eaf9b7d8e..66e021889c64b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -269,7 +269,7 @@ static int kfd_get_cu_occupancy(struct attribute *attr, char *buffer)
 	int cu_cnt;
 	int wave_cnt;
 	int max_waves_per_cu;
-	struct kfd_dev *dev = NULL;
+	struct kfd_node *dev = NULL;
 	struct kfd_process *proc = NULL;
 	struct kfd_process_device *pdd = NULL;
 
@@ -691,7 +691,7 @@ void kfd_process_destroy_wq(void)
 static void kfd_process_free_gpuvm(struct kgd_mem *mem,
 			struct kfd_process_device *pdd, void **kptr)
 {
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 
 	if (kptr && *kptr) {
 		amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(mem);
@@ -713,7 +713,7 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd,
 				   uint64_t gpu_va, uint32_t size,
 				   uint32_t flags, struct kgd_mem **mem, void **kptr)
 {
-	struct kfd_dev *kdev = pdd->dev;
+	struct kfd_node *kdev = pdd->dev;
 	int err;
 
 	err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(kdev->adev, gpu_va, size,
@@ -982,7 +982,7 @@ static void kfd_process_device_free_bos(struct kfd_process_device *pdd)
 static void kfd_process_kunmap_signal_bo(struct kfd_process *p)
 {
 	struct kfd_process_device *pdd;
-	struct kfd_dev *kdev;
+	struct kfd_node *kdev;
 	void *mem;
 
 	kdev = kfd_device_by_id(GET_GPU_ID(p->signal_handle));
@@ -1040,9 +1040,9 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 		bitmap_free(pdd->qpd.doorbell_bitmap);
 		idr_destroy(&pdd->alloc_idr);
 
-		kfd_free_process_doorbells(pdd->dev, pdd->doorbell_index);
+		kfd_free_process_doorbells(pdd->dev->kfd, pdd->doorbell_index);
 
-		if (pdd->dev->shared_resources.enable_mes)
+		if (pdd->dev->kfd->shared_resources.enable_mes)
 			amdgpu_amdkfd_free_gtt_mem(pdd->dev->adev,
 						   pdd->proc_ctx_bo);
 		/*
@@ -1259,10 +1259,10 @@ static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 	int i;
 
 	for (i = 0; i < p->n_pdds; i++) {
-		struct kfd_dev *dev = p->pdds[i]->dev;
+		struct kfd_node *dev = p->pdds[i]->dev;
 		struct qcm_process_device *qpd = &p->pdds[i]->qpd;
 
-		if (!dev->cwsr_enabled || qpd->cwsr_kaddr || qpd->cwsr_base)
+		if (!dev->kfd->cwsr_enabled || qpd->cwsr_kaddr || qpd->cwsr_base)
 			continue;
 
 		offset = KFD_MMAP_TYPE_RESERVED_MEM | KFD_MMAP_GPU_ID(dev->id);
@@ -1279,7 +1279,7 @@ static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 			return err;
 		}
 
-		memcpy(qpd->cwsr_kaddr, dev->cwsr_isa, dev->cwsr_isa_size);
+		memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size);
 
 		qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET;
 		pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n",
@@ -1291,7 +1291,7 @@ static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 
 static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd)
 {
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 	struct qcm_process_device *qpd = &pdd->qpd;
 	uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT
 			| KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE
@@ -1300,7 +1300,7 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd)
 	void *kaddr;
 	int ret;
 
-	if (!dev->cwsr_enabled || qpd->cwsr_kaddr || !qpd->cwsr_base)
+	if (!dev->kfd->cwsr_enabled || qpd->cwsr_kaddr || !qpd->cwsr_base)
 		return 0;
 
 	/* cwsr_base is only set for dGPU */
@@ -1313,7 +1313,7 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd)
 	qpd->cwsr_kaddr = kaddr;
 	qpd->tba_addr = qpd->cwsr_base;
 
-	memcpy(qpd->cwsr_kaddr, dev->cwsr_isa, dev->cwsr_isa_size);
+	memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size);
 
 	qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET;
 	pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n",
@@ -1324,10 +1324,10 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd)
 
 static void kfd_process_device_destroy_cwsr_dgpu(struct kfd_process_device *pdd)
 {
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 	struct qcm_process_device *qpd = &pdd->qpd;
 
-	if (!dev->cwsr_enabled || !qpd->cwsr_kaddr || !qpd->cwsr_base)
+	if (!dev->kfd->cwsr_enabled || !qpd->cwsr_kaddr || !qpd->cwsr_base)
 		return;
 
 	kfd_process_free_gpuvm(qpd->cwsr_mem, pdd, &qpd->cwsr_kaddr);
@@ -1371,7 +1371,7 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported)
 	 * support retry.
 	 */
 	for (i = 0; i < p->n_pdds; i++) {
-		struct kfd_dev *dev = p->pdds[i]->dev;
+		struct kfd_node *dev = p->pdds[i]->dev;
 
 		/* Only consider GFXv9 and higher GPUs. Older GPUs don't
 		 * support the SVM APIs and don't need to be considered
@@ -1394,7 +1394,7 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported)
 		if (KFD_GC_VERSION(dev) >= IP_VERSION(10, 1, 1))
 			return false;
 
-		if (dev->noretry)
+		if (dev->kfd->noretry)
 			return false;
 	}
 
@@ -1528,7 +1528,7 @@ static int init_doorbell_bitmap(struct qcm_process_device *qpd,
 	return 0;
 }
 
-struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev,
+struct kfd_process_device *kfd_get_process_device_data(struct kfd_node *dev,
 							struct kfd_process *p)
 {
 	int i;
@@ -1540,7 +1540,7 @@ struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev,
 	return NULL;
 }
 
-struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
+struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 							struct kfd_process *p)
 {
 	struct kfd_process_device *pdd = NULL;
@@ -1552,7 +1552,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 	if (!pdd)
 		return NULL;
 
-	if (init_doorbell_bitmap(&pdd->qpd, dev)) {
+	if (init_doorbell_bitmap(&pdd->qpd, dev->kfd)) {
 		pr_err("Failed to init doorbell for process\n");
 		goto err_free_pdd;
 	}
@@ -1573,7 +1573,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev,
 	pdd->user_gpu_id = dev->id;
 	atomic64_set(&pdd->evict_duration_counter, 0);
 
-	if (dev->shared_resources.enable_mes) {
+	if (dev->kfd->shared_resources.enable_mes) {
 		retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev,
 						AMDGPU_MES_PROC_CTX_SIZE,
 						&pdd->proc_ctx_bo,
@@ -1619,7 +1619,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd,
 	struct amdgpu_fpriv *drv_priv;
 	struct amdgpu_vm *avm;
 	struct kfd_process *p;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	int ret;
 
 	if (!drm_file)
@@ -1679,7 +1679,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd,
  *
  * Assumes that the process lock is held.
  */
-struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev,
+struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev,
 							struct kfd_process *p)
 {
 	struct kfd_process_device *pdd;
@@ -1811,7 +1811,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger)
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_process_device *pdd = p->pdds[i];
 
-		kfd_smi_event_queue_eviction(pdd->dev, p->lead_thread->pid,
+		kfd_smi_event_queue_eviction(pdd->dev->kfd, p->lead_thread->pid,
 					     trigger);
 
 		r = pdd->dev->dqm->ops.evict_process_queues(pdd->dev->dqm,
@@ -1839,7 +1839,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger)
 		if (n_evicted == 0)
 			break;
 
-		kfd_smi_event_queue_restore(pdd->dev, p->lead_thread->pid);
+		kfd_smi_event_queue_restore(pdd->dev->kfd, p->lead_thread->pid);
 
 		if (pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm,
 							      &pdd->qpd))
@@ -1860,7 +1860,7 @@ int kfd_process_restore_queues(struct kfd_process *p)
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_process_device *pdd = p->pdds[i];
 
-		kfd_smi_event_queue_restore(pdd->dev, p->lead_thread->pid);
+		kfd_smi_event_queue_restore(pdd->dev->kfd, p->lead_thread->pid);
 
 		r = pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm,
 							      &pdd->qpd);
@@ -2016,7 +2016,7 @@ int kfd_resume_all_processes(void)
 	return ret;
 }
 
-int kfd_reserved_mem_mmap(struct kfd_dev *dev, struct kfd_process *process,
+int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process,
 			  struct vm_area_struct *vma)
 {
 	struct kfd_process_device *pdd;
@@ -2051,7 +2051,7 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 {
 	struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv);
 	uint64_t tlb_seq = amdgpu_vm_tlb_seq(vm);
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 
 	/*
 	 * It can be that we race and lose here, but that is extremely unlikely
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 4236539d9f932..5602498e713fa 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -81,7 +81,7 @@ static int find_available_queue_slot(struct process_queue_manager *pqm,
 
 void kfd_process_dequeue_from_device(struct kfd_process_device *pdd)
 {
-	struct kfd_dev *dev = pdd->dev;
+	struct kfd_node *dev = pdd->dev;
 
 	if (pdd->already_dequeued)
 		return;
@@ -93,7 +93,7 @@ void kfd_process_dequeue_from_device(struct kfd_process_device *pdd)
 int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid,
 			void *gws)
 {
-	struct kfd_dev *dev = NULL;
+	struct kfd_node *dev = NULL;
 	struct process_queue_node *pqn;
 	struct kfd_process_device *pdd;
 	struct kgd_mem *mem = NULL;
@@ -178,7 +178,7 @@ void pqm_uninit(struct process_queue_manager *pqm)
 }
 
 static int init_user_queue(struct process_queue_manager *pqm,
-				struct kfd_dev *dev, struct queue **q,
+				struct kfd_node *dev, struct queue **q,
 				struct queue_properties *q_properties,
 				struct file *f, struct amdgpu_bo *wptr_bo,
 				unsigned int qid)
@@ -199,7 +199,7 @@ static int init_user_queue(struct process_queue_manager *pqm,
 	(*q)->device = dev;
 	(*q)->process = pqm->process;
 
-	if (dev->shared_resources.enable_mes) {
+	if (dev->kfd->shared_resources.enable_mes) {
 		retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev,
 						AMDGPU_MES_GANG_CTX_SIZE,
 						&(*q)->gang_ctx_bo,
@@ -224,7 +224,7 @@ static int init_user_queue(struct process_queue_manager *pqm,
 }
 
 int pqm_create_queue(struct process_queue_manager *pqm,
-			    struct kfd_dev *dev,
+			    struct kfd_node *dev,
 			    struct file *f,
 			    struct queue_properties *properties,
 			    unsigned int *qid,
@@ -258,7 +258,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 	 * Hence we also check the type as well
 	 */
 	if ((pdd->qpd.is_debug) || (type == KFD_QUEUE_TYPE_DIQ))
-		max_queues = dev->device_info.max_no_of_hqd/2;
+		max_queues = dev->kfd->device_info.max_no_of_hqd/2;
 
 	if (pdd->qpd.queue_count >= max_queues)
 		return -ENOSPC;
@@ -354,7 +354,7 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		 */
 		*p_doorbell_offset_in_process =
 			(q->properties.doorbell_off * sizeof(uint32_t)) &
-			(kfd_doorbell_process_slice(dev) - 1);
+			(kfd_doorbell_process_slice(dev->kfd) - 1);
 
 	pr_debug("PQM After DQM create queue\n");
 
@@ -387,7 +387,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid)
 	struct process_queue_node *pqn;
 	struct kfd_process_device *pdd;
 	struct device_queue_manager *dqm;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	int retval;
 
 	dqm = NULL;
@@ -439,7 +439,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid)
 			pdd->qpd.num_gws = 0;
 		}
 
-		if (dev->shared_resources.enable_mes) {
+		if (dev->kfd->shared_resources.enable_mes) {
 			amdgpu_amdkfd_free_gtt_mem(dev->adev,
 						   pqn->q->gang_ctx_bo);
 			if (pqn->q->wptr_bo)
@@ -859,7 +859,7 @@ int kfd_criu_restore_queue(struct kfd_process *p,
 	}
 
 	if (!pdd->doorbell_index &&
-	    kfd_alloc_process_doorbells(pdd->dev, &pdd->doorbell_index) < 0) {
+	    kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) {
 		ret = -ENOMEM;
 		goto exit;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index 0472b56de2457..a0bf6558f4acc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -36,7 +36,7 @@ struct kfd_smi_client {
 	wait_queue_head_t wait_queue;
 	/* events enabled */
 	uint64_t events;
-	struct kfd_dev *dev;
+	struct kfd_node *dev;
 	spinlock_t lock;
 	struct rcu_head rcu;
 	pid_t pid;
@@ -149,7 +149,7 @@ static void kfd_smi_ev_client_free(struct rcu_head *p)
 static int kfd_smi_ev_release(struct inode *inode, struct file *filep)
 {
 	struct kfd_smi_client *client = filep->private_data;
-	struct kfd_dev *dev = client->dev;
+	struct kfd_node *dev = client->dev;
 
 	spin_lock(&dev->smi_lock);
 	list_del_rcu(&client->list);
@@ -171,7 +171,7 @@ static bool kfd_smi_ev_enabled(pid_t pid, struct kfd_smi_client *client,
 	return events & KFD_SMI_EVENT_MASK_FROM_INDEX(event);
 }
 
-static void add_event_to_kfifo(pid_t pid, struct kfd_dev *dev,
+static void add_event_to_kfifo(pid_t pid, struct kfd_node *dev,
 			       unsigned int smi_event, char *event_msg, int len)
 {
 	struct kfd_smi_client *client;
@@ -196,7 +196,7 @@ static void add_event_to_kfifo(pid_t pid, struct kfd_dev *dev,
 }
 
 __printf(4, 5)
-static void kfd_smi_event_add(pid_t pid, struct kfd_dev *dev,
+static void kfd_smi_event_add(pid_t pid, struct kfd_node *dev,
 			      unsigned int event, char *fmt, ...)
 {
 	char fifo_in[KFD_SMI_EVENT_MSG_SIZE];
@@ -215,7 +215,7 @@ static void kfd_smi_event_add(pid_t pid, struct kfd_dev *dev,
 	add_event_to_kfifo(pid, dev, event, fifo_in, len);
 }
 
-void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset)
+void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset)
 {
 	unsigned int event;
 
@@ -228,7 +228,7 @@ void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset)
 	kfd_smi_event_add(0, dev, event, "%x\n", dev->reset_seq_num);
 }
 
-void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev,
+void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev,
 					     uint64_t throttle_bitmask)
 {
 	kfd_smi_event_add(0, dev, KFD_SMI_EVENT_THERMAL_THROTTLE, "%llx:%llx\n",
@@ -236,7 +236,7 @@ void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev,
 			  amdgpu_dpm_get_thermal_throttling_counter(dev->adev));
 }
 
-void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid)
+void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid)
 {
 	struct amdgpu_task_info task_info;
 
@@ -254,17 +254,17 @@ void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid,
 				    unsigned long address, bool write_fault,
 				    ktime_t ts)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_PAGE_FAULT_START,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_PAGE_FAULT_START,
 			  "%lld -%d @%lx(%x) %c\n", ktime_to_ns(ts), pid,
-			  address, dev->id, write_fault ? 'W' : 'R');
+			  address, dev->node->id, write_fault ? 'W' : 'R');
 }
 
 void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid,
 				  unsigned long address, bool migration)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_PAGE_FAULT_END,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_PAGE_FAULT_END,
 			  "%lld -%d @%lx(%x) %c\n", ktime_get_boottime_ns(),
-			  pid, address, dev->id, migration ? 'M' : 'U');
+			  pid, address, dev->node->id, migration ? 'M' : 'U');
 }
 
 void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
@@ -273,7 +273,7 @@ void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
 				   uint32_t prefetch_loc, uint32_t preferred_loc,
 				   uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_MIGRATE_START,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_MIGRATE_START,
 			  "%lld -%d @%lx(%lx) %x->%x %x:%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, prefetch_loc, preferred_loc, trigger);
@@ -283,7 +283,7 @@ void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
 				 unsigned long start, unsigned long end,
 				 uint32_t from, uint32_t to, uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_MIGRATE_END,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_MIGRATE_END,
 			  "%lld -%d @%lx(%lx) %x->%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, trigger);
@@ -292,16 +292,16 @@ void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
 void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_QUEUE_EVICTION,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_QUEUE_EVICTION,
 			  "%lld -%d %x %d\n", ktime_get_boottime_ns(), pid,
-			  dev->id, trigger);
+			  dev->node->id, trigger);
 }
 
 void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_QUEUE_RESTORE,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_QUEUE_RESTORE,
 			  "%lld -%d %x\n", ktime_get_boottime_ns(), pid,
-			  dev->id);
+			  dev->node->id);
 }
 
 void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
@@ -328,12 +328,12 @@ void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid,
 				  unsigned long address, unsigned long last,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_UNMAP_FROM_GPU,
+	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_UNMAP_FROM_GPU,
 			  "%lld -%d @%lx(%lx) %x %d\n", ktime_get_boottime_ns(),
-			  pid, address, last - address + 1, dev->id, trigger);
+			  pid, address, last - address + 1, dev->node->id, trigger);
 }
 
-int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd)
+int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd)
 {
 	struct kfd_smi_client *client;
 	int ret;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
index 76fe4e0ec2d23..59cd089f80d18 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
@@ -24,11 +24,11 @@
 #ifndef KFD_SMI_EVENTS_H_INCLUDED
 #define KFD_SMI_EVENTS_H_INCLUDED
 
-int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd);
-void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid);
-void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev,
+int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd);
+void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid);
+void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev,
 					     uint64_t throttle_bitmask);
-void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset);
+void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset);
 void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid,
 				    unsigned long address, bool write_fault,
 				    ktime_t ts);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index c02430537e9c8..96ccff79902ce 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1266,7 +1266,7 @@ svm_range_unmap_from_gpus(struct svm_range *prange, unsigned long start,
 			return -EINVAL;
 		}
 
-		kfd_smi_event_unmap_from_gpu(pdd->dev, p->lead_thread->pid,
+		kfd_smi_event_unmap_from_gpu(pdd->dev->kfd, p->lead_thread->pid,
 					     start, last, trigger);
 
 		r = svm_range_unmap_from_gpu(pdd->dev->adev,
@@ -3083,7 +3083,7 @@ int svm_range_list_init(struct kfd_process *p)
 	spin_lock_init(&svms->deferred_list_lock);
 
 	for (i = 0; i < p->n_pdds; i++)
-		if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev))
+		if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev->kfd))
 			bitmap_set(svms->bitmap_supported, i, 1);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 8e4124dcb6e4c..06a11186d947c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -96,7 +96,7 @@ struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id)
 	return ret;
 }
 
-struct kfd_dev *kfd_device_by_id(uint32_t gpu_id)
+struct kfd_node *kfd_device_by_id(uint32_t gpu_id)
 {
 	struct kfd_topology_device *top_dev;
 
@@ -107,10 +107,10 @@ struct kfd_dev *kfd_device_by_id(uint32_t gpu_id)
 	return top_dev->gpu;
 }
 
-struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev)
+struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev)
 {
 	struct kfd_topology_device *top_dev;
-	struct kfd_dev *device = NULL;
+	struct kfd_node *device = NULL;
 
 	down_read(&topology_lock);
 
@@ -125,10 +125,10 @@ struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev)
 	return device;
 }
 
-struct kfd_dev *kfd_device_by_adev(const struct amdgpu_device *adev)
+struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev)
 {
 	struct kfd_topology_device *top_dev;
-	struct kfd_dev *device = NULL;
+	struct kfd_node *device = NULL;
 
 	down_read(&topology_lock);
 
@@ -526,7 +526,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 
 	if (dev->gpu) {
 		log_max_watch_addr =
-			__ilog2_u32(dev->gpu->device_info.num_of_watch_points);
+			__ilog2_u32(dev->gpu->kfd->device_info.num_of_watch_points);
 
 		if (log_max_watch_addr) {
 			dev->node_props.capability |=
@@ -548,11 +548,11 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 		sysfs_show_64bit_prop(buffer, offs, "local_mem_size", 0ULL);
 
 		sysfs_show_32bit_prop(buffer, offs, "fw_version",
-				      dev->gpu->mec_fw_version);
+				      dev->gpu->kfd->mec_fw_version);
 		sysfs_show_32bit_prop(buffer, offs, "capability",
 				      dev->node_props.capability);
 		sysfs_show_32bit_prop(buffer, offs, "sdma_fw_version",
-				      dev->gpu->sdma_fw_version);
+				      dev->gpu->kfd->sdma_fw_version);
 		sysfs_show_64bit_prop(buffer, offs, "unique_id",
 				      dev->gpu->adev->unique_id);
 
@@ -1157,7 +1157,7 @@ void kfd_topology_shutdown(void)
 	up_write(&topology_lock);
 }
 
-static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
+static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 {
 	uint32_t hashout;
 	uint32_t buf[7];
@@ -1167,8 +1167,8 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
 	if (!gpu)
 		return 0;
 
-	local_mem_size = gpu->local_mem_info.local_mem_size_private +
-			gpu->local_mem_info.local_mem_size_public;
+	local_mem_size = gpu->kfd->local_mem_info.local_mem_size_private +
+			gpu->kfd->local_mem_info.local_mem_size_public;
 	buf[0] = gpu->adev->pdev->devfn;
 	buf[1] = gpu->adev->pdev->subsystem_vendor |
 		(gpu->adev->pdev->subsystem_device << 16);
@@ -1188,7 +1188,7 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
  *		list then return NULL. This means a new topology device has to
  *		be created for this GPU.
  */
-static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
+static struct kfd_topology_device *kfd_assign_gpu(struct kfd_node *gpu)
 {
 	struct kfd_topology_device *dev;
 	struct kfd_topology_device *out_dev = NULL;
@@ -1201,7 +1201,7 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu)
 		/* Discrete GPUs need their own topology device list
 		 * entries. Don't assign them to CPU/APU nodes.
 		 */
-		if (!gpu->use_iommu_v2 &&
+		if (!gpu->kfd->use_iommu_v2 &&
 		    dev->node_props.cpu_cores_count)
 			continue;
 
@@ -1275,7 +1275,7 @@ static void kfd_set_iolink_no_atomics(struct kfd_topology_device *dev,
 				CRAT_IOLINK_FLAGS_NO_ATOMICS_64_BIT;
 	/* set gpu (dev) flags. */
 	} else {
-		if (!dev->gpu->pci_atomic_requested ||
+		if (!dev->gpu->kfd->pci_atomic_requested ||
 				dev->gpu->adev->asic_type == CHIP_HAWAII)
 			link->flags |= CRAT_IOLINK_FLAGS_NO_ATOMICS_32_BIT |
 				CRAT_IOLINK_FLAGS_NO_ATOMICS_64_BIT;
@@ -1569,8 +1569,8 @@ static int kfd_dev_create_p2p_links(void)
 		if (dev == new_dev)
 			break;
 		if (!dev->gpu || !dev->gpu->adev ||
-		    (dev->gpu->hive_id &&
-		     dev->gpu->hive_id == new_dev->gpu->hive_id))
+		    (dev->gpu->kfd->hive_id &&
+		     dev->gpu->kfd->hive_id == new_dev->gpu->kfd->hive_id))
 			goto next;
 
 		/* check if node(s) is/are peer accessible in one direction or bi-direction */
@@ -1590,7 +1590,6 @@ static int kfd_dev_create_p2p_links(void)
 	return ret;
 }
 
-
 /* Helper function. See kfd_fill_gpu_cache_info for parameter description */
 static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext,
 				struct kfd_gpu_cache_info *pcache_info,
@@ -1723,7 +1722,7 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext,
 /* kfd_fill_cache_non_crat_info - Fill GPU cache info using kfd_gpu_cache_info
  * tables
  */
-static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct kfd_dev *kdev)
+static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct kfd_node *kdev)
 {
 	struct kfd_gpu_cache_info *pcache_info = NULL;
 	int i, j, k;
@@ -1805,7 +1804,7 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct
 	pr_debug("Added [%d] GPU cache entries\n", num_of_entries);
 }
 
-static int kfd_topology_add_device_locked(struct kfd_dev *gpu, uint32_t gpu_id,
+static int kfd_topology_add_device_locked(struct kfd_node *gpu, uint32_t gpu_id,
 					  struct kfd_topology_device **dev)
 {
 	int proximity_domain = ++topology_crat_proximity_domain;
@@ -1865,7 +1864,7 @@ static int kfd_topology_add_device_locked(struct kfd_dev *gpu, uint32_t gpu_id,
 	return res;
 }
 
-int kfd_topology_add_device(struct kfd_dev *gpu)
+int kfd_topology_add_device(struct kfd_node *gpu)
 {
 	uint32_t gpu_id;
 	struct kfd_topology_device *dev;
@@ -1916,7 +1915,8 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	dev->node_props.simd_arrays_per_engine =
 		cu_info.num_shader_arrays_per_engine;
 
-	dev->node_props.gfx_target_version = gpu->device_info.gfx_target_version;
+	dev->node_props.gfx_target_version =
+				gpu->kfd->device_info.gfx_target_version;
 	dev->node_props.vendor_id = gpu->adev->pdev->vendor;
 	dev->node_props.device_id = gpu->adev->pdev->device;
 	dev->node_props.capability |=
@@ -1929,15 +1929,15 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	dev->node_props.max_engine_clk_ccompute =
 		cpufreq_quick_get_max(0) / 1000;
 	dev->node_props.drm_render_minor =
-		gpu->shared_resources.drm_render_minor;
+		gpu->kfd->shared_resources.drm_render_minor;
 
-	dev->node_props.hive_id = gpu->hive_id;
+	dev->node_props.hive_id = gpu->kfd->hive_id;
 	dev->node_props.num_sdma_engines = kfd_get_num_sdma_engines(gpu);
 	dev->node_props.num_sdma_xgmi_engines =
 					kfd_get_num_xgmi_sdma_engines(gpu);
 	dev->node_props.num_sdma_queues_per_engine =
-				gpu->device_info.num_sdma_queues_per_engine -
-				gpu->device_info.num_reserved_sdma_queues_per_engine;
+				gpu->kfd->device_info.num_sdma_queues_per_engine -
+				gpu->kfd->device_info.num_reserved_sdma_queues_per_engine;
 	dev->node_props.num_gws = (dev->gpu->gws &&
 		dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ?
 		dev->gpu->adev->gds.gws_size : 0;
@@ -1979,7 +1979,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
 	 * Overwrite ATS capability according to needs_iommu_device to fix
 	 * potential missing corresponding bit in CRAT of BIOS.
 	 */
-	if (dev->gpu->use_iommu_v2)
+	if (dev->gpu->kfd->use_iommu_v2)
 		dev->node_props.capability |= HSA_CAP_ATS_PRESENT;
 	else
 		dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT;
@@ -2079,7 +2079,7 @@ static void kfd_topology_update_io_links(int proximity_domain)
 	}
 }
 
-int kfd_topology_remove_device(struct kfd_dev *gpu)
+int kfd_topology_remove_device(struct kfd_node *gpu)
 {
 	struct kfd_topology_device *dev, *tmp;
 	uint32_t gpu_id;
@@ -2119,7 +2119,7 @@ int kfd_topology_remove_device(struct kfd_dev *gpu)
  * Return -	0: On success (@kdev will be NULL for non GPU nodes)
  *		-1: If end of list
  */
-int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev)
+int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev)
 {
 
 	struct kfd_topology_device *top_dev;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
index fca30d00a9bb3..3b8afb6aba79e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
@@ -75,7 +75,7 @@ struct kfd_mem_properties {
 	uint32_t		flags;
 	uint32_t		width;
 	uint32_t		mem_clk_max;
-	struct kfd_dev		*gpu;
+	struct kfd_node		*gpu;
 	struct kobject		*kobj;
 	struct attribute	attr;
 };
@@ -93,7 +93,7 @@ struct kfd_cache_properties {
 	uint32_t		cache_latency;
 	uint32_t		cache_type;
 	uint8_t			sibling_map[CACHE_SIBLINGMAP_SIZE];
-	struct kfd_dev		*gpu;
+	struct kfd_node		*gpu;
 	struct kobject		*kobj;
 	struct attribute	attr;
 	uint32_t		sibling_map_size;
@@ -113,7 +113,7 @@ struct kfd_iolink_properties {
 	uint32_t		max_bandwidth;
 	uint32_t		rec_transfer_size;
 	uint32_t		flags;
-	struct kfd_dev		*gpu;
+	struct kfd_node		*gpu;
 	struct kobject		*kobj;
 	struct attribute	attr;
 };
@@ -135,7 +135,7 @@ struct kfd_topology_device {
 	struct list_head		io_link_props;
 	struct list_head		p2p_link_props;
 	struct list_head		perf_props;
-	struct kfd_dev			*gpu;
+	struct kfd_node			*gpu;
 	struct kobject			*kobj_node;
 	struct kobject			*kobj_mem;
 	struct kobject			*kobj_cache;
-- 
GitLab


From 74c5b85da75475c73a8f040397610fbfcc2c3e78 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 16:33:38 -0400
Subject: [PATCH 1046/1791] drm/amdkfd: Add spatial partitioning support in KFD

This patch introduces multi-partition support in KFD.
This patch includes:
- Support for maximum 8 spatial partitions in KFD.
- Initialize one HIQ per partition.
- Management of VMID range depending on partition mode.
- Management of doorbell aperture space between all
  partitions.
- Each partition does its own queue management, interrupt
  handling, SMI event reporting.
- IOMMU, if enabled with multiple partitions, will only work
  on first partition.
- SPM is only supported on the first partition.
- Currently, there is no support for resetting individual
  partitions. All partitions will reset together.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       | 214 +++++++++++++-----
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |   2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_iommu.c        |  13 +-
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c      |   8 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  16 +-
 drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c   |  24 +-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c     |   8 +-
 7 files changed, 208 insertions(+), 77 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 23d9a7f770551..37c6dc5c37bfb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -567,23 +567,27 @@ static int kfd_init_node(struct kfd_node *node)
 	return err;
 }
 
-static void kfd_cleanup_node(struct kfd_dev *kfd)
+static void kfd_cleanup_nodes(struct kfd_dev *kfd, unsigned int num_nodes)
 {
-	struct kfd_node *knode = kfd->node;
-
-	device_queue_manager_uninit(knode->dqm);
-	kfd_interrupt_exit(knode);
-	kfd_topology_remove_device(knode);
-	if (knode->gws)
-		amdgpu_amdkfd_free_gws(knode->adev, knode->gws);
-	kfree(knode);
-	kfd->node = NULL;
+	struct kfd_node *knode;
+	unsigned int i;
+
+	for (i = 0; i < num_nodes; i++) {
+		knode = kfd->nodes[i];
+		device_queue_manager_uninit(knode->dqm);
+		kfd_interrupt_exit(knode);
+		kfd_topology_remove_device(knode);
+		if (knode->gws)
+			amdgpu_amdkfd_free_gws(knode->adev, knode->gws);
+		kfree(knode);
+		kfd->nodes[i] = NULL;
+	}
 }
 
 bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			 const struct kgd2kfd_shared_resources *gpu_resources)
 {
-	unsigned int size, map_process_packet_size;
+	unsigned int size, map_process_packet_size, i;
 	struct kfd_node *node;
 	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
 	unsigned int max_proc_per_quantum;
@@ -596,9 +600,18 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			KGD_ENGINE_SDMA1);
 	kfd->shared_resources = *gpu_resources;
 
-	first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1;
-	last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1;
-	vmid_num_kfd = last_vmid_kfd - first_vmid_kfd + 1;
+	if (kfd->adev->gfx.num_xcd == 0 || kfd->adev->gfx.num_xcd == 1 ||
+	    kfd->adev->gfx.num_xcc_per_xcp == 0)
+		kfd->num_nodes = 1;
+	else
+		kfd->num_nodes =
+			kfd->adev->gfx.num_xcd/kfd->adev->gfx.num_xcc_per_xcp;
+	if (kfd->num_nodes == 0) {
+		dev_err(kfd_device,
+			"KFD num nodes cannot be 0, GC inst: %d, num_xcc_in_node: %d\n",
+			kfd->adev->gfx.num_xcd, kfd->adev->gfx.num_xcc_per_xcp);
+		goto out;
+	}
 
 	/* Allow BIF to recode atomics to PCIe 3.0 AtomicOps.
 	 * 32 and 64-bit requests are possible and must be
@@ -617,6 +630,26 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		return false;
 	}
 
+	first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1;
+	last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1;
+	vmid_num_kfd = last_vmid_kfd - first_vmid_kfd + 1;
+
+	/* For GFX9.4.3, we need special handling for VMIDs depending on
+	 * partition mode.
+	 * In CPX mode, the VMID range needs to be shared between XCDs.
+	 * Additionally, there are 13 VMIDs (3-15) available for KFD. To
+	 * divide them equally, we change starting VMID to 4 and not use
+	 * VMID 3.
+	 * If the VMID range changes for GFX9.4.3, then this code MUST be
+	 * revisited.
+	 */
+	if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
+	    kfd->adev->gfx.partition_mode == AMDGPU_CPX_PARTITION_MODE &&
+	    kfd->num_nodes != 1) {
+		vmid_num_kfd /= 2;
+		first_vmid_kfd = last_vmid_kfd + 1 - vmid_num_kfd*2;
+	}
+
 	/* Verify module parameters regarding mapped process number*/
 	if (hws_max_conc_proc >= 0)
 		max_proc_per_quantum = min((u32)hws_max_conc_proc, vmid_num_kfd);
@@ -682,6 +715,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd_cwsr_init(kfd);
 
+	/* TODO: Needs to be updated for memory partitioning */
 	svm_migrate_init(kfd->adev);
 
 	/* Allocate the KFD node */
@@ -700,12 +734,51 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	node->max_proc_per_quantum = max_proc_per_quantum;
 	atomic_set(&node->sram_ecc_flag, 0);
 
-	/* Initialize the KFD node */
-	if (kfd_init_node(node)) {
-		dev_err(kfd_device, "Error initializing KFD node\n");
-		goto node_init_error;
+	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
+				kfd->num_nodes);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL);
+		if (!node)
+			goto node_alloc_error;
+
+		node->adev = kfd->adev;
+		node->kfd = kfd;
+		node->kfd2kgd = kfd->kfd2kgd;
+		node->vm_info.vmid_num_kfd = vmid_num_kfd;
+		node->num_xcc_per_node = max(1U, kfd->adev->gfx.num_xcc_per_xcp);
+		node->start_xcc_id = node->num_xcc_per_node * i;
+
+		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
+		    kfd->adev->gfx.partition_mode == AMDGPU_CPX_PARTITION_MODE &&
+		    kfd->num_nodes != 1) {
+			/* For GFX9.4.3 and CPX mode, first XCD gets VMID range
+			 * 4-9 and second XCD gets VMID range 10-15.
+			 */
+
+			node->vm_info.first_vmid_kfd = (i%2 == 0) ?
+						first_vmid_kfd :
+						first_vmid_kfd+vmid_num_kfd;
+			node->vm_info.last_vmid_kfd = (i%2 == 0) ?
+						last_vmid_kfd-vmid_num_kfd :
+						last_vmid_kfd;
+			node->compute_vmid_bitmap =
+				((0x1 << (node->vm_info.last_vmid_kfd + 1)) - 1) -
+				((0x1 << (node->vm_info.first_vmid_kfd)) - 1);
+		} else {
+			node->vm_info.first_vmid_kfd = first_vmid_kfd;
+			node->vm_info.last_vmid_kfd = last_vmid_kfd;
+			node->compute_vmid_bitmap =
+				gpu_resources->compute_vmid_bitmap;
+		}
+		node->max_proc_per_quantum = max_proc_per_quantum;
+		atomic_set(&node->sram_ecc_flag, 0);
+		/* Initialize the KFD node */
+		if (kfd_init_node(node)) {
+			dev_err(kfd_device, "Error initializing KFD node\n");
+			goto node_init_error;
+		}
+		kfd->nodes[i] = node;
 	}
-	kfd->node = node;
 
 	if (kfd_resume_iommu(kfd))
 		goto kfd_resume_iommu_error;
@@ -722,9 +795,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	goto out;
 
 kfd_resume_iommu_error:
-	kfd_cleanup_node(kfd);
 node_init_error:
 node_alloc_error:
+	kfd_cleanup_nodes(kfd, i);
 device_iommu_error:
 	kfd_doorbell_fini(kfd);
 kfd_doorbell_error:
@@ -742,7 +815,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 void kgd2kfd_device_exit(struct kfd_dev *kfd)
 {
 	if (kfd->init_complete) {
-		kfd_cleanup_node(kfd);
+		/* Cleanup KFD nodes */
+		kfd_cleanup_nodes(kfd, kfd->num_nodes);
+		/* Cleanup common/shared resources */
 		kfd_doorbell_fini(kfd);
 		ida_destroy(&kfd->doorbell_ida);
 		kfd_gtt_sa_fini(kfd);
@@ -754,18 +829,23 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
 
 int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 {
-	struct kfd_node *node = kfd->node;
+	struct kfd_node *node;
+	int i;
 
 	if (!kfd->init_complete)
 		return 0;
 
-	kfd_smi_event_update_gpu_reset(node, false);
-
-	node->dqm->ops.pre_reset(node->dqm);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		kfd_smi_event_update_gpu_reset(node, false);
+		node->dqm->ops.pre_reset(node->dqm);
+	}
 
 	kgd2kfd_suspend(kfd, false);
 
-	kfd_signal_reset_event(node);
+	for (i = 0; i < kfd->num_nodes; i++)
+		kfd_signal_reset_event(kfd->nodes[i]);
+
 	return 0;
 }
 
@@ -778,19 +858,25 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd)
 int kgd2kfd_post_reset(struct kfd_dev *kfd)
 {
 	int ret;
-	struct kfd_node *node = kfd->node;
+	struct kfd_node *node;
+	int i;
 
 	if (!kfd->init_complete)
 		return 0;
 
-	ret = kfd_resume(node);
-	if (ret)
-		return ret;
-	atomic_dec(&kfd_locked);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		ret = kfd_resume(kfd->nodes[i]);
+		if (ret)
+			return ret;
+	}
 
-	atomic_set(&node->sram_ecc_flag, 0);
+	atomic_dec(&kfd_locked);
 
-	kfd_smi_event_update_gpu_reset(node, true);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		atomic_set(&node->sram_ecc_flag, 0);
+		kfd_smi_event_update_gpu_reset(node, true);
+	}
 
 	return 0;
 }
@@ -802,7 +888,8 @@ bool kfd_is_locked(void)
 
 void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
-	struct kfd_node *node = kfd->node;
+	struct kfd_node *node;
+	int i;
 
 	if (!kfd->init_complete)
 		return;
@@ -814,21 +901,25 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 			kfd_suspend_all_processes();
 	}
 
-	node->dqm->ops.stop(node->dqm);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		node->dqm->ops.stop(node->dqm);
+	}
 	kfd_iommu_suspend(kfd);
 }
 
 int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
-	int ret, count;
-	struct kfd_node *node = kfd->node;
+	int ret, count, i;
 
 	if (!kfd->init_complete)
 		return 0;
 
-	ret = kfd_resume(node);
-	if (ret)
-		return ret;
+	for (i = 0; i < kfd->num_nodes; i++) {
+		ret = kfd_resume(kfd->nodes[i]);
+		if (ret)
+			return ret;
+	}
 
 	/* for runtime resume, skip unlocking kfd */
 	if (!run_pm) {
@@ -892,10 +983,10 @@ static inline void kfd_queue_work(struct workqueue_struct *wq,
 /* This is called directly from KGD at ISR. */
 void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 {
-	uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE];
+	uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE], i;
 	bool is_patched = false;
 	unsigned long flags;
-	struct kfd_node *node = kfd->node;
+	struct kfd_node *node;
 
 	if (!kfd->init_complete)
 		return;
@@ -905,16 +996,22 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 		return;
 	}
 
-	spin_lock_irqsave(&node->interrupt_lock, flags);
-
-	if (node->interrupts_active
-	    && interrupt_is_wanted(node, ih_ring_entry,
-				   patched_ihre, &is_patched)
-	    && enqueue_ih_ring_entry(node,
-				     is_patched ? patched_ihre : ih_ring_entry))
-		kfd_queue_work(node->ih_wq, &node->interrupt_work);
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		spin_lock_irqsave(&node->interrupt_lock, flags);
+
+		if (node->interrupts_active
+		    && interrupt_is_wanted(node, ih_ring_entry,
+			    	patched_ihre, &is_patched)
+		    && enqueue_ih_ring_entry(node,
+			    	is_patched ? patched_ihre : ih_ring_entry)) {
+			kfd_queue_work(node->ih_wq, &node->interrupt_work);
+			spin_unlock_irqrestore(&node->interrupt_lock, flags);
+				return;
+		}
+		spin_unlock_irqrestore(&node->interrupt_lock, flags);
+	}
 
-	spin_unlock_irqrestore(&node->interrupt_lock, flags);
 }
 
 int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger)
@@ -1181,8 +1278,13 @@ int kfd_gtt_sa_free(struct kfd_node *node, struct kfd_mem_obj *mem_obj)
 
 void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd)
 {
+	/*
+	 * TODO: Currently update SRAM ECC flag for first node.
+	 * This needs to be updated later when we can
+	 * identify SRAM ECC error on other nodes also.
+	 */
 	if (kfd)
-		atomic_inc(&kfd->node->sram_ecc_flag);
+		atomic_inc(&kfd->nodes[0]->sram_ecc_flag);
 }
 
 void kfd_inc_compute_active(struct kfd_node *node)
@@ -1202,8 +1304,14 @@ void kfd_dec_compute_active(struct kfd_node *node)
 
 void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask)
 {
+	/*
+	 * TODO: For now, raise the throttling event only on first node.
+	 * This will need to change after we are able to determine
+	 * which node raised the throttling event.
+	 */
 	if (kfd && kfd->init_complete)
-		kfd_smi_event_update_thermal_throttling(kfd->node, throttle_bitmask);
+		kfd_smi_event_update_thermal_throttling(kfd->nodes[0],
+							throttle_bitmask);
 }
 
 /* kfd_get_num_sdma_engines returns the number of PCIe optimized SDMA and
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 34977d89f01c4..6ee17100c3336 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1426,7 +1426,7 @@ static int set_sched_resources(struct device_queue_manager *dqm)
 	int i, mec;
 	struct scheduling_resources res;
 
-	res.vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap;
+	res.vmid_mask = dqm->dev->compute_vmid_bitmap;
 
 	res.queue_mask = 0;
 	for (i = 0; i < KGD_MAX_QUEUES; ++i) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
index 6eee9a0944f3e..808ee010520a2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c
@@ -121,6 +121,12 @@ int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd)
 		return -EINVAL;
 	}
 
+	if (!kfd_is_first_node(dev)) {
+		dev_warn_once(kfd_device,
+				"IOMMU supported only on first node\n");
+		return 0;
+	}
+
 	err = amd_iommu_bind_pasid(dev->adev->pdev, p->pasid, p->lead_thread);
 	if (!err)
 		pdd->bound = PDD_BOUND;
@@ -138,7 +144,8 @@ void kfd_iommu_unbind_process(struct kfd_process *p)
 	int i;
 
 	for (i = 0; i < p->n_pdds; i++)
-		if (p->pdds[i]->bound == PDD_BOUND)
+		if ((p->pdds[i]->bound == PDD_BOUND) &&
+		    (kfd_is_first_node((p->pdds[i]->dev))))
 			amd_iommu_unbind_pasid(p->pdds[i]->dev->adev->pdev,
 					       p->pasid);
 }
@@ -281,7 +288,7 @@ void kfd_iommu_suspend(struct kfd_dev *kfd)
 	if (!kfd->use_iommu_v2)
 		return;
 
-	kfd_unbind_processes_from_device(kfd->node);
+	kfd_unbind_processes_from_device(kfd->nodes[0]);
 
 	amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL);
 	amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL);
@@ -312,7 +319,7 @@ int kfd_iommu_resume(struct kfd_dev *kfd)
 	amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev,
 				     iommu_invalid_ppr_cb);
 
-	err = kfd_bind_processes_to_device(kfd->node);
+	err = kfd_bind_processes_to_device(kfd->nodes[0]);
 	if (err) {
 		amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL);
 		amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 1e187677c90ac..5f4dc2a45bd03 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -423,7 +423,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      0, adev->kfd.dev->node->id, prange->prefetch_loc,
+				      0, adev->kfd.dev->nodes[0]->id, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -456,7 +456,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    0, adev->kfd.dev->node->id, trigger);
+				    0, adev->kfd.dev->nodes[0]->id, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 	svm_range_free_dma_mappings(prange);
@@ -701,7 +701,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      adev->kfd.dev->node->id, 0, prange->prefetch_loc,
+				      adev->kfd.dev->nodes[0]->id, 0, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -737,7 +737,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    adev->kfd.dev->node->id, 0, trigger);
+				    adev->kfd.dev->nodes[0]->id, 0, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index fdb97e5d0c01a..873b49238dc15 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -255,6 +255,8 @@ struct kfd_vmid_info {
 	uint32_t vmid_num_kfd;
 };
 
+#define MAX_KFD_NODES	8
+
 struct kfd_dev;
 
 struct kfd_node {
@@ -267,6 +269,10 @@ struct kfd_node {
 					      */
 	struct kfd_vmid_info vm_info;
 	unsigned int id;                /* topology stub index */
+	unsigned int num_xcc_per_node;
+	unsigned int start_xcc_id;	/* Starting XCC instance
+					 * number for the node
+					 */
 	/* Interrupts */
 	struct kfifo ih_fifo;
 	struct workqueue_struct *ih_wq;
@@ -300,6 +306,8 @@ struct kfd_node {
 	/* Maximum process number mapped to HW scheduler */
 	unsigned int max_proc_per_quantum;
 
+	unsigned int compute_vmid_bitmap;
+
 	struct kfd_dev *kfd;
 };
 
@@ -368,7 +376,8 @@ struct kfd_dev {
 	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
 	struct dev_pagemap pgmap;
 
-	struct kfd_node *node;
+	struct kfd_node *nodes[MAX_KFD_NODES];
+	unsigned int num_nodes;
 };
 
 enum kfd_mempool {
@@ -1397,6 +1406,11 @@ static inline int kfd_devcgroup_check_permission(struct kfd_node *kfd)
 #endif
 }
 
+static inline bool kfd_is_first_node(struct kfd_node *node)
+{
+	return (node == node->kfd->nodes[0]);
+}
+
 /* Debugfs */
 #if defined(CONFIG_DEBUG_FS)
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index a0bf6558f4acc..b703da59e0677 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -254,17 +254,17 @@ void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid,
 				    unsigned long address, bool write_fault,
 				    ktime_t ts)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_PAGE_FAULT_START,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_PAGE_FAULT_START,
 			  "%lld -%d @%lx(%x) %c\n", ktime_to_ns(ts), pid,
-			  address, dev->node->id, write_fault ? 'W' : 'R');
+			  address, dev->nodes[0]->id, write_fault ? 'W' : 'R');
 }
 
 void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid,
 				  unsigned long address, bool migration)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_PAGE_FAULT_END,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_PAGE_FAULT_END,
 			  "%lld -%d @%lx(%x) %c\n", ktime_get_boottime_ns(),
-			  pid, address, dev->node->id, migration ? 'M' : 'U');
+			  pid, address, dev->nodes[0]->id, migration ? 'M' : 'U');
 }
 
 void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
@@ -273,7 +273,7 @@ void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
 				   uint32_t prefetch_loc, uint32_t preferred_loc,
 				   uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_MIGRATE_START,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_MIGRATE_START,
 			  "%lld -%d @%lx(%lx) %x->%x %x:%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, prefetch_loc, preferred_loc, trigger);
@@ -283,7 +283,7 @@ void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
 				 unsigned long start, unsigned long end,
 				 uint32_t from, uint32_t to, uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_MIGRATE_END,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_MIGRATE_END,
 			  "%lld -%d @%lx(%lx) %x->%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, trigger);
@@ -292,16 +292,16 @@ void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
 void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_QUEUE_EVICTION,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_QUEUE_EVICTION,
 			  "%lld -%d %x %d\n", ktime_get_boottime_ns(), pid,
-			  dev->node->id, trigger);
+			  dev->nodes[0]->id, trigger);
 }
 
 void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_QUEUE_RESTORE,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_QUEUE_RESTORE,
 			  "%lld -%d %x\n", ktime_get_boottime_ns(), pid,
-			  dev->node->id);
+			  dev->nodes[0]->id);
 }
 
 void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
@@ -328,9 +328,9 @@ void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid,
 				  unsigned long address, unsigned long last,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->node, KFD_SMI_EVENT_UNMAP_FROM_GPU,
+	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_UNMAP_FROM_GPU,
 			  "%lld -%d @%lx(%lx) %x %d\n", ktime_get_boottime_ns(),
-			  pid, address, last - address + 1, dev->node->id, trigger);
+			  pid, address, last - address + 1, dev->nodes[0]->id, trigger);
 }
 
 int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 06a11186d947c..94af37df3ed2f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -555,7 +555,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 				      dev->gpu->kfd->sdma_fw_version);
 		sysfs_show_64bit_prop(buffer, offs, "unique_id",
 				      dev->gpu->adev->unique_id);
-
+		sysfs_show_32bit_prop(buffer, offs, "num_xcc",
+				      dev->gpu->num_xcc_per_node);
 	}
 
 	return sysfs_show_32bit_prop(buffer, offs, "max_engine_clk_ccompute",
@@ -1160,7 +1161,7 @@ void kfd_topology_shutdown(void)
 static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 {
 	uint32_t hashout;
-	uint32_t buf[7];
+	uint32_t buf[8];
 	uint64_t local_mem_size;
 	int i;
 
@@ -1177,8 +1178,9 @@ static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 	buf[4] = gpu->adev->pdev->bus->number;
 	buf[5] = lower_32_bits(local_mem_size);
 	buf[6] = upper_32_bits(local_mem_size);
+	buf[7] = gpu->start_xcc_id | (gpu->num_xcc_per_node << 16);
 
-	for (i = 0, hashout = 0; i < 7; i++)
+	for (i = 0, hashout = 0; i < 8; i++)
 		hashout ^= hash_32(buf[i], KFD_GPU_ID_HASH_WIDTH);
 
 	return hashout;
-- 
GitLab


From 2f77b9a242a2e01822efc80c8b63eaa31df0f8b4 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 21:45:50 -0400
Subject: [PATCH 1047/1791] drm/amdkfd: Update MQD management on multi XCC
 setup

Update MQD management for both HIQ and user-mode compute
queues on a multi XCC setup. MQDs needs to be allocated,
initialized, loaded and destroyed for each XCC in the KFD
node.

v2: squash in fix "drm/amdkfd: Fix SDMA+HIQ HQD allocation on GFX9.4.3"

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c       |  51 ++-
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |   3 +-
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c  |  28 +-
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h  |   8 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c  |   3 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |   3 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   | 292 ++++++++++++++++--
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c   |   3 +
 .../amd/amdkfd/kfd_process_queue_manager.c    |  16 +-
 drivers/gpu/drm/amd/include/v9_structs.h      |  30 +-
 10 files changed, 380 insertions(+), 57 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index f61f07575f63b..6bbe3b89aef5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -800,6 +800,41 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev,
 	sg_free_table(ttm->sg);
 }
 
+/*
+ * total_pages is constructed as MQD0+CtrlStack0 + MQD1+CtrlStack1 + ...
+ * MQDn+CtrlStackn where n is the number of XCCs per partition.
+ * pages_per_xcc is the size of one MQD+CtrlStack. The first page is MQD
+ * and uses memory type default, UC. The rest of pages_per_xcc are
+ * Ctrl stack and modify their memory type to NC.
+ */
+static void amdgpu_ttm_gart_bind_gfx9_mqd(struct amdgpu_device *adev,
+				struct ttm_tt *ttm, uint64_t flags)
+{
+	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	uint64_t total_pages = ttm->num_pages;
+	int num_xcc = max(1U, adev->gfx.num_xcc_per_xcp);
+	uint64_t page_idx, pages_per_xcc = total_pages / num_xcc;
+	int i;
+	uint64_t ctrl_flags = (flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
+			AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC);
+
+	for (i = 0, page_idx = 0; i < num_xcc; i++, page_idx += pages_per_xcc) {
+		/* MQD page: use default flags */
+		amdgpu_gart_bind(adev,
+				gtt->offset + (page_idx << PAGE_SHIFT),
+				1, &gtt->ttm.dma_address[page_idx], flags);
+		/*
+		 * Ctrl pages - modify the memory type to NC (ctrl_flags) from
+		 * the second page of the BO onward.
+		 */
+		amdgpu_gart_bind(adev,
+				gtt->offset + ((page_idx + 1) << PAGE_SHIFT),
+				pages_per_xcc - 1,
+				&gtt->ttm.dma_address[page_idx + 1],
+				ctrl_flags);
+	}
+}
+
 static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev,
 				 struct ttm_buffer_object *tbo,
 				 uint64_t flags)
@@ -812,21 +847,7 @@ static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev,
 		flags |= AMDGPU_PTE_TMZ;
 
 	if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) {
-		uint64_t page_idx = 1;
-
-		amdgpu_gart_bind(adev, gtt->offset, page_idx,
-				 gtt->ttm.dma_address, flags);
-
-		/* The memory type of the first page defaults to UC. Now
-		 * modify the memory type to NC from the second page of
-		 * the BO onward.
-		 */
-		flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK;
-		flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC);
-
-		amdgpu_gart_bind(adev, gtt->offset + (page_idx << PAGE_SHIFT),
-				 ttm->num_pages - page_idx,
-				 &(gtt->ttm.dma_address[page_idx]), flags);
+		amdgpu_ttm_gart_bind_gfx9_mqd(adev, ttm, flags);
 	} else {
 		amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages,
 				 gtt->ttm.dma_address, flags);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 6ee17100c3336..9afd3295ca850 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2247,7 +2247,8 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm)
 	uint32_t size = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size *
 		get_num_all_sdma_engines(dqm) *
 		dev->kfd->device_info.num_sdma_queues_per_engine +
-		dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size;
+		(dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size *
+		dqm->dev->num_xcc_per_node);
 
 	retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size,
 		&(mem_obj->gtt_mem), &(mem_obj->gpu_addr),
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
index 61f6dd68c84b3..074f6075ccc71 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
@@ -76,7 +76,8 @@ struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev,
 		q->sdma_queue_id) *
 		dev->dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size;
 
-	offset += dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size;
+	offset += dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size *
+		  dev->num_xcc_per_node;
 
 	mqd_mem_obj->gtt_mem = (void *)((uint64_t)dev->dqm->hiq_sdma_mqd.gtt_mem
 				+ offset);
@@ -246,3 +247,28 @@ bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd,
 {
 	return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd);
 }
+
+uint64_t kfd_hiq_mqd_stride(struct kfd_node *dev)
+{
+	return dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size;
+}
+
+void kfd_get_hiq_xcc_mqd(struct kfd_node *dev, struct kfd_mem_obj *mqd_mem_obj,
+		     uint32_t virtual_xcc_id)
+{
+	uint64_t offset;
+
+	offset = kfd_hiq_mqd_stride(dev) * virtual_xcc_id;
+
+	mqd_mem_obj->gtt_mem = (virtual_xcc_id == 0) ?
+			dev->dqm->hiq_sdma_mqd.gtt_mem : NULL;
+	mqd_mem_obj->gpu_addr = dev->dqm->hiq_sdma_mqd.gpu_addr + offset;
+	mqd_mem_obj->cpu_ptr = (uint32_t *)((uintptr_t)
+				dev->dqm->hiq_sdma_mqd.cpu_ptr + offset);
+}
+
+uint64_t kfd_mqd_stride(struct mqd_manager *mm,
+			struct queue_properties *q)
+{
+	return mm->mqd_size;
+}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
index 46fc3f273d0dd..eb18be74f5599 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
@@ -119,6 +119,8 @@ struct mqd_manager {
 	int	(*debugfs_show_mqd)(struct seq_file *m, void *data);
 #endif
 	uint32_t (*read_doorbell_id)(void *mqd);
+	uint64_t (*mqd_stride)(struct mqd_manager *mm,
+				struct queue_properties *p);
 
 	struct mutex	mqd_mutex;
 	struct kfd_node	*dev;
@@ -164,4 +166,10 @@ bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd,
 		uint64_t queue_address, uint32_t pipe_id,
 		uint32_t queue_id);
 
+void kfd_get_hiq_xcc_mqd(struct kfd_node *dev,
+		struct kfd_mem_obj *mqd_mem_obj, uint32_t virtual_xcc_id);
+
+uint64_t kfd_hiq_mqd_stride(struct kfd_node *dev);
+uint64_t kfd_mqd_stride(struct mqd_manager *mm,
+			struct queue_properties *q);
 #endif /* KFD_MQD_MANAGER_H_ */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
index 03e04d5e5a11c..ca19664667590 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
@@ -428,6 +428,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
 		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct cik_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -442,6 +443,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
 		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct cik_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -457,6 +459,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type,
 		mqd->checkpoint_mqd = checkpoint_mqd_sdma;
 		mqd->restore_mqd = restore_mqd_sdma;
 		mqd->mqd_size = sizeof(struct cik_sdma_rlc_registers);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd_sdma;
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index 7a93be0ebb19b..c9565ea99df53 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -432,6 +432,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type,
 		mqd->get_wave_state = get_wave_state;
 		mqd->checkpoint_mqd = checkpoint_mqd;
 		mqd->restore_mqd = restore_mqd;
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -447,6 +448,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type,
 		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct v10_compute_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -478,6 +480,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type,
 		mqd->checkpoint_mqd = checkpoint_mqd_sdma;
 		mqd->restore_mqd = restore_mqd_sdma;
 		mqd->mqd_size = sizeof(struct v10_sdma_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd_sdma;
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 943a738e73f95..c677322057dd9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -33,6 +33,21 @@
 #include "sdma0/sdma0_4_0_sh_mask.h"
 #include "amdgpu_amdkfd.h"
 
+static void update_mqd(struct mqd_manager *mm, void *mqd,
+		       struct queue_properties *q,
+		       struct mqd_update_info *minfo);
+
+static uint64_t mqd_stride_v9(struct mqd_manager *mm,
+				struct queue_properties *q)
+{
+	if (mm->dev->kfd->cwsr_enabled &&
+	    q->type == KFD_QUEUE_TYPE_COMPUTE)
+		return ALIGN(q->ctl_stack_size, PAGE_SIZE) +
+			ALIGN(sizeof(struct v9_mqd), PAGE_SIZE);
+
+	return mm->mqd_size;
+}
+
 static inline struct v9_mqd *get_mqd(void *mqd)
 {
 	return (struct v9_mqd *)mqd;
@@ -110,8 +125,9 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node,
 		if (!mqd_mem_obj)
 			return NULL;
 		retval = amdgpu_amdkfd_alloc_gtt_mem(node->adev,
-			ALIGN(q->ctl_stack_size, PAGE_SIZE) +
-				ALIGN(sizeof(struct v9_mqd), PAGE_SIZE),
+			(ALIGN(q->ctl_stack_size, PAGE_SIZE) +
+			ALIGN(sizeof(struct v9_mqd), PAGE_SIZE)) *
+			node->num_xcc_per_node,
 			&(mqd_mem_obj->gtt_mem),
 			&(mqd_mem_obj->gpu_addr),
 			(void *)&(mqd_mem_obj->cpu_ptr), true);
@@ -165,24 +181,9 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT |
 			1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT;
 
-	if (q->format == KFD_QUEUE_FORMAT_AQL) {
+	if (q->format == KFD_QUEUE_FORMAT_AQL)
 		m->cp_hqd_aql_control =
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
-		if (KFD_GC_VERSION(mm->dev) == IP_VERSION(9, 4, 3)) {
-			/* On GC 9.4.3, DW 41 is re-purposed as
-			 * compute_tg_chunk_size.
-			 * TODO: review this setting when active CUs in the
-			 * partition play a role
-			 */
-			m->compute_static_thread_mgmt_se6 = 1;
-		}
-	} else {
-		/* PM4 queue */
-		if (KFD_GC_VERSION(mm->dev) == IP_VERSION(9, 4, 3)) {
-			m->compute_static_thread_mgmt_se6 = 0;
-			/* TODO: program pm4_target_xcc */
-		}
-	}
 
 	if (q->tba_addr) {
 		m->compute_pgm_rsrc2 |=
@@ -205,7 +206,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	*mqd = m;
 	if (gart_addr)
 		*gart_addr = addr;
-	mm->update_mqd(mm, m, q, NULL);
+	update_mqd(mm, m, q, NULL);
 }
 
 static int load_mqd(struct mqd_manager *mm, void *mqd,
@@ -269,13 +270,10 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 	m->cp_hqd_vmid = q->vmid;
 
 	if (q->format == KFD_QUEUE_FORMAT_AQL) {
-		m->cp_hqd_pq_control |=
+		m->cp_hqd_pq_control |= CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK |
 				2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT |
 				1 << CP_HQD_PQ_CONTROL__QUEUE_FULL_EN__SHIFT |
 				1 << CP_HQD_PQ_CONTROL__WPP_CLAMP_EN__SHIFT;
-		if (KFD_GC_VERSION(mm->dev) != IP_VERSION(9, 4, 3))
-			m->cp_hqd_pq_control |=
-				CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
 		m->cp_hqd_pq_doorbell_control |= 1 <<
 			CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT;
 	}
@@ -466,6 +464,224 @@ static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd,
 	qp->is_active = 0;
 }
 
+static void init_mqd_hiq_v9_4_3(struct mqd_manager *mm, void **mqd,
+			struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr,
+			struct queue_properties *q)
+{
+	struct v9_mqd *m;
+	int xcc = 0;
+	struct kfd_mem_obj xcc_mqd_mem_obj;
+	uint64_t xcc_gart_addr = 0;
+
+	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		kfd_get_hiq_xcc_mqd(mm->dev, &xcc_mqd_mem_obj, xcc);
+
+		init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q);
+
+		m->cp_hqd_pq_control |= CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK |
+					1 << CP_HQD_PQ_CONTROL__PRIV_STATE__SHIFT |
+					1 << CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT;
+		m->cp_mqd_stride_size = kfd_hiq_mqd_stride(mm->dev);
+		if (xcc == 0) {
+			/* Set no_update_rptr = 0 in Master XCC */
+			m->cp_hqd_pq_control &= ~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
+
+			/* Set the MQD pointer and gart address to XCC0 MQD */
+			*mqd = m;
+			*gart_addr = xcc_gart_addr;
+		}
+	}
+}
+
+static int hiq_load_mqd_kiq_v9_4_3(struct mqd_manager *mm, void *mqd,
+			uint32_t pipe_id, uint32_t queue_id,
+			struct queue_properties *p, struct mm_struct *mms)
+{
+	int xcc, err;
+	void *xcc_mqd;
+	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		xcc_mqd = mqd + hiq_mqd_size * xcc;
+		err = mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, xcc_mqd,
+						     pipe_id, queue_id,
+						     p->doorbell_off);
+		if (err) {
+			pr_debug("Failed to load HIQ MQD for XCC: %d\n", xcc);
+			break;
+		}
+	}
+
+	return err;
+}
+
+static int destroy_hiq_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
+			enum kfd_preempt_type type, unsigned int timeout,
+			uint32_t pipe_id, uint32_t queue_id)
+{
+	int xcc = 0, err;
+	void *xcc_mqd;
+	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		xcc_mqd = mqd + hiq_mqd_size * xcc;
+		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
+						    type, timeout, pipe_id,
+						    queue_id);
+		if (err) {
+			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
+			break;
+		}
+	}
+
+	return err;
+}
+
+static void get_xcc_mqd(struct kfd_mem_obj *mqd_mem_obj,
+			       struct kfd_mem_obj *xcc_mqd_mem_obj,
+			       uint64_t offset)
+{
+	xcc_mqd_mem_obj->gtt_mem = (offset == 0) ?
+					mqd_mem_obj->gtt_mem : NULL;
+	xcc_mqd_mem_obj->gpu_addr = mqd_mem_obj->gpu_addr + offset;
+	xcc_mqd_mem_obj->cpu_ptr = (uint32_t *)((uintptr_t)mqd_mem_obj->cpu_ptr
+						+ offset);
+}
+
+static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
+			struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr,
+			struct queue_properties *q)
+{
+	struct v9_mqd *m;
+	int xcc = 0;
+	struct kfd_mem_obj xcc_mqd_mem_obj;
+	uint64_t xcc_gart_addr = 0;
+	uint64_t offset = mm->mqd_stride(mm, q);
+
+	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		get_xcc_mqd(mqd_mem_obj, &xcc_mqd_mem_obj, offset*xcc);
+
+		init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q);
+
+		m->cp_mqd_stride_size = offset;
+		if (q->format == KFD_QUEUE_FORMAT_AQL) {
+			m->compute_tg_chunk_size = 1;
+
+			switch (xcc) {
+			case 0:
+				/* Master XCC */
+				m->cp_hqd_pq_control &=
+					~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
+				m->compute_current_logic_xcc_id =
+					mm->dev->num_xcc_per_node - 1;
+				break;
+			default:
+				m->compute_current_logic_xcc_id =
+					xcc - 1;
+				break;
+			}
+		} else {
+			/* PM4 Queue */
+			m->compute_current_logic_xcc_id = 0;
+			m->compute_tg_chunk_size = 0;
+		}
+
+		if (xcc == 0) {
+			/* Set the MQD pointer and gart address to XCC0 MQD */
+			*mqd = m;
+			*gart_addr = xcc_gart_addr;
+		}
+	}
+}
+
+static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
+		      struct queue_properties *q, struct mqd_update_info *minfo)
+{
+	struct v9_mqd *m;
+	int xcc = 0;
+	uint64_t size = mm->mqd_stride(mm, q);
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		m = get_mqd(mqd + size * xcc);
+		update_mqd(mm, m, q, minfo);
+
+		if (q->format == KFD_QUEUE_FORMAT_AQL) {
+			switch (xcc) {
+			case 0:
+				/* Master XCC */
+				m->cp_hqd_pq_control &=
+					~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
+				m->compute_current_logic_xcc_id =
+					mm->dev->num_xcc_per_node - 1;
+				break;
+			default:
+				m->compute_current_logic_xcc_id =
+					xcc - 1;
+				break;
+			}
+			m->compute_tg_chunk_size = 1;
+		} else {
+			/* PM4 Queue */
+			m->compute_current_logic_xcc_id = 0;
+			m->compute_tg_chunk_size = 0;
+		}
+	}
+}
+
+static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
+		   enum kfd_preempt_type type, unsigned int timeout,
+		   uint32_t pipe_id, uint32_t queue_id)
+{
+	int xcc = 0, err;
+	void *xcc_mqd;
+	struct v9_mqd *m;
+	uint64_t mqd_offset;
+
+	m = get_mqd(mqd);
+	mqd_offset = m->cp_mqd_stride_size;
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		xcc_mqd = mqd + mqd_offset * xcc;
+		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
+						    type, timeout, pipe_id,
+						    queue_id);
+		if (err) {
+			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
+			break;
+		}
+	}
+
+	return err;
+}
+
+static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
+			uint32_t pipe_id, uint32_t queue_id,
+			struct queue_properties *p, struct mm_struct *mms)
+{
+	/* AQL write pointer counts in 64B packets, PM4/CP counts in dwords. */
+	uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0);
+	int xcc = 0, err;
+	void *xcc_mqd;
+	uint64_t mqd_stride_size = mm->mqd_stride(mm, p);
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		xcc_mqd = mqd + mqd_stride_size * xcc;
+		err = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, xcc_mqd,
+					 pipe_id, queue_id,
+					(uint32_t __user *)p->write_ptr,
+					wptr_shift, 0, mms);
+		if (err) {
+			pr_debug("Load MQD failed for xcc: %d\n", xcc);
+			break;
+		}
+	}
+
+	return err;
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 static int debugfs_show_mqd(struct seq_file *m, void *data)
@@ -501,34 +717,49 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 	switch (type) {
 	case KFD_MQD_TYPE_CP:
 		mqd->allocate_mqd = allocate_mqd;
-		mqd->init_mqd = init_mqd;
 		mqd->free_mqd = kfd_free_mqd_cp;
-		mqd->load_mqd = load_mqd;
-		mqd->update_mqd = update_mqd;
-		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->get_wave_state = get_wave_state;
 		mqd->get_checkpoint_info = get_checkpoint_info;
 		mqd->checkpoint_mqd = checkpoint_mqd;
 		mqd->restore_mqd = restore_mqd;
 		mqd->mqd_size = sizeof(struct v9_mqd);
+		mqd->mqd_stride = mqd_stride_v9;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
+		if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)) {
+			mqd->init_mqd = init_mqd_v9_4_3;
+			mqd->load_mqd = load_mqd_v9_4_3;
+			mqd->update_mqd = update_mqd_v9_4_3;
+			mqd->destroy_mqd = destroy_mqd_v9_4_3;
+		} else {
+			mqd->init_mqd = init_mqd;
+			mqd->load_mqd = load_mqd;
+			mqd->update_mqd = update_mqd;
+			mqd->destroy_mqd = kfd_destroy_mqd_cp;
+		}
 		break;
 	case KFD_MQD_TYPE_HIQ:
 		mqd->allocate_mqd = allocate_hiq_mqd;
-		mqd->init_mqd = init_mqd_hiq;
 		mqd->free_mqd = free_mqd_hiq_sdma;
-		mqd->load_mqd = kfd_hiq_load_mqd_kiq;
 		mqd->update_mqd = update_mqd;
-		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct v9_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
 		mqd->read_doorbell_id = read_doorbell_id;
+		if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)) {
+			mqd->init_mqd = init_mqd_hiq_v9_4_3;
+			mqd->load_mqd = hiq_load_mqd_kiq_v9_4_3;
+			mqd->destroy_mqd = destroy_hiq_mqd_v9_4_3;
+		} else {
+			mqd->init_mqd = init_mqd_hiq;
+			mqd->load_mqd = kfd_hiq_load_mqd_kiq;
+			mqd->destroy_mqd = kfd_destroy_mqd_cp;
+		}
 		break;
 	case KFD_MQD_TYPE_DIQ:
 		mqd->allocate_mqd = allocate_mqd;
@@ -554,6 +785,7 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 		mqd->checkpoint_mqd = checkpoint_mqd_sdma;
 		mqd->restore_mqd = restore_mqd_sdma;
 		mqd->mqd_size = sizeof(struct v9_sdma_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd_sdma;
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index f6b4a5686dcb2..8736a3cdbe1e8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -486,6 +486,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
 		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct vi_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -500,6 +501,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
 		mqd->destroy_mqd = kfd_destroy_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->mqd_size = sizeof(struct vi_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd;
 #endif
@@ -515,6 +517,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type,
 		mqd->checkpoint_mqd = checkpoint_mqd_sdma;
 		mqd->restore_mqd = restore_mqd_sdma;
 		mqd->mqd_size = sizeof(struct vi_sdma_mqd);
+		mqd->mqd_stride = kfd_mqd_stride;
 #if defined(CONFIG_DEBUG_FS)
 		mqd->debugfs_show_mqd = debugfs_show_mqd_sdma;
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 5602498e713fa..b1fb017b2ef88 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -927,7 +927,9 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data)
 	struct queue *q;
 	enum KFD_MQD_TYPE mqd_type;
 	struct mqd_manager *mqd_mgr;
-	int r = 0;
+	int r = 0, xcc, num_xccs = 1;
+	void *mqd;
+	uint64_t size = 0;
 
 	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
 		if (pqn->q) {
@@ -943,6 +945,7 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data)
 				seq_printf(m, "  Compute queue on device %x\n",
 					   q->device->id);
 				mqd_type = KFD_MQD_TYPE_CP;
+				num_xccs = q->device->num_xcc_per_node;
 				break;
 			default:
 				seq_printf(m,
@@ -951,6 +954,8 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data)
 				continue;
 			}
 			mqd_mgr = q->device->dqm->mqd_mgrs[mqd_type];
+			size = mqd_mgr->mqd_stride(mqd_mgr,
+							&q->properties);
 		} else if (pqn->kq) {
 			q = pqn->kq->queue;
 			mqd_mgr = pqn->kq->mqd_mgr;
@@ -972,9 +977,12 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data)
 			continue;
 		}
 
-		r = mqd_mgr->debugfs_show_mqd(m, q->mqd);
-		if (r != 0)
-			break;
+		for (xcc = 0; xcc < num_xccs; xcc++) {
+			mqd = q->mqd + size * xcc;
+			r = mqd_mgr->debugfs_show_mqd(m, mqd);
+			if (r != 0)
+				break;
+		}
 	}
 
 	return r;
diff --git a/drivers/gpu/drm/amd/include/v9_structs.h b/drivers/gpu/drm/amd/include/v9_structs.h
index a0c672889fe43..a2f81b9c38afc 100644
--- a/drivers/gpu/drm/amd/include/v9_structs.h
+++ b/drivers/gpu/drm/amd/include/v9_structs.h
@@ -196,10 +196,20 @@ struct v9_mqd {
 	uint32_t compute_wave_restore_addr_lo;
 	uint32_t compute_wave_restore_addr_hi;
 	uint32_t compute_wave_restore_control;
-	uint32_t compute_static_thread_mgmt_se4;
-	uint32_t compute_static_thread_mgmt_se5;
-	uint32_t compute_static_thread_mgmt_se6;
-	uint32_t compute_static_thread_mgmt_se7;
+	union {
+		struct {
+			uint32_t compute_static_thread_mgmt_se4;
+			uint32_t compute_static_thread_mgmt_se5;
+			uint32_t compute_static_thread_mgmt_se6;
+			uint32_t compute_static_thread_mgmt_se7;
+		};
+		struct {
+			uint32_t compute_current_logic_xcc_id; // offset: 39  (0x27)
+			uint32_t compute_restart_cg_tg_id; // offset: 40  (0x28)
+			uint32_t compute_tg_chunk_size; // offset: 41  (0x29)
+			uint32_t compute_restore_tg_chunk_size; // offset: 42  (0x2A)
+		};
+	};
 	uint32_t reserved_43;
 	uint32_t reserved_44;
 	uint32_t reserved_45;
@@ -382,8 +392,16 @@ struct v9_mqd {
 	uint32_t iqtimer_pkt_dw29;
 	uint32_t iqtimer_pkt_dw30;
 	uint32_t iqtimer_pkt_dw31;
-	uint32_t reserved_225;
-	uint32_t reserved_226;
+	union {
+		struct {
+			uint32_t reserved_225;
+			uint32_t reserved_226;
+		};
+		struct {
+			uint32_t pm4_target_xcc_in_xcp; // offset: 225  (0xE1)
+			uint32_t cp_mqd_stride_size; // offset: 226  (0xE2)
+		};
+	};
 	uint32_t reserved_227;
 	uint32_t set_resources_header;
 	uint32_t set_resources_dw1;
-- 
GitLab


From 3c8bdb51be0e895010da62dfa173bb1227ff3b6f Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 21:50:43 -0400
Subject: [PATCH 1048/1791] drm/amdkfd: Add PM4 target XCC

In a device that supports multiple XCCs, unlike AQL queues, the PM4 queue
will be only processed in one XCC in the partitioning. This patch
re-purposes the queue percentage variable in create queue and update
queue ioctl for the user space to specify the target XCC.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      | 22 +++++++++++++++----
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  2 ++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  1 +
 .../amd/amdkfd/kfd_process_queue_manager.c    |  1 +
 4 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index eb0b0b38f10eb..45e8da125f70b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -186,7 +186,12 @@ static int kfd_ioctl_get_version(struct file *filep, struct kfd_process *p,
 static int set_queue_properties_from_user(struct queue_properties *q_properties,
 				struct kfd_ioctl_create_queue_args *args)
 {
-	if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) {
+	/*
+	 * Repurpose queue percentage to accommodate new features:
+	 * bit 0-7: queue percentage
+	 * bit 8-15: pm4_target_xcc
+	 */
+	if ((args->queue_percentage & 0xFF) > KFD_MAX_QUEUE_PERCENTAGE) {
 		pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
 		return -EINVAL;
 	}
@@ -236,7 +241,9 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties,
 
 	q_properties->is_interop = false;
 	q_properties->is_gws = false;
-	q_properties->queue_percent = args->queue_percentage;
+	q_properties->queue_percent = args->queue_percentage & 0xFF;
+	/* bit 8-15 are repurposed to be PM4 target XCC */
+	q_properties->pm4_target_xcc = (args->queue_percentage >> 8) & 0xFF;
 	q_properties->priority = args->queue_priority;
 	q_properties->queue_address = args->ring_base_address;
 	q_properties->queue_size = args->ring_size;
@@ -442,7 +449,12 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
 	struct kfd_ioctl_update_queue_args *args = data;
 	struct queue_properties properties;
 
-	if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) {
+	/*
+	 * Repurpose queue percentage to accommodate new features:
+	 * bit 0-7: queue percentage
+	 * bit 8-15: pm4_target_xcc
+	 */
+	if ((args->queue_percentage & 0xFF) > KFD_MAX_QUEUE_PERCENTAGE) {
 		pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n");
 		return -EINVAL;
 	}
@@ -466,7 +478,9 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p,
 
 	properties.queue_address = args->ring_base_address;
 	properties.queue_size = args->ring_size;
-	properties.queue_percent = args->queue_percentage;
+	properties.queue_percent = args->queue_percentage & 0xFF;
+	/* bit 8-15 are repurposed to be PM4 target XCC */
+	properties.pm4_target_xcc = (args->queue_percentage >> 8) & 0xFF;
 	properties.priority = args->queue_priority;
 
 	pr_debug("Updating queue id %d for pasid 0x%x\n",
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index c677322057dd9..b46c984b3a179 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -587,6 +587,7 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 			/* PM4 Queue */
 			m->compute_current_logic_xcc_id = 0;
 			m->compute_tg_chunk_size = 0;
+			m->pm4_target_xcc_in_xcp = q->pm4_target_xcc;
 		}
 
 		if (xcc == 0) {
@@ -627,6 +628,7 @@ static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 			/* PM4 Queue */
 			m->compute_current_logic_xcc_id = 0;
 			m->compute_tg_chunk_size = 0;
+			m->pm4_target_xcc_in_xcp = q->pm4_target_xcc;
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 873b49238dc15..1337fcdf8958a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -509,6 +509,7 @@ struct queue_properties {
 	bool is_evicted;
 	bool is_active;
 	bool is_gws;
+	uint32_t pm4_target_xcc;
 	/* Not relevant for user mode queues in cp scheduling */
 	unsigned int vmid;
 	/* Relevant only for sdma queues*/
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index b1fb017b2ef88..2b2ae0c9902bd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -477,6 +477,7 @@ int pqm_update_queue_properties(struct process_queue_manager *pqm,
 	pqn->q->properties.queue_size = p->queue_size;
 	pqn->q->properties.queue_percent = p->queue_percent;
 	pqn->q->properties.priority = p->priority;
+	pqn->q->properties.pm4_target_xcc = p->pm4_target_xcc;
 
 	retval = pqn->q->device->dqm->ops.update_queue(pqn->q->device->dqm,
 							pqn->q, NULL);
-- 
GitLab


From e2069a7b0880ccdc6fa6530b6091e47168705425 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:20:52 -0400
Subject: [PATCH 1049/1791] drm/amdkfd: Add XCC instance to kgd2kfd interface
 (v3)

Gfx 9 starts to have multiple XCC instances in one device. Add instance
parameter to kgd2kfd functions where XCC instance was hard coded as 0.
Also, update code to pass the correct instance number when running
on a multi-XCC setup.

v2: introduce the XCC instance to gfx v11 (Morris)
v3: rebase (Alex)

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Morris Zhang <Shiwu.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   |  38 ++---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    |  22 +--
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |  27 +--
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    |  19 ++-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c |  17 +-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c |  17 +-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 160 +++++++++---------
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |  29 ++--
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 103 +++++++----
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c  |   6 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c  |   2 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |   2 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  |   2 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  15 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c   |   2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |   2 +-
 .../gpu/drm/amd/include/kgd_kfd_interface.h   |  25 +--
 17 files changed, 270 insertions(+), 218 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 562e1a04160f1..49d8087e469ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -33,7 +33,7 @@
 #include "soc15.h"
 
 static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
-				u32 pasid, unsigned int vmid)
+				u32 pasid, unsigned int vmid, uint32_t inst)
 {
 	unsigned long timeout;
 
@@ -47,11 +47,11 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 	uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid |
 			ATC_VMID0_PASID_MAPPING__VALID_MASK;
 
-	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
+	WREG32(SOC15_REG_OFFSET(ATHUB, inst,
 		regATC_VMID0_PASID_MAPPING) + vmid, pasid_mapping);
 
 	timeout = jiffies + msecs_to_jiffies(10);
-	while (!(RREG32(SOC15_REG_OFFSET(ATHUB, 0,
+	while (!(RREG32(SOC15_REG_OFFSET(ATHUB, inst,
 			regATC_VMID_PASID_MAPPING_UPDATE_STATUS)) &
 			(1U << vmid))) {
 		if (time_after(jiffies, timeout)) {
@@ -61,13 +61,13 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 		cpu_relax();
 	}
 
-	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
+	WREG32(SOC15_REG_OFFSET(ATHUB, inst,
 		regATC_VMID_PASID_MAPPING_UPDATE_STATUS),
 		1U << vmid);
 
-	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid,
+	WREG32(SOC15_REG_OFFSET(OSSSYS, inst, mmIH_VMID_0_LUT) + vmid,
 		pasid_mapping);
-	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid,
+	WREG32(SOC15_REG_OFFSET(OSSSYS, inst, mmIH_VMID_0_LUT_MM) + vmid,
 		pasid_mapping);
 
 	return 0;
@@ -81,7 +81,7 @@ static inline struct v9_mqd *get_mqd(void *mqd)
 static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst)
 {
 	struct v9_mqd *m;
 	uint32_t *mqd_hqd;
@@ -89,12 +89,12 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	m = get_mqd(mqd);
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
 	/* HQD registers extend to CP_HQD_AQL_DISPATCH_ID_HI */
 	mqd_hqd = &m->cp_mqd_base_addr_lo;
-	hqd_base = SOC15_REG_OFFSET(GC, 0, regCP_MQD_BASE_ADDR);
-	hqd_end = SOC15_REG_OFFSET(GC, 0, regCP_HQD_AQL_DISPATCH_ID_HI);
+	hqd_base = SOC15_REG_OFFSET(GC, inst, regCP_MQD_BASE_ADDR);
+	hqd_end = SOC15_REG_OFFSET(GC, inst, regCP_HQD_AQL_DISPATCH_ID_HI);
 
 	for (reg = hqd_base; reg <= hqd_end; reg++)
 		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
@@ -103,7 +103,7 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_DOORBELL_CONTROL),
 				data);
 
 	if (wptr) {
@@ -133,29 +133,29 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_LO),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_LO),
 		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_HI),
 		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_PQ_WPTR_POLL_ADDR),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_POLL_ADDR),
 		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0,
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst,
 			regCP_HQD_PQ_WPTR_POLL_ADDR_HI),
 			upper_32_bits((uintptr_t)wptr));
-		WREG32(SOC15_REG_OFFSET(GC, 0, regCP_PQ_WPTR_POLL_CNTL1),
+		WREG32(SOC15_REG_OFFSET(GC, inst, regCP_PQ_WPTR_POLL_CNTL1),
 		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id,
 			       queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_EOP_RPTR),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_EOP_RPTR),
 	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
 			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, regCP_HQD_ACTIVE), data);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_ACTIVE), data);
 
-	kgd_gfx_v9_release_queue(adev);
+	kgd_gfx_v9_release_queue(adev, inst);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index f599e1e74fcc2..7b60268d93c0c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -79,7 +79,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -91,7 +91,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 }
 
 static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -135,7 +135,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
  * but still works
  */
 
-static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id)
+static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -205,7 +206,7 @@ static inline struct v10_sdma_mqd *get_sdma_mqd(void *mqd)
 static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst)
 {
 	struct v10_compute_mqd *m;
 	uint32_t *mqd_hqd;
@@ -286,7 +287,7 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
-			    uint32_t doorbell_off)
+			    uint32_t doorbell_off, uint32_t inst)
 {
 	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v10_compute_mqd *m;
@@ -338,7 +339,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 
 static int kgd_hqd_dump(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS 56
@@ -469,7 +470,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev,
 
 static bool kgd_hqd_is_occupied(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
@@ -510,7 +511,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	enum hqd_dequeue_request_type type;
 	unsigned long end_jiffies;
@@ -673,7 +674,7 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev,
 
 static int kgd_wave_control_execute(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data = 0;
 
@@ -709,7 +710,8 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev,
 }
 
 static void program_trap_handler_settings(struct amdgpu_device *adev,
-		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr)
+		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
+		uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 5c4152ae44da0..52d0d35fb58db 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -80,7 +80,7 @@ static void program_sh_mem_settings_v10_3(struct amdgpu_device *adev, uint32_t v
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -93,7 +93,7 @@ static void program_sh_mem_settings_v10_3(struct amdgpu_device *adev, uint32_t v
 
 /* ATC is defeatured on Sienna_Cichlid */
 static int set_pasid_vmid_mapping_v10_3(struct amdgpu_device *adev, unsigned int pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	uint32_t value = pasid << IH_VMID_0_LUT__PASID__SHIFT;
 
@@ -105,7 +105,8 @@ static int set_pasid_vmid_mapping_v10_3(struct amdgpu_device *adev, unsigned int
 	return 0;
 }
 
-static int init_interrupts_v10_3(struct amdgpu_device *adev, uint32_t pipe_id)
+static int init_interrupts_v10_3(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -177,7 +178,7 @@ static inline struct v10_sdma_mqd *get_sdma_mqd(void *mqd)
 static int hqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst)
 {
 	struct v10_compute_mqd *m;
 	uint32_t *mqd_hqd;
@@ -273,7 +274,7 @@ static int hqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 
 static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
-			    uint32_t doorbell_off)
+			    uint32_t doorbell_off, uint32_t inst)
 {
 	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v10_compute_mqd *m;
@@ -325,7 +326,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd,
 
 static int hqd_dump_v10_3(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS 56
@@ -456,7 +457,7 @@ static int hqd_sdma_dump_v10_3(struct amdgpu_device *adev,
 
 static bool hqd_is_occupied_v10_3(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
@@ -498,7 +499,7 @@ static bool hqd_sdma_is_occupied_v10_3(struct amdgpu_device *adev,
 static int hqd_destroy_v10_3(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	enum hqd_dequeue_request_type type;
 	unsigned long end_jiffies;
@@ -586,7 +587,7 @@ static int hqd_sdma_destroy_v10_3(struct amdgpu_device *adev, void *mqd,
 
 static int wave_control_execute_v10_3(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data = 0;
 
@@ -628,7 +629,8 @@ static void set_vm_context_page_table_base_v10_3(struct amdgpu_device *adev,
 }
 
 static void program_trap_handler_settings_v10_3(struct amdgpu_device *adev,
-			uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr)
+			uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
+			uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -765,7 +767,7 @@ uint32_t set_wave_launch_mode_v10_3(struct amdgpu_device *adev,
  *	deq_retry_wait_time      -- Wait Count for Global Wave Syncs.
  */
 void get_iq_wait_times_v10_3(struct amdgpu_device *adev,
-					uint32_t *wait_times)
+					uint32_t *wait_times, uint32_t inst)
 
 {
 	*wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2));
@@ -775,7 +777,8 @@ void build_grace_period_packet_info_v10_3(struct amdgpu_device *adev,
 						uint32_t wait_times,
 						uint32_t grace_period,
 						uint32_t *reg_offset,
-						uint32_t *reg_data)
+						uint32_t *reg_data,
+						uint32_t inst)
 {
 	*reg_data = wait_times;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 5cdb7289d35b8..7deff8a547fb1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -78,7 +78,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -89,7 +89,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi
 }
 
 static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	uint32_t value = pasid << IH_VMID_0_LUT__PASID__SHIFT;
 
@@ -101,7 +101,8 @@ static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int p
 	return 0;
 }
 
-static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id)
+static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -162,7 +163,7 @@ static inline struct v11_sdma_mqd *get_sdma_mqd(void *mqd)
 static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id,
 			uint32_t queue_id, uint32_t __user *wptr,
 			uint32_t wptr_shift, uint32_t wptr_mask,
-			struct mm_struct *mm)
+			struct mm_struct *mm, uint32_t inst)
 {
 	struct v11_compute_mqd *m;
 	uint32_t *mqd_hqd;
@@ -258,7 +259,7 @@ static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id,
 
 static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd,
 			      uint32_t pipe_id, uint32_t queue_id,
-			      uint32_t doorbell_off)
+			      uint32_t doorbell_off, uint32_t inst)
 {
 	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
 	struct v11_compute_mqd *m;
@@ -310,7 +311,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd,
 
 static int hqd_dump_v11(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS 56
@@ -445,7 +446,7 @@ static int hqd_sdma_dump_v11(struct amdgpu_device *adev,
 }
 
 static bool hqd_is_occupied_v11(struct amdgpu_device *adev, uint64_t queue_address,
-				uint32_t pipe_id, uint32_t queue_id)
+				uint32_t pipe_id, uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
@@ -486,7 +487,7 @@ static bool hqd_sdma_is_occupied_v11(struct amdgpu_device *adev, void *mqd)
 static int hqd_destroy_v11(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	enum hqd_dequeue_request_type type;
 	unsigned long end_jiffies;
@@ -571,7 +572,7 @@ static int hqd_sdma_destroy_v11(struct amdgpu_device *adev, void *mqd,
 
 static int wave_control_execute_v11(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index e83cb1c09610c..6bf448ab3dffc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -78,7 +78,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -91,7 +91,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 }
 
 static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -114,7 +114,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
 	return 0;
 }
 
-static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id)
+static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -158,7 +159,7 @@ static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd)
 static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst)
 {
 	struct cik_mqd *m;
 	uint32_t *mqd_hqd;
@@ -202,7 +203,7 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 static int kgd_hqd_dump(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS (35+4)
@@ -318,7 +319,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev,
 
 static bool kgd_hqd_is_occupied(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
@@ -358,7 +359,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t temp;
 	enum hqd_dequeue_request_type type;
@@ -494,7 +495,7 @@ static int kgd_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
 
 static int kgd_wave_control_execute(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index 870f352837fc6..cd06e4a6d1da4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -72,7 +72,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
 	lock_srbm(adev, 0, 0, 0, vmid);
 
@@ -85,7 +85,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 }
 
 static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -109,7 +109,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
 	return 0;
 }
 
-static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id)
+static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -153,7 +154,7 @@ static inline struct vi_sdma_mqd *get_sdma_mqd(void *mqd)
 static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst)
 {
 	struct vi_mqd *m;
 	uint32_t *mqd_hqd;
@@ -226,7 +227,7 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 static int kgd_hqd_dump(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS (54+4)
@@ -350,7 +351,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev,
 
 static bool kgd_hqd_is_occupied(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
@@ -390,7 +391,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t temp;
 	enum hqd_dequeue_request_type type;
@@ -540,7 +541,7 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev,
 
 static int kgd_wave_control_execute(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index d36219ecd3dd0..7918a00cbb5a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -46,26 +46,26 @@ enum hqd_dequeue_request_type {
 	SAVE_WAVES
 };
 
-static void lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe,
-			uint32_t queue, uint32_t vmid)
+static void kgd_gfx_v9_lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe,
+			uint32_t queue, uint32_t vmid, uint32_t inst)
 {
 	mutex_lock(&adev->srbm_mutex);
-	soc15_grbm_select(adev, mec, pipe, queue, vmid, 0);
+	soc15_grbm_select(adev, mec, pipe, queue, vmid, inst);
 }
 
-static void unlock_srbm(struct amdgpu_device *adev)
+static void kgd_gfx_v9_unlock_srbm(struct amdgpu_device *adev, uint32_t inst)
 {
-	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, inst);
 	mutex_unlock(&adev->srbm_mutex);
 }
 
 void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	uint32_t pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
 
-	lock_srbm(adev, mec, pipe, queue_id, 0);
+	kgd_gfx_v9_lock_srbm(adev, mec, pipe, queue_id, 0, inst);
 }
 
 uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
@@ -77,28 +77,28 @@ uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
 	return 1ull << bit;
 }
 
-void kgd_gfx_v9_release_queue(struct amdgpu_device *adev)
+void kgd_gfx_v9_release_queue(struct amdgpu_device *adev, uint32_t inst)
 {
-	unlock_srbm(adev);
+	kgd_gfx_v9_unlock_srbm(adev, inst);
 }
 
 void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid,
 					uint32_t sh_mem_config,
 					uint32_t sh_mem_ape1_base,
 					uint32_t sh_mem_ape1_limit,
-					uint32_t sh_mem_bases)
+					uint32_t sh_mem_bases, uint32_t inst)
 {
-	lock_srbm(adev, 0, 0, 0, vmid);
+	kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst);
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmSH_MEM_CONFIG), sh_mem_config);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmSH_MEM_BASES), sh_mem_bases);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmSH_MEM_CONFIG), sh_mem_config);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmSH_MEM_BASES), sh_mem_bases);
 	/* APE1 no longer exists on GFX9 */
 
-	unlock_srbm(adev);
+	kgd_gfx_v9_unlock_srbm(adev, inst);
 }
 
 int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
-					unsigned int vmid)
+					unsigned int vmid, uint32_t inst)
 {
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -156,7 +156,8 @@ int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
  * but still works
  */
 
-int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id)
+int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst)
 {
 	uint32_t mec;
 	uint32_t pipe;
@@ -164,13 +165,13 @@ int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id)
 	mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
 
-	lock_srbm(adev, mec, pipe, 0, 0);
+	kgd_gfx_v9_lock_srbm(adev, mec, pipe, 0, 0, inst);
 
-	WREG32_SOC15(GC, 0, mmCPC_INT_CNTL,
+	WREG32_SOC15(GC, inst, mmCPC_INT_CNTL,
 		CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK |
 		CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK);
 
-	unlock_srbm(adev);
+	kgd_gfx_v9_unlock_srbm(adev, inst);
 
 	return 0;
 }
@@ -220,7 +221,8 @@ static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd)
 int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			uint32_t __user *wptr, uint32_t wptr_shift,
-			uint32_t wptr_mask, struct mm_struct *mm)
+			uint32_t wptr_mask, struct mm_struct *mm,
+			uint32_t inst)
 {
 	struct v9_mqd *m;
 	uint32_t *mqd_hqd;
@@ -228,21 +230,22 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	m = get_mqd(mqd);
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
 	/* HQD registers extend from CP_MQD_BASE_ADDR to CP_HQD_EOP_WPTR_MEM. */
 	mqd_hqd = &m->cp_mqd_base_addr_lo;
-	hqd_base = SOC15_REG_OFFSET(GC, 0, mmCP_MQD_BASE_ADDR);
+	hqd_base = SOC15_REG_OFFSET(GC, inst, mmCP_MQD_BASE_ADDR);
 
 	for (reg = hqd_base;
-	     reg <= SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI); reg++)
+	     reg <= SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI); reg++)
 		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
 
 
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_DOORBELL_CONTROL), data);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_DOORBELL_CONTROL),
+					data);
 
 	if (wptr) {
 		/* Don't read wptr with get_user because the user
@@ -271,43 +274,43 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_LO),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_LO),
 		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI),
 		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_POLL_ADDR),
 		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
 		       upper_32_bits((uintptr_t)wptr));
 		WREG32_SOC15(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1,
 		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_EOP_RPTR),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_EOP_RPTR),
 	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
 			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_ACTIVE), data);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_ACTIVE), data);
 
-	kgd_gfx_v9_release_queue(adev);
+	kgd_gfx_v9_release_queue(adev, inst);
 
 	return 0;
 }
 
 int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
-			    uint32_t doorbell_off)
+			    uint32_t doorbell_off, uint32_t inst)
 {
-	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring;
+	struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[inst].ring;
 	struct v9_mqd *m;
 	uint32_t mec, pipe;
 	int r;
 
 	m = get_mqd(mqd);
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
 	mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1;
 	pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec);
@@ -315,7 +318,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n",
 		 mec, pipe, queue_id);
 
-	spin_lock(&adev->gfx.kiq[0].ring_lock);
+	spin_lock(&adev->gfx.kiq[inst].ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, 7);
 	if (r) {
 		pr_err("Failed to alloc KIQ (%d).\n", r);
@@ -342,15 +345,15 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 	amdgpu_ring_commit(kiq_ring);
 
 out_unlock:
-	spin_unlock(&adev->gfx.kiq[0].ring_lock);
-	kgd_gfx_v9_release_queue(adev);
+	spin_unlock(&adev->gfx.kiq[inst].ring_lock);
+	kgd_gfx_v9_release_queue(adev, inst);
 
 	return r;
 }
 
 int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs)
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst)
 {
 	uint32_t i = 0, reg;
 #define HQD_N_REGS 56
@@ -365,13 +368,13 @@ int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev,
 	if (*dump == NULL)
 		return -ENOMEM;
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
-	for (reg = SOC15_REG_OFFSET(GC, 0, mmCP_MQD_BASE_ADDR);
-	     reg <= SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI); reg++)
+	for (reg = SOC15_REG_OFFSET(GC, inst, mmCP_MQD_BASE_ADDR);
+	     reg <= SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI); reg++)
 		DUMP_REG(reg);
 
-	kgd_gfx_v9_release_queue(adev);
+	kgd_gfx_v9_release_queue(adev, inst);
 
 	WARN_ON_ONCE(i != HQD_N_REGS);
 	*n_regs = i;
@@ -481,23 +484,23 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev,
 
 bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	uint32_t act;
 	bool retval = false;
 	uint32_t low, high;
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
-	act = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
+	act = RREG32_SOC15(GC, inst, mmCP_HQD_ACTIVE);
 	if (act) {
 		low = lower_32_bits(queue_address >> 8);
 		high = upper_32_bits(queue_address >> 8);
 
-		if (low == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE) &&
-		   high == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE_HI))
+		if (low == RREG32_SOC15(GC, inst, mmCP_HQD_PQ_BASE) &&
+		   high == RREG32_SOC15(GC, inst, mmCP_HQD_PQ_BASE_HI))
 			retval = true;
 	}
-	kgd_gfx_v9_release_queue(adev);
+	kgd_gfx_v9_release_queue(adev, inst);
 	return retval;
 }
 
@@ -522,7 +525,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id)
+				uint32_t queue_id, uint32_t inst)
 {
 	enum hqd_dequeue_request_type type;
 	unsigned long end_jiffies;
@@ -532,10 +535,10 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 	if (amdgpu_in_reset(adev))
 		return -EIO;
 
-	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id);
+	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
 	if (m->cp_hqd_vmid == 0)
-		WREG32_FIELD15_RLC(GC, 0, RLC_CP_SCHEDULERS, scheduler1, 0);
+		WREG32_FIELD15_RLC(GC, inst, RLC_CP_SCHEDULERS, scheduler1, 0);
 
 	switch (reset_type) {
 	case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN:
@@ -552,22 +555,22 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 		break;
 	}
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_DEQUEUE_REQUEST), type);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_DEQUEUE_REQUEST), type);
 
 	end_jiffies = (utimeout * HZ / 1000) + jiffies;
 	while (true) {
-		temp = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE);
+		temp = RREG32_SOC15(GC, inst, mmCP_HQD_ACTIVE);
 		if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK))
 			break;
 		if (time_after(jiffies, end_jiffies)) {
 			pr_err("cp queue preemption time out.\n");
-			kgd_gfx_v9_release_queue(adev);
+			kgd_gfx_v9_release_queue(adev, inst);
 			return -ETIME;
 		}
 		usleep_range(500, 1000);
 	}
 
-	kgd_gfx_v9_release_queue(adev);
+	kgd_gfx_v9_release_queue(adev, inst);
 	return 0;
 }
 
@@ -624,14 +627,14 @@ bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev,
 
 int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd)
+					uint32_t sq_cmd, uint32_t inst)
 {
 	uint32_t data = 0;
 
 	mutex_lock(&adev->grbm_idx_mutex);
 
-	WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_INDEX, gfx_index_val);
-	WREG32_SOC15(GC, 0, mmSQ_CMD, sq_cmd);
+	WREG32_SOC15_RLC_SHADOW(GC, inst, mmGRBM_GFX_INDEX, gfx_index_val);
+	WREG32_SOC15(GC, inst, mmSQ_CMD, sq_cmd);
 
 	data = REG_SET_FIELD(data, GRBM_GFX_INDEX,
 		INSTANCE_BROADCAST_WRITES, 1);
@@ -640,7 +643,7 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 	data = REG_SET_FIELD(data, GRBM_GFX_INDEX,
 		SE_BROADCAST_WRITES, 1);
 
-	WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_INDEX, data);
+	WREG32_SOC15_RLC_SHADOW(GC, inst, mmGRBM_GFX_INDEX, data);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	return 0;
@@ -685,7 +688,7 @@ static void unlock_spi_csq_mutexes(struct amdgpu_device *adev)
  * is being collected
  */
 static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
-		int *wave_cnt, int *vmid)
+		int *wave_cnt, int *vmid, uint32_t inst)
 {
 	int pipe_idx;
 	int queue_slot;
@@ -700,12 +703,12 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
 	*wave_cnt = 0;
 	pipe_idx = queue_idx / adev->gfx.mec.num_queue_per_pipe;
 	queue_slot = queue_idx % adev->gfx.mec.num_queue_per_pipe;
-	soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0, 0);
-	reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) +
+	soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0, inst);
+	reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, inst, mmSPI_CSQ_WF_ACTIVE_COUNT_0) +
 			 queue_slot);
 	*wave_cnt = reg_val & SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK;
 	if (*wave_cnt != 0)
-		*vmid = (RREG32_SOC15(GC, 0, mmCP_HQD_VMID) &
+		*vmid = (RREG32_SOC15(GC, inst, mmCP_HQD_VMID) &
 			 CP_HQD_VMID__VMID_MASK) >> CP_HQD_VMID__VMID__SHIFT;
 }
 
@@ -756,7 +759,7 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
  *  Reading registers referenced above involves programming GRBM appropriately
  */
 void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
-		int *pasid_wave_cnt, int *max_waves_per_cu)
+		int *pasid_wave_cnt, int *max_waves_per_cu, uint32_t inst)
 {
 	int qidx;
 	int vmid;
@@ -772,7 +775,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	DECLARE_BITMAP(cp_queue_bitmap, KGD_MAX_QUEUES);
 
 	lock_spi_csq_mutexes(adev);
-	soc15_grbm_select(adev, 1, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 1, 0, 0, 0, inst);
 
 	/*
 	 * Iterate through the shader engines and arrays of the device
@@ -787,8 +790,8 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	for (se_idx = 0; se_idx < se_cnt; se_idx++) {
 		for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) {
 
-			amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff, 0);
-			queue_map = RREG32_SOC15(GC, 0, mmSPI_CSQ_WF_ACTIVE_STATUS);
+			amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff, inst);
+			queue_map = RREG32_SOC15(GC, inst, mmSPI_CSQ_WF_ACTIVE_STATUS);
 
 			/*
 			 * Assumption: queue map encodes following schema: four
@@ -808,10 +811,11 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 					continue;
 
 				/* Get number of waves in flight and aggregate them */
-				get_wave_count(adev, qidx, &wave_cnt, &vmid);
+				get_wave_count(adev, qidx, &wave_cnt, &vmid,
+						inst);
 				if (wave_cnt != 0) {
 					pasid_tmp =
-					  RREG32(SOC15_REG_OFFSET(OSSSYS, 0,
+					  RREG32(SOC15_REG_OFFSET(OSSSYS, inst,
 						 mmIH_VMID_0_LUT) + vmid);
 					if (pasid_tmp == pasid)
 						vmid_wave_cnt += wave_cnt;
@@ -820,8 +824,8 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 		}
 	}
 
-	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
-	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, inst);
+	soc15_grbm_select(adev, 0, 0, 0, 0, inst);
 	unlock_spi_csq_mutexes(adev);
 
 	/* Update the output parameters and return */
@@ -831,27 +835,27 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 }
 
 void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
-                        uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr)
+		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, uint32_t inst)
 {
-	lock_srbm(adev, 0, 0, 0, vmid);
+	kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst);
 
 	/*
 	 * Program TBA registers
 	 */
-	WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_LO,
+	WREG32_SOC15(GC, inst, mmSQ_SHADER_TBA_LO,
                         lower_32_bits(tba_addr >> 8));
-	WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_HI,
+	WREG32_SOC15(GC, inst, mmSQ_SHADER_TBA_HI,
                         upper_32_bits(tba_addr >> 8));
 
 	/*
 	 * Program TMA registers
 	 */
-	WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_LO,
+	WREG32_SOC15(GC, inst, mmSQ_SHADER_TMA_LO,
 			lower_32_bits(tma_addr >> 8));
-	WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_HI,
+	WREG32_SOC15(GC, inst, mmSQ_SHADER_TMA_HI,
 			upper_32_bits(tma_addr >> 8));
 
-	unlock_srbm(adev);
+	kgd_gfx_v9_unlock_srbm(adev, inst);
 }
 
 const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index 491273a02e309..a241299f4fbcc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -25,41 +25,42 @@
 void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid,
 		uint32_t sh_mem_config,
 		uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit,
-		uint32_t sh_mem_bases);
+		uint32_t sh_mem_bases, uint32_t inst);
 int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid,
-		unsigned int vmid);
-int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id);
+		unsigned int vmid, uint32_t inst);
+int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
+				uint32_t inst);
 int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id,
 			uint32_t queue_id, uint32_t __user *wptr,
 			uint32_t wptr_shift, uint32_t wptr_mask,
-			struct mm_struct *mm);
+			struct mm_struct *mm, uint32_t inst);
 int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
-			    uint32_t doorbell_off);
+			    uint32_t doorbell_off, uint32_t inst);
 int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs);
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst);
 bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev,
 			uint64_t queue_address, uint32_t pipe_id,
-			uint32_t queue_id);
+			uint32_t queue_id, uint32_t inst);
 int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int utimeout, uint32_t pipe_id,
-				uint32_t queue_id);
+				uint32_t queue_id, uint32_t inst);
 int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd);
+					uint32_t sq_cmd, uint32_t inst);
 bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev,
 					uint8_t vmid, uint16_t *p_pasid);
-
 void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev,
 			uint32_t vmid, uint64_t page_table_base);
 void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
-		int *pasid_wave_cnt, int *max_waves_per_cu);
+		int *pasid_wave_cnt, int *max_waves_per_cu, uint32_t inst);
 void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
-		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr);
+		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
+		uint32_t inst);
 void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
-				uint32_t queue_id);
+				uint32_t queue_id, uint32_t inst);
 uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
 				uint32_t pipe_id, uint32_t queue_id);
-void kgd_gfx_v9_release_queue(struct amdgpu_device *adev);
+void kgd_gfx_v9_release_queue(struct amdgpu_device *adev, uint32_t inst);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 9afd3295ca850..33a9394f9e58c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -127,12 +127,16 @@ static inline uint64_t get_reserved_sdma_queues_bitmap(struct device_queue_manag
 void program_sh_mem_settings(struct device_queue_manager *dqm,
 					struct qcm_process_device *qpd)
 {
-	return dqm->dev->kfd2kgd->program_sh_mem_settings(
+	int xcc = 0;
+
+	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+		dqm->dev->kfd2kgd->program_sh_mem_settings(
 						dqm->dev->adev, qpd->vmid,
 						qpd->sh_mem_config,
 						qpd->sh_mem_ape1_base,
 						qpd->sh_mem_ape1_limit,
-						qpd->sh_mem_bases);
+						qpd->sh_mem_bases,
+						dqm->dev->start_xcc_id + xcc);
 }
 
 static void kfd_hws_hang(struct device_queue_manager *dqm)
@@ -405,10 +409,14 @@ static void deallocate_doorbell(struct qcm_process_device *qpd,
 static void program_trap_handler_settings(struct device_queue_manager *dqm,
 				struct qcm_process_device *qpd)
 {
+	int xcc = 0;
+
 	if (dqm->dev->kfd2kgd->program_trap_handler_settings)
-		dqm->dev->kfd2kgd->program_trap_handler_settings(
+		for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+			dqm->dev->kfd2kgd->program_trap_handler_settings(
 						dqm->dev->adev, qpd->vmid,
-						qpd->tba_addr, qpd->tma_addr);
+						qpd->tba_addr, qpd->tma_addr,
+						dqm->dev->start_xcc_id + xcc);
 }
 
 static int allocate_vmid(struct device_queue_manager *dqm,
@@ -671,6 +679,7 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process
 	struct kfd_process_device *pdd;
 	int first_vmid_to_scan = dev->vm_info.first_vmid_kfd;
 	int last_vmid_to_scan = dev->vm_info.last_vmid_kfd;
+	int xcc = 0;
 
 	reg_sq_cmd.u32All = 0;
 	reg_gfx_index.u32All = 0;
@@ -715,9 +724,11 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process
 	reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_KILL;
 	reg_sq_cmd.bits.vm_id = vmid;
 
-	dev->kfd2kgd->wave_control_execute(dev->adev,
+	for (xcc = 0; xcc < dev->num_xcc_per_node; xcc++)
+		dev->kfd2kgd->wave_control_execute(dev->adev,
 					reg_gfx_index.u32All,
-					reg_sq_cmd.u32All);
+					reg_sq_cmd.u32All,
+					dev->start_xcc_id + xcc);
 
 	return 0;
 }
@@ -1229,17 +1240,32 @@ static int
 set_pasid_vmid_mapping(struct device_queue_manager *dqm, u32 pasid,
 			unsigned int vmid)
 {
-	return dqm->dev->kfd2kgd->set_pasid_vmid_mapping(
-						dqm->dev->adev, pasid, vmid);
+	int xcc = 0, ret;
+
+	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++) {
+		ret = dqm->dev->kfd2kgd->set_pasid_vmid_mapping(
+						dqm->dev->adev, pasid, vmid,
+						dqm->dev->start_xcc_id + xcc);
+		if (ret)
+			break;
+	}
+
+	return ret;
 }
 
 static void init_interrupts(struct device_queue_manager *dqm)
 {
-	unsigned int i;
+	unsigned int i, xcc;
 
-	for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++)
-		if (is_pipe_enabled(dqm, 0, i))
-			dqm->dev->kfd2kgd->init_interrupts(dqm->dev->adev, i);
+	for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) {
+		if (is_pipe_enabled(dqm, 0, i)) {
+			for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+				dqm->dev->kfd2kgd->init_interrupts(
+							dqm->dev->adev, i,
+							dqm->dev->start_xcc_id +
+							xcc);
+		}
+	}
 }
 
 static void init_sdma_bitmaps(struct device_queue_manager *dqm)
@@ -2455,44 +2481,49 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 	struct device_queue_manager *dqm = data;
 	uint32_t (*dump)[2], n_regs;
 	int pipe, queue;
-	int r = 0;
+	int r = 0, xcc;
+	uint32_t inst;
 
 	if (!dqm->sched_running) {
 		seq_puts(m, " Device is stopped\n");
 		return 0;
 	}
 
-	r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev,
+	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++) {
+		inst = dqm->dev->start_xcc_id + xcc;
+		r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev,
 					KFD_CIK_HIQ_PIPE, KFD_CIK_HIQ_QUEUE,
-					&dump, &n_regs);
-	if (!r) {
-		seq_printf(m, "  HIQ on MEC %d Pipe %d Queue %d\n",
-			   KFD_CIK_HIQ_PIPE/get_pipes_per_mec(dqm)+1,
-			   KFD_CIK_HIQ_PIPE%get_pipes_per_mec(dqm),
-			   KFD_CIK_HIQ_QUEUE);
-		seq_reg_dump(m, dump, n_regs);
+					&dump, &n_regs, inst);
+		if (!r) {
+			seq_printf(m,
+				"   Inst %d, HIQ on MEC %d Pipe %d Queue %d\n",
+				inst, KFD_CIK_HIQ_PIPE/get_pipes_per_mec(dqm)+1,
+				KFD_CIK_HIQ_PIPE%get_pipes_per_mec(dqm),
+				KFD_CIK_HIQ_QUEUE);
+			seq_reg_dump(m, dump, n_regs);
 
-		kfree(dump);
-	}
+			kfree(dump);
+		}
 
-	for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) {
-		int pipe_offset = pipe * get_queues_per_pipe(dqm);
+		for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) {
+			int pipe_offset = pipe * get_queues_per_pipe(dqm);
 
-		for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) {
-			if (!test_bit(pipe_offset + queue,
+			for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) {
+				if (!test_bit(pipe_offset + queue,
 				      dqm->dev->kfd->shared_resources.cp_queue_bitmap))
-				continue;
+					continue;
 
-			r = dqm->dev->kfd2kgd->hqd_dump(
-				dqm->dev->adev, pipe, queue, &dump, &n_regs);
-			if (r)
-				break;
+				r = dqm->dev->kfd2kgd->hqd_dump(
+					dqm->dev->adev, pipe, queue, &dump, &n_regs, inst);
+				if (r)
+					break;
 
-			seq_printf(m, "  CP Pipe %d, Queue %d\n",
-				  pipe, queue);
-			seq_reg_dump(m, dump, n_regs);
+				seq_printf(m, " Inst %d,  CP Pipe %d, Queue %d\n",
+					  inst, pipe, queue);
+				seq_reg_dump(m, dump, n_regs);
 
-			kfree(dump);
+				kfree(dump);
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
index 074f6075ccc71..d81125421aaff 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
@@ -190,7 +190,7 @@ int kfd_hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd,
 		     struct queue_properties *p, struct mm_struct *mms)
 {
 	return mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, mqd, pipe_id,
-					      queue_id, p->doorbell_off);
+					      queue_id, p->doorbell_off, 0);
 }
 
 int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd,
@@ -198,7 +198,7 @@ int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd,
 		uint32_t pipe_id, uint32_t queue_id)
 {
 	return mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, mqd, type, timeout,
-						pipe_id, queue_id);
+						pipe_id, queue_id, 0);
 }
 
 void kfd_free_mqd_cp(struct mqd_manager *mm, void *mqd,
@@ -217,7 +217,7 @@ bool kfd_is_occupied_cp(struct mqd_manager *mm, void *mqd,
 		 uint32_t queue_id)
 {
 	return mm->dev->kfd2kgd->hqd_is_occupied(mm->dev->adev, queue_address,
-						pipe_id, queue_id);
+						pipe_id, queue_id, 0);
 }
 
 int kfd_load_mqd_sdma(struct mqd_manager *mm, void *mqd,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
index ca19664667590..eb11940bec34c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
@@ -167,7 +167,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, uint32_t pipe_id,
 
 	return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id,
 					  (uint32_t __user *)p->write_ptr,
-					  wptr_shift, wptr_mask, mms);
+					  wptr_shift, wptr_mask, mms, 0);
 }
 
 static void __update_mqd(struct mqd_manager *mm, void *mqd,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index c9565ea99df53..d54c6fdebbb6b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -151,7 +151,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd,
 
 	r = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id,
 					  (uint32_t __user *)p->write_ptr,
-					  wptr_shift, 0, mms);
+					  wptr_shift, 0, mms, 0);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index dff171b54b5c8..338d639c18980 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -184,7 +184,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd,
 
 	r = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id,
 					  (uint32_t __user *)p->write_ptr,
-					  wptr_shift, 0, mms);
+					  wptr_shift, 0, mms, 0);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index b46c984b3a179..b53cd8f9620bb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -218,7 +218,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd,
 
 	return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id,
 					  (uint32_t __user *)p->write_ptr,
-					  wptr_shift, 0, mms);
+					  wptr_shift, 0, mms, 0);
 }
 
 static void update_mqd(struct mqd_manager *mm, void *mqd,
@@ -501,13 +501,15 @@ static int hiq_load_mqd_kiq_v9_4_3(struct mqd_manager *mm, void *mqd,
 {
 	int xcc, err;
 	void *xcc_mqd;
+	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
 
 	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
 		xcc_mqd = mqd + hiq_mqd_size * xcc;
 		err = mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, xcc_mqd,
 						     pipe_id, queue_id,
-						     p->doorbell_off);
+						     p->doorbell_off,
+						     start_inst+xcc);
 		if (err) {
 			pr_debug("Failed to load HIQ MQD for XCC: %d\n", xcc);
 			break;
@@ -523,13 +525,14 @@ static int destroy_hiq_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 {
 	int xcc = 0, err;
 	void *xcc_mqd;
+	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
 
 	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
 		xcc_mqd = mqd + hiq_mqd_size * xcc;
 		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
 						    type, timeout, pipe_id,
-						    queue_id);
+						    queue_id, start_inst+xcc);
 		if (err) {
 			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
 			break;
@@ -641,6 +644,7 @@ static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 	void *xcc_mqd;
 	struct v9_mqd *m;
 	uint64_t mqd_offset;
+	uint32_t start_inst = mm->dev->start_xcc_id;
 
 	m = get_mqd(mqd);
 	mqd_offset = m->cp_mqd_stride_size;
@@ -649,7 +653,7 @@ static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 		xcc_mqd = mqd + mqd_offset * xcc;
 		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
 						    type, timeout, pipe_id,
-						    queue_id);
+						    queue_id, start_inst+xcc);
 		if (err) {
 			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
 			break;
@@ -667,6 +671,7 @@ static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 	uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0);
 	int xcc = 0, err;
 	void *xcc_mqd;
+	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t mqd_stride_size = mm->mqd_stride(mm, p);
 
 	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
@@ -674,7 +679,7 @@ static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 		err = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, xcc_mqd,
 					 pipe_id, queue_id,
 					(uint32_t __user *)p->write_ptr,
-					wptr_shift, 0, mms);
+					wptr_shift, 0, mms, start_inst+xcc);
 		if (err) {
 			pr_debug("Load MQD failed for xcc: %d\n", xcc);
 			break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index 8736a3cdbe1e8..ebf963f42b51f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -165,7 +165,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd,
 
 	return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id,
 					  (uint32_t __user *)p->write_ptr,
-					  wptr_shift, wptr_mask, mms);
+					  wptr_shift, wptr_mask, mms, 0);
 }
 
 static void __update_mqd(struct mqd_manager *mm, void *mqd,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 66e021889c64b..888590dfa646f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -290,7 +290,7 @@ static int kfd_get_cu_occupancy(struct attribute *attr, char *buffer)
 	wave_cnt = 0;
 	max_waves_per_cu = 0;
 	dev->kfd2kgd->get_cu_occupancy(dev->adev, proc->pasid, &wave_cnt,
-			&max_waves_per_cu);
+			&max_waves_per_cu, 0);
 
 	/* Translate wave count to number of compute units */
 	cu_cnt = (wave_cnt + (max_waves_per_cu - 1)) / max_waves_per_cu;
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 5cb3e8634739d..8cb3dbcae3e47 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -230,28 +230,30 @@ struct kfd2kgd_calls {
 	/* Register access functions */
 	void (*program_sh_mem_settings)(struct amdgpu_device *adev, uint32_t vmid,
 			uint32_t sh_mem_config,	uint32_t sh_mem_ape1_base,
-			uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases);
+			uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases,
+			uint32_t inst);
 
 	int (*set_pasid_vmid_mapping)(struct amdgpu_device *adev, u32 pasid,
-					unsigned int vmid);
+					unsigned int vmid, uint32_t inst);
 
-	int (*init_interrupts)(struct amdgpu_device *adev, uint32_t pipe_id);
+	int (*init_interrupts)(struct amdgpu_device *adev, uint32_t pipe_id,
+			uint32_t inst);
 
 	int (*hqd_load)(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id,
 			uint32_t queue_id, uint32_t __user *wptr,
 			uint32_t wptr_shift, uint32_t wptr_mask,
-			struct mm_struct *mm);
+			struct mm_struct *mm, uint32_t inst);
 
 	int (*hiq_mqd_load)(struct amdgpu_device *adev, void *mqd,
 			    uint32_t pipe_id, uint32_t queue_id,
-			    uint32_t doorbell_off);
+			    uint32_t doorbell_off, uint32_t inst);
 
 	int (*hqd_sdma_load)(struct amdgpu_device *adev, void *mqd,
 			     uint32_t __user *wptr, struct mm_struct *mm);
 
 	int (*hqd_dump)(struct amdgpu_device *adev,
 			uint32_t pipe_id, uint32_t queue_id,
-			uint32_t (**dump)[2], uint32_t *n_regs);
+			uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst);
 
 	int (*hqd_sdma_dump)(struct amdgpu_device *adev,
 			     uint32_t engine_id, uint32_t queue_id,
@@ -259,12 +261,12 @@ struct kfd2kgd_calls {
 
 	bool (*hqd_is_occupied)(struct amdgpu_device *adev,
 				uint64_t queue_address, uint32_t pipe_id,
-				uint32_t queue_id);
+				uint32_t queue_id, uint32_t inst);
 
 	int (*hqd_destroy)(struct amdgpu_device *adev, void *mqd,
 				enum kfd_preempt_type reset_type,
 				unsigned int timeout, uint32_t pipe_id,
-				uint32_t queue_id);
+				uint32_t queue_id, uint32_t inst);
 
 	bool (*hqd_sdma_is_occupied)(struct amdgpu_device *adev, void *mqd);
 
@@ -273,7 +275,7 @@ struct kfd2kgd_calls {
 
 	int (*wave_control_execute)(struct amdgpu_device *adev,
 					uint32_t gfx_index_val,
-					uint32_t sq_cmd);
+					uint32_t sq_cmd, uint32_t inst);
 	bool (*get_atc_vmid_pasid_mapping_info)(struct amdgpu_device *adev,
 					uint8_t vmid,
 					uint16_t *p_pasid);
@@ -290,9 +292,10 @@ struct kfd2kgd_calls {
 	uint32_t (*read_vmid_from_vmfault_reg)(struct amdgpu_device *adev);
 
 	void (*get_cu_occupancy)(struct amdgpu_device *adev, int pasid,
-			int *wave_cnt, int *max_waves_per_cu);
+			int *wave_cnt, int *max_waves_per_cu, uint32_t inst);
 	void (*program_trap_handler_settings)(struct amdgpu_device *adev,
-			uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr);
+			uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
+			uint32_t inst);
 };
 
 #endif	/* KGD_KFD_INTERFACE_H_INCLUDED */
-- 
GitLab


From f87f686482c6d2d4465245356854710b01f312c1 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:22:20 -0400
Subject: [PATCH 1050/1791] drm/amdgpu: Add XCC inst to PASID TLB flushing

Add XCC instance to select the correct KIQ ring when
flushing TLBs on a multi-XCC setup.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c |  6 ++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h    |  7 ++++---
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c      |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c      | 12 ++++++------
 drivers/gpu/drm/amd/amdkfd/kfd_process.c   |  7 +++++--
 9 files changed, 25 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 5afbcc390d89b..9d19c7ceda3f6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -743,7 +743,9 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev,
 }
 
 int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
-				      uint16_t pasid, enum TLB_FLUSH_TYPE flush_type)
+				      uint16_t pasid,
+				      enum TLB_FLUSH_TYPE flush_type,
+				      uint32_t inst)
 {
 	bool all_hub = false;
 
@@ -751,7 +753,7 @@ int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	    adev->family == AMDGPU_FAMILY_RV)
 		all_hub = true;
 
-	return amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, flush_type, all_hub);
+	return amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, flush_type, all_hub, inst);
 }
 
 bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 01ba3589b60a0..df07e212c21e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -160,7 +160,8 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev);
 int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev,
 				uint16_t vmid);
 int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
-				uint16_t pasid, enum TLB_FLUSH_TYPE flush_type);
+				uint16_t pasid, enum TLB_FLUSH_TYPE flush_type,
+				uint32_t inst);
 
 bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 6d105d7fb98bb..572ef5be539fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -119,7 +119,8 @@ struct amdgpu_gmc_funcs {
 				uint32_t vmhub, uint32_t flush_type);
 	/* flush the vm tlb via pasid */
 	int (*flush_gpu_tlb_pasid)(struct amdgpu_device *adev, uint16_t pasid,
-					uint32_t flush_type, bool all_hub);
+					uint32_t flush_type, bool all_hub,
+					uint32_t inst);
 	/* flush the vm tlb via ring */
 	uint64_t (*emit_flush_gpu_tlb)(struct amdgpu_ring *ring, unsigned vmid,
 				       uint64_t pd_addr);
@@ -296,9 +297,9 @@ struct amdgpu_gmc {
 };
 
 #define amdgpu_gmc_flush_gpu_tlb(adev, vmid, vmhub, type) ((adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (vmhub), (type)))
-#define amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, type, allhub) \
+#define amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, type, allhub, inst) \
 	((adev)->gmc.gmc_funcs->flush_gpu_tlb_pasid \
-	((adev), (pasid), (type), (allhub)))
+	((adev), (pasid), (type), (allhub), (inst)))
 #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr))
 #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid))
 #define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags))
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index ff96f11c2adf0..d76f5c8d4977e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -419,7 +419,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  */
 static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 					uint16_t pasid, uint32_t flush_type,
-					bool all_hub)
+					bool all_hub, uint32_t inst)
 {
 	int vmid, i;
 	signed long r;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 3453f1c0e0660..4bf807d825c03 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -324,7 +324,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  */
 static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 					uint16_t pasid, uint32_t flush_type,
-					bool all_hub)
+					bool all_hub, uint32_t inst)
 {
 	int vmid, i;
 	signed long r;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 81609a2b226ff..6f53049619cdd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -424,7 +424,7 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev)
  */
 static int gmc_v7_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 					uint16_t pasid, uint32_t flush_type,
-					bool all_hub)
+					bool all_hub, uint32_t inst)
 {
 	int vmid;
 	unsigned int tmp;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index d48e33738a882..48475077ca92f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -622,7 +622,7 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev)
  */
 static int gmc_v8_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 					uint16_t pasid, uint32_t flush_type,
-					bool all_hub)
+					bool all_hub, uint32_t inst)
 {
 	int vmid;
 	unsigned int tmp;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 6da85365e5aad..0163a761ccf06 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -924,7 +924,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  */
 static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 					uint16_t pasid, uint32_t flush_type,
-					bool all_hub)
+					bool all_hub, uint32_t inst)
 {
 	int vmid, i;
 	signed long r;
@@ -932,8 +932,8 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint16_t queried_pasid;
 	bool ret;
 	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout;
-	struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring;
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst];
 
 	if (amdgpu_in_reset(adev))
 		return -EIO;
@@ -953,7 +953,7 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		if (vega20_xgmi_wa)
 			ndw += kiq->pmf->invalidate_tlbs_size;
 
-		spin_lock(&adev->gfx.kiq[0].ring_lock);
+		spin_lock(&adev->gfx.kiq[inst].ring_lock);
 		/* 2 dwords flush + 8 dwords fence */
 		amdgpu_ring_alloc(ring, ndw);
 		if (vega20_xgmi_wa)
@@ -964,13 +964,13 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
 		if (r) {
 			amdgpu_ring_undo(ring);
-			spin_unlock(&adev->gfx.kiq[0].ring_lock);
+			spin_unlock(&adev->gfx.kiq[inst].ring_lock);
 			up_read(&adev->reset_domain->sem);
 			return -ETIME;
 		}
 
 		amdgpu_ring_commit(ring);
-		spin_unlock(&adev->gfx.kiq[0].ring_lock);
+		spin_unlock(&adev->gfx.kiq[inst].ring_lock);
 		r = amdgpu_fence_wait_polling(ring, seq, usec_timeout);
 		if (r < 1) {
 			dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 888590dfa646f..9b1e84d33cdc7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -2052,6 +2052,7 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 	struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv);
 	uint64_t tlb_seq = amdgpu_vm_tlb_seq(vm);
 	struct kfd_node *dev = pdd->dev;
+	int xcc = 0;
 
 	/*
 	 * It can be that we race and lose here, but that is extremely unlikely
@@ -2069,8 +2070,10 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 			amdgpu_amdkfd_flush_gpu_tlb_vmid(dev->adev,
 							pdd->qpd.vmid);
 	} else {
-		amdgpu_amdkfd_flush_gpu_tlb_pasid(dev->adev,
-					pdd->process->pasid, type);
+		for (xcc = 0; xcc < dev->num_xcc_per_node; xcc++)
+			amdgpu_amdkfd_flush_gpu_tlb_pasid(dev->adev,
+					pdd->process->pasid, type,
+					dev->start_xcc_id + xcc);
 	}
 }
 
-- 
GitLab


From 7fe51e6fd2368b358441a1f6e0c94f4cd7e0720f Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:25:25 -0400
Subject: [PATCH 1051/1791] drm/amdkfd: Update context save handling for multi
 XCC setup (v2)

Context save handling needs to be updated for a multi XCC
setup:
- On a multi XCC setup, KFD needs to report context save base
  address and size for each XCC in MQD.
- Thunk will allocate a large context save area covering all
  XCCs which will be equal to: num_of_xccs in a partition * size
  of context save area for 1 XCC. However, it will report only the
  size of context save area for 1 XCC only in the ioctl call.
- Driver then setups the MQD correctly using the size passed from
  Thunk and information about number of XCCs in a partition.
- Update get_wave_state function to return context save area
  for all XCCs in the partition.

v2: update the get_wave_state function for mqd manager v11 (Morris)

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Morris Zhang <Shiwu.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h  |  1 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |  1 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  |  1 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   | 62 ++++++++++++++++++-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c   |  1 +
 6 files changed, 67 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 33a9394f9e58c..01ea038cab6d7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2095,8 +2095,8 @@ static int get_wave_state(struct device_queue_manager *dqm,
 	 * and the queue should be protected against destruction by the process
 	 * lock.
 	 */
-	return mqd_mgr->get_wave_state(mqd_mgr, q->mqd, ctl_stack,
-			ctl_stack_used_size, save_area_used_size);
+	return mqd_mgr->get_wave_state(mqd_mgr, q->mqd, &q->properties,
+			ctl_stack, ctl_stack_used_size, save_area_used_size);
 }
 
 static void get_queue_checkpoint_info(struct device_queue_manager *dqm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
index eb18be74f5599..23158db7da035 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h
@@ -97,6 +97,7 @@ struct mqd_manager {
 				uint32_t queue_id);
 
 	int	(*get_wave_state)(struct mqd_manager *mm, void *mqd,
+				  struct queue_properties *q,
 				  void __user *ctl_stack,
 				  u32 *ctl_stack_used_size,
 				  u32 *save_area_used_size);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index d54c6fdebbb6b..772c09b5821b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -227,6 +227,7 @@ static uint32_t read_doorbell_id(void *mqd)
 }
 
 static int get_wave_state(struct mqd_manager *mm, void *mqd,
+			  struct queue_properties *q,
 			  void __user *ctl_stack,
 			  u32 *ctl_stack_used_size,
 			  u32 *save_area_used_size)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index 338d639c18980..632344b95d905 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -260,6 +260,7 @@ static uint32_t read_doorbell_id(void *mqd)
 }
 
 static int get_wave_state(struct mqd_manager *mm, void *mqd,
+			  struct queue_properties *q,
 			  void __user *ctl_stack,
 			  u32 *ctl_stack_used_size,
 			  u32 *save_area_used_size)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index b53cd8f9620bb..5c9b3392758e3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -295,6 +295,7 @@ static uint32_t read_doorbell_id(void *mqd)
 }
 
 static int get_wave_state(struct mqd_manager *mm, void *mqd,
+			  struct queue_properties *q,
 			  void __user *ctl_stack,
 			  u32 *ctl_stack_used_size,
 			  u32 *save_area_used_size)
@@ -561,6 +562,7 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 	int xcc = 0;
 	struct kfd_mem_obj xcc_mqd_mem_obj;
 	uint64_t xcc_gart_addr = 0;
+	uint64_t xcc_ctx_save_restore_area_address;
 	uint64_t offset = mm->mqd_stride(mm, q);
 
 	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
@@ -570,6 +572,23 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 		init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q);
 
 		m->cp_mqd_stride_size = offset;
+
+		/*
+		 * Update the CWSR address for each XCC if CWSR is enabled
+		 * and CWSR area is allocated in thunk
+		 */
+		if (mm->dev->kfd->cwsr_enabled &&
+		    q->ctx_save_restore_area_address) {
+			xcc_ctx_save_restore_area_address =
+				q->ctx_save_restore_area_address +
+				(xcc * q->ctx_save_restore_area_size);
+
+			m->cp_hqd_ctx_save_base_addr_lo =
+				lower_32_bits(xcc_ctx_save_restore_area_address);
+			m->cp_hqd_ctx_save_base_addr_hi =
+				upper_32_bits(xcc_ctx_save_restore_area_address);
+		}
+
 		if (q->format == KFD_QUEUE_FORMAT_AQL) {
 			m->compute_tg_chunk_size = 1;
 
@@ -689,6 +708,46 @@ static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 	return err;
 }
 
+static int get_wave_state_v9_4_3(struct mqd_manager *mm, void *mqd,
+				 struct queue_properties *q,
+				 void __user *ctl_stack,
+				 u32 *ctl_stack_used_size,
+				 u32 *save_area_used_size)
+{
+	int xcc, err = 0;
+	void *xcc_mqd;
+	void __user *xcc_ctl_stack;
+	uint64_t mqd_stride_size = mm->mqd_stride(mm, q);
+	u32 tmp_ctl_stack_used_size = 0, tmp_save_area_used_size = 0;
+
+	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+		xcc_mqd = mqd + mqd_stride_size * xcc;
+		xcc_ctl_stack = (void __user *)((uintptr_t)ctl_stack +
+					q->ctx_save_restore_area_size * xcc);
+
+		err = get_wave_state(mm, xcc_mqd, q, xcc_ctl_stack,
+				     &tmp_ctl_stack_used_size,
+				     &tmp_save_area_used_size);
+		if (err)
+			break;
+
+		/*
+		 * Set the ctl_stack_used_size and save_area_used_size to
+		 * ctl_stack_used_size and save_area_used_size of XCC 0 when
+		 * passing the info the user-space.
+		 * For multi XCC, user-space would have to look at the header
+		 * info of each Control stack area to determine the control
+		 * stack size and save area used.
+		 */
+		if (xcc == 0) {
+			*ctl_stack_used_size = tmp_ctl_stack_used_size;
+			*save_area_used_size = tmp_save_area_used_size;
+		}
+	}
+
+	return err;
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 static int debugfs_show_mqd(struct seq_file *m, void *data)
@@ -726,7 +785,6 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 		mqd->allocate_mqd = allocate_mqd;
 		mqd->free_mqd = kfd_free_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
-		mqd->get_wave_state = get_wave_state;
 		mqd->get_checkpoint_info = get_checkpoint_info;
 		mqd->checkpoint_mqd = checkpoint_mqd;
 		mqd->restore_mqd = restore_mqd;
@@ -740,11 +798,13 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 			mqd->load_mqd = load_mqd_v9_4_3;
 			mqd->update_mqd = update_mqd_v9_4_3;
 			mqd->destroy_mqd = destroy_mqd_v9_4_3;
+			mqd->get_wave_state = get_wave_state_v9_4_3;
 		} else {
 			mqd->init_mqd = init_mqd;
 			mqd->load_mqd = load_mqd;
 			mqd->update_mqd = update_mqd;
 			mqd->destroy_mqd = kfd_destroy_mqd_cp;
+			mqd->get_wave_state = get_wave_state;
 		}
 		break;
 	case KFD_MQD_TYPE_HIQ:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index ebf963f42b51f..fe69492b1bb39 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -261,6 +261,7 @@ static void update_mqd_tonga(struct mqd_manager *mm, void *mqd,
 }
 
 static int get_wave_state(struct mqd_manager *mm, void *mqd,
+			  struct queue_properties *q,
 			  void __user *ctl_stack,
 			  u32 *ctl_stack_used_size,
 			  u32 *save_area_used_size)
-- 
GitLab


From 21e1217b4c0e0234704d50ea303c7603266604ac Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:30:57 -0400
Subject: [PATCH 1052/1791] drm/amdgpu: Fix VM fault reporting on XCC1

Fix VM fault reporting and clear VM fault register
for XCC1.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 0163a761ccf06..681bc9d354fee 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -557,6 +557,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	u64 addr;
 	uint32_t cam_index = 0;
 	int ret;
+	uint32_t node_id;
 
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
@@ -611,7 +612,9 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
 	} else {
 		hub_name = "gfxhub0";
-		hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+		node_id = (adev->ip_versions[GC_HWIP][0] ==
+			   IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
+		hub = &adev->vmhub[node_id/2];
 	}
 
 	memset(&task_info, 0, sizeof(struct amdgpu_task_info));
@@ -645,11 +648,10 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	rw = REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, RW);
 	WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
 
-
 	dev_err(adev->dev,
 		"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
 		status);
-	if (hub == &adev->vmhub[AMDGPU_GFXHUB(0)]) {
+	if (entry->vmid_src == AMDGPU_GFXHUB(0)) {
 		dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
 			cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" :
 			gfxhub_client_ids[cid],
-- 
GitLab


From 5e4060123687c4f2c9fb855874f77b14f07526d6 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:34:38 -0400
Subject: [PATCH 1053/1791] drm/amdkfd: Call DQM stop during DQM uninitialize

During DQM tear down, call DQM stop to unitialize HIQ and
associated memory allocated during packet manager init.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 01ea038cab6d7..f78c1e7aad57d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1348,9 +1348,16 @@ static int start_nocpsch(struct device_queue_manager *dqm)
 
 static int stop_nocpsch(struct device_queue_manager *dqm)
 {
+	dqm_lock(dqm);
+	if (!dqm->sched_running) {
+		dqm_unlock(dqm);
+		return 0;
+	}
+
 	if (dqm->dev->adev->asic_type == CHIP_HAWAII)
 		pm_uninit(&dqm->packet_mgr, false);
 	dqm->sched_running = false;
+	dqm_unlock(dqm);
 
 	return 0;
 }
@@ -2423,6 +2430,7 @@ static void deallocate_hiq_sdma_mqd(struct kfd_node *dev,
 
 void device_queue_manager_uninit(struct device_queue_manager *dqm)
 {
+	dqm->ops.stop(dqm);
 	dqm->ops.uninitialize(dqm);
 	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		deallocate_hiq_sdma_mqd(dqm->dev, &dqm->hiq_sdma_mqd);
-- 
GitLab


From f38f147ab3121adbd7510a82e6eb0b41a356c26e Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:35:55 -0400
Subject: [PATCH 1054/1791] drm/amdkfd: Update sysfs node properties for multi
 XCC

Update simd_count and array_count node properties to report
values multiplied by number of XCCs in the partition.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 94af37df3ed2f..6d958bf0fe908 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -468,7 +468,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 	sysfs_show_32bit_prop(buffer, offs, "cpu_cores_count",
 			      dev->node_props.cpu_cores_count);
 	sysfs_show_32bit_prop(buffer, offs, "simd_count",
-			      dev->gpu ? dev->node_props.simd_count : 0);
+			      dev->gpu ? (dev->node_props.simd_count *
+					  dev->gpu->num_xcc_per_node) : 0);
 	sysfs_show_32bit_prop(buffer, offs, "mem_banks_count",
 			      dev->node_props.mem_banks_count);
 	sysfs_show_32bit_prop(buffer, offs, "caches_count",
@@ -492,7 +493,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 	sysfs_show_32bit_prop(buffer, offs, "wave_front_size",
 			      dev->node_props.wave_front_size);
 	sysfs_show_32bit_prop(buffer, offs, "array_count",
-			      dev->node_props.array_count);
+			      dev->gpu ? (dev->node_props.array_count *
+					  dev->gpu->num_xcc_per_node) : 0);
 	sysfs_show_32bit_prop(buffer, offs, "simd_arrays_per_engine",
 			      dev->node_props.simd_arrays_per_engine);
 	sysfs_show_32bit_prop(buffer, offs, "cu_per_simd_array",
-- 
GitLab


From a805889a15315f7fa78c1c4bb2f1875c7c43f919 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 9 May 2022 22:52:39 -0400
Subject: [PATCH 1055/1791] drm/amdkfd: Update SDMA queue management for
 GFX9.4.3

This patch updates SDMA queue management for multi XCC in GFX9.4.3.
- Allocate/deallocate SDMA queues from the correct SDMA engines
  based on the partition mode.
- Updates the kgd2kfd interface to fetch the correct SDMA register
  addresses.
- It also fixes dumping correct SDMA queue info in debugfs.

v2: squash in fix "drm/amdkfd: Fix XGMI SDMA user-mode queue allocation"

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   | 194 +++++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |   8 +-
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |  59 +++---
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |   4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |   3 +
 5 files changed, 227 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 49d8087e469ec..e81bdca53f42f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -31,6 +31,192 @@
 #include "oss/osssys_4_0_sh_mask.h"
 #include "v9_structs.h"
 #include "soc15.h"
+#include "sdma/sdma_4_4_2_offset.h"
+#include "sdma/sdma_4_4_2_sh_mask.h"
+
+static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd)
+{
+	return (struct v9_sdma_mqd *)mqd;
+}
+
+static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
+					unsigned int engine_id,
+					unsigned int queue_id)
+{
+	uint32_t sdma_engine_reg_base = SOC15_REG_OFFSET(SDMA0, engine_id,
+					regSDMA_RLC0_RB_CNTL) -
+					regSDMA_RLC0_RB_CNTL;
+	uint32_t retval = sdma_engine_reg_base +
+		  queue_id * (regSDMA_RLC1_RB_CNTL - regSDMA_RLC0_RB_CNTL);
+
+	pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", engine_id,
+							queue_id, retval);
+	return retval;
+}
+
+int kgd_gfx_v9_4_3_hqd_sdma_load(struct amdgpu_device *adev, void *mqd,
+				 uint32_t __user *wptr, struct mm_struct *mm)
+{
+	struct v9_sdma_mqd *m;
+	uint32_t sdma_rlc_reg_offset;
+	unsigned long end_jiffies;
+	uint32_t data;
+	uint64_t data64;
+	uint64_t __user *wptr64 = (uint64_t __user *)wptr;
+
+	m = get_sdma_mqd(mqd);
+	sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
+							m->sdma_queue_id);
+
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL,
+		m->sdmax_rlcx_rb_cntl & (~SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK));
+
+	end_jiffies = msecs_to_jiffies(2000) + jiffies;
+	while (true) {
+		data = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_CONTEXT_STATUS);
+		if (data & SDMA_RLC0_CONTEXT_STATUS__IDLE_MASK)
+			break;
+		if (time_after(jiffies, end_jiffies)) {
+			pr_err("SDMA RLC not idle in %s\n", __func__);
+			return -ETIME;
+		}
+		usleep_range(500, 1000);
+	}
+
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL_OFFSET,
+		m->sdmax_rlcx_doorbell_offset);
+
+	data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA_RLC0_DOORBELL,
+				ENABLE, 1);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL, data);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR,
+					m->sdmax_rlcx_rb_rptr);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_HI,
+					m->sdmax_rlcx_rb_rptr_hi);
+
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_MINOR_PTR_UPDATE, 1);
+	if (read_user_wptr(mm, wptr64, data64)) {
+		WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR,
+			lower_32_bits(data64));
+		WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR_HI,
+			upper_32_bits(data64));
+	} else {
+		WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR,
+			m->sdmax_rlcx_rb_rptr);
+		WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR_HI,
+			m->sdmax_rlcx_rb_rptr_hi);
+	}
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_MINOR_PTR_UPDATE, 0);
+
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_BASE, m->sdmax_rlcx_rb_base);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_BASE_HI,
+			m->sdmax_rlcx_rb_base_hi);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_ADDR_LO,
+			m->sdmax_rlcx_rb_rptr_addr_lo);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_ADDR_HI,
+			m->sdmax_rlcx_rb_rptr_addr_hi);
+
+	data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA_RLC0_RB_CNTL,
+				RB_ENABLE, 1);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, data);
+
+	return 0;
+}
+
+int kgd_gfx_v9_4_3_hqd_sdma_dump(struct amdgpu_device *adev,
+				 uint32_t engine_id, uint32_t queue_id,
+				 uint32_t (**dump)[2], uint32_t *n_regs)
+{
+	uint32_t sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev,
+							engine_id, queue_id);
+	uint32_t i = 0, reg;
+#undef HQD_N_REGS
+#define HQD_N_REGS (19+6+7+12)
+#define DUMP_REG(addr) do {				\
+		if (WARN_ON_ONCE(i >= HQD_N_REGS))      \
+			break;				\
+		(*dump)[i][0] = (addr) << 2;            \
+		(*dump)[i++][1] = RREG32(addr);         \
+	} while (0)
+
+	*dump = kmalloc_array(HQD_N_REGS * 2, sizeof(uint32_t), GFP_KERNEL);
+	if (*dump == NULL)
+		return -ENOMEM;
+
+	for (reg = regSDMA_RLC0_RB_CNTL; reg <= regSDMA_RLC0_DOORBELL; reg++)
+		DUMP_REG(sdma_rlc_reg_offset + reg);
+	for (reg = regSDMA_RLC0_STATUS; reg <= regSDMA_RLC0_CSA_ADDR_HI; reg++)
+		DUMP_REG(sdma_rlc_reg_offset + reg);
+	for (reg = regSDMA_RLC0_IB_SUB_REMAIN;
+	     reg <= regSDMA_RLC0_MINOR_PTR_UPDATE; reg++)
+		DUMP_REG(sdma_rlc_reg_offset + reg);
+	for (reg = regSDMA_RLC0_MIDCMD_DATA0;
+	     reg <= regSDMA_RLC0_MIDCMD_CNTL; reg++)
+		DUMP_REG(sdma_rlc_reg_offset + reg);
+
+	WARN_ON_ONCE(i != HQD_N_REGS);
+	*n_regs = i;
+
+	return 0;
+}
+
+bool kgd_gfx_v9_4_3_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
+{
+	struct v9_sdma_mqd *m;
+	uint32_t sdma_rlc_reg_offset;
+	uint32_t sdma_rlc_rb_cntl;
+
+	m = get_sdma_mqd(mqd);
+	sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
+							m->sdma_queue_id);
+
+	sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL);
+
+	if (sdma_rlc_rb_cntl & SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK)
+		return true;
+
+	return false;
+}
+
+int kgd_gfx_v9_4_3_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
+				    unsigned int utimeout)
+{
+	struct v9_sdma_mqd *m;
+	uint32_t sdma_rlc_reg_offset;
+	uint32_t temp;
+	unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies;
+
+	m = get_sdma_mqd(mqd);
+	sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id,
+							m->sdma_queue_id);
+
+	temp = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL);
+	temp = temp & ~SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK;
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, temp);
+
+	while (true) {
+		temp = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_CONTEXT_STATUS);
+		if (temp & SDMA_RLC0_CONTEXT_STATUS__IDLE_MASK)
+			break;
+		if (time_after(jiffies, end_jiffies)) {
+			pr_err("SDMA RLC not idle in %s\n", __func__);
+			return -ETIME;
+		}
+		usleep_range(500, 1000);
+	}
+
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL, 0);
+	WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL,
+		RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL) |
+		SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK);
+
+	m->sdmax_rlcx_rb_rptr =
+			RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR);
+	m->sdmax_rlcx_rb_rptr_hi =
+			RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_HI);
+
+	return 0;
+}
 
 static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 				u32 pasid, unsigned int vmid, uint32_t inst)
@@ -166,13 +352,13 @@ const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = {
 	.init_interrupts = kgd_gfx_v9_init_interrupts,
 	.hqd_load = kgd_gfx_v9_4_3_hqd_load,
 	.hiq_mqd_load = kgd_gfx_v9_hiq_mqd_load,
-	.hqd_sdma_load = kgd_arcturus_hqd_sdma_load,
+	.hqd_sdma_load = kgd_gfx_v9_4_3_hqd_sdma_load,
 	.hqd_dump = kgd_gfx_v9_hqd_dump,
-	.hqd_sdma_dump = kgd_arcturus_hqd_sdma_dump,
+	.hqd_sdma_dump = kgd_gfx_v9_4_3_hqd_sdma_dump,
 	.hqd_is_occupied = kgd_gfx_v9_hqd_is_occupied,
-	.hqd_sdma_is_occupied = kgd_arcturus_hqd_sdma_is_occupied,
+	.hqd_sdma_is_occupied = kgd_gfx_v9_4_3_hqd_sdma_is_occupied,
 	.hqd_destroy = kgd_gfx_v9_hqd_destroy,
-	.hqd_sdma_destroy = kgd_arcturus_hqd_sdma_destroy,
+	.hqd_sdma_destroy = kgd_gfx_v9_4_3_hqd_sdma_destroy,
 	.wave_control_execute = kgd_gfx_v9_wave_control_execute,
 	.get_atc_vmid_pasid_mapping_info =
 				kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 37c6dc5c37bfb..ec5f85ff34e51 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -741,6 +741,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		if (!node)
 			goto node_alloc_error;
 
+		node->node_id = i;
 		node->adev = kfd->adev;
 		node->kfd = kfd;
 		node->kfd2kgd = kfd->kfd2kgd;
@@ -1323,15 +1324,16 @@ unsigned int kfd_get_num_sdma_engines(struct kfd_node *node)
 {
 	/* If XGMI is not supported, all SDMA engines are PCIe */
 	if (!node->adev->gmc.xgmi.supported)
-		return node->adev->sdma.num_instances;
+		return node->adev->sdma.num_instances/(int)node->kfd->num_nodes;
 
-	return min(node->adev->sdma.num_instances, 2);
+	return min(node->adev->sdma.num_instances/(int)node->kfd->num_nodes, 2);
 }
 
 unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node)
 {
 	/* After reserved for PCIe, the rest of engines are XGMI */
-	return node->adev->sdma.num_instances - kfd_get_num_sdma_engines(node);
+	return node->adev->sdma.num_instances/(int)node->kfd->num_nodes -
+		kfd_get_num_sdma_engines(node);
 }
 
 #if defined(CONFIG_DEBUG_FS)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index f78c1e7aad57d..69419a53a14e5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -124,6 +124,15 @@ static inline uint64_t get_reserved_sdma_queues_bitmap(struct device_queue_manag
 	return dqm->dev->kfd->device_info.reserved_sdma_queues_bitmap;
 }
 
+static void init_sdma_bitmaps(struct device_queue_manager *dqm)
+{
+	bitmap_zero(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES);
+	bitmap_set(dqm->sdma_bitmap, 0, get_num_sdma_queues(dqm));
+
+	bitmap_zero(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES);
+	bitmap_set(dqm->xgmi_sdma_bitmap, 0, get_num_xgmi_sdma_queues(dqm));
+}
+
 void program_sh_mem_settings(struct device_queue_manager *dqm,
 					struct qcm_process_device *qpd)
 {
@@ -1268,24 +1277,6 @@ static void init_interrupts(struct device_queue_manager *dqm)
 	}
 }
 
-static void init_sdma_bitmaps(struct device_queue_manager *dqm)
-{
-	unsigned int num_sdma_queues =
-		min_t(unsigned int, sizeof(dqm->sdma_bitmap)*8,
-		      get_num_sdma_queues(dqm));
-	unsigned int num_xgmi_sdma_queues =
-		min_t(unsigned int, sizeof(dqm->xgmi_sdma_bitmap)*8,
-		      get_num_xgmi_sdma_queues(dqm));
-
-	if (num_sdma_queues)
-		dqm->sdma_bitmap = GENMASK_ULL(num_sdma_queues-1, 0);
-	if (num_xgmi_sdma_queues)
-		dqm->xgmi_sdma_bitmap = GENMASK_ULL(num_xgmi_sdma_queues-1, 0);
-
-	dqm->sdma_bitmap &= ~get_reserved_sdma_queues_bitmap(dqm);
-	pr_info("sdma_bitmap: %llx\n", dqm->sdma_bitmap);
-}
-
 static int initialize_nocpsch(struct device_queue_manager *dqm)
 {
 	int pipe, queue;
@@ -1375,46 +1366,49 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 	int bit;
 
 	if (q->properties.type == KFD_QUEUE_TYPE_SDMA) {
-		if (dqm->sdma_bitmap == 0) {
+		if (bitmap_empty(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES)) {
 			pr_err("No more SDMA queue to allocate\n");
 			return -ENOMEM;
 		}
 
 		if (restore_sdma_id) {
 			/* Re-use existing sdma_id */
-			if (!(dqm->sdma_bitmap & (1ULL << *restore_sdma_id))) {
+			if (!test_bit(*restore_sdma_id, dqm->sdma_bitmap)) {
 				pr_err("SDMA queue already in use\n");
 				return -EBUSY;
 			}
-			dqm->sdma_bitmap &= ~(1ULL << *restore_sdma_id);
+			clear_bit(*restore_sdma_id, dqm->sdma_bitmap);
 			q->sdma_id = *restore_sdma_id;
 		} else {
 			/* Find first available sdma_id */
-			bit = __ffs64(dqm->sdma_bitmap);
-			dqm->sdma_bitmap &= ~(1ULL << bit);
+			bit = find_first_bit(dqm->sdma_bitmap,
+					     get_num_sdma_queues(dqm));
+			clear_bit(bit, dqm->sdma_bitmap);
 			q->sdma_id = bit;
 		}
 
-		q->properties.sdma_engine_id = q->sdma_id %
-				kfd_get_num_sdma_engines(dqm->dev);
+		q->properties.sdma_engine_id =
+			dqm->dev->node_id * get_num_all_sdma_engines(dqm) +
+			q->sdma_id % kfd_get_num_sdma_engines(dqm->dev);
 		q->properties.sdma_queue_id = q->sdma_id /
 				kfd_get_num_sdma_engines(dqm->dev);
 	} else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) {
-		if (dqm->xgmi_sdma_bitmap == 0) {
+		if (bitmap_empty(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES)) {
 			pr_err("No more XGMI SDMA queue to allocate\n");
 			return -ENOMEM;
 		}
 		if (restore_sdma_id) {
 			/* Re-use existing sdma_id */
-			if (!(dqm->xgmi_sdma_bitmap & (1ULL << *restore_sdma_id))) {
+			if (!test_bit(*restore_sdma_id, dqm->xgmi_sdma_bitmap)) {
 				pr_err("SDMA queue already in use\n");
 				return -EBUSY;
 			}
-			dqm->xgmi_sdma_bitmap &= ~(1ULL << *restore_sdma_id);
+			clear_bit(*restore_sdma_id, dqm->xgmi_sdma_bitmap);
 			q->sdma_id = *restore_sdma_id;
 		} else {
-			bit = __ffs64(dqm->xgmi_sdma_bitmap);
-			dqm->xgmi_sdma_bitmap &= ~(1ULL << bit);
+			bit = find_first_bit(dqm->xgmi_sdma_bitmap,
+					     get_num_xgmi_sdma_queues(dqm));
+			clear_bit(bit, dqm->xgmi_sdma_bitmap);
 			q->sdma_id = bit;
 		}
 		/* sdma_engine_id is sdma id including
@@ -1424,6 +1418,7 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 		 * PCIe-optimized ones
 		 */
 		q->properties.sdma_engine_id =
+			dqm->dev->node_id * get_num_all_sdma_engines(dqm) +
 			kfd_get_num_sdma_engines(dqm->dev) +
 			q->sdma_id % kfd_get_num_xgmi_sdma_engines(dqm->dev);
 		q->properties.sdma_queue_id = q->sdma_id /
@@ -1442,11 +1437,11 @@ static void deallocate_sdma_queue(struct device_queue_manager *dqm,
 	if (q->properties.type == KFD_QUEUE_TYPE_SDMA) {
 		if (q->sdma_id >= get_num_sdma_queues(dqm))
 			return;
-		dqm->sdma_bitmap |= (1ULL << q->sdma_id);
+		set_bit(q->sdma_id, dqm->sdma_bitmap);
 	} else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) {
 		if (q->sdma_id >= get_num_xgmi_sdma_queues(dqm))
 			return;
-		dqm->xgmi_sdma_bitmap |= (1ULL << q->sdma_id);
+		set_bit(q->sdma_id, dqm->xgmi_sdma_bitmap);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index e554a48f30542..b11c474d40671 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -239,8 +239,8 @@ struct device_queue_manager {
 	unsigned int		total_queue_count;
 	unsigned int		next_pipe_to_allocate;
 	unsigned int		*allocated_queues;
-	uint64_t		sdma_bitmap;
-	uint64_t		xgmi_sdma_bitmap;
+	DECLARE_BITMAP(sdma_bitmap, KFD_MAX_SDMA_QUEUES);
+	DECLARE_BITMAP(xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES);
 	/* the pasid mapping for each kfd vmid */
 	uint16_t		vmid_pasid[VMID_NUM];
 	uint64_t		pipelines_addr;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 1337fcdf8958a..5cfebcc8b3059 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -113,6 +113,8 @@
 
 #define KFD_UNMAP_LATENCY_MS	(4000)
 
+#define KFD_MAX_SDMA_QUEUES	128
+
 /*
  * 512 = 0x200
  * The doorbell index distance between SDMA RLC (2*i) and (2*i+1) in the
@@ -260,6 +262,7 @@ struct kfd_vmid_info {
 struct kfd_dev;
 
 struct kfd_node {
+	unsigned int node_id;
 	struct amdgpu_device *adev;     /* Duplicated here along with keeping
 					 * a copy in kfd_dev to save a hop
 					 */
-- 
GitLab


From a8027fcd08f9127d38edeb59600ecb76c56a121a Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 2 Mar 2022 21:40:38 -0500
Subject: [PATCH 1056/1791] drm/amdgpu: Fix CP_HYP_XCP_CTL register programming
 in CPX mode

Currently, in CPX mode, the CP_HYP_XCP_CTL register is programmed
incorrectly with the number of XCCs in the partition. As a result,
HIQ doesn't work in CPX mode. Fix this by programming the correct
number of XCCs in a partition, which is 1, in CPX mode.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 232feb387a408..1dcb69b4816f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1161,7 +1161,7 @@ static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
 		break;
 	case 2:
 		tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID);
-		tmp = tmp | (adev->gfx.num_xcd << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
+		tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
 		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, tmp);
 
 		tmp = xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, PHYSICAL_XCC_ID);
-- 
GitLab


From ef75a6ef37235e211bbdb17c25e5f79c55df1750 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Thu, 3 Mar 2022 10:56:05 -0500
Subject: [PATCH 1057/1791] drm/amdkfd: Update coherence settings for svm
 ranges

Recently introduced commit "drm/amdgpu: Set cache coherency
for GC 9.4.3" did not update the settings applicable for svm ranges.
Add the coherence settings for svm ranges for GFX IP 9.4.3.

Reviewed-by: Amber Lin <amber.lin@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 17 +++++++++++++++++
 include/uapi/linux/kfd_ioctl.h       |  2 ++
 2 files changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 96ccff79902ce..4b4f3bf8b823c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1159,6 +1159,7 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 	uint64_t pte_flags;
 	bool snoop = (domain != SVM_RANGE_VRAM_DOMAIN);
 	bool coherent = flags & KFD_IOCTL_SVM_FLAG_COHERENT;
+	bool uncached = flags & KFD_IOCTL_SVM_FLAG_UNCACHED;
 
 	if (domain == SVM_RANGE_VRAM_DOMAIN)
 		bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev);
@@ -1198,6 +1199,22 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 				AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
 		}
 		break;
+	case IP_VERSION(9, 4, 3):
+		//TODO: Need more work for handling multiple memory partitions
+		//e.g. NPS4. Current approch is only applicable without memory
+		//partitions.
+		snoop = true;
+		if (uncached)
+			mapping_flags |= AMDGPU_VM_MTYPE_UC;
+		/* local HBM region close to partition*/
+		else if (bo_adev == adev)
+			mapping_flags |= AMDGPU_VM_MTYPE_RW;
+		/* local HBM region far from partition or remote XGMI GPU or
+		 * system memory
+		 */
+		else
+			mapping_flags |= AMDGPU_VM_MTYPE_NC;
+		break;
 	default:
 		mapping_flags |= coherent ?
 			AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 2da5c3ad71bd0..2a9671e1ddb5a 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -623,6 +623,8 @@ enum kfd_mmio_remap {
 #define KFD_IOCTL_SVM_FLAG_GPU_READ_MOSTLY     0x00000020
 /* Keep GPU memory mapping always valid as if XNACK is disable */
 #define KFD_IOCTL_SVM_FLAG_GPU_ALWAYS_MAPPED   0x00000040
+/* Uncached access to memory */
+#define KFD_IOCTL_SVM_FLAG_UNCACHED 0x00000080
 
 /**
  * kfd_ioctl_svm_op - SVM ioctl operations
-- 
GitLab


From 5db392a04575120de4e73ee10c0dc727426100c7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 10 Mar 2022 16:45:53 +0530
Subject: [PATCH 1058/1791] drm/amdgpu: Use new atomfirmware init for GC 9.4.3

Use the new atomfirmware initialization logic for GC 9.4.3 based ASICs
also. ASIC init logic doesn't consider boot clocks during init.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 82a3d0ff65607..54be85539ecb8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -942,7 +942,8 @@ static int amdgpu_device_asic_init(struct amdgpu_device *adev)
 {
 	amdgpu_asic_pre_asic_init(adev);
 
-	if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0))
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) ||
+	    adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0))
 		return amdgpu_atomfirmware_asic_init(adev, true);
 	else
 		return amdgpu_atom_asic_init(adev->mode_info.atom_context);
-- 
GitLab


From d25555866172a0454b0dc1374b0ff29e1ae5f676 Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Wed, 23 Mar 2022 11:01:52 -0400
Subject: [PATCH 1059/1791] drm/amdgpu: Set XNACK per process on GC 9.4.3

Set RETRY_PERMISSION_OR_INVALID_PAGE_FAULT bit in VM_CONTEXT1_CNTL
as well so XNACK can be enabled in the SQ per process.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 6 ++++--
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c  | 2 +-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index e5016fea1f280..d74621662311e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -328,13 +328,15 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 					    PAGE_TABLE_BLOCK_SIZE,
 					    block_size);
 			/* Send no-retry XNACK on fault to suppress VM fault storm.
-			 * On Aldebaran, XNACK can be enabled in the SQ per-process.
+			 * On 9.4.2 and 9.4.3, XNACK can be enabled in
+			 * the SQ per-process.
 			 * Retry faults need to be enabled for that to work.
 			 */
 			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 					    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
 					    !adev->gmc.noretry ||
-					    adev->asic_type == CHIP_ALDEBARAN);
+					    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
+					    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3));
 			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL,
 					    i * hub->ctx_distance, tmp);
 			WREG32_SOC15_OFFSET(GC, j,
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 9ec06f9db7617..3883758b79933 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -288,7 +288,7 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
 				    PAGE_TABLE_BLOCK_SIZE,
 				    block_size);
-		/* On Aldebaran, XNACK can be enabled in the SQ per-process.
+		/* On 9.4.3, XNACK can be enabled in the SQ per-process.
 		 * Retry faults need to be enabled for that to work.
 		 */
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-- 
GitLab


From 0ee20b86966cff359f51051a56f8c2d89b09aed4 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 28 Feb 2022 19:30:10 +0800
Subject: [PATCH 1060/1791] drm/amdgpu: assign the doorbell index in 1st page
 to sdma page queue

Previously for vega10, the sdma_doorbell_range is only enough for sdma
gfx queue, thus the index on second doorbell page is allocated for sdma
page queue. From vega20, the sdma_doorbell_range on 1st page is enlarged.
Therefore, just leverage these index instead of allocation on 2nd page.

v2: change "(x << 1) + 2" to "(x + 1) << 1" for readability and add comments.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |  3 ++-
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c     | 14 ++++++++++++--
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c   |  7 +++----
 3 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 54be85539ecb8..38b6bbcaf186f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1091,7 +1091,8 @@ static int amdgpu_device_doorbell_init(struct amdgpu_device *adev)
 		 * doorbells are in the first page. So with paging queue enabled,
 		 * the max num_kernel_doorbells should + 1 page (0x400 in dword)
 		 */
-		if (adev->asic_type >= CHIP_VEGA10)
+		if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(4, 0, 0) &&
+		    adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(4, 2, 0))
 			adev->doorbell.num_kernel_doorbells += 0x400;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 50b6eb9bcfda7..70b0d1fd9868b 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1847,8 +1847,18 @@ static int sdma_v4_0_sw_init(void *handle)
 			/* paging queue use same doorbell index/routing as gfx queue
 			 * with 0x400 (4096 dwords) offset on second doorbell page
 			 */
-			ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
-			ring->doorbell_index += 0x400;
+			if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(4, 0, 0) &&
+			    adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(4, 2, 0)) {
+				ring->doorbell_index =
+					adev->doorbell_index.sdma_engine[i] << 1;
+				ring->doorbell_index += 0x400;
+			} else {
+				/* From vega20, the sdma_doorbell_range in 1st
+				 * doorbell page is reserved for page queue.
+				 */
+				ring->doorbell_index =
+					(adev->doorbell_index.sdma_engine[i] + 1) << 1;
+			}
 
 			if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) && i >= 5)
 				ring->vm_hub = AMDGPU_MMHUB1(0);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 7efe7c43fffbe..441d6911fd208 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1323,11 +1323,10 @@ static int sdma_v4_4_2_sw_init(void *handle)
 			ring->ring_obj = NULL;
 			ring->use_doorbell = true;
 
-			/* paging queue use same doorbell index/routing as gfx queue
-			 * with 0x400 (4096 dwords) offset on second doorbell page
+			/* doorbell index of page queue is assigned right after
+			 * gfx queue on the same instance
 			 */
-			ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
-			ring->doorbell_index += 0x400;
+			ring->doorbell_index = (adev->doorbell_index.sdma_engine[i] + 1) << 1;
 			ring->vm_hub = AMDGPU_MMHUB0(0);
 
 			sprintf(ring->name, "page%d", i);
-- 
GitLab


From 386ea27c3b0bcdd5b5be74bdf26022ab931eae7b Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 23 Feb 2022 11:43:01 +0800
Subject: [PATCH 1061/1791] drm/amdgpu: adjust some basic elements for multiple
 AID case

add some elements below:
 - num_aid
 - aid_id for each sdma instance
 - num_inst_per_aid for sdma

and extend macro size below:
 - SDMA_MAX_INSTANCES to 16
 - AMDGPU_MAX_RINGS to 96
 - AMDGPU_MAX_HWIP_RINGS to 32

v2: move aid_id from amdgpu_ring to amdgpu_sdma_instance. (Lijo)

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      | 1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 4 +++-
 3 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 00c52caeebeb0..5f4396185a2e6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1051,6 +1051,7 @@ struct amdgpu_device {
 
 	bool                            job_hang;
 	bool                            dc_enabled;
+	uint32_t			num_aid;
 };
 
 static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index b0dc0a0c2631f..8f4b416a92e6e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -37,8 +37,8 @@ struct amdgpu_job;
 struct amdgpu_vm;
 
 /* max number of rings */
-#define AMDGPU_MAX_RINGS		28
-#define AMDGPU_MAX_HWIP_RINGS		8
+#define AMDGPU_MAX_RINGS		96
+#define AMDGPU_MAX_HWIP_RINGS		32
 #define AMDGPU_MAX_GFX_RINGS		2
 #define AMDGPU_MAX_SW_GFX_RINGS         2
 #define AMDGPU_MAX_COMPUTE_RINGS	8
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index fc8528812598b..67975dcede5d7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -26,7 +26,7 @@
 #include "amdgpu_ras.h"
 
 /* max number of IP instances */
-#define AMDGPU_MAX_SDMA_INSTANCES		8
+#define AMDGPU_MAX_SDMA_INSTANCES		16
 
 enum amdgpu_sdma_irq {
 	AMDGPU_SDMA_IRQ_INSTANCE0  = 0,
@@ -49,6 +49,7 @@ struct amdgpu_sdma_instance {
 	struct amdgpu_ring	ring;
 	struct amdgpu_ring	page;
 	bool			burst_nop;
+	uint32_t		aid_id;
 };
 
 struct amdgpu_sdma_ras {
@@ -66,6 +67,7 @@ struct amdgpu_sdma {
 	struct amdgpu_irq_src	srbm_write_irq;
 
 	int			num_instances;
+	int			num_inst_per_aid;
 	uint32_t                    srbm_soft_reset;
 	bool			has_page_queue;
 	struct ras_common_if	*ras_if;
-- 
GitLab


From f786b1d4ec778a5fc23911f06a0e38c9f4953b0c Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 20 Apr 2022 17:03:00 +0800
Subject: [PATCH 1062/1791] drm/amdgpu: add support for SDMA on multiple AIDs

Initialize SDMA instances on each AID.

v2: revise coding fault in hw_fini

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h |  8 ++++++++
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 20 +++++++++++++-------
 2 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 67975dcede5d7..632b77138fe40 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -37,6 +37,14 @@ enum amdgpu_sdma_irq {
 	AMDGPU_SDMA_IRQ_INSTANCE5,
 	AMDGPU_SDMA_IRQ_INSTANCE6,
 	AMDGPU_SDMA_IRQ_INSTANCE7,
+	AMDGPU_SDMA_IRQ_INSTANCE8,
+	AMDGPU_SDMA_IRQ_INSTANCE9,
+	AMDGPU_SDMA_IRQ_INSTANCE10,
+	AMDGPU_SDMA_IRQ_INSTANCE11,
+	AMDGPU_SDMA_IRQ_INSTANCE12,
+	AMDGPU_SDMA_IRQ_INSTANCE13,
+	AMDGPU_SDMA_IRQ_INSTANCE14,
+	AMDGPU_SDMA_IRQ_INSTANCE15,
 	AMDGPU_SDMA_IRQ_LAST
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 441d6911fd208..7deadea03caa5 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1253,9 +1253,10 @@ static int sdma_v4_4_2_sw_init(void *handle)
 	struct amdgpu_ring *ring;
 	int r, i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	u32 aid_id;
 
 	/* SDMA trap event */
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for (i = 0; i < adev->sdma.num_inst_per_aid; i++) {
 		r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i),
 				      SDMA0_4_0__SRCID__SDMA_TRAP,
 				      &adev->sdma.trap_irq);
@@ -1264,7 +1265,7 @@ static int sdma_v4_4_2_sw_init(void *handle)
 	}
 
 	/* SDMA SRAM ECC event */
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for (i = 0; i < adev->sdma.num_inst_per_aid; i++) {
 		r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i),
 				      SDMA0_4_0__SRCID__SDMA_SRAM_ECC,
 				      &adev->sdma.ecc_irq);
@@ -1273,7 +1274,7 @@ static int sdma_v4_4_2_sw_init(void *handle)
 	}
 
 	/* SDMA VM_HOLE/DOORBELL_INV/POLL_TIMEOUT/SRBM_WRITE_PROTECTION event*/
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for (i = 0; i < adev->sdma.num_inst_per_aid; i++) {
 		r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i),
 				      SDMA0_4_0__SRCID__SDMA_VM_HOLE,
 				      &adev->sdma.vm_hole_irq);
@@ -1303,15 +1304,17 @@ static int sdma_v4_4_2_sw_init(void *handle)
 		ring = &adev->sdma.instance[i].ring;
 		ring->ring_obj = NULL;
 		ring->use_doorbell = true;
+		aid_id = adev->sdma.instance[i].aid_id;
 
 		DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i,
 				ring->use_doorbell?"true":"false");
 
 		/* doorbell size is 2 dwords, get DWORD offset */
 		ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
-		ring->vm_hub = AMDGPU_MMHUB0(0);
+		ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
-		sprintf(ring->name, "sdma%d", i);
+		sprintf(ring->name, "sdma%d.%d", aid_id,
+				i % adev->sdma.num_inst_per_aid);
 		r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq,
 				     AMDGPU_SDMA_IRQ_INSTANCE0 + i,
 				     AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -1327,9 +1330,10 @@ static int sdma_v4_4_2_sw_init(void *handle)
 			 * gfx queue on the same instance
 			 */
 			ring->doorbell_index = (adev->doorbell_index.sdma_engine[i] + 1) << 1;
-			ring->vm_hub = AMDGPU_MMHUB0(0);
+			ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
-			sprintf(ring->name, "page%d", i);
+			sprintf(ring->name, "page%d.%d", aid_id,
+					i % adev->sdma.num_inst_per_aid);
 			r = amdgpu_ring_init(adev, ring, 1024,
 					     &adev->sdma.trap_irq,
 					     AMDGPU_SDMA_IRQ_INSTANCE0 + i,
@@ -1811,6 +1815,8 @@ static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev)
 				&sdma_v4_4_2_page_ring_funcs;
 			adev->sdma.instance[i].page.me = i;
 		}
+
+		adev->sdma.instance[i].aid_id = i / adev->sdma.num_inst_per_aid;
 	}
 }
 
-- 
GitLab


From 2a47a2d90e5cd96c24503061c8920a1e6ee248a0 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 1 Mar 2022 19:42:29 +0800
Subject: [PATCH 1063/1791] drm/amdgpu: assign the doorbell index for sdma on
 non-AID0

Allocate new sdma doorbell index for the instances only on AID1 for now.

Todo: there's limitation that SDMA doorbell index on SDMA 4.4.2 needs to be
less than 0x1FF, so the tail part in _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT is not
enough to store sdma doorbell range on maximum 4 AIDs if doorbell_range is 20.
So it looks better to create a new doorbell index assignment table for 4.4.2.

v2: change "(x << 1) + 2" to "(x + 1) << 1" for readability.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h |  6 +++++-
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c     | 18 ++++++++++++++++--
 2 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index ffb75d23d2fc6..b036d2f019300 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -85,6 +85,7 @@ struct amdgpu_doorbell_index {
 	uint32_t last_non_cp;
 	uint32_t xcc1_kiq_start;
 	uint32_t xcc1_mec_ring0_start;
+	uint32_t aid1_sdma_start;
 	uint32_t max_assignment;
 	/* Per engine SDMA doorbell size in dword */
 	uint32_t sdma_doorbell_range;
@@ -171,7 +172,10 @@ typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT
 	/* 8 compute rings per GC. Max to 0x1CE */
 	AMDGPU_VEGA20_DOORBELL_XCC1_MEC_RING0_START       = 0x197,
 
-	AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT            = 0x1CE,
+	/* AID1 SDMA: 0x1D0 ~ 0x1F7 */
+	AMDGPU_VEGA20_DOORBELL_AID1_sDMA_START           = 0x1D0,
+
+	AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT            = 0x1F7,
 	AMDGPU_VEGA20_DOORBELL_INVALID                   = 0xFFFF
 } AMDGPU_VEGA20_DOORBELL_ASSIGNMENT;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 7deadea03caa5..6935a24d1e89f 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1310,7 +1310,14 @@ static int sdma_v4_4_2_sw_init(void *handle)
 				ring->use_doorbell?"true":"false");
 
 		/* doorbell size is 2 dwords, get DWORD offset */
-		ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
+		if (aid_id > 0)
+			ring->doorbell_index =
+				(adev->doorbell_index.aid1_sdma_start << 1)
+				+ adev->doorbell_index.sdma_doorbell_range
+				* (i - adev->sdma.num_inst_per_aid);
+		else
+			ring->doorbell_index =
+				adev->doorbell_index.sdma_engine[i] << 1;
 		ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
 		sprintf(ring->name, "sdma%d.%d", aid_id,
@@ -1329,7 +1336,14 @@ static int sdma_v4_4_2_sw_init(void *handle)
 			/* doorbell index of page queue is assigned right after
 			 * gfx queue on the same instance
 			 */
-			ring->doorbell_index = (adev->doorbell_index.sdma_engine[i] + 1) << 1;
+			if (aid_id > 0)
+				ring->doorbell_index =
+					((adev->doorbell_index.aid1_sdma_start + 1) << 1)
+					+ adev->doorbell_index.sdma_doorbell_range
+					* (i - adev->sdma.num_inst_per_aid);
+			else
+				ring->doorbell_index =
+					(adev->doorbell_index.sdma_engine[i] + 1) << 1;
 			ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
 			sprintf(ring->name, "page%d.%d", aid_id,
-- 
GitLab


From 3a1083873ba7730970665d04c33680f96b27e3b4 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 2 Mar 2022 17:33:24 +0800
Subject: [PATCH 1064/1791] drm/amdgpu: do mmhub init for multiple AIDs

Mmhub on each AID needs to be initialized respectively

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 609 ++++++++++++++----------
 1 file changed, 348 insertions(+), 261 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 3883758b79933..67338cb3d7bc3 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -53,18 +53,27 @@ static u64 mmhub_v1_8_get_fb_location(struct amdgpu_device *adev)
 static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
 				uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
-
-	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
-			    hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base));
+	struct amdgpu_vmhub *hub;
+	int i;
 
-	WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
-			    hub->ctx_addr_distance * vmid, upper_32_bits(page_table_base));
+	for (i = 0; i < adev->num_aid; i++) {
+		hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+		WREG32_SOC15_OFFSET(MMHUB, i,
+				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+				    hub->ctx_addr_distance * vmid,
+				    lower_32_bits(page_table_base));
+
+		WREG32_SOC15_OFFSET(MMHUB, i,
+				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+				    hub->ctx_addr_distance * vmid,
+				    upper_32_bits(page_table_base));
+	}
 }
 
 static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t pt_base;
+	int i;
 
 	if (adev->gmc.pdb0_bo)
 		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
@@ -76,27 +85,37 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
 	/* If use GART for FB translation, vmid0 page table covers both
 	 * vram and system memory (gart)
 	 */
-	if (adev->gmc.pdb0_bo) {
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
-			     (u32)(adev->gmc.fb_start >> 12));
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
-			     (u32)(adev->gmc.fb_start >> 44));
-
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
-			     (u32)(adev->gmc.gart_end >> 12));
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
-			     (u32)(adev->gmc.gart_end >> 44));
-
-	} else {
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
-			     (u32)(adev->gmc.gart_start >> 12));
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
-			     (u32)(adev->gmc.gart_start >> 44));
-
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
-			     (u32)(adev->gmc.gart_end >> 12));
-		WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
-			     (u32)(adev->gmc.gart_end >> 44));
+	for (i = 0; i < adev->num_aid; i++) {
+		if (adev->gmc.pdb0_bo) {
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				     (u32)(adev->gmc.fb_start >> 12));
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				     (u32)(adev->gmc.fb_start >> 44));
+
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				     (u32)(adev->gmc.gart_end >> 12));
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				     (u32)(adev->gmc.gart_end >> 44));
+
+		} else {
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
+				     (u32)(adev->gmc.gart_start >> 12));
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
+				     (u32)(adev->gmc.gart_start >> 44));
+
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
+				     (u32)(adev->gmc.gart_end >> 12));
+			WREG32_SOC15(MMHUB, i,
+				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
+				     (u32)(adev->gmc.gart_end >> 44));
+		}
 	}
 }
 
@@ -104,159 +123,202 @@ static void mmhub_v1_8_init_system_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t value;
 	uint32_t tmp;
+	int i;
 
-	/* Program the AGP BAR */
-	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BASE, 0);
-	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
-	WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
+	for (i = 0; i < adev->num_aid; i++) {
+		/* Program the AGP BAR */
+		WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BASE, 0);
+		WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BOT,
+			     adev->gmc.agp_start >> 24);
+		WREG32_SOC15(MMHUB, i, regMC_VM_AGP_TOP,
+			     adev->gmc.agp_end >> 24);
 
-	/* Program the system aperture low logical page number. */
-	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
-		     min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
+		if (amdgpu_sriov_vf(adev))
+			return;
 
-	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
-		     max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
+		/* Program the system aperture low logical page number. */
+		WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+			min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
 
-	/* In the case squeezing vram into GART aperture, we don't use
-	 * FB aperture and AGP aperture. Disable them.
-	 */
-	if (adev->gmc.pdb0_bo) {
-		WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, 0xFFFFFF);
-		WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, 0);
-		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP, 0);
-		WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
-		WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
-		WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
-	}
-	if (amdgpu_sriov_vf(adev))
-		return;
+		WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
+			max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
 
-	/* Set default page address. */
-	value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
-	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
-		     (u32)(value >> 12));
-	WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
-		     (u32)(value >> 44));
-
-	/* Program "protection fault". */
-	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
-		     (u32)(adev->dummy_page_addr >> 12));
-	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
-		     (u32)((u64)adev->dummy_page_addr >> 44));
-
-	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
-			    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+		/* In the case squeezing vram into GART aperture, we don't use
+		 * FB aperture and AGP aperture. Disable them.
+		 */
+		if (adev->gmc.pdb0_bo) {
+			WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BOT, 0xFFFFFF);
+			WREG32_SOC15(MMHUB, i, regMC_VM_AGP_TOP, 0);
+			WREG32_SOC15(MMHUB, i, regMC_VM_FB_LOCATION_TOP, 0);
+			WREG32_SOC15(MMHUB, i, regMC_VM_FB_LOCATION_BASE,
+				     0x00FFFFFF);
+			WREG32_SOC15(MMHUB, i,
+				     regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+				     0x3FFFFFFF);
+			WREG32_SOC15(MMHUB, i,
+				     regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+		}
+
+		/* Set default page address. */
+		value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
+		WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+			     (u32)(value >> 12));
+		WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+			     (u32)(value >> 44));
+
+		/* Program "protection fault". */
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+			     (u32)(adev->dummy_page_addr >> 12));
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+			     (u32)((u64)adev->dummy_page_addr >> 44));
+
+		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL2);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
+				    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
+		WREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+	}
 }
 
 static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
 	/* Setup TLB control */
-	tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
-
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    ENABLE_ADVANCED_DRIVER_MODEL, 1);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    MTYPE, MTYPE_UC);/* XXX for emulation. */
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
-
-	WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
+	for (i = 0; i < adev->num_aid; i++) {
+		tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL);
+
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
+				    1);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    SYSTEM_ACCESS_MODE, 3);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    ENABLE_ADVANCED_DRIVER_MODEL, 1);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    SYSTEM_APERTURE_UNMAPPED_ACCESS, 0);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    MTYPE, MTYPE_UC);/* XXX for emulation. */
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
+
+		WREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+	}
 }
 
 static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return;
 
 	/* Setup L2 cache */
-	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
-	/* XXX for emulation, Refer to closed source code.*/
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE,
-			    0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp);
-
-	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
-	tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2, tmp);
-
-	tmp = regVM_L2_CNTL3_DEFAULT;
-	if (adev->gmc.translate_further) {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
-				    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
-	} else {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
-				    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
-	}
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, tmp);
-
-	tmp = regVM_L2_CNTL4_DEFAULT;
-	if (adev->gmc.xgmi.connected_to_cpu) {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
-				    VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
-				    VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
-	} else {
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
-				    VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
-				    VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
-	}
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL4, tmp);
+	for (i = 0; i < adev->num_aid; i++) {
+		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL,
+				    ENABLE_L2_FRAGMENT_PROCESSING, 1);
+		/* XXX for emulation, Refer to closed source code.*/
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL,
+				    L2_PDE0_CACHE_TAG_GENERATION_MODE, 0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION,
+				    0);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL,
+				    CONTEXT1_IDENTITY_ACCESS_MODE, 1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL,
+				    IDENTITY_MODE_FRAGMENT_SIZE, 0);
+		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL, tmp);
+
+		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_CNTL2);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS,
+				    1);
+		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
+		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL2, tmp);
+
+		tmp = regVM_L2_CNTL3_DEFAULT;
+		if (adev->gmc.translate_further) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+					    L2_CACHE_BIGK_FRAGMENT_SIZE, 9);
+		} else {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
+					    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
+		}
+		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL3, tmp);
+
+		tmp = regVM_L2_CNTL4_DEFAULT;
+		if (adev->gmc.xgmi.connected_to_cpu) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+					    VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+					    VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
+		} else {
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+					    VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
+					    VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
+		}
+		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL4, tmp);
+ 	}
 }
 
 static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
+	int i;
 
-	tmp = RREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
-			adev->gmc.vmid0_page_table_depth);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
-			adev->gmc.vmid0_page_table_block_size);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
-			    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
-	WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL, tmp);
+	for (i = 0; i < adev->num_aid; i++) {
+		tmp = RREG32_SOC15(MMHUB, i, regVM_CONTEXT0_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
+				adev->gmc.vmid0_page_table_depth);
+		tmp = REG_SET_FIELD(tmp,
+				    VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE,
+				    adev->gmc.vmid0_page_table_block_size);
+		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
+				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
+		WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_CNTL, tmp);
+	}
 }
 
 static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev)
 {
+	int i;
+
 	if (amdgpu_sriov_vf(adev))
 		return;
 
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, 0xFFFFFFFF);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, 0x0000000F);
-
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, 0);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, 0);
-
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
-	WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+	for (i = 0; i < adev->num_aid; i++) {
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
+			     0XFFFFFFFF);
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
+			     0x0000000F);
+
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
+			     0);
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
+			     0);
+
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
+		WREG32_SOC15(MMHUB, i,
+			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
+	}
 }
 
 static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
+	struct amdgpu_vmhub *hub;
 	unsigned num_level, block_size;
 	uint32_t tmp;
-	int i;
+	int i, j;
 
 	num_level = adev->vm_manager.num_level;
 	block_size = adev->vm_manager.block_size;
@@ -265,60 +327,73 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 	else
 		block_size -= 9;
 
-	for (i = 0; i <= 14; i++) {
-		tmp = RREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL, i);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
-				    num_level);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT,
-				    1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    PAGE_TABLE_BLOCK_SIZE,
-				    block_size);
-		/* On 9.4.3, XNACK can be enabled in the SQ per-process.
-		 * Retry faults need to be enabled for that to work.
-		 */
-		tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
-				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT,
-				    1);
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL,
-				    i * hub->ctx_distance, tmp);
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
-				    i * hub->ctx_addr_distance, 0);
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
-				    i * hub->ctx_addr_distance, 0);
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
-				    i * hub->ctx_addr_distance,
-				    lower_32_bits(adev->vm_manager.max_pfn - 1));
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
-				    i * hub->ctx_addr_distance,
-				    upper_32_bits(adev->vm_manager.max_pfn - 1));
+	for (j = 0; j < adev->num_aid; j++) {
+		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+		for (i = 0; i <= 14; i++) {
+			tmp = RREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT1_CNTL,
+						  i);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    ENABLE_CONTEXT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    PAGE_TABLE_DEPTH, num_level);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+					    PAGE_TABLE_BLOCK_SIZE,
+					    block_size);
+			/* On 9.4.3, XNACK can be enabled in the SQ
+			 * per-process. Retry faults need to be enabled for
+			 * that to work.
+			 */
+			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+				RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1);
+			WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT1_CNTL,
+					    i * hub->ctx_distance, tmp);
+			WREG32_SOC15_OFFSET(MMHUB, j,
+				regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
+				i * hub->ctx_addr_distance, 0);
+			WREG32_SOC15_OFFSET(MMHUB, j,
+				regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
+				i * hub->ctx_addr_distance, 0);
+			WREG32_SOC15_OFFSET(MMHUB, j,
+				regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
+				i * hub->ctx_addr_distance,
+				lower_32_bits(adev->vm_manager.max_pfn - 1));
+			WREG32_SOC15_OFFSET(MMHUB, j,
+				regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
+				i * hub->ctx_addr_distance,
+				upper_32_bits(adev->vm_manager.max_pfn - 1));
+		}
 	}
 }
 
 static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
-	unsigned i;
-
-	for (i = 0; i < 18; ++i) {
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
-				    i * hub->eng_addr_distance, 0xffffffff);
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
-				    i * hub->eng_addr_distance, 0x1f);
+	struct amdgpu_vmhub *hub;
+	unsigned i, j;
+
+	for (j = 0; j < adev->num_aid; j++) {
+		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+		for (i = 0; i < 18; ++i) {
+			WREG32_SOC15_OFFSET(MMHUB, j,
+					regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+					i * hub->eng_addr_distance, 0xffffffff);
+			WREG32_SOC15_OFFSET(MMHUB, j,
+					regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+					i * hub->eng_addr_distance, 0x1f);
+		}
 	}
 }
 
@@ -352,28 +427,33 @@ static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
+	struct amdgpu_vmhub *hub;
 	u32 tmp;
-	u32 i;
+	u32 i, j;
 
 	/* Disable all tables */
-	for (i = 0; i < 16; i++)
-		WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL,
-				    i * hub->ctx_distance, 0);
-
-	/* Setup TLB control */
-	tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
-	tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
-			    ENABLE_ADVANCED_DRIVER_MODEL, 0);
-	WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp);
-
-	if (!amdgpu_sriov_vf(adev)) {
-		/* Setup L2 cache */
-		tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL);
-		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
-		WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp);
-		WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, 0);
+	for (j = 0; j < adev->num_aid; j++) {
+		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
+		for (i = 0; i < 16; i++)
+			WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT0_CNTL,
+					    i * hub->ctx_distance, 0);
+
+		/* Setup TLB control */
+		tmp = RREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
+				    0);
+		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
+				    ENABLE_ADVANCED_DRIVER_MODEL, 0);
+		WREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL, tmp);
+
+		if (!amdgpu_sriov_vf(adev)) {
+			/* Setup L2 cache */
+			tmp = RREG32_SOC15(MMHUB, j, regVM_L2_CNTL);
+			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE,
+					    0);
+			WREG32_SOC15(MMHUB, j, regVM_L2_CNTL, tmp);
+			WREG32_SOC15(MMHUB, j, regVM_L2_CNTL3, 0);
+		}
 	}
 }
 
@@ -386,72 +466,79 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 {
 	u32 tmp;
+	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return;
 
-	tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
-			    value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-			    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
-	if (!value) {
+	for (i = 0; i < adev->num_aid; i++) {
+		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-				    CRASH_ON_NO_RETRY_FAULT, 1);
+				PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
-				    CRASH_ON_RETRY_FAULT, 1);
+				PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+			TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT,
+			value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+				EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+		if (!value) {
+			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+					    CRASH_ON_NO_RETRY_FAULT, 1);
+			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
+					    CRASH_ON_RETRY_FAULT, 1);
+		}
+
+		WREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
 	}
-
-	WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
 }
 
 static void mmhub_v1_8_init(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
-
-	hub->ctx0_ptb_addr_lo32 =
-		SOC15_REG_OFFSET(MMHUB, 0,
-				 regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
-	hub->ctx0_ptb_addr_hi32 =
-		SOC15_REG_OFFSET(MMHUB, 0,
-				 regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
-	hub->vm_inv_eng0_req =
-		SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_REQ);
-	hub->vm_inv_eng0_ack =
-		SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ACK);
-	hub->vm_context0_cntl =
-		SOC15_REG_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL);
-	hub->vm_l2_pro_fault_status =
-		SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_STATUS);
-	hub->vm_l2_pro_fault_cntl =
-		SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL);
-
-	hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
-	hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
-		regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
-	hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ;
-	hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
-		regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+	struct amdgpu_vmhub *hub;
+	int i;
 
+	for (i = 0; i < adev->num_aid; i++) {
+		hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
+
+		hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0,
+			regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
+		hub->ctx0_ptb_addr_hi32 = SOC15_REG_OFFSET(MMHUB, 0,
+			regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
+		hub->vm_inv_eng0_req =
+			SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_REQ);
+		hub->vm_inv_eng0_ack =
+			SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ACK);
+		hub->vm_context0_cntl =
+			SOC15_REG_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL);
+		hub->vm_l2_pro_fault_status = SOC15_REG_OFFSET(MMHUB, 0,
+			regVM_L2_PROTECTION_FAULT_STATUS);
+		hub->vm_l2_pro_fault_cntl = SOC15_REG_OFFSET(MMHUB, 0,
+			regVM_L2_PROTECTION_FAULT_CNTL);
+
+		hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
+		hub->ctx_addr_distance =
+			regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 -
+			regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
+		hub->eng_distance = regVM_INVALIDATE_ENG1_REQ -
+			regVM_INVALIDATE_ENG0_REQ;
+		hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 -
+			regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32;
+	}
 }
 
 static int mmhub_v1_8_set_clockgating(struct amdgpu_device *adev,
-- 
GitLab


From ed42f2cc3b56955310a16da726886e684ed88432 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 24 Feb 2022 16:26:07 +0800
Subject: [PATCH 1065/1791] drm/amdgpu: correct the vmhub reference for each
 XCD in gfxhub init

Correct this though the value is same across different vmhub.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index d74621662311e..d3424ce97aa89 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -42,10 +42,11 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 					 uint32_t vmid,
 					 uint64_t page_table_base)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+	struct amdgpu_vmhub *hub;
 	int i;
 
 	for (i = 0; i < adev->gfx.num_xcd; i++) {
+		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 		WREG32_SOC15_OFFSET(GC, i,
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 				    hub->ctx_addr_distance * vmid,
@@ -291,7 +292,7 @@ static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+	struct amdgpu_vmhub *hub;
 	unsigned num_level, block_size;
 	uint32_t tmp;
 	int i, j;
@@ -304,6 +305,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 		block_size -= 9;
 
 	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		for (i = 0; i <= 14; i++) {
 			tmp = RREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL, i);
 			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
@@ -359,10 +361,11 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+	struct amdgpu_vmhub *hub;
 	unsigned i, j;
 
 	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		for (i = 0 ; i < 18; ++i) {
 			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
 					    i * hub->eng_addr_distance, 0xffffffff);
@@ -408,11 +411,12 @@ static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 {
-	struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)];
+	struct amdgpu_vmhub *hub;
 	u32 tmp;
 	u32 i, j;
 
 	for (j = 0; j < adev->gfx.num_xcd; j++) {
+		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		/* Disable all tables */
 		for (i = 0; i < 16; i++)
 			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT0_CNTL,
-- 
GitLab


From 0d81101c190d1835f1bfba85dc3d65b9f9cacd68 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 25 Feb 2022 15:14:19 +0800
Subject: [PATCH 1066/1791] drm/amdgpu: complement the IH node_id table for
 multiple AIDs

With different node_id, the SDMA interrupt from multiple AIDs can be
distinguished by sw driver.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c  | 4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h  | 6 +++++-
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 3 +++
 3 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 031610c1340a3..3dafaf70c9878 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -100,12 +100,16 @@ const char *soc15_ih_clientid_name[] = {
 };
 
 const int node_id_to_phys_map[NODEID_MAX] = {
+	[AID0_NODEID] = 0,
 	[XCD0_NODEID] = 0,
 	[XCD1_NODEID] = 1,
+	[AID1_NODEID] = 1,
 	[XCD2_NODEID] = 2,
 	[XCD3_NODEID] = 3,
+	[AID2_NODEID] = 2,
 	[XCD4_NODEID] = 4,
 	[XCD5_NODEID] = 5,
+	[AID3_NODEID] = 3,
 	[XCD6_NODEID] = 6,
 	[XCD7_NODEID] = 7,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
index efe8a278cbdf9..04c0b4fa17a4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h
@@ -102,13 +102,17 @@ struct amdgpu_irq {
 	bool                            retry_cam_enabled;
 };
 
-enum interrupt_node_id_per_xcp {
+enum interrupt_node_id_per_aid {
+	AID0_NODEID = 0,
 	XCD0_NODEID = 1,
 	XCD1_NODEID = 2,
+	AID1_NODEID = 4,
 	XCD2_NODEID = 5,
 	XCD3_NODEID = 6,
+	AID2_NODEID = 8,
 	XCD4_NODEID = 9,
 	XCD5_NODEID = 10,
+	AID3_NODEID = 12,
 	XCD6_NODEID = 13,
 	XCD7_NODEID = 14,
 	NODEID_MAX,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 6935a24d1e89f..d3c7f9a43ef13 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1492,6 +1492,9 @@ static int sdma_v4_4_2_process_trap_irq(struct amdgpu_device *adev,
 
 	DRM_DEBUG("IH: SDMA trap\n");
 	instance = sdma_v4_4_2_irq_id_to_seq(entry->client_id);
+	instance += node_id_to_phys_map[entry->node_id] *
+			adev->sdma.num_inst_per_aid;
+
 	switch (entry->ring_id) {
 	case 0:
 		amdgpu_fence_process(&adev->sdma.instance[instance].ring);
-- 
GitLab


From 5de6bd6a13f1c717279c870eb8290e466c8f6a80 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 25 Feb 2022 15:47:20 +0800
Subject: [PATCH 1067/1791] drm/amdgpu: set mmhub bitmask for multiple AIDs

Like GFXHUB, set MMHUB0 bitmask for each AID.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 681bc9d354fee..59be0c0293c4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1730,7 +1730,7 @@ static int gmc_v9_0_sw_init(void *handle)
 		break;
 	case IP_VERSION(9, 4, 3):
 		bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), adev->gfx.num_xcd);
-		bitmap_set(adev->vmhubs_mask, AMDGPU_MMHUB0(0), 1);
+		bitmap_set(adev->vmhubs_mask, AMDGPU_MMHUB0(0), adev->num_aid);
 
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
 		break;
-- 
GitLab


From feb36dd01403689140a42f906fc75769d3158400 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 18 Mar 2022 16:46:04 +0800
Subject: [PATCH 1068/1791] drm/amdgpu: convert the doorbell_index to 2 dwords
 offset for kiq

KIQ doorbell_index is non-zero from XCC1, thus need to left-shift it like
other rings.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 682c157f2d8f9..49b18cf987aa3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -313,14 +313,13 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->adev = NULL;
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
-	ring->doorbell_index = adev->doorbell_index.kiq;
 	ring->xcc_id = xcc_id;
 	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
 	if (xcc_id >= 1)
-		ring->doorbell_index = adev->doorbell_index.xcc1_kiq_start +
-					xcc_id - 1;
+		ring->doorbell_index = (adev->doorbell_index.xcc1_kiq_start +
+					xcc_id - 1) << 1;
 	else
-		ring->doorbell_index = adev->doorbell_index.kiq;
+		ring->doorbell_index = adev->doorbell_index.kiq << 1;
 
 	r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id);
 	if (r)
-- 
GitLab


From 36be0181eab50abbb043a087988e6c2bef59dd45 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 20 Apr 2022 23:25:48 +0800
Subject: [PATCH 1069/1791] drm/amdgpu: program GRBM_MCM_ADDR for non-AID0 GRBM

Otherwise the EOP interrupt on non-AID0 cannot route to IH0.

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 1dcb69b4816f9..a9fab8de29e89 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -184,7 +184,10 @@ static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 
 static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
 {
+	int i;
 
+	for (i = 2; i < adev->gfx.num_xcd; i++)
+		WREG32_SOC15(GC, i, regGRBM_MCM_ADDR, 0x4);
 }
 
 static void gfx_v9_4_3_write_data_to_reg(struct amdgpu_ring *ring, int eng_sel,
-- 
GitLab


From 20bedf1379b1d4b060e3f43661f35e5447d0cfed Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 25 Apr 2022 22:19:58 +0800
Subject: [PATCH 1070/1791] drm/amdgpu: introduce new doorbell assignment table
 for GC 9.4.3

Four basic reasons as below to do the change:
  1. number of ring expand a lot on GC 9.4.3, and adjustment on old
     assignment cannot make each ring in a continuous doorbell space.
  2. the SDMA doorbell index should not exceed 0x1FF on SDMA 4.2.2 due to
     regDOORBELLx_CTRL_ENTRY.BIF_DOORBELLx_RANGE_OFFSET_ENTRY field width.
  3. re-design the doorbell assignment and unify the calculation as
     "start + ring/inst id" will make the code much concise.
  4. only defining the START/END makes the table look simple

v2: (Lijo)
  1. replace name
  2. use num_inst_per_aid/sdma_doorbell_range instead of hardcoding

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h | 32 +++++++++++++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c      |  6 +---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c      |  8 +----
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c     | 19 ++----------
 4 files changed, 33 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index b036d2f019300..a29a018ec84ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -59,7 +59,7 @@ struct amdgpu_doorbell_index {
 	uint32_t gfx_ring1;
 	uint32_t gfx_userqueue_start;
 	uint32_t gfx_userqueue_end;
-	uint32_t sdma_engine[8];
+	uint32_t sdma_engine[16];
 	uint32_t mes_ring0;
 	uint32_t mes_ring1;
 	uint32_t ih;
@@ -83,9 +83,6 @@ struct amdgpu_doorbell_index {
 	};
 	uint32_t first_non_cp;
 	uint32_t last_non_cp;
-	uint32_t xcc1_kiq_start;
-	uint32_t xcc1_mec_ring0_start;
-	uint32_t aid1_sdma_start;
 	uint32_t max_assignment;
 	/* Per engine SDMA doorbell size in dword */
 	uint32_t sdma_doorbell_range;
@@ -312,6 +309,33 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT
 	AMDGPU_DOORBELL64_INVALID                 = 0xFFFF
 } AMDGPU_DOORBELL64_ASSIGNMENT;
 
+typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1
+{
+	/* KIQ: 0~7 for maximum 8 XCD */
+	AMDGPU_DOORBELL_LAYOUT1_KIQ_START               = 0x000,
+	AMDGPU_DOORBELL_LAYOUT1_HIQ                     = 0x008,
+	AMDGPU_DOORBELL_LAYOUT1_DIQ                     = 0x009,
+	/* Compute: 0x0A ~ 0x49 */
+	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START          = 0x00A,
+	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_END            = 0x049,
+	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START         = 0x04A,
+	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END           = 0x0C9,
+	/* SDMA: 0x100 ~ 0x19F */
+	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START       = 0x100,
+	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END         = 0x19F,
+	/* IH: 0x1A0 ~ 0x1AF */
+	AMDGPU_DOORBELL_LAYOUT1_IH                      = 0x1A0,
+	/* VCN: 0x1B0 ~ 0x1C2 */
+	AMDGPU_DOORBELL_LAYOUT1_VCN_START               = 0x1B0,
+	AMDGPU_DOORBELL_LAYOUT1_VCN_END                 = 0x1C2,
+
+	AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP            = AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START,
+	AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP             = AMDGPU_DOORBELL_LAYOUT1_VCN_END,
+
+	AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT          = 0x1C2,
+	AMDGPU_DOORBELL_LAYOUT1_INVALID                 = 0xFFFF
+} AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1;
+
 u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index);
 void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v);
 u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 49b18cf987aa3..76438f197de1b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -315,11 +315,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->use_doorbell = true;
 	ring->xcc_id = xcc_id;
 	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
-	if (xcc_id >= 1)
-		ring->doorbell_index = (adev->doorbell_index.xcc1_kiq_start +
-					xcc_id - 1) << 1;
-	else
-		ring->doorbell_index = adev->doorbell_index.kiq << 1;
+	ring->doorbell_index = (adev->doorbell_index.kiq + xcc_id) << 1;
 
 	r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id);
 	if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index a9fab8de29e89..52185b1d5d310 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -817,13 +817,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
-	if (xcc_id >= 1)
-		ring->doorbell_index =
-				(adev->doorbell_index.xcc1_mec_ring0_start +
-				ring_id - adev->gfx.num_compute_rings) << 1;
-	else
-		ring->doorbell_index =
-				(adev->doorbell_index.mec_ring0 + ring_id) << 1;
+	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
 	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
 				+ (ring_id * GFX9_MEC_HPD_SIZE);
 	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index d3c7f9a43ef13..9b53174925f8c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1310,14 +1310,7 @@ static int sdma_v4_4_2_sw_init(void *handle)
 				ring->use_doorbell?"true":"false");
 
 		/* doorbell size is 2 dwords, get DWORD offset */
-		if (aid_id > 0)
-			ring->doorbell_index =
-				(adev->doorbell_index.aid1_sdma_start << 1)
-				+ adev->doorbell_index.sdma_doorbell_range
-				* (i - adev->sdma.num_inst_per_aid);
-		else
-			ring->doorbell_index =
-				adev->doorbell_index.sdma_engine[i] << 1;
+		ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1;
 		ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
 		sprintf(ring->name, "sdma%d.%d", aid_id,
@@ -1336,14 +1329,8 @@ static int sdma_v4_4_2_sw_init(void *handle)
 			/* doorbell index of page queue is assigned right after
 			 * gfx queue on the same instance
 			 */
-			if (aid_id > 0)
-				ring->doorbell_index =
-					((adev->doorbell_index.aid1_sdma_start + 1) << 1)
-					+ adev->doorbell_index.sdma_doorbell_range
-					* (i - adev->sdma.num_inst_per_aid);
-			else
-				ring->doorbell_index =
-					(adev->doorbell_index.sdma_engine[i] + 1) << 1;
+			ring->doorbell_index =
+				(adev->doorbell_index.sdma_engine[i] + 1) << 1;
 			ring->vm_hub = AMDGPU_MMHUB0(aid_id);
 
 			sprintf(ring->name, "page%d.%d", aid_id,
-- 
GitLab


From 92085240ef9c0ec60c27a60b3cc0d4f5266fa511 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 3 May 2022 10:16:46 -0400
Subject: [PATCH 1071/1791] drm/amdkfd: add gpu compute cores io links for
 gfx9.4.3

The PSP TA will only provide xGMI topology info for links between GPU
sockets so links between partitions from different sockets will be
hardcoded as 3 xGMI hops with 1 hops weighted as xGMI and 2 hops
weighted with a new intra-socket weight to indicate the longest
possible distance.

If the link between a partition and the CPU is non-PCIe, then assume
the CPU (CCDs) is located within the same socket as the partition
and represent the link as an intra-socket weighted single hop XGMI link
with memory bandwidth.

Links between partitions within a single socket will be abstracted as
single hop xGMI links weighted with the new intra-socket weight and
will have memory bandwidth.

Finally, use the unused function bits in the location ID to represent the
coordinates of the compute partition within its socket.

A follow on patch will resolve the requirement for GPU socket xGMI
link representation sometime later.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c     | 49 ++++++++++++++++-------
 drivers/gpu/drm/amd/amdkfd/kfd_crat.h     |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c   |  8 ++++
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c |  4 ++
 4 files changed, 47 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index f5aebba31e886..dc93a67257e10 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -1166,7 +1166,7 @@ static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink,
 			if (props->iolink_type == CRAT_IOLINK_TYPE_PCIEXPRESS)
 				props->weight = 20;
 			else if (props->iolink_type == CRAT_IOLINK_TYPE_XGMI)
-				props->weight = 15 * iolink->num_hops_xgmi;
+				props->weight = iolink->weight_xgmi;
 			else
 				props->weight = node_distance(id_from, id_to);
 
@@ -1972,6 +1972,9 @@ static void kfd_find_numa_node_in_srat(struct kfd_node *kdev)
 }
 #endif
 
+#define KFD_CRAT_INTRA_SOCKET_WEIGHT	13
+#define KFD_CRAT_XGMI_WEIGHT		15
+
 /* kfd_fill_gpu_direct_io_link - Fill in direct io link from GPU
  * to its NUMA node
  *	@avail_size: Available size in the memory
@@ -2003,6 +2006,12 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
 	 * TODO: Fill-in other fields of iolink subtype
 	 */
 	if (kdev->adev->gmc.xgmi.connected_to_cpu) {
+		bool ext_cpu = KFD_GC_VERSION(kdev) != IP_VERSION(9, 4, 3);
+		int mem_bw = 819200, weight = ext_cpu ? KFD_CRAT_XGMI_WEIGHT :
+							KFD_CRAT_INTRA_SOCKET_WEIGHT;
+		uint32_t bandwidth = ext_cpu ? amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(
+							kdev->adev, NULL, true) : mem_bw;
+
 		/*
 		 * with host gpu xgmi link, host can access gpu memory whether
 		 * or not pcie bar type is large, so always create bidirectional
@@ -2010,14 +2019,9 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
 		 */
 		sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL;
 		sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI;
-		sub_type_hdr->num_hops_xgmi = 1;
-		if (KFD_GC_VERSION(kdev) == IP_VERSION(9, 4, 2)) {
-			sub_type_hdr->minimum_bandwidth_mbs =
-					amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(
-							kdev->adev, NULL, true);
-			sub_type_hdr->maximum_bandwidth_mbs =
-					sub_type_hdr->minimum_bandwidth_mbs;
-		}
+		sub_type_hdr->weight_xgmi = weight;
+		sub_type_hdr->minimum_bandwidth_mbs = bandwidth;
+		sub_type_hdr->maximum_bandwidth_mbs = bandwidth;
 	} else {
 		sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_PCIEXPRESS;
 		sub_type_hdr->minimum_bandwidth_mbs =
@@ -2050,6 +2054,8 @@ static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size,
 			uint32_t proximity_domain_from,
 			uint32_t proximity_domain_to)
 {
+	bool use_ta_info = kdev->kfd->num_nodes == 1;
+
 	*avail_size -= sizeof(struct crat_subtype_iolink);
 	if (*avail_size < 0)
 		return -ENOMEM;
@@ -2064,12 +2070,25 @@ static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size,
 	sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI;
 	sub_type_hdr->proximity_domain_from = proximity_domain_from;
 	sub_type_hdr->proximity_domain_to = proximity_domain_to;
-	sub_type_hdr->num_hops_xgmi =
-		amdgpu_amdkfd_get_xgmi_hops_count(kdev->adev, peer_kdev->adev);
-	sub_type_hdr->maximum_bandwidth_mbs =
-		amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, peer_kdev->adev, false);
-	sub_type_hdr->minimum_bandwidth_mbs = sub_type_hdr->maximum_bandwidth_mbs ?
-		amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, NULL, true) : 0;
+
+	if (use_ta_info) {
+		sub_type_hdr->weight_xgmi = KFD_CRAT_XGMI_WEIGHT *
+			amdgpu_amdkfd_get_xgmi_hops_count(kdev->adev, peer_kdev->adev);
+		sub_type_hdr->maximum_bandwidth_mbs =
+			amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev,
+							peer_kdev->adev, false);
+		sub_type_hdr->minimum_bandwidth_mbs = sub_type_hdr->maximum_bandwidth_mbs ?
+			amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, NULL, true) : 0;
+	} else {
+		bool is_single_hop = kdev->kfd == peer_kdev->kfd;
+		int weight = is_single_hop ? KFD_CRAT_INTRA_SOCKET_WEIGHT :
+			(2 * KFD_CRAT_INTRA_SOCKET_WEIGHT) + KFD_CRAT_XGMI_WEIGHT;
+		int mem_bw = 819200;
+
+		sub_type_hdr->weight_xgmi = weight;
+		sub_type_hdr->maximum_bandwidth_mbs = is_single_hop ? mem_bw : 0;
+		sub_type_hdr->minimum_bandwidth_mbs = is_single_hop ? mem_bw : 0;
+	}
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
index 3d0e533b93b9e..fc719389b5d65 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
@@ -275,7 +275,7 @@ struct crat_subtype_iolink {
 	uint32_t	maximum_bandwidth_mbs;
 	uint32_t	recommended_transfer_size;
 	uint8_t		reserved2[CRAT_IOLINK_RESERVED_LENGTH - 1];
-	uint8_t		num_hops_xgmi;
+	uint8_t		weight_xgmi;
 };
 
 /*
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index ec5f85ff34e51..971a3aa3294ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -702,6 +702,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	if (amdgpu_use_xgmi_p2p)
 		kfd->hive_id = kfd->adev->gmc.xgmi.hive_id;
 
+	/*
+	 * For GFX9.4.3, the KFD abstracts all partitions within a socket as
+	 * xGMI connected in the topology so assign a unique hive id per
+	 * device based on the pci device location if device is in PCIe mode.
+	 */
+	if (!kfd->hive_id && (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3)) && kfd->num_nodes > 1)
+		kfd->hive_id = pci_dev_id(kfd->adev->pdev);
+
 	kfd->noretry = kfd->adev->gmc.noretry;
 
 	/* If CRAT is broken, won't set iommu enabled */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 6d958bf0fe908..d3e70341dfadf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1926,7 +1926,11 @@ int kfd_topology_add_device(struct kfd_node *gpu)
 	dev->node_props.capability |=
 		((dev->gpu->adev->rev_id << HSA_CAP_ASIC_REVISION_SHIFT) &
 			HSA_CAP_ASIC_REVISION_MASK);
+
 	dev->node_props.location_id = pci_dev_id(gpu->adev->pdev);
+	if (KFD_GC_VERSION(dev->gpu->kfd) == IP_VERSION(9, 4, 3))
+		dev->node_props.location_id |= dev->gpu->node_id;
+
 	dev->node_props.domain = pci_domain_nr(gpu->adev->pdev->bus);
 	dev->node_props.max_engine_clk_fcompute =
 		amdgpu_amdkfd_get_max_engine_clock_in_mhz(dev->gpu->adev);
-- 
GitLab


From 643e40d4c06f8c887af1789c7bf8d279e9c8e4cf Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 31 May 2022 16:25:16 -0400
Subject: [PATCH 1072/1791] drm/amdkfd: Fix SDMA in CPX mode

When creating a user-mode SDMA queue, CP FW expects
driver to use/set virtual SDMA engine id in MAP_QUEUES
packet instead of using the physical SDMA engine id.
Each partition node's virtual SDMA number should start
from 0. However, when allocating doorbell for the queue,
KFD needs to allocate the doorbell from doorbell space
corresponding to the physical SDMA engine id, otherwise
the hwardware will not see the doorbell press.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 69419a53a14e5..2b5c4b2dd2423 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -363,7 +363,16 @@ static int allocate_doorbell(struct qcm_process_device *qpd,
 		 */
 
 		uint32_t *idx_offset = dev->kfd->shared_resources.sdma_doorbell_idx;
-		uint32_t valid_id = idx_offset[q->properties.sdma_engine_id]
+
+		/*
+		 * q->properties.sdma_engine_id corresponds to the virtual
+		 * sdma engine number. However, for doorbell allocation,
+		 * we need the physical sdma engine id in order to get the
+		 * correct doorbell offset.
+		 */
+		uint32_t valid_id = idx_offset[qpd->dqm->dev->node_id *
+					       get_num_all_sdma_engines(qpd->dqm) +
+					       q->properties.sdma_engine_id]
 						+ (q->properties.sdma_queue_id & 1)
 						* KFD_QUEUE_DOORBELL_MIRROR_OFFSET
 						+ (q->properties.sdma_queue_id >> 1);
@@ -1388,7 +1397,6 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 		}
 
 		q->properties.sdma_engine_id =
-			dqm->dev->node_id * get_num_all_sdma_engines(dqm) +
 			q->sdma_id % kfd_get_num_sdma_engines(dqm->dev);
 		q->properties.sdma_queue_id = q->sdma_id /
 				kfd_get_num_sdma_engines(dqm->dev);
@@ -1418,7 +1426,6 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm,
 		 * PCIe-optimized ones
 		 */
 		q->properties.sdma_engine_id =
-			dqm->dev->node_id * get_num_all_sdma_engines(dqm) +
 			kfd_get_num_sdma_engines(dqm->dev) +
 			q->sdma_id % kfd_get_num_xgmi_sdma_engines(dqm->dev);
 		q->properties.sdma_queue_id = q->sdma_id /
@@ -2486,6 +2493,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 	int pipe, queue;
 	int r = 0, xcc;
 	uint32_t inst;
+	uint32_t sdma_engine_start;
 
 	if (!dqm->sched_running) {
 		seq_puts(m, " Device is stopped\n");
@@ -2530,7 +2538,10 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 		}
 	}
 
-	for (pipe = 0; pipe < get_num_all_sdma_engines(dqm); pipe++) {
+	sdma_engine_start = dqm->dev->node_id * get_num_all_sdma_engines(dqm);
+	for (pipe = sdma_engine_start;
+	     pipe < (sdma_engine_start + get_num_all_sdma_engines(dqm));
+	     pipe++) {
 		for (queue = 0;
 		     queue < dqm->dev->kfd->device_info.num_sdma_queues_per_engine;
 		     queue++) {
-- 
GitLab


From 1bd6dd21fcd53ac78a9018b96699ef1aa99a3e59 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 31 May 2022 16:31:28 -0400
Subject: [PATCH 1073/1791] drm/amdkfd: Add SDMA info for SDMA 4.4.2

Update SDMA queue information for SDMA 4.4.2.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 971a3aa3294ac..8e5d785b88243 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -82,6 +82,7 @@ static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd)
 	case IP_VERSION(4, 2, 0):/* VEGA20 */
 	case IP_VERSION(4, 2, 2):/* ARCTURUS */
 	case IP_VERSION(4, 4, 0):/* ALDEBARAN */
+	case IP_VERSION(4, 4, 2):
 	case IP_VERSION(5, 0, 0):/* NAVI10 */
 	case IP_VERSION(5, 0, 1):/* CYAN_SKILLFISH */
 	case IP_VERSION(5, 0, 2):/* NAVI14 */
-- 
GitLab


From d1d22df174ae512c57374f517e346b608f61555c Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 7 Jun 2022 14:46:18 -0400
Subject: [PATCH 1074/1791] drm/amdkfd: Populate memory info before adding GPU
 node to topology

The local memory info needs to be fetched before the GPU node is added
to topology. Without this, the sysfs is incorrectly populated and the
size is reported as 0. This was causing rocr tests to fail. This issue
was caused because of a bad merge.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 8e5d785b88243..829e32433faf2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -743,6 +743,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	node->max_proc_per_quantum = max_proc_per_quantum;
 	atomic_set(&node->sram_ecc_flag, 0);
 
+	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
+
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
 				kfd->num_nodes);
 	for (i = 0; i < kfd->num_nodes; i++) {
@@ -793,8 +795,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	if (kfd_resume_iommu(kfd))
 		goto kfd_resume_iommu_error;
 
-	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
-
 	kfd->init_complete = true;
 	dev_info(kfd_device, "added device %x:%x\n", kfd->adev->pdev->vendor,
 		 kfd->adev->pdev->device);
-- 
GitLab


From cd8d77f328c53aad8915c9c4d64cf557742bb257 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Fri, 27 May 2022 13:47:24 +0800
Subject: [PATCH 1075/1791] drm/amdgpu: add new vram type for dgpu

hbm3 will be supported in some dgpu program

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c | 1 +
 drivers/gpu/drm/amd/include/atomfirmware.h       | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
index ac6fe0ae4609f..ef4b9a41f20ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c
@@ -272,6 +272,7 @@ static int convert_atom_mem_type_to_vram_type(struct amdgpu_device *adev,
 			break;
 		case ATOM_DGPU_VRAM_TYPE_HBM2:
 		case ATOM_DGPU_VRAM_TYPE_HBM2E:
+		case ATOM_DGPU_VRAM_TYPE_HBM3:
 			vram_type = AMDGPU_VRAM_TYPE_HBM;
 			break;
 		case ATOM_DGPU_VRAM_TYPE_GDDR6:
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index bbe1337a8cee3..e68c1e280322a 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -182,6 +182,7 @@ enum atom_dgpu_vram_type {
   ATOM_DGPU_VRAM_TYPE_HBM2  = 0x60,
   ATOM_DGPU_VRAM_TYPE_HBM2E = 0x61,
   ATOM_DGPU_VRAM_TYPE_GDDR6 = 0x70,
+  ATOM_DGPU_VRAM_TYPE_HBM3 = 0x80,
 };
 
 enum atom_dp_vs_preemph_def{
-- 
GitLab


From 9eb7681f760c77adece36bc62953245c9f44a3be Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 21 Feb 2022 15:38:39 +0800
Subject: [PATCH 1076/1791] drm/amdgpu: add the support of XGMI link for GC
 9.4.3

Add the xgmi LFB_CNTL/LBF_SIZE reg addresses to fetch the xgmi info from.

v2: move get_xgmi_info() to GC_V9_4_3 sepecific source files to utilize
the register definitions specific for GC_V9_4_3
v3: remove the duplicated register definitions
v4: enable xgmi based on asic_type as XGMI_IP ver is not available
yet for IP discovery

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Ack-by: Lijo Lazar <Lijo.Lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 41 +++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c    | 10 ++++--
 2 files changed, 47 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index d3424ce97aa89..1bb17d95f7208 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -528,6 +528,45 @@ static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 	}
 }
 
+static int gfxhub_v1_2_get_xgmi_info(struct amdgpu_device *adev)
+{
+	u32 max_num_physical_nodes;
+	u32 max_physical_node_id;
+	u32 xgmi_lfb_cntl;
+	u32 max_region;
+	u64 seg_size;
+
+	xgmi_lfb_cntl = RREG32_SOC15(GC, 0, regMC_VM_XGMI_LFB_CNTL);
+	seg_size = REG_GET_FIELD(
+		RREG32_SOC15(GC, 0, regMC_VM_XGMI_LFB_SIZE),
+		MC_VM_XGMI_LFB_SIZE, PF_LFB_SIZE) << 24;
+	max_region =
+		REG_GET_FIELD(xgmi_lfb_cntl, MC_VM_XGMI_LFB_CNTL, PF_MAX_REGION);
+
+
+
+	max_num_physical_nodes   = 8;
+	max_physical_node_id     = 7;
+
+	/* PF_MAX_REGION=0 means xgmi is disabled */
+	if (max_region || adev->gmc.xgmi.connected_to_cpu) {
+		adev->gmc.xgmi.num_physical_nodes = max_region + 1;
+
+		if (adev->gmc.xgmi.num_physical_nodes > max_num_physical_nodes)
+			return -EINVAL;
+
+		adev->gmc.xgmi.physical_node_id =
+			REG_GET_FIELD(xgmi_lfb_cntl, MC_VM_XGMI_LFB_CNTL,
+					PF_LFB_REGION);
+
+		if (adev->gmc.xgmi.physical_node_id > max_physical_node_id)
+			return -EINVAL;
+
+		adev->gmc.xgmi.node_segment_size = seg_size;
+	}
+
+	return 0;
+}
 
 const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = {
 	.get_mc_fb_offset = gfxhub_v1_2_get_mc_fb_offset,
@@ -536,5 +575,5 @@ const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = {
 	.gart_disable = gfxhub_v1_2_gart_disable,
 	.set_fault_enable_default = gfxhub_v1_2_set_fault_enable_default,
 	.init = gfxhub_v1_2_init,
-	.get_xgmi_info = gfxhub_v1_1_get_xgmi_info,
+	.get_xgmi_info = gfxhub_v1_2_get_xgmi_info,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 59be0c0293c4d..4b2c4ecd72538 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1419,9 +1419,13 @@ static int gmc_v9_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	/* ARCT and VEGA20 don't have XGMI defined in their IP discovery tables */
-	if (adev->asic_type == CHIP_VEGA20 ||
-	    adev->asic_type == CHIP_ARCTURUS)
+	/*
+	 * 9.4.0, 9.4.1 and 9.4.3 don't have XGMI defined
+	 * in their IP discovery tables
+	 */
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 0) ||
+	    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 1) ||
+	    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
 		adev->gmc.xgmi.supported = true;
 
 	if (adev->ip_versions[XGMI_HWIP][0] == IP_VERSION(6, 1, 0)) {
-- 
GitLab


From 8078f1c610fdcdd8003e2c538fb04af41fa5c269 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 11:41:53 +0530
Subject: [PATCH 1077/1791] drm/amdgpu: Change num_xcd to xcc_mask

Instead of number of XCCs, keep a mask of XCCs for the exact XCCs
available on the ASIC. XCC configuration could differ based on
different ASIC configs.

v2:
	Rename num_xcd to num_xcc (Hawking)
	Use smaller xcc_mask size, changed to u16 (Le)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c  |  21 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h  |   4 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c    |   2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c  | 133 +++++++++++++----------
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c |  67 +++++++-----
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c    |   3 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c  |  10 +-
 7 files changed, 141 insertions(+), 99 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 76438f197de1b..069b259f384c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -209,12 +209,12 @@ void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev)
 	int max_queues_per_mec = min(adev->gfx.mec.num_pipe_per_mec *
 				     adev->gfx.mec.num_queue_per_pipe,
 				     adev->gfx.num_compute_rings);
-	int num_xcd = (adev->gfx.num_xcd > 1) ? adev->gfx.num_xcd : 1;
+	int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1;
 
 	if (multipipe_policy) {
 		/* policy: make queues evenly cross all pipes on MEC1 only
 		 * for multiple xcc, just use the original policy for simplicity */
-		for (j = 0; j < num_xcd; j++) {
+		for (j = 0; j < num_xcc; j++) {
 			for (i = 0; i < max_queues_per_mec; i++) {
 				pipe = i % adev->gfx.mec.num_pipe_per_mec;
 				queue = (i / adev->gfx.mec.num_pipe_per_mec) %
@@ -226,13 +226,13 @@ void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev)
 		}
 	} else {
 		/* policy: amdgpu owns all queues in the given pipe */
-		for (j = 0; j < num_xcd; j++) {
+		for (j = 0; j < num_xcc; j++) {
 			for (i = 0; i < max_queues_per_mec; ++i)
 				set_bit(i, adev->gfx.mec_bitmap[j].queue_bitmap);
 		}
 	}
 
-	for (j = 0; j < num_xcd; j++) {
+	for (j = 0; j < num_xcc; j++) {
 		dev_dbg(adev->dev, "mec queue bitmap weight=%d\n",
 			bitmap_weight(adev->gfx.mec_bitmap[j].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES));
 	}
@@ -1207,23 +1207,24 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = drm_to_adev(ddev);
 	enum amdgpu_gfx_partition mode;
-	int ret;
+	int ret = 0, num_xcc;
 
-	if (adev->gfx.num_xcd % 2 != 0)
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	if (num_xcc % 2 != 0)
 		return -EINVAL;
 
 	if (!strncasecmp("SPX", buf, strlen("SPX"))) {
 		mode = AMDGPU_SPX_PARTITION_MODE;
 	} else if (!strncasecmp("DPX", buf, strlen("DPX"))) {
-		if (adev->gfx.num_xcd != 4 || adev->gfx.num_xcd != 8)
+		if (num_xcc != 4 || num_xcc != 8)
 			return -EINVAL;
 		mode = AMDGPU_DPX_PARTITION_MODE;
 	} else if (!strncasecmp("TPX", buf, strlen("TPX"))) {
-		if (adev->gfx.num_xcd != 6)
+		if (num_xcc != 6)
 			return -EINVAL;
 		mode = AMDGPU_TPX_PARTITION_MODE;
 	} else if (!strncasecmp("QPX", buf, strlen("QPX"))) {
-		if (adev->gfx.num_xcd != 8)
+		if (num_xcc != 8)
 			return -EINVAL;
 		mode = AMDGPU_QPX_PARTITION_MODE;
 	} else if (!strncasecmp("CPX", buf, strlen("CPX"))) {
@@ -1253,7 +1254,7 @@ static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev,
 	char *supported_partition;
 
 	/* TBD */
-	switch (adev->gfx.num_xcd) {
+	switch (NUM_XCC(adev->gfx.xcc_mask)) {
 	case 8:
 		supported_partition = "SPX, DPX, QPX, CPX";
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 8df36527aee94..93f9875154dbb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -64,6 +64,8 @@ enum amdgpu_gfx_partition {
 	AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE,
 };
 
+#define NUM_XCC(x) hweight16(x)
+
 struct amdgpu_mec {
 	struct amdgpu_bo	*hpd_eop_obj;
 	u64			hpd_eop_gpu_addr;
@@ -396,7 +398,7 @@ struct amdgpu_gfx {
 	bool				cp_gfx_shadow; /* for gfx11 */
 
 	enum amdgpu_gfx_partition	partition_mode;
-	uint32_t			num_xcd;
+	uint16_t 			xcc_mask;
 	uint32_t			num_xcc_per_xcp;
 	struct mutex			partition_mutex;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 91814dc083c9e..da69177dc76f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4536,7 +4536,7 @@ static int gfx_v9_0_early_init(void *handle)
 		adev->gfx.num_gfx_rings = 0;
 	else
 		adev->gfx.num_gfx_rings = GFX9_NUM_GFX_RINGS;
-	adev->gfx.num_xcd = 1;
+	adev->gfx.xcc_mask = 1;
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
 	gfx_v9_0_set_kiq_pm4_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 52185b1d5d310..c776fc5884def 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -177,16 +177,19 @@ static const struct kiq_pm4_funcs gfx_v9_4_3_kiq_pm4_funcs = {
 
 static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 {
-	int i;
-	for (i = 0; i < adev->gfx.num_xcd; i++)
+	int i, num_xcc;
+
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++)
 		adev->gfx.kiq[i].pmf = &gfx_v9_4_3_kiq_pm4_funcs;
 }
 
 static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
 {
-	int i;
+	int i, num_xcc;
 
-	for (i = 2; i < adev->gfx.num_xcd; i++)
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 2; i < num_xcc; i++)
 		WREG32_SOC15(GC, i, regGRBM_MCM_ADDR, 0x4);
 }
 
@@ -499,7 +502,7 @@ static void gfx_v9_4_3_mec_fini(struct amdgpu_device *adev)
 
 static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 {
-	int r, i;
+	int r, i, num_xcc;
 	u32 *hpd;
 	const __le32 *fw_data;
 	unsigned fw_size;
@@ -508,7 +511,8 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 
 	const struct gfx_firmware_header_v1_0 *mec_hdr;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++)
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++)
 		bitmap_zero(adev->gfx.mec_bitmap[i].queue_bitmap,
 			AMDGPU_MAX_COMPUTE_QUEUES);
 
@@ -683,23 +687,24 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 						enum amdgpu_gfx_partition mode)
 {
 	u32 tmp = 0;
-	int num_xcc_per_partition, i;
+	int num_xcc_per_partition, i, num_xcc;
 
 	if (mode == adev->gfx.partition_mode)
 		return mode;
 
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
-		num_xcc_per_partition = adev->gfx.num_xcd;
+		num_xcc_per_partition = num_xcc;
 		break;
 	case AMDGPU_DPX_PARTITION_MODE:
-		num_xcc_per_partition = adev->gfx.num_xcd / 2;
+		num_xcc_per_partition = num_xcc / 2;
 		break;
 	case AMDGPU_TPX_PARTITION_MODE:
-		num_xcc_per_partition = adev->gfx.num_xcd / 3;
+		num_xcc_per_partition = num_xcc / 3;
 		break;
 	case AMDGPU_QPX_PARTITION_MODE:
-		num_xcc_per_partition = adev->gfx.num_xcd / 4;
+		num_xcc_per_partition = num_xcc / 4;
 		break;
 	case AMDGPU_CPX_PARTITION_MODE:
 		num_xcc_per_partition = 1;
@@ -712,7 +717,7 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 	 * Stop user queues and threads, and make sure GPU is empty of work.
 	 */
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	for (i = 0; i < num_xcc; i++) {
 		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP,
 				    num_xcc_per_partition);
 		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID,
@@ -836,7 +841,7 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 
 static int gfx_v9_4_3_sw_init(void *handle)
 {
-	int i, j, k, r, ring_id, xcc_id;
+	int i, j, k, r, ring_id, xcc_id, num_xcc;
 	struct amdgpu_kiq *kiq;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
@@ -844,6 +849,8 @@ static int gfx_v9_4_3_sw_init(void *handle)
 	adev->gfx.mec.num_pipe_per_mec = 4;
 	adev->gfx.mec.num_queue_per_pipe = 8;
 
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+
 	/* EOP Event */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_EOP_INTERRUPT, &adev->gfx.eop_irq);
 	if (r)
@@ -877,8 +884,7 @@ static int gfx_v9_4_3_sw_init(void *handle)
 
 	/* set up the compute queues - allocate horizontally across pipes */
 	ring_id = 0;
-	for (xcc_id = 0; xcc_id < adev->gfx.num_xcd; xcc_id++) {
-
+	for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) {
 		for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 			for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 				for (k = 0; k < adev->gfx.mec.num_pipe_per_mec;
@@ -930,14 +936,14 @@ static int gfx_v9_4_3_sw_init(void *handle)
 
 static int gfx_v9_4_3_sw_fini(void *handle)
 {
-	int i;
+	int i, num_xcc;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	for (i = 0; i < adev->gfx.num_compute_rings *
-		adev->gfx.num_xcd; i++)
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < adev->gfx.num_compute_rings * num_xcc; i++)
 		amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	for (i = 0; i < num_xcc; i++) {
 		amdgpu_gfx_mqd_sw_fini(adev, i);
 		amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[i].ring);
 		amdgpu_gfx_kiq_fini(adev, i);
@@ -1050,9 +1056,10 @@ static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 {
 	u32 tmp;
-	int i, j;
+	int i, j, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		WREG32_FIELD15_PREREG(GC, i, GRBM_CNTL, READ_TIMEOUT, 0xff);
 		gfx_v9_4_3_setup_rb(adev, i);
 	}
@@ -1064,7 +1071,7 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
-		for (j = 0; j < adev->gfx.num_xcd; j++) {
+		for (j = 0; j < num_xcc; j++) {
 			soc15_grbm_select(adev, 0, 0, 0, i, j);
 			/* CP and shaders */
 			if (i == 0) {
@@ -1092,7 +1099,7 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 
 	mutex_unlock(&adev->srbm_mutex);
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	for (i = 0; i < num_xcc; i++) {
 		gfx_v9_4_3_init_compute_vmid(adev, i);
 		gfx_v9_4_3_init_gds_vmid(adev, i);
 	}
@@ -1150,8 +1157,10 @@ static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t tmp = 0;
+	int num_xcc;
 
-	switch (adev->gfx.num_xcd) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	switch (num_xcc) {
 	/* directly config VIRTUAL_XCC_ID to 0 for 1-XCC */
 	case 1:
 		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, 0x8);
@@ -1288,9 +1297,10 @@ static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
 
 static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 {
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 0);
 		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
 		gfx_v9_4_3_wait_for_rlc_serdes(adev, i);
@@ -1299,9 +1309,10 @@ static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 
 static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev)
 {
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
 		udelay(50);
 		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
@@ -1314,9 +1325,10 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 #ifdef AMDGPU_RLC_DEBUG_RETRY
 	u32 rlc_ucode_ver;
 #endif
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 1);
 		udelay(50);
 
@@ -1377,11 +1389,12 @@ static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev, int xcc_id)
 
 static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 {
-	int r, i;
+	int r, i, num_xcc;
 
 	adev->gfx.rlc.funcs->stop(adev);
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		/* disable CG */
 		WREG32_SOC15(GC, i, regRLC_CGCG_CGLS_CTRL, 0);
 
@@ -1954,10 +1967,11 @@ static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev, int xcc_id)
 
 static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 {
-	int r, i, j;
+	int r, i, j, num_xcc;
 	struct amdgpu_ring *ring;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
 
 		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
@@ -2021,12 +2035,13 @@ static int gfx_v9_4_3_hw_init(void *handle)
 static int gfx_v9_4_3_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int i;
+	int i, num_xcc;
 
 	amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
 	amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		if (amdgpu_gfx_disable_kcq(adev, i))
 			DRM_ERROR("XCD %d KCQ disable failed\n", i);
 
@@ -2069,9 +2084,10 @@ static int gfx_v9_4_3_resume(void *handle)
 static bool gfx_v9_4_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		if (REG_GET_FIELD(RREG32_SOC15(GC, i, regGRBM_STATUS),
 					GRBM_STATUS, GUI_ACTIVE))
 			return false;
@@ -2183,30 +2199,30 @@ static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring,
 static int gfx_v9_4_3_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int num_xcc;
 
-	/* hardcode in emulation phase */
-	adev->gfx.num_xcd = 1;
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
 	adev->gfx.partition_mode = amdgpu_user_partt_mode;
 	/* calculate the num_xcc_in_xcp for the partition mode*/
 	switch (amdgpu_user_partt_mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd;
+		adev->gfx.num_xcc_per_xcp = num_xcc;
 		break;
 	case AMDGPU_DPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 2;
+		adev->gfx.num_xcc_per_xcp = num_xcc / 2;
 		break;
 	case AMDGPU_TPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 3;
+		adev->gfx.num_xcc_per_xcp = num_xcc / 3;
 		break;
 	case AMDGPU_QPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd / 4;
+		adev->gfx.num_xcc_per_xcp = num_xcc / 4;
 		break;
 	case AMDGPU_CPX_PARTITION_MODE:
 		adev->gfx.num_xcc_per_xcp = 1;
 		break;
 	default:
-		adev->gfx.num_xcc_per_xcp = adev->gfx.num_xcd;
+		adev->gfx.num_xcc_per_xcp = num_xcc;
 		break;
 	}
 
@@ -2404,14 +2420,15 @@ static int gfx_v9_4_3_set_clockgating_state(void *handle,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int i;
+	int i, num_xcc;
 
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 3):
-		for (i = 0; i < adev->gfx.num_xcd; i++)
+		for (i = 0; i < num_xcc; i++)
 			gfx_v9_4_3_update_gfx_clock_gating(adev,
 						state == AMD_CG_STATE_GATE, i);
 		break;
@@ -2739,12 +2756,13 @@ static int gfx_v9_4_3_set_priv_reg_fault_state(struct amdgpu_device *adev,
 					     unsigned type,
 					     enum amdgpu_interrupt_state state)
 {
-	int i;
+	int i, num_xcc;
 
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
-		for (i = 0; i < adev->gfx.num_xcd; i++)
+		for (i = 0; i < num_xcc; i++)
 			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
 				PRIV_REG_INT_ENABLE,
 				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
@@ -2761,12 +2779,13 @@ static int gfx_v9_4_3_set_priv_inst_fault_state(struct amdgpu_device *adev,
 					      unsigned type,
 					      enum amdgpu_interrupt_state state)
 {
-	int i;
+	int i, num_xcc;
 
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
-		for (i = 0; i < adev->gfx.num_xcd; i++)
+		for (i = 0; i < num_xcc; i++)
 			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
 				PRIV_INSTR_INT_ENABLE,
 				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
@@ -2783,8 +2802,10 @@ static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
 					    unsigned type,
 					    enum amdgpu_interrupt_state state)
 {
-	int i;
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	int i, num_xcc;
+
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		switch (type) {
 		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP:
 			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 0, state, i);
@@ -2842,6 +2863,7 @@ static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 			/* Per-queue interrupt is supported for MEC starting from VI.
 			  * The interrupt can only be enabled/disabled per pipe instead of per queue.
 			  */
+
 			if ((ring->me == me_id) && (ring->pipe == pipe_id) && (ring->queue == queue_id))
 				amdgpu_fence_process(ring);
 		}
@@ -3056,9 +3078,10 @@ static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_kiq = {
 
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev)
 {
-	int i, j;
+	int i, j, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		adev->gfx.kiq[i].ring.funcs = &gfx_v9_4_3_ring_funcs_kiq;
 
 		for (j = 0; j < adev->gfx.num_compute_rings; j++)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index 1bb17d95f7208..e35365ab3f1fb 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -43,9 +43,10 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 					 uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 		WREG32_SOC15_OFFSET(GC, i,
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
@@ -56,13 +57,14 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
 				    hub->ctx_addr_distance * vmid,
 				    upper_32_bits(page_table_base));
+
 	}
 }
 
 static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t pt_base;
-	int i;
+	int i, num_xcc;
 
 	if (adev->gmc.pdb0_bo)
 		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
@@ -74,7 +76,8 @@ static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 	/* If use GART for FB translation, vmid0 page table covers both
 	 * vram and system memory (gart)
 	 */
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		if (adev->gmc.pdb0_bo) {
 			WREG32_SOC15(GC, i,
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
@@ -111,9 +114,10 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t value;
 	uint32_t tmp;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		/* Program the AGP BAR */
 		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BASE, 0);
 		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
@@ -177,9 +181,10 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		/* Setup TLB control */
 		tmp = RREG32_SOC15(GC, i, regMC_VM_MX_L1_TLB_CNTL);
 
@@ -202,9 +207,10 @@ static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		/* Setup L2 cache */
 		tmp = RREG32_SOC15(GC, i, regVM_L2_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
@@ -249,9 +255,10 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		tmp = RREG32_SOC15(GC, i, regVM_CONTEXT0_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
@@ -266,9 +273,10 @@ static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 
 static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 {
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		WREG32_SOC15(GC, i,
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
 			     0XFFFFFFFF);
@@ -295,7 +303,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	struct amdgpu_vmhub *hub;
 	unsigned num_level, block_size;
 	uint32_t tmp;
-	int i, j;
+	int i, j, num_xcc;
 
 	num_level = adev->vm_manager.num_level;
 	block_size = adev->vm_manager.block_size;
@@ -304,7 +312,8 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	else
 		block_size -= 9;
 
-	for (j = 0; j < adev->gfx.num_xcd; j++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (j = 0; j < num_xcc; j++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		for (i = 0; i <= 14; i++) {
 			tmp = RREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL, i);
@@ -362,10 +371,12 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
-	unsigned i, j;
+	unsigned i, j, num_xcc;
 
-	for (j = 0; j < adev->gfx.num_xcd; j++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (j = 0; j < num_xcc; j++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
+
 		for (i = 0 ; i < 18; ++i) {
 			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
 					    i * hub->eng_addr_distance, 0xffffffff);
@@ -377,9 +388,10 @@ static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 
 static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 {
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		if (amdgpu_sriov_vf(adev)) {
 		/*
 		 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are
@@ -413,9 +425,10 @@ static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
 	u32 tmp;
-	u32 i, j;
+	u32 i, j, num_xcc;
 
-	for (j = 0; j < adev->gfx.num_xcd; j++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (j = 0; j < num_xcc; j++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		/* Disable all tables */
 		for (i = 0; i < 16; i++)
@@ -449,9 +462,10 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 						 bool value)
 {
 	u32 tmp;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		tmp = RREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
@@ -490,9 +504,10 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
-	int i;
+	int i, num_xcc;
 
-	for (i = 0; i < adev->gfx.num_xcd; i++) {
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 
 		hub->ctx0_ptb_addr_lo32 =
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 4b2c4ecd72538..2c322a25bf1cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1733,7 +1733,8 @@ static int gmc_v9_0_sw_init(void *handle)
 		adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
 	case IP_VERSION(9, 4, 3):
-		bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), adev->gfx.num_xcd);
+		bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0),
+				  NUM_XCC(adev->gfx.xcc_mask));
 		bitmap_set(adev->vmhubs_mask, AMDGPU_MMHUB0(0), adev->num_aid);
 
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 829e32433faf2..df96c4c508a0c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -592,6 +592,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	struct kfd_node *node;
 	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
 	unsigned int max_proc_per_quantum;
+	int num_xcd;
 
 	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev,
 			KGD_ENGINE_MEC1);
@@ -601,16 +602,15 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			KGD_ENGINE_SDMA1);
 	kfd->shared_resources = *gpu_resources;
 
-	if (kfd->adev->gfx.num_xcd == 0 || kfd->adev->gfx.num_xcd == 1 ||
-	    kfd->adev->gfx.num_xcc_per_xcp == 0)
+	num_xcd = NUM_XCC(kfd->adev->gfx.xcc_mask);
+	if (num_xcd == 0 || num_xcd == 1 || kfd->adev->gfx.num_xcc_per_xcp == 0)
 		kfd->num_nodes = 1;
 	else
-		kfd->num_nodes =
-			kfd->adev->gfx.num_xcd/kfd->adev->gfx.num_xcc_per_xcp;
+		kfd->num_nodes = num_xcd / kfd->adev->gfx.num_xcc_per_xcp;
 	if (kfd->num_nodes == 0) {
 		dev_err(kfd_device,
 			"KFD num nodes cannot be 0, GC inst: %d, num_xcc_in_node: %d\n",
-			kfd->adev->gfx.num_xcd, kfd->adev->gfx.num_xcc_per_xcp);
+			num_xcd, kfd->adev->gfx.num_xcc_per_xcp);
 		goto out;
 	}
 
-- 
GitLab


From d244a4167a0276db054d588e11f8142942bcf529 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 16:04:26 -0400
Subject: [PATCH 1078/1791] drm/amdgpu/vcn: use vcn4 irqsrc header for VCN
 4.0.3

Use vcn4 irqsrc header for VCN 4.0.3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index ae2cc47d344a0..fafce2beb6cfe 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -35,7 +35,7 @@
 
 #include "vcn/vcn_4_0_3_offset.h"
 #include "vcn/vcn_4_0_3_sh_mask.h"
-#include "ivsrcid/vcn/irqsrcs_vcn_2_0.h"
+#include "ivsrcid/vcn/irqsrcs_vcn_4_0.h"
 
 #define mmUVD_DPG_LMA_CTL		regUVD_DPG_LMA_CTL
 #define mmUVD_DPG_LMA_CTL_BASE_IDX	regUVD_DPG_LMA_CTL_BASE_IDX
@@ -104,7 +104,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 
 	/* VCN DEC TRAP */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-		VCN_2_0__SRCID__UVD_SYSTEM_MESSAGE_INTERRUPT, &adev->vcn.inst->irq);
+		VCN_4_0__SRCID__UVD_TRAP, &adev->vcn.inst->irq);
 	if (r)
 		return r;
 
@@ -1380,7 +1380,7 @@ static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 	DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n");
 
 	switch (entry->src_id) {
-	case VCN_2_0__SRCID__UVD_SYSTEM_MESSAGE_INTERRUPT:
+	case VCN_4_0__SRCID__UVD_TRAP:
 		amdgpu_fence_process(&adev->vcn.inst->ring_dec);
 		break;
 	default:
-- 
GitLab


From f9f74df58faa4d6744acc6d91a81a86895bc7fe8 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 16:41:16 -0400
Subject: [PATCH 1079/1791] drm/amdgpu/vcn: update vcn header to support
 multiple AIDs

Add aid_id in vcn header to support multiple AIDs

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index c730949ece7d9..1024a06359ca6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -242,6 +242,7 @@ struct amdgpu_vcn_inst {
 	uint32_t		*dpg_sram_curr_addr;
 	atomic_t		dpg_enc_submission_cnt;
 	struct amdgpu_vcn_fw_shared fw_shared;
+	uint8_t			aid_id;
 };
 
 struct amdgpu_vcn_ras {
@@ -271,6 +272,8 @@ struct amdgpu_vcn {
 
 	struct ras_common_if    *ras_if;
 	struct amdgpu_vcn_ras   *ras;
+
+	uint8_t	num_inst_per_aid;
 };
 
 struct amdgpu_fw_shared_rb_ptrs_struct {
-- 
GitLab


From 492c464750587ea033d7a2fcaa21f902e345b383 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 16:41:52 -0400
Subject: [PATCH 1080/1791] drm/amdgpu/jpeg: update jpeg header to support
 multiple AIDs

Add aid_id in jpeg header to support multiple AIDs.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index cb6c127ab81d4..5c200a508fa3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -40,6 +40,7 @@ struct amdgpu_jpeg_inst {
 	struct amdgpu_ring ring_dec[AMDGPU_MAX_JPEG_RINGS];
 	struct amdgpu_irq_src irq;
 	struct amdgpu_jpeg_reg external;
+	uint8_t aid_id;
 };
 
 struct amdgpu_jpeg_ras {
@@ -58,6 +59,7 @@ struct amdgpu_jpeg {
 	atomic_t total_submission_cnt;
 	struct ras_common_if	*ras_if;
 	struct amdgpu_jpeg_ras	*ras;
+	uint8_t num_inst_per_aid;
 };
 
 int amdgpu_jpeg_sw_init(struct amdgpu_device *adev);
-- 
GitLab


From 53054e9a7775c228ada4d052f3e7849e71072811 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 19:34:00 -0400
Subject: [PATCH 1081/1791] drm/amdgpu/vcn: update new doorbell map

New doorbell map is used for VCN 4.0.3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c  | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 8914f3c6c80f3..e12e3646c49ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -102,7 +102,7 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 		ring = &adev->jpeg.inst->ring_dec[i];
 		ring->use_doorbell = true;
 		ring->vm_hub = AMDGPU_MMHUB0(0);
-		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + (i?8:1) + i;
+		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + i;
 		sprintf(ring->name, "jpeg_dec_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 					AMDGPU_RING_PRIO_DEFAULT, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 266b504fd83e7..9626270059619 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -161,7 +161,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
-				0x10);
+				0x9);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
@@ -174,7 +174,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x4);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
-				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x10);
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x9);
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index fafce2beb6cfe..ddd844cca02e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -110,7 +110,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 
 	ring = &adev->vcn.inst->ring_dec;
 	ring->use_doorbell = true;
-	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 5;
+	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1);
 	ring->vm_hub = AMDGPU_MMHUB0(0);
 	sprintf(ring->name, "vcn_dec");
 	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
@@ -176,7 +176,7 @@ static int vcn_v4_0_3_hw_init(void *handle)
 	int r;
 
 	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), ring->me);
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
 	if (ring->use_doorbell)
 		WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
 			ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
-- 
GitLab


From 429a111851bc1f7388fe44af36166d399583a18f Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 20:00:41 -0400
Subject: [PATCH 1082/1791] drm/amdgpu/vcn: update ucode setup

Use common amdgpu_vcn_setup_ucode for ucode setup.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index ddd844cca02e5..5eaaac531ab66 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -87,16 +87,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
-		const struct common_firmware_header *hdr;
-		hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
-		adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].ucode_id = AMDGPU_UCODE_ID_VCN;
-		adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].fw = adev->vcn.fw;
-		adev->firmware.fw_size +=
-			ALIGN(le32_to_cpu(hdr->ucode_size_bytes), PAGE_SIZE);
-
-		DRM_DEV_INFO(adev->dev, "Will use PSP to load VCN firmware\n");
-	}
+	amdgpu_vcn_setup_ucode(adev);
 
 	r = amdgpu_vcn_resume(adev);
 	if (r)
-- 
GitLab


From 7f0af02ab362df7f064df690fd59659a1cdb9b44 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 20:03:18 -0400
Subject: [PATCH 1083/1791] drm/amdgpu/vcn: remove unused code

Remove unused code.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 121 ------------------------
 1 file changed, 121 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 5eaaac531ab66..93c18fd7de77f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -416,117 +416,6 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indir
 		VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
 }
 
-/**
- * vcn_v4_0_disable_static_power_gating - disable VCN static power gating
- *
- * @adev: amdgpu_device pointer
- *
- * Disable static power gating for VCN block
- */
-static void vcn_v4_0_3_disable_static_power_gating(struct amdgpu_device *adev)
-{
-	uint32_t data = 0;
-
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN) {
-		data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
-
-		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
-		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS,
-			UVD_PGFSM_STATUS__UVDM_UVDU_UVDLM_PWR_ON_3_0, 0x3F3FFFFF);
-	} else {
-		data = (1 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
-			| 1 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
-
-		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
-		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS, 0,  0x3F3FFFFF);
-	}
-
-	data = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
-	data &= ~0x103;
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN)
-		data |= UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON |
-			UVD_POWER_STATUS__UVD_PG_EN_MASK;
-
-	WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, data);
-}
-
-/**
- * vcn_v4_0_3_enable_static_power_gating - enable VCN static power gating
- *
- * @adev: amdgpu_device pointer
- *
- * Enable static power gating for VCN block
- */
-static void vcn_v4_0_3_enable_static_power_gating(struct amdgpu_device *adev)
-{
-	uint32_t data;
-
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN) {
-		/* Before power off, this indicator has to be turned on */
-		data = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
-		data &= ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK;
-		data |= UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF;
-		WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, data);
-
-		data = (2 << UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT
-			| 2 << UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT);
-		WREG32_SOC15(VCN, 0, regUVD_PGFSM_CONFIG, data);
-
-		data = (2 << UVD_PGFSM_STATUS__UVDM_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDS_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDF_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDTC_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDB_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDTA_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDLM_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDTD_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDTE_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDE_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDAB_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDTB_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDNA_PWR_STATUS__SHIFT
-			| 2 << UVD_PGFSM_STATUS__UVDNB_PWR_STATUS__SHIFT);
-		SOC15_WAIT_ON_RREG(VCN, 0, regUVD_PGFSM_STATUS, data, 0x3F3FFFFF);
-	}
-}
-
 /**
  * vcn_v4_0_3_disable_clock_gating - disable VCN clock gating
  *
@@ -869,9 +758,6 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
 		return vcn_v4_0_3_start_dpg_mode(adev, adev->vcn.indirect_sram);
 
-	/* disable VCN power gating */
-	vcn_v4_0_3_disable_static_power_gating(adev);
-
 	/* set VCN status busy */
 	tmp = RREG32_SOC15(VCN, 0, regUVD_STATUS) | UVD_STATUS__UVD_BUSY;
 	WREG32_SOC15(VCN, 0, regUVD_STATUS, tmp);
@@ -1119,19 +1005,12 @@ static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 	tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
 	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
 
-	tmp = RREG32_SOC15(VCN, 0, regUVD_VCPU_CNTL);
-	tmp |= UVD_VCPU_CNTL__BLK_RST_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
-
 	/* clear VCN status */
 	WREG32_SOC15(VCN, 0, regUVD_STATUS, 0);
 
 	/* apply HW clock gating */
 	vcn_v4_0_3_enable_clock_gating(adev);
 
-	/* enable VCN power gating */
-	vcn_v4_0_3_enable_static_power_gating(adev);
-
 Done:
 	if (adev->pm.dpm_enabled)
 		amdgpu_dpm_enable_uvd(adev, false);
-- 
GitLab


From cf1f3737c2d336b1473596cc7356bdffa186bc71 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 11 Jul 2022 10:58:40 -0400
Subject: [PATCH 1084/1791] drm/amdgpu/vcn: update amdgpu_fw_shared to
 amdgpu_vcn4_fw_shared

Use amdgpu_vcn4_fw_shared for vcn 4.0.3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 40 +++++++------------------
 1 file changed, 11 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 93c18fd7de77f..0b2b97593baca 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -79,7 +79,7 @@ static int vcn_v4_0_3_early_init(void *handle)
 static int vcn_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	int r;
 
@@ -111,10 +111,8 @@ static int vcn_v4_0_3_sw_init(void *handle)
 		return r;
 
 	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
-	fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_SW_RING_FLAG) |
-				     cpu_to_le32(AMDGPU_VCN_MULTI_QUEUE_FLAG) |
-				     cpu_to_le32(AMDGPU_VCN_FW_SHARED_FLAG_0_RB);
-	fw_shared->sw_ring.is_enabled = cpu_to_le32(true);
+	fw_shared->present_flag_0 = 0;
+	fw_shared->sq.is_enabled = cpu_to_le32(true);
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
 		adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode;
@@ -135,11 +133,11 @@ static int vcn_v4_0_3_sw_fini(void *handle)
 	int r, idx;
 
 	if (drm_dev_enter(&adev->ddev, &idx)) {
-		volatile struct amdgpu_fw_shared *fw_shared;
+		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 
 		fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 		fw_shared->present_flag_0 = 0;
-		fw_shared->sw_ring.is_enabled = cpu_to_le32(false);
+		fw_shared->sq.is_enabled = cpu_to_le32(false);
 
 		drm_dev_exit(idx);
 	}
@@ -304,7 +302,7 @@ static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev)
 		upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr));
 	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0, 0);
 	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_SIZE0,
-		AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)));
+		AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)));
 }
 
 /**
@@ -407,7 +405,7 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indir
 			VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect);
 	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_NONCACHE_SIZE0),
-			AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_fw_shared)), 0, indirect);
+			AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)), 0, indirect);
 
 	/* VCN global tiling registers */
 	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
@@ -612,7 +610,7 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
  */
 static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
 {
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 
@@ -702,7 +700,6 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
 				(uintptr_t)adev->vcn.inst->dpg_sram_cpu_addr));
 
 	ring = &adev->vcn.inst->ring_dec;
-	fw_shared->multi_queue.decode_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
 
 	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
@@ -728,12 +725,8 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
 
 	WREG32_SOC15(VCN, 0, regUVD_SCRATCH2, 0);
 
-	/* Reset FW shared memory RBC WPTR/RPTR */
-	fw_shared->rb.rptr = 0;
-	fw_shared->rb.wptr = lower_32_bits(ring->wptr);
-
 	/*resetting done, fw can check RB ring */
-	fw_shared->multi_queue.decode_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+	fw_shared->sq.queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
 
 	return 0;
 }
@@ -747,7 +740,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
  */
 static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 {
-	volatile struct amdgpu_fw_shared *fw_shared;
+	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int j, k, r;
@@ -878,7 +871,6 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 	ring = &adev->vcn.inst->ring_dec;
 
 	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
-	fw_shared->multi_queue.decode_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
 
 	/* program the RB_BASE for ring buffer */
 	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
@@ -902,8 +894,7 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
 
 	ring->wptr = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
-	fw_shared->rb.wptr = cpu_to_le32(lower_32_bits(ring->wptr));
-	fw_shared->multi_queue.decode_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+	fw_shared->sq.queue_mode &= cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF));
 
 	return 0;
 }
@@ -1075,15 +1066,6 @@ static uint64_t vcn_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
 static void vcn_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	volatile struct amdgpu_fw_shared *fw_shared;
-
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
-		/*whenever update RBC_RB_WPTR, we save the wptr in shared rb.wptr and scratch2 */
-		fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
-		fw_shared->rb.wptr = lower_32_bits(ring->wptr);
-		WREG32_SOC15(VCN, ring->me, regUVD_SCRATCH2,
-			lower_32_bits(ring->wptr));
-	}
 
 	if (ring->use_doorbell) {
 		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
-- 
GitLab


From 07bc768afb0a1c638f4eb054aaed9eaa291132de Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 9 Aug 2022 12:58:58 -0400
Subject: [PATCH 1085/1791] drm/amdgpu: reflect psp xgmi topology info for
 gfx9.4.3

Similar to GFX9.4.2 non-A+A devices, GFX9.4.3 psp xgmi topology info is
half duplex and requires the driver to fill in the bidirectional info.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Shiwu Zhang <shiwu.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index aa37b703c7186..dd7f6d688449c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1254,8 +1254,9 @@ int psp_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id)
 
 static bool psp_xgmi_peer_link_info_supported(struct psp_context *psp)
 {
-	return psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2) &&
-		psp->xgmi_context.context.bin_desc.fw_version >= 0x2000000b;
+	return (psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2) &&
+		psp->xgmi_context.context.bin_desc.fw_version >= 0x2000000b) ||
+		psp->adev->ip_versions[MP0_HWIP][0] >= IP_VERSION(13, 0, 6);
 }
 
 /*
@@ -1363,6 +1364,9 @@ int psp_xgmi_get_topology_info(struct psp_context *psp,
 	/* Invoke xgmi ta again to get the link information */
 	if (psp_xgmi_peer_link_info_supported(psp)) {
 		struct ta_xgmi_cmd_get_peer_link_info_output *link_info_output;
+		bool requires_reflection =
+			(psp->xgmi_context.supports_extended_data && get_extended_data) ||
+				psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6);
 
 		xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_PEER_LINKS;
 
@@ -1380,8 +1384,7 @@ int psp_xgmi_get_topology_info(struct psp_context *psp,
 					link_info_output->nodes[i].num_links;
 
 			/* reflect the topology information for bi-directionality */
-			if (psp->xgmi_context.supports_extended_data &&
-					get_extended_data && topology->nodes[i].num_hops)
+			if (requires_reflection && topology->nodes[i].num_hops)
 				psp_xgmi_reflect_topology_info(psp, topology->nodes[i]);
 		}
 	}
-- 
GitLab


From 1526ec9a3ed90e7ad36de7ef8aa2768b60af14df Mon Sep 17 00:00:00 2001
From: David Belanger <david.belanger@amd.com>
Date: Wed, 16 Feb 2022 12:07:28 -0500
Subject: [PATCH 1086/1791] drm/amdkfd: EOP Removal - Handle size 0 correctly

On GC 9.4.3, we are removing the EOP buffer.
If we specify 0 for the size, CP_HQD_EOP_CONTROL ends up with
incorrect value as order_size_2 calculations does not handle 0.

Fix it by using zero for the MQD entry for EOP size 0.

v2: Reworked code with a conditional assignment and fixed style issues.

Signed-off-by: David Belanger <david.belanger@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 5c9b3392758e3..2e2a0f8586f7a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -257,9 +257,14 @@ static void update_mqd(struct mqd_manager *mm, void *mqd,
 	 * Limit is 0xFF EOP entries (= 0x7F8 dwords). CP will not submit
 	 * more than (EOP entry count - 1) so a queue size of 0x800 dwords
 	 * is safe, giving a maximum field value of 0xA.
+	 *
+	 * Also, do calculation only if EOP is used (size > 0), otherwise
+	 * the order_base_2 calculation provides incorrect result.
+	 *
 	 */
-	m->cp_hqd_eop_control = min(0xA,
-		order_base_2(q->eop_ring_buffer_size / 4) - 1);
+	m->cp_hqd_eop_control = q->eop_ring_buffer_size ?
+		min(0xA, order_base_2(q->eop_ring_buffer_size / 4) - 1) : 0;
+
 	m->cp_hqd_eop_base_addr_lo =
 			lower_32_bits(q->eop_ring_buffer_address >> 8);
 	m->cp_hqd_eop_base_addr_hi =
-- 
GitLab


From 0c552ed38780f24b7ac235c3d10c6c94686ecfdf Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Sat, 2 Apr 2022 19:39:59 +0800
Subject: [PATCH 1087/1791] drm/amdgpu: add indirect r/w interface for smn
 address greater than 32bits

On multiple AIDs platform, bit[34:32] in SMD address is leveraged to access
nonAID0 register smn address and new PCI_INDEX_HI register is introduced
to access the higher bits.

v2: rebase on latest register accessors (Alex)

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        | 11 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 98 ++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h   |  1 +
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c     |  6 ++
 drivers/gpu/drm/amd/amdgpu/soc15.c         |  2 +
 5 files changed, 118 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 5f4396185a2e6..729d5eedae49f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -608,6 +608,9 @@ void amdgpu_cgs_destroy_device(struct cgs_device *cgs_device);
 typedef uint32_t (*amdgpu_rreg_t)(struct amdgpu_device*, uint32_t);
 typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
 
+typedef uint32_t (*amdgpu_rreg_ext_t)(struct amdgpu_device*, uint64_t);
+typedef void (*amdgpu_wreg_ext_t)(struct amdgpu_device*, uint64_t, uint32_t);
+
 typedef uint64_t (*amdgpu_rreg64_t)(struct amdgpu_device*, uint32_t);
 typedef void (*amdgpu_wreg64_t)(struct amdgpu_device*, uint32_t, uint64_t);
 
@@ -798,6 +801,8 @@ struct amdgpu_device {
 	amdgpu_wreg_t			pcie_wreg;
 	amdgpu_rreg_t			pciep_rreg;
 	amdgpu_wreg_t			pciep_wreg;
+	amdgpu_rreg_ext_t		pcie_rreg_ext;
+	amdgpu_wreg_ext_t		pcie_wreg_ext;
 	amdgpu_rreg64_t			pcie_rreg64;
 	amdgpu_wreg64_t			pcie_wreg64;
 	/* protects concurrent UVD register access */
@@ -1088,9 +1093,13 @@ uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev,
 			    uint32_t expected_value, uint32_t mask);
 uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
 			    uint32_t reg, uint32_t acc_flags);
+u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev,
+				    u64 reg_addr);
 void amdgpu_device_wreg(struct amdgpu_device *adev,
 			uint32_t reg, uint32_t v,
 			uint32_t acc_flags);
+void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev,
+				     u64 reg_addr, u32 reg_data);
 void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev,
 			     uint32_t reg, uint32_t v);
 void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value);
@@ -1142,6 +1151,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define WREG32_PCIE(reg, v) adev->pcie_wreg(adev, (reg), (v))
 #define RREG32_PCIE_PORT(reg) adev->pciep_rreg(adev, (reg))
 #define WREG32_PCIE_PORT(reg, v) adev->pciep_wreg(adev, (reg), (v))
+#define RREG32_PCIE_EXT(reg) adev->pcie_rreg_ext(adev, (reg))
+#define WREG32_PCIE_EXT(reg, v) adev->pcie_wreg_ext(adev, (reg), (v))
 #define RREG64_PCIE(reg) adev->pcie_rreg64(adev, (reg))
 #define WREG64_PCIE(reg, v) adev->pcie_wreg64(adev, (reg), (v))
 #define RREG32_SMC(reg) adev->smc_rreg(adev, (reg))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 38b6bbcaf186f..c190365d67e27 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -707,6 +707,48 @@ u32 amdgpu_device_indirect_rreg(struct amdgpu_device *adev,
 	return r;
 }
 
+u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev,
+				    u64 reg_addr)
+{
+	unsigned long flags, pcie_index, pcie_index_hi, pcie_data;
+	u32 r;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_index_hi_offset;
+	void __iomem *pcie_data_offset;
+
+	pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev);
+	pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev);
+	if (adev->nbio.funcs->get_pcie_index_hi_offset)
+		pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev);
+	else
+		pcie_index_hi = 0;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+	if (pcie_index_hi != 0)
+		pcie_index_hi_offset = (void __iomem *)adev->rmmio +
+				pcie_index_hi * 4;
+
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	if (pcie_index_hi != 0) {
+		writel((reg_addr >> 32) & 0xff, pcie_index_hi_offset);
+		readl(pcie_index_hi_offset);
+	}
+	r = readl(pcie_data_offset);
+
+	/* clear the high bits */
+	if (pcie_index_hi != 0) {
+		writel(0, pcie_index_hi_offset);
+		readl(pcie_index_hi_offset);
+	}
+
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+
+	return r;
+}
+
 /**
  * amdgpu_device_indirect_rreg64 - read a 64bits indirect register
  *
@@ -774,6 +816,46 @@ void amdgpu_device_indirect_wreg(struct amdgpu_device *adev,
 	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
 }
 
+void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev,
+				     u64 reg_addr, u32 reg_data)
+{
+	unsigned long flags, pcie_index, pcie_index_hi, pcie_data;
+	void __iomem *pcie_index_offset;
+	void __iomem *pcie_index_hi_offset;
+	void __iomem *pcie_data_offset;
+
+	pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev);
+	pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev);
+	if (adev->nbio.funcs->get_pcie_index_hi_offset)
+		pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev);
+	else
+		pcie_index_hi = 0;
+
+	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
+	pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4;
+	pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4;
+	if (pcie_index_hi != 0)
+		pcie_index_hi_offset = (void __iomem *)adev->rmmio +
+				pcie_index_hi * 4;
+
+	writel(reg_addr, pcie_index_offset);
+	readl(pcie_index_offset);
+	if (pcie_index_hi != 0) {
+		writel((reg_addr >> 32) & 0xff, pcie_index_hi_offset);
+		readl(pcie_index_hi_offset);
+	}
+	writel(reg_data, pcie_data_offset);
+	readl(pcie_data_offset);
+
+	/* clear the high bits */
+	if (pcie_index_hi != 0) {
+		writel(0, pcie_index_hi_offset);
+		readl(pcie_index_hi_offset);
+	}
+
+	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
+}
+
 /**
  * amdgpu_device_indirect_wreg64 - write a 64bits indirect register address
  *
@@ -840,6 +922,13 @@ static uint32_t amdgpu_invalid_rreg(struct amdgpu_device *adev, uint32_t reg)
 	return 0;
 }
 
+static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg)
+{
+	DRM_ERROR("Invalid callback to read register 0x%llX\n", reg);
+	BUG();
+	return 0;
+}
+
 /**
  * amdgpu_invalid_wreg - dummy reg write function
  *
@@ -857,6 +946,13 @@ static void amdgpu_invalid_wreg(struct amdgpu_device *adev, uint32_t reg, uint32
 	BUG();
 }
 
+static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, uint32_t v)
+{
+	DRM_ERROR("Invalid callback to write register 0x%llX with 0x%08X\n",
+		  reg, v);
+	BUG();
+}
+
 /**
  * amdgpu_invalid_rreg64 - dummy 64 bit reg read function
  *
@@ -3611,6 +3707,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	adev->smc_wreg = &amdgpu_invalid_wreg;
 	adev->pcie_rreg = &amdgpu_invalid_rreg;
 	adev->pcie_wreg = &amdgpu_invalid_wreg;
+	adev->pcie_rreg_ext = &amdgpu_invalid_rreg_ext;
+	adev->pcie_wreg_ext = &amdgpu_invalid_wreg_ext;
 	adev->pciep_rreg = &amdgpu_invalid_rreg;
 	adev->pciep_wreg = &amdgpu_invalid_wreg;
 	adev->pcie_rreg64 = &amdgpu_invalid_rreg64;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index 6a636c34b7178..8fa3a1f3b1812 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -61,6 +61,7 @@ struct amdgpu_nbio_funcs {
 	u32 (*get_hdp_flush_done_offset)(struct amdgpu_device *adev);
 	u32 (*get_pcie_index_offset)(struct amdgpu_device *adev);
 	u32 (*get_pcie_data_offset)(struct amdgpu_device *adev);
+	u32 (*get_pcie_index_hi_offset)(struct amdgpu_device *adev);
 	u32 (*get_pcie_port_index_offset)(struct amdgpu_device *adev);
 	u32 (*get_pcie_port_data_offset)(struct amdgpu_device *adev);
 	u32 (*get_rev_id)(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 9626270059619..92e9c5ed95df8 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -319,6 +319,11 @@ static u32 nbio_v7_9_get_pcie_data_offset(struct amdgpu_device *adev)
 	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_DATA2);
 }
 
+static u32 nbio_v7_9_get_pcie_index_hi_offset(struct amdgpu_device *adev)
+{
+	return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_INDEX2_HI);
+}
+
 const struct nbio_hdp_flush_reg nbio_v7_9_hdp_flush_reg = {
 	.ref_and_mask_cp0 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP0_MASK,
 	.ref_and_mask_cp1 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP1_MASK,
@@ -382,6 +387,7 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset,
 	.get_pcie_index_offset = nbio_v7_9_get_pcie_index_offset,
 	.get_pcie_data_offset = nbio_v7_9_get_pcie_data_offset,
+	.get_pcie_index_hi_offset = nbio_v7_9_get_pcie_index_hi_offset,
 	.get_rev_id = nbio_v7_9_get_rev_id,
 	.mc_access_enable = nbio_v7_9_mc_access_enable,
 	.get_memsize = nbio_v7_9_get_memsize,
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index b7e8af56df846..b9bcb12bff91e 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -866,6 +866,8 @@ static int soc15_common_early_init(void *handle)
 	adev->smc_wreg = NULL;
 	adev->pcie_rreg = &amdgpu_device_indirect_rreg;
 	adev->pcie_wreg = &amdgpu_device_indirect_wreg;
+	adev->pcie_rreg_ext = &amdgpu_device_indirect_rreg_ext;
+	adev->pcie_wreg_ext = &amdgpu_device_indirect_wreg_ext;
 	adev->pcie_rreg64 = &amdgpu_device_indirect_rreg64;
 	adev->pcie_wreg64 = &amdgpu_device_indirect_wreg64;
 	adev->uvd_ctx_rreg = &soc15_uvd_ctx_rreg;
-- 
GitLab


From 6b22ef25748fb77030bac02e2147f6d738212ac3 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Sat, 2 Apr 2022 20:21:35 +0800
Subject: [PATCH 1088/1791] drm/amdgpu: configure the doorbell settings for
 sdma on non-AID0

Configure the sdma doorbell settings on NBIF0 and SYSHUB of each AID

v2: fetch aid_id from amdgpu_sdma_instance (Lijo)

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 65 ++++++++++++++++++++++----
 1 file changed, 56 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 92e9c5ed95df8..5a5bab665f600 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -62,10 +62,23 @@ static u32 nbio_v7_9_get_memsize(struct amdgpu_device *adev)
 	return RREG32_SOC15(NBIO, 0, regRCC_DEV0_EPF0_RCC_CONFIG_MEMSIZE);
 }
 
+#define S2A_DOORBELL_REG_LSD_OFFSET	0x40
+
+/* Temporarily add 2 macros below. Range is 0 ~ 3 as total AID number is 4.
+ * They will be obsoleted after the latest ip offset header
+ * is imported in driver in near future.
+ */
+#define AMDGPU_SMN_TARGET_AID(x)	((u64)(x) << 32)
+#define AMDGPU_SMN_CROSS_AID		(1ULL << 34)
+
 static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
 			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
 	u32 doorbell_range = 0, doorbell_ctrl = 0;
+	int aid_id = adev->sdma.instance[instance].aid_id;
+
+	if (use_doorbell == false)
+		return;
 
 	doorbell_range =
 		REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0,
@@ -80,9 +93,10 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 		REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL,
 			S2A_DOORBELL_PORT1_RANGE_SIZE, doorbell_size);
 
-	switch (instance) {
+	switch (instance % adev->sdma.num_inst_per_aid) {
 	case 0:
-		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1, doorbell_range);
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1) +
+			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 					S2A_DOORBELL_ENTRY_1_CTRL,
@@ -94,10 +108,15 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x1);
-		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_1_CTRL, doorbell_ctrl);
+		WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_1_CTRL)
+			+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
+			+ AMDGPU_SMN_TARGET_AID(aid_id)
+			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+			doorbell_ctrl);
 		break;
 	case 1:
-		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2, doorbell_range);
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2) +
+			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 					S2A_DOORBELL_ENTRY_1_CTRL,
@@ -109,10 +128,15 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x2);
-		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_2_CTRL, doorbell_ctrl);
+		WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_2_CTRL)
+			+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
+			+ AMDGPU_SMN_TARGET_AID(aid_id)
+			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+			doorbell_ctrl);
 		break;
 	case 2:
-		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3, doorbell_range);
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3) +
+			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 					S2A_DOORBELL_ENTRY_1_CTRL,
@@ -124,10 +148,22 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x8);
-		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_5_CTRL, doorbell_ctrl);
+		if (aid_id != 0)
+			WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0,
+				regS2A_DOORBELL_ENTRY_3_CTRL)
+				+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
+				+ AMDGPU_SMN_TARGET_AID(aid_id)
+				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+				doorbell_ctrl);
+		else
+			WREG32(SOC15_REG_OFFSET(NBIO, 0,
+				regS2A_DOORBELL_ENTRY_5_CTRL)
+				+ S2A_DOORBELL_REG_LSD_OFFSET,
+				doorbell_ctrl);
 		break;
 	case 3:
-		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4, doorbell_range);
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4) +
+			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 					S2A_DOORBELL_ENTRY_1_CTRL,
@@ -139,7 +175,18 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x9);
-		WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_6_CTRL, doorbell_ctrl);
+		if (aid_id != 0)
+			WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0,
+				regS2A_DOORBELL_ENTRY_4_CTRL)
+				+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
+				+ AMDGPU_SMN_TARGET_AID(aid_id)
+				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+				doorbell_ctrl);
+		else
+			WREG32(SOC15_REG_OFFSET(NBIO, 0,
+				regS2A_DOORBELL_ENTRY_6_CTRL)
+				+ S2A_DOORBELL_REG_LSD_OFFSET,
+				doorbell_ctrl);
 		break;
 	default:
 		break;
-- 
GitLab


From fe1f05df5919c67c3add49efb55e251a8d78ee4e Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 31 May 2022 14:39:36 -0400
Subject: [PATCH 1089/1791] drm/amdkfd: Rework kfd_locked handling

Currently, even if kfd_locked is set, a process is first
created and then removed to work around a race condition
in updating kfd_locked flag. Rework kfd_locked handling to
ensure no processes is created if kfd_locked is set. This
is achieved by updating kfd_locked under kfd_processes_mutex.
With this there is no need for kfd_locked to be an atomic
counter. Instead, it can be a regular integer.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  7 -------
 drivers/gpu/drm/amd/amdkfd/kfd_device.c  | 21 ++++++++++++++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |  2 ++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c |  8 +++++++-
 4 files changed, 25 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 45e8da125f70b..8b9accecf49b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -146,13 +146,6 @@ static int kfd_open(struct inode *inode, struct file *filep)
 	if (IS_ERR(process))
 		return PTR_ERR(process);
 
-	if (kfd_is_locked()) {
-		dev_dbg(kfd_device, "kfd is locked!\n"
-				"process %d unreferenced", process->pasid);
-		kfd_unref_process(process);
-		return -EAGAIN;
-	}
-
 	/* filep now owns the reference returned by kfd_create_process */
 	filep->private_data = process;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index df96c4c508a0c..eb2b44fddf746 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -42,7 +42,7 @@
  * once locked, kfd driver will stop any further GPU execution.
  * create process (open) will return -EAGAIN.
  */
-static atomic_t kfd_locked = ATOMIC_INIT(0);
+static int kfd_locked;
 
 #ifdef CONFIG_DRM_AMDGPU_CIK
 extern const struct kfd2kgd_calls gfx_v7_kfd2kgd;
@@ -880,7 +880,9 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd)
 			return ret;
 	}
 
-	atomic_dec(&kfd_locked);
+	mutex_lock(&kfd_processes_mutex);
+	--kfd_locked;
+	mutex_unlock(&kfd_processes_mutex);
 
 	for (i = 0; i < kfd->num_nodes; i++) {
 		node = kfd->nodes[i];
@@ -893,21 +895,27 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd)
 
 bool kfd_is_locked(void)
 {
-	return  (atomic_read(&kfd_locked) > 0);
+	lockdep_assert_held(&kfd_processes_mutex);
+	return  (kfd_locked > 0);
 }
 
 void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 	struct kfd_node *node;
 	int i;
+	int count;
 
 	if (!kfd->init_complete)
 		return;
 
 	/* for runtime suspend, skip locking kfd */
 	if (!run_pm) {
+		mutex_lock(&kfd_processes_mutex);
+		count = ++kfd_locked;
+		mutex_unlock(&kfd_processes_mutex);
+
 		/* For first KFD device suspend all the KFD processes */
-		if (atomic_inc_return(&kfd_locked) == 1)
+		if (count == 1)
 			kfd_suspend_all_processes();
 	}
 
@@ -933,7 +941,10 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 
 	/* for runtime resume, skip unlocking kfd */
 	if (!run_pm) {
-		count = atomic_dec_return(&kfd_locked);
+		mutex_lock(&kfd_processes_mutex);
+		count = --kfd_locked;
+		mutex_unlock(&kfd_processes_mutex);
+
 		WARN_ONCE(count < 0, "KFD suspend / resume ref. error");
 		if (count == 0)
 			ret = kfd_resume_all_processes();
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 5cfebcc8b3059..400b4dcbdf053 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -201,6 +201,8 @@ extern int amdgpu_no_queue_eviction_on_vm_fault;
 /* Enable eviction debug messages */
 extern bool debug_evictions;
 
+extern struct mutex kfd_processes_mutex;
+
 enum cache_policy {
 	cache_policy_coherent,
 	cache_policy_noncoherent
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 9b1e84d33cdc7..c3d43e6e52361 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -50,7 +50,7 @@ struct mm_struct;
  * Unique/indexed by mm_struct*
  */
 DEFINE_HASHTABLE(kfd_processes_table, KFD_PROCESS_TABLE_SIZE);
-static DEFINE_MUTEX(kfd_processes_mutex);
+DEFINE_MUTEX(kfd_processes_mutex);
 
 DEFINE_SRCU(kfd_processes_srcu);
 
@@ -818,6 +818,12 @@ struct kfd_process *kfd_create_process(struct file *filep)
 	 */
 	mutex_lock(&kfd_processes_mutex);
 
+	if (kfd_is_locked()) {
+		mutex_unlock(&kfd_processes_mutex);
+		pr_debug("KFD is locked! Cannot create process");
+		return ERR_PTR(-EINVAL);
+	}
+
 	/* A prior open of /dev/kfd could have already created the process. */
 	process = find_process(thread, false);
 	if (process) {
-- 
GitLab


From 0c7315e7d5ef9b36ca4db32ffeb34a187cbaf231 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Fri, 10 Jun 2022 09:41:29 -0400
Subject: [PATCH 1090/1791] drm/amdkfd: Add device repartition support

GFX9.4.3 will support dynamic repartitioning of the GPU through sysfs.
Add device repartitioning support in KFD to repartition GPU from one
mode to other.

v2: squash in fix ("drm/amdkfd: Fix warning kgd2kfd_unlock_kfd defined but not used")

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 10 ++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c    | 22 +++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c    |  5 +----
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    | 21 +++++++++++++++++++++
 5 files changed, 66 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 9d19c7ceda3f6..bbbfe9ec4adf8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -773,3 +773,13 @@ bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev)
 	else
 		return false;
 }
+
+int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev)
+{
+	return kgd2kfd_check_and_lock_kfd();
+}
+
+void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev)
+{
+	kgd2kfd_unlock_kfd();
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index df07e212c21e9..d1d643a050a1d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -151,6 +151,8 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev);
 void amdgpu_amdkfd_device_init(struct amdgpu_device *adev);
 void amdgpu_amdkfd_device_fini_sw(struct amdgpu_device *adev);
+int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev);
+void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev);
 int amdgpu_amdkfd_submit_ib(struct amdgpu_device *adev,
 				enum kgd_engine_type engine,
 				uint32_t vmid, uint64_t gpu_addr,
@@ -373,6 +375,8 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd);
 void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
 void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd);
 void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask);
+int kgd2kfd_check_and_lock_kfd(void);
+void kgd2kfd_unlock_kfd(void);
 #else
 static inline int kgd2kfd_init(void)
 {
@@ -438,5 +442,14 @@ static inline
 void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask)
 {
 }
+
+static inline int kgd2kfd_check_and_lock_kfd(void)
+{
+	return 0;
+}
+
+static inline void kgd2kfd_unlock_kfd(void)
+{
+}
 #endif
 #endif /* AMDGPU_AMDKFD_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 069b259f384c3..69bac5b801cee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1233,10 +1233,30 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 		return -EINVAL;
 	}
 
+	if (!adev->kfd.init_complete)
+		return -EPERM;
+
 	mutex_lock(&adev->gfx.partition_mutex);
 
-	ret = adev->gfx.funcs->switch_partition_mode(adev, mode);
+	if (mode == adev->gfx.funcs->query_partition_mode(adev))
+		goto out;
+
+	ret = amdgpu_amdkfd_check_and_lock_kfd(adev);
+	if (ret)
+		goto out;
+
+	amdgpu_amdkfd_device_fini_sw(adev);
+
+	adev->gfx.funcs->switch_partition_mode(adev, mode);
+
+	amdgpu_amdkfd_device_probe(adev);
+	amdgpu_amdkfd_device_init(adev);
+	/* If KFD init failed, return failure */
+	if (!adev->kfd.init_complete)
+		ret = -EIO;
 
+	amdgpu_amdkfd_unlock_kfd(adev);
+out:
 	mutex_unlock(&adev->gfx.partition_mutex);
 
 	if (ret)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index c776fc5884def..47d8ac64e877c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -675,7 +675,7 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 static enum amdgpu_gfx_partition
 gfx_v9_4_3_query_compute_partition(struct amdgpu_device *adev)
 {
-	enum amdgpu_gfx_partition mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+	enum amdgpu_gfx_partition mode = adev->gfx.partition_mode;
 
 	if (adev->nbio.funcs->get_compute_partition_mode)
 		mode = adev->nbio.funcs->get_compute_partition_mode(adev);
@@ -689,9 +689,6 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 	u32 tmp = 0;
 	int num_xcc_per_partition, i, num_xcc;
 
-	if (mode == adev->gfx.partition_mode)
-		return mode;
-
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index eb2b44fddf746..293787290e36b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -1356,6 +1356,27 @@ unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node)
 		kfd_get_num_sdma_engines(node);
 }
 
+int kgd2kfd_check_and_lock_kfd(void)
+{
+	mutex_lock(&kfd_processes_mutex);
+	if (!hash_empty(kfd_processes_table) || kfd_is_locked()) {
+		mutex_unlock(&kfd_processes_mutex);
+		return -EBUSY;
+	}
+
+	++kfd_locked;
+	mutex_unlock(&kfd_processes_mutex);
+
+	return 0;
+}
+
+void kgd2kfd_unlock_kfd(void)
+{
+	mutex_lock(&kfd_processes_mutex);
+	--kfd_locked;
+	mutex_unlock(&kfd_processes_mutex);
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 /* This function will send a package to HIQ to hang the HWS
-- 
GitLab


From 5d30cbb4dbf096bb964fb9c5fd4417cad13445a4 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 14:59:04 +0530
Subject: [PATCH 1091/1791] drm/amdgpu: Add map of logical to physical inst

Add a map for logical to physical instances of an IP. For ex: on some device
configurations, the first logical XCC may not be the first physical XCC.
Software may continue to access in logical IP instance order. The map
provides a convenient way to get to the actual physical instance.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 729d5eedae49f..32112db5c841e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -669,6 +669,15 @@ enum amd_hw_ip_block_type {
 #define IP_VERSION_MIN(ver) (((ver) >> 8) & 0xFF)
 #define IP_VERSION_REV(ver) ((ver) & 0xFF)
 
+struct amdgpu_ip_map_info {
+	/* Map of logical to actual dev instances */
+	uint32_t 		dev_inst[MAX_HWIP][HWIP_MAX_INSTANCE];
+	int8_t (*logical_to_dev_inst)(struct amdgpu_device *adev,
+				      enum amd_hw_ip_block_type block,
+				      int8_t inst);
+
+};
+
 struct amd_powerplay {
 	void *pp_handle;
 	const struct amd_pm_funcs *pp_funcs;
@@ -968,6 +977,7 @@ struct amdgpu_device {
 
 	/* soc15 register offset based on ip, instance and  segment */
 	uint32_t		*reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE];
+	struct amdgpu_ip_map_info	ip_map;
 
 	/* delayed work_func for deferring clockgating during resume */
 	struct delayed_work     delayed_init_work;
-- 
GitLab


From 659a4ab8e27eb39cc61cb74cc714ba1a8f8c9a61 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 5 Jul 2022 09:56:41 +0530
Subject: [PATCH 1092/1791] drm/amdgpu: Use instance lookup table for GC 9.4.3

Register accesses need to be based on physical instance on bare metal.
Pass the right instance using logical to physical instance lookup
table before accessing registers. Add a macro GET_INST to get the right
physical instance of an IP corresponding to a logical instance.

v2: fix gfx_v9_4_3_check_rlcg_range() (Alex)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   |  20 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 381 +++++++++---------
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c      | 146 +++----
 drivers/gpu/drm/amd/amdgpu/soc15_common.h     |   3 +
 4 files changed, 279 insertions(+), 271 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index e81bdca53f42f..5daec0b45545b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -279,8 +279,8 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	/* HQD registers extend to CP_HQD_AQL_DISPATCH_ID_HI */
 	mqd_hqd = &m->cp_mqd_base_addr_lo;
-	hqd_base = SOC15_REG_OFFSET(GC, inst, regCP_MQD_BASE_ADDR);
-	hqd_end = SOC15_REG_OFFSET(GC, inst, regCP_HQD_AQL_DISPATCH_ID_HI);
+	hqd_base = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_MQD_BASE_ADDR);
+	hqd_end = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_AQL_DISPATCH_ID_HI);
 
 	for (reg = hqd_base; reg <= hqd_end; reg++)
 		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
@@ -289,7 +289,7 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_DOORBELL_CONTROL),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_DOORBELL_CONTROL),
 				data);
 
 	if (wptr) {
@@ -319,27 +319,27 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_LO),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_LO),
 		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_HI),
 		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_PQ_WPTR_POLL_ADDR),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_POLL_ADDR),
 		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst,
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
 			regCP_HQD_PQ_WPTR_POLL_ADDR_HI),
 			upper_32_bits((uintptr_t)wptr));
-		WREG32(SOC15_REG_OFFSET(GC, inst, regCP_PQ_WPTR_POLL_CNTL1),
+		WREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_PQ_WPTR_POLL_CNTL1),
 		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id,
 			       queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_EOP_RPTR),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_EOP_RPTR),
 	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
 			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, regCP_HQD_ACTIVE), data);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_ACTIVE), data);
 
 	kgd_gfx_v9_release_queue(adev, inst);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 47d8ac64e877c..e9c12b4970f97 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -190,7 +190,7 @@ static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 2; i < num_xcc; i++)
-		WREG32_SOC15(GC, i, regGRBM_MCM_ADDR, 0x4);
+		WREG32_SOC15(GC, GET_INST(GC, i), regGRBM_MCM_ADDR, 0x4);
 }
 
 static void gfx_v9_4_3_write_data_to_reg(struct amdgpu_ring *ring, int eng_sel,
@@ -236,7 +236,7 @@ static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring)
 	/* scratch_reg0_offset is 32bit even with full XCD config */
 	uint32_t scratch_reg0_offset;
 
-	scratch_reg0_offset = SOC15_REG_OFFSET(GC, ring->xcc_id, regSCRATCH_REG0);
+	scratch_reg0_offset = SOC15_REG_OFFSET(GC, GET_INST(GC, ring->xcc_id), regSCRATCH_REG0);
 	WREG32(scratch_reg0_offset, 0xCAFEDEAD);
 
 	r = amdgpu_ring_alloc(ring, 3);
@@ -324,9 +324,9 @@ static uint64_t gfx_v9_4_3_get_gpu_clock_counter(struct amdgpu_device *adev)
 
 	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
-	WREG32_SOC15(GC, 0, regRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
-	clock = (uint64_t)RREG32_SOC15(GC, 0, regRLC_GPU_CLOCK_COUNT_LSB) |
-		((uint64_t)RREG32_SOC15(GC, 0, regRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+	WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CAPTURE_GPU_CLOCK_COUNT, 1);
+	clock = (uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_LSB) |
+		((uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
 	amdgpu_gfx_off_ctrl(adev, true);
 
@@ -598,24 +598,24 @@ static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
 	else
 		data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SH_INDEX, sh_num);
 
-	WREG32_SOC15_RLC_SHADOW_EX(reg, GC, xcc_id, regGRBM_GFX_INDEX, data);
+	WREG32_SOC15_RLC_SHADOW_EX(reg, GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX, data);
 }
 
 static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address)
 {
-	WREG32_SOC15_RLC(GC, 0, regSQ_IND_INDEX,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, 0), regSQ_IND_INDEX,
 		(wave << SQ_IND_INDEX__WAVE_ID__SHIFT) |
 		(simd << SQ_IND_INDEX__SIMD_ID__SHIFT) |
 		(address << SQ_IND_INDEX__INDEX__SHIFT) |
 		(SQ_IND_INDEX__FORCE_READ_MASK));
-	return RREG32_SOC15(GC, 0, regSQ_IND_DATA);
+	return RREG32_SOC15(GC, GET_INST(GC, 0), regSQ_IND_DATA);
 }
 
 static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 			   uint32_t wave, uint32_t thread,
 			   uint32_t regno, uint32_t num, uint32_t *out)
 {
-	WREG32_SOC15_RLC(GC, 0, regSQ_IND_INDEX,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, 0), regSQ_IND_INDEX,
 		(wave << SQ_IND_INDEX__WAVE_ID__SHIFT) |
 		(simd << SQ_IND_INDEX__SIMD_ID__SHIFT) |
 		(regno << SQ_IND_INDEX__INDEX__SHIFT) |
@@ -623,7 +623,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		(SQ_IND_INDEX__FORCE_READ_MASK) |
 		(SQ_IND_INDEX__AUTO_INCR_MASK));
 	while (num--)
-		*(out++) = RREG32_SOC15(GC, 0, regSQ_IND_DATA);
+		*(out++) = RREG32_SOC15(GC, GET_INST(GC, 0), regSQ_IND_DATA);
 }
 
 static void gfx_v9_4_3_read_wave_data(struct amdgpu_device *adev,
@@ -669,7 +669,7 @@ static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd
 static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 					u32 me, u32 pipe, u32 q, u32 vm)
 {
-	soc15_grbm_select(adev, me, pipe, q, vm, 0);
+	soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, 0));
 }
 
 static enum amdgpu_gfx_partition
@@ -719,7 +719,7 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 				    num_xcc_per_partition);
 		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID,
 				    i % num_xcc_per_partition);
-		WREG32_SOC15(GC, i, regCP_HYP_XCP_CTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, i), regCP_HYP_XCP_CTL, tmp);
 	}
 
 	if (adev->nbio.funcs->set_compute_partition_mode)
@@ -755,7 +755,7 @@ static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
 		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
 		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
 		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0;
-		gb_addr_config = RREG32_SOC15(GC, 0, regGB_ADDR_CONFIG);
+		gb_addr_config = RREG32_SOC15(GC, GET_INST(GC, 0), regGB_ADDR_CONFIG);
 		break;
 	default:
 		BUG();
@@ -957,8 +957,8 @@ static u32 gfx_v9_4_3_get_rb_active_bitmap(struct amdgpu_device *adev)
 {
 	u32 data, mask;
 
-	data = RREG32_SOC15(GC, 0, regCC_RB_BACKEND_DISABLE);
-	data |= RREG32_SOC15(GC, 0, regGC_USER_RB_BACKEND_DISABLE);
+	data = RREG32_SOC15(GC, GET_INST(GC, 0), regCC_RB_BACKEND_DISABLE);
+	data |= RREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_RB_BACKEND_DISABLE);
 
 	data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK;
 	data >>= GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT;
@@ -1014,21 +1014,21 @@ static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev, int xcc_id)
 
 	mutex_lock(&adev->srbm_mutex);
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		soc15_grbm_select(adev, 0, 0, 0, i, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id));
 		/* CP and shaders */
-		WREG32_SOC15_RLC(GC, xcc_id, regSH_MEM_CONFIG, sh_mem_config);
-		WREG32_SOC15_RLC(GC, xcc_id, regSH_MEM_BASES, sh_mem_bases);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_CONFIG, sh_mem_config);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_BASES, sh_mem_bases);
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
+	soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 	mutex_unlock(&adev->srbm_mutex);
 
 	/* Initialize all compute VMIDs to have no GDS, GWS, or OA
 	   acccess. These should be enabled by FW for target VMIDs. */
 	for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) {
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_BASE, 2 * i, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_SIZE, 2 * i, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_GWS_VMID0, i, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_OA_VMID0, i, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_BASE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_SIZE, 2 * i, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_GWS_VMID0, i, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_OA_VMID0, i, 0);
 	}
 }
 
@@ -1043,10 +1043,10 @@ static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 	 * access so that HWS firmware can save/restore entries.
 	 */
 	for (vmid = 1; vmid < AMDGPU_NUM_VMID; vmid++) {
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_BASE, 2 * vmid, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_VMID0_SIZE, 2 * vmid, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_GWS_VMID0, vmid, 0);
-		WREG32_SOC15_OFFSET(GC, xcc_id, regGDS_OA_VMID0, vmid, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_BASE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_SIZE, 2 * vmid, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_GWS_VMID0, vmid, 0);
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_OA_VMID0, vmid, 0);
 	}
 }
 
@@ -1057,42 +1057,42 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, i, GRBM_CNTL, READ_TIMEOUT, 0xff);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_CNTL, READ_TIMEOUT, 0xff);
 		gfx_v9_4_3_setup_rb(adev, i);
 	}
 
 	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
-	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, 0, regDB_DEBUG2);
+	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2);
 
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
 		for (j = 0; j < num_xcc; j++) {
-			soc15_grbm_select(adev, 0, 0, 0, i, j);
+			soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, j));
 			/* CP and shaders */
 			if (i == 0) {
 				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
 						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
 						    !!adev->gmc.noretry);
-				WREG32_SOC15_RLC(GC, j, regSH_MEM_CONFIG, tmp);
-				WREG32_SOC15_RLC(GC, j, regSH_MEM_BASES, 0);
+				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_CONFIG, tmp);
+				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_BASES, 0);
 			} else {
 				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
 						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
 						    !!adev->gmc.noretry);
-				WREG32_SOC15_RLC(GC, j, regSH_MEM_CONFIG, tmp);
+				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_CONFIG, tmp);
 				tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
 					(adev->gmc.private_aperture_start >> 48));
 				tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
 					(adev->gmc.shared_aperture_start >> 48));
-				WREG32_SOC15_RLC(GC, j, regSH_MEM_BASES, tmp);
+				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_BASES, tmp);
 			}
 		}
 	}
-	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
+	soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, 0));
 
 	mutex_unlock(&adev->srbm_mutex);
 
@@ -1105,18 +1105,18 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev,
 						   int xcc_id)
 {
-	WREG32_FIELD15_PREREG(GC, xcc_id, RLC_SRM_CNTL, SRM_ENABLE, 1);
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_SRM_CNTL, SRM_ENABLE, 1);
 }
 
 static void gfx_v9_4_3_init_csb(struct amdgpu_device *adev, int xcc_id)
 {
 	adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr);
 	/* csib */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_ADDR_HI),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_HI),
 			adev->gfx.rlc.clear_state_gpu_addr >> 32);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_ADDR_LO),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_LO),
 			adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, xcc_id, regRLC_CSIB_LENGTH),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_LENGTH),
 			adev->gfx.rlc.clear_state_size);
 }
 
@@ -1137,7 +1137,7 @@ static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 			      AMD_PG_SUPPORT_CP |
 			      AMD_PG_SUPPORT_GDS |
 			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
-		WREG32_SOC15(GC, 0, regRLC_JUMP_TABLE_RESTORE,
+		WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_JUMP_TABLE_RESTORE,
 		       adev->gfx.rlc.cp_table_gpu_addr >> 8);
 	}
 }
@@ -1146,9 +1146,9 @@ static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
-	data = RREG32_SOC15(GC, xcc_id, regCPC_PSP_DEBUG);
+	data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG);
 	data |= CPC_PSP_DEBUG__UTCL2IUGPAOVERRIDE_MASK;
-	WREG32_SOC15(GC, xcc_id, regCPC_PSP_DEBUG, data);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG, data);
 }
 
 static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
@@ -1160,16 +1160,16 @@ static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
 	switch (num_xcc) {
 	/* directly config VIRTUAL_XCC_ID to 0 for 1-XCC */
 	case 1:
-		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, 0x8);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, 0x8);
 		break;
 	case 2:
 		tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID);
 		tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
-		WREG32_SOC15(GC, xcc_id, regCP_HYP_XCP_CTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, tmp);
 
 		tmp = xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, PHYSICAL_XCC_ID);
 		tmp = tmp | (xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, XCC_DIE_ID));
-		WREG32_SOC15(GC, xcc_id, regCP_PSP_XCP_CTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_PSP_XCP_CTL, tmp);
 		break;
 	default:
 		break;
@@ -1181,7 +1181,7 @@ static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
 	uint32_t rlc_setting;
 
 	/* if RLC is not enabled, do nothing */
-	rlc_setting = RREG32_SOC15(GC, 0, regRLC_CNTL);
+	rlc_setting = RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CNTL);
 	if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK))
 		return false;
 
@@ -1195,11 +1195,11 @@ static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 
 	data = RLC_SAFE_MODE__CMD_MASK;
 	data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
-	WREG32_SOC15(GC, xcc_id, regRLC_SAFE_MODE, data);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data);
 
 	/* wait for RLC_SAFE_MODE */
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (!REG_GET_FIELD(RREG32_SOC15(GC, 0, regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
+		if (!REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
 			break;
 		udelay(1);
 	}
@@ -1210,7 +1210,7 @@ static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 	uint32_t data;
 
 	data = RLC_SAFE_MODE__CMD_MASK;
-	WREG32_SOC15(GC, xcc_id, regRLC_SAFE_MODE, data);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data);
 }
 
 static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
@@ -1247,7 +1247,7 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
 			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, xcc_id);
 			for (k = 0; k < adev->usec_timeout; k++) {
-				if (RREG32_SOC15(GC, 0, regRLC_SERDES_CU_MASTER_BUSY) == 0)
+				if (RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
 				udelay(1);
 			}
@@ -1270,7 +1270,7 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 		RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY_MASK |
 		RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY_MASK;
 	for (k = 0; k < adev->usec_timeout; k++) {
-		if ((RREG32_SOC15(GC, 0, regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0)
+		if ((RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0)
 			break;
 		udelay(1);
 	}
@@ -1283,13 +1283,13 @@ static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
 
 	/* These interrupts should be enabled to drive DS clock */
 
-	tmp = RREG32_SOC15(GC, xcc_id, regCP_INT_CNTL_RING0);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0);
 
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0);
 	tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0);
 
-	WREG32_SOC15(GC, xcc_id, regCP_INT_CNTL_RING0, tmp);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0, tmp);
 }
 
 static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
@@ -1298,7 +1298,7 @@ static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 0);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), RLC_CNTL, RLC_ENABLE_F32, 0);
 		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
 		gfx_v9_4_3_wait_for_rlc_serdes(adev, i);
 	}
@@ -1310,9 +1310,9 @@ static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
 		udelay(50);
-		WREG32_FIELD15_PREREG(GC, i, GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
 		udelay(50);
 	}
 }
@@ -1326,7 +1326,7 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, i, RLC_CNTL, RLC_ENABLE_F32, 1);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), RLC_CNTL, RLC_ENABLE_F32, 1);
 		udelay(50);
 
 		/* carrizo do enable cp interrupt after cp inited */
@@ -1337,18 +1337,18 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 
 #ifdef AMDGPU_RLC_DEBUG_RETRY
 		/* RLC_GPM_GENERAL_6 : RLC Ucode version */
-		rlc_ucode_ver = RREG32_SOC15(GC, i, regRLC_GPM_GENERAL_6);
+		rlc_ucode_ver = RREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_GENERAL_6);
 		if (rlc_ucode_ver == 0x108) {
 			dev_info(adev->dev,
 				 "Using rlc debug ucode. regRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n",
 				 rlc_ucode_ver, adev->gfx.rlc_fw_version);
 			/* RLC_GPM_TIMER_INT_3 : Timer interval in RefCLK cycles,
 			 * default is 0x9C4 to create a 100us interval */
-			WREG32_SOC15(GC, i, regRLC_GPM_TIMER_INT_3, 0x9C4);
+			WREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_TIMER_INT_3, 0x9C4);
 			/* RLC_GPM_GENERAL_12 : Minimum gap between wptr and rptr
 			 * to disable the page fault retry interrupts, default is
 			 * 0x100 (256) */
-			WREG32_SOC15(GC, i, regRLC_GPM_GENERAL_12, 0x100);
+			WREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_GENERAL_12, 0x100);
 		}
 #endif
 	}
@@ -1370,16 +1370,16 @@ static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev, int xcc_id)
 			   le32_to_cpu(hdr->header.ucode_array_offset_bytes));
 	fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
 
-	WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_ADDR,
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR,
 			RLCG_UCODE_LOADING_START_ADDRESS);
 	for (i = 0; i < fw_size; i++) {
 		if (amdgpu_emu_mode == 1 && i % 100 == 0) {
 			dev_info(adev->dev, "Write RLC ucode data %u DWs\n", i);
 			msleep(1);
 		}
-		WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++));
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++));
 	}
-	WREG32_SOC15(GC, xcc_id, regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version);
 
 	return 0;
 }
@@ -1393,7 +1393,7 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		/* disable CG */
-		WREG32_SOC15(GC, i, regRLC_CGCG_CGLS_CTRL, 0);
+		WREG32_SOC15(GC, GET_INST(GC, i), regRLC_CGCG_CGLS_CTRL, 0);
 
 		gfx_v9_4_3_init_pg(adev, i);
 
@@ -1415,7 +1415,7 @@ static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev,
 {
 	u32 reg, data;
 
-	reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
+	reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL);
 	if (amdgpu_sriov_is_pp_one_vf(adev))
 		data = RREG32_NO_KIQ(reg);
 	else
@@ -1425,9 +1425,9 @@ static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev,
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
 	if (amdgpu_sriov_is_pp_one_vf(adev))
-		WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
+		WREG32_SOC15_NO_KIQ(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
 	else
-		WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data);
+		WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data);
 }
 
 static const struct soc15_reg_rlcg rlcg_access_gc_9_4_3[] = {
@@ -1439,7 +1439,7 @@ static bool gfx_v9_4_3_check_rlcg_range(struct amdgpu_device *adev,
 					uint32_t offset,
 					struct soc15_reg_rlcg *entries, int arr_size)
 {
-	int i;
+	int i, inst;
 	uint32_t reg;
 
 	if (!entries)
@@ -1449,7 +1449,12 @@ static bool gfx_v9_4_3_check_rlcg_range(struct amdgpu_device *adev,
 		const struct soc15_reg_rlcg *entry;
 
 		entry = &entries[i];
-		reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg;
+		inst = adev->ip_map.logical_to_dev_inst ?
+			       adev->ip_map.logical_to_dev_inst(
+				       adev, entry->hwip, entry->instance) :
+			       entry->instance;
+		reg = adev->reg_offset[entry->hwip][inst][entry->segment] +
+		      entry->reg;
 		if (offset == reg)
 			return true;
 	}
@@ -1468,9 +1473,9 @@ static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev,
 					 bool enable, int xcc_id)
 {
 	if (enable) {
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_MEC_CNTL, 0);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MEC_CNTL, 0);
 	} else {
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_MEC_CNTL,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MEC_CNTL,
 			(CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
 		adev->gfx.kiq[xcc_id].ring.sched.ready = false;
 	}
@@ -1501,17 +1506,17 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev,
 	tmp = 0;
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, VMID, 0);
 	tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0);
-	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_CNTL, tmp);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_CNTL, tmp);
 
-	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_LO,
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_LO,
 		adev->gfx.mec.mec_fw_gpu_addr & 0xFFFFF000);
-	WREG32_SOC15(GC, xcc_id, regCP_CPC_IC_BASE_HI,
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_HI,
 		upper_32_bits(adev->gfx.mec.mec_fw_gpu_addr));
 
 	mec_ucode_addr_offset =
-		SOC15_REG_OFFSET(GC, xcc_id, regCP_MEC_ME1_UCODE_ADDR);
+		SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_MEC_ME1_UCODE_ADDR);
 	mec_ucode_data_offset =
-		SOC15_REG_OFFSET(GC, xcc_id, regCP_MEC_ME1_UCODE_DATA);
+		SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_MEC_ME1_UCODE_DATA);
 
 	/* MEC1 */
 	WREG32(mec_ucode_addr_offset, mec_hdr->jt_offset);
@@ -1532,12 +1537,12 @@ static void gfx_v9_4_3_kiq_setting(struct amdgpu_ring *ring, int xcc_id)
 	struct amdgpu_device *adev = ring->adev;
 
 	/* tell RLC which is KIQ queue */
-	tmp = RREG32_SOC15(GC, xcc_id, regRLC_CP_SCHEDULERS);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS);
 	tmp &= 0xffffff00;
 	tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue);
-	WREG32_SOC15_RLC(GC, xcc_id, regRLC_CP_SCHEDULERS, tmp);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp);
 	tmp |= 0x80;
-	WREG32_SOC15_RLC(GC, xcc_id, regRLC_CP_SCHEDULERS, tmp);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp);
 }
 
 static void gfx_v9_4_3_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd)
@@ -1580,14 +1585,14 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_hqd_eop_base_addr_hi = upper_32_bits(eop_base_addr);
 
 	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
-	tmp = RREG32_SOC15(GC, 0, regCP_HQD_EOP_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_EOP_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_EOP_CONTROL, EOP_SIZE,
 			(order_base_2(GFX9_MEC_HPD_SIZE / 4) - 1));
 
 	mqd->cp_hqd_eop_control = tmp;
 
 	/* enable doorbell? */
-	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PQ_DOORBELL_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_DOORBELL_CONTROL);
 
 	if (ring->use_doorbell) {
 		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
@@ -1617,7 +1622,7 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_mqd_base_addr_hi = upper_32_bits(ring->mqd_gpu_addr);
 
 	/* set MQD vmid to 0 */
-	tmp = RREG32_SOC15(GC, 0, regCP_MQD_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_MQD_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_MQD_CONTROL, VMID, 0);
 	mqd->cp_mqd_control = tmp;
 
@@ -1627,7 +1632,7 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_hqd_pq_base_hi = upper_32_bits(hqd_gpu_addr);
 
 	/* set up the HQD, this is similar to CP_RB0_CNTL */
-	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PQ_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, QUEUE_SIZE,
 			    (order_base_2(ring->ring_size / 4) - 1));
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE,
@@ -1654,23 +1659,23 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
 	ring->wptr = 0;
-	mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, 0, regCP_HQD_PQ_RPTR);
+	mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_RPTR);
 
 	/* set the vmid for the queue */
 	mqd->cp_hqd_vmid = 0;
 
-	tmp = RREG32_SOC15(GC, 0, regCP_HQD_PERSISTENT_STATE);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PERSISTENT_STATE);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PERSISTENT_STATE, PRELOAD_SIZE, 0x53);
 	mqd->cp_hqd_persistent_state = tmp;
 
 	/* set MIN_IB_AVAIL_SIZE */
-	tmp = RREG32_SOC15(GC, 0, regCP_HQD_IB_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_IB_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3);
 	mqd->cp_hqd_ib_control = tmp;
 
 	/* set static priority for a queue/ring */
 	gfx_v9_4_3_mqd_set_priority(ring, mqd);
-	mqd->cp_hqd_quantum = RREG32_SOC15(GC, 0, regCP_HQD_QUANTUM);
+	mqd->cp_hqd_quantum = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_QUANTUM);
 
 	/* map_queues packet doesn't need activate the queue,
 	 * so only kiq need set this field.
@@ -1690,94 +1695,94 @@ static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
 	/* disable wptr polling */
 	WREG32_FIELD15_PREREG(GC, xcc_id, CP_PQ_WPTR_POLL_CNTL, EN, 0);
 
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_BASE_ADDR,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR,
 	       mqd->cp_hqd_eop_base_addr_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_BASE_ADDR_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR_HI,
 	       mqd->cp_hqd_eop_base_addr_hi);
 
 	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_EOP_CONTROL,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_CONTROL,
 	       mqd->cp_hqd_eop_control);
 
 	/* enable doorbell? */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL,
 	       mqd->cp_hqd_pq_doorbell_control);
 
 	/* disable the queue if it's active */
-	if (RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1) {
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST, 1);
+	if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1) {
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, 1);
 		for (j = 0; j < adev->usec_timeout; j++) {
-			if (!(RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1))
+			if (!(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1))
 				break;
 			udelay(1);
 		}
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST,
 		       mqd->cp_hqd_dequeue_request);
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR,
 		       mqd->cp_hqd_pq_rptr);
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO,
 		       mqd->cp_hqd_pq_wptr_lo);
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI,
 		       mqd->cp_hqd_pq_wptr_hi);
 	}
 
 	/* set the pointer to the MQD */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_BASE_ADDR,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR,
 	       mqd->cp_mqd_base_addr_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_BASE_ADDR_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR_HI,
 	       mqd->cp_mqd_base_addr_hi);
 
 	/* set MQD vmid to 0 */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_MQD_CONTROL,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_CONTROL,
 	       mqd->cp_mqd_control);
 
 	/* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_BASE,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE,
 	       mqd->cp_hqd_pq_base_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_BASE_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE_HI,
 	       mqd->cp_hqd_pq_base_hi);
 
 	/* set up the HQD, this is similar to CP_RB0_CNTL */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_CONTROL,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_CONTROL,
 	       mqd->cp_hqd_pq_control);
 
 	/* set the wb address whether it's enabled or not */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR_REPORT_ADDR,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR,
 				mqd->cp_hqd_pq_rptr_report_addr_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR_REPORT_ADDR_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR_HI,
 				mqd->cp_hqd_pq_rptr_report_addr_hi);
 
 	/* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_POLL_ADDR,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR,
 	       mqd->cp_hqd_pq_wptr_poll_addr_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
 	       mqd->cp_hqd_pq_wptr_poll_addr_hi);
 
 	/* enable the doorbell if requested */
 	if (ring->use_doorbell) {
-		WREG32_SOC15(GC, xcc_id, regCP_MEC_DOORBELL_RANGE_LOWER,
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_LOWER,
 					(adev->doorbell_index.kiq * 2) << 2);
-		WREG32_SOC15(GC, xcc_id, regCP_MEC_DOORBELL_RANGE_UPPER,
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_UPPER,
 				(adev->doorbell_index.userqueue_end * 2) << 2);
 	}
 
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL,
 	       mqd->cp_hqd_pq_doorbell_control);
 
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO,
 	       mqd->cp_hqd_pq_wptr_lo);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI,
 	       mqd->cp_hqd_pq_wptr_hi);
 
 	/* set the vmid for the queue */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_VMID, mqd->cp_hqd_vmid);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_VMID, mqd->cp_hqd_vmid);
 
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PERSISTENT_STATE,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE,
 	       mqd->cp_hqd_persistent_state);
 
 	/* activate the queue */
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_ACTIVE,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE,
 	       mqd->cp_hqd_active);
 
 	if (ring->use_doorbell)
@@ -1792,12 +1797,12 @@ static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring, int xcc_id)
 	int j;
 
 	/* disable the queue if it's active */
-	if (RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1) {
+	if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1) {
 
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST, 1);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, 1);
 
 		for (j = 0; j < adev->usec_timeout; j++) {
-			if (!(RREG32_SOC15(GC, xcc_id, regCP_HQD_ACTIVE) & 1))
+			if (!(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1))
 				break;
 			udelay(1);
 		}
@@ -1806,21 +1811,21 @@ static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring, int xcc_id)
 			DRM_DEBUG("KIQ dequeue request failed.\n");
 
 			/* Manual disable if dequeue request times out */
-			WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_ACTIVE, 0);
+			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, 0);
 		}
 
-		WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_DEQUEUE_REQUEST,
+		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST,
 		      0);
 	}
 
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_IQ_TIMER, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_IB_CONTROL, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PERSISTENT_STATE, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_DOORBELL_CONTROL, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_RPTR, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_HI, 0);
-	WREG32_SOC15_RLC(GC, xcc_id, regCP_HQD_PQ_WPTR_LO, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IQ_TIMER, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IB_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, 0);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, 0);
 
 	return 0;
 }
@@ -1848,19 +1853,19 @@ static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		ring->wptr = 0;
 		amdgpu_ring_clear_ring(ring);
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
 		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
-		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
 		gfx_v9_4_3_mqd_init(ring);
 		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
-		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.kiq[xcc_id].mqd_backup)
@@ -1888,9 +1893,9 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
-		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, xcc_id);
+		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
 		gfx_v9_4_3_mqd_init(ring);
-		soc15_grbm_select(adev, 0, 0, 0, 0, xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
@@ -2049,9 +2054,9 @@ static int gfx_v9_4_3_hw_fini(void *handle)
 			mutex_lock(&adev->srbm_mutex);
 			soc15_grbm_select(adev, adev->gfx.kiq[i].ring.me,
 					adev->gfx.kiq[i].ring.pipe,
-					adev->gfx.kiq[i].ring.queue, 0, i);
+					adev->gfx.kiq[i].ring.queue, 0, GET_INST(GC, i));
 			gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[i].ring, i);
-			soc15_grbm_select(adev, 0, 0, 0, 0, i);
+			soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, i));
 			mutex_unlock(&adev->srbm_mutex);
 		}
 
@@ -2085,7 +2090,7 @@ static bool gfx_v9_4_3_is_idle(void *handle)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		if (REG_GET_FIELD(RREG32_SOC15(GC, i, regGRBM_STATUS),
+		if (REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, i), regGRBM_STATUS),
 					GRBM_STATUS, GUI_ACTIVE))
 			return false;
 	}
@@ -2112,7 +2117,7 @@ static int gfx_v9_4_3_soft_reset(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	/* GRBM_STATUS */
-	tmp = RREG32_SOC15(GC, 0, regGRBM_STATUS);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_STATUS);
 	if (tmp & (GRBM_STATUS__PA_BUSY_MASK | GRBM_STATUS__SC_BUSY_MASK |
 		   GRBM_STATUS__BCI_BUSY_MASK | GRBM_STATUS__SX_BUSY_MASK |
 		   GRBM_STATUS__TA_BUSY_MASK | GRBM_STATUS__VGT_BUSY_MASK |
@@ -2131,7 +2136,7 @@ static int gfx_v9_4_3_soft_reset(void *handle)
 	}
 
 	/* GRBM_STATUS2 */
-	tmp = RREG32_SOC15(GC, 0, regGRBM_STATUS2);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_STATUS2);
 	if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY))
 		grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
 						GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
@@ -2145,17 +2150,17 @@ static int gfx_v9_4_3_soft_reset(void *handle)
 		gfx_v9_4_3_cp_compute_enable(adev, false, 0);
 
 		if (grbm_soft_reset) {
-			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
 			tmp |= grbm_soft_reset;
 			dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
-			WREG32_SOC15(GC, 0, regGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+			WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
+			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
 
 			udelay(50);
 
 			tmp &= ~grbm_soft_reset;
-			WREG32_SOC15(GC, 0, regGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, regGRBM_SOFT_RESET);
+			WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
+			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
 		}
 
 		/* Wait a little for things to settle down */
@@ -2174,22 +2179,22 @@ static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring,
 
 	/* GDS Base */
 	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
-				   SOC15_REG_OFFSET(GC, 0, regGDS_VMID0_BASE) + 2 * vmid,
+				   SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_VMID0_BASE) + 2 * vmid,
 				   gds_base);
 
 	/* GDS Size */
 	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
-				   SOC15_REG_OFFSET(GC, 0, regGDS_VMID0_SIZE) + 2 * vmid,
+				   SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_VMID0_SIZE) + 2 * vmid,
 				   gds_size);
 
 	/* GWS */
 	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
-				   SOC15_REG_OFFSET(GC, 0, regGDS_GWS_VMID0) + vmid,
+				   SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_GWS_VMID0) + vmid,
 				   gws_size << GDS_GWS_VMID0__SIZE__SHIFT | gws_base);
 
 	/* OA */
 	gfx_v9_4_3_write_data_to_reg(ring, 0, false,
-				   SOC15_REG_OFFSET(GC, 0, regGDS_OA_VMID0) + vmid,
+				   SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_OA_VMID0) + vmid,
 				   (1 << (oa_size + oa_base)) - (1 << oa_base));
 }
 
@@ -2260,7 +2265,7 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
 		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
-		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE);
 
 		data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
@@ -2270,28 +2275,28 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 		data |= RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK;
 
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* MGLS is a global flag to control all MGLS in GFX */
 		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) {
 			/* 2 - RLC memory Light sleep */
 			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_RLC_LS) {
-				def = data = RREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL);
+				def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL);
 				data |= RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
 				if (def != data)
-					WREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL, data);
+					WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL, data);
 			}
 			/* 3 - CP memory Light sleep */
 			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) {
-				def = data = RREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL);
+				def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL);
 				data |= CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
 				if (def != data)
-					WREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL, data);
+					WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL, data);
 			}
 		}
 	} else {
 		/* 1 - MGCG_OVERRIDE */
-		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE);
 
 		data |= (RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK |
 			 RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
@@ -2299,20 +2304,20 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 			 RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK);
 
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* 2 - disable MGLS in RLC */
-		data = RREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL);
+		data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL);
 		if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) {
 			data &= ~RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK;
-			WREG32_SOC15(GC, xcc_id, regRLC_MEM_SLP_CNTL, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL, data);
 		}
 
 		/* 3 - disable MGLS in CP */
-		data = RREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL);
+		data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL);
 		if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) {
 			data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
-			WREG32_SOC15(GC, xcc_id, regCP_MEM_SLP_CNTL, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL, data);
 		}
 	}
 
@@ -2327,7 +2332,7 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
-		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE);
+		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE);
 		/* unset CGCG override */
 		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK;
 		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS)
@@ -2336,10 +2341,10 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 			data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK;
 		/* update CGCG and CGLS override bits */
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regRLC_CGTT_MGCG_OVERRIDE, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data);
 
 		/* enable cgcg FSM(0x0000363F) */
-		def = RREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL);
+		def = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL);
 
 		if (adev->asic_type == CHIP_ARCTURUS)
 			data = (0x2000 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
@@ -2351,21 +2356,21 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 			data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) |
 				RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK;
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data);
 
 		/* set IDLE_POLL_COUNT(0x00900100) */
-		def = RREG32_SOC15(GC, xcc_id, regCP_RB_WPTR_POLL_CNTL);
+		def = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL);
 		data = (0x0100 << CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT) |
 			(0x0090 << CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT);
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regCP_RB_WPTR_POLL_CNTL, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL, data);
 	} else {
-		def = data = RREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL);
+		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL);
 		/* reset CGCG/CGLS bits */
 		data &= ~(RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK);
 		/* disable cgcg and cgls in FSM */
 		if (def != data)
-			WREG32_SOC15(GC, xcc_id, regRLC_CGCG_CGLS_CTRL, data);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data);
 	}
 
 	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
@@ -2444,12 +2449,12 @@ static void gfx_v9_4_3_get_clockgating_state(void *handle, u64 *flags)
 		*flags = 0;
 
 	/* AMD_CG_SUPPORT_GFX_MGCG */
-	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_CGTT_MGCG_OVERRIDE));
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_CGTT_MGCG_OVERRIDE));
 	if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK))
 		*flags |= AMD_CG_SUPPORT_GFX_MGCG;
 
 	/* AMD_CG_SUPPORT_GFX_CGCG */
-	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_CGCG_CGLS_CTRL));
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_CGCG_CGLS_CTRL));
 	if (data & RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK)
 		*flags |= AMD_CG_SUPPORT_GFX_CGCG;
 
@@ -2458,12 +2463,12 @@ static void gfx_v9_4_3_get_clockgating_state(void *handle, u64 *flags)
 		*flags |= AMD_CG_SUPPORT_GFX_CGLS;
 
 	/* AMD_CG_SUPPORT_GFX_RLC_LS */
-	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regRLC_MEM_SLP_CNTL));
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_MEM_SLP_CNTL));
 	if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK)
 		*flags |= AMD_CG_SUPPORT_GFX_RLC_LS | AMD_CG_SUPPORT_GFX_MGLS;
 
 	/* AMD_CG_SUPPORT_GFX_CP_LS */
-	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, regCP_MEM_SLP_CNTL));
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regCP_MEM_SLP_CNTL));
 	if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK)
 		*flags |= AMD_CG_SUPPORT_GFX_CP_LS | AMD_CG_SUPPORT_GFX_MGLS;
 }
@@ -2634,7 +2639,7 @@ static void gfx_v9_4_3_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
 		amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 		amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
 					 WRITE_DATA_DST_SEL(0) | WR_CONFIRM));
-		amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, 0, regCPC_INT_STATUS));
+		amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regCPC_INT_STATUS));
 		amdgpu_ring_write(ring, 0);
 		amdgpu_ring_write(ring, 0x20000000); /* src_id is 178 */
 	}
@@ -2710,16 +2715,16 @@ static void gfx_v9_4_3_set_compute_eop_interrupt_state(struct amdgpu_device *ade
 	if (me == 1) {
 		switch (pipe) {
 		case 0:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE0_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE0_INT_CNTL);
 			break;
 		case 1:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE1_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE1_INT_CNTL);
 			break;
 		case 2:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE2_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE2_INT_CNTL);
 			break;
 		case 3:
-			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, xcc_id, regCP_ME1_PIPE3_INT_CNTL);
+			mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE3_INT_CNTL);
 			break;
 		default:
 			DRM_DEBUG("invalid pipe %d\n", pipe);
@@ -2760,7 +2765,7 @@ static int gfx_v9_4_3_set_priv_reg_fault_state(struct amdgpu_device *adev,
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
 		for (i = 0; i < num_xcc; i++)
-			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
+			WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0,
 				PRIV_REG_INT_ENABLE,
 				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
 		break;
@@ -2783,7 +2788,7 @@ static int gfx_v9_4_3_set_priv_inst_fault_state(struct amdgpu_device *adev,
 	case AMDGPU_IRQ_STATE_DISABLE:
 	case AMDGPU_IRQ_STATE_ENABLE:
 		for (i = 0; i < num_xcc; i++)
-			WREG32_FIELD15_PREREG(GC, i, CP_INT_CNTL_RING0,
+			WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0,
 				PRIV_INSTR_INT_ENABLE,
 				state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
 		break;
@@ -2943,16 +2948,16 @@ static void gfx_v9_4_3_emit_wave_limit_cs(struct amdgpu_ring *ring,
 
 	switch (pipe) {
 	case 0:
-		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS0);
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS0);
 		break;
 	case 1:
-		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS1);
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS1);
 		break;
 	case 2:
-		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS2);
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS2);
 		break;
 	case 3:
-		wcl_cs_reg = SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_CS3);
+		wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS3);
 		break;
 	default:
 		DRM_DEBUG("invalid pipe %d\n", pipe);
@@ -2974,7 +2979,7 @@ static void gfx_v9_4_3_emit_wave_limit(struct amdgpu_ring *ring, bool enable)
 	 */
 	val = enable ? 0x1f : 0x07ffffff;
 	amdgpu_ring_emit_wreg(ring,
-			      SOC15_REG_OFFSET(GC, 0, regSPI_WCL_PIPE_PERCENT_GFX),
+			      SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_GFX),
 			      val);
 
 	/* Restrict waves for normal/low priority compute queues as well
@@ -3161,15 +3166,15 @@ static void gfx_v9_4_3_set_user_cu_inactive_bitmap(struct amdgpu_device *adev,
 	data = bitmap << GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
 	data &= GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
 
-	WREG32_SOC15(GC, 0, regGC_USER_SHADER_ARRAY_CONFIG, data);
+	WREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG, data);
 }
 
 static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev)
 {
 	u32 data, mask;
 
-	data = RREG32_SOC15(GC, 0, regCC_GC_SHADER_ARRAY_CONFIG);
-	data |= RREG32_SOC15(GC, 0, regGC_USER_SHADER_ARRAY_CONFIG);
+	data = RREG32_SOC15(GC, GET_INST(GC, 0), regCC_GC_SHADER_ARRAY_CONFIG);
+	data |= RREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG);
 
 	data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
 	data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index e35365ab3f1fb..c26ac0662c7e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -35,7 +35,7 @@
 
 static u64 gfxhub_v1_2_get_mc_fb_offset(struct amdgpu_device *adev)
 {
-	return (u64)RREG32_SOC15(GC, 0, regMC_VM_FB_OFFSET) << 24;
+	return (u64)RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_FB_OFFSET) << 24;
 }
 
 static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
@@ -48,12 +48,12 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
-		WREG32_SOC15_OFFSET(GC, i,
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, i),
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
 				    hub->ctx_addr_distance * vmid,
 				    lower_32_bits(page_table_base));
 
-		WREG32_SOC15_OFFSET(GC, i,
+		WREG32_SOC15_OFFSET(GC, GET_INST(GC, i),
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
 				    hub->ctx_addr_distance * vmid,
 				    upper_32_bits(page_table_base));
@@ -79,31 +79,31 @@ static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		if (adev->gmc.pdb0_bo) {
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
 				     (u32)(adev->gmc.fb_start >> 12));
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
 				     (u32)(adev->gmc.fb_start >> 44));
 
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
 				     (u32)(adev->gmc.gart_end >> 12));
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
 				     (u32)(adev->gmc.gart_end >> 44));
 		} else {
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
 				     (u32)(adev->gmc.gart_start >> 12));
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32,
 				     (u32)(adev->gmc.gart_start >> 44));
 
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32,
 				     (u32)(adev->gmc.gart_end >> 12));
-			WREG32_SOC15(GC, i,
+			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32,
 				     (u32)(adev->gmc.gart_end >> 44));
 		}
@@ -119,13 +119,13 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		/* Program the AGP BAR */
-		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BASE, 0);
-		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
-		WREG32_SOC15_RLC(GC, i, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BASE, 0);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24);
 
 		if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) {
 			/* Program the system aperture low logical page number. */
-			WREG32_SOC15_RLC(GC, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
+			WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_LOW_ADDR,
 				min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
 
 			if (adev->apu_flags & AMD_APU_IS_RAVEN2)
@@ -136,44 +136,44 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 				* aperture high address (add 1) to get rid of the VM
 				* fault and hardware hang.
 				*/
-				WREG32_SOC15_RLC(GC, i,
+				WREG32_SOC15_RLC(GC, GET_INST(GC, i),
 						 regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
 						 max((adev->gmc.fb_end >> 18) + 0x1,
 						     adev->gmc.agp_end >> 18));
 			else
-				WREG32_SOC15_RLC(GC, i,
+				WREG32_SOC15_RLC(GC, GET_INST(GC, i),
 					regMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
 					max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
 
 			/* Set default page address. */
 			value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr);
-			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB,
 				     (u32)(value >> 12));
-			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB,
 				     (u32)(value >> 44));
 
 			/* Program "protection fault". */
-			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
+			WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32,
 				     (u32)(adev->dummy_page_addr >> 12));
-			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
+			WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32,
 				     (u32)((u64)adev->dummy_page_addr >> 44));
 
-			tmp = RREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL2);
+			tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL2);
 			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2,
 					    ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1);
-			WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
+			WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL2, tmp);
 		}
 
 		/* In the case squeezing vram into GART aperture, we don't use
 		 * FB aperture and AGP aperture. Disable them.
 		 */
 		if (adev->gmc.pdb0_bo) {
-			WREG32_SOC15(GC, i, regMC_VM_FB_LOCATION_TOP, 0);
-			WREG32_SOC15(GC, i, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
-			WREG32_SOC15(GC, i, regMC_VM_AGP_TOP, 0);
-			WREG32_SOC15(GC, i, regMC_VM_AGP_BOT, 0xFFFFFF);
-			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
-			WREG32_SOC15(GC, i, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, 0);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, 0);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_BOT, 0xFFFFFF);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF);
+			WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0);
 		}
 	}
 }
@@ -186,7 +186,7 @@ static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		/* Setup TLB control */
-		tmp = RREG32_SOC15(GC, i, regMC_VM_MX_L1_TLB_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_MX_L1_TLB_CNTL);
 
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL,
 				    ENABLE_L1_TLB, 1);
@@ -200,7 +200,7 @@ static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 				    MTYPE, MTYPE_UC);/* XXX for emulation. */
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1);
 
-		WREG32_SOC15_RLC(GC, i, regMC_VM_MX_L1_TLB_CNTL, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_MX_L1_TLB_CNTL, tmp);
 	}
 }
 
@@ -212,7 +212,7 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		/* Setup L2 cache */
-		tmp = RREG32_SOC15(GC, i, regVM_L2_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1);
 		/* XXX for emulation, Refer to closed source code.*/
@@ -221,12 +221,12 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0);
-		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL, tmp);
 
-		tmp = RREG32_SOC15(GC, i, regVM_L2_CNTL2);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_CNTL2);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1);
-		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL2, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL2, tmp);
 
 		tmp = regVM_L2_CNTL3_DEFAULT;
 		if (adev->gmc.translate_further) {
@@ -238,7 +238,7 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3,
 					    L2_CACHE_BIGK_FRAGMENT_SIZE, 6);
 		}
-		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL3, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL3, tmp);
 
 		tmp = regVM_L2_CNTL4_DEFAULT;
 		if (adev->gmc.xgmi.connected_to_cpu) {
@@ -248,7 +248,7 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0);
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
 		}
-		WREG32_SOC15_RLC(GC, i, regVM_L2_CNTL4, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL4, tmp);
 	}
 }
 
@@ -259,7 +259,7 @@ static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		tmp = RREG32_SOC15(GC, i, regVM_CONTEXT0_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
 				adev->gmc.vmid0_page_table_depth);
@@ -267,7 +267,7 @@ static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 				adev->gmc.vmid0_page_table_block_size);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL,
 				    RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0);
-		WREG32_SOC15(GC, i, regVM_CONTEXT0_CNTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL, tmp);
 	}
 }
 
@@ -277,23 +277,23 @@ static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
 			     0XFFFFFFFF);
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32,
 			     0x0000000F);
 
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32,
 			     0);
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32,
 			     0);
 
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0);
-		WREG32_SOC15(GC, i,
+		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0);
 	}
 }
@@ -316,7 +316,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	for (j = 0; j < num_xcc; j++) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		for (i = 0; i <= 14; i++) {
-			tmp = RREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL, i);
+			tmp = RREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT1_CNTL, i);
 			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
 			tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH,
 					    num_level);
@@ -348,19 +348,19 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 					    !adev->gmc.noretry ||
 					    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ||
 					    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3));
-			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT1_CNTL,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT1_CNTL,
 					    i * hub->ctx_distance, tmp);
-			WREG32_SOC15_OFFSET(GC, j,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j),
 					    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32,
 					    i * hub->ctx_addr_distance, 0);
-			WREG32_SOC15_OFFSET(GC, j,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j),
 					    regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32,
 					    i * hub->ctx_addr_distance, 0);
-			WREG32_SOC15_OFFSET(GC, j,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j),
 					    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32,
 					    i * hub->ctx_addr_distance,
 					    lower_32_bits(adev->vm_manager.max_pfn - 1));
-			WREG32_SOC15_OFFSET(GC, j,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j),
 					    regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32,
 					    i * hub->ctx_addr_distance,
 					    upper_32_bits(adev->vm_manager.max_pfn - 1));
@@ -378,9 +378,9 @@ static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 
 		for (i = 0 ; i < 18; ++i) {
-			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32,
 					    i * hub->eng_addr_distance, 0xffffffff);
-			WREG32_SOC15_OFFSET(GC, j, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32,
 					    i * hub->eng_addr_distance, 0x1f);
 		}
 	}
@@ -398,9 +398,9 @@ static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 		 * VF copy registers so vbios post doesn't program them, for
 		 * SRIOV driver need to program them
 		 */
-			WREG32_SOC15_RLC(GC, i, regMC_VM_FB_LOCATION_BASE,
+			WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE,
 				     adev->gmc.vram_start >> 24);
-			WREG32_SOC15_RLC(GC, i, regMC_VM_FB_LOCATION_TOP,
+			WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP,
 				     adev->gmc.vram_end >> 24);
 		}
 	}
@@ -432,23 +432,23 @@ static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		/* Disable all tables */
 		for (i = 0; i < 16; i++)
-			WREG32_SOC15_OFFSET(GC, j, regVM_CONTEXT0_CNTL,
+			WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT0_CNTL,
 					    i * hub->ctx_distance, 0);
 
 		/* Setup TLB control */
-		tmp = RREG32_SOC15(GC, j, regMC_VM_MX_L1_TLB_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, j), regMC_VM_MX_L1_TLB_CNTL);
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0);
 		tmp = REG_SET_FIELD(tmp,
 					MC_VM_MX_L1_TLB_CNTL,
 					ENABLE_ADVANCED_DRIVER_MODEL,
 					0);
-		WREG32_SOC15_RLC(GC, j, regMC_VM_MX_L1_TLB_CNTL, tmp);
+		WREG32_SOC15_RLC(GC, GET_INST(GC, j), regMC_VM_MX_L1_TLB_CNTL, tmp);
 
 		/* Setup L2 cache */
-		tmp = RREG32_SOC15(GC, j, regVM_L2_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0);
-		WREG32_SOC15(GC, j, regVM_L2_CNTL, tmp);
-		WREG32_SOC15(GC, j, regVM_L2_CNTL3, 0);
+		WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL3, 0);
 	}
 }
 
@@ -466,7 +466,7 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		tmp = RREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
@@ -497,7 +497,7 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 			tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
 					CRASH_ON_RETRY_FAULT, 1);
 		}
-		WREG32_SOC15(GC, i, regVM_L2_PROTECTION_FAULT_CNTL, tmp);
+		WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL, tmp);
 	}
 }
 
@@ -511,24 +511,24 @@ static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 
 		hub->ctx0_ptb_addr_lo32 =
-			SOC15_REG_OFFSET(GC, i,
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i),
 				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
 		hub->ctx0_ptb_addr_hi32 =
-			SOC15_REG_OFFSET(GC, i,
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i),
 				regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
 		hub->vm_inv_eng0_sem =
-			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_SEM);
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_SEM);
 		hub->vm_inv_eng0_req =
-			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_REQ);
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_REQ);
 		hub->vm_inv_eng0_ack =
-			SOC15_REG_OFFSET(GC, i, regVM_INVALIDATE_ENG0_ACK);
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_ACK);
 		hub->vm_context0_cntl =
-			SOC15_REG_OFFSET(GC, i, regVM_CONTEXT0_CNTL);
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL);
 		hub->vm_l2_pro_fault_status =
-			SOC15_REG_OFFSET(GC, i,
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i),
 				regVM_L2_PROTECTION_FAULT_STATUS);
 		hub->vm_l2_pro_fault_cntl =
-			SOC15_REG_OFFSET(GC, i, regVM_L2_PROTECTION_FAULT_CNTL);
+			SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL);
 
 		hub->ctx_distance = regVM_CONTEXT1_CNTL -
 				regVM_CONTEXT0_CNTL;
@@ -551,9 +551,9 @@ static int gfxhub_v1_2_get_xgmi_info(struct amdgpu_device *adev)
 	u32 max_region;
 	u64 seg_size;
 
-	xgmi_lfb_cntl = RREG32_SOC15(GC, 0, regMC_VM_XGMI_LFB_CNTL);
+	xgmi_lfb_cntl = RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_XGMI_LFB_CNTL);
 	seg_size = REG_GET_FIELD(
-		RREG32_SOC15(GC, 0, regMC_VM_XGMI_LFB_SIZE),
+		RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_XGMI_LFB_SIZE),
 		MC_VM_XGMI_LFB_SIZE, PF_LFB_SIZE) << 24;
 	max_region =
 		REG_GET_FIELD(xgmi_lfb_cntl, MC_VM_XGMI_LFB_CNTL, PF_MAX_REGION);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index eb35096756b8e..39e4406da4aee 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -24,6 +24,9 @@
 #ifndef __SOC15_COMMON_H__
 #define __SOC15_COMMON_H__
 
+/* GET_INST returns the physical instance corresponding to a logical instance */
+#define GET_INST(ip, inst) (adev->ip_map.logical_to_dev_inst? adev->ip_map.logical_to_dev_inst(adev, ip##_HWIP, inst): inst)
+
 /* Register Access Macros */
 #define SOC15_REG_OFFSET(ip, inst, reg)	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg)
 #define SOC15_REG_OFFSET1(ip, inst, reg, offset) \
-- 
GitLab


From 4db6f200a5fef12666ab66e54bacb65c5d675e9d Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 16:26:49 +0530
Subject: [PATCH 1093/1791] drm/amdgpu: Add mask for SDMA instances

Add a mask of SDMA instances available for use. On certain ASIC configs,
not all SDMA instances are available for software use.

v2:
	Change sdma mask type to uint32_t (Le)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 632b77138fe40..62afb282a3ff1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -48,6 +48,8 @@ enum amdgpu_sdma_irq {
 	AMDGPU_SDMA_IRQ_LAST
 };
 
+#define NUM_SDMA(x) hweight32(x)
+
 struct amdgpu_sdma_instance {
 	/* SDMA firmware */
 	const struct firmware	*fw;
@@ -75,6 +77,7 @@ struct amdgpu_sdma {
 	struct amdgpu_irq_src	srbm_write_irq;
 
 	int			num_instances;
+	uint32_t 		sdma_mask;
 	int			num_inst_per_aid;
 	uint32_t                    srbm_soft_reset;
 	bool			has_page_queue;
-- 
GitLab


From f8b34a0518701bae8bb02fabe129f01b1dc33336 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 21:04:39 +0530
Subject: [PATCH 1094/1791] drm/amdgpu: Use instance table for sdma 4.4.2

For ASICs with sdma IP v4.4.2, add mapping for logical to physical
instances.

v2:
	Register accesses on bare metal should be based on physical
instance. Use GET_INST() to get physical instance.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   |  7 ++--
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c        |  7 +++-
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c      | 38 ++++++++++++++-----
 3 files changed, 38 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 5daec0b45545b..772774615cb81 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -43,9 +43,10 @@ static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
 					unsigned int engine_id,
 					unsigned int queue_id)
 {
-	uint32_t sdma_engine_reg_base = SOC15_REG_OFFSET(SDMA0, engine_id,
-					regSDMA_RLC0_RB_CNTL) -
-					regSDMA_RLC0_RB_CNTL;
+	uint32_t sdma_engine_reg_base =
+		SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, engine_id),
+				 regSDMA_RLC0_RB_CNTL) -
+		regSDMA_RLC0_RB_CNTL;
 	uint32_t retval = sdma_engine_reg_base +
 		  queue_id * (regSDMA_RLC1_RB_CNTL - regSDMA_RLC0_RB_CNTL);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 5a5bab665f600..e6c1004e4e765 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -75,7 +75,10 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
 	u32 doorbell_range = 0, doorbell_ctrl = 0;
-	int aid_id = adev->sdma.instance[instance].aid_id;
+	int aid_id, dev_inst;
+
+	dev_inst = GET_INST(SDMA0, instance);
+	aid_id = adev->sdma.instance[instance].aid_id;
 
 	if (use_doorbell == false)
 		return;
@@ -93,7 +96,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 		REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL,
 			S2A_DOORBELL_PORT1_RANGE_SIZE, doorbell_size);
 
-	switch (instance % adev->sdma.num_inst_per_aid) {
+	switch (dev_inst % adev->sdma.num_inst_per_aid) {
 	case 0:
 		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1) +
 			4 * aid_id, doorbell_range);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 9b53174925f8c..4350939105c51 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -57,7 +57,9 @@ static void sdma_v4_4_2_set_irq_funcs(struct amdgpu_device *adev);
 static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev,
 		u32 instance, u32 offset)
 {
-	return (adev->reg_offset[SDMA0_HWIP][instance][0] + offset);
+	u32 dev_inst = GET_INST(SDMA0, instance);
+
+	return (adev->reg_offset[SDMA0_HWIP][dev_inst][0] + offset);
 }
 
 static unsigned sdma_v4_4_2_seq_to_irq_id(int seq_num)
@@ -1475,16 +1477,31 @@ static int sdma_v4_4_2_process_trap_irq(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
-	uint32_t instance;
+	uint32_t instance, i;
 
 	DRM_DEBUG("IH: SDMA trap\n");
 	instance = sdma_v4_4_2_irq_id_to_seq(entry->client_id);
-	instance += node_id_to_phys_map[entry->node_id] *
-			adev->sdma.num_inst_per_aid;
+
+	/* Client id gives the SDMA instance in AID. To know the exact SDMA
+	 * instance, interrupt entry gives the node id which corresponds to the AID instance.
+	 * Match node id with the AID id associated with the SDMA instance. */
+	for (i = instance; i < adev->sdma.num_instances;
+	     i += adev->sdma.num_inst_per_aid) {
+		if (adev->sdma.instance[i].aid_id ==
+		    node_id_to_phys_map[entry->node_id])
+			break;
+	}
+
+	if (i >= adev->sdma.num_instances) {
+		dev_WARN_ONCE(
+			adev->dev, 1,
+			"Couldn't find the right sdma instance in trap handler");
+		return 0;
+	}
 
 	switch (entry->ring_id) {
 	case 0:
-		amdgpu_fence_process(&adev->sdma.instance[instance].ring);
+		amdgpu_fence_process(&adev->sdma.instance[i].ring);
 		break;
 	default:
 		break;
@@ -1717,12 +1734,12 @@ static void sdma_v4_4_2_get_clockgating_state(void *handle, u64 *flags)
 		*flags = 0;
 
 	/* AMD_CG_SUPPORT_SDMA_MGCG */
-	data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, regSDMA_CLK_CTRL));
+	data = RREG32(SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, 0), regSDMA_CLK_CTRL));
 	if (!(data & SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK))
 		*flags |= AMD_CG_SUPPORT_SDMA_MGCG;
 
 	/* AMD_CG_SUPPORT_SDMA_LS */
-	data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, regSDMA_POWER_CNTL));
+	data = RREG32(SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, 0), regSDMA_POWER_CNTL));
 	if (data & SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK)
 		*flags |= AMD_CG_SUPPORT_SDMA_LS;
 }
@@ -1809,7 +1826,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = {
 
 static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev)
 {
-	int i;
+	int i, dev_inst;
 
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		adev->sdma.instance[i].ring.funcs = &sdma_v4_4_2_ring_funcs;
@@ -1820,7 +1837,10 @@ static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev)
 			adev->sdma.instance[i].page.me = i;
 		}
 
-		adev->sdma.instance[i].aid_id = i / adev->sdma.num_inst_per_aid;
+		dev_inst = GET_INST(SDMA0, i);
+		/* AID to which SDMA belongs depends on physical instance */
+		adev->sdma.instance[i].aid_id =
+			dev_inst / adev->sdma.num_inst_per_aid;
 	}
 }
 
-- 
GitLab


From 7aa8a266aaa25e9e2f85d9d2d594cdff6b5635f2 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 6 Jul 2022 13:50:45 +0530
Subject: [PATCH 1095/1791] drm/amdgpu: Fix GRBM programming sequence

It needs to be done only for XCC instances in non-AID0. Use the physical
instance to determine non-AID0 XCC instances.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index e9c12b4970f97..43126f7b70eac 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -186,11 +186,14 @@ static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev)
 
 static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
 {
-	int i, num_xcc;
+	int i, num_xcc, dev_inst;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 2; i < num_xcc; i++)
-		WREG32_SOC15(GC, GET_INST(GC, i), regGRBM_MCM_ADDR, 0x4);
+	for (i = 0; i < num_xcc; i++) {
+		dev_inst = GET_INST(GC, i);
+		if (dev_inst >= 2)
+			WREG32_SOC15(GC, dev_inst, regGRBM_MCM_ADDR, 0x4);
+	}
 }
 
 static void gfx_v9_4_3_write_data_to_reg(struct amdgpu_ring *ring, int eng_sel,
-- 
GitLab


From 2e10ced47ff261d9dba439c5bcecf68d0c1b7de5 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 19:39:34 -0400
Subject: [PATCH 1096/1791] drm/amdgpu/nbio: add vcn doorbell multiple AIDs
 support

Update vcn doorbell range to support multiple AIDs.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h |  1 +
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c   | 20 +++++++++++++++++---
 2 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
index 70a5f030d5f7d..22483dc663518 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h
@@ -42,6 +42,7 @@
 #define regUVD_JRBC_STATUS_INTERNAL_OFFSET				0x4089
 #define regUVD_JPEG_PITCH_INTERNAL_OFFSET				0x4043
 #define regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET			0x4094
+#define regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET			0x1bffe
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index e6c1004e4e765..bcd3e08d01cd2 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -202,6 +202,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 					 int doorbell_index, int instance)
 {
 	u32 doorbell_range = 0, doorbell_ctrl = 0;
+	u32 aid_id = instance;
 
 	if (use_doorbell) {
 		doorbell_range = REG_SET_FIELD(doorbell_range,
@@ -212,6 +213,11 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
 				0x9);
+		if (aid_id)
+			doorbell_range = REG_SET_FIELD(doorbell_range,
+					DOORBELL0_CTRL_ENTRY_0,
+					DOORBELL0_FENCE_ENABLE_ENTRY,
+					0x4);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
@@ -228,6 +234,13 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
+
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17) +
+					aid_id, doorbell_range);
+		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL) * 4
+				+ AMDGPU_SMN_TARGET_AID(aid_id)
+				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+				doorbell_ctrl);
 	} else {
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
@@ -235,10 +248,11 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_RANGE_SIZE, 0);
-	}
 
-	WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, doorbell_range);
-	WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL, doorbell_ctrl);
+		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, doorbell_range);
+		WREG32(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL),
+						doorbell_ctrl);
+	}
 }
 
 static void nbio_v7_9_enable_doorbell_aperture(struct amdgpu_device *adev,
-- 
GitLab


From d4ad24a0b796ad429403bf17ba97ee7e2470ad68 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 2 Jul 2022 19:53:36 -0400
Subject: [PATCH 1097/1791] drm/amdgpu/jpeg: add JPEG multiple AIDs support

Add JPEG multiple AIDs support.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 380 ++++++++++++++---------
 1 file changed, 227 insertions(+), 153 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index e12e3646c49ad..aa14a6619e9a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -63,6 +63,8 @@ static int jpeg_v4_0_3_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->jpeg.num_jpeg_rings = AMDGPU_MAX_JPEG_RINGS;
+
 	jpeg_v4_0_3_set_dec_ring_funcs(adev);
 	jpeg_v4_0_3_set_irq_funcs(adev);
 
@@ -80,12 +82,12 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int i, r;
+	int i, j, r;
 
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
+	for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 		/* JPEG TRAP */
 		r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-				amdgpu_ih_srcid_jpeg[i], &adev->jpeg.inst->irq);
+				amdgpu_ih_srcid_jpeg[j], &adev->jpeg.inst->irq);
 		if (r)
 			return r;
 	}
@@ -98,22 +100,27 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		ring = &adev->jpeg.inst->ring_dec[i];
-		ring->use_doorbell = true;
-		ring->vm_hub = AMDGPU_MMHUB0(0);
-		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + i;
-		sprintf(ring->name, "jpeg_dec_%d", i);
-		r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
-					AMDGPU_RING_PRIO_DEFAULT, NULL);
-		if (r)
-			return r;
-
-		adev->jpeg.internal.jpeg_pitch[i] =
-			regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET;
-		adev->jpeg.inst->external.jpeg_pitch[i] =
-			SOC15_REG_OFFSET1(JPEG, 0, regUVD_JRBC0_UVD_JRBC_SCRATCH0,
-			(i?(0x40 * i - 0xc80):0));
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			ring = &adev->jpeg.inst[i].ring_dec[j];
+			ring->use_doorbell = true;
+			ring->vm_hub = AMDGPU_MMHUB0(adev->jpeg.inst[i].aid_id);
+			ring->doorbell_index =
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + j + 9 * i;
+			sprintf(ring->name, "jpeg_dec_%d.%d", i, j);
+			r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
+						AMDGPU_RING_PRIO_DEFAULT, NULL);
+			if (r)
+				return r;
+
+			adev->jpeg.internal.jpeg_pitch[j] =
+				regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET;
+			adev->jpeg.inst[i].external.jpeg_pitch[j] =
+				SOC15_REG_OFFSET1(JPEG, i, regUVD_JRBC0_UVD_JRBC_SCRATCH0,
+				(j?(0x40 * j - 0xc80):0));
+		}
 	}
 
 	return 0;
@@ -149,22 +156,30 @@ static int jpeg_v4_0_3_sw_fini(void *handle)
 static int jpeg_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
-	int i, r;
+	struct amdgpu_ring *ring;
+	int i, j, r;
 
-	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-			(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		ring = adev->jpeg.inst[i].ring_dec;
 
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		ring = &adev->jpeg.inst->ring_dec[i];
 		if (ring->use_doorbell)
-			WREG32_SOC15_OFFSET(VCN, 0, regVCN_JPEG_DB_CTRL,
-				(ring->pipe?(ring->pipe - 0x15):0),
-				ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
-				VCN_JPEG_DB_CTRL__EN_MASK);
-		r = amdgpu_ring_test_helper(ring);
-		if (r)
-			return r;
+			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i,
+				adev->jpeg.inst[i].aid_id);
+
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			ring = &adev->jpeg.inst[i].ring_dec[j];
+			if (ring->use_doorbell)
+				WREG32_SOC15_OFFSET(VCN, i, regVCN_JPEG_DB_CTRL,
+					(ring->pipe?(ring->pipe - 0x15):0),
+					ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
+					VCN_JPEG_DB_CTRL__EN_MASK);
+			r = amdgpu_ring_test_helper(ring);
+			if (r)
+				return r;
+		}
 	}
 	DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n");
 
@@ -233,48 +248,52 @@ static int jpeg_v4_0_3_resume(void *handle)
 	return r;
 }
 
-static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
+static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
 	int i;
 
-	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
-	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
+	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) {
 		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
-	else
+		data &= (~(JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1));
+	} else {
 		data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	}
 
 	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
+	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL, data);
 
-	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
+	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE);
 	data &= ~(JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
 		data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
-	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
+	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE, data);
 }
 
-static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
+static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
 	int i;
 
-	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL);
-	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG)
+	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL);
+	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) {
 		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
-	else
+		data |= (JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1);
+	} else {
 		data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+	}
 
 	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(JPEG, 0, regJPEG_CGC_CTRL, data);
+	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL, data);
 
-	data = RREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE);
+	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE);
 	data |= (JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
 		data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
-	WREG32_SOC15(JPEG, 0, regJPEG_CGC_GATE, data);
+	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE, data);
 }
 
 /**
@@ -286,58 +305,63 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
  */
 static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
-	int i;
-
-	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
-		1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
-	SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_PGFSM_STATUS,
-		UVD_PGFSM_STATUS__UVDJ_PWR_ON <<
-		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
-		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
-
-	/* disable anti hang mechanism */
-	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_POWER_STATUS), 0,
-		~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
-
-	/* JPEG disable CGC */
-	jpeg_v4_0_3_disable_clock_gating(adev);
-
-	/* MJPEG global tiling registers */
-	WREG32_SOC15(JPEG, 0, regJPEG_DEC_GFX8_ADDR_CONFIG,
-		adev->gfx.config.gb_addr_config);
-	WREG32_SOC15(JPEG, 0, regJPEG_DEC_GFX10_ADDR_CONFIG,
-		adev->gfx.config.gb_addr_config);
-
-	/* enable JMI channel */
-	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI_CNTL), 0,
-		~UVD_JMI_CNTL__SOFT_RESET_MASK);
-
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
-
-		ring = &adev->jpeg.inst->ring_dec[i];
-
-		/* enable System Interrupt for JRBC */
-		WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regJPEG_SYS_INT_EN),
-			JPEG_SYS_INT_EN__DJRBC0_MASK << i,
-			~(JPEG_SYS_INT_EN__DJRBC0_MASK << i));
-
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, reg_offset, 0);
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
-			(0x00000001L | 0x00000002L));
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
-			reg_offset, lower_32_bits(ring->gpu_addr));
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
-			reg_offset, upper_32_bits(ring->gpu_addr));
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_RPTR, reg_offset, 0);
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR, reg_offset, 0);
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
-			0x00000002L);
-		WREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_SIZE, reg_offset,
-			ring->ring_size / 4);
-		ring->wptr = RREG32_SOC15_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
-			reg_offset);
+	struct amdgpu_ring *ring;
+	int i, j;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		WREG32_SOC15(JPEG, i, regUVD_PGFSM_CONFIG,
+			1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+		SOC15_WAIT_ON_RREG(JPEG, i, regUVD_PGFSM_STATUS,
+			UVD_PGFSM_STATUS__UVDJ_PWR_ON <<
+			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
+
+		/* disable anti hang mechanism */
+		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JPEG_POWER_STATUS), 0,
+			~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+
+		/* JPEG disable CGC */
+		jpeg_v4_0_3_disable_clock_gating(adev, i);
+
+		/* MJPEG global tiling registers */
+		WREG32_SOC15(JPEG, i, regJPEG_DEC_GFX8_ADDR_CONFIG,
+			adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(JPEG, i, regJPEG_DEC_GFX10_ADDR_CONFIG,
+			adev->gfx.config.gb_addr_config);
+
+		/* enable JMI channel */
+		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JMI_CNTL), 0,
+			~UVD_JMI_CNTL__SOFT_RESET_MASK);
+
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
+
+			ring = &adev->jpeg.inst[i].ring_dec[j];
+
+			/* enable System Interrupt for JRBC */
+			WREG32_P(SOC15_REG_OFFSET(JPEG, i, regJPEG_SYS_INT_EN),
+				JPEG_SYS_INT_EN__DJRBC0_MASK << j,
+				~(JPEG_SYS_INT_EN__DJRBC0_MASK << j));
+
+			WREG32_SOC15_OFFSET(JPEG, i,
+				regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
+				(0x00000001L | 0x00000002L));
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
+				reg_offset, lower_32_bits(ring->gpu_addr));
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
+				reg_offset, upper_32_bits(ring->gpu_addr));
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_RPTR, reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_WPTR, reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
+				0x00000002L);
+			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_SIZE, reg_offset,
+				ring->ring_size / 4);
+			ring->wptr = RREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+				reg_offset);
+		}
 	}
 
 	return 0;
@@ -352,24 +376,31 @@ static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
  */
 static int jpeg_v4_0_3_stop(struct amdgpu_device *adev)
 {
-	/* reset JMI */
-	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI_CNTL),
-		UVD_JMI_CNTL__SOFT_RESET_MASK,
-		~UVD_JMI_CNTL__SOFT_RESET_MASK);
+	int i;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
 
-	jpeg_v4_0_3_enable_clock_gating(adev);
+		/* reset JMI */
+		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JMI_CNTL),
+			UVD_JMI_CNTL__SOFT_RESET_MASK,
+			~UVD_JMI_CNTL__SOFT_RESET_MASK);
 
-	/* enable anti hang mechanism */
-	WREG32_P(SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_POWER_STATUS),
-		UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK,
-		~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+		jpeg_v4_0_3_enable_clock_gating(adev, i);
 
-	WREG32_SOC15(JPEG, 0, regUVD_PGFSM_CONFIG,
-		2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
-	SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_PGFSM_STATUS,
-		UVD_PGFSM_STATUS__UVDJ_PWR_OFF <<
-		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
-		UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
+		/* enable anti hang mechanism */
+		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JPEG_POWER_STATUS),
+			UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK,
+			~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+
+		WREG32_SOC15(JPEG, i, regUVD_PGFSM_CONFIG,
+			2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+		SOC15_WAIT_ON_RREG(JPEG, i, regUVD_PGFSM_STATUS,
+			UVD_PGFSM_STATUS__UVDJ_PWR_OFF <<
+			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
+	}
 
 	return 0;
 }
@@ -502,10 +533,28 @@ static void jpeg_v4_0_3_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
 		0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE4));
 	amdgpu_ring_write(ring, 0);
 
+	if (ring->adev->jpeg.inst[ring->me].aid_id) {
+		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET,
+			0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0));
+		amdgpu_ring_write(ring, 0x4);
+	} else {
+		amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
+		amdgpu_ring_write(ring, 0);
+	}
+
 	amdgpu_ring_write(ring,	PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
 	amdgpu_ring_write(ring, 0x3fbc);
 
+	if (ring->adev->jpeg.inst[ring->me].aid_id) {
+		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET,
+			0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0));
+		amdgpu_ring_write(ring, 0x0);
+	} else {
+		amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6));
+		amdgpu_ring_write(ring, 0);
+	}
+
 	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
 		0, 0, PACKETJ_TYPE0));
 	amdgpu_ring_write(ring, 0x1);
@@ -651,15 +700,19 @@ static bool jpeg_v4_0_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool ret;
-	int i;
-
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
-
-		ret &= ((RREG32_SOC15_OFFSET(JPEG, 0,
-				regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset) &
-				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
-				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+	int i, j;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
+
+			ret &= ((RREG32_SOC15_OFFSET(JPEG, i,
+					regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset) &
+					UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
+					UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+		}
 	}
 
 	return ret;
@@ -669,17 +722,20 @@ static int jpeg_v4_0_3_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int ret;
-	int i;
+	int i, j;
 
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		unsigned int reg_offset = (i?(0x40 * i - 0xc80):0);
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
 
-		ret &= SOC15_WAIT_ON_RREG_OFFSET(JPEG, 0,
-			regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset,
-			UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
-			UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+			ret &= SOC15_WAIT_ON_RREG_OFFSET(JPEG, i,
+				regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset,
+				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
+				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+		}
 	}
-
 	return ret;
 }
 
@@ -688,15 +744,19 @@ static int jpeg_v4_0_3_set_clockgating_state(void *handle,
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	int i;
 
-	if (enable) {
-		if (!jpeg_v4_0_3_is_idle(handle))
-			return -EBUSY;
-		jpeg_v4_0_3_enable_clock_gating(adev);
-	} else {
-		jpeg_v4_0_3_disable_clock_gating(adev);
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		if (enable) {
+			if (!jpeg_v4_0_3_is_idle(handle))
+				return -EBUSY;
+			jpeg_v4_0_3_enable_clock_gating(adev, i);
+		} else {
+			jpeg_v4_0_3_disable_clock_gating(adev, i);
+		}
 	}
-
 	return 0;
 }
 
@@ -732,32 +792,35 @@ static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
+	uint32_t i;
+
+	i = node_id_to_phys_map[entry->node_id];
 	DRM_DEV_DEBUG(adev->dev, "IH: JPEG TRAP\n");
 
 	switch (entry->src_id) {
 	case VCN_4_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[0]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[0]);
 		break;
 	case VCN_4_0__SRCID__JPEG1_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[1]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[1]);
 		break;
 	case VCN_4_0__SRCID__JPEG2_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[2]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[2]);
 		break;
 	case VCN_4_0__SRCID__JPEG3_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[3]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[3]);
 		break;
 	case VCN_4_0__SRCID__JPEG4_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[4]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[4]);
 		break;
 	case VCN_4_0__SRCID__JPEG5_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[5]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[5]);
 		break;
 	case VCN_4_0__SRCID__JPEG6_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[6]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[6]);
 		break;
 	case VCN_4_0__SRCID__JPEG7_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst->ring_dec[7]);
+		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[7]);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -798,7 +861,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
 		8 + /* jpeg_v4_0_3_dec_ring_emit_vm_flush */
-		18 + 18 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */
+		22 + 22 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */
 		8 + 16,
 	.emit_ib_size = 22, /* jpeg_v4_0_3_dec_ring_emit_ib */
 	.emit_ib = jpeg_v4_0_3_dec_ring_emit_ib,
@@ -819,12 +882,17 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 
 static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	int i;
-
-	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) {
-		adev->jpeg.inst->ring_dec[i].funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
-		adev->jpeg.inst->ring_dec[i].me = 0;
-		adev->jpeg.inst->ring_dec[i].pipe = i;
+	int i, j;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
+			adev->jpeg.inst[i].ring_dec[j].funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
+			adev->jpeg.inst[i].ring_dec[j].me = i;
+			adev->jpeg.inst[i].ring_dec[j].pipe = j;
+		}
+		adev->jpeg.inst[i].aid_id = i / adev->jpeg.num_inst_per_aid;
 	}
 	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
 }
@@ -836,7 +904,13 @@ static const struct amdgpu_irq_src_funcs jpeg_v4_0_3_irq_funcs = {
 
 static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 {
-	adev->jpeg.inst->irq.num_types = adev->jpeg.num_jpeg_rings;
+	int i;
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+		if (adev->jpeg.harvest_config & (1 << i))
+			continue;
+		adev->jpeg.inst->irq.num_types += adev->jpeg.num_jpeg_rings;
+	}
 	adev->jpeg.inst->irq.funcs = &jpeg_v4_0_3_irq_funcs;
 }
 
-- 
GitLab


From 7229bd6fe02865a9fc324b4f062268f53190b5f4 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 11 Jul 2022 11:05:05 -0400
Subject: [PATCH 1098/1791] drm/amdgpu/vcn: update clock gate setting for VCN
 4.0.3

Update clock gate setting.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 30 +++++++++++++------------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 0b2b97593baca..a9f06f3b00eb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -424,13 +424,14 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indir
 static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 {
 	uint32_t data;
+	int inst_idx = 0;
 
-	/* VCN disable CGC */
-	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
-		data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
-	else
-		data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK;
+		return;
+
+	/* VCN disable CGC */
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
 	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
@@ -517,11 +518,11 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
 {
 	uint32_t reg_data = 0;
 
-	/* enable sw clock gating control */
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
-		reg_data = 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
-	else
-		reg_data = 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+		return;
+
+	/* enable sw clock gating control */
+	reg_data = 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 	reg_data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	reg_data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
 	reg_data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK |
@@ -563,13 +564,14 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
 static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 {
 	uint32_t data;
+	int inst_idx = 0;
 
-	/* enable VCN CGC */
-	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
-		data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
-	else
-		data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
+		return;
+
+	/* enable VCN CGC */
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
 	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
-- 
GitLab


From 762702ebd324596def832a2b2a1cfd24fff78338 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 11 Jul 2022 11:06:46 -0400
Subject: [PATCH 1099/1791] drm/amdgpu/vcn: add vcn multiple AIDs support

add vcn multiple AIDs support.

v2: squash in FW setting fix (Alex)

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 810 +++++++++++++-----------
 1 file changed, 434 insertions(+), 376 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index a9f06f3b00eb2..8db50a50b868b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -79,9 +79,8 @@ static int vcn_v4_0_3_early_init(void *handle)
 static int vcn_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
-	int r;
+	int i, r;
 
 	r = amdgpu_vcn_sw_init(adev);
 	if (r)
@@ -99,20 +98,27 @@ static int vcn_v4_0_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	ring = &adev->vcn.inst->ring_dec;
-	ring->use_doorbell = true;
-	ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1);
-	ring->vm_hub = AMDGPU_MMHUB0(0);
-	sprintf(ring->name, "vcn_dec");
-	r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
-			     AMDGPU_RING_PRIO_DEFAULT,
-			     &adev->vcn.inst->sched_score);
-	if (r)
-		return r;
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 
-	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
-	fw_shared->present_flag_0 = 0;
-	fw_shared->sq.is_enabled = cpu_to_le32(true);
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+
+		ring = &adev->vcn.inst[i].ring_dec;
+		ring->use_doorbell = true;
+		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i;
+		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
+		sprintf(ring->name, "vcn_dec_%d", i);
+		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
+				     AMDGPU_RING_PRIO_DEFAULT,
+				     &adev->vcn.inst[i].sched_score);
+		if (r)
+			return r;
+
+		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
+		fw_shared->present_flag_0 = 0;
+		fw_shared->sq.is_enabled = cpu_to_le32(true);
+	}
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
 		adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode;
@@ -130,15 +136,18 @@ static int vcn_v4_0_3_sw_init(void *handle)
 static int vcn_v4_0_3_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int r, idx;
+	int i, r, idx;
 
 	if (drm_dev_enter(&adev->ddev, &idx)) {
-		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
-
-		fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
-		fw_shared->present_flag_0 = 0;
-		fw_shared->sq.is_enabled = cpu_to_le32(false);
-
+		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+			volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+
+			if (adev->vcn.harvest_config & (1 << i))
+				continue;
+			fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
+			fw_shared->present_flag_0 = 0;
+			fw_shared->sq.is_enabled = cpu_to_le32(false);
+		}
 		drm_dev_exit(idx);
 	}
 
@@ -161,18 +170,30 @@ static int vcn_v4_0_3_sw_fini(void *handle)
 static int vcn_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
-	int r;
+	struct amdgpu_ring *ring;
+	int i, r;
 
-	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
-	if (ring->use_doorbell)
-		WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
-			ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
-			VCN_RB4_DB_CTRL__EN_MASK);
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		ring = &adev->vcn.inst[i].ring_dec;
 
-	r = amdgpu_ring_test_helper(ring);
+		if (ring->use_doorbell)
+			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i,
+					adev->vcn.inst[i].aid_id);
 
+		if (ring->use_doorbell)
+			WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
+				ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
+				VCN_RB4_DB_CTRL__EN_MASK);
+
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
+			goto done;
+	}
+
+done:
 	if (!r)
 		DRM_DEV_INFO(adev->dev, "VCN decode initialized successfully(under %s).\n",
 			(adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode");
@@ -245,10 +266,11 @@ static int vcn_v4_0_3_resume(void *handle)
  * vcn_v4_0_3_mc_resume - memory controller programming
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number
  *
  * Let the VCN memory controller know it's offsets
  */
-static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev)
+static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t offset, size;
 	const struct common_firmware_header *hdr;
@@ -258,50 +280,46 @@ static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev)
 
 	/* cache window 0: fw */
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
-		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
-			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_lo));
-		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
-			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_hi));
-		WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0, 0);
+		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_lo));
+		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_hi));
+		WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET0, 0);
 		offset = 0;
 	} else {
-		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
-			lower_32_bits(adev->vcn.inst->gpu_addr));
-		WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
-			upper_32_bits(adev->vcn.inst->gpu_addr));
+		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
+		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
 		offset = size;
-		if (amdgpu_emu_mode == 1)
-			/* No signed header here */
-			WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0, 0);
-		else
-			WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET0,
-				AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
+		WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET0,
+			AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
 	}
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE0, size);
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE0, size);
 
 	/* cache window 1: stack */
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW,
-		lower_32_bits(adev->vcn.inst->gpu_addr + offset));
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH,
-		upper_32_bits(adev->vcn.inst->gpu_addr + offset));
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET1, 0);
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE);
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET1, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE);
 
 	/* cache window 2: context */
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW,
-		lower_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH,
-		upper_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_OFFSET2, 0);
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE);
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET2, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE);
 
 	/* non-cache window */
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW,
-		lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr));
-	WREG32_SOC15(VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH,
-		upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr));
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0, 0);
-	WREG32_SOC15(VCN, 0, regUVD_VCPU_NONCACHE_SIZE0,
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW,
+		lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr));
+	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH,
+		upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr));
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_NONCACHE_OFFSET0, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_NONCACHE_SIZE0,
 		AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)));
 }
 
@@ -309,11 +327,12 @@ static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev)
  * vcn_v4_0_mc_resume_dpg_mode - memory controller programming for dpg mode
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number index
  * @indirect: indirectly write sram
  *
  * Let the VCN memory controller know it's offsets with dpg mode
  */
-static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indirect)
+static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect)
 {
 	uint32_t offset, size;
 	const struct common_firmware_header *hdr;
@@ -324,93 +343,97 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indir
 	/* cache window 0: fw */
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
 		if (!indirect) {
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
-				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_lo), 0, indirect);
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN +
+					inst_idx].tmr_mc_addr_lo), 0, indirect);
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
-				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN].tmr_mc_addr_hi), 0, indirect);
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+				(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN +
+					inst_idx].tmr_mc_addr_hi), 0, indirect);
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect);
 		} else {
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), 0, 0, indirect);
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), 0, 0, indirect);
-			WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 				VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect);
 		}
 		offset = 0;
 	} else {
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW),
-			lower_32_bits(adev->vcn.inst->gpu_addr), 0, indirect);
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect);
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH),
-			upper_32_bits(adev->vcn.inst->gpu_addr), 0, indirect);
+			upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect);
 		offset = size;
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_OFFSET0),
-			0, 0, indirect);
+			AMDGPU_UVD_FIRMWARE_OFFSET >> 3, 0, indirect);
 	}
 
 	if (!indirect)
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_SIZE0), size, 0, indirect);
 	else
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_SIZE0), 0, 0, indirect);
 
 	/* cache window 1: stack */
 	if (!indirect) {
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW),
-			lower_32_bits(adev->vcn.inst->gpu_addr + offset), 0, indirect);
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect);
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH),
-			upper_32_bits(adev->vcn.inst->gpu_addr + offset), 0, indirect);
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect);
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect);
 	} else {
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), 0, 0, indirect);
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), 0, 0, indirect);
-		WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+		WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect);
 	}
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE, 0, indirect);
 
 	/* cache window 2: context */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW),
-			lower_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset +
+				AMDGPU_VCN_STACK_SIZE), 0, indirect);
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH),
-			upper_32_bits(adev->vcn.inst->gpu_addr + offset + AMDGPU_VCN_STACK_SIZE), 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset +
+				AMDGPU_VCN_STACK_SIZE), 0, indirect);
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_OFFSET2), 0, 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE, 0, indirect);
 
 	/* non-cache window */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW),
-			lower_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect);
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH),
-			upper_32_bits(adev->vcn.inst->fw_shared.gpu_addr), 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+			upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect);
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 			VCN, 0, regUVD_VCPU_NONCACHE_SIZE0),
 			AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)), 0, indirect);
 
 	/* VCN global tiling registers */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_GFX8_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect);
 }
 
@@ -418,13 +441,13 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, bool indir
  * vcn_v4_0_3_disable_clock_gating - disable VCN clock gating
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number
  *
  * Disable clock gating for VCN block
  */
-static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
+static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
-	int inst_idx = 0;
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
 		return;
@@ -434,9 +457,9 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 	data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_CGC_GATE);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_GATE);
 	data &= ~(UVD_CGC_GATE__SYS_MASK
 		| UVD_CGC_GATE__MPEG2_MASK
 		| UVD_CGC_GATE__REGS_MASK
@@ -450,10 +473,10 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 		| UVD_CGC_GATE__VCPU_MASK
 		| UVD_CGC_GATE__MMSCH_MASK);
 
-	WREG32_SOC15(VCN, 0, regUVD_CGC_GATE, data);
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_CGC_GATE, 0,  0xFFFFFFFF);
+	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_GATE, data);
+	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_CGC_GATE, 0,  0xFFFFFFFF);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
 	data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK
 		| UVD_CGC_CTRL__MPEG2_MODE_MASK
 		| UVD_CGC_CTRL__REGS_MODE_MASK
@@ -466,9 +489,9 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 		| UVD_CGC_CTRL__WCB_MODE_MASK
 		| UVD_CGC_CTRL__VCPU_MODE_MASK
 		| UVD_CGC_CTRL__MMSCH_MODE_MASK);
-	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_GATE);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_GATE);
 	data |= (UVD_SUVD_CGC_GATE__SRE_MASK
 		| UVD_SUVD_CGC_GATE__SIT_MASK
 		| UVD_SUVD_CGC_GATE__SMP_MASK
@@ -490,9 +513,9 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 		| UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK
 		| UVD_SUVD_CGC_GATE__SDB_VP9_MASK
 		| UVD_SUVD_CGC_GATE__IME_HEVC_MASK);
-	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_GATE, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_GATE, data);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL);
 	data &= ~(UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
@@ -501,7 +524,7 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
 		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
-	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL, data);
 }
 
 /**
@@ -509,12 +532,13 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev)
  *
  * @adev: amdgpu_device pointer
  * @sram_sel: sram select
+ * @inst_idx: instance number index
  * @indirect: indirectly write sram
  *
  * Disable clock gating for VCN block with dpg mode
  */
 static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev, uint8_t sram_sel,
-				uint8_t indirect)
+				int inst_idx, uint8_t indirect)
 {
 	uint32_t reg_data = 0;
 
@@ -538,19 +562,19 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
 		 UVD_CGC_CTRL__LRBBM_MODE_MASK |
 		 UVD_CGC_CTRL__WCB_MODE_MASK |
 		 UVD_CGC_CTRL__VCPU_MODE_MASK);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_CGC_CTRL), reg_data, sram_sel, indirect);
 
 	/* turn off clock gating */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_CGC_GATE), 0, sram_sel, indirect);
 
 	/* turn on SUVD clock gating */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_SUVD_CGC_GATE), 1, sram_sel, indirect);
 
 	/* turn on sw mode in UVD_SUVD_CGC_CTRL */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_SUVD_CGC_CTRL), 0, sram_sel, indirect);
 }
 
@@ -558,13 +582,13 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
  * vcn_v4_0_enable_clock_gating - enable VCN clock gating
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number
  *
  * Enable clock gating for VCN block
  */
-static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
+static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
-	int inst_idx = 0;
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
 		return;
@@ -574,9 +598,9 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 	data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
 	data |= (UVD_CGC_CTRL__SYS_MODE_MASK
 		| UVD_CGC_CTRL__MPEG2_MODE_MASK
 		| UVD_CGC_CTRL__REGS_MODE_MASK
@@ -588,9 +612,9 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 		| UVD_CGC_CTRL__LRBBM_MODE_MASK
 		| UVD_CGC_CTRL__WCB_MODE_MASK
 		| UVD_CGC_CTRL__VCPU_MODE_MASK);
-	WREG32_SOC15(VCN, 0, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL);
 	data |= (UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
@@ -599,48 +623,51 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev)
 		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
-	WREG32_SOC15(VCN, 0, regUVD_SUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL, data);
 }
 
 /**
  * vcn_v4_0_3_start_dpg_mode - VCN start with dpg mode
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number index
  * @indirect: indirectly write sram
  *
  * Start VCN block with dpg mode
  */
-static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
+static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect)
 {
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	volatile struct amdgpu_vcn4_fw_shared *fw_shared =
+						adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 
 	/* disable register anti-hang mechanism */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_POWER_STATUS), 1,
+	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1,
 		~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 	/* enable dynamic power gating mode */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_POWER_STATUS);
+	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_POWER_STATUS);
 	tmp |= UVD_POWER_STATUS__UVD_PG_MODE_MASK;
 	tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_POWER_STATUS, tmp);
+	WREG32_SOC15(VCN, inst_idx, regUVD_POWER_STATUS, tmp);
 
 	if (indirect)
-		adev->vcn.inst->dpg_sram_curr_addr = (uint32_t *)adev->vcn.inst->dpg_sram_cpu_addr;
+		adev->vcn.inst[inst_idx].dpg_sram_curr_addr =
+				(uint32_t *)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr;
 
 	/* enable clock gating */
-	vcn_v4_0_3_disable_clock_gating_dpg_mode(adev, 0, indirect);
+	vcn_v4_0_3_disable_clock_gating_dpg_mode(adev, 0, inst_idx, indirect);
 
 	/* enable VCPU clock */
 	tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT);
 	tmp |= UVD_VCPU_CNTL__CLK_EN_MASK;
 	tmp |= UVD_VCPU_CNTL__BLK_RST_MASK;
 
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect);
 
 	/* disable master interrupt */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MASTINT_EN), 0, 0, indirect);
 
 	/* setup regUVD_LMI_CTRL */
@@ -652,80 +679,80 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, bool indirect)
 		UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK |
 		(8 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) |
 		0x00100000L);
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_LMI_CTRL), tmp, 0, indirect);
 
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MPC_CNTL),
 		0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT, 0, indirect);
 
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MPC_SET_MUXA0),
 		((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
 		 (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
 		 (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
 		 (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)), 0, indirect);
 
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MPC_SET_MUXB0),
 		 ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
 		 (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
 		 (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
 		 (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)), 0, indirect);
 
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MPC_SET_MUX),
 		((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
 		 (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
 		 (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)), 0, indirect);
 
-	vcn_v4_0_3_mc_resume_dpg_mode(adev, indirect);
+	vcn_v4_0_3_mc_resume_dpg_mode(adev, inst_idx, indirect);
 
 	tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT);
 	tmp |= UVD_VCPU_CNTL__CLK_EN_MASK;
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect);
 
 	/* enable LMI MC and UMC channels */
 	tmp = 0x1f << UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM__SHIFT;
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_LMI_CTRL2), tmp, 0, indirect);
 
 	/* enable master interrupt */
-	WREG32_SOC15_DPG_MODE(0, SOC15_DPG_MODE_OFFSET(
+	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
 	if (indirect)
-		psp_update_vcn_sram(adev, 0, adev->vcn.inst->dpg_sram_gpu_addr,
-			(uint32_t)((uintptr_t)adev->vcn.inst->dpg_sram_curr_addr -
-				(uintptr_t)adev->vcn.inst->dpg_sram_cpu_addr));
+		psp_update_vcn_sram(adev, inst_idx, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr,
+			(uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr -
+				(uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr));
 
-	ring = &adev->vcn.inst->ring_dec;
+	ring = &adev->vcn.inst[inst_idx].ring_dec;
 
 	/* program the RB_BASE for ring buffer */
-	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_LO4,
 		lower_32_bits(ring->gpu_addr));
-	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_HI4,
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_HI4,
 		upper_32_bits(ring->gpu_addr));
 
-	WREG32_SOC15(VCN, 0, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
 
 	/* resetting ring, fw should not check RB ring */
-	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
 	tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
-	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
 
 	/* Initialize the ring buffer's read and write pointers */
-	WREG32_SOC15(VCN, 0, regUVD_RB_RPTR4, 0);
-	WREG32_SOC15(VCN, 0, regUVD_RB_WPTR4, 0);
-	ring->wptr = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_RPTR4, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4, 0);
+	ring->wptr = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4);
 
-	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
+	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
 	tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
-	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
 
-	WREG32_SOC15(VCN, 0, regUVD_SCRATCH2, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_SCRATCH2, 0);
 
 	/*resetting done, fw can check RB ring */
 	fw_shared->sq.queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
@@ -745,159 +772,162 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
-	int j, k, r;
+	int i, j, k, r;
 
 	if (adev->pm.dpm_enabled)
 		amdgpu_dpm_enable_uvd(adev, true);
 
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
-		return vcn_v4_0_3_start_dpg_mode(adev, adev->vcn.indirect_sram);
-
-	/* set VCN status busy */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_STATUS) | UVD_STATUS__UVD_BUSY;
-	WREG32_SOC15(VCN, 0, regUVD_STATUS, tmp);
-
-	/*SW clock gating */
-	vcn_v4_0_3_disable_clock_gating(adev);
-
-	/* enable VCPU clock */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
-		UVD_VCPU_CNTL__CLK_EN_MASK, ~UVD_VCPU_CNTL__CLK_EN_MASK);
-
-	/* disable master interrupt */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_MASTINT_EN), 0,
-		~UVD_MASTINT_EN__VCPU_EN_MASK);
-
-	/* enable LMI MC and UMC channels */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_LMI_CTRL2), 0,
-		~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK);
-
-	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
-	tmp &= ~UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
-	tmp &= ~UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
-
-	/* setup regUVD_LMI_CTRL */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_LMI_CTRL);
-	WREG32_SOC15(VCN, 0, regUVD_LMI_CTRL, tmp |
-		UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
-		UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
-		UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
-		UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK);
-
-	/* setup regUVD_MPC_CNTL */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_MPC_CNTL);
-	tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK;
-	tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT;
-	WREG32_SOC15(VCN, 0, regUVD_MPC_CNTL, tmp);
-
-	/* setup UVD_MPC_SET_MUXA0 */
-	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUXA0,
-		((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
-		(0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
-		(0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
-		(0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)));
-
-	/* setup UVD_MPC_SET_MUXB0 */
-	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUXB0,
-		((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
-		(0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
-		(0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
-		(0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)));
-
-	/* setup UVD_MPC_SET_MUX */
-	WREG32_SOC15(VCN, 0, regUVD_MPC_SET_MUX,
-		((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
-		(0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
-		(0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)));
-
-	vcn_v4_0_3_mc_resume(adev);
-
-	/* VCN global tiling registers */
-	WREG32_SOC15(VCN, 0, regUVD_GFX8_ADDR_CONFIG,
-		adev->gfx.config.gb_addr_config);
-	WREG32_SOC15(VCN, 0, regUVD_GFX10_ADDR_CONFIG,
-		adev->gfx.config.gb_addr_config);
-
-	/* unblock VCPU register access */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_RB_ARB_CTRL), 0,
-		~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
+			r = vcn_v4_0_3_start_dpg_mode(adev, i, adev->vcn.indirect_sram);
+			continue;
+		}
 
-	/* release VCPU reset to boot */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
-		~UVD_VCPU_CNTL__BLK_RST_MASK);
+		/* set VCN status busy */
+		tmp = RREG32_SOC15(VCN, i, regUVD_STATUS) | UVD_STATUS__UVD_BUSY;
+		WREG32_SOC15(VCN, i, regUVD_STATUS, tmp);
+
+		/*SW clock gating */
+		vcn_v4_0_3_disable_clock_gating(adev, i);
+
+		/* enable VCPU clock */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
+			UVD_VCPU_CNTL__CLK_EN_MASK, ~UVD_VCPU_CNTL__CLK_EN_MASK);
+
+		/* disable master interrupt */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_MASTINT_EN), 0,
+			~UVD_MASTINT_EN__VCPU_EN_MASK);
+
+		/* enable LMI MC and UMC channels */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_LMI_CTRL2), 0,
+			~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK);
+
+		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp &= ~UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
+		tmp &= ~UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
+		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
+
+		/* setup regUVD_LMI_CTRL */
+		tmp = RREG32_SOC15(VCN, i, regUVD_LMI_CTRL);
+		WREG32_SOC15(VCN, i, regUVD_LMI_CTRL, tmp |
+			UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
+			UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
+			UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
+			UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK);
+
+		/* setup regUVD_MPC_CNTL */
+		tmp = RREG32_SOC15(VCN, i, regUVD_MPC_CNTL);
+		tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK;
+		tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT;
+		WREG32_SOC15(VCN, i, regUVD_MPC_CNTL, tmp);
+
+		/* setup UVD_MPC_SET_MUXA0 */
+		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUXA0,
+			((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
+			(0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
+			(0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
+			(0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)));
+
+		/* setup UVD_MPC_SET_MUXB0 */
+		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUXB0,
+			((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
+			(0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
+			(0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
+			(0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)));
+
+		/* setup UVD_MPC_SET_MUX */
+		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUX,
+			((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
+			(0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
+			(0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)));
+
+		vcn_v4_0_3_mc_resume(adev, i);
+
+		/* VCN global tiling registers */
+		WREG32_SOC15(VCN, i, regUVD_GFX8_ADDR_CONFIG,
+			adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(VCN, i, regUVD_GFX10_ADDR_CONFIG,
+			adev->gfx.config.gb_addr_config);
+
+		/* unblock VCPU register access */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_RB_ARB_CTRL), 0,
+			~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+
+		/* release VCPU reset to boot */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
+			~UVD_VCPU_CNTL__BLK_RST_MASK);
 
-	for (j = 0; j < 10; ++j) {
-		uint32_t status;
+		for (j = 0; j < 10; ++j) {
+			uint32_t status;
 
-		for (k = 0; k < 100; ++k) {
-			status = RREG32_SOC15(VCN, 0, regUVD_STATUS);
+			for (k = 0; k < 100; ++k) {
+				status = RREG32_SOC15(VCN, i, regUVD_STATUS);
+				if (status & 2)
+					break;
+				mdelay(10);
+			}
+			r = 0;
 			if (status & 2)
 				break;
-			if (amdgpu_emu_mode == 1)
-				msleep(500);
-			else
-				mdelay(10);
-		}
-		r = 0;
-		if (status & 2)
-			break;
-
-		DRM_DEV_ERROR(adev->dev,
-			"VCN decode not responding, trying to reset the VCPU!!!\n");
-		WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
-			UVD_VCPU_CNTL__BLK_RST_MASK,
-			~UVD_VCPU_CNTL__BLK_RST_MASK);
-		mdelay(10);
-		WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
-			~UVD_VCPU_CNTL__BLK_RST_MASK);
-
-		mdelay(10);
-		r = -1;
-	}
 
-	if (r) {
-		DRM_DEV_ERROR(adev->dev, "VCN decode not responding, giving up!!!\n");
-		return r;
-	}
+			DRM_DEV_ERROR(adev->dev,
+				"VCN decode not responding, trying to reset the VCPU!!!\n");
+			WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
+				UVD_VCPU_CNTL__BLK_RST_MASK,
+				~UVD_VCPU_CNTL__BLK_RST_MASK);
+			mdelay(10);
+			WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
+				~UVD_VCPU_CNTL__BLK_RST_MASK);
+
+			mdelay(10);
+			r = -1;
+		}
 
-	/* enable master interrupt */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_MASTINT_EN),
-		UVD_MASTINT_EN__VCPU_EN_MASK,
-		~UVD_MASTINT_EN__VCPU_EN_MASK);
+		if (r) {
+			DRM_DEV_ERROR(adev->dev, "VCN decode not responding, giving up!!!\n");
+			return r;
+		}
 
-	/* clear the busy bit of VCN_STATUS */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_STATUS), 0,
-		~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
+		/* enable master interrupt */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_MASTINT_EN),
+			UVD_MASTINT_EN__VCPU_EN_MASK,
+			~UVD_MASTINT_EN__VCPU_EN_MASK);
 
-	ring = &adev->vcn.inst->ring_dec;
+		/* clear the busy bit of VCN_STATUS */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_STATUS), 0,
+			~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
 
-	fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+		ring = &adev->vcn.inst[i].ring_dec;
+		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 
-	/* program the RB_BASE for ring buffer */
-	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_LO4,
-		lower_32_bits(ring->gpu_addr));
-	WREG32_SOC15(VCN, 0, regUVD_RB_BASE_HI4,
-		upper_32_bits(ring->gpu_addr));
+		/* program the RB_BASE for ring buffer */
+		WREG32_SOC15(VCN, i, regUVD_RB_BASE_LO4,
+			lower_32_bits(ring->gpu_addr));
+		WREG32_SOC15(VCN, i, regUVD_RB_BASE_HI4,
+			upper_32_bits(ring->gpu_addr));
 
-	WREG32_SOC15(VCN, 0, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+		WREG32_SOC15(VCN, i, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
 
-	/* resetting ring, fw should not check RB ring */
-	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
-	tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
-	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+		/* resetting ring, fw should not check RB ring */
+		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
+		tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
+		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
 
-	/* Initialize the ring buffer's read and write pointers */
-	WREG32_SOC15(VCN, 0, regUVD_RB_RPTR4, 0);
-	WREG32_SOC15(VCN, 0, regUVD_RB_WPTR4, 0);
+		/* Initialize the ring buffer's read and write pointers */
+		WREG32_SOC15(VCN, i, regUVD_RB_RPTR4, 0);
+		WREG32_SOC15(VCN, i, regUVD_RB_WPTR4, 0);
 
-	tmp = RREG32_SOC15(VCN, 0, regVCN_RB_ENABLE);
-	tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
-	WREG32_SOC15(VCN, 0, regVCN_RB_ENABLE, tmp);
+		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
+		tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
+		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
 
-	ring->wptr = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
-	fw_shared->sq.queue_mode &= cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF));
+		ring->wptr = RREG32_SOC15(VCN, i, regUVD_RB_WPTR4);
+		fw_shared->sq.queue_mode &=
+			cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF));
 
+	}
 	return 0;
 }
 
@@ -905,32 +935,27 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
  * vcn_v4_0_3_stop_dpg_mode - VCN stop with dpg mode
  *
  * @adev: amdgpu_device pointer
+ * @inst_idx: instance number index
  *
  * Stop VCN block with dpg mode
  */
-static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev)
+static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t tmp;
 
 	/* Wait for power status to be 1 */
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_POWER_STATUS, 1,
+	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_POWER_STATUS, 1,
 		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 	/* wait for read ptr to be equal to write ptr */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR);
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR, tmp, 0xFFFFFFFF);
-
-	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR2);
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR2, tmp, 0xFFFFFFFF);
-
-	tmp = RREG32_SOC15(VCN, 0, regUVD_RB_WPTR4);
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_RB_RPTR4, tmp, 0xFFFFFFFF);
+	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4);
+	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_RB_RPTR4, tmp, 0xFFFFFFFF);
 
-	SOC15_WAIT_ON_RREG(VCN, 0, regUVD_POWER_STATUS, 1,
+	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_POWER_STATUS, 1,
 		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 	/* disable dynamic power gating mode */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_POWER_STATUS), 0,
+	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 0,
 		~UVD_POWER_STATUS__UVD_PG_MODE_MASK);
 	return 0;
 }
@@ -945,65 +970,68 @@ static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev)
 static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 {
 	uint32_t tmp;
-	int r = 0;
+	int i, r = 0;
 
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
-		r = vcn_v4_0_3_stop_dpg_mode(adev);
-		goto Done;
-	}
-
-	/* wait for vcn idle */
-	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_STATUS, UVD_STATUS__IDLE, 0x7);
-	if (r)
-		return r;
-
-	tmp = UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK |
-		UVD_LMI_STATUS__READ_CLEAN_MASK |
-		UVD_LMI_STATUS__WRITE_CLEAN_MASK |
-		UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK;
-	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_LMI_STATUS, tmp, tmp);
-	if (r)
-		return r;
-
-	/* stall UMC channel */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_LMI_CTRL2);
-	tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_LMI_CTRL2, tmp);
-	tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK|
-		UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK;
-	r = SOC15_WAIT_ON_RREG(VCN, 0, regUVD_LMI_STATUS, tmp, tmp);
-	if (r)
-		return r;
-
-	/* Unblock VCPU Register access */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_RB_ARB_CTRL),
-		UVD_RB_ARB_CTRL__VCPU_DIS_MASK,
-		~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
-
-	/* release VCPU reset to boot */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL),
-		UVD_VCPU_CNTL__BLK_RST_MASK,
-		~UVD_VCPU_CNTL__BLK_RST_MASK);
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
+			r = vcn_v4_0_3_stop_dpg_mode(adev, i);
+			goto Done;
+		}
 
-	/* disable VCPU clock */
-	WREG32_P(SOC15_REG_OFFSET(VCN, 0, regUVD_VCPU_CNTL), 0,
-		~(UVD_VCPU_CNTL__CLK_EN_MASK));
+		/* wait for vcn idle */
+		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_STATUS, UVD_STATUS__IDLE, 0x7);
+		if (r)
+			goto Done;
+
+		tmp = UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK |
+			UVD_LMI_STATUS__READ_CLEAN_MASK |
+			UVD_LMI_STATUS__WRITE_CLEAN_MASK |
+			UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK;
+		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_LMI_STATUS, tmp, tmp);
+		if (r)
+			goto Done;
+
+		/* stall UMC channel */
+		tmp = RREG32_SOC15(VCN, i, regUVD_LMI_CTRL2);
+		tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK;
+		WREG32_SOC15(VCN, i, regUVD_LMI_CTRL2, tmp);
+		tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK|
+			UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK;
+		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_LMI_STATUS, tmp, tmp);
+		if (r)
+			goto Done;
+
+		/* Unblock VCPU Register access */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_RB_ARB_CTRL),
+			UVD_RB_ARB_CTRL__VCPU_DIS_MASK,
+			~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+
+		/* release VCPU reset to boot */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
+			UVD_VCPU_CNTL__BLK_RST_MASK,
+			~UVD_VCPU_CNTL__BLK_RST_MASK);
 
-	/* reset LMI UMC/LMI/VCPU */
-	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
-	tmp |= UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+		/* disable VCPU clock */
+		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
+			~(UVD_VCPU_CNTL__CLK_EN_MASK));
 
-	tmp = RREG32_SOC15(VCN, 0, regUVD_SOFT_RESET);
-	tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
-	WREG32_SOC15(VCN, 0, regUVD_SOFT_RESET, tmp);
+		/* reset LMI UMC/LMI/VCPU */
+		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp |= UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
+		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
 
-	/* clear VCN status */
-	WREG32_SOC15(VCN, 0, regUVD_STATUS, 0);
+		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
+		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
 
-	/* apply HW clock gating */
-	vcn_v4_0_3_enable_clock_gating(adev);
+		/* clear VCN status */
+		WREG32_SOC15(VCN, i, regUVD_STATUS, 0);
 
+		/* apply HW clock gating */
+		vcn_v4_0_3_enable_clock_gating(adev, i);
+	}
 Done:
 	if (adev->pm.dpm_enabled)
 		amdgpu_dpm_enable_uvd(adev, false);
@@ -1113,8 +1141,15 @@ static const struct amdgpu_ring_funcs vcn_v4_0_3_dec_sw_ring_vm_funcs = {
  */
 static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->vcn.inst->ring_dec.funcs = &vcn_v4_0_3_dec_sw_ring_vm_funcs;
-	adev->vcn.inst->ring_dec.me = 0;
+	int i;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		adev->vcn.inst[i].ring_dec.funcs = &vcn_v4_0_3_dec_sw_ring_vm_funcs;
+		adev->vcn.inst[i].ring_dec.me = i;
+		adev->vcn.inst[i].aid_id = i / adev->vcn.num_inst_per_aid;
+	}
 	DRM_DEV_INFO(adev->dev, "VCN decode(Software Ring) is enabled in VM mode\n");
 }
 
@@ -1128,8 +1163,15 @@ static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 static bool vcn_v4_0_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i, ret = 1;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		ret &= (RREG32_SOC15(VCN, i, regUVD_STATUS) == UVD_STATUS__IDLE);
+	}
 
-	return (RREG32_SOC15(VCN, 0, regUVD_STATUS) == UVD_STATUS__IDLE);
+	return ret;
 }
 
 /**
@@ -1142,13 +1184,21 @@ static bool vcn_v4_0_3_is_idle(void *handle)
 static int vcn_v4_0_3_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i, ret = 0;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		ret = SOC15_WAIT_ON_RREG(VCN, i, regUVD_STATUS, UVD_STATUS__IDLE,
+				UVD_STATUS__IDLE);
+		if (ret)
+			return ret;
+	}
 
-	return SOC15_WAIT_ON_RREG(VCN, 0, regUVD_STATUS, UVD_STATUS__IDLE,
-			UVD_STATUS__IDLE);
+	return ret;
 }
 
-/**
- * vcn_v4_0_3_set_clockgating_state - set VCN block clockgating state
+/* vcn_v4_0_3_set_clockgating_state - set VCN block clockgating state
  *
  * @handle: amdgpu_device pointer
  * @state: clock gating state
@@ -1160,15 +1210,19 @@ static int vcn_v4_0_3_set_clockgating_state(void *handle,
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	int i;
 
-	if (enable) {
-		if (RREG32_SOC15(VCN, 0, regUVD_STATUS) != UVD_STATUS__IDLE)
-			return -EBUSY;
-		vcn_v4_0_3_enable_clock_gating(adev);
-	} else {
-		vcn_v4_0_3_disable_clock_gating(adev);
+	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+		if (enable) {
+			if (RREG32_SOC15(VCN, i, regUVD_STATUS) != UVD_STATUS__IDLE)
+				return -EBUSY;
+			vcn_v4_0_3_enable_clock_gating(adev, i);
+		} else {
+			vcn_v4_0_3_disable_clock_gating(adev, i);
+		}
 	}
-
 	return 0;
 }
 
@@ -1231,11 +1285,15 @@ static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
+	uint32_t i;
+
+	i = node_id_to_phys_map[entry->node_id];
+
 	DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n");
 
 	switch (entry->src_id) {
 	case VCN_4_0__SRCID__UVD_TRAP:
-		amdgpu_fence_process(&adev->vcn.inst->ring_dec);
+		amdgpu_fence_process(&adev->vcn.inst[i].ring_dec);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -1266,9 +1324,9 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
 
-		adev->vcn.inst->irq.num_types = 1;
-		adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs;
+		adev->vcn.inst->irq.num_types++;
 	}
+	adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs;
 }
 
 static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = {
-- 
GitLab


From 0b02364e03caecbe30bdd9db0b3e6ba0196bb0ef Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 29 Jul 2022 14:56:59 +0530
Subject: [PATCH 1100/1791] drm/amdgpu: Conform to SET_UCONFIG_REG spec

The packet expects only 16 bits register offset. Hence pass register
offset which is local to each XCC.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 43126f7b70eac..14b96b43f02c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -232,13 +232,14 @@ static void gfx_v9_4_3_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel,
 
 static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring)
 {
+	uint32_t scratch_reg0_offset, xcc_offset;
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t tmp = 0;
 	unsigned i;
 	int r;
-	/* scratch_reg0_offset is 32bit even with full XCD config */
-	uint32_t scratch_reg0_offset;
 
+	/* Use register offset which is local to XCC in the packet */
+	xcc_offset = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0);
 	scratch_reg0_offset = SOC15_REG_OFFSET(GC, GET_INST(GC, ring->xcc_id), regSCRATCH_REG0);
 	WREG32(scratch_reg0_offset, 0xCAFEDEAD);
 
@@ -247,7 +248,7 @@ static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring)
 		return r;
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
-	amdgpu_ring_write(ring, scratch_reg0_offset - PACKET3_SET_UCONFIG_REG_START);
+	amdgpu_ring_write(ring, xcc_offset - PACKET3_SET_UCONFIG_REG_START);
 	amdgpu_ring_write(ring, 0xDEADBEEF);
 	amdgpu_ring_commit(ring);
 
-- 
GitLab


From 5fb34bd9cf9e248d7e84e431a4a6b731334ab564 Mon Sep 17 00:00:00 2001
From: Alex Sierra <alex.sierra@amd.com>
Date: Tue, 24 May 2022 10:22:12 -0500
Subject: [PATCH 1101/1791] drm/amdkfd: pass kfd_node ref to svm migration api

This work is required for GC 9.4.3, previous to support memory
partitions per node at SVM. When multiple partition is configured,
every BO should be allocated inside one specific partition which
corresponds to the current amdgpu_device and kfd_node.

v2: squash in compilation fix (Alex)
v3: squash in fix for pre-gfx 9.4.3 (Alex)
v4: squash in best_loc fix (Alex)

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c   |   7 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h   |   3 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c   |   2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c    |  33 ++---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  45 +++----
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |  30 ++++-
 drivers/gpu/drm/amd/amdkfd/kfd_process.c |   6 +-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c     | 152 +++++++++++------------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h     |  21 ++--
 9 files changed, 166 insertions(+), 133 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index c3964c14f215b..c390b2856cc93 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2441,7 +2441,8 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
  * shouldn't be reported any more.
  */
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    uint64_t addr, bool write_fault)
+			    u32 client_id, u32 node_id, uint64_t addr,
+			    bool write_fault)
 {
 	bool is_compute_context = false;
 	struct amdgpu_bo *root;
@@ -2465,8 +2466,8 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 
 	addr /= AMDGPU_GPU_PAGE_SIZE;
 
-	if (is_compute_context &&
-	    !svm_range_restore_pages(adev, pasid, addr, write_fault)) {
+	if (is_compute_context && !svm_range_restore_pages(adev, pasid, client_id,
+	    node_id, addr, write_fault)) {
 		amdgpu_bo_unref(&root);
 		return true;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 9f5d32b0fda1f..dbab31647186b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -455,7 +455,8 @@ void amdgpu_vm_check_compute_bug(struct amdgpu_device *adev);
 void amdgpu_vm_get_task_info(struct amdgpu_device *adev, u32 pasid,
 			     struct amdgpu_task_info *task_info);
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    uint64_t addr, bool write_fault);
+			    u32 client_id, u32 node_id, uint64_t addr,
+			    bool write_fault);
 
 void amdgpu_vm_set_task_info(struct amdgpu_vm *vm);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index d76f5c8d4977e..01bd456513821 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -139,7 +139,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,
 		/* Try to handle the recoverable page faults by filling page
 		 * tables
 		 */
-		if (amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault))
+		if (amdgpu_vm_handle_fault(adev, entry->pasid, 0, 0, addr, write_fault))
 			return 1;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 2c322a25bf1cd..c5752a349f3d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -557,11 +557,24 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	u64 addr;
 	uint32_t cam_index = 0;
 	int ret;
-	uint32_t node_id;
+	uint32_t node_id = 0;
 
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
 
+	if (entry->client_id == SOC15_IH_CLIENTID_VMC) {
+		hub_name = "mmhub0";
+		hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
+	} else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) {
+		hub_name = "mmhub1";
+		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
+	} else {
+		hub_name = "gfxhub0";
+		node_id = (adev->ip_versions[GC_HWIP][0] ==
+			   IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
+		hub = &adev->vmhub[node_id/2];
+	}
+
 	if (retry_fault) {
 		if (adev->irq.retry_cam_enabled) {
 			/* Delegate it to a different ring if the hardware hasn't
@@ -574,7 +587,8 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 
 			cam_index = entry->src_data[2] & 0x3ff;
 
-			ret = amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault);
+			ret = amdgpu_vm_handle_fault(adev, entry->pasid, entry->client_id, node_id,
+						     addr, write_fault);
 			WDOORBELL32(adev->irq.retry_cam_doorbell_index, cam_index);
 			if (ret)
 				return 1;
@@ -596,7 +610,8 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 			/* Try to handle the recoverable page faults by filling page
 			 * tables
 			 */
-			if (amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault))
+			if (amdgpu_vm_handle_fault(adev, entry->pasid, entry->client_id, node_id,
+						   addr, write_fault))
 				return 1;
 		}
 	}
@@ -604,18 +619,6 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	if (!printk_ratelimit())
 		return 0;
 
-	if (entry->client_id == SOC15_IH_CLIENTID_VMC) {
-		hub_name = "mmhub0";
-		hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
-	} else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) {
-		hub_name = "mmhub1";
-		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
-	} else {
-		hub_name = "gfxhub0";
-		node_id = (adev->ip_versions[GC_HWIP][0] ==
-			   IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
-		hub = &adev->vmhub[node_id/2];
-	}
 
 	memset(&task_info, 0, sizeof(struct amdgpu_task_info));
 	amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 5f4dc2a45bd03..e7e5abc32c84c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -287,11 +287,12 @@ static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate)
 }
 
 static int
-svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
+svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 			 struct migrate_vma *migrate, struct dma_fence **mfence,
 			 dma_addr_t *scratch, uint64_t ttm_res_offset)
 {
-	uint64_t npages = migrate->npages;
+	uint64_t npages = migrate->cpages;
+	struct amdgpu_device *adev = node->adev;
 	struct device *dev = adev->dev;
 	struct amdgpu_res_cursor cursor;
 	dma_addr_t *src;
@@ -321,7 +322,7 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 					      DMA_TO_DEVICE);
 			r = dma_mapping_error(dev, src[i]);
 			if (r) {
-				dev_err(adev->dev, "%s: fail %d dma_map_page\n",
+				dev_err(dev, "%s: fail %d dma_map_page\n",
 					__func__, r);
 				goto out_free_vram_pages;
 			}
@@ -390,12 +391,13 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 }
 
 static long
-svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
+svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 			struct vm_area_struct *vma, uint64_t start,
 			uint64_t end, uint32_t trigger, uint64_t ttm_res_offset)
 {
 	struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms);
 	uint64_t npages = (end - start) >> PAGE_SHIFT;
+	struct amdgpu_device *adev = node->adev;
 	struct kfd_process_device *pdd;
 	struct dma_fence *mfence = NULL;
 	struct migrate_vma migrate = { 0 };
@@ -445,7 +447,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 	else
 		pr_debug("0x%lx pages migrated\n", cpages);
 
-	r = svm_migrate_copy_to_vram(adev, prange, &migrate, &mfence, scratch, ttm_res_offset);
+	r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset);
 	migrate_vma_pages(&migrate);
 
 	pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
@@ -465,7 +467,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange,
 	kvfree(buf);
 out:
 	if (!r && cpages) {
-		pdd = svm_range_get_pdd_by_adev(prange, adev);
+		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_in, pdd->page_in + cpages);
 
@@ -492,8 +494,8 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 {
 	unsigned long addr, start, end;
 	struct vm_area_struct *vma;
-	struct amdgpu_device *adev;
 	uint64_t ttm_res_offset;
+	struct kfd_node *node;
 	unsigned long cpages = 0;
 	long r = 0;
 
@@ -503,9 +505,9 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 		return 0;
 	}
 
-	adev = svm_range_get_adev_by_id(prange, best_loc);
-	if (!adev) {
-		pr_debug("failed to get device by id 0x%x\n", best_loc);
+	node = svm_range_get_node_by_id(prange, best_loc);
+	if (!node) {
+		pr_debug("failed to get kfd node by id 0x%x\n", best_loc);
 		return -ENODEV;
 	}
 
@@ -515,9 +517,9 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 	start = prange->start << PAGE_SHIFT;
 	end = (prange->last + 1) << PAGE_SHIFT;
 
-	r = svm_range_vram_node_new(adev, prange, true);
+	r = svm_range_vram_node_new(node, prange, true);
 	if (r) {
-		dev_dbg(adev->dev, "fail %ld to alloc vram\n", r);
+		dev_dbg(node->adev->dev, "fail %ld to alloc vram\n", r);
 		return r;
 	}
 	ttm_res_offset = prange->offset << PAGE_SHIFT;
@@ -530,7 +532,7 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 			break;
 
 		next = min(vma->vm_end, end);
-		r = svm_migrate_vma_to_vram(adev, prange, vma, addr, next, trigger, ttm_res_offset);
+		r = svm_migrate_vma_to_vram(node, prange, vma, addr, next, trigger, ttm_res_offset);
 		if (r < 0) {
 			pr_debug("failed %ld to migrate\n", r);
 			break;
@@ -663,7 +665,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
  *   positive values - partial migration, number of pages not migrated
  */
 static long
-svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
+svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 		       struct vm_area_struct *vma, uint64_t start, uint64_t end,
 		       uint32_t trigger, struct page *fault_page)
 {
@@ -671,6 +673,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 	uint64_t npages = (end - start) >> PAGE_SHIFT;
 	unsigned long upages = npages;
 	unsigned long cpages = 0;
+	struct amdgpu_device *adev = node->adev;
 	struct kfd_process_device *pdd;
 	struct dma_fence *mfence = NULL;
 	struct migrate_vma migrate = { 0 };
@@ -745,7 +748,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 	kvfree(buf);
 out:
 	if (!r && cpages) {
-		pdd = svm_range_get_pdd_by_adev(prange, adev);
+		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_out, pdd->page_out + cpages);
 	}
@@ -766,7 +769,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 			    uint32_t trigger, struct page *fault_page)
 {
-	struct amdgpu_device *adev;
+	struct kfd_node *node;
 	struct vm_area_struct *vma;
 	unsigned long addr;
 	unsigned long start;
@@ -780,13 +783,11 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 		return 0;
 	}
 
-	adev = svm_range_get_adev_by_id(prange, prange->actual_loc);
-	if (!adev) {
-		pr_debug("failed to get device by id 0x%x\n",
-			 prange->actual_loc);
+	node = svm_range_get_node_by_id(prange, prange->actual_loc);
+	if (!node) {
+		pr_debug("failed to get kfd node by id 0x%x\n", prange->actual_loc);
 		return -ENODEV;
 	}
-
 	pr_debug("svms 0x%p prange 0x%p [0x%lx 0x%lx] from gpu 0x%x to ram\n",
 		 prange->svms, prange, prange->start, prange->last,
 		 prange->actual_loc);
@@ -805,7 +806,7 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 		}
 
 		next = min(vma->vm_end, end);
-		r = svm_migrate_vma_to_ram(adev, prange, vma, addr, next, trigger,
+		r = svm_migrate_vma_to_ram(node, prange, vma, addr, next, trigger,
 			fault_page);
 		if (r < 0) {
 			pr_debug("failed %ld to migrate prange %p\n", r, prange);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 400b4dcbdf053..df372de6b056c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -289,6 +289,7 @@ struct kfd_node {
 	 * from the HW ring into a SW ring.
 	 */
 	bool interrupts_active;
+	uint32_t interrupt_bitmap; /* Only used for GFX 9.4.3 */
 
 	/* QCM Device instance */
 	struct device_queue_manager *dqm;
@@ -971,9 +972,8 @@ struct kfd_process *kfd_lookup_process_by_pasid(u32 pasid);
 struct kfd_process *kfd_lookup_process_by_mm(const struct mm_struct *mm);
 
 int kfd_process_gpuidx_from_gpuid(struct kfd_process *p, uint32_t gpu_id);
-int kfd_process_gpuid_from_adev(struct kfd_process *p,
-			       struct amdgpu_device *adev, uint32_t *gpuid,
-			       uint32_t *gpuidx);
+int kfd_process_gpuid_from_node(struct kfd_process *p, struct kfd_node *node,
+				uint32_t *gpuid, uint32_t *gpuidx);
 static inline int kfd_process_gpuid_from_gpuidx(struct kfd_process *p,
 				uint32_t gpuidx, uint32_t *gpuid) {
 	return gpuidx < p->n_pdds ? p->pdds[gpuidx]->dev->id : -EINVAL;
@@ -1073,6 +1073,30 @@ struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev);
 struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev);
+static inline bool kfd_irq_is_from_node(struct kfd_node *node, uint32_t client_id,
+				     uint32_t node_id)
+{
+	if ((node->interrupt_bitmap & (0x1U << node_id)) ||
+	    ((node_id % 4) == 0 &&
+	    (node->interrupt_bitmap >> 16) & (0x1U << client_id)))
+		return true;
+
+	return false;
+}
+static inline struct kfd_node *kfd_node_by_irq_ids(struct amdgpu_device *adev,
+					uint32_t client_id, uint32_t node_id) {
+	struct kfd_dev *dev = adev->kfd.dev;
+	uint32_t i;
+
+	if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3))
+		return dev->nodes[0];
+
+	for (i = 0; i < dev->num_nodes; i++)
+		if (kfd_irq_is_from_node(dev->nodes[i], client_id, node_id))
+			return dev->nodes[i];
+
+	return NULL;
+}
 int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev);
 int kfd_numa_node_to_apic_id(int numa_node_id);
 void kfd_double_confirm_iommu_support(struct kfd_dev *gpu);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index c3d43e6e52361..666815b227a80 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1891,13 +1891,13 @@ int kfd_process_gpuidx_from_gpuid(struct kfd_process *p, uint32_t gpu_id)
 }
 
 int
-kfd_process_gpuid_from_adev(struct kfd_process *p, struct amdgpu_device *adev,
-			   uint32_t *gpuid, uint32_t *gpuidx)
+kfd_process_gpuid_from_node(struct kfd_process *p, struct kfd_node *node,
+			    uint32_t *gpuid, uint32_t *gpuidx)
 {
 	int i;
 
 	for (i = 0; i < p->n_pdds; i++)
-		if (p->pdds[i] && p->pdds[i]->dev->adev == adev) {
+		if (p->pdds[i] && p->pdds[i]->dev == node) {
 			*gpuid = p->pdds[i]->user_gpu_id;
 			*gpuidx = i;
 			return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 4b4f3bf8b823c..639831fbb6caf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -170,8 +170,7 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
 
 		page = hmm_pfn_to_page(hmm_pfns[i]);
 		if (is_zone_device_page(page)) {
-			struct amdgpu_device *bo_adev =
-					amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev);
+			struct amdgpu_device *bo_adev = prange->svm_bo->node->adev;
 
 			addr[i] = (hmm_pfns[i] << PAGE_SHIFT) +
 				   bo_adev->vm_manager.vram_base_offset -
@@ -424,10 +423,8 @@ static void svm_range_bo_unref(struct svm_range_bo *svm_bo)
 }
 
 static bool
-svm_range_validate_svm_bo(struct amdgpu_device *adev, struct svm_range *prange)
+svm_range_validate_svm_bo(struct kfd_node *node, struct svm_range *prange)
 {
-	struct amdgpu_device *bo_adev;
-
 	mutex_lock(&prange->lock);
 	if (!prange->svm_bo) {
 		mutex_unlock(&prange->lock);
@@ -440,12 +437,11 @@ svm_range_validate_svm_bo(struct amdgpu_device *adev, struct svm_range *prange)
 	}
 	if (svm_bo_ref_unless_zero(prange->svm_bo)) {
 		/*
-		 * Migrate from GPU to GPU, remove range from source bo_adev
-		 * svm_bo range list, and return false to allocate svm_bo from
-		 * destination adev.
+		 * Migrate from GPU to GPU, remove range from source svm_bo->node
+		 * range list, and return false to allocate svm_bo from destination
+		 * node.
 		 */
-		bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev);
-		if (bo_adev != adev) {
+		if (prange->svm_bo->node != node) {
 			mutex_unlock(&prange->lock);
 
 			spin_lock(&prange->svm_bo->list_lock);
@@ -513,7 +509,7 @@ static struct svm_range_bo *svm_range_bo_new(void)
 }
 
 int
-svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange,
+svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange,
 			bool clear)
 {
 	struct amdgpu_bo_param bp;
@@ -528,7 +524,7 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange,
 	pr_debug("pasid: %x svms 0x%p [0x%lx 0x%lx]\n", p->pasid, prange->svms,
 		 prange->start, prange->last);
 
-	if (svm_range_validate_svm_bo(adev, prange))
+	if (svm_range_validate_svm_bo(node, prange))
 		return 0;
 
 	svm_bo = svm_range_bo_new();
@@ -542,6 +538,7 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange,
 		kfree(svm_bo);
 		return -ESRCH;
 	}
+	svm_bo->node = node;
 	svm_bo->eviction_fence =
 		amdgpu_amdkfd_fence_create(dma_fence_context_alloc(1),
 					   mm,
@@ -559,7 +556,10 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange,
 	bp.type = ttm_bo_type_device;
 	bp.resv = NULL;
 
-	r = amdgpu_bo_create_user(adev, &bp, &ubo);
+	/* TODO: Allocate memory from the right memory partition. We can sort
+	 * out the details later, once basic memory partitioning is working
+	 */
+	r = amdgpu_bo_create_user(node->adev, &bp, &ubo);
 	if (r) {
 		pr_debug("failed %d to create bo\n", r);
 		goto create_bo_failed;
@@ -617,45 +617,30 @@ void svm_range_vram_node_free(struct svm_range *prange)
 	prange->ttm_res = NULL;
 }
 
-struct amdgpu_device *
-svm_range_get_adev_by_id(struct svm_range *prange, uint32_t gpu_id)
+struct kfd_node *
+svm_range_get_node_by_id(struct svm_range *prange, uint32_t gpu_id)
 {
-	struct kfd_process_device *pdd;
 	struct kfd_process *p;
-	int32_t gpu_idx;
+	struct kfd_process_device *pdd;
 
 	p = container_of(prange->svms, struct kfd_process, svms);
-
-	gpu_idx = kfd_process_gpuidx_from_gpuid(p, gpu_id);
-	if (gpu_idx < 0) {
-		pr_debug("failed to get device by id 0x%x\n", gpu_id);
-		return NULL;
-	}
-	pdd = kfd_process_device_from_gpuidx(p, gpu_idx);
+	pdd = kfd_process_device_data_by_id(p, gpu_id);
 	if (!pdd) {
-		pr_debug("failed to get device by idx 0x%x\n", gpu_idx);
+		pr_debug("failed to get kfd process device by id 0x%x\n", gpu_id);
 		return NULL;
 	}
 
-	return pdd->dev->adev;
+	return pdd->dev;
 }
 
 struct kfd_process_device *
-svm_range_get_pdd_by_adev(struct svm_range *prange, struct amdgpu_device *adev)
+svm_range_get_pdd_by_node(struct svm_range *prange, struct kfd_node *node)
 {
 	struct kfd_process *p;
-	int32_t gpu_idx, gpuid;
-	int r;
 
 	p = container_of(prange->svms, struct kfd_process, svms);
 
-	r = kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpu_idx);
-	if (r) {
-		pr_debug("failed to get device id by adev %p\n", adev);
-		return NULL;
-	}
-
-	return kfd_process_device_from_gpuidx(p, gpu_idx);
+	return kfd_get_process_device_data(node, p);
 }
 
 static int svm_range_bo_validate(void *param, struct amdgpu_bo *bo)
@@ -1148,12 +1133,18 @@ svm_range_split_by_granularity(struct kfd_process *p, struct mm_struct *mm,
 	}
 	return 0;
 }
+static bool
+svm_nodes_in_same_hive(struct kfd_node *node_a, struct kfd_node *node_b)
+{
+	return (node_a->adev == node_b->adev ||
+		amdgpu_xgmi_same_hive(node_a->adev, node_b->adev));
+}
 
 static uint64_t
-svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
-			int domain)
+svm_range_get_pte_flags(struct kfd_node *node,
+			struct svm_range *prange, int domain)
 {
-	struct amdgpu_device *bo_adev;
+	struct kfd_node *bo_node;
 	uint32_t flags = prange->flags;
 	uint32_t mapping_flags = 0;
 	uint64_t pte_flags;
@@ -1162,18 +1153,18 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 	bool uncached = flags & KFD_IOCTL_SVM_FLAG_UNCACHED;
 
 	if (domain == SVM_RANGE_VRAM_DOMAIN)
-		bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev);
+		bo_node = prange->svm_bo->node;
 
-	switch (KFD_GC_VERSION(adev->kfd.dev)) {
+	switch (node->adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 1):
 		if (domain == SVM_RANGE_VRAM_DOMAIN) {
-			if (bo_adev == adev) {
+			if (bo_node == node) {
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
 			} else {
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
-				if (amdgpu_xgmi_same_hive(adev, bo_adev))
+				if (svm_nodes_in_same_hive(node, bo_node))
 					snoop = true;
 			}
 		} else {
@@ -1183,15 +1174,15 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 		break;
 	case IP_VERSION(9, 4, 2):
 		if (domain == SVM_RANGE_VRAM_DOMAIN) {
-			if (bo_adev == adev) {
+			if (bo_node == node) {
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
-				if (adev->gmc.xgmi.connected_to_cpu)
+				if (node->adev->gmc.xgmi.connected_to_cpu)
 					snoop = true;
 			} else {
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
-				if (amdgpu_xgmi_same_hive(adev, bo_adev))
+				if (svm_nodes_in_same_hive(node, bo_node))
 					snoop = true;
 			}
 		} else {
@@ -1207,7 +1198,7 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 		if (uncached)
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		/* local HBM region close to partition*/
-		else if (bo_adev == adev)
+		else if (bo_node == node)
 			mapping_flags |= AMDGPU_VM_MTYPE_RW;
 		/* local HBM region far from partition or remote XGMI GPU or
 		 * system memory
@@ -1231,7 +1222,7 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange,
 	pte_flags |= (domain == SVM_RANGE_VRAM_DOMAIN) ? 0 : AMDGPU_PTE_SYSTEM;
 	pte_flags |= snoop ? AMDGPU_PTE_SNOOPED : 0;
 
-	pte_flags |= amdgpu_gem_va_map_flags(adev, mapping_flags);
+	pte_flags |= amdgpu_gem_va_map_flags(node->adev, mapping_flags);
 	return pte_flags;
 }
 
@@ -1338,7 +1329,7 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 		pr_debug("Mapping range [0x%lx 0x%llx] on domain: %s\n",
 			 last_start, prange->start + i, last_domain ? "GPU" : "CPU");
 
-		pte_flags = svm_range_get_pte_flags(adev, prange, last_domain);
+		pte_flags = svm_range_get_pte_flags(pdd->dev, prange, last_domain);
 		if (readonly)
 			pte_flags &= ~AMDGPU_PTE_WRITEABLE;
 
@@ -1347,6 +1338,9 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 			 (last_domain == SVM_RANGE_VRAM_DOMAIN) ? 1 : 0,
 			 pte_flags);
 
+		/* TODO: we still need to determine the vm_manager.vram_base_offset based on
+		 * the memory partition.
+		 */
 		r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb, NULL,
 					   last_start, prange->start + i,
 					   pte_flags,
@@ -1384,16 +1378,14 @@ svm_range_map_to_gpus(struct svm_range *prange, unsigned long offset,
 		      unsigned long *bitmap, bool wait, bool flush_tlb)
 {
 	struct kfd_process_device *pdd;
-	struct amdgpu_device *bo_adev;
+	struct amdgpu_device *bo_adev = NULL;
 	struct kfd_process *p;
 	struct dma_fence *fence = NULL;
 	uint32_t gpuidx;
 	int r = 0;
 
 	if (prange->svm_bo && prange->ttm_res)
-		bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev);
-	else
-		bo_adev = NULL;
+		bo_adev = prange->svm_bo->node->adev;
 
 	p = container_of(prange->svms, struct kfd_process, svms);
 	for_each_set_bit(gpuidx, bitmap, MAX_GPU_INSTANCE) {
@@ -2526,17 +2518,17 @@ svm_range_from_addr(struct svm_range_list *svms, unsigned long addr,
  */
 static int32_t
 svm_range_best_restore_location(struct svm_range *prange,
-				struct amdgpu_device *adev,
+				struct kfd_node *node,
 				int32_t *gpuidx)
 {
-	struct amdgpu_device *bo_adev, *preferred_adev;
+	struct kfd_node *bo_node, *preferred_node;
 	struct kfd_process *p;
 	uint32_t gpuid;
 	int r;
 
 	p = container_of(prange->svms, struct kfd_process, svms);
 
-	r = kfd_process_gpuid_from_adev(p, adev, &gpuid, gpuidx);
+	r = kfd_process_gpuid_from_node(p, node, &gpuid, gpuidx);
 	if (r < 0) {
 		pr_debug("failed to get gpuid from kgd\n");
 		return -1;
@@ -2546,9 +2538,8 @@ svm_range_best_restore_location(struct svm_range *prange,
 	    prange->preferred_loc == KFD_IOCTL_SVM_LOCATION_SYSMEM) {
 		return prange->preferred_loc;
 	} else if (prange->preferred_loc != KFD_IOCTL_SVM_LOCATION_UNDEFINED) {
-		preferred_adev = svm_range_get_adev_by_id(prange,
-							prange->preferred_loc);
-		if (amdgpu_xgmi_same_hive(adev, preferred_adev))
+		preferred_node = svm_range_get_node_by_id(prange, prange->preferred_loc);
+		if (preferred_node && svm_nodes_in_same_hive(node, preferred_node))
 			return prange->preferred_loc;
 		/* fall through */
 	}
@@ -2560,8 +2551,8 @@ svm_range_best_restore_location(struct svm_range *prange,
 		if (!prange->actual_loc)
 			return 0;
 
-		bo_adev = svm_range_get_adev_by_id(prange, prange->actual_loc);
-		if (amdgpu_xgmi_same_hive(adev, bo_adev))
+		bo_node = svm_range_get_node_by_id(prange, prange->actual_loc);
+		if (bo_node && svm_nodes_in_same_hive(node, bo_node))
 			return prange->actual_loc;
 		else
 			return 0;
@@ -2678,7 +2669,7 @@ svm_range_check_vm_userptr(struct kfd_process *p, uint64_t start, uint64_t last,
 }
 
 static struct
-svm_range *svm_range_create_unregistered_range(struct amdgpu_device *adev,
+svm_range *svm_range_create_unregistered_range(struct kfd_node *node,
 						struct kfd_process *p,
 						struct mm_struct *mm,
 						int64_t addr)
@@ -2713,7 +2704,7 @@ svm_range *svm_range_create_unregistered_range(struct amdgpu_device *adev,
 		pr_debug("Failed to create prange in address [0x%llx]\n", addr);
 		return NULL;
 	}
-	if (kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpuidx)) {
+	if (kfd_process_gpuid_from_node(p, node, &gpuid, &gpuidx)) {
 		pr_debug("failed to get gpuid from kgd\n");
 		svm_range_free(prange, true);
 		return NULL;
@@ -2767,7 +2758,7 @@ static bool svm_range_skip_recover(struct svm_range *prange)
 }
 
 static void
-svm_range_count_fault(struct amdgpu_device *adev, struct kfd_process *p,
+svm_range_count_fault(struct kfd_node *node, struct kfd_process *p,
 		      int32_t gpuidx)
 {
 	struct kfd_process_device *pdd;
@@ -2780,7 +2771,7 @@ svm_range_count_fault(struct amdgpu_device *adev, struct kfd_process *p,
 		uint32_t gpuid;
 		int r;
 
-		r = kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpuidx);
+		r = kfd_process_gpuid_from_node(p, node, &gpuid, &gpuidx);
 		if (r < 0)
 			return;
 	}
@@ -2808,6 +2799,7 @@ svm_fault_allowed(struct vm_area_struct *vma, bool write_fault)
 
 int
 svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
+			uint32_t client_id, uint32_t node_id,
 			uint64_t addr, bool write_fault)
 {
 	struct mm_struct *mm = NULL;
@@ -2815,6 +2807,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 	struct svm_range *prange;
 	struct kfd_process *p;
 	ktime_t timestamp = ktime_get_boottime();
+	struct kfd_node *node;
 	int32_t best_loc;
 	int32_t gpuidx = MAX_GPU_INSTANCE;
 	bool write_locked = false;
@@ -2858,6 +2851,13 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		goto out;
 	}
 
+	node = kfd_node_by_irq_ids(adev, node_id, client_id);
+	if (!node) {
+		pr_debug("kfd node does not exist node_id: %d, client_id: %d\n", node_id,
+			 client_id);
+		r = -EFAULT;
+		goto out;
+	}
 	mmap_read_lock(mm);
 retry_write_locked:
 	mutex_lock(&svms->lock);
@@ -2876,7 +2876,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 			write_locked = true;
 			goto retry_write_locked;
 		}
-		prange = svm_range_create_unregistered_range(adev, p, mm, addr);
+		prange = svm_range_create_unregistered_range(node, p, mm, addr);
 		if (!prange) {
 			pr_debug("failed to create unregistered range svms 0x%p address [0x%llx]\n",
 				 svms, addr);
@@ -2891,7 +2891,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 	mutex_lock(&prange->migrate_mutex);
 
 	if (svm_range_skip_recover(prange)) {
-		amdgpu_gmc_filter_faults_remove(adev, addr, pasid);
+		amdgpu_gmc_filter_faults_remove(node->adev, addr, pasid);
 		r = 0;
 		goto out_unlock_range;
 	}
@@ -2922,7 +2922,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		goto out_unlock_range;
 	}
 
-	best_loc = svm_range_best_restore_location(prange, adev, &gpuidx);
+	best_loc = svm_range_best_restore_location(prange, node, &gpuidx);
 	if (best_loc == -1) {
 		pr_debug("svms %p failed get best restore loc [0x%lx 0x%lx]\n",
 			 svms, prange->start, prange->last);
@@ -2981,7 +2981,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 	mutex_unlock(&svms->lock);
 	mmap_read_unlock(mm);
 
-	svm_range_count_fault(adev, p, gpuidx);
+	svm_range_count_fault(node, p, gpuidx);
 
 	mmput(mm);
 out:
@@ -2989,7 +2989,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 
 	if (r == -EAGAIN) {
 		pr_debug("recover vm fault later\n");
-		amdgpu_gmc_filter_faults_remove(adev, addr, pasid);
+		amdgpu_gmc_filter_faults_remove(node->adev, addr, pasid);
 		r = 0;
 	}
 	return r;
@@ -3231,7 +3231,7 @@ svm_range_best_prefetch_location(struct svm_range *prange)
 	DECLARE_BITMAP(bitmap, MAX_GPU_INSTANCE);
 	uint32_t best_loc = prange->prefetch_loc;
 	struct kfd_process_device *pdd;
-	struct amdgpu_device *bo_adev;
+	struct kfd_node *bo_node;
 	struct kfd_process *p;
 	uint32_t gpuidx;
 
@@ -3240,9 +3240,9 @@ svm_range_best_prefetch_location(struct svm_range *prange)
 	if (!best_loc || best_loc == KFD_IOCTL_SVM_LOCATION_UNDEFINED)
 		goto out;
 
-	bo_adev = svm_range_get_adev_by_id(prange, best_loc);
-	if (!bo_adev) {
-		WARN_ONCE(1, "failed to get device by id 0x%x\n", best_loc);
+	bo_node = svm_range_get_node_by_id(prange, best_loc);
+	if (!bo_node) {
+		WARN_ONCE(1, "failed to get valid kfd node at id%x\n", best_loc);
 		best_loc = 0;
 		goto out;
 	}
@@ -3260,10 +3260,10 @@ svm_range_best_prefetch_location(struct svm_range *prange)
 			continue;
 		}
 
-		if (pdd->dev->adev == bo_adev)
+		if (pdd->dev->adev == bo_node->adev)
 			continue;
 
-		if (!amdgpu_xgmi_same_hive(pdd->dev->adev, bo_adev)) {
+		if (!svm_nodes_in_same_hive(pdd->dev, bo_node)) {
 			best_loc = 0;
 			break;
 		}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index 7a33b93f9df6f..a165c73b40b28 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -48,6 +48,7 @@ struct svm_range_bo {
 	struct work_struct		eviction_work;
 	uint32_t			evicting;
 	struct work_struct		release_work;
+	struct kfd_node			*node;
 };
 
 enum svm_work_list_ops {
@@ -163,16 +164,17 @@ int svm_ioctl(struct kfd_process *p, enum kfd_ioctl_svm_op op, uint64_t start,
 struct svm_range *svm_range_from_addr(struct svm_range_list *svms,
 				      unsigned long addr,
 				      struct svm_range **parent);
-struct amdgpu_device *svm_range_get_adev_by_id(struct svm_range *prange,
-					       uint32_t id);
-int svm_range_vram_node_new(struct amdgpu_device *adev,
-			    struct svm_range *prange, bool clear);
+struct kfd_node *svm_range_get_node_by_id(struct svm_range *prange,
+					  uint32_t gpu_id);
+int svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange,
+			    bool clear);
 void svm_range_vram_node_free(struct svm_range *prange);
 int svm_range_split_by_granularity(struct kfd_process *p, struct mm_struct *mm,
 			       unsigned long addr, struct svm_range *parent,
 			       struct svm_range *prange);
-int svm_range_restore_pages(struct amdgpu_device *adev,
-			    unsigned int pasid, uint64_t addr, bool write_fault);
+int svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
+			    uint32_t client_id, uint32_t node_id, uint64_t addr,
+			    bool write_fault);
 int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence);
 void svm_range_add_list_work(struct svm_range_list *svms,
 			     struct svm_range *prange, struct mm_struct *mm,
@@ -192,7 +194,7 @@ int kfd_criu_restore_svm(struct kfd_process *p,
 			 uint64_t max_priv_data_size);
 int kfd_criu_resume_svm(struct kfd_process *p);
 struct kfd_process_device *
-svm_range_get_pdd_by_adev(struct svm_range *prange, struct amdgpu_device *adev);
+svm_range_get_pdd_by_node(struct svm_range *prange, struct kfd_node *node);
 void svm_range_list_lock_and_flush_work(struct svm_range_list *svms, struct mm_struct *mm);
 
 /* SVM API and HMM page migration work together, device memory type
@@ -219,8 +221,9 @@ static inline void svm_range_list_fini(struct kfd_process *p)
 }
 
 static inline int svm_range_restore_pages(struct amdgpu_device *adev,
-					  unsigned int pasid, uint64_t addr,
-					  bool write_fault)
+					  unsigned int pasid,
+					  uint32_t client_id, uint32_t node_id,
+					  uint64_t addr, bool write_fault)
 {
 	return -EFAULT;
 }
-- 
GitLab


From cd321e6fd611db983fb7cdb52daf089fabe4f9c7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 8 Aug 2022 11:20:36 +0530
Subject: [PATCH 1102/1791] drm/amdgpu: Use status register for partition mode

Program partition status register to reflect the current partition mode.
Partition capability register is for capability and is a one-time setting.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 28 +++++++++++---------------
 1 file changed, 12 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index bcd3e08d01cd2..ef0b557e9b3eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -418,20 +418,13 @@ static void nbio_v7_9_enable_doorbell_interrupt(struct amdgpu_device *adev,
 
 static enum amdgpu_gfx_partition nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
 {
-	u32 tmp;
+	u32 tmp, px;
 
-	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP);
+	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS);
+	px = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS,
+			   PARTITION_MODE);
 
-	if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, SPX_SUPPORT))
-		return AMDGPU_SPX_PARTITION_MODE;
-	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, DPX_SUPPORT))
-		return AMDGPU_DPX_PARTITION_MODE;
-	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, TPX_SUPPORT))
-		return AMDGPU_TPX_PARTITION_MODE;
-	else if (REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_CAP, CPX_SUPPORT))
-		return AMDGPU_CPX_PARTITION_MODE;
-	else
-		return AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+	return ffs(px);
 }
 
 static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
@@ -439,11 +432,14 @@ static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
 {
 	u32 tmp;
 
-	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP);
-	tmp &= ~0x1f;
-	tmp |= 1 << mode;
+	/* Each bit represents DPX,TPX,QPX,CPX mode. No bit set means default
+	 * SPX mode.
+	 */
+	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS);
+	tmp = REG_SET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS,
+			    PARTITION_MODE, mode ? BIT(mode - 1) : mode);
 
-	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_CAP, tmp);
+	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
 }
 
 const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
-- 
GitLab


From d6e924ad85a0cebc9e39eb956a23386ce32cc9f9 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 9 Aug 2022 14:56:53 -0400
Subject: [PATCH 1103/1791] drm/amdkfd: Update SMI events for GFX9.4.3

On GFX 9.4.3, there can be multiple KFD nodes. As a result,
SMI events for SVM, queue evict/restore should be raised for
each node independently.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c    | 16 ++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_process.c    |  6 ++--
 drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c | 38 ++++++++++-----------
 drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h | 14 ++++----
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c        |  6 ++--
 5 files changed, 40 insertions(+), 40 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index e7e5abc32c84c..42e599912e52b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -423,9 +423,9 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	migrate.dst = migrate.src + npages;
 	scratch = (dma_addr_t *)(migrate.dst + npages);
 
-	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
+	kfd_smi_event_migration_start(node, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      0, adev->kfd.dev->nodes[0]->id, prange->prefetch_loc,
+				      0, node->id, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -456,9 +456,9 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
+	kfd_smi_event_migration_end(node, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    0, adev->kfd.dev->nodes[0]->id, trigger);
+				    0, node->id, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 	svm_range_free_dma_mappings(prange);
@@ -702,9 +702,9 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	migrate.fault_page = fault_page;
 	scratch = (dma_addr_t *)(migrate.dst + npages);
 
-	kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid,
+	kfd_smi_event_migration_start(node, p->lead_thread->pid,
 				      start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				      adev->kfd.dev->nodes[0]->id, 0, prange->prefetch_loc,
+				      node->id, 0, prange->prefetch_loc,
 				      prange->preferred_loc, trigger);
 
 	r = migrate_vma_setup(&migrate);
@@ -738,9 +738,9 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid,
+	kfd_smi_event_migration_end(node, p->lead_thread->pid,
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    adev->kfd.dev->nodes[0]->id, 0, trigger);
+				    node->id, 0, trigger);
 
 	svm_range_dma_unmap(adev->dev, scratch, 0, npages);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 666815b227a80..a6ff57f114725 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1817,7 +1817,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger)
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_process_device *pdd = p->pdds[i];
 
-		kfd_smi_event_queue_eviction(pdd->dev->kfd, p->lead_thread->pid,
+		kfd_smi_event_queue_eviction(pdd->dev, p->lead_thread->pid,
 					     trigger);
 
 		r = pdd->dev->dqm->ops.evict_process_queues(pdd->dev->dqm,
@@ -1845,7 +1845,7 @@ int kfd_process_evict_queues(struct kfd_process *p, uint32_t trigger)
 		if (n_evicted == 0)
 			break;
 
-		kfd_smi_event_queue_restore(pdd->dev->kfd, p->lead_thread->pid);
+		kfd_smi_event_queue_restore(pdd->dev, p->lead_thread->pid);
 
 		if (pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm,
 							      &pdd->qpd))
@@ -1866,7 +1866,7 @@ int kfd_process_restore_queues(struct kfd_process *p)
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_process_device *pdd = p->pdds[i];
 
-		kfd_smi_event_queue_restore(pdd->dev->kfd, p->lead_thread->pid);
+		kfd_smi_event_queue_restore(pdd->dev, p->lead_thread->pid);
 
 		r = pdd->dev->dqm->ops.restore_process_queues(pdd->dev->dqm,
 							      &pdd->qpd);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
index b703da59e0677..d9953c2b26614 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c
@@ -250,58 +250,58 @@ void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid)
 			  task_info.pid, task_info.task_name);
 }
 
-void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_page_fault_start(struct kfd_node *node, pid_t pid,
 				    unsigned long address, bool write_fault,
 				    ktime_t ts)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_PAGE_FAULT_START,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_START,
 			  "%lld -%d @%lx(%x) %c\n", ktime_to_ns(ts), pid,
-			  address, dev->nodes[0]->id, write_fault ? 'W' : 'R');
+			  address, node->id, write_fault ? 'W' : 'R');
 }
 
-void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_page_fault_end(struct kfd_node *node, pid_t pid,
 				  unsigned long address, bool migration)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_PAGE_FAULT_END,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_END,
 			  "%lld -%d @%lx(%x) %c\n", ktime_get_boottime_ns(),
-			  pid, address, dev->nodes[0]->id, migration ? 'M' : 'U');
+			  pid, address, node->id, migration ? 'M' : 'U');
 }
 
-void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
 				   unsigned long start, unsigned long end,
 				   uint32_t from, uint32_t to,
 				   uint32_t prefetch_loc, uint32_t preferred_loc,
 				   uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_MIGRATE_START,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_START,
 			  "%lld -%d @%lx(%lx) %x->%x %x:%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, prefetch_loc, preferred_loc, trigger);
 }
 
-void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
 				 unsigned long start, unsigned long end,
 				 uint32_t from, uint32_t to, uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_MIGRATE_END,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_END,
 			  "%lld -%d @%lx(%lx) %x->%x %d\n",
 			  ktime_get_boottime_ns(), pid, start, end - start,
 			  from, to, trigger);
 }
 
-void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_QUEUE_EVICTION,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_EVICTION,
 			  "%lld -%d %x %d\n", ktime_get_boottime_ns(), pid,
-			  dev->nodes[0]->id, trigger);
+			  node->id, trigger);
 }
 
-void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid)
+void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_QUEUE_RESTORE,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_RESTORE,
 			  "%lld -%d %x\n", ktime_get_boottime_ns(), pid,
-			  dev->nodes[0]->id);
+			  node->id);
 }
 
 void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
@@ -324,13 +324,13 @@ void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm)
 	kfd_unref_process(p);
 }
 
-void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid,
 				  unsigned long address, unsigned long last,
 				  uint32_t trigger)
 {
-	kfd_smi_event_add(pid, dev->nodes[0], KFD_SMI_EVENT_UNMAP_FROM_GPU,
+	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_UNMAP_FROM_GPU,
 			  "%lld -%d @%lx(%lx) %x %d\n", ktime_get_boottime_ns(),
-			  pid, address, last - address + 1, dev->nodes[0]->id, trigger);
+			  pid, address, last - address + 1, node->id, trigger);
 }
 
 int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
index 59cd089f80d18..fa95c2dfd587f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h
@@ -29,24 +29,24 @@ void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid);
 void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev,
 					     uint64_t throttle_bitmask);
 void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset);
-void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_page_fault_start(struct kfd_node *node, pid_t pid,
 				    unsigned long address, bool write_fault,
 				    ktime_t ts);
-void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_page_fault_end(struct kfd_node *node, pid_t pid,
 				  unsigned long address, bool migration);
-void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
 			     unsigned long start, unsigned long end,
 			     uint32_t from, uint32_t to,
 			     uint32_t prefetch_loc, uint32_t preferred_loc,
 			     uint32_t trigger);
-void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
 			     unsigned long start, unsigned long end,
 			     uint32_t from, uint32_t to, uint32_t trigger);
-void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid,
 				  uint32_t trigger);
-void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid);
+void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid);
 void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm);
-void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid,
+void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid,
 				  unsigned long address, unsigned long last,
 				  uint32_t trigger);
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 639831fbb6caf..0dafbbe954ca4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1274,7 +1274,7 @@ svm_range_unmap_from_gpus(struct svm_range *prange, unsigned long start,
 			return -EINVAL;
 		}
 
-		kfd_smi_event_unmap_from_gpu(pdd->dev->kfd, p->lead_thread->pid,
+		kfd_smi_event_unmap_from_gpu(pdd->dev, p->lead_thread->pid,
 					     start, last, trigger);
 
 		r = svm_range_unmap_from_gpu(pdd->dev->adev,
@@ -2934,7 +2934,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		 svms, prange->start, prange->last, best_loc,
 		 prange->actual_loc);
 
-	kfd_smi_event_page_fault_start(adev->kfd.dev, p->lead_thread->pid, addr,
+	kfd_smi_event_page_fault_start(node, p->lead_thread->pid, addr,
 				       write_fault, timestamp);
 
 	if (prange->actual_loc != best_loc) {
@@ -2972,7 +2972,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		pr_debug("failed %d to map svms 0x%p [0x%lx 0x%lx] to gpus\n",
 			 r, svms, prange->start, prange->last);
 
-	kfd_smi_event_page_fault_end(adev->kfd.dev, p->lead_thread->pid, addr,
+	kfd_smi_event_page_fault_end(node, p->lead_thread->pid, addr,
 				     migration);
 
 out_unlock_range:
-- 
GitLab


From 58bd8023752f3d7e6888d5cbb27387853b04c431 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 24 Aug 2022 17:41:47 +0800
Subject: [PATCH 1104/1791] drm/amdgpu: adjust s2a entry register for sdma
 doorbell trans decoding

Use s2a entry 5/6 registers to decode sdma doorbell trans on different AIDs,
which aligns the entry table in SHUB spec, and leave entry 4 dedicated for VCN
doorbell to avoid conflict.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 32 +++++++-------------------
 1 file changed, 8 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index ef0b557e9b3eb..5dbab8ab9b274 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -151,18 +151,10 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x8);
-		if (aid_id != 0)
-			WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0,
-				regS2A_DOORBELL_ENTRY_3_CTRL)
-				+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
-				+ AMDGPU_SMN_TARGET_AID(aid_id)
-				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-				doorbell_ctrl);
-		else
-			WREG32(SOC15_REG_OFFSET(NBIO, 0,
-				regS2A_DOORBELL_ENTRY_5_CTRL)
-				+ S2A_DOORBELL_REG_LSD_OFFSET,
-				doorbell_ctrl);
+		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_5_CTRL) * 4
+			+ AMDGPU_SMN_TARGET_AID(aid_id)
+			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+			doorbell_ctrl);
 		break;
 	case 3:
 		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4) +
@@ -178,18 +170,10 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x9);
-		if (aid_id != 0)
-			WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0,
-				regS2A_DOORBELL_ENTRY_4_CTRL)
-				+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
-				+ AMDGPU_SMN_TARGET_AID(aid_id)
-				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-				doorbell_ctrl);
-		else
-			WREG32(SOC15_REG_OFFSET(NBIO, 0,
-				regS2A_DOORBELL_ENTRY_6_CTRL)
-				+ S2A_DOORBELL_REG_LSD_OFFSET,
-				doorbell_ctrl);
+		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_6_CTRL) * 4
+			+ AMDGPU_SMN_TARGET_AID(aid_id)
+			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
+			doorbell_ctrl);
 		break;
 	default:
 		break;
-- 
GitLab


From 9cd51d53695e1df134301c1bdb8a8f965506e35b Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 25 Aug 2022 15:51:43 +0800
Subject: [PATCH 1105/1791] drm/amdgpu: drop redundant csb init for gfx943

It's not required for compute pipeline and will cause soft lockup on emulation
due to long-time writing.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 99 -------------------------
 1 file changed, 99 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 14b96b43f02c3..73f652ad5b00b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -429,75 +429,6 @@ static int gfx_v9_4_3_init_microcode(struct amdgpu_device *adev)
 	return r;
 }
 
-static u32 gfx_v9_4_3_get_csb_size(struct amdgpu_device *adev)
-{
-	u32 count = 0;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
-
-	/* begin clear state */
-	count += 2;
-	/* context control state */
-	count += 3;
-
-	for (sect = gfx9_cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT)
-				count += 2 + ext->reg_count;
-			else
-				return 0;
-		}
-	}
-
-	/* end clear state */
-	count += 2;
-	/* clear state */
-	count += 2;
-
-	return count;
-}
-
-static void gfx_v9_4_3_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
-{
-	u32 count = 0, i;
-	const struct cs_section_def *sect = NULL;
-	const struct cs_extent_def *ext = NULL;
-
-	if (adev->gfx.rlc.cs_data == NULL)
-		return;
-	if (buffer == NULL)
-		return;
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_BEGIN_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CONTEXT_CONTROL, 1));
-	buffer[count++] = cpu_to_le32(0x80000000);
-	buffer[count++] = cpu_to_le32(0x80000000);
-
-	for (sect = adev->gfx.rlc.cs_data; sect->section != NULL; ++sect) {
-		for (ext = sect->section; ext->extent != NULL; ++ext) {
-			if (sect->id == SECT_CONTEXT) {
-				buffer[count++] =
-					cpu_to_le32(PACKET3(PACKET3_SET_CONTEXT_REG, ext->reg_count));
-				buffer[count++] = cpu_to_le32(ext->reg_index -
-						PACKET3_SET_CONTEXT_REG_START);
-				for (i = 0; i < ext->reg_count; i++)
-					buffer[count++] = cpu_to_le32(ext->extent[i]);
-			} else {
-				return;
-			}
-		}
-	}
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
-	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
-
-	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_CLEAR_STATE, 0));
-	buffer[count++] = cpu_to_le32(0);
-}
-
 static void gfx_v9_4_3_mec_fini(struct amdgpu_device *adev)
 {
 	amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL);
@@ -1112,22 +1043,8 @@ static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev,
 	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_SRM_CNTL, SRM_ENABLE, 1);
 }
 
-static void gfx_v9_4_3_init_csb(struct amdgpu_device *adev, int xcc_id)
-{
-	adev->gfx.rlc.funcs->get_csb_buffer(adev, adev->gfx.rlc.cs_ptr);
-	/* csib */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_HI),
-			adev->gfx.rlc.clear_state_gpu_addr >> 32);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_ADDR_LO),
-			adev->gfx.rlc.clear_state_gpu_addr & 0xfffffffc);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_CSIB_LENGTH),
-			adev->gfx.rlc.clear_state_size);
-}
-
 static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 {
-	gfx_v9_4_3_init_csb(adev, xcc_id);
-
 	/*
 	 * Rlc save restore list is workable since v2_1.
 	 * And it's needed by gfxoff feature.
@@ -1219,20 +1136,6 @@ static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
 
 static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
 {
-	const struct cs_section_def *cs_data;
-	int r;
-
-	adev->gfx.rlc.cs_data = gfx9_cs_data;
-
-	cs_data = adev->gfx.rlc.cs_data;
-
-	if (cs_data) {
-		/* init clear state block */
-		r = amdgpu_gfx_rlc_init_csb(adev);
-		if (r)
-			return r;
-	}
-
 	/* init spm vmid with 0xf */
 	if (adev->gfx.rlc.funcs->update_spm_vmid)
 		adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf);
@@ -2406,8 +2309,6 @@ static const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = {
 	.set_safe_mode = gfx_v9_4_3_set_safe_mode,
 	.unset_safe_mode = gfx_v9_4_3_unset_safe_mode,
 	.init = gfx_v9_4_3_rlc_init,
-	.get_csb_size = gfx_v9_4_3_get_csb_size,
-	.get_csb_buffer = gfx_v9_4_3_get_csb_buffer,
 	.resume = gfx_v9_4_3_rlc_resume,
 	.stop = gfx_v9_4_3_rlc_stop,
 	.reset = gfx_v9_4_3_rlc_reset,
-- 
GitLab


From a058799923af7984552c6c07d6bbb088c1537e6a Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Tue, 12 Apr 2022 15:37:15 -0400
Subject: [PATCH 1106/1791] drm/amdgpu: Fix SWS on multi-XCD GPU

GFX_9_4_3 supports multi-XCDs and multi-AIDs in one GPU device. SWS needs
to program IH_VMID_x_LUT with specified XCC instance and corresponded
AID instance.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   | 31 +++++++++++++------
 1 file changed, 22 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 772774615cb81..58adbf2262cb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -21,14 +21,13 @@
  */
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
-#include "amdgpu_amdkfd_arcturus.h"
 #include "amdgpu_amdkfd_gfx_v9.h"
 #include "gc/gc_9_4_3_offset.h"
 #include "gc/gc_9_4_3_sh_mask.h"
 #include "athub/athub_1_8_0_offset.h"
 #include "athub/athub_1_8_0_sh_mask.h"
-#include "oss/osssys_4_0_offset.h"
-#include "oss/osssys_4_0_sh_mask.h"
+#include "oss/osssys_4_4_2_offset.h"
+#include "oss/osssys_4_4_2_sh_mask.h"
 #include "v9_structs.h"
 #include "soc15.h"
 #include "sdma/sdma_4_4_2_offset.h"
@@ -220,9 +219,12 @@ int kgd_gfx_v9_4_3_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
 }
 
 static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
-				u32 pasid, unsigned int vmid, uint32_t inst)
+			u32 pasid, unsigned int vmid, uint32_t xcc_inst)
 {
 	unsigned long timeout;
+	unsigned int reg;
+	/* Every two XCCs share one AID */
+	unsigned int aid = xcc_inst / 2;
 
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -234,11 +236,11 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 	uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid |
 			ATC_VMID0_PASID_MAPPING__VALID_MASK;
 
-	WREG32(SOC15_REG_OFFSET(ATHUB, inst,
+	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
 		regATC_VMID0_PASID_MAPPING) + vmid, pasid_mapping);
 
 	timeout = jiffies + msecs_to_jiffies(10);
-	while (!(RREG32(SOC15_REG_OFFSET(ATHUB, inst,
+	while (!(RREG32(SOC15_REG_OFFSET(ATHUB, 0,
 			regATC_VMID_PASID_MAPPING_UPDATE_STATUS)) &
 			(1U << vmid))) {
 		if (time_after(jiffies, timeout)) {
@@ -248,14 +250,25 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 		cpu_relax();
 	}
 
-	WREG32(SOC15_REG_OFFSET(ATHUB, inst,
+	WREG32(SOC15_REG_OFFSET(ATHUB, 0,
 		regATC_VMID_PASID_MAPPING_UPDATE_STATUS),
 		1U << vmid);
 
-	WREG32(SOC15_REG_OFFSET(OSSSYS, inst, mmIH_VMID_0_LUT) + vmid,
+	reg = RREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX));
+	/* Every 4 numbers is a cycle. 1st is AID, 2nd and 3rd are XCDs,
+	 * and the 4th is reserved. Therefore "aid * 4 + (xcc_inst % 2) + 1"
+	 * programs _LUT for XCC and "aid * 4" for AID where the XCC connects
+	 * to.
+	 */
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX),
+		aid * 4 + (xcc_inst % 2) + 1);
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid,
 		pasid_mapping);
-	WREG32(SOC15_REG_OFFSET(OSSSYS, inst, mmIH_VMID_0_LUT_MM) + vmid,
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX),
+		aid * 4);
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT_MM) + vmid,
 		pasid_mapping);
+	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX), reg);
 
 	return 0;
 }
-- 
GitLab


From 02ee3b02d77661c593c7b0e49f5baa4aa0974024 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 13 Sep 2022 15:13:18 -0400
Subject: [PATCH 1107/1791] drm/amdkfd: Use instance table for GFX 9.4.3

For GFX 9.4.3, use the logical to physical mapping table,
to get the correct XCD instance when accessing registers on
bare metal.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c   |  5 +-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 60 +++++++++----------
 2 files changed, 33 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 58adbf2262cb2..81dfbe39fd8ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -223,8 +223,9 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 {
 	unsigned long timeout;
 	unsigned int reg;
+	unsigned int phy_inst = GET_INST(GC, xcc_inst);
 	/* Every two XCCs share one AID */
-	unsigned int aid = xcc_inst / 2;
+	unsigned int aid = phy_inst / 2;
 
 	/*
 	 * We have to assume that there is no outstanding mapping.
@@ -261,7 +262,7 @@ static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev,
 	 * to.
 	 */
 	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX),
-		aid * 4 + (xcc_inst % 2) + 1);
+		aid * 4 + (phy_inst % 2) + 1);
 	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid,
 		pasid_mapping);
 	WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX),
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 7918a00cbb5a0..9fa9aab22fe98 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -50,12 +50,12 @@ static void kgd_gfx_v9_lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint3
 			uint32_t queue, uint32_t vmid, uint32_t inst)
 {
 	mutex_lock(&adev->srbm_mutex);
-	soc15_grbm_select(adev, mec, pipe, queue, vmid, inst);
+	soc15_grbm_select(adev, mec, pipe, queue, vmid, GET_INST(GC, inst));
 }
 
 static void kgd_gfx_v9_unlock_srbm(struct amdgpu_device *adev, uint32_t inst)
 {
-	soc15_grbm_select(adev, 0, 0, 0, 0, inst);
+	soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, inst));
 	mutex_unlock(&adev->srbm_mutex);
 }
 
@@ -90,8 +90,8 @@ void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 {
 	kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst);
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmSH_MEM_CONFIG), sh_mem_config);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmSH_MEM_BASES), sh_mem_bases);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_CONFIG), sh_mem_config);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_BASES), sh_mem_bases);
 	/* APE1 no longer exists on GFX9 */
 
 	kgd_gfx_v9_unlock_srbm(adev, inst);
@@ -167,7 +167,7 @@ int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id,
 
 	kgd_gfx_v9_lock_srbm(adev, mec, pipe, 0, 0, inst);
 
-	WREG32_SOC15(GC, inst, mmCPC_INT_CNTL,
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmCPC_INT_CNTL,
 		CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK |
 		CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK);
 
@@ -234,17 +234,17 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	/* HQD registers extend from CP_MQD_BASE_ADDR to CP_HQD_EOP_WPTR_MEM. */
 	mqd_hqd = &m->cp_mqd_base_addr_lo;
-	hqd_base = SOC15_REG_OFFSET(GC, inst, mmCP_MQD_BASE_ADDR);
+	hqd_base = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_MQD_BASE_ADDR);
 
 	for (reg = hqd_base;
-	     reg <= SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI); reg++)
+	     reg <= SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI); reg++)
 		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
 
 
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_DOORBELL_CONTROL),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_DOORBELL_CONTROL),
 					data);
 
 	if (wptr) {
@@ -274,25 +274,25 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_LO),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_LO),
 		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI),
 		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_POLL_ADDR),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR),
 		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
+		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
 		       upper_32_bits((uintptr_t)wptr));
-		WREG32_SOC15(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1,
+		WREG32_SOC15(GC, GET_INST(GC, inst), mmCP_PQ_WPTR_POLL_CNTL1,
 		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_EOP_RPTR),
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_EOP_RPTR),
 	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
 			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_ACTIVE), data);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE), data);
 
 	kgd_gfx_v9_release_queue(adev, inst);
 
@@ -370,8 +370,8 @@ int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev,
 
 	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
-	for (reg = SOC15_REG_OFFSET(GC, inst, mmCP_MQD_BASE_ADDR);
-	     reg <= SOC15_REG_OFFSET(GC, inst, mmCP_HQD_PQ_WPTR_HI); reg++)
+	for (reg = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_MQD_BASE_ADDR);
+	     reg <= SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI); reg++)
 		DUMP_REG(reg);
 
 	kgd_gfx_v9_release_queue(adev, inst);
@@ -491,13 +491,13 @@ bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev,
 	uint32_t low, high;
 
 	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
-	act = RREG32_SOC15(GC, inst, mmCP_HQD_ACTIVE);
+	act = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE);
 	if (act) {
 		low = lower_32_bits(queue_address >> 8);
 		high = upper_32_bits(queue_address >> 8);
 
-		if (low == RREG32_SOC15(GC, inst, mmCP_HQD_PQ_BASE) &&
-		   high == RREG32_SOC15(GC, inst, mmCP_HQD_PQ_BASE_HI))
+		if (low == RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE) &&
+		   high == RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE_HI))
 			retval = true;
 	}
 	kgd_gfx_v9_release_queue(adev, inst);
@@ -538,7 +538,7 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 	kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst);
 
 	if (m->cp_hqd_vmid == 0)
-		WREG32_FIELD15_RLC(GC, inst, RLC_CP_SCHEDULERS, scheduler1, 0);
+		WREG32_FIELD15_RLC(GC, GET_INST(GC, inst), RLC_CP_SCHEDULERS, scheduler1, 0);
 
 	switch (reset_type) {
 	case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN:
@@ -555,11 +555,11 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 		break;
 	}
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, inst, mmCP_HQD_DEQUEUE_REQUEST), type);
+	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_DEQUEUE_REQUEST), type);
 
 	end_jiffies = (utimeout * HZ / 1000) + jiffies;
 	while (true) {
-		temp = RREG32_SOC15(GC, inst, mmCP_HQD_ACTIVE);
+		temp = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE);
 		if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK))
 			break;
 		if (time_after(jiffies, end_jiffies)) {
@@ -633,8 +633,8 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 
 	mutex_lock(&adev->grbm_idx_mutex);
 
-	WREG32_SOC15_RLC_SHADOW(GC, inst, mmGRBM_GFX_INDEX, gfx_index_val);
-	WREG32_SOC15(GC, inst, mmSQ_CMD, sq_cmd);
+	WREG32_SOC15_RLC_SHADOW(GC, GET_INST(GC, inst), mmGRBM_GFX_INDEX, gfx_index_val);
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_CMD, sq_cmd);
 
 	data = REG_SET_FIELD(data, GRBM_GFX_INDEX,
 		INSTANCE_BROADCAST_WRITES, 1);
@@ -643,7 +643,7 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 	data = REG_SET_FIELD(data, GRBM_GFX_INDEX,
 		SE_BROADCAST_WRITES, 1);
 
-	WREG32_SOC15_RLC_SHADOW(GC, inst, mmGRBM_GFX_INDEX, data);
+	WREG32_SOC15_RLC_SHADOW(GC, GET_INST(GC, inst), mmGRBM_GFX_INDEX, data);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	return 0;
@@ -842,17 +842,17 @@ void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
 	/*
 	 * Program TBA registers
 	 */
-	WREG32_SOC15(GC, inst, mmSQ_SHADER_TBA_LO,
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_LO,
                         lower_32_bits(tba_addr >> 8));
-	WREG32_SOC15(GC, inst, mmSQ_SHADER_TBA_HI,
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_HI,
                         upper_32_bits(tba_addr >> 8));
 
 	/*
 	 * Program TMA registers
 	 */
-	WREG32_SOC15(GC, inst, mmSQ_SHADER_TMA_LO,
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TMA_LO,
 			lower_32_bits(tma_addr >> 8));
-	WREG32_SOC15(GC, inst, mmSQ_SHADER_TMA_HI,
+	WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TMA_HI,
 			upper_32_bits(tma_addr >> 8));
 
 	kgd_gfx_v9_unlock_srbm(adev, inst);
-- 
GitLab


From cb30544e3cadf2164a123859519521dc474d21eb Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Thu, 29 Sep 2022 12:07:49 -0400
Subject: [PATCH 1108/1791] drm/amdgpu: Fix failure when switching to DPX mode

Fix the if condition which causes dynamic repartitioning
to fail when trying to switch to DPX mode.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 69bac5b801cee..2f6d85090b55e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1216,7 +1216,11 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 	if (!strncasecmp("SPX", buf, strlen("SPX"))) {
 		mode = AMDGPU_SPX_PARTITION_MODE;
 	} else if (!strncasecmp("DPX", buf, strlen("DPX"))) {
-		if (num_xcc != 4 || num_xcc != 8)
+		/*
+		 * DPX mode needs AIDs to be in multiple of 2.
+		 * Each AID connects 2 XCCs.
+		 */
+		if (num_xcc%4)
 			return -EINVAL;
 		mode = AMDGPU_DPX_PARTITION_MODE;
 	} else if (!strncasecmp("TPX", buf, strlen("TPX"))) {
-- 
GitLab


From f5fe7edfd6ce62cd23fbd707e7f9fe0f56a45e94 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Fri, 30 Sep 2022 09:16:21 -0400
Subject: [PATCH 1109/1791] drm/amdkfd: Update interrupt handling for GFX9.4.3

Update interrupt handling in CPX mode for GFX9.4.3 by using the
VMID space instead of SDMA client id to determine if an interrupt
should be processed by a KFD node. This is especially needed for
handling retry faults from MMHUB.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c |  7 +++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c  |  4 ++--
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h  | 16 ++++++----------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c   |  8 ++++----
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h   |  2 +-
 6 files changed, 19 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index c390b2856cc93..22a900f298f2e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2434,6 +2434,9 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
  * amdgpu_vm_handle_fault - graceful handling of VM faults.
  * @adev: amdgpu device pointer
  * @pasid: PASID of the VM
+ * @vmid: VMID, only used for GFX 9.4.3.
+ * @node_id: Node_id received in IH cookie. Only applicable for
+ *           GFX 9.4.3.
  * @addr: Address of the fault
  * @write_fault: true is write fault, false is read fault
  *
@@ -2441,7 +2444,7 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm)
  * shouldn't be reported any more.
  */
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    u32 client_id, u32 node_id, uint64_t addr,
+			    u32 vmid, u32 node_id, uint64_t addr,
 			    bool write_fault)
 {
 	bool is_compute_context = false;
@@ -2466,7 +2469,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
 
 	addr /= AMDGPU_GPU_PAGE_SIZE;
 
-	if (is_compute_context && !svm_range_restore_pages(adev, pasid, client_id,
+	if (is_compute_context && !svm_range_restore_pages(adev, pasid, vmid,
 	    node_id, addr, write_fault)) {
 		amdgpu_bo_unref(&root);
 		return true;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index dbab31647186b..8add5f5eb92a2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -455,7 +455,7 @@ void amdgpu_vm_check_compute_bug(struct amdgpu_device *adev);
 void amdgpu_vm_get_task_info(struct amdgpu_device *adev, u32 pasid,
 			     struct amdgpu_task_info *task_info);
 bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
-			    u32 client_id, u32 node_id, uint64_t addr,
+			    u32 vmid, u32 node_id, uint64_t addr,
 			    bool write_fault);
 
 void amdgpu_vm_set_task_info(struct amdgpu_vm *vm);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index c5752a349f3d9..f2814270da401 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -587,7 +587,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 
 			cam_index = entry->src_data[2] & 0x3ff;
 
-			ret = amdgpu_vm_handle_fault(adev, entry->pasid, entry->client_id, node_id,
+			ret = amdgpu_vm_handle_fault(adev, entry->pasid, entry->vmid, node_id,
 						     addr, write_fault);
 			WDOORBELL32(adev->irq.retry_cam_doorbell_index, cam_index);
 			if (ret)
@@ -610,7 +610,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 			/* Try to handle the recoverable page faults by filling page
 			 * tables
 			 */
-			if (amdgpu_vm_handle_fault(adev, entry->pasid, entry->client_id, node_id,
+			if (amdgpu_vm_handle_fault(adev, entry->pasid, entry->vmid, node_id,
 						   addr, write_fault))
 				return 1;
 		}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index df372de6b056c..fb3cf2c51da80 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1073,18 +1073,14 @@ struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev);
 struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev);
-static inline bool kfd_irq_is_from_node(struct kfd_node *node, uint32_t client_id,
-				     uint32_t node_id)
+static inline bool kfd_irq_is_from_node(struct kfd_node *node, uint32_t node_id,
+					uint32_t vmid)
 {
-	if ((node->interrupt_bitmap & (0x1U << node_id)) ||
-	    ((node_id % 4) == 0 &&
-	    (node->interrupt_bitmap >> 16) & (0x1U << client_id)))
-		return true;
-
-	return false;
+	return (node->interrupt_bitmap & (1 << node_id)) != 0 &&
+	       (node->compute_vmid_bitmap & (1 << vmid)) != 0;
 }
 static inline struct kfd_node *kfd_node_by_irq_ids(struct amdgpu_device *adev,
-					uint32_t client_id, uint32_t node_id) {
+					uint32_t node_id, uint32_t vmid) {
 	struct kfd_dev *dev = adev->kfd.dev;
 	uint32_t i;
 
@@ -1092,7 +1088,7 @@ static inline struct kfd_node *kfd_node_by_irq_ids(struct amdgpu_device *adev,
 		return dev->nodes[0];
 
 	for (i = 0; i < dev->num_nodes; i++)
-		if (kfd_irq_is_from_node(dev->nodes[i], client_id, node_id))
+		if (kfd_irq_is_from_node(dev->nodes[i], node_id, vmid))
 			return dev->nodes[i];
 
 	return NULL;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 0dafbbe954ca4..5d6e02559d8ee 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -2799,7 +2799,7 @@ svm_fault_allowed(struct vm_area_struct *vma, bool write_fault)
 
 int
 svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
-			uint32_t client_id, uint32_t node_id,
+			uint32_t vmid, uint32_t node_id,
 			uint64_t addr, bool write_fault)
 {
 	struct mm_struct *mm = NULL;
@@ -2851,10 +2851,10 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		goto out;
 	}
 
-	node = kfd_node_by_irq_ids(adev, node_id, client_id);
+	node = kfd_node_by_irq_ids(adev, node_id, vmid);
 	if (!node) {
-		pr_debug("kfd node does not exist node_id: %d, client_id: %d\n", node_id,
-			 client_id);
+		pr_debug("kfd node does not exist node_id: %d, vmid: %d\n", node_id,
+			 vmid);
 		r = -EFAULT;
 		goto out;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index a165c73b40b28..5116786718b66 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -173,7 +173,7 @@ int svm_range_split_by_granularity(struct kfd_process *p, struct mm_struct *mm,
 			       unsigned long addr, struct svm_range *parent,
 			       struct svm_range *prange);
 int svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
-			    uint32_t client_id, uint32_t node_id, uint64_t addr,
+			    uint32_t vmid, uint32_t node_id, uint64_t addr,
 			    bool write_fault);
 int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence);
 void svm_range_add_list_work(struct svm_range_list *svms,
-- 
GitLab


From 955220b04d42c41050158fec0f53957f320b96f9 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 6 Oct 2022 15:25:08 +0530
Subject: [PATCH 1110/1791] drm/amdgpu: Fix programming of initial XCP mode

On initialization set the partition mode correctly to SPX (default) or
any other user specified partition mode. Use switch_compute_partition
API so that all settings are initialized correctly.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 28 ++++++-------------------
 1 file changed, 6 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 73f652ad5b00b..b6b7dbb62448e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1891,6 +1891,11 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 				return r;
 		}
 
+		if (adev->gfx.partition_mode ==
+		    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
+			gfx_v9_4_3_switch_compute_partition(
+				adev, amdgpu_user_partt_mode);
+
 		/* set the virtual and physical id based on partition_mode */
 		gfx_v9_4_3_program_xcc_id(adev, i);
 
@@ -2112,28 +2117,7 @@ static int gfx_v9_4_3_early_init(void *handle)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
-	adev->gfx.partition_mode = amdgpu_user_partt_mode;
-	/* calculate the num_xcc_in_xcp for the partition mode*/
-	switch (amdgpu_user_partt_mode) {
-	case AMDGPU_SPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = num_xcc;
-		break;
-	case AMDGPU_DPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = num_xcc / 2;
-		break;
-	case AMDGPU_TPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = num_xcc / 3;
-		break;
-	case AMDGPU_QPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = num_xcc / 4;
-		break;
-	case AMDGPU_CPX_PARTITION_MODE:
-		adev->gfx.num_xcc_per_xcp = 1;
-		break;
-	default:
-		adev->gfx.num_xcc_per_xcp = num_xcc;
-		break;
-	}
+	adev->gfx.partition_mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
 
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
-- 
GitLab


From 5a8b26a88639d69453d592ee11c03a24e0b62b9a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 19 Sep 2022 18:08:25 +0530
Subject: [PATCH 1111/1791] drm/amdgpu: Fix register access on GC v9.4.3

In GC v9.4.3 there are multiple XCCs. It's required to use
physical instance number to get the right register offset. Use
GET_INST API for that.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index b6b7dbb62448e..f1e6da4d62a34 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1600,7 +1600,7 @@ static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
 	int j;
 
 	/* disable wptr polling */
-	WREG32_FIELD15_PREREG(GC, xcc_id, CP_PQ_WPTR_POLL_CNTL, EN, 0);
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_WPTR_POLL_CNTL, EN, 0);
 
 	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR,
 	       mqd->cp_hqd_eop_base_addr_lo);
@@ -1693,7 +1693,7 @@ static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
 	       mqd->cp_hqd_active);
 
 	if (ring->use_doorbell)
-		WREG32_FIELD15_PREREG(GC, xcc_id, CP_PQ_STATUS, DOORBELL_ENABLE, 1);
+		WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_STATUS, DOORBELL_ENABLE, 1);
 
 	return 0;
 }
-- 
GitLab


From 6df442a03d1a839242397259fa13168858c52413 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Mon, 25 Apr 2022 22:19:58 +0800
Subject: [PATCH 1112/1791] drm/amdgpu: add new doorbell assignment table for
 aqua_vanjaram

Four basic reasons as below to do the change:
  1. number of ring expand a lot on aqua_vanjaram, and adjustment on old
     assignment cannot make each ring in a continuous doorbell space.
  2. the SDMA doorbell index should not exceed 0x1FF on aqua_vanjaram due to
     regDOORBELLx_CTRL_ENTRY.BIF_DOORBELLx_RANGE_OFFSET_ENTRY field width.
  3. re-design the doorbell assignment and unify the calculation as
     "start + ring/inst id" will make the code much concise.
  4. only defining the START/END makes the table look simple

v2: (Lijo)
  1. replace name
  2. use num_inst_per_aid/sdma_doorbell_range instead of hardcoding

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile           |  2 +-
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 52 +++++++++++++++++++
 2 files changed, 53 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 594e303084c52..90f771423c947 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -78,7 +78,7 @@ amdgpu-y += \
 	vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \
 	nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \
 	sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o \
-	nbio_v7_9.o
+	nbio_v7_9.o aqua_vanjaram_reg_init.o
 
 # add DF block
 amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
new file mode 100644
index 0000000000000..3b97bc922d4ae
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -0,0 +1,52 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "soc15.h"
+
+#include "soc15_common.h"
+
+void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
+{
+	int i;
+
+	adev->doorbell_index.kiq = AMDGPU_DOORBELL_LAYOUT1_KIQ_START;
+
+	adev->doorbell_index.mec_ring0 = AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START;
+
+	adev->doorbell_index.userqueue_start = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START;
+	adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END;
+
+	adev->doorbell_index.sdma_doorbell_range = 20;
+	for (i = 0; i < adev->num_aid * adev->sdma.num_inst_per_aid; i++)
+		adev->doorbell_index.sdma_engine[i] =
+			AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START +
+			i * (adev->doorbell_index.sdma_doorbell_range >> 1);
+
+	adev->doorbell_index.ih = AMDGPU_DOORBELL_LAYOUT1_IH;
+	adev->doorbell_index.vcn.vcn_ring0_1 = AMDGPU_DOORBELL_LAYOUT1_VCN_START;
+
+	adev->doorbell_index.first_non_cp = AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP;
+	adev->doorbell_index.last_non_cp = AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP;
+
+	adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1;
+}
-- 
GitLab


From cab7d478da112e66f2ad8eec7dcfc0aa2a5babe1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 15:45:06 +0530
Subject: [PATCH 1113/1791] drm/amdgpu: Add IP instance map for aqua vanjaram

Add XCC logical to physical instance map for aqua vanjaram

v2:
	Keep look up table only for required IPs, for others return
default mapping (Felix).

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 13 +++++++
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 39 +++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/soc15.h            |  1 +
 3 files changed, 53 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index b58d94dc19246..915ae0bcdab21 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1951,6 +1951,17 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev)
 	return 0;
 }
 
+static void amdgpu_discovery_ip_map_init(struct amdgpu_device *adev)
+{
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 3):
+		aqua_vanjaram_ip_map_init(adev);
+		break;
+	default:
+		break;
+	}
+}
+
 int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 {
 	int r;
@@ -2128,6 +2139,8 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		break;
 	}
 
+	amdgpu_discovery_ip_map_init(adev);
+
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 0, 1):
 	case IP_VERSION(9, 2, 1):
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 3b97bc922d4ae..dbff8220fc198 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -50,3 +50,42 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
 
 	adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1;
 }
+
+static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
+					 enum amd_hw_ip_block_type block,
+					 int8_t inst)
+{
+	int8_t dev_inst;
+
+	switch (block) {
+	case GC_HWIP:
+		dev_inst = adev->ip_map.dev_inst[block][inst];
+		break;
+	default:
+		/* For rest of the IPs, no look up required.
+		 * Assume 'logical instance == physical instance' for all configs. */
+		dev_inst = inst;
+		break;
+	}
+
+	return dev_inst;
+}
+
+void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
+{
+	int xcc_mask;
+	int l, i;
+
+	/* Map GC instances */
+	l = 0;
+	xcc_mask = adev->gfx.xcc_mask;
+	while (xcc_mask) {
+		i = ffs(xcc_mask) - 1;
+		adev->ip_map.dev_inst[GC_HWIP][l++] = i;
+		xcc_mask &= ~(1 << i);
+	}
+	for (; l < HWIP_MAX_INSTANCE; l++)
+		adev->ip_map.dev_inst[GC_HWIP][l] = -1;
+
+	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index 2b41ee968dd10..d8a2a6c2c6e3f 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -111,6 +111,7 @@ int vega10_reg_base_init(struct amdgpu_device *adev);
 int vega20_reg_base_init(struct amdgpu_device *adev);
 int arct_reg_base_init(struct amdgpu_device *adev);
 int aldebaran_reg_base_init(struct amdgpu_device *adev);
+void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev);
 
 void vega10_doorbell_index_init(struct amdgpu_device *adev);
 void vega20_doorbell_index_init(struct amdgpu_device *adev);
-- 
GitLab


From cf24f6a0d7137bc703a23187ffa4a65ed3f17820 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 16:26:49 +0530
Subject: [PATCH 1114/1791] drm/amdgpu: Add mask for SDMA instances

Add a mask of SDMA instances available for use. On certain ASIC configs,
not all SDMA instances are available for software use.

v2:
	Change sdma mask type to uint32_t (Le)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index dbff8220fc198..5a1511a223677 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -37,7 +37,7 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END;
 
 	adev->doorbell_index.sdma_doorbell_range = 20;
-	for (i = 0; i < adev->num_aid * adev->sdma.num_inst_per_aid; i++)
+	for (i = 0; i < adev->sdma.num_instances; i++)
 		adev->doorbell_index.sdma_engine[i] =
 			AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START +
 			i * (adev->doorbell_index.sdma_doorbell_range >> 1);
-- 
GitLab


From 5c606836eff823ea14c481ad6374bc6d87121182 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 29 Jun 2022 21:04:39 +0530
Subject: [PATCH 1115/1791] drm/amdgpu: Use SDMA instance table for aqua
 vanjaram

For aqua vanjaram, add mapping for logical to physical
instances.

v2:
	Register accesses on bare metal should be based on physical
instance. Use GET_INST() to get physical instance.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 5a1511a223677..12379c374457f 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -59,6 +59,7 @@ static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 
 	switch (block) {
 	case GC_HWIP:
+	case SDMA0_HWIP:
 		dev_inst = adev->ip_map.dev_inst[block][inst];
 		break;
 	default:
@@ -73,7 +74,7 @@ static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 
 void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 {
-	int xcc_mask;
+	int xcc_mask, sdma_mask;
 	int l, i;
 
 	/* Map GC instances */
@@ -87,5 +88,15 @@ void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 	for (; l < HWIP_MAX_INSTANCE; l++)
 		adev->ip_map.dev_inst[GC_HWIP][l] = -1;
 
+	l = 0;
+	sdma_mask = adev->sdma.sdma_mask;
+	while (sdma_mask) {
+		i = ffs(sdma_mask) - 1;
+		adev->ip_map.dev_inst[SDMA0_HWIP][l++] = i;
+		sdma_mask &= ~(1 << i);
+	}
+	for (; l < HWIP_MAX_INSTANCE; l++)
+		adev->ip_map.dev_inst[SDMA0_HWIP][l] = -1;
+
 	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
 }
-- 
GitLab


From 1dfcdc30270a80ba5b45f922833c0c0e56d82576 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 24 May 2022 19:44:27 +0800
Subject: [PATCH 1116/1791] drm/amdgpu: switch to
 aqua_vanjaram_doorbell_index_init

New doorbell index assignment is used by aqua_vanjaram.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 24 +++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/soc15.h |  1 +
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index b9bcb12bff91e..29d2e08834c78 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -853,6 +853,28 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs =
 	.query_video_codecs = &soc15_query_video_codecs,
 };
 
+static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs =
+{
+	.read_disabled_bios = &soc15_read_disabled_bios,
+	.read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom,
+	.read_register = &soc15_read_register,
+	.reset = &soc15_asic_reset,
+	.reset_method = &soc15_asic_reset_method,
+	.set_vga_state = &soc15_vga_set_state,
+	.get_xclk = &soc15_get_xclk,
+	.set_uvd_clocks = &soc15_set_uvd_clocks,
+	.set_vce_clocks = &soc15_set_vce_clocks,
+	.get_config_memsize = &soc15_get_config_memsize,
+	.need_full_reset = &soc15_need_full_reset,
+	.init_doorbell_index = &aqua_vanjaram_doorbell_index_init,
+	.get_pcie_usage = &vega20_get_pcie_usage,
+	.need_reset_on_init = &soc15_need_reset_on_init,
+	.get_pcie_replay_count = &soc15_get_pcie_replay_count,
+	.supports_baco = &soc15_supports_baco,
+	.pre_asic_init = &soc15_pre_asic_init,
+	.query_video_codecs = &soc15_query_video_codecs,
+};
+
 static int soc15_common_early_init(void *handle)
 {
 #define MMIO_REG_HOLE_OFFSET (0x80000 - PAGE_SIZE)
@@ -1096,7 +1118,7 @@ static int soc15_common_early_init(void *handle)
 		adev->external_rev_id = adev->rev_id + 0x3c;
 		break;
 	case IP_VERSION(9, 4, 3):
-		adev->asic_funcs = &vega20_asic_funcs;
+		adev->asic_funcs = &aqua_vanjaram_asic_funcs;
 		adev->cg_flags =
 			AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG;
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index d8a2a6c2c6e3f..9cc2dda087c4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -115,4 +115,5 @@ void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev);
 
 void vega10_doorbell_index_init(struct amdgpu_device *adev);
 void vega20_doorbell_index_init(struct amdgpu_device *adev);
+void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev);
 #endif
-- 
GitLab


From fa9e78d14070c55a47dc092f6d18364680b83f44 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 31 Aug 2022 17:11:59 +0800
Subject: [PATCH 1117/1791] drm/amdgpu: update ip discovery header to v4

version 4 supports 64bit ip base address

Signed-off-by: Le Ma <le.ma@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/include/discovery.h | 30 ++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/include/discovery.h b/drivers/gpu/drm/amd/include/discovery.h
index f150404ffc682..9181e57887db7 100644
--- a/drivers/gpu/drm/amd/include/discovery.h
+++ b/drivers/gpu/drm/amd/include/discovery.h
@@ -79,7 +79,14 @@ typedef struct ip_discovery_header
 	uint32_t id;           /* Table ID */
 	uint16_t num_dies;     /* Number of Dies */
 	die_info die_info[16]; /* list die information for up to 16 dies */
-	uint16_t padding[1];   /* padding */
+	union {
+		uint16_t padding[1];	/* version <= 3 */
+		struct {		/* version == 4 */
+			uint8_t base_addr_64_bit : 1; /* ip structures are using 64 bit base address */
+			uint8_t reserved : 7;
+			uint8_t reserved2;
+		};
+	};
 } ip_discovery_header;
 
 typedef struct ip
@@ -118,6 +125,26 @@ typedef struct ip_v3
 	uint32_t base_address[1];               /* Base Address list. Corresponds to the num_base_address field*/
 } ip_v3;
 
+typedef struct ip_v4 {
+	uint16_t hw_id;                         /* Hardware ID */
+	uint8_t instance_number;                /* Instance number for the IP */
+	uint8_t num_base_address;               /* Number of base addresses*/
+	uint8_t major;                          /* Hardware ID.major version */
+	uint8_t minor;                          /* Hardware ID.minor version */
+	uint8_t revision;                       /* Hardware ID.revision version */
+#if defined(LITTLEENDIAN_CPU)
+	uint8_t sub_revision : 4;               /* HCID Sub-Revision */
+	uint8_t variant : 4;                    /* HW variant */
+#elif defined(BIGENDIAN_CPU)
+	uint8_t variant : 4;                    /* HW variant */
+	uint8_t sub_revision : 4;               /* HCID Sub-Revision */
+#endif
+	union {
+		uint32_t base_address[0];               /* 32-bit Base Address list. Corresponds to the num_base_address field*/
+		uint64_t base_address_64[0];            /* 64-bit Base Address list. Corresponds to the num_base_address field*/
+	} __packed;
+} ip_v4;
+
 typedef struct die_header
 {
 	uint16_t die_id;
@@ -134,6 +161,7 @@ typedef struct ip_structure
 		{
 			ip *ip_list;
 			ip_v3 *ip_v3_list;
+			ip_v4 *ip_v4_list;
 		};                                  /* IP list. Variable size*/
 	} die;
 } ip_structure;
-- 
GitLab


From c3729997a23e8955f017b6286bd6c73b386fbe49 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 6 Sep 2022 15:22:44 +0800
Subject: [PATCH 1118/1791] drm/amdgpu: increase DISCOVERY_TMR_SIZE

New ip_discovery binary size is increased.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
index 63ec6924b907a..3a2f347bd50d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h
@@ -24,7 +24,7 @@
 #ifndef __AMDGPU_DISCOVERY__
 #define __AMDGPU_DISCOVERY__
 
-#define DISCOVERY_TMR_SIZE      (4 << 10)
+#define DISCOVERY_TMR_SIZE      (8 << 10)
 #define DISCOVERY_TMR_OFFSET    (64 << 10)
 
 void amdgpu_discovery_fini(struct amdgpu_device *adev);
-- 
GitLab


From 7e0eebdc4745b7e0f031df571621204fd9d6288e Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 14 Sep 2022 15:07:23 +0800
Subject: [PATCH 1119/1791] drm/amdgpu: extend max instances

Number of instances is extended.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 32112db5c841e..e17a3697a23f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -661,7 +661,7 @@ enum amd_hw_ip_block_type {
 	MAX_HWIP
 };
 
-#define HWIP_MAX_INSTANCE	28
+#define HWIP_MAX_INSTANCE	44
 
 #define HW_ID_MAX		300
 #define IP_VERSION(mj, mn, rv) (((mj) << 16) | ((mn) << 8) | (rv))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 1024a06359ca6..6ea3f076257e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -32,7 +32,7 @@
 #define AMDGPU_VCN_FIRMWARE_OFFSET	256
 #define AMDGPU_VCN_MAX_ENC_RINGS	3
 
-#define AMDGPU_MAX_VCN_INSTANCES	2
+#define AMDGPU_MAX_VCN_INSTANCES	4
 #define AMDGPU_MAX_VCN_ENC_RINGS  AMDGPU_VCN_MAX_ENC_RINGS * AMDGPU_MAX_VCN_INSTANCES
 
 #define AMDGPU_VCN_HARVEST_VCN0 (1 << 0)
-- 
GitLab


From 3955b14134285f99e0763ba74b8b1b35faed37ed Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 20 Sep 2022 15:14:48 +0800
Subject: [PATCH 1120/1791] drm/amdgpu: do some register access cleanup in nbio
 v7_9

Use WREG_SOC15x() instead of WREG32(SOC15_REG_OFFSET())

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 5dbab8ab9b274..b8e4afb8c7914 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -98,7 +98,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 
 	switch (dev_inst % adev->sdma.num_inst_per_aid) {
 	case 0:
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1) +
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1,
 			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
@@ -118,7 +118,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 			doorbell_ctrl);
 		break;
 	case 1:
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2) +
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2,
 			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
@@ -138,7 +138,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 			doorbell_ctrl);
 		break;
 	case 2:
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3) +
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3,
 			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
@@ -157,7 +157,7 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 			doorbell_ctrl);
 		break;
 	case 3:
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4) +
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4,
 			4 * aid_id, doorbell_range);
 
 		doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl,
@@ -219,7 +219,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4);
 
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17) +
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17,
 					aid_id, doorbell_range);
 		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL) * 4
 				+ AMDGPU_SMN_TARGET_AID(aid_id)
@@ -233,7 +233,8 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_RANGE_SIZE, 0);
 
-		WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, doorbell_range);
+		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17,
+					aid_id, doorbell_range);
 		WREG32(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL),
 						doorbell_ctrl);
 	}
-- 
GitLab


From aabb478421f5ca2be0f1343d02873394e935c582 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 7 Sep 2022 21:17:03 +0800
Subject: [PATCH 1121/1791] drm/amdgpu: upgrade amdgpu_discovery struct ip to
 ip_v4

version 4 supports 64bit ip base address

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 34 +++++++++----------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 915ae0bcdab21..2b573675bc7e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -471,11 +471,11 @@ void amdgpu_discovery_fini(struct amdgpu_device *adev)
 	adev->mman.discovery_bin = NULL;
 }
 
-static int amdgpu_discovery_validate_ip(const struct ip *ip)
+static int amdgpu_discovery_validate_ip(const struct ip_v4 *ip)
 {
-	if (ip->number_instance >= HWIP_MAX_INSTANCE) {
-		DRM_ERROR("Unexpected number_instance (%d) from ip discovery blob\n",
-			  ip->number_instance);
+	if (ip->instance_number >= HWIP_MAX_INSTANCE) {
+		DRM_ERROR("Unexpected instance_number (%d) from ip discovery blob\n",
+			  ip->instance_number);
 		return -EINVAL;
 	}
 	if (le16_to_cpu(ip->hw_id) >= HW_ID_MAX) {
@@ -493,7 +493,7 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev,
 	struct binary_header *bhdr;
 	struct ip_discovery_header *ihdr;
 	struct die_header *dhdr;
-	struct ip *ip;
+	struct ip_v4 *ip;
 	uint16_t die_offset, ip_offset, num_dies, num_ips;
 	int i, j;
 
@@ -510,16 +510,16 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev,
 		ip_offset = die_offset + sizeof(*dhdr);
 
 		for (j = 0; j < num_ips; j++) {
-			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
+			ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset);
 
 			if (amdgpu_discovery_validate_ip(ip))
 				goto next_ip;
 
-			if (le16_to_cpu(ip->harvest) == 1) {
+			if (le16_to_cpu(ip->variant) == 1) {
 				switch (le16_to_cpu(ip->hw_id)) {
 				case VCN_HWID:
 					(*vcn_harvest_count)++;
-					if (ip->number_instance == 0)
+					if (ip->instance_number == 0)
 						adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN0;
 					else
 						adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1;
@@ -852,10 +852,10 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 		size_t ip_offset = _ip_offset;
 
 		for (jj = 0; jj < num_ips; jj++) {
-			struct ip *ip;
+			struct ip_v4 *ip;
 			struct ip_hw_instance *ip_hw_instance;
 
-			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
+			ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset);
 			if (amdgpu_discovery_validate_ip(ip) ||
 			    le16_to_cpu(ip->hw_id) != ii)
 				goto next_ip;
@@ -903,11 +903,11 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 				return -ENOMEM;
 			}
 			ip_hw_instance->hw_id = le16_to_cpu(ip->hw_id); /* == ii */
-			ip_hw_instance->num_instance = ip->number_instance;
+			ip_hw_instance->num_instance = ip->instance_number;
 			ip_hw_instance->major = ip->major;
 			ip_hw_instance->minor = ip->minor;
 			ip_hw_instance->revision = ip->revision;
-			ip_hw_instance->harvest = ip->harvest;
+			ip_hw_instance->harvest = ip->variant;
 			ip_hw_instance->num_base_addresses = ip->num_base_address;
 
 			for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++)
@@ -1082,7 +1082,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 	struct binary_header *bhdr;
 	struct ip_discovery_header *ihdr;
 	struct die_header *dhdr;
-	struct ip *ip;
+	struct ip_v4 *ip;
 	uint16_t die_offset;
 	uint16_t ip_offset;
 	uint16_t num_dies;
@@ -1121,7 +1121,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				le16_to_cpu(dhdr->die_id), num_ips);
 
 		for (j = 0; j < num_ips; j++) {
-			ip = (struct ip *)(adev->mman.discovery_bin + ip_offset);
+			ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset);
 
 			if (amdgpu_discovery_validate_ip(ip))
 				goto next_ip;
@@ -1131,7 +1131,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 			DRM_DEBUG("%s(%d) #%d v%d.%d.%d:\n",
 				  hw_id_names[le16_to_cpu(ip->hw_id)],
 				  le16_to_cpu(ip->hw_id),
-				  ip->number_instance,
+				  ip->instance_number,
 				  ip->major, ip->minor,
 				  ip->revision);
 
@@ -1182,7 +1182,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id)) {
 					DRM_DEBUG("set register base offset for %s\n",
 							hw_id_names[le16_to_cpu(ip->hw_id)]);
-					adev->reg_offset[hw_ip][ip->number_instance] =
+					adev->reg_offset[hw_ip][ip->instance_number] =
 						ip->base_address;
 					/* Instance support is somewhat inconsistent.
 					 * SDMA is a good example.  Sienna cichlid has 4 total
@@ -1193,7 +1193,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 					 * example.  On most chips there are multiple instances
 					 * with the same HWID.
 					 */
-					adev->ip_versions[hw_ip][ip->number_instance] =
+					adev->ip_versions[hw_ip][ip->instance_number] =
 						IP_VERSION(ip->major, ip->minor, ip->revision);
 				}
 			}
-- 
GitLab


From 7d158f52bfbf82279ea8f1c3dc7a3ab3065a76e4 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 27 Sep 2022 17:26:27 +0800
Subject: [PATCH 1122/1791] drm/amdgpu: parse base address from new ip
 discovery with 64bit ip base address

Truncate the 64bit base address from ip discovery and only store lower 32bit
ip base in reg_offset[].

Bits > 32 follows ASIC specific format, thus just discard them and handle it
within specific ASIC.

By this way reg_offset[] and related helpers can stay unchanged.

v2: make comments more generic

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 49 +++++++++++++++----
 1 file changed, 40 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 2b573675bc7e0..71293db09009f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -532,7 +532,10 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev,
 				}
 			}
 next_ip:
-			ip_offset += struct_size(ip, base_address, ip->num_base_address);
+			if (ihdr->base_addr_64_bit)
+				ip_offset += struct_size(ip, base_address_64, ip->num_base_address);
+			else
+				ip_offset += struct_size(ip, base_address, ip->num_base_address);
 		}
 	}
 }
@@ -838,7 +841,8 @@ static void ip_disc_release(struct kobject *kobj)
 
 static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 				      struct ip_die_entry *ip_die_entry,
-				      const size_t _ip_offset, const int num_ips)
+				      const size_t _ip_offset, const int num_ips,
+				      bool reg_base_64)
 {
 	int ii, jj, kk, res;
 
@@ -910,15 +914,25 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 			ip_hw_instance->harvest = ip->variant;
 			ip_hw_instance->num_base_addresses = ip->num_base_address;
 
-			for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++)
-				ip_hw_instance->base_addr[kk] = ip->base_address[kk];
+			for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++) {
+				if (reg_base_64)
+					ip_hw_instance->base_addr[kk] =
+						lower_32_bits(le64_to_cpu(ip->base_address_64[kk])) & 0x3FFFFFFF;
+				else
+					ip_hw_instance->base_addr[kk] = ip->base_address[kk];
+			}
 
 			kobject_init(&ip_hw_instance->kobj, &ip_hw_instance_ktype);
 			ip_hw_instance->kobj.kset = &ip_hw_id->hw_id_kset;
 			res = kobject_add(&ip_hw_instance->kobj, NULL,
 					  "%d", ip_hw_instance->num_instance);
 next_ip:
-			ip_offset += struct_size(ip, base_address, ip->num_base_address);
+			if (reg_base_64)
+				ip_offset += struct_size(ip, base_address_64,
+							 ip->num_base_address);
+			else
+				ip_offset += struct_size(ip, base_address,
+							 ip->num_base_address);
 		}
 	}
 
@@ -972,7 +986,7 @@ static int amdgpu_discovery_sysfs_recurse(struct amdgpu_device *adev)
 			return res;
 		}
 
-		amdgpu_discovery_sysfs_ips(adev, ip_die_entry, ip_offset, num_ips);
+		amdgpu_discovery_sysfs_ips(adev, ip_die_entry, ip_offset, num_ips, !!ihdr->base_addr_64_bit);
 	}
 
 	return 0;
@@ -1174,12 +1188,26 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				 * convert the endianness of base addresses in place,
 				 * so that we don't need to convert them when accessing adev->reg_offset.
 				 */
-				ip->base_address[k] = le32_to_cpu(ip->base_address[k]);
+				if (ihdr->base_addr_64_bit)
+					/* Truncate the 64bit base address from ip discovery
+					 * and only store lower 32bit ip base in reg_offset[].
+					 * Bits > 32 follows ASIC specific format, thus just
+					 * discard them and handle it within specific ASIC.
+					 * By this way reg_offset[] and related helpers can
+					 * stay unchanged.
+					 * The base address is in dwords, thus clear the
+					 * highest 2 bits to store.
+					 */
+					ip->base_address[k] =
+						lower_32_bits(le64_to_cpu(ip->base_address_64[k])) & 0x3FFFFFFF;
+				else
+					ip->base_address[k] = le32_to_cpu(ip->base_address[k]);
 				DRM_DEBUG("\t0x%08x\n", ip->base_address[k]);
 			}
 
 			for (hw_ip = 0; hw_ip < MAX_HWIP; hw_ip++) {
-				if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id)) {
+				if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id) &&
+				    hw_id_map[hw_ip] != 0) {
 					DRM_DEBUG("set register base offset for %s\n",
 							hw_id_names[le16_to_cpu(ip->hw_id)]);
 					adev->reg_offset[hw_ip][ip->instance_number] =
@@ -1199,7 +1227,10 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 			}
 
 next_ip:
-			ip_offset += struct_size(ip, base_address, ip->num_base_address);
+			if (ihdr->base_addr_64_bit)
+				ip_offset += struct_size(ip, base_address_64, ip->num_base_address);
+			else
+				ip_offset += struct_size(ip, base_address, ip->num_base_address);
 		}
 	}
 
-- 
GitLab


From 2fa480d36eb302712e48dce4d2f6564b24426be3 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 27 Sep 2022 17:51:33 +0800
Subject: [PATCH 1123/1791] drm/amdgpu: add helpers to access registers on
 different AIDs

SMN address which is larger than 32bit has different indications
through bit[34:32] on different AIDs.

v2: put smn addressing of different AIDs into asic specific place
v3: change to ext_id/ext_offset naming

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h           |  2 ++
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 19 +++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/soc15.c            |  1 +
 drivers/gpu/drm/amd/amdgpu/soc15.h            |  1 +
 drivers/gpu/drm/amd/amdgpu/soc15_common.h     | 10 ++++++++++
 5 files changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index e17a3697a23f2..4c4ce33c83598 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -574,6 +574,8 @@ struct amdgpu_asic_funcs {
 	/* query video codecs */
 	int (*query_video_codecs)(struct amdgpu_device *adev, bool encode,
 				  const struct amdgpu_video_codecs **codecs);
+	/* encode "> 32bits" smn addressing */
+	u64 (*encode_ext_smn_addressing)(int ext_id);
 };
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 12379c374457f..2616bdb40418e 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -100,3 +100,22 @@ void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 
 	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
 }
+
+/* Fixed pattern for smn addressing on different AIDs:
+ *   bit[34]: indicate cross AID access
+ *   bit[33:32]: indicate target AID id
+ * AID id range is 0 ~ 3 as maximum AID number is 4.
+ */
+u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id)
+{
+	u64 ext_offset;
+
+	/* local routing and bit[34:32] will be zeros */
+	if (ext_id == 0)
+		return 0;
+
+	/* Initiated from host, accessing to all non-zero aids are cross traffic */
+	ext_offset = ((u64)(ext_id & 0x3) << 32) | (1ULL << 34);
+
+	return ext_offset;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 29d2e08834c78..206013d276c1a 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -873,6 +873,7 @@ static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs =
 	.supports_baco = &soc15_supports_baco,
 	.pre_asic_init = &soc15_pre_asic_init,
 	.query_video_codecs = &soc15_query_video_codecs,
+	.encode_ext_smn_addressing = &aqua_vanjaram_encode_ext_smn_addressing,
 };
 
 static int soc15_common_early_init(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index 9cc2dda087c4d..dd48db09aa511 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -112,6 +112,7 @@ int vega20_reg_base_init(struct amdgpu_device *adev);
 int arct_reg_base_init(struct amdgpu_device *adev);
 int aldebaran_reg_base_init(struct amdgpu_device *adev);
 void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev);
+u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id);
 
 void vega10_doorbell_index_init(struct amdgpu_device *adev);
 void vega20_doorbell_index_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index 39e4406da4aee..1c9e924b5f8cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -188,4 +188,14 @@
 #define RREG32_SOC15_OFFSET_RLC(ip, inst, reg, offset) \
 	__RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, AMDGPU_REGS_RLC, ip##_HWIP)
 
+/* inst equals to ext for some IPs */
+#define RREG32_SOC15_EXT(ip, inst, reg, ext) \
+	RREG32_PCIE_EXT((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) * 4 \
+			+ adev->asic_funcs->encode_ext_smn_addressing(ext)) \
+
+#define WREG32_SOC15_EXT(ip, inst, reg, ext, value) \
+	WREG32_PCIE_EXT((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) * 4 \
+			+ adev->asic_funcs->encode_ext_smn_addressing(ext), \
+			value) \
+
 #endif
-- 
GitLab


From 369576c2d5a8e8c3d7efb9d598970ba3f006b07e Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 22 Sep 2022 15:46:54 +0800
Subject: [PATCH 1124/1791] drm/amdgpu: consolidate the access helpers in nbio
 v7_9

Use WREG32_SOC15_EXT to write registers with address larger than 32bit.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 45 +++++++-------------------
 1 file changed, 12 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index b8e4afb8c7914..78eab4d48e383 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -62,15 +62,6 @@ static u32 nbio_v7_9_get_memsize(struct amdgpu_device *adev)
 	return RREG32_SOC15(NBIO, 0, regRCC_DEV0_EPF0_RCC_CONFIG_MEMSIZE);
 }
 
-#define S2A_DOORBELL_REG_LSD_OFFSET	0x40
-
-/* Temporarily add 2 macros below. Range is 0 ~ 3 as total AID number is 4.
- * They will be obsoleted after the latest ip offset header
- * is imported in driver in near future.
- */
-#define AMDGPU_SMN_TARGET_AID(x)	((u64)(x) << 32)
-#define AMDGPU_SMN_CROSS_AID		(1ULL << 34)
-
 static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instance,
 			bool use_doorbell, int doorbell_index, int doorbell_size)
 {
@@ -111,11 +102,8 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x1);
-		WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_1_CTRL)
-			+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
-			+ AMDGPU_SMN_TARGET_AID(aid_id)
-			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-			doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_1_CTRL,
+			aid_id, doorbell_ctrl);
 		break;
 	case 1:
 		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2,
@@ -131,11 +119,8 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x2);
-		WREG32_PCIE_EXT((SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_2_CTRL)
-			+ S2A_DOORBELL_REG_LSD_OFFSET) * 4
-			+ AMDGPU_SMN_TARGET_AID(aid_id)
-			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-			doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_2_CTRL,
+			aid_id, doorbell_ctrl);
 		break;
 	case 2:
 		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3,
@@ -151,10 +136,8 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x8);
-		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_5_CTRL) * 4
-			+ AMDGPU_SMN_TARGET_AID(aid_id)
-			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-			doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_5_CTRL,
+			aid_id, doorbell_ctrl);
 		break;
 	case 3:
 		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4,
@@ -170,10 +153,8 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 					S2A_DOORBELL_ENTRY_1_CTRL,
 					S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE,
 					0x9);
-		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_6_CTRL) * 4
-			+ AMDGPU_SMN_TARGET_AID(aid_id)
-			+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-			doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_6_CTRL,
+			aid_id, doorbell_ctrl);
 		break;
 	default:
 		break;
@@ -221,10 +202,8 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 
 		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17,
 					aid_id, doorbell_range);
-		WREG32_PCIE_EXT(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL) * 4
-				+ AMDGPU_SMN_TARGET_AID(aid_id)
-				+ AMDGPU_SMN_CROSS_AID * !!aid_id,
-				doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_4_CTRL,
+				aid_id, doorbell_ctrl);
 	} else {
 		doorbell_range = REG_SET_FIELD(doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
@@ -235,8 +214,8 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do
 
 		WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17,
 					aid_id, doorbell_range);
-		WREG32(SOC15_REG_OFFSET(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL),
-						doorbell_ctrl);
+		WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_4_CTRL,
+				aid_id, doorbell_ctrl);
 	}
 }
 
-- 
GitLab


From 870d1e5afca58261a147e9080abb8cc75fccb849 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 14 Oct 2022 13:06:18 +0530
Subject: [PATCH 1125/1791] drm/amdgpu: Fix interrupt handling in GFX v9.4.3

IH follows a different identification scheme for its clients. Get the
right mapping of xcc instance from IH node id.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 35 +++++++++++++++++++++----
 1 file changed, 30 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index f1e6da4d62a34..c5ffcd70ec7ba 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2730,11 +2730,24 @@ static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node)
+{
+	int xcc;
+
+	xcc = hweight8(adev->gfx.xcc_mask & GENMASK(ih_node / 2, 0));
+	if (!xcc) {
+		dev_err(adev->dev, "Couldn't find xcc mapping from IH node");
+		return -EINVAL;
+	}
+
+	return xcc - 1;
+}
+
 static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 			    struct amdgpu_irq_src *source,
 			    struct amdgpu_iv_entry *entry)
 {
-	int i, phys_id;
+	int i, xcc_id;
 	u8 me_id, pipe_id, queue_id;
 	struct amdgpu_ring *ring;
 
@@ -2743,14 +2756,19 @@ static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 	pipe_id = (entry->ring_id & 0x03) >> 0;
 	queue_id = (entry->ring_id & 0x70) >> 4;
 
-	phys_id = node_id_to_phys_map[entry->node_id];
+	xcc_id = gfx_v9_4_3_ih_to_xcc_inst(adev, entry->node_id);
+
+	if (xcc_id == -EINVAL)
+		return -EINVAL;
 
 	switch (me_id) {
 	case 0:
 	case 1:
 	case 2:
 		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			ring = &adev->gfx.compute_ring[i + phys_id * adev->gfx.num_compute_rings];
+			ring = &adev->gfx.compute_ring
+					[i +
+					 xcc_id * adev->gfx.num_compute_rings];
 			/* Per-queue interrupt is supported for MEC starting from VI.
 			  * The interrupt can only be enabled/disabled per pipe instead of per queue.
 			  */
@@ -2768,18 +2786,25 @@ static void gfx_v9_4_3_fault(struct amdgpu_device *adev,
 {
 	u8 me_id, pipe_id, queue_id;
 	struct amdgpu_ring *ring;
-	int i;
+	int i, xcc_id;
 
 	me_id = (entry->ring_id & 0x0c) >> 2;
 	pipe_id = (entry->ring_id & 0x03) >> 0;
 	queue_id = (entry->ring_id & 0x70) >> 4;
 
+	xcc_id = gfx_v9_4_3_ih_to_xcc_inst(adev, entry->node_id);
+
+	if (xcc_id == -EINVAL)
+		return;
+
 	switch (me_id) {
 	case 0:
 	case 1:
 	case 2:
 		for (i = 0; i < adev->gfx.num_compute_rings; i++) {
-			ring = &adev->gfx.compute_ring[i];
+			ring = &adev->gfx.compute_ring
+					[i +
+					 xcc_id * adev->gfx.num_compute_rings];
 			if (ring->me == me_id && ring->pipe == pipe_id &&
 			    ring->queue == queue_id)
 				drm_sched_fault(&ring->sched);
-- 
GitLab


From 57a83b2dafff055698190d3fe3d7197c969c2dcc Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 23 Oct 2022 20:27:59 +0800
Subject: [PATCH 1126/1791] drm/amdgpu: init gfx_v9_4_3 external_rev_id

it is used for user space driver to identify gfx_v9_4_3 chip

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 206013d276c1a..06a18b2f6e041 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1127,6 +1127,7 @@ static int soc15_common_early_init(void *handle)
 			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_VCN_DPG |
 			AMD_PG_SUPPORT_JPEG;
+		adev->external_rev_id = adev->rev_id + 0x46;
 		break;
 	default:
 		/* FIXME: not supported yet */
-- 
GitLab


From 5f09237b82e51be8a1849eb658dcb153748f8f60 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 31 Oct 2022 10:46:05 +0530
Subject: [PATCH 1127/1791] drm/amdgpu: Add SDMA v4.4.2 golden settings

Add programming of SDMA golden settings for v4.4.2

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 4350939105c51..016813b295ac0 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -96,11 +96,22 @@ static int sdma_v4_4_2_irq_id_to_seq(unsigned client_id)
 
 static void sdma_v4_4_2_init_golden_registers(struct amdgpu_device *adev)
 {
-	switch (adev->ip_versions[SDMA0_HWIP][0]) {
-	case IP_VERSION(4, 4, 2):
-		break;
-	default:
-		break;
+	u32 val;
+	int i;
+
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		val = RREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG);
+		val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG, NUM_BANKS, 4);
+		val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG,
+				    PIPE_INTERLEAVE_SIZE, 0);
+		WREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG, val);
+
+		val = RREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG_READ);
+		val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG_READ, NUM_BANKS,
+				    4);
+		val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG_READ,
+				    PIPE_INTERLEAVE_SIZE, 0);
+		WREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG_READ, val);
 	}
 }
 
-- 
GitLab


From 553f973a0d7bbe95ea5da46979d926a9c0ada109 Mon Sep 17 00:00:00 2001
From: Tom St Denis <tom.stdenis@amd.com>
Date: Tue, 11 Oct 2022 09:52:58 -0400
Subject: [PATCH 1128/1791] drm/amd/amdgpu: Update debugfs for XCC support (v3)

This patch updates the 'regs2' interface for MMIO
registers to add a new IOCTL command for a 'v2' state
data that includes the XCC ID.

This patch then updates amdgpu_gfx_select_se_sh()
and amdgpu_gfx_select_me_pipe_q() (and the implementations
in the gfx drivers) to support an additional parameter.

This patch then creates a new debugfs interface "gprwave"
which is a merge of shader GPR and wave status access.  This
new inteface uses an IOCTL to select banks as well as XCC identity.

(v2) Fix missing xcc_id in wave_ind function

(v3) Fix pm runtime calls and mutex locking

(v4) Fix bad label

Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 202 ++++++++++++++++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h     |  10 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h     |  36 +++-
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c      |   8 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c      |   8 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c       |   6 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c       |   6 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c       |   6 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c       |   8 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c     |  60 +++---
 10 files changed, 275 insertions(+), 75 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index df94cd2c4b39f..e94d0cf3f7936 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -139,7 +139,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 					sh_bank, instance_bank, 0);
 	} else if (use_ring) {
 		mutex_lock(&adev->srbm_mutex);
-		amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid);
+		amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid, 0);
 	}
 
 	if (pm_pg_lock)
@@ -172,7 +172,7 @@ static int  amdgpu_debugfs_process_reg_op(bool read, struct file *f,
 		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
 		mutex_unlock(&adev->grbm_idx_mutex);
 	} else if (use_ring) {
-		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0);
+		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	}
 
@@ -263,14 +263,14 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off
 		}
 		mutex_lock(&adev->grbm_idx_mutex);
 		amdgpu_gfx_select_se_sh(adev, rd->id.grbm.se,
-								rd->id.grbm.sh,
-								rd->id.grbm.instance, 0);
+						  rd->id.grbm.sh,
+						  rd->id.grbm.instance, rd->id.xcc_id);
 	}
 
 	if (rd->id.use_srbm) {
 		mutex_lock(&adev->srbm_mutex);
 		amdgpu_gfx_select_me_pipe_q(adev, rd->id.srbm.me, rd->id.srbm.pipe,
-									rd->id.srbm.queue, rd->id.srbm.vmid);
+					    rd->id.srbm.queue, rd->id.srbm.vmid, rd->id.xcc_id);
 	}
 
 	if (rd->id.pg_lock)
@@ -296,12 +296,12 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off
 	}
 end:
 	if (rd->id.use_grbm) {
-		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
+		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, rd->id.xcc_id);
 		mutex_unlock(&adev->grbm_idx_mutex);
 	}
 
 	if (rd->id.use_srbm) {
-		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0);
+		amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0, rd->id.xcc_id);
 		mutex_unlock(&adev->srbm_mutex);
 	}
 
@@ -320,19 +320,45 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off
 static long amdgpu_debugfs_regs2_ioctl(struct file *f, unsigned int cmd, unsigned long data)
 {
 	struct amdgpu_debugfs_regs2_data *rd = f->private_data;
+	struct amdgpu_debugfs_regs2_iocdata v1_data;
 	int r;
 
+	mutex_lock(&rd->lock);
+
 	switch (cmd) {
-	case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE:
-		mutex_lock(&rd->lock);
-		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata *)data,
+	case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE_V2:
+		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata_v2 *)data,
 				   sizeof(rd->id));
-		mutex_unlock(&rd->lock);
-		return r ? -EINVAL : 0;
+		if (r)
+			r = -EINVAL;
+		goto done;
+	case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE:
+		r = copy_from_user(&v1_data, (struct amdgpu_debugfs_regs2_iocdata *)data,
+				   sizeof(v1_data));
+		if (r) {
+			r = -EINVAL;
+			goto done;
+		}
+		goto v1_copy;
 	default:
-		return -EINVAL;
-	}
-	return 0;
+		r = -EINVAL;
+		goto done;
+	}
+
+v1_copy:
+	rd->id.use_srbm = v1_data.use_srbm;
+	rd->id.use_grbm = v1_data.use_grbm;
+	rd->id.pg_lock = v1_data.pg_lock;
+	rd->id.grbm.se = v1_data.grbm.se;
+	rd->id.grbm.sh = v1_data.grbm.sh;
+	rd->id.grbm.instance = v1_data.grbm.instance;
+	rd->id.srbm.me = v1_data.srbm.me;
+	rd->id.srbm.pipe = v1_data.srbm.pipe;
+	rd->id.srbm.queue = v1_data.srbm.queue;
+	rd->id.xcc_id = 0;
+done:
+	mutex_unlock(&rd->lock);
+	return r;
 }
 
 static ssize_t amdgpu_debugfs_regs2_read(struct file *f, char __user *buf, size_t size, loff_t *pos)
@@ -345,6 +371,135 @@ static ssize_t amdgpu_debugfs_regs2_write(struct file *f, const char __user *buf
 	return amdgpu_debugfs_regs2_op(f, (char __user *)buf, *pos, size, 1);
 }
 
+static int amdgpu_debugfs_gprwave_open(struct inode *inode, struct file *file)
+{
+	struct amdgpu_debugfs_gprwave_data *rd;
+
+	rd = kzalloc(sizeof *rd, GFP_KERNEL);
+	if (!rd)
+		return -ENOMEM;
+	rd->adev = file_inode(file)->i_private;
+	file->private_data = rd;
+	mutex_init(&rd->lock);
+
+	return 0;
+}
+
+static int amdgpu_debugfs_gprwave_release(struct inode *inode, struct file *file)
+{
+	struct amdgpu_debugfs_gprwave_data *rd = file->private_data;
+	mutex_destroy(&rd->lock);
+	kfree(file->private_data);
+	return 0;
+}
+
+static ssize_t amdgpu_debugfs_gprwave_read(struct file *f, char __user *buf, size_t size, loff_t *pos)
+{
+	struct amdgpu_debugfs_gprwave_data *rd = f->private_data;
+	struct amdgpu_device *adev = rd->adev;
+	ssize_t result = 0;
+	int r;
+	uint32_t *data, x;
+
+	if (size & 0x3 || *pos & 0x3)
+		return -EINVAL;
+
+	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
+	if (r < 0) {
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+		return r;
+	}
+
+	r = amdgpu_virt_enable_access_debugfs(adev);
+	if (r < 0) {
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+		return r;
+	}
+
+	data = kcalloc(1024, sizeof(*data), GFP_KERNEL);
+	if (!data) {
+		pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+		amdgpu_virt_disable_access_debugfs(adev);
+		return -ENOMEM;
+	}
+
+	/* switch to the specific se/sh/cu */
+	mutex_lock(&adev->grbm_idx_mutex);
+	amdgpu_gfx_select_se_sh(adev, rd->id.se, rd->id.sh, rd->id.cu, rd->id.xcc_id);
+
+	if (!rd->id.gpr_or_wave) {
+		x = 0;
+		if (adev->gfx.funcs->read_wave_data)
+			adev->gfx.funcs->read_wave_data(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, data, &x);
+	} else {
+		x = size >> 2;
+		if (rd->id.gpr.vpgr_or_sgpr) {
+			if (adev->gfx.funcs->read_wave_vgprs)
+				adev->gfx.funcs->read_wave_vgprs(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, rd->id.gpr.thread, *pos, size>>2, data);
+		} else {
+			if (adev->gfx.funcs->read_wave_sgprs)
+				adev->gfx.funcs->read_wave_sgprs(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, *pos, size>>2, data);
+		}
+	}
+
+	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, rd->id.xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	pm_runtime_mark_last_busy(adev_to_drm(adev)->dev);
+	pm_runtime_put_autosuspend(adev_to_drm(adev)->dev);
+
+	if (!x) {
+		result = -EINVAL;
+		goto done;
+	}
+
+	while (size && (*pos < x * 4)) {
+		uint32_t value;
+
+		value = data[*pos >> 2];
+		r = put_user(value, (uint32_t *)buf);
+		if (r) {
+			result = r;
+			goto done;
+		}
+
+		result += 4;
+		buf += 4;
+		*pos += 4;
+		size -= 4;
+	}
+
+done:
+	amdgpu_virt_disable_access_debugfs(adev);
+	kfree(data);
+	return result;
+}
+
+static long amdgpu_debugfs_gprwave_ioctl(struct file *f, unsigned int cmd, unsigned long data)
+{
+	struct amdgpu_debugfs_gprwave_data *rd = f->private_data;
+	int r;
+
+	mutex_lock(&rd->lock);
+
+	switch (cmd) {
+	case AMDGPU_DEBUGFS_GPRWAVE_IOC_SET_STATE:
+		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_gprwave_iocdata *)data, sizeof rd->id);
+		if (r)
+			return r ? -EINVAL : 0;
+		goto done;
+	default:
+		r = -EINVAL;
+		goto done;
+	}
+
+done:
+	mutex_unlock(&rd->lock);
+	return r;
+}
+
+
+
 
 /**
  * amdgpu_debugfs_regs_pcie_read - Read from a PCIE register
@@ -913,7 +1068,7 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf,
 
 	x = 0;
 	if (adev->gfx.funcs->read_wave_data)
-		adev->gfx.funcs->read_wave_data(adev, simd, wave, data, &x);
+		adev->gfx.funcs->read_wave_data(adev, 0, simd, wave, data, &x);
 
 	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0);
 	mutex_unlock(&adev->grbm_idx_mutex);
@@ -1007,10 +1162,10 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf,
 
 	if (bank == 0) {
 		if (adev->gfx.funcs->read_wave_vgprs)
-			adev->gfx.funcs->read_wave_vgprs(adev, simd, wave, thread, offset, size>>2, data);
+			adev->gfx.funcs->read_wave_vgprs(adev, 0, simd, wave, thread, offset, size>>2, data);
 	} else {
 		if (adev->gfx.funcs->read_wave_sgprs)
-			adev->gfx.funcs->read_wave_sgprs(adev, simd, wave, offset, size>>2, data);
+			adev->gfx.funcs->read_wave_sgprs(adev, 0, simd, wave, offset, size>>2, data);
 	}
 
 	amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0);
@@ -1341,6 +1496,15 @@ static const struct file_operations amdgpu_debugfs_regs2_fops = {
 	.llseek = default_llseek
 };
 
+static const struct file_operations amdgpu_debugfs_gprwave_fops = {
+	.owner = THIS_MODULE,
+	.unlocked_ioctl = amdgpu_debugfs_gprwave_ioctl,
+	.read = amdgpu_debugfs_gprwave_read,
+	.open = amdgpu_debugfs_gprwave_open,
+	.release = amdgpu_debugfs_gprwave_release,
+	.llseek = default_llseek
+};
+
 static const struct file_operations amdgpu_debugfs_regs_fops = {
 	.owner = THIS_MODULE,
 	.read = amdgpu_debugfs_regs_read,
@@ -1418,6 +1582,7 @@ static const struct file_operations amdgpu_debugfs_gfxoff_residency_fops = {
 static const struct file_operations *debugfs_regs[] = {
 	&amdgpu_debugfs_regs_fops,
 	&amdgpu_debugfs_regs2_fops,
+	&amdgpu_debugfs_gprwave_fops,
 	&amdgpu_debugfs_regs_didt_fops,
 	&amdgpu_debugfs_regs_pcie_fops,
 	&amdgpu_debugfs_regs_smc_fops,
@@ -1434,6 +1599,7 @@ static const struct file_operations *debugfs_regs[] = {
 static const char * const debugfs_regs_names[] = {
 	"amdgpu_regs",
 	"amdgpu_regs2",
+	"amdgpu_gprwave",
 	"amdgpu_regs_didt",
 	"amdgpu_regs_pcie",
 	"amdgpu_regs_smc",
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 93f9875154dbb..2b2d75763875e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -247,16 +247,16 @@ struct amdgpu_gfx_funcs {
 	uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev);
 	void (*select_se_sh)(struct amdgpu_device *adev, u32 se_num,
 			     u32 sh_num, u32 instance, int xcc_id);
-	void (*read_wave_data)(struct amdgpu_device *adev, uint32_t simd,
+	void (*read_wave_data)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 			       uint32_t wave, uint32_t *dst, int *no_fields);
-	void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t simd,
+	void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				uint32_t wave, uint32_t thread, uint32_t start,
 				uint32_t size, uint32_t *dst);
-	void (*read_wave_sgprs)(struct amdgpu_device *adev, uint32_t simd,
+	void (*read_wave_sgprs)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				uint32_t wave, uint32_t start, uint32_t size,
 				uint32_t *dst);
 	void (*select_me_pipe_q)(struct amdgpu_device *adev, u32 me, u32 pipe,
-				 u32 queue, u32 vmid);
+				 u32 queue, u32 vmid, u32 xcc_id);
 	void (*init_spm_golden)(struct amdgpu_device *adev);
 	void (*update_perfmon_mgcg)(struct amdgpu_device *adev, bool enable);
 	int (*get_gfx_shadow_info)(struct amdgpu_device *adev,
@@ -405,7 +405,7 @@ struct amdgpu_gfx {
 
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id)))
-#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid))
+#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid, xcc_id) ((adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid), (xcc_id)))
 #define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev))
 #define amdgpu_gfx_get_gfx_shadow_info(adev, si) ((adev)->gfx.funcs->get_gfx_shadow_info((adev), (si)))
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h
index 919d9d4017509..107f9bb0e24f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h
@@ -35,17 +35,51 @@ struct amdgpu_debugfs_regs2_iocdata {
 	} srbm;
 };
 
+struct amdgpu_debugfs_regs2_iocdata_v2 {
+	__u32 use_srbm, use_grbm, pg_lock;
+	struct {
+		__u32 se, sh, instance;
+	} grbm;
+	struct {
+		__u32 me, pipe, queue, vmid;
+	} srbm;
+	u32 xcc_id;
+};
+
+struct amdgpu_debugfs_gprwave_iocdata {
+	u32 gpr_or_wave, se, sh, cu, wave, simd, xcc_id;
+	struct {
+		u32 thread, vpgr_or_sgpr;
+	} gpr;
+};
+
 /*
  * MMIO debugfs state data (per file* handle)
  */
 struct amdgpu_debugfs_regs2_data {
 	struct amdgpu_device *adev;
 	struct mutex lock;
-	struct amdgpu_debugfs_regs2_iocdata id;
+	struct amdgpu_debugfs_regs2_iocdata_v2 id;
+};
+
+struct amdgpu_debugfs_gprwave_data {
+	struct amdgpu_device *adev;
+	struct mutex lock;
+	struct amdgpu_debugfs_gprwave_iocdata id;
 };
 
 enum AMDGPU_DEBUGFS_REGS2_CMDS {
 	AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE=0,
+	AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE_V2,
+};
+
+enum AMDGPU_DEBUGFS_GPRWAVE_CMDS {
+	AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE=0,
 };
 
+//reg2 interface
 #define AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE _IOWR(0x20, AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE, struct amdgpu_debugfs_regs2_iocdata)
+#define AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE_V2 _IOWR(0x20, AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE_V2, struct amdgpu_debugfs_regs2_iocdata_v2)
+
+//gprwave interface
+#define AMDGPU_DEBUGFS_GPRWAVE_IOC_SET_STATE _IOWR(0x20, AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE, struct amdgpu_debugfs_gprwave_iocdata)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 7b585141e10e0..89158c72753e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4291,7 +4291,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t wave,
 		*(out++) = RREG32_SOC15(GC, 0, mmSQ_IND_DATA);
 }
 
-static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* in gfx10 the SIMD_ID is specified as part of the INSTANCE
 	 * field when performing a select_se_sh so it should be
@@ -4318,7 +4318,7 @@ static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd,
 	dst[(*no_fields)++] = wave_read_ind(adev, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
@@ -4329,7 +4329,7 @@ static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 		dst);
 }
 
-static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				      uint32_t wave, uint32_t thread,
 				      uint32_t start, uint32_t size,
 				      uint32_t *dst)
@@ -4340,7 +4340,7 @@ static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v10_0_select_me_pipe_q(struct amdgpu_device *adev,
-				       u32 me, u32 pipe, u32 q, u32 vm)
+				       u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	nv_grbm_select(adev, me, pipe, q, vm);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 790df2cc3480f..4b7224de879ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -765,7 +765,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t wave,
 		*(out++) = RREG32_SOC15(GC, 0, regSQ_IND_DATA);
 }
 
-static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* in gfx11 the SIMD_ID is specified as part of the INSTANCE
 	 * field when performing a select_se_sh so it should be
@@ -791,7 +791,7 @@ static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd,
 	dst[(*no_fields)++] = wave_read_ind(adev, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
@@ -802,7 +802,7 @@ static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 		dst);
 }
 
-static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				      uint32_t wave, uint32_t thread,
 				      uint32_t start, uint32_t size,
 				      uint32_t *dst)
@@ -813,7 +813,7 @@ static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v11_0_select_me_pipe_q(struct amdgpu_device *adev,
-									  u32 me, u32 pipe, u32 q, u32 vm)
+					u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	soc21_grbm_select(adev, me, pipe, q, vm);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 7cb72bf1acdd3..809558c718e3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -2968,7 +2968,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		*(out++) = RREG32(mmSQ_IND_DATA);
 }
 
-static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* type 0 wave data */
 	dst[(*no_fields)++] = 0;
@@ -2993,7 +2993,7 @@ static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u
 	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
@@ -3003,7 +3003,7 @@ static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v6_0_select_me_pipe_q(struct amdgpu_device *adev,
-				  u32 me, u32 pipe, u32 q, u32 vm)
+				  u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	DRM_INFO("Not implemented\n");
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index d56dda5fc5883..0f0c12bbe2285 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -4112,7 +4112,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		*(out++) = RREG32(mmSQ_IND_DATA);
 }
 
-static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* type 0 wave data */
 	dst[(*no_fields)++] = 0;
@@ -4137,7 +4137,7 @@ static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u
 	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
@@ -4147,7 +4147,7 @@ static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v7_0_select_me_pipe_q(struct amdgpu_device *adev,
-				  u32 me, u32 pipe, u32 q, u32 vm)
+				  u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	cik_srbm_select(adev, me, pipe, q, vm);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 2ae7f167985fe..6d0589dc1d6cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -3419,7 +3419,7 @@ static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev,
 }
 
 static void gfx_v8_0_select_me_pipe_q(struct amdgpu_device *adev,
-				  u32 me, u32 pipe, u32 q, u32 vm)
+				  u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	vi_srbm_select(adev, me, pipe, q, vm);
 }
@@ -5217,7 +5217,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		*(out++) = RREG32(mmSQ_IND_DATA);
 }
 
-static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* type 0 wave data */
 	dst[(*no_fields)++] = 0;
@@ -5242,7 +5242,7 @@ static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u
 	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v8_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v8_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index da69177dc76f5..cc005e3bcd400 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1788,7 +1788,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		*(out++) = RREG32_SOC15(GC, 0, mmSQ_IND_DATA);
 }
 
-static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
+static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields)
 {
 	/* type 1 wave data */
 	dst[(*no_fields)++] = 1;
@@ -1809,7 +1809,7 @@ static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u
 	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE);
 }
 
-static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t start,
 				     uint32_t size, uint32_t *dst)
 {
@@ -1818,7 +1818,7 @@ static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
 		start + SQIND_WAVE_SGPRS_OFFSET, size, dst);
 }
 
-static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				     uint32_t wave, uint32_t thread,
 				     uint32_t start, uint32_t size,
 				     uint32_t *dst)
@@ -1829,7 +1829,7 @@ static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v9_0_select_me_pipe_q(struct amdgpu_device *adev,
-				  u32 me, u32 pipe, u32 q, u32 vm)
+				  u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
 	soc15_grbm_select(adev, me, pipe, q, vm, 0);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index c5ffcd70ec7ba..76e3571ec5c8d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -536,21 +536,21 @@ static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
 	WREG32_SOC15_RLC_SHADOW_EX(reg, GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX, data);
 }
 
-static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address)
+static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t address)
 {
-	WREG32_SOC15_RLC(GC, GET_INST(GC, 0), regSQ_IND_INDEX,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX,
 		(wave << SQ_IND_INDEX__WAVE_ID__SHIFT) |
 		(simd << SQ_IND_INDEX__SIMD_ID__SHIFT) |
 		(address << SQ_IND_INDEX__INDEX__SHIFT) |
 		(SQ_IND_INDEX__FORCE_READ_MASK));
-	return RREG32_SOC15(GC, GET_INST(GC, 0), regSQ_IND_DATA);
+	return RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA);
 }
 
-static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
+static void wave_read_regs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 			   uint32_t wave, uint32_t thread,
 			   uint32_t regno, uint32_t num, uint32_t *out)
 {
-	WREG32_SOC15_RLC(GC, GET_INST(GC, 0), regSQ_IND_INDEX,
+	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX,
 		(wave << SQ_IND_INDEX__WAVE_ID__SHIFT) |
 		(simd << SQ_IND_INDEX__SIMD_ID__SHIFT) |
 		(regno << SQ_IND_INDEX__INDEX__SHIFT) |
@@ -558,53 +558,53 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd,
 		(SQ_IND_INDEX__FORCE_READ_MASK) |
 		(SQ_IND_INDEX__AUTO_INCR_MASK));
 	while (num--)
-		*(out++) = RREG32_SOC15(GC, GET_INST(GC, 0), regSQ_IND_DATA);
+		*(out++) = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA);
 }
 
 static void gfx_v9_4_3_read_wave_data(struct amdgpu_device *adev,
-				      uint32_t simd, uint32_t wave,
+				      uint32_t xcc_id, uint32_t simd, uint32_t wave,
 				      uint32_t *dst, int *no_fields)
 {
 	/* type 1 wave data */
 	dst[(*no_fields)++] = 1;
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_STATUS);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_PC_LO);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_PC_HI);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_EXEC_LO);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_EXEC_HI);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_HW_ID);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_INST_DW0);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_INST_DW1);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_GPR_ALLOC);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_LDS_ALLOC);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_TRAPSTS);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_IB_STS);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_IB_DBG0);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_M0);
-	dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE);
-}
-
-static void gfx_v9_4_3_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd,
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_STATUS);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_LO);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_HI);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_LO);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_HI);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_HW_ID);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW0);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW1);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_GPR_ALLOC);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_LDS_ALLOC);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_TRAPSTS);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_STS);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_DBG0);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_M0);
+	dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_MODE);
+}
+
+static void gfx_v9_4_3_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				       uint32_t wave, uint32_t start,
 				       uint32_t size, uint32_t *dst)
 {
-	wave_read_regs(adev, simd, wave, 0,
+	wave_read_regs(adev, xcc_id, simd, wave, 0,
 		       start + SQIND_WAVE_SGPRS_OFFSET, size, dst);
 }
 
-static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
+static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd,
 				       uint32_t wave, uint32_t thread,
 				       uint32_t start, uint32_t size,
 				       uint32_t *dst)
 {
-	wave_read_regs(adev, simd, wave, thread,
+	wave_read_regs(adev, xcc_id, simd, wave, thread,
 		       start + SQIND_WAVE_VGPRS_OFFSET, size, dst);
 }
 
 static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
-					u32 me, u32 pipe, u32 q, u32 vm)
+					u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id)
 {
-	soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, 0));
+	soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, xcc_id));
 }
 
 static enum amdgpu_gfx_partition
-- 
GitLab


From ebadc1061e045e961339e0df7b8a07f3e589579c Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 7 Nov 2022 11:11:29 +0800
Subject: [PATCH 1129/1791] drm/amdgpu: retire render backend setup from
 gfx_v9_4_3

gfx v9_4_3 only support compute. render backend
doesn't need to be involved in any compute shader
execution.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 44 -------------------------
 1 file changed, 44 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 76e3571ec5c8d..6ed97371ff1e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -888,46 +888,6 @@ static int gfx_v9_4_3_sw_fini(void *handle)
 	return 0;
 }
 
-static u32 gfx_v9_4_3_get_rb_active_bitmap(struct amdgpu_device *adev)
-{
-	u32 data, mask;
-
-	data = RREG32_SOC15(GC, GET_INST(GC, 0), regCC_RB_BACKEND_DISABLE);
-	data |= RREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_RB_BACKEND_DISABLE);
-
-	data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK;
-	data >>= GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT;
-
-	mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_backends_per_se /
-					 adev->gfx.config.max_sh_per_se);
-
-	return (~data) & mask;
-}
-
-static void gfx_v9_4_3_setup_rb(struct amdgpu_device *adev, int xcc_id)
-{
-	int i, j;
-	u32 data;
-	u32 active_rbs = 0;
-	u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se /
-					adev->gfx.config.max_sh_per_se;
-
-	mutex_lock(&adev->grbm_idx_mutex);
-	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
-		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, xcc_id);
-			data = gfx_v9_4_3_get_rb_active_bitmap(adev);
-			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
-					       rb_bitmap_width_per_sh);
-		}
-	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id);
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	adev->gfx.config.backend_enable_mask = active_rbs;
-	adev->gfx.config.num_rbs = hweight32(active_rbs);
-}
-
 #define DEFAULT_SH_MEM_BASES	(0x6000)
 static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev, int xcc_id)
 {
@@ -991,10 +951,6 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	int i, j, num_xcc;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_CNTL, READ_TIMEOUT, 0xff);
-		gfx_v9_4_3_setup_rb(adev, i);
-	}
 
 	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
 	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2);
-- 
GitLab


From 63121b11a95ccd30763e6def363f8fbe992b7a3f Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 24 Apr 2022 15:23:08 +0800
Subject: [PATCH 1130/1791] drm/amdgpu: add smuio v13_0_3 ip headers

Add smuio v13_0_3 register offset and shift masks
header files

v2: update headers (Alex)

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../asic_reg/smuio/smuio_13_0_3_offset.h      | 177 ++++++++
 .../asic_reg/smuio/smuio_13_0_3_sh_mask.h     | 428 ++++++++++++++++++
 2 files changed, 605 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h
 create mode 100644 drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h

diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h
new file mode 100644
index 0000000000000..b62b489402c54
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h
@@ -0,0 +1,177 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef _smuio_13_0_3_OFFSET_HEADER
+#define _smuio_13_0_3_OFFSET_HEADER
+
+
+
+// addressBlock: aid_smuio_smuio_reset_SmuSmuioDec
+// base address: 0x5a300
+#define regSMUIO_MP_RESET_INTR                                                                          0x00c1
+#define regSMUIO_MP_RESET_INTR_BASE_IDX                                                                 1
+#define regSMUIO_SOC_HALT                                                                               0x00c2
+#define regSMUIO_SOC_HALT_BASE_IDX                                                                      1
+
+
+// addressBlock: aid_smuio_smuio_tsc_SmuSmuioDec
+// base address: 0x5a8a0
+#define regPWROK_REFCLK_GAP_CYCLES                                                                      0x0028
+#define regPWROK_REFCLK_GAP_CYCLES_BASE_IDX                                                             2
+#define regGOLDEN_TSC_INCREMENT_UPPER                                                                   0x002b
+#define regGOLDEN_TSC_INCREMENT_UPPER_BASE_IDX                                                          2
+#define regGOLDEN_TSC_INCREMENT_LOWER                                                                   0x002c
+#define regGOLDEN_TSC_INCREMENT_LOWER_BASE_IDX                                                          2
+#define regGOLDEN_TSC_COUNT_UPPER                                                                       0x002d
+#define regGOLDEN_TSC_COUNT_UPPER_BASE_IDX                                                              2
+#define regGOLDEN_TSC_COUNT_LOWER                                                                       0x002e
+#define regGOLDEN_TSC_COUNT_LOWER_BASE_IDX                                                              2
+#define regSOC_GOLDEN_TSC_SHADOW_UPPER                                                                  0x002f
+#define regSOC_GOLDEN_TSC_SHADOW_UPPER_BASE_IDX                                                         2
+#define regSOC_GOLDEN_TSC_SHADOW_LOWER                                                                  0x0030
+#define regSOC_GOLDEN_TSC_SHADOW_LOWER_BASE_IDX                                                         2
+#define regSOC_GAP_PWROK                                                                                0x0031
+#define regSOC_GAP_PWROK_BASE_IDX                                                                       2
+
+
+// addressBlock: aid_smuio_smuio_swtimer_SmuSmuioDec
+// base address: 0x5ac70
+#define regPWR_VIRT_RESET_REQ                                                                           0x011c
+#define regPWR_VIRT_RESET_REQ_BASE_IDX                                                                  2
+#define regPWR_DISP_TIMER_CONTROL                                                                       0x011d
+#define regPWR_DISP_TIMER_CONTROL_BASE_IDX                                                              2
+#define regPWR_DISP_TIMER_DEBUG                                                                         0x011e
+#define regPWR_DISP_TIMER_DEBUG_BASE_IDX                                                                2
+#define regPWR_DISP_TIMER2_CONTROL                                                                      0x011f
+#define regPWR_DISP_TIMER2_CONTROL_BASE_IDX                                                             2
+#define regPWR_DISP_TIMER2_DEBUG                                                                        0x0120
+#define regPWR_DISP_TIMER2_DEBUG_BASE_IDX                                                               2
+#define regPWR_DISP_TIMER_GLOBAL_CONTROL                                                                0x0121
+#define regPWR_DISP_TIMER_GLOBAL_CONTROL_BASE_IDX                                                       2
+#define regPWR_IH_CONTROL                                                                               0x0122
+#define regPWR_IH_CONTROL_BASE_IDX                                                                      2
+
+
+// addressBlock: aid_smuio_smuio_misc_SmuSmuioDec
+// base address: 0x5a000
+#define regSMUIO_MCM_CONFIG                                                                             0x0023
+#define regSMUIO_MCM_CONFIG_BASE_IDX                                                                    1
+#define regIP_DISCOVERY_VERSION                                                                         0x0000
+#define regIP_DISCOVERY_VERSION_BASE_IDX                                                                2
+#define regSCRATCH_REGISTER0                                                                            0x01bd
+#define regSCRATCH_REGISTER0_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER1                                                                            0x01be
+#define regSCRATCH_REGISTER1_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER2                                                                            0x01bf
+#define regSCRATCH_REGISTER2_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER3                                                                            0x01c0
+#define regSCRATCH_REGISTER3_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER4                                                                            0x01c1
+#define regSCRATCH_REGISTER4_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER5                                                                            0x01c2
+#define regSCRATCH_REGISTER5_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER6                                                                            0x01c3
+#define regSCRATCH_REGISTER6_BASE_IDX                                                                   2
+#define regSCRATCH_REGISTER7                                                                            0x01c4
+#define regSCRATCH_REGISTER7_BASE_IDX                                                                   2
+
+
+// addressBlock: aid_smuio_smuio_gpio_SmuSmuioDec
+// base address: 0x5a500
+#define regSMU_GPIOPAD_SW_INT_STAT                                                                      0x0140
+#define regSMU_GPIOPAD_SW_INT_STAT_BASE_IDX                                                             1
+#define regSMU_GPIOPAD_MASK                                                                             0x0141
+#define regSMU_GPIOPAD_MASK_BASE_IDX                                                                    1
+#define regSMU_GPIOPAD_A                                                                                0x0142
+#define regSMU_GPIOPAD_A_BASE_IDX                                                                       1
+#define regSMU_GPIOPAD_TXIMPSEL                                                                         0x0143
+#define regSMU_GPIOPAD_TXIMPSEL_BASE_IDX                                                                1
+#define regSMU_GPIOPAD_EN                                                                               0x0144
+#define regSMU_GPIOPAD_EN_BASE_IDX                                                                      1
+#define regSMU_GPIOPAD_Y                                                                                0x0145
+#define regSMU_GPIOPAD_Y_BASE_IDX                                                                       1
+#define regSMU_GPIOPAD_RXEN                                                                             0x0146
+#define regSMU_GPIOPAD_RXEN_BASE_IDX                                                                    1
+#define regSMU_GPIOPAD_RCVR_SEL0                                                                        0x0147
+#define regSMU_GPIOPAD_RCVR_SEL0_BASE_IDX                                                               1
+#define regSMU_GPIOPAD_RCVR_SEL1                                                                        0x0148
+#define regSMU_GPIOPAD_RCVR_SEL1_BASE_IDX                                                               1
+#define regSMU_GPIOPAD_PU_EN                                                                            0x0149
+#define regSMU_GPIOPAD_PU_EN_BASE_IDX                                                                   1
+#define regSMU_GPIOPAD_PD_EN                                                                            0x014a
+#define regSMU_GPIOPAD_PD_EN_BASE_IDX                                                                   1
+#define regSMU_GPIOPAD_PINSTRAPS                                                                        0x014b
+#define regSMU_GPIOPAD_PINSTRAPS_BASE_IDX                                                               1
+#define regDFT_PINSTRAPS                                                                                0x014c
+#define regDFT_PINSTRAPS_BASE_IDX                                                                       1
+#define regSMU_GPIOPAD_INT_STAT_EN                                                                      0x014d
+#define regSMU_GPIOPAD_INT_STAT_EN_BASE_IDX                                                             1
+#define regSMU_GPIOPAD_INT_STAT                                                                         0x014e
+#define regSMU_GPIOPAD_INT_STAT_BASE_IDX                                                                1
+#define regSMU_GPIOPAD_INT_STAT_AK                                                                      0x014f
+#define regSMU_GPIOPAD_INT_STAT_AK_BASE_IDX                                                             1
+#define regSMU_GPIOPAD_INT_EN                                                                           0x0150
+#define regSMU_GPIOPAD_INT_EN_BASE_IDX                                                                  1
+#define regSMU_GPIOPAD_INT_TYPE                                                                         0x0151
+#define regSMU_GPIOPAD_INT_TYPE_BASE_IDX                                                                1
+#define regSMU_GPIOPAD_INT_POLARITY                                                                     0x0152
+#define regSMU_GPIOPAD_INT_POLARITY_BASE_IDX                                                            1
+#define regSMUIO_PCC_GPIO_SELECT                                                                        0x0155
+#define regSMUIO_PCC_GPIO_SELECT_BASE_IDX                                                               1
+#define regSMU_GPIOPAD_S0                                                                               0x0156
+#define regSMU_GPIOPAD_S0_BASE_IDX                                                                      1
+#define regSMU_GPIOPAD_S1                                                                               0x0157
+#define regSMU_GPIOPAD_S1_BASE_IDX                                                                      1
+#define regSMU_GPIOPAD_SCHMEN                                                                           0x0158
+#define regSMU_GPIOPAD_SCHMEN_BASE_IDX                                                                  1
+#define regSMU_GPIOPAD_SCL_EN                                                                           0x0159
+#define regSMU_GPIOPAD_SCL_EN_BASE_IDX                                                                  1
+#define regSMU_GPIOPAD_SDA_EN                                                                           0x015a
+#define regSMU_GPIOPAD_SDA_EN_BASE_IDX                                                                  1
+#define regSMUIO_GPIO_INT0_SELECT                                                                       0x015b
+#define regSMUIO_GPIO_INT0_SELECT_BASE_IDX                                                              1
+#define regSMUIO_GPIO_INT1_SELECT                                                                       0x015c
+#define regSMUIO_GPIO_INT1_SELECT_BASE_IDX                                                              1
+#define regSMUIO_GPIO_INT2_SELECT                                                                       0x015d
+#define regSMUIO_GPIO_INT2_SELECT_BASE_IDX                                                              1
+#define regSMUIO_GPIO_INT3_SELECT                                                                       0x015e
+#define regSMUIO_GPIO_INT3_SELECT_BASE_IDX                                                              1
+#define regSMU_GPIOPAD_MP_INT0_STAT                                                                     0x015f
+#define regSMU_GPIOPAD_MP_INT0_STAT_BASE_IDX                                                            1
+#define regSMU_GPIOPAD_MP_INT1_STAT                                                                     0x0160
+#define regSMU_GPIOPAD_MP_INT1_STAT_BASE_IDX                                                            1
+#define regSMU_GPIOPAD_MP_INT2_STAT                                                                     0x0161
+#define regSMU_GPIOPAD_MP_INT2_STAT_BASE_IDX                                                            1
+#define regSMU_GPIOPAD_MP_INT3_STAT                                                                     0x0162
+#define regSMU_GPIOPAD_MP_INT3_STAT_BASE_IDX                                                            1
+#define regSMIO_INDEX                                                                                   0x0163
+#define regSMIO_INDEX_BASE_IDX                                                                          1
+#define regS0_VID_SMIO_CNTL                                                                             0x0164
+#define regS0_VID_SMIO_CNTL_BASE_IDX                                                                    1
+#define regS1_VID_SMIO_CNTL                                                                             0x0165
+#define regS1_VID_SMIO_CNTL_BASE_IDX                                                                    1
+#define regOPEN_DRAIN_SELECT                                                                            0x0166
+#define regOPEN_DRAIN_SELECT_BASE_IDX                                                                   1
+#define regSMIO_ENABLE                                                                                  0x0167
+#define regSMIO_ENABLE_BASE_IDX                                                                         1
+
+#endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h
new file mode 100644
index 0000000000000..be896f3089fe2
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h
@@ -0,0 +1,428 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef _smuio_13_0_3_SH_MASK_HEADER
+#define _smuio_13_0_3_SH_MASK_HEADER
+
+
+// addressBlock: aid_smuio_smuio_reset_SmuSmuioDec
+//SMUIO_MP_RESET_INTR
+#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR__SHIFT                                                       0x0
+#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR_MASK                                                         0x00000001L
+//SMUIO_SOC_HALT
+#define SMUIO_SOC_HALT__WDT_FORCE_PWROK_EN__SHIFT                                                             0x2
+#define SMUIO_SOC_HALT__WDT_FORCE_RESETn_EN__SHIFT                                                            0x3
+#define SMUIO_SOC_HALT__WDT_FORCE_PWROK_EN_MASK                                                               0x00000004L
+#define SMUIO_SOC_HALT__WDT_FORCE_RESETn_EN_MASK                                                              0x00000008L
+
+
+// addressBlock: aid_smuio_smuio_tsc_SmuSmuioDec
+//PWROK_REFCLK_GAP_CYCLES
+#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PreAssertion_clkgap_cycles__SHIFT                                      0x0
+#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PostAssertion_clkgap_cycles__SHIFT                                     0x8
+#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PreAssertion_clkgap_cycles_MASK                                        0x000000FFL
+#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PostAssertion_clkgap_cycles_MASK                                       0x0000FF00L
+//GOLDEN_TSC_INCREMENT_UPPER
+#define GOLDEN_TSC_INCREMENT_UPPER__GoldenTscIncrementUpper__SHIFT                                            0x0
+#define GOLDEN_TSC_INCREMENT_UPPER__GoldenTscIncrementUpper_MASK                                              0x00FFFFFFL
+//GOLDEN_TSC_INCREMENT_LOWER
+#define GOLDEN_TSC_INCREMENT_LOWER__GoldenTscIncrementLower__SHIFT                                            0x0
+#define GOLDEN_TSC_INCREMENT_LOWER__GoldenTscIncrementLower_MASK                                              0xFFFFFFFFL
+//GOLDEN_TSC_COUNT_UPPER
+#define GOLDEN_TSC_COUNT_UPPER__GoldenTscCountUpper__SHIFT                                                    0x0
+#define GOLDEN_TSC_COUNT_UPPER__GoldenTscCountUpper_MASK                                                      0x00FFFFFFL
+//GOLDEN_TSC_COUNT_LOWER
+#define GOLDEN_TSC_COUNT_LOWER__GoldenTscCountLower__SHIFT                                                    0x0
+#define GOLDEN_TSC_COUNT_LOWER__GoldenTscCountLower_MASK                                                      0xFFFFFFFFL
+//SOC_GOLDEN_TSC_SHADOW_UPPER
+#define SOC_GOLDEN_TSC_SHADOW_UPPER__SocGoldenTscShadowUpper__SHIFT                                           0x0
+#define SOC_GOLDEN_TSC_SHADOW_UPPER__SocGoldenTscShadowUpper_MASK                                             0x00FFFFFFL
+//SOC_GOLDEN_TSC_SHADOW_LOWER
+#define SOC_GOLDEN_TSC_SHADOW_LOWER__SocGoldenTscShadowLower__SHIFT                                           0x0
+#define SOC_GOLDEN_TSC_SHADOW_LOWER__SocGoldenTscShadowLower_MASK                                             0xFFFFFFFFL
+//SOC_GAP_PWROK
+#define SOC_GAP_PWROK__soc_gap_pwrok__SHIFT                                                                   0x0
+#define SOC_GAP_PWROK__soc_gap_pwrok_MASK                                                                     0x00000001L
+
+
+// addressBlock: aid_smuio_smuio_swtimer_SmuSmuioDec
+//PWR_VIRT_RESET_REQ
+#define PWR_VIRT_RESET_REQ__VF_FLR__SHIFT                                                                     0x0
+#define PWR_VIRT_RESET_REQ__PF_FLR__SHIFT                                                                     0x1f
+#define PWR_VIRT_RESET_REQ__VF_FLR_MASK                                                                       0x7FFFFFFFL
+#define PWR_VIRT_RESET_REQ__PF_FLR_MASK                                                                       0x80000000L
+//PWR_DISP_TIMER_CONTROL
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_COUNT__SHIFT                                                   0x0
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_ENABLE__SHIFT                                                  0x19
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_DISABLE__SHIFT                                                 0x1a
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MASK__SHIFT                                                    0x1b
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_STAT_AK__SHIFT                                                 0x1c
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_TYPE__SHIFT                                                    0x1d
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MODE__SHIFT                                                    0x1e
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_COUNT_MASK                                                     0x01FFFFFFL
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_ENABLE_MASK                                                    0x02000000L
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_DISABLE_MASK                                                   0x04000000L
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MASK_MASK                                                      0x08000000L
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_STAT_AK_MASK                                                   0x10000000L
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_TYPE_MASK                                                      0x20000000L
+#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MODE_MASK                                                      0x40000000L
+//PWR_DISP_TIMER_DEBUG
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_RUNNING__SHIFT                                                   0x0
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_STAT__SHIFT                                                      0x1
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT__SHIFT                                                           0x2
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_RUN_VAL__SHIFT                                                       0x7
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_RUNNING_MASK                                                     0x00000001L
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_STAT_MASK                                                        0x00000002L
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_MASK                                                             0x00000004L
+#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_RUN_VAL_MASK                                                         0xFFFFFF80L
+//PWR_DISP_TIMER2_CONTROL
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_COUNT__SHIFT                                                  0x0
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_ENABLE__SHIFT                                                 0x19
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_DISABLE__SHIFT                                                0x1a
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MASK__SHIFT                                                   0x1b
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_STAT_AK__SHIFT                                                0x1c
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_TYPE__SHIFT                                                   0x1d
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MODE__SHIFT                                                   0x1e
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_COUNT_MASK                                                    0x01FFFFFFL
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_ENABLE_MASK                                                   0x02000000L
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_DISABLE_MASK                                                  0x04000000L
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MASK_MASK                                                     0x08000000L
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_STAT_AK_MASK                                                  0x10000000L
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_TYPE_MASK                                                     0x20000000L
+#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MODE_MASK                                                     0x40000000L
+//PWR_DISP_TIMER2_DEBUG
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_RUNNING__SHIFT                                                  0x0
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_STAT__SHIFT                                                     0x1
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT__SHIFT                                                          0x2
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_RUN_VAL__SHIFT                                                      0x7
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_RUNNING_MASK                                                    0x00000001L
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_STAT_MASK                                                       0x00000002L
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_MASK                                                            0x00000004L
+#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_RUN_VAL_MASK                                                        0xFFFFFF80L
+//PWR_DISP_TIMER_GLOBAL_CONTROL
+#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_WIDTH__SHIFT                                          0x0
+#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_EN__SHIFT                                             0xa
+#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_WIDTH_MASK                                            0x000003FFL
+#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_EN_MASK                                               0x00000400L
+//PWR_IH_CONTROL
+#define PWR_IH_CONTROL__MAX_CREDIT__SHIFT                                                                     0x0
+#define PWR_IH_CONTROL__DISP_TIMER_TRIGGER_MASK__SHIFT                                                        0x5
+#define PWR_IH_CONTROL__DISP_TIMER2_TRIGGER_MASK__SHIFT                                                       0x6
+#define PWR_IH_CONTROL__PWR_IH_CLK_GATE_EN__SHIFT                                                             0x1f
+#define PWR_IH_CONTROL__MAX_CREDIT_MASK                                                                       0x0000001FL
+#define PWR_IH_CONTROL__DISP_TIMER_TRIGGER_MASK_MASK                                                          0x00000020L
+#define PWR_IH_CONTROL__DISP_TIMER2_TRIGGER_MASK_MASK                                                         0x00000040L
+#define PWR_IH_CONTROL__PWR_IH_CLK_GATE_EN_MASK                                                               0x80000000L
+
+
+// addressBlock: aid_smuio_smuio_misc_SmuSmuioDec
+//SMUIO_MCM_CONFIG
+#define SMUIO_MCM_CONFIG__DIE_ID__SHIFT                                                                       0x0
+#define SMUIO_MCM_CONFIG__PKG_TYPE__SHIFT                                                                     0x2
+#define SMUIO_MCM_CONFIG__SOCKET_ID__SHIFT                                                                    0x8
+#define SMUIO_MCM_CONFIG__PKG_SUBTYPE__SHIFT                                                                  0xc
+#define SMUIO_MCM_CONFIG__CONSOLE_K__SHIFT                                                                    0x10
+#define SMUIO_MCM_CONFIG__CONSOLE_A__SHIFT                                                                    0x11
+#define SMUIO_MCM_CONFIG__TOPOLOGY_ID__SHIFT                                                                  0x12
+#define SMUIO_MCM_CONFIG__DIE_ID_MASK                                                                         0x00000003L
+#define SMUIO_MCM_CONFIG__PKG_TYPE_MASK                                                                       0x0000003CL
+#define SMUIO_MCM_CONFIG__SOCKET_ID_MASK                                                                      0x00000F00L
+#define SMUIO_MCM_CONFIG__PKG_SUBTYPE_MASK                                                                    0x00001000L
+#define SMUIO_MCM_CONFIG__CONSOLE_K_MASK                                                                      0x00010000L
+#define SMUIO_MCM_CONFIG__CONSOLE_A_MASK                                                                      0x00020000L
+#define SMUIO_MCM_CONFIG__TOPOLOGY_ID_MASK                                                                    0x007C0000L
+//IP_DISCOVERY_VERSION
+#define IP_DISCOVERY_VERSION__IP_DISCOVERY_VERSION__SHIFT                                                     0x0
+#define IP_DISCOVERY_VERSION__IP_DISCOVERY_VERSION_MASK                                                       0xFFFFFFFFL
+//SCRATCH_REGISTER0
+#define SCRATCH_REGISTER0__ScratchPad0__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER0__ScratchPad0_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER1
+#define SCRATCH_REGISTER1__ScratchPad1__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER1__ScratchPad1_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER2
+#define SCRATCH_REGISTER2__ScratchPad2__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER2__ScratchPad2_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER3
+#define SCRATCH_REGISTER3__ScratchPad3__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER3__ScratchPad3_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER4
+#define SCRATCH_REGISTER4__ScratchPad4__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER4__ScratchPad4_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER5
+#define SCRATCH_REGISTER5__ScratchPad5__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER5__ScratchPad5_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER6
+#define SCRATCH_REGISTER6__ScratchPad6__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER6__ScratchPad6_MASK                                                                   0xFFFFFFFFL
+//SCRATCH_REGISTER7
+#define SCRATCH_REGISTER7__ScratchPad7__SHIFT                                                                 0x0
+#define SCRATCH_REGISTER7__ScratchPad7_MASK                                                                   0xFFFFFFFFL
+
+
+// addressBlock: aid_smuio_smuio_gpio_SmuSmuioDec
+//SMU_GPIOPAD_SW_INT_STAT
+#define SMU_GPIOPAD_SW_INT_STAT__SW_INT_STAT__SHIFT                                                           0x0
+#define SMU_GPIOPAD_SW_INT_STAT__SW_INT_STAT_MASK                                                             0x00000001L
+//SMU_GPIOPAD_MASK
+#define SMU_GPIOPAD_MASK__GPIO_MASK__SHIFT                                                                    0x0
+#define SMU_GPIOPAD_MASK__GPIO_MASK_MASK                                                                      0x7FFFFFFFL
+//SMU_GPIOPAD_A
+#define SMU_GPIOPAD_A__GPIO_A__SHIFT                                                                          0x0
+#define SMU_GPIOPAD_A__GPIO_A_MASK                                                                            0x7FFFFFFFL
+//SMU_GPIOPAD_TXIMPSEL
+#define SMU_GPIOPAD_TXIMPSEL__GPIO_TXIMPSEL__SHIFT                                                            0x0
+#define SMU_GPIOPAD_TXIMPSEL__GPIO_TXIMPSEL_MASK                                                              0x7FFFFFFFL
+//SMU_GPIOPAD_EN
+#define SMU_GPIOPAD_EN__GPIO_EN__SHIFT                                                                        0x0
+#define SMU_GPIOPAD_EN__GPIO_EN_MASK                                                                          0x7FFFFFFFL
+//SMU_GPIOPAD_Y
+#define SMU_GPIOPAD_Y__GPIO_Y__SHIFT                                                                          0x0
+#define SMU_GPIOPAD_Y__GPIO_Y_MASK                                                                            0x7FFFFFFFL
+//SMU_GPIOPAD_RXEN
+#define SMU_GPIOPAD_RXEN__GPIO_RXEN__SHIFT                                                                    0x0
+#define SMU_GPIOPAD_RXEN__GPIO_RXEN_MASK                                                                      0x7FFFFFFFL
+//SMU_GPIOPAD_RCVR_SEL0
+#define SMU_GPIOPAD_RCVR_SEL0__GPIO_RCVR_SEL0__SHIFT                                                          0x0
+#define SMU_GPIOPAD_RCVR_SEL0__GPIO_RCVR_SEL0_MASK                                                            0x7FFFFFFFL
+//SMU_GPIOPAD_RCVR_SEL1
+#define SMU_GPIOPAD_RCVR_SEL1__GPIO_RCVR_SEL1__SHIFT                                                          0x0
+#define SMU_GPIOPAD_RCVR_SEL1__GPIO_RCVR_SEL1_MASK                                                            0x7FFFFFFFL
+//SMU_GPIOPAD_PU_EN
+#define SMU_GPIOPAD_PU_EN__GPIO_PU_EN__SHIFT                                                                  0x0
+#define SMU_GPIOPAD_PU_EN__GPIO_PU_EN_MASK                                                                    0x7FFFFFFFL
+//SMU_GPIOPAD_PD_EN
+#define SMU_GPIOPAD_PD_EN__GPIO_PD_EN__SHIFT                                                                  0x0
+#define SMU_GPIOPAD_PD_EN__GPIO_PD_EN_MASK                                                                    0x7FFFFFFFL
+//SMU_GPIOPAD_PINSTRAPS
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_0__SHIFT                                                         0x0
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_1__SHIFT                                                         0x1
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_2__SHIFT                                                         0x2
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_3__SHIFT                                                         0x3
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_4__SHIFT                                                         0x4
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_5__SHIFT                                                         0x5
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_6__SHIFT                                                         0x6
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_7__SHIFT                                                         0x7
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_8__SHIFT                                                         0x8
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_9__SHIFT                                                         0x9
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_10__SHIFT                                                        0xa
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_11__SHIFT                                                        0xb
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_12__SHIFT                                                        0xc
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_13__SHIFT                                                        0xd
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_14__SHIFT                                                        0xe
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_15__SHIFT                                                        0xf
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_16__SHIFT                                                        0x10
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_17__SHIFT                                                        0x11
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_18__SHIFT                                                        0x12
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_19__SHIFT                                                        0x13
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_20__SHIFT                                                        0x14
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_21__SHIFT                                                        0x15
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_22__SHIFT                                                        0x16
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_23__SHIFT                                                        0x17
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_24__SHIFT                                                        0x18
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_25__SHIFT                                                        0x19
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_26__SHIFT                                                        0x1a
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_27__SHIFT                                                        0x1b
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_28__SHIFT                                                        0x1c
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_29__SHIFT                                                        0x1d
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_30__SHIFT                                                        0x1e
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_0_MASK                                                           0x00000001L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_1_MASK                                                           0x00000002L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_2_MASK                                                           0x00000004L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_3_MASK                                                           0x00000008L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_4_MASK                                                           0x00000010L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_5_MASK                                                           0x00000020L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_6_MASK                                                           0x00000040L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_7_MASK                                                           0x00000080L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_8_MASK                                                           0x00000100L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_9_MASK                                                           0x00000200L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_10_MASK                                                          0x00000400L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_11_MASK                                                          0x00000800L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_12_MASK                                                          0x00001000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_13_MASK                                                          0x00002000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_14_MASK                                                          0x00004000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_15_MASK                                                          0x00008000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_16_MASK                                                          0x00010000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_17_MASK                                                          0x00020000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_18_MASK                                                          0x00040000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_19_MASK                                                          0x00080000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_20_MASK                                                          0x00100000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_21_MASK                                                          0x00200000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_22_MASK                                                          0x00400000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_23_MASK                                                          0x00800000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_24_MASK                                                          0x01000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_25_MASK                                                          0x02000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_26_MASK                                                          0x04000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_27_MASK                                                          0x08000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_28_MASK                                                          0x10000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_29_MASK                                                          0x20000000L
+#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_30_MASK                                                          0x40000000L
+//DFT_PINSTRAPS
+#define DFT_PINSTRAPS__DFT_PINSTRAPS__SHIFT                                                                   0x0
+#define DFT_PINSTRAPS__DFT_PINSTRAPS_MASK                                                                     0x000003FFL
+//SMU_GPIOPAD_INT_STAT_EN
+#define SMU_GPIOPAD_INT_STAT_EN__GPIO_INT_STAT_EN__SHIFT                                                      0x0
+#define SMU_GPIOPAD_INT_STAT_EN__SW_INITIATED_INT_STAT_EN__SHIFT                                              0x1f
+#define SMU_GPIOPAD_INT_STAT_EN__GPIO_INT_STAT_EN_MASK                                                        0x1FFFFFFFL
+#define SMU_GPIOPAD_INT_STAT_EN__SW_INITIATED_INT_STAT_EN_MASK                                                0x80000000L
+//SMU_GPIOPAD_INT_STAT
+#define SMU_GPIOPAD_INT_STAT__GPIO_INT_STAT__SHIFT                                                            0x0
+#define SMU_GPIOPAD_INT_STAT__SW_INITIATED_INT_STAT__SHIFT                                                    0x1f
+#define SMU_GPIOPAD_INT_STAT__GPIO_INT_STAT_MASK                                                              0x1FFFFFFFL
+#define SMU_GPIOPAD_INT_STAT__SW_INITIATED_INT_STAT_MASK                                                      0x80000000L
+//SMU_GPIOPAD_INT_STAT_AK
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_0__SHIFT                                                    0x0
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_1__SHIFT                                                    0x1
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_2__SHIFT                                                    0x2
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_3__SHIFT                                                    0x3
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_4__SHIFT                                                    0x4
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_5__SHIFT                                                    0x5
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_6__SHIFT                                                    0x6
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_7__SHIFT                                                    0x7
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_8__SHIFT                                                    0x8
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_9__SHIFT                                                    0x9
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_10__SHIFT                                                   0xa
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_11__SHIFT                                                   0xb
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_12__SHIFT                                                   0xc
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_13__SHIFT                                                   0xd
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_14__SHIFT                                                   0xe
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_15__SHIFT                                                   0xf
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_16__SHIFT                                                   0x10
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_17__SHIFT                                                   0x11
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_18__SHIFT                                                   0x12
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_19__SHIFT                                                   0x13
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_20__SHIFT                                                   0x14
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_21__SHIFT                                                   0x15
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_22__SHIFT                                                   0x16
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_23__SHIFT                                                   0x17
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_24__SHIFT                                                   0x18
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_25__SHIFT                                                   0x19
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_26__SHIFT                                                   0x1a
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_27__SHIFT                                                   0x1b
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_28__SHIFT                                                   0x1c
+#define SMU_GPIOPAD_INT_STAT_AK__SW_INITIATED_INT_STAT_AK__SHIFT                                              0x1f
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_0_MASK                                                      0x00000001L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_1_MASK                                                      0x00000002L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_2_MASK                                                      0x00000004L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_3_MASK                                                      0x00000008L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_4_MASK                                                      0x00000010L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_5_MASK                                                      0x00000020L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_6_MASK                                                      0x00000040L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_7_MASK                                                      0x00000080L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_8_MASK                                                      0x00000100L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_9_MASK                                                      0x00000200L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_10_MASK                                                     0x00000400L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_11_MASK                                                     0x00000800L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_12_MASK                                                     0x00001000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_13_MASK                                                     0x00002000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_14_MASK                                                     0x00004000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_15_MASK                                                     0x00008000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_16_MASK                                                     0x00010000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_17_MASK                                                     0x00020000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_18_MASK                                                     0x00040000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_19_MASK                                                     0x00080000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_20_MASK                                                     0x00100000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_21_MASK                                                     0x00200000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_22_MASK                                                     0x00400000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_23_MASK                                                     0x00800000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_24_MASK                                                     0x01000000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_25_MASK                                                     0x02000000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_26_MASK                                                     0x04000000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_27_MASK                                                     0x08000000L
+#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_28_MASK                                                     0x10000000L
+#define SMU_GPIOPAD_INT_STAT_AK__SW_INITIATED_INT_STAT_AK_MASK                                                0x80000000L
+//SMU_GPIOPAD_INT_EN
+#define SMU_GPIOPAD_INT_EN__GPIO_INT_EN__SHIFT                                                                0x0
+#define SMU_GPIOPAD_INT_EN__SW_INITIATED_INT_EN__SHIFT                                                        0x1f
+#define SMU_GPIOPAD_INT_EN__GPIO_INT_EN_MASK                                                                  0x1FFFFFFFL
+#define SMU_GPIOPAD_INT_EN__SW_INITIATED_INT_EN_MASK                                                          0x80000000L
+//SMU_GPIOPAD_INT_TYPE
+#define SMU_GPIOPAD_INT_TYPE__GPIO_INT_TYPE__SHIFT                                                            0x0
+#define SMU_GPIOPAD_INT_TYPE__SW_INITIATED_INT_TYPE__SHIFT                                                    0x1f
+#define SMU_GPIOPAD_INT_TYPE__GPIO_INT_TYPE_MASK                                                              0x1FFFFFFFL
+#define SMU_GPIOPAD_INT_TYPE__SW_INITIATED_INT_TYPE_MASK                                                      0x80000000L
+//SMU_GPIOPAD_INT_POLARITY
+#define SMU_GPIOPAD_INT_POLARITY__GPIO_INT_POLARITY__SHIFT                                                    0x0
+#define SMU_GPIOPAD_INT_POLARITY__SW_INITIATED_INT_POLARITY__SHIFT                                            0x1f
+#define SMU_GPIOPAD_INT_POLARITY__GPIO_INT_POLARITY_MASK                                                      0x1FFFFFFFL
+#define SMU_GPIOPAD_INT_POLARITY__SW_INITIATED_INT_POLARITY_MASK                                              0x80000000L
+//SMUIO_PCC_GPIO_SELECT
+#define SMUIO_PCC_GPIO_SELECT__GPIO__SHIFT                                                                    0x0
+#define SMUIO_PCC_GPIO_SELECT__GPIO_MASK                                                                      0xFFFFFFFFL
+//SMU_GPIOPAD_S0
+#define SMU_GPIOPAD_S0__GPIO_S0__SHIFT                                                                        0x0
+#define SMU_GPIOPAD_S0__GPIO_S0_MASK                                                                          0x7FFFFFFFL
+//SMU_GPIOPAD_S1
+#define SMU_GPIOPAD_S1__GPIO_S1__SHIFT                                                                        0x0
+#define SMU_GPIOPAD_S1__GPIO_S1_MASK                                                                          0x7FFFFFFFL
+//SMU_GPIOPAD_SCHMEN
+#define SMU_GPIOPAD_SCHMEN__GPIO_SCHMEN__SHIFT                                                                0x0
+#define SMU_GPIOPAD_SCHMEN__GPIO_SCHMEN_MASK                                                                  0x7FFFFFFFL
+//SMU_GPIOPAD_SCL_EN
+#define SMU_GPIOPAD_SCL_EN__GPIO_SCL_EN__SHIFT                                                                0x0
+#define SMU_GPIOPAD_SCL_EN__GPIO_SCL_EN_MASK                                                                  0x7FFFFFFFL
+//SMU_GPIOPAD_SDA_EN
+#define SMU_GPIOPAD_SDA_EN__GPIO_SDA_EN__SHIFT                                                                0x0
+#define SMU_GPIOPAD_SDA_EN__GPIO_SDA_EN_MASK                                                                  0x7FFFFFFFL
+//SMUIO_GPIO_INT0_SELECT
+#define SMUIO_GPIO_INT0_SELECT__GPIO_INT0_SELECT__SHIFT                                                       0x0
+#define SMUIO_GPIO_INT0_SELECT__GPIO_INT0_SELECT_MASK                                                         0xFFFFFFFFL
+//SMUIO_GPIO_INT1_SELECT
+#define SMUIO_GPIO_INT1_SELECT__GPIO_INT1_SELECT__SHIFT                                                       0x0
+#define SMUIO_GPIO_INT1_SELECT__GPIO_INT1_SELECT_MASK                                                         0xFFFFFFFFL
+//SMUIO_GPIO_INT2_SELECT
+#define SMUIO_GPIO_INT2_SELECT__GPIO_INT2_SELECT__SHIFT                                                       0x0
+#define SMUIO_GPIO_INT2_SELECT__GPIO_INT2_SELECT_MASK                                                         0xFFFFFFFFL
+//SMUIO_GPIO_INT3_SELECT
+#define SMUIO_GPIO_INT3_SELECT__GPIO_INT3_SELECT__SHIFT                                                       0x0
+#define SMUIO_GPIO_INT3_SELECT__GPIO_INT3_SELECT_MASK                                                         0xFFFFFFFFL
+//SMU_GPIOPAD_MP_INT0_STAT
+#define SMU_GPIOPAD_MP_INT0_STAT__GPIO_MP_INT0_STAT__SHIFT                                                    0x0
+#define SMU_GPIOPAD_MP_INT0_STAT__GPIO_MP_INT0_STAT_MASK                                                      0x1FFFFFFFL
+//SMU_GPIOPAD_MP_INT1_STAT
+#define SMU_GPIOPAD_MP_INT1_STAT__GPIO_MP_INT1_STAT__SHIFT                                                    0x0
+#define SMU_GPIOPAD_MP_INT1_STAT__GPIO_MP_INT1_STAT_MASK                                                      0x1FFFFFFFL
+//SMU_GPIOPAD_MP_INT2_STAT
+#define SMU_GPIOPAD_MP_INT2_STAT__GPIO_MP_INT2_STAT__SHIFT                                                    0x0
+#define SMU_GPIOPAD_MP_INT2_STAT__GPIO_MP_INT2_STAT_MASK                                                      0x1FFFFFFFL
+//SMU_GPIOPAD_MP_INT3_STAT
+#define SMU_GPIOPAD_MP_INT3_STAT__GPIO_MP_INT3_STAT__SHIFT                                                    0x0
+#define SMU_GPIOPAD_MP_INT3_STAT__GPIO_MP_INT3_STAT_MASK                                                      0x1FFFFFFFL
+//SMIO_INDEX
+#define SMIO_INDEX__SW_SMIO_INDEX__SHIFT                                                                      0x0
+#define SMIO_INDEX__SW_SMIO_INDEX_MASK                                                                        0x00000001L
+//S0_VID_SMIO_CNTL
+#define S0_VID_SMIO_CNTL__S0_SMIO_VALUES__SHIFT                                                               0x0
+#define S0_VID_SMIO_CNTL__S0_SMIO_VALUES_MASK                                                                 0xFFFFFFFFL
+//S1_VID_SMIO_CNTL
+#define S1_VID_SMIO_CNTL__S1_SMIO_VALUES__SHIFT                                                               0x0
+#define S1_VID_SMIO_CNTL__S1_SMIO_VALUES_MASK                                                                 0xFFFFFFFFL
+//OPEN_DRAIN_SELECT
+#define OPEN_DRAIN_SELECT__OPEN_DRAIN_SELECT__SHIFT                                                           0x0
+#define OPEN_DRAIN_SELECT__RESERVED__SHIFT                                                                    0x1f
+#define OPEN_DRAIN_SELECT__OPEN_DRAIN_SELECT_MASK                                                             0x7FFFFFFFL
+#define OPEN_DRAIN_SELECT__RESERVED_MASK                                                                      0x80000000L
+//SMIO_ENABLE
+#define SMIO_ENABLE__SMIO_ENABLE__SHIFT                                                                       0x0
+#define SMIO_ENABLE__SMIO_ENABLE_MASK                                                                         0xFFFFFFFFL
+
+#endif
-- 
GitLab


From 3d2ea552b229495050316e84d7cb0257cb3cd13b Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Tue, 12 Apr 2022 23:33:20 -0400
Subject: [PATCH 1131/1791] drm/amdgpu: implement smuio v13_0_3 callbacks

Add smuio v13_0_3 callbacks for SMUIO.

Tested-by: Ori Messinger <Ori.Messinger@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile        |   1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h    |   5 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h  |   1 +
 drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c | 103 +++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h |  28 ++++++
 5 files changed, 138 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c
 create mode 100644 drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 90f771423c947..8418a90df4933 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -205,6 +205,7 @@ amdgpu-y += \
 	smuio_v11_0.o \
 	smuio_v11_0_6.o \
 	smuio_v13_0.o \
+	smuio_v13_0_3.o \
 	smuio_v13_0_6.o
 
 # add reset block
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 2b2d75763875e..789115f5505f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -66,6 +66,11 @@ enum amdgpu_gfx_partition {
 
 #define NUM_XCC(x) hweight16(x)
 
+enum amdgpu_pkg_type {
+	AMDGPU_PKG_TYPE_APU = 2,
+	AMDGPU_PKG_TYPE_UNKNOWN,
+};
+
 struct amdgpu_mec {
 	struct amdgpu_bo	*hpd_eop_obj;
 	u64			hpd_eop_gpu_addr;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h
index c7a823f3f2c59..89c38d8644718 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h
@@ -30,6 +30,7 @@ struct amdgpu_smuio_funcs {
 	void (*get_clock_gating_state)(struct amdgpu_device *adev, u64 *flags);
 	u32 (*get_die_id)(struct amdgpu_device *adev);
 	u32 (*get_socket_id)(struct amdgpu_device *adev);
+	enum amdgpu_pkg_type (*get_pkg_type)(struct amdgpu_device *adev);
 	bool (*is_host_gpu_xgmi_supported)(struct amdgpu_device *adev);
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c
new file mode 100644
index 0000000000000..4368a5891eebd
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c
@@ -0,0 +1,103 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "smuio_v13_0_3.h"
+#include "soc15_common.h"
+#include "smuio/smuio_13_0_3_offset.h"
+#include "smuio/smuio_13_0_3_sh_mask.h"
+
+#define PKG_TYPE_MASK		0x00000003L
+
+/**
+ * smuio_v13_0_3_get_die_id - query die id from FCH.
+ *
+ * @adev: amdgpu device pointer
+ *
+ * Returns die id
+ */
+static u32 smuio_v13_0_3_get_die_id(struct amdgpu_device *adev)
+{
+	u32 data, die_id;
+
+	data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG);
+	die_id = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, DIE_ID);
+
+	return die_id;
+}
+
+/**
+ * smuio_v13_0_3_get_socket_id - query socket id from FCH
+ *
+ * @adev: amdgpu device pointer
+ *
+ * Returns socket id
+ */
+static u32 smuio_v13_0_3_get_socket_id(struct amdgpu_device *adev)
+{
+	u32 data, socket_id;
+
+	data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG);
+	socket_id = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, SOCKET_ID);
+
+	return socket_id;
+}
+
+/**
+ * smuio_v13_0_3_get_pkg_type - query package type set by MP1/bootcode
+ *
+ * @adev: amdgpu device pointer
+ *
+ * Returns package type
+ */
+
+static enum amdgpu_pkg_type smuio_v13_0_3_get_pkg_type(struct amdgpu_device *adev)
+{
+	enum amdgpu_pkg_type pkg_type;
+	u32 data;
+
+	data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG);
+	data = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, PKG_TYPE);
+	/* pkg_type[4:0]
+	 *
+	 * bit 1 == 1 APU form factor
+	 *
+	 * b0100 - b1111 - Reserved
+	 */
+	switch (data & PKG_TYPE_MASK) {
+	case 0x2:
+		pkg_type = AMDGPU_PKG_TYPE_APU;
+		break;
+	default:
+		pkg_type = AMDGPU_PKG_TYPE_UNKNOWN;
+		break;
+	}
+
+	return pkg_type;
+}
+
+
+const struct amdgpu_smuio_funcs smuio_v13_0_3_funcs = {
+	.get_die_id = smuio_v13_0_3_get_die_id,
+	.get_socket_id = smuio_v13_0_3_get_socket_id,
+	.get_pkg_type = smuio_v13_0_3_get_pkg_type,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h
new file mode 100644
index 0000000000000..aec35f7efb589
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __SMUIO_V13_0_3_H__
+#define __SMUIO_V13_0_3_H__
+
+extern const struct amdgpu_smuio_funcs smuio_v13_0_3_funcs;
+
+#endif /* __SMUIO_V13_0_3_H__ */
-- 
GitLab


From 6b7ec18b045ff524eab94340e18feefe9a783d2e Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Tue, 23 Nov 2021 22:27:17 +0800
Subject: [PATCH 1132/1791] drm/amdgpu: init smuio funcs for smuio v13_0_3

Add callbacks for SMUIO 13.0.3

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 4 ++++
 drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h    | 2 ++
 2 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 71293db09009f..b602df4b445a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -82,6 +82,7 @@
 #include "smuio_v11_0.h"
 #include "smuio_v11_0_6.h"
 #include "smuio_v13_0.h"
+#include "smuio_v13_0_3.h"
 #include "smuio_v13_0_6.h"
 
 #define FIRMWARE_IP_DISCOVERY "amdgpu/ip_discovery.bin"
@@ -2382,6 +2383,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(13, 0, 2):
 		adev->smuio.funcs = &smuio_v13_0_funcs;
 		break;
+	case IP_VERSION(13, 0, 3):
+		adev->smuio.funcs = &smuio_v13_0_3_funcs;
+		break;
 	case IP_VERSION(13, 0, 6):
 	case IP_VERSION(13, 0, 8):
 		adev->smuio.funcs = &smuio_v13_0_6_funcs;
diff --git a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h
index aec35f7efb589..795f66c5a58b8 100644
--- a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h
@@ -23,6 +23,8 @@
 #ifndef __SMUIO_V13_0_3_H__
 #define __SMUIO_V13_0_3_H__
 
+#include "soc15_common.h"
+
 extern const struct amdgpu_smuio_funcs smuio_v13_0_3_funcs;
 
 #endif /* __SMUIO_V13_0_3_H__ */
-- 
GitLab


From ea2d2f8ececdd4c778b66e19b82ce278dfc5e1c4 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Tue, 5 Apr 2022 13:00:13 -0400
Subject: [PATCH 1133/1791] drm/amdgpu: detect current GPU memory partition
 mode

 - Add helpers to detect the current GPU memory partition.
 - Add current memory partition mode sysfs node.

Tested-by: Ori Messinger <Ori.Messinger@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c  | 25 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h  | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c  | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c   | 12 ++++++++++++
 5 files changed, 60 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 2f6d85090b55e..c2e92664031b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1200,6 +1200,24 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 	return sysfs_emit(buf, "%s\n", partition_mode);
 }
 
+static ssize_t amdgpu_gfx_get_current_memory_partition(struct device *dev,
+						struct device_attribute *addr,
+						char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	enum amdgpu_memory_partition mode;
+	static const char *partition_modes[] = {
+		"UNKNOWN", "NPS1", "NPS2", "NPS4", "NPS8"
+	};
+	BUILD_BUG_ON(ARRAY_SIZE(partition_modes) <= AMDGPU_NPS8_PARTITION_MODE);
+
+	mode = min((int)adev->gfx.funcs->query_mem_partition_mode(adev),
+		AMDGPU_NPS8_PARTITION_MODE);
+
+	return sysfs_emit(buf, "%s\n", partition_modes[mode]);
+}
+
 static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 						struct device_attribute *addr,
 						const char *buf, size_t count)
@@ -1307,6 +1325,9 @@ static DEVICE_ATTR(current_compute_partition, S_IRUGO | S_IWUSR,
 static DEVICE_ATTR(available_compute_partition, S_IRUGO,
 		   amdgpu_gfx_get_available_compute_partition, NULL);
 
+static DEVICE_ATTR(current_memory_partition, S_IRUGO,
+		   amdgpu_gfx_get_current_memory_partition, NULL);
+
 int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
 {
 	int r;
@@ -1319,5 +1340,9 @@ int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
 	if (r)
 		return r;
 
+	r = device_create_file(adev->dev, &dev_attr_current_memory_partition);
+	if (r)
+		return r;
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 789115f5505f9..e37501685fa94 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -71,6 +71,14 @@ enum amdgpu_pkg_type {
 	AMDGPU_PKG_TYPE_UNKNOWN,
 };
 
+enum amdgpu_memory_partition {
+	UNKNOWN_MEMORY_PARTITION_MODE = 0,
+	AMDGPU_NPS1_PARTITION_MODE = 1,
+	AMDGPU_NPS2_PARTITION_MODE = 2,
+	AMDGPU_NPS4_PARTITION_MODE = 3,
+	AMDGPU_NPS8_PARTITION_MODE = 4,
+};
+
 struct amdgpu_mec {
 	struct amdgpu_bo	*hpd_eop_obj;
 	u64			hpd_eop_gpu_addr;
@@ -268,6 +276,8 @@ struct amdgpu_gfx_funcs {
 				   struct amdgpu_gfx_shadow_info *shadow_info);
 	enum amdgpu_gfx_partition
 			(*query_partition_mode)(struct amdgpu_device *adev);
+	enum amdgpu_memory_partition
+			(*query_mem_partition_mode)(struct amdgpu_device *adev);
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
 				     enum amdgpu_gfx_partition mode);
 };
@@ -404,6 +414,7 @@ struct amdgpu_gfx {
 
 	enum amdgpu_gfx_partition	partition_mode;
 	uint16_t 			xcc_mask;
+	enum amdgpu_memory_partition	mem_partition_mode;
 	uint32_t			num_xcc_per_xcp;
 	struct mutex			partition_mutex;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index 8fa3a1f3b1812..639b86c4d6640 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -97,6 +97,7 @@ struct amdgpu_nbio_funcs {
 	void (*clear_doorbell_interrupt)(struct amdgpu_device *adev);
 	u32 (*get_rom_offset)(struct amdgpu_device *adev);
 	u32 (*get_compute_partition_mode)(struct amdgpu_device *adev);
+	u32 (*get_memory_partition_mode)(struct amdgpu_device *adev);
 	void (*set_compute_partition_mode)(struct amdgpu_device *adev,
 					   enum amdgpu_gfx_partition mode);
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 6ed97371ff1e2..3c19c5a0069d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -606,6 +606,16 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 {
 	soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, xcc_id));
 }
+static enum amdgpu_memory_partition
+gfx_v9_4_3_query_memory_partition(struct amdgpu_device *adev)
+{
+	enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE;
+
+	if (adev->nbio.funcs->get_memory_partition_mode)
+		mode = adev->nbio.funcs->get_memory_partition_mode(adev);
+
+	return mode;
+}
 
 static enum amdgpu_gfx_partition
 gfx_v9_4_3_query_compute_partition(struct amdgpu_device *adev)
@@ -675,6 +685,7 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
 	.query_partition_mode = &gfx_v9_4_3_query_compute_partition,
 	.switch_partition_mode = &gfx_v9_4_3_switch_compute_partition,
+	.query_mem_partition_mode = &gfx_v9_4_3_query_memory_partition,
 };
 
 static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 78eab4d48e383..fa4b423c399b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -30,6 +30,8 @@
 #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h"
 #include <uapi/linux/kfd_ioctl.h>
 
+#define NPS_MODE_MASK 0x000000FFL
+
 static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev)
 {
 	WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL,
@@ -406,6 +408,15 @@ static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
 	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
 }
 
+static enum amdgpu_memory_partition nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev)
+{
+	u32 tmp;
+	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_STATUS);
+	tmp = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_MEM_STATUS, NPS_MODE);
+
+	return ffs(tmp);
+}
+
 const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_hdp_flush_req_offset = nbio_v7_9_get_hdp_flush_req_offset,
 	.get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset,
@@ -428,4 +439,5 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.remap_hdp_registers = nbio_v7_9_remap_hdp_registers,
 	.get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode,
 	.set_compute_partition_mode = nbio_v7_9_set_compute_partition_mode,
+	.get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode,
 };
-- 
GitLab


From d55391c2138c1a5bcd1316ccedf1835dd067d568 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 9 Nov 2022 19:47:38 +0530
Subject: [PATCH 1134/1791] drm/amdgpu: Revert programming of CP_PSP_XCP_CTL

Programming of this register is taken care by PSP. Incorrect programming
causes CP not to detect its XCC.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reported-by: Alexander Turek <Alexander.Turek@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 3c19c5a0069d4..9d7852ffd7084 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1055,9 +1055,6 @@ static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
 		tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
 		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, tmp);
 
-		tmp = xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, PHYSICAL_XCC_ID);
-		tmp = tmp | (xcc_id << REG_FIELD_SHIFT(CP_PSP_XCP_CTL, XCC_DIE_ID));
-		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_PSP_XCP_CTL, tmp);
 		break;
 	default:
 		break;
-- 
GitLab


From 4bc615a5237a641ddb2235236c21b1eaadb76963 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Sat, 12 Nov 2022 17:01:05 +0800
Subject: [PATCH 1135/1791] drm/amdgpu: fix vm context register assignment in
 mmhub v1.8

Assign the vm context register addr per aid instance.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 67338cb3d7bc3..6f469b9aa9a02 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -515,19 +515,19 @@ static void mmhub_v1_8_init(struct amdgpu_device *adev)
 	for (i = 0; i < adev->num_aid; i++) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
 
-		hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0,
+		hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, i,
 			regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32);
-		hub->ctx0_ptb_addr_hi32 = SOC15_REG_OFFSET(MMHUB, 0,
+		hub->ctx0_ptb_addr_hi32 = SOC15_REG_OFFSET(MMHUB, i,
 			regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32);
 		hub->vm_inv_eng0_req =
-			SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_REQ);
+			SOC15_REG_OFFSET(MMHUB, i, regVM_INVALIDATE_ENG0_REQ);
 		hub->vm_inv_eng0_ack =
-			SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ACK);
+			SOC15_REG_OFFSET(MMHUB, i, regVM_INVALIDATE_ENG0_ACK);
 		hub->vm_context0_cntl =
-			SOC15_REG_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL);
-		hub->vm_l2_pro_fault_status = SOC15_REG_OFFSET(MMHUB, 0,
+			SOC15_REG_OFFSET(MMHUB, i, regVM_CONTEXT0_CNTL);
+		hub->vm_l2_pro_fault_status = SOC15_REG_OFFSET(MMHUB, i,
 			regVM_L2_PROTECTION_FAULT_STATUS);
-		hub->vm_l2_pro_fault_cntl = SOC15_REG_OFFSET(MMHUB, 0,
+		hub->vm_l2_pro_fault_cntl = SOC15_REG_OFFSET(MMHUB, i,
 			regVM_L2_PROTECTION_FAULT_CNTL);
 
 		hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL;
-- 
GitLab


From cbd442ce91bdeb8f618511d65e674894a80e4c31 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Sat, 12 Nov 2022 16:30:40 +0530
Subject: [PATCH 1136/1791] drm/amdgpu: Skip runtime db read for PSP 13.0.6

Skip reading runtime db information for PSP 13.0.6.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index dd7f6d688449c..1319df5796cc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -329,6 +329,9 @@ static bool psp_get_runtime_db_entry(struct amdgpu_device *adev,
 	bool ret = false;
 	int i;
 
+	if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6))
+		return false;
+
 	db_header_pos = adev->gmc.mc_vram_size - PSP_RUNTIME_DB_OFFSET;
 	db_dir_pos = db_header_pos + sizeof(struct psp_runtime_data_header);
 
-- 
GitLab


From 37dd9d58a595f4cda5a7f01703592cc4e00f69b4 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Fri, 11 Nov 2022 15:54:52 +0800
Subject: [PATCH 1137/1791] drm/amdgpu: fix kcq mqd_backup buffer double free
 for multi-XCD

For gfx_v9_4_3 and beyond, struct kiq has its own mqd_backup pointer
rather than using the last pointer from mec struct. Then the kfree
operation on the pointer from the mec struct should be removed otherwise
it will cause double free on the first kcq's mqd_backup buffer on XCD1.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index c2e92664031b7..8aea9e0232755 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -480,7 +480,6 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id)
 
 	ring = &kiq->ring;
 	kfree(kiq->mqd_backup);
-	kfree(adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS]);
 	amdgpu_bo_free_kernel(&ring->mqd_obj,
 			      &ring->mqd_gpu_addr,
 			      &ring->mqd_ptr);
-- 
GitLab


From 993d218f82211b1e17fcea7a5f727fa16efba353 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 14 Nov 2022 15:52:19 +0800
Subject: [PATCH 1138/1791] drm/amdgpu: remove partition attributes sys file
 for gfx_v9_4_3

For driver de-init like rmmod operations those partition specific
attributes need to be removed accordingly.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 7 +++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 1 +
 3 files changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 8aea9e0232755..5ff49737d7c60 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1345,3 +1345,10 @@ int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev)
+{
+	device_remove_file(adev->dev, &dev_attr_current_compute_partition);
+	device_remove_file(adev->dev, &dev_attr_available_compute_partition);
+	device_remove_file(adev->dev, &dev_attr_current_memory_partition);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index e37501685fa94..023c5b08929ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -502,4 +502,5 @@ int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev,
 
 bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev);
+void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 9d7852ffd7084..75ad5176e99ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -895,6 +895,7 @@ static int gfx_v9_4_3_sw_fini(void *handle)
 	gfx_v9_4_3_mec_fini(adev);
 	amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj);
 	gfx_v9_4_3_free_microcode(adev);
+	amdgpu_gfx_sysfs_fini(adev);
 
 	return 0;
 }
-- 
GitLab


From eaae4beee8a94b30f37341c9d14837c82e7e2647 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Mon, 14 Nov 2022 17:35:43 -0500
Subject: [PATCH 1139/1791] drm/amdgpu: more GPU page fault info for GC v9.4.3

Output IH cookie node_id and translate it to the corresponding AID id
and XCC id, to help debug the GPU page fault.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index f2814270da401..2966aca9545d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -557,7 +557,9 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	u64 addr;
 	uint32_t cam_index = 0;
 	int ret;
-	uint32_t node_id = 0;
+	uint32_t node_id;
+
+	node_id = (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
 
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
@@ -570,8 +572,6 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
 	} else {
 		hub_name = "gfxhub0";
-		node_id = (adev->ip_versions[GC_HWIP][0] ==
-			   IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
 		hub = &adev->vmhub[node_id/2];
 	}
 
@@ -634,6 +634,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 		addr, entry->client_id,
 		soc15_ih_clientid_name[entry->client_id]);
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
+		dev_err(adev->dev, "  cookie node_id %d fault from die %s%d%s\n",
+			node_id, node_id % 4 == 3 ? "RSV" : "AID", node_id / 4,
+			node_id % 4 == 1 ? ".XCD0" : node_id % 4 == 2 ? ".XCD1" : "");
+
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-- 
GitLab


From 497db7ea33f7cec2a0019894e844789f003dbd22 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Tue, 8 Nov 2022 23:04:30 -0500
Subject: [PATCH 1140/1791] drm/amdgpu: Check APU supports true APP mode

On GPXIP 9.4.3 APU, in no carveout mode there is no real vram heap and
could be emulated by the driver over the interleaved NUMA system memory
and the APU could also  be in the carveout mode during early development
stage or otherwise for debugging purpose so introduce a new member in
amdgpu_gmc to figure out whether the APU is in the native mode as per
the production configuration. AMD_IS_APU cannot be used for Accelerated
Processing Platform APUs as it might be used in a different context on
previous generations or on small APUs.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Graham Sider <graham.sider@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h |  1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 14 ++++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 572ef5be539fc..e408abfc2daf3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -251,6 +251,7 @@ struct amdgpu_gmc {
 	uint64_t		last_fault:AMDGPU_GMC_FAULT_RING_ORDER;
 
 	bool tmz_enabled;
+	bool is_app_apu;
 
 	const struct amdgpu_gmc_funcs	*gmc_funcs;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 2966aca9545d6..0792c48fe347d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1442,6 +1442,20 @@ static int gmc_v9_0_early_init(void *handle)
 			adev->smuio.funcs->is_host_gpu_xgmi_supported(adev);
 	}
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		enum amdgpu_pkg_type pkg_type =
+			adev->smuio.funcs->get_pkg_type(adev);
+		/* On GFXIP 9.4.3. APU, there is no physical VRAM domain present
+		 * and the APU, can be in used two possible modes:
+		 *  - carveout mode
+		 *  - native APU mode
+		 * "is_app_apu" can be used to identify the APU in the native
+		 * mode.
+		 */
+		adev->gmc.is_app_apu = (pkg_type == AMDGPU_PKG_TYPE_APU &&
+					!pci_resource_len(adev->pdev, 0));
+	}
+
 	gmc_v9_0_set_gmc_funcs(adev);
 	gmc_v9_0_set_irq_funcs(adev);
 	gmc_v9_0_set_umc_funcs(adev);
-- 
GitLab


From 880f8b3f8032a022c85351857ba7020fd3345592 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 20 Jul 2022 13:45:30 +0530
Subject: [PATCH 1141/1791] drm/amdgpu: Rename xcc specific functions

Add 'xcc' prefix to xcc specific functions to distinguish from IP block
functions.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 205 +++++++++++++-----------
 1 file changed, 113 insertions(+), 92 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 75ad5176e99ea..93420c7db93c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -506,11 +506,8 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev,
-				    u32 se_num,
-				    u32 sh_num,
-				    u32 instance,
-				    int xcc_id)
+static void gfx_v9_4_3_xcc_select_se_sh(struct amdgpu_device *adev, u32 se_num,
+					u32 sh_num, u32 instance, int xcc_id)
 {
 	u32 data;
 
@@ -678,7 +675,7 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 
 static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter,
-	.select_se_sh = &gfx_v9_4_3_select_se_sh,
+	.select_se_sh = &gfx_v9_4_3_xcc_select_se_sh,
 	.read_wave_data = &gfx_v9_4_3_read_wave_data,
 	.read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs,
 	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
@@ -901,7 +898,8 @@ static int gfx_v9_4_3_sw_fini(void *handle)
 }
 
 #define DEFAULT_SH_MEM_BASES	(0x6000)
-static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_init_compute_vmid(struct amdgpu_device *adev,
+					     int xcc_id)
 {
 	int i;
 	uint32_t sh_mem_config;
@@ -939,7 +937,7 @@ static void gfx_v9_4_3_init_compute_vmid(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-static void gfx_v9_4_3_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 {
 	int vmid;
 
@@ -1000,25 +998,26 @@ static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
 	mutex_unlock(&adev->srbm_mutex);
 
 	for (i = 0; i < num_xcc; i++) {
-		gfx_v9_4_3_init_compute_vmid(adev, i);
-		gfx_v9_4_3_init_gds_vmid(adev, i);
+		gfx_v9_4_3_xcc_init_compute_vmid(adev, i);
+		gfx_v9_4_3_xcc_init_gds_vmid(adev, i);
 	}
 }
 
-static void gfx_v9_4_3_enable_save_restore_machine(struct amdgpu_device *adev,
-						   int xcc_id)
+static void
+gfx_v9_4_3_xcc_enable_save_restore_machine(struct amdgpu_device *adev,
+					   int xcc_id)
 {
 	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_SRM_CNTL, SRM_ENABLE, 1);
 }
 
-static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_init_pg(struct amdgpu_device *adev, int xcc_id)
 {
 	/*
 	 * Rlc save restore list is workable since v2_1.
 	 * And it's needed by gfxoff feature.
 	 */
 	if (adev->gfx.rlc.is_rlc_v2_1)
-		gfx_v9_4_3_enable_save_restore_machine(adev, xcc_id);
+		gfx_v9_4_3_xcc_enable_save_restore_machine(adev, xcc_id);
 
 	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
 			      AMD_PG_SUPPORT_GFX_SMG |
@@ -1031,7 +1030,7 @@ static void gfx_v9_4_3_init_pg(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 
@@ -1040,7 +1039,8 @@ static void gfx_v9_4_3_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
 	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG, data);
 }
 
-static void gfx_v9_4_3_program_xcc_id(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_program_xcc_id(struct amdgpu_device *adev,
+					  int xcc_id)
 {
 	uint32_t tmp = 0;
 	int num_xcc;
@@ -1074,7 +1074,7 @@ static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev)
 	return true;
 }
 
-static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 {
 	uint32_t data;
 	unsigned i;
@@ -1091,7 +1091,8 @@ static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev, int xcc_id)
+static void gfx_v9_4_3_xcc_unset_safe_mode(struct amdgpu_device *adev,
+					   int xcc_id)
 {
 	uint32_t data;
 
@@ -1108,8 +1109,8 @@ static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
-					   int xcc_id)
+static void gfx_v9_4_3_xcc_wait_for_rlc_serdes(struct amdgpu_device *adev,
+					       int xcc_id)
 {
 	u32 i, j, k;
 	u32 mask;
@@ -1117,16 +1118,17 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, xcc_id);
+			gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff,
+						    xcc_id);
 			for (k = 0; k < adev->usec_timeout; k++) {
 				if (RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
 				udelay(1);
 			}
 			if (k == adev->usec_timeout) {
-				gfx_v9_4_3_select_se_sh(adev, 0xffffffff,
-							0xffffffff, 0xffffffff,
-							xcc_id);
+				gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff,
+							    0xffffffff,
+							    0xffffffff, xcc_id);
 				mutex_unlock(&adev->grbm_idx_mutex);
 				DRM_INFO("Timeout wait for RLC serdes %u,%u\n",
 					 i, j);
@@ -1134,7 +1136,8 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 			}
 		}
 	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, xcc_id);
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+				    xcc_id);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK |
@@ -1148,8 +1151,8 @@ static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev,
 	}
 }
 
-static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev,
-						 bool enable, int xcc_id)
+static void gfx_v9_4_3_xcc_enable_gui_idle_interrupt(struct amdgpu_device *adev,
+						     bool enable, int xcc_id)
 {
 	u32 tmp;
 
@@ -1171,8 +1174,8 @@ static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), RLC_CNTL, RLC_ENABLE_F32, 0);
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
-		gfx_v9_4_3_wait_for_rlc_serdes(adev, i);
+		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, i);
+		gfx_v9_4_3_xcc_wait_for_rlc_serdes(adev, i);
 	}
 }
 
@@ -1203,7 +1206,7 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 
 		/* carrizo do enable cp interrupt after cp inited */
 		if (!(adev->flags & AMD_IS_APU)) {
-			gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
+			gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, i);
 			udelay(50);
 		}
 
@@ -1226,7 +1229,8 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 	}
 }
 
-static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev, int xcc_id)
+static int gfx_v9_4_3_xcc_rlc_load_microcode(struct amdgpu_device *adev,
+					     int xcc_id)
 {
 	const struct rlc_firmware_header_v2_0 *hdr;
 	const __le32 *fw_data;
@@ -1267,11 +1271,11 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 		/* disable CG */
 		WREG32_SOC15(GC, GET_INST(GC, i), regRLC_CGCG_CGLS_CTRL, 0);
 
-		gfx_v9_4_3_init_pg(adev, i);
+		gfx_v9_4_3_xcc_init_pg(adev, i);
 
 		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
 			/* legacy rlc firmware loading */
-			r = gfx_v9_4_3_rlc_load_microcode(adev, i);
+			r = gfx_v9_4_3_xcc_rlc_load_microcode(adev, i);
 			if (r)
 				return r;
 		}
@@ -1341,8 +1345,8 @@ static bool gfx_v9_4_3_is_rlcg_access_range(struct amdgpu_device *adev, u32 offs
 					ARRAY_SIZE(rlcg_access_gc_9_4_3));
 }
 
-static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev,
-					 bool enable, int xcc_id)
+static void gfx_v9_4_3_xcc_cp_compute_enable(struct amdgpu_device *adev,
+					     bool enable, int xcc_id)
 {
 	if (enable) {
 		WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MEC_CNTL, 0);
@@ -1354,8 +1358,8 @@ static void gfx_v9_4_3_cp_compute_enable(struct amdgpu_device *adev,
 	udelay(50);
 }
 
-static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev,
-						int xcc_id)
+static int gfx_v9_4_3_xcc_cp_compute_load_microcode(struct amdgpu_device *adev,
+						    int xcc_id)
 {
 	const struct gfx_firmware_header_v1_0 *mec_hdr;
 	const __le32 *fw_data;
@@ -1367,7 +1371,7 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev,
 	if (!adev->gfx.mec_fw)
 		return -EINVAL;
 
-	gfx_v9_4_3_cp_compute_enable(adev, false, xcc_id);
+	gfx_v9_4_3_xcc_cp_compute_enable(adev, false, xcc_id);
 
 	mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
 	amdgpu_ucode_print_gfx_hdr(&mec_hdr->header);
@@ -1403,7 +1407,7 @@ static int gfx_v9_4_3_cp_compute_load_microcode(struct amdgpu_device *adev,
 }
 
 /* KIQ functions */
-static void gfx_v9_4_3_kiq_setting(struct amdgpu_ring *ring, int xcc_id)
+static void gfx_v9_4_3_xcc_kiq_setting(struct amdgpu_ring *ring, int xcc_id)
 {
 	uint32_t tmp;
 	struct amdgpu_device *adev = ring->adev;
@@ -1558,7 +1562,8 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
+static int gfx_v9_4_3_xcc_kiq_init_register(struct amdgpu_ring *ring,
+					    int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
@@ -1663,7 +1668,8 @@ static int gfx_v9_4_3_kiq_init_register(struct amdgpu_ring *ring, int xcc_id)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring, int xcc_id)
+static int gfx_v9_4_3_xcc_kiq_fini_register(struct amdgpu_ring *ring,
+					    int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	int j;
@@ -1702,13 +1708,13 @@ static int gfx_v9_4_3_kiq_fini_register(struct amdgpu_ring *ring, int xcc_id)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
+static int gfx_v9_4_3_xcc_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
 	struct v9_mqd *tmp_mqd;
 
-	gfx_v9_4_3_kiq_setting(ring, xcc_id);
+	gfx_v9_4_3_xcc_kiq_setting(ring, xcc_id);
 
 	/* GPU could be in bad state during probe, driver trigger the reset
 	 * after load the SMU, in this case , the mqd is not be initialized.
@@ -1726,7 +1732,7 @@ static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		amdgpu_ring_clear_ring(ring);
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
-		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
+		gfx_v9_4_3_xcc_kiq_init_register(ring, xcc_id);
 		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
@@ -1736,7 +1742,7 @@ static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
 		gfx_v9_4_3_mqd_init(ring);
-		gfx_v9_4_3_kiq_init_register(ring, xcc_id);
+		gfx_v9_4_3_xcc_kiq_init_register(ring, xcc_id);
 		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 
@@ -1747,7 +1753,7 @@ static int gfx_v9_4_3_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 	return 0;
 }
 
-static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
+static int gfx_v9_4_3_xcc_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
@@ -1785,7 +1791,7 @@ static int gfx_v9_4_3_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 	return 0;
 }
 
-static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev, int xcc_id)
+static int gfx_v9_4_3_xcc_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring;
 	int r;
@@ -1802,7 +1808,7 @@ static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 		return r;
 	}
 
-	gfx_v9_4_3_kiq_init_queue(ring, xcc_id);
+	gfx_v9_4_3_xcc_kiq_init_queue(ring, xcc_id);
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
@@ -1810,12 +1816,12 @@ static int gfx_v9_4_3_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 	return 0;
 }
 
-static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev, int xcc_id)
+static int gfx_v9_4_3_xcc_kcq_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring = NULL;
 	int r = 0, i;
 
-	gfx_v9_4_3_cp_compute_enable(adev, true, xcc_id);
+	gfx_v9_4_3_xcc_cp_compute_enable(adev, true, xcc_id);
 
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
 		ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings];
@@ -1825,7 +1831,7 @@ static int gfx_v9_4_3_kcq_resume(struct amdgpu_device *adev, int xcc_id)
 			goto done;
 		r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
 		if (!r) {
-			r = gfx_v9_4_3_kcq_init_queue(ring, xcc_id);
+			r = gfx_v9_4_3_xcc_kcq_init_queue(ring, xcc_id);
 			amdgpu_bo_kunmap(ring->mqd_obj);
 			ring->mqd_ptr = NULL;
 		}
@@ -1846,12 +1852,12 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, false, i);
+		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, i);
 
 		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-			gfx_v9_4_3_disable_gpa_mode(adev, i);
+			gfx_v9_4_3_xcc_disable_gpa_mode(adev, i);
 
-			r = gfx_v9_4_3_cp_compute_load_microcode(adev, i);
+			r = gfx_v9_4_3_xcc_cp_compute_load_microcode(adev, i);
 			if (r)
 				return r;
 		}
@@ -1862,13 +1868,13 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 				adev, amdgpu_user_partt_mode);
 
 		/* set the virtual and physical id based on partition_mode */
-		gfx_v9_4_3_program_xcc_id(adev, i);
+		gfx_v9_4_3_xcc_program_xcc_id(adev, i);
 
-		r = gfx_v9_4_3_kiq_resume(adev, i);
+		r = gfx_v9_4_3_xcc_kiq_resume(adev, i);
 		if (r)
 			return r;
 
-		r = gfx_v9_4_3_kcq_resume(adev, i);
+		r = gfx_v9_4_3_xcc_kcq_resume(adev, i);
 		if (r)
 			return r;
 
@@ -1879,16 +1885,16 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 				return r;
 		}
 
-		gfx_v9_4_3_enable_gui_idle_interrupt(adev, true, i);
+		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, i);
 	}
 
 	return 0;
 }
 
-static void gfx_v9_4_3_cp_enable(struct amdgpu_device *adev, bool enable,
-				int xcc_id)
+static void gfx_v9_4_3_xcc_cp_enable(struct amdgpu_device *adev, bool enable,
+				     int xcc_id)
 {
-	gfx_v9_4_3_cp_compute_enable(adev, enable, xcc_id);
+	gfx_v9_4_3_xcc_cp_compute_enable(adev, enable, xcc_id);
 }
 
 static int gfx_v9_4_3_hw_init(void *handle)
@@ -1932,12 +1938,13 @@ static int gfx_v9_4_3_hw_fini(void *handle)
 			soc15_grbm_select(adev, adev->gfx.kiq[i].ring.me,
 					adev->gfx.kiq[i].ring.pipe,
 					adev->gfx.kiq[i].ring.queue, 0, GET_INST(GC, i));
-			gfx_v9_4_3_kiq_fini_register(&adev->gfx.kiq[i].ring, i);
+			gfx_v9_4_3_xcc_kiq_fini_register(&adev->gfx.kiq[i].ring,
+							 i);
 			soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, i));
 			mutex_unlock(&adev->srbm_mutex);
 		}
 
-		gfx_v9_4_3_cp_enable(adev, false, i);
+		gfx_v9_4_3_xcc_cp_enable(adev, false, i);
 	}
 
 	/* Skip suspend with A+A reset */
@@ -2024,7 +2031,7 @@ static int gfx_v9_4_3_soft_reset(void *handle)
 		adev->gfx.rlc.funcs->stop(adev);
 
 		/* Disable MEC parsing/prefetching */
-		gfx_v9_4_3_cp_compute_enable(adev, false, 0);
+		gfx_v9_4_3_xcc_cp_compute_enable(adev, false, 0);
 
 		if (grbm_soft_reset) {
 			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
@@ -2111,8 +2118,9 @@ static int gfx_v9_4_3_late_init(void *handle)
 	return 0;
 }
 
-static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *adev,
-						      bool enable, int xcc_id)
+static void
+gfx_v9_4_3_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev,
+						bool enable, int xcc_id)
 {
 	uint32_t data, def;
 
@@ -2180,8 +2188,9 @@ static void gfx_v9_4_3_update_medium_grain_clock_gating(struct amdgpu_device *ad
 	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
-static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
-						      bool enable, int xcc_id)
+static void
+gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
+						bool enable, int xcc_id)
 {
 	uint32_t def, data;
 
@@ -2232,31 +2241,35 @@ static void gfx_v9_4_3_update_coarse_grain_clock_gating(struct amdgpu_device *ad
 	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
-static int gfx_v9_4_3_update_gfx_clock_gating(struct amdgpu_device *adev,
-					    bool enable, int xcc_id)
+static int gfx_v9_4_3_xcc_update_gfx_clock_gating(struct amdgpu_device *adev,
+						  bool enable, int xcc_id)
 {
 	if (enable) {
 		/* CGCG/CGLS should be enabled after MGCG/MGLS
 		 * ===  MGCG + MGLS ===
 		 */
-		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_medium_grain_clock_gating(adev, enable,
+								xcc_id);
 		/* ===  CGCG + CGLS === */
-		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(adev, enable,
+								xcc_id);
 	} else {
 		/* CGCG/CGLS should be disabled before MGCG/MGLS
 		 * ===  CGCG + CGLS ===
 		 */
-		gfx_v9_4_3_update_coarse_grain_clock_gating(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(adev, enable,
+								xcc_id);
 		/* ===  MGCG + MGLS === */
-		gfx_v9_4_3_update_medium_grain_clock_gating(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_medium_grain_clock_gating(adev, enable,
+								xcc_id);
 	}
 	return 0;
 }
 
 static const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = {
 	.is_rlc_enabled = gfx_v9_4_3_is_rlc_enabled,
-	.set_safe_mode = gfx_v9_4_3_set_safe_mode,
-	.unset_safe_mode = gfx_v9_4_3_unset_safe_mode,
+	.set_safe_mode = gfx_v9_4_3_xcc_set_safe_mode,
+	.unset_safe_mode = gfx_v9_4_3_xcc_unset_safe_mode,
 	.init = gfx_v9_4_3_rlc_init,
 	.resume = gfx_v9_4_3_rlc_resume,
 	.stop = gfx_v9_4_3_rlc_stop,
@@ -2285,8 +2298,8 @@ static int gfx_v9_4_3_set_clockgating_state(void *handle,
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 3):
 		for (i = 0; i < num_xcc; i++)
-			gfx_v9_4_3_update_gfx_clock_gating(adev,
-						state == AMD_CG_STATE_GATE, i);
+			gfx_v9_4_3_xcc_update_gfx_clock_gating(
+				adev, state == AMD_CG_STATE_GATE, i);
 		break;
 	default:
 		break;
@@ -2553,10 +2566,9 @@ static void gfx_v9_4_3_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring,
 						   ref, mask);
 }
 
-static void gfx_v9_4_3_set_compute_eop_interrupt_state(struct amdgpu_device *adev,
-						       int me, int pipe,
-						       enum amdgpu_interrupt_state state,
-						       int xcc_id)
+static void gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+	struct amdgpu_device *adev, int me, int pipe,
+	enum amdgpu_interrupt_state state, int xcc_id)
 {
 	u32 mec_int_cntl, mec_int_cntl_reg;
 
@@ -2664,28 +2676,36 @@ static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
 	for (i = 0; i < num_xcc; i++) {
 		switch (type) {
 		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 0, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 1, 0, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 1, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 1, 1, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 2, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 1, 2, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 1, 3, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 1, 3, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 0, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 2, 0, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 1, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 2, 1, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 2, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 2, 2, state, i);
 			break;
 		case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP:
-			gfx_v9_4_3_set_compute_eop_interrupt_state(adev, 2, 3, state, i);
+			gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
+				adev, 2, 3, state, i);
 			break;
 		default:
 			break;
@@ -3090,7 +3110,7 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
 			mask = 1;
 			ao_bitmap = 0;
 			counter = 0;
-			gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff, 0);
+			gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff, 0);
 			gfx_v9_4_3_set_user_cu_inactive_bitmap(
 				adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]);
 			bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev);
@@ -3123,7 +3143,8 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev,
 			cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap;
 		}
 	}
-	gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0);
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+				    0);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	cu_info->number = active_cu_number;
-- 
GitLab


From 44b5cf2e0f7952856f48b9be56b9eb2f688d70f0 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 16 Nov 2022 10:47:18 +0530
Subject: [PATCH 1142/1791] drm/amdgpu: Add xcc specific functions

Add more XCC specific functions and use them from IP block functions.
RLC, CP functions are further split to have xcc specific versions.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 295 ++++++++++++++----------
 1 file changed, 176 insertions(+), 119 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 93420c7db93c2..93a0baa4515c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -955,52 +955,64 @@ static void gfx_v9_4_3_xcc_init_gds_vmid(struct amdgpu_device *adev, int xcc_id)
 	}
 }
 
-static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
+static void gfx_v9_4_3_xcc_constants_init(struct amdgpu_device *adev,
+					  int xcc_id)
 {
 	u32 tmp;
-	int i, j, num_xcc;
-
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-
-	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
-	adev->gfx.config.db_debug2 = RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2);
+	int i;
 
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
 	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) {
-		for (j = 0; j < num_xcc; j++) {
-			soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, j));
-			/* CP and shaders */
-			if (i == 0) {
-				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
-						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
-				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-						    !!adev->gmc.noretry);
-				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_CONFIG, tmp);
-				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_BASES, 0);
-			} else {
-				tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
-						    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
-				tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
-						    !!adev->gmc.noretry);
-				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_CONFIG, tmp);
-				tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
-					(adev->gmc.private_aperture_start >> 48));
-				tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
-					(adev->gmc.shared_aperture_start >> 48));
-				WREG32_SOC15_RLC(GC, GET_INST(GC, j), regSH_MEM_BASES, tmp);
-			}
+		soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id));
+		/* CP and shaders */
+		if (i == 0) {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+					    !!adev->gmc.noretry);
+			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id),
+					 regSH_MEM_CONFIG, tmp);
+			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id),
+					 regSH_MEM_BASES, 0);
+		} else {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE,
+					    !!adev->gmc.noretry);
+			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id),
+					 regSH_MEM_CONFIG, tmp);
+			tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE,
+					    (adev->gmc.private_aperture_start >>
+					     48));
+			tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE,
+					    (adev->gmc.shared_aperture_start >>
+					     48));
+			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id),
+					 regSH_MEM_BASES, tmp);
 		}
 	}
 	soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, 0));
 
 	mutex_unlock(&adev->srbm_mutex);
 
-	for (i = 0; i < num_xcc; i++) {
-		gfx_v9_4_3_xcc_init_compute_vmid(adev, i);
-		gfx_v9_4_3_xcc_init_gds_vmid(adev, i);
-	}
+	gfx_v9_4_3_xcc_init_compute_vmid(adev, xcc_id);
+	gfx_v9_4_3_xcc_init_gds_vmid(adev, xcc_id);
+}
+
+static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev)
+{
+	int i, num_xcc;
+
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+
+	gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info);
+	adev->gfx.config.db_debug2 =
+		RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2);
+
+	for (i = 0; i < num_xcc; i++)
+		gfx_v9_4_3_xcc_constants_init(adev, i);
 }
 
 static void
@@ -1167,16 +1179,31 @@ static void gfx_v9_4_3_xcc_enable_gui_idle_interrupt(struct amdgpu_device *adev,
 	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0, tmp);
 }
 
+static void gfx_v9_4_3_xcc_rlc_stop(struct amdgpu_device *adev, int xcc_id)
+{
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_CNTL,
+			      RLC_ENABLE_F32, 0);
+	gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, xcc_id);
+	gfx_v9_4_3_xcc_wait_for_rlc_serdes(adev, xcc_id);
+}
+
 static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev)
 {
 	int i, num_xcc;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), RLC_CNTL, RLC_ENABLE_F32, 0);
-		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, i);
-		gfx_v9_4_3_xcc_wait_for_rlc_serdes(adev, i);
-	}
+	for (i = 0; i < num_xcc; i++)
+		gfx_v9_4_3_xcc_rlc_stop(adev, i);
+}
+
+static void gfx_v9_4_3_xcc_rlc_reset(struct amdgpu_device *adev, int xcc_id)
+{
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET,
+			      SOFT_RESET_RLC, 1);
+	udelay(50);
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET,
+			      SOFT_RESET_RLC, 0);
+	udelay(50);
 }
 
 static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev)
@@ -1184,10 +1211,19 @@ static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev)
 	int i, num_xcc;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_SOFT_RESET, SOFT_RESET_RLC, 1);
-		udelay(50);
-		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), GRBM_SOFT_RESET, SOFT_RESET_RLC, 0);
+	for (i = 0; i < num_xcc; i++)
+		gfx_v9_4_3_xcc_rlc_reset(adev, i);
+}
+
+static void gfx_v9_4_3_xcc_rlc_start(struct amdgpu_device *adev, int xcc_id)
+{
+	WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_CNTL,
+			      RLC_ENABLE_F32, 1);
+	udelay(50);
+
+	/* carrizo do enable cp interrupt after cp inited */
+	if (!(adev->flags & AMD_IS_APU)) {
+		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, xcc_id);
 		udelay(50);
 	}
 }
@@ -1201,15 +1237,7 @@ static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), RLC_CNTL, RLC_ENABLE_F32, 1);
-		udelay(50);
-
-		/* carrizo do enable cp interrupt after cp inited */
-		if (!(adev->flags & AMD_IS_APU)) {
-			gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, i);
-			udelay(50);
-		}
-
+		gfx_v9_4_3_xcc_rlc_start(adev, i);
 #ifdef AMDGPU_RLC_DEBUG_RETRY
 		/* RLC_GPM_GENERAL_6 : RLC Ucode version */
 		rlc_ucode_ver = RREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_GENERAL_6);
@@ -1260,28 +1288,39 @@ static int gfx_v9_4_3_xcc_rlc_load_microcode(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
+static int gfx_v9_4_3_xcc_rlc_resume(struct amdgpu_device *adev, int xcc_id)
 {
-	int r, i, num_xcc;
+	int r;
 
-	adev->gfx.rlc.funcs->stop(adev);
+	gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id);
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		/* disable CG */
-		WREG32_SOC15(GC, GET_INST(GC, i), regRLC_CGCG_CGLS_CTRL, 0);
+	/* disable CG */
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, 0);
 
-		gfx_v9_4_3_xcc_init_pg(adev, i);
+	gfx_v9_4_3_xcc_init_pg(adev, xcc_id);
 
-		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-			/* legacy rlc firmware loading */
-			r = gfx_v9_4_3_xcc_rlc_load_microcode(adev, i);
-			if (r)
-				return r;
-		}
+	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+		/* legacy rlc firmware loading */
+		r = gfx_v9_4_3_xcc_rlc_load_microcode(adev, xcc_id);
+		if (r)
+			return r;
 	}
 
-	adev->gfx.rlc.funcs->start(adev);
+	gfx_v9_4_3_xcc_rlc_start(adev, xcc_id);
+
+	return 0;
+}
+
+static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
+{
+	int r, i, num_xcc;
+
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
+		r = gfx_v9_4_3_xcc_rlc_resume(adev, i);
+		if (r)
+			return r;
+	}
 
 	return 0;
 }
@@ -1845,47 +1884,58 @@ static int gfx_v9_4_3_xcc_kcq_resume(struct amdgpu_device *adev, int xcc_id)
 	return r;
 }
 
-static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
+static int gfx_v9_4_3_xcc_cp_resume(struct amdgpu_device *adev, int xcc_id)
 {
-	int r, i, j, num_xcc;
 	struct amdgpu_ring *ring;
+	int r, j;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, i);
+	gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, xcc_id);
 
-		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
-			gfx_v9_4_3_xcc_disable_gpa_mode(adev, i);
+	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+		gfx_v9_4_3_xcc_disable_gpa_mode(adev, xcc_id);
 
-			r = gfx_v9_4_3_xcc_cp_compute_load_microcode(adev, i);
-			if (r)
-				return r;
-		}
+		r = gfx_v9_4_3_xcc_cp_compute_load_microcode(adev, xcc_id);
+		if (r)
+			return r;
+	}
 
-		if (adev->gfx.partition_mode ==
-		    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
-			gfx_v9_4_3_switch_compute_partition(
-				adev, amdgpu_user_partt_mode);
+	if (adev->gfx.partition_mode == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
+		gfx_v9_4_3_switch_compute_partition(adev,
+						    amdgpu_user_partt_mode);
 
-		/* set the virtual and physical id based on partition_mode */
-		gfx_v9_4_3_xcc_program_xcc_id(adev, i);
+	/* set the virtual and physical id based on partition_mode */
+	gfx_v9_4_3_xcc_program_xcc_id(adev, xcc_id);
 
-		r = gfx_v9_4_3_xcc_kiq_resume(adev, i);
-		if (r)
-			return r;
+	r = gfx_v9_4_3_xcc_kiq_resume(adev, xcc_id);
+	if (r)
+		return r;
+
+	r = gfx_v9_4_3_xcc_kcq_resume(adev, xcc_id);
+	if (r)
+		return r;
 
-		r = gfx_v9_4_3_xcc_kcq_resume(adev, i);
+	for (j = 0; j < adev->gfx.num_compute_rings; j++) {
+		ring = &adev->gfx.compute_ring
+				[j + xcc_id * adev->gfx.num_compute_rings];
+		r = amdgpu_ring_test_helper(ring);
 		if (r)
 			return r;
+	}
 
-		for (j = 0; j < adev->gfx.num_compute_rings; j++) {
-			ring = &adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings];
-			r = amdgpu_ring_test_helper(ring);
-			if (r)
-				return r;
-		}
+	gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, xcc_id);
+
+	return 0;
+}
+
+static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
+{
+	int r, i, num_xcc;
 
-		gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, i);
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	for (i = 0; i < num_xcc; i++) {
+		r = gfx_v9_4_3_xcc_cp_resume(adev, i);
+		if (r)
+			return r;
 	}
 
 	return 0;
@@ -1897,6 +1947,37 @@ static void gfx_v9_4_3_xcc_cp_enable(struct amdgpu_device *adev, bool enable,
 	gfx_v9_4_3_xcc_cp_compute_enable(adev, enable, xcc_id);
 }
 
+static void gfx_v9_4_3_xcc_fini(struct amdgpu_device *adev, int xcc_id)
+{
+	if (amdgpu_gfx_disable_kcq(adev, xcc_id))
+		DRM_ERROR("XCD %d KCQ disable failed\n", xcc_id);
+
+	/* Use deinitialize sequence from CAIL when unbinding device
+	 * from driver, otherwise KIQ is hanging when binding back
+	 */
+	if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
+		mutex_lock(&adev->srbm_mutex);
+		soc15_grbm_select(adev, adev->gfx.kiq[xcc_id].ring.me,
+				  adev->gfx.kiq[xcc_id].ring.pipe,
+				  adev->gfx.kiq[xcc_id].ring.queue, 0,
+				  GET_INST(GC, xcc_id));
+		gfx_v9_4_3_xcc_kiq_fini_register(&adev->gfx.kiq[xcc_id].ring,
+						 xcc_id);
+		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
+		mutex_unlock(&adev->srbm_mutex);
+	}
+
+	gfx_v9_4_3_xcc_cp_enable(adev, false, xcc_id);
+
+	/* Skip suspend with A+A reset */
+	if (adev->gmc.xgmi.connected_to_cpu && amdgpu_in_reset(adev)) {
+		dev_dbg(adev->dev, "Device in reset. Skipping RLC halt\n");
+		return;
+	}
+
+	gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id);
+}
+
 static int gfx_v9_4_3_hw_init(void *handle)
 {
 	int r;
@@ -1927,33 +2008,9 @@ static int gfx_v9_4_3_hw_fini(void *handle)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-		if (amdgpu_gfx_disable_kcq(adev, i))
-			DRM_ERROR("XCD %d KCQ disable failed\n", i);
-
-		/* Use deinitialize sequence from CAIL when unbinding device
-		 * from driver, otherwise KIQ is hanging when binding back
-		 */
-		if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
-			mutex_lock(&adev->srbm_mutex);
-			soc15_grbm_select(adev, adev->gfx.kiq[i].ring.me,
-					adev->gfx.kiq[i].ring.pipe,
-					adev->gfx.kiq[i].ring.queue, 0, GET_INST(GC, i));
-			gfx_v9_4_3_xcc_kiq_fini_register(&adev->gfx.kiq[i].ring,
-							 i);
-			soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, i));
-			mutex_unlock(&adev->srbm_mutex);
-		}
-
-		gfx_v9_4_3_xcc_cp_enable(adev, false, i);
-	}
-
-	/* Skip suspend with A+A reset */
-	if (adev->gmc.xgmi.connected_to_cpu && amdgpu_in_reset(adev)) {
-		dev_dbg(adev->dev, "Device in reset. Skipping RLC halt\n");
-		return 0;
+		gfx_v9_4_3_xcc_fini(adev, i);
 	}
 
-	adev->gfx.rlc.funcs->stop(adev);
 	return 0;
 }
 
-- 
GitLab


From dd1a02e280dae1904c8858c8cb572a61a84ba7c0 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 14 Sep 2022 12:16:48 +0530
Subject: [PATCH 1143/1791] drm/amdgpu: Add xcc specific functions for gfxhub

GFXHUB 1.2 supports multiple XCC instances. Add XCC specific functions
to handle XCC instances separately.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      |   4 +
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 202 ++++++++++++++---------
 2 files changed, 128 insertions(+), 78 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 4c4ce33c83598..c2dd100bbd60f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1263,6 +1263,10 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 
 #define amdgpu_inc_vram_lost(adev) atomic_inc(&((adev)->vram_lost_counter));
 
+#define for_each_inst(i, inst_mask)                                            \
+	for (i = ffs(inst_mask) - 1; inst_mask;                                \
+	     inst_mask &= ~(1U << i), i = ffs(inst_mask) - 1)
+
 #define MIN(X, Y) ((X) < (Y) ? (X) : (Y))
 
 /* Common functions */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index c26ac0662c7e8..b9316bfb7263b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -38,15 +38,15 @@ static u64 gfxhub_v1_2_get_mc_fb_offset(struct amdgpu_device *adev)
 	return (u64)RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_FB_OFFSET) << 24;
 }
 
-static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
-					 uint32_t vmid,
-					 uint64_t page_table_base)
+static void gfxhub_v1_2_xcc_setup_vm_pt_regs(struct amdgpu_device *adev,
+					     uint32_t vmid,
+					     uint64_t page_table_base,
+					     uint32_t xcc_mask)
 {
 	struct amdgpu_vmhub *hub;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 		WREG32_SOC15_OFFSET(GC, GET_INST(GC, i),
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
@@ -57,27 +57,36 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
 				    hub->ctx_addr_distance * vmid,
 				    upper_32_bits(page_table_base));
-
 	}
 }
 
-static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
+static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev,
+					 uint32_t vmid,
+					 uint64_t page_table_base)
+{
+	uint32_t xcc_mask;
+
+	xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0);
+	gfxhub_v1_2_xcc_setup_vm_pt_regs(adev, vmid, page_table_base, xcc_mask);
+}
+
+static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev,
+						    uint32_t xcc_mask)
 {
 	uint64_t pt_base;
-	int i, num_xcc;
+	int i;
 
 	if (adev->gmc.pdb0_bo)
 		pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo);
 	else
 		pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
 
-	gfxhub_v1_2_setup_vm_pt_regs(adev, 0, pt_base);
+	gfxhub_v1_2_xcc_setup_vm_pt_regs(adev, 0, pt_base, xcc_mask);
 
 	/* If use GART for FB translation, vmid0 page table covers both
 	 * vram and system memory (gart)
 	 */
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		if (adev->gmc.pdb0_bo) {
 			WREG32_SOC15(GC, GET_INST(GC, i),
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
@@ -110,14 +119,15 @@ static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
+static void
+gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev,
+					  uint32_t xcc_mask)
 {
 	uint64_t value;
 	uint32_t tmp;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		/* Program the AGP BAR */
 		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BASE, 0);
 		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24);
@@ -178,13 +188,13 @@ static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
+static void gfxhub_v1_2_xcc_init_tlb_regs(struct amdgpu_device *adev,
+					  uint32_t xcc_mask)
 {
 	uint32_t tmp;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		/* Setup TLB control */
 		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_MX_L1_TLB_CNTL);
 
@@ -204,13 +214,13 @@ static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
+static void gfxhub_v1_2_xcc_init_cache_regs(struct amdgpu_device *adev,
+					    uint32_t xcc_mask)
 {
 	uint32_t tmp;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		/* Setup L2 cache */
 		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
@@ -252,13 +262,13 @@ static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
+static void gfxhub_v1_2_xcc_enable_system_domain(struct amdgpu_device *adev,
+						 uint32_t xcc_mask)
 {
 	uint32_t tmp;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
@@ -271,12 +281,13 @@ static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
+static void
+gfxhub_v1_2_xcc_disable_identity_aperture(struct amdgpu_device *adev,
+					  uint32_t xcc_mask)
 {
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		WREG32_SOC15(GC, GET_INST(GC, i),
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
 			     0XFFFFFFFF);
@@ -298,12 +309,13 @@ static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
+static void gfxhub_v1_2_xcc_setup_vmid_config(struct amdgpu_device *adev,
+					      uint32_t xcc_mask)
 {
 	struct amdgpu_vmhub *hub;
 	unsigned num_level, block_size;
 	uint32_t tmp;
-	int i, j, num_xcc;
+	int i, j;
 
 	num_level = adev->vm_manager.num_level;
 	block_size = adev->vm_manager.block_size;
@@ -312,8 +324,7 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	else
 		block_size -= 9;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (j = 0; j < num_xcc; j++) {
+	for_each_inst(j, xcc_mask) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		for (i = 0; i <= 14; i++) {
 			tmp = RREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT1_CNTL, i);
@@ -368,13 +379,13 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev)
 	}
 }
 
-static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
+static void gfxhub_v1_2_xcc_program_invalidation(struct amdgpu_device *adev,
+						 uint32_t xcc_mask)
 {
 	struct amdgpu_vmhub *hub;
-	unsigned i, j, num_xcc;
+	unsigned int i, j;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (j = 0; j < num_xcc; j++) {
+	for_each_inst(j, xcc_mask) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 
 		for (i = 0 ; i < 18; ++i) {
@@ -386,18 +397,21 @@ static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev)
 	}
 }
 
-static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
+static int gfxhub_v1_2_xcc_gart_enable(struct amdgpu_device *adev,
+				       uint32_t xcc_mask)
 {
-	int i, num_xcc;
-
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
-		if (amdgpu_sriov_vf(adev)) {
-		/*
-		 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are
-		 * VF copy registers so vbios post doesn't program them, for
-		 * SRIOV driver need to program them
-		 */
+	uint32_t tmp_mask;
+	int i;
+
+	tmp_mask = xcc_mask;
+	/*
+	 * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, because they are
+	 * VF copy registers so vbios post doesn't program them, for
+	 * SRIOV driver need to program them
+	 */
+	if (amdgpu_sriov_vf(adev)) {
+		for_each_inst(i, tmp_mask) {
+			i = ffs(tmp_mask) - 1;
 			WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE,
 				     adev->gmc.vram_start >> 24);
 			WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP,
@@ -406,29 +420,37 @@ static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
 	}
 
 	/* GART Enable. */
-	gfxhub_v1_2_init_gart_aperture_regs(adev);
-	gfxhub_v1_2_init_system_aperture_regs(adev);
-	gfxhub_v1_2_init_tlb_regs(adev);
+	gfxhub_v1_2_xcc_init_gart_aperture_regs(adev, xcc_mask);
+	gfxhub_v1_2_xcc_init_system_aperture_regs(adev, xcc_mask);
+	gfxhub_v1_2_xcc_init_tlb_regs(adev, xcc_mask);
 	if (!amdgpu_sriov_vf(adev))
-		gfxhub_v1_2_init_cache_regs(adev);
+		gfxhub_v1_2_xcc_init_cache_regs(adev, xcc_mask);
 
-	gfxhub_v1_2_enable_system_domain(adev);
+	gfxhub_v1_2_xcc_enable_system_domain(adev, xcc_mask);
 	if (!amdgpu_sriov_vf(adev))
-		gfxhub_v1_2_disable_identity_aperture(adev);
-	gfxhub_v1_2_setup_vmid_config(adev);
-	gfxhub_v1_2_program_invalidation(adev);
+		gfxhub_v1_2_xcc_disable_identity_aperture(adev, xcc_mask);
+	gfxhub_v1_2_xcc_setup_vmid_config(adev, xcc_mask);
+	gfxhub_v1_2_xcc_program_invalidation(adev, xcc_mask);
 
 	return 0;
 }
 
-static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
+static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev)
+{
+	uint32_t xcc_mask;
+
+	xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0);
+	return gfxhub_v1_2_xcc_gart_enable(adev, xcc_mask);
+}
+
+static void gfxhub_v1_2_xcc_gart_disable(struct amdgpu_device *adev,
+					 uint32_t xcc_mask)
 {
 	struct amdgpu_vmhub *hub;
 	u32 tmp;
-	u32 i, j, num_xcc;
+	u32 i, j;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (j = 0; j < num_xcc; j++) {
+	for_each_inst(j, xcc_mask) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(j)];
 		/* Disable all tables */
 		for (i = 0; i < 16; i++)
@@ -452,20 +474,22 @@ static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
 	}
 }
 
-/**
- * gfxhub_v1_2_set_fault_enable_default - update GART/VM fault handling
- *
- * @adev: amdgpu_device pointer
- * @value: true redirects VM faults to the default page
- */
-static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
-						 bool value)
+static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev)
+{
+	uint32_t xcc_mask;
+
+	xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0);
+	gfxhub_v1_2_xcc_gart_disable(adev, xcc_mask);
+}
+
+static void gfxhub_v1_2_xcc_set_fault_enable_default(struct amdgpu_device *adev,
+						     bool value,
+						     uint32_t xcc_mask)
 {
 	u32 tmp;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
@@ -501,13 +525,27 @@ static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
 	}
 }
 
-static void gfxhub_v1_2_init(struct amdgpu_device *adev)
+/**
+ * gfxhub_v1_2_set_fault_enable_default - update GART/VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev,
+						 bool value)
+{
+	uint32_t xcc_mask;
+
+	xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0);
+	gfxhub_v1_2_xcc_set_fault_enable_default(adev, value, xcc_mask);
+}
+
+static void gfxhub_v1_2_xcc_init(struct amdgpu_device *adev, uint32_t xcc_mask)
 {
 	struct amdgpu_vmhub *hub;
-	int i, num_xcc;
+	int i;
 
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	for (i = 0; i < num_xcc; i++) {
+	for_each_inst(i, xcc_mask) {
 		hub = &adev->vmhub[AMDGPU_GFXHUB(i)];
 
 		hub->ctx0_ptb_addr_lo32 =
@@ -543,6 +581,14 @@ static void gfxhub_v1_2_init(struct amdgpu_device *adev)
 	}
 }
 
+static void gfxhub_v1_2_init(struct amdgpu_device *adev)
+{
+	uint32_t xcc_mask;
+
+	xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0);
+	gfxhub_v1_2_xcc_init(adev, xcc_mask);
+}
+
 static int gfxhub_v1_2_get_xgmi_info(struct amdgpu_device *adev)
 {
 	u32 max_num_physical_nodes;
-- 
GitLab


From 527c670e5323414dbef8f4719dc9b348a50ac1c8 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 14 Sep 2022 12:48:08 +0530
Subject: [PATCH 1144/1791] drm/amdgpu: Add sdma instance specific functions

SDMA 4.4.2 supports multiple instances. Add functions to support
handling of each SDMA instance separately.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 162 +++++++++++++----------
 1 file changed, 91 insertions(+), 71 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 016813b295ac0..6a971e15b4b06 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -94,7 +94,8 @@ static int sdma_v4_4_2_irq_id_to_seq(unsigned client_id)
 	}
 }
 
-static void sdma_v4_4_2_init_golden_registers(struct amdgpu_device *adev)
+static void sdma_v4_4_2_inst_init_golden_registers(struct amdgpu_device *adev,
+						   uint32_t inst_mask)
 {
 	u32 val;
 	int i;
@@ -418,13 +419,14 @@ static void sdma_v4_4_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
  *
  * Stop the gfx async dma ring buffers.
  */
-static void sdma_v4_4_2_gfx_stop(struct amdgpu_device *adev)
+static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev,
+				      uint32_t inst_mask)
 {
 	struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES];
 	u32 rb_cntl, ib_cntl;
 	int i, unset = 0;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		sdma[i] = &adev->sdma.instance[i].ring;
 
 		if ((adev->mman.buffer_funcs_ring == sdma[i]) && unset != 1) {
@@ -448,7 +450,8 @@ static void sdma_v4_4_2_gfx_stop(struct amdgpu_device *adev)
  *
  * Stop the compute async dma queues.
  */
-static void sdma_v4_4_2_rlc_stop(struct amdgpu_device *adev)
+static void sdma_v4_4_2_inst_rlc_stop(struct amdgpu_device *adev,
+				      uint32_t inst_mask)
 {
 	/* XXX todo */
 }
@@ -460,14 +463,15 @@ static void sdma_v4_4_2_rlc_stop(struct amdgpu_device *adev)
  *
  * Stop the page async dma ring buffers.
  */
-static void sdma_v4_4_2_page_stop(struct amdgpu_device *adev)
+static void sdma_v4_4_2_inst_page_stop(struct amdgpu_device *adev,
+				       uint32_t inst_mask)
 {
 	struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES];
 	u32 rb_cntl, ib_cntl;
 	int i;
 	bool unset = false;
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		sdma[i] = &adev->sdma.instance[i].page;
 
 		if ((adev->mman.buffer_funcs_ring == sdma[i]) &&
@@ -495,7 +499,8 @@ static void sdma_v4_4_2_page_stop(struct amdgpu_device *adev)
  *
  * Halt or unhalt the async dma engines context switch.
  */
-static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
+static void sdma_v4_4_2_inst_ctx_switch_enable(struct amdgpu_device *adev,
+					       bool enable, uint32_t inst_mask)
 {
 	u32 f32_cntl, phase_quantum = 0;
 	int i;
@@ -524,7 +529,7 @@ static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enabl
 			unit  << SDMA_PHASE0_QUANTUM__UNIT__SHIFT;
 	}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		f32_cntl = RREG32_SDMA(i, regSDMA_CNTL);
 		f32_cntl = REG_SET_FIELD(f32_cntl, SDMA_CNTL,
 				AUTO_CTXSW_ENABLE, enable ? 1 : 0);
@@ -538,7 +543,6 @@ static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enabl
 		/* Extend page fault timeout to avoid interrupt storm */
 		WREG32_SDMA(i, regSDMA_UTCL1_TIMEOUT, 0x00800080);
 	}
-
 }
 
 /**
@@ -546,22 +550,24 @@ static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enabl
  *
  * @adev: amdgpu_device pointer
  * @enable: enable/disable the DMA MEs.
+ * @inst_mask: mask of dma engine instances to be enabled
  *
  * Halt or unhalt the async dma engines.
  */
-static void sdma_v4_4_2_enable(struct amdgpu_device *adev, bool enable)
+static void sdma_v4_4_2_inst_enable(struct amdgpu_device *adev, bool enable,
+				    uint32_t inst_mask)
 {
 	u32 f32_cntl;
 	int i;
 
 	if (!enable) {
-		sdma_v4_4_2_gfx_stop(adev);
-		sdma_v4_4_2_rlc_stop(adev);
+		sdma_v4_4_2_inst_gfx_stop(adev, inst_mask);
+		sdma_v4_4_2_inst_rlc_stop(adev, inst_mask);
 		if (adev->sdma.has_page_queue)
-			sdma_v4_4_2_page_stop(adev);
+			sdma_v4_4_2_inst_page_stop(adev, inst_mask);
 	}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		f32_cntl = RREG32_SDMA(i, regSDMA_F32_CNTL);
 		f32_cntl = REG_SET_FIELD(f32_cntl, SDMA_F32_CNTL, HALT, enable ? 0 : 1);
 		WREG32_SDMA(i, regSDMA_F32_CNTL, f32_cntl);
@@ -780,7 +786,8 @@ static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev)
  * Set up the compute DMA queues and enable them.
  * Returns 0 for success, error for failure.
  */
-static int sdma_v4_4_2_rlc_resume(struct amdgpu_device *adev)
+static int sdma_v4_4_2_inst_rlc_resume(struct amdgpu_device *adev,
+				       uint32_t inst_mask)
 {
 	sdma_v4_4_2_init_pg(adev);
 
@@ -795,7 +802,8 @@ static int sdma_v4_4_2_rlc_resume(struct amdgpu_device *adev)
  * Loads the sDMA0/1 ucode.
  * Returns 0 for success, -EINVAL if the ucode is not available.
  */
-static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev)
+static int sdma_v4_4_2_inst_load_microcode(struct amdgpu_device *adev,
+					   uint32_t inst_mask)
 {
 	const struct sdma_firmware_header_v1_0 *hdr;
 	const __le32 *fw_data;
@@ -803,9 +811,9 @@ static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev)
 	int i, j;
 
 	/* halt the MEs */
-	sdma_v4_4_2_enable(adev, false);
+	sdma_v4_4_2_inst_enable(adev, false, inst_mask);
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	for_each_inst(i, inst_mask) {
 		if (!adev->sdma.instance[i].fw)
 			return -EINVAL;
 
@@ -831,38 +839,41 @@ static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev)
 }
 
 /**
- * sdma_v4_4_2_start - setup and start the async dma engines
+ * sdma_v4_4_2_inst_start - setup and start the async dma engines
  *
  * @adev: amdgpu_device pointer
  *
  * Set up the DMA engines and enable them.
  * Returns 0 for success, error for failure.
  */
-static int sdma_v4_4_2_start(struct amdgpu_device *adev)
+static int sdma_v4_4_2_inst_start(struct amdgpu_device *adev,
+				  uint32_t inst_mask)
 {
 	struct amdgpu_ring *ring;
+	uint32_t tmp_mask;
 	int i, r = 0;
 
 	if (amdgpu_sriov_vf(adev)) {
-		sdma_v4_4_2_ctx_switch_enable(adev, false);
-		sdma_v4_4_2_enable(adev, false);
+		sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask);
+		sdma_v4_4_2_inst_enable(adev, false, inst_mask);
 	} else {
 		/* bypass sdma microcode loading on Gopher */
 		if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP &&
-		    !(adev->pdev->device == 0x49) && !(adev->pdev->device == 0x50)) {
-			r = sdma_v4_4_2_load_microcode(adev);
+		    adev->sdma.instance[0].fw) {
+			r = sdma_v4_4_2_inst_load_microcode(adev, inst_mask);
 			if (r)
 				return r;
 		}
 
 		/* unhalt the MEs */
-		sdma_v4_4_2_enable(adev, true);
+		sdma_v4_4_2_inst_enable(adev, true, inst_mask);
 		/* enable sdma ring preemption */
-		sdma_v4_4_2_ctx_switch_enable(adev, true);
+		sdma_v4_4_2_inst_ctx_switch_enable(adev, true, inst_mask);
 	}
 
 	/* start the gfx rings and rlc compute queues */
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	tmp_mask = inst_mask;
+	for_each_inst(i, tmp_mask) {
 		uint32_t temp;
 
 		WREG32_SDMA(i, regSDMA_SEM_WAIT_FAIL_TIMER_CNTL, 0);
@@ -889,15 +900,16 @@ static int sdma_v4_4_2_start(struct amdgpu_device *adev)
 	}
 
 	if (amdgpu_sriov_vf(adev)) {
-		sdma_v4_4_2_ctx_switch_enable(adev, true);
-		sdma_v4_4_2_enable(adev, true);
+		sdma_v4_4_2_inst_ctx_switch_enable(adev, true, inst_mask);
+		sdma_v4_4_2_inst_enable(adev, true, inst_mask);
 	} else {
-		r = sdma_v4_4_2_rlc_resume(adev);
+		r = sdma_v4_4_2_inst_rlc_resume(adev, inst_mask);
 		if (r)
 			return r;
 	}
 
-	for (i = 0; i < adev->sdma.num_instances; i++) {
+	tmp_mask = inst_mask;
+	for_each_inst(i, tmp_mask) {
 		ring = &adev->sdma.instance[i].ring;
 
 		r = amdgpu_ring_test_helper(ring);
@@ -1383,14 +1395,17 @@ static int sdma_v4_4_2_hw_init(void *handle)
 {
 	int r;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t inst_mask;
 
+	/* TODO: Check if this is needed */
 	if (adev->flags & AMD_IS_APU)
 		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_SDMA, false);
 
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	if (!amdgpu_sriov_vf(adev))
-		sdma_v4_4_2_init_golden_registers(adev);
+		sdma_v4_4_2_inst_init_golden_registers(adev, inst_mask);
 
-	r = sdma_v4_4_2_start(adev);
+	r = sdma_v4_4_2_inst_start(adev, inst_mask);
 
 	return r;
 }
@@ -1398,22 +1413,27 @@ static int sdma_v4_4_2_hw_init(void *handle)
 static int sdma_v4_4_2_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t inst_mask;
 	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	for (i = 0; i < adev->sdma.num_instances; i++) {
 		amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
 			       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
 	}
 
-	sdma_v4_4_2_ctx_switch_enable(adev, false);
-	sdma_v4_4_2_enable(adev, false);
+	sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask);
+	sdma_v4_4_2_inst_enable(adev, false, inst_mask);
 
 	return 0;
 }
 
+static int sdma_v4_4_2_set_clockgating_state(void *handle,
+					     enum amd_clockgating_state state);
+
 static int sdma_v4_4_2_suspend(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -1650,15 +1670,39 @@ static int sdma_v4_4_2_process_srbm_write_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
-static void sdma_v4_4_2_update_medium_grain_clock_gating(
-		struct amdgpu_device *adev,
-		bool enable)
+static void sdma_v4_4_2_inst_update_medium_grain_light_sleep(
+	struct amdgpu_device *adev, bool enable, uint32_t inst_mask)
+{
+	uint32_t data, def;
+	int i;
+
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) {
+		for_each_inst(i, inst_mask) {
+			/* 1-not override: enable sdma mem light sleep */
+			def = data = RREG32_SDMA(i, regSDMA_POWER_CNTL);
+			data |= SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
+			if (def != data)
+				WREG32_SDMA(i, regSDMA_POWER_CNTL, data);
+		}
+	} else {
+		for_each_inst(i, inst_mask) {
+			/* 0-override:disable sdma mem light sleep */
+			def = data = RREG32_SDMA(i, regSDMA_POWER_CNTL);
+			data &= ~SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
+			if (def != data)
+				WREG32_SDMA(i, regSDMA_POWER_CNTL, data);
+		}
+	}
+}
+
+static void sdma_v4_4_2_inst_update_medium_grain_clock_gating(
+	struct amdgpu_device *adev, bool enable, uint32_t inst_mask)
 {
 	uint32_t data, def;
 	int i;
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) {
-		for (i = 0; i < adev->sdma.num_instances; i++) {
+		for_each_inst(i, inst_mask) {
 			def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL);
 			data &= ~(SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK |
 				  SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK |
@@ -1672,7 +1716,7 @@ static void sdma_v4_4_2_update_medium_grain_clock_gating(
 				WREG32_SDMA(i, regSDMA_CLK_CTRL, data);
 		}
 	} else {
-		for (i = 0; i < adev->sdma.num_instances; i++) {
+		for_each_inst(i, inst_mask) {
 			def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL);
 			data |= (SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK |
 				 SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK |
@@ -1688,45 +1732,21 @@ static void sdma_v4_4_2_update_medium_grain_clock_gating(
 	}
 }
 
-
-static void sdma_v4_4_2_update_medium_grain_light_sleep(
-		struct amdgpu_device *adev,
-		bool enable)
-{
-	uint32_t data, def;
-	int i;
-
-	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) {
-		for (i = 0; i < adev->sdma.num_instances; i++) {
-			/* 1-not override: enable sdma mem light sleep */
-			def = data = RREG32_SDMA(0, regSDMA_POWER_CNTL);
-			data |= SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
-			if (def != data)
-				WREG32_SDMA(0, regSDMA_POWER_CNTL, data);
-		}
-	} else {
-		for (i = 0; i < adev->sdma.num_instances; i++) {
-		/* 0-override:disable sdma mem light sleep */
-			def = data = RREG32_SDMA(0, regSDMA_POWER_CNTL);
-			data &= ~SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK;
-			if (def != data)
-				WREG32_SDMA(0, regSDMA_POWER_CNTL, data);
-		}
-	}
-}
-
 static int sdma_v4_4_2_set_clockgating_state(void *handle,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t inst_mask;
 
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-	sdma_v4_4_2_update_medium_grain_clock_gating(adev,
-			state == AMD_CG_STATE_GATE);
-	sdma_v4_4_2_update_medium_grain_light_sleep(adev,
-			state == AMD_CG_STATE_GATE);
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
+
+	sdma_v4_4_2_inst_update_medium_grain_clock_gating(
+		adev, state == AMD_CG_STATE_GATE, inst_mask);
+	sdma_v4_4_2_inst_update_medium_grain_light_sleep(
+		adev, state == AMD_CG_STATE_GATE, inst_mask);
 	return 0;
 }
 
-- 
GitLab


From 75d1692393cb78b510ff18733457f91d002452f7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 16 Sep 2022 12:43:35 +0530
Subject: [PATCH 1145/1791] drm/amdgpu: Add initial version of XCP routines

Within a device, an accelerator core partition can be constituted with
different IP instances. These partitions are spatial in nature. Number
of partitions which can exist at the same time depends on the 'partition
mode'. Add a manager entity which is responsible for switching between
different partition modes and maintaining partitions. It is also
responsible for suspend/resume of different partitions.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile      |   2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      |   2 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h  |   2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h |   2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c  | 244 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h  | 107 ++++++++++
 6 files changed, 356 insertions(+), 3 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
 create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 8418a90df4933..74a9aa6fe18c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -60,7 +60,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \
 	amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o \
 	amdgpu_fw_attestation.o amdgpu_securedisplay.o \
 	amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \
-	amdgpu_ring_mux.o
+	amdgpu_ring_mux.o amdgpu_xcp.o
 
 amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index c2dd100bbd60f..3858d29baef10 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -283,6 +283,7 @@ extern uint amdgpu_user_partt_mode;
 #define AMDGPU_SMARTSHIFT_MAX_BIAS (100)
 #define AMDGPU_SMARTSHIFT_MIN_BIAS (-100)
 
+struct amdgpu_xcp_mgr;
 struct amdgpu_device;
 struct amdgpu_irq_src;
 struct amdgpu_fpriv;
@@ -765,6 +766,7 @@ struct amdgpu_device {
 	struct amdgpu_acp		acp;
 #endif
 	struct amdgpu_hive_info *hive;
+	struct amdgpu_xcp_mgr *xcp_mgr;
 	/* ASIC */
 	enum amd_asic_type		asic_type;
 	uint32_t			family;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 023c5b08929ff..5adfe42776412 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -61,7 +61,7 @@ enum amdgpu_gfx_partition {
 	AMDGPU_TPX_PARTITION_MODE = 2,
 	AMDGPU_QPX_PARTITION_MODE = 3,
 	AMDGPU_CPX_PARTITION_MODE = 4,
-	AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE,
+	AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE = -1,
 };
 
 #define NUM_XCC(x) hweight16(x)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index 639b86c4d6640..eb25ac98903f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -96,7 +96,7 @@ struct amdgpu_nbio_funcs {
 	void (*apply_l1_link_width_reconfig_wa)(struct amdgpu_device *adev);
 	void (*clear_doorbell_interrupt)(struct amdgpu_device *adev);
 	u32 (*get_rom_offset)(struct amdgpu_device *adev);
-	u32 (*get_compute_partition_mode)(struct amdgpu_device *adev);
+	int (*get_compute_partition_mode)(struct amdgpu_device *adev);
 	u32 (*get_memory_partition_mode)(struct amdgpu_device *adev);
 	void (*set_compute_partition_mode)(struct amdgpu_device *adev,
 					   enum amdgpu_gfx_partition mode);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
new file mode 100644
index 0000000000000..f59bc450cabea
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -0,0 +1,244 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "amdgpu_xcp.h"
+
+static int __amdgpu_xcp_run(struct amdgpu_xcp_mgr *xcp_mgr,
+			    struct amdgpu_xcp_ip *xcp_ip, int xcp_state)
+{
+	int (*run_func)(void *handle, uint32_t inst_mask);
+	int ret = 0;
+
+	if (!xcp_ip || !xcp_ip->valid || !xcp_ip->ip_funcs)
+		return 0;
+
+	run_func = NULL;
+
+	switch (xcp_state) {
+	case AMDGPU_XCP_PREPARE_SUSPEND:
+		run_func = xcp_ip->ip_funcs->prepare_suspend;
+		break;
+	case AMDGPU_XCP_SUSPEND:
+		run_func = xcp_ip->ip_funcs->suspend;
+		break;
+	case AMDGPU_XCP_PREPARE_RESUME:
+		run_func = xcp_ip->ip_funcs->prepare_resume;
+		break;
+	case AMDGPU_XCP_RESUME:
+		run_func = xcp_ip->ip_funcs->resume;
+		break;
+	}
+
+	if (run_func)
+		ret = run_func(xcp_mgr->adev, xcp_ip->inst_mask);
+
+	return ret;
+}
+
+static int amdgpu_xcp_run_transition(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+				     int state)
+{
+	struct amdgpu_xcp_ip *xcp_ip;
+	struct amdgpu_xcp *xcp;
+	int i, ret;
+
+	if (xcp_id > MAX_XCP || !xcp_mgr->xcp[xcp_id].valid)
+		return -EINVAL;
+
+	xcp = &xcp_mgr->xcp[xcp_id];
+	for (i = 0; i < AMDGPU_XCP_MAX_BLOCKS; ++i) {
+		xcp_ip = &xcp->ip[i];
+		ret = __amdgpu_xcp_run(xcp_mgr, xcp_ip, state);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id)
+{
+	return amdgpu_xcp_run_transition(xcp_mgr, xcp_id,
+					 AMDGPU_XCP_PREPARE_SUSPEND);
+}
+
+int amdgpu_xcp_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id)
+{
+	return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, AMDGPU_XCP_SUSPEND);
+}
+
+int amdgpu_xcp_prepare_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id)
+{
+	return amdgpu_xcp_run_transition(xcp_mgr, xcp_id,
+					 AMDGPU_XCP_PREPARE_RESUME);
+}
+
+int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id)
+{
+	return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, AMDGPU_XCP_RESUME);
+}
+
+static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+				   struct amdgpu_xcp_ip *ip)
+{
+	struct amdgpu_xcp *xcp;
+
+	if (!ip)
+		return;
+
+	xcp = &xcp_mgr->xcp[xcp_id];
+	xcp->ip[ip->ip_id] = *ip;
+	xcp->ip[ip->ip_id].valid = true;
+
+	xcp->valid = true;
+}
+
+static int __amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps)
+{
+	struct amdgpu_xcp_ip ip;
+	int i, j, ret;
+
+	for (i = 0; i < MAX_XCP; ++i)
+		xcp_mgr->xcp[i].valid = false;
+
+	for (i = 0; i < num_xcps; ++i) {
+		for (j = AMDGPU_XCP_GFXHUB; j < AMDGPU_XCP_MAX_BLOCKS; ++j) {
+			ret = xcp_mgr->funcs->get_ip_details(xcp_mgr, i, j,
+							     &ip);
+			if (ret)
+				continue;
+
+			__amdgpu_xcp_add_block(xcp_mgr, i, &ip);
+		}
+	}
+
+	xcp_mgr->num_xcps = num_xcps;
+
+	return 0;
+}
+
+int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
+{
+	int ret, num_xcps = 0;
+
+	if (!xcp_mgr || mode == AMDGPU_XCP_MODE_NONE)
+		return -EINVAL;
+
+	if (xcp_mgr->mode == mode)
+		return 0;
+
+	if (!xcp_mgr->funcs || !xcp_mgr->funcs->switch_partition_mode)
+		return 0;
+
+	mutex_lock(&xcp_mgr->xcp_lock);
+
+	ret = xcp_mgr->funcs->switch_partition_mode(xcp_mgr, mode, &num_xcps);
+
+	if (ret)
+		goto out;
+
+	if (!num_xcps || num_xcps > MAX_XCP) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	xcp_mgr->mode = mode;
+	__amdgpu_xcp_init(xcp_mgr, num_xcps);
+out:
+	mutex_unlock(&xcp_mgr->xcp_lock);
+
+	return ret;
+}
+
+int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
+{
+	int mode;
+
+	if (xcp_mgr->mode == AMDGPU_XCP_MODE_NONE)
+		return xcp_mgr->mode;
+
+	if (!xcp_mgr->funcs || !xcp_mgr->funcs->query_partition_mode)
+		return xcp_mgr->mode;
+
+	mutex_lock(&xcp_mgr->xcp_lock);
+	mode = xcp_mgr->funcs->query_partition_mode(xcp_mgr);
+	if (mode != xcp_mgr->mode)
+		dev_WARN(
+			xcp_mgr->adev->dev,
+			"Cached partition mode %d not matching with device mode %d",
+			xcp_mgr->mode, mode);
+
+	mutex_unlock(&xcp_mgr->xcp_lock);
+
+	return mode;
+}
+
+int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
+			int init_num_xcps,
+			struct amdgpu_xcp_mgr_funcs *xcp_funcs)
+{
+	struct amdgpu_xcp_mgr *xcp_mgr;
+
+	if (!xcp_funcs || !xcp_funcs->switch_partition_mode ||
+	    !xcp_funcs->get_ip_details)
+		return -EINVAL;
+
+	xcp_mgr = kzalloc(sizeof(*xcp_mgr), GFP_KERNEL);
+
+	if (!xcp_mgr)
+		return -ENOMEM;
+
+	xcp_mgr->adev = adev;
+	xcp_mgr->funcs = xcp_funcs;
+	xcp_mgr->mode = init_mode;
+	mutex_init(&xcp_mgr->xcp_lock);
+
+	if (init_mode != AMDGPU_XCP_MODE_NONE)
+		__amdgpu_xcp_init(xcp_mgr, init_num_xcps);
+
+	adev->xcp_mgr = xcp_mgr;
+
+	return 0;
+}
+
+int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
+			     enum AMDGPU_XCP_IP_BLOCK ip, int instance)
+{
+	struct amdgpu_xcp *xcp;
+	int i, id_mask = 0;
+
+	if (ip >= AMDGPU_XCP_MAX_BLOCKS)
+		return -EINVAL;
+
+	for (i = 0; i < xcp_mgr->num_xcps; ++i) {
+		xcp = &xcp_mgr->xcp[i];
+		if ((xcp->valid) && (xcp->ip[ip].valid) &&
+		    (xcp->ip[ip].inst_mask & BIT(instance)))
+			id_mask |= BIT(i);
+	}
+
+	if (!id_mask)
+		id_mask = -ENXIO;
+
+	return id_mask;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
new file mode 100644
index 0000000000000..f0b973c6092fd
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -0,0 +1,107 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef AMDGPU_XCP_H
+#define AMDGPU_XCP_H
+
+#include <linux/xarray.h>
+
+#define MAX_XCP 8
+
+#define AMDGPU_XCP_MODE_NONE -1
+
+enum AMDGPU_XCP_IP_BLOCK {
+	AMDGPU_XCP_GFXHUB,
+	AMDGPU_XCP_GFX,
+	AMDGPU_XCP_SDMA,
+	AMDGPU_XCP_VCN,
+	AMDGPU_XCP_MAX_BLOCKS
+};
+
+enum AMDGPU_XCP_STATE {
+	AMDGPU_XCP_PREPARE_SUSPEND,
+	AMDGPU_XCP_SUSPEND,
+	AMDGPU_XCP_PREPARE_RESUME,
+	AMDGPU_XCP_RESUME,
+};
+
+struct amdgpu_xcp_ip_funcs {
+	int (*prepare_suspend)(void *handle, uint32_t inst_mask);
+	int (*suspend)(void *handle, uint32_t inst_mask);
+	int (*prepare_resume)(void *handle, uint32_t inst_mask);
+	int (*resume)(void *handle, uint32_t inst_mask);
+};
+
+struct amdgpu_xcp_ip {
+	struct amdgpu_xcp_ip_funcs *ip_funcs;
+	uint32_t inst_mask;
+
+	enum AMDGPU_XCP_IP_BLOCK ip_id;
+	bool valid;
+};
+
+struct amdgpu_xcp {
+	struct amdgpu_xcp_ip ip[AMDGPU_XCP_MAX_BLOCKS];
+
+	uint8_t id;
+	uint8_t mem_node;
+	bool valid;
+};
+
+struct amdgpu_xcp_mgr {
+	struct amdgpu_device *adev;
+	struct mutex xcp_lock;
+	struct amdgpu_xcp_mgr_funcs *funcs;
+
+	struct amdgpu_xcp xcp[MAX_XCP];
+	uint8_t num_xcps;
+	int8_t mode;
+};
+
+struct amdgpu_xcp_mgr_funcs {
+	int (*switch_partition_mode)(struct amdgpu_xcp_mgr *xcp_mgr, int mode,
+				     int *num_xcps);
+	int (*query_partition_mode)(struct amdgpu_xcp_mgr *xcp_mgr);
+	int (*get_ip_details)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+			      enum AMDGPU_XCP_IP_BLOCK ip_id,
+			      struct amdgpu_xcp_ip *ip);
+
+	int (*prepare_suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+	int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+	int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+	int (*resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+};
+
+int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+int amdgpu_xcp_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+int amdgpu_xcp_prepare_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+
+int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
+			int init_xcps, struct amdgpu_xcp_mgr_funcs *xcp_funcs);
+int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr);
+int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode);
+int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
+			     enum AMDGPU_XCP_IP_BLOCK ip, int instance);
+
+#endif
-- 
GitLab


From 9cb18287d8f1c4549d95280e2cf60f4d1bab64f8 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 19 Sep 2022 17:34:02 +0530
Subject: [PATCH 1146/1791] drm/amdgpu: Add SOC partition funcs for GC v9.4.3

Switching the partition mode configuration of ASIC is SOC
level function rather than something at GFX core level. Add
partition mode switch functions as SOC specific callbacks.
Implement the XCP manager callbacks needed for partition
switch for GC 9.4.3 based ASICs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |   4 +
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 232 ++++++++++++++++++
 2 files changed, 236 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 5adfe42776412..8be4ab50b171d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -278,8 +278,12 @@ struct amdgpu_gfx_funcs {
 			(*query_partition_mode)(struct amdgpu_device *adev);
 	enum amdgpu_memory_partition
 			(*query_mem_partition_mode)(struct amdgpu_device *adev);
+
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
 				     enum amdgpu_gfx_partition mode);
+
+	int (*switch_gfx_partition_mode)(struct amdgpu_device *adev,
+				     int num_xccs_per_xcp);
 };
 
 struct sq_work {
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 2616bdb40418e..4264fbd267ae3 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -24,6 +24,12 @@
 #include "soc15.h"
 
 #include "soc15_common.h"
+#include "amdgpu_xcp.h"
+
+#define XCP_INST_MASK(num_inst, xcp_id)                                        \
+	(num_inst ? GENMASK(num_inst - 1, 0) << (xcp_id * num_inst) : 0)
+
+#define AMDGPU_XCP_OPS_KFD	(1 << 0)
 
 void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
 {
@@ -119,3 +125,229 @@ u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id)
 
 	return ext_offset;
 }
+
+static int aqua_vanjaram_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
+{
+	enum amdgpu_gfx_partition mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+	struct amdgpu_device *adev = xcp_mgr->adev;
+
+	if (adev->nbio.funcs->get_compute_partition_mode)
+		mode = adev->nbio.funcs->get_compute_partition_mode(adev);
+
+	return mode;
+}
+
+int __aqua_vanjaram_get_xcc_per_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
+{
+	int num_xcc, num_xcc_per_xcp = 0;
+
+	num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask);
+
+	switch (mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		num_xcc_per_xcp = num_xcc;
+		break;
+	case AMDGPU_DPX_PARTITION_MODE:
+		num_xcc_per_xcp = num_xcc / 2;
+		break;
+	case AMDGPU_TPX_PARTITION_MODE:
+		num_xcc_per_xcp = num_xcc / 3;
+		break;
+	case AMDGPU_QPX_PARTITION_MODE:
+		num_xcc_per_xcp = num_xcc / 4;
+		break;
+	case AMDGPU_CPX_PARTITION_MODE:
+		num_xcc_per_xcp = 1;
+		break;
+	}
+
+	return num_xcc_per_xcp;
+}
+
+int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+				    enum AMDGPU_XCP_IP_BLOCK ip_id,
+				    struct amdgpu_xcp_ip *ip)
+{
+	struct amdgpu_device *adev = xcp_mgr->adev;
+	int num_xcc_xcp, num_sdma_xcp, num_vcn_xcp;
+	int num_sdma, num_vcn;
+
+	num_sdma = adev->sdma.num_instances;
+	num_vcn = adev->vcn.num_vcn_inst;
+
+	switch (xcp_mgr->mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		num_sdma_xcp = num_sdma;
+		num_vcn_xcp = num_vcn;
+		break;
+	case AMDGPU_DPX_PARTITION_MODE:
+		num_sdma_xcp = num_sdma / 2;
+		num_vcn_xcp = num_vcn / 2;
+		break;
+	case AMDGPU_TPX_PARTITION_MODE:
+		num_sdma_xcp = num_sdma / 3;
+		num_vcn_xcp = num_vcn / 3;
+		break;
+	case AMDGPU_QPX_PARTITION_MODE:
+		num_sdma_xcp = num_sdma / 4;
+		num_vcn_xcp = num_vcn / 4;
+		break;
+	case AMDGPU_CPX_PARTITION_MODE:
+		num_sdma_xcp = 2;
+		num_vcn_xcp = num_vcn ? 1 : 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	num_xcc_xcp = adev->gfx.num_xcc_per_xcp;
+
+	switch (ip_id) {
+	case AMDGPU_XCP_GFXHUB:
+		ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id);
+		/* TODO : Assign IP funcs */
+		break;
+	case AMDGPU_XCP_GFX:
+		ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id);
+		/* TODO : Assign IP funcs */
+		break;
+	case AMDGPU_XCP_SDMA:
+		ip->inst_mask = XCP_INST_MASK(num_sdma_xcp, xcp_id);
+		/* TODO : Assign IP funcs */
+		break;
+	case AMDGPU_XCP_VCN:
+		ip->inst_mask = XCP_INST_MASK(num_vcn_xcp, xcp_id);
+		/* TODO : Assign IP funcs */
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ip->ip_id = ip_id;
+
+	return 0;
+}
+
+static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
+					  enum amdgpu_gfx_partition mode)
+{
+	int num_xcc, num_xccs_per_xcp;
+
+	num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask);
+	switch (mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		return num_xcc > 0;
+	case AMDGPU_DPX_PARTITION_MODE:
+		return (num_xcc % 4) == 0;
+	case AMDGPU_TPX_PARTITION_MODE:
+		return (num_xcc % 3) == 0;
+	case AMDGPU_QPX_PARTITION_MODE:
+		num_xccs_per_xcp = num_xcc / 4;
+		return (num_xccs_per_xcp >= 2);
+	case AMDGPU_CPX_PARTITION_MODE:
+		return (num_xcc > 1);
+	default:
+		return false;
+	}
+
+	return false;
+}
+
+static int __aqua_vanjaram_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
+{
+	/* TODO:
+	 * Stop user queues and threads, and make sure GPU is empty of work.
+	 */
+
+	if (flags & AMDGPU_XCP_OPS_KFD)
+		amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev);
+
+	return 0;
+}
+
+static int __aqua_vanjaram_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
+{
+	int ret = 0;
+
+	if (flags & AMDGPU_XCP_OPS_KFD) {
+		amdgpu_amdkfd_device_probe(xcp_mgr->adev);
+		amdgpu_amdkfd_device_init(xcp_mgr->adev);
+		/* If KFD init failed, return failure */
+		if (!xcp_mgr->adev->kfd.init_complete)
+			ret = -EIO;
+	}
+
+	return ret;
+}
+
+static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
+					       int mode, int *num_xcps)
+{
+	int num_xcc_per_xcp, num_xcc, ret;
+	struct amdgpu_device *adev;
+	u32 flags = 0;
+
+	adev = xcp_mgr->adev;
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+
+	if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode))
+		return -EINVAL;
+
+	if (adev->kfd.init_complete)
+		flags |= AMDGPU_XCP_OPS_KFD;
+
+	if (flags & AMDGPU_XCP_OPS_KFD) {
+		ret = amdgpu_amdkfd_check_and_lock_kfd(adev);
+		if (ret)
+			goto out;
+	}
+
+	ret = __aqua_vanjaram_pre_partition_switch(xcp_mgr, flags);
+	if (ret)
+		goto unlock;
+
+	num_xcc_per_xcp = __aqua_vanjaram_get_xcc_per_xcp(xcp_mgr, mode);
+	if (adev->gfx.funcs->switch_gfx_partition_mode)
+		adev->gfx.funcs->switch_gfx_partition_mode(xcp_mgr->adev,
+						       num_xcc_per_xcp);
+
+	if (adev->nbio.funcs->set_compute_partition_mode)
+		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
+
+	ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags);
+unlock:
+	if (flags & AMDGPU_XCP_OPS_KFD)
+		amdgpu_amdkfd_unlock_kfd(adev);
+out:
+	return ret;
+}
+
+int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+				     enum AMDGPU_XCP_IP_BLOCK ip_id,
+				     struct amdgpu_xcp_ip *ip)
+{
+	if (!ip)
+		return -EINVAL;
+
+	return __aqua_vanjaram_get_xcp_ip_info(xcp_mgr, xcp_id, ip_id, ip);
+}
+
+struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = {
+	.switch_partition_mode = &aqua_vanjaram_switch_partition_mode,
+	.query_partition_mode = &aqua_vanjaram_query_partition_mode,
+	.get_ip_details = &aqua_vanjaram_get_xcp_ip_details
+};
+
+static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
+{
+	int ret;
+
+	ret = amdgpu_xcp_mgr_init(adev, AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE, 1,
+				  &aqua_vanjaram_xcp_funcs);
+	if (ret)
+		return ret;
+
+	/* TODO: Default memory node affinity init */
+
+	return ret;
+}
-- 
GitLab


From e56c9ef6cb35f33dc83f635419ae55adf69db9fc Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 23 Sep 2022 14:43:17 +0530
Subject: [PATCH 1147/1791] drm/amdgpu: Add soc config init for GC9.4.3 ASICs

Add function to initialize soc configuration information for GC 9.4.3
ASICs. Use it to map IPs and other SOC related information once IP
configuration information is available through discovery.

For GC9.4.3 compute partition related callbacks are initialized as part
of configuration init.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c       |  6 +++---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/soc15.h                  |  1 +
 3 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index b602df4b445a9..e6d10a3f1753c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1983,11 +1983,11 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void amdgpu_discovery_ip_map_init(struct amdgpu_device *adev)
+static void amdgpu_discovery_init_soc_config(struct amdgpu_device *adev)
 {
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 3):
-		aqua_vanjaram_ip_map_init(adev);
+		aqua_vanjaram_init_soc_config(adev);
 		break;
 	default:
 		break;
@@ -2171,7 +2171,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		break;
 	}
 
-	amdgpu_discovery_ip_map_init(adev);
+	amdgpu_discovery_init_soc_config(adev);
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 0, 1):
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 4264fbd267ae3..a9de229a28286 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -351,3 +351,16 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
 
 	return ret;
 }
+
+int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
+{
+	int ret;
+
+	ret = aqua_vanjaram_xcp_mgr_init(adev);
+	if (ret)
+		return ret;
+
+	aqua_vanjaram_ip_map_init(adev);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h
index dd48db09aa511..eac54042c6c0e 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.h
@@ -113,6 +113,7 @@ int arct_reg_base_init(struct amdgpu_device *adev);
 int aldebaran_reg_base_init(struct amdgpu_device *adev);
 void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev);
 u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id);
+int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev);
 
 void vega10_doorbell_index_init(struct amdgpu_device *adev);
 void vega20_doorbell_index_init(struct amdgpu_device *adev);
-- 
GitLab


From 8e7fd19380f9187dae3ad18a61793b1752dfa097 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 16 Nov 2022 17:15:47 +0530
Subject: [PATCH 1148/1791] drm/amdgpu: Switch to SOC partition funcs

For GFXv9.4.3, use SOC level partition switch implementation rather than
keeping them at GFX IP level. Change the exisiting implementation in
GFX IP for keeping partition mode and restrict it to only GFX related
switch.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c       | 31 ++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |  5 --
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   |  4 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 59 +++----------------
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |  8 ++-
 5 files changed, 20 insertions(+), 87 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 5ff49737d7c60..f895a4b8ca0de 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -28,6 +28,7 @@
 #include "amdgpu_gfx.h"
 #include "amdgpu_rlc.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_xcp.h"
 
 /* delay 0.1 second to enable gfx off feature */
 #define GFX_OFF_DELAY_ENABLE         msecs_to_jiffies(100)
@@ -1170,10 +1171,10 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 {
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = drm_to_adev(ddev);
-	enum amdgpu_gfx_partition mode;
+	int mode;
 	char *partition_mode;
 
-	mode = adev->gfx.funcs->query_partition_mode(adev);
+	mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr);
 
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
@@ -1254,31 +1255,7 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 		return -EINVAL;
 	}
 
-	if (!adev->kfd.init_complete)
-		return -EPERM;
-
-	mutex_lock(&adev->gfx.partition_mutex);
-
-	if (mode == adev->gfx.funcs->query_partition_mode(adev))
-		goto out;
-
-	ret = amdgpu_amdkfd_check_and_lock_kfd(adev);
-	if (ret)
-		goto out;
-
-	amdgpu_amdkfd_device_fini_sw(adev);
-
-	adev->gfx.funcs->switch_partition_mode(adev, mode);
-
-	amdgpu_amdkfd_device_probe(adev);
-	amdgpu_amdkfd_device_init(adev);
-	/* If KFD init failed, return failure */
-	if (!adev->kfd.init_complete)
-		ret = -EIO;
-
-	amdgpu_amdkfd_unlock_kfd(adev);
-out:
-	mutex_unlock(&adev->gfx.partition_mutex);
+	ret = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, mode);
 
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 8be4ab50b171d..2287768ed141a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -278,11 +278,7 @@ struct amdgpu_gfx_funcs {
 			(*query_partition_mode)(struct amdgpu_device *adev);
 	enum amdgpu_memory_partition
 			(*query_mem_partition_mode)(struct amdgpu_device *adev);
-
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
-				     enum amdgpu_gfx_partition mode);
-
-	int (*switch_gfx_partition_mode)(struct amdgpu_device *adev,
 				     int num_xccs_per_xcp);
 };
 
@@ -416,7 +412,6 @@ struct amdgpu_gfx {
 
 	bool				cp_gfx_shadow; /* for gfx11 */
 
-	enum amdgpu_gfx_partition	partition_mode;
 	uint16_t 			xcc_mask;
 	enum amdgpu_memory_partition	mem_partition_mode;
 	uint32_t			num_xcc_per_xcp;
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index a9de229a28286..bbcdececfd2f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -307,8 +307,8 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 		goto unlock;
 
 	num_xcc_per_xcp = __aqua_vanjaram_get_xcc_per_xcp(xcp_mgr, mode);
-	if (adev->gfx.funcs->switch_gfx_partition_mode)
-		adev->gfx.funcs->switch_gfx_partition_mode(xcp_mgr->adev,
+	if (adev->gfx.funcs->switch_partition_mode)
+		adev->gfx.funcs->switch_partition_mode(xcp_mgr->adev,
 						       num_xcc_per_xcp);
 
 	if (adev->nbio.funcs->set_compute_partition_mode)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 93a0baa4515c2..d684037a7a5d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -38,6 +38,7 @@
 #include "gc/gc_9_4_3_sh_mask.h"
 
 #include "gfx_v9_4_3.h"
+#include "amdgpu_xcp.h"
 
 MODULE_FIRMWARE("amdgpu/gc_9_4_3_mec.bin");
 MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
@@ -614,61 +615,23 @@ gfx_v9_4_3_query_memory_partition(struct amdgpu_device *adev)
 	return mode;
 }
 
-static enum amdgpu_gfx_partition
-gfx_v9_4_3_query_compute_partition(struct amdgpu_device *adev)
-{
-	enum amdgpu_gfx_partition mode = adev->gfx.partition_mode;
-
-	if (adev->nbio.funcs->get_compute_partition_mode)
-		mode = adev->nbio.funcs->get_compute_partition_mode(adev);
-
-	return mode;
-}
-
 static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
-						enum amdgpu_gfx_partition mode)
+						int num_xccs_per_xcp)
 {
+	int i, num_xcc;
 	u32 tmp = 0;
-	int num_xcc_per_partition, i, num_xcc;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
-	switch (mode) {
-	case AMDGPU_SPX_PARTITION_MODE:
-		num_xcc_per_partition = num_xcc;
-		break;
-	case AMDGPU_DPX_PARTITION_MODE:
-		num_xcc_per_partition = num_xcc / 2;
-		break;
-	case AMDGPU_TPX_PARTITION_MODE:
-		num_xcc_per_partition = num_xcc / 3;
-		break;
-	case AMDGPU_QPX_PARTITION_MODE:
-		num_xcc_per_partition = num_xcc / 4;
-		break;
-	case AMDGPU_CPX_PARTITION_MODE:
-		num_xcc_per_partition = 1;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	/* TODO:
-	 * Stop user queues and threads, and make sure GPU is empty of work.
-	 */
 
 	for (i = 0; i < num_xcc; i++) {
 		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP,
-				    num_xcc_per_partition);
+				    num_xccs_per_xcp);
 		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID,
-				    i % num_xcc_per_partition);
+				    i % num_xccs_per_xcp);
 		WREG32_SOC15(GC, GET_INST(GC, i), regCP_HYP_XCP_CTL, tmp);
 	}
 
-	if (adev->nbio.funcs->set_compute_partition_mode)
-		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
-
-	adev->gfx.num_xcc_per_xcp = num_xcc_per_partition;
-	adev->gfx.partition_mode = mode;
+	adev->gfx.num_xcc_per_xcp = num_xccs_per_xcp;
 
 	return 0;
 }
@@ -680,7 +643,6 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs,
 	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
-	.query_partition_mode = &gfx_v9_4_3_query_compute_partition,
 	.switch_partition_mode = &gfx_v9_4_3_switch_compute_partition,
 	.query_mem_partition_mode = &gfx_v9_4_3_query_memory_partition,
 };
@@ -1899,10 +1861,6 @@ static int gfx_v9_4_3_xcc_cp_resume(struct amdgpu_device *adev, int xcc_id)
 			return r;
 	}
 
-	if (adev->gfx.partition_mode == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
-		gfx_v9_4_3_switch_compute_partition(adev,
-						    amdgpu_user_partt_mode);
-
 	/* set the virtual and physical id based on partition_mode */
 	gfx_v9_4_3_xcc_program_xcc_id(adev, xcc_id);
 
@@ -1931,6 +1889,9 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 {
 	int r, i, num_xcc;
 
+	if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr) == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
+		amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, amdgpu_user_partt_mode);
+
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		r = gfx_v9_4_3_xcc_cp_resume(adev, i);
@@ -2146,8 +2107,6 @@ static int gfx_v9_4_3_early_init(void *handle)
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
-	adev->gfx.partition_mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
-
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
 	gfx_v9_4_3_set_kiq_pm4_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 293787290e36b..7a963d0a34e2a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -34,6 +34,7 @@
 #include "kfd_smi_events.h"
 #include "kfd_migrate.h"
 #include "amdgpu.h"
+#include "amdgpu_xcp.h"
 
 #define MQD_SIZE_ALIGNED 768
 
@@ -592,7 +593,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	struct kfd_node *node;
 	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
 	unsigned int max_proc_per_quantum;
-	int num_xcd;
+	int num_xcd, partition_mode;
 
 	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev,
 			KGD_ENGINE_MEC1);
@@ -644,8 +645,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	 * If the VMID range changes for GFX9.4.3, then this code MUST be
 	 * revisited.
 	 */
+	partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr);
 	if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
-	    kfd->adev->gfx.partition_mode == AMDGPU_CPX_PARTITION_MODE &&
+	    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
 	    kfd->num_nodes != 1) {
 		vmid_num_kfd /= 2;
 		first_vmid_kfd = last_vmid_kfd + 1 - vmid_num_kfd*2;
@@ -761,7 +763,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		node->start_xcc_id = node->num_xcc_per_node * i;
 
 		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
-		    kfd->adev->gfx.partition_mode == AMDGPU_CPX_PARTITION_MODE &&
+		    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
 		    kfd->num_nodes != 1) {
 			/* For GFX9.4.3 and CPX mode, first XCD gets VMID range
 			 * 4-9 and second XCD gets VMID range 10-15.
-- 
GitLab


From dfdd6f89c4c7f3315d9cabd9bf80a8174ecb5753 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 23 Sep 2022 15:20:08 +0530
Subject: [PATCH 1149/1791] drm/amdgpu: Add GFXHUB v1.2 XCP funcs

Add functions required for suspend/resume of GFXHUB instances which are
part of an XCP.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 35 ++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index b9316bfb7263b..8ba59ffe0e9f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -21,6 +21,7 @@
  *
  */
 #include "amdgpu.h"
+#include "amdgpu_xcp.h"
 #include "gfxhub_v1_2.h"
 #include "gfxhub_v1_1.h"
 
@@ -638,3 +639,37 @@ const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = {
 	.init = gfxhub_v1_2_init,
 	.get_xgmi_info = gfxhub_v1_2_get_xgmi_info,
 };
+
+static int gfxhub_v1_2_xcp_resume(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool value;
+	int ret;
+
+	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
+		value = false;
+	else
+		value = true;
+
+	gfxhub_v1_2_xcc_set_fault_enable_default(adev, value, inst_mask);
+
+	if (!amdgpu_sriov_vf(adev))
+		ret = gfxhub_v1_2_xcc_gart_enable(adev, inst_mask);
+
+	return ret;
+}
+
+static int gfxhub_v1_2_xcp_suspend(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (!amdgpu_sriov_vf(adev))
+		gfxhub_v1_2_xcc_gart_disable(adev, inst_mask);
+
+	return 0;
+}
+
+struct amdgpu_xcp_ip_funcs gfxhub_v1_2_xcp_funcs = {
+	.suspend = &gfxhub_v1_2_xcp_suspend,
+	.resume = &gfxhub_v1_2_xcp_resume
+};
-- 
GitLab


From 3446cb78f3a8033fda416825282e7cafecc83129 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 23 Sep 2022 15:40:15 +0530
Subject: [PATCH 1150/1791] drm/amdgpu: Add SDMA v4.4.2 XCP funcs

Add functions required to suspend/resume instances of SDMA which
are part of an XCP.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 36 ++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 6a971e15b4b06..184eb79027223 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -27,6 +27,7 @@
 #include <linux/pci.h>
 
 #include "amdgpu.h"
+#include "amdgpu_xcp.h"
 #include "amdgpu_ucode.h"
 #include "amdgpu_trace.h"
 
@@ -2025,3 +2026,38 @@ const struct amdgpu_ip_block_version sdma_v4_4_2_ip_block = {
 	.rev = 0,
 	.funcs = &sdma_v4_4_2_ip_funcs,
 };
+
+static int sdma_v4_4_2_xcp_resume(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int r;
+
+	if (!amdgpu_sriov_vf(adev))
+		sdma_v4_4_2_inst_init_golden_registers(adev, inst_mask);
+
+	r = sdma_v4_4_2_inst_start(adev, inst_mask);
+
+	return r;
+}
+
+static int sdma_v4_4_2_xcp_suspend(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t tmp_mask = inst_mask;
+	int i;
+
+	for_each_inst(i, tmp_mask) {
+		amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
+			       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+	}
+
+	sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask);
+	sdma_v4_4_2_inst_enable(adev, false, inst_mask);
+
+	return 0;
+}
+
+struct amdgpu_xcp_ip_funcs sdma_v4_4_2_xcp_funcs = {
+	.suspend = &sdma_v4_4_2_xcp_suspend,
+	.resume = &sdma_v4_4_2_xcp_resume
+};
-- 
GitLab


From 73c84f7c478aeb98bce494cac73f2d20f4a81c6e Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 23 Sep 2022 16:48:43 +0530
Subject: [PATCH 1151/1791] drm/amdgpu: Add XCP functions for GFX v9.4.3

Add functions to suspend/resume GFX instances belonging to an XCP.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 47 +++++++++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index d684037a7a5d8..aaa67592bbb5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -23,6 +23,7 @@
 #include <linux/firmware.h>
 
 #include "amdgpu.h"
+#include "amdgpu_xcp.h"
 #include "amdgpu_gfx.h"
 #include "soc15.h"
 #include "soc15d.h"
@@ -3177,3 +3178,49 @@ const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block = {
 	.rev = 0,
 	.funcs = &gfx_v9_4_3_ip_funcs,
 };
+
+static int gfx_v9_4_3_xcp_resume(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	uint32_t tmp_mask;
+	int i, r;
+
+	/* TODO : Initialize golden regs */
+	/* gfx_v9_4_3_init_golden_registers(adev); */
+
+	tmp_mask = inst_mask;
+	for_each_inst(i, tmp_mask)
+		gfx_v9_4_3_xcc_constants_init(adev, i);
+
+	tmp_mask = inst_mask;
+	for_each_inst(i, tmp_mask) {
+		r = gfx_v9_4_3_xcc_rlc_resume(adev, i);
+		if (r)
+			return r;
+	}
+
+	tmp_mask = inst_mask;
+	for_each_inst(i, tmp_mask) {
+		r = gfx_v9_4_3_xcc_cp_resume(adev, i);
+		if (r)
+			return r;
+	}
+
+	return 0;
+}
+
+static int gfx_v9_4_3_xcp_suspend(void *handle, uint32_t inst_mask)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
+
+	for_each_inst(i, inst_mask)
+		gfx_v9_4_3_xcc_fini(adev, i);
+
+	return 0;
+}
+
+struct amdgpu_xcp_ip_funcs gfx_v9_4_3_xcp_funcs = {
+	.suspend = &gfx_v9_4_3_xcp_suspend,
+	.resume = &gfx_v9_4_3_xcp_resume
+};
-- 
GitLab


From 845c9b313f3122191c847fcc4092492ce039542a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 23 Sep 2022 17:15:15 +0530
Subject: [PATCH 1152/1791] drm/amdgpu: Add XCP IP callback funcs for each IP

Initialize with the IP specific functions needed for GFXHUB, GFX and
SDMA.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 9 ++++++---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h             | 2 ++
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h            | 2 ++
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h            | 2 ++
 4 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index bbcdececfd2f7..2cbac0bccd802 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -25,6 +25,9 @@
 
 #include "soc15_common.h"
 #include "amdgpu_xcp.h"
+#include "gfx_v9_4_3.h"
+#include "gfxhub_v1_2.h"
+#include "sdma_v4_4_2.h"
 
 #define XCP_INST_MASK(num_inst, xcp_id)                                        \
 	(num_inst ? GENMASK(num_inst - 1, 0) << (xcp_id * num_inst) : 0)
@@ -205,15 +208,15 @@ int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 	switch (ip_id) {
 	case AMDGPU_XCP_GFXHUB:
 		ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id);
-		/* TODO : Assign IP funcs */
+		ip->ip_funcs = &gfxhub_v1_2_xcp_funcs;
 		break;
 	case AMDGPU_XCP_GFX:
 		ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id);
-		/* TODO : Assign IP funcs */
+		ip->ip_funcs = &gfx_v9_4_3_xcp_funcs;
 		break;
 	case AMDGPU_XCP_SDMA:
 		ip->inst_mask = XCP_INST_MASK(num_sdma_xcp, xcp_id);
-		/* TODO : Assign IP funcs */
+		ip->ip_funcs = &sdma_v4_4_2_xcp_funcs;
 		break;
 	case AMDGPU_XCP_VCN:
 		ip->inst_mask = XCP_INST_MASK(num_vcn_xcp, xcp_id);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
index 4b530f4c12953..42d67ee0e7ef6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h
@@ -26,4 +26,6 @@
 
 extern const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block;
 
+extern struct amdgpu_xcp_ip_funcs gfx_v9_4_3_xcp_funcs;
+
 #endif /* __GFX_V9_4_3_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h
index e2d508f5a7ee3..997e9f90c9900 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h
@@ -26,4 +26,6 @@
 
 extern const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs;
 
+extern struct amdgpu_xcp_ip_funcs gfxhub_v1_2_xcp_funcs;
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h
index 4814e8a074d61..d516145529bb8 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h
@@ -27,4 +27,6 @@
 extern const struct amd_ip_funcs sdma_v4_4_2_ip_funcs;
 extern const struct amdgpu_ip_block_version sdma_v4_4_2_ip_block;
 
+extern struct amdgpu_xcp_ip_funcs sdma_v4_4_2_xcp_funcs;
+
 #endif
-- 
GitLab


From 5b03127d4745d6848f208463390e6a76d489eb03 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 24 Nov 2022 14:23:58 +0530
Subject: [PATCH 1153/1791] drm/amdgpu: Skip TMR allocation if not required

On ASICs with PSPv13.0.6, TMR is reserved at boot time. There is no need
to allocate TMR region by driver. However, it's still required to send
SETUP_TMR command to PSP.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 34 +++++++++++++++++++------
 1 file changed, 26 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 1319df5796cc6..863fa331e6ff3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -703,8 +703,13 @@ static void psp_prep_tmr_cmd_buf(struct psp_context *psp,
 				 uint64_t tmr_mc, struct amdgpu_bo *tmr_bo)
 {
 	struct amdgpu_device *adev = psp->adev;
-	uint32_t size = amdgpu_bo_size(tmr_bo);
-	uint64_t tmr_pa = amdgpu_gmc_vram_pa(adev, tmr_bo);
+	uint32_t size = 0;
+	uint64_t tmr_pa = 0;
+
+	if (tmr_bo) {
+		size = amdgpu_bo_size(tmr_bo);
+		tmr_pa = amdgpu_gmc_vram_pa(adev, tmr_bo);
+	}
 
 	if (amdgpu_sriov_vf(psp->adev))
 		cmd->cmd_id = GFX_CMD_ID_SETUP_VMR;
@@ -749,6 +754,16 @@ static int psp_load_toc(struct psp_context *psp,
 	return ret;
 }
 
+static bool psp_boottime_tmr(struct psp_context *psp)
+{
+	switch (psp->adev->ip_versions[MP0_HWIP][0]) {
+	case IP_VERSION(13, 0, 6):
+		return true;
+	default:
+		return false;
+	}
+}
+
 /* Set up Trusted Memory Region */
 static int psp_tmr_init(struct psp_context *psp)
 {
@@ -820,8 +835,9 @@ static int psp_tmr_load(struct psp_context *psp)
 	cmd = acquire_psp_cmd_buf(psp);
 
 	psp_prep_tmr_cmd_buf(psp, cmd, psp->tmr_mc_addr, psp->tmr_bo);
-	DRM_INFO("reserve 0x%lx from 0x%llx for PSP TMR\n",
-		 amdgpu_bo_size(psp->tmr_bo), psp->tmr_mc_addr);
+	if (psp->tmr_bo)
+		DRM_INFO("reserve 0x%lx from 0x%llx for PSP TMR\n",
+			 amdgpu_bo_size(psp->tmr_bo), psp->tmr_mc_addr);
 
 	ret = psp_cmd_submit_buf(psp, NULL, cmd,
 				 psp->fence_buf_mc_addr);
@@ -2080,10 +2096,12 @@ static int psp_hw_start(struct psp_context *psp)
 	if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev))
 		goto skip_pin_bo;
 
-	ret = psp_tmr_init(psp);
-	if (ret) {
-		DRM_ERROR("PSP tmr init failed!\n");
-		return ret;
+	if (!psp_boottime_tmr(psp)) {
+		ret = psp_tmr_init(psp);
+		if (ret) {
+			DRM_ERROR("PSP tmr init failed!\n");
+			return ret;
+		}
 	}
 
 skip_pin_bo:
-- 
GitLab


From fee500fa7cb7e11a4d2d66e75e65e67c156e27c6 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Fri, 18 Nov 2022 14:21:15 +0800
Subject: [PATCH 1154/1791] drm/amdgpu: Fix the KCQ hang when binding back

Just like the KIQ, KCQ need to clear the doorbell related regs as well
to avoid hangs when to load driver again after unloading.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 28 ++++++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index aaa67592bbb5d..ef552c9b19b55 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1670,7 +1670,7 @@ static int gfx_v9_4_3_xcc_kiq_init_register(struct amdgpu_ring *ring,
 	return 0;
 }
 
-static int gfx_v9_4_3_xcc_kiq_fini_register(struct amdgpu_ring *ring,
+static int gfx_v9_4_3_xcc_q_fini_register(struct amdgpu_ring *ring,
 					    int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -1688,7 +1688,7 @@ static int gfx_v9_4_3_xcc_kiq_fini_register(struct amdgpu_ring *ring,
 		}
 
 		if (j == AMDGPU_MAX_USEC_TIMEOUT) {
-			DRM_DEBUG("KIQ dequeue request failed.\n");
+			DRM_DEBUG("%s dequeue request failed.\n", ring->name);
 
 			/* Manual disable if dequeue request times out */
 			WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, 0);
@@ -1793,6 +1793,27 @@ static int gfx_v9_4_3_xcc_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 	return 0;
 }
 
+static int gfx_v9_4_3_xcc_kcq_fini_register(struct amdgpu_device *adev, int xcc_id)
+{
+	struct amdgpu_ring *ring;
+	int j;
+
+	for (j = 0; j < adev->gfx.num_compute_rings; j++) {
+		ring = &adev->gfx.compute_ring[j +  xcc_id * adev->gfx.num_compute_rings];
+		if (!amdgpu_in_reset(adev) && !adev->in_suspend) {
+			mutex_lock(&adev->srbm_mutex);
+			soc15_grbm_select(adev, ring->me,
+					ring->pipe,
+					ring->queue, 0, GET_INST(GC, xcc_id));
+			gfx_v9_4_3_xcc_q_fini_register(ring, xcc_id);
+			soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
+			mutex_unlock(&adev->srbm_mutex);
+		}
+	}
+
+	return 0;
+}
+
 static int gfx_v9_4_3_xcc_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	struct amdgpu_ring *ring;
@@ -1923,12 +1944,13 @@ static void gfx_v9_4_3_xcc_fini(struct amdgpu_device *adev, int xcc_id)
 				  adev->gfx.kiq[xcc_id].ring.pipe,
 				  adev->gfx.kiq[xcc_id].ring.queue, 0,
 				  GET_INST(GC, xcc_id));
-		gfx_v9_4_3_xcc_kiq_fini_register(&adev->gfx.kiq[xcc_id].ring,
+		gfx_v9_4_3_xcc_q_fini_register(&adev->gfx.kiq[xcc_id].ring,
 						 xcc_id);
 		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 	}
 
+	gfx_v9_4_3_xcc_kcq_fini_register(adev, xcc_id);
 	gfx_v9_4_3_xcc_cp_enable(adev, false, xcc_id);
 
 	/* Skip suspend with A+A reset */
-- 
GitLab


From dc6df2095deaaefe38a94d62a51b0d07c0794eaf Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 28 Nov 2022 09:57:51 +0530
Subject: [PATCH 1155/1791] drm/amdgpu: Move generic logic to soc config

Move soc specific configuration details to aqua vanjaram specific file.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 2cbac0bccd802..a6204b588829b 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -359,6 +359,15 @@ int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 {
 	int ret;
 
+	/* generally 1 AID supports 4 instances */
+	adev->sdma.num_inst_per_aid = 4;
+	adev->sdma.num_instances = NUM_SDMA(adev->sdma.sdma_mask);
+
+	adev->vcn.num_inst_per_aid = 1;
+	adev->vcn.num_vcn_inst = adev->vcn.num_inst_per_aid * adev->num_aid;
+	adev->jpeg.num_inst_per_aid = 1;
+	adev->jpeg.num_jpeg_inst = adev->jpeg.num_inst_per_aid * adev->num_aid;
+
 	ret = aqua_vanjaram_xcp_mgr_init(adev);
 	if (ret)
 		return ret;
-- 
GitLab


From bbca579fd2ea8cbc170df33587f8a4b572a4f025 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 28 Nov 2022 11:17:15 +0530
Subject: [PATCH 1156/1791] drm/amdgpu: Derive active clusters from SDMA

SDMA instances per active cluster and SDMA instance mask are used
to find the number of active clusters.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index a6204b588829b..6f7226b5d4461 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -357,12 +357,22 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
 
 int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 {
+	u32 inst_mask = adev->sdma.sdma_mask;
 	int ret;
 
 	/* generally 1 AID supports 4 instances */
 	adev->sdma.num_inst_per_aid = 4;
 	adev->sdma.num_instances = NUM_SDMA(adev->sdma.sdma_mask);
 
+	adev->num_aid = 1;
+	inst_mask >>= adev->sdma.num_inst_per_aid;
+
+	for (const u32 mask = (1 << adev->sdma.num_inst_per_aid) - 1; inst_mask;
+	     inst_mask >>= adev->sdma.num_inst_per_aid) {
+		if ((inst_mask & mask) == mask)
+			adev->num_aid++;
+	}
+
 	adev->vcn.num_inst_per_aid = 1;
 	adev->vcn.num_vcn_inst = adev->vcn.num_inst_per_aid * adev->num_aid;
 	adev->jpeg.num_inst_per_aid = 1;
-- 
GitLab


From 7a1efad04c210594069c4ab9f9c25039cd6915e4 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 29 Nov 2022 14:00:37 +0530
Subject: [PATCH 1157/1791] drm/amdgpu: Use mask for active clusters

Use a mask of available active clusters instead of using only the number
of active clusters.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h           |  3 +-
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 13 +++--
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c         |  5 +-
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c       | 56 ++++++++++++-------
 4 files changed, 49 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 3858d29baef10..279057ec7a0b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1070,7 +1070,8 @@ struct amdgpu_device {
 
 	bool                            job_hang;
 	bool                            dc_enabled;
-	uint32_t			num_aid;
+	/* Mask of active clusters */
+	uint32_t			aid_mask;
 };
 
 static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev)
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 6f7226b5d4461..0d7bc212def13 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -358,25 +358,26 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
 int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 {
 	u32 inst_mask = adev->sdma.sdma_mask;
-	int ret;
+	int ret, i, num_inst;
 
 	/* generally 1 AID supports 4 instances */
 	adev->sdma.num_inst_per_aid = 4;
 	adev->sdma.num_instances = NUM_SDMA(adev->sdma.sdma_mask);
 
-	adev->num_aid = 1;
+	adev->aid_mask = i = 1;
 	inst_mask >>= adev->sdma.num_inst_per_aid;
 
 	for (const u32 mask = (1 << adev->sdma.num_inst_per_aid) - 1; inst_mask;
-	     inst_mask >>= adev->sdma.num_inst_per_aid) {
+	     inst_mask >>= adev->sdma.num_inst_per_aid, ++i) {
 		if ((inst_mask & mask) == mask)
-			adev->num_aid++;
+			adev->aid_mask |= (1 << i);
 	}
 
+	num_inst = hweight32(adev->aid_mask);
 	adev->vcn.num_inst_per_aid = 1;
-	adev->vcn.num_vcn_inst = adev->vcn.num_inst_per_aid * adev->num_aid;
+	adev->vcn.num_vcn_inst = adev->vcn.num_inst_per_aid * num_inst;
 	adev->jpeg.num_inst_per_aid = 1;
-	adev->jpeg.num_jpeg_inst = adev->jpeg.num_inst_per_aid * adev->num_aid;
+	adev->jpeg.num_jpeg_inst = adev->jpeg.num_inst_per_aid * num_inst;
 
 	ret = aqua_vanjaram_xcp_mgr_init(adev);
 	if (ret)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 0792c48fe347d..b3f64f2f306d7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1674,6 +1674,7 @@ static int gmc_v9_0_sw_init(void *handle)
 {
 	int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	unsigned long inst_mask = adev->aid_mask;
 
 	adev->gfxhub.funcs->init(adev);
 
@@ -1757,7 +1758,9 @@ static int gmc_v9_0_sw_init(void *handle)
 	case IP_VERSION(9, 4, 3):
 		bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0),
 				  NUM_XCC(adev->gfx.xcc_mask));
-		bitmap_set(adev->vmhubs_mask, AMDGPU_MMHUB0(0), adev->num_aid);
+
+		inst_mask <<= AMDGPU_MMHUB0(0);
+		bitmap_or(adev->vmhubs_mask, adev->vmhubs_mask, &inst_mask, 32);
 
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 6f469b9aa9a02..a530e2a3cc287 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -54,9 +54,11 @@ static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmi
 				uint64_t page_table_base)
 {
 	struct amdgpu_vmhub *hub;
+	u32 inst_mask;
 	int i;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
 		WREG32_SOC15_OFFSET(MMHUB, i,
 				    regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
@@ -73,6 +75,7 @@ static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmi
 static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
 {
 	uint64_t pt_base;
+	u32 inst_mask;
 	int i;
 
 	if (adev->gmc.pdb0_bo)
@@ -85,7 +88,8 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
 	/* If use GART for FB translation, vmid0 page table covers both
 	 * vram and system memory (gart)
 	 */
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		if (adev->gmc.pdb0_bo) {
 			WREG32_SOC15(MMHUB, i,
 				     regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
@@ -121,11 +125,12 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_init_system_aperture_regs(struct amdgpu_device *adev)
 {
+	uint32_t tmp, inst_mask;
 	uint64_t value;
-	uint32_t tmp;
 	int i;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		/* Program the AGP BAR */
 		WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BASE, 0);
 		WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BOT,
@@ -183,11 +188,12 @@ static void mmhub_v1_8_init_system_aperture_regs(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
 {
-	uint32_t tmp;
+	uint32_t tmp, inst_mask;
 	int i;
 
 	/* Setup TLB control */
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL);
 
 		tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB,
@@ -208,14 +214,15 @@ static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
 {
-	uint32_t tmp;
+	uint32_t tmp, inst_mask;
 	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return;
 
 	/* Setup L2 cache */
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1);
 		tmp = REG_SET_FIELD(tmp, VM_L2_CNTL,
@@ -267,10 +274,11 @@ static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev)
 {
-	uint32_t tmp;
+	uint32_t tmp, inst_mask;
 	int i;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		tmp = RREG32_SOC15(MMHUB, i, regVM_CONTEXT0_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1);
 		tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH,
@@ -286,12 +294,14 @@ static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev)
 
 static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev)
 {
+	u32 inst_mask;
 	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		WREG32_SOC15(MMHUB, i,
 			     regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32,
 			     0XFFFFFFFF);
@@ -317,7 +327,7 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
 	unsigned num_level, block_size;
-	uint32_t tmp;
+	uint32_t tmp, inst_mask;
 	int i, j;
 
 	num_level = adev->vm_manager.num_level;
@@ -327,7 +337,8 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 	else
 		block_size -= 9;
 
-	for (j = 0; j < adev->num_aid; j++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(j, inst_mask) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
 		for (i = 0; i <= 14; i++) {
 			tmp = RREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT1_CNTL,
@@ -382,9 +393,10 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
-	unsigned i, j;
+	u32 i, j, inst_mask;
 
-	for (j = 0; j < adev->num_aid; j++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(j, inst_mask) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
 		for (i = 0; i < 18; ++i) {
 			WREG32_SOC15_OFFSET(MMHUB, j,
@@ -429,10 +441,11 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
 	u32 tmp;
-	u32 i, j;
+	u32 i, j, inst_mask;
 
 	/* Disable all tables */
-	for (j = 0; j < adev->num_aid; j++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(j, inst_mask) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(j)];
 		for (i = 0; i < 16; i++)
 			WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT0_CNTL,
@@ -465,13 +478,14 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev)
  */
 static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool value)
 {
-	u32 tmp;
+	u32 tmp, inst_mask;
 	int i;
 
 	if (amdgpu_sriov_vf(adev))
 		return;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		tmp = RREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL);
 		tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL,
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
@@ -510,9 +524,11 @@ static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool
 static void mmhub_v1_8_init(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
+	u32 inst_mask;
 	int i;
 
-	for (i = 0; i < adev->num_aid; i++) {
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask) {
 		hub = &adev->vmhub[AMDGPU_MMHUB0(i)];
 
 		hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, i,
-- 
GitLab


From 753b999afe47900531282f86bf430aec250b4232 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Wed, 7 Dec 2022 00:29:40 -0500
Subject: [PATCH 1158/1791] drm/amdgpu: set MTYPE in PTE for GFXIP 9.4.3

Apply the GFXIP 9.4.3 specific snoop and mtype settings for various
scenarios such as APU, APU in Carveout mode and dGPU mode.

Note: This is expected to change due to:
1 - NPS > 1 support in future
2 - Hardware bugs found during initial asic bringup.

Cc: Graham Sider <graham.sider@amd.com>
Cc: Hawking Zhang <hawking.zhang@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 40 ++++++++++++++++++++++++++-
 1 file changed, 39 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index b3f64f2f306d7..3765178e6fc52 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1173,7 +1173,6 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 1):
 	case IP_VERSION(9, 4, 2):
-	case IP_VERSION(9, 4, 3):
 		if (is_vram) {
 			if (bo_adev == adev) {
 				if (uncached)
@@ -1207,6 +1206,45 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 			 */
 			snoop = true;
 		}
+		break;
+	case IP_VERSION(9, 4, 3):
+		/* FIXME: Needs more work for handling multiple memory
+		 * partitions (> NPS1 mode) e.g. NPS4 for both APU and dGPU
+		 * modes.
+		 */
+		snoop = true;
+		if (uncached) {
+			mtype = MTYPE_UC;
+		} else if (adev->gmc.is_app_apu) {
+			/* FIXME: APU in native mode, NPS1 single socket only
+			 *
+			 * For suporting NUMA partitioned APU e.g. in NPS4 mode,
+			 * this need to look at the NUMA node on which the
+			 * system memory allocation was done.
+			 *
+			 * Memory access by a different partition within same
+			 * socket should be treated as remote access so MTYPE_RW
+			 * cannot be used always.
+			 */
+			mtype = MTYPE_RW;
+		} else if (adev->flags & AMD_IS_APU) {
+			/* APU on carve out mode */
+			mtype = MTYPE_RW;
+		} else {
+			/* dGPU */
+			/*
+			if ((mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) &&
+			    bo_adev == adev)
+				mapping_flags |= AMDGPU_VM_MTYPE_RW;
+			else
+			*/
+			/* Temporarily comment out above lines and use MTYPE_NC
+			 * on both VRAM and system memory access until
+			 * MTYPE_RW can properly work on VRAM access
+			 */
+			mtype = MTYPE_NC;
+		}
+
 		break;
 	default:
 		if (uncached || coherent)
-- 
GitLab


From 1794e9d7e78cb52605234d0ddc3f46084937f4e9 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Thu, 8 Dec 2022 12:08:17 -0500
Subject: [PATCH 1159/1791] drm/amdkfd: Update packet manager for GFX9.4.3

In GFX 9.4.3, there can be more than 8 SDMA engines.
As a result, extended_engine_sel and engine_sel fields
in MAP_QUEUES packet need to be updated to allow correct
mapping of SDMA queues to these SDMA engines.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c   | 16 +++++++++++++---
 drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h  |  3 ++-
 2 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 54d7d4665ad22..44cf3a5f6fdba 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -225,9 +225,19 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 			packet->bitfields2.engine_sel = q->properties.sdma_engine_id +
 				engine_sel__mes_map_queues__sdma0_vi;
 		else {
-			packet->bitfields2.extended_engine_sel =
-				extended_engine_sel__mes_map_queues__sdma0_to_7_sel;
-			packet->bitfields2.engine_sel = q->properties.sdma_engine_id;
+			/*
+			 * For GFX9.4.3, SDMA engine id can be greater than 8.
+			 * For such cases, set extended_engine_sel to 2 and
+			 * ensure engine_sel lies between 0-7.
+			 */
+			if (q->properties.sdma_engine_id >= 8)
+				packet->bitfields2.extended_engine_sel =
+					extended_engine_sel__mes_map_queues__sdma8_to_15_sel;
+			else
+				packet->bitfields2.extended_engine_sel =
+					extended_engine_sel__mes_map_queues__sdma0_to_7_sel;
+
+			packet->bitfields2.engine_sel = q->properties.sdma_engine_id % 8;
 		}
 		break;
 	default:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
index a666710ed4035..2ad708c640125 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
@@ -263,7 +263,8 @@ enum mes_map_queues_engine_sel_enum {
 
 enum mes_map_queues_extended_engine_sel_enum {
 	extended_engine_sel__mes_map_queues__legacy_engine_sel = 0,
-	extended_engine_sel__mes_map_queues__sdma0_to_7_sel = 1
+	extended_engine_sel__mes_map_queues__sdma0_to_7_sel  = 1,
+	extended_engine_sel__mes_map_queues__sdma8_to_15_sel = 2
 };
 
 struct pm4_mes_map_queues {
-- 
GitLab


From 98b2e9cad2279132e3aa4b9caf9164b2e35c1a52 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 9 Dec 2022 19:44:05 +0800
Subject: [PATCH 1160/1791] drm/amdgpu: correct the vmhub index when page fault
 occurs

The AMDGPU_GFXHUB was bind to each xcc in the logical order.
Thus convert the node_id to logical xcc_id to index the
correct AMDGPU_GFXHUB. And "node_id / 4" can get the correct
AMDGPU_MMHUB0 index.

Signed-off-by: Le Ma <le.ma@amd.com>
Tested-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 27 +++++++++++++------------
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 14 +++++++++----
 3 files changed, 25 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 2287768ed141a..81b4c7e684af2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -280,6 +280,7 @@ struct amdgpu_gfx_funcs {
 			(*query_mem_partition_mode)(struct amdgpu_device *adev);
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
 				     int num_xccs_per_xcp);
+	int (*ih_node_to_logical_xcc)(struct amdgpu_device *adev, int ih_node);
 };
 
 struct sq_work {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ef552c9b19b55..6aaa810ea0446 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -637,6 +637,19 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node)
+{
+	int xcc;
+
+	xcc = hweight8(adev->gfx.xcc_mask & GENMASK(ih_node / 2, 0));
+	if (!xcc) {
+		dev_err(adev->dev, "Couldn't find xcc mapping from IH node");
+		return -EINVAL;
+	}
+
+	return xcc - 1;
+}
+
 static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter,
 	.select_se_sh = &gfx_v9_4_3_xcc_select_se_sh,
@@ -646,6 +659,7 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
 	.switch_partition_mode = &gfx_v9_4_3_switch_compute_partition,
 	.query_mem_partition_mode = &gfx_v9_4_3_query_memory_partition,
+	.ih_node_to_logical_xcc = &gfx_v9_4_3_ih_to_xcc_inst,
 };
 
 static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
@@ -2754,19 +2768,6 @@ static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
-static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node)
-{
-	int xcc;
-
-	xcc = hweight8(adev->gfx.xcc_mask & GENMASK(ih_node / 2, 0));
-	if (!xcc) {
-		dev_err(adev->dev, "Couldn't find xcc mapping from IH node");
-		return -EINVAL;
-	}
-
-	return xcc - 1;
-}
-
 static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev,
 			    struct amdgpu_irq_src *source,
 			    struct amdgpu_iv_entry *entry)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 3765178e6fc52..841333148610e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -557,22 +557,28 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	u64 addr;
 	uint32_t cam_index = 0;
 	int ret;
-	uint32_t node_id;
+	uint32_t node_id, xcc_id = 0;
 
-	node_id = (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) ? entry->node_id : 0;
+	node_id = entry->node_id;
 
 	addr = (u64)entry->src_data[0] << 12;
 	addr |= ((u64)entry->src_data[1] & 0xf) << 44;
 
 	if (entry->client_id == SOC15_IH_CLIENTID_VMC) {
 		hub_name = "mmhub0";
-		hub = &adev->vmhub[AMDGPU_MMHUB0(0)];
+		hub = &adev->vmhub[AMDGPU_MMHUB0(node_id / 4)];
 	} else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) {
 		hub_name = "mmhub1";
 		hub = &adev->vmhub[AMDGPU_MMHUB1(0)];
 	} else {
 		hub_name = "gfxhub0";
-		hub = &adev->vmhub[node_id/2];
+		if (adev->gfx.funcs->ih_node_to_logical_xcc) {
+			xcc_id = adev->gfx.funcs->ih_node_to_logical_xcc(adev,
+				node_id);
+			if (xcc_id < 0)
+				xcc_id = 0;
+		}
+		hub = &adev->vmhub[xcc_id];
 	}
 
 	if (retry_fault) {
-- 
GitLab


From a820d3ca8ed3a2851e1c9d5713e807f84e88019d Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 1 Dec 2022 17:22:01 +0530
Subject: [PATCH 1161/1791] drm/amdgpu: Remove unnecessary return value check

There is no need to check return value, as the function internally
used - amdgpu_discovery_read_binary_from_vram() - returns void.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index e6d10a3f1753c..5b6088fc2b2bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -201,14 +201,13 @@ static int hw_id_map[MAX_HWIP] = {
 	[PCIE_HWIP]	= PCIE_HWID,
 };
 
-static int amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary)
+static void amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary)
 {
 	uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20;
 	uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET;
 
 	amdgpu_device_vram_access(adev, pos, (uint32_t *)binary,
 				  adev->mman.discovery_tmr_size, false);
-	return 0;
 }
 
 static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, uint8_t *binary)
@@ -302,12 +301,7 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	if (!adev->mman.discovery_bin)
 		return -ENOMEM;
 
-	r = amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin);
-	if (r) {
-		dev_err(adev->dev, "failed to read ip discovery binary from vram\n");
-		r = -EINVAL;
-		goto out;
-	}
+	amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin);
 
 	if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin) || amdgpu_discovery == 2) {
 		/* ignore the discovery binary from vram if discovery=2 in kernel module parameter */
-- 
GitLab


From c2d43918a14f7b0f04932f5a45728e0fe8161da0 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Fri, 9 Dec 2022 09:03:01 -0500
Subject: [PATCH 1162/1791] drm/amdkfd: Setup current_logical_xcc_id in MQD

Setup rolling current_logical_xcc_id in MQD for GFX9.4.3
to ensure each queue starts at a different place and prevent
hotspotting issues. Also, remove updating current_logical_xcc_id
during queue update.

Suggested-by: Joseph Greathouse <Joseph.Greathouse@amd.com>
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.h   |  3 +++
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c     | 13 +++++--------
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index b11c474d40671..cd4383bb207f4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -256,6 +256,9 @@ struct device_queue_manager {
 	struct work_struct	hw_exception_work;
 	struct kfd_mem_obj	hiq_sdma_mqd;
 	bool			sched_running;
+
+	/* used for GFX 9.4.3 only */
+	uint32_t		current_logical_xcc_start;
 };
 
 void device_queue_manager_init_cik(
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 2e2a0f8586f7a..c781314b213c5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -32,6 +32,7 @@
 #include "gc/gc_9_0_sh_mask.h"
 #include "sdma0/sdma0_4_0_sh_mask.h"
 #include "amdgpu_amdkfd.h"
+#include "kfd_device_queue_manager.h"
 
 static void update_mqd(struct mqd_manager *mm, void *mqd,
 		       struct queue_properties *q,
@@ -569,6 +570,7 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 	uint64_t xcc_gart_addr = 0;
 	uint64_t xcc_ctx_save_restore_area_address;
 	uint64_t offset = mm->mqd_stride(mm, q);
+	uint32_t local_xcc_start = mm->dev->dqm->current_logical_xcc_start++;
 
 	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
 	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
@@ -596,18 +598,17 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 
 		if (q->format == KFD_QUEUE_FORMAT_AQL) {
 			m->compute_tg_chunk_size = 1;
+			m->compute_current_logic_xcc_id =
+					(local_xcc_start + xcc) %
+					mm->dev->num_xcc_per_node;
 
 			switch (xcc) {
 			case 0:
 				/* Master XCC */
 				m->cp_hqd_pq_control &=
 					~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
-				m->compute_current_logic_xcc_id =
-					mm->dev->num_xcc_per_node - 1;
 				break;
 			default:
-				m->compute_current_logic_xcc_id =
-					xcc - 1;
 				break;
 			}
 		} else {
@@ -642,12 +643,8 @@ static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 				/* Master XCC */
 				m->cp_hqd_pq_control &=
 					~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK;
-				m->compute_current_logic_xcc_id =
-					mm->dev->num_xcc_per_node - 1;
 				break;
 			default:
-				m->compute_current_logic_xcc_id =
-					xcc - 1;
 				break;
 			}
 			m->compute_tg_chunk_size = 1;
-- 
GitLab


From a0a0c69c05bff025abf49ec66b2bfb94aeabcc6e Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 14 Dec 2022 10:28:50 +0530
Subject: [PATCH 1163/1791] drm/amdgpu: Fix semaphore release

Use the right register for semaphore release during invalidation.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 841333148610e..1e43641208450 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -915,9 +915,9 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		 * write with 0 means semaphore release
 		 */
 		if (vmhub >= AMDGPU_MMHUB0(0))
-			WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
+			WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0);
 		else
-			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req);
+			WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0);
 	}
 
 	spin_unlock(&adev->gmc.invalidate_lock);
-- 
GitLab


From 12c4d7edfb7238ded6c7a2584995d888b4d877ec Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 15 Dec 2022 13:13:29 +0530
Subject: [PATCH 1164/1791] drm/amdgpu: Fix GFX 9.4.3 dma address capability
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

ASICs with GFX 9.4.3 support 48-bit addressing.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 1e43641208450..444441c6b7e39 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1846,7 +1846,7 @@ static int gmc_v9_0_sw_init(void *handle)
 	 */
 	adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */
 
-	dma_addr_bits = adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ? 48:44;
+	dma_addr_bits = adev->ip_versions[GC_HWIP][0] >= IP_VERSION(9, 4, 2) ? 48:44;
 	r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(dma_addr_bits));
 	if (r) {
 		printk(KERN_WARNING "amdgpu: No suitable DMA available.\n");
-- 
GitLab


From d524180b88009d9158bff7fd20f3916455e0c32c Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 19 Dec 2022 17:39:42 +0530
Subject: [PATCH 1165/1791] drm/amdgpu: Fix GFX v9.4.3 EOP buffer allocation

Each compute cluster gets 8 compute queues in GFX v9.4.3. Fix the EOP
buffer allocation so that compute queue on every XCC gets a unique
address.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Tested-and-Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 6aaa810ea0446..b56fa2945464a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -455,7 +455,8 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 
 	/* take ownership of the relevant compute queues */
 	amdgpu_gfx_compute_queue_acquire(adev);
-	mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
+	mec_hpd_size =
+		adev->gfx.num_compute_rings * num_xcc * GFX9_MEC_HPD_SIZE;
 	if (mec_hpd_size) {
 		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
 					      AMDGPU_GEM_DOMAIN_VRAM,
-- 
GitLab


From 85b45b60722f506322393320bb6cc195378f2e4f Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Mon, 28 Nov 2022 11:26:02 -0500
Subject: [PATCH 1166/1791] amd/amdgpu: Set MTYPE_UC for access over PCIe

For GFX v9_4_3, set MTYPE_UC for memory access over PCIe.

v4 - add missing indentation pointed out by Felix and add his
reviewed-by tag.
v3 - add missing logic for the svm path.
v2 - add amdgpu_xgmi_same_hive to separate access over xgmi from pcie

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 29 ++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 5d6e02559d8ee..6daba0582bf37 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1195,16 +1195,29 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		//e.g. NPS4. Current approch is only applicable without memory
 		//partitions.
 		snoop = true;
-		if (uncached)
+		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
-		/* local HBM region close to partition*/
-		else if (bo_node == node)
-			mapping_flags |= AMDGPU_VM_MTYPE_RW;
-		/* local HBM region far from partition or remote XGMI GPU or
-		 * system memory
-		 */
-		else
+		} else if (domain == SVM_RANGE_VRAM_DOMAIN) {
+			/* local HBM region close to partition with a workaround
+			 * for Endpoint systems.
+			 */
+			if (bo_node == node)
+				mapping_flags |=
+					(node->adev->flags & AMD_IS_APU) ?
+					AMDGPU_VM_MTYPE_RW : AMDGPU_VM_MTYPE_NC;
+			/* local HBM region far from partition or remote XGMI GPU */
+			else if (svm_nodes_in_same_hive(bo_node, node))
+				mapping_flags |= AMDGPU_VM_MTYPE_NC;
+			/* PCIe P2P */
+			else
+				mapping_flags |= AMDGPU_VM_MTYPE_UC;
+		/* system memory accessed by the APU */
+		} else if (node->adev->flags & AMD_IS_APU) {
 			mapping_flags |= AMDGPU_VM_MTYPE_NC;
+		/* system memory accessed by the dGPU */
+		} else {
+			mapping_flags |= AMDGPU_VM_MTYPE_UC;
+		}
 		break;
 	default:
 		mapping_flags |= coherent ?
-- 
GitLab


From 7a7aaab021a6772b29c81c22db9f4d2d8cd66ecd Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Tue, 20 Dec 2022 15:37:57 -0500
Subject: [PATCH 1167/1791] drm/amdgpu: Make UTCL2 snoop CPU caches

On AMD APP APUs, to make UTCL2 snoop CPU caches, its not sufficient to
rely on xgmi connected flag so add the logic to use is_app_apu to
program the PDE_REQUEST_PHYSICAL bit correctly for gfxhub and mmhub
both.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 3 ++-
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c  | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index 8ba59ffe0e9f5..8901e73fd7005 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -252,7 +252,8 @@ static void gfxhub_v1_2_xcc_init_cache_regs(struct amdgpu_device *adev,
 		WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL3, tmp);
 
 		tmp = regVM_L2_CNTL4_DEFAULT;
-		if (adev->gmc.xgmi.connected_to_cpu) {
+		/* For AMD APP APUs setup WC memory */
+		if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) {
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1);
 		} else {
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index a530e2a3cc287..a8faf66b68780 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -257,7 +257,8 @@ static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
 		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL3, tmp);
 
 		tmp = regVM_L2_CNTL4_DEFAULT;
-		if (adev->gmc.xgmi.connected_to_cpu) {
+		/* For AMD APP APUs setup WC memory */
+		if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) {
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
 					    VMC_TAP_PDE_REQUEST_PHYSICAL, 1);
 			tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4,
-- 
GitLab


From 34fd9d686772f6725242e900913ca2be987c12dd Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 20 Dec 2022 14:21:57 +0530
Subject: [PATCH 1168/1791] drm/amdgpu: Add FGCG logic for GFX v9.4.3

Add logic for fine grain clock gating logic for GFX v9.4.3. The feature
will be controlled using CG flags. Also, make a change so that RLC safe
mode entry/exit is done only once during CG update sequence.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 71 ++++++++++++++++++++++---
 1 file changed, 65 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index b56fa2945464a..55d99c4ea48c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2172,14 +2172,64 @@ static int gfx_v9_4_3_late_init(void *handle)
 	return 0;
 }
 
+static void gfx_v9_4_3_xcc_update_sram_fgcg(struct amdgpu_device *adev,
+					    bool enable, int xcc_id)
+{
+	uint32_t def, data;
+
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_GFX_FGCG))
+		return;
+
+	def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id),
+				  regRLC_CGTT_MGCG_OVERRIDE);
+
+	if (enable)
+		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK;
+	else
+		data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK;
+
+	if (def != data)
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id),
+			     regRLC_CGTT_MGCG_OVERRIDE, data);
+
+	def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL);
+
+	if (enable)
+		data &= ~RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK;
+	else
+		data |= RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK;
+
+	if (def != data)
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL, data);
+}
+
+static void gfx_v9_4_3_xcc_update_repeater_fgcg(struct amdgpu_device *adev,
+						bool enable, int xcc_id)
+{
+	uint32_t def, data;
+
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_REPEATER_FGCG))
+		return;
+
+	def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id),
+				  regRLC_CGTT_MGCG_OVERRIDE);
+
+	if (enable)
+		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_REP_FGCG_OVERRIDE_MASK;
+	else
+		data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_REP_FGCG_OVERRIDE_MASK;
+
+	if (def != data)
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id),
+			     regRLC_CGTT_MGCG_OVERRIDE, data);
+}
+
 static void
 gfx_v9_4_3_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 						bool enable, int xcc_id)
 {
 	uint32_t data, def;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
-
 	/* It is disabled by HW by default */
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
 		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
@@ -2239,7 +2289,6 @@ gfx_v9_4_3_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 		}
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
 static void
@@ -2248,8 +2297,6 @@ gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
 {
 	uint32_t def, data;
 
-	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
-
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
 		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE);
 		/* unset CGCG override */
@@ -2292,13 +2339,18 @@ gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
 			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data);
 	}
 
-	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 }
 
 static int gfx_v9_4_3_xcc_update_gfx_clock_gating(struct amdgpu_device *adev,
 						  bool enable, int xcc_id)
 {
+	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
+
 	if (enable) {
+		/* FGCG */
+		gfx_v9_4_3_xcc_update_sram_fgcg(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_repeater_fgcg(adev, enable, xcc_id);
+
 		/* CGCG/CGLS should be enabled after MGCG/MGLS
 		 * ===  MGCG + MGLS ===
 		 */
@@ -2316,7 +2368,14 @@ static int gfx_v9_4_3_xcc_update_gfx_clock_gating(struct amdgpu_device *adev,
 		/* ===  MGCG + MGLS === */
 		gfx_v9_4_3_xcc_update_medium_grain_clock_gating(adev, enable,
 								xcc_id);
+
+		/* FGCG */
+		gfx_v9_4_3_xcc_update_sram_fgcg(adev, enable, xcc_id);
+		gfx_v9_4_3_xcc_update_repeater_fgcg(adev, enable, xcc_id);
 	}
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
+
 	return 0;
 }
 
-- 
GitLab


From c9a502e981a961053f3f873b14677d95e804251e Mon Sep 17 00:00:00 2001
From: Felix Kuehling <Felix.Kuehling@amd.com>
Date: Tue, 29 Nov 2022 12:45:26 -0500
Subject: [PATCH 1169/1791] drm/amdgpu: Allocate GART table in RAM for AMD APU

Some AMD APUs may not have a dedicated VRAM. On such platforms the GART
table should be allocated on the system memory. When real vram size is
zero, place the GART table in system memory and create an SG BO to make
it GPU accessible.

v2: fix includes

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
(rajneesh: removed set_memory_wc workaround)
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c | 137 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h |   2 +
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c    |  23 +++-
 3 files changed, 156 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index 6b12f4a75fc37..a070adf30c88c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -35,6 +35,7 @@
 #endif
 #include "amdgpu.h"
 #include <drm/drm_drv.h>
+#include <drm/ttm/ttm_tt.h>
 
 /*
  * GART
@@ -102,6 +103,142 @@ void amdgpu_gart_dummy_page_fini(struct amdgpu_device *adev)
 	adev->dummy_page_addr = 0;
 }
 
+/**
+ * amdgpu_gart_table_ram_alloc - allocate system ram for gart page table
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Allocate system memory for GART page table for ASICs that don't have
+ * dedicated VRAM.
+ * Returns 0 for success, error for failure.
+ */
+int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev)
+{
+	unsigned int order = get_order(adev->gart.table_size);
+	gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO;
+	struct amdgpu_bo *bo = NULL;
+	struct sg_table *sg = NULL;
+	struct amdgpu_bo_param bp;
+	dma_addr_t dma_addr;
+	struct page *p;
+	int ret;
+
+	if (adev->gart.bo != NULL)
+		return 0;
+
+	p = alloc_pages(gfp_flags, order);
+	if (!p)
+		return -ENOMEM;
+
+	/* If the hardware does not support UTCL2 snooping of the CPU caches
+	 * then set_memory_wc() could be used as a workaround to mark the pages
+	 * as write combine memory.
+	 */
+	dma_addr = dma_map_page(&adev->pdev->dev, p, 0, adev->gart.table_size,
+				DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(&adev->pdev->dev, dma_addr)) {
+		dev_err(&adev->pdev->dev, "Failed to DMA MAP the GART BO page\n");
+		__free_pages(p, order);
+		p = NULL;
+		return -EFAULT;
+	}
+
+	dev_info(adev->dev, "%s dma_addr:%llx\n", __func__, dma_addr);
+	/* Create SG table */
+	sg = kmalloc(sizeof(*sg), GFP_KERNEL);
+	if (!sg) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	ret = sg_alloc_table(sg, 1, GFP_KERNEL);
+	if (ret)
+		goto error;
+
+	sg_dma_address(sg->sgl) = dma_addr;
+	sg->sgl->length = adev->gart.table_size;
+#ifdef CONFIG_NEED_SG_DMA_LENGTH
+	sg->sgl->dma_length = adev->gart.table_size;
+#endif
+	/* Create SG BO */
+	memset(&bp, 0, sizeof(bp));
+	bp.size = adev->gart.table_size;
+	bp.byte_align = PAGE_SIZE;
+	bp.domain = AMDGPU_GEM_DOMAIN_CPU;
+	bp.type = ttm_bo_type_sg;
+	bp.resv = NULL;
+	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+	bp.flags = 0;
+	ret = amdgpu_bo_create(adev, &bp, &bo);
+	if (ret)
+		goto error;
+
+	bo->tbo.sg = sg;
+	bo->tbo.ttm->sg = sg;
+	bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
+	bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
+
+	ret = amdgpu_bo_reserve(bo, true);
+	if (ret) {
+		dev_err(adev->dev, "(%d) failed to reserve bo for GART system bo\n", ret);
+		goto error;
+	}
+
+	ret = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
+	WARN(ret, "Pinning the GART table failed");
+	if (ret)
+		goto error_resv;
+
+	adev->gart.bo = bo;
+	adev->gart.ptr = page_to_virt(p);
+	/* Make GART table accessible in VMID0 */
+	ret = amdgpu_ttm_alloc_gart(&adev->gart.bo->tbo);
+	if (ret)
+		amdgpu_gart_table_ram_free(adev);
+	amdgpu_bo_unreserve(bo);
+
+	return 0;
+
+error_resv:
+	amdgpu_bo_unreserve(bo);
+error:
+	amdgpu_bo_unref(&bo);
+	if (sg) {
+		sg_free_table(sg);
+		kfree(sg);
+	}
+	__free_pages(p, order);
+	return ret;
+}
+
+/**
+ * amdgpu_gart_table_ram_free - free gart page table system ram
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Free the system memory used for the GART page tableon ASICs that don't
+ * have dedicated VRAM.
+ */
+void amdgpu_gart_table_ram_free(struct amdgpu_device *adev)
+{
+	unsigned int order = get_order(adev->gart.table_size);
+	struct sg_table *sg = adev->gart.bo->tbo.sg;
+	struct page *p;
+	int ret;
+
+	ret = amdgpu_bo_reserve(adev->gart.bo, false);
+	if (!ret) {
+		amdgpu_bo_unpin(adev->gart.bo);
+		amdgpu_bo_unreserve(adev->gart.bo);
+	}
+	amdgpu_bo_unref(&adev->gart.bo);
+	sg_free_table(sg);
+	kfree(sg);
+	p = virt_to_page(adev->gart.ptr);
+	__free_pages(p, order);
+
+	adev->gart.ptr = NULL;
+}
+
 /**
  * amdgpu_gart_table_vram_alloc - allocate vram for gart page table
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
index 8fea3e04e4110..8283d682f543b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
@@ -51,6 +51,8 @@ struct amdgpu_gart {
 	uint64_t			gart_pte_flags;
 };
 
+int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev);
+void amdgpu_gart_table_ram_free(struct amdgpu_device *adev);
 int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev);
 void amdgpu_gart_table_vram_free(struct amdgpu_device *adev);
 int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 444441c6b7e39..aca8489635b8f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1688,12 +1688,18 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev)
 	adev->gart.gart_pte_flags = AMDGPU_PTE_MTYPE_VG10(MTYPE_UC) |
 				 AMDGPU_PTE_EXECUTABLE;
 
-	r = amdgpu_gart_table_vram_alloc(adev);
-	if (r)
-		return r;
+	if (!adev->gmc.real_vram_size) {
+		dev_info(adev->dev, "Put GART in system memory for APU\n");
+		r = amdgpu_gart_table_ram_alloc(adev);
+		if (r)
+			dev_err(adev->dev, "Failed to allocate GART in system memory\n");
+	} else {
+		r = amdgpu_gart_table_vram_alloc(adev);
+		if (r)
+			return r;
 
-	if (adev->gmc.xgmi.connected_to_cpu) {
-		r = amdgpu_gmc_pdb0_alloc(adev);
+		if (adev->gmc.xgmi.connected_to_cpu)
+			r = amdgpu_gmc_pdb0_alloc(adev);
 	}
 
 	return r;
@@ -1902,7 +1908,12 @@ static int gmc_v9_0_sw_fini(void *handle)
 	amdgpu_gmc_ras_fini(adev);
 	amdgpu_gem_force_release(adev);
 	amdgpu_vm_manager_fini(adev);
-	amdgpu_gart_table_vram_free(adev);
+	if (!adev->gmc.real_vram_size) {
+		dev_info(adev->dev, "Put GART in system memory for APU free\n");
+		amdgpu_gart_table_ram_free(adev);
+	} else {
+		amdgpu_gart_table_vram_free(adev);
+	}
 	amdgpu_bo_free_kernel(&adev->gmc.pdb0_bo, NULL, &adev->gmc.ptr_pdb0);
 	amdgpu_bo_fini(adev);
 
-- 
GitLab


From 44cbc4534bbe7cc1f7dd25976a044c7a84628978 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 1 Dec 2022 17:27:47 +0530
Subject: [PATCH 1170/1791] drm/amdgpu: Make VRAM discovery read optional

When overridden with module param, directly read discovery info
from discovery binary instead of reading from VRAM.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 31 +++++++++----------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 5b6088fc2b2bc..700750433d0f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -301,28 +301,27 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 	if (!adev->mman.discovery_bin)
 		return -ENOMEM;
 
-	amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin);
-
-	if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin) || amdgpu_discovery == 2) {
-		/* ignore the discovery binary from vram if discovery=2 in kernel module parameter */
-		if (amdgpu_discovery == 2)
-			dev_info(adev->dev, "force read ip discovery binary from file");
-		else
-			dev_warn(adev->dev, "get invalid ip discovery binary signature from vram\n");
-
-		/* retry read ip discovery binary from file */
+	/* Read from file if it is the preferred option */
+	if (amdgpu_discovery == 2) {
+		dev_info(adev->dev, "use ip discovery information from file");
 		r = amdgpu_discovery_read_binary_from_file(adev, adev->mman.discovery_bin);
+
 		if (r) {
 			dev_err(adev->dev, "failed to read ip discovery binary from file\n");
 			r = -EINVAL;
 			goto out;
 		}
-		/* check the ip discovery binary signature */
-		if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) {
-			dev_warn(adev->dev, "get invalid ip discovery binary signature from file\n");
-			r = -EINVAL;
-			goto out;
-		}
+
+	} else {
+		amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin);
+	}
+
+	/* check the ip discovery binary signature */
+	if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) {
+		dev_err(adev->dev,
+			"get invalid ip discovery binary signature\n");
+		r = -EINVAL;
+		goto out;
 	}
 
 	bhdr = (struct binary_header *)adev->mman.discovery_bin;
-- 
GitLab


From 73fa255328263e525d7d46d511b088e80e3cd579 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 28 Nov 2022 12:02:14 +0530
Subject: [PATCH 1171/1791] drm/amdgpu: Use discovery to get XCC/SDMA mask

Get information about active XCC and SDMAs from discovery table.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 25 ++++++++++++++++---
 1 file changed, 22 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 700750433d0f9..a1f06945b8ee6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -574,6 +574,14 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev,
 				1 << (le16_to_cpu(harvest_info->list[i].number_instance));
 			(*umc_harvest_count)++;
 			break;
+		case GC_HWID:
+			adev->gfx.xcc_mask &=
+				~(1U << harvest_info->list[i].number_instance);
+			break;
+		case SDMA0_HWID:
+			adev->sdma.sdma_mask &=
+				~(1U << harvest_info->list[i].number_instance);
+			break;
 		default:
 			break;
 		}
@@ -1106,6 +1114,8 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 		return r;
 	}
 
+	adev->gfx.xcc_mask = 0;
+	adev->sdma.sdma_mask = 0;
 	bhdr = (struct binary_header *)adev->mman.discovery_bin;
 	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
@@ -1164,12 +1174,16 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 			    le16_to_cpu(ip->hw_id) == SDMA1_HWID ||
 			    le16_to_cpu(ip->hw_id) == SDMA2_HWID ||
 			    le16_to_cpu(ip->hw_id) == SDMA3_HWID) {
-				if (adev->sdma.num_instances < AMDGPU_MAX_SDMA_INSTANCES)
+				if (adev->sdma.num_instances <
+				    AMDGPU_MAX_SDMA_INSTANCES) {
 					adev->sdma.num_instances++;
-				else
+					adev->sdma.sdma_mask |=
+						(1U << ip->instance_number);
+				} else {
 					dev_err(adev->dev, "Too many SDMA instances: %d vs %d\n",
 						adev->sdma.num_instances + 1,
 						AMDGPU_MAX_SDMA_INSTANCES);
+				}
 			}
 
 			if (le16_to_cpu(ip->hw_id) == UMC_HWID) {
@@ -1177,6 +1191,10 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				adev->umc.node_inst_num++;
 			}
 
+			if (le16_to_cpu(ip->hw_id) == GC_HWID)
+				adev->gfx.xcc_mask |=
+					(1U << ip->instance_number);
+
 			for (k = 0; k < num_base_address; k++) {
 				/*
 				 * convert the endianness of base addresses in place,
@@ -1243,7 +1261,8 @@ static void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev)
 	 * so read harvest bit per IP data structure to set
 	 * harvest configuration.
 	 */
-	if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 2, 0)) {
+	if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 2, 0) &&
+	    adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3)) {
 		if ((adev->pdev->device == 0x731E &&
 			(adev->pdev->revision == 0xC6 ||
 			 adev->pdev->revision == 0xC7)) ||
-- 
GitLab


From 45ed97ad36b744dfc2754c47cfd4423aab6322a2 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 12 Dec 2022 12:29:04 -0500
Subject: [PATCH 1172/1791] drm/amdgpu: increase MAX setting to hold more jpeg
 instances

vcn_v4_0_3 increased jpeg instances,
need increasing MAX resources setting accordlingly.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h | 6 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h     | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h     | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index a29a018ec84ea..3c3ae2b4dbc8a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -325,14 +325,14 @@ typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1
 	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END         = 0x19F,
 	/* IH: 0x1A0 ~ 0x1AF */
 	AMDGPU_DOORBELL_LAYOUT1_IH                      = 0x1A0,
-	/* VCN: 0x1B0 ~ 0x1C2 */
+	/* VCN: 0x1B0 ~ 0x1D4 */
 	AMDGPU_DOORBELL_LAYOUT1_VCN_START               = 0x1B0,
-	AMDGPU_DOORBELL_LAYOUT1_VCN_END                 = 0x1C2,
+	AMDGPU_DOORBELL_LAYOUT1_VCN_END                 = 0x1D4,
 
 	AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP            = AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START,
 	AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP             = AMDGPU_DOORBELL_LAYOUT1_VCN_END,
 
-	AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT          = 0x1C2,
+	AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT          = 0x1D4,
 	AMDGPU_DOORBELL_LAYOUT1_INVALID                 = 0xFFFF
 } AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index 5c200a508fa3e..bb700a2b97c21 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -26,7 +26,7 @@
 
 #include "amdgpu_ras.h"
 
-#define AMDGPU_MAX_JPEG_INSTANCES	2
+#define AMDGPU_MAX_JPEG_INSTANCES	4
 #define AMDGPU_MAX_JPEG_RINGS		8
 
 #define AMDGPU_JPEG_HARVEST_JPEG0 (1 << 0)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 8f4b416a92e6e..bed118d7add0c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -37,7 +37,7 @@ struct amdgpu_job;
 struct amdgpu_vm;
 
 /* max number of rings */
-#define AMDGPU_MAX_RINGS		96
+#define AMDGPU_MAX_RINGS		102
 #define AMDGPU_MAX_HWIP_RINGS		32
 #define AMDGPU_MAX_GFX_RINGS		2
 #define AMDGPU_MAX_SW_GFX_RINGS         2
-- 
GitLab


From 2d7f1d51c1e9812c6a085b6e5bfd99e31b1442cb Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 12 Dec 2022 12:31:23 -0500
Subject: [PATCH 1173/1791] drm/amdgpu: add fwlog support on vcn_v4_0_3

Add fwlog support on vcn_v4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 8db50a50b868b..40de2852c5caa 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -118,6 +118,9 @@ static int vcn_v4_0_3_sw_init(void *handle)
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 		fw_shared->present_flag_0 = 0;
 		fw_shared->sq.is_enabled = cpu_to_le32(true);
+
+		if (amdgpu_vcnfw_log)
+			amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]);
 	}
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
-- 
GitLab


From da044aaeb31ac11e733a667763487508433f3ede Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 17 Dec 2022 10:51:04 -0500
Subject: [PATCH 1174/1791] drm/amdgpu: add unified queue support on vcn_v4_0_3

Add unified queue support on vcn_v4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c |   9 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 173 +++++++++++++-----------
 2 files changed, 102 insertions(+), 80 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 72eb12aa2e8d4..2e28b19e741e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -1004,11 +1004,14 @@ int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 int amdgpu_vcn_unified_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 {
+	struct amdgpu_device *adev = ring->adev;
 	long r;
 
-	r = amdgpu_vcn_enc_ring_test_ib(ring, timeout);
-	if (r)
-		goto error;
+	if (adev->ip_versions[UVD_HWIP][0] != IP_VERSION(4, 0, 3)) {
+		r = amdgpu_vcn_enc_ring_test_ib(ring, timeout);
+		if (r)
+			goto error;
+	}
 
 	r =  amdgpu_vcn_dec_sw_ring_test_ib(ring, timeout);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 40de2852c5caa..18320eebdb4c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -31,7 +31,6 @@
 #include "soc15d.h"
 #include "soc15_hw_ip.h"
 #include "vcn_v2_0.h"
-#include "vcn_sw_ring.h"
 
 #include "vcn/vcn_4_0_3_offset.h"
 #include "vcn/vcn_4_0_3_sh_mask.h"
@@ -45,12 +44,13 @@
 #define VCN_VID_SOC_ADDRESS_2_0		0x1fb00
 #define VCN1_VID_SOC_ADDRESS_3_0	0x48300
 
-static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev);
+static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev);
 static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev);
 static int vcn_v4_0_3_set_powergating_state(void *handle,
 		enum amd_powergating_state state);
 static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev,
 		int inst_idx, struct dpg_pause_state *new_state);
+static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring);
 
 /**
  * vcn_v4_0_3_early_init - set function pointers
@@ -63,7 +63,10 @@ static int vcn_v4_0_3_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	vcn_v4_0_3_set_dec_ring_funcs(adev);
+	/* re-use enc ring as unified ring */
+	adev->vcn.num_enc_rings = 1;
+
+	vcn_v4_0_3_set_unified_ring_funcs(adev);
 	vcn_v4_0_3_set_irq_funcs(adev);
 
 	return 0;
@@ -94,7 +97,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 
 	/* VCN DEC TRAP */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-		VCN_4_0__SRCID__UVD_TRAP, &adev->vcn.inst->irq);
+		VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE, &adev->vcn.inst->irq);
 	if (r)
 		return r;
 
@@ -104,11 +107,11 @@ static int vcn_v4_0_3_sw_init(void *handle)
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
 
-		ring = &adev->vcn.inst[i].ring_dec;
+		ring = &adev->vcn.inst[i].ring_enc[0];
 		ring->use_doorbell = true;
 		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i;
 		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
-		sprintf(ring->name, "vcn_dec_%d", i);
+		sprintf(ring->name, "vcn_unified_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
 				     AMDGPU_RING_PRIO_DEFAULT,
 				     &adev->vcn.inst[i].sched_score);
@@ -116,7 +119,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 			return r;
 
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
-		fw_shared->present_flag_0 = 0;
+		fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE);
 		fw_shared->sq.is_enabled = cpu_to_le32(true);
 
 		if (amdgpu_vcnfw_log)
@@ -179,17 +182,17 @@ static int vcn_v4_0_3_hw_init(void *handle)
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
-		ring = &adev->vcn.inst[i].ring_dec;
+		ring = &adev->vcn.inst[i].ring_enc[0];
 
-		if (ring->use_doorbell)
+		if (ring->use_doorbell) {
 			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
 					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i,
 					adev->vcn.inst[i].aid_id);
 
-		if (ring->use_doorbell)
-			WREG32_SOC15(VCN, ring->me, regVCN_RB4_DB_CTRL,
-				ring->doorbell_index << VCN_RB4_DB_CTRL__OFFSET__SHIFT |
-				VCN_RB4_DB_CTRL__EN_MASK);
+			WREG32_SOC15(VCN, ring->me, regVCN_RB1_DB_CTRL,
+				ring->doorbell_index << VCN_RB1_DB_CTRL__OFFSET__SHIFT |
+				VCN_RB1_DB_CTRL__EN_MASK);
+		}
 
 		r = amdgpu_ring_test_helper(ring);
 		if (r)
@@ -731,31 +734,31 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 			(uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr -
 				(uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr));
 
-	ring = &adev->vcn.inst[inst_idx].ring_dec;
+	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
 	/* program the RB_BASE for ring buffer */
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_LO4,
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_LO,
 		lower_32_bits(ring->gpu_addr));
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_HI4,
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_HI,
 		upper_32_bits(ring->gpu_addr));
 
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_SIZE, ring->ring_size / sizeof(uint32_t));
 
 	/* resetting ring, fw should not check RB ring */
 	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
-	tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
+	tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK);
 	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
+	fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET;
 
 	/* Initialize the ring buffer's read and write pointers */
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_RPTR4, 0);
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4, 0);
-	ring->wptr = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4);
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_RPTR, 0);
+	WREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR, 0);
+	ring->wptr = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR);
 
 	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
-	tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
+	tmp |= VCN_RB_ENABLE__RB_EN_MASK;
 	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
-
-	WREG32_SOC15(VCN, inst_idx, regUVD_SCRATCH2, 0);
+	fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF);
 
 	/*resetting done, fw can check RB ring */
 	fw_shared->sq.queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
@@ -902,31 +905,31 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_STATUS), 0,
 			~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
 
-		ring = &adev->vcn.inst[i].ring_dec;
+		ring = &adev->vcn.inst[i].ring_enc[0];
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 
 		/* program the RB_BASE for ring buffer */
-		WREG32_SOC15(VCN, i, regUVD_RB_BASE_LO4,
+		WREG32_SOC15(VCN, i, regUVD_RB_BASE_LO,
 			lower_32_bits(ring->gpu_addr));
-		WREG32_SOC15(VCN, i, regUVD_RB_BASE_HI4,
+		WREG32_SOC15(VCN, i, regUVD_RB_BASE_HI,
 			upper_32_bits(ring->gpu_addr));
 
-		WREG32_SOC15(VCN, i, regUVD_RB_SIZE4, ring->ring_size / sizeof(uint32_t));
+		WREG32_SOC15(VCN, i, regUVD_RB_SIZE, ring->ring_size / sizeof(uint32_t));
 
 		/* resetting ring, fw should not check RB ring */
 		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
-		tmp &= ~(VCN_RB_ENABLE__RB4_EN_MASK);
+		tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK);
 		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
 
 		/* Initialize the ring buffer's read and write pointers */
-		WREG32_SOC15(VCN, i, regUVD_RB_RPTR4, 0);
-		WREG32_SOC15(VCN, i, regUVD_RB_WPTR4, 0);
+		WREG32_SOC15(VCN, i, regUVD_RB_RPTR, 0);
+		WREG32_SOC15(VCN, i, regUVD_RB_WPTR, 0);
 
 		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
-		tmp |= VCN_RB_ENABLE__RB4_EN_MASK;
+		tmp |= VCN_RB_ENABLE__RB_EN_MASK;
 		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
 
-		ring->wptr = RREG32_SOC15(VCN, i, regUVD_RB_WPTR4);
+		ring->wptr = RREG32_SOC15(VCN, i, regUVD_RB_WPTR);
 		fw_shared->sq.queue_mode &=
 			cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF));
 
@@ -951,8 +954,8 @@ static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 	/* wait for read ptr to be equal to write ptr */
-	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR4);
-	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_RB_RPTR4, tmp, 0xFFFFFFFF);
+	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR);
+	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_RB_RPTR, tmp, 0xFFFFFFFF);
 
 	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_POWER_STATUS, 1,
 		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
@@ -972,15 +975,20 @@ static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
  */
 static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 {
+	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	uint32_t tmp;
 	int i, r = 0;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
+
+		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
+		fw_shared->sq.queue_mode |= FW_QUEUE_DPG_HOLD_OFF;
+
 		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
-			r = vcn_v4_0_3_stop_dpg_mode(adev, i);
-			goto Done;
+			vcn_v4_0_3_stop_dpg_mode(adev, i);
+			continue;
 		}
 
 		/* wait for vcn idle */
@@ -1000,7 +1008,7 @@ static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 		tmp = RREG32_SOC15(VCN, i, regUVD_LMI_CTRL2);
 		tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK;
 		WREG32_SOC15(VCN, i, regUVD_LMI_CTRL2, tmp);
-		tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK|
+		tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK |
 			UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK;
 		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_LMI_STATUS, tmp, tmp);
 		if (r)
@@ -1059,101 +1067,112 @@ static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev, int inst_idx,
 }
 
 /**
- * vcn_v4_0_3_dec_ring_get_rptr - get read pointer
+ * vcn_v4_0_3_unified_ring_get_rptr - get unified read pointer
  *
  * @ring: amdgpu_ring pointer
  *
- * Returns the current hardware read pointer
+ * Returns the current hardware unified read pointer
  */
-static uint64_t vcn_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring)
+static uint64_t vcn_v4_0_3_unified_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 
-	return RREG32_SOC15(VCN, ring->me, regUVD_RB_RPTR4);
+	if (ring != &adev->vcn.inst[ring->me].ring_enc[0])
+		DRM_ERROR("wrong ring id is identified in %s", __func__);
+
+	return RREG32_SOC15(VCN, ring->me, regUVD_RB_RPTR);
 }
 
 /**
- * vcn_v4_0_3_dec_ring_get_wptr - get write pointer
+ * vcn_v4_0_3_unified_ring_get_wptr - get unified write pointer
  *
  * @ring: amdgpu_ring pointer
  *
- * Returns the current hardware write pointer
+ * Returns the current hardware unified write pointer
  */
-static uint64_t vcn_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
+static uint64_t vcn_v4_0_3_unified_ring_get_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 
+	if (ring != &adev->vcn.inst[ring->me].ring_enc[0])
+		DRM_ERROR("wrong ring id is identified in %s", __func__);
+
 	if (ring->use_doorbell)
-		return adev->wb.wb[ring->wptr_offs];
+		return *ring->wptr_cpu_addr;
 	else
-		return RREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR4);
+		return RREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR);
 }
 
 /**
- * vcn_v4_0_3_dec_ring_set_wptr - set write pointer
+ * vcn_v4_0_3_unified_ring_set_wptr - set enc write pointer
  *
  * @ring: amdgpu_ring pointer
  *
- * Commits the write pointer to the hardware
+ * Commits the enc write pointer to the hardware
  */
-static void vcn_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
+static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 
+	if (ring != &adev->vcn.inst[ring->me].ring_enc[0])
+		DRM_ERROR("wrong ring id is identified in %s", __func__);
+
 	if (ring->use_doorbell) {
-		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
+		*ring->wptr_cpu_addr = lower_32_bits(ring->wptr);
 		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
 	} else {
-		WREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR4, lower_32_bits(ring->wptr));
+		WREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR, lower_32_bits(ring->wptr));
 	}
 }
 
-static const struct amdgpu_ring_funcs vcn_v4_0_3_dec_sw_ring_vm_funcs = {
-	.type = AMDGPU_RING_TYPE_VCN_DEC,
+static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = {
+	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
-	.nop = VCN_DEC_SW_CMD_NO_OP,
-	.get_rptr = vcn_v4_0_3_dec_ring_get_rptr,
-	.get_wptr = vcn_v4_0_3_dec_ring_get_wptr,
-	.set_wptr = vcn_v4_0_3_dec_ring_set_wptr,
+	.nop = VCN_ENC_CMD_NO_OP,
+	.get_rptr = vcn_v4_0_3_unified_ring_get_rptr,
+	.get_wptr = vcn_v4_0_3_unified_ring_get_wptr,
+	.set_wptr = vcn_v4_0_3_unified_ring_set_wptr,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 4 +
-		VCN_SW_RING_EMIT_FRAME_SIZE,
-	.emit_ib_size = 5, /* vcn_dec_sw_ring_emit_ib */
-	.emit_ib = vcn_dec_sw_ring_emit_ib,
-	.emit_fence = vcn_dec_sw_ring_emit_fence,
-	.emit_vm_flush = vcn_dec_sw_ring_emit_vm_flush,
-	.test_ring = amdgpu_vcn_dec_sw_ring_test_ring,
-	.test_ib = amdgpu_vcn_dec_sw_ring_test_ib,
+		4 + /* vcn_v2_0_enc_ring_emit_vm_flush */
+		5 + 5 + /* vcn_v2_0_enc_ring_emit_fence x2 vm fence */
+		1, /* vcn_v2_0_enc_ring_insert_end */
+	.emit_ib_size = 5, /* vcn_v2_0_enc_ring_emit_ib */
+	.emit_ib = vcn_v2_0_enc_ring_emit_ib,
+	.emit_fence = vcn_v2_0_enc_ring_emit_fence,
+	.emit_vm_flush = vcn_v2_0_enc_ring_emit_vm_flush,
+	.test_ring = amdgpu_vcn_enc_ring_test_ring,
+	.test_ib = amdgpu_vcn_unified_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
-	.insert_end = vcn_dec_sw_ring_insert_end,
+	.insert_end = vcn_v2_0_enc_ring_insert_end,
 	.pad_ib = amdgpu_ring_generic_pad_ib,
 	.begin_use = amdgpu_vcn_ring_begin_use,
 	.end_use = amdgpu_vcn_ring_end_use,
-	.emit_wreg = vcn_dec_sw_ring_emit_wreg,
-	.emit_reg_wait = vcn_dec_sw_ring_emit_reg_wait,
+	.emit_wreg = vcn_v2_0_enc_ring_emit_wreg,
+	.emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
 };
 
 /**
- * vcn_v4_0_3_set_dec_ring_funcs - set dec ring functions
+ * vcn_v4_0_3_set_unified_ring_funcs - set unified ring functions
  *
  * @adev: amdgpu_device pointer
  *
- * Set decode ring functions
+ * Set unified ring functions
  */
-static void vcn_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
+static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev)
 {
 	int i;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
-		adev->vcn.inst[i].ring_dec.funcs = &vcn_v4_0_3_dec_sw_ring_vm_funcs;
-		adev->vcn.inst[i].ring_dec.me = i;
+		adev->vcn.inst[i].ring_enc[0].funcs = &vcn_v4_0_3_unified_ring_vm_funcs;
+		adev->vcn.inst[i].ring_enc[0].me = i;
 		adev->vcn.inst[i].aid_id = i / adev->vcn.num_inst_per_aid;
 	}
-	DRM_DEV_INFO(adev->dev, "VCN decode(Software Ring) is enabled in VM mode\n");
+	DRM_DEV_INFO(adev->dev, "VCN decode is enabled in VM mode\n");
 }
 
 /**
@@ -1276,7 +1295,7 @@ static int vcn_v4_0_3_set_interrupt_state(struct amdgpu_device *adev,
 }
 
 /**
- * vcn_v4_0_process_interrupt - process VCN block interrupt
+ * vcn_v4_0_3_process_interrupt - process VCN block interrupt
  *
  * @adev: amdgpu_device pointer
  * @source: interrupt sources
@@ -1295,8 +1314,8 @@ static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 	DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n");
 
 	switch (entry->src_id) {
-	case VCN_4_0__SRCID__UVD_TRAP:
-		amdgpu_fence_process(&adev->vcn.inst[i].ring_dec);
+	case VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE:
+		amdgpu_fence_process(&adev->vcn.inst[i].ring_enc[0]);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
-- 
GitLab


From e40b4b9a474887653263c138b0172b0e2fde57cb Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 12 Dec 2022 13:14:05 -0500
Subject: [PATCH 1175/1791] drm/amdgpu: enable indirect_sram mode on vcn_v4_0_3

Enable indirect_sram mode on vcn_v4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 2e28b19e741e8..daa69ae766d3d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -113,18 +113,9 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
 		atomic_set(&adev->vcn.inst[i].dpg_enc_submission_cnt, 0);
 
-	switch (adev->ip_versions[UVD_HWIP][0]) {
-	case IP_VERSION(4, 0, 3):
-		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
-		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
-			adev->vcn.indirect_sram = false;
-		break;
-	default:
-		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
-		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
-			adev->vcn.indirect_sram = true;
-		break;
-	}
+	if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
+	    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
+		adev->vcn.indirect_sram = true;
 
 	/*
 	 * Some Steam Deck's BIOS versions are incompatible with the
-- 
GitLab


From ed1f42f03c46767df7f55d6a75c39051a55cc656 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Sat, 17 Dec 2022 19:44:15 -0500
Subject: [PATCH 1176/1791] drm/amdgpu: enable vcn/jpeg on vcn_v4_0_3

Enable vcn/jpeg on vcn_v4_0_3.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index a1f06945b8ee6..d8f55a99646e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -76,6 +76,8 @@
 #include "jpeg_v3_0.h"
 #include "vcn_v4_0.h"
 #include "jpeg_v4_0.h"
+#include "vcn_v4_0_3.h"
+#include "jpeg_v4_0_3.h"
 #include "amdgpu_vkms.h"
 #include "mes_v10_1.h"
 #include "mes_v11_0.h"
@@ -1947,7 +1949,11 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev)
 		case IP_VERSION(4, 0, 4):
 			amdgpu_device_ip_block_add(adev, &vcn_v4_0_ip_block);
 			amdgpu_device_ip_block_add(adev, &jpeg_v4_0_ip_block);
-			return 0;
+			break;
+		case IP_VERSION(4, 0, 3):
+			amdgpu_device_ip_block_add(adev, &vcn_v4_0_3_ip_block);
+			amdgpu_device_ip_block_add(adev, &jpeg_v4_0_3_ip_block);
+			break;
 		default:
 			dev_err(adev->dev,
 				"Failed to add vcn/jpeg ip block(UVD_HWIP:0x%x)\n",
-- 
GitLab


From 30b52995c294b3c9a002ff2f90f773cbf8850714 Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Tue, 3 Jan 2023 13:14:58 +0800
Subject: [PATCH 1177/1791] drm/amdgpu: Remove SMU powergate message call for
 SDMA

SDMA v4.4.2 doesn't need explicit power gating control through PMFW

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 184eb79027223..7c91cbd91ac80 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1398,10 +1398,6 @@ static int sdma_v4_4_2_hw_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	uint32_t inst_mask;
 
-	/* TODO: Check if this is needed */
-	if (adev->flags & AMD_IS_APU)
-		amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_SDMA, false);
-
 	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
 	if (!amdgpu_sriov_vf(adev))
 		sdma_v4_4_2_inst_init_golden_registers(adev, inst_mask);
-- 
GitLab


From d839a158b2480814bc438f9f46f440a7b9f63cb6 Mon Sep 17 00:00:00 2001
From: Graham Sider <Graham.Sider@amd.com>
Date: Thu, 5 Jan 2023 10:58:07 -0500
Subject: [PATCH 1178/1791] drm/amdgpu: Correct dGPU MTYPE settings for gfx943

Revert temporary dGPU VRAM MTYPE setting and align with expected
coherency protocol.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 15 +++++----------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c  |  8 ++------
 2 files changed, 7 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index aca8489635b8f..b6c500be6f703 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1238,17 +1238,12 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 			mtype = MTYPE_RW;
 		} else {
 			/* dGPU */
-			/*
-			if ((mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) &&
-			    bo_adev == adev)
-				mapping_flags |= AMDGPU_VM_MTYPE_RW;
+			if (is_vram && bo_adev == adev)
+				mtype = MTYPE_RW;
+			else if (is_vram)
+				mtype = MTYPE_NC;
 			else
-			*/
-			/* Temporarily comment out above lines and use MTYPE_NC
-			 * on both VRAM and system memory access until
-			 * MTYPE_RW can properly work on VRAM access
-			 */
-			mtype = MTYPE_NC;
+				mtype = MTYPE_UC;
 		}
 
 		break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 6daba0582bf37..2b79849ddd30f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1198,13 +1198,9 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		} else if (domain == SVM_RANGE_VRAM_DOMAIN) {
-			/* local HBM region close to partition with a workaround
-			 * for Endpoint systems.
-			 */
+			/* local HBM region close to partition */
 			if (bo_node == node)
-				mapping_flags |=
-					(node->adev->flags & AMD_IS_APU) ?
-					AMDGPU_VM_MTYPE_RW : AMDGPU_VM_MTYPE_NC;
+				mapping_flags |= AMDGPU_VM_MTYPE_RW;
 			/* local HBM region far from partition or remote XGMI GPU */
 			else if (svm_nodes_in_same_hive(bo_node, node))
 				mapping_flags |= AMDGPU_VM_MTYPE_NC;
-- 
GitLab


From 52c293ab0653b3e57d2202a5002f59593ed12d1b Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 10 Jan 2023 09:52:53 +0530
Subject: [PATCH 1179/1791] drm/amdgpu: Populate VCN/JPEG harvest information

Certain instances of VCN/JPEG IPs may not be usable. Fetch the information
from harvest table.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index d8f55a99646e5..6701f17a4db62 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -563,10 +563,10 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev,
 		switch (le16_to_cpu(harvest_info->list[i].hw_id)) {
 		case VCN_HWID:
 			(*vcn_harvest_count)++;
-			if (harvest_info->list[i].number_instance == 0)
-				adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN0;
-			else
-				adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1;
+			adev->vcn.harvest_config |=
+				(1 << harvest_info->list[i].number_instance);
+			adev->jpeg.harvest_config |=
+				(1 << harvest_info->list[i].number_instance);
 			break;
 		case DMU_HWID:
 			adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK;
-- 
GitLab


From f471de2586e8ef388eea2cafa911695d94ec1d88 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 19 Dec 2022 20:11:11 -0500
Subject: [PATCH 1180/1791] drm/amdgpu: vcn_v4_0_3 load vcn fw once for all
 AIDs

Signed-off-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 3 +++
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index daa69ae766d3d..57dabfe1a1bee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -1046,6 +1046,9 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev)
 			adev->firmware.ucode[idx].fw = adev->vcn.fw;
 			adev->firmware.fw_size +=
 				ALIGN(le32_to_cpu(hdr->ucode_size_bytes), PAGE_SIZE);
+
+			if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(4, 0, 3))
+				break;
 		}
 		dev_info(adev->dev, "Will use PSP to load VCN firmware\n");
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 18320eebdb4c8..746df23b2eaa6 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -730,7 +730,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
 	if (indirect)
-		psp_update_vcn_sram(adev, inst_idx, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr,
+		psp_update_vcn_sram(adev, 0, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr,
 			(uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr -
 				(uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr));
 
-- 
GitLab


From 1bd99ca2695a76f15d7f7862d6ef878588e854dc Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 10 Jan 2023 09:01:33 -0500
Subject: [PATCH 1181/1791] drm/amdgpu: increase AMDGPU_MAX_HWIP_RINGS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[WA] Increase AMDGPU_MAX_HWIP_RINGS to 64 to support more compute
ring resource. Later need redesign with queue/prirority/scheduler
factors to reduce AMDGPU_MAX_HWIP_RINGS.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index bed118d7add0c..add7cc2831b03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -38,7 +38,7 @@ struct amdgpu_vm;
 
 /* max number of rings */
 #define AMDGPU_MAX_RINGS		102
-#define AMDGPU_MAX_HWIP_RINGS		32
+#define AMDGPU_MAX_HWIP_RINGS		64
 #define AMDGPU_MAX_GFX_RINGS		2
 #define AMDGPU_MAX_SW_GFX_RINGS         2
 #define AMDGPU_MAX_COMPUTE_RINGS	8
-- 
GitLab


From 13a94f3f130ee4db6b4d2a0843104807a7299aa4 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 10 Jan 2023 09:05:35 -0500
Subject: [PATCH 1182/1791] drm/amdgpu: add num_xcps return

Add num_xcps return.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 0d7bc212def13..6591d39c6518c 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -318,6 +318,8 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
 
 	ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags);
+
+	*num_xcps = num_xcc / num_xcc_per_xcp;
 unlock:
 	if (flags & AMDGPU_XCP_OPS_KFD)
 		amdgpu_amdkfd_unlock_kfd(adev);
-- 
GitLab


From b2ef2fdffed2a7fd5bf3f178a6a0427487dba5dd Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Thu, 5 Jan 2023 11:39:34 -0500
Subject: [PATCH 1183/1791] drm/amdkfd: Report XGMI IOLINKs for GFXIP9.4.3

GFXIP 9.4.3 could be in APU or carveout mode but we cannot use the
xgmi.connected_to_cpu flag to identify the iolinks type. Use appropriate
APU or Carveout mode based condition to report xgmi connection in kfd
topology.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c     | 5 ++++-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 5 ++---
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index dc93a67257e10..16475921587bf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -2005,7 +2005,10 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
 	/* Fill in IOLINK subtype.
 	 * TODO: Fill-in other fields of iolink subtype
 	 */
-	if (kdev->adev->gmc.xgmi.connected_to_cpu) {
+	if (kdev->adev->gmc.xgmi.connected_to_cpu ||
+	    (KFD_GC_VERSION(kdev) == IP_VERSION(9, 4, 3) &&
+	     kdev->adev->smuio.funcs->get_pkg_type(kdev->adev) ==
+	     AMDGPU_PKG_TYPE_APU)) {
 		bool ext_cpu = KFD_GC_VERSION(kdev) != IP_VERSION(9, 4, 3);
 		int mem_bw = 819200, weight = ext_cpu ? KFD_CRAT_XGMI_WEIGHT :
 							KFD_CRAT_INTRA_SOCKET_WEIGHT;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index d3e70341dfadf..5373a79ac6a10 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1327,9 +1327,8 @@ static void kfd_fill_iolink_non_crat_info(struct kfd_topology_device *dev)
 			continue;
 
 		/* Include the CPU peer in GPU hive if connected over xGMI. */
-		if (!peer_dev->gpu && !peer_dev->node_props.hive_id &&
-				dev->node_props.hive_id &&
-				dev->gpu->adev->gmc.xgmi.connected_to_cpu)
+		if (!peer_dev->gpu &&
+		    link->iolink_type == CRAT_IOLINK_TYPE_XGMI)
 			peer_dev->node_props.hive_id = dev->node_props.hive_id;
 
 		list_for_each_entry(inbound_link, &peer_dev->io_link_props,
-- 
GitLab


From 7389c75114c53b061d686f19dff5833adaf96cb8 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 17 Jan 2023 16:54:49 +0530
Subject: [PATCH 1184/1791] drm/amdgpu: Keep SDMAv4.4.2 active during reset

During ASIC wide reset, SDMA shouldn't be clockgated and be ready to
accept freeze requests from PMFW. For that, don't stop SDMA engine
during reset and keep the clocks active.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 7c91cbd91ac80..729e26a4a2e74 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -566,6 +566,11 @@ static void sdma_v4_4_2_inst_enable(struct amdgpu_device *adev, bool enable,
 		sdma_v4_4_2_inst_rlc_stop(adev, inst_mask);
 		if (adev->sdma.has_page_queue)
 			sdma_v4_4_2_inst_page_stop(adev, inst_mask);
+
+		/* SDMA FW needs to respond to FREEZE requests during reset.
+		 * Keep it running during reset */
+		if (!amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev))
+			return;
 	}
 
 	for_each_inst(i, inst_mask) {
@@ -1435,6 +1440,9 @@ static int sdma_v4_4_2_suspend(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (amdgpu_in_reset(adev))
+		sdma_v4_4_2_set_clockgating_state(adev, AMD_CG_STATE_UNGATE);
+
 	return sdma_v4_4_2_hw_fini(adev);
 }
 
-- 
GitLab


From 233bb3733bd43966696f4a5e95129476e86bf4e3 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 19 Jan 2023 14:47:22 +0530
Subject: [PATCH 1185/1791] drm/amdgpu: Use unique doorbell range per xcc

Program different ranges in each XCC with MEC_DOORBELL_RANGE_LOWER/HIGHER.
Keeping the same range causes CPF in other XCCs also to be busy when an IB
packet is submitted to KCQ. Only the XCC which processes the packet
comes back to idle afterwards and this causes other CPs not be idle.
This in turn affects clockgating behavior as RLC doesn't get idle
interrupt.

LOWER/HIGHER covers only KIQ/KCQs which are per XCC queues. Assigning
different ranges doesn't seem to have any side effect as user queue ranges
are outside of this range. User queue tests - PM4 through KFD and AQL
through rocr - have the same results after this change.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h  | 35 +++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c       |  5 ++-
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       | 31 +++++++++++-----
 4 files changed, 47 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
index 3c3ae2b4dbc8a..f637574644c0e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h
@@ -86,6 +86,8 @@ struct amdgpu_doorbell_index {
 	uint32_t max_assignment;
 	/* Per engine SDMA doorbell size in dword */
 	uint32_t sdma_doorbell_range;
+	/* Per xcc doorbell size for KIQ/KCQ */
+	uint32_t xcc_doorbell_range;
 };
 
 typedef enum _AMDGPU_DOORBELL_ASSIGNMENT
@@ -309,28 +311,31 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT
 	AMDGPU_DOORBELL64_INVALID                 = 0xFFFF
 } AMDGPU_DOORBELL64_ASSIGNMENT;
 
-typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1
-{
-	/* KIQ: 0~7 for maximum 8 XCD */
-	AMDGPU_DOORBELL_LAYOUT1_KIQ_START               = 0x000,
-	AMDGPU_DOORBELL_LAYOUT1_HIQ                     = 0x008,
-	AMDGPU_DOORBELL_LAYOUT1_DIQ                     = 0x009,
-	/* Compute: 0x0A ~ 0x49 */
-	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START          = 0x00A,
-	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_END            = 0x049,
-	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START         = 0x04A,
-	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END           = 0x0C9,
+typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1 {
+	/* XCC0: 0x00 ~20, XCC1: 20 ~ 2F ... */
+
+	/* KIQ/HIQ/DIQ */
+	AMDGPU_DOORBELL_LAYOUT1_KIQ_START		= 0x000,
+	AMDGPU_DOORBELL_LAYOUT1_HIQ			= 0x001,
+	AMDGPU_DOORBELL_LAYOUT1_DIQ			= 0x002,
+	/* Compute: 0x08 ~ 0x20  */
+	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START		= 0x008,
+	AMDGPU_DOORBELL_LAYOUT1_MEC_RING_END		= 0x00F,
+	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START		= 0x010,
+	AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END		= 0x01F,
+	AMDGPU_DOORBELL_LAYOUT1_XCC_RANGE		= 0x020,
+
 	/* SDMA: 0x100 ~ 0x19F */
-	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START       = 0x100,
-	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END         = 0x19F,
+	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START	= 0x100,
+	AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END		= 0x19F,
 	/* IH: 0x1A0 ~ 0x1AF */
 	AMDGPU_DOORBELL_LAYOUT1_IH                      = 0x1A0,
 	/* VCN: 0x1B0 ~ 0x1D4 */
 	AMDGPU_DOORBELL_LAYOUT1_VCN_START               = 0x1B0,
 	AMDGPU_DOORBELL_LAYOUT1_VCN_END                 = 0x1D4,
 
-	AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP            = AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START,
-	AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP             = AMDGPU_DOORBELL_LAYOUT1_VCN_END,
+	AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP		= AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START,
+	AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP		= AMDGPU_DOORBELL_LAYOUT1_VCN_END,
 
 	AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT          = 0x1D4,
 	AMDGPU_DOORBELL_LAYOUT1_INVALID                 = 0xFFFF
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index f895a4b8ca0de..70c6099353b82 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -316,7 +316,10 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev,
 	ring->use_doorbell = true;
 	ring->xcc_id = xcc_id;
 	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
-	ring->doorbell_index = (adev->doorbell_index.kiq + xcc_id) << 1;
+	ring->doorbell_index =
+		(adev->doorbell_index.kiq +
+		 xcc_id * adev->doorbell_index.xcc_doorbell_range)
+		<< 1;
 
 	r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id);
 	if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 6591d39c6518c..55a6ebb940baa 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -44,6 +44,7 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
 
 	adev->doorbell_index.userqueue_start = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START;
 	adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END;
+	adev->doorbell_index.xcc_doorbell_range = AMDGPU_DOORBELL_LAYOUT1_XCC_RANGE;
 
 	adev->doorbell_index.sdma_doorbell_range = 20;
 	for (i = 0; i < adev->sdma.num_instances; i++)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 55d99c4ea48c6..557a2458ef5e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -729,8 +729,10 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	unsigned irq_type;
 	struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id];
 	unsigned int hw_prio;
+	uint32_t xcc_doorbell_start;
 
-	ring = &adev->gfx.compute_ring[ring_id];
+	ring = &adev->gfx.compute_ring[xcc_id * adev->gfx.num_compute_rings +
+				       ring_id];
 
 	/* mec0 is me1 */
 	ring->xcc_id = xcc_id;
@@ -740,9 +742,12 @@ static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 
 	ring->ring_obj = NULL;
 	ring->use_doorbell = true;
-	ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1;
-	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr
-				+ (ring_id * GFX9_MEC_HPD_SIZE);
+	xcc_doorbell_start = adev->doorbell_index.mec_ring0 +
+			     xcc_id * adev->doorbell_index.xcc_doorbell_range;
+	ring->doorbell_index = (xcc_doorbell_start + ring_id) << 1;
+	ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr +
+			     (ring_id + xcc_id * adev->gfx.num_compute_rings) *
+				     GFX9_MEC_HPD_SIZE;
 	ring->vm_hub = AMDGPU_GFXHUB(xcc_id);
 	sprintf(ring->name, "comp_%d.%d.%d.%d",
 			ring->xcc_id, ring->me, ring->pipe, ring->queue);
@@ -801,8 +806,8 @@ static int gfx_v9_4_3_sw_init(void *handle)
 	}
 
 	/* set up the compute queues - allocate horizontally across pipes */
-	ring_id = 0;
 	for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) {
+		ring_id = 0;
 		for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
 			for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
 				for (k = 0; k < adev->gfx.mec.num_pipe_per_mec;
@@ -1654,10 +1659,18 @@ static int gfx_v9_4_3_xcc_kiq_init_register(struct amdgpu_ring *ring,
 
 	/* enable the doorbell if requested */
 	if (ring->use_doorbell) {
-		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_LOWER,
-					(adev->doorbell_index.kiq * 2) << 2);
-		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEC_DOORBELL_RANGE_UPPER,
-				(adev->doorbell_index.userqueue_end * 2) << 2);
+		WREG32_SOC15(
+			GC, GET_INST(GC, xcc_id),
+			regCP_MEC_DOORBELL_RANGE_LOWER,
+			((adev->doorbell_index.kiq +
+			  xcc_id * adev->doorbell_index.xcc_doorbell_range) *
+			 2) << 2);
+		WREG32_SOC15(
+			GC, GET_INST(GC, xcc_id),
+			regCP_MEC_DOORBELL_RANGE_UPPER,
+			((adev->doorbell_index.userqueue_end +
+			  xcc_id * adev->doorbell_index.xcc_doorbell_range) *
+			 2) << 2);
 	}
 
 	WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL,
-- 
GitLab


From b7c7011e67b09efc486b1de38f6bfbed75139989 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 19 Jan 2023 15:00:45 +0530
Subject: [PATCH 1186/1791] drm/amdgpu: Enable CGCG/LS for GC 9.4.3

Enable coarse grain clockgating/light sleep for GC v9.4.3. Remove
programming that is not meant for GC 9.4.3.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 14 +++++---------
 drivers/gpu/drm/amd/amdgpu/soc15.c      |  3 ++-
 2 files changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 557a2458ef5e1..52041d4cb14af 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2250,11 +2250,9 @@ gfx_v9_4_3_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev,
 
 		data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
+			  RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK);
 
-		/* only for Vega10 & Raven1 */
-		data |= RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK;
-
 		if (def != data)
 			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data);
 
@@ -2311,6 +2309,7 @@ gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
 	uint32_t def, data;
 
 	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
+
 		def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE);
 		/* unset CGCG override */
 		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK;
@@ -2325,12 +2324,9 @@ gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
 		/* enable cgcg FSM(0x0000363F) */
 		def = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL);
 
-		if (adev->asic_type == CHIP_ARCTURUS)
-			data = (0x2000 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
-				RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK;
-		else
-			data = (0x36 << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
-				RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK;
+		data = (0x36
+			<< RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) |
+		       RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK;
 		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS)
 			data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) |
 				RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK;
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 06a18b2f6e041..4138d96342662 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1121,7 +1121,8 @@ static int soc15_common_early_init(void *handle)
 	case IP_VERSION(9, 4, 3):
 		adev->asic_funcs = &aqua_vanjaram_asic_funcs;
 		adev->cg_flags =
-			AMD_CG_SUPPORT_VCN_MGCG |
+			AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGCG |
+			AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_VCN_MGCG |
 			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
-- 
GitLab


From 5ca1ceebab140b36928dabc9a5b36b9a3010e844 Mon Sep 17 00:00:00 2001
From: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Date: Sat, 21 Jan 2023 15:47:11 -0500
Subject: [PATCH 1187/1791] drm/amd: fix compiler error to support older
 compilers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

‘for’ loop initial declarations are only allowed in C99 or C11 mode

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 55a6ebb940baa..fdc728f678d71 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -360,7 +360,7 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
 
 int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 {
-	u32 inst_mask = adev->sdma.sdma_mask;
+	u32 mask, inst_mask = adev->sdma.sdma_mask;
 	int ret, i, num_inst;
 
 	/* generally 1 AID supports 4 instances */
@@ -370,7 +370,7 @@ int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 	adev->aid_mask = i = 1;
 	inst_mask >>= adev->sdma.num_inst_per_aid;
 
-	for (const u32 mask = (1 << adev->sdma.num_inst_per_aid) - 1; inst_mask;
+	for (mask = (1 << adev->sdma.num_inst_per_aid) - 1; inst_mask;
 	     inst_mask >>= adev->sdma.num_inst_per_aid, ++i) {
 		if ((inst_mask & mask) == mask)
 			adev->aid_mask |= (1 << i);
-- 
GitLab


From c1d3f627ff33bf1ae145209030a114c4985beddf Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 20 Jan 2023 12:42:00 +0530
Subject: [PATCH 1188/1791] drm/amdgpu: Fix mqd init on GFX v9.4.3

For MQD init, an XCC's queue is selected with GRBM select. However, for
initialization of MQD, values read from logical XCC0 registers are used.
This results in garbage values being read from XCC0 whose queue is not
selected. Change to read from the right XCC for MQD initialization.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 52041d4cb14af..1f1268cd5e09f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1456,7 +1456,7 @@ static void gfx_v9_4_3_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd
 	}
 }
 
-static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
+static int gfx_v9_4_3_xcc_mqd_init(struct amdgpu_ring *ring, int xcc_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_mqd *mqd = ring->mqd_ptr;
@@ -1483,14 +1483,14 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_hqd_eop_base_addr_hi = upper_32_bits(eop_base_addr);
 
 	/* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_EOP_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_EOP_CONTROL, EOP_SIZE,
 			(order_base_2(GFX9_MEC_HPD_SIZE / 4) - 1));
 
 	mqd->cp_hqd_eop_control = tmp;
 
 	/* enable doorbell? */
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_DOORBELL_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL);
 
 	if (ring->use_doorbell) {
 		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
@@ -1520,7 +1520,7 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_mqd_base_addr_hi = upper_32_bits(ring->mqd_gpu_addr);
 
 	/* set MQD vmid to 0 */
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_MQD_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MQD_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_MQD_CONTROL, VMID, 0);
 	mqd->cp_mqd_control = tmp;
 
@@ -1530,7 +1530,7 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_hqd_pq_base_hi = upper_32_bits(hqd_gpu_addr);
 
 	/* set up the HQD, this is similar to CP_RB0_CNTL */
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, QUEUE_SIZE,
 			    (order_base_2(ring->ring_size / 4) - 1));
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE,
@@ -1557,23 +1557,23 @@ static int gfx_v9_4_3_mqd_init(struct amdgpu_ring *ring)
 
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
 	ring->wptr = 0;
-	mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PQ_RPTR);
+	mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR);
 
 	/* set the vmid for the queue */
 	mqd->cp_hqd_vmid = 0;
 
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_PERSISTENT_STATE);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PERSISTENT_STATE, PRELOAD_SIZE, 0x53);
 	mqd->cp_hqd_persistent_state = tmp;
 
 	/* set MIN_IB_AVAIL_SIZE */
-	tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_IB_CONTROL);
+	tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_IB_CONTROL);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3);
 	mqd->cp_hqd_ib_control = tmp;
 
 	/* set static priority for a queue/ring */
 	gfx_v9_4_3_mqd_set_priority(ring, mqd);
-	mqd->cp_hqd_quantum = RREG32_SOC15(GC, GET_INST(GC, 0), regCP_HQD_QUANTUM);
+	mqd->cp_hqd_quantum = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_QUANTUM);
 
 	/* map_queues packet doesn't need activate the queue,
 	 * so only kiq need set this field.
@@ -1771,7 +1771,7 @@ static int gfx_v9_4_3_xcc_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
-		gfx_v9_4_3_mqd_init(ring);
+		gfx_v9_4_3_xcc_mqd_init(ring, xcc_id);
 		gfx_v9_4_3_xcc_kiq_init_register(ring, xcc_id);
 		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
@@ -1802,7 +1802,7 @@ static int gfx_v9_4_3_xcc_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id)
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
 		mutex_lock(&adev->srbm_mutex);
 		soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id));
-		gfx_v9_4_3_mqd_init(ring);
+		gfx_v9_4_3_xcc_mqd_init(ring, xcc_id);
 		soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id));
 		mutex_unlock(&adev->srbm_mutex);
 
-- 
GitLab


From 4482d3c94d7f1d6912521e6de23bb051bfcd084d Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Wed, 12 Oct 2022 21:58:29 -0400
Subject: [PATCH 1189/1791] drm/ttm: add NUMA node id to the pool
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This allows backing ttm_tt structure with pages from different NUMA
pools.

Tested-by: Graham Sider <graham.sider@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/ttm/ttm_device.c |  2 +-
 drivers/gpu/drm/ttm/ttm_pool.c   | 13 ++++++++-----
 include/drm/ttm/ttm_pool.h       |  4 +++-
 3 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c
index 64a59f46f6c34..43e27ab77f957 100644
--- a/drivers/gpu/drm/ttm/ttm_device.c
+++ b/drivers/gpu/drm/ttm/ttm_device.c
@@ -213,7 +213,7 @@ int ttm_device_init(struct ttm_device *bdev, struct ttm_device_funcs *funcs,
 	bdev->funcs = funcs;
 
 	ttm_sys_man_init(bdev);
-	ttm_pool_init(&bdev->pool, dev, use_dma_alloc, use_dma32);
+	ttm_pool_init(&bdev->pool, dev, NUMA_NO_NODE, use_dma_alloc, use_dma32);
 
 	bdev->vma_manager = vma_manager;
 	spin_lock_init(&bdev->lru_lock);
diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c
index 18c342a919a2e..afc4476c446fa 100644
--- a/drivers/gpu/drm/ttm/ttm_pool.c
+++ b/drivers/gpu/drm/ttm/ttm_pool.c
@@ -98,7 +98,7 @@ static struct page *ttm_pool_alloc_page(struct ttm_pool *pool, gfp_t gfp_flags,
 			__GFP_KSWAPD_RECLAIM;
 
 	if (!pool->use_dma_alloc) {
-		p = alloc_pages(gfp_flags, order);
+		p = alloc_pages_node(pool->nid, gfp_flags, order);
 		if (p)
 			p->private = order;
 		return p;
@@ -292,7 +292,7 @@ static struct ttm_pool_type *ttm_pool_select_type(struct ttm_pool *pool,
 						  enum ttm_caching caching,
 						  unsigned int order)
 {
-	if (pool->use_dma_alloc)
+	if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE)
 		return &pool->caching[caching].orders[order];
 
 #ifdef CONFIG_X86
@@ -550,29 +550,32 @@ EXPORT_SYMBOL(ttm_pool_free);
  *
  * @pool: the pool to initialize
  * @dev: device for DMA allocations and mappings
+ * @nid: NUMA node to use for allocations
  * @use_dma_alloc: true if coherent DMA alloc should be used
  * @use_dma32: true if GFP_DMA32 should be used
  *
  * Initialize the pool and its pool types.
  */
 void ttm_pool_init(struct ttm_pool *pool, struct device *dev,
-		   bool use_dma_alloc, bool use_dma32)
+		   int nid, bool use_dma_alloc, bool use_dma32)
 {
 	unsigned int i, j;
 
 	WARN_ON(!dev && use_dma_alloc);
 
 	pool->dev = dev;
+	pool->nid = nid;
 	pool->use_dma_alloc = use_dma_alloc;
 	pool->use_dma32 = use_dma32;
 
-	if (use_dma_alloc) {
+	if (use_dma_alloc || nid != NUMA_NO_NODE) {
 		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
 			for (j = 0; j < TTM_DIM_ORDER; ++j)
 				ttm_pool_type_init(&pool->caching[i].orders[j],
 						   pool, i, j);
 	}
 }
+EXPORT_SYMBOL(ttm_pool_init);
 
 /**
  * ttm_pool_fini - Cleanup a pool
@@ -586,7 +589,7 @@ void ttm_pool_fini(struct ttm_pool *pool)
 {
 	unsigned int i, j;
 
-	if (pool->use_dma_alloc) {
+	if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE) {
 		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
 			for (j = 0; j < TTM_DIM_ORDER; ++j)
 				ttm_pool_type_fini(&pool->caching[i].orders[j]);
diff --git a/include/drm/ttm/ttm_pool.h b/include/drm/ttm/ttm_pool.h
index ef09b23d29e31..23bd8be6d4f85 100644
--- a/include/drm/ttm/ttm_pool.h
+++ b/include/drm/ttm/ttm_pool.h
@@ -61,12 +61,14 @@ struct ttm_pool_type {
  * struct ttm_pool - Pool for all caching and orders
  *
  * @dev: the device we allocate pages for
+ * @nid: which numa node to use
  * @use_dma_alloc: if coherent DMA allocations should be used
  * @use_dma32: if GFP_DMA32 should be used
  * @caching: pools for each caching/order
  */
 struct ttm_pool {
 	struct device *dev;
+	int nid;
 
 	bool use_dma_alloc;
 	bool use_dma32;
@@ -81,7 +83,7 @@ int ttm_pool_alloc(struct ttm_pool *pool, struct ttm_tt *tt,
 void ttm_pool_free(struct ttm_pool *pool, struct ttm_tt *tt);
 
 void ttm_pool_init(struct ttm_pool *pool, struct device *dev,
-		   bool use_dma_alloc, bool use_dma32);
+		   int nid, bool use_dma_alloc, bool use_dma32);
 void ttm_pool_fini(struct ttm_pool *pool);
 
 int ttm_pool_debugfs(struct ttm_pool *pool, struct seq_file *m);
-- 
GitLab


From 8c45a8340dd097ea0d6be6f718c4882283d9645d Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Tue, 24 Jan 2023 10:10:14 -0500
Subject: [PATCH 1190/1791] drm/amdkfd: Cleanup KFD nodes creation

kfd node allocation outside kfd->num_nodes loop is not needed and causes
memory leak because kfd->num_nodes is at least equal to 1.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 7a963d0a34e2a..d7cffd91f1d7d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -729,26 +729,12 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	/* TODO: Needs to be updated for memory partitioning */
 	svm_migrate_init(kfd->adev);
 
-	/* Allocate the KFD node */
-	node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL);
-	if (!node) {
-		dev_err(kfd_device, "Error allocating KFD node\n");
-		goto node_alloc_error;
-	}
-
-	node->adev = kfd->adev;
-	node->kfd = kfd;
-	node->kfd2kgd = kfd->kfd2kgd;
-	node->vm_info.vmid_num_kfd = vmid_num_kfd;
-	node->vm_info.first_vmid_kfd = first_vmid_kfd;
-	node->vm_info.last_vmid_kfd = last_vmid_kfd;
-	node->max_proc_per_quantum = max_proc_per_quantum;
-	atomic_set(&node->sram_ecc_flag, 0);
-
 	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
 
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
 				kfd->num_nodes);
+
+	/* Allocate the KFD nodes */
 	for (i = 0; i < kfd->num_nodes; i++) {
 		node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL);
 		if (!node)
-- 
GitLab


From 1698e200e88db96aef7d16aa3d63df68a209ffbd Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 2 Feb 2023 11:10:08 -0500
Subject: [PATCH 1191/1791] drm/amdkfd: bind cpu and hiveless gpu to a hive if
 xgmi connected

If a CPU and GPU are xGMI connected but the GPU is hiveless with
respect to other GPUs, create a new CPU-GPU hive using the GPU's PCI
device location ID as the new hive ID to maintain fine grain memory
access usage.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 5373a79ac6a10..c7072fff778ec 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1328,8 +1328,15 @@ static void kfd_fill_iolink_non_crat_info(struct kfd_topology_device *dev)
 
 		/* Include the CPU peer in GPU hive if connected over xGMI. */
 		if (!peer_dev->gpu &&
-		    link->iolink_type == CRAT_IOLINK_TYPE_XGMI)
+		    link->iolink_type == CRAT_IOLINK_TYPE_XGMI) {
+			/*
+			 * If the GPU is not part of a GPU hive, use its pci
+			 * device location as the hive ID to bind with the CPU.
+			 */
+			if (!dev->node_props.hive_id)
+				dev->node_props.hive_id = pci_dev_id(dev->gpu->adev->pdev);
 			peer_dev->node_props.hive_id = dev->node_props.hive_id;
+		}
 
 		list_for_each_entry(inbound_link, &peer_dev->io_link_props,
 									list) {
-- 
GitLab


From 7b08b2e1caa04757bc7a35a46b3d5c83b1748423 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonny.jiang@amd.com>
Date: Tue, 31 Jan 2023 16:44:28 -0500
Subject: [PATCH 1192/1791] drm/amdgpu: add vcn_4_0_3 codec query

Add support for vcn_4_0_3 video codec query

Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 4138d96342662..5a63ec441a5bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -153,6 +153,24 @@ static const struct amdgpu_video_codecs rn_video_codecs_decode =
 	.codec_array = rn_video_codecs_decode_array,
 };
 
+static const struct amdgpu_video_codec_info vcn_4_0_3_video_codecs_decode_array[] = {
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 8192, 4352, 186)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG, 4096, 4096, 0)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)},
+	{codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)},
+};
+
+static const struct amdgpu_video_codecs vcn_4_0_3_video_codecs_decode = {
+	.codec_count = ARRAY_SIZE(vcn_4_0_3_video_codecs_decode_array),
+	.codec_array = vcn_4_0_3_video_codecs_decode_array,
+};
+
+static const struct amdgpu_video_codecs vcn_4_0_3_video_codecs_encode = {
+	.codec_count = 0,
+	.codec_array = NULL,
+};
+
 static int soc15_query_video_codecs(struct amdgpu_device *adev, bool encode,
 				    const struct amdgpu_video_codecs **codecs)
 {
@@ -185,6 +203,12 @@ static int soc15_query_video_codecs(struct amdgpu_device *adev, bool encode,
 			else
 				*codecs = &rn_video_codecs_decode;
 			return 0;
+		case IP_VERSION(4, 0, 3):
+			if (encode)
+				*codecs = &vcn_4_0_3_video_codecs_encode;
+			else
+				*codecs = &vcn_4_0_3_video_codecs_decode;
+			return 0;
 		default:
 			return -EINVAL;
 		}
-- 
GitLab


From 35ff4301ebc37bd45c18edae08afd2983dc9a338 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Fri, 3 Feb 2023 14:38:33 +0800
Subject: [PATCH 1193/1791] drm/amdgpu: enable context empty interrupt on sdma
 v4.4.2

With SDMA_CTNL.CTXEMPTY_INT_ENABLE set, the F32 clock can be gated when
SDMA finishes all job and goes to idle.

And no specific interrupt handling is required in driver.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 729e26a4a2e74..7aa26e716a2dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -890,6 +890,8 @@ static int sdma_v4_4_2_inst_start(struct amdgpu_device *adev,
 		/* set utc l1 enable flag always to 1 */
 		temp = RREG32_SDMA(i, regSDMA_CNTL);
 		temp = REG_SET_FIELD(temp, SDMA_CNTL, UTC_L1_ENABLE, 1);
+		/* enable context empty interrupt during initialization */
+		temp = REG_SET_FIELD(temp, SDMA_CNTL, CTXEMPTY_INT_ENABLE, 1);
 		WREG32_SDMA(i, regSDMA_CNTL, temp);
 
 		if (!amdgpu_sriov_vf(adev)) {
-- 
GitLab


From cbf9e46ab0fec29ac39d05d9a87fa66122bc9783 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 3 Feb 2023 13:17:51 +0530
Subject: [PATCH 1194/1791] drm/amdgpu: Enable MGCG on SDMAv4.4.2

Enable clock gating on SDMAv4.4.2 versions. Leave memory light sleep to
default.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 22 +++++++++++++---------
 drivers/gpu/drm/amd/amdgpu/soc15.c       |  4 ++--
 2 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 7aa26e716a2dd..dca0beec0252a 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1683,7 +1683,11 @@ static void sdma_v4_4_2_inst_update_medium_grain_light_sleep(
 	uint32_t data, def;
 	int i;
 
-	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) {
+	/* leave as default if it is not driver controlled */
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS))
+		return;
+
+	if (enable) {
 		for_each_inst(i, inst_mask) {
 			/* 1-not override: enable sdma mem light sleep */
 			def = data = RREG32_SDMA(i, regSDMA_POWER_CNTL);
@@ -1708,12 +1712,14 @@ static void sdma_v4_4_2_inst_update_medium_grain_clock_gating(
 	uint32_t data, def;
 	int i;
 
-	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) {
+	/* leave as default if it is not driver controlled */
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG))
+		return;
+
+	if (enable) {
 		for_each_inst(i, inst_mask) {
 			def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL);
-			data &= ~(SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK |
-				  SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK |
-				  SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK |
+			data &= ~(SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK |
 				  SDMA_CLK_CTRL__SOFT_OVERRIDE4_MASK |
 				  SDMA_CLK_CTRL__SOFT_OVERRIDE3_MASK |
 				  SDMA_CLK_CTRL__SOFT_OVERRIDE2_MASK |
@@ -1725,9 +1731,7 @@ static void sdma_v4_4_2_inst_update_medium_grain_clock_gating(
 	} else {
 		for_each_inst(i, inst_mask) {
 			def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL);
-			data |= (SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK |
-				 SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK |
-				 SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK |
+			data |= (SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK |
 				 SDMA_CLK_CTRL__SOFT_OVERRIDE4_MASK |
 				 SDMA_CLK_CTRL__SOFT_OVERRIDE3_MASK |
 				 SDMA_CLK_CTRL__SOFT_OVERRIDE2_MASK |
@@ -1773,7 +1777,7 @@ static void sdma_v4_4_2_get_clockgating_state(void *handle, u64 *flags)
 
 	/* AMD_CG_SUPPORT_SDMA_MGCG */
 	data = RREG32(SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, 0), regSDMA_CLK_CTRL));
-	if (!(data & SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK))
+	if (!(data & SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK))
 		*flags |= AMD_CG_SUPPORT_SDMA_MGCG;
 
 	/* AMD_CG_SUPPORT_SDMA_LS */
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 5a63ec441a5bd..100fdf5074b43 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1146,8 +1146,8 @@ static int soc15_common_early_init(void *handle)
 		adev->asic_funcs = &aqua_vanjaram_asic_funcs;
 		adev->cg_flags =
 			AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGCG |
-			AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_VCN_MGCG |
-			AMD_CG_SUPPORT_JPEG_MGCG;
+			AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_SDMA_MGCG |
+			AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_VCN_DPG |
-- 
GitLab


From 322a7e005db78b8a46ead91b7e3df3514cb658f0 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Feb 2023 20:54:08 +0800
Subject: [PATCH 1195/1791] drm/amdgpu: Add common helper to query ras error
 (v2)

Add common helper to query ras error status and
log error information, including memory block id
and erorr count. The helpers are applicable to IP
blocks that follow the new ras error logging design.
For IP blocks that don't support the new design,
please still implement ip specific helper to query
ras error.

v2: optimize struct amdgpu_ras_err_status_reg_entry
and the implementaion in helper (Lijo/Tao)

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 119 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  54 +++++++++++
 2 files changed, 173 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 22f401fd1901c..57e86af0c9064 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3103,3 +3103,122 @@ int amdgpu_ras_register_ras_block(struct amdgpu_device *adev,
 
 	return 0;
 }
+
+void amdgpu_ras_get_error_type_name(uint32_t err_type, char *err_type_name)
+{
+	if (!err_type_name)
+		return;
+
+	switch (err_type) {
+	case AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE:
+		sprintf(err_type_name, "correctable");
+		break;
+	case AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE:
+		sprintf(err_type_name, "uncorrectable");
+		break;
+	default:
+		sprintf(err_type_name, "unknown");
+		break;
+	}
+}
+
+bool amdgpu_ras_inst_get_memory_id_field(struct amdgpu_device *adev,
+					 const struct amdgpu_ras_err_status_reg_entry *reg_entry,
+					 uint32_t instance,
+					 uint32_t *memory_id)
+{
+	uint32_t err_status_lo_data, err_status_lo_offset;
+
+	if (!reg_entry)
+		return false;
+
+	err_status_lo_offset =
+		AMDGPU_RAS_REG_ENTRY_OFFSET(reg_entry->hwip, instance,
+					    reg_entry->seg_lo, reg_entry->reg_lo);
+	err_status_lo_data = RREG32(err_status_lo_offset);
+
+	if ((reg_entry->flags & AMDGPU_RAS_ERR_STATUS_VALID) &&
+	    !REG_GET_FIELD(err_status_lo_data, ERR_STATUS_LO, ERR_STATUS_VALID_FLAG))
+		return false;
+
+	*memory_id = REG_GET_FIELD(err_status_lo_data, ERR_STATUS_LO, MEMORY_ID);
+
+	return true;
+}
+
+bool amdgpu_ras_inst_get_err_cnt_field(struct amdgpu_device *adev,
+				       const struct amdgpu_ras_err_status_reg_entry *reg_entry,
+				       uint32_t instance,
+				       unsigned long *err_cnt)
+{
+	uint32_t err_status_hi_data, err_status_hi_offset;
+
+	if (!reg_entry)
+		return false;
+
+	err_status_hi_offset =
+		AMDGPU_RAS_REG_ENTRY_OFFSET(reg_entry->hwip, instance,
+					    reg_entry->seg_hi, reg_entry->reg_hi);
+	err_status_hi_data = RREG32(err_status_hi_offset);
+
+	if ((reg_entry->flags & AMDGPU_RAS_ERR_INFO_VALID) &&
+	    !REG_GET_FIELD(err_status_hi_data, ERR_STATUS_HI, ERR_INFO_VALID_FLAG))
+		return false;
+
+	/* read err count */
+	*err_cnt = REG_GET_FIELD(err_status_hi_data, ERR_STATUS, ERR_CNT);
+
+	return true;
+}
+
+void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev,
+					   const struct amdgpu_ras_err_status_reg_entry *reg_list,
+					   uint32_t reg_list_size,
+					   const struct amdgpu_ras_memory_id_entry *mem_list,
+					   uint32_t mem_list_size,
+					   uint32_t instance,
+					   uint32_t err_type,
+					   unsigned long *err_count)
+{
+	uint32_t memory_id;
+	unsigned long err_cnt;
+	char err_type_name[16];
+	uint32_t i, j;
+
+	for (i = 0; i < reg_list_size; i++) {
+		/* query err_cnt from err_status_hi */
+		if (!amdgpu_ras_inst_get_err_cnt_field(adev, &reg_list[i],
+						       instance, &err_cnt) ||
+		    !err_cnt)
+			continue;
+
+		/* query memory_id from err_status_lo */
+		if (!amdgpu_ras_inst_get_memory_id_field(adev, &reg_list[i],
+							 instance, &memory_id))
+			continue;
+
+		*err_count += err_cnt;
+
+		/* log the errors */
+		amdgpu_ras_get_error_type_name(err_type, err_type_name);
+		if (!mem_list) {
+			/* memory_list is not supported */
+			dev_info(adev->dev,
+				 "%ld %s hardware errors detected in %s, instance: %d, memory_id: %d\n",
+				 err_cnt, err_type_name,
+				 reg_list[i].block_name,
+				 instance, memory_id);
+		} else {
+			for (j = 0; j < mem_list_size; j++) {
+				if (memory_id == mem_list[j].memory_id) {
+					dev_info(adev->dev,
+						 "%ld %s hardware errors detected in %s, instance: %d, memory block: %s\n",
+						 err_cnt, err_type_name,
+						 reg_list[i].block_name,
+						 instance, mem_list[j].name);
+					break;
+				}
+			}
+		}
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 17b3d1992e808..c820af7f1a4b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -314,6 +314,43 @@ enum amdgpu_ras_ret {
 	AMDGPU_RAS_PT,
 };
 
+/* ras error status reisger fields */
+#define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT	0x0
+#define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK	0x00000001L
+#define ERR_STATUS_LO__MEMORY_ID__SHIFT			0x18
+#define ERR_STATUS_LO__MEMORY_ID_MASK			0xFF000000L
+#define ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT	0x2
+#define ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK		0x00000004L
+#define ERR_STATUS__ERR_CNT__SHIFT			0x17
+#define ERR_STATUS__ERR_CNT_MASK			0x03800000L
+
+#define AMDGPU_RAS_REG_ENTRY(ip, inst, reg_lo, reg_hi) \
+	ip##_HWIP, inst, reg_lo##_BASE_IDX, reg_lo, reg_hi##_BASE_IDX, reg_hi
+
+#define AMDGPU_RAS_REG_ENTRY_OFFSET(hwip, ip_inst, segment, reg) \
+	(adev->reg_offset[hwip][ip_inst][segment] + (reg))
+
+#define AMDGPU_RAS_ERR_INFO_VALID	(1 << 0)
+#define AMDGPU_RAS_ERR_STATUS_VALID	(1 << 1)
+#define AMDGPU_RAS_ERR_ADDRESS_VALID	(1 << 2)
+
+struct amdgpu_ras_err_status_reg_entry {
+	uint32_t hwip;
+	uint32_t ip_inst;
+	uint32_t seg_lo;
+	uint32_t reg_lo;
+	uint32_t seg_hi;
+	uint32_t reg_hi;
+	uint32_t reg_inst;
+	uint32_t flags;
+	const char *block_name;
+};
+
+struct amdgpu_ras_memory_id_entry {
+	uint32_t memory_id;
+	const char *name;
+};
+
 struct ras_common_if {
 	enum amdgpu_ras_block block;
 	enum amdgpu_ras_error_type type;
@@ -696,4 +733,21 @@ int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_co
 int amdgpu_ras_register_ras_block(struct amdgpu_device *adev,
 				struct amdgpu_ras_block_object *ras_block_obj);
 void amdgpu_ras_interrupt_fatal_error_handler(struct amdgpu_device *adev);
+void amdgpu_ras_get_error_type_name(uint32_t err_type, char *err_type_name);
+bool amdgpu_ras_inst_get_memory_id_field(struct amdgpu_device *adev,
+					 const struct amdgpu_ras_err_status_reg_entry *reg_entry,
+					 uint32_t instance,
+					 uint32_t *memory_id);
+bool amdgpu_ras_inst_get_err_cnt_field(struct amdgpu_device *adev,
+				       const struct amdgpu_ras_err_status_reg_entry *reg_entry,
+				       uint32_t instance,
+				       unsigned long *err_cnt);
+void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev,
+					   const struct amdgpu_ras_err_status_reg_entry *reg_list,
+					   uint32_t reg_list_size,
+					   const struct amdgpu_ras_memory_id_entry *mem_list,
+					   uint32_t mem_list_size,
+					   uint32_t instance,
+					   uint32_t err_type,
+					   unsigned long *err_count);
 #endif
-- 
GitLab


From e53a3250f76b8a0dd5b533bd0ce0dc821055e77d Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Fri, 3 Feb 2023 16:10:37 +0800
Subject: [PATCH 1196/1791] drm/amdgpu: Add common helper to reset ras error

Add common helper to reset ras error status. It
applies to IP blocks that follow the new ras error
logging register design, and need to write 0 to
reset the error status. For IP blocks that don't
support the new design, please still implement ip
specific helper.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 20 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  4 ++++
 2 files changed, 24 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 57e86af0c9064..8a16a06cb78a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3222,3 +3222,23 @@ void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev,
 		}
 	}
 }
+
+void amdgpu_ras_inst_reset_ras_error_count(struct amdgpu_device *adev,
+					   const struct amdgpu_ras_err_status_reg_entry *reg_list,
+					   uint32_t reg_list_size,
+					   uint32_t instance)
+{
+	uint32_t err_status_lo_offset, err_status_hi_offset;
+	uint32_t i;
+
+	for (i = 0; i < reg_list_size; i++) {
+		err_status_lo_offset =
+			AMDGPU_RAS_REG_ENTRY_OFFSET(reg_list[i].hwip, instance,
+						    reg_list[i].seg_lo, reg_list[i].reg_lo);
+		err_status_hi_offset =
+			AMDGPU_RAS_REG_ENTRY_OFFSET(reg_list[i].hwip, instance,
+						    reg_list[i].seg_hi, reg_list[i].reg_hi);
+		WREG32(err_status_lo_offset, 0);
+		WREG32(err_status_hi_offset, 0);
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index c820af7f1a4b0..e96333d0c2694 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -750,4 +750,8 @@ void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev,
 					   uint32_t instance,
 					   uint32_t err_type,
 					   unsigned long *err_count);
+void amdgpu_ras_inst_reset_ras_error_count(struct amdgpu_device *adev,
+					   const struct amdgpu_ras_err_status_reg_entry *reg_list,
+					   uint32_t reg_list_size,
+					   uint32_t instance);
 #endif
-- 
GitLab


From d90d90a1978af6530c7d8b201c4ab117d0506b1a Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Fri, 23 Dec 2022 15:54:43 +0800
Subject: [PATCH 1197/1791] drm/amdgpu: Add sdma v4_4_2 ras registers

SDMA_UE_ERR_STATUS_HI|LO are introduced in v4_4_2
to replace SDMA_EDC_COUNTER/COUNTER2 registers to
log SDMA RAS errors

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../include/asic_reg/sdma/sdma_4_4_2_offset.h |  4 ++++
 .../asic_reg/sdma/sdma_4_4_2_sh_mask.h        | 24 +++++++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h
index 31bef0776ded5..ead81aeffd679 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h
@@ -211,6 +211,10 @@
 #define regSDMA_RAS_STATUS_BASE_IDX                                                                     0
 #define regSDMA_CLK_STATUS                                                                              0x0068
 #define regSDMA_CLK_STATUS_BASE_IDX                                                                     0
+#define regSDMA_UE_ERR_STATUS_LO                                                                        0x0069
+#define regSDMA_UE_ERR_STATUS_LO_BASE_IDX                                                               0
+#define regSDMA_UE_ERR_STATUS_HI                                                                        0x006a
+#define regSDMA_UE_ERR_STATUS_HI_BASE_IDX                                                               0
 #define regSDMA_POWER_CNTL                                                                              0x006b
 #define regSDMA_POWER_CNTL_BASE_IDX                                                                     0
 #define regSDMA_CLK_CTRL                                                                                0x006c
diff --git a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h
index e46cb33393554..290953bdf1d69 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h
@@ -1171,6 +1171,30 @@
 #define SDMA_CLK_STATUS__F32_CLK_MASK                                                                         0x00000008L
 #define SDMA_CLK_STATUS__CE_CLK_MASK                                                                          0x00000010L
 #define SDMA_CLK_STATUS__PERF_CLK_MASK                                                                        0x00000020L
+//SDMA_UE_ERR_STATUS_LO
+#define SDMA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                   0x0
+#define SDMA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define SDMA_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                 0x2
+#define SDMA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                               0x18
+#define SDMA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                     0x00000001L
+#define SDMA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define SDMA_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define SDMA_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                 0xFF000000L
+//SDMA_UE_ERR_STATUS_HI
+#define SDMA_UE_ERR_STATUS_HI__ECC__SHIFT                                                                     0x0
+#define SDMA_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                  0x1
+#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                     0x2
+#define SDMA_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                0x3
+#define SDMA_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                  0x17
+#define SDMA_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                 0x1a
+#define SDMA_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                0x1d
+#define SDMA_UE_ERR_STATUS_HI__ECC_MASK                                                                       0x00000001L
+#define SDMA_UE_ERR_STATUS_HI__PARITY_MASK                                                                    0x00000002L
+#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                       0x00000004L
+#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                  0x007FFFF8L
+#define SDMA_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                    0x03800000L
+#define SDMA_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                   0x1C000000L
+#define SDMA_UE_ERR_STATUS_HI__RESERVED_MASK                                                                  0xE0000000L
 //SDMA_POWER_CNTL
 #define SDMA_POWER_CNTL__PG_CNTL_ENABLE__SHIFT                                                                0x0
 #define SDMA_POWER_CNTL__EXT_PG_POWER_ON_REQ__SHIFT                                                           0x1
-- 
GitLab


From dc37a9194ad20c4f09f22ec79cc2b5e5eb57c5a2 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 5 Feb 2023 22:54:50 +0800
Subject: [PATCH 1198/1791] drm/amdgpu: Add query_ras_error_count for sdma
 v4_4_2

Add query_ras_error_count callback for sdma
v4_4_2. It will be used to query and log sdma
uncorrectable error count and memory block.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h | 28 +++++++++++
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 64 ++++++++++++++++++++++++
 2 files changed, 92 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 62afb282a3ff1..513ac22120c1f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -62,6 +62,34 @@ struct amdgpu_sdma_instance {
 	uint32_t		aid_id;
 };
 
+enum amdgpu_sdma_ras_memory_id {
+	AMDGPU_SDMA_MBANK_DATA_BUF0 = 1,
+	AMDGPU_SDMA_MBANK_DATA_BUF1 = 2,
+	AMDGPU_SDMA_MBANK_DATA_BUF2 = 3,
+	AMDGPU_SDMA_MBANK_DATA_BUF3 = 4,
+	AMDGPU_SDMA_MBANK_DATA_BUF4 = 5,
+	AMDGPU_SDMA_MBANK_DATA_BUF5 = 6,
+	AMDGPU_SDMA_MBANK_DATA_BUF6 = 7,
+	AMDGPU_SDMA_MBANK_DATA_BUF7 = 8,
+	AMDGPU_SDMA_MBANK_DATA_BUF8 = 9,
+	AMDGPU_SDMA_MBANK_DATA_BUF9 = 10,
+	AMDGPU_SDMA_MBANK_DATA_BUF10 = 11,
+	AMDGPU_SDMA_MBANK_DATA_BUF11 = 12,
+	AMDGPU_SDMA_MBANK_DATA_BUF12 = 13,
+	AMDGPU_SDMA_MBANK_DATA_BUF13 = 14,
+	AMDGPU_SDMA_MBANK_DATA_BUF14 = 15,
+	AMDGPU_SDMA_MBANK_DATA_BUF15 = 16,
+	AMDGPU_SDMA_UCODE_BUF = 17,
+	AMDGPU_SDMA_RB_CMD_BUF = 18,
+	AMDGPU_SDMA_IB_CMD_BUF = 19,
+	AMDGPU_SDMA_UTCL1_RD_FIFO = 20,
+	AMDGPU_SDMA_UTCL1_RDBST_FIFO = 21,
+	AMDGPU_SDMA_UTCL1_WR_FIFO = 22,
+	AMDGPU_SDMA_DATA_LUT_FIFO = 23,
+	AMDGPU_SDMA_SPLIT_DAT_BUF = 24,
+	AMDGPU_SDMA_MEMORY_BLOCK_LAST,
+};
+
 struct amdgpu_sdma_ras {
 	struct amdgpu_ras_block_object ras_block;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index dca0beec0252a..925ca6218a5e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -2071,3 +2071,67 @@ struct amdgpu_xcp_ip_funcs sdma_v4_4_2_xcp_funcs = {
 	.suspend = &sdma_v4_4_2_xcp_suspend,
 	.resume = &sdma_v4_4_2_xcp_resume
 };
+
+static const struct amdgpu_ras_err_status_reg_entry sdma_v4_2_2_ue_reg_list[] = {
+	{AMDGPU_RAS_REG_ENTRY(SDMA0, 0, regSDMA_UE_ERR_STATUS_LO, regSDMA_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SDMA"},
+};
+
+static const struct amdgpu_ras_memory_id_entry sdma_v4_4_2_ras_memory_list[] = {
+	{AMDGPU_SDMA_MBANK_DATA_BUF0, "SDMA_MBANK_DATA_BUF0"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF1, "SDMA_MBANK_DATA_BUF1"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF2, "SDMA_MBANK_DATA_BUF2"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF3, "SDMA_MBANK_DATA_BUF3"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF4, "SDMA_MBANK_DATA_BUF4"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF5, "SDMA_MBANK_DATA_BUF5"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF6, "SDMA_MBANK_DATA_BUF6"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF7, "SDMA_MBANK_DATA_BUF7"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF8, "SDMA_MBANK_DATA_BUF8"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF9, "SDMA_MBANK_DATA_BUF9"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF10, "SDMA_MBANK_DATA_BUF10"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF11, "SDMA_MBANK_DATA_BUF11"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF12, "SDMA_MBANK_DATA_BUF12"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF13, "SDMA_MBANK_DATA_BUF13"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF14, "SDMA_MBANK_DATA_BUF14"},
+	{AMDGPU_SDMA_MBANK_DATA_BUF15, "SDMA_MBANK_DATA_BUF15"},
+	{AMDGPU_SDMA_UCODE_BUF, "SDMA_UCODE_BUF"},
+	{AMDGPU_SDMA_RB_CMD_BUF, "SDMA_RB_CMD_BUF"},
+	{AMDGPU_SDMA_IB_CMD_BUF, "SDMA_IB_CMD_BUF"},
+	{AMDGPU_SDMA_UTCL1_RD_FIFO, "SDMA_UTCL1_RD_FIFO"},
+	{AMDGPU_SDMA_UTCL1_RDBST_FIFO, "SDMA_UTCL1_RDBST_FIFO"},
+	{AMDGPU_SDMA_UTCL1_WR_FIFO, "SDMA_UTCL1_WR_FIFO"},
+	{AMDGPU_SDMA_DATA_LUT_FIFO, "SDMA_DATA_LUT_FIFO"},
+	{AMDGPU_SDMA_SPLIT_DAT_BUF, "SDMA_SPLIT_DAT_BUF"},
+};
+
+static void sdma_v4_4_2_inst_query_ras_error_count(struct amdgpu_device *adev,
+						   uint32_t sdma_inst,
+						   void *ras_err_status)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status;
+
+	/* sdma v4_4_2 doesn't support query ce counts */
+	amdgpu_ras_inst_query_ras_error_count(adev,
+					sdma_v4_2_2_ue_reg_list,
+					ARRAY_SIZE(sdma_v4_2_2_ue_reg_list),
+					sdma_v4_4_2_ras_memory_list,
+					ARRAY_SIZE(sdma_v4_4_2_ras_memory_list),
+					sdma_inst,
+					AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+					&err_data->ue_count);
+}
+
+static void sdma_v4_4_2_query_ras_error_count(struct amdgpu_device *adev,
+					      void *ras_err_status)
+{
+	uint32_t inst_mask;
+	int i = 0;
+
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		for_each_inst(i, inst_mask)
+			sdma_v4_4_2_inst_query_ras_error_count(adev, i, ras_err_status);
+	} else {
+		dev_warn(adev->dev, "SDMA RAS is not supported\n");
+	}
+}
-- 
GitLab


From a64b15520cc3a19bc6ca807870b51b37774a4d3d Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 12:19:57 +0800
Subject: [PATCH 1199/1791] drm/amdgpu: Add reset_ras_error_count for sdma
 v4_4_2

Add reset_ras_error_count callback for sdma
v4_4_2. It will be used to reset sdma ras error
count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 925ca6218a5e9..f0333822df784 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -2135,3 +2135,26 @@ static void sdma_v4_4_2_query_ras_error_count(struct amdgpu_device *adev,
 		dev_warn(adev->dev, "SDMA RAS is not supported\n");
 	}
 }
+
+static void sdma_v4_4_2_inst_reset_ras_error_count(struct amdgpu_device *adev,
+						   uint32_t sdma_inst)
+{
+	amdgpu_ras_inst_reset_ras_error_count(adev,
+					sdma_v4_2_2_ue_reg_list,
+					ARRAY_SIZE(sdma_v4_2_2_ue_reg_list),
+					sdma_inst);
+}
+
+static void sdma_v4_4_2_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	uint32_t inst_mask;
+	int i = 0;
+
+	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		for_each_inst(i, inst_mask)
+			sdma_v4_4_2_inst_reset_ras_error_count(adev, i);
+	} else {
+		dev_warn(adev->dev, "SDMA RAS is not supported\n");
+	}
+}
-- 
GitLab


From 1e69fde70030e2fd2b729de5998d3fd6c94be238 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 23:29:28 +0800
Subject: [PATCH 1200/1791] drm/amdgpu: Initialize sdma v4_4_2 ras function

Initialize sdma v4_4_2 ras function and interrupt
handler.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 48 +++++++++++++++++++-----
 1 file changed, 39 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index f0333822df784..570ea68c521fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -54,6 +54,7 @@ static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_buffer_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_vm_pte_funcs(struct amdgpu_device *adev);
 static void sdma_v4_4_2_set_irq_funcs(struct amdgpu_device *adev);
+static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev);
 
 static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev,
 		u32 instance, u32 offset)
@@ -1254,6 +1255,7 @@ static int sdma_v4_4_2_early_init(void *handle)
 	sdma_v4_4_2_set_buffer_funcs(adev);
 	sdma_v4_4_2_set_vm_pte_funcs(adev);
 	sdma_v4_4_2_set_irq_funcs(adev);
+	sdma_v4_4_2_set_ras_funcs(adev);
 
 	return 0;
 }
@@ -1377,6 +1379,11 @@ static int sdma_v4_4_2_sw_init(void *handle)
 		}
 	}
 
+	if (amdgpu_sdma_ras_sw_init(adev)) {
+		dev_err(adev->dev, "fail to initialize sdma ras block\n");
+		return -EINVAL;
+	}
+
 	return r;
 }
 
@@ -1558,7 +1565,7 @@ static int sdma_v4_4_2_process_ras_data_cb(struct amdgpu_device *adev,
 	 * be disabled and the driver should only look for the aggregated
 	 * interrupt via sync flood
 	 */
-	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA))
 		goto out;
 
 	instance = sdma_v4_4_2_irq_id_to_seq(entry->client_id);
@@ -1597,15 +1604,22 @@ static int sdma_v4_4_2_set_ecc_irq_state(struct amdgpu_device *adev,
 					unsigned type,
 					enum amdgpu_interrupt_state state)
 {
-	u32 sdma_edc_config;
+	u32 sdma_cntl;
 
-	sdma_edc_config = RREG32_SDMA(type, regCC_SDMA_EDC_CONFIG);
-	/*
-	 * FIXME: This was inherited from Aldebaran, but no this field
-	 * definition in the regspec of both Aldebaran and SDMA 4.4.2
-	 */
-	sdma_edc_config |= (state == AMDGPU_IRQ_STATE_ENABLE) ? (1 << 2) : 0;
-	WREG32_SDMA(type, regCC_SDMA_EDC_CONFIG, sdma_edc_config);
+	sdma_cntl = RREG32_SDMA(type, regSDMA_CNTL);
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+		sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA_CNTL,
+					  DRAM_ECC_INT_ENABLE, 0);
+		WREG32_SDMA(type, regSDMA_CNTL, sdma_cntl);
+		break;
+	/* sdma ecc interrupt is enabled by default
+	 * driver doesn't need to do anything to
+	 * enable the interrupt */
+	case AMDGPU_IRQ_STATE_ENABLE:
+	default:
+		break;
+	}
 
 	return 0;
 }
@@ -2158,3 +2172,19 @@ static void sdma_v4_4_2_reset_ras_error_count(struct amdgpu_device *adev)
 		dev_warn(adev->dev, "SDMA RAS is not supported\n");
 	}
 }
+
+static const struct amdgpu_ras_block_hw_ops sdma_v4_4_2_ras_hw_ops = {
+	.query_ras_error_count = sdma_v4_4_2_query_ras_error_count,
+	.reset_ras_error_count = sdma_v4_4_2_reset_ras_error_count,
+};
+
+static struct amdgpu_sdma_ras sdma_v4_4_2_ras = {
+	.ras_block = {
+		.hw_ops = &sdma_v4_4_2_ras_hw_ops,
+	},
+};
+
+static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev)
+{
+	adev->sdma.ras = &sdma_v4_4_2_ras;
+}
-- 
GitLab


From 90cbee204e6619e47b1ec9fc14ebe03852585dac Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Wed, 28 Dec 2022 18:18:38 +0800
Subject: [PATCH 1201/1791] drm/amdgpu: Add mmhub v1_8_0 ras err status
 registers

add new ras error status registers introduced in
mmhub v1_8_0 to log mmea and mm_cane ras err, including
MMEAx_UE|CE_ERR_STATUS_LO|HI
MM_CANE_UE|CE_ERR_STATUS_LO|HI

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../asic_reg/mmhub/mmhub_1_8_0_offset.h       |  56 ++-
 .../asic_reg/mmhub/mmhub_1_8_0_sh_mask.h      | 325 +++++++++++++++++-
 2 files changed, 373 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h
index 8bcc81f2dfc0b..879ee9de3ff31 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h
@@ -1491,6 +1491,10 @@
 #define regMMEA0_PERFCOUNTER1_CFG_BASE_IDX                                                              0
 #define regMMEA0_PERFCOUNTER_RSLT_CNTL                                                                  0x0400
 #define regMMEA0_PERFCOUNTER_RSLT_CNTL_BASE_IDX                                                         0
+#define regMMEA0_UE_ERR_STATUS_LO                                                                       0x0406
+#define regMMEA0_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regMMEA0_UE_ERR_STATUS_HI                                                                       0x0407
+#define regMMEA0_UE_ERR_STATUS_HI_BASE_IDX                                                              0
 #define regMMEA0_DSM_CNTL                                                                               0x0408
 #define regMMEA0_DSM_CNTL_BASE_IDX                                                                      0
 #define regMMEA0_DSM_CNTLA                                                                              0x0409
@@ -1511,8 +1515,12 @@
 #define regMMEA0_ERR_STATUS_BASE_IDX                                                                    0
 #define regMMEA0_MISC2                                                                                  0x0412
 #define regMMEA0_MISC2_BASE_IDX                                                                         0
+#define regMMEA0_CE_ERR_STATUS_LO                                                                       0x0414
+#define regMMEA0_CE_ERR_STATUS_LO_BASE_IDX                                                              0
 #define regMMEA0_MISC_AON                                                                               0x0415
 #define regMMEA0_MISC_AON_BASE_IDX                                                                      0
+#define regMMEA0_CE_ERR_STATUS_HI                                                                       0x0416
+#define regMMEA0_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 
 // addressBlock: aid_mmhub_ea_mmeadec1
@@ -1709,6 +1717,10 @@
 #define regMMEA1_PERFCOUNTER1_CFG_BASE_IDX                                                              0
 #define regMMEA1_PERFCOUNTER_RSLT_CNTL                                                                  0x0540
 #define regMMEA1_PERFCOUNTER_RSLT_CNTL_BASE_IDX                                                         0
+#define regMMEA1_UE_ERR_STATUS_LO                                                                       0x0546
+#define regMMEA1_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regMMEA1_UE_ERR_STATUS_HI                                                                       0x0547
+#define regMMEA1_UE_ERR_STATUS_HI_BASE_IDX                                                              0
 #define regMMEA1_DSM_CNTL                                                                               0x0548
 #define regMMEA1_DSM_CNTL_BASE_IDX                                                                      0
 #define regMMEA1_DSM_CNTLA                                                                              0x0549
@@ -1729,8 +1741,12 @@
 #define regMMEA1_ERR_STATUS_BASE_IDX                                                                    0
 #define regMMEA1_MISC2                                                                                  0x0552
 #define regMMEA1_MISC2_BASE_IDX                                                                         0
+#define regMMEA1_CE_ERR_STATUS_LO                                                                       0x0554
+#define regMMEA1_CE_ERR_STATUS_LO_BASE_IDX                                                              0
 #define regMMEA1_MISC_AON                                                                               0x0555
 #define regMMEA1_MISC_AON_BASE_IDX                                                                      0
+#define regMMEA1_CE_ERR_STATUS_HI                                                                       0x0556
+#define regMMEA1_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 
 // addressBlock: aid_mmhub_ea_mmeadec2
@@ -1927,6 +1943,10 @@
 #define regMMEA2_PERFCOUNTER1_CFG_BASE_IDX                                                              0
 #define regMMEA2_PERFCOUNTER_RSLT_CNTL                                                                  0x0680
 #define regMMEA2_PERFCOUNTER_RSLT_CNTL_BASE_IDX                                                         0
+#define regMMEA2_UE_ERR_STATUS_LO                                                                       0x0686
+#define regMMEA2_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regMMEA2_UE_ERR_STATUS_HI                                                                       0x0687
+#define regMMEA2_UE_ERR_STATUS_HI_BASE_IDX                                                              0
 #define regMMEA2_DSM_CNTL                                                                               0x0688
 #define regMMEA2_DSM_CNTL_BASE_IDX                                                                      0
 #define regMMEA2_DSM_CNTLA                                                                              0x0689
@@ -1947,8 +1967,12 @@
 #define regMMEA2_ERR_STATUS_BASE_IDX                                                                    0
 #define regMMEA2_MISC2                                                                                  0x0692
 #define regMMEA2_MISC2_BASE_IDX                                                                         0
+#define regMMEA2_CE_ERR_STATUS_LO                                                                       0x0694
+#define regMMEA2_CE_ERR_STATUS_LO_BASE_IDX                                                              0
 #define regMMEA2_MISC_AON                                                                               0x0695
 #define regMMEA2_MISC_AON_BASE_IDX                                                                      0
+#define regMMEA2_CE_ERR_STATUS_HI                                                                       0x0696
+#define regMMEA2_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 
 // addressBlock: aid_mmhub_ea_mmeadec3
@@ -2145,6 +2169,10 @@
 #define regMMEA3_PERFCOUNTER1_CFG_BASE_IDX                                                              0
 #define regMMEA3_PERFCOUNTER_RSLT_CNTL                                                                  0x07c0
 #define regMMEA3_PERFCOUNTER_RSLT_CNTL_BASE_IDX                                                         0
+#define regMMEA3_UE_ERR_STATUS_LO                                                                       0x07c6
+#define regMMEA3_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regMMEA3_UE_ERR_STATUS_HI                                                                       0x07c7
+#define regMMEA3_UE_ERR_STATUS_HI_BASE_IDX                                                              0
 #define regMMEA3_DSM_CNTL                                                                               0x07c8
 #define regMMEA3_DSM_CNTL_BASE_IDX                                                                      0
 #define regMMEA3_DSM_CNTLA                                                                              0x07c9
@@ -2165,9 +2193,12 @@
 #define regMMEA3_ERR_STATUS_BASE_IDX                                                                    0
 #define regMMEA3_MISC2                                                                                  0x07d2
 #define regMMEA3_MISC2_BASE_IDX                                                                         0
+#define regMMEA3_CE_ERR_STATUS_LO                                                                       0x07d4
+#define regMMEA3_CE_ERR_STATUS_LO_BASE_IDX                                                              0
 #define regMMEA3_MISC_AON                                                                               0x07d5
 #define regMMEA3_MISC_AON_BASE_IDX                                                                      0
-
+#define regMMEA3_CE_ERR_STATUS_HI                                                                       0x07d6
+#define regMMEA3_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 // addressBlock: aid_mmhub_ea_mmeadec4
 // base address: 0x62000
@@ -2363,6 +2394,10 @@
 #define regMMEA4_PERFCOUNTER1_CFG_BASE_IDX                                                              0
 #define regMMEA4_PERFCOUNTER_RSLT_CNTL                                                                  0x0900
 #define regMMEA4_PERFCOUNTER_RSLT_CNTL_BASE_IDX                                                         0
+#define regMMEA4_UE_ERR_STATUS_LO                                                                       0x0906
+#define regMMEA4_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regMMEA4_UE_ERR_STATUS_HI                                                                       0x0907
+#define regMMEA4_UE_ERR_STATUS_HI_BASE_IDX                                                              0
 #define regMMEA4_DSM_CNTL                                                                               0x0908
 #define regMMEA4_DSM_CNTL_BASE_IDX                                                                      0
 #define regMMEA4_DSM_CNTLA                                                                              0x0909
@@ -2383,9 +2418,12 @@
 #define regMMEA4_ERR_STATUS_BASE_IDX                                                                    0
 #define regMMEA4_MISC2                                                                                  0x0912
 #define regMMEA4_MISC2_BASE_IDX                                                                         0
+#define regMMEA4_CE_ERR_STATUS_LO                                                                       0x0914
+#define regMMEA4_CE_ERR_STATUS_LO_BASE_IDX                                                              0
 #define regMMEA4_MISC_AON                                                                               0x0915
 #define regMMEA4_MISC_AON_BASE_IDX                                                                      0
-
+#define regMMEA4_CE_ERR_STATUS_HI                                                                       0x0916
+#define regMMEA4_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 // addressBlock: aid_mmhub_pctldec0
 // base address: 0x62a00
@@ -3310,5 +3348,19 @@
 #define regL2TLB_PERFCOUNTER_HI                                                                         0x0d2d
 #define regL2TLB_PERFCOUNTER_HI_BASE_IDX                                                                0
 
+// addressBlock: aid_mmhub_mm_cane_mmcanedec
+// base address: 0x635f0
+#define regMM_CANE_ICG_CTRL                                                                             0x0d8a
+#define regMM_CANE_ICG_CTRL_BASE_IDX                                                                    0
+#define regMM_CANE_ERR_STATUS                                                                           0x0d8c
+#define regMM_CANE_ERR_STATUS_BASE_IDX                                                                  0
+#define regMM_CANE_UE_ERR_STATUS_LO                                                                     0x0d8d
+#define regMM_CANE_UE_ERR_STATUS_LO_BASE_IDX                                                            0
+#define regMM_CANE_UE_ERR_STATUS_HI                                                                     0x0d8e
+#define regMM_CANE_UE_ERR_STATUS_HI_BASE_IDX                                                            0
+#define regMM_CANE_CE_ERR_STATUS_LO                                                                     0x0d8f
+#define regMM_CANE_CE_ERR_STATUS_LO_BASE_IDX                                                            0
+#define regMM_CANE_CE_ERR_STATUS_HI                                                                     0x0d90
+#define regMM_CANE_CE_ERR_STATUS_HI_BASE_IDX                                                            0
 
 #endif
diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h
index af41468ce69ff..088c1f02aa433 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h
@@ -10470,6 +10470,30 @@
 #define MMEA0_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK                                                          0x01000000L
 #define MMEA0_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK                                                           0x02000000L
 #define MMEA0_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK                                                0x04000000L
+//MMEA0_UE_ERR_STATUS_LO
+#define MMEA0_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA0_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA0_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA0_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA0_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//MMEA0_UE_ERR_STATUS_HI
+#define MMEA0_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA0_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA0_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define MMEA0_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define MMEA0_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                         0x1d
+#define MMEA0_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA0_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA0_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define MMEA0_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define MMEA0_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                           0xE0000000L
 //MMEA0_DSM_CNTL
 #define MMEA0_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                               0x0
 #define MMEA0_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                              0x2
@@ -10718,12 +10742,35 @@
 #define MMEA0_MISC2__DRAM_WR_THROTTLE_MASK                                                                    0x00020000L
 #define MMEA0_MISC2__GMI_RD_THROTTLE_MASK                                                                     0x00040000L
 #define MMEA0_MISC2__GMI_WR_THROTTLE_MASK                                                                     0x00080000L
+//MMEA0_CE_ERR_STATUS_LO
+#define MMEA0_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA0_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA0_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA0_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA0_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
 //MMEA0_MISC_AON
 #define MMEA0_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT                                                     0x0
 #define MMEA0_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT                                                  0x2
 #define MMEA0_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK                                                       0x00000003L
 #define MMEA0_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK                                                    0x00000004L
-
+//MMEA0_CE_ERR_STATUS_HI
+#define MMEA0_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                        0x1
+#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA0_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define MMEA0_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                        0x1b
+#define MMEA0_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                          0x00000002L
+#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA0_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define MMEA0_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                          0xF8000000L
 
 // addressBlock: aid_mmhub_ea_mmeadec1
 //MMEA1_DRAM_RD_CLI2GRP_MAP0
@@ -12418,6 +12465,30 @@
 #define MMEA1_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK                                                          0x01000000L
 #define MMEA1_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK                                                           0x02000000L
 #define MMEA1_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK                                                0x04000000L
+//MMEA1_UE_ERR_STATUS_LO
+#define MMEA1_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA1_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA1_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA1_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA1_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//MMEA1_UE_ERR_STATUS_HI
+#define MMEA1_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA1_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA1_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define MMEA1_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define MMEA1_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                         0x1d
+#define MMEA1_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA1_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA1_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define MMEA1_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define MMEA1_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                           0xE0000000L
 //MMEA1_DSM_CNTL
 #define MMEA1_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                               0x0
 #define MMEA1_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                              0x2
@@ -12666,12 +12737,35 @@
 #define MMEA1_MISC2__DRAM_WR_THROTTLE_MASK                                                                    0x00020000L
 #define MMEA1_MISC2__GMI_RD_THROTTLE_MASK                                                                     0x00040000L
 #define MMEA1_MISC2__GMI_WR_THROTTLE_MASK                                                                     0x00080000L
+//MMEA1_CE_ERR_STATUS_LO
+#define MMEA1_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA1_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA1_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA1_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA1_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
 //MMEA1_MISC_AON
 #define MMEA1_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT                                                     0x0
 #define MMEA1_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT                                                  0x2
 #define MMEA1_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK                                                       0x00000003L
 #define MMEA1_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK                                                    0x00000004L
-
+//MMEA1_CE_ERR_STATUS_HI
+#define MMEA1_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                        0x1
+#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA1_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define MMEA1_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                        0x1b
+#define MMEA1_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                          0x00000002L
+#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA1_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define MMEA1_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                          0xF8000000L
 
 // addressBlock: aid_mmhub_ea_mmeadec2
 //MMEA2_DRAM_RD_CLI2GRP_MAP0
@@ -14366,6 +14460,30 @@
 #define MMEA2_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK                                                          0x01000000L
 #define MMEA2_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK                                                           0x02000000L
 #define MMEA2_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK                                                0x04000000L
+//MMEA2_UE_ERR_STATUS_LO
+#define MMEA2_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA2_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA2_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA2_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//MMEA2_UE_ERR_STATUS_HI
+#define MMEA2_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA2_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA2_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define MMEA2_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define MMEA2_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                         0x1d
+#define MMEA2_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA2_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA2_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define MMEA2_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define MMEA2_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                           0xE0000000L
 //MMEA2_DSM_CNTL
 #define MMEA2_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                               0x0
 #define MMEA2_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                              0x2
@@ -14614,12 +14732,35 @@
 #define MMEA2_MISC2__DRAM_WR_THROTTLE_MASK                                                                    0x00020000L
 #define MMEA2_MISC2__GMI_RD_THROTTLE_MASK                                                                     0x00040000L
 #define MMEA2_MISC2__GMI_WR_THROTTLE_MASK                                                                     0x00080000L
+//MMEA2_CE_ERR_STATUS_LO
+#define MMEA2_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA2_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA2_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA2_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
 //MMEA2_MISC_AON
 #define MMEA2_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT                                                     0x0
 #define MMEA2_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT                                                  0x2
 #define MMEA2_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK                                                       0x00000003L
 #define MMEA2_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK                                                    0x00000004L
-
+//MMEA2_CE_ERR_STATUS_HI
+#define MMEA2_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                        0x1
+#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA2_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define MMEA2_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                        0x1b
+#define MMEA2_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                          0x00000002L
+#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA2_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define MMEA2_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                          0xF8000000L
 
 // addressBlock: aid_mmhub_ea_mmeadec3
 //MMEA3_DRAM_RD_CLI2GRP_MAP0
@@ -16314,6 +16455,30 @@
 #define MMEA3_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK                                                          0x01000000L
 #define MMEA3_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK                                                           0x02000000L
 #define MMEA3_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK                                                0x04000000L
+//MMEA3_UE_ERR_STATUS_LO
+#define MMEA3_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA3_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA3_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA3_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA3_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//MMEA3_UE_ERR_STATUS_HI
+#define MMEA3_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA3_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA3_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define MMEA3_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define MMEA3_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                         0x1d
+#define MMEA3_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA3_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA3_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define MMEA3_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define MMEA3_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                           0xE0000000L
 //MMEA3_DSM_CNTL
 #define MMEA3_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                               0x0
 #define MMEA3_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                              0x2
@@ -16562,12 +16727,35 @@
 #define MMEA3_MISC2__DRAM_WR_THROTTLE_MASK                                                                    0x00020000L
 #define MMEA3_MISC2__GMI_RD_THROTTLE_MASK                                                                     0x00040000L
 #define MMEA3_MISC2__GMI_WR_THROTTLE_MASK                                                                     0x00080000L
+//MMEA3_CE_ERR_STATUS_LO
+#define MMEA3_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA3_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA3_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA3_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA3_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
 //MMEA3_MISC_AON
 #define MMEA3_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT                                                     0x0
 #define MMEA3_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT                                                  0x2
 #define MMEA3_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK                                                       0x00000003L
 #define MMEA3_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK                                                    0x00000004L
-
+//MMEA3_CE_ERR_STATUS_HI
+#define MMEA3_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                        0x1
+#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA3_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define MMEA3_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                        0x1b
+#define MMEA3_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                          0x00000002L
+#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA3_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define MMEA3_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                          0xF8000000L
 
 // addressBlock: aid_mmhub_ea_mmeadec4
 //MMEA4_DRAM_RD_CLI2GRP_MAP0
@@ -18262,6 +18450,30 @@
 #define MMEA4_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK                                                          0x01000000L
 #define MMEA4_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK                                                           0x02000000L
 #define MMEA4_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK                                                0x04000000L
+//MMEA4_UE_ERR_STATUS_LO
+#define MMEA4_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA4_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA4_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA4_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA4_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//MMEA4_UE_ERR_STATUS_HI
+#define MMEA4_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA4_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA4_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define MMEA4_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define MMEA4_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                         0x1d
+#define MMEA4_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA4_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA4_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define MMEA4_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define MMEA4_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                           0xE0000000L
 //MMEA4_DSM_CNTL
 #define MMEA4_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                               0x0
 #define MMEA4_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                              0x2
@@ -18510,12 +18722,35 @@
 #define MMEA4_MISC2__DRAM_WR_THROTTLE_MASK                                                                    0x00020000L
 #define MMEA4_MISC2__GMI_RD_THROTTLE_MASK                                                                     0x00040000L
 #define MMEA4_MISC2__GMI_WR_THROTTLE_MASK                                                                     0x00080000L
+//MMEA4_CE_ERR_STATUS_LO
+#define MMEA4_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                      0x0
+#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define MMEA4_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define MMEA4_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define MMEA4_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                        0x00000001L
+#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define MMEA4_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
 //MMEA4_MISC_AON
 #define MMEA4_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT                                                     0x0
 #define MMEA4_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT                                                  0x2
 #define MMEA4_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK                                                       0x00000003L
 #define MMEA4_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK                                                    0x00000004L
-
+//MMEA4_CE_ERR_STATUS_HI
+#define MMEA4_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                        0x1
+#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define MMEA4_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define MMEA4_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                        0x1b
+#define MMEA4_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                          0x00000002L
+#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define MMEA4_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define MMEA4_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                          0xF8000000L
 
 // addressBlock: aid_mmhub_pctldec0
 //PCTL0_CTRL
@@ -22311,5 +22546,83 @@
 #define L2TLB_PERFCOUNTER_HI__COUNTER_HI_MASK                                                                 0x0000FFFFL
 #define L2TLB_PERFCOUNTER_HI__COMPARE_VALUE_MASK                                                              0xFFFF0000L
 
-
+// addressBlock: aid_mmhub_mm_cane_mmcanedec
+//MM_CANE_ICG_CTRL
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_IREQ0__SHIFT                                                          0x0
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_ATRET__SHIFT                                                          0x1
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_OREQ__SHIFT                                                           0x2
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_REGISTER__SHIFT                                                       0x3
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_SDPM_RETURN__SHIFT                                                    0x4
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_IREQ0_MASK                                                            0x00000001L
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_ATRET_MASK                                                            0x00000002L
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_OREQ_MASK                                                             0x00000004L
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_REGISTER_MASK                                                         0x00000008L
+#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_SDPM_RETURN_MASK                                                      0x00000010L
+//MM_CANE_ERR_STATUS
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_STATUS__SHIFT                                                          0x0
+#define MM_CANE_ERR_STATUS__SDPM_WRRSP_STATUS__SHIFT                                                          0x4
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS__SHIFT                                                      0x8
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR__SHIFT                                                0xa
+#define MM_CANE_ERR_STATUS__SDPS_DAT_ERROR__SHIFT                                                             0xb
+#define MM_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR__SHIFT                                                      0xc
+#define MM_CANE_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT                                                         0xd
+#define MM_CANE_ERR_STATUS__BUSY_ON_ERROR__SHIFT                                                              0xe
+#define MM_CANE_ERR_STATUS__BUSY_ON_UER_ERROR__SHIFT                                                          0xf
+#define MM_CANE_ERR_STATUS__FUE_FLAG__SHIFT                                                                   0x10
+#define MM_CANE_ERR_STATUS__INTERRUPT_ON_FATAL__SHIFT                                                         0x11
+#define MM_CANE_ERR_STATUS__LEVEL_INTERRUPT__SHIFT                                                            0x12
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_STATUS_MASK                                                            0x0000000FL
+#define MM_CANE_ERR_STATUS__SDPM_WRRSP_STATUS_MASK                                                            0x000000F0L
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS_MASK                                                        0x00000300L
+#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR_MASK                                                  0x00000400L
+#define MM_CANE_ERR_STATUS__SDPS_DAT_ERROR_MASK                                                               0x00000800L
+#define MM_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR_MASK                                                        0x00001000L
+#define MM_CANE_ERR_STATUS__CLEAR_ERROR_STATUS_MASK                                                           0x00002000L
+#define MM_CANE_ERR_STATUS__BUSY_ON_ERROR_MASK                                                                0x00004000L
+#define MM_CANE_ERR_STATUS__BUSY_ON_UER_ERROR_MASK                                                            0x00008000L
+#define MM_CANE_ERR_STATUS__FUE_FLAG_MASK                                                                     0x00010000L
+#define MM_CANE_ERR_STATUS__INTERRUPT_ON_FATAL_MASK                                                           0x00020000L
+#define MM_CANE_ERR_STATUS__LEVEL_INTERRUPT_MASK                                                              0x00040000L
+//MM_CANE_UE_ERR_STATUS_LO
+#define MM_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                0x0
+#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                   0x1
+#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                              0x2
+#define MM_CANE_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                            0x18
+#define MM_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                  0x00000001L
+#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                     0x00000002L
+#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                0x00FFFFFCL
+#define MM_CANE_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                              0xFF000000L
+//MM_CANE_UE_ERR_STATUS_HI
+#define MM_CANE_UE_ERR_STATUS_HI__ECC__SHIFT                                                                  0x0
+#define MM_CANE_UE_ERR_STATUS_HI__PARITY__SHIFT                                                               0x1
+#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                  0x2
+#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                             0x3
+#define MM_CANE_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                               0x17
+#define MM_CANE_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                              0x1a
+#define MM_CANE_UE_ERR_STATUS_HI__ECC_MASK                                                                    0x00000001L
+#define MM_CANE_UE_ERR_STATUS_HI__PARITY_MASK                                                                 0x00000002L
+#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                    0x00000004L
+#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                               0x007FFFF8L
+#define MM_CANE_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                 0x03800000L
+#define MM_CANE_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                0x1C000000L
+//MM_CANE_CE_ERR_STATUS_LO
+#define MM_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                0x0
+#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                   0x1
+#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                              0x2
+#define MM_CANE_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                            0x18
+#define MM_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                  0x00000001L
+#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                     0x00000002L
+#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                0x00FFFFFCL
+#define MM_CANE_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                              0xFF000000L
+//MM_CANE_CE_ERR_STATUS_HI
+#define MM_CANE_CE_ERR_STATUS_HI__ECC__SHIFT                                                                  0x0
+#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                  0x2
+#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                             0x3
+#define MM_CANE_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                               0x17
+#define MM_CANE_CE_ERR_STATUS_HI__POISON__SHIFT                                                               0x1a
+#define MM_CANE_CE_ERR_STATUS_HI__ECC_MASK                                                                    0x00000001L
+#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                    0x00000004L
+#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                               0x007FFFF8L
+#define MM_CANE_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                 0x03800000L
+#define MM_CANE_CE_ERR_STATUS_HI__POISON_MASK                                                                 0x04000000L
 #endif
-- 
GitLab


From bc069d823bffd774294f5c3b12757a50fb726fd0 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Feb 2023 21:00:39 +0800
Subject: [PATCH 1202/1791] drm/amdgpu: Add query_ras_error_count for mmhub
 v1_8

Add query_ras_error_count callback for mmhub v1_8.
It will be used to query and log mmhub error count
and memory block.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h | 23 ++++++
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c   | 93 +++++++++++++++++++++++
 2 files changed, 116 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
index d21bb6dae56eb..1ca9d4ed8063a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
@@ -21,6 +21,29 @@
 #ifndef __AMDGPU_MMHUB_H__
 #define __AMDGPU_MMHUB_H__
 
+enum amdgpu_mmhub_ras_memory_id {
+	AMDGPU_MMHUB_WGMI_PAGEMEM = 0,
+	AMDGPU_MMHUB_RGMI_PAGEMEM = 1,
+	AMDGPU_MMHUB_WDRAM_PAGEMEM = 2,
+	AMDGPU_MMHUB_RDRAM_PAGEMEM = 3,
+	AMDGPU_MMHUB_WIO_CMDMEM = 4,
+	AMDGPU_MMHUB_RIO_CMDMEM = 5,
+	AMDGPU_MMHUB_WGMI_CMDMEM = 6,
+	AMDGPU_MMHUB_RGMI_CMDMEM = 7,
+	AMDGPU_MMHUB_WDRAM_CMDMEM = 8,
+	AMDGPU_MMHUB_RDRAM_CMDMEM = 9,
+	AMDGPU_MMHUB_MAM_DMEM0 = 10,
+	AMDGPU_MMHUB_MAM_DMEM1 = 11,
+	AMDGPU_MMHUB_MAM_DMEM2 = 12,
+	AMDGPU_MMHUB_MAM_DMEM3 = 13,
+	AMDGPU_MMHUB_WRET_TAGMEM = 19,
+	AMDGPU_MMHUB_RRET_TAGMEM = 20,
+	AMDGPU_MMHUB_WIO_DATAMEM = 21,
+	AMDGPU_MMHUB_WGMI_DATAMEM = 22,
+	AMDGPU_MMHUB_WDRAM_DATAMEM = 23,
+	AMDGPU_MMHUB_MEMORY_BLOCK_LAST,
+};
+
 struct amdgpu_mmhub_ras {
 	struct amdgpu_ras_block_object ras_block;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index a8faf66b68780..11240ca5ad83c 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -29,6 +29,7 @@
 
 #include "soc15_common.h"
 #include "soc15.h"
+#include "amdgpu_ras.h"
 
 #define regVM_L2_CNTL3_DEFAULT	0x80100007
 #define regVM_L2_CNTL4_DEFAULT	0x000000c1
@@ -579,3 +580,95 @@ const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs = {
 	.set_clockgating = mmhub_v1_8_set_clockgating,
 	.get_clockgating = mmhub_v1_8_get_clockgating,
 };
+
+static const struct amdgpu_ras_err_status_reg_entry mmhub_v1_8_ce_reg_list[] = {
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA0_CE_ERR_STATUS_LO, regMMEA0_CE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA0"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA1_CE_ERR_STATUS_LO, regMMEA1_CE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA1"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA2_CE_ERR_STATUS_LO, regMMEA2_CE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA2"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA3_CE_ERR_STATUS_LO, regMMEA3_CE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA3"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA4_CE_ERR_STATUS_LO, regMMEA4_CE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA4"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMM_CANE_CE_ERR_STATUS_LO, regMM_CANE_CE_ERR_STATUS_HI),
+	1, 0, "MM_CANE"},
+};
+
+static const struct amdgpu_ras_err_status_reg_entry mmhub_v1_8_ue_reg_list[] = {
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA0_UE_ERR_STATUS_LO, regMMEA0_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA0"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA1_UE_ERR_STATUS_LO, regMMEA1_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA1"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA2_UE_ERR_STATUS_LO, regMMEA2_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA2"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA3_UE_ERR_STATUS_LO, regMMEA3_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA3"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA4_UE_ERR_STATUS_LO, regMMEA4_UE_ERR_STATUS_HI),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA4"},
+	{AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMM_CANE_UE_ERR_STATUS_LO, regMM_CANE_UE_ERR_STATUS_HI),
+	1, 0, "MM_CANE"},
+};
+
+static const struct amdgpu_ras_memory_id_entry mmhub_v1_8_ras_memory_list[] = {
+	{AMDGPU_MMHUB_WGMI_PAGEMEM, "MMEA_WGMI_PAGEMEM"},
+	{AMDGPU_MMHUB_RGMI_PAGEMEM, "MMEA_RGMI_PAGEMEM"},
+	{AMDGPU_MMHUB_WDRAM_PAGEMEM, "MMEA_WDRAM_PAGEMEM"},
+	{AMDGPU_MMHUB_RDRAM_PAGEMEM, "MMEA_RDRAM_PAGEMEM"},
+	{AMDGPU_MMHUB_WIO_CMDMEM, "MMEA_WIO_CMDMEM"},
+	{AMDGPU_MMHUB_RIO_CMDMEM, "MMEA_RIO_CMDMEM"},
+	{AMDGPU_MMHUB_WGMI_CMDMEM, "MMEA_WGMI_CMDMEM"},
+	{AMDGPU_MMHUB_RGMI_CMDMEM, "MMEA_RGMI_CMDMEM"},
+	{AMDGPU_MMHUB_WDRAM_CMDMEM, "MMEA_WDRAM_CMDMEM"},
+	{AMDGPU_MMHUB_RDRAM_CMDMEM, "MMEA_RDRAM_CMDMEM"},
+	{AMDGPU_MMHUB_MAM_DMEM0, "MMEA_MAM_DMEM0"},
+	{AMDGPU_MMHUB_MAM_DMEM1, "MMEA_MAM_DMEM1"},
+	{AMDGPU_MMHUB_MAM_DMEM2, "MMEA_MAM_DMEM2"},
+	{AMDGPU_MMHUB_MAM_DMEM3, "MMEA_MAM_DMEM3"},
+	{AMDGPU_MMHUB_WRET_TAGMEM, "MMEA_WRET_TAGMEM"},
+	{AMDGPU_MMHUB_RRET_TAGMEM, "MMEA_RRET_TAGMEM"},
+	{AMDGPU_MMHUB_WIO_DATAMEM, "MMEA_WIO_DATAMEM"},
+	{AMDGPU_MMHUB_WGMI_DATAMEM, "MMEA_WGMI_DATAMEM"},
+	{AMDGPU_MMHUB_WDRAM_DATAMEM, "MMEA_WDRAM_DATAMEM"},
+};
+
+static void mmhub_v1_8_inst_query_ras_error_count(struct amdgpu_device *adev,
+						  uint32_t mmhub_inst,
+						  void *ras_err_status)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status;
+
+	amdgpu_ras_inst_query_ras_error_count(adev,
+					mmhub_v1_8_ce_reg_list,
+					ARRAY_SIZE(mmhub_v1_8_ce_reg_list),
+					mmhub_v1_8_ras_memory_list,
+					ARRAY_SIZE(mmhub_v1_8_ras_memory_list),
+					mmhub_inst,
+					AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE,
+					&err_data->ce_count);
+	amdgpu_ras_inst_query_ras_error_count(adev,
+					mmhub_v1_8_ue_reg_list,
+					ARRAY_SIZE(mmhub_v1_8_ue_reg_list),
+					mmhub_v1_8_ras_memory_list,
+					ARRAY_SIZE(mmhub_v1_8_ras_memory_list),
+					mmhub_inst,
+					AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+					&err_data->ue_count);
+}
+
+static void mmhub_v1_8_query_ras_error_count(struct amdgpu_device *adev,
+					     void *ras_err_status)
+{
+	uint32_t inst_mask;
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) {
+		dev_warn(adev->dev, "MMHUB RAS is not supported\n");
+		return;
+	}
+
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask)
+		mmhub_v1_8_inst_query_ras_error_count(adev, i, ras_err_status);
+}
-- 
GitLab


From a0cdb3d09332900e145fb99f52d1d571d7030183 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 22:22:06 +0800
Subject: [PATCH 1203/1791] drm/amdgpu: Add reset_ras_error_count for mmhub
 v1_8

Add reset_ras_error_count callback for mmhub
v1_8. It will be used to reset mmhub ras error
count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 28 +++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 11240ca5ad83c..f40a6a9f3c6df 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -672,3 +672,31 @@ static void mmhub_v1_8_query_ras_error_count(struct amdgpu_device *adev,
 	for_each_inst(i, inst_mask)
 		mmhub_v1_8_inst_query_ras_error_count(adev, i, ras_err_status);
 }
+
+static void mmhub_v1_8_inst_reset_ras_error_count(struct amdgpu_device *adev,
+						  uint32_t mmhub_inst)
+{
+	amdgpu_ras_inst_reset_ras_error_count(adev,
+					mmhub_v1_8_ce_reg_list,
+					ARRAY_SIZE(mmhub_v1_8_ce_reg_list),
+					mmhub_inst);
+	amdgpu_ras_inst_reset_ras_error_count(adev,
+					mmhub_v1_8_ue_reg_list,
+					ARRAY_SIZE(mmhub_v1_8_ue_reg_list),
+					mmhub_inst);
+}
+
+static void mmhub_v1_8_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	uint32_t inst_mask;
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) {
+		dev_warn(adev->dev, "MMHUB RAS is not supported\n");
+		return;
+	}
+
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask)
+		mmhub_v1_8_inst_reset_ras_error_count(adev, i);
+}
-- 
GitLab


From 00c145222b6ac93ee9926a8ef2e41dae55b97b7e Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 23:36:25 +0800
Subject: [PATCH 1204/1791] drm/amdgpu: Add query_ras_error_status for mmhub
 v1_8

Add query_ras_error_status callback for mmhub
v1_8. It will be used to log mmhub error status.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 56 +++++++++++++++++++++++++
 1 file changed, 56 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index f40a6a9f3c6df..2923304e653a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -700,3 +700,59 @@ static void mmhub_v1_8_reset_ras_error_count(struct amdgpu_device *adev)
 	for_each_inst(i, inst_mask)
 		mmhub_v1_8_inst_reset_ras_error_count(adev, i);
 }
+
+static const uint32_t mmhub_v1_8_mmea_err_status_reg[] = {
+	regMMEA0_ERR_STATUS,
+	regMMEA1_ERR_STATUS,
+	regMMEA2_ERR_STATUS,
+	regMMEA3_ERR_STATUS,
+	regMMEA4_ERR_STATUS,
+};
+
+static void mmhub_v1_8_inst_query_ras_err_status(struct amdgpu_device *adev,
+						 uint32_t mmhub_inst)
+{
+	uint32_t reg_value;
+	uint32_t mmea_err_status_addr_dist;
+	uint32_t i;
+
+	/* query mmea ras err status */
+	mmea_err_status_addr_dist = regMMEA1_ERR_STATUS - regMMEA0_ERR_STATUS;
+	for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i++) {
+		reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+						regMMEA0_ERR_STATUS,
+						i * mmea_err_status_addr_dist);
+		if (REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_RDRSP_STATUS) ||
+		    REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_WRRSP_STATUS) ||
+		    REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_RDRSP_DATAPARITY_ERROR)) {
+			dev_warn(adev->dev,
+				 "Detected MMEA%d err in MMHUB%d, status: 0x%x\n",
+				 i, mmhub_inst, reg_value);
+		}
+	}
+
+	/* query mm_cane ras err status */
+	reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS);
+	if (REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_RDRSP_STATUS) ||
+	    REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_WRRSP_STATUS) ||
+	    REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_RDRSP_DATAPARITY_ERROR)) {
+		dev_warn(adev->dev,
+			 "Detected MM CANE err in MMHUB%d, status: 0x%x\n",
+			 mmhub_inst, reg_value);
+	}
+}
+
+static void mmhub_v1_8_query_ras_error_status(struct amdgpu_device *adev)
+{
+	uint32_t inst_mask;
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) {
+		dev_warn(adev->dev, "MMHUB RAS is not supported\n");
+		return;
+	}
+
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask)
+		mmhub_v1_8_inst_query_ras_err_status(adev, i);
+}
-- 
GitLab


From ccfdbd4bdc0875ee8b8fe00691a6b5f883227bdb Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 23:20:09 +0800
Subject: [PATCH 1205/1791] drm/amdgpu: Add reset_ras_error_status for mmhub
 v1_8

Add reset_ras_error_status callback for mmhub
v1_8. It will be used to reset mmhub error status.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 91 +++++++++++++++++++++++++
 1 file changed, 91 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 2923304e653a0..4f274c7db591c 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -756,3 +756,94 @@ static void mmhub_v1_8_query_ras_error_status(struct amdgpu_device *adev)
 	for_each_inst(i, inst_mask)
 		mmhub_v1_8_inst_query_ras_err_status(adev, i);
 }
+
+static const uint32_t mmhub_v1_8_mmea_cgtt_clk_cntl_reg[] = {
+	regMMEA0_CGTT_CLK_CTRL,
+	regMMEA1_CGTT_CLK_CTRL,
+	regMMEA2_CGTT_CLK_CTRL,
+	regMMEA3_CGTT_CLK_CTRL,
+	regMMEA4_CGTT_CLK_CTRL,
+};
+
+static void mmhub_v1_8_inst_reset_ras_err_status(struct amdgpu_device *adev,
+						 uint32_t mmhub_inst)
+{
+	uint32_t mmea_cgtt_clk_cntl_addr_dist;
+	uint32_t mmea_err_status_addr_dist;
+	uint32_t reg_value;
+	uint32_t i;
+
+	/* reset mmea ras err status */
+	mmea_cgtt_clk_cntl_addr_dist = regMMEA1_CGTT_CLK_CTRL - regMMEA0_CGTT_CLK_CTRL;
+	mmea_err_status_addr_dist = regMMEA1_ERR_STATUS - regMMEA0_ERR_STATUS;
+	for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i ++) {
+		/* force clk branch on for response path
+		 * set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 1 */
+		reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+						regMMEA0_CGTT_CLK_CTRL,
+						i * mmea_cgtt_clk_cntl_addr_dist);
+		reg_value = REG_SET_FIELD(reg_value, MMEA0_CGTT_CLK_CTRL,
+					  SOFT_OVERRIDE_RETURN, 1);
+		WREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+				    regMMEA0_CGTT_CLK_CTRL,
+				    i * mmea_cgtt_clk_cntl_addr_dist,
+				    reg_value);
+
+		/* set MMEA0_ERR_STATUS.CLEAR_ERROR_STATUS = 1 */
+		reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+						regMMEA0_ERR_STATUS,
+						i * mmea_err_status_addr_dist);
+		reg_value = REG_SET_FIELD(reg_value, MMEA0_ERR_STATUS,
+					  CLEAR_ERROR_STATUS, 1);
+		WREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+				    regMMEA0_ERR_STATUS,
+				    i * mmea_err_status_addr_dist,
+				    reg_value);
+
+		/* set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 0 */
+		reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+						regMMEA0_CGTT_CLK_CTRL,
+						i * mmea_cgtt_clk_cntl_addr_dist);
+		reg_value = REG_SET_FIELD(reg_value, MMEA0_CGTT_CLK_CTRL,
+					  SOFT_OVERRIDE_RETURN, 0);
+		WREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
+				    regMMEA0_CGTT_CLK_CTRL,
+				    i * mmea_cgtt_clk_cntl_addr_dist,
+				    reg_value);
+	}
+
+	/* reset mm_cane ras err status
+	 * force clk branch on for response path
+	 * set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 1 */
+	reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL);
+	reg_value = REG_SET_FIELD(reg_value, MM_CANE_ICG_CTRL,
+				  SOFT_OVERRIDE_ATRET, 1);
+	WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL, reg_value);
+
+	/* set MM_CANE_ERR_STATUS.CLEAR_ERROR_STATUS = 1 */
+	reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS);
+	reg_value = REG_SET_FIELD(reg_value, MM_CANE_ERR_STATUS,
+				  CLEAR_ERROR_STATUS, 1);
+	WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS, reg_value);
+
+	/* set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 0 */
+	reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL);
+	reg_value = REG_SET_FIELD(reg_value, MM_CANE_ICG_CTRL,
+				  SOFT_OVERRIDE_ATRET, 0);
+	WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL, reg_value);
+}
+
+static void mmhub_v1_8_reset_ras_error_status(struct amdgpu_device *adev)
+{
+	uint32_t inst_mask;
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) {
+		dev_warn(adev->dev, "MMHUB RAS is not supported\n");
+		return;
+	}
+
+	inst_mask = adev->aid_mask;
+	for_each_inst(i, inst_mask)
+		mmhub_v1_8_inst_reset_ras_err_status(adev, i);
+}
-- 
GitLab


From 73c2b3fd2c515bcb819d801c5c4bf053fdb1e5cb Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 22 Jan 2023 23:26:40 +0800
Subject: [PATCH 1206/1791] drm/amdgpu: Initialize mmhub v1_8 ras function

Initialize mmhub v1_8 ras function.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   |  3 +++
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h |  1 +
 3 files changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index b6c500be6f703..16634a791e10c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1419,6 +1419,9 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev)
 	case IP_VERSION(9, 4, 2):
 		adev->mmhub.ras = &mmhub_v1_7_ras;
 		break;
+	case IP_VERSION(1, 8, 0):
+		adev->mmhub.ras = &mmhub_v1_8_ras;
+		break;
 	default:
 		/* mmhub ras is not available */
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 4f274c7db591c..3648994724c26 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -847,3 +847,16 @@ static void mmhub_v1_8_reset_ras_error_status(struct amdgpu_device *adev)
 	for_each_inst(i, inst_mask)
 		mmhub_v1_8_inst_reset_ras_err_status(adev, i);
 }
+
+static const struct amdgpu_ras_block_hw_ops mmhub_v1_8_ras_hw_ops = {
+	.query_ras_error_count = mmhub_v1_8_query_ras_error_count,
+	.reset_ras_error_count = mmhub_v1_8_reset_ras_error_count,
+	.query_ras_error_status = mmhub_v1_8_query_ras_error_status,
+	.reset_ras_error_status = mmhub_v1_8_reset_ras_error_status,
+};
+
+struct amdgpu_mmhub_ras mmhub_v1_8_ras = {
+	.ras_block = {
+		.hw_ops = &mmhub_v1_8_ras_hw_ops,
+	},
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h
index 0bb36200e4e5b..126f0075ac508 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h
@@ -24,5 +24,6 @@
 #define __MMHUB_V1_8_H__
 
 extern const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs;
+extern struct amdgpu_mmhub_ras mmhub_v1_8_ras;
 
 #endif
-- 
GitLab


From 8107e4996f4ec3304485c608b5880185840c464c Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Sun, 29 Jan 2023 22:48:15 +0800
Subject: [PATCH 1207/1791] drm/amdgpu: Enable persistent edc harvesting in APP
 APU

Persistent edc harvesting is supported in APP APU

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 8a16a06cb78a1..5ae89602a1161 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2624,7 +2624,8 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 
 int amdgpu_persistent_edc_harvesting_supported(struct amdgpu_device *adev)
 {
-	if (adev->gmc.xgmi.connected_to_cpu)
+	if (adev->gmc.xgmi.connected_to_cpu ||
+	    adev->gmc.is_app_apu)
 		return 1;
 	return 0;
 }
-- 
GitLab


From 9faf929fbf6b457d13064d57017b2d4f62670b58 Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Tue, 7 Feb 2023 20:55:24 +0800
Subject: [PATCH 1208/1791] drm/amdgpu: Enable CG for IH v4.4.2

Enable clock gating on IH v4.4.2 versions.

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 100fdf5074b43..9fbfd0811d067 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1147,7 +1147,8 @@ static int soc15_common_early_init(void *handle)
 		adev->cg_flags =
 			AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGCG |
 			AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_SDMA_MGCG |
-			AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG;
+			AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_VCN_MGCG |
+			AMD_CG_SUPPORT_JPEG_MGCG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_VCN_DPG |
-- 
GitLab


From 228ce176434b0f61451019065393040d58e1668d Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Fri, 27 Jan 2023 21:57:00 -0500
Subject: [PATCH 1209/1791] drm/amdgpu: Handle VRAM dependencies on GFXIP9.4.3

[For 1P NPS1 mode driver bringup]

Changes required to initialize the amdgpu driver with frontdoor firmware
loading and discovery=2 with the native mode SBIOS that enables CPU GPU
unified interleaved memory.

sudo modprobe amdgpu discovery=2

Once PSP TMR region is reported via the ACPI interface, the dependency
on the ip_discovery.bin will be removed.

Choice of where to allocate driver table is given to each IP version. In
general, both GTT and VRAM domains will be considered. If one of the
tables has a strict restriction for VRAM domain, then only VRAM domain
is considered.

Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
(lijo: Modified the handling for SMU Tables)
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  |  3 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c    |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c       |  6 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c       | 89 +++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c     |  7 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c       |  3 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c         |  9 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  6 ++
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c         |  5 ++
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 10 ++-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |  6 +-
 11 files changed, 99 insertions(+), 47 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index af37f2ef4438b..4e179e50de25c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -2292,8 +2292,9 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev,
 	(*mem)->dmabuf = dma_buf;
 	(*mem)->bo = bo;
 	(*mem)->va = va;
-	(*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ?
+	(*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) && !adev->gmc.is_app_apu ?
 		AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT;
+
 	(*mem)->mapped_to_gpu_memory = 0;
 	(*mem)->process_info = avm->process_info;
 	add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, false);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 9f0d5f02119ed..f431205e1077f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1044,7 +1044,7 @@ static const char * const amdgpu_vram_names[] = {
 int amdgpu_bo_init(struct amdgpu_device *adev)
 {
 	/* On A+A platform, VRAM can be mapped as WB */
-	if (!adev->gmc.xgmi.connected_to_cpu) {
+	if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) {
 		/* reserve PAT memory space to WC for VRAM */
 		int r = arch_io_reserve_memtype_wc(adev->gmc.aper_base,
 				adev->gmc.aper_size);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 863fa331e6ff3..4395c53d09d83 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -476,7 +476,8 @@ static int psp_sw_init(void *handle)
 		return ret;
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_FENCE_BUFFER_SIZE, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &psp->fence_buf_bo,
 				      &psp->fence_buf_mc_addr,
 				      &psp->fence_buf);
@@ -484,7 +485,8 @@ static int psp_sw_init(void *handle)
 		goto failed1;
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_CMD_BUFFER_SIZE, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &psp->cmd_buf_bo, &psp->cmd_buf_mc_addr,
 				      (void **)&psp->cmd_buf_mem);
 	if (ret)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 6bbe3b89aef5d..bc11ae56bba5b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1708,15 +1708,20 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 		ctx->init = PSP_MEM_TRAIN_RESERVE_SUCCESS;
 	}
 
-	ret = amdgpu_bo_create_kernel_at(adev,
-					 adev->gmc.real_vram_size - adev->mman.discovery_tmr_size,
-					 adev->mman.discovery_tmr_size,
-					 &adev->mman.discovery_memory,
-					 NULL);
-	if (ret) {
-		DRM_ERROR("alloc tmr failed(%d)!\n", ret);
-		amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
-		return ret;
+	if (!adev->gmc.is_app_apu) {
+		ret = amdgpu_bo_create_kernel_at(adev,
+						 adev->gmc.real_vram_size -
+						 adev->mman.discovery_tmr_size,
+						 adev->mman.discovery_tmr_size,
+						 &adev->mman.discovery_memory,
+						 NULL);
+		if (ret) {
+			DRM_ERROR("alloc tmr failed(%d)!\n", ret);
+			amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
+			return ret;
+		}
+	} else {
+		DRM_DEBUG_DRIVER("backdoor fw loading path for PSP TMR, no reservation needed\n");
 	}
 
 	return 0;
@@ -1765,10 +1770,12 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		adev->mman.aper_base_kaddr = ioremap_cache(adev->gmc.aper_base,
 				adev->gmc.visible_vram_size);
 
-	else
+	else if (!adev->gmc.is_app_apu)
 #endif
 		adev->mman.aper_base_kaddr = ioremap_wc(adev->gmc.aper_base,
 				adev->gmc.visible_vram_size);
+	else
+		DRM_DEBUG_DRIVER("No need to ioremap when real vram size is 0\n");
 #endif
 
 	/*
@@ -1803,23 +1810,32 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	 * This is used for VGA emulation and pre-OS scanout buffers to
 	 * avoid display artifacts while transitioning between pre-OS
 	 * and driver.  */
-	r = amdgpu_bo_create_kernel_at(adev, 0, adev->mman.stolen_vga_size,
-				       &adev->mman.stolen_vga_memory,
-				       NULL);
-	if (r)
-		return r;
-	r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_vga_size,
-				       adev->mman.stolen_extended_size,
-				       &adev->mman.stolen_extended_memory,
-				       NULL);
-	if (r)
-		return r;
-	r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_reserved_offset,
-				       adev->mman.stolen_reserved_size,
-				       &adev->mman.stolen_reserved_memory,
-				       NULL);
-	if (r)
-		return r;
+	if (!adev->gmc.is_app_apu) {
+		r = amdgpu_bo_create_kernel_at(adev, 0,
+					       adev->mman.stolen_vga_size,
+					       &adev->mman.stolen_vga_memory,
+					       NULL);
+		if (r)
+			return r;
+
+		r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_vga_size,
+					       adev->mman.stolen_extended_size,
+					       &adev->mman.stolen_extended_memory,
+					       NULL);
+
+		if (r)
+			return r;
+
+		r = amdgpu_bo_create_kernel_at(adev,
+					       adev->mman.stolen_reserved_offset,
+					       adev->mman.stolen_reserved_size,
+					       &adev->mman.stolen_reserved_memory,
+					       NULL);
+		if (r)
+			return r;
+	} else {
+		DRM_DEBUG_DRIVER("Skipped stolen memory reservation\n");
+	}
 
 	DRM_INFO("amdgpu: %uM of VRAM memory ready\n",
 		 (unsigned) (adev->gmc.real_vram_size / (1024 * 1024)));
@@ -1866,7 +1882,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		DRM_ERROR("Failed initializing oa heap.\n");
 		return r;
 	}
-
 	if (amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE,
 				AMDGPU_GEM_DOMAIN_GTT,
 				&adev->mman.sdma_access_bo, NULL,
@@ -1887,13 +1902,15 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 
 	amdgpu_ttm_training_reserve_vram_fini(adev);
 	/* return the stolen vga memory back to VRAM */
-	amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
-	amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
-	/* return the IP Discovery TMR memory back to VRAM */
-	amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
-	if (adev->mman.stolen_reserved_size)
-		amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory,
-				      NULL, NULL);
+	if (!adev->gmc.is_app_apu) {
+		amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
+		amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
+		/* return the IP Discovery TMR memory back to VRAM */
+		amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
+		if (adev->mman.stolen_reserved_size)
+			amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory,
+					      NULL, NULL);
+	}
 	amdgpu_bo_free_kernel(&adev->mman.sdma_access_bo, NULL,
 					&adev->mman.sdma_access_ptr);
 	amdgpu_ttm_fw_reserve_vram_fini(adev);
@@ -1935,7 +1952,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 	int r;
 
 	if (!adev->mman.initialized || amdgpu_in_reset(adev) ||
-	    adev->mman.buffer_funcs_enabled == enable)
+	    adev->mman.buffer_funcs_enabled == enable || adev->gmc.is_app_apu)
 		return;
 
 	if (enable) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index df63dc3bca18c..bc5d126b600b4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -512,7 +512,12 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	bp.size = amdgpu_vm_pt_size(adev, level);
 	bp.byte_align = AMDGPU_GPU_PAGE_SIZE;
-	bp.domain = AMDGPU_GEM_DOMAIN_VRAM;
+
+	if (!adev->gmc.is_app_apu)
+		bp.domain = AMDGPU_GEM_DOMAIN_VRAM;
+	else
+		bp.domain = AMDGPU_GEM_DOMAIN_GTT;
+
 	bp.domain = amdgpu_bo_get_preferred_domain(adev, bp.domain);
 	bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
 		AMDGPU_GEM_CREATE_CPU_GTT_USWC;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 1f1268cd5e09f..42877c4505f1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -459,7 +459,8 @@ static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev)
 		adev->gfx.num_compute_rings * num_xcc * GFX9_MEC_HPD_SIZE;
 	if (mec_hpd_size) {
 		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-					      AMDGPU_GEM_DOMAIN_VRAM,
+					      AMDGPU_GEM_DOMAIN_VRAM |
+					      AMDGPU_GEM_DOMAIN_GTT,
 					      &adev->gfx.mec.hpd_eop_obj,
 					      &adev->gfx.mec.hpd_eop_gpu_addr,
 					      (void **)&hpd);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 16634a791e10c..245de27c75402 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1593,8 +1593,13 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev)
 	int r;
 
 	/* size in MB on si */
-	adev->gmc.mc_vram_size =
-		adev->nbio.funcs->get_memsize(adev) * 1024ULL * 1024ULL;
+	if (!adev->gmc.is_app_apu) {
+		adev->gmc.mc_vram_size =
+			adev->nbio.funcs->get_memsize(adev) * 1024ULL * 1024ULL;
+	} else {
+		DRM_DEBUG("Set mc_vram_size = 0 for APP APU\n");
+		adev->gmc.mc_vram_size = 0;
+	}
 	adev->gmc.real_vram_size = adev->gmc.mc_vram_size;
 
 	if (!(adev->flags & AMD_IS_APU) &&
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 8b9accecf49b8..f85ac4dbc6737 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1026,6 +1026,12 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev)
 	if (dev->kfd->local_mem_info.local_mem_size_private == 0 &&
 	    dev->kfd->local_mem_info.local_mem_size_public > 0)
 		return true;
+
+	if (dev->kfd->local_mem_info.local_mem_size_public == 0 && dev->kfd->adev->gmc.is_app_apu) {
+		pr_debug("APP APU, Consider like a large bar system\n");
+		return true;
+	}
+
 	return false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 16475921587bf..1aaf933f9f488 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -30,6 +30,9 @@
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 
+/* Fixme: Fake 32GB for 1PNPS1 mode bringup */
+#define DUMMY_VRAM_SIZE 31138512896
+
 /* GPU Processor ID base for dGPUs for which VCRAT needs to be created.
  * GPU processor ID are expressed with Bit[31]=1.
  * The base is set to 0x8000_0000 + 0x1000 to avoid collision with GPU IDs
@@ -1053,6 +1056,8 @@ static int kfd_parse_subtype_mem(struct crat_subtype_memory *mem,
 
 			props->heap_type = heap_type;
 			props->flags = flags;
+			if (size_in_bytes == 0)
+				size_in_bytes = DUMMY_VRAM_SIZE; /* Fixme: TBD */
 			props->size_in_bytes = size_in_bytes;
 			props->width = width;
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 2ddf5198e5c48..4dea79a0c5b5e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -822,11 +822,20 @@ static int smu_init_fb_allocations(struct smu_context *smu)
 		}
 	}
 
+	driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT;
 	/* VRAM allocation for driver table */
 	for (i = 0; i < SMU_TABLE_COUNT; i++) {
 		if (tables[i].size == 0)
 			continue;
 
+		/* If one of the tables has VRAM domain restriction, keep it in
+		 * VRAM
+		 */
+		if ((tables[i].domain &
+		    (AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT)) ==
+			    AMDGPU_GEM_DOMAIN_VRAM)
+			driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM;
+
 		if (i == SMU_TABLE_PMSTATUSLOG)
 			continue;
 
@@ -836,7 +845,6 @@ static int smu_init_fb_allocations(struct smu_context *smu)
 
 	driver_table->size = max_table_size;
 	driver_table->align = PAGE_SIZE;
-	driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM;
 
 	ret = amdgpu_bo_create_kernel(adev,
 				      driver_table->size,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index ea8f3d6fb98b3..8969b3ff5c8fc 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -220,10 +220,12 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 			       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 
 	SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(MetricsTable_t),
-		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+		       PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT);
 
 	SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t),
-		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+		       PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT);
 
 	smu_table->metrics_table = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
-- 
GitLab


From f431393d605f55f8865dbf8ba8236760fbb0a3dc Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Fri, 27 Jan 2023 21:46:59 -0500
Subject: [PATCH 1210/1791] drm/amdgpu: Implement new dummy vram manager

This adds dummy vram manager to support ASICs that do not have a
dedicated or carvedout vram domain.

Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 67 ++++++++++++++++++--
 1 file changed, 60 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 43d6a9d6a5384..89d35d194f2c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -370,6 +370,45 @@ int amdgpu_vram_mgr_query_page_status(struct amdgpu_vram_mgr *mgr,
 	return ret;
 }
 
+static void amdgpu_dummy_vram_mgr_debug(struct ttm_resource_manager *man,
+				  struct drm_printer *printer)
+{
+	DRM_DEBUG_DRIVER("Dummy vram mgr debug\n");
+}
+
+static bool amdgpu_dummy_vram_mgr_compatible(struct ttm_resource_manager *man,
+				       struct ttm_resource *res,
+				       const struct ttm_place *place,
+				       size_t size)
+{
+	DRM_DEBUG_DRIVER("Dummy vram mgr compatible\n");
+	return false;
+}
+
+static bool amdgpu_dummy_vram_mgr_intersects(struct ttm_resource_manager *man,
+				       struct ttm_resource *res,
+				       const struct ttm_place *place,
+				       size_t size)
+{
+	DRM_DEBUG_DRIVER("Dummy vram mgr intersects\n");
+	return true;
+}
+
+static void amdgpu_dummy_vram_mgr_del(struct ttm_resource_manager *man,
+				struct ttm_resource *res)
+{
+	DRM_DEBUG_DRIVER("Dummy vram mgr deleted\n");
+}
+
+static int amdgpu_dummy_vram_mgr_new(struct ttm_resource_manager *man,
+			       struct ttm_buffer_object *tbo,
+			       const struct ttm_place *place,
+			       struct ttm_resource **res)
+{
+	DRM_DEBUG_DRIVER("Dummy vram mgr new\n");
+	return -ENOSPC;
+}
+
 /**
  * amdgpu_vram_mgr_new - allocate new ranges
  *
@@ -817,6 +856,14 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man,
 	mutex_unlock(&mgr->lock);
 }
 
+static const struct ttm_resource_manager_func amdgpu_dummy_vram_mgr_func = {
+	.alloc	= amdgpu_dummy_vram_mgr_new,
+	.free	= amdgpu_dummy_vram_mgr_del,
+	.intersects = amdgpu_dummy_vram_mgr_intersects,
+	.compatible = amdgpu_dummy_vram_mgr_compatible,
+	.debug	= amdgpu_dummy_vram_mgr_debug
+};
+
 static const struct ttm_resource_manager_func amdgpu_vram_mgr_func = {
 	.alloc	= amdgpu_vram_mgr_new,
 	.free	= amdgpu_vram_mgr_del,
@@ -841,17 +888,22 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev)
 	ttm_resource_manager_init(man, &adev->mman.bdev,
 				  adev->gmc.real_vram_size);
 
-	man->func = &amdgpu_vram_mgr_func;
-
-	err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE);
-	if (err)
-		return err;
-
 	mutex_init(&mgr->lock);
 	INIT_LIST_HEAD(&mgr->reservations_pending);
 	INIT_LIST_HEAD(&mgr->reserved_pages);
 	mgr->default_page_size = PAGE_SIZE;
 
+	if (!adev->gmc.is_app_apu) {
+		man->func = &amdgpu_vram_mgr_func;
+
+		err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE);
+		if (err)
+			return err;
+	} else {
+		man->func = &amdgpu_dummy_vram_mgr_func;
+		DRM_INFO("Setup dummy vram mgr\n");
+	}
+
 	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_VRAM, &mgr->manager);
 	ttm_resource_manager_set_used(man, true);
 	return 0;
@@ -886,7 +938,8 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev)
 		drm_buddy_free_list(&mgr->mm, &rsv->allocated);
 		kfree(rsv);
 	}
-	drm_buddy_fini(&mgr->mm);
+	if (!adev->gmc.is_app_apu)
+		drm_buddy_fini(&mgr->mm);
 	mutex_unlock(&mgr->lock);
 
 	ttm_resource_manager_cleanup(man);
-- 
GitLab


From 970c1646b5ac93a13496d3429aca3e799fa6cf07 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Fri, 27 Jan 2023 21:48:06 -0500
Subject: [PATCH 1211/1791] drm/amdgpu: Create VRAM BOs on GTT for GFXIP9.4.3

On GFXIP9.4.3 APP APU where there is no dedicated VRAM domain handle
VRAM BO allocation requests on CPU domain and validate them on GTT.

Support for handling multi-socket and multi-numa partitions within a
socket will be added by future patches, this enables 1P NPS1 asic
bringup configuration.

Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c   | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 4e179e50de25c..bbdd5e3aa18e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1649,9 +1649,16 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	 */
 	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) {
 		domain = alloc_domain = AMDGPU_GEM_DOMAIN_VRAM;
-		alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE;
-		alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ?
+
+		if (adev->gmc.is_app_apu) {
+			domain = AMDGPU_GEM_DOMAIN_GTT;
+			alloc_domain = AMDGPU_GEM_DOMAIN_CPU;
+			alloc_flags = 0;
+		} else {
+			alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE;
+			alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ?
 			AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : 0;
+		}
 	} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) {
 		domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT;
 		alloc_flags = 0;
@@ -1738,6 +1745,13 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	(*mem)->domain = domain;
 	(*mem)->mapped_to_gpu_memory = 0;
 	(*mem)->process_info = avm->process_info;
+
+	if (adev->gmc.is_app_apu &&
+	    ((*mem)->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)) {
+		bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
+		bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
+	}
+
 	add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr);
 
 	if (user_addr) {
-- 
GitLab


From bb0ed57b4450e46de3651b310dcb4273032c3122 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 16 Mar 2023 11:08:06 +0800
Subject: [PATCH 1212/1791] drm/amdgpu: increase AMDGPU_MAX_RINGS

On newer GPUs, the number of kernel rings are increased.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index add7cc2831b03..d2b1a88546038 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -37,7 +37,7 @@ struct amdgpu_job;
 struct amdgpu_vm;
 
 /* max number of rings */
-#define AMDGPU_MAX_RINGS		102
+#define AMDGPU_MAX_RINGS		124
 #define AMDGPU_MAX_HWIP_RINGS		64
 #define AMDGPU_MAX_GFX_RINGS		2
 #define AMDGPU_MAX_SW_GFX_RINGS         2
-- 
GitLab


From 9e4216cf2dcccacd0e5dc932f26e35d18527ff41 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Fri, 5 May 2023 11:54:38 -0400
Subject: [PATCH 1213/1791] drm/amdgpu: Increase Max GPU instance to 64

Increase Max GPU instances to 64 to handle multi-socket
system with GFX 9.4.3 asic.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 279057ec7a0b2..39743d44b567b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -108,7 +108,7 @@
 #include "amdgpu_mca.h"
 #include "amdgpu_ras.h"
 
-#define MAX_GPU_INSTANCE		16
+#define MAX_GPU_INSTANCE		64
 
 struct amdgpu_gpu_instance
 {
-- 
GitLab


From 0c451baf3bff8e2a9fa45ef6471f9f25da00f39b Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Thu, 16 Mar 2023 17:42:49 +0800
Subject: [PATCH 1214/1791] drm/amdgpu: change the print level to warn for ip
 block disabled

Avoid to mislead users as it's not a real error.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index c190365d67e27..7f62826fcaca9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2292,7 +2292,7 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 	total = true;
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
-			DRM_ERROR("disabled ip block: %d <%s>\n",
+			DRM_WARN("disabled ip block: %d <%s>\n",
 				  i, adev->ip_blocks[i].version->funcs->name);
 			adev->ip_blocks[i].status.valid = false;
 		} else {
-- 
GitLab


From 1ad29cb3438175a9517f3b6cfeb0e331be213d8c Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Wed, 22 Mar 2023 11:16:53 +0800
Subject: [PATCH 1215/1791] drm/amdgpu: fix sdma instance

It should change logical instance to device instance
to query ras info

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 570ea68c521fd..bf47eb33c12e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -2123,6 +2123,7 @@ static void sdma_v4_4_2_inst_query_ras_error_count(struct amdgpu_device *adev,
 						   void *ras_err_status)
 {
 	struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status;
+	uint32_t sdma_dev_inst = GET_INST(SDMA0, sdma_inst);
 
 	/* sdma v4_4_2 doesn't support query ce counts */
 	amdgpu_ras_inst_query_ras_error_count(adev,
@@ -2130,7 +2131,7 @@ static void sdma_v4_4_2_inst_query_ras_error_count(struct amdgpu_device *adev,
 					ARRAY_SIZE(sdma_v4_2_2_ue_reg_list),
 					sdma_v4_4_2_ras_memory_list,
 					ARRAY_SIZE(sdma_v4_4_2_ras_memory_list),
-					sdma_inst,
+					sdma_dev_inst,
 					AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
 					&err_data->ue_count);
 }
@@ -2153,10 +2154,12 @@ static void sdma_v4_4_2_query_ras_error_count(struct amdgpu_device *adev,
 static void sdma_v4_4_2_inst_reset_ras_error_count(struct amdgpu_device *adev,
 						   uint32_t sdma_inst)
 {
+	uint32_t sdma_dev_inst = GET_INST(SDMA0, sdma_inst);
+
 	amdgpu_ras_inst_reset_ras_error_count(adev,
 					sdma_v4_2_2_ue_reg_list,
 					ARRAY_SIZE(sdma_v4_2_2_ue_reg_list),
-					sdma_inst);
+					sdma_dev_inst);
 }
 
 static void sdma_v4_4_2_reset_ras_error_count(struct amdgpu_device *adev)
-- 
GitLab


From 8d1c1bc13134ab90d773cb73c0298f2459703bee Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Wed, 8 Feb 2023 20:19:13 +0800
Subject: [PATCH 1216/1791] drm/amd/pm: Update pmfw header files for SMU
 v13.0.6

Update driver metrics table for SMU v13.0.6 to be
compatible with PMFW v85.47 version

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h  | 4 ++--
 drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h | 6 +++++-
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
index be596777cd2ca..370c6125d7189 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
@@ -26,7 +26,7 @@
 // *** IMPORTANT ***
 // PMFW TEAM: Always increment the interface version if
 // anything is changed in this file
-#define SMU13_0_6_DRIVER_IF_VERSION 0x08042022
+#define SMU13_0_6_DRIVER_IF_VERSION 0x08042023
 
 //I2C Interface
 #define NUM_I2C_CONTROLLERS                8
@@ -106,7 +106,7 @@ typedef enum {
 } UCLK_DPM_MODE_e;
 
 typedef struct {
-  //0-26 SOC, 27-29 SOCIO
+  //0-23 SOC, 24-26 SOCIO, 27-29 SOC
   uint16_t avgPsmCount[30];
   uint16_t minPsmCount[30];
   float    avgPsmVoltage[30];
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
index bdccbb4a62763..3fe403615d86b 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
@@ -123,7 +123,7 @@ typedef enum {
   VOLTAGE_GUARDBAND_COUNT
 } GFX_GUARDBAND_e;
 
-#define SMU_METRICS_TABLE_VERSION 0x1
+#define SMU_METRICS_TABLE_VERSION 0x3
 
 typedef struct {
   uint32_t AccumulationCounter;
@@ -198,6 +198,10 @@ typedef struct {
   uint32_t SocketThmResidencyAcc;
   uint32_t VrThmResidencyAcc;
   uint32_t HbmThmResidencyAcc;
+  uint32_t spare;
+
+  // New Items at end to maintain driver compatibility
+  uint32_t GfxclkFrequency[8];
 } MetricsTable_t;
 
 #define SMU_VF_METRICS_TABLE_VERSION 0x1
-- 
GitLab


From a1b0dafafca414cf8b3a51225215a21df2b3ddf8 Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Wed, 8 Feb 2023 23:04:25 +0800
Subject: [PATCH 1217/1791] drm/amd/pm: Update gfx clock frequency for SMU
 v13.0.6

Update gfx clock frequency from metric table for SMU v13.0.6

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 8969b3ff5c8fc..d0c49e8883e7a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -660,7 +660,10 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu,
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
 	MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table;
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t smu_version;
 	int ret = 0;
+	int xcc_id;
 
 	ret = smu_v13_0_6_get_metrics_table(smu, NULL, false);
 	if (ret)
@@ -670,7 +673,13 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu,
 	switch (member) {
 	case METRICS_CURR_GFXCLK:
 	case METRICS_AVERAGE_GFXCLK:
-		*value = 0;
+		smu_cmn_get_smc_version(smu, NULL, &smu_version);
+		if (smu_version >= 0x552F00) {
+			xcc_id = GET_INST(GC, 0);
+			*value = SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc_id]);
+		} else {
+			*value = 0;
+		}
 		break;
 	case METRICS_CURR_SOCCLK:
 	case METRICS_AVERAGE_SOCCLK:
-- 
GitLab


From 6d5f5eaf6acf26ce0dc986fe7240dc4a0c981119 Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Mon, 13 Feb 2023 19:52:56 +0800
Subject: [PATCH 1218/1791] drm/amd/pm: Update pmfw header files for SMU
 v13.0.6

Update driver interface for SMU v13.0.6 to be
compatible with PMFW v85.48 version

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h   | 12 ++++++++++++
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c |  1 -
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
index 370c6125d7189..8b7fa0fa59c3c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
@@ -121,6 +121,18 @@ typedef struct {
   float    minPsmVoltage[30];
 } AvfsDebugTableXcd_t;
 
+// Defines used for IH-based thermal interrupts to GFX driver - A/X only
+#define IH_INTERRUPT_ID_TO_DRIVER                   0xFE
+#define IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING  0x7
+
+//thermal over-temp mask defines
+#define THROTTLER_TEMP_CCD_BIT     5
+#define THROTTLER_TEMP_XCD_BIT     6
+#define THROTTLER_TEMP_HBM_BIT     7
+#define THROTTLER_TEMP_AID_BIT     8
+#define THROTTLER_VRHOT_BIT        9
+
+
 // These defines are used with the following messages:
 // SMC_MSG_TransferTableDram2Smu
 // SMC_MSG_TransferTableSmu2Dram
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index d0c49e8883e7a..b08608caafd06 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -175,7 +175,6 @@ static const struct cmn2asic_mapping smu_v13_0_6_table_map[SMU_TABLE_COUNT] = {
 #define THROTTLER_PPT_BIT 1
 #define THROTTLER_TEMP_SOC_BIT 2
 #define THROTTLER_TEMP_VR_GFX_BIT 3
-#define THROTTLER_TEMP_HBM_BIT 4
 
 static const uint8_t smu_v13_0_6_throttler_map[] = {
 	[THROTTLER_PPT_BIT]		= (SMU_THROTTLER_PPT0_BIT),
-- 
GitLab


From 676915e4108f3a8d98944e16e6ce00a6440ac701 Mon Sep 17 00:00:00 2001
From: Asad kamal <asad.kamal@amd.com>
Date: Wed, 15 Feb 2023 15:53:15 +0800
Subject: [PATCH 1219/1791] drm/amd/pm: Add ih for SMU v13.0.6 thermal
 throttling

Add interrupt handler for thermal throttler events from
PMFW on SMUv13.0.6

Signed-off-by: Asad kamal <asad.kamal@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 107 +++++++++++++++++-
 1 file changed, 104 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index b08608caafd06..43a855de7e9e2 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1297,6 +1297,109 @@ static int smu_v13_0_6_set_power_limit(struct smu_context *smu,
 	return smu_v13_0_set_power_limit(smu, limit_type, limit);
 }
 
+static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
+				   struct amdgpu_irq_src *source,
+				   struct amdgpu_iv_entry *entry)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	uint32_t client_id = entry->client_id;
+	uint32_t src_id = entry->src_id;
+	/*
+	 * ctxid is used to distinguish different
+	 * events for SMCToHost interrupt
+	 */
+	uint32_t ctxid = entry->src_data[0];
+	uint32_t data;
+
+	if (client_id == SOC15_IH_CLIENTID_MP1) {
+		if (src_id == IH_INTERRUPT_ID_TO_DRIVER) {
+			/* ACK SMUToHost interrupt */
+			data = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL);
+			data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1);
+			WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, data);
+
+			switch (ctxid) {
+			case IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING:
+				/*
+				 * Increment the throttle interrupt counter
+				 */
+				atomic64_inc(&smu->throttle_int_counter);
+
+				if (!atomic_read(&adev->throttling_logging_enabled))
+					return 0;
+
+				if (__ratelimit(&adev->throttling_logging_rs))
+					schedule_work(&smu->throttling_logging_work);
+
+				break;
+			}
+		}
+	}
+
+	return 0;
+}
+
+int smu_v13_0_6_set_irq_state(struct amdgpu_device *adev,
+			      struct amdgpu_irq_src *source,
+			      unsigned tyep,
+			      enum amdgpu_interrupt_state state)
+{
+	uint32_t val = 0;
+
+	switch (state) {
+	case AMDGPU_IRQ_STATE_DISABLE:
+		/* For MP1 SW irqs */
+		val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL);
+		val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT_CTRL, INT_MASK, 1);
+		WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, val);
+
+		break;
+	case AMDGPU_IRQ_STATE_ENABLE:
+		/* For MP1 SW irqs */
+		val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT);
+		val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT, ID, 0xFE);
+		val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT, VALID, 0);
+		WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT, val);
+
+		val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL);
+		val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT_CTRL, INT_MASK, 0);
+		WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, val);
+
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static const struct amdgpu_irq_src_funcs smu_v13_0_6_irq_funcs =
+{
+	.set = smu_v13_0_6_set_irq_state,
+	.process = smu_v13_0_6_irq_process,
+};
+
+int smu_v13_0_6_register_irq_handler(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+	struct amdgpu_irq_src *irq_src = &smu->irq_source;
+	int ret = 0;
+
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
+	irq_src->num_types = 1;
+	irq_src->funcs = &smu_v13_0_6_irq_funcs;
+
+	ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1,
+				IH_INTERRUPT_ID_TO_DRIVER,
+				irq_src);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
 static int smu_v13_0_6_system_features_control(struct smu_context *smu,
 					       bool enable)
 {
@@ -2042,11 +2145,9 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.feature_is_enabled = smu_cmn_feature_is_enabled,
 	.set_power_limit = smu_v13_0_6_set_power_limit,
 	.set_xgmi_pstate = smu_v13_0_set_xgmi_pstate,
-	/* TODO: Thermal limits unknown, skip these for now
-	.register_irq_handler = smu_v13_0_register_irq_handler,
+	.register_irq_handler = smu_v13_0_6_register_irq_handler,
 	.enable_thermal_alert = smu_v13_0_enable_thermal_alert,
 	.disable_thermal_alert = smu_v13_0_disable_thermal_alert,
-	*/
 	.setup_pptable = smu_v13_0_6_setup_pptable,
 	.baco_is_support = smu_v13_0_6_is_baco_supported,
 	.get_dpm_ultimate_freq = smu_v13_0_6_get_dpm_ultimate_freq,
-- 
GitLab


From 9661bf687623f628729566cc3c58207c44e56258 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 21 Feb 2023 14:47:51 +0530
Subject: [PATCH 1220/1791] drm/amd/pm: Keep interface version in PMFW header

Use the interface version directly from PMFW interface header file rather
than keeping another definition in common smu13 file.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../inc/pmfw_if/smu13_driver_if_aldebaran.h   |  2 +
 .../inc/pmfw_if/smu13_driver_if_v13_0_0.h     |  2 +
 .../inc/pmfw_if/smu13_driver_if_v13_0_4.h     |  2 +-
 .../inc/pmfw_if/smu13_driver_if_v13_0_5.h     |  2 +-
 .../inc/pmfw_if/smu13_driver_if_v13_0_7.h     |  2 +-
 .../inc/pmfw_if/smu13_driver_if_yellow_carp.h |  2 +-
 drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h  | 11 ------
 .../drm/amd/pm/swsmu/smu13/aldebaran_ppt.c    |  1 +
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c    | 39 +------------------
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c  |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c  |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c  |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c  |  1 +
 .../drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c  |  1 +
 15 files changed, 17 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h
index 90200f31ff526..cddf45eebee8d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h
@@ -24,6 +24,8 @@
 #ifndef SMU13_DRIVER_IF_ALDEBARAN_H
 #define SMU13_DRIVER_IF_ALDEBARAN_H
 
+#define SMU13_DRIVER_IF_VERSION_ALDE 0x08
+
 #define NUM_VCLK_DPM_LEVELS   8
 #define NUM_DCLK_DPM_LEVELS   8
 #define NUM_SOCCLK_DPM_LEVELS 8
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
index b686fb68a6e76..fe995651c6f55 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
@@ -24,6 +24,8 @@
 #ifndef SMU13_DRIVER_IF_V13_0_0_H
 #define SMU13_DRIVER_IF_V13_0_0_H
 
+#define SMU13_0_0_DRIVER_IF_VERSION 0x32
+
 //Increment this version if SkuTable_t or BoardTable_t change
 #define PPTABLE_VERSION 0x26
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h
index 2162ecd1057d1..fee9293b3f970 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h
@@ -27,7 +27,7 @@
 // *** IMPORTANT ***
 // SMU TEAM: Always increment the interface version if
 // any structure is changed in this file
-#define PMFW_DRIVER_IF_VERSION 8
+#define SMU13_0_4_DRIVER_IF_VERSION 8
 
 typedef struct {
   int32_t value;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h
index aa971412b4347..7589faa0232d8 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h
@@ -23,7 +23,7 @@
 #ifndef __SMU13_DRIVER_IF_V13_0_5_H__
 #define __SMU13_DRIVER_IF_V13_0_5_H__
 
-#define PMFW_DRIVER_IF_VERSION 4
+#define SMU13_0_5_DRIVER_IF_VERSION 4
 
 // Throttler Status Bitmask
 #define THROTTLER_STATUS_BIT_SPL            0
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
index 4c46a03924512..44e879c51cae9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
@@ -25,7 +25,7 @@
 
 // *** IMPORTANT ***
 // PMFW TEAM: Always increment the interface version on any change to this file
-#define SMU13_DRIVER_IF_VERSION  0x35
+#define SMU13_0_7_DRIVER_IF_VERSION  0x35
 
 //Increment this version if SkuTable_t or BoardTable_t change
 #define PPTABLE_VERSION 0x27
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h
index 25540cb282088..7417634827ad6 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h
@@ -26,7 +26,7 @@
 // *** IMPORTANT ***
 // SMU TEAM: Always increment the interface version if
 // any structure is changed in this file
-#define SMU13_DRIVER_IF_VERSION 4
+#define SMU13_YELLOW_CARP_DRIVER_IF_VERSION 4
 
 typedef struct {
   int32_t value;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
index df3baaab00378..3ae8d5d252a3c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
@@ -25,17 +25,6 @@
 
 #include "amdgpu_smu.h"
 
-#define SMU13_DRIVER_IF_VERSION_INV 0xFFFFFFFF
-#define SMU13_DRIVER_IF_VERSION_YELLOW_CARP 0x04
-#define SMU13_DRIVER_IF_VERSION_ALDE 0x08
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0 0x37
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x08
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10 0x32
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x37
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_10 0x1D
-#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_6 0x0
-
 #define SMU13_MODE1_RESET_WAIT_TIME_IN_MS 500  //500ms
 
 /* MP Apertures */
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index d30ec3005ea19..e80f122d8aec5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -2147,5 +2147,6 @@ void aldebaran_set_ppt_funcs(struct smu_context *smu)
 	smu->clock_map = aldebaran_clk_map;
 	smu->feature_map = aldebaran_feature_mask_map;
 	smu->table_map = aldebaran_table_map;
+	smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_ALDE;
 	smu_v13_0_set_smu_mailbox_registers(smu);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index 393c6a7b96096..b24c8549d0b11 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -269,45 +269,10 @@ int smu_v13_0_check_fw_version(struct smu_context *smu)
 	smu_major = (smu_version >> 16) & 0xff;
 	smu_minor = (smu_version >> 8) & 0xff;
 	smu_debug = (smu_version >> 0) & 0xff;
-	if (smu->is_apu)
+	if (smu->is_apu ||
+	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 6))
 		adev->pm.fw_version = smu_version;
 
-	switch (adev->ip_versions[MP1_HWIP][0]) {
-	case IP_VERSION(13, 0, 2):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_ALDE;
-		break;
-	case IP_VERSION(13, 0, 0):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0;
-		break;
-	case IP_VERSION(13, 0, 10):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10;
-		break;
-	case IP_VERSION(13, 0, 7):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_7;
-		break;
-	case IP_VERSION(13, 0, 1):
-	case IP_VERSION(13, 0, 3):
-	case IP_VERSION(13, 0, 8):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_YELLOW_CARP;
-		break;
-	case IP_VERSION(13, 0, 4):
-	case IP_VERSION(13, 0, 11):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_4;
-		break;
-	case IP_VERSION(13, 0, 5):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_5;
-		break;
-	case IP_VERSION(13, 0, 6):
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_6;
-		adev->pm.fw_version = smu_version;
-		break;
-	default:
-		dev_err(adev->dev, "smu unsupported IP version: 0x%x.\n",
-			adev->ip_versions[MP1_HWIP][0]);
-		smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_INV;
-		break;
-	}
-
 	/* only for dGPU w/ SMU13*/
 	if (adev->pm.fw)
 		dev_dbg(smu->adev->dev, "smu fw reported program %d, version = 0x%08x (%d.%d.%d)\n",
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index 09405ef1e3c83..f4783e685bf83 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -2199,5 +2199,6 @@ void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu)
 	smu->table_map = smu_v13_0_0_table_map;
 	smu->pwr_src_map = smu_v13_0_0_pwr_src_map;
 	smu->workload_map = smu_v13_0_0_workload_map;
+	smu->smc_driver_if_version = SMU13_0_0_DRIVER_IF_VERSION;
 	smu_v13_0_0_set_smu_mailbox_registers(smu);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
index 8fa9a36c38b64..0a0a7debb3ae7 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
@@ -1043,6 +1043,7 @@ void smu_v13_0_4_set_ppt_funcs(struct smu_context *smu)
 	smu->message_map = smu_v13_0_4_message_map;
 	smu->feature_map = smu_v13_0_4_feature_mask_map;
 	smu->table_map = smu_v13_0_4_table_map;
+	smu->smc_driver_if_version = SMU13_0_4_DRIVER_IF_VERSION;
 	smu->is_apu = true;
 
 	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 4))
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
index 66445964efbd1..165b2470b017d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
@@ -1068,6 +1068,7 @@ void smu_v13_0_5_set_ppt_funcs(struct smu_context *smu)
 	smu->feature_map = smu_v13_0_5_feature_mask_map;
 	smu->table_map = smu_v13_0_5_table_map;
 	smu->is_apu = true;
+	smu->smc_driver_if_version = SMU13_0_5_DRIVER_IF_VERSION;
 	smu->param_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_34);
 	smu->msg_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_2);
 	smu->resp_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_33);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 43a855de7e9e2..b46e0414be605 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -2176,5 +2176,6 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
 	smu->clock_map = smu_v13_0_6_clk_map;
 	smu->feature_map = smu_v13_0_6_feature_mask_map;
 	smu->table_map = smu_v13_0_6_table_map;
+	smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION;
 	smu_v13_0_set_smu_mailbox_registers(smu);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 98a33f8ee2095..1b6116cf8b4cd 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -1797,5 +1797,6 @@ void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu)
 	smu->table_map = smu_v13_0_7_table_map;
 	smu->pwr_src_map = smu_v13_0_7_pwr_src_map;
 	smu->workload_map = smu_v13_0_7_workload_map;
+	smu->smc_driver_if_version = SMU13_0_7_DRIVER_IF_VERSION;
 	smu_v13_0_set_smu_mailbox_registers(smu);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
index 04e56b0b3033e..ac5fcca0e47f9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
@@ -1234,5 +1234,6 @@ void yellow_carp_set_ppt_funcs(struct smu_context *smu)
 	smu->feature_map = yellow_carp_feature_mask_map;
 	smu->table_map = yellow_carp_table_map;
 	smu->is_apu = true;
+	smu->smc_driver_if_version = SMU13_YELLOW_CARP_DRIVER_IF_VERSION;
 	smu_v13_0_set_smu_mailbox_registers(smu);
 }
-- 
GitLab


From 909ae7155faebc62af461924a91071c0b9cc4e39 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 27 Feb 2023 16:51:16 +0530
Subject: [PATCH 1221/1791] drm/amd/pm: Initialize power limit for SMU v13.0.6

PMFW will initialize the power limit values even if PPT throttler
feature is disabled. Fetch the limit value from FW.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index b46e0414be605..7474d3ffab93e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1256,21 +1256,6 @@ static int smu_v13_0_6_get_power_limit(struct smu_context *smu,
 	uint32_t power_limit = 0;
 	int ret;
 
-	if (!smu_cmn_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT)) {
-		if (current_power_limit)
-			*current_power_limit = 0;
-		if (default_power_limit)
-			*default_power_limit = 0;
-		if (max_power_limit)
-			*max_power_limit = 0;
-
-		dev_warn(
-			smu->adev->dev,
-			"PPT feature is not enabled, power values can't be fetched.");
-
-		return 0;
-	}
-
 	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetPptLimit, &power_limit);
 
 	if (ret) {
-- 
GitLab


From 8572fa2aa517d2512abba661ddd5e9a44a893dc9 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Fri, 3 Mar 2023 12:20:21 +0800
Subject: [PATCH 1222/1791] drm/amd/pm: Update hw mon attributes for GC version
 9.4.3

Update hw mon attributes for GC Version 9.4.3 to valid ones
on APU and Non APU systems

v2: Group checks along existing one
Added power limit & mclock for gc version 9.4.3

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 36 +++++++++++++++++++++---------
 1 file changed, 25 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 58c2246918fda..385d83eb87069 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -3362,7 +3362,8 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 		return 0;
 
 	/* Skip crit temp on APU */
-	if ((adev->flags & AMD_IS_APU) && (adev->family >= AMDGPU_FAMILY_CZ) &&
+	if ((((adev->flags & AMD_IS_APU) && (adev->family >= AMDGPU_FAMILY_CZ)) ||
+	    (gc_ver == IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr))
 		return 0;
@@ -3395,9 +3396,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	      attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */
 		effective_mode &= ~S_IWUSR;
 
-	/* In the case of APUs, this is only implemented on Vangogh */
+	/* not implemented yet for APUs other than GC 10.3.1 (vangogh) and 9.4.3 */
 	if (((adev->family == AMDGPU_FAMILY_SI) ||
-	     ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)))) &&
+	     ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)) &&
+	      (gc_ver != IP_VERSION(9, 4, 3)))) &&
 	    (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
 	     attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr ||
 	     attr == &sensor_dev_attr_power1_cap.dev_attr.attr ||
@@ -3426,25 +3428,39 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 		return 0;
 
 	if ((adev->family == AMDGPU_FAMILY_SI ||	/* not implemented yet */
-	     adev->family == AMDGPU_FAMILY_KV) &&	/* not implemented yet */
+	     adev->family == AMDGPU_FAMILY_KV ||	/* not implemented yet */
+	     (gc_ver == IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_in0_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_in0_label.dev_attr.attr))
 		return 0;
 
-	/* only APUs have vddnb */
-	if (!(adev->flags & AMD_IS_APU) &&
+	/* only APUs other than gc 9,4,3 have vddnb */
+	if ((!(adev->flags & AMD_IS_APU) || (gc_ver == IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_in1_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_in1_label.dev_attr.attr))
 		return 0;
 
-	/* no mclk on APUs */
-	if ((adev->flags & AMD_IS_APU) &&
+	/* no mclk on APUs other than gc 9,4,3*/
+	if (((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_freq2_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_freq2_label.dev_attr.attr))
 		return 0;
 
-	/* only SOC15 dGPUs support hotspot and mem temperatures */
 	if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0)) &&
+	    (gc_ver != IP_VERSION(9, 4, 3)) &&
+	    (attr == &sensor_dev_attr_temp2_input.dev_attr.attr ||
+	     attr == &sensor_dev_attr_temp2_label.dev_attr.attr))
+		return 0;
+
+	/* Only hotspot temperature for gc 9,4,3*/
+	if ((gc_ver == IP_VERSION(9, 4, 3)) &&
+	    (attr == &sensor_dev_attr_temp1_input.dev_attr.attr ||
+	     attr == &sensor_dev_attr_temp1_label.dev_attr.attr))
+		return 0;
+
+	/* only SOC15 dGPUs support hotspot and mem temperatures */
+	if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0) ||
+	    (gc_ver == IP_VERSION(9, 4, 3))) &&
 	    (attr == &sensor_dev_attr_temp2_crit.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp2_crit_hyst.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp3_crit.dev_attr.attr ||
@@ -3452,9 +3468,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_temp1_emergency.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp2_emergency.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp3_emergency.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp2_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp3_input.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp2_label.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp3_label.dev_attr.attr))
 		return 0;
 
-- 
GitLab


From bfb4fd20b3e6997876068f469c14d963b227d896 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Wed, 8 Mar 2023 22:30:58 +0800
Subject: [PATCH 1223/1791] drm/amd/pm: Expose mem temperature for GC version
 9.4.3

Add mem temperature as part of hw mon attributes for GC version 9.4.3

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 385d83eb87069..40100b77b2d9a 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -3449,10 +3449,12 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0)) &&
 	    (gc_ver != IP_VERSION(9, 4, 3)) &&
 	    (attr == &sensor_dev_attr_temp2_input.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp2_label.dev_attr.attr))
+	     attr == &sensor_dev_attr_temp2_label.dev_attr.attr ||
+	     attr == &sensor_dev_attr_temp3_input.dev_attr.attr ||
+	     attr == &sensor_dev_attr_temp3_label.dev_attr.attr))
 		return 0;
 
-	/* Only hotspot temperature for gc 9,4,3*/
+	/* hotspot temperature for gc 9,4,3*/
 	if ((gc_ver == IP_VERSION(9, 4, 3)) &&
 	    (attr == &sensor_dev_attr_temp1_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp1_label.dev_attr.attr))
@@ -3467,9 +3469,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_temp3_crit_hyst.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp1_emergency.dev_attr.attr ||
 	     attr == &sensor_dev_attr_temp2_emergency.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp3_emergency.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp3_input.dev_attr.attr ||
-	     attr == &sensor_dev_attr_temp3_label.dev_attr.attr))
+	     attr == &sensor_dev_attr_temp3_emergency.dev_attr.attr))
 		return 0;
 
 	/* only Vangogh has fast PPT limit and power labels */
-- 
GitLab


From fe9e5f56feb287b3f14b0a5892061a1da2b89b5b Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 10 Mar 2023 18:11:25 +0530
Subject: [PATCH 1224/1791] drm/amd/pm: Update PMFW headers for version 85.54

It adds message support for FW notification on driver unload.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../inc/pmfw_if/smu13_driver_if_v13_0_6.h      | 18 ------------------
 .../pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h   |  3 ++-
 2 files changed, 2 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
index 8b7fa0fa59c3c..de84fff39799a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
@@ -132,22 +132,4 @@ typedef struct {
 #define THROTTLER_TEMP_AID_BIT     8
 #define THROTTLER_VRHOT_BIT        9
 
-
-// These defines are used with the following messages:
-// SMC_MSG_TransferTableDram2Smu
-// SMC_MSG_TransferTableSmu2Dram
-// #define TABLE_PPTABLE                 0
-// #define TABLE_AVFS_PSM_DEBUG          1
-// #define TABLE_AVFS_FUSE_OVERRIDE      2
-// #define TABLE_PMSTATUSLOG             3
-// #define TABLE_SMU_METRICS             4
-// #define TABLE_DRIVER_SMU_CONFIG       5
-// #define TABLE_I2C_COMMANDS            6
-// #define TABLE_COUNT                   7
-
-// // Table transfer status
-// #define TABLE_TRANSFER_OK         0x0
-// #define TABLE_TRANSFER_FAILED     0xFF
-// #define TABLE_TRANSFER_PENDING    0xAB
-
 #endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
index b838e8db395ac..ae4f44c4b8771 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
@@ -82,7 +82,8 @@
 #define PPSMC_MSG_SetSoftMaxGfxClk                  0x31
 #define PPSMC_MSG_GetMinGfxDpmFreq                  0x32
 #define PPSMC_MSG_GetMaxGfxDpmFreq                  0x33
-#define PPSMC_Message_Count                         0x34
+#define PPSMC_MSG_PrepareForDriverUnload            0x34
+#define PPSMC_Message_Count                         0x35
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1
-- 
GitLab


From 7214c08c168046aadf15e3d731ee673f26c77213 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 10 Mar 2023 11:33:37 +0530
Subject: [PATCH 1225/1791] drm/amd/pm: Notify PMFW about driver unload cases

On SMU v13.0.6 APUs, FW will need to take some actions if driver is going
to halt RLC. Notify PMFW that driver is not going to manage device so
that FW takes care of the required actions.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 25 +++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 7474d3ffab93e..6dcafd04c98d7 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -122,6 +122,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU
 	MSG_MAP(GetMaxGfxclkFrequency,               PPSMC_MSG_GetMaxGfxDpmFreq,                0),
 	MSG_MAP(SetSoftMinGfxclk,                    PPSMC_MSG_SetSoftMinGfxClk,                0),
 	MSG_MAP(SetSoftMaxGfxClk,                    PPSMC_MSG_SetSoftMaxGfxClk,                0),
+	MSG_MAP(PrepareMp1ForUnload,                 PPSMC_MSG_PrepareForDriverUnload,          0),
 };
 
 static const struct cmn2asic_mapping smu_v13_0_6_clk_map[SMU_CLK_COUNT] = {
@@ -1385,14 +1386,34 @@ int smu_v13_0_6_register_irq_handler(struct smu_context *smu)
 	return ret;
 }
 
+static int smu_v13_0_6_notify_unload(struct smu_context *smu)
+{
+	uint32_t smu_version;
+
+	smu_cmn_get_smc_version(smu, NULL, &smu_version);
+	if (smu_version <= 0x553500)
+		return 0;
+
+	dev_dbg(smu->adev->dev, "Notify PMFW about driver unload");
+	/* Ignore return, just intimate FW that driver is not going to be there */
+	smu_cmn_send_smc_msg(smu, SMU_MSG_PrepareMp1ForUnload, NULL);
+
+	return 0;
+}
+
 static int smu_v13_0_6_system_features_control(struct smu_context *smu,
 					       bool enable)
 {
+	struct amdgpu_device *adev = smu->adev;
 	int ret;
 
-	/* Nothing to be done for APU */
-	if (smu->adev->flags & AMD_IS_APU)
+	/* On APUs, notify FW that the device is no longer driver managed */
+	if (adev->flags & AMD_IS_APU) {
+		if (!enable)
+			smu_v13_0_6_notify_unload(smu);
+
 		return 0;
+	}
 
 	ret = smu_v13_0_system_features_control(smu, enable);
 
-- 
GitLab


From 8f2ccaaa373815ff94223dc2e3f6d53ff2f3ecb3 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 9 Mar 2023 13:04:56 +0530
Subject: [PATCH 1226/1791] drm/amdgpu: Add mode-2 reset in SMU v13.0.6

Modifications to mode-2 reset flow for SMU v13.0.6 ASICs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c               |  9 +++++++++
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 16 +++++++---------
 2 files changed, 16 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 9fbfd0811d067..082c1e9308d04 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -556,6 +556,15 @@ soc15_asic_reset_method(struct amdgpu_device *adev)
 		if (connected_to_cpu)
 			return AMD_RESET_METHOD_MODE2;
 		break;
+	case IP_VERSION(13, 0, 6):
+		/* Use gpu_recovery param to target a reset method.
+		 * Enable triggering of GPU reset only if specified
+		 * by module parameter.
+		 */
+		if (amdgpu_gpu_recovery == 4 || amdgpu_gpu_recovery == 5)
+			return AMD_RESET_METHOD_MODE2;
+		else
+			return AMD_RESET_METHOD_NONE;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 6dcafd04c98d7..4b808c0addc26 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -2024,27 +2024,27 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 
 static int smu_v13_0_6_mode2_reset(struct smu_context *smu)
 {
-	u32 smu_version;
 	int ret = 0, index;
 	struct amdgpu_device *adev = smu->adev;
 	int timeout = 10;
 
-	smu_cmn_get_smc_version(smu, NULL, &smu_version);
-
 	index = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_MSG,
 					       SMU_MSG_GfxDeviceDriverReset);
 
 	mutex_lock(&smu->message_lock);
+
 	ret = smu_cmn_send_msg_without_waiting(smu, (uint16_t)index,
 					       SMU_RESET_MODE_2);
+
 	/* This is similar to FLR, wait till max FLR timeout */
 	msleep(100);
+
 	dev_dbg(smu->adev->dev, "restore config space...\n");
 	/* Restore the config space saved during init */
 	amdgpu_device_load_pci_state(adev->pdev);
 
 	dev_dbg(smu->adev->dev, "wait for reset ack\n");
-	while (ret == -ETIME && timeout) {
+	do {
 		ret = smu_cmn_wait_for_response(smu);
 		/* Wait a bit more time for getting ACK */
 		if (ret == -ETIME) {
@@ -2053,16 +2053,14 @@ static int smu_v13_0_6_mode2_reset(struct smu_context *smu)
 			continue;
 		}
 
-		if (ret != 1) {
+		if (ret) {
 			dev_err(adev->dev,
-				"failed to send mode2 message \tparam: 0x%08x response %#x\n",
+				"failed to send mode2 message \tparam: 0x%08x error code %d\n",
 				SMU_RESET_MODE_2, ret);
 			goto out;
 		}
-	}
+	} while (ret == -ETIME && timeout);
 
-	if (ret == 1)
-		ret = 0;
 out:
 	mutex_unlock(&smu->message_lock);
 
-- 
GitLab


From 463e953ea2eda25fbde70e0e72900f5bafeff93c Mon Sep 17 00:00:00 2001
From: Yang Wang <KevinYang.Wang@amd.com>
Date: Thu, 27 Apr 2023 10:36:51 +0800
Subject: [PATCH 1227/1791] drm/amd/pm: fix wrong smu socclk value

fix typo about smu socclk value.

Signed-off-by: Yang Wang <KevinYang.Wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 4b808c0addc26..a712b2bf2d25a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -387,7 +387,7 @@ static int smu_v13_0_6_get_dpm_ultimate_freq(struct smu_context *smu,
 			break;
 		case SMU_SOCCLK:
 			if (pptable->Init)
-				clock_limit = pptable->UclkFrequencyTable[0];
+				clock_limit = pptable->SocclkFrequencyTable[0];
 			break;
 		case SMU_FCLK:
 			if (pptable->Init)
-- 
GitLab


From ded7d99eb5b78931cec30dd49cd4097d0ac770e1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 16 Jan 2023 10:55:38 +0530
Subject: [PATCH 1228/1791] drm/amdgpu: Add flags for partition mode query

It's not required to take lock on all cases while querying partition
mode. Querying partition mode during KFD init process doesn't need to
take a lock. Init process after a switch will already be happening under
lock. Control the behaviour by adding flags to xcp_query_partition_mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 8 +++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h | 5 ++++-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 4 +++-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 2 +-
 5 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 70c6099353b82..1487ecac27054 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1177,7 +1177,8 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 	int mode;
 	char *partition_mode;
 
-	mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr);
+	mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
+					       AMDGPU_XCP_FL_NONE);
 
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index f59bc450cabea..5b999e5334bbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -170,7 +170,7 @@ int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 	return ret;
 }
 
-int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
+int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
 {
 	int mode;
 
@@ -180,7 +180,8 @@ int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
 	if (!xcp_mgr->funcs || !xcp_mgr->funcs->query_partition_mode)
 		return xcp_mgr->mode;
 
-	mutex_lock(&xcp_mgr->xcp_lock);
+	if (!(flags & AMDGPU_XCP_FL_LOCKED))
+		mutex_lock(&xcp_mgr->xcp_lock);
 	mode = xcp_mgr->funcs->query_partition_mode(xcp_mgr);
 	if (mode != xcp_mgr->mode)
 		dev_WARN(
@@ -188,7 +189,8 @@ int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
 			"Cached partition mode %d not matching with device mode %d",
 			xcp_mgr->mode, mode);
 
-	mutex_unlock(&xcp_mgr->xcp_lock);
+	if (!(flags & AMDGPU_XCP_FL_LOCKED))
+		mutex_unlock(&xcp_mgr->xcp_lock);
 
 	return mode;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index f0b973c6092fd..9fa6f0ea20612 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -30,6 +30,9 @@
 
 #define AMDGPU_XCP_MODE_NONE -1
 
+#define AMDGPU_XCP_FL_NONE 0
+#define AMDGPU_XCP_FL_LOCKED (1 << 0)
+
 enum AMDGPU_XCP_IP_BLOCK {
 	AMDGPU_XCP_GFXHUB,
 	AMDGPU_XCP_GFX,
@@ -99,7 +102,7 @@ int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
 
 int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
 			int init_xcps, struct amdgpu_xcp_mgr_funcs *xcp_funcs);
-int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr);
+int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags);
 int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode);
 int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
 			     enum AMDGPU_XCP_IP_BLOCK ip, int instance);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 42877c4505f1e..69867294117ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1940,7 +1940,9 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 {
 	int r, i, num_xcc;
 
-	if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr) == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
+	if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
+					    AMDGPU_XCP_FL_NONE) ==
+	    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
 		amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, amdgpu_user_partt_mode);
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index d7cffd91f1d7d..4293cbf9ceb04 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -645,7 +645,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	 * If the VMID range changes for GFX9.4.3, then this code MUST be
 	 * revisited.
 	 */
-	partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr);
+	partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr, AMDGPU_XCP_FL_LOCKED);
 	if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
 	    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
 	    kfd->num_nodes != 1) {
-- 
GitLab


From 46d79cbf9ac64a5e63f0c85f256ba6400a1f2024 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 20 Jan 2023 15:53:47 +0530
Subject: [PATCH 1229/1791] drm/amdgpu: Use transient mode during xcp switch

During partition switch, keep the state as transient mode. Fetch the
latest state if switch fails.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 18 +++++++++++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  1 +
 2 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 5b999e5334bbd..e8aa4d6c6b621 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -139,7 +139,7 @@ static int __amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps)
 
 int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 {
-	int ret, num_xcps = 0;
+	int ret, curr_mode, num_xcps = 0;
 
 	if (!xcp_mgr || mode == AMDGPU_XCP_MODE_NONE)
 		return -EINVAL;
@@ -152,10 +152,22 @@ int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 
 	mutex_lock(&xcp_mgr->xcp_lock);
 
+	curr_mode = xcp_mgr->mode;
+	/* State set to transient mode */
+	xcp_mgr->mode = AMDGPU_XCP_MODE_TRANS;
+
 	ret = xcp_mgr->funcs->switch_partition_mode(xcp_mgr, mode, &num_xcps);
 
-	if (ret)
+	if (ret) {
+		/* Failed, get whatever mode it's at now */
+		if (xcp_mgr->funcs->query_partition_mode)
+			xcp_mgr->mode = amdgpu_xcp_query_partition_mode(
+				xcp_mgr, AMDGPU_XCP_FL_LOCKED);
+		else
+			xcp_mgr->mode = curr_mode;
+
 		goto out;
+	}
 
 	if (!num_xcps || num_xcps > MAX_XCP) {
 		ret = -EINVAL;
@@ -183,7 +195,7 @@ int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
 	if (!(flags & AMDGPU_XCP_FL_LOCKED))
 		mutex_lock(&xcp_mgr->xcp_lock);
 	mode = xcp_mgr->funcs->query_partition_mode(xcp_mgr);
-	if (mode != xcp_mgr->mode)
+	if (xcp_mgr->mode != AMDGPU_XCP_MODE_TRANS && mode != xcp_mgr->mode)
 		dev_WARN(
 			xcp_mgr->adev->dev,
 			"Cached partition mode %d not matching with device mode %d",
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 9fa6f0ea20612..1d3dc7d68f543 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -29,6 +29,7 @@
 #define MAX_XCP 8
 
 #define AMDGPU_XCP_MODE_NONE -1
+#define AMDGPU_XCP_MODE_TRANS -2
 
 #define AMDGPU_XCP_FL_NONE 0
 #define AMDGPU_XCP_FL_LOCKED (1 << 0)
-- 
GitLab


From 01ef47477d05b784ab6ac26fa6878987eda436f1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 2 Feb 2023 15:13:12 +0530
Subject: [PATCH 1230/1791] drm/amdgpu: Add FGCG for GFX v9.4.3

It's not fine grain, behaves similar to MGCG.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 082c1e9308d04..122ba1a505c34 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1156,8 +1156,9 @@ static int soc15_common_early_init(void *handle)
 		adev->cg_flags =
 			AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGCG |
 			AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_SDMA_MGCG |
-			AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_VCN_MGCG |
-			AMD_CG_SUPPORT_JPEG_MGCG;
+			AMD_CG_SUPPORT_GFX_FGCG | AMD_CG_SUPPORT_REPEATER_FGCG |
+			AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG |
+			AMD_CG_SUPPORT_IH_CG;
 		adev->pg_flags =
 			AMD_PG_SUPPORT_VCN |
 			AMD_PG_SUPPORT_VCN_DPG |
-- 
GitLab


From f4d8b6f5c61ab5e98258bd0072d733741c76bd8d Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 31 Jan 2023 11:23:50 -0500
Subject: [PATCH 1231/1791] drm/amdkfd: Enable SVM on Native mode

This patch enables SVM capability on GFX9.4.3 when
run in Native mode. It also sets best_prefetch and
best_restore locations to CPU as there is no VRAM.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 8 ++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h | 3 ++-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 2b79849ddd30f..cf354f9e42856 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -2543,6 +2543,9 @@ svm_range_best_restore_location(struct svm_range *prange,
 		return -1;
 	}
 
+	if (node->adev->gmc.is_app_apu)
+		return 0;
+
 	if (prange->preferred_loc == gpuid ||
 	    prange->preferred_loc == KFD_IOCTL_SVM_LOCATION_SYSMEM) {
 		return prange->preferred_loc;
@@ -3256,6 +3259,11 @@ svm_range_best_prefetch_location(struct svm_range *prange)
 		goto out;
 	}
 
+	if (bo_node->adev->gmc.is_app_apu) {
+		best_loc = 0;
+		goto out;
+	}
+
 	if (p->xnack_enabled)
 		bitmap_copy(bitmap, prange->bitmap_aip, MAX_GPU_INSTANCE);
 	else
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index 5116786718b66..7515ddade3ae9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -200,7 +200,8 @@ void svm_range_list_lock_and_flush_work(struct svm_range_list *svms, struct mm_s
 /* SVM API and HMM page migration work together, device memory type
  * is initialized to not 0 when page migration register device memory.
  */
-#define KFD_IS_SVM_API_SUPPORTED(dev) ((dev)->pgmap.type != 0)
+#define KFD_IS_SVM_API_SUPPORTED(dev) ((dev)->pgmap.type != 0 ||\
+				       (dev)->adev->gmc.is_app_apu)
 
 void svm_range_bo_unref_async(struct svm_range_bo *svm_bo);
 
-- 
GitLab


From 4d5275ab0b18d17697392aafd93e206e6b9de647 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 27 Jan 2023 18:18:17 +0530
Subject: [PATCH 1232/1791] drm/amdgpu: Add parsing of acpi xcc objects

Add parsing of ACPI xcc objects and fill in relevant info from them by
invoking the DSM methods.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      |   2 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 294 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c  |   1 +
 3 files changed, 297 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 39743d44b567b..880bf9d672842 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1404,11 +1404,13 @@ int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev);
 void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps);
 bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev);
 void amdgpu_acpi_detect(void);
+void amdgpu_acpi_release(void);
 #else
 static inline int amdgpu_acpi_init(struct amdgpu_device *adev) { return 0; }
 static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { }
 static inline bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev) { return false; }
 static inline void amdgpu_acpi_detect(void) { }
+static inline void amdgpu_acpi_release(void) { }
 static inline bool amdgpu_acpi_is_power_shift_control_supported(void) { return false; }
 static inline int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev,
 						  u8 dev_state, bool drv_state) { return 0; }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index aeeec211861c4..a3a2ef43abfc7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -38,6 +38,43 @@
 #include "amd_acpi.h"
 #include "atom.h"
 
+/* Declare GUID for AMD _DSM method for XCCs */
+static const guid_t amd_xcc_dsm_guid = GUID_INIT(0x8267f5d5, 0xa556, 0x44f2,
+						 0xb8, 0xb4, 0x45, 0x56, 0x2e,
+						 0x8c, 0x5b, 0xec);
+
+#define AMD_XCC_HID_START 3000
+#define AMD_XCC_DSM_GET_NUM_FUNCS 0
+#define AMD_XCC_DSM_GET_SUPP_MODE 1
+#define AMD_XCC_DSM_GET_XCP_MODE 2
+#define AMD_XCC_DSM_GET_VF_XCC_MAPPING 4
+#define AMD_XCC_DSM_GET_TMR_INFO 5
+#define AMD_XCC_DSM_NUM_FUNCS 5
+
+#define AMD_XCC_MAX_HID 24
+
+/* Encapsulates the XCD acpi object information */
+struct amdgpu_acpi_xcc_info {
+	struct list_head list;
+	int mem_node;
+	uint8_t xcp_node;
+	uint8_t phy_id;
+	acpi_handle handle;
+};
+
+struct amdgpu_acpi_dev_info {
+	struct list_head list;
+	struct list_head xcc_list;
+	uint16_t bdf;
+	uint16_t supp_xcp_mode;
+	uint16_t xcp_mode;
+	uint16_t mem_mode;
+	uint64_t tmr_base;
+	uint64_t tmr_size;
+};
+
+struct list_head amdgpu_acpi_dev_list;
+
 struct amdgpu_atif_notification_cfg {
 	bool enabled;
 	int command_code;
@@ -801,6 +838,240 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta
 	return r;
 }
 
+/**
+ * amdgpu_acpi_get_node_id - obtain the NUMA node id for corresponding amdgpu
+ * acpi device handle
+ *
+ * @handle: acpi handle
+ * @nid: NUMA Node id returned by the platform firmware
+ *
+ * Queries the ACPI interface to fetch the corresponding NUMA Node ID for a
+ * given amdgpu acpi device.
+ *
+ * Returns ACPI STATUS OK with Node ID on success or the corresponding failure reason
+ */
+acpi_status amdgpu_acpi_get_node_id(acpi_handle handle, int *nid)
+{
+#ifdef CONFIG_ACPI_NUMA
+	u64 pxm;
+	acpi_status status;
+
+	status = acpi_evaluate_integer(handle, "_PXM", NULL, &pxm);
+
+	if (ACPI_FAILURE(status))
+		return status;
+
+	*nid = pxm_to_node(pxm);
+
+	return_ACPI_STATUS(AE_OK);
+#else
+	return_ACPI_STATUS(AE_NOT_EXIST);
+#endif
+}
+
+struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u16 bdf)
+{
+	struct amdgpu_acpi_dev_info *acpi_dev;
+
+	if (list_empty(&amdgpu_acpi_dev_list))
+		return NULL;
+
+	list_for_each_entry(acpi_dev, &amdgpu_acpi_dev_list, list)
+		if (acpi_dev->bdf == bdf)
+			return acpi_dev;
+
+	return NULL;
+}
+
+static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info,
+				struct amdgpu_acpi_xcc_info *xcc_info, u16 bdf)
+{
+	struct amdgpu_acpi_dev_info *tmp;
+	union acpi_object *obj;
+	int ret = -ENOENT;
+
+	*dev_info = NULL;
+	tmp = kzalloc(sizeof(struct amdgpu_acpi_dev_info), GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&tmp->xcc_list);
+	INIT_LIST_HEAD(&tmp->list);
+	tmp->bdf = bdf;
+
+	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
+				      AMD_XCC_DSM_GET_SUPP_MODE, NULL,
+				      ACPI_TYPE_INTEGER);
+
+	if (!obj) {
+		acpi_handle_debug(xcc_info->handle,
+				  "_DSM function %d evaluation failed",
+				  AMD_XCC_DSM_GET_SUPP_MODE);
+		ret = -ENOENT;
+		goto out;
+	}
+
+	tmp->supp_xcp_mode = obj->integer.value & 0xFFFF;
+	ACPI_FREE(obj);
+
+	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
+				      AMD_XCC_DSM_GET_XCP_MODE, NULL,
+				      ACPI_TYPE_INTEGER);
+
+	if (!obj) {
+		acpi_handle_debug(xcc_info->handle,
+				  "_DSM function %d evaluation failed",
+				  AMD_XCC_DSM_GET_XCP_MODE);
+		ret = -ENOENT;
+		goto out;
+	}
+
+	tmp->xcp_mode = obj->integer.value & 0xFFFF;
+	tmp->mem_mode = (obj->integer.value >> 32) & 0xFFFF;
+	ACPI_FREE(obj);
+
+	/* Evaluate DSMs and fill XCC information */
+	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
+				      AMD_XCC_DSM_GET_TMR_INFO, NULL,
+				      ACPI_TYPE_PACKAGE);
+
+	if (!obj || obj->package.count < 2) {
+		acpi_handle_debug(xcc_info->handle,
+				  "_DSM function %d evaluation failed",
+				  AMD_XCC_DSM_GET_TMR_INFO);
+		ret = -ENOENT;
+		goto out;
+	}
+
+	tmp->tmr_base = obj->package.elements[0].integer.value;
+	tmp->tmr_size = obj->package.elements[1].integer.value;
+	ACPI_FREE(obj);
+
+	DRM_DEBUG_DRIVER(
+		"New dev(%x): Supported xcp mode: %x curr xcp_mode : %x mem mode : %x, tmr base: %llx tmr size: %llx  ",
+		tmp->bdf, tmp->supp_xcp_mode, tmp->xcp_mode, tmp->mem_mode,
+		tmp->tmr_base, tmp->tmr_size);
+	list_add_tail(&tmp->list, &amdgpu_acpi_dev_list);
+	*dev_info = tmp;
+
+	return 0;
+
+out:
+	if (obj)
+		ACPI_FREE(obj);
+	kfree(tmp);
+
+	return ret;
+}
+
+static int amdgpu_acpi_get_xcc_info(struct amdgpu_acpi_xcc_info *xcc_info,
+				    u16 *bdf)
+{
+	union acpi_object *obj;
+	acpi_status status;
+	int ret = -ENOENT;
+
+	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
+				      AMD_XCC_DSM_GET_NUM_FUNCS, NULL,
+				      ACPI_TYPE_INTEGER);
+
+	if (!obj || obj->integer.value != AMD_XCC_DSM_NUM_FUNCS)
+		goto out;
+	ACPI_FREE(obj);
+
+	/* Evaluate DSMs and fill XCC information */
+	obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0,
+				      AMD_XCC_DSM_GET_VF_XCC_MAPPING, NULL,
+				      ACPI_TYPE_INTEGER);
+
+	if (!obj) {
+		acpi_handle_debug(xcc_info->handle,
+				  "_DSM function %d evaluation failed",
+				  AMD_XCC_DSM_GET_VF_XCC_MAPPING);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* PF xcc id [39:32] */
+	xcc_info->phy_id = (obj->integer.value >> 32) & 0xFF;
+	/* xcp node of this xcc [47:40] */
+	xcc_info->xcp_node = (obj->integer.value >> 40) & 0xFF;
+	/* PF bus/dev/fn of this xcc [63:48] */
+	*bdf = (obj->integer.value >> 48) & 0xFFFF;
+	ACPI_FREE(obj);
+	obj = NULL;
+
+	status = amdgpu_acpi_get_node_id(xcc_info->handle, &xcc_info->mem_node);
+
+	/* TODO: check if this check is required */
+	if (ACPI_SUCCESS(status))
+		ret = 0;
+out:
+	if (obj)
+		ACPI_FREE(obj);
+
+	return ret;
+}
+
+static int amdgpu_acpi_enumerate_xcc(void)
+{
+	struct amdgpu_acpi_dev_info *dev_info = NULL;
+	struct amdgpu_acpi_xcc_info *xcc_info;
+	struct acpi_device *acpi_dev;
+	char hid[ACPI_ID_LEN];
+	int ret, id;
+	u16 bdf;
+
+	INIT_LIST_HEAD(&amdgpu_acpi_dev_list);
+
+	for (id = 0; id < AMD_XCC_MAX_HID; id++) {
+		sprintf(hid, "%s%d", "AMD", AMD_XCC_HID_START + id);
+		acpi_dev = acpi_dev_get_first_match_dev(hid, NULL, -1);
+		/* These ACPI objects are expected to be in sequential order. If
+		 * one is not found, no need to check the rest.
+		 */
+		if (!acpi_dev) {
+			DRM_DEBUG_DRIVER("No matching acpi device found for %s",
+					 hid);
+			break;
+		}
+
+		xcc_info = kzalloc(sizeof(struct amdgpu_acpi_xcc_info),
+				   GFP_KERNEL);
+		if (!xcc_info) {
+			DRM_ERROR("Failed to allocate memory for xcc info\n");
+			return -ENOMEM;
+		}
+
+		INIT_LIST_HEAD(&xcc_info->list);
+		xcc_info->handle = acpi_device_handle(acpi_dev);
+		acpi_dev_put(acpi_dev);
+
+		ret = amdgpu_acpi_get_xcc_info(xcc_info, &bdf);
+		if (ret) {
+			kfree(xcc_info);
+			continue;
+		}
+
+		dev_info = amdgpu_acpi_get_dev(bdf);
+
+		if (!dev_info)
+			ret = amdgpu_acpi_dev_init(&dev_info, xcc_info, bdf);
+
+		if (ret == -ENOMEM)
+			return ret;
+
+		if (!dev_info) {
+			kfree(xcc_info);
+			continue;
+		}
+
+		list_add_tail(&xcc_info->list, &dev_info->xcc_list);
+	}
+
+	return 0;
+}
+
 /**
  * amdgpu_acpi_event - handle notify events
  *
@@ -1054,6 +1325,29 @@ void amdgpu_acpi_detect(void)
 	} else {
 		atif->backlight_caps.caps_valid = false;
 	}
+
+	amdgpu_acpi_enumerate_xcc();
+}
+
+void amdgpu_acpi_release(void)
+{
+	struct amdgpu_acpi_dev_info *dev_info, *dev_tmp;
+	struct amdgpu_acpi_xcc_info *xcc_info, *xcc_tmp;
+
+	if (list_empty(&amdgpu_acpi_dev_list))
+		return;
+
+	list_for_each_entry_safe(dev_info, dev_tmp, &amdgpu_acpi_dev_list,
+				 list) {
+		list_for_each_entry_safe(xcc_info, xcc_tmp, &dev_info->xcc_list,
+					 list) {
+			list_del(&xcc_info->list);
+			kfree(xcc_info);
+		}
+
+		list_del(&dev_info->list);
+		kfree(dev_info);
+	}
 }
 
 #if IS_ENABLED(CONFIG_SUSPEND)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 9e9da2ac5c82e..f319318a88134 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2898,6 +2898,7 @@ static void __exit amdgpu_exit(void)
 	amdgpu_amdkfd_fini();
 	pci_unregister_driver(&amdgpu_kms_pci_driver);
 	amdgpu_unregister_atpx_handler();
+	amdgpu_acpi_release();
 	amdgpu_sync_fini();
 	amdgpu_fence_slab_fini();
 	mmu_notifier_synchronize();
-- 
GitLab


From 6e01882267a696b022cfe3473a0d3e5ccbe54010 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 27 Jan 2023 18:40:14 +0530
Subject: [PATCH 1233/1791] drm/amdgpu: Add API to get tmr info from acpi

In certain configs, TMR information is available from ACPI. Add API to
fetch the information.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      |  7 +++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 20 ++++++++++++++++++++
 2 files changed, 27 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 880bf9d672842..f4461bc8b1fdd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1400,6 +1400,8 @@ int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev,
 				    u8 dev_state, bool drv_state);
 int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_state);
 int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev);
+int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset,
+			     u64 *tmr_size);
 
 void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps);
 bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev);
@@ -1407,6 +1409,11 @@ void amdgpu_acpi_detect(void);
 void amdgpu_acpi_release(void);
 #else
 static inline int amdgpu_acpi_init(struct amdgpu_device *adev) { return 0; }
+static inline int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev,
+					   u64 *tmr_offset, u64 *tmr_size)
+{
+	return -EINVAL;
+}
 static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { }
 static inline bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev) { return false; }
 static inline void amdgpu_acpi_detect(void) { }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index a3a2ef43abfc7..9dbdd699dcea6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1072,6 +1072,26 @@ static int amdgpu_acpi_enumerate_xcc(void)
 	return 0;
 }
 
+int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset,
+			     u64 *tmr_size)
+{
+	struct amdgpu_acpi_dev_info *dev_info;
+	u16 bdf;
+
+	if (!tmr_offset || !tmr_size)
+		return -EINVAL;
+
+	bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn;
+	dev_info = amdgpu_acpi_get_dev(bdf);
+	if (!dev_info)
+		return -ENOENT;
+
+	*tmr_offset = dev_info->tmr_base;
+	*tmr_size = dev_info->tmr_size;
+
+	return 0;
+}
+
 /**
  * amdgpu_acpi_event - handle notify events
  *
-- 
GitLab


From 368bb1bcfb3a3bc70793cd347abe0bc60c01d94b Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 30 Jan 2023 09:38:09 +0530
Subject: [PATCH 1234/1791] drm/amdgpu: Read discovery info from system memory

On certain ASICs, discovery info is available at reserved region in system
memory. The location is available through ACPI interface. Add API to read
discovery info from there.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 23 +++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 6701f17a4db62..246070938c41c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -203,6 +203,29 @@ static int hw_id_map[MAX_HWIP] = {
 	[PCIE_HWIP]	= PCIE_HWID,
 };
 
+static int amdgpu_discovery_read_binary_from_sysmem(struct amdgpu_device *adev, uint8_t *binary)
+{
+	u64 tmr_offset, tmr_size, pos;
+	void *discv_regn;
+	int ret;
+
+	ret = amdgpu_acpi_get_tmr_info(adev, &tmr_offset, &tmr_size);
+	if (ret)
+		return ret;
+
+	pos = tmr_offset + tmr_size - DISCOVERY_TMR_OFFSET;
+
+	/* This region is read-only and reserved from system use */
+	discv_regn = memremap(pos, adev->mman.discovery_tmr_size, MEMREMAP_WC);
+	if (discv_regn) {
+		memcpy(binary, discv_regn, adev->mman.discovery_tmr_size);
+		memunmap(discv_regn);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
 static void amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary)
 {
 	uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20;
-- 
GitLab


From ac772a3c07cad66df46b2781f39121be211d383e Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 30 Jan 2023 09:48:39 +0530
Subject: [PATCH 1235/1791] drm/amdgpu: Add fallback path for discovery info

If SOC doesn't expose dedicated vram, discovery region may be
available through system memory. Rename the existing interface to
generic read_binary_from_mem and add a fallback path to read from system
memory.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 21 ++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 246070938c41c..01b21988c1aed 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -226,13 +226,21 @@ static int amdgpu_discovery_read_binary_from_sysmem(struct amdgpu_device *adev,
 	return -ENOENT;
 }
 
-static void amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary)
+static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev,
+						 uint8_t *binary)
 {
 	uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20;
-	uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET;
+	int ret = 0;
 
-	amdgpu_device_vram_access(adev, pos, (uint32_t *)binary,
-				  adev->mman.discovery_tmr_size, false);
+	if (vram_size) {
+		uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET;
+		amdgpu_device_vram_access(adev, pos, (uint32_t *)binary,
+					  adev->mman.discovery_tmr_size, false);
+	} else {
+		ret = amdgpu_discovery_read_binary_from_sysmem(adev, binary);
+	}
+
+	return ret;
 }
 
 static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, uint8_t *binary)
@@ -338,7 +346,10 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev)
 		}
 
 	} else {
-		amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin);
+		r = amdgpu_discovery_read_binary_from_mem(
+			adev, adev->mman.discovery_bin);
+		if (r)
+			goto out;
 	}
 
 	/* check the ip discovery binary signature */
-- 
GitLab


From 75dda67c7213c3e0d17244a8c42547c27ee746f8 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 9 Feb 2023 18:23:16 -0500
Subject: [PATCH 1236/1791] drm/amdkfd: Flush TLB after unmapping for GFX
 v9.4.3

kfd_flush_tlb_after_unmap should return true for GFX v9.4.3, to do TLB
heavyweight flush after unmapping from GPU to guarantee that the GPU
will not access pages after they have been unmapped. This also helps
improve the mapping to GPU performance.

Without this, KFD accidently flush TLB after mapping to GPU because the
vm update sequence number is increased by previous unmapping.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index fb3cf2c51da80..6e1c15682c285 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1405,9 +1405,9 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type);
 
 static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev)
 {
-	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
-	       (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
-	       dev->adev->sdma.instance[0].fw_version >= 18) ||
+	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3) ||
+	       KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
+	       (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) && dev->sdma_fw_version >= 18) ||
 	       KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
 }
 
-- 
GitLab


From f2b8447b1f309901c3fdd4045febfe5cab545d87 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 10 Feb 2023 12:05:59 +0530
Subject: [PATCH 1237/1791] drm/amdgpu: Fix discovery sys node harvest info

Initalize syfs nodes after harvest information is fetched and fetch the
correct harvest info based on each IP instance.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 41 +++++++++++++++++--
 1 file changed, 38 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 01b21988c1aed..d81b2e1e8aee4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -877,6 +877,36 @@ static void ip_disc_release(struct kobject *kobj)
 	kfree(ip_top);
 }
 
+static uint8_t amdgpu_discovery_get_harvest_info(struct amdgpu_device *adev,
+						 uint16_t hw_id, uint8_t inst)
+{
+	uint8_t harvest = 0;
+
+	/* Until a uniform way is figured, get mask based on hwid */
+	switch (hw_id) {
+	case VCN_HWID:
+		harvest = (1 << inst) & adev->vcn.harvest_config;
+		break;
+	case DMU_HWID:
+		if (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)
+			harvest = 0x1;
+		break;
+	case UMC_HWID:
+		/* TODO: It needs another parsing; for now, ignore.*/
+		break;
+	case GC_HWID:
+		harvest = ((1 << inst) & adev->gfx.xcc_mask) == 0;
+		break;
+	case SDMA0_HWID:
+		harvest = ((1 << inst) & adev->sdma.sdma_mask) == 0;
+		break;
+	default:
+		break;
+	}
+
+	return harvest;
+}
+
 static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 				      struct ip_die_entry *ip_die_entry,
 				      const size_t _ip_offset, const int num_ips,
@@ -949,7 +979,10 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev,
 			ip_hw_instance->major = ip->major;
 			ip_hw_instance->minor = ip->minor;
 			ip_hw_instance->revision = ip->revision;
-			ip_hw_instance->harvest = ip->variant;
+			ip_hw_instance->harvest =
+				amdgpu_discovery_get_harvest_info(
+					adev, ip_hw_instance->hw_id,
+					ip_hw_instance->num_instance);
 			ip_hw_instance->num_base_addresses = ip->num_base_address;
 
 			for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++) {
@@ -1035,6 +1068,9 @@ static int amdgpu_discovery_sysfs_init(struct amdgpu_device *adev)
 	struct kset *die_kset;
 	int res, ii;
 
+	if (!adev->mman.discovery_bin)
+		return -EINVAL;
+
 	adev->ip_top = kzalloc(sizeof(*adev->ip_top), GFP_KERNEL);
 	if (!adev->ip_top)
 		return -ENOMEM;
@@ -1282,8 +1318,6 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 		}
 	}
 
-	amdgpu_discovery_sysfs_init(adev);
-
 	return 0;
 }
 
@@ -2224,6 +2258,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 	}
 
 	amdgpu_discovery_init_soc_config(adev);
+	amdgpu_discovery_sysfs_init(adev);
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 0, 1):
-- 
GitLab


From 5ae0ec8b8045d72467d4e7417b34a5ab2fa72138 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 13 Feb 2023 15:45:42 +0800
Subject: [PATCH 1238/1791] drm/amdgpu: preserve the num_links in case of
 reflection

For topology reflection, each socket to every other socket has the
exactly same topology info as the other way around. So it is safe
to keep the reflected num_links value otherwise it will be overriden
by the link info output of GET_PEER_LINKS command.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 4395c53d09d83..ea47012795e7f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1402,7 +1402,8 @@ int psp_xgmi_get_topology_info(struct psp_context *psp,
 			topology->nodes[i].num_links = get_extended_data ?
 					topology->nodes[i].num_links +
 							link_info_output->nodes[i].num_links :
-					link_info_output->nodes[i].num_links;
+					((requires_reflection && topology->nodes[i].num_links) ? topology->nodes[i].num_links :
+					 link_info_output->nodes[i].num_links);
 
 			/* reflect the topology information for bi-directionality */
 			if (requires_reflection && topology->nodes[i].num_hops)
-- 
GitLab


From 48d19834ea551af2932090ff6de04730007e9876 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonjiang@amd.com>
Date: Wed, 15 Feb 2023 13:56:57 -0500
Subject: [PATCH 1239/1791] drm/amdgpu: Load vcn_v4_0_3 ucode during early_init

VCN loading ucode is moved to early_init with using 'amdgpu_ucode_*'
helpers.

Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Sonny Jiang <sonjiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 746df23b2eaa6..308dfe80a87c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -69,7 +69,7 @@ static int vcn_v4_0_3_early_init(void *handle)
 	vcn_v4_0_3_set_unified_ring_funcs(adev);
 	vcn_v4_0_3_set_irq_funcs(adev);
 
-	return 0;
+	return amdgpu_vcn_early_init(adev);
 }
 
 /**
-- 
GitLab


From aaf1090a6cb66e8b6d9da63ee983604a7eca8ffd Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 17 Feb 2023 18:27:51 +0530
Subject: [PATCH 1240/1791] drm/amdgpu: Add instance mask for VCN and JPEG

Keep an instance mask formed by physical instance numbers for VCN and JPEG
IPs. Populate the mask from discovery table information.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 30 ++++++++++++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h      |  2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h       |  1 +
 3 files changed, 29 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index d81b2e1e8aee4..ae5852f805497 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -549,10 +549,19 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev,
 				switch (le16_to_cpu(ip->hw_id)) {
 				case VCN_HWID:
 					(*vcn_harvest_count)++;
-					if (ip->instance_number == 0)
+					if (ip->instance_number == 0) {
 						adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN0;
-					else
+						adev->vcn.inst_mask &=
+							~AMDGPU_VCN_HARVEST_VCN0;
+						adev->jpeg.inst_mask &=
+							~AMDGPU_VCN_HARVEST_VCN0;
+					} else {
 						adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1;
+						adev->vcn.inst_mask &=
+							~AMDGPU_VCN_HARVEST_VCN1;
+						adev->jpeg.inst_mask &=
+							~AMDGPU_VCN_HARVEST_VCN1;
+					}
 					break;
 				case DMU_HWID:
 					adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK;
@@ -601,6 +610,11 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev,
 				(1 << harvest_info->list[i].number_instance);
 			adev->jpeg.harvest_config |=
 				(1 << harvest_info->list[i].number_instance);
+
+			adev->vcn.inst_mask &=
+				~(1U << harvest_info->list[i].number_instance);
+			adev->jpeg.inst_mask &=
+				~(1U << harvest_info->list[i].number_instance);
 			break;
 		case DMU_HWID:
 			adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK;
@@ -1188,6 +1202,8 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 
 	adev->gfx.xcc_mask = 0;
 	adev->sdma.sdma_mask = 0;
+	adev->vcn.inst_mask = 0;
+	adev->jpeg.inst_mask = 0;
 	bhdr = (struct binary_header *)adev->mman.discovery_bin;
 	ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin +
 			le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset));
@@ -1235,12 +1251,18 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 				adev->vcn.vcn_config[adev->vcn.num_vcn_inst] =
 					ip->revision & 0xc0;
 				ip->revision &= ~0xc0;
-				if (adev->vcn.num_vcn_inst < AMDGPU_MAX_VCN_INSTANCES)
+				if (adev->vcn.num_vcn_inst <
+				    AMDGPU_MAX_VCN_INSTANCES) {
 					adev->vcn.num_vcn_inst++;
-				else
+					adev->vcn.inst_mask |=
+						(1U << ip->instance_number);
+					adev->jpeg.inst_mask |=
+						(1U << ip->instance_number);
+				} else {
 					dev_err(adev->dev, "Too many VCN instances: %d vs %d\n",
 						adev->vcn.num_vcn_inst + 1,
 						AMDGPU_MAX_VCN_INSTANCES);
+				}
 			}
 			if (le16_to_cpu(ip->hw_id) == SDMA0_HWID ||
 			    le16_to_cpu(ip->hw_id) == SDMA1_HWID ||
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index bb700a2b97c21..90516f623f561 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -59,6 +59,8 @@ struct amdgpu_jpeg {
 	atomic_t total_submission_cnt;
 	struct ras_common_if	*ras_if;
 	struct amdgpu_jpeg_ras	*ras;
+
+	uint16_t inst_mask;
 	uint8_t num_inst_per_aid;
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 6ea3f076257e3..47463ef10fce2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -273,6 +273,7 @@ struct amdgpu_vcn {
 	struct ras_common_if    *ras_if;
 	struct amdgpu_vcn_ras   *ras;
 
+	uint16_t inst_mask;
 	uint8_t	num_inst_per_aid;
 };
 
-- 
GitLab


From 07bc0ac8ff49c9868a66526634fbc21cb194afca Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 17 Feb 2023 18:44:20 +0530
Subject: [PATCH 1241/1791] drm/amdgpu: Add VCN logical to physical id mapping

Add mappings for logical to physical id for VCN/JPEG 4.0.3

v2: make local function static (Alex)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c    | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index fdc728f678d71..90fe77db9bee5 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -82,6 +82,21 @@ static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 	return dev_inst;
 }
 
+static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev,
+					  enum amd_hw_ip_block_type ip_block,
+					  uint32_t inst_mask)
+{
+	int l = 0, i;
+
+	while (inst_mask) {
+		i = ffs(inst_mask) - 1;
+		adev->ip_map.dev_inst[ip_block][l++] = i;
+		inst_mask &= ~(1 << i);
+	}
+	for (; l < HWIP_MAX_INSTANCE; l++)
+		adev->ip_map.dev_inst[ip_block][l] = -1;
+}
+
 void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 {
 	int xcc_mask, sdma_mask;
@@ -108,6 +123,9 @@ void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 	for (; l < HWIP_MAX_INSTANCE; l++)
 		adev->ip_map.dev_inst[SDMA0_HWIP][l] = -1;
 
+	/* This covers both VCN and JPEG, JPEG is only alias of VCN */
+	aqua_vanjaram_populate_ip_map(adev, VCN_HWIP, adev->vcn.inst_mask);
+
 	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
 }
 
-- 
GitLab


From fd91d38b5275959a5b0804d4b4dbc5a4c0a8aac9 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 17 Feb 2023 19:41:06 +0530
Subject: [PATCH 1242/1791] drm/amdgpu: Use logical ids for VCN/JPEG v4.0.3

Address VCN/JPEG instances using logical ids. Whenever register access is
required, get the physical instance using GET_INST.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h       |  29 +-
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   |  14 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c      | 251 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c       | 431 ++++++++++--------
 4 files changed, 408 insertions(+), 317 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 47463ef10fce2..1eb9ccd1d83dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -141,18 +141,23 @@
 		RREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_DATA);				\
 	})
 
-#define WREG32_SOC15_DPG_MODE(inst_idx, offset, value, mask_en, indirect)			\
-	do {											\
-		if (!indirect) {								\
-			WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_DATA, value);			\
-			WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_CTL, 				\
-				(0x1 << UVD_DPG_LMA_CTL__READ_WRITE__SHIFT |			\
-				 mask_en << UVD_DPG_LMA_CTL__MASK_EN__SHIFT |			\
-				 offset << UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT));		\
-		} else {									\
-			*adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = offset;		\
-			*adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = value;			\
-		}										\
+#define WREG32_SOC15_DPG_MODE(inst_idx, offset, value, mask_en, indirect)             \
+	do {                                                                          \
+		if (!indirect) {                                                      \
+			WREG32_SOC15(VCN, GET_INST(VCN, inst_idx),                    \
+				     mmUVD_DPG_LMA_DATA, value);                      \
+			WREG32_SOC15(                                                 \
+				VCN, GET_INST(VCN, inst_idx),                         \
+				mmUVD_DPG_LMA_CTL,                                    \
+				(0x1 << UVD_DPG_LMA_CTL__READ_WRITE__SHIFT |          \
+				 mask_en << UVD_DPG_LMA_CTL__MASK_EN__SHIFT |         \
+				 offset << UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT)); \
+		} else {                                                              \
+			*adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ =              \
+				offset;                                               \
+			*adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ =              \
+				value;                                                \
+		}                                                                     \
 	} while (0)
 
 #define AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE (1 << 2)
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 90fe77db9bee5..51d3cb81e37a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -70,6 +70,8 @@ static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 	switch (block) {
 	case GC_HWIP:
 	case SDMA0_HWIP:
+	/* Both JPEG and VCN as JPEG is only alias of VCN */
+	case VCN_HWIP:
 		dev_inst = adev->ip_map.dev_inst[block][inst];
 		break;
 	default:
@@ -379,7 +381,7 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
 int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 {
 	u32 mask, inst_mask = adev->sdma.sdma_mask;
-	int ret, i, num_inst;
+	int ret, i;
 
 	/* generally 1 AID supports 4 instances */
 	adev->sdma.num_inst_per_aid = 4;
@@ -394,11 +396,15 @@ int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev)
 			adev->aid_mask |= (1 << i);
 	}
 
-	num_inst = hweight32(adev->aid_mask);
+	/* Harvest config is not used for aqua vanjaram. VCN and JPEGs will be
+	 * addressed based on logical instance ids.
+	 */
+	adev->vcn.harvest_config = 0;
 	adev->vcn.num_inst_per_aid = 1;
-	adev->vcn.num_vcn_inst = adev->vcn.num_inst_per_aid * num_inst;
+	adev->vcn.num_vcn_inst = hweight32(adev->vcn.inst_mask);
+	adev->jpeg.harvest_config = 0;
 	adev->jpeg.num_inst_per_aid = 1;
-	adev->jpeg.num_jpeg_inst = adev->jpeg.num_inst_per_aid * num_inst;
+	adev->jpeg.num_jpeg_inst = hweight32(adev->jpeg.inst_mask);
 
 	ret = aqua_vanjaram_xcp_mgr_init(adev);
 	if (ret)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index aa14a6619e9a1..c0e90e27f24bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -82,7 +82,7 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int i, j, r;
+	int i, j, r, jpeg_inst;
 
 	for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 		/* JPEG TRAP */
@@ -101,14 +101,15 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 		return r;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
+		jpeg_inst = GET_INST(JPEG, i);
+
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			ring = &adev->jpeg.inst[i].ring_dec[j];
 			ring->use_doorbell = true;
 			ring->vm_hub = AMDGPU_MMHUB0(adev->jpeg.inst[i].aid_id);
 			ring->doorbell_index =
-				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + j + 9 * i;
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+				1 + j + 9 * jpeg_inst;
 			sprintf(ring->name, "jpeg_dec_%d.%d", i, j);
 			r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 						AMDGPU_RING_PRIO_DEFAULT, NULL);
@@ -118,8 +119,10 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 			adev->jpeg.internal.jpeg_pitch[j] =
 				regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET;
 			adev->jpeg.inst[i].external.jpeg_pitch[j] =
-				SOC15_REG_OFFSET1(JPEG, i, regUVD_JRBC0_UVD_JRBC_SCRATCH0,
-				(j?(0x40 * j - 0xc80):0));
+				SOC15_REG_OFFSET1(
+					JPEG, jpeg_inst,
+					regUVD_JRBC0_UVD_JRBC_SCRATCH0,
+					(j ? (0x40 * j - 0xc80) : 0));
 		}
 	}
 
@@ -157,25 +160,30 @@ static int jpeg_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int i, j, r;
+	int i, j, r, jpeg_inst;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
+		jpeg_inst = GET_INST(JPEG, i);
+
 		ring = adev->jpeg.inst[i].ring_dec;
 
 		if (ring->use_doorbell)
-			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i,
+			adev->nbio.funcs->vcn_doorbell_range(
+				adev, ring->use_doorbell,
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+					9 * jpeg_inst,
 				adev->jpeg.inst[i].aid_id);
 
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			ring = &adev->jpeg.inst[i].ring_dec[j];
 			if (ring->use_doorbell)
-				WREG32_SOC15_OFFSET(VCN, i, regVCN_JPEG_DB_CTRL,
-					(ring->pipe?(ring->pipe - 0x15):0),
-					ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
-					VCN_JPEG_DB_CTRL__EN_MASK);
+				WREG32_SOC15_OFFSET(
+					VCN, GET_INST(VCN, i),
+					regVCN_JPEG_DB_CTRL,
+					(ring->pipe ? (ring->pipe - 0x15) : 0),
+					ring->doorbell_index
+							<< VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
+						VCN_JPEG_DB_CTRL__EN_MASK);
 			r = amdgpu_ring_test_helper(ring);
 			if (r)
 				return r;
@@ -250,10 +258,11 @@ static int jpeg_v4_0_3_resume(void *handle)
 
 static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
+	int i, jpeg_inst;
 	uint32_t data;
-	int i;
 
-	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL);
+	jpeg_inst = GET_INST(JPEG, inst_idx);
+	data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) {
 		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 		data &= (~(JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1));
@@ -263,21 +272,22 @@ static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int ins
 
 	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL, data);
+	WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL, data);
 
-	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE);
+	data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE);
 	data &= ~(JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
 		data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
-	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE, data);
+	WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE, data);
 }
 
 static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
+	int i, jpeg_inst;
 	uint32_t data;
-	int i;
 
-	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL);
+	jpeg_inst = GET_INST(JPEG, inst_idx);
+	data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL);
 	if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) {
 		data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 		data |= (JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1);
@@ -287,13 +297,13 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst
 
 	data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_CTRL, data);
+	WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL, data);
 
-	data = RREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE);
+	data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE);
 	data |= (JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK);
 	for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i)
 		data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK << i);
-	WREG32_SOC15(JPEG, inst_idx, regJPEG_CGC_GATE, data);
+	WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE, data);
 }
 
 /**
@@ -306,34 +316,36 @@ static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst
 static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 {
 	struct amdgpu_ring *ring;
-	int i, j;
+	int i, j, jpeg_inst;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
-		WREG32_SOC15(JPEG, i, regUVD_PGFSM_CONFIG,
-			1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
-		SOC15_WAIT_ON_RREG(JPEG, i, regUVD_PGFSM_STATUS,
-			UVD_PGFSM_STATUS__UVDJ_PWR_ON <<
-			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+		jpeg_inst = GET_INST(JPEG, i);
+
+		WREG32_SOC15(JPEG, jpeg_inst, regUVD_PGFSM_CONFIG,
+			     1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+		SOC15_WAIT_ON_RREG(
+			JPEG, jpeg_inst, regUVD_PGFSM_STATUS,
+			UVD_PGFSM_STATUS__UVDJ_PWR_ON
+				<< UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
 			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
 
 		/* disable anti hang mechanism */
-		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JPEG_POWER_STATUS), 0,
-			~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+		WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst,
+					  regUVD_JPEG_POWER_STATUS),
+			 0, ~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
 
 		/* JPEG disable CGC */
 		jpeg_v4_0_3_disable_clock_gating(adev, i);
 
 		/* MJPEG global tiling registers */
-		WREG32_SOC15(JPEG, i, regJPEG_DEC_GFX8_ADDR_CONFIG,
-			adev->gfx.config.gb_addr_config);
-		WREG32_SOC15(JPEG, i, regJPEG_DEC_GFX10_ADDR_CONFIG,
-			adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(JPEG, jpeg_inst, regJPEG_DEC_GFX8_ADDR_CONFIG,
+			     adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(JPEG, jpeg_inst, regJPEG_DEC_GFX10_ADDR_CONFIG,
+			     adev->gfx.config.gb_addr_config);
 
 		/* enable JMI channel */
-		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JMI_CNTL), 0,
-			~UVD_JMI_CNTL__SOFT_RESET_MASK);
+		WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, regUVD_JMI_CNTL), 0,
+			 ~UVD_JMI_CNTL__SOFT_RESET_MASK);
 
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
@@ -341,25 +353,40 @@ static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
 			ring = &adev->jpeg.inst[i].ring_dec[j];
 
 			/* enable System Interrupt for JRBC */
-			WREG32_P(SOC15_REG_OFFSET(JPEG, i, regJPEG_SYS_INT_EN),
-				JPEG_SYS_INT_EN__DJRBC0_MASK << j,
-				~(JPEG_SYS_INT_EN__DJRBC0_MASK << j));
-
-			WREG32_SOC15_OFFSET(JPEG, i,
-				regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, reg_offset, 0);
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
-				(0x00000001L | 0x00000002L));
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
+			WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst,
+						  regJPEG_SYS_INT_EN),
+				 JPEG_SYS_INT_EN__DJRBC0_MASK << j,
+				 ~(JPEG_SYS_INT_EN__DJRBC0_MASK << j));
+
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JMI0_UVD_LMI_JRBC_RB_VMID,
+					    reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JRBC0_UVD_JRBC_RB_CNTL,
+					    reg_offset,
+					    (0x00000001L | 0x00000002L));
+			WREG32_SOC15_OFFSET(
+				JPEG, jpeg_inst,
+				regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW,
 				reg_offset, lower_32_bits(ring->gpu_addr));
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
+			WREG32_SOC15_OFFSET(
+				JPEG, jpeg_inst,
+				regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH,
 				reg_offset, upper_32_bits(ring->gpu_addr));
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_RPTR, reg_offset, 0);
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_WPTR, reg_offset, 0);
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_CNTL, reg_offset,
-				0x00000002L);
-			WREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_SIZE, reg_offset,
-				ring->ring_size / 4);
-			ring->wptr = RREG32_SOC15_OFFSET(JPEG, i, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JRBC0_UVD_JRBC_RB_RPTR,
+					    reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+					    reg_offset, 0);
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JRBC0_UVD_JRBC_RB_CNTL,
+					    reg_offset, 0x00000002L);
+			WREG32_SOC15_OFFSET(JPEG, jpeg_inst,
+					    regUVD_JRBC0_UVD_JRBC_RB_SIZE,
+					    reg_offset, ring->ring_size / 4);
+			ring->wptr = RREG32_SOC15_OFFSET(
+				JPEG, jpeg_inst, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
 				reg_offset);
 		}
 	}
@@ -376,29 +403,29 @@ static int jpeg_v4_0_3_start(struct amdgpu_device *adev)
  */
 static int jpeg_v4_0_3_stop(struct amdgpu_device *adev)
 {
-	int i;
+	int i, jpeg_inst;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
-
+		jpeg_inst = GET_INST(JPEG, i);
 		/* reset JMI */
-		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JMI_CNTL),
-			UVD_JMI_CNTL__SOFT_RESET_MASK,
-			~UVD_JMI_CNTL__SOFT_RESET_MASK);
+		WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, regUVD_JMI_CNTL),
+			 UVD_JMI_CNTL__SOFT_RESET_MASK,
+			 ~UVD_JMI_CNTL__SOFT_RESET_MASK);
 
 		jpeg_v4_0_3_enable_clock_gating(adev, i);
 
 		/* enable anti hang mechanism */
-		WREG32_P(SOC15_REG_OFFSET(JPEG, i, regUVD_JPEG_POWER_STATUS),
-			UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK,
-			~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
-
-		WREG32_SOC15(JPEG, i, regUVD_PGFSM_CONFIG,
-			2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
-		SOC15_WAIT_ON_RREG(JPEG, i, regUVD_PGFSM_STATUS,
-			UVD_PGFSM_STATUS__UVDJ_PWR_OFF <<
-			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
+		WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst,
+					  regUVD_JPEG_POWER_STATUS),
+			 UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK,
+			 ~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK);
+
+		WREG32_SOC15(JPEG, jpeg_inst, regUVD_PGFSM_CONFIG,
+			     2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT);
+		SOC15_WAIT_ON_RREG(
+			JPEG, jpeg_inst, regUVD_PGFSM_STATUS,
+			UVD_PGFSM_STATUS__UVDJ_PWR_OFF
+				<< UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT,
 			UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK);
 	}
 
@@ -416,8 +443,9 @@ static uint64_t jpeg_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 
-	return RREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_RPTR,
-			ring->pipe?(0x40 * ring->pipe - 0xc80):0);
+	return RREG32_SOC15_OFFSET(
+		JPEG, GET_INST(JPEG, ring->me), regUVD_JRBC0_UVD_JRBC_RB_RPTR,
+		ring->pipe ? (0x40 * ring->pipe - 0xc80) : 0);
 }
 
 /**
@@ -434,8 +462,10 @@ static uint64_t jpeg_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring)
 	if (ring->use_doorbell)
 		return adev->wb.wb[ring->wptr_offs];
 	else
-		return RREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
-			ring->pipe?(0x40 * ring->pipe - 0xc80):0);
+		return RREG32_SOC15_OFFSET(
+			JPEG, GET_INST(JPEG, ring->me),
+			regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+			ring->pipe ? (0x40 * ring->pipe - 0xc80) : 0);
 }
 
 /**
@@ -453,8 +483,11 @@ static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
 		adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr);
 		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
 	} else {
-		WREG32_SOC15_OFFSET(JPEG, ring->me, regUVD_JRBC0_UVD_JRBC_RB_WPTR,
-			(ring->pipe?(0x40 * ring->pipe - 0xc80):0), lower_32_bits(ring->wptr));
+		WREG32_SOC15_OFFSET(JPEG, GET_INST(JPEG, ring->me),
+				    regUVD_JRBC0_UVD_JRBC_RB_WPTR,
+				    (ring->pipe ? (0x40 * ring->pipe - 0xc80) :
+						  0),
+				    lower_32_bits(ring->wptr));
 	}
 }
 
@@ -703,15 +736,15 @@ static bool jpeg_v4_0_3_is_idle(void *handle)
 	int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
 
-			ret &= ((RREG32_SOC15_OFFSET(JPEG, i,
-					regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset) &
-					UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
-					UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
+			ret &= ((RREG32_SOC15_OFFSET(
+					 JPEG, GET_INST(JPEG, i),
+					 regUVD_JRBC0_UVD_JRBC_STATUS,
+					 reg_offset) &
+				 UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) ==
+				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
 		}
 	}
 
@@ -725,12 +758,11 @@ static int jpeg_v4_0_3_wait_for_idle(void *handle)
 	int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			unsigned int reg_offset = (j?(0x40 * j - 0xc80):0);
 
-			ret &= SOC15_WAIT_ON_RREG_OFFSET(JPEG, i,
+			ret &= SOC15_WAIT_ON_RREG_OFFSET(
+				JPEG, GET_INST(JPEG, i),
 				regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset,
 				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK,
 				UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK);
@@ -747,8 +779,6 @@ static int jpeg_v4_0_3_set_clockgating_state(void *handle,
 	int i;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
 		if (enable) {
 			if (!jpeg_v4_0_3_is_idle(handle))
 				return -EBUSY;
@@ -792,35 +822,46 @@ static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
-	uint32_t i;
+	uint32_t i, inst;
 
 	i = node_id_to_phys_map[entry->node_id];
 	DRM_DEV_DEBUG(adev->dev, "IH: JPEG TRAP\n");
 
+	for (inst = 0; inst < adev->jpeg.num_jpeg_inst; ++inst)
+		if (adev->jpeg.inst[inst].aid_id == i)
+			break;
+
+	if (inst >= adev->jpeg.num_jpeg_inst) {
+		dev_WARN_ONCE(adev->dev, 1,
+			      "Interrupt received for unknown JPEG instance %d",
+			      entry->node_id);
+		return 0;
+	}
+
 	switch (entry->src_id) {
 	case VCN_4_0__SRCID__JPEG_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[0]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[0]);
 		break;
 	case VCN_4_0__SRCID__JPEG1_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[1]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[1]);
 		break;
 	case VCN_4_0__SRCID__JPEG2_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[2]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[2]);
 		break;
 	case VCN_4_0__SRCID__JPEG3_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[3]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[3]);
 		break;
 	case VCN_4_0__SRCID__JPEG4_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[4]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[4]);
 		break;
 	case VCN_4_0__SRCID__JPEG5_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[5]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[5]);
 		break;
 	case VCN_4_0__SRCID__JPEG6_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[6]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[6]);
 		break;
 	case VCN_4_0__SRCID__JPEG7_DECODE:
-		amdgpu_fence_process(&adev->jpeg.inst[i].ring_dec[7]);
+		amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[7]);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -882,17 +923,17 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 
 static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev)
 {
-	int i, j;
+	int i, j, jpeg_inst;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			adev->jpeg.inst[i].ring_dec[j].funcs = &jpeg_v4_0_3_dec_ring_vm_funcs;
 			adev->jpeg.inst[i].ring_dec[j].me = i;
 			adev->jpeg.inst[i].ring_dec[j].pipe = j;
 		}
-		adev->jpeg.inst[i].aid_id = i / adev->jpeg.num_inst_per_aid;
+		jpeg_inst = GET_INST(JPEG, i);
+		adev->jpeg.inst[i].aid_id =
+			jpeg_inst / adev->jpeg.num_inst_per_aid;
 	}
 	DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n");
 }
@@ -907,8 +948,6 @@ static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 	int i;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-		if (adev->jpeg.harvest_config & (1 << i))
-			continue;
 		adev->jpeg.inst->irq.num_types += adev->jpeg.num_jpeg_rings;
 	}
 	adev->jpeg.inst->irq.funcs = &jpeg_v4_0_3_irq_funcs;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 308dfe80a87c7..49b07843efd14 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -83,7 +83,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int i, r;
+	int i, r, vcn_inst;
 
 	r = amdgpu_vcn_sw_init(adev);
 	if (r)
@@ -104,12 +104,13 @@ static int vcn_v4_0_3_sw_init(void *handle)
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
+		vcn_inst = GET_INST(VCN, i);
 
 		ring = &adev->vcn.inst[i].ring_enc[0];
 		ring->use_doorbell = true;
-		ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i;
+		ring->doorbell_index =
+			(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+			9 * vcn_inst;
 		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
 		sprintf(ring->name, "vcn_unified_%d", i);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
@@ -148,8 +149,6 @@ static int vcn_v4_0_3_sw_fini(void *handle)
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 			volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 
-			if (adev->vcn.harvest_config & (1 << i))
-				continue;
 			fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 			fw_shared->present_flag_0 = 0;
 			fw_shared->sq.is_enabled = cpu_to_le32(false);
@@ -177,21 +176,25 @@ static int vcn_v4_0_3_hw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	struct amdgpu_ring *ring;
-	int i, r;
+	int i, r, vcn_inst;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
+		vcn_inst = GET_INST(VCN, i);
 		ring = &adev->vcn.inst[i].ring_enc[0];
 
 		if (ring->use_doorbell) {
-			adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-					(adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 9 * i,
-					adev->vcn.inst[i].aid_id);
-
-			WREG32_SOC15(VCN, ring->me, regVCN_RB1_DB_CTRL,
-				ring->doorbell_index << VCN_RB1_DB_CTRL__OFFSET__SHIFT |
-				VCN_RB1_DB_CTRL__EN_MASK);
+			adev->nbio.funcs->vcn_doorbell_range(
+				adev, ring->use_doorbell,
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+					9 * vcn_inst,
+				adev->vcn.inst[i].aid_id);
+
+			WREG32_SOC15(
+				VCN, GET_INST(VCN, ring->me),
+				regVCN_RB1_DB_CTRL,
+				ring->doorbell_index
+						<< VCN_RB1_DB_CTRL__OFFSET__SHIFT |
+					VCN_RB1_DB_CTRL__EN_MASK);
 		}
 
 		r = amdgpu_ring_test_helper(ring);
@@ -278,54 +281,67 @@ static int vcn_v4_0_3_resume(void *handle)
  */
 static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev, int inst_idx)
 {
-	uint32_t offset, size;
+	uint32_t offset, size, vcn_inst;
 	const struct common_firmware_header *hdr;
 
 	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
 	size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8);
 
+	vcn_inst = GET_INST(VCN, inst_idx);
 	/* cache window 0: fw */
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
-		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
-			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_lo));
-		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
-			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx].tmr_mc_addr_hi));
-		WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET0, 0);
+		WREG32_SOC15(
+			VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx]
+				 .tmr_mc_addr_lo));
+		WREG32_SOC15(
+			VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			(adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx]
+				 .tmr_mc_addr_hi));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET0, 0);
 		offset = 0;
 	} else {
-		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
-			lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
-		WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
-			upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW,
+			     lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
+		WREG32_SOC15(VCN, vcn_inst,
+			     regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH,
+			     upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr));
 		offset = size;
-		WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET0,
-			AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET0,
+			     AMDGPU_UVD_FIRMWARE_OFFSET >> 3);
 	}
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE0, size);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE0, size);
 
 	/* cache window 1: stack */
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW,
-		lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH,
-		upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET1, 0);
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE1, AMDGPU_VCN_STACK_SIZE);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW,
+		     lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH,
+		     upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET1, 0);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE1,
+		     AMDGPU_VCN_STACK_SIZE);
 
 	/* cache window 2: context */
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW,
-		lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH,
-		upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + AMDGPU_VCN_STACK_SIZE));
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_OFFSET2, 0);
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_CACHE_SIZE2, AMDGPU_VCN_CONTEXT_SIZE);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW,
+		     lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset +
+				   AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH,
+		     upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset +
+				   AMDGPU_VCN_STACK_SIZE));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET2, 0);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE2,
+		     AMDGPU_VCN_CONTEXT_SIZE);
 
 	/* non-cache window */
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW,
+	WREG32_SOC15(
+		VCN, vcn_inst, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW,
 		lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr));
-	WREG32_SOC15(VCN, inst_idx, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH,
+	WREG32_SOC15(
+		VCN, vcn_inst, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH,
 		upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr));
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_NONCACHE_OFFSET0, 0);
-	WREG32_SOC15(VCN, inst_idx, regUVD_VCPU_NONCACHE_SIZE0,
+	WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_NONCACHE_OFFSET0, 0);
+	WREG32_SOC15(
+		VCN, vcn_inst, regUVD_VCPU_NONCACHE_SIZE0,
 		AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)));
 }
 
@@ -454,18 +470,21 @@ static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_i
 static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
+	int vcn_inst;
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
 		return;
 
+	vcn_inst = GET_INST(VCN, inst_idx);
+
 	/* VCN disable CGC */
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL);
 	data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_GATE);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_GATE);
 	data &= ~(UVD_CGC_GATE__SYS_MASK
 		| UVD_CGC_GATE__MPEG2_MASK
 		| UVD_CGC_GATE__REGS_MASK
@@ -479,10 +498,10 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst
 		| UVD_CGC_GATE__VCPU_MASK
 		| UVD_CGC_GATE__MMSCH_MASK);
 
-	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_GATE, data);
-	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_CGC_GATE, 0,  0xFFFFFFFF);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_GATE, data);
+	SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_CGC_GATE, 0, 0xFFFFFFFF);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL);
 	data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK
 		| UVD_CGC_CTRL__MPEG2_MODE_MASK
 		| UVD_CGC_CTRL__REGS_MODE_MASK
@@ -495,9 +514,9 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst
 		| UVD_CGC_CTRL__WCB_MODE_MASK
 		| UVD_CGC_CTRL__VCPU_MODE_MASK
 		| UVD_CGC_CTRL__MMSCH_MODE_MASK);
-	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_GATE);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_GATE);
 	data |= (UVD_SUVD_CGC_GATE__SRE_MASK
 		| UVD_SUVD_CGC_GATE__SIT_MASK
 		| UVD_SUVD_CGC_GATE__SMP_MASK
@@ -519,9 +538,9 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst
 		| UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK
 		| UVD_SUVD_CGC_GATE__SDB_VP9_MASK
 		| UVD_SUVD_CGC_GATE__IME_HEVC_MASK);
-	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_GATE, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_GATE, data);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL);
 	data &= ~(UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
@@ -530,7 +549,7 @@ static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst
 		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
-	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL, data);
 }
 
 /**
@@ -595,18 +614,21 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
 static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t data;
+	int vcn_inst;
 
 	if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG)
 		return;
 
+	vcn_inst = GET_INST(VCN, inst_idx);
+
 	/* enable VCN CGC */
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL);
 	data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT;
 	data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT;
 	data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT;
-	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL);
 	data |= (UVD_CGC_CTRL__SYS_MODE_MASK
 		| UVD_CGC_CTRL__MPEG2_MODE_MASK
 		| UVD_CGC_CTRL__REGS_MODE_MASK
@@ -618,9 +640,9 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_
 		| UVD_CGC_CTRL__LRBBM_MODE_MASK
 		| UVD_CGC_CTRL__WCB_MODE_MASK
 		| UVD_CGC_CTRL__VCPU_MODE_MASK);
-	WREG32_SOC15(VCN, inst_idx, regUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data);
 
-	data = RREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL);
+	data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL);
 	data |= (UVD_SUVD_CGC_CTRL__SRE_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SIT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SMP_MODE_MASK
@@ -629,7 +651,7 @@ static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_
 		| UVD_SUVD_CGC_CTRL__ENT_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__IME_MODE_MASK
 		| UVD_SUVD_CGC_CTRL__SITE_MODE_MASK);
-	WREG32_SOC15(VCN, inst_idx, regUVD_SUVD_CGC_CTRL, data);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL, data);
 }
 
 /**
@@ -646,16 +668,18 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared =
 						adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
+	int vcn_inst;
 	uint32_t tmp;
 
+	vcn_inst = GET_INST(VCN, inst_idx);
 	/* disable register anti-hang mechanism */
-	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1,
-		~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+	WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_POWER_STATUS), 1,
+		 ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 	/* enable dynamic power gating mode */
-	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_POWER_STATUS);
+	tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_POWER_STATUS);
 	tmp |= UVD_POWER_STATUS__UVD_PG_MODE_MASK;
 	tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK;
-	WREG32_SOC15(VCN, inst_idx, regUVD_POWER_STATUS, tmp);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_POWER_STATUS, tmp);
 
 	if (indirect)
 		adev->vcn.inst[inst_idx].dpg_sram_curr_addr =
@@ -737,27 +761,28 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
 	/* program the RB_BASE for ring buffer */
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_LO,
-		lower_32_bits(ring->gpu_addr));
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_BASE_HI,
-		upper_32_bits(ring->gpu_addr));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_LO,
+		     lower_32_bits(ring->gpu_addr));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_HI,
+		     upper_32_bits(ring->gpu_addr));
 
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_SIZE, ring->ring_size / sizeof(uint32_t));
+	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_SIZE,
+		     ring->ring_size / sizeof(uint32_t));
 
 	/* resetting ring, fw should not check RB ring */
-	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
+	tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE);
 	tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK);
-	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
+	WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp);
 	fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET;
 
 	/* Initialize the ring buffer's read and write pointers */
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_RPTR, 0);
-	WREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR, 0);
-	ring->wptr = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0);
+	WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0);
+	ring->wptr = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR);
 
-	tmp = RREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE);
+	tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE);
 	tmp |= VCN_RB_ENABLE__RB_EN_MASK;
-	WREG32_SOC15(VCN, inst_idx, regVCN_RB_ENABLE, tmp);
+	WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp);
 	fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF);
 
 	/*resetting done, fw can check RB ring */
@@ -777,99 +802,101 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 {
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
+	int i, j, k, r, vcn_inst;
 	uint32_t tmp;
-	int i, j, k, r;
 
 	if (adev->pm.dpm_enabled)
 		amdgpu_dpm_enable_uvd(adev, true);
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
 		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
 			r = vcn_v4_0_3_start_dpg_mode(adev, i, adev->vcn.indirect_sram);
 			continue;
 		}
 
+		vcn_inst = GET_INST(VCN, i);
 		/* set VCN status busy */
-		tmp = RREG32_SOC15(VCN, i, regUVD_STATUS) | UVD_STATUS__UVD_BUSY;
-		WREG32_SOC15(VCN, i, regUVD_STATUS, tmp);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_STATUS) |
+		      UVD_STATUS__UVD_BUSY;
+		WREG32_SOC15(VCN, vcn_inst, regUVD_STATUS, tmp);
 
 		/*SW clock gating */
 		vcn_v4_0_3_disable_clock_gating(adev, i);
 
 		/* enable VCPU clock */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
-			UVD_VCPU_CNTL__CLK_EN_MASK, ~UVD_VCPU_CNTL__CLK_EN_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL),
+			 UVD_VCPU_CNTL__CLK_EN_MASK,
+			 ~UVD_VCPU_CNTL__CLK_EN_MASK);
 
 		/* disable master interrupt */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_MASTINT_EN), 0,
-			~UVD_MASTINT_EN__VCPU_EN_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_MASTINT_EN), 0,
+			 ~UVD_MASTINT_EN__VCPU_EN_MASK);
 
 		/* enable LMI MC and UMC channels */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_LMI_CTRL2), 0,
-			~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_LMI_CTRL2), 0,
+			 ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK);
 
-		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET);
 		tmp &= ~UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
 		tmp &= ~UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
-		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp);
 
 		/* setup regUVD_LMI_CTRL */
-		tmp = RREG32_SOC15(VCN, i, regUVD_LMI_CTRL);
-		WREG32_SOC15(VCN, i, regUVD_LMI_CTRL, tmp |
-			UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
-			UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
-			UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
-			UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL,
+			     tmp | UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK |
+				     UVD_LMI_CTRL__MASK_MC_URGENT_MASK |
+				     UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK |
+				     UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK);
 
 		/* setup regUVD_MPC_CNTL */
-		tmp = RREG32_SOC15(VCN, i, regUVD_MPC_CNTL);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_MPC_CNTL);
 		tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK;
 		tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT;
-		WREG32_SOC15(VCN, i, regUVD_MPC_CNTL, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_CNTL, tmp);
 
 		/* setup UVD_MPC_SET_MUXA0 */
-		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUXA0,
-			((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
-			(0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
-			(0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
-			(0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUXA0,
+			     ((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) |
+			      (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) |
+			      (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) |
+			      (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)));
 
 		/* setup UVD_MPC_SET_MUXB0 */
-		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUXB0,
-			((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
-			(0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
-			(0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
-			(0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUXB0,
+			     ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) |
+			      (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) |
+			      (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) |
+			      (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)));
 
 		/* setup UVD_MPC_SET_MUX */
-		WREG32_SOC15(VCN, i, regUVD_MPC_SET_MUX,
-			((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
-			(0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
-			(0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUX,
+			     ((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) |
+			      (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) |
+			      (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)));
 
 		vcn_v4_0_3_mc_resume(adev, i);
 
 		/* VCN global tiling registers */
-		WREG32_SOC15(VCN, i, regUVD_GFX8_ADDR_CONFIG,
-			adev->gfx.config.gb_addr_config);
-		WREG32_SOC15(VCN, i, regUVD_GFX10_ADDR_CONFIG,
-			adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_GFX8_ADDR_CONFIG,
+			     adev->gfx.config.gb_addr_config);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_GFX10_ADDR_CONFIG,
+			     adev->gfx.config.gb_addr_config);
 
 		/* unblock VCPU register access */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_RB_ARB_CTRL), 0,
-			~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_RB_ARB_CTRL), 0,
+			 ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
 
 		/* release VCPU reset to boot */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
-			~UVD_VCPU_CNTL__BLK_RST_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), 0,
+			 ~UVD_VCPU_CNTL__BLK_RST_MASK);
 
 		for (j = 0; j < 10; ++j) {
 			uint32_t status;
 
 			for (k = 0; k < 100; ++k) {
-				status = RREG32_SOC15(VCN, i, regUVD_STATUS);
+				status = RREG32_SOC15(VCN, vcn_inst,
+						      regUVD_STATUS);
 				if (status & 2)
 					break;
 				mdelay(10);
@@ -880,12 +907,14 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 
 			DRM_DEV_ERROR(adev->dev,
 				"VCN decode not responding, trying to reset the VCPU!!!\n");
-			WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
-				UVD_VCPU_CNTL__BLK_RST_MASK,
-				~UVD_VCPU_CNTL__BLK_RST_MASK);
+			WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst,
+						  regUVD_VCPU_CNTL),
+				 UVD_VCPU_CNTL__BLK_RST_MASK,
+				 ~UVD_VCPU_CNTL__BLK_RST_MASK);
 			mdelay(10);
-			WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
-				~UVD_VCPU_CNTL__BLK_RST_MASK);
+			WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst,
+						  regUVD_VCPU_CNTL),
+				 0, ~UVD_VCPU_CNTL__BLK_RST_MASK);
 
 			mdelay(10);
 			r = -1;
@@ -897,39 +926,40 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 		}
 
 		/* enable master interrupt */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_MASTINT_EN),
-			UVD_MASTINT_EN__VCPU_EN_MASK,
-			~UVD_MASTINT_EN__VCPU_EN_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_MASTINT_EN),
+			 UVD_MASTINT_EN__VCPU_EN_MASK,
+			 ~UVD_MASTINT_EN__VCPU_EN_MASK);
 
 		/* clear the busy bit of VCN_STATUS */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_STATUS), 0,
-			~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_STATUS), 0,
+			 ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT));
 
 		ring = &adev->vcn.inst[i].ring_enc[0];
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 
 		/* program the RB_BASE for ring buffer */
-		WREG32_SOC15(VCN, i, regUVD_RB_BASE_LO,
-			lower_32_bits(ring->gpu_addr));
-		WREG32_SOC15(VCN, i, regUVD_RB_BASE_HI,
-			upper_32_bits(ring->gpu_addr));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_LO,
+			     lower_32_bits(ring->gpu_addr));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_HI,
+			     upper_32_bits(ring->gpu_addr));
 
-		WREG32_SOC15(VCN, i, regUVD_RB_SIZE, ring->ring_size / sizeof(uint32_t));
+		WREG32_SOC15(VCN, vcn_inst, regUVD_RB_SIZE,
+			     ring->ring_size / sizeof(uint32_t));
 
 		/* resetting ring, fw should not check RB ring */
-		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE);
 		tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK);
-		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp);
 
 		/* Initialize the ring buffer's read and write pointers */
-		WREG32_SOC15(VCN, i, regUVD_RB_RPTR, 0);
-		WREG32_SOC15(VCN, i, regUVD_RB_WPTR, 0);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0);
 
-		tmp = RREG32_SOC15(VCN, i, regVCN_RB_ENABLE);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE);
 		tmp |= VCN_RB_ENABLE__RB_EN_MASK;
-		WREG32_SOC15(VCN, i, regVCN_RB_ENABLE, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp);
 
-		ring->wptr = RREG32_SOC15(VCN, i, regUVD_RB_WPTR);
+		ring->wptr = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR);
 		fw_shared->sq.queue_mode &=
 			cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF));
 
@@ -948,21 +978,24 @@ static int vcn_v4_0_3_start(struct amdgpu_device *adev)
 static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 {
 	uint32_t tmp;
+	int vcn_inst;
+
+	vcn_inst = GET_INST(VCN, inst_idx);
 
 	/* Wait for power status to be 1 */
-	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_POWER_STATUS, 1,
-		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+	SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_POWER_STATUS, 1,
+			   UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 	/* wait for read ptr to be equal to write ptr */
-	tmp = RREG32_SOC15(VCN, inst_idx, regUVD_RB_WPTR);
-	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_RB_RPTR, tmp, 0xFFFFFFFF);
+	tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR);
+	SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_RB_RPTR, tmp, 0xFFFFFFFF);
 
-	SOC15_WAIT_ON_RREG(VCN, inst_idx, regUVD_POWER_STATUS, 1,
-		UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
+	SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_POWER_STATUS, 1,
+			   UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 	/* disable dynamic power gating mode */
-	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 0,
-		~UVD_POWER_STATUS__UVD_PG_MODE_MASK);
+	WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_POWER_STATUS), 0,
+		 ~UVD_POWER_STATUS__UVD_PG_MODE_MASK);
 	return 0;
 }
 
@@ -976,12 +1009,11 @@ static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx)
 static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 {
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	int i, r = 0, vcn_inst;
 	uint32_t tmp;
-	int i, r = 0;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
+		vcn_inst = GET_INST(VCN, i);
 
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 		fw_shared->sq.queue_mode |= FW_QUEUE_DPG_HOLD_OFF;
@@ -992,7 +1024,8 @@ static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 		}
 
 		/* wait for vcn idle */
-		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_STATUS, UVD_STATUS__IDLE, 0x7);
+		r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_STATUS,
+				       UVD_STATUS__IDLE, 0x7);
 		if (r)
 			goto Done;
 
@@ -1000,45 +1033,47 @@ static int vcn_v4_0_3_stop(struct amdgpu_device *adev)
 			UVD_LMI_STATUS__READ_CLEAN_MASK |
 			UVD_LMI_STATUS__WRITE_CLEAN_MASK |
 			UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK;
-		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_LMI_STATUS, tmp, tmp);
+		r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_LMI_STATUS, tmp,
+				       tmp);
 		if (r)
 			goto Done;
 
 		/* stall UMC channel */
-		tmp = RREG32_SOC15(VCN, i, regUVD_LMI_CTRL2);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL2);
 		tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK;
-		WREG32_SOC15(VCN, i, regUVD_LMI_CTRL2, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL2, tmp);
 		tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK |
 			UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK;
-		r = SOC15_WAIT_ON_RREG(VCN, i, regUVD_LMI_STATUS, tmp, tmp);
+		r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_LMI_STATUS, tmp,
+				       tmp);
 		if (r)
 			goto Done;
 
 		/* Unblock VCPU Register access */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_RB_ARB_CTRL),
-			UVD_RB_ARB_CTRL__VCPU_DIS_MASK,
-			~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_RB_ARB_CTRL),
+			 UVD_RB_ARB_CTRL__VCPU_DIS_MASK,
+			 ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK);
 
 		/* release VCPU reset to boot */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL),
-			UVD_VCPU_CNTL__BLK_RST_MASK,
-			~UVD_VCPU_CNTL__BLK_RST_MASK);
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL),
+			 UVD_VCPU_CNTL__BLK_RST_MASK,
+			 ~UVD_VCPU_CNTL__BLK_RST_MASK);
 
 		/* disable VCPU clock */
-		WREG32_P(SOC15_REG_OFFSET(VCN, i, regUVD_VCPU_CNTL), 0,
-			~(UVD_VCPU_CNTL__CLK_EN_MASK));
+		WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), 0,
+			 ~(UVD_VCPU_CNTL__CLK_EN_MASK));
 
 		/* reset LMI UMC/LMI/VCPU */
-		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET);
 		tmp |= UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK;
-		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp);
 
-		tmp = RREG32_SOC15(VCN, i, regUVD_SOFT_RESET);
+		tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET);
 		tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK;
-		WREG32_SOC15(VCN, i, regUVD_SOFT_RESET, tmp);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp);
 
 		/* clear VCN status */
-		WREG32_SOC15(VCN, i, regUVD_STATUS, 0);
+		WREG32_SOC15(VCN, vcn_inst, regUVD_STATUS, 0);
 
 		/* apply HW clock gating */
 		vcn_v4_0_3_enable_clock_gating(adev, i);
@@ -1080,7 +1115,7 @@ static uint64_t vcn_v4_0_3_unified_ring_get_rptr(struct amdgpu_ring *ring)
 	if (ring != &adev->vcn.inst[ring->me].ring_enc[0])
 		DRM_ERROR("wrong ring id is identified in %s", __func__);
 
-	return RREG32_SOC15(VCN, ring->me, regUVD_RB_RPTR);
+	return RREG32_SOC15(VCN, GET_INST(VCN, ring->me), regUVD_RB_RPTR);
 }
 
 /**
@@ -1100,7 +1135,8 @@ static uint64_t vcn_v4_0_3_unified_ring_get_wptr(struct amdgpu_ring *ring)
 	if (ring->use_doorbell)
 		return *ring->wptr_cpu_addr;
 	else
-		return RREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR);
+		return RREG32_SOC15(VCN, GET_INST(VCN, ring->me),
+				    regUVD_RB_WPTR);
 }
 
 /**
@@ -1121,7 +1157,8 @@ static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring)
 		*ring->wptr_cpu_addr = lower_32_bits(ring->wptr);
 		WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr));
 	} else {
-		WREG32_SOC15(VCN, ring->me, regUVD_RB_WPTR, lower_32_bits(ring->wptr));
+		WREG32_SOC15(VCN, GET_INST(VCN, ring->me), regUVD_RB_WPTR,
+			     lower_32_bits(ring->wptr));
 	}
 }
 
@@ -1163,14 +1200,14 @@ static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = {
  */
 static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev)
 {
-	int i;
+	int i, vcn_inst;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
 		adev->vcn.inst[i].ring_enc[0].funcs = &vcn_v4_0_3_unified_ring_vm_funcs;
 		adev->vcn.inst[i].ring_enc[0].me = i;
-		adev->vcn.inst[i].aid_id = i / adev->vcn.num_inst_per_aid;
+		vcn_inst = GET_INST(VCN, i);
+		adev->vcn.inst[i].aid_id =
+			vcn_inst / adev->vcn.num_inst_per_aid;
 	}
 	DRM_DEV_INFO(adev->dev, "VCN decode is enabled in VM mode\n");
 }
@@ -1188,9 +1225,8 @@ static bool vcn_v4_0_3_is_idle(void *handle)
 	int i, ret = 1;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-		ret &= (RREG32_SOC15(VCN, i, regUVD_STATUS) == UVD_STATUS__IDLE);
+		ret &= (RREG32_SOC15(VCN, GET_INST(VCN, i), regUVD_STATUS) ==
+			UVD_STATUS__IDLE);
 	}
 
 	return ret;
@@ -1209,10 +1245,8 @@ static int vcn_v4_0_3_wait_for_idle(void *handle)
 	int i, ret = 0;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-		ret = SOC15_WAIT_ON_RREG(VCN, i, regUVD_STATUS, UVD_STATUS__IDLE,
-				UVD_STATUS__IDLE);
+		ret = SOC15_WAIT_ON_RREG(VCN, GET_INST(VCN, i), regUVD_STATUS,
+					 UVD_STATUS__IDLE, UVD_STATUS__IDLE);
 		if (ret)
 			return ret;
 	}
@@ -1235,10 +1269,9 @@ static int vcn_v4_0_3_set_clockgating_state(void *handle,
 	int i;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
 		if (enable) {
-			if (RREG32_SOC15(VCN, i, regUVD_STATUS) != UVD_STATUS__IDLE)
+			if (RREG32_SOC15(VCN, GET_INST(VCN, i),
+					 regUVD_STATUS) != UVD_STATUS__IDLE)
 				return -EBUSY;
 			vcn_v4_0_3_enable_clock_gating(adev, i);
 		} else {
@@ -1307,15 +1340,26 @@ static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
 {
-	uint32_t i;
+	uint32_t i, inst;
 
 	i = node_id_to_phys_map[entry->node_id];
 
 	DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n");
 
+	for (inst = 0; inst < adev->vcn.num_vcn_inst; ++inst)
+		if (adev->vcn.inst[inst].aid_id == i)
+			break;
+
+	if (inst >= adev->vcn.num_vcn_inst) {
+		dev_WARN_ONCE(adev->dev, 1,
+			      "Interrupt received for unknown VCN instance %d",
+			      entry->node_id);
+		return 0;
+	}
+
 	switch (entry->src_id) {
 	case VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE:
-		amdgpu_fence_process(&adev->vcn.inst[i].ring_enc[0]);
+		amdgpu_fence_process(&adev->vcn.inst[inst].ring_enc[0]);
 		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
@@ -1343,9 +1387,6 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 	int i;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
 		adev->vcn.inst->irq.num_types++;
 	}
 	adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs;
-- 
GitLab


From 6a944ccbf5f5059de1a9b3d48971a50cb5857ebf Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 27 Feb 2023 16:07:39 +0530
Subject: [PATCH 1243/1791] drm/amdgpu: Fix harvest reporting of VCN

Use VCN instance mask to check if an instance is harvested or not.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index ae5852f805497..caae6bf2fd306 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -313,6 +313,7 @@ static void amdgpu_discovery_harvest_config_quirk(struct amdgpu_device *adev)
 		case 0xCF:
 		case 0xDF:
 			adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1;
+			adev->vcn.inst_mask &= ~AMDGPU_VCN_HARVEST_VCN1;
 			break;
 		default:
 			break;
@@ -899,7 +900,7 @@ static uint8_t amdgpu_discovery_get_harvest_info(struct amdgpu_device *adev,
 	/* Until a uniform way is figured, get mask based on hwid */
 	switch (hw_id) {
 	case VCN_HWID:
-		harvest = (1 << inst) & adev->vcn.harvest_config;
+		harvest = ((1 << inst) & adev->vcn.inst_mask) == 0;
 		break;
 	case DMU_HWID:
 		if (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)
-- 
GitLab


From d3e53452b0f4cdd210432a268cabdbf65e98ddab Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 15 Mar 2023 03:54:55 -0400
Subject: [PATCH 1244/1791] drm/amdgpu/vcn: use dummy register selects AID for
 VCN_RAM ucode

Use dummy register 0xDEADBEEF selects AID for PSP VCN_RAM ucode.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 49b07843efd14..759f64a4acf46 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -681,9 +681,15 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 	tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK;
 	WREG32_SOC15(VCN, vcn_inst, regUVD_POWER_STATUS, tmp);
 
-	if (indirect)
+	if (indirect) {
+		DRM_DEV_DEBUG(adev->dev, "VCN %d start: on AID %d",
+			inst_idx, adev->vcn.inst[inst_idx].aid_id);
 		adev->vcn.inst[inst_idx].dpg_sram_curr_addr =
 				(uint32_t *)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr;
+		/* Use dummy register 0xDEADBEEF passing AID selection to PSP FW */
+		WREG32_SOC15_DPG_MODE(inst_idx, 0xDEADBEEF,
+			adev->vcn.inst[inst_idx].aid_id, 0, true);
+	}
 
 	/* enable clock gating */
 	vcn_v4_0_3_disable_clock_gating_dpg_mode(adev, 0, inst_idx, indirect);
-- 
GitLab


From 358e6c38300b7d2b7d7122d4fe485d8a4580dc1e Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 15 Mar 2023 04:09:25 -0400
Subject: [PATCH 1245/1791] drm/amdgpu: use physical AID index for ring name

Use physical AID index for VCN/JPEG ring name instead of
logical AID index.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index c0e90e27f24bf..ea9cb098a144a 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -110,7 +110,7 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 			ring->doorbell_index =
 				(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
 				1 + j + 9 * jpeg_inst;
-			sprintf(ring->name, "jpeg_dec_%d.%d", i, j);
+			sprintf(ring->name, "jpeg_dec_%d.%d", adev->jpeg.inst[i].aid_id, j);
 			r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0,
 						AMDGPU_RING_PRIO_DEFAULT, NULL);
 			if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 759f64a4acf46..b0e28d611f2d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -112,7 +112,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 			(adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
 			9 * vcn_inst;
 		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
-		sprintf(ring->name, "vcn_unified_%d", i);
+		sprintf(ring->name, "vcn_unified_%d", adev->vcn.inst[i].aid_id);
 		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
 				     AMDGPU_RING_PRIO_DEFAULT,
 				     &adev->vcn.inst[i].sched_score);
-- 
GitLab


From 26dc0448ef36ba83be43a7c4da94d55ec626db1a Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonny.jiang@amd.com>
Date: Mon, 20 Mar 2023 16:17:21 -0400
Subject: [PATCH 1246/1791] drm/amdgpu: A workaround for JPEG_v4_0_3 ring test
 fail

The jpeg_v4_0_3 jpeg_pitch register uses UVD_JRBC_SCRATCH0. It needs to
move WREG() to after jpeg_start.
Switch to a posted register write when doing the ring test to make sure
the register write lands before we test the result.

Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 388466a5f730e..9a1db2bd03e78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -124,11 +124,14 @@ int amdgpu_jpeg_dec_ring_test_ring(struct amdgpu_ring *ring)
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-	WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe], 0xCAFEDEAD);
 	r = amdgpu_ring_alloc(ring, 3);
 	if (r)
 		return r;
 
+	WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe], 0xCAFEDEAD);
+	/* Add a read register to make sure the write register is executed. */
+	RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]);
+
 	amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0));
 	amdgpu_ring_write(ring, 0xDEADBEEF);
 	amdgpu_ring_commit(ring);
-- 
GitLab


From 55ff23d9eb814dce8393a4c471259ded5a85d461 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonjiang@amd.com>
Date: Wed, 22 Mar 2023 15:49:29 -0400
Subject: [PATCH 1247/1791] drm/amdgpu: fixes a JPEG get write/read pointer bug

Need parentheses for the micro parameters.

Signed-off-by: Sonny Jiang <sonjiang@amd.com>
Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15_common.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index 1c9e924b5f8cd..3730c5ec202fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -71,7 +71,8 @@
 			 AMDGPU_REGS_NO_KIQ, ip##_HWIP)
 
 #define RREG32_SOC15_OFFSET(ip, inst, reg, offset) \
-	 __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, 0, ip##_HWIP)
+	 __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)) + \
+			 (offset), 0, ip##_HWIP)
 
 #define WREG32_SOC15(ip, inst, reg, value) \
 	 __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg), \
-- 
GitLab


From 96e693ad78a0778efb5256684a2f3712298f3e18 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonny.jiang@amd.com>
Date: Tue, 28 Mar 2023 13:19:11 -0400
Subject: [PATCH 1248/1791] drm/amdgpu: Add a read after write DB_CTRL for
 vcn_v4_0_3

To make sure VCN DB_CTRL is delivered before doorbell write.

Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index b0e28d611f2d1..9d0c3dc765472 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -195,6 +195,11 @@ static int vcn_v4_0_3_hw_init(void *handle)
 				ring->doorbell_index
 						<< VCN_RB1_DB_CTRL__OFFSET__SHIFT |
 					VCN_RB1_DB_CTRL__EN_MASK);
+
+			/* Read DB_CTRL to flush the write DB_CTRL command. */
+			RREG32_SOC15(
+				VCN, GET_INST(VCN, ring->me),
+				regVCN_RB1_DB_CTRL);
 		}
 
 		r = amdgpu_ring_test_helper(ring);
-- 
GitLab


From e7947c021a8b394677ab875288565d4dfcac779a Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonny.jiang@amd.com>
Date: Fri, 24 Mar 2023 17:44:59 -0400
Subject: [PATCH 1249/1791] drm/amdgpu: Use a different value than 0xDEADBEEF
 for jpeg ring test

The 0xDEADBEEF standard anti-hang value. Use it may cause
fake pass.

Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 9a1db2bd03e78..8c479669c459a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -133,12 +133,12 @@ int amdgpu_jpeg_dec_ring_test_ring(struct amdgpu_ring *ring)
 	RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]);
 
 	amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0));
-	amdgpu_ring_write(ring, 0xDEADBEEF);
+	amdgpu_ring_write(ring, 0xABADCAFE);
 	amdgpu_ring_commit(ring);
 
 	for (i = 0; i < adev->usec_timeout; i++) {
 		tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]);
-		if (tmp == 0xDEADBEEF)
+		if (tmp == 0xABADCAFE)
 			break;
 		udelay(1);
 	}
-- 
GitLab


From a3edd1ac706243fe5ca1c0925ce120b5a2661975 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 8 Mar 2023 12:41:22 +0530
Subject: [PATCH 1250/1791] drm/amdgpu/vcn: Use buffer object's deletion logic
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

VCN DPG buffer object is intialized to NULL. If allotted, buffer object
deletion logic will take care of NULL check and delete accordingly. This
is useful for cases where indirect sram flag could be manipulated later
after buffer allocation.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 57dabfe1a1bee..06ec2dc558571 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -235,11 +235,11 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev)
 		if (adev->vcn.harvest_config & (1 << j))
 			continue;
 
-		if (adev->vcn.indirect_sram) {
-			amdgpu_bo_free_kernel(&adev->vcn.inst[j].dpg_sram_bo,
-						  &adev->vcn.inst[j].dpg_sram_gpu_addr,
-						  (void **)&adev->vcn.inst[j].dpg_sram_cpu_addr);
-		}
+		amdgpu_bo_free_kernel(
+			&adev->vcn.inst[j].dpg_sram_bo,
+			&adev->vcn.inst[j].dpg_sram_gpu_addr,
+			(void **)&adev->vcn.inst[j].dpg_sram_cpu_addr);
+
 		kvfree(adev->vcn.inst[j].saved_bo);
 
 		amdgpu_bo_free_kernel(&adev->vcn.inst[j].vcpu_bo,
-- 
GitLab


From 672c883c26c68fe49b161d7ceab94bdc69e57b0e Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 20 Feb 2023 12:04:30 +0530
Subject: [PATCH 1251/1791] drm/amdgpu: Simplify aquavanjram instance mapping

Simplify so as to use the same sequence to assign logical to physical
ids for all IPs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 33 +++++--------------
 1 file changed, 8 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 51d3cb81e37a7..68d732dd9ecbf 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -101,32 +101,15 @@ static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev,
 
 void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 {
-	int xcc_mask, sdma_mask;
-	int l, i;
-
-	/* Map GC instances */
-	l = 0;
-	xcc_mask = adev->gfx.xcc_mask;
-	while (xcc_mask) {
-		i = ffs(xcc_mask) - 1;
-		adev->ip_map.dev_inst[GC_HWIP][l++] = i;
-		xcc_mask &= ~(1 << i);
-	}
-	for (; l < HWIP_MAX_INSTANCE; l++)
-		adev->ip_map.dev_inst[GC_HWIP][l] = -1;
-
-	l = 0;
-	sdma_mask = adev->sdma.sdma_mask;
-	while (sdma_mask) {
-		i = ffs(sdma_mask) - 1;
-		adev->ip_map.dev_inst[SDMA0_HWIP][l++] = i;
-		sdma_mask &= ~(1 << i);
-	}
-	for (; l < HWIP_MAX_INSTANCE; l++)
-		adev->ip_map.dev_inst[SDMA0_HWIP][l] = -1;
+	u32 ip_map[][2] = {
+		{ GC_HWIP, adev->gfx.xcc_mask },
+		{ SDMA0_HWIP, adev->sdma.sdma_mask },
+		{ VCN_HWIP, adev->vcn.inst_mask },
+	};
+	int i;
 
-	/* This covers both VCN and JPEG, JPEG is only alias of VCN */
-	aqua_vanjaram_populate_ip_map(adev, VCN_HWIP, adev->vcn.inst_mask);
+	for (i = 0; i < ARRAY_SIZE(ip_map); ++i)
+		aqua_vanjaram_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]);
 
 	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
 }
-- 
GitLab


From bc71daff4f147377674e14589fb651bb36f44d4b Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 23 Feb 2023 20:13:56 +0530
Subject: [PATCH 1252/1791] drm/amdgpu: Use GPU VA space for IH v4.4.2 in APU

For IH ring buffer and read/write pointers, use GPU VA space rather than
Guest PA on APU configs. Access through Guest PA doesn't work when IOMMU
is enabled. It is also beneficial in NUMA configs as it allocates from
the closest numa pool in a numa enabled system.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vega20_ih.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
index 536128447b71d..17ccf02462abe 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
@@ -526,6 +526,7 @@ static int vega20_ih_early_init(void *handle)
 static int vega20_ih_sw_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	bool use_bus_addr = true;
 	int r;
 
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0,
@@ -533,14 +534,18 @@ static int vega20_ih_sw_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, true);
+	if ((adev->flags & AMD_IS_APU) &&
+	    (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2)))
+		use_bus_addr = false;
+
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
 	if (r)
 		return r;
 
 	adev->irq.ih.use_doorbell = true;
 	adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
@@ -559,7 +564,7 @@ static int vega20_ih_sw_init(void *handle)
 	/* initialize ih control registers offset */
 	vega20_ih_init_register_offset(adev);
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
-- 
GitLab


From db3b5cb64a9ca301d14ed027e470834316720e42 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 24 Feb 2023 18:01:38 +0530
Subject: [PATCH 1253/1791] drm/amdgpu: Use apt name for FW reserved region

Use the generic term fw_reserved_memory for FW reserve region. This
region may also hold discovery TMR in addition to other reserve
regions. This region size could be larger than discovery tmr size, hence
don't change the discovery tmr size based on this.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 36 +++++++++++++------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h |  3 ++-
 2 files changed, 21 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index bc11ae56bba5b..c8a2d030f226c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1644,14 +1644,15 @@ static int amdgpu_ttm_training_reserve_vram_fini(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev)
+static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev,
+						uint32_t reserve_size)
 {
 	struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx;
 
 	memset(ctx, 0, sizeof(*ctx));
 
 	ctx->c2p_train_data_offset =
-		ALIGN((adev->gmc.mc_vram_size - adev->mman.discovery_tmr_size - SZ_1M), SZ_1M);
+		ALIGN((adev->gmc.mc_vram_size - reserve_size - SZ_1M), SZ_1M);
 	ctx->p2c_train_data_offset =
 		(adev->gmc.mc_vram_size - GDDR6_MEM_TRAINING_OFFSET);
 	ctx->train_data_size =
@@ -1669,9 +1670,10 @@ static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev)
  */
 static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 {
-	int ret;
 	struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx;
 	bool mem_train_support = false;
+	uint32_t reserve_size = 0;
+	int ret;
 
 	if (!amdgpu_sriov_vf(adev)) {
 		if (amdgpu_atomfirmware_mem_training_supported(adev))
@@ -1687,14 +1689,15 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	 * Otherwise, fallback to legacy approach to check and reserve tmr block for ip
 	 * discovery data and G6 memory training data respectively
 	 */
-	adev->mman.discovery_tmr_size =
-		amdgpu_atomfirmware_get_fw_reserved_fb_size(adev);
-	if (!adev->mman.discovery_tmr_size)
-		adev->mman.discovery_tmr_size = DISCOVERY_TMR_OFFSET;
+	if (adev->bios)
+		reserve_size =
+			amdgpu_atomfirmware_get_fw_reserved_fb_size(adev);
+	if (!reserve_size)
+		reserve_size = DISCOVERY_TMR_OFFSET;
 
 	if (mem_train_support) {
 		/* reserve vram for mem train according to TMR location */
-		amdgpu_ttm_training_data_block_init(adev);
+		amdgpu_ttm_training_data_block_init(adev, reserve_size);
 		ret = amdgpu_bo_create_kernel_at(adev,
 						 ctx->c2p_train_data_offset,
 						 ctx->train_data_size,
@@ -1709,15 +1712,13 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	}
 
 	if (!adev->gmc.is_app_apu) {
-		ret = amdgpu_bo_create_kernel_at(adev,
-						 adev->gmc.real_vram_size -
-						 adev->mman.discovery_tmr_size,
-						 adev->mman.discovery_tmr_size,
-						 &adev->mman.discovery_memory,
-						 NULL);
+		ret = amdgpu_bo_create_kernel_at(
+			adev, adev->gmc.real_vram_size - reserve_size,
+			reserve_size, &adev->mman.fw_reserved_memory, NULL);
 		if (ret) {
 			DRM_ERROR("alloc tmr failed(%d)!\n", ret);
-			amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
+			amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory,
+					      NULL, NULL);
 			return ret;
 		}
 	} else {
@@ -1905,8 +1906,9 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	if (!adev->gmc.is_app_apu) {
 		amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
 		amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
-		/* return the IP Discovery TMR memory back to VRAM */
-		amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL);
+		/* return the FW reserved memory back to VRAM */
+		amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL,
+				      NULL);
 		if (adev->mman.stolen_reserved_size)
 			amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory,
 					      NULL, NULL);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index e2cd5894afc9d..da6544fdc8ddd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -78,7 +78,8 @@ struct amdgpu_mman {
 	/* discovery */
 	uint8_t				*discovery_bin;
 	uint32_t			discovery_tmr_size;
-	struct amdgpu_bo		*discovery_memory;
+	/* fw reserved memory */
+	struct amdgpu_bo		*fw_reserved_memory;
 
 	/* firmware VRAM reservation */
 	u64		fw_vram_usage_start_offset;
-- 
GitLab


From 4bdca2057933ef08a2ca7f44e30a8894ff78c472 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 25 Jan 2023 20:04:52 +0530
Subject: [PATCH 1254/1791] drm/amdgpu: Add utility functions for xcp

Add utility functions to get details of xcp and iterate through
available xcps.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 12 ++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h | 31 +++++++++++++++++++++++++
 2 files changed, 43 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index e8aa4d6c6b621..337d558a3145a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -256,3 +256,15 @@ int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
 
 	return id_mask;
 }
+
+int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp,
+				enum AMDGPU_XCP_IP_BLOCK ip,
+				uint32_t *inst_mask)
+{
+	if (!xcp->valid || !inst_mask || !(xcp->ip[ip].valid))
+		return -EINVAL;
+
+	*inst_mask = xcp->ip[ip].inst_mask;
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 1d3dc7d68f543..45d590d7fd95d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -108,4 +108,35 @@ int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode);
 int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
 			     enum AMDGPU_XCP_IP_BLOCK ip, int instance);
 
+int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp,
+				enum AMDGPU_XCP_IP_BLOCK ip,
+				uint32_t *inst_mask);
+
+static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr)
+{
+	if (!xcp_mgr)
+		return 1;
+	else
+		return xcp_mgr->num_xcps;
+}
+
+static inline struct amdgpu_xcp *
+amdgpu_get_next_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int *from)
+{
+	if (!xcp_mgr)
+		return NULL;
+
+	while (*from < MAX_XCP) {
+		if (xcp_mgr->xcp[*from].valid)
+			return &xcp_mgr->xcp[*from];
+		++(*from);
+	}
+
+	return NULL;
+}
+
+#define for_each_xcp(xcp_mgr, xcp, i)                            \
+	for (i = 0, xcp = amdgpu_get_next_xcp(xcp_mgr, &i); xcp; \
+	     xcp = amdgpu_get_next_xcp(xcp_mgr, &i))
+
 #endif
-- 
GitLab


From b6f90baafe267a0705c5d9b1429c875d3c39fbc7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 31 Jan 2023 12:39:49 +0530
Subject: [PATCH 1255/1791] drm/amdgpu: Move memory partition query to gmc

GMC block handles memory related information, it makes more sense to
keep memory partition functions in gmc block.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 30 +----------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 11 -------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 44 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h | 16 +++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 10 ------
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 18 ++++++++++
 6 files changed, 79 insertions(+), 50 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 1487ecac27054..2f7a101593e78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1204,24 +1204,6 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 	return sysfs_emit(buf, "%s\n", partition_mode);
 }
 
-static ssize_t amdgpu_gfx_get_current_memory_partition(struct device *dev,
-						struct device_attribute *addr,
-						char *buf)
-{
-	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct amdgpu_device *adev = drm_to_adev(ddev);
-	enum amdgpu_memory_partition mode;
-	static const char *partition_modes[] = {
-		"UNKNOWN", "NPS1", "NPS2", "NPS4", "NPS8"
-	};
-	BUILD_BUG_ON(ARRAY_SIZE(partition_modes) <= AMDGPU_NPS8_PARTITION_MODE);
-
-	mode = min((int)adev->gfx.funcs->query_mem_partition_mode(adev),
-		AMDGPU_NPS8_PARTITION_MODE);
-
-	return sysfs_emit(buf, "%s\n", partition_modes[mode]);
-}
-
 static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
 						struct device_attribute *addr,
 						const char *buf, size_t count)
@@ -1305,9 +1287,6 @@ static DEVICE_ATTR(current_compute_partition, S_IRUGO | S_IWUSR,
 static DEVICE_ATTR(available_compute_partition, S_IRUGO,
 		   amdgpu_gfx_get_available_compute_partition, NULL);
 
-static DEVICE_ATTR(current_memory_partition, S_IRUGO,
-		   amdgpu_gfx_get_current_memory_partition, NULL);
-
 int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
 {
 	int r;
@@ -1317,19 +1296,12 @@ int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev)
 		return r;
 
 	r = device_create_file(adev->dev, &dev_attr_available_compute_partition);
-	if (r)
-		return r;
 
-	r = device_create_file(adev->dev, &dev_attr_current_memory_partition);
-	if (r)
-		return r;
-
-	return 0;
+	return r;
 }
 
 void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev)
 {
 	device_remove_file(adev->dev, &dev_attr_current_compute_partition);
 	device_remove_file(adev->dev, &dev_attr_available_compute_partition);
-	device_remove_file(adev->dev, &dev_attr_current_memory_partition);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 81b4c7e684af2..728977f8afe7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -71,14 +71,6 @@ enum amdgpu_pkg_type {
 	AMDGPU_PKG_TYPE_UNKNOWN,
 };
 
-enum amdgpu_memory_partition {
-	UNKNOWN_MEMORY_PARTITION_MODE = 0,
-	AMDGPU_NPS1_PARTITION_MODE = 1,
-	AMDGPU_NPS2_PARTITION_MODE = 2,
-	AMDGPU_NPS4_PARTITION_MODE = 3,
-	AMDGPU_NPS8_PARTITION_MODE = 4,
-};
-
 struct amdgpu_mec {
 	struct amdgpu_bo	*hpd_eop_obj;
 	u64			hpd_eop_gpu_addr;
@@ -276,8 +268,6 @@ struct amdgpu_gfx_funcs {
 				   struct amdgpu_gfx_shadow_info *shadow_info);
 	enum amdgpu_gfx_partition
 			(*query_partition_mode)(struct amdgpu_device *adev);
-	enum amdgpu_memory_partition
-			(*query_mem_partition_mode)(struct amdgpu_device *adev);
 	int (*switch_partition_mode)(struct amdgpu_device *adev,
 				     int num_xccs_per_xcp);
 	int (*ih_node_to_logical_xcc)(struct amdgpu_device *adev, int ih_node);
@@ -414,7 +404,6 @@ struct amdgpu_gfx {
 	bool				cp_gfx_shadow; /* for gfx11 */
 
 	uint16_t 			xcc_mask;
-	enum amdgpu_memory_partition	mem_partition_mode;
 	uint32_t			num_xcc_per_xcp;
 	struct mutex			partition_mutex;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index b8825a0670a41..d12625f1de5ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -891,3 +891,47 @@ int amdgpu_gmc_vram_checking(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+static ssize_t current_memory_partition_show(
+	struct device *dev, struct device_attribute *addr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	enum amdgpu_memory_partition mode;
+
+	mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+	switch (mode) {
+	case AMDGPU_NPS1_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS1\n");
+	case AMDGPU_NPS2_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS2\n");
+	case AMDGPU_NPS3_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS3\n");
+	case AMDGPU_NPS4_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS4\n");
+	case AMDGPU_NPS6_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS6\n");
+	case AMDGPU_NPS8_PARTITION_MODE:
+		return sysfs_emit(buf, "NPS8\n");
+	default:
+		return sysfs_emit(buf, "UNKNOWN\n");
+	}
+
+	return sysfs_emit(buf, "UNKNOWN\n");
+}
+
+static DEVICE_ATTR_RO(current_memory_partition);
+
+int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev)
+{
+	if (!adev->gmc.gmc_funcs->query_mem_partition_mode)
+		return 0;
+
+	return device_create_file(adev->dev,
+				  &dev_attr_current_memory_partition);
+}
+
+void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev)
+{
+	device_remove_file(adev->dev, &dev_attr_current_memory_partition);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index e408abfc2daf3..2bd3b9665ebfd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -63,6 +63,16 @@
 
 struct firmware;
 
+enum amdgpu_memory_partition {
+	UNKNOWN_MEMORY_PARTITION_MODE = 0,
+	AMDGPU_NPS1_PARTITION_MODE = 1,
+	AMDGPU_NPS2_PARTITION_MODE = 2,
+	AMDGPU_NPS3_PARTITION_MODE = 3,
+	AMDGPU_NPS4_PARTITION_MODE = 4,
+	AMDGPU_NPS6_PARTITION_MODE = 6,
+	AMDGPU_NPS8_PARTITION_MODE = 8,
+};
+
 /*
  * GMC page fault information
  */
@@ -140,6 +150,9 @@ struct amdgpu_gmc_funcs {
 			   uint64_t *flags);
 	/* get the amount of memory used by the vbios for pre-OS console */
 	unsigned int (*get_vbios_fb_size)(struct amdgpu_device *adev);
+
+	enum amdgpu_memory_partition (*query_mem_partition_mode)(
+		struct amdgpu_device *adev);
 };
 
 struct amdgpu_xgmi_ras {
@@ -375,4 +388,7 @@ uint64_t amdgpu_gmc_vram_mc2pa(struct amdgpu_device *adev, uint64_t mc_addr);
 uint64_t amdgpu_gmc_vram_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo);
 uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo);
 int amdgpu_gmc_vram_checking(struct amdgpu_device *adev);
+int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev);
+void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 69867294117ea..81ab3cd2f2296 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -607,16 +607,7 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 {
 	soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, xcc_id));
 }
-static enum amdgpu_memory_partition
-gfx_v9_4_3_query_memory_partition(struct amdgpu_device *adev)
-{
-	enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE;
-
-	if (adev->nbio.funcs->get_memory_partition_mode)
-		mode = adev->nbio.funcs->get_memory_partition_mode(adev);
 
-	return mode;
-}
 
 static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 						int num_xccs_per_xcp)
@@ -660,7 +651,6 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = {
 	.read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q,
 	.switch_partition_mode = &gfx_v9_4_3_switch_compute_partition,
-	.query_mem_partition_mode = &gfx_v9_4_3_query_memory_partition,
 	.ih_node_to_logical_xcc = &gfx_v9_4_3_ih_to_xcc_inst,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 245de27c75402..db157a31a780e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1330,6 +1330,17 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
 	return size;
 }
 
+static enum amdgpu_memory_partition
+gmc_v9_0_query_memory_partition(struct amdgpu_device *adev)
+{
+	enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE;
+
+	if (adev->nbio.funcs->get_memory_partition_mode)
+		mode = adev->nbio.funcs->get_memory_partition_mode(adev);
+
+	return mode;
+}
+
 static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
 	.flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid,
@@ -1339,6 +1350,7 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.get_vm_pde = gmc_v9_0_get_vm_pde,
 	.get_vm_pte = gmc_v9_0_get_vm_pte,
 	.get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size,
+	.query_mem_partition_mode = &gmc_v9_0_query_memory_partition,
 };
 
 static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev)
@@ -1901,6 +1913,9 @@ static int gmc_v9_0_sw_init(void *handle)
 	if (r)
 		return r;
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
+		amdgpu_gmc_sysfs_init(adev);
+
 	return 0;
 }
 
@@ -1908,6 +1923,9 @@ static int gmc_v9_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
+		amdgpu_gmc_sysfs_fini(adev);
+
 	amdgpu_gmc_ras_fini(adev);
 	amdgpu_gem_force_release(adev);
 	amdgpu_vm_manager_fini(adev);
-- 
GitLab


From 0f2e1d620eca56c4ceebc041aabb1eda26b2cfd0 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 17 Feb 2023 09:32:44 +0530
Subject: [PATCH 1256/1791] drm/amdgpu: Get supported memory partition modes

Expand the interface to get supported memory partition modes also along
with the current memory partition mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h |  3 ++-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c    | 11 +++++++++--
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c   |  9 ++++++++-
 3 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index eb25ac98903f1..095aecfb201ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -97,7 +97,8 @@ struct amdgpu_nbio_funcs {
 	void (*clear_doorbell_interrupt)(struct amdgpu_device *adev);
 	u32 (*get_rom_offset)(struct amdgpu_device *adev);
 	int (*get_compute_partition_mode)(struct amdgpu_device *adev);
-	u32 (*get_memory_partition_mode)(struct amdgpu_device *adev);
+	u32 (*get_memory_partition_mode)(struct amdgpu_device *adev,
+					 u32 *supp_modes);
 	void (*set_compute_partition_mode)(struct amdgpu_device *adev,
 					   enum amdgpu_gfx_partition mode);
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index db157a31a780e..d6a1dac01952e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1331,16 +1331,23 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
 }
 
 static enum amdgpu_memory_partition
-gmc_v9_0_query_memory_partition(struct amdgpu_device *adev)
+gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes)
 {
 	enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE;
 
 	if (adev->nbio.funcs->get_memory_partition_mode)
-		mode = adev->nbio.funcs->get_memory_partition_mode(adev);
+		mode = adev->nbio.funcs->get_memory_partition_mode(adev,
+								   supp_modes);
 
 	return mode;
 }
 
+static enum amdgpu_memory_partition
+gmc_v9_0_query_memory_partition(struct amdgpu_device *adev)
+{
+	return gmc_v9_0_get_memory_partition(adev, NULL);
+}
+
 static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
 	.flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid,
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index fa4b423c399b2..e1552d6453083 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -408,12 +408,19 @@ static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
 	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
 }
 
-static enum amdgpu_memory_partition nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev)
+static enum amdgpu_memory_partition
+nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev, u32 *supp_modes)
 {
 	u32 tmp;
+
 	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_STATUS);
 	tmp = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_MEM_STATUS, NPS_MODE);
 
+	if (supp_modes) {
+		*supp_modes =
+			RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_CAP);
+	}
+
 	return ffs(tmp);
 }
 
-- 
GitLab


From 1cc823011a23fa0e3497e9f6655172b2507ce2cd Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 14 Feb 2023 18:33:51 +0530
Subject: [PATCH 1257/1791] drm/amdgpu: Store additional numa node information

Use a struct to store additional numa node information including size
and base address. Add numa_info pointer to xcc object to point to the
relevant structure based on its proximity domain.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 79 ++++++++++++++++++++++--
 1 file changed, 75 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 9dbdd699dcea6..6a13e9c275507 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -26,6 +26,7 @@
 #include <linux/acpi.h>
 #include <linux/backlight.h>
 #include <linux/slab.h>
+#include <linux/xarray.h>
 #include <linux/power_supply.h>
 #include <linux/pm_runtime.h>
 #include <linux/suspend.h>
@@ -53,10 +54,18 @@ static const guid_t amd_xcc_dsm_guid = GUID_INIT(0x8267f5d5, 0xa556, 0x44f2,
 
 #define AMD_XCC_MAX_HID 24
 
+struct amdgpu_numa_info {
+	uint64_t size;
+	int pxm;
+	int nid;
+};
+
+struct xarray numa_info_xa;
+
 /* Encapsulates the XCD acpi object information */
 struct amdgpu_acpi_xcc_info {
 	struct list_head list;
-	int mem_node;
+	struct amdgpu_numa_info *numa_info;
 	uint8_t xcp_node;
 	uint8_t phy_id;
 	acpi_handle handle;
@@ -838,6 +847,52 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta
 	return r;
 }
 
+static inline uint64_t amdgpu_acpi_get_numa_size(int nid)
+{
+	/* This is directly using si_meminfo_node implementation as the
+	 * function is not exported.
+	 */
+	int zone_type;
+	uint64_t managed_pages = 0;
+
+	pg_data_t *pgdat = NODE_DATA(nid);
+
+	for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++)
+		managed_pages +=
+			zone_managed_pages(&pgdat->node_zones[zone_type]);
+	return managed_pages * PAGE_SIZE;
+}
+
+static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm)
+{
+	struct amdgpu_numa_info *numa_info;
+	int nid;
+
+	numa_info = xa_load(&numa_info_xa, pxm);
+
+	if (!numa_info) {
+		struct sysinfo info;
+
+		numa_info = kzalloc(sizeof *numa_info, GFP_KERNEL);
+		if (!numa_info)
+			return NULL;
+
+		nid = pxm_to_node(pxm);
+		numa_info->pxm = pxm;
+		numa_info->nid = nid;
+
+		if (numa_info->nid == NUMA_NO_NODE) {
+			si_meminfo(&info);
+			numa_info->size = info.totalram * info.mem_unit;
+		} else {
+			numa_info->size = amdgpu_acpi_get_numa_size(nid);
+		}
+		xa_store(&numa_info_xa, numa_info->pxm, numa_info, GFP_KERNEL);
+	}
+
+	return numa_info;
+}
+
 /**
  * amdgpu_acpi_get_node_id - obtain the NUMA node id for corresponding amdgpu
  * acpi device handle
@@ -850,18 +905,25 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta
  *
  * Returns ACPI STATUS OK with Node ID on success or the corresponding failure reason
  */
-acpi_status amdgpu_acpi_get_node_id(acpi_handle handle, int *nid)
+acpi_status amdgpu_acpi_get_node_id(acpi_handle handle,
+				    struct amdgpu_numa_info **numa_info)
 {
 #ifdef CONFIG_ACPI_NUMA
 	u64 pxm;
 	acpi_status status;
 
+	if (!numa_info)
+		return_ACPI_STATUS(AE_ERROR);
+
 	status = acpi_evaluate_integer(handle, "_PXM", NULL, &pxm);
 
 	if (ACPI_FAILURE(status))
 		return status;
 
-	*nid = pxm_to_node(pxm);
+	*numa_info = amdgpu_acpi_get_numa_info(pxm);
+
+	if (!*numa_info)
+		return_ACPI_STATUS(AE_ERROR);
 
 	return_ACPI_STATUS(AE_OK);
 #else
@@ -1001,7 +1063,8 @@ static int amdgpu_acpi_get_xcc_info(struct amdgpu_acpi_xcc_info *xcc_info,
 	ACPI_FREE(obj);
 	obj = NULL;
 
-	status = amdgpu_acpi_get_node_id(xcc_info->handle, &xcc_info->mem_node);
+	status =
+		amdgpu_acpi_get_node_id(xcc_info->handle, &xcc_info->numa_info);
 
 	/* TODO: check if this check is required */
 	if (ACPI_SUCCESS(status))
@@ -1023,6 +1086,7 @@ static int amdgpu_acpi_enumerate_xcc(void)
 	u16 bdf;
 
 	INIT_LIST_HEAD(&amdgpu_acpi_dev_list);
+	xa_init(&numa_info_xa);
 
 	for (id = 0; id < AMD_XCC_MAX_HID; id++) {
 		sprintf(hid, "%s%d", "AMD", AMD_XCC_HID_START + id);
@@ -1353,6 +1417,13 @@ void amdgpu_acpi_release(void)
 {
 	struct amdgpu_acpi_dev_info *dev_info, *dev_tmp;
 	struct amdgpu_acpi_xcc_info *xcc_info, *xcc_tmp;
+	struct amdgpu_numa_info *numa_info;
+	unsigned long index;
+
+	xa_for_each(&numa_info_xa, index, numa_info) {
+		kfree(numa_info);
+		xa_erase(&numa_info_xa, index);
+	}
 
 	if (list_empty(&amdgpu_acpi_dev_list))
 		return;
-- 
GitLab


From fa0497c34eb7dd9db9a09963917382e924c3fbc5 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 14 Feb 2023 18:59:40 +0530
Subject: [PATCH 1258/1791] drm/amdgpu: Add API to get numa information of XCC

Add interface to get numa information of ACPI XCC object. The interface
uses logical id to identify an XCC.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      | 14 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 32 +++++++++++++++++++-----
 2 files changed, 40 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index f4461bc8b1fdd..f2bafab15cebd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1383,6 +1383,12 @@ struct amdgpu_afmt_acr amdgpu_afmt_acr(uint32_t clock);
 
 /* amdgpu_acpi.c */
 
+struct amdgpu_numa_info {
+	uint64_t size;
+	int pxm;
+	int nid;
+};
+
 /* ATCS Device/Driver State */
 #define AMDGPU_ATCS_PSC_DEV_STATE_D0		0
 #define AMDGPU_ATCS_PSC_DEV_STATE_D3_HOT	3
@@ -1402,6 +1408,8 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta
 int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev);
 int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset,
 			     u64 *tmr_size);
+int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id,
+			     struct amdgpu_numa_info *numa_info);
 
 void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps);
 bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev);
@@ -1414,6 +1422,12 @@ static inline int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev,
 {
 	return -EINVAL;
 }
+static inline int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev,
+					   int xcc_id,
+					   struct amdgpu_numa_info *numa_info)
+{
+	return -EINVAL;
+}
 static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { }
 static inline bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev) { return false; }
 static inline void amdgpu_acpi_detect(void) { }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 6a13e9c275507..873532c4adbe3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -54,12 +54,6 @@ static const guid_t amd_xcc_dsm_guid = GUID_INIT(0x8267f5d5, 0xa556, 0x44f2,
 
 #define AMD_XCC_MAX_HID 24
 
-struct amdgpu_numa_info {
-	uint64_t size;
-	int pxm;
-	int nid;
-};
-
 struct xarray numa_info_xa;
 
 /* Encapsulates the XCD acpi object information */
@@ -1156,6 +1150,32 @@ int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset,
 	return 0;
 }
 
+int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id,
+			     struct amdgpu_numa_info *numa_info)
+{
+	struct amdgpu_acpi_dev_info *dev_info;
+	struct amdgpu_acpi_xcc_info *xcc_info;
+	u16 bdf;
+
+	if (!numa_info)
+		return -EINVAL;
+
+	bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn;
+	dev_info = amdgpu_acpi_get_dev(bdf);
+	if (!dev_info)
+		return -ENOENT;
+
+	list_for_each_entry(xcc_info, &dev_info->xcc_list, list) {
+		if (xcc_info->phy_id == xcc_id) {
+			memcpy(numa_info, xcc_info->numa_info,
+			       sizeof(*numa_info));
+			return 0;
+		}
+	}
+
+	return -ENOENT;
+}
+
 /**
  * amdgpu_acpi_event - handle notify events
  *
-- 
GitLab


From 14493cb99b71dbaff58dc0dc0b2cc0a56a88ef05 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 14 Feb 2023 14:37:53 +0530
Subject: [PATCH 1259/1791] drm/amdgpu: Add memory partitions to gmc

Some ASICs have the device memory divided into multiple partitions. The
parititions could be denoted by a numa node or by a range of pages.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 2bd3b9665ebfd..43357d699e6ee 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -178,6 +178,21 @@ struct amdgpu_xgmi {
 	struct amdgpu_xgmi_ras *ras;
 };
 
+struct amdgpu_mem_partition_info {
+	union {
+		struct {
+			uint32_t fpfn;
+			uint32_t lpfn;
+		} range;
+		struct {
+			int node;
+		} numa;
+	};
+	uint64_t size;
+};
+
+#define INVALID_PFN    -1
+
 struct amdgpu_gmc {
 	/* FB's physical address in MMIO space (for CPU to
 	 * map FB). This is different compared to the agp/
@@ -266,6 +281,8 @@ struct amdgpu_gmc {
 	bool tmz_enabled;
 	bool is_app_apu;
 
+	struct amdgpu_mem_partition_info *mem_partitions;
+	uint8_t num_mem_partitions;
 	const struct amdgpu_gmc_funcs	*gmc_funcs;
 
 	struct amdgpu_xgmi xgmi;
-- 
GitLab


From a433f1f59484fba7a7743a3c5a5f320d9e828b3a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 14 Feb 2023 14:45:45 +0530
Subject: [PATCH 1260/1791] drm/amdgpu: Initialize memory ranges for GC 9.4.3

GC 9.4.3 ASICS may have memory split into multiple partitions.Initialize
the memory partition information for each range. The information may be
in the form of a numa node id or a range of pages.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 172 ++++++++++++++++++++++++++
 1 file changed, 172 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index d6a1dac01952e..1653d77df3ba3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -79,6 +79,7 @@
 #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2                                                          0x05ea
 #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2_BASE_IDX                                                 2
 
+#define MAX_MEM_RANGES 8
 
 static const char *gfxhub_client_ids[] = {
 	"CB",
@@ -1742,6 +1743,169 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
 		adev->gmc.sdpif_register = RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0);
 }
 
+static bool gmc_v9_0_validate_partition_info(struct amdgpu_device *adev)
+{
+	enum amdgpu_memory_partition mode;
+	u32 supp_modes;
+	bool valid;
+
+	mode = gmc_v9_0_get_memory_partition(adev, &supp_modes);
+
+	/* Mode detected by hardware not present in supported modes */
+	if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) &&
+	    !(BIT(mode - 1) & supp_modes))
+		return false;
+
+	switch (mode) {
+	case UNKNOWN_MEMORY_PARTITION_MODE:
+	case AMDGPU_NPS1_PARTITION_MODE:
+		valid = (adev->gmc.num_mem_partitions == 1);
+		break;
+	case AMDGPU_NPS2_PARTITION_MODE:
+		valid = (adev->gmc.num_mem_partitions == 2);
+		break;
+	case AMDGPU_NPS4_PARTITION_MODE:
+		valid = (adev->gmc.num_mem_partitions == 3 ||
+			 adev->gmc.num_mem_partitions == 4);
+		break;
+	default:
+		valid = false;
+	}
+
+	return valid;
+}
+
+static bool gmc_v9_0_is_node_present(int *node_ids, int num_ids, int nid)
+{
+	int i;
+
+	/* Check if node with id 'nid' is present in 'node_ids' array */
+	for (i = 0; i < num_ids; ++i)
+		if (node_ids[i] == nid)
+			return true;
+
+	return false;
+}
+
+static void
+gmc_v9_0_init_acpi_mem_ranges(struct amdgpu_device *adev,
+			      struct amdgpu_mem_partition_info *mem_ranges)
+{
+	int num_ranges = 0, ret, mem_groups;
+	struct amdgpu_numa_info numa_info;
+	int node_ids[MAX_MEM_RANGES];
+	int num_xcc, xcc_id;
+	uint32_t xcc_mask;
+
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	xcc_mask = (1U << num_xcc) - 1;
+	mem_groups = hweight32(adev->aid_mask);
+
+	for_each_inst(xcc_id, xcc_mask)	{
+		ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info);
+		if (ret)
+			continue;
+
+		if (numa_info.nid == NUMA_NO_NODE) {
+			mem_ranges[0].size = numa_info.size;
+			mem_ranges[0].numa.node = numa_info.nid;
+			num_ranges = 1;
+			break;
+		}
+
+		if (gmc_v9_0_is_node_present(node_ids, num_ranges,
+					     numa_info.nid))
+			continue;
+
+		node_ids[num_ranges] = numa_info.nid;
+		mem_ranges[num_ranges].numa.node = numa_info.nid;
+		mem_ranges[num_ranges].size = numa_info.size;
+		++num_ranges;
+	}
+
+	adev->gmc.num_mem_partitions = num_ranges;
+
+	/* If there is only partition, don't use entire size */
+	if (adev->gmc.num_mem_partitions == 1)
+		mem_ranges[0].size =
+			(mem_ranges[0].size * (mem_groups - 1) / mem_groups);
+}
+
+static void
+gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev,
+			    struct amdgpu_mem_partition_info *mem_ranges)
+{
+	enum amdgpu_memory_partition mode;
+	u32 start_addr = 0, size;
+	int i;
+
+	mode = gmc_v9_0_query_memory_partition(adev);
+
+	switch (mode) {
+	case UNKNOWN_MEMORY_PARTITION_MODE:
+	case AMDGPU_NPS1_PARTITION_MODE:
+		adev->gmc.num_mem_partitions = 1;
+		break;
+	case AMDGPU_NPS2_PARTITION_MODE:
+		adev->gmc.num_mem_partitions = 2;
+		break;
+	case AMDGPU_NPS4_PARTITION_MODE:
+		if (adev->flags & AMD_IS_APU)
+			adev->gmc.num_mem_partitions = 3;
+		else
+			adev->gmc.num_mem_partitions = 4;
+		break;
+	default:
+		adev->gmc.num_mem_partitions = 1;
+		break;
+	}
+
+	size = (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) /
+	       adev->gmc.num_mem_partitions;
+
+	for (i = 0; i < adev->gmc.num_mem_partitions; ++i) {
+		mem_ranges[i].range.fpfn = start_addr;
+		mem_ranges[i].size = ((u64)size << AMDGPU_GPU_PAGE_SHIFT);
+		mem_ranges[i].range.lpfn = start_addr + size - 1;
+		start_addr += size;
+	}
+
+	/* Adjust the last one */
+	mem_ranges[adev->gmc.num_mem_partitions - 1].range.lpfn =
+		(adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1;
+	mem_ranges[adev->gmc.num_mem_partitions - 1].size =
+		adev->gmc.real_vram_size -
+		((u64)mem_ranges[adev->gmc.num_mem_partitions - 1].range.fpfn
+		 << AMDGPU_GPU_PAGE_SHIFT);
+}
+
+static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev)
+{
+	bool valid;
+
+	adev->gmc.mem_partitions = kzalloc(
+		MAX_MEM_RANGES * sizeof(struct amdgpu_mem_partition_info),
+		GFP_KERNEL);
+
+	if (!adev->gmc.mem_partitions)
+		return -ENOMEM;
+
+	/* TODO : Get the range from PSP/Discovery for dGPU */
+	if (adev->gmc.is_app_apu)
+		gmc_v9_0_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions);
+	else
+		gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions);
+
+	valid = gmc_v9_0_validate_partition_info(adev);
+	if (!valid) {
+		/* TODO: handle invalid case */
+		dev_WARN(adev->dev,
+			 "Mem ranges not matching with hardware config");
+	}
+
+	return 0;
+}
+
 static int gmc_v9_0_sw_init(void *handle)
 {
 	int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits;
@@ -1888,6 +2052,12 @@ static int gmc_v9_0_sw_init(void *handle)
 
 	amdgpu_gmc_get_vbios_allocations(adev);
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		r = gmc_v9_0_init_mem_ranges(adev);
+		if (r)
+			return r;
+	}
+
 	/* Memory manager */
 	r = amdgpu_bo_init(adev);
 	if (r)
@@ -1932,6 +2102,8 @@ static int gmc_v9_0_sw_fini(void *handle)
 
 	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
 		amdgpu_gmc_sysfs_fini(adev);
+	adev->gmc.num_mem_partitions = 0;
+	kfree(adev->gmc.mem_partitions);
 
 	amdgpu_gmc_ras_fini(adev);
 	amdgpu_gem_force_release(adev);
-- 
GitLab


From da539b213d7952741499283636f70406383b9570 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 3 Feb 2023 17:12:10 +0530
Subject: [PATCH 1261/1791] drm/amdgpu: Add callback to fill xcp memory id

Add callback in xcp interface to fill xcp memory id information. Memory
id is used to identify the range/partition of an XCP from the available
memory partitions in device. Also, fill the id information.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 12 ++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  4 +++-
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 337d558a3145a..e1d3727036a11 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -116,6 +116,7 @@ static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 static int __amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps)
 {
 	struct amdgpu_xcp_ip ip;
+	uint8_t mem_id;
 	int i, j, ret;
 
 	for (i = 0; i < MAX_XCP; ++i)
@@ -130,6 +131,17 @@ static int __amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps)
 
 			__amdgpu_xcp_add_block(xcp_mgr, i, &ip);
 		}
+
+		xcp_mgr->xcp[i].id = i;
+
+		if (xcp_mgr->funcs->get_xcp_mem_id) {
+			ret = xcp_mgr->funcs->get_xcp_mem_id(
+				xcp_mgr, &xcp_mgr->xcp[i], &mem_id);
+			if (ret)
+				continue;
+			else
+				xcp_mgr->xcp[i].mem_id = mem_id;
+		}
 	}
 
 	xcp_mgr->num_xcps = num_xcps;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 45d590d7fd95d..7e7e458d307ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -68,7 +68,7 @@ struct amdgpu_xcp {
 	struct amdgpu_xcp_ip ip[AMDGPU_XCP_MAX_BLOCKS];
 
 	uint8_t id;
-	uint8_t mem_node;
+	uint8_t mem_id;
 	bool valid;
 };
 
@@ -89,6 +89,8 @@ struct amdgpu_xcp_mgr_funcs {
 	int (*get_ip_details)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 			      enum AMDGPU_XCP_IP_BLOCK ip_id,
 			      struct amdgpu_xcp_ip *ip);
+	int (*get_xcp_mem_id)(struct amdgpu_xcp_mgr *xcp_mgr,
+			      struct amdgpu_xcp *xcp, uint8_t *mem_id);
 
 	int (*prepare_suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
 	int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
-- 
GitLab


From 15e3eee8d3939d1f28cd314a5db2590ab94109d6 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 3 Feb 2023 17:14:12 +0530
Subject: [PATCH 1262/1791] drm/amdgpu: Fill xcp mem node in aquavanjaram

Implement callbacks to fill memory node information in aquavanjaram.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 61 ++++++++++++++++++-
 1 file changed, 60 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 68d732dd9ecbf..aa1bb78831582 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -331,6 +331,64 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 	return ret;
 }
 
+static int __aqua_vanjaram_get_xcp_mem_id(struct amdgpu_device *adev,
+					  int xcc_id, uint8_t *mem_id)
+{
+	/* TODO: Check if any validation is required based on current
+	 * memory/spatial modes
+	 */
+	*mem_id = xcc_id / adev->gfx.num_xcc_per_xcp;
+
+	return 0;
+}
+
+static int aqua_vanjaram_get_xcp_mem_id(struct amdgpu_xcp_mgr *xcp_mgr,
+					struct amdgpu_xcp *xcp, uint8_t *mem_id)
+{
+	struct amdgpu_numa_info numa_info;
+	struct amdgpu_device *adev;
+	uint32_t xcc_mask;
+	int r, i, xcc_id;
+
+	adev = xcp_mgr->adev;
+	/* TODO: BIOS is not returning the right info now
+	 * Check on this later
+	 */
+	/*
+	if (adev->gmc.gmc_funcs->query_mem_partition_mode)
+		mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+	*/
+	if (adev->gmc.num_mem_partitions == 1) {
+		/* Only one range */
+		*mem_id = 0;
+		return 0;
+	}
+
+	r = amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_GFX, &xcc_mask);
+	if (r || !xcc_mask)
+		return -EINVAL;
+
+	xcc_id = ffs(xcc_mask) - 1;
+	if (!adev->gmc.is_app_apu)
+		return __aqua_vanjaram_get_xcp_mem_id(adev, xcc_id, mem_id);
+
+	r = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info);
+
+	if (r)
+		return r;
+
+	r = -EINVAL;
+	for (i = 0; i < adev->gmc.num_mem_partitions; ++i) {
+		if (adev->gmc.mem_partitions[i].numa.node == numa_info.nid) {
+			*mem_id = i;
+			r = 0;
+			break;
+		}
+	}
+
+	return r;
+}
+
 int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 				     enum AMDGPU_XCP_IP_BLOCK ip_id,
 				     struct amdgpu_xcp_ip *ip)
@@ -344,7 +402,8 @@ int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = {
 	.switch_partition_mode = &aqua_vanjaram_switch_partition_mode,
 	.query_partition_mode = &aqua_vanjaram_query_partition_mode,
-	.get_ip_details = &aqua_vanjaram_get_xcp_ip_details
+	.get_ip_details = &aqua_vanjaram_get_xcp_ip_details,
+	.get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id
 };
 
 static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
-- 
GitLab


From e47947abb9e71176ea2d9c8f55e03134dabd2605 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 3 Feb 2023 18:46:40 +0530
Subject: [PATCH 1263/1791] drm/amdgpu: Move initialization of xcp before kfd

After partition switch, fill all relevant xcp information before kfd
starts initialization.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c          | 16 +++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h          |  1 +
 .../gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c  |  6 ++++--
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index e1d3727036a11..bca226cc4e0b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -113,12 +113,17 @@ static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 	xcp->valid = true;
 }
 
-static int __amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps)
+int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 {
 	struct amdgpu_xcp_ip ip;
 	uint8_t mem_id;
 	int i, j, ret;
 
+	if (!num_xcps || num_xcps > MAX_XCP)
+		return -EINVAL;
+
+	xcp_mgr->mode = mode;
+
 	for (i = 0; i < MAX_XCP; ++i)
 		xcp_mgr->xcp[i].valid = false;
 
@@ -181,13 +186,6 @@ int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 		goto out;
 	}
 
-	if (!num_xcps || num_xcps > MAX_XCP) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	xcp_mgr->mode = mode;
-	__amdgpu_xcp_init(xcp_mgr, num_xcps);
 out:
 	mutex_unlock(&xcp_mgr->xcp_lock);
 
@@ -240,7 +238,7 @@ int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
 	mutex_init(&xcp_mgr->xcp_lock);
 
 	if (init_mode != AMDGPU_XCP_MODE_NONE)
-		__amdgpu_xcp_init(xcp_mgr, init_num_xcps);
+		amdgpu_xcp_init(xcp_mgr, init_num_xcps, init_mode);
 
 	adev->xcp_mgr = xcp_mgr;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 7e7e458d307ec..e1319b887bf3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -105,6 +105,7 @@ int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
 
 int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
 			int init_xcps, struct amdgpu_xcp_mgr_funcs *xcp_funcs);
+int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode);
 int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags);
 int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode);
 int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index aa1bb78831582..004400fb89b01 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -321,9 +321,11 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 	if (adev->nbio.funcs->set_compute_partition_mode)
 		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
 
-	ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags);
-
+	/* Init info about new xcps */
 	*num_xcps = num_xcc / num_xcc_per_xcp;
+	amdgpu_xcp_init(xcp_mgr, *num_xcps, mode);
+
+	ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags);
 unlock:
 	if (flags & AMDGPU_XCP_OPS_KFD)
 		amdgpu_amdkfd_unlock_kfd(adev);
-- 
GitLab


From a75f2271a4936265c8a189ab06f9eb89e343b441 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 9 Feb 2023 14:44:13 +0530
Subject: [PATCH 1264/1791] drm/amdkfd: Add xcp reference to kfd node

Fetch xcp information from xcp_mgr and also add xcc_mask to kfd node.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 19 +++++++++++++------
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h   |  3 +++
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 4293cbf9ceb04..647c3313c27e7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -594,6 +594,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
 	unsigned int max_proc_per_quantum;
 	int num_xcd, partition_mode;
+	int xcp_idx;
 
 	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev,
 			KGD_ENGINE_MEC1);
@@ -603,11 +604,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			KGD_ENGINE_SDMA1);
 	kfd->shared_resources = *gpu_resources;
 
-	num_xcd = NUM_XCC(kfd->adev->gfx.xcc_mask);
-	if (num_xcd == 0 || num_xcd == 1 || kfd->adev->gfx.num_xcc_per_xcp == 0)
-		kfd->num_nodes = 1;
-	else
-		kfd->num_nodes = num_xcd / kfd->adev->gfx.num_xcc_per_xcp;
+	kfd->num_nodes = amdgpu_xcp_get_num_xcp(kfd->adev->xcp_mgr);
+
 	if (kfd->num_nodes == 0) {
 		dev_err(kfd_device,
 			"KFD num nodes cannot be 0, GC inst: %d, num_xcc_in_node: %d\n",
@@ -735,7 +733,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 				kfd->num_nodes);
 
 	/* Allocate the KFD nodes */
-	for (i = 0; i < kfd->num_nodes; i++) {
+	for (i = 0, xcp_idx = 0; i < kfd->num_nodes; i++) {
 		node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL);
 		if (!node)
 			goto node_alloc_error;
@@ -745,6 +743,15 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		node->kfd = kfd;
 		node->kfd2kgd = kfd->kfd2kgd;
 		node->vm_info.vmid_num_kfd = vmid_num_kfd;
+		node->xcp = amdgpu_get_next_xcp(kfd->adev->xcp_mgr, &xcp_idx);
+		/* TODO : Check if error handling is needed */
+		if (node->xcp)
+			amdgpu_xcp_get_inst_details(node->xcp, AMDGPU_XCP_GFX,
+						    &node->xcc_mask);
+		else
+			node->xcc_mask =
+				(1U << NUM_XCC(kfd->adev->gfx.xcc_mask)) - 1;
+
 		node->num_xcc_per_node = max(1U, kfd->adev->gfx.num_xcc_per_xcp);
 		node->start_xcc_id = node->num_xcc_per_node * i;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 6e1c15682c285..559ac5efdc260 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -278,6 +278,9 @@ struct kfd_node {
 	unsigned int start_xcc_id;	/* Starting XCC instance
 					 * number for the node
 					 */
+	uint32_t xcc_mask; /* Instance mask of XCCs present */
+	struct amdgpu_xcp *xcp;
+
 	/* Interrupts */
 	struct kfifo ih_fifo;
 	struct workqueue_struct *ih_wq;
-- 
GitLab


From c4050ff1a43eec08498b1ed876efc6213592dba0 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 9 Feb 2023 16:30:53 +0530
Subject: [PATCH 1265/1791] drm/amdkfd: Use xcc mask for identifying xcc

Instead of start xcc id and number of xcc per node, use the xcc mask
which is the mask of logical ids of xccs belonging to a parition.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |  9 +-
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 86 ++++++++++---------
 drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c  |  2 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   | 71 +++++++--------
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  4 -
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |  8 +-
 .../amd/amdkfd/kfd_process_queue_manager.c    |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c     |  8 +-
 8 files changed, 95 insertions(+), 95 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 647c3313c27e7..b5497d2ee984e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -745,15 +745,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		node->vm_info.vmid_num_kfd = vmid_num_kfd;
 		node->xcp = amdgpu_get_next_xcp(kfd->adev->xcp_mgr, &xcp_idx);
 		/* TODO : Check if error handling is needed */
-		if (node->xcp)
+		if (node->xcp) {
 			amdgpu_xcp_get_inst_details(node->xcp, AMDGPU_XCP_GFX,
 						    &node->xcc_mask);
-		else
+			++xcp_idx;
+		} else {
 			node->xcc_mask =
 				(1U << NUM_XCC(kfd->adev->gfx.xcc_mask)) - 1;
-
-		node->num_xcc_per_node = max(1U, kfd->adev->gfx.num_xcc_per_xcp);
-		node->start_xcc_id = node->num_xcc_per_node * i;
+		}
 
 		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
 		    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 2b5c4b2dd2423..493b4b66f1803 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -136,16 +136,14 @@ static void init_sdma_bitmaps(struct device_queue_manager *dqm)
 void program_sh_mem_settings(struct device_queue_manager *dqm,
 					struct qcm_process_device *qpd)
 {
-	int xcc = 0;
+	uint32_t xcc_mask = dqm->dev->xcc_mask;
+	int xcc_id;
 
-	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+	for_each_inst(xcc_id, xcc_mask)
 		dqm->dev->kfd2kgd->program_sh_mem_settings(
-						dqm->dev->adev, qpd->vmid,
-						qpd->sh_mem_config,
-						qpd->sh_mem_ape1_base,
-						qpd->sh_mem_ape1_limit,
-						qpd->sh_mem_bases,
-						dqm->dev->start_xcc_id + xcc);
+			dqm->dev->adev, qpd->vmid, qpd->sh_mem_config,
+			qpd->sh_mem_ape1_base, qpd->sh_mem_ape1_limit,
+			qpd->sh_mem_bases, xcc_id);
 }
 
 static void kfd_hws_hang(struct device_queue_manager *dqm)
@@ -427,14 +425,14 @@ static void deallocate_doorbell(struct qcm_process_device *qpd,
 static void program_trap_handler_settings(struct device_queue_manager *dqm,
 				struct qcm_process_device *qpd)
 {
-	int xcc = 0;
+	uint32_t xcc_mask = dqm->dev->xcc_mask;
+	int xcc_id;
 
 	if (dqm->dev->kfd2kgd->program_trap_handler_settings)
-		for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+		for_each_inst(xcc_id, xcc_mask)
 			dqm->dev->kfd2kgd->program_trap_handler_settings(
-						dqm->dev->adev, qpd->vmid,
-						qpd->tba_addr, qpd->tma_addr,
-						dqm->dev->start_xcc_id + xcc);
+				dqm->dev->adev, qpd->vmid, qpd->tba_addr,
+				qpd->tma_addr, xcc_id);
 }
 
 static int allocate_vmid(struct device_queue_manager *dqm,
@@ -697,7 +695,8 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process
 	struct kfd_process_device *pdd;
 	int first_vmid_to_scan = dev->vm_info.first_vmid_kfd;
 	int last_vmid_to_scan = dev->vm_info.last_vmid_kfd;
-	int xcc = 0;
+	uint32_t xcc_mask = dev->xcc_mask;
+	int xcc_id;
 
 	reg_sq_cmd.u32All = 0;
 	reg_gfx_index.u32All = 0;
@@ -742,11 +741,10 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process
 	reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_KILL;
 	reg_sq_cmd.bits.vm_id = vmid;
 
-	for (xcc = 0; xcc < dev->num_xcc_per_node; xcc++)
-		dev->kfd2kgd->wave_control_execute(dev->adev,
-					reg_gfx_index.u32All,
-					reg_sq_cmd.u32All,
-					dev->start_xcc_id + xcc);
+	for_each_inst(xcc_id, xcc_mask)
+		dev->kfd2kgd->wave_control_execute(
+			dev->adev, reg_gfx_index.u32All,
+			reg_sq_cmd.u32All, xcc_id);
 
 	return 0;
 }
@@ -1258,12 +1256,12 @@ static int
 set_pasid_vmid_mapping(struct device_queue_manager *dqm, u32 pasid,
 			unsigned int vmid)
 {
-	int xcc = 0, ret;
+	uint32_t xcc_mask = dqm->dev->xcc_mask;
+	int xcc_id, ret;
 
-	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++) {
+	for_each_inst(xcc_id, xcc_mask) {
 		ret = dqm->dev->kfd2kgd->set_pasid_vmid_mapping(
-						dqm->dev->adev, pasid, vmid,
-						dqm->dev->start_xcc_id + xcc);
+			dqm->dev->adev, pasid, vmid, xcc_id);
 		if (ret)
 			break;
 	}
@@ -1273,15 +1271,14 @@ set_pasid_vmid_mapping(struct device_queue_manager *dqm, u32 pasid,
 
 static void init_interrupts(struct device_queue_manager *dqm)
 {
-	unsigned int i, xcc;
+	uint32_t xcc_mask = dqm->dev->xcc_mask;
+	unsigned int i, xcc_id;
 
 	for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) {
 		if (is_pipe_enabled(dqm, 0, i)) {
-			for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++)
+			for_each_inst(xcc_id, xcc_mask)
 				dqm->dev->kfd2kgd->init_interrupts(
-							dqm->dev->adev, i,
-							dqm->dev->start_xcc_id +
-							xcc);
+					dqm->dev->adev, i, xcc_id);
 		}
 	}
 }
@@ -2283,7 +2280,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm)
 		get_num_all_sdma_engines(dqm) *
 		dev->kfd->device_info.num_sdma_queues_per_engine +
 		(dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size *
-		dqm->dev->num_xcc_per_node);
+		NUM_XCC(dqm->dev->xcc_mask));
 
 	retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size,
 		&(mem_obj->gtt_mem), &(mem_obj->gpu_addr),
@@ -2489,10 +2486,10 @@ static void seq_reg_dump(struct seq_file *m,
 int dqm_debugfs_hqds(struct seq_file *m, void *data)
 {
 	struct device_queue_manager *dqm = data;
+	uint32_t xcc_mask = dqm->dev->xcc_mask;
 	uint32_t (*dump)[2], n_regs;
 	int pipe, queue;
-	int r = 0, xcc;
-	uint32_t inst;
+	int r = 0, xcc_id;
 	uint32_t sdma_engine_start;
 
 	if (!dqm->sched_running) {
@@ -2500,16 +2497,18 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 		return 0;
 	}
 
-	for (xcc = 0; xcc < dqm->dev->num_xcc_per_node; xcc++) {
-		inst = dqm->dev->start_xcc_id + xcc;
+	for_each_inst(xcc_id, xcc_mask) {
 		r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev,
-					KFD_CIK_HIQ_PIPE, KFD_CIK_HIQ_QUEUE,
-					&dump, &n_regs, inst);
+						KFD_CIK_HIQ_PIPE,
+						KFD_CIK_HIQ_QUEUE, &dump,
+						&n_regs, xcc_id);
 		if (!r) {
-			seq_printf(m,
+			seq_printf(
+				m,
 				"   Inst %d, HIQ on MEC %d Pipe %d Queue %d\n",
-				inst, KFD_CIK_HIQ_PIPE/get_pipes_per_mec(dqm)+1,
-				KFD_CIK_HIQ_PIPE%get_pipes_per_mec(dqm),
+				xcc_id,
+				KFD_CIK_HIQ_PIPE / get_pipes_per_mec(dqm) + 1,
+				KFD_CIK_HIQ_PIPE % get_pipes_per_mec(dqm),
 				KFD_CIK_HIQ_QUEUE);
 			seq_reg_dump(m, dump, n_regs);
 
@@ -2524,13 +2523,16 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data)
 				      dqm->dev->kfd->shared_resources.cp_queue_bitmap))
 					continue;
 
-				r = dqm->dev->kfd2kgd->hqd_dump(
-					dqm->dev->adev, pipe, queue, &dump, &n_regs, inst);
+				r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev,
+								pipe, queue,
+								&dump, &n_regs,
+								xcc_id);
 				if (r)
 					break;
 
-				seq_printf(m, " Inst %d,  CP Pipe %d, Queue %d\n",
-					  inst, pipe, queue);
+				seq_printf(m,
+					   " Inst %d,  CP Pipe %d, Queue %d\n",
+					   xcc_id, pipe, queue);
 				seq_reg_dump(m, dump, n_regs);
 
 				kfree(dump);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
index d81125421aaff..863cf060af484 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
@@ -77,7 +77,7 @@ struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev,
 		dev->dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size;
 
 	offset += dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size *
-		  dev->num_xcc_per_node;
+		  NUM_XCC(dev->xcc_mask);
 
 	mqd_mem_obj->gtt_mem = (void *)((uint64_t)dev->dqm->hiq_sdma_mqd.gtt_mem
 				+ offset);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index c781314b213c5..226132ec37145 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -128,7 +128,7 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node,
 		retval = amdgpu_amdkfd_alloc_gtt_mem(node->adev,
 			(ALIGN(q->ctl_stack_size, PAGE_SIZE) +
 			ALIGN(sizeof(struct v9_mqd), PAGE_SIZE)) *
-			node->num_xcc_per_node,
+			NUM_XCC(node->xcc_mask),
 			&(mqd_mem_obj->gtt_mem),
 			&(mqd_mem_obj->gpu_addr),
 			(void *)&(mqd_mem_obj->cpu_ptr), true);
@@ -482,7 +482,7 @@ static void init_mqd_hiq_v9_4_3(struct mqd_manager *mm, void **mqd,
 
 	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+	for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) {
 		kfd_get_hiq_xcc_mqd(mm->dev, &xcc_mqd_mem_obj, xcc);
 
 		init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q);
@@ -506,21 +506,21 @@ static int hiq_load_mqd_kiq_v9_4_3(struct mqd_manager *mm, void *mqd,
 			uint32_t pipe_id, uint32_t queue_id,
 			struct queue_properties *p, struct mm_struct *mms)
 {
-	int xcc, err;
+	uint32_t xcc_mask = mm->dev->xcc_mask;
+	int xcc_id, err, inst = 0;
 	void *xcc_mqd;
-	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
-		xcc_mqd = mqd + hiq_mqd_size * xcc;
+	for_each_inst(xcc_id, xcc_mask) {
+		xcc_mqd = mqd + hiq_mqd_size * inst;
 		err = mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, xcc_mqd,
 						     pipe_id, queue_id,
-						     p->doorbell_off,
-						     start_inst+xcc);
+						     p->doorbell_off, xcc_id);
 		if (err) {
-			pr_debug("Failed to load HIQ MQD for XCC: %d\n", xcc);
+			pr_debug("Failed to load HIQ MQD for XCC: %d\n", inst);
 			break;
 		}
+		++inst;
 	}
 
 	return err;
@@ -530,20 +530,21 @@ static int destroy_hiq_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 			enum kfd_preempt_type type, unsigned int timeout,
 			uint32_t pipe_id, uint32_t queue_id)
 {
-	int xcc = 0, err;
+	uint32_t xcc_mask = mm->dev->xcc_mask;
+	int xcc_id, err, inst = 0;
 	void *xcc_mqd;
-	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev);
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
-		xcc_mqd = mqd + hiq_mqd_size * xcc;
+	for_each_inst(xcc_id, xcc_mask) {
+		xcc_mqd = mqd + hiq_mqd_size * inst;
 		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
 						    type, timeout, pipe_id,
-						    queue_id, start_inst+xcc);
+						    queue_id, xcc_id);
 		if (err) {
-			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
+			pr_debug("Destroy MQD failed for xcc: %d\n", inst);
 			break;
 		}
+		++inst;
 	}
 
 	return err;
@@ -573,7 +574,7 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 	uint32_t local_xcc_start = mm->dev->dqm->current_logical_xcc_start++;
 
 	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+	for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) {
 		get_xcc_mqd(mqd_mem_obj, &xcc_mqd_mem_obj, offset*xcc);
 
 		init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q);
@@ -600,7 +601,7 @@ static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 			m->compute_tg_chunk_size = 1;
 			m->compute_current_logic_xcc_id =
 					(local_xcc_start + xcc) %
-					mm->dev->num_xcc_per_node;
+					NUM_XCC(mm->dev->xcc_mask);
 
 			switch (xcc) {
 			case 0:
@@ -633,7 +634,7 @@ static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 	int xcc = 0;
 	uint64_t size = mm->mqd_stride(mm, q);
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+	for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) {
 		m = get_mqd(mqd + size * xcc);
 		update_mqd(mm, m, q, minfo);
 
@@ -661,24 +662,25 @@ static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 		   enum kfd_preempt_type type, unsigned int timeout,
 		   uint32_t pipe_id, uint32_t queue_id)
 {
-	int xcc = 0, err;
+	uint32_t xcc_mask = mm->dev->xcc_mask;
+	int xcc_id, err, inst = 0;
 	void *xcc_mqd;
 	struct v9_mqd *m;
 	uint64_t mqd_offset;
-	uint32_t start_inst = mm->dev->start_xcc_id;
 
 	m = get_mqd(mqd);
 	mqd_offset = m->cp_mqd_stride_size;
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
-		xcc_mqd = mqd + mqd_offset * xcc;
+	for_each_inst(xcc_id, xcc_mask) {
+		xcc_mqd = mqd + mqd_offset * inst;
 		err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd,
 						    type, timeout, pipe_id,
-						    queue_id, start_inst+xcc);
+						    queue_id, xcc_id);
 		if (err) {
-			pr_debug("Destroy MQD failed for xcc: %d\n", xcc);
+			pr_debug("Destroy MQD failed for xcc: %d\n", inst);
 			break;
 		}
+		++inst;
 	}
 
 	return err;
@@ -690,21 +692,22 @@ static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
 {
 	/* AQL write pointer counts in 64B packets, PM4/CP counts in dwords. */
 	uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0);
-	int xcc = 0, err;
+	uint32_t xcc_mask = mm->dev->xcc_mask;
+	int xcc_id, err, inst = 0;
 	void *xcc_mqd;
-	uint32_t start_inst = mm->dev->start_xcc_id;
 	uint64_t mqd_stride_size = mm->mqd_stride(mm, p);
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
-		xcc_mqd = mqd + mqd_stride_size * xcc;
-		err = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, xcc_mqd,
-					 pipe_id, queue_id,
-					(uint32_t __user *)p->write_ptr,
-					wptr_shift, 0, mms, start_inst+xcc);
+	for_each_inst(xcc_id, xcc_mask) {
+		xcc_mqd = mqd + mqd_stride_size * inst;
+		err = mm->dev->kfd2kgd->hqd_load(
+			mm->dev->adev, xcc_mqd, pipe_id, queue_id,
+			(uint32_t __user *)p->write_ptr, wptr_shift, 0, mms,
+			xcc_id);
 		if (err) {
-			pr_debug("Load MQD failed for xcc: %d\n", xcc);
+			pr_debug("Load MQD failed for xcc: %d\n", inst);
 			break;
 		}
+		++inst;
 	}
 
 	return err;
@@ -722,7 +725,7 @@ static int get_wave_state_v9_4_3(struct mqd_manager *mm, void *mqd,
 	uint64_t mqd_stride_size = mm->mqd_stride(mm, q);
 	u32 tmp_ctl_stack_used_size = 0, tmp_save_area_used_size = 0;
 
-	for (xcc = 0; xcc < mm->dev->num_xcc_per_node; xcc++) {
+	for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) {
 		xcc_mqd = mqd + mqd_stride_size * xcc;
 		xcc_ctl_stack = (void __user *)((uintptr_t)ctl_stack +
 					q->ctx_save_restore_area_size * xcc);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 559ac5efdc260..02a90fd7f646b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -274,10 +274,6 @@ struct kfd_node {
 					      */
 	struct kfd_vmid_info vm_info;
 	unsigned int id;                /* topology stub index */
-	unsigned int num_xcc_per_node;
-	unsigned int start_xcc_id;	/* Starting XCC instance
-					 * number for the node
-					 */
 	uint32_t xcc_mask; /* Instance mask of XCCs present */
 	struct amdgpu_xcp *xcp;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index a6ff57f114725..7f7d1378a2f89 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -2058,6 +2058,7 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 	struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv);
 	uint64_t tlb_seq = amdgpu_vm_tlb_seq(vm);
 	struct kfd_node *dev = pdd->dev;
+	uint32_t xcc_mask = dev->xcc_mask;
 	int xcc = 0;
 
 	/*
@@ -2076,10 +2077,9 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 			amdgpu_amdkfd_flush_gpu_tlb_vmid(dev->adev,
 							pdd->qpd.vmid);
 	} else {
-		for (xcc = 0; xcc < dev->num_xcc_per_node; xcc++)
-			amdgpu_amdkfd_flush_gpu_tlb_pasid(dev->adev,
-					pdd->process->pasid, type,
-					dev->start_xcc_id + xcc);
+		for_each_inst(xcc, xcc_mask)
+			amdgpu_amdkfd_flush_gpu_tlb_pasid(
+				dev->adev, pdd->process->pasid, type, xcc);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 2b2ae0c9902bd..a3c23d07c7df5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -946,7 +946,7 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data)
 				seq_printf(m, "  Compute queue on device %x\n",
 					   q->device->id);
 				mqd_type = KFD_MQD_TYPE_CP;
-				num_xccs = q->device->num_xcc_per_node;
+				num_xccs = NUM_XCC(q->device->xcc_mask);
 				break;
 			default:
 				seq_printf(m,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index c7072fff778ec..d2a42b6b1fa89 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -469,7 +469,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 			      dev->node_props.cpu_cores_count);
 	sysfs_show_32bit_prop(buffer, offs, "simd_count",
 			      dev->gpu ? (dev->node_props.simd_count *
-					  dev->gpu->num_xcc_per_node) : 0);
+					  NUM_XCC(dev->gpu->xcc_mask)) : 0);
 	sysfs_show_32bit_prop(buffer, offs, "mem_banks_count",
 			      dev->node_props.mem_banks_count);
 	sysfs_show_32bit_prop(buffer, offs, "caches_count",
@@ -494,7 +494,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 			      dev->node_props.wave_front_size);
 	sysfs_show_32bit_prop(buffer, offs, "array_count",
 			      dev->gpu ? (dev->node_props.array_count *
-					  dev->gpu->num_xcc_per_node) : 0);
+					  NUM_XCC(dev->gpu->xcc_mask)) : 0);
 	sysfs_show_32bit_prop(buffer, offs, "simd_arrays_per_engine",
 			      dev->node_props.simd_arrays_per_engine);
 	sysfs_show_32bit_prop(buffer, offs, "cu_per_simd_array",
@@ -558,7 +558,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 		sysfs_show_64bit_prop(buffer, offs, "unique_id",
 				      dev->gpu->adev->unique_id);
 		sysfs_show_32bit_prop(buffer, offs, "num_xcc",
-				      dev->gpu->num_xcc_per_node);
+				      NUM_XCC(dev->gpu->xcc_mask));
 	}
 
 	return sysfs_show_32bit_prop(buffer, offs, "max_engine_clk_ccompute",
@@ -1180,7 +1180,7 @@ static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 	buf[4] = gpu->adev->pdev->bus->number;
 	buf[5] = lower_32_bits(local_mem_size);
 	buf[6] = upper_32_bits(local_mem_size);
-	buf[7] = gpu->start_xcc_id | (gpu->num_xcc_per_node << 16);
+	buf[7] = (ffs(gpu->xcc_mask) - 1) | (NUM_XCC(gpu->xcc_mask) << 16);
 
 	for (i = 0, hashout = 0; i < 8; i++)
 		hashout ^= hash_32(buf[i], KFD_GPU_ID_HASH_WIDTH);
-- 
GitLab


From 1589c82a10852c6de742e5d6a92042a3fd68d753 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 13 Feb 2023 18:50:07 +0530
Subject: [PATCH 1266/1791] drm/amdgpu: Check memory ranges for valid xcp mode

Check the memory ranges available to the device also for deciding a
valid partition mode. Only select combinations are valid for a
particular mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Philip Yang <philip.yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 004400fb89b01..7469de3fd6fe1 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -238,21 +238,28 @@ int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 					  enum amdgpu_gfx_partition mode)
 {
+	struct amdgpu_device *adev = xcp_mgr->adev;
 	int num_xcc, num_xccs_per_xcp;
 
-	num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask);
+	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	switch (mode) {
 	case AMDGPU_SPX_PARTITION_MODE:
-		return num_xcc > 0;
+		return adev->gmc.num_mem_partitions == 1 && num_xcc > 0;
 	case AMDGPU_DPX_PARTITION_MODE:
-		return (num_xcc % 4) == 0;
+		return adev->gmc.num_mem_partitions != 8 && (num_xcc % 4) == 0;
 	case AMDGPU_TPX_PARTITION_MODE:
-		return (num_xcc % 3) == 0;
+		return (adev->gmc.num_mem_partitions == 1 ||
+			adev->gmc.num_mem_partitions == 3) &&
+		       ((num_xcc % 3) == 0);
 	case AMDGPU_QPX_PARTITION_MODE:
 		num_xccs_per_xcp = num_xcc / 4;
-		return (num_xccs_per_xcp >= 2);
+		return (adev->gmc.num_mem_partitions == 1 ||
+			adev->gmc.num_mem_partitions == 4) &&
+		       (num_xccs_per_xcp >= 2);
 	case AMDGPU_CPX_PARTITION_MODE:
-		return (num_xcc > 1);
+		return (num_xcc > 1) &&
+		       (adev->gmc.num_mem_partitions == 1 ||
+			adev->gmc.num_mem_partitions == num_xcc);
 	default:
 		return false;
 	}
-- 
GitLab


From 570de94b9c5d93e1c5bc4e357946efb93c662da9 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 13 Feb 2023 19:26:18 +0530
Subject: [PATCH 1267/1791] drm/amdgpu: Add auto mode for compute partition

When auto mode is specified, driver will choose the right compute
partition mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Philip Yang <philip.yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h           |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c       |  8 ++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h       |  2 ++
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 28 ++++++++++++++++++-
 4 files changed, 35 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index f2bafab15cebd..cb9373f8c25ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -242,7 +242,7 @@ extern int amdgpu_num_kcq;
 extern int amdgpu_vcnfw_log;
 extern int amdgpu_sg_display;
 
-extern uint amdgpu_user_partt_mode;
+extern int amdgpu_user_partt_mode;
 
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index f319318a88134..da4e50aef95a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -193,7 +193,7 @@ int amdgpu_smartshift_bias;
 int amdgpu_use_xgmi_p2p = 1;
 int amdgpu_vcnfw_log;
 int amdgpu_sg_display = -1; /* auto */
-uint amdgpu_user_partt_mode;
+int amdgpu_user_partt_mode = AMDGPU_AUTO_COMPUTE_PARTITION_MODE;
 
 static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work);
 
@@ -955,8 +955,10 @@ module_param_named(smu_pptable_id, amdgpu_smu_pptable_id, int, 0444);
  * DOC: partition_mode (int)
  * Used to override the default SPX mode.
  */
-MODULE_PARM_DESC(user_partt_mode,
-	"specify partition mode to be used (0 = AMDGPU_SPX_PARTITION_MODE(default value), \
+MODULE_PARM_DESC(
+	user_partt_mode,
+	"specify partition mode to be used (-2 = AMDGPU_AUTO_COMPUTE_PARTITION_MODE(default value) \
+						0 = AMDGPU_SPX_PARTITION_MODE, \
 						1 = AMDGPU_DPX_PARTITION_MODE, \
 						2 = AMDGPU_TPX_PARTITION_MODE, \
 						3 = AMDGPU_QPX_PARTITION_MODE, \
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 728977f8afe7a..e9c93f6e12b88 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -62,6 +62,8 @@ enum amdgpu_gfx_partition {
 	AMDGPU_QPX_PARTITION_MODE = 3,
 	AMDGPU_CPX_PARTITION_MODE = 4,
 	AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE = -1,
+	/* Automatically choose the right mode */
+	AMDGPU_AUTO_COMPUTE_PARTITION_MODE = -2,
 };
 
 #define NUM_XCC(x) hweight16(x)
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 7469de3fd6fe1..a165b51e9e58f 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -235,6 +235,30 @@ int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 	return 0;
 }
 
+static enum amdgpu_gfx_partition
+__aqua_vanjaram_get_auto_mode(struct amdgpu_xcp_mgr *xcp_mgr)
+{
+	struct amdgpu_device *adev = xcp_mgr->adev;
+	int num_xcc;
+
+	num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask);
+
+	if (adev->gmc.num_mem_partitions == 1)
+		return AMDGPU_SPX_PARTITION_MODE;
+
+	if (adev->gmc.num_mem_partitions == num_xcc)
+		return AMDGPU_CPX_PARTITION_MODE;
+
+	if (adev->gmc.num_mem_partitions == num_xcc / 2)
+		return (adev->flags & AMD_IS_APU) ? AMDGPU_TPX_PARTITION_MODE :
+						    AMDGPU_QPX_PARTITION_MODE;
+
+	if (adev->gmc.num_mem_partitions == 2 && !(adev->flags & AMD_IS_APU))
+		return AMDGPU_DPX_PARTITION_MODE;
+
+	return AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE;
+}
+
 static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 					  enum amdgpu_gfx_partition mode)
 {
@@ -304,7 +328,9 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 	adev = xcp_mgr->adev;
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
-	if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode))
+	if (mode == AMDGPU_AUTO_COMPUTE_PARTITION_MODE)
+		mode = __aqua_vanjaram_get_auto_mode(xcp_mgr);
+	else if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode))
 		return -EINVAL;
 
 	if (adev->kfd.init_complete)
-- 
GitLab


From 6b43e1a05cb764196c5158b2447a9bfad1f2b531 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Mon, 13 Feb 2023 23:51:07 -0500
Subject: [PATCH 1268/1791] drm/ttm: export ttm_pool_fini for cleanup

ttm_pool_init is exported and used outside of ttm subsystem with
amdgpu_ttm interface, similarly export ttm_pool_fini for proper cleanup.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/ttm/ttm_pool.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c
index afc4476c446fa..668abf63f2bd6 100644
--- a/drivers/gpu/drm/ttm/ttm_pool.c
+++ b/drivers/gpu/drm/ttm/ttm_pool.c
@@ -600,6 +600,7 @@ void ttm_pool_fini(struct ttm_pool *pool)
 	 */
 	synchronize_shrinkers();
 }
+EXPORT_SYMBOL(ttm_pool_fini);
 
 /* As long as pages are available make sure to release at least one */
 static unsigned long ttm_pool_shrinker_scan(struct shrinker *shrink,
-- 
GitLab


From 1e03322cfef9b83aa87ea0a508588f9f05a47dfc Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Mon, 27 Feb 2023 11:16:09 -0500
Subject: [PATCH 1269/1791] drm/amdgpu: Set TTM pools for memory partitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For native mode only, create TTM pool for each memory partition to store
the NUMA node id, then the TTM pool will be selected using memory
partition id to allocate memory from the correct partition.

Acked-by: Christian König <christian.koenig@amd.com>
(rajneesh: changed need_swiotlb and need_dma32 to false for pool init)
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-and-tested-by:  Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 61 ++++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h |  1 +
 2 files changed, 60 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index c8a2d030f226c..7674109810b06 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -631,6 +631,7 @@ struct amdgpu_ttm_tt {
 	struct task_struct	*usertask;
 	uint32_t		userflags;
 	bool			bound;
+	int32_t			pool_id;
 };
 
 #define ttm_to_amdgpu_ttm_tt(ptr)	container_of(ptr, struct amdgpu_ttm_tt, ttm)
@@ -1059,6 +1060,7 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 		return NULL;
 	}
 	gtt->gobj = &bo->base;
+	gtt->pool_id = NUMA_NO_NODE;
 
 	if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 		caching = ttm_write_combined;
@@ -1085,6 +1087,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev,
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bdev);
 	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
+	struct ttm_pool *pool;
 	pgoff_t i;
 	int ret;
 
@@ -1099,7 +1102,11 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev,
 	if (ttm->page_flags & TTM_TT_FLAG_EXTERNAL)
 		return 0;
 
-	ret = ttm_pool_alloc(&adev->mman.bdev.pool, ttm, ctx);
+	if (adev->mman.ttm_pools && gtt->pool_id >= 0)
+		pool = &adev->mman.ttm_pools[gtt->pool_id];
+	else
+		pool = &adev->mman.bdev.pool;
+	ret = ttm_pool_alloc(pool, ttm, ctx);
 	if (ret)
 		return ret;
 
@@ -1120,6 +1127,7 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev,
 {
 	struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm);
 	struct amdgpu_device *adev;
+	struct ttm_pool *pool;
 	pgoff_t i;
 
 	amdgpu_ttm_backend_unbind(bdev, ttm);
@@ -1138,7 +1146,13 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev,
 		ttm->pages[i]->mapping = NULL;
 
 	adev = amdgpu_ttm_adev(bdev);
-	return ttm_pool_free(&adev->mman.bdev.pool, ttm);
+
+	if (adev->mman.ttm_pools && gtt->pool_id >= 0)
+		pool = &adev->mman.ttm_pools[gtt->pool_id];
+	else
+		pool = &adev->mman.bdev.pool;
+
+	return ttm_pool_free(pool, ttm);
 }
 
 /**
@@ -1728,6 +1742,41 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	return 0;
 }
 
+static int amdgpu_ttm_pools_init(struct amdgpu_device *adev)
+{
+	int i;
+
+	if (!adev->gmc.is_app_apu || !adev->gmc.num_mem_partitions)
+		return 0;
+
+	adev->mman.ttm_pools = kcalloc(adev->gmc.num_mem_partitions,
+				       sizeof(*adev->mman.ttm_pools),
+				       GFP_KERNEL);
+	if (!adev->mman.ttm_pools)
+		return -ENOMEM;
+
+	for (i = 0; i < adev->gmc.num_mem_partitions; i++) {
+		ttm_pool_init(&adev->mman.ttm_pools[i], adev->dev,
+			      adev->gmc.mem_partitions[i].numa.node,
+			      false, false);
+	}
+	return 0;
+}
+
+static void amdgpu_ttm_pools_fini(struct amdgpu_device *adev)
+{
+	int i;
+
+	if (!adev->gmc.is_app_apu || !adev->mman.ttm_pools)
+		return;
+
+	for (i = 0; i < adev->gmc.num_mem_partitions; i++)
+		ttm_pool_fini(&adev->mman.ttm_pools[i]);
+
+	kfree(adev->mman.ttm_pools);
+	adev->mman.ttm_pools = NULL;
+}
+
 /*
  * amdgpu_ttm_init - Init the memory management (ttm) as well as various
  * gtt/vram related fields.
@@ -1754,6 +1803,12 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		DRM_ERROR("failed initializing buffer object driver(%d).\n", r);
 		return r;
 	}
+
+	r = amdgpu_ttm_pools_init(adev);
+	if (r) {
+		DRM_ERROR("failed to init ttm pools(%d).\n", r);
+		return r;
+	}
 	adev->mman.initialized = true;
 
 	/* Initialize VRAM pool with all of VRAM divided into pages */
@@ -1901,6 +1956,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	if (!adev->mman.initialized)
 		return;
 
+	amdgpu_ttm_pools_fini(adev);
+
 	amdgpu_ttm_training_reserve_vram_fini(adev);
 	/* return the stolen vga memory back to VRAM */
 	if (!adev->gmc.is_app_apu) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index da6544fdc8ddd..8ef048a0a33e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -49,6 +49,7 @@ struct amdgpu_gtt_mgr {
 
 struct amdgpu_mman {
 	struct ttm_device		bdev;
+	struct ttm_pool			*ttm_pools;
 	bool				initialized;
 	void __iomem			*aper_base_kaddr;
 
-- 
GitLab


From fcfefd85f18a0004c7c7b499f0701fd2c76d4c68 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Mon, 27 Feb 2023 20:08:29 -0500
Subject: [PATCH 1270/1791] drm/amdkfd: Native mode memory partition support

For native mode, after amdgpu_bo is created on CPU domain, then call
amdgpu_ttm_tt_set_mem_pool to select the TTM pool using bo->mem_id.
ttm_bo_validate will allocate the memory to the correct memory partition
before mapping to GPUs.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-and-tested-by:  Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c   |  7 +++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c        | 18 ++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h        |  1 +
 3 files changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index bbdd5e3aa18e9..59404b3e6b876 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1643,6 +1643,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	uint64_t aligned_size;
 	u64 alloc_flags;
 	int ret;
+	int mem_id = 0; /* Fixme : to be changed when mem_id support patch lands, until then NPS1, SPX only */
 
 	/*
 	 * Check on which domain to allocate BO
@@ -1750,6 +1751,11 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	    ((*mem)->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)) {
 		bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
 		bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
+		ret = amdgpu_ttm_tt_set_mem_pool(&bo->tbo, mem_id);
+		if (ret) {
+			pr_debug("failed to set ttm mem pool %d\n", ret);
+			goto err_set_mem_partition;
+		}
 	}
 
 	add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr);
@@ -1778,6 +1784,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 allocate_init_user_pages_failed:
 err_pin_bo:
 	remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info);
+err_set_mem_partition:
 	drm_vma_node_revoke(&gobj->vma_node, drm_priv);
 err_node_allow:
 	/* Don't unreserve system mem limit twice */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 7674109810b06..3933432daaac8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1155,6 +1155,24 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev,
 	return ttm_pool_free(pool, ttm);
 }
 
+/**
+ * amdgpu_ttm_tt_set_mem_pool - Set the TTM memory pool for the TTM BO
+ * @tbo: The ttm_buffer_object that backs the VRAM bo
+ * @mem_id: to select the initialized ttm pool corresponding to the memory partition
+ */
+int amdgpu_ttm_tt_set_mem_pool(struct ttm_buffer_object *tbo, int mem_id)
+{
+	struct ttm_tt *ttm = tbo->ttm;
+	struct amdgpu_ttm_tt *gtt;
+
+	if (!ttm && !ttm_tt_is_populated(ttm))
+		return -EINVAL;
+
+	gtt = ttm_to_amdgpu_ttm_tt(ttm);
+	gtt->pool_id = mem_id;
+	return 0;
+}
+
 /**
  * amdgpu_ttm_tt_get_userptr - Return the userptr GTT ttm_tt for the current
  * task
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 8ef048a0a33e7..fe32de1bf4d51 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -192,6 +192,7 @@ bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm);
 struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm);
 bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
 				  unsigned long end, unsigned long *userptr);
+int amdgpu_ttm_tt_set_mem_pool(struct ttm_buffer_object *tbo, int mem_id);
 bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
 				       int *last_invalidated);
 bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm);
-- 
GitLab


From e181be58ccc2ac48e4b79996c8dd6dd9f34fa4b5 Mon Sep 17 00:00:00 2001
From: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Date: Mon, 27 Feb 2023 13:17:14 -0500
Subject: [PATCH 1271/1791] drm/amdgpu: Fix xGMI access P2P mapping failure on
 GFXIP 9.4.3

On GFXIP 9.4.3, we dont need to rely on xGMI hive info to determine P2P
access.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-and-tested-by:  Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 59404b3e6b876..c4b949d17e14c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -814,7 +814,7 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem,
 	 * if peer device has large BAR. In contrast, access over xGMI is
 	 * allowed for both small and large BAR configurations of peer device
 	 */
-	if ((adev != bo_adev) &&
+	if ((adev != bo_adev && !adev->gmc.is_app_apu) &&
 	    ((mem->domain == AMDGPU_GEM_DOMAIN_VRAM) ||
 	     (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) ||
 	     (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP))) {
-- 
GitLab


From a0ba127960982b8827ba8b410c272ec8f3ee7e6a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 3 Mar 2023 18:03:00 +0530
Subject: [PATCH 1272/1791] drm/amdgpu: Fix unmapping of aperture

When aperture size is zero, there is no mapping done.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 7 ++++---
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 7f62826fcaca9..19a3f9ba545b4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -4121,7 +4121,7 @@ static void amdgpu_device_unmap_mmio(struct amdgpu_device *adev)
 	adev->mman.aper_base_kaddr = NULL;
 
 	/* Memory manager related */
-	if (!adev->gmc.xgmi.connected_to_cpu) {
+	if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) {
 		arch_phys_wc_del(adev->gmc.vram_mtrr);
 		arch_io_free_memtype_wc(adev->gmc.aper_base, adev->gmc.aper_size);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index f431205e1077f..725530eb4e0c1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1080,8 +1080,7 @@ void amdgpu_bo_fini(struct amdgpu_device *adev)
 	amdgpu_ttm_fini(adev);
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
-
-		if (!adev->gmc.xgmi.connected_to_cpu) {
+		if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) {
 			arch_phys_wc_del(adev->gmc.vram_mtrr);
 			arch_io_free_memtype_wc(adev->gmc.aper_base, adev->gmc.aper_size);
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 3933432daaac8..09d1a98bd11ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1844,12 +1844,13 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		adev->mman.aper_base_kaddr = ioremap_cache(adev->gmc.aper_base,
 				adev->gmc.visible_vram_size);
 
-	else if (!adev->gmc.is_app_apu)
+	else if (adev->gmc.is_app_apu)
+		DRM_DEBUG_DRIVER(
+			"No need to ioremap when real vram size is 0\n");
+	else
 #endif
 		adev->mman.aper_base_kaddr = ioremap_wc(adev->gmc.aper_base,
 				adev->gmc.visible_vram_size);
-	else
-		DRM_DEBUG_DRIVER("No need to ioremap when real vram size is 0\n");
 #endif
 
 	/*
-- 
GitLab


From f9632096be49ed31e86541e3e79960e21e8f1578 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 7 Mar 2023 10:33:05 +0530
Subject: [PATCH 1273/1791] drm/amdgpu: Add compute mode descriptor function

Keep a helper function to get description of compute partition mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 24 +-----------------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 2f7a101593e78..6098b8b1ae5b1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1175,33 +1175,11 @@ static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = drm_to_adev(ddev);
 	int mode;
-	char *partition_mode;
 
 	mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
 					       AMDGPU_XCP_FL_NONE);
 
-	switch (mode) {
-	case AMDGPU_SPX_PARTITION_MODE:
-		partition_mode = "SPX";
-		break;
-	case AMDGPU_DPX_PARTITION_MODE:
-		partition_mode = "DPX";
-		break;
-	case AMDGPU_TPX_PARTITION_MODE:
-		partition_mode = "TPX";
-		break;
-	case AMDGPU_QPX_PARTITION_MODE:
-		partition_mode = "QPX";
-		break;
-	case AMDGPU_CPX_PARTITION_MODE:
-		partition_mode = "CPX";
-		break;
-	default:
-		partition_mode = "UNKNOWN";
-		break;
-	}
-
-	return sysfs_emit(buf, "%s\n", partition_mode);
+	return sysfs_emit(buf, "%s\n", amdgpu_gfx_compute_mode_desc(mode));
 }
 
 static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index e9c93f6e12b88..3d11b7a0bd759 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -494,4 +494,25 @@ int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev,
 bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev);
 void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev);
+
+static inline const char *amdgpu_gfx_compute_mode_desc(int mode)
+{
+	switch (mode) {
+	case AMDGPU_SPX_PARTITION_MODE:
+		return "SPX";
+	case AMDGPU_DPX_PARTITION_MODE:
+		return "DPX";
+	case AMDGPU_TPX_PARTITION_MODE:
+		return "TPX";
+	case AMDGPU_QPX_PARTITION_MODE:
+		return "QPX";
+	case AMDGPU_CPX_PARTITION_MODE:
+		return "CPX";
+	default:
+		return "UNKNOWN";
+	}
+
+	return "UNKNOWN";
+}
+
 #endif
-- 
GitLab


From b6b85c8b43a85988ecd06f039f8f90c041842812 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 7 Mar 2023 10:36:08 +0530
Subject: [PATCH 1274/1791] drm/amdgpu: Return error on invalid compute mode

Return error if an invalid compute partition mode is requested.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 8 ++++++--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c             | 6 +++++-
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index a165b51e9e58f..848049db00ab1 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -328,10 +328,14 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 	adev = xcp_mgr->adev;
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
-	if (mode == AMDGPU_AUTO_COMPUTE_PARTITION_MODE)
+	if (mode == AMDGPU_AUTO_COMPUTE_PARTITION_MODE) {
 		mode = __aqua_vanjaram_get_auto_mode(xcp_mgr);
-	else if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode))
+	} else if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode)) {
+		dev_err(adev->dev,
+			"Invalid compute partition mode requested, requested: %s, available memory partitions: %d",
+			amdgpu_gfx_compute_mode_desc(mode), adev->gmc.num_mem_partitions);
 		return -EINVAL;
+	}
 
 	if (adev->kfd.init_complete)
 		flags |= AMDGPU_XCP_OPS_KFD;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 81ab3cd2f2296..d0ddcd751432f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1933,7 +1933,11 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 	if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
 					    AMDGPU_XCP_FL_NONE) ==
 	    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
-		amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, amdgpu_user_partt_mode);
+		r = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr,
+						     amdgpu_user_partt_mode);
+
+	if (r)
+		return r;
 
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
-- 
GitLab


From ba08e9cb6ff87acdb2f28f013fe695a252533f0e Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 10 Mar 2023 15:38:19 +0530
Subject: [PATCH 1275/1791] drm/amdgpu: Add PSP spatial parition interface

Add PSP ring command interface for spatial partitioning.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 21 +++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h |  2 ++
 drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h |  9 +++++++++
 3 files changed, 32 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index ea47012795e7f..d62746b596f59 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -991,6 +991,27 @@ static int psp_rl_load(struct amdgpu_device *adev)
 	return ret;
 }
 
+int psp_spatial_partition(struct psp_context *psp, int mode)
+{
+	struct psp_gfx_cmd_resp *cmd;
+	int ret;
+
+	if (amdgpu_sriov_vf(psp->adev))
+		return 0;
+
+	cmd = acquire_psp_cmd_buf(psp);
+
+	cmd->cmd_id = GFX_CMD_ID_SRIOV_SPATIAL_PART;
+	cmd->cmd.cmd_spatial_part.mode = mode;
+
+	dev_info(psp->adev->dev, "Requesting %d paritions through PSP", mode);
+	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
+
+	release_psp_cmd_buf(psp);
+
+	return ret;
+}
+
 static int psp_asd_initialize(struct psp_context *psp)
 {
 	int ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index cf4f60c661223..0a409da749d10 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -519,6 +519,8 @@ int psp_load_fw_list(struct psp_context *psp,
 		     struct amdgpu_firmware_info **ucode_list, int ucode_count);
 void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size);
 
+int psp_spatial_partition(struct psp_context *psp, int mode);
+
 int is_psp_fw_valid(struct psp_bin_desc bin);
 
 int amdgpu_psp_sysfs_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
index 22c775f39119b..18917df785eca 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h
@@ -102,6 +102,7 @@ enum psp_gfx_cmd_id
     GFX_CMD_ID_LOAD_TOC           = 0x00000020,   /* Load TOC and obtain TMR size */
     GFX_CMD_ID_AUTOLOAD_RLC       = 0x00000021,   /* Indicates all graphics fw loaded, start RLC autoload */
     GFX_CMD_ID_BOOT_CFG           = 0x00000022,   /* Boot Config */
+    GFX_CMD_ID_SRIOV_SPATIAL_PART = 0x00000027,   /* Configure spatial partitioning mode */
 };
 
 /* PSP boot config sub-commands */
@@ -338,6 +339,13 @@ struct psp_gfx_cmd_boot_cfg
     uint32_t                        boot_config_valid;    /* dynamic boot configuration valid bits bitmask */
 };
 
+struct psp_gfx_cmd_sriov_spatial_part {
+	uint32_t mode;
+	uint32_t override_ips;
+	uint32_t override_xcds_avail;
+	uint32_t override_this_aid;
+};
+
 /* All GFX ring buffer commands. */
 union psp_gfx_commands
 {
@@ -351,6 +359,7 @@ union psp_gfx_commands
     struct psp_gfx_cmd_setup_tmr        cmd_setup_vmr;
     struct psp_gfx_cmd_load_toc         cmd_load_toc;
     struct psp_gfx_cmd_boot_cfg         boot_cfg;
+    struct psp_gfx_cmd_sriov_spatial_part cmd_spatial_part;
 };
 
 struct psp_gfx_uresp_reserved
-- 
GitLab


From 63630c9e5c3481c51ac3b4ee058628a01ac91ba8 Mon Sep 17 00:00:00 2001
From: Gavin Wan <Gavin.Wan@amd.com>
Date: Thu, 16 Mar 2023 13:44:41 -0400
Subject: [PATCH 1276/1791] drm/amdgpu: Add PSP supporting PSP 13.0.6 SRIOV
 ucode init.

Add PSP supporting PSP 13.0.6 SRIOV ucode init.

Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index d62746b596f59..ec79a5c2f500c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -146,6 +146,9 @@ static int psp_init_sriov_microcode(struct psp_context *psp)
 	case IP_VERSION(13, 0, 0):
 		adev->virt.autoload_ucode_id = 0;
 		break;
+	case IP_VERSION(13, 0, 6):
+		ret = psp_init_cap_microcode(psp, ucode_prefix);
+		break;
 	case IP_VERSION(13, 0, 10):
 		adev->virt.autoload_ucode_id = AMDGPU_UCODE_ID_CP_MES1_DATA;
 		ret = psp_init_cap_microcode(psp, ucode_prefix);
-- 
GitLab


From b0a3bbf947f6ed690336cec1f6cde2a30d082dbb Mon Sep 17 00:00:00 2001
From: Gavin Wan <Gavin.Wan@amd.com>
Date: Mon, 3 Apr 2023 17:49:41 -0400
Subject: [PATCH 1277/1791] drm/amdgpu: Skip using MC FB Offset when APU flag
 is set for SRIOV.

The MC_VM_FB_OFFSET is PF only register. It cannot be read on VF.
So, the driver should not use MC_VM_FB_OFFSET address to set the
address of dev->gmc.aper_base.

Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Reviewed-by: Zhigang Luo <zhigang.luo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 1653d77df3ba3..58bcd1e1f1b6e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1644,7 +1644,8 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev)
 	 */
 
 	/* check whether both host-gpu and gpu-gpu xgmi links exist */
-	if (((adev->flags & AMD_IS_APU) && !amdgpu_passthrough(adev)) ||
+	if ((!amdgpu_sriov_vf(adev) &&
+		(adev->flags & AMD_IS_APU) && !amdgpu_passthrough(adev)) ||
 	    (adev->gmc.xgmi.supported &&
 	     adev->gmc.xgmi.connected_to_cpu)) {
 		adev->gmc.aper_base =
-- 
GitLab


From 46f7b4deb30558593c1d2e62d561a3cee21f558a Mon Sep 17 00:00:00 2001
From: Gavin Wan <Gavin.Wan@amd.com>
Date: Mon, 10 Apr 2023 15:04:26 -0400
Subject: [PATCH 1278/1791] drm/amdgpu: Set memory partitions to 1 for SRIOV.

For SRIOV, the memory partitions are set on host drover. Each VF only
has one memory partition. We need set the memory partitions to 1 on
guest driver for SRIOV.

V2: sqaush in fix ("drm/amdgpu: Fix memory range info of GC 9.4.3 VFs")

Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Acked-by: Zhigang Luo <zhigang.luo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 58bcd1e1f1b6e..95c3253e240ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1346,6 +1346,9 @@ gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes)
 static enum amdgpu_memory_partition
 gmc_v9_0_query_memory_partition(struct amdgpu_device *adev)
 {
+	if (amdgpu_sriov_vf(adev))
+		return AMDGPU_NPS1_PARTITION_MODE;
+
 	return gmc_v9_0_get_memory_partition(adev, NULL);
 }
 
@@ -1897,7 +1900,10 @@ static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev)
 	else
 		gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions);
 
-	valid = gmc_v9_0_validate_partition_info(adev);
+	if (amdgpu_sriov_vf(adev))
+		valid = true;
+	else
+		valid = gmc_v9_0_validate_partition_info(adev);
 	if (!valid) {
 		/* TODO: handle invalid case */
 		dev_WARN(adev->dev,
-- 
GitLab


From b4520bfd801338c48d887f0eec74a53ab80f2167 Mon Sep 17 00:00:00 2001
From: Gavin Wan <Gavin.Wan@amd.com>
Date: Fri, 17 Mar 2023 18:42:30 -0400
Subject: [PATCH 1279/1791] drm/amdgpu: Checked if the pointer NULL before use
 it.

For SRIOV on some parts, the host driver does not post VBIOS. So the guest
cannot get bios information. Therefore, adev->virt.fw_reserve.p_pf2vf
and adev->mode_info.atom_context are NULL.

Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Reviewed-by: Zhigang Luo <Zhigang.Luo@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 19a3f9ba545b4..0c6f983fb2adc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3855,21 +3855,24 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	}
 
 	/* enable PCIE atomic ops */
-	if (amdgpu_sriov_vf(adev))
-		adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *)
-			adev->virt.fw_reserve.p_pf2vf)->pcie_atomic_ops_support_flags ==
-			(PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64);
+	if (amdgpu_sriov_vf(adev)) {
+		if (adev->virt.fw_reserve.p_pf2vf)
+			adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *)
+						      adev->virt.fw_reserve.p_pf2vf)->pcie_atomic_ops_support_flags ==
+				(PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64);
 	/* APUs w/ gfx9 onwards doesn't reply on PCIe atomics, rather it is a
 	 * internal path natively support atomics, set have_atomics_support to true.
 	 */
-	else if ((adev->flags & AMD_IS_APU) &&
-		(adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0)))
+	} else if ((adev->flags & AMD_IS_APU) &&
+		   (adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0))) {
 		adev->have_atomics_support = true;
-	else
+	} else {
 		adev->have_atomics_support =
 			!pci_enable_atomic_ops_to_root(adev->pdev,
 					  PCI_EXP_DEVCAP2_ATOMIC_COMP32 |
 					  PCI_EXP_DEVCAP2_ATOMIC_COMP64);
+	}
+
 	if (!adev->have_atomics_support)
 		dev_info(adev->dev, "PCIE atomic ops is not supported\n");
 
@@ -3885,7 +3888,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	amdgpu_reset_init(adev);
 
 	/* detect if we are with an SRIOV vbios */
-	amdgpu_device_detect_sriov_bios(adev);
+	if (adev->bios)
+		amdgpu_device_detect_sriov_bios(adev);
 
 	/* check if we need to reset the asic
 	 *  E.g., driver was not cleanly unloaded previously, etc.
-- 
GitLab


From 6c2bebfca43c14967cfb9cf6c2c074b0d755ddd7 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Wed, 1 Mar 2023 10:05:17 +0800
Subject: [PATCH 1280/1791] drm/amdgpu: Add vcn/jpeg ras err status registers

Add new ras error status registers introduced in
vcn v4_0_3 to log vcn and jpeg ras error.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../include/asic_reg/vcn/vcn_4_0_3_offset.h   |  78 +++
 .../include/asic_reg/vcn/vcn_4_0_3_sh_mask.h  | 495 ++++++++++++++++++
 2 files changed, 573 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
index facad93cd06f7..e9742d10de1c6 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h
@@ -1500,6 +1500,84 @@
 #define regVCN_RAS_CNTL_MMSCH                                                                           0x0914
 #define regVCN_RAS_CNTL_MMSCH_BASE_IDX                                                                  1
 
+// addressBlock: aid_uvd0_vcn_edcc_dec
+// base address: 0x21d20
+#define regVCN_UE_ERR_STATUS_LO_VIDD                                                                    0x094c
+#define regVCN_UE_ERR_STATUS_LO_VIDD_BASE_IDX                                                           1
+#define regVCN_UE_ERR_STATUS_HI_VIDD                                                                    0x094d
+#define regVCN_UE_ERR_STATUS_HI_VIDD_BASE_IDX                                                           1
+#define regVCN_UE_ERR_STATUS_LO_VIDV                                                                    0x094e
+#define regVCN_UE_ERR_STATUS_LO_VIDV_BASE_IDX                                                           1
+#define regVCN_UE_ERR_STATUS_HI_VIDV                                                                    0x094f
+#define regVCN_UE_ERR_STATUS_HI_VIDV_BASE_IDX                                                           1
+#define regVCN_CE_ERR_STATUS_LO_MMSCHD                                                                  0x0950
+#define regVCN_CE_ERR_STATUS_LO_MMSCHD_BASE_IDX                                                         1
+#define regVCN_CE_ERR_STATUS_HI_MMSCHD                                                                  0x0951
+#define regVCN_CE_ERR_STATUS_HI_MMSCHD_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG0S                                                                  0x0952
+#define regVCN_UE_ERR_STATUS_LO_JPEG0S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG0S                                                                  0x0953
+#define regVCN_UE_ERR_STATUS_HI_JPEG0S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG0D                                                                  0x0954
+#define regVCN_UE_ERR_STATUS_LO_JPEG0D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG0D                                                                  0x0955
+#define regVCN_UE_ERR_STATUS_HI_JPEG0D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG1S                                                                  0x0956
+#define regVCN_UE_ERR_STATUS_LO_JPEG1S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG1S                                                                  0x0957
+#define regVCN_UE_ERR_STATUS_HI_JPEG1S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG1D                                                                  0x0958
+#define regVCN_UE_ERR_STATUS_LO_JPEG1D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG1D                                                                  0x0959
+#define regVCN_UE_ERR_STATUS_HI_JPEG1D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG2S                                                                  0x095a
+#define regVCN_UE_ERR_STATUS_LO_JPEG2S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG2S                                                                  0x095b
+#define regVCN_UE_ERR_STATUS_HI_JPEG2S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG2D                                                                  0x095c
+#define regVCN_UE_ERR_STATUS_LO_JPEG2D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG2D                                                                  0x095d
+#define regVCN_UE_ERR_STATUS_HI_JPEG2D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG3S                                                                  0x095e
+#define regVCN_UE_ERR_STATUS_LO_JPEG3S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG3S                                                                  0x095f
+#define regVCN_UE_ERR_STATUS_HI_JPEG3S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG3D                                                                  0x0960
+#define regVCN_UE_ERR_STATUS_LO_JPEG3D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG3D                                                                  0x0961
+#define regVCN_UE_ERR_STATUS_HI_JPEG3D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG4S                                                                  0x0962
+#define regVCN_UE_ERR_STATUS_LO_JPEG4S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG4S                                                                  0x0963
+#define regVCN_UE_ERR_STATUS_HI_JPEG4S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG4D                                                                  0x0964
+#define regVCN_UE_ERR_STATUS_LO_JPEG4D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG4D                                                                  0x0965
+#define regVCN_UE_ERR_STATUS_HI_JPEG4D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG5S                                                                  0x0966
+#define regVCN_UE_ERR_STATUS_LO_JPEG5S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG5S                                                                  0x0967
+#define regVCN_UE_ERR_STATUS_HI_JPEG5S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG5D                                                                  0x0968
+#define regVCN_UE_ERR_STATUS_LO_JPEG5D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG5D                                                                  0x0969
+#define regVCN_UE_ERR_STATUS_HI_JPEG5D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG6S                                                                  0x096a
+#define regVCN_UE_ERR_STATUS_LO_JPEG6S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG6S                                                                  0x096b
+#define regVCN_UE_ERR_STATUS_HI_JPEG6S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG6D                                                                  0x096c
+#define regVCN_UE_ERR_STATUS_LO_JPEG6D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG6D                                                                  0x096d
+#define regVCN_UE_ERR_STATUS_HI_JPEG6D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG7S                                                                  0x096e
+#define regVCN_UE_ERR_STATUS_LO_JPEG7S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG7S                                                                  0x096f
+#define regVCN_UE_ERR_STATUS_HI_JPEG7S_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_LO_JPEG7D                                                                  0x0970
+#define regVCN_UE_ERR_STATUS_LO_JPEG7D_BASE_IDX                                                         1
+#define regVCN_UE_ERR_STATUS_HI_JPEG7D                                                                  0x0971
+#define regVCN_UE_ERR_STATUS_HI_JPEG7D_BASE_IDX                                                         1
 
 // addressBlock: aid_uvd0_uvd_jrbc1_uvd_jrbc_dec
 // base address: 0x1e000
diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h
index be643ea0e5698..5bd8111bf04ad 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h
@@ -5312,6 +5312,501 @@
 #define UVD_JPEG_DEC_SOFT_RST__SOFT_RESET_MASK                                                                0x00000001L
 #define UVD_JPEG_DEC_SOFT_RST__RESET_STATUS_MASK                                                              0x00010000L
 
+// addressBlock: aid_uvd0_vcn_edcc_dec
+//VCN_UE_ERR_STATUS_LO_VIDD
+#define VCN_UE_ERR_STATUS_LO_VIDD__Err_Status_Valid_Flag__SHIFT                                               0x0
+#define VCN_UE_ERR_STATUS_LO_VIDD__Address_Valid_Flag__SHIFT                                                  0x1
+#define VCN_UE_ERR_STATUS_LO_VIDD__Address__SHIFT                                                             0x2
+#define VCN_UE_ERR_STATUS_LO_VIDD__Memory_id__SHIFT                                                           0x18
+#define VCN_UE_ERR_STATUS_LO_VIDD__Err_Status_Valid_Flag_MASK                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_LO_VIDD__Address_Valid_Flag_MASK                                                    0x00000002L
+#define VCN_UE_ERR_STATUS_LO_VIDD__Address_MASK                                                               0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_VIDD__Memory_id_MASK                                                             0xFF000000L
+//VCN_UE_ERR_STATUS_HI_VIDD
+#define VCN_UE_ERR_STATUS_HI_VIDD__ECC__SHIFT                                                                 0x0
+#define VCN_UE_ERR_STATUS_HI_VIDD__Parity__SHIFT                                                              0x1
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_Valid_Flag__SHIFT                                                 0x2
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info__SHIFT                                                            0x3
+#define VCN_UE_ERR_STATUS_HI_VIDD__UE_Cnt__SHIFT                                                              0x17
+#define VCN_UE_ERR_STATUS_HI_VIDD__FED_Cnt__SHIFT                                                             0x1a
+#define VCN_UE_ERR_STATUS_HI_VIDD__RESERVED__SHIFT                                                            0x1d
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_clr__SHIFT                                                             0x1f
+#define VCN_UE_ERR_STATUS_HI_VIDD__ECC_MASK                                                                   0x00000001L
+#define VCN_UE_ERR_STATUS_HI_VIDD__Parity_MASK                                                                0x00000002L
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_Valid_Flag_MASK                                                   0x00000004L
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_MASK                                                              0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_VIDD__UE_Cnt_MASK                                                                0x03800000L
+#define VCN_UE_ERR_STATUS_HI_VIDD__FED_Cnt_MASK                                                               0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_VIDD__RESERVED_MASK                                                              0x60000000L
+#define VCN_UE_ERR_STATUS_HI_VIDD__Err_clr_MASK                                                               0x80000000L
+//VCN_UE_ERR_STATUS_LO_VIDV
+#define VCN_UE_ERR_STATUS_LO_VIDV__Err_Status_Valid_Flag__SHIFT                                               0x0
+#define VCN_UE_ERR_STATUS_LO_VIDV__Address_Valid_Flag__SHIFT                                                  0x1
+#define VCN_UE_ERR_STATUS_LO_VIDV__Address__SHIFT                                                             0x2
+#define VCN_UE_ERR_STATUS_LO_VIDV__Memory_id__SHIFT                                                           0x18
+#define VCN_UE_ERR_STATUS_LO_VIDV__Err_Status_Valid_Flag_MASK                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_LO_VIDV__Address_Valid_Flag_MASK                                                    0x00000002L
+#define VCN_UE_ERR_STATUS_LO_VIDV__Address_MASK                                                               0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_VIDV__Memory_id_MASK                                                             0xFF000000L
+//VCN_UE_ERR_STATUS_HI_VIDV
+#define VCN_UE_ERR_STATUS_HI_VIDV__ECC__SHIFT                                                                 0x0
+#define VCN_UE_ERR_STATUS_HI_VIDV__Parity__SHIFT                                                              0x1
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_Valid_Flag__SHIFT                                                 0x2
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info__SHIFT                                                            0x3
+#define VCN_UE_ERR_STATUS_HI_VIDV__UE_Cnt__SHIFT                                                              0x17
+#define VCN_UE_ERR_STATUS_HI_VIDV__FED_Cnt__SHIFT                                                             0x1a
+#define VCN_UE_ERR_STATUS_HI_VIDV__RESERVED__SHIFT                                                            0x1d
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_clr__SHIFT                                                             0x1f
+#define VCN_UE_ERR_STATUS_HI_VIDV__ECC_MASK                                                                   0x00000001L
+#define VCN_UE_ERR_STATUS_HI_VIDV__Parity_MASK                                                                0x00000002L
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_Valid_Flag_MASK                                                   0x00000004L
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_MASK                                                              0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_VIDV__UE_Cnt_MASK                                                                0x03800000L
+#define VCN_UE_ERR_STATUS_HI_VIDV__FED_Cnt_MASK                                                               0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_VIDV__RESERVED_MASK                                                              0x60000000L
+#define VCN_UE_ERR_STATUS_HI_VIDV__Err_clr_MASK                                                               0x80000000L
+//VCN_CE_ERR_STATUS_LO_MMSCHD
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address__SHIFT                                                           0x2
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Memory_id__SHIFT                                                         0x18
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_MASK                                                             0x00FFFFFCL
+#define VCN_CE_ERR_STATUS_LO_MMSCHD__Memory_id_MASK                                                           0xFF000000L
+//VCN_CE_ERR_STATUS_HI_MMSCHD
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__ECC__SHIFT                                                               0x0
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Other__SHIFT                                                             0x1
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info__SHIFT                                                          0x3
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__CE_Cnt__SHIFT                                                            0x17
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Poison__SHIFT                                                            0x1c
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__RESERVED__SHIFT                                                          0x1d
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_clr__SHIFT                                                           0x1f
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__ECC_MASK                                                                 0x00000001L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Other_MASK                                                               0x00000002L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__CE_Cnt_MASK                                                              0x03800000L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Poison_MASK                                                              0x10000000L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__RESERVED_MASK                                                            0x60000000L
+#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG0S
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG0S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG0S
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG0D
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG0D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG0D
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG1S
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG1S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG1S
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG1D
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG1D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG1D
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG2S
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG2S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG2S
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG2D
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG2D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG2D
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG3S
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG3S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG3S
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG3D
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG3D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG3D
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG4S
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG4S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG4S
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG4D
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG4D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG4D
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG5S
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG5S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG5S
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG5D
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG5D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG5D
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG6S
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG6S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG6S
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG6D
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG6D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG6D
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG7S
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG7S__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG7S
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_clr_MASK                                                             0x80000000L
+//VCN_UE_ERR_STATUS_LO_JPEG7D
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Err_Status_Valid_Flag__SHIFT                                             0x0
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_Valid_Flag__SHIFT                                                0x1
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address__SHIFT                                                           0x2
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Memory_id__SHIFT                                                         0x18
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Err_Status_Valid_Flag_MASK                                               0x00000001L
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_Valid_Flag_MASK                                                  0x00000002L
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_MASK                                                             0x00FFFFFCL
+#define VCN_UE_ERR_STATUS_LO_JPEG7D__Memory_id_MASK                                                           0xFF000000L
+//VCN_UE_ERR_STATUS_HI_JPEG7D
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__ECC__SHIFT                                                               0x0
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Parity__SHIFT                                                            0x1
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_Valid_Flag__SHIFT                                               0x2
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info__SHIFT                                                          0x3
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__UE_Cnt__SHIFT                                                            0x17
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__FED_Cnt__SHIFT                                                           0x1a
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__RESERVED__SHIFT                                                          0x1d
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_clr__SHIFT                                                           0x1f
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__ECC_MASK                                                                 0x00000001L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Parity_MASK                                                              0x00000002L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_Valid_Flag_MASK                                                 0x00000004L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_MASK                                                            0x007FFFF8L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__UE_Cnt_MASK                                                              0x03800000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__FED_Cnt_MASK                                                             0x1C000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__RESERVED_MASK                                                            0x60000000L
+#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_clr_MASK                                                             0x80000000L
 
 // addressBlock: aid_uvd0_uvd_jrbc0_uvd_jrbc_dec
 //UVD_JRBC0_UVD_JRBC_RB_WPTR
-- 
GitLab


From 5e1e227fb7cbea2624b4b9375a9b888d02fed4f5 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Wed, 1 Mar 2023 20:37:56 +0800
Subject: [PATCH 1281/1791] drm/amdgpu: Add query_ras_error_count for vcn
 v4_0_3

Add query_ras_error_count callback for vcn v4_0_3.
It will be used to query and log vcn error count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 36 +++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 9d0c3dc765472..e5037d6f884bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -1430,3 +1430,39 @@ const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block = {
 	.rev = 3,
 	.funcs = &vcn_v4_0_3_ip_funcs,
 };
+
+static const struct amdgpu_ras_err_status_reg_entry vcn_v4_0_3_ue_reg_list[] = {
+	{AMDGPU_RAS_REG_ENTRY(VCN, 0, regVCN_UE_ERR_STATUS_LO_VIDD, regVCN_UE_ERR_STATUS_HI_VIDD),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "VIDD"},
+	{AMDGPU_RAS_REG_ENTRY(VCN, 0, regVCN_UE_ERR_STATUS_LO_VIDV, regVCN_UE_ERR_STATUS_HI_VIDV),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "VIDV"},
+};
+
+static void vcn_v4_0_3_inst_query_ras_error_count(struct amdgpu_device *adev,
+						  uint32_t vcn_inst,
+						  void *ras_err_status)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status;
+
+	/* vcn v4_0_3 only support query uncorrectable errors */
+	amdgpu_ras_inst_query_ras_error_count(adev,
+			vcn_v4_0_3_ue_reg_list,
+			ARRAY_SIZE(vcn_v4_0_3_ue_reg_list),
+			NULL, 0, GET_INST(VCN, vcn_inst),
+			AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+			&err_data->ue_count);
+}
+
+static void vcn_v4_0_3_query_ras_error_count(struct amdgpu_device *adev,
+					     void *ras_err_status)
+{
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) {
+		dev_warn(adev->dev, "VCN RAS is not supported\n");
+		return;
+	}
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		vcn_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status);
+}
-- 
GitLab


From 6d39fa3fc802e511241898a6890a9b2ec7f958e3 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Mar 2023 14:23:47 +0800
Subject: [PATCH 1282/1791] drm/amdgpu: Add reset_ras_error_count for vcn
 v4_0_3

Add reset_ras_error_count callback for vcn v4_0_3.
It will be used to reset vcn ras error count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index e5037d6f884bd..e8933039bcd68 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -1466,3 +1466,25 @@ static void vcn_v4_0_3_query_ras_error_count(struct amdgpu_device *adev,
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
 		vcn_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status);
 }
+
+static void vcn_v4_0_3_inst_reset_ras_error_count(struct amdgpu_device *adev,
+						  uint32_t vcn_inst)
+{
+	amdgpu_ras_inst_reset_ras_error_count(adev,
+					vcn_v4_0_3_ue_reg_list,
+					ARRAY_SIZE(vcn_v4_0_3_ue_reg_list),
+					GET_INST(VCN, vcn_inst));
+}
+
+static void vcn_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) {
+		dev_warn(adev->dev, "VCN RAS is not supported\n");
+		return;
+	}
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		vcn_v4_0_3_inst_reset_ras_error_count(adev, i);
+}
-- 
GitLab


From c3f05ab8c40f8a5a8576e3d936cff450a39b0360 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 6 Mar 2023 11:00:11 +0800
Subject: [PATCH 1283/1791] drm/amdgpu: Initialize vcn v4_0_3 ras function

Initialize vcn v4_0_3 ras function

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 26 +++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index e8933039bcd68..7558095ecf6a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -51,6 +51,7 @@ static int vcn_v4_0_3_set_powergating_state(void *handle,
 static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev,
 		int inst_idx, struct dpg_pause_state *new_state);
 static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring);
+static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev);
 
 /**
  * vcn_v4_0_3_early_init - set function pointers
@@ -68,6 +69,7 @@ static int vcn_v4_0_3_early_init(void *handle)
 
 	vcn_v4_0_3_set_unified_ring_funcs(adev);
 	vcn_v4_0_3_set_irq_funcs(adev);
+	vcn_v4_0_3_set_ras_funcs(adev);
 
 	return amdgpu_vcn_early_init(adev);
 }
@@ -130,6 +132,14 @@ static int vcn_v4_0_3_sw_init(void *handle)
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)
 		adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode;
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) {
+		r = amdgpu_vcn_ras_sw_init(adev);
+		if (r) {
+			dev_err(adev->dev, "Failed to initialize vcn ras block!\n");
+			return r;
+		}
+	}
+
 	return 0;
 }
 
@@ -1488,3 +1498,19 @@ static void vcn_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev)
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
 		vcn_v4_0_3_inst_reset_ras_error_count(adev, i);
 }
+
+static const struct amdgpu_ras_block_hw_ops vcn_v4_0_3_ras_hw_ops = {
+	.query_ras_error_count = vcn_v4_0_3_query_ras_error_count,
+	.reset_ras_error_count = vcn_v4_0_3_reset_ras_error_count,
+};
+
+static struct amdgpu_vcn_ras vcn_v4_0_3_ras = {
+	.ras_block = {
+		.hw_ops = &vcn_v4_0_3_ras_hw_ops,
+	},
+};
+
+static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev)
+{
+	adev->vcn.ras = &vcn_v4_0_3_ras;
+}
-- 
GitLab


From 85f23b0a8ce31f1e22d7bfb4666b4a7830563347 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Mar 2023 16:38:38 +0800
Subject: [PATCH 1284/1791] drm/amdgpu: Re-enable VCN RAS if DPG is enabled

VCN RAS enablement sequence needs to be added in
DPG HW init sequence.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 27 ++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 7558095ecf6a1..c77ceaf53dcd9 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -52,7 +52,8 @@ static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev,
 		int inst_idx, struct dpg_pause_state *new_state);
 static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring);
 static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev);
-
+static void vcn_v4_0_3_enable_ras(struct amdgpu_device *adev,
+				  int inst_idx, bool indirect);
 /**
  * vcn_v4_0_3_early_init - set function pointers
  *
@@ -769,6 +770,8 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_LMI_CTRL2), tmp, 0, indirect);
 
+	vcn_v4_0_3_enable_ras(adev, inst_idx, indirect);
+
 	/* enable master interrupt */
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, 0, regUVD_MASTINT_EN),
@@ -1514,3 +1517,25 @@ static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev)
 {
 	adev->vcn.ras = &vcn_v4_0_3_ras;
 }
+
+static void vcn_v4_0_3_enable_ras(struct amdgpu_device *adev,
+				  int inst_idx, bool indirect)
+{
+	uint32_t tmp;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN))
+		return;
+
+	tmp = VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK |
+	      VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK;
+	WREG32_SOC15_DPG_MODE(inst_idx,
+			      SOC15_DPG_MODE_OFFSET(VCN, 0, regVCN_RAS_CNTL),
+			      tmp, 0, indirect);
+
+	tmp = UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK;
+	WREG32_SOC15_DPG_MODE(inst_idx,
+			      SOC15_DPG_MODE_OFFSET(VCN, 0, regUVD_SYS_INT_EN),
+			      tmp, 0, indirect);
+}
-- 
GitLab


From 41e491d8b606ea55b7234967f802cec8e6d77952 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Mar 2023 17:56:59 +0800
Subject: [PATCH 1285/1791] drm/amdgpu: Add query_ras_error_count for jpeg
 v4_0_3

Add query_ras_error_count callback for jpeg v4_0_3.
It will be used to query and log jpeg error count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 64 ++++++++++++++++++++++++
 1 file changed, 64 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index ea9cb098a144a..5dedba91fa32c 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -960,3 +960,67 @@ const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block = {
 	.rev = 3,
 	.funcs = &jpeg_v4_0_3_ip_funcs,
 };
+
+static const struct amdgpu_ras_err_status_reg_entry jpeg_v4_0_3_ue_reg_list[] = {
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG0S, regVCN_UE_ERR_STATUS_HI_JPEG0S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG0S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG0D, regVCN_UE_ERR_STATUS_HI_JPEG0D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG0D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG1S, regVCN_UE_ERR_STATUS_HI_JPEG1S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG1S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG1D, regVCN_UE_ERR_STATUS_HI_JPEG1D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG1D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG2S, regVCN_UE_ERR_STATUS_HI_JPEG2S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG2S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG2D, regVCN_UE_ERR_STATUS_HI_JPEG2D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG2D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG3S, regVCN_UE_ERR_STATUS_HI_JPEG3S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG3S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG3D, regVCN_UE_ERR_STATUS_HI_JPEG3D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG3D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG4S, regVCN_UE_ERR_STATUS_HI_JPEG4S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG4S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG4D, regVCN_UE_ERR_STATUS_HI_JPEG4D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG4D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG5S, regVCN_UE_ERR_STATUS_HI_JPEG5S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG5S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG5D, regVCN_UE_ERR_STATUS_HI_JPEG5D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG5D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG6S, regVCN_UE_ERR_STATUS_HI_JPEG6S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG6S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG6D, regVCN_UE_ERR_STATUS_HI_JPEG6D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG6D"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG7S, regVCN_UE_ERR_STATUS_HI_JPEG7S),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG7S"},
+	{AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG7D, regVCN_UE_ERR_STATUS_HI_JPEG7D),
+	1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG7D"},
+};
+
+static void jpeg_v4_0_3_inst_query_ras_error_count(struct amdgpu_device *adev,
+						   uint32_t jpeg_inst,
+						   void *ras_err_status)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status;
+
+	/* jpeg v4_0_3 only support uncorrectable errors */
+	amdgpu_ras_inst_query_ras_error_count(adev,
+			jpeg_v4_0_3_ue_reg_list,
+			ARRAY_SIZE(jpeg_v4_0_3_ue_reg_list),
+			NULL, 0, GET_INST(VCN, jpeg_inst),
+			AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+			&err_data->ue_count);
+}
+
+static void jpeg_v4_0_3_query_ras_error_count(struct amdgpu_device *adev,
+					      void *ras_err_status)
+{
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) {
+		dev_warn(adev->dev, "JPEG RAS is not supported\n");
+		return;
+	}
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++)
+		jpeg_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status);
+}
-- 
GitLab


From 570df4bca6187f493a1315a7373d7eb1285b3e86 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 2 Mar 2023 18:04:24 +0800
Subject: [PATCH 1286/1791] drm/amdgpu: Add reset_ras_error_count for jpeg
 v4_0_3

Add reset_ras_error_count callback for jpeg v4_0_3.
It will be used to reset jpeg ras error count.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 5dedba91fa32c..21226d6d26f83 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -1024,3 +1024,25 @@ static void jpeg_v4_0_3_query_ras_error_count(struct amdgpu_device *adev,
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++)
 		jpeg_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status);
 }
+
+static void jpeg_v4_0_3_inst_reset_ras_error_count(struct amdgpu_device *adev,
+						   uint32_t jpeg_inst)
+{
+	amdgpu_ras_inst_reset_ras_error_count(adev,
+			jpeg_v4_0_3_ue_reg_list,
+			ARRAY_SIZE(jpeg_v4_0_3_ue_reg_list),
+			GET_INST(VCN, jpeg_inst));
+}
+
+static void jpeg_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	uint32_t i;
+
+	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) {
+		dev_warn(adev->dev, "JPEG RAS is not supported\n");
+		return;
+	}
+
+	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++)
+		jpeg_v4_0_3_inst_reset_ras_error_count(adev, i);
+}
-- 
GitLab


From 35d54e21e002198c13647b6cd8c77586f683cf39 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 6 Mar 2023 11:03:27 +0800
Subject: [PATCH 1287/1791] drm/amdgpu: Initialize jpeg v4_0_3 ras function

Initialize jpeg v4_0_3 ras function.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 26 ++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 21226d6d26f83..ede15a3a4701d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -40,6 +40,7 @@ static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev);
 static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev);
 static int jpeg_v4_0_3_set_powergating_state(void *handle,
 				enum amd_powergating_state state);
+static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev);
 
 static int amdgpu_ih_srcid_jpeg[] = {
 	VCN_4_0__SRCID__JPEG_DECODE,
@@ -67,6 +68,7 @@ static int jpeg_v4_0_3_early_init(void *handle)
 
 	jpeg_v4_0_3_set_dec_ring_funcs(adev);
 	jpeg_v4_0_3_set_irq_funcs(adev);
+	jpeg_v4_0_3_set_ras_funcs(adev);
 
 	return 0;
 }
@@ -126,6 +128,14 @@ static int jpeg_v4_0_3_sw_init(void *handle)
 		}
 	}
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) {
+		r = amdgpu_jpeg_ras_sw_init(adev);
+		if (r) {
+			dev_err(adev->dev, "Failed to initialize jpeg ras block!\n");
+			return r;
+		}
+	}
+
 	return 0;
 }
 
@@ -1046,3 +1056,19 @@ static void jpeg_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev)
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++)
 		jpeg_v4_0_3_inst_reset_ras_error_count(adev, i);
 }
+
+static const struct amdgpu_ras_block_hw_ops jpeg_v4_0_3_ras_hw_ops = {
+	.query_ras_error_count = jpeg_v4_0_3_query_ras_error_count,
+	.reset_ras_error_count = jpeg_v4_0_3_reset_ras_error_count,
+};
+
+static struct amdgpu_jpeg_ras jpeg_v4_0_3_ras = {
+	.ras_block = {
+		.hw_ops = &jpeg_v4_0_3_ras_hw_ops,
+	},
+};
+
+static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev)
+{
+	adev->jpeg.ras = &jpeg_v4_0_3_ras;
+}
-- 
GitLab


From 9b337b7d628a5e97b4dd72bb1d75f1716567b416 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 20 Mar 2023 17:51:30 +0800
Subject: [PATCH 1288/1791] drm/amdgpu: Adjust the sequence to query ras error
 info

It turns out STATUS_VALID_FLAG needs to be checked
ahead of any other fields. ADDRESS_VALID_FLAG and
ERR_INFO_VALID_FLAG only manages ADDRESS and ERR_INFO
field respectively. driver should continue poll
ERR CNT field even ERR_INFO_VALD_FLAG is not set.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 5ae89602a1161..64f80e8cbd638 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3164,7 +3164,8 @@ bool amdgpu_ras_inst_get_err_cnt_field(struct amdgpu_device *adev,
 
 	if ((reg_entry->flags & AMDGPU_RAS_ERR_INFO_VALID) &&
 	    !REG_GET_FIELD(err_status_hi_data, ERR_STATUS_HI, ERR_INFO_VALID_FLAG))
-		return false;
+		/* keep the check here in case we need to refer to the result later */
+		dev_dbg(adev->dev, "Invalid err_info field\n");
 
 	/* read err count */
 	*err_cnt = REG_GET_FIELD(err_status_hi_data, ERR_STATUS, ERR_CNT);
@@ -3187,17 +3188,17 @@ void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev,
 	uint32_t i, j;
 
 	for (i = 0; i < reg_list_size; i++) {
+		/* query memory_id from err_status_lo */
+		if (!amdgpu_ras_inst_get_memory_id_field(adev, &reg_list[i],
+							 instance, &memory_id))
+			continue;
+
 		/* query err_cnt from err_status_hi */
 		if (!amdgpu_ras_inst_get_err_cnt_field(adev, &reg_list[i],
 						       instance, &err_cnt) ||
 		    !err_cnt)
 			continue;
 
-		/* query memory_id from err_status_lo */
-		if (!amdgpu_ras_inst_get_memory_id_field(adev, &reg_list[i],
-							 instance, &memory_id))
-			continue;
-
 		*err_count += err_cnt;
 
 		/* log the errors */
-- 
GitLab


From 3697b9bd7c69910cb6543d8441211ecfb2f013ca Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 15 Mar 2023 14:04:33 -0400
Subject: [PATCH 1289/1791] drm/amdkfd: Increase queue number per process to
 255 on GFX9.4.3

Increase the maximum number of queues that can be created per process
to 255 on GFX 9.4.3. There is no HWS limitation restricting the number
queues that can be created.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index a3c23d07c7df5..b100933340d27 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -242,6 +242,13 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 	enum kfd_queue_type type = properties->type;
 	unsigned int max_queues = 127; /* HWS limit */
 
+	/*
+	 * On GFX 9.4.3, increase the number of queues that
+	 * can be created to 255. No HWS limit on GFX 9.4.3.
+	 */
+	if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3))
+		max_queues = 255;
+
 	q = NULL;
 	kq = NULL;
 
-- 
GitLab


From 1e91a5f79110b96baf7ad21d3c7b5c3e18cdf2a5 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 16 Mar 2023 14:59:27 +0530
Subject: [PATCH 1290/1791] drm/amdgpu: Fix register accesses in GFX v9.4.3

Access registers with the right xcc id. Also, remove the unused logic as
PG is not used in GFX v9.4.3

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index d0ddcd751432f..7ef2c9b515ef8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1004,16 +1004,6 @@ static void gfx_v9_4_3_xcc_init_pg(struct amdgpu_device *adev, int xcc_id)
 	 */
 	if (adev->gfx.rlc.is_rlc_v2_1)
 		gfx_v9_4_3_xcc_enable_save_restore_machine(adev, xcc_id);
-
-	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
-			      AMD_PG_SUPPORT_GFX_SMG |
-			      AMD_PG_SUPPORT_GFX_DMG |
-			      AMD_PG_SUPPORT_CP |
-			      AMD_PG_SUPPORT_GDS |
-			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
-		WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_JUMP_TABLE_RESTORE,
-		       adev->gfx.rlc.cp_table_gpu_addr >> 8);
-	}
 }
 
 static void gfx_v9_4_3_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id)
@@ -1071,7 +1061,7 @@ static void gfx_v9_4_3_xcc_set_safe_mode(struct amdgpu_device *adev, int xcc_id)
 
 	/* wait for RLC_SAFE_MODE */
 	for (i = 0; i < adev->usec_timeout; i++) {
-		if (!REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
+		if (!REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
 			break;
 		udelay(1);
 	}
@@ -1107,7 +1097,7 @@ static void gfx_v9_4_3_xcc_wait_for_rlc_serdes(struct amdgpu_device *adev,
 			gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff,
 						    xcc_id);
 			for (k = 0; k < adev->usec_timeout; k++) {
-				if (RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SERDES_CU_MASTER_BUSY) == 0)
+				if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SERDES_CU_MASTER_BUSY) == 0)
 					break;
 				udelay(1);
 			}
@@ -1131,7 +1121,7 @@ static void gfx_v9_4_3_xcc_wait_for_rlc_serdes(struct amdgpu_device *adev,
 		RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY_MASK |
 		RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY_MASK;
 	for (k = 0; k < adev->usec_timeout; k++) {
-		if ((RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0)
+		if ((RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0)
 			break;
 		udelay(1);
 	}
-- 
GitLab


From 00e1ab02c2ba31b2bd446979949193eb3ca2561c Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 24 Mar 2023 15:21:30 +0530
Subject: [PATCH 1291/1791] drm/amdgpu: Skip halting RLC on GFX v9.4.3

RLC-PMFW handshake happens periodically when GFXCLK DPM is enabled and
halting RLC may cause unexpected results. Avoid halting RLC from driver
side.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 23 +++++++----------------
 1 file changed, 7 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 7ef2c9b515ef8..6cde05421a100 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1256,21 +1256,20 @@ static int gfx_v9_4_3_xcc_rlc_resume(struct amdgpu_device *adev, int xcc_id)
 {
 	int r;
 
-	gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id);
-
-	/* disable CG */
-	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, 0);
-
-	gfx_v9_4_3_xcc_init_pg(adev, xcc_id);
-
 	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+		gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id);
 		/* legacy rlc firmware loading */
 		r = gfx_v9_4_3_xcc_rlc_load_microcode(adev, xcc_id);
 		if (r)
 			return r;
+		gfx_v9_4_3_xcc_rlc_start(adev, xcc_id);
 	}
 
-	gfx_v9_4_3_xcc_rlc_start(adev, xcc_id);
+	amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id);
+	/* disable CG */
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, 0);
+	gfx_v9_4_3_xcc_init_pg(adev, xcc_id);
+	amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id);
 
 	return 0;
 }
@@ -1967,14 +1966,6 @@ static void gfx_v9_4_3_xcc_fini(struct amdgpu_device *adev, int xcc_id)
 
 	gfx_v9_4_3_xcc_kcq_fini_register(adev, xcc_id);
 	gfx_v9_4_3_xcc_cp_enable(adev, false, xcc_id);
-
-	/* Skip suspend with A+A reset */
-	if (adev->gmc.xgmi.connected_to_cpu && amdgpu_in_reset(adev)) {
-		dev_dbg(adev->dev, "Device in reset. Skipping RLC halt\n");
-		return;
-	}
-
-	gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id);
 }
 
 static int gfx_v9_4_3_hw_init(void *handle)
-- 
GitLab


From 610dab118ff5013d46069c828b58d576e0907b66 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Fri, 31 Mar 2023 11:13:40 -0400
Subject: [PATCH 1292/1791] drm/amdkfd: Move pgmap to amdgpu_kfd_dev structure

VRAM pgmap resource is allocated every time when switching compute
partitions because kfd_dev is re-initialized by post_partition_switch,
As a result, it causes memory region resource leaking and system
memory usage accounting unbalanced.

pgmap resource should be allocated and registered only once when loading
driver and freed when unloading driver, move it from kfd_dev to
amdgpu_kfd_dev.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 4 ++++
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c   | 8 ++++----
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h      | 3 ---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c       | 6 +++---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h       | 4 ++--
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c  | 2 +-
 6 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index d1d643a050a1d..e4e1dbba060aa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -30,6 +30,7 @@
 #include <linux/kthread.h>
 #include <linux/workqueue.h>
 #include <linux/mmu_notifier.h>
+#include <linux/memremap.h>
 #include <kgd_kfd_interface.h>
 #include <drm/ttm/ttm_execbuf_util.h>
 #include "amdgpu_sync.h"
@@ -101,6 +102,9 @@ struct amdgpu_kfd_dev {
 	uint64_t vram_used_aligned;
 	bool init_complete;
 	struct work_struct reset_work;
+
+	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
+	struct dev_pagemap pgmap;
 };
 
 enum kgd_engine_type {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 42e599912e52b..199d32c7c289f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -206,7 +206,7 @@ svm_migrate_copy_done(struct amdgpu_device *adev, struct dma_fence *mfence)
 unsigned long
 svm_migrate_addr_to_pfn(struct amdgpu_device *adev, unsigned long addr)
 {
-	return (addr + adev->kfd.dev->pgmap.range.start) >> PAGE_SHIFT;
+	return (addr + adev->kfd.pgmap.range.start) >> PAGE_SHIFT;
 }
 
 static void
@@ -236,7 +236,7 @@ svm_migrate_addr(struct amdgpu_device *adev, struct page *page)
 	unsigned long addr;
 
 	addr = page_to_pfn(page) << PAGE_SHIFT;
-	return (addr - adev->kfd.dev->pgmap.range.start);
+	return (addr - adev->kfd.pgmap.range.start);
 }
 
 static struct page *
@@ -990,14 +990,14 @@ static const struct dev_pagemap_ops svm_migrate_pgmap_ops = {
 
 int svm_migrate_init(struct amdgpu_device *adev)
 {
-	struct kfd_dev *kfddev = adev->kfd.dev;
+	struct amdgpu_kfd_dev *kfddev = &adev->kfd;
 	struct dev_pagemap *pgmap;
 	struct resource *res = NULL;
 	unsigned long size;
 	void *r;
 
 	/* Page migration works on Vega10 or newer */
-	if (!KFD_IS_SOC15(kfddev))
+	if (!KFD_IS_SOC15(kfddev->dev))
 		return -EINVAL;
 
 	pgmap = &kfddev->pgmap;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 02a90fd7f646b..214d950f948ee 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -378,9 +378,6 @@ struct kfd_dev {
 
 	int noretry;
 
-	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
-	struct dev_pagemap pgmap;
-
 	struct kfd_node *nodes[MAX_KFD_NODES];
 	unsigned int num_nodes;
 };
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index cf354f9e42856..2b2129dd1e4a6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -174,7 +174,7 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
 
 			addr[i] = (hmm_pfns[i] << PAGE_SHIFT) +
 				   bo_adev->vm_manager.vram_base_offset -
-				   bo_adev->kfd.dev->pgmap.range.start;
+				   bo_adev->kfd.pgmap.range.start;
 			addr[i] |= SVM_RANGE_VRAM_DOMAIN;
 			pr_debug_ratelimited("vram address: 0x%llx\n", addr[i]);
 			continue;
@@ -2827,7 +2827,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 	bool migration = false;
 	int r = 0;
 
-	if (!KFD_IS_SVM_API_SUPPORTED(adev->kfd.dev)) {
+	if (!KFD_IS_SVM_API_SUPPORTED(adev)) {
 		pr_debug("device does not support SVM\n");
 		return -EFAULT;
 	}
@@ -3112,7 +3112,7 @@ int svm_range_list_init(struct kfd_process *p)
 	spin_lock_init(&svms->deferred_list_lock);
 
 	for (i = 0; i < p->n_pdds; i++)
-		if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev->kfd))
+		if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev->adev))
 			bitmap_set(svms->bitmap_supported, i, 1);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index 7515ddade3ae9..021def496f5ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -200,8 +200,8 @@ void svm_range_list_lock_and_flush_work(struct svm_range_list *svms, struct mm_s
 /* SVM API and HMM page migration work together, device memory type
  * is initialized to not 0 when page migration register device memory.
  */
-#define KFD_IS_SVM_API_SUPPORTED(dev) ((dev)->pgmap.type != 0 ||\
-				       (dev)->adev->gmc.is_app_apu)
+#define KFD_IS_SVM_API_SUPPORTED(adev) ((adev)->kfd.pgmap.type != 0 ||\
+					(adev)->gmc.is_app_apu)
 
 void svm_range_bo_unref_async(struct svm_range_bo *svm_bo);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index d2a42b6b1fa89..6d6243b978e1c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -2021,7 +2021,7 @@ int kfd_topology_add_device(struct kfd_node *gpu)
 		dev->node_props.capability |= (dev->gpu->adev->ras_enabled != 0) ?
 			HSA_CAP_RASEVENTNOTIFY : 0;
 
-	if (KFD_IS_SVM_API_SUPPORTED(dev->gpu->adev->kfd.dev))
+	if (KFD_IS_SVM_API_SUPPORTED(dev->gpu->adev))
 		dev->node_props.capability |= HSA_CAP_SVMAPI_SUPPORTED;
 
 	kfd_debug_print_topology();
-- 
GitLab


From fc021438d0ab7863dc93f84a557af6dc6255b881 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Wed, 19 Apr 2023 17:43:26 -0400
Subject: [PATCH 1293/1791] drm/amdgpu: Enable NPS4 CPX mode

CPX compute mode is valid mode for NPS4 memory partition mode.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 848049db00ab1..97011e7e031d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -281,9 +281,9 @@ static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 			adev->gmc.num_mem_partitions == 4) &&
 		       (num_xccs_per_xcp >= 2);
 	case AMDGPU_CPX_PARTITION_MODE:
-		return (num_xcc > 1) &&
-		       (adev->gmc.num_mem_partitions == 1 ||
-			adev->gmc.num_mem_partitions == num_xcc);
+		return ((num_xcc > 1) &&
+		       (adev->gmc.num_mem_partitions == 1 || adev->gmc.num_mem_partitions == 4) &&
+		       (num_xcc % adev->gmc.num_mem_partitions) == 0);
 	default:
 		return false;
 	}
-- 
GitLab


From f915f3af9984464c308787102990d85d4e988d2c Mon Sep 17 00:00:00 2001
From: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Date: Fri, 28 Apr 2023 14:20:00 -0400
Subject: [PATCH 1294/1791] drm/amdgpu: For GFX 9.4.3 APU fix vram_usage value

For GFX 9.4.3 APP APU VRAM is allocated in GTT domain. While freeing
memory check for GTT domain instead of VRAM if it is APP APU

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index c4b949d17e14c..fd395cd61b540 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1886,11 +1886,14 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
 	}
 
 	/* Update the size of the BO being freed if it was allocated from
-	 * VRAM and is not imported.
+	 * VRAM and is not imported. For APP APU VRAM allocations are done
+	 * in GTT domain
 	 */
 	if (size) {
-		if ((mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_VRAM) &&
-		    (!is_imported))
+		if (!is_imported &&
+		   (mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_VRAM ||
+		   (adev->gmc.is_app_apu &&
+		    mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_GTT)))
 			*size = bo_size;
 		else
 			*size = 0;
-- 
GitLab


From 2e8cc5d317d12f7fb4f66361a3ce5427f0abe2cd Mon Sep 17 00:00:00 2001
From: Graham Sider <Graham.Sider@amd.com>
Date: Wed, 8 Feb 2023 11:10:57 -0500
Subject: [PATCH 1295/1791] drm/amdgpu: Use legacy TLB flush for gfx943

Invalidate TLBs via a legacy flush request (flush_type=0) prior to the
heavyweight flush requests (flush_type=2) in gmc_v9_0.c. This is
temporarily required to mitigate a bug causing CPC UTCL1 to return stale
translations after invalidation requests in address range mode.

v2: squash in long term fix "drm/amdgpu: disable extra gfx943 legacy flush on rev1+"

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 95c3253e240ac..2eb67b53e497d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -833,6 +833,11 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		 */
 		inv_req = gmc_v9_0_get_invalidate_req(vmid, 2);
 		inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type);
+	} else if (flush_type == 2 &&
+		   adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) &&
+		   adev->rev_id == 0) {
+		inv_req = gmc_v9_0_get_invalidate_req(vmid, 0);
+		inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type);
 	} else {
 		inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type);
 		inv_req2 = 0;
@@ -976,6 +981,13 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 		if (vega20_xgmi_wa)
 			kiq->pmf->kiq_invalidate_tlbs(ring,
 						      pasid, 2, all_hub);
+
+		if (flush_type == 2 &&
+		    adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) &&
+		    adev->rev_id == 0)
+			kiq->pmf->kiq_invalidate_tlbs(ring,
+						pasid, 0, all_hub);
+
 		kiq->pmf->kiq_invalidate_tlbs(ring,
 					pasid, flush_type, all_hub);
 		r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT);
-- 
GitLab


From 895797d9193b38e759bc01268a8e3887e521f682 Mon Sep 17 00:00:00 2001
From: Graham Sider <Graham.Sider@amd.com>
Date: Mon, 6 Feb 2023 14:04:42 -0500
Subject: [PATCH 1296/1791] drm/amdgpu/bu: Add use_mtype_cc_wa module param

By default, set use_mtype_cc_wa to 1 to set PTE coherence flag MTYPE_CC
instead of MTYPE_RW by default. This is required for the time being to
mitigate a bug causing XCCs to hit stale data due to TCC marking fully
dirty lines as exclusive.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Joseph Greathouse <Joseph.Greathouse@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  7 +++++++
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 10 +++++++---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c    |  7 +++++--
 4 files changed, 20 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index cb9373f8c25ae..cd2a29a7e26d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -212,6 +212,7 @@ extern int amdgpu_noretry;
 extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
 extern int amdgpu_use_xgmi_p2p;
+extern bool amdgpu_use_mtype_cc_wa;
 #ifdef CONFIG_HSA_AMD
 extern int sched_policy;
 extern bool debug_evictions;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index da4e50aef95a3..8bc37826a99ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -822,6 +822,13 @@ MODULE_PARM_DESC(no_queue_eviction_on_vm_fault, "No queue eviction on VM fault (
 module_param_named(no_queue_eviction_on_vm_fault, amdgpu_no_queue_eviction_on_vm_fault, int, 0444);
 #endif
 
+/**
+ * DOC: use_mtype_cc_wa (bool)
+ */
+bool amdgpu_use_mtype_cc_wa = true;
+MODULE_PARM_DESC(use_mtype_cc_wa, "Use MTYPE_CC workaround (0 = use MTYPE_RW where applicable, 1 = use MTYPE_CC where applicable (default))");
+module_param_named(use_mtype_cc_wa, amdgpu_use_mtype_cc_wa, bool, 0444);
+
 /**
  * DOC: pcie_p2p (bool)
  * Enable PCIe P2P (requires large-BAR). Default value: true (on)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 2eb67b53e497d..8623b93c05eea 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1187,6 +1187,7 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 	bool coherent = bo->flags & AMDGPU_GEM_CREATE_COHERENT;
 	bool uncached = bo->flags & AMDGPU_GEM_CREATE_UNCACHED;
 	unsigned int mtype;
+	unsigned int mtype_default;
 	bool snoop = false;
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
@@ -1230,7 +1231,10 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		/* FIXME: Needs more work for handling multiple memory
 		 * partitions (> NPS1 mode) e.g. NPS4 for both APU and dGPU
 		 * modes.
+		 * FIXME: Temporarily using MTYPE_CC instead of MTYPE_RW where applicable.
+		 * To force use of MTYPE_RW, set use_mtype_cc_wa=0
 		 */
+		mtype_default = amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
 		snoop = true;
 		if (uncached) {
 			mtype = MTYPE_UC;
@@ -1245,14 +1249,14 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 			 * socket should be treated as remote access so MTYPE_RW
 			 * cannot be used always.
 			 */
-			mtype = MTYPE_RW;
+			mtype = mtype_default;
 		} else if (adev->flags & AMD_IS_APU) {
 			/* APU on carve out mode */
-			mtype = MTYPE_RW;
+			mtype = mtype_default;
 		} else {
 			/* dGPU */
 			if (is_vram && bo_adev == adev)
-				mtype = MTYPE_RW;
+				mtype = mtype_default;
 			else if (is_vram)
 				mtype = MTYPE_NC;
 			else
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 2b2129dd1e4a6..477ef92942037 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1198,9 +1198,12 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		} else if (domain == SVM_RANGE_VRAM_DOMAIN) {
-			/* local HBM region close to partition */
+			/* local HBM region close to partition
+			 * FIXME: Temporarily using MTYPE_CC instead of MTYPE_RW where applicable.
+			 * To force use of MTYPE_RW, set use_mtype_cc_wa=0
+			 */
 			if (bo_node == node)
-				mapping_flags |= AMDGPU_VM_MTYPE_RW;
+				mapping_flags |= amdgpu_use_mtype_cc_wa ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
 			/* local HBM region far from partition or remote XGMI GPU */
 			else if (svm_nodes_in_same_hive(bo_node, node))
 				mapping_flags |= AMDGPU_VM_MTYPE_NC;
-- 
GitLab


From 1e4a00334add40f609162914af7a24bc92951008 Mon Sep 17 00:00:00 2001
From: Felix Kuehling <Felix.Kuehling@amd.com>
Date: Tue, 21 Feb 2023 17:31:32 -0500
Subject: [PATCH 1297/1791] drm/amdgpu: Fix per-BO MTYPE selection for
 GFXv9.4.3

Treat system memory on NUMA systems as remote by default. Overriding with
a more efficient MTYPE per page will be implemented in the next patch.

No need for a special case for APP APUs. System memory is handled the same
for carve-out and native mode. And VRAM doesn't exist in native mode.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 40 +++++++++++----------------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c  | 24 +++++++++-------
 2 files changed, 30 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 8623b93c05eea..cf976b5b7b63e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1186,9 +1186,10 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 	bool is_vram = bo->tbo.resource->mem_type == TTM_PL_VRAM;
 	bool coherent = bo->flags & AMDGPU_GEM_CREATE_COHERENT;
 	bool uncached = bo->flags & AMDGPU_GEM_CREATE_UNCACHED;
-	unsigned int mtype;
-	unsigned int mtype_default;
+	/* TODO: memory partitions struct amdgpu_vm *vm = mapping->bo_va->base.vm;*/
+	unsigned int mtype_local, mtype;
 	bool snoop = false;
+	bool is_local;
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 1):
@@ -1228,35 +1229,26 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		}
 		break;
 	case IP_VERSION(9, 4, 3):
-		/* FIXME: Needs more work for handling multiple memory
-		 * partitions (> NPS1 mode) e.g. NPS4 for both APU and dGPU
-		 * modes.
-		 * FIXME: Temporarily using MTYPE_CC instead of MTYPE_RW where applicable.
-		 * To force use of MTYPE_RW, set use_mtype_cc_wa=0
+		/* Only local VRAM BOs or system memory on non-NUMA APUs
+		 * can be assumed to be local in their entirety. Choose
+		 * MTYPE_NC as safe fallback for all system memory BOs on
+		 * NUMA systems. Their MTYPE can be overridden per-page in
+		 * gmc_v9_0_override_vm_pte_flags.
 		 */
-		mtype_default = amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
+		mtype_local = amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
+		is_local = (!is_vram && (adev->flags & AMD_IS_APU) &&
+			    num_possible_nodes() <= 1) ||
+			   (is_vram && adev == bo_adev /* TODO: memory partitions &&
+			    bo->mem_id == vm->mem_id*/);
 		snoop = true;
 		if (uncached) {
 			mtype = MTYPE_UC;
-		} else if (adev->gmc.is_app_apu) {
-			/* FIXME: APU in native mode, NPS1 single socket only
-			 *
-			 * For suporting NUMA partitioned APU e.g. in NPS4 mode,
-			 * this need to look at the NUMA node on which the
-			 * system memory allocation was done.
-			 *
-			 * Memory access by a different partition within same
-			 * socket should be treated as remote access so MTYPE_RW
-			 * cannot be used always.
-			 */
-			mtype = mtype_default;
 		} else if (adev->flags & AMD_IS_APU) {
-			/* APU on carve out mode */
-			mtype = mtype_default;
+			mtype = is_local ? mtype_local : MTYPE_NC;
 		} else {
 			/* dGPU */
-			if (is_vram && bo_adev == adev)
-				mtype = mtype_default;
+			if (is_local)
+				mtype = mtype_local;
 			else if (is_vram)
 				mtype = MTYPE_NC;
 			else
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 477ef92942037..4eec75b28917a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1151,6 +1151,7 @@ svm_range_get_pte_flags(struct kfd_node *node,
 	bool snoop = (domain != SVM_RANGE_VRAM_DOMAIN);
 	bool coherent = flags & KFD_IOCTL_SVM_FLAG_COHERENT;
 	bool uncached = flags & KFD_IOCTL_SVM_FLAG_UNCACHED;
+	unsigned int mtype_local;
 
 	if (domain == SVM_RANGE_VRAM_DOMAIN)
 		bo_node = prange->svm_bo->node;
@@ -1191,19 +1192,16 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		}
 		break;
 	case IP_VERSION(9, 4, 3):
-		//TODO: Need more work for handling multiple memory partitions
-		//e.g. NPS4. Current approch is only applicable without memory
-		//partitions.
+		mtype_local = amdgpu_use_mtype_cc_wa ? AMDGPU_VM_MTYPE_CC :
+						       AMDGPU_VM_MTYPE_RW;
 		snoop = true;
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		} else if (domain == SVM_RANGE_VRAM_DOMAIN) {
-			/* local HBM region close to partition
-			 * FIXME: Temporarily using MTYPE_CC instead of MTYPE_RW where applicable.
-			 * To force use of MTYPE_RW, set use_mtype_cc_wa=0
-			 */
-			if (bo_node == node)
-				mapping_flags |= amdgpu_use_mtype_cc_wa ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
+			/* local HBM region close to partition */
+			if (bo_node->adev == node->adev /* TODO: memory partitions &&
+			    bo_node->mem_id == node->mem_id*/)
+				mapping_flags |= mtype_local;
 			/* local HBM region far from partition or remote XGMI GPU */
 			else if (svm_nodes_in_same_hive(bo_node, node))
 				mapping_flags |= AMDGPU_VM_MTYPE_NC;
@@ -1212,7 +1210,13 @@ svm_range_get_pte_flags(struct kfd_node *node,
 				mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		/* system memory accessed by the APU */
 		} else if (node->adev->flags & AMD_IS_APU) {
-			mapping_flags |= AMDGPU_VM_MTYPE_NC;
+			/* On NUMA systems, locality is determined per-page
+			 * in amdgpu_gmc_override_vm_pte_flags
+			 */
+			if (num_possible_nodes() <= 1)
+				mapping_flags |= mtype_local;
+			else
+				mapping_flags |= AMDGPU_VM_MTYPE_NC;
 		/* system memory accessed by the dGPU */
 		} else {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
-- 
GitLab


From 352b919c1e5ff50c71d665395b27acbd1bf23a05 Mon Sep 17 00:00:00 2001
From: Felix Kuehling <Felix.Kuehling@amd.com>
Date: Tue, 21 Feb 2023 17:44:18 -0500
Subject: [PATCH 1298/1791] drm/amdgpu: Override MTYPE per page on GFXv9.4.3
 APUs

On GFXv9.4.3 NUMA APUs, system memory locality must be determined per
page to choose the correct MTYPE. This patch adds a GMC callback that
can provide this per-page override and implements it for native mode.

Carve-out mode is not yet supported and will use the safe default
(remote) MTYPE for system memory.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h   |  7 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c | 22 ++++++--
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c     | 64 +++++++++++++++++++++++
 3 files changed, 90 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 43357d699e6ee..6794edd1d2d2a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -148,6 +148,10 @@ struct amdgpu_gmc_funcs {
 	void (*get_vm_pte)(struct amdgpu_device *adev,
 			   struct amdgpu_bo_va_mapping *mapping,
 			   uint64_t *flags);
+	/* override per-page pte flags */
+	void (*override_vm_pte_flags)(struct amdgpu_device *dev,
+				      struct amdgpu_vm *vm,
+				      uint64_t addr, uint64_t *flags);
 	/* get the amount of memory used by the vbios for pre-OS console */
 	unsigned int (*get_vbios_fb_size)(struct amdgpu_device *adev);
 
@@ -336,6 +340,9 @@ struct amdgpu_gmc {
 #define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags))
 #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags))
 #define amdgpu_gmc_get_vm_pte(adev, mapping, flags) (adev)->gmc.gmc_funcs->get_vm_pte((adev), (mapping), (flags))
+#define amdgpu_gmc_override_vm_pte_flags(adev, vm, addr, pte_flags)	\
+	(adev)->gmc.gmc_funcs->override_vm_pte_flags			\
+		((adev), (vm), (addr), (pte_flags))
 #define amdgpu_gmc_get_vbios_fb_size(adev) (adev)->gmc.gmc_funcs->get_vbios_fb_size((adev))
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index bc5d126b600b4..60b1da93b06dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -786,13 +786,14 @@ static void amdgpu_vm_pte_update_flags(struct amdgpu_vm_update_params *params,
 				       uint64_t pe, uint64_t addr,
 				       unsigned int count, uint32_t incr,
 				       uint64_t flags)
-
 {
+	struct amdgpu_device *adev = params->adev;
+
 	if (level != AMDGPU_VM_PTB) {
 		flags |= AMDGPU_PDE_PTE;
-		amdgpu_gmc_get_vm_pde(params->adev, level, &addr, &flags);
+		amdgpu_gmc_get_vm_pde(adev, level, &addr, &flags);
 
-	} else if (params->adev->asic_type >= CHIP_VEGA10 &&
+	} else if (adev->asic_type >= CHIP_VEGA10 &&
 		   !(flags & AMDGPU_PTE_VALID) &&
 		   !(flags & AMDGPU_PTE_PRT)) {
 
@@ -800,6 +801,21 @@ static void amdgpu_vm_pte_update_flags(struct amdgpu_vm_update_params *params,
 		flags |= AMDGPU_PTE_EXECUTABLE;
 	}
 
+	/* APUs mapping system memory may need different MTYPEs on different
+	 * NUMA nodes. Only do this for contiguous ranges that can be assumed
+	 * to be on the same NUMA node.
+	 */
+	if ((flags & AMDGPU_PTE_SYSTEM) && (adev->flags & AMD_IS_APU) &&
+	    adev->gmc.gmc_funcs->override_vm_pte_flags &&
+	    num_possible_nodes() > 1) {
+		if (!params->pages_addr)
+			amdgpu_gmc_override_vm_pte_flags(adev, params->vm,
+							 addr, &flags);
+		else
+			dev_dbg(adev->dev,
+				"override_vm_pte_flags skipped: non-contiguous\n");
+	}
+
 	params->vm->update_funcs->update(params, pt, pe, addr, count, incr,
 					 flags);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index cf976b5b7b63e..c64a69f75da28 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1297,6 +1297,69 @@ static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev,
 					     mapping, flags);
 }
 
+static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
+					   struct amdgpu_vm *vm,
+					   uint64_t addr, uint64_t *flags)
+{
+	int local_node, nid;
+
+	/* Only GFX 9.4.3 APUs associate GPUs with NUMA nodes. Local system
+	 * memory can use more efficient MTYPEs.
+	 */
+	if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3))
+		return;
+
+	/* Only direct-mapped memory allows us to determine the NUMA node from
+	 * the DMA address.
+	 */
+	if (!adev->ram_is_direct_mapped) {
+		dev_dbg(adev->dev, "RAM is not direct mapped\n");
+		return;
+	}
+
+	/* Only override mappings with MTYPE_NC, which is the safe default for
+	 * cacheable memory.
+	 */
+	if ((*flags & AMDGPU_PTE_MTYPE_VG10_MASK) !=
+	    AMDGPU_PTE_MTYPE_VG10(MTYPE_NC)) {
+		dev_dbg(adev->dev, "MTYPE is not NC\n");
+		return;
+	}
+
+	/* TODO: memory partitions. mem_id is hard-coded to 0 for now.
+	 * FIXME: Only supported on native mode for now. For carve-out, the
+	 * NUMA affinity of the GPU/VM needs to come from the PCI info because
+	 * memory partitions are not associated with different NUMA nodes.
+	 */
+	if (adev->gmc.is_app_apu) {
+		local_node = adev->gmc.mem_partitions[/*vm->mem_id*/0].numa.node;
+	} else {
+		dev_dbg(adev->dev, "Only native mode APU is supported.\n");
+		return;
+	}
+
+	/* Only handle real RAM. Mappings of PCIe resources don't have struct
+	 * page or NUMA nodes.
+	 */
+	if (!page_is_ram(addr >> PAGE_SHIFT)) {
+		dev_dbg(adev->dev, "Page is not RAM.\n");
+		return;
+	}
+	nid = pfn_to_nid(addr >> PAGE_SHIFT);
+	dev_dbg(adev->dev, "vm->mem_id=%d, local_node=%d, nid=%d\n",
+		/*vm->mem_id*/0, local_node, nid);
+	if (nid == local_node) {
+		unsigned int mtype_local =
+			amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
+		uint64_t old_flags = *flags;
+
+		*flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
+			 AMDGPU_PTE_MTYPE_VG10(mtype_local);
+		dev_dbg(adev->dev, "flags updated from %llx to %llx\n",
+			old_flags, *flags);
+	}
+}
+
 static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
 {
 	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
@@ -1368,6 +1431,7 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.map_mtype = gmc_v9_0_map_mtype,
 	.get_vm_pde = gmc_v9_0_get_vm_pde,
 	.get_vm_pte = gmc_v9_0_get_vm_pte,
+	.override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags,
 	.get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size,
 	.query_mem_partition_mode = &gmc_v9_0_query_memory_partition,
 };
-- 
GitLab


From 76eb9c95a409ea820b2e7c968c220e7a38f27d76 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Mon, 27 Feb 2023 10:33:11 -0500
Subject: [PATCH 1299/1791] drm/amdgpu/bu: add mtype_local as a module
 parameter

Selects the MTYPE to be used for local memory,
(0 = MTYPE_CC (default), 1 = MTYPE_NC, 2 = MTYPE_RW)

v2: squash in build fix (Alex)

Reviewed-by: Graham Sider <Graham.Sider@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  8 ++++----
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 19 ++++++++++++++++---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c    |  3 +--
 4 files changed, 22 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index cd2a29a7e26d6..c2feaf2fd0702 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -212,7 +212,7 @@ extern int amdgpu_noretry;
 extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
 extern int amdgpu_use_xgmi_p2p;
-extern bool amdgpu_use_mtype_cc_wa;
+extern int amdgpu_mtype_local;
 #ifdef CONFIG_HSA_AMD
 extern int sched_policy;
 extern bool debug_evictions;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 8bc37826a99ff..706ba4af062fe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -823,11 +823,11 @@ module_param_named(no_queue_eviction_on_vm_fault, amdgpu_no_queue_eviction_on_vm
 #endif
 
 /**
- * DOC: use_mtype_cc_wa (bool)
+ * DOC: mtype_local (int)
  */
-bool amdgpu_use_mtype_cc_wa = true;
-MODULE_PARM_DESC(use_mtype_cc_wa, "Use MTYPE_CC workaround (0 = use MTYPE_RW where applicable, 1 = use MTYPE_CC where applicable (default))");
-module_param_named(use_mtype_cc_wa, amdgpu_use_mtype_cc_wa, bool, 0444);
+int amdgpu_mtype_local;
+MODULE_PARM_DESC(mtype_local, "MTYPE for local memory (0 = MTYPE_CC (default), 1 = MTYPE_NC, 2 = MTYPE_RW)");
+module_param_named(mtype_local, amdgpu_mtype_local, int, 0444);
 
 /**
  * DOC: pcie_p2p (bool)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index c64a69f75da28..5a1414300271d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1235,7 +1235,16 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		 * NUMA systems. Their MTYPE can be overridden per-page in
 		 * gmc_v9_0_override_vm_pte_flags.
 		 */
-		mtype_local = amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
+		mtype_local = MTYPE_CC;
+		if (amdgpu_mtype_local == 1) {
+			DRM_INFO_ONCE("Using MTYPE_NC for local memory\n");
+			mtype_local = MTYPE_NC;
+		} else if (amdgpu_mtype_local == 2) {
+			DRM_INFO_ONCE("Using MTYPE_RW for local memory\n");
+			mtype_local = MTYPE_RW;
+		} else {
+			DRM_INFO_ONCE("Using MTYPE_CC for local memory\n");
+		}
 		is_local = (!is_vram && (adev->flags & AMD_IS_APU) &&
 			    num_possible_nodes() <= 1) ||
 			   (is_vram && adev == bo_adev /* TODO: memory partitions &&
@@ -1349,9 +1358,13 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 	dev_dbg(adev->dev, "vm->mem_id=%d, local_node=%d, nid=%d\n",
 		/*vm->mem_id*/0, local_node, nid);
 	if (nid == local_node) {
-		unsigned int mtype_local =
-			amdgpu_use_mtype_cc_wa ? MTYPE_CC : MTYPE_RW;
 		uint64_t old_flags = *flags;
+		unsigned int mtype_local = MTYPE_CC;
+
+		if (amdgpu_mtype_local == 1)
+			mtype_local = MTYPE_NC;
+		else if (amdgpu_mtype_local == 2)
+			mtype_local = MTYPE_RW;
 
 		*flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
 			 AMDGPU_PTE_MTYPE_VG10(mtype_local);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 4eec75b28917a..df0ed56776094 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1192,8 +1192,7 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		}
 		break;
 	case IP_VERSION(9, 4, 3):
-		mtype_local = amdgpu_use_mtype_cc_wa ? AMDGPU_VM_MTYPE_CC :
-						       AMDGPU_VM_MTYPE_RW;
+		mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC : (amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_RW : AMDGPU_VM_MTYPE_CC);
 		snoop = true;
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
-- 
GitLab


From b9cbd51000ad3541351ca832b00600870ac08e5c Mon Sep 17 00:00:00 2001
From: Graham Sider <Graham.Sider@amd.com>
Date: Mon, 6 Mar 2023 17:56:44 -0500
Subject: [PATCH 1300/1791] drm/amdgpu/bu: update mtype_local parameter
 settings

Update mtype_local module parameter to use MTYPE_RW by default.

0: MTYPE_RW (default)
1: MTYPE_NC
2: MTYPE_CC

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 12 ++++++------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c    |  3 ++-
 3 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 706ba4af062fe..aa466a9eb956f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -826,7 +826,7 @@ module_param_named(no_queue_eviction_on_vm_fault, amdgpu_no_queue_eviction_on_vm
  * DOC: mtype_local (int)
  */
 int amdgpu_mtype_local;
-MODULE_PARM_DESC(mtype_local, "MTYPE for local memory (0 = MTYPE_CC (default), 1 = MTYPE_NC, 2 = MTYPE_RW)");
+MODULE_PARM_DESC(mtype_local, "MTYPE for local memory (0 = MTYPE_RW (default), 1 = MTYPE_NC, 2 = MTYPE_CC)");
 module_param_named(mtype_local, amdgpu_mtype_local, int, 0444);
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 5a1414300271d..32eb4f4f54922 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1235,15 +1235,15 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		 * NUMA systems. Their MTYPE can be overridden per-page in
 		 * gmc_v9_0_override_vm_pte_flags.
 		 */
-		mtype_local = MTYPE_CC;
+		mtype_local = MTYPE_RW;
 		if (amdgpu_mtype_local == 1) {
 			DRM_INFO_ONCE("Using MTYPE_NC for local memory\n");
 			mtype_local = MTYPE_NC;
 		} else if (amdgpu_mtype_local == 2) {
-			DRM_INFO_ONCE("Using MTYPE_RW for local memory\n");
-			mtype_local = MTYPE_RW;
-		} else {
 			DRM_INFO_ONCE("Using MTYPE_CC for local memory\n");
+			mtype_local = MTYPE_CC;
+		} else {
+			DRM_INFO_ONCE("Using MTYPE_RW for local memory\n");
 		}
 		is_local = (!is_vram && (adev->flags & AMD_IS_APU) &&
 			    num_possible_nodes() <= 1) ||
@@ -1359,12 +1359,12 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 		/*vm->mem_id*/0, local_node, nid);
 	if (nid == local_node) {
 		uint64_t old_flags = *flags;
-		unsigned int mtype_local = MTYPE_CC;
+		unsigned int mtype_local = MTYPE_RW;
 
 		if (amdgpu_mtype_local == 1)
 			mtype_local = MTYPE_NC;
 		else if (amdgpu_mtype_local == 2)
-			mtype_local = MTYPE_RW;
+			mtype_local = MTYPE_CC;
 
 		*flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
 			 AMDGPU_PTE_MTYPE_VG10(mtype_local);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index df0ed56776094..e6348d4133fd8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1192,7 +1192,8 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		}
 		break;
 	case IP_VERSION(9, 4, 3):
-		mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC : (amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_RW : AMDGPU_VM_MTYPE_CC);
+		mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC :
+			     (amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW);
 		snoop = true;
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
-- 
GitLab


From 2c1c7ba457d4ecf475c0e220ac5359971355c6eb Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 16:55:02 -0400
Subject: [PATCH 1301/1791] drm/amdgpu: support partition drm devices
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Support partition drm devices on GC_HWIP IP_VERSION(9, 4, 3).

This is a temporary solution and will be superceded.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-and-tested-by: Philip Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c    | 32 ++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h    |  2 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c    | 59 +++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h    |  5 ++
 6 files changed, 99 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index c2feaf2fd0702..eb2fb968e3af5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -107,6 +107,7 @@
 #include "amdgpu_fdinfo.h"
 #include "amdgpu_mca.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_xcp.h"
 
 #define MAX_GPU_INSTANCE		64
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 0c6f983fb2adc..4d9c535bcb0cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6065,6 +6065,7 @@ void amdgpu_device_halt(struct amdgpu_device *adev)
 	struct pci_dev *pdev = adev->pdev;
 	struct drm_device *ddev = adev_to_drm(adev);
 
+	amdgpu_xcp_dev_unplug(adev);
 	drm_dev_unplug(ddev);
 
 	amdgpu_irq_disable_all(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index aa466a9eb956f..7300df2a342c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2185,6 +2185,10 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 		goto err_pci;
 	}
 
+	ret = amdgpu_xcp_dev_register(adev, ent);
+	if (ret)
+		goto err_pci;
+
 	/*
 	 * 1. don't init fbdev on hw without DCE
 	 * 2. don't init fbdev if there are no connectors
@@ -2257,6 +2261,7 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
 
+	amdgpu_xcp_dev_unplug(adev);
 	drm_dev_unplug(dev);
 
 	if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) {
@@ -2840,6 +2845,33 @@ static const struct drm_driver amdgpu_kms_driver = {
 	.patchlevel = KMS_DRIVER_PATCHLEVEL,
 };
 
+const struct drm_driver amdgpu_partition_driver = {
+	.driver_features =
+	    DRIVER_GEM | DRIVER_RENDER | DRIVER_SYNCOBJ |
+	    DRIVER_SYNCOBJ_TIMELINE,
+	.open = amdgpu_driver_open_kms,
+	.postclose = amdgpu_driver_postclose_kms,
+	.lastclose = amdgpu_driver_lastclose_kms,
+	.ioctls = amdgpu_ioctls_kms,
+	.num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms),
+	.dumb_create = amdgpu_mode_dumb_create,
+	.dumb_map_offset = amdgpu_mode_dumb_mmap,
+	.fops = &amdgpu_driver_kms_fops,
+	.release = &amdgpu_driver_release_kms,
+
+	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
+	.gem_prime_import = amdgpu_gem_prime_import,
+	.gem_prime_mmap = drm_gem_prime_mmap,
+
+	.name = DRIVER_NAME,
+	.desc = DRIVER_DESC,
+	.date = DRIVER_DATE,
+	.major = KMS_DRIVER_MAJOR,
+	.minor = KMS_DRIVER_MINOR,
+	.patchlevel = KMS_DRIVER_PATCHLEVEL,
+};
+
 static struct pci_error_handlers amdgpu_pci_err_handler = {
 	.error_detected	= amdgpu_pci_error_detected,
 	.mmio_enabled	= amdgpu_pci_mmio_enabled,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h
index 8178323e4beff..5bc2cb661af7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h
@@ -42,6 +42,8 @@
 #define DRIVER_DESC		"AMD GPU"
 #define DRIVER_DATE		"20150101"
 
+extern const struct drm_driver amdgpu_partition_driver;
+
 long amdgpu_drm_ioctl(struct file *filp,
 		      unsigned int cmd, unsigned long arg);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index bca226cc4e0b0..8b28b18e42915 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -22,6 +22,9 @@
  */
 #include "amdgpu.h"
 #include "amdgpu_xcp.h"
+#include "amdgpu_drv.h"
+
+#include <drm/drm_drv.h>
 
 static int __amdgpu_xcp_run(struct amdgpu_xcp_mgr *xcp_mgr,
 			    struct amdgpu_xcp_ip *xcp_ip, int xcp_state)
@@ -217,6 +220,31 @@ int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
 	return mode;
 }
 
+static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev)
+{
+	struct drm_device *p_ddev;
+	struct pci_dev *pdev;
+	struct drm_device *ddev;
+	int i;
+
+	pdev = adev->pdev;
+	ddev = adev_to_drm(adev);
+
+	for (i = 0; i < MAX_XCP; i++) {
+		p_ddev = drm_dev_alloc(&amdgpu_partition_driver,
+			&pci_upstream_bridge(pdev)->dev);
+		if (IS_ERR(p_ddev))
+			return PTR_ERR(p_ddev);
+
+		/* Redirect all IOCTLs to the primary device */
+		p_ddev->render->dev = ddev;
+		p_ddev->vma_offset_manager = ddev->vma_offset_manager;
+		adev->xcp_mgr->xcp[i].ddev = p_ddev;
+	}
+
+	return 0;
+}
+
 int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
 			int init_num_xcps,
 			struct amdgpu_xcp_mgr_funcs *xcp_funcs)
@@ -242,7 +270,7 @@ int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode,
 
 	adev->xcp_mgr = xcp_mgr;
 
-	return 0;
+	return amdgpu_xcp_dev_alloc(adev);
 }
 
 int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr,
@@ -278,3 +306,32 @@ int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp,
 
 	return 0;
 }
+
+int amdgpu_xcp_dev_register(struct amdgpu_device *adev,
+			const struct pci_device_id *ent)
+{
+	int i, ret;
+
+	if (!adev->xcp_mgr)
+		return 0;
+
+	for (i = 0; i < MAX_XCP; i++) {
+		ret = drm_dev_register(adev->xcp_mgr->xcp[i].ddev, ent->driver_data);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev)
+{
+	int i;
+
+	if (!adev->xcp_mgr)
+		return;
+
+	for (i = 0; i < MAX_XCP; i++)
+		drm_dev_unplug(adev->xcp_mgr->xcp[i].ddev);
+}
+
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index e1319b887bf3e..dad0b98d1ae7c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -70,6 +70,7 @@ struct amdgpu_xcp {
 	uint8_t id;
 	uint8_t mem_id;
 	bool valid;
+	struct drm_device *ddev;
 };
 
 struct amdgpu_xcp_mgr {
@@ -115,6 +116,10 @@ int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp,
 				enum AMDGPU_XCP_IP_BLOCK ip,
 				uint32_t *inst_mask);
 
+int amdgpu_xcp_dev_register(struct amdgpu_device *adev,
+				const struct pci_device_id *ent);
+void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev);
+
 static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr)
 {
 	if (!xcp_mgr)
-- 
GitLab


From be3800f57c3b0fb39dc732345279db76a50559a3 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 28 Feb 2023 14:16:38 -0500
Subject: [PATCH 1302/1791] drm/amdgpu: find partition ID when open device
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Find partition ID when open device from render device minor.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-and-tested-by: Philip Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c |  4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 29 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  3 +++
 4 files changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index eb2fb968e3af5..bba8cfeff71f0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -467,6 +467,8 @@ struct amdgpu_fpriv {
 	struct mutex		bo_list_lock;
 	struct idr		bo_list_handles;
 	struct amdgpu_ctx_mgr	ctx_mgr;
+	/** GPU partition selection */
+	uint32_t		xcp_id;
 };
 
 int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 44997c7ee89d2..879718598fa4c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1223,6 +1223,10 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 		goto out_suspend;
 	}
 
+	r = amdgpu_xcp_open_device(adev, fpriv, file_priv);
+	if (r)
+		return r;
+
 	pasid = amdgpu_pasid_alloc(16);
 	if (pasid < 0) {
 		dev_warn(adev->dev, "No more PASIDs available!");
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 8b28b18e42915..9b627a8b1d5c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -335,3 +335,32 @@ void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev)
 		drm_dev_unplug(adev->xcp_mgr->xcp[i].ddev);
 }
 
+int amdgpu_xcp_open_device(struct amdgpu_device *adev,
+			   struct amdgpu_fpriv *fpriv,
+			   struct drm_file *file_priv)
+{
+	int i;
+
+	if (!adev->xcp_mgr)
+		return 0;
+
+	fpriv->xcp_id = ~0;
+	for (i = 0; i < MAX_XCP; ++i) {
+		if (!adev->xcp_mgr->xcp[i].ddev)
+			break;
+
+		if (file_priv->minor == adev->xcp_mgr->xcp[i].ddev->render) {
+			if (adev->xcp_mgr->xcp[i].valid == FALSE) {
+				dev_err(adev->dev, "renderD%d partition %d not valid!",
+						file_priv->minor->index, i);
+				return -ENOENT;
+			}
+			dev_dbg(adev->dev, "renderD%d partition %d openned!",
+					file_priv->minor->index, i);
+			fpriv->xcp_id = i;
+			break;
+		}
+	}
+	return 0;
+}
+
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index dad0b98d1ae7c..ad60520f952cf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -119,6 +119,9 @@ int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp,
 int amdgpu_xcp_dev_register(struct amdgpu_device *adev,
 				const struct pci_device_id *ent);
 void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev);
+int amdgpu_xcp_open_device(struct amdgpu_device *adev,
+			   struct amdgpu_fpriv *fpriv,
+			   struct drm_file *file_priv);
 
 static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr)
 {
-- 
GitLab


From 797a0a142ca7f3b823ae1032983111c055bc50fb Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 16:45:12 -0400
Subject: [PATCH 1303/1791] drm/amdgpu: add partition ID track in ring

Keep track partition ID in ring.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h      |  1 +
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 41 +++++++++++++++++++
 2 files changed, 42 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index d2b1a88546038..4a4c9f89c302a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -252,6 +252,7 @@ struct amdgpu_ring {
 	uint32_t		buf_mask;
 	u32			idx;
 	u32			xcc_id;
+	u32			xcp_id;
 	u32			me;
 	u32			pipe;
 	u32			queue;
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 97011e7e031d4..c90ea34ef9ec1 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -61,6 +61,47 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev)
 	adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1;
 }
 
+static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev,
+			     uint32_t inst_idx, struct amdgpu_ring *ring)
+{
+	int xcp_id;
+	enum AMDGPU_XCP_IP_BLOCK ip_blk;
+	uint32_t inst_mask;
+
+	ring->xcp_id = ~0;
+	if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE)
+		return;
+
+	inst_mask = 1 << inst_idx;
+
+	switch (ring->funcs->type) {
+	case AMDGPU_HW_IP_GFX:
+	case AMDGPU_RING_TYPE_COMPUTE:
+	case AMDGPU_RING_TYPE_KIQ:
+		ip_blk = AMDGPU_XCP_GFX;
+		break;
+	case AMDGPU_RING_TYPE_SDMA:
+		ip_blk = AMDGPU_XCP_SDMA;
+		break;
+	case AMDGPU_RING_TYPE_VCN_ENC:
+	case AMDGPU_RING_TYPE_VCN_JPEG:
+		ip_blk = AMDGPU_XCP_VCN;
+		if (adev->xcp_mgr->mode == AMDGPU_CPX_PARTITION_MODE)
+			inst_mask = 1 << (inst_idx * 2);
+		break;
+	default:
+		DRM_ERROR("Not support ring type %d!", ring->funcs->type);
+		return;
+	}
+
+	for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) {
+		if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) {
+			ring->xcp_id = xcp_id;
+			break;
+		}
+	}
+}
+
 static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 					 enum amd_hw_ip_block_type block,
 					 int8_t inst)
-- 
GitLab


From 0a9115fd952a5de27dc360dc0c4618b6a1846c58 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 16:55:02 -0400
Subject: [PATCH 1304/1791] drm/amdgpu: update header to support partition
 scheduling

Update header to support partition scheduling.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index ad60520f952cf..cca06d38b03dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -70,7 +70,9 @@ struct amdgpu_xcp {
 	uint8_t id;
 	uint8_t mem_id;
 	bool valid;
+	atomic_t	ref_cnt;
 	struct drm_device *ddev;
+	struct amdgpu_sched	gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX];
 };
 
 struct amdgpu_xcp_mgr {
@@ -97,6 +99,10 @@ struct amdgpu_xcp_mgr_funcs {
 	int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
 	int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
 	int (*resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
+	int (*select_scheds)(struct amdgpu_device *adev,
+				  u32 hw_ip, u32 hw_prio, struct amdgpu_fpriv *fpriv,
+				  unsigned int *num_scheds, struct drm_gpu_scheduler ***scheds);
+	int (*update_partition_sched_list)(struct amdgpu_device *adev);
 };
 
 int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id);
@@ -123,6 +129,15 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev,
 			   struct amdgpu_fpriv *fpriv,
 			   struct drm_file *file_priv);
 
+#define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \
+	((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \
+	(adev)->xcp_mgr->funcs->select_scheds ? \
+	(adev)->xcp_mgr->funcs->select_scheds((adev), (e), (c), (d), (x), (y)) : -ENOENT)
+#define amdgpu_xcp_update_partition_sched_list(adev) \
+	((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \
+	(adev)->xcp_mgr->funcs->update_partition_sched_list ? \
+	(adev)->xcp_mgr->funcs->update_partition_sched_list(adev) : 0)
+
 static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr)
 {
 	if (!xcp_mgr)
-- 
GitLab


From d425c6f48b189f0a5a7c7d26980fd7a2114fb35d Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 17:19:11 -0400
Subject: [PATCH 1305/1791] drm/amdgpu: add partition scheduler list update

Add partition scheduler list update in late init
and xcp partition mode switch.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c    |  2 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c       |  2 +
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 67 ++++++++++++++++++-
 3 files changed, 70 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 4d9c535bcb0cb..02ee79b7b56d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2474,6 +2474,8 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev)
 		}
 	}
 
+	amdgpu_xcp_update_partition_sched_list(adev);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 9b627a8b1d5c4..78fce5aab2180 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -118,6 +118,7 @@ static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 
 int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 {
+	struct amdgpu_device *adev = xcp_mgr->adev;
 	struct amdgpu_xcp_ip ip;
 	uint8_t mem_id;
 	int i, j, ret;
@@ -153,6 +154,7 @@ int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 	}
 
 	xcp_mgr->num_xcps = num_xcps;
+	amdgpu_xcp_update_partition_sched_list(adev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index c90ea34ef9ec1..073ae95e6dd61 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -102,6 +102,70 @@ static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev,
 	}
 }
 
+static void aqua_vanjaram_xcp_gpu_sched_update(
+		struct amdgpu_device *adev,
+		struct amdgpu_ring *ring,
+		unsigned int sel_xcp_id)
+{
+	unsigned int *num_gpu_sched;
+
+	num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id]
+			.gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds;
+	adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio]
+			.sched[(*num_gpu_sched)++] = &ring->sched;
+	DRM_DEBUG("%s :[%d] gpu_sched[%d][%d] = %d", ring->name,
+			sel_xcp_id, ring->funcs->type,
+			ring->hw_prio, *num_gpu_sched);
+}
+
+static int aqua_vanjaram_xcp_sched_list_update(
+		struct amdgpu_device *adev)
+{
+	struct amdgpu_ring *ring;
+	int i;
+
+	for (i = 0; i < MAX_XCP; i++) {
+		atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0);
+		memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched));
+	}
+
+	if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE)
+		return 0;
+
+	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
+		ring = adev->rings[i];
+		if (!ring || !ring->sched.ready)
+			continue;
+
+		aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id);
+
+		/* VCN is shared by two partitions under CPX MODE */
+		if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC ||
+			ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) &&
+			adev->xcp_mgr->mode == AMDGPU_CPX_PARTITION_MODE)
+			aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1);
+	}
+
+	return 0;
+}
+
+static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev)
+{
+	int i;
+
+	for (i = 0; i < adev->num_rings; i++) {
+		struct amdgpu_ring *ring = adev->rings[i];
+
+		if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE ||
+			ring->funcs->type == AMDGPU_RING_TYPE_KIQ)
+			aqua_vanjaram_set_xcp_id(adev, ring->xcc_id, ring);
+		else
+			aqua_vanjaram_set_xcp_id(adev, ring->me, ring);
+	}
+
+	return aqua_vanjaram_xcp_sched_list_update(adev);
+}
+
 static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 					 enum amd_hw_ip_block_type block,
 					 int8_t inst)
@@ -483,7 +547,8 @@ struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = {
 	.switch_partition_mode = &aqua_vanjaram_switch_partition_mode,
 	.query_partition_mode = &aqua_vanjaram_query_partition_mode,
 	.get_ip_details = &aqua_vanjaram_get_xcp_ip_details,
-	.get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id
+	.get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id,
+	.update_partition_sched_list = &aqua_vanjaram_update_partition_sched_list
 };
 
 static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev)
-- 
GitLab


From c30e326e488ec43f6b0931f16ddba16a3370ed4a Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 17:12:01 -0400
Subject: [PATCH 1306/1791] drm/amdgpu: keep amdgpu_ctx_mgr in ctx structure

Keep amdgpu_ctx_mgr in ctx structure to track fpriv.

v2: add missing fpriv declaration lost in rebase

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index e1f642a3dc2fb..ef279481d8fb1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -303,6 +303,7 @@ static int amdgpu_ctx_get_stable_pstate(struct amdgpu_ctx *ctx,
 static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority,
 			   struct drm_file *filp, struct amdgpu_ctx *ctx)
 {
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	u32 current_stable_pstate;
 	int r;
 
@@ -331,6 +332,7 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority,
 	else
 		ctx->stable_pstate = current_stable_pstate;
 
+	ctx->ctx_mgr = &(fpriv->ctx_mgr);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
index 0fa0e56daf67e..f1e27b6e16f48 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
@@ -57,6 +57,7 @@ struct amdgpu_ctx {
 	unsigned long			ras_counter_ce;
 	unsigned long			ras_counter_ue;
 	uint32_t			stable_pstate;
+	struct amdgpu_ctx_mgr		*ctx_mgr;
 };
 
 struct amdgpu_ctx_mgr {
-- 
GitLab


From cd7d8400aa04ba989a87949cf4611b7e16af274f Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 17:00:54 -0400
Subject: [PATCH 1307/1791] drm/amdgpu: add partition schedule for GC(9, 4, 3)

Implement partition schedule for GC(9, 4, 3).

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c   | 41 +++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 073ae95e6dd61..4ca932a62ce67 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -166,6 +166,46 @@ static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev)
 	return aqua_vanjaram_xcp_sched_list_update(adev);
 }
 
+int aqua_vanjaram_select_scheds(
+		struct amdgpu_device *adev,
+		u32 hw_ip,
+		u32 hw_prio,
+		struct amdgpu_fpriv *fpriv,
+		unsigned int *num_scheds,
+		struct drm_gpu_scheduler ***scheds)
+{
+	u32 sel_xcp_id;
+	int i;
+
+	if (fpriv->xcp_id == ~0) {
+		u32 least_ref_cnt = ~0;
+
+		fpriv->xcp_id = 0;
+		for (i = 0; i < adev->xcp_mgr->num_xcps; i++) {
+			u32 total_ref_cnt;
+
+			total_ref_cnt = atomic_read(&adev->xcp_mgr->xcp[i].ref_cnt);
+			if (total_ref_cnt < least_ref_cnt) {
+				fpriv->xcp_id = i;
+				least_ref_cnt = total_ref_cnt;
+			}
+		}
+	}
+	sel_xcp_id = fpriv->xcp_id;
+
+	if (adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) {
+		*num_scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds;
+		*scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched;
+		atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt);
+		DRM_DEBUG("Selected partition #%d", sel_xcp_id);
+	} else {
+		DRM_ERROR("Failed to schedule partition #%d.", sel_xcp_id);
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
 static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 					 enum amd_hw_ip_block_type block,
 					 int8_t inst)
@@ -548,6 +588,7 @@ struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = {
 	.query_partition_mode = &aqua_vanjaram_query_partition_mode,
 	.get_ip_details = &aqua_vanjaram_get_xcp_ip_details,
 	.get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id,
+	.select_scheds = &aqua_vanjaram_select_scheds,
 	.update_partition_sched_list = &aqua_vanjaram_update_partition_sched_list
 };
 
-- 
GitLab


From 9a18292d41ae201333fc4203b3e7987ce5c1de6e Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 17:15:02 -0400
Subject: [PATCH 1308/1791] drm/amdgpu: run partition schedule if it is
 supported

Run partition schedule if it is supported during ctx init entity.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index ef279481d8fb1..333757486098c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -222,8 +222,19 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 	drm_prio = amdgpu_ctx_to_drm_sched_prio(ctx_prio);
 
 	hw_ip = array_index_nospec(hw_ip, AMDGPU_HW_IP_NUM);
-	scheds = adev->gpu_sched[hw_ip][hw_prio].sched;
-	num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds;
+
+	if (!(adev)->xcp_mgr) {
+		scheds = adev->gpu_sched[hw_ip][hw_prio].sched;
+		num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds;
+	} else {
+		struct amdgpu_fpriv *fpriv;
+
+		fpriv = container_of(ctx->ctx_mgr, struct amdgpu_fpriv, ctx_mgr);
+		r = amdgpu_xcp_select_scheds(adev, hw_ip, hw_prio, fpriv,
+						&num_scheds, &scheds);
+		if (r)
+			goto cleanup_entity;
+	}
 
 	/* disable load balance if the hw engine retains context among dependent jobs */
 	if (hw_ip == AMDGPU_HW_IP_VCN_ENC ||
-- 
GitLab


From 3e7c6fe38724eab767033f9d26b496bc2e815157 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 15 Aug 2022 17:21:44 -0400
Subject: [PATCH 1309/1791] drm/amdgpu: update ref_cnt before ctx free

Update ref_cnt before ctx free.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c |  7 +++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 16 ++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  2 ++
 3 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 333757486098c..410acdd4554ca 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -266,7 +266,8 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 	return r;
 }
 
-static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity)
+static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_device *adev,
+				  struct amdgpu_ctx_entity *entity)
 {
 	ktime_t res = ns_to_ktime(0);
 	int i;
@@ -279,6 +280,8 @@ static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity)
 		dma_fence_put(entity->fences[i]);
 	}
 
+	amdgpu_xcp_release_sched(adev, entity);
+
 	kfree(entity);
 	return res;
 }
@@ -412,7 +415,7 @@ static void amdgpu_ctx_fini(struct kref *ref)
 		for (j = 0; j < AMDGPU_MAX_ENTITY_NUM; ++j) {
 			ktime_t spend;
 
-			spend = amdgpu_ctx_fini_entity(ctx->entities[i][j]);
+			spend = amdgpu_ctx_fini_entity(adev, ctx->entities[i][j]);
 			atomic64_add(ktime_to_ns(spend), &mgr->time_spend[i]);
 		}
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 78fce5aab2180..9b960ba0b7ac2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -366,3 +366,19 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev,
 	return 0;
 }
 
+void amdgpu_xcp_release_sched(struct amdgpu_device *adev,
+				  struct amdgpu_ctx_entity *entity)
+{
+	struct drm_gpu_scheduler *sched;
+	struct amdgpu_ring *ring;
+
+	if (!adev->xcp_mgr)
+		return;
+
+	sched = entity->entity.rq->sched;
+	if (sched->ready) {
+		ring = to_amdgpu_ring(entity->entity.rq->sched);
+		atomic_dec(&adev->xcp_mgr->xcp[ring->xcp_id].ref_cnt);
+	}
+}
+
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index cca06d38b03dd..39aca87ce204c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -128,6 +128,8 @@ void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev);
 int amdgpu_xcp_open_device(struct amdgpu_device *adev,
 			   struct amdgpu_fpriv *fpriv,
 			   struct drm_file *file_priv);
+void amdgpu_xcp_release_sched(struct amdgpu_device *adev,
+			      struct amdgpu_ctx_entity *entity);
 
 #define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \
 	((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \
-- 
GitLab


From d26ea1b346e71c07aa00956c32fe2d2dbec068ec Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Fri, 3 Mar 2023 19:45:45 -0500
Subject: [PATCH 1310/1791] drm/amdgpu: Add xcp manager
 num_xcp_per_mem_partition

Used by KFD to check memory limit accounting.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 9b960ba0b7ac2..f2981d21d4e07 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -156,6 +156,7 @@ int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 	xcp_mgr->num_xcps = num_xcps;
 	amdgpu_xcp_update_partition_sched_list(adev);
 
+	xcp_mgr->num_xcp_per_mem_partition = num_xcps / xcp_mgr->adev->gmc.num_mem_partitions;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 39aca87ce204c..68b63b970ce8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -83,6 +83,9 @@ struct amdgpu_xcp_mgr {
 	struct amdgpu_xcp xcp[MAX_XCP];
 	uint8_t num_xcps;
 	int8_t mode;
+
+	 /* Used to determine KFD memory size limits per XCP */
+	unsigned int num_xcp_per_mem_partition;
 };
 
 struct amdgpu_xcp_mgr_funcs {
-- 
GitLab


From a476c0c645535cc0361938becb440b4239996079 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 23 Feb 2023 11:03:37 -0500
Subject: [PATCH 1311/1791] drm/amdkfd: Store drm node minor number for kfd
 nodes

From KFD topology, application will find kfd node with the corresponding
drm device node minor number, for example if partition drm node starts
from /dev/dri/renderD129, then KFD node 0 with store drm node minor
number 129. Application will open drm node /dev/dri/renderD129 to create
amdgpu vm for kfd node 0 with the correct vm->mem_id to indicate the
memory partition.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 6d6243b978e1c..a8e25aecf839b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1942,8 +1942,12 @@ int kfd_topology_add_device(struct kfd_node *gpu)
 		amdgpu_amdkfd_get_max_engine_clock_in_mhz(dev->gpu->adev);
 	dev->node_props.max_engine_clk_ccompute =
 		cpufreq_quick_get_max(0) / 1000;
-	dev->node_props.drm_render_minor =
-		gpu->kfd->shared_resources.drm_render_minor;
+
+	if (gpu->xcp)
+		dev->node_props.drm_render_minor = gpu->xcp->ddev->render->index;
+	else
+		dev->node_props.drm_render_minor =
+				gpu->kfd->shared_resources.drm_render_minor;
 
 	dev->node_props.hive_id = gpu->kfd->hive_id;
 	dev->node_props.num_sdma_engines = kfd_get_num_sdma_engines(gpu);
-- 
GitLab


From 934deb64fdf220d2caf978d22615bcc7c9f6897e Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 23 Feb 2023 19:58:22 -0500
Subject: [PATCH 1312/1791] drm/amdgpu: Add memory partition id to amdgpu_vm

If xcp_mgr is initialized, add mem_id to amdgpu_vm structure to store
memory partition number when creating amdgpu_vm for the xcp. The xcp
number is decided when opening the render device, for example
/dev/dri/renderD129 is xcp_id 0, /dev/dri/renderD130 is xcp_id 1.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 8 ++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h  | 3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 3 +++
 3 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 879718598fa4c..815098be4c2fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1223,10 +1223,6 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 		goto out_suspend;
 	}
 
-	r = amdgpu_xcp_open_device(adev, fpriv, file_priv);
-	if (r)
-		return r;
-
 	pasid = amdgpu_pasid_alloc(16);
 	if (pasid < 0) {
 		dev_warn(adev->dev, "No more PASIDs available!");
@@ -1237,6 +1233,10 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 	if (r)
 		goto error_pasid;
 
+	r = amdgpu_xcp_open_device(adev, fpriv, file_priv);
+	if (r)
+		goto error_vm;
+
 	r = amdgpu_vm_set_pasid(adev, &fpriv->vm, pasid);
 	if (r)
 		goto error_vm;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 8add5f5eb92a2..14f9a2bf3acb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -329,6 +329,9 @@ struct amdgpu_vm {
 	struct ttm_lru_bulk_move lru_bulk_move;
 	/* Flag to indicate if VM is used for compute */
 	bool			is_compute_context;
+
+	/* Memory partition number, -1 means any partition */
+	int8_t			mem_id;
 };
 
 struct amdgpu_vm_manager {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index f2981d21d4e07..610c32c4f5afb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -364,6 +364,9 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev,
 			break;
 		}
 	}
+
+	fpriv->vm.mem_id = fpriv->xcp_id == ~0 ? -1 :
+				adev->xcp_mgr->xcp[fpriv->xcp_id].mem_id;
 	return 0;
 }
 
-- 
GitLab


From 4c6ce75fdd628c43aea11448ed41b52119dae42b Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 26 Jan 2023 18:11:29 -0500
Subject: [PATCH 1313/1791] drm/amdkfd: Show KFD node memory partition info

Show KFD node memory partition id and size, add helper function
KFD_XCP_MEMORY_SIZE to get kfd node memory size, will be used
later to support memory accounting per partition.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 5 +++++
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    | 7 ++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index e4e1dbba060aa..324cb566ca2f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -330,6 +330,11 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		uint64_t size, u32 alloc_flag);
 
+#define KFD_XCP_MEMORY_SIZE(n) ((n)->adev->gmc.num_mem_partitions ?\
+		(n)->adev->gmc.mem_partitions[(n)->xcp->mem_id].size /\
+		(n)->adev->xcp_mgr->num_xcp_per_mem_partition :\
+		(n)->adev->gmc.real_vram_size)
+
 #if IS_ENABLED(CONFIG_HSA_AMD)
 void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
 void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index b5497d2ee984e..db5b53fcdf11a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -724,7 +724,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd_cwsr_init(kfd);
 
-	/* TODO: Needs to be updated for memory partitioning */
 	svm_migrate_init(kfd->adev);
 
 	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
@@ -754,6 +753,12 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 				(1U << NUM_XCC(kfd->adev->gfx.xcc_mask)) - 1;
 		}
 
+		if (node->xcp) {
+			dev_info(kfd_device, "KFD node %d partition %d size %lldM\n",
+				node->node_id, node->xcp->mem_id,
+				KFD_XCP_MEMORY_SIZE(node) >> 20);
+		}
+
 		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
 		    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
 		    kfd->num_nodes != 1) {
-- 
GitLab


From f24e924b7e8aba7b62671e7e1a19d83301a08597 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 26 Jan 2023 18:25:28 -0500
Subject: [PATCH 1314/1791] drm/amdgpu: Add memory partition mem_id to
 amdgpu_bo

Add mem_id_plus1 parameter to amdgpu_gem_object_create and pass it to
amdgpu_bo_create. For dGPU mode allocation, mem_id is used by VRAM
manager to get the memory partition fpfn, lpfn from xcp manager. For APU
native mode allocation, mem_id is used to get NUMA node id from xcp
manager, then pass to TTM as numa pool id to alloc memory from the
specific NUMA node. mem_id -1 means for entire VRAM or any NUMA nodes.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c      | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c          | 9 +++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h          | 3 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c       | 3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h       | 5 +++++
 6 files changed, 17 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index fd395cd61b540..71f1e19909254 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -290,7 +290,7 @@ create_dmamap_sg_bo(struct amdgpu_device *adev,
 
 	ret = amdgpu_gem_object_create(adev, mem->bo->tbo.base.size, 1,
 			AMDGPU_GEM_DOMAIN_CPU, AMDGPU_GEM_CREATE_PREEMPTIBLE | flags,
-			ttm_bo_type_sg, mem->bo->tbo.base.resv, &gem_obj);
+			ttm_bo_type_sg, mem->bo->tbo.base.resv, &gem_obj, 0);
 
 	amdgpu_bo_unreserve(mem->bo);
 
@@ -1721,7 +1721,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 			va, (*mem)->aql_queue ? size << 1 : size, domain_string(alloc_domain));
 
 	ret = amdgpu_gem_object_create(adev, aligned_size, 1, alloc_domain, alloc_flags,
-				       bo_type, NULL, &gobj);
+				       bo_type, NULL, &gobj, 0);
 	if (ret) {
 		pr_debug("Failed to create BO on domain %s. ret %d\n",
 			 domain_string(alloc_domain), ret);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index 6dceaf40625b2..12210598e5b8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -336,7 +336,7 @@ amdgpu_dma_buf_create_obj(struct drm_device *dev, struct dma_buf *dma_buf)
 
 	ret = amdgpu_gem_object_create(adev, dma_buf->size, PAGE_SIZE,
 				       AMDGPU_GEM_DOMAIN_CPU, flags,
-				       ttm_bo_type_sg, resv, &gobj);
+				       ttm_bo_type_sg, resv, &gobj, 0);
 	if (ret)
 		goto error;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 863cb668e0005..33ebee18b80d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -98,7 +98,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 			     int alignment, u32 initial_domain,
 			     u64 flags, enum ttm_bo_type type,
 			     struct dma_resv *resv,
-			     struct drm_gem_object **obj)
+			     struct drm_gem_object **obj, int8_t mem_id_plus1)
 {
 	struct amdgpu_bo *bo;
 	struct amdgpu_bo_user *ubo;
@@ -116,6 +116,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 	bp.flags = flags;
 	bp.domain = initial_domain;
 	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+	bp.mem_id_plus1 = mem_id_plus1;
 
 	r = amdgpu_bo_create_user(adev, &bp, &ubo);
 	if (r)
@@ -336,7 +337,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 retry:
 	r = amdgpu_gem_object_create(adev, size, args->in.alignment,
 				     initial_domain,
-				     flags, ttm_bo_type_device, resv, &gobj);
+				     flags, ttm_bo_type_device, resv, &gobj, 0);
 	if (r && r != -ERESTARTSYS) {
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) {
 			flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
@@ -405,7 +406,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 
 	/* create a gem object to contain this object in */
 	r = amdgpu_gem_object_create(adev, args->size, 0, AMDGPU_GEM_DOMAIN_CPU,
-				     0, ttm_bo_type_device, NULL, &gobj);
+				     0, ttm_bo_type_device, NULL, &gobj, 0);
 	if (r)
 		return r;
 
@@ -931,7 +932,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 	domain = amdgpu_bo_get_preferred_domain(adev,
 				amdgpu_display_supported_domains(adev, flags));
 	r = amdgpu_gem_object_create(adev, args->size, 0, domain, flags,
-				     ttm_bo_type_device, NULL, &gobj);
+				     ttm_bo_type_device, NULL, &gobj, 0);
 	if (r)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
index 637bf51dbf06f..646c4fcc8e404 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
@@ -43,8 +43,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 			     int alignment, u32 initial_domain,
 			     u64 flags, enum ttm_bo_type type,
 			     struct dma_resv *resv,
-			     struct drm_gem_object **obj);
-
+			     struct drm_gem_object **obj, int8_t mem_id_plus1);
 int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 			    struct drm_device *dev,
 			    struct drm_mode_create_dumb *args);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 725530eb4e0c1..628632efabc89 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -574,6 +574,9 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 
 	bo->flags = bp->flags;
 
+	/* bo->mem_id -1 means any partition */
+	bo->mem_id = bp->mem_id_plus1 - 1;
+
 	if (!amdgpu_bo_support_uswc(bo->flags))
 		bo->flags &= ~AMDGPU_GEM_CREATE_CPU_GTT_USWC;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 35b8106816a13..eb24a66ccee59 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -56,6 +56,8 @@ struct amdgpu_bo_param {
 	bool				no_wait_gpu;
 	struct dma_resv			*resv;
 	void				(*destroy)(struct ttm_buffer_object *bo);
+	/* memory partition number plus 1, 0 means any partition */
+	int8_t				mem_id_plus1;
 };
 
 /* bo virtual addresses in a vm */
@@ -108,6 +110,9 @@ struct amdgpu_bo {
 	struct mmu_interval_notifier	notifier;
 #endif
 	struct kgd_mem                  *kfd_bo;
+
+	/* memory partition number, -1 means any partition */
+	int8_t				mem_id;
 };
 
 struct amdgpu_bo_user {
-- 
GitLab


From 53c5692e7a3c8e8eed3ec6b876a3c982d217a5d7 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 26 Jan 2023 18:50:09 -0500
Subject: [PATCH 1315/1791] drm/amdkfd: Alloc memory of GPU support memory
 partition

For dGPU mode VRAM allocation, create amdgpu_bo from amdgpu_vm->mem_id,
to alloc from the correct memory range.

For APU mode VRAM allocation, set alloc domain to GTT, and set
bp->mem_id_plus1 from amdgpu_vm->mem_id + 1 to create amdgpu_bo, to
allocate system memory from correct NUMA node.

For GTT allocation, use mem_id -1 to allocate system memory from any
NUMA nodes.

Remove amdgpu_ttm_tt_set_mem_pool, to avoid the confusion that memory
maybe allocated from different mem_id.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 24 ++++++-------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c       | 20 +---------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h       |  1 -
 3 files changed, 8 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 71f1e19909254..c234dc0db7991 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1641,9 +1641,9 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	struct drm_gem_object *gobj = NULL;
 	u32 domain, alloc_domain;
 	uint64_t aligned_size;
+	int8_t mem_id = -1;
 	u64 alloc_flags;
 	int ret;
-	int mem_id = 0; /* Fixme : to be changed when mem_id support patch lands, until then NPS1, SPX only */
 
 	/*
 	 * Check on which domain to allocate BO
@@ -1653,13 +1653,14 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 
 		if (adev->gmc.is_app_apu) {
 			domain = AMDGPU_GEM_DOMAIN_GTT;
-			alloc_domain = AMDGPU_GEM_DOMAIN_CPU;
+			alloc_domain = AMDGPU_GEM_DOMAIN_GTT;
 			alloc_flags = 0;
 		} else {
 			alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE;
 			alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ?
 			AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : 0;
 		}
+		mem_id = avm->mem_id;
 	} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) {
 		domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT;
 		alloc_flags = 0;
@@ -1717,11 +1718,12 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		goto err_reserve_limit;
 	}
 
-	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s\n",
-			va, (*mem)->aql_queue ? size << 1 : size, domain_string(alloc_domain));
+	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s mem_id %d\n",
+		 va, (*mem)->aql_queue ? size << 1 : size,
+		 domain_string(alloc_domain), mem_id);
 
 	ret = amdgpu_gem_object_create(adev, aligned_size, 1, alloc_domain, alloc_flags,
-				       bo_type, NULL, &gobj, 0);
+				       bo_type, NULL, &gobj, mem_id + 1);
 	if (ret) {
 		pr_debug("Failed to create BO on domain %s. ret %d\n",
 			 domain_string(alloc_domain), ret);
@@ -1747,17 +1749,6 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	(*mem)->mapped_to_gpu_memory = 0;
 	(*mem)->process_info = avm->process_info;
 
-	if (adev->gmc.is_app_apu &&
-	    ((*mem)->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)) {
-		bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT;
-		bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT;
-		ret = amdgpu_ttm_tt_set_mem_pool(&bo->tbo, mem_id);
-		if (ret) {
-			pr_debug("failed to set ttm mem pool %d\n", ret);
-			goto err_set_mem_partition;
-		}
-	}
-
 	add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr);
 
 	if (user_addr) {
@@ -1784,7 +1775,6 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 allocate_init_user_pages_failed:
 err_pin_bo:
 	remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info);
-err_set_mem_partition:
 	drm_vma_node_revoke(&gobj->vma_node, drm_priv);
 err_node_allow:
 	/* Don't unreserve system mem limit twice */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 09d1a98bd11ef..129c593cb2bd1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1060,7 +1060,7 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 		return NULL;
 	}
 	gtt->gobj = &bo->base;
-	gtt->pool_id = NUMA_NO_NODE;
+	gtt->pool_id = abo->mem_id;
 
 	if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 		caching = ttm_write_combined;
@@ -1155,24 +1155,6 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev,
 	return ttm_pool_free(pool, ttm);
 }
 
-/**
- * amdgpu_ttm_tt_set_mem_pool - Set the TTM memory pool for the TTM BO
- * @tbo: The ttm_buffer_object that backs the VRAM bo
- * @mem_id: to select the initialized ttm pool corresponding to the memory partition
- */
-int amdgpu_ttm_tt_set_mem_pool(struct ttm_buffer_object *tbo, int mem_id)
-{
-	struct ttm_tt *ttm = tbo->ttm;
-	struct amdgpu_ttm_tt *gtt;
-
-	if (!ttm && !ttm_tt_is_populated(ttm))
-		return -EINVAL;
-
-	gtt = ttm_to_amdgpu_ttm_tt(ttm);
-	gtt->pool_id = mem_id;
-	return 0;
-}
-
 /**
  * amdgpu_ttm_tt_get_userptr - Return the userptr GTT ttm_tt for the current
  * task
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index fe32de1bf4d51..8ef048a0a33e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -192,7 +192,6 @@ bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm);
 struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm);
 bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
 				  unsigned long end, unsigned long *userptr);
-int amdgpu_ttm_tt_set_mem_pool(struct ttm_buffer_object *tbo, int mem_id);
 bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
 				       int *last_invalidated);
 bool amdgpu_ttm_tt_is_userptr(struct ttm_tt *ttm);
-- 
GitLab


From 2046ed6c8aa951e4ae83c5022bb0a7c777386097 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 26 Jan 2023 18:45:32 -0500
Subject: [PATCH 1316/1791] drm/amdkfd: SVM range allocation support memory
 partition

Pass kfd node->xcp->mem_id to amdgpu bo create parameter mem_id_plus1 to
allocate new svm_bo on the specified memory partition.

This is only for dGPU mode as we don't migrate with APU mode.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index e6348d4133fd8..62aa7fb2eaa54 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -555,16 +555,20 @@ svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange,
 	bp.flags |= AMDGPU_GEM_CREATE_DISCARDABLE;
 	bp.type = ttm_bo_type_device;
 	bp.resv = NULL;
+	if (node->xcp)
+		bp.mem_id_plus1 = node->xcp->mem_id + 1;
 
-	/* TODO: Allocate memory from the right memory partition. We can sort
-	 * out the details later, once basic memory partitioning is working
-	 */
 	r = amdgpu_bo_create_user(node->adev, &bp, &ubo);
 	if (r) {
 		pr_debug("failed %d to create bo\n", r);
 		goto create_bo_failed;
 	}
 	bo = &ubo->bo;
+
+	pr_debug("alloc bo at offset 0x%lx size 0x%lx on partition %d\n",
+		 bo->tbo.resource->start << PAGE_SHIFT, bp.size,
+		 bp.mem_id_plus1 - 1);
+
 	r = amdgpu_bo_reserve(bo, true);
 	if (r) {
 		pr_debug("failed %d to reserve bo\n", r);
-- 
GitLab


From 7f6db89418f9d26eb68e050ff16de8e9827011ca Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 26 Jan 2023 18:54:29 -0500
Subject: [PATCH 1317/1791] drm/amdgpu: dGPU mode placement support memory
 partition

dGPU mode uses VRAM manager to validate bo, amdgpu bo placement use the
mem_id  to get the allocation range first, last page frame number
from xcp manager, pass to drm buddy allocator as the allowed range.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 628632efabc89..85ad355815fe6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -132,13 +132,18 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 	if (domain & AMDGPU_GEM_DOMAIN_VRAM) {
 		unsigned int visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
 
-		places[c].fpfn = 0;
-		places[c].lpfn = 0;
+		if (adev->gmc.mem_partitions && abo->mem_id >= 0) {
+			places[c].fpfn = adev->gmc.mem_partitions[abo->mem_id].range.fpfn;
+			places[c].lpfn = adev->gmc.mem_partitions[abo->mem_id].range.lpfn;
+		} else {
+			places[c].fpfn = 0;
+			places[c].lpfn = 0;
+		}
 		places[c].mem_type = TTM_PL_VRAM;
 		places[c].flags = 0;
 
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
-			places[c].lpfn = visible_pfn;
+			places[c].lpfn = min_not_zero(places[c].lpfn, visible_pfn);
 		else if (adev->gmc.real_vram_size != adev->gmc.visible_vram_size)
 			places[c].flags |= TTM_PL_FLAG_TOPDOWN;
 
-- 
GitLab


From dc12f9eddedb8b41f4dc948e5e636e5221fb4d43 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 2 Feb 2023 11:07:53 -0500
Subject: [PATCH 1318/1791] drm/amdkfd: Update MTYPE for far memory partition

Use MTYPE RW/MTYPE_CC for mapping system memory or VRAM to KFD node
within the same memory partition, use MTYPE_NC for mapping on KFD node
from the far memory partition of the same socket or from another socket
on same XGMI hive.

On NPS4 or 4P system, MTYPE will be overridden per page depending on
the memory NUMA node id and vm->mem_id.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 15 +++++++--------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c  |  9 +++++----
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 32eb4f4f54922..263d17a8b4334 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1186,7 +1186,7 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 	bool is_vram = bo->tbo.resource->mem_type == TTM_PL_VRAM;
 	bool coherent = bo->flags & AMDGPU_GEM_CREATE_COHERENT;
 	bool uncached = bo->flags & AMDGPU_GEM_CREATE_UNCACHED;
-	/* TODO: memory partitions struct amdgpu_vm *vm = mapping->bo_va->base.vm;*/
+	struct amdgpu_vm *vm = mapping->bo_va->base.vm;
 	unsigned int mtype_local, mtype;
 	bool snoop = false;
 	bool is_local;
@@ -1247,8 +1247,8 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		}
 		is_local = (!is_vram && (adev->flags & AMD_IS_APU) &&
 			    num_possible_nodes() <= 1) ||
-			   (is_vram && adev == bo_adev /* TODO: memory partitions &&
-			    bo->mem_id == vm->mem_id*/);
+			   (is_vram && adev == bo_adev &&
+			    bo->mem_id == vm->mem_id);
 		snoop = true;
 		if (uncached) {
 			mtype = MTYPE_UC;
@@ -1335,13 +1335,12 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 		return;
 	}
 
-	/* TODO: memory partitions. mem_id is hard-coded to 0 for now.
-	 * FIXME: Only supported on native mode for now. For carve-out, the
+	/* FIXME: Only supported on native mode for now. For carve-out, the
 	 * NUMA affinity of the GPU/VM needs to come from the PCI info because
 	 * memory partitions are not associated with different NUMA nodes.
 	 */
-	if (adev->gmc.is_app_apu) {
-		local_node = adev->gmc.mem_partitions[/*vm->mem_id*/0].numa.node;
+	if (adev->gmc.is_app_apu && vm->mem_id >= 0) {
+		local_node = adev->gmc.mem_partitions[vm->mem_id].numa.node;
 	} else {
 		dev_dbg(adev->dev, "Only native mode APU is supported.\n");
 		return;
@@ -1356,7 +1355,7 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 	}
 	nid = pfn_to_nid(addr >> PAGE_SHIFT);
 	dev_dbg(adev->dev, "vm->mem_id=%d, local_node=%d, nid=%d\n",
-		/*vm->mem_id*/0, local_node, nid);
+		vm->mem_id, local_node, nid);
 	if (nid == local_node) {
 		uint64_t old_flags = *flags;
 		unsigned int mtype_local = MTYPE_RW;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 62aa7fb2eaa54..a700d9ccd0548 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1203,8 +1203,8 @@ svm_range_get_pte_flags(struct kfd_node *node,
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
 		} else if (domain == SVM_RANGE_VRAM_DOMAIN) {
 			/* local HBM region close to partition */
-			if (bo_node->adev == node->adev /* TODO: memory partitions &&
-			    bo_node->mem_id == node->mem_id*/)
+			if (bo_node->adev == node->adev &&
+			    (!bo_node->xcp || !node->xcp || bo_node->xcp->mem_id == node->xcp->mem_id))
 				mapping_flags |= mtype_local;
 			/* local HBM region far from partition or remote XGMI GPU */
 			else if (svm_nodes_in_same_hive(bo_node, node))
@@ -1358,8 +1358,9 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 			 (last_domain == SVM_RANGE_VRAM_DOMAIN) ? 1 : 0,
 			 pte_flags);
 
-		/* TODO: we still need to determine the vm_manager.vram_base_offset based on
-		 * the memory partition.
+		/* For dGPU mode, we use same vm_manager to allocate VRAM for
+		 * different memory partition based on fpfn/lpfn, we should use
+		 * same vm_manager.vram_base_offset regardless memory partition.
 		 */
 		r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb, NULL,
 					   last_start, prange->start + i,
-- 
GitLab


From ea7bf2f22061a7bd77b17ddd4ac1bb3500ae823f Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 23 Feb 2023 20:00:05 -0500
Subject: [PATCH 1319/1791] drm/amdgpu: Alloc page table on correct memory
 partition

Alloc kernel mode page table bo uses the amdgpu_vm->mem_id + 1 as bp
mem_id_plus1 parameter. For APU mode, select the correct TTM pool to
alloc page from the corresponding memory partition, this will be the
closest NUMA node. For dGPU mode, select the correct address range for
vram manager.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index 60b1da93b06dd..62fc7e8d326ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -534,6 +534,8 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	bp.type = ttm_bo_type_kernel;
 	bp.no_wait_gpu = immediate;
+	bp.mem_id_plus1 = vm->mem_id + 1;
+
 	if (vm->root.bo)
 		bp.resv = vm->root.bo->tbo.base.resv;
 
@@ -558,6 +560,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	bp.type = ttm_bo_type_kernel;
 	bp.resv = bo->tbo.base.resv;
 	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+	bp.mem_id_plus1 = vm->mem_id + 1;
 
 	r = amdgpu_bo_create(adev, &bp, &(*vmbo)->shadow);
 
-- 
GitLab


From 6cfba94a77c5cbad2d7c106ac5f026b6a8b1efe6 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Tue, 7 Mar 2023 11:30:24 -0500
Subject: [PATCH 1320/1791] drm/amdgpu: dGPU mode set VRAM range lpfn as
 exclusive

TTM place lpfn is exclusive used as end (start + size) in drm and buddy
allocator, adev->gmc memory partition range lpfn is inclusive (start +
size - 1), should plus 1 to set TTM place lpfn.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 85ad355815fe6..b2d11c4f39b07 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -134,7 +134,11 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 
 		if (adev->gmc.mem_partitions && abo->mem_id >= 0) {
 			places[c].fpfn = adev->gmc.mem_partitions[abo->mem_id].range.fpfn;
-			places[c].lpfn = adev->gmc.mem_partitions[abo->mem_id].range.lpfn;
+			/*
+			 * memory partition range lpfn is inclusive start + size - 1
+			 * TTM place lpfn is exclusive start + size
+			 */
+			places[c].lpfn = adev->gmc.mem_partitions[abo->mem_id].range.lpfn + 1;
 		} else {
 			places[c].fpfn = 0;
 			places[c].lpfn = 0;
-- 
GitLab


From 3ebfd221c1a83e5f0edadb87d173d8fd93d1d125 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Wed, 8 Mar 2023 11:57:00 -0500
Subject: [PATCH 1321/1791] drm/amdkfd: Store xcp partition id to amdgpu bo

For memory accounting per compute partition and export drm amdgpu bo and
then import to KFD, we need the xcp id to account the memory usage or
find the KFD node of the original amdgpu bo to create the KFD bo on the
correct adev KFD node.

Set xcp_id_plus1 of amdgpu_bo_param to create bo and store xcp_id to
amddgpu bo. Add helper macro to get the mem_id from adev and xcp_id.

v2: squash in fix ("drm/amdgpu: Fix BO creation failure on GFX 9.4.3 dGPU")

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h       |  4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 11 ++++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c          |  4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h          |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c       | 15 ++++++++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h       | 12 ++++++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c          |  6 +++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c        |  5 +++--
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c            |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c             |  4 ++--
 10 files changed, 42 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 324cb566ca2f3..05c54776951b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -330,6 +330,10 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		uint64_t size, u32 alloc_flag);
 
+#define KFD_XCP_MEM_ID(adev, xcp_id) \
+		((adev)->xcp_mgr && (xcp_id) >= 0 ?\
+		(adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1)
+
 #define KFD_XCP_MEMORY_SIZE(n) ((n)->adev->gmc.num_mem_partitions ?\
 		(n)->adev->gmc.mem_partitions[(n)->xcp->mem_id].size /\
 		(n)->adev->xcp_mgr->num_xcp_per_mem_partition :\
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index c234dc0db7991..8724a0be31b8f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1634,6 +1634,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		uint64_t *offset, uint32_t flags, bool criu_resume)
 {
 	struct amdgpu_vm *avm = drm_priv_to_vm(drm_priv);
+	struct amdgpu_fpriv *fpriv = container_of(avm, struct amdgpu_fpriv, vm);
 	enum ttm_bo_type bo_type = ttm_bo_type_device;
 	struct sg_table *sg = NULL;
 	uint64_t user_addr = 0;
@@ -1641,7 +1642,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	struct drm_gem_object *gobj = NULL;
 	u32 domain, alloc_domain;
 	uint64_t aligned_size;
-	int8_t mem_id = -1;
+	int8_t xcp_id = -1;
 	u64 alloc_flags;
 	int ret;
 
@@ -1660,7 +1661,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 			alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ?
 			AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : 0;
 		}
-		mem_id = avm->mem_id;
+		xcp_id = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id;
 	} else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) {
 		domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT;
 		alloc_flags = 0;
@@ -1718,12 +1719,12 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		goto err_reserve_limit;
 	}
 
-	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s mem_id %d\n",
+	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s xcp_id %d\n",
 		 va, (*mem)->aql_queue ? size << 1 : size,
-		 domain_string(alloc_domain), mem_id);
+		 domain_string(alloc_domain), xcp_id);
 
 	ret = amdgpu_gem_object_create(adev, aligned_size, 1, alloc_domain, alloc_flags,
-				       bo_type, NULL, &gobj, mem_id + 1);
+				       bo_type, NULL, &gobj, xcp_id + 1);
 	if (ret) {
 		pr_debug("Failed to create BO on domain %s. ret %d\n",
 			 domain_string(alloc_domain), ret);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 33ebee18b80d9..7e8839cc6f58b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -98,7 +98,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 			     int alignment, u32 initial_domain,
 			     u64 flags, enum ttm_bo_type type,
 			     struct dma_resv *resv,
-			     struct drm_gem_object **obj, int8_t mem_id_plus1)
+			     struct drm_gem_object **obj, int8_t xcp_id_plus1)
 {
 	struct amdgpu_bo *bo;
 	struct amdgpu_bo_user *ubo;
@@ -116,7 +116,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 	bp.flags = flags;
 	bp.domain = initial_domain;
 	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
-	bp.mem_id_plus1 = mem_id_plus1;
+	bp.xcp_id_plus1 = xcp_id_plus1;
 
 	r = amdgpu_bo_create_user(adev, &bp, &ubo);
 	if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
index 646c4fcc8e404..f30264782ba27 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
@@ -43,7 +43,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
 			     int alignment, u32 initial_domain,
 			     u64 flags, enum ttm_bo_type type,
 			     struct dma_resv *resv,
-			     struct drm_gem_object **obj, int8_t mem_id_plus1);
+			     struct drm_gem_object **obj, int8_t xcp_id_plus1);
 int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 			    struct drm_device *dev,
 			    struct drm_mode_create_dumb *args);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index b2d11c4f39b07..42c02f48c3a87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -131,14 +131,15 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 
 	if (domain & AMDGPU_GEM_DOMAIN_VRAM) {
 		unsigned int visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT;
+		int8_t mem_id = KFD_XCP_MEM_ID(adev, abo->xcp_id);
 
-		if (adev->gmc.mem_partitions && abo->mem_id >= 0) {
-			places[c].fpfn = adev->gmc.mem_partitions[abo->mem_id].range.fpfn;
+		if (adev->gmc.mem_partitions && mem_id >= 0) {
+			places[c].fpfn = adev->gmc.mem_partitions[mem_id].range.fpfn;
 			/*
 			 * memory partition range lpfn is inclusive start + size - 1
 			 * TTM place lpfn is exclusive start + size
 			 */
-			places[c].lpfn = adev->gmc.mem_partitions[abo->mem_id].range.lpfn + 1;
+			places[c].lpfn = adev->gmc.mem_partitions[mem_id].range.lpfn + 1;
 		} else {
 			places[c].fpfn = 0;
 			places[c].lpfn = 0;
@@ -583,8 +584,12 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 
 	bo->flags = bp->flags;
 
-	/* bo->mem_id -1 means any partition */
-	bo->mem_id = bp->mem_id_plus1 - 1;
+	if (adev->gmc.mem_partitions)
+		/* For GPUs with spatial partitioning, bo->xcp_id=-1 means any partition */
+		bo->xcp_id = bp->xcp_id_plus1 - 1;
+	else
+		/* For GPUs without spatial partitioning */
+		bo->xcp_id = 0;
 
 	if (!amdgpu_bo_support_uswc(bo->flags))
 		bo->flags &= ~AMDGPU_GEM_CREATE_CPU_GTT_USWC;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index eb24a66ccee59..05496b97ef930 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -56,8 +56,8 @@ struct amdgpu_bo_param {
 	bool				no_wait_gpu;
 	struct dma_resv			*resv;
 	void				(*destroy)(struct ttm_buffer_object *bo);
-	/* memory partition number plus 1, 0 means any partition */
-	int8_t				mem_id_plus1;
+	/* xcp partition number plus 1, 0 means any partition */
+	int8_t				xcp_id_plus1;
 };
 
 /* bo virtual addresses in a vm */
@@ -111,8 +111,12 @@ struct amdgpu_bo {
 #endif
 	struct kgd_mem                  *kfd_bo;
 
-	/* memory partition number, -1 means any partition */
-	int8_t				mem_id;
+	/*
+	 * For GPUs with spatial partitioning, xcp partition number, -1 means
+	 * any partition. For other ASICs without spatial partition, always 0
+	 * for memory accounting.
+	 */
+	int8_t				xcp_id;
 };
 
 struct amdgpu_bo_user {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 129c593cb2bd1..23101c82519af 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1051,6 +1051,7 @@ static void amdgpu_ttm_backend_destroy(struct ttm_device *bdev,
 static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 					   uint32_t page_flags)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo);
 	struct amdgpu_ttm_tt *gtt;
 	enum ttm_caching caching;
@@ -1060,7 +1061,10 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 		return NULL;
 	}
 	gtt->gobj = &bo->base;
-	gtt->pool_id = abo->mem_id;
+	if (adev->gmc.mem_partitions && abo->xcp_id >= 0)
+		gtt->pool_id = KFD_XCP_MEM_ID(adev, abo->xcp_id);
+	else
+		gtt->pool_id = abo->xcp_id;
 
 	if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC)
 		caching = ttm_write_combined;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index 62fc7e8d326ef..cc3b1b596e561 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -502,6 +502,7 @@ int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			int level, bool immediate, struct amdgpu_bo_vm **vmbo)
 {
+	struct amdgpu_fpriv *fpriv = container_of(vm, struct amdgpu_fpriv, vm);
 	struct amdgpu_bo_param bp;
 	struct amdgpu_bo *bo;
 	struct dma_resv *resv;
@@ -534,7 +535,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	bp.type = ttm_bo_type_kernel;
 	bp.no_wait_gpu = immediate;
-	bp.mem_id_plus1 = vm->mem_id + 1;
+	bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1;
 
 	if (vm->root.bo)
 		bp.resv = vm->root.bo->tbo.base.resv;
@@ -560,7 +561,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	bp.type = ttm_bo_type_kernel;
 	bp.resv = bo->tbo.base.resv;
 	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
-	bp.mem_id_plus1 = vm->mem_id + 1;
+	bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1;
 
 	r = amdgpu_bo_create(adev, &bp, &(*vmbo)->shadow);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 263d17a8b4334..7ea80bdf8e1ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1248,7 +1248,7 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		is_local = (!is_vram && (adev->flags & AMD_IS_APU) &&
 			    num_possible_nodes() <= 1) ||
 			   (is_vram && adev == bo_adev &&
-			    bo->mem_id == vm->mem_id);
+			    KFD_XCP_MEM_ID(adev, bo->xcp_id) == vm->mem_id);
 		snoop = true;
 		if (uncached) {
 			mtype = MTYPE_UC;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index a700d9ccd0548..45959892bc0f0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -556,7 +556,7 @@ svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange,
 	bp.type = ttm_bo_type_device;
 	bp.resv = NULL;
 	if (node->xcp)
-		bp.mem_id_plus1 = node->xcp->mem_id + 1;
+		bp.xcp_id_plus1 = node->xcp->id + 1;
 
 	r = amdgpu_bo_create_user(node->adev, &bp, &ubo);
 	if (r) {
@@ -567,7 +567,7 @@ svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange,
 
 	pr_debug("alloc bo at offset 0x%lx size 0x%lx on partition %d\n",
 		 bo->tbo.resource->start << PAGE_SHIFT, bp.size,
-		 bp.mem_id_plus1 - 1);
+		 bp.xcp_id_plus1 - 1);
 
 	r = amdgpu_bo_reserve(bo, true);
 	if (r) {
-- 
GitLab


From 2fa9ff25de08e598af051c76b216d2f073b2ee89 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 9 Mar 2023 19:30:02 -0500
Subject: [PATCH 1322/1791] drm/amdgpu: KFD graphics interop support compute
 partition

kfd_ioctl_get_dmabuf use the amdgpu bo xcp_id to get the gpu_id of the
KFD node from the exported dmabuf_adev, and then create kfd bo on the
correct adev and KFD node when importing the amdgpu bo to KFD.

Remove function kfd_device_by_adev, it is not needed as it is the same
result as dmabuf_adev->kfd.dev->nodes[0]->id.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c |  4 +++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c   | 14 ++++++--------
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h      |  1 -
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c  | 18 ------------------
 5 files changed, 10 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index bbbfe9ec4adf8..00edb13d21242 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -498,7 +498,7 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd,
 				  struct amdgpu_device **dmabuf_adev,
 				  uint64_t *bo_size, void *metadata_buffer,
 				  size_t buffer_size, uint32_t *metadata_size,
-				  uint32_t *flags)
+				  uint32_t *flags, int8_t *xcp_id)
 {
 	struct dma_buf *dma_buf;
 	struct drm_gem_object *obj;
@@ -542,6 +542,8 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd,
 		if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
 			*flags |= KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC;
 	}
+	if (xcp_id)
+		*xcp_id = bo->xcp_id;
 
 out_put:
 	dma_buf_put(dma_buf);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 05c54776951b0..4e6221bccffe5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -241,7 +241,7 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd,
 				  struct amdgpu_device **dmabuf_adev,
 				  uint64_t *bo_size, void *metadata_buffer,
 				  size_t buffer_size, uint32_t *metadata_size,
-				  uint32_t *flags);
+				  uint32_t *flags, int8_t *xcp_id);
 uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct amdgpu_device *dst,
 					  struct amdgpu_device *src);
 int amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(struct amdgpu_device *dst,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index f85ac4dbc6737..fcad90d53c9be 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1499,6 +1499,7 @@ static int kfd_ioctl_get_dmabuf_info(struct file *filep,
 	struct amdgpu_device *dmabuf_adev;
 	void *metadata_buffer = NULL;
 	uint32_t flags;
+	int8_t xcp_id;
 	unsigned int i;
 	int r;
 
@@ -1519,17 +1520,14 @@ static int kfd_ioctl_get_dmabuf_info(struct file *filep,
 	r = amdgpu_amdkfd_get_dmabuf_info(dev->adev, args->dmabuf_fd,
 					  &dmabuf_adev, &args->size,
 					  metadata_buffer, args->metadata_size,
-					  &args->metadata_size, &flags);
+					  &args->metadata_size, &flags, &xcp_id);
 	if (r)
 		goto exit;
 
-	/* Reverse-lookup gpu_id from kgd pointer */
-	dev = kfd_device_by_adev(dmabuf_adev);
-	if (!dev) {
-		r = -EINVAL;
-		goto exit;
-	}
-	args->gpu_id = dev->id;
+	if (xcp_id >= 0)
+		args->gpu_id = dmabuf_adev->kfd.dev->nodes[xcp_id]->id;
+	else
+		args->gpu_id = dmabuf_adev->kfd.dev->nodes[0]->id;
 	args->flags = flags;
 
 	/* Copy metadata buffer to user mode */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 214d950f948ee..44f4d5509db6f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1068,7 +1068,6 @@ struct kfd_topology_device *kfd_topology_device_by_proximity_domain_no_lock(
 struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_id(uint32_t gpu_id);
 struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev);
-struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev);
 static inline bool kfd_irq_is_from_node(struct kfd_node *node, uint32_t node_id,
 					uint32_t vmid)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index a8e25aecf839b..dbb6159344b35 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -125,24 +125,6 @@ struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev)
 	return device;
 }
 
-struct kfd_node *kfd_device_by_adev(const struct amdgpu_device *adev)
-{
-	struct kfd_topology_device *top_dev;
-	struct kfd_node *device = NULL;
-
-	down_read(&topology_lock);
-
-	list_for_each_entry(top_dev, &topology_device_list, list)
-		if (top_dev->gpu && top_dev->gpu->adev == adev) {
-			device = top_dev->gpu;
-			break;
-		}
-
-	up_read(&topology_lock);
-
-	return device;
-}
-
 /* Called with write topology_lock acquired */
 static void kfd_release_topology_device(struct kfd_topology_device *dev)
 {
-- 
GitLab


From b125b80bd546d72d08ab64d63cfc8efa397b6779 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Mon, 13 Mar 2023 12:03:18 -0400
Subject: [PATCH 1323/1791] drm/amdgpu: use xcp partition ID for amdgpu_gem

Find xcp_id from amdgpu_fpriv, use it for amdgpu_gem_object_create.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 7e8839cc6f58b..fec9df3546525 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -337,7 +337,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 retry:
 	r = amdgpu_gem_object_create(adev, size, args->in.alignment,
 				     initial_domain,
-				     flags, ttm_bo_type_device, resv, &gobj, 0);
+				     flags, ttm_bo_type_device, resv, &gobj, fpriv->xcp_id + 1);
 	if (r && r != -ERESTARTSYS) {
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) {
 			flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
@@ -380,6 +380,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 	struct ttm_operation_ctx ctx = { true, false };
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct drm_amdgpu_gem_userptr *args = data;
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	struct drm_gem_object *gobj;
 	struct hmm_range *range;
 	struct amdgpu_bo *bo;
@@ -406,7 +407,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 
 	/* create a gem object to contain this object in */
 	r = amdgpu_gem_object_create(adev, args->size, 0, AMDGPU_GEM_DOMAIN_CPU,
-				     0, ttm_bo_type_device, NULL, &gobj, 0);
+				     0, ttm_bo_type_device, NULL, &gobj, fpriv->xcp_id + 1);
 	if (r)
 		return r;
 
@@ -909,6 +910,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 			    struct drm_mode_create_dumb *args)
 {
 	struct amdgpu_device *adev = drm_to_adev(dev);
+	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
 	struct drm_gem_object *gobj;
 	uint32_t handle;
 	u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
@@ -932,7 +934,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 	domain = amdgpu_bo_get_preferred_domain(adev,
 				amdgpu_display_supported_domains(adev, flags));
 	r = amdgpu_gem_object_create(adev, args->size, 0, domain, flags,
-				     ttm_bo_type_device, NULL, &gobj, 0);
+				     ttm_bo_type_device, NULL, &gobj, fpriv->xcp_id + 1);
 	if (r)
 		return -ENOMEM;
 
-- 
GitLab


From 315e29eca57f85107cc6f687c2d510aa532fb3f0 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 20 Mar 2023 11:21:38 -0400
Subject: [PATCH 1324/1791] drm/amdkfd: Move local_mem_info to kfd_node

We need to track memory usage on a per partition basis. To do
that, store the local memory information in KFD node instead
of kfd device.

v2: squash in fix ("amdkfd: Use mem_id to access mem_partition info")

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 17 +++++++++++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 12 +++++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c   |  7 ++++---
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c      |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    |  7 +++++--
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h      |  3 ++-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c  |  7 ++++---
 7 files changed, 36 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 00edb13d21242..85df73f2c85ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -428,14 +428,23 @@ uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 }
 
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
-				      struct kfd_local_mem_info *mem_info)
+				      struct kfd_local_mem_info *mem_info,
+				      uint8_t xcp_id)
 {
 	memset(mem_info, 0, sizeof(*mem_info));
 
-	mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
-	mem_info->local_mem_size_private = adev->gmc.real_vram_size -
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		if (adev->gmc.real_vram_size == adev->gmc.visible_vram_size)
+			mem_info->local_mem_size_public =
+					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+		else
+			mem_info->local_mem_size_private =
+					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+	} else {
+		mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
+		mem_info->local_mem_size_private = adev->gmc.real_vram_size -
 						adev->gmc.visible_vram_size;
-
+	}
 	mem_info->vram_width = adev->gmc.vram_width;
 
 	pr_debug("Address base: %pap public 0x%llx private 0x%llx\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 4e6221bccffe5..4bf6f56595688 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -231,7 +231,8 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem);
 uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 				      enum kgd_engine_type type);
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
-				      struct kfd_local_mem_info *mem_info);
+				      struct kfd_local_mem_info *mem_info,
+				      uint8_t xcp_id);
 uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev);
 
 uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev);
@@ -334,10 +335,11 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		((adev)->xcp_mgr && (xcp_id) >= 0 ?\
 		(adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1)
 
-#define KFD_XCP_MEMORY_SIZE(n) ((n)->adev->gmc.num_mem_partitions ?\
-		(n)->adev->gmc.mem_partitions[(n)->xcp->mem_id].size /\
-		(n)->adev->xcp_mgr->num_xcp_per_mem_partition :\
-		(n)->adev->gmc.real_vram_size)
+#define KFD_XCP_MEMORY_SIZE(adev, xcp_id)\
+		((adev)->gmc.num_mem_partitions && (xcp_id) >= 0 ?\
+		(adev)->gmc.mem_partitions[KFD_XCP_MEM_ID((adev), (xcp_id))].size /\
+		(adev)->xcp_mgr->num_xcp_per_mem_partition :\
+		(adev)->gmc.real_vram_size)
 
 #if IS_ENABLED(CONFIG_HSA_AMD)
 void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index fcad90d53c9be..1ae867482bc76 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1023,11 +1023,12 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev)
 	if (dev->kfd->use_iommu_v2)
 		return false;
 
-	if (dev->kfd->local_mem_info.local_mem_size_private == 0 &&
-	    dev->kfd->local_mem_info.local_mem_size_public > 0)
+	if (dev->local_mem_info.local_mem_size_private == 0 &&
+	    dev->local_mem_info.local_mem_size_public > 0)
 		return true;
 
-	if (dev->kfd->local_mem_info.local_mem_size_public == 0 && dev->kfd->adev->gmc.is_app_apu) {
+	if (dev->local_mem_info.local_mem_size_public == 0 &&
+	    dev->kfd->adev->gmc.is_app_apu) {
 		pr_debug("APP APU, Consider like a large bar system\n");
 		return true;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 1aaf933f9f488..950af6820153b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -2191,7 +2191,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 	 * report the total FB size (public+private) as a single
 	 * private heap.
 	 */
-	local_mem_info = kdev->kfd->local_mem_info;
+	local_mem_info = kdev->local_mem_info;
 	sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
 			sub_type_hdr->length);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index db5b53fcdf11a..d41da964d2f5a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -726,7 +726,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	svm_migrate_init(kfd->adev);
 
-	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
 
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
 				kfd->num_nodes);
@@ -756,7 +755,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		if (node->xcp) {
 			dev_info(kfd_device, "KFD node %d partition %d size %lldM\n",
 				node->node_id, node->xcp->mem_id,
-				KFD_XCP_MEMORY_SIZE(node) >> 20);
+				KFD_XCP_MEMORY_SIZE(node->adev, node->node_id) >> 20);
 		}
 
 		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
@@ -783,6 +782,10 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		}
 		node->max_proc_per_quantum = max_proc_per_quantum;
 		atomic_set(&node->sram_ecc_flag, 0);
+
+		amdgpu_amdkfd_get_local_mem_info(kfd->adev,
+					&node->local_mem_info, node->xcp->id);
+
 		/* Initialize the KFD node */
 		if (kfd_init_node(node)) {
 			dev_err(kfd_device, "Error initializing KFD node\n");
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 44f4d5509db6f..3bd222e8f6c3c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -313,6 +313,8 @@ struct kfd_node {
 
 	unsigned int compute_vmid_bitmap;
 
+	struct kfd_local_mem_info local_mem_info;
+
 	struct kfd_dev *kfd;
 };
 
@@ -335,7 +337,6 @@ struct kfd_dev {
 					   */
 
 	struct kgd2kfd_shared_resources shared_resources;
-	struct kfd_local_mem_info local_mem_info;
 
 	const struct kfd2kgd_calls *kfd2kgd;
 	struct mutex doorbell_mutex;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index dbb6159344b35..e0bacf017a405 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1152,8 +1152,8 @@ static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 	if (!gpu)
 		return 0;
 
-	local_mem_size = gpu->kfd->local_mem_info.local_mem_size_private +
-			gpu->kfd->local_mem_info.local_mem_size_public;
+	local_mem_size = gpu->local_mem_info.local_mem_size_private +
+			gpu->local_mem_info.local_mem_size_public;
 	buf[0] = gpu->adev->pdev->devfn;
 	buf[1] = gpu->adev->pdev->subsystem_vendor |
 		(gpu->adev->pdev->subsystem_device << 16);
@@ -1234,7 +1234,8 @@ static void kfd_fill_mem_clk_max_info(struct kfd_topology_device *dev)
 	 * for APUs - If CRAT from ACPI reports more than one bank, then
 	 *	all the banks will report the same mem_clk_max information
 	 */
-	amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info);
+	amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info,
+					 dev->gpu->xcp->id);
 
 	list_for_each_entry(mem, &dev->mem_props, list)
 		mem->mem_clk_max = local_mem_info.mem_clk_max;
-- 
GitLab


From 1c77527a69d5ca19cb276e2728992d922b687f35 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 20 Mar 2023 11:22:30 -0400
Subject: [PATCH 1325/1791] drm/amdkfd: Fix memory reporting on GFX 9.4.3

This patch fixes memory reporting on the GFX 9.4.3 APU and dGPU
by reporting available memory on a per partition basis. If its an
APU, available and used memory calculations take into account
system and TTM memory.

v2: squash in fix ("drm/amdkfd: Fix array out of bound warning")
    squash in fix ("drm/amdgpu: Update memory reporting for GFX9.4.3")

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h    | 12 +--
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 81 ++++++++++++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h       |  5 ++
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  3 +-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c          | 14 ++--
 5 files changed, 84 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 4bf6f56595688..948d362adabbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -35,6 +35,7 @@
 #include <drm/ttm/ttm_execbuf_util.h>
 #include "amdgpu_sync.h"
 #include "amdgpu_vm.h"
+#include "amdgpu_xcp.h"
 
 extern uint64_t amdgpu_amdkfd_total_mem_size;
 
@@ -98,8 +99,8 @@ struct amdgpu_amdkfd_fence {
 
 struct amdgpu_kfd_dev {
 	struct kfd_dev *dev;
-	int64_t vram_used;
-	uint64_t vram_used_aligned;
+	int64_t vram_used[MAX_XCP];
+	uint64_t vram_used_aligned[MAX_XCP];
 	bool init_complete;
 	struct work_struct reset_work;
 
@@ -287,7 +288,8 @@ int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev,
 void amdgpu_amdkfd_gpuvm_release_process_vm(struct amdgpu_device *adev,
 					void *drm_priv);
 uint64_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *drm_priv);
-size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev);
+size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev,
+					uint8_t xcp_id);
 int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		struct amdgpu_device *adev, uint64_t va, uint64_t size,
 		void *drm_priv, struct kgd_mem **mem,
@@ -327,9 +329,9 @@ void amdgpu_amdkfd_block_mmu_notifications(void *p);
 int amdgpu_amdkfd_criu_resume(void *p);
 bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev);
 int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
-		uint64_t size, u32 alloc_flag);
+		uint64_t size, u32 alloc_flag, int8_t xcp_id);
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
-		uint64_t size, u32 alloc_flag);
+		uint64_t size, u32 alloc_flag, int8_t xcp_id);
 
 #define KFD_XCP_MEM_ID(adev, xcp_id) \
 		((adev)->xcp_mgr && (xcp_id) >= 0 ?\
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 8724a0be31b8f..cc37f04651e29 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -157,12 +157,13 @@ void amdgpu_amdkfd_reserve_system_mem(uint64_t size)
  * Return: returns -ENOMEM in case of error, ZERO otherwise
  */
 int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
-		uint64_t size, u32 alloc_flag)
+		uint64_t size, u32 alloc_flag, int8_t xcp_id)
 {
 	uint64_t reserved_for_pt =
 		ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size);
 	size_t system_mem_needed, ttm_mem_needed, vram_needed;
 	int ret = 0;
+	uint64_t vram_size = 0;
 
 	system_mem_needed = 0;
 	ttm_mem_needed = 0;
@@ -177,6 +178,17 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 		 * 2M BO chunk.
 		 */
 		vram_needed = size;
+		/*
+		 * For GFX 9.4.3, get the VRAM size from XCP structs
+		 */
+		if (WARN_ONCE(xcp_id < 0, "invalid XCP ID %d", xcp_id))
+			return -EINVAL;
+
+		vram_size = KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+		if (adev->gmc.is_app_apu) {
+			system_mem_needed = size;
+			ttm_mem_needed = size;
+		}
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
 		system_mem_needed = size;
 	} else if (!(alloc_flag &
@@ -196,8 +208,8 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	     kfd_mem_limit.max_system_mem_limit && !no_system_mem_limit) ||
 	    (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
 	     kfd_mem_limit.max_ttm_mem_limit) ||
-	    (adev && adev->kfd.vram_used + vram_needed >
-	     adev->gmc.real_vram_size - reserved_for_pt)) {
+	    (adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] + vram_needed >
+	     vram_size - reserved_for_pt)) {
 		ret = -ENOMEM;
 		goto release;
 	}
@@ -207,9 +219,11 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	 */
 	WARN_ONCE(vram_needed && !adev,
 		  "adev reference can't be null when vram is used");
-	if (adev) {
-		adev->kfd.vram_used += vram_needed;
-		adev->kfd.vram_used_aligned += ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN);
+	if (adev && xcp_id >= 0) {
+		adev->kfd.vram_used[xcp_id] += vram_needed;
+		adev->kfd.vram_used_aligned[xcp_id] += adev->gmc.is_app_apu ?
+				vram_needed :
+				ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN);
 	}
 	kfd_mem_limit.system_mem_used += system_mem_needed;
 	kfd_mem_limit.ttm_mem_used += ttm_mem_needed;
@@ -220,7 +234,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 }
 
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
-		uint64_t size, u32 alloc_flag)
+		uint64_t size, u32 alloc_flag, int8_t xcp_id)
 {
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
 
@@ -230,9 +244,19 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) {
 		WARN_ONCE(!adev,
 			  "adev reference can't be null when alloc mem flags vram is set");
+		if (WARN_ONCE(xcp_id < 0, "invalid XCP ID %d", xcp_id))
+			goto release;
+
 		if (adev) {
-			adev->kfd.vram_used -= size;
-			adev->kfd.vram_used_aligned -= ALIGN(size, VRAM_AVAILABLITY_ALIGN);
+			adev->kfd.vram_used[xcp_id] -= size;
+			if (adev->gmc.is_app_apu) {
+				adev->kfd.vram_used_aligned[xcp_id] -= size;
+				kfd_mem_limit.system_mem_used -= size;
+				kfd_mem_limit.ttm_mem_used -= size;
+			} else {
+				adev->kfd.vram_used_aligned[xcp_id] -=
+					ALIGN(size, VRAM_AVAILABLITY_ALIGN);
+			}
 		}
 	} else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) {
 		kfd_mem_limit.system_mem_used -= size;
@@ -242,8 +266,8 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		pr_err("%s: Invalid BO type %#x\n", __func__, alloc_flag);
 		goto release;
 	}
-	WARN_ONCE(adev && adev->kfd.vram_used < 0,
-		  "KFD VRAM memory accounting unbalanced");
+	WARN_ONCE(adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] < 0,
+		  "KFD VRAM memory accounting unbalanced for xcp: %d", xcp_id);
 	WARN_ONCE(kfd_mem_limit.ttm_mem_used < 0,
 		  "KFD TTM memory accounting unbalanced");
 	WARN_ONCE(kfd_mem_limit.system_mem_used < 0,
@@ -259,7 +283,8 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo)
 	u32 alloc_flags = bo->kfd_bo->alloc_flags;
 	u64 size = amdgpu_bo_size(bo);
 
-	amdgpu_amdkfd_unreserve_mem_limit(adev, size, alloc_flags);
+	amdgpu_amdkfd_unreserve_mem_limit(adev, size, alloc_flags,
+					  bo->xcp_id);
 
 	kfree(bo->kfd_bo);
 }
@@ -1609,23 +1634,42 @@ int amdgpu_amdkfd_criu_resume(void *p)
 	return ret;
 }
 
-size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev)
+size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev,
+					  uint8_t xcp_id)
 {
 	uint64_t reserved_for_pt =
 		ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size);
 	ssize_t available;
+	uint64_t vram_available, system_mem_available, ttm_mem_available;
 
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
-	available = adev->gmc.real_vram_size
-		- adev->kfd.vram_used_aligned
+	vram_available = KFD_XCP_MEMORY_SIZE(adev, xcp_id)
+		- adev->kfd.vram_used_aligned[xcp_id]
 		- atomic64_read(&adev->vram_pin_size)
 		- reserved_for_pt;
+
+	if (adev->gmc.is_app_apu) {
+		system_mem_available = no_system_mem_limit ?
+					kfd_mem_limit.max_system_mem_limit :
+					kfd_mem_limit.max_system_mem_limit -
+					kfd_mem_limit.system_mem_used;
+
+		ttm_mem_available = kfd_mem_limit.max_ttm_mem_limit -
+				kfd_mem_limit.ttm_mem_used;
+
+		available = min3(system_mem_available, ttm_mem_available,
+				 vram_available);
+		available = ALIGN_DOWN(available, PAGE_SIZE);
+	} else {
+		available = ALIGN_DOWN(vram_available, VRAM_AVAILABLITY_ALIGN);
+	}
+
 	spin_unlock(&kfd_mem_limit.mem_limit_lock);
 
 	if (available < 0)
 		available = 0;
 
-	return ALIGN_DOWN(available, VRAM_AVAILABLITY_ALIGN);
+	return available;
 }
 
 int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
@@ -1713,7 +1757,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 
 	amdgpu_sync_create(&(*mem)->sync);
 
-	ret = amdgpu_amdkfd_reserve_mem_limit(adev, aligned_size, flags);
+	ret = amdgpu_amdkfd_reserve_mem_limit(adev, aligned_size, flags,
+					      xcp_id);
 	if (ret) {
 		pr_debug("Insufficient memory\n");
 		goto err_reserve_limit;
@@ -1781,7 +1826,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	/* Don't unreserve system mem limit twice */
 	goto err_reserve_limit;
 err_bo_create:
-	amdgpu_amdkfd_unreserve_mem_limit(adev, aligned_size, flags);
+	amdgpu_amdkfd_unreserve_mem_limit(adev, aligned_size, flags, xcp_id);
 err_reserve_limit:
 	mutex_destroy(&(*mem)->lock);
 	if (gobj)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 68b63b970ce8e..9c5912b9d8bd0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -24,8 +24,11 @@
 #ifndef AMDGPU_XCP_H
 #define AMDGPU_XCP_H
 
+#include <linux/pci.h>
 #include <linux/xarray.h>
 
+#include "amdgpu_ctx.h"
+
 #define MAX_XCP 8
 
 #define AMDGPU_XCP_MODE_NONE -1
@@ -34,6 +37,8 @@
 #define AMDGPU_XCP_FL_NONE 0
 #define AMDGPU_XCP_FL_LOCKED (1 << 0)
 
+struct amdgpu_fpriv;
+
 enum AMDGPU_XCP_IP_BLOCK {
 	AMDGPU_XCP_GFXHUB,
 	AMDGPU_XCP_GFX,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 1ae867482bc76..a9efff94390bf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1044,7 +1044,8 @@ static int kfd_ioctl_get_available_memory(struct file *filep,
 
 	if (!pdd)
 		return -EINVAL;
-	args->available = amdgpu_amdkfd_get_available_memory(pdd->dev->adev);
+	args->available = amdgpu_amdkfd_get_available_memory(pdd->dev->adev,
+							pdd->dev->node_id);
 	kfd_unlock_pdd(pdd);
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 45959892bc0f0..c1ab70faf36e5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -280,7 +280,7 @@ static void svm_range_free(struct svm_range *prange, bool update_mem_usage)
 	if (update_mem_usage && !p->xnack_enabled) {
 		pr_debug("unreserve prange 0x%p size: 0x%llx\n", prange, size);
 		amdgpu_amdkfd_unreserve_mem_limit(NULL, size,
-					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 	}
 	mutex_destroy(&prange->lock);
 	mutex_destroy(&prange->migrate_mutex);
@@ -313,7 +313,7 @@ svm_range *svm_range_new(struct svm_range_list *svms, uint64_t start,
 	p = container_of(svms, struct kfd_process, svms);
 	if (!p->xnack_enabled && update_mem_usage &&
 	    amdgpu_amdkfd_reserve_mem_limit(NULL, size << PAGE_SHIFT,
-					    KFD_IOC_ALLOC_MEM_FLAGS_USERPTR)) {
+				    KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0)) {
 		pr_info("SVM mapping failed, exceeds resident system memory limit\n");
 		kfree(prange);
 		return NULL;
@@ -3037,10 +3037,10 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled)
 			size = (pchild->last - pchild->start + 1) << PAGE_SHIFT;
 			if (xnack_enabled) {
 				amdgpu_amdkfd_unreserve_mem_limit(NULL, size,
-						KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 			} else {
 				r = amdgpu_amdkfd_reserve_mem_limit(NULL, size,
-						KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 				if (r)
 					goto out_unlock;
 				reserved_size += size;
@@ -3050,10 +3050,10 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled)
 		size = (prange->last - prange->start + 1) << PAGE_SHIFT;
 		if (xnack_enabled) {
 			amdgpu_amdkfd_unreserve_mem_limit(NULL, size,
-						KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 		} else {
 			r = amdgpu_amdkfd_reserve_mem_limit(NULL, size,
-						KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 			if (r)
 				goto out_unlock;
 			reserved_size += size;
@@ -3066,7 +3066,7 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled)
 
 	if (r)
 		amdgpu_amdkfd_unreserve_mem_limit(NULL, reserved_size,
-						KFD_IOC_ALLOC_MEM_FLAGS_USERPTR);
+					KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0);
 	else
 		/* Change xnack mode must be inside svms lock, to avoid race with
 		 * svm_range_deferred_list_work unreserve memory in parallel.
-- 
GitLab


From 25f50704343de1bea70100ad41621b5737a6a96b Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 23 Mar 2023 08:45:56 -0400
Subject: [PATCH 1326/1791] drm/amdkfd: APU mode set max svm range pages

svm_migrate_init set the max svm range pages based on the KFD nodes
partition size. APU mode don't init pgmap because there is no migration.

kgd2kfd_device_init calls svm_migrate_init after KFD nodes allocation
and initialization.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c  |  5 ++---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  7 +++++--
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c     | 15 ++++++++++-----
 3 files changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index d41da964d2f5a..882ff86bba082 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -724,9 +724,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd_cwsr_init(kfd);
 
-	svm_migrate_init(kfd->adev);
-
-
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
 				kfd->num_nodes);
 
@@ -794,6 +791,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		kfd->nodes[i] = node;
 	}
 
+	svm_migrate_init(kfd->adev);
+
 	if (kfd_resume_iommu(kfd))
 		goto kfd_resume_iommu_error;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 199d32c7c289f..2512bf6811124 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -1000,6 +1000,11 @@ int svm_migrate_init(struct amdgpu_device *adev)
 	if (!KFD_IS_SOC15(kfddev->dev))
 		return -EINVAL;
 
+	svm_range_set_max_pages(adev);
+
+	if (adev->gmc.is_app_apu)
+		return 0;
+
 	pgmap = &kfddev->pgmap;
 	memset(pgmap, 0, sizeof(*pgmap));
 
@@ -1042,8 +1047,6 @@ int svm_migrate_init(struct amdgpu_device *adev)
 
 	amdgpu_amdkfd_reserve_system_mem(SVM_HMM_PAGE_STRUCT_SIZE(size));
 
-	svm_range_set_max_pages(adev);
-
 	pr_info("HMM registered %ldMB device memory\n", size >> 20);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index c1ab70faf36e5..206851c9e6424 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1938,14 +1938,19 @@ void svm_range_set_max_pages(struct amdgpu_device *adev)
 {
 	uint64_t max_pages;
 	uint64_t pages, _pages;
+	uint64_t min_pages = 0;
+	int i;
+
+	for (i = 0; i < adev->kfd.dev->num_nodes; i++) {
+		pages = KFD_XCP_MEMORY_SIZE(adev, adev->kfd.dev->nodes[i]->xcp->id) >> 17;
+		pages = clamp(pages, 1ULL << 9, 1ULL << 18);
+		pages = rounddown_pow_of_two(pages);
+		min_pages = min_not_zero(min_pages, pages);
+	}
 
-	/* 1/32 VRAM size in pages */
-	pages = adev->gmc.real_vram_size >> 17;
-	pages = clamp(pages, 1ULL << 9, 1ULL << 18);
-	pages = rounddown_pow_of_two(pages);
 	do {
 		max_pages = READ_ONCE(max_svm_range_pages);
-		_pages = min_not_zero(max_pages, pages);
+		_pages = min_not_zero(max_pages, min_pages);
 	} while (cmpxchg(&max_svm_range_pages, max_pages, _pages) != max_pages);
 }
 
-- 
GitLab


From 44a976655597b60bd501148abe66462bdc33fb6b Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Fri, 31 Mar 2023 17:16:41 +0800
Subject: [PATCH 1327/1791] drm/amdgpu: route ioctls on primary node of XCPs to
 primary device

During XCP init, unlike the primary device, there is no amdgpu_device
attached to each XCP's drm_device

In case that user trying to open/close the primary node of XCP drm_device
this rerouting is to solve the NULL pointer issue causing by referring
to any member of the amdgpu_device

 BUG: unable to handle page fault for address: 0000000000020c80
 #PF: supervisor write access in kernel mode
 #PF: error_code(0x0002) - not-present page
 Oops: 0002 [#1] PREEMPT SMP NOPTI
 Call Trace:
  <TASK>
  lock_timer_base+0x6b/0x90
  try_to_del_timer_sync+0x2b/0x80
  del_timer_sync+0x29/0x40
  flush_delayed_work+0x1c/0x50
  amdgpu_driver_open_kms+0x2c/0x280 [amdgpu]
  drm_file_alloc+0x1b3/0x260 [drm]
  drm_open+0xaa/0x280 [drm]
  drm_stub_open+0xa2/0x120 [drm]
  chrdev_open+0xa6/0x1c0

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 610c32c4f5afb..daeb6bcc9245a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -241,6 +241,7 @@ static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev)
 
 		/* Redirect all IOCTLs to the primary device */
 		p_ddev->render->dev = ddev;
+		p_ddev->primary->dev = ddev;
 		p_ddev->vma_offset_manager = ddev->vma_offset_manager;
 		adev->xcp_mgr->xcp[i].ddev = p_ddev;
 	}
-- 
GitLab


From 84b4dd3f84de424a68e1fda0d483530ddaa92b45 Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Fri, 31 Mar 2023 11:18:12 -0400
Subject: [PATCH 1328/1791] drm/amdkfd: Refactor migrate init to support
 partition switch

Rename smv_migrate_init to a better name kgd2kfd_init_zone_device
because it setup zone devive pgmap for page migration and keep it in
kfd_migrate.c to access static functions svm_migrate_pgmap_ops. Call it
only once in amdgpu_device_ip_init after adev ip blocks are initialized,
but before amdgpu_amdkfd_device_init initialize kfd nodes which enable
SVM support based on pgmap.

svm_range_set_max_pages is called by kgd2kfd_device_init everytime after
switching compute partition mode.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |  4 +++-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    |  3 ++-
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c   |  8 +++-----
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.h   |  9 ---------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h       |  4 ++++
 6 files changed, 23 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 948d362adabbc..48d12dbff9681 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -372,6 +372,17 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo)
 {
 }
 #endif
+
+#if IS_ENABLED(CONFIG_HSA_AMD_SVM)
+int kgd2kfd_init_zone_device(struct amdgpu_device *adev);
+#else
+static inline
+int kgd2kfd_init_zone_device(struct amdgpu_device *adev)
+{
+	return 0;
+}
+#endif
+
 /* KGD2KFD callbacks */
 int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger);
 int kgd2kfd_resume_mm(struct mm_struct *mm);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 02ee79b7b56d1..f0666230b2ed0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2633,8 +2633,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 		goto init_failed;
 
 	/* Don't init kfd if whole hive need to be reset during init */
-	if (!adev->gmc.xgmi.pending_reset)
+	if (!adev->gmc.xgmi.pending_reset) {
+		kgd2kfd_init_zone_device(adev);
 		amdgpu_amdkfd_device_init(adev);
+	}
 
 	amdgpu_fru_get_product_info(adev);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 882ff86bba082..bf32e547182c9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -32,6 +32,7 @@
 #include "kfd_iommu.h"
 #include "amdgpu_amdkfd.h"
 #include "kfd_smi_events.h"
+#include "kfd_svm.h"
 #include "kfd_migrate.h"
 #include "amdgpu.h"
 #include "amdgpu_xcp.h"
@@ -791,7 +792,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		kfd->nodes[i] = node;
 	}
 
-	svm_migrate_init(kfd->adev);
+	svm_range_set_max_pages(kfd->adev);
 
 	if (kfd_resume_iommu(kfd))
 		goto kfd_resume_iommu_error;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 2512bf6811124..35cf6558cf1b4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -988,7 +988,7 @@ static const struct dev_pagemap_ops svm_migrate_pgmap_ops = {
 /* Each VRAM page uses sizeof(struct page) on system memory */
 #define SVM_HMM_PAGE_STRUCT_SIZE(size) ((size)/PAGE_SIZE * sizeof(struct page))
 
-int svm_migrate_init(struct amdgpu_device *adev)
+int kgd2kfd_init_zone_device(struct amdgpu_device *adev)
 {
 	struct amdgpu_kfd_dev *kfddev = &adev->kfd;
 	struct dev_pagemap *pgmap;
@@ -996,12 +996,10 @@ int svm_migrate_init(struct amdgpu_device *adev)
 	unsigned long size;
 	void *r;
 
-	/* Page migration works on Vega10 or newer */
-	if (!KFD_IS_SOC15(kfddev->dev))
+	/* Page migration works on gfx9 or newer */
+	if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 0, 1))
 		return -EINVAL;
 
-	svm_range_set_max_pages(adev);
-
 	if (adev->gmc.is_app_apu)
 		return 0;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h
index a5d7e6d222646..487f263681648 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h
@@ -47,15 +47,6 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 unsigned long
 svm_migrate_addr_to_pfn(struct amdgpu_device *adev, unsigned long addr);
 
-int svm_migrate_init(struct amdgpu_device *adev);
-
-#else
-
-static inline int svm_migrate_init(struct amdgpu_device *adev)
-{
-	return 0;
-}
-
 #endif /* IS_ENABLED(CONFIG_HSA_AMD_SVM) */
 
 #endif /* KFD_MIGRATE_H_ */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index 021def496f5ac..762679835e31b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -265,6 +265,10 @@ static inline int kfd_criu_resume_svm(struct kfd_process *p)
 	return 0;
 }
 
+static inline void svm_range_set_max_pages(struct amdgpu_device *adev)
+{
+}
+
 #define KFD_IS_SVM_API_SUPPORTED(dev) false
 
 #endif /* IS_ENABLED(CONFIG_HSA_AMD_SVM) */
-- 
GitLab


From 3cde91172d2e9d8d8dc6e0d62b7c829de503825c Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Wed, 19 Apr 2023 17:39:35 -0400
Subject: [PATCH 1329/1791] drm/amdgpu: Correct get_xcp_mem_id calculation

Current calculation only works for NPS4/QPX mode, correct it for
NPS4/CPX mode.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 4ca932a62ce67..93e9f947a85d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -518,10 +518,9 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 static int __aqua_vanjaram_get_xcp_mem_id(struct amdgpu_device *adev,
 					  int xcc_id, uint8_t *mem_id)
 {
-	/* TODO: Check if any validation is required based on current
-	 * memory/spatial modes
-	 */
+	/* memory/spatial modes validation check is already done */
 	*mem_id = xcc_id / adev->gfx.num_xcc_per_xcp;
+	*mem_id /= adev->xcp_mgr->num_xcp_per_mem_partition;
 
 	return 0;
 }
-- 
GitLab


From 40b832aac03249ebc70479da9f3ecf2789deaeed Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 12 Apr 2023 16:56:29 -0400
Subject: [PATCH 1330/1791] drm/amdgpu: Enable IH CAM on GFX9.4.3

This patch enables IH CAM on GFX9.4.3 ASIC.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vega20_ih.c | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index e1552d6453083..755259e96bbca 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -265,7 +265,7 @@ static void nbio_v7_9_ih_doorbell_range(struct amdgpu_device *adev,
 		ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range,
 				DOORBELL0_CTRL_ENTRY_0,
 				BIF_DOORBELL0_RANGE_SIZE_ENTRY,
-				0x4);
+				0x8);
 
 		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
@@ -278,7 +278,7 @@ static void nbio_v7_9_ih_doorbell_range(struct amdgpu_device *adev,
 				S2A_DOORBELL_PORT1_RANGE_OFFSET, 0);
 		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
-				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x4);
+				S2A_DOORBELL_PORT1_RANGE_SIZE, 0x8);
 		ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl,
 				S2A_DOORBELL_ENTRY_1_CTRL,
 				S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
index 17ccf02462abe..4d719df376a72 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
@@ -334,7 +334,8 @@ static int vega20_ih_irq_init(struct amdgpu_device *adev)
 		vega20_setup_retry_doorbell(adev->irq.retry_cam_doorbell_index));
 
 	/* Enable IH Retry CAM */
-	if (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 0))
+	if (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 0) ||
+	    adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2))
 		WREG32_FIELD15(OSSSYS, 0, IH_RETRY_INT_CAM_CNTL_ALDEBARAN,
 			       ENABLE, 1);
 	else
-- 
GitLab


From af2ba368838ee4913e758f34e3d8bbfeb110be36 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 27 Feb 2023 16:31:56 +0800
Subject: [PATCH 1331/1791] drm/amdgpu: convert logical instance mask to
 physical one

Convert instance mask for the convenience of RAS TA.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h            |  6 ++++--
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c    | 18 ++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/soc15_common.h      |  7 ++++++-
 3 files changed, 28 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index bba8cfeff71f0..a4d0bc80ac92a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -677,12 +677,14 @@ enum amd_hw_ip_block_type {
 #define IP_VERSION_REV(ver) ((ver) & 0xFF)
 
 struct amdgpu_ip_map_info {
-	/* Map of logical to actual dev instances */
+	/* Map of logical to actual dev instances/mask */
 	uint32_t 		dev_inst[MAX_HWIP][HWIP_MAX_INSTANCE];
 	int8_t (*logical_to_dev_inst)(struct amdgpu_device *adev,
 				      enum amd_hw_ip_block_type block,
 				      int8_t inst);
-
+	uint32_t (*logical_to_dev_mask)(struct amdgpu_device *adev,
+					enum amd_hw_ip_block_type block,
+					uint32_t mask);
 };
 
 struct amd_powerplay {
diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 93e9f947a85d5..68d1a0fc5f5d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -229,6 +229,23 @@ static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev,
 	return dev_inst;
 }
 
+static uint32_t aqua_vanjaram_logical_to_dev_mask(struct amdgpu_device *adev,
+					 enum amd_hw_ip_block_type block,
+					 uint32_t mask)
+{
+	uint32_t dev_mask = 0;
+	int8_t log_inst, dev_inst;
+
+	while (mask) {
+		log_inst = ffs(mask) - 1;
+		dev_inst = aqua_vanjaram_logical_to_dev_inst(adev, block, log_inst);
+		dev_mask |= (1 << dev_inst);
+		mask &= ~(1 << log_inst);
+	}
+
+	return dev_mask;
+}
+
 static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev,
 					  enum amd_hw_ip_block_type ip_block,
 					  uint32_t inst_mask)
@@ -257,6 +274,7 @@ void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev)
 		aqua_vanjaram_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]);
 
 	adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst;
+	adev->ip_map.logical_to_dev_mask = aqua_vanjaram_logical_to_dev_mask;
 }
 
 /* Fixed pattern for smn addressing on different AIDs:
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index 3730c5ec202fc..96948a59f8dd5 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -25,7 +25,12 @@
 #define __SOC15_COMMON_H__
 
 /* GET_INST returns the physical instance corresponding to a logical instance */
-#define GET_INST(ip, inst) (adev->ip_map.logical_to_dev_inst? adev->ip_map.logical_to_dev_inst(adev, ip##_HWIP, inst): inst)
+#define GET_INST(ip, inst) \
+	(adev->ip_map.logical_to_dev_inst ? \
+	adev->ip_map.logical_to_dev_inst(adev, ip##_HWIP, inst) : inst)
+#define GET_MASK(ip, mask) \
+	(adev->ip_map.logical_to_dev_mask ? \
+	adev->ip_map.logical_to_dev_mask(adev, ip##_HWIP, mask) : mask)
 
 /* Register Access Macros */
 #define SOC15_REG_OFFSET(ip, inst, reg)	(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg)
-- 
GitLab


From 2c22ed0bdb0cb6da9408593eafa6137325576017 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 27 Feb 2023 18:25:23 +0800
Subject: [PATCH 1332/1791] drm/amdgpu: add instance mask for RAS inject

User can specify injected instances by the mask. For backward
compatibility, the mask value is incorporated into sub block index
without interface change of RAS TA.
User uses logical mask and driver should convert it to physical value
before sending it to RAS TA.

v2: update parameter name.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c  | 21 ++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h  |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c  | 23 ++++++++++++++++-------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h  |  9 ++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c |  5 +++--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c    |  6 +++---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c    |  4 ++--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c  |  5 +++--
 8 files changed, 56 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index ec79a5c2f500c..59b8b26e2caf2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1672,14 +1672,33 @@ int psp_ras_initialize(struct psp_context *psp)
 }
 
 int psp_ras_trigger_error(struct psp_context *psp,
-			  struct ta_ras_trigger_error_input *info)
+			  struct ta_ras_trigger_error_input *info, uint32_t instance_mask)
 {
 	struct ta_ras_shared_memory *ras_cmd;
+	struct amdgpu_device *adev = psp->adev;
 	int ret;
+	uint32_t dev_mask;
 
 	if (!psp->ras_context.context.initialized)
 		return -EINVAL;
 
+	switch (info->block_id) {
+	case TA_RAS_BLOCK__GFX:
+		dev_mask = GET_MASK(GC, instance_mask);
+		break;
+	case TA_RAS_BLOCK__SDMA:
+		dev_mask = GET_MASK(SDMA0, instance_mask);
+		break;
+	default:
+		dev_mask = instance_mask;
+		break;
+	}
+
+	/* reuse sub_block_index for backward compatibility */
+	dev_mask <<= AMDGPU_RAS_INST_SHIFT;
+	dev_mask &= AMDGPU_RAS_INST_MASK;
+	info->sub_block_index |= dev_mask;
+
 	ras_cmd = (struct ta_ras_shared_memory *)psp->ras_context.context.mem_context.shared_buf;
 	memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory));
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 0a409da749d10..d84323923a3fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -486,7 +486,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
 int psp_ras_enable_features(struct psp_context *psp,
 		union ta_ras_cmd_input *info, bool enable);
 int psp_ras_trigger_error(struct psp_context *psp,
-			  struct ta_ras_trigger_error_input *info);
+			  struct ta_ras_trigger_error_input *info, uint32_t instance_mask);
 int psp_ras_terminate(struct psp_context *psp);
 
 int psp_hdcp_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 64f80e8cbd638..7ae08f168f990 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -256,6 +256,8 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 	int block_id;
 	uint32_t sub_block;
 	u64 address, value;
+	/* default value is 0 if the mask is not set by user */
+	u32 instance_mask = 0;
 
 	if (*pos)
 		return -EINVAL;
@@ -306,7 +308,11 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 		data->op = op;
 
 		if (op == 2) {
-			if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx",
+			if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx 0x%x",
+				   &sub_block, &address, &value, &instance_mask) != 4 &&
+			    sscanf(str, "%*s %*s %*s %u %llu %llu %u",
+				   &sub_block, &address, &value, &instance_mask) != 4 &&
+				sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx",
 				   &sub_block, &address, &value) != 3 &&
 			    sscanf(str, "%*s %*s %*s %u %llu %llu",
 				   &sub_block, &address, &value) != 3)
@@ -314,6 +320,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 			data->head.sub_block_index = sub_block;
 			data->inject.address = address;
 			data->inject.value = value;
+			data->inject.instance_mask = instance_mask;
 		}
 	} else {
 		if (size < sizeof(*data))
@@ -341,7 +348,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
  * sub_block_index: some IPs have subcomponets. say, GFX, sDMA.
  * name: the name of IP.
  *
- * inject has two more members than head, they are address, value.
+ * inject has three more members than head, they are address, value and mask.
  * As their names indicate, inject operation will write the
  * value to the address.
  *
@@ -365,7 +372,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
  *
  *	echo "disable <block>" > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
  *	echo "enable  <block> <error>" > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
- *	echo "inject  <block> <error> <sub-block> <address> <value> > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
+ *	echo "inject  <block> <error> <sub-block> <address> <value> <mask>" > /sys/kernel/debug/dri/<N>/ras/ras_ctrl
  *
  * Where N, is the card which you want to affect.
  *
@@ -382,13 +389,14 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
  *
  * The sub-block is a the sub-block index, pass 0 if there is no sub-block.
  * The address and value are hexadecimal numbers, leading 0x is optional.
+ * The mask means instance mask, is optional, default value is 0x1.
  *
  * For instance,
  *
  * .. code-block:: bash
  *
  *	echo inject umc ue 0x0 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
- *	echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl
+ *	echo inject umc ce 0 0 0 3 > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *	echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl
  *
  * How to check the result of the operation?
@@ -1117,13 +1125,14 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev,
 
 	if (info->head.block == AMDGPU_RAS_BLOCK__GFX) {
 		if (block_obj->hw_ops->ras_error_inject)
-			ret = block_obj->hw_ops->ras_error_inject(adev, info);
+			ret = block_obj->hw_ops->ras_error_inject(adev, info, info->instance_mask);
 	} else {
 		/* If defined special ras_error_inject(e.g: xgmi), implement special ras_error_inject */
 		if (block_obj->hw_ops->ras_error_inject)
-			ret = block_obj->hw_ops->ras_error_inject(adev, &block_info);
+			ret = block_obj->hw_ops->ras_error_inject(adev, &block_info,
+						info->instance_mask);
 		else  /*If not defined .ras_error_inject, use default ras_error_inject*/
-			ret = psp_ras_trigger_error(&adev->psp, &block_info);
+			ret = psp_ras_trigger_error(&adev->psp, &block_info, info->instance_mask);
 	}
 
 	if (ret)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index e96333d0c2694..bc43f7db17ccf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -32,6 +32,11 @@
 struct amdgpu_iv_entry;
 
 #define AMDGPU_RAS_FLAG_INIT_BY_VBIOS		(0x1 << 0)
+/* position of instance value in sub_block_index of
+ * ta_ras_trigger_error_input, the sub block uses lower 12 bits
+ */
+#define AMDGPU_RAS_INST_MASK 0xfffff000
+#define AMDGPU_RAS_INST_SHIFT 0xc
 
 enum amdgpu_ras_block {
 	AMDGPU_RAS_BLOCK__UMC = 0,
@@ -508,6 +513,7 @@ struct ras_inject_if {
 	struct ras_common_if head;
 	uint64_t address;
 	uint64_t value;
+	uint32_t instance_mask;
 };
 
 struct ras_cure_if {
@@ -545,7 +551,8 @@ struct amdgpu_ras_block_object {
 };
 
 struct amdgpu_ras_block_hw_ops {
-	int  (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if);
+	int  (*ras_error_inject)(struct amdgpu_device *adev,
+			void *inject_if, uint32_t instance_mask);
 	void (*query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status);
 	void (*query_ras_error_status)(struct amdgpu_device *adev);
 	void (*query_ras_error_address)(struct amdgpu_device *adev, void *ras_error_status);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 439925477fb89..85ee1af963dde 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -1014,7 +1014,8 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
 }
 
 /* Trigger XGMI/WAFL error */
-static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,  void *inject_if)
+static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,
+			void *inject_if, uint32_t instance_mask)
 {
 	int ret = 0;
 	struct ta_ras_trigger_error_input *block_info =
@@ -1026,7 +1027,7 @@ static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,  void *injec
 	if (amdgpu_dpm_allow_xgmi_power_down(adev, false))
 		dev_warn(adev->dev, "Failed to disallow XGMI power down");
 
-	ret = psp_ras_trigger_error(&adev->psp, block_info);
+	ret = psp_ras_trigger_error(&adev->psp, block_info, instance_mask);
 
 	if (amdgpu_ras_intr_triggered())
 		return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index cc005e3bcd400..de8e70b3db75c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -770,7 +770,7 @@ static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring);
 static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
 					  void *ras_error_status);
 static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev,
-				     void *inject_if);
+				     void *inject_if, uint32_t instance_mask);
 static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev);
 
 static void gfx_v9_0_kiq_set_resources(struct amdgpu_ring *kiq_ring,
@@ -6335,7 +6335,7 @@ static const struct soc15_ras_field_entry gfx_v9_0_ras_fields[] = {
 };
 
 static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev,
-				     void *inject_if)
+				     void *inject_if, uint32_t instance_mask)
 {
 	struct ras_inject_if *info = (struct ras_inject_if *)inject_if;
 	int ret;
@@ -6374,7 +6374,7 @@ static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev,
 	block_info.value = info->value;
 
 	mutex_lock(&adev->grbm_idx_mutex);
-	ret = psp_ras_trigger_error(&adev->psp, &block_info);
+	ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
index c67e387a97f5c..59abe162bbaf9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
@@ -971,7 +971,7 @@ static void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev)
 }
 
 static int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev,
-				     void *inject_if)
+				     void *inject_if, uint32_t instance_mask)
 {
 	struct ras_inject_if *info = (struct ras_inject_if *)inject_if;
 	int ret;
@@ -987,7 +987,7 @@ static int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev,
 	block_info.value = info->value;
 
 	mutex_lock(&adev->grbm_idx_mutex);
-	ret = psp_ras_trigger_error(&adev->psp, &block_info);
+	ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index ec7c049c59521..4906affa6f8cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -1699,7 +1699,8 @@ static void gfx_v9_4_2_reset_ras_error_count(struct amdgpu_device *adev)
 	gfx_v9_4_2_query_utc_edc_count(adev, NULL, NULL);
 }
 
-static int gfx_v9_4_2_ras_error_inject(struct amdgpu_device *adev, void *inject_if)
+static int gfx_v9_4_2_ras_error_inject(struct amdgpu_device *adev,
+			void *inject_if, uint32_t instance_mask)
 {
 	struct ras_inject_if *info = (struct ras_inject_if *)inject_if;
 	int ret;
@@ -1715,7 +1716,7 @@ static int gfx_v9_4_2_ras_error_inject(struct amdgpu_device *adev, void *inject_
 	block_info.value = info->value;
 
 	mutex_lock(&adev->grbm_idx_mutex);
-	ret = psp_ras_trigger_error(&adev->psp, &block_info);
+	ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	return ret;
-- 
GitLab


From 27c5f29526d17a33946a6b977a0274eae320c489 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 13 Mar 2023 16:24:11 +0800
Subject: [PATCH 1333/1791] drm/amdgpu: reorganize RAS injection flow

So GFX RAS injection could use default function if it doesn't define its
own injection interface.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 7ae08f168f990..b7d8250a92813 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1123,16 +1123,15 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev,
 							  block_info.address);
 	}
 
-	if (info->head.block == AMDGPU_RAS_BLOCK__GFX) {
-		if (block_obj->hw_ops->ras_error_inject)
+	if (block_obj->hw_ops->ras_error_inject) {
+		if (info->head.block == AMDGPU_RAS_BLOCK__GFX)
 			ret = block_obj->hw_ops->ras_error_inject(adev, info, info->instance_mask);
-	} else {
-		/* If defined special ras_error_inject(e.g: xgmi), implement special ras_error_inject */
-		if (block_obj->hw_ops->ras_error_inject)
+		else /* Special ras_error_inject is defined (e.g: xgmi) */
 			ret = block_obj->hw_ops->ras_error_inject(adev, &block_info,
 						info->instance_mask);
-		else  /*If not defined .ras_error_inject, use default ras_error_inject*/
-			ret = psp_ras_trigger_error(&adev->psp, &block_info, info->instance_mask);
+	} else {
+		/* default path */
+		ret = psp_ras_trigger_error(&adev->psp, &block_info, info->instance_mask);
 	}
 
 	if (ret)
-- 
GitLab


From 6e3c51a5814aff70b72e8b2a9953bdac7aea2f44 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 13 Mar 2023 16:34:19 +0800
Subject: [PATCH 1334/1791] drm/amdgpu: remove RAS GFX injection for
 gfx_v9_4/gfx_v9_4_2

No special requirement in RAS injection for the two versions, switch to
use default injection interface.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c   | 24 ------------------------
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c | 24 ------------------------
 2 files changed, 48 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
index 59abe162bbaf9..bc8416afb62c5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c
@@ -970,29 +970,6 @@ static void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev)
 	WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_DSM_INDEX, 255);
 }
 
-static int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev,
-				     void *inject_if, uint32_t instance_mask)
-{
-	struct ras_inject_if *info = (struct ras_inject_if *)inject_if;
-	int ret;
-	struct ta_ras_trigger_error_input block_info = { 0 };
-
-	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
-		return -EINVAL;
-
-	block_info.block_id = amdgpu_ras_block_to_ta(info->head.block);
-	block_info.sub_block_index = info->head.sub_block_index;
-	block_info.inject_error_type = amdgpu_ras_error_to_ta(info->head.type);
-	block_info.address = info->address;
-	block_info.value = info->value;
-
-	mutex_lock(&adev->grbm_idx_mutex);
-	ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask);
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return ret;
-}
-
 static const struct soc15_reg_entry gfx_v9_4_ea_err_status_regs =
 	{ SOC15_REG_ENTRY(GC, 0, mmGCEA_ERR_STATUS), 0, 1, 32 };
 
@@ -1030,7 +1007,6 @@ static void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev)
 
 
 const struct amdgpu_ras_block_hw_ops  gfx_v9_4_ras_ops = {
-	.ras_error_inject = &gfx_v9_4_ras_error_inject,
 	.query_ras_error_count = &gfx_v9_4_query_ras_error_count,
 	.reset_ras_error_count = &gfx_v9_4_reset_ras_error_count,
 	.query_ras_error_status = &gfx_v9_4_query_ras_error_status,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index 4906affa6f8cb..2cc3a7cb1f542 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -1699,29 +1699,6 @@ static void gfx_v9_4_2_reset_ras_error_count(struct amdgpu_device *adev)
 	gfx_v9_4_2_query_utc_edc_count(adev, NULL, NULL);
 }
 
-static int gfx_v9_4_2_ras_error_inject(struct amdgpu_device *adev,
-			void *inject_if, uint32_t instance_mask)
-{
-	struct ras_inject_if *info = (struct ras_inject_if *)inject_if;
-	int ret;
-	struct ta_ras_trigger_error_input block_info = { 0 };
-
-	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX))
-		return -EINVAL;
-
-	block_info.block_id = amdgpu_ras_block_to_ta(info->head.block);
-	block_info.sub_block_index = info->head.sub_block_index;
-	block_info.inject_error_type = amdgpu_ras_error_to_ta(info->head.type);
-	block_info.address = info->address;
-	block_info.value = info->value;
-
-	mutex_lock(&adev->grbm_idx_mutex);
-	ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask);
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return ret;
-}
-
 static void gfx_v9_4_2_query_ea_err_status(struct amdgpu_device *adev)
 {
 	uint32_t i, j;
@@ -1945,7 +1922,6 @@ static bool gfx_v9_4_2_query_uctl2_poison_status(struct amdgpu_device *adev)
 }
 
 struct amdgpu_ras_block_hw_ops  gfx_v9_4_2_ras_ops = {
-		.ras_error_inject = &gfx_v9_4_2_ras_error_inject,
 		.query_ras_error_count = &gfx_v9_4_2_query_ras_error_count,
 		.reset_ras_error_count = &gfx_v9_4_2_reset_ras_error_count,
 		.query_ras_error_status = &gfx_v9_4_2_query_ras_error_status,
-- 
GitLab


From f464c5dd4d918d4dd84eda7e68d4a0b6d41fe37f Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 20 Mar 2023 18:21:14 +0800
Subject: [PATCH 1335/1791] drm/amdgpu: add check for RAS instance mask

The mask is only needed to be set when RAS block instance number is
more than 1 and invalid bits should be also masked out.
We only check valid bits for GFX and SDMA block for now, and will
add check for other RAS blocks in the future.

v2: move the check under injection operation since the mask is only
    used by RAS error inject.
v3: add valid bits handling for SDMA.
v4: print message if the mask is adjusted.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 38 +++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index b7d8250a92813..6bb438642cc03 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -333,6 +333,42 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 	return 0;
 }
 
+static void amdgpu_ras_instance_mask_check(struct amdgpu_device *adev,
+				struct ras_debug_if *data)
+{
+	int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1;
+	uint32_t mask, inst_mask = data->inject.instance_mask;
+
+	/* no need to set instance mask if there is only one instance */
+	if (num_xcc <= 1 && inst_mask) {
+		data->inject.instance_mask = 0;
+		dev_dbg(adev->dev,
+			"RAS inject mask(0x%x) isn't supported and force it to 0.\n",
+			inst_mask);
+
+		return;
+	}
+
+	switch (data->head.block) {
+	case AMDGPU_RAS_BLOCK__GFX:
+		mask = GENMASK(num_xcc - 1, 0);
+		break;
+	case AMDGPU_RAS_BLOCK__SDMA:
+		mask = GENMASK(adev->sdma.num_instances - 1, 0);
+		break;
+	default:
+		mask = 0;
+		break;
+	}
+
+	/* remove invalid bits in instance mask */
+	data->inject.instance_mask &= mask;
+	if (inst_mask != data->inject.instance_mask)
+		dev_dbg(adev->dev,
+			"Adjust RAS inject mask 0x%x to 0x%x\n",
+			inst_mask, data->inject.instance_mask);
+}
+
 /**
  * DOC: AMDGPU RAS debugfs control interface
  *
@@ -468,6 +504,8 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f,
 			break;
 		}
 
+		amdgpu_ras_instance_mask_check(adev, &data);
+
 		/* data.inject.address is offset instead of absolute gpu address */
 		ret = amdgpu_ras_error_inject(adev, &data.inject);
 		break;
-- 
GitLab


From 0409022c15623d91c112e51f38cb62633becd432 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 11 May 2023 17:01:03 +0800
Subject: [PATCH 1336/1791] drm/amdkfd: Fix null ptr access

Avoid access null xcp_mgr pointer.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index bf32e547182c9..2cfef3f9456f1 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -644,12 +644,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	 * If the VMID range changes for GFX9.4.3, then this code MUST be
 	 * revisited.
 	 */
-	partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr, AMDGPU_XCP_FL_LOCKED);
-	if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
-	    partition_mode == AMDGPU_CPX_PARTITION_MODE &&
-	    kfd->num_nodes != 1) {
-		vmid_num_kfd /= 2;
-		first_vmid_kfd = last_vmid_kfd + 1 - vmid_num_kfd*2;
+	if (kfd->adev->xcp_mgr) {
+		partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr,
+								 AMDGPU_XCP_FL_LOCKED);
+		if (partition_mode == AMDGPU_CPX_PARTITION_MODE &&
+		    kfd->num_nodes != 1) {
+			vmid_num_kfd /= 2;
+			first_vmid_kfd = last_vmid_kfd + 1 - vmid_num_kfd*2;
+		}
 	}
 
 	/* Verify module parameters regarding mapped process number*/
-- 
GitLab


From 9a3ce1a7a9e5372d8c275bf3fbef4456c8407145 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Fri, 12 May 2023 13:22:57 +0800
Subject: [PATCH 1337/1791] drm/amdgpu: Do not access members of xcp w/o check
 (v2)

Not all the asic needs xcp. ensure check xcp availabity
before accessing its member.

v2: add missing change in kfd_topology.c

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 8 ++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    | 2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c  | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 85df73f2c85ed..739eb7c0d133b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -429,17 +429,17 @@ uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
 				      struct kfd_local_mem_info *mem_info,
-				      uint8_t xcp_id)
+				      struct amdgpu_xcp *xcp)
 {
 	memset(mem_info, 0, sizeof(*mem_info));
 
-	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+	if (xcp) {
 		if (adev->gmc.real_vram_size == adev->gmc.visible_vram_size)
 			mem_info->local_mem_size_public =
-					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+					KFD_XCP_MEMORY_SIZE(adev, xcp->id);
 		else
 			mem_info->local_mem_size_private =
-					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+					KFD_XCP_MEMORY_SIZE(adev, xcp->id);
 	} else {
 		mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
 		mem_info->local_mem_size_private = adev->gmc.real_vram_size -
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 48d12dbff9681..be43d71ba7ef4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -233,7 +233,7 @@ uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 				      enum kgd_engine_type type);
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
 				      struct kfd_local_mem_info *mem_info,
-				      uint8_t xcp_id);
+				      struct amdgpu_xcp *xcp);
 uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev);
 
 uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 2cfef3f9456f1..986543a000bf5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -784,7 +784,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		atomic_set(&node->sram_ecc_flag, 0);
 
 		amdgpu_amdkfd_get_local_mem_info(kfd->adev,
-					&node->local_mem_info, node->xcp->id);
+					&node->local_mem_info, node->xcp);
 
 		/* Initialize the KFD node */
 		if (kfd_init_node(node)) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e0bacf017a405..8302d8967158b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1235,7 +1235,7 @@ static void kfd_fill_mem_clk_max_info(struct kfd_topology_device *dev)
 	 *	all the banks will report the same mem_clk_max information
 	 */
 	amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info,
-					 dev->gpu->xcp->id);
+					 dev->gpu->xcp);
 
 	list_for_each_entry(mem, &dev->mem_props, list)
 		mem->mem_clk_max = local_mem_info.mem_clk_max;
-- 
GitLab


From d78c71321ea963378cd3c5646ac6c6483d8d1a50 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Thu, 2 Feb 2023 18:57:04 +0800
Subject: [PATCH 1338/1791] drm/amdgpu: add GFX RAS common function

The common function can help reduce redundant code.

v2: remove xcp operation, only need to do RAS operations for all
instances.
v3: remove check for GFX RAS support, will be checked in higher level.
    add amdgpu prefix for the function name.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 19 +++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h |  4 ++++
 2 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 6098b8b1ae5b1..8883d5ee13cb5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -884,6 +884,25 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev,
+		void *ras_error_status,
+		void (*func)(struct amdgpu_device *adev, void *ras_error_status,
+				int xcc_id))
+{
+	int i;
+	int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1;
+	uint32_t xcc_mask = GENMASK(num_xcc - 1, 0);
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
+
+	if (err_data) {
+		err_data->ue_count = 0;
+		err_data->ce_count = 0;
+	}
+
+	for_each_inst(i, xcc_mask)
+		func(adev, ras_error_status, i);
+}
+
 uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
 {
 	signed long r, cnt = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 3d11b7a0bd759..0df53fe7b1999 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -494,6 +494,10 @@ int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev,
 bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id);
 int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev);
 void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev);
+void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev,
+		void *ras_error_status,
+		void (*func)(struct amdgpu_device *adev, void *ras_error_status,
+				int xcc_id));
 
 static inline const char *amdgpu_gfx_compute_mode_desc(int mode)
 {
-- 
GitLab


From bf16235b39d4ca9c8dd47ec1b2faded6ea58f7a2 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Thu, 2 Feb 2023 17:20:23 +0800
Subject: [PATCH 1339/1791] drm/amdgpu: add RAS status query for gfx_v9_4_3

Query GFX RAS status.

v2: remove xcp operation.
v3: change instance from 0 to xcc_id for register access.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 75 +++++++++++++++++++++++++
 1 file changed, 75 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 6cde05421a100..f178e3f565e92 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2980,6 +2980,81 @@ static void gfx_v9_4_3_emit_wave_limit(struct amdgpu_ring *ring, bool enable)
 	}
 }
 
+static const struct soc15_reg_entry gfx_v9_4_3_ea_err_status_regs = {
+	SOC15_REG_ENTRY(GC, 0, regGCEA_ERR_STATUS), 0, 1, 16
+};
+
+static void gfx_v9_4_3_inst_query_ea_err_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	uint32_t i, j;
+	uint32_t reg_value;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	for (i = 0; i < gfx_v9_4_3_ea_err_status_regs.se_num; i++) {
+		for (j = 0; j < gfx_v9_4_3_ea_err_status_regs.instance; j++) {
+			gfx_v9_4_3_xcc_select_se_sh(adev, i, 0, j, xcc_id);
+			reg_value = RREG32_SOC15(GC, GET_INST(GC, xcc_id),
+					regGCEA_ERR_STATUS);
+			if (REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_RDRSP_STATUS) ||
+			    REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_WRRSP_STATUS) ||
+			    REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_RDRSP_DATAPARITY_ERROR)) {
+				dev_warn(adev->dev,
+					"GCEA err detected at instance: %d, status: 0x%x!\n",
+					j, reg_value);
+			}
+			/* clear after read */
+			reg_value = REG_SET_FIELD(reg_value, GCEA_ERR_STATUS,
+						  CLEAR_ERROR_STATUS, 0x1);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS,
+					reg_value);
+		}
+	}
+
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
+
+static void gfx_v9_4_3_inst_query_utc_err_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	uint32_t data;
+
+	data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS);
+	if (data) {
+		dev_warn(adev->dev, "GFX UTCL2 Mem Ecc Status: 0x%x!\n", data);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS, 0x3);
+	}
+
+	data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS);
+	if (data) {
+		dev_warn(adev->dev, "GFX VML2 Mem Ecc Status: 0x%x!\n", data);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS, 0x3);
+	}
+
+	data = RREG32_SOC15(GC, GET_INST(GC, xcc_id),
+				regVML2_WALKER_MEM_ECC_STATUS);
+	if (data) {
+		dev_warn(adev->dev, "GFX VML2 Walker Mem Ecc Status: 0x%x!\n", data);
+		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_WALKER_MEM_ECC_STATUS,
+				0x3);
+	}
+}
+
+static void gfx_v9_4_3_inst_query_ras_err_status(struct amdgpu_device *adev,
+					void *ras_error_status, int xcc_id)
+{
+	gfx_v9_4_3_inst_query_ea_err_status(adev, xcc_id);
+	gfx_v9_4_3_inst_query_utc_err_status(adev, xcc_id);
+}
+
+static void gfx_v9_4_3_query_ras_error_status(struct amdgpu_device *adev)
+{
+	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_query_ras_err_status);
+}
+
 static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = {
 	.name = "gfx_v9_4_3",
 	.early_init = gfx_v9_4_3_early_init,
-- 
GitLab


From 47e7f527c8256a2fe3e61fcd5f59c18bc3fb53fc Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Fri, 3 Feb 2023 10:41:26 +0800
Subject: [PATCH 1340/1791] drm/amdgpu: add RAS status reset for gfx_v9_4_3

Reset GFX RAS status registers.

v2: fix typo in title.
    remove xcp operation.
v3: change instance from 0 to xcc_id for register access.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 41 +++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index f178e3f565e92..e6069d081f71d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -3050,11 +3050,52 @@ static void gfx_v9_4_3_inst_query_ras_err_status(struct amdgpu_device *adev,
 	gfx_v9_4_3_inst_query_utc_err_status(adev, xcc_id);
 }
 
+static void gfx_v9_4_3_inst_reset_utc_err_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS, 0x3);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS, 0x3);
+	WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_WALKER_MEM_ECC_STATUS, 0x3);
+}
+
+static void gfx_v9_4_3_inst_reset_ea_err_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	uint32_t i, j;
+	uint32_t value;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	for (i = 0; i < gfx_v9_4_3_ea_err_status_regs.se_num; i++) {
+		for (j = 0; j < gfx_v9_4_3_ea_err_status_regs.instance; j++) {
+			gfx_v9_4_3_xcc_select_se_sh(adev, i, 0, j, xcc_id);
+			value = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS);
+			value = REG_SET_FIELD(value, GCEA_ERR_STATUS,
+						CLEAR_ERROR_STATUS, 0x1);
+			WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS, value);
+		}
+	}
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
+
+static void gfx_v9_4_3_inst_reset_ras_err_status(struct amdgpu_device *adev,
+					void *ras_error_status, int xcc_id)
+{
+	gfx_v9_4_3_inst_reset_utc_err_status(adev, xcc_id);
+	gfx_v9_4_3_inst_reset_ea_err_status(adev, xcc_id);
+}
+
 static void gfx_v9_4_3_query_ras_error_status(struct amdgpu_device *adev)
 {
 	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_query_ras_err_status);
 }
 
+static void gfx_v9_4_3_reset_ras_error_status(struct amdgpu_device *adev)
+{
+	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_reset_ras_err_status);
+}
+
 static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = {
 	.name = "gfx_v9_4_3",
 	.early_init = gfx_v9_4_3_early_init,
-- 
GitLab


From 2b80ffc2d86cce8444d5fc0237afd77f7d18cd1d Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Mon, 27 Feb 2023 17:36:19 +0800
Subject: [PATCH 1341/1791] drm/amdgpu: Add gc v9_4_3 ras error status
 registers

GC v9_4_3 introduces UE|CE_ERR_STATUS_LO|HI to log
hardware errors

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/include/asic_reg/gc/gc_9_4_3_offset.h |  192 +++
 .../include/asic_reg/gc/gc_9_4_3_sh_mask.h    | 1112 +++++++++++++++++
 2 files changed, 1304 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h
index 3100de8b38810..393963502b7a6 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h
@@ -705,6 +705,46 @@
 #define regSQC_ICACHE_UTCL1_STATUS_BASE_IDX                                                             0
 #define regSQC_DCACHE_UTCL1_STATUS                                                                      0x03d8
 #define regSQC_DCACHE_UTCL1_STATUS_BASE_IDX                                                             0
+#define regSQC_UE_EDC_LO                                                                                0x03d9
+#define regSQC_UE_EDC_LO_BASE_IDX                                                                       0
+#define regSQC_UE_EDC_HI                                                                                0x03da
+#define regSQC_UE_EDC_HI_BASE_IDX                                                                       0
+#define regSQC_CE_EDC_LO                                                                                0x03db
+#define regSQC_CE_EDC_LO_BASE_IDX                                                                       0
+#define regSQC_CE_EDC_HI                                                                                0x03dc
+#define regSQC_CE_EDC_HI_BASE_IDX                                                                       0
+#define regSQ_UE_ERR_STATUS_LO                                                                          0x03dd
+#define regSQ_UE_ERR_STATUS_LO_BASE_IDX                                                                 0
+#define regSQ_UE_ERR_STATUS_HI                                                                          0x03de
+#define regSQ_UE_ERR_STATUS_HI_BASE_IDX                                                                 0
+#define regSQ_CE_ERR_STATUS_LO                                                                          0x03df
+#define regSQ_CE_ERR_STATUS_LO_BASE_IDX                                                                 0
+#define regSQ_CE_ERR_STATUS_HI                                                                          0x03e0
+#define regSQ_CE_ERR_STATUS_HI_BASE_IDX                                                                 0
+#define regLDS_UE_ERR_STATUS_LO                                                                         0x03e1
+#define regLDS_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regLDS_UE_ERR_STATUS_HI                                                                         0x03e2
+#define regLDS_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regLDS_CE_ERR_STATUS_LO                                                                         0x03e3
+#define regLDS_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regLDS_CE_ERR_STATUS_HI                                                                         0x03e4
+#define regLDS_CE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regSP0_UE_ERR_STATUS_LO                                                                         0x03e5
+#define regSP0_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSP0_UE_ERR_STATUS_HI                                                                         0x03e6
+#define regSP0_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regSP0_CE_ERR_STATUS_LO                                                                         0x03e7
+#define regSP0_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSP0_CE_ERR_STATUS_HI                                                                         0x03e8
+#define regSP0_CE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regSP1_UE_ERR_STATUS_LO                                                                         0x03e9
+#define regSP1_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSP1_UE_ERR_STATUS_HI                                                                         0x03ea
+#define regSP1_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regSP1_CE_ERR_STATUS_LO                                                                         0x03eb
+#define regSP1_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSP1_CE_ERR_STATUS_HI                                                                         0x03ec
+#define regSP1_CE_ERR_STATUS_HI_BASE_IDX                                                                0
 
 
 // addressBlock: xcd0_gc_shsdec
@@ -727,6 +767,14 @@
 #define regSPI_DSM_CNTL2_BASE_IDX                                                                       0
 #define regSPI_EDC_CNT                                                                                  0x0445
 #define regSPI_EDC_CNT_BASE_IDX                                                                         0
+#define regSPI_UE_ERR_STATUS_LO                                                                         0x0446
+#define regSPI_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSPI_UE_ERR_STATUS_HI                                                                         0x0447
+#define regSPI_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regSPI_CE_ERR_STATUS_LO                                                                         0x0448
+#define regSPI_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regSPI_CE_ERR_STATUS_HI                                                                         0x0449
+#define regSPI_CE_ERR_STATUS_HI_BASE_IDX                                                                0
 #define regSPI_DEBUG_BUSY                                                                               0x0450
 #define regSPI_DEBUG_BUSY_BASE_IDX                                                                      0
 #define regSPI_CONFIG_PS_CU_EN                                                                          0x0452
@@ -871,6 +919,14 @@
 #define regTD_STATUS_BASE_IDX                                                                           0
 #define regTD_POWER_CNTL                                                                                0x052a
 #define regTD_POWER_CNTL_BASE_IDX                                                                       0
+#define regTD_UE_EDC_LO                                                                                 0x052b
+#define regTD_UE_EDC_LO_BASE_IDX                                                                        0
+#define regTD_UE_EDC_HI                                                                                 0x052c
+#define regTD_UE_EDC_HI_BASE_IDX                                                                        0
+#define regTD_CE_EDC_LO                                                                                 0x052d
+#define regTD_CE_EDC_LO_BASE_IDX                                                                        0
+#define regTD_CE_EDC_HI                                                                                 0x052e
+#define regTD_CE_EDC_HI_BASE_IDX                                                                        0
 #define regTD_DSM_CNTL                                                                                  0x052f
 #define regTD_DSM_CNTL_BASE_IDX                                                                         0
 #define regTD_DSM_CNTL2                                                                                 0x0530
@@ -893,6 +949,14 @@
 #define regTA_DSM_CNTL_BASE_IDX                                                                         0
 #define regTA_DSM_CNTL2                                                                                 0x0585
 #define regTA_DSM_CNTL2_BASE_IDX                                                                        0
+#define regTA_UE_EDC_LO                                                                                 0x0587
+#define regTA_UE_EDC_LO_BASE_IDX                                                                        0
+#define regTA_UE_EDC_HI                                                                                 0x0588
+#define regTA_UE_EDC_HI_BASE_IDX                                                                        0
+#define regTA_CE_EDC_LO                                                                                 0x0589
+#define regTA_CE_EDC_LO_BASE_IDX                                                                        0
+#define regTA_CE_EDC_HI                                                                                 0x058a
+#define regTA_CE_EDC_HI_BASE_IDX                                                                        0
 
 
 // addressBlock: xcd0_gc_gdsdec
@@ -923,6 +987,14 @@
 #define regGDS_DSM_CNTL2_BASE_IDX                                                                       0
 #define regGDS_WD_GDS_CSB                                                                               0x05ce
 #define regGDS_WD_GDS_CSB_BASE_IDX                                                                      0
+#define regGDS_UE_ERR_STATUS_LO                                                                         0x05cf
+#define regGDS_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regGDS_UE_ERR_STATUS_HI                                                                         0x05d0
+#define regGDS_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regGDS_CE_ERR_STATUS_LO                                                                         0x05d1
+#define regGDS_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regGDS_CE_ERR_STATUS_HI                                                                         0x05d2
+#define regGDS_CE_ERR_STATUS_HI_BASE_IDX                                                                0
 
 
 // addressBlock: xcd0_gc_rbdec
@@ -1243,6 +1315,10 @@
 #define regGCEA_MAM_CTRL_BASE_IDX                                                                       0
 #define regGCEA_MAM_CTRL2                                                                               0x0702
 #define regGCEA_MAM_CTRL2_BASE_IDX                                                                      0
+#define regGCEA_UE_ERR_STATUS_LO                                                                        0x0706
+#define regGCEA_UE_ERR_STATUS_LO_BASE_IDX                                                               0
+#define regGCEA_UE_ERR_STATUS_HI                                                                        0x0707
+#define regGCEA_UE_ERR_STATUS_HI_BASE_IDX                                                               0
 #define regGCEA_DSM_CNTL                                                                                0x0708
 #define regGCEA_DSM_CNTL_BASE_IDX                                                                       0
 #define regGCEA_DSM_CNTLA                                                                               0x0709
@@ -1277,6 +1353,10 @@
 #define regGCEA_SDP_BACKDOOR_DATACREDITS1_BASE_IDX                                                      0
 #define regGCEA_SDP_BACKDOOR_MISCCREDITS                                                                0x0719
 #define regGCEA_SDP_BACKDOOR_MISCCREDITS_BASE_IDX                                                       0
+#define regGCEA_CE_ERR_STATUS_LO                                                                        0x071b
+#define regGCEA_CE_ERR_STATUS_LO_BASE_IDX                                                               0
+#define regGCEA_CE_ERR_STATUS_HI                                                                        0x071d
+#define regGCEA_CE_ERR_STATUS_HI_BASE_IDX                                                               0
 #define regGCEA_SDP_ENABLE                                                                              0x071f
 #define regGCEA_SDP_ENABLE_BASE_IDX                                                                     0
 
@@ -1389,6 +1469,14 @@
 #define regATC_L2_CNTL4_BASE_IDX                                                                        0
 #define regATC_L2_MM_GROUP_RT_CLASSES                                                                   0x0816
 #define regATC_L2_MM_GROUP_RT_CLASSES_BASE_IDX                                                          0
+#define regATC_L2_UE_ERR_STATUS_LO                                                                      0x081a
+#define regATC_L2_UE_ERR_STATUS_LO_BASE_IDX                                                             0
+#define regATC_L2_UE_ERR_STATUS_HI                                                                      0x081b
+#define regATC_L2_UE_ERR_STATUS_HI_BASE_IDX                                                             0
+#define regATC_L2_CE_ERR_STATUS_LO                                                                      0x081c
+#define regATC_L2_CE_ERR_STATUS_LO_BASE_IDX                                                             0
+#define regATC_L2_CE_ERR_STATUS_HI                                                                      0x081d
+#define regATC_L2_CE_ERR_STATUS_HI_BASE_IDX                                                             0
 
 
 // addressBlock: xcd0_gc_utcl2_vml2pfdec
@@ -1475,6 +1563,30 @@
 #define regUTCL2_EDC_MODE_BASE_IDX                                                                      0
 #define regUTCL2_EDC_CONFIG                                                                             0x084c
 #define regUTCL2_EDC_CONFIG_BASE_IDX                                                                    0
+#define regVML2_UE_ERR_STATUS_LO                                                                        0x084d
+#define regVML2_UE_ERR_STATUS_LO_BASE_IDX                                                               0
+#define regVML2_WALKER_UE_ERR_STATUS_LO                                                                 0x084e
+#define regVML2_WALKER_UE_ERR_STATUS_LO_BASE_IDX                                                        0
+#define regUTCL2_UE_ERR_STATUS_LO                                                                       0x084f
+#define regUTCL2_UE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regVML2_UE_ERR_STATUS_HI                                                                        0x0850
+#define regVML2_UE_ERR_STATUS_HI_BASE_IDX                                                               0
+#define regVML2_WALKER_UE_ERR_STATUS_HI                                                                 0x0851
+#define regVML2_WALKER_UE_ERR_STATUS_HI_BASE_IDX                                                        0
+#define regUTCL2_UE_ERR_STATUS_HI                                                                       0x0852
+#define regUTCL2_UE_ERR_STATUS_HI_BASE_IDX                                                              0
+#define regVML2_CE_ERR_STATUS_LO                                                                        0x0853
+#define regVML2_CE_ERR_STATUS_LO_BASE_IDX                                                               0
+#define regVML2_WALKER_CE_ERR_STATUS_LO                                                                 0x0854
+#define regVML2_WALKER_CE_ERR_STATUS_LO_BASE_IDX                                                        0
+#define regUTCL2_CE_ERR_STATUS_LO                                                                       0x0855
+#define regUTCL2_CE_ERR_STATUS_LO_BASE_IDX                                                              0
+#define regVML2_CE_ERR_STATUS_HI                                                                        0x0856
+#define regVML2_CE_ERR_STATUS_HI_BASE_IDX                                                               0
+#define regVML2_WALKER_CE_ERR_STATUS_HI                                                                 0x0857
+#define regVML2_WALKER_CE_ERR_STATUS_HI_BASE_IDX                                                        0
+#define regUTCL2_CE_ERR_STATUS_HI                                                                       0x0858
+#define regUTCL2_CE_ERR_STATUS_HI_BASE_IDX                                                              0
 
 
 // addressBlock: xcd0_gc_utcl2_vml2vcdec
@@ -2011,6 +2123,22 @@
 #define regTC_CFG_L1_VOLATILE_BASE_IDX                                                                  0
 #define regTC_CFG_L2_VOLATILE                                                                           0x0b23
 #define regTC_CFG_L2_VOLATILE_BASE_IDX                                                                  0
+#define regTCP_UE_EDC_HI_REG                                                                            0x0b54
+#define regTCP_UE_EDC_HI_REG_BASE_IDX                                                                   0
+#define regTCP_UE_EDC_LO_REG                                                                            0x0b55
+#define regTCP_UE_EDC_LO_REG_BASE_IDX                                                                   0
+#define regTCP_CE_EDC_HI_REG                                                                            0x0b56
+#define regTCP_CE_EDC_HI_REG_BASE_IDX                                                                   0
+#define regTCP_CE_EDC_LO_REG                                                                            0x0b57
+#define regTCP_CE_EDC_LO_REG_BASE_IDX                                                                   0
+#define regTCI_UE_EDC_HI_REG                                                                            0x0b58
+#define regTCI_UE_EDC_HI_REG_BASE_IDX                                                                   0
+#define regTCI_UE_EDC_LO_REG                                                                            0x0b59
+#define regTCI_UE_EDC_LO_REG_BASE_IDX                                                                   0
+#define regTCI_CE_EDC_HI_REG                                                                            0x0b5a
+#define regTCI_CE_EDC_HI_REG_BASE_IDX                                                                   0
+#define regTCI_CE_EDC_LO_REG                                                                            0x0b5b
+#define regTCI_CE_EDC_LO_REG_BASE_IDX                                                                   0
 #define regTCI_MISC                                                                                     0x0b5c
 #define regTCI_MISC_BASE_IDX                                                                            0
 #define regTCI_CNTL_3                                                                                   0x0b5d
@@ -2061,6 +2189,26 @@
 #define regTCX_DSM_CNTL_BASE_IDX                                                                        0
 #define regTCX_DSM_CNTL2                                                                                0x0bc8
 #define regTCX_DSM_CNTL2_BASE_IDX                                                                       0
+#define regTCA_UE_ERR_STATUS_LO                                                                         0x0bc9
+#define regTCA_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regTCA_UE_ERR_STATUS_HI                                                                         0x0bca
+#define regTCA_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regTCX_UE_ERR_STATUS_LO                                                                         0x0bcb
+#define regTCX_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regTCX_UE_ERR_STATUS_HI                                                                         0x0bcc
+#define regTCX_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regTCX_CE_ERR_STATUS_LO                                                                         0x0bcd
+#define regTCX_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regTCX_CE_ERR_STATUS_HI                                                                         0x0bce
+#define regTCX_CE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regTCC_UE_ERR_STATUS_LO                                                                         0x0bcf
+#define regTCC_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regTCC_UE_ERR_STATUS_HI                                                                         0x0bd0
+#define regTCC_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regTCC_CE_ERR_STATUS_LO                                                                         0x0bd1
+#define regTCC_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regTCC_CE_ERR_STATUS_HI                                                                         0x0bd2
+#define regTCC_CE_ERR_STATUS_HI_BASE_IDX                                                                0
 
 
 // addressBlock: xcd0_gc_shdec
@@ -2905,6 +3053,30 @@
 #define regCP_MEC2_F32_INT_DIS_BASE_IDX                                                                 0
 #define regCP_VMID_STATUS                                                                               0x10bf
 #define regCP_VMID_STATUS_BASE_IDX                                                                      0
+#define regCPC_UE_ERR_STATUS_LO                                                                         0x10e0
+#define regCPC_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPC_UE_ERR_STATUS_HI                                                                         0x10e1
+#define regCPC_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regCPC_CE_ERR_STATUS_LO                                                                         0x10e2
+#define regCPC_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPC_CE_ERR_STATUS_HI                                                                         0x10e3
+#define regCPC_CE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regCPF_UE_ERR_STATUS_LO                                                                         0x10e4
+#define regCPF_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPF_UE_ERR_STATUS_HI                                                                         0x10e5
+#define regCPF_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regCPF_CE_ERR_STATUS_LO                                                                         0x10e6
+#define regCPF_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPF_CE_ERR_STATUS_HI                                                                         0x10e7
+#define regCPF_CE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regCPG_UE_ERR_STATUS_LO                                                                         0x10e8
+#define regCPG_UE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPG_UE_ERR_STATUS_HI                                                                         0x10e9
+#define regCPG_UE_ERR_STATUS_HI_BASE_IDX                                                                0
+#define regCPG_CE_ERR_STATUS_LO                                                                         0x10ea
+#define regCPG_CE_ERR_STATUS_LO_BASE_IDX                                                                0
+#define regCPG_CE_ERR_STATUS_HI                                                                         0x10eb
+#define regCPG_CE_ERR_STATUS_HI_BASE_IDX                                                                0
 
 
 // addressBlock: xcd0_gc_cppdec2
@@ -5364,6 +5536,18 @@
 #define regSPI_WAVE_LIMIT_CNTL                                                                          0x2443
 #define regSPI_WAVE_LIMIT_CNTL_BASE_IDX                                                                 1
 
+// addressBlock: xcd0_gc_gccanedec
+// base address: 0x33d00
+#define regGC_CANE_ERR_STATUS                                                                           0x2f4d
+#define regGC_CANE_ERR_STATUS_BASE_IDX                                                                  1
+#define regGC_CANE_UE_ERR_STATUS_LO                                                                     0x2f4e
+#define regGC_CANE_UE_ERR_STATUS_LO_BASE_IDX                                                            1
+#define regGC_CANE_UE_ERR_STATUS_HI                                                                     0x2f4f
+#define regGC_CANE_UE_ERR_STATUS_HI_BASE_IDX                                                            1
+#define regGC_CANE_CE_ERR_STATUS_LO                                                                     0x2f50
+#define regGC_CANE_CE_ERR_STATUS_LO_BASE_IDX                                                            1
+#define regGC_CANE_CE_ERR_STATUS_HI                                                                     0x2f51
+#define regGC_CANE_CE_ERR_STATUS_HI_BASE_IDX                                                            1
 
 // addressBlock: xcd0_gc_perfddec
 // base address: 0x34000
@@ -6583,6 +6767,10 @@
 #define regRLC_CAPTURE_GPU_CLOCK_COUNT_2_BASE_IDX                                                       1
 #define regRLC_CPG_STAT_INVAL                                                                           0x4d09
 #define regRLC_CPG_STAT_INVAL_BASE_IDX                                                                  1
+#define regRLC_UE_ERR_STATUS_LOW                                                                        0x4d40
+#define regRLC_UE_ERR_STATUS_LOW_BASE_IDX                                                               1
+#define regRLC_UE_ERR_STATUS_HIGH                                                                       0x4d41
+#define regRLC_UE_ERR_STATUS_HIGH_BASE_IDX                                                              1
 #define regRLC_DSM_CNTL                                                                                 0x4d42
 #define regRLC_DSM_CNTL_BASE_IDX                                                                        1
 #define regRLC_DSM_CNTLA                                                                                0x4d43
@@ -6591,6 +6779,10 @@
 #define regRLC_DSM_CNTL2_BASE_IDX                                                                       1
 #define regRLC_DSM_CNTL2A                                                                               0x4d45
 #define regRLC_DSM_CNTL2A_BASE_IDX                                                                      1
+#define regRLC_CE_ERR_STATUS_LOW                                                                        0x4d49
+#define regRLC_CE_ERR_STATUS_LOW_BASE_IDX                                                               1
+#define regRLC_CE_ERR_STATUS_HIGH                                                                       0x4d4a
+#define regRLC_CE_ERR_STATUS_HIGH_BASE_IDX                                                              1
 #define regRLC_RLCV_SPARE_INT                                                                           0x4f30
 #define regRLC_RLCV_SPARE_INT_BASE_IDX                                                                  1
 #define regRLC_SMU_CLK_REQ                                                                              0x4f97
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h
index 84a75b58347f7..2bd9f3f1026f9 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h
@@ -4129,6 +4129,240 @@
 #define SQC_DCACHE_UTCL1_STATUS__FAULT_DETECTED_MASK                                                          0x00000001L
 #define SQC_DCACHE_UTCL1_STATUS__RETRY_DETECTED_MASK                                                          0x00000002L
 #define SQC_DCACHE_UTCL1_STATUS__PRT_DETECTED_MASK                                                            0x00000004L
+//SQC_UE_EDC_LO
+#define SQC_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                               0x0
+#define SQC_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                         0x1
+#define SQC_UE_EDC_LO__ADDRESS__SHIFT                                                                         0x2
+#define SQC_UE_EDC_LO__MEM_ID__SHIFT                                                                          0x18
+#define SQC_UE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                 0x00000001L
+#define SQC_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                           0x00000002L
+#define SQC_UE_EDC_LO__ADDRESS_MASK                                                                           0x00FFFFFCL
+#define SQC_UE_EDC_LO__MEM_ID_MASK                                                                            0xFF000000L
+//SQC_UE_EDC_HI
+#define SQC_UE_EDC_HI__ECC__SHIFT                                                                             0x0
+#define SQC_UE_EDC_HI__PARITY__SHIFT                                                                          0x1
+#define SQC_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                             0x2
+#define SQC_UE_EDC_HI__ERR_INFO__SHIFT                                                                        0x3
+#define SQC_UE_EDC_HI__UE_CNT__SHIFT                                                                          0x17
+#define SQC_UE_EDC_HI__FED_CNT__SHIFT                                                                         0x1a
+#define SQC_UE_EDC_HI__ECC_MASK                                                                               0x00000001L
+#define SQC_UE_EDC_HI__PARITY_MASK                                                                            0x00000002L
+#define SQC_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                               0x00000004L
+#define SQC_UE_EDC_HI__ERR_INFO_MASK                                                                          0x007FFFF8L
+#define SQC_UE_EDC_HI__UE_CNT_MASK                                                                            0x03800000L
+#define SQC_UE_EDC_HI__FED_CNT_MASK                                                                           0x1C000000L
+//SQC_CE_EDC_LO
+#define SQC_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                               0x0
+#define SQC_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                         0x1
+#define SQC_CE_EDC_LO__ADDRESS__SHIFT                                                                         0x2
+#define SQC_CE_EDC_LO__MEM_ID__SHIFT                                                                          0x18
+#define SQC_CE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                 0x00000001L
+#define SQC_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                           0x00000002L
+#define SQC_CE_EDC_LO__ADDRESS_MASK                                                                           0x00FFFFFCL
+#define SQC_CE_EDC_LO__MEM_ID_MASK                                                                            0xFF000000L
+//SQC_CE_EDC_HI
+#define SQC_CE_EDC_HI__ECC__SHIFT                                                                             0x0
+#define SQC_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                             0x2
+#define SQC_CE_EDC_HI__ERR_INFO__SHIFT                                                                        0x3
+#define SQC_CE_EDC_HI__CE_CNT__SHIFT                                                                          0x17
+#define SQC_CE_EDC_HI__POSION__SHIFT                                                                          0x1a
+#define SQC_CE_EDC_HI__ECC_MASK                                                                               0x00000001L
+#define SQC_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                               0x00000004L
+#define SQC_CE_EDC_HI__ERR_INFO_MASK                                                                          0x007FFFF8L
+#define SQC_CE_EDC_HI__CE_CNT_MASK                                                                            0x03800000L
+#define SQC_CE_EDC_HI__POSION_MASK                                                                            0x04000000L
+//SQ_UE_ERR_STATUS_LO
+#define SQ_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                     0x0
+#define SQ_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                        0x1
+#define SQ_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                   0x2
+#define SQ_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                 0x18
+#define SQ_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                       0x00000001L
+#define SQ_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                          0x00000002L
+#define SQ_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                     0x00FFFFFCL
+#define SQ_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                   0xFF000000L
+//SQ_UE_ERR_STATUS_HI
+#define SQ_UE_ERR_STATUS_HI__ECC__SHIFT                                                                       0x0
+#define SQ_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                    0x1
+#define SQ_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                       0x2
+#define SQ_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                  0x3
+#define SQ_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                    0x17
+#define SQ_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                   0x1a
+#define SQ_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                  0x1d
+#define SQ_UE_ERR_STATUS_HI__ECC_MASK                                                                         0x00000001L
+#define SQ_UE_ERR_STATUS_HI__PARITY_MASK                                                                      0x00000002L
+#define SQ_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                         0x00000004L
+#define SQ_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                    0x007FFFF8L
+#define SQ_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                      0x03800000L
+#define SQ_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                     0x1C000000L
+#define SQ_UE_ERR_STATUS_HI__RESERVED_MASK                                                                    0xE0000000L
+//SQ_CE_ERR_STATUS_LO
+#define SQ_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                     0x0
+#define SQ_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                        0x1
+#define SQ_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                   0x2
+#define SQ_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                 0x18
+#define SQ_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                       0x00000001L
+#define SQ_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                          0x00000002L
+#define SQ_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                     0x00FFFFFCL
+#define SQ_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                   0xFF000000L
+//SQ_CE_ERR_STATUS_HI
+#define SQ_CE_ERR_STATUS_HI__ECC__SHIFT                                                                       0x0
+#define SQ_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                     0x1
+#define SQ_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                       0x2
+#define SQ_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                  0x3
+#define SQ_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                    0x17
+#define SQ_CE_ERR_STATUS_HI__POISON__SHIFT                                                                    0x1a
+#define SQ_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                  0x1b
+#define SQ_CE_ERR_STATUS_HI__ECC_MASK                                                                         0x00000001L
+#define SQ_CE_ERR_STATUS_HI__OTHER_MASK                                                                       0x00000002L
+#define SQ_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                         0x00000004L
+#define SQ_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                    0x007FFFF8L
+#define SQ_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                      0x03800000L
+#define SQ_CE_ERR_STATUS_HI__POISON_MASK                                                                      0x04000000L
+#define SQ_CE_ERR_STATUS_HI__RESERVED_MASK                                                                    0xF8000000L
+//LDS_UE_ERR_STATUS_LO
+#define LDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define LDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define LDS_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define LDS_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define LDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define LDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define LDS_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define LDS_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//LDS_UE_ERR_STATUS_HI
+#define LDS_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define LDS_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define LDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define LDS_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define LDS_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define LDS_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define LDS_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define LDS_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define LDS_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define LDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define LDS_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define LDS_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define LDS_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define LDS_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//LDS_CE_ERR_STATUS_LO
+#define LDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define LDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define LDS_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define LDS_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define LDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define LDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define LDS_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define LDS_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//LDS_CE_ERR_STATUS_HI
+#define LDS_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define LDS_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define LDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define LDS_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define LDS_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define LDS_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define LDS_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define LDS_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define LDS_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define LDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define LDS_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define LDS_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define LDS_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define LDS_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
+//SP0_UE_ERR_STATUS_LO
+#define SP0_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SP0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SP0_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SP0_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SP0_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SP0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SP0_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SP0_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SP0_UE_ERR_STATUS_HI
+#define SP0_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SP0_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define SP0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SP0_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SP0_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define SP0_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define SP0_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define SP0_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SP0_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define SP0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SP0_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SP0_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define SP0_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define SP0_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//SP0_CE_ERR_STATUS_LO
+#define SP0_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SP0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SP0_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SP0_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SP0_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SP0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SP0_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SP0_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SP0_CE_ERR_STATUS_HI
+#define SP0_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SP0_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define SP0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SP0_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SP0_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define SP0_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define SP0_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define SP0_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SP0_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define SP0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SP0_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SP0_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define SP0_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define SP0_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
+//SP1_UE_ERR_STATUS_LO
+#define SP1_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SP1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SP1_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SP1_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SP1_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SP1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SP1_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SP1_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SP1_UE_ERR_STATUS_HI
+#define SP1_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SP1_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define SP1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SP1_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SP1_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define SP1_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define SP1_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define SP1_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SP1_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define SP1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SP1_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SP1_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define SP1_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define SP1_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//SP1_CE_ERR_STATUS_LO
+#define SP1_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SP1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SP1_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SP1_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SP1_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SP1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SP1_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SP1_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SP1_CE_ERR_STATUS_HI
+#define SP1_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SP1_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define SP1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SP1_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SP1_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define SP1_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define SP1_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define SP1_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SP1_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define SP1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SP1_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SP1_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define SP1_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define SP1_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
 
 
 // addressBlock: xcd0_gc_shsdec
@@ -4235,6 +4469,54 @@
 #define SPI_EDC_CNT__SPI_LIFE_CNT_SEC_COUNT_MASK                                                              0x00030000L
 #define SPI_EDC_CNT__SPI_LIFE_CNT_DED_COUNT_MASK                                                              0x000C0000L
 #define SPI_EDC_CNT__UNUSED_MASK                                                                              0xFFF00000L
+//SPI_UE_ERR_STATUS_LO
+#define SPI_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SPI_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SPI_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SPI_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SPI_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SPI_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SPI_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SPI_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SPI_UE_ERR_STATUS_HI
+#define SPI_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SPI_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define SPI_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SPI_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SPI_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define SPI_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define SPI_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define SPI_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SPI_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define SPI_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SPI_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SPI_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define SPI_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define SPI_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//SPI_CE_ERR_STATUS_LO
+#define SPI_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define SPI_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define SPI_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define SPI_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define SPI_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define SPI_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define SPI_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define SPI_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//SPI_CE_ERR_STATUS_HI
+#define SPI_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define SPI_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define SPI_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define SPI_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define SPI_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define SPI_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define SPI_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define SPI_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define SPI_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define SPI_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define SPI_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define SPI_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define SPI_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define SPI_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
 //SPI_DEBUG_BUSY
 #define SPI_DEBUG_BUSY__HS_BUSY__SHIFT                                                                        0x0
 #define SPI_DEBUG_BUSY__GS_BUSY__SHIFT                                                                        0x1
@@ -4622,6 +4904,48 @@
 #define TD_POWER_CNTL__MGCG_OUTPUTSTAGE_MASK                                                                  0x00000002L
 #define TD_POWER_CNTL__MID0_THREAD_DATA_MASK                                                                  0x00000004L
 #define TD_POWER_CNTL__MID2_ACCUM_DATA_MASK                                                                   0x00000008L
+//TD_UE_EDC_LO
+#define TD_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                                0x0
+#define TD_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                          0x1
+#define TD_UE_EDC_LO__ADDRESS__SHIFT                                                                          0x2
+#define TD_UE_EDC_LO__MEM_ID__SHIFT                                                                           0x18
+#define TD_UE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                  0x00000001L
+#define TD_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                            0x00000002L
+#define TD_UE_EDC_LO__ADDRESS_MASK                                                                            0x00FFFFFCL
+#define TD_UE_EDC_LO__MEM_ID_MASK                                                                             0xFF000000L
+//TD_UE_EDC_HI
+#define TD_UE_EDC_HI__ECC__SHIFT                                                                              0x0
+#define TD_UE_EDC_HI__PARITY__SHIFT                                                                           0x1
+#define TD_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                              0x2
+#define TD_UE_EDC_HI__ERR_INFO__SHIFT                                                                         0x3
+#define TD_UE_EDC_HI__UE_CNT__SHIFT                                                                           0x17
+#define TD_UE_EDC_HI__FED_CNT__SHIFT                                                                          0x1a
+#define TD_UE_EDC_HI__ECC_MASK                                                                                0x00000001L
+#define TD_UE_EDC_HI__PARITY_MASK                                                                             0x00000002L
+#define TD_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                                0x00000004L
+#define TD_UE_EDC_HI__ERR_INFO_MASK                                                                           0x007FFFF8L
+#define TD_UE_EDC_HI__UE_CNT_MASK                                                                             0x03800000L
+#define TD_UE_EDC_HI__FED_CNT_MASK                                                                            0x1C000000L
+//TD_CE_EDC_LO
+#define TD_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                                0x0
+#define TD_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                          0x1
+#define TD_CE_EDC_LO__ADDRESS__SHIFT                                                                          0x2
+#define TD_CE_EDC_LO__MEM_ID__SHIFT                                                                           0x18
+#define TD_CE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                  0x00000001L
+#define TD_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                            0x00000002L
+#define TD_CE_EDC_LO__ADDRESS_MASK                                                                            0x00FFFFFCL
+#define TD_CE_EDC_LO__MEM_ID_MASK                                                                             0xFF000000L
+//TD_CE_EDC_HI
+#define TD_CE_EDC_HI__ECC__SHIFT                                                                              0x0
+#define TD_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                              0x2
+#define TD_CE_EDC_HI__ERR_INFO__SHIFT                                                                         0x3
+#define TD_CE_EDC_HI__CE_CNT__SHIFT                                                                           0x17
+#define TD_CE_EDC_HI__POISON__SHIFT                                                                           0x1a
+#define TD_CE_EDC_HI__ECC_MASK                                                                                0x00000001L
+#define TD_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                                0x00000004L
+#define TD_CE_EDC_HI__ERR_INFO_MASK                                                                           0x007FFFF8L
+#define TD_CE_EDC_HI__CE_CNT_MASK                                                                             0x03800000L
+#define TD_CE_EDC_HI__POISON_MASK                                                                             0x04000000L
 //TD_DSM_CNTL
 #define TD_DSM_CNTL__TD_SS_FIFO_LO_DSM_IRRITATOR_DATA__SHIFT                                                  0x0
 #define TD_DSM_CNTL__TD_SS_FIFO_LO_ENABLE_SINGLE_WRITE__SHIFT                                                 0x2
@@ -4771,6 +5095,48 @@
 #define TA_DSM_CNTL2__TA_FS_AFIFO_HI_ENABLE_ERROR_INJECT_MASK                                                 0x000C0000L
 #define TA_DSM_CNTL2__TA_FS_AFIFO_HI_SELECT_INJECT_DELAY_MASK                                                 0x00100000L
 #define TA_DSM_CNTL2__TA_INJECT_DELAY_MASK                                                                    0xFC000000L
+//TA_UE_EDC_LO
+#define TA_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                                0x0
+#define TA_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                          0x1
+#define TA_UE_EDC_LO__ADDRESS__SHIFT                                                                          0x2
+#define TA_UE_EDC_LO__MEM_ID__SHIFT                                                                           0x18
+#define TA_UE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                  0x00000001L
+#define TA_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                            0x00000002L
+#define TA_UE_EDC_LO__ADDRESS_MASK                                                                            0x00FFFFFCL
+#define TA_UE_EDC_LO__MEM_ID_MASK                                                                             0xFF000000L
+//TA_UE_EDC_HI
+#define TA_UE_EDC_HI__ECC__SHIFT                                                                              0x0
+#define TA_UE_EDC_HI__PARITY__SHIFT                                                                           0x1
+#define TA_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                              0x2
+#define TA_UE_EDC_HI__ERR_INFO__SHIFT                                                                         0x3
+#define TA_UE_EDC_HI__UE_CNT__SHIFT                                                                           0x17
+#define TA_UE_EDC_HI__FED_CNT__SHIFT                                                                          0x1a
+#define TA_UE_EDC_HI__ECC_MASK                                                                                0x00000001L
+#define TA_UE_EDC_HI__PARITY_MASK                                                                             0x00000002L
+#define TA_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                                0x00000004L
+#define TA_UE_EDC_HI__ERR_INFO_MASK                                                                           0x007FFFF8L
+#define TA_UE_EDC_HI__UE_CNT_MASK                                                                             0x03800000L
+#define TA_UE_EDC_HI__FED_CNT_MASK                                                                            0x1C000000L
+//TA_CE_EDC_LO
+#define TA_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT                                                                0x0
+#define TA_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT                                                          0x1
+#define TA_CE_EDC_LO__ADDRESS__SHIFT                                                                          0x2
+#define TA_CE_EDC_LO__MEM_ID__SHIFT                                                                           0x18
+#define TA_CE_EDC_LO__STATUS_VALID_FLAG_MASK                                                                  0x00000001L
+#define TA_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK                                                            0x00000002L
+#define TA_CE_EDC_LO__ADDRESS_MASK                                                                            0x00FFFFFCL
+#define TA_CE_EDC_LO__MEM_ID_MASK                                                                             0xFF000000L
+//TA_CE_EDC_HI
+#define TA_CE_EDC_HI__ECC__SHIFT                                                                              0x0
+#define TA_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT                                                              0x2
+#define TA_CE_EDC_HI__ERR_INFO__SHIFT                                                                         0x3
+#define TA_CE_EDC_HI__CE_CNT__SHIFT                                                                           0x17
+#define TA_CE_EDC_HI__POISON__SHIFT                                                                           0x1a
+#define TA_CE_EDC_HI__ECC_MASK                                                                                0x00000001L
+#define TA_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK                                                                0x00000004L
+#define TA_CE_EDC_HI__ERR_INFO_MASK                                                                           0x007FFFF8L
+#define TA_CE_EDC_HI__CE_CNT_MASK                                                                             0x03800000L
+#define TA_CE_EDC_HI__POISON_MASK                                                                             0x04000000L
 
 
 // addressBlock: xcd0_gc_gdsdec
@@ -5015,6 +5381,54 @@
 #define GDS_WD_GDS_CSB__UNUSED__SHIFT                                                                         0xd
 #define GDS_WD_GDS_CSB__COUNTER_MASK                                                                          0x00001FFFL
 #define GDS_WD_GDS_CSB__UNUSED_MASK                                                                           0xFFFFE000L
+//GDS_UE_ERR_STATUS_LO
+#define GDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define GDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define GDS_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define GDS_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define GDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define GDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define GDS_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define GDS_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//GDS_UE_ERR_STATUS_HI
+#define GDS_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define GDS_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define GDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define GDS_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define GDS_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define GDS_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define GDS_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define GDS_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define GDS_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define GDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define GDS_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define GDS_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define GDS_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define GDS_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//GDS_CE_ERR_STATUS_LO
+#define GDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define GDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define GDS_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define GDS_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define GDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define GDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define GDS_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define GDS_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//GDS_CE_ERR_STATUS_HI
+#define GDS_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define GDS_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define GDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define GDS_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define GDS_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define GDS_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define GDS_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define GDS_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define GDS_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define GDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define GDS_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define GDS_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define GDS_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define GDS_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
 
 
 // addressBlock: xcd0_gc_rbdec
@@ -7370,6 +7784,30 @@
 #define GCEA_MAM_CTRL2__ARAM_FLUSH_NOALLOC_MASK                                                               0x00000040L
 #define GCEA_MAM_CTRL2__RESERVED_FIELD_MASK                                                                   0x00FFFF80L
 #define GCEA_MAM_CTRL2__ADDR_HI_MASK                                                                          0xFF000000L
+//GCEA_UE_ERR_STATUS_LO
+#define GCEA_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                       0x0
+#define GCEA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define GCEA_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                 0x2
+#define GCEA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                               0x18
+#define GCEA_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                         0x00000001L
+#define GCEA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define GCEA_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define GCEA_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                 0xFF000000L
+//GCEA_UE_ERR_STATUS_HI
+#define GCEA_UE_ERR_STATUS_HI__ECC__SHIFT                                                                     0x0
+#define GCEA_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                  0x1
+#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                     0x2
+#define GCEA_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                0x3
+#define GCEA_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                  0x17
+#define GCEA_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                 0x1a
+#define GCEA_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT                                                          0x1d
+#define GCEA_UE_ERR_STATUS_HI__ECC_MASK                                                                       0x00000001L
+#define GCEA_UE_ERR_STATUS_HI__PARITY_MASK                                                                    0x00000002L
+#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                       0x00000004L
+#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                  0x007FFFF8L
+#define GCEA_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                    0x03800000L
+#define GCEA_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                   0x1C000000L
+#define GCEA_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK                                                            0xE0000000L
 //GCEA_DSM_CNTL
 #define GCEA_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT                                                0x0
 #define GCEA_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT                                               0x2
@@ -7745,6 +8183,30 @@
 #define GCEA_SDP_BACKDOOR_MISCCREDITS__WRRSP_CREDITS_RELEASED_MASK                                            0x0000FF00L
 #define GCEA_SDP_BACKDOOR_MISCCREDITS__PRB_REQ_CREDITS_RELEASED_MASK                                          0x007F0000L
 #define GCEA_SDP_BACKDOOR_MISCCREDITS__PRB_RSP_CREDITS_RECEIVED_MASK                                          0x3F800000L
+//GCEA_CE_ERR_STATUS_LO
+#define GCEA_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT                                                       0x0
+#define GCEA_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define GCEA_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                 0x2
+#define GCEA_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                               0x18
+#define GCEA_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK                                                         0x00000001L
+#define GCEA_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define GCEA_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define GCEA_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                 0xFF000000L
+//GCEA_CE_ERR_STATUS_HI
+#define GCEA_CE_ERR_STATUS_HI__ECC__SHIFT                                                                     0x0
+#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT                                                         0x1
+#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                     0x2
+#define GCEA_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                0x3
+#define GCEA_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                  0x17
+#define GCEA_CE_ERR_STATUS_HI__POISON__SHIFT                                                                  0x1a
+#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT                                                         0x1b
+#define GCEA_CE_ERR_STATUS_HI__ECC_MASK                                                                       0x00000001L
+#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK                                                           0x00000002L
+#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                       0x00000004L
+#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                  0x007FFFF8L
+#define GCEA_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                    0x03800000L
+#define GCEA_CE_ERR_STATUS_HI__POISON_MASK                                                                    0x04000000L
+#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK                                                           0xF8000000L
 //GCEA_SDP_ENABLE
 #define GCEA_SDP_ENABLE__ENABLE__SHIFT                                                                        0x0
 #define GCEA_SDP_ENABLE__ENABLE_MASK                                                                          0x00000001L
@@ -8440,6 +8902,54 @@
 //ATC_L2_MM_GROUP_RT_CLASSES
 #define ATC_L2_MM_GROUP_RT_CLASSES__GROUP_RT_CLASS__SHIFT                                                     0x0
 #define ATC_L2_MM_GROUP_RT_CLASSES__GROUP_RT_CLASS_MASK                                                       0xFFFFFFFFL
+//ATC_L2_UE_ERR_STATUS_LO
+#define ATC_L2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                 0x0
+#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                    0x1
+#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                               0x2
+#define ATC_L2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                             0x18
+#define ATC_L2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                   0x00000001L
+#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                      0x00000002L
+#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                 0x00FFFFFCL
+#define ATC_L2_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                               0xFF000000L
+//ATC_L2_UE_ERR_STATUS_HI
+#define ATC_L2_UE_ERR_STATUS_HI__ECC__SHIFT                                                                   0x0
+#define ATC_L2_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                0x1
+#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                   0x2
+#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                              0x3
+#define ATC_L2_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                0x17
+#define ATC_L2_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                               0x1a
+#define ATC_L2_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                              0x1d
+#define ATC_L2_UE_ERR_STATUS_HI__ECC_MASK                                                                     0x00000001L
+#define ATC_L2_UE_ERR_STATUS_HI__PARITY_MASK                                                                  0x00000002L
+#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                     0x00000004L
+#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                0x007FFFF8L
+#define ATC_L2_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                  0x03800000L
+#define ATC_L2_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                 0x1C000000L
+#define ATC_L2_UE_ERR_STATUS_HI__RESERVED_MASK                                                                0x60000000L
+//ATC_L2_CE_ERR_STATUS_LO
+#define ATC_L2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                 0x0
+#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                    0x1
+#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                               0x2
+#define ATC_L2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                             0x18
+#define ATC_L2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                   0x00000001L
+#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                      0x00000002L
+#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                 0x00FFFFFCL
+#define ATC_L2_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                               0xFF000000L
+//ATC_L2_CE_ERR_STATUS_HI
+#define ATC_L2_CE_ERR_STATUS_HI__ECC__SHIFT                                                                   0x0
+#define ATC_L2_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                 0x1
+#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                   0x2
+#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                              0x3
+#define ATC_L2_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                0x17
+#define ATC_L2_CE_ERR_STATUS_HI__POISON__SHIFT                                                                0x1a
+#define ATC_L2_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                              0x1b
+#define ATC_L2_CE_ERR_STATUS_HI__ECC_MASK                                                                     0x00000001L
+#define ATC_L2_CE_ERR_STATUS_HI__OTHER_MASK                                                                   0x00000002L
+#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                     0x00000004L
+#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                0x007FFFF8L
+#define ATC_L2_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                  0x03800000L
+#define ATC_L2_CE_ERR_STATUS_HI__POISON_MASK                                                                  0x04000000L
+#define ATC_L2_CE_ERR_STATUS_HI__RESERVED_MASK                                                                0xF8000000L
 
 
 // addressBlock: xcd0_gc_utcl2_vml2pfdec
@@ -8888,6 +9398,150 @@
 #define UTCL2_EDC_CONFIG__DIS_EDC__SHIFT                                                                      0x1
 #define UTCL2_EDC_CONFIG__WRITE_DIS_MASK                                                                      0x00000001L
 #define UTCL2_EDC_CONFIG__DIS_EDC_MASK                                                                        0x00000002L
+//VML2_UE_ERR_STATUS_LO
+#define VML2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                   0x0
+#define VML2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define VML2_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                 0x2
+#define VML2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                               0x18
+#define VML2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                     0x00000001L
+#define VML2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define VML2_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define VML2_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                 0xFF000000L
+//VML2_WALKER_UE_ERR_STATUS_LO
+#define VML2_WALKER_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                            0x0
+#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                               0x1
+#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                          0x2
+#define VML2_WALKER_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                        0x18
+#define VML2_WALKER_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                              0x00000001L
+#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                 0x00000002L
+#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_MASK                                                            0x00FFFFFCL
+#define VML2_WALKER_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                          0xFF000000L
+//UTCL2_UE_ERR_STATUS_LO
+#define UTCL2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                  0x0
+#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define UTCL2_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define UTCL2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define UTCL2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                    0x00000001L
+#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define UTCL2_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//VML2_UE_ERR_STATUS_HI
+#define VML2_UE_ERR_STATUS_HI__ECC__SHIFT                                                                     0x0
+#define VML2_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                  0x1
+#define VML2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                     0x2
+#define VML2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                0x3
+#define VML2_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                  0x17
+#define VML2_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                 0x1a
+#define VML2_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                0x1d
+#define VML2_UE_ERR_STATUS_HI__ECC_MASK                                                                       0x00000001L
+#define VML2_UE_ERR_STATUS_HI__PARITY_MASK                                                                    0x00000002L
+#define VML2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                       0x00000004L
+#define VML2_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                  0x007FFFF8L
+#define VML2_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                    0x03800000L
+#define VML2_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                   0x1C000000L
+#define VML2_UE_ERR_STATUS_HI__RESERVED_MASK                                                                  0xE0000000L
+//VML2_WALKER_UE_ERR_STATUS_HI
+#define VML2_WALKER_UE_ERR_STATUS_HI__ECC__SHIFT                                                              0x0
+#define VML2_WALKER_UE_ERR_STATUS_HI__PARITY__SHIFT                                                           0x1
+#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                              0x2
+#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                         0x3
+#define VML2_WALKER_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                           0x17
+#define VML2_WALKER_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                          0x1a
+#define VML2_WALKER_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                         0x1d
+#define VML2_WALKER_UE_ERR_STATUS_HI__ECC_MASK                                                                0x00000001L
+#define VML2_WALKER_UE_ERR_STATUS_HI__PARITY_MASK                                                             0x00000002L
+#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                0x00000004L
+#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                           0x007FFFF8L
+#define VML2_WALKER_UE_ERR_STATUS_HI__UE_CNT_MASK                                                             0x03800000L
+#define VML2_WALKER_UE_ERR_STATUS_HI__FED_CNT_MASK                                                            0x1C000000L
+#define VML2_WALKER_UE_ERR_STATUS_HI__RESERVED_MASK                                                           0xE0000000L
+//UTCL2_UE_ERR_STATUS_HI
+#define UTCL2_UE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define UTCL2_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                 0x1
+#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define UTCL2_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                 0x17
+#define UTCL2_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                0x1a
+#define UTCL2_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                               0x1d
+#define UTCL2_UE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define UTCL2_UE_ERR_STATUS_HI__PARITY_MASK                                                                   0x00000002L
+#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define UTCL2_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                   0x03800000L
+#define UTCL2_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                  0x1C000000L
+#define UTCL2_UE_ERR_STATUS_HI__RESERVED_MASK                                                                 0xE0000000L
+//VML2_CE_ERR_STATUS_LO
+#define VML2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                   0x0
+#define VML2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define VML2_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                 0x2
+#define VML2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                               0x18
+#define VML2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                     0x00000001L
+#define VML2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define VML2_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define VML2_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                 0xFF000000L
+//VML2_WALKER_CE_ERR_STATUS_LO
+#define VML2_WALKER_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                            0x0
+#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                               0x1
+#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                          0x2
+#define VML2_WALKER_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                        0x18
+#define VML2_WALKER_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                              0x00000001L
+#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                 0x00000002L
+#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_MASK                                                            0x00FFFFFCL
+#define VML2_WALKER_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                          0xFF000000L
+//UTCL2_CE_ERR_STATUS_LO
+#define UTCL2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                  0x0
+#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                     0x1
+#define UTCL2_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                0x2
+#define UTCL2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                              0x18
+#define UTCL2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                    0x00000001L
+#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                       0x00000002L
+#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                  0x00FFFFFCL
+#define UTCL2_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                0xFF000000L
+//VML2_CE_ERR_STATUS_HI
+#define VML2_CE_ERR_STATUS_HI__ECC__SHIFT                                                                     0x0
+#define VML2_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                   0x1
+#define VML2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                     0x2
+#define VML2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                0x3
+#define VML2_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                  0x17
+#define VML2_CE_ERR_STATUS_HI__POISON__SHIFT                                                                  0x1a
+#define VML2_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                0x1b
+#define VML2_CE_ERR_STATUS_HI__ECC_MASK                                                                       0x00000001L
+#define VML2_CE_ERR_STATUS_HI__OTHER_MASK                                                                     0x00000002L
+#define VML2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                       0x00000004L
+#define VML2_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                  0x007FFFF8L
+#define VML2_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                    0x03800000L
+#define VML2_CE_ERR_STATUS_HI__POISON_MASK                                                                    0x04000000L
+#define VML2_CE_ERR_STATUS_HI__RESERVED_MASK                                                                  0xF8000000L
+//VML2_WALKER_CE_ERR_STATUS_HI
+#define VML2_WALKER_CE_ERR_STATUS_HI__ECC__SHIFT                                                              0x0
+#define VML2_WALKER_CE_ERR_STATUS_HI__OTHER__SHIFT                                                            0x1
+#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                              0x2
+#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                         0x3
+#define VML2_WALKER_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                           0x17
+#define VML2_WALKER_CE_ERR_STATUS_HI__POISON__SHIFT                                                           0x1a
+#define VML2_WALKER_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                         0x1b
+#define VML2_WALKER_CE_ERR_STATUS_HI__ECC_MASK                                                                0x00000001L
+#define VML2_WALKER_CE_ERR_STATUS_HI__OTHER_MASK                                                              0x00000002L
+#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                0x00000004L
+#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                           0x007FFFF8L
+#define VML2_WALKER_CE_ERR_STATUS_HI__CE_CNT_MASK                                                             0x03800000L
+#define VML2_WALKER_CE_ERR_STATUS_HI__POISON_MASK                                                             0x04000000L
+#define VML2_WALKER_CE_ERR_STATUS_HI__RESERVED_MASK                                                           0xF8000000L
+//UTCL2_CE_ERR_STATUS_HI
+#define UTCL2_CE_ERR_STATUS_HI__ECC__SHIFT                                                                    0x0
+#define UTCL2_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                  0x1
+#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                               0x3
+#define UTCL2_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                 0x17
+#define UTCL2_CE_ERR_STATUS_HI__POISON__SHIFT                                                                 0x1a
+#define UTCL2_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                               0x1b
+#define UTCL2_CE_ERR_STATUS_HI__ECC_MASK                                                                      0x00000001L
+#define UTCL2_CE_ERR_STATUS_HI__OTHER_MASK                                                                    0x00000002L
+#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define UTCL2_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                   0x03800000L
+#define UTCL2_CE_ERR_STATUS_HI__POISON_MASK                                                                   0x04000000L
+#define UTCL2_CE_ERR_STATUS_HI__RESERVED_MASK                                                                 0xF8000000L
 
 
 // addressBlock: xcd0_gc_utcl2_vml2vcdec
@@ -11139,6 +11793,98 @@
 //TC_CFG_L2_VOLATILE
 #define TC_CFG_L2_VOLATILE__VOL__SHIFT                                                                        0x0
 #define TC_CFG_L2_VOLATILE__VOL_MASK                                                                          0x0000000FL
+//TCP_UE_EDC_HI_REG
+#define TCP_UE_EDC_HI_REG__ECC__SHIFT                                                                         0x0
+#define TCP_UE_EDC_HI_REG__PARITY__SHIFT                                                                      0x1
+#define TCP_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT                                                         0x2
+#define TCP_UE_EDC_HI_REG__ERR_INFO__SHIFT                                                                    0x3
+#define TCP_UE_EDC_HI_REG__UE_CNT__SHIFT                                                                      0x17
+#define TCP_UE_EDC_HI_REG__FED_CNT__SHIFT                                                                     0x1a
+#define TCP_UE_EDC_HI_REG__RESERVED__SHIFT                                                                    0x1d
+#define TCP_UE_EDC_HI_REG__ECC_MASK                                                                           0x00000001L
+#define TCP_UE_EDC_HI_REG__PARITY_MASK                                                                        0x00000002L
+#define TCP_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK                                                           0x00000004L
+#define TCP_UE_EDC_HI_REG__ERR_INFO_MASK                                                                      0x007FFFF8L
+#define TCP_UE_EDC_HI_REG__UE_CNT_MASK                                                                        0x03800000L
+#define TCP_UE_EDC_HI_REG__FED_CNT_MASK                                                                       0x1C000000L
+#define TCP_UE_EDC_HI_REG__RESERVED_MASK                                                                      0xE0000000L
+//TCP_UE_EDC_LO_REG
+#define TCP_UE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT                                                           0x0
+#define TCP_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT                                                     0x1
+#define TCP_UE_EDC_LO_REG__ADDRESS__SHIFT                                                                     0x2
+#define TCP_UE_EDC_LO_REG__MEM_ID__SHIFT                                                                      0x18
+#define TCP_UE_EDC_LO_REG__STATUS_VALID_FLAG_MASK                                                             0x00000001L
+#define TCP_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK                                                       0x00000002L
+#define TCP_UE_EDC_LO_REG__ADDRESS_MASK                                                                       0x00FFFFFCL
+#define TCP_UE_EDC_LO_REG__MEM_ID_MASK                                                                        0xFF000000L
+//TCP_CE_EDC_HI_REG
+#define TCP_CE_EDC_HI_REG__ECC__SHIFT                                                                         0x0
+#define TCP_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT                                                         0x2
+#define TCP_CE_EDC_HI_REG__ERR_INFO__SHIFT                                                                    0x3
+#define TCP_CE_EDC_HI_REG__CE_CNT__SHIFT                                                                      0x17
+#define TCP_CE_EDC_HI_REG__POISON__SHIFT                                                                      0x1a
+#define TCP_CE_EDC_HI_REG__RESERVED__SHIFT                                                                    0x1b
+#define TCP_CE_EDC_HI_REG__ECC_MASK                                                                           0x00000001L
+#define TCP_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK                                                           0x00000004L
+#define TCP_CE_EDC_HI_REG__ERR_INFO_MASK                                                                      0x007FFFF8L
+#define TCP_CE_EDC_HI_REG__CE_CNT_MASK                                                                        0x03800000L
+#define TCP_CE_EDC_HI_REG__POISON_MASK                                                                        0x04000000L
+#define TCP_CE_EDC_HI_REG__RESERVED_MASK                                                                      0xF8000000L
+//TCP_CE_EDC_LO_REG
+#define TCP_CE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT                                                           0x0
+#define TCP_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT                                                     0x1
+#define TCP_CE_EDC_LO_REG__ADDRESS__SHIFT                                                                     0x2
+#define TCP_CE_EDC_LO_REG__MEM_ID__SHIFT                                                                      0x18
+#define TCP_CE_EDC_LO_REG__STATUS_VALID_FLAG_MASK                                                             0x00000001L
+#define TCP_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK                                                       0x00000002L
+#define TCP_CE_EDC_LO_REG__ADDRESS_MASK                                                                       0x00FFFFFCL
+#define TCP_CE_EDC_LO_REG__MEM_ID_MASK                                                                        0xFF000000L
+//TCI_UE_EDC_HI_REG
+#define TCI_UE_EDC_HI_REG__ECC__SHIFT                                                                         0x0
+#define TCI_UE_EDC_HI_REG__PARITY__SHIFT                                                                      0x1
+#define TCI_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT                                                         0x2
+#define TCI_UE_EDC_HI_REG__ERR_INFO__SHIFT                                                                    0x3
+#define TCI_UE_EDC_HI_REG__UE_CNT__SHIFT                                                                      0x17
+#define TCI_UE_EDC_HI_REG__FED_CNT__SHIFT                                                                     0x1a
+#define TCI_UE_EDC_HI_REG__RESERVED__SHIFT                                                                    0x1d
+#define TCI_UE_EDC_HI_REG__ECC_MASK                                                                           0x00000001L
+#define TCI_UE_EDC_HI_REG__PARITY_MASK                                                                        0x00000002L
+#define TCI_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK                                                           0x00000004L
+#define TCI_UE_EDC_HI_REG__ERR_INFO_MASK                                                                      0x007FFFF8L
+#define TCI_UE_EDC_HI_REG__UE_CNT_MASK                                                                        0x03800000L
+#define TCI_UE_EDC_HI_REG__FED_CNT_MASK                                                                       0x1C000000L
+#define TCI_UE_EDC_HI_REG__RESERVED_MASK                                                                      0xE0000000L
+//TCI_UE_EDC_LO_REG
+#define TCI_UE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT                                                           0x0
+#define TCI_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT                                                     0x1
+#define TCI_UE_EDC_LO_REG__ADDRESS__SHIFT                                                                     0x2
+#define TCI_UE_EDC_LO_REG__MEM_ID__SHIFT                                                                      0x18
+#define TCI_UE_EDC_LO_REG__STATUS_VALID_FLAG_MASK                                                             0x00000001L
+#define TCI_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK                                                       0x00000002L
+#define TCI_UE_EDC_LO_REG__ADDRESS_MASK                                                                       0x00FFFFFCL
+#define TCI_UE_EDC_LO_REG__MEM_ID_MASK                                                                        0xFF000000L
+//TCI_CE_EDC_HI_REG
+#define TCI_CE_EDC_HI_REG__ECC__SHIFT                                                                         0x0
+#define TCI_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT                                                         0x2
+#define TCI_CE_EDC_HI_REG__ERR_INFO__SHIFT                                                                    0x3
+#define TCI_CE_EDC_HI_REG__CE_CNT__SHIFT                                                                      0x17
+#define TCI_CE_EDC_HI_REG__POISON__SHIFT                                                                      0x1a
+#define TCI_CE_EDC_HI_REG__RESERVED__SHIFT                                                                    0x1b
+#define TCI_CE_EDC_HI_REG__ECC_MASK                                                                           0x00000001L
+#define TCI_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK                                                           0x00000004L
+#define TCI_CE_EDC_HI_REG__ERR_INFO_MASK                                                                      0x007FFFF8L
+#define TCI_CE_EDC_HI_REG__CE_CNT_MASK                                                                        0x03800000L
+#define TCI_CE_EDC_HI_REG__POISON_MASK                                                                        0x04000000L
+#define TCI_CE_EDC_HI_REG__RESERVED_MASK                                                                      0xF8000000L
+//TCI_CE_EDC_LO_REG
+#define TCI_CE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT                                                           0x0
+#define TCI_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT                                                     0x1
+#define TCI_CE_EDC_LO_REG__ADDRESS__SHIFT                                                                     0x2
+#define TCI_CE_EDC_LO_REG__MEM_ID__SHIFT                                                                      0x18
+#define TCI_CE_EDC_LO_REG__STATUS_VALID_FLAG_MASK                                                             0x00000001L
+#define TCI_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK                                                       0x00000002L
+#define TCI_CE_EDC_LO_REG__ADDRESS_MASK                                                                       0x00FFFFFCL
+#define TCI_CE_EDC_LO_REG__MEM_ID_MASK                                                                        0xFF000000L
 //TCI_MISC
 #define TCI_MISC__FGCG_REPEATER_DISABLE__SHIFT                                                                0x0
 #define TCI_MISC__LEGACY_MGCG_DISABLE__SHIFT                                                                  0x1
@@ -11560,6 +12306,112 @@
 #define TCX_DSM_CNTL2__SED_ENABLE_ERROR_INJECT_MASK                                                           0x00000003L
 #define TCX_DSM_CNTL2__SED_SELECT_INJECT_DELAY_MASK                                                           0x00000004L
 #define TCX_DSM_CNTL2__INJECT_DELAY_MASK                                                                      0xFC000000L
+//TCA_UE_ERR_STATUS_LO
+#define TCA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define TCA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define TCA_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define TCA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define TCA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define TCA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define TCA_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define TCA_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//TCA_UE_ERR_STATUS_HI
+#define TCA_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define TCA_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define TCA_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT                                                               0x3
+#define TCA_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define TCA_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define TCA_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define TCA_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_MASK                                                                 0x007FFFF8L
+#define TCA_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define TCA_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+//TCX_UE_ERR_STATUS_LO
+#define TCX_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define TCX_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define TCX_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define TCX_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define TCX_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define TCX_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define TCX_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define TCX_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//TCX_UE_ERR_STATUS_HI
+#define TCX_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define TCX_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define TCX_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT                                                               0x3
+#define TCX_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define TCX_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define TCX_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define TCX_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_MASK                                                                 0x007FFFF8L
+#define TCX_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define TCX_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+//TCX_CE_ERR_STATUS_LO
+#define TCX_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define TCX_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define TCX_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define TCX_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define TCX_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define TCX_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define TCX_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define TCX_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//TCX_CE_ERR_STATUS_HI
+#define TCX_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define TCX_CE_ERR_STATUS_HI__ERROR_INFO__SHIFT                                                               0x3
+#define TCX_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define TCX_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define TCX_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_MASK                                                                 0x007FFFF8L
+#define TCX_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define TCX_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+//TCC_UE_ERR_STATUS_LO
+#define TCC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define TCC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define TCC_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define TCC_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define TCC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define TCC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define TCC_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define TCC_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//TCC_UE_ERR_STATUS_HI
+#define TCC_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define TCC_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define TCC_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT                                                               0x3
+#define TCC_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define TCC_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define TCC_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define TCC_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_MASK                                                                 0x007FFFF8L
+#define TCC_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define TCC_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+//TCC_CE_ERR_STATUS_LO
+#define TCC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define TCC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define TCC_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define TCC_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define TCC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define TCC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define TCC_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define TCC_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//TCC_CE_ERR_STATUS_HI
+#define TCC_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define TCC_CE_ERR_STATUS_HI__ERROR_INFO__SHIFT                                                               0x3
+#define TCC_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define TCC_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define TCC_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_MASK                                                                 0x007FFFF8L
+#define TCC_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define TCC_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
 
 
 // addressBlock: xcd0_gc_shdec
@@ -14384,6 +15236,150 @@
 #define CP_VMID_STATUS__PREEMPT_CE_STATUS__SHIFT                                                              0x10
 #define CP_VMID_STATUS__PREEMPT_DE_STATUS_MASK                                                                0x0000FFFFL
 #define CP_VMID_STATUS__PREEMPT_CE_STATUS_MASK                                                                0xFFFF0000L
+//CPC_UE_ERR_STATUS_LO
+#define CPC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPC_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPC_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPC_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPC_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPC_UE_ERR_STATUS_HI
+#define CPC_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPC_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define CPC_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPC_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPC_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define CPC_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define CPC_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define CPC_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPC_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define CPC_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPC_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPC_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define CPC_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define CPC_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//CPC_CE_ERR_STATUS_LO
+#define CPC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPC_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPC_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPC_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPC_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPC_CE_ERR_STATUS_HI
+#define CPC_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPC_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define CPC_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPC_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPC_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define CPC_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define CPC_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define CPC_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPC_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define CPC_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPC_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPC_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define CPC_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define CPC_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
+//CPF_UE_ERR_STATUS_LO
+#define CPF_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPF_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPF_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPF_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPF_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPF_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPF_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPF_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPF_UE_ERR_STATUS_HI
+#define CPF_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPF_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define CPF_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPF_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPF_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define CPF_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define CPF_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define CPF_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPF_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define CPF_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPF_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPF_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define CPF_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define CPF_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//CPF_CE_ERR_STATUS_LO
+#define CPF_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPF_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPF_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPF_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPF_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPF_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPF_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPF_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPF_CE_ERR_STATUS_HI
+#define CPF_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPF_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define CPF_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPF_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPF_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define CPF_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define CPF_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define CPF_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPF_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define CPF_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPF_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPF_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define CPF_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define CPF_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
+//CPG_UE_ERR_STATUS_LO
+#define CPG_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPG_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPG_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPG_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPG_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPG_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPG_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPG_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPG_UE_ERR_STATUS_HI
+#define CPG_UE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPG_UE_ERR_STATUS_HI__PARITY__SHIFT                                                                   0x1
+#define CPG_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPG_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPG_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                                   0x17
+#define CPG_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                                  0x1a
+#define CPG_UE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1d
+#define CPG_UE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPG_UE_ERR_STATUS_HI__PARITY_MASK                                                                     0x00000002L
+#define CPG_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPG_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPG_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                     0x03800000L
+#define CPG_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                    0x1C000000L
+#define CPG_UE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xE0000000L
+//CPG_CE_ERR_STATUS_LO
+#define CPG_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                    0x0
+#define CPG_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                       0x1
+#define CPG_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                                  0x2
+#define CPG_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                                0x18
+#define CPG_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                      0x00000001L
+#define CPG_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                         0x00000002L
+#define CPG_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                    0x00FFFFFCL
+#define CPG_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                                  0xFF000000L
+//CPG_CE_ERR_STATUS_HI
+#define CPG_CE_ERR_STATUS_HI__ECC__SHIFT                                                                      0x0
+#define CPG_CE_ERR_STATUS_HI__OTHER__SHIFT                                                                    0x1
+#define CPG_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                      0x2
+#define CPG_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                                 0x3
+#define CPG_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                                   0x17
+#define CPG_CE_ERR_STATUS_HI__POISON__SHIFT                                                                   0x1a
+#define CPG_CE_ERR_STATUS_HI__RESERVED__SHIFT                                                                 0x1b
+#define CPG_CE_ERR_STATUS_HI__ECC_MASK                                                                        0x00000001L
+#define CPG_CE_ERR_STATUS_HI__OTHER_MASK                                                                      0x00000002L
+#define CPG_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                        0x00000004L
+#define CPG_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                                   0x007FFFF8L
+#define CPG_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                     0x03800000L
+#define CPG_CE_ERR_STATUS_HI__POISON_MASK                                                                     0x04000000L
+#define CPG_CE_ERR_STATUS_HI__RESERVED_MASK                                                                   0xF8000000L
 
 
 // addressBlock: xcd0_gc_cppdec2
@@ -22764,6 +23760,74 @@
 #define SPI_WAVE_LIMIT_CNTL__GS_WAVE_GRAN_MASK                                                                0x00000030L
 #define SPI_WAVE_LIMIT_CNTL__HS_WAVE_GRAN_MASK                                                                0x000000C0L
 
+// addressBlock: xcd0_gc_gccanedec
+//GC_CANE_ERR_STATUS
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_STATUS__SHIFT                                                          0x0
+#define GC_CANE_ERR_STATUS__SDPM_WRRSP_STATUS__SHIFT                                                          0x4
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS__SHIFT                                                      0x8
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR__SHIFT                                                0xa
+#define GC_CANE_ERR_STATUS__SDPS_DAT_ERROR__SHIFT                                                             0xb
+#define GC_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR__SHIFT                                                      0xc
+#define GC_CANE_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT                                                         0xd
+#define GC_CANE_ERR_STATUS__BUSY_ON_ERROR__SHIFT                                                              0xe
+#define GC_CANE_ERR_STATUS__BUSY_ON_UER_ERROR__SHIFT                                                          0xf
+#define GC_CANE_ERR_STATUS__FUE_FLAG__SHIFT                                                                   0x10
+#define GC_CANE_ERR_STATUS__INTERRUPT_ON_FATAL__SHIFT                                                         0x11
+#define GC_CANE_ERR_STATUS__LEVEL_INTERRUPT__SHIFT                                                            0x12
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_STATUS_MASK                                                            0x0000000FL
+#define GC_CANE_ERR_STATUS__SDPM_WRRSP_STATUS_MASK                                                            0x000000F0L
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS_MASK                                                        0x00000300L
+#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR_MASK                                                  0x00000400L
+#define GC_CANE_ERR_STATUS__SDPS_DAT_ERROR_MASK                                                               0x00000800L
+#define GC_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR_MASK                                                        0x00001000L
+#define GC_CANE_ERR_STATUS__CLEAR_ERROR_STATUS_MASK                                                           0x00002000L
+#define GC_CANE_ERR_STATUS__BUSY_ON_ERROR_MASK                                                                0x00004000L
+#define GC_CANE_ERR_STATUS__BUSY_ON_UER_ERROR_MASK                                                            0x00008000L
+#define GC_CANE_ERR_STATUS__FUE_FLAG_MASK                                                                     0x00010000L
+#define GC_CANE_ERR_STATUS__INTERRUPT_ON_FATAL_MASK                                                           0x00020000L
+#define GC_CANE_ERR_STATUS__LEVEL_INTERRUPT_MASK                                                              0x00040000L
+//GC_CANE_UE_ERR_STATUS_LO
+#define GC_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                0x0
+#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                   0x1
+#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS__SHIFT                                                              0x2
+#define GC_CANE_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                            0x18
+#define GC_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                  0x00000001L
+#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                     0x00000002L
+#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_MASK                                                                0x00FFFFFCL
+#define GC_CANE_UE_ERR_STATUS_LO__MEMORY_ID_MASK                                                              0xFF000000L
+//GC_CANE_UE_ERR_STATUS_HI
+#define GC_CANE_UE_ERR_STATUS_HI__ECC__SHIFT                                                                  0x0
+#define GC_CANE_UE_ERR_STATUS_HI__PARITY__SHIFT                                                               0x1
+#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                  0x2
+#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                             0x3
+#define GC_CANE_UE_ERR_STATUS_HI__UE_CNT__SHIFT                                                               0x17
+#define GC_CANE_UE_ERR_STATUS_HI__FED_CNT__SHIFT                                                              0x1a
+#define GC_CANE_UE_ERR_STATUS_HI__ECC_MASK                                                                    0x00000001L
+#define GC_CANE_UE_ERR_STATUS_HI__PARITY_MASK                                                                 0x00000002L
+#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                    0x00000004L
+#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_MASK                                                               0x007FFFF8L
+#define GC_CANE_UE_ERR_STATUS_HI__UE_CNT_MASK                                                                 0x03800000L
+#define GC_CANE_UE_ERR_STATUS_HI__FED_CNT_MASK                                                                0x1C000000L
+//GC_CANE_CE_ERR_STATUS_LO
+#define GC_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT                                                0x0
+#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT                                                   0x1
+#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS__SHIFT                                                              0x2
+#define GC_CANE_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT                                                            0x18
+#define GC_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK                                                  0x00000001L
+#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK                                                     0x00000002L
+#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_MASK                                                                0x00FFFFFCL
+#define GC_CANE_CE_ERR_STATUS_LO__MEMORY_ID_MASK                                                              0xFF000000L
+//GC_CANE_CE_ERR_STATUS_HI
+#define GC_CANE_CE_ERR_STATUS_HI__ECC__SHIFT                                                                  0x0
+#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT                                                  0x2
+#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO__SHIFT                                                             0x3
+#define GC_CANE_CE_ERR_STATUS_HI__CE_CNT__SHIFT                                                               0x17
+#define GC_CANE_CE_ERR_STATUS_HI__POISON__SHIFT                                                               0x1a
+#define GC_CANE_CE_ERR_STATUS_HI__ECC_MASK                                                                    0x00000001L
+#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK                                                    0x00000004L
+#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_MASK                                                               0x007FFFF8L
+#define GC_CANE_CE_ERR_STATUS_HI__CE_CNT_MASK                                                                 0x03800000L
+#define GC_CANE_CE_ERR_STATUS_HI__POISON_MASK                                                                 0x04000000L
 
 // addressBlock: xcd0_gc_perfddec
 //CPG_PERFCOUNTER1_LO
@@ -26471,6 +27535,30 @@
 //RLC_CPG_STAT_INVAL
 #define RLC_CPG_STAT_INVAL__CPG_stat_inval__SHIFT                                                             0x0
 #define RLC_CPG_STAT_INVAL__CPG_stat_inval_MASK                                                               0x00000001L
+//RLC_UE_ERR_STATUS_LOW
+#define RLC_UE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG__SHIFT                                                   0x0
+#define RLC_UE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define RLC_UE_ERR_STATUS_LOW__ADDRESS__SHIFT                                                                 0x2
+#define RLC_UE_ERR_STATUS_LOW__MEMORY_ID__SHIFT                                                               0x18
+#define RLC_UE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG_MASK                                                     0x00000001L
+#define RLC_UE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define RLC_UE_ERR_STATUS_LOW__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define RLC_UE_ERR_STATUS_LOW__MEMORY_ID_MASK                                                                 0xFF000000L
+//RLC_UE_ERR_STATUS_HIGH
+#define RLC_UE_ERR_STATUS_HIGH__ECC__SHIFT                                                                    0x0
+#define RLC_UE_ERR_STATUS_HIGH__PARITY__SHIFT                                                                 0x1
+#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO__SHIFT                                                               0x3
+#define RLC_UE_ERR_STATUS_HIGH__UE_CNT__SHIFT                                                                 0x17
+#define RLC_UE_ERR_STATUS_HIGH__FED_CNT__SHIFT                                                                0x1a
+#define RLC_UE_ERR_STATUS_HIGH__RESERVED__SHIFT                                                               0x1d
+#define RLC_UE_ERR_STATUS_HIGH__ECC_MASK                                                                      0x00000001L
+#define RLC_UE_ERR_STATUS_HIGH__PARITY_MASK                                                                   0x00000002L
+#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define RLC_UE_ERR_STATUS_HIGH__UE_CNT_MASK                                                                   0x03800000L
+#define RLC_UE_ERR_STATUS_HIGH__FED_CNT_MASK                                                                  0x1C000000L
+#define RLC_UE_ERR_STATUS_HIGH__RESERVED_MASK                                                                 0xE0000000L
 //RLC_DSM_CNTL
 #define RLC_DSM_CNTL__RLCG_INSTR_RAM_IRRITATOR_DATA_SEL__SHIFT                                                0x0
 #define RLC_DSM_CNTL__RLCG_INSTR_RAM_IRRITATOR_SINGLE_WRITE__SHIFT                                            0x2
@@ -26573,6 +27661,30 @@
 #define RLC_DSM_CNTL2A__RLC_SPM_SE2_SCRATCH_RAM_SELECT_INJECT_DELAY_MASK                                      0x00000100L
 #define RLC_DSM_CNTL2A__RLC_SPM_SE3_SCRATCH_RAM_ENABLE_ERROR_INJECT_MASK                                      0x00000600L
 #define RLC_DSM_CNTL2A__RLC_SPM_SE3_SCRATCH_RAM_SELECT_INJECT_DELAY_MASK                                      0x00000800L
+//RLC_CE_ERR_STATUS_LOW
+#define RLC_CE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG__SHIFT                                                   0x0
+#define RLC_CE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG__SHIFT                                                      0x1
+#define RLC_CE_ERR_STATUS_LOW__ADDRESS__SHIFT                                                                 0x2
+#define RLC_CE_ERR_STATUS_LOW__MEMORY_ID__SHIFT                                                               0x18
+#define RLC_CE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG_MASK                                                     0x00000001L
+#define RLC_CE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG_MASK                                                        0x00000002L
+#define RLC_CE_ERR_STATUS_LOW__ADDRESS_MASK                                                                   0x00FFFFFCL
+#define RLC_CE_ERR_STATUS_LOW__MEMORY_ID_MASK                                                                 0xFF000000L
+//RLC_CE_ERR_STATUS_HIGH
+#define RLC_CE_ERR_STATUS_HIGH__ECC__SHIFT                                                                    0x0
+#define RLC_CE_ERR_STATUS_HIGH__OTHER__SHIFT                                                                  0x1
+#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG__SHIFT                                                    0x2
+#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO__SHIFT                                                               0x3
+#define RLC_CE_ERR_STATUS_HIGH__CE_CNT__SHIFT                                                                 0x17
+#define RLC_CE_ERR_STATUS_HIGH__POISON__SHIFT                                                                 0x1a
+#define RLC_CE_ERR_STATUS_HIGH__RESERVED__SHIFT                                                               0x1b
+#define RLC_CE_ERR_STATUS_HIGH__ECC_MASK                                                                      0x00000001L
+#define RLC_CE_ERR_STATUS_HIGH__OTHER_MASK                                                                    0x00000002L
+#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG_MASK                                                      0x00000004L
+#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_MASK                                                                 0x007FFFF8L
+#define RLC_CE_ERR_STATUS_HIGH__CE_CNT_MASK                                                                   0x03800000L
+#define RLC_CE_ERR_STATUS_HIGH__POISON_MASK                                                                   0x04000000L
+#define RLC_CE_ERR_STATUS_HIGH__RESERVED_MASK                                                                 0xF8000000L
 //RLC_RLCV_SPARE_INT
 #define RLC_RLCV_SPARE_INT__INTERRUPT__SHIFT                                                                  0x0
 #define RLC_RLCV_SPARE_INT__RESERVED__SHIFT                                                                   0x1
-- 
GitLab


From 77462ab8c62b0dc65261c042771efea44a111131 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Tue, 7 Feb 2023 18:30:55 +0800
Subject: [PATCH 1342/1791] drm/amdgpu: add RAS definitions for GFX

Add common GFX RAS definitions.

v2: remove instance from amdgpu_gfx_ras_reg_entry,
    amdgpu_ras_err_status_reg_entry has already defined it.
v3: remove memory id definitions from amdgpu_gfx.h, they are
    related to IP version.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 39 +++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 0df53fe7b1999..ce0f7a8ad4b87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -73,6 +73,32 @@ enum amdgpu_pkg_type {
 	AMDGPU_PKG_TYPE_UNKNOWN,
 };
 
+enum amdgpu_gfx_ras_mem_id_type {
+	AMDGPU_GFX_CP_MEM = 0,
+	AMDGPU_GFX_GCEA_MEM,
+	AMDGPU_GFX_GC_CANE_MEM,
+	AMDGPU_GFX_GCUTCL2_MEM,
+	AMDGPU_GFX_GDS_MEM,
+	AMDGPU_GFX_LDS_MEM,
+	AMDGPU_GFX_RLC_MEM,
+	AMDGPU_GFX_SP_MEM,
+	AMDGPU_GFX_SPI_MEM,
+	AMDGPU_GFX_SQC_MEM,
+	AMDGPU_GFX_SQ_MEM,
+	AMDGPU_GFX_TA_MEM,
+	AMDGPU_GFX_TCC_MEM,
+	AMDGPU_GFX_TCA_MEM,
+	AMDGPU_GFX_TCI_MEM,
+	AMDGPU_GFX_TCP_MEM,
+	AMDGPU_GFX_TD_MEM,
+	AMDGPU_GFX_TCX_MEM,
+	AMDGPU_GFX_ATC_L2_MEM,
+	AMDGPU_GFX_UTCL2_MEM,
+	AMDGPU_GFX_VML2_MEM,
+	AMDGPU_GFX_VML2_WALKER_MEM,
+	AMDGPU_GFX_MEM_TYPE_NUM
+};
+
 struct amdgpu_mec {
 	struct amdgpu_bo	*hpd_eop_obj;
 	u64			hpd_eop_gpu_addr;
@@ -410,6 +436,19 @@ struct amdgpu_gfx {
 	struct mutex			partition_mutex;
 };
 
+struct amdgpu_gfx_ras_reg_entry {
+	struct amdgpu_ras_err_status_reg_entry reg_entry;
+	enum amdgpu_gfx_ras_mem_id_type mem_id_type;
+	uint32_t se_num;
+};
+
+struct amdgpu_gfx_ras_mem_id_entry {
+	const struct amdgpu_ras_memory_id_entry *mem_id_ent;
+	uint32_t size;
+};
+
+#define AMDGPU_GFX_MEMID_ENT(x) {(x), ARRAY_SIZE(x)},
+
 #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
 #define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id)))
 #define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid, xcc_id) ((adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid), (xcc_id)))
-- 
GitLab


From 5c1c09a71634423604c47502d8059a5c098c6f40 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Mon, 6 Feb 2023 11:38:19 +0800
Subject: [PATCH 1343/1791] drm/amdgpu: add RAS error count definitions for
 gfx_v9_4_3

Prepare for the query of GFX RAS ce/ue count.

v2: remove xcp operation.
    only select_se_sh when instance number is more than 1.
v3: add more CE/UE registsers to query list.
    add check for se_num before select_se_sh.
    change instance from 0 to xcc_id for register access.
v4: move gfx memory id definitions to gfx_v9_4_3.
v5: create a dedicated patch for adding error count query function.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 740 ++++++++++++++++++++++++
 1 file changed, 740 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index e6069d081f71d..188b4d9a2cbb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2980,6 +2980,746 @@ static void gfx_v9_4_3_emit_wave_limit(struct amdgpu_ring *ring, bool enable)
 	}
 }
 
+enum amdgpu_gfx_cp_ras_mem_id {
+	AMDGPU_GFX_CP_MEM1 = 1,
+	AMDGPU_GFX_CP_MEM2,
+	AMDGPU_GFX_CP_MEM3,
+	AMDGPU_GFX_CP_MEM4,
+	AMDGPU_GFX_CP_MEM5,
+};
+
+enum amdgpu_gfx_gcea_ras_mem_id {
+	AMDGPU_GFX_GCEA_IOWR_CMDMEM = 4,
+	AMDGPU_GFX_GCEA_IORD_CMDMEM,
+	AMDGPU_GFX_GCEA_GMIWR_CMDMEM,
+	AMDGPU_GFX_GCEA_GMIRD_CMDMEM,
+	AMDGPU_GFX_GCEA_DRAMWR_CMDMEM,
+	AMDGPU_GFX_GCEA_DRAMRD_CMDMEM,
+	AMDGPU_GFX_GCEA_MAM_DMEM0,
+	AMDGPU_GFX_GCEA_MAM_DMEM1,
+	AMDGPU_GFX_GCEA_MAM_DMEM2,
+	AMDGPU_GFX_GCEA_MAM_DMEM3,
+	AMDGPU_GFX_GCEA_MAM_AMEM0,
+	AMDGPU_GFX_GCEA_MAM_AMEM1,
+	AMDGPU_GFX_GCEA_MAM_AMEM2,
+	AMDGPU_GFX_GCEA_MAM_AMEM3,
+	AMDGPU_GFX_GCEA_MAM_AFLUSH_BUFFER,
+	AMDGPU_GFX_GCEA_WRET_TAGMEM,
+	AMDGPU_GFX_GCEA_RRET_TAGMEM,
+	AMDGPU_GFX_GCEA_IOWR_DATAMEM,
+	AMDGPU_GFX_GCEA_GMIWR_DATAMEM,
+	AMDGPU_GFX_GCEA_DRAM_DATAMEM,
+};
+
+enum amdgpu_gfx_gc_cane_ras_mem_id {
+	AMDGPU_GFX_GC_CANE_MEM0 = 0,
+};
+
+enum amdgpu_gfx_gcutcl2_ras_mem_id {
+	AMDGPU_GFX_GCUTCL2_MEM2P512X95 = 160,
+};
+
+enum amdgpu_gfx_gds_ras_mem_id {
+	AMDGPU_GFX_GDS_MEM0 = 0,
+};
+
+enum amdgpu_gfx_lds_ras_mem_id {
+	AMDGPU_GFX_LDS_BANK0 = 0,
+	AMDGPU_GFX_LDS_BANK1,
+	AMDGPU_GFX_LDS_BANK2,
+	AMDGPU_GFX_LDS_BANK3,
+	AMDGPU_GFX_LDS_BANK4,
+	AMDGPU_GFX_LDS_BANK5,
+	AMDGPU_GFX_LDS_BANK6,
+	AMDGPU_GFX_LDS_BANK7,
+	AMDGPU_GFX_LDS_BANK8,
+	AMDGPU_GFX_LDS_BANK9,
+	AMDGPU_GFX_LDS_BANK10,
+	AMDGPU_GFX_LDS_BANK11,
+	AMDGPU_GFX_LDS_BANK12,
+	AMDGPU_GFX_LDS_BANK13,
+	AMDGPU_GFX_LDS_BANK14,
+	AMDGPU_GFX_LDS_BANK15,
+	AMDGPU_GFX_LDS_BANK16,
+	AMDGPU_GFX_LDS_BANK17,
+	AMDGPU_GFX_LDS_BANK18,
+	AMDGPU_GFX_LDS_BANK19,
+	AMDGPU_GFX_LDS_BANK20,
+	AMDGPU_GFX_LDS_BANK21,
+	AMDGPU_GFX_LDS_BANK22,
+	AMDGPU_GFX_LDS_BANK23,
+	AMDGPU_GFX_LDS_BANK24,
+	AMDGPU_GFX_LDS_BANK25,
+	AMDGPU_GFX_LDS_BANK26,
+	AMDGPU_GFX_LDS_BANK27,
+	AMDGPU_GFX_LDS_BANK28,
+	AMDGPU_GFX_LDS_BANK29,
+	AMDGPU_GFX_LDS_BANK30,
+	AMDGPU_GFX_LDS_BANK31,
+	AMDGPU_GFX_LDS_SP_BUFFER_A,
+	AMDGPU_GFX_LDS_SP_BUFFER_B,
+};
+
+enum amdgpu_gfx_rlc_ras_mem_id {
+	AMDGPU_GFX_RLC_GPMF32 = 1,
+	AMDGPU_GFX_RLC_RLCVF32,
+	AMDGPU_GFX_RLC_SCRATCH,
+	AMDGPU_GFX_RLC_SRM_ARAM,
+	AMDGPU_GFX_RLC_SRM_DRAM,
+	AMDGPU_GFX_RLC_TCTAG,
+	AMDGPU_GFX_RLC_SPM_SE,
+	AMDGPU_GFX_RLC_SPM_GRBMT,
+};
+
+enum amdgpu_gfx_sp_ras_mem_id {
+	AMDGPU_GFX_SP_SIMDID0 = 0,
+};
+
+enum amdgpu_gfx_spi_ras_mem_id {
+	AMDGPU_GFX_SPI_MEM0 = 0,
+	AMDGPU_GFX_SPI_MEM1,
+	AMDGPU_GFX_SPI_MEM2,
+	AMDGPU_GFX_SPI_MEM3,
+};
+
+enum amdgpu_gfx_sqc_ras_mem_id {
+	AMDGPU_GFX_SQC_INST_CACHE_A = 100,
+	AMDGPU_GFX_SQC_INST_CACHE_B = 101,
+	AMDGPU_GFX_SQC_INST_CACHE_TAG_A = 102,
+	AMDGPU_GFX_SQC_INST_CACHE_TAG_B = 103,
+	AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_A = 104,
+	AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_B = 105,
+	AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_A = 106,
+	AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_B = 107,
+	AMDGPU_GFX_SQC_DATA_CACHE_A = 200,
+	AMDGPU_GFX_SQC_DATA_CACHE_B = 201,
+	AMDGPU_GFX_SQC_DATA_CACHE_TAG_A = 202,
+	AMDGPU_GFX_SQC_DATA_CACHE_TAG_B = 203,
+	AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_A = 204,
+	AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_B = 205,
+	AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_A = 206,
+	AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_B = 207,
+	AMDGPU_GFX_SQC_DIRTY_BIT_A = 208,
+	AMDGPU_GFX_SQC_DIRTY_BIT_B = 209,
+	AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU0 = 210,
+	AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU1 = 211,
+	AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A = 212,
+	AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B = 213,
+	AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_INST_CACHE = 108,
+};
+
+enum amdgpu_gfx_sq_ras_mem_id {
+	AMDGPU_GFX_SQ_SGPR_MEM0 = 0,
+	AMDGPU_GFX_SQ_SGPR_MEM1,
+	AMDGPU_GFX_SQ_SGPR_MEM2,
+	AMDGPU_GFX_SQ_SGPR_MEM3,
+};
+
+enum amdgpu_gfx_ta_ras_mem_id {
+	AMDGPU_GFX_TA_FS_AFIFO_RAM_LO = 1,
+	AMDGPU_GFX_TA_FS_AFIFO_RAM_HI,
+	AMDGPU_GFX_TA_FS_CFIFO_RAM,
+	AMDGPU_GFX_TA_FSX_LFIFO,
+	AMDGPU_GFX_TA_FS_DFIFO_RAM,
+};
+
+enum amdgpu_gfx_tcc_ras_mem_id {
+	AMDGPU_GFX_TCC_MEM1 = 1,
+};
+
+enum amdgpu_gfx_tca_ras_mem_id {
+	AMDGPU_GFX_TCA_MEM1 = 1,
+};
+
+enum amdgpu_gfx_tci_ras_mem_id {
+	AMDGPU_GFX_TCIW_MEM = 1,
+};
+
+enum amdgpu_gfx_tcp_ras_mem_id {
+	AMDGPU_GFX_TCP_LFIFO0 = 1,
+	AMDGPU_GFX_TCP_SET0BANK0_RAM,
+	AMDGPU_GFX_TCP_SET0BANK1_RAM,
+	AMDGPU_GFX_TCP_SET0BANK2_RAM,
+	AMDGPU_GFX_TCP_SET0BANK3_RAM,
+	AMDGPU_GFX_TCP_SET1BANK0_RAM,
+	AMDGPU_GFX_TCP_SET1BANK1_RAM,
+	AMDGPU_GFX_TCP_SET1BANK2_RAM,
+	AMDGPU_GFX_TCP_SET1BANK3_RAM,
+	AMDGPU_GFX_TCP_SET2BANK0_RAM,
+	AMDGPU_GFX_TCP_SET2BANK1_RAM,
+	AMDGPU_GFX_TCP_SET2BANK2_RAM,
+	AMDGPU_GFX_TCP_SET2BANK3_RAM,
+	AMDGPU_GFX_TCP_SET3BANK0_RAM,
+	AMDGPU_GFX_TCP_SET3BANK1_RAM,
+	AMDGPU_GFX_TCP_SET3BANK2_RAM,
+	AMDGPU_GFX_TCP_SET3BANK3_RAM,
+	AMDGPU_GFX_TCP_VM_FIFO,
+	AMDGPU_GFX_TCP_DB_TAGRAM0,
+	AMDGPU_GFX_TCP_DB_TAGRAM1,
+	AMDGPU_GFX_TCP_DB_TAGRAM2,
+	AMDGPU_GFX_TCP_DB_TAGRAM3,
+	AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE0,
+	AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE1,
+	AMDGPU_GFX_TCP_CMD_FIFO,
+};
+
+enum amdgpu_gfx_td_ras_mem_id {
+	AMDGPU_GFX_TD_UTD_CS_FIFO_MEM = 1,
+	AMDGPU_GFX_TD_UTD_SS_FIFO_LO_MEM,
+	AMDGPU_GFX_TD_UTD_SS_FIFO_HI_MEM,
+};
+
+enum amdgpu_gfx_tcx_ras_mem_id {
+	AMDGPU_GFX_TCX_FIFOD0 = 0,
+	AMDGPU_GFX_TCX_FIFOD1,
+	AMDGPU_GFX_TCX_FIFOD2,
+	AMDGPU_GFX_TCX_FIFOD3,
+	AMDGPU_GFX_TCX_FIFOD4,
+	AMDGPU_GFX_TCX_FIFOD5,
+	AMDGPU_GFX_TCX_FIFOD6,
+	AMDGPU_GFX_TCX_FIFOD7,
+	AMDGPU_GFX_TCX_FIFOB0,
+	AMDGPU_GFX_TCX_FIFOB1,
+	AMDGPU_GFX_TCX_FIFOB2,
+	AMDGPU_GFX_TCX_FIFOB3,
+	AMDGPU_GFX_TCX_FIFOB4,
+	AMDGPU_GFX_TCX_FIFOB5,
+	AMDGPU_GFX_TCX_FIFOB6,
+	AMDGPU_GFX_TCX_FIFOB7,
+	AMDGPU_GFX_TCX_FIFOA0,
+	AMDGPU_GFX_TCX_FIFOA1,
+	AMDGPU_GFX_TCX_FIFOA2,
+	AMDGPU_GFX_TCX_FIFOA3,
+	AMDGPU_GFX_TCX_FIFOA4,
+	AMDGPU_GFX_TCX_FIFOA5,
+	AMDGPU_GFX_TCX_FIFOA6,
+	AMDGPU_GFX_TCX_FIFOA7,
+	AMDGPU_GFX_TCX_CFIFO0,
+	AMDGPU_GFX_TCX_CFIFO1,
+	AMDGPU_GFX_TCX_CFIFO2,
+	AMDGPU_GFX_TCX_CFIFO3,
+	AMDGPU_GFX_TCX_CFIFO4,
+	AMDGPU_GFX_TCX_CFIFO5,
+	AMDGPU_GFX_TCX_CFIFO6,
+	AMDGPU_GFX_TCX_CFIFO7,
+	AMDGPU_GFX_TCX_FIFO_ACKB0,
+	AMDGPU_GFX_TCX_FIFO_ACKB1,
+	AMDGPU_GFX_TCX_FIFO_ACKB2,
+	AMDGPU_GFX_TCX_FIFO_ACKB3,
+	AMDGPU_GFX_TCX_FIFO_ACKB4,
+	AMDGPU_GFX_TCX_FIFO_ACKB5,
+	AMDGPU_GFX_TCX_FIFO_ACKB6,
+	AMDGPU_GFX_TCX_FIFO_ACKB7,
+	AMDGPU_GFX_TCX_FIFO_ACKD0,
+	AMDGPU_GFX_TCX_FIFO_ACKD1,
+	AMDGPU_GFX_TCX_FIFO_ACKD2,
+	AMDGPU_GFX_TCX_FIFO_ACKD3,
+	AMDGPU_GFX_TCX_FIFO_ACKD4,
+	AMDGPU_GFX_TCX_FIFO_ACKD5,
+	AMDGPU_GFX_TCX_FIFO_ACKD6,
+	AMDGPU_GFX_TCX_FIFO_ACKD7,
+	AMDGPU_GFX_TCX_DST_FIFOA0,
+	AMDGPU_GFX_TCX_DST_FIFOA1,
+	AMDGPU_GFX_TCX_DST_FIFOA2,
+	AMDGPU_GFX_TCX_DST_FIFOA3,
+	AMDGPU_GFX_TCX_DST_FIFOA4,
+	AMDGPU_GFX_TCX_DST_FIFOA5,
+	AMDGPU_GFX_TCX_DST_FIFOA6,
+	AMDGPU_GFX_TCX_DST_FIFOA7,
+	AMDGPU_GFX_TCX_DST_FIFOB0,
+	AMDGPU_GFX_TCX_DST_FIFOB1,
+	AMDGPU_GFX_TCX_DST_FIFOB2,
+	AMDGPU_GFX_TCX_DST_FIFOB3,
+	AMDGPU_GFX_TCX_DST_FIFOB4,
+	AMDGPU_GFX_TCX_DST_FIFOB5,
+	AMDGPU_GFX_TCX_DST_FIFOB6,
+	AMDGPU_GFX_TCX_DST_FIFOB7,
+	AMDGPU_GFX_TCX_DST_FIFOD0,
+	AMDGPU_GFX_TCX_DST_FIFOD1,
+	AMDGPU_GFX_TCX_DST_FIFOD2,
+	AMDGPU_GFX_TCX_DST_FIFOD3,
+	AMDGPU_GFX_TCX_DST_FIFOD4,
+	AMDGPU_GFX_TCX_DST_FIFOD5,
+	AMDGPU_GFX_TCX_DST_FIFOD6,
+	AMDGPU_GFX_TCX_DST_FIFOD7,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB0,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB1,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB2,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB3,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB4,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB5,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB6,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKB7,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD0,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD1,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD2,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD3,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD4,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD5,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD6,
+	AMDGPU_GFX_TCX_DST_FIFO_ACKD7,
+};
+
+enum amdgpu_gfx_atc_l2_ras_mem_id {
+	AMDGPU_GFX_ATC_L2_MEM0 = 0,
+};
+
+enum amdgpu_gfx_utcl2_ras_mem_id {
+	AMDGPU_GFX_UTCL2_MEM0 = 0,
+};
+
+enum amdgpu_gfx_vml2_ras_mem_id {
+	AMDGPU_GFX_VML2_MEM0 = 0,
+};
+
+enum amdgpu_gfx_vml2_walker_ras_mem_id {
+	AMDGPU_GFX_VML2_WALKER_MEM0 = 0,
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_cp_mem_list[] = {
+	{AMDGPU_GFX_CP_MEM1, "CP_MEM1"},
+	{AMDGPU_GFX_CP_MEM2, "CP_MEM2"},
+	{AMDGPU_GFX_CP_MEM3, "CP_MEM3"},
+	{AMDGPU_GFX_CP_MEM4, "CP_MEM4"},
+	{AMDGPU_GFX_CP_MEM5, "CP_MEM5"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gcea_mem_list[] = {
+	{AMDGPU_GFX_GCEA_IOWR_CMDMEM, "GCEA_IOWR_CMDMEM"},
+	{AMDGPU_GFX_GCEA_IORD_CMDMEM, "GCEA_IORD_CMDMEM"},
+	{AMDGPU_GFX_GCEA_GMIWR_CMDMEM, "GCEA_GMIWR_CMDMEM"},
+	{AMDGPU_GFX_GCEA_GMIRD_CMDMEM, "GCEA_GMIRD_CMDMEM"},
+	{AMDGPU_GFX_GCEA_DRAMWR_CMDMEM, "GCEA_DRAMWR_CMDMEM"},
+	{AMDGPU_GFX_GCEA_DRAMRD_CMDMEM, "GCEA_DRAMRD_CMDMEM"},
+	{AMDGPU_GFX_GCEA_MAM_DMEM0, "GCEA_MAM_DMEM0"},
+	{AMDGPU_GFX_GCEA_MAM_DMEM1, "GCEA_MAM_DMEM1"},
+	{AMDGPU_GFX_GCEA_MAM_DMEM2, "GCEA_MAM_DMEM2"},
+	{AMDGPU_GFX_GCEA_MAM_DMEM3, "GCEA_MAM_DMEM3"},
+	{AMDGPU_GFX_GCEA_MAM_AMEM0, "GCEA_MAM_AMEM0"},
+	{AMDGPU_GFX_GCEA_MAM_AMEM1, "GCEA_MAM_AMEM1"},
+	{AMDGPU_GFX_GCEA_MAM_AMEM2, "GCEA_MAM_AMEM2"},
+	{AMDGPU_GFX_GCEA_MAM_AMEM3, "GCEA_MAM_AMEM3"},
+	{AMDGPU_GFX_GCEA_MAM_AFLUSH_BUFFER, "GCEA_MAM_AFLUSH_BUFFER"},
+	{AMDGPU_GFX_GCEA_WRET_TAGMEM, "GCEA_WRET_TAGMEM"},
+	{AMDGPU_GFX_GCEA_RRET_TAGMEM, "GCEA_RRET_TAGMEM"},
+	{AMDGPU_GFX_GCEA_IOWR_DATAMEM, "GCEA_IOWR_DATAMEM"},
+	{AMDGPU_GFX_GCEA_GMIWR_DATAMEM, "GCEA_GMIWR_DATAMEM"},
+	{AMDGPU_GFX_GCEA_DRAM_DATAMEM, "GCEA_DRAM_DATAMEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gc_cane_mem_list[] = {
+	{AMDGPU_GFX_GC_CANE_MEM0, "GC_CANE_MEM0"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gcutcl2_mem_list[] = {
+	{AMDGPU_GFX_GCUTCL2_MEM2P512X95, "GCUTCL2_MEM2P512X95"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gds_mem_list[] = {
+	{AMDGPU_GFX_GDS_MEM0, "GDS_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_lds_mem_list[] = {
+	{AMDGPU_GFX_LDS_BANK0, "LDS_BANK0"},
+	{AMDGPU_GFX_LDS_BANK1, "LDS_BANK1"},
+	{AMDGPU_GFX_LDS_BANK2, "LDS_BANK2"},
+	{AMDGPU_GFX_LDS_BANK3, "LDS_BANK3"},
+	{AMDGPU_GFX_LDS_BANK4, "LDS_BANK4"},
+	{AMDGPU_GFX_LDS_BANK5, "LDS_BANK5"},
+	{AMDGPU_GFX_LDS_BANK6, "LDS_BANK6"},
+	{AMDGPU_GFX_LDS_BANK7, "LDS_BANK7"},
+	{AMDGPU_GFX_LDS_BANK8, "LDS_BANK8"},
+	{AMDGPU_GFX_LDS_BANK9, "LDS_BANK9"},
+	{AMDGPU_GFX_LDS_BANK10, "LDS_BANK10"},
+	{AMDGPU_GFX_LDS_BANK11, "LDS_BANK11"},
+	{AMDGPU_GFX_LDS_BANK12, "LDS_BANK12"},
+	{AMDGPU_GFX_LDS_BANK13, "LDS_BANK13"},
+	{AMDGPU_GFX_LDS_BANK14, "LDS_BANK14"},
+	{AMDGPU_GFX_LDS_BANK15, "LDS_BANK15"},
+	{AMDGPU_GFX_LDS_BANK16, "LDS_BANK16"},
+	{AMDGPU_GFX_LDS_BANK17, "LDS_BANK17"},
+	{AMDGPU_GFX_LDS_BANK18, "LDS_BANK18"},
+	{AMDGPU_GFX_LDS_BANK19, "LDS_BANK19"},
+	{AMDGPU_GFX_LDS_BANK20, "LDS_BANK20"},
+	{AMDGPU_GFX_LDS_BANK21, "LDS_BANK21"},
+	{AMDGPU_GFX_LDS_BANK22, "LDS_BANK22"},
+	{AMDGPU_GFX_LDS_BANK23, "LDS_BANK23"},
+	{AMDGPU_GFX_LDS_BANK24, "LDS_BANK24"},
+	{AMDGPU_GFX_LDS_BANK25, "LDS_BANK25"},
+	{AMDGPU_GFX_LDS_BANK26, "LDS_BANK26"},
+	{AMDGPU_GFX_LDS_BANK27, "LDS_BANK27"},
+	{AMDGPU_GFX_LDS_BANK28, "LDS_BANK28"},
+	{AMDGPU_GFX_LDS_BANK29, "LDS_BANK29"},
+	{AMDGPU_GFX_LDS_BANK30, "LDS_BANK30"},
+	{AMDGPU_GFX_LDS_BANK31, "LDS_BANK31"},
+	{AMDGPU_GFX_LDS_SP_BUFFER_A, "LDS_SP_BUFFER_A"},
+	{AMDGPU_GFX_LDS_SP_BUFFER_B, "LDS_SP_BUFFER_B"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_rlc_mem_list[] = {
+	{AMDGPU_GFX_RLC_GPMF32, "RLC_GPMF32"},
+	{AMDGPU_GFX_RLC_RLCVF32, "RLC_RLCVF32"},
+	{AMDGPU_GFX_RLC_SCRATCH, "RLC_SCRATCH"},
+	{AMDGPU_GFX_RLC_SRM_ARAM, "RLC_SRM_ARAM"},
+	{AMDGPU_GFX_RLC_SRM_DRAM, "RLC_SRM_DRAM"},
+	{AMDGPU_GFX_RLC_TCTAG, "RLC_TCTAG"},
+	{AMDGPU_GFX_RLC_SPM_SE, "RLC_SPM_SE"},
+	{AMDGPU_GFX_RLC_SPM_GRBMT, "RLC_SPM_GRBMT"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sp_mem_list[] = {
+	{AMDGPU_GFX_SP_SIMDID0, "SP_SIMDID0"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_spi_mem_list[] = {
+	{AMDGPU_GFX_SPI_MEM0, "SPI_MEM0"},
+	{AMDGPU_GFX_SPI_MEM1, "SPI_MEM1"},
+	{AMDGPU_GFX_SPI_MEM2, "SPI_MEM2"},
+	{AMDGPU_GFX_SPI_MEM3, "SPI_MEM3"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sqc_mem_list[] = {
+	{AMDGPU_GFX_SQC_INST_CACHE_A, "SQC_INST_CACHE_A"},
+	{AMDGPU_GFX_SQC_INST_CACHE_B, "SQC_INST_CACHE_B"},
+	{AMDGPU_GFX_SQC_INST_CACHE_TAG_A, "SQC_INST_CACHE_TAG_A"},
+	{AMDGPU_GFX_SQC_INST_CACHE_TAG_B, "SQC_INST_CACHE_TAG_B"},
+	{AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_A, "SQC_INST_CACHE_MISS_FIFO_A"},
+	{AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_B, "SQC_INST_CACHE_MISS_FIFO_B"},
+	{AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_A, "SQC_INST_CACHE_GATCL1_MISS_FIFO_A"},
+	{AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_B, "SQC_INST_CACHE_GATCL1_MISS_FIFO_B"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_A, "SQC_DATA_CACHE_A"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_B, "SQC_DATA_CACHE_B"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_TAG_A, "SQC_DATA_CACHE_TAG_A"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_TAG_B, "SQC_DATA_CACHE_TAG_B"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_A, "SQC_DATA_CACHE_MISS_FIFO_A"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_B, "SQC_DATA_CACHE_MISS_FIFO_B"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_A, "SQC_DATA_CACHE_HIT_FIFO_A"},
+	{AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_B, "SQC_DATA_CACHE_HIT_FIFO_B"},
+	{AMDGPU_GFX_SQC_DIRTY_BIT_A, "SQC_DIRTY_BIT_A"},
+	{AMDGPU_GFX_SQC_DIRTY_BIT_B, "SQC_DIRTY_BIT_B"},
+	{AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU0, "SQC_WRITE_DATA_BUFFER_CU0"},
+	{AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU1, "SQC_WRITE_DATA_BUFFER_CU1"},
+	{AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A, "SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A"},
+	{AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B, "SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B"},
+	{AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_INST_CACHE, "SQC_UTCL1_MISS_LFIFO_INST_CACHE"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sq_mem_list[] = {
+	{AMDGPU_GFX_SQ_SGPR_MEM0, "SQ_SGPR_MEM0"},
+	{AMDGPU_GFX_SQ_SGPR_MEM1, "SQ_SGPR_MEM1"},
+	{AMDGPU_GFX_SQ_SGPR_MEM2, "SQ_SGPR_MEM2"},
+	{AMDGPU_GFX_SQ_SGPR_MEM3, "SQ_SGPR_MEM3"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_ta_mem_list[] = {
+	{AMDGPU_GFX_TA_FS_AFIFO_RAM_LO, "TA_FS_AFIFO_RAM_LO"},
+	{AMDGPU_GFX_TA_FS_AFIFO_RAM_HI, "TA_FS_AFIFO_RAM_HI"},
+	{AMDGPU_GFX_TA_FS_CFIFO_RAM, "TA_FS_CFIFO_RAM"},
+	{AMDGPU_GFX_TA_FSX_LFIFO, "TA_FSX_LFIFO"},
+	{AMDGPU_GFX_TA_FS_DFIFO_RAM, "TA_FS_DFIFO_RAM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcc_mem_list[] = {
+	{AMDGPU_GFX_TCC_MEM1, "TCC_MEM1"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tca_mem_list[] = {
+	{AMDGPU_GFX_TCA_MEM1, "TCA_MEM1"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tci_mem_list[] = {
+	{AMDGPU_GFX_TCIW_MEM, "TCIW_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcp_mem_list[] = {
+	{AMDGPU_GFX_TCP_LFIFO0, "TCP_LFIFO0"},
+	{AMDGPU_GFX_TCP_SET0BANK0_RAM, "TCP_SET0BANK0_RAM"},
+	{AMDGPU_GFX_TCP_SET0BANK1_RAM, "TCP_SET0BANK1_RAM"},
+	{AMDGPU_GFX_TCP_SET0BANK2_RAM, "TCP_SET0BANK2_RAM"},
+	{AMDGPU_GFX_TCP_SET0BANK3_RAM, "TCP_SET0BANK3_RAM"},
+	{AMDGPU_GFX_TCP_SET1BANK0_RAM, "TCP_SET1BANK0_RAM"},
+	{AMDGPU_GFX_TCP_SET1BANK1_RAM, "TCP_SET1BANK1_RAM"},
+	{AMDGPU_GFX_TCP_SET1BANK2_RAM, "TCP_SET1BANK2_RAM"},
+	{AMDGPU_GFX_TCP_SET1BANK3_RAM, "TCP_SET1BANK3_RAM"},
+	{AMDGPU_GFX_TCP_SET2BANK0_RAM, "TCP_SET2BANK0_RAM"},
+	{AMDGPU_GFX_TCP_SET2BANK1_RAM, "TCP_SET2BANK1_RAM"},
+	{AMDGPU_GFX_TCP_SET2BANK2_RAM, "TCP_SET2BANK2_RAM"},
+	{AMDGPU_GFX_TCP_SET2BANK3_RAM, "TCP_SET2BANK3_RAM"},
+	{AMDGPU_GFX_TCP_SET3BANK0_RAM, "TCP_SET3BANK0_RAM"},
+	{AMDGPU_GFX_TCP_SET3BANK1_RAM, "TCP_SET3BANK1_RAM"},
+	{AMDGPU_GFX_TCP_SET3BANK2_RAM, "TCP_SET3BANK2_RAM"},
+	{AMDGPU_GFX_TCP_SET3BANK3_RAM, "TCP_SET3BANK3_RAM"},
+	{AMDGPU_GFX_TCP_VM_FIFO, "TCP_VM_FIFO"},
+	{AMDGPU_GFX_TCP_DB_TAGRAM0, "TCP_DB_TAGRAM0"},
+	{AMDGPU_GFX_TCP_DB_TAGRAM1, "TCP_DB_TAGRAM1"},
+	{AMDGPU_GFX_TCP_DB_TAGRAM2, "TCP_DB_TAGRAM2"},
+	{AMDGPU_GFX_TCP_DB_TAGRAM3, "TCP_DB_TAGRAM3"},
+	{AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE0, "TCP_UTCL1_LFIFO_PROBE0"},
+	{AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE1, "TCP_UTCL1_LFIFO_PROBE1"},
+	{AMDGPU_GFX_TCP_CMD_FIFO, "TCP_CMD_FIFO"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_td_mem_list[] = {
+	{AMDGPU_GFX_TD_UTD_CS_FIFO_MEM, "TD_UTD_CS_FIFO_MEM"},
+	{AMDGPU_GFX_TD_UTD_SS_FIFO_LO_MEM, "TD_UTD_SS_FIFO_LO_MEM"},
+	{AMDGPU_GFX_TD_UTD_SS_FIFO_HI_MEM, "TD_UTD_SS_FIFO_HI_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcx_mem_list[] = {
+	{AMDGPU_GFX_TCX_FIFOD0, "TCX_FIFOD0"},
+	{AMDGPU_GFX_TCX_FIFOD1, "TCX_FIFOD1"},
+	{AMDGPU_GFX_TCX_FIFOD2, "TCX_FIFOD2"},
+	{AMDGPU_GFX_TCX_FIFOD3, "TCX_FIFOD3"},
+	{AMDGPU_GFX_TCX_FIFOD4, "TCX_FIFOD4"},
+	{AMDGPU_GFX_TCX_FIFOD5, "TCX_FIFOD5"},
+	{AMDGPU_GFX_TCX_FIFOD6, "TCX_FIFOD6"},
+	{AMDGPU_GFX_TCX_FIFOD7, "TCX_FIFOD7"},
+	{AMDGPU_GFX_TCX_FIFOB0, "TCX_FIFOB0"},
+	{AMDGPU_GFX_TCX_FIFOB1, "TCX_FIFOB1"},
+	{AMDGPU_GFX_TCX_FIFOB2, "TCX_FIFOB2"},
+	{AMDGPU_GFX_TCX_FIFOB3, "TCX_FIFOB3"},
+	{AMDGPU_GFX_TCX_FIFOB4, "TCX_FIFOB4"},
+	{AMDGPU_GFX_TCX_FIFOB5, "TCX_FIFOB5"},
+	{AMDGPU_GFX_TCX_FIFOB6, "TCX_FIFOB6"},
+	{AMDGPU_GFX_TCX_FIFOB7, "TCX_FIFOB7"},
+	{AMDGPU_GFX_TCX_FIFOA0, "TCX_FIFOA0"},
+	{AMDGPU_GFX_TCX_FIFOA1, "TCX_FIFOA1"},
+	{AMDGPU_GFX_TCX_FIFOA2, "TCX_FIFOA2"},
+	{AMDGPU_GFX_TCX_FIFOA3, "TCX_FIFOA3"},
+	{AMDGPU_GFX_TCX_FIFOA4, "TCX_FIFOA4"},
+	{AMDGPU_GFX_TCX_FIFOA5, "TCX_FIFOA5"},
+	{AMDGPU_GFX_TCX_FIFOA6, "TCX_FIFOA6"},
+	{AMDGPU_GFX_TCX_FIFOA7, "TCX_FIFOA7"},
+	{AMDGPU_GFX_TCX_CFIFO0, "TCX_CFIFO0"},
+	{AMDGPU_GFX_TCX_CFIFO1, "TCX_CFIFO1"},
+	{AMDGPU_GFX_TCX_CFIFO2, "TCX_CFIFO2"},
+	{AMDGPU_GFX_TCX_CFIFO3, "TCX_CFIFO3"},
+	{AMDGPU_GFX_TCX_CFIFO4, "TCX_CFIFO4"},
+	{AMDGPU_GFX_TCX_CFIFO5, "TCX_CFIFO5"},
+	{AMDGPU_GFX_TCX_CFIFO6, "TCX_CFIFO6"},
+	{AMDGPU_GFX_TCX_CFIFO7, "TCX_CFIFO7"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB0, "TCX_FIFO_ACKB0"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB1, "TCX_FIFO_ACKB1"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB2, "TCX_FIFO_ACKB2"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB3, "TCX_FIFO_ACKB3"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB4, "TCX_FIFO_ACKB4"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB5, "TCX_FIFO_ACKB5"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB6, "TCX_FIFO_ACKB6"},
+	{AMDGPU_GFX_TCX_FIFO_ACKB7, "TCX_FIFO_ACKB7"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD0, "TCX_FIFO_ACKD0"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD1, "TCX_FIFO_ACKD1"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD2, "TCX_FIFO_ACKD2"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD3, "TCX_FIFO_ACKD3"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD4, "TCX_FIFO_ACKD4"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD5, "TCX_FIFO_ACKD5"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD6, "TCX_FIFO_ACKD6"},
+	{AMDGPU_GFX_TCX_FIFO_ACKD7, "TCX_FIFO_ACKD7"},
+	{AMDGPU_GFX_TCX_DST_FIFOA0, "TCX_DST_FIFOA0"},
+	{AMDGPU_GFX_TCX_DST_FIFOA1, "TCX_DST_FIFOA1"},
+	{AMDGPU_GFX_TCX_DST_FIFOA2, "TCX_DST_FIFOA2"},
+	{AMDGPU_GFX_TCX_DST_FIFOA3, "TCX_DST_FIFOA3"},
+	{AMDGPU_GFX_TCX_DST_FIFOA4, "TCX_DST_FIFOA4"},
+	{AMDGPU_GFX_TCX_DST_FIFOA5, "TCX_DST_FIFOA5"},
+	{AMDGPU_GFX_TCX_DST_FIFOA6, "TCX_DST_FIFOA6"},
+	{AMDGPU_GFX_TCX_DST_FIFOA7, "TCX_DST_FIFOA7"},
+	{AMDGPU_GFX_TCX_DST_FIFOB0, "TCX_DST_FIFOB0"},
+	{AMDGPU_GFX_TCX_DST_FIFOB1, "TCX_DST_FIFOB1"},
+	{AMDGPU_GFX_TCX_DST_FIFOB2, "TCX_DST_FIFOB2"},
+	{AMDGPU_GFX_TCX_DST_FIFOB3, "TCX_DST_FIFOB3"},
+	{AMDGPU_GFX_TCX_DST_FIFOB4, "TCX_DST_FIFOB4"},
+	{AMDGPU_GFX_TCX_DST_FIFOB5, "TCX_DST_FIFOB5"},
+	{AMDGPU_GFX_TCX_DST_FIFOB6, "TCX_DST_FIFOB6"},
+	{AMDGPU_GFX_TCX_DST_FIFOB7, "TCX_DST_FIFOB7"},
+	{AMDGPU_GFX_TCX_DST_FIFOD0, "TCX_DST_FIFOD0"},
+	{AMDGPU_GFX_TCX_DST_FIFOD1, "TCX_DST_FIFOD1"},
+	{AMDGPU_GFX_TCX_DST_FIFOD2, "TCX_DST_FIFOD2"},
+	{AMDGPU_GFX_TCX_DST_FIFOD3, "TCX_DST_FIFOD3"},
+	{AMDGPU_GFX_TCX_DST_FIFOD4, "TCX_DST_FIFOD4"},
+	{AMDGPU_GFX_TCX_DST_FIFOD5, "TCX_DST_FIFOD5"},
+	{AMDGPU_GFX_TCX_DST_FIFOD6, "TCX_DST_FIFOD6"},
+	{AMDGPU_GFX_TCX_DST_FIFOD7, "TCX_DST_FIFOD7"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB0, "TCX_DST_FIFO_ACKB0"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB1, "TCX_DST_FIFO_ACKB1"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB2, "TCX_DST_FIFO_ACKB2"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB3, "TCX_DST_FIFO_ACKB3"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB4, "TCX_DST_FIFO_ACKB4"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB5, "TCX_DST_FIFO_ACKB5"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB6, "TCX_DST_FIFO_ACKB6"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKB7, "TCX_DST_FIFO_ACKB7"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD0, "TCX_DST_FIFO_ACKD0"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD1, "TCX_DST_FIFO_ACKD1"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD2, "TCX_DST_FIFO_ACKD2"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD3, "TCX_DST_FIFO_ACKD3"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD4, "TCX_DST_FIFO_ACKD4"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD5, "TCX_DST_FIFO_ACKD5"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD6, "TCX_DST_FIFO_ACKD6"},
+	{AMDGPU_GFX_TCX_DST_FIFO_ACKD7, "TCX_DST_FIFO_ACKD7"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_atc_l2_mem_list[] = {
+	{AMDGPU_GFX_ATC_L2_MEM, "ATC_L2_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_utcl2_mem_list[] = {
+	{AMDGPU_GFX_UTCL2_MEM, "UTCL2_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_vml2_mem_list[] = {
+	{AMDGPU_GFX_VML2_MEM, "VML2_MEM"},
+};
+
+static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_vml2_walker_mem_list[] = {
+	{AMDGPU_GFX_VML2_WALKER_MEM, "VML2_WALKER_MEM"},
+};
+
+static const struct amdgpu_gfx_ras_mem_id_entry gfx_v9_4_3_ras_mem_list_array[AMDGPU_GFX_MEM_TYPE_NUM] = {
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_cp_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gcea_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gc_cane_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gcutcl2_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gds_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_lds_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_rlc_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sp_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_spi_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sqc_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sq_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_ta_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcc_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tca_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tci_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcp_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_td_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcx_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_atc_l2_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_utcl2_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_vml2_mem_list)
+	AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_vml2_walker_mem_list)
+};
+
+static const struct amdgpu_gfx_ras_reg_entry gfx_v9_4_3_ce_reg_list[] = {
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regRLC_CE_ERR_STATUS_LOW, regRLC_CE_ERR_STATUS_HIGH),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "RLC"},
+	    AMDGPU_GFX_RLC_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPC_CE_ERR_STATUS_LO, regCPC_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPC"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPF_CE_ERR_STATUS_LO, regCPF_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPF"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPG_CE_ERR_STATUS_LO, regCPG_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPG"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGDS_CE_ERR_STATUS_LO, regGDS_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GDS"},
+	    AMDGPU_GFX_GDS_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGC_CANE_CE_ERR_STATUS_LO, regGC_CANE_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CANE"},
+	    AMDGPU_GFX_GC_CANE_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSPI_CE_ERR_STATUS_LO, regSPI_CE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SPI"},
+	    AMDGPU_GFX_SPI_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP0_CE_ERR_STATUS_LO, regSP0_CE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP0"},
+	    AMDGPU_GFX_SP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP1_CE_ERR_STATUS_LO, regSP1_CE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP1"},
+	    AMDGPU_GFX_SP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQ_CE_ERR_STATUS_LO, regSQ_CE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQ"},
+	    AMDGPU_GFX_SQ_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQC_CE_EDC_LO, regSQC_CE_EDC_HI),
+	    5, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQC"},
+	    AMDGPU_GFX_SQC_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCX_CE_ERR_STATUS_LO, regTCX_CE_ERR_STATUS_HI),
+	    2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCX"},
+	    AMDGPU_GFX_TCX_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCC_CE_ERR_STATUS_LO, regTCC_CE_ERR_STATUS_HI),
+	    16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCC"},
+	    AMDGPU_GFX_TCC_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTA_CE_EDC_LO, regTA_CE_EDC_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TA"},
+	    AMDGPU_GFX_TA_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCI_CE_EDC_LO_REG, regTCI_CE_EDC_HI_REG),
+	    31, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCI"},
+	    AMDGPU_GFX_TCI_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCP_CE_EDC_LO_REG, regTCP_CE_EDC_HI_REG),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCP"},
+	    AMDGPU_GFX_TCP_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTD_CE_EDC_LO, regTD_CE_EDC_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TD"},
+	    AMDGPU_GFX_TD_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGCEA_CE_ERR_STATUS_LO, regGCEA_CE_ERR_STATUS_HI),
+	    16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GCEA"},
+	    AMDGPU_GFX_GCEA_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regLDS_CE_ERR_STATUS_LO, regLDS_CE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "LDS"},
+	    AMDGPU_GFX_LDS_MEM, 1},
+};
+
+static const struct amdgpu_gfx_ras_reg_entry gfx_v9_4_3_ue_reg_list[] = {
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regRLC_UE_ERR_STATUS_LOW, regRLC_UE_ERR_STATUS_HIGH),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "RLC"},
+	    AMDGPU_GFX_RLC_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPC_UE_ERR_STATUS_LO, regCPC_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPC"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPF_UE_ERR_STATUS_LO, regCPF_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPF"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPG_UE_ERR_STATUS_LO, regCPG_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPG"},
+	    AMDGPU_GFX_CP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGDS_UE_ERR_STATUS_LO, regGDS_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GDS"},
+	    AMDGPU_GFX_GDS_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGC_CANE_UE_ERR_STATUS_LO, regGC_CANE_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CANE"},
+	    AMDGPU_GFX_GC_CANE_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSPI_UE_ERR_STATUS_LO, regSPI_UE_ERR_STATUS_HI),
+	    1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SPI"},
+	    AMDGPU_GFX_SPI_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP0_UE_ERR_STATUS_LO, regSP0_UE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP0"},
+	    AMDGPU_GFX_SP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP1_UE_ERR_STATUS_LO, regSP1_UE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP1"},
+	    AMDGPU_GFX_SP_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQ_UE_ERR_STATUS_LO, regSQ_UE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQ"},
+	    AMDGPU_GFX_SQ_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQC_UE_EDC_LO, regSQC_UE_EDC_HI),
+	    5, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQC"},
+	    AMDGPU_GFX_SQC_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCX_UE_ERR_STATUS_LO, regTCX_UE_ERR_STATUS_HI),
+	    2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCX"},
+	    AMDGPU_GFX_TCX_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCC_UE_ERR_STATUS_LO, regTCC_UE_ERR_STATUS_HI),
+	    16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCC"},
+	    AMDGPU_GFX_TCC_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTA_UE_EDC_LO, regTA_UE_EDC_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TA"},
+	    AMDGPU_GFX_TA_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCI_UE_EDC_LO_REG, regTCI_UE_EDC_HI_REG),
+	    31, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCI"},
+	    AMDGPU_GFX_TCI_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCP_UE_EDC_LO_REG, regTCP_UE_EDC_HI_REG),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCP"},
+	    AMDGPU_GFX_TCP_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTD_UE_EDC_LO, regTD_UE_EDC_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TD"},
+	    AMDGPU_GFX_TD_MEM, 8},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCA_UE_ERR_STATUS_LO, regTCA_UE_ERR_STATUS_HI),
+	    2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCA"},
+	    AMDGPU_GFX_TCA_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regGCEA_UE_ERR_STATUS_LO, regGCEA_UE_ERR_STATUS_HI),
+	    16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GCEA"},
+	    AMDGPU_GFX_GCEA_MEM, 1},
+	{{AMDGPU_RAS_REG_ENTRY(GC, 0, regLDS_UE_ERR_STATUS_LO, regLDS_UE_ERR_STATUS_HI),
+	    10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "LDS"},
+	    AMDGPU_GFX_LDS_MEM, 1},
+};
+
 static const struct soc15_reg_entry gfx_v9_4_3_ea_err_status_regs = {
 	SOC15_REG_ENTRY(GC, 0, regGCEA_ERR_STATUS), 0, 1, 16
 };
-- 
GitLab


From bfa84da6185cb1897fcee0ac3815625d162d39f0 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Fri, 17 Mar 2023 17:13:46 +0800
Subject: [PATCH 1344/1791] drm/amdgpu: add RAS error count query for
 gfx_v9_4_3

Query GFX RAS ce/ue count.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 56 +++++++++++++++++++++++++
 1 file changed, 56 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 188b4d9a2cbb9..bfd041ba51d65 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -3724,6 +3724,55 @@ static const struct soc15_reg_entry gfx_v9_4_3_ea_err_status_regs = {
 	SOC15_REG_ENTRY(GC, 0, regGCEA_ERR_STATUS), 0, 1, 16
 };
 
+static void gfx_v9_4_3_inst_query_ras_err_count(struct amdgpu_device *adev,
+					void *ras_error_status, int xcc_id)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
+	unsigned long ce_count = 0, ue_count = 0;
+	uint32_t i, j, k;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	for (i = 0; i < ARRAY_SIZE(gfx_v9_4_3_ce_reg_list); i++) {
+		for (j = 0; j < gfx_v9_4_3_ce_reg_list[i].se_num; j++) {
+			for (k = 0; k < gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst; k++) {
+				/* no need to select if instance number is 1 */
+				if (gfx_v9_4_3_ce_reg_list[i].se_num > 1 ||
+				    gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst > 1)
+					gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id);
+
+				amdgpu_ras_inst_query_ras_error_count(adev,
+					&(gfx_v9_4_3_ce_reg_list[i].reg_entry),
+					1,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ce_reg_list[i].mem_id_type].mem_id_ent,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ce_reg_list[i].mem_id_type].size,
+					GET_INST(GC, xcc_id),
+					AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE,
+					&ce_count);
+
+				amdgpu_ras_inst_query_ras_error_count(adev,
+					&(gfx_v9_4_3_ue_reg_list[i].reg_entry),
+					1,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].mem_id_ent,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].size,
+					GET_INST(GC, xcc_id),
+					AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+					&ue_count);
+			}
+		}
+	}
+
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	/* the caller should make sure initialize value of
+	 * err_data->ue_count and err_data->ce_count
+	 */
+	err_data->ce_count += ce_count;
+	err_data->ue_count += ue_count;
+}
+
 static void gfx_v9_4_3_inst_query_ea_err_status(struct amdgpu_device *adev,
 					int xcc_id)
 {
@@ -3826,6 +3875,13 @@ static void gfx_v9_4_3_inst_reset_ras_err_status(struct amdgpu_device *adev,
 	gfx_v9_4_3_inst_reset_ea_err_status(adev, xcc_id);
 }
 
+static void gfx_v9_4_3_query_ras_error_count(struct amdgpu_device *adev,
+					void *ras_error_status)
+{
+	amdgpu_gfx_ras_error_func(adev, ras_error_status,
+			gfx_v9_4_3_inst_query_ras_err_count);
+}
+
 static void gfx_v9_4_3_query_ras_error_status(struct amdgpu_device *adev)
 {
 	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_query_ras_err_status);
-- 
GitLab


From 30feef0676092bdb4b8697e68b8d5864d54f096f Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Wed, 8 Feb 2023 14:54:01 +0800
Subject: [PATCH 1345/1791] drm/amdgpu: add RAS error count reset for
 gfx_v9_4_3

Add GFX RAS error count reset function.

v2: remove xcp operation.
    only select_se_sh when instance number is more than 1.
v3: add check for se_num before select_se_sh.
    change instance from 0 to xcc_id for register access.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 38 +++++++++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index bfd041ba51d65..ac5270d5eff4d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -3773,6 +3773,39 @@ static void gfx_v9_4_3_inst_query_ras_err_count(struct amdgpu_device *adev,
 	err_data->ue_count += ue_count;
 }
 
+static void gfx_v9_4_3_inst_reset_ras_err_count(struct amdgpu_device *adev,
+					void *ras_error_status, int xcc_id)
+{
+	uint32_t i, j, k;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	for (i = 0; i < ARRAY_SIZE(gfx_v9_4_3_ce_reg_list); i++) {
+		for (j = 0; j < gfx_v9_4_3_ce_reg_list[i].se_num; j++) {
+			for (k = 0; k < gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst; k++) {
+				/* no need to select if instance number is 1 */
+				if (gfx_v9_4_3_ce_reg_list[i].se_num > 1 ||
+				    gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst > 1)
+					gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id);
+
+				amdgpu_ras_inst_reset_ras_error_count(adev,
+					&(gfx_v9_4_3_ce_reg_list[i].reg_entry),
+					1,
+					GET_INST(GC, xcc_id));
+
+				amdgpu_ras_inst_reset_ras_error_count(adev,
+					&(gfx_v9_4_3_ue_reg_list[i].reg_entry),
+					1,
+					GET_INST(GC, xcc_id));
+			}
+		}
+	}
+
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
+
 static void gfx_v9_4_3_inst_query_ea_err_status(struct amdgpu_device *adev,
 					int xcc_id)
 {
@@ -3882,6 +3915,11 @@ static void gfx_v9_4_3_query_ras_error_count(struct amdgpu_device *adev,
 			gfx_v9_4_3_inst_query_ras_err_count);
 }
 
+static void gfx_v9_4_3_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_reset_ras_err_count);
+}
+
 static void gfx_v9_4_3_query_ras_error_status(struct amdgpu_device *adev)
 {
 	amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_query_ras_err_status);
-- 
GitLab


From 0386d52d1516d80b81a25552df74b8a82dfb77f3 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Fri, 10 Feb 2023 18:41:28 +0800
Subject: [PATCH 1346/1791] drm/amdgpu: add sq timeout status functions for
 gfx_v9_4_3

Query and reset sq timeout status.

v2: change instance from 0 to xcc_id for register access.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 98 +++++++++++++++++++++++++
 1 file changed, 98 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ac5270d5eff4d..5bd2f40a817e6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -3865,11 +3865,87 @@ static void gfx_v9_4_3_inst_query_utc_err_status(struct amdgpu_device *adev,
 	}
 }
 
+static void gfx_v9_4_3_log_cu_timeout_status(struct amdgpu_device *adev,
+					uint32_t status, int xcc_id)
+{
+	struct amdgpu_cu_info *cu_info = &adev->gfx.cu_info;
+	uint32_t i, simd, wave;
+	uint32_t wave_status;
+	uint32_t wave_pc_lo, wave_pc_hi;
+	uint32_t wave_exec_lo, wave_exec_hi;
+	uint32_t wave_inst_dw0, wave_inst_dw1;
+	uint32_t wave_ib_sts;
+
+	for (i = 0; i < 32; i++) {
+		if (!((i << 1) & status))
+			continue;
+
+		simd = i / cu_info->max_waves_per_simd;
+		wave = i % cu_info->max_waves_per_simd;
+
+		wave_status = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_STATUS);
+		wave_pc_lo = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_LO);
+		wave_pc_hi = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_HI);
+		wave_exec_lo =
+			wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_LO);
+		wave_exec_hi =
+			wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_HI);
+		wave_inst_dw0 =
+			wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW0);
+		wave_inst_dw1 =
+			wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW1);
+		wave_ib_sts = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_STS);
+
+		dev_info(
+			adev->dev,
+			"\t SIMD %d, Wave %d: status 0x%x, pc 0x%llx, exec 0x%llx, inst 0x%llx, ib_sts 0x%x\n",
+			simd, wave, wave_status,
+			((uint64_t)wave_pc_hi << 32 | wave_pc_lo),
+			((uint64_t)wave_exec_hi << 32 | wave_exec_lo),
+			((uint64_t)wave_inst_dw1 << 32 | wave_inst_dw0),
+			wave_ib_sts);
+	}
+}
+
+static void gfx_v9_4_3_inst_query_sq_timeout_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	uint32_t se_idx, sh_idx, cu_idx;
+	uint32_t status;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	for (se_idx = 0; se_idx < adev->gfx.config.max_shader_engines; se_idx++) {
+		for (sh_idx = 0; sh_idx < adev->gfx.config.max_sh_per_se; sh_idx++) {
+			for (cu_idx = 0; cu_idx < adev->gfx.config.max_cu_per_sh; cu_idx++) {
+				gfx_v9_4_3_xcc_select_se_sh(adev, se_idx, sh_idx,
+							cu_idx, xcc_id);
+				status = RREG32_SOC15(GC, GET_INST(GC, xcc_id),
+						      regSQ_TIMEOUT_STATUS);
+				if (status != 0) {
+					dev_info(
+						adev->dev,
+						"GFX Watchdog Timeout: SE %d, SH %d, CU %d\n",
+						se_idx, sh_idx, cu_idx);
+					gfx_v9_4_3_log_cu_timeout_status(
+						adev, status, xcc_id);
+				}
+				/* clear old status */
+				WREG32_SOC15(GC, GET_INST(GC, xcc_id),
+						regSQ_TIMEOUT_STATUS, 0);
+			}
+		}
+	}
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
+
 static void gfx_v9_4_3_inst_query_ras_err_status(struct amdgpu_device *adev,
 					void *ras_error_status, int xcc_id)
 {
 	gfx_v9_4_3_inst_query_ea_err_status(adev, xcc_id);
 	gfx_v9_4_3_inst_query_utc_err_status(adev, xcc_id);
+	gfx_v9_4_3_inst_query_sq_timeout_status(adev, xcc_id);
 }
 
 static void gfx_v9_4_3_inst_reset_utc_err_status(struct amdgpu_device *adev,
@@ -3901,11 +3977,33 @@ static void gfx_v9_4_3_inst_reset_ea_err_status(struct amdgpu_device *adev,
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
+static void gfx_v9_4_3_inst_reset_sq_timeout_status(struct amdgpu_device *adev,
+					int xcc_id)
+{
+	uint32_t se_idx, sh_idx, cu_idx;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+	for (se_idx = 0; se_idx < adev->gfx.config.max_shader_engines; se_idx++) {
+		for (sh_idx = 0; sh_idx < adev->gfx.config.max_sh_per_se; sh_idx++) {
+			for (cu_idx = 0; cu_idx < adev->gfx.config.max_cu_per_sh; cu_idx++) {
+				gfx_v9_4_3_xcc_select_se_sh(adev, se_idx, sh_idx,
+							cu_idx, xcc_id);
+				WREG32_SOC15(GC, GET_INST(GC, xcc_id),
+						regSQ_TIMEOUT_STATUS, 0);
+			}
+		}
+	}
+	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
+			xcc_id);
+	mutex_unlock(&adev->grbm_idx_mutex);
+}
+
 static void gfx_v9_4_3_inst_reset_ras_err_status(struct amdgpu_device *adev,
 					void *ras_error_status, int xcc_id)
 {
 	gfx_v9_4_3_inst_reset_utc_err_status(adev, xcc_id);
 	gfx_v9_4_3_inst_reset_ea_err_status(adev, xcc_id);
+	gfx_v9_4_3_inst_reset_sq_timeout_status(adev, xcc_id);
 }
 
 static void gfx_v9_4_3_query_ras_error_count(struct amdgpu_device *adev,
-- 
GitLab


From 92ecb92ccc839c4c4b51ab1025cde5dd82c2fb4b Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Wed, 8 Feb 2023 17:05:08 +0800
Subject: [PATCH 1347/1791] drm/amdgpu: initialize RAS for gfx_v9_4_3

Register GFX RAS functions and initialize GFX RAS.

v2: remove xcp operations.
v3: reuse the return value of gfx_ras_sw_init.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 5bd2f40a817e6..e5cfb3adb3b3f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -47,6 +47,8 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
 #define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
 
+struct amdgpu_gfx_ras gfx_v9_4_3_ras;
+
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev);
 static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev);
@@ -659,6 +661,7 @@ static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev)
 	u32 gb_addr_config;
 
 	adev->gfx.funcs = &gfx_v9_4_3_gfx_funcs;
+	adev->gfx.ras = &gfx_v9_4_3_ras;
 
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 3):
@@ -845,7 +848,7 @@ static int gfx_v9_4_3_sw_init(void *handle)
 	if (r)
 		return r;
 
-	return 0;
+	return amdgpu_gfx_ras_sw_init(adev);
 }
 
 static int gfx_v9_4_3_sw_fini(void *handle)
@@ -4342,3 +4345,16 @@ struct amdgpu_xcp_ip_funcs gfx_v9_4_3_xcp_funcs = {
 	.suspend = &gfx_v9_4_3_xcp_suspend,
 	.resume = &gfx_v9_4_3_xcp_resume
 };
+
+struct amdgpu_ras_block_hw_ops  gfx_v9_4_3_ras_ops = {
+	.query_ras_error_count = &gfx_v9_4_3_query_ras_error_count,
+	.reset_ras_error_count = &gfx_v9_4_3_reset_ras_error_count,
+	.query_ras_error_status = &gfx_v9_4_3_query_ras_error_status,
+	.reset_ras_error_status = &gfx_v9_4_3_reset_ras_error_status,
+};
+
+struct amdgpu_gfx_ras gfx_v9_4_3_ras = {
+	.ras_block = {
+		.hw_ops = &gfx_v9_4_3_ras_ops,
+	},
+};
-- 
GitLab


From 45b3a914d40e63d2c9e3a3e02fb2014be975b9b0 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 16 May 2023 17:16:30 -0400
Subject: [PATCH 1348/1791] drm/amdgpu/gmc9: fix 64 bit division in partition
 code

Rework logic or use do_div() to avoid problems on 32 bit.

v2: add a missing case for XCP macro
v3: fix out of bounds array access
v4: fix xcp handling harder

Acked-by: Guchun Chen <guchun.chen@amd.com> (v1)
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> (v3)
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 15 +++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h |  9 ++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    |  5 ++++-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c      | 11 ++++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c       |  8 ++++++--
 5 files changed, 35 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 739eb7c0d133b..5de92c9ab18f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -794,3 +794,18 @@ void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev)
 {
 	kgd2kfd_unlock_kfd();
 }
+
+
+u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id)
+{
+	u64 tmp;
+	s8 mem_id = KFD_XCP_MEM_ID(adev, xcp_id);
+
+	if (adev->gmc.num_mem_partitions && xcp_id >= 0 && mem_id >= 0) {
+		tmp = adev->gmc.mem_partitions[mem_id].size;
+		do_div(tmp, adev->xcp_mgr->num_xcp_per_mem_partition);
+		return tmp;
+	} else {
+		return adev->gmc.real_vram_size;
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index be43d71ba7ef4..94cc456761e57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -333,15 +333,14 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		uint64_t size, u32 alloc_flag, int8_t xcp_id);
 
+u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id);
+
 #define KFD_XCP_MEM_ID(adev, xcp_id) \
 		((adev)->xcp_mgr && (xcp_id) >= 0 ?\
 		(adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1)
 
-#define KFD_XCP_MEMORY_SIZE(adev, xcp_id)\
-		((adev)->gmc.num_mem_partitions && (xcp_id) >= 0 ?\
-		(adev)->gmc.mem_partitions[KFD_XCP_MEM_ID((adev), (xcp_id))].size /\
-		(adev)->xcp_mgr->num_xcp_per_mem_partition :\
-		(adev)->gmc.real_vram_size)
+#define KFD_XCP_MEMORY_SIZE(adev, xcp_id) amdgpu_amdkfd_xcp_memory_size((adev), (xcp_id))
+
 
 #if IS_ENABLED(CONFIG_HSA_AMD)
 void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 23101c82519af..902773ce41b5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -814,11 +814,14 @@ static void amdgpu_ttm_gart_bind_gfx9_mqd(struct amdgpu_device *adev,
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
 	uint64_t total_pages = ttm->num_pages;
 	int num_xcc = max(1U, adev->gfx.num_xcc_per_xcp);
-	uint64_t page_idx, pages_per_xcc = total_pages / num_xcc;
+	uint64_t page_idx, pages_per_xcc;
 	int i;
 	uint64_t ctrl_flags = (flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
 			AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC);
 
+	pages_per_xcc = total_pages;
+	do_div(pages_per_xcc, num_xcc);
+
 	for (i = 0, page_idx = 0; i < num_xcc; i++, page_idx += pages_per_xcc) {
 		/* MQD page: use default flags */
 		amdgpu_gart_bind(adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 7ea80bdf8e1ea..f70e666cecf24 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1914,9 +1914,10 @@ gmc_v9_0_init_acpi_mem_ranges(struct amdgpu_device *adev,
 	adev->gmc.num_mem_partitions = num_ranges;
 
 	/* If there is only partition, don't use entire size */
-	if (adev->gmc.num_mem_partitions == 1)
-		mem_ranges[0].size =
-			(mem_ranges[0].size * (mem_groups - 1) / mem_groups);
+	if (adev->gmc.num_mem_partitions == 1) {
+		mem_ranges[0].size = mem_ranges[0].size * (mem_groups - 1);
+		do_div(mem_ranges[0].size, mem_groups);
+	}
 }
 
 static void
@@ -1948,8 +1949,8 @@ gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev,
 		break;
 	}
 
-	size = (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) /
-	       adev->gmc.num_mem_partitions;
+	size = adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT;
+	size /= adev->gmc.num_mem_partitions;
 
 	for (i = 0; i < adev->gmc.num_mem_partitions; ++i) {
 		mem_ranges[i].range.fpfn = start_addr;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 206851c9e6424..b0f0d31bf3e63 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1939,10 +1939,14 @@ void svm_range_set_max_pages(struct amdgpu_device *adev)
 	uint64_t max_pages;
 	uint64_t pages, _pages;
 	uint64_t min_pages = 0;
-	int i;
+	int i, id;
 
 	for (i = 0; i < adev->kfd.dev->num_nodes; i++) {
-		pages = KFD_XCP_MEMORY_SIZE(adev, adev->kfd.dev->nodes[i]->xcp->id) >> 17;
+		if (adev->kfd.dev->nodes[i]->xcp)
+			id = adev->kfd.dev->nodes[i]->xcp->id;
+		else
+			id = -1;
+		pages = KFD_XCP_MEMORY_SIZE(adev, id) >> 17;
 		pages = clamp(pages, 1ULL << 9, 1ULL << 18);
 		pages = rounddown_pow_of_two(pages);
 		min_pages = min_not_zero(min_pages, pages);
-- 
GitLab


From 0ce50b2efe08e56224d11d735310b353e0e4e222 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 16 May 2023 16:56:49 -0400
Subject: [PATCH 1349/1791] drm/amdgpu/vcn4: fix endian conversion

sq.is_enabled is a byte so there is no need to endian swap it.

Acked-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index c77ceaf53dcd9..339842382a1ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -124,7 +124,7 @@ static int vcn_v4_0_3_sw_init(void *handle)
 
 		fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 		fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE);
-		fw_shared->sq.is_enabled = cpu_to_le32(true);
+		fw_shared->sq.is_enabled = true;
 
 		if (amdgpu_vcnfw_log)
 			amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]);
-- 
GitLab


From 01c3f464743b64e6e65cb9bad951458986819a42 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 17 May 2023 20:10:48 +0530
Subject: [PATCH 1350/1791] drm/amd/amdgpu: Fix errors & warnings in
 amdgpu_ttm.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below checkpatch insisted error & warnings:

ERROR: Macros with complex values should be enclosed in parentheses
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: braces {} are not necessary for single statement blocks
WARNING: Block comments use a trailing */ on a separate line
WARNING: Missing a blank line after declarations

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 902773ce41b5d..b8adcebb0d89a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -65,7 +65,7 @@
 
 MODULE_IMPORT_NS(DMA_BUF);
 
-#define AMDGPU_TTM_VRAM_MAX_DW_READ	(size_t)128
+#define AMDGPU_TTM_VRAM_MAX_DW_READ	((size_t)128)
 
 static int amdgpu_ttm_backend_bind(struct ttm_device *bdev,
 				   struct ttm_tt *ttm,
@@ -184,11 +184,11 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
 static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo,
 				 struct ttm_resource *mem,
 				 struct amdgpu_res_cursor *mm_cur,
-				 unsigned window, struct amdgpu_ring *ring,
+				 unsigned int window, struct amdgpu_ring *ring,
 				 bool tmz, uint64_t *size, uint64_t *addr)
 {
 	struct amdgpu_device *adev = ring->adev;
-	unsigned offset, num_pages, num_dw, num_bytes;
+	unsigned int offset, num_pages, num_dw, num_bytes;
 	uint64_t src_addr, dst_addr;
 	struct amdgpu_job *job;
 	void *cpu_addr;
@@ -1060,9 +1060,9 @@ static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo,
 	enum ttm_caching caching;
 
 	gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL);
-	if (gtt == NULL) {
+	if (!gtt)
 		return NULL;
-	}
+
 	gtt->gobj = &bo->base;
 	if (adev->gmc.mem_partitions && abo->xcp_id >= 0)
 		gtt->pool_id = KFD_XCP_MEM_ID(adev, abo->xcp_id);
@@ -1847,9 +1847,8 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	 *place on the VRAM, so reserve it early.
 	 */
 	r = amdgpu_ttm_fw_reserve_vram_init(adev);
-	if (r) {
+	if (r)
 		return r;
-	}
 
 	/*
 	 *The reserved vram for driver must be pinned to the specified
@@ -1873,7 +1872,8 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	/* allocate memory as required for VGA
 	 * This is used for VGA emulation and pre-OS scanout buffers to
 	 * avoid display artifacts while transitioning between pre-OS
-	 * and driver.  */
+	 * and driver.
+	 */
 	if (!adev->gmc.is_app_apu) {
 		r = amdgpu_bo_create_kernel_at(adev, 0,
 					       adev->mman.stolen_vga_size,
@@ -1902,7 +1902,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	}
 
 	DRM_INFO("amdgpu: %uM of VRAM memory ready\n",
-		 (unsigned) (adev->gmc.real_vram_size / (1024 * 1024)));
+		 (unsigned int)(adev->gmc.real_vram_size / (1024 * 1024)));
 
 	/* Compute GTT size, either based on TTM limit
 	 * or whatever the user passed on module init.
@@ -1919,7 +1919,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		return r;
 	}
 	DRM_INFO("amdgpu: %uM of GTT memory ready.\n",
-		 (unsigned)(gtt_size / (1024 * 1024)));
+		 (unsigned int)(gtt_size / (1024 * 1024)));
 
 	/* Initialize preemptible memory pool */
 	r = amdgpu_preempt_mgr_init(adev);
@@ -1961,6 +1961,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 void amdgpu_ttm_fini(struct amdgpu_device *adev)
 {
 	int idx;
+
 	if (!adev->mman.initialized)
 		return;
 
@@ -2089,10 +2090,10 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 		       bool vm_needs_flush, bool tmz)
 {
 	struct amdgpu_device *adev = ring->adev;
-	unsigned num_loops, num_dw;
+	unsigned int num_loops, num_dw;
 	struct amdgpu_job *job;
 	uint32_t max_bytes;
-	unsigned i;
+	unsigned int i;
 	int r;
 
 	if (!direct_submit && !ring->sched.ready) {
-- 
GitLab


From f2cd6b26922e68ffafd14a9128e20630296e430d Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 17 May 2023 15:04:35 -0400
Subject: [PATCH 1351/1791] drm/amdkfd: fix stack size in
 svm_range_validate_and_map
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Allocate large local variable on heap to avoid exceeding the
stack size:
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c: In function ‘svm_range_validate_and_map’:
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c:1690:1: warning: the frame size of 2360 bytes is larger than 2048 bytes [-Wframe-larger-than=]

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 56 ++++++++++++++++------------
 1 file changed, 33 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index b0f0d31bf3e63..ee16130ddc75e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1554,48 +1554,54 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 				      struct svm_range *prange, int32_t gpuidx,
 				      bool intr, bool wait, bool flush_tlb)
 {
-	struct svm_validate_context ctx;
+	struct svm_validate_context *ctx;
 	unsigned long start, end, addr;
 	struct kfd_process *p;
 	void *owner;
 	int32_t idx;
 	int r = 0;
 
-	ctx.process = container_of(prange->svms, struct kfd_process, svms);
-	ctx.prange = prange;
-	ctx.intr = intr;
+	ctx = kzalloc(sizeof(struct svm_validate_context), GFP_KERNEL);
+	if (!ctx)
+		return -ENOMEM;
+	ctx->process = container_of(prange->svms, struct kfd_process, svms);
+	ctx->prange = prange;
+	ctx->intr = intr;
 
 	if (gpuidx < MAX_GPU_INSTANCE) {
-		bitmap_zero(ctx.bitmap, MAX_GPU_INSTANCE);
-		bitmap_set(ctx.bitmap, gpuidx, 1);
-	} else if (ctx.process->xnack_enabled) {
-		bitmap_copy(ctx.bitmap, prange->bitmap_aip, MAX_GPU_INSTANCE);
+		bitmap_zero(ctx->bitmap, MAX_GPU_INSTANCE);
+		bitmap_set(ctx->bitmap, gpuidx, 1);
+	} else if (ctx->process->xnack_enabled) {
+		bitmap_copy(ctx->bitmap, prange->bitmap_aip, MAX_GPU_INSTANCE);
 
 		/* If prefetch range to GPU, or GPU retry fault migrate range to
 		 * GPU, which has ACCESS attribute to the range, create mapping
 		 * on that GPU.
 		 */
 		if (prange->actual_loc) {
-			gpuidx = kfd_process_gpuidx_from_gpuid(ctx.process,
+			gpuidx = kfd_process_gpuidx_from_gpuid(ctx->process,
 							prange->actual_loc);
 			if (gpuidx < 0) {
 				WARN_ONCE(1, "failed get device by id 0x%x\n",
 					 prange->actual_loc);
-				return -EINVAL;
+				r = -EINVAL;
+				goto free_ctx;
 			}
 			if (test_bit(gpuidx, prange->bitmap_access))
-				bitmap_set(ctx.bitmap, gpuidx, 1);
+				bitmap_set(ctx->bitmap, gpuidx, 1);
 		}
 	} else {
-		bitmap_or(ctx.bitmap, prange->bitmap_access,
+		bitmap_or(ctx->bitmap, prange->bitmap_access,
 			  prange->bitmap_aip, MAX_GPU_INSTANCE);
 	}
 
-	if (bitmap_empty(ctx.bitmap, MAX_GPU_INSTANCE)) {
-		if (!prange->mapped_to_gpu)
-			return 0;
+	if (bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) {
+		if (!prange->mapped_to_gpu) {
+			r = 0;
+			goto free_ctx;
+		}
 
-		bitmap_copy(ctx.bitmap, prange->bitmap_access, MAX_GPU_INSTANCE);
+		bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE);
 	}
 
 	if (prange->actual_loc && !prange->ttm_res) {
@@ -1603,15 +1609,16 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 		 * svm_migrate_ram_to_vram after allocating a BO.
 		 */
 		WARN_ONCE(1, "VRAM BO missing during validation\n");
-		return -EINVAL;
+		r = -EINVAL;
+		goto free_ctx;
 	}
 
-	svm_range_reserve_bos(&ctx);
+	svm_range_reserve_bos(ctx);
 
 	p = container_of(prange->svms, struct kfd_process, svms);
-	owner = kfd_svm_page_owner(p, find_first_bit(ctx.bitmap,
+	owner = kfd_svm_page_owner(p, find_first_bit(ctx->bitmap,
 						MAX_GPU_INSTANCE));
-	for_each_set_bit(idx, ctx.bitmap, MAX_GPU_INSTANCE) {
+	for_each_set_bit(idx, ctx->bitmap, MAX_GPU_INSTANCE) {
 		if (kfd_svm_page_owner(p, idx) != owner) {
 			owner = NULL;
 			break;
@@ -1648,7 +1655,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 		}
 
 		offset = (addr - start) >> PAGE_SHIFT;
-		r = svm_range_dma_map(prange, ctx.bitmap, offset, npages,
+		r = svm_range_dma_map(prange, ctx->bitmap, offset, npages,
 				      hmm_range->hmm_pfns);
 		if (r) {
 			pr_debug("failed %d to dma map range\n", r);
@@ -1668,7 +1675,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 		}
 
 		r = svm_range_map_to_gpus(prange, offset, npages, readonly,
-					  ctx.bitmap, wait, flush_tlb);
+					  ctx->bitmap, wait, flush_tlb);
 
 unlock_out:
 		svm_range_unlock(prange);
@@ -1682,11 +1689,14 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 	}
 
 unreserve_out:
-	svm_range_unreserve_bos(&ctx);
+	svm_range_unreserve_bos(ctx);
 
 	if (!r)
 		prange->validate_timestamp = ktime_get_boottime();
 
+free_ctx:
+	kfree(ctx);
+
 	return r;
 }
 
-- 
GitLab


From 29f187f71e7fd4e4516f235c400e457c50e2ab70 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 17 May 2023 18:38:00 +0530
Subject: [PATCH 1352/1791] drm/amd/amdgpu: Fix warnings in amdgpu_encoders.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below checkpatch warnings:

WARNING: Missing a blank line after declarations
+                       struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
+                       amdgpu_encoder->active_device = amdgpu_encoder->devices & amdgpu_connector->devices;

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
index 27a782a9dc728..3aaeed2d35620 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c
@@ -70,6 +70,7 @@ void amdgpu_encoder_set_active_device(struct drm_encoder *encoder)
 	drm_for_each_connector_iter(connector, &iter) {
 		if (connector->encoder == encoder) {
 			struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
+
 			amdgpu_encoder->active_device = amdgpu_encoder->devices & amdgpu_connector->devices;
 			DRM_DEBUG_KMS("setting active device to %08x from %08x %08x for encoder %d\n",
 				  amdgpu_encoder->active_device, amdgpu_encoder->devices,
@@ -165,12 +166,12 @@ void amdgpu_panel_mode_fixup(struct drm_encoder *encoder,
 {
 	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
 	struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode;
-	unsigned hblank = native_mode->htotal - native_mode->hdisplay;
-	unsigned vblank = native_mode->vtotal - native_mode->vdisplay;
-	unsigned hover = native_mode->hsync_start - native_mode->hdisplay;
-	unsigned vover = native_mode->vsync_start - native_mode->vdisplay;
-	unsigned hsync_width = native_mode->hsync_end - native_mode->hsync_start;
-	unsigned vsync_width = native_mode->vsync_end - native_mode->vsync_start;
+	unsigned int hblank = native_mode->htotal - native_mode->hdisplay;
+	unsigned int vblank = native_mode->vtotal - native_mode->vdisplay;
+	unsigned int hover = native_mode->hsync_start - native_mode->hdisplay;
+	unsigned int vover = native_mode->vsync_start - native_mode->vdisplay;
+	unsigned int hsync_width = native_mode->hsync_end - native_mode->hsync_start;
+	unsigned int vsync_width = native_mode->vsync_end - native_mode->vsync_start;
 
 	adjusted_mode->clock = native_mode->clock;
 	adjusted_mode->flags = native_mode->flags;
-- 
GitLab


From 5d0622705ff76e017b32cb763cbc7b00694f3b92 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 17 May 2023 20:31:02 +0530
Subject: [PATCH 1353/1791] drm/amd/amdgpu: Fix errors & warnings in
 amdgpu_vcn.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below checkpatch insisted error & warnings:

ERROR: space required before the open brace '{'
WARNING: braces {} are not necessary for any arm of this statement
+       if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK)) {
[...]
+       } else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK)) {
[...]
+       } else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK)) {
[...]
ERROR: code indent should use tabs where possible
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: braces {} are not necessary for single statement blocks
+               for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
+                       fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_enc[i]);
+
ERROR: space required before the open parenthesis '('
WARNING: Missing a blank line after declarations
WARNING: please, no spaces at the start of a line
WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 35 ++++++++++++-------------
 1 file changed, 17 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 06ec2dc558571..c088111c23218 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -169,7 +169,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 	if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
 		bo_size += AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8);
 
-	if (adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0)){
+	if (adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0)) {
 		fw_shared_size = AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared));
 		log_offset = offsetof(struct amdgpu_vcn4_fw_shared, fw_log);
 	} else {
@@ -276,20 +276,19 @@ bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type t
 	bool ret = false;
 	int vcn_config = adev->vcn.vcn_config[vcn_instance];
 
-	if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK)) {
+	if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK))
 		ret = true;
-	} else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK)) {
+	else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK))
 		ret = true;
-	} else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK)) {
+	else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK))
 		ret = true;
-	}
 
 	return ret;
 }
 
 int amdgpu_vcn_suspend(struct amdgpu_device *adev)
 {
-	unsigned size;
+	unsigned int size;
 	void *ptr;
 	int i, idx;
 
@@ -318,7 +317,7 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev)
 
 int amdgpu_vcn_resume(struct amdgpu_device *adev)
 {
-	unsigned size;
+	unsigned int size;
 	void *ptr;
 	int i, idx;
 
@@ -340,7 +339,7 @@ int amdgpu_vcn_resume(struct amdgpu_device *adev)
 			adev->vcn.inst[i].saved_bo = NULL;
 		} else {
 			const struct common_firmware_header *hdr;
-			unsigned offset;
+			unsigned int offset;
 
 			hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
 			if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
@@ -371,9 +370,8 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 		if (adev->vcn.harvest_config & (1 << j))
 			continue;
 
-		for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
+		for (i = 0; i < adev->vcn.num_enc_rings; ++i)
 			fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_enc[i]);
-		}
 
 		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)	{
 			struct dpg_pause_state new_state;
@@ -460,7 +458,7 @@ int amdgpu_vcn_dec_ring_test_ring(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t tmp = 0;
-	unsigned i;
+	unsigned int i;
 	int r;
 
 	/* VCN in SRIOV does not support direct register read/write */
@@ -797,7 +795,7 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t rptr;
-	unsigned i;
+	unsigned int i;
 	int r;
 
 	if (amdgpu_sriov_vf(adev))
@@ -1012,7 +1010,7 @@ int amdgpu_vcn_unified_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 enum amdgpu_ring_priority_level amdgpu_vcn_get_enc_ring_prio(int ring)
 {
-	switch(ring) {
+	switch (ring) {
 	case 0:
 		return AMDGPU_RING_PRIO_0;
 	case 1:
@@ -1031,6 +1029,7 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev)
 
 	if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) {
 		const struct common_firmware_header *hdr;
+
 		hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
 
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
@@ -1059,7 +1058,7 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev)
  */
 #if defined(CONFIG_DEBUG_FS)
 static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf,
-                                             size_t size, loff_t *pos)
+					     size_t size, loff_t *pos)
 {
 	struct amdgpu_vcn_inst *vcn;
 	void *log_buf;
@@ -1105,7 +1104,7 @@ static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf,
 			if (read_pos == AMDGPU_VCNFW_LOG_SIZE)
 				read_pos = plog->header_size;
 			if (read_num[i] == copy_to_user((buf + read_bytes),
-			                                (log_buf + read_pos), read_num[i]))
+							(log_buf + read_pos), read_num[i]))
 				return -EFAULT;
 
 			read_bytes += read_num[i];
@@ -1126,7 +1125,7 @@ static const struct file_operations amdgpu_debugfs_vcnfwlog_fops = {
 #endif
 
 void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i,
-                                   struct amdgpu_vcn_inst *vcn)
+				   struct amdgpu_vcn_inst *vcn)
 {
 #if defined(CONFIG_DEBUG_FS)
 	struct drm_minor *minor = adev_to_drm(adev)->primary;
@@ -1134,7 +1133,7 @@ void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i,
 	char name[32];
 
 	sprintf(name, "amdgpu_vcn_%d_fwlog", i);
-	debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, vcn,
+	debugfs_create_file_size(name, S_IFREG | 0444, root, vcn,
 				 &amdgpu_debugfs_vcnfwlog_fops,
 				 AMDGPU_VCNFW_LOG_SIZE);
 #endif
@@ -1148,7 +1147,7 @@ void amdgpu_vcn_fwlog_init(struct amdgpu_vcn_inst *vcn)
 	uint64_t fw_log_gpu_addr = vcn->fw_shared.gpu_addr + vcn->fw_shared.mem_size;
 	volatile struct amdgpu_vcn_fwlog *log_buf = fw_log_cpu_addr;
 	volatile struct amdgpu_fw_shared_fw_logging *fw_log = vcn->fw_shared.cpu_addr
-                                                         + vcn->fw_shared.log_offset;
+							 + vcn->fw_shared.log_offset;
 	*flag |= cpu_to_le32(AMDGPU_VCN_FW_LOGGING_FLAG);
 	fw_log->is_enabled = 1;
 	fw_log->addr_lo = cpu_to_le32(fw_log_gpu_addr & 0xFFFFFFFF);
-- 
GitLab


From 6c47a79b3b8ba91faf89f9866da2ec16aac979e7 Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Tue, 16 May 2023 17:34:17 +0800
Subject: [PATCH 1354/1791] drm/amdgpu: perform mode2 reset for sdma fed error
 on gfx v11_0_3

perform mode2 reset for sdma fed error on gfx v11_0_3.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c  |  8 +++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h  |  5 +++++
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c | 14 +++++++++++++-
 3 files changed, 25 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 6bb438642cc03..f2da69adcd9d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2053,9 +2053,15 @@ static void amdgpu_ras_do_recovery(struct work_struct *work)
 		/* Perform full reset in fatal error mode */
 		if (!amdgpu_ras_is_poison_mode_supported(ras->adev))
 			set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags);
-		else
+		else {
 			clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags);
 
+			if (ras->gpu_reset_flags & AMDGPU_RAS_GPU_RESET_MODE2_RESET) {
+				ras->gpu_reset_flags &= ~AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+				reset_context.method = AMD_RESET_METHOD_MODE2;
+			}
+		}
+
 		amdgpu_device_gpu_recover(ras->adev, NULL, &reset_context);
 	}
 	atomic_set(&ras->in_recovery, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index bc43f7db17ccf..46bf1889a9d75 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -339,6 +339,8 @@ enum amdgpu_ras_ret {
 #define AMDGPU_RAS_ERR_STATUS_VALID	(1 << 1)
 #define AMDGPU_RAS_ERR_ADDRESS_VALID	(1 << 2)
 
+#define AMDGPU_RAS_GPU_RESET_MODE2_RESET  (0x1 << 0)
+
 struct amdgpu_ras_err_status_reg_entry {
 	uint32_t hwip;
 	uint32_t ip_inst;
@@ -427,6 +429,9 @@ struct amdgpu_ras {
 
 	/* Indicates smu whether need update bad channel info */
 	bool update_channel_flag;
+
+	/* Record special requirements of gpu reset caller */
+	uint32_t  gpu_reset_flags;
 };
 
 struct ras_fs_data {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
index 068b9586a223e..26d6286d86c99 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
@@ -84,8 +84,20 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev,
 	/* Workaround: when vmid and pasid are both zero, trigger gpu reset in KGD. */
 	if (entry && (entry->client_id == SOC21_IH_CLIENTID_GFX) &&
 	    (entry->src_id == GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT) &&
-	     !entry->vmid && !entry->pasid)
+	     !entry->vmid && !entry->pasid) {
+		uint32_t rlc_status0 = 0;
+
+		rlc_status0 = RREG32_SOC15(GC, 0, regRLC_RLCS_FED_STATUS_0);
+
+		if (REG_GET_FIELD(rlc_status0, RLC_RLCS_FED_STATUS_0, SDMA0_FED_ERR) ||
+		    REG_GET_FIELD(rlc_status0, RLC_RLCS_FED_STATUS_0, SDMA1_FED_ERR)) {
+			struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
+
+			ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		}
+
 		amdgpu_ras_reset_gpu(adev);
+	}
 
 	return 0;
 }
-- 
GitLab


From f10984a353c87e696872de8f9614a58689deb0d0 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 17 May 2023 21:24:43 +0530
Subject: [PATCH 1355/1791] drm/amd/amdgpu: Fix errors & warnings in amdgpu
 _uvd, _vce.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below checkpatch errors & warnings:

In amdgpu_uvd.c:

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: Prefer 'unsigned int *' to bare use of 'unsigned *'
WARNING: Missing a blank line after declarations
WARNING: %Lx is non-standard C, use %llx
ERROR: space required before the open parenthesis '('
ERROR: space required before the open brace '{'
WARNING: %LX is non-standard C, use %llX
WARNING: Block comments use * on subsequent lines
+/* multiple fence commands without any stream commands in between can
+   crash the vcpu so just try to emmit a dummy create/destroy msg to
WARNING: Block comments use a trailing */ on a separate line
+   avoid this */
WARNING: braces {} are not necessary for single statement blocks
+               for (j = 0; j < adev->uvd.num_enc_rings; ++j) {
+                       fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring_enc[j]);
+               }

In amdgpu_vce.c:

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: Missing a blank line after declarations
WARNING: %Lx is non-standard C, use %llx
WARNING: Possible repeated word: 'we'
ERROR: space required before the open parenthesis '('

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 83 +++++++++++++------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c | 39 ++++++------
 2 files changed, 63 insertions(+), 59 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index 6887109abb139..b7441654e6fa7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -96,16 +96,16 @@
  */
 struct amdgpu_uvd_cs_ctx {
 	struct amdgpu_cs_parser *parser;
-	unsigned reg, count;
-	unsigned data0, data1;
-	unsigned idx;
+	unsigned int reg, count;
+	unsigned int data0, data1;
+	unsigned int idx;
 	struct amdgpu_ib *ib;
 
 	/* does the IB has a msg command */
 	bool has_msg_cmd;
 
 	/* minimum buffer sizes */
-	unsigned *buf_sizes;
+	unsigned int *buf_sizes;
 };
 
 #ifdef CONFIG_DRM_AMDGPU_SI
@@ -186,7 +186,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 	unsigned long bo_size;
 	const char *fw_name;
 	const struct common_firmware_header *hdr;
-	unsigned family_id;
+	unsigned int family_id;
 	int i, j, r;
 
 	INIT_DELAYED_WORK(&adev->uvd.idle_work, amdgpu_uvd_idle_work_handler);
@@ -275,7 +275,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 	family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
 
 	if (adev->asic_type < CHIP_VEGA20) {
-		unsigned version_major, version_minor;
+		unsigned int version_major, version_minor;
 
 		version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
 		version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
@@ -420,7 +420,7 @@ int amdgpu_uvd_entity_init(struct amdgpu_device *adev)
 
 int amdgpu_uvd_suspend(struct amdgpu_device *adev)
 {
-	unsigned size;
+	unsigned int size;
 	void *ptr;
 	int i, j, idx;
 	bool in_ras_intr = amdgpu_ras_intr_triggered();
@@ -469,7 +469,7 @@ int amdgpu_uvd_suspend(struct amdgpu_device *adev)
 
 int amdgpu_uvd_resume(struct amdgpu_device *adev)
 {
-	unsigned size;
+	unsigned int size;
 	void *ptr;
 	int i, idx;
 
@@ -491,7 +491,7 @@ int amdgpu_uvd_resume(struct amdgpu_device *adev)
 			adev->uvd.inst[i].saved_bo = NULL;
 		} else {
 			const struct common_firmware_header *hdr;
-			unsigned offset;
+			unsigned int offset;
 
 			hdr = (const struct common_firmware_header *)adev->uvd.fw->data;
 			if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
@@ -542,6 +542,7 @@ void amdgpu_uvd_free_handles(struct amdgpu_device *adev, struct drm_file *filp)
 static void amdgpu_uvd_force_into_uvd_segment(struct amdgpu_bo *abo)
 {
 	int i;
+
 	for (i = 0; i < abo->placement.num_placement; ++i) {
 		abo->placements[i].fpfn = 0 >> PAGE_SHIFT;
 		abo->placements[i].lpfn = (256 * 1024 * 1024) >> PAGE_SHIFT;
@@ -579,7 +580,7 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx)
 
 	r = amdgpu_cs_find_mapping(ctx->parser, addr, &bo, &mapping);
 	if (r) {
-		DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr);
+		DRM_ERROR("Can't find BO for addr 0x%08llx\n", addr);
 		return r;
 	}
 
@@ -589,6 +590,7 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx)
 		if (cmd == 0x0 || cmd == 0x3) {
 			/* yes, force it into VRAM */
 			uint32_t domain = AMDGPU_GEM_DOMAIN_VRAM;
+
 			amdgpu_bo_placement_from_domain(bo, domain);
 		}
 		amdgpu_uvd_force_into_uvd_segment(bo);
@@ -609,21 +611,21 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx)
  * Peek into the decode message and calculate the necessary buffer sizes.
  */
 static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
-	unsigned buf_sizes[])
+	unsigned int buf_sizes[])
 {
-	unsigned stream_type = msg[4];
-	unsigned width = msg[6];
-	unsigned height = msg[7];
-	unsigned dpb_size = msg[9];
-	unsigned pitch = msg[28];
-	unsigned level = msg[57];
+	unsigned int stream_type = msg[4];
+	unsigned int width = msg[6];
+	unsigned int height = msg[7];
+	unsigned int dpb_size = msg[9];
+	unsigned int pitch = msg[28];
+	unsigned int level = msg[57];
 
-	unsigned width_in_mb = width / 16;
-	unsigned height_in_mb = ALIGN(height / 16, 2);
-	unsigned fs_in_mb = width_in_mb * height_in_mb;
+	unsigned int width_in_mb = width / 16;
+	unsigned int height_in_mb = ALIGN(height / 16, 2);
+	unsigned int fs_in_mb = width_in_mb * height_in_mb;
 
-	unsigned image_size, tmp, min_dpb_size, num_dpb_buffer;
-	unsigned min_ctx_size = ~0;
+	unsigned int image_size, tmp, min_dpb_size, num_dpb_buffer;
+	unsigned int min_ctx_size = ~0;
 
 	image_size = width * height;
 	image_size += image_size / 2;
@@ -631,7 +633,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
 
 	switch (stream_type) {
 	case 0: /* H264 */
-		switch(level) {
+		switch (level) {
 		case 30:
 			num_dpb_buffer = 8100 / fs_in_mb;
 			break;
@@ -709,7 +711,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
 		break;
 
 	case 7: /* H264 Perf */
-		switch(level) {
+		switch (level) {
 		case 30:
 			num_dpb_buffer = 8100 / fs_in_mb;
 			break;
@@ -742,7 +744,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
 		/* reference picture buffer */
 		min_dpb_size = image_size * num_dpb_buffer;
 
-		if (!adev->uvd.use_ctx_buf){
+		if (!adev->uvd.use_ctx_buf) {
 			/* macroblock context buffer */
 			min_dpb_size +=
 				width_in_mb * height_in_mb * num_dpb_buffer * 192;
@@ -805,7 +807,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg,
  * Make sure that we don't open up to many sessions.
  */
 static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx,
-			     struct amdgpu_bo *bo, unsigned offset)
+			     struct amdgpu_bo *bo, unsigned int offset)
 {
 	struct amdgpu_device *adev = ctx->parser->adev;
 	int32_t *msg, msg_type, handle;
@@ -911,7 +913,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx)
 
 	r = amdgpu_cs_find_mapping(ctx->parser, addr, &bo, &mapping);
 	if (r) {
-		DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr);
+		DRM_ERROR("Can't find BO for addr 0x%08llx\n", addr);
 		return r;
 	}
 
@@ -930,7 +932,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx)
 	if (cmd < 0x4) {
 		if ((end - start) < ctx->buf_sizes[cmd]) {
 			DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd,
-				  (unsigned)(end - start),
+				  (unsigned int)(end - start),
 				  ctx->buf_sizes[cmd]);
 			return -EINVAL;
 		}
@@ -938,7 +940,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx)
 	} else if (cmd == 0x206) {
 		if ((end - start) < ctx->buf_sizes[4]) {
 			DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd,
-					  (unsigned)(end - start),
+					  (unsigned int)(end - start),
 					  ctx->buf_sizes[4]);
 			return -EINVAL;
 		}
@@ -949,14 +951,14 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx)
 
 	if (!ctx->parser->adev->uvd.address_64_bit) {
 		if ((start >> 28) != ((end - 1) >> 28)) {
-			DRM_ERROR("reloc %LX-%LX crossing 256MB boundary!\n",
+			DRM_ERROR("reloc %llx-%llx crossing 256MB boundary!\n",
 				  start, end);
 			return -EINVAL;
 		}
 
 		if ((cmd == 0 || cmd == 0x3) &&
 		    (start >> 28) != (ctx->parser->adev->uvd.inst->gpu_addr >> 28)) {
-			DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n",
+			DRM_ERROR("msg/fb buffer %llx-%llx out of 256MB segment!\n",
 				  start, end);
 			return -EINVAL;
 		}
@@ -990,7 +992,7 @@ static int amdgpu_uvd_cs_reg(struct amdgpu_uvd_cs_ctx *ctx,
 
 	ctx->idx++;
 	for (i = 0; i <= ctx->count; ++i) {
-		unsigned reg = ctx->reg + i;
+		unsigned int reg = ctx->reg + i;
 
 		if (ctx->idx >= ctx->ib->length_dw) {
 			DRM_ERROR("Register command after end of CS!\n");
@@ -1036,7 +1038,8 @@ static int amdgpu_uvd_cs_packets(struct amdgpu_uvd_cs_ctx *ctx,
 
 	for (ctx->idx = 0 ; ctx->idx < ctx->ib->length_dw; ) {
 		uint32_t cmd = amdgpu_ib_get_value(ctx->ib, ctx->idx);
-		unsigned type = CP_PACKET_GET_TYPE(cmd);
+		unsigned int type = CP_PACKET_GET_TYPE(cmd);
+
 		switch (type) {
 		case PACKET_TYPE0:
 			ctx->reg = CP_PACKET0_GET_REG(cmd);
@@ -1070,7 +1073,7 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser,
 			     struct amdgpu_ib *ib)
 {
 	struct amdgpu_uvd_cs_ctx ctx = {};
-	unsigned buf_sizes[] = {
+	unsigned int buf_sizes[] = {
 		[0x00000000]	=	2048,
 		[0x00000001]	=	0xFFFFFFFF,
 		[0x00000002]	=	0xFFFFFFFF,
@@ -1185,8 +1188,9 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo,
 }
 
 /* multiple fence commands without any stream commands in between can
-   crash the vcpu so just try to emmit a dummy create/destroy msg to
-   avoid this */
+ * crash the vcpu so just try to emmit a dummy create/destroy msg to
+ * avoid this
+ */
 int amdgpu_uvd_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 			      struct dma_fence **fence)
 {
@@ -1252,15 +1256,14 @@ static void amdgpu_uvd_idle_work_handler(struct work_struct *work)
 {
 	struct amdgpu_device *adev =
 		container_of(work, struct amdgpu_device, uvd.idle_work.work);
-	unsigned fences = 0, i, j;
+	unsigned int fences = 0, i, j;
 
 	for (i = 0; i < adev->uvd.num_uvd_inst; ++i) {
 		if (adev->uvd.harvest_config & (1 << i))
 			continue;
 		fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring);
-		for (j = 0; j < adev->uvd.num_enc_rings; ++j) {
+		for (j = 0; j < adev->uvd.num_enc_rings; ++j)
 			fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring_enc[j]);
-		}
 	}
 
 	if (fences == 0) {
@@ -1356,7 +1359,7 @@ int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout)
  */
 uint32_t amdgpu_uvd_used_handles(struct amdgpu_device *adev)
 {
-	unsigned i;
+	unsigned int i;
 	uint32_t used_handles = 0;
 
 	for (i = 0; i < adev->uvd.max_handles; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index e2b7324a70cbb..1904edf684071 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -99,7 +99,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 {
 	const char *fw_name;
 	const struct common_firmware_header *hdr;
-	unsigned ucode_version, version_major, version_minor, binary_id;
+	unsigned int ucode_version, version_major, version_minor, binary_id;
 	int i, r;
 
 	switch (adev->asic_type) {
@@ -207,7 +207,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
  */
 int amdgpu_vce_sw_fini(struct amdgpu_device *adev)
 {
-	unsigned i;
+	unsigned int i;
 
 	if (adev->vce.vcpu_bo == NULL)
 		return 0;
@@ -286,7 +286,7 @@ int amdgpu_vce_resume(struct amdgpu_device *adev)
 {
 	void *cpu_addr;
 	const struct common_firmware_header *hdr;
-	unsigned offset;
+	unsigned int offset;
 	int r, idx;
 
 	if (adev->vce.vcpu_bo == NULL)
@@ -332,7 +332,7 @@ static void amdgpu_vce_idle_work_handler(struct work_struct *work)
 {
 	struct amdgpu_device *adev =
 		container_of(work, struct amdgpu_device, vce.idle_work.work);
-	unsigned i, count = 0;
+	unsigned int i, count = 0;
 
 	for (i = 0; i < adev->vce.num_rings; i++)
 		count += amdgpu_fence_count_emitted(&adev->vce.ring[i]);
@@ -409,6 +409,7 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp)
 {
 	struct amdgpu_ring *ring = &adev->vce.ring[0];
 	int i, r;
+
 	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
 		uint32_t handle = atomic_read(&adev->vce.handles[i]);
 
@@ -436,7 +437,7 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp)
 static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 				     struct dma_fence **fence)
 {
-	const unsigned ib_size_dw = 1024;
+	const unsigned int ib_size_dw = 1024;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct amdgpu_ib ib_msg;
@@ -528,7 +529,7 @@ static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 				      bool direct, struct dma_fence **fence)
 {
-	const unsigned ib_size_dw = 1024;
+	const unsigned int ib_size_dw = 1024;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct dma_fence *f = NULL;
@@ -596,12 +597,12 @@ static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
  */
 static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p,
 				  struct amdgpu_ib *ib, int lo, int hi,
-				  unsigned size, int32_t index)
+				  unsigned int size, int32_t index)
 {
 	int64_t offset = ((uint64_t)size) * ((int64_t)index);
 	struct ttm_operation_ctx ctx = { false, false };
 	struct amdgpu_bo_va_mapping *mapping;
-	unsigned i, fpfn, lpfn;
+	unsigned int i, fpfn, lpfn;
 	struct amdgpu_bo *bo;
 	uint64_t addr;
 	int r;
@@ -619,7 +620,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p,
 
 	r = amdgpu_cs_find_mapping(p, addr, &bo, &mapping);
 	if (r) {
-		DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d %d %d\n",
+		DRM_ERROR("Can't find BO for addr 0x%010llx %d %d %d %d\n",
 			  addr, lo, hi, size, index);
 		return r;
 	}
@@ -646,7 +647,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p,
  * Patch relocation inside command stream with real buffer address
  */
 static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib,
-			       int lo, int hi, unsigned size, uint32_t index)
+			       int lo, int hi, unsigned int size, uint32_t index)
 {
 	struct amdgpu_bo_va_mapping *mapping;
 	struct amdgpu_bo *bo;
@@ -662,14 +663,14 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib,
 
 	r = amdgpu_cs_find_mapping(p, addr, &bo, &mapping);
 	if (r) {
-		DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d %d %d\n",
+		DRM_ERROR("Can't find BO for addr 0x%010llx %d %d %d %d\n",
 			  addr, lo, hi, size, index);
 		return r;
 	}
 
 	if ((addr + (uint64_t)size) >
 	    (mapping->last + 1) * AMDGPU_GPU_PAGE_SIZE) {
-		DRM_ERROR("BO too small for addr 0x%010Lx %d %d\n",
+		DRM_ERROR("BO too small for addr 0x%010llx %d %d\n",
 			  addr, lo, hi);
 		return -EINVAL;
 	}
@@ -692,12 +693,12 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib,
  * @allocated: allocated a new handle?
  *
  * Validates the handle and return the found session index or -EINVAL
- * we we don't have another free session index.
+ * we don't have another free session index.
  */
 static int amdgpu_vce_validate_handle(struct amdgpu_cs_parser *p,
 				      uint32_t handle, uint32_t *allocated)
 {
-	unsigned i;
+	unsigned int i;
 
 	/* validate the handle */
 	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
@@ -735,14 +736,14 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p,
 			     struct amdgpu_job *job,
 			     struct amdgpu_ib *ib)
 {
-	unsigned fb_idx = 0, bs_idx = 0;
+	unsigned int fb_idx = 0, bs_idx = 0;
 	int session_idx = -1;
 	uint32_t destroyed = 0;
 	uint32_t created = 0;
 	uint32_t allocated = 0;
 	uint32_t tmp, handle = 0;
 	uint32_t *size = &tmp;
-	unsigned idx;
+	unsigned int idx;
 	int i, r = 0;
 
 	job->vm = NULL;
@@ -1084,7 +1085,7 @@ void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring,
  *
  */
 void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
-				unsigned flags)
+				unsigned int flags)
 {
 	WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT);
 
@@ -1106,7 +1107,7 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	uint32_t rptr;
-	unsigned i;
+	unsigned int i;
 	int r, timeout = adev->usec_timeout;
 
 	/* skip ring test for sriov*/
@@ -1171,7 +1172,7 @@ int amdgpu_vce_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 
 enum amdgpu_ring_priority_level amdgpu_vce_get_ring_prio(int ring)
 {
-	switch(ring) {
+	switch (ring) {
 	case 0:
 		return AMDGPU_RING_PRIO_0;
 	case 1:
-- 
GitLab


From e03f04b84901644c81b4348a813a8d17facbd277 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 18 May 2023 07:52:06 +0530
Subject: [PATCH 1356/1791] drm/amdgpu: Fix warnings in amdgpu _sdma, _ucode.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below checkpatch warnings:

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: Comparisons should place the constant on the right side of the test
WARNING: Missing a blank line after declarations

Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c  | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c | 5 +++--
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
index 231ca06bc9c78..9568adaad5cbe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
@@ -64,7 +64,7 @@ int amdgpu_sdma_get_index_from_ring(struct amdgpu_ring *ring, uint32_t *index)
 }
 
 uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring,
-				     unsigned vmid)
+				     unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
 	uint64_t csa_mc_addr;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
index f76b1cb8baf85..16807ff96dc9b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c
@@ -748,7 +748,7 @@ static int amdgpu_ucode_init_single_fw(struct amdgpu_device *adev,
 	const struct imu_firmware_header_v1_0 *imu_hdr = NULL;
 	u8 *ucode_addr;
 
-	if (NULL == ucode->fw)
+	if (!ucode->fw)
 		return 0;
 
 	ucode->mc_addr = mc_addr;
@@ -972,7 +972,7 @@ static int amdgpu_ucode_patch_jt(struct amdgpu_firmware_info *ucode,
 	uint8_t *src_addr = NULL;
 	uint8_t *dst_addr = NULL;
 
-	if (NULL == ucode->fw)
+	if (!ucode->fw)
 		return 0;
 
 	comm_hdr = (const struct common_firmware_header *)ucode->fw->data;
@@ -1043,6 +1043,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev)
 			if (i == AMDGPU_UCODE_ID_CP_MEC1 &&
 			    adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
 				const struct gfx_firmware_header_v1_0 *cp_hdr;
+
 				cp_hdr = (const struct gfx_firmware_header_v1_0 *)ucode->fw->data;
 				amdgpu_ucode_patch_jt(ucode,  adev->firmware.fw_buf_mc + fw_offset,
 						    adev->firmware.fw_buf_ptr + fw_offset);
-- 
GitLab


From e602157ec089240861cd641ee2c7c64eeaec09bf Mon Sep 17 00:00:00 2001
From: Jack Xiao <Jack.Xiao@amd.com>
Date: Wed, 17 May 2023 17:07:01 +0800
Subject: [PATCH 1357/1791] drm/amdgpu: fix S3 issue if MQD in VRAM

1. Need flush HDP for MQD putting in vram
2. Zero out mes MQD

Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 4 ++++
 drivers/gpu/drm/amd/amdgpu/mes_v10_1.c  | 3 +++
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c  | 3 +++
 3 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 8883d5ee13cb5..f2d0b1d55d771 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -593,6 +593,8 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id)
 
 	DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe,
 							kiq_ring->queue);
+	amdgpu_device_flush_hdp(adev, NULL);
+
 	spin_lock(&kiq->ring_lock);
 	r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size *
 					adev->gfx.num_compute_rings +
@@ -630,6 +632,8 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id)
 	if (!kiq->pmf || !kiq->pmf->kiq_map_queues)
 		return -EINVAL;
 
+	amdgpu_device_flush_hdp(adev, NULL);
+
 	spin_lock(&kiq->ring_lock);
 	/* No need to map kcq on the slave */
 	if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) {
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index f1a6abdad21b6..88262f10ef7c5 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -632,6 +632,8 @@ static int mes_v10_1_mqd_init(struct amdgpu_ring *ring)
 	uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr;
 	uint32_t tmp;
 
+	memset(mqd, 0, sizeof(*mqd));
+
 	mqd->header = 0xC0310800;
 	mqd->compute_pipelinestat_enable = 0x00000001;
 	mqd->compute_static_thread_mgmt_se0 = 0xffffffff;
@@ -728,6 +730,7 @@ static int mes_v10_1_mqd_init(struct amdgpu_ring *ring)
 	/* offset: 184 - this is used for CP_HQD_GFX_CONTROL */
 	mqd->cp_hqd_suspend_cntl_stack_offset = tmp;
 
+	amdgpu_device_flush_hdp(ring->adev, NULL);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 9791f35817865..9a48328c6572d 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -704,6 +704,8 @@ static int mes_v11_0_mqd_init(struct amdgpu_ring *ring)
 	uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr;
 	uint32_t tmp;
 
+	memset(mqd, 0, sizeof(*mqd));
+
 	mqd->header = 0xC0310800;
 	mqd->compute_pipelinestat_enable = 0x00000001;
 	mqd->compute_static_thread_mgmt_se0 = 0xffffffff;
@@ -797,6 +799,7 @@ static int mes_v11_0_mqd_init(struct amdgpu_ring *ring)
 	mqd->cp_hqd_iq_timer = regCP_HQD_IQ_TIMER_DEFAULT;
 	mqd->cp_hqd_quantum = regCP_HQD_QUANTUM_DEFAULT;
 
+	amdgpu_device_flush_hdp(ring->adev, NULL);
 	return 0;
 }
 
-- 
GitLab


From 48dd83c0fb6c68742f7fefca907036942dd358be Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Wed, 17 May 2023 13:31:45 -0400
Subject: [PATCH 1358/1791] drm/amd/display: drop redundant memset() in
 get_available_dsc_slices()

get_available_dsc_slices() returns the number of indices set, and all of
the users of get_available_dsc_slices() don't cross the returned bound
when iterating over available_slices[]. So, the memset() in
get_available_dsc_slices() is redundant and can be dropped.

Fixes: 97bda0322b8a ("drm/amd/display: Add DSC support for Navi (v2)")
Reported-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index b9a05bb025db6..58dd62cce4bb9 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -645,8 +645,6 @@ static int get_available_dsc_slices(union dsc_enc_slice_caps slice_caps, int *av
 {
 	int idx = 0;
 
-	memset(available_slices, -1, MIN_AVAILABLE_SLICES_SIZE);
-
 	if (slice_caps.bits.NUM_SLICES_1)
 		available_slices[idx++] = 1;
 
-- 
GitLab


From fd73c8507675f6bccc039cf319f183e41e447cb7 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Wed, 17 May 2023 13:49:29 -0400
Subject: [PATCH 1359/1791] drm/amd/display: drop unused function
 set_abm_event()

set_abm_event() is never actually used. So, drop it.

Fixes: b8fe56375f78 ("drm/amd/display: Refactor ABM feature")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Tom Rix <trix@redhat.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c | 12 ------------
 drivers/gpu/drm/amd/display/dc/inc/hw/abm.h   |  2 --
 2 files changed, 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
index a66f83a61402d..2fb9572ce25db 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c
@@ -131,17 +131,6 @@ static bool dmub_abm_set_pipe_ex(struct abm *abm, uint32_t otg_inst, uint32_t op
 	return ret;
 }
 
-static bool dmub_abm_set_event_ex(struct abm *abm, unsigned int full_screen, unsigned int video_mode,
-		unsigned int hdr_mode, unsigned int panel_inst)
-{
-	bool ret = false;
-	unsigned int feature_support;
-
-	feature_support = abm_feature_support(abm, panel_inst);
-
-	return ret;
-}
-
 static bool dmub_abm_set_backlight_level_pwm_ex(struct abm *abm,
 		unsigned int backlight_pwm_u16_16,
 		unsigned int frame_ramp,
@@ -167,7 +156,6 @@ static const struct abm_funcs abm_funcs = {
 	.init_abm_config = dmub_abm_init_config_ex,
 	.set_abm_pause = dmub_abm_set_pause_ex,
 	.set_pipe_ex = dmub_abm_set_pipe_ex,
-	.set_abm_event = dmub_abm_set_event_ex,
 	.set_backlight_level_pwm = dmub_abm_set_backlight_level_pwm_ex,
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
index db5cf9acafe6d..d2190a3320f64 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
@@ -59,8 +59,6 @@ struct abm_funcs {
 			unsigned int otg_inst,
 			unsigned int option,
 			unsigned int panel_inst);
-	bool (*set_abm_event)(struct abm *abm, unsigned int full_screen, unsigned int video_mode,
-			unsigned int hdr_mode, unsigned int panel_inst);
 };
 
 #endif
-- 
GitLab


From 8789989b476b5f3bb0bf1a63b5223f6e76cfd13d Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Wed, 17 May 2023 14:09:39 -0400
Subject: [PATCH 1360/1791] drm/amd/display: drop unused count variable in
 create_eml_sink()

Since, we are only interested in having
drm_edid_override_connector_update(), update the value of
connector->edid_blob_ptr. We don't care about the return value of
drm_edid_override_connector_update() here. So, drop count.

Fixes: 550e5d23f147 ("drm/amd/display: assign edid_blob_ptr with edid from debugfs")
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index e1c4db673dea4..dfdd419310d94 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6397,9 +6397,8 @@ static void create_eml_sink(struct amdgpu_dm_connector *aconnector)
 		/* if connector->edid_override valid, pass
 		 * it to edid_override to edid_blob_ptr
 		 */
-		int count;
 
-		count = drm_edid_override_connector_update(&aconnector->base);
+		drm_edid_override_connector_update(&aconnector->base);
 
 		if (!aconnector->base.edid_blob_ptr) {
 			DRM_ERROR("No EDID firmware found on connector: %s ,forcing to OFF!\n",
-- 
GitLab


From 04e8595819480bec2754bd2cc4377270e7e6c799 Mon Sep 17 00:00:00 2001
From: Tong Liu01 <Tong.Liu01@amd.com>
Date: Tue, 16 May 2023 14:50:04 +0800
Subject: [PATCH 1361/1791] drm/amdgpu: fix incorrect pcie_gen_mask in
 passthrough case

[why]
Passthrough case is treated as root bus and pcie_gen_mask is set as
default value that does not support GEN 3 and GEN 4 for PCIe link
speed. So PCIe link speed will be downgraded at smu hw init in
passthrough condition

[how]
Move get pci info after detect virtualization and check if it is
passthrough case when set pcie_gen_mask

Signed-off-by: Tong Liu01 <Tong.Liu01@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index f0666230b2ed0..ff9ca0dbeb5ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3814,8 +3814,6 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base);
 	DRM_INFO("register mmio size: %u\n", (unsigned)adev->rmmio_size);
 
-	amdgpu_device_get_pcie_info(adev);
-
 	if (amdgpu_mcbp)
 		DRM_INFO("MCBP is enabled\n");
 
@@ -3831,6 +3829,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	/* detect hw virtualization here */
 	amdgpu_detect_virtualization(adev);
 
+	amdgpu_device_get_pcie_info(adev);
+
 	r = amdgpu_device_get_job_timeout_settings(adev);
 	if (r) {
 		dev_err(adev->dev, "invalid lockup_timeout parameter syntax\n");
@@ -5588,7 +5588,7 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev)
 		adev->pm.pcie_mlw_mask = amdgpu_pcie_lane_cap;
 
 	/* covers APUs as well */
-	if (pci_is_root_bus(adev->pdev->bus)) {
+	if (pci_is_root_bus(adev->pdev->bus) && !amdgpu_passthrough(adev)) {
 		if (adev->pm.pcie_gen_mask == 0)
 			adev->pm.pcie_gen_mask = AMDGPU_DEFAULT_PCIE_GEN_MASK;
 		if (adev->pm.pcie_mlw_mask == 0)
-- 
GitLab


From ea2a50ae3d4c9857c8967df44b0b2b42eee7d7c9 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 18 May 2023 09:30:26 -0500
Subject: [PATCH 1362/1791] drm/amd: Update driver-misc.html for Phoenix

AMD has added marketing information publicly for Phoenix, so
update the APU table with matching versions.

Link: https://www.amd.com/en/product/13036
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/apu-asic-info-table.csv | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index 395a7b7bfaefb..9ef623bd684e2 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -8,3 +8,4 @@ Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barce
 Ryzen 6000 series / Ryzen 7x35 series, YELLOW CARP / Rembrandt / Rembrandt+, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3
 Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
 Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8
+Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
\ No newline at end of file
-- 
GitLab


From c0dba2d34ac153414c4c09e073f5582eeb67a252 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 18 May 2023 09:35:17 -0500
Subject: [PATCH 1363/1791] drm/amd: Update driver-misc.html for Dragon Range

AMD has added marketing information publicly for Dragon Range, so
update the APU table with matching versions.

Link: https://www.amd.com/en/product/13016
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/apu-asic-info-table.csv | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index 9ef623bd684e2..b5c67059e9bea 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -7,5 +7,6 @@ SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0
 Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1
 Ryzen 6000 series / Ryzen 7x35 series, YELLOW CARP / Rembrandt / Rembrandt+, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3
 Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
+Ryzen 7x45 series (FL1), / Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
 Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8
 Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
\ No newline at end of file
-- 
GitLab


From 109b4d8cfe4279da1cbcbcd99ae54cb2b2aee521 Mon Sep 17 00:00:00 2001
From: Su Hui <suhui@nfschina.com>
Date: Mon, 15 May 2023 09:34:28 +0800
Subject: [PATCH 1364/1791] drm/amdgpu: remove unnecessary (void*) conversions

No need cast (void*) to (struct amdgpu_device *).

Signed-off-by: Su Hui <suhui@nfschina.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c   | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c     | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c      | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c     | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c     | 2 +-
 6 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index e94d0cf3f7936..c657bed350acc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -1638,7 +1638,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
 
 static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 	struct drm_device *dev = adev_to_drm(adev);
 	int r = 0, i;
 
@@ -1749,7 +1749,7 @@ static int amdgpu_debugfs_benchmark(void *data, u64 val)
 
 static int amdgpu_debugfs_vm_info_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_file *file;
 	int r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 928814492d5b0..876ec35b8f83d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -833,7 +833,7 @@ static const struct dma_fence_ops amdgpu_job_fence_ops = {
 #if defined(CONFIG_DEBUG_FS)
 static int amdgpu_debugfs_fence_info_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 	int i;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index fec9df3546525..74055cba3dc9a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -951,7 +951,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv,
 #if defined(CONFIG_DEBUG_FS)
 static int amdgpu_debugfs_gem_info_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 	struct drm_device *dev = adev_to_drm(adev);
 	struct drm_file *file;
 	int r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index b27ac48ca123e..ebeddc9a37e9b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -463,7 +463,7 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
 
 static int amdgpu_debugfs_sa_info_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 
 	seq_printf(m, "--------------------- DELAYED --------------------- \n");
 	amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DELAYED],
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 815098be4c2fd..41d047e5de696 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1468,7 +1468,7 @@ void amdgpu_disable_vblank_kms(struct drm_crtc *crtc)
 
 static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 	struct drm_amdgpu_info_firmware fw_info;
 	struct drm_amdgpu_query_fw query_fw;
 	struct atom_context *ctx = adev->mode_info.atom_context;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index b8adcebb0d89a..509b26e9d0c4a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -2257,7 +2257,7 @@ int amdgpu_ttm_evict_resources(struct amdgpu_device *adev, int mem_type)
 
 static int amdgpu_ttm_page_pool_show(struct seq_file *m, void *unused)
 {
-	struct amdgpu_device *adev = (struct amdgpu_device *)m->private;
+	struct amdgpu_device *adev = m->private;
 
 	return ttm_pool_debugfs(&adev->mman.bdev.pool, m);
 }
-- 
GitLab


From e5edd9ab0e291912cf68abf5dc84d393d29a60a7 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 18 May 2023 09:40:09 -0500
Subject: [PATCH 1365/1791] drm/amd: Update driver-misc.html for Rembrandt-R

AMD has added marketing information publicly for Rembrandt-R, so
update the APU table with matching versions.

Link: https://www.amd.com/en/product/13086
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/apu-asic-info-table.csv | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index b5c67059e9bea..2e76b427ba1ee 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -5,7 +5,7 @@ Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3
 Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1
 SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0
 Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1
-Ryzen 6000 series / Ryzen 7x35 series, YELLOW CARP / Rembrandt / Rembrandt+, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3
+Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3
 Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
 Ryzen 7x45 series (FL1), / Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
 Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8
-- 
GitLab


From a2b308044dcaca8d3e580959a4f867a1d5c37fac Mon Sep 17 00:00:00 2001
From: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Date: Sat, 13 May 2023 14:51:00 +0200
Subject: [PATCH 1366/1791] drm/amdgpu: Validate VM ioctl flags.

None have been defined yet, so reject anybody setting any. Mesa sets
it to 0 anyway.

Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 22a900f298f2e..22f9a65ca0fc7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2370,6 +2370,10 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	int r;
 
+	/* No valid flags defined yet */
+	if (args->in.flags)
+		return -EINVAL;
+
 	switch (args->in.op) {
 	case AMDGPU_VM_OP_RESERVE_VMID:
 		/* We only have requirement to reserve vmid from gfxhub */
-- 
GitLab


From b18f05a0666aecd5cb19c26a8305bcfa4e9d6502 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Wed, 17 May 2023 14:39:46 -0400
Subject: [PATCH 1367/1791] drm/amd/display: Fix artifacting on eDP panels when
 engaging freesync video mode

[Why]
When freesync video mode is enabled, switching resolution from native
mode to one of the freesync video compatible modes can trigger continous
artifacts on some eDP panels when running under KDE. The articating can be seen in the
attached bug report.

[How]
Fix this by restricting updates that require full commit by using the same checks
for stream and scaling changes in the the enable pass of dm_update_crtc_state()
along with the check for compatible timings for freesync vide mode.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2162
Fixes: da5e14909776 ("drm/amd/display: Fix hang when skipping modeset")
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index dfdd419310d94..abef3c2f59c8d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -9285,6 +9285,8 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 
 		/* Now check if we should set freesync video mode */
 		if (amdgpu_freesync_vid_mode && dm_new_crtc_state->stream &&
+		    dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) &&
+		    dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream) &&
 		    is_timing_unchanged_for_freesync(new_crtc_state,
 						     old_crtc_state)) {
 			new_crtc_state->mode_changed = false;
-- 
GitLab


From 1385d88c6aa774332f1a88562b6f1bf04de6d710 Mon Sep 17 00:00:00 2001
From: Sukrut Bellary <sukrut.bellary@linux.com>
Date: Wed, 3 May 2023 16:15:07 -0700
Subject: [PATCH 1368/1791] drm:amd:amdgpu: Fix missing buffer object unlock in
 failure path

smatch warning -
1) drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c:3615 gfx_v9_0_kiq_resume()
warn: inconsistent returns 'ring->mqd_obj->tbo.base.resv'.

2) drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c:6901 gfx_v10_0_kiq_resume()
warn: inconsistent returns 'ring->mqd_obj->tbo.base.resv'.

Signed-off-by: Sukrut Bellary <sukrut.bellary@linux.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 4 +++-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c  | 4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 89158c72753e8..0d15002eac691 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -6803,8 +6803,10 @@ static int gfx_v10_0_kiq_resume(struct amdgpu_device *adev)
 		return r;
 
 	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
-	if (unlikely(r != 0))
+	if (unlikely(r != 0)) {
+		amdgpu_bo_unreserve(ring->mqd_obj);
 		return r;
+	}
 
 	gfx_v10_0_kiq_init_queue(ring);
 	amdgpu_bo_kunmap(ring->mqd_obj);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index de8e70b3db75c..8bf95a6b07677 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -3614,8 +3614,10 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev)
 		return r;
 
 	r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
-	if (unlikely(r != 0))
+	if (unlikely(r != 0)) {
+		amdgpu_bo_unreserve(ring->mqd_obj);
 		return r;
+	}
 
 	gfx_v9_0_kiq_init_queue(ring);
 	amdgpu_bo_kunmap(ring->mqd_obj);
-- 
GitLab


From 6091ede913015fd3c07cb16298505bbd71f41689 Mon Sep 17 00:00:00 2001
From: Su Hui <suhui@nfschina.com>
Date: Wed, 17 May 2023 10:52:19 +0800
Subject: [PATCH 1369/1791] drm/radeon: Remove unnecessary (void*) conversions

No need cast (void*) to (struct radeon_device *)
or (struct radeon_ring *).

Signed-off-by: Su Hui <suhui@nfschina.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/r100.c         | 8 ++++----
 drivers/gpu/drm/radeon/r300.c         | 2 +-
 drivers/gpu/drm/radeon/r420.c         | 2 +-
 drivers/gpu/drm/radeon/r600.c         | 2 +-
 drivers/gpu/drm/radeon/radeon_fence.c | 2 +-
 drivers/gpu/drm/radeon/radeon_gem.c   | 2 +-
 drivers/gpu/drm/radeon/radeon_ib.c    | 2 +-
 drivers/gpu/drm/radeon/radeon_pm.c    | 2 +-
 drivers/gpu/drm/radeon/radeon_ring.c  | 2 +-
 drivers/gpu/drm/radeon/radeon_ttm.c   | 2 +-
 drivers/gpu/drm/radeon/rs400.c        | 2 +-
 drivers/gpu/drm/radeon/rv515.c        | 4 ++--
 12 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index d4f09ecc3d221..affa9e0309b27 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -2929,7 +2929,7 @@ static void r100_set_safe_registers(struct radeon_device *rdev)
 #if defined(CONFIG_DEBUG_FS)
 static int r100_debugfs_rbbm_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t reg, value;
 	unsigned i;
 
@@ -2948,7 +2948,7 @@ static int r100_debugfs_rbbm_info_show(struct seq_file *m, void *unused)
 
 static int r100_debugfs_cp_ring_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 	uint32_t rdp, wdp;
 	unsigned count, i, j;
@@ -2974,7 +2974,7 @@ static int r100_debugfs_cp_ring_info_show(struct seq_file *m, void *unused)
 
 static int r100_debugfs_cp_csq_fifo_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t csq_stat, csq2_stat, tmp;
 	unsigned r_rptr, r_wptr, ib1_rptr, ib1_wptr, ib2_rptr, ib2_wptr;
 	unsigned i;
@@ -3022,7 +3022,7 @@ static int r100_debugfs_cp_csq_fifo_show(struct seq_file *m, void *unused)
 
 static int r100_debugfs_mc_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32(RADEON_CONFIG_MEMSIZE);
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 7b0cfeaddcec3..9c1a92fa2af6d 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -589,7 +589,7 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 #if defined(CONFIG_DEBUG_FS)
 static int rv370_debugfs_pcie_gart_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32_PCIE(RADEON_PCIE_TX_GART_CNTL);
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 7e6320e8c6a0c..eae8a6389f5ea 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -474,7 +474,7 @@ int r420_init(struct radeon_device *rdev)
 #if defined(CONFIG_DEBUG_FS)
 static int r420_debugfs_pipes_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32(R400_GB_PIPE_SELECT);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index dd78fc4994024..382795a8b3c06 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -4345,7 +4345,7 @@ int r600_irq_process(struct radeon_device *rdev)
 
 static int r600_debugfs_mc_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 
 	DREG32_SYS(m, rdev, R_000E50_SRBM_STATUS);
 	DREG32_SYS(m, rdev, VM_L2_STATUS);
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 73e3117420bf0..2749dde5838f1 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -955,7 +955,7 @@ void radeon_fence_driver_force_completion(struct radeon_device *rdev, int ring)
 #if defined(CONFIG_DEBUG_FS)
 static int radeon_debugfs_fence_info_show(struct seq_file *m, void *data)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	int i, j;
 
 	for (i = 0; i < RADEON_NUM_RINGS; ++i) {
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index bdc5af23f005a..5de99ffa072f0 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -879,7 +879,7 @@ int radeon_mode_dumb_create(struct drm_file *file_priv,
 #if defined(CONFIG_DEBUG_FS)
 static int radeon_debugfs_gem_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	struct radeon_bo *rbo;
 	unsigned i = 0;
 
diff --git a/drivers/gpu/drm/radeon/radeon_ib.c b/drivers/gpu/drm/radeon/radeon_ib.c
index 6a45a72488f97..fb9ecf5dbe2b7 100644
--- a/drivers/gpu/drm/radeon/radeon_ib.c
+++ b/drivers/gpu/drm/radeon/radeon_ib.c
@@ -292,7 +292,7 @@ int radeon_ib_ring_tests(struct radeon_device *rdev)
 
 static int radeon_debugfs_sa_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 
 	radeon_sa_bo_dump_debug_info(&rdev->ring_tmp_bo, m);
 
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index cbc554928bccf..b73fd9ab02522 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1916,7 +1916,7 @@ static void radeon_dynpm_idle_work_handler(struct work_struct *work)
 
 static int radeon_debugfs_pm_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	struct drm_device *ddev = rdev->ddev;
 
 	if  ((rdev->flags & RADEON_IS_PX) &&
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 7e207276df374..e6534fa9f1fb5 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -464,7 +464,7 @@ void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *ring)
 
 static int radeon_debugfs_ring_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_ring *ring = (struct radeon_ring *) m->private;
+	struct radeon_ring *ring = m->private;
 	struct radeon_device *rdev = ring->rdev;
 
 	uint32_t rptr, wptr, rptr_next;
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 2220cdf6a3f68..06a53ecc04a25 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -780,7 +780,7 @@ void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size)
 
 static int radeon_ttm_page_pool_show(struct seq_file *m, void *data)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 
 	return ttm_pool_debugfs(&rdev->mman.bdev.pool, m);
 }
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 6383f7a34bd8c..922a29e588802 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -307,7 +307,7 @@ void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
 #if defined(CONFIG_DEBUG_FS)
 static int rs400_debugfs_gart_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32(RADEON_HOST_PATH_CNTL);
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 63fb06e8e2d7c..76260fdfbaa72 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -221,7 +221,7 @@ void rv515_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
 #if defined(CONFIG_DEBUG_FS)
 static int rv515_debugfs_pipes_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32(GB_PIPE_SELECT);
@@ -237,7 +237,7 @@ static int rv515_debugfs_pipes_info_show(struct seq_file *m, void *unused)
 
 static int rv515_debugfs_ga_info_show(struct seq_file *m, void *unused)
 {
-	struct radeon_device *rdev = (struct radeon_device *)m->private;
+	struct radeon_device *rdev = m->private;
 	uint32_t tmp;
 
 	tmp = RREG32(0x2140);
-- 
GitLab


From b3122c92697a870d75a516537f094f589dfe7ca1 Mon Sep 17 00:00:00 2001
From: Jesse Zhang <jesse.zhang@amd.com>
Date: Thu, 18 May 2023 09:46:22 +0800
Subject: [PATCH 1370/1791] drm/amdgpu: don't enable secure display on
 incompatible platforms

[why]
[drm] psp gfx command LOAD_TA(0x1) failed and response status is (0x7)
[drm] psp gfx command INVOKE_CMD(0x3) failed and response status is (0x4)
amdgpu 0000:04:00.0: amdgpu: Secure display: Generic Failure.

[how]
don't enable secure display on incompatible platforms

Suggested-by: Aaron Liu <aaron.liu@amd.com>
Signed-off-by: Jesse zhang <jesse.zhang@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/psp_v10_0.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
index e1b7fca096660..5f10883da6a23 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c
@@ -57,7 +57,13 @@ static int psp_v10_0_init_microcode(struct psp_context *psp)
 	if (err)
 		return err;
 
-	return psp_init_ta_microcode(psp, ucode_prefix);
+	err = psp_init_ta_microcode(psp, ucode_prefix);
+	if ((adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 1, 0)) &&
+		(adev->pdev->revision == 0xa1) &&
+		(psp->securedisplay_context.context.bin_desc.fw_version >= 0x27000008)) {
+		adev->psp.securedisplay_context.context.bin_desc.size_bytes = 0;
+	}
+	return err;
 }
 
 static int psp_v10_0_ring_create(struct psp_context *psp,
-- 
GitLab


From c796d7e039b57f9407523b8c4a3ba5358dd2d8ff Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Wed, 17 May 2023 13:49:08 +0800
Subject: [PATCH 1371/1791] drm/amdgpu: add the smu_v13_0_6 and gfx_v9_4_3 ip
 block

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index caae6bf2fd306..a0685a5ebda66 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -30,6 +30,7 @@
 
 #include "soc15.h"
 #include "gfx_v9_0.h"
+#include "gfx_v9_4_3.h"
 #include "gmc_v9_0.h"
 #include "df_v1_7.h"
 #include "df_v3_6.h"
@@ -1795,6 +1796,7 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(13, 0, 3):
 	case IP_VERSION(13, 0, 4):
 	case IP_VERSION(13, 0, 5):
+	case IP_VERSION(13, 0, 6):
 	case IP_VERSION(13, 0, 7):
 	case IP_VERSION(13, 0, 8):
 	case IP_VERSION(13, 0, 10):
@@ -1893,6 +1895,9 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(9, 4, 2):
 		amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
 		break;
+	case IP_VERSION(9, 4, 3):
+		amdgpu_device_ip_block_add(adev, &gfx_v9_4_3_ip_block);
+		break;
 	case IP_VERSION(10, 1, 10):
 	case IP_VERSION(10, 1, 2):
 	case IP_VERSION(10, 1, 1):
-- 
GitLab


From e825fb641b4c78a8dab5101559dd27e64d2f24aa Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Wed, 17 May 2023 13:21:15 +0800
Subject: [PATCH 1372/1791] drm/amdgpu: fix the memory override in kiq ring
 struct

This is introduced by the code merge and will let the
adev->gfx.kiq[0].ring struct being overrided

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index f2d0b1d55d771..2b4bf6c11ae4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -448,8 +448,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 
 			ring->mqd_size = mqd_size;
 			/* prepare MQD backup */
-			adev->gfx.mec.mqd_backup[j + xcc_id * adev->gfx.num_compute_rings] = kmalloc(mqd_size, GFP_KERNEL);
-			if (!adev->gfx.mec.mqd_backup[j + xcc_id * adev->gfx.num_compute_rings])
+			adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL);
+			if (!adev->gfx.mec.mqd_backup[j])
 				dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name);
 		}
 	}
-- 
GitLab


From c77b3608b8ec79a33ac36a9d281e0395a3343d6e Mon Sep 17 00:00:00 2001
From: YuanShang <YuanShang.Mao@amd.com>
Date: Mon, 8 May 2023 12:12:28 +0800
Subject: [PATCH 1373/1791] drm/amdgpu: Remove IMU ucode in vf2pf

The IMU firmware is loaded on the host side, not the guest.
Remove IMU in vf2pf ucode id enum.

Signed-off-by: YuanShang <YuanShang.Mao@amd.com>
Reviewed-By: Horace Chen <horace.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c    | 1 -
 drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index d0ad7cb0fa05d..2c1fbed245353 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -560,7 +560,6 @@ static void amdgpu_virt_populate_vf2pf_ucode_info(struct amdgpu_device *adev)
 	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC_SRLS, adev->gfx.rlc_srls_fw_version);
 	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC,      adev->gfx.mec_fw_version);
 	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC2,     adev->gfx.mec2_fw_version);
-	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_IMU,      adev->gfx.imu_fw_version);
 	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SOS,      adev->psp.sos.fw_version);
 	POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_ASD,
 			    adev->psp.asd_context.bin_desc.fw_version);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
index 24d42d24e6a01..104a5ad8397da 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
@@ -70,7 +70,6 @@ enum amd_sriov_ucode_engine_id {
 	AMD_SRIOV_UCODE_ID_RLC_SRLS,
 	AMD_SRIOV_UCODE_ID_MEC,
 	AMD_SRIOV_UCODE_ID_MEC2,
-	AMD_SRIOV_UCODE_ID_IMU,
 	AMD_SRIOV_UCODE_ID_SOS,
 	AMD_SRIOV_UCODE_ID_ASD,
 	AMD_SRIOV_UCODE_ID_TA_RAS,
-- 
GitLab


From 2ecf927b1730a4addba7ef775f433046fbcc423b Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 21:45:51 -0400
Subject: [PATCH 1374/1791] drm/amdgpu: separate ras irq from vcn instance irq
 for UVD_POISON

Separate vcn RAS poison consumption handling from the instance irq, and
register dedicated ras_poison_irq src and funcs for UVD_POISON.

v2:
- Separate ras irq from vcn instance irq
- Improve the subject and code comments

v3:
- Split the patch into three parts
- Improve the code comments

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 27 ++++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h |  3 +++
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index c088111c23218..acbef1a24b9c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -1188,6 +1188,31 @@ int amdgpu_vcn_process_poison_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+int amdgpu_vcn_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block)
+{
+	int r, i;
+
+	r = amdgpu_ras_block_late_init(adev, ras_block);
+	if (r)
+		return r;
+
+	if (amdgpu_ras_is_supported(adev, ras_block->block)) {
+		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+			if (adev->vcn.harvest_config & (1 << i))
+				continue;
+
+			r = amdgpu_irq_get(adev, &adev->vcn.inst[i].ras_poison_irq, 0);
+			if (r)
+				goto late_fini;
+		}
+	}
+	return 0;
+
+late_fini:
+	amdgpu_ras_block_late_fini(adev, ras_block);
+	return r;
+}
+
 int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev)
 {
 	int err;
@@ -1209,7 +1234,7 @@ int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev)
 	adev->vcn.ras_if = &ras->ras_block.ras_comm;
 
 	if (!ras->ras_block.ras_late_init)
-		ras->ras_block.ras_late_init = amdgpu_ras_block_late_init;
+		ras->ras_block.ras_late_init = amdgpu_vcn_ras_late_init;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 1eb9ccd1d83dd..92d5534df5f42 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -239,6 +239,7 @@ struct amdgpu_vcn_inst {
 	struct amdgpu_ring	ring_enc[AMDGPU_VCN_MAX_ENC_RINGS];
 	atomic_t		sched_score;
 	struct amdgpu_irq_src	irq;
+	struct amdgpu_irq_src	ras_poison_irq;
 	struct amdgpu_vcn_reg	external;
 	struct amdgpu_bo	*dpg_sram_bo;
 	struct dpg_pause_state	pause_state;
@@ -409,6 +410,8 @@ void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev,
 int amdgpu_vcn_process_poison_irq(struct amdgpu_device *adev,
 			struct amdgpu_irq_src *source,
 			struct amdgpu_iv_entry *entry);
+int amdgpu_vcn_ras_late_init(struct amdgpu_device *adev,
+			struct ras_common_if *ras_block);
 int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev);
 
 #endif
-- 
GitLab


From 46d75d23005f87057881c460a94f9357d079087f Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 21:52:00 -0400
Subject: [PATCH 1375/1791] drm/amdgpu: add RAS POISON interrupt funcs for
 vcn_v2_6

Add ras_poison_irq and functions.

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index 7044bd7c9f620..bb1875f926f19 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -143,7 +143,7 @@ static int vcn_v2_5_sw_init(void *handle)
 
 		/* VCN POISON TRAP */
 		r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[j],
-			VCN_2_6__SRCID_UVD_POISON, &adev->vcn.inst[j].irq);
+			VCN_2_6__SRCID_UVD_POISON, &adev->vcn.inst[j].ras_poison_irq);
 		if (r)
 			return r;
 	}
@@ -354,6 +354,9 @@ static int vcn_v2_5_hw_fini(void *handle)
 		    (adev->vcn.cur_state != AMD_PG_STATE_GATE &&
 		     RREG32_SOC15(VCN, i, mmUVD_STATUS)))
 			vcn_v2_5_set_powergating_state(adev, AMD_PG_STATE_GATE);
+
+		if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN))
+			amdgpu_irq_put(adev, &adev->vcn.inst[i].ras_poison_irq, 0);
 	}
 
 	return 0;
@@ -1807,6 +1810,14 @@ static int vcn_v2_5_set_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int vcn_v2_6_set_ras_interrupt_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned int type,
+					enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
 static int vcn_v2_5_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
@@ -1837,9 +1848,6 @@ static int vcn_v2_5_process_interrupt(struct amdgpu_device *adev,
 	case VCN_2_0__SRCID__UVD_ENC_LOW_LATENCY:
 		amdgpu_fence_process(&adev->vcn.inst[ip_instance].ring_enc[1]);
 		break;
-	case VCN_2_6__SRCID_UVD_POISON:
-		amdgpu_vcn_process_poison_irq(adev, source, entry);
-		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
 			  entry->src_id, entry->src_data[0]);
@@ -1854,6 +1862,11 @@ static const struct amdgpu_irq_src_funcs vcn_v2_5_irq_funcs = {
 	.process = vcn_v2_5_process_interrupt,
 };
 
+static const struct amdgpu_irq_src_funcs vcn_v2_6_ras_irq_funcs = {
+	.set = vcn_v2_6_set_ras_interrupt_state,
+	.process = amdgpu_vcn_process_poison_irq,
+};
+
 static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev)
 {
 	int i;
@@ -1863,6 +1876,9 @@ static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev)
 			continue;
 		adev->vcn.inst[i].irq.num_types = adev->vcn.num_enc_rings + 1;
 		adev->vcn.inst[i].irq.funcs = &vcn_v2_5_irq_funcs;
+
+		adev->vcn.inst[i].ras_poison_irq.num_types = adev->vcn.num_enc_rings + 1;
+		adev->vcn.inst[i].ras_poison_irq.funcs = &vcn_v2_6_ras_irq_funcs;
 	}
 }
 
@@ -1965,6 +1981,7 @@ const struct amdgpu_ras_block_hw_ops vcn_v2_6_ras_hw_ops = {
 static struct amdgpu_vcn_ras vcn_v2_6_ras = {
 	.ras_block = {
 		.hw_ops = &vcn_v2_6_ras_hw_ops,
+		.ras_late_init = amdgpu_vcn_ras_late_init,
 	},
 };
 
-- 
GitLab


From 66a11ecbdebbc8ab29a4076df0b95f0bfd61b1c5 Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 22:09:43 -0400
Subject: [PATCH 1376/1791] drm/amdgpu: add RAS POISON interrupt funcs for
 vcn_v4_0

Add ras_poison_irq and functions. And fix the amdgpu_irq_put
call trace in vcn_v4_0_hw_fini.

[   44.563572] RIP: 0010:amdgpu_irq_put+0xa4/0xc0 [amdgpu]
[   44.563629] RSP: 0018:ffffb36740edfc90 EFLAGS: 00010246
[   44.563630] RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
[   44.563630] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[   44.563631] RBP: ffffb36740edfcb0 R08: 0000000000000000 R09: 0000000000000000
[   44.563631] R10: 0000000000000000 R11: 0000000000000000 R12: ffff954c568e2ea8
[   44.563631] R13: 0000000000000000 R14: ffff954c568c0000 R15: ffff954c568e2ea8
[   44.563632] FS:  0000000000000000(0000) GS:ffff954f584c0000(0000) knlGS:0000000000000000
[   44.563632] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   44.563633] CR2: 00007f028741ba70 CR3: 000000026ca10000 CR4: 0000000000750ee0
[   44.563633] PKRU: 55555554
[   44.563633] Call Trace:
[   44.563634]  <TASK>
[   44.563634]  vcn_v4_0_hw_fini+0x62/0x160 [amdgpu]
[   44.563700]  vcn_v4_0_suspend+0x13/0x30 [amdgpu]
[   44.563755]  amdgpu_device_ip_suspend_phase2+0x240/0x470 [amdgpu]
[   44.563806]  amdgpu_device_ip_suspend+0x41/0x80 [amdgpu]
[   44.563858]  amdgpu_device_pre_asic_reset+0xd9/0x4a0 [amdgpu]
[   44.563909]  amdgpu_device_gpu_recover.cold+0x548/0xcf1 [amdgpu]
[   44.564006]  amdgpu_debugfs_reset_work+0x4c/0x80 [amdgpu]
[   44.564061]  process_one_work+0x21f/0x400
[   44.564062]  worker_thread+0x200/0x3f0
[   44.564063]  ? process_one_work+0x400/0x400
[   44.564064]  kthread+0xee/0x120
[   44.564065]  ? kthread_complete_and_exit+0x20/0x20
[   44.564066]  ret_from_fork+0x22/0x30

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 36 ++++++++++++++++++++++-----
 1 file changed, 30 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 81446e6996df6..60c3fd20e8ced 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -139,7 +139,7 @@ static int vcn_v4_0_sw_init(void *handle)
 
 		/* VCN POISON TRAP */
 		r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[i],
-				VCN_4_0__SRCID_UVD_POISON, &adev->vcn.inst[i].irq);
+				VCN_4_0__SRCID_UVD_POISON, &adev->vcn.inst[i].ras_poison_irq);
 		if (r)
 			return r;
 
@@ -305,8 +305,8 @@ static int vcn_v4_0_hw_fini(void *handle)
                         vcn_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
 			}
 		}
-
-		amdgpu_irq_put(adev, &adev->vcn.inst[i].irq, 0);
+		if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN))
+			amdgpu_irq_put(adev, &adev->vcn.inst[i].ras_poison_irq, 0);
 	}
 
 	return 0;
@@ -1975,6 +1975,24 @@ static int vcn_v4_0_set_interrupt_state(struct amdgpu_device *adev, struct amdgp
 	return 0;
 }
 
+/**
+ * vcn_v4_0_set_ras_interrupt_state - set VCN block RAS interrupt state
+ *
+ * @adev: amdgpu_device pointer
+ * @source: interrupt sources
+ * @type: interrupt types
+ * @state: interrupt states
+ *
+ * Set VCN block RAS interrupt state
+ */
+static int vcn_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev,
+	struct amdgpu_irq_src *source,
+	unsigned int type,
+	enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
 /**
  * vcn_v4_0_process_interrupt - process VCN block interrupt
  *
@@ -2007,9 +2025,6 @@ static int vcn_v4_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_
 	case VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE:
 		amdgpu_fence_process(&adev->vcn.inst[ip_instance].ring_enc[0]);
 		break;
-	case VCN_4_0__SRCID_UVD_POISON:
-		amdgpu_vcn_process_poison_irq(adev, source, entry);
-		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
 			  entry->src_id, entry->src_data[0]);
@@ -2024,6 +2039,11 @@ static const struct amdgpu_irq_src_funcs vcn_v4_0_irq_funcs = {
 	.process = vcn_v4_0_process_interrupt,
 };
 
+static const struct amdgpu_irq_src_funcs vcn_v4_0_ras_irq_funcs = {
+	.set = vcn_v4_0_set_ras_interrupt_state,
+	.process = amdgpu_vcn_process_poison_irq,
+};
+
 /**
  * vcn_v4_0_set_irq_funcs - set VCN block interrupt irq functions
  *
@@ -2041,6 +2061,9 @@ static void vcn_v4_0_set_irq_funcs(struct amdgpu_device *adev)
 
 		adev->vcn.inst[i].irq.num_types = adev->vcn.num_enc_rings + 1;
 		adev->vcn.inst[i].irq.funcs = &vcn_v4_0_irq_funcs;
+
+		adev->vcn.inst[i].ras_poison_irq.num_types = adev->vcn.num_enc_rings + 1;
+		adev->vcn.inst[i].ras_poison_irq.funcs = &vcn_v4_0_ras_irq_funcs;
 	}
 }
 
@@ -2114,6 +2137,7 @@ const struct amdgpu_ras_block_hw_ops vcn_v4_0_ras_hw_ops = {
 static struct amdgpu_vcn_ras vcn_v4_0_ras = {
 	.ras_block = {
 		.hw_ops = &vcn_v4_0_ras_hw_ops,
+		.ras_late_init = amdgpu_vcn_ras_late_init,
 	},
 };
 
-- 
GitLab


From 18dad20c3dcbd7789f3d07056cd78394c8278a75 Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 22:57:19 -0400
Subject: [PATCH 1377/1791] drm/amdgpu: separate ras irq from jpeg instance irq
 for UVD_POISON

Separate jpegbRAS poison consumption handling from the instance irq, and
register dedicated ras_poison_irq src and funcs for UVD_POISON.

v2:
- Separate ras irq from jpeg instance irq
- Improve the subject and code comments

v3:
- Split the patch into three parts
- Improve the code comments

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 27 +++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h |  3 +++
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 8c479669c459a..3add4b4f0667f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -245,6 +245,31 @@ int amdgpu_jpeg_process_poison_irq(struct amdgpu_device *adev,
 	return 0;
 }
 
+int amdgpu_jpeg_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block)
+{
+	int r, i;
+
+	r = amdgpu_ras_block_late_init(adev, ras_block);
+	if (r)
+		return r;
+
+	if (amdgpu_ras_is_supported(adev, ras_block->block)) {
+		for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
+			if (adev->jpeg.harvest_config & (1 << i))
+				continue;
+
+			r = amdgpu_irq_get(adev, &adev->jpeg.inst[i].ras_poison_irq, 0);
+			if (r)
+				goto late_fini;
+		}
+	}
+	return 0;
+
+late_fini:
+	amdgpu_ras_block_late_fini(adev, ras_block);
+	return r;
+}
+
 int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev)
 {
 	int err;
@@ -266,7 +291,7 @@ int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev)
 	adev->jpeg.ras_if = &ras->ras_block.ras_comm;
 
 	if (!ras->ras_block.ras_late_init)
-		ras->ras_block.ras_late_init = amdgpu_ras_block_late_init;
+		ras->ras_block.ras_late_init = amdgpu_jpeg_ras_late_init;
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index 90516f623f561..ffe47e9f5bf2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -39,6 +39,7 @@ struct amdgpu_jpeg_reg{
 struct amdgpu_jpeg_inst {
 	struct amdgpu_ring ring_dec[AMDGPU_MAX_JPEG_RINGS];
 	struct amdgpu_irq_src irq;
+	struct amdgpu_irq_src ras_poison_irq;
 	struct amdgpu_jpeg_reg external;
 	uint8_t aid_id;
 };
@@ -78,6 +79,8 @@ int amdgpu_jpeg_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout);
 int amdgpu_jpeg_process_poison_irq(struct amdgpu_device *adev,
 				struct amdgpu_irq_src *source,
 				struct amdgpu_iv_entry *entry);
+int amdgpu_jpeg_ras_late_init(struct amdgpu_device *adev,
+				struct ras_common_if *ras_block);
 int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev);
 
 #endif /*__AMDGPU_JPEG_H__*/
-- 
GitLab


From 674f90f83bc941d0cd5a85a714c5bfeb8789163a Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 22:59:54 -0400
Subject: [PATCH 1378/1791] drm/amdgpu: add RAS POISON interrupt funcs for
 jpeg_v2_6

Add ras_poison_irq and functions.

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 28 ++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index b79edb12b90e8..f533ede484d4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -102,13 +102,13 @@ static int jpeg_v2_5_sw_init(void *handle)
 
 		/* JPEG DJPEG POISON EVENT */
 		r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_jpeg[i],
-			VCN_2_6__SRCID_DJPEG0_POISON, &adev->jpeg.inst[i].irq);
+			VCN_2_6__SRCID_DJPEG0_POISON, &adev->jpeg.inst[i].ras_poison_irq);
 		if (r)
 			return r;
 
 		/* JPEG EJPEG POISON EVENT */
 		r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_jpeg[i],
-			VCN_2_6__SRCID_EJPEG0_POISON, &adev->jpeg.inst[i].irq);
+			VCN_2_6__SRCID_EJPEG0_POISON, &adev->jpeg.inst[i].ras_poison_irq);
 		if (r)
 			return r;
 	}
@@ -221,6 +221,9 @@ static int jpeg_v2_5_hw_fini(void *handle)
 		if (adev->jpeg.cur_state != AMD_PG_STATE_GATE &&
 		      RREG32_SOC15(JPEG, i, mmUVD_JRBC_STATUS))
 			jpeg_v2_5_set_powergating_state(adev, AMD_PG_STATE_GATE);
+
+		if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG))
+			amdgpu_irq_put(adev, &adev->jpeg.inst[i].ras_poison_irq, 0);
 	}
 
 	return 0;
@@ -569,6 +572,14 @@ static int jpeg_v2_5_set_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int jpeg_v2_6_set_ras_interrupt_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned int type,
+					enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
 static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
@@ -593,10 +604,6 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev,
 	case VCN_2_0__SRCID__JPEG_DECODE:
 		amdgpu_fence_process(adev->jpeg.inst[ip_instance].ring_dec);
 		break;
-	case VCN_2_6__SRCID_DJPEG0_POISON:
-	case VCN_2_6__SRCID_EJPEG0_POISON:
-		amdgpu_jpeg_process_poison_irq(adev, source, entry);
-		break;
 	default:
 		DRM_ERROR("Unhandled interrupt: %d %d\n",
 			  entry->src_id, entry->src_data[0]);
@@ -725,6 +732,11 @@ static const struct amdgpu_irq_src_funcs jpeg_v2_5_irq_funcs = {
 	.process = jpeg_v2_5_process_interrupt,
 };
 
+static const struct amdgpu_irq_src_funcs jpeg_v2_6_ras_irq_funcs = {
+	.set = jpeg_v2_6_set_ras_interrupt_state,
+	.process = amdgpu_jpeg_process_poison_irq,
+};
+
 static void jpeg_v2_5_set_irq_funcs(struct amdgpu_device *adev)
 {
 	int i;
@@ -735,6 +747,9 @@ static void jpeg_v2_5_set_irq_funcs(struct amdgpu_device *adev)
 
 		adev->jpeg.inst[i].irq.num_types = 1;
 		adev->jpeg.inst[i].irq.funcs = &jpeg_v2_5_irq_funcs;
+
+		adev->jpeg.inst[i].ras_poison_irq.num_types = 1;
+		adev->jpeg.inst[i].ras_poison_irq.funcs = &jpeg_v2_6_ras_irq_funcs;
 	}
 }
 
@@ -800,6 +815,7 @@ const struct amdgpu_ras_block_hw_ops jpeg_v2_6_ras_hw_ops = {
 static struct amdgpu_jpeg_ras jpeg_v2_6_ras = {
 	.ras_block = {
 		.hw_ops = &jpeg_v2_6_ras_hw_ops,
+		.ras_late_init = amdgpu_jpeg_ras_late_init,
 	},
 };
 
-- 
GitLab


From a34b09060a3b95c0341b444ea49558a807988e34 Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 15 May 2023 23:02:10 -0400
Subject: [PATCH 1379/1791] drm/amdgpu: add RAS POISON interrupt funcs for
 jpeg_v4_0

Add ras_poison_irq and functions. And fix the amdgpu_irq_put
call trace in jpeg_v4_0_hw_fini.

[   50.497562] RIP: 0010:amdgpu_irq_put+0xa4/0xc0 [amdgpu]
[   50.497619] RSP: 0018:ffffaa2400fcfcb0 EFLAGS: 00010246
[   50.497620] RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
[   50.497621] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[   50.497621] RBP: ffffaa2400fcfcd0 R08: 0000000000000000 R09: 0000000000000000
[   50.497622] R10: 0000000000000000 R11: 0000000000000000 R12: ffff99b2105242d8
[   50.497622] R13: 0000000000000000 R14: ffff99b210500000 R15: ffff99b210500000
[   50.497623] FS:  0000000000000000(0000) GS:ffff99b518480000(0000) knlGS:0000000000000000
[   50.497623] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   50.497624] CR2: 00007f9d32aa91e8 CR3: 00000001ba210000 CR4: 0000000000750ee0
[   50.497624] PKRU: 55555554
[   50.497625] Call Trace:
[   50.497625]  <TASK>
[   50.497627]  jpeg_v4_0_hw_fini+0x43/0xc0 [amdgpu]
[   50.497693]  jpeg_v4_0_suspend+0x13/0x30 [amdgpu]
[   50.497751]  amdgpu_device_ip_suspend_phase2+0x240/0x470 [amdgpu]
[   50.497802]  amdgpu_device_ip_suspend+0x41/0x80 [amdgpu]
[   50.497854]  amdgpu_device_pre_asic_reset+0xd9/0x4a0 [amdgpu]
[   50.497905]  amdgpu_device_gpu_recover.cold+0x548/0xcf1 [amdgpu]
[   50.498005]  amdgpu_debugfs_reset_work+0x4c/0x80 [amdgpu]
[   50.498060]  process_one_work+0x21f/0x400
[   50.498063]  worker_thread+0x200/0x3f0
[   50.498064]  ? process_one_work+0x400/0x400
[   50.498065]  kthread+0xee/0x120
[   50.498067]  ? kthread_complete_and_exit+0x20/0x20
[   50.498068]  ret_from_fork+0x22/0x30

Suggested-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c | 28 +++++++++++++++++++-------
 1 file changed, 21 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index 495facb885f4d..86383fbf9358a 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -87,13 +87,13 @@ static int jpeg_v4_0_sw_init(void *handle)
 
 	/* JPEG DJPEG POISON EVENT */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-			VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->irq);
+			VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq);
 	if (r)
 		return r;
 
 	/* JPEG EJPEG POISON EVENT */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
-			VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->irq);
+			VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq);
 	if (r)
 		return r;
 
@@ -202,7 +202,8 @@ static int jpeg_v4_0_hw_fini(void *handle)
 			RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS))
 			jpeg_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE);
 	}
-	amdgpu_irq_put(adev, &adev->jpeg.inst->irq, 0);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG))
+		amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0);
 
 	return 0;
 }
@@ -670,6 +671,14 @@ static int jpeg_v4_0_set_interrupt_state(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int jpeg_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev,
+					struct amdgpu_irq_src *source,
+					unsigned int type,
+					enum amdgpu_interrupt_state state)
+{
+	return 0;
+}
+
 static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev,
 				      struct amdgpu_irq_src *source,
 				      struct amdgpu_iv_entry *entry)
@@ -680,10 +689,6 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev,
 	case VCN_4_0__SRCID__JPEG_DECODE:
 		amdgpu_fence_process(adev->jpeg.inst->ring_dec);
 		break;
-	case VCN_4_0__SRCID_DJPEG0_POISON:
-	case VCN_4_0__SRCID_EJPEG0_POISON:
-		amdgpu_jpeg_process_poison_irq(adev, source, entry);
-		break;
 	default:
 		DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n",
 			  entry->src_id, entry->src_data[0]);
@@ -753,10 +758,18 @@ static const struct amdgpu_irq_src_funcs jpeg_v4_0_irq_funcs = {
 	.process = jpeg_v4_0_process_interrupt,
 };
 
+static const struct amdgpu_irq_src_funcs jpeg_v4_0_ras_irq_funcs = {
+	.set = jpeg_v4_0_set_ras_interrupt_state,
+	.process = amdgpu_jpeg_process_poison_irq,
+};
+
 static void jpeg_v4_0_set_irq_funcs(struct amdgpu_device *adev)
 {
 	adev->jpeg.inst->irq.num_types = 1;
 	adev->jpeg.inst->irq.funcs = &jpeg_v4_0_irq_funcs;
+
+	adev->jpeg.inst->ras_poison_irq.num_types = 1;
+	adev->jpeg.inst->ras_poison_irq.funcs = &jpeg_v4_0_ras_irq_funcs;
 }
 
 const struct amdgpu_ip_block_version jpeg_v4_0_ip_block = {
@@ -811,6 +824,7 @@ const struct amdgpu_ras_block_hw_ops jpeg_v4_0_ras_hw_ops = {
 static struct amdgpu_jpeg_ras jpeg_v4_0_ras = {
 	.ras_block = {
 		.hw_ops = &jpeg_v4_0_ras_hw_ops,
+		.ras_late_init = amdgpu_jpeg_ras_late_init,
 	},
 };
 
-- 
GitLab


From d7b8e68dc04ad89809832bebe9ab5d7965a6eef5 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 11:36:57 +0530
Subject: [PATCH 1380/1791] drm/amdgpu: Fix uninitialized variable in
 gfx_v9_4_3_cp_resume
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1925:6: error: variable 'r' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
        if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1931:6: note: uninitialized use occurs here
        if (r)
            ^
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1925:2: note: remove the 'if' if its condition is always true
        if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:1923:7: note: initialize the variable 'r' to silence this warning
        int r, i, num_xcc;
             ^
              = 0
1 error generated.

Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index e5cfb3adb3b3f..6a1a28df606c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1920,7 +1920,7 @@ static int gfx_v9_4_3_xcc_cp_resume(struct amdgpu_device *adev, int xcc_id)
 
 static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 {
-	int r, i, num_xcc;
+	int r = 0, i, num_xcc;
 
 	if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
 					    AMDGPU_XCP_FL_NONE) ==
-- 
GitLab


From b336c681bdb5d1814acf8f19d1225a93f36ddfa2 Mon Sep 17 00:00:00 2001
From: Likun Gao <Likun.Gao@amd.com>
Date: Fri, 19 May 2023 16:01:57 +0800
Subject: [PATCH 1381/1791] drm/amdgpu: fix vga_set_state NULL pointer issue

Fix NULL pointer issue for vga_set_state function
as not all the ASIC need this operation.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index a4d0bc80ac92a..8b4a89a6c3fdd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1241,7 +1241,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 /*
  * ASICs macro.
  */
-#define amdgpu_asic_set_vga_state(adev, state) (adev)->asic_funcs->set_vga_state((adev), (state))
+#define amdgpu_asic_set_vga_state(adev, state) \
+    ((adev)->asic_funcs->set_vga_state ? (adev)->asic_funcs->set_vga_state((adev), (state)) : 0)
 #define amdgpu_asic_reset(adev) (adev)->asic_funcs->reset((adev))
 #define amdgpu_asic_reset_method(adev) (adev)->asic_funcs->reset_method((adev))
 #define amdgpu_asic_get_xclk(adev) (adev)->asic_funcs->get_xclk((adev))
-- 
GitLab


From 9788d087caffd8358d6e14349ee69d9385666719 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Fri, 5 May 2023 13:16:32 -0400
Subject: [PATCH 1382/1791] drm/amd/display: improve the message printed when
 loading DC

[Why&How]
Change how DC version and hardware version is printed when driver is
loaded.

- Remove exclamation
- Add DC version and hardware version to both success and failure cases
- Add version in between appropriate filler words to make a complete
  statement.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index abef3c2f59c8d..4edbcc9e535bd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1667,10 +1667,11 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	adev->dm.dc = dc_create(&init_data);
 
 	if (adev->dm.dc) {
-		DRM_INFO("Display Core initialized with v%s! %s\n", DC_VER,
+		DRM_INFO("Display Core v%s initialized on %s\n", DC_VER,
 			 dce_version_to_string(adev->dm.dc->ctx->dce_version));
 	} else {
-		DRM_INFO("Display Core failed to initialize with v%s!\n", DC_VER);
+		DRM_INFO("Display Core v%s failed to initialize on %s\n", DC_VER,
+			 dce_version_to_string(adev->dm.dc->ctx->dce_version));
 		goto error;
 	}
 
-- 
GitLab


From 20a29ac09192f9d9705015261652f277e8162f09 Mon Sep 17 00:00:00 2001
From: Likun Gao <Likun.Gao@amd.com>
Date: Fri, 19 May 2023 18:54:18 +0800
Subject: [PATCH 1383/1791] drm/amdgpu: retire set_vga_state for some ASIC

set_vga_state operation only allowed on SI generation
ASIC, retire the realted function on those ASIC which
did not do anything.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nv.c    | 6 ------
 drivers/gpu/drm/amd/amdgpu/soc15.c | 8 --------
 drivers/gpu/drm/amd/amdgpu/soc21.c | 6 ------
 drivers/gpu/drm/amd/amdgpu/vi.c    | 6 ------
 4 files changed, 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index a5f76c9538c4c..51523b27a186c 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -341,11 +341,6 @@ void nv_grbm_select(struct amdgpu_device *adev,
 	WREG32_SOC15(GC, 0, mmGRBM_GFX_CNTL, grbm_gfx_cntl);
 }
 
-static void nv_vga_set_state(struct amdgpu_device *adev, bool state)
-{
-	/* todo */
-}
-
 static bool nv_read_disabled_bios(struct amdgpu_device *adev)
 {
 	/* todo */
@@ -654,7 +649,6 @@ static const struct amdgpu_asic_funcs nv_asic_funcs =
 	.read_register = &nv_read_register,
 	.reset = &nv_asic_reset,
 	.reset_method = &nv_asic_reset_method,
-	.set_vga_state = &nv_vga_set_state,
 	.get_xclk = &nv_get_xclk,
 	.set_uvd_clocks = &nv_set_uvd_clocks,
 	.set_vce_clocks = &nv_set_vce_clocks,
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 122ba1a505c34..135440b5afe9d 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -346,11 +346,6 @@ void soc15_grbm_select(struct amdgpu_device *adev,
 	WREG32_SOC15_RLC_SHADOW(GC, xcc_id, mmGRBM_GFX_CNTL, grbm_gfx_cntl);
 }
 
-static void soc15_vga_set_state(struct amdgpu_device *adev, bool state)
-{
-	/* todo */
-}
-
 static bool soc15_read_disabled_bios(struct amdgpu_device *adev)
 {
 	/* todo */
@@ -849,7 +844,6 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs =
 	.read_register = &soc15_read_register,
 	.reset = &soc15_asic_reset,
 	.reset_method = &soc15_asic_reset_method,
-	.set_vga_state = &soc15_vga_set_state,
 	.get_xclk = &soc15_get_xclk,
 	.set_uvd_clocks = &soc15_set_uvd_clocks,
 	.set_vce_clocks = &soc15_set_vce_clocks,
@@ -871,7 +865,6 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs =
 	.read_register = &soc15_read_register,
 	.reset = &soc15_asic_reset,
 	.reset_method = &soc15_asic_reset_method,
-	.set_vga_state = &soc15_vga_set_state,
 	.get_xclk = &soc15_get_xclk,
 	.set_uvd_clocks = &soc15_set_uvd_clocks,
 	.set_vce_clocks = &soc15_set_vce_clocks,
@@ -893,7 +886,6 @@ static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs =
 	.read_register = &soc15_read_register,
 	.reset = &soc15_asic_reset,
 	.reset_method = &soc15_asic_reset_method,
-	.set_vga_state = &soc15_vga_set_state,
 	.get_xclk = &soc15_get_xclk,
 	.set_uvd_clocks = &soc15_set_uvd_clocks,
 	.set_vce_clocks = &soc15_set_vce_clocks,
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index 6bff936a6e552..e5e5d68a4d702 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -248,11 +248,6 @@ void soc21_grbm_select(struct amdgpu_device *adev,
 	WREG32_SOC15(GC, 0, regGRBM_GFX_CNTL, grbm_gfx_cntl);
 }
 
-static void soc21_vga_set_state(struct amdgpu_device *adev, bool state)
-{
-	/* todo */
-}
-
 static bool soc21_read_disabled_bios(struct amdgpu_device *adev)
 {
 	/* todo */
@@ -559,7 +554,6 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs =
 	.read_register = &soc21_read_register,
 	.reset = &soc21_asic_reset,
 	.reset_method = &soc21_asic_reset_method,
-	.set_vga_state = &soc21_vga_set_state,
 	.get_xclk = &soc21_get_xclk,
 	.set_uvd_clocks = &soc21_set_uvd_clocks,
 	.set_vce_clocks = &soc21_set_vce_clocks,
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 8e70581960fbb..770f2d7a371fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -580,11 +580,6 @@ void vi_srbm_select(struct amdgpu_device *adev,
 	WREG32(mmSRBM_GFX_CNTL, srbm_gfx_cntl);
 }
 
-static void vi_vga_set_state(struct amdgpu_device *adev, bool state)
-{
-	/* todo */
-}
-
 static bool vi_read_disabled_bios(struct amdgpu_device *adev)
 {
 	u32 bus_cntl;
@@ -1435,7 +1430,6 @@ static const struct amdgpu_asic_funcs vi_asic_funcs =
 	.read_register = &vi_read_register,
 	.reset = &vi_asic_reset,
 	.reset_method = &vi_asic_reset_method,
-	.set_vga_state = &vi_vga_set_state,
 	.get_xclk = &vi_get_xclk,
 	.set_uvd_clocks = &vi_set_uvd_clocks,
 	.set_vce_clocks = &vi_set_vce_clocks,
-- 
GitLab


From ff6b11cc7263d4d6f17bc7b94e81ffcaae5fe107 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 16:24:55 +0530
Subject: [PATCH 1384/1791] drm/amd/amdgpu: Fix errors & warnings in
 mmhub_v1_8.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below errors & warnings reported by checkpatch:

ERROR: code indent should use tabs where possible
WARNING: please, no space before tabs
WARNING: please, no spaces at the start of a line
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
ERROR: space prohibited before that '++' (ctx:WxB)
WARNING: Block comments use a trailing */ on a separate line

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 3648994724c26..8600e42434e70 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -271,7 +271,7 @@ static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev)
 					    VMC_TAP_PTE_REQUEST_PHYSICAL, 0);
 		}
 		WREG32_SOC15(MMHUB, i, regVM_L2_CNTL4, tmp);
- 	}
+	}
 }
 
 static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev)
@@ -328,7 +328,7 @@ static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev)
 static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev)
 {
 	struct amdgpu_vmhub *hub;
-	unsigned num_level, block_size;
+	unsigned int num_level, block_size;
 	uint32_t tmp, inst_mask;
 	int i, j;
 
@@ -776,9 +776,10 @@ static void mmhub_v1_8_inst_reset_ras_err_status(struct amdgpu_device *adev,
 	/* reset mmea ras err status */
 	mmea_cgtt_clk_cntl_addr_dist = regMMEA1_CGTT_CLK_CTRL - regMMEA0_CGTT_CLK_CTRL;
 	mmea_err_status_addr_dist = regMMEA1_ERR_STATUS - regMMEA0_ERR_STATUS;
-	for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i ++) {
+	for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i++) {
 		/* force clk branch on for response path
-		 * set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 1 */
+		 * set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 1
+		 */
 		reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst,
 						regMMEA0_CGTT_CLK_CTRL,
 						i * mmea_cgtt_clk_cntl_addr_dist);
@@ -814,7 +815,8 @@ static void mmhub_v1_8_inst_reset_ras_err_status(struct amdgpu_device *adev,
 
 	/* reset mm_cane ras err status
 	 * force clk branch on for response path
-	 * set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 1 */
+	 * set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 1
+	 */
 	reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL);
 	reg_value = REG_SET_FIELD(reg_value, MM_CANE_ICG_CTRL,
 				  SOFT_OVERRIDE_ATRET, 1);
-- 
GitLab


From b3b0e016ec44d94db48a7d01b69570b5de37a31c Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 10:34:50 +0530
Subject: [PATCH 1385/1791] drm/amdgpu: Fix uninitalized variable in
 jpeg_v4_0_3_is_idle & jpeg_v4_0_3_wait_for_idle
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c:752:4: error: variable 'ret' is uninitialized when used here [-Werror,-Wuninitialized]
                        ret &= ((RREG32_SOC15_OFFSET(
                        ^~~
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c:745:10: note: initialize the variable 'ret' to silence this warning
        bool ret;
                ^
                 = 0
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c:774:4: error: variable 'ret' is uninitialized when used here [-Werror,-Wuninitialized]
                        ret &= SOC15_WAIT_ON_RREG_OFFSET(
                        ^~~
drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c:767:9: note: initialize the variable 'ret' to silence this warning
        int ret;
               ^
                = 0
2 errors generated.

Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: James Zhu <James.Zhu@amd.com>
Cc: Leo Liu <leo.liu@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index ede15a3a4701d..ce2b22f7e4e4b 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -742,7 +742,7 @@ static void jpeg_v4_0_3_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count)
 static bool jpeg_v4_0_3_is_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	bool ret;
+	bool ret = false;
 	int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
@@ -764,7 +764,7 @@ static bool jpeg_v4_0_3_is_idle(void *handle)
 static int jpeg_v4_0_3_wait_for_idle(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int ret;
+	int ret = 0;
 	int i, j;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
-- 
GitLab


From 2ad00e753ae13b0c523a579fb04372787f77cce9 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 10:44:41 +0530
Subject: [PATCH 1386/1791] drm/amdgpu: Fix uninitalized variable in
 kgd2kfd_device_init
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_device.c:613:4: error: variable 'num_xcd' is uninitialized when used here [-Werror,-Wuninitialized]
                        num_xcd, kfd->adev->gfx.num_xcc_per_xcp);
                        ^~~~~~~
include/linux/dev_printk.h:144:65: note: expanded from macro 'dev_err'
        dev_printk_index_wrap(_dev_err, KERN_ERR, dev, dev_fmt(fmt), ##__VA_ARGS__)
                                                                       ^~~~~~~~~~~
include/linux/dev_printk.h:110:23: note: expanded from macro 'dev_printk_index_wrap'
                _p_func(dev, fmt, ##__VA_ARGS__);                       \
                                    ^~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_device.c:597:13: note: initialize the variable 'num_xcd' to silence this warning
        int num_xcd, partition_mode;
                   ^
                    = 0
1 error generated.

Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 986543a000bf5..e92b93b2c14c3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -594,7 +594,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	struct kfd_node *node;
 	uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd;
 	unsigned int max_proc_per_quantum;
-	int num_xcd, partition_mode;
+	int partition_mode;
 	int xcp_idx;
 
 	kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev,
@@ -609,8 +609,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	if (kfd->num_nodes == 0) {
 		dev_err(kfd_device,
-			"KFD num nodes cannot be 0, GC inst: %d, num_xcc_in_node: %d\n",
-			num_xcd, kfd->adev->gfx.num_xcc_per_xcp);
+			"KFD num nodes cannot be 0, num_xcc_in_node: %d\n",
+			kfd->adev->gfx.num_xcc_per_xcp);
 		goto out;
 	}
 
-- 
GitLab


From 4c44a51c78447ae97a6d7ef27ffdb790c3f61cca Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 18 May 2023 12:38:22 -0400
Subject: [PATCH 1387/1791] drm/radeon: reintroduce radeon_dp_work_func content

Put back the radeon_dp_work_func logic.  It seems that
handling DP RX interrupts is necessary to make some
panels work.  This was removed with the MST support,
but it regresses some systems so add it back.  While
we are here, add the proper mutex locking.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2567
Fixes: 01ad1d9c2888 ("drm/radeon: Drop legacy MST support")
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Lyude Paul <lyude@redhat.com>
---
 drivers/gpu/drm/radeon/radeon_irq_kms.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 3377fbc71f654..c4dda908666cf 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -99,6 +99,16 @@ static void radeon_hotplug_work_func(struct work_struct *work)
 
 static void radeon_dp_work_func(struct work_struct *work)
 {
+	struct radeon_device *rdev = container_of(work, struct radeon_device,
+						  dp_work);
+	struct drm_device *dev = rdev->ddev;
+	struct drm_mode_config *mode_config = &dev->mode_config;
+	struct drm_connector *connector;
+
+	mutex_lock(&mode_config->mutex);
+	list_for_each_entry(connector, &mode_config->connector_list, head)
+		radeon_connector_hotplug(connector);
+	mutex_unlock(&mode_config->mutex);
 }
 
 /**
-- 
GitLab


From 73ade646c545feda7c5df9b9c78c5d011ce76463 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 18 May 2023 15:51:38 +0200
Subject: [PATCH 1388/1791] drm/amdgpu: stop including swiotlb.h

amdgpu does not need swiotlb.h, so stop including it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 509b26e9d0c4a..0abad5f894213 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -38,7 +38,6 @@
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
-#include <linux/swiotlb.h>
 #include <linux/dma-buf.h>
 #include <linux/sizes.h>
 #include <linux/module.h>
-- 
GitLab


From f36f2648f32c184ffc285a836b1ce3757e966925 Mon Sep 17 00:00:00 2001
From: Cruise Hung <cruise.hung@amd.com>
Date: Tue, 9 May 2023 21:36:35 +0800
Subject: [PATCH 1389/1791] drm/amd/display: Fix DMUB debugging print issue

[Why]
The DMUB diagnostic data was not printed out correctly.

[How]
Print the DMUB diagnostic data line by line.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Cruise Hung <cruise.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c     |  12 +++
 drivers/gpu/drm/amd/display/dc/dc.h          |   2 +
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 103 +++++++------------
 3 files changed, 49 insertions(+), 68 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index e664a77d05eb8..9f8efee277210 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -4832,6 +4832,18 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc,
 	DC_LOG_DEBUG("%s: hpd_int_enable(%d)\n", __func__, hpd_int_enable);
 }
 
+/**
+ * dc_print_dmub_diagnostic_data - Print DMUB diagnostic data for debugging
+ *
+ * @dc: [in] dc structure
+ *
+ *
+ */
+void dc_print_dmub_diagnostic_data(const struct dc *dc)
+{
+	dc_dmub_srv_log_diagnostic_data(dc->ctx->dmub_srv);
+}
+
 /**
  * dc_disable_accelerated_mode - disable accelerated mode
  * @dc: dc structure
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 2dff1a5cf3b10..7fc087f85d398 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -2236,6 +2236,8 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc,
 void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc,
 				uint32_t hpd_int_enable);
 
+void dc_print_dmub_diagnostic_data(const struct dc *dc);
+
 /* DSC Interfaces */
 #include "dc_dsc.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 0319a30f2d5c1..c52c40b163872 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -169,6 +169,7 @@ bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int coun
 
 		if (status != DMUB_STATUS_OK) {
 			DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
+			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
 			return false;
 		}
 
@@ -797,74 +798,40 @@ void dc_dmub_srv_log_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv)
 		return;
 	}
 
-	DC_LOG_DEBUG(
-		"DMCUB STATE\n"
-		"    dmcub_version      : %08x\n"
-		"    scratch  [0]       : %08x\n"
-		"    scratch  [1]       : %08x\n"
-		"    scratch  [2]       : %08x\n"
-		"    scratch  [3]       : %08x\n"
-		"    scratch  [4]       : %08x\n"
-		"    scratch  [5]       : %08x\n"
-		"    scratch  [6]       : %08x\n"
-		"    scratch  [7]       : %08x\n"
-		"    scratch  [8]       : %08x\n"
-		"    scratch  [9]       : %08x\n"
-		"    scratch [10]       : %08x\n"
-		"    scratch [11]       : %08x\n"
-		"    scratch [12]       : %08x\n"
-		"    scratch [13]       : %08x\n"
-		"    scratch [14]       : %08x\n"
-		"    scratch [15]       : %08x\n"
-		"    pc                 : %08x\n"
-		"    unk_fault_addr     : %08x\n"
-		"    inst_fault_addr    : %08x\n"
-		"    data_fault_addr    : %08x\n"
-		"    inbox1_rptr        : %08x\n"
-		"    inbox1_wptr        : %08x\n"
-		"    inbox1_size        : %08x\n"
-		"    inbox0_rptr        : %08x\n"
-		"    inbox0_wptr        : %08x\n"
-		"    inbox0_size        : %08x\n"
-		"    is_enabled         : %d\n"
-		"    is_soft_reset      : %d\n"
-		"    is_secure_reset    : %d\n"
-		"    is_traceport_en    : %d\n"
-		"    is_cw0_en          : %d\n"
-		"    is_cw6_en          : %d\n",
-		diag_data.dmcub_version,
-		diag_data.scratch[0],
-		diag_data.scratch[1],
-		diag_data.scratch[2],
-		diag_data.scratch[3],
-		diag_data.scratch[4],
-		diag_data.scratch[5],
-		diag_data.scratch[6],
-		diag_data.scratch[7],
-		diag_data.scratch[8],
-		diag_data.scratch[9],
-		diag_data.scratch[10],
-		diag_data.scratch[11],
-		diag_data.scratch[12],
-		diag_data.scratch[13],
-		diag_data.scratch[14],
-		diag_data.scratch[15],
-		diag_data.pc,
-		diag_data.undefined_address_fault_addr,
-		diag_data.inst_fetch_fault_addr,
-		diag_data.data_write_fault_addr,
-		diag_data.inbox1_rptr,
-		diag_data.inbox1_wptr,
-		diag_data.inbox1_size,
-		diag_data.inbox0_rptr,
-		diag_data.inbox0_wptr,
-		diag_data.inbox0_size,
-		diag_data.is_dmcub_enabled,
-		diag_data.is_dmcub_soft_reset,
-		diag_data.is_dmcub_secure_reset,
-		diag_data.is_traceport_en,
-		diag_data.is_cw0_enabled,
-		diag_data.is_cw6_enabled);
+	DC_LOG_DEBUG("DMCUB STATE:");
+	DC_LOG_DEBUG("    dmcub_version      : %08x", diag_data.dmcub_version);
+	DC_LOG_DEBUG("    scratch  [0]       : %08x", diag_data.scratch[0]);
+	DC_LOG_DEBUG("    scratch  [1]       : %08x", diag_data.scratch[1]);
+	DC_LOG_DEBUG("    scratch  [2]       : %08x", diag_data.scratch[2]);
+	DC_LOG_DEBUG("    scratch  [3]       : %08x", diag_data.scratch[3]);
+	DC_LOG_DEBUG("    scratch  [4]       : %08x", diag_data.scratch[4]);
+	DC_LOG_DEBUG("    scratch  [5]       : %08x", diag_data.scratch[5]);
+	DC_LOG_DEBUG("    scratch  [6]       : %08x", diag_data.scratch[6]);
+	DC_LOG_DEBUG("    scratch  [7]       : %08x", diag_data.scratch[7]);
+	DC_LOG_DEBUG("    scratch  [8]       : %08x", diag_data.scratch[8]);
+	DC_LOG_DEBUG("    scratch  [9]       : %08x", diag_data.scratch[9]);
+	DC_LOG_DEBUG("    scratch [10]       : %08x", diag_data.scratch[10]);
+	DC_LOG_DEBUG("    scratch [11]       : %08x", diag_data.scratch[11]);
+	DC_LOG_DEBUG("    scratch [12]       : %08x", diag_data.scratch[12]);
+	DC_LOG_DEBUG("    scratch [13]       : %08x", diag_data.scratch[13]);
+	DC_LOG_DEBUG("    scratch [14]       : %08x", diag_data.scratch[14]);
+	DC_LOG_DEBUG("    scratch [15]       : %08x", diag_data.scratch[15]);
+	DC_LOG_DEBUG("    pc                 : %08x", diag_data.pc);
+	DC_LOG_DEBUG("    unk_fault_addr     : %08x", diag_data.undefined_address_fault_addr);
+	DC_LOG_DEBUG("    inst_fault_addr    : %08x", diag_data.inst_fetch_fault_addr);
+	DC_LOG_DEBUG("    data_fault_addr    : %08x", diag_data.data_write_fault_addr);
+	DC_LOG_DEBUG("    inbox1_rptr        : %08x", diag_data.inbox1_rptr);
+	DC_LOG_DEBUG("    inbox1_wptr        : %08x", diag_data.inbox1_wptr);
+	DC_LOG_DEBUG("    inbox1_size        : %08x", diag_data.inbox1_size);
+	DC_LOG_DEBUG("    inbox0_rptr        : %08x", diag_data.inbox0_rptr);
+	DC_LOG_DEBUG("    inbox0_wptr        : %08x", diag_data.inbox0_wptr);
+	DC_LOG_DEBUG("    inbox0_size        : %08x", diag_data.inbox0_size);
+	DC_LOG_DEBUG("    is_enabled         : %d", diag_data.is_dmcub_enabled);
+	DC_LOG_DEBUG("    is_soft_reset      : %d", diag_data.is_dmcub_soft_reset);
+	DC_LOG_DEBUG("    is_secure_reset    : %d", diag_data.is_dmcub_secure_reset);
+	DC_LOG_DEBUG("    is_traceport_en    : %d", diag_data.is_traceport_en);
+	DC_LOG_DEBUG("    is_cw0_en          : %d", diag_data.is_cw0_enabled);
+	DC_LOG_DEBUG("    is_cw6_en          : %d", diag_data.is_cw6_enabled);
 }
 
 static bool dc_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx)
-- 
GitLab


From 75589226372ce5255ffade2ec6dea862338f7595 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Tue, 9 May 2023 09:38:29 -0400
Subject: [PATCH 1390/1791] drm/amd/display: disable dcn315 pixel rate crb when
 scaling

The rough calculation does not account for scaling. Also, make 2
segments the minimum allowed per surface to avoid potential 0 detile
with mpc/odm combine on such outputs.

Reviewed-by: Ariel Bernstein <eric.bernstein@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn315/dcn315_resource.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index 42a0157fd8133..c6bc2d603ab8b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -1666,6 +1666,14 @@ static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context)
 		if (!res_ctx->pipe_ctx[i].stream)
 			continue;
 
+		/*Don't apply if scaling*/
+		if (res_ctx->pipe_ctx[i].stream->src.width != res_ctx->pipe_ctx[i].stream->dst.width ||
+				res_ctx->pipe_ctx[i].stream->src.height != res_ctx->pipe_ctx[i].stream->dst.height ||
+				(res_ctx->pipe_ctx[i].top_pipe->plane_state && (res_ctx->pipe_ctx[i].top_pipe->plane_state->src_rect.width
+														!= res_ctx->pipe_ctx[i].top_pipe->plane_state->dst_rect.width ||
+					res_ctx->pipe_ctx[i].top_pipe->plane_state->src_rect.height
+														!= res_ctx->pipe_ctx[i].top_pipe->plane_state->dst_rect.height)))
+			return false;
 		/*Don't apply if MPO to avoid transition issues*/
 		if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state != res_ctx->pipe_ctx[i].plane_state)
 			return false;
@@ -1715,10 +1723,15 @@ static int dcn315_populate_dml_pipes_from_context(
 			/* Ceil to crb segment size */
 			int approx_det_segs_required_for_pstate = dcn_get_approx_det_segs_required_for_pstate(
 					&context->bw_ctx.dml.soc, timing->pix_clk_100hz, bpp, DCN3_15_CRB_SEGMENT_SIZE_KB);
+
 			if (approx_det_segs_required_for_pstate <= 2 * DCN3_15_MAX_DET_SEGS) {
 				bool split_required = approx_det_segs_required_for_pstate > DCN3_15_MAX_DET_SEGS;
 				split_required = split_required || timing->pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc);
 				split_required = split_required || (pipe->plane_state && pipe->plane_state->src_rect.width > 5120);
+
+				/* Minimum 2 segments to allow mpc/odm combine if its used later */
+				if (approx_det_segs_required_for_pstate < 2)
+					approx_det_segs_required_for_pstate = 2;
 				if (split_required)
 					approx_det_segs_required_for_pstate += approx_det_segs_required_for_pstate % 2;
 				pipes[pipe_cnt].pipe.src.det_size_override = approx_det_segs_required_for_pstate;
-- 
GitLab


From 6812d74803740100a0c422b9bc1fda947af4da6a Mon Sep 17 00:00:00 2001
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Date: Mon, 8 May 2023 16:08:37 -0400
Subject: [PATCH 1391/1791] drm/amd/display: Update SR watermarks for DCN314

[Why & How]
Update parameters for SR watermarks for DCN314

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c   | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 4d5cd59f6433a..2a0c696f58619 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -403,32 +403,32 @@ static struct wm_table lpddr5_wm_table = {
 			.wm_inst = WM_A,
 			.wm_type = WM_TYPE_PSTATE_CHG,
 			.pstate_latency_us = 11.65333,
-			.sr_exit_time_us = 16.5,
-			.sr_enter_plus_exit_time_us = 18.5,
+			.sr_exit_time_us = 30.0,
+			.sr_enter_plus_exit_time_us = 32.0,
 			.valid = true,
 		},
 		{
 			.wm_inst = WM_B,
 			.wm_type = WM_TYPE_PSTATE_CHG,
 			.pstate_latency_us = 11.65333,
-			.sr_exit_time_us = 16.5,
-			.sr_enter_plus_exit_time_us = 18.5,
+			.sr_exit_time_us = 30.0,
+			.sr_enter_plus_exit_time_us = 32.0,
 			.valid = true,
 		},
 		{
 			.wm_inst = WM_C,
 			.wm_type = WM_TYPE_PSTATE_CHG,
 			.pstate_latency_us = 11.65333,
-			.sr_exit_time_us = 16.5,
-			.sr_enter_plus_exit_time_us = 18.5,
+			.sr_exit_time_us = 30.0,
+			.sr_enter_plus_exit_time_us = 32.0,
 			.valid = true,
 		},
 		{
 			.wm_inst = WM_D,
 			.wm_type = WM_TYPE_PSTATE_CHG,
 			.pstate_latency_us = 11.65333,
-			.sr_exit_time_us = 16.5,
-			.sr_enter_plus_exit_time_us = 18.5,
+			.sr_exit_time_us = 30.0,
+			.sr_enter_plus_exit_time_us = 32.0,
 			.valid = true,
 		},
 	}
-- 
GitLab


From cd465a670087f94e62100622f9cbb894f524268a Mon Sep 17 00:00:00 2001
From: Alan Liu <haoping.liu@amd.com>
Date: Tue, 2 May 2023 17:54:50 +0800
Subject: [PATCH 1392/1791] drm/amd/display: Fix warning in disabling vblank
 irq

[Why]
During gpu-reset, we toggle vblank irq by calling dc_interrupt_set()
instead of amdgpu_irq_get/put() because we don't want to change the irq
source's refcount. However, we see the warning when vblank irq is enabled
by dc_interrupt_set() during gpu-reset but disabled by amdgpu_irq_put()
after gpu-reset.

[How]
Only in dm_gpureset_toggle_interrupts() we toggle vblank interrupts by
calling dc_interrupt_set(). Apart from this we call dm_set_vblank()
which uses amdgpu_irq_get/put() to operate vblank irq.

Reviewed-by: Bhawanpreet Lakha <bhawanpreet.lakha@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alan Liu <haoping.liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 21 ++++++++++++-------
 .../amd/display/amdgpu_dm/amdgpu_dm_crtc.c    | 16 +++-----------
 2 files changed, 16 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 4edbcc9e535bd..f59cb4d659dc9 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2475,20 +2475,25 @@ static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev,
 		if (acrtc && state->stream_status[i].plane_count != 0) {
 			irq_source = IRQ_TYPE_PFLIP + acrtc->otg_inst;
 			rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY;
-			DRM_DEBUG_VBL("crtc %d - vupdate irq %sabling: r=%d\n",
-				      acrtc->crtc_id, enable ? "en" : "dis", rc);
 			if (rc)
 				DRM_WARN("Failed to %s pflip interrupts\n",
 					 enable ? "enable" : "disable");
 
 			if (enable) {
-				rc = amdgpu_dm_crtc_enable_vblank(&acrtc->base);
-				if (rc)
-					DRM_WARN("Failed to enable vblank interrupts\n");
-			} else {
-				amdgpu_dm_crtc_disable_vblank(&acrtc->base);
-			}
+				if (amdgpu_dm_crtc_vrr_active(to_dm_crtc_state(acrtc->base.state)))
+					rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, true);
+			} else
+				rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, false);
 
+			if (rc)
+				DRM_WARN("Failed to %sable vupdate interrupt\n", enable ? "en" : "dis");
+
+			irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
+			/* During gpu-reset we disable and then enable vblank irq, so
+			 * don't use amdgpu_irq_get/put() to avoid refcount change.
+			 */
+			if (!dc_interrupt_set(adev->dm.dc, irq_source, enable))
+				DRM_WARN("Failed to %sable vblank interrupt\n", enable ? "en" : "dis");
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index e3762e806617c..440fc0869a34b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -146,7 +146,6 @@ static void vblank_control_worker(struct work_struct *work)
 
 static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
 {
-	enum dc_irq_source irq_source;
 	struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
 	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state);
@@ -169,18 +168,9 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable)
 	if (rc)
 		return rc;
 
-	if (amdgpu_in_reset(adev)) {
-		irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
-		/* During gpu-reset we disable and then enable vblank irq, so
-		 * don't use amdgpu_irq_get/put() to avoid refcount change.
-		 */
-		if (!dc_interrupt_set(adev->dm.dc, irq_source, enable))
-			rc = -EBUSY;
-	} else {
-		rc = (enable)
-			? amdgpu_irq_get(adev, &adev->crtc_irq, acrtc->crtc_id)
-			: amdgpu_irq_put(adev, &adev->crtc_irq, acrtc->crtc_id);
-	}
+	rc = (enable)
+		? amdgpu_irq_get(adev, &adev->crtc_irq, acrtc->crtc_id)
+		: amdgpu_irq_put(adev, &adev->crtc_irq, acrtc->crtc_id);
 
 	if (rc)
 		return rc;
-- 
GitLab


From 38ff516bb00cd8e974c8b5e70ab6e1b354b8f424 Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Mon, 24 Apr 2023 11:39:20 -0400
Subject: [PATCH 1393/1791] drm/amd/display: lower dp link training message
 level

[Why] some test apps report dp link training waring even dp
training pass. there are 4 tries of lt within
perform_link_training_with_retries. if lt pass within 4 tries,
it will NOT be reated as lt failure. for each try of lt, if lt
fails, current driver implementation prints message at warning
level. this let people think dp lt does not work properly.

[How] for 1st, 2nd and 3rd try of lt, print message at debug
level. for the 4th try of lt, print message at warning level.

Reviewed-by: Jerry Zuo <jerry.zuo@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/link/protocols/link_dp_training.c        | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index 579fa222810d4..e011df4bdaf21 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -1653,10 +1653,19 @@ bool perform_link_training_with_retries(
 				break;
 		}
 
-		DC_LOG_WARNING("%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n",
-			       __func__, link->link_index, (unsigned int)j + 1, attempts,
-			      cur_link_settings.link_rate, cur_link_settings.lane_count,
-			      cur_link_settings.link_spread, status);
+		if (j == (attempts - 1)) {
+			DC_LOG_WARNING(
+				"%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n",
+				__func__, link->link_index, (unsigned int)j + 1, attempts,
+				cur_link_settings.link_rate, cur_link_settings.lane_count,
+				cur_link_settings.link_spread, status);
+		} else {
+			DC_LOG_HW_LINK_TRAINING(
+				"%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n",
+				__func__, link->link_index, (unsigned int)j + 1, attempts,
+				cur_link_settings.link_rate, cur_link_settings.lane_count,
+				cur_link_settings.link_spread, status);
+		}
 
 		dp_disable_link_phy(link, &pipe_ctx->link_res, signal);
 
-- 
GitLab


From 0ab720d506252a28983baabafa2605eb6c94b1d7 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Wed, 10 May 2023 15:27:19 -0400
Subject: [PATCH 1394/1791] drm/amd/display: fix dcn315 pixel rate crb scaling
 check

fix dcn315 pixel rate crb scaling check error

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index c6bc2d603ab8b..95fd3d087ea35 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -1669,10 +1669,10 @@ static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context)
 		/*Don't apply if scaling*/
 		if (res_ctx->pipe_ctx[i].stream->src.width != res_ctx->pipe_ctx[i].stream->dst.width ||
 				res_ctx->pipe_ctx[i].stream->src.height != res_ctx->pipe_ctx[i].stream->dst.height ||
-				(res_ctx->pipe_ctx[i].top_pipe->plane_state && (res_ctx->pipe_ctx[i].top_pipe->plane_state->src_rect.width
-														!= res_ctx->pipe_ctx[i].top_pipe->plane_state->dst_rect.width ||
-					res_ctx->pipe_ctx[i].top_pipe->plane_state->src_rect.height
-														!= res_ctx->pipe_ctx[i].top_pipe->plane_state->dst_rect.height)))
+				(res_ctx->pipe_ctx[i].plane_state && (res_ctx->pipe_ctx[i].plane_state->src_rect.width
+														!= res_ctx->pipe_ctx[i].plane_state->dst_rect.width ||
+					res_ctx->pipe_ctx[i].plane_state->src_rect.height
+														!= res_ctx->pipe_ctx[i].plane_state->dst_rect.height)))
 			return false;
 		/*Don't apply if MPO to avoid transition issues*/
 		if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state != res_ctx->pipe_ctx[i].plane_state)
-- 
GitLab


From 91b38ca1b331ef1af3b77e2ffdb41654e0fba127 Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <jerry.zuo@amd.com>
Date: Wed, 10 May 2023 16:43:30 -0400
Subject: [PATCH 1395/1791] drm/amd/display: Have Payload Properly Created
 After Resume

At drm suspend sequence, MST dc_sink is removed. When commit cached
MST stream back in drm resume sequence, the MST stream payload is not
properly created and added into the payload table. After resume, topology
change is reprobed by removing existing streams first. That leads to
no payload is found in the existing payload table as below error
"[drm] ERROR No payload for [MST PORT:] found in mst state"

1. In encoder .atomic_check routine, remove check existance of dc_sink
2. Bypass MST by checking existence of MST root port. dc_link_type cannot
differentiate MST port before topology is rediscovered.

Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index f59cb4d659dc9..c98c57deb5ebf 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2853,7 +2853,7 @@ static int dm_resume(void *handle)
 		 * this is the case when traversing through already created
 		 * MST connectors, should be skipped
 		 */
-		if (aconnector->dc_link->type == dc_connection_mst_branch)
+		if (aconnector && aconnector->mst_root)
 			continue;
 
 		mutex_lock(&aconnector->hpd_lock);
@@ -6776,7 +6776,7 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder,
 	int clock, bpp = 0;
 	bool is_y420 = false;
 
-	if (!aconnector->mst_output_port || !aconnector->dc_sink)
+	if (!aconnector->mst_output_port)
 		return 0;
 
 	mst_port = aconnector->mst_output_port;
-- 
GitLab


From 3e8d74cb128fb1a4d56270ffbecea6056c55739a Mon Sep 17 00:00:00 2001
From: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Date: Tue, 9 May 2023 14:41:59 -0400
Subject: [PATCH 1396/1791] drm/amd/display: Trigger DIO FIFO resync on commit
 streams

[WHY]
Currently, there is an intermittent issue where a screen can either go
blank or be corrupted.

[HOW]
To resolve the issue we trigger the ramping logic for DIO FIFO so that
it goes back up to the correct speed.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dce110/dce110_hw_sequencer.c   |  3 +++
 .../gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h |  4 ++-
 .../drm/amd/display/dc/dcn314/dcn314_dccg.c   | 11 ++++++++
 .../drm/amd/display/dc/dcn314/dcn314_dccg.h   |  5 +++-
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  | 27 +++++++++++++++++++
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |  2 ++
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |  1 +
 drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h  |  3 +++
 .../amd/display/dc/inc/hw_sequencer_private.h |  2 ++
 9 files changed, 56 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 8d2460d06bced..1a0be40d125c6 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -2291,6 +2291,9 @@ enum dc_status dce110_apply_ctx_to_hw(
 
 		if (DC_OK != status)
 			return status;
+
+		if (hws->funcs.resync_fifo_dccg_dio)
+			hws->funcs.resync_fifo_dccg_dio(hws, dc, context);
 	}
 
 	if (dc->fbc_compressor)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
index 7bdc146f7cb59..c8602bcfa393a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h
@@ -208,7 +208,9 @@
 #define DCCG314_REG_FIELD_LIST(type) \
 	type DSCCLK3_DTO_PHASE;\
 	type DSCCLK3_DTO_MODULO;\
-	type DSCCLK3_DTO_ENABLE;
+	type DSCCLK3_DTO_ENABLE;\
+	type DENTIST_DISPCLK_RDIVIDER;\
+	type DENTIST_DISPCLK_WDIVIDER;
 
 #define DCCG32_REG_FIELD_LIST(type) \
 	type DPSTREAMCLK0_EN;\
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
index de7bfba2c1798..e0e7d32bb1a0e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
@@ -45,6 +45,16 @@
 #define DC_LOGGER \
 	dccg->ctx->logger
 
+static void dccg314_trigger_dio_fifo_resync(
+	struct dccg *dccg)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	uint32_t dispclk_rdivider_value = 0;
+
+	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value);
+	REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value);
+}
+
 static void dccg314_get_pixel_rate_div(
 		struct dccg *dccg,
 		uint32_t otg_inst,
@@ -357,6 +367,7 @@ static const struct dccg_funcs dccg314_funcs = {
 	.disable_dsc = dccg31_disable_dscclk,
 	.enable_dsc = dccg31_enable_dscclk,
 	.set_pixel_rate_div = dccg314_set_pixel_rate_div,
+	.trigger_dio_fifo_resync = dccg314_trigger_dio_fifo_resync,
 	.set_valid_pixel_rate = dccg314_set_valid_pixel_rate,
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
index 90687a9e8fddd..8e07d3151f915 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h
@@ -192,7 +192,10 @@
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYDSYMCLK_GATE_DISABLE, mask_sh),\
-	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh)
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh),\
+	DCCG_SF(HDMISTREAMCLK0_DTO_PARAM, HDMISTREAMCLK0_DTO_MODULO, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh)
 
 struct dccg *dccg314_create(
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index f7a3e0d71d016..70fac2ebb7576 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -390,6 +390,33 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 				pix_per_cycle);
 }
 
+void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context)
+{
+	uint8_t i;
+	struct pipe_ctx *pipe = NULL;
+	bool otg_disabled[MAX_PIPES] = {false};
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe->top_pipe || pipe->prev_odm_pipe)
+			continue;
+
+		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+			pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg);
+			reset_sync_context_for_pipe(dc, context, i);
+			otg_disabled[i] = true;
+		}
+	}
+
+	hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg);
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (otg_disabled[i])
+			pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg);
+	}
+}
+
 void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on)
 {
 	if (!hws->ctx->dc->debug.root_clock_optimization.bits.dpp)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
index 6d0b62503caa6..559d71002e8a4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
@@ -41,6 +41,8 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 
 void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
 
+void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context);
+
 void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on);
 
 void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index 7a28c7bb25d20..90be62c058229 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -152,6 +152,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = {
 	.setup_hpo_hw_control = dcn31_setup_hpo_hw_control,
 	.calculate_dccg_k1_k2_values = dcn314_calculate_dccg_k1_k2_values,
 	.set_pixels_per_cycle = dcn314_set_pixels_per_cycle,
+	.resync_fifo_dccg_dio = dcn314_resync_fifo_dccg_dio,
 };
 
 void dcn314_hw_sequencer_construct(struct dc *dc)
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
index ad6acd1b34e1d..0b700b3d7d972 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -159,6 +159,9 @@ struct dccg_funcs {
 			int otg_inst,
 			int pixclk_khz);
 
+	void (*trigger_dio_fifo_resync)(
+			struct dccg *dccg);
+
 	void (*dpp_root_clock_control)(
 			struct dccg *dccg,
 			unsigned int dpp_inst,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
index 4513544559be2..4ca4192c1e127 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
@@ -160,6 +160,8 @@ struct hwseq_private_funcs {
 			unsigned int *k1_div,
 			unsigned int *k2_div);
 	void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx);
+	void (*resync_fifo_dccg_dio)(struct dce_hwseq *hws, struct dc *dc,
+			struct dc_state *context);
 	bool (*is_dp_dig_pixel_rate_div_policy)(struct pipe_ctx *pipe_ctx);
 #endif
 };
-- 
GitLab


From c02b04633c4f4654331c53966cb937df1c73a9bb Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Thu, 11 May 2023 09:12:09 -0400
Subject: [PATCH 1397/1791] drm/amd/display: Revert vblank change that causes
 null pointer crash

Revert commit 1a4bcdbea431 ("drm/amd/display: Fix possible underflow for displays with large vblank")
Because it cause some regression

Fixes: 1a4bcdbea431 ("drm/amd/display: Fix possible underflow for displays with large vblank")
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 554152371eb53..1d00eb9e73c62 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -33,7 +33,7 @@
 #include "dml/display_mode_vba.h"
 
 struct _vcs_dpi_ip_params_st dcn3_14_ip = {
-	.VBlankNomDefaultUS = 800,
+	.VBlankNomDefaultUS = 668,
 	.gpuvm_enable = 1,
 	.gpuvm_max_page_table_levels = 1,
 	.hostvm_enable = 1,
@@ -286,7 +286,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *pipe;
 	bool upscaled = false;
-	const unsigned int max_allowed_vblank_nom = 1023;
+	bool isFreesyncVideo = false;
 
 	dc_assert_fp_enabled();
 
@@ -300,11 +300,16 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 		pipe = &res_ctx->pipe_ctx[i];
 		timing = &pipe->stream->timing;
 
-		pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
-		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
-		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, dcn3_14_ip.VBlankNomDefaultUS);
-		pipes[pipe_cnt].pipe.dest.vblank_nom = max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width);
-		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
+		isFreesyncVideo = pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min;
+		isFreesyncVideo = isFreesyncVideo && pipe->stream->adjust.v_total_min > timing->v_total;
+
+		if (!isFreesyncVideo) {
+			pipes[pipe_cnt].pipe.dest.vblank_nom =
+				dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0));
+		} else {
+			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		}
 
 		if (pipe->plane_state &&
 				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
-- 
GitLab


From de231189e7bf1a38c0f889ec5f8911af473aa792 Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Thu, 11 May 2023 10:51:27 -0400
Subject: [PATCH 1398/1791] drm/amd/display: Fix possible underflow for
 displays with large vblank

[Why]
Underflow observed when using a display with a large vblank region
and low refresh rate

[How]
Simplify calculation of vblank_nom

Increase value for VBlankNomDefaultUS to 800us

Fixed a null pointer from previous commit of this change

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 1d00eb9e73c62..554152371eb53 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -33,7 +33,7 @@
 #include "dml/display_mode_vba.h"
 
 struct _vcs_dpi_ip_params_st dcn3_14_ip = {
-	.VBlankNomDefaultUS = 668,
+	.VBlankNomDefaultUS = 800,
 	.gpuvm_enable = 1,
 	.gpuvm_max_page_table_levels = 1,
 	.hostvm_enable = 1,
@@ -286,7 +286,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 	struct resource_context *res_ctx = &context->res_ctx;
 	struct pipe_ctx *pipe;
 	bool upscaled = false;
-	bool isFreesyncVideo = false;
+	const unsigned int max_allowed_vblank_nom = 1023;
 
 	dc_assert_fp_enabled();
 
@@ -300,16 +300,11 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c
 		pipe = &res_ctx->pipe_ctx[i];
 		timing = &pipe->stream->timing;
 
-		isFreesyncVideo = pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min;
-		isFreesyncVideo = isFreesyncVideo && pipe->stream->adjust.v_total_min > timing->v_total;
-
-		if (!isFreesyncVideo) {
-			pipes[pipe_cnt].pipe.dest.vblank_nom =
-				dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0));
-		} else {
-			pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
-			pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
-		}
+		pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive;
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, dcn3_14_ip.VBlankNomDefaultUS);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width);
+		pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom);
 
 		if (pipe->plane_state &&
 				(pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height ||
-- 
GitLab


From 7e60ab4eb3e4ba2adac46d737fdbbc5732bebd58 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Thu, 11 May 2023 15:25:01 -0400
Subject: [PATCH 1399/1791] drm/amd/display: Apply 60us prefetch for DCFCLK <=
 300Mhz

[Description]
- Previously we wanted to apply extra 60us of prefetch for min DCFCLK
  (200Mhz), but DCFCLK can be calculated to be 201Mhz which underflows
  also without the extra prefetch
- Instead, apply the the extra 60us prefetch for any DCFCLK freq <=
  300Mhz

Reviewed-by: Nevenko Stupar <nevenko.stupar@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c    | 4 ++--
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h    | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
index 958d27224f644..cbdfb762c10c5 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c
@@ -811,7 +811,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 					v->SwathHeightC[k],
 					TWait,
 					(v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ||
-						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= MIN_DCFCLK_FREQ_MHZ) ?
+						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 							mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 					/* Output */
 					&v->DSTXAfterScaler[k],
@@ -3315,7 +3315,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->swath_width_chroma_ub_this_state[k],
 							v->SwathHeightYThisState[k],
 							v->SwathHeightCThisState[k], v->TWait,
-							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= MIN_DCFCLK_FREQ_MHZ) ?
+							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 									mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 
 							/* Output */
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
index d98e36a9a09cc..c4745d63039bb 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h
@@ -53,7 +53,7 @@
 #define BPP_BLENDED_PIPE 0xffffffff
 
 #define MEM_STROBE_FREQ_MHZ 1600
-#define MIN_DCFCLK_FREQ_MHZ 200
+#define DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ 300
 #define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0
 
 struct display_mode_lib;
-- 
GitLab


From 25879d7b4986beba3f0d84762fe40d09fdc8b219 Mon Sep 17 00:00:00 2001
From: Qingqing Zhuo <qingqing.zhuo@amd.com>
Date: Thu, 16 Mar 2023 09:05:58 -0400
Subject: [PATCH 1400/1791] drm/amd/display: Clean FPGA code in dc

[Why]
Drop dead code for Linux.

[How]
Remove all IS_FPGA_MAXIMUS_DC and IS_DIAG_DC

Reviewed-by: Ariel Bernstein <eric.bernstein@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/clk_mgr/dce112/dce112_clk_mgr.c        |  20 +-
 .../dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c  |  10 +-
 .../display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c  |  58 +++--
 .../dc/clk_mgr/dcn201/dcn201_clk_mgr.c        |  22 +-
 .../amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c |  73 +++---
 .../dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c   |  17 +-
 .../display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c  |  32 ++-
 .../display/dc/clk_mgr/dcn301/vg_clk_mgr.c    |  50 ++--
 .../display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c  |  38 ++-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        |  38 ++-
 .../dc/clk_mgr/dcn315/dcn315_clk_mgr.c        |  38 ++-
 .../dc/clk_mgr/dcn316/dcn316_clk_mgr.c        |  49 ++--
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  |   1 -
 drivers/gpu/drm/amd/display/dc/core/dc.c      |  40 ++--
 .../gpu/drm/amd/display/dc/core/dc_stream.c   |  30 ---
 drivers/gpu/drm/amd/display/dc/dc_helper.c    |   6 +-
 drivers/gpu/drm/amd/display/dc/dc_types.h     |   7 -
 .../drm/amd/display/dc/dce/dce_clock_source.c |  38 ---
 drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c |   8 +-
 .../display/dc/dce110/dce110_hw_sequencer.c   |   3 -
 .../display/dc/dce112/dce112_hw_sequencer.c   |   3 -
 .../display/dc/dce120/dce120_hw_sequencer.c   |   3 -
 .../amd/display/dc/dcn10/dcn10_hw_sequencer.c | 126 ++++------
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c |   8 +-
 .../drm/amd/display/dc/dcn10/dcn10_resource.c |   5 +-
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.c    |  51 ++--
 .../gpu/drm/amd/display/dc/dcn20/dcn20_init.c |   4 -
 .../gpu/drm/amd/display/dc/dcn20/dcn20_optc.c |   7 +-
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |  15 +-
 .../drm/amd/display/dc/dcn201/dcn201_hwseq.c  |  79 +++----
 .../drm/amd/display/dc/dcn201/dcn201_optc.c   |   7 +-
 .../amd/display/dc/dcn201/dcn201_resource.c   |   5 +-
 .../gpu/drm/amd/display/dc/dcn21/dcn21_init.c |   4 -
 .../drm/amd/display/dc/dcn21/dcn21_resource.c |  10 +-
 .../drm/amd/display/dc/dcn30/dcn30_hwseq.c    |  54 ++---
 .../gpu/drm/amd/display/dc/dcn30/dcn30_init.c |   4 -
 .../gpu/drm/amd/display/dc/dcn30/dcn30_optc.c |   7 +-
 .../drm/amd/display/dc/dcn30/dcn30_resource.c |   8 +-
 .../amd/display/dc/dcn301/dcn301_resource.c   |  10 +-
 .../amd/display/dc/dcn302/dcn302_resource.c   |   5 +-
 .../amd/display/dc/dcn303/dcn303_resource.c   |   5 +-
 .../drm/amd/display/dc/dcn31/dcn31_hwseq.c    | 106 ++++-----
 .../gpu/drm/amd/display/dc/dcn31/dcn31_init.c |   4 -
 .../drm/amd/display/dc/dcn31/dcn31_resource.c |  17 +-
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |   4 -
 .../amd/display/dc/dcn314/dcn314_resource.c   |  10 +-
 .../amd/display/dc/dcn315/dcn315_resource.c   |  17 +-
 .../amd/display/dc/dcn316/dcn316_resource.c   |  17 +-
 .../gpu/drm/amd/display/dc/dcn32/dcn32_init.c |   4 -
 .../drm/amd/display/dc/dcn32/dcn32_resource.c |  13 +-
 .../amd/display/dc/dcn321/dcn321_resource.c   |  13 +-
 .../amd/display/dc/dml/dcn301/dcn301_fpu.c    |  70 +++---
 .../drm/amd/display/dc/dml/dcn31/dcn31_fpu.c  | 219 ++++++++---------
 .../dc/dml/dcn31/display_rq_dlg_calc_31.c     |   8 -
 .../amd/display/dc/dml/dcn314/dcn314_fpu.c    |   6 +-
 .../dc/dml/dcn314/display_rq_dlg_calc_314.c   |   8 -
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 120 +++++-----
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    | 120 +++++-----
 .../drm/amd/display/dc/dml/display_mode_lib.c |   1 -
 .../drm/amd/display/dc/dml/display_mode_lib.h |   1 -
 .../display/dc/link/hwss/link_hwss_hpo_dp.c   |  83 +------
 .../gpu/drm/amd/display/dc/link/link_dpms.c   | 220 ++++++++----------
 .../dc/link/protocols/link_dp_capability.c    |   3 +-
 63 files changed, 770 insertions(+), 1292 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
index 934e6423dc1ad..1f36ad8a7de46 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c
@@ -111,12 +111,10 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz)
 
 	bp->funcs->set_dce_clock(bp, &dce_clk_params);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
-			if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock)
-				dmcu->funcs->set_psr_wait_loop(dmcu,
-						actual_clock / 1000 / 7);
-		}
+	if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
+		if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock)
+			dmcu->funcs->set_psr_wait_loop(dmcu,
+					actual_clock / 1000 / 7);
 	}
 
 	clk_mgr_dce->dfs_bypass_disp_clk = actual_clock;
@@ -153,12 +151,10 @@ int dce112_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_clk_khz)
 		clk_mgr->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
 
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
-			if (clk_mgr->dfs_bypass_disp_clk != actual_clock)
-				dmcu->funcs->set_psr_wait_loop(dmcu,
-						actual_clock / 1000 / 7);
-		}
+	if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
+		if (clk_mgr->dfs_bypass_disp_clk != actual_clock)
+			dmcu->funcs->set_psr_wait_loop(dmcu,
+					actual_clock / 1000 / 7);
 	}
 
 	clk_mgr->dfs_bypass_disp_clk = actual_clock;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
index 450eaead4f20c..89b79dd396283 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c
@@ -135,12 +135,10 @@ int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_di
 			VBIOSSMC_MSG_SetDispclkFreq,
 			khz_to_mhz_ceil(requested_dispclk_khz));
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
-			if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz)
-				dmcu->funcs->set_psr_wait_loop(dmcu,
-						actual_dispclk_set_mhz / 7);
-		}
+	if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
+		if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz)
+			dmcu->funcs->set_psr_wait_loop(dmcu,
+					actual_dispclk_set_mhz / 7);
 	}
 
 	return actual_dispclk_set_mhz * 1000;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
index 650f3b4b562e9..c435f7632e8e8 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c
@@ -531,6 +531,11 @@ void dcn20_clk_mgr_construct(
 		struct pp_smu_funcs *pp_smu,
 		struct dccg *dccg)
 {
+	int dprefclk_did;
+	int target_div;
+	uint32_t pll_req_reg;
+	struct fixed31_32 pll_req;
+
 	clk_mgr->base.ctx = ctx;
 	clk_mgr->pp_smu = pp_smu;
 	clk_mgr->base.funcs = &dcn2_funcs;
@@ -547,42 +552,34 @@ void dcn20_clk_mgr_construct(
 
 	clk_mgr->base.dprefclk_khz = 700000; // 700 MHz planned if VCO is 3.85 GHz, will be retrieved
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		dcn2_funcs.update_clocks = dcn2_update_clocks_fpga;
-		clk_mgr->base.dentist_vco_freq_khz = 3850000;
+	/* DFS Slice 2 should be used for DPREFCLK */
+	dprefclk_did = REG_READ(CLK3_CLK2_DFS_CNTL);
+	/* Convert DPREFCLK DFS Slice DID to actual divider */
+	target_div = dentist_get_divider_from_did(dprefclk_did);
+	/* get FbMult value */
+	pll_req_reg = REG_READ(CLK3_CLK_PLL_REQ);
 
-	} else {
-		/* DFS Slice 2 should be used for DPREFCLK */
-		int dprefclk_did = REG_READ(CLK3_CLK2_DFS_CNTL);
-		/* Convert DPREFCLK DFS Slice DID to actual divider*/
-		int target_div = dentist_get_divider_from_did(dprefclk_did);
-
-		/* get FbMult value */
-		uint32_t pll_req_reg = REG_READ(CLK3_CLK_PLL_REQ);
-		struct fixed31_32 pll_req;
-
-		/* set up a fixed-point number
-		 * this works because the int part is on the right edge of the register
-		 * and the frac part is on the left edge
-		 */
+	/* set up a fixed-point number
+	 * this works because the int part is on the right edge of the register
+	 * and the frac part is on the left edge
+	 */
 
-		pll_req = dc_fixpt_from_int(pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_int);
-		pll_req.value |= pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_frac;
+	pll_req = dc_fixpt_from_int(pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_int);
+	pll_req.value |= pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_frac;
 
-		/* multiply by REFCLK period */
-		pll_req = dc_fixpt_mul_int(pll_req, 100000);
+	/* multiply by REFCLK period */
+	pll_req = dc_fixpt_mul_int(pll_req, 100000);
 
-		/* integer part is now VCO frequency in kHz */
-		clk_mgr->base.dentist_vco_freq_khz = dc_fixpt_floor(pll_req);
+	/* integer part is now VCO frequency in kHz */
+	clk_mgr->base.dentist_vco_freq_khz = dc_fixpt_floor(pll_req);
 
-		/* in case we don't get a value from the register, use default */
-		if (clk_mgr->base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.dentist_vco_freq_khz = 3850000;
+	/* in case we don't get a value from the register, use default */
+	if (clk_mgr->base.dentist_vco_freq_khz == 0)
+		clk_mgr->base.dentist_vco_freq_khz = 3850000;
 
-		/* Calculate the DPREFCLK in kHz.*/
-		clk_mgr->base.dprefclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
-			* clk_mgr->base.dentist_vco_freq_khz) / target_div;
-	}
+	/* Calculate the DPREFCLK in kHz.*/
+	clk_mgr->base.dprefclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
+		* clk_mgr->base.dentist_vco_freq_khz) / target_div;
 	//Integrated_info table does not exist on dGPU projects so should not be referenced
 	//anywhere in code for dGPUs.
 	//Also there is no plan for now that DFS BYPASS will be used on NV10/12/14.
@@ -590,4 +587,3 @@ void dcn20_clk_mgr_construct(
 
 	dce_clock_read_ss_info(clk_mgr);
 }
-
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c
index 811720749faf8..694fe4271b4d9 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c
@@ -190,23 +190,17 @@ void dcn201_clk_mgr_construct(struct dc_context *ctx,
 	clk_mgr->dprefclk_ss_divider = 1000;
 	clk_mgr->ss_on_dprefclk = false;
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		dcn201_funcs.update_clocks = dcn2_update_clocks_fpga;
-		clk_mgr->base.dprefclk_khz = 600000;
-		clk_mgr->base.dentist_vco_freq_khz = 3000000;
-	} else {
-		clk_mgr->base.dprefclk_khz = REG_READ(CLK4_CLK2_CURRENT_CNT);
-		clk_mgr->base.dprefclk_khz *= 100;
+	clk_mgr->base.dprefclk_khz = REG_READ(CLK4_CLK2_CURRENT_CNT);
+	clk_mgr->base.dprefclk_khz *= 100;
 
-		if (clk_mgr->base.dprefclk_khz == 0)
-			clk_mgr->base.dprefclk_khz = 600000;
+	if (clk_mgr->base.dprefclk_khz == 0)
+		clk_mgr->base.dprefclk_khz = 600000;
 
-		REG_GET(CLK4_CLK_PLL_REQ, FbMult_int, &clk_mgr->base.dentist_vco_freq_khz);
-		clk_mgr->base.dentist_vco_freq_khz *= 100000;
+	REG_GET(CLK4_CLK_PLL_REQ, FbMult_int, &clk_mgr->base.dentist_vco_freq_khz);
+	clk_mgr->base.dentist_vco_freq_khz *= 100000;
 
-		if (clk_mgr->base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.dentist_vco_freq_khz = 3000000;
-	}
+	if (clk_mgr->base.dentist_vco_freq_khz == 0)
+		clk_mgr->base.dentist_vco_freq_khz = 3000000;
 
 	if (!debug->disable_dfs_bypass && bp->integrated_info)
 		if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
index bd9fd0b54f46a..0c6a4ab72b1d2 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c
@@ -705,6 +705,7 @@ void rn_clk_mgr_construct(
 	struct dpm_clocks clock_table = { 0 };
 	enum pp_smu_status status = 0;
 	int is_green_sardine = 0;
+	struct clk_log_info log_info = {0};
 
 #if defined(CONFIG_DRM_AMD_DC_FP)
 	is_green_sardine = ASICREV_IS_GREEN_SARDINE(ctx->asic_id.hw_internal_rev);
@@ -725,48 +726,41 @@ void rn_clk_mgr_construct(
 
 	clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		dcn21_funcs.update_clocks = dcn2_update_clocks_fpga;
+	clk_mgr->periodic_retraining_disabled = rn_vbios_smu_is_periodic_retraining_disabled(clk_mgr);
+
+	/* SMU Version 55.51.0 and up no longer have an issue
+	 * that needs to limit minimum dispclk */
+	if (clk_mgr->smu_ver >= SMU_VER_55_51_0)
+		debug->min_disp_clk_khz = 0;
+
+	/* TODO: Check we get what we expect during bringup */
+	clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr);
+
+	/* in case we don't get a value from the register, use default */
+	if (clk_mgr->base.dentist_vco_freq_khz == 0)
 		clk_mgr->base.dentist_vco_freq_khz = 3600000;
-	} else {
-		struct clk_log_info log_info = {0};
-
-		clk_mgr->periodic_retraining_disabled = rn_vbios_smu_is_periodic_retraining_disabled(clk_mgr);
-
-		/* SMU Version 55.51.0 and up no longer have an issue
-		 * that needs to limit minimum dispclk */
-		if (clk_mgr->smu_ver >= SMU_VER_55_51_0)
-			debug->min_disp_clk_khz = 0;
-
-		/* TODO: Check we get what we expect during bringup */
-		clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr);
-
-		/* in case we don't get a value from the register, use default */
-		if (clk_mgr->base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.dentist_vco_freq_khz = 3600000;
-
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) {
-			if (clk_mgr->periodic_retraining_disabled) {
-				rn_bw_params.wm_table = lpddr4_wm_table_with_disabled_ppt;
-			} else {
-				if (is_green_sardine)
-					rn_bw_params.wm_table = lpddr4_wm_table_gs;
-				else
-					rn_bw_params.wm_table = lpddr4_wm_table_rn;
-			}
+
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) {
+		if (clk_mgr->periodic_retraining_disabled) {
+			rn_bw_params.wm_table = lpddr4_wm_table_with_disabled_ppt;
 		} else {
 			if (is_green_sardine)
-				rn_bw_params.wm_table = ddr4_wm_table_gs;
-			else {
-				if (ctx->dc->config.is_single_rank_dimm)
-					rn_bw_params.wm_table = ddr4_1R_wm_table_rn;
-				else
-					rn_bw_params.wm_table = ddr4_wm_table_rn;
-			}
+				rn_bw_params.wm_table = lpddr4_wm_table_gs;
+			else
+				rn_bw_params.wm_table = lpddr4_wm_table_rn;
+		}
+	} else {
+		if (is_green_sardine)
+			rn_bw_params.wm_table = ddr4_wm_table_gs;
+		else {
+			if (ctx->dc->config.is_single_rank_dimm)
+				rn_bw_params.wm_table = ddr4_1R_wm_table_rn;
+			else
+				rn_bw_params.wm_table = ddr4_wm_table_rn;
 		}
-		/* Saved clocks configured at boot for debug purposes */
-		rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info);
 	}
+	/* Saved clocks configured at boot for debug purposes */
+	rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info);
 
 	clk_mgr->base.dprefclk_khz = 600000;
 	dce_clock_read_ss_info(clk_mgr);
@@ -786,9 +780,8 @@ void rn_clk_mgr_construct(
 		}
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->smu_ver >= 0x00371500) {
-		/* enable powerfeatures when displaycount goes to 0 */
+	/* enable powerfeatures when displaycount goes to 0 */
+	if (clk_mgr->smu_ver >= 0x00371500)
 		rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn);
-	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
index 27fbe906682f9..8c9d45e5b13b8 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c
@@ -147,17 +147,14 @@ int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dis
 			VBIOSSMC_MSG_SetDispclkFreq,
 			khz_to_mhz_ceil(requested_dispclk_khz));
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
-			if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz)
-				dmcu->funcs->set_psr_wait_loop(dmcu,
-						actual_dispclk_set_mhz / 7);
-		}
+	if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) {
+		if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz)
+			dmcu->funcs->set_psr_wait_loop(dmcu,
+					actual_dispclk_set_mhz / 7);
 	}
 
 	// pmfw always set clock more than or equal requested clock
-	if (!IS_DIAG_DC(dc->ctx->dce_environment))
-		ASSERT(actual_dispclk_set_mhz >= khz_to_mhz_ceil(requested_dispclk_khz));
+	ASSERT(actual_dispclk_set_mhz >= khz_to_mhz_ceil(requested_dispclk_khz));
 
 	return actual_dispclk_set_mhz * 1000;
 }
@@ -221,15 +218,13 @@ void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phy
 int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz)
 {
 	int actual_dppclk_set_mhz = -1;
-	struct dc *dc = clk_mgr->base.ctx->dc;
 
 	actual_dppclk_set_mhz = rn_vbios_smu_send_msg_with_param(
 			clk_mgr,
 			VBIOSSMC_MSG_SetDppclkFreq,
 			khz_to_mhz_ceil(requested_dpp_khz));
 
-	if (!IS_DIAG_DC(dc->ctx->dce_environment))
-		ASSERT(actual_dppclk_set_mhz >= khz_to_mhz_ceil(requested_dpp_khz));
+	ASSERT(actual_dppclk_set_mhz >= khz_to_mhz_ceil(requested_dpp_khz));
 
 	return actual_dppclk_set_mhz * 1000;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
index 3908e7cfd6cbf..3271c8c7905dd 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c
@@ -521,6 +521,8 @@ void dcn3_clk_mgr_construct(
 		struct pp_smu_funcs *pp_smu,
 		struct dccg *dccg)
 {
+	struct clk_state_registers_and_bypass s = { 0 };
+
 	clk_mgr->base.ctx = ctx;
 	clk_mgr->base.funcs = &dcn3_funcs;
 	clk_mgr->regs = &clk_mgr_regs;
@@ -537,27 +539,19 @@ void dcn3_clk_mgr_construct(
 
 	clk_mgr->base.dprefclk_khz = 730000; // 700 MHz planned if VCO is 3.85 GHz, will be retrieved
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		clk_mgr->base.funcs  = &dcn3_fpga_funcs;
-		clk_mgr->base.dentist_vco_freq_khz = 3650000;
-
-	} else {
-		struct clk_state_registers_and_bypass s = { 0 };
+	/* integer part is now VCO frequency in kHz */
+	clk_mgr->base.dentist_vco_freq_khz = dcn30_get_vco_frequency_from_reg(clk_mgr);
 
-		/* integer part is now VCO frequency in kHz */
-		clk_mgr->base.dentist_vco_freq_khz = dcn30_get_vco_frequency_from_reg(clk_mgr);
-
-		/* in case we don't get a value from the register, use default */
-		if (clk_mgr->base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.dentist_vco_freq_khz = 3650000;
-		/* Convert dprefclk units from MHz to KHz */
-		/* Value already divided by 10, some resolution lost */
+	/* in case we don't get a value from the register, use default */
+	if (clk_mgr->base.dentist_vco_freq_khz == 0)
+		clk_mgr->base.dentist_vco_freq_khz = 3650000;
+	/* Convert dprefclk units from MHz to KHz */
+	/* Value already divided by 10, some resolution lost */
 
-		/*TODO: uncomment assert once dcn3_dump_clk_registers is implemented */
-		//ASSERT(s.dprefclk != 0);
-		if (s.dprefclk != 0)
-			clk_mgr->base.dprefclk_khz = s.dprefclk * 1000;
-	}
+	/*TODO: uncomment assert once dcn3_dump_clk_registers is implemented */
+	//ASSERT(s.dprefclk != 0);
+	if (s.dprefclk != 0)
+		clk_mgr->base.dprefclk_khz = s.dprefclk * 1000;
 
 	clk_mgr->dfs_bypass_enabled = false;
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
index 01383aac6b419..a5489fe6875f4 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
@@ -117,7 +117,7 @@ static void vg_update_clocks(struct clk_mgr *clk_mgr_base,
 
 			display_count = vg_get_active_display_cnt_wa(dc, context);
 			/* if we can go lower, go lower */
-			if (display_count == 0 && !IS_DIAG_DC(dc->ctx->dce_environment)) {
+			if (display_count == 0) {
 				union display_idle_optimization_u idle_info = { 0 };
 
 				idle_info.idle_info.df_request_disabled = 1;
@@ -151,10 +151,8 @@ static void vg_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	// workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (new_clocks->dppclk_khz < 100000)
-			new_clocks->dppclk_khz = 100000;
-	}
+	if (new_clocks->dppclk_khz < 100000)
+		new_clocks->dppclk_khz = 100000;
 
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
@@ -664,6 +662,7 @@ void vg_clk_mgr_construct(
 		struct dccg *dccg)
 {
 	struct smu_dpm_clks smu_dpm_clks = { 0 };
+	struct clk_log_info log_info = {0};
 
 	clk_mgr->base.base.ctx = ctx;
 	clk_mgr->base.base.funcs = &vg_funcs;
@@ -703,32 +702,25 @@ void vg_clk_mgr_construct(
 
 	ASSERT(smu_dpm_clks.dpm_clks);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		vg_funcs.update_clocks = dcn2_update_clocks_fpga;
-		clk_mgr->base.base.dentist_vco_freq_khz = 3600000;
-	} else {
-		struct clk_log_info log_info = {0};
+	clk_mgr->base.smu_ver = dcn301_smu_get_smu_version(&clk_mgr->base);
 
-		clk_mgr->base.smu_ver = dcn301_smu_get_smu_version(&clk_mgr->base);
+	if (clk_mgr->base.smu_ver)
+		clk_mgr->base.smu_present = true;
 
-		if (clk_mgr->base.smu_ver)
-			clk_mgr->base.smu_present = true;
+	/* TODO: Check we get what we expect during bringup */
+	clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
 
-		/* TODO: Check we get what we expect during bringup */
-		clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
-
-		/* in case we don't get a value from the register, use default */
-		if (clk_mgr->base.base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.base.dentist_vco_freq_khz = 3600000;
+	/* in case we don't get a value from the register, use default */
+	if (clk_mgr->base.base.dentist_vco_freq_khz == 0)
+		clk_mgr->base.base.dentist_vco_freq_khz = 3600000;
 
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
-			vg_bw_params.wm_table = lpddr5_wm_table;
-		} else {
-			vg_bw_params.wm_table = ddr4_wm_table;
-		}
-		/* Saved clocks configured at boot for debug purposes */
-		vg_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, &clk_mgr->base.base, &log_info);
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
+		vg_bw_params.wm_table = lpddr5_wm_table;
+	} else {
+		vg_bw_params.wm_table = ddr4_wm_table;
 	}
+	/* Saved clocks configured at boot for debug purposes */
+	vg_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, &clk_mgr->base.base, &log_info);
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	dce_clock_read_ss_info(&clk_mgr->base);
@@ -746,12 +738,6 @@ void vg_clk_mgr_construct(
 	if (smu_dpm_clks.dpm_clks && smu_dpm_clks.mc_address.quad_part != 0)
 		dm_helpers_free_gpu_mem(clk_mgr->base.base.ctx, DC_MEM_ALLOC_TYPE_FRAME_BUFFER,
 				smu_dpm_clks.dpm_clks);
-/*
-	if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->base.smu_ver) {
-		 enable powerfeatures when displaycount goes to 0
-		dcn301_smu_enable_phy_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn);
-	}
-*/
 }
 
 void vg_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr_int)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
index 3c743cd3d3f05..7ccd96959256d 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
@@ -205,10 +205,8 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	// workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (new_clocks->dppclk_khz < 100000)
-			new_clocks->dppclk_khz = 100000;
-	}
+	if (new_clocks->dppclk_khz < 100000)
+		new_clocks->dppclk_khz = 100000;
 
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
@@ -672,6 +670,7 @@ void dcn31_clk_mgr_construct(
 		struct dccg *dccg)
 {
 	struct dcn31_smu_dpm_clks smu_dpm_clks = { 0 };
+	struct clk_log_info log_info = {0};
 
 	clk_mgr->base.base.ctx = ctx;
 	clk_mgr->base.base.funcs = &dcn31_funcs;
@@ -711,29 +710,22 @@ void dcn31_clk_mgr_construct(
 
 	ASSERT(smu_dpm_clks.dpm_clks);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		clk_mgr->base.base.funcs = &dcn3_fpga_funcs;
-	} else {
-		struct clk_log_info log_info = {0};
-
-		clk_mgr->base.smu_ver = dcn31_smu_get_smu_version(&clk_mgr->base);
+	clk_mgr->base.smu_ver = dcn31_smu_get_smu_version(&clk_mgr->base);
 
-		if (clk_mgr->base.smu_ver)
-			clk_mgr->base.smu_present = true;
+	if (clk_mgr->base.smu_ver)
+		clk_mgr->base.smu_present = true;
 
-		/* TODO: Check we get what we expect during bringup */
-		clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
-
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
-			dcn31_bw_params.wm_table = lpddr5_wm_table;
-		} else {
-			dcn31_bw_params.wm_table = ddr5_wm_table;
-		}
-		/* Saved clocks configured at boot for debug purposes */
-		dcn31_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
-					 &clk_mgr->base.base, &log_info);
+	/* TODO: Check we get what we expect during bringup */
+	clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
 
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
+		dcn31_bw_params.wm_table = lpddr5_wm_table;
+	} else {
+		dcn31_bw_params.wm_table = ddr5_wm_table;
 	}
+	/* Saved clocks configured at boot for debug purposes */
+	dcn31_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
+				 &clk_mgr->base.base, &log_info);
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	clk_mgr->base.base.clks.ref_dtbclk_khz = 600000;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 2a0c696f58619..2f7c8996b19d3 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -241,10 +241,8 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	// workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (new_clocks->dppclk_khz < 100000)
-			new_clocks->dppclk_khz = 100000;
-	}
+	if (new_clocks->dppclk_khz < 100000)
+		new_clocks->dppclk_khz = 100000;
 
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
@@ -724,6 +722,7 @@ void dcn314_clk_mgr_construct(
 		struct dccg *dccg)
 {
 	struct dcn314_smu_dpm_clks smu_dpm_clks = { 0 };
+	struct clk_log_info log_info = {0};
 
 	clk_mgr->base.base.ctx = ctx;
 	clk_mgr->base.base.funcs = &dcn314_funcs;
@@ -763,29 +762,22 @@ void dcn314_clk_mgr_construct(
 
 	ASSERT(smu_dpm_clks.dpm_clks);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		clk_mgr->base.base.funcs = &dcn3_fpga_funcs;
-	} else {
-		struct clk_log_info log_info = {0};
-
-		clk_mgr->base.smu_ver = dcn314_smu_get_smu_version(&clk_mgr->base);
+	clk_mgr->base.smu_ver = dcn314_smu_get_smu_version(&clk_mgr->base);
 
-		if (clk_mgr->base.smu_ver)
-			clk_mgr->base.smu_present = true;
+	if (clk_mgr->base.smu_ver)
+		clk_mgr->base.smu_present = true;
 
-		/* TODO: Check we get what we expect during bringup */
-		clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
+	/* TODO: Check we get what we expect during bringup */
+	clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
 
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType)
-			dcn314_bw_params.wm_table = lpddr5_wm_table;
-		else
-			dcn314_bw_params.wm_table = ddr5_wm_table;
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType)
+		dcn314_bw_params.wm_table = lpddr5_wm_table;
+	else
+		dcn314_bw_params.wm_table = ddr5_wm_table;
 
-		/* Saved clocks configured at boot for debug purposes */
-		dcn314_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
-					  &clk_mgr->base.base, &log_info);
-
-	}
+	/* Saved clocks configured at boot for debug purposes */
+	dcn314_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
+				  &clk_mgr->base.base, &log_info);
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	clk_mgr->base.base.clks.ref_dtbclk_khz = 600000;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
index 300c6a5872d0b..b2c4f97afc8b4 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
@@ -184,12 +184,10 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	// workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (new_clocks->dppclk_khz < MIN_DPP_DISP_CLK)
-			new_clocks->dppclk_khz = MIN_DPP_DISP_CLK;
-		if (new_clocks->dispclk_khz < MIN_DPP_DISP_CLK)
-			new_clocks->dispclk_khz = MIN_DPP_DISP_CLK;
-	}
+	if (new_clocks->dppclk_khz < MIN_DPP_DISP_CLK)
+		new_clocks->dppclk_khz = MIN_DPP_DISP_CLK;
+	if (new_clocks->dispclk_khz < MIN_DPP_DISP_CLK)
+		new_clocks->dispclk_khz = MIN_DPP_DISP_CLK;
 
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
@@ -600,6 +598,7 @@ void dcn315_clk_mgr_construct(
 		struct dccg *dccg)
 {
 	struct dcn315_smu_dpm_clks smu_dpm_clks = { 0 };
+	struct clk_log_info log_info = {0};
 
 	clk_mgr->base.base.ctx = ctx;
 	clk_mgr->base.base.funcs = &dcn315_funcs;
@@ -639,26 +638,19 @@ void dcn315_clk_mgr_construct(
 
 	ASSERT(smu_dpm_clks.dpm_clks);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		clk_mgr->base.base.funcs = &dcn3_fpga_funcs;
-	} else {
-		struct clk_log_info log_info = {0};
-
-		clk_mgr->base.smu_ver = dcn315_smu_get_smu_version(&clk_mgr->base);
+	clk_mgr->base.smu_ver = dcn315_smu_get_smu_version(&clk_mgr->base);
 
-		if (clk_mgr->base.smu_ver > 0)
-			clk_mgr->base.smu_present = true;
-
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
-			dcn315_bw_params.wm_table = lpddr5_wm_table;
-		} else {
-			dcn315_bw_params.wm_table = ddr5_wm_table;
-		}
-		/* Saved clocks configured at boot for debug purposes */
-		dcn315_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
-					  &clk_mgr->base.base, &log_info);
+	if (clk_mgr->base.smu_ver > 0)
+		clk_mgr->base.smu_present = true;
 
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
+		dcn315_bw_params.wm_table = lpddr5_wm_table;
+	} else {
+		dcn315_bw_params.wm_table = ddr5_wm_table;
 	}
+	/* Saved clocks configured at boot for debug purposes */
+	dcn315_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
+				  &clk_mgr->base.base, &log_info);
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	clk_mgr->base.base.dprefclk_khz = dcn315_smu_get_dpref_clk(&clk_mgr->base);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
index 538126cefd4d4..d7de756301cf7 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
@@ -207,12 +207,10 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	// workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (new_clocks->dppclk_khz < 100000)
-			new_clocks->dppclk_khz = 100000;
-		if (new_clocks->dispclk_khz < 100000)
-			new_clocks->dispclk_khz = 100000;
-	}
+	if (new_clocks->dppclk_khz < 100000)
+		new_clocks->dppclk_khz = 100000;
+	if (new_clocks->dispclk_khz < 100000)
+		new_clocks->dispclk_khz = 100000;
 
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
@@ -616,6 +614,7 @@ void dcn316_clk_mgr_construct(
 		struct dccg *dccg)
 {
 	struct dcn316_smu_dpm_clks smu_dpm_clks = { 0 };
+	struct clk_log_info log_info = {0};
 
 	clk_mgr->base.base.ctx = ctx;
 	clk_mgr->base.base.funcs = &dcn316_funcs;
@@ -655,35 +654,27 @@ void dcn316_clk_mgr_construct(
 
 	ASSERT(smu_dpm_clks.dpm_clks);
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		clk_mgr->base.base.funcs = &dcn3_fpga_funcs;
-		clk_mgr->base.base.dentist_vco_freq_khz = 2500000;
-	} else {
-		struct clk_log_info log_info = {0};
-
-		clk_mgr->base.smu_ver = dcn316_smu_get_smu_version(&clk_mgr->base);
+	clk_mgr->base.smu_ver = dcn316_smu_get_smu_version(&clk_mgr->base);
 
-		if (clk_mgr->base.smu_ver > 0)
-			clk_mgr->base.smu_present = true;
+	if (clk_mgr->base.smu_ver > 0)
+		clk_mgr->base.smu_present = true;
 
-		// Skip this for now as it did not work on DCN315, renable during bring up
-		clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
+	// Skip this for now as it did not work on DCN315, renable during bring up
+	clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base);
 
-		/* in case we don't get a value from the register, use default */
-		if (clk_mgr->base.base.dentist_vco_freq_khz == 0)
-			clk_mgr->base.base.dentist_vco_freq_khz = 2500000; /* 2400MHz */
+	/* in case we don't get a value from the register, use default */
+	if (clk_mgr->base.base.dentist_vco_freq_khz == 0)
+		clk_mgr->base.base.dentist_vco_freq_khz = 2500000; /* 2400MHz */
 
 
-		if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
-			dcn316_bw_params.wm_table = lpddr5_wm_table;
-		} else {
-			dcn316_bw_params.wm_table = ddr4_wm_table;
-		}
-		/* Saved clocks configured at boot for debug purposes */
-		dcn316_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
-					  &clk_mgr->base.base, &log_info);
-
+	if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) {
+		dcn316_bw_params.wm_table = lpddr5_wm_table;
+	} else {
+		dcn316_bw_params.wm_table = ddr4_wm_table;
 	}
+	/* Saved clocks configured at boot for debug purposes */
+	dcn316_dump_clk_registers(&clk_mgr->base.base.boot_snapshot,
+				  &clk_mgr->base.base, &log_info);
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	clk_mgr->base.base.dprefclk_khz = dcn316_smu_get_dpref_clk(&clk_mgr->base);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 1df623b298a92..20bff6a346baa 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -941,7 +941,6 @@ void dcn32_clk_mgr_construct(
 			clk_mgr->base.clks.ref_dtbclk_khz = 268750;
 	}
 
-
 	/* integer part is now VCO frequency in kHz */
 	clk_mgr->base.dentist_vco_freq_khz = dcn32_get_vco_frequency_from_reg(clk_mgr);
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 9f8efee277210..af929ee11af52 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3588,29 +3588,27 @@ static void commit_planes_for_stream(struct dc *dc,
 		}
 	}
 
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		for (i = 0; i < surface_count; i++) {
-			struct dc_plane_state *plane_state = srf_updates[i].surface;
-			/*set logical flag for lock/unlock use*/
-			for (j = 0; j < dc->res_pool->pipe_count; j++) {
-				struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
-				if (!pipe_ctx->plane_state)
-					continue;
-				if (should_update_pipe_for_plane(context, pipe_ctx, plane_state))
-					continue;
-				pipe_ctx->plane_state->triplebuffer_flips = false;
-				if (update_type == UPDATE_TYPE_FAST &&
-					dc->hwss.program_triplebuffer != NULL &&
-					!pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) {
-						/*triple buffer for VUpdate  only*/
-						pipe_ctx->plane_state->triplebuffer_flips = true;
-				}
-			}
-			if (update_type == UPDATE_TYPE_FULL) {
-				/* force vsync flip when reconfiguring pipes to prevent underflow */
-				plane_state->flip_immediate = false;
+	for (i = 0; i < surface_count; i++) {
+		struct dc_plane_state *plane_state = srf_updates[i].surface;
+		/*set logical flag for lock/unlock use*/
+		for (j = 0; j < dc->res_pool->pipe_count; j++) {
+			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+			if (!pipe_ctx->plane_state)
+				continue;
+			if (should_update_pipe_for_plane(context, pipe_ctx, plane_state))
+				continue;
+			pipe_ctx->plane_state->triplebuffer_flips = false;
+			if (update_type == UPDATE_TYPE_FAST &&
+				dc->hwss.program_triplebuffer != NULL &&
+				!pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) {
+					/*triple buffer for VUpdate  only*/
+					pipe_ctx->plane_state->triplebuffer_flips = true;
 			}
 		}
+		if (update_type == UPDATE_TYPE_FULL) {
+			/* force vsync flip when reconfiguring pipes to prevent underflow */
+			plane_state->flip_immediate = false;
+		}
 	}
 
 	// Update Type FULL, Surface updates
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 72b261ad95870..0d3ec50b13851 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -490,25 +490,6 @@ bool dc_stream_add_writeback(struct dc *dc,
 		struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst];
 		dwb->otg_inst = stream_status->primary_otg_inst;
 	}
-	if (IS_DIAG_DC(dc->ctx->dce_environment)) {
-		if (!dc->hwss.update_bandwidth(dc, dc->current_state)) {
-			dm_error("DC: update_bandwidth failed!\n");
-			return false;
-		}
-
-		/* enable writeback */
-		if (dc->hwss.enable_writeback) {
-			struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst];
-
-			if (dwb->funcs->is_enabled(dwb)) {
-				/* writeback pipe already enabled, only need to update */
-				dc->hwss.update_writeback(dc, wb_info, dc->current_state);
-			} else {
-				/* Enable writeback pipe from scratch*/
-				dc->hwss.enable_writeback(dc, wb_info, dc->current_state);
-			}
-		}
-	}
 	return true;
 }
 
@@ -553,17 +534,6 @@ bool dc_stream_remove_writeback(struct dc *dc,
 	}
 	stream->num_wb_info = j;
 
-	if (IS_DIAG_DC(dc->ctx->dce_environment)) {
-		/* recalculate and apply DML parameters */
-		if (!dc->hwss.update_bandwidth(dc, dc->current_state)) {
-			dm_error("DC: update_bandwidth failed!\n");
-			return false;
-		}
-
-		/* disable writeback */
-		if (dc->hwss.disable_writeback)
-			dc->hwss.disable_writeback(dc, dwb_pipe_inst);
-	}
 	return true;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 3db7a2e11af52..3907eeff560ce 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -464,8 +464,7 @@ void generic_reg_wait(const struct dc_context *ctx,
 		field_value = get_reg_field_value_ex(reg_val, mask, shift);
 
 		if (field_value == condition_value) {
-			if (i * delay_between_poll_us > 1000 &&
-					!IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
+			if (i * delay_between_poll_us > 1000)
 				DC_LOG_DC("REG_WAIT taking a while: %dms in %s line:%d\n",
 						delay_between_poll_us * i / 1000,
 						func_name, line);
@@ -477,8 +476,7 @@ void generic_reg_wait(const struct dc_context *ctx,
 			delay_between_poll_us, time_out_num_tries,
 			func_name, line);
 
-	if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-		BREAK_TO_DEBUGGER();
+	BREAK_TO_DEBUGGER();
 }
 
 void generic_write_indirect_reg(const struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 150c19286d67f..6b4731b5e975f 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -69,13 +69,6 @@ enum dce_environment {
 	DCE_ENV_VIRTUAL_HW
 };
 
-/* Note: use these macro definitions instead of direct comparison! */
-#define IS_FPGA_MAXIMUS_DC(dce_environment) \
-	(dce_environment == DCE_ENV_FPGA_MAXIMUS)
-
-#define IS_DIAG_DC(dce_environment) \
-	(IS_FPGA_MAXIMUS_DC(dce_environment) || (dce_environment == DCE_ENV_DIAG))
-
 struct dc_perf_trace {
 	unsigned long read_count;
 	unsigned long write_count;
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
index 462c7a3ec3ccb..ed8936405dfa5 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c
@@ -920,25 +920,6 @@ static bool dce112_program_pix_clk(
 	struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source);
 	struct bp_pixel_clock_parameters bp_pc_params = {0};
 
-	if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) {
-		unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0;
-		unsigned dp_dto_ref_100hz = 7000000;
-		unsigned clock_100hz = pll_settings->actual_pix_clk_100hz;
-
-		/* Set DTO values: phase = target clock, modulo = reference clock */
-		REG_WRITE(PHASE[inst], clock_100hz);
-		REG_WRITE(MODULO[inst], dp_dto_ref_100hz);
-
-		/* Enable DTO */
-		if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL)
-			REG_UPDATE_2(PIXEL_RATE_CNTL[inst],
-					DP_DTO0_ENABLE, 1,
-					PIPE0_DTO_SRC_SEL, 1);
-		else
-			REG_UPDATE(PIXEL_RATE_CNTL[inst],
-					DP_DTO0_ENABLE, 1);
-		return true;
-	}
 	/* First disable SS
 	 * ATOMBIOS will enable by default SS on PLL for DP,
 	 * do not disable it here
@@ -1015,25 +996,6 @@ static bool dcn31_program_pix_clk(
 			REG_UPDATE(PIXEL_RATE_CNTL[inst],
 					DP_DTO0_ENABLE, 1);
 	} else {
-		if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) {
-			unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0;
-			unsigned dp_dto_ref_100hz = 7000000;
-			unsigned clock_100hz = pll_settings->actual_pix_clk_100hz;
-
-			/* Set DTO values: phase = target clock, modulo = reference clock */
-			REG_WRITE(PHASE[inst], clock_100hz);
-			REG_WRITE(MODULO[inst], dp_dto_ref_100hz);
-
-			/* Enable DTO */
-			if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL)
-				REG_UPDATE_2(PIXEL_RATE_CNTL[inst],
-						DP_DTO0_ENABLE, 1,
-						PIPE0_DTO_SRC_SEL, 1);
-			else
-				REG_UPDATE(PIXEL_RATE_CNTL[inst],
-						DP_DTO0_ENABLE, 1);
-			return true;
-		}
 
 		if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL)
 			REG_UPDATE(PIXEL_RATE_CNTL[inst],
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
index e74266cc00986..63009db8b5a72 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c
@@ -1093,11 +1093,9 @@ static void dcn21_dmcu_construct(
 
 	dce_dmcu_construct(dmcu_dce, ctx, regs, dmcu_shift, dmcu_mask);
 
-	if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) {
-		psp_version = dm_read_reg(ctx, mmMP0_SMN_C2PMSG_58);
-		dmcu_dce->base.auto_load_dmcu = ((psp_version & 0x00FF00FF) > 0x00110029);
-		dmcu_dce->base.psp_version = psp_version;
-	}
+	psp_version = dm_read_reg(ctx, mmMP0_SMN_C2PMSG_58);
+	dmcu_dce->base.auto_load_dmcu = ((psp_version & 0x00FF00FF) > 0x00110029);
+	dmcu_dce->base.psp_version = psp_version;
 }
 
 struct dmcu *dce_dmcu_create(
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index 1a0be40d125c6..c6fe2c00aedb6 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -209,9 +209,6 @@ static bool dce110_enable_display_power_gating(
 	struct dc_context *ctx = dc->ctx;
 	unsigned int underlay_idx = dc->res_pool->underlay_pipe_index;
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-		return true;
-
 	if (power_gating == PIPE_GATING_CONTROL_INIT)
 		cntl = ASIC_PIPE_INIT;
 	else if (power_gating == PIPE_GATING_CONTROL_ENABLE)
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c
index 19873ee1f78dd..690caaaff0194 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c
@@ -120,9 +120,6 @@ static bool dce112_enable_display_power_gating(
 	enum bp_pipe_control_action cntl;
 	struct dc_context *ctx = dc->ctx;
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-		return true;
-
 	if (power_gating == PIPE_GATING_CONTROL_INIT)
 		cntl = ASIC_PIPE_INIT;
 	else if (power_gating == PIPE_GATING_CONTROL_ENABLE)
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
index d4afe6c824d2c..45e08c4d58618 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c
@@ -159,9 +159,6 @@ static bool dce120_enable_display_power_gating(
 	enum bp_pipe_control_action cntl;
 	struct dc_context *ctx = dc->ctx;
 
-	if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-		return true;
-
 	if (power_gating == PIPE_GATING_CONTROL_INIT)
 		cntl = ASIC_PIPE_INIT;
 	else if (power_gating == PIPE_GATING_CONTROL_ENABLE)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 905246a2ece45..13b4e5118459b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -1012,31 +1012,29 @@ static void dcn10_reset_back_end_for_pipe(
 		return;
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		link = pipe_ctx->stream->link;
-		/* DPMS may already disable or */
-		/* dpms_off status is incorrect due to fastboot
-		 * feature. When system resume from S4 with second
-		 * screen only, the dpms_off would be true but
-		 * VBIOS lit up eDP, so check link status too.
-		 */
-		if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
-			dc->link_srv->set_dpms_off(pipe_ctx);
-		else if (pipe_ctx->stream_res.audio)
-			dc->hwss.disable_audio_stream(pipe_ctx);
-
-		if (pipe_ctx->stream_res.audio) {
-			/*disable az_endpoint*/
-			pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
-
-			/*free audio*/
-			if (dc->caps.dynamic_audio == true) {
-				/*we have to dynamic arbitrate the audio endpoints*/
-				/*we free the resource, need reset is_audio_acquired*/
-				update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
-						pipe_ctx->stream_res.audio, false);
-				pipe_ctx->stream_res.audio = NULL;
-			}
+	link = pipe_ctx->stream->link;
+	/* DPMS may already disable or */
+	/* dpms_off status is incorrect due to fastboot
+	 * feature. When system resume from S4 with second
+	 * screen only, the dpms_off would be true but
+	 * VBIOS lit up eDP, so check link status too.
+	 */
+	if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
+		dc->link_srv->set_dpms_off(pipe_ctx);
+	else if (pipe_ctx->stream_res.audio)
+		dc->hwss.disable_audio_stream(pipe_ctx);
+
+	if (pipe_ctx->stream_res.audio) {
+		/*disable az_endpoint*/
+		pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
+
+		/*free audio*/
+		if (dc->caps.dynamic_audio == true) {
+			/*we have to dynamic arbitrate the audio endpoints*/
+			/*we free the resource, need reset is_audio_acquired*/
+			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
+					pipe_ctx->stream_res.audio, false);
+			pipe_ctx->stream_res.audio = NULL;
 		}
 	}
 
@@ -1499,28 +1497,6 @@ void dcn10_init_hw(struct dc *dc)
 	if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->dccg_init)
 		dc->res_pool->dccg->funcs->dccg_init(res_pool->dccg);
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-
-		REG_WRITE(REFCLK_CNTL, 0);
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
-		REG_WRITE(DIO_MEM_PWR_CTRL, 0);
-
-		if (!dc->debug.disable_clock_gate) {
-			/* enable all DCN clock gating */
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
-
-			REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
-		}
-
-		//Enable ability to power gate / don't force power on permanently
-		if (hws->funcs.enable_power_gating_plane)
-			hws->funcs.enable_power_gating_plane(hws, true);
-
-		return;
-	}
-
 	if (!dcb->funcs->is_accelerated_mode(dcb))
 		hws->funcs.disable_vga(dc->hwseq);
 
@@ -1532,23 +1508,21 @@ void dcn10_init_hw(struct dc *dc)
 		res_pool->ref_clocks.xtalin_clock_inKhz =
 				dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency;
 
-		if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-			if (res_pool->dccg && res_pool->hubbub) {
+		if (res_pool->dccg && res_pool->hubbub) {
 
-				(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
-						dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
-						&res_pool->ref_clocks.dccg_ref_clock_inKhz);
+			(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
+					dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
+					&res_pool->ref_clocks.dccg_ref_clock_inKhz);
 
-				(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
-						res_pool->ref_clocks.dccg_ref_clock_inKhz,
-						&res_pool->ref_clocks.dchub_ref_clock_inKhz);
-			} else {
-				// Not all ASICs have DCCG sw component
-				res_pool->ref_clocks.dccg_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-				res_pool->ref_clocks.dchub_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-			}
+			(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
+					res_pool->ref_clocks.dccg_ref_clock_inKhz,
+					&res_pool->ref_clocks.dchub_ref_clock_inKhz);
+		} else {
+			// Not all ASICs have DCCG sw component
+			res_pool->ref_clocks.dccg_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
+			res_pool->ref_clocks.dchub_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
 		}
 	} else
 		ASSERT_CRITICAL(false);
@@ -3070,15 +3044,13 @@ void dcn10_prepare_bandwidth(
 	if (dc->debug.sanity_checks)
 		hws->funcs.verify_allow_pstate_change_high(dc);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (context->stream_count == 0)
-			context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
+	if (context->stream_count == 0)
+		context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
 
-		dc->clk_mgr->funcs->update_clocks(
-				dc->clk_mgr,
-				context,
-				false);
-	}
+	dc->clk_mgr->funcs->update_clocks(
+			dc->clk_mgr,
+			context,
+			false);
 
 	dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub,
 			&context->bw_ctx.bw.dcn.watermarks,
@@ -3110,15 +3082,13 @@ void dcn10_optimize_bandwidth(
 	if (dc->debug.sanity_checks)
 		hws->funcs.verify_allow_pstate_change_high(dc);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		if (context->stream_count == 0)
-			context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
+	if (context->stream_count == 0)
+		context->bw_ctx.bw.dcn.clk.phyclk_khz = 0;
 
-		dc->clk_mgr->funcs->update_clocks(
-				dc->clk_mgr,
-				context,
-				true);
-	}
+	dc->clk_mgr->funcs->update_clocks(
+			dc->clk_mgr,
+			context,
+			true);
 
 	hubbub->funcs->program_watermarks(hubbub,
 			&context->bw_ctx.bw.dcn.watermarks,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index c9e53dc49c92b..e1975991e075d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -653,11 +653,9 @@ void optc1_lock(struct timing_generator *optc)
 	REG_SET(OTG_MASTER_UPDATE_LOCK, 0,
 			OTG_MASTER_UPDATE_LOCK, 1);
 
-	/* Should be fast, status does not update on maximus */
-	if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS)
-		REG_WAIT(OTG_MASTER_UPDATE_LOCK,
-				UPDATE_LOCK_STATUS, 1,
-				1, 10);
+	REG_WAIT(OTG_MASTER_UPDATE_LOCK,
+			UPDATE_LOCK_STATUS, 1,
+			1, 10);
 
 	TRACE_OPTC_LOCK_UNLOCK_STATE(optc1, optc->inst, true);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index 21ec1ba5ed756..a0625209c86d6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -1651,9 +1651,8 @@ static bool dcn10_resource_construct(
 	}
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto fail;
+			&res_create_funcs))
+		goto fail;
 
 	dcn10_hw_sequencer_construct(dc);
 	dc->caps.max_planes =  pool->base.pipe_count;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index e74c3ce561ab6..b0a13eb8318cd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2480,36 +2480,31 @@ static void dcn20_reset_back_end_for_pipe(
 		return;
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		/* DPMS may already disable or */
-		/* dpms_off status is incorrect due to fastboot
-		 * feature. When system resume from S4 with second
-		 * screen only, the dpms_off would be true but
-		 * VBIOS lit up eDP, so check link status too.
-		 */
-		if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
-			dc->link_srv->set_dpms_off(pipe_ctx);
-		else if (pipe_ctx->stream_res.audio)
-			dc->hwss.disable_audio_stream(pipe_ctx);
-
-		/* free acquired resources */
-		if (pipe_ctx->stream_res.audio) {
-			/*disable az_endpoint*/
-			pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
-
-			/*free audio*/
-			if (dc->caps.dynamic_audio == true) {
-				/*we have to dynamic arbitrate the audio endpoints*/
-				/*we free the resource, need reset is_audio_acquired*/
-				update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
-						pipe_ctx->stream_res.audio, false);
-				pipe_ctx->stream_res.audio = NULL;
-			}
+	/* DPMS may already disable or */
+	/* dpms_off status is incorrect due to fastboot
+	 * feature. When system resume from S4 with second
+	 * screen only, the dpms_off would be true but
+	 * VBIOS lit up eDP, so check link status too.
+	 */
+	if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
+		dc->link_srv->set_dpms_off(pipe_ctx);
+	else if (pipe_ctx->stream_res.audio)
+		dc->hwss.disable_audio_stream(pipe_ctx);
+
+	/* free acquired resources */
+	if (pipe_ctx->stream_res.audio) {
+		/*disable az_endpoint*/
+		pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
+
+		/*free audio*/
+		if (dc->caps.dynamic_audio == true) {
+			/*we have to dynamic arbitrate the audio endpoints*/
+			/*we free the resource, need reset is_audio_acquired*/
+			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
+					pipe_ctx->stream_res.audio, false);
+			pipe_ctx->stream_res.audio = NULL;
 		}
 	}
-	else if (pipe_ctx->stream_res.dsc) {
-		dc->link_srv->set_dsc_enable(pipe_ctx, false);
-	}
 
 	/* by upper caller loop, parent pipe: pipe0, will be reset last.
 	 * back end share by all pipes and will be disable only when disable
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
index 4192c522e59aa..e4b44e691ce66 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c
@@ -145,8 +145,4 @@ void dcn20_hw_sequencer_construct(struct dc *dc)
 	dc->hwss = dcn20_funcs;
 	dc->hwseq->funcs = dcn20_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
index a08c335b73838..e0edc163d7677 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
@@ -391,10 +391,9 @@ void optc2_triplebuffer_lock(struct timing_generator *optc)
 	REG_SET(OTG_MASTER_UPDATE_LOCK, 0,
 		OTG_MASTER_UPDATE_LOCK, 1);
 
-	if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS)
-		REG_WAIT(OTG_MASTER_UPDATE_LOCK,
-				UPDATE_LOCK_STATUS, 1,
-				1, 10);
+	REG_WAIT(OTG_MASTER_UPDATE_LOCK,
+			UPDATE_LOCK_STATUS, 1,
+			1, 10);
 }
 
 void optc2_triplebuffer_unlock(struct timing_generator *optc)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index 1d8c5805ef20c..7dcae3183e078 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -2488,15 +2488,9 @@ static bool dcn20_resource_construct(
 
 	dc->caps.dp_hdmi21_pcon_support = true;
 
-	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) {
+	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	} else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		pool->base.pipe_count = 4;
-		pool->base.mpcc_count = pool->base.pipe_count;
-		dc->debug = debug_defaults_diags;
-	} else {
-		dc->debug = debug_defaults_diags;
-	}
+
 	//dcn2.0x
 	dc->work_arounds.dedcn20_305_wa = true;
 
@@ -2734,9 +2728,8 @@ static bool dcn20_resource_construct(
 	}
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	dcn20_hw_sequencer_construct(dc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
index 75472d53ff52f..9e027db6d752c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c
@@ -231,52 +231,39 @@ void dcn201_init_hw(struct dc *dc)
 	if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks)
 		dc->clk_mgr->funcs->init_clocks(dc->clk_mgr);
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF);
-		REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF);
-
-		hws->funcs.dccg_init(hws);
-
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2);
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
-		REG_WRITE(REFCLK_CNTL, 0);
-	} else {
-		hws->funcs.bios_golden_init(dc);
-
-		if (dc->ctx->dc_bios->fw_info_valid) {
-			res_pool->ref_clocks.xtalin_clock_inKhz =
-				dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency;
-
-			if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-				if (res_pool->dccg && res_pool->hubbub) {
-					(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
-							dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
-							&res_pool->ref_clocks.dccg_ref_clock_inKhz);
-
-					(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
-							res_pool->ref_clocks.dccg_ref_clock_inKhz,
-							&res_pool->ref_clocks.dchub_ref_clock_inKhz);
-				} else {
-					res_pool->ref_clocks.dccg_ref_clock_inKhz =
-							res_pool->ref_clocks.xtalin_clock_inKhz;
-					res_pool->ref_clocks.dchub_ref_clock_inKhz =
-							res_pool->ref_clocks.xtalin_clock_inKhz;
-				}
-			}
-		} else
-			ASSERT_CRITICAL(false);
-		for (i = 0; i < dc->link_count; i++) {
-			/* Power up AND update implementation according to the
-			 * required signal (which may be different from the
-			 * default signal on connector).
-			 */
-			struct dc_link *link = dc->links[i];
-
-			link->link_enc->funcs->hw_init(link->link_enc);
+	hws->funcs.bios_golden_init(dc);
+
+	if (dc->ctx->dc_bios->fw_info_valid) {
+		res_pool->ref_clocks.xtalin_clock_inKhz =
+			dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency;
+
+		if (res_pool->dccg && res_pool->hubbub) {
+			(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
+					dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
+					&res_pool->ref_clocks.dccg_ref_clock_inKhz);
+
+			(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
+					res_pool->ref_clocks.dccg_ref_clock_inKhz,
+					&res_pool->ref_clocks.dchub_ref_clock_inKhz);
+		} else {
+			res_pool->ref_clocks.dccg_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
+			res_pool->ref_clocks.dchub_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
 		}
-		if (hws->fb_offset.quad_part == 0)
-			read_mmhub_vm_setup(hws);
+	} else
+		ASSERT_CRITICAL(false);
+	for (i = 0; i < dc->link_count; i++) {
+		/* Power up AND update implementation according to the
+		 * required signal (which may be different from the
+		 * default signal on connector).
+		 */
+		struct dc_link *link = dc->links[i];
+
+		link->link_enc->funcs->hw_init(link->link_enc);
 	}
+	if (hws->fb_offset.quad_part == 0)
+		read_mmhub_vm_setup(hws);
 
 	/* Blank pixel data with OPP DPG */
 	for (i = 0; i < res_pool->timing_generator_count; i++) {
@@ -362,10 +349,6 @@ void dcn201_init_hw(struct dc *dc)
 		tg->funcs->tg_init(tg);
 	}
 
-	/* end of FPGA. Below if real ASIC */
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		return;
-
 	for (i = 0; i < res_pool->audio_count; i++) {
 		struct audio *audio = res_pool->audios[i];
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c
index 730875dfd8b44..70fcbec03fb6b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c
@@ -55,10 +55,9 @@ static void optc201_triplebuffer_lock(struct timing_generator *optc)
 	REG_SET(OTG_MASTER_UPDATE_LOCK, 0,
 		OTG_MASTER_UPDATE_LOCK, 1);
 
-	if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS)
-		REG_WAIT(OTG_MASTER_UPDATE_LOCK,
-				UPDATE_LOCK_STATUS, 1,
-				1, 10);
+	REG_WAIT(OTG_MASTER_UPDATE_LOCK,
+			UPDATE_LOCK_STATUS, 1,
+			1, 10);
 }
 
 static void optc201_triplebuffer_unlock(struct timing_generator *optc)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
index 6ea70da28aaaf..212c475d95cbc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
@@ -1272,9 +1272,8 @@ static bool dcn201_resource_construct(
 	}
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	dcn201_hw_sequencer_construct(dc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
index 8b58ce1db0352..f024157bd6eb6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c
@@ -151,8 +151,4 @@ void dcn21_hw_sequencer_construct(struct dc *dc)
 	dc->hwss = dcn21_funcs;
 	dc->hwseq->funcs = dcn21_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index 19aaa557b2db0..76268a7b7934d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -1503,11 +1503,6 @@ static bool dcn21_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		pool->base.pipe_count = 4;
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
 
 	// Init the vm_helper
 	if (dc->vm_helper)
@@ -1721,9 +1716,8 @@ static bool dcn21_resource_construct(
 	}
 
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	dcn21_hw_sequencer_construct(dc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index 3303c9aae0686..ad4e1023b00e2 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -330,10 +330,6 @@ void dcn30_enable_writeback(
 	DC_LOG_DWB("%s dwb_pipe_inst = %d, mpcc_inst = %d",\
 		__func__, wb_info->dwb_pipe_inst,\
 		wb_info->mpcc_inst);
-	if (IS_DIAG_DC(dc->ctx->dce_environment)) {
-		/*till diags switch to warmup interface*/
-		dcn30_mmhubbub_warmup(dc, 1, wb_info);
-	}
 	/* Update writeback pipe */
 	dcn30_set_writeback(dc, wb_info, context);
 
@@ -447,28 +443,6 @@ void dcn30_init_hw(struct dc *dc)
 	if (res_pool->dccg->funcs->dccg_init)
 		res_pool->dccg->funcs->dccg_init(res_pool->dccg);
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-
-		REG_WRITE(REFCLK_CNTL, 0);
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
-		REG_WRITE(DIO_MEM_PWR_CTRL, 0);
-
-		if (!dc->debug.disable_clock_gate) {
-			/* enable all DCN clock gating */
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
-
-			REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
-		}
-
-		//Enable ability to power gate / don't force power on permanently
-		if (hws->funcs.enable_power_gating_plane)
-			hws->funcs.enable_power_gating_plane(hws, true);
-
-		return;
-	}
-
 	if (!dcb->funcs->is_accelerated_mode(dcb)) {
 		hws->funcs.bios_golden_init(dc);
 		hws->funcs.disable_vga(dc->hwseq);
@@ -491,23 +465,21 @@ void dcn30_init_hw(struct dc *dc)
 		res_pool->ref_clocks.xtalin_clock_inKhz =
 				dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency;
 
-		if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-			if (res_pool->dccg && res_pool->hubbub) {
+		if (res_pool->dccg && res_pool->hubbub) {
 
-				(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
-						dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
-						&res_pool->ref_clocks.dccg_ref_clock_inKhz);
+			(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
+					dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
+					&res_pool->ref_clocks.dccg_ref_clock_inKhz);
 
-				(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
-						res_pool->ref_clocks.dccg_ref_clock_inKhz,
-						&res_pool->ref_clocks.dchub_ref_clock_inKhz);
-			} else {
-				// Not all ASICs have DCCG sw component
-				res_pool->ref_clocks.dccg_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-				res_pool->ref_clocks.dchub_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-			}
+			(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
+					res_pool->ref_clocks.dccg_ref_clock_inKhz,
+					&res_pool->ref_clocks.dchub_ref_clock_inKhz);
+		} else {
+			// Not all ASICs have DCCG sw component
+			res_pool->ref_clocks.dccg_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
+			res_pool->ref_clocks.dchub_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
 		}
 	} else
 		ASSERT_CRITICAL(false);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
index 18e94d8ae54fc..3d19acaa12f3c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c
@@ -151,8 +151,4 @@ void dcn30_hw_sequencer_construct(struct dc *dc)
 	dc->hwss = dcn30_funcs;
 	dc->hwseq->funcs = dcn30_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
index 34b08d90dc1da..c6f5f3df8061a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
@@ -55,10 +55,9 @@ void optc3_triplebuffer_lock(struct timing_generator *optc)
 	REG_SET(OTG_MASTER_UPDATE_LOCK, 0,
 		OTG_MASTER_UPDATE_LOCK, 1);
 
-	if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS)
-		REG_WAIT(OTG_MASTER_UPDATE_LOCK,
-				UPDATE_LOCK_STATUS, 1,
-				1, 10);
+	REG_WAIT(OTG_MASTER_UPDATE_LOCK,
+			UPDATE_LOCK_STATUS, 1,
+			1, 10);
 
 	TRACE_OPTC_LOCK_UNLOCK_STATE(optc1, optc->inst, true);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index 67a34cda3774a..cd94b5f5fdb4e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -2376,10 +2376,7 @@ static bool dcn30_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2577,8 +2574,7 @@ static bool dcn30_resource_construct(
 
 	/* Audio, Stream Encoders including DIG and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
+			&res_create_funcs))
 		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
index 5ac2a272c380f..1cf84a086fecb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
@@ -1513,10 +1513,7 @@ static bool dcn301_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -1710,9 +1707,8 @@ static bool dcn301_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
 	dcn301_hw_sequencer_construct(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
index 9f93c43115ba3..efd98d64588d3 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
@@ -1309,8 +1309,6 @@ static bool dcn302_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else
-		dc->debug = debug_defaults_diags;
 
 	// Init the vm_helper
 	if (dc->vm_helper)
@@ -1489,8 +1487,7 @@ static bool dcn302_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, pool,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-					&res_create_funcs : &res_create_maximus_funcs)))
+			&res_create_funcs))
 		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
index 7f72ef882ca41..fcd1266021789 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
@@ -1232,8 +1232,6 @@ static bool dcn303_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else
-		dc->debug = debug_defaults_diags;
 
 	// Init the vm_helper
 	if (dc->vm_helper)
@@ -1400,8 +1398,7 @@ static bool dcn303_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, pool,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-					&res_create_funcs : &res_create_maximus_funcs)))
+			&res_create_funcs))
 		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
index 55494730e5008..2a7f47642a447 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c
@@ -117,28 +117,6 @@ void dcn31_init_hw(struct dc *dc)
 	if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks)
 		dc->clk_mgr->funcs->init_clocks(dc->clk_mgr);
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-
-		REG_WRITE(REFCLK_CNTL, 0);
-		REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1);
-		REG_WRITE(DIO_MEM_PWR_CTRL, 0);
-
-		if (!dc->debug.disable_clock_gate) {
-			/* enable all DCN clock gating */
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-
-			REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
-
-			REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
-		}
-
-		//Enable ability to power gate / don't force power on permanently
-		if (hws->funcs.enable_power_gating_plane)
-			hws->funcs.enable_power_gating_plane(hws, true);
-
-		return;
-	}
-
 	if (!dcb->funcs->is_accelerated_mode(dcb)) {
 		hws->funcs.bios_golden_init(dc);
 		if (hws->funcs.disable_vga)
@@ -154,23 +132,21 @@ void dcn31_init_hw(struct dc *dc)
 		res_pool->ref_clocks.xtalin_clock_inKhz =
 				dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency;
 
-		if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-			if (res_pool->dccg && res_pool->hubbub) {
-
-				(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
-						dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
-						&res_pool->ref_clocks.dccg_ref_clock_inKhz);
-
-				(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
-						res_pool->ref_clocks.dccg_ref_clock_inKhz,
-						&res_pool->ref_clocks.dchub_ref_clock_inKhz);
-			} else {
-				// Not all ASICs have DCCG sw component
-				res_pool->ref_clocks.dccg_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-				res_pool->ref_clocks.dchub_ref_clock_inKhz =
-						res_pool->ref_clocks.xtalin_clock_inKhz;
-			}
+		if (res_pool->dccg && res_pool->hubbub) {
+
+			(res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg,
+					dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency,
+					&res_pool->ref_clocks.dccg_ref_clock_inKhz);
+
+			(res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub,
+					res_pool->ref_clocks.dccg_ref_clock_inKhz,
+					&res_pool->ref_clocks.dchub_ref_clock_inKhz);
+		} else {
+			// Not all ASICs have DCCG sw component
+			res_pool->ref_clocks.dccg_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
+			res_pool->ref_clocks.dchub_ref_clock_inKhz =
+					res_pool->ref_clocks.xtalin_clock_inKhz;
 		}
 	} else
 		ASSERT_CRITICAL(false);
@@ -553,35 +529,31 @@ static void dcn31_reset_back_end_for_pipe(
 		pipe_ctx->stream_res.tg->funcs->set_drr(
 				pipe_ctx->stream_res.tg, NULL);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		link = pipe_ctx->stream->link;
-		/* DPMS may already disable or */
-		/* dpms_off status is incorrect due to fastboot
-		 * feature. When system resume from S4 with second
-		 * screen only, the dpms_off would be true but
-		 * VBIOS lit up eDP, so check link status too.
-		 */
-		if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
-			dc->link_srv->set_dpms_off(pipe_ctx);
-		else if (pipe_ctx->stream_res.audio)
-			dc->hwss.disable_audio_stream(pipe_ctx);
-
-		/* free acquired resources */
-		if (pipe_ctx->stream_res.audio) {
-			/*disable az_endpoint*/
-			pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
-
-			/*free audio*/
-			if (dc->caps.dynamic_audio == true) {
-				/*we have to dynamic arbitrate the audio endpoints*/
-				/*we free the resource, need reset is_audio_acquired*/
-				update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
-						pipe_ctx->stream_res.audio, false);
-				pipe_ctx->stream_res.audio = NULL;
-			}
+	link = pipe_ctx->stream->link;
+	/* DPMS may already disable or */
+	/* dpms_off status is incorrect due to fastboot
+	 * feature. When system resume from S4 with second
+	 * screen only, the dpms_off would be true but
+	 * VBIOS lit up eDP, so check link status too.
+	 */
+	if (!pipe_ctx->stream->dpms_off || link->link_status.link_active)
+		dc->link_srv->set_dpms_off(pipe_ctx);
+	else if (pipe_ctx->stream_res.audio)
+		dc->hwss.disable_audio_stream(pipe_ctx);
+
+	/* free acquired resources */
+	if (pipe_ctx->stream_res.audio) {
+		/*disable az_endpoint*/
+		pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio);
+
+		/*free audio*/
+		if (dc->caps.dynamic_audio == true) {
+			/*we have to dynamic arbitrate the audio endpoints*/
+			/*we free the resource, need reset is_audio_acquired*/
+			update_audio_usage(&dc->current_state->res_ctx, dc->res_pool,
+					pipe_ctx->stream_res.audio, false);
+			pipe_ctx->stream_res.audio = NULL;
 		}
-	} else if (pipe_ctx->stream_res.dsc) {
-		dc->link_srv->set_dsc_enable(pipe_ctx, false);
 	}
 
 	pipe_ctx->stream = NULL;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
index ba9e7dee6e5e5..fc25cc300a175 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c
@@ -154,8 +154,4 @@ void dcn31_hw_sequencer_construct(struct dc *dc)
 	dc->hwss = dcn31_funcs;
 	dc->hwseq->funcs = dcn31_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
index ff8cd50764348..9c637c895f914 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
@@ -1341,13 +1341,6 @@ static struct dce_hwseq *dcn31_hwseq_create(
 		hws->regs = &hwseq_reg;
 		hws->shifts = &hwseq_shift;
 		hws->masks = &hwseq_mask;
-		/* DCN3.1 FPGA Workaround
-		 * Need to enable HPO DP Stream Encoder before setting OTG master enable.
-		 * To do so, move calling function enable_stream_timing to only be done AFTER calling
-		 * function core_link_enable_stream
-		 */
-		if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-			hws->wa.dp_hpo_and_otg_sequence = true;
 	}
 	return hws;
 }
@@ -1988,10 +1981,7 @@ static bool dcn31_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2195,9 +2185,8 @@ static bool dcn31_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
 	dcn31_hw_sequencer_construct(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
index 90be62c058229..86d6a514dec0b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c
@@ -160,8 +160,4 @@ void dcn314_hw_sequencer_construct(struct dc *dc)
 	dc->hwss = dcn314_funcs;
 	dc->hwseq->funcs = dcn314_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index abeeede38fb39..2483d37e425da 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -1375,13 +1375,6 @@ static struct dce_hwseq *dcn314_hwseq_create(
 		hws->regs = &hwseq_reg;
 		hws->shifts = &hwseq_shift;
 		hws->masks = &hwseq_mask;
-		/* DCN3.1 FPGA Workaround
-		 * Need to enable HPO DP Stream Encoder before setting OTG master enable.
-		 * To do so, move calling function enable_stream_timing to only be done AFTER calling
-		 * function core_link_enable_stream
-		 */
-		if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-			hws->wa.dp_hpo_and_otg_sequence = true;
 	}
 	return hws;
 }
@@ -2101,8 +2094,7 @@ static bool dcn314_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-				(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-				 &res_create_funcs : &res_create_maximus_funcs)))
+			&res_create_funcs))
 		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index 95fd3d087ea35..e6d87c162d26c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -1342,13 +1342,6 @@ static struct dce_hwseq *dcn31_hwseq_create(
 		hws->regs = &hwseq_reg;
 		hws->shifts = &hwseq_shift;
 		hws->masks = &hwseq_mask;
-		/* DCN3.1 FPGA Workaround
-		 * Need to enable HPO DP Stream Encoder before setting OTG master enable.
-		 * To do so, move calling function enable_stream_timing to only be done AFTER calling
-		 * function core_link_enable_stream
-		 */
-		if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-			hws->wa.dp_hpo_and_otg_sequence = true;
 	}
 	return hws;
 }
@@ -1947,10 +1940,7 @@ static bool dcn315_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2131,9 +2121,8 @@ static bool dcn315_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
 	dcn31_hw_sequencer_construct(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
index 9ead347a33e93..a3be61cc541f7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
@@ -1340,13 +1340,6 @@ static struct dce_hwseq *dcn31_hwseq_create(
 		hws->regs = &hwseq_reg;
 		hws->shifts = &hwseq_shift;
 		hws->masks = &hwseq_mask;
-		/* DCN3.1 FPGA Workaround
-		 * Need to enable HPO DP Stream Encoder before setting OTG master enable.
-		 * To do so, move calling function enable_stream_timing to only be done AFTER calling
-		 * function core_link_enable_stream
-		 */
-		if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment))
-			hws->wa.dp_hpo_and_otg_sequence = true;
 	}
 	return hws;
 }
@@ -1844,10 +1837,7 @@ static bool dcn316_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2028,9 +2018,8 @@ static bool dcn316_resource_construct(
 
 	/* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer and Plane caps */
 	dcn31_hw_sequencer_construct(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
index 24a890d879b80..6f9a165c1eab9 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
@@ -161,8 +161,4 @@ void dcn32_hw_sequencer_init_functions(struct dc *dc)
 	dc->hwss = dcn32_funcs;
 	dc->hwseq->funcs = dcn32_private_funcs;
 
-	if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		dc->hwss.init_hw = dcn20_fpga_init_hw;
-		dc->hwseq->funcs.init_pipes = NULL;
-	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 98c394f9f8cfc..33abc8c9d4bec 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -2260,10 +2260,7 @@ static bool dcn32_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -2319,8 +2316,7 @@ static bool dcn32_resource_construct(
 	}
 
 	/* DML */
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_2_soc, &dcn3_2_ip, DML_PROJECT_DCN32);
+	dml_init_instance(&dc->dml, &dcn3_2_soc, &dcn3_2_ip, DML_PROJECT_DCN32);
 
 	/* IRQ Service */
 	init_data.ctx = dc->ctx;
@@ -2457,9 +2453,8 @@ static bool dcn32_resource_construct(
 
 	/* Audio, HWSeq, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer init functions and Plane caps */
 	dcn32_hw_sequencer_init_functions(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index f4cd9749ffdf8..af0bb3e94250a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -1801,10 +1801,7 @@ static bool dcn321_resource_construct(
 
 	if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV)
 		dc->debug = debug_defaults_drv;
-	else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) {
-		dc->debug = debug_defaults_diags;
-	} else
-		dc->debug = debug_defaults_diags;
+
 	// Init the vm_helper
 	if (dc->vm_helper)
 		vm_helper_init(dc->vm_helper, 16);
@@ -1860,8 +1857,7 @@ static bool dcn321_resource_construct(
 	}
 
 	/* DML */
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
+	dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
 
 	/* IRQ Service */
 	init_data.ctx = dc->ctx;
@@ -1993,9 +1989,8 @@ static bool dcn321_resource_construct(
 
 	/* Audio, HWSeq, Stream Encoders including HPO and virtual, MPC 3D LUTs */
 	if (!resource_construct(num_virtual_links, dc, &pool->base,
-			(!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ?
-			&res_create_funcs : &res_create_maximus_funcs)))
-			goto create_fail;
+			&res_create_funcs))
+		goto create_fail;
 
 	/* HW Sequencer init functions and Plane caps */
 	dcn32_hw_sequencer_init_functions(dc);
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
index 422f17aefd4a6..6ce90678b33c0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
@@ -333,45 +333,43 @@ void dcn301_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
 	memcpy(s, dcn3_01_soc.clock_limits, sizeof(dcn3_01_soc.clock_limits));
 
 	/* Default clock levels are used for diags, which may lead to overclocking. */
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		dcn3_01_ip.max_num_otg = pool->base.res_cap->num_timing_generator;
-		dcn3_01_ip.max_num_dpp = pool->base.pipe_count;
-		dcn3_01_soc.num_chans = bw_params->num_channels;
-
-		ASSERT(clk_table->num_entries);
-		for (i = 0; i < clk_table->num_entries; i++) {
-			/* loop backwards*/
-			for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) {
-				if ((unsigned int) dcn3_01_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
-					closest_clk_lvl = j;
-					break;
-				}
+	dcn3_01_ip.max_num_otg = pool->base.res_cap->num_timing_generator;
+	dcn3_01_ip.max_num_dpp = pool->base.pipe_count;
+	dcn3_01_soc.num_chans = bw_params->num_channels;
+
+	ASSERT(clk_table->num_entries);
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_01_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
 			}
-
-			s[i].state = i;
-			s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
-			s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
-			s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
-			s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
-
-			s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
-			s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
-			s[i].dram_bw_per_chan_gbps =
-				dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
-			s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
-			s[i].dtbclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
-			s[i].phyclk_d18_mhz =
-				dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
-			s[i].phyclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
 		}
 
-		if (clk_table->num_entries) {
-			dcn3_01_soc.num_states = clk_table->num_entries;
-			/* duplicate last level */
-			s[dcn3_01_soc.num_states] =
-				dcn3_01_soc.clock_limits[dcn3_01_soc.num_states - 1];
-			s[dcn3_01_soc.num_states].state = dcn3_01_soc.num_states;
-		}
+		s[i].state = i;
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
+
+		s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+		s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+
+	if (clk_table->num_entries) {
+		dcn3_01_soc.num_states = clk_table->num_entries;
+		/* duplicate last level */
+		s[dcn3_01_soc.num_states] =
+			dcn3_01_soc.clock_limits[dcn3_01_soc.num_states - 1];
+		s[dcn3_01_soc.num_states].state = dcn3_01_soc.num_states;
 	}
 
 	memcpy(dcn3_01_soc.clock_limits, s, sizeof(dcn3_01_soc.clock_limits));
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
index 19d034341e640..deb6d162a2d5c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c
@@ -582,6 +582,7 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params
 	struct _vcs_dpi_voltage_scaling_st *s = dc->scratch.update_bw_bounding_box.clock_limits;
 	struct clk_limit_table *clk_table = &bw_params->clk_table;
 	unsigned int i, closest_clk_lvl;
+	int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
 	int j;
 
 	dc_assert_fp_enabled();
@@ -589,59 +590,55 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params
 	memcpy(s, dcn3_1_soc.clock_limits, sizeof(dcn3_1_soc.clock_limits));
 
 	// Default clock levels are used for diags, which may lead to overclocking.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-		int max_dispclk_mhz = 0, max_dppclk_mhz = 0;
+	dcn3_1_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+	dcn3_1_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_1_soc.num_chans = bw_params->num_channels;
 
-		dcn3_1_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
-		dcn3_1_ip.max_num_dpp = dc->res_pool->pipe_count;
-		dcn3_1_soc.num_chans = bw_params->num_channels;
+	ASSERT(clk_table->num_entries);
 
-		ASSERT(clk_table->num_entries);
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
 
-		/* Prepass to find max clocks independent of voltage level. */
-		for (i = 0; i < clk_table->num_entries; ++i) {
-			if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
-				max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
-			if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
-				max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_1_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
+			}
 		}
 
-		for (i = 0; i < clk_table->num_entries; i++) {
-			/* loop backwards*/
-			for (closest_clk_lvl = 0, j = dcn3_1_soc.num_states - 1; j >= 0; j--) {
-				if ((unsigned int) dcn3_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) {
-					closest_clk_lvl = j;
-					break;
-				}
-			}
+		s[i].state = i;
 
-			s[i].state = i;
-
-			/* Clocks dependent on voltage level. */
-			s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
-			s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
-			s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
-			s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
-				2 * clk_table->entries[i].wck_ratio;
-
-			/* Clocks independent of voltage level. */
-			s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
-				dcn3_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
-
-			s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
-				dcn3_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
-
-			s[i].dram_bw_per_chan_gbps =
-				dcn3_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
-			s[i].dscclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
-			s[i].dtbclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
-			s[i].phyclk_d18_mhz =
-				dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
-			s[i].phyclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
-		}
-		if (clk_table->num_entries) {
-			dcn3_1_soc.num_states = clk_table->num_entries;
-		}
+		/* Clocks dependent on voltage level. */
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
+			2 * clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+	if (clk_table->num_entries) {
+		dcn3_1_soc.num_states = clk_table->num_entries;
 	}
 
 	memcpy(dcn3_1_soc.clock_limits, s, sizeof(dcn3_1_soc.clock_limits));
@@ -655,10 +652,7 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params
 		dcn3_1_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31);
-	else
-		dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31_FPGA);
+	dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31);
 }
 
 void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
@@ -719,10 +713,7 @@ void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
 		dcn3_15_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN315);
-	else
-		dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN31_FPGA);
+	dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN315);
 }
 
 void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params)
@@ -738,71 +729,68 @@ void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
 	memcpy(s, dcn3_16_soc.clock_limits, sizeof(dcn3_16_soc.clock_limits));
 
 	// Default clock levels are used for diags, which may lead to overclocking.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment)) {
-
-		dcn3_16_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
-		dcn3_16_ip.max_num_dpp = dc->res_pool->pipe_count;
-		dcn3_16_soc.num_chans = bw_params->num_channels;
+	dcn3_16_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
+	dcn3_16_ip.max_num_dpp = dc->res_pool->pipe_count;
+	dcn3_16_soc.num_chans = bw_params->num_channels;
 
-		ASSERT(clk_table->num_entries);
+	ASSERT(clk_table->num_entries);
 
-		/* Prepass to find max clocks independent of voltage level. */
-		for (i = 0; i < clk_table->num_entries; ++i) {
-			if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
-				max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
-			if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
-				max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
-		}
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
 
-		for (i = 0; i < clk_table->num_entries; i++) {
-			/* loop backwards*/
-			for (closest_clk_lvl = 0, j = dcn3_16_soc.num_states - 1; j >= 0; j--) {
-				if ((unsigned int) dcn3_16_soc.clock_limits[j].dcfclk_mhz <=
-				    clk_table->entries[i].dcfclk_mhz) {
-					closest_clk_lvl = j;
-					break;
-				}
-			}
-			// Ported from DCN315
-			if (clk_table->num_entries == 1) {
-				/*smu gives one DPM level, let's take the highest one*/
-				closest_clk_lvl = dcn3_16_soc.num_states - 1;
+	for (i = 0; i < clk_table->num_entries; i++) {
+		/* loop backwards*/
+		for (closest_clk_lvl = 0, j = dcn3_16_soc.num_states - 1; j >= 0; j--) {
+			if ((unsigned int) dcn3_16_soc.clock_limits[j].dcfclk_mhz <=
+			    clk_table->entries[i].dcfclk_mhz) {
+				closest_clk_lvl = j;
+				break;
 			}
+		}
+		// Ported from DCN315
+		if (clk_table->num_entries == 1) {
+			/*smu gives one DPM level, let's take the highest one*/
+			closest_clk_lvl = dcn3_16_soc.num_states - 1;
+		}
 
-			s[i].state = i;
+		s[i].state = i;
 
-			/* Clocks dependent on voltage level. */
-			s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
-			if (clk_table->num_entries == 1 &&
-			    s[i].dcfclk_mhz <
-			    dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
-				/*SMU fix not released yet*/
-				s[i].dcfclk_mhz =
-					dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
-			}
-			s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
-			s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
-			s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
-				2 * clk_table->entries[i].wck_ratio;
-
-			/* Clocks independent of voltage level. */
-			s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
-				dcn3_16_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
-
-			s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
-				dcn3_16_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
-
-			s[i].dram_bw_per_chan_gbps =
-				dcn3_16_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
-			s[i].dscclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
-			s[i].dtbclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
-			s[i].phyclk_d18_mhz =
-				dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
-			s[i].phyclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
-		}
-		if (clk_table->num_entries) {
-			dcn3_16_soc.num_states = clk_table->num_entries;
+		/* Clocks dependent on voltage level. */
+		s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz;
+		if (clk_table->num_entries == 1 &&
+		    s[i].dcfclk_mhz <
+		    dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) {
+			/*SMU fix not released yet*/
+			s[i].dcfclk_mhz =
+				dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz;
 		}
+		s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz;
+		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
+		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz *
+			2 * clk_table->entries[i].wck_ratio;
+
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
+		s[i].dram_bw_per_chan_gbps =
+			dcn3_16_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
+		s[i].dscclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dscclk_mhz;
+		s[i].dtbclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dtbclk_mhz;
+		s[i].phyclk_d18_mhz =
+			dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz;
+		s[i].phyclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_mhz;
+	}
+	if (clk_table->num_entries) {
+		dcn3_16_soc.num_states = clk_table->num_entries;
 	}
 
 	memcpy(dcn3_16_soc.clock_limits, s, sizeof(dcn3_16_soc.clock_limits));
@@ -817,10 +805,7 @@ void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param
 		dcn3_16_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000;
 	}
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31);
-	else
-		dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31_FPGA);
+	dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31);
 }
 
 int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
index fcde8f21b8be0..4113ce79c4aff 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
@@ -1432,14 +1432,6 @@ static void dml_rq_dlg_get_dlg_params(
 	dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip    = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip);
 	dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip   = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip);
 
-	// hack for FPGA
-	if (mode_lib->project == DML_PROJECT_DCN31_FPGA) {
-		if (disp_dlg_regs->vratio_prefetch >= (unsigned int) dml_pow(2, 22)) {
-			disp_dlg_regs->vratio_prefetch = (unsigned int) dml_pow(2, 22) - 1;
-			dml_print("vratio_prefetch exceed the max value, the register field is [21:0]\n");
-		}
-	}
-
 	disp_dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l);
 	ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13));
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 554152371eb53..318b9c2bc9be3 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -190,8 +190,7 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	dc_assert_fp_enabled();
 
 	// Default clock levels are used for diags, which may lead to overclocking.
-	if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc->config.use_default_clock_table == false) {
-
+	if (dc->config.use_default_clock_table == false) {
 		dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator;
 		dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count;
 
@@ -267,10 +266,7 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 
 	dcn20_patch_bounding_box(dc, &dcn3_14_soc);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
 		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314);
-	else
-		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA);
 }
 
 static bool is_dual_plane(enum surface_pixel_format format)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
index 61ba3e33bb11d..b3e8dc08030cb 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
@@ -1520,14 +1520,6 @@ static void dml_rq_dlg_get_dlg_params(
 	dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip    = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip);
 	dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip   = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip);
 
-	// hack for FPGA
-	if (mode_lib->project == DML_PROJECT_DCN31_FPGA) {
-		if (disp_dlg_regs->vratio_prefetch >= (unsigned int) dml_pow(2, 22)) {
-			disp_dlg_regs->vratio_prefetch = (unsigned int) dml_pow(2, 22) - 1;
-			dml_print("vratio_prefetch exceed the max value, the register field is [21:0]\n");
-		}
-	}
-
 	disp_dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l);
 	ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13));
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index d8d8fcd5ef1ff..137ff970c9aae 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2530,80 +2530,78 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 {
 	dc_assert_fp_enabled();
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		/* Overrides from dc->config options */
-		dcn3_2_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
-
-		/* Override from passed dc->bb_overrides if available*/
-		if ((int)(dcn3_2_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns
-				&& dc->bb_overrides.sr_exit_time_ns) {
-			dcn3_2_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
-		}
+	/* Overrides from dc->config options */
+	dcn3_2_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
 
-		if ((int)(dcn3_2_soc.sr_enter_plus_exit_time_us * 1000)
-				!= dc->bb_overrides.sr_enter_plus_exit_time_ns
-				&& dc->bb_overrides.sr_enter_plus_exit_time_ns) {
-			dcn3_2_soc.sr_enter_plus_exit_time_us =
-				dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
-		}
+	/* Override from passed dc->bb_overrides if available*/
+	if ((int)(dcn3_2_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns
+			&& dc->bb_overrides.sr_exit_time_ns) {
+		dcn3_2_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_2_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
-			&& dc->bb_overrides.urgent_latency_ns) {
-			dcn3_2_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
-			dcn3_2_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_2_soc.sr_enter_plus_exit_time_us * 1000)
+			!= dc->bb_overrides.sr_enter_plus_exit_time_ns
+			&& dc->bb_overrides.sr_enter_plus_exit_time_ns) {
+		dcn3_2_soc.sr_enter_plus_exit_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_2_soc.dram_clock_change_latency_us * 1000)
-				!= dc->bb_overrides.dram_clock_change_latency_ns
-				&& dc->bb_overrides.dram_clock_change_latency_ns) {
-			dcn3_2_soc.dram_clock_change_latency_us =
-				dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_2_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
+		&& dc->bb_overrides.urgent_latency_ns) {
+		dcn3_2_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+		dcn3_2_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000)
-				!= dc->bb_overrides.fclk_clock_change_latency_ns
-				&& dc->bb_overrides.fclk_clock_change_latency_ns) {
-			dcn3_2_soc.fclk_change_latency_us =
-				dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
-		}
+	if ((int)(dcn3_2_soc.dram_clock_change_latency_us * 1000)
+			!= dc->bb_overrides.dram_clock_change_latency_ns
+			&& dc->bb_overrides.dram_clock_change_latency_ns) {
+		dcn3_2_soc.dram_clock_change_latency_us =
+			dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000)
-				!= dc->bb_overrides.dummy_clock_change_latency_ns
-				&& dc->bb_overrides.dummy_clock_change_latency_ns) {
-			dcn3_2_soc.dummy_pstate_latency_us =
-				dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000)
+			!= dc->bb_overrides.fclk_clock_change_latency_ns
+			&& dc->bb_overrides.fclk_clock_change_latency_ns) {
+		dcn3_2_soc.fclk_change_latency_us =
+			dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
+	}
 
-		/* Override from VBIOS if VBIOS bb_info available */
-		if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
-			struct bp_soc_bb_info bb_info = {0};
+	if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000)
+			!= dc->bb_overrides.dummy_clock_change_latency_ns
+			&& dc->bb_overrides.dummy_clock_change_latency_ns) {
+		dcn3_2_soc.dummy_pstate_latency_us =
+			dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0;
+	}
 
-			if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
-				if (bb_info.dram_clock_change_latency_100ns > 0)
-					dcn3_2_soc.dram_clock_change_latency_us =
-						bb_info.dram_clock_change_latency_100ns * 10;
+	/* Override from VBIOS if VBIOS bb_info available */
+	if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
+		struct bp_soc_bb_info bb_info = {0};
 
-				if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
-					dcn3_2_soc.sr_enter_plus_exit_time_us =
-						bb_info.dram_sr_enter_exit_latency_100ns * 10;
+		if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dcn3_2_soc.dram_clock_change_latency_us =
+					bb_info.dram_clock_change_latency_100ns * 10;
 
-				if (bb_info.dram_sr_exit_latency_100ns > 0)
-					dcn3_2_soc.sr_exit_time_us =
-						bb_info.dram_sr_exit_latency_100ns * 10;
-			}
-		}
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dcn3_2_soc.sr_enter_plus_exit_time_us =
+					bb_info.dram_sr_enter_exit_latency_100ns * 10;
 
-		/* Override from VBIOS for num_chan */
-		if (dc->ctx->dc_bios->vram_info.num_chans) {
-			dcn3_2_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
-			dcn3_2_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc,
-				dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel);
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dcn3_2_soc.sr_exit_time_us =
+					bb_info.dram_sr_exit_latency_100ns * 10;
 		}
+	}
 
-		if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
-			dcn3_2_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+	/* Override from VBIOS for num_chan */
+	if (dc->ctx->dc_bios->vram_info.num_chans) {
+		dcn3_2_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
+		dcn3_2_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc,
+			dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel);
 	}
 
+	if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
+		dcn3_2_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+
 	/* DML DSC delay factor workaround */
 	dcn3_2_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0;
 
@@ -2614,7 +2612,7 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
 
 	/* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */
-	if ((!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) && (bw_params->clk_table.entries[0].memclk_mhz)) {
+	if (bw_params->clk_table.entries[0].memclk_mhz) {
 		if (dc->debug.use_legacy_soc_bb_mechanism) {
 			unsigned int i = 0, j = 0, num_states = 0;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index 342a1bcb49273..ffd7c3c1b1420 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -471,80 +471,78 @@ static void dcn321_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts,
 void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params)
 {
 	dc_assert_fp_enabled();
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		/* Overrides from dc->config options */
-		dcn3_21_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
-
-		/* Override from passed dc->bb_overrides if available*/
-		if ((int)(dcn3_21_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns
-				&& dc->bb_overrides.sr_exit_time_ns) {
-			dcn3_21_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
-		}
+	/* Overrides from dc->config options */
+	dcn3_21_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk;
 
-		if ((int)(dcn3_21_soc.sr_enter_plus_exit_time_us * 1000)
-				!= dc->bb_overrides.sr_enter_plus_exit_time_ns
-				&& dc->bb_overrides.sr_enter_plus_exit_time_ns) {
-			dcn3_21_soc.sr_enter_plus_exit_time_us =
-				dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
-		}
+	/* Override from passed dc->bb_overrides if available*/
+	if ((int)(dcn3_21_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns
+			&& dc->bb_overrides.sr_exit_time_ns) {
+		dcn3_21_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_21_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
-			&& dc->bb_overrides.urgent_latency_ns) {
-			dcn3_21_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
-			dcn3_21_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_21_soc.sr_enter_plus_exit_time_us * 1000)
+			!= dc->bb_overrides.sr_enter_plus_exit_time_ns
+			&& dc->bb_overrides.sr_enter_plus_exit_time_ns) {
+		dcn3_21_soc.sr_enter_plus_exit_time_us =
+			dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_21_soc.dram_clock_change_latency_us * 1000)
-				!= dc->bb_overrides.dram_clock_change_latency_ns
-				&& dc->bb_overrides.dram_clock_change_latency_ns) {
-			dcn3_21_soc.dram_clock_change_latency_us =
-				dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_21_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns
+		&& dc->bb_overrides.urgent_latency_ns) {
+		dcn3_21_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+		dcn3_21_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000)
-				!= dc->bb_overrides.fclk_clock_change_latency_ns
-				&& dc->bb_overrides.fclk_clock_change_latency_ns) {
-			dcn3_21_soc.fclk_change_latency_us =
-				dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
-		}
+	if ((int)(dcn3_21_soc.dram_clock_change_latency_us * 1000)
+			!= dc->bb_overrides.dram_clock_change_latency_ns
+			&& dc->bb_overrides.dram_clock_change_latency_ns) {
+		dcn3_21_soc.dram_clock_change_latency_us =
+			dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+	}
 
-		if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000)
-				!= dc->bb_overrides.dummy_clock_change_latency_ns
-				&& dc->bb_overrides.dummy_clock_change_latency_ns) {
-			dcn3_21_soc.dummy_pstate_latency_us =
-				dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0;
-		}
+	if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000)
+			!= dc->bb_overrides.fclk_clock_change_latency_ns
+			&& dc->bb_overrides.fclk_clock_change_latency_ns) {
+		dcn3_21_soc.fclk_change_latency_us =
+			dc->bb_overrides.fclk_clock_change_latency_ns / 1000;
+	}
 
-		/* Override from VBIOS if VBIOS bb_info available */
-		if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
-			struct bp_soc_bb_info bb_info = {0};
+	if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000)
+			!= dc->bb_overrides.dummy_clock_change_latency_ns
+			&& dc->bb_overrides.dummy_clock_change_latency_ns) {
+		dcn3_21_soc.dummy_pstate_latency_us =
+			dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0;
+	}
 
-			if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
-				if (bb_info.dram_clock_change_latency_100ns > 0)
-					dcn3_21_soc.dram_clock_change_latency_us =
-						bb_info.dram_clock_change_latency_100ns * 10;
+	/* Override from VBIOS if VBIOS bb_info available */
+	if (dc->ctx->dc_bios->funcs->get_soc_bb_info) {
+		struct bp_soc_bb_info bb_info = {0};
 
-				if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
-					dcn3_21_soc.sr_enter_plus_exit_time_us =
-						bb_info.dram_sr_enter_exit_latency_100ns * 10;
+		if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) {
+			if (bb_info.dram_clock_change_latency_100ns > 0)
+				dcn3_21_soc.dram_clock_change_latency_us =
+					bb_info.dram_clock_change_latency_100ns * 10;
 
-				if (bb_info.dram_sr_exit_latency_100ns > 0)
-					dcn3_21_soc.sr_exit_time_us =
-						bb_info.dram_sr_exit_latency_100ns * 10;
-			}
-		}
+			if (bb_info.dram_sr_enter_exit_latency_100ns > 0)
+				dcn3_21_soc.sr_enter_plus_exit_time_us =
+					bb_info.dram_sr_enter_exit_latency_100ns * 10;
 
-		/* Override from VBIOS for num_chan */
-		if (dc->ctx->dc_bios->vram_info.num_chans) {
-			dcn3_21_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
-			dcn3_21_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc,
-				dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel);
+			if (bb_info.dram_sr_exit_latency_100ns > 0)
+				dcn3_21_soc.sr_exit_time_us =
+					bb_info.dram_sr_exit_latency_100ns * 10;
 		}
+	}
 
-		if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
-			dcn3_21_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+	/* Override from VBIOS for num_chan */
+	if (dc->ctx->dc_bios->vram_info.num_chans) {
+		dcn3_21_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans;
+		dcn3_21_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc,
+			dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel);
 	}
 
+	if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes)
+		dcn3_21_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
+
 	/* DML DSC delay factor workaround */
 	dcn3_21_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0;
 
@@ -555,7 +553,6 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
 
 	/* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */
-	if ((!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) && (bw_params->clk_table.entries[0].memclk_mhz)) {
 		if (dc->debug.use_legacy_soc_bb_mechanism) {
 			unsigned int i = 0, j = 0, num_states = 0;
 
@@ -699,6 +696,5 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 		dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
 		if (dc->current_state)
 			dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
-	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
index bdf3ac6cadd55..da0cfbb071e6a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c
@@ -113,7 +113,6 @@ void dml_init_instance(struct display_mode_lib *lib,
 		lib->funcs = dml30_funcs;
 		break;
 	case DML_PROJECT_DCN31:
-	case DML_PROJECT_DCN31_FPGA:
 	case DML_PROJECT_DCN315:
 		lib->funcs = dml31_funcs;
 		break;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
index a9d49ef58fb59..5edf69fa40d11 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h
@@ -41,7 +41,6 @@ enum dml_project {
 	DML_PROJECT_DCN30,
 	DML_PROJECT_DCN31,
 	DML_PROJECT_DCN315,
-	DML_PROJECT_DCN31_FPGA,
 	DML_PROJECT_DCN314,
 	DML_PROJECT_DCN32,
 };
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c
index edd7d026a762a..586fe25c17029 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c
@@ -28,24 +28,6 @@
 #include "dccg.h"
 #include "clk_mgr.h"
 
-static enum phyd32clk_clock_source get_phyd32clk_src(struct dc_link *link)
-{
-	switch (link->link_enc->transmitter) {
-	case TRANSMITTER_UNIPHY_A:
-		return PHYD32CLKA;
-	case TRANSMITTER_UNIPHY_B:
-		return PHYD32CLKB;
-	case TRANSMITTER_UNIPHY_C:
-		return PHYD32CLKC;
-	case TRANSMITTER_UNIPHY_D:
-		return PHYD32CLKD;
-	case TRANSMITTER_UNIPHY_E:
-		return PHYD32CLKE;
-	default:
-		return PHYD32CLKA;
-	}
-}
-
 static void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx,
 		struct fixed31_32 throttled_vcp_size)
 {
@@ -120,81 +102,26 @@ static void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx)
 			DPCD_SOURCE_SEQ_AFTER_DP_STREAM_ATTR);
 }
 
-static void enable_hpo_dp_fpga_link_output(struct dc_link *link,
-		const struct link_resource *link_res,
-		enum signal_type signal,
-		enum clock_source_id clock_source,
-		const struct dc_link_settings *link_settings)
-{
-	const struct dc *dc = link->dc;
-	enum phyd32clk_clock_source phyd32clk = get_phyd32clk_src(link);
-	int phyd32clk_freq_khz = link_settings->link_rate == LINK_RATE_UHBR10 ? 312500 :
-			link_settings->link_rate == LINK_RATE_UHBR13_5 ? 412875 :
-			link_settings->link_rate == LINK_RATE_UHBR20 ? 625000 : 0;
-
-	dm_set_phyd32clk(dc->ctx, phyd32clk_freq_khz);
-	dc->res_pool->dccg->funcs->set_physymclk(
-			dc->res_pool->dccg,
-			link->link_enc_hw_inst,
-			PHYSYMCLK_FORCE_SRC_PHYD32CLK,
-			true);
-	dc->res_pool->dccg->funcs->enable_symclk32_le(
-			dc->res_pool->dccg,
-			link_res->hpo_dp_link_enc->inst,
-			phyd32clk);
-	link_res->hpo_dp_link_enc->funcs->link_enable(
-			link_res->hpo_dp_link_enc,
-			link_settings->lane_count);
-
-}
-
 static void enable_hpo_dp_link_output(struct dc_link *link,
 		const struct link_resource *link_res,
 		enum signal_type signal,
 		enum clock_source_id clock_source,
 		const struct dc_link_settings *link_settings)
 {
-	if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment))
-		enable_hpo_dp_fpga_link_output(link, link_res, signal,
-				clock_source, link_settings);
-	else
-		link_res->hpo_dp_link_enc->funcs->enable_link_phy(
-				link_res->hpo_dp_link_enc,
-				link_settings,
-				link->link_enc->transmitter,
-				link->link_enc->hpd_source);
-}
-
-
-static void disable_hpo_dp_fpga_link_output(struct dc_link *link,
-		const struct link_resource *link_res,
-		enum signal_type signal)
-{
-	const struct dc *dc = link->dc;
-
-	link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc);
-	dc->res_pool->dccg->funcs->disable_symclk32_le(
-			dc->res_pool->dccg,
-			link_res->hpo_dp_link_enc->inst);
-	dc->res_pool->dccg->funcs->set_physymclk(
-			dc->res_pool->dccg,
-			link->link_enc_hw_inst,
-			PHYSYMCLK_FORCE_SRC_SYMCLK,
-			false);
-	dm_set_phyd32clk(dc->ctx, 0);
+	link_res->hpo_dp_link_enc->funcs->enable_link_phy(
+			link_res->hpo_dp_link_enc,
+			link_settings,
+			link->link_enc->transmitter,
+			link->link_enc->hpd_source);
 }
 
 static void disable_hpo_dp_link_output(struct dc_link *link,
 		const struct link_resource *link_res,
 		enum signal_type signal)
 {
-	if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment)) {
-		disable_hpo_dp_fpga_link_output(link, link_res, signal);
-	} else {
 		link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc);
 		link_res->hpo_dp_link_enc->funcs->disable_link_phy(
 				link_res->hpo_dp_link_enc, signal);
-	}
 }
 
 static void set_hpo_dp_link_test_pattern(struct dc_link *link,
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 2267fb0978305..2963267fe74ab 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -765,7 +765,7 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable)
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	bool result = false;
 
-	if (dc_is_virtual_signal(stream->signal) || IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
+	if (dc_is_virtual_signal(stream->signal))
 		result = true;
 	else
 		result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable);
@@ -778,7 +778,6 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable)
 void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 {
 	struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc;
-	struct dc *dc = pipe_ctx->stream->ctx->dc;
 	struct dc_stream_state *stream = pipe_ctx->stream;
 	struct pipe_ctx *odm_pipe;
 	int opp_cnt = 1;
@@ -816,8 +815,7 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 		optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED;
 
 		/* Enable DSC in encoder */
-		if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)
-				&& !dp_is_128b_132b_signal(pipe_ctx)) {
+		if (dc_is_dp_signal(stream->signal) && !dp_is_128b_132b_signal(pipe_ctx)) {
 			DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id);
 			dsc_optc_config_log(dsc, &dsc_optc_cfg);
 			pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc,
@@ -849,7 +847,7 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 										false,
 										NULL,
 										true);
-			else if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+			else {
 				pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(
 						pipe_ctx->stream_res.stream_enc,
 						OPTC_DSC_DISABLED, 0, 0);
@@ -2271,8 +2269,7 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 		}
 	}
 
-	if (!IS_DIAG_DC(dc->ctx->dce_environment) &&
-			dc_is_virtual_signal(pipe_ctx->stream->signal))
+	if (dc_is_virtual_signal(pipe_ctx->stream->signal))
 		return;
 
 	if (!pipe_ctx->stream->sink->edid_caps.panel_patch.skip_avmute) {
@@ -2358,6 +2355,8 @@ void link_set_dpms_on(
 	enum otg_out_mux_dest otg_out_dest = OUT_MUX_DIO;
 	struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg;
 	const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res);
+	bool apply_edp_fast_boot_optimization =
+		pipe_ctx->stream->apply_edp_fast_boot_optimization;
 
 	ASSERT(is_master_pipe_for_link(link, pipe_ctx));
 
@@ -2375,9 +2374,8 @@ void link_set_dpms_on(
 		}
 	}
 
-	if (!IS_DIAG_DC(dc->ctx->dce_environment) &&
-			dc_is_virtual_signal(pipe_ctx->stream->signal))
-		return;
+		if (dc_is_virtual_signal(pipe_ctx->stream->signal))
+			return;
 
 	link_enc = link_enc_cfg_get_link_enc(link);
 	ASSERT(link_enc);
@@ -2402,138 +2400,126 @@ void link_set_dpms_on(
 
 	link_hwss->setup_stream_attribute(pipe_ctx);
 
-	if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
-		bool apply_edp_fast_boot_optimization =
-			pipe_ctx->stream->apply_edp_fast_boot_optimization;
-
-		pipe_ctx->stream->apply_edp_fast_boot_optimization = false;
+	pipe_ctx->stream->apply_edp_fast_boot_optimization = false;
 
-		// Enable VPG before building infoframe
-		if (vpg && vpg->funcs->vpg_poweron)
-			vpg->funcs->vpg_poweron(vpg);
+	// Enable VPG before building infoframe
+	if (vpg && vpg->funcs->vpg_poweron)
+		vpg->funcs->vpg_poweron(vpg);
 
-		resource_build_info_frame(pipe_ctx);
-		dc->hwss.update_info_frame(pipe_ctx);
+	resource_build_info_frame(pipe_ctx);
+	dc->hwss.update_info_frame(pipe_ctx);
 
-		if (dc_is_dp_signal(pipe_ctx->stream->signal))
-			dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME);
+	if (dc_is_dp_signal(pipe_ctx->stream->signal))
+		dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME);
 
-		/* Do not touch link on seamless boot optimization. */
-		if (pipe_ctx->stream->apply_seamless_boot_optimization) {
-			pipe_ctx->stream->dpms_off = false;
+	/* Do not touch link on seamless boot optimization. */
+	if (pipe_ctx->stream->apply_seamless_boot_optimization) {
+		pipe_ctx->stream->dpms_off = false;
 
-			/* Still enable stream features & audio on seamless boot for DP external displays */
-			if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT) {
-				enable_stream_features(pipe_ctx);
-				dc->hwss.enable_audio_stream(pipe_ctx);
-			}
-
-			update_psp_stream_config(pipe_ctx, false);
-			return;
-		}
-
-		/* eDP lit up by bios already, no need to enable again. */
-		if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP &&
-					apply_edp_fast_boot_optimization &&
-					!pipe_ctx->stream->timing.flags.DSC &&
-					!pipe_ctx->next_odm_pipe) {
-			pipe_ctx->stream->dpms_off = false;
-			update_psp_stream_config(pipe_ctx, false);
-			return;
+		/* Still enable stream features & audio on seamless boot for DP external displays */
+		if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT) {
+			enable_stream_features(pipe_ctx);
+			dc->hwss.enable_audio_stream(pipe_ctx);
 		}
 
-		if (pipe_ctx->stream->dpms_off)
-			return;
+		update_psp_stream_config(pipe_ctx, false);
+		return;
+	}
 
-		/* Have to setup DSC before DIG FE and BE are connected (which happens before the
-		 * link training). This is to make sure the bandwidth sent to DIG BE won't be
-		 * bigger than what the link and/or DIG BE can handle. VBID[6]/CompressedStream_flag
-		 * will be automatically set at a later time when the video is enabled
-		 * (DP_VID_STREAM_EN = 1).
-		 */
-		if (pipe_ctx->stream->timing.flags.DSC) {
-			if (dc_is_dp_signal(pipe_ctx->stream->signal) ||
-				dc_is_virtual_signal(pipe_ctx->stream->signal))
-			link_set_dsc_enable(pipe_ctx, true);
+	/* eDP lit up by bios already, no need to enable again. */
+	if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP &&
+				apply_edp_fast_boot_optimization &&
+				!pipe_ctx->stream->timing.flags.DSC &&
+				!pipe_ctx->next_odm_pipe) {
+		pipe_ctx->stream->dpms_off = false;
+		update_psp_stream_config(pipe_ctx, false);
+		return;
+	}
 
-		}
+	if (pipe_ctx->stream->dpms_off)
+		return;
 
-		status = enable_link(state, pipe_ctx);
+	/* Have to setup DSC before DIG FE and BE are connected (which happens before the
+	 * link training). This is to make sure the bandwidth sent to DIG BE won't be
+	 * bigger than what the link and/or DIG BE can handle. VBID[6]/CompressedStream_flag
+	 * will be automatically set at a later time when the video is enabled
+	 * (DP_VID_STREAM_EN = 1).
+	 */
+	if (pipe_ctx->stream->timing.flags.DSC) {
+		if (dc_is_dp_signal(pipe_ctx->stream->signal) ||
+			dc_is_virtual_signal(pipe_ctx->stream->signal))
+		link_set_dsc_enable(pipe_ctx, true);
 
-		if (status != DC_OK) {
-			DC_LOG_WARNING("enabling link %u failed: %d\n",
-			pipe_ctx->stream->link->link_index,
-			status);
+	}
 
-			/* Abort stream enable *unless* the failure was due to
-			 * DP link training - some DP monitors will recover and
-			 * show the stream anyway. But MST displays can't proceed
-			 * without link training.
-			 */
-			if (status != DC_FAIL_DP_LINK_TRAINING ||
-					pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
-				if (false == stream->link->link_status.link_active)
-					disable_link(stream->link, &pipe_ctx->link_res,
-							pipe_ctx->stream->signal);
-				BREAK_TO_DEBUGGER();
-				return;
-			}
-		}
+	status = enable_link(state, pipe_ctx);
 
-		/* turn off otg test pattern if enable */
-		if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
-			pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
-					CONTROLLER_DP_TEST_PATTERN_VIDEOMODE,
-					COLOR_DEPTH_UNDEFINED);
+	if (status != DC_OK) {
+		DC_LOG_WARNING("enabling link %u failed: %d\n",
+		pipe_ctx->stream->link->link_index,
+		status);
 
-		/* This second call is needed to reconfigure the DIG
-		 * as a workaround for the incorrect value being applied
-		 * from transmitter control.
+		/* Abort stream enable *unless* the failure was due to
+		 * DP link training - some DP monitors will recover and
+		 * show the stream anyway. But MST displays can't proceed
+		 * without link training.
 		 */
-		if (!(dc_is_virtual_signal(pipe_ctx->stream->signal) ||
-				dp_is_128b_132b_signal(pipe_ctx))) {
-				if (link_enc)
-					link_enc->funcs->setup(
-						link_enc,
+		if (status != DC_FAIL_DP_LINK_TRAINING ||
+				pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
+			if (false == stream->link->link_status.link_active)
+				disable_link(stream->link, &pipe_ctx->link_res,
 						pipe_ctx->stream->signal);
-			}
+			BREAK_TO_DEBUGGER();
+			return;
+		}
+	}
 
-		dc->hwss.enable_stream(pipe_ctx);
+	/* turn off otg test pattern if enable */
+	if (pipe_ctx->stream_res.tg->funcs->set_test_pattern)
+		pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg,
+				CONTROLLER_DP_TEST_PATTERN_VIDEOMODE,
+				COLOR_DEPTH_UNDEFINED);
 
-		/* Set DPS PPS SDP (AKA "info frames") */
-		if (pipe_ctx->stream->timing.flags.DSC) {
-			if (dc_is_dp_signal(pipe_ctx->stream->signal) ||
-					dc_is_virtual_signal(pipe_ctx->stream->signal)) {
-				dp_set_dsc_on_rx(pipe_ctx, true);
-				link_set_dsc_pps_packet(pipe_ctx, true, true);
-			}
+	/* This second call is needed to reconfigure the DIG
+	 * as a workaround for the incorrect value being applied
+	 * from transmitter control.
+	 */
+	if (!(dc_is_virtual_signal(pipe_ctx->stream->signal) ||
+			dp_is_128b_132b_signal(pipe_ctx))) {
+			if (link_enc)
+				link_enc->funcs->setup(
+					link_enc,
+					pipe_ctx->stream->signal);
 		}
 
-		if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
-			allocate_mst_payload(pipe_ctx);
-		else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT &&
-				dp_is_128b_132b_signal(pipe_ctx))
-			update_sst_payload(pipe_ctx, true);
+	dc->hwss.enable_stream(pipe_ctx);
+
+	/* Set DPS PPS SDP (AKA "info frames") */
+	if (pipe_ctx->stream->timing.flags.DSC) {
+		if (dc_is_dp_signal(pipe_ctx->stream->signal) ||
+				dc_is_virtual_signal(pipe_ctx->stream->signal)) {
+			dp_set_dsc_on_rx(pipe_ctx, true);
+			link_set_dsc_pps_packet(pipe_ctx, true, true);
+		}
+	}
 
-		dc->hwss.unblank_stream(pipe_ctx,
-			&pipe_ctx->stream->link->cur_link_settings);
+	if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
+		allocate_mst_payload(pipe_ctx);
+	else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT &&
+			dp_is_128b_132b_signal(pipe_ctx))
+		update_sst_payload(pipe_ctx, true);
 
-		if (stream->sink_patches.delay_ignore_msa > 0)
-			msleep(stream->sink_patches.delay_ignore_msa);
+	dc->hwss.unblank_stream(pipe_ctx,
+		&pipe_ctx->stream->link->cur_link_settings);
 
-		if (dc_is_dp_signal(pipe_ctx->stream->signal))
-			enable_stream_features(pipe_ctx);
-		update_psp_stream_config(pipe_ctx, false);
+	if (stream->sink_patches.delay_ignore_msa > 0)
+		msleep(stream->sink_patches.delay_ignore_msa);
 
-		dc->hwss.enable_audio_stream(pipe_ctx);
+	if (dc_is_dp_signal(pipe_ctx->stream->signal))
+		enable_stream_features(pipe_ctx);
+	update_psp_stream_config(pipe_ctx, false);
 
-	} else { // if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment))
-		if (dp_is_128b_132b_signal(pipe_ctx))
-			dp_fpga_hpo_enable_link_and_stream(state, pipe_ctx);
-		if (dc_is_dp_signal(pipe_ctx->stream->signal) ||
-				dc_is_virtual_signal(pipe_ctx->stream->signal))
-			link_set_dsc_enable(pipe_ctx, true);
-	}
+	dc->hwss.enable_audio_stream(pipe_ctx);
 
 	if (dc_is_hdmi_signal(pipe_ctx->stream->signal)) {
 		set_avmute(pipe_ctx, false);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index f46864630506f..3a5e80b577116 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -326,8 +326,7 @@ bool dp_is_fec_supported(const struct dc_link *link)
 
 	return (dc_is_dp_signal(link->connector_signal) && link_enc &&
 			link_enc->features.fec_supported &&
-			link->dpcd_caps.fec_cap.bits.FEC_CAPABLE &&
-			!IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment));
+			link->dpcd_caps.fec_cap.bits.FEC_CAPABLE);
 }
 
 bool dp_should_enable_fec(const struct dc_link *link)
-- 
GitLab


From 268182606f26434c5d3ebd0e86efcb0418dec487 Mon Sep 17 00:00:00 2001
From: Cruise Hung <cruise.hung@amd.com>
Date: Fri, 12 May 2023 23:33:46 +0800
Subject: [PATCH 1401/1791] drm/amd/display: Update correct DCN314 register
 header

[Why]
The register header for DCN314 is not correct.

[How]
Update correct DCN314 register header.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Cruise Hung <cruise.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/src/Makefile |  2 +-
 .../drm/amd/display/dmub/src/dmub_dcn314.c    | 62 +++++++++++++++++++
 .../drm/amd/display/dmub/src/dmub_dcn314.h    | 33 ++++++++++
 .../gpu/drm/amd/display/dmub/src/dmub_srv.c   |  5 +-
 4 files changed, 100 insertions(+), 2 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c
 create mode 100644 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h

diff --git a/drivers/gpu/drm/amd/display/dmub/src/Makefile b/drivers/gpu/drm/amd/display/dmub/src/Makefile
index 0589ad4778eea..caf095aca8f3f 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/Makefile
+++ b/drivers/gpu/drm/amd/display/dmub/src/Makefile
@@ -22,7 +22,7 @@
 
 DMUB = dmub_srv.o dmub_srv_stat.o dmub_reg.o dmub_dcn20.o dmub_dcn21.o
 DMUB += dmub_dcn30.o dmub_dcn301.o dmub_dcn302.o dmub_dcn303.o
-DMUB += dmub_dcn31.o dmub_dcn315.o dmub_dcn316.o
+DMUB += dmub_dcn31.o dmub_dcn314.o dmub_dcn315.o dmub_dcn316.o
 DMUB += dmub_dcn32.o
 
 AMD_DAL_DMUB = $(addprefix $(AMDDALPATH)/dmub/src/,$(DMUB))
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c
new file mode 100644
index 0000000000000..48a06dbd9be78
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c
@@ -0,0 +1,62 @@
+/*
+ * Copyright 2021 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include "../dmub_srv.h"
+#include "dmub_reg.h"
+#include "dmub_dcn314.h"
+
+#include "dcn/dcn_3_1_4_offset.h"
+#include "dcn/dcn_3_1_4_sh_mask.h"
+
+#define DCN_BASE__INST0_SEG0                       0x00000012
+#define DCN_BASE__INST0_SEG1                       0x000000C0
+#define DCN_BASE__INST0_SEG2                       0x000034C0
+#define DCN_BASE__INST0_SEG3                       0x00009000
+#define DCN_BASE__INST0_SEG4                       0x02403C00
+#define DCN_BASE__INST0_SEG5                       0
+
+#define BASE_INNER(seg) DCN_BASE__INST0_SEG##seg
+#define CTX dmub
+#define REGS dmub->regs_dcn31
+#define REG_OFFSET_EXP(reg_name) (BASE(reg##reg_name##_BASE_IDX) + reg##reg_name)
+
+/* Registers. */
+
+const struct dmub_srv_dcn31_regs dmub_srv_dcn314_regs = {
+#define DMUB_SR(reg) REG_OFFSET_EXP(reg),
+	{
+		DMUB_DCN31_REGS()
+		DMCUB_INTERNAL_REGS()
+	},
+#undef DMUB_SR
+
+#define DMUB_SF(reg, field) FD_MASK(reg, field),
+	{ DMUB_DCN31_FIELDS() },
+#undef DMUB_SF
+
+#define DMUB_SF(reg, field) FD_SHIFT(reg, field),
+	{ DMUB_DCN31_FIELDS() },
+#undef DMUB_SF
+};
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h
new file mode 100644
index 0000000000000..674267a2940e9
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright 2021 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#ifndef _DMUB_DCN314_H_
+#define _DMUB_DCN314_H_
+
+#include "dmub_dcn31.h"
+
+extern const struct dmub_srv_dcn31_regs dmub_srv_dcn314_regs;
+
+#endif /* _DMUB_DCN314_H_ */
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index ea3bed70a2297..9e9a6a44a7ac4 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -32,6 +32,7 @@
 #include "dmub_dcn302.h"
 #include "dmub_dcn303.h"
 #include "dmub_dcn31.h"
+#include "dmub_dcn314.h"
 #include "dmub_dcn315.h"
 #include "dmub_dcn316.h"
 #include "dmub_dcn32.h"
@@ -225,7 +226,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic)
 	case DMUB_ASIC_DCN314:
 	case DMUB_ASIC_DCN315:
 	case DMUB_ASIC_DCN316:
-		if (asic == DMUB_ASIC_DCN315)
+		if (asic == DMUB_ASIC_DCN314)
+			dmub->regs_dcn31 = &dmub_srv_dcn314_regs;
+		else if (asic == DMUB_ASIC_DCN315)
 			dmub->regs_dcn31 = &dmub_srv_dcn315_regs;
 		else if (asic == DMUB_ASIC_DCN316)
 			dmub->regs_dcn31 = &dmub_srv_dcn316_regs;
-- 
GitLab


From 2da3556c8650798606c0d3f2288b2f87c6665a69 Mon Sep 17 00:00:00 2001
From: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Date: Thu, 11 May 2023 15:16:35 -0400
Subject: [PATCH 1402/1791] drm/amd/display: Trigger DIO FIFO resync on commit
 streams for DCN32

[WHY and HOW]
Currently, on DCN32 we have an old workaround to resolve a DIO FIFO
speed issue when writing to the OTG DIVIDER register. However, this
workaround is not safe as we should be applying the DIO FIFO rampup
logic when the OTG re disabled along with the encoders. This new
workaround accounts for this. If the workaround sequence is incorrect,
like it is was, there is a chance we might hang. this new
workaround was first implemented in DCN314.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  |  2 ++
 .../gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c | 21 ++++----------
 .../gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h |  5 ++--
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    | 29 +++++++++++++++++++
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.h    |  2 ++
 .../gpu/drm/amd/display/dc/dcn32/dcn32_init.c |  1 +
 6 files changed, 42 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 70fac2ebb7576..46b6f4f9e1fd1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -412,6 +412,8 @@ void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc
 	hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
 		if (otg_disabled[i])
 			pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg);
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
index ffbb739d85b69..11e28e056cf7c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
@@ -42,18 +42,14 @@
 #define DC_LOGGER \
 	dccg->ctx->logger
 
-/* This function is a workaround for writing to OTG_PIXEL_RATE_DIV
- * without the probability of causing a DIG FIFO error.
- */
-static void dccg32_wait_for_dentist_change_done(
+static void dccg32_trigger_dio_fifo_resync(
 	struct dccg *dccg)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	uint32_t dispclk_rdivider_value = 0;
 
-	uint32_t dentist_dispclk_value = REG_READ(DENTIST_DISPCLK_CNTL);
-
-	REG_WRITE(DENTIST_DISPCLK_CNTL, dentist_dispclk_value);
-	REG_WAIT(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, 1, 50, 2000);
+	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value);
+	REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value);
 }
 
 static void dccg32_get_pixel_rate_div(
@@ -124,29 +120,21 @@ static void dccg32_set_pixel_rate_div(
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG0_PIXEL_RATE_DIVK1, k1,
 				OTG0_PIXEL_RATE_DIVK2, k2);
-
-		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 1:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG1_PIXEL_RATE_DIVK1, k1,
 				OTG1_PIXEL_RATE_DIVK2, k2);
-
-		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 2:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG2_PIXEL_RATE_DIVK1, k1,
 				OTG2_PIXEL_RATE_DIVK2, k2);
-
-		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	case 3:
 		REG_UPDATE_2(OTG_PIXEL_RATE_DIV,
 				OTG3_PIXEL_RATE_DIVK1, k1,
 				OTG3_PIXEL_RATE_DIVK2, k2);
-
-		dccg32_wait_for_dentist_change_done(dccg);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
@@ -352,6 +340,7 @@ static const struct dccg_funcs dccg32_funcs = {
 	.otg_add_pixel = dccg32_otg_add_pixel,
 	.otg_drop_pixel = dccg32_otg_drop_pixel,
 	.set_pixel_rate_div = dccg32_set_pixel_rate_div,
+	.trigger_dio_fifo_resync = dccg32_trigger_dio_fifo_resync,
 };
 
 struct dccg *dccg32_create(
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h
index 8071ab98d7084..cf55087181222 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h
@@ -112,8 +112,9 @@
 	DCCG_SF(DTBCLK_P_CNTL, DTBCLK_P3_EN, mask_sh),\
 	DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, mask_sh),\
 	DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL, mask_sh),\
-	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)
-
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh)
 
 struct dccg *dccg32_create(
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 2de910e0ce75f..7f5cd8c8d49b7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -1175,6 +1175,35 @@ void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 				pix_per_cycle);
 }
 
+void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context)
+{
+	uint8_t i;
+	struct pipe_ctx *pipe = NULL;
+	bool otg_disabled[MAX_PIPES] = {false};
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe->top_pipe || pipe->prev_odm_pipe)
+			continue;
+
+		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+			pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg);
+			reset_sync_context_for_pipe(dc, context, i);
+			otg_disabled[i] = true;
+		}
+	}
+
+	hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg);
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (otg_disabled[i])
+			pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg);
+	}
+}
+
 void dcn32_unblank_stream(struct pipe_ctx *pipe_ctx,
 		struct dc_link_settings *link_settings)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
index 6694c1d14aa31..6dbe929cf599a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
@@ -75,6 +75,8 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 
 void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
 
+void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context);
+
 void dcn32_subvp_pipe_control_lock(struct dc *dc,
 		struct dc_state *context,
 		bool lock,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
index 6f9a165c1eab9..8356b31e1d9af 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
@@ -153,6 +153,7 @@ static const struct hwseq_private_funcs dcn32_private_funcs = {
 	.update_mall_sel = dcn32_update_mall_sel,
 	.calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values,
 	.set_pixels_per_cycle = dcn32_set_pixels_per_cycle,
+	.resync_fifo_dccg_dio = dcn32_resync_fifo_dccg_dio,
 	.is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy,
 };
 
-- 
GitLab


From 5b466b28fa943aa9441cd27a9a469e1330814299 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Date: Tue, 9 May 2023 13:59:36 -0600
Subject: [PATCH 1403/1791] drm/amd/display: Reorganize DCN30 Makefile

DCN30 has a lot of files in the Makefile, and adding each one next to
the other makes it hard to read and can increase the chance of merge
conflicts. This commit just reorganize the Makefile to put each file
associated with DCN30 in its own line.

Reviewed-by: Chris Park <chris.park@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn30/Makefile | 38 +++++++++++++------
 1 file changed, 26 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile
index b7c2ae9ddfda3..4a3e9e47b6b6f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile
@@ -1,16 +1,16 @@
-# 
+#
 # Copyright 2020 Advanced Micro Devices, Inc.
-# 
+#
 # Permission is hereby granted, free of charge, to any person obtaining a
 # copy of this software and associated documentation files (the "Software"),
 # to deal in the Software without restriction, including without limitation
 # the rights to use, copy, modify, merge, publish, distribute, sublicense,
 # and/or sell copies of the Software, and to permit persons to whom the
 # Software is furnished to do so, subject to the following conditions:
-# 
+#
 # The above copyright notice and this permission notice shall be included in
 # all copies or substantial portions of the Software.
-# 
+#
 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
@@ -18,17 +18,31 @@
 # OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
 # ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
 # OTHER DEALINGS IN THE SOFTWARE.
-# 
+#
 # Authors: AMD
-# 
-# 
+#
+#
+
+DCN30 := \
+	dcn30_init.o \
+	dcn30_hubbub.o \
+	dcn30_hubp.o \
+	dcn30_dpp.o \
+	dcn30_optc.o \
+	dcn30_dccg.o \
+	dcn30_hwseq.o \
+	dcn30_mpc.o dcn30_vpg.o \
+	dcn30_afmt.o \
+	dcn30_dio_stream_encoder.o \
+	dcn30_dwb.o \
+	dcn30_dpp_cm.o \
+	dcn30_dwb_cm.o \
+	dcn30_cm_common.o \
+	dcn30_mmhubbub.o \
+	dcn30_resource.o \
+	dcn30_dio_link_encoder.o
 
 
-DCN30 = dcn30_init.o dcn30_hubbub.o dcn30_hubp.o dcn30_dpp.o dcn30_optc.o \
-	dcn30_dccg.o dcn30_hwseq.o dcn30_mpc.o dcn30_vpg.o \
-	dcn30_afmt.o dcn30_dio_stream_encoder.o dcn30_dwb.o \
-	dcn30_dpp_cm.o dcn30_dwb_cm.o dcn30_cm_common.o dcn30_mmhubbub.o \
-	dcn30_dio_link_encoder.o dcn30_resource.o
 
 AMD_DAL_DCN30 = $(addprefix $(AMDDALPATH)/dc/dcn30/,$(DCN30))
 
-- 
GitLab


From 257e9891db0b961b79c9f0ca50c808a738000e70 Mon Sep 17 00:00:00 2001
From: Josip Pavic <josip.pavic@amd.com>
Date: Fri, 12 May 2023 09:56:05 -0400
Subject: [PATCH 1404/1791] drm/amd/display: cache trace buffer size

[Why & How]
Cache the trace buffer size retrieved from DMUB FW in the driver

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Josip Pavic <josip.pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h     | 1 +
 drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 7c9a2b34bd05c..e210cb082ebd9 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -428,6 +428,7 @@ struct dmub_srv {
 	enum dmub_asic asic;
 	void *user_ctx;
 	uint32_t fw_version;
+	uint32_t trace_buffer_size;
 	bool is_virtual;
 	struct dmub_fb scratch_mem_fb;
 	volatile const struct dmub_fw_state *fw_state;
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index 9e9a6a44a7ac4..8b9af18e84fe0 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -427,6 +427,8 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub,
 			dmub->fw_version = fw_info->fw_version;
 	}
 
+	dmub->trace_buffer_size = trace_buffer_size;
+
 	trace_buff->base = dmub_align(mail->top, 256);
 	trace_buff->top = trace_buff->base + dmub_align(trace_buffer_size, 64);
 
-- 
GitLab


From 0d1ff99a3398ad4b7165ecd8e69d360090b32250 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Mon, 15 May 2023 00:14:00 -0400
Subject: [PATCH 1405/1791] drm/amd/display: 3.2.237

This version brings along following fixes:
- Improve the message printed
- Disable dcn315 pixel rate crb when scaling
- Update SR watermarks for DCN314
- Fix dcn315 pixel rate crb scaling check
- Reset CRTC State Before Restore from S2idle
- Have Payload Properly Created After Resume
- Trigger DIO FIFO resync on commit streams
- Revert vblank change that causes null pointer crash
- Fix possible underflow for displays with large vblank
- Apply 60us prefetch for DCFCLK <= 300Mhz
- Update correct DCN314 register header

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7fc087f85d398..7fd9f5a9e1915 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.236"
+#define DC_VER "3.2.237"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From 40e39d72277fc014e7b8149def35831998c8df2f Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 17:48:12 +0530
Subject: [PATCH 1406/1791] drm/amdgpu: Fix unused amdgpu_acpi_get_numa_info
 function in amdgpu_acpi_get_node_id()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the below compiler complaining error:

drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:860:33: error: unused function 'amdgpu_acpi_get_numa_info' [-Werror,-Wunused-function]
static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm)
                                ^
1 error generated.

By guarding amdgpu_acpi_get_numa_info & amdgpu_acpi_get_numa_size
function, only when CONFIG_ACPI_NUMA is enabled.

Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Lijo Lazar <lijo.lazar@amd.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 873532c4adbe3..02a66844e73e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -841,6 +841,7 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta
 	return r;
 }
 
+#ifdef CONFIG_ACPI_NUMA
 static inline uint64_t amdgpu_acpi_get_numa_size(int nid)
 {
 	/* This is directly using si_meminfo_node implementation as the
@@ -886,6 +887,7 @@ static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm)
 
 	return numa_info;
 }
+#endif
 
 /**
  * amdgpu_acpi_get_node_id - obtain the NUMA node id for corresponding amdgpu
-- 
GitLab


From 1893549af62135c788a66e7ff27c81459c532fb2 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 10:25:28 +0530
Subject: [PATCH 1407/1791] drm/amdgpu: Fix uninitialized variable in
 gfxhub_v1_2_xcp_resume
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:657:6: error: variable 'ret' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
        if (!amdgpu_sriov_vf(adev))
            ^~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:660:9: note: uninitialized use occurs here
        return ret;
               ^~~
drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:657:2: note: remove the 'if' if its condition is always true
        if (!amdgpu_sriov_vf(adev))
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:648:9: note: initialize the variable 'ret' to silence this warning
        int ret;
               ^
                = 0
1 error generated.

Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Lijo Lazar <lijo.lazar@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
index 8901e73fd7005..4dabf910334b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c
@@ -645,7 +645,6 @@ static int gfxhub_v1_2_xcp_resume(void *handle, uint32_t inst_mask)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool value;
-	int ret;
 
 	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
 		value = false;
@@ -655,9 +654,9 @@ static int gfxhub_v1_2_xcp_resume(void *handle, uint32_t inst_mask)
 	gfxhub_v1_2_xcc_set_fault_enable_default(adev, value, inst_mask);
 
 	if (!amdgpu_sriov_vf(adev))
-		ret = gfxhub_v1_2_xcc_gart_enable(adev, inst_mask);
+		return gfxhub_v1_2_xcc_gart_enable(adev, inst_mask);
 
-	return ret;
+	return 0;
 }
 
 static int gfxhub_v1_2_xcp_suspend(void *handle, uint32_t inst_mask)
-- 
GitLab


From 332bb09352a69b8e7cf0825575f90581d3695135 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Fri, 19 May 2023 11:34:04 +0800
Subject: [PATCH 1408/1791] drm/amdgpu: remove unused definition

mmhub_v1_8_mmea_cgtt_clk_cntl_reg is defined but not used.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index 8600e42434e70..bf3b0f382c896 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -757,14 +757,6 @@ static void mmhub_v1_8_query_ras_error_status(struct amdgpu_device *adev)
 		mmhub_v1_8_inst_query_ras_err_status(adev, i);
 }
 
-static const uint32_t mmhub_v1_8_mmea_cgtt_clk_cntl_reg[] = {
-	regMMEA0_CGTT_CLK_CTRL,
-	regMMEA1_CGTT_CLK_CTRL,
-	regMMEA2_CGTT_CLK_CTRL,
-	regMMEA3_CGTT_CLK_CTRL,
-	regMMEA4_CGTT_CLK_CTRL,
-};
-
 static void mmhub_v1_8_inst_reset_ras_err_status(struct amdgpu_device *adev,
 						 uint32_t mmhub_inst)
 {
-- 
GitLab


From 89fb3020d68d46807e1341ad8acae53cdf197234 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Tue, 16 May 2023 09:55:38 +0800
Subject: [PATCH 1409/1791] drm/amdgpu: init the XCC_DOORBELL_FENCE regs

add the the init_registers callback for nbio_v7_9

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index 755259e96bbca..ad70086de9b5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -424,6 +424,22 @@ nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev, u32 *supp_modes)
 	return ffs(tmp);
 }
 
+static void nbio_v7_9_init_registers(struct amdgpu_device *adev)
+{
+	u32 inst_mask;
+	int i;
+
+	WREG32_SOC15(NBIO, 0, regXCC_DOORBELL_FENCE,
+		0xff & ~(adev->gfx.xcc_mask));
+
+	inst_mask = adev->aid_mask & ~1U;
+	for_each_inst(i, inst_mask) {
+		WREG32_SOC15_EXT(NBIO, i, regXCC_DOORBELL_FENCE, i,
+			XCC_DOORBELL_FENCE__SHUB_SLV_MODE_MASK);
+
+	}
+}
+
 const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_hdp_flush_req_offset = nbio_v7_9_get_hdp_flush_req_offset,
 	.get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset,
@@ -447,4 +463,5 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode,
 	.set_compute_partition_mode = nbio_v7_9_set_compute_partition_mode,
 	.get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode,
+	.init_registers = nbio_v7_9_init_registers,
 };
-- 
GitLab


From a64f7eb026ea7205bf9dfadabd746e47c5717b5d Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Thu, 18 May 2023 15:52:08 +0200
Subject: [PATCH 1410/1791] drm/radeon: stop including swiotlb.h

radeon does not need swiotlb.h, so stop including it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_ttm.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 06a53ecc04a25..10794be302390 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -36,7 +36,6 @@
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
-#include <linux/swiotlb.h>
 
 #include <drm/drm_device.h>
 #include <drm/drm_file.h>
-- 
GitLab


From 1becc57cd1a905e2aa0e1eca60d2a37744525c4a Mon Sep 17 00:00:00 2001
From: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Date: Fri, 19 May 2023 08:33:27 -0700
Subject: [PATCH 1411/1791] drm/radeon: fix possible division-by-zero errors

Function rv740_get_decoded_reference_divider() may return 0 due to
unpredictable reference divider value calculated in
radeon_atom_get_clock_dividers(). This will lead to
division-by-zero error once that value is used as a divider
in calculating 'clk_s'.
While unlikely, this issue should nonetheless be prevented so add a
sanity check for such cases by testing 'decoded_ref' value against 0.

Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.

v2: minor coding style fixes (Alex)
In practice this should actually happen as the vbios should be
properly populated.

Fixes: 66229b200598 ("drm/radeon/kms: add dpm support for rv7xx (v4)")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/cypress_dpm.c | 8 ++++++--
 drivers/gpu/drm/radeon/ni_dpm.c      | 8 ++++++--
 drivers/gpu/drm/radeon/rv740_dpm.c   | 8 ++++++--
 3 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index fdddbbaecbb74..72a0768df00f7 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -557,8 +557,12 @@ static int cypress_populate_mclk_value(struct radeon_device *rdev,
 						     ASIC_INTERNAL_MEMORY_SS, vco_freq)) {
 			u32 reference_clock = rdev->clock.mpll.reference_freq;
 			u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div);
-			u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
-			u32 clk_v = ss.percentage *
+			u32 clk_s, clk_v;
+
+			if (!decoded_ref)
+				return -EINVAL;
+			clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
+			clk_v = ss.percentage *
 				(0x4000 * dividers.whole_fb_div + 0x800 * dividers.frac_fb_div) / (clk_s * 625);
 
 			mpll_ss1 &= ~CLKV_MASK;
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index 672d2239293e0..3e1c1a392fb7b 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -2241,8 +2241,12 @@ static int ni_populate_mclk_value(struct radeon_device *rdev,
 						     ASIC_INTERNAL_MEMORY_SS, vco_freq)) {
 			u32 reference_clock = rdev->clock.mpll.reference_freq;
 			u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div);
-			u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
-			u32 clk_v = ss.percentage *
+			u32 clk_s, clk_v;
+
+			if (!decoded_ref)
+				return -EINVAL;
+			clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
+			clk_v = ss.percentage *
 				(0x4000 * dividers.whole_fb_div + 0x800 * dividers.frac_fb_div) / (clk_s * 625);
 
 			mpll_ss1 &= ~CLKV_MASK;
diff --git a/drivers/gpu/drm/radeon/rv740_dpm.c b/drivers/gpu/drm/radeon/rv740_dpm.c
index d57a3e1df8d63..4464fd21a3029 100644
--- a/drivers/gpu/drm/radeon/rv740_dpm.c
+++ b/drivers/gpu/drm/radeon/rv740_dpm.c
@@ -249,8 +249,12 @@ int rv740_populate_mclk_value(struct radeon_device *rdev,
 						     ASIC_INTERNAL_MEMORY_SS, vco_freq)) {
 			u32 reference_clock = rdev->clock.mpll.reference_freq;
 			u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div);
-			u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
-			u32 clk_v = 0x40000 * ss.percentage *
+			u32 clk_s, clk_v;
+
+			if (!decoded_ref)
+				return -EINVAL;
+			clk_s = reference_clock * 5 / (decoded_ref * ss.rate);
+			clk_v = 0x40000 * ss.percentage *
 				(dividers.whole_fb_div + (dividers.frac_fb_div / 8)) / (clk_s * 10000);
 
 			mpll_ss1 &= ~CLKV_MASK;
-- 
GitLab


From d4281b49c1b6afc66c470c85019fc1eceb676a78 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Fri, 19 May 2023 14:20:17 +0800
Subject: [PATCH 1412/1791] drm/amd/pm: add missing NotifyPowerSource message
 mapping for SMU13.0.7

Otherwise, the power source switching will fail due to message
unavailable.

Fixes: bf4823267a81 ("drm/amd/pm: fix possible power mode mismatch between driver and PMFW")
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 1b6116cf8b4cd..cf6827179fd13 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -125,6 +125,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(ArmD3,				PPSMC_MSG_ArmD3,                       0),
 	MSG_MAP(AllowGpo,			PPSMC_MSG_SetGpoAllow,           0),
 	MSG_MAP(GetPptLimit,			PPSMC_MSG_GetPptLimit,                 0),
+	MSG_MAP(NotifyPowerSource,		PPSMC_MSG_NotifyPowerSource,           0),
 };
 
 static struct cmn2asic_mapping smu_v13_0_7_clk_map[SMU_CLK_COUNT] = {
-- 
GitLab


From a3ffabb25077059427434368a1c65c176a0f93d0 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 19 Apr 2023 16:02:44 +0530
Subject: [PATCH 1413/1791] drm/amdgpu: Disable interrupt tracker on NBIOv7.9

Enabling nBIF interrupt history tracker prevents LCLK deep sleep,
hence disable it

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c                        | 2 ++
 drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index ad70086de9b5d..e082f6343d209 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -432,6 +432,8 @@ static void nbio_v7_9_init_registers(struct amdgpu_device *adev)
 	WREG32_SOC15(NBIO, 0, regXCC_DOORBELL_FENCE,
 		0xff & ~(adev->gfx.xcc_mask));
 
+	WREG32_SOC15(NBIO, 0, regBIFC_GFX_INT_MONITOR_MASK, 0x7ff);
+
 	inst_mask = adev->aid_mask & ~1U;
 	for_each_inst(i, inst_mask) {
 		WREG32_SOC15_EXT(NBIO, i, regXCC_DOORBELL_FENCE, i,
diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h
index 033f2796c1e37..c8a15c8f48228 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h
@@ -6201,6 +6201,8 @@
 #define regNBIF_SHUB_TODET_SYNCFLOOD_CTRL2_BASE_IDX 8
 #define regBIFC_BME_ERR_LOG_HB                                                                          0xe8ab
 #define regBIFC_BME_ERR_LOG_HB_BASE_IDX 8
+#define regBIFC_GFX_INT_MONITOR_MASK                                                                    0xe8ad
+#define regBIFC_GFX_INT_MONITOR_MASK_BASE_IDX 8
 #define regBIFC_HRP_SDP_WRRSP_POOLCRED_ALLOC                                                            0xe8c0
 #define regBIFC_HRP_SDP_WRRSP_POOLCRED_ALLOC_BASE_IDX 8
 #define regBIFC_HRP_SDP_RDRSP_POOLCRED_ALLOC                                                            0xe8c1
-- 
GitLab


From ab22ecabe99922db4bec8e2b439336f865bbc117 Mon Sep 17 00:00:00 2001
From: Jonatas Esteves <jntesteves@gmail.com>
Date: Sat, 20 May 2023 10:39:52 -0300
Subject: [PATCH 1414/1791] drm/amd/pm: Fix output of pp_od_clk_voltage

Printing the other clock types should not be conditioned on being able
to print OD_SCLK. Some GPUs currently have limited capability of only
printing a subset of these.

Since this condition was introduced in v5.18-rc1, reading from
`pp_od_clk_voltage` has been returning empty on the Asus ROG Strix G15
(2021).

Fixes: 79c65f3fcbb1 ("drm/amd/pm: do not expose power implementation details to amdgpu_pm.c")
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Jonatas Esteves <jntesteves@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 40100b77b2d9a..da0da03569e85 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -871,13 +871,11 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev,
 	}
 	if (ret == -ENOENT) {
 		size = amdgpu_dpm_print_clock_levels(adev, OD_SCLK, buf);
-		if (size > 0) {
-			size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf + size);
-			size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf + size);
-			size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf + size);
-			size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf + size);
-			size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf + size);
-		}
+		size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf + size);
+		size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf + size);
+		size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf + size);
+		size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf + size);
+		size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf + size);
 	}
 
 	if (size == 0)
-- 
GitLab


From 9f77af014cbc3b77a2f5b8cbce8262ff97e94aa7 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Mon, 22 May 2023 10:02:52 +0100
Subject: [PATCH 1415/1791] drm/amdgpu: Fix a couple of spelling mistakes in
 info and debug messages

There are a couple of spelling mistakes, one in a dev_info message
and the other in a dev_debug message. Fix them.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 59b8b26e2caf2..789cc16e1be7e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1007,7 +1007,7 @@ int psp_spatial_partition(struct psp_context *psp, int mode)
 	cmd->cmd_id = GFX_CMD_ID_SRIOV_SPATIAL_PART;
 	cmd->cmd.cmd_spatial_part.mode = mode;
 
-	dev_info(psp->adev->dev, "Requesting %d paritions through PSP", mode);
+	dev_info(psp->adev->dev, "Requesting %d partitions through PSP", mode);
 	ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr);
 
 	release_psp_cmd_buf(psp);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index daeb6bcc9245a..e9586a0dc335d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -359,7 +359,7 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev,
 						file_priv->minor->index, i);
 				return -ENOENT;
 			}
-			dev_dbg(adev->dev, "renderD%d partition %d openned!",
+			dev_dbg(adev->dev, "renderD%d partition %d opened!",
 					file_priv->minor->index, i);
 			fpriv->xcp_id = i;
 			break;
-- 
GitLab


From 6dabce860d40703d7c27b71a120317f09293cf9c Mon Sep 17 00:00:00 2001
From: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Date: Mon, 22 May 2023 00:30:15 -0700
Subject: [PATCH 1416/1791] drm/amdgpu: Fix unsigned comparison with zero in
 gmc_v9_0_process_interrupt()

Smatch warns:
	drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c:579:
	unsigned 'xcc_id' is never less than zero.

gfx_v9_4_3_ih_to_xcc_inst() returns negative numbers as well.
Fix this by changing type of xcc_id to int.

Fixes: 98b2e9cad227 ("drm/amdgpu: correct the vmhub index when page fault occurs")
Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index f70e666cecf24..1e8b2aaa48c16 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -557,8 +557,8 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
 	const char *hub_name;
 	u64 addr;
 	uint32_t cam_index = 0;
-	int ret;
-	uint32_t node_id, xcc_id = 0;
+	int ret, xcc_id = 0;
+	uint32_t node_id;
 
 	node_id = entry->node_id;
 
-- 
GitLab


From 87f4c2d9205c6646b25081581e810a05cc9d2799 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 22 May 2023 13:50:28 +0200
Subject: [PATCH 1417/1791] drm/amd/pm: mark irq functions as 'static'

Two newly added functions cause a warning because they lack a prototype:

drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_6_ppt.c:1328:5: error: no previous prototype for 'smu_v13_0_6_set_irq_state' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_6_ppt.c:1368:5: error: no previous prototype for 'smu_v13_0_6_register_irq_handler' [-Werror=missing-prototypes]

They are only used locally, so just mark them static.

Fixes: 676915e4108f ("drm/amd/pm: Add ih for SMU v13.0.6 thermal throttling")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index a712b2bf2d25a..41b49cc827cd5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1325,7 +1325,7 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
 	return 0;
 }
 
-int smu_v13_0_6_set_irq_state(struct amdgpu_device *adev,
+static int smu_v13_0_6_set_irq_state(struct amdgpu_device *adev,
 			      struct amdgpu_irq_src *source,
 			      unsigned tyep,
 			      enum amdgpu_interrupt_state state)
@@ -1365,7 +1365,7 @@ static const struct amdgpu_irq_src_funcs smu_v13_0_6_irq_funcs =
 	.process = smu_v13_0_6_irq_process,
 };
 
-int smu_v13_0_6_register_irq_handler(struct smu_context *smu)
+static int smu_v13_0_6_register_irq_handler(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
 	struct amdgpu_irq_src *irq_src = &smu->irq_source;
-- 
GitLab


From d522458e63136ccccea18077687ceff1d31527ca Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 22 May 2023 13:50:29 +0200
Subject: [PATCH 1418/1791] drm/amdkfd: mark local functions as static

The file was newly added and causes some -Wmissing-prototype warnings:

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:57:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_load' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:126:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_dump' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:163:6: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_is_occupied' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:181:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_destroy' [-Werror=missing-prototypes]

Mark these all as 'static' since there are no outside callers.

Fixes: a805889a1531 ("drm/amdkfd: Update SDMA queue management for GFX9.4.3")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 81dfbe39fd8ef..5b4b7f8b92a51 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -54,7 +54,7 @@ static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev,
 	return retval;
 }
 
-int kgd_gfx_v9_4_3_hqd_sdma_load(struct amdgpu_device *adev, void *mqd,
+static int kgd_gfx_v9_4_3_hqd_sdma_load(struct amdgpu_device *adev, void *mqd,
 				 uint32_t __user *wptr, struct mm_struct *mm)
 {
 	struct v9_sdma_mqd *m;
@@ -123,7 +123,7 @@ int kgd_gfx_v9_4_3_hqd_sdma_load(struct amdgpu_device *adev, void *mqd,
 	return 0;
 }
 
-int kgd_gfx_v9_4_3_hqd_sdma_dump(struct amdgpu_device *adev,
+static int kgd_gfx_v9_4_3_hqd_sdma_dump(struct amdgpu_device *adev,
 				 uint32_t engine_id, uint32_t queue_id,
 				 uint32_t (**dump)[2], uint32_t *n_regs)
 {
@@ -160,7 +160,7 @@ int kgd_gfx_v9_4_3_hqd_sdma_dump(struct amdgpu_device *adev,
 	return 0;
 }
 
-bool kgd_gfx_v9_4_3_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
+static bool kgd_gfx_v9_4_3_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 {
 	struct v9_sdma_mqd *m;
 	uint32_t sdma_rlc_reg_offset;
@@ -178,7 +178,7 @@ bool kgd_gfx_v9_4_3_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd)
 	return false;
 }
 
-int kgd_gfx_v9_4_3_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
+static int kgd_gfx_v9_4_3_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
 				    unsigned int utimeout)
 {
 	struct v9_sdma_mqd *m;
-- 
GitLab


From 1b177b5c6846f20be013b45c36c24264049c81bf Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 22 May 2023 13:50:30 +0200
Subject: [PATCH 1419/1791] drm/amdgpu:mark aqua_vanjaram_reg_init.c function
 as static

A few newly added global functions have no prototype, which causes warnings:

drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:169:5: error: no previous prototype for 'aqua_vanjaram_select_scheds' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:310:5: error: no previous prototype for '__aqua_vanjaram_get_xcc_per_xcp' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:337:5: error: no previous prototype for '__aqua_vanjaram_get_xcp_ip_info' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:593:5: error: no previous prototype for 'aqua_vanjaram_get_xcp_ip_details' [-Werror=missing-prototypes]

There are no callers from other files, so just mark them as 'static'.

Fixes: cd7d8400aa04 ("drm/amdgpu: add partition schedule for GC(9, 4, 3)")
Fixes: 9cb18287d8f1 ("drm/amdgpu: Add SOC partition funcs for GC v9.4.3")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 68d1a0fc5f5d9..a595bb958215e 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -166,7 +166,7 @@ static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev)
 	return aqua_vanjaram_xcp_sched_list_update(adev);
 }
 
-int aqua_vanjaram_select_scheds(
+static int aqua_vanjaram_select_scheds(
 		struct amdgpu_device *adev,
 		u32 hw_ip,
 		u32 hw_prio,
@@ -307,7 +307,7 @@ static int aqua_vanjaram_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr)
 	return mode;
 }
 
-int __aqua_vanjaram_get_xcc_per_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
+static int __aqua_vanjaram_get_xcc_per_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 {
 	int num_xcc, num_xcc_per_xcp = 0;
 
@@ -334,7 +334,7 @@ int __aqua_vanjaram_get_xcc_per_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int mode)
 	return num_xcc_per_xcp;
 }
 
-int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+static int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 				    enum AMDGPU_XCP_IP_BLOCK ip_id,
 				    struct amdgpu_xcp_ip *ip)
 {
@@ -590,7 +590,7 @@ static int aqua_vanjaram_get_xcp_mem_id(struct amdgpu_xcp_mgr *xcp_mgr,
 	return r;
 }
 
-int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
+static int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 				     enum AMDGPU_XCP_IP_BLOCK ip_id,
 				     struct amdgpu_xcp_ip *ip)
 {
-- 
GitLab


From 1f9bb94f128f7d13a67fbff5eca730cc2e8842a6 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 22 May 2023 13:50:31 +0200
Subject: [PATCH 1420/1791] drm/amdgpu: use %pad format string for dma_addr_t

DMA addresses can be shorter than u64, which results in a broken debug output:

drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c: In function 'amdgpu_gart_table_ram_alloc':
drivers/gpu/drm/amd/amdgpu/amdgpu.h:41:22: error: format '%llx' expects argument of type 'long long unsigned int', but argument 4 has type 'dma_addr_t' {aka 'unsigned int'} [-Werror=format=]
drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c:146:9: note: in expansion of macro 'dev_info'
  146 |         dev_info(adev->dev, "%s dma_addr:%llx\n", __func__, dma_addr);

Use the special %pad format string and pass the DMA address by reference.

Fixes: c9a502e981a9 ("drm/amdgpu: Allocate GART table in RAM for AMD APU")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index a070adf30c88c..73b8cca35bab8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -143,7 +143,7 @@ int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev)
 		return -EFAULT;
 	}
 
-	dev_info(adev->dev, "%s dma_addr:%llx\n", __func__, dma_addr);
+	dev_info(adev->dev, "%s dma_addr:%pad\n", __func__, &dma_addr);
 	/* Create SG table */
 	sg = kmalloc(sizeof(*sg), GFP_KERNEL);
 	if (!sg) {
-- 
GitLab


From 1501fe94eedd18243b84008aecc25f4f3c4fa48d Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 22 May 2023 13:50:32 +0200
Subject: [PATCH 1421/1791] drm/amdgpu: fix acpi build warnings

Two newly introduced functions are in the global namespace but have no prototypes
or callers outside of amdgpu_acpi.c, another function is static but only has
a caller inside of an #ifdef:

drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:902:13: error: no previous prototype for 'amdgpu_acpi_get_node_id' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:928:30: error: no previous prototype for 'amdgpu_acpi_get_dev' [-Werror=missing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:860:33: error: 'amdgpu_acpi_get_numa_info' defined but not used [-Werror=unused-function]

Avoid the warnings by marking all of them static and ensuring that the compiler is
able to see the callsites.

v2: rebase on latest code (Alex)

Fixes: fa0497c34eb7 ("drm/amdgpu: Add API to get numa information of XCC")
Fixes: 1cc823011a23 ("drm/amdgpu: Store additional numa node information")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 02a66844e73e2..3fb2c3af0998a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -901,7 +901,7 @@ static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm)
  *
  * Returns ACPI STATUS OK with Node ID on success or the corresponding failure reason
  */
-acpi_status amdgpu_acpi_get_node_id(acpi_handle handle,
+static acpi_status amdgpu_acpi_get_node_id(acpi_handle handle,
 				    struct amdgpu_numa_info **numa_info)
 {
 #ifdef CONFIG_ACPI_NUMA
@@ -927,7 +927,7 @@ acpi_status amdgpu_acpi_get_node_id(acpi_handle handle,
 #endif
 }
 
-struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u16 bdf)
+static struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u16 bdf)
 {
 	struct amdgpu_acpi_dev_info *acpi_dev;
 
-- 
GitLab


From 803e4c9efc79c96796efbecab9ed53267d051256 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Mon, 22 May 2023 08:30:13 -0400
Subject: [PATCH 1422/1791] drm/amdgpu: remove unused variable num_xcc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc with W=1 reports
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:2138:13: error: variable
  ‘num_xcc’ set but not used [-Werror=unused-but-set-variable]
 2138 |         int num_xcc;
      |             ^~~~~~~

This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 6a1a28df606c7..cb34da8967370 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2135,9 +2135,6 @@ static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring,
 static int gfx_v9_4_3_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int num_xcc;
-
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
-- 
GitLab


From 665d49c27eff01c91a155a37f025b981c2f73a3b Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Sun, 21 May 2023 09:24:00 +0800
Subject: [PATCH 1423/1791] drm/amd/pm: reverse mclk and fclk clocks levels for
 SMU v13.0.4

This patch reverses the DPM clocks levels output of pp_dpm_mclk
and pp_dpm_fclk.

On dGPUs and older APUs we expose the levels from lowest clocks
to highest clocks. But for some APUs, the clocks levels that from
the DFPstateTable are given the reversed orders by PMFW. Like the
memory DPM clocks that are exposed by pp_dpm_mclk.

It's not intuitive that they are reversed on these APUs. All tools
and software that talks to the driver then has to know different ways
to interpret the data depending on the asic.

So we need to reverse them to expose the clocks levels from the
driver consistently.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
index 0a0a7debb3ae7..46a8a366f2879 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
@@ -478,7 +478,7 @@ static int smu_v13_0_4_get_dpm_level_count(struct smu_context *smu,
 static int smu_v13_0_4_print_clk_levels(struct smu_context *smu,
 					enum smu_clk_type clk_type, char *buf)
 {
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	uint32_t min, max;
 
@@ -512,7 +512,8 @@ static int smu_v13_0_4_print_clk_levels(struct smu_context *smu,
 			break;
 
 		for (i = 0; i < count; i++) {
-			ret = smu_v13_0_4_get_dpm_freq_by_index(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = smu_v13_0_4_get_dpm_freq_by_index(smu, clk_type, idx, &value);
 			if (ret)
 				break;
 
-- 
GitLab


From d9ed111b76e3ebe1d15b7db746d498666a396de1 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Sun, 21 May 2023 10:28:05 +0800
Subject: [PATCH 1424/1791] drm/amd/pm: reverse mclk clocks levels for SMU
 v13.0.5

This patch reverses the DPM clocks levels output of pp_dpm_mclk.

On dGPUs and older APUs we expose the levels from lowest clocks
to highest clocks. But for some APUs, the clocks levels that from
the DFPstateTable are given the reversed orders by PMFW. Like the
memory DPM clocks that are exposed by pp_dpm_mclk.

It's not intuitive that they are reversed on these APUs. All tools
and software that talks to the driver then has to know different ways
to interpret the data depending on the asic.

So we need to reverse them to expose the clocks levels from the
driver consistently.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
index 165b2470b017d..7c3ac535f68a5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
@@ -866,7 +866,7 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu,
 static int smu_v13_0_5_print_clk_levels(struct smu_context *smu,
 				enum smu_clk_type clk_type, char *buf)
 {
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	uint32_t min = 0, max = 0;
 
@@ -898,7 +898,8 @@ static int smu_v13_0_5_print_clk_levels(struct smu_context *smu,
 			goto print_clk_out;
 
 		for (i = 0; i < count; i++) {
-			ret = smu_v13_0_5_get_dpm_freq_by_index(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = smu_v13_0_5_get_dpm_freq_by_index(smu, clk_type, idx, &value);
 			if (ret)
 				goto print_clk_out;
 
-- 
GitLab


From 06aade19bb2433001f4d10f1424a803b3f63734a Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Sun, 21 May 2023 10:35:59 +0800
Subject: [PATCH 1425/1791] drm/amd/pm: reverse mclk and fclk clocks levels for
 yellow carp

This patch reverses the DPM clocks levels output of pp_dpm_mclk
and pp_dpm_fclk.

On dGPUs and older APUs we expose the levels from lowest clocks
to highest clocks. But for some APUs, the clocks levels that from
the DFPstateTable are given the reversed orders by PMFW. Like the
memory DPM clocks that are exposed by pp_dpm_mclk.

It's not intuitive that they are reversed on these APUs. All tools
and software that talks to the driver then has to know different ways
to interpret the data depending on the asic.

So we need to reverse them to expose the clocks levels from the
driver consistently.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
index ac5fcca0e47f9..a92da336ecec6 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
@@ -1000,7 +1000,7 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu,
 static int yellow_carp_print_clk_levels(struct smu_context *smu,
 				enum smu_clk_type clk_type, char *buf)
 {
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	uint32_t min, max;
 
@@ -1033,7 +1033,8 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu,
 			goto print_clk_out;
 
 		for (i = 0; i < count; i++) {
-			ret = yellow_carp_get_dpm_freq_by_index(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = yellow_carp_get_dpm_freq_by_index(smu, clk_type, idx, &value);
 			if (ret)
 				goto print_clk_out;
 
-- 
GitLab


From 63b9acdf06200bb6537a3a479741b4cb52488a89 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Sun, 21 May 2023 11:10:19 +0800
Subject: [PATCH 1426/1791] drm/amd/pm: reverse mclk and fclk clocks levels for
 vangogh

This patch reverses the DPM clocks levels output of pp_dpm_mclk
and pp_dpm_fclk.

On dGPUs and older APUs we expose the levels from lowest clocks
to highest clocks. But for some APUs, the clocks levels that from
the DFPstateTable are given the reversed orders by PMFW. Like the
memory DPM clocks that are exposed by pp_dpm_mclk.

It's not intuitive that they are reversed on these APUs. All tools
and software that talks to the driver then has to know different ways
to interpret the data depending on the asic.

So we need to reverse them to expose the clocks levels from the
driver consistently.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
index 7433dcaa16e04..067b4e0b026c0 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c
@@ -582,7 +582,7 @@ static int vangogh_print_legacy_clk_levels(struct smu_context *smu,
 	DpmClocks_t *clk_table = smu->smu_table.clocks_table;
 	SmuMetrics_legacy_t metrics;
 	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	bool cur_value_match_level = false;
 
@@ -656,7 +656,8 @@ static int vangogh_print_legacy_clk_levels(struct smu_context *smu,
 	case SMU_MCLK:
 	case SMU_FCLK:
 		for (i = 0; i < count; i++) {
-			ret = vangogh_get_dpm_clk_limited(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = vangogh_get_dpm_clk_limited(smu, clk_type, idx, &value);
 			if (ret)
 				return ret;
 			if (!value)
@@ -683,7 +684,7 @@ static int vangogh_print_clk_levels(struct smu_context *smu,
 	DpmClocks_t *clk_table = smu->smu_table.clocks_table;
 	SmuMetrics_t metrics;
 	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	bool cur_value_match_level = false;
 	uint32_t min, max;
@@ -765,7 +766,8 @@ static int vangogh_print_clk_levels(struct smu_context *smu,
 	case SMU_MCLK:
 	case SMU_FCLK:
 		for (i = 0; i < count; i++) {
-			ret = vangogh_get_dpm_clk_limited(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = vangogh_get_dpm_clk_limited(smu, clk_type, idx, &value);
 			if (ret)
 				return ret;
 			if (!value)
-- 
GitLab


From acf429dcac1440169a3b28da784cbda72f3b678b Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Thu, 18 May 2023 15:08:16 -0400
Subject: [PATCH 1427/1791] drm/amdkfd: Align partition memory size to page
 size

The compute partition memory size calculated from KFD_XCP_MEMORY_SIZE
may not align to page size if xcp_mgr->num_xcp_per_mem_partition is 6.

Change the KFD_XCP_MEMORY_SIZE macro to return page align size, so KFD
node memory size reported in sysfs is page align size, to avoid
application VRAM allocation failure because application may use the size
directly and Thunk requires the memory allocation size is page size
align.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 5de92c9ab18f5..66f80b9ab0c5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -804,7 +804,7 @@ u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id)
 	if (adev->gmc.num_mem_partitions && xcp_id >= 0 && mem_id >= 0) {
 		tmp = adev->gmc.mem_partitions[mem_id].size;
 		do_div(tmp, adev->xcp_mgr->num_xcp_per_mem_partition);
-		return tmp;
+		return ALIGN_DOWN(tmp, PAGE_SIZE);
 	} else {
 		return adev->gmc.real_vram_size;
 	}
-- 
GitLab


From 9f173a80305d84f6b41bfb2a5482179e6e4957b1 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 23 May 2023 10:34:30 +0200
Subject: [PATCH 1428/1791] drm/amd/display: avoid calling missing
 .resync_fifo_dccg_dio()

The .resync_fifo_dccg_dio() callback pointer was added in an #ifdef block,
but is called unconditionally:

drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:2292:31: error: 'struct hwseq_private_funcs' has no member named 'resync_fifo_dccg_dio'

Add the same #ifdef around the caller as well.

Fixes: 3e8d74cb128f ("drm/amd/display: Trigger DIO FIFO resync on commit streams")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index c6fe2c00aedb6..d4cacb8df6310 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -2289,8 +2289,10 @@ enum dc_status dce110_apply_ctx_to_hw(
 		if (DC_OK != status)
 			return status;
 
+#ifdef CONFIG_DRM_AMD_DC_FP
 		if (hws->funcs.resync_fifo_dccg_dio)
 			hws->funcs.resync_fifo_dccg_dio(hws, dc, context);
+#endif
 	}
 
 	if (dc->fbc_compressor)
-- 
GitLab


From 025723e059ab454823e6aa21277976178f23d120 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Tue, 23 May 2023 07:49:37 -0400
Subject: [PATCH 1429/1791] drm/amd/display: remove unused variables
 res_create_maximus_funcs and debug_defaults_diags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc with W=1 reports
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:1069:43: error:
  ‘res_create_maximus_funcs’ defined but not used [-Werror=unused-const-variable=]
 1069 | static const struct resource_create_funcs res_create_maximus_funcs = {
      |                                           ^~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:727:38: error:
  ‘debug_defaults_diags’ defined but not used [-Werror=unused-const-variable=]
  727 | static const struct dc_debug_options debug_defaults_diags = {
      |                                      ^~~~~~~~~~~~~~~~~~~~

These variables are not used so remove them.

Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn20/dcn20_resource.c | 23 -------------------
 1 file changed, 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index 7dcae3183e078..6ef7e26349918 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -724,22 +724,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.underflow_assert_delay_us = 0xFFFFFFFF,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-		.disable_dmcu = false,
-		.force_abm_enable = false,
-		.timing_trace = true,
-		.clock_trace = true,
-		.disable_dpp_power_gate = true,
-		.disable_hubp_power_gate = true,
-		.disable_clock_gate = true,
-		.disable_pplib_clock_request = true,
-		.disable_pplib_wm_range = true,
-		.disable_stutter = true,
-		.scl_reset_length10 = true,
-		.underflow_assert_delay_us = 0xFFFFFFFF,
-		.enable_tri_buf = true,
-};
-
 void dcn20_dpp_destroy(struct dpp **dpp)
 {
 	kfree(TO_DCN20_DPP(*dpp));
@@ -1066,13 +1050,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn20_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn20_hwseq_create,
-};
-
 static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu);
 
 void dcn20_clock_source_destroy(struct clock_source **clk_src)
-- 
GitLab


From e7665d0ca7938a8f921760a780bdc55c5eda6df0 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Tue, 23 May 2023 10:18:11 +0800
Subject: [PATCH 1430/1791] drm/amdgpu: Remove duplicate include

./drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c: amdgpu_xcp.h is included more than once.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5281
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index cb34da8967370..c54bf0b38fe14 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -23,7 +23,6 @@
 #include <linux/firmware.h>
 
 #include "amdgpu.h"
-#include "amdgpu_xcp.h"
 #include "amdgpu_gfx.h"
 #include "soc15.h"
 #include "soc15d.h"
-- 
GitLab


From aaff9c089947cdb366ffddf4d4fb9747b6469d3e Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Tue, 23 May 2023 11:17:09 +0800
Subject: [PATCH 1431/1791] drm/amdgpu: Modify mismatched function name

No functional modification involved.

drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:426: warning: expecting prototype for sdma_v4_4_2_gfx_stop(). Prototype was for sdma_v4_4_2_inst_gfx_stop() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:457: warning: expecting prototype for sdma_v4_4_2_rlc_stop(). Prototype was for sdma_v4_4_2_inst_rlc_stop() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:470: warning: expecting prototype for sdma_v4_4_2_page_stop(). Prototype was for sdma_v4_4_2_inst_page_stop() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:506: warning: expecting prototype for sdma_v4_4_2_ctx_switch_enable(). Prototype was for sdma_v4_4_2_inst_ctx_switch_enable() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:561: warning: expecting prototype for sdma_v4_4_2_enable(). Prototype was for sdma_v4_4_2_inst_enable() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:798: warning: expecting prototype for sdma_v4_4_2_rlc_resume(). Prototype was for sdma_v4_4_2_inst_rlc_resume() instead.
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:814: warning: expecting prototype for sdma_v4_4_2_load_microcode(). Prototype was for sdma_v4_4_2_inst_load_microcode() instead.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5283
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index bf47eb33c12e0..590b08585901c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -415,7 +415,7 @@ static void sdma_v4_4_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
 
 
 /**
- * sdma_v4_4_2_gfx_stop - stop the gfx async dma engines
+ * sdma_v4_4_2_inst_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
  *
@@ -446,7 +446,7 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev,
 }
 
 /**
- * sdma_v4_4_2_rlc_stop - stop the compute async dma engines
+ * sdma_v4_4_2_inst_rlc_stop - stop the compute async dma engines
  *
  * @adev: amdgpu_device pointer
  *
@@ -459,7 +459,7 @@ static void sdma_v4_4_2_inst_rlc_stop(struct amdgpu_device *adev,
 }
 
 /**
- * sdma_v4_4_2_page_stop - stop the page async dma engines
+ * sdma_v4_4_2_inst_page_stop - stop the page async dma engines
  *
  * @adev: amdgpu_device pointer
  *
@@ -494,7 +494,7 @@ static void sdma_v4_4_2_inst_page_stop(struct amdgpu_device *adev,
 }
 
 /**
- * sdma_v4_4_2_ctx_switch_enable - stop the async dma engines context switch
+ * sdma_v4_4_2_inst_ctx_switch_enable - stop the async dma engines context switch
  *
  * @adev: amdgpu_device pointer
  * @enable: enable/disable the DMA MEs context switch.
@@ -548,7 +548,7 @@ static void sdma_v4_4_2_inst_ctx_switch_enable(struct amdgpu_device *adev,
 }
 
 /**
- * sdma_v4_4_2_enable - stop the async dma engines
+ * sdma_v4_4_2_inst_enable - stop the async dma engines
  *
  * @adev: amdgpu_device pointer
  * @enable: enable/disable the DMA MEs.
@@ -786,7 +786,7 @@ static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev)
 }
 
 /**
- * sdma_v4_4_2_rlc_resume - setup and start the async dma engines
+ * sdma_v4_4_2_inst_rlc_resume - setup and start the async dma engines
  *
  * @adev: amdgpu_device pointer
  *
@@ -802,7 +802,7 @@ static int sdma_v4_4_2_inst_rlc_resume(struct amdgpu_device *adev,
 }
 
 /**
- * sdma_v4_4_2_load_microcode - load the sDMA ME ucode
+ * sdma_v4_4_2_inst_load_microcode - load the sDMA ME ucode
  *
  * @adev: amdgpu_device pointer
  *
-- 
GitLab


From 1fbc69b8f543864fa0a6c4b5d95fb5f9f5d23d33 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Tue, 23 May 2023 11:32:02 +0800
Subject: [PATCH 1432/1791] drm/amdgpu/vcn: Modify mismatched function name

No functional modification involved.

drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:374: warning: expecting prototype for vcn_v4_0_mc_resume_dpg_mode(). Prototype was for vcn_v4_0_3_mc_resume_dpg_mode() instead.
drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:631: warning: expecting prototype for vcn_v4_0_enable_clock_gating(). Prototype was for vcn_v4_0_3_enable_clock_gating() instead.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5284
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 339842382a1ea..5d67b8b8a3d6b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -362,7 +362,7 @@ static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev, int inst_idx)
 }
 
 /**
- * vcn_v4_0_mc_resume_dpg_mode - memory controller programming for dpg mode
+ * vcn_v4_0_3_mc_resume_dpg_mode - memory controller programming for dpg mode
  *
  * @adev: amdgpu_device pointer
  * @inst_idx: instance number index
@@ -620,7 +620,7 @@ static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev,
 }
 
 /**
- * vcn_v4_0_enable_clock_gating - enable VCN clock gating
+ * vcn_v4_0_3_enable_clock_gating - enable VCN clock gating
  *
  * @adev: amdgpu_device pointer
  * @inst_idx: instance number
-- 
GitLab


From c3aaca43fb07ce05f3a3bd85288eb3d500469be5 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 17 May 2023 15:53:50 -0400
Subject: [PATCH 1433/1791] drm/amdgpu: Add a low priority scheduler for VRAM
 clearing

Add a low priority DRM scheduler for VRAM clearing instead of using
the exisiting high priority scheduler. Use the high priority scheduler
for migrations and evictions.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c |  4 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 37 ++++++++++++++++------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h    |  5 ++-
 3 files changed, 34 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 42c02f48c3a87..d9e3315083899 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -627,7 +627,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 	    bo->tbo.resource->mem_type == TTM_PL_VRAM) {
 		struct dma_fence *fence;
 
-		r = amdgpu_fill_buffer(bo, 0, bo->tbo.base.resv, &fence);
+		r = amdgpu_fill_buffer(bo, 0, bo->tbo.base.resv, &fence, true);
 		if (unlikely(r))
 			goto fail_unreserve;
 
@@ -1354,7 +1354,7 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)
 	if (WARN_ON_ONCE(!dma_resv_trylock(bo->base.resv)))
 		return;
 
-	r = amdgpu_fill_buffer(abo, AMDGPU_POISON, bo->base.resv, &fence);
+	r = amdgpu_fill_buffer(abo, AMDGPU_POISON, bo->base.resv, &fence, true);
 	if (!WARN_ON(r)) {
 		amdgpu_bo_fence(abo, fence, false);
 		dma_fence_put(fence);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 0abad5f894213..473eeac1f03bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -383,7 +383,8 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
 	    (abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) {
 		struct dma_fence *wipe_fence = NULL;
 
-		r = amdgpu_fill_buffer(abo, AMDGPU_POISON, NULL, &wipe_fence);
+		r = amdgpu_fill_buffer(abo, AMDGPU_POISON, NULL, &wipe_fence,
+					false);
 		if (r) {
 			goto error;
 		} else if (wipe_fence) {
@@ -2036,8 +2037,18 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 				  r);
 			return;
 		}
+
+		r = drm_sched_entity_init(&adev->mman.delayed,
+					  DRM_SCHED_PRIORITY_NORMAL, &sched,
+					  1, NULL);
+		if (r) {
+			DRM_ERROR("Failed setting up TTM BO move entity (%d)\n",
+				  r);
+			goto error_free_entity;
+		}
 	} else {
 		drm_sched_entity_destroy(&adev->mman.entity);
+		drm_sched_entity_destroy(&adev->mman.delayed);
 		dma_fence_put(man->move);
 		man->move = NULL;
 	}
@@ -2049,6 +2060,11 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 		size = adev->gmc.visible_vram_size;
 	man->size = size;
 	adev->mman.buffer_funcs_enabled = enable;
+
+	return;
+
+error_free_entity:
+	drm_sched_entity_destroy(&adev->mman.entity);
 }
 
 static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
@@ -2056,14 +2072,16 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
 				  unsigned int num_dw,
 				  struct dma_resv *resv,
 				  bool vm_needs_flush,
-				  struct amdgpu_job **job)
+				  struct amdgpu_job **job,
+				  bool delayed)
 {
 	enum amdgpu_ib_pool_type pool = direct_submit ?
 		AMDGPU_IB_POOL_DIRECT :
 		AMDGPU_IB_POOL_DELAYED;
 	int r;
-
-	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity,
+	struct drm_sched_entity *entity = delayed ? &adev->mman.delayed :
+						    &adev->mman.entity;
+	r = amdgpu_job_alloc_with_ib(adev, entity,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4, pool, job);
 	if (r)
@@ -2104,7 +2122,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 	num_loops = DIV_ROUND_UP(byte_count, max_bytes);
 	num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8);
 	r = amdgpu_ttm_prepare_job(adev, direct_submit, num_dw,
-				   resv, vm_needs_flush, &job);
+				   resv, vm_needs_flush, &job, false);
 	if (r)
 		return r;
 
@@ -2140,7 +2158,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data,
 			       uint64_t dst_addr, uint32_t byte_count,
 			       struct dma_resv *resv,
 			       struct dma_fence **fence,
-			       bool vm_needs_flush)
+			       bool vm_needs_flush, bool delayed)
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned int num_loops, num_dw;
@@ -2153,7 +2171,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data,
 	num_loops = DIV_ROUND_UP_ULL(byte_count, max_bytes);
 	num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->fill_num_dw, 8);
 	r = amdgpu_ttm_prepare_job(adev, false, num_dw, resv, vm_needs_flush,
-				   &job);
+				   &job, delayed);
 	if (r)
 		return r;
 
@@ -2176,7 +2194,8 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data,
 int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			uint32_t src_data,
 			struct dma_resv *resv,
-			struct dma_fence **f)
+			struct dma_fence **f,
+			bool delayed)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
@@ -2205,7 +2224,7 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			goto error;
 
 		r = amdgpu_ttm_fill_mem(ring, src_data, to, cur_size, resv,
-					&next, true);
+					&next, true, delayed);
 		if (r)
 			goto error;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 8ef048a0a33e7..e82b1edee7a4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -61,6 +61,8 @@ struct amdgpu_mman {
 	struct mutex				gtt_window_lock;
 	/* Scheduler entity for buffer moves */
 	struct drm_sched_entity			entity;
+	/* Scheduler entity for VRAM clearing */
+	struct drm_sched_entity			delayed;
 
 	struct amdgpu_vram_mgr vram_mgr;
 	struct amdgpu_gtt_mgr gtt_mgr;
@@ -152,7 +154,8 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev,
 int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			uint32_t src_data,
 			struct dma_resv *resv,
-			struct dma_fence **fence);
+			struct dma_fence **fence,
+			bool delayed);
 
 int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo);
 void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo);
-- 
GitLab


From 353491c48697df8a133bc468dc1b8ef65045254a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michel=20D=C3=A4nzer?= <mdaenzer@redhat.com>
Date: Mon, 22 May 2023 15:08:22 +0200
Subject: [PATCH 1434/1791] Revert "drm/amd/display: Block optimize on
 consecutive FAMS enables"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit 751e17147953bc30036b8fe0eaaf780b6951404c.

It depends on its parent commit, which we want to revert.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
[Hamza: fix a whitespace issue in dcn30_prepare_bandwidth()]
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn20/dcn20_hwseq.c    |  3 ---
 .../drm/amd/display/dc/dcn30/dcn30_hwseq.c    | 22 +++----------------
 2 files changed, 3 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index b0a13eb8318cd..b56f799f10e12 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2126,9 +2126,6 @@ void dcn20_optimize_bandwidth(
 		dc_dmub_srv_p_state_delegate(dc,
 			true, context);
 		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
-		dc->clk_mgr->clks.fw_based_mclk_switching = true;
-	} else {
-		dc->clk_mgr->clks.fw_based_mclk_switching = false;
 	}
 
 	dc->clk_mgr->funcs->update_clocks(
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index ad4e1023b00e2..368fd5a716806 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -949,13 +949,9 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 }
 
 void dcn30_prepare_bandwidth(struct dc *dc,
-	struct dc_state *context)
+			     struct dc_state *context)
 {
-	bool p_state_change_support = context->bw_ctx.bw.dcn.clk.p_state_change_support;
-	/* Any transition into an FPO config should disable MCLK switching first to avoid
-	 * driver and FW P-State synchronization issues.
-	 */
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) {
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
 		dc->optimized_required = true;
 		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
 	}
@@ -966,19 +962,7 @@ void dcn30_prepare_bandwidth(struct dc *dc,
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
-	/*
-	 * enabled -> enabled: do not disable
-	 * enabled -> disabled: disable
-	 * disabled -> enabled: don't care
-	 * disabled -> disabled: don't care
-	 */
-	if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching)
-		dc_dmub_srv_p_state_delegate(dc, false, context);
 
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) {
-		/* After disabling P-State, restore the original value to ensure we get the correct P-State
-		 * on the next optimize. */
-		context->bw_ctx.bw.dcn.clk.p_state_change_support = p_state_change_support;
-	}
+	dc_dmub_srv_p_state_delegate(dc, false, context);
 }
 
-- 
GitLab


From 413521a4c9f4bbb4637b9ff3427070325890b08f Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 24 May 2023 01:17:48 +0530
Subject: [PATCH 1435/1791] drm/amd/amdgpu: Fix warnings in amdgpu_irq.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

checkpatch reports:

WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: braces {} are not necessary for any arm of this statement
+               if (nvec <= 0) {
[...]
+               } else {
[...]
WARNING: Block comments use a trailing */ on a separate line

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 27 +++++++++++++------------
 1 file changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 3dafaf70c9878..5667053c295c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -124,7 +124,7 @@ const int node_id_to_phys_map[NODEID_MAX] = {
 void amdgpu_irq_disable_all(struct amdgpu_device *adev)
 {
 	unsigned long irqflags;
-	unsigned i, j, k;
+	unsigned int i, j, k;
 	int r;
 
 	spin_lock_irqsave(&adev->irq.lock, irqflags);
@@ -283,11 +283,11 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
 		int nvec = pci_msix_vec_count(adev->pdev);
 		unsigned int flags;
 
-		if (nvec <= 0) {
+		if (nvec <= 0)
 			flags = PCI_IRQ_MSI;
-		} else {
+		else
 			flags = PCI_IRQ_MSI | PCI_IRQ_MSIX;
-		}
+
 		/* we only need one vector */
 		nvec = pci_alloc_irq_vectors(adev->pdev, 1, 1, flags);
 		if (nvec > 0) {
@@ -346,7 +346,7 @@ void amdgpu_irq_fini_hw(struct amdgpu_device *adev)
  */
 void amdgpu_irq_fini_sw(struct amdgpu_device *adev)
 {
-	unsigned i, j;
+	unsigned int i, j;
 
 	for (i = 0; i < AMDGPU_IRQ_CLIENTID_MAX; ++i) {
 		if (!adev->irq.client[i].sources)
@@ -380,7 +380,7 @@ void amdgpu_irq_fini_sw(struct amdgpu_device *adev)
  * 0 on success or error code otherwise
  */
 int amdgpu_irq_add_id(struct amdgpu_device *adev,
-		      unsigned client_id, unsigned src_id,
+		      unsigned int client_id, unsigned int src_id,
 		      struct amdgpu_irq_src *source)
 {
 	if (client_id >= AMDGPU_IRQ_CLIENTID_MAX)
@@ -432,7 +432,7 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
 {
 	u32 ring_index = ih->rptr >> 2;
 	struct amdgpu_iv_entry entry;
-	unsigned client_id, src_id;
+	unsigned int client_id, src_id;
 	struct amdgpu_irq_src *src;
 	bool handled = false;
 	int r;
@@ -507,7 +507,7 @@ void amdgpu_irq_delegate(struct amdgpu_device *adev,
  * Updates interrupt state for the specific source (all ASICs).
  */
 int amdgpu_irq_update(struct amdgpu_device *adev,
-			     struct amdgpu_irq_src *src, unsigned type)
+			     struct amdgpu_irq_src *src, unsigned int type)
 {
 	unsigned long irqflags;
 	enum amdgpu_interrupt_state state;
@@ -516,7 +516,8 @@ int amdgpu_irq_update(struct amdgpu_device *adev,
 	spin_lock_irqsave(&adev->irq.lock, irqflags);
 
 	/* We need to determine after taking the lock, otherwise
-	   we might disable just enabled interrupts again */
+	 * we might disable just enabled interrupts again
+	 */
 	if (amdgpu_irq_enabled(adev, src, type))
 		state = AMDGPU_IRQ_STATE_ENABLE;
 	else
@@ -570,7 +571,7 @@ void amdgpu_irq_gpu_reset_resume_helper(struct amdgpu_device *adev)
  * 0 on success or error code otherwise
  */
 int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
-		   unsigned type)
+		   unsigned int type)
 {
 	if (!adev->irq.installed)
 		return -ENOENT;
@@ -600,7 +601,7 @@ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
  * 0 on success or error code otherwise
  */
 int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
-		   unsigned type)
+		   unsigned int type)
 {
 	if (!adev->irq.installed)
 		return -ENOENT;
@@ -634,7 +635,7 @@ int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
  * invalid parameters
  */
 bool amdgpu_irq_enabled(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
-			unsigned type)
+			unsigned int type)
 {
 	if (!adev->irq.installed)
 		return false;
@@ -747,7 +748,7 @@ void amdgpu_irq_remove_domain(struct amdgpu_device *adev)
  * Returns:
  * Linux IRQ
  */
-unsigned amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned src_id)
+unsigned int amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned int src_id)
 {
 	adev->irq.virq[src_id] = irq_create_mapping(adev->irq.domain, src_id);
 
-- 
GitLab


From 360930985ec9f394c82ba0b235403b4a366d1560 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michel=20D=C3=A4nzer?= <mdaenzer@redhat.com>
Date: Mon, 22 May 2023 15:08:23 +0200
Subject: [PATCH 1436/1791] Revert "drm/amd/display: Do not set drr on pipe
 commit"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit e101bf95ea87ccc03ac2f48dfc0757c6364ff3c7.

Caused a regression:

Samsung Odyssey Neo G9, running at 5120x1440@240/VRR, connected to Navi
21 via DisplayPort, blanks and the GPU hangs while starting the Steam
game Assetto Corsa Competizione (via Proton 7.0).

Example dmesg excerpt:

 amdgpu 0000:0c:00.0: [drm] ERROR [CRTC:82:crtc-0] flip_done timed out
 NMI watchdog: Watchdog detected hard LOCKUP on cpu 6
 [...]
 RIP: 0010:amdgpu_device_rreg.part.0+0x2f/0xf0 [amdgpu]
 Code: 41 54 44 8d 24 b5 00 00 00 00 55 89 f5 53 48 89 fb 4c 3b a7 60 0b 00 00 73 6a 83 e2 02 74 29 4c 03 a3 68 0b 00 00 45 8b 24 24 <48> 8b 43 08 0f b7 70 3e 66 90 44 89 e0 5b 5d 41 5c 31 d2 31 c9 31
 RSP: 0000:ffffb39a119dfb88 EFLAGS: 00000086
 RAX: ffffffffc0eb96a0 RBX: ffff9e7963dc0000 RCX: 0000000000007fff
 RDX: 0000000000000000 RSI: 0000000000004ff6 RDI: ffff9e7963dc0000
 RBP: 0000000000004ff6 R08: ffffb39a119dfc40 R09: 0000000000000010
 R10: ffffb39a119dfc40 R11: ffffb39a119dfc44 R12: 00000000000e05ae
 R13: 0000000000000000 R14: ffff9e7963dc0010 R15: 0000000000000000
 FS:  000000001012f6c0(0000) GS:ffff9e805eb80000(0000) knlGS:000000007fd40000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00000000461ca000 CR3: 00000002a8a20000 CR4: 0000000000350ee0
 Call Trace:
  <TASK>
  dm_read_reg_func+0x37/0xc0 [amdgpu]
  generic_reg_get2+0x22/0x60 [amdgpu]
  optc1_get_crtc_scanoutpos+0x6a/0xc0 [amdgpu]
  dc_stream_get_scanoutpos+0x74/0x90 [amdgpu]
  dm_crtc_get_scanoutpos+0x82/0xf0 [amdgpu]
  amdgpu_display_get_crtc_scanoutpos+0x91/0x190 [amdgpu]
  ? dm_read_reg_func+0x37/0xc0 [amdgpu]
  amdgpu_get_vblank_counter_kms+0xb4/0x1a0 [amdgpu]
  dm_pflip_high_irq+0x213/0x2f0 [amdgpu]
  amdgpu_dm_irq_handler+0x8a/0x200 [amdgpu]
  amdgpu_irq_dispatch+0xd4/0x220 [amdgpu]
  amdgpu_ih_process+0x7f/0x110 [amdgpu]
  amdgpu_irq_handler+0x1f/0x70 [amdgpu]
  __handle_irq_event_percpu+0x46/0x1b0
  handle_irq_event+0x34/0x80
  handle_edge_irq+0x9f/0x240
  __common_interrupt+0x66/0x110
  common_interrupt+0x5c/0xd0
  asm_common_interrupt+0x22/0x40

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 6 ------
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c | 7 -------
 2 files changed, 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index b56f799f10e12..eaf9e9ccad2a7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -2122,12 +2122,6 @@ void dcn20_optimize_bandwidth(
 	if (hubbub->funcs->program_compbuf_size)
 		hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true);
 
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
-		dc_dmub_srv_p_state_delegate(dc,
-			true, context);
-		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
-	}
-
 	dc->clk_mgr->funcs->update_clocks(
 			dc->clk_mgr,
 			context,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
index 368fd5a716806..b9753867d97b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c
@@ -951,18 +951,11 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 void dcn30_prepare_bandwidth(struct dc *dc,
 			     struct dc_state *context)
 {
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
-		dc->optimized_required = true;
-		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
-	}
-
 	if (dc->clk_mgr->dc_mode_softmax_enabled)
 		if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 &&
 				context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000)
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
-
-	dc_dmub_srv_p_state_delegate(dc, false, context);
 }
 
-- 
GitLab


From 55a6dc60b47c817c644af2b505d46815d8b9219e Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 23 May 2023 11:59:53 -0400
Subject: [PATCH 1437/1791] drm/amdkfd: Set event interrupt class for GFX 9.4.3

Fix the warning during driver load because the event
interrupt class is not set for GFX9.4.3.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index e92b93b2c14c3..862a50f7b490d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -139,6 +139,7 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd)
 	case IP_VERSION(9, 4, 0): /* VEGA20 */
 	case IP_VERSION(9, 4, 1): /* ARCTURUS */
 	case IP_VERSION(9, 4, 2): /* ALDEBARAN */
+	case IP_VERSION(9, 4, 3): /* GC 9.4.3 */
 	case IP_VERSION(10, 3, 1): /* VANGOGH */
 	case IP_VERSION(10, 3, 3): /* YELLOW_CARP */
 	case IP_VERSION(10, 3, 6): /* GC 10.3.6 */
-- 
GitLab


From d48a4f2c2809b882b58b428577707fe1b6f52673 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Mon, 22 May 2023 23:17:28 +0800
Subject: [PATCH 1438/1791] drm/amd/pm: reverse mclk and fclk clocks levels for
 renoir

This patch reverses the DPM clocks levels output of pp_dpm_mclk
and pp_dpm_fclk for renoir.

On dGPUs and older APUs we expose the levels from lowest clocks
to highest clocks. But for some APUs, the clocks levels are
given the reversed orders by PMFW. Like the memory DPM clocks
that are exposed by pp_dpm_mclk.

It's not intuitive that they are reversed on these APUs. All tools
and software that talks to the driver then has to know different ways
to interpret the data depending on the asic.

So we need to reverse them to expose the clocks levels from the
driver consistently.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
index 5cdc07165480b..8a8ba25c9ad7c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
@@ -494,7 +494,7 @@ static int renoir_set_fine_grain_gfx_freq_parameters(struct smu_context *smu)
 static int renoir_print_clk_levels(struct smu_context *smu,
 			enum smu_clk_type clk_type, char *buf)
 {
-	int i, size = 0, ret = 0;
+	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0, min = 0, max = 0;
 	SmuMetrics_t metrics;
 	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
@@ -594,7 +594,8 @@ static int renoir_print_clk_levels(struct smu_context *smu,
 	case SMU_VCLK:
 	case SMU_DCLK:
 		for (i = 0; i < count; i++) {
-			ret = renoir_get_dpm_clk_limited(smu, clk_type, i, &value);
+			idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i;
+			ret = renoir_get_dpm_clk_limited(smu, clk_type, idx, &value);
 			if (ret)
 				return ret;
 			if (!value)
-- 
GitLab


From 194224a54c8bbc896b1fdb4a10ca5789ea4b9e7d Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 23 May 2023 13:26:39 +0530
Subject: [PATCH 1439/1791] drm/amdgpu: Fix warnings

Fix warnings reported by kernel test bot/smatch

smatch warnings:
drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c:65 amdgpu_xcp_run_transition()
error: buffer overflow 'xcp_mgr->xcp' 8 <= 8

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/r/202305231453.I0bXngYn-lkp@intel.com/
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index e9586a0dc335d..fcdc0862d2584 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -65,7 +65,7 @@ static int amdgpu_xcp_run_transition(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 	struct amdgpu_xcp *xcp;
 	int i, ret;
 
-	if (xcp_id > MAX_XCP || !xcp_mgr->xcp[xcp_id].valid)
+	if (xcp_id >= MAX_XCP || !xcp_mgr->xcp[xcp_id].valid)
 		return -EINVAL;
 
 	xcp = &xcp_mgr->xcp[xcp_id];
-- 
GitLab


From a569552f76889c8162b73add69afa161ff660ae4 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Wed, 24 May 2023 15:03:02 +0800
Subject: [PATCH 1440/1791] drm/amd/pm: resolve reboot exception for si oland

During reboot test on arm64 platform, it may failure on boot.

The error message are as follows:
[    1.706570][ 3] [  T273] [drm:si_thermal_enable_alert [amdgpu]] *ERROR* Could not enable thermal interrupts.
[    1.716547][ 3] [  T273] [drm:amdgpu_device_ip_late_init [amdgpu]] *ERROR* late_init of IP block <si_dpm> failed -22
[    1.727064][ 3] [  T273] amdgpu 0000:02:00.0: amdgpu_device_ip_late_init failed
[    1.734367][ 3] [  T273] amdgpu 0000:02:00.0: Fatal error during GPU init

v2: squash in built warning fix (Alex)

Signed-off-by: Zhenneng Li <lizhenneng@kylinos.cn>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 29 ----------------------
 1 file changed, 29 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index d6d9e3b1b2c0e..02e69ccff3bac 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -6925,23 +6925,6 @@ static int si_dpm_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
-static int si_set_temperature_range(struct amdgpu_device *adev)
-{
-	int ret;
-
-	ret = si_thermal_enable_alert(adev, false);
-	if (ret)
-		return ret;
-	ret = si_thermal_set_temperature_range(adev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX);
-	if (ret)
-		return ret;
-	ret = si_thermal_enable_alert(adev, true);
-	if (ret)
-		return ret;
-
-	return ret;
-}
-
 static void si_dpm_disable(struct amdgpu_device *adev)
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(adev);
@@ -7626,18 +7609,6 @@ static int si_dpm_process_interrupt(struct amdgpu_device *adev,
 
 static int si_dpm_late_init(void *handle)
 {
-	int ret;
-	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-
-	if (!adev->pm.dpm_enabled)
-		return 0;
-
-	ret = si_set_temperature_range(adev);
-	if (ret)
-		return ret;
-#if 0 //TODO ?
-	si_dpm_powergate_uvd(adev, true);
-#endif
 	return 0;
 }
 
-- 
GitLab


From 61c31b8b6c5e386a9c2ddc4e2cf9d8ae46c8a1f9 Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Fri, 12 May 2023 16:04:56 +0800
Subject: [PATCH 1441/1791] drm/amdgpu/sdma: set sched.ready status after
 ring/IB test in sdma

sched.ready is nothing with ring initialization, it needs to set
to be true after ring/IB test in amdgpu_ring_test_helper to tell
the ring is ready for submission.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/cik_sdma.c    | 2 --
 drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c   | 2 --
 drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c   | 2 --
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c   | 4 ----
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 4 ----
 drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c   | 2 --
 drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c   | 6 +-----
 drivers/gpu/drm/amd/amdgpu/si_dma.c      | 2 --
 9 files changed, 3 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 67d16236b2168..52598fbc9b39d 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -489,8 +489,6 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
 #endif
 		/* enable DMA IBs */
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
-
-		ring->sched.ready = true;
 	}
 
 	cik_sdma_enable(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index fd2a7b66ac562..51afc92994a80 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -466,8 +466,6 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev)
 #endif
 		/* enable DMA IBs */
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
-
-		ring->sched.ready = true;
 	}
 
 	sdma_v2_4_enable(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index e572389089d24..344202870aebe 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -734,8 +734,6 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
 #endif
 		/* enable DMA IBs */
 		WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
-
-		ring->sched.ready = true;
 	}
 
 	/* unhalt the MEs */
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 70b0d1fd9868b..1f83eebfc8a74 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1114,8 +1114,6 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 #endif
 	/* enable DMA IBs */
 	WREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL, ib_cntl);
-
-	ring->sched.ready = true;
 }
 
 /**
@@ -1202,8 +1200,6 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i)
 #endif
 	/* enable DMA IBs */
 	WREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL, ib_cntl);
-
-	ring->sched.ready = true;
 }
 
 static void
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 590b08585901c..ff41fb577cdd5 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -685,8 +685,6 @@ static void sdma_v4_4_2_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 #endif
 	/* enable DMA IBs */
 	WREG32_SDMA(i, regSDMA_GFX_IB_CNTL, ib_cntl);
-
-	ring->sched.ready = true;
 }
 
 /**
@@ -776,8 +774,6 @@ static void sdma_v4_4_2_page_resume(struct amdgpu_device *adev, unsigned int i)
 #endif
 	/* enable DMA IBs */
 	WREG32_SDMA(i, regSDMA_PAGE_IB_CNTL, ib_cntl);
-
-	ring->sched.ready = true;
 }
 
 static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
index a0077cf412955..5c4d4df9cf94c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c
@@ -819,8 +819,6 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev)
 		/* enable DMA IBs */
 		WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
 
-		ring->sched.ready = true;
-
 		if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */
 			sdma_v5_0_ctx_switch_enable(adev, true);
 			sdma_v5_0_enable(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index efa2c84ee78e5..6aae62b68f32d 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -617,8 +617,6 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev)
 		/* enable DMA IBs */
 		WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
 
-		ring->sched.ready = true;
-
 		if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */
 			sdma_v5_2_ctx_switch_enable(adev, true);
 			sdma_v5_2_enable(adev, true);
@@ -630,6 +628,8 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev)
 			return r;
 		}
 
+		ring->sched.ready = true;
+
 		if (adev->mman.buffer_funcs_ring == ring)
 			amdgpu_ttm_set_buffer_funcs_status(adev, true);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index 79d09792d2ce3..1c90b5c661fb7 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -585,16 +585,12 @@ static int sdma_v6_0_gfx_resume(struct amdgpu_device *adev)
 		/* enable DMA IBs */
 		WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL), ib_cntl);
 
-		ring->sched.ready = true;
-
 		if (amdgpu_sriov_vf(adev))
 			sdma_v6_0_enable(adev, true);
 
 		r = amdgpu_ring_test_helper(ring);
-		if (r) {
-			ring->sched.ready = false;
+		if (r)
 			return r;
-		}
 
 		if (adev->mman.buffer_funcs_ring == ring)
 			amdgpu_ttm_set_buffer_funcs_status(adev, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index abca8b529721e..42c4547f32ec9 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -174,8 +174,6 @@ static int si_dma_start(struct amdgpu_device *adev)
 		WREG32(DMA_RB_WPTR + sdma_offsets[i], ring->wptr << 2);
 		WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl | DMA_RB_ENABLE);
 
-		ring->sched.ready = true;
-
 		r = amdgpu_ring_test_helper(ring);
 		if (r)
 			return r;
-- 
GitLab


From 232f2431899cbe6c00c1350e35cfba91ea0c1c0b Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Fri, 12 May 2023 16:10:40 +0800
Subject: [PATCH 1442/1791] drm/amdgpu/gfx: set sched.ready status after
 ring/IB test in gfx

sched.ready is nothing with ring initialization, it needs to set
to be true after ring/IB test in amdgpu_ring_test_helper to tell
the ring is ready for submission.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c  | 24 ++++--------------------
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  | 25 ++++---------------------
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c   |  2 --
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   |  2 --
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c |  1 -
 5 files changed, 8 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 0d15002eac691..f7ad883a70fa2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -6073,7 +6073,6 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev)
 	u32 tmp;
 	u32 rb_bufsz;
 	u64 rb_addr, rptr_addr, wptr_gpu_addr;
-	u32 i;
 
 	/* Set the write pointer delay */
 	WREG32_SOC15(GC, 0, mmCP_RB_WPTR_DELAY, 0);
@@ -6168,11 +6167,6 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev)
 	/* start the ring */
 	gfx_v10_0_cp_gfx_start(adev);
 
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-		ring = &adev->gfx.gfx_ring[i];
-		ring->sched.ready = true;
-	}
-
 	return 0;
 }
 
@@ -6470,7 +6464,7 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 
 		r = amdgpu_bo_reserve(ring->mqd_obj, false);
 		if (unlikely(r != 0))
-			goto done;
+			return r;
 
 		r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
 		if (!r) {
@@ -6480,23 +6474,14 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 		}
 		amdgpu_bo_unreserve(ring->mqd_obj);
 		if (r)
-			goto done;
+			return r;
 	}
 
 	r = amdgpu_gfx_enable_kgq(adev, 0);
 	if (r)
-		goto done;
-
-	r = gfx_v10_0_cp_gfx_start(adev);
-	if (r)
-		goto done;
+		return r;
 
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-		ring = &adev->gfx.gfx_ring[i];
-		ring->sched.ready = true;
-	}
-done:
-	return r;
+	return gfx_v10_0_cp_gfx_start(adev);
 }
 
 static int gfx_v10_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
@@ -6812,7 +6797,6 @@ static int gfx_v10_0_kiq_resume(struct amdgpu_device *adev)
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
-	ring->sched.ready = true;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 4b7224de879ec..da21bf868080e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -3228,7 +3228,6 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev)
 	u32 tmp;
 	u32 rb_bufsz;
 	u64 rb_addr, rptr_addr, wptr_gpu_addr;
-	u32 i;
 
 	/* Set the write pointer delay */
 	WREG32_SOC15(GC, 0, regCP_RB_WPTR_DELAY, 0);
@@ -3320,11 +3319,6 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev)
 	/* start the ring */
 	gfx_v11_0_cp_gfx_start(adev);
 
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-		ring = &adev->gfx.gfx_ring[i];
-		ring->sched.ready = true;
-	}
-
 	return 0;
 }
 
@@ -3370,8 +3364,6 @@ static void gfx_v11_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 		WREG32_SOC15(GC, 0, regCP_MEC_CNTL, data);
 	}
 
-	adev->gfx.kiq[0].ring.sched.ready = enable;
-
 	udelay(50);
 }
 
@@ -3711,7 +3703,7 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 
 		r = amdgpu_bo_reserve(ring->mqd_obj, false);
 		if (unlikely(r != 0))
-			goto done;
+			return r;
 
 		r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr);
 		if (!r) {
@@ -3721,23 +3713,14 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev)
 		}
 		amdgpu_bo_unreserve(ring->mqd_obj);
 		if (r)
-			goto done;
+			return r;
 	}
 
 	r = amdgpu_gfx_enable_kgq(adev, 0);
 	if (r)
-		goto done;
-
-	r = gfx_v11_0_cp_gfx_start(adev);
-	if (r)
-		goto done;
+		return r;
 
-	for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
-		ring = &adev->gfx.gfx_ring[i];
-		ring->sched.ready = true;
-	}
-done:
-	return r;
+	return gfx_v11_0_cp_gfx_start(adev);
 }
 
 static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 6d0589dc1d6cc..2f1ef75e126c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4283,7 +4283,6 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev)
 	/* start the ring */
 	amdgpu_ring_clear_ring(ring);
 	gfx_v8_0_cp_gfx_start(adev);
-	ring->sched.ready = true;
 
 	return 0;
 }
@@ -4693,7 +4692,6 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
-	ring->sched.ready = true;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 8bf95a6b07677..fe090eafa3665 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -3144,7 +3144,6 @@ static int gfx_v9_0_cp_gfx_resume(struct amdgpu_device *adev)
 
 	/* start the ring */
 	gfx_v9_0_cp_gfx_start(adev);
-	ring->sched.ready = true;
 
 	return 0;
 }
@@ -3623,7 +3622,6 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev)
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
-	ring->sched.ready = true;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index c54bf0b38fe14..ed41a7862d9f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1845,7 +1845,6 @@ static int gfx_v9_4_3_xcc_kiq_resume(struct amdgpu_device *adev, int xcc_id)
 	amdgpu_bo_kunmap(ring->mqd_obj);
 	ring->mqd_ptr = NULL;
 	amdgpu_bo_unreserve(ring->mqd_obj);
-	ring->sched.ready = true;
 	return 0;
 }
 
-- 
GitLab


From 93ab59ac6d8311244a76ddb31e7ced4cb1e8f22c Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Fri, 12 May 2023 16:14:25 +0800
Subject: [PATCH 1443/1791] drm/amdgpu: switch to unified
 amdgpu_ring_test_helper

This will simplify code.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 9 ++-------
 drivers/gpu/drm/amd/amdgpu/mes_v10_1.c  | 8 +-------
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c  | 7 +------
 drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c  | 8 ++------
 4 files changed, 6 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index f0f00466b59f2..49bb6c03d6067 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -1305,14 +1305,9 @@ static int amdgpu_mes_test_queues(struct amdgpu_ring **added_rings)
 		if (!ring)
 			continue;
 
-		r = amdgpu_ring_test_ring(ring);
-		if (r) {
-			DRM_DEV_ERROR(ring->adev->dev,
-				      "ring %s test failed (%d)\n",
-				      ring->name, r);
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
 			return r;
-		} else
-			DRM_INFO("ring %s test pass\n", ring->name);
 
 		r = amdgpu_ring_test_ib(ring, 1000 * 10);
 		if (r) {
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
index 88262f10ef7c5..36a123e6c8ee7 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c
@@ -815,13 +815,7 @@ static int mes_v10_1_kiq_enable_queue(struct amdgpu_device *adev)
 
 	kiq->pmf->kiq_map_queues(kiq_ring, &adev->mes.ring);
 
-	r = amdgpu_ring_test_ring(kiq_ring);
-	if (r) {
-		DRM_ERROR("kfq enable failed\n");
-		kiq_ring->sched.ready = false;
-	}
-
-	return r;
+	return amdgpu_ring_test_helper(kiq_ring);
 }
 
 static int mes_v10_1_queue_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 9a48328c6572d..90b4a74ccf01a 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -882,12 +882,7 @@ static int mes_v11_0_kiq_enable_queue(struct amdgpu_device *adev)
 
 	kiq->pmf->kiq_map_queues(kiq_ring, &adev->mes.ring);
 
-	r = amdgpu_ring_test_ring(kiq_ring);
-	if (r) {
-		DRM_ERROR("kfq enable failed\n");
-		kiq_ring->sched.ready = false;
-	}
-	return r;
+	return amdgpu_ring_test_helper(kiq_ring);
 }
 
 static int mes_v11_0_queue_init(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index 6aae62b68f32d..a7b230e5a26d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -622,13 +622,9 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev)
 			sdma_v5_2_enable(adev, true);
 		}
 
-		r = amdgpu_ring_test_ring(ring);
-		if (r) {
-			ring->sched.ready = false;
+		r = amdgpu_ring_test_helper(ring);
+		if (r)
 			return r;
-		}
-
-		ring->sched.ready = true;
 
 		if (adev->mman.buffer_funcs_ring == ring)
 			amdgpu_ttm_set_buffer_funcs_status(adev, true);
-- 
GitLab


From 3525844d483bfb2236c1dd00f7a490297721ef78 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 28 Jul 2022 11:42:44 +0530
Subject: [PATCH 1444/1791] drm/amdgpu: Use single copy per SDMA instance type
 (v2)

v1: Only single copy per instance type is required for PSP. All instance types
use the same firmware copy. (Lijo)

v2: Apply the change into amdgpu_sdma_init_microcode() due to rebase. (Morris)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
index 9568adaad5cbe..78ec3420ef853 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
@@ -252,6 +252,13 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev,
 				if (!duplicate && (instance != i))
 					continue;
 				else {
+					/* Use a single copy per SDMA firmware type. PSP uses the same instance for all
+					 * groups of SDMAs */
+					if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 4, 2) &&
+					    adev->firmware.load_type == AMDGPU_FW_LOAD_PSP &&
+					    adev->sdma.num_inst_per_aid == i) {
+						break;
+					}
 					info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i];
 					info->ucode_id = AMDGPU_UCODE_ID_SDMA0 + i;
 					info->fw = adev->sdma.instance[i].fw;
-- 
GitLab


From bea35f7d4219999b5e4a9c8cdf6a17683b2ed1bd Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:40 +0800
Subject: [PATCH 1445/1791] drm/amd/display: remove unused definition

Eliminate the following warning:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_resource.c:889:43: warning: unused variable 'res_create_maximus_funcs'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index a0625209c86d6..26ddf73fd5b11 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -886,13 +886,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn10_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn10_hwseq_create,
-};
-
 static void dcn10_clock_source_destroy(struct clock_source **clk_src)
 {
 	kfree(TO_DCE110_CLK_SRC(*clk_src));
-- 
GitLab


From 164f0791c5d10a2f0e947f6872a7c14ccd860085 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:41 +0800
Subject: [PATCH 1446/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn21/dcn21_resource.c:1222:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn21/dcn21_resource.c:658:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn21/dcn21_resource.c | 23 -------------------
 1 file changed, 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index 76268a7b7934d..da6d42de0554a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -655,22 +655,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.use_max_lb = true,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-		.disable_dmcu = false,
-		.force_abm_enable = false,
-		.timing_trace = true,
-		.clock_trace = true,
-		.disable_dpp_power_gate = true,
-		.disable_hubp_power_gate = true,
-		.disable_clock_gate = true,
-		.disable_pplib_clock_request = true,
-		.disable_pplib_wm_range = true,
-		.disable_stutter = true,
-		.disable_48mhz_pwrdwn = true,
-		.enable_tri_buf = true,
-		.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 		.psr = {
 			.disable_psr = false,
@@ -1219,13 +1203,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn21_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn21_hwseq_create,
-};
-
 static const struct encoder_feature_support link_enc_feature = {
 		.max_hdmi_deep_color = COLOR_DEPTH_121212,
 		.max_hdmi_pixel_clock = 600000,
-- 
GitLab


From 95c2f89895930538879e1c3f1467a8095e03d1a5 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:42 +0800
Subject: [PATCH 1447/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn30/dcn30_resource.c:1079:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn30/dcn30_resource.c:731:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn30/dcn30_resource.c | 25 -------------------
 1 file changed, 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index cd94b5f5fdb4e..f4ee4b3df5962 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -728,24 +728,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.exit_idle_opt_for_cursor_updates = true
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true, //No dmcu on DCN30
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 	.psr = {
 		.disable_psr = false,
@@ -1076,13 +1058,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn30_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn30_hwseq_create,
-};
-
 static void dcn30_resource_destruct(struct dcn30_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 200c7c8132318378feffc1e4b0189482f4c7f1b6 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 May 2023 12:51:23 -0500
Subject: [PATCH 1448/1791] drm/amd: Drop debugfs access to the DPCD

There is already access to the DPCD from userspace through
`CONFIG_DRM_DP_AUX_CHARDEV`, so it's unnecessary to reinvent the wheel
with a set of extra debugfs nodes specific to amdgpu.

The character device interface behaves more like you would expect in that
you can seek/read/write all from the same file.

Reviewed-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |   4 -
 .../amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 107 ------------------
 2 files changed, 111 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 2e2413fd73a4f..4561f55afa99d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -661,10 +661,6 @@ struct amdgpu_dm_connector {
 	struct mutex hpd_lock;
 
 	bool fake_enable;
-#ifdef CONFIG_DEBUG_FS
-	uint32_t debugfs_dpcd_address;
-	uint32_t debugfs_dpcd_size;
-#endif
 	bool force_yuv420_output;
 	struct dsc_preferred_settings dsc_settings;
 	union dp_downstream_port_present mst_downstream_port_present;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index 827fcb4fb3b3b..82234397dd44e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -1039,88 +1039,6 @@ static ssize_t dp_sdp_message_debugfs_write(struct file *f, const char __user *b
 	return write_size;
 }
 
-static ssize_t dp_dpcd_address_write(struct file *f, const char __user *buf,
-				 size_t size, loff_t *pos)
-{
-	int r;
-	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
-
-	if (size < sizeof(connector->debugfs_dpcd_address))
-		return -EINVAL;
-
-	r = copy_from_user(&connector->debugfs_dpcd_address,
-			buf, sizeof(connector->debugfs_dpcd_address));
-
-	return size - r;
-}
-
-static ssize_t dp_dpcd_size_write(struct file *f, const char __user *buf,
-				 size_t size, loff_t *pos)
-{
-	int r;
-	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
-
-	if (size < sizeof(connector->debugfs_dpcd_size))
-		return -EINVAL;
-
-	r = copy_from_user(&connector->debugfs_dpcd_size,
-			buf, sizeof(connector->debugfs_dpcd_size));
-
-	if (connector->debugfs_dpcd_size > 256)
-		connector->debugfs_dpcd_size = 0;
-
-	return size - r;
-}
-
-static ssize_t dp_dpcd_data_write(struct file *f, const char __user *buf,
-				 size_t size, loff_t *pos)
-{
-	int r;
-	char *data;
-	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
-	struct dc_link *link = connector->dc_link;
-	uint32_t write_size = connector->debugfs_dpcd_size;
-
-	if (!write_size || size < write_size)
-		return -EINVAL;
-
-	data = kzalloc(write_size, GFP_KERNEL);
-	if (!data)
-		return 0;
-
-	r = copy_from_user(data, buf, write_size);
-
-	dm_helpers_dp_write_dpcd(link->ctx, link,
-			connector->debugfs_dpcd_address, data, write_size - r);
-	kfree(data);
-	return write_size - r;
-}
-
-static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf,
-				 size_t size, loff_t *pos)
-{
-	int r;
-	char *data;
-	struct amdgpu_dm_connector *connector = file_inode(f)->i_private;
-	struct dc_link *link = connector->dc_link;
-	uint32_t read_size = connector->debugfs_dpcd_size;
-
-	if (!read_size || size < read_size)
-		return 0;
-
-	data = kzalloc(read_size, GFP_KERNEL);
-	if (!data)
-		return 0;
-
-	dm_helpers_dp_read_dpcd(link->ctx, link,
-			connector->debugfs_dpcd_address, data, read_size);
-
-	r = copy_to_user(buf, data, read_size);
-
-	kfree(data);
-	return read_size - r;
-}
-
 /* function: Read link's DSC & FEC capabilities
  *
  *
@@ -2682,25 +2600,6 @@ static const struct file_operations sdp_message_fops = {
 	.llseek = default_llseek
 };
 
-static const struct file_operations dp_dpcd_address_debugfs_fops = {
-	.owner = THIS_MODULE,
-	.write = dp_dpcd_address_write,
-	.llseek = default_llseek
-};
-
-static const struct file_operations dp_dpcd_size_debugfs_fops = {
-	.owner = THIS_MODULE,
-	.write = dp_dpcd_size_write,
-	.llseek = default_llseek
-};
-
-static const struct file_operations dp_dpcd_data_debugfs_fops = {
-	.owner = THIS_MODULE,
-	.read = dp_dpcd_data_read,
-	.write = dp_dpcd_data_write,
-	.llseek = default_llseek
-};
-
 static const struct file_operations dp_max_bpc_debugfs_fops = {
 	.owner = THIS_MODULE,
 	.read = dp_max_bpc_read,
@@ -2724,9 +2623,6 @@ static const struct {
 		{"test_pattern", &dp_phy_test_pattern_fops},
 		{"hdcp_sink_capability", &hdcp_sink_capability_fops},
 		{"sdp_message", &sdp_message_fops},
-		{"aux_dpcd_address", &dp_dpcd_address_debugfs_fops},
-		{"aux_dpcd_size", &dp_dpcd_size_debugfs_fops},
-		{"aux_dpcd_data", &dp_dpcd_data_debugfs_fops},
 		{"dsc_clock_en", &dp_dsc_clock_en_debugfs_fops},
 		{"dsc_slice_width", &dp_dsc_slice_width_debugfs_fops},
 		{"dsc_slice_height", &dp_dsc_slice_height_debugfs_fops},
@@ -3025,9 +2921,6 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector)
 				    connector_debugfs_entries[i].fops);
 	}
 
-	connector->debugfs_dpcd_address = 0;
-	connector->debugfs_dpcd_size = 0;
-
 	if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA) {
 		for (i = 0; i < ARRAY_SIZE(hdmi_debugfs_entries); i++) {
 			debugfs_create_file(hdmi_debugfs_entries[i].name,
-- 
GitLab


From 4ba439b0b38456c61505568a7fa4fa364850a236 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:43 +0800
Subject: [PATCH 1449/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn31/dcn31_resource.c:1356:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn31/dcn31_resource.c:894:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn31/dcn31_resource.c | 27 -------------------
 1 file changed, 27 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
index 9c637c895f914..f2d589c505a80 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
@@ -891,24 +891,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.dml_hostvm_override = DML_HOSTVM_OVERRIDE_FALSE,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 	.psr = {
 		.disable_psr = false,
@@ -1353,15 +1335,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn31_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn31_hwseq_create,
-};
-
 static void dcn31_resource_destruct(struct dcn31_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 6c882a573bc1d6130274ef74d1697dd769f6a9e4 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 24 May 2023 09:44:02 -0700
Subject: [PATCH 1450/1791] drm/amdgpu: Fix return types of certain NBIOv7.9
 callbacks

When building with clang's -Wincompatible-function-pointer-types-strict,
which ensures that function pointer signatures match exactly to avoid
tripping clang's Control Flow Integrity (kCFI) checks at run time and
will eventually be turned on for the kernel, the following instances
appear in the NBIOv7.9 code:

  drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c:465:32: error: incompatible function pointer types initializing 'int (*)(struct amdgpu_device *)' with an expression of type 'enum amdgpu_gfx_partition (struct amdgpu_device *)' [-Werror,-Wincompatible-function-pointer-types-strict]
          .get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode,
                                        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c:467:31: error: incompatible function pointer types initializing 'u32 (*)(struct amdgpu_device *, u32 *)' (aka 'unsigned int (*)(struct amdgpu_device *, unsigned int *)') with an expression of type 'enum amdgpu_memory_partition (struct amdgpu_device *, u32 *)' (aka 'enum amdgpu_memory_partition (struct amdgpu_device *, unsigned int *)') [-Werror,-Wincompatible-function-pointer-types-strict]
          .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode,
                                       ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  2 errors generated.

Change the return types of these callbacks to match the prototypes to
clear up the warning and avoid tripping kCFI at run time. Both functions
return a value from ffs(), which is an integer that can fit into either
int or unsigned int.

Fixes: 98a54e88e87f ("drm/amdgpu: add sysfs node for compute partition mode")
Fixes: ea2d2f8ececd ("drm/amdgpu: detect current GPU memory partition mode")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index e082f6343d209..d193254767521 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -382,7 +382,7 @@ static void nbio_v7_9_enable_doorbell_interrupt(struct amdgpu_device *adev,
 			      DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1);
 }
 
-static enum amdgpu_gfx_partition nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
+static int nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
 {
 	u32 tmp, px;
 
@@ -408,8 +408,8 @@ static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
 	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
 }
 
-static enum amdgpu_memory_partition
-nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev, u32 *supp_modes)
+static u32 nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev,
+					       u32 *supp_modes)
 {
 	u32 tmp;
 
-- 
GitLab


From a09e2065101a343ac3a709aa6236cdac874627eb Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 24 May 2023 20:37:26 +0530
Subject: [PATCH 1451/1791] drm/amdgpu: Fix defined but not used gfx9_cs_data
 in gfx_v9_4_3.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc with W=1
In file included from drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:33:
drivers/gpu/drm/amd/amdgpu/clearstate_gfx9.h:939:36: warning: ‘gfx9_cs_data’ defined but not used [-Wunused-const-variable=]
  939 | static const struct cs_section_def gfx9_cs_data[] = {
      |

gfx9_cs_data is not used in gfx_v9_4_3.c, hence remove its
include in gfx_v9_4_3.c

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ed41a7862d9f9..bdaea50cafe7b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -29,7 +29,6 @@
 #include "soc15_common.h"
 #include "vega10_enum.h"
 
-#include "clearstate_gfx9.h"
 #include "v9_structs.h"
 
 #include "ivsrcid/gfx/irqsrcs_gfx_9_0.h"
-- 
GitLab


From 8cce16826f5e154a3463b7eafa0f6beebeb48e49 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 24 May 2023 00:26:28 +0530
Subject: [PATCH 1452/1791] drm/amdgpu: Fix unused variable in amdgpu_gfx.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c: In function ‘amdgpu_gfx_disable_kcq’:
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c:497:6: warning: variable ‘j’ set but not used [-Wunused-but-set-variable]
  497 |  int j;
      |      ^
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c: In function ‘amdgpu_gfx_disable_kgq’:
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c:528:6: warning: variable ‘j’ set but not used [-Wunused-but-set-variable]
  528 |  int j;
      |      ^
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c: In function ‘amdgpu_gfx_enable_kgq’:
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c:630:12: warning: variable ‘j’ set but not used [-Wunused-but-set-variable]
  630 |  int r, i, j;
      |

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 2b4bf6c11ae4e..a33d4bc34cee7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -509,7 +509,7 @@ int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id)
 	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
 		j = i + xcc_id * adev->gfx.num_compute_rings;
 		kiq->pmf->kiq_unmap_queues(kiq_ring,
-					   &adev->gfx.compute_ring[i],
+					   &adev->gfx.compute_ring[j],
 					   RESET_QUEUES, 0, 0);
 	}
 
@@ -541,7 +541,7 @@ int amdgpu_gfx_disable_kgq(struct amdgpu_device *adev, int xcc_id)
 		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
 			j = i + xcc_id * adev->gfx.num_gfx_rings;
 			kiq->pmf->kiq_unmap_queues(kiq_ring,
-						   &adev->gfx.gfx_ring[i],
+						   &adev->gfx.gfx_ring[j],
 						   PREEMPT_QUEUES, 0, 0);
 		}
 	}
@@ -648,7 +648,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id)
 		for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
 			j = i + xcc_id * adev->gfx.num_gfx_rings;
 			kiq->pmf->kiq_map_queues(kiq_ring,
-						 &adev->gfx.gfx_ring[i]);
+						 &adev->gfx.gfx_ring[j]);
 		}
 	}
 
-- 
GitLab


From 423502d4f622ad90414bd38d834763277fbab26c Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:44 +0800
Subject: [PATCH 1453/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource.c:1360:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource.c:737:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn32/dcn32_resource.c | 29 -------------------
 1 file changed, 29 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 33abc8c9d4bec..2e6b39fe26132 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -734,26 +734,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_max_blank_us = 1000,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_dsc_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true,
-	.force_disable_subvp = true
-};
-
 static struct dce_aux *dcn32_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
@@ -1357,15 +1337,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn32_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn32_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn32_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn32_hwseq_create,
-};
-
 static void dcn32_resource_destruct(struct dcn32_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 24bc366a4309f407ea77110ba15e3581005def6b Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:45 +0800
Subject: [PATCH 1454/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn301/dcn301_resource.c:1050:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn301/dcn301_resource.c:705:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn301/dcn301_resource.c   | 24 -------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
index 1cf84a086fecb..3485fbb1093ef 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c
@@ -702,23 +702,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.exit_idle_opt_for_cursor_updates = true
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = false,
-	.disable_hubp_power_gate = false,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = true,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.use_max_lb = false,
-};
-
 static void dcn301_dpp_destroy(struct dpp **dpp)
 {
 	kfree(TO_DCN20_DPP(*dpp));
@@ -1047,13 +1030,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn301_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn301_hwseq_create,
-};
-
 static void dcn301_destruct(struct dcn301_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From b7941e2fef13baabd3eade31601e70adf729e887 Mon Sep 17 00:00:00 2001
From: Jiadong Zhu <Jiadong.Zhu@amd.com>
Date: Wed, 24 May 2023 11:42:19 +0800
Subject: [PATCH 1455/1791] drm/amdgpu: Reset CP_VMID_PREEMPT after trailing
 fence signaled

When MEC executes unmap_queue for mid command buffer preemption, it will
kick the write pointer of the gfx ring, set CP_VMID_PREEMPT to trigger the
preemption and wait for CP_VMID_PREEMPT becomes zero after the preemption
done. There is a race condition that PFP may excute the resetting command
before MEC set CP_VMID_PREEMPT. As a result, hang happens as
CP_VMID_PREEMPT is always 0xffff.

To avoid this, we send resetting CP_VMID_PREEMPT command after the trailing
fence is siganled and update gfx write pointer explicitly.

Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index fe090eafa3665..45fa02063491d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -5400,10 +5400,6 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
 	amdgpu_ring_alloc(ring, 13);
 	gfx_v9_0_ring_emit_fence(ring, ring->trail_fence_gpu_addr,
 				 ring->trail_seq, AMDGPU_FENCE_FLAG_EXEC | AMDGPU_FENCE_FLAG_INT);
-	/*reset the CP_VMID_PREEMPT after trailing fence*/
-	amdgpu_ring_emit_wreg(ring,
-			      SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT),
-			      0x0);
 
 	/* assert IB preemption, emit the trailing fence */
 	kiq->pmf->kiq_unmap_queues(kiq_ring, ring, PREEMPT_QUEUES_NO_UNMAP,
@@ -5426,6 +5422,10 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
 		DRM_WARN("ring %d timeout to preempt ib\n", ring->idx);
 	}
 
+	/*reset the CP_VMID_PREEMPT after trailing fence*/
+	amdgpu_ring_emit_wreg(ring,
+			      SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT),
+			      0x0);
 	amdgpu_ring_commit(ring);
 
 	/* deassert preemption condition */
-- 
GitLab


From cfdce594171cea19ba033e8d7ff57a767c0ccd63 Mon Sep 17 00:00:00 2001
From: Jiadong Zhu <Jiadong.Zhu@amd.com>
Date: Wed, 24 May 2023 16:51:32 +0800
Subject: [PATCH 1456/1791] drm/amdgpu: Program gds backup address as zero if
 no gds allocated

It is firmware requirement to set gds_backup_addrlo and gds_backup_addrhi
of DE meta both zero if no gds partition is allocated for the frame.

Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 45fa02063491d..0189e50bd89f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -765,7 +765,7 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev);
 static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev,
 				struct amdgpu_cu_info *cu_info);
 static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev);
-static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume);
+static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds);
 static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring);
 static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev,
 					  void *ras_error_status);
@@ -5158,7 +5158,8 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 			gfx_v9_0_ring_emit_de_meta(ring,
 						   (!amdgpu_sriov_vf(ring->adev) &&
 						   flags & AMDGPU_IB_PREEMPTED) ?
-						   true : false);
+						   true : false,
+						   job->gds_size > 0 && job->gds_base != 0);
 	}
 
 	amdgpu_ring_write(ring, header);
@@ -5433,7 +5434,7 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring)
 	return r;
 }
 
-static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
+static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct v9_de_ib_state de_payload = {0};
@@ -5464,8 +5465,10 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume)
 				 PAGE_SIZE);
 	}
 
-	de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
-	de_payload.gds_backup_addrhi = upper_32_bits(gds_addr);
+	if (usegds) {
+		de_payload.gds_backup_addrlo = lower_32_bits(gds_addr);
+		de_payload.gds_backup_addrhi = upper_32_bits(gds_addr);
+	}
 
 	cnt = (sizeof(de_payload) >> 2) + 4 - 2;
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, cnt));
-- 
GitLab


From 9535a86a4072babc37dc6bdadae52bdbb88166f5 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Wed, 17 May 2023 14:15:05 +0800
Subject: [PATCH 1457/1791] drm/amdgpu: bypass bios dependent operations

Since bios reading does not work currently so just bypass all operations
related to bios

v2: hardcode the vram info for APP_APU case (hawking)
v3: correct the vram_width with channel number * channel size (lijo)

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 66 ++++++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    |  7 ++-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c      | 63 +++++++++++++--------
 3 files changed, 85 insertions(+), 51 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index ff9ca0dbeb5ae..8f50ca2bee97b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1389,6 +1389,15 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev)
 	return 0;
 }
 
+static bool amdgpu_device_read_bios(struct amdgpu_device *adev)
+{
+	if (hweight32(adev->aid_mask) && (adev->flags & AMD_IS_APU)) {
+		return false;
+	}
+
+	return true;
+}
+
 /*
  * GPU helpers function.
  */
@@ -1408,6 +1417,9 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev)
 	if (amdgpu_sriov_vf(adev))
 		return false;
 
+	if (!amdgpu_device_read_bios(adev))
+		return false;
+
 	if (amdgpu_passthrough(adev)) {
 		/* for FIJI: In whole GPU pass-through virtualization case, after VM reboot
 		 * some old smc fw still need driver do vPost otherwise gpu hang, while
@@ -2318,14 +2330,16 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 				return r;
 
 			/* Read BIOS */
-			if (!amdgpu_get_bios(adev))
-				return -EINVAL;
+			if (amdgpu_device_read_bios(adev)) {
+				if (!amdgpu_get_bios(adev))
+					return -EINVAL;
 
-			r = amdgpu_atombios_init(adev);
-			if (r) {
-				dev_err(adev->dev, "amdgpu_atombios_init failed\n");
-				amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0);
-				return r;
+				r = amdgpu_atombios_init(adev);
+				if (r) {
+					dev_err(adev->dev, "amdgpu_atombios_init failed\n");
+					amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0);
+					return r;
+				}
 			}
 
 			/*get pf2vf msg info at it's earliest time*/
@@ -3945,25 +3959,27 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 		}
 	}
 
-	if (adev->is_atom_fw) {
-		/* Initialize clocks */
-		r = amdgpu_atomfirmware_get_clock_info(adev);
-		if (r) {
-			dev_err(adev->dev, "amdgpu_atomfirmware_get_clock_info failed\n");
-			amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0);
-			goto failed;
-		}
-	} else {
-		/* Initialize clocks */
-		r = amdgpu_atombios_get_clock_info(adev);
-		if (r) {
-			dev_err(adev->dev, "amdgpu_atombios_get_clock_info failed\n");
-			amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0);
-			goto failed;
+	if (adev->bios) {
+		if (adev->is_atom_fw) {
+			/* Initialize clocks */
+			r = amdgpu_atomfirmware_get_clock_info(adev);
+			if (r) {
+				dev_err(adev->dev, "amdgpu_atomfirmware_get_clock_info failed\n");
+				amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0);
+				goto failed;
+			}
+		} else {
+			/* Initialize clocks */
+			r = amdgpu_atombios_get_clock_info(adev);
+			if (r) {
+				dev_err(adev->dev, "amdgpu_atombios_get_clock_info failed\n");
+				amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0);
+				goto failed;
+			}
+			/* init i2c buses */
+			if (!amdgpu_device_has_dc_support(adev))
+				amdgpu_atombios_i2c_init(adev);
 		}
-		/* init i2c buses */
-		if (!amdgpu_device_has_dc_support(adev))
-			amdgpu_atombios_i2c_init(adev);
 	}
 
 fence_driver_init:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 473eeac1f03bc..d2d0d27f90539 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1696,7 +1696,7 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	uint32_t reserve_size = 0;
 	int ret;
 
-	if (!amdgpu_sriov_vf(adev)) {
+	if (adev->bios && !amdgpu_sriov_vf(adev)) {
 		if (amdgpu_atomfirmware_mem_training_supported(adev))
 			mem_train_support = true;
 		else
@@ -1713,7 +1713,10 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev)
 	if (adev->bios)
 		reserve_size =
 			amdgpu_atomfirmware_get_fw_reserved_fb_size(adev);
-	if (!reserve_size)
+
+	if (!adev->bios && adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3))
+		reserve_size = max(reserve_size, (uint32_t)280 << 20);
+	else if (!reserve_size)
 		reserve_size = DISCOVERY_TMR_OFFSET;
 
 	if (mem_train_support) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 1e8b2aaa48c16..be7823d82150d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -2010,34 +2010,49 @@ static int gmc_v9_0_sw_init(void *handle)
 
 	spin_lock_init(&adev->gmc.invalidate_lock);
 
-	r = amdgpu_atomfirmware_get_vram_info(adev,
-		&vram_width, &vram_type, &vram_vendor);
-	if (amdgpu_sriov_vf(adev))
-		/* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN,
-		 * and DF related registers is not readable, seems hardcord is the
-		 * only way to set the correct vram_width
-		 */
-		adev->gmc.vram_width = 2048;
-	else if (amdgpu_emu_mode != 1)
-		adev->gmc.vram_width = vram_width;
+	if (!(adev->bios) || adev->gmc.is_app_apu) {
+		if (adev->flags & AMD_IS_APU) {
+			if (adev->gmc.is_app_apu) {
+				adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM;
+				adev->gmc.vram_width = 128 * 64;
+			} else {
+				adev->gmc.vram_type = AMDGPU_VRAM_TYPE_DDR4;
+				adev->gmc.vram_width = 64 * 64;
+			}
+		} else {
+			adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM;
+			adev->gmc.vram_width = 128 * 64;
+		}
+	} else {
+		r = amdgpu_atomfirmware_get_vram_info(adev,
+			&vram_width, &vram_type, &vram_vendor);
+		if (amdgpu_sriov_vf(adev))
+			/* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN,
+			 * and DF related registers is not readable, seems hardcord is the
+			 * only way to set the correct vram_width
+			 */
+			adev->gmc.vram_width = 2048;
+		else if (amdgpu_emu_mode != 1)
+			adev->gmc.vram_width = vram_width;
 
-	if (!adev->gmc.vram_width) {
-		int chansize, numchan;
+		if (!adev->gmc.vram_width) {
+			int chansize, numchan;
 
-		/* hbm memory channel size */
-		if (adev->flags & AMD_IS_APU)
-			chansize = 64;
-		else
-			chansize = 128;
-		if (adev->df.funcs &&
-		    adev->df.funcs->get_hbm_channel_number) {
-			numchan = adev->df.funcs->get_hbm_channel_number(adev);
-			adev->gmc.vram_width = numchan * chansize;
+			/* hbm memory channel size */
+			if (adev->flags & AMD_IS_APU)
+				chansize = 64;
+			else
+				chansize = 128;
+			if (adev->df.funcs &&
+			    adev->df.funcs->get_hbm_channel_number) {
+				numchan = adev->df.funcs->get_hbm_channel_number(adev);
+				adev->gmc.vram_width = numchan * chansize;
+			}
 		}
-	}
 
-	adev->gmc.vram_type = vram_type;
-	adev->gmc.vram_vendor = vram_vendor;
+		adev->gmc.vram_type = vram_type;
+		adev->gmc.vram_vendor = vram_vendor;
+	}
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 1, 0):
 	case IP_VERSION(9, 2, 2):
-- 
GitLab


From 89f85765555caccec0a31b604639cea53942e522 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Tue, 16 May 2023 10:31:49 +0800
Subject: [PATCH 1458/1791] drm/amdgpu: golden settings for ASIC rev_id 0

Suggested by FW team that GB_ADDR_CONFIG is handled by golden
settings in driver to get the expected value

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index bdaea50cafe7b..35d359de0aada 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -45,6 +45,8 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
 #define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
 
+#define GOLDEN_GB_ADDR_CONFIG 0x2a114042
+
 struct amdgpu_gfx_ras gfx_v9_4_3_ras;
 
 static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev);
@@ -195,6 +197,15 @@ static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev)
 		dev_inst = GET_INST(GC, i);
 		if (dev_inst >= 2)
 			WREG32_SOC15(GC, dev_inst, regGRBM_MCM_ADDR, 0x4);
+
+		/* Golden settings applied by driver for ASIC with rev_id 0 */
+		if (adev->rev_id == 0) {
+			WREG32_SOC15(GC, dev_inst, regGB_ADDR_CONFIG,
+				     GOLDEN_GB_ADDR_CONFIG);
+
+			WREG32_FIELD15_PREREG(GC, dev_inst, TCP_UTCL1_CNTL1,
+					      REDUCE_FIFO_DEPTH_BY_2, 2);
+		}
 	}
 }
 
-- 
GitLab


From 491ae27829cda38df3ab6d2fe5d94a80ec1bbe22 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Wed, 17 May 2023 13:40:04 +0800
Subject: [PATCH 1459/1791] drm/amdgpu: complement the 4, 6 and 8 XCC cases

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 35d359de0aada..ebdd7fa985d60 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1040,6 +1040,9 @@ static void gfx_v9_4_3_xcc_program_xcc_id(struct amdgpu_device *adev,
 		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, 0x8);
 		break;
 	case 2:
+	case 4:
+	case 6:
+	case 8:
 		tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID);
 		tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP));
 		WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, tmp);
-- 
GitLab


From 28bb7f13e70dcd3a6c736ee1567cf91c47af2600 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 24 May 2023 10:48:40 -0400
Subject: [PATCH 1460/1791] drm/jpeg: add init value for num_jpeg_rings

Need init new num_jpeg_rings to 1 on jpeg.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Richard Liang <rliang1@amd.com>
Tested-by: Ying Li <ying.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 1 +
 drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c | 1 +
 5 files changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index 1c5b60604a193..77595e9622da3 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -460,6 +460,7 @@ int jpeg_v1_0_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	adev->jpeg.num_jpeg_inst = 1;
+	adev->jpeg.num_jpeg_rings = 1;
 
 	jpeg_v1_0_set_dec_ring_funcs(adev);
 	jpeg_v1_0_set_irq_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
index 3aeeceae34a5e..c25d4a07350b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
@@ -49,6 +49,7 @@ static int jpeg_v2_0_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	adev->jpeg.num_jpeg_inst = 1;
+	adev->jpeg.num_jpeg_rings = 1;
 
 	jpeg_v2_0_set_dec_ring_funcs(adev);
 	jpeg_v2_0_set_irq_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index f533ede484d4e..aadb74de52bcf 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -60,6 +60,7 @@ static int jpeg_v2_5_early_init(void *handle)
 	u32 harvest;
 	int i;
 
+	adev->jpeg.num_jpeg_rings = 1;
 	adev->jpeg.num_jpeg_inst = JPEG25_MAX_HW_INSTANCES_ARCTURUS;
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) {
 		harvest = RREG32_SOC15(JPEG, i, mmCC_UVD_HARVESTING);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index cb5494effc0fd..79791379fc2b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -64,6 +64,7 @@ static int jpeg_v3_0_early_init(void *handle)
 	}
 
 	adev->jpeg.num_jpeg_inst = 1;
+	adev->jpeg.num_jpeg_rings = 1;
 
 	jpeg_v3_0_set_dec_ring_funcs(adev);
 	jpeg_v3_0_set_irq_funcs(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index 86383fbf9358a..a707d407fbd0d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -58,6 +58,7 @@ static int jpeg_v4_0_early_init(void *handle)
 
 
 	adev->jpeg.num_jpeg_inst = 1;
+	adev->jpeg.num_jpeg_rings = 1;
 
 	jpeg_v4_0_set_dec_ring_funcs(adev);
 	jpeg_v4_0_set_irq_funcs(adev);
-- 
GitLab


From 20997c04b7168b1833da77f882eb5a6f246c4b96 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 22 May 2023 17:11:59 +0800
Subject: [PATCH 1461/1791] drm/amdgpu: set the APU flag based on package type

Since currently APU and dGPU share the same pcie class
while gmc init needs the flag to set up correctly for upcomming
memory allocations

v2: call get_pkg_type in smuio 13_0_3 is enough (hawking)

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index a0685a5ebda66..859882109f55d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -2500,6 +2500,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		break;
 	case IP_VERSION(13, 0, 3):
 		adev->smuio.funcs = &smuio_v13_0_3_funcs;
+		if (adev->smuio.funcs->get_pkg_type(adev) == AMDGPU_PKG_TYPE_APU) {
+			adev->flags |= AMD_IS_APU;
+		}
 		break;
 	case IP_VERSION(13, 0, 6):
 	case IP_VERSION(13, 0, 8):
-- 
GitLab


From 23105541727cd6b702c9ee66d98ba50a129fbd5e Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 25 Apr 2023 16:55:56 -0400
Subject: [PATCH 1462/1791] drm/amdgpu: save/restore part of xcp drm_device
 fields

Redirect xcp allocated drm_device::rdev/pdev/driver with
amdgpu pci_device/drm_device setting. They need be saved
before redirect and restored after unregister xcp drm_device.

-v2: fix warning discarded-qualifiers

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 16 ++++++++++++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  4 ++++
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index fcdc0862d2584..86087faab689f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -240,9 +240,14 @@ static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev)
 			return PTR_ERR(p_ddev);
 
 		/* Redirect all IOCTLs to the primary device */
+		adev->xcp_mgr->xcp[i].rdev = p_ddev->render->dev;
+		adev->xcp_mgr->xcp[i].pdev = p_ddev->primary->dev;
+		adev->xcp_mgr->xcp[i].driver = (struct drm_driver *)p_ddev->driver;
+		adev->xcp_mgr->xcp[i].vma_offset_manager = p_ddev->vma_offset_manager;
 		p_ddev->render->dev = ddev;
 		p_ddev->primary->dev = ddev;
 		p_ddev->vma_offset_manager = ddev->vma_offset_manager;
+		p_ddev->driver = &amdgpu_partition_driver;
 		adev->xcp_mgr->xcp[i].ddev = p_ddev;
 	}
 
@@ -330,13 +335,20 @@ int amdgpu_xcp_dev_register(struct amdgpu_device *adev,
 
 void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev)
 {
+	struct drm_device *p_ddev;
 	int i;
 
 	if (!adev->xcp_mgr)
 		return;
 
-	for (i = 0; i < MAX_XCP; i++)
-		drm_dev_unplug(adev->xcp_mgr->xcp[i].ddev);
+	for (i = 0; i < MAX_XCP; i++) {
+		p_ddev = adev->xcp_mgr->xcp[i].ddev;
+		drm_dev_unplug(p_ddev);
+		p_ddev->render->dev = adev->xcp_mgr->xcp[i].rdev;
+		p_ddev->primary->dev = adev->xcp_mgr->xcp[i].pdev;
+		p_ddev->driver =  adev->xcp_mgr->xcp[i].driver;
+		p_ddev->vma_offset_manager = adev->xcp_mgr->xcp[i].vma_offset_manager;
+	}
 }
 
 int amdgpu_xcp_open_device(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 9c5912b9d8bd0..0f8026d64ea51 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -77,6 +77,10 @@ struct amdgpu_xcp {
 	bool valid;
 	atomic_t	ref_cnt;
 	struct drm_device *ddev;
+	struct drm_device *rdev;
+	struct drm_device *pdev;
+	struct drm_driver *driver;
+	struct drm_vma_offset_manager *vma_offset_manager;
 	struct amdgpu_sched	gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX];
 };
 
-- 
GitLab


From 5d6cd20075c823565e7550f8de70d7615ec3c8b7 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Mon, 22 May 2023 15:58:10 +0800
Subject: [PATCH 1463/1791] drm/amdgpu: add the accelerator pcie class

v2: add the base class id for accelerator (lijo)
v3: add the new pci class in amdgpu tree (hawking)

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c  | 5 +++++
 drivers/gpu/drm/amd/include/amd_shared.h | 1 +
 2 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 7300df2a342c7..422c36ed8f365 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2041,6 +2041,11 @@ static const struct pci_device_id pciidlist[] = {
 	  .class_mask = 0xffffff,
 	  .driver_data = CHIP_IP_DISCOVERY },
 
+	{ PCI_DEVICE(0x1002, PCI_ANY_ID),
+	  .class = AMD_ACCELERATOR_PROCESSING << 8,
+	  .class_mask = 0xffffff,
+	  .driver_data = CHIP_IP_DISCOVERY },
+
 	{0, 0, 0}
 };
 
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index f175e65b853a0..57d95e2cc54bb 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -27,6 +27,7 @@
 
 
 #define AMD_MAX_USEC_TIMEOUT		1000000  /* 1000 ms */
+#define AMD_ACCELERATOR_PROCESSING	0x1200   /* hardcoded pci class */
 
 /*
  * Chip flags
-- 
GitLab


From e0ac8656e75f1f26833c14313811043e36b0bba8 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:47 +0800
Subject: [PATCH 1464/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn303/dcn303_resource.c:884:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn303/dcn303_resource.c:84:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn303/dcn303_resource.c   | 24 -------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
index fcd1266021789..f35514188a5cd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
@@ -81,23 +81,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.disable_idle_power_optimizations = false,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-		.disable_dmcu = true,
-		.force_abm_enable = false,
-		.timing_trace = true,
-		.clock_trace = true,
-		.disable_dpp_power_gate = true,
-		.disable_hubp_power_gate = true,
-		.disable_clock_gate = true,
-		.disable_pplib_clock_request = true,
-		.disable_pplib_wm_range = true,
-		.disable_stutter = false,
-		.scl_reset_length10 = true,
-		.dwb_fi_phase = -1, // -1 = disable
-		.dmub_command_table = true,
-		.enable_tri_buf = true,
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 		.psr = {
 			.disable_psr = false,
@@ -881,13 +864,6 @@ static const struct resource_create_funcs res_create_funcs = {
 		.create_hwseq = dcn303_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-		.read_dce_straps = NULL,
-		.create_audio = NULL,
-		.create_stream_encoder = NULL,
-		.create_hwseq = dcn303_hwseq_create,
-};
-
 static bool is_soc_bounding_box_valid(struct dc *dc)
 {
 	uint32_t hw_internal_rev = dc->ctx->asic_id.hw_internal_rev;
-- 
GitLab


From aab9b215d96e0c7e89109821f738f80d84270461 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:46 +0800
Subject: [PATCH 1465/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn302/dcn302_resource.c:957:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn302/dcn302_resource.c:101:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn302/dcn302_resource.c   | 25 -------------------
 1 file changed, 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
index efd98d64588d3..93f42132c900f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
@@ -98,24 +98,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.exit_idle_opt_for_cursor_updates = true
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-		.disable_dmcu = true,
-		.force_abm_enable = false,
-		.timing_trace = true,
-		.clock_trace = true,
-		.disable_dpp_power_gate = true,
-		.disable_hubp_power_gate = true,
-		.disable_clock_gate = true,
-		.disable_pplib_clock_request = true,
-		.disable_pplib_wm_range = true,
-		.disable_stutter = false,
-		.scl_reset_length10 = true,
-		.dwb_fi_phase = -1, // -1 = disable
-		.dmub_command_table = true,
-		.enable_tri_buf = true,
-		.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 		.psr = {
 			.disable_psr = false,
@@ -954,13 +936,6 @@ static const struct resource_create_funcs res_create_funcs = {
 		.create_hwseq = dcn302_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-		.read_dce_straps = NULL,
-		.create_audio = NULL,
-		.create_stream_encoder = NULL,
-		.create_hwseq = dcn302_hwseq_create,
-};
-
 static bool is_soc_bounding_box_valid(struct dc *dc)
 {
 	uint32_t hw_internal_rev = dc->ctx->asic_id.hw_internal_rev;
-- 
GitLab


From abc7e24275a35975e58e699ff2afdbcc47e124d6 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:49 +0800
Subject: [PATCH 1466/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn315/dcn315_resource.c:1357:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn315/dcn315_resource.c:893:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn315/dcn315_resource.c   | 27 -------------------
 1 file changed, 27 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index e6d87c162d26c..cb95e978417b6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -890,24 +890,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.psr_power_use_phy_fsm = 0,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 	.psr = {
 		.disable_psr = false,
@@ -1354,15 +1336,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn31_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn31_hwseq_create,
-};
-
 static void dcn315_resource_destruct(struct dcn315_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From c0c2c51c40fef6960c11a3f132acf91878fa0de0 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Wed, 24 May 2023 16:57:08 +0800
Subject: [PATCH 1467/1791] drm/amd/display: clean up some inconsistent
 indenting

No functional modification involved.

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn314/dcn314_fpu.c:269 dcn314_update_bw_bounding_box_fpu() warn: inconsistent indenting.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5305
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
index 318b9c2bc9be3..c9afddd11589b 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c
@@ -265,8 +265,7 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	}
 
 	dcn20_patch_bounding_box(dc, &dcn3_14_soc);
-
-		dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314);
+	dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314);
 }
 
 static bool is_dual_plane(enum surface_pixel_format format)
-- 
GitLab


From 2c4993bf88ef1e0ed6c81d2fb56f30f32c3d2e74 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:50 +0800
Subject: [PATCH 1468/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn316/dcn316_resource.c:1355:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn316/dcn316_resource.c:899:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn316/dcn316_resource.c   | 27 -------------------
 1 file changed, 27 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
index a3be61cc541f7..b6e0aa2ab27a0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
@@ -886,24 +886,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	},
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true
-};
-
 static const struct dc_panel_config panel_config_defaults = {
 	.psr = {
 		.disable_psr = false,
@@ -1352,15 +1334,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn31_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn31_hwseq_create,
-};
-
 static void dcn316_resource_destruct(struct dcn316_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 3a10a44a3e00d0227d13210ffeef50daa3a326bb Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Wed, 24 May 2023 16:57:09 +0800
Subject: [PATCH 1469/1791] drm/amd/display: clean up some inconsistent
 indenting

No functional modification involved.

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn321/dcn321_fpu.c:556 dcn321_update_bw_bounding_box_fpu() warn: inconsistent indenting.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5304
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    | 242 +++++++++---------
 1 file changed, 121 insertions(+), 121 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index ffd7c3c1b1420..1aaff6f2d4532 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -553,148 +553,148 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 	dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
 
 	/* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */
-		if (dc->debug.use_legacy_soc_bb_mechanism) {
-			unsigned int i = 0, j = 0, num_states = 0;
-
-			unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0};
-			unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0};
-			unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0};
-			unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0};
-
-			unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {615, 906, 1324, 1564};
-			unsigned int num_dcfclk_sta_targets = 4, num_uclk_states = 0;
-			unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0;
-
-			for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
-				if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz)
-					max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
-				if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz)
-					max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
-				if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz)
-					max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
-				if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz)
-					max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
-			}
-			if (!max_dcfclk_mhz)
-				max_dcfclk_mhz = dcn3_21_soc.clock_limits[0].dcfclk_mhz;
-			if (!max_dispclk_mhz)
-				max_dispclk_mhz = dcn3_21_soc.clock_limits[0].dispclk_mhz;
-			if (!max_dppclk_mhz)
-				max_dppclk_mhz = dcn3_21_soc.clock_limits[0].dppclk_mhz;
-			if (!max_phyclk_mhz)
-				max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz;
-
-			if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
-				// If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array
-				dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz;
-				num_dcfclk_sta_targets++;
-			} else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
-				// If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates
-				for (i = 0; i < num_dcfclk_sta_targets; i++) {
-					if (dcfclk_sta_targets[i] > max_dcfclk_mhz) {
-						dcfclk_sta_targets[i] = max_dcfclk_mhz;
-						break;
-					}
+	if (dc->debug.use_legacy_soc_bb_mechanism) {
+		unsigned int i = 0, j = 0, num_states = 0;
+
+		unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0};
+		unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0};
+		unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0};
+		unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0};
+
+		unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {615, 906, 1324, 1564};
+		unsigned int num_dcfclk_sta_targets = 4, num_uclk_states = 0;
+		unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0;
+
+		for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
+			if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz)
+				max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
+			if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz)
+				max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
+			if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz)
+				max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
+			if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz)
+				max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
+		}
+		if (!max_dcfclk_mhz)
+			max_dcfclk_mhz = dcn3_21_soc.clock_limits[0].dcfclk_mhz;
+		if (!max_dispclk_mhz)
+			max_dispclk_mhz = dcn3_21_soc.clock_limits[0].dispclk_mhz;
+		if (!max_dppclk_mhz)
+			max_dppclk_mhz = dcn3_21_soc.clock_limits[0].dppclk_mhz;
+		if (!max_phyclk_mhz)
+			max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz;
+
+		if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
+			// If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array
+			dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz;
+			num_dcfclk_sta_targets++;
+		} else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) {
+			// If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates
+			for (i = 0; i < num_dcfclk_sta_targets; i++) {
+				if (dcfclk_sta_targets[i] > max_dcfclk_mhz) {
+					dcfclk_sta_targets[i] = max_dcfclk_mhz;
+					break;
 				}
-				// Update size of array since we "removed" duplicates
-				num_dcfclk_sta_targets = i + 1;
 			}
+			// Update size of array since we "removed" duplicates
+			num_dcfclk_sta_targets = i + 1;
+		}
 
-			num_uclk_states = bw_params->clk_table.num_entries;
+		num_uclk_states = bw_params->clk_table.num_entries;
 
-			// Calculate optimal dcfclk for each uclk
-			for (i = 0; i < num_uclk_states; i++) {
-				dcn321_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16,
-						&optimal_dcfclk_for_uclk[i], NULL);
-				if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) {
-					optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz;
-				}
+		// Calculate optimal dcfclk for each uclk
+		for (i = 0; i < num_uclk_states; i++) {
+			dcn321_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16,
+					&optimal_dcfclk_for_uclk[i], NULL);
+			if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) {
+				optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz;
 			}
+		}
 
-			// Calculate optimal uclk for each dcfclk sta target
-			for (i = 0; i < num_dcfclk_sta_targets; i++) {
-				for (j = 0; j < num_uclk_states; j++) {
-					if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
-						optimal_uclk_for_dcfclk_sta_targets[i] =
-								bw_params->clk_table.entries[j].memclk_mhz * 16;
-						break;
-					}
+		// Calculate optimal uclk for each dcfclk sta target
+		for (i = 0; i < num_dcfclk_sta_targets; i++) {
+			for (j = 0; j < num_uclk_states; j++) {
+				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
+					optimal_uclk_for_dcfclk_sta_targets[i] =
+							bw_params->clk_table.entries[j].memclk_mhz * 16;
+					break;
 				}
 			}
+		}
 
-			i = 0;
-			j = 0;
-			// create the final dcfclk and uclk table
-			while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
-					dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
-					dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+		i = 0;
+		j = 0;
+		// create the final dcfclk and uclk table
+		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
+				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+			} else {
+				if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) {
+					dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
+					dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
 				} else {
-					if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) {
-						dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
-						dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
-					} else {
-						j = num_uclk_states;
-					}
+					j = num_uclk_states;
 				}
 			}
+		}
 
-			while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) {
-				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
-				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
-			}
+		while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) {
+			dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
+			dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
+		}
 
-			while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES &&
-					optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) {
-				dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
-				dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
-			}
+		while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES &&
+				optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) {
+			dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j];
+			dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16;
+		}
 
-			dcn3_21_soc.num_states = num_states;
-			for (i = 0; i < dcn3_21_soc.num_states; i++) {
-				dcn3_21_soc.clock_limits[i].state = i;
-				dcn3_21_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i];
-				dcn3_21_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i];
-
-				/* Fill all states with max values of all these clocks */
-				dcn3_21_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
-				dcn3_21_soc.clock_limits[i].dppclk_mhz  = max_dppclk_mhz;
-				dcn3_21_soc.clock_limits[i].phyclk_mhz  = max_phyclk_mhz;
-				dcn3_21_soc.clock_limits[i].dscclk_mhz  = max_dispclk_mhz / 3;
-
-				/* Populate from bw_params for DTBCLK, SOCCLK */
-				if (i > 0) {
-					if (!bw_params->clk_table.entries[i].dtbclk_mhz) {
-						dcn3_21_soc.clock_limits[i].dtbclk_mhz  = dcn3_21_soc.clock_limits[i-1].dtbclk_mhz;
-					} else {
-						dcn3_21_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
-					}
-				} else if (bw_params->clk_table.entries[i].dtbclk_mhz) {
+		dcn3_21_soc.num_states = num_states;
+		for (i = 0; i < dcn3_21_soc.num_states; i++) {
+			dcn3_21_soc.clock_limits[i].state = i;
+			dcn3_21_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i];
+			dcn3_21_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i];
+
+			/* Fill all states with max values of all these clocks */
+			dcn3_21_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz;
+			dcn3_21_soc.clock_limits[i].dppclk_mhz  = max_dppclk_mhz;
+			dcn3_21_soc.clock_limits[i].phyclk_mhz  = max_phyclk_mhz;
+			dcn3_21_soc.clock_limits[i].dscclk_mhz  = max_dispclk_mhz / 3;
+
+			/* Populate from bw_params for DTBCLK, SOCCLK */
+			if (i > 0) {
+				if (!bw_params->clk_table.entries[i].dtbclk_mhz) {
+					dcn3_21_soc.clock_limits[i].dtbclk_mhz  = dcn3_21_soc.clock_limits[i-1].dtbclk_mhz;
+				} else {
 					dcn3_21_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
 				}
+			} else if (bw_params->clk_table.entries[i].dtbclk_mhz) {
+				dcn3_21_soc.clock_limits[i].dtbclk_mhz  = bw_params->clk_table.entries[i].dtbclk_mhz;
+			}
 
-				if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0)
-					dcn3_21_soc.clock_limits[i].socclk_mhz = dcn3_21_soc.clock_limits[i-1].socclk_mhz;
-				else
-					dcn3_21_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz;
+			if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0)
+				dcn3_21_soc.clock_limits[i].socclk_mhz = dcn3_21_soc.clock_limits[i-1].socclk_mhz;
+			else
+				dcn3_21_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz;
 
-				if (!dram_speed_mts[i] && i > 0)
-					dcn3_21_soc.clock_limits[i].dram_speed_mts = dcn3_21_soc.clock_limits[i-1].dram_speed_mts;
-				else
-					dcn3_21_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i];
+			if (!dram_speed_mts[i] && i > 0)
+				dcn3_21_soc.clock_limits[i].dram_speed_mts = dcn3_21_soc.clock_limits[i-1].dram_speed_mts;
+			else
+				dcn3_21_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i];
 
-				/* These clocks cannot come from bw_params, always fill from dcn3_21_soc[0] */
-				/* PHYCLK_D18, PHYCLK_D32 */
-				dcn3_21_soc.clock_limits[i].phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz;
-				dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
-			}
-		} else {
-			build_synthetic_soc_states(bw_params, dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states);
+			/* These clocks cannot come from bw_params, always fill from dcn3_21_soc[0] */
+			/* PHYCLK_D18, PHYCLK_D32 */
+			dcn3_21_soc.clock_limits[i].phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz;
+			dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
 		}
+	} else {
+		build_synthetic_soc_states(bw_params, dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states);
+	}
 
-		/* Re-init DML with updated bb */
-		dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
-		if (dc->current_state)
-			dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
+	/* Re-init DML with updated bb */
+	dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
+	if (dc->current_state)
+		dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32);
 }
 
-- 
GitLab


From 70a6267753c1f755157e3bfb63d8fce4137729c6 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:51 +0800
Subject: [PATCH 1470/1791] drm/amd/display: remove unused definition

Eliminate the following warnings:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn321/dcn321_resource.c:1346:43: warning: unused variable 'res_create_maximus_funcs'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn321/dcn321_resource.c:735:38: warning: unused variable 'debug_defaults_diags'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dcn321/dcn321_resource.c   | 30 -------------------
 1 file changed, 30 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index af0bb3e94250a..bbcd3579fea61 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -732,27 +732,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_max_blank_us = 1000,
 };
 
-static const struct dc_debug_options debug_defaults_diags = {
-	.disable_dmcu = true,
-	.force_abm_enable = false,
-	.timing_trace = true,
-	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
-	.disable_dsc_power_gate = true,
-	.disable_clock_gate = true,
-	.disable_pplib_clock_request = true,
-	.disable_pplib_wm_range = true,
-	.disable_stutter = false,
-	.scl_reset_length10 = true,
-	.dwb_fi_phase = -1, // -1 = disable
-	.dmub_command_table = true,
-	.enable_tri_buf = true,
-	.use_max_lb = true,
-	.force_disable_subvp = true,
-};
-
-
 static struct dce_aux *dcn321_aux_engine_create(
 	struct dc_context *ctx,
 	uint32_t inst)
@@ -1343,15 +1322,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn321_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn321_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn321_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn321_hwseq_create,
-};
-
 static void dcn321_resource_destruct(struct dcn321_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 3034983db355daefc4463defce802b8e6d86539f Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 19 May 2023 18:10:40 +0530
Subject: [PATCH 1471/1791] drm/amdgpu: Mark mmhub_v1_8_mmea_err_status_reg as
 __maybe_unused
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Silencing the compiler from below compilation error:

drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c:704:23: error: variable 'mmhub_v1_8_mmea_err_status_reg' is not needed and will not be emitted [-Werror,-Wunneeded-internal-declaration]
static const uint32_t mmhub_v1_8_mmea_err_status_reg[] = {
                      ^
1 error generated.

Mark the variable as __maybe_unused to make it clear to clang that this
is expected, so there is no more warning.

Cc: Christian König <christian.koenig@amd.com>
Cc: Lijo Lazar <lijo.lazar@amd.com>
Cc: Luben Tuikov <luben.tuikov@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
index bf3b0f382c896..5e8b493f86995 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c
@@ -701,7 +701,7 @@ static void mmhub_v1_8_reset_ras_error_count(struct amdgpu_device *adev)
 		mmhub_v1_8_inst_reset_ras_error_count(adev, i);
 }
 
-static const uint32_t mmhub_v1_8_mmea_err_status_reg[] = {
+static const u32 mmhub_v1_8_mmea_err_status_reg[] __maybe_unused = {
 	regMMEA0_ERR_STATUS,
 	regMMEA1_ERR_STATUS,
 	regMMEA2_ERR_STATUS,
-- 
GitLab


From 3b60b70dbec9bb2450ecf012a7b8b6e5dce7168d Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 25 Apr 2023 17:00:50 -0400
Subject: [PATCH 1472/1791] drm/amdxcp: add platform device driver for amdxcp

Add platform device driver for amdxcp to support
amdgpu spatial partition.

-v2: fix build warning

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/Makefile                    |   1 +
 drivers/gpu/drm/amd/amdxcp/Makefile         |  25 +++++
 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c | 110 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h |  29 ++++++
 4 files changed, 165 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/amdxcp/Makefile
 create mode 100644 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
 create mode 100644 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index a33257d2bc7fd..b119089c312f9 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -140,6 +140,7 @@ obj-$(CONFIG_DRM_TTM)	+= ttm/
 obj-$(CONFIG_DRM_SCHED)	+= scheduler/
 obj-$(CONFIG_DRM_RADEON)+= radeon/
 obj-$(CONFIG_DRM_AMDGPU)+= amd/amdgpu/
+obj-$(CONFIG_DRM_AMDGPU)+= amd/amdxcp/
 obj-$(CONFIG_DRM_I915)	+= i915/
 obj-$(CONFIG_DRM_KMB_DISPLAY)  += kmb/
 obj-$(CONFIG_DRM_MGAG200) += mgag200/
diff --git a/drivers/gpu/drm/amd/amdxcp/Makefile b/drivers/gpu/drm/amd/amdxcp/Makefile
new file mode 100644
index 0000000000000..5e1bd70748d46
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdxcp/Makefile
@@ -0,0 +1,25 @@
+#
+# Copyright 2023 Advanced Micro Devices, Inc.
+#
+# Permission is hereby granted, free of charge, to any person obtaining a
+# copy of this software and associated documentation files (the "Software"),
+# to deal in the Software without restriction, including without limitation
+# the rights to use, copy, modify, merge, publish, distribute, sublicense,
+# and/or sell copies of the Software, and to permit persons to whom the
+# Software is furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in
+# all copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+# OTHER DEALINGS IN THE SOFTWARE.
+#
+
+amdgpu-y := amdgpu_xcp_drv.o
+
+obj-$(CONFIG_DRM_AMDGPU) += amdgpu_xcp_drv.o
diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
new file mode 100644
index 0000000000000..353597fc908d9
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
@@ -0,0 +1,110 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#include <drm/drm_drv.h>
+
+#include "amdgpu_xcp_drv.h"
+
+#define MAX_XCP_PLATFORM_DEVICE 64
+
+struct xcp_device {
+	struct drm_device drm;
+	struct platform_device *pdev;
+};
+
+static const struct drm_driver amdgpu_xcp_driver = {
+	.driver_features = DRIVER_GEM | DRIVER_RENDER,
+	.name = "amdgpu_xcp_drv",
+	.major = 1,
+	.minor = 0,
+};
+
+static int pdev_num;
+static struct xcp_device *xcp_dev[MAX_XCP_PLATFORM_DEVICE];
+
+int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev)
+{
+	struct platform_device *pdev;
+	struct xcp_device *pxcp_dev;
+	int ret;
+
+	if (pdev_num >= MAX_XCP_PLATFORM_DEVICE)
+		return -ENODEV;
+
+	pdev = platform_device_register_simple("amdgpu_xcp", pdev_num, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) {
+		ret = -ENOMEM;
+		goto out_unregister;
+	}
+
+	pxcp_dev = devm_drm_dev_alloc(&pdev->dev, &amdgpu_xcp_driver, struct xcp_device, drm);
+	if (IS_ERR(pxcp_dev)) {
+		ret = PTR_ERR(pxcp_dev);
+		goto out_devres;
+	}
+
+	xcp_dev[pdev_num] = pxcp_dev;
+	xcp_dev[pdev_num]->pdev = pdev;
+	*ddev = &pxcp_dev->drm;
+	pdev_num++;
+
+	return 0;
+
+out_devres:
+	devres_release_group(&pdev->dev, NULL);
+out_unregister:
+	platform_device_unregister(pdev);
+
+	return ret;
+}
+EXPORT_SYMBOL(amdgpu_xcp_drm_dev_alloc);
+
+void amdgpu_xcp_drv_release(void)
+{
+	for (--pdev_num; pdev_num >= 0; --pdev_num) {
+		devres_release_group(&xcp_dev[pdev_num]->pdev->dev, NULL);
+		platform_device_unregister(xcp_dev[pdev_num]->pdev);
+		xcp_dev[pdev_num]->pdev = NULL;
+		xcp_dev[pdev_num] = NULL;
+	}
+	pdev_num = 0;
+}
+EXPORT_SYMBOL(amdgpu_xcp_drv_release);
+
+static void __exit amdgpu_xcp_drv_exit(void)
+{
+	amdgpu_xcp_drv_release();
+}
+
+module_exit(amdgpu_xcp_drv_exit);
+
+MODULE_AUTHOR("AMD linux driver team");
+MODULE_DESCRIPTION("AMD XCP PLATFORM DEVICES");
+MODULE_LICENSE("GPL and additional rights");
diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h
new file mode 100644
index 0000000000000..c1c4b679bf95c
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef _AMDGPU_XCP_DRV_H_
+#define _AMDGPU_XCP_DRV_H_
+
+int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev);
+void amdgpu_xcp_drv_release(void);
+#endif /* _AMDGPU_XCP_DRV_H_ */
-- 
GitLab


From ab1270a29b4f2b23aaa28d590d8361903c68b770 Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:48 +0800
Subject: [PATCH 1473/1791] drm/amd/display: remove unused definition

Eliminate the following warning:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn314/dcn314_resource.c:1390:43: warning: unused variable 'res_create_maximus_funcs'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 2483d37e425da..3592efcc7fae0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -1387,15 +1387,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn314_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create,
-	.create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create,
-	.create_hwseq = dcn314_hwseq_create,
-};
-
 static void dcn314_resource_destruct(struct dcn314_resource_pool *pool)
 {
 	unsigned int i;
-- 
GitLab


From 9938333a46c9e20539c85ca7df42a739541b0493 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Tue, 25 Apr 2023 17:02:48 -0400
Subject: [PATCH 1474/1791] drm/amdgpu: use amdxcp platform device as spatial
 partition

Use amdxcp platform device as spatial partition device.

-v2: remove unused variable

Signed-off-by: James Zhu <James.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 12 +++++-------
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 422c36ed8f365..7489b2b1a0d01 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -50,6 +50,7 @@
 #include "amdgpu_ras.h"
 #include "amdgpu_xgmi.h"
 #include "amdgpu_reset.h"
+#include "../amdxcp/amdgpu_xcp_drv.h"
 
 /*
  * KMS wrapper.
@@ -2948,6 +2949,7 @@ static void __exit amdgpu_exit(void)
 	amdgpu_sync_fini();
 	amdgpu_fence_slab_fini();
 	mmu_notifier_synchronize();
+	amdgpu_xcp_drv_release();
 }
 
 module_init(amdgpu_init);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 86087faab689f..d733fa6e7477e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -25,6 +25,7 @@
 #include "amdgpu_drv.h"
 
 #include <drm/drm_drv.h>
+#include "../amdxcp/amdgpu_xcp_drv.h"
 
 static int __amdgpu_xcp_run(struct amdgpu_xcp_mgr *xcp_mgr,
 			    struct amdgpu_xcp_ip *xcp_ip, int xcp_state)
@@ -226,18 +227,15 @@ int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags)
 static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev)
 {
 	struct drm_device *p_ddev;
-	struct pci_dev *pdev;
 	struct drm_device *ddev;
-	int i;
+	int i, ret;
 
-	pdev = adev->pdev;
 	ddev = adev_to_drm(adev);
 
 	for (i = 0; i < MAX_XCP; i++) {
-		p_ddev = drm_dev_alloc(&amdgpu_partition_driver,
-			&pci_upstream_bridge(pdev)->dev);
-		if (IS_ERR(p_ddev))
-			return PTR_ERR(p_ddev);
+		ret = amdgpu_xcp_drm_dev_alloc(&p_ddev);
+		if (ret)
+			return ret;
 
 		/* Redirect all IOCTLs to the primary device */
 		adev->xcp_mgr->xcp[i].rdev = p_ddev->render->dev;
-- 
GitLab


From 321488d180c2f5c1811a0ba7b18d18c7af87739b Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 24 May 2023 11:59:52 +0800
Subject: [PATCH 1475/1791] drm/amd/display: remove unused definition

Eliminate the following warning:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn201/dcn201_resource.c:899:43: warning: unused variable 'res_create_maximus_funcs'

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5296
Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc")
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
index 212c475d95cbc..5ff09bf4bc308 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
@@ -896,13 +896,6 @@ static const struct resource_create_funcs res_create_funcs = {
 	.create_hwseq = dcn201_hwseq_create,
 };
 
-static const struct resource_create_funcs res_create_maximus_funcs = {
-	.read_dce_straps = NULL,
-	.create_audio = NULL,
-	.create_stream_encoder = NULL,
-	.create_hwseq = dcn201_hwseq_create,
-};
-
 static void dcn201_clock_source_destroy(struct clock_source **clk_src)
 {
 	kfree(TO_DCE110_CLK_SRC(*clk_src));
-- 
GitLab


From 9c9d501b28a00f4365632260df6cae488a905af7 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 25 May 2023 11:04:51 +0300
Subject: [PATCH 1476/1791] drm/amd/amdgpu: Fix up locking etc in
 amdgpu_debugfs_gprwave_ioctl()

There are two bugs here.
1) Drop the lock if copy_from_user() fails.
2) If the copy fails then the correct error code is -EFAULT instead of
   -EINVAL.

I also broke up the long line and changed "sizeof rd->id" to
"sizeof(rd->id)".

Fixes: 553f973a0d7b ("drm/amd/amdgpu: Update debugfs for XCC support (v3)")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index c657bed350acc..56e89e76ff179 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -478,15 +478,16 @@ static ssize_t amdgpu_debugfs_gprwave_read(struct file *f, char __user *buf, siz
 static long amdgpu_debugfs_gprwave_ioctl(struct file *f, unsigned int cmd, unsigned long data)
 {
 	struct amdgpu_debugfs_gprwave_data *rd = f->private_data;
-	int r;
+	int r = 0;
 
 	mutex_lock(&rd->lock);
 
 	switch (cmd) {
 	case AMDGPU_DEBUGFS_GPRWAVE_IOC_SET_STATE:
-		r = copy_from_user(&rd->id, (struct amdgpu_debugfs_gprwave_iocdata *)data, sizeof rd->id);
-		if (r)
-			return r ? -EINVAL : 0;
+		if (copy_from_user(&rd->id,
+				   (struct amdgpu_debugfs_gprwave_iocdata *)data,
+				   sizeof(rd->id)))
+			r = -EFAULT;
 		goto done;
 	default:
 		r = -EINVAL;
-- 
GitLab


From 8ffd6f0442674f32c048ec8dffdbc5ec67829beb Mon Sep 17 00:00:00 2001
From: Guchun Chen <guchun.chen@amd.com>
Date: Thu, 25 May 2023 17:24:31 +0800
Subject: [PATCH 1477/1791] drm/amdgpu: keep irq count in
 amdgpu_irq_disable_all
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This can clean up all irq warnings because of unbalanced
amdgpu_irq_get/put when unplugging/unbinding device, and leave
irq count decrease in each ip fini function.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 5667053c295c3..3481d2808ce52 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -139,7 +139,6 @@ void amdgpu_irq_disable_all(struct amdgpu_device *adev)
 				continue;
 
 			for (k = 0; k < src->num_types; ++k) {
-				atomic_set(&src->enabled_types[k], 0);
 				r = src->funcs->set(adev, src, k,
 						    AMDGPU_IRQ_STATE_DISABLE);
 				if (r)
-- 
GitLab


From 28ebbb4981cb1fad12e0b1227dbecc88810b1ee8 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 24 May 2023 14:30:12 -0400
Subject: [PATCH 1478/1791] drm/amdkfd: fix gfx_target_version for certain
 11.0.3 devices

Certain boards with GC IP 11.0.3 need slightly different handling
in the shader compiler due to board specific bounding box
optimizations.

Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 862a50f7b490d..ebc3c3f965f94 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -411,8 +411,15 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 			f2g = &gfx_v11_kfd2kgd;
 			break;
 		case IP_VERSION(11, 0, 3):
-			/* Note: Compiler version is 11.0.1 while HW version is 11.0.3 */
-			gfx_target_version = 110001;
+			if ((adev->pdev->device == 0x7460 &&
+			     adev->pdev->revision == 0x00) ||
+			    (adev->pdev->device == 0x7461 &&
+			     adev->pdev->revision == 0x00))
+				/* Note: Compiler version is 11.0.5 while HW version is 11.0.3 */
+				gfx_target_version = 110005;
+			else
+				/* Note: Compiler version is 11.0.1 while HW version is 11.0.3 */
+				gfx_target_version = 110001;
 			f2g = &gfx_v11_kfd2kgd;
 			break;
 		default:
-- 
GitLab


From 3a25071a970885a2bd3f63cfc7c729e0d536e10f Mon Sep 17 00:00:00 2001
From: Ikshwaku Chauhan <ikshwaku.chauhan@amd.com>
Date: Thu, 25 May 2023 10:57:26 +0530
Subject: [PATCH 1479/1791] drm/amdgpu: enable tmz by default for GC 11.0.1

Add IP GC 11.0.1 in the list of target to have
tmz enabled by default.

Signed-off-by: Ikshwaku Chauhan <ikshwaku.chauhan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index d12625f1de5ab..d78bd97325434 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -592,6 +592,8 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev)
 	case IP_VERSION(9, 3, 0):
 	/* GC 10.3.7 */
 	case IP_VERSION(10, 3, 7):
+	/* GC 11.0.1 */
+	case IP_VERSION(11, 0, 1):
 		if (amdgpu_tmz == 0) {
 			adev->gmc.tmz_enabled = false;
 			dev_info(adev->dev,
@@ -615,7 +617,6 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev)
 	case IP_VERSION(10, 3, 1):
 	/* YELLOW_CARP*/
 	case IP_VERSION(10, 3, 3):
-	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 4):
 		/* Don't enable it by default yet.
 		 */
-- 
GitLab


From 23616d1ff31d6e8ffd4f1e12b6b1e2e783fa8280 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 25 May 2023 22:32:35 +0530
Subject: [PATCH 1480/1791] drm/amdgpu: Fix up kdoc in sdma_v4_4_2.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Address a bunch of kdoc warnings:

gcc with W=1
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:426: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_gfx_stop'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:457: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_rlc_stop'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:470: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_page_stop'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:506: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_ctx_switch_enable'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:794: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_rlc_resume'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:810: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_load_microcode'
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:854: warning: Function parameter or member 'inst_mask' not described in 'sdma_v4_4_2_inst_start'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index ff41fb577cdd5..8eebf9c2bbcdb 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -418,6 +418,7 @@ static void sdma_v4_4_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
  * sdma_v4_4_2_inst_gfx_stop - stop the gfx async dma engines
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be disabled
  *
  * Stop the gfx async dma ring buffers.
  */
@@ -449,6 +450,7 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev,
  * sdma_v4_4_2_inst_rlc_stop - stop the compute async dma engines
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be disabled
  *
  * Stop the compute async dma queues.
  */
@@ -462,6 +464,7 @@ static void sdma_v4_4_2_inst_rlc_stop(struct amdgpu_device *adev,
  * sdma_v4_4_2_inst_page_stop - stop the page async dma engines
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be disabled
  *
  * Stop the page async dma ring buffers.
  */
@@ -498,6 +501,7 @@ static void sdma_v4_4_2_inst_page_stop(struct amdgpu_device *adev,
  *
  * @adev: amdgpu_device pointer
  * @enable: enable/disable the DMA MEs context switch.
+ * @inst_mask: mask of dma engine instances to be enabled
  *
  * Halt or unhalt the async dma engines context switch.
  */
@@ -785,6 +789,7 @@ static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev)
  * sdma_v4_4_2_inst_rlc_resume - setup and start the async dma engines
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be enabled
  *
  * Set up the compute DMA queues and enable them.
  * Returns 0 for success, error for failure.
@@ -801,6 +806,7 @@ static int sdma_v4_4_2_inst_rlc_resume(struct amdgpu_device *adev,
  * sdma_v4_4_2_inst_load_microcode - load the sDMA ME ucode
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be enabled
  *
  * Loads the sDMA0/1 ucode.
  * Returns 0 for success, -EINVAL if the ucode is not available.
@@ -845,6 +851,7 @@ static int sdma_v4_4_2_inst_load_microcode(struct amdgpu_device *adev,
  * sdma_v4_4_2_inst_start - setup and start the async dma engines
  *
  * @adev: amdgpu_device pointer
+ * @inst_mask: mask of dma engine instances to be enabled
  *
  * Set up the DMA engines and enable them.
  * Returns 0 for success, error for failure.
-- 
GitLab


From 66dadf1ab196fd2cf8c41f07a4745ad7fb84726e Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 25 May 2023 23:30:59 +0530
Subject: [PATCH 1481/1791] drm/amdgpu: Fix up kdoc in amdgpu_acpi.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding & deleting the deviant arguments.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:906: warning: Function parameter or member 'numa_info' not described in 'amdgpu_acpi_get_node_id'
drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:906: warning: Excess function parameter 'nid' description in 'amdgpu_acpi_get_node_id'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 3fb2c3af0998a..be881f21b340a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -894,7 +894,7 @@ static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm)
  * acpi device handle
  *
  * @handle: acpi handle
- * @nid: NUMA Node id returned by the platform firmware
+ * @numa_info: amdgpu_numa_info structure holding numa information
  *
  * Queries the ACPI interface to fetch the corresponding NUMA Node ID for a
  * given amdgpu acpi device.
-- 
GitLab


From 9eba1b8b70f6488e944fdd1928ef758917bf7229 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 25 May 2023 21:37:35 +0530
Subject: [PATCH 1482/1791] drm/amdgpu: Fix up missing kdoc in sdma_v6_0.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Address a bunch of kdoc warnings:

gcc with W=1
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:248: warning: Function parameter or member 'job' not described in 'sdma_v6_0_ring_emit_ib'
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:248: warning: Function parameter or member 'flags' not described in 'sdma_v6_0_ring_emit_ib'
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:946: warning: Function parameter or member 'timeout' not described in 'sdma_v6_0_ring_test_ib'
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:1125: warning: Function parameter or member 'ring' not described in 'sdma_v6_0_ring_pad_ib'
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:1176: warning: Function parameter or member 'vmid' not described in 'sdma_v6_0_ring_emit_vm_flush'
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:1176: warning: Function parameter or member 'pd_addr' not described in 'sdma_v6_0_ring_emit_vm_flush'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index 1c90b5c661fb7..967849c59ebe7 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -238,6 +238,8 @@ static void sdma_v6_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
  *
  * @ring: amdgpu ring pointer
  * @ib: IB object to schedule
+ * @flags: unused
+ * @job: job to retrieve vmid from
  *
  * Schedule an IB in the DMA ring.
  */
@@ -938,6 +940,7 @@ static int sdma_v6_0_ring_test_ring(struct amdgpu_ring *ring)
  * sdma_v6_0_ring_test_ib - test an IB on the DMA engine
  *
  * @ring: amdgpu_ring structure holding ring information
+ * @timeout: timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT
  *
  * Test a simple IB in the DMA ring.
  * Returns 0 on success, error on failure.
@@ -1167,6 +1170,8 @@ static void sdma_v6_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
  * sdma_v6_0_ring_emit_vm_flush - vm flush using sDMA
  *
  * @ring: amdgpu_ring pointer
+ * @vmid: vmid number to use
+ * @pd_addr: address
  *
  * Update the page table base and flush the VM TLB
  * using sDMA.
-- 
GitLab


From 07a1475279244cd8eea81bec44fd5f0a9d6871f8 Mon Sep 17 00:00:00 2001
From: Graham Sider <Graham.Sider@amd.com>
Date: Mon, 3 Apr 2023 15:31:53 -0400
Subject: [PATCH 1483/1791] drm/amdkfd: Add new gfx_target_versions for GC
 9.4.3

For GC 9.4.3, set gfx_target_version to 90402 for rev 1 and later (APU
or dGPU), 90401 for rev 0 dGPU, and 90400 for rev 0 APU.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Acked-by: Alex Deucher <Alexander.Deucher@amd.com>
Reviewed-by: Amber Lin <Amber.Lin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index ebc3c3f965f94..e84ad1c5ef446 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -332,7 +332,9 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 			f2g = &aldebaran_kfd2kgd;
 			break;
 		case IP_VERSION(9, 4, 3):
-			gfx_target_version = 90400;
+			gfx_target_version = adev->rev_id >= 1 ? 90402
+					   : adev->flags & AMD_IS_APU ? 90400
+					   : 90401;
 			f2g = &gc_9_4_3_kfd2kgd;
 			break;
 		/* Navi10 */
-- 
GitLab


From b695c97b580a1949d0dd96aa17b01d4de738eda3 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 29 May 2023 19:18:54 +0530
Subject: [PATCH 1484/1791] drm/amdkfd: Fix MEC pipe interrupt enablement

for_each_inst modifies xcc_mask and therefore the loop doesn't
initialize properly interrupts on all pipes. Keep looping through xcc as
the outer loop to fix this issue.

Fixes: c4050ff1a43e ("drm/amdkfd: Use xcc mask for identifying xcc")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 493b4b66f1803..80cddb46657fa 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1274,11 +1274,12 @@ static void init_interrupts(struct device_queue_manager *dqm)
 	uint32_t xcc_mask = dqm->dev->xcc_mask;
 	unsigned int i, xcc_id;
 
-	for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) {
-		if (is_pipe_enabled(dqm, 0, i)) {
-			for_each_inst(xcc_id, xcc_mask)
+	for_each_inst(xcc_id, xcc_mask) {
+		for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) {
+			if (is_pipe_enabled(dqm, 0, i)) {
 				dqm->dev->kfd2kgd->init_interrupts(
 					dqm->dev->adev, i, xcc_id);
+			}
 		}
 	}
 }
-- 
GitLab


From 837d4e071d250d695eba7a08c55c77f6a5b4bb5e Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 27 May 2023 22:24:53 +0530
Subject: [PATCH 1485/1791] drm/amdgpu: Fix create_dmamap_sg_bo kdoc warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c:292: warning: Cannot understand  * @create_dmamap_sg_bo: Creates a amdgpu_bo object to reflect information

Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index cc37f04651e29..23390b5932fb6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -290,8 +290,9 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo)
 }
 
 /**
- * @create_dmamap_sg_bo: Creates a amdgpu_bo object to reflect information
+ * create_dmamap_sg_bo() - Creates a amdgpu_bo object to reflect information
  * about USERPTR or DOOREBELL or MMIO BO.
+ *
  * @adev: Device for which dmamap BO is being created
  * @mem: BO of peer device that is being DMA mapped. Provides parameters
  *	 in building the dmamap BO
-- 
GitLab


From 21d81681c298e9712fe37df4b001e8476fc7d03b Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 27 May 2023 19:45:52 +0530
Subject: [PATCH 1486/1791] drm/amd/display: Add missing kdoc entries in
 update_planes_and_stream_adapter

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'dc' not described in 'update_planes_and_stream_adapter'
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'update_type' not described in 'update_planes_and_stream_adapter'
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'planes_count' not described in 'update_planes_and_stream_adapter'
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'stream' not described in 'update_planes_and_stream_adapter'
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'stream_update' not described in 'update_planes_and_stream_adapter'
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:374: warning: Function parameter or member 'array_of_surface_update' not described in 'update_planes_and_stream_adapter'

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c98c57deb5ebf..d0532d7e83414 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -365,6 +365,14 @@ static inline void reverse_planes_order(struct dc_surface_update *array_of_surfa
  * adjustments and preparation before calling it. This function is a wrapper
  * for the dc_update_planes_and_stream that does any required configuration
  * before passing control to DC.
+ *
+ * @dc: Display Core control structure
+ * @update_type: specify whether it is FULL/MEDIUM/FAST update
+ * @planes_count: planes count to update
+ * @stream: stream state
+ * @stream_update: stream update
+ * @array_of_surface_update: dc surface update pointer
+ *
  */
 static inline bool update_planes_and_stream_adapter(struct dc *dc,
 						    int update_type,
-- 
GitLab


From 0be5ccd518031be41266ef952db2202900d519cc Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 27 May 2023 22:05:40 +0530
Subject: [PATCH 1487/1791] drm/amd/display: Fix up kdoc formats in dcn32_fpu.c

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2806: warning: Cannot understand  * *************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2855: warning: Cannot understand  * *************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2900: warning: Function parameter or member 'dc' not described in 'dcn32_assign_fpo_vactive_candidate'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2900: warning: Function parameter or member 'context' not described in 'dcn32_assign_fpo_vactive_candidate'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2900: warning: Function parameter or member 'fpo_candidate_stream' not described in 'dcn32_assign_fpo_vactive_candidate'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2929: warning: Function parameter or member 'dc' not described in 'dcn32_find_vactive_pipe'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2929: warning: Function parameter or member 'context' not described in 'dcn32_find_vactive_pipe'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:2929: warning: Function parameter or member 'vactive_margin_req_us' not described in 'dcn32_find_vactive_pipe'

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 33 ++++++++-----------
 1 file changed, 13 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 137ff970c9aae..7e03c844b05e0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2803,12 +2803,11 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe)
 }
 
 /**
- * ************************************************************************************************
  * dcn32_allow_subvp_high_refresh_rate: Determine if the high refresh rate config will allow subvp
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- * @param [in]: pipe: Pipe to be considered for use in subvp
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @pipe: Pipe to be considered for use in subvp
  *
  * On high refresh rate display configs, we will allow subvp under the following conditions:
  * 1. Resolution is 3840x2160, 3440x1440, or 2560x1440
@@ -2817,9 +2816,7 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe)
  * 4. Freesync is inactive
  * 5. For single display cases, freesync must be disabled
  *
- * @return: True if pipe can be used for subvp, false otherwise
- *
- * ************************************************************************************************
+ * Return: True if pipe can be used for subvp, false otherwise
  */
 bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe)
 {
@@ -2852,15 +2849,12 @@ bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context
 }
 
 /**
- * *******************************************************************************************
  * dcn32_determine_max_vratio_prefetch: Determine max Vratio for prefetch by driver policy
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- *
- * @return: Max vratio for prefetch
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
  *
- * *******************************************************************************************
+ * Return: Max vratio for prefetch
  */
 double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *context)
 {
@@ -2890,9 +2884,9 @@ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *conte
  * ActiveMargin <= 0 to be the FPO stream candidate if found.
  *
  *
- * @param [in]: dc - current dc state
- * @param [in]: context - new dc state
- * @param [out]: fpo_candidate_stream - pointer to FPO stream candidate if one is found
+ * @dc: current dc state
+ * @context: new dc state
+ * @fpo_candidate_stream: pointer to FPO stream candidate if one is found
  *
  * Return: void
  */
@@ -2918,10 +2912,9 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co
 /**
  * dcn32_find_vactive_pipe - Determines if the config has a pipe that can switch in VACTIVE
  *
- * @param [in]: dc - current dc state
- * @param [in]: context - new dc state
- * @param [in]: vactive_margin_req_us - The vactive marign required for a vactive pipe to be
- *                                      considered "found"
+ * @dc: current dc state
+ * @context: new dc state
+ * @vactive_margin_req_us: The vactive marign required for a vactive pipe to be considered "found"
  *
  * Return: True if VACTIVE display is found, false otherwise
  */
-- 
GitLab


From c6a64ad9b7f7182b5e2439a740574300b2e61951 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 30 May 2023 11:52:45 +0530
Subject: [PATCH 1488/1791] drm/amdgpu: Initialize xcc mask

For ASICs which are not initialized through discovery, initialize GFX
cluster as 1.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 1 +
 3 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 809558c718e3c..da6caff78c22b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -3028,6 +3028,7 @@ static int gfx_v6_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->gfx.xcc_mask = 1;
 	adev->gfx.num_gfx_rings = GFX6_NUM_GFX_RINGS;
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  GFX6_NUM_COMPUTE_RINGS);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 0f0c12bbe2285..20fcd86a3e798 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -4179,6 +4179,7 @@ static int gfx_v7_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->gfx.xcc_mask = 1;
 	adev->gfx.num_gfx_rings = GFX7_NUM_GFX_RINGS;
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 2f1ef75e126c7..51c1745c83697 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -5262,6 +5262,7 @@ static int gfx_v8_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->gfx.xcc_mask = 1;
 	adev->gfx.num_gfx_rings = GFX8_NUM_GFX_RINGS;
 	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
 					  AMDGPU_MAX_COMPUTE_RINGS);
-- 
GitLab


From c22b044070971e474dd0ff81a9830df93751f726 Mon Sep 17 00:00:00 2001
From: Alex Sierra <alex.sierra@amd.com>
Date: Mon, 29 May 2023 16:01:37 -0500
Subject: [PATCH 1489/1791] drm/amdkfd: flag added to handle errors from svm
 validate and map

If a return error is raised during validation and mapping of a
prange, this flag is set. It is a rare occurrence, but it could happen
when `amdgpu_hmm_range_get_pages_done` returns true. In such cases,
the caller should retry. However, it is important to ensure that the
prange is updated correctly during the retry.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 ++-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.h | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index ee16130ddc75e..9c88d6e90c3b5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -809,7 +809,7 @@ svm_range_is_same_attrs(struct kfd_process *p, struct svm_range *prange,
 		}
 	}
 
-	return true;
+	return !prange->is_error_flag;
 }
 
 /**
@@ -1691,6 +1691,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 unreserve_out:
 	svm_range_unreserve_bos(ctx);
 
+	prange->is_error_flag = !!r;
 	if (!r)
 		prange->validate_timestamp = ktime_get_boottime();
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
index 762679835e31b..21b14510882b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h
@@ -134,6 +134,7 @@ struct svm_range {
 	DECLARE_BITMAP(bitmap_aip, MAX_GPU_INSTANCE);
 	bool				validated_once;
 	bool				mapped_to_gpu;
+	bool				is_error_flag;
 };
 
 static inline void svm_range_lock(struct svm_range *prange)
-- 
GitLab


From ca2943fe0acecfc89937dcf0abef2d7c1bccf9f4 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 29 May 2023 19:29:35 +0530
Subject: [PATCH 1490/1791] drm/amdgpu: Fix missing parameter desc for 'xcc_id'
 in gfx_v7_0.c & amdgpu_rlc.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding 'xcc_id' arguments.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c:1557: warning: Function parameter or member 'xcc_id' not described in 'gfx_v7_0_select_se_sh'
drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c:38: warning: Function parameter or member 'xcc_id' not described in 'amdgpu_gfx_rlc_enter_safe_mode'
drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c:62: warning: Function parameter or member 'xcc_id' not described in 'amdgpu_gfx_rlc_exit_safe_mode'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c | 2 ++
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c   | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
index d3bed9a3e61fa..35e0ae9acadcd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
@@ -31,6 +31,7 @@
  * amdgpu_gfx_rlc_enter_safe_mode - Set RLC into safe mode
  *
  * @adev: amdgpu_device pointer
+ * @xcc_id: xcc accelerated compute core id
  *
  * Set RLC enter into safe mode if RLC is enabled and haven't in safe mode.
  */
@@ -55,6 +56,7 @@ void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id)
  * amdgpu_gfx_rlc_exit_safe_mode - Set RLC out of safe mode
  *
  * @adev: amdgpu_device pointer
+ * @xcc_id: xcc accelerated compute core id
  *
  * Set RLC exit safe mode if RLC is enabled and have entered into safe mode.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 20fcd86a3e798..8c174c11eaee0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -1548,7 +1548,7 @@ static void gfx_v7_0_tiling_mode_table_init(struct amdgpu_device *adev)
  * @sh_num: sh block to address
  * @instance: Certain registers are instanced per SE or SH.
  *            0xffffffff means broadcast to all SEs or SHs (CIK).
- *
+ * @xcc_id: xcc accelerated compute core id
  * Select which SE, SH combinations to address.
  */
 static void gfx_v7_0_select_se_sh(struct amdgpu_device *adev,
-- 
GitLab


From 3eeb0d037a543588cf2b8890224ec26841e1069a Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 30 May 2023 18:17:17 +0530
Subject: [PATCH 1491/1791] drm/amdgpu: Fix up missing kdoc parameter 'inst' in
 get_wave_count() & kgd_gfx_v9_get_cu_occupancy()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding 'inst' arguments to kdocs.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:692: warning: Function parameter or member 'inst' not described in 'get_wave_count'
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:763: warning: Function parameter or member 'inst' not described in 'kgd_gfx_v9_get_cu_occupancy'

Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 9fa9aab22fe98..34bf030f31377 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -685,7 +685,8 @@ static void unlock_spi_csq_mutexes(struct amdgpu_device *adev)
  * @queue_idx: Index of queue in the queue-map bit-field
  * @wave_cnt: Output parameter updated with number of waves in flight
  * @vmid: Output parameter updated with VMID of queue whose wave count
- * is being collected
+ *        is being collected
+ * @inst: xcc's instance number on a multi-XCC setup
  */
 static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
 		int *wave_cnt, int *vmid, uint32_t inst)
@@ -721,9 +722,10 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx,
  * @adev: Handle of device from which to get number of waves in flight
  * @pasid: Identifies the process for which this query call is invoked
  * @pasid_wave_cnt: Output parameter updated with number of waves in flight that
- * belong to process with given pasid
+ *                  belong to process with given pasid
  * @max_waves_per_cu: Output parameter updated with maximum number of waves
- * possible per Compute Unit
+ *                    possible per Compute Unit
+ * @inst: xcc's instance number on a multi-XCC setup
  *
  * Note: It's possible that the device has too many queues (oversubscription)
  * in which case a VMID could be remapped to a different PASID. This could lead
-- 
GitLab


From 1bae03aab2b41770b9198b3ef1ddc7dc7efb0678 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 30 May 2023 14:43:14 +0530
Subject: [PATCH 1492/1791] drm/amdgpu: Fix up missing parameter in kdoc for
 'inst' in gmc_ v7, v8, v9, v10, v11.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding 'inst' arguments to kdocs.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c:428: warning: Function parameter or member 'inst' not described in 'gmc_v7_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c:626: warning: Function parameter or member 'inst' not described in 'gmc_v8_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:423: warning: Function parameter or member 'inst' not described in 'gmc_v10_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c:328: warning: Function parameter or member 'inst' not described in 'gmc_v11_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c:950: warning: Function parameter or member 'inst' not described in 'gmc_v9_0_flush_gpu_tlb_pasid'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c  | 1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c  | 1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c  | 1 +
 5 files changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 01bd456513821..b2e42f1b0f12d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -414,6 +414,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  * @pasid: pasid to be flush
  * @flush_type: the flush type
  * @all_hub: Used with PACKET3_INVALIDATE_TLBS_ALL_HUB()
+ * @inst: is used to select which instance of KIQ to use for the invalidation
  *
  * Flush the TLB for the requested pasid.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 4bf807d825c03..c571f0d959946 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -319,6 +319,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  * @pasid: pasid to be flush
  * @flush_type: the flush type
  * @all_hub: flush all hubs
+ * @inst: is used to select which instance of KIQ to use for the invalidation
  *
  * Flush the TLB for the requested pasid.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 6f53049619cdd..acd2b407860f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -419,6 +419,7 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev)
  * @pasid: pasid to be flush
  * @flush_type: type of flush
  * @all_hub: flush all hubs
+ * @inst: is used to select which instance of KIQ to use for the invalidation
  *
  * Flush the TLB for the requested pasid.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 48475077ca92f..85dead2a57021 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -617,6 +617,7 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev)
  * @pasid: pasid to be flush
  * @flush_type: type of flush
  * @all_hub: flush all hubs
+ * @inst: is used to select which instance of KIQ to use for the invalidation
  *
  * Flush the TLB for the requested pasid.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index be7823d82150d..3ed286b72cae9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -941,6 +941,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
  * @pasid: pasid to be flush
  * @flush_type: the flush type
  * @all_hub: flush all hubs
+ * @inst: is used to select which instance of KIQ to use for the invalidation
  *
  * Flush the TLB for the requested pasid.
  */
-- 
GitLab


From 932fc49479303961c1da54a1112eb26cdc890c76 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 29 May 2023 20:23:33 +0530
Subject: [PATCH 1493/1791] drm/amdgpu: Fix missing parameter desc for 'xcp_id'
 in amdgpu_amdkfd_reserve_mem_limit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding 'xcp_id' argument.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c:160: warning: Function parameter or member 'xcp_id' not described in 'amdgpu_amdkfd_reserve_mem_limit'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 23390b5932fb6..f61527b800e62 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -153,8 +153,11 @@ void amdgpu_amdkfd_reserve_system_mem(uint64_t size)
  * @size: Size of buffer, in bytes, encapsulated by B0. This should be
  * equivalent to amdgpu_bo_size(BO)
  * @alloc_flag: Flag used in allocating a BO as noted above
+ * @xcp_id: xcp_id is used to get xcp from xcp manager, one xcp is
+ * managed as one compute node in driver for app
  *
- * Return: returns -ENOMEM in case of error, ZERO otherwise
+ * Return:
+ *	returns -ENOMEM in case of error, ZERO otherwise
  */
 int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 		uint64_t size, u32 alloc_flag, int8_t xcp_id)
-- 
GitLab


From 7a66ad6c087ee3863cc9a8d696ac2191d1c2e904 Mon Sep 17 00:00:00 2001
From: ZhenGuo Yin <zhenguo.yin@amd.com>
Date: Tue, 9 May 2023 17:42:11 +0800
Subject: [PATCH 1494/1791] drm/amdgpu: set finished fence error if job
 timedout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set finished fence to ETIME error if job timedout.

Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index c3d9d75143f4f..aca3a2bfe8d23 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -65,6 +65,8 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
 	DRM_ERROR("Process information: process %s pid %d thread %s pid %d\n",
 		  ti.process_name, ti.tgid, ti.task_name, ti.pid);
 
+	dma_fence_set_error(&s_job->s_fence->finished, -ETIME);
+
 	if (amdgpu_device_should_recover_gpu(ring->adev)) {
 		struct amdgpu_reset_context reset_context;
 		memset(&reset_context, 0, sizeof(reset_context));
-- 
GitLab


From b3a02e8b61c19a0380870c713bc704d7e4f9e0dd Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 30 May 2023 04:03:16 +0530
Subject: [PATCH 1495/1791] drm/amdgpu: Fix up missing parameters kdoc in
 svm_migrate_vma_to_ram
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by adding & deleting the deviant arguments.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_migrate.c:671: warning: Function parameter or member 'node' not described in 'svm_migrate_vma_to_ram'
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_migrate.c:671: warning: Function parameter or member 'trigger' not described in 'svm_migrate_vma_to_ram'
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_migrate.c:671: warning: Function parameter or member 'fault_page' not described in 'svm_migrate_vma_to_ram'
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_migrate.c:671: warning: Excess function parameter 'adev' description in 'svm_migrate_vma_to_ram'
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_migrate.c:771: warning: Function parameter or member 'fault_page' not described in 'svm_migrate_vram_to_ram'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 35cf6558cf1b4..58d95fb995959 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -651,11 +651,13 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 /**
  * svm_migrate_vma_to_ram - migrate range inside one vma from device to system
  *
- * @adev: amdgpu device to migrate from
  * @prange: svm range structure
  * @vma: vm_area_struct that range [start, end] belongs to
  * @start: range start virtual address in pages
  * @end: range end virtual address in pages
+ * @node: kfd node device to migrate from
+ * @trigger: reason of migration
+ * @fault_page: is from vmf->page, svm_migrate_to_ram(), this is CPU page fault callback
  *
  * Context: Process context, caller hold mmap read lock, prange->migrate_mutex
  *
@@ -760,6 +762,7 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
  * @prange: range structure
  * @mm: process mm, use current->mm if NULL
  * @trigger: reason of migration
+ * @fault_page: is from vmf->page, svm_migrate_to_ram(), this is CPU page fault callback
  *
  * Context: Process context, caller hold mmap read lock, prange->migrate_mutex
  *
-- 
GitLab


From 23ed8833f231a3ea36d1c352737ef6f1eadfc899 Mon Sep 17 00:00:00 2001
From: Bob Zhou <bob.zhou@amd.com>
Date: Tue, 30 May 2023 14:48:02 +0800
Subject: [PATCH 1496/1791] drm/amdxcp: fix Makefile to build amdxcp module

After drm conduct amdgpu Makefile, amdgpu.ko has been created
and "amdgpu-y +=" in amdxcp Makefile isn't used.
So modify amdgpu-y to amdxcp-y and build amdxcp module.

Signed-off-by: Bob Zhou <bob.zhou@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdxcp/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdxcp/Makefile b/drivers/gpu/drm/amd/amdxcp/Makefile
index 5e1bd70748d46..870501a4bb8c0 100644
--- a/drivers/gpu/drm/amd/amdxcp/Makefile
+++ b/drivers/gpu/drm/amd/amdxcp/Makefile
@@ -20,6 +20,6 @@
 # OTHER DEALINGS IN THE SOFTWARE.
 #
 
-amdgpu-y := amdgpu_xcp_drv.o
+amdxcp-y := amdgpu_xcp_drv.o
 
-obj-$(CONFIG_DRM_AMDGPU) += amdgpu_xcp_drv.o
+obj-$(CONFIG_DRM_AMDGPU) += amdxcp.o
-- 
GitLab


From c00ebe9aeec6df816fa8a5a167cd1c102d02dd28 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 14:52:02 +0530
Subject: [PATCH 1497/1791] drm/amd/display: Fix up kdoc formatting in
 dcn32_resource_helpers.c

Fixes the following W=1 kernel build warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:97: warning: Cannot understand  * **************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:264: warning: Cannot understand  * *************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:435: warning: Cannot understand  * *************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:475: warning: Cannot understand  * *************************************************************************

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:599:
warning: Function parameter or member 'dc' not described in
'dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:599:
warning: Function parameter or member 'context' not described in
'dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch'

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:587:
warning: Function parameter or member 'dc' not described in
'dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:587:
warning: Function parameter or member 'context' not described in
'dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch'

Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dcn32/dcn32_resource_helpers.c | 57 ++++++++-----------
 1 file changed, 23 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index a8082580df92a..1d13fd7972122 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -94,18 +94,15 @@ uint32_t dcn32_helper_calculate_mall_bytes_for_cursor(
 }
 
 /**
- * ********************************************************************************************
- * dcn32_helper_calculate_num_ways_for_subvp: Calculate number of ways needed for SubVP
+ * dcn32_helper_calculate_num_ways_for_subvp(): Calculate number of ways needed for SubVP
  *
  * Gets total allocation required for the phantom viewport calculated by DML in bytes and
  * converts to number of cache ways.
  *
- * @param [in] dc: current dc state
- * @param [in] context: new dc state
+ * @dc: current dc state
+ * @context: new dc state
  *
- * @return: number of ways required for SubVP
- *
- * ********************************************************************************************
+ * Return: number of ways required for SubVP
  */
 uint32_t dcn32_helper_calculate_num_ways_for_subvp(
 		struct dc *dc,
@@ -261,8 +258,7 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
 #define DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER 7
 
 /**
- * *******************************************************************************************
- * dcn32_determine_det_override: Determine DET allocation for each pipe
+ * dcn32_determine_det_override(): Determine DET allocation for each pipe
  *
  * This function determines how much DET to allocate for each pipe. The total number of
  * DET segments will be split equally among each of the streams, and after that the DET
@@ -290,13 +286,11 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
  * 3. Assign smaller DET size for lower pixel display and higher DET size for
  *    higher pixel display
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- * @param [in]: pipes: Array of DML pipes
- *
- * @return: void
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @pipes: Array of DML pipes
  *
- * *******************************************************************************************
+ * Return: void
  */
 void dcn32_determine_det_override(struct dc *dc,
 		struct dc_state *context,
@@ -432,8 +426,7 @@ void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context,
 }
 
 /**
- * *******************************************************************************************
- * dcn32_save_mall_state: Save MALL (SubVP) state for fast validation cases
+ * dcn32_save_mall_state(): Save MALL (SubVP) state for fast validation cases
  *
  * This function saves the MALL (SubVP) case for fast validation cases. For fast validation,
  * there are situations where a shallow copy of the dc->current_state is created for the
@@ -446,13 +439,11 @@ void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context,
  * NOTE: This function ONLY works if the streams are not moved to a different pipe in the
  *       validation. We don't expect this to happen in fast_validation=1 cases.
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- * @param [out]: temp_config: struct used to cache the existing MALL state
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @temp_config: struct used to cache the existing MALL state
  *
- * @return: void
- *
- * *******************************************************************************************
+ * Return: void
  */
 void dcn32_save_mall_state(struct dc *dc,
 		struct dc_state *context,
@@ -472,18 +463,15 @@ void dcn32_save_mall_state(struct dc *dc,
 }
 
 /**
- * *******************************************************************************************
- * dcn32_restore_mall_state: Restore MALL (SubVP) state for fast validation cases
+ * dcn32_restore_mall_state(): Restore MALL (SubVP) state for fast validation cases
  *
  * Restore the MALL state based on the previously saved state from dcn32_save_mall_state
  *
- * @param [in]: dc: Current DC state
- * @param [in/out]: context: New DC state to be programmed, restore MALL state into here
- * @param [in]: temp_config: struct that has the cached MALL state
- *
- * @return: void
+ * @dc: Current DC state
+ * @context: New DC state to be programmed, restore MALL state into here
+ * @temp_config: struct that has the cached MALL state
  *
- * *******************************************************************************************
+ * Return: void
  */
 void dcn32_restore_mall_state(struct dc *dc,
 		struct dc_state *context,
@@ -588,10 +576,11 @@ static int get_refresh_rate(struct dc_stream_state *fpo_candidate_stream)
 }
 
 /**
- * dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch - Determines if config can support FPO
+ * dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch() - Determines if config can
+ *								    support FPO
  *
- * @param [in]: dc - current dc state
- * @param [in]: context - new dc state
+ * @dc: current dc state
+ * @context: new dc state
  *
  * Return: Pointer to FPO stream candidate if config can support FPO, otherwise NULL
  */
-- 
GitLab


From 3808c34b291925d8a0fda2f23136381c1feb1dd2 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 10:41:59 +0530
Subject: [PATCH 1498/1791] drm/amd/display: Fix up missing 'dc' & 'pipe_ctx'
 kdoc parameters in delay_cursor_until_vupdate()

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:1904: warning: Function parameter or member 'dc' not described in 'delay_cursor_until_vupdate'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hw_sequencer.c:1904: warning: Function parameter or member 'pipe_ctx' not described in 'delay_cursor_until_vupdate'

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 13b4e5118459b..20a1582be0b13 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -1899,6 +1899,11 @@ void dcn10_pipe_control_lock(
  *
  * TODO: Optimize cursor programming to be once per frame before VUPDATE
  *       to avoid the need for this workaround.
+ *
+ * @dc: Current DC state
+ * @pipe_ctx: Pipe_ctx pointer for delayed cursor update
+ *
+ * Return: void
  */
 static void delay_cursor_until_vupdate(struct dc *dc, struct pipe_ctx *pipe_ctx)
 {
-- 
GitLab


From b5c07eaefc8b6e4aaa433f52ce74e619cd0ec386 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 09:22:36 +0530
Subject: [PATCH 1499/1791] drm/amd/display: Correct kdoc formatting for
 DCN32_CRB_SEGMENT_SIZE_KB in dcn32_hubbub.c

Fixes the following W=1 kernel build warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_hubbub.c:45: warning: Cannot understand  * @DCN32_CRB_SEGMENT_SIZE_KB: Maximum Configurable Return Buffer size for
 on line 45 - I thought it was a doc line

Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
index eb08ccc38e798..a18b9c0c5709c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c
@@ -42,8 +42,8 @@
 	hubbub2->shifts->field_name, hubbub2->masks->field_name
 
 /**
- * @DCN32_CRB_SEGMENT_SIZE_KB: Maximum Configurable Return Buffer size for
- * DCN32
+ * DCN32_CRB_SEGMENT_SIZE_KB: Maximum Configurable Return Buffer size for
+ *                            DCN32
  */
 #define DCN32_CRB_SEGMENT_SIZE_KB 64
 
-- 
GitLab


From 09521b5d49222d5ae932c4d738b2d55fb7abb415 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 30 May 2023 11:57:59 -0500
Subject: [PATCH 1500/1791] drm/amd: Disallow s0ix without BIOS support again
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

commit cf488dcd0ab7 ("drm/amd: Allow s0ix without BIOS support") showed
improvements to power consumption over suspend when s0ix wasn't enabled in
BIOS and the system didn't support S3.

This patch however was misguided because the reason the system didn't
support S3 was because SMT was disabled in OEM BIOS setup.
This prevented the BIOS from allowing S3.

Also allowing GPUs to use the s2idle path actually causes problems if
they're invoked on systems that may not support s2idle in the platform
firmware. `systemd` has a tendency to try to use `s2idle` if `deep` fails
for any reason, which could lead to unexpected flows.

The original commit also fixed a problem during resume from suspend to idle
without hardware support, but this is no longer necessary with commit
ca4751866397 ("drm/amd: Don't allow s0ix on APUs older than Raven")

Revert commit cf488dcd0ab7 ("drm/amd: Allow s0ix without BIOS support")
to make it match the expected behavior again.

Cc: Rafael Ávila de Espíndola <rafael@espindo.la>
Link: https://github.com/torvalds/linux/blob/v6.1/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c#L1060
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/2599
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index be881f21b340a..73e4434d9b546 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1499,16 +1499,20 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev)
 	 * S0ix even though the system is suspending to idle, so return false
 	 * in that case.
 	 */
-	if (!(acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0))
+	if (!(acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0)) {
 		dev_warn_once(adev->dev,
 			      "Power consumption will be higher as BIOS has not been configured for suspend-to-idle.\n"
 			      "To use suspend-to-idle change the sleep mode in BIOS setup.\n");
+		return false;
+	}
 
 #if !IS_ENABLED(CONFIG_AMD_PMC)
 	dev_warn_once(adev->dev,
 		      "Power consumption will be higher as the kernel has not been compiled with CONFIG_AMD_PMC.\n");
-#endif /* CONFIG_AMD_PMC */
+	return false;
+#else
 	return true;
+#endif /* CONFIG_AMD_PMC */
 }
 
 #endif /* CONFIG_SUSPEND */
-- 
GitLab


From 0df1106bfd13a9fe1eb7c33666bec091bc37c2a7 Mon Sep 17 00:00:00 2001
From: Tom Rix <trix@redhat.com>
Date: Thu, 30 Mar 2023 11:20:40 -0400
Subject: [PATCH 1501/1791] drm/amdkfd: remove unused sq_int_priv variable

clang with W=1 reports
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_int_process_v11.c:282:38: error: variable
  'sq_int_priv' set but not used [-Werror,-Wunused-but-set-variable]
        uint8_t sq_int_enc, sq_int_errtype, sq_int_priv;
                                            ^
This variable is not used so remove it.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index 0f0fdea4cd8ad..c2166bf964ef0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -279,7 +279,7 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 {
 	uint16_t source_id, client_id, ring_id, pasid, vmid;
 	uint32_t context_id0, context_id1;
-	uint8_t sq_int_enc, sq_int_errtype, sq_int_priv;
+	uint8_t sq_int_enc, sq_int_errtype;
 	struct kfd_vm_fault_info info = {0};
 	struct kfd_hsa_memory_exception_data exception_data;
 
@@ -348,13 +348,6 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_INST:
 				print_sq_intr_info_inst(context_id0, context_id1);
-				sq_int_priv = REG_GET_FIELD(context_id0,
-						SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV);
-				/*if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid,
-						KFD_CTXID0_DOORBELL_ID(context_id0),
-						KFD_CTXID0_TRAP_CODE(context_id0),
-						NULL, 0)))
-					return;*/
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				print_sq_intr_info_error(context_id0, context_id1);
-- 
GitLab


From 1b320ad3f5a88602aef2f207bc211539a5496702 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Wed, 24 May 2023 14:59:32 -0400
Subject: [PATCH 1502/1791] drm/amd/amdgpu: introduce DRM_AMDGPU_WERROR

We want to do -Werror builds on our CI. However, non-amdgpu breakages
have prevented us from doing so thus far. Also, there are a number of
additional checks that we should enable, that the community cares about
and are hidden behind -Wextra. So, define DRM_AMDGPU_WERROR to only
enable -Werror for the amdgpu kernel module and enable -Wextra while
disabling all of the checks that are too noisy.

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Kenny Ho <kenny.ho@amd.com>
Suggested-by: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Kenny Ho <Kenny.Ho@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Kconfig  | 10 ++++++++++
 drivers/gpu/drm/amd/amdgpu/Makefile |  9 +++++++++
 2 files changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig
index 12adca8c7819c..b91e79c721e2b 100644
--- a/drivers/gpu/drm/amd/amdgpu/Kconfig
+++ b/drivers/gpu/drm/amd/amdgpu/Kconfig
@@ -69,6 +69,16 @@ config DRM_AMDGPU_USERPTR
 	  This option selects CONFIG_HMM and CONFIG_HMM_MIRROR if it
 	  isn't already selected to enabled full userptr support.
 
+config DRM_AMDGPU_WERROR
+	bool "Force the compiler to throw an error instead of a warning when compiling"
+	depends on DRM_AMDGPU
+	depends on EXPERT
+	depends on !COMPILE_TEST
+	default n
+	help
+	  Add -Werror to the build flags for amdgpu.ko.
+	  Only enable this if you are warning code for amdgpu.ko.
+
 source "drivers/gpu/drm/amd/acp/Kconfig"
 source "drivers/gpu/drm/amd/display/Kconfig"
 source "drivers/gpu/drm/amd/amdkfd/Kconfig"
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 74a9aa6fe18c8..7ee68b1bbfedf 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -39,6 +39,15 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm \
 	-I$(FULL_AMD_PATH)/amdkfd
 
+subdir-ccflags-y := -Wextra
+subdir-ccflags-y += -Wunused-but-set-variable
+subdir-ccflags-y += -Wno-unused-parameter
+subdir-ccflags-y += -Wno-type-limits
+subdir-ccflags-y += -Wno-sign-compare
+subdir-ccflags-y += -Wno-missing-field-initializers
+subdir-ccflags-y += -Wno-override-init
+subdir-ccflags-$(CONFIG_DRM_AMDGPU_WERROR) += -Werror
+
 amdgpu-y := amdgpu_drv.o
 
 # add KMS driver
-- 
GitLab


From c60c9a5f9ab659e5ca9fa0e485a28e82fba761ce Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 00:34:10 +0530
Subject: [PATCH 1503/1791] drm/amd/display: Fix up kdoc formatting in
 display_mode_vba.c

Fixes the following W=1 kernel build warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/display_mode_vba.c:936: warning: Cannot understand  * *************************************************************************

Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
index 2f63ae954826c..9a3ded3111952 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c
@@ -933,18 +933,16 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib)
 }
 
 /**
- * ********************************************************************************************
  * cache_debug_params: Cache any params that needed to be maintained from the initial validation
  * for debug purposes.
  *
  * The DML getters can modify some of the VBA params that we are interested in (for example when
  * calculating with dummy p-state latency), so cache any params here that we want for debugging
  *
- * @param [in] mode_lib: mode_lib input/output of validate call
+ * @mode_lib: mode_lib input/output of validate call
  *
- * @return: void
+ * Return: void
  *
- * ********************************************************************************************
  */
 static void cache_debug_params(struct display_mode_lib *mode_lib)
 {
-- 
GitLab


From ebe884e8b93351382290ae107c880230d3a1f125 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 00:12:22 +0530
Subject: [PATCH 1504/1791] drm/amdgpu: Fix up kdoc 'ring' parameter in
 sdma_v6_0_ring_pad_ib
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix this warning by adding 'ring' arguments to kdoc.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c:1128: warning: Function parameter or member 'ring' not described in 'sdma_v6_0_ring_pad_ib'

Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index 967849c59ebe7..3b03dda854fdc 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -1121,6 +1121,7 @@ static void sdma_v6_0_vm_set_pte_pde(struct amdgpu_ib *ib,
 /**
  * sdma_v6_0_ring_pad_ib - pad the IB
  * @ib: indirect buffer to fill with padding
+ * @ring: amdgpu ring pointer
  *
  * Pad the IB with NOPs to a boundary multiple of 8.
  */
-- 
GitLab


From 2e9fee9b8e0e10fb9a4cba3ace607cebf7021bc1 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 25 May 2023 22:56:17 +0530
Subject: [PATCH 1505/1791] drm/amdgpu: Fix up kdoc in amdgpu_device.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix these warnings by deleting the deviant arguments.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:799: warning: Excess function parameter 'pcie_index' description in 'amdgpu_device_indirect_wreg'
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:799: warning: Excess function parameter 'pcie_data' description in 'amdgpu_device_indirect_wreg'
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:870: warning: Excess function parameter 'pcie_index' description in 'amdgpu_device_indirect_wreg64'
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c:870: warning: Excess function parameter 'pcie_data' description in 'amdgpu_device_indirect_wreg64'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 8f50ca2bee97b..e25f085ee8867 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -789,8 +789,6 @@ u64 amdgpu_device_indirect_rreg64(struct amdgpu_device *adev,
  * amdgpu_device_indirect_wreg - write an indirect register address
  *
  * @adev: amdgpu_device pointer
- * @pcie_index: mmio register offset
- * @pcie_data: mmio register offset
  * @reg_addr: indirect register offset
  * @reg_data: indirect register data
  *
@@ -860,8 +858,6 @@ void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev,
  * amdgpu_device_indirect_wreg64 - write a 64bits indirect register address
  *
  * @adev: amdgpu_device pointer
- * @pcie_index: mmio register offset
- * @pcie_data: mmio register offset
  * @reg_addr: indirect register offset
  * @reg_data: indirect register data
  *
-- 
GitLab


From 16cc3a221537bb3588ec2a568d7bd0e7972b25a8 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 23:33:35 +0530
Subject: [PATCH 1506/1791] drm/amdgpu: Add function parameter 'event' to kdoc
 in svm_range_evict()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_svm.c:1841: warning: Function parameter or member 'event' not described in 'svm_range_evict'

Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 9c88d6e90c3b5..615eab3f78c93 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1826,6 +1826,7 @@ static void svm_range_restore_work(struct work_struct *work)
  * @mm: current process mm_struct
  * @start: starting process queue number
  * @last: last process queue number
+ * @event: mmu notifier event when range is evicted or migrated
  *
  * Stop all queues of the process to ensure GPU doesn't access the memory, then
  * return to let CPU evict the buffer and proceed CPU pagetable update.
-- 
GitLab


From cbb63eccc05626d0d111b335e44f111a3bb92871 Mon Sep 17 00:00:00 2001
From: Horatio Zhang <Hongkun.Zhang@amd.com>
Date: Mon, 29 May 2023 14:23:37 -0400
Subject: [PATCH 1507/1791] drm/amdgpu: fix Null pointer dereference error in
 amdgpu_device_recover_vram

Use the function of amdgpu_bo_vm_destroy to handle the resource release
of shadow bo. During the amdgpu_mes_self_test, shadow bo released, but
vmbo->shadow_list was not, which caused a null pointer reference error
in amdgpu_device_recover_vram when GPU reset.

Fixes: 6c032c37ac3e ("drm/amdgpu: Fix vram recover doesn't work after whole GPU reset (v2)")
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Acked-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 10 ++++------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c  |  1 -
 2 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index d9e3315083899..f76649e523a0d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -79,9 +79,10 @@ static void amdgpu_bo_user_destroy(struct ttm_buffer_object *tbo)
 static void amdgpu_bo_vm_destroy(struct ttm_buffer_object *tbo)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev);
-	struct amdgpu_bo *bo = ttm_to_amdgpu_bo(tbo);
+	struct amdgpu_bo *shadow_bo = ttm_to_amdgpu_bo(tbo), *bo;
 	struct amdgpu_bo_vm *vmbo;
 
+	bo = shadow_bo->parent;
 	vmbo = to_amdgpu_bo_vm(bo);
 	/* in case amdgpu_device_recover_vram got NULL of bo->parent */
 	if (!list_empty(&vmbo->shadow_list)) {
@@ -711,11 +712,6 @@ int amdgpu_bo_create_vm(struct amdgpu_device *adev,
 		return r;
 
 	*vmbo_ptr = to_amdgpu_bo_vm(bo_ptr);
-	INIT_LIST_HEAD(&(*vmbo_ptr)->shadow_list);
-	/* Set destroy callback to amdgpu_bo_vm_destroy after vmbo->shadow_list
-	 * is initialized.
-	 */
-	bo_ptr->tbo.destroy = &amdgpu_bo_vm_destroy;
 	return r;
 }
 
@@ -732,6 +728,8 @@ void amdgpu_bo_add_to_shadow_list(struct amdgpu_bo_vm *vmbo)
 
 	mutex_lock(&adev->shadow_list_lock);
 	list_add_tail(&vmbo->shadow_list, &adev->shadow_list);
+	vmbo->shadow->parent = amdgpu_bo_ref(&vmbo->bo);
+	vmbo->shadow->tbo.destroy = &amdgpu_bo_vm_destroy;
 	mutex_unlock(&adev->shadow_list_lock);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index cc3b1b596e561..dea1a64be44d0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -573,7 +573,6 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		return r;
 	}
 
-	(*vmbo)->shadow->parent = amdgpu_bo_ref(bo);
 	amdgpu_bo_add_to_shadow_list(*vmbo);
 
 	return 0;
-- 
GitLab


From 023f4d60747cadd96115c3c3b55986798322f3f6 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 3 Apr 2023 11:38:17 +0530
Subject: [PATCH 1508/1791] drm/amd/pm: Update SMUv13.0.6 PMFW headers

Update PMFW interface headers to for new metrics table format and
throttling information.

v2: Added dummy definition for compilation error

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../inc/pmfw_if/smu13_driver_if_v13_0_6.h     | 31 ++++++++++++++-----
 .../pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h   | 13 +++++---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |  2 ++
 3 files changed, 35 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
index de84fff39799a..ca4a5e99ccd1f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h
@@ -26,7 +26,7 @@
 // *** IMPORTANT ***
 // PMFW TEAM: Always increment the interface version if
 // anything is changed in this file
-#define SMU13_0_6_DRIVER_IF_VERSION 0x08042023
+#define SMU13_0_6_DRIVER_IF_VERSION 0x08042024
 
 //I2C Interface
 #define NUM_I2C_CONTROLLERS                8
@@ -125,11 +125,28 @@ typedef struct {
 #define IH_INTERRUPT_ID_TO_DRIVER                   0xFE
 #define IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING  0x7
 
-//thermal over-temp mask defines
-#define THROTTLER_TEMP_CCD_BIT     5
-#define THROTTLER_TEMP_XCD_BIT     6
-#define THROTTLER_TEMP_HBM_BIT     7
-#define THROTTLER_TEMP_AID_BIT     8
-#define THROTTLER_VRHOT_BIT        9
+//thermal over-temp mask defines for IH interrupt to host
+#define THROTTLER_PROCHOT_BIT           0
+#define THROTTLER_PPT_BIT               1
+#define THROTTLER_THERMAL_SOCKET_BIT    2//AID, XCD, CCD throttling
+#define THROTTLER_THERMAL_VR_BIT        3//VRHOT
+#define THROTTLER_THERMAL_HBM_BIT       4
+
+// These defines are used with the following messages:
+// SMC_MSG_TransferTableDram2Smu
+// SMC_MSG_TransferTableSmu2Dram
+// #define TABLE_PPTABLE                 0
+// #define TABLE_AVFS_PSM_DEBUG          1
+// #define TABLE_AVFS_FUSE_OVERRIDE      2
+// #define TABLE_PMSTATUSLOG             3
+// #define TABLE_SMU_METRICS             4
+// #define TABLE_DRIVER_SMU_CONFIG       5
+// #define TABLE_I2C_COMMANDS            6
+// #define TABLE_COUNT                   7
+
+// // Table transfer status
+// #define TABLE_TRANSFER_OK         0x0
+// #define TABLE_TRANSFER_FAILED     0xFF
+// #define TABLE_TRANSFER_PENDING    0xAB
 
 #endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
index 3fe403615d86b..252aef190c5c2 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h
@@ -123,9 +123,9 @@ typedef enum {
   VOLTAGE_GUARDBAND_COUNT
 } GFX_GUARDBAND_e;
 
-#define SMU_METRICS_TABLE_VERSION 0x3
+#define SMU_METRICS_TABLE_VERSION 0x5
 
-typedef struct {
+typedef struct __attribute__((packed, aligned(4))) {
   uint32_t AccumulationCounter;
 
   //TEMPERATURE
@@ -202,11 +202,16 @@ typedef struct {
 
   // New Items at end to maintain driver compatibility
   uint32_t GfxclkFrequency[8];
+
+  //PSNs
+  uint64_t PublicSerialNumber_AID[4];
+  uint64_t PublicSerialNumber_XCD[8];
+  uint64_t PublicSerialNumber_CCD[12];
 } MetricsTable_t;
 
-#define SMU_VF_METRICS_TABLE_VERSION 0x1
+#define SMU_VF_METRICS_TABLE_VERSION 0x3
 
-typedef struct {
+typedef struct __attribute__((packed, aligned(4))) {
   uint32_t AccumulationCounter;
   uint32_t InstGfxclk_TargFreq;
   uint64_t AccGfxclk_TargFreq;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 41b49cc827cd5..27fd71afc73f7 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -82,6 +82,8 @@
 
 #define smnPCIE_ESM_CTRL 0x111003D0
 
+#define THROTTLER_TEMP_HBM_BIT 2
+
 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
 	MSG_MAP(GetSmuVersion,			     PPSMC_MSG_GetSmuVersion,			1),
-- 
GitLab


From 93682f8a196718c2caf9b9b3de7894d5c0318f1f Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 31 Mar 2023 16:34:15 +0530
Subject: [PATCH 1509/1791] drm/amd/pm: Fix SMUv13.0.6 throttle status report

Add throttle status in power context
Keep throttle status indicator in SMUv13 power context

v2: Removed Dummy definition

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h  |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 95 +++++++++----------
 2 files changed, 46 insertions(+), 50 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
index 3ae8d5d252a3c..5a99a091965e3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
@@ -119,6 +119,7 @@ struct smu_13_0_power_context {
 	uint32_t	power_source;
 	uint8_t		in_power_limit_boost_mode;
 	enum smu_13_0_power_state power_state;
+	atomic_t	throttle_status;
 };
 
 #if defined(SWSMU_CODE_LAYER_L2) || defined(SWSMU_CODE_LAYER_L3)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 27fd71afc73f7..b9f32e0364db1 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -82,8 +82,6 @@
 
 #define smnPCIE_ESM_CTRL 0x111003D0
 
-#define THROTTLER_TEMP_HBM_BIT 2
-
 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
 	MSG_MAP(GetSmuVersion,			     PPSMC_MSG_GetSmuVersion,			1),
@@ -174,17 +172,12 @@ static const struct cmn2asic_mapping smu_v13_0_6_table_map[SMU_TABLE_COUNT] = {
 	TAB_MAP(I2C_COMMANDS),
 };
 
-#define THROTTLER_PROCHOT_GFX_BIT  0
-#define THROTTLER_PPT_BIT 1
-#define THROTTLER_TEMP_SOC_BIT 2
-#define THROTTLER_TEMP_VR_GFX_BIT 3
-
 static const uint8_t smu_v13_0_6_throttler_map[] = {
 	[THROTTLER_PPT_BIT]		= (SMU_THROTTLER_PPT0_BIT),
-	[THROTTLER_TEMP_SOC_BIT]	= (SMU_THROTTLER_TEMP_GPU_BIT),
-	[THROTTLER_TEMP_HBM_BIT]	= (SMU_THROTTLER_TEMP_MEM_BIT),
-	[THROTTLER_TEMP_VR_GFX_BIT]	= (SMU_THROTTLER_TEMP_VR_GFX_BIT),
-	[THROTTLER_PROCHOT_GFX_BIT]	= (SMU_THROTTLER_PROCHOT_GFX_BIT),
+	[THROTTLER_THERMAL_SOCKET_BIT]	= (SMU_THROTTLER_TEMP_GPU_BIT),
+	[THROTTLER_THERMAL_HBM_BIT]	= (SMU_THROTTLER_TEMP_MEM_BIT),
+	[THROTTLER_THERMAL_VR_BIT]	= (SMU_THROTTLER_TEMP_VR_GFX_BIT),
+	[THROTTLER_PROCHOT_BIT]		= (SMU_THROTTLER_PROCHOT_GFX_BIT),
 };
 
 struct PPTable_t {
@@ -642,16 +635,14 @@ static int smu_v13_0_6_freqs_in_same_level(int32_t frequency1,
 	return (abs(frequency1 - frequency2) <= EPSILON);
 }
 
-static uint32_t smu_v13_0_6_get_throttler_status(struct smu_context *smu,
-						 MetricsTable_t *metrics)
+static uint32_t smu_v13_0_6_get_throttler_status(struct smu_context *smu)
 {
+	struct smu_power_context *smu_power = &smu->smu_power;
+	struct smu_13_0_power_context *power_context = smu_power->power_context;
 	uint32_t  throttler_status = 0;
 
-	throttler_status |= metrics->ProchotResidencyAcc > 0 ? 1U << THROTTLER_PROCHOT_GFX_BIT : 0;
-	throttler_status |= metrics->PptResidencyAcc > 0 ? 1U << THROTTLER_PPT_BIT : 0;
-	throttler_status |= metrics->SocketThmResidencyAcc > 0 ?  1U << THROTTLER_TEMP_SOC_BIT : 0;
-	throttler_status |= metrics->VrThmResidencyAcc > 0 ? 1U << THROTTLER_TEMP_VR_GFX_BIT : 0;
-	throttler_status |= metrics->HbmThmResidencyAcc > 0 ? 1U << THROTTLER_TEMP_HBM_BIT : 0;
+	throttler_status = atomic_read(&power_context->throttle_status);
+	dev_dbg(smu->adev->dev, "SMU Throttler status: %u", throttler_status);
 
 	return throttler_status;
 }
@@ -721,9 +712,6 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu,
 	case METRICS_TEMPERATURE_VRSOC:
 		*value = SMUQ10_TO_UINT(metrics->MaxVrTemperature);
 		break;
-	case METRICS_THROTTLER_STATUS:
-		*value = smu_v13_0_6_get_throttler_status(smu, metrics);
-		break;
 	default:
 		*value = UINT_MAX;
 		break;
@@ -1290,13 +1278,11 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
 				   struct amdgpu_iv_entry *entry)
 {
 	struct smu_context *smu = adev->powerplay.pp_handle;
+	struct smu_power_context *smu_power = &smu->smu_power;
+	struct smu_13_0_power_context *power_context = smu_power->power_context;
 	uint32_t client_id = entry->client_id;
-	uint32_t src_id = entry->src_id;
-	/*
-	 * ctxid is used to distinguish different
-	 * events for SMCToHost interrupt
-	 */
 	uint32_t ctxid = entry->src_data[0];
+	uint32_t src_id = entry->src_id;
 	uint32_t data;
 
 	if (client_id == SOC15_IH_CLIENTID_MP1) {
@@ -1305,7 +1291,10 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
 			data = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL);
 			data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1);
 			WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, data);
-
+			/*
+			 * ctxid is used to distinguish different events for SMCToHost
+			 * interrupt.
+			 */
 			switch (ctxid) {
 			case IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING:
 				/*
@@ -1316,8 +1305,17 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev,
 				if (!atomic_read(&adev->throttling_logging_enabled))
 					return 0;
 
-				if (__ratelimit(&adev->throttling_logging_rs))
+				/* This uses the new method which fixes the
+				 * incorrect throttling status reporting
+				 * through metrics table. For older FWs,
+				 * it will be ignored.
+				 */
+				if (__ratelimit(&adev->throttling_logging_rs)) {
+					atomic_set(
+						&power_context->throttle_status,
+							entry->src_data[1]);
 					schedule_work(&smu->throttling_logging_work);
+				}
 
 				break;
 			}
@@ -1895,37 +1893,35 @@ static int smu_v13_0_6_allow_xgmi_power_down(struct smu_context *smu, bool en)
 					       en ? 0 : 1, NULL);
 }
 
-static const struct throttling_logging_label {
-	uint32_t feature_mask;
-	const char *label;
-} logging_label[] = {
-	{ (1U << THROTTLER_TEMP_HBM_BIT), "HBM" },
-	{ (1U << THROTTLER_TEMP_SOC_BIT), "SOC" },
-	{ (1U << THROTTLER_TEMP_VR_GFX_BIT), "VR limit" },
+static const char *const throttling_logging_label[] = {
+	[THROTTLER_PROCHOT_BIT] = "Prochot",
+	[THROTTLER_PPT_BIT] = "PPT",
+	[THROTTLER_THERMAL_SOCKET_BIT] = "SOC",
+	[THROTTLER_THERMAL_VR_BIT] = "VR",
+	[THROTTLER_THERMAL_HBM_BIT] = "HBM"
 };
+
 static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu)
 {
-	int ret;
 	int throttler_idx, throtting_events = 0, buf_idx = 0;
 	struct amdgpu_device *adev = smu->adev;
 	uint32_t throttler_status;
 	char log_buf[256];
 
-	ret = smu_v13_0_6_get_smu_metrics_data(smu, METRICS_THROTTLER_STATUS,
-					      &throttler_status);
-	if (ret)
+	throttler_status = smu_v13_0_6_get_throttler_status(smu);
+	if (!throttler_status)
 		return;
 
 	memset(log_buf, 0, sizeof(log_buf));
-	for (throttler_idx = 0; throttler_idx < ARRAY_SIZE(logging_label);
+	for (throttler_idx = 0;
+	     throttler_idx < ARRAY_SIZE(throttling_logging_label);
 	     throttler_idx++) {
-		if (throttler_status &
-		    logging_label[throttler_idx].feature_mask) {
+		if (throttler_status & (1U << throttler_idx)) {
 			throtting_events++;
-			buf_idx += snprintf(log_buf + buf_idx,
-					    sizeof(log_buf) - buf_idx, "%s%s",
-					    throtting_events > 1 ? " and " : "",
-					    logging_label[throttler_idx].label);
+			buf_idx += snprintf(
+				log_buf + buf_idx, sizeof(log_buf) - buf_idx,
+				"%s%s", throtting_events > 1 ? " and " : "",
+				throttling_logging_label[throttler_idx]);
 			if (buf_idx >= sizeof(log_buf)) {
 				dev_err(adev->dev, "buffer overflow!\n");
 				log_buf[sizeof(log_buf) - 1] = '\0';
@@ -1934,10 +1930,9 @@ static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu)
 		}
 	}
 
-	dev_warn(
-		adev->dev,
-		"WARN: GPU thermal throttling temperature reached, expect performance decrease. %s.\n",
-		log_buf);
+	dev_warn(adev->dev,
+		 "WARN: GPU is throttled, expect performance decrease. %s.\n",
+		 log_buf);
 	kgd2kfd_smi_event_throttle(
 		smu->adev->kfd.dev,
 		smu_cmn_get_indep_throttler_status(throttler_status,
-- 
GitLab


From 5e86aa29a338f5c25e2d10d021bffc6b1b560ad5 Mon Sep 17 00:00:00 2001
From: Yang Wang <KevinYang.Wang@amd.com>
Date: Wed, 24 May 2023 13:54:26 +0800
Subject: [PATCH 1510/1791] drm/amd/pm: add unique serial number support for
 smu_v13_0_6

add unique serial number support for smu_v13_0_6.
(use aid0 serial number by default)

Signed-off-by: Yang Wang <KevinYang.Wang@amd.com>
Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 20 ++++++++-----------
 1 file changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index b9f32e0364db1..75255e0baf919 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -192,6 +192,7 @@ struct PPTable_t {
 	uint32_t LclkFrequencyTable[4];
 	uint32_t MaxLclkDpmRange;
 	uint32_t MinLclkDpmRange;
+	uint64_t PublicSerialNumber_AID;
 	bool Init;
 };
 
@@ -352,6 +353,9 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 				SMUQ10_TO_UINT(metrics->LclkFrequencyTable[i]);
 		}
 
+		/* use AID0 serial number by default */
+		pptable->PublicSerialNumber_AID = metrics->PublicSerialNumber_AID[0];
+
 		pptable->Init = true;
 	}
 
@@ -1856,19 +1860,11 @@ static void smu_v13_0_6_i2c_control_fini(struct smu_context *smu)
 static void smu_v13_0_6_get_unique_id(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	//SmuMetrics_t *metrics = smu->smu_table.metrics_table;
-	uint32_t upper32 = 0, lower32 = 0;
-	int ret;
-
-	ret = smu_cmn_get_metrics_table(smu, NULL, false);
-	if (ret)
-		goto out;
-
-	//upper32 = metrics->PublicSerialNumUpper32;
-	//lower32 = metrics->PublicSerialNumLower32;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct PPTable_t *pptable =
+		(struct PPTable_t *)smu_table->driver_pptable;
 
-out:
-	adev->unique_id = ((uint64_t)upper32 << 32) | lower32;
+	adev->unique_id = pptable->PublicSerialNumber_AID;
 	if (adev->serial[0] == '\0')
 		sprintf(adev->serial, "%016llx", adev->unique_id);
 }
-- 
GitLab


From 3c87de6d034fcb756a10523367219c5564a85fd3 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 31 Mar 2023 16:30:01 +0530
Subject: [PATCH 1511/1791] drm/amd/pm: Fix power context allocation in SMU13

Use the right data structure for allocation.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index b24c8549d0b11..70db36d459743 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -538,11 +538,11 @@ int smu_v13_0_init_power(struct smu_context *smu)
 	if (smu_power->power_context || smu_power->power_context_size != 0)
 		return -EINVAL;
 
-	smu_power->power_context = kzalloc(sizeof(struct smu_13_0_dpm_context),
+	smu_power->power_context = kzalloc(sizeof(struct smu_13_0_power_context),
 					   GFP_KERNEL);
 	if (!smu_power->power_context)
 		return -ENOMEM;
-	smu_power->power_context_size = sizeof(struct smu_13_0_dpm_context);
+	smu_power->power_context_size = sizeof(struct smu_13_0_power_context);
 
 	return 0;
 }
-- 
GitLab


From 6ff5a1cff70441e1cd27614c359a66d29649e872 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Thu, 6 Apr 2023 12:08:21 +0800
Subject: [PATCH 1512/1791] drm/amd/pm: conditionally disable pcie lane
 switching for some sienna_cichlid SKUs

Disable the pcie lane switching for some sienna_cichlid SKUs since it
might not work well on some platforms.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/pm/swsmu/smu11/sienna_cichlid_ppt.c   | 92 +++++++++++++++----
 1 file changed, 74 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index 75f18681e984c..85d53597eb07a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -2067,33 +2067,94 @@ static int sienna_cichlid_display_disable_memory_clock_switch(struct smu_context
 	return ret;
 }
 
+static void sienna_cichlid_get_override_pcie_settings(struct smu_context *smu,
+						      uint32_t *gen_speed_override,
+						      uint32_t *lane_width_override)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	*gen_speed_override = 0xff;
+	*lane_width_override = 0xff;
+
+	switch (adev->pdev->device) {
+	case 0x73A0:
+	case 0x73A1:
+	case 0x73A2:
+	case 0x73A3:
+	case 0x73AB:
+	case 0x73AE:
+		/* Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32 */
+		*lane_width_override = 6;
+		break;
+	case 0x73E0:
+	case 0x73E1:
+	case 0x73E3:
+		*lane_width_override = 4;
+		break;
+	case 0x7420:
+	case 0x7421:
+	case 0x7422:
+	case 0x7423:
+	case 0x7424:
+		*lane_width_override = 3;
+		break;
+	default:
+		break;
+	}
+}
+
+#define MAX(a, b)	((a) > (b) ? (a) : (b))
+
 static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu,
 					 uint32_t pcie_gen_cap,
 					 uint32_t pcie_width_cap)
 {
 	struct smu_11_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context;
-
-	uint32_t smu_pcie_arg;
+	struct smu_11_0_pcie_table *pcie_table = &dpm_context->dpm_tables.pcie_table;
+	uint32_t gen_speed_override, lane_width_override;
 	uint8_t *table_member1, *table_member2;
+	uint32_t min_gen_speed, max_gen_speed;
+	uint32_t min_lane_width, max_lane_width;
+	uint32_t smu_pcie_arg;
 	int ret, i;
 
 	GET_PPTABLE_MEMBER(PcieGenSpeed, &table_member1);
 	GET_PPTABLE_MEMBER(PcieLaneCount, &table_member2);
 
-	/* lclk dpm table setup */
-	for (i = 0; i < MAX_PCIE_CONF; i++) {
-		dpm_context->dpm_tables.pcie_table.pcie_gen[i] = table_member1[i];
-		dpm_context->dpm_tables.pcie_table.pcie_lane[i] = table_member2[i];
+	sienna_cichlid_get_override_pcie_settings(smu,
+						  &gen_speed_override,
+						  &lane_width_override);
+
+	/* PCIE gen speed override */
+	if (gen_speed_override != 0xff) {
+		min_gen_speed = MIN(pcie_gen_cap, gen_speed_override);
+		max_gen_speed = MIN(pcie_gen_cap, gen_speed_override);
+	} else {
+		min_gen_speed = MAX(0, table_member1[0]);
+		max_gen_speed = MIN(pcie_gen_cap, table_member1[1]);
+		min_gen_speed = min_gen_speed > max_gen_speed ?
+				max_gen_speed : min_gen_speed;
 	}
+	pcie_table->pcie_gen[0] = min_gen_speed;
+	pcie_table->pcie_gen[1] = max_gen_speed;
+
+	/* PCIE lane width override */
+	if (lane_width_override != 0xff) {
+		min_lane_width = MIN(pcie_width_cap, lane_width_override);
+		max_lane_width = MIN(pcie_width_cap, lane_width_override);
+	} else {
+		min_lane_width = MAX(1, table_member2[0]);
+		max_lane_width = MIN(pcie_width_cap, table_member2[1]);
+		min_lane_width = min_lane_width > max_lane_width ?
+				 max_lane_width : min_lane_width;
+	}
+	pcie_table->pcie_lane[0] = min_lane_width;
+	pcie_table->pcie_lane[1] = max_lane_width;
 
 	for (i = 0; i < NUM_LINK_LEVELS; i++) {
-		smu_pcie_arg = (i << 16) |
-			((table_member1[i] <= pcie_gen_cap) ?
-			 (table_member1[i] << 8) :
-			 (pcie_gen_cap << 8)) |
-			((table_member2[i] <= pcie_width_cap) ?
-			 table_member2[i] :
-			 pcie_width_cap);
+		smu_pcie_arg = (i << 16 |
+				pcie_table->pcie_gen[i] << 8 |
+				pcie_table->pcie_lane[i]);
 
 		ret = smu_cmn_send_smc_msg_with_param(smu,
 				SMU_MSG_OverridePcieParameters,
@@ -2101,11 +2162,6 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu,
 				NULL);
 		if (ret)
 			return ret;
-
-		if (table_member1[i] > pcie_gen_cap)
-			dpm_context->dpm_tables.pcie_table.pcie_gen[i] = pcie_gen_cap;
-		if (table_member2[i] > pcie_width_cap)
-			dpm_context->dpm_tables.pcie_table.pcie_lane[i] = pcie_width_cap;
 	}
 
 	return 0;
-- 
GitLab


From ba3c87fffb79311f54464288c66421d19c2c1234 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Jun 2023 12:58:05 -0400
Subject: [PATCH 1513/1791] amd/amdkfd: drop unused KFD_IOCTL_SVM_FLAG_UNCACHED
 flag

Was leftover from GC 9.4.3 bring up and is currently
unused.  Drop it for now.

Cc: Philip.Yang@amd.com
Cc: rajneesh.bhardwaj@amd.com
Cc: Felix.Kuehling@amd.com
Reviewed-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 +-
 include/uapi/linux/kfd_ioctl.h       | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 615eab3f78c93..5ff1a5a89d968 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1154,7 +1154,7 @@ svm_range_get_pte_flags(struct kfd_node *node,
 	uint64_t pte_flags;
 	bool snoop = (domain != SVM_RANGE_VRAM_DOMAIN);
 	bool coherent = flags & KFD_IOCTL_SVM_FLAG_COHERENT;
-	bool uncached = flags & KFD_IOCTL_SVM_FLAG_UNCACHED;
+	bool uncached = false; /*flags & KFD_IOCTL_SVM_FLAG_UNCACHED;*/
 	unsigned int mtype_local;
 
 	if (domain == SVM_RANGE_VRAM_DOMAIN)
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 2a9671e1ddb5a..2da5c3ad71bd0 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -623,8 +623,6 @@ enum kfd_mmio_remap {
 #define KFD_IOCTL_SVM_FLAG_GPU_READ_MOSTLY     0x00000020
 /* Keep GPU memory mapping always valid as if XNACK is disable */
 #define KFD_IOCTL_SVM_FLAG_GPU_ALWAYS_MAPPED   0x00000040
-/* Uncached access to memory */
-#define KFD_IOCTL_SVM_FLAG_UNCACHED 0x00000080
 
 /**
  * kfd_ioctl_svm_op - SVM ioctl operations
-- 
GitLab


From 4f98cf2baf9faee5b6f2f7889dad7c0f7686a787 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 2 Mar 2022 14:30:12 -0500
Subject: [PATCH 1514/1791] drm/amdkfd: add debug and runtime enable interface

Introduce the GPU debug operations interface.

For ROCm-GDB to extend the GNU Debugger's ability to inspect the AMD GPU
instruction set, provide the necessary interface to allow the debugger
to HW debug-mode set and query exceptions per HSA queue, process or
device.

The runtime_enable interface coordinates exception handling with the
HSA runtime.

Usage is available in the kern docs at uapi/linux/kfd_ioctl.h.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  48 ++
 include/uapi/linux/kfd_ioctl.h           | 668 ++++++++++++++++++++++-
 2 files changed, 715 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index a9efff94390bf..00e34125987c1 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2729,6 +2729,48 @@ static int kfd_ioctl_criu(struct file *filep, struct kfd_process *p, void *data)
 	return ret;
 }
 
+static int kfd_ioctl_runtime_enable(struct file *filep, struct kfd_process *p, void *data)
+{
+	return 0;
+}
+
+static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, void *data)
+{
+	struct kfd_ioctl_dbg_trap_args *args = data;
+	int r = 0;
+
+	if (sched_policy == KFD_SCHED_POLICY_NO_HWS) {
+		pr_err("Debugging does not support sched_policy %i", sched_policy);
+		return -EINVAL;
+	}
+
+	switch (args->op) {
+	case KFD_IOC_DBG_TRAP_ENABLE:
+	case KFD_IOC_DBG_TRAP_DISABLE:
+	case KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT:
+	case KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED:
+	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE:
+	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE:
+	case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES:
+	case KFD_IOC_DBG_TRAP_RESUME_QUEUES:
+	case KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH:
+	case KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH:
+	case KFD_IOC_DBG_TRAP_SET_FLAGS:
+	case KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT:
+	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
+	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
+	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
+		pr_warn("Debugging not supported yet\n");
+		r = -EACCES;
+		break;
+	default:
+		pr_err("Invalid option: %i\n", args->op);
+		r = -EINVAL;
+	}
+
+	return r;
+}
+
 #define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \
 	[_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, \
 			    .cmd_drv = 0, .name = #ioctl}
@@ -2841,6 +2883,12 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = {
 
 	AMDKFD_IOCTL_DEF(AMDKFD_IOC_EXPORT_DMABUF,
 				kfd_ioctl_export_dmabuf, 0),
+
+	AMDKFD_IOCTL_DEF(AMDKFD_IOC_RUNTIME_ENABLE,
+			kfd_ioctl_runtime_enable, 0),
+
+	AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_TRAP,
+			kfd_ioctl_set_debug_trap, 0),
 };
 
 #define AMDKFD_CORE_IOCTL_COUNT	ARRAY_SIZE(amdkfd_ioctls)
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 2da5c3ad71bd0..32913d674d38b 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -110,6 +110,32 @@ struct kfd_ioctl_get_available_memory_args {
 	__u32 pad;
 };
 
+struct kfd_dbg_device_info_entry {
+	__u64 exception_status;
+	__u64 lds_base;
+	__u64 lds_limit;
+	__u64 scratch_base;
+	__u64 scratch_limit;
+	__u64 gpuvm_base;
+	__u64 gpuvm_limit;
+	__u32 gpu_id;
+	__u32 location_id;
+	__u32 vendor_id;
+	__u32 device_id;
+	__u32 revision_id;
+	__u32 subsystem_vendor_id;
+	__u32 subsystem_device_id;
+	__u32 fw_version;
+	__u32 gfx_target_version;
+	__u32 simd_count;
+	__u32 max_waves_per_simd;
+	__u32 array_count;
+	__u32 simd_arrays_per_engine;
+	__u32 num_xcc;
+	__u32 capability;
+	__u32 debug_prop;
+};
+
 /* For kfd_ioctl_set_memory_policy_args.default_policy and alternate_policy */
 #define KFD_IOC_CACHE_POLICY_COHERENT 0
 #define KFD_IOC_CACHE_POLICY_NONCOHERENT 1
@@ -773,6 +799,640 @@ struct kfd_ioctl_set_xnack_mode_args {
 	__s32 xnack_enabled;
 };
 
+/* Wave launch override modes */
+enum kfd_dbg_trap_override_mode {
+	KFD_DBG_TRAP_OVERRIDE_OR = 0,
+	KFD_DBG_TRAP_OVERRIDE_REPLACE = 1
+};
+
+/* Wave launch overrides */
+enum kfd_dbg_trap_mask {
+	KFD_DBG_TRAP_MASK_FP_INVALID = 1,
+	KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL = 2,
+	KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO = 4,
+	KFD_DBG_TRAP_MASK_FP_OVERFLOW = 8,
+	KFD_DBG_TRAP_MASK_FP_UNDERFLOW = 16,
+	KFD_DBG_TRAP_MASK_FP_INEXACT = 32,
+	KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO = 64,
+	KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH = 128,
+	KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION = 256,
+	KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START = (1 << 30),
+	KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END = (1 << 31)
+};
+
+/* Wave launch modes */
+enum kfd_dbg_trap_wave_launch_mode {
+	KFD_DBG_TRAP_WAVE_LAUNCH_MODE_NORMAL = 0,
+	KFD_DBG_TRAP_WAVE_LAUNCH_MODE_HALT = 1,
+	KFD_DBG_TRAP_WAVE_LAUNCH_MODE_DEBUG = 3
+};
+
+/* Address watch modes */
+enum kfd_dbg_trap_address_watch_mode {
+	KFD_DBG_TRAP_ADDRESS_WATCH_MODE_READ = 0,
+	KFD_DBG_TRAP_ADDRESS_WATCH_MODE_NONREAD = 1,
+	KFD_DBG_TRAP_ADDRESS_WATCH_MODE_ATOMIC = 2,
+	KFD_DBG_TRAP_ADDRESS_WATCH_MODE_ALL = 3
+};
+
+/* Additional wave settings */
+enum kfd_dbg_trap_flags {
+	KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP = 1,
+};
+
+/* Trap exceptions */
+enum kfd_dbg_trap_exception_code {
+	EC_NONE = 0,
+	/* per queue */
+	EC_QUEUE_WAVE_ABORT = 1,
+	EC_QUEUE_WAVE_TRAP = 2,
+	EC_QUEUE_WAVE_MATH_ERROR = 3,
+	EC_QUEUE_WAVE_ILLEGAL_INSTRUCTION = 4,
+	EC_QUEUE_WAVE_MEMORY_VIOLATION = 5,
+	EC_QUEUE_WAVE_APERTURE_VIOLATION = 6,
+	EC_QUEUE_PACKET_DISPATCH_DIM_INVALID = 16,
+	EC_QUEUE_PACKET_DISPATCH_GROUP_SEGMENT_SIZE_INVALID = 17,
+	EC_QUEUE_PACKET_DISPATCH_CODE_INVALID = 18,
+	EC_QUEUE_PACKET_RESERVED = 19,
+	EC_QUEUE_PACKET_UNSUPPORTED = 20,
+	EC_QUEUE_PACKET_DISPATCH_WORK_GROUP_SIZE_INVALID = 21,
+	EC_QUEUE_PACKET_DISPATCH_REGISTER_INVALID = 22,
+	EC_QUEUE_PACKET_VENDOR_UNSUPPORTED = 23,
+	EC_QUEUE_PREEMPTION_ERROR = 30,
+	EC_QUEUE_NEW = 31,
+	/* per device */
+	EC_DEVICE_QUEUE_DELETE = 32,
+	EC_DEVICE_MEMORY_VIOLATION = 33,
+	EC_DEVICE_RAS_ERROR = 34,
+	EC_DEVICE_FATAL_HALT = 35,
+	EC_DEVICE_NEW = 36,
+	/* per process */
+	EC_PROCESS_RUNTIME = 48,
+	EC_PROCESS_DEVICE_REMOVE = 49,
+	EC_MAX
+};
+
+/* Mask generated by ecode in kfd_dbg_trap_exception_code */
+#define KFD_EC_MASK(ecode)	(1ULL << (ecode - 1))
+
+/* Masks for exception code type checks below */
+#define KFD_EC_MASK_QUEUE	(KFD_EC_MASK(EC_QUEUE_WAVE_ABORT) |	\
+				 KFD_EC_MASK(EC_QUEUE_WAVE_TRAP) |	\
+				 KFD_EC_MASK(EC_QUEUE_WAVE_MATH_ERROR) |	\
+				 KFD_EC_MASK(EC_QUEUE_WAVE_ILLEGAL_INSTRUCTION) |	\
+				 KFD_EC_MASK(EC_QUEUE_WAVE_MEMORY_VIOLATION) |	\
+				 KFD_EC_MASK(EC_QUEUE_WAVE_APERTURE_VIOLATION) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_DIM_INVALID) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_GROUP_SEGMENT_SIZE_INVALID) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_CODE_INVALID) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_RESERVED) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_UNSUPPORTED) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_WORK_GROUP_SIZE_INVALID) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_DISPATCH_REGISTER_INVALID) |	\
+				 KFD_EC_MASK(EC_QUEUE_PACKET_VENDOR_UNSUPPORTED)	|	\
+				 KFD_EC_MASK(EC_QUEUE_PREEMPTION_ERROR)	|	\
+				 KFD_EC_MASK(EC_QUEUE_NEW))
+#define KFD_EC_MASK_DEVICE	(KFD_EC_MASK(EC_DEVICE_QUEUE_DELETE) |		\
+				 KFD_EC_MASK(EC_DEVICE_RAS_ERROR) |		\
+				 KFD_EC_MASK(EC_DEVICE_FATAL_HALT) |		\
+				 KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION) |	\
+				 KFD_EC_MASK(EC_DEVICE_NEW))
+#define KFD_EC_MASK_PROCESS	(KFD_EC_MASK(EC_PROCESS_RUNTIME) |	\
+				 KFD_EC_MASK(EC_PROCESS_DEVICE_REMOVE))
+
+/* Checks for exception code types for KFD search */
+#define KFD_DBG_EC_TYPE_IS_QUEUE(ecode)					\
+			(!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_QUEUE))
+#define KFD_DBG_EC_TYPE_IS_DEVICE(ecode)				\
+			(!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_DEVICE))
+#define KFD_DBG_EC_TYPE_IS_PROCESS(ecode)				\
+			(!!(KFD_EC_MASK(ecode) & KFD_EC_MASK_PROCESS))
+
+
+/* Runtime enable states */
+enum kfd_dbg_runtime_state {
+	DEBUG_RUNTIME_STATE_DISABLED = 0,
+	DEBUG_RUNTIME_STATE_ENABLED = 1,
+	DEBUG_RUNTIME_STATE_ENABLED_BUSY = 2,
+	DEBUG_RUNTIME_STATE_ENABLED_ERROR = 3
+};
+
+/* Runtime enable status */
+struct kfd_runtime_info {
+	__u64 r_debug;
+	__u32 runtime_state;
+	__u32 ttmp_setup;
+};
+
+/* Enable modes for runtime enable */
+#define KFD_RUNTIME_ENABLE_MODE_ENABLE_MASK	1
+#define KFD_RUNTIME_ENABLE_MODE_TTMP_SAVE_MASK	2
+
+/**
+ * kfd_ioctl_runtime_enable_args - Arguments for runtime enable
+ *
+ * Coordinates debug exception signalling and debug device enablement with runtime.
+ *
+ * @r_debug - pointer to user struct for sharing information between ROCr and the debuggger
+ * @mode_mask - mask to set mode
+ *	KFD_RUNTIME_ENABLE_MODE_ENABLE_MASK - enable runtime for debugging, otherwise disable
+ *	KFD_RUNTIME_ENABLE_MODE_TTMP_SAVE_MASK - enable trap temporary setup (ignore on disable)
+ * @capabilities_mask - mask to notify runtime on what KFD supports
+ *
+ * Return - 0 on SUCCESS.
+ *	  - EBUSY if runtime enable call already pending.
+ *	  - EEXIST if user queues already active prior to call.
+ *	    If process is debug enabled, runtime enable will enable debug devices and
+ *	    wait for debugger process to send runtime exception EC_PROCESS_RUNTIME
+ *	    to unblock - see kfd_ioctl_dbg_trap_args.
+ *
+ */
+struct kfd_ioctl_runtime_enable_args {
+	__u64 r_debug;
+	__u32 mode_mask;
+	__u32 capabilities_mask;
+};
+
+/* Queue information */
+struct kfd_queue_snapshot_entry {
+	__u64 exception_status;
+	__u64 ring_base_address;
+	__u64 write_pointer_address;
+	__u64 read_pointer_address;
+	__u64 ctx_save_restore_address;
+	__u32 queue_id;
+	__u32 gpu_id;
+	__u32 ring_size;
+	__u32 queue_type;
+	__u32 ctx_save_restore_area_size;
+	__u32 reserved;
+};
+
+/* Queue status return for suspend/resume */
+#define KFD_DBG_QUEUE_ERROR_BIT		30
+#define KFD_DBG_QUEUE_INVALID_BIT	31
+#define KFD_DBG_QUEUE_ERROR_MASK	(1 << KFD_DBG_QUEUE_ERROR_BIT)
+#define KFD_DBG_QUEUE_INVALID_MASK	(1 << KFD_DBG_QUEUE_INVALID_BIT)
+
+/* Context save area header information */
+struct kfd_context_save_area_header {
+	struct {
+		__u32 control_stack_offset;
+		__u32 control_stack_size;
+		__u32 wave_state_offset;
+		__u32 wave_state_size;
+	} wave_state;
+	__u32 debug_offset;
+	__u32 debug_size;
+	__u64 err_payload_addr;
+	__u32 err_event_id;
+	__u32 reserved1;
+};
+
+/*
+ * Debug operations
+ *
+ * For specifics on usage and return values, see documentation per operation
+ * below.  Otherwise, generic error returns apply:
+ *	- ESRCH if the process to debug does not exist.
+ *
+ *	- EINVAL (with KFD_IOC_DBG_TRAP_ENABLE exempt) if operation
+ *		 KFD_IOC_DBG_TRAP_ENABLE has not succeeded prior.
+ *		 Also returns this error if GPU hardware scheduling is not supported.
+ *
+ *	- EPERM (with KFD_IOC_DBG_TRAP_DISABLE exempt) if target process is not
+ *		 PTRACE_ATTACHED.  KFD_IOC_DBG_TRAP_DISABLE is exempt to allow
+ *		 clean up of debug mode as long as process is debug enabled.
+ *
+ *	- EACCES if any DBG_HW_OP (debug hardware operation) is requested when
+ *		 AMDKFD_IOC_RUNTIME_ENABLE has not succeeded prior.
+ *
+ *	- ENODEV if any GPU does not support debugging on a DBG_HW_OP call.
+ *
+ *	- Other errors may be returned when a DBG_HW_OP occurs while the GPU
+ *	  is in a fatal state.
+ *
+ */
+enum kfd_dbg_trap_operations {
+	KFD_IOC_DBG_TRAP_ENABLE = 0,
+	KFD_IOC_DBG_TRAP_DISABLE = 1,
+	KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT = 2,
+	KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED = 3,
+	KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE = 4,  /* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE = 5,      /* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_SUSPEND_QUEUES = 6,		/* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_RESUME_QUEUES = 7,		/* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH = 8,	/* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH = 9,	/* DBG_HW_OP */
+	KFD_IOC_DBG_TRAP_SET_FLAGS = 10,
+	KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT = 11,
+	KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO = 12,
+	KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT = 13,
+	KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT = 14
+};
+
+/**
+ * kfd_ioctl_dbg_trap_enable_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_ENABLE.
+ *
+ *     Enables debug session for target process. Call @op KFD_IOC_DBG_TRAP_DISABLE in
+ *     kfd_ioctl_dbg_trap_args to disable debug session.
+ *
+ *     @exception_mask (IN)	- exceptions to raise to the debugger
+ *     @rinfo_ptr      (IN)	- pointer to runtime info buffer (see kfd_runtime_info)
+ *     @rinfo_size     (IN/OUT)	- size of runtime info buffer in bytes
+ *     @dbg_fd	       (IN)	- fd the KFD will nofify the debugger with of raised
+ *				  exceptions set in exception_mask.
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *		Copies KFD saved kfd_runtime_info to @rinfo_ptr on enable.
+ *		Size of kfd_runtime saved by the KFD returned to @rinfo_size.
+ *            - EBADF if KFD cannot get a reference to dbg_fd.
+ *            - EFAULT if KFD cannot copy runtime info to rinfo_ptr.
+ *            - EINVAL if target process is already debug enabled.
+ *
+ */
+struct kfd_ioctl_dbg_trap_enable_args {
+	__u64 exception_mask;
+	__u64 rinfo_ptr;
+	__u32 rinfo_size;
+	__u32 dbg_fd;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_send_runtime_event_args
+ *
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT.
+ *     Raises exceptions to runtime.
+ *
+ *     @exception_mask (IN) - exceptions to raise to runtime
+ *     @gpu_id	       (IN) - target device id
+ *     @queue_id       (IN) - target queue id
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *	      - ENODEV if gpu_id not found.
+ *		If exception_mask contains EC_PROCESS_RUNTIME, unblocks pending
+ *		AMDKFD_IOC_RUNTIME_ENABLE call - see kfd_ioctl_runtime_enable_args.
+ *		All other exceptions are raised to runtime through err_payload_addr.
+ *		See kfd_context_save_area_header.
+ */
+struct kfd_ioctl_dbg_trap_send_runtime_event_args {
+	__u64 exception_mask;
+	__u32 gpu_id;
+	__u32 queue_id;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_set_exceptions_enabled_args
+ *
+ *     Arguments for KFD_IOC_SET_EXCEPTIONS_ENABLED
+ *     Set new exceptions to be raised to the debugger.
+ *
+ *     @exception_mask (IN) - new exceptions to raise the debugger
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ */
+struct kfd_ioctl_dbg_trap_set_exceptions_enabled_args {
+	__u64 exception_mask;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_set_wave_launch_override_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE
+ *     Enable HW exceptions to raise trap.
+ *
+ *     @override_mode	     (IN)     - see kfd_dbg_trap_override_mode
+ *     @enable_mask	     (IN/OUT) - reference kfd_dbg_trap_mask.
+ *					IN is the override modes requested to be enabled.
+ *					OUT is referenced in Return below.
+ *     @support_request_mask (IN/OUT) - reference kfd_dbg_trap_mask.
+ *					IN is the override modes requested for support check.
+ *					OUT is referenced in Return below.
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *		Previous enablement is returned in @enable_mask.
+ *		Actual override support is returned in @support_request_mask.
+ *	      - EINVAL if override mode is not supported.
+ *	      - EACCES if trap support requested is not actually supported.
+ *		i.e. enable_mask (IN) is not a subset of support_request_mask (OUT).
+ *		Otherwise it is considered a generic error (see kfd_dbg_trap_operations).
+ */
+struct kfd_ioctl_dbg_trap_set_wave_launch_override_args {
+	__u32 override_mode;
+	__u32 enable_mask;
+	__u32 support_request_mask;
+	__u32 pad;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_set_wave_launch_mode_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE
+ *     Set wave launch mode.
+ *
+ *     @mode (IN) - see kfd_dbg_trap_wave_launch_mode
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ */
+struct kfd_ioctl_dbg_trap_set_wave_launch_mode_args {
+	__u32 launch_mode;
+	__u32 pad;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_suspend_queues_ags
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SUSPEND_QUEUES
+ *     Suspend queues.
+ *
+ *     @exception_mask	(IN) - raised exceptions to clear
+ *     @queue_array_ptr (IN) - pointer to array of queue ids (u32 per queue id)
+ *			       to suspend
+ *     @num_queues	(IN) - number of queues to suspend in @queue_array_ptr
+ *     @grace_period	(IN) - wave time allowance before preemption
+ *			       per 1K GPU clock cycle unit
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Destruction of a suspended queue is blocked until the queue is
+ *     resumed.  This allows the debugger to access queue information and
+ *     the its context save area without running into a race condition on
+ *     queue destruction.
+ *     Automatically copies per queue context save area header information
+ *     into the save area base
+ *     (see kfd_queue_snapshot_entry and kfd_context_save_area_header).
+ *
+ *     Return - Number of queues suspended on SUCCESS.
+ *	.	KFD_DBG_QUEUE_ERROR_MASK and KFD_DBG_QUEUE_INVALID_MASK masked
+ *		for each queue id in @queue_array_ptr array reports unsuccessful
+ *		suspend reason.
+ *		KFD_DBG_QUEUE_ERROR_MASK = HW failure.
+ *		KFD_DBG_QUEUE_INVALID_MASK = queue does not exist, is new or
+ *		is being destroyed.
+ */
+struct kfd_ioctl_dbg_trap_suspend_queues_args {
+	__u64 exception_mask;
+	__u64 queue_array_ptr;
+	__u32 num_queues;
+	__u32 grace_period;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_resume_queues_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_RESUME_QUEUES
+ *     Resume queues.
+ *
+ *     @queue_array_ptr (IN) - pointer to array of queue ids (u32 per queue id)
+ *			       to resume
+ *     @num_queues	(IN) - number of queues to resume in @queue_array_ptr
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - Number of queues resumed on SUCCESS.
+ *		KFD_DBG_QUEUE_ERROR_MASK and KFD_DBG_QUEUE_INVALID_MASK mask
+ *		for each queue id in @queue_array_ptr array reports unsuccessful
+ *		resume reason.
+ *		KFD_DBG_QUEUE_ERROR_MASK = HW failure.
+ *		KFD_DBG_QUEUE_INVALID_MASK = queue does not exist.
+ */
+struct kfd_ioctl_dbg_trap_resume_queues_args {
+	__u64 queue_array_ptr;
+	__u32 num_queues;
+	__u32 pad;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_set_node_address_watch_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH
+ *     Sets address watch for device.
+ *
+ *     @address	(IN)  - watch address to set
+ *     @mode    (IN)  - see kfd_dbg_trap_address_watch_mode
+ *     @mask    (IN)  - watch address mask
+ *     @gpu_id  (IN)  - target gpu to set watch point
+ *     @id      (OUT) - watch id allocated
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *		Allocated watch ID returned to @id.
+ *	      - ENODEV if gpu_id not found.
+ *	      - ENOMEM if watch IDs can be allocated
+ */
+struct kfd_ioctl_dbg_trap_set_node_address_watch_args {
+	__u64 address;
+	__u32 mode;
+	__u32 mask;
+	__u32 gpu_id;
+	__u32 id;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_clear_node_address_watch_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH
+ *     Clear address watch for device.
+ *
+ *     @gpu_id  (IN)  - target device to clear watch point
+ *     @id      (IN) - allocated watch id to clear
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *	      - ENODEV if gpu_id not found.
+ *	      - EINVAL if watch ID has not been allocated.
+ */
+struct kfd_ioctl_dbg_trap_clear_node_address_watch_args {
+	__u32 gpu_id;
+	__u32 id;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_set_flags_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_SET_FLAGS
+ *     Sets flags for wave behaviour.
+ *
+ *     @flags (IN/OUT) - IN = flags to enable, OUT = flags previously enabled
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *	      - EACCESS if any debug device does not allow flag options.
+ */
+struct kfd_ioctl_dbg_trap_set_flags_args {
+	__u32 flags;
+	__u32 pad;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_query_debug_event_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT
+ *
+ *     Find one or more raised exceptions. This function can return multiple
+ *     exceptions from a single queue or a single device with one call. To find
+ *     all raised exceptions, this function must be called repeatedly until it
+ *     returns -EAGAIN. Returned exceptions can optionally be cleared by
+ *     setting the corresponding bit in the @exception_mask input parameter.
+ *     However, clearing an exception prevents retrieving further information
+ *     about it with KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO.
+ *
+ *     @exception_mask (IN/OUT) - exception to clear (IN) and raised (OUT)
+ *     @gpu_id	       (OUT)    - gpu id of exceptions raised
+ *     @queue_id       (OUT)    - queue id of exceptions raised
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on raised exception found
+ *              Raised exceptions found are returned in @exception mask
+ *              with reported source id returned in @gpu_id or @queue_id.
+ *            - EAGAIN if no raised exception has been found
+ */
+struct kfd_ioctl_dbg_trap_query_debug_event_args {
+	__u64 exception_mask;
+	__u32 gpu_id;
+	__u32 queue_id;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_query_exception_info_args
+ *
+ *     Arguments KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO
+ *     Get additional info on raised exception.
+ *
+ *     @info_ptr	(IN)	 - pointer to exception info buffer to copy to
+ *     @info_size	(IN/OUT) - exception info buffer size (bytes)
+ *     @source_id	(IN)     - target gpu or queue id
+ *     @exception_code	(IN)     - target exception
+ *     @clear_exception	(IN)     - clear raised @exception_code exception
+ *				   (0 = false, 1 = true)
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *              If @exception_code is EC_DEVICE_MEMORY_VIOLATION, copy @info_size(OUT)
+ *		bytes of memory exception data to @info_ptr.
+ *              If @exception_code is EC_PROCESS_RUNTIME, copy saved
+ *              kfd_runtime_info to @info_ptr.
+ *              Actual required @info_ptr size (bytes) is returned in @info_size.
+ */
+struct kfd_ioctl_dbg_trap_query_exception_info_args {
+	__u64 info_ptr;
+	__u32 info_size;
+	__u32 source_id;
+	__u32 exception_code;
+	__u32 clear_exception;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_get_queue_snapshot_args
+ *
+ *     Arguments KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT
+ *     Get queue information.
+ *
+ *     @exception_mask	 (IN)	  - exceptions raised to clear
+ *     @snapshot_buf_ptr (IN)	  - queue snapshot entry buffer (see kfd_queue_snapshot_entry)
+ *     @num_queues	 (IN/OUT) - number of queue snapshot entries
+ *         The debugger specifies the size of the array allocated in @num_queues.
+ *         KFD returns the number of queues that actually existed. If this is
+ *         larger than the size specified by the debugger, KFD will not overflow
+ *         the array allocated by the debugger.
+ *
+ *     @entry_size	 (IN/OUT) - size per entry in bytes
+ *         The debugger specifies sizeof(struct kfd_queue_snapshot_entry) in
+ *         @entry_size. KFD returns the number of bytes actually populated per
+ *         entry. The debugger should use the KFD_IOCTL_MINOR_VERSION to determine,
+ *         which fields in struct kfd_queue_snapshot_entry are valid. This allows
+ *         growing the ABI in a backwards compatible manner.
+ *         Note that entry_size(IN) should still be used to stride the snapshot buffer in the
+ *         event that it's larger than actual kfd_queue_snapshot_entry.
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *              Copies @num_queues(IN) queue snapshot entries of size @entry_size(IN)
+ *              into @snapshot_buf_ptr if @num_queues(IN) > 0.
+ *              Otherwise return @num_queues(OUT) queue snapshot entries that exist.
+ */
+struct kfd_ioctl_dbg_trap_queue_snapshot_args {
+	__u64 exception_mask;
+	__u64 snapshot_buf_ptr;
+	__u32 num_queues;
+	__u32 entry_size;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_get_device_snapshot_args
+ *
+ *     Arguments for KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT
+ *     Get device information.
+ *
+ *     @exception_mask	 (IN)	  - exceptions raised to clear
+ *     @snapshot_buf_ptr (IN)	  - pointer to snapshot buffer (see kfd_dbg_device_info_entry)
+ *     @num_devices	 (IN/OUT) - number of debug devices to snapshot
+ *         The debugger specifies the size of the array allocated in @num_devices.
+ *         KFD returns the number of devices that actually existed. If this is
+ *         larger than the size specified by the debugger, KFD will not overflow
+ *         the array allocated by the debugger.
+ *
+ *     @entry_size	 (IN/OUT) - size per entry in bytes
+ *         The debugger specifies sizeof(struct kfd_dbg_device_info_entry) in
+ *         @entry_size. KFD returns the number of bytes actually populated. The
+ *         debugger should use KFD_IOCTL_MINOR_VERSION to determine, which fields
+ *         in struct kfd_dbg_device_info_entry are valid. This allows growing the
+ *         ABI in a backwards compatible manner.
+ *         Note that entry_size(IN) should still be used to stride the snapshot buffer in the
+ *         event that it's larger than actual kfd_dbg_device_info_entry.
+ *
+ *     Generic errors apply (see kfd_dbg_trap_operations).
+ *     Return - 0 on SUCCESS.
+ *              Copies @num_devices(IN) device snapshot entries of size @entry_size(IN)
+ *              into @snapshot_buf_ptr if @num_devices(IN) > 0.
+ *              Otherwise return @num_devices(OUT) queue snapshot entries that exist.
+ */
+struct kfd_ioctl_dbg_trap_device_snapshot_args {
+	__u64 exception_mask;
+	__u64 snapshot_buf_ptr;
+	__u32 num_devices;
+	__u32 entry_size;
+};
+
+/**
+ * kfd_ioctl_dbg_trap_args
+ *
+ * Arguments to debug target process.
+ *
+ *     @pid - target process to debug
+ *     @op  - debug operation (see kfd_dbg_trap_operations)
+ *
+ *     @op determines which union struct args to use.
+ *     Refer to kern docs for each kfd_ioctl_dbg_trap_*_args struct.
+ */
+struct kfd_ioctl_dbg_trap_args {
+	__u32 pid;
+	__u32 op;
+
+	union {
+		struct kfd_ioctl_dbg_trap_enable_args enable;
+		struct kfd_ioctl_dbg_trap_send_runtime_event_args send_runtime_event;
+		struct kfd_ioctl_dbg_trap_set_exceptions_enabled_args set_exceptions_enabled;
+		struct kfd_ioctl_dbg_trap_set_wave_launch_override_args launch_override;
+		struct kfd_ioctl_dbg_trap_set_wave_launch_mode_args launch_mode;
+		struct kfd_ioctl_dbg_trap_suspend_queues_args suspend_queues;
+		struct kfd_ioctl_dbg_trap_resume_queues_args resume_queues;
+		struct kfd_ioctl_dbg_trap_set_node_address_watch_args set_node_address_watch;
+		struct kfd_ioctl_dbg_trap_clear_node_address_watch_args clear_node_address_watch;
+		struct kfd_ioctl_dbg_trap_set_flags_args set_flags;
+		struct kfd_ioctl_dbg_trap_query_debug_event_args query_debug_event;
+		struct kfd_ioctl_dbg_trap_query_exception_info_args query_exception_info;
+		struct kfd_ioctl_dbg_trap_queue_snapshot_args queue_snapshot;
+		struct kfd_ioctl_dbg_trap_device_snapshot_args device_snapshot;
+	};
+};
+
 #define AMDKFD_IOCTL_BASE 'K'
 #define AMDKFD_IO(nr)			_IO(AMDKFD_IOCTL_BASE, nr)
 #define AMDKFD_IOR(nr, type)		_IOR(AMDKFD_IOCTL_BASE, nr, type)
@@ -887,7 +1547,13 @@ struct kfd_ioctl_set_xnack_mode_args {
 #define AMDKFD_IOC_EXPORT_DMABUF		\
 		AMDKFD_IOWR(0x24, struct kfd_ioctl_export_dmabuf_args)
 
+#define AMDKFD_IOC_RUNTIME_ENABLE		\
+		AMDKFD_IOWR(0x25, struct kfd_ioctl_runtime_enable_args)
+
+#define AMDKFD_IOC_DBG_TRAP			\
+		AMDKFD_IOWR(0x26, struct kfd_ioctl_dbg_trap_args)
+
 #define AMDKFD_COMMAND_START		0x01
-#define AMDKFD_COMMAND_END		0x25
+#define AMDKFD_COMMAND_END		0x27
 
 #endif
-- 
GitLab


From d230f1bfe7a1977565ce1e2804ddb7b7a3d911ff Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 25 Mar 2022 12:39:06 -0400
Subject: [PATCH 1515/1791] drm/amdkfd: display debug capabilities

Expose debug capabilities in the KFD topology node's HSA capabilities and
debug properties flags.

Ensure correct capabilities are exposed based on firmware support.

Flag definitions can be referenced in uapi/linux/kfd_sysfs.h.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 101 ++++++++++++++++++++--
 drivers/gpu/drm/amd/amdkfd/kfd_topology.h |   6 ++
 include/uapi/linux/kfd_sysfs.h            |  15 ++++
 3 files changed, 117 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 8302d8967158b..3def25b2bdbbc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -535,6 +535,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 				      dev->gpu->kfd->mec_fw_version);
 		sysfs_show_32bit_prop(buffer, offs, "capability",
 				      dev->node_props.capability);
+		sysfs_show_64bit_prop(buffer, offs, "debug_prop",
+				      dev->node_props.debug_prop);
 		sysfs_show_32bit_prop(buffer, offs, "sdma_fw_version",
 				      dev->gpu->kfd->sdma_fw_version);
 		sysfs_show_64bit_prop(buffer, offs, "unique_id",
@@ -1857,6 +1859,97 @@ static int kfd_topology_add_device_locked(struct kfd_node *gpu, uint32_t gpu_id,
 	return res;
 }
 
+static void kfd_topology_set_dbg_firmware_support(struct kfd_topology_device *dev)
+{
+	bool firmware_supported = true;
+
+	if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(11, 0, 0) &&
+			KFD_GC_VERSION(dev->gpu) < IP_VERSION(12, 0, 0)) {
+		firmware_supported =
+			(dev->gpu->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 9;
+		goto out;
+	}
+
+	/*
+	 * Note: Any unlisted devices here are assumed to support exception handling.
+	 * Add additional checks here as needed.
+	 */
+	switch (KFD_GC_VERSION(dev->gpu)) {
+	case IP_VERSION(9, 0, 1):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 459 + 32768;
+		break;
+	case IP_VERSION(9, 1, 0):
+	case IP_VERSION(9, 2, 1):
+	case IP_VERSION(9, 2, 2):
+	case IP_VERSION(9, 3, 0):
+	case IP_VERSION(9, 4, 0):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 459;
+		break;
+	case IP_VERSION(9, 4, 1):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 60;
+		break;
+	case IP_VERSION(9, 4, 2):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 51;
+		break;
+	case IP_VERSION(10, 1, 10):
+	case IP_VERSION(10, 1, 2):
+	case IP_VERSION(10, 1, 1):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 144;
+		break;
+	case IP_VERSION(10, 3, 0):
+	case IP_VERSION(10, 3, 2):
+	case IP_VERSION(10, 3, 1):
+	case IP_VERSION(10, 3, 4):
+	case IP_VERSION(10, 3, 5):
+		firmware_supported = dev->gpu->kfd->mec_fw_version >= 89;
+		break;
+	case IP_VERSION(10, 1, 3):
+	case IP_VERSION(10, 3, 3):
+		firmware_supported = false;
+		break;
+	default:
+		break;
+	}
+
+out:
+	if (firmware_supported)
+		dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_FIRMWARE_SUPPORTED;
+}
+
+static void kfd_topology_set_capabilities(struct kfd_topology_device *dev)
+{
+	dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 <<
+				HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) &
+				HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK);
+
+	dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_SUPPORT |
+			HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_TRAP_OVERRIDE_SUPPORTED |
+			HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_MODE_SUPPORTED;
+
+	if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) {
+		dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 |
+						HSA_DBG_WATCH_ADDR_MASK_HI_BIT;
+
+		if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(9, 4, 2))
+			dev->node_props.debug_prop |=
+				HSA_DBG_DISPATCH_INFO_ALWAYS_VALID;
+		else
+			dev->node_props.capability |=
+				HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED;
+	} else {
+		dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 |
+					HSA_DBG_WATCH_ADDR_MASK_HI_BIT;
+
+		if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(11, 0, 0))
+			dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID;
+		else
+			dev->node_props.capability |=
+				HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED;
+	}
+
+	kfd_topology_set_dbg_firmware_support(dev);
+}
+
 int kfd_topology_add_device(struct kfd_node *gpu)
 {
 	uint32_t gpu_id;
@@ -1967,13 +2060,11 @@ int kfd_topology_add_device(struct kfd_node *gpu)
 			HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK);
 		break;
 	default:
-		if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(9, 0, 1))
-			dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 <<
-				HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) &
-				HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK);
-		else
+		if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(9, 0, 1))
 			WARN(1, "Unexpected ASIC family %u",
 			     dev->gpu->adev->asic_type);
+		else
+			kfd_topology_set_capabilities(dev);
 	}
 
 	/*
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
index 3b8afb6aba79e..cba2cd5ed9d19 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h
@@ -31,6 +31,11 @@
 
 #define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 32
 
+#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9	6
+#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10	7
+#define HSA_DBG_WATCH_ADDR_MASK_HI_BIT  \
+			(29 << HSA_DBG_WATCH_ADDR_MASK_HI_BIT_SHIFT)
+
 struct kfd_node_properties {
 	uint64_t hive_id;
 	uint32_t cpu_cores_count;
@@ -42,6 +47,7 @@ struct kfd_node_properties {
 	uint32_t cpu_core_id_base;
 	uint32_t simd_id_base;
 	uint32_t capability;
+	uint64_t debug_prop;
 	uint32_t max_waves_per_simd;
 	uint32_t lds_size_in_kb;
 	uint32_t gds_size_in_kb;
diff --git a/include/uapi/linux/kfd_sysfs.h b/include/uapi/linux/kfd_sysfs.h
index 3e330f368917d..a51b7331e0b4b 100644
--- a/include/uapi/linux/kfd_sysfs.h
+++ b/include/uapi/linux/kfd_sysfs.h
@@ -43,6 +43,11 @@
 #define HSA_CAP_DOORBELL_TYPE_2_0		0x2
 #define HSA_CAP_AQL_QUEUE_DOUBLE_MAP		0x00004000
 
+#define HSA_CAP_TRAP_DEBUG_SUPPORT              0x00008000
+#define HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_TRAP_OVERRIDE_SUPPORTED  0x00010000
+#define HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_MODE_SUPPORTED           0x00020000
+#define HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED  0x00040000
+
 /* Old buggy user mode depends on this being 0 */
 #define HSA_CAP_RESERVED_WAS_SRAM_EDCSUPPORTED	0x00080000
 
@@ -53,8 +58,18 @@
 #define HSA_CAP_SRAM_EDCSUPPORTED		0x04000000
 #define HSA_CAP_SVMAPI_SUPPORTED		0x08000000
 #define HSA_CAP_FLAGS_COHERENTHOSTACCESS	0x10000000
+#define HSA_CAP_TRAP_DEBUG_FIRMWARE_SUPPORTED   0x20000000
 #define HSA_CAP_RESERVED			0xe00f8000
 
+/* debug_prop bits in node properties */
+#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_MASK     0x0000000f
+#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_SHIFT    0
+#define HSA_DBG_WATCH_ADDR_MASK_HI_BIT_MASK     0x000003f0
+#define HSA_DBG_WATCH_ADDR_MASK_HI_BIT_SHIFT    4
+#define HSA_DBG_DISPATCH_INFO_ALWAYS_VALID      0x00000400
+#define HSA_DBG_WATCHPOINTS_EXCLUSIVE           0x00000800
+#define HSA_DBG_RESERVED                0xfffffffffffff000ull
+
 /* Heap types in memory properties */
 #define HSA_MEM_HEAP_TYPE_SYSTEM	0
 #define HSA_MEM_HEAP_TYPE_FB_PUBLIC	1
-- 
GitLab


From 0ab2d7532b05a3e7c06fd3b0c8bd6b46c1dfb508 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 25 Mar 2022 14:55:30 -0400
Subject: [PATCH 1516/1791] drm/amdkfd: prepare per-process debug enable and
 disable

The ROCm debugger will attach to a process to debug by PTRACE and will
expect the KFD to prepare a process for the target PID, whether the
target PID has opened the KFD device or not.

This patch is to explicity handle this requirement.  Further HW mode
setting and runtime coordination requirements will be handled in
following patches.

In the case where the target process has not opened the KFD device,
a new KFD process must be created for the target PID.
The debugger as well as the target process for this case will have not
acquired any VMs so handle process restoration to correctly account for
this.

To coordinate with HSA runtime, the debugger must be aware of the target
process' runtime enablement status and will copy the runtime status
information into the debugged KFD process for later query.

On enablement, the debugger will subscribe to a set of exceptions where
each exception events will notify the debugger through a pollable FIFO
file descriptor that the debugger provides to the KFD to manage.

Finally on process termination of either the debugger or the target,
debugging must be disabled if it has not been done so.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/Makefile           |   3 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      | 102 +++++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        |  80 ++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  32 ++++++
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |  26 +++--
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  31 +++++-
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |  60 +++++++----
 7 files changed, 304 insertions(+), 30 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/amdkfd/kfd_debug.c
 create mode 100644 drivers/gpu/drm/amd/amdkfd/kfd_debug.h

diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile
index e758c2a24cd08..7477544280736 100644
--- a/drivers/gpu/drm/amd/amdkfd/Makefile
+++ b/drivers/gpu/drm/amd/amdkfd/Makefile
@@ -55,7 +55,8 @@ AMDKFD_FILES	:= $(AMDKFD_PATH)/kfd_module.o \
 		$(AMDKFD_PATH)/kfd_int_process_v9.o \
 		$(AMDKFD_PATH)/kfd_int_process_v11.o \
 		$(AMDKFD_PATH)/kfd_smi_events.o \
-		$(AMDKFD_PATH)/kfd_crat.o
+		$(AMDKFD_PATH)/kfd_crat.o \
+		$(AMDKFD_PATH)/kfd_debug.o
 
 ifneq ($(CONFIG_AMD_IOMMU_V2),)
 AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 00e34125987c1..ee086a0a46dfe 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -44,6 +44,7 @@
 #include "amdgpu_amdkfd.h"
 #include "kfd_smi_events.h"
 #include "amdgpu_dma_buf.h"
+#include "kfd_debug.h"
 
 static long kfd_ioctl(struct file *, unsigned int, unsigned long);
 static int kfd_open(struct inode *, struct file *);
@@ -142,10 +143,15 @@ static int kfd_open(struct inode *inode, struct file *filep)
 		return -EPERM;
 	}
 
-	process = kfd_create_process(filep);
+	process = kfd_create_process(current);
 	if (IS_ERR(process))
 		return PTR_ERR(process);
 
+	if (kfd_process_init_cwsr_apu(process, filep)) {
+		kfd_unref_process(process);
+		return -EFAULT;
+	}
+
 	/* filep now owns the reference returned by kfd_create_process */
 	filep->private_data = process;
 
@@ -2737,6 +2743,10 @@ static int kfd_ioctl_runtime_enable(struct file *filep, struct kfd_process *p, v
 static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, void *data)
 {
 	struct kfd_ioctl_dbg_trap_args *args = data;
+	struct task_struct *thread = NULL;
+	struct mm_struct *mm = NULL;
+	struct pid *pid = NULL;
+	struct kfd_process *target = NULL;
 	int r = 0;
 
 	if (sched_policy == KFD_SCHED_POLICY_NO_HWS) {
@@ -2744,9 +2754,81 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		return -EINVAL;
 	}
 
+	pid = find_get_pid(args->pid);
+	if (!pid) {
+		pr_debug("Cannot find pid info for %i\n", args->pid);
+		r = -ESRCH;
+		goto out;
+	}
+
+	thread = get_pid_task(pid, PIDTYPE_PID);
+	if (!thread) {
+		r = -ESRCH;
+		goto out;
+	}
+
+	mm = get_task_mm(thread);
+	if (!mm) {
+		r = -ESRCH;
+		goto out;
+	}
+
+	if (args->op == KFD_IOC_DBG_TRAP_ENABLE) {
+		bool create_process;
+
+		rcu_read_lock();
+		create_process = thread && thread != current && ptrace_parent(thread) == current;
+		rcu_read_unlock();
+
+		target = create_process ? kfd_create_process(thread) :
+					kfd_lookup_process_by_pid(pid);
+	} else {
+		target = kfd_lookup_process_by_pid(pid);
+	}
+
+	if (!target) {
+		pr_debug("Cannot find process PID %i to debug\n", args->pid);
+		r = -ESRCH;
+		goto out;
+	}
+
+	/* Check if target is still PTRACED. */
+	rcu_read_lock();
+	if (target != p && args->op != KFD_IOC_DBG_TRAP_DISABLE
+				&& ptrace_parent(target->lead_thread) != current) {
+		pr_err("PID %i is not PTRACED and cannot be debugged\n", args->pid);
+		r = -EPERM;
+	}
+	rcu_read_unlock();
+
+	if (r)
+		goto out;
+
+	mutex_lock(&target->mutex);
+
+	if (args->op != KFD_IOC_DBG_TRAP_ENABLE && !target->debug_trap_enabled) {
+		pr_err("PID %i not debug enabled for op %i\n", args->pid, args->op);
+		r = -EINVAL;
+		goto unlock_out;
+	}
+
 	switch (args->op) {
 	case KFD_IOC_DBG_TRAP_ENABLE:
+		if (target != p)
+			target->debugger_process = p;
+
+		r = kfd_dbg_trap_enable(target,
+					args->enable.dbg_fd,
+					(void __user *)args->enable.rinfo_ptr,
+					&args->enable.rinfo_size);
+		if (!r)
+			target->exception_enable_mask = args->enable.exception_mask;
+
+		pr_warn("Debug functions limited\n");
+		break;
 	case KFD_IOC_DBG_TRAP_DISABLE:
+		r = kfd_dbg_trap_disable(target);
+		break;
 	case KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT:
 	case KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED:
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE:
@@ -2760,7 +2842,7 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
 	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
 	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
-		pr_warn("Debugging not supported yet\n");
+		pr_warn("Debug op %i not supported yet\n", args->op);
 		r = -EACCES;
 		break;
 	default:
@@ -2768,6 +2850,22 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		r = -EINVAL;
 	}
 
+unlock_out:
+	mutex_unlock(&target->mutex);
+
+out:
+	if (thread)
+		put_task_struct(thread);
+
+	if (mm)
+		mmput(mm);
+
+	if (pid)
+		put_pid(pid);
+
+	if (target)
+		kfd_unref_process(target);
+
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
new file mode 100644
index 0000000000000..898cc1fe3d135
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -0,0 +1,80 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "kfd_debug.h"
+#include <linux/file.h>
+
+int kfd_dbg_trap_disable(struct kfd_process *target)
+{
+	if (!target->debug_trap_enabled)
+		return 0;
+
+	fput(target->dbg_ev_file);
+	target->dbg_ev_file = NULL;
+
+	if (target->debugger_process) {
+		atomic_dec(&target->debugger_process->debugged_process_count);
+		target->debugger_process = NULL;
+	}
+
+	target->debug_trap_enabled = false;
+	kfd_unref_process(target);
+
+	return 0;
+}
+
+int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
+			void __user *runtime_info, uint32_t *runtime_size)
+{
+	struct file *f;
+	uint32_t copy_size;
+	int r = 0;
+
+	if (target->debug_trap_enabled)
+		return -EALREADY;
+
+	copy_size = min((size_t)(*runtime_size), sizeof(target->runtime_info));
+
+	f = fget(fd);
+	if (!f) {
+		pr_err("Failed to get file for (%i)\n", fd);
+		return -EBADF;
+	}
+
+	target->dbg_ev_file = f;
+
+	/* We already hold the process reference but hold another one for the
+	 * debug session.
+	 */
+	kref_get(&target->ref);
+	target->debug_trap_enabled = true;
+
+	if (target->debugger_process)
+		atomic_inc(&target->debugger_process->debugged_process_count);
+
+	if (copy_to_user(runtime_info, (void *)&target->runtime_info, copy_size))
+		r = -EFAULT;
+
+	*runtime_size = sizeof(target->runtime_info);
+
+	return r;
+}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
new file mode 100644
index 0000000000000..a8abfe2a0a140
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef KFD_DEBUG_EVENTS_H_INCLUDED
+#define KFD_DEBUG_EVENTS_H_INCLUDED
+
+#include "kfd_priv.h"
+
+int kfd_dbg_trap_disable(struct kfd_process *target);
+int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
+			void __user *runtime_info,
+			uint32_t *runtime_info_size);
+#endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 80cddb46657fa..2baa0781eafc2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1006,6 +1006,14 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		goto out;
 
 	pdd = qpd_to_pdd(qpd);
+
+	/* The debugger creates processes that temporarily have not acquired
+	 * all VMs for all devices and has no VMs itself.
+	 * Skip queue eviction on process eviction.
+	 */
+	if (!pdd->drm_priv)
+		goto out;
+
 	pr_debug_ratelimited("Evicting PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
@@ -1127,13 +1135,10 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 {
 	struct queue *q;
 	struct kfd_process_device *pdd;
-	uint64_t pd_base;
 	uint64_t eviction_duration;
 	int retval = 0;
 
 	pdd = qpd_to_pdd(qpd);
-	/* Retrieve PD base */
-	pd_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->drm_priv);
 
 	dqm_lock(dqm);
 	if (WARN_ON_ONCE(!qpd->evicted)) /* already restored, do nothing */
@@ -1143,12 +1148,19 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 		goto out;
 	}
 
+	/* The debugger creates processes that temporarily have not acquired
+	 * all VMs for all devices and has no VMs itself.
+	 * Skip queue restore on process restore.
+	 */
+	if (!pdd->drm_priv)
+		goto vm_not_acquired;
+
 	pr_debug_ratelimited("Restoring PASID 0x%x queues\n",
 			    pdd->process->pasid);
 
 	/* Update PD Base in QPD */
-	qpd->page_table_base = pd_base;
-	pr_debug("Updated PD address to 0x%llx\n", pd_base);
+	qpd->page_table_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->drm_priv);
+	pr_debug("Updated PD address to 0x%llx\n", qpd->page_table_base);
 
 	/* activate all active queues on the qpd */
 	list_for_each_entry(q, &qpd->queues_list, list) {
@@ -1171,9 +1183,11 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		retval = execute_queues_cpsch(dqm,
 					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
-	qpd->evicted = 0;
+
 	eviction_duration = get_jiffies_64() - pdd->last_evict_timestamp;
 	atomic64_add(eviction_duration, &pdd->evict_duration_counter);
+vm_not_acquired:
+	qpd->evicted = 0;
 out:
 	dqm_unlock(dqm);
 	return retval;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 3bd222e8f6c3c..1b272f879b4c4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -920,11 +920,33 @@ struct kfd_process {
 	 */
 	unsigned long last_restore_timestamp;
 
+	/* Indicates device process is debug attached with reserved vmid. */
+	bool debug_trap_enabled;
+
+	/* per-process-per device debug event fd file */
+	struct file *dbg_ev_file;
+
+	/* If the process is a kfd debugger, we need to know so we can clean
+	 * up at exit time.  If a process enables debugging on itself, it does
+	 * its own clean-up, so we don't set the flag here.  We track this by
+	 * counting the number of processes this process is debugging.
+	 */
+	atomic_t debugged_process_count;
+
+	/* If the process is a debugged, this is the debugger process */
+	struct kfd_process *debugger_process;
+
 	/* Kobj for our procfs */
 	struct kobject *kobj;
 	struct kobject *kobj_queues;
 	struct attribute attr_pasid;
 
+	/* Keep track cwsr init */
+	bool has_cwsr;
+
+	/* Exception code enable mask and status */
+	uint64_t exception_enable_mask;
+
 	/* shared virtual memory registered by this process */
 	struct svm_range_list svms;
 
@@ -933,6 +955,10 @@ struct kfd_process {
 	atomic_t poison;
 	/* Queues are in paused stated because we are in the process of doing a CRIU checkpoint */
 	bool queues_paused;
+
+	/* Tracks runtime enable status */
+	struct kfd_runtime_info runtime_info;
+
 };
 
 #define KFD_PROCESS_TABLE_SIZE 5 /* bits: 32 entries */
@@ -963,7 +989,7 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev);
 int kfd_process_create_wq(void);
 void kfd_process_destroy_wq(void);
 void kfd_cleanup_processes(void);
-struct kfd_process *kfd_create_process(struct file *filep);
+struct kfd_process *kfd_create_process(struct task_struct *thread);
 struct kfd_process *kfd_get_process(const struct task_struct *task);
 struct kfd_process *kfd_lookup_process_by_pasid(u32 pasid);
 struct kfd_process *kfd_lookup_process_by_mm(const struct mm_struct *mm);
@@ -1108,6 +1134,9 @@ void kfd_process_set_trap_handler(struct qcm_process_device *qpd,
 				  uint64_t tba_addr,
 				  uint64_t tma_addr);
 
+/* CWSR initialization */
+int kfd_process_init_cwsr_apu(struct kfd_process *process, struct file *filep);
+
 /* CRIU */
 /*
  * Need to increment KFD_CRIU_PRIV_VERSION each time a change is made to any of the CRIU private
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 7f7d1378a2f89..ef67f5e37c6f4 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -44,6 +44,7 @@ struct mm_struct;
 #include "kfd_iommu.h"
 #include "kfd_svm.h"
 #include "kfd_smi_events.h"
+#include "kfd_debug.h"
 
 /*
  * List of struct kfd_process (field kfd_process).
@@ -69,7 +70,6 @@ static struct kfd_process *find_process(const struct task_struct *thread,
 					bool ref);
 static void kfd_process_ref_release(struct kref *ref);
 static struct kfd_process *create_process(const struct task_struct *thread);
-static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep);
 
 static void evict_process_worker(struct work_struct *work);
 static void restore_process_worker(struct work_struct *work);
@@ -798,18 +798,19 @@ static void kfd_process_device_destroy_ib_mem(struct kfd_process_device *pdd)
 	kfd_process_free_gpuvm(qpd->ib_mem, pdd, &qpd->ib_kaddr);
 }
 
-struct kfd_process *kfd_create_process(struct file *filep)
+struct kfd_process *kfd_create_process(struct task_struct *thread)
 {
 	struct kfd_process *process;
-	struct task_struct *thread = current;
 	int ret;
 
-	if (!thread->mm)
+	if (!(thread->mm && mmget_not_zero(thread->mm)))
 		return ERR_PTR(-EINVAL);
 
 	/* Only the pthreads threading model is supported. */
-	if (thread->group_leader->mm != thread->mm)
+	if (thread->group_leader->mm != thread->mm) {
+		mmput(thread->mm);
 		return ERR_PTR(-EINVAL);
+	}
 
 	/*
 	 * take kfd processes mutex before starting of process creation
@@ -833,10 +834,6 @@ struct kfd_process *kfd_create_process(struct file *filep)
 		if (IS_ERR(process))
 			goto out;
 
-		ret = kfd_process_init_cwsr_apu(process, filep);
-		if (ret)
-			goto out_destroy;
-
 		if (!procfs.kobj)
 			goto out;
 
@@ -870,16 +867,9 @@ struct kfd_process *kfd_create_process(struct file *filep)
 	if (!IS_ERR(process))
 		kref_get(&process->ref);
 	mutex_unlock(&kfd_processes_mutex);
+	mmput(thread->mm);
 
 	return process;
-
-out_destroy:
-	hash_del_rcu(&process->kfd_processes);
-	mutex_unlock(&kfd_processes_mutex);
-	synchronize_srcu(&kfd_processes_srcu);
-	/* kfd_process_free_notifier will trigger the cleanup */
-	mmu_notifier_put(&process->mmu_notifier);
-	return ERR_PTR(ret);
 }
 
 struct kfd_process *kfd_get_process(const struct task_struct *thread)
@@ -1180,6 +1170,25 @@ static void kfd_process_notifier_release_internal(struct kfd_process *p)
 
 	/* Indicate to other users that MM is no longer valid */
 	p->mm = NULL;
+	kfd_dbg_trap_disable(p);
+
+	if (atomic_read(&p->debugged_process_count) > 0) {
+		struct kfd_process *target;
+		unsigned int temp;
+		int idx = srcu_read_lock(&kfd_processes_srcu);
+
+		hash_for_each_rcu(kfd_processes_table, temp, target, kfd_processes) {
+			if (target->debugger_process && target->debugger_process == p) {
+				mutex_lock_nested(&target->mutex, 1);
+				kfd_dbg_trap_disable(target);
+				mutex_unlock(&target->mutex);
+				if (atomic_read(&p->debugged_process_count) == 0)
+					break;
+			}
+		}
+
+		srcu_read_unlock(&kfd_processes_srcu, idx);
+	}
 
 	mmu_notifier_put(&p->mmu_notifier);
 }
@@ -1259,11 +1268,14 @@ void kfd_cleanup_processes(void)
 	mmu_notifier_synchronize();
 }
 
-static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
+int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 {
 	unsigned long  offset;
 	int i;
 
+	if (p->has_cwsr)
+		return 0;
+
 	for (i = 0; i < p->n_pdds; i++) {
 		struct kfd_node *dev = p->pdds[i]->dev;
 		struct qcm_process_device *qpd = &p->pdds[i]->qpd;
@@ -1292,6 +1304,8 @@ static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 			qpd->tba_addr, qpd->tma_addr, qpd->cwsr_kaddr);
 	}
 
+	p->has_cwsr = true;
+
 	return 0;
 }
 
@@ -1434,6 +1448,10 @@ static struct kfd_process *create_process(const struct task_struct *thread)
 	if (err)
 		goto err_event_init;
 	process->is_32bit_user_mode = in_compat_syscall();
+	process->debug_trap_enabled = false;
+	process->debugger_process = NULL;
+	process->exception_enable_mask = 0;
+	atomic_set(&process->debugged_process_count, 0);
 
 	process->pasid = kfd_pasid_alloc();
 	if (process->pasid == 0) {
@@ -1967,8 +1985,10 @@ static void restore_process_worker(struct work_struct *work)
 	 */
 
 	p->last_restore_timestamp = get_jiffies_64();
-	ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info,
-						     &p->ef);
+	/* VMs may not have been acquired yet during debugging. */
+	if (p->kgd_process_info)
+		ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info,
+							     &p->ef);
 	if (ret) {
 		pr_debug("Failed to restore BOs of pasid 0x%x, retry after %d ms\n",
 			 p->pasid, PROCESS_BACK_OFF_TIME_MS);
-- 
GitLab


From 08ca712270028111b22e4b159d11dbd6b770135e Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 30 Mar 2022 14:54:16 -0400
Subject: [PATCH 1517/1791] drm/amdgpu: add kgd hw debug mode setting interface

Introduce the require KGD debug calls that will execute hardware debug
mode setting.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/include/kgd_kfd_interface.h   | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 8cb3dbcae3e47..d0df3381539f0 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -291,6 +291,40 @@ struct kfd2kgd_calls {
 			uint32_t vmid, uint64_t page_table_base);
 	uint32_t (*read_vmid_from_vmfault_reg)(struct amdgpu_device *adev);
 
+	uint32_t (*enable_debug_trap)(struct amdgpu_device *adev,
+					bool restore_dbg_registers,
+					uint32_t vmid);
+	uint32_t (*disable_debug_trap)(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid);
+	int (*validate_trap_override_request)(struct amdgpu_device *adev,
+					uint32_t trap_override,
+					uint32_t *trap_mask_supported);
+	uint32_t (*set_wave_launch_trap_override)(struct amdgpu_device *adev,
+					     uint32_t vmid,
+					     uint32_t trap_override,
+					     uint32_t trap_mask_bits,
+					     uint32_t trap_mask_request,
+					     uint32_t *trap_mask_prev,
+					     uint32_t kfd_dbg_trap_cntl_prev);
+	uint32_t (*set_wave_launch_mode)(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid);
+	uint32_t (*set_address_watch)(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid);
+	uint32_t (*clear_address_watch)(struct amdgpu_device *adev,
+			uint32_t watch_id);
+	void (*get_iq_wait_times)(struct amdgpu_device *adev,
+			uint32_t *wait_times);
+	void (*build_grace_period_packet_info)(struct amdgpu_device *adev,
+			uint32_t wait_times,
+			uint32_t grace_period,
+			uint32_t *reg_offset,
+			uint32_t *reg_data);
 	void (*get_cu_occupancy)(struct amdgpu_device *adev, int pasid,
 			int *wave_cnt, int *max_waves_per_cu, uint32_t inst);
 	void (*program_trap_handler_settings)(struct amdgpu_device *adev,
-- 
GitLab


From 4504f14338cdc43586189558113faafa8acb9ffe Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 31 Mar 2022 12:05:00 -0400
Subject: [PATCH 1518/1791] drm/amdgpu: setup hw debug registers on driver
 initialization

Add missing debug trap registers references and initialize all debug
registers on boot by clearing the hardware exception overrides and the
wave allocation ID index.

The debugger requires that TTMPs 6 & 7 save the dispatch ID to map
waves onto dispatch during compute context inspection.
In order to correctly set this up, set the special reserved CP bit by
default whenever the MQD is initailized.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        | 26 +++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c        |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         | 30 ++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c       |  3 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |  5 ++
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  |  5 ++
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  5 ++
 .../include/asic_reg/gc/gc_10_1_0_offset.h    | 14 ++++
 .../include/asic_reg/gc/gc_10_1_0_sh_mask.h   | 69 +++++++++++++++++++
 .../include/asic_reg/gc/gc_10_3_0_offset.h    | 10 +++
 .../include/asic_reg/gc/gc_10_3_0_sh_mask.h   |  4 ++
 .../include/asic_reg/gc/gc_11_0_0_sh_mask.h   |  4 ++
 12 files changed, 176 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index f7ad883a70fa2..be984f8c71c7b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4825,6 +4825,29 @@ static u32 gfx_v10_0_init_pa_sc_tile_steering_override(struct amdgpu_device *ade
 
 #define DEFAULT_SH_MEM_BASES	(0x6000)
 
+static void gfx_v10_0_debug_trap_config_init(struct amdgpu_device *adev,
+				uint32_t first_vmid,
+				uint32_t last_vmid)
+{
+	uint32_t data;
+	uint32_t trap_config_vmid_mask = 0;
+	int i;
+
+	/* Calculate trap config vmid mask */
+	for (i = first_vmid; i < last_vmid; i++)
+		trap_config_vmid_mask |= (1 << i);
+
+	data = REG_SET_FIELD(0, SPI_GDBG_TRAP_CONFIG,
+			VMID_SEL, trap_config_vmid_mask);
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG,
+			TRAP_EN, 1);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0);
+}
+
 static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev)
 {
 	int i;
@@ -4856,6 +4879,9 @@ static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev)
 		WREG32_SOC15_OFFSET(GC, 0, mmGDS_GWS_VMID0, i, 0);
 		WREG32_SOC15_OFFSET(GC, 0, mmGDS_OA_VMID0, i, 0);
 	}
+
+	gfx_v10_0_debug_trap_config_init(adev, adev->vm_manager.first_kfd_vmid,
+					AMDGPU_NUM_VMID);
 }
 
 static void gfx_v10_0_init_gds_vmid(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index da21bf868080e..690e121d9ddaa 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1638,6 +1638,7 @@ static void gfx_v11_0_init_compute_vmid(struct amdgpu_device *adev)
 		/* Enable trap for each kfd vmid. */
 		data = RREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL);
 		data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
+		WREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL, data);
 	}
 	soc21_grbm_select(adev, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 0189e50bd89f3..7f17e0061027c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2303,6 +2303,29 @@ static void gfx_v9_0_setup_rb(struct amdgpu_device *adev)
 	adev->gfx.config.num_rbs = hweight32(active_rbs);
 }
 
+static void gfx_v9_0_debug_trap_config_init(struct amdgpu_device *adev,
+				uint32_t first_vmid,
+				uint32_t last_vmid)
+{
+	uint32_t data;
+	uint32_t trap_config_vmid_mask = 0;
+	int i;
+
+	/* Calculate trap config vmid mask */
+	for (i = first_vmid; i < last_vmid; i++)
+		trap_config_vmid_mask |= (1 << i);
+
+	data = REG_SET_FIELD(0, SPI_GDBG_TRAP_CONFIG,
+			VMID_SEL, trap_config_vmid_mask);
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG,
+			TRAP_EN, 1);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0);
+}
+
 #define DEFAULT_SH_MEM_BASES	(0x6000)
 static void gfx_v9_0_init_compute_vmid(struct amdgpu_device *adev)
 {
@@ -4602,6 +4625,13 @@ static int gfx_v9_0_late_init(void *handle)
 	if (r)
 		return r;
 
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2))
+		gfx_v9_4_2_debug_trap_config_init(adev,
+			adev->vm_manager.first_kfd_vmid, AMDGPU_NUM_VMID);
+	else
+		gfx_v9_0_debug_trap_config_init(adev,
+			adev->vm_manager.first_kfd_vmid, AMDGPU_NUM_VMID);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index 2cc3a7cb1f542..63f6843a069e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -771,6 +771,9 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev,
 
 	soc15_grbm_select(adev, 0, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_TRAP_DATA0), 0);
+	WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_TRAP_DATA1), 0);
 }
 
 void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index 772c09b5821b8..eaaa4f4ddaaaf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -117,6 +117,11 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT |
 			1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT;
 
+	/* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the
+	 * DISPATCH_PTR.  This is required for the kfd debugger
+	 */
+	m->cp_hqd_hq_scheduler0 = 1 << 14;
+
 	if (q->format == KFD_QUEUE_FORMAT_AQL) {
 		m->cp_hqd_aql_control =
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index 632344b95d905..3a48bbc589fe2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -143,6 +143,11 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT |
 			1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT;
 
+	/* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the
+	 * DISPATCH_PTR.  This is required for the kfd debugger
+	 */
+	m->cp_hqd_hq_status0 = 1 << 14;
+
 	/*
 	 * GFX11 RS64 CPFW version >= 509 supports PCIe atomics support
 	 * acknowledgment.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 226132ec37145..b7c95158d4a0f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -182,6 +182,11 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 			1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT |
 			1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT;
 
+	/* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the
+	 * DISPATCH_PTR.  This is required for the kfd debugger
+	 */
+	m->cp_hqd_hq_status0 = 1 << 14;
+
 	if (q->format == KFD_QUEUE_FORMAT_AQL)
 		m->cp_hqd_aql_control =
 			1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT;
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h
index 79c41004c0b62..4908044f74090 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h
@@ -5194,6 +5194,20 @@
 #define mmSPI_WCL_PIPE_PERCENT_CS6_BASE_IDX                                                            0
 #define mmSPI_WCL_PIPE_PERCENT_CS7                                                                     0x1f70
 #define mmSPI_WCL_PIPE_PERCENT_CS7_BASE_IDX                                                            0
+#define mmSPI_GDBG_WAVE_CNTL                                                                           0x1f71
+#define mmSPI_GDBG_WAVE_CNTL_BASE_IDX                                                                  0
+#define mmSPI_GDBG_TRAP_CONFIG                                                                         0x1f72
+#define mmSPI_GDBG_TRAP_CONFIG_BASE_IDX                                                                0
+#define mmSPI_GDBG_TRAP_MASK                                                                           0x1f73
+#define mmSPI_GDBG_TRAP_MASK_BASE_IDX                                                                  0
+#define mmSPI_GDBG_WAVE_CNTL2                                                                          0x1f74
+#define mmSPI_GDBG_WAVE_CNTL2_BASE_IDX                                                                 0
+#define mmSPI_GDBG_WAVE_CNTL3                                                                          0x1f75
+#define mmSPI_GDBG_WAVE_CNTL3_BASE_IDX                                                                 0
+#define mmSPI_GDBG_TRAP_DATA0                                                                          0x1f78
+#define mmSPI_GDBG_TRAP_DATA0_BASE_IDX                                                                 0
+#define mmSPI_GDBG_TRAP_DATA1                                                                          0x1f79
+#define mmSPI_GDBG_TRAP_DATA1_BASE_IDX                                                                 0
 #define mmSPI_COMPUTE_QUEUE_RESET                                                                      0x1f7b
 #define mmSPI_COMPUTE_QUEUE_RESET_BASE_IDX                                                             0
 #define mmSPI_RESOURCE_RESERVE_CU_0                                                                    0x1f7c
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h
index 52043e1430678..9b7d219e7954f 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h
@@ -19700,6 +19700,75 @@
 //SPI_WCL_PIPE_PERCENT_CS7
 #define SPI_WCL_PIPE_PERCENT_CS7__VALUE__SHIFT                                                                0x0
 #define SPI_WCL_PIPE_PERCENT_CS7__VALUE_MASK                                                                  0x7FL
+//SPI_GDBG_WAVE_CNTL
+#define SPI_GDBG_WAVE_CNTL__STALL_RA__SHIFT                                                                   0x0
+#define SPI_GDBG_WAVE_CNTL__STALL_VMID__SHIFT                                                                 0x1
+#define SPI_GDBG_WAVE_CNTL__STALL_RA_MASK                                                                     0x00000001L
+#define SPI_GDBG_WAVE_CNTL__STALL_VMID_MASK                                                                   0x0001FFFEL
+//SPI_GDBG_TRAP_CONFIG
+#define SPI_GDBG_TRAP_CONFIG__ME_SEL__SHIFT                                                                   0x0
+#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL__SHIFT                                                                 0x2
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL__SHIFT                                                                0x4
+#define SPI_GDBG_TRAP_CONFIG__ME_MATCH__SHIFT                                                                 0x7
+#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH__SHIFT                                                               0x8
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH__SHIFT                                                              0x9
+#define SPI_GDBG_TRAP_CONFIG__TRAP_EN__SHIFT                                                                  0xf
+#define SPI_GDBG_TRAP_CONFIG__VMID_SEL__SHIFT                                                                 0x10
+#define SPI_GDBG_TRAP_CONFIG__ME_SEL_MASK                                                                     0x00000003L
+#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL_MASK                                                                   0x0000000CL
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL_MASK                                                                  0x00000070L
+#define SPI_GDBG_TRAP_CONFIG__ME_MATCH_MASK                                                                   0x00000080L
+#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH_MASK                                                                 0x00000100L
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH_MASK                                                                0x00000200L
+#define SPI_GDBG_TRAP_CONFIG__TRAP_EN_MASK                                                                    0x00008000L
+#define SPI_GDBG_TRAP_CONFIG__VMID_SEL_MASK                                                                   0xFFFF0000L
+//SPI_GDBG_TRAP_MASK
+#define SPI_GDBG_TRAP_MASK__EXCP_EN__SHIFT                                                                    0x0
+#define SPI_GDBG_TRAP_MASK__REPLACE__SHIFT                                                                    0x9
+#define SPI_GDBG_TRAP_MASK__EXCP_EN_MASK                                                                      0x01FFL
+#define SPI_GDBG_TRAP_MASK__REPLACE_MASK                                                                      0x0200L
+//SPI_GDBG_WAVE_CNTL2
+#define SPI_GDBG_WAVE_CNTL2__VMID_MASK__SHIFT                                                                 0x0
+#define SPI_GDBG_WAVE_CNTL2__MODE__SHIFT                                                                      0x10
+#define SPI_GDBG_WAVE_CNTL2__VMID_MASK_MASK                                                                   0x0000FFFFL
+#define SPI_GDBG_WAVE_CNTL2__MODE_MASK                                                                        0x00030000L
+//SPI_GDBG_WAVE_CNTL3
+#define SPI_GDBG_WAVE_CNTL3__STALL_PS__SHIFT                                                                  0x0
+#define SPI_GDBG_WAVE_CNTL3__STALL_VS__SHIFT                                                                  0x1
+#define SPI_GDBG_WAVE_CNTL3__STALL_GS__SHIFT                                                                  0x2
+#define SPI_GDBG_WAVE_CNTL3__STALL_HS__SHIFT                                                                  0x3
+#define SPI_GDBG_WAVE_CNTL3__STALL_CSG__SHIFT                                                                 0x4
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS0__SHIFT                                                                 0x5
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS1__SHIFT                                                                 0x6
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS2__SHIFT                                                                 0x7
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS3__SHIFT                                                                 0x8
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS4__SHIFT                                                                 0x9
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS5__SHIFT                                                                 0xa
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS6__SHIFT                                                                 0xb
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS7__SHIFT                                                                 0xc
+#define SPI_GDBG_WAVE_CNTL3__STALL_DURATION__SHIFT                                                            0xd
+#define SPI_GDBG_WAVE_CNTL3__STALL_MULT__SHIFT                                                                0x1c
+#define SPI_GDBG_WAVE_CNTL3__STALL_PS_MASK                                                                    0x00000001L
+#define SPI_GDBG_WAVE_CNTL3__STALL_VS_MASK                                                                    0x00000002L
+#define SPI_GDBG_WAVE_CNTL3__STALL_GS_MASK                                                                    0x00000004L
+#define SPI_GDBG_WAVE_CNTL3__STALL_HS_MASK                                                                    0x00000008L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CSG_MASK                                                                   0x00000010L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS0_MASK                                                                   0x00000020L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS1_MASK                                                                   0x00000040L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS2_MASK                                                                   0x00000080L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS3_MASK                                                                   0x00000100L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS4_MASK                                                                   0x00000200L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS5_MASK                                                                   0x00000400L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS6_MASK                                                                   0x00000800L
+#define SPI_GDBG_WAVE_CNTL3__STALL_CS7_MASK                                                                   0x00001000L
+#define SPI_GDBG_WAVE_CNTL3__STALL_DURATION_MASK                                                              0x0FFFE000L
+#define SPI_GDBG_WAVE_CNTL3__STALL_MULT_MASK                                                                  0x10000000L
+//SPI_GDBG_TRAP_DATA0
+#define SPI_GDBG_TRAP_DATA0__DATA__SHIFT                                                                      0x0
+#define SPI_GDBG_TRAP_DATA0__DATA_MASK                                                                        0xFFFFFFFFL
+//SPI_GDBG_TRAP_DATA1
+#define SPI_GDBG_TRAP_DATA1__DATA__SHIFT                                                                      0x0
+#define SPI_GDBG_TRAP_DATA1__DATA_MASK                                                                        0xFFFFFFFFL
 //SPI_COMPUTE_QUEUE_RESET
 #define SPI_COMPUTE_QUEUE_RESET__RESET__SHIFT                                                                 0x0
 #define SPI_COMPUTE_QUEUE_RESET__RESET_MASK                                                                   0x01L
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
index a734abaa91a59..5e15ac14b63c8 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h
@@ -26,6 +26,8 @@
 #define mmSQ_DEBUG_STS_GLOBAL_BASE_IDX                                                                 0
 #define mmSQ_DEBUG_STS_GLOBAL2                                                                         0x10B0
 #define mmSQ_DEBUG_STS_GLOBAL2_BASE_IDX                                                                0
+#define mmSQ_DEBUG                                                                                     0x10B1
+#define mmSQ_DEBUG_BASE_IDX                                                                            0
 
 // addressBlock: gc_sdma0_sdma0dec
 // base address: 0x4980
@@ -4853,10 +4855,18 @@
 #define mmSPI_WCL_PIPE_PERCENT_CS3_BASE_IDX                                                            0
 #define mmSPI_GDBG_WAVE_CNTL                                                                           0x1f71
 #define mmSPI_GDBG_WAVE_CNTL_BASE_IDX                                                                  0
+#define mmSPI_GDBG_TRAP_CONFIG                                                                         0x1f72
+#define mmSPI_GDBG_TRAP_CONFIG_BASE_IDX                                                                0
 #define mmSPI_GDBG_TRAP_MASK                                                                           0x1f73
 #define mmSPI_GDBG_TRAP_MASK_BASE_IDX                                                                  0
 #define mmSPI_GDBG_WAVE_CNTL2                                                                          0x1f74
 #define mmSPI_GDBG_WAVE_CNTL2_BASE_IDX                                                                 0
+#define mmSPI_GDBG_WAVE_CNTL3                                                                          0x1f75
+#define mmSPI_GDBG_WAVE_CNTL3_BASE_IDX                                                                 0
+#define mmSPI_GDBG_TRAP_DATA0                                                                          0x1f78
+#define mmSPI_GDBG_TRAP_DATA0_BASE_IDX                                                                 0
+#define mmSPI_GDBG_TRAP_DATA1                                                                          0x1f79
+#define mmSPI_GDBG_TRAP_DATA1_BASE_IDX                                                                 0
 #define mmSPI_COMPUTE_QUEUE_RESET                                                                      0x1f7b
 #define mmSPI_COMPUTE_QUEUE_RESET_BASE_IDX                                                             0
 #define mmSPI_RESOURCE_RESERVE_CU_0                                                                    0x1f7c
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
index d7a17bae25846..e4ecd6c2d20e2 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h
@@ -47907,6 +47907,10 @@
 
 
 // addressBlock: sqind
+//SQ_DEBUG
+#define SQ_DEBUG__SINGLE_MEMOP_MASK 0x00000001L
+#define SQ_DEBUG__SINGLE_MEMOP__SHIFT 0x00000000
+
 //SQ_DEBUG_STS_GLOBAL
 #define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0_MASK 0x000000ffL
 #define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0__SHIFT 0x00000000
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h
index 4f08f90856fc7..3088a4a13cb50 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h
@@ -17216,11 +17216,15 @@
 #define SPI_GDBG_PER_VMID_CNTL__TRAP_EN__SHIFT                                                                0x3
 #define SPI_GDBG_PER_VMID_CNTL__EXCP_EN__SHIFT                                                                0x4
 #define SPI_GDBG_PER_VMID_CNTL__EXCP_REPLACE__SHIFT                                                           0xd
+#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_START__SHIFT                                                          0xe
+#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_END__SHIFT                                                            0xf
 #define SPI_GDBG_PER_VMID_CNTL__STALL_VMID_MASK                                                               0x00000001L
 #define SPI_GDBG_PER_VMID_CNTL__LAUNCH_MODE_MASK                                                              0x00000006L
 #define SPI_GDBG_PER_VMID_CNTL__TRAP_EN_MASK                                                                  0x00000008L
 #define SPI_GDBG_PER_VMID_CNTL__EXCP_EN_MASK                                                                  0x00001FF0L
 #define SPI_GDBG_PER_VMID_CNTL__EXCP_REPLACE_MASK                                                             0x00002000L
+#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_START_MASK                                                            0x00004000L
+#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_END_MASK                                                              0x00008000L
 //SPI_COMPUTE_QUEUE_RESET
 #define SPI_COMPUTE_QUEUE_RESET__RESET__SHIFT                                                                 0x0
 #define SPI_COMPUTE_QUEUE_RESET__RESET_MASK                                                                   0x01L
-- 
GitLab


From 257d7b7be26d83768cb07585480d90e875365d5c Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 30 May 2023 12:44:30 -0500
Subject: [PATCH 1519/1791] drm/amd: Make lack of `ACPI_FADT_LOW_POWER_S0` or
 `CONFIG_AMD_PMC` louder during suspend path

Users have reported that s2idle wasn't working on OEM Phoenix systems,
but it was root caused to be because `CONFIG_AMD_PMC` wasn't set in
the distribution kernel config.

To make this more apparent, raise the messaging to err instead of warn.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=217497
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index 73e4434d9b546..385c6acb5728b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1500,14 +1500,14 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev)
 	 * in that case.
 	 */
 	if (!(acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0)) {
-		dev_warn_once(adev->dev,
+		dev_err_once(adev->dev,
 			      "Power consumption will be higher as BIOS has not been configured for suspend-to-idle.\n"
 			      "To use suspend-to-idle change the sleep mode in BIOS setup.\n");
 		return false;
 	}
 
 #if !IS_ENABLED(CONFIG_AMD_PMC)
-	dev_warn_once(adev->dev,
+	dev_err_once(adev->dev,
 		      "Power consumption will be higher as the kernel has not been compiled with CONFIG_AMD_PMC.\n");
 	return false;
 #else
-- 
GitLab


From 34941e5dc72daff878267c55a9bdb5e055804953 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 31 May 2023 09:03:27 +0530
Subject: [PATCH 1520/1791] drm/amd/display: Drop unused DCN_BASE variable in
 dcn314_resource.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following W=1 kernel build warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn314/dcn314_resource.c:128:29: warning: ‘DCN_BASE’ defined but not used [-Wunused-const-variable=]
  128 | static const struct IP_BASE DCN_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C0, 0x00009000, 0x02403C00, 0, 0, 0 } },
      |                             ^~~~~~~~

Suggested-by: Roman Li <Roman.Li@amd.com>
Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Roman Li <Roman.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn314/dcn314_resource.c | 17 -----------------
 1 file changed, 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 3592efcc7fae0..837884c4f03a6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -117,23 +117,6 @@
 #define regBIF_BX2_BIOS_SCRATCH_6			0x003e
 #define regBIF_BX2_BIOS_SCRATCH_6_BASE_IDX		1
 
-struct IP_BASE_INSTANCE {
-	unsigned int segment[MAX_SEGMENT];
-};
-
-struct IP_BASE {
-	struct IP_BASE_INSTANCE instance[MAX_INSTANCE];
-};
-
-static const struct IP_BASE DCN_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C0, 0x00009000, 0x02403C00, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } },
-					{ { 0, 0, 0, 0, 0, 0, 0, 0 } } } };
-
-
 #define DC_LOGGER_INIT(logger)
 
 enum dcn31_clk_src_array_id {
-- 
GitLab


From d3116d9f27b89d363dd528e42fcf4895a15e0c3c Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Wed, 31 May 2023 10:08:11 +0800
Subject: [PATCH 1521/1791] drm/amdkfd: clean up one inconsistent indenting

drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_device.c:1036 kgd2kfd_interrupt() warn: inconsistent indenting

Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index e84ad1c5ef446..f0ed6e6416c37 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -1042,7 +1042,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 			    	is_patched ? patched_ihre : ih_ring_entry)) {
 			kfd_queue_work(node->ih_wq, &node->interrupt_work);
 			spin_unlock_irqrestore(&node->interrupt_lock, flags);
-				return;
+			return;
 		}
 		spin_unlock_irqrestore(&node->interrupt_lock, flags);
 	}
-- 
GitLab


From cde2e087a320bff5d772d82c9fbddaea18daa94a Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 30 Mar 2022 15:09:11 -0400
Subject: [PATCH 1522/1791] drm/amdgpu: add gfx9 hw debug mode enable and
 disable calls

Implement the per-device calls to enable or disable HW debug mode for
GFX9 prior to GFX9.4.1.

GFX9.4.1 and onward will require their own enable/disable sequence as
follow on patches.

When hardware debug mode setting is requested, waves will inherit
these settings in the Shader Processor Input's (SPI) Sequencer Global
Block (SQG). This means that the KGD must drain all waves from the SPI
into SQG (approximately 96 SPI clock cycles) prior to debug mode setting
to ensure that the order of operations that the debugger expects with
regards to debug mode setting transaction requests and wave inheritence
of that mode is upheld.

Also ensure that exception overrides are reset to their original state
prior to debug enable or disable.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 92 +++++++++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |  9 ++
 2 files changed, 101 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 34bf030f31377..8d7d04704b00f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -649,6 +649,96 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev,
 	return 0;
 }
 
+/*
+ * GFX9 helper for wave launch stall requirements on debug trap setting.
+ *
+ * vmid:
+ *   Target VMID to stall/unstall.
+ *
+ * stall:
+ *   0-unstall wave launch (enable), 1-stall wave launch (disable).
+ *   After wavefront launch has been stalled, allocated waves must drain from
+ *   SPI in order for debug trap settings to take effect on those waves.
+ *   This is roughly a ~96 clock cycle wait on SPI where a read on
+ *   SPI_GDBG_WAVE_CNTL translates to ~32 clock cycles.
+ *   KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY indicates the number of reads required.
+ *
+ *   NOTE: We can afford to clear the entire STALL_VMID field on unstall
+ *   because GFX9.4.1 cannot support multi-process debugging due to trap
+ *   configuration and masking being limited to global scope.  Always assume
+ *   single process conditions.
+ */
+#define KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY	3
+void kgd_gfx_v9_set_wave_launch_stall(struct amdgpu_device *adev,
+					uint32_t vmid,
+					bool stall)
+{
+	int i;
+	uint32_t data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 1))
+		data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_VMID,
+							stall ? 1 << vmid : 0);
+	else
+		data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA,
+							stall ? 1 : 0);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
+
+	if (!stall)
+		return;
+
+	for (i = 0; i < KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY; i++)
+		RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+}
+
+/*
+ * restore_dbg_registers is ignored here but is a general interface requirement
+ * for devices that support GFXOFF and where the RLC save/restore list
+ * does not support hw registers for debugging i.e. the driver has to manually
+ * initialize the debug mode registers after it has disabled GFX off during the
+ * debug session.
+ */
+uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev,
+				bool restore_dbg_registers,
+				uint32_t vmid)
+{
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
+/*
+ * keep_trap_enabled is ignored here but is a general interface requirement
+ * for devices that support multi-process debugging where the performance
+ * overhead from trap temporary setup needs to be bypassed when the debug
+ * session has ended.
+ */
+uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid)
+{
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev,
 			uint32_t vmid, uint64_t page_table_base)
 {
@@ -877,6 +967,8 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info =
 			kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
+	.enable_debug_trap = kgd_gfx_v9_enable_debug_trap,
+	.disable_debug_trap = kgd_gfx_v9_disable_debug_trap,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index a241299f4fbcc..9588ff055393d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -64,3 +64,12 @@ void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id,
 uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev,
 				uint32_t pipe_id, uint32_t queue_id);
 void kgd_gfx_v9_release_queue(struct amdgpu_device *adev, uint32_t inst);
+void kgd_gfx_v9_set_wave_launch_stall(struct amdgpu_device *adev,
+					uint32_t vmid,
+					bool stall);
+uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev,
+				      bool restore_dbg_registers,
+				      uint32_t vmid);
+uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid);
-- 
GitLab


From 01f648202c5390f4c366793b34c27cddad4ca8d7 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 30 Mar 2022 15:31:00 -0400
Subject: [PATCH 1523/1791] drm/amdgpu: add gfx9.4.1 hw debug mode enable and
 disable calls

On GFX9.4.1, the implicit wait count instruction on s_barrier is
disabled by default in the driver during normal operation for
performance requirements.

There is a hardware bug in GFX9.4.1 where if the implicit wait count
instruction after an s_barrier instruction is disabled, any wave that
hits an exception may step over the s_barrier when returning from the
trap handler with the barrier logic having no ability to be
aware of this, thereby causing other waves to wait at the barrier
indefinitely resulting in a shader hang.  This bug has been corrected
for GFX9.4.2 and onward.

Since the debugger subscribes to hardware exceptions, in order to avoid
this bug, the debugger must enable implicit wait count on s_barrier
for a debug session and disable it on detach.

In order to change this setting in the in the device global SQ_CONFIG
register, the GFX pipeline must be idle.  GFX9.4.1 as a compute device
will either dispatch work through the compute ring buffers used for
image post processing or through the hardware scheduler by the KFD.

Have the KGD suspend and drain the compute ring buffer, then suspend the
hardware scheduler and block any future KFD process job requests before
changing the implicit wait count setting.  Once set, resume all work.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h           |   3 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c   | 116 ++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         |   4 +-
 3 files changed, 121 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 8b4a89a6c3fdd..74104d46a7bea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1046,6 +1046,9 @@ struct amdgpu_device {
 	struct pci_saved_state          *pci_state;
 	pci_channel_state_t		pci_channel_state;
 
+	/* Track auto wait count on s_barrier settings */
+	bool				barrier_has_auto_waitcnt;
+
 	struct amdgpu_reset_control     *reset_cntl;
 	uint32_t                        ip_versions[MAX_HWIP][HWIP_MAX_INSTANCE];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 4191af5a3f132..d2918e5c0dea3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -26,6 +26,7 @@
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_amdkfd_arcturus.h"
+#include "amdgpu_reset.h"
 #include "sdma0/sdma0_4_2_2_offset.h"
 #include "sdma0/sdma0_4_2_2_sh_mask.h"
 #include "sdma1/sdma1_4_2_2_offset.h"
@@ -48,6 +49,8 @@
 #include "amdgpu_amdkfd_gfx_v9.h"
 #include "gfxhub_v1_0.h"
 #include "mmhub_v9_4.h"
+#include "gc/gc_9_0_offset.h"
+#include "gc/gc_9_0_sh_mask.h"
 
 #define HQD_N_REGS 56
 #define DUMP_REG(addr) do {				\
@@ -276,6 +279,117 @@ int kgd_arcturus_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd,
 	return 0;
 }
 
+/*
+ * Helper used to suspend/resume gfx pipe for image post process work to set
+ * barrier behaviour.
+ */
+static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool suspend)
+{
+	int i, r = 0;
+
+	for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+		struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
+
+		if (!(ring && ring->sched.thread))
+			continue;
+
+		/* stop secheduler and drain ring. */
+		if (suspend) {
+			drm_sched_stop(&ring->sched, NULL);
+			r = amdgpu_fence_wait_empty(ring);
+			if (r)
+				goto out;
+		} else {
+			drm_sched_start(&ring->sched, false);
+		}
+	}
+
+out:
+	/* return on resume or failure to drain rings. */
+	if (!suspend || r)
+		return r;
+
+	return amdgpu_device_ip_wait_for_idle(adev, GC_HWIP);
+}
+
+static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_waitcnt)
+{
+	uint32_t data;
+
+	WRITE_ONCE(adev->barrier_has_auto_waitcnt, enable_waitcnt);
+
+	if (!down_read_trylock(&adev->reset_domain->sem))
+		return;
+
+	amdgpu_amdkfd_suspend(adev, false);
+
+	if (suspend_resume_compute_scheduler(adev, true))
+		goto out;
+
+	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_CONFIG));
+	data = REG_SET_FIELD(data, SQ_CONFIG, DISABLE_BARRIER_WAITCNT,
+						!enable_waitcnt);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_CONFIG), data);
+
+out:
+	suspend_resume_compute_scheduler(adev, false);
+
+	amdgpu_amdkfd_resume(adev, false);
+
+	up_read(&adev->reset_domain->sem);
+}
+
+/*
+ * restore_dbg_registers is ignored here but is a general interface requirement
+ * for devices that support GFXOFF and where the RLC save/restore list
+ * does not support hw registers for debugging i.e. the driver has to manually
+ * initialize the debug mode registers after it has disabled GFX off during the
+ * debug session.
+ */
+static uint32_t kgd_arcturus_enable_debug_trap(struct amdgpu_device *adev,
+				bool restore_dbg_registers,
+				uint32_t vmid)
+{
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	set_barrier_auto_waitcnt(adev, true);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
+/*
+ * keep_trap_enabled is ignored here but is a general interface requirement
+ * for devices that support multi-process debugging where the performance
+ * overhead from trap temporary setup needs to be bypassed when the debug
+ * session has ended.
+ */
+static uint32_t kgd_arcturus_disable_debug_trap(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid)
+{
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	set_barrier_auto_waitcnt(adev, false);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
 const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -294,6 +408,8 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 				kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base =
 				kgd_gfx_v9_set_vm_context_page_table_base,
+	.enable_debug_trap = kgd_arcturus_enable_debug_trap,
+	.disable_debug_trap = kgd_arcturus_disable_debug_trap,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 7f17e0061027c..f092a1dbdb568 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2390,8 +2390,8 @@ static void gfx_v9_0_init_sq_config(struct amdgpu_device *adev)
 	switch (adev->ip_versions[GC_HWIP][0]) {
 	case IP_VERSION(9, 4, 1):
 		tmp = RREG32_SOC15(GC, 0, mmSQ_CONFIG);
-		tmp = REG_SET_FIELD(tmp, SQ_CONFIG,
-					DISABLE_BARRIER_WAITCNT, 1);
+		tmp = REG_SET_FIELD(tmp, SQ_CONFIG, DISABLE_BARRIER_WAITCNT,
+				!READ_ONCE(adev->barrier_has_auto_waitcnt));
 		WREG32_SOC15(GC, 0, mmSQ_CONFIG, tmp);
 		break;
 	default:
-- 
GitLab


From bb13d763f251c28b08d996671c5146a2113fc9e7 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 24 Mar 2023 16:19:27 -0400
Subject: [PATCH 1524/1791] drm/amdkfd: fix kfd_suspend_all_processes

Flush delayed restore work in kfd_suspend_all_queues instead of
cancelling. Cancelling the work before it runs results in the queues
becoming permanently disabled. Flushing the work ensures that the
queue suspend/resume state stays balanced.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index ef67f5e37c6f4..d75dac92775c7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -2014,7 +2014,7 @@ void kfd_suspend_all_processes(void)
 	WARN(debug_evictions, "Evicting all processes");
 	hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
 		cancel_delayed_work_sync(&p->eviction_work);
-		cancel_delayed_work_sync(&p->restore_work);
+		flush_delayed_work(&p->restore_work);
 
 		if (kfd_process_evict_queues(p, KFD_QUEUE_EVICTION_TRIGGER_SUSPEND))
 			pr_err("Failed to suspend process 0x%x\n", p->pasid);
-- 
GitLab


From d13f050fee94a454323f864fb005c4355600cdbd Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 31 Mar 2022 13:14:01 -0400
Subject: [PATCH 1525/1791] drm/amdgpu: add gfx10 hw debug mode enable and
 disable calls

Similar to GFX9 debug devices, set the hardware debug mode by draining
the SPI appropriately prior the mode setting request.

Because GFX10 has waves allocated by the work group boundary and each
SE's SPI instances do not communicate, the SPI drain time is much longer.
This long drain time will be fixed for GFX11 onwards.

Also remove a bunch of deprecated misplaced references for GFX10.3.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    |  96 ++++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h    |  28 ++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  | 148 +-----------------
 3 files changed, 127 insertions(+), 145 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 7b60268d93c0c..240f5006e278d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -21,6 +21,7 @@
  */
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
+#include "amdgpu_amdkfd_gfx_v10.h"
 #include "gc/gc_10_1_0_offset.h"
 #include "gc/gc_10_1_0_sh_mask.h"
 #include "athub/athub_2_0_0_offset.h"
@@ -709,6 +710,99 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev,
 	adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
 }
 
+/*
+ * GFX10 helper for wave launch stall requirements on debug trap setting.
+ *
+ * vmid:
+ *   Target VMID to stall/unstall.
+ *
+ * stall:
+ *   0-unstall wave launch (enable), 1-stall wave launch (disable).
+ *   After wavefront launch has been stalled, allocated waves must drain from
+ *   SPI in order for debug trap settings to take effect on those waves.
+ *   This is roughly a ~3500 clock cycle wait on SPI where a read on
+ *   SPI_GDBG_WAVE_CNTL translates to ~32 clock cycles.
+ *   KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY indicates the number of reads required.
+ *
+ *   NOTE: We can afford to clear the entire STALL_VMID field on unstall
+ *   because current GFX10 chips cannot support multi-process debugging due to
+ *   trap configuration and masking being limited to global scope.  Always
+ *   assume single process conditions.
+ *
+ */
+
+#define KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY	110
+static void kgd_gfx_v10_set_wave_launch_stall(struct amdgpu_device *adev, uint32_t vmid, bool stall)
+{
+	uint32_t data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+	int i;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_VMID,
+							stall ? 1 << vmid : 0);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
+
+	if (!stall)
+		return;
+
+	for (i = 0; i < KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY; i++)
+		RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+}
+
+uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev,
+				bool restore_dbg_registers,
+				uint32_t vmid)
+{
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true);
+
+	/* assume gfx off is disabled for the debug session if rlc restore not supported. */
+	if (restore_dbg_registers) {
+		uint32_t data = 0;
+
+		data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG,
+				VMID_SEL, 1 << vmid);
+		data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG,
+				TRAP_EN, 1);
+		WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data);
+		WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0);
+		WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0);
+
+		kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false);
+
+		mutex_unlock(&adev->grbm_idx_mutex);
+
+		return 0;
+	}
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
+uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid)
+{
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true);
+
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 static void program_trap_handler_settings(struct amdgpu_device *adev,
 		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
 		uint32_t inst)
@@ -752,5 +846,7 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info =
 			get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
+	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
+	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
 	.program_trap_handler_settings = program_trap_handler_settings,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
new file mode 100644
index 0000000000000..251d61fbde079
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev,
+				      bool restore_dbg_registers,
+				      uint32_t vmid);
+uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
+					bool keep_trap_enabled,
+					uint32_t vmid);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 52d0d35fb58db..8b293f3dcbd2b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -22,6 +22,7 @@
 #include <linux/mmu_context.h>
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
+#include "amdgpu_amdkfd_gfx_v10.h"
 #include "gc/gc_10_3_0_offset.h"
 #include "gc/gc_10_3_0_sh_mask.h"
 #include "oss/osssys_5_0_0_offset.h"
@@ -654,143 +655,6 @@ static void program_trap_handler_settings_v10_3(struct amdgpu_device *adev,
 	unlock_srbm(adev);
 }
 
-#if 0
-uint32_t enable_debug_trap_v10_3(struct amdgpu_device *adev,
-				uint32_t trap_debug_wave_launch_mode,
-				uint32_t vmid)
-{
-	uint32_t data = 0;
-	uint32_t orig_wave_cntl_value;
-	uint32_t orig_stall_vmid;
-
-	mutex_lock(&adev->grbm_idx_mutex);
-
-	orig_wave_cntl_value = RREG32(SOC15_REG_OFFSET(GC,
-				0,
-				mmSPI_GDBG_WAVE_CNTL));
-	orig_stall_vmid = REG_GET_FIELD(orig_wave_cntl_value,
-			SPI_GDBG_WAVE_CNTL,
-			STALL_VMID);
-
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 1);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
-
-	data = 0;
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data);
-
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), orig_stall_vmid);
-
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return 0;
-}
-
-uint32_t disable_debug_trap_v10_3(struct amdgpu_device *adev)
-{
-	mutex_lock(&adev->grbm_idx_mutex);
-
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0);
-
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return 0;
-}
-
-uint32_t set_wave_launch_trap_override_v10_3(struct amdgpu_device *adev,
-						uint32_t trap_override,
-						uint32_t trap_mask)
-{
-	uint32_t data = 0;
-
-	mutex_lock(&adev->grbm_idx_mutex);
-
-	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 1);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
-
-	data = 0;
-	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK,
-			EXCP_EN, trap_mask);
-	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK,
-			REPLACE, trap_override);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data);
-
-	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 0);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
-
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return 0;
-}
-
-uint32_t set_wave_launch_mode_v10_3(struct amdgpu_device *adev,
-					uint8_t wave_launch_mode,
-					uint32_t vmid)
-{
-	uint32_t data = 0;
-	bool is_stall_mode;
-	bool is_mode_set;
-
-	is_stall_mode = (wave_launch_mode == 4);
-	is_mode_set = (wave_launch_mode != 0 && wave_launch_mode != 4);
-
-	mutex_lock(&adev->grbm_idx_mutex);
-
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
-			VMID_MASK, is_mode_set ? 1 << vmid : 0);
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
-			MODE, is_mode_set ? wave_launch_mode : 0);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data);
-
-	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL,
-			STALL_VMID, is_stall_mode ? 1 << vmid : 0);
-	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL,
-			STALL_RA, is_stall_mode ? 1 : 0);
-	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data);
-
-	mutex_unlock(&adev->grbm_idx_mutex);
-
-	return 0;
-}
-
-/* kgd_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
- * The values read are:
- *	ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
- *	atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads.
- *	wrm_offload_wait_time    -- Wait Count for WAIT_REG_MEM Offloads.
- *	gws_wait_time            -- Wait Count for Global Wave Syncs.
- *	que_sleep_wait_time      -- Wait Count for Dequeue Retry.
- *	sch_wave_wait_time       -- Wait Count for Scheduling Wave Message.
- *	sem_rearm_wait_time      -- Wait Count for Semaphore re-arm.
- *	deq_retry_wait_time      -- Wait Count for Global Wave Syncs.
- */
-void get_iq_wait_times_v10_3(struct amdgpu_device *adev,
-					uint32_t *wait_times, uint32_t inst)
-
-{
-	*wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2));
-}
-
-void build_grace_period_packet_info_v10_3(struct amdgpu_device *adev,
-						uint32_t wait_times,
-						uint32_t grace_period,
-						uint32_t *reg_offset,
-						uint32_t *reg_data,
-						uint32_t inst)
-{
-	*reg_data = wait_times;
-
-	*reg_data = REG_SET_FIELD(*reg_data,
-			CP_IQ_WAIT_TIME2,
-			SCH_WAVE,
-			grace_period);
-
-	*reg_offset = mmCP_IQ_WAIT_TIME2;
-}
-#endif
-
 const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v10_3,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v10_3,
@@ -808,12 +672,6 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info_v10_3,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3,
 	.program_trap_handler_settings = program_trap_handler_settings_v10_3,
-#if 0
-	.enable_debug_trap = enable_debug_trap_v10_3,
-	.disable_debug_trap = disable_debug_trap_v10_3,
-	.set_wave_launch_trap_override = set_wave_launch_trap_override_v10_3,
-	.set_wave_launch_mode = set_wave_launch_mode_v10_3,
-	.get_iq_wait_times = get_iq_wait_times_v10_3,
-	.build_grace_period_packet_info = build_grace_period_packet_info_v10_3,
-#endif
+	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
+	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap
 };
-- 
GitLab


From be6f94039e1a91df4b0efc2da9167ca9c17bb532 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 1 Apr 2022 13:31:57 -0400
Subject: [PATCH 1526/1791] drm/amdgpu: add gfx9.4.2 hw debug mode enable and
 disable calls

GFX9.4.2 now supports per-VMID debug mode controls registers
(SPI_GDBG_PER_VMID_CNTL).

Because the KFD lets the HWS handle PASID-VMID mapping, the KFD will
forward all debug mode setting register writes to the HWS scheduler
using a new MAP_PROCESS API, so instead of writing to registers, return
the required register values that the HWS needs to write on debug enable
and disable.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c  | 42 ++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index 4485bb29bec96..a6f98141c29c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -23,6 +23,44 @@
 #include "amdgpu_amdkfd.h"
 #include "amdgpu_amdkfd_arcturus.h"
 #include "amdgpu_amdkfd_gfx_v9.h"
+#include "gc/gc_9_4_2_offset.h"
+#include "gc/gc_9_4_2_sh_mask.h"
+
+/*
+ * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE.
+ *
+ * restore_dbg_registers is ignored here but is a general interface requirement
+ * for devices that support GFXOFF and where the RLC save/restore list
+ * does not support hw registers for debugging i.e. the driver has to manually
+ * initialize the debug mode registers after it has disabled GFX off during the
+ * debug session.
+ */
+static uint32_t kgd_aldebaran_enable_debug_trap(struct amdgpu_device *adev,
+					    bool restore_dbg_registers,
+					    uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0);
+
+	return data;
+}
+
+/* returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */
+static uint32_t kgd_aldebaran_disable_debug_trap(struct amdgpu_device *adev,
+						bool keep_trap_enabled,
+						uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0);
+
+	return data;
+}
 
 const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
@@ -42,5 +80,7 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 				kgd_gfx_v9_get_atc_vmid_pasid_mapping_info,
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
-	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings
+	.enable_debug_trap = kgd_aldebaran_enable_debug_trap,
+	.disable_debug_trap = kgd_aldebaran_disable_debug_trap,
+	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
 };
-- 
GitLab


From 33f3437ae1194ef5dedbf275dcf74ed9c114647d Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 26 Aug 2022 22:35:50 -0400
Subject: [PATCH 1527/1791] drm/amdgpu: add gfx11 hw debug mode enable and
 disable calls

Implement the per-device calls to enable or disable HW debug mode
for GFX11.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    | 38 +++++++++++++++++++
 1 file changed, 38 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 7deff8a547fb1..cc954cf248ca8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -607,6 +607,42 @@ static void set_vm_context_page_table_base_v11(struct amdgpu_device *adev,
 	adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base);
 }
 
+/*
+ * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE.
+ *
+ * restore_dbg_registers is ignored here but is a general interface requirement
+ * for devices that support GFXOFF and where the RLC save/restore list
+ * does not support hw registers for debugging i.e. the driver has to manually
+ * initialize the debug mode registers after it has disabled GFX off during the
+ * debug session.
+ */
+static uint32_t kgd_gfx_v11_enable_debug_trap(struct amdgpu_device *adev,
+					    bool restore_dbg_registers,
+					    uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0);
+
+	return data;
+}
+
+/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */
+static uint32_t kgd_gfx_v11_disable_debug_trap(struct amdgpu_device *adev,
+						bool keep_trap_enabled,
+						uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0);
+
+	return data;
+}
+
 const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v11,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11,
@@ -623,4 +659,6 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.wave_control_execute = wave_control_execute_v11,
 	.get_atc_vmid_pasid_mapping_info = NULL,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base_v11,
+	.enable_debug_trap = kgd_gfx_v11_enable_debug_trap,
+	.disable_debug_trap = kgd_gfx_v11_disable_debug_trap
 };
-- 
GitLab


From 7cee6a6824a0429a6255abe91b5af01b9a01cd03 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 23 Mar 2023 17:17:20 -0400
Subject: [PATCH 1528/1791] drm/amdgpu: add configurable grace period for unmap
 queues

The HWS schedule allows a grace period for wave completion prior to
preemption for better performance by avoiding CWSR on waves that can
potentially complete quickly. The debugger, on the other hand, will
want to inspect wave status immediately after it actively triggers
preemption (a suspend function to be provided).

To minimize latency between preemption and debugger wave inspection, allow
immediate preemption by setting the grace period to 0.

Note that setting the preepmtion grace period to 0 will result in an
infinite grace period being set due to a CP FW bug so set it to 1 for now.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c  |  2 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c   |  2 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    | 43 ++++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h    |  6 ++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |  2 +
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 43 ++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |  8 ++-
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 63 +++++++++++++-----
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |  3 +
 .../gpu/drm/amd/amdkfd/kfd_packet_manager.c   | 32 +++++++++
 .../drm/amd/amdkfd/kfd_packet_manager_v9.c    | 39 +++++++++++
 .../drm/amd/amdkfd/kfd_packet_manager_vi.c    |  2 +
 .../gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h   | 65 +++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  5 ++
 14 files changed, 295 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index a6f98141c29c7..b811a09850503 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -82,5 +82,7 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.enable_debug_trap = kgd_aldebaran_enable_debug_trap,
 	.disable_debug_trap = kgd_aldebaran_disable_debug_trap,
+	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
+	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index d2918e5c0dea3..a62bd0068515c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -410,6 +410,8 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 				kgd_gfx_v9_set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_arcturus_enable_debug_trap,
 	.disable_debug_trap = kgd_arcturus_disable_debug_trap,
+	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
+	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 240f5006e278d..98006c7021ddf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -803,6 +803,47 @@ uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
 	return 0;
 }
 
+/* kgd_gfx_v10_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
+ * The values read are:
+ *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
+ *     atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads.
+ *     wrm_offload_wait_time    -- Wait Count for WAIT_REG_MEM Offloads.
+ *     gws_wait_time            -- Wait Count for Global Wave Syncs.
+ *     que_sleep_wait_time      -- Wait Count for Dequeue Retry.
+ *     sch_wave_wait_time       -- Wait Count for Scheduling Wave Message.
+ *     sem_rearm_wait_time      -- Wait Count for Semaphore re-arm.
+ *     deq_retry_wait_time      -- Wait Count for Global Wave Syncs.
+ */
+void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev,
+					uint32_t *wait_times)
+
+{
+	*wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2));
+}
+
+void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
+						uint32_t wait_times,
+						uint32_t grace_period,
+						uint32_t *reg_offset,
+						uint32_t *reg_data)
+{
+	*reg_data = wait_times;
+
+	/*
+	 * The CP cannont handle a 0 grace period input and will result in
+	 * an infinite grace period being set so set to 1 to prevent this.
+	 */
+	if (grace_period == 0)
+		grace_period = 1;
+
+	*reg_data = REG_SET_FIELD(*reg_data,
+			CP_IQ_WAIT_TIME2,
+			SCH_WAVE,
+			grace_period);
+
+	*reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2);
+}
+
 static void program_trap_handler_settings(struct amdgpu_device *adev,
 		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr,
 		uint32_t inst)
@@ -848,5 +889,7 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
+	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
+	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.program_trap_handler_settings = program_trap_handler_settings,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
index 251d61fbde079..1e993a2136464 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
@@ -26,3 +26,9 @@ uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev,
 uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
 					bool keep_trap_enabled,
 					uint32_t vmid);
+void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
+void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
+					       uint32_t wait_times,
+					       uint32_t grace_period,
+					       uint32_t *reg_offset,
+					       uint32_t *reg_data);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 8b293f3dcbd2b..387bdf4823c90 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -672,6 +672,8 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info_v10_3,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3,
 	.program_trap_handler_settings = program_trap_handler_settings_v10_3,
+	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
+	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 8d7d04704b00f..829ee720cc448 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -739,6 +739,24 @@ uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
 	return 0;
 }
 
+/* kgd_gfx_v9_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
+ * The values read are:
+ *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
+ *     atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads.
+ *     wrm_offload_wait_time    -- Wait Count for WAIT_REG_MEM Offloads.
+ *     gws_wait_time            -- Wait Count for Global Wave Syncs.
+ *     que_sleep_wait_time      -- Wait Count for Dequeue Retry.
+ *     sch_wave_wait_time       -- Wait Count for Scheduling Wave Message.
+ *     sem_rearm_wait_time      -- Wait Count for Semaphore re-arm.
+ *     deq_retry_wait_time      -- Wait Count for Global Wave Syncs.
+ */
+void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev,
+					uint32_t *wait_times)
+
+{
+	*wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2));
+}
+
 void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev,
 			uint32_t vmid, uint64_t page_table_base)
 {
@@ -926,6 +944,29 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 				adev->gfx.cu_info.max_waves_per_simd;
 }
 
+void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev,
+		uint32_t wait_times,
+		uint32_t grace_period,
+		uint32_t *reg_offset,
+		uint32_t *reg_data)
+{
+	*reg_data = wait_times;
+
+	/*
+	 * The CP cannont handle a 0 grace period input and will result in
+	 * an infinite grace period being set so set to 1 to prevent this.
+	 */
+	if (grace_period == 0)
+		grace_period = 1;
+
+	*reg_data = REG_SET_FIELD(*reg_data,
+			CP_IQ_WAIT_TIME2,
+			SCH_WAVE,
+			grace_period);
+
+	*reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2);
+}
+
 void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev,
 		uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, uint32_t inst)
 {
@@ -969,6 +1010,8 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_gfx_v9_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v9_disable_debug_trap,
+	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
+	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index 9588ff055393d..fed5b7f18b1ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -20,8 +20,6 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  */
 
-
-
 void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid,
 		uint32_t sh_mem_config,
 		uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit,
@@ -73,3 +71,9 @@ uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev,
 uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
 					bool keep_trap_enabled,
 					uint32_t vmid);
+void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
+void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev,
+					       uint32_t wait_times,
+					       uint32_t grace_period,
+					       uint32_t *reg_offset,
+					       uint32_t *reg_data);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 2baa0781eafc2..0b88a64e61fe2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -46,10 +46,13 @@ static int set_pasid_vmid_mapping(struct device_queue_manager *dqm,
 
 static int execute_queues_cpsch(struct device_queue_manager *dqm,
 				enum kfd_unmap_queues_filter filter,
-				uint32_t filter_param);
+				uint32_t filter_param,
+				uint32_t grace_period);
 static int unmap_queues_cpsch(struct device_queue_manager *dqm,
 				enum kfd_unmap_queues_filter filter,
-				uint32_t filter_param, bool reset);
+				uint32_t filter_param,
+				uint32_t grace_period,
+				bool reset);
 
 static int map_queues_cpsch(struct device_queue_manager *dqm);
 
@@ -866,7 +869,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 	if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) {
 		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			retval = unmap_queues_cpsch(dqm,
-						    KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, false);
+						    KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD, false);
 		else if (prev_active)
 			retval = remove_queue_mes(dqm, q, &pdd->qpd);
 
@@ -1042,7 +1045,8 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 		retval = execute_queues_cpsch(dqm,
 					      qpd->is_debug ?
 					      KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES :
-					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
+					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
+					      USE_DEFAULT_GRACE_PERIOD);
 
 out:
 	dqm_unlock(dqm);
@@ -1182,8 +1186,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 	}
 	if (!dqm->dev->kfd->shared_resources.enable_mes)
 		retval = execute_queues_cpsch(dqm,
-					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
-
+					      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD);
 	eviction_duration = get_jiffies_64() - pdd->last_evict_timestamp;
 	atomic64_add(eviction_duration, &pdd->evict_duration_counter);
 vm_not_acquired:
@@ -1525,6 +1528,9 @@ static int initialize_cpsch(struct device_queue_manager *dqm)
 
 	init_sdma_bitmaps(dqm);
 
+	if (dqm->dev->kfd2kgd->get_iq_wait_times)
+		dqm->dev->kfd2kgd->get_iq_wait_times(dqm->dev->adev,
+					&dqm->wait_times);
 	return 0;
 }
 
@@ -1563,8 +1569,9 @@ static int start_cpsch(struct device_queue_manager *dqm)
 	dqm->is_hws_hang = false;
 	dqm->is_resetting = false;
 	dqm->sched_running = true;
+
 	if (!dqm->dev->kfd->shared_resources.enable_mes)
-		execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
+		execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD);
 	dqm_unlock(dqm);
 
 	return 0;
@@ -1589,7 +1596,7 @@ static int stop_cpsch(struct device_queue_manager *dqm)
 
 	if (!dqm->is_hws_hang) {
 		if (!dqm->dev->kfd->shared_resources.enable_mes)
-			unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, false);
+			unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD, false);
 		else
 			remove_all_queues_mes(dqm);
 	}
@@ -1631,7 +1638,8 @@ static int create_kernel_queue_cpsch(struct device_queue_manager *dqm,
 	list_add(&kq->list, &qpd->priv_queue_list);
 	increment_queue_count(dqm, qpd, kq->queue);
 	qpd->is_debug = true;
-	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
+	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
+			USE_DEFAULT_GRACE_PERIOD);
 	dqm_unlock(dqm);
 
 	return 0;
@@ -1645,7 +1653,8 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm,
 	list_del(&kq->list);
 	decrement_queue_count(dqm, qpd, kq->queue);
 	qpd->is_debug = false;
-	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0);
+	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0,
+			USE_DEFAULT_GRACE_PERIOD);
 	/*
 	 * Unconditionally decrement this counter, regardless of the queue's
 	 * type.
@@ -1722,7 +1731,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 
 		if (!dqm->dev->kfd->shared_resources.enable_mes)
 			retval = execute_queues_cpsch(dqm,
-					KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
+					KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD);
 		else
 			retval = add_queue_mes(dqm, q, qpd);
 		if (retval)
@@ -1811,7 +1820,9 @@ static int map_queues_cpsch(struct device_queue_manager *dqm)
 /* dqm->lock mutex has to be locked before calling this function */
 static int unmap_queues_cpsch(struct device_queue_manager *dqm,
 				enum kfd_unmap_queues_filter filter,
-				uint32_t filter_param, bool reset)
+				uint32_t filter_param,
+				uint32_t grace_period,
+				bool reset)
 {
 	int retval = 0;
 	struct mqd_manager *mqd_mgr;
@@ -1823,6 +1834,12 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm,
 	if (!dqm->active_runlist)
 		return retval;
 
+	if (grace_period != USE_DEFAULT_GRACE_PERIOD) {
+		retval = pm_update_grace_period(&dqm->packet_mgr, grace_period);
+		if (retval)
+			return retval;
+	}
+
 	retval = pm_send_unmap_queue(&dqm->packet_mgr, filter, filter_param, reset);
 	if (retval)
 		return retval;
@@ -1855,6 +1872,13 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm,
 		return -ETIME;
 	}
 
+	/* We need to reset the grace period value for this device */
+	if (grace_period != USE_DEFAULT_GRACE_PERIOD) {
+		if (pm_update_grace_period(&dqm->packet_mgr,
+					USE_DEFAULT_GRACE_PERIOD))
+			pr_err("Failed to reset grace period\n");
+	}
+
 	pm_release_ib(&dqm->packet_mgr);
 	dqm->active_runlist = false;
 
@@ -1870,7 +1894,7 @@ static int reset_queues_cpsch(struct device_queue_manager *dqm,
 	dqm_lock(dqm);
 
 	retval = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_BY_PASID,
-			pasid, true);
+			pasid, USE_DEFAULT_GRACE_PERIOD, true);
 
 	dqm_unlock(dqm);
 	return retval;
@@ -1879,13 +1903,14 @@ static int reset_queues_cpsch(struct device_queue_manager *dqm,
 /* dqm->lock mutex has to be locked before calling this function */
 static int execute_queues_cpsch(struct device_queue_manager *dqm,
 				enum kfd_unmap_queues_filter filter,
-				uint32_t filter_param)
+				uint32_t filter_param,
+				uint32_t grace_period)
 {
 	int retval;
 
 	if (dqm->is_hws_hang)
 		return -EIO;
-	retval = unmap_queues_cpsch(dqm, filter, filter_param, false);
+	retval = unmap_queues_cpsch(dqm, filter, filter_param, grace_period, false);
 	if (retval)
 		return retval;
 
@@ -1943,7 +1968,8 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 		if (!dqm->dev->kfd->shared_resources.enable_mes) {
 			decrement_queue_count(dqm, qpd, q);
 			retval = execute_queues_cpsch(dqm,
-						      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
+						      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
+						      USE_DEFAULT_GRACE_PERIOD);
 			if (retval == -ETIME)
 				qpd->reset_wavefronts = true;
 		} else {
@@ -2228,7 +2254,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm,
 	}
 
 	if (!dqm->dev->kfd->shared_resources.enable_mes)
-		retval = execute_queues_cpsch(dqm, filter, 0);
+		retval = execute_queues_cpsch(dqm, filter, 0, USE_DEFAULT_GRACE_PERIOD);
 
 	if ((!dqm->is_hws_hang) && (retval || qpd->reset_wavefronts)) {
 		pr_warn("Resetting wave fronts (cpsch) on dev %p\n", dqm->dev);
@@ -2589,7 +2615,8 @@ int dqm_debugfs_hang_hws(struct device_queue_manager *dqm)
 		return r;
 	}
 	dqm->active_runlist = true;
-	r = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0);
+	r = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES,
+				0, USE_DEFAULT_GRACE_PERIOD);
 	dqm_unlock(dqm);
 
 	return r;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index cd4383bb207f4..d4dd3b4acbf0a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -37,6 +37,7 @@
 
 #define KFD_MES_PROCESS_QUANTUM		100000
 #define KFD_MES_GANG_QUANTUM		10000
+#define USE_DEFAULT_GRACE_PERIOD 0xffffffff
 
 struct device_process_node {
 	struct qcm_process_device *qpd;
@@ -259,6 +260,8 @@ struct device_queue_manager {
 
 	/* used for GFX 9.4.3 only */
 	uint32_t		current_logical_xcc_start;
+
+	uint32_t		wait_times;
 };
 
 void device_queue_manager_init_cik(
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
index 2f54172e9175c..401096c103b2f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c
@@ -370,6 +370,38 @@ int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address,
 	return retval;
 }
 
+int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period)
+{
+	int retval = 0;
+	uint32_t *buffer, size;
+
+	size = pm->pmf->set_grace_period_size;
+
+	mutex_lock(&pm->lock);
+
+	if (size) {
+		kq_acquire_packet_buffer(pm->priv_queue,
+			size / sizeof(uint32_t),
+			(unsigned int **)&buffer);
+
+		if (!buffer) {
+			pr_err("Failed to allocate buffer on kernel queue\n");
+			retval = -ENOMEM;
+			goto out;
+		}
+
+		retval = pm->pmf->set_grace_period(pm, buffer, grace_period);
+		if (!retval)
+			kq_submit_packet(pm->priv_queue);
+		else
+			kq_rollback_packet(pm->priv_queue);
+	}
+
+out:
+	mutex_unlock(&pm->lock);
+	return retval;
+}
+
 int pm_send_unmap_queue(struct packet_manager *pm,
 			enum kfd_unmap_queues_filter filter,
 			uint32_t filter_param, bool reset)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 44cf3a5f6fdba..1fda6dcf84b16 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -262,6 +262,41 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 	return 0;
 }
 
+static int pm_set_grace_period_v9(struct packet_manager *pm,
+		uint32_t *buffer,
+		uint32_t grace_period)
+{
+	struct pm4_mec_write_data_mmio *packet;
+	uint32_t reg_offset = 0;
+	uint32_t reg_data = 0;
+
+	pm->dqm->dev->kfd2kgd->build_grace_period_packet_info(
+			pm->dqm->dev->adev,
+			pm->dqm->wait_times,
+			grace_period,
+			&reg_offset,
+			&reg_data);
+
+	if (grace_period == USE_DEFAULT_GRACE_PERIOD)
+		reg_data = pm->dqm->wait_times;
+
+	packet = (struct pm4_mec_write_data_mmio *)buffer;
+	memset(buffer, 0, sizeof(struct pm4_mec_write_data_mmio));
+
+	packet->header.u32All = pm_build_pm4_header(IT_WRITE_DATA,
+					sizeof(struct pm4_mec_write_data_mmio));
+
+	packet->bitfields2.dst_sel  = dst_sel___write_data__mem_mapped_register;
+	packet->bitfields2.addr_incr =
+			addr_incr___write_data__do_not_increment_address;
+
+	packet->bitfields3.dst_mmreg_addr = reg_offset;
+
+	packet->data = reg_data;
+
+	return 0;
+}
+
 static int pm_unmap_queues_v9(struct packet_manager *pm, uint32_t *buffer,
 			enum kfd_unmap_queues_filter filter,
 			uint32_t filter_param, bool reset)
@@ -345,6 +380,7 @@ const struct packet_manager_funcs kfd_v9_pm_funcs = {
 	.set_resources		= pm_set_resources_v9,
 	.map_queues		= pm_map_queues_v9,
 	.unmap_queues		= pm_unmap_queues_v9,
+	.set_grace_period       = pm_set_grace_period_v9,
 	.query_status		= pm_query_status_v9,
 	.release_mem		= NULL,
 	.map_process_size	= sizeof(struct pm4_mes_map_process),
@@ -352,6 +388,7 @@ const struct packet_manager_funcs kfd_v9_pm_funcs = {
 	.set_resources_size	= sizeof(struct pm4_mes_set_resources),
 	.map_queues_size	= sizeof(struct pm4_mes_map_queues),
 	.unmap_queues_size	= sizeof(struct pm4_mes_unmap_queues),
+	.set_grace_period_size  = sizeof(struct pm4_mec_write_data_mmio),
 	.query_status_size	= sizeof(struct pm4_mes_query_status),
 	.release_mem_size	= 0,
 };
@@ -362,6 +399,7 @@ const struct packet_manager_funcs kfd_aldebaran_pm_funcs = {
 	.set_resources		= pm_set_resources_v9,
 	.map_queues		= pm_map_queues_v9,
 	.unmap_queues		= pm_unmap_queues_v9,
+	.set_grace_period       = pm_set_grace_period_v9,
 	.query_status		= pm_query_status_v9,
 	.release_mem		= NULL,
 	.map_process_size	= sizeof(struct pm4_mes_map_process_aldebaran),
@@ -369,6 +407,7 @@ const struct packet_manager_funcs kfd_aldebaran_pm_funcs = {
 	.set_resources_size	= sizeof(struct pm4_mes_set_resources),
 	.map_queues_size	= sizeof(struct pm4_mes_map_queues),
 	.unmap_queues_size	= sizeof(struct pm4_mes_unmap_queues),
+	.set_grace_period_size  = sizeof(struct pm4_mec_write_data_mmio),
 	.query_status_size	= sizeof(struct pm4_mes_query_status),
 	.release_mem_size	= 0,
 };
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
index faf4772ed317c..c1199d06d131b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c
@@ -303,6 +303,7 @@ const struct packet_manager_funcs kfd_vi_pm_funcs = {
 	.set_resources		= pm_set_resources_vi,
 	.map_queues		= pm_map_queues_vi,
 	.unmap_queues		= pm_unmap_queues_vi,
+	.set_grace_period	= NULL,
 	.query_status		= pm_query_status_vi,
 	.release_mem		= pm_release_mem_vi,
 	.map_process_size	= sizeof(struct pm4_mes_map_process),
@@ -310,6 +311,7 @@ const struct packet_manager_funcs kfd_vi_pm_funcs = {
 	.set_resources_size	= sizeof(struct pm4_mes_set_resources),
 	.map_queues_size	= sizeof(struct pm4_mes_map_queues),
 	.unmap_queues_size	= sizeof(struct pm4_mes_unmap_queues),
+	.set_grace_period_size	= 0,
 	.query_status_size	= sizeof(struct pm4_mes_query_status),
 	.release_mem_size	= sizeof(struct pm4_mec_release_mem)
 };
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
index 2ad708c640125..206f1960857fb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
@@ -584,6 +584,71 @@ struct pm4_mec_release_mem {
 
 #endif
 
+#ifndef PM4_MEC_WRITE_DATA_DEFINED
+#define PM4_MEC_WRITE_DATA_DEFINED
+
+enum WRITE_DATA_dst_sel_enum {
+	dst_sel___write_data__mem_mapped_register = 0,
+	dst_sel___write_data__tc_l2 = 2,
+	dst_sel___write_data__gds = 3,
+	dst_sel___write_data__memory = 5,
+	dst_sel___write_data__memory_mapped_adc_persistent_state = 6,
+};
+
+enum WRITE_DATA_addr_incr_enum {
+	addr_incr___write_data__increment_address = 0,
+	addr_incr___write_data__do_not_increment_address = 1
+};
+
+enum WRITE_DATA_wr_confirm_enum {
+	wr_confirm___write_data__do_not_wait_for_write_confirmation = 0,
+	wr_confirm___write_data__wait_for_write_confirmation = 1
+};
+
+enum WRITE_DATA_cache_policy_enum {
+	cache_policy___write_data__lru = 0,
+	cache_policy___write_data__stream = 1
+};
+
+
+struct pm4_mec_write_data_mmio {
+	union {
+		union PM4_MES_TYPE_3_HEADER header;     /*header */
+		unsigned int ordinal1;
+	};
+
+	union {
+		struct {
+			unsigned int reserved1:8;
+			unsigned int dst_sel:4;
+			unsigned int reserved2:4;
+			unsigned int addr_incr:1;
+			unsigned int reserved3:2;
+			unsigned int resume_vf:1;
+			unsigned int wr_confirm:1;
+			unsigned int reserved4:4;
+			unsigned int cache_policy:2;
+			unsigned int reserved5:5;
+		} bitfields2;
+		unsigned int ordinal2;
+	};
+
+	union {
+		struct {
+			unsigned int dst_mmreg_addr:18;
+			unsigned int reserved6:14;
+		} bitfields3;
+		unsigned int ordinal3;
+	};
+
+	uint32_t reserved7;
+
+	uint32_t data;
+
+};
+
+#endif
+
 enum {
 	CACHE_FLUSH_AND_INV_TS_EVENT = 0x00000014
 };
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 1b272f879b4c4..4c912b7735b5b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1350,6 +1350,8 @@ struct packet_manager_funcs {
 	int (*unmap_queues)(struct packet_manager *pm, uint32_t *buffer,
 			enum kfd_unmap_queues_filter mode,
 			uint32_t filter_param, bool reset);
+	int (*set_grace_period)(struct packet_manager *pm, uint32_t *buffer,
+			uint32_t grace_period);
 	int (*query_status)(struct packet_manager *pm, uint32_t *buffer,
 			uint64_t fence_address,	uint64_t fence_value);
 	int (*release_mem)(uint64_t gpu_addr, uint32_t *buffer);
@@ -1360,6 +1362,7 @@ struct packet_manager_funcs {
 	int set_resources_size;
 	int map_queues_size;
 	int unmap_queues_size;
+	int set_grace_period_size;
 	int query_status_size;
 	int release_mem_size;
 };
@@ -1382,6 +1385,8 @@ int pm_send_unmap_queue(struct packet_manager *pm,
 
 void pm_release_ib(struct packet_manager *pm);
 
+int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period);
+
 /* Following PM funcs can be shared among VI and AI */
 unsigned int pm_build_pm4_header(unsigned int opcode, size_t packet_size);
 
-- 
GitLab


From 97ae3c8cce96f3bebf883d0812cef5d3fdbe3e64 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 4 Apr 2022 12:27:43 -0400
Subject: [PATCH 1529/1791] drm/amdkfd: prepare map process for single process
 debug devices

Older HW only supports debugging on a single process because the
SPI debug mode setting registers are device global.

The HWS has supplied a single pinned VMID (0xf) for MAP_PROCESS
for debug purposes. To pin the VMID, the KFD will remove the VMID from
the HWS dynamic VMID allocation via SET_RESOUCES so that a debugged
process will never migrate away from its pinned VMID.

The KFD is responsible for reserving and releasing this pinned VMID
accordingly whenever the debugger attaches and detaches respectively.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 93 +++++++++++++++++++
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |  5 +
 .../drm/amd/amdkfd/kfd_packet_manager_v9.c    |  9 ++
 .../gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h   |  5 +-
 4 files changed, 111 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 0b88a64e61fe2..cfe5bd59070e8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -1525,6 +1525,7 @@ static int initialize_cpsch(struct device_queue_manager *dqm)
 	dqm->gws_queue_count = 0;
 	dqm->active_runlist = false;
 	INIT_WORK(&dqm->hw_exception_work, kfd_process_hw_exception);
+	dqm->trap_debug_vmid = 0;
 
 	init_sdma_bitmaps(dqm);
 
@@ -2501,6 +2502,98 @@ static void kfd_process_hw_exception(struct work_struct *work)
 	amdgpu_amdkfd_gpu_reset(dqm->dev->adev);
 }
 
+int reserve_debug_trap_vmid(struct device_queue_manager *dqm,
+				struct qcm_process_device *qpd)
+{
+	int r;
+	int updated_vmid_mask;
+
+	if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) {
+		pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy);
+		return -EINVAL;
+	}
+
+	dqm_lock(dqm);
+
+	if (dqm->trap_debug_vmid != 0) {
+		pr_err("Trap debug id already reserved\n");
+		r = -EBUSY;
+		goto out_unlock;
+	}
+
+	r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0,
+			USE_DEFAULT_GRACE_PERIOD, false);
+	if (r)
+		goto out_unlock;
+
+	updated_vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap;
+	updated_vmid_mask &= ~(1 << dqm->dev->vm_info.last_vmid_kfd);
+
+	dqm->dev->kfd->shared_resources.compute_vmid_bitmap = updated_vmid_mask;
+	dqm->trap_debug_vmid = dqm->dev->vm_info.last_vmid_kfd;
+	r = set_sched_resources(dqm);
+	if (r)
+		goto out_unlock;
+
+	r = map_queues_cpsch(dqm);
+	if (r)
+		goto out_unlock;
+
+	pr_debug("Reserved VMID for trap debug: %i\n", dqm->trap_debug_vmid);
+
+out_unlock:
+	dqm_unlock(dqm);
+	return r;
+}
+
+/*
+ * Releases vmid for the trap debugger
+ */
+int release_debug_trap_vmid(struct device_queue_manager *dqm,
+			struct qcm_process_device *qpd)
+{
+	int r;
+	int updated_vmid_mask;
+	uint32_t trap_debug_vmid;
+
+	if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) {
+		pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy);
+		return -EINVAL;
+	}
+
+	dqm_lock(dqm);
+	trap_debug_vmid = dqm->trap_debug_vmid;
+	if (dqm->trap_debug_vmid == 0) {
+		pr_err("Trap debug id is not reserved\n");
+		r = -EINVAL;
+		goto out_unlock;
+	}
+
+	r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0,
+			USE_DEFAULT_GRACE_PERIOD, false);
+	if (r)
+		goto out_unlock;
+
+	updated_vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap;
+	updated_vmid_mask |= (1 << dqm->dev->vm_info.last_vmid_kfd);
+
+	dqm->dev->kfd->shared_resources.compute_vmid_bitmap = updated_vmid_mask;
+	dqm->trap_debug_vmid = 0;
+	r = set_sched_resources(dqm);
+	if (r)
+		goto out_unlock;
+
+	r = map_queues_cpsch(dqm);
+	if (r)
+		goto out_unlock;
+
+	pr_debug("Released VMID for trap debug: %i\n", trap_debug_vmid);
+
+out_unlock:
+	dqm_unlock(dqm);
+	return r;
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 static void seq_reg_dump(struct seq_file *m,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index d4dd3b4acbf0a..bf7aa3f841821 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -250,6 +250,7 @@ struct device_queue_manager {
 	struct kfd_mem_obj	*fence_mem;
 	bool			active_runlist;
 	int			sched_policy;
+	uint32_t		trap_debug_vmid;
 
 	/* hw exception  */
 	bool			is_hws_hang;
@@ -285,6 +286,10 @@ unsigned int get_queues_per_pipe(struct device_queue_manager *dqm);
 unsigned int get_pipes_per_mec(struct device_queue_manager *dqm);
 unsigned int get_num_sdma_queues(struct device_queue_manager *dqm);
 unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm);
+int reserve_debug_trap_vmid(struct device_queue_manager *dqm,
+			struct qcm_process_device *qpd);
+int release_debug_trap_vmid(struct device_queue_manager *dqm,
+			struct qcm_process_device *qpd);
 
 static inline unsigned int get_sh_mem_bases_32(struct kfd_process_device *pdd)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 1fda6dcf84b16..0fe73dbd28afa 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -34,6 +34,9 @@ static int pm_map_process_v9(struct packet_manager *pm,
 {
 	struct pm4_mes_map_process *packet;
 	uint64_t vm_page_table_base_addr = qpd->page_table_base;
+	struct kfd_node *kfd = pm->dqm->dev;
+	struct kfd_process_device *pdd =
+			container_of(qpd, struct kfd_process_device, qpd);
 
 	packet = (struct pm4_mes_map_process *)buffer;
 	memset(buffer, 0, sizeof(struct pm4_mes_map_process));
@@ -49,6 +52,12 @@ static int pm_map_process_v9(struct packet_manager *pm,
 	packet->bitfields14.sdma_enable = 1;
 	packet->bitfields14.num_queues = (qpd->is_debug) ? 0 : qpd->queue_count;
 
+	if (kfd->dqm->trap_debug_vmid && pdd->process->debug_trap_enabled &&
+			pdd->process->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) {
+		packet->bitfields2.debug_vmid = kfd->dqm->trap_debug_vmid;
+		packet->bitfields2.new_debug = 1;
+	}
+
 	packet->sh_mem_config = qpd->sh_mem_config;
 	packet->sh_mem_bases = qpd->sh_mem_bases;
 	if (qpd->tba_addr) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
index 206f1960857fb..8b6b2bd5c148f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h
@@ -146,7 +146,10 @@ struct pm4_mes_map_process {
 	union {
 		struct {
 			uint32_t pasid:16;
-			uint32_t reserved1:8;
+			uint32_t reserved1:2;
+			uint32_t debug_vmid:4;
+			uint32_t new_debug:1;
+			uint32_t reserved2:1;
 			uint32_t diq_enable:1;
 			uint32_t process_quantum:7;
 		} bitfields2;
-- 
GitLab


From 0de4ec9a03537bd2b189b5afbf83acd6b72b0258 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 4 Apr 2022 13:38:11 -0400
Subject: [PATCH 1530/1791] drm/amdgpu: prepare map process for multi-process
 debug devices

Unlike single process debug devices, multi-process debug devices allow
debug mode setting per-VMID (non-device-global).

Because the HWS manages PASID-VMID mapping, the new MAP_PROCESS API allows
the KFD to forward the required SPI debug register write requests.

To request a new debug mode setting change, the KFD must be able to
preempt all queues then remap all queues with these new setting
requests for MAP_PROCESS to take effect.

Note that by default, trap enablement in non-debug mode must be disabled
for performance reasons for multi-process debug devices due to setup
overhead in FW.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  5 ++
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 51 +++++++++++++++++++
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |  3 ++
 .../drm/amd/amdkfd/kfd_packet_manager_v9.c    | 14 +++++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  9 ++++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |  5 ++
 6 files changed, 87 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index a8abfe2a0a140..db6d72e7930f7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -29,4 +29,9 @@ int kfd_dbg_trap_disable(struct kfd_process *target);
 int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			void __user *runtime_info,
 			uint32_t *runtime_info_size);
+static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
+{
+	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2);
+}
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index cfe5bd59070e8..495c9238254e7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -36,6 +36,7 @@
 #include "kfd_kernel_queue.h"
 #include "amdgpu_amdkfd.h"
 #include "mes_api_def.h"
+#include "kfd_debug.h"
 
 /* Size of the per-pipe EOP queue */
 #define CIK_HPD_EOP_BYTES_LOG2 11
@@ -2594,6 +2595,56 @@ int release_debug_trap_vmid(struct device_queue_manager *dqm,
 	return r;
 }
 
+int debug_lock_and_unmap(struct device_queue_manager *dqm)
+{
+	int r;
+
+	if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) {
+		pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy);
+		return -EINVAL;
+	}
+
+	if (!kfd_dbg_is_per_vmid_supported(dqm->dev))
+		return 0;
+
+	dqm_lock(dqm);
+
+	r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, 0, false);
+	if (r)
+		dqm_unlock(dqm);
+
+	return r;
+}
+
+int debug_map_and_unlock(struct device_queue_manager *dqm)
+{
+	int r;
+
+	if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) {
+		pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy);
+		return -EINVAL;
+	}
+
+	if (!kfd_dbg_is_per_vmid_supported(dqm->dev))
+		return 0;
+
+	r = map_queues_cpsch(dqm);
+
+	dqm_unlock(dqm);
+
+	return r;
+}
+
+int debug_refresh_runlist(struct device_queue_manager *dqm)
+{
+	int r = debug_lock_and_unmap(dqm);
+
+	if (r)
+		return r;
+
+	return debug_map_and_unlock(dqm);
+}
+
 #if defined(CONFIG_DEBUG_FS)
 
 static void seq_reg_dump(struct seq_file *m,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index bf7aa3f841821..bb75d93712eb2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -290,6 +290,9 @@ int reserve_debug_trap_vmid(struct device_queue_manager *dqm,
 			struct qcm_process_device *qpd);
 int release_debug_trap_vmid(struct device_queue_manager *dqm,
 			struct qcm_process_device *qpd);
+int debug_lock_and_unmap(struct device_queue_manager *dqm);
+int debug_map_and_unlock(struct device_queue_manager *dqm);
+int debug_refresh_runlist(struct device_queue_manager *dqm);
 
 static inline unsigned int get_sh_mem_bases_32(struct kfd_process_device *pdd)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
index 0fe73dbd28afa..29a2d0499b674 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c
@@ -88,6 +88,10 @@ static int pm_map_process_aldebaran(struct packet_manager *pm,
 {
 	struct pm4_mes_map_process_aldebaran *packet;
 	uint64_t vm_page_table_base_addr = qpd->page_table_base;
+	struct kfd_dev *kfd = pm->dqm->dev->kfd;
+	struct kfd_process_device *pdd =
+			container_of(qpd, struct kfd_process_device, qpd);
+	int i;
 
 	packet = (struct pm4_mes_map_process_aldebaran *)buffer;
 	memset(buffer, 0, sizeof(struct pm4_mes_map_process_aldebaran));
@@ -102,6 +106,16 @@ static int pm_map_process_aldebaran(struct packet_manager *pm,
 	packet->bitfields14.num_oac = qpd->num_oac;
 	packet->bitfields14.sdma_enable = 1;
 	packet->bitfields14.num_queues = (qpd->is_debug) ? 0 : qpd->queue_count;
+	packet->spi_gdbg_per_vmid_cntl = pdd->spi_dbg_override |
+						pdd->spi_dbg_launch_mode;
+
+	if (pdd->process->debug_trap_enabled) {
+		for (i = 0; i < kfd->device_info.num_of_watch_points; i++)
+			packet->tcp_watch_cntl[i] = pdd->watch_points[i];
+
+		packet->bitfields2.single_memops =
+				!!(pdd->process->dbg_flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP);
+	}
 
 	packet->sh_mem_config = qpd->sh_mem_config;
 	packet->sh_mem_bases = qpd->sh_mem_bases;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 4c912b7735b5b..8fca7175daabf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -816,6 +816,12 @@ struct kfd_process_device {
 	uint64_t faults;
 	uint64_t page_in;
 	uint64_t page_out;
+
+	/* Tracks debug per-vmid request settings */
+	uint32_t spi_dbg_override;
+	uint32_t spi_dbg_launch_mode;
+	uint32_t watch_points[4];
+
 	/*
 	 * If this process has been checkpointed before, then the user
 	 * application will use the original gpu_id on the
@@ -952,6 +958,9 @@ struct kfd_process {
 
 	bool xnack_enabled;
 
+	/* Tracks debug per-vmid request for debug flags */
+	bool dbg_flags;
+
 	atomic_t poison;
 	/* Queues are in paused stated because we are in the process of doing a CRIU checkpoint */
 	bool queues_paused;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index d75dac92775c7..725d936b2cc7d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1612,6 +1612,11 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 	}
 
 	p->pdds[p->n_pdds++] = pdd;
+	if (kfd_dbg_is_per_vmid_supported(pdd->dev))
+		pdd->spi_dbg_override = pdd->dev->kfd2kgd->disable_debug_trap(
+							pdd->dev->adev,
+							false,
+							0);
 
 	/* Init idr used for memory handle translation */
 	idr_init(&pdd->alloc_idr);
-- 
GitLab


From a9818854ea7870ec5464d37b72c89f5fc198708e Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 26 Aug 2022 22:04:15 -0400
Subject: [PATCH 1531/1791] drm/amdgpu: expose debug api for mes

Similar to the F32 HWS, the RS64 HWS for GFX11 now supports a multi-process
debug API.

The skip_process_ctx_clear ADD_QUEUE requirement is to prevent the MES
from clearing the process context when the first queue is added to the
scheduler in order to maintain debug mode settings during queue preemption
and restore.  The MES clears the process context in this case due to an
unresolved FW caching bug during normal mode operations.
During debug mode, the KFD will hold a reference to the target process
so the process context should never go stale and MES can afford to skip
this requirement.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c       | 32 +++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h       | 20 ++++++++++++
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c        | 12 +++++++
 drivers/gpu/drm/amd/include/mes_v11_api_def.h | 21 +++++++++++-
 4 files changed, 84 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index 49bb6c03d6067..20cc3fffe921a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -924,6 +924,38 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg,
 	return r;
 }
 
+int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
+				uint64_t process_context_addr,
+				uint32_t spi_gdbg_per_vmid_cntl,
+				const uint32_t *tcp_watch_cntl,
+				uint32_t flags)
+{
+	struct mes_misc_op_input op_input = {0};
+	int r;
+
+	if (!adev->mes.funcs->misc_op) {
+		DRM_ERROR("mes set shader debugger is not supported!\n");
+		return -EINVAL;
+	}
+
+	op_input.op = MES_MISC_OP_SET_SHADER_DEBUGGER;
+	op_input.set_shader_debugger.process_context_addr = process_context_addr;
+	op_input.set_shader_debugger.flags.u32all = flags;
+	op_input.set_shader_debugger.spi_gdbg_per_vmid_cntl = spi_gdbg_per_vmid_cntl;
+	memcpy(op_input.set_shader_debugger.tcp_watch_cntl, tcp_watch_cntl,
+			sizeof(op_input.set_shader_debugger.tcp_watch_cntl));
+
+	amdgpu_mes_lock(&adev->mes);
+
+	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
+	if (r)
+		DRM_ERROR("failed to set_shader_debugger\n");
+
+	amdgpu_mes_unlock(&adev->mes);
+
+	return r;
+}
+
 static void
 amdgpu_mes_ring_to_queue_props(struct amdgpu_device *adev,
 			       struct amdgpu_ring *ring,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index 547ec35691fac..d20df0cf0d88b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -256,6 +256,7 @@ enum mes_misc_opcode {
 	MES_MISC_OP_READ_REG,
 	MES_MISC_OP_WRM_REG_WAIT,
 	MES_MISC_OP_WRM_REG_WR_WAIT,
+	MES_MISC_OP_SET_SHADER_DEBUGGER,
 };
 
 struct mes_misc_op_input {
@@ -278,6 +279,20 @@ struct mes_misc_op_input {
 			uint32_t                   reg0;
 			uint32_t                   reg1;
 		} wrm_reg;
+
+		struct {
+			uint64_t process_context_addr;
+			union {
+				struct {
+					uint64_t single_memop : 1;
+					uint64_t single_alu_op : 1;
+					uint64_t reserved: 30;
+				};
+				uint32_t u32all;
+			} flags;
+			uint32_t spi_gdbg_per_vmid_cntl;
+			uint32_t tcp_watch_cntl[4];
+		} set_shader_debugger;
 	};
 };
 
@@ -340,6 +355,11 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg,
 int amdgpu_mes_reg_write_reg_wait(struct amdgpu_device *adev,
 				  uint32_t reg0, uint32_t reg1,
 				  uint32_t ref, uint32_t mask);
+int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
+				uint64_t process_context_addr,
+				uint32_t spi_gdbg_per_vmid_cntl,
+				const uint32_t *tcp_watch_cntl,
+				uint32_t flags);
 
 int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id,
 			int queue_type, int idx,
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 90b4a74ccf01a..861910a6662d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -339,6 +339,18 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes,
 		misc_pkt.wait_reg_mem.reg_offset1 = input->wrm_reg.reg0;
 		misc_pkt.wait_reg_mem.reg_offset2 = input->wrm_reg.reg1;
 		break;
+	case MES_MISC_OP_SET_SHADER_DEBUGGER:
+		misc_pkt.opcode = MESAPI_MISC__SET_SHADER_DEBUGGER;
+		misc_pkt.set_shader_debugger.process_context_addr =
+				input->set_shader_debugger.process_context_addr;
+		misc_pkt.set_shader_debugger.flags.u32all =
+				input->set_shader_debugger.flags.u32all;
+		misc_pkt.set_shader_debugger.spi_gdbg_per_vmid_cntl =
+				input->set_shader_debugger.spi_gdbg_per_vmid_cntl;
+		memcpy(misc_pkt.set_shader_debugger.tcp_watch_cntl,
+				input->set_shader_debugger.tcp_watch_cntl,
+				sizeof(misc_pkt.set_shader_debugger.tcp_watch_cntl));
+		break;
 	default:
 		DRM_ERROR("unsupported misc op (%d) \n", input->op);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
index dc694cb246d9f..f3c15f18ddb56 100644
--- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
@@ -274,7 +274,8 @@ union MESAPI__ADD_QUEUE {
 			uint32_t is_kfd_process		: 1;
 			uint32_t trap_en		: 1;
 			uint32_t is_aql_queue		: 1;
-			uint32_t reserved		: 20;
+			uint32_t skip_process_ctx_clear : 1;
+			uint32_t reserved		: 19;
 		};
 		struct MES_API_STATUS		api_status;
 		uint64_t                        tma_addr;
@@ -523,6 +524,7 @@ enum MESAPI_MISC_OPCODE {
 	MESAPI_MISC__QUERY_STATUS,
 	MESAPI_MISC__READ_REG,
 	MESAPI_MISC__WAIT_REG_MEM,
+	MESAPI_MISC__SET_SHADER_DEBUGGER,
 	MESAPI_MISC__MAX,
 };
 
@@ -561,6 +563,20 @@ struct QUERY_STATUS {
 	uint32_t context_id;
 };
 
+struct SET_SHADER_DEBUGGER {
+	uint64_t process_context_addr;
+	union {
+		struct {
+			uint32_t single_memop : 1;  /* SQ_DEBUG.single_memop */
+			uint32_t single_alu_op : 1; /* SQ_DEBUG.single_alu_op */
+			uint32_t reserved : 30;
+		};
+		uint32_t u32all;
+	} flags;
+	uint32_t spi_gdbg_per_vmid_cntl;
+	uint32_t tcp_watch_cntl[4]; /* TCP_WATCHx_CNTL */
+};
+
 union MESAPI__MISC {
 	struct {
 		union MES_API_HEADER	header;
@@ -573,6 +589,9 @@ union MESAPI__MISC {
 			struct		QUERY_STATUS query_status;
 			struct		READ_REG read_reg;
 			struct          WAIT_REG_MEM wait_reg_mem;
+			struct		SET_SHADER_DEBUGGER set_shader_debugger;
+			enum MES_AMD_PRIORITY_LEVEL queue_sch_level;
+
 			uint32_t	data[MISC_DATA_MAX_SIZE_IN_DWORDS];
 		};
 	};
-- 
GitLab


From 218895820e6fccade42a7c3ab9c0a44dec0a1ebc Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 5 Apr 2022 12:34:55 -0400
Subject: [PATCH 1532/1791] drm/amdkfd: add per process hw trap enable and
 disable functions

To enable HW debug mode per process, all devices must be debug enabled
successfully.  If a failure occures, rewind the enablement of debug mode
on the enabled devices.

A power management scenario that needs to be considered is HW
debug mode setting during GFXOFF.  During GFXOFF, these registers
will be unreachable so we have to transiently disable GFXOFF when
setting.  Also, some devices don't support the RLC save restore
function for these debug registers so we have to disable GFXOFF
completely during a debug session.

Cooperative launch also has debugging restriction based on HW/FW bugs.
If such bugs exists, the debugger cannot attach to a process that uses GWS
resources nor can GWS resources be requested if a process is being
debugged.

Multi-process debug devices can only enable trap temporaries based
on certain runtime scenerios, which will be explained when the
runtime enable functions are implemented in a follow up patch.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |   5 +
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 148 ++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |  29 +++++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c |  10 ++
 4 files changed, 190 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index ee086a0a46dfe..826a99acb6fb9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1488,6 +1488,11 @@ static int kfd_ioctl_alloc_queue_gws(struct file *filep,
 		goto out_unlock;
 	}
 
+	if (!kfd_dbg_has_gws_support(dev) && p->debug_trap_enabled) {
+		retval = -EBUSY;
+		goto out_unlock;
+	}
+
 	retval = pqm_set_gws(&p->pqm, args->queue_id, args->num_gws ? dev->gws : NULL);
 	mutex_unlock(&p->mutex);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 898cc1fe3d135..73b07b5f17f14 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -21,13 +21,78 @@
  */
 
 #include "kfd_debug.h"
+#include "kfd_device_queue_manager.h"
 #include <linux/file.h>
 
+static int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
+{
+	uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode;
+	uint32_t flags = pdd->process->dbg_flags;
+
+	if (!kfd_dbg_is_per_vmid_supported(pdd->dev))
+		return 0;
+
+	return amdgpu_mes_set_shader_debugger(pdd->dev->adev, pdd->proc_ctx_gpu_addr, spi_dbg_cntl,
+						pdd->watch_points, flags);
+}
+
+/* kfd_dbg_trap_deactivate:
+ *	target: target process
+ *	unwind: If this is unwinding a failed kfd_dbg_trap_enable()
+ *	unwind_count:
+ *		If unwind == true, how far down the pdd list we need
+ *				to unwind
+ *		else: ignored
+ */
+static void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count)
+{
+	int i;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		/* If this is an unwind, and we have unwound the required
+		 * enable calls on the pdd list, we need to stop now
+		 * otherwise we may mess up another debugger session.
+		 */
+		if (unwind && i == unwind_count)
+			break;
+
+		/* GFX off is already disabled by debug activate if not RLC restore supported. */
+		if (kfd_dbg_is_rlc_restore_supported(pdd->dev))
+			amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+		pdd->spi_dbg_override =
+				pdd->dev->kfd2kgd->disable_debug_trap(
+				pdd->dev->adev,
+				target->runtime_info.ttmp_setup,
+				pdd->dev->vm_info.last_vmid_kfd);
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+		if (!kfd_dbg_is_per_vmid_supported(pdd->dev) &&
+				release_debug_trap_vmid(pdd->dev->dqm, &pdd->qpd))
+			pr_err("Failed to release debug vmid on [%i]\n", pdd->dev->id);
+
+		if (!pdd->dev->kfd->shared_resources.enable_mes)
+			debug_refresh_runlist(pdd->dev->dqm);
+		else
+			kfd_dbg_set_mes_debug_mode(pdd);
+	}
+}
+
 int kfd_dbg_trap_disable(struct kfd_process *target)
 {
 	if (!target->debug_trap_enabled)
 		return 0;
 
+	/*
+	 * Defer deactivation to runtime if runtime not enabled otherwise reset
+	 * attached running target runtime state to enable for re-attach.
+	 */
+	if (target->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED)
+		kfd_dbg_trap_deactivate(target, false, 0);
+	else if (target->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_DISABLED)
+		target->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_ENABLED;
+
 	fput(target->dbg_ev_file);
 	target->dbg_ev_file = NULL;
 
@@ -42,16 +107,89 @@ int kfd_dbg_trap_disable(struct kfd_process *target)
 	return 0;
 }
 
+static int kfd_dbg_trap_activate(struct kfd_process *target)
+{
+	int i, r = 0;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) {
+			r = reserve_debug_trap_vmid(pdd->dev->dqm, &pdd->qpd);
+
+			if (r) {
+				target->runtime_info.runtime_state = (r == -EBUSY) ?
+							DEBUG_RUNTIME_STATE_ENABLED_BUSY :
+							DEBUG_RUNTIME_STATE_ENABLED_ERROR;
+
+				goto unwind_err;
+			}
+		}
+
+		/* Disable GFX OFF to prevent garbage read/writes to debug registers.
+		 * If RLC restore of debug registers is not supported and runtime enable
+		 * hasn't done so already on ttmp setup request, restore the trap config registers.
+		 *
+		 * If RLC restore of debug registers is not supported, keep gfx off disabled for
+		 * the debug session.
+		 */
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+		if (!(kfd_dbg_is_rlc_restore_supported(pdd->dev) ||
+						target->runtime_info.ttmp_setup))
+			pdd->dev->kfd2kgd->enable_debug_trap(pdd->dev->adev, true,
+								pdd->dev->vm_info.last_vmid_kfd);
+
+		pdd->spi_dbg_override = pdd->dev->kfd2kgd->enable_debug_trap(
+					pdd->dev->adev,
+					false,
+					pdd->dev->vm_info.last_vmid_kfd);
+
+		if (kfd_dbg_is_rlc_restore_supported(pdd->dev))
+			amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+		if (!pdd->dev->kfd->shared_resources.enable_mes)
+			r = debug_refresh_runlist(pdd->dev->dqm);
+		else
+			r = kfd_dbg_set_mes_debug_mode(pdd);
+
+		if (r) {
+			target->runtime_info.runtime_state =
+					DEBUG_RUNTIME_STATE_ENABLED_ERROR;
+			goto unwind_err;
+		}
+	}
+
+	return 0;
+
+unwind_err:
+	/* Enabling debug failed, we need to disable on
+	 * all GPUs so the enable is all or nothing.
+	 */
+	kfd_dbg_trap_deactivate(target, true, i);
+	return r;
+}
+
 int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			void __user *runtime_info, uint32_t *runtime_size)
 {
 	struct file *f;
 	uint32_t copy_size;
-	int r = 0;
+	int i, r = 0;
 
 	if (target->debug_trap_enabled)
 		return -EALREADY;
 
+	/* Enable pre-checks */
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		if (!KFD_IS_SOC15(pdd->dev))
+			return -ENODEV;
+
+		if (!kfd_dbg_has_gws_support(pdd->dev) && pdd->qpd.num_gws)
+			return -EBUSY;
+	}
+
 	copy_size = min((size_t)(*runtime_size), sizeof(target->runtime_info));
 
 	f = fget(fd);
@@ -62,6 +200,10 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 
 	target->dbg_ev_file = f;
 
+	/* defer activation to runtime if not runtime enabled */
+	if (target->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED)
+		kfd_dbg_trap_activate(target);
+
 	/* We already hold the process reference but hold another one for the
 	 * debug session.
 	 */
@@ -71,8 +213,10 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 	if (target->debugger_process)
 		atomic_inc(&target->debugger_process->debugged_process_count);
 
-	if (copy_to_user(runtime_info, (void *)&target->runtime_info, copy_size))
+	if (copy_to_user(runtime_info, (void *)&target->runtime_info, copy_size)) {
+		kfd_dbg_trap_deactivate(target, false, 0);
 		r = -EFAULT;
+	}
 
 	*runtime_size = sizeof(target->runtime_info);
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index db6d72e7930f7..17481f824647d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -34,4 +34,33 @@ static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2);
 }
 
+/*
+ * If GFX off is enabled, chips that do not support RLC restore for the debug
+ * registers will disable GFX off temporarily for the entire debug session.
+ * See disable_on_trap_action_entry and enable_on_trap_action_exit for details.
+ */
+static inline bool kfd_dbg_is_rlc_restore_supported(struct kfd_node *dev)
+{
+	return !(KFD_GC_VERSION(dev) == IP_VERSION(10, 1, 10) ||
+		 KFD_GC_VERSION(dev) == IP_VERSION(10, 1, 1));
+}
+
+static inline bool kfd_dbg_has_gws_support(struct kfd_node *dev)
+{
+	if ((KFD_GC_VERSION(dev) == IP_VERSION(9, 0, 1)
+			&& dev->kfd->mec2_fw_version < 0x81b6) ||
+		(KFD_GC_VERSION(dev) >= IP_VERSION(9, 1, 0)
+			&& KFD_GC_VERSION(dev) <= IP_VERSION(9, 2, 2)
+			&& dev->kfd->mec2_fw_version < 0x1b6) ||
+		(KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0)
+			&& dev->kfd->mec2_fw_version < 0x1b6) ||
+		(KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1)
+			&& dev->kfd->mec2_fw_version < 0x30) ||
+		(KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0) &&
+			KFD_GC_VERSION(dev) < IP_VERSION(12, 0, 0)))
+		return false;
+
+	/* Assume debugging and cooperative launch supported otherwise. */
+	return true;
+}
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 725d936b2cc7d..e77cadadb09b6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1165,9 +1165,19 @@ static void kfd_process_free_notifier(struct mmu_notifier *mn)
 
 static void kfd_process_notifier_release_internal(struct kfd_process *p)
 {
+	int i;
+
 	cancel_delayed_work_sync(&p->eviction_work);
 	cancel_delayed_work_sync(&p->restore_work);
 
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+
+		/* re-enable GFX OFF since runtime enable with ttmp setup disabled it. */
+		if (!kfd_dbg_is_rlc_restore_supported(pdd->dev) && p->runtime_info.ttmp_setup)
+			amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+	}
+
 	/* Indicate to other users that MM is no longer valid */
 	p->mm = NULL;
 	kfd_dbg_trap_disable(p);
-- 
GitLab


From 69a8c3ae2dea84a6d571e4c1aad306f630f3ccfd Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 1 Sep 2022 11:27:15 -0400
Subject: [PATCH 1533/1791] drm/amdkfd: apply trap workaround for gfx11

Due to a HW bug, waves in only half the shader arrays can enter trap.

When starting a debug session, relocate all waves to the first shader
array of each shader engine and mask off the 2nd shader array as
unavailable.

When ending a debug session, re-enable the 2nd shader array per
shader engine.

User CU masking per queue cannot be guaranteed to remain functional
if requested during debugging (e.g. user cu mask requests only 2nd shader
array as an available resource leading to zero HW resources available)
nor can runtime be alerted of any of these changes during execution.

Make user CU masking and debugging mutual exclusive with respect to
availability.

If the debugger tries to attach to a process with a user cu masked
queue, return the runtime status as enabled but busy.

If the debugger tries to attach and fails to reallocate queue waves to
the first shader array of each shader engine, return the runtime status
as enabled but with an error.

In addition, like any other mutli-process debug supported devices,
disable trap temporary setup per-process to avoid performance impact from
setup overhead.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h       |  2 +
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c        |  7 +--
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  2 -
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        | 57 +++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  3 +-
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |  7 +++
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c  |  3 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |  3 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  | 42 ++++++++++----
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  3 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c   |  3 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  5 +-
 .../amd/amdkfd/kfd_process_queue_manager.c    |  9 ++-
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c     |  7 ++-
 14 files changed, 122 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index d20df0cf0d88b..b5f5eed2b5efe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -219,6 +219,8 @@ struct mes_add_queue_input {
 	uint32_t        gws_size;
 	uint64_t	tba_addr;
 	uint64_t	tma_addr;
+	uint32_t	trap_en;
+	uint32_t	skip_process_ctx_clear;
 	uint32_t	is_kfd_process;
 	uint32_t	is_aql_queue;
 	uint32_t	queue_size;
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 861910a6662d9..c4e3cb8d44de3 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -202,17 +202,14 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes,
 	mes_add_queue_pkt.gws_size = input->gws_size;
 	mes_add_queue_pkt.trap_handler_addr = input->tba_addr;
 	mes_add_queue_pkt.tma_addr = input->tma_addr;
+	mes_add_queue_pkt.trap_en = input->trap_en;
+	mes_add_queue_pkt.skip_process_ctx_clear = input->skip_process_ctx_clear;
 	mes_add_queue_pkt.is_kfd_process = input->is_kfd_process;
 
 	/* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */
 	mes_add_queue_pkt.is_aql_queue = input->is_aql_queue;
 	mes_add_queue_pkt.gds_size = input->queue_size;
 
-	if (!(((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 4) &&
-		  (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) &&
-		  (adev->ip_versions[GC_HWIP][0] <= IP_VERSION(11, 0, 3))))
-		mes_add_queue_pkt.trap_en = 1;
-
 	/* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */
 	mes_add_queue_pkt.is_aql_queue = input->is_aql_queue;
 	mes_add_queue_pkt.gds_size = input->queue_size;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 826a99acb6fb9..d4df424e45144 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -537,8 +537,6 @@ static int kfd_ioctl_set_cu_mask(struct file *filp, struct kfd_process *p,
 		goto out;
 	}
 
-	minfo.update_flag = UPDATE_FLAG_CU_MASK;
-
 	mutex_lock(&p->mutex);
 
 	retval = pqm_update_mqd(&p->pqm, args->queue_id, &minfo);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 73b07b5f17f14..5e2ee2d1acc48 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -24,6 +24,57 @@
 #include "kfd_device_queue_manager.h"
 #include <linux/file.h>
 
+static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable)
+{
+	struct mqd_update_info minfo = {0};
+	int err;
+
+	if (!q)
+		return 0;
+
+	if (KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) ||
+	    KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0))
+		return 0;
+
+	if (enable && q->properties.is_user_cu_masked)
+		return -EBUSY;
+
+	minfo.update_flag = enable ? UPDATE_FLAG_DBG_WA_ENABLE : UPDATE_FLAG_DBG_WA_DISABLE;
+
+	q->properties.is_dbg_wa = enable;
+	err = q->device->dqm->ops.update_queue(q->device->dqm, q, &minfo);
+	if (err)
+		q->properties.is_dbg_wa = false;
+
+	return err;
+}
+
+static int kfd_dbg_set_workaround(struct kfd_process *target, bool enable)
+{
+	struct process_queue_manager *pqm = &target->pqm;
+	struct process_queue_node *pqn;
+	int r = 0;
+
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
+		r = kfd_dbg_set_queue_workaround(pqn->q, enable);
+		if (enable && r)
+			goto unwind;
+	}
+
+	return 0;
+
+unwind:
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list)
+		kfd_dbg_set_queue_workaround(pqn->q, false);
+
+	if (enable)
+		target->runtime_info.runtime_state = r == -EBUSY ?
+				DEBUG_RUNTIME_STATE_ENABLED_BUSY :
+				DEBUG_RUNTIME_STATE_ENABLED_ERROR;
+
+	return r;
+}
+
 static int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
 {
 	uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode;
@@ -77,6 +128,8 @@ static void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int
 		else
 			kfd_dbg_set_mes_debug_mode(pdd);
 	}
+
+	kfd_dbg_set_workaround(target, false);
 }
 
 int kfd_dbg_trap_disable(struct kfd_process *target)
@@ -111,6 +164,10 @@ static int kfd_dbg_trap_activate(struct kfd_process *target)
 {
 	int i, r = 0;
 
+	r = kfd_dbg_set_workaround(target, true);
+	if (r)
+		return r;
+
 	for (i = 0; i < target->n_pdds; i++) {
 		struct kfd_process_device *pdd = target->pdds[i];
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 17481f824647d..3e56225f6ef62 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -31,7 +31,8 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			uint32_t *runtime_info_size);
 static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 {
-	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2);
+	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
+	       KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0);
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 495c9238254e7..44d87943e40ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -226,6 +226,10 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q,
 	queue_input.paging = false;
 	queue_input.tba_addr = qpd->tba_addr;
 	queue_input.tma_addr = qpd->tma_addr;
+	queue_input.trap_en = KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) ||
+			      KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0) ||
+			      q->properties.is_dbg_wa;
+	queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled;
 
 	queue_type = convert_to_mes_queue_type(q->properties.type);
 	if (queue_type < 0) {
@@ -1716,6 +1720,9 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	 * updates the is_evicted flag but is a no-op otherwise.
 	 */
 	q->properties.is_evicted = !!qpd->evicted;
+	q->properties.is_dbg_wa = qpd->pqm->process->debug_trap_enabled &&
+			KFD_GC_VERSION(q->device) >= IP_VERSION(11, 0, 0) &&
+			KFD_GC_VERSION(q->device) < IP_VERSION(12, 0, 0);
 
 	if (qd)
 		mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
index eb11940bec34c..65c9f01a1f86c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c
@@ -48,8 +48,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
 	struct cik_mqd *m;
 	uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */
 
-	if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) ||
-	    !minfo->cu_mask.ptr)
+	if (!minfo || !minfo->cu_mask.ptr)
 		return;
 
 	mqd_symmetrically_map_cu_mask(mm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index eaaa4f4ddaaaf..a0ac4f2fe6b58 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -48,8 +48,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
 	struct v10_compute_mqd *m;
 	uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */
 
-	if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) ||
-	    !minfo->cu_mask.ptr)
+	if (!minfo || !minfo->cu_mask.ptr)
 		return;
 
 	mqd_symmetrically_map_cu_mask(mm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index 3a48bbc589fe2..9a9b4e8535162 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -46,15 +46,33 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
 {
 	struct v11_compute_mqd *m;
 	uint32_t se_mask[KFD_MAX_NUM_SE] = {0};
+	bool has_wa_flag = minfo && (minfo->update_flag & (UPDATE_FLAG_DBG_WA_ENABLE |
+			UPDATE_FLAG_DBG_WA_DISABLE));
 
-	if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) ||
-	    !minfo->cu_mask.ptr)
+	if (!minfo || !(has_wa_flag || minfo->cu_mask.ptr))
 		return;
 
+	m = get_mqd(mqd);
+
+	if (has_wa_flag) {
+		uint32_t wa_mask = minfo->update_flag == UPDATE_FLAG_DBG_WA_ENABLE ?
+						0xffff : 0xffffffff;
+
+		m->compute_static_thread_mgmt_se0 = wa_mask;
+		m->compute_static_thread_mgmt_se1 = wa_mask;
+		m->compute_static_thread_mgmt_se2 = wa_mask;
+		m->compute_static_thread_mgmt_se3 = wa_mask;
+		m->compute_static_thread_mgmt_se4 = wa_mask;
+		m->compute_static_thread_mgmt_se5 = wa_mask;
+		m->compute_static_thread_mgmt_se6 = wa_mask;
+		m->compute_static_thread_mgmt_se7 = wa_mask;
+
+		return;
+	}
+
 	mqd_symmetrically_map_cu_mask(mm,
 		minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask);
 
-	m = get_mqd(mqd);
 	m->compute_static_thread_mgmt_se0 = se_mask[0];
 	m->compute_static_thread_mgmt_se1 = se_mask[1];
 	m->compute_static_thread_mgmt_se2 = se_mask[2];
@@ -109,6 +127,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 	uint64_t addr;
 	struct v11_compute_mqd *m;
 	int size;
+	uint32_t wa_mask = q->is_dbg_wa ? 0xffff : 0xffffffff;
 
 	m = (struct v11_compute_mqd *) mqd_mem_obj->cpu_ptr;
 	addr = mqd_mem_obj->gpu_addr;
@@ -122,14 +141,15 @@ static void init_mqd(struct mqd_manager *mm, void **mqd,
 
 	m->header = 0xC0310800;
 	m->compute_pipelinestat_enable = 1;
-	m->compute_static_thread_mgmt_se0 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se1 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se2 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se3 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se4 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se5 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se6 = 0xFFFFFFFF;
-	m->compute_static_thread_mgmt_se7 = 0xFFFFFFFF;
+
+	m->compute_static_thread_mgmt_se0 = wa_mask;
+	m->compute_static_thread_mgmt_se1 = wa_mask;
+	m->compute_static_thread_mgmt_se2 = wa_mask;
+	m->compute_static_thread_mgmt_se3 = wa_mask;
+	m->compute_static_thread_mgmt_se4 = wa_mask;
+	m->compute_static_thread_mgmt_se5 = wa_mask;
+	m->compute_static_thread_mgmt_se6 = wa_mask;
+	m->compute_static_thread_mgmt_se7 = wa_mask;
 
 	m->cp_hqd_persistent_state = CP_HQD_PERSISTENT_STATE__PRELOAD_REQ_MASK |
 			0x55 << CP_HQD_PERSISTENT_STATE__PRELOAD_SIZE__SHIFT;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index b7c95158d4a0f..5b87c244e909f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -65,8 +65,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
 	struct v9_mqd *m;
 	uint32_t se_mask[KFD_MAX_NUM_SE] = {0};
 
-	if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) ||
-	    !minfo->cu_mask.ptr)
+	if (!minfo || !minfo->cu_mask.ptr)
 		return;
 
 	mqd_symmetrically_map_cu_mask(mm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
index fe69492b1bb39..d1e962da51dd3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c
@@ -51,8 +51,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd,
 	struct vi_mqd *m;
 	uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */
 
-	if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) ||
-	    !minfo->cu_mask.ptr)
+	if (!minfo || !minfo->cu_mask.ptr)
 		return;
 
 	mqd_symmetrically_map_cu_mask(mm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 8fca7175daabf..f0a45d184c8fc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -513,6 +513,8 @@ struct queue_properties {
 	bool is_active;
 	bool is_gws;
 	uint32_t pm4_target_xcc;
+	bool is_dbg_wa;
+	bool is_user_cu_masked;
 	/* Not relevant for user mode queues in cp scheduling */
 	unsigned int vmid;
 	/* Relevant only for sdma queues*/
@@ -535,7 +537,8 @@ struct queue_properties {
 			    !(q).is_evicted)
 
 enum mqd_update_flag {
-	UPDATE_FLAG_CU_MASK = 0,
+	UPDATE_FLAG_DBG_WA_ENABLE = 1,
+	UPDATE_FLAG_DBG_WA_DISABLE = 2,
 };
 
 struct mqd_update_info {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index b100933340d27..43d432b5c5bcb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -506,8 +506,12 @@ int pqm_update_mqd(struct process_queue_manager *pqm,
 		return -EFAULT;
 	}
 
+	/* CUs are masked for debugger requirements so deny user mask  */
+	if (pqn->q->properties.is_dbg_wa && minfo && minfo->cu_mask.ptr)
+		return -EBUSY;
+
 	/* ASICs that have WGPs must enforce pairwise enabled mask checks. */
-	if (minfo && minfo->update_flag == UPDATE_FLAG_CU_MASK && minfo->cu_mask.ptr &&
+	if (minfo && minfo->cu_mask.ptr &&
 			KFD_GC_VERSION(pqn->q->device) >= IP_VERSION(10, 0, 0)) {
 		int i;
 
@@ -526,6 +530,9 @@ int pqm_update_mqd(struct process_queue_manager *pqm,
 	if (retval != 0)
 		return retval;
 
+	if (minfo && minfo->cu_mask.ptr)
+		pqn->q->properties.is_user_cu_masked = true;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 3def25b2bdbbc..faa7939f35bde 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1863,10 +1863,13 @@ static void kfd_topology_set_dbg_firmware_support(struct kfd_topology_device *de
 {
 	bool firmware_supported = true;
 
+	/*
+	 * FIXME: GFX11 FW currently not sufficient to deal with CWSR WA.
+	 * Updated FW with API changes coming soon.
+	 */
 	if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(11, 0, 0) &&
 			KFD_GC_VERSION(dev->gpu) < IP_VERSION(12, 0, 0)) {
-		firmware_supported =
-			(dev->gpu->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 9;
+		firmware_supported = false;
 		goto out;
 	}
 
-- 
GitLab


From 44b87bb0836c65d1b9d21b01503eb6e9b9297771 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 6 Apr 2022 12:03:31 -0400
Subject: [PATCH 1534/1791] drm/amdkfd: add raise exception event function

Exception events can be generated from interrupts or queue activitity.

The raise event function will save exception status of a queue, device
or process then notify the debugger of the status change by writing to
a debugger polled file descriptor that the debugger provides during
debug attach.

For memory violation exceptions, extra exception data will be saved.

The debugger will be able to query the saved exception states by query
operation that will be provided by follow up patches.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 104 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |   7 ++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |  10 +++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c |   2 +
 4 files changed, 123 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 5e2ee2d1acc48..dccb27fc764b0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -24,6 +24,107 @@
 #include "kfd_device_queue_manager.h"
 #include <linux/file.h>
 
+void debug_event_write_work_handler(struct work_struct *work)
+{
+	struct kfd_process *process;
+
+	static const char write_data = '.';
+	loff_t pos = 0;
+
+	process = container_of(work,
+			struct kfd_process,
+			debug_event_workarea);
+
+	kernel_write(process->dbg_ev_file, &write_data, 1, &pos);
+}
+
+/* update process/device/queue exception status, write to descriptor
+ * only if exception_status is enabled.
+ */
+bool kfd_dbg_ev_raise(uint64_t event_mask,
+			struct kfd_process *process, struct kfd_node *dev,
+			unsigned int source_id, bool use_worker,
+			void *exception_data, size_t exception_data_size)
+{
+	struct process_queue_manager *pqm;
+	struct process_queue_node *pqn;
+	int i;
+	static const char write_data = '.';
+	loff_t pos = 0;
+	bool is_subscribed = true;
+
+	if (!(process && process->debug_trap_enabled))
+		return false;
+
+	mutex_lock(&process->event_mutex);
+
+	if (event_mask & KFD_EC_MASK_DEVICE) {
+		for (i = 0; i < process->n_pdds; i++) {
+			struct kfd_process_device *pdd = process->pdds[i];
+
+			if (pdd->dev != dev)
+				continue;
+
+			pdd->exception_status |= event_mask & KFD_EC_MASK_DEVICE;
+
+			if (event_mask & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) {
+				if (!pdd->vm_fault_exc_data) {
+					pdd->vm_fault_exc_data = kmemdup(
+							exception_data,
+							exception_data_size,
+							GFP_KERNEL);
+					if (!pdd->vm_fault_exc_data)
+						pr_debug("Failed to allocate exception data memory");
+				} else {
+					pr_debug("Debugger exception data not saved\n");
+					print_hex_dump_bytes("exception data: ",
+							DUMP_PREFIX_OFFSET,
+							exception_data,
+							exception_data_size);
+				}
+			}
+			break;
+		}
+	} else if (event_mask & KFD_EC_MASK_PROCESS) {
+		process->exception_status |= event_mask & KFD_EC_MASK_PROCESS;
+	} else {
+		pqm = &process->pqm;
+		list_for_each_entry(pqn, &pqm->queues,
+				process_queue_list) {
+			int target_id;
+
+			if (!pqn->q)
+				continue;
+
+			target_id = event_mask & KFD_EC_MASK(EC_QUEUE_NEW) ?
+					pqn->q->properties.queue_id :
+							pqn->q->doorbell_id;
+
+			if (pqn->q->device != dev || target_id != source_id)
+				continue;
+
+			pqn->q->properties.exception_status |= event_mask;
+			break;
+		}
+	}
+
+	if (process->exception_enable_mask & event_mask) {
+		if (use_worker)
+			schedule_work(&process->debug_event_workarea);
+		else
+			kernel_write(process->dbg_ev_file,
+					&write_data,
+					1,
+					&pos);
+	} else {
+		is_subscribed = false;
+	}
+
+	mutex_unlock(&process->event_mutex);
+
+	return is_subscribed;
+}
+
 static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable)
 {
 	struct mqd_update_info minfo = {0};
@@ -99,6 +200,9 @@ static void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int
 {
 	int i;
 
+	if (!unwind)
+		cancel_work_sync(&target->debug_event_workarea);
+
 	for (i = 0; i < target->n_pdds; i++) {
 		struct kfd_process_device *pdd = target->pdds[i];
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 3e56225f6ef62..66ee7b95d08a9 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -25,6 +25,11 @@
 
 #include "kfd_priv.h"
 
+bool kfd_dbg_ev_raise(uint64_t event_mask,
+			struct kfd_process *process, struct kfd_node *dev,
+			unsigned int source_id, bool use_worker,
+			void *exception_data,
+			size_t exception_data_size);
 int kfd_dbg_trap_disable(struct kfd_process *target);
 int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			void __user *runtime_info,
@@ -35,6 +40,8 @@ static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 	       KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0);
 }
 
+void debug_event_write_work_handler(struct work_struct *work);
+
 /*
  * If GFX off is enabled, chips that do not support RLC restore for the debug
  * registers will disable GFX off temporarily for the entire debug session.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index f0a45d184c8fc..b18cd4bf76bfb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -529,6 +529,7 @@ struct queue_properties {
 	uint32_t ctl_stack_size;
 	uint64_t tba_addr;
 	uint64_t tma_addr;
+	uint64_t exception_status;
 };
 
 #define QUEUE_IS_ACTIVE(q) ((q).queue_size > 0 &&	\
@@ -820,6 +821,11 @@ struct kfd_process_device {
 	uint64_t page_in;
 	uint64_t page_out;
 
+	/* Exception code status*/
+	uint64_t exception_status;
+	void *vm_fault_exc_data;
+	size_t vm_fault_exc_data_size;
+
 	/* Tracks debug per-vmid request settings */
 	uint32_t spi_dbg_override;
 	uint32_t spi_dbg_launch_mode;
@@ -955,12 +961,16 @@ struct kfd_process {
 
 	/* Exception code enable mask and status */
 	uint64_t exception_enable_mask;
+	uint64_t exception_status;
 
 	/* shared virtual memory registered by this process */
 	struct svm_range_list svms;
 
 	bool xnack_enabled;
 
+	/* Work area for debugger event writer worker. */
+	struct work_struct debug_event_workarea;
+
 	/* Tracks debug per-vmid request for debug flags */
 	bool dbg_flags;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index e77cadadb09b6..f904d6d6e01cd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1509,6 +1509,8 @@ static struct kfd_process *create_process(const struct task_struct *thread)
 	kfd_unref_process(process);
 	get_task_struct(process->lead_thread);
 
+	INIT_WORK(&process->debug_event_workarea, debug_event_write_work_handler);
+
 	return process;
 
 err_register_notifier:
-- 
GitLab


From c2d2588c702364ff53916ddd97e2b26fd4f4a317 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 8 Apr 2022 12:49:48 -0400
Subject: [PATCH 1535/1791] drm/amdkfd: add send exception operation

Add a debug operation that allows the debugger to send an exception
directly to runtime through a payload address.

For memory violations, normal vmfault signals will be applied to
notify runtime instead after passing in the saved exception data
when a memory violation was raised to the debugger.

For runtime exceptions, this will unblock the runtime enable
function which will be explained and implemented in a follow up
patch.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdkfd/cik_event_interrupt.c  |  4 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  5 ++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        | 43 +++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  6 ++
 drivers/gpu/drm/amd/amdkfd/kfd_events.c       |  3 +-
 .../gpu/drm/amd/amdkfd/kfd_int_process_v9.c   |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  7 +-
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      | 71 ++++++++++++++++++-
 8 files changed, 135 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
index 4ebfff6b6c55a..795382b55e0a9 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
@@ -118,9 +118,9 @@ static void cik_event_interrupt_wq(struct kfd_node *dev,
 			return;
 
 		if (info.vmid == vmid)
-			kfd_signal_vm_fault_event(dev, pasid, &info);
+			kfd_signal_vm_fault_event(dev, pasid, &info, NULL);
 		else
-			kfd_signal_vm_fault_event(dev, pasid, NULL);
+			kfd_signal_vm_fault_event(dev, pasid, NULL, NULL);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index d4df424e45144..5e57b3e96ff90 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2833,6 +2833,11 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		r = kfd_dbg_trap_disable(target);
 		break;
 	case KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT:
+		r = kfd_dbg_send_exception_to_runtime(target,
+				args->send_runtime_event.gpu_id,
+				args->send_runtime_event.queue_id,
+				args->send_runtime_event.exception_mask);
+		break;
 	case KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED:
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE:
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index dccb27fc764b0..61098975bb0ed 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -125,6 +125,49 @@ bool kfd_dbg_ev_raise(uint64_t event_mask,
 	return is_subscribed;
 }
 
+int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
+					unsigned int dev_id,
+					unsigned int queue_id,
+					uint64_t error_reason)
+{
+	if (error_reason & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) {
+		struct kfd_process_device *pdd = NULL;
+		struct kfd_hsa_memory_exception_data *data;
+		int i;
+
+		for (i = 0; i < p->n_pdds; i++) {
+			if (p->pdds[i]->dev->id == dev_id) {
+				pdd = p->pdds[i];
+				break;
+			}
+		}
+
+		if (!pdd)
+			return -ENODEV;
+
+		data = (struct kfd_hsa_memory_exception_data *)
+						pdd->vm_fault_exc_data;
+
+		kfd_dqm_evict_pasid(pdd->dev->dqm, p->pasid);
+		kfd_signal_vm_fault_event(pdd->dev, p->pasid, NULL, data);
+		error_reason &= ~KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION);
+	}
+
+	if (error_reason & (KFD_EC_MASK(EC_PROCESS_RUNTIME))) {
+		/*
+		 * block should only happen after the debugger receives runtime
+		 * enable notice.
+		 */
+		up(&p->runtime_enable_sema);
+		error_reason &= ~KFD_EC_MASK(EC_PROCESS_RUNTIME);
+	}
+
+	if (error_reason)
+		return kfd_send_exception_to_runtime(p, queue_id, error_reason);
+
+	return 0;
+}
+
 static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable)
 {
 	struct mqd_update_info minfo = {0};
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 66ee7b95d08a9..2c6866bb88506 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -34,6 +34,12 @@ int kfd_dbg_trap_disable(struct kfd_process *target);
 int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			void __user *runtime_info,
 			uint32_t *runtime_info_size);
+
+int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
+					unsigned int dev_id,
+					unsigned int queue_id,
+					uint64_t error_reason);
+
 static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 {
 	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 9926186f88a65..0a5e7b172a648 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -1222,7 +1222,8 @@ void kfd_signal_hw_exception_event(u32 pasid)
 }
 
 void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
-			       struct kfd_vm_fault_info *info)
+				struct kfd_vm_fault_info *info,
+				struct kfd_hsa_memory_exception_data *data)
 {
 	struct kfd_event *ev;
 	uint32_t id;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 861bccb1e9dc5..8cf58be80f4ea 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -362,7 +362,7 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 
 		kfd_smi_event_update_vmfault(dev, pasid);
 		kfd_dqm_evict_pasid(dev->dqm, pasid);
-		kfd_signal_vm_fault_event(dev, pasid, &info);
+		kfd_signal_vm_fault_event(dev, pasid, &info, NULL);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index b18cd4bf76bfb..58b82fa59584d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -979,6 +979,7 @@ struct kfd_process {
 	bool queues_paused;
 
 	/* Tracks runtime enable status */
+	struct semaphore runtime_enable_sema;
 	struct kfd_runtime_info runtime_info;
 
 };
@@ -1447,7 +1448,8 @@ int kfd_get_num_events(struct kfd_process *p);
 int kfd_event_destroy(struct kfd_process *p, uint32_t event_id);
 
 void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
-				struct kfd_vm_fault_info *info);
+				struct kfd_vm_fault_info *info,
+				struct kfd_hsa_memory_exception_data *data);
 
 void kfd_signal_reset_event(struct kfd_node *dev);
 
@@ -1463,6 +1465,9 @@ static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev)
 	       KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
 }
 
+int kfd_send_exception_to_runtime(struct kfd_process *p,
+				unsigned int queue_id,
+				uint64_t error_reason);
 bool kfd_is_locked(void);
 
 /* Compute profile */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index f904d6d6e01cd..5cbfcaf08c8f6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1462,6 +1462,7 @@ static struct kfd_process *create_process(const struct task_struct *thread)
 	process->debugger_process = NULL;
 	process->exception_enable_mask = 0;
 	atomic_set(&process->debugged_process_count, 0);
+	sema_init(&process->runtime_enable_sema, 0);
 
 	process->pasid = kfd_pasid_alloc();
 	if (process->pasid == 0) {
@@ -2120,6 +2121,75 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 	}
 }
 
+struct send_exception_work_handler_workarea {
+	struct work_struct work;
+	struct kfd_process *p;
+	unsigned int queue_id;
+	uint64_t error_reason;
+};
+
+static void send_exception_work_handler(struct work_struct *work)
+{
+	struct send_exception_work_handler_workarea *workarea;
+	struct kfd_process *p;
+	struct queue *q;
+	struct mm_struct *mm;
+	struct kfd_context_save_area_header __user *csa_header;
+	uint64_t __user *err_payload_ptr;
+	uint64_t cur_err;
+	uint32_t ev_id;
+
+	workarea = container_of(work,
+				struct send_exception_work_handler_workarea,
+				work);
+	p = workarea->p;
+
+	mm = get_task_mm(p->lead_thread);
+
+	if (!mm)
+		return;
+
+	kthread_use_mm(mm);
+
+	q = pqm_get_user_queue(&p->pqm, workarea->queue_id);
+
+	if (!q)
+		goto out;
+
+	csa_header = (void __user *)q->properties.ctx_save_restore_area_address;
+
+	get_user(err_payload_ptr, (uint64_t __user **)&csa_header->err_payload_addr);
+	get_user(cur_err, err_payload_ptr);
+	cur_err |= workarea->error_reason;
+	put_user(cur_err, err_payload_ptr);
+	get_user(ev_id, &csa_header->err_event_id);
+
+	kfd_set_event(p, ev_id);
+
+out:
+	kthread_unuse_mm(mm);
+	mmput(mm);
+}
+
+int kfd_send_exception_to_runtime(struct kfd_process *p,
+			unsigned int queue_id,
+			uint64_t error_reason)
+{
+	struct send_exception_work_handler_workarea worker;
+
+	INIT_WORK_ONSTACK(&worker.work, send_exception_work_handler);
+
+	worker.p = p;
+	worker.queue_id = queue_id;
+	worker.error_reason = error_reason;
+
+	schedule_work(&worker.work);
+	flush_work(&worker.work);
+	destroy_work_on_stack(&worker.work);
+
+	return 0;
+}
+
 struct kfd_process_device *kfd_process_device_data_by_id(struct kfd_process *p, uint32_t gpu_id)
 {
 	int i;
@@ -2179,4 +2249,3 @@ int kfd_debugfs_mqds_by_process(struct seq_file *m, void *data)
 }
 
 #endif
-
-- 
GitLab


From 455227c4642c5e1867213cea73a527e431779060 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 8 Apr 2022 13:12:24 -0400
Subject: [PATCH 1536/1791] drm/amdkfd: add runtime enable operation

The debugger can attach to a process prior to HSA enablement (i.e.
inferior is spawned by the debugger and attached to immediately before
target process has been enabled for HSA dispatches) or it
can attach to a running target that is already HSA enabled.  Either
way, the debugger needs to know the enablement status to know when
it can inspect queues.

For the scenario where the debugger spawns the target process,
it will have to wait for ROCr's runtime enable request from the target.
The runtime enable request will be able to see that its process has been
debug attached.  ROCr raises an EC_PROCESS_RUNTIME signal to the
debugger then blocks the target process while waiting the debugger's
response. Once the debugger has received the runtime signal, it will
unblock the target process.

For the scenario where the debugger attaches to a running target
process, ROCr will set the target process' runtime status as enabled so
that on an attach request, the debugger will be able to see this
status and will continue with debug enablement as normal.

A secondary requirement is to conditionally enable the trap tempories only
if the user requests it (env var HSA_ENABLE_DEBUG=1) or if the debugger
attaches with HSA runtime enabled.  This is because setting up the trap
temporaries incurs a performance overhead that is unacceptable for
microbench performance in normal mode for certain customers.

In the scenario where the debugger spawns the target process, when ROCr
detects that the debugger has attached during the runtime enable
request, it will enable the trap temporaries before it blocks the target
process while waiting for the debugger to respond.

In the scenario where the debugger attaches to a running target process,
it will enable to trap temporaries itself.

Finally, there is an additional restriction that is required to be
enforced with runtime enable and HW debug mode setting. The debugger must
first ensure that HW debug mode has been enabled before permitting HW debug
mode operations.

With single process debug devices, allowing the debugger to set debug
HW modes prior to trap activation means that debug HW mode setting can
occur before the KFD has reserved the debug VMID (0xf) from the hardware
scheduler's VMID allocation resource pool.  This can result in the
hardware scheduler assigning VMID 0xf to a non-debugged process and
having that process inherit debug HW mode settings intended for the
debugged target process instead, which is both incorrect and potentially
fatal for normal mode operation.

With multi process debug devices, allowing the debugger to set debug
HW modes prior to trap activation means that non-debugged processes
migrating to a new VMID could inherit unintended debug settings.

All debug operations that touch HW settings must require trap activation
where trap activation is triggered by both debug attach and runtime
enablement (target has KFD opened and is ready to dispatch work).

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 143 ++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   |   6 +-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |   4 +
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |   1 +
 4 files changed, 150 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 5e57b3e96ff90..615fa9ab36b7e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2738,11 +2738,140 @@ static int kfd_ioctl_criu(struct file *filep, struct kfd_process *p, void *data)
 	return ret;
 }
 
-static int kfd_ioctl_runtime_enable(struct file *filep, struct kfd_process *p, void *data)
+static int runtime_enable(struct kfd_process *p, uint64_t r_debug,
+			bool enable_ttmp_setup)
+{
+	int i = 0, ret = 0;
+
+	if (p->is_runtime_retry)
+		goto retry;
+
+	if (p->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_DISABLED)
+		return -EBUSY;
+
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+
+		if (pdd->qpd.queue_count)
+			return -EEXIST;
+	}
+
+	p->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_ENABLED;
+	p->runtime_info.r_debug = r_debug;
+	p->runtime_info.ttmp_setup = enable_ttmp_setup;
+
+	if (p->runtime_info.ttmp_setup) {
+		for (i = 0; i < p->n_pdds; i++) {
+			struct kfd_process_device *pdd = p->pdds[i];
+
+			if (!kfd_dbg_is_rlc_restore_supported(pdd->dev)) {
+				amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+				pdd->dev->kfd2kgd->enable_debug_trap(
+						pdd->dev->adev,
+						true,
+						pdd->dev->vm_info.last_vmid_kfd);
+			} else if (kfd_dbg_is_per_vmid_supported(pdd->dev)) {
+				pdd->spi_dbg_override = pdd->dev->kfd2kgd->enable_debug_trap(
+						pdd->dev->adev,
+						false,
+						0);
+			}
+		}
+	}
+
+retry:
+	if (p->debug_trap_enabled) {
+		if (!p->is_runtime_retry) {
+			kfd_dbg_trap_activate(p);
+			kfd_dbg_ev_raise(KFD_EC_MASK(EC_PROCESS_RUNTIME),
+					p, NULL, 0, false, NULL, 0);
+		}
+
+		mutex_unlock(&p->mutex);
+		ret = down_interruptible(&p->runtime_enable_sema);
+		mutex_lock(&p->mutex);
+
+		p->is_runtime_retry = !!ret;
+	}
+
+	return ret;
+}
+
+static int runtime_disable(struct kfd_process *p)
 {
+	int i = 0, ret;
+	bool was_enabled = p->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED;
+
+	p->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_DISABLED;
+	p->runtime_info.r_debug = 0;
+
+	if (p->debug_trap_enabled) {
+		if (was_enabled)
+			kfd_dbg_trap_deactivate(p, false, 0);
+
+		if (!p->is_runtime_retry)
+			kfd_dbg_ev_raise(KFD_EC_MASK(EC_PROCESS_RUNTIME),
+					p, NULL, 0, false, NULL, 0);
+
+		mutex_unlock(&p->mutex);
+		ret = down_interruptible(&p->runtime_enable_sema);
+		mutex_lock(&p->mutex);
+
+		p->is_runtime_retry = !!ret;
+		if (ret)
+			return ret;
+	}
+
+	if (was_enabled && p->runtime_info.ttmp_setup) {
+		for (i = 0; i < p->n_pdds; i++) {
+			struct kfd_process_device *pdd = p->pdds[i];
+
+			if (!kfd_dbg_is_rlc_restore_supported(pdd->dev))
+				amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+		}
+	}
+
+	p->runtime_info.ttmp_setup = false;
+
+	/* disable ttmp setup */
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+
+		if (kfd_dbg_is_per_vmid_supported(pdd->dev)) {
+			pdd->spi_dbg_override =
+					pdd->dev->kfd2kgd->disable_debug_trap(
+					pdd->dev->adev,
+					false,
+					pdd->dev->vm_info.last_vmid_kfd);
+
+			if (!pdd->dev->kfd->shared_resources.enable_mes)
+				debug_refresh_runlist(pdd->dev->dqm);
+			else
+				kfd_dbg_set_mes_debug_mode(pdd);
+		}
+	}
+
 	return 0;
 }
 
+static int kfd_ioctl_runtime_enable(struct file *filep, struct kfd_process *p, void *data)
+{
+	struct kfd_ioctl_runtime_enable_args *args = data;
+	int r;
+
+	mutex_lock(&p->mutex);
+
+	if (args->mode_mask & KFD_RUNTIME_ENABLE_MODE_ENABLE_MASK)
+		r = runtime_enable(p, args->r_debug,
+				!!(args->mode_mask & KFD_RUNTIME_ENABLE_MODE_TTMP_SAVE_MASK));
+	else
+		r = runtime_disable(p);
+
+	mutex_unlock(&p->mutex);
+
+	return r;
+}
+
 static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, void *data)
 {
 	struct kfd_ioctl_dbg_trap_args *args = data;
@@ -2815,6 +2944,18 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		goto unlock_out;
 	}
 
+	if (target->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_ENABLED &&
+			(args->op == KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE ||
+			 args->op == KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE ||
+			 args->op == KFD_IOC_DBG_TRAP_SUSPEND_QUEUES ||
+			 args->op == KFD_IOC_DBG_TRAP_RESUME_QUEUES ||
+			 args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH ||
+			 args->op == KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH ||
+			 args->op == KFD_IOC_DBG_TRAP_SET_FLAGS)) {
+		r = -EPERM;
+		goto unlock_out;
+	}
+
 	switch (args->op) {
 	case KFD_IOC_DBG_TRAP_ENABLE:
 		if (target != p)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 61098975bb0ed..a19c21d04438a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -219,7 +219,7 @@ static int kfd_dbg_set_workaround(struct kfd_process *target, bool enable)
 	return r;
 }
 
-static int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
+int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
 {
 	uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode;
 	uint32_t flags = pdd->process->dbg_flags;
@@ -239,7 +239,7 @@ static int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
  *				to unwind
  *		else: ignored
  */
-static void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count)
+void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count)
 {
 	int i;
 
@@ -307,7 +307,7 @@ int kfd_dbg_trap_disable(struct kfd_process *target)
 	return 0;
 }
 
-static int kfd_dbg_trap_activate(struct kfd_process *target)
+int kfd_dbg_trap_activate(struct kfd_process *target)
 {
 	int i, r = 0;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 2c6866bb88506..fca9285649483 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -25,6 +25,8 @@
 
 #include "kfd_priv.h"
 
+void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count);
+int kfd_dbg_trap_activate(struct kfd_process *target);
 bool kfd_dbg_ev_raise(uint64_t event_mask,
 			struct kfd_process *process, struct kfd_node *dev,
 			unsigned int source_id, bool use_worker,
@@ -77,4 +79,6 @@ static inline bool kfd_dbg_has_gws_support(struct kfd_node *dev)
 	/* Assume debugging and cooperative launch supported otherwise. */
 	return true;
 }
+
+int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd);
 #endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 58b82fa59584d..4b80f74b9de04 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -980,6 +980,7 @@ struct kfd_process {
 
 	/* Tracks runtime enable status */
 	struct semaphore runtime_enable_sema;
+	bool is_runtime_retry;
 	struct kfd_runtime_info runtime_info;
 
 };
-- 
GitLab


From 50cff45e274896235d371f16eab67a180e12a732 Mon Sep 17 00:00:00 2001
From: Jay Cornwall <jay.cornwall@amd.com>
Date: Mon, 1 Mar 2021 18:34:39 -0600
Subject: [PATCH 1537/1791] drm/amdkfd: add debug trap enabled flag to tma

Trap handler behavior will differ when a debugger is attached.

Make the debug trap flag available in the trap handler TMA.
Update it when the debug trap ioctl is invoked.

Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 11 +++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |  2 ++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c | 15 +++++++++++++++
 3 files changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index a19c21d04438a..17e8e9edccbf0 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -256,6 +256,8 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 		if (unwind && i == unwind_count)
 			break;
 
+		kfd_process_set_trap_debug_flag(&pdd->qpd, false);
+
 		/* GFX off is already disabled by debug activate if not RLC restore supported. */
 		if (kfd_dbg_is_rlc_restore_supported(pdd->dev))
 			amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
@@ -351,6 +353,15 @@ int kfd_dbg_trap_activate(struct kfd_process *target)
 		if (kfd_dbg_is_rlc_restore_supported(pdd->dev))
 			amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
 
+		/*
+		 * Setting the debug flag in the trap handler requires that the TMA has been
+		 * allocated, which occurs during CWSR initialization.
+		 * In the event that CWSR has not been initialized at this point, setting the
+		 * flag will be called again during CWSR initialization if the target process
+		 * is still debug enabled.
+		 */
+		kfd_process_set_trap_debug_flag(&pdd->qpd, true);
+
 		if (!pdd->dev->kfd->shared_resources.enable_mes)
 			r = debug_refresh_runlist(pdd->dev->dqm);
 		else
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 4b80f74b9de04..a02fb939614af 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1157,6 +1157,8 @@ int kfd_init_apertures(struct kfd_process *process);
 void kfd_process_set_trap_handler(struct qcm_process_device *qpd,
 				  uint64_t tba_addr,
 				  uint64_t tma_addr);
+void kfd_process_set_trap_debug_flag(struct qcm_process_device *qpd,
+				     bool enabled);
 
 /* CWSR initialization */
 int kfd_process_init_cwsr_apu(struct kfd_process *process, struct file *filep);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 5cbfcaf08c8f6..3b7f219c9d06f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1309,6 +1309,8 @@ int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep)
 
 		memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size);
 
+		kfd_process_set_trap_debug_flag(qpd, p->debug_trap_enabled);
+
 		qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET;
 		pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n",
 			qpd->tba_addr, qpd->tma_addr, qpd->cwsr_kaddr);
@@ -1345,6 +1347,9 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd)
 
 	memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size);
 
+	kfd_process_set_trap_debug_flag(&pdd->qpd,
+					pdd->process->debug_trap_enabled);
+
 	qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET;
 	pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n",
 		 qpd->tba_addr, qpd->tma_addr, qpd->cwsr_kaddr);
@@ -1431,6 +1436,16 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported)
 	return true;
 }
 
+void kfd_process_set_trap_debug_flag(struct qcm_process_device *qpd,
+				     bool enabled)
+{
+	if (qpd->cwsr_kaddr) {
+		uint64_t *tma =
+			(uint64_t *)(qpd->cwsr_kaddr + KFD_CWSR_TMA_OFFSET);
+		tma[2] = enabled;
+	}
+}
+
 /*
  * On return the kfd_process is fully operational and will be freed when the
  * mm is released
-- 
GitLab


From a37d23f816b18a324c24d066d5bc453308913bf9 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Tue, 11 Apr 2023 11:25:52 +0800
Subject: [PATCH 1538/1791] drm/amd/pm: update SMU13 header files for coming OD
 support

Correct the data structures for OD feature support.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h  | 16 +++++-----
 .../inc/pmfw_if/smu13_driver_if_v13_0_0.h     | 18 +++++++-----
 .../inc/pmfw_if/smu13_driver_if_v13_0_7.h     | 29 ++++---------------
 .../amd/pm/swsmu/inc/smu_v13_0_7_pptable.h    | 16 +++++-----
 4 files changed, 34 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h
index 566a0da59e530..1dc7a065a6d48 100644
--- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h
+++ b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h
@@ -38,13 +38,12 @@
 #define SMU_13_0_0_PP_THERMALCONTROLLER_NONE 0
 #define SMU_13_0_0_PP_THERMALCONTROLLER_NAVI21 28
 
-#define SMU_13_0_0_PP_OVERDRIVE_VERSION 0x81        // OverDrive 8 Table Version 0.2
+#define SMU_13_0_0_PP_OVERDRIVE_VERSION 0x83        // OverDrive 8 Table Version 0.2
 #define SMU_13_0_0_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00
 
 enum SMU_13_0_0_ODFEATURE_CAP
 {
     SMU_13_0_0_ODCAP_GFXCLK_LIMITS = 0,
-    SMU_13_0_0_ODCAP_GFXCLK_CURVE,
     SMU_13_0_0_ODCAP_UCLK_LIMITS,
     SMU_13_0_0_ODCAP_POWER_LIMIT,
     SMU_13_0_0_ODCAP_FAN_ACOUSTIC_LIMIT,
@@ -59,13 +58,13 @@ enum SMU_13_0_0_ODFEATURE_CAP
     SMU_13_0_0_ODCAP_FAN_CURVE,
     SMU_13_0_0_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT,
     SMU_13_0_0_ODCAP_POWER_MODE,
+    SMU_13_0_0_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET,
     SMU_13_0_0_ODCAP_COUNT,
 };
 
 enum SMU_13_0_0_ODFEATURE_ID
 {
     SMU_13_0_0_ODFEATURE_GFXCLK_LIMITS           = 1 << SMU_13_0_0_ODCAP_GFXCLK_LIMITS,           //GFXCLK Limit feature
-    SMU_13_0_0_ODFEATURE_GFXCLK_CURVE            = 1 << SMU_13_0_0_ODCAP_GFXCLK_CURVE,            //GFXCLK Curve feature
     SMU_13_0_0_ODFEATURE_UCLK_LIMITS             = 1 << SMU_13_0_0_ODCAP_UCLK_LIMITS,             //UCLK Limit feature
     SMU_13_0_0_ODFEATURE_POWER_LIMIT             = 1 << SMU_13_0_0_ODCAP_POWER_LIMIT,             //Power Limit feature
     SMU_13_0_0_ODFEATURE_FAN_ACOUSTIC_LIMIT      = 1 << SMU_13_0_0_ODCAP_FAN_ACOUSTIC_LIMIT,      //Fan Acoustic RPM feature
@@ -80,6 +79,7 @@ enum SMU_13_0_0_ODFEATURE_ID
     SMU_13_0_0_ODFEATURE_FAN_CURVE               = 1 << SMU_13_0_0_ODCAP_FAN_CURVE,               //Fan Curve feature
     SMU_13_0_0_ODFEATURE_AUTO_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_0_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, //Auto Fan Acoustic RPM feature
     SMU_13_0_0_ODFEATURE_POWER_MODE              = 1 << SMU_13_0_0_ODCAP_POWER_MODE,              //Optimized GPU Power Mode feature
+    SMU_13_0_0_ODFEATURE_PER_ZONE_GFX_VOLTAGE_OFFSET  = 1 << SMU_13_0_0_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET,  //Perzone voltage offset feature
     SMU_13_0_0_ODFEATURE_COUNT                   = 16,
 };
 
@@ -89,10 +89,6 @@ enum SMU_13_0_0_ODSETTING_ID
 {
     SMU_13_0_0_ODSETTING_GFXCLKFMAX = 0,
     SMU_13_0_0_ODSETTING_GFXCLKFMIN,
-    SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_A,
-    SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_B,
-    SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_C,
-    SMU_13_0_0_ODSETTING_CUSTOM_CURVE_VFT_FMIN,
     SMU_13_0_0_ODSETTING_UCLKFMIN,
     SMU_13_0_0_ODSETTING_UCLKFMAX,
     SMU_13_0_0_ODSETTING_POWERPERCENTAGE,
@@ -117,6 +113,12 @@ enum SMU_13_0_0_ODSETTING_ID
     SMU_13_0_0_ODSETTING_FAN_CURVE_SPEED_5,
     SMU_13_0_0_ODSETTING_AUTO_FAN_ACOUSTIC_LIMIT,
     SMU_13_0_0_ODSETTING_POWER_MODE,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_1,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_2,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_3,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_4,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_5,
+    SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_6,
     SMU_13_0_0_ODSETTING_COUNT,
 };
 #define SMU_13_0_0_MAX_ODSETTING 64 //Maximum Number of ODSettings
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
index fe995651c6f55..e656e82a0154a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
@@ -667,7 +667,14 @@ typedef enum {
 
 #define PP_NUM_RTAVFS_PWL_ZONES 5
 
-
+#define PP_OD_FEATURE_GFX_VF_CURVE_BIT  0
+#define PP_OD_FEATURE_PPT_BIT       2
+#define PP_OD_FEATURE_FAN_CURVE_BIT 3
+#define PP_OD_FEATURE_GFXCLK_BIT      7
+#define PP_OD_FEATURE_UCLK_BIT      8
+#define PP_OD_FEATURE_ZERO_FAN_BIT      9
+#define PP_OD_FEATURE_TEMPERATURE_BIT 10
+#define PP_OD_FEATURE_COUNT 13
 
 // VBIOS or PPLIB configures telemetry slope and offset. Only slope expected to be set for SVI3
 // Slope Q1.7, Offset Q1.2
@@ -689,10 +696,8 @@ typedef struct {
 
   //Voltage control
   int16_t                VoltageOffsetPerZoneBoundary[PP_NUM_OD_VF_CURVE_POINTS];
-  uint16_t               VddGfxVmax;         // in mV
 
-  uint8_t                IdlePwrSavingFeaturesCtrl;
-  uint8_t                RuntimePwrSavingFeaturesCtrl;
+  uint32_t               Reserved;
 
   //Frequency changes
   int16_t                GfxclkFmin;           // MHz
@@ -729,10 +734,9 @@ typedef struct {
   uint32_t FeatureCtrlMask;
 
   int16_t VoltageOffsetPerZoneBoundary;
-  uint16_t               VddGfxVmax;         // in mV
+  uint16_t               Reserved1;
 
-  uint8_t                IdlePwrSavingFeaturesCtrl;
-  uint8_t                RuntimePwrSavingFeaturesCtrl;
+  uint16_t               Reserved2;
 
   int16_t               GfxclkFmin;           // MHz
   int16_t               GfxclkFmax;           // MHz
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
index 44e879c51cae9..62b7c0daff687 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h
@@ -683,18 +683,12 @@ typedef struct {
 
 
 #define PP_OD_FEATURE_GFX_VF_CURVE_BIT  0
-#define PP_OD_FEATURE_VMAX_BIT      1
 #define PP_OD_FEATURE_PPT_BIT       2
 #define PP_OD_FEATURE_FAN_CURVE_BIT 3
-#define PP_OD_FEATURE_FREQ_DETER_BIT 4
-#define PP_OD_FEATURE_FULL_CTRL_BIT 5
-#define PP_OD_FEATURE_TDC_BIT      6
 #define PP_OD_FEATURE_GFXCLK_BIT      7
 #define PP_OD_FEATURE_UCLK_BIT      8
 #define PP_OD_FEATURE_ZERO_FAN_BIT      9
 #define PP_OD_FEATURE_TEMPERATURE_BIT 10
-#define PP_OD_FEATURE_POWER_FEATURE_CTRL_BIT 11
-#define PP_OD_FEATURE_ASIC_TDC_BIT 12
 #define PP_OD_FEATURE_COUNT 13
 
 typedef enum {
@@ -713,10 +707,8 @@ typedef struct {
 
   //Voltage control
   int16_t                VoltageOffsetPerZoneBoundary[PP_NUM_OD_VF_CURVE_POINTS];
-  uint16_t               VddGfxVmax;         // in mV
 
-  uint8_t                IdlePwrSavingFeaturesCtrl;
-  uint8_t                RuntimePwrSavingFeaturesCtrl;
+  uint32_t               Reserved;
 
   //Frequency changes
   int16_t                GfxclkFmin;           // MHz
@@ -741,12 +733,7 @@ typedef struct {
   uint8_t                MaxOpTemp;
   uint8_t                Padding[4];
 
-  uint16_t               GfxVoltageFullCtrlMode;
-  uint16_t               GfxclkFullCtrlMode;
-  uint16_t               UclkFullCtrlMode;
-  int16_t                AsicTdc;
-
-  uint32_t               Spare[10];
+  uint32_t               Spare[12];
   uint32_t               MmHubPadding[8]; // SMU internal use. Adding here instead of external as a workaround
 } OverDriveTable_t;
 
@@ -759,10 +746,9 @@ typedef struct {
   uint32_t FeatureCtrlMask;
 
   int16_t VoltageOffsetPerZoneBoundary;
-  uint16_t               VddGfxVmax;         // in mV
+  uint16_t               Reserved1;
 
-  uint8_t                IdlePwrSavingFeaturesCtrl;
-  uint8_t                RuntimePwrSavingFeaturesCtrl;
+  uint16_t               Reserved2;
 
   int16_t                GfxclkFmin;           // MHz
   int16_t                GfxclkFmax;           // MHz
@@ -785,12 +771,7 @@ typedef struct {
   uint8_t                MaxOpTemp;
   uint8_t                Padding[4];
 
-  uint16_t               GfxVoltageFullCtrlMode;
-  uint16_t               GfxclkFullCtrlMode;
-  uint16_t               UclkFullCtrlMode;
-  int16_t                AsicTdc;
-
-  uint32_t               Spare[10];
+  uint32_t               Spare[12];
 
 } OverDriveLimits_t;
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h
index 478862ded0bde..eadbe0149cae5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h
@@ -38,13 +38,12 @@
 #define SMU_13_0_7_PP_THERMALCONTROLLER_NONE 0
 #define SMU_13_0_7_PP_THERMALCONTROLLER_NAVI21 28
 
-#define SMU_13_0_7_PP_OVERDRIVE_VERSION 0x81        // OverDrive 8 Table Version 0.2
+#define SMU_13_0_7_PP_OVERDRIVE_VERSION 0x83        // OverDrive 8 Table Version 0.2
 #define SMU_13_0_7_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00
 
 enum SMU_13_0_7_ODFEATURE_CAP
 {
     SMU_13_0_7_ODCAP_GFXCLK_LIMITS = 0,
-    SMU_13_0_7_ODCAP_GFXCLK_CURVE,
     SMU_13_0_7_ODCAP_UCLK_LIMITS,
     SMU_13_0_7_ODCAP_POWER_LIMIT,
     SMU_13_0_7_ODCAP_FAN_ACOUSTIC_LIMIT,
@@ -59,13 +58,13 @@ enum SMU_13_0_7_ODFEATURE_CAP
     SMU_13_0_7_ODCAP_FAN_CURVE,
     SMU_13_0_7_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT,
     SMU_13_0_7_ODCAP_POWER_MODE,
+    SMU_13_0_7_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET,
     SMU_13_0_7_ODCAP_COUNT,
 };
 
 enum SMU_13_0_7_ODFEATURE_ID
 {
     SMU_13_0_7_ODFEATURE_GFXCLK_LIMITS           = 1 << SMU_13_0_7_ODCAP_GFXCLK_LIMITS,           //GFXCLK Limit feature
-    SMU_13_0_7_ODFEATURE_GFXCLK_CURVE            = 1 << SMU_13_0_7_ODCAP_GFXCLK_CURVE,            //GFXCLK Curve feature
     SMU_13_0_7_ODFEATURE_UCLK_LIMITS             = 1 << SMU_13_0_7_ODCAP_UCLK_LIMITS,             //UCLK Limit feature
     SMU_13_0_7_ODFEATURE_POWER_LIMIT             = 1 << SMU_13_0_7_ODCAP_POWER_LIMIT,             //Power Limit feature
     SMU_13_0_7_ODFEATURE_FAN_ACOUSTIC_LIMIT      = 1 << SMU_13_0_7_ODCAP_FAN_ACOUSTIC_LIMIT,      //Fan Acoustic RPM feature
@@ -80,6 +79,7 @@ enum SMU_13_0_7_ODFEATURE_ID
     SMU_13_0_7_ODFEATURE_FAN_CURVE               = 1 << SMU_13_0_7_ODCAP_FAN_CURVE,               //Fan Curve feature
     SMU_13_0_7_ODFEATURE_AUTO_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_7_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, //Auto Fan Acoustic RPM feature
     SMU_13_0_7_ODFEATURE_POWER_MODE              = 1 << SMU_13_0_7_ODCAP_POWER_MODE,              //Optimized GPU Power Mode feature
+    SMU_13_0_7_ODFEATURE_PER_ZONE_GFX_VOLTAGE_OFFSET  = 1 << SMU_13_0_7_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET,  //Perzone voltage offset feature
     SMU_13_0_7_ODFEATURE_COUNT                   = 16,
 };
 
@@ -89,10 +89,6 @@ enum SMU_13_0_7_ODSETTING_ID
 {
     SMU_13_0_7_ODSETTING_GFXCLKFMAX = 0,
     SMU_13_0_7_ODSETTING_GFXCLKFMIN,
-    SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_A,
-    SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_B,
-    SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_C,
-    SMU_13_0_7_ODSETTING_CUSTOM_CURVE_VFT_FMIN,
     SMU_13_0_7_ODSETTING_UCLKFMIN,
     SMU_13_0_7_ODSETTING_UCLKFMAX,
     SMU_13_0_7_ODSETTING_POWERPERCENTAGE,
@@ -117,6 +113,12 @@ enum SMU_13_0_7_ODSETTING_ID
     SMU_13_0_7_ODSETTING_FAN_CURVE_SPEED_5,
     SMU_13_0_7_ODSETTING_AUTO_FAN_ACOUSTIC_LIMIT,
     SMU_13_0_7_ODSETTING_POWER_MODE,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_1,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_2,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_3,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_4,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_5,
+    SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_6,
     SMU_13_0_7_ODSETTING_COUNT,
 };
 #define SMU_13_0_7_MAX_ODSETTING 64 //Maximum Number of ODSettings
-- 
GitLab


From 12fb1ad70d65edc3405884792d044fa79df7244f Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 22 Apr 2022 12:26:18 -0400
Subject: [PATCH 1539/1791] drm/amdkfd: update process interrupt handling for
 debug events

The debugger must be notified by any debugger subscribed exception
that comes from hardware interrupts.

If a debugger session exits, any exceptions it subscribed to may still
have interrupts in the interrupt ring buffer or KGD/KFD pipeline.
To prevent a new session from inheriting stale interrupts, when a new
queue is created, open an interrupt drain and allow the IH ring to drain
from a timestamped checkpoint.  Then inject a custom IV so that once
the custom IV is picked up by the KFD, it's safe to close the drain
and proceed with queue creation.

The drain must also be on debug disable as SW interrupts may still
be processed.  Drain at this time and clear all the exception status.

The debugger may also not be attached nor subscibed to certain
exceptions so forward them directly to the runtime.

GFX10 also requires its own IV processing, hence the creation of
kfd_int_process_v10.c.  This is because the IV from SQ interrupts are
packed into a new continguous format unlike GFX9. To make this clear,
a separate interrupting handling code file was created.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c    |  16 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h    |   2 +
 drivers/gpu/drm/amd/amdkfd/Makefile           |   1 +
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        |  84 ++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |   6 +
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |   4 +-
 .../gpu/drm/amd/amdkfd/kfd_int_process_v10.c  | 405 ++++++++++++++++++
 .../gpu/drm/amd/amdkfd/kfd_int_process_v11.c  |  26 +-
 .../gpu/drm/amd/amdkfd/kfd_int_process_v9.c   |  98 ++++-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  12 +
 drivers/gpu/drm/amd/amdkfd/kfd_process.c      |  47 ++
 .../amd/amdkfd/kfd_process_queue_manager.c    |   4 +
 12 files changed, 686 insertions(+), 19 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 66f80b9ab0c5d..98cd52bb005fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -777,6 +777,22 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bo
 	amdgpu_umc_poison_handler(adev, reset);
 }
 
+int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev,
+					uint32_t *payload)
+{
+	int ret;
+
+	/* Device or IH ring is not ready so bail. */
+	ret = amdgpu_ih_wait_on_checkpoint_process_ts(adev, &adev->irq.ih);
+	if (ret)
+		return ret;
+
+	/* Send payload to fence KFD interrupts */
+	amdgpu_amdkfd_interrupt(adev, payload);
+
+	return 0;
+}
+
 bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev)
 {
 	if (adev->gfx.ras && adev->gfx.ras->query_utcl2_poison_status)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 94cc456761e57..dd740e64e6e15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -250,6 +250,8 @@ int amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(struct amdgpu_device *dst,
 					    struct amdgpu_device *src,
 					    bool is_min);
 int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_min);
+int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev,
+					uint32_t *payload);
 
 /* Read user wptr from a specified user address space with page fault
  * disabled. The memory must be pinned and mapped to the hardware when
diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile
index 7477544280736..2ec8f27c5366c 100644
--- a/drivers/gpu/drm/amd/amdkfd/Makefile
+++ b/drivers/gpu/drm/amd/amdkfd/Makefile
@@ -53,6 +53,7 @@ AMDKFD_FILES	:= $(AMDKFD_PATH)/kfd_module.o \
 		$(AMDKFD_PATH)/kfd_events.o \
 		$(AMDKFD_PATH)/cik_event_interrupt.o \
 		$(AMDKFD_PATH)/kfd_int_process_v9.o \
+		$(AMDKFD_PATH)/kfd_int_process_v10.o \
 		$(AMDKFD_PATH)/kfd_int_process_v11.o \
 		$(AMDKFD_PATH)/kfd_smi_events.o \
 		$(AMDKFD_PATH)/kfd_crat.o \
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 17e8e9edccbf0..68b657398d417 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -125,6 +125,64 @@ bool kfd_dbg_ev_raise(uint64_t event_mask,
 	return is_subscribed;
 }
 
+/* set pending event queue entry from ring entry  */
+bool kfd_set_dbg_ev_from_interrupt(struct kfd_node *dev,
+				   unsigned int pasid,
+				   uint32_t doorbell_id,
+				   uint64_t trap_mask,
+				   void *exception_data,
+				   size_t exception_data_size)
+{
+	struct kfd_process *p;
+	bool signaled_to_debugger_or_runtime = false;
+
+	p = kfd_lookup_process_by_pasid(pasid);
+
+	if (!p)
+		return false;
+
+	if (!kfd_dbg_ev_raise(trap_mask, p, dev, doorbell_id, true,
+			      exception_data, exception_data_size)) {
+		struct process_queue_manager *pqm;
+		struct process_queue_node *pqn;
+
+		if (!!(trap_mask & KFD_EC_MASK_QUEUE) &&
+		       p->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) {
+			mutex_lock(&p->mutex);
+
+			pqm = &p->pqm;
+			list_for_each_entry(pqn, &pqm->queues,
+							process_queue_list) {
+
+				if (!(pqn->q && pqn->q->device == dev &&
+				      pqn->q->doorbell_id == doorbell_id))
+					continue;
+
+				kfd_send_exception_to_runtime(p, pqn->q->properties.queue_id,
+							      trap_mask);
+
+				signaled_to_debugger_or_runtime = true;
+
+				break;
+			}
+
+			mutex_unlock(&p->mutex);
+		} else if (trap_mask & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) {
+			kfd_dqm_evict_pasid(dev->dqm, p->pasid);
+			kfd_signal_vm_fault_event(dev, p->pasid, NULL,
+							exception_data);
+
+			signaled_to_debugger_or_runtime = true;
+		}
+	} else {
+		signaled_to_debugger_or_runtime = true;
+	}
+
+	kfd_unref_process(p);
+
+	return signaled_to_debugger_or_runtime;
+}
+
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
 					unsigned int queue_id,
@@ -281,6 +339,31 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 	kfd_dbg_set_workaround(target, false);
 }
 
+static void kfd_dbg_clean_exception_status(struct kfd_process *target)
+{
+	struct process_queue_manager *pqm;
+	struct process_queue_node *pqn;
+	int i;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		kfd_process_drain_interrupts(pdd);
+
+		pdd->exception_status = 0;
+	}
+
+	pqm = &target->pqm;
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
+		if (!pqn->q)
+			continue;
+
+		pqn->q->properties.exception_status = 0;
+	}
+
+	target->exception_status = 0;
+}
+
 int kfd_dbg_trap_disable(struct kfd_process *target)
 {
 	if (!target->debug_trap_enabled)
@@ -304,6 +387,7 @@ int kfd_dbg_trap_disable(struct kfd_process *target)
 	}
 
 	target->debug_trap_enabled = false;
+	kfd_dbg_clean_exception_status(target);
 	kfd_unref_process(target);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index fca9285649483..5153ccbd7fd17 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -27,6 +27,12 @@
 
 void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count);
 int kfd_dbg_trap_activate(struct kfd_process *target);
+bool kfd_set_dbg_ev_from_interrupt(struct kfd_node *dev,
+				   unsigned int pasid,
+				   uint32_t doorbell_id,
+				   uint64_t trap_mask,
+				   void *exception_data,
+				   size_t exception_data_size);
 bool kfd_dbg_ev_raise(uint64_t event_mask,
 			struct kfd_process *process, struct kfd_node *dev,
 			unsigned int source_id, bool use_worker,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index f0ed6e6416c37..2c36bb5786338 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -140,6 +140,8 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd)
 	case IP_VERSION(9, 4, 1): /* ARCTURUS */
 	case IP_VERSION(9, 4, 2): /* ALDEBARAN */
 	case IP_VERSION(9, 4, 3): /* GC 9.4.3 */
+		kfd->device_info.event_interrupt_class = &event_interrupt_class_v9;
+		break;
 	case IP_VERSION(10, 3, 1): /* VANGOGH */
 	case IP_VERSION(10, 3, 3): /* YELLOW_CARP */
 	case IP_VERSION(10, 3, 6): /* GC 10.3.6 */
@@ -153,7 +155,7 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd)
 	case IP_VERSION(10, 3, 2): /* NAVY_FLOUNDER */
 	case IP_VERSION(10, 3, 4): /* DIMGREY_CAVEFISH */
 	case IP_VERSION(10, 3, 5): /* BEIGE_GOBY */
-		kfd->device_info.event_interrupt_class = &event_interrupt_class_v9;
+		kfd->device_info.event_interrupt_class = &event_interrupt_class_v10;
 		break;
 	case IP_VERSION(11, 0, 0):
 	case IP_VERSION(11, 0, 1):
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
new file mode 100644
index 0000000000000..c7991e07b6be5
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
@@ -0,0 +1,405 @@
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "kfd_events.h"
+#include "kfd_debug.h"
+#include "soc15_int.h"
+#include "kfd_device_queue_manager.h"
+
+/*
+ * GFX10 SQ Interrupts
+ *
+ * There are 3 encoding types of interrupts sourced from SQ sent as a 44-bit
+ * packet to the Interrupt Handler:
+ * Auto - Generated by the SQG (various cmd overflows, timestamps etc)
+ * Wave - Generated by S_SENDMSG through a shader program
+ * Error - HW generated errors (Illegal instructions, Memviols, EDC etc)
+ *
+ * The 44-bit packet is mapped as {context_id1[7:0],context_id0[31:0]} plus
+ * 4-bits for VMID (SOC15_VMID_FROM_IH_ENTRY) as such:
+ *
+ * - context_id1[7:6]
+ * Encoding type (0 = Auto, 1 = Wave, 2 = Error)
+ *
+ * - context_id0[24]
+ * PRIV bit indicates that Wave S_SEND or error occurred within trap
+ *
+ * - context_id0[22:0]
+ * 23-bit data with the following layout per encoding type:
+ * Auto - only context_id0[8:0] is used, which reports various interrupts
+ * generated by SQG.  The rest is 0.
+ * Wave - user data sent from m0 via S_SENDMSG
+ * Error - Error type (context_id0[22:19]), Error Details (rest of bits)
+ *
+ * The other context_id bits show coordinates (SE/SH/CU/SIMD/WGP) for wave
+ * S_SENDMSG and Errors.  These are 0 for Auto.
+ */
+
+enum SQ_INTERRUPT_WORD_ENCODING {
+	SQ_INTERRUPT_WORD_ENCODING_AUTO = 0x0,
+	SQ_INTERRUPT_WORD_ENCODING_INST,
+	SQ_INTERRUPT_WORD_ENCODING_ERROR,
+};
+
+enum SQ_INTERRUPT_ERROR_TYPE {
+	SQ_INTERRUPT_ERROR_TYPE_EDC_FUE = 0x0,
+	SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST,
+	SQ_INTERRUPT_ERROR_TYPE_MEMVIOL,
+	SQ_INTERRUPT_ERROR_TYPE_EDC_FED,
+};
+
+/* SQ_INTERRUPT_WORD_AUTO_CTXID */
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE__SHIFT 0
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__WLT__SHIFT 1
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF0_FULL__SHIFT 2
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF1_FULL__SHIFT 3
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_UTC_ERROR__SHIFT 7
+#define SQ_INTERRUPT_WORD_AUTO_CTXID1__SE_ID__SHIFT 4
+#define SQ_INTERRUPT_WORD_AUTO_CTXID1__ENCODING__SHIFT 6
+
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_MASK 0x00000001
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__WLT_MASK 0x00000002
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF0_FULL_MASK 0x00000004
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF1_FULL_MASK 0x00000008
+#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_UTC_ERROR_MASK 0x00000080
+#define SQ_INTERRUPT_WORD_AUTO_CTXID1__SE_ID_MASK 0x030
+#define SQ_INTERRUPT_WORD_AUTO_CTXID1__ENCODING_MASK 0x0c0
+
+/* SQ_INTERRUPT_WORD_WAVE_CTXID */
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__DATA__SHIFT 0
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SA_ID__SHIFT 23
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV__SHIFT 24
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__WAVE_ID__SHIFT 25
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SIMD_ID__SHIFT 30
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__WGP_ID__SHIFT 0
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__SE_ID__SHIFT 4
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__ENCODING__SHIFT 6
+
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__DATA_MASK 0x000007fffff
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SA_ID_MASK 0x0000800000
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV_MASK 0x00001000000
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__WAVE_ID_MASK 0x0003e000000
+#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SIMD_ID_MASK 0x000c0000000
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__WGP_ID_MASK 0x00f
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__SE_ID_MASK 0x030
+#define SQ_INTERRUPT_WORD_WAVE_CTXID1__ENCODING_MASK 0x0c0
+
+#define KFD_CTXID0__ERR_TYPE_MASK 0x780000
+#define KFD_CTXID0__ERR_TYPE__SHIFT 19
+
+/* GFX10 SQ interrupt ENC type bit (context_id1[7:6]) for wave s_sendmsg */
+#define KFD_CONTEXT_ID1_ENC_TYPE_WAVE_MASK	0x40
+/* GFX10 SQ interrupt PRIV bit (context_id0[24]) for s_sendmsg inside trap */
+#define KFD_CONTEXT_ID0_PRIV_MASK		0x1000000
+/*
+ * The debugger will send user data(m0) with PRIV=1 to indicate it requires
+ * notification from the KFD with the following queue id (DOORBELL_ID) and
+ * trap code (TRAP_CODE).
+ */
+#define KFD_CONTEXT_ID0_DEBUG_DOORBELL_MASK	0x0003ff
+#define KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_SHIFT	10
+#define KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_MASK	0x07fc00
+#define KFD_DEBUG_DOORBELL_ID(ctxid0)	((ctxid0) &	\
+				KFD_CONTEXT_ID0_DEBUG_DOORBELL_MASK)
+#define KFD_DEBUG_TRAP_CODE(ctxid0)	(((ctxid0) &	\
+				KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_MASK)	\
+				>> KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_SHIFT)
+#define KFD_DEBUG_CP_BAD_OP_ECODE_MASK		0x3fffc00
+#define KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT		10
+#define KFD_DEBUG_CP_BAD_OP_ECODE(ctxid0) (((ctxid0) &			\
+				KFD_DEBUG_CP_BAD_OP_ECODE_MASK)		\
+				>> KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT)
+
+static void event_interrupt_poison_consumption(struct kfd_node *dev,
+				uint16_t pasid, uint16_t client_id)
+{
+	int old_poison, ret = -EINVAL;
+	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
+
+	if (!p)
+		return;
+
+	/* all queues of a process will be unmapped in one time */
+	old_poison = atomic_cmpxchg(&p->poison, 0, 1);
+	kfd_unref_process(p);
+	if (old_poison)
+		return;
+
+	switch (client_id) {
+	case SOC15_IH_CLIENTID_SE0SH:
+	case SOC15_IH_CLIENTID_SE1SH:
+	case SOC15_IH_CLIENTID_SE2SH:
+	case SOC15_IH_CLIENTID_SE3SH:
+	case SOC15_IH_CLIENTID_UTCL2:
+		ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
+		break;
+	case SOC15_IH_CLIENTID_SDMA0:
+	case SOC15_IH_CLIENTID_SDMA1:
+	case SOC15_IH_CLIENTID_SDMA2:
+	case SOC15_IH_CLIENTID_SDMA3:
+	case SOC15_IH_CLIENTID_SDMA4:
+		break;
+	default:
+		break;
+	}
+
+	kfd_signal_poison_consumed_event(dev, pasid);
+
+	/* resetting queue passes, do page retirement without gpu reset
+	 * resetting queue fails, fallback to gpu reset solution
+	 */
+	if (!ret) {
+		dev_warn(dev->adev->dev,
+			"RAS poison consumption, unmap queue flow succeeded: client id %d\n",
+			client_id);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false);
+	} else {
+		dev_warn(dev->adev->dev,
+			"RAS poison consumption, fall back to gpu reset flow: client id %d\n",
+			client_id);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true);
+	}
+}
+
+static bool event_interrupt_isr_v10(struct kfd_node *dev,
+					const uint32_t *ih_ring_entry,
+					uint32_t *patched_ihre,
+					bool *patched_flag)
+{
+	uint16_t source_id, client_id, pasid, vmid;
+	const uint32_t *data = ih_ring_entry;
+
+	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
+	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
+
+	/* Only handle interrupts from KFD VMIDs */
+	vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry);
+	if (!KFD_IRQ_IS_FENCE(client_id, source_id) &&
+	   (vmid < dev->vm_info.first_vmid_kfd ||
+	    vmid > dev->vm_info.last_vmid_kfd))
+		return false;
+
+	pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry);
+
+	/* Only handle clients we care about */
+	if (client_id != SOC15_IH_CLIENTID_GRBM_CP &&
+	    client_id != SOC15_IH_CLIENTID_SDMA0 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA1 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA2 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA3 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA4 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA5 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA6 &&
+	    client_id != SOC15_IH_CLIENTID_SDMA7 &&
+	    client_id != SOC15_IH_CLIENTID_VMC &&
+	    client_id != SOC15_IH_CLIENTID_VMC1 &&
+	    client_id != SOC15_IH_CLIENTID_UTCL2 &&
+	    client_id != SOC15_IH_CLIENTID_SE0SH &&
+	    client_id != SOC15_IH_CLIENTID_SE1SH &&
+	    client_id != SOC15_IH_CLIENTID_SE2SH &&
+	    client_id != SOC15_IH_CLIENTID_SE3SH)
+		return false;
+
+	pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n",
+		 client_id, source_id, vmid, pasid);
+	pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
+		 data[0], data[1], data[2], data[3],
+		 data[4], data[5], data[6], data[7]);
+
+	/* If there is no valid PASID, it's likely a bug */
+	if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt"))
+		return 0;
+
+	/* Interrupt types we care about: various signals and faults.
+	 * They will be forwarded to a work queue (see below).
+	 */
+	return source_id == SOC15_INTSRC_CP_END_OF_PIPE ||
+		source_id == SOC15_INTSRC_SDMA_TRAP ||
+		source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG ||
+		source_id == SOC15_INTSRC_CP_BAD_OPCODE ||
+		client_id == SOC15_IH_CLIENTID_VMC ||
+		client_id == SOC15_IH_CLIENTID_VMC1 ||
+		client_id == SOC15_IH_CLIENTID_UTCL2 ||
+		KFD_IRQ_IS_FENCE(client_id, source_id);
+}
+
+static void event_interrupt_wq_v10(struct kfd_node *dev,
+					const uint32_t *ih_ring_entry)
+{
+	uint16_t source_id, client_id, pasid, vmid;
+	uint32_t context_id0, context_id1;
+	uint32_t encoding, sq_intr_err_type;
+
+	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
+	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
+	pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry);
+	vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry);
+	context_id0 = SOC15_CONTEXT_ID0_FROM_IH_ENTRY(ih_ring_entry);
+	context_id1 = SOC15_CONTEXT_ID1_FROM_IH_ENTRY(ih_ring_entry);
+
+	if (client_id == SOC15_IH_CLIENTID_GRBM_CP ||
+	    client_id == SOC15_IH_CLIENTID_SE0SH ||
+	    client_id == SOC15_IH_CLIENTID_SE1SH ||
+	    client_id == SOC15_IH_CLIENTID_SE2SH ||
+	    client_id == SOC15_IH_CLIENTID_SE3SH) {
+		if (source_id == SOC15_INTSRC_CP_END_OF_PIPE)
+			kfd_signal_event_interrupt(pasid, context_id0, 32);
+		else if (source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG) {
+			encoding = REG_GET_FIELD(context_id1,
+						SQ_INTERRUPT_WORD_WAVE_CTXID1, ENCODING);
+			switch (encoding) {
+			case SQ_INTERRUPT_WORD_ENCODING_AUTO:
+				pr_debug(
+					"sq_intr: auto, se %d, ttrace %d, wlt %d, ttrac_buf0_full %d, ttrac_buf1_full %d, ttrace_utc_err %d\n",
+					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_AUTO_CTXID1,
+							SE_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+							THREAD_TRACE),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+							WLT),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+							THREAD_TRACE_BUF0_FULL),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+							THREAD_TRACE_BUF1_FULL),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0,
+							THREAD_TRACE_UTC_ERROR));
+				break;
+			case SQ_INTERRUPT_WORD_ENCODING_INST:
+				pr_debug("sq_intr: inst, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n",
+					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+							SE_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							DATA),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							SA_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							PRIV),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							WAVE_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							SIMD_ID),
+					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+							WGP_ID));
+				if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV_MASK) {
+					if (kfd_set_dbg_ev_from_interrupt(dev, pasid,
+							KFD_DEBUG_DOORBELL_ID(context_id0),
+							KFD_DEBUG_TRAP_CODE(context_id0),
+							NULL, 0))
+						return;
+				}
+				break;
+			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
+				sq_intr_err_type = REG_GET_FIELD(context_id0, KFD_CTXID0,
+								ERR_TYPE);
+				pr_warn("sq_intr: error, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d, err_type %d\n",
+					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+							SE_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							DATA),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							SA_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							PRIV),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							WAVE_ID),
+					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0,
+							SIMD_ID),
+					REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1,
+							WGP_ID),
+					sq_intr_err_type);
+				if (sq_intr_err_type != SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST &&
+					sq_intr_err_type != SQ_INTERRUPT_ERROR_TYPE_MEMVIOL) {
+					event_interrupt_poison_consumption(dev, pasid, source_id);
+					return;
+				}
+				break;
+			default:
+				break;
+			}
+			kfd_signal_event_interrupt(pasid, context_id0 & 0x7fffff, 23);
+		} else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) {
+			kfd_set_dbg_ev_from_interrupt(dev, pasid,
+				KFD_DEBUG_DOORBELL_ID(context_id0),
+				KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)),
+				NULL,
+				0);
+		}
+	} else if (client_id == SOC15_IH_CLIENTID_SDMA0 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA1 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA2 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA3 ||
+		   (client_id == SOC15_IH_CLIENTID_SDMA3_Sienna_Cichlid &&
+		    KFD_GC_VERSION(dev) == IP_VERSION(10, 3, 0)) ||
+		   client_id == SOC15_IH_CLIENTID_SDMA4 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA5 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA6 ||
+		   client_id == SOC15_IH_CLIENTID_SDMA7) {
+		if (source_id == SOC15_INTSRC_SDMA_TRAP) {
+			kfd_signal_event_interrupt(pasid, context_id0 & 0xfffffff, 28);
+		} else if (source_id == SOC15_INTSRC_SDMA_ECC) {
+			event_interrupt_poison_consumption(dev, pasid, source_id);
+			return;
+		}
+	} else if (client_id == SOC15_IH_CLIENTID_VMC ||
+		   client_id == SOC15_IH_CLIENTID_VMC1 ||
+		   client_id == SOC15_IH_CLIENTID_UTCL2) {
+		struct kfd_vm_fault_info info = {0};
+		uint16_t ring_id = SOC15_RING_ID_FROM_IH_ENTRY(ih_ring_entry);
+		struct kfd_hsa_memory_exception_data exception_data;
+
+		if (client_id == SOC15_IH_CLIENTID_UTCL2 &&
+				amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) {
+			event_interrupt_poison_consumption(dev, pasid, client_id);
+			return;
+		}
+
+		info.vmid = vmid;
+		info.mc_id = client_id;
+		info.page_addr = ih_ring_entry[4] |
+			(uint64_t)(ih_ring_entry[5] & 0xf) << 32;
+		info.prot_valid = ring_id & 0x08;
+		info.prot_read  = ring_id & 0x10;
+		info.prot_write = ring_id & 0x20;
+
+		memset(&exception_data, 0, sizeof(exception_data));
+		exception_data.gpu_id = dev->id;
+		exception_data.va = (info.page_addr) << PAGE_SHIFT;
+		exception_data.failure.NotPresent = info.prot_valid ? 1 : 0;
+		exception_data.failure.NoExecute = info.prot_exec ? 1 : 0;
+		exception_data.failure.ReadOnly = info.prot_write ? 1 : 0;
+		exception_data.failure.imprecise = 0;
+
+		kfd_set_dbg_ev_from_interrupt(dev,
+						pasid,
+						-1,
+						KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION),
+						&exception_data,
+						sizeof(exception_data));
+	} else if (KFD_IRQ_IS_FENCE(client_id, source_id)) {
+		kfd_process_close_interrupt_drain(pasid);
+	}
+}
+
+const struct kfd_event_interrupt_class event_interrupt_class_v10 = {
+	.interrupt_isr = event_interrupt_isr_v10,
+	.interrupt_wq = event_interrupt_wq_v10,
+};
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index c2166bf964ef0..f933bd231fb9c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -26,6 +26,7 @@
 #include "kfd_device_queue_manager.h"
 #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"
 #include "kfd_smi_events.h"
+#include "kfd_debug.h"
 
 /*
  * GFX11 SQ Interrupts
@@ -238,7 +239,7 @@ static bool event_interrupt_isr_v11(struct kfd_node *dev,
 	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
 	/* Only handle interrupts from KFD VMIDs */
 	vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry);
-	if (/*!KFD_IRQ_IS_FENCE(client_id, source_id) &&*/
+	if (!KFD_IRQ_IS_FENCE(client_id, source_id) &&
 	    (vmid < dev->vm_info.first_vmid_kfd ||
 	    vmid > dev->vm_info.last_vmid_kfd))
 		return false;
@@ -267,7 +268,7 @@ static bool event_interrupt_isr_v11(struct kfd_node *dev,
 		source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG ||
 		source_id == SOC15_INTSRC_CP_BAD_OPCODE ||
 		source_id == SOC21_INTSRC_SDMA_TRAP ||
-		/* KFD_IRQ_IS_FENCE(client_id, source_id) || */
+		KFD_IRQ_IS_FENCE(client_id, source_id) ||
 		(((client_id == SOC21_IH_CLIENTID_VMC) ||
 		 ((client_id == SOC21_IH_CLIENTID_GFX) &&
 		  (source_id == UTCL2_1_0__SRCID__FAULT))) &&
@@ -279,7 +280,7 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 {
 	uint16_t source_id, client_id, ring_id, pasid, vmid;
 	uint32_t context_id0, context_id1;
-	uint8_t sq_int_enc, sq_int_errtype;
+	uint8_t sq_int_enc, sq_int_priv, sq_int_errtype;
 	struct kfd_vm_fault_info info = {0};
 	struct kfd_hsa_memory_exception_data exception_data;
 
@@ -312,9 +313,9 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 		exception_data.failure.ReadOnly = info.prot_write ? 1 : 0;
 		exception_data.failure.imprecise = 0;
 
-		/*kfd_set_dbg_ev_from_interrupt(dev, pasid, -1,
+		kfd_set_dbg_ev_from_interrupt(dev, pasid, -1,
 					      KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION),
-					      &exception_data, sizeof(exception_data));*/
+					      &exception_data, sizeof(exception_data));
 		kfd_smi_event_update_vmfault(dev, pasid);
 
 	/* GRBM, SDMA, SE, PMM */
@@ -324,11 +325,11 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 		/* CP */
 		if (source_id == SOC15_INTSRC_CP_END_OF_PIPE)
 			kfd_signal_event_interrupt(pasid, context_id0, 32);
-		/*else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE)
+		else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE)
 			kfd_set_dbg_ev_from_interrupt(dev, pasid,
 				KFD_CTXID0_DOORBELL_ID(context_id0),
 				KFD_EC_MASK(KFD_CTXID0_CP_BAD_OP_ECODE(context_id0)),
-				NULL, 0);*/
+				NULL, 0);
 
 		/* SDMA */
 		else if (source_id == SOC21_INTSRC_SDMA_TRAP)
@@ -348,6 +349,13 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_INST:
 				print_sq_intr_info_inst(context_id0, context_id1);
+				sq_int_priv = REG_GET_FIELD(context_id0,
+						SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV);
+				if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid,
+						KFD_CTXID0_DOORBELL_ID(context_id0),
+						KFD_CTXID0_TRAP_CODE(context_id0),
+						NULL, 0)))
+					return;
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				print_sq_intr_info_error(context_id0, context_id1);
@@ -366,8 +374,8 @@ static void event_interrupt_wq_v11(struct kfd_node *dev,
 			kfd_signal_event_interrupt(pasid, context_id0 & 0xffffff, 24);
 		}
 
-	/*} else if (KFD_IRQ_IS_FENCE(client_id, source_id)) {
-		kfd_process_close_interrupt_drain(pasid);*/
+	} else if (KFD_IRQ_IS_FENCE(client_id, source_id)) {
+		kfd_process_close_interrupt_drain(pasid);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 8cf58be80f4ea..d5c9f30552e3b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -23,10 +23,40 @@
 
 #include "kfd_priv.h"
 #include "kfd_events.h"
+#include "kfd_debug.h"
 #include "soc15_int.h"
 #include "kfd_device_queue_manager.h"
 #include "kfd_smi_events.h"
 
+/*
+ * GFX9 SQ Interrupts
+ *
+ * There are 3 encoding types of interrupts sourced from SQ sent as a 44-bit
+ * packet to the Interrupt Handler:
+ * Auto - Generated by the SQG (various cmd overflows, timestamps etc)
+ * Wave - Generated by S_SENDMSG through a shader program
+ * Error - HW generated errors (Illegal instructions, Memviols, EDC etc)
+ *
+ * The 44-bit packet is mapped as {context_id1[7:0],context_id0[31:0]} plus
+ * 4-bits for VMID (SOC15_VMID_FROM_IH_ENTRY) as such:
+ *
+ * - context_id0[27:26]
+ * Encoding type (0 = Auto, 1 = Wave, 2 = Error)
+ *
+ * - context_id0[13]
+ * PRIV bit indicates that Wave S_SEND or error occurred within trap
+ *
+ * - {context_id1[7:0],context_id0[31:28],context_id0[11:0]}
+ * 24-bit data with the following layout per encoding type:
+ * Auto - only context_id0[8:0] is used, which reports various interrupts
+ * generated by SQG.  The rest is 0.
+ * Wave - user data sent from m0 via S_SENDMSG
+ * Error - Error type (context_id1[7:4]), Error Details (rest of bits)
+ *
+ * The other context_id bits show coordinates (SE/SH/CU/SIMD/WAVE) for wave
+ * S_SENDMSG and Errors.  These are 0 for Auto.
+ */
+
 enum SQ_INTERRUPT_WORD_ENCODING {
 	SQ_INTERRUPT_WORD_ENCODING_AUTO = 0x0,
 	SQ_INTERRUPT_WORD_ENCODING_INST,
@@ -84,12 +114,32 @@ enum SQ_INTERRUPT_ERROR_TYPE {
 #define SQ_INTERRUPT_WORD_WAVE_CTXID__SE_ID_MASK 0x03000000
 #define SQ_INTERRUPT_WORD_WAVE_CTXID__ENCODING_MASK 0x0c000000
 
+/* GFX9 SQ interrupt 24-bit data from context_id<0,1> */
 #define KFD_CONTEXT_ID_GET_SQ_INT_DATA(ctx0, ctx1)                             \
 	((ctx0 & 0xfff) | ((ctx0 >> 16) & 0xf000) | ((ctx1 << 16) & 0xff0000))
 
 #define KFD_SQ_INT_DATA__ERR_TYPE_MASK 0xF00000
 #define KFD_SQ_INT_DATA__ERR_TYPE__SHIFT 20
 
+/*
+ * The debugger will send user data(m0) with PRIV=1 to indicate it requires
+ * notification from the KFD with the following queue id (DOORBELL_ID) and
+ * trap code (TRAP_CODE).
+ */
+#define KFD_INT_DATA_DEBUG_DOORBELL_MASK	0x0003ff
+#define KFD_INT_DATA_DEBUG_TRAP_CODE_SHIFT	10
+#define KFD_INT_DATA_DEBUG_TRAP_CODE_MASK	0x07fc00
+#define KFD_DEBUG_DOORBELL_ID(sq_int_data)	((sq_int_data) &	\
+				KFD_INT_DATA_DEBUG_DOORBELL_MASK)
+#define KFD_DEBUG_TRAP_CODE(sq_int_data)	(((sq_int_data) &	\
+				KFD_INT_DATA_DEBUG_TRAP_CODE_MASK)	\
+				>> KFD_INT_DATA_DEBUG_TRAP_CODE_SHIFT)
+#define KFD_DEBUG_CP_BAD_OP_ECODE_MASK		0x3fffc00
+#define KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT		10
+#define KFD_DEBUG_CP_BAD_OP_ECODE(ctxid0)	(((ctxid0) &		\
+				KFD_DEBUG_CP_BAD_OP_ECODE_MASK)		\
+				>> KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT)
+
 static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 				uint16_t pasid, uint16_t client_id)
 {
@@ -168,14 +218,16 @@ static bool event_interrupt_isr_v9(struct kfd_node *dev,
 	uint16_t source_id, client_id, pasid, vmid;
 	const uint32_t *data = ih_ring_entry;
 
+	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
+	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
+
 	/* Only handle interrupts from KFD VMIDs */
 	vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry);
-	if (vmid < dev->vm_info.first_vmid_kfd ||
-	    vmid > dev->vm_info.last_vmid_kfd)
+	if (!KFD_IRQ_IS_FENCE(client_id, source_id) &&
+	   (vmid < dev->vm_info.first_vmid_kfd ||
+	    vmid > dev->vm_info.last_vmid_kfd))
 		return false;
 
-	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
-	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
 	pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry);
 
 	/* Only handle clients we care about */
@@ -194,7 +246,8 @@ static bool event_interrupt_isr_v9(struct kfd_node *dev,
 	    client_id != SOC15_IH_CLIENTID_SE0SH &&
 	    client_id != SOC15_IH_CLIENTID_SE1SH &&
 	    client_id != SOC15_IH_CLIENTID_SE2SH &&
-	    client_id != SOC15_IH_CLIENTID_SE3SH)
+	    client_id != SOC15_IH_CLIENTID_SE3SH &&
+	    !KFD_IRQ_IS_FENCE(client_id, source_id))
 		return false;
 
 	/* This is a known issue for gfx9. Under non HWS, pasid is not set
@@ -247,6 +300,7 @@ static bool event_interrupt_isr_v9(struct kfd_node *dev,
 		source_id == SOC15_INTSRC_SDMA_ECC ||
 		source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG ||
 		source_id == SOC15_INTSRC_CP_BAD_OPCODE ||
+		KFD_IRQ_IS_FENCE(client_id, source_id) ||
 		((client_id == SOC15_IH_CLIENTID_VMC ||
 		client_id == SOC15_IH_CLIENTID_VMC1 ||
 		client_id == SOC15_IH_CLIENTID_UTCL2) &&
@@ -302,6 +356,13 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SIMD_ID),
 					REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, CU_ID),
 					sq_int_data);
+				if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID__PRIV_MASK) {
+					if (kfd_set_dbg_ev_from_interrupt(dev, pasid,
+							KFD_DEBUG_DOORBELL_ID(sq_int_data),
+							KFD_DEBUG_TRAP_CODE(sq_int_data),
+							NULL, 0))
+						return;
+				}
 				break;
 			case SQ_INTERRUPT_WORD_ENCODING_ERROR:
 				sq_intr_err = REG_GET_FIELD(sq_int_data, KFD_SQ_INT_DATA, ERR_TYPE);
@@ -324,8 +385,12 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 				break;
 			}
 			kfd_signal_event_interrupt(pasid, context_id0 & 0xffffff, 24);
-		} else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE)
-			kfd_signal_hw_exception_event(pasid);
+		} else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) {
+			kfd_set_dbg_ev_from_interrupt(dev, pasid,
+				KFD_DEBUG_DOORBELL_ID(context_id0),
+				KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)),
+				NULL, 0);
+		}
 	} else if (client_id == SOC15_IH_CLIENTID_SDMA0 ||
 		   client_id == SOC15_IH_CLIENTID_SDMA1 ||
 		   client_id == SOC15_IH_CLIENTID_SDMA2 ||
@@ -345,6 +410,7 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 		   client_id == SOC15_IH_CLIENTID_UTCL2) {
 		struct kfd_vm_fault_info info = {0};
 		uint16_t ring_id = SOC15_RING_ID_FROM_IH_ENTRY(ih_ring_entry);
+		struct kfd_hsa_memory_exception_data exception_data;
 
 		if (client_id == SOC15_IH_CLIENTID_UTCL2 &&
 		    amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) {
@@ -360,9 +426,23 @@ static void event_interrupt_wq_v9(struct kfd_node *dev,
 		info.prot_read  = ring_id & 0x10;
 		info.prot_write = ring_id & 0x20;
 
+		memset(&exception_data, 0, sizeof(exception_data));
+		exception_data.gpu_id = dev->id;
+		exception_data.va = (info.page_addr) << PAGE_SHIFT;
+		exception_data.failure.NotPresent = info.prot_valid ? 1 : 0;
+		exception_data.failure.NoExecute = info.prot_exec ? 1 : 0;
+		exception_data.failure.ReadOnly = info.prot_write ? 1 : 0;
+		exception_data.failure.imprecise = 0;
+
+		kfd_set_dbg_ev_from_interrupt(dev,
+						pasid,
+						-1,
+						KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION),
+						&exception_data,
+						sizeof(exception_data));
 		kfd_smi_event_update_vmfault(dev, pasid);
-		kfd_dqm_evict_pasid(dev->dqm, pasid);
-		kfd_signal_vm_fault_event(dev, pasid, &info, NULL);
+	} else if (KFD_IRQ_IS_FENCE(client_id, source_id)) {
+		kfd_process_close_interrupt_drain(pasid);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index a02fb939614af..cd2d56e5cdf05 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -963,6 +963,10 @@ struct kfd_process {
 	uint64_t exception_enable_mask;
 	uint64_t exception_status;
 
+	/* Used to drain stale interrupts */
+	wait_queue_head_t wait_irq_drain;
+	bool irq_drain_is_open;
+
 	/* shared virtual memory registered by this process */
 	struct svm_range_list svms;
 
@@ -1144,12 +1148,19 @@ int kfd_numa_node_to_apic_id(int numa_node_id);
 void kfd_double_confirm_iommu_support(struct kfd_dev *gpu);
 
 /* Interrupts */
+#define	KFD_IRQ_FENCE_CLIENTID	0xff
+#define	KFD_IRQ_FENCE_SOURCEID	0xff
+#define	KFD_IRQ_IS_FENCE(client, source)				\
+				((client) == KFD_IRQ_FENCE_CLIENTID &&	\
+				(source) == KFD_IRQ_FENCE_SOURCEID)
 int kfd_interrupt_init(struct kfd_node *dev);
 void kfd_interrupt_exit(struct kfd_node *dev);
 bool enqueue_ih_ring_entry(struct kfd_node *kfd, const void *ih_ring_entry);
 bool interrupt_is_wanted(struct kfd_node *dev,
 				const uint32_t *ih_ring_entry,
 				uint32_t *patched_ihre, bool *flag);
+int kfd_process_drain_interrupts(struct kfd_process_device *pdd);
+void kfd_process_close_interrupt_drain(unsigned int pasid);
 
 /* amdkfd Apertures */
 int kfd_init_apertures(struct kfd_process *process);
@@ -1421,6 +1432,7 @@ uint64_t kfd_get_number_elems(struct kfd_dev *kfd);
 /* Events */
 extern const struct kfd_event_interrupt_class event_interrupt_class_cik;
 extern const struct kfd_event_interrupt_class event_interrupt_class_v9;
+extern const struct kfd_event_interrupt_class event_interrupt_class_v10;
 extern const struct kfd_event_interrupt_class event_interrupt_class_v11;
 
 extern const struct kfd_device_global_init_class device_global_init_class_cik;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 3b7f219c9d06f..3d3611705d419 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -862,6 +862,8 @@ struct kfd_process *kfd_create_process(struct task_struct *thread)
 		kfd_procfs_add_sysfs_stats(process);
 		kfd_procfs_add_sysfs_files(process);
 		kfd_procfs_add_sysfs_counters(process);
+
+		init_waitqueue_head(&process->wait_irq_drain);
 	}
 out:
 	if (!IS_ERR(process))
@@ -2136,6 +2138,51 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type)
 	}
 }
 
+/* assumes caller holds process lock. */
+int kfd_process_drain_interrupts(struct kfd_process_device *pdd)
+{
+	uint32_t irq_drain_fence[8];
+	int r = 0;
+
+	if (!KFD_IS_SOC15(pdd->dev))
+		return 0;
+
+	pdd->process->irq_drain_is_open = true;
+
+	memset(irq_drain_fence, 0, sizeof(irq_drain_fence));
+	irq_drain_fence[0] = (KFD_IRQ_FENCE_SOURCEID << 8) |
+							KFD_IRQ_FENCE_CLIENTID;
+	irq_drain_fence[3] = pdd->process->pasid;
+
+	/* ensure stale irqs scheduled KFD interrupts and send drain fence. */
+	if (amdgpu_amdkfd_send_close_event_drain_irq(pdd->dev->adev,
+						     irq_drain_fence)) {
+		pdd->process->irq_drain_is_open = false;
+		return 0;
+	}
+
+	r = wait_event_interruptible(pdd->process->wait_irq_drain,
+				     !READ_ONCE(pdd->process->irq_drain_is_open));
+	if (r)
+		pdd->process->irq_drain_is_open = false;
+
+	return r;
+}
+
+void kfd_process_close_interrupt_drain(unsigned int pasid)
+{
+	struct kfd_process *p;
+
+	p = kfd_lookup_process_by_pasid(pasid);
+
+	if (!p)
+		return;
+
+	WRITE_ONCE(p->irq_drain_is_open, false);
+	wake_up_all(&p->wait_irq_drain);
+	kfd_unref_process(p);
+}
+
 struct send_exception_work_handler_workarea {
 	struct work_struct work;
 	struct kfd_process *p;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 43d432b5c5bcb..70852a200d8f7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -337,6 +337,10 @@ int pqm_create_queue(struct process_queue_manager *pqm,
 		kq->queue->properties.queue_id = *qid;
 		pqn->kq = kq;
 		pqn->q = NULL;
+		retval = kfd_process_drain_interrupts(pdd);
+		if (retval)
+			break;
+
 		retval = dev->dqm->ops.create_kernel_queue(dev->dqm,
 							kq, &pdd->qpd);
 		break;
-- 
GitLab


From e90bf919f714ae2a658cdfd03238e7be9ce9185c Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 27 Apr 2022 10:24:37 -0400
Subject: [PATCH 1540/1791] drm/amdkfd: add debug set exceptions enabled
 operation

The debugger subscibes to nofication for requested exceptions on attach.
Allow the debugger to change its subsciption later on.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  3 ++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 36 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |  2 ++
 3 files changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 615fa9ab36b7e..dcf4b5f3886e3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2980,6 +2980,9 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				args->send_runtime_event.exception_mask);
 		break;
 	case KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED:
+		kfd_dbg_set_enabled_debug_exception_mask(target,
+				args->set_exceptions_enabled.exception_mask);
+		break;
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE:
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE:
 	case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 68b657398d417..48a4e3cc22343 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -521,3 +521,39 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 
 	return r;
 }
+
+void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
+					uint64_t exception_set_mask)
+{
+	uint64_t found_mask = 0;
+	struct process_queue_manager *pqm;
+	struct process_queue_node *pqn;
+	static const char write_data = '.';
+	loff_t pos = 0;
+	int i;
+
+	mutex_lock(&target->event_mutex);
+
+	found_mask |= target->exception_status;
+
+	pqm = &target->pqm;
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
+		if (!pqn)
+			continue;
+
+		found_mask |= pqn->q->properties.exception_status;
+	}
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		found_mask |= pdd->exception_status;
+	}
+
+	if (exception_set_mask & found_mask)
+		kernel_write(target->dbg_ev_file, &write_data, 1, &pos);
+
+	target->exception_enable_mask = exception_set_mask;
+
+	mutex_unlock(&target->event_mutex);
+}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 5153ccbd7fd17..6c1054a088723 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -56,6 +56,8 @@ static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 
 void debug_event_write_work_handler(struct work_struct *work);
 
+void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
+					uint64_t exception_set_mask);
 /*
  * If GFX off is enabled, chips that do not support RLC restore for the debug
  * registers will disable GFX off temporarily for the entire debug session.
-- 
GitLab


From 101827e13026a981e887527620fe9710adc0e481 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 27 Apr 2022 13:18:10 -0400
Subject: [PATCH 1541/1791] drm/amdkfd: add debug wave launch override
 operation

This operation allows the debugger to override the enabled HW
exceptions on the device.

On debug devices that only support the debugging of a single process,
the HW exceptions are global and set through the SPI_GDBG_TRAP_MASK
register.
Because they are global, only address watch exceptions are allowed to
be enabled.  In other words, the debugger must preserve all non-address
watch exception states in normal mode operation by barring a full
replacement override or a non-address watch override request.

For multi-process debugging, all HW exception overrides are per-VMID so
all exceptions can be overridden or fully replaced.

In order for the debugger to know what is permissible, returned the
supported override mask back to the debugger along with the previously
enable overrides.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c  | 47 ++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c   |  2 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    | 55 ++++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h    | 10 +++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |  5 +-
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    | 87 ++++++++++++++++++-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 55 ++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h | 10 +++
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  7 ++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        | 69 +++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  6 ++
 11 files changed, 351 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index b811a09850503..d7881bbd828d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -25,6 +25,7 @@
 #include "amdgpu_amdkfd_gfx_v9.h"
 #include "gc/gc_9_4_2_offset.h"
 #include "gc/gc_9_4_2_sh_mask.h"
+#include <uapi/linux/kfd_ioctl.h>
 
 /*
  * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE.
@@ -62,6 +63,50 @@ static uint32_t kgd_aldebaran_disable_debug_trap(struct amdgpu_device *adev,
 	return data;
 }
 
+static int kgd_aldebaran_validate_trap_override_request(struct amdgpu_device *adev,
+							uint32_t trap_override,
+							uint32_t *trap_mask_supported)
+{
+	*trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID |
+				KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL |
+				KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO |
+				KFD_DBG_TRAP_MASK_FP_OVERFLOW |
+				KFD_DBG_TRAP_MASK_FP_UNDERFLOW |
+				KFD_DBG_TRAP_MASK_FP_INEXACT |
+				KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO |
+				KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH |
+				KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION;
+
+	if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR &&
+			trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE)
+		return -EPERM;
+
+	return 0;
+}
+
+/* returns TRAP_EN, EXCP_EN and EXCP_RPLACE. */
+static uint32_t kgd_aldebaran_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					uint32_t vmid,
+					uint32_t trap_override,
+					uint32_t trap_mask_bits,
+					uint32_t trap_mask_request,
+					uint32_t *trap_mask_prev,
+					uint32_t kfd_dbg_trap_cntl_prev)
+
+{
+	uint32_t data = 0;
+
+	*trap_mask_prev = REG_GET_FIELD(kfd_dbg_trap_cntl_prev, SPI_GDBG_PER_VMID_CNTL, EXCP_EN);
+	trap_mask_bits = (trap_mask_bits & trap_mask_request) |
+		(*trap_mask_prev & ~trap_mask_request);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, trap_mask_bits);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override);
+
+	return data;
+}
+
 const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -82,6 +127,8 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
 	.enable_debug_trap = kgd_aldebaran_enable_debug_trap,
 	.disable_debug_trap = kgd_aldebaran_disable_debug_trap,
+	.validate_trap_override_request = kgd_aldebaran_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_aldebaran_set_wave_launch_trap_override,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index a62bd0068515c..ec25876640018 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -410,6 +410,8 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 				kgd_gfx_v9_set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_arcturus_enable_debug_trap,
 	.disable_debug_trap = kgd_arcturus_disable_debug_trap,
+	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 98006c7021ddf..7ea0362dcab32 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -32,6 +32,7 @@
 #include "v10_structs.h"
 #include "nv.h"
 #include "nvd.h"
+#include <uapi/linux/kfd_ioctl.h>
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -803,6 +804,58 @@ uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
 	return 0;
 }
 
+int kgd_gfx_v10_validate_trap_override_request(struct amdgpu_device *adev,
+					      uint32_t trap_override,
+					      uint32_t *trap_mask_supported)
+{
+	*trap_mask_supported &= KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH;
+
+	/* The SPI_GDBG_TRAP_MASK register is global and affects all
+	 * processes. Only allow OR-ing the address-watch bit, since
+	 * this only affects processes under the debugger. Other bits
+	 * should stay 0 to avoid the debugger interfering with other
+	 * processes.
+	 */
+	if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR)
+		return -EINVAL;
+
+	return 0;
+}
+
+uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					      uint32_t vmid,
+					      uint32_t trap_override,
+					      uint32_t trap_mask_bits,
+					      uint32_t trap_mask_request,
+					      uint32_t *trap_mask_prev,
+					      uint32_t kfd_dbg_trap_cntl_prev)
+{
+	uint32_t data, wave_cntl_prev;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	wave_cntl_prev = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true);
+
+	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK));
+	*trap_mask_prev = REG_GET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN);
+
+	trap_mask_bits = (trap_mask_bits & trap_mask_request) |
+		(*trap_mask_prev & ~trap_mask_request);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN, trap_mask_bits);
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, REPLACE, trap_override);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data);
+
+	/* We need to preserve wave launch mode stall settings. */
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), wave_cntl_prev);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 /* kgd_gfx_v10_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -889,6 +942,8 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
+	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
 	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.program_trap_handler_settings = program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
index 1e993a2136464..57339fa128079 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
@@ -26,6 +26,16 @@ uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev,
 uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev,
 					bool keep_trap_enabled,
 					uint32_t vmid);
+int kgd_gfx_v10_validate_trap_override_request(struct amdgpu_device *adev,
+					     uint32_t trap_override,
+					     uint32_t *trap_mask_supported);
+uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					     uint32_t vmid,
+					     uint32_t trap_override,
+					     uint32_t trap_mask_bits,
+					     uint32_t trap_mask_request,
+					     uint32_t *trap_mask_prev,
+					     uint32_t kfd_dbg_trap_cntl_prev);
 void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
 void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t wait_times,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 387bdf4823c90..7120927fed15f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -675,5 +675,8 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
-	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap
+	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
+	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override
+
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index cc954cf248ca8..ae0c4707919f1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -30,6 +30,7 @@
 #include "soc15d.h"
 #include "v11_structs.h"
 #include "soc21.h"
+#include <uapi/linux/kfd_ioctl.h>
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -643,6 +644,88 @@ static uint32_t kgd_gfx_v11_disable_debug_trap(struct amdgpu_device *adev,
 	return data;
 }
 
+static int kgd_gfx_v11_validate_trap_override_request(struct amdgpu_device *adev,
+							uint32_t trap_override,
+							uint32_t *trap_mask_supported)
+{
+	*trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID |
+				KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL |
+				KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO |
+				KFD_DBG_TRAP_MASK_FP_OVERFLOW |
+				KFD_DBG_TRAP_MASK_FP_UNDERFLOW |
+				KFD_DBG_TRAP_MASK_FP_INEXACT |
+				KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO |
+				KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH |
+				KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION;
+
+	if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 4))
+		*trap_mask_supported |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START |
+					KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END;
+
+	if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR &&
+			trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE)
+		return -EPERM;
+
+	return 0;
+}
+
+static uint32_t trap_mask_map_sw_to_hw(uint32_t mask)
+{
+	uint32_t trap_on_start = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START) ? 1 : 0;
+	uint32_t trap_on_end = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END) ? 1 : 0;
+	uint32_t excp_en = mask & (KFD_DBG_TRAP_MASK_FP_INVALID |
+			KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL |
+			KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO |
+			KFD_DBG_TRAP_MASK_FP_OVERFLOW |
+			KFD_DBG_TRAP_MASK_FP_UNDERFLOW |
+			KFD_DBG_TRAP_MASK_FP_INEXACT |
+			KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO |
+			KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH |
+			KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION);
+	uint32_t ret;
+
+	ret = REG_SET_FIELD(0, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, excp_en);
+	ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START, trap_on_start);
+	ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END, trap_on_end);
+
+	return ret;
+}
+
+static uint32_t trap_mask_map_hw_to_sw(uint32_t mask)
+{
+	uint32_t ret = REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, EXCP_EN);
+
+	if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START))
+		ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START;
+
+	if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END))
+		ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END;
+
+	return ret;
+}
+
+/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */
+static uint32_t kgd_gfx_v11_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					uint32_t vmid,
+					uint32_t trap_override,
+					uint32_t trap_mask_bits,
+					uint32_t trap_mask_request,
+					uint32_t *trap_mask_prev,
+					uint32_t kfd_dbg_trap_cntl_prev)
+{
+	uint32_t data = 0;
+
+	*trap_mask_prev = trap_mask_map_hw_to_sw(kfd_dbg_trap_cntl_prev);
+
+	data = (trap_mask_bits & trap_mask_request) | (*trap_mask_prev & ~trap_mask_request);
+	data = trap_mask_map_sw_to_hw(data);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1);
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override);
+
+	return data;
+}
+
 const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v11,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11,
@@ -660,5 +743,7 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.get_atc_vmid_pasid_mapping_info = NULL,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base_v11,
 	.enable_debug_trap = kgd_gfx_v11_enable_debug_trap,
-	.disable_debug_trap = kgd_gfx_v11_disable_debug_trap
+	.disable_debug_trap = kgd_gfx_v11_disable_debug_trap,
+	.validate_trap_override_request = kgd_gfx_v11_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 829ee720cc448..ce5c256f05176 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -38,6 +38,7 @@
 #include "soc15d.h"
 #include "gfx_v9_0.h"
 #include "amdgpu_amdkfd_gfx_v9.h"
+#include <uapi/linux/kfd_ioctl.h>
 
 enum hqd_dequeue_request_type {
 	NO_ACTION = 0,
@@ -739,6 +740,58 @@ uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
 	return 0;
 }
 
+int kgd_gfx_v9_validate_trap_override_request(struct amdgpu_device *adev,
+					uint32_t trap_override,
+					uint32_t *trap_mask_supported)
+{
+	*trap_mask_supported &= KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH;
+
+	/* The SPI_GDBG_TRAP_MASK register is global and affects all
+	 * processes. Only allow OR-ing the address-watch bit, since
+	 * this only affects processes under the debugger. Other bits
+	 * should stay 0 to avoid the debugger interfering with other
+	 * processes.
+	 */
+	if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR)
+		return -EINVAL;
+
+	return 0;
+}
+
+uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					     uint32_t vmid,
+					     uint32_t trap_override,
+					     uint32_t trap_mask_bits,
+					     uint32_t trap_mask_request,
+					     uint32_t *trap_mask_prev,
+					     uint32_t kfd_dbg_cntl_prev)
+{
+	uint32_t data, wave_cntl_prev;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	wave_cntl_prev = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL));
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK));
+	*trap_mask_prev = REG_GET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN);
+
+	trap_mask_bits = (trap_mask_bits & trap_mask_request) |
+		(*trap_mask_prev & ~trap_mask_request);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN, trap_mask_bits);
+	data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, REPLACE, trap_override);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data);
+
+	/* We need to preserve wave launch mode stall settings. */
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), wave_cntl_prev);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 /* kgd_gfx_v9_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -1010,6 +1063,8 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base,
 	.enable_debug_trap = kgd_gfx_v9_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v9_disable_debug_trap,
+	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
+	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index fed5b7f18b1ad..76812ddd35b1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -71,6 +71,16 @@ uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev,
 uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
 					bool keep_trap_enabled,
 					uint32_t vmid);
+int kgd_gfx_v9_validate_trap_override_request(struct amdgpu_device *adev,
+					     uint32_t trap_override,
+					     uint32_t *trap_mask_supported);
+uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev,
+					     uint32_t vmid,
+					     uint32_t trap_override,
+					     uint32_t trap_mask_bits,
+					     uint32_t trap_mask_request,
+					     uint32_t *trap_mask_prev,
+					     uint32_t kfd_dbg_trap_cntl_prev);
 void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
 void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t wait_times,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index dcf4b5f3886e3..7fa249807671d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2984,6 +2984,13 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				args->set_exceptions_enabled.exception_mask);
 		break;
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE:
+		r = kfd_dbg_trap_set_wave_launch_override(target,
+				args->launch_override.override_mode,
+				args->launch_override.enable_mask,
+				args->launch_override.support_request_mask,
+				&args->launch_override.enable_mask,
+				&args->launch_override.support_request_mask);
+		break;
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE:
 	case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES:
 	case KFD_IOC_DBG_TRAP_RESUME_QUEUES:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 48a4e3cc22343..733390fb24590 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -522,6 +522,75 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 	return r;
 }
 
+static int kfd_dbg_validate_trap_override_request(struct kfd_process *p,
+						uint32_t trap_override,
+						uint32_t trap_mask_request,
+						uint32_t *trap_mask_supported)
+{
+	int i = 0;
+
+	*trap_mask_supported = 0xffffffff;
+
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+		int err = pdd->dev->kfd2kgd->validate_trap_override_request(
+								pdd->dev->adev,
+								trap_override,
+								trap_mask_supported);
+
+		if (err)
+			return err;
+	}
+
+	if (trap_mask_request & ~*trap_mask_supported)
+		return -EACCES;
+
+	return 0;
+}
+
+int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
+					uint32_t trap_override,
+					uint32_t trap_mask_bits,
+					uint32_t trap_mask_request,
+					uint32_t *trap_mask_prev,
+					uint32_t *trap_mask_supported)
+{
+	int r = 0, i;
+
+	r = kfd_dbg_validate_trap_override_request(target,
+						trap_override,
+						trap_mask_request,
+						trap_mask_supported);
+
+	if (r)
+		return r;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+		pdd->spi_dbg_override = pdd->dev->kfd2kgd->set_wave_launch_trap_override(
+				pdd->dev->adev,
+				pdd->dev->vm_info.last_vmid_kfd,
+				trap_override,
+				trap_mask_bits,
+				trap_mask_request,
+				trap_mask_prev,
+				pdd->spi_dbg_override);
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+		if (!pdd->dev->kfd->shared_resources.enable_mes)
+			r = debug_refresh_runlist(pdd->dev->dqm);
+		else
+			r = kfd_dbg_set_mes_debug_mode(pdd);
+
+		if (r)
+			break;
+	}
+
+	return r;
+}
+
 void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 					uint64_t exception_set_mask)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 6c1054a088723..c9245221aa763 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -42,6 +42,12 @@ int kfd_dbg_trap_disable(struct kfd_process *target);
 int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd,
 			void __user *runtime_info,
 			uint32_t *runtime_info_size);
+int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
+					uint32_t trap_override,
+					uint32_t trap_mask_bits,
+					uint32_t trap_mask_request,
+					uint32_t *trap_mask_prev,
+					uint32_t *trap_mask_supported);
 
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
-- 
GitLab


From aea1b4738bebd8092bd437ce0b03aa9587fc20a7 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 2 May 2022 11:45:05 -0400
Subject: [PATCH 1542/1791] drm/amdkfd: add debug wave launch mode operation

Allow the debugger to set wave behaviour on to either normally operate,
halt at launch, trap on every instruction, terminate immediately or
stall on allocation.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c  | 12 +++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c   |  1 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    | 25 +++++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h    |  3 ++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |  3 +-
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    | 14 +++++++-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 25 +++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |  3 ++
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  3 ++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        | 36 ++++++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |  2 ++
 11 files changed, 124 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index d7881bbd828d6..774ecfc3451a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -107,6 +107,17 @@ static uint32_t kgd_aldebaran_set_wave_launch_trap_override(struct amdgpu_device
 	return data;
 }
 
+static uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, LAUNCH_MODE, wave_launch_mode);
+
+	return data;
+}
+
 const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -129,6 +140,7 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.disable_debug_trap = kgd_aldebaran_disable_debug_trap,
 	.validate_trap_override_request = kgd_aldebaran_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_aldebaran_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index ec25876640018..fbdc1b7b1e42a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -412,6 +412,7 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.disable_debug_trap = kgd_arcturus_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 7ea0362dcab32..a7a6edda557fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -856,6 +856,30 @@ uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev,
 	return 0;
 }
 
+uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid)
+{
+	uint32_t data = 0;
+	bool is_mode_set = !!wave_launch_mode;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
+			VMID_MASK, is_mode_set ? 1 << vmid : 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
+			MODE, is_mode_set ? wave_launch_mode : 0);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data);
+
+	kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 /* kgd_gfx_v10_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -944,6 +968,7 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode,
 	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.program_trap_handler_settings = program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
index 57339fa128079..3a6aca2b0eaa9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
@@ -36,6 +36,9 @@ uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev,
 					     uint32_t trap_mask_request,
 					     uint32_t *trap_mask_prev,
 					     uint32_t kfd_dbg_trap_cntl_prev);
+uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev,
+					 uint8_t wave_launch_mode,
+					 uint32_t vmid);
 void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
 void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t wait_times,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 7120927fed15f..ed36b433a48b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -677,6 +677,7 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.enable_debug_trap = kgd_gfx_v10_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
-	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override
+	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode
 
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index ae0c4707919f1..9711d5128d090 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -726,6 +726,17 @@ static uint32_t kgd_gfx_v11_set_wave_launch_trap_override(struct amdgpu_device *
 	return data;
 }
 
+static uint32_t kgd_gfx_v11_set_wave_launch_mode(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid)
+{
+	uint32_t data = 0;
+
+	data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, LAUNCH_MODE, wave_launch_mode);
+
+	return data;
+}
+
 const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v11,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11,
@@ -745,5 +756,6 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.enable_debug_trap = kgd_gfx_v11_enable_debug_trap,
 	.disable_debug_trap = kgd_gfx_v11_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v11_validate_trap_override_request,
-	.set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override
+	.set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index ce5c256f05176..0acc0c18dfe6c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -792,6 +792,30 @@ uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev,
 	return 0;
 }
 
+uint32_t kgd_gfx_v9_set_wave_launch_mode(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid)
+{
+	uint32_t data = 0;
+	bool is_mode_set = !!wave_launch_mode;
+
+	mutex_lock(&adev->grbm_idx_mutex);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true);
+
+	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
+		VMID_MASK, is_mode_set ? 1 << vmid : 0);
+	data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2,
+		MODE, is_mode_set ? wave_launch_mode : 0);
+	WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data);
+
+	kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false);
+
+	mutex_unlock(&adev->grbm_idx_mutex);
+
+	return 0;
+}
+
 /* kgd_gfx_v9_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -1065,6 +1089,7 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.disable_debug_trap = kgd_gfx_v9_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
+	.set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index 76812ddd35b1e..18f4970ac8e41 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -74,6 +74,9 @@ uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev,
 int kgd_gfx_v9_validate_trap_override_request(struct amdgpu_device *adev,
 					     uint32_t trap_override,
 					     uint32_t *trap_mask_supported);
+uint32_t kgd_gfx_v9_set_wave_launch_mode(struct amdgpu_device *adev,
+					uint8_t wave_launch_mode,
+					uint32_t vmid);
 uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev,
 					     uint32_t vmid,
 					     uint32_t trap_override,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 7fa249807671d..a6570b124b2b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2992,6 +2992,9 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				&args->launch_override.support_request_mask);
 		break;
 	case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE:
+		r = kfd_dbg_trap_set_wave_launch_mode(target,
+				args->launch_mode.launch_mode);
+		break;
 	case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES:
 	case KFD_IOC_DBG_TRAP_RESUME_QUEUES:
 	case KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 733390fb24590..53c3418562d4c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -301,8 +301,10 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 {
 	int i;
 
-	if (!unwind)
+	if (!unwind) {
 		cancel_work_sync(&target->debug_event_workarea);
+		kfd_dbg_trap_set_wave_launch_mode(target, 0);
+	}
 
 	for (i = 0; i < target->n_pdds; i++) {
 		struct kfd_process_device *pdd = target->pdds[i];
@@ -591,6 +593,38 @@ int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
 	return r;
 }
 
+int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target,
+					uint8_t wave_launch_mode)
+{
+	int r = 0, i;
+
+	if (wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_NORMAL &&
+			wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_HALT &&
+			wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_DEBUG)
+		return -EINVAL;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+		pdd->spi_dbg_launch_mode = pdd->dev->kfd2kgd->set_wave_launch_mode(
+				pdd->dev->adev,
+				wave_launch_mode,
+				pdd->dev->vm_info.last_vmid_kfd);
+		amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+		if (!pdd->dev->kfd->shared_resources.enable_mes)
+			r = debug_refresh_runlist(pdd->dev->dqm);
+		else
+			r = kfd_dbg_set_mes_debug_mode(pdd);
+
+		if (r)
+			break;
+	}
+
+	return r;
+}
+
 void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 					uint64_t exception_set_mask)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index c9245221aa763..cb17869437c5b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -48,6 +48,8 @@ int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
 					uint32_t trap_mask_request,
 					uint32_t *trap_mask_prev,
 					uint32_t *trap_mask_supported);
+int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target,
+					uint8_t wave_launch_mode);
 
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
-- 
GitLab


From a70a93fa568b4f05aba548dadb673703eccf5480 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Thu, 5 May 2022 16:15:37 -0400
Subject: [PATCH 1543/1791] drm/amdkfd: add debug suspend and resume process
 queues operation

In order to inspect waves from the saved context at any point during a
debug session, the debugger must be able to preempt queues to trigger
context save by suspending them.

On queue suspend, the KFD will copy the context save header information
so that the debugger can correctly crawl the appropriate size of the saved
context. The debugger must then also be allowed to resume suspended queues.

A queue that is newly created cannot be suspended because queue ids are
recycled after destruction so the debugger needs to know that this has
occurred.  Query functions will be later added that will clear a given
queue of its new queue status.

A queue cannot be destroyed while it is suspended to preserve its saved
context during debugger inspection.  Have queue destruction block while
a queue is suspended and unblocked when it is resumed.  Likewise, if a
queue is about to be destroyed, it cannot be suspended.

Return the number of queues successfully suspended or resumed along with
a per queue status array where the upper bits per queue status show that
the request was invalid (new/destroyed queue suspend request, missing
queue) or an error occurred (HWS in a fatal state so it can't suspend or
resume queues).

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c    |   5 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h    |   1 +
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  11 +
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        |   7 +
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 447 +++++++++++++++++-
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |  10 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c  |  10 +
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c  |  15 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   |  14 +-
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |   5 +-
 .../amd/amdkfd/kfd_process_queue_manager.c    |   1 +
 11 files changed, 512 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 98cd52bb005fd..b4fcad0e62f7e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -772,6 +772,11 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev)
 	return adev->have_atomics_support;
 }
 
+void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev)
+{
+	amdgpu_device_flush_hdp(adev, NULL);
+}
+
 void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset)
 {
 	amdgpu_umc_poison_handler(adev, reset);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index dd740e64e6e15..2d0406bff84ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -322,6 +322,7 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev,
 				      uint64_t *mmap_offset);
 int amdgpu_amdkfd_gpuvm_export_dmabuf(struct kgd_mem *mem,
 				      struct dma_buf **dmabuf);
+void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev);
 int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev,
 				struct tile_config *config);
 void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index a6570b124b2b8..1fae97df7a1ef 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -410,6 +410,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p,
 	pr_debug("Write ptr address   == 0x%016llX\n",
 			args->write_pointer_address);
 
+	kfd_dbg_ev_raise(KFD_EC_MASK(EC_QUEUE_NEW), p, dev, queue_id, false, NULL, 0);
 	return 0;
 
 err_create_queue:
@@ -2996,7 +2997,17 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				args->launch_mode.launch_mode);
 		break;
 	case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES:
+		r = suspend_queues(target,
+				args->suspend_queues.num_queues,
+				args->suspend_queues.grace_period,
+				args->suspend_queues.exception_mask,
+				(uint32_t *)args->suspend_queues.queue_array_ptr);
+
+		break;
 	case KFD_IOC_DBG_TRAP_RESUME_QUEUES:
+		r = resume_queues(target, args->resume_queues.num_queues,
+				(uint32_t *)args->resume_queues.queue_array_ptr);
+		break;
 	case KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH:
 	case KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH:
 	case KFD_IOC_DBG_TRAP_SET_FLAGS:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 53c3418562d4c..f4d3dfb35cb39 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -339,6 +339,13 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 	}
 
 	kfd_dbg_set_workaround(target, false);
+
+	if (!unwind) {
+		int resume_count = resume_queues(target, 0, NULL);
+
+		if (resume_count)
+			pr_debug("Resumed %d queues\n", resume_count);
+	}
 }
 
 static void kfd_dbg_clean_exception_status(struct kfd_process *target)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 44d87943e40ac..bc9e812931655 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -952,6 +952,92 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 	return retval;
 }
 
+/* suspend_single_queue does not lock the dqm like the
+ * evict_process_queues_cpsch or evict_process_queues_nocpsch. You should
+ * lock the dqm before calling, and unlock after calling.
+ *
+ * The reason we don't lock the dqm is because this function may be
+ * called on multiple queues in a loop, so rather than locking/unlocking
+ * multiple times, we will just keep the dqm locked for all of the calls.
+ */
+static int suspend_single_queue(struct device_queue_manager *dqm,
+				      struct kfd_process_device *pdd,
+				      struct queue *q)
+{
+	bool is_new;
+
+	if (q->properties.is_suspended)
+		return 0;
+
+	pr_debug("Suspending PASID %u queue [%i]\n",
+			pdd->process->pasid,
+			q->properties.queue_id);
+
+	is_new = q->properties.exception_status & KFD_EC_MASK(EC_QUEUE_NEW);
+
+	if (is_new || q->properties.is_being_destroyed) {
+		pr_debug("Suspend: skip %s queue id %i\n",
+				is_new ? "new" : "destroyed",
+				q->properties.queue_id);
+		return -EBUSY;
+	}
+
+	q->properties.is_suspended = true;
+	if (q->properties.is_active) {
+		if (dqm->dev->kfd->shared_resources.enable_mes) {
+			int r = remove_queue_mes(dqm, q, &pdd->qpd);
+
+			if (r)
+				return r;
+		}
+
+		decrement_queue_count(dqm, &pdd->qpd, q);
+		q->properties.is_active = false;
+	}
+
+	return 0;
+}
+
+/* resume_single_queue does not lock the dqm like the functions
+ * restore_process_queues_cpsch or restore_process_queues_nocpsch. You should
+ * lock the dqm before calling, and unlock after calling.
+ *
+ * The reason we don't lock the dqm is because this function may be
+ * called on multiple queues in a loop, so rather than locking/unlocking
+ * multiple times, we will just keep the dqm locked for all of the calls.
+ */
+static int resume_single_queue(struct device_queue_manager *dqm,
+				      struct qcm_process_device *qpd,
+				      struct queue *q)
+{
+	struct kfd_process_device *pdd;
+
+	if (!q->properties.is_suspended)
+		return 0;
+
+	pdd = qpd_to_pdd(qpd);
+
+	pr_debug("Restoring from suspend PASID %u queue [%i]\n",
+			    pdd->process->pasid,
+			    q->properties.queue_id);
+
+	q->properties.is_suspended = false;
+
+	if (QUEUE_IS_ACTIVE(q->properties)) {
+		if (dqm->dev->kfd->shared_resources.enable_mes) {
+			int r = add_queue_mes(dqm, q, &pdd->qpd);
+
+			if (r)
+				return r;
+		}
+
+		q->properties.is_active = true;
+		increment_queue_count(dqm, qpd, q);
+	}
+
+	return 0;
+}
+
 static int evict_process_queues_nocpsch(struct device_queue_manager *dqm,
 					struct qcm_process_device *qpd)
 {
@@ -1926,6 +2012,31 @@ static int execute_queues_cpsch(struct device_queue_manager *dqm,
 	return map_queues_cpsch(dqm);
 }
 
+static int wait_on_destroy_queue(struct device_queue_manager *dqm,
+				 struct queue *q)
+{
+	struct kfd_process_device *pdd = kfd_get_process_device_data(q->device,
+								q->process);
+	int ret = 0;
+
+	if (pdd->qpd.is_debug)
+		return ret;
+
+	q->properties.is_being_destroyed = true;
+
+	if (pdd->process->debug_trap_enabled && q->properties.is_suspended) {
+		dqm_unlock(dqm);
+		mutex_unlock(&q->process->mutex);
+		ret = wait_event_interruptible(dqm->destroy_wait,
+						!q->properties.is_suspended);
+
+		mutex_lock(&q->process->mutex);
+		dqm_lock(dqm);
+	}
+
+	return ret;
+}
+
 static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 				struct qcm_process_device *qpd,
 				struct queue *q)
@@ -1945,11 +2056,16 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 				q->properties.queue_id);
 	}
 
-	retval = 0;
-
 	/* remove queue from list to prevent rescheduling after preemption */
 	dqm_lock(dqm);
 
+	retval = wait_on_destroy_queue(dqm, q);
+
+	if (retval) {
+		dqm_unlock(dqm);
+		return retval;
+	}
+
 	if (qpd->is_debug) {
 		/*
 		 * error, currently we do not allow to destroy a queue
@@ -1996,7 +2112,14 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 
 	dqm_unlock(dqm);
 
-	/* Do free_mqd after dqm_unlock(dqm) to avoid circular locking */
+	/*
+	 * Do free_mqd and raise delete event after dqm_unlock(dqm) to avoid
+	 * circular locking
+	 */
+	kfd_dbg_ev_raise(KFD_EC_MASK(EC_DEVICE_QUEUE_DELETE),
+				qpd->pqm->process, q->device,
+				-1, false, NULL, 0);
+
 	mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
 
 	return retval;
@@ -2461,8 +2584,10 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev)
 		goto out_free;
 	}
 
-	if (!dqm->ops.initialize(dqm))
+	if (!dqm->ops.initialize(dqm)) {
+		init_waitqueue_head(&dqm->destroy_wait);
 		return dqm;
+	}
 
 out_free:
 	kfree(dqm);
@@ -2602,6 +2727,320 @@ int release_debug_trap_vmid(struct device_queue_manager *dqm,
 	return r;
 }
 
+#define QUEUE_NOT_FOUND		-1
+/* invalidate queue operation in array */
+static void q_array_invalidate(uint32_t num_queues, uint32_t *queue_ids)
+{
+	int i;
+
+	for (i = 0; i < num_queues; i++)
+		queue_ids[i] |= KFD_DBG_QUEUE_INVALID_MASK;
+}
+
+/* find queue index in array */
+static int q_array_get_index(unsigned int queue_id,
+		uint32_t num_queues,
+		uint32_t *queue_ids)
+{
+	int i;
+
+	for (i = 0; i < num_queues; i++)
+		if (queue_id == (queue_ids[i] & ~KFD_DBG_QUEUE_INVALID_MASK))
+			return i;
+
+	return QUEUE_NOT_FOUND;
+}
+
+struct copy_context_work_handler_workarea {
+	struct work_struct copy_context_work;
+	struct kfd_process *p;
+};
+
+static void copy_context_work_handler (struct work_struct *work)
+{
+	struct copy_context_work_handler_workarea *workarea;
+	struct mqd_manager *mqd_mgr;
+	struct queue *q;
+	struct mm_struct *mm;
+	struct kfd_process *p;
+	uint32_t tmp_ctl_stack_used_size, tmp_save_area_used_size;
+	int i;
+
+	workarea = container_of(work,
+			struct copy_context_work_handler_workarea,
+			copy_context_work);
+
+	p = workarea->p;
+	mm = get_task_mm(p->lead_thread);
+
+	if (!mm)
+		return;
+
+	kthread_use_mm(mm);
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+		struct device_queue_manager *dqm = pdd->dev->dqm;
+		struct qcm_process_device *qpd = &pdd->qpd;
+
+		list_for_each_entry(q, &qpd->queues_list, list) {
+			mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_CP];
+
+			/* We ignore the return value from get_wave_state
+			 * because
+			 * i) right now, it always returns 0, and
+			 * ii) if we hit an error, we would continue to the
+			 *      next queue anyway.
+			 */
+			mqd_mgr->get_wave_state(mqd_mgr,
+					q->mqd,
+					&q->properties,
+					(void __user *)	q->properties.ctx_save_restore_area_address,
+					&tmp_ctl_stack_used_size,
+					&tmp_save_area_used_size);
+		}
+	}
+	kthread_unuse_mm(mm);
+	mmput(mm);
+}
+
+static uint32_t *get_queue_ids(uint32_t num_queues, uint32_t *usr_queue_id_array)
+{
+	size_t array_size = num_queues * sizeof(uint32_t);
+	uint32_t *queue_ids = NULL;
+
+	if (!usr_queue_id_array)
+		return NULL;
+
+	queue_ids = kzalloc(array_size, GFP_KERNEL);
+	if (!queue_ids)
+		return ERR_PTR(-ENOMEM);
+
+	if (copy_from_user(queue_ids, usr_queue_id_array, array_size))
+		return ERR_PTR(-EFAULT);
+
+	return queue_ids;
+}
+
+int resume_queues(struct kfd_process *p,
+		uint32_t num_queues,
+		uint32_t *usr_queue_id_array)
+{
+	uint32_t *queue_ids = NULL;
+	int total_resumed = 0;
+	int i;
+
+	if (usr_queue_id_array) {
+		queue_ids = get_queue_ids(num_queues, usr_queue_id_array);
+
+		if (IS_ERR(queue_ids))
+			return PTR_ERR(queue_ids);
+
+		/* mask all queues as invalid.  unmask per successful request */
+		q_array_invalidate(num_queues, queue_ids);
+	}
+
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+		struct device_queue_manager *dqm = pdd->dev->dqm;
+		struct qcm_process_device *qpd = &pdd->qpd;
+		struct queue *q;
+		int r, per_device_resumed = 0;
+
+		dqm_lock(dqm);
+
+		/* unmask queues that resume or already resumed as valid */
+		list_for_each_entry(q, &qpd->queues_list, list) {
+			int q_idx = QUEUE_NOT_FOUND;
+
+			if (queue_ids)
+				q_idx = q_array_get_index(
+						q->properties.queue_id,
+						num_queues,
+						queue_ids);
+
+			if (!queue_ids || q_idx != QUEUE_NOT_FOUND) {
+				int err = resume_single_queue(dqm, &pdd->qpd, q);
+
+				if (queue_ids) {
+					if (!err) {
+						queue_ids[q_idx] &=
+							~KFD_DBG_QUEUE_INVALID_MASK;
+					} else {
+						queue_ids[q_idx] |=
+							KFD_DBG_QUEUE_ERROR_MASK;
+						break;
+					}
+				}
+
+				if (dqm->dev->kfd->shared_resources.enable_mes) {
+					wake_up_all(&dqm->destroy_wait);
+					if (!err)
+						total_resumed++;
+				} else {
+					per_device_resumed++;
+				}
+			}
+		}
+
+		if (!per_device_resumed) {
+			dqm_unlock(dqm);
+			continue;
+		}
+
+		r = execute_queues_cpsch(dqm,
+					KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES,
+					0,
+					USE_DEFAULT_GRACE_PERIOD);
+		if (r) {
+			pr_err("Failed to resume process queues\n");
+			if (queue_ids) {
+				list_for_each_entry(q, &qpd->queues_list, list) {
+					int q_idx = q_array_get_index(
+							q->properties.queue_id,
+							num_queues,
+							queue_ids);
+
+					/* mask queue as error on resume fail */
+					if (q_idx != QUEUE_NOT_FOUND)
+						queue_ids[q_idx] |=
+							KFD_DBG_QUEUE_ERROR_MASK;
+				}
+			}
+		} else {
+			wake_up_all(&dqm->destroy_wait);
+			total_resumed += per_device_resumed;
+		}
+
+		dqm_unlock(dqm);
+	}
+
+	if (queue_ids) {
+		if (copy_to_user((void __user *)usr_queue_id_array, queue_ids,
+				num_queues * sizeof(uint32_t)))
+			pr_err("copy_to_user failed on queue resume\n");
+
+		kfree(queue_ids);
+	}
+
+	return total_resumed;
+}
+
+int suspend_queues(struct kfd_process *p,
+			uint32_t num_queues,
+			uint32_t grace_period,
+			uint64_t exception_clear_mask,
+			uint32_t *usr_queue_id_array)
+{
+	uint32_t *queue_ids = get_queue_ids(num_queues, usr_queue_id_array);
+	int total_suspended = 0;
+	int i;
+
+	if (IS_ERR(queue_ids))
+		return PTR_ERR(queue_ids);
+
+	/* mask all queues as invalid.  umask on successful request */
+	q_array_invalidate(num_queues, queue_ids);
+
+	for (i = 0; i < p->n_pdds; i++) {
+		struct kfd_process_device *pdd = p->pdds[i];
+		struct device_queue_manager *dqm = pdd->dev->dqm;
+		struct qcm_process_device *qpd = &pdd->qpd;
+		struct queue *q;
+		int r, per_device_suspended = 0;
+
+		mutex_lock(&p->event_mutex);
+		dqm_lock(dqm);
+
+		/* unmask queues that suspend or already suspended */
+		list_for_each_entry(q, &qpd->queues_list, list) {
+			int q_idx = q_array_get_index(q->properties.queue_id,
+							num_queues,
+							queue_ids);
+
+			if (q_idx != QUEUE_NOT_FOUND) {
+				int err = suspend_single_queue(dqm, pdd, q);
+				bool is_mes = dqm->dev->kfd->shared_resources.enable_mes;
+
+				if (!err) {
+					queue_ids[q_idx] &= ~KFD_DBG_QUEUE_INVALID_MASK;
+					if (exception_clear_mask && is_mes)
+						q->properties.exception_status &=
+							~exception_clear_mask;
+
+					if (is_mes)
+						total_suspended++;
+					else
+						per_device_suspended++;
+				} else if (err != -EBUSY) {
+					r = err;
+					queue_ids[q_idx] |= KFD_DBG_QUEUE_ERROR_MASK;
+					break;
+				}
+			}
+		}
+
+		if (!per_device_suspended) {
+			dqm_unlock(dqm);
+			mutex_unlock(&p->event_mutex);
+			if (total_suspended)
+				amdgpu_amdkfd_debug_mem_fence(dqm->dev->adev);
+			continue;
+		}
+
+		r = execute_queues_cpsch(dqm,
+			KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
+			grace_period);
+
+		if (r)
+			pr_err("Failed to suspend process queues.\n");
+		else
+			total_suspended += per_device_suspended;
+
+		list_for_each_entry(q, &qpd->queues_list, list) {
+			int q_idx = q_array_get_index(q->properties.queue_id,
+						num_queues, queue_ids);
+
+			if (q_idx == QUEUE_NOT_FOUND)
+				continue;
+
+			/* mask queue as error on suspend fail */
+			if (r)
+				queue_ids[q_idx] |= KFD_DBG_QUEUE_ERROR_MASK;
+			else if (exception_clear_mask)
+				q->properties.exception_status &=
+							~exception_clear_mask;
+		}
+
+		dqm_unlock(dqm);
+		mutex_unlock(&p->event_mutex);
+		amdgpu_device_flush_hdp(dqm->dev->adev, NULL);
+	}
+
+	if (total_suspended) {
+		struct copy_context_work_handler_workarea copy_context_worker;
+
+		INIT_WORK_ONSTACK(
+				&copy_context_worker.copy_context_work,
+				copy_context_work_handler);
+
+		copy_context_worker.p = p;
+
+		schedule_work(&copy_context_worker.copy_context_work);
+
+
+		flush_work(&copy_context_worker.copy_context_work);
+		destroy_work_on_stack(&copy_context_worker.copy_context_work);
+	}
+
+	if (copy_to_user((void __user *)usr_queue_id_array, queue_ids,
+			num_queues * sizeof(uint32_t)))
+		pr_err("copy_to_user failed on queue suspend\n");
+
+	kfree(queue_ids);
+
+	return total_suspended;
+}
+
 int debug_lock_and_unmap(struct device_queue_manager *dqm)
 {
 	int r;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index bb75d93712eb2..d4e6dbffe8c21 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -263,6 +263,8 @@ struct device_queue_manager {
 	uint32_t		current_logical_xcc_start;
 
 	uint32_t		wait_times;
+
+	wait_queue_head_t	destroy_wait;
 };
 
 void device_queue_manager_init_cik(
@@ -290,6 +292,14 @@ int reserve_debug_trap_vmid(struct device_queue_manager *dqm,
 			struct qcm_process_device *qpd);
 int release_debug_trap_vmid(struct device_queue_manager *dqm,
 			struct qcm_process_device *qpd);
+int suspend_queues(struct kfd_process *p,
+			uint32_t num_queues,
+			uint32_t grace_period,
+			uint64_t exception_clear_mask,
+			uint32_t *usr_queue_id_array);
+int resume_queues(struct kfd_process *p,
+		uint32_t num_queues,
+		uint32_t *usr_queue_id_array);
 int debug_lock_and_unmap(struct device_queue_manager *dqm);
 int debug_map_and_unlock(struct device_queue_manager *dqm);
 int debug_refresh_runlist(struct device_queue_manager *dqm);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
index a0ac4f2fe6b58..94c0fc2e57b7f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c
@@ -237,6 +237,7 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 			  u32 *save_area_used_size)
 {
 	struct v10_compute_mqd *m;
+	struct kfd_context_save_area_header header;
 
 	m = get_mqd(mqd);
 
@@ -255,6 +256,15 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 	 * accessible to user mode
 	 */
 
+	header.wave_state.control_stack_size = *ctl_stack_used_size;
+	header.wave_state.wave_state_size = *save_area_used_size;
+
+	header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset;
+	header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset;
+
+	if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state)))
+		return -EFAULT;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
index 9a9b4e8535162..31fec5e70d135 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c
@@ -291,7 +291,7 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 			  u32 *save_area_used_size)
 {
 	struct v11_compute_mqd *m;
-	/*struct mqd_user_context_save_area_header header;*/
+	struct kfd_context_save_area_header header;
 
 	m = get_mqd(mqd);
 
@@ -309,16 +309,15 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 	 * it's part of the context save area that is already
 	 * accessible to user mode
 	 */
-/*
-	header.control_stack_size = *ctl_stack_used_size;
-	header.wave_state_size = *save_area_used_size;
+	header.wave_state.control_stack_size = *ctl_stack_used_size;
+	header.wave_state.wave_state_size = *save_area_used_size;
 
-	header.wave_state_offset = m->cp_hqd_wg_state_offset;
-	header.control_stack_offset = m->cp_hqd_cntl_stack_offset;
+	header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset;
+	header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset;
 
-	if (copy_to_user(ctl_stack, &header, sizeof(header)))
+	if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state)))
 		return -EFAULT;
-*/
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 5b87c244e909f..601bb9f68048c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -311,6 +311,7 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 			  u32 *save_area_used_size)
 {
 	struct v9_mqd *m;
+	struct kfd_context_save_area_header header;
 
 	/* Control stack is located one page after MQD. */
 	void *mqd_ctl_stack = (void *)((uintptr_t)mqd + PAGE_SIZE);
@@ -322,7 +323,18 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd,
 	*save_area_used_size = m->cp_hqd_wg_state_offset -
 		m->cp_hqd_cntl_stack_size;
 
-	if (copy_to_user(ctl_stack, mqd_ctl_stack, m->cp_hqd_cntl_stack_size))
+	header.wave_state.control_stack_size = *ctl_stack_used_size;
+	header.wave_state.wave_state_size = *save_area_used_size;
+
+	header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset;
+	header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset;
+
+	if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state)))
+		return -EFAULT;
+
+	if (copy_to_user(ctl_stack + m->cp_hqd_cntl_stack_offset,
+				mqd_ctl_stack + m->cp_hqd_cntl_stack_offset,
+				*ctl_stack_used_size))
 		return -EFAULT;
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index cd2d56e5cdf05..05da43bf233a5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -510,6 +510,8 @@ struct queue_properties {
 	uint32_t doorbell_off;
 	bool is_interop;
 	bool is_evicted;
+	bool is_suspended;
+	bool is_being_destroyed;
 	bool is_active;
 	bool is_gws;
 	uint32_t pm4_target_xcc;
@@ -535,7 +537,8 @@ struct queue_properties {
 #define QUEUE_IS_ACTIVE(q) ((q).queue_size > 0 &&	\
 			    (q).queue_address != 0 &&	\
 			    (q).queue_percent > 0 &&	\
-			    !(q).is_evicted)
+			    !(q).is_evicted &&		\
+			    !(q).is_suspended)
 
 enum mqd_update_flag {
 	UPDATE_FLAG_DBG_WA_ENABLE = 1,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 70852a200d8f7..01ccab607a699 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -187,6 +187,7 @@ static int init_user_queue(struct process_queue_manager *pqm,
 
 	/* Doorbell initialized in user space*/
 	q_properties->doorbell_ptr = NULL;
+	q_properties->exception_status = KFD_EC_MASK(EC_QUEUE_NEW);
 
 	/* let DQM handle it*/
 	q_properties->vmid = 0;
-- 
GitLab


From e0f85f4690d089cc1a60337decafb1acf7eec45e Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 6 May 2022 14:58:55 -0400
Subject: [PATCH 1544/1791] drm/amdkfd: add debug set and clear address watch
 points operation

Shader read, write and atomic memory operations can be alerted to the
debugger as an address watch exception.

Allow the debugger to pass in a watch point to a particular memory
address per device.

Note that there exists only 4 watch points per devices to date, so have
the KFD keep track of what watch points are allocated or not.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c  |  51 +++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c   |   2 +
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c    |  78 ++++++++++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h    |   8 ++
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c  |   5 +-
 .../drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c    |  52 ++++++-
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c |  77 ++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h |   8 ++
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  24 ++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c        | 136 ++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h        |   8 +-
 drivers/gpu/drm/amd/amdkfd/kfd_device.c       |   2 +
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |   6 +-
 13 files changed, 452 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index 774ecfc3451a6..efd6a72aab4eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -118,6 +118,55 @@ static uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev,
 	return data;
 }
 
+#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H)
+static uint32_t kgd_gfx_aldebaran_set_address_watch(
+					struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid)
+{
+	uint32_t watch_address_high;
+	uint32_t watch_address_low;
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+	watch_address_low = lower_32_bits(watch_address);
+	watch_address_high = upper_32_bits(watch_address) & 0xffff;
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MODE,
+			watch_mode);
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MASK,
+			watch_address_mask >> 6);
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			1);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_H) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_high);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_L) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_low);
+
+	return watch_address_cntl;
+}
+
+uint32_t kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id)
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping,
@@ -141,6 +190,8 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = {
 	.validate_trap_override_request = kgd_aldebaran_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_aldebaran_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_aldebaran_set_address_watch,
+	.clear_address_watch = kgd_gfx_aldebaran_clear_address_watch,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index fbdc1b7b1e42a..6df215aba4c4e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -413,6 +413,8 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = {
 	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_v9_set_address_watch,
+	.clear_address_watch = kgd_gfx_v9_clear_address_watch,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index a7a6edda557fa..8ad7a7779e147 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -880,6 +880,82 @@ uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev,
 	return 0;
 }
 
+#define TCP_WATCH_STRIDE (mmTCP_WATCH1_ADDR_H - mmTCP_WATCH0_ADDR_H)
+uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid)
+{
+	uint32_t watch_address_high;
+	uint32_t watch_address_low;
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+
+	watch_address_low = lower_32_bits(watch_address);
+	watch_address_high = upper_32_bits(watch_address) & 0xffff;
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VMID,
+			debug_vmid);
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MODE,
+			watch_mode);
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MASK,
+			watch_address_mask >> 7);
+
+	/* Turning off this watch point until we set all the registers */
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			0);
+
+	WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_H) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_high);
+
+	WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_L) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_low);
+
+	/* Enable the watch point */
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			1);
+
+	WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	return 0;
+}
+
+uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id)
+{
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+
+	WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	return 0;
+}
+
+
 /* kgd_gfx_v10_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -969,6 +1045,8 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = {
 	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_v10_set_address_watch,
+	.clear_address_watch = kgd_gfx_v10_clear_address_watch,
 	.get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info,
 	.program_trap_handler_settings = program_trap_handler_settings,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
index 3a6aca2b0eaa9..e6b70196071ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h
@@ -39,6 +39,14 @@ uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev,
 uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev,
 					 uint8_t wave_launch_mode,
 					 uint32_t vmid);
+uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid);
+uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id);
 void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
 void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t wait_times,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index ed36b433a48b2..8c8437a4383f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -678,6 +678,7 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = {
 	.disable_debug_trap = kgd_gfx_v10_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override,
-	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode
-
+	.set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_v10_set_address_watch,
+	.clear_address_watch = kgd_gfx_v10_clear_address_watch
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 9711d5128d090..52efa690a3c21 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -737,6 +737,54 @@ static uint32_t kgd_gfx_v11_set_wave_launch_mode(struct amdgpu_device *adev,
 	return data;
 }
 
+#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H)
+static uint32_t kgd_gfx_v11_set_address_watch(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid)
+{
+	uint32_t watch_address_high;
+	uint32_t watch_address_low;
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+	watch_address_low = lower_32_bits(watch_address);
+	watch_address_high = upper_32_bits(watch_address) & 0xffff;
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MODE,
+			watch_mode);
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MASK,
+			watch_address_mask >> 7);
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			1);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_H) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_high);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_L) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_low);
+
+	return watch_address_cntl;
+}
+
+uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id)
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.program_sh_mem_settings = program_sh_mem_settings_v11,
 	.set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11,
@@ -757,5 +805,7 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = {
 	.disable_debug_trap = kgd_gfx_v11_disable_debug_trap,
 	.validate_trap_override_request = kgd_gfx_v11_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override,
-	.set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode
+	.set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_v11_set_address_watch,
+	.clear_address_watch = kgd_gfx_v11_clear_address_watch
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 0acc0c18dfe6c..51d93fb13ea3b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -816,6 +816,81 @@ uint32_t kgd_gfx_v9_set_wave_launch_mode(struct amdgpu_device *adev,
 	return 0;
 }
 
+#define TCP_WATCH_STRIDE (mmTCP_WATCH1_ADDR_H - mmTCP_WATCH0_ADDR_H)
+uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid)
+{
+	uint32_t watch_address_high;
+	uint32_t watch_address_low;
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+
+	watch_address_low = lower_32_bits(watch_address);
+	watch_address_high = upper_32_bits(watch_address) & 0xffff;
+
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VMID,
+			debug_vmid);
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MODE,
+			watch_mode);
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			MASK,
+			watch_address_mask >> 6);
+
+	/* Turning off this watch point until we set all the registers */
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			0);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_H) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_high);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_L) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_low);
+
+	/* Enable the watch point */
+	watch_address_cntl = REG_SET_FIELD(watch_address_cntl,
+			TCP_WATCH0_CNTL,
+			VALID,
+			1);
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	return 0;
+}
+
+uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id)
+{
+	uint32_t watch_address_cntl;
+
+	watch_address_cntl = 0;
+
+	WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) +
+			(watch_id * TCP_WATCH_STRIDE)),
+			watch_address_cntl);
+
+	return 0;
+}
+
 /* kgd_gfx_v9_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values
  * The values read are:
  *     ib_offload_wait_time     -- Wait Count for Indirect Buffer Offloads.
@@ -1090,6 +1165,8 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = {
 	.validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request,
 	.set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override,
 	.set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode,
+	.set_address_watch = kgd_gfx_v9_set_address_watch,
+	.clear_address_watch = kgd_gfx_v9_clear_address_watch,
 	.get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times,
 	.build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info,
 	.get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
index 18f4970ac8e41..5f54bff0db496 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h
@@ -84,6 +84,14 @@ uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev,
 					     uint32_t trap_mask_request,
 					     uint32_t *trap_mask_prev,
 					     uint32_t kfd_dbg_trap_cntl_prev);
+uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t watch_id,
+					uint32_t watch_mode,
+					uint32_t debug_vmid);
+uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev,
+					uint32_t watch_id);
 void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times);
 void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev,
 					       uint32_t wait_times,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 1fae97df7a1ef..016724c82928d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2880,6 +2880,7 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 	struct mm_struct *mm = NULL;
 	struct pid *pid = NULL;
 	struct kfd_process *target = NULL;
+	struct kfd_process_device *pdd = NULL;
 	int r = 0;
 
 	if (sched_policy == KFD_SCHED_POLICY_NO_HWS) {
@@ -2957,6 +2958,20 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		goto unlock_out;
 	}
 
+	if (args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH ||
+	    args->op == KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH) {
+		int user_gpu_id = kfd_process_get_user_gpu_id(target,
+				args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH ?
+					args->set_node_address_watch.gpu_id :
+					args->clear_node_address_watch.gpu_id);
+
+		pdd = kfd_process_device_data_by_id(target, user_gpu_id);
+		if (user_gpu_id == -EINVAL || !pdd) {
+			r = -ENODEV;
+			goto unlock_out;
+		}
+	}
+
 	switch (args->op) {
 	case KFD_IOC_DBG_TRAP_ENABLE:
 		if (target != p)
@@ -3009,7 +3024,16 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				(uint32_t *)args->resume_queues.queue_array_ptr);
 		break;
 	case KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH:
+		r = kfd_dbg_trap_set_dev_address_watch(pdd,
+				args->set_node_address_watch.address,
+				args->set_node_address_watch.mask,
+				&args->set_node_address_watch.id,
+				args->set_node_address_watch.mode);
+		break;
 	case KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH:
+		r = kfd_dbg_trap_clear_dev_address_watch(pdd,
+				args->clear_node_address_watch.id);
+		break;
 	case KFD_IOC_DBG_TRAP_SET_FLAGS:
 	case KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT:
 	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index f4d3dfb35cb39..4b36cc8b5fb7c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -24,6 +24,8 @@
 #include "kfd_device_queue_manager.h"
 #include <linux/file.h>
 
+#define MAX_WATCH_ADDRESSES	4
+
 void debug_event_write_work_handler(struct work_struct *work)
 {
 	struct kfd_process *process;
@@ -289,6 +291,139 @@ int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
 						pdd->watch_points, flags);
 }
 
+#define KFD_DEBUGGER_INVALID_WATCH_POINT_ID -1
+static int kfd_dbg_get_dev_watch_id(struct kfd_process_device *pdd, int *watch_id)
+{
+	int i;
+
+	*watch_id = KFD_DEBUGGER_INVALID_WATCH_POINT_ID;
+
+	spin_lock(&pdd->dev->kfd->watch_points_lock);
+
+	for (i = 0; i < MAX_WATCH_ADDRESSES; i++) {
+		/* device watchpoint in use so skip */
+		if ((pdd->dev->kfd->alloc_watch_ids >> i) & 0x1)
+			continue;
+
+		pdd->alloc_watch_ids |= 0x1 << i;
+		pdd->dev->kfd->alloc_watch_ids |= 0x1 << i;
+		*watch_id = i;
+		spin_unlock(&pdd->dev->kfd->watch_points_lock);
+		return 0;
+	}
+
+	spin_unlock(&pdd->dev->kfd->watch_points_lock);
+
+	return -ENOMEM;
+}
+
+static void kfd_dbg_clear_dev_watch_id(struct kfd_process_device *pdd, int watch_id)
+{
+	spin_lock(&pdd->dev->kfd->watch_points_lock);
+
+	/* process owns device watch point so safe to clear */
+	if ((pdd->alloc_watch_ids >> watch_id) & 0x1) {
+		pdd->alloc_watch_ids &= ~(0x1 << watch_id);
+		pdd->dev->kfd->alloc_watch_ids &= ~(0x1 << watch_id);
+	}
+
+	spin_unlock(&pdd->dev->kfd->watch_points_lock);
+}
+
+static bool kfd_dbg_owns_dev_watch_id(struct kfd_process_device *pdd, int watch_id)
+{
+	bool owns_watch_id = false;
+
+	spin_lock(&pdd->dev->kfd->watch_points_lock);
+	owns_watch_id = watch_id < MAX_WATCH_ADDRESSES &&
+			((pdd->alloc_watch_ids >> watch_id) & 0x1);
+
+	spin_unlock(&pdd->dev->kfd->watch_points_lock);
+
+	return owns_watch_id;
+}
+
+int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd,
+					uint32_t watch_id)
+{
+	int r;
+
+	if (!kfd_dbg_owns_dev_watch_id(pdd, watch_id))
+		return -EINVAL;
+
+	if (!pdd->dev->kfd->shared_resources.enable_mes) {
+		r = debug_lock_and_unmap(pdd->dev->dqm);
+		if (r)
+			return r;
+	}
+
+	amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+	pdd->watch_points[watch_id] = pdd->dev->kfd2kgd->clear_address_watch(
+							pdd->dev->adev,
+							watch_id);
+	amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+	if (!pdd->dev->kfd->shared_resources.enable_mes)
+		r = debug_map_and_unlock(pdd->dev->dqm);
+	else
+		r = kfd_dbg_set_mes_debug_mode(pdd);
+
+	kfd_dbg_clear_dev_watch_id(pdd, watch_id);
+
+	return r;
+}
+
+int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t *watch_id,
+					uint32_t watch_mode)
+{
+	int r = kfd_dbg_get_dev_watch_id(pdd, watch_id);
+
+	if (r)
+		return r;
+
+	if (!pdd->dev->kfd->shared_resources.enable_mes) {
+		r = debug_lock_and_unmap(pdd->dev->dqm);
+		if (r) {
+			kfd_dbg_clear_dev_watch_id(pdd, *watch_id);
+			return r;
+		}
+	}
+
+	amdgpu_gfx_off_ctrl(pdd->dev->adev, false);
+	pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch(
+				pdd->dev->adev,
+				watch_address,
+				watch_address_mask,
+				*watch_id,
+				watch_mode,
+				pdd->dev->vm_info.last_vmid_kfd);
+	amdgpu_gfx_off_ctrl(pdd->dev->adev, true);
+
+	if (!pdd->dev->kfd->shared_resources.enable_mes)
+		r = debug_map_and_unlock(pdd->dev->dqm);
+	else
+		r = kfd_dbg_set_mes_debug_mode(pdd);
+
+	/* HWS is broken so no point in HW rollback but release the watchpoint anyways */
+	if (r)
+		kfd_dbg_clear_dev_watch_id(pdd, *watch_id);
+
+	return 0;
+}
+
+static void kfd_dbg_clear_process_address_watch(struct kfd_process *target)
+{
+	int i, j;
+
+	for (i = 0; i < target->n_pdds; i++)
+		for (j = 0; j < MAX_WATCH_ADDRESSES; j++)
+			kfd_dbg_trap_clear_dev_address_watch(target->pdds[i], j);
+}
+
+
 /* kfd_dbg_trap_deactivate:
  *	target: target process
  *	unwind: If this is unwinding a failed kfd_dbg_trap_enable()
@@ -303,6 +438,7 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 
 	if (!unwind) {
 		cancel_work_sync(&target->debug_event_workarea);
+		kfd_dbg_clear_process_address_watch(target);
 		kfd_dbg_trap_set_wave_launch_mode(target, 0);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index cb17869437c5b..7f0757c2af2ce 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -50,7 +50,13 @@ int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target,
 					uint32_t *trap_mask_supported);
 int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target,
 					uint8_t wave_launch_mode);
-
+int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd,
+					uint32_t watch_id);
+int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
+					uint64_t watch_address,
+					uint32_t watch_address_mask,
+					uint32_t *watch_id,
+					uint32_t watch_mode);
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
 					unsigned int queue_id,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 2c36bb5786338..9fc9d32cb5796 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -811,6 +811,8 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 	if (kfd_resume_iommu(kfd))
 		goto kfd_resume_iommu_error;
 
+	spin_lock_init(&kfd->watch_points_lock);
+
 	kfd->init_complete = true;
 	dev_info(kfd_device, "added device %x:%x\n", kfd->adev->pdev->vendor,
 		 kfd->adev->pdev->device);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 05da43bf233a5..8ec87bc8ba823 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -381,6 +381,10 @@ struct kfd_dev {
 
 	struct kfd_node *nodes[MAX_KFD_NODES];
 	unsigned int num_nodes;
+
+	/* Track per device allocated watch points */
+	uint32_t alloc_watch_ids;
+	spinlock_t watch_points_lock;
 };
 
 enum kfd_mempool {
@@ -833,6 +837,7 @@ struct kfd_process_device {
 	uint32_t spi_dbg_override;
 	uint32_t spi_dbg_launch_mode;
 	uint32_t watch_points[4];
+	uint32_t alloc_watch_ids;
 
 	/*
 	 * If this process has been checkpointed before, then the user
@@ -989,7 +994,6 @@ struct kfd_process {
 	struct semaphore runtime_enable_sema;
 	bool is_runtime_retry;
 	struct kfd_runtime_info runtime_info;
-
 };
 
 #define KFD_PROCESS_TABLE_SIZE 5 /* bits: 32 entries */
-- 
GitLab


From 103d5f08ff42b666c61c350be2c3e724c1646918 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 9 May 2022 10:51:56 -0400
Subject: [PATCH 1545/1791] drm/amdkfd: add debug set flags operation

Allow the debugger to set single memory and single ALU operations.

Some exceptions are imprecise (memory violations, address watch) in the
sense that a trap occurs only when the exception interrupt occurs and
not at the non-halting faulty instruction.  Trap temporaries 0 & 1 save
the program counter address, which means that these values will not point
to the faulty instruction address but to whenever the interrupt was
raised.

Setting the Single Memory Operations flag will inject an automatic wait
on every memory operation instruction forcing imprecise memory exceptions
to become precise at the cost of performance.  This setting is not
permitted on debug devices that support only a global setting of this
option.

Return the previous set flags to the debugger as well.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  2 +
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 58 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |  1 +
 3 files changed, 61 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 016724c82928d..5ee38614ed9b2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3035,6 +3035,8 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				args->clear_node_address_watch.id);
 		break;
 	case KFD_IOC_DBG_TRAP_SET_FLAGS:
+		r = kfd_dbg_trap_set_flags(target, &args->set_flags.flags);
+		break;
 	case KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT:
 	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
 	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 4b36cc8b5fb7c..43c3170998d38 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -23,6 +23,7 @@
 #include "kfd_debug.h"
 #include "kfd_device_queue_manager.h"
 #include <linux/file.h>
+#include <uapi/linux/kfd_ioctl.h>
 
 #define MAX_WATCH_ADDRESSES	4
 
@@ -423,6 +424,59 @@ static void kfd_dbg_clear_process_address_watch(struct kfd_process *target)
 			kfd_dbg_trap_clear_dev_address_watch(target->pdds[i], j);
 }
 
+int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags)
+{
+	uint32_t prev_flags = target->dbg_flags;
+	int i, r = 0, rewind_count = 0;
+
+	for (i = 0; i < target->n_pdds; i++) {
+		if (!kfd_dbg_is_per_vmid_supported(target->pdds[i]->dev) &&
+			(*flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP)) {
+			*flags = prev_flags;
+			return -EACCES;
+		}
+	}
+
+	target->dbg_flags = *flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP;
+	*flags = prev_flags;
+	for (i = 0; i < target->n_pdds; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+
+		if (!kfd_dbg_is_per_vmid_supported(pdd->dev))
+			continue;
+
+		if (!pdd->dev->kfd->shared_resources.enable_mes)
+			r = debug_refresh_runlist(pdd->dev->dqm);
+		else
+			r = kfd_dbg_set_mes_debug_mode(pdd);
+
+		if (r) {
+			target->dbg_flags = prev_flags;
+			break;
+		}
+
+		rewind_count++;
+	}
+
+	/* Rewind flags */
+	if (r) {
+		target->dbg_flags = prev_flags;
+
+		for (i = 0; i < rewind_count; i++) {
+			struct kfd_process_device *pdd = target->pdds[i];
+
+			if (!kfd_dbg_is_per_vmid_supported(pdd->dev))
+				continue;
+
+			if (!pdd->dev->kfd->shared_resources.enable_mes)
+				debug_refresh_runlist(pdd->dev->dqm);
+			else
+				kfd_dbg_set_mes_debug_mode(pdd);
+		}
+	}
+
+	return r;
+}
 
 /* kfd_dbg_trap_deactivate:
  *	target: target process
@@ -437,9 +491,13 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 	int i;
 
 	if (!unwind) {
+		uint32_t flags = 0;
+
 		cancel_work_sync(&target->debug_event_workarea);
 		kfd_dbg_clear_process_address_watch(target);
 		kfd_dbg_trap_set_wave_launch_mode(target, 0);
+
+		kfd_dbg_trap_set_flags(target, &flags);
 	}
 
 	for (i = 0; i < target->n_pdds; i++) {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 7f0757c2af2ce..ef8e9f7f17165 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -57,6 +57,7 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
 					uint32_t watch_address_mask,
 					uint32_t *watch_id,
 					uint32_t watch_mode);
+int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags);
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
 					unsigned int queue_id,
-- 
GitLab


From 5bc20c224bcb863571e8831cdbba23cd61b10ac3 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 9 May 2022 11:10:32 -0400
Subject: [PATCH 1546/1791] drm/amdkfd: add debug query event operation

Allow the debugger to query a single queue, device and process
exception.
The KFD should also return the GPU or Queue id of the exception.
The debugger also has the option of clearing exceptions after
being queried.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  6 +++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 64 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |  5 ++
 3 files changed, 75 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 5ee38614ed9b2..498859259b556 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3038,6 +3038,12 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		r = kfd_dbg_trap_set_flags(target, &args->set_flags.flags);
 		break;
 	case KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT:
+		r = kfd_dbg_ev_query_debug_event(target,
+				&args->query_debug_event.queue_id,
+				&args->query_debug_event.gpu_id,
+				args->query_debug_event.exception_mask,
+				&args->query_debug_event.exception_mask);
+		break;
 	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
 	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
 	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 43c3170998d38..e9530e682e85c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -27,6 +27,70 @@
 
 #define MAX_WATCH_ADDRESSES	4
 
+int kfd_dbg_ev_query_debug_event(struct kfd_process *process,
+		      unsigned int *queue_id,
+		      unsigned int *gpu_id,
+		      uint64_t exception_clear_mask,
+		      uint64_t *event_status)
+{
+	struct process_queue_manager *pqm;
+	struct process_queue_node *pqn;
+	int i;
+
+	if (!(process && process->debug_trap_enabled))
+		return -ENODATA;
+
+	mutex_lock(&process->event_mutex);
+	*event_status = 0;
+	*queue_id = 0;
+	*gpu_id = 0;
+
+	/* find and report queue events */
+	pqm = &process->pqm;
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
+		uint64_t tmp = process->exception_enable_mask;
+
+		if (!pqn->q)
+			continue;
+
+		tmp &= pqn->q->properties.exception_status;
+
+		if (!tmp)
+			continue;
+
+		*event_status = pqn->q->properties.exception_status;
+		*queue_id = pqn->q->properties.queue_id;
+		*gpu_id = pqn->q->device->id;
+		pqn->q->properties.exception_status &= ~exception_clear_mask;
+		goto out;
+	}
+
+	/* find and report device events */
+	for (i = 0; i < process->n_pdds; i++) {
+		struct kfd_process_device *pdd = process->pdds[i];
+		uint64_t tmp = process->exception_enable_mask
+						& pdd->exception_status;
+
+		if (!tmp)
+			continue;
+
+		*event_status = pdd->exception_status;
+		*gpu_id = pdd->dev->id;
+		pdd->exception_status &= ~exception_clear_mask;
+		goto out;
+	}
+
+	/* report process events */
+	if (process->exception_enable_mask & process->exception_status) {
+		*event_status = process->exception_status;
+		process->exception_status &= ~exception_clear_mask;
+	}
+
+out:
+	mutex_unlock(&process->event_mutex);
+	return *event_status ? 0 : -EAGAIN;
+}
+
 void debug_event_write_work_handler(struct work_struct *work)
 {
 	struct kfd_process *process;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index ef8e9f7f17165..e78f954c06849 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -27,6 +27,11 @@
 
 void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count);
 int kfd_dbg_trap_activate(struct kfd_process *target);
+int kfd_dbg_ev_query_debug_event(struct kfd_process *process,
+			unsigned int *queue_id,
+			unsigned int *gpu_id,
+			uint64_t exception_clear_mask,
+			uint64_t *event_status);
 bool kfd_set_dbg_ev_from_interrupt(struct kfd_node *dev,
 				   unsigned int pasid,
 				   uint32_t doorbell_id,
-- 
GitLab


From 2b36de971d25daa2ad287114ae3ca11a8f8d49d7 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 9 May 2022 13:37:36 -0400
Subject: [PATCH 1547/1791] drm/amdkfd: add debug query exception info
 operation

Allow the debugger to query additional info based on an exception code.
For device exceptions, it's currently only memory violation information.
For process exceptions, it's currently only runtime information.
Queue exception only report the queue exception status.

The debugger has the option of clearing the target exception on query.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |   7 ++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 120 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |   6 ++
 3 files changed, 133 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 498859259b556..b7ee79b5220a8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3045,6 +3045,13 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				&args->query_debug_event.exception_mask);
 		break;
 	case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO:
+		r = kfd_dbg_trap_query_exception_info(target,
+				args->query_exception_info.source_id,
+				args->query_exception_info.exception_code,
+				args->query_exception_info.clear_exception,
+				(void __user *)args->query_exception_info.info_ptr,
+				&args->query_exception_info.info_size);
+		break;
 	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
 	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
 		pr_warn("Debug op %i not supported yet\n", args->op);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index e9530e682e85c..24e2b285448ab 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -890,6 +890,126 @@ int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target,
 	return r;
 }
 
+int kfd_dbg_trap_query_exception_info(struct kfd_process *target,
+		uint32_t source_id,
+		uint32_t exception_code,
+		bool clear_exception,
+		void __user *info,
+		uint32_t *info_size)
+{
+	bool found = false;
+	int r = 0;
+	uint32_t copy_size, actual_info_size = 0;
+	uint64_t *exception_status_ptr = NULL;
+
+	if (!target)
+		return -EINVAL;
+
+	if (!info || !info_size)
+		return -EINVAL;
+
+	mutex_lock(&target->event_mutex);
+
+	if (KFD_DBG_EC_TYPE_IS_QUEUE(exception_code)) {
+		/* Per queue exceptions */
+		struct queue *queue = NULL;
+		int i;
+
+		for (i = 0; i < target->n_pdds; i++) {
+			struct kfd_process_device *pdd = target->pdds[i];
+			struct qcm_process_device *qpd = &pdd->qpd;
+
+			list_for_each_entry(queue, &qpd->queues_list, list) {
+				if (!found && queue->properties.queue_id == source_id) {
+					found = true;
+					break;
+				}
+			}
+			if (found)
+				break;
+		}
+
+		if (!found) {
+			r = -EINVAL;
+			goto out;
+		}
+
+		if (!(queue->properties.exception_status & KFD_EC_MASK(exception_code))) {
+			r = -ENODATA;
+			goto out;
+		}
+		exception_status_ptr = &queue->properties.exception_status;
+	} else if (KFD_DBG_EC_TYPE_IS_DEVICE(exception_code)) {
+		/* Per device exceptions */
+		struct kfd_process_device *pdd = NULL;
+		int i;
+
+		for (i = 0; i < target->n_pdds; i++) {
+			pdd = target->pdds[i];
+			if (pdd->dev->id == source_id) {
+				found = true;
+				break;
+			}
+		}
+
+		if (!found) {
+			r = -EINVAL;
+			goto out;
+		}
+
+		if (!(pdd->exception_status & KFD_EC_MASK(exception_code))) {
+			r = -ENODATA;
+			goto out;
+		}
+
+		if (exception_code == EC_DEVICE_MEMORY_VIOLATION) {
+			copy_size = min((size_t)(*info_size), pdd->vm_fault_exc_data_size);
+
+			if (copy_to_user(info, pdd->vm_fault_exc_data, copy_size)) {
+				r = -EFAULT;
+				goto out;
+			}
+			actual_info_size = pdd->vm_fault_exc_data_size;
+			if (clear_exception) {
+				kfree(pdd->vm_fault_exc_data);
+				pdd->vm_fault_exc_data = NULL;
+				pdd->vm_fault_exc_data_size = 0;
+			}
+		}
+		exception_status_ptr = &pdd->exception_status;
+	} else if (KFD_DBG_EC_TYPE_IS_PROCESS(exception_code)) {
+		/* Per process exceptions */
+		if (!(target->exception_status & KFD_EC_MASK(exception_code))) {
+			r = -ENODATA;
+			goto out;
+		}
+
+		if (exception_code == EC_PROCESS_RUNTIME) {
+			copy_size = min((size_t)(*info_size), sizeof(target->runtime_info));
+
+			if (copy_to_user(info, (void *)&target->runtime_info, copy_size)) {
+				r = -EFAULT;
+				goto out;
+			}
+
+			actual_info_size = sizeof(target->runtime_info);
+		}
+
+		exception_status_ptr = &target->exception_status;
+	} else {
+		pr_debug("Bad exception type [%i]\n", exception_code);
+		r = -EINVAL;
+		goto out;
+	}
+
+	*info_size = actual_info_size;
+	if (clear_exception)
+		*exception_status_ptr &= ~KFD_EC_MASK(exception_code);
+out:
+	mutex_unlock(&target->event_mutex);
+	return r;
+}
+
 void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 					uint64_t exception_set_mask)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index e78f954c06849..234e2ccafa878 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -63,6 +63,12 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd,
 					uint32_t *watch_id,
 					uint32_t watch_mode);
 int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags);
+int kfd_dbg_trap_query_exception_info(struct kfd_process *target,
+		uint32_t source_id,
+		uint32_t exception_code,
+		bool clear_exception,
+		void __user *info,
+		uint32_t *info_size);
 int kfd_dbg_send_exception_to_runtime(struct kfd_process *p,
 					unsigned int dev_id,
 					unsigned int queue_id,
-- 
GitLab


From b17bd5dbf64677682a3bca249c64521d5eabcb38 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 10 May 2022 11:15:29 -0400
Subject: [PATCH 1548/1791] drm/amdkfd: add debug queue snapshot operation

Allow the debugger to get a snapshot of a specified number of queues
containing various queue property information that is copied to the
debugger.

Since the debugger doesn't know how many queues exist at any given time,
allow the debugger to pass the requested number of snapshots as 0 to get
the actual number of potential snapshots to use for a subsequent snapshot
request for actual information.

To prevent future ABI breakage, pass in the requested entry_size.
The KFD will return it's own entry_size in case the debugger still wants
log the information in a core dump on sizing failure.

Also allow the debugger to clear exceptions when doing a snapshot.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c      |  6 +++
 .../drm/amd/amdkfd/kfd_device_queue_manager.c | 36 +++++++++++++++++
 .../drm/amd/amdkfd/kfd_device_queue_manager.h |  3 ++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h         |  5 +++
 .../amd/amdkfd/kfd_process_queue_manager.c    | 40 +++++++++++++++++++
 5 files changed, 90 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index b7ee79b5220a8..24066756e478a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3053,6 +3053,12 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				&args->query_exception_info.info_size);
 		break;
 	case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT:
+		r = pqm_get_queue_snapshot(&target->pqm,
+				args->queue_snapshot.exception_mask,
+				(void __user *)args->queue_snapshot.snapshot_buf_ptr,
+				&args->queue_snapshot.num_queues,
+				&args->queue_snapshot.entry_size);
+		break;
 	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
 		pr_warn("Debug op %i not supported yet\n", args->op);
 		r = -EACCES;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index bc9e812931655..0c1be91a87c69 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -3041,6 +3041,42 @@ int suspend_queues(struct kfd_process *p,
 	return total_suspended;
 }
 
+static uint32_t set_queue_type_for_user(struct queue_properties *q_props)
+{
+	switch (q_props->type) {
+	case KFD_QUEUE_TYPE_COMPUTE:
+		return q_props->format == KFD_QUEUE_FORMAT_PM4
+					? KFD_IOC_QUEUE_TYPE_COMPUTE
+					: KFD_IOC_QUEUE_TYPE_COMPUTE_AQL;
+	case KFD_QUEUE_TYPE_SDMA:
+		return KFD_IOC_QUEUE_TYPE_SDMA;
+	case KFD_QUEUE_TYPE_SDMA_XGMI:
+		return KFD_IOC_QUEUE_TYPE_SDMA_XGMI;
+	default:
+		WARN_ONCE(true, "queue type not recognized!");
+		return 0xffffffff;
+	};
+}
+
+void set_queue_snapshot_entry(struct queue *q,
+			      uint64_t exception_clear_mask,
+			      struct kfd_queue_snapshot_entry *qss_entry)
+{
+	qss_entry->ring_base_address = q->properties.queue_address;
+	qss_entry->write_pointer_address = (uint64_t)q->properties.write_ptr;
+	qss_entry->read_pointer_address = (uint64_t)q->properties.read_ptr;
+	qss_entry->ctx_save_restore_address =
+				q->properties.ctx_save_restore_area_address;
+	qss_entry->ctx_save_restore_area_size =
+				q->properties.ctx_save_restore_area_size;
+	qss_entry->exception_status = q->properties.exception_status;
+	qss_entry->queue_id = q->properties.queue_id;
+	qss_entry->gpu_id = q->device->id;
+	qss_entry->ring_size = (uint32_t)q->properties.queue_size;
+	qss_entry->queue_type = set_queue_type_for_user(&q->properties);
+	q->properties.exception_status &= ~exception_clear_mask;
+}
+
 int debug_lock_and_unmap(struct device_queue_manager *dqm)
 {
 	int r;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index d4e6dbffe8c21..7dd4b177219de 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -300,6 +300,9 @@ int suspend_queues(struct kfd_process *p,
 int resume_queues(struct kfd_process *p,
 		uint32_t num_queues,
 		uint32_t *usr_queue_id_array);
+void set_queue_snapshot_entry(struct queue *q,
+			      uint64_t exception_clear_mask,
+			      struct kfd_queue_snapshot_entry *qss_entry);
 int debug_lock_and_unmap(struct device_queue_manager *dqm);
 int debug_map_and_unlock(struct device_queue_manager *dqm);
 int debug_refresh_runlist(struct device_queue_manager *dqm);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 8ec87bc8ba823..023b17e0116be 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1355,6 +1355,11 @@ int pqm_get_wave_state(struct process_queue_manager *pqm,
 		       void __user *ctl_stack,
 		       u32 *ctl_stack_used_size,
 		       u32 *save_area_used_size);
+int pqm_get_queue_snapshot(struct process_queue_manager *pqm,
+			   uint64_t exception_clear_mask,
+			   void __user *buf,
+			   int *num_qss_entries,
+			   uint32_t *entry_size);
 
 int amdkfd_fence_wait_timeout(uint64_t *fence_addr,
 			      uint64_t fence_value,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 01ccab607a699..9ad1a2186a245 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -585,6 +585,46 @@ int pqm_get_wave_state(struct process_queue_manager *pqm,
 						       save_area_used_size);
 }
 
+int pqm_get_queue_snapshot(struct process_queue_manager *pqm,
+			   uint64_t exception_clear_mask,
+			   void __user *buf,
+			   int *num_qss_entries,
+			   uint32_t *entry_size)
+{
+	struct process_queue_node *pqn;
+	struct kfd_queue_snapshot_entry src;
+	uint32_t tmp_entry_size = *entry_size, tmp_qss_entries = *num_qss_entries;
+	int r = 0;
+
+	*num_qss_entries = 0;
+	if (!(*entry_size))
+		return -EINVAL;
+
+	*entry_size = min_t(size_t, *entry_size, sizeof(struct kfd_queue_snapshot_entry));
+	mutex_lock(&pqm->process->event_mutex);
+
+	memset(&src, 0, sizeof(src));
+
+	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
+		if (!pqn->q)
+			continue;
+
+		if (*num_qss_entries < tmp_qss_entries) {
+			set_queue_snapshot_entry(pqn->q, exception_clear_mask, &src);
+
+			if (copy_to_user(buf, &src, *entry_size)) {
+				r = -EFAULT;
+				break;
+			}
+			buf += tmp_entry_size;
+		}
+		*num_qss_entries += 1;
+	}
+
+	mutex_unlock(&pqm->process->event_mutex);
+	return r;
+}
+
 static int get_queue_data_sizes(struct kfd_process_device *pdd,
 				struct queue *q,
 				uint32_t *mqd_size,
-- 
GitLab


From 12976e6a5ab8fc3766c0304d72f7eec81a109b55 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 10 May 2022 12:47:45 -0400
Subject: [PATCH 1549/1791] drm/amdkfd: add debug device snapshot operation

Similar to queue snapshot, return an array of device information using
an entry_size check and return.
Unlike queue snapshots, the debugger needs to pass to correct number of
devices that exist.  If it fails to do so, the KFD will return the
number of actual devices so that the debugger can make a subsequent
successful call.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c |  7 ++-
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c   | 73 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_debug.h   |  5 ++
 3 files changed, 83 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 24066756e478a..f54ff5c3387df 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3060,8 +3060,11 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 				&args->queue_snapshot.entry_size);
 		break;
 	case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT:
-		pr_warn("Debug op %i not supported yet\n", args->op);
-		r = -EACCES;
+		r = kfd_dbg_trap_device_snapshot(target,
+				args->device_snapshot.exception_mask,
+				(void __user *)args->device_snapshot.snapshot_buf_ptr,
+				&args->device_snapshot.num_devices,
+				&args->device_snapshot.entry_size);
 		break;
 	default:
 		pr_err("Invalid option: %i\n", args->op);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 24e2b285448ab..125274445f439 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -22,6 +22,7 @@
 
 #include "kfd_debug.h"
 #include "kfd_device_queue_manager.h"
+#include "kfd_topology.h"
 #include <linux/file.h>
 #include <uapi/linux/kfd_ioctl.h>
 
@@ -1010,6 +1011,78 @@ int kfd_dbg_trap_query_exception_info(struct kfd_process *target,
 	return r;
 }
 
+int kfd_dbg_trap_device_snapshot(struct kfd_process *target,
+		uint64_t exception_clear_mask,
+		void __user *user_info,
+		uint32_t *number_of_device_infos,
+		uint32_t *entry_size)
+{
+	struct kfd_dbg_device_info_entry device_info;
+	uint32_t tmp_entry_size = *entry_size, tmp_num_devices;
+	int i, r = 0;
+
+	if (!(target && user_info && number_of_device_infos && entry_size))
+		return -EINVAL;
+
+	tmp_num_devices = min_t(size_t, *number_of_device_infos, target->n_pdds);
+	*number_of_device_infos = target->n_pdds;
+	*entry_size = min_t(size_t, *entry_size, sizeof(device_info));
+
+	if (!tmp_num_devices)
+		return 0;
+
+	memset(&device_info, 0, sizeof(device_info));
+
+	mutex_lock(&target->event_mutex);
+
+	/* Run over all pdd of the process */
+	for (i = 0; i < tmp_num_devices; i++) {
+		struct kfd_process_device *pdd = target->pdds[i];
+		struct kfd_topology_device *topo_dev = kfd_topology_device_by_id(pdd->dev->id);
+
+		device_info.gpu_id = pdd->dev->id;
+		device_info.exception_status = pdd->exception_status;
+		device_info.lds_base = pdd->lds_base;
+		device_info.lds_limit = pdd->lds_limit;
+		device_info.scratch_base = pdd->scratch_base;
+		device_info.scratch_limit = pdd->scratch_limit;
+		device_info.gpuvm_base = pdd->gpuvm_base;
+		device_info.gpuvm_limit = pdd->gpuvm_limit;
+		device_info.location_id = topo_dev->node_props.location_id;
+		device_info.vendor_id = topo_dev->node_props.vendor_id;
+		device_info.device_id = topo_dev->node_props.device_id;
+		device_info.revision_id = pdd->dev->adev->pdev->revision;
+		device_info.subsystem_vendor_id = pdd->dev->adev->pdev->subsystem_vendor;
+		device_info.subsystem_device_id = pdd->dev->adev->pdev->subsystem_device;
+		device_info.fw_version = pdd->dev->kfd->mec_fw_version;
+		device_info.gfx_target_version =
+			topo_dev->node_props.gfx_target_version;
+		device_info.simd_count = topo_dev->node_props.simd_count;
+		device_info.max_waves_per_simd =
+			topo_dev->node_props.max_waves_per_simd;
+		device_info.array_count = topo_dev->node_props.array_count;
+		device_info.simd_arrays_per_engine =
+			topo_dev->node_props.simd_arrays_per_engine;
+		device_info.num_xcc = NUM_XCC(pdd->dev->xcc_mask);
+		device_info.capability = topo_dev->node_props.capability;
+		device_info.debug_prop = topo_dev->node_props.debug_prop;
+
+		if (exception_clear_mask)
+			pdd->exception_status &= ~exception_clear_mask;
+
+		if (copy_to_user(user_info, &device_info, *entry_size)) {
+			r = -EFAULT;
+			break;
+		}
+
+		user_info += tmp_entry_size;
+	}
+
+	mutex_unlock(&target->event_mutex);
+
+	return r;
+}
+
 void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 					uint64_t exception_set_mask)
 {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
index 234e2ccafa878..a289e59ceb797 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h
@@ -81,6 +81,11 @@ static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev)
 }
 
 void debug_event_write_work_handler(struct work_struct *work);
+int kfd_dbg_trap_device_snapshot(struct kfd_process *target,
+		uint64_t exception_clear_mask,
+		void __user *user_info,
+		uint32_t *number_of_device_infos,
+		uint32_t *entry_size);
 
 void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 					uint64_t exception_set_mask);
-- 
GitLab


From a159afdad2f6b97e4d18549cff2b53d17e68a412 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 10 May 2022 12:51:26 -0400
Subject: [PATCH 1550/1791] drm/amdkfd: bump kfd ioctl  minor version for debug
 api availability

Bump the minor version to declare debugging capability is now
available.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 1 -
 include/uapi/linux/kfd_ioctl.h           | 3 ++-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index f54ff5c3387df..cce2abe12e1b8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2984,7 +2984,6 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		if (!r)
 			target->exception_enable_mask = args->enable.exception_mask;
 
-		pr_warn("Debug functions limited\n");
 		break;
 	case KFD_IOC_DBG_TRAP_DISABLE:
 		r = kfd_dbg_trap_disable(target);
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 32913d674d38b..1781e7669982b 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -38,9 +38,10 @@
  * - 1.10 - Add SMI profiler event log
  * - 1.11 - Add unified memory for ctx save/restore area
  * - 1.12 - Add DMA buf export ioctl
+ * - 1.13 - Add debugger API
  */
 #define KFD_IOCTL_MAJOR_VERSION 1
-#define KFD_IOCTL_MINOR_VERSION 12
+#define KFD_IOCTL_MINOR_VERSION 13
 
 struct kfd_ioctl_get_version_args {
 	__u32 major_version;	/* from KFD */
-- 
GitLab


From 8f4f5f0b901a444c2317ef0fb29f35bc296daa55 Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Tue, 11 Apr 2023 11:49:09 +0800
Subject: [PATCH 1551/1791] drm/amd/pm: fulfill SMU13 OD settings init and
 restore

Gfxclk fmin/fmax, Uclk fmin/fmax and Gfx v/f curve voltage offset
OD settings are supported for SMU13.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c            |  26 +++-
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c    |  13 +-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c  | 129 +++++++++++++++++-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c  | 128 ++++++++++++++++-
 4 files changed, 286 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index da0da03569e85..a57952b93e73f 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -678,7 +678,12 @@ static ssize_t amdgpu_set_pp_table(struct device *dev,
  *   clock labeled OD_MCLK
  *
  * - three <frequency, voltage> points labeled OD_VDDC_CURVE.
- *   They can be used to calibrate the sclk voltage curve.
+ *   They can be used to calibrate the sclk voltage curve. This is
+ *   available for Vega20 and NV1X.
+ *
+ * - voltage offset for the six anchor points of the v/f curve labeled
+ *   OD_VDDC_CURVE. They can be used to calibrate the v/f curve. This
+ *   is only availabe for some SMU13 ASICs.
  *
  * - voltage offset(in mV) applied on target voltage calculation.
  *   This is available for Sienna Cichlid, Navy Flounder and Dimgrey
@@ -719,12 +724,19 @@ static ssize_t amdgpu_set_pp_table(struct device *dev,
  *   E.g., "p 2 0 800" would set the minimum core clock on core
  *   2 to 800Mhz.
  *
- *   For sclk voltage curve, enter the new values by writing a
- *   string that contains "vc point clock voltage" to the file. The
- *   points are indexed by 0, 1 and 2. E.g., "vc 0 300 600" will
- *   update point1 with clock set as 300Mhz and voltage as
- *   600mV. "vc 2 1000 1000" will update point3 with clock set
- *   as 1000Mhz and voltage 1000mV.
+ *   For sclk voltage curve,
+ *     - For NV1X, enter the new values by writing a string that
+ *       contains "vc point clock voltage" to the file. The points
+ *       are indexed by 0, 1 and 2. E.g., "vc 0 300 600" will update
+ *       point1 with clock set as 300Mhz and voltage as 600mV. "vc 2
+ *       1000 1000" will update point3 with clock set as 1000Mhz and
+ *       voltage 1000mV.
+ *     - For SMU13 ASICs, enter the new values by writing a string that
+ *       contains "vc anchor_point_index voltage_offset" to the file.
+ *       There are total six anchor points defined on the v/f curve with
+ *       index as 0 - 5.
+ *       - "vc 0 10" will update the voltage offset for point1 as 10mv.
+ *       - "vc 5 -10" will update the voltage offset for point6 as -10mv.
  *
  *   To update the voltage offset applied for gfxclk/voltage calculation,
  *   enter the new value by writing a string that contains "vo offset".
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index 70db36d459743..e52c563f0dacc 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -467,17 +467,26 @@ int smu_v13_0_init_smc_tables(struct smu_context *smu)
 			ret = -ENOMEM;
 			goto err3_out;
 		}
+
+		smu_table->user_overdrive_table =
+			kzalloc(tables[SMU_TABLE_OVERDRIVE].size, GFP_KERNEL);
+		if (!smu_table->user_overdrive_table) {
+			ret = -ENOMEM;
+			goto err4_out;
+		}
 	}
 
 	smu_table->combo_pptable =
 		kzalloc(tables[SMU_TABLE_COMBO_PPTABLE].size, GFP_KERNEL);
 	if (!smu_table->combo_pptable) {
 		ret = -ENOMEM;
-		goto err4_out;
+		goto err5_out;
 	}
 
 	return 0;
 
+err5_out:
+	kfree(smu_table->user_overdrive_table);
 err4_out:
 	kfree(smu_table->boot_overdrive_table);
 err3_out:
@@ -497,12 +506,14 @@ int smu_v13_0_fini_smc_tables(struct smu_context *smu)
 
 	kfree(smu_table->gpu_metrics_table);
 	kfree(smu_table->combo_pptable);
+	kfree(smu_table->user_overdrive_table);
 	kfree(smu_table->boot_overdrive_table);
 	kfree(smu_table->overdrive_table);
 	kfree(smu_table->max_sustainable_clocks);
 	kfree(smu_table->driver_pptable);
 	smu_table->gpu_metrics_table = NULL;
 	smu_table->combo_pptable = NULL;
+	smu_table->user_overdrive_table = NULL;
 	smu_table->boot_overdrive_table = NULL;
 	smu_table->overdrive_table = NULL;
 	smu_table->max_sustainable_clocks = NULL;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index f4783e685bf83..faa772db106c1 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -237,6 +237,7 @@ static struct cmn2asic_mapping smu_v13_0_0_table_map[SMU_TABLE_COUNT] = {
 	[SMU_TABLE_COMBO_PPTABLE] = {1, TABLE_COMBO_PPTABLE},
 	TAB_MAP(I2C_COMMANDS),
 	TAB_MAP(ECCINFO),
+	TAB_MAP(OVERDRIVE),
 };
 
 static struct cmn2asic_mapping smu_v13_0_0_pwr_src_map[SMU_POWER_SOURCE_COUNT] = {
@@ -331,6 +332,11 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu)
 	struct smu_13_0_0_powerplay_table *powerplay_table =
 		table_context->power_play_table;
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&pptable->SkuTable.OverDriveLimitsBasicMax;
+	const OverDriveLimits_t * const overdrive_lowerlimits =
+				&pptable->SkuTable.OverDriveLimitsMin;
 
 	if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_HARDWAREDC)
 		smu->dc_controlled_by_gpio = true;
@@ -342,6 +348,10 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu)
 	if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_MACO)
 		smu_baco->maco_support = true;
 
+	if (!overdrive_lowerlimits->FeatureCtrlMask ||
+	    !overdrive_upperlimits->FeatureCtrlMask)
+		smu->od_enabled = false;
+
 	table_context->thermal_controller_type =
 		powerplay_table->thermal_controller_type;
 
@@ -461,7 +471,7 @@ static int smu_v13_0_0_tables_init(struct smu_context *smu)
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
-	SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTable_t),
+	SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTableExternal_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE,
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
@@ -1384,6 +1394,121 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu,
 	return sizeof(struct gpu_metrics_v1_3);
 }
 
+static int smu_v13_0_0_get_overdrive_table(struct smu_context *smu,
+					   OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   false);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
+
+	return ret;
+}
+
+static int smu_v13_0_0_upload_overdrive_table(struct smu_context *smu,
+					      OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   true);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
+
+	return ret;
+}
+
+static void smu_v13_0_0_dump_od_table(struct smu_context *smu,
+				      OverDriveTableExternal_t *od_table)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
+						     od_table->OverDriveTable.GfxclkFmax);
+	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
+						   od_table->OverDriveTable.UclkFmax);
+}
+
+static int smu_v13_0_0_set_default_od_settings(struct smu_context *smu)
+{
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.overdrive_table;
+	OverDriveTableExternal_t *boot_od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.boot_overdrive_table;
+	OverDriveTableExternal_t *user_od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.user_overdrive_table;
+	OverDriveTableExternal_t user_od_table_bak;
+	int ret = 0;
+	int i;
+
+	ret = smu_v13_0_0_get_overdrive_table(smu, boot_od_table);
+	if (ret)
+		return ret;
+
+	smu_v13_0_0_dump_od_table(smu, boot_od_table);
+
+	memcpy(od_table,
+	       boot_od_table,
+	       sizeof(OverDriveTableExternal_t));
+
+	/*
+	 * For S3/S4/Runpm resume, we need to setup those overdrive tables again,
+	 * but we have to preserve user defined values in "user_od_table".
+	 */
+	if (!smu->adev->in_suspend) {
+		memcpy(user_od_table,
+		       boot_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		smu->user_dpm_profile.user_od = false;
+	} else if (smu->user_dpm_profile.user_od) {
+		memcpy(&user_od_table_bak,
+		       user_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		memcpy(user_od_table,
+		       boot_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		user_od_table->OverDriveTable.GfxclkFmin =
+				user_od_table_bak.OverDriveTable.GfxclkFmin;
+		user_od_table->OverDriveTable.GfxclkFmax =
+				user_od_table_bak.OverDriveTable.GfxclkFmax;
+		user_od_table->OverDriveTable.UclkFmin =
+				user_od_table_bak.OverDriveTable.UclkFmin;
+		user_od_table->OverDriveTable.UclkFmax =
+				user_od_table_bak.OverDriveTable.UclkFmax;
+		for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++)
+			user_od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i] =
+				user_od_table_bak.OverDriveTable.VoltageOffsetPerZoneBoundary[i];
+	}
+
+	return 0;
+}
+
+static int smu_v13_0_0_restore_user_od_settings(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTableExternal_t *od_table = table_context->overdrive_table;
+	OverDriveTableExternal_t *user_od_table = table_context->user_overdrive_table;
+	int res;
+
+	user_od_table->OverDriveTable.FeatureCtrlMask = 1U << PP_OD_FEATURE_GFXCLK_BIT |
+							1U << PP_OD_FEATURE_UCLK_BIT |
+							1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT;
+	res = smu_v13_0_0_upload_overdrive_table(smu, user_od_table);
+	user_od_table->OverDriveTable.FeatureCtrlMask = 0;
+	if (res == 0)
+		memcpy(od_table, user_od_table, sizeof(OverDriveTableExternal_t));
+
+	return res;
+}
+
 static int smu_v13_0_0_populate_umd_state_clk(struct smu_context *smu)
 {
 	struct smu_13_0_dpm_context *dpm_context =
@@ -2150,6 +2275,8 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.notify_memory_pool_location = smu_v13_0_notify_memory_pool_location,
 	.get_gpu_metrics = smu_v13_0_0_get_gpu_metrics,
 	.set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range,
+	.set_default_od_settings = smu_v13_0_0_set_default_od_settings,
+	.restore_user_od_settings = smu_v13_0_0_restore_user_od_settings,
 	.init_pptable_microcode = smu_v13_0_init_pptable_microcode,
 	.populate_umd_state_clk = smu_v13_0_0_populate_umd_state_clk,
 	.set_performance_level = smu_v13_0_set_performance_level,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index cf6827179fd13..20a8be3bb49a3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -206,6 +206,7 @@ static struct cmn2asic_mapping smu_v13_0_7_table_map[SMU_TABLE_COUNT] = {
 	TAB_MAP(DRIVER_SMU_CONFIG),
 	TAB_MAP(ACTIVITY_MONITOR_COEFF),
 	[SMU_TABLE_COMBO_PPTABLE] = {1, TABLE_COMBO_PPTABLE},
+	TAB_MAP(OVERDRIVE),
 };
 
 static struct cmn2asic_mapping smu_v13_0_7_pwr_src_map[SMU_POWER_SOURCE_COUNT] = {
@@ -322,6 +323,10 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu)
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
 	PPTable_t *smc_pptable = table_context->driver_pptable;
 	BoardTable_t *BoardTable = &smc_pptable->BoardTable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&smc_pptable->SkuTable.OverDriveLimitsBasicMax;
+	const OverDriveLimits_t * const overdrive_lowerlimits =
+				&smc_pptable->SkuTable.OverDriveLimitsMin;
 
 	if (powerplay_table->platform_caps & SMU_13_0_7_PP_PLATFORM_CAP_HARDWAREDC)
 		smu->dc_controlled_by_gpio = true;
@@ -333,6 +338,10 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu)
 	if (smu_baco->platform_support && (BoardTable->HsrEnabled || BoardTable->VddqOffEnabled))
 		smu_baco->maco_support = true;
 
+	if (!overdrive_lowerlimits->FeatureCtrlMask ||
+	    !overdrive_upperlimits->FeatureCtrlMask)
+		smu->od_enabled = false;
+
 	table_context->thermal_controller_type =
 		powerplay_table->thermal_controller_type;
 
@@ -479,7 +488,7 @@ static int smu_v13_0_7_tables_init(struct smu_context *smu)
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
-	SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTable_t),
+	SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTableExternal_t),
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE,
 		       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
@@ -1371,6 +1380,121 @@ static ssize_t smu_v13_0_7_get_gpu_metrics(struct smu_context *smu,
 	return sizeof(struct gpu_metrics_v1_3);
 }
 
+static int smu_v13_0_7_get_overdrive_table(struct smu_context *smu,
+					   OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   false);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
+
+	return ret;
+}
+
+static int smu_v13_0_7_upload_overdrive_table(struct smu_context *smu,
+					      OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   true);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
+
+	return ret;
+}
+
+static void smu_v13_0_7_dump_od_table(struct smu_context *smu,
+				      OverDriveTableExternal_t *od_table)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
+						     od_table->OverDriveTable.GfxclkFmax);
+	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
+						   od_table->OverDriveTable.UclkFmax);
+}
+
+static int smu_v13_0_7_set_default_od_settings(struct smu_context *smu)
+{
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.overdrive_table;
+	OverDriveTableExternal_t *boot_od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.boot_overdrive_table;
+	OverDriveTableExternal_t *user_od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.user_overdrive_table;
+	OverDriveTableExternal_t user_od_table_bak;
+	int ret = 0;
+	int i;
+
+	ret = smu_v13_0_7_get_overdrive_table(smu, boot_od_table);
+	if (ret)
+		return ret;
+
+	smu_v13_0_7_dump_od_table(smu, boot_od_table);
+
+	memcpy(od_table,
+	       boot_od_table,
+	       sizeof(OverDriveTableExternal_t));
+
+	/*
+	 * For S3/S4/Runpm resume, we need to setup those overdrive tables again,
+	 * but we have to preserve user defined values in "user_od_table".
+	 */
+	if (!smu->adev->in_suspend) {
+		memcpy(user_od_table,
+		       boot_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		smu->user_dpm_profile.user_od = false;
+	} else if (smu->user_dpm_profile.user_od) {
+		memcpy(&user_od_table_bak,
+		       user_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		memcpy(user_od_table,
+		       boot_od_table,
+		       sizeof(OverDriveTableExternal_t));
+		user_od_table->OverDriveTable.GfxclkFmin =
+				user_od_table_bak.OverDriveTable.GfxclkFmin;
+		user_od_table->OverDriveTable.GfxclkFmax =
+				user_od_table_bak.OverDriveTable.GfxclkFmax;
+		user_od_table->OverDriveTable.UclkFmin =
+				user_od_table_bak.OverDriveTable.UclkFmin;
+		user_od_table->OverDriveTable.UclkFmax =
+				user_od_table_bak.OverDriveTable.UclkFmax;
+		for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++)
+			user_od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i] =
+				user_od_table_bak.OverDriveTable.VoltageOffsetPerZoneBoundary[i];
+	}
+
+	return 0;
+}
+
+static int smu_v13_0_7_restore_user_od_settings(struct smu_context *smu)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTableExternal_t *od_table = table_context->overdrive_table;
+	OverDriveTableExternal_t *user_od_table = table_context->user_overdrive_table;
+	int res;
+
+	user_od_table->OverDriveTable.FeatureCtrlMask = 1U << PP_OD_FEATURE_GFXCLK_BIT |
+							1U << PP_OD_FEATURE_UCLK_BIT |
+							1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT;
+	res = smu_v13_0_7_upload_overdrive_table(smu, user_od_table);
+	user_od_table->OverDriveTable.FeatureCtrlMask = 0;
+	if (res == 0)
+		memcpy(od_table, user_od_table, sizeof(OverDriveTableExternal_t));
+
+	return res;
+}
+
 static int smu_v13_0_7_populate_umd_state_clk(struct smu_context *smu)
 {
 	struct smu_13_0_dpm_context *dpm_context =
@@ -1760,6 +1884,8 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.notify_memory_pool_location = smu_v13_0_notify_memory_pool_location,
 	.get_gpu_metrics = smu_v13_0_7_get_gpu_metrics,
 	.set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range,
+	.set_default_od_settings = smu_v13_0_7_set_default_od_settings,
+	.restore_user_od_settings = smu_v13_0_7_restore_user_od_settings,
 	.set_performance_level = smu_v13_0_set_performance_level,
 	.gfx_off_control = smu_v13_0_gfx_off_control,
 	.get_fan_speed_pwm = smu_v13_0_7_get_fan_speed_pwm,
-- 
GitLab


From 2e8452ea4ef6406927e4c5a71d1a7ed6881c5a9b Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Mon, 8 May 2023 16:57:02 +0800
Subject: [PATCH 1552/1791] drm/amd/pm: fulfill the OD support for SMU13.0.0

Fulfill the interfaces for OD settings retrieving and setting.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c  | 445 ++++++++++++++++--
 1 file changed, 402 insertions(+), 43 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index faa772db106c1..5ac5ea770c1c0 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -1032,17 +1032,119 @@ static int smu_v13_0_0_get_current_clk_freq_by_table(struct smu_context *smu,
 						value);
 }
 
+static bool smu_v13_0_0_is_od_feature_supported(struct smu_context *smu,
+						int od_feature_bit)
+{
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&pptable->SkuTable.OverDriveLimitsBasicMax;
+
+	return overdrive_upperlimits->FeatureCtrlMask & (1U << od_feature_bit);
+}
+
+static void smu_v13_0_0_get_od_setting_limits(struct smu_context *smu,
+					      int od_feature_bit,
+					      bool lower_boundary,
+					      int32_t *min,
+					      int32_t *max)
+{
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&pptable->SkuTable.OverDriveLimitsBasicMax;
+	const OverDriveLimits_t * const overdrive_lowerlimits =
+				&pptable->SkuTable.OverDriveLimitsMin;
+	int32_t od_min_setting, od_max_setting;
+
+	switch (od_feature_bit) {
+	case PP_OD_FEATURE_GFXCLK_BIT:
+		if (lower_boundary) {
+			od_min_setting = overdrive_lowerlimits->GfxclkFmin;
+			od_max_setting = overdrive_upperlimits->GfxclkFmin;
+		} else {
+			od_min_setting = overdrive_lowerlimits->GfxclkFmax;
+			od_max_setting = overdrive_upperlimits->GfxclkFmax;
+		}
+		break;
+	case PP_OD_FEATURE_UCLK_BIT:
+		if (lower_boundary) {
+			od_min_setting = overdrive_lowerlimits->UclkFmin;
+			od_max_setting = overdrive_upperlimits->UclkFmin;
+		} else {
+			od_min_setting = overdrive_lowerlimits->UclkFmax;
+			od_max_setting = overdrive_upperlimits->UclkFmax;
+		}
+		break;
+	case PP_OD_FEATURE_GFX_VF_CURVE_BIT:
+		od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary;
+		od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary;
+		break;
+	default:
+		break;
+	}
+
+	if (min)
+		*min = od_min_setting;
+	if (max)
+		*max = od_max_setting;
+}
+
+static void smu_v13_0_0_dump_od_table(struct smu_context *smu,
+				      OverDriveTableExternal_t *od_table)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
+						     od_table->OverDriveTable.GfxclkFmax);
+	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
+						   od_table->OverDriveTable.UclkFmax);
+}
+
+static int smu_v13_0_0_get_overdrive_table(struct smu_context *smu,
+					   OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   false);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
+
+	return ret;
+}
+
+static int smu_v13_0_0_upload_overdrive_table(struct smu_context *smu,
+					      OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   true);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
+
+	return ret;
+}
+
 static int smu_v13_0_0_print_clk_levels(struct smu_context *smu,
 					enum smu_clk_type clk_type,
 					char *buf)
 {
 	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
 	struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context;
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.overdrive_table;
 	struct smu_13_0_dpm_table *single_dpm_table;
 	struct smu_13_0_pcie_table *pcie_table;
 	const int link_width[] = {0, 1, 2, 4, 8, 12, 16};
 	uint32_t gen_speed, lane_width;
 	int i, curr_freq, size = 0;
+	int32_t min_value, max_value;
 	int ret = 0;
 
 	smu_cmn_get_sysfs_buf(&buf, &size);
@@ -1159,6 +1261,89 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu,
 					"*" : "");
 		break;
 
+	case SMU_OD_SCLK:
+		if (!smu_v13_0_0_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_GFXCLK_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_SCLK:\n");
+		size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMhz\n",
+					od_table->OverDriveTable.GfxclkFmin,
+					od_table->OverDriveTable.GfxclkFmax);
+		break;
+
+	case SMU_OD_MCLK:
+		if (!smu_v13_0_0_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_UCLK_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_MCLK:\n");
+		size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMHz\n",
+					od_table->OverDriveTable.UclkFmin,
+					od_table->OverDriveTable.UclkFmax);
+		break;
+
+	case SMU_OD_VDDC_CURVE:
+		if (!smu_v13_0_0_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_GFX_VF_CURVE_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_VDDC_CURVE:\n");
+		for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++)
+			size += sysfs_emit_at(buf, size, "%d: %dmv\n",
+						i,
+						od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i]);
+		break;
+
+	case SMU_OD_RANGE:
+		if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT) &&
+		    !smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT) &&
+		    !smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE");
+
+		if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) {
+			smu_v13_0_0_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFXCLK_BIT,
+							  true,
+							  &min_value,
+							  NULL);
+			smu_v13_0_0_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFXCLK_BIT,
+							  false,
+							  NULL,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n",
+					      min_value, max_value);
+		}
+
+		if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) {
+			smu_v13_0_0_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_UCLK_BIT,
+							  true,
+							  &min_value,
+							  NULL);
+			smu_v13_0_0_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_UCLK_BIT,
+							  false,
+							  NULL,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n",
+					      min_value, max_value);
+		}
+
+		if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) {
+			smu_v13_0_0_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFX_VF_CURVE_BIT,
+							  true,
+							  &min_value,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n",
+					      min_value, max_value);
+		}
+		break;
+
 	default:
 		break;
 	}
@@ -1166,6 +1351,222 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu,
 	return size;
 }
 
+static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu,
+					 enum PP_OD_DPM_TABLE_COMMAND type,
+					 long input[],
+					 uint32_t size)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)table_context->overdrive_table;
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t offset_of_featurectrlmask;
+	int32_t minimum, maximum;
+	uint32_t feature_ctrlmask;
+	int i, ret = 0;
+
+	switch (type) {
+	case PP_OD_EDIT_SCLK_VDDC_TABLE:
+		if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) {
+			dev_warn(adev->dev, "GFXCLK_LIMITS setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				dev_info(adev->dev, "invalid number of input parameters %d\n", size);
+				return -EINVAL;
+			}
+
+			switch (input[i]) {
+			case 0:
+				smu_v13_0_0_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_GFXCLK_BIT,
+								  true,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "GfxclkFmin (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.GfxclkFmin = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT;
+				break;
+
+			case 1:
+				smu_v13_0_0_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_GFXCLK_BIT,
+								  false,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "GfxclkFmax (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.GfxclkFmax = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT;
+				break;
+
+			default:
+				dev_info(adev->dev, "Invalid SCLK_VDDC_TABLE index: %ld\n", input[i]);
+				dev_info(adev->dev, "Supported indices: [0:min,1:max]\n");
+				return -EINVAL;
+			}
+		}
+
+		if (od_table->OverDriveTable.GfxclkFmin > od_table->OverDriveTable.GfxclkFmax) {
+			dev_err(adev->dev,
+				"Invalid setting: GfxclkFmin(%u) is bigger than GfxclkFmax(%u)\n",
+				(uint32_t)od_table->OverDriveTable.GfxclkFmin,
+				(uint32_t)od_table->OverDriveTable.GfxclkFmax);
+			return -EINVAL;
+		}
+		break;
+
+	case PP_OD_EDIT_MCLK_VDDC_TABLE:
+		if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) {
+			dev_warn(adev->dev, "UCLK_LIMITS setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				dev_info(adev->dev, "invalid number of input parameters %d\n", size);
+				return -EINVAL;
+			}
+
+			switch (input[i]) {
+			case 0:
+				smu_v13_0_0_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_UCLK_BIT,
+								  true,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "UclkFmin (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.UclkFmin = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT;
+				break;
+
+			case 1:
+				smu_v13_0_0_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_UCLK_BIT,
+								  false,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "UclkFmax (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.UclkFmax = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT;
+				break;
+
+			default:
+				dev_info(adev->dev, "Invalid MCLK_VDDC_TABLE index: %ld\n", input[i]);
+				dev_info(adev->dev, "Supported indices: [0:min,1:max]\n");
+				return -EINVAL;
+			}
+		}
+
+		if (od_table->OverDriveTable.UclkFmin > od_table->OverDriveTable.UclkFmax) {
+			dev_err(adev->dev,
+				"Invalid setting: UclkFmin(%u) is bigger than UclkFmax(%u)\n",
+				(uint32_t)od_table->OverDriveTable.UclkFmin,
+				(uint32_t)od_table->OverDriveTable.UclkFmax);
+			return -EINVAL;
+		}
+		break;
+
+	case PP_OD_EDIT_VDDC_CURVE:
+		if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) {
+			dev_warn(adev->dev, "VF curve setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		if (input[0] >= PP_NUM_OD_VF_CURVE_POINTS ||
+		    input[0] < 0)
+			return -EINVAL;
+
+		smu_v13_0_0_get_od_setting_limits(smu,
+						  PP_OD_FEATURE_GFX_VF_CURVE_BIT,
+						  true,
+						  &minimum,
+						  &maximum);
+		if (input[1] < minimum ||
+		    input[1] > maximum) {
+			dev_info(adev->dev, "Voltage offset (%ld) must be within [%d, %d]!\n",
+				 input[1], minimum, maximum);
+			return -EINVAL;
+		}
+
+		od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[input[0]] = input[1];
+		od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT;
+		break;
+
+	case PP_OD_RESTORE_DEFAULT_TABLE:
+		feature_ctrlmask = od_table->OverDriveTable.FeatureCtrlMask;
+		memcpy(od_table,
+		       table_context->boot_overdrive_table,
+		       sizeof(OverDriveTableExternal_t));
+		od_table->OverDriveTable.FeatureCtrlMask = feature_ctrlmask;
+		fallthrough;
+
+	case PP_OD_COMMIT_DPM_TABLE:
+		/*
+		 * The member below instructs PMFW the settings focused in
+		 * this single operation.
+		 * `uint32_t FeatureCtrlMask;`
+		 * It does not contain actual informations about user's custom
+		 * settings. Thus we do not cache it.
+		 */
+		offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask);
+		if (memcmp(od_table + offset_of_featurectrlmask,
+			   table_context->user_overdrive_table + offset_of_featurectrlmask,
+			   sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) {
+			smu_v13_0_0_dump_od_table(smu, od_table);
+
+			ret = smu_v13_0_0_upload_overdrive_table(smu, od_table);
+			if (ret) {
+				dev_err(adev->dev, "Failed to upload overdrive table!\n");
+				return ret;
+			}
+
+			od_table->OverDriveTable.FeatureCtrlMask = 0;
+			memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask,
+			       od_table + offset_of_featurectrlmask,
+			       sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask);
+
+			if (!memcmp(table_context->user_overdrive_table,
+				    table_context->boot_overdrive_table,
+				    sizeof(OverDriveTableExternal_t)))
+				smu->user_dpm_profile.user_od = false;
+			else
+				smu->user_dpm_profile.user_od = true;
+		}
+		break;
+
+	default:
+		return -ENOSYS;
+	}
+
+	return ret;
+}
+
 static int smu_v13_0_0_force_clk_levels(struct smu_context *smu,
 					enum smu_clk_type clk_type,
 					uint32_t mask)
@@ -1394,49 +1795,6 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu,
 	return sizeof(struct gpu_metrics_v1_3);
 }
 
-static int smu_v13_0_0_get_overdrive_table(struct smu_context *smu,
-					   OverDriveTableExternal_t *od_table)
-{
-	int ret = 0;
-
-	ret = smu_cmn_update_table(smu,
-				   SMU_TABLE_OVERDRIVE,
-				   0,
-				   (void *)od_table,
-				   false);
-	if (ret)
-		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
-
-	return ret;
-}
-
-static int smu_v13_0_0_upload_overdrive_table(struct smu_context *smu,
-					      OverDriveTableExternal_t *od_table)
-{
-	int ret = 0;
-
-	ret = smu_cmn_update_table(smu,
-				   SMU_TABLE_OVERDRIVE,
-				   0,
-				   (void *)od_table,
-				   true);
-	if (ret)
-		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
-
-	return ret;
-}
-
-static void smu_v13_0_0_dump_od_table(struct smu_context *smu,
-				      OverDriveTableExternal_t *od_table)
-{
-	struct amdgpu_device *adev = smu->adev;
-
-	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
-						     od_table->OverDriveTable.GfxclkFmax);
-	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
-						   od_table->OverDriveTable.UclkFmax);
-}
-
 static int smu_v13_0_0_set_default_od_settings(struct smu_context *smu)
 {
 	OverDriveTableExternal_t *od_table =
@@ -2277,6 +2635,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range,
 	.set_default_od_settings = smu_v13_0_0_set_default_od_settings,
 	.restore_user_od_settings = smu_v13_0_0_restore_user_od_settings,
+	.od_edit_dpm_table = smu_v13_0_0_od_edit_dpm_table,
 	.init_pptable_microcode = smu_v13_0_init_pptable_microcode,
 	.populate_umd_state_clk = smu_v13_0_0_populate_umd_state_clk,
 	.set_performance_level = smu_v13_0_set_performance_level,
-- 
GitLab


From 1718e973e3d23b653cd77994073a9deda3875689 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 1 Jun 2023 17:53:45 +0530
Subject: [PATCH 1553/1791] drm/amd/pm: Fill metrics data for SMUv13.0.6

Populate metrics data table for SMU v13.0.6. Add PCIe link speed/width
information also.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 107 +++++++++++-------
 1 file changed, 66 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 75255e0baf919..2611241a5ff16 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -80,7 +80,10 @@
 /* possible frequency drift (1Mhz) */
 #define EPSILON 1
 
-#define smnPCIE_ESM_CTRL 0x111003D0
+#define smnPCIE_ESM_CTRL 0x193D0
+#define smnPCIE_LC_LINK_WIDTH_CNTL 0x1ab40288
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK 0x00000070L
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT 0x4
 
 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
@@ -197,6 +200,7 @@ struct PPTable_t {
 };
 
 #define SMUQ10_TO_UINT(x) ((x) >> 10)
+#define SMUQ16_TO_UINT(x) ((x) >> 16)
 
 struct smu_v13_0_6_dpm_map {
 	enum smu_clk_type clk_type;
@@ -1935,6 +1939,15 @@ static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu)
 						   smu_v13_0_6_throttler_map));
 }
 
+static int
+smu_v13_0_6_get_current_pcie_link_width_level(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	return REG_GET_FIELD(RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL),
+			     PCIE_LC_LINK_WIDTH_CNTL, LC_LINK_WIDTH_RD);
+}
+
 static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
@@ -1953,8 +1966,12 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 	struct smu_table_context *smu_table = &smu->smu_table;
 	struct gpu_metrics_v1_3 *gpu_metrics =
 		(struct gpu_metrics_v1_3 *)smu_table->gpu_metrics_table;
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0, inst0, xcc0;
 	MetricsTable_t *metrics;
-	int i, ret = 0;
+
+	inst0 = adev->sdma.instance[0].aid_id;
+	xcc0 = GET_INST(GC, 0);
 
 	metrics = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL);
 	ret = smu_v13_0_6_get_metrics_table(smu, metrics, true);
@@ -1963,51 +1980,59 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 
 	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3);
 
-	/* TODO: Decide on how to fill in zero value fields */
-	gpu_metrics->temperature_edge = 0;
-	gpu_metrics->temperature_hotspot = 0;
-	gpu_metrics->temperature_mem = 0;
-	gpu_metrics->temperature_vrgfx = 0;
-	gpu_metrics->temperature_vrsoc = 0;
-	gpu_metrics->temperature_vrmem = 0;
-
-	gpu_metrics->average_gfx_activity = 0;
-	gpu_metrics->average_umc_activity = 0;
-	gpu_metrics->average_mm_activity = 0;
-
-	gpu_metrics->average_socket_power = 0;
-	gpu_metrics->energy_accumulator = 0;
-
-	gpu_metrics->average_gfxclk_frequency = 0;
-	gpu_metrics->average_socclk_frequency = 0;
-	gpu_metrics->average_uclk_frequency = 0;
-	gpu_metrics->average_vclk0_frequency = 0;
-	gpu_metrics->average_dclk0_frequency = 0;
-
-	gpu_metrics->current_gfxclk = 0;
-	gpu_metrics->current_socclk = 0;
-	gpu_metrics->current_uclk = 0;
-	gpu_metrics->current_vclk0 = 0;
-	gpu_metrics->current_dclk0 = 0;
-
+	gpu_metrics->temperature_hotspot =
+		SMUQ10_TO_UINT(metrics->MaxSocketTemperature);
+	/* Individual HBM stack temperature is not reported */
+	gpu_metrics->temperature_mem =
+		SMUQ10_TO_UINT(metrics->MaxHbmTemperature);
+	/* Reports max temperature of all voltage rails */
+	gpu_metrics->temperature_vrsoc =
+		SMUQ10_TO_UINT(metrics->MaxVrTemperature);
+
+	gpu_metrics->average_gfx_activity =
+		SMUQ10_TO_UINT(metrics->SocketGfxBusy);
+	gpu_metrics->average_umc_activity =
+		SMUQ10_TO_UINT(metrics->DramBandwidthUtilization);
+
+	gpu_metrics->average_socket_power =
+		SMUQ10_TO_UINT(metrics->SocketPower);
+	gpu_metrics->energy_accumulator =
+		SMUQ16_TO_UINT(metrics->SocketEnergyAcc);
+
+	gpu_metrics->current_gfxclk =
+		SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc0]);
+	gpu_metrics->current_socclk =
+		SMUQ10_TO_UINT(metrics->SocclkFrequency[inst0]);
+	gpu_metrics->current_uclk = SMUQ10_TO_UINT(metrics->UclkFrequency);
+	gpu_metrics->current_vclk0 =
+		SMUQ10_TO_UINT(metrics->VclkFrequency[inst0]);
+	gpu_metrics->current_dclk0 =
+		SMUQ10_TO_UINT(metrics->DclkFrequency[inst0]);
+
+	gpu_metrics->average_gfxclk_frequency = gpu_metrics->current_gfxclk;
+	gpu_metrics->average_socclk_frequency = gpu_metrics->current_socclk;
+	gpu_metrics->average_uclk_frequency = gpu_metrics->current_uclk;
+	gpu_metrics->average_vclk0_frequency = gpu_metrics->current_vclk0;
+	gpu_metrics->average_dclk0_frequency = gpu_metrics->current_dclk0;
+
+	/* Throttle status is not reported through metrics now */
 	gpu_metrics->throttle_status = 0;
-	gpu_metrics->indep_throttle_status = smu_cmn_get_indep_throttler_status(
-		gpu_metrics->throttle_status, smu_v13_0_6_throttler_map);
-
-	gpu_metrics->current_fan_speed = 0;
 
-	gpu_metrics->pcie_link_width = 0;
-	gpu_metrics->pcie_link_speed = smu_v13_0_6_get_current_pcie_link_speed(smu);
+	if (!(adev->flags & AMD_IS_APU)) {
+		gpu_metrics->pcie_link_width =
+			smu_v13_0_6_get_current_pcie_link_width_level(smu);
+		gpu_metrics->pcie_link_speed =
+			smu_v13_0_6_get_current_pcie_link_speed(smu);
+	}
 
 	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
 
-	gpu_metrics->gfx_activity_acc = 0;
-	gpu_metrics->mem_activity_acc = 0;
-
-	for (i = 0; i < NUM_HBM_INSTANCES; i++)
-		gpu_metrics->temperature_hbm[i] = 0;
+	gpu_metrics->gfx_activity_acc =
+		SMUQ10_TO_UINT(metrics->SocketGfxBusyAcc);
+	gpu_metrics->mem_activity_acc =
+		SMUQ10_TO_UINT(metrics->DramBandwidthUtilizationAcc);
 
-	gpu_metrics->firmware_timestamp = 0;
+	gpu_metrics->firmware_timestamp = metrics->Timestamp;
 
 	*table = (void *)gpu_metrics;
 	kfree(metrics);
-- 
GitLab


From 2a9aa52e4617c777fb0c885f0c02bf5ac65a786c Mon Sep 17 00:00:00 2001
From: Evan Quan <evan.quan@amd.com>
Date: Tue, 9 May 2023 10:21:52 +0800
Subject: [PATCH 1554/1791] drm/amd/pm: fulfill the OD support for SMU13.0.7

Fulfill the interfaces for OD settings retrieving and setting.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c  | 445 ++++++++++++++++--
 1 file changed, 402 insertions(+), 43 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 20a8be3bb49a3..0bd086360efae 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -1022,16 +1022,118 @@ static int smu_v13_0_7_get_current_clk_freq_by_table(struct smu_context *smu,
 						value);
 }
 
+static bool smu_v13_0_7_is_od_feature_supported(struct smu_context *smu,
+						int od_feature_bit)
+{
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&pptable->SkuTable.OverDriveLimitsBasicMax;
+
+	return overdrive_upperlimits->FeatureCtrlMask & (1U << od_feature_bit);
+}
+
+static void smu_v13_0_7_get_od_setting_limits(struct smu_context *smu,
+					      int od_feature_bit,
+					      bool lower_boundary,
+					      int32_t *min,
+					      int32_t *max)
+{
+	PPTable_t *pptable = smu->smu_table.driver_pptable;
+	const OverDriveLimits_t * const overdrive_upperlimits =
+				&pptable->SkuTable.OverDriveLimitsBasicMax;
+	const OverDriveLimits_t * const overdrive_lowerlimits =
+				&pptable->SkuTable.OverDriveLimitsMin;
+	int32_t od_min_setting, od_max_setting;
+
+	switch (od_feature_bit) {
+	case PP_OD_FEATURE_GFXCLK_BIT:
+		if (lower_boundary) {
+			od_min_setting = overdrive_lowerlimits->GfxclkFmin;
+			od_max_setting = overdrive_upperlimits->GfxclkFmin;
+		} else {
+			od_min_setting = overdrive_lowerlimits->GfxclkFmax;
+			od_max_setting = overdrive_upperlimits->GfxclkFmax;
+		}
+		break;
+	case PP_OD_FEATURE_UCLK_BIT:
+		if (lower_boundary) {
+			od_min_setting = overdrive_lowerlimits->UclkFmin;
+			od_max_setting = overdrive_upperlimits->UclkFmin;
+		} else {
+			od_min_setting = overdrive_lowerlimits->UclkFmax;
+			od_max_setting = overdrive_upperlimits->UclkFmax;
+		}
+		break;
+	case PP_OD_FEATURE_GFX_VF_CURVE_BIT:
+		od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary;
+		od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary;
+		break;
+	default:
+		break;
+	}
+
+	if (min)
+		*min = od_min_setting;
+	if (max)
+		*max = od_max_setting;
+}
+
+static void smu_v13_0_7_dump_od_table(struct smu_context *smu,
+				      OverDriveTableExternal_t *od_table)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
+						     od_table->OverDriveTable.GfxclkFmax);
+	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
+						   od_table->OverDriveTable.UclkFmax);
+}
+
+static int smu_v13_0_7_get_overdrive_table(struct smu_context *smu,
+					   OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   false);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
+
+	return ret;
+}
+
+static int smu_v13_0_7_upload_overdrive_table(struct smu_context *smu,
+					      OverDriveTableExternal_t *od_table)
+{
+	int ret = 0;
+
+	ret = smu_cmn_update_table(smu,
+				   SMU_TABLE_OVERDRIVE,
+				   0,
+				   (void *)od_table,
+				   true);
+	if (ret)
+		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
+
+	return ret;
+}
+
 static int smu_v13_0_7_print_clk_levels(struct smu_context *smu,
 					enum smu_clk_type clk_type,
 					char *buf)
 {
 	struct smu_dpm_context *smu_dpm = &smu->smu_dpm;
 	struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context;
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)smu->smu_table.overdrive_table;
 	struct smu_13_0_dpm_table *single_dpm_table;
 	struct smu_13_0_pcie_table *pcie_table;
 	uint32_t gen_speed, lane_width;
 	int i, curr_freq, size = 0;
+	int32_t min_value, max_value;
 	int ret = 0;
 
 	smu_cmn_get_sysfs_buf(&buf, &size);
@@ -1148,6 +1250,89 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu,
 					"*" : "");
 		break;
 
+	case SMU_OD_SCLK:
+		if (!smu_v13_0_7_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_GFXCLK_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_SCLK:\n");
+		size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMhz\n",
+					od_table->OverDriveTable.GfxclkFmin,
+					od_table->OverDriveTable.GfxclkFmax);
+		break;
+
+	case SMU_OD_MCLK:
+		if (!smu_v13_0_7_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_UCLK_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_MCLK:\n");
+		size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMHz\n",
+					od_table->OverDriveTable.UclkFmin,
+					od_table->OverDriveTable.UclkFmax);
+		break;
+
+	case SMU_OD_VDDC_CURVE:
+		if (!smu_v13_0_7_is_od_feature_supported(smu,
+							 PP_OD_FEATURE_GFX_VF_CURVE_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "OD_VDDC_CURVE:\n");
+		for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++)
+			size += sysfs_emit_at(buf, size, "%d: %dmv\n",
+						i,
+						od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i]);
+		break;
+
+	case SMU_OD_RANGE:
+		if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT) &&
+		    !smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT) &&
+		    !smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT))
+			break;
+
+		size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE");
+
+		if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) {
+			smu_v13_0_7_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFXCLK_BIT,
+							  true,
+							  &min_value,
+							  NULL);
+			smu_v13_0_7_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFXCLK_BIT,
+							  false,
+							  NULL,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n",
+					      min_value, max_value);
+		}
+
+		if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) {
+			smu_v13_0_7_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_UCLK_BIT,
+							  true,
+							  &min_value,
+							  NULL);
+			smu_v13_0_7_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_UCLK_BIT,
+							  false,
+							  NULL,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n",
+					      min_value, max_value);
+		}
+
+		if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) {
+			smu_v13_0_7_get_od_setting_limits(smu,
+							  PP_OD_FEATURE_GFX_VF_CURVE_BIT,
+							  true,
+							  &min_value,
+							  &max_value);
+			size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n",
+					      min_value, max_value);
+		}
+		break;
+
 	default:
 		break;
 	}
@@ -1155,6 +1340,222 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu,
 	return size;
 }
 
+static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu,
+					 enum PP_OD_DPM_TABLE_COMMAND type,
+					 long input[],
+					 uint32_t size)
+{
+	struct smu_table_context *table_context = &smu->smu_table;
+	OverDriveTableExternal_t *od_table =
+		(OverDriveTableExternal_t *)table_context->overdrive_table;
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t offset_of_featurectrlmask;
+	int32_t minimum, maximum;
+	uint32_t feature_ctrlmask;
+	int i, ret = 0;
+
+	switch (type) {
+	case PP_OD_EDIT_SCLK_VDDC_TABLE:
+		if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) {
+			dev_warn(adev->dev, "GFXCLK_LIMITS setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				dev_info(adev->dev, "invalid number of input parameters %d\n", size);
+				return -EINVAL;
+			}
+
+			switch (input[i]) {
+			case 0:
+				smu_v13_0_7_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_GFXCLK_BIT,
+								  true,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "GfxclkFmin (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.GfxclkFmin = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT;
+				break;
+
+			case 1:
+				smu_v13_0_7_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_GFXCLK_BIT,
+								  false,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "GfxclkFmax (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.GfxclkFmax = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT;
+				break;
+
+			default:
+				dev_info(adev->dev, "Invalid SCLK_VDDC_TABLE index: %ld\n", input[i]);
+				dev_info(adev->dev, "Supported indices: [0:min,1:max]\n");
+				return -EINVAL;
+			}
+		}
+
+		if (od_table->OverDriveTable.GfxclkFmin > od_table->OverDriveTable.GfxclkFmax) {
+			dev_err(adev->dev,
+				"Invalid setting: GfxclkFmin(%u) is bigger than GfxclkFmax(%u)\n",
+				(uint32_t)od_table->OverDriveTable.GfxclkFmin,
+				(uint32_t)od_table->OverDriveTable.GfxclkFmax);
+			return -EINVAL;
+		}
+		break;
+
+	case PP_OD_EDIT_MCLK_VDDC_TABLE:
+		if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) {
+			dev_warn(adev->dev, "UCLK_LIMITS setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		for (i = 0; i < size; i += 2) {
+			if (i + 2 > size) {
+				dev_info(adev->dev, "invalid number of input parameters %d\n", size);
+				return -EINVAL;
+			}
+
+			switch (input[i]) {
+			case 0:
+				smu_v13_0_7_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_UCLK_BIT,
+								  true,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "UclkFmin (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.UclkFmin = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT;
+				break;
+
+			case 1:
+				smu_v13_0_7_get_od_setting_limits(smu,
+								  PP_OD_FEATURE_UCLK_BIT,
+								  false,
+								  &minimum,
+								  &maximum);
+				if (input[i + 1] < minimum ||
+				    input[i + 1] > maximum) {
+					dev_info(adev->dev, "UclkFmax (%ld) must be within [%u, %u]!\n",
+						input[i + 1], minimum, maximum);
+					return -EINVAL;
+				}
+
+				od_table->OverDriveTable.UclkFmax = input[i + 1];
+				od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT;
+				break;
+
+			default:
+				dev_info(adev->dev, "Invalid MCLK_VDDC_TABLE index: %ld\n", input[i]);
+				dev_info(adev->dev, "Supported indices: [0:min,1:max]\n");
+				return -EINVAL;
+			}
+		}
+
+		if (od_table->OverDriveTable.UclkFmin > od_table->OverDriveTable.UclkFmax) {
+			dev_err(adev->dev,
+				"Invalid setting: UclkFmin(%u) is bigger than UclkFmax(%u)\n",
+				(uint32_t)od_table->OverDriveTable.UclkFmin,
+				(uint32_t)od_table->OverDriveTable.UclkFmax);
+			return -EINVAL;
+		}
+		break;
+
+	case PP_OD_EDIT_VDDC_CURVE:
+		if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) {
+			dev_warn(adev->dev, "VF curve setting not supported!\n");
+			return -ENOTSUPP;
+		}
+
+		if (input[0] >= PP_NUM_OD_VF_CURVE_POINTS ||
+		    input[0] < 0)
+			return -EINVAL;
+
+		smu_v13_0_7_get_od_setting_limits(smu,
+						  PP_OD_FEATURE_GFX_VF_CURVE_BIT,
+						  true,
+						  &minimum,
+						  &maximum);
+		if (input[1] < minimum ||
+		    input[1] > maximum) {
+			dev_info(adev->dev, "Voltage offset (%ld) must be within [%d, %d]!\n",
+				 input[1], minimum, maximum);
+			return -EINVAL;
+		}
+
+		od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[input[0]] = input[1];
+		od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT;
+		break;
+
+	case PP_OD_RESTORE_DEFAULT_TABLE:
+		feature_ctrlmask = od_table->OverDriveTable.FeatureCtrlMask;
+		memcpy(od_table,
+		       table_context->boot_overdrive_table,
+		       sizeof(OverDriveTableExternal_t));
+		od_table->OverDriveTable.FeatureCtrlMask = feature_ctrlmask;
+		fallthrough;
+
+	case PP_OD_COMMIT_DPM_TABLE:
+		/*
+		 * The member below instructs PMFW the settings focused in
+		 * this single operation.
+		 * `uint32_t FeatureCtrlMask;`
+		 * It does not contain actual informations about user's custom
+		 * settings. Thus we do not cache it.
+		 */
+		offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask);
+		if (memcmp(od_table + offset_of_featurectrlmask,
+			   table_context->user_overdrive_table + offset_of_featurectrlmask,
+			   sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) {
+			smu_v13_0_7_dump_od_table(smu, od_table);
+
+			ret = smu_v13_0_7_upload_overdrive_table(smu, od_table);
+			if (ret) {
+				dev_err(adev->dev, "Failed to upload overdrive table!\n");
+				return ret;
+			}
+
+			od_table->OverDriveTable.FeatureCtrlMask = 0;
+			memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask,
+			       od_table + offset_of_featurectrlmask,
+			       sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask);
+
+			if (!memcmp(table_context->user_overdrive_table,
+				    table_context->boot_overdrive_table,
+				    sizeof(OverDriveTableExternal_t)))
+				smu->user_dpm_profile.user_od = false;
+			else
+				smu->user_dpm_profile.user_od = true;
+		}
+		break;
+
+	default:
+		return -ENOSYS;
+	}
+
+	return ret;
+}
+
 static int smu_v13_0_7_force_clk_levels(struct smu_context *smu,
 					enum smu_clk_type clk_type,
 					uint32_t mask)
@@ -1380,49 +1781,6 @@ static ssize_t smu_v13_0_7_get_gpu_metrics(struct smu_context *smu,
 	return sizeof(struct gpu_metrics_v1_3);
 }
 
-static int smu_v13_0_7_get_overdrive_table(struct smu_context *smu,
-					   OverDriveTableExternal_t *od_table)
-{
-	int ret = 0;
-
-	ret = smu_cmn_update_table(smu,
-				   SMU_TABLE_OVERDRIVE,
-				   0,
-				   (void *)od_table,
-				   false);
-	if (ret)
-		dev_err(smu->adev->dev, "Failed to get overdrive table!\n");
-
-	return ret;
-}
-
-static int smu_v13_0_7_upload_overdrive_table(struct smu_context *smu,
-					      OverDriveTableExternal_t *od_table)
-{
-	int ret = 0;
-
-	ret = smu_cmn_update_table(smu,
-				   SMU_TABLE_OVERDRIVE,
-				   0,
-				   (void *)od_table,
-				   true);
-	if (ret)
-		dev_err(smu->adev->dev, "Failed to upload overdrive table!\n");
-
-	return ret;
-}
-
-static void smu_v13_0_7_dump_od_table(struct smu_context *smu,
-				      OverDriveTableExternal_t *od_table)
-{
-	struct amdgpu_device *adev = smu->adev;
-
-	dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin,
-						     od_table->OverDriveTable.GfxclkFmax);
-	dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin,
-						   od_table->OverDriveTable.UclkFmax);
-}
-
 static int smu_v13_0_7_set_default_od_settings(struct smu_context *smu)
 {
 	OverDriveTableExternal_t *od_table =
@@ -1886,6 +2244,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range,
 	.set_default_od_settings = smu_v13_0_7_set_default_od_settings,
 	.restore_user_od_settings = smu_v13_0_7_restore_user_od_settings,
+	.od_edit_dpm_table = smu_v13_0_7_od_edit_dpm_table,
 	.set_performance_level = smu_v13_0_set_performance_level,
 	.gfx_off_control = smu_v13_0_gfx_off_control,
 	.get_fan_speed_pwm = smu_v13_0_7_get_fan_speed_pwm,
-- 
GitLab


From a4d4db727320e0f80df605ccb877743359448a36 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 2 Jun 2023 15:19:22 +0530
Subject: [PATCH 1555/1791] drm/amd/display: Fix dc/dcn20/dcn20_optc.c kdoc

Fix all kdoc warnings in dc/dcn20/dcn20_optc.c:

display/dc/dcn20/dcn20_optc.c:41: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Enable CRTC
display/dc/dcn20/dcn20_optc.c:76: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 *For the below, I'm not sure how your GSL parameters are stored in your
  env,
display/dc/dcn20/dcn20_optc.c:85: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * There are (MAX_OPTC+1)/2 gsl groups available for use.

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn20/dcn20_optc.c   | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
index e0edc163d7677..d6f095b4555dc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
@@ -38,8 +38,12 @@
 	optc1->tg_shift->field_name, optc1->tg_mask->field_name
 
 /**
- * Enable CRTC
- * Enable CRTC - call ASIC Control Object to enable Timing generator.
+ * optc2_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator.
+ *
+ * @optc: timing_generator instance.
+ *
+ * Return: If CRTC is enabled, return true.
+ *
  */
 bool optc2_enable_crtc(struct timing_generator *optc)
 {
@@ -73,15 +77,18 @@ bool optc2_enable_crtc(struct timing_generator *optc)
 }
 
 /**
- *For the below, I'm not sure how your GSL parameters are stored in your env,
- * so I will assume a gsl_params struct for now
+ * optc2_set_gsl() - Assign OTG to GSL groups,
+ *                   set one of the OTGs to be master & rest are slaves
+ *
+ * @optc: timing_generator instance.
+ * @params: pointer to gsl_params
  */
 void optc2_set_gsl(struct timing_generator *optc,
 		   const struct gsl_params *params)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
 
-/**
+/*
  * There are (MAX_OPTC+1)/2 gsl groups available for use.
  * In each group (assign an OTG to a group by setting OTG_GSLX_EN = 1,
  * set one of the OTGs to be the master (OTG_GSL_MASTER_EN = 1) and the rest are slaves.
-- 
GitLab


From 2b607025797543433e7733c276ec34381edffd71 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Fri, 2 Jun 2023 14:17:53 +0800
Subject: [PATCH 1556/1791] drm/amd/display: clean up some inconsistent
 indenting

No functional modification involved.

drivers/gpu/drm/amd/amdgpu/../display/dc/link/link_dpms.c:2377 link_set_dpms_on() warn: inconsistent indenting.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5376
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 2963267fe74ab..f7f1a1586f3b2 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2374,8 +2374,8 @@ void link_set_dpms_on(
 		}
 	}
 
-		if (dc_is_virtual_signal(pipe_ctx->stream->signal))
-			return;
+	if (dc_is_virtual_signal(pipe_ctx->stream->signal))
+		return;
 
 	link_enc = link_enc_cfg_get_link_enc(link);
 	ASSERT(link_enc);
-- 
GitLab


From 24e461e84f1c6d58fa1032f06d97e277dd0b4adf Mon Sep 17 00:00:00 2001
From: Samson Tam <samson.tam@amd.com>
Date: Tue, 9 May 2023 16:40:19 -0400
Subject: [PATCH 1557/1791] drm/amd/display: add ODM case when looking for
 first split pipe

[Why]
When going from ODM 2:1 single display case to max displays, second
odm pipe needs to be repurposed for one of the new single displays.
However, acquire_first_split_pipe() only handles MPC case and not
ODM case

[How]
Add ODM conditions in acquire_first_split_pipe()
Add commit_minimal_transition_state() in commit_streams() to handle
odm 2:1 exit first, and then process new streams
Handle ODM condition in commit_minimal_transition_state()

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Samson Tam <samson.tam@amd.com>
Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 36 ++++++++++++++++++-
 .../gpu/drm/amd/display/dc/core/dc_resource.c | 20 +++++++++++
 2 files changed, 55 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index af929ee11af52..0e9403f21dae9 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2008,6 +2008,9 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
 	return result;
 }
 
+static bool commit_minimal_transition_state(struct dc *dc,
+		struct dc_state *transition_base_context);
+
 /**
  * dc_commit_streams - Commit current stream state
  *
@@ -2029,6 +2032,8 @@ enum dc_status dc_commit_streams(struct dc *dc,
 	struct dc_state *context;
 	enum dc_status res = DC_OK;
 	struct dc_validation_set set[MAX_STREAMS] = {0};
+	struct pipe_ctx *pipe;
+	bool handle_exit_odm2to1 = false;
 
 	if (dc->ctx->dce_environment == DCE_ENV_VIRTUAL_HW)
 		return res;
@@ -2053,6 +2058,22 @@ enum dc_status dc_commit_streams(struct dc *dc,
 		}
 	}
 
+	/* Check for case where we are going from odm 2:1 to max
+	 *  pipe scenario.  For these cases, we will call
+	 *  commit_minimal_transition_state() to exit out of odm 2:1
+	 *  first before processing new streams
+	 */
+	if (stream_count == dc->res_pool->pipe_count) {
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+			if (pipe->next_odm_pipe)
+				handle_exit_odm2to1 = true;
+		}
+	}
+
+	if (handle_exit_odm2to1)
+		res = commit_minimal_transition_state(dc, dc->current_state);
+
 	context = dc_create_state(dc);
 	if (!context)
 		goto context_alloc_fail;
@@ -3912,6 +3933,7 @@ static bool commit_minimal_transition_state(struct dc *dc,
 	unsigned int i, j;
 	unsigned int pipe_in_use = 0;
 	bool subvp_in_use = false;
+	bool odm_in_use = false;
 
 	if (!transition_context)
 		return false;
@@ -3940,6 +3962,18 @@ static bool commit_minimal_transition_state(struct dc *dc,
 		}
 	}
 
+	/* If ODM is enabled and we are adding or removing planes from any ODM
+	 * pipe, we must use the minimal transition.
+	 */
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe->stream && pipe->next_odm_pipe) {
+			odm_in_use = true;
+			break;
+		}
+	}
+
 	/* When the OS add a new surface if we have been used all of pipes with odm combine
 	 * and mpc split feature, it need use commit_minimal_transition_state to transition safely.
 	 * After OS exit MPO, it will back to use odm and mpc split with all of pipes, we need
@@ -3948,7 +3982,7 @@ static bool commit_minimal_transition_state(struct dc *dc,
 	 * Reduce the scenarios to use dc_commit_state_no_check in the stage of flip. Especially
 	 * enter/exit MPO when DCN still have enough resources.
 	 */
-	if (pipe_in_use != dc->res_pool->pipe_count && !subvp_in_use) {
+	if (pipe_in_use != dc->res_pool->pipe_count && !subvp_in_use && !odm_in_use) {
 		dc_release_state(transition_context);
 		return true;
 	}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 7e1e5532f88f3..c72540d37aefd 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -1445,6 +1445,26 @@ static int acquire_first_split_pipe(
 			split_pipe->plane_res.mpcc_inst = pool->dpps[i]->inst;
 			split_pipe->pipe_idx = i;
 
+			split_pipe->stream = stream;
+			return i;
+		} else if (split_pipe->prev_odm_pipe &&
+				split_pipe->prev_odm_pipe->plane_state == split_pipe->plane_state) {
+			split_pipe->prev_odm_pipe->next_odm_pipe = split_pipe->next_odm_pipe;
+			if (split_pipe->next_odm_pipe)
+				split_pipe->next_odm_pipe->prev_odm_pipe = split_pipe->prev_odm_pipe;
+
+			if (split_pipe->prev_odm_pipe->plane_state)
+				resource_build_scaling_params(split_pipe->prev_odm_pipe);
+
+			memset(split_pipe, 0, sizeof(*split_pipe));
+			split_pipe->stream_res.tg = pool->timing_generators[i];
+			split_pipe->plane_res.hubp = pool->hubps[i];
+			split_pipe->plane_res.ipp = pool->ipps[i];
+			split_pipe->plane_res.dpp = pool->dpps[i];
+			split_pipe->stream_res.opp = pool->opps[i];
+			split_pipe->plane_res.mpcc_inst = pool->dpps[i]->inst;
+			split_pipe->pipe_idx = i;
+
 			split_pipe->stream = stream;
 			return i;
 		}
-- 
GitLab


From 75c2b7ed080d7421157c03064be82275364136e7 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Tue, 18 Apr 2023 10:11:56 -0400
Subject: [PATCH 1558/1791] drm/amd/display: fix seamless odm transitions

Add missing programming and function pointers

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 11 +++++++++++
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c  |  2 +-
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h  |  1 +
 3 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index eaf9e9ccad2a7..20f668d28364c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -1741,6 +1741,17 @@ static void dcn20_program_pipe(
 
 		if (hws->funcs.setup_vupdate_interrupt)
 			hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx);
+
+		if (hws->funcs.calculate_dccg_k1_k2_values && dc->res_pool->dccg->funcs->set_pixel_rate_div) {
+			unsigned int k1_div, k2_div;
+
+			hws->funcs.calculate_dccg_k1_k2_values(pipe_ctx, &k1_div, &k2_div);
+
+			dc->res_pool->dccg->funcs->set_pixel_rate_div(
+				dc->res_pool->dccg,
+				pipe_ctx->stream_res.tg->inst,
+				k1_div, k2_div);
+		}
 	}
 
 	if (pipe_ctx->update_flags.bits.odm)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
index 6ef56fb32131d..2cffedea2df56 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
@@ -98,7 +98,7 @@ static void optc32_set_odm_combine(struct timing_generator *optc, int *opp_id, i
 	optc1->opp_count = opp_cnt;
 }
 
-static void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode)
+void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h
index b92ba8c756940..abf0121a10060 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h
@@ -179,5 +179,6 @@
 	SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh)
 
 void dcn32_timing_generator_init(struct optc *optc1);
+void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode);
 
 #endif /* __DC_OPTC_DCN32_H__ */
-- 
GitLab


From 9bd443cb74bd47d820c3cc31ee0ed3008d004d73 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Fri, 2 Jun 2023 13:52:04 -0400
Subject: [PATCH 1559/1791] drm/amdgpu: fix debug wait on idle for gfx9.4.1

Wait calls for amd_ip_block_type not amd_hw_ip_block_type.

Reported-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 6df215aba4c4e..625db444df1cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -309,7 +309,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
 	if (!suspend || r)
 		return r;
 
-	return amdgpu_device_ip_wait_for_idle(adev, GC_HWIP);
+	return amdgpu_device_ip_wait_for_idle(adev, AMD_IP_BLOCK_TYPE_GFX);
 }
 
 static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_waitcnt)
-- 
GitLab


From 9f0bcf49e9895cb005d78b33a5eebfa11711b425 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 1 Jun 2023 15:44:12 -0700
Subject: [PATCH 1560/1791] amdgpu: validate offset_in_bo of drm_amdgpu_gem_va
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is motivated by OOB access in amdgpu_vm_update_range when
offset_in_bo+map_size overflows.

v2: keep the validations in amdgpu_vm_bo_map
v3: add the validations to amdgpu_vm_bo_map/amdgpu_vm_bo_replace_map
    rather than to amdgpu_gem_va_ioctl

Fixes: 9f7eb5367d00 ("drm/amdgpu: actually use the VM map parameters")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 22f9a65ca0fc7..76d57bc7ac620 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1434,14 +1434,14 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	uint64_t eaddr;
 
 	/* validate the parameters */
-	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
-	    size == 0 || size & ~PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || size & ~PAGE_MASK)
+		return -EINVAL;
+	if (saddr + size <= saddr || offset + size <= offset)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
 	eaddr = saddr + size - 1;
-	if (saddr >= eaddr ||
-	    (bo && offset + size > amdgpu_bo_size(bo)) ||
+	if ((bo && offset + size > amdgpu_bo_size(bo)) ||
 	    (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT))
 		return -EINVAL;
 
@@ -1500,14 +1500,14 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 	int r;
 
 	/* validate the parameters */
-	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK ||
-	    size == 0 || size & ~PAGE_MASK)
+	if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || size & ~PAGE_MASK)
+		return -EINVAL;
+	if (saddr + size <= saddr || offset + size <= offset)
 		return -EINVAL;
 
 	/* make sure object fit at this offset */
 	eaddr = saddr + size - 1;
-	if (saddr >= eaddr ||
-	    (bo && offset + size > amdgpu_bo_size(bo)) ||
+	if ((bo && offset + size > amdgpu_bo_size(bo)) ||
 	    (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT))
 		return -EINVAL;
 
-- 
GitLab


From 5be7d4e3cf9ef9853934daa03cf573723bae1650 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Sun, 28 May 2023 14:26:37 -0600
Subject: [PATCH 1561/1791] drm/amdgpu/discovery: Replace fake flex-arrays with
 flexible-array members
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zero-length and one-element arrays are deprecated, and we are moving
towards adopting C99 flexible-array members, instead.

Use the DECLARE_FLEX_ARRAY() helper macro to transform zero-length
arrays in a union into flexible-array members. And replace a one-element
array with a C99 flexible-array member.

Address the following warnings found with GCC-13 and
-fstrict-flex-arrays=3 enabled:
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1009:89: warning: array subscript kk is outside array bounds of ‘uint32_t[0]’ {aka ‘unsigned int[]’} [-Warray-bounds=]
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1007:94: warning: array subscript kk is outside array bounds of ‘uint64_t[0]’ {aka ‘long long unsigned int[]’} [-Warray-bounds=]
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1310:94: warning: array subscript k is outside array bounds of ‘uint64_t[0]’ {aka ‘long long unsigned int[]’} [-Warray-bounds=]
drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c:1309:57: warning: array subscript k is outside array bounds of ‘uint32_t[0]’ {aka ‘unsigned int[]’} [-Warray-bounds=]

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

This results in no differences in binary output.

Link: https://github.com/KSPP/linux/issues/21
Link: https://github.com/KSPP/linux/issues/193
Link: https://github.com/KSPP/linux/issues/300
Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1]
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/include/discovery.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/include/discovery.h b/drivers/gpu/drm/amd/include/discovery.h
index 9181e57887db7..f43e29722ef71 100644
--- a/drivers/gpu/drm/amd/include/discovery.h
+++ b/drivers/gpu/drm/amd/include/discovery.h
@@ -122,7 +122,7 @@ typedef struct ip_v3
 	uint8_t sub_revision : 4;               /* HCID Sub-Revision */
 	uint8_t variant : 4;                    /* HW variant */
 #endif
-	uint32_t base_address[1];               /* Base Address list. Corresponds to the num_base_address field*/
+	uint32_t base_address[];		/* Base Address list. Corresponds to the num_base_address field*/
 } ip_v3;
 
 typedef struct ip_v4 {
@@ -140,8 +140,8 @@ typedef struct ip_v4 {
 	uint8_t sub_revision : 4;               /* HCID Sub-Revision */
 #endif
 	union {
-		uint32_t base_address[0];               /* 32-bit Base Address list. Corresponds to the num_base_address field*/
-		uint64_t base_address_64[0];            /* 64-bit Base Address list. Corresponds to the num_base_address field*/
+		DECLARE_FLEX_ARRAY(uint32_t, base_address);	/* 32-bit Base Address list. Corresponds to the num_base_address field*/
+		DECLARE_FLEX_ARRAY(uint64_t, base_address_64);	/* 64-bit Base Address list. Corresponds to the num_base_address field*/
 	} __packed;
 } ip_v4;
 
-- 
GitLab


From 2890662822def3dcc1b2e690d6fcea694c666083 Mon Sep 17 00:00:00 2001
From: "GONG, Ruiqi" <gongruiqi@huaweicloud.com>
Date: Fri, 2 Jun 2023 18:12:33 +0800
Subject: [PATCH 1562/1791] drm/amd/display: fix compilation error due to
 shifting negative value
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently compiling linux-next with allmodconfig triggers the following
error:

./drivers/gpu/drm/amd/amdgpu/../display/include/fixed31_32.h: In function ‘dc_fixpt_truncate’:
./drivers/gpu/drm/amd/amdgpu/../display/include/fixed31_32.h:528:22: error: left shift of negative value [-Werror=shift-negative-value]
  528 |  arg.value &= (~0LL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits);
      |                      ^~

Use `unsigned long long` instead.

Signed-off-by: GONG, Ruiqi <gongruiqi@huaweicloud.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/include/fixed31_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/include/fixed31_32.h b/drivers/gpu/drm/amd/display/include/fixed31_32.h
index ece97ae0e826c..d4cf7ead1d877 100644
--- a/drivers/gpu/drm/amd/display/include/fixed31_32.h
+++ b/drivers/gpu/drm/amd/display/include/fixed31_32.h
@@ -525,7 +525,7 @@ static inline struct fixed31_32 dc_fixpt_truncate(struct fixed31_32 arg, unsigne
 
 	if (negative)
 		arg.value = -arg.value;
-	arg.value &= (~0LL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits);
+	arg.value &= (~0ULL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits);
 	if (negative)
 		arg.value = -arg.value;
 	return arg;
-- 
GitLab


From 6b37fee590ec842f6e172c4f9c7dc4baadbdfda2 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 2 Jun 2023 14:02:45 +0530
Subject: [PATCH 1563/1791] drm/amd/display: Address kdoc warnings in
 dcn30_fpu.c

Fixes the following gcc with W=1:

display/dc/dml/dcn30/dcn30_fpu.c:677: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Finds dummy_latency_index when MCLK switching using firmware based
display/dc/dml/dcn30/dcn30_fpu.c:688: warning: Function parameter or member 'dc' not described in 'dcn30_find_dummy_latency_index_for_fw_based_mclk_switch'
display/dc/dml/dcn30/dcn30_fpu.c:688: warning: Function parameter or member 'context' not described in 'dcn30_find_dummy_latency_index_for_fw_based_mclk_switch'
display/dc/dml/dcn30/dcn30_fpu.c:688: warning: Function parameter or member 'pipes' not described in 'dcn30_find_dummy_latency_index_for_fw_based_mclk_switch'
display/dc/dml/dcn30/dcn30_fpu.c:688: warning: Function parameter or member 'pipe_cnt' not described in 'dcn30_find_dummy_latency_index_for_fw_based_mclk_switch'
display/dc/dml/dcn30/dcn30_fpu.c:688: warning: Function parameter or member 'vlevel' not described in 'dcn30_find_dummy_latency_index_for_fw_based_mclk_switch'

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c  | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
index a352c703e2589..ccb4ad78f6670 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c
@@ -674,10 +674,19 @@ void dcn30_fpu_update_bw_bounding_box(struct dc *dc,
 }
 
 /**
- * Finds dummy_latency_index when MCLK switching using firmware based
- * vblank stretch is enabled. This function will iterate through the
- * table of dummy pstate latencies until the lowest value that allows
+ * dcn30_find_dummy_latency_index_for_fw_based_mclk_switch() - Finds
+ * dummy_latency_index when MCLK switching using firmware based vblank stretch
+ * is enabled. This function will iterate through the table of dummy pstate
+ * latencies until the lowest value that allows
  * dm_allow_self_refresh_and_mclk_switch to happen is found
+ *
+ * @dc: Current DC state
+ * @context: new dc state
+ * @pipes: DML pipe params
+ * @pipe_cnt: number of DML pipes
+ * @vlevel: Voltage level calculated by DML
+ *
+ * Return: lowest dummy_latency_index value
  */
 int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc,
 							    struct dc_state *context,
-- 
GitLab


From d6634d4d92eac068e2136afab49dfb15a9efae74 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 2 Jun 2023 21:08:49 +0530
Subject: [PATCH 1564/1791] drm/amd/display: Add gnu_printf format attribute
 for snprintf_count()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following W=1 kernel build warning:

display/dc/dcn10/dcn10_hw_sequencer_debug.c: In function ‘snprintf_count’:
display/dc/dcn10/dcn10_hw_sequencer_debug.c:56:2: warning: function ‘snprintf_count’ might be a candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]

Use the __printf() attribute to let the compiler warn if
invalid format strings are passed in.

And fix the following checks:

CHECK: Avoid CamelCase: <pBuf> +unsigned int __printf(3, 4)
snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...)

CHECK: Avoid CamelCase: <bufSize> +unsigned int __printf(3, 4)
snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...)

Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c  | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
index a0f8e31d2adc9..46a2ebcabd1ac 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c
@@ -45,7 +45,8 @@
 #include "dcn10_cm_common.h"
 #include "clk_mgr.h"
 
-unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...)
+__printf(3, 4)
+unsigned int snprintf_count(char *pbuf, unsigned int bufsize, char *fmt, ...)
 {
 	int ret_vsnprintf;
 	unsigned int chars_printed;
@@ -53,15 +54,15 @@ unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...)
 	va_list args;
 	va_start(args, fmt);
 
-	ret_vsnprintf = vsnprintf(pBuf, bufSize, fmt, args);
+	ret_vsnprintf = vsnprintf(pbuf, bufsize, fmt, args);
 
 	va_end(args);
 
 	if (ret_vsnprintf > 0) {
-		if (ret_vsnprintf < bufSize)
+		if (ret_vsnprintf < bufsize)
 			chars_printed = ret_vsnprintf;
 		else
-			chars_printed = bufSize - 1;
+			chars_printed = bufsize - 1;
 	} else
 		chars_printed = 0;
 
-- 
GitLab


From 3cb4807dbf7f47f0a1368f78e7c37b1dd515c2dd Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Tue, 4 Apr 2023 12:53:54 -0400
Subject: [PATCH 1565/1791] Revert "drm/amd/display: Only use ODM2:1 policy for
 high pixel rate displays"

This reverts commit 047783cdd5f604d87398236beb4971abb4d43293 since it
causes higher power consumption for single display use case (4k60).

Also, this patch introduced a 35% performance drop in a Vulkan benchmark.

* The patch disabled the ODM-combination on most popular monitors, including 4K, 2K and FHD monitors at 60Hz.

* ODM-combination can halve the DPP clock to save power, that is the reason why we introduce ODM-combination, and the PM log shows single pipe consumes more power at 4K@60Hz.

* ODM-combination has 2 de-tiled buffer involved, which provides longer self-sustained time, that benefit to the memory power optimization.

Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c | 1 -
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 2e6b39fe26132..8d68f02f5147b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -1918,7 +1918,6 @@ int dcn32_populate_dml_pipes_from_context(
 				context->stream_status[0].plane_count == 1 &&
 				!dc_is_hdmi_signal(res_ctx->pipe_ctx[i].stream->signal) &&
 				is_h_timing_divisible_by_2(res_ctx->pipe_ctx[i].stream) &&
-				pipe->stream->timing.pix_clk_100hz * 100 > DCN3_2_VMIN_DISPCLK_HZ &&
 				dc->debug.enable_single_display_2to1_odm_policy &&
 				!vsr_odm_support) { //excluding 2to1 ODM combine on >= 5k vsr
 			pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 58826e0aa76e3..80bebdf43eca1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -37,7 +37,6 @@
 #define DCN3_2_MBLK_WIDTH 128
 #define DCN3_2_MBLK_HEIGHT_4BPE 128
 #define DCN3_2_MBLK_HEIGHT_8BPE 64
-#define DCN3_2_VMIN_DISPCLK_HZ 717000000
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
 #define SUBVP_HIGH_REFRESH_LIST_LEN 3
 #define DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ 1800
-- 
GitLab


From 1598fc576420207e5c89088fc46610c2318e2f5c Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Thu, 1 Jun 2023 13:09:44 -0400
Subject: [PATCH 1566/1791] drm/amd/display: Program OTG vtotal min/max
 selectors unconditionally for DCN1+

For FPO/FAMS, DMCUB will try to change the output timings by writing to
the OTG registers. However, the timings written directly to the OTG
registers will not be honoured unless VMIN/VMAX selector registers are
programmed with the right bits and trigger source is selected correctly.
Proper solution needs to go into DMCUB but will require additional state
tracking to ensure that the selectors are set and reset correctly as per
driver state. Until fix is merged into firmware, apply the workaround in
driver to unconditionally write OTG vmin/vmax selectors.

Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c | 15 +++------------
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c | 10 ++++++++++
 2 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index e1975991e075d..633989fd2514d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -930,19 +930,10 @@ void optc1_set_drr(
 				OTG_FORCE_LOCK_ON_EVENT, 0,
 				OTG_SET_V_TOTAL_MIN_MASK_EN, 0,
 				OTG_SET_V_TOTAL_MIN_MASK, 0);
-
-		// Setup manual flow control for EOF via TRIG_A
-		optc->funcs->setup_manual_trigger(optc);
-
-	} else {
-		REG_UPDATE_4(OTG_V_TOTAL_CONTROL,
-				OTG_SET_V_TOTAL_MIN_MASK, 0,
-				OTG_V_TOTAL_MIN_SEL, 0,
-				OTG_V_TOTAL_MAX_SEL, 0,
-				OTG_FORCE_LOCK_ON_EVENT, 0);
-
-		optc->funcs->set_vtotal_min_max(optc, 0, 0);
 	}
+
+	// Setup manual flow control for EOF via TRIG_A
+	optc->funcs->setup_manual_trigger(optc);
 }
 
 void optc1_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
index d6f095b4555dc..58bdbd859bf9b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c
@@ -462,6 +462,16 @@ void optc2_setup_manual_trigger(struct timing_generator *optc)
 {
 	struct optc *optc1 = DCN10TG_FROM_TG(optc);
 
+	/* Set the min/max selectors unconditionally so that
+	 * DMCUB fw may change OTG timings when necessary
+	 * TODO: Remove the w/a after fixing the issue in DMCUB firmware
+	 */
+	REG_UPDATE_4(OTG_V_TOTAL_CONTROL,
+				 OTG_V_TOTAL_MIN_SEL, 1,
+				 OTG_V_TOTAL_MAX_SEL, 1,
+				 OTG_FORCE_LOCK_ON_EVENT, 0,
+				 OTG_SET_V_TOTAL_MIN_MASK, (1 << 1)); /* TRIGA */
+
 	REG_SET_8(OTG_TRIGA_CNTL, 0,
 			OTG_TRIGA_SOURCE_SELECT, 21,
 			OTG_TRIGA_SOURCE_PIPE_SELECT, optc->inst,
-- 
GitLab


From 25c30a12d718bd68ad91f58c7546eceaaf0feca5 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 2 Jun 2023 21:14:21 +0530
Subject: [PATCH 1567/1791] drm/amdgpu: Mark
 'kgd_gfx_aldebaran_clear_address_watch' & 'kgd_gfx_v11_clear_address_watch'
 functions as static
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Below two functions cause a warning because they lack a prototype:

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c:164:10: warning: no previous prototype for ‘kgd_gfx_aldebaran_clear_address_watch’ [-Wmissing-prototypes]
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c:782:10: warning: no previous prototype for ‘kgd_gfx_v11_clear_address_watch’ [-Wmissing-prototypes]

There are no callers from other files, so just mark them as 'static'.

Also fixes the following checks:

CHECK: Alignment should match open parenthesis +static uint32_t
kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev,
uint32_t watch_id)

CHECK: Alignment should match open parenthesis +static uint32_t
kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev, uint32_t
watch_id)

Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c   | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
index efd6a72aab4eb..60f9e027fb660 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c
@@ -161,8 +161,8 @@ static uint32_t kgd_gfx_aldebaran_set_address_watch(
 	return watch_address_cntl;
 }
 
-uint32_t kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev,
-					uint32_t watch_id)
+static uint32_t kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev,
+						      uint32_t watch_id)
 {
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index 52efa690a3c21..91c3574ebed30 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -779,8 +779,8 @@ static uint32_t kgd_gfx_v11_set_address_watch(struct amdgpu_device *adev,
 	return watch_address_cntl;
 }
 
-uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev,
-					uint32_t watch_id)
+static uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev,
+						uint32_t watch_id)
 {
 	return 0;
 }
-- 
GitLab


From 09a77a40b51a979557521d5a2d39e431564d5d23 Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Wed, 31 May 2023 16:08:50 +0800
Subject: [PATCH 1568/1791] drm/amdgpu/pm: notify driver unloading to PMFW for
 SMU v13.0.6 dGPU

Per requested, follow the same sequence as APU to send only
PPSMC_MSG_PrepareForDriverUnload to PMFW during driver unloading.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 2611241a5ff16..a92ea4601ea47 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1413,18 +1413,16 @@ static int smu_v13_0_6_system_features_control(struct smu_context *smu,
 					       bool enable)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int ret;
-
-	/* On APUs, notify FW that the device is no longer driver managed */
-	if (adev->flags & AMD_IS_APU) {
-		if (!enable)
-			smu_v13_0_6_notify_unload(smu);
+	int ret = 0;
 
-		return 0;
+	if (enable) {
+		if (!(adev->flags & AMD_IS_APU))
+			ret = smu_v13_0_system_features_control(smu, enable);
+	} else {
+		/* Notify FW that the device is no longer driver managed */
+		smu_v13_0_6_notify_unload(smu);
 	}
 
-	ret = smu_v13_0_system_features_control(smu, enable);
-
 	return ret;
 }
 
-- 
GitLab


From d522ca2714b77e15ebe6e77c1db7468c11a81180 Mon Sep 17 00:00:00 2001
From: Kenneth Feng <kenneth.feng@amd.com>
Date: Mon, 5 Jun 2023 11:15:34 +0800
Subject: [PATCH 1569/1791] drm/amd/pm: update smu-driver if header for smu
 13.0.0 and smu 13.0.10

update smu-driver if header for smu 13.0.0 and smu 13.0.10

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../inc/pmfw_if/smu13_driver_if_v13_0_0.h     | 33 ++++++++++++++-----
 1 file changed, 25 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
index e656e82a0154a..9dd1ed5b89406 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h
@@ -24,10 +24,10 @@
 #ifndef SMU13_DRIVER_IF_V13_0_0_H
 #define SMU13_DRIVER_IF_V13_0_0_H
 
-#define SMU13_0_0_DRIVER_IF_VERSION 0x32
+#define SMU13_0_0_DRIVER_IF_VERSION 0x3D
 
 //Increment this version if SkuTable_t or BoardTable_t change
-#define PPTABLE_VERSION 0x26
+#define PPTABLE_VERSION 0x2B
 
 #define NUM_GFXCLK_DPM_LEVELS    16
 #define NUM_SOCCLK_DPM_LEVELS    8
@@ -96,7 +96,7 @@
 #define FEATURE_ATHUB_MMHUB_PG_BIT            48
 #define FEATURE_SOC_PCC_BIT                   49
 #define FEATURE_EDC_PWRBRK_BIT                50
-#define FEATURE_SPARE_51_BIT                  51
+#define FEATURE_BOMXCO_SVI3_PROG_BIT          51
 #define FEATURE_SPARE_52_BIT                  52
 #define FEATURE_SPARE_53_BIT                  53
 #define FEATURE_SPARE_54_BIT                  54
@@ -312,6 +312,7 @@ typedef enum {
 	I2C_CONTROLLER_PROTOCOL_VR_IR35217,
 	I2C_CONTROLLER_PROTOCOL_TMP_MAX31875,
 	I2C_CONTROLLER_PROTOCOL_INA3221,
+	I2C_CONTROLLER_PROTOCOL_TMP_MAX6604,
 	I2C_CONTROLLER_PROTOCOL_COUNT,
 } I2cControllerProtocol_e;
 
@@ -570,6 +571,7 @@ typedef enum {
 } POWER_SOURCE_e;
 
 typedef enum {
+  MEM_VENDOR_PLACEHOLDER0,
   MEM_VENDOR_SAMSUNG,
   MEM_VENDOR_INFINEON,
   MEM_VENDOR_ELPIDA,
@@ -579,7 +581,6 @@ typedef enum {
   MEM_VENDOR_MOSEL,
   MEM_VENDOR_WINBOND,
   MEM_VENDOR_ESMT,
-  MEM_VENDOR_PLACEHOLDER0,
   MEM_VENDOR_PLACEHOLDER1,
   MEM_VENDOR_PLACEHOLDER2,
   MEM_VENDOR_PLACEHOLDER3,
@@ -812,6 +813,9 @@ typedef enum {
 
 #define INVALID_BOARD_GPIO 0xFF
 
+#define MARKETING_BASE_CLOCKS         0
+#define MARKETING_GAME_CLOCKS         1
+#define MARKETING_BOOST_CLOCKS        2
 
 typedef struct {
   //PLL 0
@@ -1102,10 +1106,15 @@ typedef struct {
   uint16_t        DcsExitHysteresis;    //The min amount of time power credit accumulator should have a value > 0 before SMU exits the DCS throttling phase.
   uint16_t        DcsTimeout;           //This is the amount of time SMU FW waits for RLC to put GFX into GFXOFF before reverting to the fallback mechanism of throttling GFXCLK to Fmin.
 
+  uint8_t         FoptEnabled;
+  uint8_t         DcsSpare2[3];
+  uint32_t        DcsFoptM;             //Tuning paramters to shift Fopt calculation
+  uint32_t        DcsFoptB;             //Tuning paramters to shift Fopt calculation
 
-  uint32_t        DcsSpare[16];
+  uint32_t        DcsSpare[11];
 
   // UCLK section
+  uint16_t     ShadowFreqTableUclk[NUM_UCLK_DPM_LEVELS];     // In MHz
   uint8_t      UseStrobeModeOptimizations; //Set to indicate that FW should use strobe mode optimizations
   uint8_t      PaddingMem[3];
 
@@ -1251,8 +1260,13 @@ typedef struct {
   QuadraticInt_t qFeffCoeffBaseClock[POWER_SOURCE_COUNT];
   QuadraticInt_t qFeffCoeffBoostClock[POWER_SOURCE_COUNT];
 
+  uint16_t TemperatureLimit_Hynix; // In degrees Celsius. Memory temperature limit associated with Hynix
+  uint16_t TemperatureLimit_Micron; // In degrees Celsius. Memory temperature limit associated with Micron
+  uint16_t TemperatureFwCtfLimit_Hynix;
+  uint16_t TemperatureFwCtfLimit_Micron;
+
   // SECTION: Sku Reserved
-  uint32_t         Spare[43];
+  uint32_t         Spare[41];
 
   // Padding for MMHUB - do not modify this
   uint32_t     MmHubPadding[8];
@@ -1324,8 +1338,9 @@ typedef struct {
   // UCLK Spread Spectrum
   uint8_t      UclkSpreadPercent[MEM_VENDOR_COUNT];
 
+  uint8_t      GfxclkSpreadEnable;
+
   // FCLK Spread Spectrum
-  uint8_t      FclkSpreadPadding;
   uint8_t      FclkSpreadPercent;   // Q4.4
   uint16_t     FclkSpreadFreq;      // kHz
 
@@ -1450,6 +1465,8 @@ typedef struct {
 
 
   uint8_t ThrottlingPercentage[THROTTLER_COUNT];
+  uint8_t VmaxThrottlingPercentage;
+  uint8_t Padding1[3];
 
   //metrics for D3hot entry/exit and driver ARM msgs
   uint32_t D3HotEntryCountPerMode[D3HOT_SEQUENCE_COUNT];
@@ -1469,7 +1486,7 @@ typedef struct {
 
 typedef struct {
   SmuMetrics_t SmuMetrics;
-  uint32_t Spare[30];
+  uint32_t Spare[29];
 
   // Padding - ignore
   uint32_t     MmHubPadding[8]; // SMU internal use
-- 
GitLab


From 61a7c162399590263bf5f1ff5d2de634bfe3ae8d Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Wed, 29 Mar 2023 22:03:09 +0800
Subject: [PATCH 1570/1791] drm/amdgpu: pass xcc mask to ras ta

pass xcc mask to ras ta, ras ta will compare
the mask with the one from chiplet topology.

Changed from V1:
	Remove IP version checking.
	Set ras_cmd->ras_init_message.init_flags.xcc_mask
	directly due to xcc_mask is common structres to
	all the devices.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 2 ++
 drivers/gpu/drm/amd/amdgpu/ta_ras_if.h  | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 789cc16e1be7e..3c7c9fca23910 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1655,6 +1655,8 @@ int psp_ras_initialize(struct psp_context *psp)
 		ras_cmd->ras_in_message.init_flags.poison_mode_en = 1;
 	if (!adev->gmc.xgmi.connected_to_cpu)
 		ras_cmd->ras_in_message.init_flags.dgpu_mode = 1;
+	ras_cmd->ras_in_message.init_flags.xcc_mask =
+		adev->gfx.xcc_mask;
 
 	ret = psp_ta_load(psp, &psp->ras_context.context);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
index 30d0482ac466c..be2984ac00a56 100644
--- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
@@ -129,6 +129,7 @@ struct ta_ras_trigger_error_input {
 struct ta_ras_init_flags {
 	uint8_t poison_mode_en;
 	uint8_t dgpu_mode;
+	uint16_t xcc_mask;
 };
 
 struct ta_ras_output_flags {
-- 
GitLab


From e3959cb5479cd24baf9687734d5591b8e2ee08d4 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Fri, 21 Apr 2023 21:14:36 +0800
Subject: [PATCH 1571/1791] drm/amdgpu: support check vcn jpeg block mask

Support VCN/JPEG instance mask checking, pass logical
mask directly except GFX/SDMA/VCN/JPEG blocks.

Changed from V1:
	correct a typo

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index f2da69adcd9d4..5bd1bdb363ca7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -356,8 +356,12 @@ static void amdgpu_ras_instance_mask_check(struct amdgpu_device *adev,
 	case AMDGPU_RAS_BLOCK__SDMA:
 		mask = GENMASK(adev->sdma.num_instances - 1, 0);
 		break;
+	case AMDGPU_RAS_BLOCK__VCN:
+	case AMDGPU_RAS_BLOCK__JPEG:
+		mask = GENMASK(adev->vcn.num_vcn_inst - 1, 0);
+		break;
 	default:
-		mask = 0;
+		mask = inst_mask;
 		break;
 	}
 
-- 
GitLab


From 3898c8fc42b247c26c43c26873895b87fe58b477 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Fri, 21 Apr 2023 20:58:39 +0800
Subject: [PATCH 1572/1791] drm/amdgpu: convert vcn/jpeg logical mask to
 physical mask

 Changed from V1:
 	Remove amdgpu_ras_logical_mask_to_physical_mask
	due to GET_MASK provides same feature.
	Support convert VCN/JPEG logical mask to physical
	mask.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3c7c9fca23910..27559a05a67b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1691,6 +1691,10 @@ int psp_ras_trigger_error(struct psp_context *psp,
 	case TA_RAS_BLOCK__SDMA:
 		dev_mask = GET_MASK(SDMA0, instance_mask);
 		break;
+	case TA_RAS_BLOCK__VCN:
+	case TA_RAS_BLOCK__JPEG:
+		dev_mask = GET_MASK(VCN, instance_mask);
+		break;
 	default:
 		dev_mask = instance_mask;
 		break;
-- 
GitLab


From a15a77c8e61d2db075cc6e2104bfdebf5c9b966c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Jun 2023 14:32:55 -0400
Subject: [PATCH 1573/1791] Revert "drm/amdgpu: change the reference clock for
 raven/raven2"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit fbc24293ca16b3b9ef891fe32ccd04735a6f8dc1.

This results in inconsistent timing reported via asynchronous
GPU queries.

Link: https://lists.freedesktop.org/archives/amd-gfx/2023-May/093731.html
Cc: Jesse.Zhang@amd.com
Cc: michel@daenzer.net
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 135440b5afe9d..afcaeadda4c74 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -325,10 +325,11 @@ static u32 soc15_get_xclk(struct amdgpu_device *adev)
 	u32 reference_clock = adev->clock.spll.reference_freq;
 
 	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 0) ||
-	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 1) ||
-	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 0) ||
-	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 1))
+	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 1))
 		return 10000;
+	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 0) ||
+	    adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 1))
+		return reference_clock / 4;
 
 	return reference_clock;
 }
-- 
GitLab


From f9bfc9fff2997abe3c1a560a38a0c359775e7ec5 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Jun 2023 14:34:12 -0400
Subject: [PATCH 1574/1791] Revert "drm/amdgpu: Differentiate between Raven2
 and Raven/Picasso according to revision id"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit 9d2d1827af295fd6971786672c41c4dba3657154.

This results in inconsistent timing reported via asynchronous
GPU queries.

Link: https://lists.freedesktop.org/archives/amd-gfx/2023-May/093731.html
Cc: Jesse.Zhang@amd.com
Cc: michel@daenzer.net
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 33 +++++++++++++++------------
 1 file changed, 19 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index f092a1dbdb568..3bc0b100936b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4023,25 +4023,30 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 		clock = clock_lo | (clock_hi << 32ULL);
 		break;
 	case IP_VERSION(9, 1, 0):
-	case IP_VERSION(9, 2, 2):
 		preempt_disable();
-		if (adev->rev_id >= 0x8) {
-			clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
-			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
-			hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
-		} else {
-			clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
+		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
+		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
+		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
+		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
+		 * roughly every 42 seconds.
+		 */
+		if (hi_check != clock_hi) {
 			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
-			hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
+			clock_hi = hi_check;
 		}
+		preempt_enable();
+		clock = clock_lo | (clock_hi << 32ULL);
+		break;
+	case IP_VERSION(9, 2, 2):
+		preempt_disable();
+		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
+		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
+		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
 		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
-		* roughly every 42 seconds.
-		*/
+		 * roughly every 42 seconds.
+		 */
 		if (hi_check != clock_hi) {
-			if (adev->rev_id >= 0x8)
-				clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
-			else
-				clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
+			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
 			clock_hi = hi_check;
 		}
 		preempt_enable();
-- 
GitLab


From 5a03159ab7ef456ba22460e47a9d0eab2f310424 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Jun 2023 14:37:13 -0400
Subject: [PATCH 1575/1791] Revert "drm/amdgpu: switch to golden tsc registers
 for raven/raven2"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit f03eb1d26c2739b75580f58bbab4ab2d5d3eba46.

This results in inconsistent timing reported via asynchronous
GPU queries.

Link: https://lists.freedesktop.org/archives/amd-gfx/2023-May/093731.html
Cc: Jesse.Zhang@amd.com
Cc: michel@daenzer.net
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 40 ---------------------------
 1 file changed, 40 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 3bc0b100936b3..4073e2ee7e6dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -149,16 +149,6 @@ MODULE_FIRMWARE("amdgpu/aldebaran_sjt_mec2.bin");
 #define mmGOLDEN_TSC_COUNT_LOWER_Renoir                0x0026
 #define mmGOLDEN_TSC_COUNT_LOWER_Renoir_BASE_IDX       1
 
-#define mmGOLDEN_TSC_COUNT_UPPER_Raven   0x007a
-#define mmGOLDEN_TSC_COUNT_UPPER_Raven_BASE_IDX 0
-#define mmGOLDEN_TSC_COUNT_LOWER_Raven   0x007b
-#define mmGOLDEN_TSC_COUNT_LOWER_Raven_BASE_IDX 0
-
-#define mmGOLDEN_TSC_COUNT_UPPER_Raven2   0x0068
-#define mmGOLDEN_TSC_COUNT_UPPER_Raven2_BASE_IDX 0
-#define mmGOLDEN_TSC_COUNT_LOWER_Raven2   0x0069
-#define mmGOLDEN_TSC_COUNT_LOWER_Raven2_BASE_IDX 0
-
 enum ta_ras_gfx_subblock {
 	/*CPC*/
 	TA_RAS_BLOCK__GFX_CPC_INDEX_START = 0,
@@ -4022,36 +4012,6 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 		preempt_enable();
 		clock = clock_lo | (clock_hi << 32ULL);
 		break;
-	case IP_VERSION(9, 1, 0):
-		preempt_disable();
-		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
-		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
-		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven);
-		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
-		 * roughly every 42 seconds.
-		 */
-		if (hi_check != clock_hi) {
-			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven);
-			clock_hi = hi_check;
-		}
-		preempt_enable();
-		clock = clock_lo | (clock_hi << 32ULL);
-		break;
-	case IP_VERSION(9, 2, 2):
-		preempt_disable();
-		clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
-		clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
-		hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2);
-		/* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over
-		 * roughly every 42 seconds.
-		 */
-		if (hi_check != clock_hi) {
-			clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2);
-			clock_hi = hi_check;
-		}
-		preempt_enable();
-		clock = clock_lo | (clock_hi << 32ULL);
-		break;
 	default:
 		amdgpu_gfx_off_ctrl(adev, false);
 		mutex_lock(&adev->gfx.gpu_clock_mutex);
-- 
GitLab


From e6850f98efc70277dc0e941e905182738e7327a0 Mon Sep 17 00:00:00 2001
From: Min Li <lm0963hack@gmail.com>
Date: Sat, 3 Jun 2023 15:43:45 +0800
Subject: [PATCH 1576/1791] drm/radeon: fix race condition UAF in
 radeon_gem_set_domain_ioctl
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Userspace can race to free the gobj(robj converted from), robj should not
be accessed again after drm_gem_object_put, otherwith it will result in
use-after-free.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Min Li <lm0963hack@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_gem.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index 5de99ffa072f0..d0119c5f7eb3e 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -459,7 +459,6 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_set_domain *args = data;
 	struct drm_gem_object *gobj;
-	struct radeon_bo *robj;
 	int r;
 
 	/* for now if someone requests domain CPU -
@@ -472,13 +471,12 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 		up_read(&rdev->exclusive_lock);
 		return -ENOENT;
 	}
-	robj = gem_to_radeon_bo(gobj);
 
 	r = radeon_gem_set_domain(gobj, args->read_domains, args->write_domain);
 
 	drm_gem_object_put(gobj);
 	up_read(&rdev->exclusive_lock);
-	r = radeon_gem_handle_lockup(robj->rdev, r);
+	r = radeon_gem_handle_lockup(rdev, r);
 	return r;
 }
 
-- 
GitLab


From 5a863904bab4f5d22012f7d68ab2becafc303a40 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 1 Jun 2023 14:48:08 -0700
Subject: [PATCH 1577/1791] drm/amdgpu: fix xclk freq on CHIP_STONEY

According to Alex, most APUs from that time seem to have the same issue
(vbios says 48Mhz, actual is 100Mhz).  I only have a CHIP_STONEY so I
limit the fixup to CHIP_STONEY

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vi.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 770f2d7a371fc..6a8494f98d3ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -542,8 +542,15 @@ static u32 vi_get_xclk(struct amdgpu_device *adev)
 	u32 reference_clock = adev->clock.spll.reference_freq;
 	u32 tmp;
 
-	if (adev->flags & AMD_IS_APU)
-		return reference_clock;
+	if (adev->flags & AMD_IS_APU) {
+		switch (adev->asic_type) {
+		case CHIP_STONEY:
+			/* vbios says 48Mhz, but the actual freq is 100Mhz */
+			return 10000;
+		default:
+			return reference_clock;
+		}
+	}
 
 	tmp = RREG32_SMC(ixCG_CLKPIN_CNTL_2);
 	if (REG_GET_FIELD(tmp, CG_CLKPIN_CNTL_2, MUX_TCLK_TO_XCLK))
-- 
GitLab


From 869bcf59fd64382e3b23b219e791e6e5ebf1114e Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Wed, 24 May 2023 17:14:15 +0800
Subject: [PATCH 1578/1791] drm/amdgpu: change reserved vram info print
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The link object of mgr->reserved_pages is the blocks
variable in struct amdgpu_vram_reservation, not the
link variable in struct drm_buddy_block.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 89d35d194f2c4..c7085a747b03b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -839,7 +839,7 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man,
 {
 	struct amdgpu_vram_mgr *mgr = to_vram_mgr(man);
 	struct drm_buddy *mm = &mgr->mm;
-	struct drm_buddy_block *block;
+	struct amdgpu_vram_reservation *rsv;
 
 	drm_printf(printer, "  vis usage:%llu\n",
 		   amdgpu_vram_mgr_vis_usage(mgr));
@@ -851,8 +851,9 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man,
 	drm_buddy_print(mm, printer);
 
 	drm_printf(printer, "reserved:\n");
-	list_for_each_entry(block, &mgr->reserved_pages, link)
-		drm_buddy_block_print(mm, block, printer);
+	list_for_each_entry(rsv, &mgr->reserved_pages, blocks)
+		drm_printf(printer, "%#018llx-%#018llx: %llu\n",
+			rsv->start, rsv->start + rsv->size, rsv->size);
 	mutex_unlock(&mgr->lock);
 }
 
-- 
GitLab


From 49f26218c344741cb3eaa740b1e44e960551a87f Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Tue, 16 May 2023 15:50:40 -0400
Subject: [PATCH 1579/1791] drm/amd/display: fix dcn315 single stream crb
 allocation

Change to improve avoiding asymetric crb calculations for single stream
scenarios.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn315/dcn315_resource.c   | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index cb95e978417b6..8570bdc292b4f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -1628,6 +1628,10 @@ static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context)
 	int i;
 	struct resource_context *res_ctx = &context->res_ctx;
 
+	/*Don't apply for single stream*/
+	if (context->stream_count < 2)
+		return false;
+
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		if (!res_ctx->pipe_ctx[i].stream)
 			continue;
@@ -1727,19 +1731,23 @@ static int dcn315_populate_dml_pipes_from_context(
 		pipe_cnt++;
 	}
 
-	/* Spread remaining unreserved crb evenly among all pipes, use default policy if not enough det or single pipe */
+	/* Spread remaining unreserved crb evenly among all pipes*/
 	if (pixel_rate_crb) {
 		for (i = 0, pipe_cnt = 0, crb_idx = 0; i < dc->res_pool->pipe_count; i++) {
 			pipe = &res_ctx->pipe_ctx[i];
 			if (!pipe->stream)
 				continue;
 
+			/* Do not use asymetric crb if not enough for pstate support */
+			if (remaining_det_segs < 0) {
+				pipes[pipe_cnt].pipe.src.det_size_override = 0;
+				continue;
+			}
+
 			if (!pipe->top_pipe && !pipe->prev_odm_pipe) {
 				bool split_required = pipe->stream->timing.pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc)
 						|| (pipe->plane_state && pipe->plane_state->src_rect.width > 5120);
 
-				if (remaining_det_segs < 0 || crb_pipes == 1)
-					pipes[pipe_cnt].pipe.src.det_size_override = 0;
 				if (remaining_det_segs > MIN_RESERVED_DET_SEGS)
 					pipes[pipe_cnt].pipe.src.det_size_override += (remaining_det_segs - MIN_RESERVED_DET_SEGS) / crb_pipes +
 							(crb_idx < (remaining_det_segs - MIN_RESERVED_DET_SEGS) % crb_pipes ? 1 : 0);
@@ -1755,6 +1763,7 @@ static int dcn315_populate_dml_pipes_from_context(
 				}
 				/* Convert segments into size for DML use */
 				pipes[pipe_cnt].pipe.src.det_size_override *= DCN3_15_CRB_SEGMENT_SIZE_KB;
+
 				crb_idx++;
 			}
 			pipe_cnt++;
-- 
GitLab


From 0baae624630788862bbd654741929007971e9d5b Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Thu, 18 May 2023 11:30:44 -0400
Subject: [PATCH 1580/1791] drm/amd/display: Refactor fast update to use new
 HWSS build sequence

[Description]
- Refactor HW sequencer to use a build / execute sequence
- Also move gamma updates to become fast

v2: squash in build fix ("drm/amd/display: Fix guarding of 'if (dc->debug.visual_confirm)'")

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 274 ++++++++++++++++--
 .../drm/amd/display/dc/core/dc_hw_sequencer.c | 255 ++++++++++++++++
 drivers/gpu/drm/amd/display/dc/dc.h           |   1 +
 .../amd/display/dc/dce100/dce100_resource.c   |   5 +
 .../amd/display/dc/dce110/dce110_resource.c   |   5 +
 .../amd/display/dc/dce112/dce112_resource.c   |   5 +
 .../amd/display/dc/dce120/dce120_resource.c   |   1 +
 .../drm/amd/display/dc/dce80/dce80_resource.c |   6 +
 .../drm/amd/display/dc/dcn10/dcn10_resource.c |   1 +
 .../drm/amd/display/dc/dcn20/dcn20_resource.c |   1 +
 .../amd/display/dc/dcn201/dcn201_resource.c   |   1 +
 .../drm/amd/display/dc/dcn21/dcn21_resource.c |   1 +
 .../drm/amd/display/dc/dcn31/dcn31_resource.c |   1 +
 .../amd/display/dc/dcn315/dcn315_resource.c   |   1 +
 .../amd/display/dc/dcn316/dcn316_resource.c   |   1 +
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.c    |  24 ++
 .../drm/amd/display/dc/dcn32/dcn32_hwseq.h    |   2 +
 .../gpu/drm/amd/display/dc/dcn32/dcn32_init.c |   1 +
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  11 +
 .../gpu/drm/amd/display/dc/inc/hw_sequencer.h | 126 ++++++++
 20 files changed, 693 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 0e9403f21dae9..60da6d834518c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2589,15 +2589,19 @@ static enum surface_update_type det_surface_update(const struct dc *dc,
 		elevate_update_type(&overall_type, type);
 	}
 
-	if (update_flags->bits.input_csc_change
-			|| update_flags->bits.coeff_reduction_change
-			|| update_flags->bits.lut_3d
-			|| update_flags->bits.gamma_change
-			|| update_flags->bits.gamut_remap_change) {
+	if (update_flags->bits.lut_3d) {
 		type = UPDATE_TYPE_FULL;
 		elevate_update_type(&overall_type, type);
 	}
 
+	if (dc->debug.enable_legacy_fast_update &&
+			(update_flags->bits.gamma_change ||
+			update_flags->bits.gamut_remap_change ||
+			update_flags->bits.input_csc_change ||
+			update_flags->bits.coeff_reduction_change)) {
+		type = UPDATE_TYPE_FULL;
+		elevate_update_type(&overall_type, type);
+	}
 	return overall_type;
 }
 
@@ -2630,7 +2634,7 @@ static enum surface_update_type check_update_surfaces_for_stream(
 			stream_update->integer_scaling_update)
 			su_flags->bits.scaling = 1;
 
-		if (stream_update->out_transfer_func)
+		if (dc->debug.enable_legacy_fast_update && stream_update->out_transfer_func)
 			su_flags->bits.out_tf = 1;
 
 		if (stream_update->abm_level)
@@ -2661,6 +2665,12 @@ static enum surface_update_type check_update_surfaces_for_stream(
 
 		if (stream_update->output_csc_transform || stream_update->output_color_space)
 			su_flags->bits.out_csc = 1;
+
+		/* Output transfer function changes do not require bandwidth recalculation,
+		 * so don't trigger a full update
+		 */
+		if (!dc->debug.enable_legacy_fast_update && stream_update->out_transfer_func)
+			su_flags->bits.out_tf = 1;
 	}
 
 	for (i = 0 ; i < surface_count; i++) {
@@ -3412,6 +3422,166 @@ void dc_dmub_update_dirty_rect(struct dc *dc,
 	}
 }
 
+static void build_dmub_update_dirty_rect(
+		struct dc *dc,
+		int surface_count,
+		struct dc_stream_state *stream,
+		struct dc_surface_update *srf_updates,
+		struct dc_state *context,
+		struct dc_dmub_cmd dc_dmub_cmd[],
+		unsigned int *dmub_cmd_count)
+{
+	union dmub_rb_cmd cmd;
+	struct dmub_cmd_update_dirty_rect_data *update_dirty_rect;
+	unsigned int i, j;
+	unsigned int panel_inst = 0;
+
+	if (!dc_dmub_should_send_dirty_rect_cmd(dc, stream))
+		return;
+
+	if (!dc_get_edp_link_panel_inst(dc, stream->link, &panel_inst))
+		return;
+
+	memset(&cmd, 0x0, sizeof(cmd));
+	cmd.update_dirty_rect.header.type = DMUB_CMD__UPDATE_DIRTY_RECT;
+	cmd.update_dirty_rect.header.sub_type = 0;
+	cmd.update_dirty_rect.header.payload_bytes =
+		sizeof(cmd.update_dirty_rect) -
+		sizeof(cmd.update_dirty_rect.header);
+	update_dirty_rect = &cmd.update_dirty_rect.update_dirty_rect_data;
+	for (i = 0; i < surface_count; i++) {
+		struct dc_plane_state *plane_state = srf_updates[i].surface;
+		const struct dc_flip_addrs *flip_addr = srf_updates[i].flip_addr;
+
+		if (!srf_updates[i].surface || !flip_addr)
+			continue;
+		/* Do not send in immediate flip mode */
+		if (srf_updates[i].surface->flip_immediate)
+			continue;
+		update_dirty_rect->cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1;
+		update_dirty_rect->dirty_rect_count = flip_addr->dirty_rect_count;
+		memcpy(update_dirty_rect->src_dirty_rects, flip_addr->dirty_rects,
+				sizeof(flip_addr->dirty_rects));
+		for (j = 0; j < dc->res_pool->pipe_count; j++) {
+			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+			if (pipe_ctx->stream != stream)
+				continue;
+			if (pipe_ctx->plane_state != plane_state)
+				continue;
+			update_dirty_rect->panel_inst = panel_inst;
+			update_dirty_rect->pipe_idx = j;
+			dc_dmub_cmd[*dmub_cmd_count].dmub_cmd = cmd;
+			dc_dmub_cmd[*dmub_cmd_count].wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT;
+			(*dmub_cmd_count)++;
+		}
+	}
+}
+
+
+/**
+ * ************************************************************************************************
+ * build_dmub_cmd_list: Build an array of DMCUB commands to be sent to DMCUB
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: srf_updates: Array of surface updates
+ * @param [in]: surface_count: Number of surfaces that have an updated
+ * @param [in]: stream: Correponding stream to be updated in the current flip
+ * @param [in]: context: New DC state to be programmed
+ *
+ * @param [out]: dc_dmub_cmd: Array of DMCUB commands to be sent to DMCUB
+ * @param [out]: dmub_cmd_count: Count indicating the number of DMCUB commands in dc_dmub_cmd array
+ *
+ * This function builds an array of DMCUB commands to be sent to DMCUB. This function is required
+ * to build an array of commands and have them sent while the OTG lock is acquired.
+ *
+ * @return: void
+ * ************************************************************************************************
+ */
+static void build_dmub_cmd_list(struct dc *dc,
+		struct dc_surface_update *srf_updates,
+		int surface_count,
+		struct dc_stream_state *stream,
+		struct dc_state *context,
+		struct dc_dmub_cmd dc_dmub_cmd[],
+		unsigned int *dmub_cmd_count)
+{
+	// Initialize cmd count to 0
+	*dmub_cmd_count = 0;
+	build_dmub_update_dirty_rect(dc, surface_count, stream, srf_updates, context, dc_dmub_cmd, dmub_cmd_count);
+}
+
+static void commit_planes_for_stream_fast(struct dc *dc,
+		struct dc_surface_update *srf_updates,
+		int surface_count,
+		struct dc_stream_state *stream,
+		struct dc_stream_update *stream_update,
+		enum surface_update_type update_type,
+		struct dc_state *context)
+{
+	int i, j;
+	struct pipe_ctx *top_pipe_to_program = NULL;
+	bool should_lock_all_pipes = (update_type != UPDATE_TYPE_FAST);
+	dc_z10_restore(dc);
+
+	for (j = 0; j < dc->res_pool->pipe_count; j++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+		if (!pipe_ctx->top_pipe &&
+			!pipe_ctx->prev_odm_pipe &&
+			pipe_ctx->stream &&
+			pipe_ctx->stream == stream) {
+			top_pipe_to_program = pipe_ctx;
+		}
+	}
+
+	if (dc->debug.visual_confirm) {
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+			if (pipe->stream && pipe->plane_state)
+				dc_update_viusal_confirm_color(dc, context, pipe);
+		}
+	}
+
+	for (i = 0; i < surface_count; i++) {
+		struct dc_plane_state *plane_state = srf_updates[i].surface;
+		/*set logical flag for lock/unlock use*/
+		for (j = 0; j < dc->res_pool->pipe_count; j++) {
+			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j];
+
+			if (!pipe_ctx->plane_state)
+				continue;
+			if (should_update_pipe_for_plane(context, pipe_ctx, plane_state))
+				continue;
+			pipe_ctx->plane_state->triplebuffer_flips = false;
+			if (update_type == UPDATE_TYPE_FAST &&
+			    dc->hwss.program_triplebuffer &&
+			    !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) {
+				/*triple buffer for VUpdate  only*/
+				pipe_ctx->plane_state->triplebuffer_flips = true;
+			}
+		}
+	}
+
+	build_dmub_cmd_list(dc,
+			srf_updates,
+			surface_count,
+			stream,
+			context,
+			context->dc_dmub_cmd,
+			&(context->dmub_cmd_count));
+	hwss_build_fast_sequence(dc,
+			context->dc_dmub_cmd,
+			context->dmub_cmd_count,
+			context->block_sequence,
+			&(context->block_sequence_steps),
+			top_pipe_to_program);
+	hwss_execute_sequence(dc,
+			context->block_sequence,
+			context->block_sequence_steps);
+}
+
 static void commit_planes_for_stream(struct dc *dc,
 		struct dc_surface_update *srf_updates,
 		int surface_count,
@@ -3449,21 +3619,6 @@ static void commit_planes_for_stream(struct dc *dc,
 		}
 	}
 
-	if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) {
-		/* Optimize seamless boot flag keeps clocks and watermarks high until
-		 * first flip. After first flip, optimization is required to lower
-		 * bandwidth. Important to note that it is expected UEFI will
-		 * only light up a single display on POST, therefore we only expect
-		 * one stream with seamless boot flag set.
-		 */
-		if (stream->apply_seamless_boot_optimization) {
-			stream->apply_seamless_boot_optimization = false;
-
-			if (get_seamless_boot_stream_count(context) == 0)
-				dc->optimized_required = true;
-		}
-	}
-
 	if (update_type == UPDATE_TYPE_FULL) {
 		dc_allow_idle_optimizations(dc, false);
 
@@ -4046,6 +4201,43 @@ static bool commit_minimal_transition_state(struct dc *dc,
 	return true;
 }
 
+/**
+ * *******************************************************************************
+ * update_seamless_boot_flags: Helper function for updating seamless boot flags
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ * @param [in]: surface_count: Number of surfaces that have an updated
+ * @param [in]: stream: Correponding stream to be updated in the current flip
+ *
+ * Updating seamless boot flags do not need to be part of the commit sequence. This
+ * helper function will update the seamless boot flags on each flip (if required)
+ * outside of the HW commit sequence (fast or slow).
+ *
+ * @return: void
+ * *******************************************************************************
+ */
+static void update_seamless_boot_flags(struct dc *dc,
+		struct dc_state *context,
+		int surface_count,
+		struct dc_stream_state *stream)
+{
+	if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) {
+		/* Optimize seamless boot flag keeps clocks and watermarks high until
+		 * first flip. After first flip, optimization is required to lower
+		 * bandwidth. Important to note that it is expected UEFI will
+		 * only light up a single display on POST, therefore we only expect
+		 * one stream with seamless boot flag set.
+		 */
+		if (stream->apply_seamless_boot_optimization) {
+			stream->apply_seamless_boot_optimization = false;
+
+			if (get_seamless_boot_stream_count(context) == 0)
+				dc->optimized_required = true;
+		}
+	}
+}
+
 bool dc_update_planes_and_stream(struct dc *dc,
 		struct dc_surface_update *srf_updates, int surface_count,
 		struct dc_stream_state *stream,
@@ -4112,14 +4304,25 @@ bool dc_update_planes_and_stream(struct dc *dc,
 		update_type = UPDATE_TYPE_FULL;
 	}
 
-	commit_planes_for_stream(
-			dc,
-			srf_updates,
-			surface_count,
-			stream,
-			stream_update,
-			update_type,
-			context);
+	update_seamless_boot_flags(dc, context, surface_count, stream);
+	if (!dc->debug.enable_legacy_fast_update && update_type == UPDATE_TYPE_FAST) {
+		commit_planes_for_stream_fast(dc,
+				srf_updates,
+				surface_count,
+				stream,
+				stream_update,
+				update_type,
+				context);
+	} else {
+		commit_planes_for_stream(
+				dc,
+				srf_updates,
+				surface_count,
+				stream,
+				stream_update,
+				update_type,
+				context);
+	}
 
 	if (dc->current_state != context) {
 
@@ -4244,7 +4447,17 @@ void dc_commit_updates_for_stream(struct dc *dc,
 
 	TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES);
 
-	commit_planes_for_stream(
+	update_seamless_boot_flags(dc, context, surface_count, stream);
+	if (!dc->debug.enable_legacy_fast_update && update_type == UPDATE_TYPE_FAST) {
+		commit_planes_for_stream_fast(dc,
+				srf_updates,
+				surface_count,
+				stream,
+				stream_update,
+				update_type,
+				context);
+	} else {
+		commit_planes_for_stream(
 				dc,
 				srf_updates,
 				surface_count,
@@ -4252,6 +4465,7 @@ void dc_commit_updates_for_stream(struct dc *dc,
 				stream_update,
 				update_type,
 				context);
+	}
 	/*update current_State*/
 	if (dc->current_state != context) {
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index 8a98b8dd008ef..182c42c63bc5e 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -27,6 +27,8 @@
 #include "core_types.h"
 #include "timing_generator.h"
 #include "hw_sequencer.h"
+#include "hw_sequencer_private.h"
+#include "basics/dc_common.h"
 
 #define NUM_ELEMENTS(a) (sizeof(a) / sizeof((a)[0]))
 
@@ -463,6 +465,259 @@ void get_subvp_visual_confirm_color(
 	}
 }
 
+void hwss_build_fast_sequence(struct dc *dc,
+		struct dc_dmub_cmd *dc_dmub_cmd,
+		unsigned int dmub_cmd_count,
+		struct block_sequence block_sequence[],
+		int *num_steps,
+		struct pipe_ctx *pipe_ctx)
+{
+	struct dc_plane_state *plane = pipe_ctx->plane_state;
+	struct dc_stream_state *stream = pipe_ctx->stream;
+	struct dce_hwseq *hws = dc->hwseq;
+	struct pipe_ctx *current_pipe = NULL;
+	struct pipe_ctx *current_mpc_pipe = NULL;
+	unsigned int i = 0;
+
+	*num_steps = 0; // Initialize to 0
+
+	if (!plane || !stream)
+		return;
+
+	if (dc->hwss.subvp_pipe_control_lock_fast) {
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc;
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = true;
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx;
+		block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST;
+		(*num_steps)++;
+	}
+	if (dc->hwss.pipe_control_lock) {
+		block_sequence[*num_steps].params.pipe_control_lock_params.dc = dc;
+		block_sequence[*num_steps].params.pipe_control_lock_params.lock = true;
+		block_sequence[*num_steps].params.pipe_control_lock_params.pipe_ctx = pipe_ctx;
+		block_sequence[*num_steps].func = OPTC_PIPE_CONTROL_LOCK;
+		(*num_steps)++;
+	}
+
+	for (i = 0; i < dmub_cmd_count; i++) {
+		block_sequence[*num_steps].params.send_dmcub_cmd_params.ctx = dc->ctx;
+		block_sequence[*num_steps].params.send_dmcub_cmd_params.cmd = &(dc_dmub_cmd[i].dmub_cmd);
+		block_sequence[*num_steps].params.send_dmcub_cmd_params.wait_type = dc_dmub_cmd[i].wait_type;
+		block_sequence[*num_steps].func = DMUB_SEND_DMCUB_CMD;
+		(*num_steps)++;
+	}
+
+	current_pipe = pipe_ctx;
+	while (current_pipe) {
+		current_mpc_pipe = current_pipe;
+		while (current_mpc_pipe) {
+			if (dc->hwss.set_flip_control_gsl && current_mpc_pipe->plane_state && current_mpc_pipe->plane_state->update_flags.raw) {
+				block_sequence[*num_steps].params.set_flip_control_gsl_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].params.set_flip_control_gsl_params.flip_immediate = current_mpc_pipe->plane_state->flip_immediate;
+				block_sequence[*num_steps].func = HUBP_SET_FLIP_CONTROL_GSL;
+				(*num_steps)++;
+			}
+			if (dc->hwss.program_triplebuffer && dc->debug.enable_tri_buf && current_mpc_pipe->plane_state->update_flags.raw) {
+				block_sequence[*num_steps].params.program_triplebuffer_params.dc = dc;
+				block_sequence[*num_steps].params.program_triplebuffer_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].params.program_triplebuffer_params.enableTripleBuffer = current_mpc_pipe->plane_state->triplebuffer_flips;
+				block_sequence[*num_steps].func = HUBP_PROGRAM_TRIPLEBUFFER;
+				(*num_steps)++;
+			}
+			if (dc->hwss.update_plane_addr && current_mpc_pipe->plane_state->update_flags.bits.addr_update) {
+				block_sequence[*num_steps].params.update_plane_addr_params.dc = dc;
+				block_sequence[*num_steps].params.update_plane_addr_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].func = HUBP_UPDATE_PLANE_ADDR;
+				(*num_steps)++;
+			}
+
+			if (hws->funcs.set_input_transfer_func && current_mpc_pipe->plane_state->update_flags.bits.gamma_change) {
+				block_sequence[*num_steps].params.set_input_transfer_func_params.dc = dc;
+				block_sequence[*num_steps].params.set_input_transfer_func_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].params.set_input_transfer_func_params.plane_state = current_mpc_pipe->plane_state;
+				block_sequence[*num_steps].func = DPP_SET_INPUT_TRANSFER_FUNC;
+				(*num_steps)++;
+			}
+
+			if (dc->hwss.program_gamut_remap && current_mpc_pipe->plane_state->update_flags.bits.gamut_remap_change) {
+				block_sequence[*num_steps].params.program_gamut_remap_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].func = DPP_PROGRAM_GAMUT_REMAP;
+				(*num_steps)++;
+			}
+			if (current_mpc_pipe->plane_state->update_flags.bits.input_csc_change) {
+				block_sequence[*num_steps].params.setup_dpp_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].func = DPP_SETUP_DPP;
+				(*num_steps)++;
+			}
+			if (current_mpc_pipe->plane_state->update_flags.bits.coeff_reduction_change) {
+				block_sequence[*num_steps].params.program_bias_and_scale_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].func = DPP_PROGRAM_BIAS_AND_SCALE;
+				(*num_steps)++;
+			}
+			if (hws->funcs.set_output_transfer_func && current_mpc_pipe->stream->update_flags.bits.out_tf) {
+				block_sequence[*num_steps].params.set_output_transfer_func_params.dc = dc;
+				block_sequence[*num_steps].params.set_output_transfer_func_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].params.set_output_transfer_func_params.stream = current_mpc_pipe->stream;
+				block_sequence[*num_steps].func = DPP_SET_OUTPUT_TRANSFER_FUNC;
+				(*num_steps)++;
+			}
+
+			current_mpc_pipe = current_mpc_pipe->bottom_pipe;
+		}
+		current_pipe = current_pipe->next_odm_pipe;
+	}
+
+	if (dc->hwss.pipe_control_lock) {
+		block_sequence[*num_steps].params.pipe_control_lock_params.dc = dc;
+		block_sequence[*num_steps].params.pipe_control_lock_params.lock = false;
+		block_sequence[*num_steps].params.pipe_control_lock_params.pipe_ctx = pipe_ctx;
+		block_sequence[*num_steps].func = OPTC_PIPE_CONTROL_LOCK;
+		(*num_steps)++;
+	}
+	if (dc->hwss.subvp_pipe_control_lock_fast) {
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc;
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = false;
+		block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx;
+		block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST;
+		(*num_steps)++;
+	}
+
+	current_pipe = pipe_ctx;
+	while (current_pipe) {
+		current_mpc_pipe = current_pipe;
+
+		while (current_mpc_pipe) {
+			if (!current_mpc_pipe->bottom_pipe && !pipe_ctx->next_odm_pipe &&
+					current_mpc_pipe->stream && current_mpc_pipe->plane_state &&
+					current_mpc_pipe->plane_state->update_flags.bits.addr_update &&
+					!current_mpc_pipe->plane_state->skip_manual_trigger) {
+				block_sequence[*num_steps].params.program_manual_trigger_params.pipe_ctx = current_mpc_pipe;
+				block_sequence[*num_steps].func = OPTC_PROGRAM_MANUAL_TRIGGER;
+				(*num_steps)++;
+			}
+			current_mpc_pipe = current_mpc_pipe->bottom_pipe;
+		}
+		current_pipe = current_pipe->next_odm_pipe;
+	}
+}
+
+void hwss_execute_sequence(struct dc *dc,
+		struct block_sequence block_sequence[],
+		int num_steps)
+{
+	unsigned int i;
+	union block_sequence_params *params;
+	struct dce_hwseq *hws = dc->hwseq;
+
+	for (i = 0; i < num_steps; i++) {
+		params = &(block_sequence[i].params);
+		switch (block_sequence[i].func) {
+
+		case DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST:
+			dc->hwss.subvp_pipe_control_lock_fast(params);
+			break;
+		case OPTC_PIPE_CONTROL_LOCK:
+			dc->hwss.pipe_control_lock(params->pipe_control_lock_params.dc,
+					params->pipe_control_lock_params.pipe_ctx,
+					params->pipe_control_lock_params.lock);
+			break;
+		case HUBP_SET_FLIP_CONTROL_GSL:
+			dc->hwss.set_flip_control_gsl(params->set_flip_control_gsl_params.pipe_ctx,
+					params->set_flip_control_gsl_params.flip_immediate);
+			break;
+		case HUBP_PROGRAM_TRIPLEBUFFER:
+			dc->hwss.program_triplebuffer(params->program_triplebuffer_params.dc,
+					params->program_triplebuffer_params.pipe_ctx,
+					params->program_triplebuffer_params.enableTripleBuffer);
+			break;
+		case HUBP_UPDATE_PLANE_ADDR:
+			dc->hwss.update_plane_addr(params->update_plane_addr_params.dc,
+					params->update_plane_addr_params.pipe_ctx);
+			break;
+		case DPP_SET_INPUT_TRANSFER_FUNC:
+			hws->funcs.set_input_transfer_func(params->set_input_transfer_func_params.dc,
+					params->set_input_transfer_func_params.pipe_ctx,
+					params->set_input_transfer_func_params.plane_state);
+			break;
+		case DPP_PROGRAM_GAMUT_REMAP:
+			dc->hwss.program_gamut_remap(params->program_gamut_remap_params.pipe_ctx);
+			break;
+		case DPP_SETUP_DPP:
+			hwss_setup_dpp(params);
+			break;
+		case DPP_PROGRAM_BIAS_AND_SCALE:
+			hwss_program_bias_and_scale(params);
+			break;
+		case OPTC_PROGRAM_MANUAL_TRIGGER:
+			hwss_program_manual_trigger(params);
+			break;
+		case DPP_SET_OUTPUT_TRANSFER_FUNC:
+			hws->funcs.set_output_transfer_func(params->set_output_transfer_func_params.dc,
+					params->set_output_transfer_func_params.pipe_ctx,
+					params->set_output_transfer_func_params.stream);
+			break;
+		case MPC_UPDATE_VISUAL_CONFIRM:
+			dc->hwss.update_visual_confirm_color(params->update_visual_confirm_params.dc,
+					params->update_visual_confirm_params.pipe_ctx,
+					params->update_visual_confirm_params.mpcc_id);
+			break;
+		case DMUB_SEND_DMCUB_CMD:
+			hwss_send_dmcub_cmd(params);
+			break;
+		default:
+			ASSERT(false);
+			break;
+		}
+	}
+}
+
+void hwss_send_dmcub_cmd(union block_sequence_params *params)
+{
+	struct dc_context *ctx = params->send_dmcub_cmd_params.ctx;
+	union dmub_rb_cmd *cmd = params->send_dmcub_cmd_params.cmd;
+	enum dm_dmub_wait_type wait_type = params->send_dmcub_cmd_params.wait_type;
+
+	dm_execute_dmub_cmd(ctx, cmd, wait_type);
+}
+
+void hwss_program_manual_trigger(union block_sequence_params *params)
+{
+	struct pipe_ctx *pipe_ctx = params->program_manual_trigger_params.pipe_ctx;
+
+	if (pipe_ctx->stream_res.tg->funcs->program_manual_trigger)
+		pipe_ctx->stream_res.tg->funcs->program_manual_trigger(pipe_ctx->stream_res.tg);
+}
+
+void hwss_setup_dpp(union block_sequence_params *params)
+{
+	struct pipe_ctx *pipe_ctx = params->setup_dpp_params.pipe_ctx;
+	struct dpp *dpp = pipe_ctx->plane_res.dpp;
+	struct dc_plane_state *plane_state = pipe_ctx->plane_state;
+
+	if (dpp && dpp->funcs->dpp_setup) {
+		// program the input csc
+		dpp->funcs->dpp_setup(dpp,
+				plane_state->format,
+				EXPANSION_MODE_ZERO,
+				plane_state->input_csc_color_matrix,
+				plane_state->color_space,
+				NULL);
+	}
+}
+
+void hwss_program_bias_and_scale(union block_sequence_params *params)
+{
+	struct pipe_ctx *pipe_ctx = params->program_bias_and_scale_params.pipe_ctx;
+	struct dpp *dpp = pipe_ctx->plane_res.dpp;
+	struct dc_plane_state *plane_state = pipe_ctx->plane_state;
+	struct dc_bias_and_scale bns_params = {0};
+
+	//TODO :for CNVC set scale and bias registers if necessary
+	build_prescale_params(&bns_params, plane_state);
+	if (dpp->funcs->dpp_program_bias_and_scale)
+		dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params);
+}
+
 void get_mclk_switch_visual_confirm_color(
 		struct dc *dc,
 		struct dc_state *context,
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7fd9f5a9e1915..7ded574f84ff1 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -896,6 +896,7 @@ struct dc_debug_options {
 	bool disable_dp_plus_plus_wa;
 	uint32_t fpo_vactive_min_active_margin_us;
 	uint32_t fpo_vactive_max_blank_us;
+	bool enable_legacy_fast_update;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 54805802cbd5a..42e9b6a529f6f 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -401,6 +401,10 @@ static const struct dc_plane_cap plane_cap = {
 	}
 };
 
+static const struct dc_debug_options debug_defaults = {
+		.enable_legacy_fast_update = true,
+};
+
 #define CTX  ctx
 #define REG(reg) mm ## reg
 
@@ -1071,6 +1075,7 @@ static bool dce100_resource_construct(
 	dc->caps.dual_link_dvi = true;
 	dc->caps.disable_dp_clk_share = true;
 	dc->caps.extended_aux_timeout_support = false;
+	dc->debug = debug_defaults;
 
 	for (i = 0; i < pool->base.pipe_count; i++) {
 		pool->base.timing_generators[i] =
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
index a4a45a6ce61e4..46eca5a21e1ce 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
@@ -424,6 +424,10 @@ static const struct dc_plane_cap plane_cap = {
 		64
 };
 
+static const struct dc_debug_options debug_defaults = {
+		.enable_legacy_fast_update = true,
+};
+
 static const struct dc_plane_cap underlay_plane_cap = {
 		.type = DC_PLANE_TYPE_DCE_UNDERLAY,
 		.per_pixel_alpha = 1,
@@ -1368,6 +1372,7 @@ static bool dce110_resource_construct(
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.is_apu = true;
 	dc->caps.extended_aux_timeout_support = false;
+	dc->debug = debug_defaults;
 
 	/*************************************************
 	 *  Create resources                             *
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
index e179e80667d1c..8088558861836 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -429,6 +429,10 @@ static const struct dc_plane_cap plane_cap = {
 	64
 };
 
+static const struct dc_debug_options debug_defaults = {
+		.enable_legacy_fast_update = true,
+};
+
 #define CTX  ctx
 #define REG(reg) mm ## reg
 
@@ -1239,6 +1243,7 @@ static bool dce112_resource_construct(
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dual_link_dvi = true;
 	dc->caps.extended_aux_timeout_support = false;
+	dc->debug = debug_defaults;
 
 	/*************************************************
 	 *  Create resources                             *
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
index af631085e88c5..18c5a86d2d614 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
@@ -526,6 +526,7 @@ static const struct dc_plane_cap plane_cap = {
 
 static const struct dc_debug_options debug_defaults = {
 		.disable_clock_gate = true,
+		.enable_legacy_fast_update = true,
 };
 
 static struct clock_source *dce120_clock_source_create(
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
index 5825e6f412bd7..3935fd455f0f9 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
@@ -418,6 +418,10 @@ static const struct dc_plane_cap plane_cap = {
 	}
 };
 
+static const struct dc_debug_options debug_defaults = {
+		.enable_legacy_fast_update = true,
+};
+
 static const struct dce_dmcu_registers dmcu_regs = {
 		DMCU_DCE80_REG_LIST()
 };
@@ -969,6 +973,7 @@ static bool dce80_construct(
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.dual_link_dvi = true;
 	dc->caps.extended_aux_timeout_support = false;
+	dc->debug = debug_defaults;
 
 	/*************************************************
 	 *  Create resources                             *
@@ -1369,6 +1374,7 @@ static bool dce83_construct(
 	dc->caps.max_cursor_size = 128;
 	dc->caps.min_horizontal_blanking_period = 80;
 	dc->caps.is_apu = true;
+	dc->debug = debug_defaults;
 
 	/*************************************************
 	 *  Create resources                             *
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index 26ddf73fd5b11..4b02f8443534e 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -553,6 +553,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.recovery_enabled = false, /*enable this by default after testing.*/
 		.max_downscale_src_width = 3840,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
+		.enable_legacy_fast_update = true,
 };
 
 static const struct dc_debug_options debug_defaults_diags = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
index 6ef7e26349918..4cc8de2627ce9 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c
@@ -722,6 +722,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.scl_reset_length10 = true,
 		.sanity_checks = false,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
+		.enable_legacy_fast_update = true,
 };
 
 void dcn20_dpp_destroy(struct dpp **dpp)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
index 5ff09bf4bc308..fdba8a9f5c301 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c
@@ -613,6 +613,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.sanity_checks = false,
 		.underflow_assert_delay_us = 0xFFFFFFFF,
 		.enable_tri_buf = false,
+		.enable_legacy_fast_update = true,
 };
 
 static void dcn201_dpp_destroy(struct dpp **dpp)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
index da6d42de0554a..d693ea42d033b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c
@@ -653,6 +653,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.usbc_combo_phy_reset_wa = true,
 		.dmub_command_table = true,
 		.use_max_lb = true,
+		.enable_legacy_fast_update = true,
 };
 
 static const struct dc_panel_config panel_config_defaults = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
index f2d589c505a80..fc33b5fcabe10 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c
@@ -887,6 +887,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 		}
 	},
 	.disable_z10 = true,
+	.enable_legacy_fast_update = true,
 	.enable_z9_disable_interface = true, /* Allow support for the PMFW interface for disable Z9*/
 	.dml_hostvm_override = DML_HOSTVM_OVERRIDE_FALSE,
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index 8570bdc292b4f..0cc853964781c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -887,6 +887,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 			.afmt = true,
 		}
 	},
+	.enable_legacy_fast_update = true,
 	.psr_power_use_phy_fsm = 0,
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
index b6e0aa2ab27a0..707cf28bbceb1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c
@@ -884,6 +884,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 			.afmt = true,
 		}
 	},
+	.enable_legacy_fast_update = true,
 };
 
 static const struct dc_panel_config panel_config_defaults = {
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 7f5cd8c8d49b7..3f11992e380b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -409,6 +409,30 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc,
 	}
 }
 
+void dcn32_subvp_pipe_control_lock_fast(union block_sequence_params *params)
+{
+	struct dc *dc = params->subvp_pipe_control_lock_fast_params.dc;
+	bool lock = params->subvp_pipe_control_lock_fast_params.lock;
+	struct pipe_ctx *pipe_ctx = params->subvp_pipe_control_lock_fast_params.pipe_ctx;
+	bool subvp_immediate_flip = false;
+
+	if (pipe_ctx && pipe_ctx->stream && pipe_ctx->plane_state) {
+		if (pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN &&
+				pipe_ctx->plane_state->flip_immediate)
+			subvp_immediate_flip = true;
+	}
+
+	// Don't need to lock for DRR VSYNC flips -- FW will wait for DRR pending update cleared.
+	if (subvp_immediate_flip) {
+		union dmub_inbox0_cmd_lock_hw hw_lock_cmd = { 0 };
+
+		hw_lock_cmd.bits.command_code = DMUB_INBOX0_CMD__HW_LOCK;
+		hw_lock_cmd.bits.hw_lock_client = HW_LOCK_CLIENT_DRIVER;
+		hw_lock_cmd.bits.lock = lock;
+		hw_lock_cmd.bits.should_release = !lock;
+		dmub_hw_lock_mgr_inbox0_cmd(dc->ctx->dmub_srv, hw_lock_cmd);
+	}
+}
 
 bool dcn32_set_mpc_shaper_3dlut(
 	struct pipe_ctx *pipe_ctx, const struct dc_stream_state *stream)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
index 6dbe929cf599a..2d2628f31bed7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
@@ -84,6 +84,8 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc,
 		struct pipe_ctx *top_pipe_to_program,
 		bool subvp_prev_use);
 
+void dcn32_subvp_pipe_control_lock_fast(union block_sequence_params *params);
+
 void dcn32_unblank_stream(struct pipe_ctx *pipe_ctx,
 		struct dc_link_settings *link_settings);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
index 8356b31e1d9af..c2490e16a66a6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c
@@ -110,6 +110,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = {
 	.enable_phantom_streams = dcn32_enable_phantom_streams,
 	.subvp_pipe_control_lock = dcn32_subvp_pipe_control_lock,
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
+	.subvp_pipe_control_lock_fast = dcn32_subvp_pipe_control_lock_fast,
 	.update_phantom_vp_position = dcn32_update_phantom_vp_position,
 	.update_dsc_pg = dcn32_update_dsc_pg,
 	.apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index d8dd143cf6ea8..034610b74a37e 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -37,6 +37,7 @@
 #include "dwb.h"
 #include "mcif_wb.h"
 #include "panel_cntl.h"
+#include "dmub/inc/dmub_cmd.h"
 
 #define MAX_CLOCK_SOURCES 7
 #define MAX_SVP_PHANTOM_STREAMS 2
@@ -499,6 +500,11 @@ struct bw_context {
 	struct display_mode_lib dml;
 };
 
+struct dc_dmub_cmd {
+	union dmub_rb_cmd dmub_cmd;
+	enum dm_dmub_wait_type wait_type;
+};
+
 /**
  * struct dc_state - The full description of a state requested by users
  */
@@ -547,6 +553,11 @@ struct dc_state {
 	 */
 	struct bw_context bw_ctx;
 
+	struct block_sequence block_sequence[50];
+	unsigned int block_sequence_steps;
+	struct dc_dmub_cmd dc_dmub_cmd[10];
+	unsigned int dmub_cmd_count;
+
 	/**
 	 * @refcount: refcount reference
 	 *
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index df160c6a630c0..cc0a3a992f7bb 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -44,6 +44,112 @@ struct dc_virtual_addr_space_config;
 struct dpp;
 struct dce_hwseq;
 struct link_resource;
+struct dc_dmub_cmd;
+
+struct subvp_pipe_control_lock_fast_params {
+	struct dc *dc;
+	bool lock;
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct pipe_control_lock_params {
+	struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+	bool lock;
+};
+
+struct set_flip_control_gsl_params {
+	struct pipe_ctx *pipe_ctx;
+	bool flip_immediate;
+};
+
+struct program_triplebuffer_params {
+	const struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+	bool enableTripleBuffer;
+};
+
+struct update_plane_addr_params {
+	struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct set_input_transfer_func_params {
+	struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+	struct dc_plane_state *plane_state;
+};
+
+struct program_gamut_remap_params {
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct program_manual_trigger_params {
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct send_dmcub_cmd_params {
+	struct dc_context *ctx;
+	union dmub_rb_cmd *cmd;
+	enum dm_dmub_wait_type wait_type;
+};
+
+struct setup_dpp_params {
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct program_bias_and_scale_params {
+	struct pipe_ctx *pipe_ctx;
+};
+
+struct set_output_transfer_func_params {
+	struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+	const struct dc_stream_state *stream;
+};
+
+struct update_visual_confirm_params {
+	struct dc *dc;
+	struct pipe_ctx *pipe_ctx;
+	int mpcc_id;
+};
+
+union block_sequence_params {
+	struct update_plane_addr_params update_plane_addr_params;
+	struct subvp_pipe_control_lock_fast_params subvp_pipe_control_lock_fast_params;
+	struct pipe_control_lock_params pipe_control_lock_params;
+	struct set_flip_control_gsl_params set_flip_control_gsl_params;
+	struct program_triplebuffer_params program_triplebuffer_params;
+	struct set_input_transfer_func_params set_input_transfer_func_params;
+	struct program_gamut_remap_params program_gamut_remap_params;
+	struct program_manual_trigger_params program_manual_trigger_params;
+	struct send_dmcub_cmd_params send_dmcub_cmd_params;
+	struct setup_dpp_params setup_dpp_params;
+	struct program_bias_and_scale_params program_bias_and_scale_params;
+	struct set_output_transfer_func_params set_output_transfer_func_params;
+	struct update_visual_confirm_params update_visual_confirm_params;
+};
+
+enum block_sequence_func {
+	DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST = 0,
+	OPTC_PIPE_CONTROL_LOCK,
+	HUBP_SET_FLIP_CONTROL_GSL,
+	HUBP_PROGRAM_TRIPLEBUFFER,
+	HUBP_UPDATE_PLANE_ADDR,
+	DPP_SET_INPUT_TRANSFER_FUNC,
+	DPP_PROGRAM_GAMUT_REMAP,
+	OPTC_PROGRAM_MANUAL_TRIGGER,
+	DMUB_SEND_DMCUB_CMD,
+	DPP_SETUP_DPP,
+	DPP_PROGRAM_BIAS_AND_SCALE,
+	DPP_SET_OUTPUT_TRANSFER_FUNC,
+	MPC_UPDATE_VISUAL_CONFIRM,
+};
+
+struct block_sequence {
+	union block_sequence_params params;
+	enum block_sequence_func func;
+};
 
 struct hw_sequencer_funcs {
 	void (*hardware_release)(struct dc *dc);
@@ -252,6 +358,7 @@ struct hw_sequencer_funcs {
 			const struct tg_color *solid_color,
 			int width, int height, int offset);
 
+	void (*subvp_pipe_control_lock_fast)(union block_sequence_params *params);
 	void (*z10_restore)(const struct dc *dc);
 	void (*z10_save_init)(struct dc *dc);
 
@@ -313,4 +420,23 @@ void get_mclk_switch_visual_confirm_color(
 		struct pipe_ctx *pipe_ctx,
 		struct tg_color *color);
 
+void hwss_execute_sequence(struct dc *dc,
+		struct block_sequence block_sequence[],
+		int num_steps);
+
+void hwss_build_fast_sequence(struct dc *dc,
+		struct dc_dmub_cmd *dc_dmub_cmd,
+		unsigned int dmub_cmd_count,
+		struct block_sequence block_sequence[],
+		int *num_steps,
+		struct pipe_ctx *pipe_ctx);
+
+void hwss_send_dmcub_cmd(union block_sequence_params *params);
+
+void hwss_program_manual_trigger(union block_sequence_params *params);
+
+void hwss_setup_dpp(union block_sequence_params *params);
+
+void hwss_program_bias_and_scale(union block_sequence_params *params);
+
 #endif /* __DC_HW_SEQUENCER_H__ */
-- 
GitLab


From 490ddccb84fe2f6165b2bdd2d00fd4ab593b95ec Mon Sep 17 00:00:00 2001
From: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Date: Thu, 18 May 2023 12:12:20 -0400
Subject: [PATCH 1581/1791] drm/amd/display: Wrong index type for pipe iterator

[Why and How]
Type mismatch in index and pipe count might cause an infinite loop. code
Change should resolve this issue.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Reviewed-by: Josip Pavic <Josip.Pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index 46b6f4f9e1fd1..ce7e6f20b31fc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -392,7 +392,7 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 
 void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context)
 {
-	uint8_t i;
+	unsigned int i;
 	struct pipe_ctx *pipe = NULL;
 	bool otg_disabled[MAX_PIPES] = {false};
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 3f11992e380b0..00f32ffe00796 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -1201,7 +1201,7 @@ void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 
 void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context)
 {
-	uint8_t i;
+	unsigned int i;
 	struct pipe_ctx *pipe = NULL;
 	bool otg_disabled[MAX_PIPES] = {false};
 
-- 
GitLab


From 8e7b3f5435b3c0751515c973972ebb11e0fc0fb5 Mon Sep 17 00:00:00 2001
From: Max Tseng <max.tseng@amd.com>
Date: Tue, 25 Apr 2023 15:05:17 +0800
Subject: [PATCH 1582/1791] drm/amd/display: Add control flag to
 dc_stream_state to skip eDP BL off/link off

Add control flag to dc_stream_state to skip eDP BL off/link off.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Max Tseng <max.tseng@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_stream.h                 | 1 +
 .../gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c    | 3 ++-
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c            | 7 ++++---
 3 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index d5b3e3a32cc6d..3697ea1d14c1b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -302,6 +302,7 @@ struct dc_stream_state {
 	bool vblank_synchronized;
 	bool fpo_in_use;
 	struct mall_stream_config mall_stream_config;
+	bool skip_edp_power_down;
 };
 
 #define ABM_LEVEL_IMMEDIATE_DISABLE 255
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index d4cacb8df6310..6c9ca43d1040b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -1216,7 +1216,8 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx)
 	struct dce_hwseq *hws = link->dc->hwseq;
 
 	if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
-		hws->funcs.edp_backlight_control(link, false);
+		if (!stream->skip_edp_power_down)
+			hws->funcs.edp_backlight_control(link, false);
 		link->dc->hwss.set_abm_immediate_disable(pipe_ctx);
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index f7f1a1586f3b2..1a7b93e41e352 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2207,9 +2207,8 @@ static enum dc_status enable_link(
 	 * link settings. Need to call disable first before enabling at
 	 * new link settings.
 	 */
-	if (link->link_status.link_active) {
+	if (link->link_status.link_active && !stream->skip_edp_power_down)
 		disable_link(link, &pipe_ctx->link_res, pipe_ctx->stream->signal);
-	}
 
 	switch (pipe_ctx->stream->signal) {
 	case SIGNAL_TYPE_DISPLAY_PORT:
@@ -2327,7 +2326,9 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 		dc->hwss.disable_stream(pipe_ctx);
 	} else {
 		dc->hwss.disable_stream(pipe_ctx);
-		disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal);
+		if (!pipe_ctx->stream->skip_edp_power_down) {
+			disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal);
+		}
 	}
 
 	if (pipe_ctx->stream->timing.flags.DSC) {
-- 
GitLab


From bbd069a860b78a087d20d91656a5026c0196586b Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Fri, 19 May 2023 11:38:15 -0400
Subject: [PATCH 1583/1791] drm/amd/display: Reduce sdp bw after urgent to 90%

[Description]
Reduce expected SDP bandwidth due to poor QoS and
arbitration issues on high bandwidth configs

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Nevenko Stupar <Nevenko.Stupar@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 7e03c844b05e0..ad6ee48580f8e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -147,7 +147,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = {
 	.urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096,
 	.urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096,
 	.urgent_out_of_order_return_per_channel_vm_only_bytes = 4096,
-	.pct_ideal_sdp_bw_after_urgent = 100.0,
+	.pct_ideal_sdp_bw_after_urgent = 90.0,
 	.pct_ideal_fabric_bw_after_urgent = 67.0,
 	.pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 20.0,
 	.pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, // N/A, for now keep as is until DML implemented
-- 
GitLab


From 35c7b59e3691cbea91c8b91e8ec4b0a4a960dd5a Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 6 Jun 2023 17:36:45 +0530
Subject: [PATCH 1584/1791] =?UTF-8?q?drm/amd/display:=20Fix=20unused=20var?=
 =?UTF-8?q?iable=20=E2=80=98should=5Flock=5Fall=5Fpipes=E2=80=99?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix below compilation error:

drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:3524:7: error: unused variable 'should_lock_all_pipes' [-Werror,-Wunused-variable]
        bool should_lock_all_pipes = (update_type != UPDATE_TYPE_FAST);

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 60da6d834518c..be72e03cdf92e 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3521,7 +3521,6 @@ static void commit_planes_for_stream_fast(struct dc *dc,
 {
 	int i, j;
 	struct pipe_ctx *top_pipe_to_program = NULL;
-	bool should_lock_all_pipes = (update_type != UPDATE_TYPE_FAST);
 	dc_z10_restore(dc);
 
 	for (j = 0; j < dc->res_pool->pipe_count; j++) {
-- 
GitLab


From caf0f98dc280b5426b1858d4e70130aa3dd9679c Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Mon, 5 Jun 2023 13:59:08 -0400
Subject: [PATCH 1585/1791] drm/amd/display: mark dml31's UseMinimumDCFCLK() as
 noinline_for_stack

clang reports:
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c:3797:6: error: stack frame size (2632) exceeds limit (2048) in 'dml31_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than]
 3797 | void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib)
      |      ^
1 error generated.

So, since UseMinimumDCFCLK() consumes a lot of stack space, mark it as
noinline_for_stack to prevent it from blowing up
dml31_ModeSupportAndSystemConfigurationFull()'s stack size.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index 01603abd75bba..43016c462251f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -7032,7 +7032,7 @@ static double CalculateUrgentLatency(
 	return ret;
 }
 
-static void UseMinimumDCFCLK(
+static noinline_for_stack void UseMinimumDCFCLK(
 		struct display_mode_lib *mode_lib,
 		int MaxPrefetchMode,
 		int ReorderingBytes)
-- 
GitLab


From 82054942472745c2caa4f6b31b4174401348b887 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Mon, 5 Jun 2023 14:18:49 -0400
Subject: [PATCH 1586/1791] drm/amd/display: mark dml314's UseMinimumDCFCLK()
 as noinline_for_stack

clang reports:
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn314/display_mode_vba_314.c:3892:6: error: stack frame size (2632) exceeds limit (2048) in 'dml314_ModeSupportAndSystemConfigurationFull' [-Werror,-Wframe-larger-than]
 3892 | void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib)
      |      ^
1 error generated.

So, since UseMinimumDCFCLK() consumes a lot of stack space, mark it as
noinline_for_stack to prevent it from blowing up
dml314_ModeSupportAndSystemConfigurationFull()'s stack size.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 27b83162ae452..1532a7e0ed6c5 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -7061,7 +7061,7 @@ static double CalculateUrgentLatency(
 	return ret;
 }
 
-static void UseMinimumDCFCLK(
+static noinline_for_stack void UseMinimumDCFCLK(
 		struct display_mode_lib *mode_lib,
 		int MaxPrefetchMode,
 		int ReorderingBytes)
-- 
GitLab


From d155cfff48499d1e973976519ca81a7d9bab2cc3 Mon Sep 17 00:00:00 2001
From: Sui Jingfeng <suijingfeng@loongson.cn>
Date: Tue, 6 Jun 2023 21:33:28 +0800
Subject: [PATCH 1587/1791] drm/amdgpu: display/Kconfig: replace leading spaces
 with tab

This patch replace the leading spaces with tab, make them keep aligned with
the rest of the config options. No functional change.

Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/Kconfig | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig
index b015c7d19531d..4e99b8836827d 100644
--- a/drivers/gpu/drm/amd/display/Kconfig
+++ b/drivers/gpu/drm/amd/display/Kconfig
@@ -42,16 +42,13 @@ config DEBUG_KERNEL_DC
 	  Choose this option if you want to hit kdgb_break in assert.
 
 config DRM_AMD_SECURE_DISPLAY
-        bool "Enable secure display support"
-        depends on DEBUG_FS
-        depends on DRM_AMD_DC_FP
-        help
-            Choose this option if you want to
-            support secure display
-
-            This option enables the calculation
-            of crc of specific region via debugfs.
-            Cooperate with specific DMCU FW.
+	bool "Enable secure display support"
+	depends on DEBUG_FS
+	depends on DRM_AMD_DC_FP
+	help
+	  Choose this option if you want to support secure display
 
+	  This option enables the calculation of crc of specific region via
+	  debugfs. Cooperate with specific DMCU FW.
 
 endmenu
-- 
GitLab


From b7588507152148eaf0f19feb98c65b72ab40a726 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 6 Jun 2023 11:33:46 +0300
Subject: [PATCH 1588/1791] drm/amd/pm: Fix memory some memory corruption

The "od_table" is a pointer to a large struct, but this code is doing
pointer math as if it were pointing to bytes.  It results in writing
far outside the struct.

Fixes: 2e8452ea4ef6 ("drm/amd/pm: fulfill the OD support for SMU13.0.0")
Fixes: 2a9aa52e4617 ("drm/amd/pm: fulfill the OD support for SMU13.0.7")
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 4 ++--
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index 5ac5ea770c1c0..413e592f0ed61 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -1535,7 +1535,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu,
 		 * settings. Thus we do not cache it.
 		 */
 		offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask);
-		if (memcmp(od_table + offset_of_featurectrlmask,
+		if (memcmp((u8 *)od_table + offset_of_featurectrlmask,
 			   table_context->user_overdrive_table + offset_of_featurectrlmask,
 			   sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) {
 			smu_v13_0_0_dump_od_table(smu, od_table);
@@ -1548,7 +1548,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu,
 
 			od_table->OverDriveTable.FeatureCtrlMask = 0;
 			memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask,
-			       od_table + offset_of_featurectrlmask,
+			       (u8 *)od_table + offset_of_featurectrlmask,
 			       sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask);
 
 			if (!memcmp(table_context->user_overdrive_table,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 0bd086360efae..cda4e818aab7e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -1524,7 +1524,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu,
 		 * settings. Thus we do not cache it.
 		 */
 		offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask);
-		if (memcmp(od_table + offset_of_featurectrlmask,
+		if (memcmp((u8 *)od_table + offset_of_featurectrlmask,
 			   table_context->user_overdrive_table + offset_of_featurectrlmask,
 			   sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) {
 			smu_v13_0_7_dump_od_table(smu, od_table);
@@ -1537,7 +1537,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu,
 
 			od_table->OverDriveTable.FeatureCtrlMask = 0;
 			memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask,
-			       od_table + offset_of_featurectrlmask,
+			       (u8 *)od_table + offset_of_featurectrlmask,
 			       sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask);
 
 			if (!memcmp(table_context->user_overdrive_table,
-- 
GitLab


From 33e82119cfb2a957f250f92a1e4c4db2b06400db Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Tue, 6 Jun 2023 11:42:53 -0400
Subject: [PATCH 1589/1791] drm/amd/display: Only use ODM2:1 policy for high
 pixel rate displays

We only gain a benefit of using the ODM2:1 dynamic policy if it allow us
to decrease DISPCLK to use the VMIN freq.  If the display config can
already achieve VMIN DISPCLK freq without ODM2:1, don't apply the
policy.

This patch was reverted but that causes some IGT regressions. To
unblock, the patch is being applied again until IGT failures are
fixed.

Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c | 1 +
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 8d68f02f5147b..2e6b39fe26132 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -1918,6 +1918,7 @@ int dcn32_populate_dml_pipes_from_context(
 				context->stream_status[0].plane_count == 1 &&
 				!dc_is_hdmi_signal(res_ctx->pipe_ctx[i].stream->signal) &&
 				is_h_timing_divisible_by_2(res_ctx->pipe_ctx[i].stream) &&
+				pipe->stream->timing.pix_clk_100hz * 100 > DCN3_2_VMIN_DISPCLK_HZ &&
 				dc->debug.enable_single_display_2to1_odm_policy &&
 				!vsr_odm_support) { //excluding 2to1 ODM combine on >= 5k vsr
 			pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 80bebdf43eca1..2f34f01b3ea13 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -40,6 +40,7 @@
 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq
 #define SUBVP_HIGH_REFRESH_LIST_LEN 3
 #define DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ 1800
+#define DCN3_2_VMIN_DISPCLK_HZ 717000000
 
 #define TO_DCN32_RES_POOL(pool)\
 	container_of(pool, struct dcn32_resource_pool, base)
-- 
GitLab


From 8be295046748432c53a2dee39c469f63c60b0ec3 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 6 Jun 2023 19:29:51 -0400
Subject: [PATCH 1590/1791] drm/amdkfd: potential error pointer dereference in
 ioctl

The "target" either comes from kfd_create_process() which returns error
pointers on error or kfd_lookup_process_by_pid() which returns NULL on
error.  So we need to check for both types of errors.

Fixes: 0ab2d7532b05 ("drm/amdkfd: prepare per-process debug enable and disable")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Jonathan Kim <jonathan.kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index cce2abe12e1b8..d655c5bc951fd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2920,9 +2920,9 @@ static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, v
 		target = kfd_lookup_process_by_pid(pid);
 	}
 
-	if (!target) {
+	if (IS_ERR_OR_NULL(target)) {
 		pr_debug("Cannot find process PID %i to debug\n", args->pid);
-		r = -ESRCH;
+		r = target ? PTR_ERR(target) : -ESRCH;
 		goto out;
 	}
 
-- 
GitLab


From f2bcc0c7db0c004f0184675e7862648e8aa197f9 Mon Sep 17 00:00:00 2001
From: Emily Deng <Emily.Deng@amd.com>
Date: Tue, 6 Jun 2023 14:27:04 +0800
Subject: [PATCH 1591/1791] drm/amdgpu/mmsch: Correct the definition for mmsch
 init header

For the header, it is version related, shouldn't use MAX_VCN_INSTANCES.

Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h | 4 +++-
 drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h | 4 +++-
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c   | 2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c   | 2 +-
 4 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h
index 3e4e858a69652..a773ef61b78c5 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h
@@ -30,6 +30,8 @@
 #define MMSCH_VERSION_MINOR	0
 #define MMSCH_VERSION	(MMSCH_VERSION_MAJOR << 16 | MMSCH_VERSION_MINOR)
 
+#define MMSCH_V3_0_VCN_INSTANCES 0x2
+
 enum mmsch_v3_0_command_type {
 	MMSCH_COMMAND__DIRECT_REG_WRITE = 0,
 	MMSCH_COMMAND__DIRECT_REG_POLLING = 2,
@@ -47,7 +49,7 @@ struct mmsch_v3_0_table_info {
 struct mmsch_v3_0_init_header {
 	uint32_t version;
 	uint32_t total_size;
-	struct mmsch_v3_0_table_info inst[AMDGPU_MAX_VCN_INSTANCES];
+	struct mmsch_v3_0_table_info inst[MMSCH_V3_0_VCN_INSTANCES];
 };
 
 struct mmsch_v3_0_cmd_direct_reg_header {
diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h
index 83653a50a1a21..796d4f8791e5f 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h
@@ -43,6 +43,8 @@
 #define MMSCH_VF_MAILBOX_RESP__OK 0x1
 #define MMSCH_VF_MAILBOX_RESP__INCOMPLETE 0x2
 
+#define MMSCH_V4_0_VCN_INSTANCES 0x2
+
 enum mmsch_v4_0_command_type {
 	MMSCH_COMMAND__DIRECT_REG_WRITE = 0,
 	MMSCH_COMMAND__DIRECT_REG_POLLING = 2,
@@ -60,7 +62,7 @@ struct mmsch_v4_0_table_info {
 struct mmsch_v4_0_init_header {
 	uint32_t version;
 	uint32_t total_size;
-	struct mmsch_v4_0_table_info inst[AMDGPU_MAX_VCN_INSTANCES];
+	struct mmsch_v4_0_table_info inst[MMSCH_V4_0_VCN_INSTANCES];
 	struct mmsch_v4_0_table_info jpegdec;
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 70fefbf26c48a..c8f63b3c6f69d 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -1313,7 +1313,7 @@ static int vcn_v3_0_start_sriov(struct amdgpu_device *adev)
 
 	header.version = MMSCH_VERSION;
 	header.total_size = sizeof(struct mmsch_v3_0_init_header) >> 2;
-	for (i = 0; i < AMDGPU_MAX_VCN_INSTANCES; i++) {
+	for (i = 0; i < MMSCH_V3_0_VCN_INSTANCES; i++) {
 		header.inst[i].init_status = 0;
 		header.inst[i].table_offset = 0;
 		header.inst[i].table_size = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 60c3fd20e8ced..8d371faaa2b3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1239,7 +1239,7 @@ static int vcn_v4_0_start_sriov(struct amdgpu_device *adev)
 
 	header.version = MMSCH_VERSION;
 	header.total_size = sizeof(struct mmsch_v4_0_init_header) >> 2;
-	for (i = 0; i < AMDGPU_MAX_VCN_INSTANCES; i++) {
+	for (i = 0; i < MMSCH_V4_0_VCN_INSTANCES; i++) {
 		header.inst[i].init_status = 0;
 		header.inst[i].table_offset = 0;
 		header.inst[i].table_size = 0;
-- 
GitLab


From 8ff865be93e642d0ad66ca7369f42fbe36dc6a90 Mon Sep 17 00:00:00 2001
From: Jiadong Zhu <Jiadong.Zhu@amd.com>
Date: Thu, 25 May 2023 16:52:55 +0800
Subject: [PATCH 1592/1791] drm/amdgpu: Modify indirect buffer packages for
 resubmission

When the preempted IB frame resubmitted to cp, we need to modify the frame
data including:
1. set PRE_RESUME 1 in CONTEXT_CONTROL.
2. use meta data(DE and CE) read from CSA in WRITE_DATA.

Add functions to save the location the first time IBs emitted and callback
to patch the package when resubmission happens.

Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c     | 18 ++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h     |  9 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c | 60 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h | 15 +++++
 4 files changed, 102 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index d8a78de1f3351..da26c555af24b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -670,3 +670,21 @@ void amdgpu_ring_ib_end(struct amdgpu_ring *ring)
 	if (ring->is_sw_ring)
 		amdgpu_sw_ring_ib_end(ring);
 }
+
+void amdgpu_ring_ib_on_emit_cntl(struct amdgpu_ring *ring)
+{
+	if (ring->is_sw_ring)
+		amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_CONTROL);
+}
+
+void amdgpu_ring_ib_on_emit_ce(struct amdgpu_ring *ring)
+{
+	if (ring->is_sw_ring)
+		amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_CE);
+}
+
+void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring)
+{
+	if (ring->is_sw_ring)
+		amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 4a4c9f89c302a..21ffb9ce32ce2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -229,6 +229,9 @@ struct amdgpu_ring_funcs {
 	int (*preempt_ib)(struct amdgpu_ring *ring);
 	void (*emit_mem_sync)(struct amdgpu_ring *ring);
 	void (*emit_wave_limit)(struct amdgpu_ring *ring, bool enable);
+	void (*patch_cntl)(struct amdgpu_ring *ring, unsigned offset);
+	void (*patch_ce)(struct amdgpu_ring *ring, unsigned offset);
+	void (*patch_de)(struct amdgpu_ring *ring, unsigned offset);
 };
 
 struct amdgpu_ring {
@@ -323,11 +326,17 @@ struct amdgpu_ring {
 #define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r))
 #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))
 #define amdgpu_ring_preempt_ib(r) (r)->funcs->preempt_ib(r)
+#define amdgpu_ring_patch_cntl(r, o) ((r)->funcs->patch_cntl((r), (o)))
+#define amdgpu_ring_patch_ce(r, o) ((r)->funcs->patch_ce((r), (o)))
+#define amdgpu_ring_patch_de(r, o) ((r)->funcs->patch_de((r), (o)))
 
 unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type);
 int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw);
 void amdgpu_ring_ib_begin(struct amdgpu_ring *ring);
 void amdgpu_ring_ib_end(struct amdgpu_ring *ring);
+void amdgpu_ring_ib_on_emit_cntl(struct amdgpu_ring *ring);
+void amdgpu_ring_ib_on_emit_ce(struct amdgpu_ring *ring);
+void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring);
 
 void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count);
 void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c
index 62079f0e3ee8f..73516abef662f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c
@@ -105,6 +105,16 @@ static void amdgpu_mux_resubmit_chunks(struct amdgpu_ring_mux *mux)
 				amdgpu_fence_update_start_timestamp(e->ring,
 								    chunk->sync_seq,
 								    ktime_get());
+				if (chunk->sync_seq ==
+					le32_to_cpu(*(e->ring->fence_drv.cpu_addr + 2))) {
+					if (chunk->cntl_offset <= e->ring->buf_mask)
+						amdgpu_ring_patch_cntl(e->ring,
+								       chunk->cntl_offset);
+					if (chunk->ce_offset <= e->ring->buf_mask)
+						amdgpu_ring_patch_ce(e->ring, chunk->ce_offset);
+					if (chunk->de_offset <= e->ring->buf_mask)
+						amdgpu_ring_patch_de(e->ring, chunk->de_offset);
+				}
 				amdgpu_ring_mux_copy_pkt_from_sw_ring(mux, e->ring,
 								      chunk->start,
 								      chunk->end);
@@ -407,6 +417,17 @@ void amdgpu_sw_ring_ib_end(struct amdgpu_ring *ring)
 	amdgpu_ring_mux_end_ib(mux, ring);
 }
 
+void amdgpu_sw_ring_ib_mark_offset(struct amdgpu_ring *ring, enum amdgpu_ring_mux_offset_type type)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_ring_mux *mux = &adev->gfx.muxer;
+	unsigned offset;
+
+	offset = ring->wptr & ring->buf_mask;
+
+	amdgpu_ring_mux_ib_mark_offset(mux, ring, offset, type);
+}
+
 void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring)
 {
 	struct amdgpu_mux_entry *e;
@@ -429,6 +450,10 @@ void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *r
 	}
 
 	chunk->start = ring->wptr;
+	/* the initialized value used to check if they are set by the ib submission*/
+	chunk->cntl_offset = ring->buf_mask + 1;
+	chunk->de_offset = ring->buf_mask + 1;
+	chunk->ce_offset = ring->buf_mask + 1;
 	list_add_tail(&chunk->entry, &e->list);
 }
 
@@ -454,6 +479,41 @@ static void scan_and_remove_signaled_chunk(struct amdgpu_ring_mux *mux, struct a
 	}
 }
 
+void amdgpu_ring_mux_ib_mark_offset(struct amdgpu_ring_mux *mux,
+				    struct amdgpu_ring *ring, u64 offset,
+				    enum amdgpu_ring_mux_offset_type type)
+{
+	struct amdgpu_mux_entry *e;
+	struct amdgpu_mux_chunk *chunk;
+
+	e = amdgpu_ring_mux_sw_entry(mux, ring);
+	if (!e) {
+		DRM_ERROR("cannot find entry!\n");
+		return;
+	}
+
+	chunk = list_last_entry(&e->list, struct amdgpu_mux_chunk, entry);
+	if (!chunk) {
+		DRM_ERROR("cannot find chunk!\n");
+		return;
+	}
+
+	switch (type) {
+	case AMDGPU_MUX_OFFSET_TYPE_CONTROL:
+		chunk->cntl_offset = offset;
+		break;
+	case AMDGPU_MUX_OFFSET_TYPE_DE:
+		chunk->de_offset = offset;
+		break;
+	case AMDGPU_MUX_OFFSET_TYPE_CE:
+		chunk->ce_offset = offset;
+		break;
+	default:
+		DRM_ERROR("invalid type (%d)\n", type);
+		break;
+	}
+}
+
 void amdgpu_ring_mux_end_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring)
 {
 	struct amdgpu_mux_entry *e;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h
index 4be45fc14954c..b22d4fb2a8470 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h
@@ -50,6 +50,12 @@ struct amdgpu_mux_entry {
 	struct list_head        list;
 };
 
+enum amdgpu_ring_mux_offset_type {
+	AMDGPU_MUX_OFFSET_TYPE_CONTROL,
+	AMDGPU_MUX_OFFSET_TYPE_DE,
+	AMDGPU_MUX_OFFSET_TYPE_CE,
+};
+
 struct amdgpu_ring_mux {
 	struct amdgpu_ring      *real_ring;
 
@@ -72,12 +78,18 @@ struct amdgpu_ring_mux {
  * @sync_seq: the fence seqno related with the saved IB.
  * @start:- start location on the software ring.
  * @end:- end location on the software ring.
+ * @control_offset:- the PRE_RESUME bit position used for resubmission.
+ * @de_offset:- the anchor in write_data for de meta of resubmission.
+ * @ce_offset:- the anchor in write_data for ce meta of resubmission.
  */
 struct amdgpu_mux_chunk {
 	struct list_head        entry;
 	uint32_t                sync_seq;
 	u64                     start;
 	u64                     end;
+	u64                     cntl_offset;
+	u64                     de_offset;
+	u64                     ce_offset;
 };
 
 int amdgpu_ring_mux_init(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring,
@@ -89,6 +101,8 @@ u64 amdgpu_ring_mux_get_wptr(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ri
 u64 amdgpu_ring_mux_get_rptr(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring);
 void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring);
 void amdgpu_ring_mux_end_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring);
+void amdgpu_ring_mux_ib_mark_offset(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring,
+				    u64 offset, enum amdgpu_ring_mux_offset_type type);
 bool amdgpu_mcbp_handle_trailing_fence_irq(struct amdgpu_ring_mux *mux);
 
 u64 amdgpu_sw_ring_get_rptr_gfx(struct amdgpu_ring *ring);
@@ -97,6 +111,7 @@ void amdgpu_sw_ring_set_wptr_gfx(struct amdgpu_ring *ring);
 void amdgpu_sw_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count);
 void amdgpu_sw_ring_ib_begin(struct amdgpu_ring *ring);
 void amdgpu_sw_ring_ib_end(struct amdgpu_ring *ring);
+void amdgpu_sw_ring_ib_mark_offset(struct amdgpu_ring *ring, enum amdgpu_ring_mux_offset_type type);
 const char *amdgpu_sw_ring_name(int idx);
 unsigned int amdgpu_sw_ring_priority(int idx);
 
-- 
GitLab


From ea791e704b97ab5abd563b6d2f88c4019940079e Mon Sep 17 00:00:00 2001
From: Jiadong Zhu <Jiadong.Zhu@amd.com>
Date: Thu, 25 May 2023 18:42:15 +0800
Subject: [PATCH 1593/1791] drm/amdgpu: Implement gfx9 patch functions for
 resubmission

Patch the packages including CONTEXT_CONTROL and WRITE_DATA for gfx9
during the resubmission scenario.

Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 80 +++++++++++++++++++++++++++
 1 file changed, 80 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 4073e2ee7e6dd..65577eca58f1c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -5165,9 +5165,83 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 #endif
 		lower_32_bits(ib->gpu_addr));
 	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+	amdgpu_ring_ib_on_emit_cntl(ring);
 	amdgpu_ring_write(ring, control);
 }
 
+static void gfx_v9_0_ring_patch_cntl(struct amdgpu_ring *ring,
+				     unsigned offset)
+{
+	u32 control = ring->ring[offset];
+
+	control |= INDIRECT_BUFFER_PRE_RESUME(1);
+	ring->ring[offset] = control;
+}
+
+static void gfx_v9_0_ring_patch_ce_meta(struct amdgpu_ring *ring,
+					unsigned offset)
+{
+	struct amdgpu_device *adev = ring->adev;
+	void *ce_payload_cpu_addr;
+	uint64_t payload_offset, payload_size;
+
+	payload_size = sizeof(struct v9_ce_ib_state);
+
+	if (ring->is_mes_queue) {
+		payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
+					  gfx[0].gfx_meta_data) +
+			offsetof(struct v9_gfx_meta_data, ce_payload);
+		ce_payload_cpu_addr =
+			amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
+	} else {
+		payload_offset = offsetof(struct v9_gfx_meta_data, ce_payload);
+		ce_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
+	}
+
+	if (offset + (payload_size >> 2) <= ring->buf_mask + 1) {
+		memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr, payload_size);
+	} else {
+		memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr,
+		       (ring->buf_mask + 1 - offset) << 2);
+		payload_size -= (ring->buf_mask + 1 - offset) << 2;
+		memcpy((void *)&ring->ring[0],
+		       ce_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2),
+		       payload_size);
+	}
+}
+
+static void gfx_v9_0_ring_patch_de_meta(struct amdgpu_ring *ring,
+					unsigned offset)
+{
+	struct amdgpu_device *adev = ring->adev;
+	void *de_payload_cpu_addr;
+	uint64_t payload_offset, payload_size;
+
+	payload_size = sizeof(struct v9_de_ib_state);
+
+	if (ring->is_mes_queue) {
+		payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data,
+					  gfx[0].gfx_meta_data) +
+			offsetof(struct v9_gfx_meta_data, de_payload);
+		de_payload_cpu_addr =
+			amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset);
+	} else {
+		payload_offset = offsetof(struct v9_gfx_meta_data, de_payload);
+		de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset;
+	}
+
+	if (offset + (payload_size >> 2) <= ring->buf_mask + 1) {
+		memcpy((void *)&ring->ring[offset], de_payload_cpu_addr, payload_size);
+	} else {
+		memcpy((void *)&ring->ring[offset], de_payload_cpu_addr,
+		       (ring->buf_mask + 1 - offset) << 2);
+		payload_size -= (ring->buf_mask + 1 - offset) << 2;
+		memcpy((void *)&ring->ring[0],
+		       de_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2),
+		       payload_size);
+	}
+}
+
 static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 					  struct amdgpu_job *job,
 					  struct amdgpu_ib *ib,
@@ -5363,6 +5437,8 @@ static void gfx_v9_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume)
 	amdgpu_ring_write(ring, lower_32_bits(ce_payload_gpu_addr));
 	amdgpu_ring_write(ring, upper_32_bits(ce_payload_gpu_addr));
 
+	amdgpu_ring_ib_on_emit_ce(ring);
+
 	if (resume)
 		amdgpu_ring_write_multiple(ring, ce_payload_cpu_addr,
 					   sizeof(ce_payload) >> 2);
@@ -5474,6 +5550,7 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bo
 	amdgpu_ring_write(ring, lower_32_bits(de_payload_gpu_addr));
 	amdgpu_ring_write(ring, upper_32_bits(de_payload_gpu_addr));
 
+	amdgpu_ring_ib_on_emit_de(ring);
 	if (resume)
 		amdgpu_ring_write_multiple(ring, de_payload_cpu_addr,
 					   sizeof(de_payload) >> 2);
@@ -6884,6 +6961,9 @@ static const struct amdgpu_ring_funcs gfx_v9_0_sw_ring_funcs_gfx = {
 	.emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait,
 	.soft_recovery = gfx_v9_0_ring_soft_recovery,
 	.emit_mem_sync = gfx_v9_0_emit_mem_sync,
+	.patch_cntl = gfx_v9_0_ring_patch_cntl,
+	.patch_de = gfx_v9_0_ring_patch_de_meta,
+	.patch_ce = gfx_v9_0_ring_patch_ce_meta,
 };
 
 static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
-- 
GitLab


From 71c79a196096bf51603322760dc6a95e2eb82ac1 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Tue, 30 May 2023 21:48:17 +0800
Subject: [PATCH 1594/1791] drm/amdgpu: Rename ras table version

Rename RAS_TABLE_VER to RAS_TABLE_VER_V1,
move RAS_TABLE_VER_V1 from amdgpu_ras_eeprom.c to amdgpu_ras_eeprom.h.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 5 ++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h | 2 ++
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index c2c2a7718613c..e9a848db3ff2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -68,7 +68,6 @@
 
 /* Table hdr is 'AMDR' */
 #define RAS_TABLE_HDR_VAL       0x414d4452
-#define RAS_TABLE_VER           0x00010000
 
 /* Bad GPU tag ‘BADG’ */
 #define RAS_TABLE_HDR_BAD       0x42414447
@@ -289,7 +288,7 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 	mutex_lock(&control->ras_tbl_mutex);
 
 	hdr->header = RAS_TABLE_HDR_VAL;
-	hdr->version = RAS_TABLE_VER;
+	hdr->version = RAS_TABLE_VER_V1;
 	hdr->first_rec_offset = RAS_RECORD_START;
 	hdr->tbl_size = RAS_TABLE_HEADER_SIZE;
 
@@ -575,7 +574,7 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 		control->tbl_hdr.header = RAS_TABLE_HDR_BAD;
 	}
 
-	control->tbl_hdr.version = RAS_TABLE_VER;
+	control->tbl_hdr.version = RAS_TABLE_VER_V1;
 	control->tbl_hdr.first_rec_offset = RAS_INDEX_TO_OFFSET(control, control->ras_fri);
 	control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE + control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
 	control->tbl_hdr.checksum = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
index 54d9bfe0881d9..6d21cfdeffa5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
@@ -26,6 +26,8 @@
 
 #include <linux/i2c.h>
 
+#define RAS_TABLE_VER_V1           0x00010000
+
 struct amdgpu_device;
 
 enum amdgpu_ras_eeprom_err_type {
-- 
GitLab


From 65183faec89f3ef2c781f2ed6803e6ed5c365d48 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Tue, 30 May 2023 22:38:17 +0800
Subject: [PATCH 1595/1791] drm/amdgpu: Add RAS table v2.1 macro definition

Add RAS EEPROM table version 2.1 macro definition.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h |  1 +
 2 files changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index e9a848db3ff2f..e835f68d1ebbf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -80,6 +80,15 @@
 #define RAS_MAX_RECORD_COUNT    ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE) \
 				 / RAS_TABLE_RECORD_SIZE)
 
+/* EEPROM Table V2_1 */
+#define RAS_TABLE_V2_1_INFO_SIZE       256
+#define RAS_TABLE_V2_1_INFO_START      RAS_TABLE_HEADER_SIZE
+#define RAS_RECORD_START_V2_1          (RAS_HDR_START + RAS_TABLE_HEADER_SIZE + \
+					RAS_TABLE_V2_1_INFO_SIZE)
+#define RAS_MAX_RECORD_COUNT_V2_1      ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE - \
+					RAS_TABLE_V2_1_INFO_SIZE) \
+					/ RAS_TABLE_RECORD_SIZE)
+
 /* Given a zero-based index of an EEPROM RAS record, yields the EEPROM
  * offset off of RAS_TABLE_START.  That is, this is something you can
  * add to control->i2c_address, and then tell I2C layer to read
@@ -102,6 +111,10 @@
 #define RAS_NUM_RECS(_tbl_hdr)  (((_tbl_hdr)->tbl_size - \
 				  RAS_TABLE_HEADER_SIZE) / RAS_TABLE_RECORD_SIZE)
 
+#define RAS_NUM_RECS_V2_1(_tbl_hdr)  (((_tbl_hdr)->tbl_size - \
+				       RAS_TABLE_HEADER_SIZE - \
+				       RAS_TABLE_V2_1_INFO_SIZE) / RAS_TABLE_RECORD_SIZE)
+
 #define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control))->adev
 
 static bool __is_ras_eeprom_supported(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
index 6d21cfdeffa5e..069249249c764 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
@@ -27,6 +27,7 @@
 #include <linux/i2c.h>
 
 #define RAS_TABLE_VER_V1           0x00010000
+#define RAS_TABLE_VER_V2_1         0x00021000
 
 struct amdgpu_device;
 
-- 
GitLab


From b573cf88c0d0a1f71873ca36edf0e20d4b9a82a8 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Tue, 30 May 2023 22:48:34 +0800
Subject: [PATCH 1596/1791] drm/amdgpu: Support setting EEPROM table version

Add setting EEPROM table version interface for umcv8.10,
Add EEPROM table v2.1 to UMC v8.10.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h | 2 ++
 drivers/gpu/drm/amd/amdgpu/umc_v8_10.c  | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
index 86133f77a9a42..43321f57f557f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
@@ -59,6 +59,8 @@ struct amdgpu_umc_ras {
 				      void *ras_error_status);
 	void (*ecc_info_query_ras_error_address)(struct amdgpu_device *adev,
 					void *ras_error_status);
+	/* support different eeprom table version for different asic */
+	void (*set_eeprom_table_version)(struct amdgpu_ras_eeprom_table_header *hdr);
 };
 
 struct amdgpu_umc_funcs {
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c
index d51ae0bc36f79..46bfdee79bfd2 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c
@@ -444,6 +444,11 @@ static void umc_v8_10_ecc_info_query_ras_error_address(struct amdgpu_device *ade
 		umc_v8_10_ecc_info_query_error_address, ras_error_status);
 }
 
+static void umc_v8_10_set_eeprom_table_version(struct amdgpu_ras_eeprom_table_header *hdr)
+{
+	hdr->version = RAS_TABLE_VER_V2_1;
+}
+
 const struct amdgpu_ras_block_hw_ops umc_v8_10_ras_hw_ops = {
 	.query_ras_error_count = umc_v8_10_query_ras_error_count,
 	.query_ras_error_address = umc_v8_10_query_ras_error_address,
@@ -457,4 +462,5 @@ struct amdgpu_umc_ras umc_v8_10_ras = {
 	.query_ras_poison_mode = umc_v8_10_query_ras_poison_mode,
 	.ecc_info_query_ras_error_count = umc_v8_10_ecc_info_query_ras_error_count,
 	.ecc_info_query_ras_error_address = umc_v8_10_ecc_info_query_ras_error_address,
+	.set_eeprom_table_version = umc_v8_10_set_eeprom_table_version,
 };
-- 
GitLab


From 7f599fed3b13fe97dcd6f68bf8a5c62abb91d0a4 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Wed, 31 May 2023 10:37:09 +0800
Subject: [PATCH 1597/1791] drm/amdgpu: Add support EEPROM table v2.1

Add ras info to EEPROM table, app can analyse device ECC
status without GPU driver through EEPROM table ras info.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c       |   2 +-
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c    | 204 ++++++++++++++++--
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h    |  12 +-
 3 files changed, 203 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 5bd1bdb363ca7..a6c3265cdbc46 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2314,7 +2314,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev)
 	atomic_set(&con->in_recovery, 0);
 	con->eeprom_control.bad_channel_bitmap = 0;
 
-	max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count();
+	max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(&con->eeprom_control);
 	amdgpu_ras_validate_threshold(adev, max_eeprom_records_count);
 
 	/* Todo: During test the SMU might fail to read the eeprom through I2C
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index e835f68d1ebbf..7f134c3b0b145 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -72,6 +72,20 @@
 /* Bad GPU tag ‘BADG’ */
 #define RAS_TABLE_HDR_BAD       0x42414447
 
+/**
+ * EEPROM Table structure v1
+ * ---------------------------------
+ * |                               |
+ * |     EEPROM TABLE HEADER       |
+ * |      ( size 20 Bytes )        |
+ * |                               |
+ * ---------------------------------
+ * |                               |
+ * |    BAD PAGE RECORD AREA       |
+ * |                               |
+ * ---------------------------------
+ */
+
 /* Assume 2-Mbit size EEPROM and take up the whole space. */
 #define RAS_TBL_SIZE_BYTES      (256 * 1024)
 #define RAS_TABLE_START         0
@@ -80,6 +94,26 @@
 #define RAS_MAX_RECORD_COUNT    ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE) \
 				 / RAS_TABLE_RECORD_SIZE)
 
+/**
+ * EEPROM Table structrue v2.1
+ * ---------------------------------
+ * |                               |
+ * |     EEPROM TABLE HEADER       |
+ * |      ( size 20 Bytes )        |
+ * |                               |
+ * ---------------------------------
+ * |                               |
+ * |     EEPROM TABLE RAS INFO     |
+ * | (available info size 4 Bytes) |
+ * |  ( reserved size 252 Bytes )  |
+ * |                               |
+ * ---------------------------------
+ * |                               |
+ * |     BAD PAGE RECORD AREA      |
+ * |                               |
+ * ---------------------------------
+ */
+
 /* EEPROM Table V2_1 */
 #define RAS_TABLE_V2_1_INFO_SIZE       256
 #define RAS_TABLE_V2_1_INFO_START      RAS_TABLE_HEADER_SIZE
@@ -242,6 +276,69 @@ static int __write_table_header(struct amdgpu_ras_eeprom_control *control)
 	return res;
 }
 
+static void
+__encode_table_ras_info_to_buf(struct amdgpu_ras_eeprom_table_ras_info *rai,
+			       unsigned char *buf)
+{
+	u32 *pp = (uint32_t *)buf;
+	u32 tmp;
+
+	tmp = ((uint32_t)(rai->rma_status) & 0xFF) |
+	      (((uint32_t)(rai->health_percent) << 8) & 0xFF00) |
+	      (((uint32_t)(rai->ecc_page_threshold) << 16) & 0xFFFF0000);
+	pp[0] = cpu_to_le32(tmp);
+}
+
+static void
+__decode_table_ras_info_from_buf(struct amdgpu_ras_eeprom_table_ras_info *rai,
+				 unsigned char *buf)
+{
+	u32 *pp = (uint32_t *)buf;
+	u32 tmp;
+
+	tmp = le32_to_cpu(pp[0]);
+	rai->rma_status = tmp & 0xFF;
+	rai->health_percent = (tmp >> 8) & 0xFF;
+	rai->ecc_page_threshold = (tmp >> 16) & 0xFFFF;
+}
+
+static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control)
+{
+	struct amdgpu_device *adev = to_amdgpu_device(control);
+	u8 *buf;
+	int res;
+
+	buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL);
+	if (!buf) {
+		DRM_ERROR("Failed to alloc buf to write table ras info\n");
+		return -ENOMEM;
+	}
+
+	__encode_table_ras_info_to_buf(&control->tbl_rai, buf);
+
+	/* i2c may be unstable in gpu reset */
+	down_read(&adev->reset_domain->sem);
+	res = amdgpu_eeprom_write(adev->pm.ras_eeprom_i2c_bus,
+				  control->i2c_address +
+				  control->ras_info_offset,
+				  buf, RAS_TABLE_V2_1_INFO_SIZE);
+	up_read(&adev->reset_domain->sem);
+
+	if (res < 0) {
+		DRM_ERROR("Failed to write EEPROM table ras info:%d", res);
+	} else if (res < RAS_TABLE_V2_1_INFO_SIZE) {
+		DRM_ERROR("Short write:%d out of %d\n",
+			  res, RAS_TABLE_V2_1_INFO_SIZE);
+		res = -EIO;
+	} else {
+		res = 0;
+	}
+
+	kfree(buf);
+
+	return res;
+}
+
 static u8 __calc_hdr_byte_sum(const struct amdgpu_ras_eeprom_control *control)
 {
 	int ii;
@@ -301,14 +398,27 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 	mutex_lock(&control->ras_tbl_mutex);
 
 	hdr->header = RAS_TABLE_HDR_VAL;
-	hdr->version = RAS_TABLE_VER_V1;
-	hdr->first_rec_offset = RAS_RECORD_START;
-	hdr->tbl_size = RAS_TABLE_HEADER_SIZE;
+	if (adev->umc.ras &&
+	    adev->umc.ras->set_eeprom_table_version)
+		adev->umc.ras->set_eeprom_table_version(hdr);
+	else
+		hdr->version = RAS_TABLE_VER_V1;
+
+	if (hdr->version == RAS_TABLE_VER_V2_1) {
+		hdr->first_rec_offset = RAS_RECORD_START_V2_1;
+		hdr->tbl_size = RAS_TABLE_HEADER_SIZE +
+				RAS_TABLE_V2_1_INFO_SIZE;
+	} else {
+		hdr->first_rec_offset = RAS_RECORD_START;
+		hdr->tbl_size = RAS_TABLE_HEADER_SIZE;
+	}
 
 	csum = __calc_hdr_byte_sum(control);
 	csum = -csum;
 	hdr->checksum = csum;
 	res = __write_table_header(control);
+	if (!res && hdr->version > RAS_TABLE_VER_V1)
+		res = __write_table_ras_info(control);
 
 	control->ras_num_recs = 0;
 	control->ras_fri = 0;
@@ -587,9 +697,13 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 		control->tbl_hdr.header = RAS_TABLE_HDR_BAD;
 	}
 
-	control->tbl_hdr.version = RAS_TABLE_VER_V1;
-	control->tbl_hdr.first_rec_offset = RAS_INDEX_TO_OFFSET(control, control->ras_fri);
-	control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE + control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
+	if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1)
+		control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE +
+					    RAS_TABLE_V2_1_INFO_SIZE +
+					    control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
+	else
+		control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE +
+					    control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
 	control->tbl_hdr.checksum = 0;
 
 	buf_size = control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
@@ -629,6 +743,8 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 	csum = -csum;
 	control->tbl_hdr.checksum = csum;
 	res = __write_table_header(control);
+	if (!res && control->tbl_hdr.version > RAS_TABLE_VER_V1)
+		res = __write_table_ras_info(control);
 Out:
 	kfree(buf);
 	return res;
@@ -819,9 +935,12 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control,
 	return res;
 }
 
-uint32_t amdgpu_ras_eeprom_max_record_count(void)
+uint32_t amdgpu_ras_eeprom_max_record_count(struct amdgpu_ras_eeprom_control *control)
 {
-	return RAS_MAX_RECORD_COUNT;
+	if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1)
+		return RAS_MAX_RECORD_COUNT_V2_1;
+	else
+		return RAS_MAX_RECORD_COUNT;
 }
 
 static ssize_t
@@ -1063,8 +1182,14 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control
 	int buf_size, res;
 	u8  csum, *buf, *pp;
 
-	buf_size = RAS_TABLE_HEADER_SIZE +
-		control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
+	if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1)
+		buf_size = RAS_TABLE_HEADER_SIZE +
+			   RAS_TABLE_V2_1_INFO_SIZE +
+			   control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
+	else
+		buf_size = RAS_TABLE_HEADER_SIZE +
+			   control->ras_num_recs * RAS_TABLE_RECORD_SIZE;
+
 	buf = kzalloc(buf_size, GFP_KERNEL);
 	if (!buf) {
 		DRM_ERROR("Out of memory checking RAS table checksum.\n");
@@ -1092,6 +1217,39 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control
 	return res < 0 ? res : csum;
 }
 
+static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control)
+{
+	struct amdgpu_ras_eeprom_table_ras_info *rai = &control->tbl_rai;
+	struct amdgpu_device *adev = to_amdgpu_device(control);
+	unsigned char *buf;
+	int res;
+
+	buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL);
+	if (!buf) {
+		DRM_ERROR("Failed to alloc buf to read EEPROM table ras info\n");
+		return -ENOMEM;
+	}
+
+	/**
+	 * EEPROM table V2_1 supports ras info,
+	 * read EEPROM table ras info
+	 */
+	res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus,
+				 control->i2c_address + control->ras_info_offset,
+				 buf, RAS_TABLE_V2_1_INFO_SIZE);
+	if (res < RAS_TABLE_V2_1_INFO_SIZE) {
+		DRM_ERROR("Failed to read EEPROM table ras info, res:%d", res);
+		res = res >= 0 ? -EIO : res;
+		goto Out;
+	}
+
+	__decode_table_ras_info_from_buf(rai, buf);
+
+Out:
+	kfree(buf);
+	return res == RAS_TABLE_V2_1_INFO_SIZE ? 0 : res;
+}
+
 int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
 			   bool *exceed_err_limit)
 {
@@ -1114,8 +1272,7 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
 		return -EINVAL;
 
 	control->ras_header_offset = RAS_HDR_START;
-	control->ras_record_offset = RAS_RECORD_START;
-	control->ras_max_record_count  = RAS_MAX_RECORD_COUNT;
+	control->ras_info_offset = RAS_TABLE_V2_1_INFO_START;
 	mutex_init(&control->ras_tbl_mutex);
 
 	/* Read the table header from EEPROM address */
@@ -1129,12 +1286,27 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
 
 	__decode_table_header_from_buf(hdr, buf);
 
-	control->ras_num_recs = RAS_NUM_RECS(hdr);
+	if (hdr->version == RAS_TABLE_VER_V2_1) {
+		control->ras_num_recs = RAS_NUM_RECS_V2_1(hdr);
+		control->ras_record_offset = RAS_RECORD_START_V2_1;
+		control->ras_max_record_count = RAS_MAX_RECORD_COUNT_V2_1;
+	} else {
+		control->ras_num_recs = RAS_NUM_RECS(hdr);
+		control->ras_record_offset = RAS_RECORD_START;
+		control->ras_max_record_count = RAS_MAX_RECORD_COUNT;
+	}
 	control->ras_fri = RAS_OFFSET_TO_INDEX(control, hdr->first_rec_offset);
 
 	if (hdr->header == RAS_TABLE_HDR_VAL) {
 		DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records",
 				 control->ras_num_recs);
+
+		if (hdr->version == RAS_TABLE_VER_V2_1) {
+			res = __read_table_ras_info(control);
+			if (res)
+				return res;
+		}
+
 		res = __verify_ras_table_checksum(control);
 		if (res)
 			DRM_ERROR("RAS table incorrect checksum or error:%d\n",
@@ -1148,6 +1320,12 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control,
 					ras->bad_page_cnt_threshold);
 	} else if (hdr->header == RAS_TABLE_HDR_BAD &&
 		   amdgpu_bad_page_threshold != 0) {
+		if (hdr->version == RAS_TABLE_VER_V2_1) {
+			res = __read_table_ras_info(control);
+			if (res)
+				return res;
+		}
+
 		res = __verify_ras_table_checksum(control);
 		if (res)
 			DRM_ERROR("RAS Table incorrect checksum or error:%d\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
index 069249249c764..3c5575c19bf87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
@@ -46,9 +46,18 @@ struct amdgpu_ras_eeprom_table_header {
 	uint32_t checksum;
 } __packed;
 
+struct amdgpu_ras_eeprom_table_ras_info {
+	u8  rma_status;
+	u8  health_percent;
+	u16 ecc_page_threshold;
+	u32 padding[64 - 1];
+} __packed;
+
 struct amdgpu_ras_eeprom_control {
 	struct amdgpu_ras_eeprom_table_header tbl_hdr;
 
+	struct amdgpu_ras_eeprom_table_ras_info tbl_rai;
+
 	/* Base I2C EEPPROM 19-bit memory address,
 	 * where the table is located. For more information,
 	 * see top of amdgpu_eeprom.c.
@@ -61,6 +70,7 @@ struct amdgpu_ras_eeprom_control {
 	 * right after the header.
 	 */
 	u32 ras_header_offset;
+	u32 ras_info_offset;
 	u32 ras_record_offset;
 
 	/* Number of records in the table.
@@ -127,7 +137,7 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control,
 int amdgpu_ras_eeprom_append(struct amdgpu_ras_eeprom_control *control,
 			     struct eeprom_table_record *records, const u32 num);
 
-uint32_t amdgpu_ras_eeprom_max_record_count(void);
+uint32_t amdgpu_ras_eeprom_max_record_count(struct amdgpu_ras_eeprom_control *control);
 
 void amdgpu_ras_debugfs_set_ret_size(struct amdgpu_ras_eeprom_control *control);
 
-- 
GitLab


From 7c2551fa1dfdb06a9dd3a6c629086fe2c348e00a Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Mon, 29 May 2023 10:17:59 +0800
Subject: [PATCH 1598/1791] drm/amdgpu: Calculate EEPROM table ras info bytes
 sum

It's more reasonable to check EEPROM table ras info bytes.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c    | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index 7f134c3b0b145..9eceb3bc10583 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -355,6 +355,21 @@ static u8 __calc_hdr_byte_sum(const struct amdgpu_ras_eeprom_control *control)
 	return csum;
 }
 
+static u8 __calc_ras_info_byte_sum(const struct amdgpu_ras_eeprom_control *control)
+{
+	int ii;
+	u8  *pp, csum;
+	size_t sz;
+
+	sz = sizeof(control->tbl_rai);
+	pp = (u8 *) &control->tbl_rai;
+	csum = 0;
+	for (ii = 0; ii < sz; ii++, pp++)
+		csum += *pp;
+
+	return csum;
+}
+
 static int amdgpu_ras_eeprom_correct_header_tag(
 	struct amdgpu_ras_eeprom_control *control,
 	uint32_t header)
@@ -414,6 +429,8 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 	}
 
 	csum = __calc_hdr_byte_sum(control);
+	if (hdr->version == RAS_TABLE_VER_V2_1)
+		csum += __calc_ras_info_byte_sum(control);
 	csum = -csum;
 	hdr->checksum = csum;
 	res = __write_table_header(control);
@@ -739,6 +756,8 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 		csum += *pp;
 
 	csum += __calc_hdr_byte_sum(control);
+	if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1)
+		csum += __calc_ras_info_byte_sum(control);
 	/* avoid sign extension when assigning to "checksum" */
 	csum = -csum;
 	control->tbl_hdr.checksum = csum;
-- 
GitLab


From 0bc3137b2157115f328859477b463c912d605c3a Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Thu, 1 Jun 2023 20:56:42 +0800
Subject: [PATCH 1599/1791] drm/amdgpu: Set EEPROM ras info

Set EEPROM ras info: rma status, health percent and bad
page threshold.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c    | 24 +++++++++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h    |  5 ++++
 2 files changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index 9eceb3bc10583..c2e8f6491ac67 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -406,6 +406,7 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 {
 	struct amdgpu_device *adev = to_amdgpu_device(control);
 	struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr;
+	struct amdgpu_ras_eeprom_table_ras_info *rai = &control->tbl_rai;
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	u8 csum;
 	int res;
@@ -423,6 +424,14 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control)
 		hdr->first_rec_offset = RAS_RECORD_START_V2_1;
 		hdr->tbl_size = RAS_TABLE_HEADER_SIZE +
 				RAS_TABLE_V2_1_INFO_SIZE;
+		rai->rma_status = GPU_HEALTH_USABLE;
+		/**
+		 * GPU health represented as a percentage.
+		 * 0 means worst health, 100 means fully health.
+		 */
+		rai->health_percent = 100;
+		/* ecc_page_threshold = 0 means disable bad page retirement */
+		rai->ecc_page_threshold = con->bad_page_cnt_threshold;
 	} else {
 		hdr->first_rec_offset = RAS_RECORD_START;
 		hdr->tbl_size = RAS_TABLE_HEADER_SIZE;
@@ -712,6 +721,10 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 			"Saved bad pages %d reaches threshold value %d\n",
 			control->ras_num_recs, ras->bad_page_cnt_threshold);
 		control->tbl_hdr.header = RAS_TABLE_HDR_BAD;
+		if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) {
+			control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD;
+			control->tbl_rai.health_percent = 0;
+		}
 	}
 
 	if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1)
@@ -749,6 +762,17 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 		goto Out;
 	}
 
+	/**
+	 * bad page records have been stored in eeprom,
+	 * now calculate gpu health percent
+	 */
+	if (amdgpu_bad_page_threshold != 0 &&
+	    control->tbl_hdr.version == RAS_TABLE_VER_V2_1 &&
+	    control->ras_num_recs < ras->bad_page_cnt_threshold)
+		control->tbl_rai.health_percent = ((ras->bad_page_cnt_threshold -
+						   control->ras_num_recs) * 100) /
+						   ras->bad_page_cnt_threshold;
+
 	/* Recalc the checksum.
 	 */
 	csum = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
index 3c5575c19bf87..6dfd667f3013d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h
@@ -31,6 +31,11 @@
 
 struct amdgpu_device;
 
+enum amdgpu_ras_gpu_health_status {
+	GPU_HEALTH_USABLE = 0,
+	GPU_RETIRED__ECC_REACH_THRESHOLD = 2,
+};
+
 enum amdgpu_ras_eeprom_err_type {
 	AMDGPU_RAS_EEPROM_ERR_NA,
 	AMDGPU_RAS_EEPROM_ERR_RECOVERABLE,
-- 
GitLab


From 7386f88ab1732af890a09ab3a7f400bb20adbe5a Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Wed, 7 Jun 2023 03:56:27 -0400
Subject: [PATCH 1600/1791] drm/amdkfd: fix vmfault signalling with additional
 data.

Exception handling for vmfaults should be raised with additional data.

Reported-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Mukul Joshi <mukul.joshi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_events.c | 34 +++++++++++++++----------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 0a5e7b172a648..ddca23ee4193e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -1240,19 +1240,24 @@ void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
 		return;
 	}
 
-	memset(&memory_exception_data, 0, sizeof(memory_exception_data));
-	memory_exception_data.gpu_id = user_gpu_id;
-	memory_exception_data.failure.imprecise = true;
-	/* Set failure reason */
-	if (info) {
-		memory_exception_data.va = (info->page_addr) << PAGE_SHIFT;
-		memory_exception_data.failure.NotPresent =
-			info->prot_valid ? 1 : 0;
-		memory_exception_data.failure.NoExecute =
-			info->prot_exec ? 1 : 0;
-		memory_exception_data.failure.ReadOnly =
-			info->prot_write ? 1 : 0;
-		memory_exception_data.failure.imprecise = 0;
+	/* SoC15 chips and onwards will pass in data from now on. */
+	if (!data) {
+		memset(&memory_exception_data, 0, sizeof(memory_exception_data));
+		memory_exception_data.gpu_id = user_gpu_id;
+		memory_exception_data.failure.imprecise = true;
+
+		/* Set failure reason */
+		if (info) {
+			memory_exception_data.va = (info->page_addr) <<
+								PAGE_SHIFT;
+			memory_exception_data.failure.NotPresent =
+				info->prot_valid ? 1 : 0;
+			memory_exception_data.failure.NoExecute =
+				info->prot_exec ? 1 : 0;
+			memory_exception_data.failure.ReadOnly =
+				info->prot_write ? 1 : 0;
+			memory_exception_data.failure.imprecise = 0;
+		}
 	}
 
 	rcu_read_lock();
@@ -1261,7 +1266,8 @@ void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
 	idr_for_each_entry_continue(&p->event_idr, ev, id)
 		if (ev->type == KFD_EVENT_TYPE_MEMORY) {
 			spin_lock(&ev->lock);
-			ev->memory_exception_data = memory_exception_data;
+			ev->memory_exception_data = data ? *data :
+							memory_exception_data;
 			set_event(ev);
 			spin_unlock(&ev->lock);
 		}
-- 
GitLab


From bbcc3514ab4f7ec3ae2273ad08b0a1b6b4aa9dd9 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Fri, 2 Jun 2023 07:18:06 -0500
Subject: [PATCH 1601/1791] drm/amd: Check that a system is a NUMA system
 before looking for SRAT

It's pointless on laptops to look for the SRAT table as these are not
NUMA.  Check the number of possible nodes is > 1 to decide whether to
look for SRAT.

Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 950af6820153b..3dcd8f8bc98e5 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -2041,7 +2041,8 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
 	sub_type_hdr->proximity_domain_from = proximity_domain;
 
 #ifdef CONFIG_ACPI_NUMA
-	if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE)
+	if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE &&
+	    num_possible_nodes() > 1)
 		kfd_find_numa_node_in_srat(kdev);
 #endif
 #ifdef CONFIG_NUMA
-- 
GitLab


From c1ac2ea802f5adfd1d128fc01375af9c5f113932 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 6 Jun 2023 11:14:04 -0400
Subject: [PATCH 1602/1791] drm/amdgpu: add missing radeon secondary PCI ID
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

0x5b70 is a missing RV370 secondary id.  Add it so
we don't try and probe it with amdgpu.

Cc: michel@daenzer.net
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Tested-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 7489b2b1a0d01..b8a1e4571cd93 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -1640,6 +1640,7 @@ static const u16 amdgpu_unsupported_pciidlist[] = {
 	0x5874,
 	0x5940,
 	0x5941,
+	0x5b70,
 	0x5b72,
 	0x5b73,
 	0x5b74,
-- 
GitLab


From 597364adc0fcf71617b3adbe647b6eec76e27554 Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Wed, 31 May 2023 11:22:03 -0400
Subject: [PATCH 1603/1791] drm/amdkfd: Fix reserved SDMA queues handling

This patch fixes a regression caused by a bad merge where
the handling of reserved SDMA queues was accidentally removed.
With the fix, the reserved SDMA queues are again correctly
marked as unavailable for allocation.

Fixes: a805889a1531 ("drm/amdkfd: Update SDMA queue management for GFX9.4.3")
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c             | 13 ++++++-------
 .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c   | 10 +++++-----
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h               |  2 +-
 3 files changed, 12 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 9fc9d32cb5796..9d4abfd8b55eb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -106,20 +106,19 @@ static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd)
 		kfd->device_info.num_sdma_queues_per_engine = 8;
 	}
 
+	bitmap_zero(kfd->device_info.reserved_sdma_queues_bitmap, KFD_MAX_SDMA_QUEUES);
+
 	switch (sdma_version) {
 	case IP_VERSION(6, 0, 0):
+	case IP_VERSION(6, 0, 1):
 	case IP_VERSION(6, 0, 2):
 	case IP_VERSION(6, 0, 3):
 		/* Reserve 1 for paging and 1 for gfx */
 		kfd->device_info.num_reserved_sdma_queues_per_engine = 2;
 		/* BIT(0)=engine-0 queue-0; BIT(1)=engine-1 queue-0; BIT(2)=engine-0 queue-1; ... */
-		kfd->device_info.reserved_sdma_queues_bitmap = 0xFULL;
-		break;
-	case IP_VERSION(6, 0, 1):
-		/* Reserve 1 for paging and 1 for gfx */
-		kfd->device_info.num_reserved_sdma_queues_per_engine = 2;
-		/* BIT(0)=engine-0 queue-0; BIT(1)=engine-0 queue-1; ... */
-		kfd->device_info.reserved_sdma_queues_bitmap = 0x3ULL;
+		bitmap_set(kfd->device_info.reserved_sdma_queues_bitmap, 0,
+			   kfd->adev->sdma.num_instances *
+			   kfd->device_info.num_reserved_sdma_queues_per_engine);
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 0c1be91a87c69..498ad7d4e7d90 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -123,11 +123,6 @@ unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm)
 		dqm->dev->kfd->device_info.num_sdma_queues_per_engine;
 }
 
-static inline uint64_t get_reserved_sdma_queues_bitmap(struct device_queue_manager *dqm)
-{
-	return dqm->dev->kfd->device_info.reserved_sdma_queues_bitmap;
-}
-
 static void init_sdma_bitmaps(struct device_queue_manager *dqm)
 {
 	bitmap_zero(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES);
@@ -135,6 +130,11 @@ static void init_sdma_bitmaps(struct device_queue_manager *dqm)
 
 	bitmap_zero(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES);
 	bitmap_set(dqm->xgmi_sdma_bitmap, 0, get_num_xgmi_sdma_queues(dqm));
+
+	/* Mask out the reserved queues */
+	bitmap_andnot(dqm->sdma_bitmap, dqm->sdma_bitmap,
+		      dqm->dev->kfd->device_info.reserved_sdma_queues_bitmap,
+		      KFD_MAX_SDMA_QUEUES);
 }
 
 void program_sh_mem_settings(struct device_queue_manager *dqm,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 023b17e0116be..7364a5d77c6ee 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -239,7 +239,7 @@ struct kfd_device_info {
 	uint32_t no_atomic_fw_version;
 	unsigned int num_sdma_queues_per_engine;
 	unsigned int num_reserved_sdma_queues_per_engine;
-	uint64_t reserved_sdma_queues_bitmap;
+	DECLARE_BITMAP(reserved_sdma_queues_bitmap, KFD_MAX_SDMA_QUEUES);
 };
 
 unsigned int kfd_get_num_sdma_engines(struct kfd_node *kdev);
-- 
GitLab


From 1626761ee4406c51d5afe9d47dd41a29e2049b71 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Fri, 3 Feb 2023 02:07:42 +0000
Subject: [PATCH 1604/1791] drm/connector: Convert DRM_MODE_COLORIMETRY to enum
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This allows us to use strongly typed arguments.

v2:
 - Bring NO_DATA back
 - Provide explicit enum values

v3:
- Drop unnecessary '&' from kerneldoc (emersion)

v4:
- Fix Normal Colorimetry comment

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/drm/display/drm_dp.h |  2 +-
 include/drm/drm_connector.h  | 49 ++++++++++++++++++------------------
 2 files changed, 26 insertions(+), 25 deletions(-)

diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h
index d735073fdd814..ccbf0c0934c36 100644
--- a/include/drm/display/drm_dp.h
+++ b/include/drm/display/drm_dp.h
@@ -1636,7 +1636,7 @@ enum dp_pixelformat {
  *
  * This enum is used to indicate DP VSC SDP Colorimetry formats.
  * It is based on DP 1.4 spec [Table 2-117: VSC SDP Payload for DB16 through
- * DB18] and a name of enum member follows DRM_MODE_COLORIMETRY definition.
+ * DB18] and a name of enum member follows enum drm_colorimetry definition.
  *
  * @DP_COLORIMETRY_DEFAULT: sRGB (IEC 61966-2-1) or
  *                          ITU-R BT.601 colorimetry format
diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index 7b5048516185c..31ecd36bc7ea2 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -427,29 +427,30 @@ enum drm_privacy_screen_status {
  * a colorspace property which will be created and exposed to
  * userspace.
  */
-
-/* For Default case, driver will set the colorspace */
-#define DRM_MODE_COLORIMETRY_DEFAULT			0
-/* CEA 861 Normal Colorimetry options */
-#define DRM_MODE_COLORIMETRY_NO_DATA			0
-#define DRM_MODE_COLORIMETRY_SMPTE_170M_YCC		1
-#define DRM_MODE_COLORIMETRY_BT709_YCC			2
-/* CEA 861 Extended Colorimetry Options */
-#define DRM_MODE_COLORIMETRY_XVYCC_601			3
-#define DRM_MODE_COLORIMETRY_XVYCC_709			4
-#define DRM_MODE_COLORIMETRY_SYCC_601			5
-#define DRM_MODE_COLORIMETRY_OPYCC_601			6
-#define DRM_MODE_COLORIMETRY_OPRGB			7
-#define DRM_MODE_COLORIMETRY_BT2020_CYCC		8
-#define DRM_MODE_COLORIMETRY_BT2020_RGB			9
-#define DRM_MODE_COLORIMETRY_BT2020_YCC			10
-/* Additional Colorimetry extension added as part of CTA 861.G */
-#define DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65		11
-#define DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER		12
-/* Additional Colorimetry Options added for DP 1.4a VSC Colorimetry Format */
-#define DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED		13
-#define DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT		14
-#define DRM_MODE_COLORIMETRY_BT601_YCC			15
+enum drm_colorspace {
+	/* For Default case, driver will set the colorspace */
+	DRM_MODE_COLORIMETRY_DEFAULT 		= 0,
+	/* CEA 861 Normal Colorimetry options */
+	DRM_MODE_COLORIMETRY_NO_DATA		= 0,
+	DRM_MODE_COLORIMETRY_SMPTE_170M_YCC	= 1,
+	DRM_MODE_COLORIMETRY_BT709_YCC		= 2,
+	/* CEA 861 Extended Colorimetry Options */
+	DRM_MODE_COLORIMETRY_XVYCC_601		= 3,
+	DRM_MODE_COLORIMETRY_XVYCC_709		= 4,
+	DRM_MODE_COLORIMETRY_SYCC_601		= 5,
+	DRM_MODE_COLORIMETRY_OPYCC_601		= 6,
+	DRM_MODE_COLORIMETRY_OPRGB		= 7,
+	DRM_MODE_COLORIMETRY_BT2020_CYCC	= 8,
+	DRM_MODE_COLORIMETRY_BT2020_RGB		= 9,
+	DRM_MODE_COLORIMETRY_BT2020_YCC		= 10,
+	/* Additional Colorimetry extension added as part of CTA 861.G */
+	DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65	= 11,
+	DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER	= 12,
+	/* Additional Colorimetry Options added for DP 1.4a VSC Colorimetry Format */
+	DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED	= 13,
+	DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT	= 14,
+	DRM_MODE_COLORIMETRY_BT601_YCC		= 15,
+};
 
 /**
  * enum drm_bus_flags - bus_flags info for &drm_display_info
@@ -901,7 +902,7 @@ struct drm_connector_state {
 	 * colorspace change on Sink. This is most commonly used to switch
 	 * to wider color gamuts like BT2020.
 	 */
-	u32 colorspace;
+	enum drm_colorspace colorspace;
 
 	/**
 	 * @writeback_job: Writeback job for writeback connectors
-- 
GitLab


From f96c61fe0383d73732aba72fabb7e2c7ce0b0835 Mon Sep 17 00:00:00 2001
From: Joshua Ashton <joshua@froggi.es>
Date: Fri, 3 Feb 2023 02:07:43 +0000
Subject: [PATCH 1605/1791] drm/connector: Add enum documentation to
 drm_colorspace
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To match the other enums, and add more information about these values.

v2:
 - Specify where an enum entry comes from
 - Clarify DEFAULT and NO_DATA behavior
 - BT.2020 CYCC is "constant luminance"
 - correct type for BT.601

v4:
- drop DP/HDMI clarifications that might create
  more questions than answers

v5:
- Add note on YCC and RGB variants

Signed-off-by: Joshua Ashton <joshua@froggi.es>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.com>
Reviewed-by: Simon Ser <contact@emersion.fr>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/drm/drm_connector.h | 70 +++++++++++++++++++++++++++++++++++--
 1 file changed, 68 insertions(+), 2 deletions(-)

diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index 31ecd36bc7ea2..e338432580e07 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -419,13 +419,79 @@ enum drm_privacy_screen_status {
 	PRIVACY_SCREEN_ENABLED_LOCKED,
 };
 
-/*
- * This is a consolidated colorimetry list supported by HDMI and
+/**
+ * enum drm_colorspace - color space
+ *
+ * This enum is a consolidated colorimetry list supported by HDMI and
  * DP protocol standard. The respective connectors will register
  * a property with the subset of this list (supported by that
  * respective protocol). Userspace will set the colorspace through
  * a colorspace property which will be created and exposed to
  * userspace.
+ *
+ * DP definitions come from the DP v2.0 spec
+ * HDMI definitions come from the CTA-861-H spec
+ *
+ * A note on YCC and RGB variants:
+ *
+ * Since userspace is not aware of the encoding on the wire
+ * (RGB or YCbCr), drivers are free to pick the appropriate
+ * variant, regardless of what userspace selects. E.g., if
+ * BT2020_RGB is selected by userspace a driver will pick
+ * BT2020_YCC if the encoding on the wire is YUV444 or YUV420.
+  *
+ * @DRM_MODE_COLORIMETRY_DEFAULT:
+ *   Driver specific behavior.
+ * @DRM_MODE_COLORIMETRY_NO_DATA:
+ *   Driver specific behavior.
+ * @DRM_MODE_COLORIMETRY_SMPTE_170M_YCC:
+ *   (HDMI)
+ *   SMPTE ST 170M colorimetry format
+ * @DRM_MODE_COLORIMETRY_BT709_YCC:
+ *   (HDMI, DP)
+ *   ITU-R BT.709 colorimetry format
+ * @DRM_MODE_COLORIMETRY_XVYCC_601:
+ *   (HDMI, DP)
+ *   xvYCC601 colorimetry format
+ * @DRM_MODE_COLORIMETRY_XVYCC_709:
+ *   (HDMI, DP)
+ *   xvYCC709 colorimetry format
+ * @DRM_MODE_COLORIMETRY_SYCC_601:
+ *   (HDMI, DP)
+ *   sYCC601 colorimetry format
+ * @DRM_MODE_COLORIMETRY_OPYCC_601:
+ *   (HDMI, DP)
+ *   opYCC601 colorimetry format
+ * @DRM_MODE_COLORIMETRY_OPRGB:
+ *   (HDMI, DP)
+ *   opRGB colorimetry format
+ * @DRM_MODE_COLORIMETRY_BT2020_CYCC:
+ *   (HDMI, DP)
+ *   ITU-R BT.2020 Y'c C'bc C'rc (constant luminance) colorimetry format
+ * @DRM_MODE_COLORIMETRY_BT2020_RGB:
+ *   (HDMI, DP)
+ *   ITU-R BT.2020 R' G' B' colorimetry format
+ * @DRM_MODE_COLORIMETRY_BT2020_YCC:
+ *   (HDMI, DP)
+ *   ITU-R BT.2020 Y' C'b C'r colorimetry format
+ * @DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65:
+ *   (HDMI)
+ *   SMPTE ST 2113 P3D65 colorimetry format
+ * @DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER:
+ *   (HDMI)
+ *   SMPTE ST 2113 P3DCI colorimetry format
+ * @DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED:
+ *   (DP)
+ *   RGB wide gamut fixed point colorimetry format
+ * @DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT:
+ *   (DP)
+ *   RGB wide gamut floating point
+ *   (scRGB (IEC 61966-2-2)) colorimetry format
+ * @DRM_MODE_COLORIMETRY_BT601_YCC:
+ *   (DP)
+ *   ITU-R BT.601 colorimetry format
+ *   The DP spec does not say whether this is the 525 or the 625
+ *   line version.
  */
 enum drm_colorspace {
 	/* For Default case, driver will set the colorspace */
-- 
GitLab


From 6120611abc05dd850eff4eb3026f977ac7e34718 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Wed, 30 Nov 2022 14:42:03 -0500
Subject: [PATCH 1606/1791] drm/connector: Pull out common
 create_colorspace_property code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Simon Ser <contact@emersion.fr>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/drm_connector.c | 54 ++++++++++++++++-----------------
 1 file changed, 27 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index 48df7a5ea503f..614dd9fd4d3f7 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -2135,33 +2135,44 @@ EXPORT_SYMBOL(drm_mode_create_aspect_ratio_property);
  * drm_mode_create_dp_colorspace_property() is used for DP connector.
  */
 
-/**
- * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property
- * @connector: connector to create the Colorspace property on.
- *
- * Called by a driver the first time it's needed, must be attached to desired
- * HDMI connectors.
- *
- * Returns:
- * Zero on success, negative errno on failure.
- */
-int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector)
+static int drm_mode_create_colorspace_property(struct drm_connector *connector,
+					const struct drm_prop_enum_list *colorspaces,
+					int size)
 {
 	struct drm_device *dev = connector->dev;
 
 	if (connector->colorspace_property)
 		return 0;
 
+	if (!colorspaces)
+		return 0;
+
 	connector->colorspace_property =
 		drm_property_create_enum(dev, DRM_MODE_PROP_ENUM, "Colorspace",
-					 hdmi_colorspaces,
-					 ARRAY_SIZE(hdmi_colorspaces));
+					colorspaces,
+					size);
 
 	if (!connector->colorspace_property)
 		return -ENOMEM;
 
 	return 0;
 }
+/**
+ * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property
+ * @connector: connector to create the Colorspace property on.
+ *
+ * Called by a driver the first time it's needed, must be attached to desired
+ * HDMI connectors.
+ *
+ * Returns:
+ * Zero on success, negative errno on failure.
+ */
+int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector)
+{
+	return drm_mode_create_colorspace_property(connector,
+						   hdmi_colorspaces,
+						   ARRAY_SIZE(hdmi_colorspaces));
+}
 EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
 
 /**
@@ -2176,20 +2187,9 @@ EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
  */
 int drm_mode_create_dp_colorspace_property(struct drm_connector *connector)
 {
-	struct drm_device *dev = connector->dev;
-
-	if (connector->colorspace_property)
-		return 0;
-
-	connector->colorspace_property =
-		drm_property_create_enum(dev, DRM_MODE_PROP_ENUM, "Colorspace",
-					 dp_colorspaces,
-					 ARRAY_SIZE(dp_colorspaces));
-
-	if (!connector->colorspace_property)
-		return -ENOMEM;
-
-	return 0;
+	return drm_mode_create_colorspace_property(connector,
+						   dp_colorspaces,
+						   ARRAY_SIZE(dp_colorspaces));
 }
 EXPORT_SYMBOL(drm_mode_create_dp_colorspace_property);
 
-- 
GitLab


From c627087cb164d1675323c7942fa29bded4263dfc Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Tue, 29 Nov 2022 15:16:31 -0500
Subject: [PATCH 1607/1791] drm/connector: Use common colorspace_names array
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We an use bitfields to track the support ones for HDMI
and DP. This allows us to print colorspaces in a consistent
manner without needing to know whether we're dealing with
DP or HDMI.

v4:
- Rename _MAX to _COUNT and leave comment to indicate
  it's not a valid value
- Fix misplaced function doc

v6:
- Drop magic in drm_mode_create_colorspace_property for
  dealing with "0" supported_colorspaces. Expect the caller
  to always provide a non-zero supported_colorspaces.
- Improve error checking and logging

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Simon Ser <contact@emersion.fr>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/drm_connector.c | 130 +++++++++++++++++++-------------
 include/drm/drm_connector.h     |   2 +
 2 files changed, 79 insertions(+), 53 deletions(-)

diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index 614dd9fd4d3f7..b823b10ed6974 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -1055,64 +1055,70 @@ static const struct drm_prop_enum_list drm_dp_subconnector_enum_list[] = {
 DRM_ENUM_NAME_FN(drm_get_dp_subconnector_name,
 		 drm_dp_subconnector_enum_list)
 
-static const struct drm_prop_enum_list hdmi_colorspaces[] = {
+
+static const char * const colorspace_names[] = {
 	/* For Default case, driver will set the colorspace */
-	{ DRM_MODE_COLORIMETRY_DEFAULT, "Default" },
+	[DRM_MODE_COLORIMETRY_DEFAULT] = "Default",
 	/* Standard Definition Colorimetry based on CEA 861 */
-	{ DRM_MODE_COLORIMETRY_SMPTE_170M_YCC, "SMPTE_170M_YCC" },
-	{ DRM_MODE_COLORIMETRY_BT709_YCC, "BT709_YCC" },
+	[DRM_MODE_COLORIMETRY_SMPTE_170M_YCC] = "SMPTE_170M_YCC",
+	[DRM_MODE_COLORIMETRY_BT709_YCC] = "BT709_YCC",
 	/* Standard Definition Colorimetry based on IEC 61966-2-4 */
-	{ DRM_MODE_COLORIMETRY_XVYCC_601, "XVYCC_601" },
+	[DRM_MODE_COLORIMETRY_XVYCC_601] = "XVYCC_601",
 	/* High Definition Colorimetry based on IEC 61966-2-4 */
-	{ DRM_MODE_COLORIMETRY_XVYCC_709, "XVYCC_709" },
+	[DRM_MODE_COLORIMETRY_XVYCC_709] = "XVYCC_709",
 	/* Colorimetry based on IEC 61966-2-1/Amendment 1 */
-	{ DRM_MODE_COLORIMETRY_SYCC_601, "SYCC_601" },
+	[DRM_MODE_COLORIMETRY_SYCC_601] = "SYCC_601",
 	/* Colorimetry based on IEC 61966-2-5 [33] */
-	{ DRM_MODE_COLORIMETRY_OPYCC_601, "opYCC_601" },
+	[DRM_MODE_COLORIMETRY_OPYCC_601] = "opYCC_601",
 	/* Colorimetry based on IEC 61966-2-5 */
-	{ DRM_MODE_COLORIMETRY_OPRGB, "opRGB" },
+	[DRM_MODE_COLORIMETRY_OPRGB] = "opRGB",
 	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_CYCC, "BT2020_CYCC" },
+	[DRM_MODE_COLORIMETRY_BT2020_CYCC] = "BT2020_CYCC",
 	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_RGB, "BT2020_RGB" },
+	[DRM_MODE_COLORIMETRY_BT2020_RGB] = "BT2020_RGB",
 	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_YCC, "BT2020_YCC" },
+	[DRM_MODE_COLORIMETRY_BT2020_YCC] = "BT2020_YCC",
 	/* Added as part of Additional Colorimetry Extension in 861.G */
-	{ DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65, "DCI-P3_RGB_D65" },
-	{ DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER, "DCI-P3_RGB_Theater" },
+	[DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65] = "DCI-P3_RGB_D65",
+	[DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER] = "DCI-P3_RGB_Theater",
+	[DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED] = "RGB_WIDE_FIXED",
+	/* Colorimetry based on scRGB (IEC 61966-2-2) */
+	[DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT] = "RGB_WIDE_FLOAT",
+	[DRM_MODE_COLORIMETRY_BT601_YCC] = "BT601_YCC",
 };
 
+static const u32 hdmi_colorspaces =
+	BIT(DRM_MODE_COLORIMETRY_SMPTE_170M_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_BT709_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_XVYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_XVYCC_709) |
+	BIT(DRM_MODE_COLORIMETRY_SYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_OPYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_OPRGB) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_CYCC) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65) |
+	BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER);
+
 /*
  * As per DP 1.4a spec, 2.2.5.7.5 VSC SDP Payload for Pixel Encoding/Colorimetry
  * Format Table 2-120
  */
-static const struct drm_prop_enum_list dp_colorspaces[] = {
-	/* For Default case, driver will set the colorspace */
-	{ DRM_MODE_COLORIMETRY_DEFAULT, "Default" },
-	{ DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED, "RGB_Wide_Gamut_Fixed_Point" },
-	/* Colorimetry based on scRGB (IEC 61966-2-2) */
-	{ DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT, "RGB_Wide_Gamut_Floating_Point" },
-	/* Colorimetry based on IEC 61966-2-5 */
-	{ DRM_MODE_COLORIMETRY_OPRGB, "opRGB" },
-	/* Colorimetry based on SMPTE RP 431-2 */
-	{ DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65, "DCI-P3_RGB_D65" },
-	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_RGB, "BT2020_RGB" },
-	{ DRM_MODE_COLORIMETRY_BT601_YCC, "BT601_YCC" },
-	{ DRM_MODE_COLORIMETRY_BT709_YCC, "BT709_YCC" },
-	/* Standard Definition Colorimetry based on IEC 61966-2-4 */
-	{ DRM_MODE_COLORIMETRY_XVYCC_601, "XVYCC_601" },
-	/* High Definition Colorimetry based on IEC 61966-2-4 */
-	{ DRM_MODE_COLORIMETRY_XVYCC_709, "XVYCC_709" },
-	/* Colorimetry based on IEC 61966-2-1/Amendment 1 */
-	{ DRM_MODE_COLORIMETRY_SYCC_601, "SYCC_601" },
-	/* Colorimetry based on IEC 61966-2-5 [33] */
-	{ DRM_MODE_COLORIMETRY_OPYCC_601, "opYCC_601" },
-	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_CYCC, "BT2020_CYCC" },
-	/* Colorimetry based on ITU-R BT.2020 */
-	{ DRM_MODE_COLORIMETRY_BT2020_YCC, "BT2020_YCC" },
-};
+static const u32 dp_colorspaces =
+	BIT(DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED) |
+	BIT(DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT) |
+	BIT(DRM_MODE_COLORIMETRY_OPRGB) |
+	BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) |
+	BIT(DRM_MODE_COLORIMETRY_BT601_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_BT709_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_XVYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_XVYCC_709) |
+	BIT(DRM_MODE_COLORIMETRY_SYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_OPYCC_601) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_CYCC) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_YCC);
 
 /**
  * DOC: standard connector properties
@@ -2136,27 +2142,49 @@ EXPORT_SYMBOL(drm_mode_create_aspect_ratio_property);
  */
 
 static int drm_mode_create_colorspace_property(struct drm_connector *connector,
-					const struct drm_prop_enum_list *colorspaces,
-					int size)
+					u32 supported_colorspaces)
 {
 	struct drm_device *dev = connector->dev;
+	u32 colorspaces = supported_colorspaces | BIT(DRM_MODE_COLORIMETRY_DEFAULT);
+	struct drm_prop_enum_list enum_list[DRM_MODE_COLORIMETRY_COUNT];
+	int i, len;
 
 	if (connector->colorspace_property)
 		return 0;
 
-	if (!colorspaces)
-		return 0;
+	if (!supported_colorspaces) {
+		drm_err(dev, "No supported colorspaces provded on [CONNECTOR:%d:%s]\n",
+			    connector->base.id, connector->name);
+		return -EINVAL;
+	}
+
+	if ((supported_colorspaces & -BIT(DRM_MODE_COLORIMETRY_COUNT)) != 0) {
+		drm_err(dev, "Unknown colorspace provded on [CONNECTOR:%d:%s]\n",
+			    connector->base.id, connector->name);
+		return -EINVAL;
+	}
+
+	len = 0;
+	for (i = 0; i < DRM_MODE_COLORIMETRY_COUNT; i++) {
+		if ((colorspaces & BIT(i)) == 0)
+			continue;
+
+		enum_list[len].type = i;
+		enum_list[len].name = colorspace_names[i];
+		len++;
+	}
 
 	connector->colorspace_property =
 		drm_property_create_enum(dev, DRM_MODE_PROP_ENUM, "Colorspace",
-					colorspaces,
-					size);
+					enum_list,
+					len);
 
 	if (!connector->colorspace_property)
 		return -ENOMEM;
 
 	return 0;
 }
+
 /**
  * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property
  * @connector: connector to create the Colorspace property on.
@@ -2169,9 +2197,7 @@ static int drm_mode_create_colorspace_property(struct drm_connector *connector,
  */
 int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector)
 {
-	return drm_mode_create_colorspace_property(connector,
-						   hdmi_colorspaces,
-						   ARRAY_SIZE(hdmi_colorspaces));
+	return drm_mode_create_colorspace_property(connector, hdmi_colorspaces);
 }
 EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
 
@@ -2187,9 +2213,7 @@ EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
  */
 int drm_mode_create_dp_colorspace_property(struct drm_connector *connector)
 {
-	return drm_mode_create_colorspace_property(connector,
-						   dp_colorspaces,
-						   ARRAY_SIZE(dp_colorspaces));
+	return drm_mode_create_colorspace_property(connector, dp_colorspaces);
 }
 EXPORT_SYMBOL(drm_mode_create_dp_colorspace_property);
 
diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index e338432580e07..6eb8e2d5d20f2 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -516,6 +516,8 @@ enum drm_colorspace {
 	DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED	= 13,
 	DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT	= 14,
 	DRM_MODE_COLORIMETRY_BT601_YCC		= 15,
+	/* not a valid value; merely used for counting */
+	DRM_MODE_COLORIMETRY_COUNT
 };
 
 /**
-- 
GitLab


From 035d53e0f36da6ce49abf7bea3d9b30a075ff247 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Wed, 30 Nov 2022 16:11:30 -0500
Subject: [PATCH 1608/1791] drm/connector: Print connector colorspace in state
 debugfs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v3: Fix kerneldocs (kernel test robot)

v4: Avoid returning NULL from drm_get_colorspace_name

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Simon Ser <contact@emersion.fr>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/drm_atomic.c    |  1 +
 drivers/gpu/drm/drm_connector.c | 15 +++++++++++++++
 include/drm/drm_connector.h     |  1 +
 3 files changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index b4c6ffc438da2..2c454568a607c 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -1131,6 +1131,7 @@ static void drm_atomic_connector_print_state(struct drm_printer *p,
 	drm_printf(p, "\tcrtc=%s\n", state->crtc ? state->crtc->name : "(null)");
 	drm_printf(p, "\tself_refresh_aware=%d\n", state->self_refresh_aware);
 	drm_printf(p, "\tmax_requested_bpc=%d\n", state->max_requested_bpc);
+	drm_printf(p, "\tcolorspace=%s\n", drm_get_colorspace_name(state->colorspace));
 
 	if (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK)
 		if (state->writeback_job && state->writeback_job->fb)
diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index b823b10ed6974..e1d7a40e17763 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -1087,6 +1087,21 @@ static const char * const colorspace_names[] = {
 	[DRM_MODE_COLORIMETRY_BT601_YCC] = "BT601_YCC",
 };
 
+/**
+ * drm_get_colorspace_name - return a string for color encoding
+ * @colorspace: color space to compute name of
+ *
+ * In contrast to the other drm_get_*_name functions this one here returns a
+ * const pointer and hence is threadsafe.
+ */
+const char *drm_get_colorspace_name(enum drm_colorspace colorspace)
+{
+	if (colorspace < ARRAY_SIZE(colorspace_names) && colorspace_names[colorspace])
+		return colorspace_names[colorspace];
+	else
+		return "(null)";
+}
+
 static const u32 hdmi_colorspaces =
 	BIT(DRM_MODE_COLORIMETRY_SMPTE_170M_YCC) |
 	BIT(DRM_MODE_COLORIMETRY_BT709_YCC) |
diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index 6eb8e2d5d20f2..880220321867d 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -2078,6 +2078,7 @@ void drm_connector_list_iter_end(struct drm_connector_list_iter *iter);
 
 bool drm_connector_has_possible_encoder(struct drm_connector *connector,
 					struct drm_encoder *encoder);
+const char *drm_get_colorspace_name(enum drm_colorspace colorspace);
 
 /**
  * drm_for_each_connector_iter - connector_list iterator macro
-- 
GitLab


From c265f340eaa87aa5f979adfb23d7463af67b7f27 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Tue, 29 Nov 2022 15:16:31 -0500
Subject: [PATCH 1609/1791] drm/connector: Allow drivers to pass list of
 supported colorspaces
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Drivers might not support all colorspaces defined in
dp_colorspaces and hdmi_colorspaces. This results in
undefined behavior when userspace is setting an
unsupported colorspace.

Allow drivers to pass the list of supported colorspaces
when creating the colorspace property.

v2:
 - Use 0 to indicate support for all colorspaces (Jani)
 - Print drm_dbg_kms message when drivers pass 0
   to signal that drivers should specify supported
   colorspaecs explicity (Jani)

v3:
 - Move changes to create a common colorspace_names array
   to separate patch

v6:
- Avoid magic when passing 0 for supported_colorspaces;
  be explicit in treating it as "all DP/HDMI"

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Sebastian Wick <sebastian.wick@redhat.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>
Reviewed-by: Simon Ser <contact@emersion.fr>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/drm_connector.c               | 24 +++++++++++++++----
 .../gpu/drm/i915/display/intel_connector.c    |  4 ++--
 drivers/gpu/drm/vc4/vc4_hdmi.c                |  2 +-
 include/drm/drm_connector.h                   |  7 ++++--
 4 files changed, 28 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c
index e1d7a40e17763..3ed4cfcb350c4 100644
--- a/drivers/gpu/drm/drm_connector.c
+++ b/drivers/gpu/drm/drm_connector.c
@@ -2210,9 +2210,17 @@ static int drm_mode_create_colorspace_property(struct drm_connector *connector,
  * Returns:
  * Zero on success, negative errno on failure.
  */
-int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector)
+int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector,
+					     u32 supported_colorspaces)
 {
-	return drm_mode_create_colorspace_property(connector, hdmi_colorspaces);
+	u32 colorspaces;
+
+	if (supported_colorspaces)
+		colorspaces = supported_colorspaces & hdmi_colorspaces;
+	else
+		colorspaces = hdmi_colorspaces;
+
+	return drm_mode_create_colorspace_property(connector, colorspaces);
 }
 EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
 
@@ -2226,9 +2234,17 @@ EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property);
  * Returns:
  * Zero on success, negative errno on failure.
  */
-int drm_mode_create_dp_colorspace_property(struct drm_connector *connector)
+int drm_mode_create_dp_colorspace_property(struct drm_connector *connector,
+					   u32 supported_colorspaces)
 {
-	return drm_mode_create_colorspace_property(connector, dp_colorspaces);
+	u32 colorspaces;
+
+	if (supported_colorspaces)
+		colorspaces = supported_colorspaces & dp_colorspaces;
+	else
+		colorspaces = dp_colorspaces;
+
+	return drm_mode_create_colorspace_property(connector, colorspaces);
 }
 EXPORT_SYMBOL(drm_mode_create_dp_colorspace_property);
 
diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c
index 257afac348397..3abfe29d0b44b 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.c
+++ b/drivers/gpu/drm/i915/display/intel_connector.c
@@ -280,14 +280,14 @@ intel_attach_aspect_ratio_property(struct drm_connector *connector)
 void
 intel_attach_hdmi_colorspace_property(struct drm_connector *connector)
 {
-	if (!drm_mode_create_hdmi_colorspace_property(connector))
+	if (!drm_mode_create_hdmi_colorspace_property(connector, 0))
 		drm_connector_attach_colorspace_property(connector);
 }
 
 void
 intel_attach_dp_colorspace_property(struct drm_connector *connector)
 {
-	if (!drm_mode_create_dp_colorspace_property(connector))
+	if (!drm_mode_create_dp_colorspace_property(connector, 0))
 		drm_connector_attach_colorspace_property(connector);
 }
 
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 06713d8b82b5d..482397d5cb48d 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -631,7 +631,7 @@ static int vc4_hdmi_connector_init(struct drm_device *dev,
 	if (ret)
 		return ret;
 
-	ret = drm_mode_create_hdmi_colorspace_property(connector);
+	ret = drm_mode_create_hdmi_colorspace_property(connector, 0);
 	if (ret)
 		return ret;
 
diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h
index 880220321867d..3cea003462059 100644
--- a/include/drm/drm_connector.h
+++ b/include/drm/drm_connector.h
@@ -30,6 +30,7 @@
 #include <linux/notifier.h>
 #include <drm/drm_mode_object.h>
 #include <drm/drm_util.h>
+#include <drm/drm_property.h>
 
 #include <uapi/drm/drm_mode.h>
 
@@ -1994,8 +1995,10 @@ int drm_connector_attach_hdr_output_metadata_property(struct drm_connector *conn
 bool drm_connector_atomic_hdr_metadata_equal(struct drm_connector_state *old_state,
 					     struct drm_connector_state *new_state);
 int drm_mode_create_aspect_ratio_property(struct drm_device *dev);
-int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector);
-int drm_mode_create_dp_colorspace_property(struct drm_connector *connector);
+int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector,
+					     u32 supported_colorspaces);
+int drm_mode_create_dp_colorspace_property(struct drm_connector *connector,
+					   u32 supported_colorspaces);
 int drm_mode_create_content_type_property(struct drm_device *dev);
 int drm_mode_create_suggested_offset_properties(struct drm_device *dev);
 
-- 
GitLab


From cb841d27b8767fd88096d06186b5f5de990fd6d0 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Fri, 1 Apr 2022 13:45:29 -0400
Subject: [PATCH 1610/1791] drm/amd/display: Always pass connector_state to
 stream validation

We need the connector_state for colorspace and scaling information
and can get it from connector->state.

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index d0532d7e83414..dd66455a8f36a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -5955,15 +5955,14 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 {
 	struct drm_display_mode *preferred_mode = NULL;
 	struct drm_connector *drm_connector;
-	const struct drm_connector_state *con_state =
-		dm_state ? &dm_state->base : NULL;
+	const struct drm_connector_state *con_state = &dm_state->base;
 	struct dc_stream_state *stream = NULL;
 	struct drm_display_mode mode;
 	struct drm_display_mode saved_mode;
 	struct drm_display_mode *freesync_mode = NULL;
 	bool native_mode_found = false;
 	bool recalculate_timing = false;
-	bool scale = dm_state ? (dm_state->scaling != RMX_OFF) : false;
+	bool scale = dm_state->scaling != RMX_OFF;
 	int mode_refresh;
 	int preferred_refresh = 0;
 	enum color_transfer_func tf = TRANSFER_FUNC_UNKNOWN;
@@ -6605,7 +6604,9 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec
 		goto fail;
 	}
 
-	stream = create_validate_stream_for_sink(aconnector, mode, NULL, NULL);
+	stream = create_validate_stream_for_sink(aconnector, mode,
+						 to_dm_connector_state(connector->state),
+						 NULL);
 	if (stream) {
 		dc_stream_release(stream);
 		result = MODE_OK;
-- 
GitLab


From 15f9dfd545a1edd604648961feadce16791d0f4f Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Fri, 25 Mar 2022 15:30:28 -0400
Subject: [PATCH 1611/1791] drm/amd/display: Register Colorspace property for
 DP and HDMI
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We want compositors to be able to set the output
colorspace on DP and HDMI outputs, based on the
caps reported from the receiver via EDID.

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index dd66455a8f36a..8e2090280f449 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7247,6 +7247,12 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector)
 	return amdgpu_dm_connector->num_modes;
 }
 
+static const u32 supported_colorspaces =
+	BIT(DRM_MODE_COLORIMETRY_BT709_YCC) |
+	BIT(DRM_MODE_COLORIMETRY_OPRGB) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) |
+	BIT(DRM_MODE_COLORIMETRY_BT2020_YCC);
+
 void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 				     struct amdgpu_dm_connector *aconnector,
 				     int connector_type,
@@ -7327,6 +7333,15 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 				adev->mode_info.abm_level_property, 0);
 	}
 
+	if (connector_type == DRM_MODE_CONNECTOR_HDMIA) {
+		if (!drm_mode_create_hdmi_colorspace_property(&aconnector->base, supported_colorspaces))
+			drm_connector_attach_colorspace_property(&aconnector->base);
+	} else if (connector_type == DRM_MODE_CONNECTOR_DisplayPort ||
+		   connector_type == DRM_MODE_CONNECTOR_eDP) {
+		if (!drm_mode_create_dp_colorspace_property(&aconnector->base, supported_colorspaces))
+			drm_connector_attach_colorspace_property(&aconnector->base);
+	}
+
 	if (connector_type == DRM_MODE_CONNECTOR_HDMIA ||
 	    connector_type == DRM_MODE_CONNECTOR_DisplayPort ||
 	    connector_type == DRM_MODE_CONNECTOR_eDP) {
-- 
GitLab


From a0b433c858ac1d2e03cbfd5bb34b9b61906600eb Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Tue, 29 Mar 2022 11:26:23 -0400
Subject: [PATCH 1612/1791] drm/amd/display: Signal mode_changed if colorspace
 changed

We need to signal mode_changed to make sure we update the output
colorspace.

v2: No need to call drm_hdmi_avi_infoframe_colorimetry as DC does its
    own infoframe packing.

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Uma Shankar <uma.shankar@intel.com>
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 8e2090280f449..49be49c85c502 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6700,6 +6700,14 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn,
 	if (!crtc)
 		return 0;
 
+	if (new_con_state->colorspace != old_con_state->colorspace) {
+		new_crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(new_crtc_state))
+			return PTR_ERR(new_crtc_state);
+
+		new_crtc_state->mode_changed = true;
+	}
+
 	if (!drm_connector_atomic_hdr_metadata_equal(old_con_state, new_con_state)) {
 		struct dc_info_packet hdr_infopacket;
 
@@ -6722,7 +6730,7 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn,
 		 * set is permissible, however. So only force a
 		 * modeset if we're entering or exiting HDR.
 		 */
-		new_crtc_state->mode_changed =
+		new_crtc_state->mode_changed = new_crtc_state->mode_changed ||
 			!old_con_state->hdr_output_metadata ||
 			!new_con_state->hdr_output_metadata;
 	}
-- 
GitLab


From 2e656827ceed9fb1ba406e7cd11d7b572010add0 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Tue, 29 Mar 2022 15:30:05 -0400
Subject: [PATCH 1613/1791] drm/amd/display: Send correct DP colorspace
 infopacket

Look at connector->colorimetry to determine output colorspace.

We don't want to impact current SDR behavior, so
DRM_MODE_COLORIMETRY_DEFAULT preserves current behavior.

Also add support to explicitly set BT601 and BT709.

v4:
- Roll support for BT709 and BT601 into this patch
- Add default case to avoid warnings for unhandled
  enum values

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 48 ++++++++++++-------
 1 file changed, 31 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 49be49c85c502..67dad847fd042 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -5339,21 +5339,44 @@ get_aspect_ratio(const struct drm_display_mode *mode_in)
 }
 
 static enum dc_color_space
-get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing)
+get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing,
+		       const struct drm_connector_state *connector_state)
 {
 	enum dc_color_space color_space = COLOR_SPACE_SRGB;
 
-	switch (dc_crtc_timing->pixel_encoding)	{
-	case PIXEL_ENCODING_YCBCR422:
-	case PIXEL_ENCODING_YCBCR444:
-	case PIXEL_ENCODING_YCBCR420:
-	{
+	switch (connector_state->colorspace) {
+	case DRM_MODE_COLORIMETRY_BT601_YCC:
+		if (dc_crtc_timing->flags.Y_ONLY)
+			color_space = COLOR_SPACE_YCBCR601_LIMITED;
+		else
+			color_space = COLOR_SPACE_YCBCR601;
+		break;
+	case DRM_MODE_COLORIMETRY_BT709_YCC:
+		if (dc_crtc_timing->flags.Y_ONLY)
+			color_space = COLOR_SPACE_YCBCR709_LIMITED;
+		else
+			color_space = COLOR_SPACE_YCBCR709;
+		break;
+	case DRM_MODE_COLORIMETRY_OPRGB:
+		color_space = COLOR_SPACE_ADOBERGB;
+		break;
+	case DRM_MODE_COLORIMETRY_BT2020_RGB:
+	case DRM_MODE_COLORIMETRY_BT2020_YCC:
+		if (dc_crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB)
+			color_space = COLOR_SPACE_2020_RGB_FULLRANGE;
+		else
+			color_space = COLOR_SPACE_2020_YCBCR;
+		break;
+	case DRM_MODE_COLORIMETRY_DEFAULT: // ITU601
+	default:
+		if (dc_crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB) {
+			color_space = COLOR_SPACE_SRGB;
 		/*
 		 * 27030khz is the separation point between HDTV and SDTV
 		 * according to HDMI spec, we use YCbCr709 and YCbCr601
 		 * respectively
 		 */
-		if (dc_crtc_timing->pix_clk_100hz > 270300) {
+		} else if (dc_crtc_timing->pix_clk_100hz > 270300) {
 			if (dc_crtc_timing->flags.Y_ONLY)
 				color_space =
 					COLOR_SPACE_YCBCR709_LIMITED;
@@ -5366,15 +5389,6 @@ get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing)
 			else
 				color_space = COLOR_SPACE_YCBCR601;
 		}
-
-	}
-	break;
-	case PIXEL_ENCODING_RGB:
-		color_space = COLOR_SPACE_SRGB;
-		break;
-
-	default:
-		WARN_ON(1);
 		break;
 	}
 
@@ -5513,7 +5527,7 @@ static void fill_stream_properties_from_drm_display_mode(
 		}
 	}
 
-	stream->output_color_space = get_output_color_space(timing_out);
+	stream->output_color_space = get_output_color_space(timing_out, connector_state);
 }
 
 static void fill_audio_info(struct audio_info *audio_info,
-- 
GitLab


From bd49f19039c1806cd10cff8aaec7f90ebf28f0e9 Mon Sep 17 00:00:00 2001
From: Joshua Ashton <joshua@froggi.es>
Date: Mon, 14 Nov 2022 19:52:30 +0000
Subject: [PATCH 1614/1791] drm/amd/display: Always set crtcinfo from
 create_stream_for_sink

Given that we always pass dm_state into here now, this won't ever
trigger anymore.

This is needed for we will always fail mode validation with invalid
clocks or link bandwidth errors.

Signed-off-by: Joshua Ashton <joshua@froggi.es>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 67dad847fd042..a6bc2c6cad5c1 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6055,7 +6055,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 
 	if (recalculate_timing)
 		drm_mode_set_crtcinfo(&saved_mode, 0);
-	else if (!dm_state)
+	else
 		drm_mode_set_crtcinfo(&mode, 0);
 
 	/*
-- 
GitLab


From fd45b6540f513887c172e2082d437209fe8f4a54 Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Tue, 29 Nov 2022 17:24:52 -0500
Subject: [PATCH 1615/1791] drm/amd/display: Add debugfs for testing output
 colorspace

In order to IGT test colorspace we'll want to print
the currently enabled colorspace on a stream. We add
a new debugfs to do so, using the same scheme as
current bpc reporting.

This might also come in handy when debugging display
issues.

v4:
- Fix function doc comment
- Fix sRGB debug print

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Joshua Ashton <joshua@froggi.es>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 57 +++++++++++++++++++
 1 file changed, 57 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index 82234397dd44e..caf13b2e8cb64 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -906,6 +906,61 @@ static int amdgpu_current_bpc_show(struct seq_file *m, void *data)
 }
 DEFINE_SHOW_ATTRIBUTE(amdgpu_current_bpc);
 
+/*
+ * Returns the current colorspace for the crtc.
+ * Example usage: cat /sys/kernel/debug/dri/0/crtc-0/amdgpu_current_colorspace
+ */
+static int amdgpu_current_colorspace_show(struct seq_file *m, void *data)
+{
+	struct drm_crtc *crtc = m->private;
+	struct drm_device *dev = crtc->dev;
+	struct dm_crtc_state *dm_crtc_state = NULL;
+	int res = -ENODEV;
+
+	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
+	if (crtc->state == NULL)
+		goto unlock;
+
+	dm_crtc_state = to_dm_crtc_state(crtc->state);
+	if (dm_crtc_state->stream == NULL)
+		goto unlock;
+
+	switch (dm_crtc_state->stream->output_color_space) {
+	case COLOR_SPACE_SRGB:
+		seq_printf(m, "sRGB");
+		break;
+	case COLOR_SPACE_YCBCR601:
+	case COLOR_SPACE_YCBCR601_LIMITED:
+		seq_printf(m, "BT601_YCC");
+		break;
+	case COLOR_SPACE_YCBCR709:
+	case COLOR_SPACE_YCBCR709_LIMITED:
+		seq_printf(m, "BT709_YCC");
+		break;
+	case COLOR_SPACE_ADOBERGB:
+		seq_printf(m, "opRGB");
+		break;
+	case COLOR_SPACE_2020_RGB_FULLRANGE:
+		seq_printf(m, "BT2020_RGB");
+		break;
+	case COLOR_SPACE_2020_YCBCR:
+		seq_printf(m, "BT2020_YCC");
+		break;
+	default:
+		goto unlock;
+	}
+	res = 0;
+
+unlock:
+	drm_modeset_unlock(&crtc->mutex);
+	mutex_unlock(&dev->mode_config.mutex);
+
+	return res;
+}
+DEFINE_SHOW_ATTRIBUTE(amdgpu_current_colorspace);
+
+
 /*
  * Example usage:
  * Disable dsc passthrough, i.e.,: have dsc decoding at converver, not external RX
@@ -3139,6 +3194,8 @@ void crtc_debugfs_init(struct drm_crtc *crtc)
 #endif
 	debugfs_create_file("amdgpu_current_bpc", 0644, crtc->debugfs_entry,
 			    crtc, &amdgpu_current_bpc_fops);
+	debugfs_create_file("amdgpu_current_colorspace", 0644, crtc->debugfs_entry,
+			    crtc, &amdgpu_current_colorspace_fops);
 }
 
 /*
-- 
GitLab


From eaa7d8301109092670c5cf3e12c502618d6adc51 Mon Sep 17 00:00:00 2001
From: Joshua Ashton <joshua@froggi.es>
Date: Tue, 10 Jan 2023 19:14:06 +0000
Subject: [PATCH 1616/1791] drm/amd/display: Refactor avi_info_frame
 colorimetry determination

Replace the messy two if-else chains here that were
on the same value with a switch on the enum.

Signed-off-by: Joshua Ashton <joshua@froggi.es>
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>

Cc: Pekka Paalanen <ppaalanen@gmail.com>
Cc: Sebastian Wick <sebastian.wick@redhat.com>
Cc: Vitaly.Prosyak@amd.com
Cc: Joshua Ashton <joshua@froggi.es>
Cc: Simon Ser <contact@emersion.fr>
Cc: Melissa Wen <mwen@igalia.com>
Cc: dri-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/core/dc_resource.c | 28 +++++++++++--------
 1 file changed, 17 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index c72540d37aefd..2f3d9a698486d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -3035,23 +3035,29 @@ static void set_avi_info_frame(
 	hdmi_info.bits.S0_S1 = scan_type;
 
 	/* C0, C1 : Colorimetry */
-	if (color_space == COLOR_SPACE_YCBCR709 ||
-			color_space == COLOR_SPACE_YCBCR709_LIMITED)
+	switch (color_space) {
+	case COLOR_SPACE_YCBCR709:
+	case COLOR_SPACE_YCBCR709_LIMITED:
 		hdmi_info.bits.C0_C1 = COLORIMETRY_ITU709;
-	else if (color_space == COLOR_SPACE_YCBCR601 ||
-			color_space == COLOR_SPACE_YCBCR601_LIMITED)
+		break;
+	case COLOR_SPACE_YCBCR601:
+	case COLOR_SPACE_YCBCR601_LIMITED:
 		hdmi_info.bits.C0_C1 = COLORIMETRY_ITU601;
-	else {
-		hdmi_info.bits.C0_C1 = COLORIMETRY_NO_DATA;
-	}
-	if (color_space == COLOR_SPACE_2020_RGB_FULLRANGE ||
-			color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE ||
-			color_space == COLOR_SPACE_2020_YCBCR) {
+		break;
+	case COLOR_SPACE_2020_RGB_FULLRANGE:
+	case COLOR_SPACE_2020_RGB_LIMITEDRANGE:
+	case COLOR_SPACE_2020_YCBCR:
 		hdmi_info.bits.EC0_EC2 = COLORIMETRYEX_BT2020RGBYCBCR;
 		hdmi_info.bits.C0_C1   = COLORIMETRY_EXTENDED;
-	} else if (color_space == COLOR_SPACE_ADOBERGB) {
+		break;
+	case COLOR_SPACE_ADOBERGB:
 		hdmi_info.bits.EC0_EC2 = COLORIMETRYEX_ADOBERGB;
 		hdmi_info.bits.C0_C1   = COLORIMETRY_EXTENDED;
+		break;
+	case COLOR_SPACE_SRGB:
+	default:
+		hdmi_info.bits.C0_C1 = COLORIMETRY_NO_DATA;
+		break;
 	}
 
 	if (pixel_encoding && color_space == COLOR_SPACE_2020_YCBCR &&
-- 
GitLab


From 5daff15cd013422bc6d1efcfe82b586800025384 Mon Sep 17 00:00:00 2001
From: Lang Yu <Lang.Yu@amd.com>
Date: Sat, 6 May 2023 12:01:33 +0800
Subject: [PATCH 1617/1791] drm/amdgpu: unmap and remove csa_va properly
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Root PD BO should be reserved before unmap and remove
a bo_va from VM otherwise lockdep will complain.

v2: check fpriv->csa_va is not NULL instead of amdgpu_mcbp (christian)

[14616.936827] WARNING: CPU: 6 PID: 1711 at drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:1762 amdgpu_vm_bo_del+0x399/0x3f0 [amdgpu]
[14616.937096] Call Trace:
[14616.937097]  <TASK>
[14616.937102]  amdgpu_driver_postclose_kms+0x249/0x2f0 [amdgpu]
[14616.937187]  drm_file_free+0x1d6/0x300 [drm]
[14616.937207]  drm_close_helper.isra.0+0x62/0x70 [drm]
[14616.937220]  drm_release+0x5e/0x100 [drm]
[14616.937234]  __fput+0x9f/0x280
[14616.937239]  ____fput+0xe/0x20
[14616.937241]  task_work_run+0x61/0x90
[14616.937246]  exit_to_user_mode_prepare+0x215/0x220
[14616.937251]  syscall_exit_to_user_mode+0x2a/0x60
[14616.937254]  do_syscall_64+0x48/0x90
[14616.937257]  entry_SYSCALL_64_after_hwframe+0x63/0xcd

Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 38 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h |  3 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 10 +++----
 3 files changed, 46 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index c6d4d41c4393e..23d054526e7c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -106,3 +106,41 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	ttm_eu_backoff_reservation(&ticket, &list);
 	return 0;
 }
+
+int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			    struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va,
+			    uint64_t csa_addr)
+{
+	struct ww_acquire_ctx ticket;
+	struct list_head list;
+	struct amdgpu_bo_list_entry pd;
+	struct ttm_validate_buffer csa_tv;
+	int r;
+
+	INIT_LIST_HEAD(&list);
+	INIT_LIST_HEAD(&csa_tv.head);
+	csa_tv.bo = &bo->tbo;
+	csa_tv.num_shared = 1;
+
+	list_add(&csa_tv.head, &list);
+	amdgpu_vm_get_pd_bo(vm, &list, &pd);
+
+	r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL);
+	if (r) {
+		DRM_ERROR("failed to reserve CSA,PD BOs: err=%d\n", r);
+		return r;
+	}
+
+	r = amdgpu_vm_bo_unmap(adev, bo_va, csa_addr);
+	if (r) {
+		DRM_ERROR("failed to do bo_unmap on static CSA, err=%d\n", r);
+		ttm_eu_backoff_reservation(&ticket, &list);
+		return r;
+	}
+
+	amdgpu_vm_bo_del(adev, bo_va);
+
+	ttm_eu_backoff_reservation(&ticket, &list);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h
index 524b4437a0217..7dfc1f2012ebf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h
@@ -34,6 +34,9 @@ int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo
 int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			  struct amdgpu_bo *bo, struct amdgpu_bo_va **bo_va,
 			  uint64_t csa_addr, uint32_t size);
+int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			    struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va,
+			    uint64_t csa_addr);
 void amdgpu_free_static_csa(struct amdgpu_bo **bo);
 
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 41d047e5de696..e3531aa3c8bd1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -1311,12 +1311,12 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
 	if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL)
 		amdgpu_vce_free_handles(adev, file_priv);
 
-	if (amdgpu_mcbp) {
-		/* TODO: how to handle reserve failure */
-		BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true));
-		amdgpu_vm_bo_del(adev, fpriv->csa_va);
+	if (fpriv->csa_va) {
+		uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK;
+
+		WARN_ON(amdgpu_unmap_static_csa(adev, &fpriv->vm, adev->virt.csa_obj,
+						fpriv->csa_va, csa_addr));
 		fpriv->csa_va = NULL;
-		amdgpu_bo_unreserve(adev->virt.csa_obj);
 	}
 
 	pasid = fpriv->vm.pasid;
-- 
GitLab


From 731b48463b0d96eda1f1684eacde6e9c8065df83 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Mon, 5 Jun 2023 11:39:30 +0800
Subject: [PATCH 1618/1791] drm/amdgpu: disable virtual display support on APP
 device

virtual display is not support on APP device.

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Gavin Wan <gavin.wan@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 2c1fbed245353..0f1ca0136f50c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -56,7 +56,8 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 
 	/* enable virtual display */
 	if (adev->asic_type != CHIP_ALDEBARAN &&
-	    adev->asic_type != CHIP_ARCTURUS) {
+	    adev->asic_type != CHIP_ARCTURUS &&
+	    ((adev->pdev->class >> 8) != AMD_ACCELERATOR_PROCESSING)) {
 		if (adev->mode_info.num_crtc == 0)
 			adev->mode_info.num_crtc = 1;
 		adev->enable_virtual_display = true;
-- 
GitLab


From cab69d36ccdbfa3fa0b5627a032150369c20b4f3 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Tue, 6 Jun 2023 14:42:39 +0800
Subject: [PATCH 1619/1791] drm/amdgpu: skip to resume rlcg for gc 9.4.3 in vf
 side

skip to resume rlcg, because rlcg is already enabled in pf side.

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index ebdd7fa985d60..f5b8d3f388ff6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1293,6 +1293,9 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev)
 {
 	int r, i, num_xcc;
 
+	if (amdgpu_sriov_vf(adev))
+		return 0;
+
 	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
 	for (i = 0; i < num_xcc; i++) {
 		r = gfx_v9_4_3_xcc_rlc_resume(adev, i);
@@ -4321,11 +4324,13 @@ static int gfx_v9_4_3_xcp_resume(void *handle, uint32_t inst_mask)
 	for_each_inst(i, tmp_mask)
 		gfx_v9_4_3_xcc_constants_init(adev, i);
 
-	tmp_mask = inst_mask;
-	for_each_inst(i, tmp_mask) {
-		r = gfx_v9_4_3_xcc_rlc_resume(adev, i);
-		if (r)
-			return r;
+	if (!amdgpu_sriov_vf(adev)) {
+		tmp_mask = inst_mask;
+		for_each_inst(i, tmp_mask) {
+			r = gfx_v9_4_3_xcc_rlc_resume(adev, i);
+			if (r)
+				return r;
+		}
 	}
 
 	tmp_mask = inst_mask;
-- 
GitLab


From 27d196c4491458ca00014cfe1cfa9d0fa87a2ff9 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Wed, 7 Jun 2023 11:24:48 +0800
Subject: [PATCH 1620/1791] drm/amd/pm: fix vclk setting failed for SMU v13.0.4

PMFW use the left-shifted 16 bits argument to set the VCLK
DPM frequency for SMU v13.0.4.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h         |  2 ++
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c | 11 +++++++++--
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
index 5a99a091965e3..6a0ac0bbaace9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h
@@ -51,6 +51,8 @@
 #define CTF_OFFSET_HOTSPOT		5
 #define CTF_OFFSET_MEM			5
 
+#define SMU_13_VCLK_SHIFT		16
+
 extern const int pmfw_decoded_link_speed[5];
 extern const int pmfw_decoded_link_width[7];
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
index 46a8a366f2879..999b07db862e5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
@@ -831,6 +831,8 @@ static int smu_v13_0_4_set_soft_freq_limited_range(struct smu_context *smu,
 						   uint32_t max)
 {
 	enum smu_message_type msg_set_min, msg_set_max;
+	uint32_t min_clk = min;
+	uint32_t max_clk = max;
 	int ret = 0;
 
 	if (!smu_v13_0_4_clk_dpm_is_enabled(smu, clk_type))
@@ -859,12 +861,17 @@ static int smu_v13_0_4_set_soft_freq_limited_range(struct smu_context *smu,
 		return -EINVAL;
 	}
 
-	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL);
+	if (clk_type == SMU_VCLK) {
+		min_clk = min << SMU_13_VCLK_SHIFT;
+		max_clk = max << SMU_13_VCLK_SHIFT;
+	}
+
+	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL);
 	if (ret)
 		return ret;
 
 	return smu_cmn_send_smc_msg_with_param(smu, msg_set_max,
-					       max, NULL);
+					       max_clk, NULL);
 }
 
 static int smu_v13_0_4_force_clk_levels(struct smu_context *smu,
-- 
GitLab


From 2d0ee64e9846ed4036fd11c5b900a21039ee8b7a Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Fri, 2 Jun 2023 16:57:12 +0800
Subject: [PATCH 1621/1791] drm/amd/pm: enable vclk and dclk Pstates for SMU
 v13.0.4

Add the ability to control the vclk and dclk frequency by
power_dpm_force_performance_level interface.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c  | 29 +++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
index 999b07db862e5..315a6d8bde2e3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
@@ -915,6 +915,8 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu,
 	uint32_t sclk_min = 0, sclk_max = 0;
 	uint32_t fclk_min = 0, fclk_max = 0;
 	uint32_t socclk_min = 0, socclk_max = 0;
+	uint32_t vclk_min = 0, vclk_max = 0;
+	uint32_t dclk_min = 0, dclk_max = 0;
 	int ret = 0;
 
 	switch (level) {
@@ -922,22 +924,32 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu,
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &fclk_max);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &socclk_max);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max);
 		sclk_min = sclk_max;
 		fclk_min = fclk_max;
 		socclk_min = socclk_max;
+		vclk_min = vclk_max;
+		dclk_min = dclk_max;
 		break;
 	case AMD_DPM_FORCED_LEVEL_LOW:
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, NULL);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, NULL);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL);
 		sclk_max = sclk_min;
 		fclk_max = fclk_min;
 		socclk_max = socclk_min;
+		vclk_max = vclk_min;
+		dclk_max = dclk_min;
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, &fclk_max);
 		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, &socclk_max);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max);
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
 	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
@@ -983,6 +995,23 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu,
 			return ret;
 	}
 
+	if (vclk_min && vclk_max) {
+		ret = smu_v13_0_4_set_soft_freq_limited_range(smu,
+							      SMU_VCLK,
+							      vclk_min,
+							      vclk_max);
+		if (ret)
+			return ret;
+	}
+
+	if (dclk_min && dclk_max) {
+		ret = smu_v13_0_4_set_soft_freq_limited_range(smu,
+							      SMU_DCLK,
+							      dclk_min,
+							      dclk_max);
+		if (ret)
+			return ret;
+	}
 	return ret;
 }
 
-- 
GitLab


From 55682a893844cc64e3a85806b0c3ca7a77b905c3 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Mon, 5 Jun 2023 16:57:45 +0800
Subject: [PATCH 1622/1791] drm/amd/pm: enable more Pstates profile levels for
 SMU v13.0.4

This patch enables following UMD stable Pstates profile
levels for power_dpm_force_performance_level interface.

- profile_peak
- profile_min_mclk
- profile_min_sclk
- profile_standard

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c  | 54 ++++++++++++++++++-
 1 file changed, 53 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
index 315a6d8bde2e3..ef37dda9908fa 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c
@@ -54,6 +54,10 @@
 
 #define FEATURE_MASK(feature) (1ULL << feature)
 
+#define SMU_13_0_4_UMD_PSTATE_GFXCLK			938
+#define SMU_13_0_4_UMD_PSTATE_SOCCLK			938
+#define SMU_13_0_4_UMD_PSTATE_FCLK			1875
+
 #define SMC_DPM_FEATURE ( \
 	FEATURE_MASK(FEATURE_CCLK_DPM_BIT) | \
 	FEATURE_MASK(FEATURE_VCN_DPM_BIT)	 | \
@@ -908,6 +912,50 @@ static int smu_v13_0_4_force_clk_levels(struct smu_context *smu,
 	return ret;
 }
 
+static int smu_v13_0_4_get_dpm_profile_freq(struct smu_context *smu,
+					enum amd_dpm_forced_level level,
+					enum smu_clk_type clk_type,
+					uint32_t *min_clk,
+					uint32_t *max_clk)
+{
+	int ret = 0;
+	uint32_t clk_limit = 0;
+
+	switch (clk_type) {
+	case SMU_GFXCLK:
+	case SMU_SCLK:
+		clk_limit = SMU_13_0_4_UMD_PSTATE_GFXCLK;
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit);
+		else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK)
+			smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL);
+		break;
+	case SMU_SOCCLK:
+		clk_limit = SMU_13_0_4_UMD_PSTATE_SOCCLK;
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &clk_limit);
+		break;
+	case SMU_FCLK:
+		clk_limit = SMU_13_0_4_UMD_PSTATE_FCLK;
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &clk_limit);
+		else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK)
+			smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &clk_limit, NULL);
+		break;
+	case SMU_VCLK:
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit);
+		break;
+	case SMU_DCLK:
+		smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	*min_clk = *max_clk = clk_limit;
+	return ret;
+}
+
 static int smu_v13_0_4_set_performance_level(struct smu_context *smu,
 					     enum amd_dpm_forced_level level)
 {
@@ -955,7 +1003,11 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu,
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
-		/* Temporarily do nothing since the optimal clocks haven't been provided yet */
+		smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max);
+		smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_FCLK, &fclk_min, &fclk_max);
+		smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_SOCCLK, &socclk_min, &socclk_max);
+		smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max);
+		smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
 	case AMD_DPM_FORCED_LEVEL_MANUAL:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT:
-- 
GitLab


From 3537d6a48c50ed37e419f89931a5acdb6c56c6d6 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 7 Jun 2023 01:41:22 -0500
Subject: [PATCH 1623/1791] drm/amd: Make sure image is written to trigger
 VBIOS image update flow

The VBIOS image update flow requires userspace to:
1) Write the image to `psp_vbflash`
2) Read `psp_vbflash`
3) Poll `psp_vbflash_status` to check for completion

If userspace reads `psp_vbflash` before writing an image, it's
possible that it causes problems that can put the dGPU into an invalid
state.

Explicitly check that an image has been written before letting a read
succeed.

Cc: stable@vger.kernel.org
Fixes: 8424f2ccb3c0 ("drm/amdgpu/psp: Add vbflash sysfs interface support")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 27559a05a67b2..8c60db1761196 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -3617,6 +3617,9 @@ static ssize_t amdgpu_psp_vbflash_read(struct file *filp, struct kobject *kobj,
 	void *fw_pri_cpu_addr;
 	int ret;
 
+	if (adev->psp.vbflash_image_size == 0)
+		return -EINVAL;
+
 	dev_info(adev->dev, "VBIOS flash to PSP started");
 
 	ret = amdgpu_bo_create_kernel(adev, adev->psp.vbflash_image_size,
-- 
GitLab


From fe56c6ee0457035ae8fbbc2aa5ddfcfac6ded787 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 7 Jun 2023 01:45:20 -0500
Subject: [PATCH 1624/1791] drm/amd: Tighten permissions on VBIOS flashing
 attributes

Non-root users shouldn't be able to try to trigger a VBIOS flash
or query the flashing status.  This should be reserved for users with the
appropriate permissions.

Cc: stable@vger.kernel.org
Fixes: 8424f2ccb3c0 ("drm/amdgpu/psp: Add vbflash sysfs interface support")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 8c60db1761196..dd865beb39a8c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -3671,13 +3671,13 @@ static ssize_t amdgpu_psp_vbflash_status(struct device *dev,
 }
 
 static const struct bin_attribute psp_vbflash_bin_attr = {
-	.attr = {.name = "psp_vbflash", .mode = 0664},
+	.attr = {.name = "psp_vbflash", .mode = 0660},
 	.size = 0,
 	.write = amdgpu_psp_vbflash_write,
 	.read = amdgpu_psp_vbflash_read,
 };
 
-static DEVICE_ATTR(psp_vbflash_status, 0444, amdgpu_psp_vbflash_status, NULL);
+static DEVICE_ATTR(psp_vbflash_status, 0440, amdgpu_psp_vbflash_status, NULL);
 
 int amdgpu_psp_sysfs_init(struct amdgpu_device *adev)
 {
-- 
GitLab


From 09d49e14ea6fd125a21f89b80f888c09be32a174 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 23 May 2023 11:57:27 -0400
Subject: [PATCH 1625/1791] drm/amdkfd: fix and enable debugging for gfx11

There are a couple of fixes required to enable gfx11 debugging.

First, ADD_QUEUE.trap_en is an inappropriate place to toggle
a per-process register so move it to SET_SHADER_DEBUGGER.trap_en.
When ADD_QUEUE.skip_process_ctx_clear is set, MES will prioritize
the SET_SHADER_DEBUGGER.trap_en setting.

Second, to preserve correct save/restore priviledged wave states
in coordination with the trap enablement setting, resume suspended
waves early in the disable call.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c            |  7 ++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h            |  4 +++-
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c             |  1 +
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c             | 14 ++++++--------
 .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c  |  3 +--
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c          | 12 +++++++-----
 drivers/gpu/drm/amd/include/mes_v11_api_def.h      |  1 +
 7 files changed, 25 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index 20cc3fffe921a..e9091ebfe230d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -928,7 +928,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 				uint64_t process_context_addr,
 				uint32_t spi_gdbg_per_vmid_cntl,
 				const uint32_t *tcp_watch_cntl,
-				uint32_t flags)
+				uint32_t flags,
+				bool trap_en)
 {
 	struct mes_misc_op_input op_input = {0};
 	int r;
@@ -945,6 +946,10 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 	memcpy(op_input.set_shader_debugger.tcp_watch_cntl, tcp_watch_cntl,
 			sizeof(op_input.set_shader_debugger.tcp_watch_cntl));
 
+	if (((adev->mes.sched_version & AMDGPU_MES_API_VERSION_MASK) >>
+			AMDGPU_MES_API_VERSION_SHIFT) >= 14)
+		op_input.set_shader_debugger.trap_en = trap_en;
+
 	amdgpu_mes_lock(&adev->mes);
 
 	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index b5f5eed2b5efe..2d6ac30b7135b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -294,6 +294,7 @@ struct mes_misc_op_input {
 			} flags;
 			uint32_t spi_gdbg_per_vmid_cntl;
 			uint32_t tcp_watch_cntl[4];
+			uint32_t trap_en;
 		} set_shader_debugger;
 	};
 };
@@ -361,7 +362,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 				uint64_t process_context_addr,
 				uint32_t spi_gdbg_per_vmid_cntl,
 				const uint32_t *tcp_watch_cntl,
-				uint32_t flags);
+				uint32_t flags,
+				bool trap_en);
 
 int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id,
 			int queue_type, int idx,
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index c4e3cb8d44de3..1bdaa00c0b466 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -347,6 +347,7 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes,
 		memcpy(misc_pkt.set_shader_debugger.tcp_watch_cntl,
 				input->set_shader_debugger.tcp_watch_cntl,
 				sizeof(misc_pkt.set_shader_debugger.tcp_watch_cntl));
+		misc_pkt.set_shader_debugger.trap_en = input->set_shader_debugger.trap_en;
 		break;
 	default:
 		DRM_ERROR("unsupported misc op (%d) \n", input->op);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index 125274445f439..cd34e7aaead49 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -349,12 +349,13 @@ int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd)
 {
 	uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode;
 	uint32_t flags = pdd->process->dbg_flags;
+	bool sq_trap_en = !!spi_dbg_cntl;
 
 	if (!kfd_dbg_is_per_vmid_supported(pdd->dev))
 		return 0;
 
 	return amdgpu_mes_set_shader_debugger(pdd->dev->adev, pdd->proc_ctx_gpu_addr, spi_dbg_cntl,
-						pdd->watch_points, flags);
+						pdd->watch_points, flags, sq_trap_en);
 }
 
 #define KFD_DEBUGGER_INVALID_WATCH_POINT_ID -1
@@ -557,6 +558,10 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 
 	if (!unwind) {
 		uint32_t flags = 0;
+		int resume_count = resume_queues(target, 0, NULL);
+
+		if (resume_count)
+			pr_debug("Resumed %d queues\n", resume_count);
 
 		cancel_work_sync(&target->debug_event_workarea);
 		kfd_dbg_clear_process_address_watch(target);
@@ -598,13 +603,6 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind
 	}
 
 	kfd_dbg_set_workaround(target, false);
-
-	if (!unwind) {
-		int resume_count = resume_queues(target, 0, NULL);
-
-		if (resume_count)
-			pr_debug("Resumed %d queues\n", resume_count);
-	}
 }
 
 static void kfd_dbg_clean_exception_status(struct kfd_process *target)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 498ad7d4e7d90..d6b15493fffdb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -227,8 +227,7 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q,
 	queue_input.tba_addr = qpd->tba_addr;
 	queue_input.tma_addr = qpd->tma_addr;
 	queue_input.trap_en = KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) ||
-			      KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0) ||
-			      q->properties.is_dbg_wa;
+			      KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0);
 	queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled;
 
 	queue_type = convert_to_mes_queue_type(q->properties.type);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index faa7939f35bde..90b86a6ac7bd6 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1863,13 +1863,15 @@ static void kfd_topology_set_dbg_firmware_support(struct kfd_topology_device *de
 {
 	bool firmware_supported = true;
 
-	/*
-	 * FIXME: GFX11 FW currently not sufficient to deal with CWSR WA.
-	 * Updated FW with API changes coming soon.
-	 */
 	if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(11, 0, 0) &&
 			KFD_GC_VERSION(dev->gpu) < IP_VERSION(12, 0, 0)) {
-		firmware_supported = false;
+		uint32_t mes_api_rev = (dev->gpu->adev->mes.sched_version &
+						AMDGPU_MES_API_VERSION_MASK) >>
+						AMDGPU_MES_API_VERSION_SHIFT;
+		uint32_t mes_rev = dev->gpu->adev->mes.sched_version &
+						AMDGPU_MES_VERSION_MASK;
+
+		firmware_supported = (mes_api_rev >= 14) && (mes_rev >= 64);
 		goto out;
 	}
 
diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
index f3c15f18ddb56..0997e999416a4 100644
--- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
@@ -575,6 +575,7 @@ struct SET_SHADER_DEBUGGER {
 	} flags;
 	uint32_t spi_gdbg_per_vmid_cntl;
 	uint32_t tcp_watch_cntl[4]; /* TCP_WATCHx_CNTL */
+	uint32_t trap_en;
 };
 
 union MESAPI__MISC {
-- 
GitLab


From 11b92df8a2f7f4605ccc764ce6ae4a72760674df Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Wed, 22 Mar 2023 14:29:26 -0400
Subject: [PATCH 1626/1791] Revert "Revert drm/amd/display: Enable Freesync
 Video Mode by default"

This reverts commit 4243c84aa082d8fba70c45f48eb2bb5c19799060.

Enables freesync video by default, since the hang and corruption issue
on eDP panels are now fixed.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Rodrigo Siqueira <rjordrigo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index a6bc2c6cad5c1..e14704b764e0f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6039,8 +6039,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
 		 */
 		DRM_DEBUG_DRIVER("No preferred mode found\n");
 	} else {
-		recalculate_timing = amdgpu_freesync_vid_mode &&
-				 is_freesync_video_mode(&mode, aconnector);
+		recalculate_timing = is_freesync_video_mode(&mode, aconnector);
 		if (recalculate_timing) {
 			freesync_mode = get_highest_refresh_rate_mode(aconnector, false);
 			drm_mode_copy(&saved_mode, &mode);
@@ -7233,7 +7232,7 @@ static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connect
 	struct amdgpu_dm_connector *amdgpu_dm_connector =
 		to_amdgpu_dm_connector(connector);
 
-	if (!(amdgpu_freesync_vid_mode && edid))
+	if (!edid)
 		return;
 
 	if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10)
@@ -9293,8 +9292,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		 * TODO: Refactor this function to allow this check to work
 		 * in all conditions.
 		 */
-		if (amdgpu_freesync_vid_mode &&
-		    dm_new_crtc_state->stream &&
+		if (dm_new_crtc_state->stream &&
 		    is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state))
 			goto skip_modeset;
 
@@ -9336,7 +9334,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 		}
 
 		/* Now check if we should set freesync video mode */
-		if (amdgpu_freesync_vid_mode && dm_new_crtc_state->stream &&
+		if (dm_new_crtc_state->stream &&
 		    dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) &&
 		    dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream) &&
 		    is_timing_unchanged_for_freesync(new_crtc_state,
@@ -9350,7 +9348,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm,
 			set_freesync_fixed_config(dm_new_crtc_state);
 
 			goto skip_modeset;
-		} else if (amdgpu_freesync_vid_mode && aconnector &&
+		} else if (aconnector &&
 			   is_freesync_video_mode(&new_crtc_state->mode,
 						  aconnector)) {
 			struct drm_display_mode *high_mode;
-- 
GitLab


From 9d65b1b4bcf3918164e17365eec169875eef8ee3 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Tue, 23 May 2023 12:02:32 +0800
Subject: [PATCH 1627/1791] drm/amdgpu: add the accelerator PCIe class

Add the accelerator PCIe class and match the
class in amdgpu for 0x1002 devices of that class.

From PCI spec:
"PCI Code and ID Assignment, r1.9, sec 1, 1.19"

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>  # pci_ids.h
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c  | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 2 +-
 drivers/gpu/drm/amd/include/amd_shared.h | 1 -
 include/linux/pci_ids.h                  | 3 +++
 4 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b8a1e4571cd93..8e58d187b1735 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2044,7 +2044,7 @@ static const struct pci_device_id pciidlist[] = {
 	  .driver_data = CHIP_IP_DISCOVERY },
 
 	{ PCI_DEVICE(0x1002, PCI_ANY_ID),
-	  .class = AMD_ACCELERATOR_PROCESSING << 8,
+	  .class = PCI_CLASS_ACCELERATOR_PROCESSING << 8,
 	  .class_mask = 0xffffff,
 	  .driver_data = CHIP_IP_DISCOVERY },
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 0f1ca0136f50c..25b4d7f0bd359 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -57,7 +57,7 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 	/* enable virtual display */
 	if (adev->asic_type != CHIP_ALDEBARAN &&
 	    adev->asic_type != CHIP_ARCTURUS &&
-	    ((adev->pdev->class >> 8) != AMD_ACCELERATOR_PROCESSING)) {
+	    ((adev->pdev->class >> 8) != PCI_CLASS_ACCELERATOR_PROCESSING)) {
 		if (adev->mode_info.num_crtc == 0)
 			adev->mode_info.num_crtc = 1;
 		adev->enable_virtual_display = true;
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 57d95e2cc54bb..f175e65b853a0 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -27,7 +27,6 @@
 
 
 #define AMD_MAX_USEC_TIMEOUT		1000000  /* 1000 ms */
-#define AMD_ACCELERATOR_PROCESSING	0x1200   /* hardcoded pci class */
 
 /*
  * Chip flags
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 45c3d62e616d8..0fbfbda3dc269 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -151,6 +151,9 @@
 #define PCI_CLASS_SP_DPIO		0x1100
 #define PCI_CLASS_SP_OTHER		0x1180
 
+#define PCI_BASE_CLASS_ACCELERATOR	0x12
+#define PCI_CLASS_ACCELERATOR_PROCESSING	0x1200
+
 #define PCI_CLASS_OTHERS		0xff
 
 /* Vendors and devices.  Sort key: vendor first, device next. */
-- 
GitLab


From 4e70da985cef954cdf7813d651c067d2c602ea71 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Thu, 8 Jun 2023 10:01:57 -0700
Subject: [PATCH 1628/1791] drm/amdgpu: Wrap -Wunused-but-set-variable in
 cc-option

-Wunused-but-set-variable was only supported in clang starting with
13.0.0, so earlier versions will emit a warning, which is turned into a
hard error for the kernel to mirror GCC:

  error: unknown warning option '-Wunused-but-set-variable'; did you mean '-Wunused-const-variable'? [-Werror,-Wunknown-warning-option]

The minimum supported version of clang for building the kernel is
11.0.0, so match the rest of the kernel and wrap
-Wunused-but-set-variable in a cc-option call, so that it is only used
when supported by the compiler.

Closes: https://github.com/ClangBuiltLinux/linux/issues/1869
Fixes: 1b320ad3f5a8 ("drm/amd/amdgpu: introduce DRM_AMDGPU_WERROR")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 7ee68b1bbfedf..86b833085f194 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -40,7 +40,7 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_PATH)/amdkfd
 
 subdir-ccflags-y := -Wextra
-subdir-ccflags-y += -Wunused-but-set-variable
+subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable)
 subdir-ccflags-y += -Wno-unused-parameter
 subdir-ccflags-y += -Wno-type-limits
 subdir-ccflags-y += -Wno-sign-compare
-- 
GitLab


From 80e709ee6ecc9eba8bd8d188218472822e1b38bd Mon Sep 17 00:00:00 2001
From: Chong Li <chongli2@amd.com>
Date: Wed, 7 Jun 2023 15:56:12 +0800
Subject: [PATCH 1629/1791] drm/amdgpu: add option params to enforce process
 isolation between graphics and compute

enforce process isolation between graphics and compute via using the same reserved vmid.

v2: remove params "struct amdgpu_vm *vm" from
    amdgpu_vmid_alloc_reserved and amdgpu_vmid_free_reserved.

Signed-off-by: Chong Li <chongli2@amd.com>
Reviewed-by: Christian Koenig <Christian.Koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 10 +++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c | 17 +++++++----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h |  6 ++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c  | 25 +++++++++++++++++--------
 5 files changed, 36 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 74104d46a7bea..a84bd4a0c4215 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -214,6 +214,7 @@ extern int amdgpu_force_asic_type;
 extern int amdgpu_smartshift_bias;
 extern int amdgpu_use_xgmi_p2p;
 extern int amdgpu_mtype_local;
+extern bool enforce_isolation;
 #ifdef CONFIG_HSA_AMD
 extern int sched_policy;
 extern bool debug_evictions;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 8e58d187b1735..999d008b6b484 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -153,7 +153,7 @@ uint amdgpu_pg_mask = 0xffffffff;
 uint amdgpu_sdma_phase_quantum = 32;
 char *amdgpu_disable_cu;
 char *amdgpu_virtual_display;
-
+bool enforce_isolation;
 /*
  * OverDrive(bit 14) disabled by default
  * GFX DCS(bit 19) disabled by default
@@ -973,6 +973,14 @@ MODULE_PARM_DESC(
 						4 = AMDGPU_CPX_PARTITION_MODE)");
 module_param_named(user_partt_mode, amdgpu_user_partt_mode, uint, 0444);
 
+
+/**
+ * DOC: enforce_isolation (bool)
+ * enforce process isolation between graphics and compute via using the same reserved vmid.
+ */
+module_param(enforce_isolation, bool, 0444);
+MODULE_PARM_DESC(enforce_isolation, "enforce process isolation between graphics and compute . enforce_isolation = on");
+
 /* These devices are not supported by amdgpu.
  * They are supported by the mach64, r128, radeon drivers
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index c991ca0b7a1c8..ff1ea99292fbf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -409,7 +409,7 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 	if (r || !idle)
 		goto error;
 
-	if (vm->reserved_vmid[vmhub]) {
+	if (vm->reserved_vmid[vmhub] || (enforce_isolation && (vmhub == AMDGPU_GFXHUB(0)))) {
 		r = amdgpu_vmid_grab_reserved(vm, ring, job, &id, fence);
 		if (r || !id)
 			goto error;
@@ -460,14 +460,11 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 }
 
 int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
-			       struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
 	mutex_lock(&id_mgr->lock);
-	if (vm->reserved_vmid[vmhub])
-		goto unlock;
 
 	++id_mgr->reserved_use_count;
 	if (!id_mgr->reserved) {
@@ -479,27 +476,23 @@ int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
 		list_del_init(&id->list);
 		id_mgr->reserved = id;
 	}
-	vm->reserved_vmid[vmhub] = true;
 
-unlock:
 	mutex_unlock(&id_mgr->lock);
 	return 0;
 }
 
 void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
-			       struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
 	mutex_lock(&id_mgr->lock);
-	if (vm->reserved_vmid[vmhub] &&
-	    !--id_mgr->reserved_use_count) {
+	if (!--id_mgr->reserved_use_count) {
 		/* give the reserved ID back to normal round robin */
 		list_add(&id_mgr->reserved->list, &id_mgr->ids_lru);
 		id_mgr->reserved = NULL;
 	}
-	vm->reserved_vmid[vmhub] = false;
+
 	mutex_unlock(&id_mgr->lock);
 }
 
@@ -578,6 +571,10 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 			list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru);
 		}
 	}
+	/* alloc a default reserved vmid to enforce isolation */
+	if (enforce_isolation)
+		amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0));
+
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
index d1cc09b45da4a..fa8c42c83d5d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
@@ -79,11 +79,9 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv,
 bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev,
 			       struct amdgpu_vmid *id);
 int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
-			       struct amdgpu_vm *vm,
-			       unsigned vmhub);
+				unsigned vmhub);
 void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
-			       struct amdgpu_vm *vm,
-			       unsigned vmhub);
+				unsigned vmhub);
 int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 		     struct amdgpu_job *job, struct dma_fence **fence);
 void amdgpu_vmid_reset(struct amdgpu_device *adev, unsigned vmhub,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 76d57bc7ac620..dc80c9c8fd140 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2284,8 +2284,14 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	}
 
 	dma_fence_put(vm->last_update);
-	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++)
-		amdgpu_vmid_free_reserved(adev, vm, i);
+
+	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) {
+		if (vm->reserved_vmid[i]) {
+			amdgpu_vmid_free_reserved(adev, i);
+			vm->reserved_vmid[i] = false;
+		}
+	}
+
 }
 
 /**
@@ -2368,7 +2374,6 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	union drm_amdgpu_vm *args = data;
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
-	int r;
 
 	/* No valid flags defined yet */
 	if (args->in.flags)
@@ -2377,13 +2382,17 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	switch (args->in.op) {
 	case AMDGPU_VM_OP_RESERVE_VMID:
 		/* We only have requirement to reserve vmid from gfxhub */
-		r = amdgpu_vmid_alloc_reserved(adev, &fpriv->vm,
-					       AMDGPU_GFXHUB(0));
-		if (r)
-			return r;
+		if (!fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
+			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0));
+			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = true;
+		}
+
 		break;
 	case AMDGPU_VM_OP_UNRESERVE_VMID:
-		amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB(0));
+		if (fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
+			amdgpu_vmid_free_reserved(adev, AMDGPU_GFXHUB(0));
+			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = false;
+		}
 		break;
 	default:
 		return -EINVAL;
-- 
GitLab


From 3ffb193969c57afd4096cfb107ca2cc3bb0c55d9 Mon Sep 17 00:00:00 2001
From: shikaguo <shikai.guo@amd.com>
Date: Fri, 9 Jun 2023 11:32:09 +0800
Subject: [PATCH 1630/1791] drm/amd/pm: enable more Pstates profile levels for
 yellow_carp

This patch enables following UMD stable Pstates profile levels for
power_dpm_force_performance_level interface.

- profile_peak
- profile_min_mclk
- profile_min_sclk
- profile_standard

Signed-off-by: shikaguo <shikai.guo@amd.com>
Reviewed-by: Tim Huang <Tim.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c  | 141 +++++++++++++++++-
 .../drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h  |   1 -
 2 files changed, 137 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
index a92da336ecec6..a1be2029ba4ab 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c
@@ -47,6 +47,14 @@
 #define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK		0x00000006L
 #define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT          0x1L
 
+#define SMU_13_0_8_UMD_PSTATE_GFXCLK                   533
+#define SMU_13_0_8_UMD_PSTATE_SOCCLK                   533
+#define SMU_13_0_8_UMD_PSTATE_FCLK                     800
+
+#define SMU_13_0_1_UMD_PSTATE_GFXCLK					700
+#define SMU_13_0_1_UMD_PSTATE_SOCCLK		              678
+#define SMU_13_0_1_UMD_PSTATE_FCLK			          1800
+
 #define FEATURE_MASK(feature) (1ULL << feature)
 #define SMC_DPM_FEATURE ( \
 	FEATURE_MASK(FEATURE_CCLK_DPM_BIT) | \
@@ -957,6 +965,9 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu,
 							uint32_t max)
 {
 	enum smu_message_type msg_set_min, msg_set_max;
+	uint32_t min_clk = min;
+	uint32_t max_clk = max;
+
 	int ret = 0;
 
 	if (!yellow_carp_clk_dpm_is_enabled(smu, clk_type))
@@ -985,11 +996,17 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu,
 		return -EINVAL;
 	}
 
-	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL);
+	if (clk_type == SMU_VCLK) {
+		min_clk = min << SMU_13_VCLK_SHIFT;
+		max_clk = max << SMU_13_VCLK_SHIFT;
+	}
+
+	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL);
+
 	if (ret)
 		goto out;
 
-	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL);
+	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max_clk, NULL);
 	if (ret)
 		goto out;
 
@@ -997,12 +1014,49 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu,
 	return ret;
 }
 
+static uint32_t yellow_carp_get_umd_pstate_clk_default(struct smu_context *smu,
+					enum smu_clk_type clk_type)
+{
+	uint32_t clk_limit = 0;
+	struct amdgpu_device *adev = smu->adev;
+
+	switch (clk_type) {
+	case SMU_GFXCLK:
+	case SMU_SCLK:
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8))
+			clk_limit = SMU_13_0_8_UMD_PSTATE_GFXCLK;
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) ||
+			(adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3))
+			clk_limit = SMU_13_0_1_UMD_PSTATE_GFXCLK;
+		break;
+	case SMU_SOCCLK:
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8))
+			clk_limit = SMU_13_0_8_UMD_PSTATE_SOCCLK;
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) ||
+			(adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3))
+			clk_limit = SMU_13_0_1_UMD_PSTATE_SOCCLK;
+		break;
+	case SMU_FCLK:
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8))
+			clk_limit = SMU_13_0_8_UMD_PSTATE_FCLK;
+		if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) ||
+			(adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3))
+			clk_limit = SMU_13_0_1_UMD_PSTATE_FCLK;
+		break;
+	default:
+		break;
+	}
+
+	return clk_limit;
+}
+
 static int yellow_carp_print_clk_levels(struct smu_context *smu,
 				enum smu_clk_type clk_type, char *buf)
 {
 	int i, idx, size = 0, ret = 0;
 	uint32_t cur_value = 0, value = 0, count = 0;
 	uint32_t min, max;
+	uint32_t clk_limit = 0;
 
 	smu_cmn_get_sysfs_buf(&buf, &size);
 
@@ -1044,6 +1098,7 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu,
 		break;
 	case SMU_GFXCLK:
 	case SMU_SCLK:
+		clk_limit = yellow_carp_get_umd_pstate_clk_default(smu, clk_type);
 		ret = yellow_carp_get_current_clk_freq(smu, clk_type, &cur_value);
 		if (ret)
 			goto print_clk_out;
@@ -1058,7 +1113,7 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu,
 		size += sysfs_emit_at(buf, size, "0: %uMhz %s\n", min,
 				i == 0 ? "*" : "");
 		size += sysfs_emit_at(buf, size, "1: %uMhz %s\n",
-				i == 1 ? cur_value : YELLOW_CARP_UMD_PSTATE_GFXCLK,
+				i == 1 ? cur_value : clk_limit,
 				i == 1 ? "*" : "");
 		size += sysfs_emit_at(buf, size, "2: %uMhz %s\n", max,
 				i == 2 ? "*" : "");
@@ -1107,6 +1162,49 @@ static int yellow_carp_force_clk_levels(struct smu_context *smu,
 	return ret;
 }
 
+static int yellow_carp_get_dpm_profile_freq(struct smu_context *smu,
+					enum amd_dpm_forced_level level,
+					enum smu_clk_type clk_type,
+					uint32_t *min_clk,
+					uint32_t *max_clk)
+{
+	int ret = 0;
+	uint32_t clk_limit = 0;
+
+	clk_limit = yellow_carp_get_umd_pstate_clk_default(smu, clk_type);
+
+	switch (clk_type) {
+	case SMU_GFXCLK:
+	case SMU_SCLK:
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit);
+		else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK)
+			yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL);
+		break;
+	case SMU_SOCCLK:
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &clk_limit);
+		break;
+	case SMU_FCLK:
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &clk_limit);
+		else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK)
+			yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &clk_limit, NULL);
+		break;
+	case SMU_VCLK:
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit);
+		break;
+	case SMU_DCLK:
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	*min_clk = *max_clk = clk_limit;
+	return ret;
+}
+
 static int yellow_carp_set_performance_level(struct smu_context *smu,
 						enum amd_dpm_forced_level level)
 {
@@ -1114,6 +1212,9 @@ static int yellow_carp_set_performance_level(struct smu_context *smu,
 	uint32_t sclk_min = 0, sclk_max = 0;
 	uint32_t fclk_min = 0, fclk_max = 0;
 	uint32_t socclk_min = 0, socclk_max = 0;
+	uint32_t vclk_min = 0, vclk_max = 0;
+	uint32_t dclk_min = 0, dclk_max = 0;
+
 	int ret = 0;
 
 	switch (level) {
@@ -1121,28 +1222,42 @@ static int yellow_carp_set_performance_level(struct smu_context *smu,
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &fclk_max);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &socclk_max);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max);
 		sclk_min = sclk_max;
 		fclk_min = fclk_max;
 		socclk_min = socclk_max;
+		vclk_min = vclk_max;
+		dclk_min = dclk_max;
 		break;
 	case AMD_DPM_FORCED_LEVEL_LOW:
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, NULL);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, NULL);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL);
 		sclk_max = sclk_min;
 		fclk_max = fclk_min;
 		socclk_max = socclk_min;
+		vclk_max = vclk_min;
+		dclk_max = dclk_min;
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, &fclk_max);
 		yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, &socclk_max);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max);
+		yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
 	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
-		/* Temporarily do nothing since the optimal clocks haven't been provided yet */
+		yellow_carp_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max);
+		yellow_carp_get_dpm_profile_freq(smu, level, SMU_FCLK, &fclk_min, &fclk_max);
+		yellow_carp_get_dpm_profile_freq(smu, level, SMU_SOCCLK, &socclk_min, &socclk_max);
+		yellow_carp_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max);
+		yellow_carp_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
 	case AMD_DPM_FORCED_LEVEL_MANUAL:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT:
@@ -1182,6 +1297,24 @@ static int yellow_carp_set_performance_level(struct smu_context *smu,
 			return ret;
 	}
 
+	if (vclk_min && vclk_max) {
+		ret = yellow_carp_set_soft_freq_limited_range(smu,
+							      SMU_VCLK,
+							      vclk_min,
+							      vclk_max);
+		if (ret)
+			return ret;
+	}
+
+	if (dclk_min && dclk_max) {
+		ret = yellow_carp_set_soft_freq_limited_range(smu,
+							      SMU_DCLK,
+							      dclk_min,
+							      dclk_max);
+		if (ret)
+			return ret;
+	}
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h
index a9205a8ea3ad2..b3ad8352c68ae 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h
@@ -24,6 +24,5 @@
 #define __YELLOW_CARP_PPT_H__
 
 extern void yellow_carp_set_ppt_funcs(struct smu_context *smu);
-#define YELLOW_CARP_UMD_PSTATE_GFXCLK       1100
 
 #endif
-- 
GitLab


From 389c6b3e120303715c018d1bfc7bab02b50ca3f4 Mon Sep 17 00:00:00 2001
From: Charlene Liu <charlene.liu@amd.com>
Date: Fri, 19 May 2023 20:43:30 -0400
Subject: [PATCH 1631/1791] drm/amd/display: add NULL pointer check

[why]
check dmub_Srv exist or not before accessing dmub.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Charlene Liu <charlene.liu@amd.com>
Reviewed-by: Zhan Liu <Zhan.Liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc_stream.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 0d3ec50b13851..6e11d2b701f82 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -276,8 +276,8 @@ static void program_cursor_attributes(
 		}
 
 		dc->hwss.set_cursor_attribute(pipe_ctx);
-
-		dc_send_update_cursor_info_to_dmu(pipe_ctx, i);
+		if (dc->ctx->dmub_srv)
+			dc_send_update_cursor_info_to_dmu(pipe_ctx, i);
 		if (dc->hwss.set_cursor_sdr_white_level)
 			dc->hwss.set_cursor_sdr_white_level(pipe_ctx);
 	}
@@ -396,8 +396,8 @@ static void program_cursor_position(
 		}
 
 		dc->hwss.set_cursor_position(pipe_ctx);
-
-		dc_send_update_cursor_info_to_dmu(pipe_ctx, i);
+		if (dc->ctx->dmub_srv)
+			dc_send_update_cursor_info_to_dmu(pipe_ctx, i);
 	}
 
 	if (pipe_to_program)
-- 
GitLab


From 57a8011512131c63cf700d42ef56ad875409a1a5 Mon Sep 17 00:00:00 2001
From: Kenneth Feng <kenneth.feng@amd.com>
Date: Thu, 8 Jun 2023 22:07:11 +0800
Subject: [PATCH 1632/1791] drm/amd/pm: workaround for compute workload type on
 some skus

On smu 13.0.0, the compute workload type cannot be set on all the skus
due to some other problems. This workaround is to make sure compute workload type
can also run on some specific skus.

v2: keep the variable consistent

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Acked-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c  | 33 +++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index 413e592f0ed61..4dd01b3f350ff 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -2179,10 +2179,39 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu,
 		}
 	}
 
-	/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
-	workload_type = smu_cmn_to_asic_specific_index(smu,
+	if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE &&
+		(((smu->adev->pdev->device == 0x744C) && (smu->adev->pdev->revision == 0xC8)) ||
+		((smu->adev->pdev->device == 0x744C) && (smu->adev->pdev->revision == 0xCC)))) {
+		ret = smu_cmn_update_table(smu,
+					   SMU_TABLE_ACTIVITY_MONITOR_COEFF,
+					   WORKLOAD_PPLIB_COMPUTE_BIT,
+					   (void *)(&activity_monitor_external),
+					   false);
+		if (ret) {
+			dev_err(smu->adev->dev, "[%s] Failed to get activity monitor!", __func__);
+			return ret;
+		}
+
+		ret = smu_cmn_update_table(smu,
+					   SMU_TABLE_ACTIVITY_MONITOR_COEFF,
+					   WORKLOAD_PPLIB_CUSTOM_BIT,
+					   (void *)(&activity_monitor_external),
+					   true);
+		if (ret) {
+			dev_err(smu->adev->dev, "[%s] Failed to set activity monitor!", __func__);
+			return ret;
+		}
+
+		workload_type = smu_cmn_to_asic_specific_index(smu,
+						       CMN2ASIC_MAPPING_WORKLOAD,
+						       PP_SMC_POWER_PROFILE_CUSTOM);
+	} else {
+		/* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */
+		workload_type = smu_cmn_to_asic_specific_index(smu,
 						       CMN2ASIC_MAPPING_WORKLOAD,
 						       smu->power_profile_mode);
+	}
+
 	if (workload_type < 0)
 		return -EINVAL;
 
-- 
GitLab


From aeb3dd7e6f91da0a8c460d61fad13db85b3b33b7 Mon Sep 17 00:00:00 2001
From: Leo Ma <hanghong.ma@amd.com>
Date: Thu, 18 May 2023 14:07:13 -0400
Subject: [PATCH 1633/1791] Revert "drm/amd/display: cache trace buffer size"

Revert commit 257e9891db0b ("drm/amd/display: cache trace buffer size")
to fix regression found in tests.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Leo Ma <hanghong.ma@amd.com>
Reviewed-by: Josip Pavic <Josip.Pavic@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h     | 1 -
 drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c | 2 --
 2 files changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index e210cb082ebd9..7c9a2b34bd05c 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -428,7 +428,6 @@ struct dmub_srv {
 	enum dmub_asic asic;
 	void *user_ctx;
 	uint32_t fw_version;
-	uint32_t trace_buffer_size;
 	bool is_virtual;
 	struct dmub_fb scratch_mem_fb;
 	volatile const struct dmub_fw_state *fw_state;
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
index 8b9af18e84fe0..9e9a6a44a7ac4 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c
@@ -427,8 +427,6 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub,
 			dmub->fw_version = fw_info->fw_version;
 	}
 
-	dmub->trace_buffer_size = trace_buffer_size;
-
 	trace_buff->base = dmub_align(mail->top, 256);
 	trace_buff->top = trace_buff->base + dmub_align(trace_buffer_size, 64);
 
-- 
GitLab


From 24e52fc20201c87912eee8f337829c036c3b0f3a Mon Sep 17 00:00:00 2001
From: Sridevi <sarvinde@amd.com>
Date: Wed, 10 May 2023 14:34:55 -0400
Subject: [PATCH 1634/1791] drm/amd/display: DSC Programming Deltas

[Why]
Programming register delta for DSC sub-block

[How]
Change DSC, resource files for programming register delta.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Sridevi <sarvinde@amd.com>
Reviewed-by: Chris Park <Chris.Park@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c  | 29 +++++++------------
 .../gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h  | 28 ++++++++++++++++++
 2 files changed, 38 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
index 5bd698cd6d20b..5eebe7f03ddc7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
@@ -30,22 +30,13 @@
 #include "dsc/dscc_types.h"
 #include "dsc/rc_calc.h"
 
-static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps);
-static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals,
-			struct dsc_optc_config *dsc_optc_cfg);
-static void dsc_init_reg_values(struct dsc_reg_values *reg_vals);
-static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params);
 static void dsc_write_to_registers(struct display_stream_compressor *dsc, const struct dsc_reg_values *reg_vals);
-static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple);
-static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth);
 
 /* Object I/F functions */
-static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz);
 static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_dsc_state *s);
 static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg);
 static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg,
 		struct dsc_optc_config *dsc_optc_cfg);
-static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps);
 static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe);
 static void dsc2_disable(struct display_stream_compressor *dsc);
 static void dsc2_disconnect(struct display_stream_compressor *dsc);
@@ -108,7 +99,7 @@ void dsc2_construct(struct dcn20_dsc *dsc,
 /* This returns the capabilities for a single DSC encoder engine. Number of slices and total throughput
  * can be doubled, tripled etc. by using additional DSC engines.
  */
-static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz)
+void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz)
 {
 	dsc_enc_caps->dsc_version = 0x21; /* v1.2 - DP spec defined it in reverse order and we kept it */
 
@@ -184,7 +175,7 @@ static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const st
 }
 
 
-static void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config)
+void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config)
 {
 	DC_LOG_DSC("\tnum_slices_h %d", config->dc_dsc_cfg.num_slices_h);
 	DC_LOG_DSC("\tnum_slices_v %d", config->dc_dsc_cfg.num_slices_v);
@@ -211,7 +202,7 @@ static void dsc2_set_config(struct display_stream_compressor *dsc, const struct
 }
 
 
-static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps)
+bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps)
 {
 	bool is_config_ok;
 	struct dsc_reg_values dsc_reg_vals;
@@ -291,7 +282,7 @@ static void dsc2_disconnect(struct display_stream_compressor *dsc)
 }
 
 /* This module's internal functions */
-static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps)
+void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps)
 {
 	int i;
 	int bits_per_pixel = pps->bits_per_pixel;
@@ -345,7 +336,7 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co
 	}
 }
 
-static void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_params_override *override)
+void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_params_override *override)
 {
 	uint8_t i;
 
@@ -372,7 +363,7 @@ static void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_
 	rc->flatness_det_thresh = override->flatness_det_thresh;
 }
 
-static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals,
+bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals,
 			struct dsc_optc_config *dsc_optc_cfg)
 {
 	struct dsc_parameters dsc_params;
@@ -463,7 +454,7 @@ static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_
 }
 
 
-static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple)
+enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple)
 {
 	enum dsc_pixel_format dsc_pix_fmt = DSC_PIXFMT_UNKNOWN;
 
@@ -495,7 +486,7 @@ static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_p
 }
 
 
-static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth)
+enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth)
 {
 	enum dsc_bits_per_comp bpc = DSC_BPC_UNKNOWN;
 
@@ -518,7 +509,7 @@ static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_co
 }
 
 
-static void dsc_init_reg_values(struct dsc_reg_values *reg_vals)
+void dsc_init_reg_values(struct dsc_reg_values *reg_vals)
 {
 	int i;
 
@@ -574,7 +565,7 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals)
  * This is required because dscc_compute_dsc_parameters returns a modified PPS, which in turn
  * affects non-PPS register values.
  */
-static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params)
+void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params)
 {
 	int i;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
index 7ce64a3c1b02b..ba869387c3c58 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h
@@ -549,6 +549,27 @@ struct dcn20_dsc {
 	int max_image_width;
 };
 
+void dsc_config_log(struct display_stream_compressor *dsc,
+		const struct dsc_config *config);
+
+void dsc_log_pps(struct display_stream_compressor *dsc,
+		struct drm_dsc_config *pps);
+
+void dsc_override_rc_params(struct rc_params *rc,
+		const struct dc_dsc_rc_params_override *override);
+
+bool dsc_prepare_config(const struct dsc_config *dsc_cfg,
+		struct dsc_reg_values *dsc_reg_vals,
+		struct dsc_optc_config *dsc_optc_cfg);
+
+enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc,
+		bool is_ycbcr422_simple);
+
+enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth);
+
+void dsc_init_reg_values(struct dsc_reg_values *reg_vals);
+
+void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params);
 
 void dsc2_construct(struct dcn20_dsc *dsc,
 		struct dc_context *ctx,
@@ -557,5 +578,12 @@ void dsc2_construct(struct dcn20_dsc *dsc,
 		const struct dcn20_dsc_shift *dsc_shift,
 		const struct dcn20_dsc_mask *dsc_mask);
 
+void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps,
+		int pixel_clock_100Hz);
+
+bool dsc2_get_packed_pps(struct display_stream_compressor *dsc,
+		const struct dsc_config *dsc_cfg,
+		uint8_t *dsc_packed_pps);
+
 #endif
 
-- 
GitLab


From 3b718dcaf163d17fe907ea098c8449e0cd6bc271 Mon Sep 17 00:00:00 2001
From: Austin Zheng <austin.zheng@amd.com>
Date: Wed, 24 May 2023 11:52:12 -0400
Subject: [PATCH 1635/1791] drm/amd/display: Filter out AC mode frequencies on
 DC mode systems

Why:
Limit maximum clock speeds to DC mode limits for DC mode systems
How:
Store DC mode limits when individual clocks are initialized and
cap the values when building the clock table

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  |  13 +-
 drivers/gpu/drm/amd/display/dc/dc.h           |   1 +
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 140 ++++++++++++-----
 .../amd/display/dc/dml/dcn321/dcn321_fpu.c    | 144 +++++++++++++-----
 .../gpu/drm/amd/display/dc/inc/hw/clk_mgr.h   |   1 +
 5 files changed, 216 insertions(+), 83 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 20bff6a346baa..96fa68f166e0b 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -182,23 +182,32 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 	dcn32_init_single_clock(clk_mgr, PPCLK_DCFCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz,
 			&num_entries_per_clk->num_dcfclk_levels);
+	clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK);
 
 	/* SOCCLK */
 	dcn32_init_single_clock(clk_mgr, PPCLK_SOCCLK,
 					&clk_mgr_base->bw_params->clk_table.entries[0].socclk_mhz,
 					&num_entries_per_clk->num_socclk_levels);
+	clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK);
 
 	/* DTBCLK */
-	if (!clk_mgr->base.ctx->dc->debug.disable_dtb_ref_clk_switch)
+	if (!clk_mgr->base.ctx->dc->debug.disable_dtb_ref_clk_switch) {
 		dcn32_init_single_clock(clk_mgr, PPCLK_DTBCLK,
 				&clk_mgr_base->bw_params->clk_table.entries[0].dtbclk_mhz,
 				&num_entries_per_clk->num_dtbclk_levels);
+		clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz =
+				dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK);
+	}
 
 	/* DISPCLK */
 	dcn32_init_single_clock(clk_mgr, PPCLK_DISPCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dispclk_mhz,
 			&num_entries_per_clk->num_dispclk_levels);
 	num_levels = num_entries_per_clk->num_dispclk_levels;
+	clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK);
+	//HW recommends limit of 1950 MHz in display clock for all DCN3.2.x
+	if (clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz > 1950)
+		clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = 1950;
 
 	if (num_entries_per_clk->num_dcfclk_levels &&
 			num_entries_per_clk->num_dtbclk_levels &&
@@ -817,6 +826,7 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 	dcn32_init_single_clock(clk_mgr, PPCLK_UCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].memclk_mhz,
 			&num_entries_per_clk->num_memclk_levels);
+	clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK);
 
 	/* memclk must have at least one level */
 	num_entries_per_clk->num_memclk_levels = num_entries_per_clk->num_memclk_levels ? num_entries_per_clk->num_memclk_levels : 1;
@@ -824,6 +834,7 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 	dcn32_init_single_clock(clk_mgr, PPCLK_FCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].fclk_mhz,
 			&num_entries_per_clk->num_fclk_levels);
+	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
 
 	if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) {
 		num_levels = num_entries_per_clk->num_memclk_levels;
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7ded574f84ff1..360dd83b1a7a0 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -897,6 +897,7 @@ struct dc_debug_options {
 	uint32_t fpo_vactive_min_active_margin_us;
 	uint32_t fpo_vactive_max_blank_us;
 	bool enable_legacy_fast_update;
+	bool disable_dc_mode_overwrite;
 };
 
 struct gpu_info_soc_bounding_box_v1_0;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index ad6ee48580f8e..fa3678342abb7 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -2325,14 +2325,48 @@ void dcn32_patch_dpm_table(struct clk_bw_params *bw_params)
 		bw_params->clk_table.entries[0].memclk_mhz = dcn3_2_soc.clock_limits[0].dram_speed_mts / 16;
 }
 
-static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
+/*
+ * override_max_clk_values - Overwrite the max clock frequencies with the max DC mode timings
+ * Input:
+ *	max_clk_limit - struct containing the desired clock timings
+ * Output:
+ *	curr_clk_limit  - struct containing the timings that need to be overwritten
+ * Return: 0 upon success, non-zero for failure
+ */
+static int override_max_clk_values(struct clk_limit_table_entry *max_clk_limit,
+		struct clk_limit_table_entry *curr_clk_limit)
+{
+	if (NULL == max_clk_limit || NULL == curr_clk_limit)
+		return -1; //invalid parameters
+
+	//only overwrite if desired max clock frequency is initialized
+	if (max_clk_limit->dcfclk_mhz != 0)
+		curr_clk_limit->dcfclk_mhz = max_clk_limit->dcfclk_mhz;
+
+	if (max_clk_limit->fclk_mhz != 0)
+		curr_clk_limit->fclk_mhz = max_clk_limit->fclk_mhz;
+
+	if (max_clk_limit->memclk_mhz != 0)
+		curr_clk_limit->memclk_mhz = max_clk_limit->memclk_mhz;
+
+	if (max_clk_limit->socclk_mhz != 0)
+		curr_clk_limit->socclk_mhz = max_clk_limit->socclk_mhz;
+
+	if (max_clk_limit->dtbclk_mhz != 0)
+		curr_clk_limit->dtbclk_mhz = max_clk_limit->dtbclk_mhz;
+
+	if (max_clk_limit->dispclk_mhz != 0)
+		curr_clk_limit->dispclk_mhz = max_clk_limit->dispclk_mhz;
+
+	return 0;
+}
+
+static int build_synthetic_soc_states(bool disable_dc_mode_overwrite, struct clk_bw_params *bw_params,
 		struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
 {
 	int i, j;
 	struct _vcs_dpi_voltage_scaling_st entry = {0};
-
-	unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0,
-			max_phyclk_mhz = 0, max_dtbclk_mhz = 0, max_fclk_mhz = 0, max_uclk_mhz = 0;
+	struct clk_limit_table_entry max_clk_data = {0};
 
 	unsigned int min_dcfclk_mhz = 199, min_fclk_mhz = 299;
 
@@ -2343,51 +2377,76 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	unsigned int num_fclk_dpms = 0;
 	unsigned int num_dcfclk_dpms = 0;
 
-	for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
-		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz)
-			max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
-		if (bw_params->clk_table.entries[i].fclk_mhz > max_fclk_mhz)
-			max_fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
-		if (bw_params->clk_table.entries[i].memclk_mhz > max_uclk_mhz)
-			max_uclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
-		if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz)
-			max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
-		if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz)
-			max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
-		if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz)
-			max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
-		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_dtbclk_mhz)
-			max_dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+	unsigned int num_dc_uclk_dpms = 0;
+	unsigned int num_dc_fclk_dpms = 0;
+	unsigned int num_dc_dcfclk_dpms = 0;
 
-		if (bw_params->clk_table.entries[i].memclk_mhz > 0)
+	for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_clk_data.dcfclk_mhz)
+			max_clk_data.dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
+		if (bw_params->clk_table.entries[i].fclk_mhz > max_clk_data.fclk_mhz)
+			max_clk_data.fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
+		if (bw_params->clk_table.entries[i].memclk_mhz > max_clk_data.memclk_mhz)
+			max_clk_data.memclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
+		if (bw_params->clk_table.entries[i].dispclk_mhz > max_clk_data.dispclk_mhz)
+			max_clk_data.dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
+		if (bw_params->clk_table.entries[i].dppclk_mhz > max_clk_data.dppclk_mhz)
+			max_clk_data.dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
+		if (bw_params->clk_table.entries[i].phyclk_mhz > max_clk_data.phyclk_mhz)
+			max_clk_data.phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
+		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_clk_data.dtbclk_mhz)
+			max_clk_data.dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+
+		if (bw_params->clk_table.entries[i].memclk_mhz > 0) {
 			num_uclk_dpms++;
-		if (bw_params->clk_table.entries[i].fclk_mhz > 0)
+			if (bw_params->clk_table.entries[i].memclk_mhz <= bw_params->dc_mode_limit.memclk_mhz)
+				num_dc_uclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].fclk_mhz > 0) {
 			num_fclk_dpms++;
-		if (bw_params->clk_table.entries[i].dcfclk_mhz > 0)
+			if (bw_params->clk_table.entries[i].fclk_mhz <= bw_params->dc_mode_limit.fclk_mhz)
+				num_dc_fclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) {
 			num_dcfclk_dpms++;
+			if (bw_params->clk_table.entries[i].dcfclk_mhz <= bw_params->dc_mode_limit.dcfclk_mhz)
+				num_dc_dcfclk_dpms++;
+		}
+	}
+
+	if (!disable_dc_mode_overwrite) {
+		//Overwrite max frequencies with max DC mode frequencies for DC mode systems
+		override_max_clk_values(&bw_params->dc_mode_limit, &max_clk_data);
+		num_uclk_dpms = num_dc_uclk_dpms;
+		num_fclk_dpms = num_dc_fclk_dpms;
+		num_dcfclk_dpms = num_dc_dcfclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_memclk_levels = num_uclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_fclk_levels = num_fclk_dpms;
 	}
 
 	if (num_dcfclk_dpms > 0 && bw_params->clk_table.entries[0].fclk_mhz > min_fclk_mhz)
 		min_fclk_mhz = bw_params->clk_table.entries[0].fclk_mhz;
 
-	if (!max_dcfclk_mhz || !max_dispclk_mhz || !max_dtbclk_mhz)
+	if (!max_clk_data.dcfclk_mhz || !max_clk_data.dispclk_mhz || !max_clk_data.dtbclk_mhz)
 		return -1;
 
-	if (max_dppclk_mhz == 0)
-		max_dppclk_mhz = max_dispclk_mhz;
+	if (max_clk_data.dppclk_mhz == 0)
+		max_clk_data.dppclk_mhz = max_clk_data.dispclk_mhz;
 
-	if (max_fclk_mhz == 0)
-		max_fclk_mhz = max_dcfclk_mhz * dcn3_2_soc.pct_ideal_sdp_bw_after_urgent / dcn3_2_soc.pct_ideal_fabric_bw_after_urgent;
+	if (max_clk_data.fclk_mhz == 0)
+		max_clk_data.fclk_mhz = max_clk_data.dcfclk_mhz *
+				dcn3_2_soc.pct_ideal_sdp_bw_after_urgent /
+				dcn3_2_soc.pct_ideal_fabric_bw_after_urgent;
 
-	if (max_phyclk_mhz == 0)
-		max_phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz;
+	if (max_clk_data.phyclk_mhz == 0)
+		max_clk_data.phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz;
 
 	*num_entries = 0;
-	entry.dispclk_mhz = max_dispclk_mhz;
-	entry.dscclk_mhz = max_dispclk_mhz / 3;
-	entry.dppclk_mhz = max_dppclk_mhz;
-	entry.dtbclk_mhz = max_dtbclk_mhz;
-	entry.phyclk_mhz = max_phyclk_mhz;
+	entry.dispclk_mhz = max_clk_data.dispclk_mhz;
+	entry.dscclk_mhz = max_clk_data.dispclk_mhz / 3;
+	entry.dppclk_mhz = max_clk_data.dppclk_mhz;
+	entry.dtbclk_mhz = max_clk_data.dtbclk_mhz;
+	entry.phyclk_mhz = max_clk_data.phyclk_mhz;
 	entry.phyclk_d18_mhz = dcn3_2_soc.clock_limits[0].phyclk_d18_mhz;
 	entry.phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz;
 
@@ -2401,7 +2460,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	}
 
 	// Insert the max DCFCLK
-	entry.dcfclk_mhz = max_dcfclk_mhz;
+	entry.dcfclk_mhz = max_clk_data.dcfclk_mhz;
 	entry.fabricclk_mhz = 0;
 	entry.dram_speed_mts = 0;
 
@@ -2429,7 +2488,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	// If FCLK fine grained, only insert max
 	else {
 		entry.dcfclk_mhz = 0;
-		entry.fabricclk_mhz = max_fclk_mhz;
+		entry.fabricclk_mhz = max_clk_data.fclk_mhz;
 		entry.dram_speed_mts = 0;
 
 		insert_entry_into_table_sorted(table, num_entries, &entry);
@@ -2441,9 +2500,9 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 
 	// Remove states that require higher clocks than are supported
 	for (i = *num_entries - 1; i >= 0 ; i--) {
-		if (table[i].dcfclk_mhz > max_dcfclk_mhz ||
-				table[i].fabricclk_mhz > max_fclk_mhz ||
-				table[i].dram_speed_mts > max_uclk_mhz * 16)
+		if (table[i].dcfclk_mhz > max_clk_data.dcfclk_mhz ||
+				table[i].fabricclk_mhz > max_clk_data.fclk_mhz ||
+				table[i].dram_speed_mts > max_clk_data.memclk_mhz * 16)
 			remove_entry_from_table_at_index(table, num_entries, i);
 	}
 
@@ -2756,7 +2815,8 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 				dcn3_2_soc.clock_limits[i].phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz;
 			}
 		} else {
-			build_synthetic_soc_states(bw_params, dcn3_2_soc.clock_limits, &dcn3_2_soc.num_states);
+			build_synthetic_soc_states(dc->debug.disable_dc_mode_overwrite, bw_params,
+					dcn3_2_soc.clock_limits, &dcn3_2_soc.num_states);
 		}
 
 		/* Re-init DML with updated bb */
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index 1aaff6f2d4532..f0683fd9d3f06 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -252,14 +252,48 @@ static void remove_entry_from_table_at_index(struct _vcs_dpi_voltage_scaling_st
 	memset(&table[--(*num_entries)], 0, sizeof(struct _vcs_dpi_voltage_scaling_st));
 }
 
-static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
+/*
+ * override_max_clk_values - Overwrite the max clock frequencies with the max DC mode timings
+ * Input:
+ *	max_clk_limit - struct containing the desired clock timings
+ * Output:
+ *	curr_clk_limit  - struct containing the timings that need to be overwritten
+ * Return: 0 upon success, non-zero for failure
+ */
+static int override_max_clk_values(struct clk_limit_table_entry *max_clk_limit,
+		struct clk_limit_table_entry *curr_clk_limit)
+{
+	if (NULL == max_clk_limit || NULL == curr_clk_limit)
+		return -1; //invalid parameters
+
+	//only overwrite if desired max clock frequency is initialized
+	if (max_clk_limit->dcfclk_mhz != 0)
+		curr_clk_limit->dcfclk_mhz = max_clk_limit->dcfclk_mhz;
+
+	if (max_clk_limit->fclk_mhz != 0)
+		curr_clk_limit->fclk_mhz = max_clk_limit->fclk_mhz;
+
+	if (max_clk_limit->memclk_mhz != 0)
+		curr_clk_limit->memclk_mhz = max_clk_limit->memclk_mhz;
+
+	if (max_clk_limit->socclk_mhz != 0)
+		curr_clk_limit->socclk_mhz = max_clk_limit->socclk_mhz;
+
+	if (max_clk_limit->dtbclk_mhz != 0)
+		curr_clk_limit->dtbclk_mhz = max_clk_limit->dtbclk_mhz;
+
+	if (max_clk_limit->dispclk_mhz != 0)
+		curr_clk_limit->dispclk_mhz = max_clk_limit->dispclk_mhz;
+
+	return 0;
+}
+
+static int build_synthetic_soc_states(bool disable_dc_mode_overwrite, struct clk_bw_params *bw_params,
 		struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries)
 {
 	int i, j;
 	struct _vcs_dpi_voltage_scaling_st entry = {0};
-
-	unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0,
-			max_phyclk_mhz = 0, max_dtbclk_mhz = 0, max_fclk_mhz = 0, max_uclk_mhz = 0;
+	struct clk_limit_table_entry max_clk_data = {0};
 
 	unsigned int min_dcfclk_mhz = 199, min_fclk_mhz = 299;
 
@@ -270,53 +304,78 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	unsigned int num_fclk_dpms = 0;
 	unsigned int num_dcfclk_dpms = 0;
 
+	unsigned int num_dc_uclk_dpms = 0;
+	unsigned int num_dc_fclk_dpms = 0;
+	unsigned int num_dc_dcfclk_dpms = 0;
+
 	for (i = 0; i < MAX_NUM_DPM_LVL; i++) {
-		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz)
-			max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
-		if (bw_params->clk_table.entries[i].fclk_mhz > max_fclk_mhz)
-			max_fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
-		if (bw_params->clk_table.entries[i].memclk_mhz > max_uclk_mhz)
-			max_uclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
-		if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz)
-			max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
-		if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz)
-			max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
-		if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz)
-			max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
-		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_dtbclk_mhz)
-			max_dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
-
-		if (bw_params->clk_table.entries[i].memclk_mhz > 0)
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > max_clk_data.dcfclk_mhz)
+			max_clk_data.dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz;
+		if (bw_params->clk_table.entries[i].fclk_mhz > max_clk_data.fclk_mhz)
+			max_clk_data.fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz;
+		if (bw_params->clk_table.entries[i].memclk_mhz > max_clk_data.memclk_mhz)
+			max_clk_data.memclk_mhz = bw_params->clk_table.entries[i].memclk_mhz;
+		if (bw_params->clk_table.entries[i].dispclk_mhz > max_clk_data.dispclk_mhz)
+			max_clk_data.dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz;
+		if (bw_params->clk_table.entries[i].dppclk_mhz > max_clk_data.dppclk_mhz)
+			max_clk_data.dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz;
+		if (bw_params->clk_table.entries[i].phyclk_mhz > max_clk_data.phyclk_mhz)
+			max_clk_data.phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz;
+		if (bw_params->clk_table.entries[i].dtbclk_mhz > max_clk_data.dtbclk_mhz)
+			max_clk_data.dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz;
+
+		if (bw_params->clk_table.entries[i].memclk_mhz > 0) {
 			num_uclk_dpms++;
-		if (bw_params->clk_table.entries[i].fclk_mhz > 0)
+			if (bw_params->clk_table.entries[i].memclk_mhz <= bw_params->dc_mode_limit.memclk_mhz)
+				num_dc_uclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].fclk_mhz > 0) {
 			num_fclk_dpms++;
-		if (bw_params->clk_table.entries[i].dcfclk_mhz > 0)
+			if (bw_params->clk_table.entries[i].fclk_mhz <= bw_params->dc_mode_limit.fclk_mhz)
+				num_dc_fclk_dpms++;
+		}
+		if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) {
 			num_dcfclk_dpms++;
+			if (bw_params->clk_table.entries[i].dcfclk_mhz <= bw_params->dc_mode_limit.dcfclk_mhz)
+				num_dc_dcfclk_dpms++;
+		}
+	}
+
+	if (!disable_dc_mode_overwrite) {
+		//Overwrite max frequencies with max DC mode frequencies for DC mode systems
+		override_max_clk_values(&bw_params->dc_mode_limit, &max_clk_data);
+		num_uclk_dpms = num_dc_uclk_dpms;
+		num_fclk_dpms = num_dc_fclk_dpms;
+		num_dcfclk_dpms = num_dc_dcfclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_memclk_levels = num_uclk_dpms;
+		bw_params->clk_table.num_entries_per_clk.num_fclk_levels = num_fclk_dpms;
 	}
 
 	if (num_dcfclk_dpms > 0 && bw_params->clk_table.entries[0].fclk_mhz > min_fclk_mhz)
 		min_fclk_mhz = bw_params->clk_table.entries[0].fclk_mhz;
 
-	if (!max_dcfclk_mhz || !max_dispclk_mhz || !max_dtbclk_mhz)
+	if (!max_clk_data.dcfclk_mhz || !max_clk_data.dispclk_mhz || !max_clk_data.dtbclk_mhz)
 		return -1;
 
-	if (max_dppclk_mhz == 0)
-		max_dppclk_mhz = max_dispclk_mhz;
+	if (max_clk_data.dppclk_mhz == 0)
+		max_clk_data.dppclk_mhz = max_clk_data.dispclk_mhz;
 
-	if (max_fclk_mhz == 0)
-		max_fclk_mhz = max_dcfclk_mhz * dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / dcn3_21_soc.pct_ideal_fabric_bw_after_urgent;
+	if (max_clk_data.fclk_mhz == 0)
+		max_clk_data.fclk_mhz = max_clk_data.dcfclk_mhz *
+				dcn3_2_soc.pct_ideal_sdp_bw_after_urgent /
+				dcn3_2_soc.pct_ideal_fabric_bw_after_urgent;
 
-	if (max_phyclk_mhz == 0)
-		max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz;
+	if (max_clk_data.phyclk_mhz == 0)
+		max_clk_data.phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz;
 
 	*num_entries = 0;
-	entry.dispclk_mhz = max_dispclk_mhz;
-	entry.dscclk_mhz = max_dispclk_mhz / 3;
-	entry.dppclk_mhz = max_dppclk_mhz;
-	entry.dtbclk_mhz = max_dtbclk_mhz;
-	entry.phyclk_mhz = max_phyclk_mhz;
-	entry.phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz;
-	entry.phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
+	entry.dispclk_mhz = max_clk_data.dispclk_mhz;
+	entry.dscclk_mhz = max_clk_data.dispclk_mhz / 3;
+	entry.dppclk_mhz = max_clk_data.dppclk_mhz;
+	entry.dtbclk_mhz = max_clk_data.dtbclk_mhz;
+	entry.phyclk_mhz = max_clk_data.phyclk_mhz;
+	entry.phyclk_d18_mhz = dcn3_2_soc.clock_limits[0].phyclk_d18_mhz;
+	entry.phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz;
 
 	// Insert all the DCFCLK STAs
 	for (i = 0; i < num_dcfclk_stas; i++) {
@@ -328,7 +387,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	}
 
 	// Insert the max DCFCLK
-	entry.dcfclk_mhz = max_dcfclk_mhz;
+	entry.dcfclk_mhz = max_clk_data.dcfclk_mhz;
 	entry.fabricclk_mhz = 0;
 	entry.dram_speed_mts = 0;
 
@@ -356,7 +415,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 	// If FCLK fine grained, only insert max
 	else {
 		entry.dcfclk_mhz = 0;
-		entry.fabricclk_mhz = max_fclk_mhz;
+		entry.fabricclk_mhz = max_clk_data.fclk_mhz;
 		entry.dram_speed_mts = 0;
 
 		dcn321_insert_entry_into_table_sorted(table, num_entries, &entry);
@@ -368,9 +427,9 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params,
 
 	// Remove states that require higher clocks than are supported
 	for (i = *num_entries - 1; i >= 0 ; i--) {
-		if (table[i].dcfclk_mhz > max_dcfclk_mhz ||
-				table[i].fabricclk_mhz > max_fclk_mhz ||
-				table[i].dram_speed_mts > max_uclk_mhz * 16)
+		if (table[i].dcfclk_mhz > max_clk_data.dcfclk_mhz ||
+				table[i].fabricclk_mhz > max_clk_data.fclk_mhz ||
+				table[i].dram_speed_mts > max_clk_data.memclk_mhz * 16)
 			remove_entry_from_table_at_index(table, num_entries, i);
 	}
 
@@ -689,7 +748,8 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 			dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz;
 		}
 	} else {
-		build_synthetic_soc_states(bw_params, dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states);
+		build_synthetic_soc_states(dc->debug.disable_dc_mode_overwrite, bw_params,
+			dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states);
 	}
 
 	/* Re-init DML with updated bb */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
index bef843cc32a1a..6faf40fa5c695 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
@@ -233,6 +233,7 @@ struct clk_bw_params {
 	struct clk_limit_table clk_table;
 	struct wm_table wm_table;
 	struct dummy_pstate_entry dummy_pstate_table[4];
+	struct clk_limit_table_entry dc_mode_limit;
 };
 /* Public interfaces */
 
-- 
GitLab


From 8d208a5eed4890f52a33ce847cbb8f8a5b1be6a7 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 8 Jun 2023 16:35:44 +0300
Subject: [PATCH 1636/1791] drm/i915: use pointer to i915 instead of rpm in
 wakeref

Currently a pointer to an intel_runtime_pm structure is stored in the
wake reference structures so the runtime data can be accessed.  We can
save the entire device information (drm_i915_private) instead, since
we'll need to reference the new workqueue we'll add in subsequent
patches.

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ec0eb5149120d04f3d9870d7671ef10103e6fc29.1686231190.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/gt/intel_engine_pm.c |  4 +---
 drivers/gpu/drm/i915/gt/intel_gt_pm.c     |  2 +-
 drivers/gpu/drm/i915/intel_runtime_pm.c   |  2 +-
 drivers/gpu/drm/i915/intel_wakeref.c      | 20 +++++++++++---------
 drivers/gpu/drm/i915/intel_wakeref.h      | 12 ++++++------
 5 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
index ee531a5c142c7..21af0ec522234 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
@@ -296,9 +296,7 @@ static const struct intel_wakeref_ops wf_ops = {
 
 void intel_engine_init__pm(struct intel_engine_cs *engine)
 {
-	struct intel_runtime_pm *rpm = engine->uncore->rpm;
-
-	intel_wakeref_init(&engine->wakeref, rpm, &wf_ops);
+	intel_wakeref_init(&engine->wakeref, engine->i915, &wf_ops);
 	intel_engine_init_heartbeat(engine);
 
 	intel_gsc_idle_msg_enable(engine);
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
index c2e69bafd02b0..5a942af0a14e6 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
@@ -137,7 +137,7 @@ void intel_gt_pm_init_early(struct intel_gt *gt)
 	 * runtime_pm is per-device rather than per-tile, so this is still the
 	 * correct structure.
 	 */
-	intel_wakeref_init(&gt->wakeref, &gt->i915->runtime_pm, &wf_ops);
+	intel_wakeref_init(&gt->wakeref, gt->i915, &wf_ops);
 	seqcount_mutex_init(&gt->stats.lock, &gt->wakeref.mutex);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index cf5122299b6b8..6d8e5e5c0cba2 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -658,5 +658,5 @@ void intel_runtime_pm_init_early(struct intel_runtime_pm *rpm)
 	init_intel_runtime_pm_wakeref(rpm);
 	INIT_LIST_HEAD(&rpm->lmem_userfault_list);
 	spin_lock_init(&rpm->lmem_userfault_lock);
-	intel_wakeref_auto_init(&rpm->userfault_wakeref, rpm);
+	intel_wakeref_auto_init(&rpm->userfault_wakeref, i915);
 }
diff --git a/drivers/gpu/drm/i915/intel_wakeref.c b/drivers/gpu/drm/i915/intel_wakeref.c
index dfd87d0822180..40aafe6760174 100644
--- a/drivers/gpu/drm/i915/intel_wakeref.c
+++ b/drivers/gpu/drm/i915/intel_wakeref.c
@@ -8,17 +8,18 @@
 
 #include "intel_runtime_pm.h"
 #include "intel_wakeref.h"
+#include "i915_drv.h"
 
 static void rpm_get(struct intel_wakeref *wf)
 {
-	wf->wakeref = intel_runtime_pm_get(wf->rpm);
+	wf->wakeref = intel_runtime_pm_get(&wf->i915->runtime_pm);
 }
 
 static void rpm_put(struct intel_wakeref *wf)
 {
 	intel_wakeref_t wakeref = fetch_and_zero(&wf->wakeref);
 
-	intel_runtime_pm_put(wf->rpm, wakeref);
+	intel_runtime_pm_put(&wf->i915->runtime_pm, wakeref);
 	INTEL_WAKEREF_BUG_ON(!wakeref);
 }
 
@@ -94,11 +95,11 @@ static void __intel_wakeref_put_work(struct work_struct *wrk)
 }
 
 void __intel_wakeref_init(struct intel_wakeref *wf,
-			  struct intel_runtime_pm *rpm,
+			  struct drm_i915_private *i915,
 			  const struct intel_wakeref_ops *ops,
 			  struct intel_wakeref_lockclass *key)
 {
-	wf->rpm = rpm;
+	wf->i915 = i915;
 	wf->ops = ops;
 
 	__mutex_init(&wf->mutex, "wakeref.mutex", &key->mutex);
@@ -137,17 +138,17 @@ static void wakeref_auto_timeout(struct timer_list *t)
 	wakeref = fetch_and_zero(&wf->wakeref);
 	spin_unlock_irqrestore(&wf->lock, flags);
 
-	intel_runtime_pm_put(wf->rpm, wakeref);
+	intel_runtime_pm_put(&wf->i915->runtime_pm, wakeref);
 }
 
 void intel_wakeref_auto_init(struct intel_wakeref_auto *wf,
-			     struct intel_runtime_pm *rpm)
+			     struct drm_i915_private *i915)
 {
 	spin_lock_init(&wf->lock);
 	timer_setup(&wf->timer, wakeref_auto_timeout, 0);
 	refcount_set(&wf->count, 0);
 	wf->wakeref = 0;
-	wf->rpm = rpm;
+	wf->i915 = i915;
 }
 
 void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout)
@@ -161,13 +162,14 @@ void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout)
 	}
 
 	/* Our mission is that we only extend an already active wakeref */
-	assert_rpm_wakelock_held(wf->rpm);
+	assert_rpm_wakelock_held(&wf->i915->runtime_pm);
 
 	if (!refcount_inc_not_zero(&wf->count)) {
 		spin_lock_irqsave(&wf->lock, flags);
 		if (!refcount_inc_not_zero(&wf->count)) {
 			INTEL_WAKEREF_BUG_ON(wf->wakeref);
-			wf->wakeref = intel_runtime_pm_get_if_in_use(wf->rpm);
+			wf->wakeref =
+				intel_runtime_pm_get_if_in_use(&wf->i915->runtime_pm);
 			refcount_set(&wf->count, 1);
 		}
 		spin_unlock_irqrestore(&wf->lock, flags);
diff --git a/drivers/gpu/drm/i915/intel_wakeref.h b/drivers/gpu/drm/i915/intel_wakeref.h
index 0b6b4852ab236..ec881b0973689 100644
--- a/drivers/gpu/drm/i915/intel_wakeref.h
+++ b/drivers/gpu/drm/i915/intel_wakeref.h
@@ -39,7 +39,7 @@ struct intel_wakeref {
 
 	intel_wakeref_t wakeref;
 
-	struct intel_runtime_pm *rpm;
+	struct drm_i915_private *i915;
 	const struct intel_wakeref_ops *ops;
 
 	struct delayed_work work;
@@ -51,13 +51,13 @@ struct intel_wakeref_lockclass {
 };
 
 void __intel_wakeref_init(struct intel_wakeref *wf,
-			  struct intel_runtime_pm *rpm,
+			  struct drm_i915_private *i915,
 			  const struct intel_wakeref_ops *ops,
 			  struct intel_wakeref_lockclass *key);
-#define intel_wakeref_init(wf, rpm, ops) do {				\
+#define intel_wakeref_init(wf, i915, ops) do {				\
 	static struct intel_wakeref_lockclass __key;			\
 									\
-	__intel_wakeref_init((wf), (rpm), (ops), &__key);		\
+	__intel_wakeref_init((wf), (i915), (ops), &__key);		\
 } while (0)
 
 int __intel_wakeref_get_first(struct intel_wakeref *wf);
@@ -262,7 +262,7 @@ __intel_wakeref_defer_park(struct intel_wakeref *wf)
 int intel_wakeref_wait_for_idle(struct intel_wakeref *wf);
 
 struct intel_wakeref_auto {
-	struct intel_runtime_pm *rpm;
+	struct drm_i915_private *i915;
 	struct timer_list timer;
 	intel_wakeref_t wakeref;
 	spinlock_t lock;
@@ -287,7 +287,7 @@ struct intel_wakeref_auto {
 void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout);
 
 void intel_wakeref_auto_init(struct intel_wakeref_auto *wf,
-			     struct intel_runtime_pm *rpm);
+			     struct drm_i915_private *i915);
 void intel_wakeref_auto_fini(struct intel_wakeref_auto *wf);
 
 #endif /* INTEL_WAKEREF_H */
-- 
GitLab


From 848a4e5c096ddf8ed1323123ae15b8d4318700ab Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 8 Jun 2023 16:35:45 +0300
Subject: [PATCH 1637/1791] drm/i915: add a dedicated workqueue inside
 drm_i915_private
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In order to avoid flush_scheduled_work() usage, add a dedicated
workqueue in the drm_i915_private structure.  In this way, we don't
need to use the system queue anymore.

This change is mostly mechanical and based on Tetsuo's original
patch[1].

v6 by Jani:
- Also create unordered_wq for mock device

Link: https://patchwork.freedesktop.org/series/114608/ [1]
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/c816ebe17ef08d363981942a096a586a7658a65e.1686231190.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  |  5 ++--
 .../drm/i915/display/intel_display_driver.c   |  2 +-
 drivers/gpu/drm/i915/display/intel_dmc.c      |  2 +-
 drivers/gpu/drm/i915/display/intel_dp.c       |  2 +-
 .../drm/i915/display/intel_dp_link_training.c |  3 ++-
 drivers/gpu/drm/i915/display/intel_drrs.c     |  4 +++-
 drivers/gpu/drm/i915/display/intel_fbc.c      |  2 +-
 drivers/gpu/drm/i915/display/intel_fbdev.c    |  3 ++-
 drivers/gpu/drm/i915/display/intel_hdcp.c     | 23 +++++++++++--------
 drivers/gpu/drm/i915/display/intel_hotplug.c  | 18 ++++++++++-----
 drivers/gpu/drm/i915/display/intel_opregion.c |  3 ++-
 drivers/gpu/drm/i915/display/intel_pps.c      |  4 +++-
 drivers/gpu/drm/i915/display/intel_psr.c      |  8 ++++---
 .../drm/i915/gt/intel_execlists_submission.c  |  5 ++--
 .../gpu/drm/i915/gt/intel_gt_buffer_pool.c    | 10 ++++----
 drivers/gpu/drm/i915/gt/intel_gt_irq.c        |  2 +-
 drivers/gpu/drm/i915/gt/intel_gt_requests.c   | 10 ++++----
 drivers/gpu/drm/i915/gt/intel_reset.c         |  2 +-
 drivers/gpu/drm/i915/gt/intel_rps.c           | 20 ++++++++--------
 drivers/gpu/drm/i915/gt/selftest_engine_cs.c  |  2 +-
 drivers/gpu/drm/i915/i915_driver.c            | 13 +++++++++++
 drivers/gpu/drm/i915/i915_drv.h               | 10 ++++++++
 drivers/gpu/drm/i915/i915_request.c           |  2 +-
 drivers/gpu/drm/i915/intel_wakeref.c          |  2 +-
 .../gpu/drm/i915/selftests/mock_gem_device.c  |  7 ++++++
 25 files changed, 110 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 5c7fdc82ac224..d8533603ad052 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7180,11 +7180,12 @@ intel_atomic_commit_ready(struct i915_sw_fence *fence,
 		break;
 	case FENCE_FREE:
 		{
+			struct drm_i915_private *i915 = to_i915(state->base.dev);
 			struct intel_atomic_helper *helper =
-				&to_i915(state->base.dev)->display.atomic_helper;
+				&i915->display.atomic_helper;
 
 			if (llist_add(&state->freed, &helper->free_list))
-				schedule_work(&helper->free_work);
+				queue_work(i915->unordered_wq, &helper->free_work);
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index dc8de861339d1..b909814ae02b2 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -442,7 +442,7 @@ void intel_display_driver_remove_noirq(struct drm_i915_private *i915)
 	intel_unregister_dsm_handler();
 
 	/* flush any delayed tasks or pending work */
-	flush_scheduled_work();
+	flush_workqueue(i915->unordered_wq);
 
 	intel_hdcp_component_fini(i915);
 
diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c
index 8a88de67ff0a9..5f479f3828bbe 100644
--- a/drivers/gpu/drm/i915/display/intel_dmc.c
+++ b/drivers/gpu/drm/i915/display/intel_dmc.c
@@ -1057,7 +1057,7 @@ void intel_dmc_init(struct drm_i915_private *i915)
 	i915->display.dmc.dmc = dmc;
 
 	drm_dbg_kms(&i915->drm, "Loading %s\n", dmc->fw_path);
-	schedule_work(&dmc->work);
+	queue_work(i915->unordered_wq, &dmc->work);
 
 	return;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index f4192fda1a76e..09dc6c88ad28e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -5251,7 +5251,7 @@ static void intel_dp_oob_hotplug_event(struct drm_connector *connector)
 	spin_lock_irq(&i915->irq_lock);
 	i915->display.hotplug.event_bits |= BIT(encoder->hpd_pin);
 	spin_unlock_irq(&i915->irq_lock);
-	queue_delayed_work(system_wq, &i915->display.hotplug.hotplug_work, 0);
+	queue_delayed_work(i915->unordered_wq, &i915->display.hotplug.hotplug_work, 0);
 }
 
 static const struct drm_connector_funcs intel_dp_connector_funcs = {
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 176b610642e75..a263773f4d68a 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1064,6 +1064,7 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp,
 						     const struct intel_crtc_state *crtc_state)
 {
 	struct intel_connector *intel_connector = intel_dp->attached_connector;
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 
 	if (!intel_digital_port_connected(&dp_to_dig_port(intel_dp)->base)) {
 		lt_dbg(intel_dp, DP_PHY_DPRX, "Link Training failed on disconnected sink.\n");
@@ -1081,7 +1082,7 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp,
 	}
 
 	/* Schedule a Hotplug Uevent to userspace to start modeset */
-	schedule_work(&intel_connector->modeset_retry_work);
+	queue_work(i915->unordered_wq, &intel_connector->modeset_retry_work);
 }
 
 /* Perform the link training on all LTTPRs and the DPRX on a link. */
diff --git a/drivers/gpu/drm/i915/display/intel_drrs.c b/drivers/gpu/drm/i915/display/intel_drrs.c
index 760e63cdc0c84..0d35b6be5b6aa 100644
--- a/drivers/gpu/drm/i915/display/intel_drrs.c
+++ b/drivers/gpu/drm/i915/display/intel_drrs.c
@@ -111,7 +111,9 @@ static void intel_drrs_set_state(struct intel_crtc *crtc,
 
 static void intel_drrs_schedule_work(struct intel_crtc *crtc)
 {
-	mod_delayed_work(system_wq, &crtc->drrs.work, msecs_to_jiffies(1000));
+	struct drm_i915_private *i915 = to_i915(crtc->base.dev);
+
+	mod_delayed_work(i915->unordered_wq, &crtc->drrs.work, msecs_to_jiffies(1000));
 }
 
 static unsigned int intel_drrs_frontbuffer_bits(const struct intel_crtc_state *crtc_state)
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 29aa029d249dd..7f8b2d7713c7a 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -1600,7 +1600,7 @@ static void __intel_fbc_handle_fifo_underrun_irq(struct intel_fbc *fbc)
 	if (READ_ONCE(fbc->underrun_detected))
 		return;
 
-	schedule_work(&fbc->underrun_work);
+	queue_work(fbc->i915->unordered_wq, &fbc->underrun_work);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c
index 2c3f7befed172..4d6209c9268b0 100644
--- a/drivers/gpu/drm/i915/display/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/display/intel_fbdev.c
@@ -694,7 +694,8 @@ void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous
 			/* Don't block our own workqueue as this can
 			 * be run in parallel with other i915.ko tasks.
 			 */
-			schedule_work(&dev_priv->display.fbdev.suspend_work);
+			queue_work(dev_priv->unordered_wq,
+				   &dev_priv->display.fbdev.suspend_work);
 			return;
 		}
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 17542c28dfd5c..5ed450111f77c 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -983,6 +983,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector,
 	struct drm_device *dev = connector->base.dev;
 	struct intel_digital_port *dig_port = intel_attached_dig_port(connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
 	drm_WARN_ON(connector->base.dev, !mutex_is_locked(&hdcp->mutex));
 
@@ -1001,7 +1002,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector,
 	hdcp->value = value;
 	if (update_property) {
 		drm_connector_get(&connector->base);
-		schedule_work(&hdcp->prop_work);
+		queue_work(i915->unordered_wq, &hdcp->prop_work);
 	}
 }
 
@@ -2090,16 +2091,17 @@ static void intel_hdcp_check_work(struct work_struct *work)
 					       struct intel_hdcp,
 					       check_work);
 	struct intel_connector *connector = intel_hdcp_to_connector(hdcp);
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
 	if (drm_connector_is_unregistered(&connector->base))
 		return;
 
 	if (!intel_hdcp2_check_link(connector))
-		schedule_delayed_work(&hdcp->check_work,
-				      DRM_HDCP2_CHECK_PERIOD_MS);
+		queue_delayed_work(i915->unordered_wq, &hdcp->check_work,
+				   DRM_HDCP2_CHECK_PERIOD_MS);
 	else if (!intel_hdcp_check_link(connector))
-		schedule_delayed_work(&hdcp->check_work,
-				      DRM_HDCP_CHECK_PERIOD_MS);
+		queue_delayed_work(i915->unordered_wq, &hdcp->check_work,
+				   DRM_HDCP_CHECK_PERIOD_MS);
 }
 
 static int i915_hdcp_component_bind(struct device *i915_kdev,
@@ -2398,7 +2400,8 @@ int intel_hdcp_enable(struct intel_atomic_state *state,
 	}
 
 	if (!ret) {
-		schedule_delayed_work(&hdcp->check_work, check_link_interval);
+		queue_delayed_work(i915->unordered_wq, &hdcp->check_work,
+				   check_link_interval);
 		intel_hdcp_update_value(connector,
 					DRM_MODE_CONTENT_PROTECTION_ENABLED,
 					true);
@@ -2447,6 +2450,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 				to_intel_connector(conn_state->connector);
 	struct intel_hdcp *hdcp = &connector->hdcp;
 	bool content_protection_type_changed, desired_and_not_enabled = false;
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
 	if (!connector->hdcp.shim)
 		return;
@@ -2473,7 +2477,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 		mutex_lock(&hdcp->mutex);
 		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
 		drm_connector_get(&connector->base);
-		schedule_work(&hdcp->prop_work);
+		queue_work(i915->unordered_wq, &hdcp->prop_work);
 		mutex_unlock(&hdcp->mutex);
 	}
 
@@ -2490,7 +2494,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state,
 		 */
 		if (!desired_and_not_enabled && !content_protection_type_changed) {
 			drm_connector_get(&connector->base);
-			schedule_work(&hdcp->prop_work);
+			queue_work(i915->unordered_wq, &hdcp->prop_work);
 		}
 	}
 
@@ -2602,6 +2606,7 @@ void intel_hdcp_atomic_check(struct drm_connector *connector,
 void intel_hdcp_handle_cp_irq(struct intel_connector *connector)
 {
 	struct intel_hdcp *hdcp = &connector->hdcp;
+	struct drm_i915_private *i915 = to_i915(connector->base.dev);
 
 	if (!hdcp->shim)
 		return;
@@ -2609,5 +2614,5 @@ void intel_hdcp_handle_cp_irq(struct intel_connector *connector)
 	atomic_inc(&connector->hdcp.cp_irq_count);
 	wake_up_all(&connector->hdcp.cp_irq_queue);
 
-	schedule_delayed_work(&hdcp->check_work, 0);
+	queue_delayed_work(i915->unordered_wq, &hdcp->check_work, 0);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index 23a5e1a875f13..1160fa20433bd 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -212,7 +212,8 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv)
 	/* Enable polling and queue hotplug re-enabling. */
 	if (hpd_disabled) {
 		drm_kms_helper_poll_enable(&dev_priv->drm);
-		mod_delayed_work(system_wq, &dev_priv->display.hotplug.reenable_work,
+		mod_delayed_work(dev_priv->unordered_wq,
+				 &dev_priv->display.hotplug.reenable_work,
 				 msecs_to_jiffies(HPD_STORM_REENABLE_DELAY));
 	}
 }
@@ -339,7 +340,8 @@ static void i915_digport_work_func(struct work_struct *work)
 		spin_lock_irq(&dev_priv->irq_lock);
 		dev_priv->display.hotplug.event_bits |= old_bits;
 		spin_unlock_irq(&dev_priv->irq_lock);
-		queue_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work, 0);
+		queue_delayed_work(dev_priv->unordered_wq,
+				   &dev_priv->display.hotplug.hotplug_work, 0);
 	}
 }
 
@@ -446,7 +448,8 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		dev_priv->display.hotplug.retry_bits |= retry;
 		spin_unlock_irq(&dev_priv->irq_lock);
 
-		mod_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work,
+		mod_delayed_work(dev_priv->unordered_wq,
+				 &dev_priv->display.hotplug.hotplug_work,
 				 msecs_to_jiffies(HPD_RETRY_DELAY));
 	}
 }
@@ -577,7 +580,8 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
 	if (queue_dig)
 		queue_work(dev_priv->display.hotplug.dp_wq, &dev_priv->display.hotplug.dig_port_work);
 	if (queue_hp)
-		queue_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work, 0);
+		queue_delayed_work(dev_priv->unordered_wq,
+				   &dev_priv->display.hotplug.hotplug_work, 0);
 }
 
 /**
@@ -687,7 +691,8 @@ void intel_hpd_poll_enable(struct drm_i915_private *dev_priv)
 	 * As well, there's no issue if we race here since we always reschedule
 	 * this worker anyway
 	 */
-	schedule_work(&dev_priv->display.hotplug.poll_init_work);
+	queue_work(dev_priv->unordered_wq,
+		   &dev_priv->display.hotplug.poll_init_work);
 }
 
 /**
@@ -715,7 +720,8 @@ void intel_hpd_poll_disable(struct drm_i915_private *dev_priv)
 		return;
 
 	WRITE_ONCE(dev_priv->display.hotplug.poll_enabled, false);
-	schedule_work(&dev_priv->display.hotplug.poll_init_work);
+	queue_work(dev_priv->unordered_wq,
+		   &dev_priv->display.hotplug.poll_init_work);
 }
 
 void intel_hpd_init_early(struct drm_i915_private *i915)
diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c
index b7973a05d022d..84078fb82b2f7 100644
--- a/drivers/gpu/drm/i915/display/intel_opregion.c
+++ b/drivers/gpu/drm/i915/display/intel_opregion.c
@@ -635,7 +635,8 @@ static void asle_work(struct work_struct *work)
 void intel_opregion_asle_intr(struct drm_i915_private *dev_priv)
 {
 	if (dev_priv->display.opregion.asle)
-		schedule_work(&dev_priv->display.opregion.asle_work);
+		queue_work(dev_priv->unordered_wq,
+			   &dev_priv->display.opregion.asle_work);
 }
 
 #define ACPI_EV_DISPLAY_SWITCH (1<<0)
diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index 5e7ba594e7e74..73f0f1714b377 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -867,6 +867,7 @@ static void edp_panel_vdd_work(struct work_struct *__work)
 
 static void edp_panel_vdd_schedule_off(struct intel_dp *intel_dp)
 {
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
 	unsigned long delay;
 
 	/*
@@ -882,7 +883,8 @@ static void edp_panel_vdd_schedule_off(struct intel_dp *intel_dp)
 	 * operations.
 	 */
 	delay = msecs_to_jiffies(intel_dp->pps.panel_power_cycle_delay * 5);
-	schedule_delayed_work(&intel_dp->pps.panel_vdd_work, delay);
+	queue_delayed_work(i915->unordered_wq,
+			   &intel_dp->pps.panel_vdd_work, delay);
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index ea0389c5f656b..d58ed9b62e677 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -341,7 +341,7 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir)
 		 */
 		intel_de_rmw(dev_priv, imr_reg, 0, psr_irq_psr_error_bit_get(intel_dp));
 
-		schedule_work(&intel_dp->psr.work);
+		queue_work(dev_priv->unordered_wq, &intel_dp->psr.work);
 	}
 }
 
@@ -2440,6 +2440,8 @@ static void
 tgl_dc3co_flush_locked(struct intel_dp *intel_dp, unsigned int frontbuffer_bits,
 		       enum fb_op_origin origin)
 {
+	struct drm_i915_private *i915 = dp_to_i915(intel_dp);
+
 	if (!intel_dp->psr.dc3co_exitline || !intel_dp->psr.psr2_enabled ||
 	    !intel_dp->psr.active)
 		return;
@@ -2453,7 +2455,7 @@ tgl_dc3co_flush_locked(struct intel_dp *intel_dp, unsigned int frontbuffer_bits,
 		return;
 
 	tgl_psr2_enable_dc3co(intel_dp);
-	mod_delayed_work(system_wq, &intel_dp->psr.dc3co_work,
+	mod_delayed_work(i915->unordered_wq, &intel_dp->psr.dc3co_work,
 			 intel_dp->psr.dc3co_exit_delay);
 }
 
@@ -2493,7 +2495,7 @@ static void _psr_flush_handle(struct intel_dp *intel_dp)
 		psr_force_hw_tracking_exit(intel_dp);
 
 		if (!intel_dp->psr.active && !intel_dp->psr.busy_frontbuffer_bits)
-			schedule_work(&intel_dp->psr.work);
+			queue_work(dev_priv->unordered_wq, &intel_dp->psr.work);
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 750326434677f..2ebd937f3b4cb 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -2327,6 +2327,7 @@ static u32 active_ccid(struct intel_engine_cs *engine)
 
 static void execlists_capture(struct intel_engine_cs *engine)
 {
+	struct drm_i915_private *i915 = engine->i915;
 	struct execlists_capture *cap;
 
 	if (!IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR))
@@ -2375,7 +2376,7 @@ static void execlists_capture(struct intel_engine_cs *engine)
 		goto err_rq;
 
 	INIT_WORK(&cap->work, execlists_capture_work);
-	schedule_work(&cap->work);
+	queue_work(i915->unordered_wq, &cap->work);
 	return;
 
 err_rq:
@@ -3680,7 +3681,7 @@ static void virtual_context_destroy(struct kref *kref)
 	 * lock, we can delegate the free of the engine to an RCU worker.
 	 */
 	INIT_RCU_WORK(&ve->rcu, rcu_virtual_context_destroy);
-	queue_rcu_work(system_wq, &ve->rcu);
+	queue_rcu_work(ve->context.engine->i915->unordered_wq, &ve->rcu);
 }
 
 static void virtual_engine_initial_hint(struct virtual_engine *ve)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
index cadfd85785b12..86b5a9ba323d5 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
@@ -88,10 +88,11 @@ static void pool_free_work(struct work_struct *wrk)
 {
 	struct intel_gt_buffer_pool *pool =
 		container_of(wrk, typeof(*pool), work.work);
+	struct intel_gt *gt = container_of(pool, struct intel_gt, buffer_pool);
 
 	if (pool_free_older_than(pool, HZ))
-		schedule_delayed_work(&pool->work,
-				      round_jiffies_up_relative(HZ));
+		queue_delayed_work(gt->i915->unordered_wq, &pool->work,
+				   round_jiffies_up_relative(HZ));
 }
 
 static void pool_retire(struct i915_active *ref)
@@ -99,6 +100,7 @@ static void pool_retire(struct i915_active *ref)
 	struct intel_gt_buffer_pool_node *node =
 		container_of(ref, typeof(*node), active);
 	struct intel_gt_buffer_pool *pool = node->pool;
+	struct intel_gt *gt = container_of(pool, struct intel_gt, buffer_pool);
 	struct list_head *list = bucket_for_size(pool, node->obj->base.size);
 	unsigned long flags;
 
@@ -116,8 +118,8 @@ static void pool_retire(struct i915_active *ref)
 	WRITE_ONCE(node->age, jiffies ?: 1); /* 0 reserved for active nodes */
 	spin_unlock_irqrestore(&pool->lock, flags);
 
-	schedule_delayed_work(&pool->work,
-			      round_jiffies_up_relative(HZ));
+	queue_delayed_work(gt->i915->unordered_wq, &pool->work,
+			   round_jiffies_up_relative(HZ));
 }
 
 void intel_gt_buffer_pool_mark_used(struct intel_gt_buffer_pool_node *node)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
index 8f888d36f16d4..62fd00c9e5192 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c
@@ -376,7 +376,7 @@ static void gen7_parity_error_irq_handler(struct intel_gt *gt, u32 iir)
 	if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT)
 		gt->i915->l3_parity.which_slice |= 1 << 0;
 
-	schedule_work(&gt->i915->l3_parity.error_work);
+	queue_work(gt->i915->unordered_wq, &gt->i915->l3_parity.error_work);
 }
 
 void gen6_gt_irq_handler(struct intel_gt *gt, u32 gt_iir)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_requests.c b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
index 1dfd01668c79c..d1a382dfaa1df 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_requests.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_requests.c
@@ -116,7 +116,7 @@ void intel_engine_add_retire(struct intel_engine_cs *engine,
 	GEM_BUG_ON(intel_engine_is_virtual(engine));
 
 	if (add_retire(engine, tl))
-		schedule_work(&engine->retire_work);
+		queue_work(engine->i915->unordered_wq, &engine->retire_work);
 }
 
 void intel_engine_init_retire(struct intel_engine_cs *engine)
@@ -207,8 +207,8 @@ static void retire_work_handler(struct work_struct *work)
 	struct intel_gt *gt =
 		container_of(work, typeof(*gt), requests.retire_work.work);
 
-	schedule_delayed_work(&gt->requests.retire_work,
-			      round_jiffies_up_relative(HZ));
+	queue_delayed_work(gt->i915->unordered_wq, &gt->requests.retire_work,
+			   round_jiffies_up_relative(HZ));
 	intel_gt_retire_requests(gt);
 }
 
@@ -224,8 +224,8 @@ void intel_gt_park_requests(struct intel_gt *gt)
 
 void intel_gt_unpark_requests(struct intel_gt *gt)
 {
-	schedule_delayed_work(&gt->requests.retire_work,
-			      round_jiffies_up_relative(HZ));
+	queue_delayed_work(gt->i915->unordered_wq, &gt->requests.retire_work,
+			   round_jiffies_up_relative(HZ));
 }
 
 void intel_gt_fini_requests(struct intel_gt *gt)
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 195ff72d7a14a..e2152f75ba2ef 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -1625,7 +1625,7 @@ void __intel_init_wedge(struct intel_wedge_me *w,
 	w->name = name;
 
 	INIT_DELAYED_WORK_ONSTACK(&w->work, intel_wedge_me);
-	schedule_delayed_work(&w->work, timeout);
+	queue_delayed_work(gt->i915->unordered_wq, &w->work, timeout);
 }
 
 void __intel_fini_wedge(struct intel_wedge_me *w)
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index e68a99205599f..e92e626d4994e 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -73,13 +73,14 @@ static void set(struct intel_uncore *uncore, i915_reg_t reg, u32 val)
 static void rps_timer(struct timer_list *t)
 {
 	struct intel_rps *rps = from_timer(rps, t, timer);
+	struct intel_gt *gt = rps_to_gt(rps);
 	struct intel_engine_cs *engine;
 	ktime_t dt, last, timestamp;
 	enum intel_engine_id id;
 	s64 max_busy[3] = {};
 
 	timestamp = 0;
-	for_each_engine(engine, rps_to_gt(rps), id) {
+	for_each_engine(engine, gt, id) {
 		s64 busy;
 		int i;
 
@@ -123,7 +124,7 @@ static void rps_timer(struct timer_list *t)
 
 			busy += div_u64(max_busy[i], 1 << i);
 		}
-		GT_TRACE(rps_to_gt(rps),
+		GT_TRACE(gt,
 			 "busy:%lld [%d%%], max:[%lld, %lld, %lld], interval:%d\n",
 			 busy, (int)div64_u64(100 * busy, dt),
 			 max_busy[0], max_busy[1], max_busy[2],
@@ -133,12 +134,12 @@ static void rps_timer(struct timer_list *t)
 		    rps->cur_freq < rps->max_freq_softlimit) {
 			rps->pm_iir |= GEN6_PM_RP_UP_THRESHOLD;
 			rps->pm_interval = 1;
-			schedule_work(&rps->work);
+			queue_work(gt->i915->unordered_wq, &rps->work);
 		} else if (100 * busy < rps->power.down_threshold * dt &&
 			   rps->cur_freq > rps->min_freq_softlimit) {
 			rps->pm_iir |= GEN6_PM_RP_DOWN_THRESHOLD;
 			rps->pm_interval = 1;
-			schedule_work(&rps->work);
+			queue_work(gt->i915->unordered_wq, &rps->work);
 		} else {
 			rps->last_adj = 0;
 		}
@@ -973,7 +974,7 @@ static int rps_set_boost_freq(struct intel_rps *rps, u32 val)
 	}
 	mutex_unlock(&rps->lock);
 	if (boost)
-		schedule_work(&rps->work);
+		queue_work(rps_to_gt(rps)->i915->unordered_wq, &rps->work);
 
 	return 0;
 }
@@ -1025,7 +1026,8 @@ void intel_rps_boost(struct i915_request *rq)
 			if (!atomic_fetch_inc(&slpc->num_waiters)) {
 				GT_TRACE(rps_to_gt(rps), "boost fence:%llx:%llx\n",
 					 rq->fence.context, rq->fence.seqno);
-				schedule_work(&slpc->boost_work);
+				queue_work(rps_to_gt(rps)->i915->unordered_wq,
+					   &slpc->boost_work);
 			}
 
 			return;
@@ -1041,7 +1043,7 @@ void intel_rps_boost(struct i915_request *rq)
 			 rq->fence.context, rq->fence.seqno);
 
 		if (READ_ONCE(rps->cur_freq) < rps->boost_freq)
-			schedule_work(&rps->work);
+			queue_work(rps_to_gt(rps)->i915->unordered_wq, &rps->work);
 
 		WRITE_ONCE(rps->boosts, rps->boosts + 1); /* debug only */
 	}
@@ -1900,7 +1902,7 @@ void gen11_rps_irq_handler(struct intel_rps *rps, u32 pm_iir)
 	gen6_gt_pm_mask_irq(gt, events);
 
 	rps->pm_iir |= events;
-	schedule_work(&rps->work);
+	queue_work(gt->i915->unordered_wq, &rps->work);
 }
 
 void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir)
@@ -1917,7 +1919,7 @@ void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir)
 		gen6_gt_pm_mask_irq(gt, events);
 		rps->pm_iir |= events;
 
-		schedule_work(&rps->work);
+		queue_work(gt->i915->unordered_wq, &rps->work);
 		spin_unlock(gt->irq_lock);
 	}
 
diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c
index 542ce6d2de192..78cdfc6f315f2 100644
--- a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c
@@ -27,7 +27,7 @@ static void perf_begin(struct intel_gt *gt)
 
 	/* Boost gpufreq to max [waitboost] and keep it fixed */
 	atomic_inc(&gt->rps.num_waiters);
-	schedule_work(&gt->rps.work);
+	queue_work(gt->i915->unordered_wq, &gt->rps.work);
 	flush_work(&gt->rps.work);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 4483be16a00b1..401f59cb4ab11 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -132,8 +132,20 @@ static int i915_workqueues_init(struct drm_i915_private *dev_priv)
 	if (dev_priv->display.hotplug.dp_wq == NULL)
 		goto out_free_wq;
 
+	/*
+	 * The unordered i915 workqueue should be used for all work
+	 * scheduling that do not require running in order, which used
+	 * to be scheduled on the system_wq before moving to a driver
+	 * instance due deprecation of flush_scheduled_work().
+	 */
+	dev_priv->unordered_wq = alloc_workqueue("i915-unordered", 0, 0);
+	if (dev_priv->unordered_wq == NULL)
+		goto out_free_dp_wq;
+
 	return 0;
 
+out_free_dp_wq:
+	destroy_workqueue(dev_priv->display.hotplug.dp_wq);
 out_free_wq:
 	destroy_workqueue(dev_priv->wq);
 out_err:
@@ -144,6 +156,7 @@ static int i915_workqueues_init(struct drm_i915_private *dev_priv)
 
 static void i915_workqueues_cleanup(struct drm_i915_private *dev_priv)
 {
+	destroy_workqueue(dev_priv->unordered_wq);
 	destroy_workqueue(dev_priv->display.hotplug.dp_wq);
 	destroy_workqueue(dev_priv->wq);
 }
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index f1205ed3ba717..f125f04b5f183 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -260,6 +260,16 @@ struct drm_i915_private {
 	 */
 	struct workqueue_struct *wq;
 
+	/**
+	 * unordered_wq - internal workqueue for unordered work
+	 *
+	 * This workqueue should be used for all unordered work
+	 * scheduling within i915, which used to be scheduled on the
+	 * system_wq before moving to a driver instance due
+	 * deprecation of flush_scheduled_work().
+	 */
+	struct workqueue_struct *unordered_wq;
+
 	/* pm private clock gating functions */
 	const struct drm_i915_clock_gating_funcs *clock_gating_funcs;
 
diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index 630a732aaecca..894068bb37b6f 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -290,7 +290,7 @@ static enum hrtimer_restart __rq_watchdog_expired(struct hrtimer *hrtimer)
 
 	if (!i915_request_completed(rq)) {
 		if (llist_add(&rq->watchdog.link, &gt->watchdog.list))
-			schedule_work(&gt->watchdog.work);
+			queue_work(gt->i915->unordered_wq, &gt->watchdog.work);
 	} else {
 		i915_request_put(rq);
 	}
diff --git a/drivers/gpu/drm/i915/intel_wakeref.c b/drivers/gpu/drm/i915/intel_wakeref.c
index 40aafe6760174..718f2f1b6174d 100644
--- a/drivers/gpu/drm/i915/intel_wakeref.c
+++ b/drivers/gpu/drm/i915/intel_wakeref.c
@@ -75,7 +75,7 @@ void __intel_wakeref_put_last(struct intel_wakeref *wf, unsigned long flags)
 
 	/* Assume we are not in process context and so cannot sleep. */
 	if (flags & INTEL_WAKEREF_PUT_ASYNC || !mutex_trylock(&wf->mutex)) {
-		mod_delayed_work(system_wq, &wf->work,
+		mod_delayed_work(wf->i915->unordered_wq, &wf->work,
 				 FIELD_GET(INTEL_WAKEREF_PUT_DELAY, flags));
 		return;
 	}
diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
index 0eda8b4ee17f5..09d4bbcdcdbfe 100644
--- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c
+++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c
@@ -69,6 +69,7 @@ static void mock_device_release(struct drm_device *dev)
 	i915_gem_drain_workqueue(i915);
 
 	mock_fini_ggtt(to_gt(i915)->ggtt);
+	destroy_workqueue(i915->unordered_wq);
 	destroy_workqueue(i915->wq);
 
 	intel_region_ttm_device_fini(i915);
@@ -208,6 +209,10 @@ struct drm_i915_private *mock_gem_device(void)
 	if (!i915->wq)
 		goto err_drv;
 
+	i915->unordered_wq = alloc_workqueue("mock-unordered", 0, 0);
+	if (!i915->unordered_wq)
+		goto err_wq;
+
 	mock_init_contexts(i915);
 
 	/* allocate the ggtt */
@@ -239,6 +244,8 @@ struct drm_i915_private *mock_gem_device(void)
 err_context:
 	intel_gt_driver_remove(to_gt(i915));
 err_unlock:
+	destroy_workqueue(i915->unordered_wq);
+err_wq:
 	destroy_workqueue(i915->wq);
 err_drv:
 	intel_region_ttm_device_fini(i915);
-- 
GitLab


From 0976b3dc021b2c5392d6abf443460e4208f85fe6 Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Thu, 8 Jun 2023 16:35:46 +0300
Subject: [PATCH 1638/1791] drm/i915/selftests: add local workqueue for SW
 fence selftest
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of using a global workqueue for the SW fence selftest,
allocate a separate one temporarily only while running the test.

Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/313f4a713053c2b4896ced5b0e9ff456eb85fe57.1686231190.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/selftests/i915_sw_fence.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
index daa985e5a19bc..8f5ce71fa453f 100644
--- a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
+++ b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c
@@ -523,12 +523,19 @@ static void task_ipc(struct work_struct *work)
 static int test_ipc(void *arg)
 {
 	struct task_ipc ipc;
+	struct workqueue_struct *wq;
 	int ret = 0;
 
+	wq = alloc_workqueue("i1915-selftest", 0, 0);
+	if (wq == NULL)
+		return -ENOMEM;
+
 	/* Test use of i915_sw_fence as an interprocess signaling mechanism */
 	ipc.in = alloc_fence();
-	if (!ipc.in)
-		return -ENOMEM;
+	if (!ipc.in) {
+		ret = -ENOMEM;
+		goto err_work;
+	}
 	ipc.out = alloc_fence();
 	if (!ipc.out) {
 		ret = -ENOMEM;
@@ -540,7 +547,7 @@ static int test_ipc(void *arg)
 
 	ipc.value = 0;
 	INIT_WORK_ONSTACK(&ipc.work, task_ipc);
-	schedule_work(&ipc.work);
+	queue_work(wq, &ipc.work);
 
 	wait_for_completion(&ipc.started);
 
@@ -563,6 +570,9 @@ static int test_ipc(void *arg)
 	free_fence(ipc.out);
 err_in:
 	free_fence(ipc.in);
+err_work:
+	destroy_workqueue(wq);
+
 	return ret;
 }
 
-- 
GitLab


From a52b66172927a2a7b861b34c908850d76bc33024 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 11 Apr 2023 01:40:34 +0300
Subject: [PATCH 1639/1791] drm/msm/adreno: fix sparse warnings in a6xx code

Sparse reports plenty of warnings against the a6xx code because of
a6xx_gmu::mmio and a6xx_gmu::rscc members. For some reason they were
defined as __iomem pointers rather than pointers to __iomem memory.
Correct the __iomem attribute.

Fixes: 02ef80c54e7c ("drm/msm/a6xx: update pdc/rscc GMU registers for A640/A650")
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/oe-kbuild-all/202304070550.NrbhJCvP-lkp@intel.com/
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/531583/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 0bc3eb443fec8..4759a8ce51e46 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -51,8 +51,8 @@ struct a6xx_gmu {
 
 	struct msm_gem_address_space *aspace;
 
-	void * __iomem mmio;
-	void * __iomem rscc;
+	void __iomem *mmio;
+	void __iomem *rscc;
 
 	int hfi_irq;
 	int gmu_irq;
-- 
GitLab


From 635455260d06b5189bdf9b400d591fabd9ef90b2 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 29 Mar 2023 03:55:29 +0300
Subject: [PATCH 1640/1791] drm/msm: drop unused ring variable in
 msm_ioctl_gem_submit()

The variable ring is not used by msm_parse_deps() and
msm_ioctl_gem_submit() and thus can be dropped.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/529340/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/msm_gem_submit.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 9f5933c75e3df..38aa048471a83 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -558,8 +558,7 @@ static struct drm_syncobj **msm_parse_deps(struct msm_gem_submit *submit,
                                            struct drm_file *file,
                                            uint64_t in_syncobjs_addr,
                                            uint32_t nr_in_syncobjs,
-                                           size_t syncobj_stride,
-                                           struct msm_ringbuffer *ring)
+                                           size_t syncobj_stride)
 {
 	struct drm_syncobj **syncobjs = NULL;
 	struct drm_msm_gem_submit_syncobj syncobj_desc = {0};
@@ -808,7 +807,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		syncobjs_to_reset = msm_parse_deps(submit, file,
 		                                   args->in_syncobjs,
 		                                   args->nr_in_syncobjs,
-		                                   args->syncobj_stride, ring);
+		                                   args->syncobj_stride);
 		if (IS_ERR(syncobjs_to_reset)) {
 			ret = PTR_ERR(syncobjs_to_reset);
 			goto out_unlock;
-- 
GitLab


From 38e27a6fbf2206b18417c5985dbcdeca0f2026b8 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 10 Apr 2023 21:52:26 +0300
Subject: [PATCH 1641/1791] drm/msm/a6xx: don't set
 IO_PGTABLE_QUIRK_ARM_OUTER_WBWA with coherent SMMU

If the Adreno SMMU is dma-coherent, allocation will fail unless we
disable IO_PGTABLE_QUIRK_ARM_OUTER_WBWA. Skip setting this quirk for the
coherent SMMUs (like we have on sm8350 platform).

Fixes: 54af0ceb7595 ("arm64: dts: qcom: sm8350: add GPU, GMU, GPU CC and SMMU nodes")
Reported-by: David Heidelberg <david@ixit.cz>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: David Heidelberg <david@ixit.cz>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Tested-by: Konrad Dybcio <konrad.dybcio@linaro.org> # SM8450 HDK
Patchwork: https://patchwork.freedesktop.org/patch/531562/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 52da3795b175d..411b7a5fa2f32 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1744,7 +1744,8 @@ a6xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev)
 	 * This allows GPU to set the bus attributes required to use system
 	 * cache on behalf of the iommu page table walker.
 	 */
-	if (!IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice))
+	if (!IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice) &&
+	    !device_iommu_capable(&pdev->dev, IOMMU_CAP_CACHE_COHERENCY))
 		quirks |= IO_PGTABLE_QUIRK_ARM_OUTER_WBWA;
 
 	return adreno_iommu_create_address_space(gpu, pdev, quirks);
-- 
GitLab


From 171f580e32096e0cbd19e458d704b9330197d627 Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Mon, 17 Apr 2023 15:55:10 -0700
Subject: [PATCH 1642/1791] drm/msm: Move cmdstream dumping out of sched
 kthread

This is something that can block for arbitrary amounts of time as
userspace consumes from the FIFO.  So we don't really want this to
be in the fence signaling path.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/532617/
---
 drivers/gpu/drm/msm/msm_gem_submit.c |  2 ++
 drivers/gpu/drm/msm/msm_gpu.c        |  4 ----
 drivers/gpu/drm/msm/msm_rd.c         | 24 ++++++++----------------
 3 files changed, 10 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 38aa048471a83..3f1aa4de3b873 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -948,6 +948,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	/* The scheduler owns a ref now: */
 	msm_gem_submit_get(submit);
 
+	msm_rd_dump_submit(priv->rd, submit, NULL);
+
 	drm_sched_entity_push_job(&submit->base);
 
 	args->fence = submit->fence_id;
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 26ebda40be4f0..b1647b8510181 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -748,8 +748,6 @@ void msm_gpu_retire(struct msm_gpu *gpu)
 /* add bo's to gpu's ring, and kick gpu: */
 void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 {
-	struct drm_device *dev = gpu->dev;
-	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_ringbuffer *ring = submit->ring;
 	unsigned long flags;
 
@@ -761,8 +759,6 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 
 	submit->seqno = submit->hw_fence->seqno;
 
-	msm_rd_dump_submit(priv->rd, submit, NULL);
-
 	update_sw_cntrs(gpu);
 
 	/*
diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c
index db2f847c8535f..8d5687d5ed786 100644
--- a/drivers/gpu/drm/msm/msm_rd.c
+++ b/drivers/gpu/drm/msm/msm_rd.c
@@ -83,15 +83,10 @@ struct msm_rd_state {
 
 	bool open;
 
-	/* current submit to read out: */
-	struct msm_gem_submit *submit;
-
 	/* fifo access is synchronized on the producer side by
-	 * gpu->lock held by submit code (otherwise we could
-	 * end up w/ cmds logged in different order than they
-	 * were executed).  And read_lock synchronizes the reads
+	 * write_lock.  And read_lock synchronizes the reads
 	 */
-	struct mutex read_lock;
+	struct mutex read_lock, write_lock;
 
 	wait_queue_head_t fifo_event;
 	struct circ_buf fifo;
@@ -243,6 +238,7 @@ static void rd_cleanup(struct msm_rd_state *rd)
 		return;
 
 	mutex_destroy(&rd->read_lock);
+	mutex_destroy(&rd->write_lock);
 	kfree(rd);
 }
 
@@ -258,6 +254,7 @@ static struct msm_rd_state *rd_init(struct drm_minor *minor, const char *name)
 	rd->fifo.buf = rd->buf;
 
 	mutex_init(&rd->read_lock);
+	mutex_init(&rd->write_lock);
 
 	init_waitqueue_head(&rd->fifo_event);
 
@@ -338,19 +335,15 @@ static void snapshot_buf(struct msm_rd_state *rd,
 	if (!(submit->bos[idx].flags & MSM_SUBMIT_BO_READ))
 		return;
 
-	msm_gem_lock(&obj->base);
 	buf = msm_gem_get_vaddr_active(&obj->base);
 	if (IS_ERR(buf))
-		goto out_unlock;
+		return;
 
 	buf += offset;
 
 	rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size);
 
 	msm_gem_put_vaddr_locked(&obj->base);
-
-out_unlock:
-	msm_gem_unlock(&obj->base);
 }
 
 /* called under gpu->lock */
@@ -364,10 +357,7 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit,
 	if (!rd->open)
 		return;
 
-	/* writing into fifo is serialized by caller, and
-	 * rd->read_lock is used to serialize the reads
-	 */
-	WARN_ON(!mutex_is_locked(&submit->gpu->lock));
+	mutex_lock(&rd->write_lock);
 
 	if (fmt) {
 		va_list args;
@@ -424,5 +414,7 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit,
 			break;
 		}
 	}
+
+	mutex_unlock(&rd->write_lock);
 }
 #endif
-- 
GitLab


From 5e7665b5e484b32872bc1f44263940f094ccd814 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@linaro.org>
Date: Tue, 30 May 2023 20:09:43 -0700
Subject: [PATCH 1643/1791] drm/msm/adreno: Add Adreno A690 support

Introduce support for the Adreno A690, found in Qualcomm SC8280XP.

Tested-by: Steev Klimaszewski <steev@kali.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/540335/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c      | 113 ++++++++++++++++++++-
 drivers/gpu/drm/msm/adreno/a6xx_hfi.c      |  33 ++++++
 drivers/gpu/drm/msm/adreno/adreno_device.c |  14 +++
 drivers/gpu/drm/msm/adreno/adreno_gpu.h    |  11 +-
 4 files changed, 166 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 411b7a5fa2f32..3c2488e63e062 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -588,6 +588,63 @@ const struct adreno_reglist a660_hwcg[] = {
 	{},
 };
 
+const struct adreno_reglist a690_hwcg[] = {
+	{REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220},
+	{REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080},
+	{REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF},
+	{REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111},
+	{REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777},
+	{REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222},
+	{REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220},
+	{REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00},
+	{REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555},
+	{REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011},
+	{REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044},
+	{REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222},
+	{REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002},
+	{REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000},
+	{REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200},
+	{REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004},
+	{REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111},
+	{REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004},
+	{REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002},
+	{REG_A6XX_RBBM_CLOCK_CNTL, 0x8AA8AA82},
+	{REG_A6XX_RBBM_ISDB_CNT, 0x00000182},
+	{REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000},
+	{REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111},
+	{REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555},
+	{REG_A6XX_GPU_GMU_AO_GMU_CGC_MODE_CNTL, 0x20200},
+	{REG_A6XX_GPU_GMU_AO_GMU_CGC_DELAY_CNTL, 0x10111},
+	{REG_A6XX_GPU_GMU_AO_GMU_CGC_HYST_CNTL, 0x5555},
+	{}
+};
+
 static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -747,6 +804,45 @@ static const u32 a660_protect[] = {
 	A6XX_PROTECT_NORDWR(0x1f8c0, 0x0000), /* note: infinite range */
 };
 
+/* These are for a690 */
+static const u32 a690_protect[] = {
+	A6XX_PROTECT_RDONLY(0x00000, 0x004ff),
+	A6XX_PROTECT_RDONLY(0x00501, 0x00001),
+	A6XX_PROTECT_RDONLY(0x0050b, 0x002f4),
+	A6XX_PROTECT_NORDWR(0x0050e, 0x00000),
+	A6XX_PROTECT_NORDWR(0x00510, 0x00000),
+	A6XX_PROTECT_NORDWR(0x00534, 0x00000),
+	A6XX_PROTECT_NORDWR(0x00800, 0x00082),
+	A6XX_PROTECT_NORDWR(0x008a0, 0x00008),
+	A6XX_PROTECT_NORDWR(0x008ab, 0x00024),
+	A6XX_PROTECT_RDONLY(0x008d0, 0x000bc),
+	A6XX_PROTECT_NORDWR(0x00900, 0x0004d),
+	A6XX_PROTECT_NORDWR(0x0098d, 0x00272),
+	A6XX_PROTECT_NORDWR(0x00e00, 0x00001),
+	A6XX_PROTECT_NORDWR(0x00e03, 0x0000c),
+	A6XX_PROTECT_NORDWR(0x03c00, 0x000c3),
+	A6XX_PROTECT_RDONLY(0x03cc4, 0x01fff),
+	A6XX_PROTECT_NORDWR(0x08630, 0x001cf),
+	A6XX_PROTECT_NORDWR(0x08e00, 0x00000),
+	A6XX_PROTECT_NORDWR(0x08e08, 0x00008),
+	A6XX_PROTECT_NORDWR(0x08e50, 0x0001f),
+	A6XX_PROTECT_NORDWR(0x08e80, 0x0027f),
+	A6XX_PROTECT_NORDWR(0x09624, 0x001db),
+	A6XX_PROTECT_NORDWR(0x09e60, 0x00011),
+	A6XX_PROTECT_NORDWR(0x09e78, 0x00187),
+	A6XX_PROTECT_NORDWR(0x0a630, 0x001cf),
+	A6XX_PROTECT_NORDWR(0x0ae02, 0x00000),
+	A6XX_PROTECT_NORDWR(0x0ae50, 0x0012f),
+	A6XX_PROTECT_NORDWR(0x0b604, 0x00000),
+	A6XX_PROTECT_NORDWR(0x0b608, 0x00006),
+	A6XX_PROTECT_NORDWR(0x0be02, 0x00001),
+	A6XX_PROTECT_NORDWR(0x0be20, 0x0015f),
+	A6XX_PROTECT_NORDWR(0x0d000, 0x005ff),
+	A6XX_PROTECT_NORDWR(0x0f000, 0x00bff),
+	A6XX_PROTECT_RDONLY(0x0fc00, 0x01fff),
+	A6XX_PROTECT_NORDWR(0x11c00, 0x00000), /*note: infiite range */
+};
+
 static void a6xx_set_cp_protect(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -758,6 +854,11 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu)
 		count = ARRAY_SIZE(a650_protect);
 		count_max = 48;
 		BUILD_BUG_ON(ARRAY_SIZE(a650_protect) > 48);
+	} else if (adreno_is_a690(adreno_gpu)) {
+		regs = a690_protect;
+		count = ARRAY_SIZE(a690_protect);
+		count_max = 48;
+		BUILD_BUG_ON(ARRAY_SIZE(a690_protect) > 48);
 	} else if (adreno_is_a660_family(adreno_gpu)) {
 		regs = a660_protect;
 		count = ARRAY_SIZE(a660_protect);
@@ -806,6 +907,13 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
 		uavflagprd_inv = 2;
 	}
 
+	if (adreno_is_a690(adreno_gpu)) {
+		lower_bit = 2;
+		amsbc = 1;
+		rgb565_predicator = 1;
+		uavflagprd_inv = 2;
+	}
+
 	if (adreno_is_7c3(adreno_gpu)) {
 		lower_bit = 1;
 		amsbc = 1;
@@ -1084,7 +1192,7 @@ static int hw_init(struct msm_gpu *gpu)
 	/* Setting the primFifo thresholds default values,
 	 * and vccCacheSkipDis=1 bit (0x200) for A640 and newer
 	*/
-	if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu))
+	if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu) || adreno_is_a690(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200);
 	else if (adreno_is_a640_family(adreno_gpu) || adreno_is_7c3(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00200200);
@@ -1994,7 +2102,8 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	info = adreno_info(config->rev);
 
 	if (info && (info->revn == 650 || info->revn == 660 ||
-			adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev)))
+		     info->revn == 690 ||
+		     adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev)))
 		adreno_gpu->base.hw_apriv = true;
 
 	a6xx_llc_slices_init(pdev, a6xx_gpu);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
index 2cc83e0496133..25b235b49ebc2 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
@@ -414,6 +414,37 @@ static void a650_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
 	msg->cnoc_cmds_data[1][0] =  0x60000001;
 }
 
+static void a690_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
+{
+	/*
+	 * Send a single "off" entry just to get things running
+	 * TODO: bus scaling
+	 */
+	msg->bw_level_num = 1;
+
+	msg->ddr_cmds_num = 3;
+	msg->ddr_wait_bitmask = 0x01;
+
+	msg->ddr_cmds_addrs[0] = 0x50004;
+	msg->ddr_cmds_addrs[1] = 0x50000;
+	msg->ddr_cmds_addrs[2] = 0x500ac;
+
+	msg->ddr_cmds_data[0][0] =  0x40000000;
+	msg->ddr_cmds_data[0][1] =  0x40000000;
+	msg->ddr_cmds_data[0][2] =  0x40000000;
+
+	/*
+	 * These are the CX (CNOC) votes - these are used by the GMU but the
+	 * votes are known and fixed for the target
+	 */
+	msg->cnoc_cmds_num = 1;
+	msg->cnoc_wait_bitmask = 0x01;
+
+	msg->cnoc_cmds_addrs[0] = 0x5003c;
+	msg->cnoc_cmds_data[0][0] =  0x40000000;
+	msg->cnoc_cmds_data[1][0] =  0x60000001;
+}
+
 static void a660_build_bw_table(struct a6xx_hfi_msg_bw_table *msg)
 {
 	/*
@@ -531,6 +562,8 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu)
 		adreno_7c3_build_bw_table(&msg);
 	else if (adreno_is_a660(adreno_gpu))
 		a660_build_bw_table(&msg);
+	else if (adreno_is_a690(adreno_gpu))
+		a690_build_bw_table(&msg);
 	else
 		a6xx_build_bw_table(&msg);
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 8cff86e9d35c5..e5a865024e942 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -355,6 +355,20 @@ static const struct adreno_info gpulist[] = {
 		.init = a6xx_gpu_init,
 		.zapfw = "a640_zap.mdt",
 		.hwcg = a640_hwcg,
+	}, {
+		.rev = ADRENO_REV(6, 9, 0, ANY_ID),
+		.revn = 690,
+		.name = "A690",
+		.fw = {
+			[ADRENO_FW_SQE] = "a660_sqe.fw",
+			[ADRENO_FW_GMU] = "a690_gmu.bin",
+		},
+		.gmem = SZ_4M,
+		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
+		.init = a6xx_gpu_init,
+		.zapfw = "a690_zap.mdt",
+		.hwcg = a690_hwcg,
+		.address_space_size = SZ_16G,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index f62612a5c70f3..ac9c429ca07b9 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -55,7 +55,7 @@ struct adreno_reglist {
 	u32 value;
 };
 
-extern const struct adreno_reglist a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[], a660_hwcg[];
+extern const struct adreno_reglist a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[], a660_hwcg[], a690_hwcg[];
 
 struct adreno_info {
 	struct adreno_rev rev;
@@ -272,6 +272,11 @@ static inline int adreno_is_a660(struct adreno_gpu *gpu)
 	return gpu->revn == 660;
 }
 
+static inline int adreno_is_a690(struct adreno_gpu *gpu)
+{
+	return gpu->revn == 690;
+};
+
 /* check for a615, a616, a618, a619 or any derivatives */
 static inline int adreno_is_a615_family(struct adreno_gpu *gpu)
 {
@@ -280,13 +285,13 @@ static inline int adreno_is_a615_family(struct adreno_gpu *gpu)
 
 static inline int adreno_is_a660_family(struct adreno_gpu *gpu)
 {
-	return adreno_is_a660(gpu) || adreno_is_7c3(gpu);
+	return adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_7c3(gpu);
 }
 
 /* check for a650, a660, or any derivatives */
 static inline int adreno_is_a650_family(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 650 || gpu->revn == 620 || adreno_is_a660_family(gpu);
+	return gpu->revn == 650 || gpu->revn == 620  || adreno_is_a660_family(gpu);
 }
 
 u64 adreno_private_address_space_size(struct msm_gpu *gpu);
-- 
GitLab


From bf08e9791ca4e598f7088f24189dc3ef802b2a9f Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@chromium.org>
Date: Sat, 3 Jun 2023 09:43:56 -0700
Subject: [PATCH 1644/1791] drm/msm/a6xx: Fix a690 CP_PROTECT settings

Allow access to CP_PERFCTR_CP_SEL[n] and RB_PERFCTR_RB_SEL which are
used by fdperf and pps-provider (perfetto).

Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 3c2488e63e062..ab5c446e4409b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -815,7 +815,7 @@ static const u32 a690_protect[] = {
 	A6XX_PROTECT_NORDWR(0x00800, 0x00082),
 	A6XX_PROTECT_NORDWR(0x008a0, 0x00008),
 	A6XX_PROTECT_NORDWR(0x008ab, 0x00024),
-	A6XX_PROTECT_RDONLY(0x008d0, 0x000bc),
+	A6XX_PROTECT_RDONLY(0x008de, 0x000ae),
 	A6XX_PROTECT_NORDWR(0x00900, 0x0004d),
 	A6XX_PROTECT_NORDWR(0x0098d, 0x00272),
 	A6XX_PROTECT_NORDWR(0x00e00, 0x00001),
@@ -824,7 +824,7 @@ static const u32 a690_protect[] = {
 	A6XX_PROTECT_RDONLY(0x03cc4, 0x01fff),
 	A6XX_PROTECT_NORDWR(0x08630, 0x001cf),
 	A6XX_PROTECT_NORDWR(0x08e00, 0x00000),
-	A6XX_PROTECT_NORDWR(0x08e08, 0x00008),
+	A6XX_PROTECT_NORDWR(0x08e08, 0x00007),
 	A6XX_PROTECT_NORDWR(0x08e50, 0x0001f),
 	A6XX_PROTECT_NORDWR(0x08e80, 0x0027f),
 	A6XX_PROTECT_NORDWR(0x09624, 0x001db),
-- 
GitLab


From 736a9327365644b460e4498b1ce172ca411efcbc Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Sun, 9 Apr 2023 04:13:29 +0300
Subject: [PATCH 1645/1791] drm/msm/a5xx: really check for A510 in
 a5xx_gpu_init

The commit 010c8bbad2cb ("drm: msm: adreno: Disable preemption on Adreno
510") added special handling for a510 (this SKU doesn't seem to support
preemption, so the driver should clamp nr_rings to 1). However the
gpu->revn is not yet set (it is set later, in adreno_gpu_init()) and
thus the condition is always false. Check config->rev instead.

Fixes: 010c8bbad2cb ("drm: msm: adreno: Disable preemption on Adreno 510")
Reported-by: Adam Skladowski <a39.skl@gmail.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Adam Skladowski <a39.skl@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/531511/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
index 1e8d2982d603c..a99310b687932 100644
--- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c
@@ -1743,6 +1743,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct platform_device *pdev = priv->gpu_pdev;
+	struct adreno_platform_config *config = pdev->dev.platform_data;
 	struct a5xx_gpu *a5xx_gpu = NULL;
 	struct adreno_gpu *adreno_gpu;
 	struct msm_gpu *gpu;
@@ -1769,7 +1770,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev)
 
 	nr_rings = 4;
 
-	if (adreno_is_a510(adreno_gpu))
+	if (adreno_cmp_rev(ADRENO_REV(5, 1, 0, ANY_ID), config->rev))
 		nr_rings = 1;
 
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, nr_rings);
-- 
GitLab


From cc943f43ece7be313617ccb8425c31e02b73cbbe Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 11 Apr 2023 19:19:02 +0300
Subject: [PATCH 1646/1791] drm/msm/adreno: warn if chip revn is verified
 before being set

The commit 010c8bbad2cb ("drm: msm: adreno: Disable preemption on Adreno
510") tried to check GPU's revn before revn being set. Add WARN_ON_ONCE
to prevent such bugs from happening again. A separate helper is
necessary so that the warning is displayed really just once instead of
being displayed for each of comparisons.

Suggested-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/531705/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/adreno_gpu.h | 63 ++++++++++++++++---------
 1 file changed, 40 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index ac9c429ca07b9..4fee467c8075b 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -145,40 +145,51 @@ struct adreno_platform_config {
 
 bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2);
 
+static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn)
+{
+	WARN_ON_ONCE(!gpu->revn);
+
+	return gpu->revn == revn;
+}
+
 static inline bool adreno_is_a2xx(struct adreno_gpu *gpu)
 {
+	WARN_ON_ONCE(!gpu->revn);
+
 	return (gpu->revn < 300);
 }
 
 static inline bool adreno_is_a20x(struct adreno_gpu *gpu)
 {
+	WARN_ON_ONCE(!gpu->revn);
+
 	return (gpu->revn < 210);
 }
 
 static inline bool adreno_is_a225(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 225;
+	return adreno_is_revn(gpu, 225);
 }
 
 static inline bool adreno_is_a305(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 305;
+	return adreno_is_revn(gpu, 305);
 }
 
 static inline bool adreno_is_a306(struct adreno_gpu *gpu)
 {
 	/* yes, 307, because a305c is 306 */
-	return gpu->revn == 307;
+	return adreno_is_revn(gpu, 307);
 }
 
 static inline bool adreno_is_a320(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 320;
+	return adreno_is_revn(gpu, 320);
 }
 
 static inline bool adreno_is_a330(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 330;
+	return adreno_is_revn(gpu, 330);
 }
 
 static inline bool adreno_is_a330v2(struct adreno_gpu *gpu)
@@ -188,77 +199,78 @@ static inline bool adreno_is_a330v2(struct adreno_gpu *gpu)
 
 static inline int adreno_is_a405(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 405;
+	return adreno_is_revn(gpu, 405);
 }
 
 static inline int adreno_is_a420(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 420;
+	return adreno_is_revn(gpu, 420);
 }
 
 static inline int adreno_is_a430(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 430;
+	return adreno_is_revn(gpu, 430);
 }
 
 static inline int adreno_is_a506(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 506;
+	return adreno_is_revn(gpu, 506);
 }
 
 static inline int adreno_is_a508(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 508;
+	return adreno_is_revn(gpu, 508);
 }
 
 static inline int adreno_is_a509(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 509;
+	return adreno_is_revn(gpu, 509);
 }
 
 static inline int adreno_is_a510(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 510;
+	return adreno_is_revn(gpu, 510);
 }
 
 static inline int adreno_is_a512(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 512;
+	return adreno_is_revn(gpu, 512);
 }
 
 static inline int adreno_is_a530(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 530;
+	return adreno_is_revn(gpu, 530);
 }
 
 static inline int adreno_is_a540(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 540;
+	return adreno_is_revn(gpu, 540);
 }
 
 static inline int adreno_is_a618(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 618;
+	return adreno_is_revn(gpu, 618);
 }
 
 static inline int adreno_is_a619(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 619;
+	return adreno_is_revn(gpu, 619);
 }
 
 static inline int adreno_is_a630(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 630;
+	return adreno_is_revn(gpu, 630);
 }
 
 static inline int adreno_is_a640_family(struct adreno_gpu *gpu)
 {
-	return (gpu->revn == 640) || (gpu->revn == 680);
+	return adreno_is_revn(gpu, 640) ||
+		adreno_is_revn(gpu, 680);
 }
 
 static inline int adreno_is_a650(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 650;
+	return adreno_is_revn(gpu, 650);
 }
 
 static inline int adreno_is_7c3(struct adreno_gpu *gpu)
@@ -269,7 +281,7 @@ static inline int adreno_is_7c3(struct adreno_gpu *gpu)
 
 static inline int adreno_is_a660(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 660;
+	return adreno_is_revn(gpu, 660);
 }
 
 static inline int adreno_is_a690(struct adreno_gpu *gpu)
@@ -280,7 +292,10 @@ static inline int adreno_is_a690(struct adreno_gpu *gpu)
 /* check for a615, a616, a618, a619 or any derivatives */
 static inline int adreno_is_a615_family(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 615 || gpu->revn == 616 || gpu->revn == 618 || gpu->revn == 619;
+	return adreno_is_revn(gpu, 615) ||
+		adreno_is_revn(gpu, 616) ||
+		adreno_is_revn(gpu, 618) ||
+		adreno_is_revn(gpu, 619);
 }
 
 static inline int adreno_is_a660_family(struct adreno_gpu *gpu)
@@ -291,7 +306,9 @@ static inline int adreno_is_a660_family(struct adreno_gpu *gpu)
 /* check for a650, a660, or any derivatives */
 static inline int adreno_is_a650_family(struct adreno_gpu *gpu)
 {
-	return gpu->revn == 650 || gpu->revn == 620  || adreno_is_a660_family(gpu);
+	return adreno_is_revn(gpu, 650) ||
+		adreno_is_revn(gpu, 620) ||
+		adreno_is_a660_family(gpu);
 }
 
 u64 adreno_private_address_space_size(struct msm_gpu *gpu);
-- 
GitLab


From 1b90e8f8879c64d4f77dd1f25134397ac075b7bd Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 11 Apr 2023 19:19:03 +0300
Subject: [PATCH 1647/1791] drm/msm/adreno: change adreno_is_* functions to
 accept const argument

All adreno_is_*() functions do not modify their argument in any way, so
they can be changed to accept const struct adreno_gpu pointer.

Suggested-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/531706/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/adreno_gpu.h | 56 ++++++++++++-------------
 1 file changed, 28 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 4fee467c8075b..1283e5fe22d24 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -152,134 +152,134 @@ static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn)
 	return gpu->revn == revn;
 }
 
-static inline bool adreno_is_a2xx(struct adreno_gpu *gpu)
+static inline bool adreno_is_a2xx(const struct adreno_gpu *gpu)
 {
 	WARN_ON_ONCE(!gpu->revn);
 
 	return (gpu->revn < 300);
 }
 
-static inline bool adreno_is_a20x(struct adreno_gpu *gpu)
+static inline bool adreno_is_a20x(const struct adreno_gpu *gpu)
 {
 	WARN_ON_ONCE(!gpu->revn);
 
 	return (gpu->revn < 210);
 }
 
-static inline bool adreno_is_a225(struct adreno_gpu *gpu)
+static inline bool adreno_is_a225(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 225);
 }
 
-static inline bool adreno_is_a305(struct adreno_gpu *gpu)
+static inline bool adreno_is_a305(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 305);
 }
 
-static inline bool adreno_is_a306(struct adreno_gpu *gpu)
+static inline bool adreno_is_a306(const struct adreno_gpu *gpu)
 {
 	/* yes, 307, because a305c is 306 */
 	return adreno_is_revn(gpu, 307);
 }
 
-static inline bool adreno_is_a320(struct adreno_gpu *gpu)
+static inline bool adreno_is_a320(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 320);
 }
 
-static inline bool adreno_is_a330(struct adreno_gpu *gpu)
+static inline bool adreno_is_a330(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 330);
 }
 
-static inline bool adreno_is_a330v2(struct adreno_gpu *gpu)
+static inline bool adreno_is_a330v2(const struct adreno_gpu *gpu)
 {
 	return adreno_is_a330(gpu) && (gpu->rev.patchid > 0);
 }
 
-static inline int adreno_is_a405(struct adreno_gpu *gpu)
+static inline int adreno_is_a405(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 405);
 }
 
-static inline int adreno_is_a420(struct adreno_gpu *gpu)
+static inline int adreno_is_a420(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 420);
 }
 
-static inline int adreno_is_a430(struct adreno_gpu *gpu)
+static inline int adreno_is_a430(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 430);
 }
 
-static inline int adreno_is_a506(struct adreno_gpu *gpu)
+static inline int adreno_is_a506(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 506);
 }
 
-static inline int adreno_is_a508(struct adreno_gpu *gpu)
+static inline int adreno_is_a508(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 508);
 }
 
-static inline int adreno_is_a509(struct adreno_gpu *gpu)
+static inline int adreno_is_a509(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 509);
 }
 
-static inline int adreno_is_a510(struct adreno_gpu *gpu)
+static inline int adreno_is_a510(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 510);
 }
 
-static inline int adreno_is_a512(struct adreno_gpu *gpu)
+static inline int adreno_is_a512(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 512);
 }
 
-static inline int adreno_is_a530(struct adreno_gpu *gpu)
+static inline int adreno_is_a530(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 530);
 }
 
-static inline int adreno_is_a540(struct adreno_gpu *gpu)
+static inline int adreno_is_a540(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 540);
 }
 
-static inline int adreno_is_a618(struct adreno_gpu *gpu)
+static inline int adreno_is_a618(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 618);
 }
 
-static inline int adreno_is_a619(struct adreno_gpu *gpu)
+static inline int adreno_is_a619(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 619);
 }
 
-static inline int adreno_is_a630(struct adreno_gpu *gpu)
+static inline int adreno_is_a630(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 630);
 }
 
-static inline int adreno_is_a640_family(struct adreno_gpu *gpu)
+static inline int adreno_is_a640_family(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 640) ||
 		adreno_is_revn(gpu, 680);
 }
 
-static inline int adreno_is_a650(struct adreno_gpu *gpu)
+static inline int adreno_is_a650(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 650);
 }
 
-static inline int adreno_is_7c3(struct adreno_gpu *gpu)
+static inline int adreno_is_7c3(const struct adreno_gpu *gpu)
 {
 	/* The order of args is important here to handle ANY_ID correctly */
 	return adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), gpu->rev);
 }
 
-static inline int adreno_is_a660(struct adreno_gpu *gpu)
+static inline int adreno_is_a660(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 660);
 }
@@ -290,7 +290,7 @@ static inline int adreno_is_a690(struct adreno_gpu *gpu)
 };
 
 /* check for a615, a616, a618, a619 or any derivatives */
-static inline int adreno_is_a615_family(struct adreno_gpu *gpu)
+static inline int adreno_is_a615_family(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 615) ||
 		adreno_is_revn(gpu, 616) ||
@@ -298,13 +298,13 @@ static inline int adreno_is_a615_family(struct adreno_gpu *gpu)
 		adreno_is_revn(gpu, 619);
 }
 
-static inline int adreno_is_a660_family(struct adreno_gpu *gpu)
+static inline int adreno_is_a660_family(const struct adreno_gpu *gpu)
 {
 	return adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_7c3(gpu);
 }
 
 /* check for a650, a660, or any derivatives */
-static inline int adreno_is_a650_family(struct adreno_gpu *gpu)
+static inline int adreno_is_a650_family(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 650) ||
 		adreno_is_revn(gpu, 620) ||
-- 
GitLab


From b84c6b26a6509f7570d099c0a1dfcebf440f1b30 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:15 +0200
Subject: [PATCH 1648/1791] dt-bindings: display: mediatek: dpi: Add compatible
 for MediaTek MT6795

Add a compatible string for the MediaTek Helio X10 MT6795 SoC, using
the same parameters as MT8183.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Alexandre Mergnat <amergnat@baylibre.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-4-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../display/mediatek/mediatek,dpi.yaml        | 23 +++++++++++--------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml
index d976380801e3b..803c00f262069 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml
@@ -17,15 +17,20 @@ description: |
 
 properties:
   compatible:
-    enum:
-      - mediatek,mt2701-dpi
-      - mediatek,mt7623-dpi
-      - mediatek,mt8173-dpi
-      - mediatek,mt8183-dpi
-      - mediatek,mt8186-dpi
-      - mediatek,mt8188-dp-intf
-      - mediatek,mt8192-dpi
-      - mediatek,mt8195-dp-intf
+    oneOf:
+      - enum:
+          - mediatek,mt2701-dpi
+          - mediatek,mt7623-dpi
+          - mediatek,mt8173-dpi
+          - mediatek,mt8183-dpi
+          - mediatek,mt8186-dpi
+          - mediatek,mt8188-dp-intf
+          - mediatek,mt8192-dpi
+          - mediatek,mt8195-dp-intf
+      - items:
+          - enum:
+              - mediatek,mt6795-dpi
+          - const: mediatek,mt8183-dpi
 
   reg:
     maxItems: 1
-- 
GitLab


From d8720452fe0fee795517d69bc73820ec6c1c4727 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:16 +0200
Subject: [PATCH 1649/1791] dt-bindings: display: mediatek: aal: Add compatible
 for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795: similarly to
MT8173, this SoC has the gamma LUT registers in DISP_AAL.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: Alexandre Mergnat <amergnat@baylibre.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-5-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,aal.yaml       | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml
index 92741486c24dd..7fd42c8fdc328 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml
@@ -27,6 +27,7 @@ properties:
       - items:
           - enum:
               - mediatek,mt2712-disp-aal
+              - mediatek,mt6795-disp-aal
           - const: mediatek,mt8173-disp-aal
       - items:
           - enum:
-- 
GitLab


From 28c143af8bc954b3167534b0475cc25d1b3eea34 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:17 +0200
Subject: [PATCH 1650/1791] dt-bindings: display: mediatek: dsi: Add compatible
 for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795, using the same
DSI block as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-6-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../display/mediatek/mediatek,dsi.yaml        | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml
index 4707b60238b07..12441b937684d 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml
@@ -22,13 +22,18 @@ allOf:
 
 properties:
   compatible:
-    enum:
-      - mediatek,mt2701-dsi
-      - mediatek,mt7623-dsi
-      - mediatek,mt8167-dsi
-      - mediatek,mt8173-dsi
-      - mediatek,mt8183-dsi
-      - mediatek,mt8186-dsi
+    oneOf:
+      - enum:
+          - mediatek,mt2701-dsi
+          - mediatek,mt7623-dsi
+          - mediatek,mt8167-dsi
+          - mediatek,mt8173-dsi
+          - mediatek,mt8183-dsi
+          - mediatek,mt8186-dsi
+      - items:
+          - enum:
+              - mediatek,mt6795-dsi
+          - const: mediatek,mt8173-dsi
 
   reg:
     maxItems: 1
-- 
GitLab


From 35b7a18c59a146635ecf368a9c10e5074a099a43 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:18 +0200
Subject: [PATCH 1651/1791] dt-bindings: display: mediatek: ovl: Add compatible
 for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's OVL block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-7-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,ovl.yaml    | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml
index 92e320d54ba2a..3e1069b00b56e 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml
@@ -31,6 +31,10 @@ properties:
               - mediatek,mt7623-disp-ovl
               - mediatek,mt2712-disp-ovl
           - const: mediatek,mt2701-disp-ovl
+      - items:
+          - enum:
+              - mediatek,mt6795-disp-ovl
+          - const: mediatek,mt8173-disp-ovl
       - items:
           - enum:
               - mediatek,mt8188-disp-ovl
-- 
GitLab


From 48ed9e648722cb2e23936df5f1b6eaf080dc9046 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:19 +0200
Subject: [PATCH 1652/1791] dt-bindings: display: mediatek: rdma: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's RDMA block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-8-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,rdma.yaml   | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml
index 42059efad45de..39dbb5c8bcf80 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml
@@ -37,6 +37,10 @@ properties:
               - mediatek,mt7623-disp-rdma
               - mediatek,mt2712-disp-rdma
           - const: mediatek,mt2701-disp-rdma
+      - items:
+          - enum:
+              - mediatek,mt6795-disp-rdma
+          - const: mediatek,mt8173-disp-rdma
       - items:
           - enum:
               - mediatek,mt8186-disp-rdma
-- 
GitLab


From 8f2a42046cb60a937481d1d3979bedb86166c6b4 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:20 +0200
Subject: [PATCH 1653/1791] dt-bindings: display: mediatek: wdma: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's WDMA block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-9-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,wdma.yaml    | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml
index 991183165d29f..a3a2b71a45230 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml
@@ -23,6 +23,9 @@ properties:
     oneOf:
       - enum:
           - mediatek,mt8173-disp-wdma
+      - items:
+          - const: mediatek,mt6795-disp-wdma
+          - const: mediatek,mt8173-disp-wdma
 
   reg:
     maxItems: 1
-- 
GitLab


From 791a9974fcaac826c4ff716a5b663f176d315ff3 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:21 +0200
Subject: [PATCH 1654/1791] dt-bindings: display: mediatek: color: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's COLOR block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-10-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,color.yaml     | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml
index d0ea77fc4b069..f21e440920436 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml
@@ -33,6 +33,7 @@ properties:
           - const: mediatek,mt2701-disp-color
       - items:
           - enum:
+              - mediatek,mt6795-disp-color
               - mediatek,mt8183-disp-color
               - mediatek,mt8186-disp-color
               - mediatek,mt8188-disp-color
-- 
GitLab


From 84e05dc1f94d53ecbafb3176b1ce8f2988a1262f Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:22 +0200
Subject: [PATCH 1655/1791] dt-bindings: display: mediatek: gamma: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's GAMMA block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-11-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,gamma.yaml  | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml
index 6c2be9d6840b1..c6641acd75d6a 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml
@@ -24,6 +24,10 @@ properties:
       - enum:
           - mediatek,mt8173-disp-gamma
           - mediatek,mt8183-disp-gamma
+      - items:
+          - enum:
+              - mediatek,mt6795-disp-gamma
+          - const: mediatek,mt8173-disp-gamma
       - items:
           - enum:
               - mediatek,mt8186-disp-gamma
-- 
GitLab


From 400ab909b7cd41a60838f0f27503d7cfc9cc7030 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:23 +0200
Subject: [PATCH 1656/1791] dt-bindings: display: mediatek: merge: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's MERGE block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-12-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,merge.yaml   | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml
index 2f8e2f4dc3b82..eead5cb8636ed 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml
@@ -24,6 +24,9 @@ properties:
       - enum:
           - mediatek,mt8173-disp-merge
           - mediatek,mt8195-disp-merge
+      - items:
+          - const: mediatek,mt6795-disp-merge
+          - const: mediatek,mt8173-disp-merge
 
   reg:
     maxItems: 1
-- 
GitLab


From 5dfd9bb376d9754c52ee6a3ef61c63869b5a36b3 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:24 +0200
Subject: [PATCH 1657/1791] dt-bindings: display: mediatek: split: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's SPLIT block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-13-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,split.yaml   | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml
index 21a4e96ecd936..a8a5c9608598f 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml
@@ -23,6 +23,9 @@ properties:
     oneOf:
       - enum:
           - mediatek,mt8173-disp-split
+      - items:
+          - const: mediatek,mt6795-disp-split
+          - const: mediatek,mt8173-disp-split
 
   reg:
     maxItems: 1
-- 
GitLab


From c769eccefbb3c6c516e38704bdca386531640170 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:25 +0200
Subject: [PATCH 1658/1791] dt-bindings: display: mediatek: ufoe: Add
 compatible for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's UFOE block: this
is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-14-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,ufoe.yaml    | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml
index 62fad23a26f5a..39e3e2d4a0db8 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml
@@ -24,6 +24,9 @@ properties:
     oneOf:
       - enum:
           - mediatek,mt8173-disp-ufoe
+      - items:
+          - const: mediatek,mt6795-disp-ufoe
+          - const: mediatek,mt8173-disp-ufoe
 
   reg:
     maxItems: 1
-- 
GitLab


From 4080a0e7b36b7bb103dc739779a338d412c78943 Mon Sep 17 00:00:00 2001
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Date: Wed, 12 Apr 2023 13:27:26 +0200
Subject: [PATCH 1659/1791] dt-bindings: display: mediatek: od: Add compatible
 for MediaTek MT6795

Add a compatible string for MediaTek Helio X10 MT6795's OverDrive (OD)
block: this is the same as MT8173.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20230412112739.160376-15-angelogioacchino.delregno@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
---
 .../devicetree/bindings/display/mediatek/mediatek,od.yaml      | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml
index 29f9fa8f82195..831c653caffd7 100644
--- a/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml
+++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml
@@ -24,6 +24,9 @@ properties:
       - enum:
           - mediatek,mt2712-disp-od
           - mediatek,mt8173-disp-od
+      - items:
+          - const: mediatek,mt6795-disp-od
+          - const: mediatek,mt8173-disp-od
 
   reg:
     maxItems: 1
-- 
GitLab


From cf683e8870bd4be0fd6b98639286700a35088660 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Wed, 14 Jun 2023 15:12:12 +0200
Subject: [PATCH 1660/1791] fbdev: Use /* */ comment in initializer macro
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use /* */ in initializer macro to avoid out-commenting the comma
at the end of the line.

Reported-by: Christian König <ckoenig.leichtzumerken@gmail.com>
Closes: https://lore.kernel.org/dri-devel/20230530150253.22758-1-tzimmermann@suse.de/T/#m356cda2679c17d7a01f30ce2b5282cd9046ea6d4
Fixes: f1061fa641b8 ("fbdev: Add initializer macros for struct fb_ops")
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-fbdev@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230614131253.10208-1-tzimmermann@suse.de
---
 include/linux/fb.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/fb.h b/include/linux/fb.h
index ce6823e157e6b..ce7d588edc3e6 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -552,7 +552,7 @@ extern ssize_t fb_io_write(struct fb_info *info, const char __user *buf,
 	.fb_imageblit	= cfb_imageblit
 
 #define __FB_DEFAULT_IO_OPS_MMAP \
-	.fb_mmap	= NULL // default implementation
+	.fb_mmap	= NULL /* default implementation */
 
 #define FB_DEFAULT_IO_OPS \
 	__FB_DEFAULT_IO_OPS_RDWR, \
@@ -585,7 +585,7 @@ extern ssize_t fb_sys_write(struct fb_info *info, const char __user *buf,
 	.fb_imageblit	= sys_imageblit
 
 #define __FB_DEFAULT_SYS_OPS_MMAP \
-	.fb_mmap	= NULL // default implementation
+	.fb_mmap	= NULL /* default implementation */
 
 #define FB_DEFAULT_SYS_OPS \
 	__FB_DEFAULT_SYS_OPS_RDWR, \
-- 
GitLab


From b263325b4f341a0e85c1d5dc4451df2c5e4349fa Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 12 Jun 2023 21:25:27 +0300
Subject: [PATCH 1661/1791] drm/msm/adreno: make adreno_is_a690()'s argument
 const
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Change adreno_is_a690() prototype to accept the const struct adreno_gpu
pointer instead of a non-const one. This fixes the following warning:

In file included from drivers/gpu/drm/msm/msm_drv.c:33:
drivers/gpu/drm/msm/adreno/adreno_gpu.h: In function ‘adreno_is_a660_family’:
drivers/gpu/drm/msm/adreno/adreno_gpu.h:303:54: warning: passing argument 1 of ‘adreno_is_a690’ discards ‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers]
  303 |         return adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_7c3(gpu);

Fixes: 1b90e8f8879c ("drm/msm/adreno: change adreno_is_* functions to accept const argument")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/542138/
Link: https://lore.kernel.org/r/20230612182527.3345786-1-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/adreno/adreno_gpu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 1283e5fe22d24..9a7626c7ac4d4 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -284,7 +284,7 @@ static inline int adreno_is_a660(const struct adreno_gpu *gpu)
 	return adreno_is_revn(gpu, 660);
 }
 
-static inline int adreno_is_a690(struct adreno_gpu *gpu)
+static inline int adreno_is_a690(const struct adreno_gpu *gpu)
 {
 	return gpu->revn == 690;
 };
-- 
GitLab


From 3bcfc7b90465efd337d39b91b43972162f0d1908 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 12 Jun 2023 21:25:33 +0300
Subject: [PATCH 1662/1791] drm/msm/dpu: do not enable color-management if
 DSPPs are not available

We can not support color management without DSPP blocks being provided
in the HW catalog. Do not enable color management for CRTCs if num_dspps
is 0.

Fixes: 4259ff7ae509 ("drm/msm/dpu: add support for pcc color block in dpu driver")
Reported-by: Yongqin Liu <yongqin.liu@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Sumit Semwal <sumit.semwal@linaro.org>
Tested-by: Yongqin Liu <yongqin.liu@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542141/
Link: https://lore.kernel.org/r/20230612182534.3345805-1-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index 6e684a7b49a17..1edf2b6b0a262 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -1463,6 +1463,8 @@ static const struct drm_crtc_helper_funcs dpu_crtc_helper_funcs = {
 struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane,
 				struct drm_plane *cursor)
 {
+	struct msm_drm_private *priv = dev->dev_private;
+	struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms);
 	struct drm_crtc *crtc = NULL;
 	struct dpu_crtc *dpu_crtc = NULL;
 	int i, ret;
@@ -1494,7 +1496,8 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane,
 
 	drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs);
 
-	drm_crtc_enable_color_mgmt(crtc, 0, true, 0);
+	if (dpu_kms->catalog->dspp_count)
+		drm_crtc_enable_color_mgmt(crtc, 0, true, 0);
 
 	/* save user friendly CRTC name for later */
 	snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id);
-- 
GitLab


From c72375172194a371935c78b0edaae75cd0c408ea Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 12 Jun 2023 21:25:34 +0300
Subject: [PATCH 1663/1791] drm/msm/dpu/catalog: define DSPP blocks found on
 sdm845

Add definitions of DSPP blocks present on the sdm845 platform. This
should enable color-management on sdm845-bassed devices.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Sumit Semwal <sumit.semwal@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542142/
Link: https://lore.kernel.org/r/20230612182534.3345805-2-dmitry.baryshkov@linaro.org
---
 .../msm/disp/dpu1/catalog/dpu_4_0_sdm845.h    | 21 +++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
index 36ea1af108949..b6098141bb9bb 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h
@@ -96,19 +96,30 @@ static const struct dpu_sspp_cfg sdm845_sspp[] = {
 
 static const struct dpu_lm_cfg sdm845_lm[] = {
 	LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_0, LM_1, 0),
+		&sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0),
 	LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_1, LM_0, 0),
+		&sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1),
 	LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_2, LM_5, 0),
+		&sdm845_lm_sblk, PINGPONG_2, LM_5, DSPP_2),
 	LM_BLK("lm_3", LM_3, 0x0, MIXER_SDM845_MASK,
-		&sdm845_lm_sblk, PINGPONG_NONE, 0, 0),
+		&sdm845_lm_sblk, PINGPONG_NONE, 0, DSPP_3),
 	LM_BLK("lm_4", LM_4, 0x0, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_NONE, 0, 0),
 	LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK,
 		&sdm845_lm_sblk, PINGPONG_3, LM_2, 0),
 };
 
+static const struct dpu_dspp_cfg sdm845_dspp[] = {
+	DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+	DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK,
+		 &sm8150_dspp_sblk),
+};
+
 static const struct dpu_pingpong_cfg sdm845_pp[] = {
 	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
@@ -193,6 +204,8 @@ const struct dpu_mdss_cfg dpu_sdm845_cfg = {
 	.sspp = sdm845_sspp,
 	.mixer_count = ARRAY_SIZE(sdm845_lm),
 	.mixer = sdm845_lm,
+	.dspp_count = ARRAY_SIZE(sdm845_dspp),
+	.dspp = sdm845_dspp,
 	.pingpong_count = ARRAY_SIZE(sdm845_pp),
 	.pingpong = sdm845_pp,
 	.dsc_count = ARRAY_SIZE(sdm845_dsc),
-- 
GitLab


From 7df1ed6ddf3da52b020ef3c3f5597bc628c3e58e Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:14 -0700
Subject: [PATCH 1664/1791] drm/display/dsc: Add flatness and initial scale
 value calculations

Add helpers to calculate det_thresh_flatness and initial_scale_value as
these calculations are defined within the DSC spec.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539282/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-1-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 24 ++++++++++++++++++++++++
 include/drm/display/drm_dsc_helper.h     |  2 ++
 2 files changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index fc187a8d8873d..4efb6236d22c9 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -1413,3 +1413,27 @@ int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg)
 	return 0;
 }
 EXPORT_SYMBOL(drm_dsc_compute_rc_parameters);
+
+/**
+ * drm_dsc_initial_scale_value() - Calculate the initial scale value for the given DSC config
+ * @dsc: Pointer to DRM DSC config struct
+ *
+ * Return: Calculated initial scale value
+ */
+u8 drm_dsc_initial_scale_value(const struct drm_dsc_config *dsc)
+{
+	return 8 * dsc->rc_model_size / (dsc->rc_model_size - dsc->initial_offset);
+}
+EXPORT_SYMBOL(drm_dsc_initial_scale_value);
+
+/**
+ * drm_dsc_flatness_det_thresh() - Calculate the flatness_det_thresh for the given DSC config
+ * @dsc: Pointer to DRM DSC config struct
+ *
+ * Return: Calculated flatness det thresh value
+ */
+u32 drm_dsc_flatness_det_thresh(const struct drm_dsc_config *dsc)
+{
+	return 2 << (dsc->bits_per_component - 8);
+}
+EXPORT_SYMBOL(drm_dsc_flatness_det_thresh);
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index fc2104415dcbd..71789fb34e172 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -24,6 +24,8 @@ void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_sdp,
 void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg);
 int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params_type type);
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
+u8 drm_dsc_initial_scale_value(const struct drm_dsc_config *dsc);
+u32 drm_dsc_flatness_det_thresh(const struct drm_dsc_config *dsc);
 
 #endif /* _DRM_DSC_HELPER_H_ */
 
-- 
GitLab


From e871a70d8ccd6dbcb30f081f5d3d8854380422fe Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 24 May 2023 10:45:15 -0700
Subject: [PATCH 1665/1791] drm/display/dsc: add helper to set semi-const
 parameters

Add a helper setting config values which are typically constant across
operating modes (table E-4 of the standard) and mux_word_size (which is
a const according to 3.5.2).

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539280/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-2-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 22 ++++++++++++++++++++++
 include/drm/display/drm_dsc_helper.h     |  1 +
 2 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index 4efb6236d22c9..b31fe98497848 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -270,6 +270,28 @@ void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload,
 }
 EXPORT_SYMBOL(drm_dsc_pps_payload_pack);
 
+/**
+ * drm_dsc_set_const_params() - Set DSC parameters considered typically
+ * constant across operation modes
+ *
+ * @vdsc_cfg:
+ * DSC Configuration data partially filled by driver
+ */
+void drm_dsc_set_const_params(struct drm_dsc_config *vdsc_cfg)
+{
+	if (!vdsc_cfg->rc_model_size)
+		vdsc_cfg->rc_model_size = DSC_RC_MODEL_SIZE_CONST;
+	vdsc_cfg->rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST;
+	vdsc_cfg->rc_tgt_offset_high = DSC_RC_TGT_OFFSET_HI_CONST;
+	vdsc_cfg->rc_tgt_offset_low = DSC_RC_TGT_OFFSET_LO_CONST;
+
+	if (vdsc_cfg->bits_per_component <= 10)
+		vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_8_10_BPC;
+	else
+		vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_12_BPC;
+}
+EXPORT_SYMBOL(drm_dsc_set_const_params);
+
 /* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */
 static const u16 drm_dsc_rc_buf_thresh[] = {
 	896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616,
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index 71789fb34e172..f4e18e5d077a8 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -21,6 +21,7 @@ void drm_dsc_dp_pps_header_init(struct dp_sdp_header *pps_header);
 int drm_dsc_dp_rc_buffer_size(u8 rc_buffer_block_size, u8 rc_buffer_size);
 void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_sdp,
 			      const struct drm_dsc_config *dsc_cfg);
+void drm_dsc_set_const_params(struct drm_dsc_config *vdsc_cfg);
 void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg);
 int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params_type type);
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
-- 
GitLab


From 68858328124162b9b42bc7b8232eee1915cc1d8f Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:16 -0700
Subject: [PATCH 1666/1791] drm/display/dsc: Add drm_dsc_get_bpp_int helper

Add helper to get the integer value of drm_dsc_config.bits_per_pixel

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539268/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-3-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/display/drm_dsc_helper.c | 13 +++++++++++++
 include/drm/display/drm_dsc_helper.h     |  1 +
 2 files changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c
index b31fe98497848..4424380c6cb63 100644
--- a/drivers/gpu/drm/display/drm_dsc_helper.c
+++ b/drivers/gpu/drm/display/drm_dsc_helper.c
@@ -1436,6 +1436,19 @@ int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg)
 }
 EXPORT_SYMBOL(drm_dsc_compute_rc_parameters);
 
+/**
+ * drm_dsc_get_bpp_int() - Get integer bits per pixel value for the given DRM DSC config
+ * @vdsc_cfg: Pointer to DRM DSC config struct
+ *
+ * Return: Integer BPP value
+ */
+u32 drm_dsc_get_bpp_int(const struct drm_dsc_config *vdsc_cfg)
+{
+	WARN_ON_ONCE(vdsc_cfg->bits_per_pixel & 0xf);
+	return vdsc_cfg->bits_per_pixel >> 4;
+}
+EXPORT_SYMBOL(drm_dsc_get_bpp_int);
+
 /**
  * drm_dsc_initial_scale_value() - Calculate the initial scale value for the given DSC config
  * @dsc: Pointer to DRM DSC config struct
diff --git a/include/drm/display/drm_dsc_helper.h b/include/drm/display/drm_dsc_helper.h
index f4e18e5d077a8..913aa20712322 100644
--- a/include/drm/display/drm_dsc_helper.h
+++ b/include/drm/display/drm_dsc_helper.h
@@ -27,6 +27,7 @@ int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params
 int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg);
 u8 drm_dsc_initial_scale_value(const struct drm_dsc_config *dsc);
 u32 drm_dsc_flatness_det_thresh(const struct drm_dsc_config *dsc);
+u32 drm_dsc_get_bpp_int(const struct drm_dsc_config *vdsc_cfg);
 
 #endif /* _DRM_DSC_HELPER_H_ */
 
-- 
GitLab


From 49fd30a7153be027b7ae9895e9061dec15e836f7 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Wed, 24 May 2023 10:45:17 -0700
Subject: [PATCH 1667/1791] drm/msm/dsi: use DRM DSC helpers for DSC setup

Use new DRM DSC helpers to setup DSI DSC configuration. The
initial_scale_value needs to be adjusted according to the standard, but
this is a separate change.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539276/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-4-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 61 ++++--------------------------
 1 file changed, 8 insertions(+), 53 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 744f2398a6d6e..7892b9de9736f 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -1728,28 +1728,9 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host,
 	return -EINVAL;
 }
 
-static u32 dsi_dsc_rc_buf_thresh[DSC_NUM_BUF_RANGES - 1] = {
-	0x0e, 0x1c, 0x2a, 0x38, 0x46, 0x54, 0x62,
-	0x69, 0x70, 0x77, 0x79, 0x7b, 0x7d, 0x7e
-};
-
-/* only 8bpc, 8bpp added */
-static char min_qp[DSC_NUM_BUF_RANGES] = {
-	0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 13
-};
-
-static char max_qp[DSC_NUM_BUF_RANGES] = {
-	4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 11, 12, 13, 13, 15
-};
-
-static char bpg_offset[DSC_NUM_BUF_RANGES] = {
-	2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -12, -12, -12, -12
-};
-
 static int dsi_populate_dsc_params(struct msm_dsi_host *msm_host, struct drm_dsc_config *dsc)
 {
-	int i;
-	u16 bpp = dsc->bits_per_pixel >> 4;
+	int ret;
 
 	if (dsc->bits_per_pixel & 0xf) {
 		DRM_DEV_ERROR(&msm_host->pdev->dev, "DSI does not support fractional bits_per_pixel\n");
@@ -1761,49 +1742,23 @@ static int dsi_populate_dsc_params(struct msm_dsi_host *msm_host, struct drm_dsc
 		return -EOPNOTSUPP;
 	}
 
-	dsc->rc_model_size = 8192;
-	dsc->first_line_bpg_offset = 12;
-	dsc->rc_edge_factor = 6;
-	dsc->rc_tgt_offset_high = 3;
-	dsc->rc_tgt_offset_low = 3;
 	dsc->simple_422 = 0;
 	dsc->convert_rgb = 1;
 	dsc->vbr_enable = 0;
 
-	/* handle only bpp = bpc = 8 */
-	for (i = 0; i < DSC_NUM_BUF_RANGES - 1 ; i++)
-		dsc->rc_buf_thresh[i] = dsi_dsc_rc_buf_thresh[i];
+	drm_dsc_set_const_params(dsc);
+	drm_dsc_set_rc_buf_thresh(dsc);
 
-	for (i = 0; i < DSC_NUM_BUF_RANGES; i++) {
-		dsc->rc_range_params[i].range_min_qp = min_qp[i];
-		dsc->rc_range_params[i].range_max_qp = max_qp[i];
-		/*
-		 * Range BPG Offset contains two's-complement signed values that fill
-		 * 8 bits, yet the registers and DCS PPS field are only 6 bits wide.
-		 */
-		dsc->rc_range_params[i].range_bpg_offset = bpg_offset[i] & DSC_RANGE_BPG_OFFSET_MASK;
+	/* handle only bpp = bpc = 8, pre-SCR panels */
+	ret = drm_dsc_setup_rc_params(dsc, DRM_DSC_1_1_PRE_SCR);
+	if (ret) {
+		DRM_DEV_ERROR(&msm_host->pdev->dev, "could not find DSC RC parameters\n");
+		return ret;
 	}
 
-	dsc->initial_offset = 6144;		/* Not bpp 12 */
-	if (bpp != 8)
-		dsc->initial_offset = 2048;	/* bpp = 12 */
-
-	if (dsc->bits_per_component <= 10)
-		dsc->mux_word_size = DSC_MUX_WORD_SIZE_8_10_BPC;
-	else
-		dsc->mux_word_size = DSC_MUX_WORD_SIZE_12_BPC;
-
-	dsc->initial_xmit_delay = 512;
 	dsc->initial_scale_value = 32;
-	dsc->first_line_bpg_offset = 12;
 	dsc->line_buf_depth = dsc->bits_per_component + 1;
 
-	/* bpc 8 */
-	dsc->flatness_min_qp = 3;
-	dsc->flatness_max_qp = 12;
-	dsc->rc_quant_incr_limit0 = 11;
-	dsc->rc_quant_incr_limit1 = 11;
-
 	return drm_dsc_compute_rc_parameters(dsc);
 }
 
-- 
GitLab


From b50f06f83e0e859d93bd7bb80a2d1911f2dc176b Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:18 -0700
Subject: [PATCH 1668/1791] drm/msm: Add MSM-specific DSC helper methods

Introduce MSM-specific DSC helper methods, as some calculations are
common between DP and DSC.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539272/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-5-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/msm_dsc_helper.h | 38 ++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
 create mode 100644 drivers/gpu/drm/msm/msm_dsc_helper.h

diff --git a/drivers/gpu/drm/msm/msm_dsc_helper.h b/drivers/gpu/drm/msm/msm_dsc_helper.h
new file mode 100644
index 0000000000000..b9049fe1e2790
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_dsc_helper.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved
+ *
+ * Helper methods for MSM-specific DSC calculations that are common between timing engine,
+ * DSI, and DP.
+ */
+
+#ifndef MSM_DSC_HELPER_H_
+#define MSM_DSC_HELPER_H_
+
+#include <linux/math.h>
+#include <drm/display/drm_dsc_helper.h>
+
+/**
+ * msm_dsc_get_slices_per_intf() - calculate number of slices per interface
+ * @dsc: Pointer to drm dsc config struct
+ * @intf_width: interface width in pixels
+ * Returns: Integer representing the number of slices for the given interface
+ */
+static inline u32 msm_dsc_get_slices_per_intf(const struct drm_dsc_config *dsc, u32 intf_width)
+{
+	return DIV_ROUND_UP(intf_width, dsc->slice_width);
+}
+
+/**
+ * msm_dsc_get_bytes_per_line() - calculate bytes per line
+ * @dsc: Pointer to drm dsc config struct
+ * Returns: Integer value representing bytes per line. DSI and DP need
+ *          to perform further calculations to turn this into pclk_per_intf,
+ *          such as dividing by different values depending on if widebus is enabled.
+ */
+static inline u32 msm_dsc_get_bytes_per_line(const struct drm_dsc_config *dsc)
+{
+	return dsc->slice_count * dsc->slice_chunk_size;
+}
+
+#endif /* MSM_DSC_HELPER_H_ */
-- 
GitLab


From 44346191a210f04c52f8389b3bd989d4a2a3e1ba Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:19 -0700
Subject: [PATCH 1669/1791] drm/msm/dpu: Use fixed DRM DSC helper for
 det_thresh_flatness

The current dpu_hw_dsc calculation for det_thresh_flatness does not
match the downstream calculation or the DSC spec.

Use the DRM DSC helper for det_thresh_flatness to match downstream
implementation and the DSC spec.

Fixes: c110cfd1753e ("drm/msm/disp/dpu1: Add support for DSC")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539275/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-6-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
index 64847912a20ff..99ba7eebf5a4f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
@@ -3,6 +3,8 @@
  * Copyright (c) 2020-2022, Linaro Limited
  */
 
+#include <drm/display/drm_dsc_helper.h>
+
 #include "dpu_kms.h"
 #include "dpu_hw_catalog.h"
 #include "dpu_hwio.h"
@@ -102,7 +104,7 @@ static void dpu_hw_dsc_config(struct dpu_hw_dsc *hw_dsc,
 	data |= dsc->final_offset;
 	DPU_REG_WRITE(c, DSC_DSC_OFFSET, data);
 
-	det_thresh_flatness = 7 + 2 * (dsc->bits_per_component - 8);
+	det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc);
 	data = det_thresh_flatness << 10;
 	data |= dsc->flatness_max_qp << 5;
 	data |= dsc->flatness_min_qp;
-- 
GitLab


From c223059e6f8340f7eac2319470984cbfc39c433b Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:20 -0700
Subject: [PATCH 1670/1791] drm/msm/dpu: Fix slice_last_group_size calculation

Correct the math for slice_last_group_size so that it matches the
calculations downstream.

Fixes: c110cfd1753e ("drm/msm/disp/dpu1: Add support for DSC")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539269/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-7-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
index 99ba7eebf5a4f..509dbaa51d878 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c
@@ -56,9 +56,10 @@ static void dpu_hw_dsc_config(struct dpu_hw_dsc *hw_dsc,
 	if (is_cmd_mode)
 		initial_lines += 1;
 
-	slice_last_group_size = 3 - (dsc->slice_width % 3);
+	slice_last_group_size = (dsc->slice_width + 2) % 3;
+
 	data = (initial_lines << 20);
-	data |= ((slice_last_group_size - 1) << 18);
+	data |= (slice_last_group_size << 18);
 	/* bpp is 6.4 format, 4 LSBs bits are for fractional part */
 	data |= (dsc->bits_per_pixel << 8);
 	data |= (dsc->block_pred_enable << 7);
-- 
GitLab


From ed1498f77419f2a636a07825492230c5562cc333 Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:21 -0700
Subject: [PATCH 1671/1791] drm/msm/dsi: Use MSM and DRM DSC helper methods

Use MSM and DRM DSC helper methods to configure DSC for DSI.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539274/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-8-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 7892b9de9736f..e693022ef79fb 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -28,6 +28,7 @@
 #include "dsi.xml.h"
 #include "sfpb.xml.h"
 #include "dsi_cfg.h"
+#include "msm_dsc_helper.h"
 #include "msm_kms.h"
 #include "msm_gem.h"
 #include "phy/dsi_phy.h"
@@ -845,7 +846,7 @@ static void dsi_update_dsc_timing(struct msm_dsi_host *msm_host, bool is_cmd_mod
 	/* first calculate dsc parameters and then program
 	 * compress mode registers
 	 */
-	slice_per_intf = DIV_ROUND_UP(hdisplay, dsc->slice_width);
+	slice_per_intf = msm_dsc_get_slices_per_intf(dsc, hdisplay);
 
 	/*
 	 * If slice_count is greater than slice_per_intf
@@ -1756,7 +1757,7 @@ static int dsi_populate_dsc_params(struct msm_dsi_host *msm_host, struct drm_dsc
 		return ret;
 	}
 
-	dsc->initial_scale_value = 32;
+	dsc->initial_scale_value = drm_dsc_initial_scale_value(dsc);
 	dsc->line_buf_depth = dsc->bits_per_component + 1;
 
 	return drm_dsc_compute_rc_parameters(dsc);
-- 
GitLab


From 149419396a9200075de8d75becd740ae48721524 Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Wed, 24 May 2023 10:45:22 -0700
Subject: [PATCH 1672/1791] drm/msm/dsi: update hdisplay calculation for
 dsi_timing_setup

Currently, hdisplay is being divided by 3 for DSC. However, this
calculation only works for cases where BPP = 8.

Update hdisplay calculation to be bytes_per_line / 3, so that it
accounts for cases where BPP != 8.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/539271/
Link: https://lore.kernel.org/r/20230329-rfc-msm-dsc-helper-v14-9-bafc7be95691@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index e693022ef79fb..1a99d75025dc4 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -949,7 +949,7 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 		 * pulse width same
 		 */
 		h_total -= hdisplay;
-		hdisplay /= 3;
+		hdisplay = msm_dsc_get_bytes_per_line(msm_host->dsc) / 3;
 		h_total += hdisplay;
 		ha_end = ha_start + hdisplay;
 	}
-- 
GitLab


From 8c4094b275f66d725a785d2314c27e00db2141dd Mon Sep 17 00:00:00 2001
From: Kuogee Hsieh <quic_khsieh@quicinc.com>
Date: Thu, 25 May 2023 10:40:54 -0700
Subject: [PATCH 1673/1791] drm/msm/dpu: add support for DSC encoder v1.2
 engine

Add support for DSC 1.2 by providing the necessary hooks to program
the DPU DSC 1.2 encoder.

Changes in v3:
-- fixed kernel test rebot report that "__iomem *off" is declared but not
   used at dpu_hw_dsc_config_1_2()
-- unrolling thresh loops

Changes in v4:
-- delete DPU_DSC_HW_REV_1_1
-- delete off and used real register name directly

Changes in v7:
-- replace offset with sblk->enc.base
-- replace ss with slice

Changes in v8:
-- fixed checkpatch warning

Changes in v9:
-- replaced __dsc_calc_ob_max_addr() with __dsc_calc_output_buf_max_addr()
-- replaced variable num_ss with num_softslice
-- remove inline from function  declaration

changes in v10:
-- rewording text of changes in v9
-- replace DPU_DSC_NATIVE_422_EN with DPU_DSC_NATIVE_42x_EN
-- replace drm_dsc_calculate_flatness_det_thresh() with drm_dsc_flatness_det_thresh()

Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539500/
Link: https://lore.kernel.org/r/1685036458-22683-7-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/Makefile                  |   1 +
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h    |  31 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h    |  14 +-
 .../gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c    | 387 ++++++++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c        |   7 +-
 5 files changed, 436 insertions(+), 4 deletions(-)
 create mode 100644 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c

diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 39d9ac05c4b6c..8d02d8c330695 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -65,6 +65,7 @@ msm-$(CONFIG_DRM_MSM_DPU) += \
 	disp/dpu1/dpu_hw_catalog.o \
 	disp/dpu1/dpu_hw_ctl.o \
 	disp/dpu1/dpu_hw_dsc.o \
+	disp/dpu1/dpu_hw_dsc_1_2.o \
 	disp/dpu1/dpu_hw_interrupts.o \
 	disp/dpu1/dpu_hw_intf.o \
 	disp/dpu1/dpu_hw_lm.o \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index ac4a9e73705cd..d3598dd9d448b 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved.
  * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved.
  */
 
@@ -240,12 +240,18 @@ enum {
 };
 
 /**
- * DSC features
+ * DSC sub-blocks/features
  * @DPU_DSC_OUTPUT_CTRL       Configure which PINGPONG block gets
  *                            the pixel output from this DSC.
+ * @DPU_DSC_HW_REV_1_2        DSC block supports DSC 1.1 and 1.2
+ * @DPU_DSC_NATIVE_42x_EN     Supports NATIVE_422_EN and NATIVE_420_EN encoding
+ * @DPU_DSC_MAX
  */
 enum {
 	DPU_DSC_OUTPUT_CTRL = 0x1,
+	DPU_DSC_HW_REV_1_2,
+	DPU_DSC_NATIVE_42x_EN,
+	DPU_DSC_MAX
 };
 
 /**
@@ -301,6 +307,14 @@ struct dpu_pp_blk {
 	u32 version;
 };
 
+/**
+ * struct dpu_dsc_blk - DSC Encoder sub-blk information
+ * @info:   HW register and features supported by this sub-blk
+ */
+struct dpu_dsc_blk {
+	DPU_HW_SUBBLK_INFO;
+};
+
 /**
  * enum dpu_qos_lut_usage - define QoS LUT use cases
  */
@@ -431,6 +445,16 @@ struct dpu_pingpong_sub_blks {
 	struct dpu_pp_blk dither;
 };
 
+/**
+ * struct dpu_dsc_sub_blks - DSC sub-blks
+ * @enc: DSC encoder sub-block
+ * @ctl: DSC controller sub-block
+ */
+struct dpu_dsc_sub_blks {
+	struct dpu_dsc_blk enc;
+	struct dpu_dsc_blk ctl;
+};
+
 /**
  * dpu_clk_ctrl_type - Defines top level clock control signals
  */
@@ -585,10 +609,13 @@ struct dpu_merge_3d_cfg  {
  * struct dpu_dsc_cfg - information of DSC blocks
  * @id                 enum identifying this block
  * @base               register offset of this block
+ * @len:               length of hardware block
  * @features           bit mask identifying sub-blocks/features
+ * @sblk:              sub-blocks information
  */
 struct dpu_dsc_cfg {
 	DPU_HW_BLK_INFO;
+	const struct dpu_dsc_sub_blks *sblk;
 };
 
 /**
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
index 138080aed9dd7..d5b597ab8c5ca 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h
@@ -1,5 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
-/* Copyright (c) 2020-2022, Linaro Limited */
+/*
+ * Copyright (c) 2020-2022, Linaro Limited
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved
+ */
 
 #ifndef _DPU_HW_DSC_H
 #define _DPU_HW_DSC_H
@@ -68,6 +71,15 @@ struct dpu_hw_dsc {
 struct dpu_hw_dsc *dpu_hw_dsc_init(const struct dpu_dsc_cfg *cfg,
 		void __iomem *addr);
 
+/**
+ * dpu_hw_dsc_init_1_2() - initializes the v1.2 DSC hw driver object
+ * @cfg:  DSC catalog entry for which driver object is required
+ * @addr: Mapped register io address of MDP
+ * Returns: Error code or allocated dpu_hw_dsc context
+ */
+struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(const struct dpu_dsc_cfg *cfg,
+				       void __iomem *addr);
+
 /**
  * dpu_hw_dsc_destroy - destroys dsc driver context
  * @dsc:   Pointer to dsc driver context returned by dpu_hw_dsc_init
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c
new file mode 100644
index 0000000000000..24fe1d98eb86a
--- /dev/null
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c
@@ -0,0 +1,387 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved.
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved
+ */
+
+#include <drm/display/drm_dsc_helper.h>
+
+#include "dpu_kms.h"
+#include "dpu_hw_catalog.h"
+#include "dpu_hwio.h"
+#include "dpu_hw_mdss.h"
+#include "dpu_hw_dsc.h"
+
+#define DSC_CMN_MAIN_CNF           0x00
+
+/* DPU_DSC_ENC register offsets */
+#define ENC_DF_CTRL                0x00
+#define ENC_GENERAL_STATUS         0x04
+#define ENC_HSLICE_STATUS          0x08
+#define ENC_OUT_STATUS             0x0C
+#define ENC_INT_STAT               0x10
+#define ENC_INT_CLR                0x14
+#define ENC_INT_MASK               0x18
+#define DSC_MAIN_CONF              0x30
+#define DSC_PICTURE_SIZE           0x34
+#define DSC_SLICE_SIZE             0x38
+#define DSC_MISC_SIZE              0x3C
+#define DSC_HRD_DELAYS             0x40
+#define DSC_RC_SCALE               0x44
+#define DSC_RC_SCALE_INC_DEC       0x48
+#define DSC_RC_OFFSETS_1           0x4C
+#define DSC_RC_OFFSETS_2           0x50
+#define DSC_RC_OFFSETS_3           0x54
+#define DSC_RC_OFFSETS_4           0x58
+#define DSC_FLATNESS_QP            0x5C
+#define DSC_RC_MODEL_SIZE          0x60
+#define DSC_RC_CONFIG              0x64
+#define DSC_RC_BUF_THRESH_0        0x68
+#define DSC_RC_BUF_THRESH_1        0x6C
+#define DSC_RC_BUF_THRESH_2        0x70
+#define DSC_RC_BUF_THRESH_3        0x74
+#define DSC_RC_MIN_QP_0            0x78
+#define DSC_RC_MIN_QP_1            0x7C
+#define DSC_RC_MIN_QP_2            0x80
+#define DSC_RC_MAX_QP_0            0x84
+#define DSC_RC_MAX_QP_1            0x88
+#define DSC_RC_MAX_QP_2            0x8C
+#define DSC_RC_RANGE_BPG_OFFSETS_0 0x90
+#define DSC_RC_RANGE_BPG_OFFSETS_1 0x94
+#define DSC_RC_RANGE_BPG_OFFSETS_2 0x98
+
+/* DPU_DSC_CTL register offsets */
+#define DSC_CTL                    0x00
+#define DSC_CFG                    0x04
+#define DSC_DATA_IN_SWAP           0x08
+#define DSC_CLK_CTRL               0x0C
+
+static int _dsc_calc_output_buf_max_addr(struct dpu_hw_dsc *hw_dsc, int num_softslice)
+{
+	int max_addr = 2400 / num_softslice;
+
+	if (hw_dsc->caps->features & BIT(DPU_DSC_NATIVE_42x_EN))
+		max_addr /= 2;
+
+	return max_addr - 1;
+};
+
+static void dpu_hw_dsc_disable_1_2(struct dpu_hw_dsc *hw_dsc)
+{
+	struct dpu_hw_blk_reg_map *hw;
+	const struct dpu_dsc_sub_blks *sblk;
+
+	if (!hw_dsc)
+		return;
+
+	hw = &hw_dsc->hw;
+	sblk = hw_dsc->caps->sblk;
+	DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, 0);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, 0);
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, 0);
+}
+
+static void dpu_hw_dsc_config_1_2(struct dpu_hw_dsc *hw_dsc,
+				  struct drm_dsc_config *dsc,
+				  u32 mode,
+				  u32 initial_lines)
+{
+	struct dpu_hw_blk_reg_map *hw;
+	const struct dpu_dsc_sub_blks *sblk;
+	u32 data = 0;
+	u32 det_thresh_flatness;
+	u32 num_active_slice_per_enc;
+	u32 bpp;
+
+	if (!hw_dsc || !dsc)
+		return;
+
+	hw = &hw_dsc->hw;
+
+	sblk = hw_dsc->caps->sblk;
+
+	if (mode & DSC_MODE_SPLIT_PANEL)
+		data |= BIT(0);
+
+	if (mode & DSC_MODE_MULTIPLEX)
+		data |= BIT(1);
+
+	num_active_slice_per_enc = dsc->slice_count;
+	if (mode & DSC_MODE_MULTIPLEX)
+		num_active_slice_per_enc = dsc->slice_count / 2;
+
+	data |= (num_active_slice_per_enc & 0x3) << 7;
+
+	DPU_REG_WRITE(hw, DSC_CMN_MAIN_CNF, data);
+
+	data = (initial_lines & 0xff);
+
+	if (mode & DSC_MODE_VIDEO)
+		data |= BIT(9);
+
+	data |= (_dsc_calc_output_buf_max_addr(hw_dsc, num_active_slice_per_enc) << 18);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, data);
+
+	data = (dsc->dsc_version_minor & 0xf) << 28;
+	if (dsc->dsc_version_minor == 0x2) {
+		if (dsc->native_422)
+			data |= BIT(22);
+		if (dsc->native_420)
+			data |= BIT(21);
+	}
+
+	bpp = dsc->bits_per_pixel;
+	/* as per hw requirement bpp should be programmed
+	 * twice the actual value in case of 420 or 422 encoding
+	 */
+	if (dsc->native_422 || dsc->native_420)
+		bpp = 2 * bpp;
+
+	data |= bpp << 10;
+
+	if (dsc->block_pred_enable)
+		data |= BIT(20);
+
+	if (dsc->convert_rgb)
+		data |= BIT(4);
+
+	data |= (dsc->line_buf_depth & 0xf) << 6;
+	data |= dsc->bits_per_component & 0xf;
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, data);
+
+	data = (dsc->pic_width & 0xffff) |
+		((dsc->pic_height & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_PICTURE_SIZE, data);
+
+	data = (dsc->slice_width & 0xffff) |
+		((dsc->slice_height & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_SLICE_SIZE, data);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_MISC_SIZE,
+		      (dsc->slice_chunk_size) & 0xffff);
+
+	data = (dsc->initial_xmit_delay & 0xffff) |
+		((dsc->initial_dec_delay & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_HRD_DELAYS, data);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE,
+		      dsc->initial_scale_value & 0x3f);
+
+	data = (dsc->scale_increment_interval & 0xffff) |
+		((dsc->scale_decrement_interval & 0x7ff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE_INC_DEC, data);
+
+	data = (dsc->first_line_bpg_offset & 0x1f) |
+		((dsc->second_line_bpg_offset & 0x1f) << 5);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_1, data);
+
+	data = (dsc->nfl_bpg_offset & 0xffff) |
+		((dsc->slice_bpg_offset & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_2, data);
+
+	data = (dsc->initial_offset & 0xffff) |
+		((dsc->final_offset & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_3, data);
+
+	data = (dsc->nsl_bpg_offset & 0xffff) |
+		((dsc->second_line_offset_adj & 0xffff) << 16);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_4, data);
+
+	det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc);
+	data = (dsc->flatness_min_qp & 0x1f) |
+		((dsc->flatness_max_qp & 0x1f) << 5) |
+		((det_thresh_flatness & 0xff) << 10);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_FLATNESS_QP, data);
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MODEL_SIZE,
+		      (dsc->rc_model_size) & 0xffff);
+
+	data = dsc->rc_edge_factor & 0xf;
+	data |= (dsc->rc_quant_incr_limit0 & 0x1f) << 8;
+	data |= (dsc->rc_quant_incr_limit1 & 0x1f) << 13;
+	data |= (dsc->rc_tgt_offset_high & 0xf) << 20;
+	data |= (dsc->rc_tgt_offset_low & 0xf) << 24;
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_CONFIG, data);
+
+	/* program the dsc wrapper */
+	data = BIT(0); /* encoder enable */
+	if (dsc->native_422)
+		data |= BIT(8);
+	else if (dsc->native_420)
+		data |= BIT(9);
+	if (!dsc->convert_rgb)
+		data |= BIT(10);
+	if (dsc->bits_per_component == 8)
+		data |= BIT(11);
+	if (mode & DSC_MODE_SPLIT_PANEL)
+		data |= BIT(12);
+	if (mode & DSC_MODE_MULTIPLEX)
+		data |= BIT(13);
+	if (!(mode & DSC_MODE_VIDEO))
+		data |= BIT(17);
+
+	DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, data);
+}
+
+static void dpu_hw_dsc_config_thresh_1_2(struct dpu_hw_dsc *hw_dsc,
+					 struct drm_dsc_config *dsc)
+{
+	struct dpu_hw_blk_reg_map *hw;
+	const struct dpu_dsc_sub_blks *sblk;
+	struct drm_dsc_rc_range_parameters *rc;
+
+	if (!hw_dsc || !dsc)
+		return;
+
+	hw = &hw_dsc->hw;
+
+	sblk = hw_dsc->caps->sblk;
+
+	rc = dsc->rc_range_params;
+
+	/*
+	 * With BUF_THRESH -- 14 in total
+	 * each register contains 4 thresh values with the last register
+	 * containing only 2 thresh values
+	 */
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_0,
+		      (dsc->rc_buf_thresh[0] << 0) |
+		      (dsc->rc_buf_thresh[1] << 8) |
+		      (dsc->rc_buf_thresh[2] << 16) |
+		      (dsc->rc_buf_thresh[3] << 24));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_1,
+		      (dsc->rc_buf_thresh[4] << 0) |
+		      (dsc->rc_buf_thresh[5] << 8) |
+		      (dsc->rc_buf_thresh[6] << 16) |
+		      (dsc->rc_buf_thresh[7] << 24));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_2,
+		      (dsc->rc_buf_thresh[8] << 0) |
+		      (dsc->rc_buf_thresh[9] << 8) |
+		      (dsc->rc_buf_thresh[10] << 16) |
+		      (dsc->rc_buf_thresh[11] << 24));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_3,
+		      (dsc->rc_buf_thresh[12] << 0) |
+		      (dsc->rc_buf_thresh[13] << 8));
+
+	/*
+	 * with min/max_QP -- 5 bits
+	 * each register contains 5 min_qp or max_qp for total of 15
+	 *
+	 * With BPG_OFFSET -- 6 bits
+	 * each register contains 5 BPG_offset for total of 15
+	 */
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_0,
+		      (rc[0].range_min_qp << 0) |
+		      (rc[1].range_min_qp << 5) |
+		      (rc[2].range_min_qp << 10) |
+		      (rc[3].range_min_qp << 15) |
+		      (rc[4].range_min_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_0,
+		      (rc[0].range_max_qp << 0) |
+		      (rc[1].range_max_qp << 5) |
+		      (rc[2].range_max_qp << 10) |
+		      (rc[3].range_max_qp << 15) |
+		      (rc[4].range_max_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_0,
+		      (rc[0].range_bpg_offset << 0) |
+		      (rc[1].range_bpg_offset << 6) |
+		      (rc[2].range_bpg_offset << 12) |
+		      (rc[3].range_bpg_offset << 18) |
+		      (rc[4].range_bpg_offset << 24));
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_1,
+		      (rc[5].range_min_qp << 0) |
+		      (rc[6].range_min_qp << 5) |
+		      (rc[7].range_min_qp << 10) |
+		      (rc[8].range_min_qp << 15) |
+		      (rc[9].range_min_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_1,
+		      (rc[5].range_max_qp << 0) |
+		      (rc[6].range_max_qp << 5) |
+		      (rc[7].range_max_qp << 10) |
+		      (rc[8].range_max_qp << 15) |
+		      (rc[9].range_max_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_1,
+		      (rc[5].range_bpg_offset << 0) |
+		      (rc[6].range_bpg_offset << 6) |
+		      (rc[7].range_bpg_offset << 12) |
+		      (rc[8].range_bpg_offset << 18) |
+		      (rc[9].range_bpg_offset << 24));
+
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_2,
+		      (rc[10].range_min_qp << 0) |
+		      (rc[11].range_min_qp << 5) |
+		      (rc[12].range_min_qp << 10) |
+		      (rc[13].range_min_qp << 15) |
+		      (rc[14].range_min_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_2,
+		      (rc[10].range_max_qp << 0) |
+		      (rc[11].range_max_qp << 5) |
+		      (rc[12].range_max_qp << 10) |
+		      (rc[13].range_max_qp << 15) |
+		      (rc[14].range_max_qp << 20));
+	DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_2,
+		      (rc[10].range_bpg_offset << 0) |
+		      (rc[11].range_bpg_offset << 6) |
+		      (rc[12].range_bpg_offset << 12) |
+		      (rc[13].range_bpg_offset << 18) |
+		      (rc[14].range_bpg_offset << 24));
+}
+
+static void dpu_hw_dsc_bind_pingpong_blk_1_2(struct dpu_hw_dsc *hw_dsc,
+					     const enum dpu_pingpong pp)
+{
+	struct dpu_hw_blk_reg_map *hw;
+	const struct dpu_dsc_sub_blks *sblk;
+	int mux_cfg = 0xf; /* Disabled */
+
+	hw = &hw_dsc->hw;
+
+	sblk = hw_dsc->caps->sblk;
+
+	if (pp)
+		mux_cfg = (pp - PINGPONG_0) & 0x7;
+
+	DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CTL, mux_cfg);
+}
+
+static void _setup_dcs_ops_1_2(struct dpu_hw_dsc_ops *ops,
+			       const unsigned long features)
+{
+	ops->dsc_disable = dpu_hw_dsc_disable_1_2;
+	ops->dsc_config = dpu_hw_dsc_config_1_2;
+	ops->dsc_config_thresh = dpu_hw_dsc_config_thresh_1_2;
+	ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk_1_2;
+}
+
+struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(const struct dpu_dsc_cfg *cfg,
+				       void __iomem *addr)
+{
+	struct dpu_hw_dsc *c;
+
+	c = kzalloc(sizeof(*c), GFP_KERNEL);
+	if (!c)
+		return ERR_PTR(-ENOMEM);
+
+	c->hw.blk_addr = addr + cfg->base;
+	c->hw.log_mask = DPU_DBG_MASK_DSC;
+
+	c->idx = cfg->id;
+	c->caps = cfg;
+	_setup_dcs_ops_1_2(&c->ops, c->caps->features);
+
+	return c;
+}
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index 2ca3c666e22ff..471842bbb9507 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
  * Copyright (c) 2016-2018, The Linux Foundation. All rights reserved.
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #define pr_fmt(fmt)	"[drm:%s] " fmt, __func__
@@ -212,7 +213,11 @@ int dpu_rm_init(struct dpu_rm *rm,
 		struct dpu_hw_dsc *hw;
 		const struct dpu_dsc_cfg *dsc = &cat->dsc[i];
 
-		hw = dpu_hw_dsc_init(dsc, mmio);
+		if (test_bit(DPU_DSC_HW_REV_1_2, &dsc->features))
+			hw = dpu_hw_dsc_init_1_2(dsc, mmio);
+		else
+			hw = dpu_hw_dsc_init(dsc, mmio);
+
 		if (IS_ERR(hw)) {
 			rc = PTR_ERR(hw);
 			DPU_ERROR("failed dsc object creation: err %d\n", rc);
-- 
GitLab


From 0d1b10c633468e84e64412b8061f6f7f678f14cb Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Thu, 25 May 2023 10:40:57 -0700
Subject: [PATCH 1674/1791] drm/msm/dpu: add DSC 1.2 hw blocks for relevant
 chipsets

Add DSC 1.2 hardware blocks to the catalog with necessary sub-block and
feature flag information.  Each display compression engine (DCE) contains
dual DSC encoders so both share same base address but with its own
different sub block address.

changes in v4:
-- delete DPU_DSC_HW_REV_1_1
-- re arrange sc8280xp_dsc[]

changes in v4:
-- fix checkpatch warning

changes in v10:
-- remove hard slice from commit text
-- replace DPU_DSC_NATIVE_422_EN with DPU_DSC_NATIVE_42x_EN
-- change DSC_BLK_1_2 .len from 0x100 to 0x29c

changes in v11:
-- remove comment at DSC_BLK_1_2 marco

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Kuogee Hsieh <quic_khsieh@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/539507/
Link: https://lore.kernel.org/r/1685036458-22683-10-git-send-email-quic_khsieh@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../msm/disp/dpu1/catalog/dpu_7_0_sm8350.h    | 14 +++++++++++++
 .../msm/disp/dpu1/catalog/dpu_7_2_sc7280.h    |  7 +++++++
 .../msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h  | 16 +++++++++++++++
 .../msm/disp/dpu1/catalog/dpu_8_1_sm8450.h    | 14 +++++++++++++
 .../msm/disp/dpu1/catalog/dpu_9_0_sm8550.h    | 14 +++++++++++++
 .../gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c    | 20 ++++++++++++++++++-
 6 files changed, 84 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
index 3c1b2c13398da..8da424eaee6a2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h
@@ -153,6 +153,18 @@ static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = {
 	MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000),
 };
 
+/*
+ * NOTE: Each display compression engine (DCE) contains dual hard
+ * slice DSC encoders so both share same base address but with
+ * its own different sub block address.
+ */
+static const struct dpu_dsc_cfg sm8350_dsc[] = {
+	DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0),
+	DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1),
+	DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0),
+	DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1),
+};
+
 static const struct dpu_intf_cfg sm8350_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
@@ -215,6 +227,8 @@ const struct dpu_mdss_cfg dpu_sm8350_cfg = {
 	.dspp = sm8350_dspp,
 	.pingpong_count = ARRAY_SIZE(sm8350_pp),
 	.pingpong = sm8350_pp,
+	.dsc_count = ARRAY_SIZE(sm8350_dsc),
+	.dsc = sm8350_dsc,
 	.merge_3d_count = ARRAY_SIZE(sm8350_merge_3d),
 	.merge_3d = sm8350_merge_3d,
 	.intf_count = ARRAY_SIZE(sm8350_intf),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index 5d894cbb0a62f..9cd361be28faa 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -94,6 +94,11 @@ static const struct dpu_pingpong_cfg sc7280_pp[] = {
 	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, -1, -1),
 };
 
+/* NOTE: sc7280 only has one DSC hard slice encoder */
+static const struct dpu_dsc_cfg sc7280_dsc[] = {
+	DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0),
+};
+
 static const struct dpu_wb_cfg sc7280_wb[] = {
 	WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6,
 			VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4),
@@ -155,6 +160,8 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = {
 	.mixer = sc7280_lm,
 	.pingpong_count = ARRAY_SIZE(sc7280_pp),
 	.pingpong = sc7280_pp,
+	.dsc_count = ARRAY_SIZE(sc7280_dsc),
+	.dsc = sc7280_dsc,
 	.wb_count = ARRAY_SIZE(sc7280_wb),
 	.wb = sc7280_wb,
 	.intf_count = ARRAY_SIZE(sc7280_intf),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
index c3f1ae000a212..f6ce6b090f718 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h
@@ -142,6 +142,20 @@ static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = {
 	MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000),
 };
 
+/*
+ * NOTE: Each display compression engine (DCE) contains dual hard
+ * slice DSC encoders so both share same base address but with
+ * its own different sub block address.
+ */
+static const struct dpu_dsc_cfg sc8280xp_dsc[] = {
+	DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0),
+	DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1),
+	DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0),
+	DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1),
+	DSC_BLK_1_2("dce_2_0", DSC_4, 0x82000, 0x29c, 0, dsc_sblk_0),
+	DSC_BLK_1_2("dce_2_1", DSC_5, 0x82000, 0x29c, 0, dsc_sblk_1),
+};
+
 /* TODO: INTF 3, 8 and 7 are used for MST, marked as INTF_NONE for now */
 static const struct dpu_intf_cfg sc8280xp_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
@@ -217,6 +231,8 @@ const struct dpu_mdss_cfg dpu_sc8280xp_cfg = {
 	.dspp = sc8280xp_dspp,
 	.pingpong_count = ARRAY_SIZE(sc8280xp_pp),
 	.pingpong = sc8280xp_pp,
+	.dsc_count = ARRAY_SIZE(sc8280xp_dsc),
+	.dsc = sc8280xp_dsc,
 	.merge_3d_count = ARRAY_SIZE(sc8280xp_merge_3d),
 	.merge_3d = sc8280xp_merge_3d,
 	.intf_count = ARRAY_SIZE(sc8280xp_intf),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
index 86c2e68ebd2c8..8d13c369213c0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h
@@ -161,6 +161,18 @@ static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = {
 	MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x65f00),
 };
 
+/*
+ * NOTE: Each display compression engine (DCE) contains dual hard
+ * slice DSC encoders so both share same base address but with
+ * its own different sub block address.
+ */
+static const struct dpu_dsc_cfg sm8450_dsc[] = {
+	DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0),
+	DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1),
+	DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0),
+	DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1),
+};
+
 static const struct dpu_intf_cfg sm8450_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
@@ -223,6 +235,8 @@ const struct dpu_mdss_cfg dpu_sm8450_cfg = {
 	.dspp = sm8450_dspp,
 	.pingpong_count = ARRAY_SIZE(sm8450_pp),
 	.pingpong = sm8450_pp,
+	.dsc_count = ARRAY_SIZE(sm8450_dsc),
+	.dsc = sm8450_dsc,
 	.merge_3d_count = ARRAY_SIZE(sm8450_merge_3d),
 	.merge_3d = sm8450_merge_3d,
 	.intf_count = ARRAY_SIZE(sm8450_intf),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
index 85dc34458b88a..f17b9a7fee851 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h
@@ -165,6 +165,18 @@ static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = {
 	MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x66700),
 };
 
+/*
+ * NOTE: Each display compression engine (DCE) contains dual hard
+ * slice DSC encoders so both share same base address but with
+ * its own different sub block address.
+ */
+static const struct dpu_dsc_cfg sm8550_dsc[] = {
+	DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0),
+	DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1),
+	DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0),
+	DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1),
+};
+
 static const struct dpu_intf_cfg sm8550_intf[] = {
 	INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24),
@@ -227,6 +239,8 @@ const struct dpu_mdss_cfg dpu_sm8550_cfg = {
 	.dspp = sm8550_dspp,
 	.pingpong_count = ARRAY_SIZE(sm8550_pp),
 	.pingpong = sm8550_pp,
+	.dsc_count = ARRAY_SIZE(sm8550_dsc),
+	.dsc = sm8550_dsc,
 	.merge_3d_count = ARRAY_SIZE(sm8550_merge_3d),
 	.merge_3d = sm8550_merge_3d,
 	.intf_count = ARRAY_SIZE(sm8550_intf),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index b9f1d58a6cace..008df60b00f05 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved.
- * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #define pr_fmt(fmt)	"[drm:%s:%d] " fmt, __func__, __LINE__
@@ -516,6 +516,16 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 /*************************************************************
  * DSC sub blocks config
  *************************************************************/
+static const struct dpu_dsc_sub_blks dsc_sblk_0 = {
+	.enc = {.base = 0x100, .len = 0x100},
+	.ctl = {.base = 0xF00, .len = 0x10},
+};
+
+static const struct dpu_dsc_sub_blks dsc_sblk_1 = {
+	.enc = {.base = 0x200, .len = 0x100},
+	.ctl = {.base = 0xF80, .len = 0x10},
+};
+
 #define DSC_BLK(_name, _id, _base, _features) \
 	{\
 	.name = _name, .id = _id, \
@@ -523,6 +533,14 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 	.features = _features, \
 	}
 
+#define DSC_BLK_1_2(_name, _id, _base, _len, _features, _sblk) \
+	{\
+	.name = _name, .id = _id, \
+	.base = _base, .len = _len, \
+	.features = BIT(DPU_DSC_HW_REV_1_2) | _features, \
+	.sblk = &_sblk, \
+	}
+
 /*************************************************************
  * INTF sub blocks config
  *************************************************************/
-- 
GitLab


From 21bf617110bab03001bd86ec1a822164bacbc747 Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Fri, 9 Jun 2023 15:57:13 -0700
Subject: [PATCH 1675/1791] msm/drm/dsi: Round up DSC hdisplay calculation

Currently, when compression is enabled, hdisplay is reduced via integer
division. This causes issues for modes where the original hdisplay is
not a multiple of 3.

To fix this, use DIV_ROUND_UP to divide hdisplay.

Suggested-by: Marijn Suijten <marijn.suijten@somainline.org>
Fixes: 08802f515c3cf ("drm/msm/dsi: Add support for DSC configuration")
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/541970/
Link: https://lore.kernel.org/r/20230405-add-dsc-support-v6-1-95eab864d1b6@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 1a99d75025dc4..a448931af804b 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -949,7 +949,7 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 		 * pulse width same
 		 */
 		h_total -= hdisplay;
-		hdisplay = msm_dsc_get_bytes_per_line(msm_host->dsc) / 3;
+		hdisplay = DIV_ROUND_UP(msm_dsc_get_bytes_per_line(msm_host->dsc), 3);
 		h_total += hdisplay;
 		ha_end = ha_start + hdisplay;
 	}
-- 
GitLab


From 7c9e4a554d4a712bc6891a49d52b1ec030ed5b4a Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Fri, 9 Jun 2023 15:57:14 -0700
Subject: [PATCH 1676/1791] drm/msm/dsi: Reduce pclk rate for compression

Adjust the pclk rate to divide hdisplay by the compression ratio when DSC
is enabled.

Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/541972/
Link: https://lore.kernel.org/r/20230405-add-dsc-support-v6-2-95eab864d1b6@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index a448931af804b..98ea1da492c79 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -561,12 +561,27 @@ void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host)
 	clk_disable_unprepare(msm_host->byte_clk);
 }
 
-static unsigned long dsi_get_pclk_rate(const struct drm_display_mode *mode, bool is_bonded_dsi)
+static unsigned long dsi_adjust_pclk_for_compression(const struct drm_display_mode *mode,
+		const struct drm_dsc_config *dsc)
+{
+	int new_hdisplay = DIV_ROUND_UP(mode->hdisplay * drm_dsc_get_bpp_int(dsc),
+			dsc->bits_per_component * 3);
+
+	int new_htotal = mode->htotal - mode->hdisplay + new_hdisplay;
+
+	return new_htotal * mode->vtotal * drm_mode_vrefresh(mode);
+}
+
+static unsigned long dsi_get_pclk_rate(const struct drm_display_mode *mode,
+		const struct drm_dsc_config *dsc, bool is_bonded_dsi)
 {
 	unsigned long pclk_rate;
 
 	pclk_rate = mode->clock * 1000;
 
+	if (dsc)
+		pclk_rate = dsi_adjust_pclk_for_compression(mode, dsc);
+
 	/*
 	 * For bonded DSI mode, the current DRM mode has the complete width of the
 	 * panel. Since, the complete panel is driven by two DSI controllers,
@@ -585,7 +600,7 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
 	u8 lanes = msm_host->lanes;
 	u32 bpp = dsi_get_bpp(msm_host->format);
-	unsigned long pclk_rate = dsi_get_pclk_rate(mode, is_bonded_dsi);
+	unsigned long pclk_rate = dsi_get_pclk_rate(mode, msm_host->dsc, is_bonded_dsi);
 	unsigned long pclk_bpp;
 
 	if (lanes == 0) {
@@ -604,7 +619,7 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d
 
 static void dsi_calc_pclk(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 {
-	msm_host->pixel_clk_rate = dsi_get_pclk_rate(msm_host->mode, is_bonded_dsi);
+	msm_host->pixel_clk_rate = dsi_get_pclk_rate(msm_host->mode, msm_host->dsc, is_bonded_dsi);
 	msm_host->byte_clk_rate = dsi_byte_clk_get_rate(&msm_host->base, is_bonded_dsi,
 							msm_host->mode);
 
-- 
GitLab


From 22598cfc94bb1d6590c49ea749c785e79463cb2f Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Fri, 9 Jun 2023 15:57:15 -0700
Subject: [PATCH 1677/1791] drm/msm/dpu: Add DPU_INTF_DATA_COMPRESS feature
 flag for DPU >= 7.0

In DPU 7.x and later, DSC/DCE enablement registers have been moved from
PINGPONG to INTF. Thus, add a DPU_INTF_DATA_COMPRESS feature flag that will
be set if the DATA_COMPRESS register is in the INTF block.

Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/541967/
Link: https://lore.kernel.org/r/20230405-add-dsc-support-v6-3-95eab864d1b6@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 008df60b00f05..36ba3f58dcdf0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -105,7 +105,7 @@
 	 BIT(DPU_INTF_STATUS_SUPPORTED) | \
 	 BIT(DPU_DATA_HCTL_EN))
 
-#define INTF_SC7280_MASK (INTF_SC7180_MASK)
+#define INTF_SC7280_MASK (INTF_SC7180_MASK | BIT(DPU_INTF_DATA_COMPRESS))
 
 #define WB_SM8250_MASK (BIT(DPU_WB_LINE_MODE) | \
 			 BIT(DPU_WB_UBWC) | \
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index d3598dd9d448b..b860784ade726 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -181,6 +181,7 @@ enum {
  * @DPU_DATA_HCTL_EN                Allows data to be transferred at different rate
  *                                  than video timing
  * @DPU_INTF_STATUS_SUPPORTED       INTF block has INTF_STATUS register
+ * @DPU_INTF_DATA_COMPRESS          INTF block has DATA_COMPRESS register
  * @DPU_INTF_MAX
  */
 enum {
@@ -188,6 +189,7 @@ enum {
 	DPU_INTF_TE,
 	DPU_DATA_HCTL_EN,
 	DPU_INTF_STATUS_SUPPORTED,
+	DPU_INTF_DATA_COMPRESS,
 	DPU_INTF_MAX
 };
 
-- 
GitLab


From 1642b580347372a985ff4fbe295c7af44090c51a Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Fri, 9 Jun 2023 15:57:16 -0700
Subject: [PATCH 1678/1791] drm/msm/dpu: Set DATA_COMPRESS on command mode for
 DCE/DSC 1.2

Add a DPU INTF op to set the DCE_DATA_COMPRESS bit to enable the
DCE/DSC 1.2 datapath

Note: For now, this op is called for command mode encoders only. Changes to
set DATA_COMPRESS for video mode encoders will be posted along with DSC
v1.2 support for DP.

Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/541966/
Link: https://lore.kernel.org/r/20230405-add-dsc-support-v6-4-95eab864d1b6@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c    |  3 +++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c         | 13 +++++++++++++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h         |  3 +++
 3 files changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
index 63ba0082b6ee4..b856c6286c85a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c
@@ -67,6 +67,9 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg(
 		phys_enc->hw_intf->ops.bind_pingpong_blk(
 				phys_enc->hw_intf,
 				phys_enc->hw_pp->idx);
+
+	if (intf_cfg.dsc != 0 && phys_enc->hw_intf->ops.enable_compression)
+		phys_enc->hw_intf->ops.enable_compression(phys_enc->hw_intf);
 }
 
 static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg, int irq_idx)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
index 530f82e34c1ec..5b0f6627e29b8 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c
@@ -91,6 +91,7 @@
 
 #define INTF_CFG2_DATABUS_WIDEN	BIT(0)
 #define INTF_CFG2_DATA_HCTL_EN	BIT(4)
+#define INTF_CFG2_DCE_DATA_COMPRESS     BIT(12)
 
 
 static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx,
@@ -512,6 +513,15 @@ static void dpu_hw_intf_disable_autorefresh(struct dpu_hw_intf *intf,
 
 }
 
+static void dpu_hw_intf_enable_compression(struct dpu_hw_intf *ctx)
+{
+	u32 intf_cfg2 = DPU_REG_READ(&ctx->hw, INTF_CONFIG2);
+
+	intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS;
+
+	DPU_REG_WRITE(&ctx->hw, INTF_CONFIG2, intf_cfg2);
+}
+
 static void _setup_intf_ops(struct dpu_hw_intf_ops *ops,
 		unsigned long cap)
 {
@@ -532,6 +542,9 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops,
 		ops->vsync_sel = dpu_hw_intf_vsync_sel;
 		ops->disable_autorefresh = dpu_hw_intf_disable_autorefresh;
 	}
+
+	if (cap & BIT(DPU_INTF_DATA_COMPRESS))
+		ops->enable_compression = dpu_hw_intf_enable_compression;
 }
 
 struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg,
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
index 33895eca12119..99e21c4137f91 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h
@@ -70,6 +70,7 @@ struct intf_status {
  * @get_autorefresh:            Retrieve autorefresh config from hardware
  *                              Return: 0 on success, -ETIMEDOUT on timeout
  * @vsync_sel:                  Select vsync signal for tear-effect configuration
+ * @enable_compression:         Enable data compression
  */
 struct dpu_hw_intf_ops {
 	void (*setup_timing_gen)(struct dpu_hw_intf *intf,
@@ -106,6 +107,8 @@ struct dpu_hw_intf_ops {
 	 * Disable autorefresh if enabled
 	 */
 	void (*disable_autorefresh)(struct dpu_hw_intf *intf, uint32_t encoder_id, u16 vdisplay);
+
+	void (*enable_compression)(struct dpu_hw_intf *intf);
 };
 
 struct dpu_hw_intf {
-- 
GitLab


From 155fa3a91d64221eb0885fd221cc8085dbef908f Mon Sep 17 00:00:00 2001
From: Jessica Zhang <quic_jesszhan@quicinc.com>
Date: Fri, 9 Jun 2023 15:57:17 -0700
Subject: [PATCH 1679/1791] drm/msm/dsi: Remove incorrect references to
 slice_count

Currently, slice_count is being used to calculate word count and
pkt_per_line. Instead, these values should be calculated using slice per
packet, which is not the same as slice_count.

Slice count represents the number of slices per interface, and its value
will not always match that of slice per packet. For example, it is possible
to have cases where there are multiple slices per interface but the panel
specifies only one slice per packet.

Thus, use the default value of one slice per packet and remove slice_count
from the aforementioned calculations.

Fixes: 08802f515c3c ("drm/msm/dsi: Add support for DSC configuration")
Fixes: bc6b6ff8135c ("drm/msm/dsi: Use DSC slice(s) packet size to compute word count")
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Signed-off-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/541965/
Link: https://lore.kernel.org/r/20230405-add-dsc-support-v6-5-95eab864d1b6@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 98ea1da492c79..fb1d3a25765f3 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -863,18 +863,17 @@ static void dsi_update_dsc_timing(struct msm_dsi_host *msm_host, bool is_cmd_mod
 	 */
 	slice_per_intf = msm_dsc_get_slices_per_intf(dsc, hdisplay);
 
-	/*
-	 * If slice_count is greater than slice_per_intf
-	 * then default to 1. This can happen during partial
-	 * update.
-	 */
-	if (dsc->slice_count > slice_per_intf)
-		dsc->slice_count = 1;
-
 	total_bytes_per_intf = dsc->slice_chunk_size * slice_per_intf;
 
 	eol_byte_num = total_bytes_per_intf % 3;
-	pkt_per_line = slice_per_intf / dsc->slice_count;
+
+	/*
+	 * Typically, pkt_per_line = slice_per_intf * slice_per_pkt.
+	 *
+	 * Since the current driver only supports slice_per_pkt = 1,
+	 * pkt_per_line will be equal to slice per intf for now.
+	 */
+	pkt_per_line = slice_per_intf;
 
 	if (is_cmd_mode) /* packet data type */
 		reg = DSI_COMMAND_COMPRESSION_MODE_CTRL_STREAM0_DATATYPE(MIPI_DSI_DCS_LONG_WRITE);
@@ -998,7 +997,14 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 		if (!msm_host->dsc)
 			wc = hdisplay * dsi_get_bpp(msm_host->format) / 8 + 1;
 		else
-			wc = msm_host->dsc->slice_chunk_size * msm_host->dsc->slice_count + 1;
+			/*
+			 * When DSC is enabled, WC = slice_chunk_size * slice_per_pkt + 1.
+			 * Currently, the driver only supports default value of slice_per_pkt = 1
+			 *
+			 * TODO: Expand mipi_dsi_device struct to hold slice_per_pkt info
+			 *       and adjust DSC math to account for slice_per_pkt.
+			 */
+			wc = msm_host->dsc->slice_chunk_size + 1;
 
 		dsi_write(msm_host, REG_DSI_CMD_MDP_STREAM0_CTRL,
 			DSI_CMD_MDP_STREAM0_CTRL_WORD_COUNT(wc) |
-- 
GitLab


From fda520976ef49d3ab66f6157053b084721d86712 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Mon, 12 Jun 2023 06:16:15 +0300
Subject: [PATCH 1680/1791] drm/msm: provide fb_dirty implemenation

Since commit 93e81e38e197 ("drm/fb_helper: Minimize damage-helper
overhead") the drm_fb_helper_funcs::fb_dirty helper is required for
proper dirty/damage processing. The drm/msm driver requires that to
function to let CMD panels to work. Use simplified version of
drm_fbdev_generic_helper_fb_dirty() to fix support for CMD mode panels.

Reported-by: Degdag Mohamed <degdagmohamed@gmail.com>
Fixes: 93e81e38e197 ("drm/fb_helper: Minimize damage-helper overhead")
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/542002/
Link: https://lore.kernel.org/r/20230612031616.3620134-1-dmitry.baryshkov@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/msm_fbdev.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index fa9c1cbffae3c..b933a85420f6d 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -139,8 +139,28 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 	return ret;
 }
 
+static int msm_fbdev_fb_dirty(struct drm_fb_helper *helper,
+			      struct drm_clip_rect *clip)
+{
+	struct drm_device *dev = helper->dev;
+	int ret;
+
+	/* Call damage handlers only if necessary */
+	if (!(clip->x1 < clip->x2 && clip->y1 < clip->y2))
+		return 0;
+
+	if (helper->fb->funcs->dirty) {
+		ret = helper->fb->funcs->dirty(helper->fb, NULL, 0, 0, clip, 1);
+		if (drm_WARN_ONCE(dev, ret, "Dirty helper failed: ret=%d\n", ret))
+			return ret;
+	}
+
+	return 0;
+}
+
 static const struct drm_fb_helper_funcs msm_fb_helper_funcs = {
 	.fb_probe = msm_fbdev_create,
+	.fb_dirty = msm_fbdev_fb_dirty,
 };
 
 /*
-- 
GitLab


From a7bfb2ad2184a1fba78be35209b6019aa8cc8d4d Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <quic_bjorande@quicinc.com>
Date: Mon, 12 Jun 2023 15:01:06 -0700
Subject: [PATCH 1681/1791] drm/msm/dp: Drop aux devices together with DP
 controller

Using devres to depopulate the aux bus made sure that upon a probe
deferral the EDP panel device would be destroyed and recreated upon next
attempt.

But the struct device which the devres is tied to is the DPUs
(drm_dev->dev), which may be happen after the DP controller is torn
down.

Indications of this can be seen in the commonly seen EDID-hexdump full
of zeros in the log, or the occasional/rare KASAN fault where the
panel's attempt to read the EDID information causes a use after free on
DP resources.

It's tempting to move the devres to the DP controller's struct device,
but the resources used by the device(s) on the aux bus are explicitly
torn down in the error path. The KASAN-reported use-after-free also
remains, as the DP aux "module" explicitly frees its devres-allocated
memory in this code path.

As such, explicitly depopulate the aux bus in the error path, and in the
component unbind path, to avoid these issues.

Fixes: 2b57f726611e ("drm/msm/dp: fix aux-bus EP lifetime")
Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/542163/
Link: https://lore.kernel.org/r/20230612220106.1884039-1-quic_bjorande@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dp/dp_display.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
index f7d4c2b709bd3..727aab2522de6 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.c
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -326,6 +326,8 @@ static void dp_display_unbind(struct device *dev, struct device *master,
 
 	kthread_stop(dp->ev_tsk);
 
+	of_dp_aux_depopulate_bus(dp->aux);
+
 	dp_power_client_deinit(dp->power);
 	dp_unregister_audio_driver(dev, dp->audio);
 	dp_aux_unregister(dp->aux);
@@ -1479,11 +1481,6 @@ void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor)
 	}
 }
 
-static void of_dp_aux_depopulate_bus_void(void *data)
-{
-	of_dp_aux_depopulate_bus(data);
-}
-
 static int dp_display_get_next_bridge(struct msm_dp *dp)
 {
 	int rc;
@@ -1511,12 +1508,6 @@ static int dp_display_get_next_bridge(struct msm_dp *dp)
 		of_node_put(aux_bus);
 		if (rc)
 			goto error;
-
-		rc = devm_add_action_or_reset(dp->drm_dev->dev,
-						of_dp_aux_depopulate_bus_void,
-						dp_priv->aux);
-		if (rc)
-			goto error;
 	} else if (dp->is_edp) {
 		DRM_ERROR("eDP aux_bus not found\n");
 		return -ENODEV;
@@ -1540,6 +1531,7 @@ static int dp_display_get_next_bridge(struct msm_dp *dp)
 
 error:
 	if (dp->is_edp) {
+		of_dp_aux_depopulate_bus(dp_priv->aux);
 		dp_display_host_phy_exit(dp_priv);
 		dp_display_host_deinit(dp_priv);
 	}
-- 
GitLab


From fa0048a4b1fa7a50c8b0e514f5b428abdf69a6f8 Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <quic_bjorande@quicinc.com>
Date: Mon, 12 Jun 2023 15:02:59 -0700
Subject: [PATCH 1682/1791] drm/msm/dp: Free resources after unregistering them

The DP component's unbind operation walks through the submodules to
unregister and clean things up. But if the unbind happens because the DP
controller itself is being removed, all the memory for those submodules
has just been freed.

Change the order of these operations to avoid the many use-after-free
that otherwise happens in this code path.

Fixes: c943b4948b58 ("drm/msm/dp: add displayPort driver support")
Signed-off-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542166/
Link: https://lore.kernel.org/r/20230612220259.1884381-1-quic_bjorande@quicinc.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 drivers/gpu/drm/msm/dp/dp_display.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c
index 727aab2522de6..76f13954015b1 100644
--- a/drivers/gpu/drm/msm/dp/dp_display.c
+++ b/drivers/gpu/drm/msm/dp/dp_display.c
@@ -1300,9 +1300,9 @@ static int dp_display_remove(struct platform_device *pdev)
 {
 	struct dp_display_private *dp = dev_get_dp_display_private(&pdev->dev);
 
+	component_del(&pdev->dev, &dp_display_comp_ops);
 	dp_display_deinit_sub_modules(dp);
 
-	component_del(&pdev->dev, &dp_display_comp_ops);
 	platform_set_drvdata(pdev, NULL);
 
 	return 0;
-- 
GitLab


From 30f90f3c1c2c63c2fa44f61233737d27b72637c2 Mon Sep 17 00:00:00 2001
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Date: Fri, 28 Apr 2023 11:23:50 -0400
Subject: [PATCH 1683/1791] drm/amd/display: Skip DPP DTO update if root clock
 is gated

[Why]
Hardware implements root clock gating by utilizing the DPP DTO registers
with a special case of DTO enabled, phase = 0, modulo = 1. This
conflicts with our policy to always update the DPPDTO for cases where
it's expected to be disabled.

The pipes unexpectedly enter a higher power state than expected because
of this programming flow.

[How]
Guard the upper layers of HWSS against this hardware quirk with
programming the register with an internal state flag in DCCG.

While technically acting as global state for the DCCG, HWSS shouldn't be
expected to understand the hardware quirk for having DTO disabled
causing more power than DTO enabled with this specific setting.

This also prevents sequencing errors from occuring in the future if
we have to program DPP DTO in multiple locations.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c   | 8 ++++++++
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c | 5 +++++
 drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h        | 1 +
 3 files changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
index 4c2fdfea162f5..65c1d754e2d6b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c
@@ -47,6 +47,14 @@ void dccg31_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+	if (dccg->dpp_clock_gated[dpp_inst]) {
+		/*
+		 * Do not update the DPPCLK DTO if the clock is stopped.
+		 * It is treated the same as if the pipe itself were in PG.
+		 */
+		return;
+	}
+
 	if (dccg->ref_dppclk && req_dppclk) {
 		int ref_dppclk = dccg->ref_dppclk;
 		int modulo, phase;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
index e0e7d32bb1a0e..cf23d7bc560a6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c
@@ -332,6 +332,9 @@ static void dccg314_dpp_root_clock_control(
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+	if (dccg->dpp_clock_gated[dpp_inst] == clock_on)
+		return;
+
 	if (clock_on) {
 		/* turn off the DTO and leave phase/modulo at max */
 		REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0);
@@ -345,6 +348,8 @@ static void dccg314_dpp_root_clock_control(
 			  DPPCLK0_DTO_PHASE, 0,
 			  DPPCLK0_DTO_MODULO, 1);
 	}
+
+	dccg->dpp_clock_gated[dpp_inst] = !clock_on;
 }
 
 static const struct dccg_funcs dccg314_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
index 0b700b3d7d972..8dc804bbe98bd 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -68,6 +68,7 @@ struct dccg {
 	const struct dccg_funcs *funcs;
 	int pipe_dppclk_khz[MAX_PIPES];
 	int ref_dppclk;
+	bool dpp_clock_gated[MAX_PIPES];
 	//int dtbclk_khz[MAX_PIPES];/* TODO needs to be removed */
 	//int audio_dtbclk_khz;/* TODO needs to be removed */
 	//int ref_dtbclk_khz;/* TODO needs to be removed */
-- 
GitLab


From c168feed5d0341b35b2f6a744f088e7625cfc1aa Mon Sep 17 00:00:00 2001
From: Lee Jones <lee@kernel.org>
Date: Fri, 9 Jun 2023 09:17:14 +0100
Subject: [PATCH 1684/1791] drm/amd/display/amdgpu_dm/amdgpu_dm_helpers: Move
 SYNAPTICS_DEVICE_ID into CONFIG_DRM_AMD_DC_DCN ifdef
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following W=1 kernel build warning(s):

 drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_helpers.c:48:22: warning: ‘SYNAPTICS_DEVICE_ID’ defined but not used [-Wunused-const-variable=]

Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com>
Cc: David Airlie <airlied@gmail.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: amd-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 09e056a647087..cd20cfc049969 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -44,9 +44,6 @@
 #include "dm_helpers.h"
 #include "ddc_service_types.h"
 
-/* MST Dock */
-static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA";
-
 /* dm_helpers_parse_edid_caps
  *
  * Parse edid caps
@@ -702,6 +699,9 @@ static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux)
 	DC_LOG_DC("Done apply_synaptics_fifo_reset_wa\n");
 }
 
+/* MST Dock */
+static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA";
+
 static uint8_t write_dsc_enable_synaptics_non_virtual_dpcd_mst(
 		struct drm_dp_aux *aux,
 		const struct dc_stream_state *stream,
-- 
GitLab


From 41ce6d6d03d5e51420ea7732c83facc8a7f2e5da Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Tue, 23 May 2023 11:55:54 -0400
Subject: [PATCH 1685/1791] drm/amdgpu: Rename DRM schedulers in amdgpu TTM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename mman.entity to mman.high_pr to make the distinction
clearer that this is a high priority scheduler. Similarly,
rename the recently added mman.delayed to mman.low_pr to
make it clear it is a low priority scheduler.
No functional change in this patch.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c  | 18 +++++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h  |  8 ++++----
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c   |  2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c |  2 +-
 4 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index d2d0d27f90539..0534ab7168094 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -228,7 +228,7 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo,
 	num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8);
 	num_bytes = num_pages * 8 * AMDGPU_GPU_PAGES_IN_CPU_PAGE;
 
-	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity,
+	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4 + num_bytes,
 				     AMDGPU_IB_POOL_DELAYED, &job);
@@ -1456,7 +1456,7 @@ static int amdgpu_ttm_access_memory_sdma(struct ttm_buffer_object *bo,
 		memcpy(adev->mman.sdma_access_ptr, buf, len);
 
 	num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8);
-	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity,
+	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4, AMDGPU_IB_POOL_DELAYED,
 				     &job);
@@ -2032,7 +2032,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 
 		ring = adev->mman.buffer_funcs_ring;
 		sched = &ring->sched;
-		r = drm_sched_entity_init(&adev->mman.entity,
+		r = drm_sched_entity_init(&adev->mman.high_pr,
 					  DRM_SCHED_PRIORITY_KERNEL, &sched,
 					  1, NULL);
 		if (r) {
@@ -2041,7 +2041,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 			return;
 		}
 
-		r = drm_sched_entity_init(&adev->mman.delayed,
+		r = drm_sched_entity_init(&adev->mman.low_pr,
 					  DRM_SCHED_PRIORITY_NORMAL, &sched,
 					  1, NULL);
 		if (r) {
@@ -2050,8 +2050,8 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 			goto error_free_entity;
 		}
 	} else {
-		drm_sched_entity_destroy(&adev->mman.entity);
-		drm_sched_entity_destroy(&adev->mman.delayed);
+		drm_sched_entity_destroy(&adev->mman.high_pr);
+		drm_sched_entity_destroy(&adev->mman.low_pr);
 		dma_fence_put(man->move);
 		man->move = NULL;
 	}
@@ -2067,7 +2067,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)
 	return;
 
 error_free_entity:
-	drm_sched_entity_destroy(&adev->mman.entity);
+	drm_sched_entity_destroy(&adev->mman.high_pr);
 }
 
 static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
@@ -2082,8 +2082,8 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
 		AMDGPU_IB_POOL_DIRECT :
 		AMDGPU_IB_POOL_DELAYED;
 	int r;
-	struct drm_sched_entity *entity = delayed ? &adev->mman.delayed :
-						    &adev->mman.entity;
+	struct drm_sched_entity *entity = delayed ? &adev->mman.low_pr :
+						    &adev->mman.high_pr;
 	r = amdgpu_job_alloc_with_ib(adev, entity,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4, pool, job);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index e82b1edee7a4e..6d0d66e40db93 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -59,10 +59,10 @@ struct amdgpu_mman {
 	bool					buffer_funcs_enabled;
 
 	struct mutex				gtt_window_lock;
-	/* Scheduler entity for buffer moves */
-	struct drm_sched_entity			entity;
-	/* Scheduler entity for VRAM clearing */
-	struct drm_sched_entity			delayed;
+	/* High priority scheduler entity for buffer moves */
+	struct drm_sched_entity			high_pr;
+	/* Low priority scheduler entity for VRAM clearing */
+	struct drm_sched_entity			low_pr;
 
 	struct amdgpu_vram_mgr vram_mgr;
 	struct amdgpu_gtt_mgr gtt_mgr;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index b2e42f1b0f12d..0c8a479895761 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -382,7 +382,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	 * translation. Avoid this by doing the invalidation from the SDMA
 	 * itself.
 	 */
-	r = amdgpu_job_alloc_with_ib(ring->adev, &adev->mman.entity,
+	r = amdgpu_job_alloc_with_ib(ring->adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     16 * 4, AMDGPU_IB_POOL_IMMEDIATE,
 				     &job);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 58d95fb995959..709ac885ca6d7 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -64,7 +64,7 @@ svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages,
 	num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8);
 	num_bytes = npages * 8;
 
-	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity,
+	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4 + num_bytes,
 				     AMDGPU_IB_POOL_DELAYED,
-- 
GitLab


From 17fbdbda9cc87ff5a013898de506212d25323ed7 Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Fri, 19 May 2023 13:02:01 -0400
Subject: [PATCH 1686/1791] drm/amd/display: Enable dcn314 DPP RCO

[Why and How]
Add back debug bits enabling RCO for dcn314 as underflow
associated with this change has been resolved

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn314/dcn314_resource.c  | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 837884c4f03a6..7957ad4d6a34a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -904,6 +904,22 @@ static const struct dc_debug_options debug_defaults_drv = {
 			.afmt = true,
 		}
 	},
+
+	.root_clock_optimization = {
+			.bits = {
+					.dpp = true,
+					.dsc = false,
+					.hdmistream = false,
+					.hdmichar = false,
+					.dpstream = false,
+					.symclk32_se = false,
+					.symclk32_le = false,
+					.symclk_fe = false,
+					.physymclk = false,
+					.dpiasymclk = false,
+			}
+	},
+
 	.seamless_boot_odm_combine = true
 };
 
-- 
GitLab


From e06da81749716ee3f0404fada97882609921d98f Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 9 Jun 2023 20:35:29 +0530
Subject: [PATCH 1687/1791] drm/amdgpu: Fix kdoc warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c:76: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * EEPROM Table structure v1
drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c:98: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * EEPROM Table structrue v2.1

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index c2e8f6491ac67..0648dfe559afc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -72,7 +72,7 @@
 /* Bad GPU tag ‘BADG’ */
 #define RAS_TABLE_HDR_BAD       0x42414447
 
-/**
+/*
  * EEPROM Table structure v1
  * ---------------------------------
  * |                               |
@@ -94,7 +94,7 @@
 #define RAS_MAX_RECORD_COUNT    ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE) \
 				 / RAS_TABLE_RECORD_SIZE)
 
-/**
+/*
  * EEPROM Table structrue v2.1
  * ---------------------------------
  * |                               |
-- 
GitLab


From 8020f0f9316b6961fe384031b4780e764eeb9652 Mon Sep 17 00:00:00 2001
From: Hamza Mahfooz <hamza.mahfooz@amd.com>
Date: Fri, 9 Jun 2023 11:31:04 -0400
Subject: [PATCH 1688/1791] drm/amd/amdgpu: enable W=1 for amdgpu

We have a clean build with W=1 as of
commit c168feed5d03 ("drm/amd/display/amdgpu_dm/amdgpu_dm_helpers: Move
SYNAPTICS_DEVICE_ID into CONFIG_DRM_AMD_DC_DCN ifdef"). So, let's enable
these checks unconditionally for the entire module to catch these errors
during development.

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 86b833085f194..8d16f280b6950 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -40,7 +40,18 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \
 	-I$(FULL_AMD_PATH)/amdkfd
 
 subdir-ccflags-y := -Wextra
-subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable)
+subdir-ccflags-y += -Wunused
+subdir-ccflags-y += -Wmissing-prototypes
+subdir-ccflags-y += -Wmissing-declarations
+subdir-ccflags-y += -Wmissing-include-dirs
+subdir-ccflags-y += -Wold-style-definition
+subdir-ccflags-y += -Wmissing-format-attribute
+# Need this to avoid recursive variable evaluation issues
+cond-flags := $(call cc-option, -Wunused-but-set-variable) \
+	$(call cc-option, -Wunused-const-variable) \
+	$(call cc-option, -Wstringop-truncation) \
+	$(call cc-option, -Wpacked-not-aligned)
+subdir-ccflags-y += $(cond-flags)
 subdir-ccflags-y += -Wno-unused-parameter
 subdir-ccflags-y += -Wno-type-limits
 subdir-ccflags-y += -Wno-sign-compare
-- 
GitLab


From c069dbbcba7319c514536820f2782a0af3361811 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 9 Jun 2023 16:03:56 +0200
Subject: [PATCH 1689/1791] drm/radeon: Disable outputs when releasing fbdev
 client
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Disable the modesetting pipeline before release the radeon's fbdev
client. Fixes the following error:

[   17.217408] WARNING: CPU: 5 PID: 1464 at drivers/gpu/drm/ttm/ttm_bo.c:326 ttm_bo_release+0x27e/0x2d0 [ttm]
[   17.217418] Modules linked in: edac_mce_amd radeon(+) drm_ttm_helper ttm video drm_suballoc_helper drm_display_helper kvm irqbypass drm_kms_helper syscopyarea crc32_pclmul sysfillrect sha512_ssse3 sysimgblt sha512_generic cfbfillrect cfbimgblt wmi_bmof aesni_intel cfbcopyarea crypto_simd cryptd k10temp acpi_cpufreq wmi dm_mod
[   17.217432] CPU: 5 PID: 1464 Comm: systemd-udevd Not tainted 6.4.0-rc4+ #1
[   17.217436] Hardware name: Micro-Star International Co., Ltd. MS-7A38/B450M PRO-VDH MAX (MS-7A38), BIOS B.G0 07/26/2022
[   17.217438] RIP: 0010:ttm_bo_release+0x27e/0x2d0 [ttm]
[   17.217444] Code: 48 89 43 38 48 89 43 40 48 8b 5c 24 30 48 8b b5 40 08 00 00 48 8b 6c 24 38 48 83 c4 58 e9 7a 49 f7 e0 48 89 ef e9 6c fe ff ff <0f> 0b 48 83 7b 20 00 0f 84 b7 fd ff ff 0f 0b 0f 1f 00 e9 ad fd ff
[   17.217448] RSP: 0018:ffffc9000095fbb0 EFLAGS: 00010202
[   17.217451] RAX: 0000000000000001 RBX: ffff8881052c8de0 RCX: 0000000000000000
[   17.217453] RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff8881052c8de0
[   17.217455] RBP: ffff888104a66e00 R08: ffff8881052c8de0 R09: ffff888104a7cf08
[   17.217457] R10: ffffc9000095fbe0 R11: ffffc9000095fbe8 R12: ffff8881052c8c78
[   17.217458] R13: ffff8881052c8c78 R14: dead000000000100 R15: ffff88810528b108
[   17.217460] FS:  00007f319fcbb8c0(0000) GS:ffff88881a540000(0000) knlGS:0000000000000000
[   17.217463] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   17.217464] CR2: 000055dc8b0224a0 CR3: 000000010373d000 CR4: 0000000000750ee0
[   17.217466] PKRU: 55555554
[   17.217468] Call Trace:
[   17.217470]  <TASK>
[   17.217472]  ? __warn+0x97/0x160
[   17.217476]  ? ttm_bo_release+0x27e/0x2d0 [ttm]
[   17.217481]  ? report_bug+0x1ec/0x200
[   17.217487]  ? handle_bug+0x3c/0x70
[   17.217490]  ? exc_invalid_op+0x1f/0x90
[   17.217493]  ? preempt_count_sub+0xb5/0x100
[   17.217496]  ? asm_exc_invalid_op+0x16/0x20
[   17.217500]  ? ttm_bo_release+0x27e/0x2d0 [ttm]
[   17.217505]  ? ttm_resource_move_to_lru_tail+0x1ab/0x1d0 [ttm]
[   17.217511]  radeon_bo_unref+0x1a/0x30 [radeon]
[   17.217547]  radeon_gem_object_free+0x20/0x30 [radeon]
[   17.217579]  radeon_fbdev_fb_destroy+0x57/0x90 [radeon]
[   17.217616]  unregister_framebuffer+0x72/0x110
[   17.217620]  drm_client_dev_unregister+0x6d/0xe0
[   17.217623]  drm_dev_unregister+0x2e/0x90
[   17.217626]  drm_put_dev+0x26/0x90
[   17.217628]  pci_device_remove+0x44/0xc0
[   17.217631]  really_probe+0x257/0x340
[   17.217635]  __driver_probe_device+0x73/0x120
[   17.217638]  driver_probe_device+0x2c/0xb0
[   17.217641]  __driver_attach+0xa0/0x150
[   17.217643]  ? __pfx___driver_attach+0x10/0x10
[   17.217646]  bus_for_each_dev+0x67/0xa0
[   17.217649]  bus_add_driver+0x10e/0x210
[   17.217651]  driver_register+0x5c/0x120
[   17.217653]  ? __pfx_radeon_module_init+0x10/0x10 [radeon]
[   17.217681]  do_one_initcall+0x44/0x220
[   17.217684]  ? kmalloc_trace+0x37/0xc0
[   17.217688]  do_init_module+0x64/0x240
[   17.217691]  __do_sys_finit_module+0xb2/0x100
[   17.217694]  do_syscall_64+0x3b/0x90
[   17.217697]  entry_SYSCALL_64_after_hwframe+0x72/0xdc
[   17.217700] RIP: 0033:0x7f319feaa5a9
[   17.217702] Code: 08 89 e8 5b 5d c3 66 2e 0f 1f 84 00 00 00 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 27 08 0d 00 f7 d8 64 89 01 48
[   17.217706] RSP: 002b:00007ffc6bf3e7f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[   17.217709] RAX: ffffffffffffffda RBX: 00005607204f3170 RCX: 00007f319feaa5a9
[   17.217710] RDX: 0000000000000000 RSI: 00007f31a002eefd RDI: 0000000000000018
[   17.217712] RBP: 00007f31a002eefd R08: 0000000000000000 R09: 00005607204f1860
[   17.217714] R10: 0000000000000018 R11: 0000000000000246 R12: 0000000000020000
[   17.217716] R13: 0000000000000000 R14: 0000560720522450 R15: 0000560720255899
[   17.217718]  </TASK>
[   17.217719] ---[ end trace 0000000000000000 ]---

The buffer object backing the fbdev emulation got pinned twice: by the
fb_probe helper radeon_fbdev_create_pinned_object() and the modesetting
code when the framebuffer got displayed. It only got unpinned once by
the fbdev helper radeon_fbdev_destroy_pinned_object(). Hence TTM's BO-
release function complains about the pin counter. Forcing the outputs
off also undoes the modesettings pin increment.

Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Reported-by: Borislav Petkov <bp@alien8.de>
Closes: https://lore.kernel.org/dri-devel/20230603174814.GCZHt83pN+wNjf63sC@fat_crate.local/
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Fixes: e317a69fe891 ("drm/radeon: Implement client-based fbdev emulation")
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: "Pan, Xinhui" <Xinhui.Pan@amd.com>
Cc: amd-gfx@lists.freedesktop.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_fbdev.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/radeon/radeon_fbdev.c b/drivers/gpu/drm/radeon/radeon_fbdev.c
index 28212c2d6c986..ab9c1abbac979 100644
--- a/drivers/gpu/drm/radeon/radeon_fbdev.c
+++ b/drivers/gpu/drm/radeon/radeon_fbdev.c
@@ -304,6 +304,7 @@ static void radeon_fbdev_client_unregister(struct drm_client_dev *client)
 
 	if (fb_helper->info) {
 		vga_switcheroo_client_fb_set(rdev->pdev, NULL);
+		drm_helper_force_disable_all(dev);
 		drm_fb_helper_unregister_info(fb_helper);
 	} else {
 		drm_client_release(&fb_helper->client);
-- 
GitLab


From 188d3f80fc6d8451ab5e570becd6a7b2d3033023 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonjiang@amd.com>
Date: Tue, 6 Jun 2023 17:18:52 -0400
Subject: [PATCH 1690/1791] drm/amdgpu: vcn_4_0 set instance 0 init sched score
 to 1

Only vcn0 can process AV1 codecx. In order to use both vcn0 and
vcn1 in h264/265 transcode to AV1 cases, set vcn0 sched score to 1
at initialization time.

Signed-off-by: Sonny Jiang <sonjiang@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 8d371faaa2b3e..b48bb52124884 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -129,7 +129,11 @@ static int vcn_v4_0_sw_init(void *handle)
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
 
-		atomic_set(&adev->vcn.inst[i].sched_score, 0);
+		/* Init instance 0 sched_score to 1, so it's scheduled after other instances */
+		if (i == 0)
+			atomic_set(&adev->vcn.inst[i].sched_score, 1);
+		else
+			atomic_set(&adev->vcn.inst[i].sched_score, 0);
 
 		/* VCN UNIFIED TRAP */
 		r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[i],
-- 
GitLab


From 8d8ffe3740b6de4a8a84817cd85195c533ed52b9 Mon Sep 17 00:00:00 2001
From: Shiwu Zhang <shiwu.zhang@amd.com>
Date: Wed, 7 Jun 2023 17:14:08 +0800
Subject: [PATCH 1691/1791] drm/amdgpu: expose num_hops and num_links xgmi info
 through dev attr

Add these two dev attrs for xgmi info details which is helpful for
developers checking the xgmi topology by catting the sys file directly.

Take 4 cards with xgmi connection as an example, get the num_hops for each
device or node through xmig_hive_info dir like,
cat /sys/bus/pci/devices/0000:41:00.0/xgmi_hive_info/node1/num_hops
will return "00 41 41 41" where "00" stands for the hops to node1 itself
and "41" is the hops in hex format to every other node in the same hive.
There are node1/node2/node3/node4 representing 4 cards in the hive.

The same for num_links dev attr.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Acked-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 46 ++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 85ee1af963dde..03dc59cbe8aac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -325,6 +325,36 @@ static ssize_t amdgpu_xgmi_show_device_id(struct device *dev,
 
 }
 
+static ssize_t amdgpu_xgmi_show_num_hops(struct device *dev,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
+	int i;
+
+	for (i = 0; i < top->num_nodes; i++)
+		sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_hops);
+
+	return sysfs_emit(buf, "%s\n", buf);
+}
+
+static ssize_t amdgpu_xgmi_show_num_links(struct device *dev,
+					struct device_attribute *attr,
+					char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info;
+	int i;
+
+	for (i = 0; i < top->num_nodes; i++)
+		sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_links);
+
+	return sysfs_emit(buf, "%s\n", buf);
+}
+
 #define AMDGPU_XGMI_SET_FICAA(o)	((o) | 0x456801)
 static ssize_t amdgpu_xgmi_show_error(struct device *dev,
 				      struct device_attribute *attr,
@@ -361,6 +391,8 @@ static ssize_t amdgpu_xgmi_show_error(struct device *dev,
 
 static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id, NULL);
 static DEVICE_ATTR(xgmi_error, S_IRUGO, amdgpu_xgmi_show_error, NULL);
+static DEVICE_ATTR(xgmi_num_hops, S_IRUGO, amdgpu_xgmi_show_num_hops, NULL);
+static DEVICE_ATTR(xgmi_num_links, S_IRUGO, amdgpu_xgmi_show_num_links, NULL);
 
 static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 					 struct amdgpu_hive_info *hive)
@@ -380,6 +412,15 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 	if (ret)
 		pr_err("failed to create xgmi_error\n");
 
+	/* Create xgmi num hops file */
+	ret = device_create_file(adev->dev, &dev_attr_xgmi_num_hops);
+	if (ret)
+		pr_err("failed to create xgmi_num_hops\n");
+
+	/* Create xgmi num links file */
+	ret = device_create_file(adev->dev, &dev_attr_xgmi_num_links);
+	if (ret)
+		pr_err("failed to create xgmi_num_links\n");
 
 	/* Create sysfs link to hive info folder on the first device */
 	if (hive->kobj.parent != (&adev->dev->kobj)) {
@@ -407,6 +448,9 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev,
 
 remove_file:
 	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
+	device_remove_file(adev->dev, &dev_attr_xgmi_error);
+	device_remove_file(adev->dev, &dev_attr_xgmi_num_hops);
+	device_remove_file(adev->dev, &dev_attr_xgmi_num_links);
 
 success:
 	return ret;
@@ -420,6 +464,8 @@ static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev,
 
 	device_remove_file(adev->dev, &dev_attr_xgmi_device_id);
 	device_remove_file(adev->dev, &dev_attr_xgmi_error);
+	device_remove_file(adev->dev, &dev_attr_xgmi_num_hops);
+	device_remove_file(adev->dev, &dev_attr_xgmi_num_links);
 
 	if (hive->kobj.parent != (&adev->dev->kobj))
 		sysfs_remove_link(&adev->dev->kobj,"xgmi_hive_info");
-- 
GitLab


From 59eddd4e215afc05f0610ebabfa05d4b099e5c13 Mon Sep 17 00:00:00 2001
From: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Date: Fri, 9 Jun 2023 06:18:41 -0700
Subject: [PATCH 1692/1791] Revert "drm/amdgpu: remove TOPDOWN flags when
 allocating VRAM in large bar system"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit c105518679b6e87232874ffc989ec403bee59664.

This patch disables the TOPDOWN flag for APU and few dGPU cards
which has the VRAM size equal to the BAR size.

When we enable the TOPDOWN flag, we get the free blocks at
the highest available memory region and we don't split the
lower order blocks. This change is required to keep off
the fragmentation related issues particularly in ASIC
which has VRAM space <= 500MiB

Hence, we are reverting this patch.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2270
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index f76649e523a0d..54cd21c06a44f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -150,7 +150,7 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 
 		if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED)
 			places[c].lpfn = min_not_zero(places[c].lpfn, visible_pfn);
-		else if (adev->gmc.real_vram_size != adev->gmc.visible_vram_size)
+		else
 			places[c].flags |= TTM_PL_FLAG_TOPDOWN;
 
 		if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)
-- 
GitLab


From 7a0e005c7957931689a327b2a4e7333a19f13f95 Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Thu, 25 May 2023 08:37:40 -0400
Subject: [PATCH 1693/1791] drm/amd/display: edp do not add non-edid timings

[Why] most edp support only timings from edid. applying
non-edid timings, especially those timings out of edp
bandwidth, may damage edp.

[How] do not add non-edid timings for edp.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Reviewed-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index b6bef202b6bbb..2f3f579efe309 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7256,7 +7256,13 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector)
 				drm_add_modes_noedid(connector, 1920, 1080);
 	} else {
 		amdgpu_dm_connector_ddc_get_modes(connector, edid);
-		amdgpu_dm_connector_add_common_modes(encoder, connector);
+		/* most eDP supports only timings from its edid,
+		 * usually only detailed timings are available
+		 * from eDP edid. timings which are not from edid
+		 * may damage eDP
+		 */
+		if (connector->connector_type != DRM_MODE_CONNECTOR_eDP)
+			amdgpu_dm_connector_add_common_modes(encoder, connector);
 		amdgpu_dm_connector_add_freesync_modes(connector, edid);
 	}
 	amdgpu_dm_fbc_init(connector);
-- 
GitLab


From ee83c930974d4afb5fec3db638a8341b5d1cd3fa Mon Sep 17 00:00:00 2001
From: Hersen Wu <hersenxs.wu@amd.com>
Date: Thu, 25 May 2023 09:00:29 -0400
Subject: [PATCH 1694/1791] drm/amd/display: add debugfs for
 allow_edp_hotplug_detection

[Why] within dc_link_detect, edp edid is read only for the first time
and saved. edid will not be read after the first time read. to run edp
edid read test, need read edp edid for each dc_link_detect. dc->config
flag allow_edp_hotplug_detection could be used for edp edid test.

[How] add debugfs for dc->config.allow_edp_hotplug_detection

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Hersen Wu <hersenxs.wu@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 32 +++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index caf13b2e8cb64..5ea3284b2b77f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -2760,6 +2760,32 @@ static int psr_read_residency(void *data, u64 *val)
 	return 0;
 }
 
+/* read allow_edp_hotplug_detection */
+static int allow_edp_hotplug_detection_get(void *data, u64 *val)
+{
+	struct amdgpu_dm_connector *aconnector = data;
+	struct drm_connector *connector = &aconnector->base;
+	struct drm_device *dev = connector->dev;
+	struct amdgpu_device *adev = drm_to_adev(dev);
+
+	*val = adev->dm.dc->config.allow_edp_hotplug_detection;
+
+	return 0;
+}
+
+/* set allow_edp_hotplug_detection */
+static int allow_edp_hotplug_detection_set(void *data, u64 val)
+{
+	struct amdgpu_dm_connector *aconnector = data;
+	struct drm_connector *connector = &aconnector->base;
+	struct drm_device *dev = connector->dev;
+	struct amdgpu_device *adev = drm_to_adev(dev);
+
+	adev->dm.dc->config.allow_edp_hotplug_detection = (uint32_t) val;
+
+	return 0;
+}
+
 /*
  * Set dmcub trace event IRQ enable or disable.
  * Usage to enable dmcub trace event IRQ: echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_dmcub_trace_event_en
@@ -2798,6 +2824,10 @@ DEFINE_DEBUGFS_ATTRIBUTE(psr_fops, psr_get, NULL, "%llu\n");
 DEFINE_DEBUGFS_ATTRIBUTE(psr_residency_fops, psr_read_residency, NULL,
 			 "%llu\n");
 
+DEFINE_DEBUGFS_ATTRIBUTE(allow_edp_hotplug_detection_fops,
+			allow_edp_hotplug_detection_get,
+			allow_edp_hotplug_detection_set, "%llu\n");
+
 DEFINE_SHOW_ATTRIBUTE(current_backlight);
 DEFINE_SHOW_ATTRIBUTE(target_backlight);
 
@@ -2968,6 +2998,8 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector)
 				    &target_backlight_fops);
 		debugfs_create_file("ilr_setting", 0644, dir, connector,
 					&edp_ilr_debugfs_fops);
+		debugfs_create_file("allow_edp_hotplug_detection", 0644, dir, connector,
+					&allow_edp_hotplug_detection_fops);
 	}
 
 	for (i = 0; i < ARRAY_SIZE(connector_debugfs_entries); i++) {
-- 
GitLab


From 3d8de40fd639c8be24e4aa557a98e20e1d09bdc3 Mon Sep 17 00:00:00 2001
From: Austin Zheng <austin.zheng@amd.com>
Date: Thu, 25 May 2023 14:01:01 -0400
Subject: [PATCH 1695/1791] drm/amd/display: Add DP2 Metrics

Why:
Log DP2 data to be used for debugging purposes

How:
Check the reported link rate of the DP connection and
translate it to the DP version.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h          | 6 ++++--
 drivers/gpu/drm/amd/display/dc/dc_dp_types.h | 2 +-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 360dd83b1a7a0..7cf3e95100435 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1,5 +1,5 @@
 /*
- * Copyright 2012-14 Advanced Micro Devices, Inc.
+ * Copyright 2012-2023 Advanced Micro Devices, Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -60,7 +60,9 @@ struct dc_versions {
 };
 
 enum dp_protocol_version {
-	DP_VERSION_1_4,
+	DP_VERSION_1_4 = 0,
+	DP_VERSION_2_1,
+	DP_VERSION_UNKNOWN,
 };
 
 enum dc_plane_type {
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 4a7f6497dc5a4..55139d7bf422c 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -61,7 +61,7 @@ enum dc_link_rate {
 	 */
 	LINK_RATE_UHBR10 = 1000,	// UHBR10 - 10.0 Gbps/Lane
 	LINK_RATE_UHBR13_5 = 1350,	// UHBR13.5 - 13.5 Gbps/Lane
-	LINK_RATE_UHBR20 = 2000,	// UHBR10 - 20.0 Gbps/Lane
+	LINK_RATE_UHBR20 = 2000,	// UHBR20 - 20.0 Gbps/Lane
 };
 
 enum dc_link_spread {
-- 
GitLab


From be3a432a9d382a09e02c3359e4f6fcd991fe7d2a Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <jerry.zuo@amd.com>
Date: Tue, 16 May 2023 15:36:45 -0400
Subject: [PATCH 1696/1791] drm/amd/display: Add Error Code for Dml Validation
 Failure

Any invalid mode from hw perspective should be given corresponding
error code, otherwise it leads to confusing warning message
"[drm] Mode Validation Warning: Validation OK failed validation."

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/dml/dcn314/display_mode_vba_314.c      | 59 +++++++++++++++++++
 .../amd/display/dc/dml/display_mode_enums.h   |  8 +++
 2 files changed, 67 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 1532a7e0ed6c5..9010c47476e92 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -5557,6 +5557,65 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_
 			}
 		}
 	}
+	for (i = v->soc.num_states; i >= 0; i--) {
+		for (j = 0; j < 2; j++) {
+			enum dm_validation_status status = DML_VALIDATION_OK;
+
+			if (!v->ScaleRatioAndTapsSupport) {
+				status = DML_FAIL_SCALE_RATIO_TAP;
+			} else if (!v->SourceFormatPixelAndScanSupport) {
+				status = DML_FAIL_SOURCE_PIXEL_FORMAT;
+			} else if (!v->ViewportSizeSupport[i][j]) {
+				status = DML_FAIL_VIEWPORT_SIZE;
+			} else if (P2IWith420) {
+				status = DML_FAIL_P2I_WITH_420;
+			} else if (DSCOnlyIfNecessaryWithBPP) {
+				status = DML_FAIL_DSC_ONLY_IF_NECESSARY_WITH_BPP;
+			} else if (DSC422NativeNotSupported) {
+				status = DML_FAIL_NOT_DSC422_NATIVE;
+			} else if (!v->ODMCombine4To1SupportCheckOK[i]) {
+				status = DML_FAIL_ODM_COMBINE4TO1;
+			} else if (v->NotEnoughDSCUnits[i]) {
+				status = DML_FAIL_NOT_ENOUGH_DSC;
+			} else if (!v->ROBSupport[i][j]) {
+				status = DML_FAIL_REORDERING_BUFFER;
+			} else if (!v->DISPCLK_DPPCLK_Support[i][j]) {
+				status = DML_FAIL_DISPCLK_DPPCLK;
+			} else if (!v->TotalAvailablePipesSupport[i][j]) {
+				status = DML_FAIL_TOTAL_AVAILABLE_PIPES;
+			} else if (!EnoughWritebackUnits) {
+				status = DML_FAIL_ENOUGH_WRITEBACK_UNITS;
+			} else if (!v->WritebackLatencySupport) {
+				status = DML_FAIL_WRITEBACK_LATENCY;
+			} else if (!v->WritebackScaleRatioAndTapsSupport) {
+				status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP;
+			} else if (!v->CursorSupport) {
+				status = DML_FAIL_CURSOR_SUPPORT;
+			} else if (!v->PitchSupport) {
+				status = DML_FAIL_PITCH_SUPPORT;
+			} else if (ViewportExceedsSurface) {
+				status = DML_FAIL_VIEWPORT_EXCEEDS_SURFACE;
+			} else if (!v->PrefetchSupported[i][j]) {
+				status = DML_FAIL_PREFETCH_SUPPORT;
+			} else if (!v->DynamicMetadataSupported[i][j]) {
+				status = DML_FAIL_DYNAMIC_METADATA;
+			} else if (!v->TotalVerticalActiveBandwidthSupport[i][j]) {
+				status = DML_FAIL_TOTAL_V_ACTIVE_BW;
+			} else if (!v->VRatioInPrefetchSupported[i][j]) {
+				status = DML_FAIL_V_RATIO_PREFETCH;
+			} else if (!v->PTEBufferSizeNotExceeded[i][j]) {
+				status = DML_FAIL_PTE_BUFFER_SIZE;
+			} else if (v->NonsupportedDSCInputBPC) {
+				status = DML_FAIL_DSC_INPUT_BPC;
+			} else if ((v->HostVMEnable
+					&& !v->ImmediateFlipSupportedForState[i][j])) {
+				status = DML_FAIL_HOST_VM_IMMEDIATE_FLIP;
+			} else if (FMTBufferExceeded) {
+				status = DML_FAIL_FMT_BUFFER_EXCEEDED;
+			}
+			mode_lib->vba.ValidationStatus[i] = status;
+		}
+	}
 
 	{
 		unsigned int MaximumMPCCombine = 0;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
index 0bffae95f3a29..d5831a34f5a19 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h
@@ -190,6 +190,14 @@ enum dm_validation_status {
 	DML_FAIL_DSC_INPUT_BPC,
 	DML_FAIL_PREFETCH_SUPPORT,
 	DML_FAIL_V_RATIO_PREFETCH,
+	DML_FAIL_P2I_WITH_420,
+	DML_FAIL_DSC_ONLY_IF_NECESSARY_WITH_BPP,
+	DML_FAIL_NOT_DSC422_NATIVE,
+	DML_FAIL_ODM_COMBINE4TO1,
+	DML_FAIL_ENOUGH_WRITEBACK_UNITS,
+	DML_FAIL_VIEWPORT_EXCEEDS_SURFACE,
+	DML_FAIL_DYNAMIC_METADATA,
+	DML_FAIL_FMT_BUFFER_EXCEEDED,
 };
 
 enum writeback_config {
-- 
GitLab


From 5b4d93eaf6b44903f4c71b4f404f65317abd8d9c Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Sun, 28 May 2023 19:48:27 -0400
Subject: [PATCH 1697/1791] drm/amd/display: Promote DAL to 3.2.238

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 7cf3e95100435..5c906b007e4d5 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.237"
+#define DC_VER "3.2.238"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From f308116676566b555ec3bab4c3f9eb20c1c9a5cb Mon Sep 17 00:00:00 2001
From: Tom Chung <chiahsuan.chung@amd.com>
Date: Mon, 29 May 2023 18:00:09 +0800
Subject: [PATCH 1698/1791] drm/amd/display: fix the system hang while disable
 PSR

[Why]
When the PSR enabled. If you try to adjust the timing parameters,
it may cause system hang. Because the timing mismatch with the
DMCUB settings.

[How]
Disable the PSR before adjusting timing parameters.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 2f3f579efe309..c6a1014e0e89c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8279,6 +8279,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		if (acrtc_state->abm_level != dm_old_crtc_state->abm_level)
 			bundle->stream_update.abm_level = &acrtc_state->abm_level;
 
+		mutex_lock(&dm->dc_lock);
+		if ((acrtc_state->update_type > UPDATE_TYPE_FAST) &&
+				acrtc_state->stream->link->psr_settings.psr_allow_active)
+			amdgpu_dm_psr_disable(acrtc_state->stream);
+		mutex_unlock(&dm->dc_lock);
+
 		/*
 		 * If FreeSync state on the stream has changed then we need to
 		 * re-adjust the min/max bounds now that DC doesn't handle this
@@ -8292,10 +8298,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags);
 		}
 		mutex_lock(&dm->dc_lock);
-		if ((acrtc_state->update_type > UPDATE_TYPE_FAST) &&
-				acrtc_state->stream->link->psr_settings.psr_allow_active)
-			amdgpu_dm_psr_disable(acrtc_state->stream);
-
 		update_planes_and_stream_adapter(dm->dc,
 					 acrtc_state->update_type,
 					 planes_count,
-- 
GitLab


From e8c49e9eead8620c7dd3c64a1f3bb44682325710 Mon Sep 17 00:00:00 2001
From: Tom Chung <chiahsuan.chung@amd.com>
Date: Mon, 29 May 2023 14:27:11 +0800
Subject: [PATCH 1699/1791] drm/amd/display: Fix disbling PSR slow response
 issue

[Why]
dmub_psr_get_state() return an invalid PSR state while disable
the PSR because convert_psr_state() doesn't recognize the state
that return from DMCUB.

[How]
Add a PSR state to make the dmub_psr_get_state() return a
correct PSR state.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_types.h     | 1 +
 drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 6b4731b5e975f..0ce7728a5a4be 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -597,6 +597,7 @@ enum dc_psr_state {
 	PSR_STATE4b_FULL_FRAME,
 	PSR_STATE4c_FULL_FRAME,
 	PSR_STATE4_FULL_FRAME_POWERUP,
+	PSR_STATE4_FULL_FRAME_HW_LOCK,
 	PSR_STATE5,
 	PSR_STATE5a,
 	PSR_STATE5b,
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
index 4000a834592c0..0f24b6fbd2201 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@@ -87,6 +87,8 @@ static enum dc_psr_state convert_psr_state(uint32_t raw_state)
 		state = PSR_STATE4c_FULL_FRAME;
 	else if (raw_state == 0x4E)
 		state = PSR_STATE4_FULL_FRAME_POWERUP;
+	else if (raw_state == 0x4F)
+		state = PSR_STATE4_FULL_FRAME_HW_LOCK;
 	else if (raw_state == 0x60)
 		state = PSR_STATE_HWLOCK_MGR;
 	else if (raw_state == 0x61)
-- 
GitLab


From 1c982c9ffefd00120f2293bfd15fec5af475dc28 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Tue, 30 May 2023 10:07:39 -0400
Subject: [PATCH 1700/1791] drm/amd/display: SubVP high refresh only if all
 displays >= 120hz

[Description]
- SubVP high refresh should only be enabled if all displays
  are >= 120hz. We do not want to accidentally enables configs
  such as 60hz[SubVP] + 120hz[SubVP]
- Ensure that the SubVP config generation code does not produce
  configs such as 60hz[SubVP] + 120hz[SubVP]
- Also add admissibility checks to ensure these configs do not
  pass as valid configs

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Dillon Varone <Dillon.Varone@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn32/dcn32_resource.h |   4 +
 .../display/dc/dcn32/dcn32_resource_helpers.c | 101 ++++++++++
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 178 +++++++++++-------
 3 files changed, 217 insertions(+), 66 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
index 2f34f01b3ea13..81e4431708295 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h
@@ -169,6 +169,10 @@ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *conte
 
 bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int width, unsigned int height);
 
+bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context);
+
+bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int vlevel);
+
 /* definitions for run time init of reg offsets */
 
 /* CLK SRC */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index 1d13fd7972122..578070e7d44b4 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -660,3 +660,104 @@ bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int widt
 
 	return is_native_scaling;
 }
+
+/**
+ * ************************************************************************************************
+ * dcn32_subvp_drr_admissable: Determine if SubVP + DRR config is admissible
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ *
+ * SubVP + DRR is admissible under the following conditions:
+ * - Config must have 2 displays (i.e., 2 non-phantom master pipes)
+ * - One display is SubVP
+ * - Other display must have Freesync enabled
+ *
+ * @return: True if admissible, false otherwise
+ *
+ * ************************************************************************************************
+ */
+bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
+{
+	bool result = false;
+	uint32_t i;
+	uint8_t subvp_count = 0;
+	uint8_t non_subvp_pipes = 0;
+	bool drr_pipe_found = false;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (pipe->plane_state && !pipe->top_pipe) {
+			if (pipe->stream->mall_stream_config.type == SUBVP_MAIN)
+				subvp_count++;
+			if (pipe->stream->mall_stream_config.type == SUBVP_NONE) {
+				non_subvp_pipes++;
+				if (pipe->stream->ignore_msa_timing_param &&
+						(pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) {
+					drr_pipe_found = true;
+				}
+			}
+		}
+	}
+
+	if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found)
+		result = true;
+
+	return result;
+}
+
+/**
+ * ************************************************************************************************
+ * dcn32_subvp_vblank_admissable: Determine if SubVP + Vblank config is admissible
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ *
+ * SubVP + Vblank is admissible under the following conditions:
+ * - Config must have 2 displays (i.e., 2 non-phantom master pipes)
+ * - One display is SubVP
+ * - Other display must not have Freesync capability
+ * - DML must have output DRAM clock change support as SubVP + Vblank
+ *
+ * @return: True if admissible, false otherwise
+ *
+ * ************************************************************************************************
+ */
+bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int vlevel)
+{
+	bool result = false;
+	uint32_t i;
+	uint8_t subvp_count = 0;
+	uint8_t non_subvp_pipes = 0;
+	bool drr_pipe_found = false;
+	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (pipe->plane_state && !pipe->top_pipe) {
+			if (pipe->stream->mall_stream_config.type == SUBVP_MAIN)
+				subvp_count++;
+			if (pipe->stream->mall_stream_config.type == SUBVP_NONE) {
+				non_subvp_pipes++;
+				if (pipe->stream->ignore_msa_timing_param &&
+						(pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) {
+					drr_pipe_found = true;
+				}
+			}
+		}
+	}
+
+	if (subvp_count == 1 && non_subvp_pipes == 1 && !drr_pipe_found &&
+			vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp)
+		result = true;
+
+	return result;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index fa3678342abb7..166123be4adcc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -679,7 +679,6 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 	unsigned int max_frame_time = 0;
 	bool valid_assignment_found = false;
 	unsigned int free_pipes = dcn32_get_num_free_pipes(dc, context);
-	bool current_assignment_freesync = false;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
@@ -720,19 +719,10 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc,
 				struct dc_stream_state *stream = pipe->stream;
 				unsigned int frame_us = (stream->timing.v_total * stream->timing.h_total /
 						(double)(stream->timing.pix_clk_100hz * 100)) * 1000000;
-				if (frame_us > max_frame_time && !stream->ignore_msa_timing_param) {
+				if (frame_us > max_frame_time) {
 					*index = i;
 					max_frame_time = frame_us;
 					valid_assignment_found = true;
-					current_assignment_freesync = false;
-				/* For the 2-Freesync display case, still choose the one with the
-			     * longest frame time
-			     */
-				} else if (stream->ignore_msa_timing_param && (!valid_assignment_found ||
-						(current_assignment_freesync && frame_us > max_frame_time))) {
-					*index = i;
-					valid_assignment_found = true;
-					current_assignment_freesync = true;
 				}
 			}
 		}
@@ -878,11 +868,12 @@ static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context)
  *
  * Return: True if the SubVP + DRR config is schedulable, false otherwise
  */
-static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struct pipe_ctx *drr_pipe)
+static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context)
 {
 	bool schedulable = false;
 	uint32_t i;
 	struct pipe_ctx *pipe = NULL;
+	struct pipe_ctx *drr_pipe = NULL;
 	struct dc_crtc_timing *main_timing = NULL;
 	struct dc_crtc_timing *phantom_timing = NULL;
 	struct dc_crtc_timing *drr_timing = NULL;
@@ -908,6 +899,19 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struc
 			break;
 	}
 
+	// Find the DRR pipe
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		drr_pipe = &context->res_ctx.pipe_ctx[i];
+
+		// We check for master pipe only
+		if (!drr_pipe->stream || !drr_pipe->plane_state || drr_pipe->top_pipe || drr_pipe->prev_odm_pipe)
+			continue;
+
+		if (drr_pipe->stream->mall_stream_config.type == SUBVP_NONE && drr_pipe->stream->ignore_msa_timing_param &&
+				(drr_pipe->stream->allow_freesync || drr_pipe->stream->vrr_active_variable))
+			break;
+	}
+
 	main_timing = &pipe->stream->timing;
 	phantom_timing = &pipe->stream->mall_stream_config.paired_stream->timing;
 	drr_timing = &drr_pipe->stream->timing;
@@ -993,13 +997,7 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context)
 		if (!subvp_pipe && pipe->stream->mall_stream_config.type == SUBVP_MAIN)
 			subvp_pipe = pipe;
 	}
-	// Use ignore_msa_timing_param and VRR active, or Freesync flag to identify as DRR On
-	if (found && context->res_ctx.pipe_ctx[vblank_index].stream->ignore_msa_timing_param &&
-			(context->res_ctx.pipe_ctx[vblank_index].stream->allow_freesync ||
-			context->res_ctx.pipe_ctx[vblank_index].stream->vrr_active_variable)) {
-		// SUBVP + DRR case -- only allowed if run through DRR validation path
-		schedulable = false;
-	} else if (found) {
+	if (found) {
 		main_timing = &subvp_pipe->stream->timing;
 		phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing;
 		vblank_timing = &context->res_ctx.pipe_ctx[vblank_index].stream->timing;
@@ -1028,6 +1026,56 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context)
 	return schedulable;
 }
 
+/**
+ * ************************************************************************************************
+ * subvp_subvp_admissable: Determine if subvp + subvp config is admissible
+ *
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ *
+ * SubVP + SubVP is admissible under the following conditions:
+ * - All SubVP pipes are < 120Hz OR
+ * - All SubVP pipes are >= 120hz
+ *
+ * @return: True if admissible, false otherwise
+ *
+ * ************************************************************************************************
+ */
+static bool subvp_subvp_admissable(struct dc *dc,
+				struct dc_state *context)
+{
+	bool result = false;
+	uint32_t i;
+	uint8_t subvp_count = 0;
+	uint32_t min_refresh = subvp_high_refresh_list.min_refresh, max_refresh = 0;
+	uint32_t refresh_rate = 0;
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe->stream)
+			continue;
+
+		if (pipe->plane_state && !pipe->top_pipe &&
+				pipe->stream->mall_stream_config.type == SUBVP_MAIN) {
+			refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 +
+					pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1)
+					/ (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total);
+			if (refresh_rate < min_refresh)
+				min_refresh = refresh_rate;
+			if (refresh_rate > max_refresh)
+				max_refresh = refresh_rate;
+			subvp_count++;
+		}
+	}
+
+	if (subvp_count == 2 && ((min_refresh < 120 && max_refresh < 120) ||
+			(min_refresh >= 120 && max_refresh >= 120)))
+		result = true;
+
+	return result;
+}
+
 /**
  * subvp_validate_static_schedulability - Check which SubVP case is calculated
  * and handle static analysis based on the case.
@@ -1046,11 +1094,12 @@ static bool subvp_validate_static_schedulability(struct dc *dc,
 				struct dc_state *context,
 				int vlevel)
 {
-	bool schedulable = true;	// true by default for single display case
+	bool schedulable = false;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
 	uint32_t i, pipe_idx;
 	uint8_t subvp_count = 0;
 	uint8_t vactive_count = 0;
+	uint8_t non_subvp_pipes = 0;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
@@ -1058,14 +1107,18 @@ static bool subvp_validate_static_schedulability(struct dc *dc,
 		if (!pipe->stream)
 			continue;
 
-		if (pipe->plane_state && !pipe->top_pipe &&
-				pipe->stream->mall_stream_config.type == SUBVP_MAIN)
-			subvp_count++;
+		if (pipe->plane_state && !pipe->top_pipe) {
+			if (pipe->stream->mall_stream_config.type == SUBVP_MAIN)
+				subvp_count++;
+			if (pipe->stream->mall_stream_config.type == SUBVP_NONE) {
+				non_subvp_pipes++;
+			}
+		}
 
 		// Count how many planes that aren't SubVP/phantom are capable of VACTIVE
 		// switching (SubVP + VACTIVE unsupported). In situations where we force
 		// SubVP for a VACTIVE plane, we don't want to increment the vactive_count.
-		if (vba->ActiveDRAMClockChangeLatencyMargin[vba->pipe_plane[pipe_idx]] > 0 &&
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vlevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 &&
 		    pipe->stream->mall_stream_config.type == SUBVP_NONE) {
 			vactive_count++;
 		}
@@ -1074,13 +1127,14 @@ static bool subvp_validate_static_schedulability(struct dc *dc,
 
 	if (subvp_count == 2) {
 		// Static schedulability check for SubVP + SubVP case
-		schedulable = subvp_subvp_schedulable(dc, context);
-	} else if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp) {
-		// Static schedulability check for SubVP + VBLANK case. Also handle the case where
-		// DML outputs SubVP + VBLANK + VACTIVE (DML will report as SubVP + VBLANK)
-		if (vactive_count > 0)
-			schedulable = false;
-		else
+		schedulable = subvp_subvp_admissable(dc, context) && subvp_subvp_schedulable(dc, context);
+	} else if (subvp_count == 1 && non_subvp_pipes == 0) {
+		// Single SubVP configs will be supported by default as long as it's suppported by DML
+		schedulable = true;
+	} else if (subvp_count == 1 && non_subvp_pipes == 1) {
+		if (dcn32_subvp_drr_admissable(dc, context))
+			schedulable = subvp_drr_schedulable(dc, context);
+		else if (dcn32_subvp_vblank_admissable(dc, context, vlevel))
 			schedulable = subvp_vblank_schedulable(dc, context);
 	} else if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vactive_w_mall_sub_vp &&
 			vactive_count > 0) {
@@ -1104,10 +1158,6 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 	unsigned int dc_pipe_idx = 0;
 	int i = 0;
 	bool found_supported_config = false;
-	struct pipe_ctx *pipe = NULL;
-	uint32_t non_subvp_pipes = 0;
-	bool drr_pipe_found = false;
-	uint32_t drr_pipe_index = 0;
 
 	dc_assert_fp_enabled();
 
@@ -1197,31 +1247,12 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 				}
 			}
 
-			if (*vlevel < context->bw_ctx.dml.soc.num_states &&
-			    vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] != dm_dram_clock_change_unsupported
-			    && subvp_validate_static_schedulability(dc, context, *vlevel)) {
+			if (*vlevel < context->bw_ctx.dml.soc.num_states
+			    && subvp_validate_static_schedulability(dc, context, *vlevel))
 				found_supported_config = true;
-			} else if (*vlevel < context->bw_ctx.dml.soc.num_states) {
-				/* Case where 1 SubVP is added, and DML reports MCLK unsupported or DRR is allowed.
-				 * This handles the case for SubVP + DRR, where the DRR display does not support MCLK
-				 * switch at it's native refresh rate / timing, or DRR is allowed for the non-subvp
-				 * display.
-				 */
-				for (i = 0; i < dc->res_pool->pipe_count; i++) {
-					pipe = &context->res_ctx.pipe_ctx[i];
-					if (pipe->stream && pipe->plane_state && !pipe->top_pipe &&
-					    pipe->stream->mall_stream_config.type == SUBVP_NONE) {
-						non_subvp_pipes++;
-						// Use ignore_msa_timing_param flag to identify as DRR
-						if (pipe->stream->ignore_msa_timing_param && pipe->stream->allow_freesync) {
-							drr_pipe_found = true;
-							drr_pipe_index = i;
-						}
-					}
-				}
-				// If there is only 1 remaining non SubVP pipe that is DRR, check static
-				// schedulability for SubVP + DRR.
-				if (non_subvp_pipes == 1 && drr_pipe_found) {
+			if (found_supported_config) {
+				// For SubVP + DRR cases, we can force the lowest vlevel that supports the mode
+				if (dcn32_subvp_drr_admissable(dc, context) && subvp_drr_schedulable(dc, context)) {
 					/* find lowest vlevel that supports the config */
 					for (i = *vlevel; i >= 0; i--) {
 						if (vba->ModeSupport[i][vba->maxMpcComb]) {
@@ -1230,9 +1261,6 @@ static void dcn32_full_validate_bw_helper(struct dc *dc,
 							break;
 						}
 					}
-
-					found_supported_config = subvp_drr_schedulable(dc, context,
-										       &context->res_ctx.pipe_ctx[drr_pipe_index]);
 				}
 			}
 		}
@@ -2882,16 +2910,34 @@ bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context
 {
 	bool allow = false;
 	uint32_t refresh_rate = 0;
-	uint32_t min_refresh = subvp_high_refresh_list.min_refresh;
-	uint32_t max_refresh = subvp_high_refresh_list.max_refresh;
+	uint32_t subvp_min_refresh = subvp_high_refresh_list.min_refresh;
+	uint32_t subvp_max_refresh = subvp_high_refresh_list.max_refresh;
+	uint32_t min_refresh = subvp_max_refresh;
 	uint32_t i;
 
-	if (!dc->debug.disable_subvp_high_refresh && pipe->stream &&
+	/* Only allow SubVP on high refresh displays if all connected displays
+	 * are considered "high refresh" (i.e. >= 120hz). We do not want to
+	 * allow combinations such as 120hz (SubVP) + 60hz (SubVP).
+	 */
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+
+		if (!pipe_ctx->stream)
+			continue;
+		refresh_rate = (pipe_ctx->stream->timing.pix_clk_100hz * 100 +
+				pipe_ctx->stream->timing.v_total * pipe_ctx->stream->timing.h_total - 1)
+						/ (double)(pipe_ctx->stream->timing.v_total * pipe_ctx->stream->timing.h_total);
+
+		if (refresh_rate < min_refresh)
+			min_refresh = refresh_rate;
+	}
+
+	if (!dc->debug.disable_subvp_high_refresh && min_refresh >= subvp_min_refresh && pipe->stream &&
 			pipe->plane_state && !(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed)) {
 		refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 +
 						pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1)
 						/ (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total);
-		if (refresh_rate >= min_refresh && refresh_rate <= max_refresh) {
+		if (refresh_rate >= subvp_min_refresh && refresh_rate <= subvp_max_refresh) {
 			for (i = 0; i < SUBVP_HIGH_REFRESH_LIST_LEN; i++) {
 				uint32_t width = subvp_high_refresh_list.res[i].width;
 				uint32_t height = subvp_high_refresh_list.res[i].height;
-- 
GitLab


From 4cc1cebe08bff0d2b75f16aa65ec61360e09a647 Mon Sep 17 00:00:00 2001
From: Daniel Miess <daniel.miess@amd.com>
Date: Tue, 23 May 2023 14:21:59 -0400
Subject: [PATCH 1701/1791] drm/amd/display: Re-enable DPP/HUBP Power Gating

[Why & How]
Bugs preventing DPP/HUBP power gating have been addressed
so this should be reenabled on dcn314 for sufficient
hardware rev versions

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn314/dcn314_resource.c   | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
index 7957ad4d6a34a..a840b008d6603 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c
@@ -874,8 +874,8 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.force_abm_enable = false,
 	.timing_trace = false,
 	.clock_trace = true,
-	.disable_dpp_power_gate = true,
-	.disable_hubp_power_gate = true,
+	.disable_dpp_power_gate = false,
+	.disable_hubp_power_gate = false,
 	.disable_pplib_clock_request = false,
 	.pipe_split_policy = MPC_SPLIT_DYNAMIC,
 	.force_single_disp_pipe_split = false,
@@ -1883,6 +1883,13 @@ static bool dcn314_resource_construct(
 	/* Use pipe context based otg sync logic */
 	dc->config.use_pipe_ctx_sync_logic = true;
 
+	/* Disable pipe power gating when unsupported */
+	if (ctx->asic_id.hw_internal_rev == 0x01 ||
+			ctx->asic_id.hw_internal_rev == 0x80) {
+		dc->debug.disable_dpp_power_gate = true;
+		dc->debug.disable_hubp_power_gate = true;
+	}
+
 	/* read VBIOS LTTPR caps */
 	{
 		if (ctx->dc_bios->funcs->get_lttpr_caps) {
-- 
GitLab


From 299004271cbf0315da327c4bd67aec3e7041cb32 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Wed, 31 May 2023 12:35:57 -0400
Subject: [PATCH 1702/1791] drm/amd/display: Re-enable SubVP high refresh

Re-enable SubVP high refresh now that it is fixed
for displays with high refresh rates.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Saaem Rizvi <SyedSaaem.Rizvi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 2e6b39fe26132..8c9e15952a490 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -728,7 +728,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_margin_us = 2000, // 2000us
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
-	.disable_subvp_high_refresh = true,
+	.disable_subvp_high_refresh = false,
 	.disable_dp_plus_plus_wa = true,
 	.fpo_vactive_min_active_margin_us = 200,
 	.fpo_vactive_max_blank_us = 1000,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index bbcd3579fea61..ee07ee3401713 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -727,7 +727,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_margin_us = 2000, // 2000us
 	.disable_fpo_vactive = false,
 	.disable_boot_optimizations = false,
-	.disable_subvp_high_refresh = true,
+	.disable_subvp_high_refresh = false,
 	.fpo_vactive_min_active_margin_us = 200,
 	.fpo_vactive_max_blank_us = 1000,
 };
-- 
GitLab


From d62088ba314ecf098871874898ed760347d1fbd8 Mon Sep 17 00:00:00 2001
From: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Date: Tue, 30 May 2023 13:21:10 -0400
Subject: [PATCH 1703/1791] drm/amd/display: Do not disable phantom pipes in
 driver

[Why and How]
We should not disable phantom pipes in this sequence, as this should be
controlled by FW. Furthermore, the previous programming sequence would
have enabled the phantom pipe in driver as well, causing corruption.
This change should avoid this from occuring.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com>
Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index 00f32ffe00796..e5bd76c6b1d34 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -1211,7 +1211,8 @@ void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_
 		if (pipe->top_pipe || pipe->prev_odm_pipe)
 			continue;
 
-		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))
+			&& pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) {
 			pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg);
 			reset_sync_context_for_pipe(dc, context, i);
 			otg_disabled[i] = true;
-- 
GitLab


From 196754951fc8187c64806d0807c467d6f435d0c5 Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Wed, 31 May 2023 20:00:54 -0400
Subject: [PATCH 1704/1791] drm/amd/display: Block SubVP + DRR if the DRR is
 PSR capable

[Description]
PSR implementation in FW has inline polling which can poll for up
to 1ms. This will interfere with SubVP because SubVP is timing
sensitive and can't tolerate up to 1ms worth of delay before
handling vertical or VLINE interrupts. Therefore block SubVP + DRR
cases if DRR is PSR capable

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Saaem Rizvi <SyedSaaem.Rizvi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dcn32/dcn32_resource_helpers.c  | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index 578070e7d44b4..7eec39576e2cc 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -672,6 +672,7 @@ bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int widt
  * - Config must have 2 displays (i.e., 2 non-phantom master pipes)
  * - One display is SubVP
  * - Other display must have Freesync enabled
+ * - The potential DRR display must not be PSR capable
  *
  * @return: True if admissible, false otherwise
  *
@@ -684,6 +685,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
 	uint8_t subvp_count = 0;
 	uint8_t non_subvp_pipes = 0;
 	bool drr_pipe_found = false;
+	bool drr_psr_capable = false;
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
@@ -696,6 +698,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
 				subvp_count++;
 			if (pipe->stream->mall_stream_config.type == SUBVP_NONE) {
 				non_subvp_pipes++;
+				drr_psr_capable = (drr_psr_capable || dcn32_is_psr_capable(pipe));
 				if (pipe->stream->ignore_msa_timing_param &&
 						(pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) {
 					drr_pipe_found = true;
@@ -704,7 +707,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
 		}
 	}
 
-	if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found)
+	if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found && !drr_psr_capable)
 		result = true;
 
 	return result;
@@ -722,6 +725,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
  * - One display is SubVP
  * - Other display must not have Freesync capability
  * - DML must have output DRAM clock change support as SubVP + Vblank
+ * - The potential vblank display must not be PSR capable
  *
  * @return: True if admissible, false otherwise
  *
@@ -735,6 +739,7 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int
 	uint8_t non_subvp_pipes = 0;
 	bool drr_pipe_found = false;
 	struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
+	bool vblank_psr_capable = false;
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
@@ -747,6 +752,7 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int
 				subvp_count++;
 			if (pipe->stream->mall_stream_config.type == SUBVP_NONE) {
 				non_subvp_pipes++;
+				vblank_psr_capable = (vblank_psr_capable || dcn32_is_psr_capable(pipe));
 				if (pipe->stream->ignore_msa_timing_param &&
 						(pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) {
 					drr_pipe_found = true;
@@ -755,7 +761,7 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int
 		}
 	}
 
-	if (subvp_count == 1 && non_subvp_pipes == 1 && !drr_pipe_found &&
+	if (subvp_count == 1 && non_subvp_pipes == 1 && !drr_pipe_found && !vblank_psr_capable &&
 			vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp)
 		result = true;
 
-- 
GitLab


From ec7282bd2688c6c741c79f8696a68c6c0403cf2d Mon Sep 17 00:00:00 2001
From: Artem Grishin <artem.grishin@amd.com>
Date: Thu, 1 Jun 2023 12:52:16 -0400
Subject: [PATCH 1705/1791] drm/amd/display: Bug fix in
 dcn315_populate_dml_pipes_from_context

[Why]
When iterating over all pipes in the loop, the CRB allocation algorithm
may potentially skip some of the pipes. Previously, the current pipe
index didn't get updated in this case, causing incorrect outcome.

[How]
Increment the pipe index when skipping over a pipe in the loop.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Artem Grishin <artem.grishin@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
index 0cc853964781c..f1153941907e6 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c
@@ -1742,6 +1742,7 @@ static int dcn315_populate_dml_pipes_from_context(
 			/* Do not use asymetric crb if not enough for pstate support */
 			if (remaining_det_segs < 0) {
 				pipes[pipe_cnt].pipe.src.det_size_override = 0;
+				pipe_cnt++;
 				continue;
 			}
 
-- 
GitLab


From da55037afde24d74a1a3f26e4d314f897f3432b4 Mon Sep 17 00:00:00 2001
From: Austin Zheng <austin.zheng@amd.com>
Date: Thu, 1 Jun 2023 15:16:45 -0400
Subject: [PATCH 1706/1791] drm/amd/display: Limit Minimum FreeSync Refresh
 Rate

Why:
Some EDIDs report a minimum refresh rate lower than what HW can support

How:
Add a check to calculate minimum supported refresh rate with current timing
and use that as the minimum if a lower one is passed in

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                   |  1 +
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c |  1 +
 .../gpu/drm/amd/display/dc/dcn302/dcn302_resource.c   |  1 +
 .../gpu/drm/amd/display/dc/dcn303/dcn303_resource.c   |  1 +
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c |  1 +
 .../gpu/drm/amd/display/dc/dcn321/dcn321_resource.c   |  1 +
 .../gpu/drm/amd/display/modules/freesync/freesync.c   | 11 +++++++++--
 7 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 5c906b007e4d5..a239dcd8e9fb0 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -266,6 +266,7 @@ struct dc_caps {
 	uint16_t subvp_pstate_allow_width_us;
 	uint16_t subvp_vertical_int_margin_us;
 	bool seamless_odm;
+	uint32_t max_v_total;
 	uint8_t subvp_drr_vblank_start_margin_us;
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
index f4ee4b3df5962..1a0284a068b2a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c
@@ -2328,6 +2328,7 @@ static bool dcn30_resource_construct(
 	dc->caps.color.mpc.ocsc = 1;
 
 	dc->caps.dp_hdmi21_pcon_support = true;
+	dc->caps.max_v_total = (1 << 15) - 1;
 
 	/* read VBIOS LTTPR caps */
 	{
diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
index 93f42132c900f..7dc065ea247a1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c
@@ -1227,6 +1227,7 @@ static bool dcn302_resource_construct(
 	dc->caps.force_dp_tps4_for_cp2520 = true;
 	dc->caps.extended_aux_timeout_support = true;
 	dc->caps.dmcub_support = true;
+	dc->caps.max_v_total = (1 << 15) - 1;
 
 	/* Color pipeline capabilities */
 	dc->caps.color.dpp.dcn_arch = 1;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
index f35514188a5cd..6d9761395288d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c
@@ -1152,6 +1152,7 @@ static bool dcn303_resource_construct(
 	dc->caps.force_dp_tps4_for_cp2520 = true;
 	dc->caps.extended_aux_timeout_support = true;
 	dc->caps.dmcub_support = true;
+	dc->caps.max_v_total = (1 << 15) - 1;
 
 	/* Color pipeline capabilities */
 	dc->caps.color.dpp.dcn_arch = 1;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
index 8c9e15952a490..19f134caa8adb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c
@@ -2175,6 +2175,7 @@ static bool dcn32_resource_construct(
 	dc->caps.extended_aux_timeout_support = true;
 	dc->caps.dmcub_support = true;
 	dc->caps.seamless_odm = true;
+	dc->caps.max_v_total = (1 << 15) - 1;
 
 	/* Color pipeline capabilities */
 	dc->caps.color.dpp.dcn_arch = 1;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
index ee07ee3401713..ea204742ad354 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c
@@ -1718,6 +1718,7 @@ static bool dcn321_resource_construct(
 	dc->caps.edp_dsc_support = true;
 	dc->caps.extended_aux_timeout_support = true;
 	dc->caps.dmcub_support = true;
+	dc->caps.max_v_total = (1 << 15) - 1;
 
 	/* Color pipeline capabilities */
 	dc->caps.color.dpp.dcn_arch = 1;
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 5798c0eafa1fb..dbd60811f95da 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -1,5 +1,5 @@
 /*
- * Copyright 2016 Advanced Micro Devices, Inc.
+ * Copyright 2016-2023 Advanced Micro Devices, Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -989,6 +989,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync,
 	unsigned int refresh_range = 0;
 	unsigned long long min_refresh_in_uhz = 0;
 	unsigned long long max_refresh_in_uhz = 0;
+	unsigned long long min_hardware_refresh_in_uhz = 0;
 
 	if (mod_freesync == NULL)
 		return;
@@ -999,7 +1000,13 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync,
 	nominal_field_rate_in_uhz =
 			mod_freesync_calc_nominal_field_rate(stream);
 
-	min_refresh_in_uhz = in_config->min_refresh_in_uhz;
+	if (stream->ctx->dc->caps.max_v_total != 0 && stream->timing.h_total != 0) {
+		min_hardware_refresh_in_uhz = div64_u64((stream->timing.pix_clk_100hz * 100000000ULL),
+			(stream->timing.h_total * stream->ctx->dc->caps.max_v_total));
+	}
+	/* Limit minimum refresh rate to what can be supported by hardware */
+	min_refresh_in_uhz = min_hardware_refresh_in_uhz > in_config->min_refresh_in_uhz ?
+		min_hardware_refresh_in_uhz : in_config->min_refresh_in_uhz;
 	max_refresh_in_uhz = in_config->max_refresh_in_uhz;
 
 	/* Full range may be larger than current video timing, so cap at nominal */
-- 
GitLab


From bbe4418f22b9b20cf2654ca710e344955380e62c Mon Sep 17 00:00:00 2001
From: Alvin Lee <alvin.lee2@amd.com>
Date: Thu, 1 Jun 2023 20:39:35 -0400
Subject: [PATCH 1707/1791] drm/amd/display: Include CSC updates in new fast
 update path

[Description]
- Missed color / CSC updates in fast update path
  which caused Custom Color to break.
- Add color / CSC updates to new fast update path
  to fix custom color

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/core/dc_hw_sequencer.c | 70 +++++++++++++++++++
 .../gpu/drm/amd/display/dc/inc/hw_sequencer.h | 32 +++++++++
 2 files changed, 102 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index 182c42c63bc5e..d7d00fefaab9b 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -562,6 +562,29 @@ void hwss_build_fast_sequence(struct dc *dc,
 				(*num_steps)++;
 			}
 
+			if (current_mpc_pipe->stream->update_flags.bits.out_csc) {
+				block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.mpc = dc->res_pool->mpc;
+				block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.mpcc_id = current_mpc_pipe->plane_res.hubp->inst;
+				block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.power_on = true;
+				block_sequence[*num_steps].func = MPC_POWER_ON_MPC_MEM_PWR;
+				(*num_steps)++;
+
+				if (current_mpc_pipe->stream->csc_color_matrix.enable_adjustment == true) {
+					block_sequence[*num_steps].params.set_output_csc_params.mpc = dc->res_pool->mpc;
+					block_sequence[*num_steps].params.set_output_csc_params.opp_id = current_mpc_pipe->stream_res.opp->inst;
+					block_sequence[*num_steps].params.set_output_csc_params.regval = current_mpc_pipe->stream->csc_color_matrix.matrix;
+					block_sequence[*num_steps].params.set_output_csc_params.ocsc_mode = MPC_OUTPUT_CSC_COEF_A;
+					block_sequence[*num_steps].func = MPC_SET_OUTPUT_CSC;
+					(*num_steps)++;
+				} else {
+					block_sequence[*num_steps].params.set_ocsc_default_params.mpc = dc->res_pool->mpc;
+					block_sequence[*num_steps].params.set_ocsc_default_params.opp_id = current_mpc_pipe->stream_res.opp->inst;
+					block_sequence[*num_steps].params.set_ocsc_default_params.color_space = current_mpc_pipe->stream->output_color_space;
+					block_sequence[*num_steps].params.set_ocsc_default_params.ocsc_mode = MPC_OUTPUT_CSC_COEF_A;
+					block_sequence[*num_steps].func = MPC_SET_OCSC_DEFAULT;
+					(*num_steps)++;
+				}
+			}
 			current_mpc_pipe = current_mpc_pipe->bottom_pipe;
 		}
 		current_pipe = current_pipe->next_odm_pipe;
@@ -661,6 +684,15 @@ void hwss_execute_sequence(struct dc *dc,
 					params->update_visual_confirm_params.pipe_ctx,
 					params->update_visual_confirm_params.mpcc_id);
 			break;
+		case MPC_POWER_ON_MPC_MEM_PWR:
+			hwss_power_on_mpc_mem_pwr(params);
+			break;
+		case MPC_SET_OUTPUT_CSC:
+			hwss_set_output_csc(params);
+			break;
+		case MPC_SET_OCSC_DEFAULT:
+			hwss_set_ocsc_default(params);
+			break;
 		case DMUB_SEND_DMCUB_CMD:
 			hwss_send_dmcub_cmd(params);
 			break;
@@ -718,6 +750,44 @@ void hwss_program_bias_and_scale(union block_sequence_params *params)
 		dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params);
 }
 
+void hwss_power_on_mpc_mem_pwr(union block_sequence_params *params)
+{
+	struct mpc *mpc = params->power_on_mpc_mem_pwr_params.mpc;
+	int mpcc_id = params->power_on_mpc_mem_pwr_params.mpcc_id;
+	bool power_on = params->power_on_mpc_mem_pwr_params.power_on;
+
+	if (mpc->funcs->power_on_mpc_mem_pwr)
+		mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, power_on);
+}
+
+void hwss_set_output_csc(union block_sequence_params *params)
+{
+	struct mpc *mpc = params->set_output_csc_params.mpc;
+	int opp_id = params->set_output_csc_params.opp_id;
+	const uint16_t *matrix = params->set_output_csc_params.regval;
+	enum mpc_output_csc_mode ocsc_mode = params->set_output_csc_params.ocsc_mode;
+
+	if (mpc->funcs->set_output_csc != NULL)
+		mpc->funcs->set_output_csc(mpc,
+				opp_id,
+				matrix,
+				ocsc_mode);
+}
+
+void hwss_set_ocsc_default(union block_sequence_params *params)
+{
+	struct mpc *mpc = params->set_ocsc_default_params.mpc;
+	int opp_id = params->set_ocsc_default_params.opp_id;
+	enum dc_color_space colorspace = params->set_ocsc_default_params.color_space;
+	enum mpc_output_csc_mode ocsc_mode = params->set_ocsc_default_params.ocsc_mode;
+
+	if (mpc->funcs->set_ocsc_default != NULL)
+		mpc->funcs->set_ocsc_default(mpc,
+				opp_id,
+				colorspace,
+				ocsc_mode);
+}
+
 void get_mclk_switch_visual_confirm_color(
 		struct dc *dc,
 		struct dc_state *context,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index cc0a3a992f7bb..02ff99f7bec2b 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -114,6 +114,26 @@ struct update_visual_confirm_params {
 	int mpcc_id;
 };
 
+struct power_on_mpc_mem_pwr_params {
+	struct mpc *mpc;
+	int mpcc_id;
+	bool power_on;
+};
+
+struct set_output_csc_params {
+	struct mpc *mpc;
+	int opp_id;
+	const uint16_t *regval;
+	enum mpc_output_csc_mode ocsc_mode;
+};
+
+struct set_ocsc_default_params {
+	struct mpc *mpc;
+	int opp_id;
+	enum dc_color_space color_space;
+	enum mpc_output_csc_mode ocsc_mode;
+};
+
 union block_sequence_params {
 	struct update_plane_addr_params update_plane_addr_params;
 	struct subvp_pipe_control_lock_fast_params subvp_pipe_control_lock_fast_params;
@@ -128,6 +148,9 @@ union block_sequence_params {
 	struct program_bias_and_scale_params program_bias_and_scale_params;
 	struct set_output_transfer_func_params set_output_transfer_func_params;
 	struct update_visual_confirm_params update_visual_confirm_params;
+	struct power_on_mpc_mem_pwr_params power_on_mpc_mem_pwr_params;
+	struct set_output_csc_params set_output_csc_params;
+	struct set_ocsc_default_params set_ocsc_default_params;
 };
 
 enum block_sequence_func {
@@ -144,6 +167,9 @@ enum block_sequence_func {
 	DPP_PROGRAM_BIAS_AND_SCALE,
 	DPP_SET_OUTPUT_TRANSFER_FUNC,
 	MPC_UPDATE_VISUAL_CONFIRM,
+	MPC_POWER_ON_MPC_MEM_PWR,
+	MPC_SET_OUTPUT_CSC,
+	MPC_SET_OCSC_DEFAULT,
 };
 
 struct block_sequence {
@@ -439,4 +465,10 @@ void hwss_setup_dpp(union block_sequence_params *params);
 
 void hwss_program_bias_and_scale(union block_sequence_params *params);
 
+void hwss_power_on_mpc_mem_pwr(union block_sequence_params *params);
+
+void hwss_set_output_csc(union block_sequence_params *params);
+
+void hwss_set_ocsc_default(union block_sequence_params *params);
+
 #endif /* __DC_HW_SEQUENCER_H__ */
-- 
GitLab


From 0e69ef6ea82e8eece7d2b2b45a0da9670eaaefff Mon Sep 17 00:00:00 2001
From: Peichen Huang <peichen.huang@amd.com>
Date: Wed, 31 May 2023 13:36:14 +0800
Subject: [PATCH 1708/1791] drm/amd/display: limit DPIA link rate to HBR3

[Why]
DPIA doesn't support UHBR, driver should not enable UHBR
for dp tunneling

[How]
limit DPIA link rate to HBR3

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Peichen Huang <peichen.huang@amd.com>
Reviewed-by: Mustapha Ghaddar <Mustapha.Ghaddar@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/link/link_detection.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index 17904de4f155b..8041b8369e45f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -984,6 +984,11 @@ static bool detect_link_and_local_sink(struct dc_link *link,
 					(link->dpcd_caps.dongle_type !=
 							DISPLAY_DONGLE_DP_HDMI_CONVERTER))
 				converter_disable_audio = true;
+
+			/* limited link rate to HBR3 for DPIA until we implement USB4 V2 */
+			if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA &&
+					link->reported_link_cap.link_rate > LINK_RATE_HIGH3)
+				link->reported_link_cap.link_rate = LINK_RATE_HIGH3;
 			break;
 		}
 
-- 
GitLab


From f4bc8a43069c6268a49f064fdbf85ead5cc2bf04 Mon Sep 17 00:00:00 2001
From: Samson Tam <samson.tam@amd.com>
Date: Tue, 30 May 2023 17:35:58 -0400
Subject: [PATCH 1709/1791] Revert "drm/amd/display: reallocate DET for dual
 displays with high pixel rate ratio"

Revert commit 5f3401eeb064 ("drm/amd/display: reallocate DET for dual displays with high pixel rate ratio")

[Why]
Previously had modified DET allocation algorithm to allocate less DET
segments for lower pixel rate display and more DET segments for higher
pixel rate display.  But noticed it is causing underflow when higher
pixel rate display is not displaying at higher mode

[How]
Roll back change

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Samson Tam <samson.tam@amd.com>
Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dcn32/dcn32_resource_helpers.c | 51 ++-----------------
 1 file changed, 5 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index 7eec39576e2cc..4882c3684b82d 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -255,8 +255,6 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
 	return psr_capable;
 }
 
-#define DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER 7
-
 /**
  * dcn32_determine_det_override(): Determine DET allocation for each pipe
  *
@@ -267,6 +265,7 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
  * If there is a plane that's driven by more than 1 pipe (i.e. pipe split), then the
  * number of DET for that given plane will be split among the pipes driving that plane.
  *
+ *
  * High level algorithm:
  * 1. Split total DET among number of streams
  * 2. For each stream, split DET among the planes
@@ -274,21 +273,9 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
  *    among those pipes.
  * 4. Assign the DET override to the DML pipes.
  *
- * Special cases:
- *
- * For two displays that have a large difference in pixel rate, we may experience
- *  underflow on the larger display when we divide the DET equally. For this, we
- *  will implement a modified algorithm to assign more DET to larger display.
- *
- * 1. Calculate difference in pixel rates ( multiplier ) between two displays
- * 2. If the multiplier exceeds DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER, then
- *    implement the modified DET override algorithm.
- * 3. Assign smaller DET size for lower pixel display and higher DET size for
- *    higher pixel display
- *
- * @dc: Current DC state
- * @context: New DC state to be programmed
- * @pipes: Array of DML pipes
+ * @param [in]: dc: Current DC state
+ * @param [in]: context: New DC state to be programmed
+ * @param [in]: pipes: Array of DML pipes
  *
  * Return: void
  */
@@ -303,31 +290,10 @@ void dcn32_determine_det_override(struct dc *dc,
 	struct dc_plane_state *current_plane = NULL;
 	uint8_t stream_count = 0;
 
-	int phy_pix_clk_mult, lower_mode_stream_index;
-	int phy_pix_clk[MAX_PIPES] = {0};
-	bool use_new_det_override_algorithm = false;
-
 	for (i = 0; i < context->stream_count; i++) {
 		/* Don't count SubVP streams for DET allocation */
-		if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM) {
-			phy_pix_clk[i] = context->streams[i]->phy_pix_clk;
+		if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM)
 			stream_count++;
-		}
-	}
-
-	/* Check for special case with two displays, one with much higher pixel rate */
-	if (stream_count == 2) {
-		ASSERT((phy_pix_clk[0] > 0) && (phy_pix_clk[1] > 0));
-		if (phy_pix_clk[0] < phy_pix_clk[1]) {
-			lower_mode_stream_index = 0;
-			phy_pix_clk_mult = phy_pix_clk[1] / phy_pix_clk[0];
-		} else {
-			lower_mode_stream_index = 1;
-			phy_pix_clk_mult = phy_pix_clk[0] / phy_pix_clk[1];
-		}
-
-		if (phy_pix_clk_mult >= DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER)
-			use_new_det_override_algorithm = true;
 	}
 
 	if (stream_count > 0) {
@@ -336,13 +302,6 @@ void dcn32_determine_det_override(struct dc *dc,
 			if (context->streams[i]->mall_stream_config.type == SUBVP_PHANTOM)
 				continue;
 
-			if (use_new_det_override_algorithm) {
-				if (i == lower_mode_stream_index)
-					stream_segments = 4;
-				else
-					stream_segments = 14;
-			}
-
 			if (context->stream_status[i].plane_count > 0)
 				plane_segments = stream_segments / context->stream_status[i].plane_count;
 			else
-- 
GitLab


From a2c7356f526dba1aa5f49ba17c822e46dcf7d6f6 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Thu, 1 Jun 2023 16:09:32 -0400
Subject: [PATCH 1710/1791] drm/amd/display: fix pixel rate update sequence

The k1/k2 pixel rate dividers in dccg should only be updated on stream enable
and do not actually depend on whether odm combine is active.

This removes an on flip update of these and fixes the calculate function
to ignore odm status for dp steams.

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Reviewed-by: Ariel Bernstein <Eric.Bernstein@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c    | 11 -----------
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c  |  9 ++-------
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h  |  2 +-
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c    |  8 ++------
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h    |  2 +-
 .../gpu/drm/amd/display/dc/inc/hw_sequencer_private.h |  2 +-
 6 files changed, 7 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
index 5934b1d70e485..b0a13eb8318cd 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c
@@ -1741,17 +1741,6 @@ static void dcn20_program_pipe(
 
 		if (hws->funcs.setup_vupdate_interrupt)
 			hws->funcs.setup_vupdate_interrupt(dc, pipe_ctx);
-
-		if (hws->funcs.calculate_dccg_k1_k2_values && dc->res_pool->dccg->funcs->set_pixel_rate_div) {
-			unsigned int k1_div, k2_div;
-
-			hws->funcs.calculate_dccg_k1_k2_values(pipe_ctx, &k1_div, &k2_div);
-
-			dc->res_pool->dccg->funcs->set_pixel_rate_div(
-				dc->res_pool->dccg,
-				pipe_ctx->stream_res.tg->inst,
-				k1_div, k2_div);
-		}
 	}
 
 	if (pipe_ctx->update_flags.bits.odm)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
index ce7e6f20b31fc..7a43f88685006 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c
@@ -337,14 +337,13 @@ void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable)
 		REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0);
 }
 
-unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div)
+void dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	unsigned int odm_combine_factor = 0;
 	bool two_pix_per_container = false;
 
 	two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing);
-	odm_combine_factor = get_odm_config(pipe_ctx, NULL);
+	get_odm_config(pipe_ctx, NULL);
 
 	if (stream->ctx->dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) {
 		*k1_div = PIXEL_RATE_DIV_BY_1;
@@ -362,15 +361,11 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 		} else {
 			*k1_div = PIXEL_RATE_DIV_BY_1;
 			*k2_div = PIXEL_RATE_DIV_BY_4;
-			if (odm_combine_factor == 2)
-				*k2_div = PIXEL_RATE_DIV_BY_2;
 		}
 	}
 
 	if ((*k1_div == PIXEL_RATE_DIV_NA) && (*k2_div == PIXEL_RATE_DIV_NA))
 		ASSERT(false);
-
-	return odm_combine_factor;
 }
 
 void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
index 559d71002e8a4..96035c75e0dfe 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h
@@ -37,7 +37,7 @@ void dcn314_dsc_pg_control(struct dce_hwseq *hws, unsigned int dsc_inst, bool po
 
 void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable);
 
-unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div);
+void dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div);
 
 void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
index e5bd76c6b1d34..c586468872e2a 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c
@@ -1141,10 +1141,9 @@ void dcn32_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *
 	}
 }
 
-unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div)
+void dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div)
 {
 	struct dc_stream_state *stream = pipe_ctx->stream;
-	unsigned int odm_combine_factor = 0;
 	bool two_pix_per_container = false;
 
 	// For phantom pipes, use the same programming as the main pipes
@@ -1152,7 +1151,6 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 		stream = pipe_ctx->stream->mall_stream_config.paired_stream;
 	}
 	two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing);
-	odm_combine_factor = get_odm_config(pipe_ctx, NULL);
 
 	if (stream->ctx->dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) {
 		*k1_div = PIXEL_RATE_DIV_BY_1;
@@ -1170,15 +1168,13 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign
 		} else {
 			*k1_div = PIXEL_RATE_DIV_BY_1;
 			*k2_div = PIXEL_RATE_DIV_BY_4;
-			if ((odm_combine_factor == 2) || dcn32_is_dp_dig_pixel_rate_div_policy(pipe_ctx))
+			if (dcn32_is_dp_dig_pixel_rate_div_policy(pipe_ctx))
 				*k2_div = PIXEL_RATE_DIV_BY_2;
 		}
 	}
 
 	if ((*k1_div == PIXEL_RATE_DIV_NA) && (*k2_div == PIXEL_RATE_DIV_NA))
 		ASSERT(false);
-
-	return odm_combine_factor;
 }
 
 void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
index 2d2628f31bed7..bf9bffabe0c03 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h
@@ -71,7 +71,7 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context);
 
 void dcn32_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx);
 
-unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div);
+void dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div);
 
 void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
index 4ca4192c1e127..a151865a3a20d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h
@@ -156,7 +156,7 @@ struct hwseq_private_funcs {
 	void (*program_mall_pipe_config)(struct dc *dc, struct dc_state *context);
 	void (*update_force_pstate)(struct dc *dc, struct dc_state *context);
 	void (*update_mall_sel)(struct dc *dc, struct dc_state *context);
-	unsigned int (*calculate_dccg_k1_k2_values)(struct pipe_ctx *pipe_ctx,
+	void (*calculate_dccg_k1_k2_values)(struct pipe_ctx *pipe_ctx,
 			unsigned int *k1_div,
 			unsigned int *k2_div);
 	void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx);
-- 
GitLab


From ddafc678913c4573d52f075af7d82152d431f322 Mon Sep 17 00:00:00 2001
From: Aric Cyr <aric.cyr@amd.com>
Date: Mon, 5 Jun 2023 00:28:51 -0400
Subject: [PATCH 1711/1791] drm/amd/display: 3.2.239

This version brings along following fixes:
- Improvement on eDP
- PSR bug fixes
- SubVP bug fixes
- Improvements on pipe handling to address potential issues
- Freesync bug fix
- DPIA bug fix
- Fix multi-display issues

Acked-by: Stylon Wang <stylon.wang@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index a239dcd8e9fb0..26d05e2250888 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -45,7 +45,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.238"
+#define DC_VER "3.2.239"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
-- 
GitLab


From e22821e6302780e2acaef7438cab828f68dde9d1 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Fri, 9 Jun 2023 13:07:56 +0800
Subject: [PATCH 1712/1791] drm/amd/pm: fix vclk setting failed for SMU v13.0.5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PMFW use the left-shifted 16 bits argument to set the VCLK
DPM frequency for SMU v13.0.5.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
index 7c3ac535f68a5..725c791ad3fc3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
@@ -831,6 +831,8 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu,
 							uint32_t max)
 {
 	enum smu_message_type msg_set_min, msg_set_max;
+	uint32_t min_clk = min;
+	uint32_t max_clk = max;
 	int ret = 0;
 
 	if (!smu_v13_0_5_clk_dpm_is_enabled(smu, clk_type))
@@ -851,11 +853,16 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu,
 		return -EINVAL;
 	}
 
-	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL);
+	if (clk_type == SMU_VCLK) {
+		min_clk = min << SMU_13_VCLK_SHIFT;
+		max_clk = max << SMU_13_VCLK_SHIFT;
+	}
+
+	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL);
 	if (ret)
 		goto out;
 
-	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL);
+	ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max_clk, NULL);
 	if (ret)
 		goto out;
 
-- 
GitLab


From d50dc746ff72b9c48812dac3344fa87fbde940a3 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Fri, 9 Jun 2023 14:06:43 +0530
Subject: [PATCH 1713/1791] drm/amdgpu: Fix memcpy() in
 sienna_cichlid_append_powerplay_table function.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following gcc with W=1:

In file included from ./include/linux/string.h:253,
                 from ./include/linux/bitmap.h:11,
                 from ./include/linux/cpumask.h:12,
                 from ./arch/x86/include/asm/cpumask.h:5,
                 from ./arch/x86/include/asm/msr.h:11,
                 from ./arch/x86/include/asm/processor.h:22,
                 from ./arch/x86/include/asm/cpufeature.h:5,
                 from ./arch/x86/include/asm/thread_info.h:53,
                 from ./include/linux/thread_info.h:60,
                 from ./arch/x86/include/asm/preempt.h:7,
                 from ./include/linux/preempt.h:78,
                 from ./include/linux/spinlock.h:56,
                 from ./include/linux/mmzone.h:8,
                 from ./include/linux/gfp.h:7,
                 from ./include/linux/firmware.h:7,
                 from drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu11/sienna_cichlid_ppt.c:26:
In function ‘fortify_memcpy_chk’,
    inlined from ‘sienna_cichlid_append_powerplay_table’ at drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu11/sienna_cichlid_ppt.c:444:2,
    inlined from ‘sienna_cichlid_setup_pptable’ at drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu11/sienna_cichlid_ppt.c:506:8,
    inlined from ‘sienna_cichlid_setup_pptable’ at drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu11/sienna_cichlid_ppt.c:494:12:
./include/linux/fortify-string.h:413:4: warning: call to ‘__read_overflow2_field’ declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Wattribute-warning]
  413 |    __read_overflow2_field(q_size_field, size);
      |    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

the compiler complains about the size calculation in the memcpy() -
"sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header)" is much
larger than what fits into table_member.

Hence, reuse 'smu_memcpy_trailing' for nv1x

Fixes: 7077b19a38240 ("drm/amd/pm: use macro to get pptable members")
Suggested-by: Evan Quan <Evan.Quan@amd.com>
Cc: Evan Quan <Evan.Quan@amd.com>
Cc: Chengming Gui <Jack.Gui@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/pm/swsmu/smu11/sienna_cichlid_ppt.c    | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index 85d53597eb07a..f7ed3e655e397 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -431,7 +431,13 @@ static int sienna_cichlid_append_powerplay_table(struct smu_context *smu)
 {
 	struct atom_smc_dpm_info_v4_9 *smc_dpm_table;
 	int index, ret;
-	I2cControllerConfig_t *table_member;
+	PPTable_beige_goby_t *ppt_beige_goby;
+	PPTable_t *ppt;
+
+	if (smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 13))
+		ppt_beige_goby = smu->smu_table.driver_pptable;
+	else
+		ppt = smu->smu_table.driver_pptable;
 
 	index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1,
 					    smc_dpm_info);
@@ -440,9 +446,13 @@ static int sienna_cichlid_append_powerplay_table(struct smu_context *smu)
 				      (uint8_t **)&smc_dpm_table);
 	if (ret)
 		return ret;
-	GET_PPTABLE_MEMBER(I2cControllers, &table_member);
-	memcpy(table_member, smc_dpm_table->I2cControllers,
-			sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header));
+
+	if (smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 13))
+		smu_memcpy_trailing(ppt_beige_goby, I2cControllers, BoardReserved,
+				    smc_dpm_table, I2cControllers);
+	else
+		smu_memcpy_trailing(ppt, I2cControllers, BoardReserved,
+				    smc_dpm_table, I2cControllers);
 
 	return 0;
 }
-- 
GitLab


From fcdb3832a4edece23c043ce97b3a1f7647bec929 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Fri, 9 Jun 2023 13:11:44 +0800
Subject: [PATCH 1714/1791] drm/amd/pm: enable vclk and dclk Pstates for SMU
 v13.0.5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add the ability to control the vclk and dclk frequency by
power_dpm_force_performance_level interface.

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c  | 29 +++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
index 725c791ad3fc3..53c508acf8951 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
@@ -983,19 +983,31 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu,
 {
 	struct amdgpu_device *adev = smu->adev;
 	uint32_t sclk_min = 0, sclk_max = 0;
+	uint32_t vclk_min = 0, vclk_max = 0;
+	uint32_t dclk_min = 0, dclk_max = 0;
 	int ret = 0;
 
 	switch (level) {
 	case AMD_DPM_FORCED_LEVEL_HIGH:
 		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max);
 		sclk_min = sclk_max;
+		vclk_min = vclk_max;
+		dclk_min = dclk_max;
 		break;
 	case AMD_DPM_FORCED_LEVEL_LOW:
 		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL);
 		sclk_max = sclk_min;
+		vclk_max = vclk_min;
+		dclk_max = dclk_min;
 		break;
 	case AMD_DPM_FORCED_LEVEL_AUTO:
 		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max);
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
 	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
@@ -1023,6 +1035,23 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu,
 		smu->gfx_actual_soft_max_freq = sclk_max;
 	}
 
+	if (vclk_min && vclk_max) {
+		ret = smu_v13_0_5_set_soft_freq_limited_range(smu,
+							      SMU_VCLK,
+							      vclk_min,
+							      vclk_max);
+		if (ret)
+			return ret;
+	}
+
+	if (dclk_min && dclk_max) {
+		ret = smu_v13_0_5_set_soft_freq_limited_range(smu,
+							      SMU_DCLK,
+							      dclk_min,
+							      dclk_max);
+		if (ret)
+			return ret;
+	}
 	return ret;
 }
 
-- 
GitLab


From 8f7bd7010dd5bca920e9d3c0c040622b2e834b57 Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Mon, 12 Jun 2023 11:42:24 -0400
Subject: [PATCH 1715/1791] drm/amdkfd: fix null queue check on debug setting
 exceptions

Null check should be done on queue struct itself and not on the
process queue list node.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_debug.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
index cd34e7aaead49..fff3ccc04fa94 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c
@@ -1097,7 +1097,7 @@ void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target,
 
 	pqm = &target->pqm;
 	list_for_each_entry(pqn, &pqm->queues, process_queue_list) {
-		if (!pqn)
+		if (!pqn->q)
 			continue;
 
 		found_mask |= pqn->q->properties.exception_status;
-- 
GitLab


From 7f80a88dd370777b86ff583f036c558c58c9f84c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 12 Jun 2023 09:54:06 -0400
Subject: [PATCH 1716/1791] Revert "drm/amd/display: fix dpms_off issue when
 disabling bios mode"

This reverts commit 58e67bb3c131da5ee14e4842b08e53f4888dce0a.

This patch was reverted, but came back again as
commit 58e67bb3c131 ("drm/amd/display: fix dpms_off issue when disabling bios mode")
Revert it again as it breaks Asus G513QY / 6800M laptops.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2259
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Zhongwei <Zhongwei.Zhang@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Mike Lothian <mike@fireburn.co.uk>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index be72e03cdf92e..5d3d61faeb28e 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -1296,7 +1296,7 @@ static void disable_vbios_mode_if_required(
 
 					if (pix_clk_100hz != requested_pix_clk_100hz) {
 						dc->link_srv->set_dpms_off(pipe);
-						pipe->stream->dpms_off = true;
+						pipe->stream->dpms_off = false;
 					}
 				}
 			}
-- 
GitLab


From e6b27cf515a7813ca2228a9aec8e61d67fb9fbf0 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 9 Jun 2023 15:40:33 -0400
Subject: [PATCH 1717/1791] drm/amd/display: don't free stolen console memory
 during suspend

Don't free the memory if we are hitting this as part of suspend.
This way we don't free any memory during suspend; see
amdgpu_bo_free_kernel().  The memory will be freed in the first
non-suspend modeset or when the driver is torn down.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2568
Acked-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c6a1014e0e89c..8964bb87e0d56 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8959,10 +8959,17 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 
 	drm_atomic_helper_cleanup_planes(dev, state);
 
-	/* return the stolen vga memory back to VRAM */
-	if (!adev->mman.keep_stolen_vga_memory)
-		amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
-	amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
+	/* Don't free the memory if we are hitting this as part of suspend.
+	 * This way we don't free any memory during suspend; see
+	 * amdgpu_bo_free_kernel().  The memory will be freed in the first
+	 * non-suspend modeset or when the driver is torn down.
+	 */
+	if (!adev->in_suspend) {
+		/* return the stolen vga memory back to VRAM */
+		if (!adev->mman.keep_stolen_vga_memory)
+			amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL);
+		amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL);
+	}
 
 	/*
 	 * Finally, drop a runtime PM reference for each newly disabled CRTC,
-- 
GitLab


From fb120e84b00ad4371c13a0f31df773fbbb16b09f Mon Sep 17 00:00:00 2001
From: Ruili Ji <ruiliji2@amd.com>
Date: Tue, 6 Jun 2023 14:06:01 +0800
Subject: [PATCH 1718/1791] drm/amdkfd: To enable traps for GC_11_0_4 and up

Flag trap_en should be enabled for trap handler.

Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index d6b15493fffdb..8a39a9e0ed5ac 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -227,7 +227,7 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q,
 	queue_input.tba_addr = qpd->tba_addr;
 	queue_input.tma_addr = qpd->tma_addr;
 	queue_input.trap_en = KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) ||
-			      KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0);
+			      KFD_GC_VERSION(q->device) > IP_VERSION(11, 0, 3);
 	queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled;
 
 	queue_type = convert_to_mes_queue_type(q->properties.type);
@@ -1807,7 +1807,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	q->properties.is_evicted = !!qpd->evicted;
 	q->properties.is_dbg_wa = qpd->pqm->process->debug_trap_enabled &&
 			KFD_GC_VERSION(q->device) >= IP_VERSION(11, 0, 0) &&
-			KFD_GC_VERSION(q->device) < IP_VERSION(12, 0, 0);
+			KFD_GC_VERSION(q->device) <= IP_VERSION(11, 0, 3);
 
 	if (qd)
 		mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr,
-- 
GitLab


From a1c23485b8ef40fbb9690fdf40f15bcb26c43e73 Mon Sep 17 00:00:00 2001
From: Ma Jun <Jun.Ma2@amd.com>
Date: Wed, 7 Jun 2023 17:19:14 +0800
Subject: [PATCH 1719/1791] drm/amdgpu: Print client id for the unregistered
 interrupt resource

Modify the debug information and print the clien id for these
interrupts as well.

Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 3481d2808ce52..5273decc5753b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -467,7 +467,8 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev,
 			handled = true;
 
 	} else {
-		DRM_DEBUG("Unhandled interrupt src_id: %d\n", src_id);
+		DRM_DEBUG("Unregistered interrupt src_id: %d of client_id:%d\n",
+			src_id, client_id);
 	}
 
 	/* Send it to amdkfd as well if it isn't already handled */
-- 
GitLab


From 47d4a680b8e2cf0502ee5a6d0191d3b7b1bdcad7 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 3 Jun 2023 20:22:13 +0530
Subject: [PATCH 1720/1791] drm/amdgpu: Add missing function parameter 'optc' &
 'enable' to kdoc in optc3_set_timing_double_buffer()

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn30/dcn30_optc.c:285: warning: Function parameter or member 'optc' not described in 'optc3_set_timing_double_buffer'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn30/dcn30_optc.c:285: warning: Function parameter or member 'enable' not described in 'optc3_set_timing_double_buffer'

Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
index c6f5f3df8061a..dfb8f62765f28 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c
@@ -279,6 +279,9 @@ static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, in
  * Sets double buffer point for V_TOTAL, H_TOTAL, VTOTAL_MIN,
  * VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers.
  *
+ * @optc: timing_generator instance.
+ * @enable: Enable DRR double buffering control if true, disable otherwise.
+ *
  * Options: any time,  start of frame, dp start of frame (range timing)
  */
 static void optc3_set_timing_double_buffer(struct timing_generator *optc, bool enable)
-- 
GitLab


From f9d9745a8603ad61937209c2431732b9abe444b7 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 5 Jun 2023 10:39:20 +0530
Subject: [PATCH 1721/1791] drm/amd/display: Correct kdoc formats in
 dcn10_opp.c

Fixes the following W=1 kernel build warning:

display/dc/dcn10/dcn10_opp.c:52: warning: Function parameter or member 'oppn10' not described in 'opp1_set_truncation'
display/dc/dcn10/dcn10_opp.c:52: warning: Function parameter or member 'params' not described in 'opp1_set_truncation'
display/dc/dcn10/dcn10_opp.c:52: warning: expecting prototype for set_truncation(). Prototype was for opp1_set_truncation() instead
display/dc/dcn10/dcn10_opp.c:161: warning: Function parameter or member 'oppn10' not described in 'opp1_set_pixel_encoding'
display/dc/dcn10/dcn10_opp.c:161: warning: Function parameter or member 'params' not described in 'opp1_set_pixel_encoding'
display/dc/dcn10/dcn10_opp.c:161: warning: expecting prototype for set_pixel_encoding(). Prototype was for opp1_set_pixel_encoding() instead
display/dc/dcn10/dcn10_opp.c:183: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 *      Set Clamping

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn10/dcn10_opp.c  | 20 +++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
index 41cec7acf51fe..0dec57679269b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c
@@ -37,14 +37,14 @@
 #define CTX \
 	oppn10->base.ctx
 
-
-/************* FORMATTER ************/
-
 /**
- *	set_truncation
+ * opp1_set_truncation():
  *	1) set truncation depth: 0 for 18 bpp or 1 for 24 bpp
  *	2) enable truncation
  *	3) HW remove 12bit FMT support for DCE11 power saving reason.
+ *
+ * @oppn10: output_pixel_processor struct instance for dcn10.
+ * @params: pointer to bit_depth_reduction_params.
  */
 static void opp1_set_truncation(
 		struct dcn10_opp *oppn10,
@@ -149,11 +149,12 @@ void opp1_program_bit_depth_reduction(
 }
 
 /**
- *	set_pixel_encoding
- *
- *	Set Pixel Encoding
+ * opp1_set_pixel_encoding():
  *		0: RGB 4:4:4 or YCbCr 4:4:4 or YOnly
  *		1: YCbCr 4:2:2
+ *
+ * @oppn10: output_pixel_processor struct instance for dcn10.
+ * @params: pointer to clamping_and_pixel_encoding_params.
  */
 static void opp1_set_pixel_encoding(
 	struct dcn10_opp *oppn10,
@@ -180,13 +181,16 @@ static void opp1_set_pixel_encoding(
 }
 
 /**
- *	Set Clamping
+ * opp1_set_clamping():
  *	1) Set clamping format based on bpc - 0 for 6bpc (No clamping)
  *		1 for 8 bpc
  *		2 for 10 bpc
  *		3 for 12 bpc
  *		7 for programable
  *	2) Enable clamp if Limited range requested
+ *
+ * @oppn10: output_pixel_processor struct instance for dcn10.
+ * @params: pointer to clamping_and_pixel_encoding_params.
  */
 static void opp1_set_clamping(
 	struct dcn10_opp *oppn10,
-- 
GitLab


From 8b42e93b349c8fa18f3a27c56f04128657f47ff0 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 12 Jun 2023 06:52:07 +0530
Subject: [PATCH 1722/1791] drm/amd/display: Correct and remove excess function
 parameter names in kdoc

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:872: warning: Excess function parameter 'drr_pipe' description in 'subvp_drr_schedulable'
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn32/dcn32_fpu.c:1030: warning: Cannot understand  * ****************************************************

Cc: Stylon Wang <stylon.wang@amd.com>
Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c   | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 166123be4adcc..e2bb2b9971f30 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -854,10 +854,9 @@ static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context)
 }
 
 /**
- * subvp_drr_schedulable - Determine if SubVP + DRR config is schedulable
+ * subvp_drr_schedulable() - Determine if SubVP + DRR config is schedulable
  * @dc: current dc state
  * @context: new dc state
- * @drr_pipe: DRR pipe_ctx for the SubVP + DRR config
  *
  * High level algorithm:
  * 1. Get timing for SubVP pipe, phantom pipe, and DRR pipe
@@ -1027,19 +1026,16 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context)
 }
 
 /**
- * ************************************************************************************************
- * subvp_subvp_admissable: Determine if subvp + subvp config is admissible
+ * subvp_subvp_admissable() - Determine if subvp + subvp config is admissible
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
  *
  * SubVP + SubVP is admissible under the following conditions:
  * - All SubVP pipes are < 120Hz OR
  * - All SubVP pipes are >= 120hz
  *
- * @return: True if admissible, false otherwise
- *
- * ************************************************************************************************
+ * Return: True if admissible, false otherwise
  */
 static bool subvp_subvp_admissable(struct dc *dc,
 				struct dc_state *context)
-- 
GitLab


From 9379c7a89f0cfbccdbe788fa189a8b29533489f6 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sat, 3 Jun 2023 19:08:17 +0530
Subject: [PATCH 1723/1791] drm/amd/display: Provide function name for
 'optc32_enable_crtc()'

Fixes the following W=1 kernel build warning(s):

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_optc.c:109: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Enable CRTC

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
index 2cffedea2df56..8abb94f60078f 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c
@@ -106,8 +106,11 @@ void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool man
 			OTG_H_TIMING_DIV_MODE_MANUAL, manual_mode ? 1 : 0);
 }
 /**
- * Enable CRTC
- * Enable CRTC - call ASIC Control Object to enable Timing generator.
+ * optc32_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator.
+ *
+ * @optc: timing_generator instance.
+ *
+ * Return: If CRTC is enabled, return true.
  */
 static bool optc32_enable_crtc(struct timing_generator *optc)
 {
-- 
GitLab


From 2e1e62c8e5e79bce38b574dd4281e6a27fd7f665 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 12 Jun 2023 09:52:33 +0530
Subject: [PATCH 1724/1791] drm/amd/display: Correct kdoc formats in
 dcn32_resource_helpers.c

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:285: warning: Function parameter or member 'dc' not described in 'dcn32_determine_det_override'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:285: warning: Function parameter or member 'context' not described in 'dcn32_determine_det_override'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:285: warning: Function parameter or member 'pipes' not described in 'dcn32_determine_det_override'
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:624: warning: Cannot understand  * *****************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn32/dcn32_resource_helpers.c:676: warning: Cannot understand  * *****************************************************************

Cc: Alvin Lee <alvin.lee2@amd.com>
Cc: Stylon Wang <stylon.wang@amd.com>
Cc: Hamza Mahfooz <hamza.mahfooz@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/dcn32/dcn32_resource_helpers.c | 29 ++++++++-----------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index 4882c3684b82d..a9c41ef0751fb 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -273,9 +273,9 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe)
  *    among those pipes.
  * 4. Assign the DET override to the DML pipes.
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- * @param [in]: pipes: Array of DML pipes
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @pipes: Array of DML pipes
  *
  * Return: void
  */
@@ -621,11 +621,10 @@ bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int widt
 }
 
 /**
- * ************************************************************************************************
- * dcn32_subvp_drr_admissable: Determine if SubVP + DRR config is admissible
+ * dcn32_subvp_drr_admissable() - Determine if SubVP + DRR config is admissible
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
  *
  * SubVP + DRR is admissible under the following conditions:
  * - Config must have 2 displays (i.e., 2 non-phantom master pipes)
@@ -633,9 +632,7 @@ bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int widt
  * - Other display must have Freesync enabled
  * - The potential DRR display must not be PSR capable
  *
- * @return: True if admissible, false otherwise
- *
- * ************************************************************************************************
+ * Return: True if admissible, false otherwise
  */
 bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
 {
@@ -673,11 +670,11 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
 }
 
 /**
- * ************************************************************************************************
- * dcn32_subvp_vblank_admissable: Determine if SubVP + Vblank config is admissible
+ * dcn32_subvp_vblank_admissable() - Determine if SubVP + Vblank config is admissible
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @vlevel: Voltage level calculated by DML
  *
  * SubVP + Vblank is admissible under the following conditions:
  * - Config must have 2 displays (i.e., 2 non-phantom master pipes)
@@ -686,9 +683,7 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context)
  * - DML must have output DRAM clock change support as SubVP + Vblank
  * - The potential vblank display must not be PSR capable
  *
- * @return: True if admissible, false otherwise
- *
- * ************************************************************************************************
+ * Return: True if admissible, false otherwise
  */
 bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int vlevel)
 {
-- 
GitLab


From ebbb0b103efdcf9b682c7fe8ea84b1cf355304c6 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 5 Jun 2023 09:41:29 +0530
Subject: [PATCH 1725/1791] drm/amd/display: Clean up dcn10_optc.c kdoc

Fixes the following W=1 kernel build warning:

display/dc/dcn10/dcn10_optc.c:45: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
* apply_front_porch_workaround  TODO FPGA still need?
display/dc/dcn10/dcn10_optc.c:136: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * program_timing_generator   used by mode timing set
display/dc/dcn10/dcn10_optc.c:391: warning: Function parameter or member 'optc' not described in 'optc1_set_timing_double_buffer'
display/dc/dcn10/dcn10_optc.c:391: warning: Function parameter or member 'enable' not described in 'optc1_set_timing_double_buffer'
display/dc/dcn10/dcn10_optc.c:404: warning: Function parameter or member 'optc' not described in 'optc1_unblank_crtc'
display/dc/dcn10/dcn10_optc.c:404: warning: expecting prototype for unblank_crtc(). Prototype was for optc1_unblank_crtc() instead
display/dc/dcn10/dcn10_optc.c:427: warning: Function parameter or member 'optc' not described in 'optc1_blank_crtc'
display/dc/dcn10/dcn10_optc.c:427: warning: expecting prototype for blank_crtc(). Prototype was for optc1_blank_crtc() instead
display/dc/dcn10/dcn10_optc.c:496: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Enable CRTC
display/dc/dcn10/dcn10_optc.c:895: warning: Cannot understand  *****************************************************************************
 on line 895 - I thought it was a doc line

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dcn10/dcn10_optc.c | 57 ++++++++++++-------
 1 file changed, 36 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index 633989fd2514d..0e8f4f36c87c5 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -42,11 +42,13 @@
 #define STATIC_SCREEN_EVENT_MASK_RANGETIMING_DOUBLE_BUFFER_UPDATE_EN 0x100
 
 /**
-* apply_front_porch_workaround  TODO FPGA still need?
-*
-* This is a workaround for a bug that has existed since R5xx and has not been
-* fixed keep Front porch at minimum 2 for Interlaced mode or 1 for progressive.
-*/
+ * apply_front_porch_workaround() - This is a workaround for a bug that has
+ *                                  existed since R5xx and has not been fixed
+ *                                  keep Front porch at minimum 2 for Interlaced
+ *                                  mode or 1 for progressive.
+ *
+ * @timing: Timing parameters used to configure DCN blocks.
+ */
 static void apply_front_porch_workaround(struct dc_crtc_timing *timing)
 {
 	if (timing->flags.INTERLACE == 1) {
@@ -133,9 +135,20 @@ void optc1_setup_vertical_interrupt2(
 }
 
 /**
- * program_timing_generator   used by mode timing set
- * Program CRTC Timing Registers - OTG_H_*, OTG_V_*, Pixel repetition.
- * Including SYNC. Call BIOS command table to program Timings.
+ * optc1_program_timing() - used by mode timing set Program
+ *                          CRTC Timing Registers - OTG_H_*,
+ *                          OTG_V_*, Pixel repetition.
+ *                          Including SYNC. Call BIOS command table to program Timings.
+ *
+ * @optc: timing_generator instance.
+ * @dc_crtc_timing: Timing parameters used to configure DCN blocks.
+ * @vready_offset: Vready's starting position.
+ * @vstartup_start: Vstartup period.
+ * @vupdate_offset: Vupdate starting position.
+ * @vupdate_width: Vupdate duration.
+ * @signal: DC signal types.
+ * @use_vbios: to program timings from BIOS command table.
+ *
  */
 void optc1_program_timing(
 	struct timing_generator *optc,
@@ -385,6 +398,9 @@ void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enab
  * Sets double buffer point for V_TOTAL, H_TOTAL, VTOTAL_MIN,
  * VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers.
  *
+ * @optc: timing_generator instance.
+ * @enable: Enable DRR double buffering control if true, disable otherwise.
+ *
  * Options: any time,  start of frame, dp start of frame (range timing)
  */
 void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable)
@@ -397,8 +413,9 @@ void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable)
 }
 
 /**
- * unblank_crtc
- * Call ASIC Control Object to UnBlank CRTC.
+ * optc1_unblank_crtc() - Call ASIC Control Object to UnBlank CRTC.
+ *
+ * @optc: timing_generator instance.
  */
 static void optc1_unblank_crtc(struct timing_generator *optc)
 {
@@ -419,8 +436,9 @@ static void optc1_unblank_crtc(struct timing_generator *optc)
 }
 
 /**
- * blank_crtc
- * Call ASIC Control Object to Blank CRTC.
+ * optc1_blank_crtc() - Call ASIC Control Object to Blank CRTC.
+ *
+ * @optc: timing_generator instance.
  */
 
 static void optc1_blank_crtc(struct timing_generator *optc)
@@ -493,8 +511,9 @@ void optc1_enable_optc_clock(struct timing_generator *optc, bool enable)
 }
 
 /**
- * Enable CRTC
- * Enable CRTC - call ASIC Control Object to enable Timing generator.
+ * optc1_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator.
+ *
+ * @optc: timing_generator instance.
  */
 static bool optc1_enable_crtc(struct timing_generator *optc)
 {
@@ -890,15 +909,11 @@ static void optc1_program_manual_trigger(struct timing_generator *optc)
 			MANUAL_FLOW_CONTROL, 0);
 }
 
-
 /**
- *****************************************************************************
- *  Function: set_drr
+ * optc1_set_drr() - Program dynamic refresh rate registers m_OTGx_OTG_V_TOTAL_*.
  *
- *  @brief
- *     Program dynamic refresh rate registers m_OTGx_OTG_V_TOTAL_*.
- *
- *****************************************************************************
+ * @optc: timing_generator instance.
+ * @params: parameters used for Dynamic Refresh Rate.
  */
 void optc1_set_drr(
 	struct timing_generator *optc,
-- 
GitLab


From d4a4ff1c8e4cd752b517af7317077939f3a25dfe Mon Sep 17 00:00:00 2001
From: Likun Gao <Likun.Gao@amd.com>
Date: Tue, 13 Jun 2023 12:10:00 +0800
Subject: [PATCH 1726/1791] drm/amdgpu: add wait_for helper for spirom update

Spirom update typically requires extremely long
duration for command execution, and special helper
function to wait for it completion.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 20 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h |  2 ++
 drivers/gpu/drm/amd/amdgpu/psp_v13_0.c  |  9 +++++----
 drivers/gpu/drm/amd/amdgpu/psp_v13_0.h  |  2 ++
 4 files changed, 29 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index dd865beb39a8c..a96f2e369d1ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -569,6 +569,26 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index,
 	return -ETIME;
 }
 
+int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index,
+			       uint32_t reg_val, uint32_t mask, uint32_t msec_timeout)
+{
+	uint32_t val;
+	int i;
+	struct amdgpu_device *adev = psp->adev;
+
+	if (psp->adev->no_hw_access)
+		return 0;
+
+	for (i = 0; i < msec_timeout; i++) {
+		val = RREG32(reg_index);
+		if ((val & mask) == reg_val)
+			return 0;
+		msleep(1);
+	}
+
+	return -ETIME;
+}
+
 static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id)
 {
 	switch (cmd_id) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index d84323923a3fc..2cae0b1a0b8ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -455,6 +455,8 @@ extern const struct amdgpu_ip_block_version psp_v13_0_4_ip_block;
 
 extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index,
 			uint32_t field_val, uint32_t mask, bool check_changed);
+extern int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index,
+			uint32_t field_val, uint32_t mask, uint32_t msec_timeout);
 
 int psp_gpu_reset(struct amdgpu_device *adev);
 int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx,
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
index caee76ab71105..f9cb0d2c89d15 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c
@@ -624,10 +624,11 @@ static int psp_v13_0_exec_spi_cmd(struct psp_context *psp, int cmd)
 	WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_73, 1);
 
 	if (cmd == C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE)
-		return 0;
-
-	ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115),
-				MBOX_READY_FLAG, MBOX_READY_MASK, false);
+		ret = psp_wait_for_spirom_update(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115),
+						 MBOX_READY_FLAG, MBOX_READY_MASK, PSP_SPIROM_UPDATE_TIMEOUT);
+	else
+		ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115),
+				   MBOX_READY_FLAG, MBOX_READY_MASK, false);
 	if (ret) {
 		dev_err(adev->dev, "SPI cmd %x timed out, ret = %d", cmd, ret);
 		return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h
index b2414a729ca17..de5677ce4330d 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h
@@ -25,6 +25,8 @@
 
 #include "amdgpu_psp.h"
 
+#define PSP_SPIROM_UPDATE_TIMEOUT   60000   /* 60s */
+
 void psp_v13_0_set_psp_funcs(struct psp_context *psp);
 
 #endif
-- 
GitLab


From 82a1f42f6aeb54e29b78aa0890ffd3087120264a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 13 Jun 2023 16:05:21 +0530
Subject: [PATCH 1727/1791] drm/amdgpu: Release SDMAv4.4.2 ecc irq properly

Release ECC irq only if irq is enabled - only when RAS feature is enabled
ECC irq gets enabled.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 8eebf9c2bbcdb..4b033b2988b2d 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1434,9 +1434,11 @@ static int sdma_v4_4_2_hw_fini(void *handle)
 		return 0;
 
 	inst_mask = GENMASK(adev->sdma.num_instances - 1, 0);
-	for (i = 0; i < adev->sdma.num_instances; i++) {
-		amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
-			       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
+				       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+		}
 	}
 
 	sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask);
@@ -2073,9 +2075,11 @@ static int sdma_v4_4_2_xcp_suspend(void *handle, uint32_t inst_mask)
 	uint32_t tmp_mask = inst_mask;
 	int i;
 
-	for_each_inst(i, tmp_mask) {
-		amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
-			       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) {
+		for_each_inst(i, tmp_mask) {
+			amdgpu_irq_put(adev, &adev->sdma.ecc_irq,
+				       AMDGPU_SDMA_IRQ_INSTANCE0 + i);
+		}
 	}
 
 	sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask);
-- 
GitLab


From 765663b7faaedf7750ff7e59c3ce5dc51fdd1fca Mon Sep 17 00:00:00 2001
From: Mukul Joshi <mukul.joshi@amd.com>
Date: Mon, 12 Jun 2023 19:02:39 -0400
Subject: [PATCH 1728/1791] drm/amdkfd: Remove DUMMY_VRAM_SIZE

Remove DUMMY_VRAM_SIZE as it is not needed and can result
in reporting incorrect memory size.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 3dcd8f8bc98e5..49f40d9f16e86 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -30,9 +30,6 @@
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 
-/* Fixme: Fake 32GB for 1PNPS1 mode bringup */
-#define DUMMY_VRAM_SIZE 31138512896
-
 /* GPU Processor ID base for dGPUs for which VCRAT needs to be created.
  * GPU processor ID are expressed with Bit[31]=1.
  * The base is set to 0x8000_0000 + 0x1000 to avoid collision with GPU IDs
@@ -1056,8 +1053,6 @@ static int kfd_parse_subtype_mem(struct crat_subtype_memory *mem,
 
 			props->heap_type = heap_type;
 			props->flags = flags;
-			if (size_in_bytes == 0)
-				size_in_bytes = DUMMY_VRAM_SIZE; /* Fixme: TBD */
 			props->size_in_bytes = size_in_bytes;
 			props->width = width;
 
-- 
GitLab


From 740f42a28f4cff9e009a17cc78666165ecca9293 Mon Sep 17 00:00:00 2001
From: Luben Tuikov <luben.tuikov@amd.com>
Date: Sat, 3 Jun 2023 00:08:04 -0400
Subject: [PATCH 1729/1791] drm/amdgpu: Report ras_num_recs in debugfs

Report the number of records stored in the RAS EEPROM table in debugfs.

This can be used by user-space to calculate the capacity of the RAS EEPROM
table since "bad_page_cnt_threshold" is also reported in the same place in
debugfs.

See commit 7fb640714547 ("drm/amdgpu: Add bad_page_cnt_threshold to debugfs").

ras_num_recs can already be inferred by dumping the RAS EEPROM table, also in
the same debugfs location, see commit reference c65b0805e77919 (drm/amdgpu:
RAS EEPROM table is now in debugfs, 2021-04-08). This commit makes it an
integer value easily shown in a single file.

Cc: Alex Deucher <Alexander.Deucher@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Cc: Tao Zhou <tao.zhou1@amd.com>
Cc: Stanley Yang <Stanley.Yang@amd.com>
Cc: John Clements <john.clements@amd.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://lore.kernel.org/r/20230603051043.211548-1-luben.tuikov@amd.com
Acked-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index a6c3265cdbc46..27a32933cbee3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1491,6 +1491,7 @@ static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
 static struct dentry *amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct amdgpu_ras_eeprom_control *eeprom = &con->eeprom_control;
 	struct drm_minor  *minor = adev_to_drm(adev)->primary;
 	struct dentry     *dir;
 
@@ -1501,6 +1502,7 @@ static struct dentry *amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *
 			    &amdgpu_ras_debugfs_eeprom_ops);
 	debugfs_create_u32("bad_page_cnt_threshold", 0444, dir,
 			   &con->bad_page_cnt_threshold);
+	debugfs_create_u32("ras_num_recs", 0444, dir, &eeprom->ras_num_recs);
 	debugfs_create_x32("ras_hw_enabled", 0444, dir, &adev->ras_hw_enabled);
 	debugfs_create_x32("ras_enabled", 0444, dir, &adev->ras_enabled);
 	debugfs_create_file("ras_eeprom_size", S_IRUGO, dir, adev,
-- 
GitLab


From e5df16d9428f5c6d2d0b1eff244d6c330ba9ef3a Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Jun 2023 12:14:00 -0400
Subject: [PATCH 1730/1791] drm/amdgpu/sdma4: set align mask to 255
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The wptr needs to be incremented at at least 64 dword intervals,
use 256 to align with windows.  This should fix potential hangs
with unaligned updates.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 1f83eebfc8a74..cd37f45e01a11 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -2312,7 +2312,7 @@ const struct amd_ip_funcs sdma_v4_0_ip_funcs = {
 
 static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = {
 	.type = AMDGPU_RING_TYPE_SDMA,
-	.align_mask = 0xf,
+	.align_mask = 0xff,
 	.nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
 	.support_64bit_ptrs = true,
 	.secure_submission_supported = true,
@@ -2344,7 +2344,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = {
 
 static const struct amdgpu_ring_funcs sdma_v4_0_page_ring_funcs = {
 	.type = AMDGPU_RING_TYPE_SDMA,
-	.align_mask = 0xf,
+	.align_mask = 0xff,
 	.nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
 	.support_64bit_ptrs = true,
 	.secure_submission_supported = true,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index 4b033b2988b2d..ea5e12390d182 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -1825,7 +1825,7 @@ const struct amd_ip_funcs sdma_v4_4_2_ip_funcs = {
 
 static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = {
 	.type = AMDGPU_RING_TYPE_SDMA,
-	.align_mask = 0xf,
+	.align_mask = 0xff,
 	.nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
 	.support_64bit_ptrs = true,
 	.get_rptr = sdma_v4_4_2_ring_get_rptr,
@@ -1856,7 +1856,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = {
 
 static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = {
 	.type = AMDGPU_RING_TYPE_SDMA,
-	.align_mask = 0xf,
+	.align_mask = 0xff,
 	.nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
 	.support_64bit_ptrs = true,
 	.get_rptr = sdma_v4_4_2_ring_get_rptr,
-- 
GitLab


From 71344a718a9fda8c551cdc4381d354f9a9907f6f Mon Sep 17 00:00:00 2001
From: Luben Tuikov <luben.tuikov@amd.com>
Date: Sat, 10 Jun 2023 06:19:15 -0400
Subject: [PATCH 1731/1791] drm/amdgpu: Fix usage of UMC fill record in RAS

The fixed commit listed in the Fixes tag below, introduced a bug in
amdgpu_ras.c::amdgpu_reserve_page_direct(), in that when introducing the new
amdgpu_umc_fill_error_record() and internally in that new function the physical
address (argument "uint64_t retired_page"--wrong name) is right-shifted by
AMDGPU_GPU_PAGE_SHIFT. Thus, in amdgpu_reserve_page_direct() when we pass
"address" to that new function, we should NOT right-shift it, since this
results, erroneously, in the page address to be 0 for first
2^(2*AMDGPU_GPU_PAGE_SHIFT) memory addresses.

This commit fixes this bug.

Cc: Tao Zhou <tao.zhou1@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Fixes: 400013b268cb ("drm/amdgpu: add umc_fill_error_record to make code more simple")
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://lore.kernel.org/r/20230610113536.10621-1-luben.tuikov@amd.com
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 27a32933cbee3..700eb180ea60f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -171,8 +171,7 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre
 
 	memset(&err_rec, 0x0, sizeof(struct eeprom_table_record));
 	err_data.err_addr = &err_rec;
-	amdgpu_umc_fill_error_record(&err_data, address,
-			(address >> AMDGPU_GPU_PAGE_SHIFT), 0, 0);
+	amdgpu_umc_fill_error_record(&err_data, address, address, 0, 0);
 
 	if (amdgpu_bad_page_threshold != 0) {
 		amdgpu_ras_add_bad_pages(adev, err_data.err_addr,
-- 
GitLab


From fdc95df9c27dd4feb4bd74ac73e69eba49843db1 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 12 Jun 2023 14:32:50 +0530
Subject: [PATCH 1732/1791] drm/radeon: Fix missing prototypes in
 radeon_atpx_handler.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes the following gcc with W=1:

drivers/gpu/drm/radeon/radeon_atpx_handler.c:64:6: warning: no previous prototype for ‘radeon_has_atpx’ [-Wmissing-prototypes]
   64 | bool 4(void) {
      |      ^~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:68:6: warning: no previous prototype for ‘radeon_has_atpx_dgpu_power_cntl’ [-Wmissing-prototypes]
   68 | bool radeon_has_atpx_dgpu_power_cntl(void) {
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:72:6: warning: no previous prototype for ‘radeon_is_atpx_hybrid’ [-Wmissing-prototypes]
   72 | bool radeon_is_atpx_hybrid(void) {
      |      ^~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:77:6: warning: no previous prototype for ‘radeon_atpx_dgpu_req_power_for_displays’ [-Wmissing-prototypes]
   77 | bool radeon_atpx_dgpu_req_power_for_displays(void) {
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:596:6: warning: no previous prototype for ‘radeon_register_atpx_handler’ [-Wmissing-prototypes]
  596 | void radeon_register_atpx_handler(void)
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:614:6: warning: no previous prototype for ‘radeon_unregister_atpx_handler’ [-Wmissing-prototypes]
  614 | void radeon_unregister_atpx_handler(void)
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/radeon/radeon_atpx_handler.c:159: warning: expecting prototype for radeon_atpx_validate_functions(). Prototype was for radeon_atpx_validate() instead

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_acpi.h         | 9 +++++++++
 drivers/gpu/drm/radeon/radeon_atpx_handler.c | 2 +-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/radeon/radeon_acpi.h b/drivers/gpu/drm/radeon/radeon_acpi.h
index 35202a453e660..974fbb4ce2c29 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.h
+++ b/drivers/gpu/drm/radeon/radeon_acpi.h
@@ -453,4 +453,13 @@ struct acpi_bus_event;
  * BYTE  - number of active lanes
  */
 
+#if defined(CONFIG_VGA_SWITCHEROO)
+void radeon_register_atpx_handler(void);
+void radeon_unregister_atpx_handler(void);
+bool radeon_has_atpx_dgpu_power_cntl(void);
+bool radeon_is_atpx_hybrid(void);
+bool radeon_has_atpx(void);
+bool radeon_atpx_dgpu_req_power_for_displays(void);
+#endif
+
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 6f93f54bf6519..d0b450a065060 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -147,7 +147,7 @@ static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mas
 }
 
 /**
- * radeon_atpx_validate_functions - validate ATPX functions
+ * radeon_atpx_validate() - validate ATPX functions
  *
  * @atpx: radeon atpx struct
  *
-- 
GitLab


From 121f17ac42df73f0869c1bdce090b31935ea37c8 Mon Sep 17 00:00:00 2001
From: Tim Huang <Tim.Huang@amd.com>
Date: Fri, 9 Jun 2023 13:29:37 +0800
Subject: [PATCH 1733/1791] drm/amd/pm: enable more Pstates profile levels for
 SMU v13.0.5

This patch enables following UMD stable Pstates profile
levels for power_dpm_force_performance_level interface.

- profile_peak
- profile_min_sclk
- profile_standard

Signed-off-by: Tim Huang <Tim.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c  | 40 ++++++++++++++++++-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h  |  2 +-
 2 files changed, 39 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
index 53c508acf8951..42f110602eb1a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c
@@ -978,6 +978,38 @@ static int smu_v13_0_5_force_clk_levels(struct smu_context *smu,
 	return ret;
 }
 
+static int smu_v13_0_5_get_dpm_profile_freq(struct smu_context *smu,
+					enum amd_dpm_forced_level level,
+					enum smu_clk_type clk_type,
+					uint32_t *min_clk,
+					uint32_t *max_clk)
+{
+	int ret = 0;
+	uint32_t clk_limit = 0;
+
+	switch (clk_type) {
+	case SMU_GFXCLK:
+	case SMU_SCLK:
+		clk_limit = SMU_13_0_5_UMD_PSTATE_GFXCLK;
+		if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)
+			smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit);
+		else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK)
+			smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL);
+		break;
+	case SMU_VCLK:
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit);
+		break;
+	case SMU_DCLK:
+		smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	*min_clk = *max_clk = clk_limit;
+	return ret;
+}
+
 static int smu_v13_0_5_set_performance_level(struct smu_context *smu,
 						enum amd_dpm_forced_level level)
 {
@@ -1011,10 +1043,14 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu,
 		break;
 	case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK:
-	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK:
-		/* Temporarily do nothing since the optimal clocks haven't been provided yet */
+		smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max);
+		smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max);
+		smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max);
 		break;
+	case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK:
+		dev_err(adev->dev, "The performance level profile_min_mclk is not supported.");
+		return -EOPNOTSUPP;
 	case AMD_DPM_FORCED_LEVEL_MANUAL:
 	case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT:
 		return 0;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h
index 40bc0f8e6d610..263cd651855ea 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h
@@ -24,6 +24,6 @@
 #define __SMU_V13_0_5_PPT_H__
 
 extern void smu_v13_0_5_set_ppt_funcs(struct smu_context *smu);
-#define SMU_13_0_5_UMD_PSTATE_GFXCLK   1100
+#define SMU_13_0_5_UMD_PSTATE_GFXCLK   700
 
 #endif
-- 
GitLab


From 80a780ab279906ec4d3b3589bc324746f12d8dbd Mon Sep 17 00:00:00 2001
From: Jonathan Kim <jonathan.kim@amd.com>
Date: Tue, 13 Jun 2023 15:57:39 -0400
Subject: [PATCH 1734/1791] drm/amdkfd: decrement queue count on mes queue
 destroy

Queue count should decrement on queue destruction regardless of HWS
support type.

Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 8a39a9e0ed5ac..f515cb8f30caf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2089,8 +2089,8 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 	list_del(&q->list);
 	qpd->queue_count--;
 	if (q->properties.is_active) {
+		decrement_queue_count(dqm, qpd, q);
 		if (!dqm->dev->kfd->shared_resources.enable_mes) {
-			decrement_queue_count(dqm, qpd, q);
 			retval = execute_queues_cpsch(dqm,
 						      KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0,
 						      USE_DEFAULT_GRACE_PERIOD);
-- 
GitLab


From c39ca69b84acbfe0a9e09f62f78f8d769d849940 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 13 Jun 2023 20:18:38 +0530
Subject: [PATCH 1735/1791] drm/amd/display: Convert to kdoc formats in
 dc/core/dc.c

Fixes the following gcc with W=1:

drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:3483: warning: Cannot understand  * *******************************************************************************
drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:4204: warning: Cannot understand  * *******************************************************************************

Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 34 +++++++++++-------------
 1 file changed, 15 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 5d3d61faeb28e..dd3a9d06c6e21 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3480,23 +3480,21 @@ static void build_dmub_update_dirty_rect(
 
 
 /**
- * ************************************************************************************************
- * build_dmub_cmd_list: Build an array of DMCUB commands to be sent to DMCUB
+ * build_dmub_cmd_list() - Build an array of DMCUB commands to be sent to DMCUB
  *
- * @param [in]: dc: Current DC state
- * @param [in]: srf_updates: Array of surface updates
- * @param [in]: surface_count: Number of surfaces that have an updated
- * @param [in]: stream: Correponding stream to be updated in the current flip
- * @param [in]: context: New DC state to be programmed
+ * @dc: Current DC state
+ * @srf_updates: Array of surface updates
+ * @surface_count: Number of surfaces that have an updated
+ * @stream: Corresponding stream to be updated in the current flip
+ * @context: New DC state to be programmed
  *
- * @param [out]: dc_dmub_cmd: Array of DMCUB commands to be sent to DMCUB
- * @param [out]: dmub_cmd_count: Count indicating the number of DMCUB commands in dc_dmub_cmd array
+ * @dc_dmub_cmd: Array of DMCUB commands to be sent to DMCUB
+ * @dmub_cmd_count: Count indicating the number of DMCUB commands in dc_dmub_cmd array
  *
  * This function builds an array of DMCUB commands to be sent to DMCUB. This function is required
  * to build an array of commands and have them sent while the OTG lock is acquired.
  *
- * @return: void
- * ************************************************************************************************
+ * Return: void
  */
 static void build_dmub_cmd_list(struct dc *dc,
 		struct dc_surface_update *srf_updates,
@@ -4201,20 +4199,18 @@ static bool commit_minimal_transition_state(struct dc *dc,
 }
 
 /**
- * *******************************************************************************
- * update_seamless_boot_flags: Helper function for updating seamless boot flags
+ * update_seamless_boot_flags() - Helper function for updating seamless boot flags
  *
- * @param [in]: dc: Current DC state
- * @param [in]: context: New DC state to be programmed
- * @param [in]: surface_count: Number of surfaces that have an updated
- * @param [in]: stream: Correponding stream to be updated in the current flip
+ * @dc: Current DC state
+ * @context: New DC state to be programmed
+ * @surface_count: Number of surfaces that have an updated
+ * @stream: Corresponding stream to be updated in the current flip
  *
  * Updating seamless boot flags do not need to be part of the commit sequence. This
  * helper function will update the seamless boot flags on each flip (if required)
  * outside of the HW commit sequence (fast or slow).
  *
- * @return: void
- * *******************************************************************************
+ * Return: void
  */
 static void update_seamless_boot_flags(struct dc *dc,
 		struct dc_state *context,
-- 
GitLab


From 4506f0bc15f42d22fc50f75a098ff9133ffdbe6f Mon Sep 17 00:00:00 2001
From: Candice Li <candice.li@amd.com>
Date: Fri, 9 Jun 2023 10:53:02 +0800
Subject: [PATCH 1736/1791] drm/amd/pm: Align eccinfo table structure with smu
 v13_0_0 interface

Update eccinfo table structure according to smu v13_0_0 interface.

v2: Calculate array size instead of using macro definition.

Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index 4dd01b3f350ff..a6083957ae51b 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -46,7 +46,6 @@
 #include "asic_reg/mp/mp_13_0_0_sh_mask.h"
 #include "smu_cmn.h"
 #include "amdgpu_ras.h"
-#include "umc_v8_10.h"
 
 /*
  * DO NOT use these for err/warn/info/debug messages.
@@ -2609,7 +2608,7 @@ static ssize_t smu_v13_0_0_get_ecc_info(struct smu_context *smu,
 
 	ecc_table = (EccInfoTable_t *)smu_table->ecc_table;
 
-	for (i = 0; i < UMC_V8_10_TOTAL_CHANNEL_NUM(adev); i++) {
+	for (i = 0; i < ARRAY_SIZE(ecc_table->EccInfo); i++) {
 		ecc_info_per_channel = &(eccinfo->ecc[i]);
 		ecc_info_per_channel->ce_count_lo_chip =
 				ecc_table->EccInfo[i].ce_count_lo_chip;
-- 
GitLab


From bcd9a5f8b9e2a705bc30e9b27ebf7b8a8625325f Mon Sep 17 00:00:00 2001
From: Candice Li <candice.li@amd.com>
Date: Sat, 10 Jun 2023 10:15:14 +0800
Subject: [PATCH 1737/1791] drm/amdgpu: Update total channel number for umc
 v8_10

Update total channel number for umc v8_10.

Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h       | 2 ++
 drivers/gpu/drm/amd/amdgpu/umc_v8_10.h        | 3 ++-
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 859882109f55d..16cf7b199457e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1515,6 +1515,7 @@ static int amdgpu_discovery_get_mall_info(struct amdgpu_device *adev)
 				mall_size += mall_size_per_umc;
 		}
 		adev->gmc.mall_size = mall_size;
+		adev->gmc.m_half_use = half_use;
 		break;
 	default:
 		dev_err(adev->dev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 6794edd1d2d2a..56d73fade5685 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -301,6 +301,8 @@ struct amdgpu_gmc {
 
 	/* MALL size */
 	u64 mall_size;
+	uint32_t m_half_use;
+
 	/* number of UMC instances */
 	int num_umc;
 	/* mode2 save restore */
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h
index c6dfd433fec7b..dc12e0af5451e 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h
@@ -33,7 +33,8 @@
 
 /* Total channel instances for all available umc nodes */
 #define UMC_V8_10_TOTAL_CHANNEL_NUM(adev) \
-	(UMC_V8_10_CHANNEL_INSTANCE_NUM * UMC_V8_10_UMC_INSTANCE_NUM * (adev)->gmc.num_umc)
+	(UMC_V8_10_CHANNEL_INSTANCE_NUM * UMC_V8_10_UMC_INSTANCE_NUM * \
+	(adev)->gmc.num_umc - hweight32((adev)->gmc.m_half_use) * 2)
 
 /* UMC regiser per channel offset */
 #define UMC_V8_10_PER_CHANNEL_OFFSET	0x400
-- 
GitLab


From 6fac3964a9092f0ac797cb30cce5fd44f80e5a09 Mon Sep 17 00:00:00 2001
From: Candice Li <candice.li@amd.com>
Date: Mon, 12 Jun 2023 11:00:41 +0800
Subject: [PATCH 1738/1791] drm/amdgpu: Add channel_dis_num to ras init flags

Add disabled channel number to ras init flags.

Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 1 +
 drivers/gpu/drm/amd/amdgpu/ta_ras_if.h  | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index a96f2e369d1ae..395e17565d594 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1677,6 +1677,7 @@ int psp_ras_initialize(struct psp_context *psp)
 		ras_cmd->ras_in_message.init_flags.dgpu_mode = 1;
 	ras_cmd->ras_in_message.init_flags.xcc_mask =
 		adev->gfx.xcc_mask;
+	ras_cmd->ras_in_message.init_flags.channel_dis_num = hweight32(adev->gmc.m_half_use) * 2;
 
 	ret = psp_ta_load(psp, &psp->ras_context.context);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
index be2984ac00a56..879bb7af297c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
+++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h
@@ -130,6 +130,7 @@ struct ta_ras_init_flags {
 	uint8_t poison_mode_en;
 	uint8_t dgpu_mode;
 	uint16_t xcc_mask;
+	uint8_t channel_dis_num;
 };
 
 struct ta_ras_output_flags {
-- 
GitLab


From 38298ce6fc35c65ba1364e4221a289dfa07bf5ea Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Mon, 12 Jun 2023 15:25:05 +0800
Subject: [PATCH 1739/1791] drm/amdgpu: Optimize checking ras supported

Using "is_app_apu" to identify device in the native
APU mode or carveout mode.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c |  8 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 34 ++++++++++++++-----------
 3 files changed, 23 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 395e17565d594..e15c27e05564c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1673,7 +1673,7 @@ int psp_ras_initialize(struct psp_context *psp)
 
 	if (amdgpu_ras_is_poison_mode_supported(adev))
 		ras_cmd->ras_in_message.init_flags.poison_mode_en = 1;
-	if (!adev->gmc.xgmi.connected_to_cpu)
+	if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu)
 		ras_cmd->ras_in_message.init_flags.dgpu_mode = 1;
 	ras_cmd->ras_in_message.init_flags.xcc_mask =
 		adev->gfx.xcc_mask;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 700eb180ea60f..89e9ba33aca1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1686,8 +1686,7 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager *
 		}
 	}
 
-	if (!adev->gmc.xgmi.connected_to_cpu)
-		amdgpu_umc_poison_handler(adev, false);
+	amdgpu_umc_poison_handler(adev, false);
 
 	if (block_obj->hw_ops && block_obj->hw_ops->handle_poison_consumption)
 		poison_stat = block_obj->hw_ops->handle_poison_consumption(adev);
@@ -2452,11 +2451,10 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev)
 {
 	adev->ras_hw_enabled = adev->ras_enabled = 0;
 
-	if (!adev->is_atom_fw ||
-	    !amdgpu_ras_asic_supported(adev))
+	if (!amdgpu_ras_asic_supported(adev))
 		return;
 
-	if (!adev->gmc.xgmi.connected_to_cpu) {
+	if (!adev->gmc.xgmi.connected_to_cpu &&	!adev->gmc.is_app_apu) {
 		if (amdgpu_atomfirmware_mem_ecc_supported(adev)) {
 			dev_info(adev->dev, "MEM ECC is active.\n");
 			adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__UMC |
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
index 1edf8e6aeb162..db0d94ca4ffc5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@@ -169,27 +169,31 @@ int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset)
 {
 	int ret = AMDGPU_RAS_SUCCESS;
 
-	if (!amdgpu_sriov_vf(adev)) {
-		if (!adev->gmc.xgmi.connected_to_cpu) {
-			struct ras_err_data err_data = {0, 0, 0, NULL};
-			struct ras_common_if head = {
-				.block = AMDGPU_RAS_BLOCK__UMC,
-			};
-			struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head);
-
-			ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset);
-
-			if (ret == AMDGPU_RAS_SUCCESS && obj) {
-				obj->err_data.ue_count += err_data.ue_count;
-				obj->err_data.ce_count += err_data.ce_count;
-			}
-		} else if (reset) {
+	if (adev->gmc.xgmi.connected_to_cpu ||
+		adev->gmc.is_app_apu) {
+		if (reset) {
 			/* MCA poison handler is only responsible for GPU reset,
 			 * let MCA notifier do page retirement.
 			 */
 			kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
 			amdgpu_ras_reset_gpu(adev);
 		}
+		return ret;
+	}
+
+	if (!amdgpu_sriov_vf(adev)) {
+		struct ras_err_data err_data = {0, 0, 0, NULL};
+		struct ras_common_if head = {
+			.block = AMDGPU_RAS_BLOCK__UMC,
+		};
+		struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head);
+
+		ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset);
+
+		if (ret == AMDGPU_RAS_SUCCESS && obj) {
+			obj->err_data.ue_count += err_data.ue_count;
+			obj->err_data.ce_count += err_data.ce_count;
+		}
 	} else {
 		if (adev->virt.ops && adev->virt.ops->ras_poison_handler)
 			adev->virt.ops->ras_poison_handler(adev);
-- 
GitLab


From 43aedbf4da1db9a9c2f9e160a4ae96dfda83774c Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Mon, 12 Jun 2023 18:25:48 +0800
Subject: [PATCH 1740/1791] drm/amdgpu: Add checking mc_vram_size

Do not compare injection address with mc_vram_size
if mc_vram_size is zero.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 89e9ba33aca1e..4769a18304d7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -489,7 +489,8 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f,
 		ret = amdgpu_ras_feature_enable(adev, &data.head, 1);
 		break;
 	case 2:
-		if ((data.inject.address >= adev->gmc.mc_vram_size) ||
+		if ((data.inject.address >= adev->gmc.mc_vram_size &&
+		    adev->gmc.mc_vram_size) ||
 		    (data.inject.address >= RAS_UMC_INJECT_ADDR_LIMIT)) {
 			dev_warn(adev->dev, "RAS WARN: input address "
 					"0x%llx is invalid.",
-- 
GitLab


From e2ad8e2df432498b1cee2af04df605723f4d75e6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 5 Jun 2023 11:11:21 +0200
Subject: [PATCH 1741/1791] drm/amdgpu: make sure BOs are locked in
 amdgpu_vm_get_memory
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We need to grab the lock of the BO or otherwise can run into a crash
when we try to inspect the current location.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Guchun Chen <guchun.chen@amd.com>
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 69 +++++++++++++++-----------
 1 file changed, 39 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index dc80c9c8fd140..5dcc8001e65c1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -920,42 +920,51 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	return r;
 }
 
+static void amdgpu_vm_bo_get_memory(struct amdgpu_bo_va *bo_va,
+				    struct amdgpu_mem_stats *stats)
+{
+	struct amdgpu_vm *vm = bo_va->base.vm;
+	struct amdgpu_bo *bo = bo_va->base.bo;
+
+	if (!bo)
+		return;
+
+	/*
+	 * For now ignore BOs which are currently locked and potentially
+	 * changing their location.
+	 */
+	if (bo->tbo.base.resv != vm->root.bo->tbo.base.resv &&
+	    !dma_resv_trylock(bo->tbo.base.resv))
+		return;
+
+	amdgpu_bo_get_memory(bo, stats);
+	if (bo->tbo.base.resv != vm->root.bo->tbo.base.resv)
+	    dma_resv_unlock(bo->tbo.base.resv);
+}
+
 void amdgpu_vm_get_memory(struct amdgpu_vm *vm,
 			  struct amdgpu_mem_stats *stats)
 {
 	struct amdgpu_bo_va *bo_va, *tmp;
 
 	spin_lock(&vm->status_lock);
-	list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
-	list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
-	list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
-	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
-	list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
-	list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) {
-		if (!bo_va->base.bo)
-			continue;
-		amdgpu_bo_get_memory(bo_va->base.bo, stats);
-	}
+	list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
+
+	list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status)
+		amdgpu_vm_bo_get_memory(bo_va, stats);
 	spin_unlock(&vm->status_lock);
 }
 
-- 
GitLab


From ca0b954a4315ca2228001c439ae1062561c81989 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 5 Jun 2023 11:11:22 +0200
Subject: [PATCH 1742/1791] drm/amdgpu: make sure that BOs have a backing store
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's perfectly possible that the BO is about to be destroyed and doesn't
have a backing store associated with it.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Guchun Chen <guchun.chen@amd.com>
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 54cd21c06a44f..f7905bce0de15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1282,8 +1282,12 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
 			  struct amdgpu_mem_stats *stats)
 {
-	unsigned int domain;
 	uint64_t size = amdgpu_bo_size(bo);
+	unsigned int domain;
+
+	/* Abort if the BO doesn't currently have a backing store */
+	if (!bo->tbo.resource)
+		return;
 
 	domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
 	switch (domain) {
-- 
GitLab


From fe381726c96d9a7c2bd6eafa30f22e968f532cd1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 13 Jun 2023 12:50:02 +0530
Subject: [PATCH 1743/1791] drm/amdgpu: Change nbio v7.9 xcp status definition

PARTITION_MODE field in PARTITION_COMPUTE_STATUS register is defined as
below by firmware.

	SPX = 0, DPX = 1, TPX = 2, QPX = 3, CPX = 4

Change driver definition accordingly.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index d193254767521..b033935d67498 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -390,7 +390,7 @@ static int nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
 	px = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS,
 			   PARTITION_MODE);
 
-	return ffs(px);
+	return px;
 }
 
 static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
@@ -398,12 +398,10 @@ static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
 {
 	u32 tmp;
 
-	/* Each bit represents DPX,TPX,QPX,CPX mode. No bit set means default
-	 * SPX mode.
-	 */
+	/* SPX=0, DPX=1, TPX=2, QPX=3, CPX=4 */
 	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS);
 	tmp = REG_SET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS,
-			    PARTITION_MODE, mode ? BIT(mode - 1) : mode);
+			    PARTITION_MODE, mode);
 
 	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
 }
-- 
GitLab


From b00f55374ccb3e3e9af6ee46761b74acb648440d Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 13 Jun 2023 13:39:36 +0530
Subject: [PATCH 1744/1791] drm/amdgpu: Use PSP FW API for partition switch

Use PSP firmware interface for switching compute partitions.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdgpu/aqua_vanjaram_reg_init.c    |  3 ---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c        | 18 ++++++------------
 2 files changed, 6 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index a595bb958215e..16471b81a1f5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -518,9 +518,6 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 		adev->gfx.funcs->switch_partition_mode(xcp_mgr->adev,
 						       num_xcc_per_xcp);
 
-	if (adev->nbio.funcs->set_compute_partition_mode)
-		adev->nbio.funcs->set_compute_partition_mode(adev, mode);
-
 	/* Init info about new xcps */
 	*num_xcps = num_xcc / num_xcc_per_xcp;
 	amdgpu_xcp_init(xcp_mgr, *num_xcps, mode);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index f5b8d3f388ff6..c1ee54d4c3d3d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -623,22 +623,16 @@ static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev,
 static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev,
 						int num_xccs_per_xcp)
 {
-	int i, num_xcc;
-	u32 tmp = 0;
-
-	num_xcc = NUM_XCC(adev->gfx.xcc_mask);
+	int ret;
 
-	for (i = 0; i < num_xcc; i++) {
-		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP,
-				    num_xccs_per_xcp);
-		tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID,
-				    i % num_xccs_per_xcp);
-		WREG32_SOC15(GC, GET_INST(GC, i), regCP_HYP_XCP_CTL, tmp);
-	}
+	ret = psp_spatial_partition(&adev->psp, NUM_XCC(adev->gfx.xcc_mask) /
+							num_xccs_per_xcp);
+	if (ret)
+		return ret;
 
 	adev->gfx.num_xcc_per_xcp = num_xccs_per_xcp;
 
-	return 0;
+	return ret;
 }
 
 static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node)
-- 
GitLab


From 2eb841bdbca819017e7483cdfbb3d401751848a5 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 15 Jun 2023 09:59:34 -0400
Subject: [PATCH 1745/1791] drm/amdgpu: mark GC 9.4.3 experimental for now

Mark as experimental for now until we get closer to production
to avoid possible undesireable behavior when mixing newer
boards with older kernels.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 16cf7b199457e..8e1cfc87122d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -1897,6 +1897,8 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev)
 		amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block);
 		break;
 	case IP_VERSION(9, 4, 3):
+		if (!amdgpu_exp_hw_support)
+			return -EINVAL;
 		amdgpu_device_ip_block_add(adev, &gfx_v9_4_3_ip_block);
 		break;
 	case IP_VERSION(10, 1, 10):
-- 
GitLab


From b13eb02ba8ba7617d41212121891756da31f1d8b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 19 Apr 2023 12:51:41 +0200
Subject: [PATCH 1746/1791] drm/amdgpu: add amdgpu_error_* debugfs file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This allows us to insert some error codes into the bottom of the pipeline
on an engine.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 24 +++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c  | 16 +++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h  |  1 +
 3 files changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 876ec35b8f83d..0c6ec9c6424c1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -691,6 +691,30 @@ void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring)
 	}
 }
 
+/**
+ * amdgpu_fence_driver_set_error - set error code on fences
+ * @ring: the ring which contains the fences
+ * @error: the error code to set
+ *
+ * Set an error code to all the fences pending on the ring.
+ */
+void amdgpu_fence_driver_set_error(struct amdgpu_ring *ring, int error)
+{
+	struct amdgpu_fence_driver *drv = &ring->fence_drv;
+	unsigned long flags;
+
+	spin_lock_irqsave(&drv->lock, flags);
+	for (unsigned int i = 0; i <= drv->num_fences_mask; ++i) {
+		struct dma_fence *fence;
+
+		fence = rcu_dereference_protected(drv->fences[i],
+						  lockdep_is_held(&drv->lock));
+		if (fence && !dma_fence_is_signaled_locked(fence))
+			dma_fence_set_error(fence, error);
+	}
+	spin_unlock_irqrestore(&drv->lock, flags);
+}
+
 /**
  * amdgpu_fence_driver_force_completion - force signal latest fence of ring
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index da26c555af24b..9fd55a10da191 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -561,6 +561,17 @@ static const struct file_operations amdgpu_debugfs_mqd_fops = {
 	.llseek = default_llseek
 };
 
+static int amdgpu_debugfs_ring_error(void *data, u64 val)
+{
+	struct amdgpu_ring *ring = data;
+
+	amdgpu_fence_driver_set_error(ring, val);
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE_SIGNED(amdgpu_debugfs_error_fops, NULL,
+				amdgpu_debugfs_ring_error, "%lld\n");
+
 #endif
 
 void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
@@ -582,6 +593,11 @@ void amdgpu_debugfs_ring_init(struct amdgpu_device *adev,
 					 &amdgpu_debugfs_mqd_fops,
 					 ring->mqd_size);
 	}
+
+	sprintf(name, "amdgpu_error_%s", ring->name);
+	debugfs_create_file(name, 0200, root, ring,
+			    &amdgpu_debugfs_error_fops);
+
 #endif
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 21ffb9ce32ce2..028ff075db512 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -126,6 +126,7 @@ struct amdgpu_fence_driver {
 extern const struct drm_sched_backend_ops amdgpu_sched_ops;
 
 void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring);
+void amdgpu_fence_driver_set_error(struct amdgpu_ring *ring, int error);
 void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring);
 
 int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring);
-- 
GitLab


From 0a33b11d26c6b7e975b54d469a739ffac29f67ab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 17 Apr 2023 12:52:18 +0200
Subject: [PATCH 1747/1791] drm/amdgpu: mark force completed fences with
 -ECANCELED
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When we force complete fences we should mark them as canceled.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 0c6ec9c6424c1..c694b41f6461b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -723,6 +723,7 @@ void amdgpu_fence_driver_set_error(struct amdgpu_ring *ring, int error)
  */
 void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring)
 {
+	amdgpu_fence_driver_set_error(ring, -ECANCELED);
 	amdgpu_fence_write(ring, ring->fence_drv.sync_seq);
 	amdgpu_fence_process(ring);
 }
-- 
GitLab


From 89fae8dc41d0a9bfc9fc1ea7ec03bf36e680774d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 17 Apr 2023 13:04:06 +0200
Subject: [PATCH 1748/1791] drm/amdgpu: mark soft recovered fences with
 -ENODATA
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set the fence error code before trying to soft-recover it.

It gets overwritten when a hard recovery is required.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 9fd55a10da191..80d6e132e4095 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -433,11 +433,18 @@ void amdgpu_ring_emit_reg_write_reg_wait_helper(struct amdgpu_ring *ring,
 bool amdgpu_ring_soft_recovery(struct amdgpu_ring *ring, unsigned int vmid,
 			       struct dma_fence *fence)
 {
+	unsigned long flags;
+
 	ktime_t deadline = ktime_add_us(ktime_get(), 10000);
 
 	if (amdgpu_sriov_vf(ring->adev) || !ring->funcs->soft_recovery || !fence)
 		return false;
 
+	spin_lock_irqsave(fence->lock, flags);
+	if (!dma_fence_is_signaled_locked(fence))
+		dma_fence_set_error(fence, -ENODATA);
+	spin_unlock_irqrestore(fence->lock, flags);
+
 	atomic_inc(&ring->adev->gpu_reset_counter);
 	while (!dma_fence_is_signaled(fence) &&
 	       ktime_to_ns(ktime_sub(deadline, ktime_get())) > 0)
-- 
GitLab


From e84e697d92d9d84ca13b4440cea36abe9a2fe079 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 17 Apr 2023 18:15:15 +0200
Subject: [PATCH 1749/1791] drm/amdgpu: abort submissions during prepare on
 error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Forward errors from previous submissions to this one.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index aca3a2bfe8d23..9e6f2fa8e2fe5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -258,16 +258,27 @@ amdgpu_job_prepare_job(struct drm_sched_job *sched_job,
 	struct dma_fence *fence = NULL;
 	int r;
 
+	/* Ignore soft recovered fences here */
+	r = drm_sched_entity_error(s_entity);
+	if (r && r != -ENODATA)
+		goto error;
+
 	if (!fence && job->gang_submit)
 		fence = amdgpu_device_switch_gang(ring->adev, job->gang_submit);
 
 	while (!fence && job->vm && !job->vmid) {
 		r = amdgpu_vmid_grab(job->vm, ring, job, &fence);
-		if (r)
+		if (r) {
 			DRM_ERROR("Error getting VM ID (%d)\n", r);
+			goto error;
+		}
 	}
 
 	return fence;
+
+error:
+	dma_fence_set_error(&job->base.s_fence->finished, r);
+	return NULL;
 }
 
 static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job)
-- 
GitLab


From 55bf196f60dfc89488c5645d112a9176c6fe4708 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 18 Apr 2023 13:18:01 +0200
Subject: [PATCH 1750/1791] drm/amdgpu: reset VM when an error is detected
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When some problem with the updates of page tables is detected reset the
state machine of the VM and re-create all page tables from scratch.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 81 +++++++++++++++++++++-----
 1 file changed, 65 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 5dcc8001e65c1..1045be4b547c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -266,6 +266,32 @@ static void amdgpu_vm_bo_done(struct amdgpu_vm_bo_base *vm_bo)
 	spin_unlock(&vm_bo->vm->status_lock);
 }
 
+/**
+ * amdgpu_vm_bo_reset_state_machine - reset the vm_bo state machine
+ * @vm: the VM which state machine to reset
+ *
+ * Move all vm_bo object in the VM into a state where they will be updated
+ * again during validation.
+ */
+static void amdgpu_vm_bo_reset_state_machine(struct amdgpu_vm *vm)
+{
+	struct amdgpu_vm_bo_base *vm_bo, *tmp;
+
+	spin_lock(&vm->status_lock);
+	list_splice_init(&vm->done, &vm->invalidated);
+	list_for_each_entry(vm_bo, &vm->invalidated, vm_status)
+		vm_bo->moved = true;
+	list_for_each_entry_safe(vm_bo, tmp, &vm->idle, vm_status) {
+		struct amdgpu_bo *bo = vm_bo->bo;
+
+		if (!bo || bo->tbo.type != ttm_bo_type_kernel)
+			list_move(&vm_bo->vm_status, &vm_bo->vm->moved);
+		else if (bo->parent)
+			list_move(&vm_bo->vm_status, &vm_bo->vm->relocated);
+	}
+	spin_unlock(&vm->status_lock);
+}
+
 /**
  * amdgpu_vm_bo_base_init - Adds bo to the list of bos associated with the vm
  *
@@ -351,6 +377,34 @@ void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev,
 	spin_unlock(&adev->mman.bdev.lru_lock);
 }
 
+/* Create scheduler entities for page table updates */
+static int amdgpu_vm_init_entities(struct amdgpu_device *adev,
+				   struct amdgpu_vm *vm)
+{
+	int r;
+
+	r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL,
+				  adev->vm_manager.vm_pte_scheds,
+				  adev->vm_manager.vm_pte_num_scheds, NULL);
+	if (r)
+		goto error;
+
+	return drm_sched_entity_init(&vm->delayed, DRM_SCHED_PRIORITY_NORMAL,
+				     adev->vm_manager.vm_pte_scheds,
+				     adev->vm_manager.vm_pte_num_scheds, NULL);
+
+error:
+	drm_sched_entity_destroy(&vm->immediate);
+	return r;
+}
+
+/* Destroy the entities for page table updates again */
+static void amdgpu_vm_fini_entities(struct amdgpu_vm *vm)
+{
+	drm_sched_entity_destroy(&vm->immediate);
+	drm_sched_entity_destroy(&vm->delayed);
+}
+
 /**
  * amdgpu_vm_validate_pt_bos - validate the page table BOs
  *
@@ -373,6 +427,14 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	struct amdgpu_bo *bo;
 	int r;
 
+	if (drm_sched_entity_error(&vm->delayed)) {
+		amdgpu_vm_bo_reset_state_machine(vm);
+		amdgpu_vm_fini_entities(vm);
+		r = amdgpu_vm_init_entities(adev, vm);
+		if (r)
+			return r;
+	}
+
 	spin_lock(&vm->status_lock);
 	while (!list_empty(&vm->evicted)) {
 		bo_base = list_first_entry(&vm->evicted,
@@ -2048,19 +2110,10 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	INIT_LIST_HEAD(&vm->pt_freed);
 	INIT_WORK(&vm->pt_free_work, amdgpu_vm_pt_free_work);
 
-	/* create scheduler entities for page table updates */
-	r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL,
-				  adev->vm_manager.vm_pte_scheds,
-				  adev->vm_manager.vm_pte_num_scheds, NULL);
+	r = amdgpu_vm_init_entities(adev, vm);
 	if (r)
 		return r;
 
-	r = drm_sched_entity_init(&vm->delayed, DRM_SCHED_PRIORITY_NORMAL,
-				  adev->vm_manager.vm_pte_scheds,
-				  adev->vm_manager.vm_pte_num_scheds, NULL);
-	if (r)
-		goto error_free_immediate;
-
 	vm->pte_support_ats = false;
 	vm->is_compute_context = false;
 
@@ -2121,10 +2174,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 error_free_delayed:
 	dma_fence_put(vm->last_tlb_flush);
 	dma_fence_put(vm->last_unlocked);
-	drm_sched_entity_destroy(&vm->delayed);
-
-error_free_immediate:
-	drm_sched_entity_destroy(&vm->immediate);
+	amdgpu_vm_fini_entities(vm);
 
 	return r;
 }
@@ -2277,8 +2327,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	amdgpu_bo_unref(&root);
 	WARN_ON(vm->root.bo);
 
-	drm_sched_entity_destroy(&vm->immediate);
-	drm_sched_entity_destroy(&vm->delayed);
+	amdgpu_vm_fini_entities(vm);
 
 	if (!RB_EMPTY_ROOT(&vm->va.rb_root)) {
 		dev_err(adev->dev, "still active bo inside vm\n");
-- 
GitLab


From f88e295e9094deee93066f32a4380307e8cb3dd9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 19 Apr 2023 15:17:57 +0200
Subject: [PATCH 1751/1791] drm/amdgpu: add VM generation token
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of using the VRAM lost counter add a 64bit token which indicates
if a context or job is still valid to use.

Should the VRAM be lost or the page tables need re-creation the token will
change indicating that userspace needs to act and re-create the contexts
and re-submit the work.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c  |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c |  4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c |  4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.h |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c  | 26 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h  |  4 ++++
 7 files changed, 37 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 6e1d331af01ff..d9503882ea972 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -309,7 +309,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 	}
 	p->gang_leader = p->jobs[p->gang_leader_idx];
 
-	if (p->ctx->vram_lost_counter != p->gang_leader->vram_lost_counter) {
+	if (p->ctx->generation != p->gang_leader->generation) {
 		ret = -ECANCELED;
 		goto free_all_kdata;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 410acdd4554ca..3ccd709ae76a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -333,7 +333,7 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority,
 
 	ctx->reset_counter = atomic_read(&mgr->adev->gpu_reset_counter);
 	ctx->reset_counter_query = ctx->reset_counter;
-	ctx->vram_lost_counter = atomic_read(&mgr->adev->vram_lost_counter);
+	ctx->generation = amdgpu_vm_generation(mgr->adev, &fpriv->vm);
 	ctx->init_priority = priority;
 	ctx->override_priority = AMDGPU_CTX_PRIORITY_UNSET;
 
@@ -586,7 +586,7 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev,
 	if (ctx->reset_counter != atomic_read(&adev->gpu_reset_counter))
 		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RESET;
 
-	if (ctx->vram_lost_counter != atomic_read(&adev->vram_lost_counter))
+	if (ctx->generation != amdgpu_vm_generation(adev, &fpriv->vm))
 		out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_VRAMLOST;
 
 	if (atomic_read(&ctx->guilty))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
index f1e27b6e16f48..85376baaa92f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h
@@ -47,7 +47,7 @@ struct amdgpu_ctx {
 	struct amdgpu_ctx_mgr		*mgr;
 	unsigned			reset_counter;
 	unsigned			reset_counter_query;
-	uint32_t			vram_lost_counter;
+	uint64_t			generation;
 	spinlock_t			ring_lock;
 	struct amdgpu_ctx_entity	*entities[AMDGPU_HW_IP_NUM][AMDGPU_MAX_ENTITY_NUM];
 	bool				preamble_presented;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 9e6f2fa8e2fe5..78476bc75b4e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -109,7 +109,7 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	(*job)->vm = vm;
 
 	amdgpu_sync_create(&(*job)->explicit_sync);
-	(*job)->vram_lost_counter = atomic_read(&adev->vram_lost_counter);
+	(*job)->generation = amdgpu_vm_generation(adev, vm);
 	(*job)->vm_pd_addr = AMDGPU_BO_INVALID_OFFSET;
 
 	if (!entity)
@@ -295,7 +295,7 @@ static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job)
 	trace_amdgpu_sched_run_job(job);
 
 	/* Skip job if VRAM is lost and never resubmit gangs */
-	if (job->vram_lost_counter != atomic_read(&adev->vram_lost_counter) ||
+	if (job->generation != amdgpu_vm_generation(adev, job->vm) ||
 	    (job->job_run_counter && job->gang_submit))
 		dma_fence_set_error(finished, -ECANCELED);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index 3f9804f956c9c..a963a25ddd620 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -61,7 +61,7 @@ struct amdgpu_job {
 	uint32_t		gds_base, gds_size;
 	uint32_t		gws_base, gws_size;
 	uint32_t		oa_base, oa_size;
-	uint32_t		vram_lost_counter;
+	uint64_t		generation;
 
 	/* user fence handling */
 	uint64_t		uf_addr;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 1045be4b547c7..143d11afe0e5a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -405,6 +405,30 @@ static void amdgpu_vm_fini_entities(struct amdgpu_vm *vm)
 	drm_sched_entity_destroy(&vm->delayed);
 }
 
+/**
+ * amdgpu_vm_generation - return the page table re-generation counter
+ * @adev: the amdgpu_device
+ * @vm: optional VM to check, might be NULL
+ *
+ * Returns a page table re-generation token to allow checking if submissions
+ * are still valid to use this VM. The VM parameter might be NULL in which case
+ * just the VRAM lost counter will be used.
+ */
+uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm)
+{
+	uint64_t result = (u64)atomic_read(&adev->vram_lost_counter) << 32;
+
+	if (!vm)
+		return result;
+
+	result += vm->generation;
+	/* Add one if the page tables will be re-generated on next CS */
+	if (drm_sched_entity_error(&vm->delayed))
+		++result;
+
+	return result;
+}
+
 /**
  * amdgpu_vm_validate_pt_bos - validate the page table BOs
  *
@@ -428,6 +452,7 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	int r;
 
 	if (drm_sched_entity_error(&vm->delayed)) {
+		++vm->generation;
 		amdgpu_vm_bo_reset_state_machine(vm);
 		amdgpu_vm_fini_entities(vm);
 		r = amdgpu_vm_init_entities(adev, vm);
@@ -2134,6 +2159,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	vm->last_update = dma_fence_get_stub();
 	vm->last_unlocked = dma_fence_get_stub();
 	vm->last_tlb_flush = dma_fence_get_stub();
+	vm->generation = 0;
 
 	mutex_init(&vm->eviction_lock);
 	vm->evicting = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 14f9a2bf3acb9..9c85d494f2a24 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -295,6 +295,9 @@ struct amdgpu_vm {
 	atomic64_t		tlb_seq;
 	struct dma_fence	*last_tlb_flush;
 
+	/* How many times we had to re-generate the page tables */
+	uint64_t		generation;
+
 	/* Last unlocked submission to the scheduler entities */
 	struct dma_fence	*last_unlocked;
 
@@ -397,6 +400,7 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
 			 struct list_head *validated,
 			 struct amdgpu_bo_list_entry *entry);
 bool amdgpu_vm_ready(struct amdgpu_vm *vm);
+uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm);
 int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			      int (*callback)(void *p, struct amdgpu_bo *bo),
 			      void *param);
-- 
GitLab


From 71eaac368dccf0619f7adc012063930e459b133e Mon Sep 17 00:00:00 2001
From: ZhenGuo Yin <zhenguo.yin@amd.com>
Date: Thu, 11 May 2023 17:29:20 +0800
Subject: [PATCH 1752/1791] drm/amdgpu: add entity error check in
 amdgpu_ctx_get_entity

[Why]
UMD is not aware of entity error, and will keep submitting jobs
into the error entity.

[How]
Add entity error check when getting entity from ctx.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 3ccd709ae76a3..0dc9c655c4fbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -432,6 +432,7 @@ int amdgpu_ctx_get_entity(struct amdgpu_ctx *ctx, u32 hw_ip, u32 instance,
 			  u32 ring, struct drm_sched_entity **entity)
 {
 	int r;
+	struct drm_sched_entity *ctx_entity;
 
 	if (hw_ip >= AMDGPU_HW_IP_NUM) {
 		DRM_ERROR("unknown HW IP type: %d\n", hw_ip);
@@ -455,7 +456,14 @@ int amdgpu_ctx_get_entity(struct amdgpu_ctx *ctx, u32 hw_ip, u32 instance,
 			return r;
 	}
 
-	*entity = &ctx->entities[hw_ip][ring]->entity;
+	ctx_entity = &ctx->entities[hw_ip][ring]->entity;
+	r = drm_sched_entity_error(ctx_entity);
+	if (r) {
+		DRM_DEBUG("error entity %p\n", ctx_entity);
+		return r;
+	}
+
+	*entity = ctx_entity;
 	return 0;
 }
 
-- 
GitLab


From 4f9b94d848696166011bead3109541ec2a523bb8 Mon Sep 17 00:00:00 2001
From: ZhenGuo Yin <zhenguo.yin@amd.com>
Date: Tue, 9 May 2023 17:46:17 +0800
Subject: [PATCH 1753/1791] drm/scheduler: avoid infinite loop if entity's
 dependency is a scheduled error fence
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
drm_sched_entity_add_dependency_cb ignores the scheduled fence and return false.
If entity's dependency is a scheduler error fence and drm_sched_stop is called
due to TDR, drm_sched_entity_pop_job will wait for the dependency infinitely.

[How]
Do not wait or ignore the scheduled error fence, add drm_sched_entity_wakeup
callback for the dependency with scheduled error fence.

Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/scheduler/sched_entity.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 68e807ae136ad..b2bbc8a68b305 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -384,7 +384,7 @@ static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity)
 	}
 
 	s_fence = to_drm_sched_fence(fence);
-	if (s_fence && s_fence->sched == sched &&
+	if (!fence->error && s_fence && s_fence->sched == sched &&
 	    !test_bit(DRM_SCHED_FENCE_DONT_PIPELINE, &fence->flags)) {
 
 		/*
-- 
GitLab


From 6f582513ad15de729ee5c91dfef946f3c266a207 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 17 May 2023 16:19:51 -0400
Subject: [PATCH 1754/1791] drm/amdkfd: add event age tracking

Add event age tracking

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/uapi/linux/kfd_ioctl.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 1781e7669982b..93f1c0bc5caff 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -320,12 +320,20 @@ struct kfd_hsa_hw_exception_data {
 	__u32 gpu_id;
 };
 
+/* hsa signal event data */
+struct kfd_hsa_signal_event_data {
+	__u64 last_event_age;	/* to and from KFD */
+};
+
 /* Event data */
 struct kfd_event_data {
 	union {
+		/* From KFD */
 		struct kfd_hsa_memory_exception_data memory_exception_data;
 		struct kfd_hsa_hw_exception_data hw_exception_data;
-	};				/* From KFD */
+		/* To and From KFD */
+		struct kfd_hsa_signal_event_data signal_event_data;
+	};
 	__u64 kfd_event_data_ext;	/* pointer to an extension structure
 					   for future exception types */
 	__u32 event_id;		/* to KFD */
-- 
GitLab


From 4057e6ce3384e079f945a7f69797fc6c2864a90f Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 17 May 2023 16:22:47 -0400
Subject: [PATCH 1755/1791] drm/amdkfd: add event_age tracking when receiving
 interrupt

Add event_age tracking when receiving interrupt.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_events.c | 6 ++++++
 drivers/gpu/drm/amd/amdkfd/kfd_events.h | 1 +
 2 files changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index ddca23ee4193e..d71ea0659d2f3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -431,6 +431,7 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p,
 	if (!ret) {
 		*event_id = ev->event_id;
 		*event_trigger_data = ev->event_id;
+		ev->event_age = 1;
 	} else {
 		kfree(ev);
 	}
@@ -629,6 +630,11 @@ static void set_event(struct kfd_event *ev)
 	 * updating the wait queues in kfd_wait_on_events.
 	 */
 	ev->signaled = !ev->auto_reset || !waitqueue_active(&ev->wq);
+	if (!(++ev->event_age)) {
+		/* Never wrap back to reserved/default event age 0/1 */
+		ev->event_age = 2;
+		WARN_ONCE(1, "event_age wrap back!");
+	}
 
 	list_for_each_entry(waiter, &ev->wq.head, wait.entry)
 		WRITE_ONCE(waiter->activated, true);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_events.h
index 1c62c8dd64601..52ccfd397c2b3 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.h
@@ -53,6 +53,7 @@ struct signal_page;
 
 struct kfd_event {
 	u32 event_id;
+	u64 event_age;
 
 	bool signaled;
 	bool auto_reset;
-- 
GitLab


From 96cdb5384d962a7d3be598f0bc9e2be73796e80c Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 17 May 2023 16:25:12 -0400
Subject: [PATCH 1756/1791] drm/amdkfd: set activated flag true when event age
 unmatchs

Set waiter's activated flag true when event age unmatchs with last_event_age.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_events.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index d71ea0659d2f3..545179535d21a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -41,6 +41,7 @@ struct kfd_event_waiter {
 	wait_queue_entry_t wait;
 	struct kfd_event *event; /* Event to wait for */
 	bool activated;		 /* Becomes true when event is signaled */
+	bool event_age_enabled;  /* set to true when last_event_age is non-zero */
 };
 
 /*
@@ -797,9 +798,9 @@ static struct kfd_event_waiter *alloc_event_waiters(uint32_t num_events)
 
 static int init_event_waiter(struct kfd_process *p,
 		struct kfd_event_waiter *waiter,
-		uint32_t event_id)
+		struct kfd_event_data *event_data)
 {
-	struct kfd_event *ev = lookup_event_by_id(p, event_id);
+	struct kfd_event *ev = lookup_event_by_id(p, event_data->event_id);
 
 	if (!ev)
 		return -EINVAL;
@@ -808,6 +809,15 @@ static int init_event_waiter(struct kfd_process *p,
 	waiter->event = ev;
 	waiter->activated = ev->signaled;
 	ev->signaled = ev->signaled && !ev->auto_reset;
+
+	/* last_event_age = 0 reserved for backward compatible */
+	if (waiter->event->type == KFD_EVENT_TYPE_SIGNAL &&
+		event_data->signal_event_data.last_event_age) {
+		waiter->event_age_enabled = true;
+		if (ev->event_age != event_data->signal_event_data.last_event_age)
+			waiter->activated = true;
+	}
+
 	if (!waiter->activated)
 		add_wait_queue(&ev->wq, &waiter->wait);
 	spin_unlock(&ev->lock);
@@ -948,8 +958,7 @@ int kfd_wait_on_events(struct kfd_process *p,
 			goto out_unlock;
 		}
 
-		ret = init_event_waiter(p, &event_waiters[i],
-					event_data.event_id);
+		ret = init_event_waiter(p, &event_waiters[i], &event_data);
 		if (ret)
 			goto out_unlock;
 	}
-- 
GitLab


From 973fddea6f3e0d2f623f13fbd5d4d2b775e157f0 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Thu, 8 Jun 2023 11:22:03 -0400
Subject: [PATCH 1757/1791] drm/amdkfd: update user space last_event_age

Update user space last_event_age when event age is enabled.
It is only for KFD_EVENT_TYPE_SIGNAL which is checked by user space.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_events.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 545179535d21a..8081a9408006e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -865,22 +865,29 @@ static int copy_signaled_event_data(uint32_t num_events,
 		struct kfd_event_waiter *event_waiters,
 		struct kfd_event_data __user *data)
 {
-	struct kfd_hsa_memory_exception_data *src;
-	struct kfd_hsa_memory_exception_data __user *dst;
+	void *src;
+	void __user *dst;
 	struct kfd_event_waiter *waiter;
 	struct kfd_event *event;
-	uint32_t i;
+	uint32_t i, size = 0;
 
 	for (i = 0; i < num_events; i++) {
 		waiter = &event_waiters[i];
 		event = waiter->event;
 		if (!event)
 			return -EINVAL; /* event was destroyed */
-		if (waiter->activated && event->type == KFD_EVENT_TYPE_MEMORY) {
-			dst = &data[i].memory_exception_data;
-			src = &event->memory_exception_data;
-			if (copy_to_user(dst, src,
-				sizeof(struct kfd_hsa_memory_exception_data)))
+		if (waiter->activated) {
+			if (event->type == KFD_EVENT_TYPE_MEMORY) {
+				dst = &data[i].memory_exception_data;
+				src = &event->memory_exception_data;
+				size = sizeof(struct kfd_hsa_memory_exception_data);
+			} else if (event->type == KFD_EVENT_TYPE_SIGNAL &&
+				waiter->event_age_enabled) {
+				dst = &data[i].signal_event_data.last_event_age;
+				src = &event->event_age;
+				size = sizeof(u64);
+			}
+			if (size && copy_to_user(dst, src, size))
 				return -EFAULT;
 		}
 	}
-- 
GitLab


From d297eedf83f5af96751c0da1e4355c19244a55a2 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 7 Jun 2023 13:27:40 -0400
Subject: [PATCH 1758/1791] drm/amdkfd: bump kfd ioctl minor version for event
 age availability

Bump the minor version to declare event age tracking feature is now
available.

In kernel amdgpu driver, kfd_wait_on_events is used to support user
space signal event wait function. For multiple threads waiting on same
event scenery, race condition could occur since some threads after
checking signal condition, before calling kfd_wait_on_events, the
event interrupt could be fired and wake up other thread which are
sleeping on this event. Then those threads could fall into sleep
without waking up again. Adding event age tracking in both kernel and
user mode, will help avoiding this race condition.

Proposed ROCT-Thunk-Interface:
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/efdbf6cfbc026bd68ac3c35d00dacf84370eb81e
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/1820ae0a2db85b6f584611dc0cde1a00e7c22915

Proposed ROCR-Runtime:
https://github.com/RadeonOpenCompute/ROCR-Runtime/compare/master...zhums:ROCR-Runtime:new_event_wait_review
https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/e1f5bdb88eb882ac798aeca2c00ea3fbb2dba459
https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/7d26afd14107b5c2a754c1a3f415d89f3aabb503

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/uapi/linux/kfd_ioctl.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index 93f1c0bc5caff..eeb2fdcbdcb70 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -39,9 +39,10 @@
  * - 1.11 - Add unified memory for ctx save/restore area
  * - 1.12 - Add DMA buf export ioctl
  * - 1.13 - Add debugger API
+ * - 1.14 - Update kfd_event_data
  */
 #define KFD_IOCTL_MAJOR_VERSION 1
-#define KFD_IOCTL_MINOR_VERSION 13
+#define KFD_IOCTL_MINOR_VERSION 14
 
 struct kfd_ioctl_get_version_args {
 	__u32 major_version;	/* from KFD */
-- 
GitLab


From 0e41639d9a46b0285cd6381482037095f196d516 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 13 Jun 2023 16:18:46 +0530
Subject: [PATCH 1759/1791] drm/amdgpu: Remove unused NBIO interface

Set compute partition mode interface in NBIO is no longer used. Remove
the only implementation from NBIO v7.9

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h |  2 --
 drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c   | 14 --------------
 2 files changed, 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
index 095aecfb201ed..8ab8ae01f87c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h
@@ -99,8 +99,6 @@ struct amdgpu_nbio_funcs {
 	int (*get_compute_partition_mode)(struct amdgpu_device *adev);
 	u32 (*get_memory_partition_mode)(struct amdgpu_device *adev,
 					 u32 *supp_modes);
-	void (*set_compute_partition_mode)(struct amdgpu_device *adev,
-					   enum amdgpu_gfx_partition mode);
 };
 
 struct amdgpu_nbio {
diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
index b033935d67498..cd1a02d30420e 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c
@@ -393,19 +393,6 @@ static int nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev)
 	return px;
 }
 
-static void nbio_v7_9_set_compute_partition_mode(struct amdgpu_device *adev,
-					enum amdgpu_gfx_partition mode)
-{
-	u32 tmp;
-
-	/* SPX=0, DPX=1, TPX=2, QPX=3, CPX=4 */
-	tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS);
-	tmp = REG_SET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS,
-			    PARTITION_MODE, mode);
-
-	WREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS, tmp);
-}
-
 static u32 nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev,
 					       u32 *supp_modes)
 {
@@ -461,7 +448,6 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = {
 	.ih_control = nbio_v7_9_ih_control,
 	.remap_hdp_registers = nbio_v7_9_remap_hdp_registers,
 	.get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode,
-	.set_compute_partition_mode = nbio_v7_9_set_compute_partition_mode,
 	.get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode,
 	.init_registers = nbio_v7_9_init_registers,
 };
-- 
GitLab


From d728eda3c59daf2df71f9aae4bb2d3a1eef081da Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Wed, 3 May 2023 13:41:40 -0400
Subject: [PATCH 1760/1791] drm/amdgpu: Enable translate further for GC v9.4.3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To extend UTCL2 reach.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 3ed286b72cae9..67e669e0141cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -2107,6 +2107,7 @@ static int gmc_v9_0_sw_init(void *handle)
 		bitmap_or(adev->vmhubs_mask, adev->vmhubs_mask, &inst_mask, 32);
 
 		amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
+		adev->gmc.translate_further = adev->vm_manager.num_level > 1;
 		break;
 	default:
 		break;
-- 
GitLab


From 5d1c70bb6e40c52ee1ff8aa786389919e6fbb09d Mon Sep 17 00:00:00 2001
From: Philip Yang <Philip.Yang@amd.com>
Date: Tue, 6 Jun 2023 15:46:33 -0400
Subject: [PATCH 1761/1791] drm/amdgpu: Increase hmm range get pages timeout

If hmm_range_fault returns -EBUSY, we should call hmm_range_fault again
to validate the remaining pages. On one system with NUMA auto balancing
enabled, hmm_range_fault takes 6 seconds for 1GB range because CPU
migrate the range one page at a time. To be safe, increase timeout value
to 1 second for 128MB range.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
index 2dadcfe43d03d..081267161d401 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
@@ -190,8 +190,8 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 		pr_debug("hmm range: start = 0x%lx, end = 0x%lx",
 			hmm_range->start, hmm_range->end);
 
-		/* Assuming 512MB takes maxmium 1 second to fault page address */
-		timeout = max((hmm_range->end - hmm_range->start) >> 29, 1UL);
+		/* Assuming 128MB takes maximum 1 second to fault page address */
+		timeout = max((hmm_range->end - hmm_range->start) >> 27, 1UL);
 		timeout *= HMM_RANGE_DEFAULT_TIMEOUT;
 		timeout = jiffies + msecs_to_jiffies(timeout);
 
-- 
GitLab


From 72f1de49ffb90b29748284f27f1d6b829ab1de95 Mon Sep 17 00:00:00 2001
From: Wayne Lin <Wayne.Lin@amd.com>
Date: Mon, 17 Apr 2023 17:08:12 +0800
Subject: [PATCH 1762/1791] drm/dp_mst: Clear MSG_RDY flag before sending new
 message

[Why]
The sequence for collecting down_reply from source perspective should
be:

Request_n->repeat (get partial reply of Request_n->clear message ready
flag to ack DPRX that the message is received) till all partial
replies for Request_n are received->new Request_n+1.

Now there is chance that drm_dp_mst_hpd_irq() will fire new down
request in the tx queue when the down reply is incomplete. Source is
restricted to generate interveleaved message transactions so we should
avoid it.

Also, while assembling partial reply packets, reading out DPCD DOWN_REP
Sideband MSG buffer + clearing DOWN_REP_MSG_RDY flag should be
wrapped up as a complete operation for reading out a reply packet.
Kicking off a new request before clearing DOWN_REP_MSG_RDY flag might
be risky. e.g. If the reply of the new request has overwritten the
DPRX DOWN_REP Sideband MSG buffer before source writing one to clear
DOWN_REP_MSG_RDY flag, source then unintentionally flushes the reply
for the new request. Should handle the up request in the same way.

[How]
Separete drm_dp_mst_hpd_irq() into 2 steps. After acking the MST IRQ
event, driver calls drm_dp_mst_hpd_irq_send_new_request() and might
trigger drm_dp_mst_kick_tx() only when there is no on going message
transaction.

Changes since v1:
* Reworked on review comments received
-> Adjust the fix to let driver explicitly kick off new down request
when mst irq event is handled and acked
-> Adjust the commit message

Changes since v2:
* Adjust the commit message
* Adjust the naming of the divided 2 functions and add a new input
  parameter "ack".
* Adjust code flow as per review comments.

Changes since v3:
* Update the function description of drm_dp_mst_hpd_irq_handle_event

Changes since v4:
* Change ack of drm_dp_mst_hpd_irq_handle_event() to be an array align
  the size of esi[]

Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 32 +++++------
 drivers/gpu/drm/display/drm_dp_mst_topology.c | 54 ++++++++++++++++---
 drivers/gpu/drm/i915/display/intel_dp.c       |  7 +--
 drivers/gpu/drm/nouveau/dispnv50/disp.c       | 12 +++--
 include/drm/display/drm_dp_mst_helper.h       |  7 ++-
 5 files changed, 81 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 8964bb87e0d56..514f6785a0209 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3267,6 +3267,7 @@ static void dm_handle_mst_sideband_msg(struct amdgpu_dm_connector *aconnector)
 
 	while (dret == dpcd_bytes_to_read &&
 		process_count < max_process_count) {
+		u8 ack[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = {};
 		u8 retry;
 		dret = 0;
 
@@ -3275,28 +3276,29 @@ static void dm_handle_mst_sideband_msg(struct amdgpu_dm_connector *aconnector)
 		DRM_DEBUG_DRIVER("ESI %02x %02x %02x\n", esi[0], esi[1], esi[2]);
 		/* handle HPD short pulse irq */
 		if (aconnector->mst_mgr.mst_state)
-			drm_dp_mst_hpd_irq(
-				&aconnector->mst_mgr,
-				esi,
-				&new_irq_handled);
+			drm_dp_mst_hpd_irq_handle_event(&aconnector->mst_mgr,
+							esi,
+							ack,
+							&new_irq_handled);
 
 		if (new_irq_handled) {
 			/* ACK at DPCD to notify down stream */
-			const int ack_dpcd_bytes_to_write =
-				dpcd_bytes_to_read - 1;
-
 			for (retry = 0; retry < 3; retry++) {
-				u8 wret;
-
-				wret = drm_dp_dpcd_write(
-					&aconnector->dm_dp_aux.aux,
-					dpcd_addr + 1,
-					&esi[1],
-					ack_dpcd_bytes_to_write);
-				if (wret == ack_dpcd_bytes_to_write)
+				ssize_t wret;
+
+				wret = drm_dp_dpcd_writeb(&aconnector->dm_dp_aux.aux,
+							  dpcd_addr + 1,
+							  ack[1]);
+				if (wret == 1)
 					break;
 			}
 
+			if (retry == 3) {
+				DRM_ERROR("Failed to ack MST event.\n");
+				return;
+			}
+
+			drm_dp_mst_hpd_irq_send_new_request(&aconnector->mst_mgr);
 			/* check if there is new irq to be handled */
 			dret = drm_dp_dpcd_read(
 				&aconnector->dm_dp_aux.aux,
diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c
index be71be95b706c..8fe7b635e5bb2 100644
--- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
@@ -4053,17 +4053,28 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr)
 }
 
 /**
- * drm_dp_mst_hpd_irq() - MST hotplug IRQ notify
+ * drm_dp_mst_hpd_irq_handle_event() - MST hotplug IRQ handle MST event
  * @mgr: manager to notify irq for.
  * @esi: 4 bytes from SINK_COUNT_ESI
+ * @ack: 4 bytes used to ack events starting from SINK_COUNT_ESI
  * @handled: whether the hpd interrupt was consumed or not
  *
- * This should be called from the driver when it detects a short IRQ,
+ * This should be called from the driver when it detects a HPD IRQ,
  * along with the value of the DEVICE_SERVICE_IRQ_VECTOR_ESI0. The
- * topology manager will process the sideband messages received as a result
- * of this.
+ * topology manager will process the sideband messages received
+ * as indicated in the DEVICE_SERVICE_IRQ_VECTOR_ESI0 and set the
+ * corresponding flags that Driver has to ack the DP receiver later.
+ *
+ * Note that driver shall also call
+ * drm_dp_mst_hpd_irq_send_new_request() if the 'handled' is set
+ * after calling this function, to try to kick off a new request in
+ * the queue if the previous message transaction is completed.
+ *
+ * See also:
+ * drm_dp_mst_hpd_irq_send_new_request()
  */
-int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handled)
+int drm_dp_mst_hpd_irq_handle_event(struct drm_dp_mst_topology_mgr *mgr, const u8 *esi,
+				    u8 *ack, bool *handled)
 {
 	int ret = 0;
 	int sc;
@@ -4078,18 +4089,47 @@ int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handl
 	if (esi[1] & DP_DOWN_REP_MSG_RDY) {
 		ret = drm_dp_mst_handle_down_rep(mgr);
 		*handled = true;
+		ack[1] |= DP_DOWN_REP_MSG_RDY;
 	}
 
 	if (esi[1] & DP_UP_REQ_MSG_RDY) {
 		ret |= drm_dp_mst_handle_up_req(mgr);
 		*handled = true;
+		ack[1] |= DP_UP_REQ_MSG_RDY;
 	}
 
-	drm_dp_mst_kick_tx(mgr);
 	return ret;
 }
-EXPORT_SYMBOL(drm_dp_mst_hpd_irq);
+EXPORT_SYMBOL(drm_dp_mst_hpd_irq_handle_event);
 
+/**
+ * drm_dp_mst_hpd_irq_send_new_request() - MST hotplug IRQ kick off new request
+ * @mgr: manager to notify irq for.
+ *
+ * This should be called from the driver when mst irq event is handled
+ * and acked. Note that new down request should only be sent when
+ * previous message transaction is completed. Source is not supposed to generate
+ * interleaved message transactions.
+ */
+void drm_dp_mst_hpd_irq_send_new_request(struct drm_dp_mst_topology_mgr *mgr)
+{
+	struct drm_dp_sideband_msg_tx *txmsg;
+	bool kick = true;
+
+	mutex_lock(&mgr->qlock);
+	txmsg = list_first_entry_or_null(&mgr->tx_msg_downq,
+					 struct drm_dp_sideband_msg_tx, next);
+	/* If last transaction is not completed yet*/
+	if (!txmsg ||
+	    txmsg->state == DRM_DP_SIDEBAND_TX_START_SEND ||
+	    txmsg->state == DRM_DP_SIDEBAND_TX_SENT)
+		kick = false;
+	mutex_unlock(&mgr->qlock);
+
+	if (kick)
+		drm_dp_mst_kick_tx(mgr);
+}
+EXPORT_SYMBOL(drm_dp_mst_hpd_irq_send_new_request);
 /**
  * drm_dp_mst_detect_port() - get connection status for an MST port
  * @connector: DRM connector for this port
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index f4192fda1a76e..c4593d025154e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -4069,9 +4069,7 @@ intel_dp_mst_hpd_irq(struct intel_dp *intel_dp, u8 *esi, u8 *ack)
 {
 	bool handled = false;
 
-	drm_dp_mst_hpd_irq(&intel_dp->mst_mgr, esi, &handled);
-	if (handled)
-		ack[1] |= esi[1] & (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY);
+	drm_dp_mst_hpd_irq_handle_event(&intel_dp->mst_mgr, esi, ack, &handled);
 
 	if (esi[1] & DP_CP_IRQ) {
 		intel_hdcp_handle_cp_irq(intel_dp->attached_connector);
@@ -4146,6 +4144,9 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp)
 
 		if (!intel_dp_ack_sink_irq_esi(intel_dp, ack))
 			drm_dbg_kms(&i915->drm, "Failed to ack ESI\n");
+
+		if (ack[1] & (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY))
+			drm_dp_mst_hpd_irq_send_new_request(&intel_dp->mst_mgr);
 	}
 
 	return link_ok;
diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c
index 9b6824f6b9e4b..42e1665ba11a3 100644
--- a/drivers/gpu/drm/nouveau/dispnv50/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c
@@ -1359,22 +1359,26 @@ nv50_mstm_service(struct nouveau_drm *drm,
 	u8 esi[8] = {};
 
 	while (handled) {
+		u8 ack[8] = {};
+
 		rc = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8);
 		if (rc != 8) {
 			ret = false;
 			break;
 		}
 
-		drm_dp_mst_hpd_irq(&mstm->mgr, esi, &handled);
+		drm_dp_mst_hpd_irq_handle_event(&mstm->mgr, esi, ack, &handled);
 		if (!handled)
 			break;
 
-		rc = drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1],
-				       3);
-		if (rc != 3) {
+		rc = drm_dp_dpcd_writeb(aux, DP_SINK_COUNT_ESI + 1, ack[1]);
+
+		if (rc != 1) {
 			ret = false;
 			break;
 		}
+
+		drm_dp_mst_hpd_irq_send_new_request(&mstm->mgr);
 	}
 
 	if (!ret)
diff --git a/include/drm/display/drm_dp_mst_helper.h b/include/drm/display/drm_dp_mst_helper.h
index f962e97880b4f..ed5c9660563c4 100644
--- a/include/drm/display/drm_dp_mst_helper.h
+++ b/include/drm/display/drm_dp_mst_helper.h
@@ -810,8 +810,11 @@ void drm_dp_mst_topology_mgr_destroy(struct drm_dp_mst_topology_mgr *mgr);
 bool drm_dp_read_mst_cap(struct drm_dp_aux *aux, const u8 dpcd[DP_RECEIVER_CAP_SIZE]);
 int drm_dp_mst_topology_mgr_set_mst(struct drm_dp_mst_topology_mgr *mgr, bool mst_state);
 
-int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handled);
-
+int drm_dp_mst_hpd_irq_handle_event(struct drm_dp_mst_topology_mgr *mgr,
+				    const u8 *esi,
+				    u8 *ack,
+				    bool *handled);
+void drm_dp_mst_hpd_irq_send_new_request(struct drm_dp_mst_topology_mgr *mgr);
 
 int
 drm_dp_mst_detect_port(struct drm_connector *connector,
-- 
GitLab


From 0b78be614c503ec03636f82cea52ad735c5085e7 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 13 Jun 2023 03:09:40 +0300
Subject: [PATCH 1763/1791] drm/msm/dpu: fix sc7280 and sc7180 PINGPONG done
 interrupts

During IRQ conversion we have lost the PP_DONE interrupts for sc7280
platform. This was left unnoticed, because this interrupt is only used
for CMD outputs and probably no sc7[12]80 systems use DSI CMD panels.

Fixes: 667e9985ee24 ("drm/msm/dpu: replace IRQ lookup with the data in hw catalog")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/542175/
Link: https://lore.kernel.org/r/20230613001004.3426676-2-dmitry.baryshkov@linaro.org
---
 .../drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h   |  8 ++++++--
 .../drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h   | 16 ++++++++++++----
 2 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
index 0b05da2592c08..67566b07195a2 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h
@@ -80,8 +80,12 @@ static const struct dpu_dspp_cfg sc7180_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sc7180_pp[] = {
-	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, -1, -1),
-	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, -1, -1),
+	PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
+			-1),
+	PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
+			-1),
 };
 
 static const struct dpu_intf_cfg sc7180_intf[] = {
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
index 9cd361be28faa..900fee410e113 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h
@@ -88,10 +88,18 @@ static const struct dpu_dspp_cfg sc7280_dspp[] = {
 };
 
 static const struct dpu_pingpong_cfg sc7280_pp[] = {
-	PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, 0, sc7280_pp_sblk, -1, -1),
-	PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, 0, sc7280_pp_sblk, -1, -1),
-	PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, 0, sc7280_pp_sblk, -1, -1),
-	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, -1, -1),
+	PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, 0, sc7280_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8),
+			-1),
+	PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, 0, sc7280_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9),
+			-1),
+	PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, 0, sc7280_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10),
+			-1),
+	PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk,
+			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11),
+			-1),
 };
 
 /* NOTE: sc7280 only has one DSC hard slice encoder */
-- 
GitLab


From 9a6c13b847d61b0c3796820ca6e976789df59cd8 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 13 Jun 2023 03:09:41 +0300
Subject: [PATCH 1764/1791] drm/msm/dpu: correct MERGE_3D length

Each MERGE_3D block has just two registers. Correct the block length
accordingly.

Fixes: 4369c93cf36b ("drm/msm/dpu: initial support for merge3D hardware block")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/542177/
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Link: https://lore.kernel.org/r/20230613001004.3426676-3-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 36ba3f58dcdf0..0de507d4d7b7a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -508,7 +508,7 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = {
 #define MERGE_3D_BLK(_name, _id, _base) \
 	{\
 	.name = _name, .id = _id, \
-	.base = _base, .len = 0x100, \
+	.base = _base, .len = 0x8, \
 	.features = MERGE_3D_SM8150_MASK, \
 	.sblk = NULL \
 	}
-- 
GitLab


From c7c4afd943cc98a123d0ee4f482bd9ea96ff18db Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Tue, 13 Jun 2023 03:09:42 +0300
Subject: [PATCH 1765/1791] drm/msm/dpu: remove unused INTF_NONE interfaces

sm6115, sm6375 and qcm2290 do not have INTF_0. Drop corresponding
interface definitions.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/542180/
Link: https://lore.kernel.org/r/20230613001004.3426676-4-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h  | 1 -
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h | 1 -
 drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h  | 1 -
 3 files changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
index ba9de008519b9..031fc8dae3c69 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h
@@ -66,7 +66,6 @@ static const struct dpu_pingpong_cfg sm6115_pp[] = {
 };
 
 static const struct dpu_intf_cfg sm6115_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
 	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
index 92ac348eea6b0..f2808098af399 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h
@@ -63,7 +63,6 @@ static const struct dpu_pingpong_cfg qcm2290_pp[] = {
 };
 
 static const struct dpu_intf_cfg qcm2290_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
 	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 			DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
index d7aae45e3e663..241fa6746674d 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h
@@ -71,7 +71,6 @@ static const struct dpu_dsc_cfg sm6375_dsc[] = {
 };
 
 static const struct dpu_intf_cfg sm6375_intf[] = {
-	INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0),
 	INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK,
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26),
 		DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27),
-- 
GitLab


From e2fd7dda3bb867012836500f1816a4837ddb5079 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 15 Jun 2023 01:44:01 +0300
Subject: [PATCH 1766/1791] drm/msm/dsi: dsi_host: drop unused clocks

Several source clocks are not used anymore, so stop handling them.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/542558/
Link: https://lore.kernel.org/r/20230614224402.296825-1-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 33 ------------------------------
 1 file changed, 33 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index fb1d3a25765f3..eaee621aa6c86 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -118,8 +118,6 @@ struct msm_dsi_host {
 	struct clk *byte_clk;
 	struct clk *esc_clk;
 	struct clk *pixel_clk;
-	struct clk *byte_clk_src;
-	struct clk *pixel_clk_src;
 	struct clk *byte_intf_clk;
 
 	unsigned long byte_clk_rate;
@@ -129,8 +127,6 @@ struct msm_dsi_host {
 
 	/* DSI v2 specific clocks */
 	struct clk *src_clk;
-	struct clk *esc_clk_src;
-	struct clk *dsi_clk_src;
 
 	unsigned long src_clk_rate;
 
@@ -267,21 +263,6 @@ int dsi_clk_init_v2(struct msm_dsi_host *msm_host)
 		return ret;
 	}
 
-	msm_host->esc_clk_src = clk_get_parent(msm_host->esc_clk);
-	if (!msm_host->esc_clk_src) {
-		ret = -ENODEV;
-		pr_err("%s: can't get esc clock parent. ret=%d\n",
-			__func__, ret);
-		return ret;
-	}
-
-	msm_host->dsi_clk_src = clk_get_parent(msm_host->src_clk);
-	if (!msm_host->dsi_clk_src) {
-		ret = -ENODEV;
-		pr_err("%s: can't get src clock parent. ret=%d\n",
-			__func__, ret);
-	}
-
 	return ret;
 }
 
@@ -346,20 +327,6 @@ static int dsi_clk_init(struct msm_dsi_host *msm_host)
 		goto exit;
 	}
 
-	msm_host->byte_clk_src = clk_get_parent(msm_host->byte_clk);
-	if (IS_ERR(msm_host->byte_clk_src)) {
-		ret = PTR_ERR(msm_host->byte_clk_src);
-		pr_err("%s: can't find byte_clk clock. ret=%d\n", __func__, ret);
-		goto exit;
-	}
-
-	msm_host->pixel_clk_src = clk_get_parent(msm_host->pixel_clk);
-	if (IS_ERR(msm_host->pixel_clk_src)) {
-		ret = PTR_ERR(msm_host->pixel_clk_src);
-		pr_err("%s: can't find pixel_clk clock. ret=%d\n", __func__, ret);
-		goto exit;
-	}
-
 	if (cfg_hnd->ops->clk_init_ver)
 		ret = cfg_hnd->ops->clk_init_ver(msm_host);
 exit:
-- 
GitLab


From 452c46ccf603ab181e7d0866f3137e3ae62b5f49 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Date: Thu, 15 Jun 2023 01:44:02 +0300
Subject: [PATCH 1767/1791] drm/msm/dsi: split dsi_ctrl_config() function

It makes no sense to pass NULL parameters to dsi_ctrl_config() in the
disable case. Split dsi_ctrl_config() into enable and disable parts and
drop unused params.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Patchwork: https://patchwork.freedesktop.org/patch/542559/
Link: https://lore.kernel.org/r/20230614224402.296825-2-dmitry.baryshkov@linaro.org
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index eaee621aa6c86..3f6dfb4f9d5a6 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -705,7 +705,12 @@ static inline enum dsi_cmd_dst_format dsi_get_cmd_fmt(
 	}
 }
 
-static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable,
+static void dsi_ctrl_disable(struct msm_dsi_host *msm_host)
+{
+	dsi_write(msm_host, REG_DSI_CTRL, 0);
+}
+
+static void dsi_ctrl_enable(struct msm_dsi_host *msm_host,
 			struct msm_dsi_phy_shared_timings *phy_shared_timings, struct msm_dsi_phy *phy)
 {
 	u32 flags = msm_host->mode_flags;
@@ -713,11 +718,6 @@ static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable,
 	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	u32 data = 0, lane_ctrl = 0;
 
-	if (!enable) {
-		dsi_write(msm_host, REG_DSI_CTRL, 0);
-		return;
-	}
-
 	if (flags & MIPI_DSI_MODE_VIDEO) {
 		if (flags & MIPI_DSI_MODE_VIDEO_HSE)
 			data |= DSI_VID_CFG0_PULSE_MODE_HSA_HE;
@@ -802,7 +802,7 @@ static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable,
 	if (!(flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)) {
 		lane_ctrl = dsi_read(msm_host, REG_DSI_LANE_CTRL);
 
-		if (msm_dsi_phy_set_continuous_clock(phy, enable))
+		if (msm_dsi_phy_set_continuous_clock(phy, true))
 			lane_ctrl &= ~DSI_LANE_CTRL_HS_REQ_SEL_PHY;
 
 		dsi_write(msm_host, REG_DSI_LANE_CTRL,
@@ -2358,7 +2358,7 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host,
 
 	dsi_timing_setup(msm_host, is_bonded_dsi);
 	dsi_sw_reset(msm_host);
-	dsi_ctrl_config(msm_host, true, phy_shared_timings, phy);
+	dsi_ctrl_enable(msm_host, phy_shared_timings, phy);
 
 	if (msm_host->disp_en_gpio)
 		gpiod_set_value(msm_host->disp_en_gpio, 1);
@@ -2390,7 +2390,7 @@ int msm_dsi_host_power_off(struct mipi_dsi_host *host)
 		goto unlock_ret;
 	}
 
-	dsi_ctrl_config(msm_host, false, NULL, NULL);
+	dsi_ctrl_disable(msm_host);
 
 	if (msm_host->disp_en_gpio)
 		gpiod_set_value(msm_host->disp_en_gpio, 0);
-- 
GitLab


From a770dc6105fbd27a8c08557c8ef720c6aa878b4d Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:42 +0200
Subject: [PATCH 1768/1791] dt-bindings: display/msm: gpu: Document GMU
 wrapper-equipped A6xx

The "GMU Wrapper" is Qualcomm's name for "let's treat the GPU blocks
we'd normally assign to the GMU as if they were a part of the GMU, even
though they are not". It's a (good) software representation of the GMU_CX
and GMU_GX register spaces within the GPUSS that helps us programatically
treat these de-facto GMU-less parts in a way that's very similar to their
GMU-equipped cousins, massively saving up on code duplication.

The "wrapper" register space was specifically designed to mimic the layout
of a real GMU, though it rather obviously does not have the M3 core et al.

GMU wrapper-equipped A6xx GPUs require clocks and clock-names to be
specified under the GPU node, just like their older cousins. Account
for that.

Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542748/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 .../devicetree/bindings/display/msm/gpu.yaml  | 61 ++++++++++++++++---
 1 file changed, 52 insertions(+), 9 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml
index 5dabe7b6794b7..58ca8912a8c34 100644
--- a/Documentation/devicetree/bindings/display/msm/gpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml
@@ -36,10 +36,7 @@ properties:
 
   reg-names:
     minItems: 1
-    items:
-      - const: kgsl_3d0_reg_memory
-      - const: cx_mem
-      - const: cx_dbgc
+    maxItems: 3
 
   interrupts:
     maxItems: 1
@@ -157,16 +154,62 @@ allOf:
       required:
         - clocks
         - clock-names
+
   - if:
       properties:
         compatible:
           contains:
-            pattern: '^qcom,adreno-6[0-9][0-9]\.[0-9]$'
-
-    then: # Since Adreno 6xx series clocks should be defined in GMU
+            enum:
+              - qcom,adreno-610.0
+              - qcom,adreno-619.1
+    then:
       properties:
-        clocks: false
-        clock-names: false
+        clocks:
+          minItems: 6
+          maxItems: 6
+
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+            - const: alt_mem_iface
+              description: GPU Alternative Memory Interface clock
+            - const: gmu
+              description: CX GMU clock
+            - const: xo
+              description: GPUCC clocksource clock
+
+        reg-names:
+          minItems: 1
+          items:
+            - const: kgsl_3d0_reg_memory
+            - const: cx_dbgc
+
+      required:
+        - clocks
+        - clock-names
+    else:
+      if:
+        properties:
+          compatible:
+            contains:
+              pattern: '^qcom,adreno-6[0-9][0-9]\.[0-9]$'
+
+      then: # Starting with A6xx, the clocks are usually defined in the GMU node
+        properties:
+          clocks: false
+          clock-names: false
+
+          reg-names:
+            minItems: 1
+            items:
+              - const: kgsl_3d0_reg_memory
+              - const: cx_mem
+              - const: cx_dbgc
 
 examples:
   - |
-- 
GitLab


From 63204be2806b09ab75e03e42c477f69bfd050efc Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:43 +0200
Subject: [PATCH 1769/1791] dt-bindings: display/msm/gmu: Add GMU wrapper

The "GMU Wrapper" is Qualcomm's name for "let's treat the GPU blocks
we'd normally assign to the GMU as if they were a part of the GMU, even
though they are not". It's a (good) software representation of the GMU_CX
and GMU_GX register spaces within the GPUSS that helps us programatically
treat these de-facto GMU-less parts in a way that's very similar to their
GMU-equipped cousins, massively saving up on code duplication.

The "wrapper" register space was specifically designed to mimic the layout
of a real GMU, though it rather obviously does not have the M3 core et al.

To sum it all up, the GMU wrapper is essentially a register space within
the GPU, which Linux sees as a dumbed-down regular GMU: there's no clocks,
interrupts, multiple reg spaces, iommus and OPP. Document it.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542750/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 .../devicetree/bindings/display/msm/gmu.yaml  | 50 ++++++++++++++-----
 1 file changed, 38 insertions(+), 12 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/gmu.yaml b/Documentation/devicetree/bindings/display/msm/gmu.yaml
index f31a26305ca9d..5fc4106110ad2 100644
--- a/Documentation/devicetree/bindings/display/msm/gmu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gmu.yaml
@@ -19,16 +19,18 @@ description: |
 
 properties:
   compatible:
-    items:
-      - pattern: '^qcom,adreno-gmu-6[0-9][0-9]\.[0-9]$'
-      - const: qcom,adreno-gmu
+    oneOf:
+      - items:
+          - pattern: '^qcom,adreno-gmu-6[0-9][0-9]\.[0-9]$'
+          - const: qcom,adreno-gmu
+      - const: qcom,adreno-gmu-wrapper
 
   reg:
-    minItems: 3
+    minItems: 1
     maxItems: 4
 
   reg-names:
-    minItems: 3
+    minItems: 1
     maxItems: 4
 
   clocks:
@@ -44,7 +46,6 @@ properties:
       - description: GMU HFI interrupt
       - description: GMU interrupt
 
-
   interrupt-names:
     items:
       - const: hfi
@@ -72,14 +73,8 @@ required:
   - compatible
   - reg
   - reg-names
-  - clocks
-  - clock-names
-  - interrupts
-  - interrupt-names
   - power-domains
   - power-domain-names
-  - iommus
-  - operating-points-v2
 
 additionalProperties: false
 
@@ -218,6 +213,28 @@ allOf:
             - const: axi
             - const: memnoc
 
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: qcom,adreno-gmu-wrapper
+    then:
+      properties:
+        reg:
+          items:
+            - description: GMU wrapper register space
+        reg-names:
+          items:
+            - const: gmu
+    else:
+      required:
+        - clocks
+        - clock-names
+        - interrupts
+        - interrupt-names
+        - iommus
+        - operating-points-v2
+
 examples:
   - |
     #include <dt-bindings/clock/qcom,gpucc-sdm845.h>
@@ -250,3 +267,12 @@ examples:
         iommus = <&adreno_smmu 5>;
         operating-points-v2 = <&gmu_opp_table>;
     };
+
+    gmu_wrapper: gmu@596a000 {
+        compatible = "qcom,adreno-gmu-wrapper";
+        reg = <0x0596a000 0x30000>;
+        reg-names = "gmu";
+        power-domains = <&gpucc GPU_CX_GDSC>,
+                        <&gpucc GPU_GX_GDSC>;
+        power-domain-names = "cx", "gx";
+    };
-- 
GitLab


From 87cc0d0a4204b437c4401864350e7c4ee7710f3f Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:44 +0200
Subject: [PATCH 1770/1791] drm/msm/adreno: Use adreno_is_revn for A690

The adreno_is_revn rework came at the same time as A690 introduction
and that resulted in it not covering all cases. Fix it.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542754/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/adreno_gpu.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 9a7626c7ac4d4..5a26c8a2de7cb 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -286,7 +286,7 @@ static inline int adreno_is_a660(const struct adreno_gpu *gpu)
 
 static inline int adreno_is_a690(const struct adreno_gpu *gpu)
 {
-	return gpu->revn == 690;
+	return adreno_is_revn(gpu, 690);
 };
 
 /* check for a615, a616, a618, a619 or any derivatives */
-- 
GitLab


From bd31afe010ea72de2323c9d95b540f08aeada732 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:45 +0200
Subject: [PATCH 1771/1791] drm/msm/a6xx: Remove static keyword from sptprac
 en/disable functions

These two will be reused by at least A619_holi in the non-gmu
paths. Turn them non-static them to make it possible.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542751/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 4 ++--
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 2 ++
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 8914992378f21..a6fa273d700ee 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -354,7 +354,7 @@ void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state)
 }
 
 /* Enable CPU control of SPTP power power collapse */
-static int a6xx_sptprac_enable(struct a6xx_gmu *gmu)
+int a6xx_sptprac_enable(struct a6xx_gmu *gmu)
 {
 	int ret;
 	u32 val;
@@ -376,7 +376,7 @@ static int a6xx_sptprac_enable(struct a6xx_gmu *gmu)
 }
 
 /* Disable CPU control of SPTP power power collapse */
-static void a6xx_sptprac_disable(struct a6xx_gmu *gmu)
+void a6xx_sptprac_disable(struct a6xx_gmu *gmu)
 {
 	u32 val;
 	int ret;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 4759a8ce51e46..236f81a43caa6 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -193,5 +193,7 @@ int a6xx_hfi_set_freq(struct a6xx_gmu *gmu, int index);
 
 bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu);
 bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu);
+void a6xx_sptprac_disable(struct a6xx_gmu *gmu);
+int a6xx_sptprac_enable(struct a6xx_gmu *gmu);
 
 #endif
-- 
GitLab


From ce8f1381fd4960634e5f2a96bdf6f4eb177bca6e Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:46 +0200
Subject: [PATCH 1772/1791] drm/msm/a6xx: Move force keepalive vote removal to
 a6xx_gmu_force_off()

As pointed out by Akhil during the review process of GMU wrapper
introduction [1], it makes sense to move this write into the function
that's responsible for forcibly shutting the GMU off.

It is also very convenient to move this to GMU-specific code, so that
it does not have to be guarded by an if-condition to avoid calling it
on GMU wrapper targets.

Move the write to the aforementioned a6xx_gmu_force_off() to achieve
that. No effective functional change.

[1] https://lore.kernel.org/linux-arm-msm/20230501194022.GA18382@akhilpo-linux.qualcomm.com/

Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542752/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 6 ++++++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index a6fa273d700ee..32852c161aabc 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -912,6 +912,12 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu)
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 	struct msm_gpu *gpu = &adreno_gpu->base;
 
+	/*
+	 * Turn off keep alive that might have been enabled by the hang
+	 * interrupt
+	 */
+	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 0);
+
 	/* Flush all the queues */
 	a6xx_hfi_stop(gmu);
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index ab5c446e4409b..eebb4bc7c0f9d 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1382,12 +1382,6 @@ static void a6xx_recover(struct msm_gpu *gpu)
 	/* Halt SQE first */
 	gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3);
 
-	/*
-	 * Turn off keep alive that might have been enabled by the hang
-	 * interrupt
-	 */
-	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 0);
-
 	pm_runtime_dont_use_autosuspend(&gpu->pdev->dev);
 
 	/* active_submit won't change until we make a submission */
-- 
GitLab


From 6e332c99d1aac34660c22fd1dffc5d6f83d330fa Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:47 +0200
Subject: [PATCH 1773/1791] drm/msm/a6xx: Move
 a6xx_bus_clear_pending_transactions to a6xx_gpu

This function is responsible for telling the GPU to halt transactions
on all of its relevant buses, drain them and leave them in a predictable
state, so that the GPU can be e.g. reset cleanly.

Move the function to a6xx_gpu.c, remove the static keyword and add a
prototype in a6xx_gpu.h to accomodate for the move.

Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542762/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 37 ---------------------------
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 36 ++++++++++++++++++++++++++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h |  2 ++
 3 files changed, 38 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 32852c161aabc..6402544f6849b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -868,43 +868,6 @@ static void a6xx_gmu_rpmh_off(struct a6xx_gmu *gmu)
 		(val & 1), 100, 1000);
 }
 
-#define GBIF_CLIENT_HALT_MASK             BIT(0)
-#define GBIF_ARB_HALT_MASK                BIT(1)
-
-static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu,
-		bool gx_off)
-{
-	struct msm_gpu *gpu = &adreno_gpu->base;
-
-	if (!a6xx_has_gbif(adreno_gpu)) {
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
-		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
-								0xf) == 0xf);
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
-
-		return;
-	}
-
-	if (gx_off) {
-		/* Halt the gx side of GBIF */
-		gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 1);
-		spin_until(gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT_ACK) & 1);
-	}
-
-	/* Halt new client requests on GBIF */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK);
-	spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) &
-			(GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK);
-
-	/* Halt all AXI requests on GBIF */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK);
-	spin_until((gpu_read(gpu,  REG_A6XX_GBIF_HALT_ACK) &
-			(GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK);
-
-	/* The GBIF halt needs to be explicitly cleared */
-	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
-}
-
 /* Force the GMU off in case it isn't responsive */
 static void a6xx_gmu_force_off(struct a6xx_gmu *gmu)
 {
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index eebb4bc7c0f9d..a48f4e3a754a0 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1705,6 +1705,42 @@ static void a6xx_llc_slices_init(struct platform_device *pdev,
 		a6xx_gpu->llc_mmio = ERR_PTR(-EINVAL);
 }
 
+#define GBIF_CLIENT_HALT_MASK             BIT(0)
+#define GBIF_ARB_HALT_MASK                BIT(1)
+
+void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off)
+{
+	struct msm_gpu *gpu = &adreno_gpu->base;
+
+	if (!a6xx_has_gbif(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
+		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
+								0xf) == 0xf);
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
+
+		return;
+	}
+
+	if (gx_off) {
+		/* Halt the gx side of GBIF */
+		gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 1);
+		spin_until(gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT_ACK) & 1);
+	}
+
+	/* Halt new client requests on GBIF */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK);
+	spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) &
+			(GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK);
+
+	/* Halt all AXI requests on GBIF */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK);
+	spin_until((gpu_read(gpu,  REG_A6XX_GBIF_HALT_ACK) &
+			(GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK);
+
+	/* The GBIF halt needs to be explicitly cleared */
+	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
+}
+
 static int a6xx_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index eea2e60ce3b7b..9580def06d45d 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -88,4 +88,6 @@ void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state,
 struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu);
 int a6xx_gpu_state_put(struct msm_gpu_state *state);
 
+void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off);
+
 #endif /* __A6XX_GPU_H__ */
-- 
GitLab


From 3773a57d9c55d77ad6615f7235b91fcac0e6c65e Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:48 +0200
Subject: [PATCH 1774/1791] drm/msm/a6xx: Improve
 a6xx_bus_clear_pending_transactions()

Unify the indentation and explain the cryptic 0xF value.

Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542756/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index a48f4e3a754a0..d5bd008c2947d 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1705,17 +1705,18 @@ static void a6xx_llc_slices_init(struct platform_device *pdev,
 		a6xx_gpu->llc_mmio = ERR_PTR(-EINVAL);
 }
 
-#define GBIF_CLIENT_HALT_MASK             BIT(0)
-#define GBIF_ARB_HALT_MASK                BIT(1)
+#define GBIF_CLIENT_HALT_MASK		BIT(0)
+#define GBIF_ARB_HALT_MASK		BIT(1)
+#define VBIF_XIN_HALT_CTRL0_MASK	GENMASK(3, 0)
 
 void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off)
 {
 	struct msm_gpu *gpu = &adreno_gpu->base;
 
 	if (!a6xx_has_gbif(adreno_gpu)) {
-		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf);
+		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, VBIF_XIN_HALT_CTRL0_MASK);
 		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
-								0xf) == 0xf);
+				(VBIF_XIN_HALT_CTRL0_MASK)) == VBIF_XIN_HALT_CTRL0_MASK);
 		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0);
 
 		return;
-- 
GitLab


From 277b967829aad7b24327f246087805078a62076e Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:49 +0200
Subject: [PATCH 1775/1791] drm/msm/a6xx: Add a helper for software-resetting
 the GPU

Introduce a6xx_gpu_sw_reset() in preparation for adding GMU wrapper
GPUs and reuse it in a6xx_gmu_force_off().

This helper, contrary to the original usage in GMU code paths, adds
a readback+delay sequence to ensure that the reset is never deasserted
too quickly due to e.g. OoO execution going crazy.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542758/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c |  3 +--
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 12 ++++++++++++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h |  1 +
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 6402544f6849b..906bed49f27de 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -899,8 +899,7 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu)
 	a6xx_bus_clear_pending_transactions(adreno_gpu, true);
 
 	/* Reset GPU core blocks */
-	gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, 1);
-	udelay(100);
+	a6xx_gpu_sw_reset(gpu, true);
 }
 
 static void a6xx_gmu_set_initial_freq(struct msm_gpu *gpu, struct a6xx_gmu *gmu)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index d5bd008c2947d..b627be3f63609 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1742,6 +1742,18 @@ void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_
 	gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0);
 }
 
+void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert)
+{
+	gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, assert);
+	/* Perform a bogus read and add a brief delay to ensure ordering. */
+	gpu_read(gpu, REG_A6XX_RBBM_SW_RESET_CMD);
+	udelay(1);
+
+	/* The reset line needs to be asserted for at least 100 us */
+	if (assert)
+		udelay(100);
+}
+
 static int a6xx_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 9580def06d45d..aa70390ee1c62 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -89,5 +89,6 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu);
 int a6xx_gpu_state_put(struct msm_gpu_state *state);
 
 void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off);
+void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert);
 
 #endif /* __A6XX_GPU_H__ */
-- 
GitLab


From 05a23a76d35b874dcfab2ad8e8d156d7f24894bd Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:50 +0200
Subject: [PATCH 1776/1791] drm/msm/a6xx: Remove both GBIF and RBBM GBIF halt
 on hw init

Currently we're only deasserting REG_A6XX_RBBM_GBIF_HALT, but we also
need REG_A6XX_GBIF_HALT to be set to 0.

This is typically done automatically on successful GX collapse, but in
case that fails, we should take care of it.

Also, add a memory barrier to ensure it's gone through before jumping
to further initialization.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542760/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index b627be3f63609..7e0d1dfcd993c 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1111,8 +1111,12 @@ static int hw_init(struct msm_gpu *gpu)
 	a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET);
 
 	/* Clear GBIF halt in case GX domain was not collapsed */
-	if (a6xx_has_gbif(adreno_gpu))
+	if (a6xx_has_gbif(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_GBIF_HALT, 0);
 		gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 0);
+		/* Let's make extra sure that the GPU can access the memory.. */
+		mb();
+	}
 
 	gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_CNTL, 0);
 
-- 
GitLab


From df5bb404fa925a1b1fa3893f4976094227fbc1f0 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:51 +0200
Subject: [PATCH 1777/1791] drm/msm/a6xx: Extend and explain UBWC config

Rename lower_bit to hbb_lo and explain what it signifies.
Add explanations (wherever possible to other tunables).

Port setting min_access_length, ubwc_mode and hbb_hi from downstream.

Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542764/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 41 ++++++++++++++++++++-------
 1 file changed, 31 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 7e0d1dfcd993c..8aa4670b4308d 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -887,10 +887,25 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu)
 static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	u32 lower_bit = 2;
-	u32 amsbc = 0;
+	/* Unknown, introduced with A650 family, related to UBWC mode/ver 4 */
 	u32 rgb565_predicator = 0;
+	/* Unknown, introduced with A650 family */
 	u32 uavflagprd_inv = 0;
+	/* Whether the minimum access length is 64 bits */
+	u32 min_acc_len = 0;
+	/* Entirely magic, per-GPU-gen value */
+	u32 ubwc_mode = 0;
+	/*
+	 * The Highest Bank Bit value represents the bit of the highest DDR bank.
+	 * We then subtract 13 from it (13 is the minimum value allowed by hw) and
+	 * write the lowest two bits of the remaining value as hbb_lo and the
+	 * one above it as hbb_hi to the hardware. This should ideally use DRAM
+	 * type detection.
+	 */
+	u32 hbb_hi = 0;
+	u32 hbb_lo = 2;
+	/* Unknown, introduced with A640/680 */
+	u32 amsbc = 0;
 
 	/* a618 is using the hw default values */
 	if (adreno_is_a618(adreno_gpu))
@@ -901,32 +916,38 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
 
 	if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) {
 		/* TODO: get ddr type from bootloader and use 2 for LPDDR4 */
-		lower_bit = 3;
+		hbb_lo = 3;
 		amsbc = 1;
 		rgb565_predicator = 1;
 		uavflagprd_inv = 2;
 	}
 
 	if (adreno_is_a690(adreno_gpu)) {
-		lower_bit = 2;
+		hbb_lo = 2;
 		amsbc = 1;
 		rgb565_predicator = 1;
 		uavflagprd_inv = 2;
 	}
 
 	if (adreno_is_7c3(adreno_gpu)) {
-		lower_bit = 1;
+		hbb_lo = 1;
 		amsbc = 1;
 		rgb565_predicator = 1;
 		uavflagprd_inv = 2;
 	}
 
 	gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL,
-		rgb565_predicator << 11 | amsbc << 4 | lower_bit << 1);
-	gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, lower_bit << 1);
-	gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL,
-		uavflagprd_inv << 4 | lower_bit << 1);
-	gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, lower_bit << 21);
+		  rgb565_predicator << 11 | hbb_hi << 10 | amsbc << 4 |
+		  min_acc_len << 3 | hbb_lo << 1 | ubwc_mode);
+
+	gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, hbb_hi << 4 |
+		  min_acc_len << 3 | hbb_lo << 1 | ubwc_mode);
+
+	gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, hbb_hi << 10 |
+		  uavflagprd_inv << 4 | min_acc_len << 3 |
+		  hbb_lo << 1 | ubwc_mode);
+
+	gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, min_acc_len << 23 | hbb_lo << 21);
 }
 
 static int a6xx_cp_init(struct msm_gpu *gpu)
-- 
GitLab


From 30f55f3f2374b53a28f08f9a670c1ba7ade3805a Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:52 +0200
Subject: [PATCH 1778/1791] drm/msm/a6xx: Move CX GMU power counter enablement
 to hw_init

Since the introduction of A6xx support, we've been enabling the CX GMU
power counter 0 in a bit of a weird spot. Move it to hw_init so that
GMU wrapper GPUs can reuse the same code paths. As a bonus, this order
makes it easier to compare mainline and downstream register access traces.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542765/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 6 ------
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 7 +++++++
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 906bed49f27de..aae7ea651607f 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -479,12 +479,6 @@ static int a6xx_rpmh_start(struct a6xx_gmu *gmu)
 
 	gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0);
 
-	/* Set up CX GMU counter 0 to count busy ticks */
-	gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000);
-	gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, 0x20);
-
-	/* Enable the power counter */
-	gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 8aa4670b4308d..0efecde2af1a3 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1256,6 +1256,13 @@ static int hw_init(struct msm_gpu *gpu)
 			0x3f0243f0);
 	}
 
+	/* Set up the CX GMU counter 0 to count busy ticks */
+	gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000);
+
+	/* Enable the power counter */
+	gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, BIT(5));
+	gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1);
+
 	/* Protect registers from the CP */
 	a6xx_set_cp_protect(gpu);
 
-- 
GitLab


From 5a903a44a98471cedf0021fac0a6a64bbe86943f Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:53 +0200
Subject: [PATCH 1779/1791] drm/msm/a6xx: Introduce GMU wrapper support

Some (particularly SMD_RPM, a.k.a non-RPMh) SoCs implement A6XX GPUs
but don't implement the associated GMUs. This is due to the fact that
the GMU directly pokes at RPMh. Sadly, this means we have to take care
of enabling & scaling power rails, clocks and bandwidth ourselves.

Reuse existing Adreno-common code and modify the deeply-GMU-infused
A6XX code to facilitate these GPUs. This involves if-ing out lots
of GMU callbacks and introducing a new type of GMU - GMU wrapper (it's
the actual name that Qualcomm uses in their downstream kernels).

This is essentially a register region which is convenient to model
as a device. We'll use it for managing the GDSCs. The register
layout matches the actual GMU_CX/GX regions on the "real GMU" devices
and lets us reuse quite a bit of gmu_read/write/rmw calls.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542766/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c       |  72 ++++++-
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c       | 201 +++++++++++++++++---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h       |   1 +
 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c |  14 +-
 drivers/gpu/drm/msm/adreno/adreno_gpu.c     |   8 +-
 drivers/gpu/drm/msm/adreno/adreno_gpu.h     |   6 +
 6 files changed, 266 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index aae7ea651607f..5deb79924897a 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1431,6 +1431,7 @@ static int a6xx_gmu_get_irq(struct a6xx_gmu *gmu, struct platform_device *pdev,
 
 void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu)
 {
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
 	struct platform_device *pdev = to_platform_device(gmu->dev);
 
@@ -1456,10 +1457,12 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu)
 	gmu->mmio = NULL;
 	gmu->rscc = NULL;
 
-	a6xx_gmu_memory_free(gmu);
+	if (!adreno_has_gmu_wrapper(adreno_gpu)) {
+		a6xx_gmu_memory_free(gmu);
 
-	free_irq(gmu->gmu_irq, gmu);
-	free_irq(gmu->hfi_irq, gmu);
+		free_irq(gmu->gmu_irq, gmu);
+		free_irq(gmu->hfi_irq, gmu);
+	}
 
 	/* Drop reference taken in of_find_device_by_node */
 	put_device(gmu->dev);
@@ -1478,6 +1481,69 @@ static int cxpd_notifier_cb(struct notifier_block *nb,
 	return 0;
 }
 
+int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
+{
+	struct platform_device *pdev = of_find_device_by_node(node);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	int ret;
+
+	if (!pdev)
+		return -ENODEV;
+
+	gmu->dev = &pdev->dev;
+
+	of_dma_configure(gmu->dev, node, true);
+
+	pm_runtime_enable(gmu->dev);
+
+	/* Mark legacy for manual SPTPRAC control */
+	gmu->legacy = true;
+
+	/* Map the GMU registers */
+	gmu->mmio = a6xx_gmu_get_mmio(pdev, "gmu");
+	if (IS_ERR(gmu->mmio)) {
+		ret = PTR_ERR(gmu->mmio);
+		goto err_mmio;
+	}
+
+	gmu->cxpd = dev_pm_domain_attach_by_name(gmu->dev, "cx");
+	if (IS_ERR(gmu->cxpd)) {
+		ret = PTR_ERR(gmu->cxpd);
+		goto err_mmio;
+	}
+
+	if (!device_link_add(gmu->dev, gmu->cxpd, DL_FLAG_PM_RUNTIME)) {
+		ret = -ENODEV;
+		goto detach_cxpd;
+	}
+
+	init_completion(&gmu->pd_gate);
+	complete_all(&gmu->pd_gate);
+	gmu->pd_nb.notifier_call = cxpd_notifier_cb;
+
+	/* Get a link to the GX power domain to reset the GPU */
+	gmu->gxpd = dev_pm_domain_attach_by_name(gmu->dev, "gx");
+	if (IS_ERR(gmu->gxpd)) {
+		ret = PTR_ERR(gmu->gxpd);
+		goto err_mmio;
+	}
+
+	gmu->initialized = true;
+
+	return 0;
+
+detach_cxpd:
+	dev_pm_domain_detach(gmu->cxpd, false);
+
+err_mmio:
+	iounmap(gmu->mmio);
+
+	/* Drop reference taken in of_find_device_by_node */
+	put_device(gmu->dev);
+
+	return ret;
+}
+
 int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 {
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 0efecde2af1a3..b91fc02eb08cb 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -21,7 +21,7 @@ static inline bool _a6xx_check_idle(struct msm_gpu *gpu)
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 
 	/* Check that the GMU is idle */
-	if (!a6xx_gmu_isidle(&a6xx_gpu->gmu))
+	if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_isidle(&a6xx_gpu->gmu))
 		return false;
 
 	/* Check tha the CX master is idle */
@@ -1126,10 +1126,13 @@ static int hw_init(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
 	int ret;
 
-	/* Make sure the GMU keeps the GPU on while we set it up */
-	a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET);
+	if (!adreno_has_gmu_wrapper(adreno_gpu)) {
+		/* Make sure the GMU keeps the GPU on while we set it up */
+		a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET);
+	}
 
 	/* Clear GBIF halt in case GX domain was not collapsed */
 	if (a6xx_has_gbif(adreno_gpu)) {
@@ -1352,6 +1355,8 @@ static int hw_init(struct msm_gpu *gpu)
 	}
 
 out:
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return ret;
 	/*
 	 * Tell the GMU that we are done touching the GPU and it can start power
 	 * management
@@ -1386,9 +1391,6 @@ static void a6xx_dump(struct msm_gpu *gpu)
 	adreno_dump(gpu);
 }
 
-#define VBIF_RESET_ACK_TIMEOUT	100
-#define VBIF_RESET_ACK_MASK	0x00f0
-
 static void a6xx_recover(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -1426,6 +1428,15 @@ static void a6xx_recover(struct msm_gpu *gpu)
 	 */
 	gpu->active_submits = 0;
 
+	if (adreno_has_gmu_wrapper(adreno_gpu)) {
+		/* Drain the outstanding traffic on memory buses */
+		a6xx_bus_clear_pending_transactions(adreno_gpu, true);
+
+		/* Reset the GPU to a clean state */
+		a6xx_gpu_sw_reset(gpu, true);
+		a6xx_gpu_sw_reset(gpu, false);
+	}
+
 	reinit_completion(&gmu->pd_gate);
 	dev_pm_genpd_add_notifier(gmu->cxpd, &gmu->pd_nb);
 	dev_pm_genpd_synced_poweroff(gmu->cxpd);
@@ -1576,7 +1587,8 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu)
 	 * Force the GPU to stay on until after we finish
 	 * collecting information
 	 */
-	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1);
+	if (!adreno_has_gmu_wrapper(adreno_gpu))
+		gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1);
 
 	DRM_DEV_ERROR(&gpu->pdev->dev,
 		"gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n",
@@ -1707,6 +1719,10 @@ static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu)
 
 static void a6xx_llc_slices_destroy(struct a6xx_gpu *a6xx_gpu)
 {
+	/* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */
+	if (adreno_has_gmu_wrapper(&a6xx_gpu->base))
+		return;
+
 	llcc_slice_putd(a6xx_gpu->llc_slice);
 	llcc_slice_putd(a6xx_gpu->htw_llc_slice);
 }
@@ -1716,6 +1732,10 @@ static void a6xx_llc_slices_init(struct platform_device *pdev,
 {
 	struct device_node *phandle;
 
+	/* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */
+	if (adreno_has_gmu_wrapper(&a6xx_gpu->base))
+		return;
+
 	/*
 	 * There is a different programming path for targets with an mmu500
 	 * attached, so detect if that is the case
@@ -1786,7 +1806,7 @@ void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert)
 		udelay(100);
 }
 
-static int a6xx_pm_resume(struct msm_gpu *gpu)
+static int a6xx_gmu_pm_resume(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
@@ -1806,10 +1826,58 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
 
 	a6xx_llc_activate(a6xx_gpu);
 
-	return 0;
+	return ret;
 }
 
-static int a6xx_pm_suspend(struct msm_gpu *gpu)
+static int a6xx_pm_resume(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	unsigned long freq = gpu->fast_rate;
+	struct dev_pm_opp *opp;
+	int ret;
+
+	gpu->needs_hw_init = true;
+
+	trace_msm_gpu_resume(0);
+
+	mutex_lock(&a6xx_gpu->gmu.lock);
+
+	opp = dev_pm_opp_find_freq_ceil(&gpu->pdev->dev, &freq);
+	if (IS_ERR(opp)) {
+		ret = PTR_ERR(opp);
+		goto err_set_opp;
+	}
+	dev_pm_opp_put(opp);
+
+	/* Set the core clock and bus bw, having VDD scaling in mind */
+	dev_pm_opp_set_opp(&gpu->pdev->dev, opp);
+
+	pm_runtime_resume_and_get(gmu->dev);
+	pm_runtime_resume_and_get(gmu->gxpd);
+
+	ret = clk_bulk_prepare_enable(gpu->nr_clocks, gpu->grp_clks);
+	if (ret)
+		goto err_bulk_clk;
+
+	/* If anything goes south, tear the GPU down piece by piece.. */
+	if (ret) {
+err_bulk_clk:
+		pm_runtime_put(gmu->gxpd);
+		pm_runtime_put(gmu->dev);
+		dev_pm_opp_set_opp(&gpu->pdev->dev, NULL);
+	}
+err_set_opp:
+	mutex_unlock(&a6xx_gpu->gmu.lock);
+
+	if (!ret)
+		msm_devfreq_resume(gpu);
+
+	return ret;
+}
+
+static int a6xx_gmu_pm_suspend(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
@@ -1836,7 +1904,40 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
 	return 0;
 }
 
-static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
+static int a6xx_pm_suspend(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	int i;
+
+	trace_msm_gpu_suspend(0);
+
+	msm_devfreq_suspend(gpu);
+
+	mutex_lock(&a6xx_gpu->gmu.lock);
+
+	/* Drain the outstanding traffic on memory buses */
+	a6xx_bus_clear_pending_transactions(adreno_gpu, true);
+
+	clk_bulk_disable_unprepare(gpu->nr_clocks, gpu->grp_clks);
+
+	pm_runtime_put_sync(gmu->gxpd);
+	dev_pm_opp_set_opp(&gpu->pdev->dev, NULL);
+	pm_runtime_put_sync(gmu->dev);
+
+	mutex_unlock(&a6xx_gpu->gmu.lock);
+
+	if (a6xx_gpu->shadow_bo)
+		for (i = 0; i < gpu->nr_rings; i++)
+			a6xx_gpu->shadow[i] = 0;
+
+	gpu->suspend_count++;
+
+	return 0;
+}
+
+static int a6xx_gmu_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
@@ -1855,6 +1956,12 @@ static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 	return 0;
 }
 
+static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
+{
+	*value = gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER);
+	return 0;
+}
+
 static struct msm_ringbuffer *a6xx_active_ring(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -2121,8 +2228,8 @@ static const struct adreno_gpu_funcs funcs = {
 		.set_param = adreno_set_param,
 		.hw_init = a6xx_hw_init,
 		.ucode_load = a6xx_ucode_load,
-		.pm_suspend = a6xx_pm_suspend,
-		.pm_resume = a6xx_pm_resume,
+		.pm_suspend = a6xx_gmu_pm_suspend,
+		.pm_resume = a6xx_gmu_pm_resume,
 		.recover = a6xx_recover,
 		.submit = a6xx_submit,
 		.active_ring = a6xx_active_ring,
@@ -2137,6 +2244,35 @@ static const struct adreno_gpu_funcs funcs = {
 #if defined(CONFIG_DRM_MSM_GPU_STATE)
 		.gpu_state_get = a6xx_gpu_state_get,
 		.gpu_state_put = a6xx_gpu_state_put,
+#endif
+		.create_address_space = a6xx_create_address_space,
+		.create_private_address_space = a6xx_create_private_address_space,
+		.get_rptr = a6xx_get_rptr,
+		.progress = a6xx_progress,
+	},
+	.get_timestamp = a6xx_gmu_get_timestamp,
+};
+
+static const struct adreno_gpu_funcs funcs_gmuwrapper = {
+	.base = {
+		.get_param = adreno_get_param,
+		.set_param = adreno_set_param,
+		.hw_init = a6xx_hw_init,
+		.ucode_load = a6xx_ucode_load,
+		.pm_suspend = a6xx_pm_suspend,
+		.pm_resume = a6xx_pm_resume,
+		.recover = a6xx_recover,
+		.submit = a6xx_submit,
+		.active_ring = a6xx_active_ring,
+		.irq = a6xx_irq,
+		.destroy = a6xx_destroy,
+#if defined(CONFIG_DRM_MSM_GPU_STATE)
+		.show = a6xx_show,
+#endif
+		.gpu_busy = a6xx_gpu_busy,
+#if defined(CONFIG_DRM_MSM_GPU_STATE)
+		.gpu_state_get = a6xx_gpu_state_get,
+		.gpu_state_put = a6xx_gpu_state_put,
 #endif
 		.create_address_space = a6xx_create_address_space,
 		.create_private_address_space = a6xx_create_private_address_space,
@@ -2169,16 +2305,31 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 
 	adreno_gpu->registers = NULL;
 
+	/* Check if there is a GMU phandle and set it up */
+	node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
+	/* FIXME: How do we gracefully handle this? */
+	BUG_ON(!node);
+
+	adreno_gpu->gmu_is_wrapper = of_device_is_compatible(node, "qcom,adreno-gmu-wrapper");
+
 	/*
 	 * We need to know the platform type before calling into adreno_gpu_init
 	 * so that the hw_apriv flag can be correctly set. Snoop into the info
 	 * and grab the revision number
 	 */
 	info = adreno_info(config->rev);
-
-	if (info && (info->revn == 650 || info->revn == 660 ||
-		     info->revn == 690 ||
-		     adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev)))
+	if (!info)
+		return ERR_PTR(-EINVAL);
+
+	/* Assign these early so that we can use the is_aXYZ helpers */
+	/* Numeric revision IDs (e.g. 630) */
+	adreno_gpu->revn = info->revn;
+	/* New-style ADRENO_REV()-only */
+	adreno_gpu->rev = info->rev;
+	/* Quirk data */
+	adreno_gpu->info = info;
+
+	if (adreno_is_a650(adreno_gpu) || adreno_is_a660_family(adreno_gpu))
 		adreno_gpu->base.hw_apriv = true;
 
 	a6xx_llc_slices_init(pdev, a6xx_gpu);
@@ -2189,7 +2340,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 		return ERR_PTR(ret);
 	}
 
-	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs_gmuwrapper, 1);
+	else
+		ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
 		return ERR_PTR(ret);
@@ -2202,13 +2356,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 	if (adreno_is_a618(adreno_gpu) || adreno_is_7c3(adreno_gpu))
 		priv->gpu_clamp_to_idle = true;
 
-	/* Check if there is a GMU phandle and set it up */
-	node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0);
-
-	/* FIXME: How do we gracefully handle this? */
-	BUG_ON(!node);
-
-	ret = a6xx_gmu_init(a6xx_gpu, node);
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		ret = a6xx_gmu_wrapper_init(a6xx_gpu, node);
+	else
+		ret = a6xx_gmu_init(a6xx_gpu, node);
 	of_node_put(node);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index aa70390ee1c62..c788b06e72da8 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -76,6 +76,7 @@ int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state);
 void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state);
 
 int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node);
+int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node);
 void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu);
 
 void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp,
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
index 30ecdff363e7f..4e5d650578c60 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
@@ -1041,16 +1041,18 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu)
 	/* Get the generic state from the adreno core */
 	adreno_gpu_state_get(gpu, &a6xx_state->base);
 
-	a6xx_get_gmu_registers(gpu, a6xx_state);
+	if (!adreno_has_gmu_wrapper(adreno_gpu)) {
+		a6xx_get_gmu_registers(gpu, a6xx_state);
 
-	a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log);
-	a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi);
-	a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug);
+		a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log);
+		a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi);
+		a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug);
 
-	a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state);
+		a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state);
+	}
 
 	/* If GX isn't on the rest of the data isn't going to be accessible */
-	if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu))
+	if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_gx_is_on(&a6xx_gpu->gmu))
 		return &a6xx_state->base;
 
 	/* Get the banks of indexed registers */
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 6934cee07d42d..5c5901d659507 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -528,6 +528,10 @@ int adreno_load_fw(struct adreno_gpu *adreno_gpu)
 		if (!adreno_gpu->info->fw[i])
 			continue;
 
+		/* Skip loading GMU firwmare with GMU Wrapper */
+		if (adreno_has_gmu_wrapper(adreno_gpu) && i == ADRENO_FW_GMU)
+			continue;
+
 		/* Skip if the firmware has already been loaded */
 		if (adreno_gpu->fw[i])
 			continue;
@@ -1074,8 +1078,8 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	u32 speedbin;
 	int ret;
 
-	/* Only handle the core clock when GMU is not in use */
-	if (config->rev.core < 6) {
+	/* Only handle the core clock when GMU is not in use (or is absent). */
+	if (adreno_has_gmu_wrapper(adreno_gpu) || config->rev.core < 6) {
 		/*
 		 * This can only be done before devm_pm_opp_of_add_table(), or
 		 * dev_pm_opp_set_config() will WARN_ON()
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 5a26c8a2de7cb..de0b03a4b5947 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -115,6 +115,7 @@ struct adreno_gpu {
 	 * code (a3xx_gpu.c) and stored in this common location.
 	 */
 	const unsigned int *reg_offsets;
+	bool gmu_is_wrapper;
 };
 #define to_adreno_gpu(x) container_of(x, struct adreno_gpu, base)
 
@@ -152,6 +153,11 @@ static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn)
 	return gpu->revn == revn;
 }
 
+static inline bool adreno_has_gmu_wrapper(const struct adreno_gpu *gpu)
+{
+	return gpu->gmu_is_wrapper;
+}
+
 static inline bool adreno_is_a2xx(const struct adreno_gpu *gpu)
 {
 	WARN_ON_ONCE(!gpu->revn);
-- 
GitLab


From 9bad37600c89eb6b3ac65faabbb26da21b48ccad Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:54 +0200
Subject: [PATCH 1780/1791] drm/msm/adreno: Disable has_cached_coherent in GMU
 wrapper configurations

A610 and A619_holi don't support the feature. Disable it to make the GPU stop
crashing after almost each and every submission - the received data on
the GPU end was simply incomplete in garbled, resulting in almost nothing
being executed properly. Extend the disablement to adreno_has_gmu_wrapper,
as none of the GMU wrapper Adrenos that don't support yet seem to feature it.

Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542774/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/adreno_device.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index e5a865024e942..6ea24b8ddcf83 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -565,7 +565,6 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
 		config.rev.minor, config.rev.patchid);
 
 	priv->is_a2xx = config.rev.core == 2;
-	priv->has_cached_coherent = config.rev.core >= 6;
 
 	gpu = info->init(drm);
 	if (IS_ERR(gpu)) {
@@ -577,6 +576,10 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		return ret;
 
+	if (config.rev.core >= 6)
+		if (!adreno_has_gmu_wrapper(to_adreno_gpu(gpu)))
+			priv->has_cached_coherent = true;
+
 	return 0;
 }
 
-- 
GitLab


From 8296ff0afdf22656683c9cf908ee8abdfa96d03a Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:55 +0200
Subject: [PATCH 1781/1791] drm/msm/a6xx: Add support for A619_holi

A619_holi is a GMU-less variant of the already-supported A619 GPU.
It's present on at least SM4350 (holi) and SM6375 (blair). No mesa
changes are required. Add the required kernel-side support for it.

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542775/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c   | 27 +++++++++++++++++++++++--
 drivers/gpu/drm/msm/adreno/adreno_gpu.h |  5 +++++
 2 files changed, 30 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index b91fc02eb08cb..2ca9e04403962 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -911,6 +911,9 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
 	if (adreno_is_a618(adreno_gpu))
 		return;
 
+	if (adreno_is_a619_holi(adreno_gpu))
+		hbb_lo = 0;
+
 	if (adreno_is_a640_family(adreno_gpu))
 		amsbc = 1;
 
@@ -1135,7 +1138,12 @@ static int hw_init(struct msm_gpu *gpu)
 	}
 
 	/* Clear GBIF halt in case GX domain was not collapsed */
-	if (a6xx_has_gbif(adreno_gpu)) {
+	if (adreno_is_a619_holi(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_GBIF_HALT, 0);
+		gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, 0);
+		/* Let's make extra sure that the GPU can access the memory.. */
+		mb();
+	} else if (a6xx_has_gbif(adreno_gpu)) {
 		gpu_write(gpu, REG_A6XX_GBIF_HALT, 0);
 		gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 0);
 		/* Let's make extra sure that the GPU can access the memory.. */
@@ -1144,6 +1152,9 @@ static int hw_init(struct msm_gpu *gpu)
 
 	gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_CNTL, 0);
 
+	if (adreno_is_a619_holi(adreno_gpu))
+		a6xx_sptprac_enable(gmu);
+
 	/*
 	 * Disable the trusted memory range - we don't actually supported secure
 	 * memory rendering at this point in time and we don't want to block off
@@ -1760,12 +1771,18 @@ static void a6xx_llc_slices_init(struct platform_device *pdev,
 #define GBIF_CLIENT_HALT_MASK		BIT(0)
 #define GBIF_ARB_HALT_MASK		BIT(1)
 #define VBIF_XIN_HALT_CTRL0_MASK	GENMASK(3, 0)
+#define VBIF_RESET_ACK_MASK		0xF0
+#define GPR0_GBIF_HALT_REQUEST		0x1E0
 
 void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off)
 {
 	struct msm_gpu *gpu = &adreno_gpu->base;
 
-	if (!a6xx_has_gbif(adreno_gpu)) {
+	if (adreno_is_a619_holi(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, GPR0_GBIF_HALT_REQUEST);
+		spin_until((gpu_read(gpu, REG_A6XX_RBBM_VBIF_GX_RESET_STATUS) &
+				(VBIF_RESET_ACK_MASK)) == VBIF_RESET_ACK_MASK);
+	} else if (!a6xx_has_gbif(adreno_gpu)) {
 		gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, VBIF_XIN_HALT_CTRL0_MASK);
 		spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) &
 				(VBIF_XIN_HALT_CTRL0_MASK)) == VBIF_XIN_HALT_CTRL0_MASK);
@@ -1861,6 +1878,9 @@ static int a6xx_pm_resume(struct msm_gpu *gpu)
 	if (ret)
 		goto err_bulk_clk;
 
+	if (adreno_is_a619_holi(adreno_gpu))
+		a6xx_sptprac_enable(gmu);
+
 	/* If anything goes south, tear the GPU down piece by piece.. */
 	if (ret) {
 err_bulk_clk:
@@ -1920,6 +1940,9 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu)
 	/* Drain the outstanding traffic on memory buses */
 	a6xx_bus_clear_pending_transactions(adreno_gpu, true);
 
+	if (adreno_is_a619_holi(adreno_gpu))
+		a6xx_sptprac_disable(gmu);
+
 	clk_bulk_disable_unprepare(gpu->nr_clocks, gpu->grp_clks);
 
 	pm_runtime_put_sync(gmu->gxpd);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index de0b03a4b5947..efd35b7bc4cfc 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -263,6 +263,11 @@ static inline int adreno_is_a619(const struct adreno_gpu *gpu)
 	return adreno_is_revn(gpu, 619);
 }
 
+static inline int adreno_is_a619_holi(const struct adreno_gpu *gpu)
+{
+	return adreno_is_a619(gpu) && adreno_has_gmu_wrapper(gpu);
+}
+
 static inline int adreno_is_a630(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 630);
-- 
GitLab


From e7fc9398e608a7bc057fe7d50dbdaf3052be59b5 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:56 +0200
Subject: [PATCH 1782/1791] drm/msm/a6xx: Add A610 support

A610 is one of (if not the) lowest-tier SKUs in the A6XX family. It
features no GMU, as it's implemented solely on SoCs with SMD_RPM.
What's more interesting is that it does not feature a VDDGX line
either, being powered solely by VDDCX and has an unfortunate hardware
quirk that makes its reset line broken - after a couple of assert/
deassert cycles, it will hang for good and will not wake up again.

This GPU requires mesa changes for proper rendering, and lots of them
at that. The command streams are quite far away from any other A6XX
GPU and hence it needs special care. This patch was validated both
by running an (incomplete) downstream mesa with some hacks (frames
rendered correctly, though some instructions made the GPU hangcheck
which is expected - garbage in, garbage out) and by replaying RD
traces captured with the downstream KGSL driver - no crashes there,
ever.

Add support for this GPU on the kernel side, which comes down to
pretty simply adding A612 HWCG tables, altering a few values and
adding a special case for handling the reset line.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542779/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c      | 99 +++++++++++++++++++---
 drivers/gpu/drm/msm/adreno/adreno_device.c | 12 +++
 drivers/gpu/drm/msm/adreno/adreno_gpu.h    |  8 +-
 3 files changed, 107 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 2ca9e04403962..47aafc9deaf82 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -252,6 +252,56 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	a6xx_flush(gpu, ring);
 }
 
+const struct adreno_reglist a612_hwcg[] = {
+	{REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220},
+	{REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000081},
+	{REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf},
+	{REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111},
+	{REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111},
+	{REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777},
+	{REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777},
+	{REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01202222},
+	{REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220},
+	{REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00},
+	{REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05522022},
+	{REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555},
+	{REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011},
+	{REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044},
+	{REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222},
+	{REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002},
+	{REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000},
+	{REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200},
+	{REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004},
+	{REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222},
+	{REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004},
+	{REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002},
+	{REG_A6XX_RBBM_ISDB_CNT, 0x00000182},
+	{REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000},
+	{REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000},
+	{REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222},
+	{REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111},
+	{REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555},
+	{},
+};
+
 /* For a615 family (a615, a616, a618 and a619) */
 const struct adreno_reglist a615_hwcg[] = {
 	{REG_A6XX_RBBM_CLOCK_CNTL_SP0,  0x02222222},
@@ -659,6 +709,8 @@ static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state)
 
 	if (adreno_is_a630(adreno_gpu))
 		clock_cntl_on = 0x8aa8aa02;
+	else if (adreno_is_a610(adreno_gpu))
+		clock_cntl_on = 0xaaa8aa82;
 	else
 		clock_cntl_on = 0x8aa8aa82;
 
@@ -669,13 +721,15 @@ static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state)
 		return;
 
 	/* Disable SP clock before programming HWCG registers */
-	gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 1, 0);
+	if (!adreno_is_a610(adreno_gpu))
+		gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 1, 0);
 
 	for (i = 0; (reg = &adreno_gpu->info->hwcg[i], reg->offset); i++)
 		gpu_write(gpu, reg->offset, state ? reg->value : 0);
 
 	/* Enable SP clock */
-	gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 0, 1);
+	if (!adreno_is_a610(adreno_gpu))
+		gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 0, 1);
 
 	gpu_write(gpu, REG_A6XX_RBBM_CLOCK_CNTL, state ? clock_cntl_on : 0);
 }
@@ -907,6 +961,13 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu)
 	/* Unknown, introduced with A640/680 */
 	u32 amsbc = 0;
 
+	if (adreno_is_a610(adreno_gpu)) {
+		/* HBB = 14 */
+		hbb_lo = 1;
+		min_acc_len = 1;
+		ubwc_mode = 1;
+	}
+
 	/* a618 is using the hw default values */
 	if (adreno_is_a618(adreno_gpu))
 		return;
@@ -1181,13 +1242,13 @@ static int hw_init(struct msm_gpu *gpu)
 	a6xx_set_hwcg(gpu, true);
 
 	/* VBIF/GBIF start*/
-	if (adreno_is_a640_family(adreno_gpu) ||
+	if (adreno_is_a610(adreno_gpu) ||
+	    adreno_is_a640_family(adreno_gpu) ||
 	    adreno_is_a650_family(adreno_gpu)) {
 		gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE0, 0x00071620);
 		gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE1, 0x00071620);
 		gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE2, 0x00071620);
 		gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE3, 0x00071620);
-		gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE3, 0x00071620);
 		gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x3);
 	} else {
 		gpu_write(gpu, REG_A6XX_RBBM_VBIF_CLIENT_QOS_CNTL, 0x3);
@@ -1215,18 +1276,26 @@ static int hw_init(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A6XX_UCHE_FILTER_CNTL, 0x804);
 	gpu_write(gpu, REG_A6XX_UCHE_CACHE_WAYS, 0x4);
 
-	if (adreno_is_a640_family(adreno_gpu) ||
-	    adreno_is_a650_family(adreno_gpu))
+	if (adreno_is_a640_family(adreno_gpu) || adreno_is_a650_family(adreno_gpu)) {
 		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x02000140);
-	else
+		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c);
+	} else if (adreno_is_a610(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x00800060);
+		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x40201b16);
+	} else {
 		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x010000c0);
-	gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c);
+		gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c);
+	}
 
 	if (adreno_is_a660_family(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_CP_LPAC_PROG_FIFO_SIZE, 0x00000020);
 
 	/* Setting the mem pool size */
-	gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 128);
+	if (adreno_is_a610(adreno_gpu)) {
+		gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 48);
+		gpu_write(gpu, REG_A6XX_CP_MEM_POOL_DBG_ADDR, 47);
+	} else
+		gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 128);
 
 	/* Setting the primFifo thresholds default values,
 	 * and vccCacheSkipDis=1 bit (0x200) for A640 and newer
@@ -1237,6 +1306,8 @@ static int hw_init(struct msm_gpu *gpu)
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00200200);
 	else if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200);
+	else if (adreno_is_a610(adreno_gpu))
+		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00080000);
 	else
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00180000);
 
@@ -1252,8 +1323,10 @@ static int hw_init(struct msm_gpu *gpu)
 	a6xx_set_ubwc_config(gpu);
 
 	/* Enable fault detection */
-	gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL,
-		(1 << 30) | 0x1fffff);
+	if (adreno_is_a610(adreno_gpu))
+		gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3ffff);
+	else
+		gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x1fffff);
 
 	gpu_write(gpu, REG_A6XX_UCHE_CLIENT_PF, 1);
 
@@ -1813,6 +1886,10 @@ void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_
 
 void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert)
 {
+	/* 11nm chips (e.g. ones with A610) have hw issues with the reset line! */
+	if (adreno_is_a610(to_adreno_gpu(gpu)))
+		return;
+
 	gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, assert);
 	/* Perform a bogus read and add a brief delay to ensure ordering. */
 	gpu_read(gpu, REG_A6XX_RBBM_SW_RESET_CMD);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 6ea24b8ddcf83..cb94cfd137a81 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -253,6 +253,18 @@ static const struct adreno_info gpulist[] = {
 		.quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE,
 		.init = a5xx_gpu_init,
 		.zapfw = "a540_zap.mdt",
+	}, {
+		.rev = ADRENO_REV(6, 1, 0, ANY_ID),
+		.revn = 610,
+		.name = "A610",
+		.fw = {
+			[ADRENO_FW_SQE] = "a630_sqe.fw",
+		},
+		.gmem = (SZ_128K + SZ_4K),
+		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
+		.init = a6xx_gpu_init,
+		.zapfw = "a610_zap.mdt",
+		.hwcg = a612_hwcg,
 	}, {
 		.rev = ADRENO_REV(6, 1, 8, ANY_ID),
 		.revn = 618,
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index efd35b7bc4cfc..3a8af5fdaea80 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -55,7 +55,8 @@ struct adreno_reglist {
 	u32 value;
 };
 
-extern const struct adreno_reglist a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[], a660_hwcg[], a690_hwcg[];
+extern const struct adreno_reglist a612_hwcg[], a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[];
+extern const struct adreno_reglist a660_hwcg[], a690_hwcg[];
 
 struct adreno_info {
 	struct adreno_rev rev;
@@ -253,6 +254,11 @@ static inline int adreno_is_a540(const struct adreno_gpu *gpu)
 	return adreno_is_revn(gpu, 540);
 }
 
+static inline int adreno_is_a610(const struct adreno_gpu *gpu)
+{
+	return adreno_is_revn(gpu, 610);
+}
+
 static inline int adreno_is_a618(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 618);
-- 
GitLab


From 3e90044d3e4941efafbfcfad8b1b7cfdb6d97a58 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:57 +0200
Subject: [PATCH 1783/1791] drm/msm/a6xx: Fix some A619 tunables

Adreno 619 expects some tunables to be set differently. Make up for it.

Fixes: b7616b5c69e6 ("drm/msm/adreno: Add A619 support")
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542782/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 47aafc9deaf82..97e261d333124 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1306,6 +1306,8 @@ static int hw_init(struct msm_gpu *gpu)
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00200200);
 	else if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200);
+	else if (adreno_is_a619(adreno_gpu))
+		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00018000);
 	else if (adreno_is_a610(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00080000);
 	else
@@ -1323,7 +1325,9 @@ static int hw_init(struct msm_gpu *gpu)
 	a6xx_set_ubwc_config(gpu);
 
 	/* Enable fault detection */
-	if (adreno_is_a610(adreno_gpu))
+	if (adreno_is_a619(adreno_gpu))
+		gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3fffff);
+	else if (adreno_is_a610(adreno_gpu))
 		gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3ffff);
 	else
 		gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x1fffff);
-- 
GitLab


From ac926549651140243c6f4e234ca83fcb29d29ecf Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:58 +0200
Subject: [PATCH 1784/1791] drm/msm/a6xx: Use "else if" in GPU speedbin rev
 matching

The GPU can only be one at a time. Turn a series of ifs into if +
elseifs to save some CPU cycles.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542770/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 97e261d333124..d0ba0844079c9 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2276,16 +2276,16 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse)
 	if (adreno_cmp_rev(ADRENO_REV(6, 1, 8, ANY_ID), rev))
 		val = a618_get_speed_bin(fuse);
 
-	if (adreno_cmp_rev(ADRENO_REV(6, 1, 9, ANY_ID), rev))
+	else if (adreno_cmp_rev(ADRENO_REV(6, 1, 9, ANY_ID), rev))
 		val = a619_get_speed_bin(fuse);
 
-	if (adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), rev))
+	else if (adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), rev))
 		val = adreno_7c3_get_speed_bin(fuse);
 
-	if (adreno_cmp_rev(ADRENO_REV(6, 4, 0, ANY_ID), rev))
+	else if (adreno_cmp_rev(ADRENO_REV(6, 4, 0, ANY_ID), rev))
 		val = a640_get_speed_bin(fuse);
 
-	if (adreno_cmp_rev(ADRENO_REV(6, 5, 0, ANY_ID), rev))
+	else if (adreno_cmp_rev(ADRENO_REV(6, 5, 0, ANY_ID), rev))
 		val = a650_get_speed_bin(fuse);
 
 	if (val == UINT_MAX) {
-- 
GitLab


From b3bcd583fa6f963be9c8a08405b2cbaeea797a3c Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:20:59 +0200
Subject: [PATCH 1785/1791] drm/msm/a6xx: Use adreno_is_aXYZ macros in speedbin
 matching

Before transitioning to using per-SoC and not per-Adreno speedbin
fuse values (need another patchset to land elsewhere), a good
improvement/stopgap solution is to use adreno_is_aXYZ macros in
place of explicit revision matching. Do so to allow differentiating
between A619 and A619_holi.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542777/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c   | 18 +++++++++---------
 drivers/gpu/drm/msm/adreno/adreno_gpu.h | 15 ++++++++++++---
 2 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index d0ba0844079c9..d7139eae0f732 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2269,23 +2269,23 @@ static u32 adreno_7c3_get_speed_bin(u32 fuse)
 	return UINT_MAX;
 }
 
-static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse)
+static u32 fuse_to_supp_hw(struct device *dev, struct adreno_gpu *adreno_gpu, u32 fuse)
 {
 	u32 val = UINT_MAX;
 
-	if (adreno_cmp_rev(ADRENO_REV(6, 1, 8, ANY_ID), rev))
+	if (adreno_is_a618(adreno_gpu))
 		val = a618_get_speed_bin(fuse);
 
-	else if (adreno_cmp_rev(ADRENO_REV(6, 1, 9, ANY_ID), rev))
+	else if (adreno_is_a619(adreno_gpu))
 		val = a619_get_speed_bin(fuse);
 
-	else if (adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), rev))
+	else if (adreno_is_7c3(adreno_gpu))
 		val = adreno_7c3_get_speed_bin(fuse);
 
-	else if (adreno_cmp_rev(ADRENO_REV(6, 4, 0, ANY_ID), rev))
+	else if (adreno_is_a640(adreno_gpu))
 		val = a640_get_speed_bin(fuse);
 
-	else if (adreno_cmp_rev(ADRENO_REV(6, 5, 0, ANY_ID), rev))
+	else if (adreno_is_a650(adreno_gpu))
 		val = a650_get_speed_bin(fuse);
 
 	if (val == UINT_MAX) {
@@ -2298,7 +2298,7 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse)
 	return (1 << val);
 }
 
-static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev)
+static int a6xx_set_supported_hw(struct device *dev, struct adreno_gpu *adreno_gpu)
 {
 	u32 supp_hw;
 	u32 speedbin;
@@ -2317,7 +2317,7 @@ static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev)
 		return ret;
 	}
 
-	supp_hw = fuse_to_supp_hw(dev, rev, speedbin);
+	supp_hw = fuse_to_supp_hw(dev, adreno_gpu, speedbin);
 
 	ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1);
 	if (ret)
@@ -2438,7 +2438,7 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev)
 
 	a6xx_llc_slices_init(pdev, a6xx_gpu);
 
-	ret = a6xx_set_supported_hw(&pdev->dev, config->rev);
+	ret = a6xx_set_supported_hw(&pdev->dev, adreno_gpu);
 	if (ret) {
 		a6xx_destroy(&(a6xx_gpu->base.base));
 		return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 3a8af5fdaea80..d8c9e8cc37535 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -279,10 +279,9 @@ static inline int adreno_is_a630(const struct adreno_gpu *gpu)
 	return adreno_is_revn(gpu, 630);
 }
 
-static inline int adreno_is_a640_family(const struct adreno_gpu *gpu)
+static inline int adreno_is_a640(const struct adreno_gpu *gpu)
 {
-	return adreno_is_revn(gpu, 640) ||
-		adreno_is_revn(gpu, 680);
+	return adreno_is_revn(gpu, 640);
 }
 
 static inline int adreno_is_a650(const struct adreno_gpu *gpu)
@@ -301,6 +300,11 @@ static inline int adreno_is_a660(const struct adreno_gpu *gpu)
 	return adreno_is_revn(gpu, 660);
 }
 
+static inline int adreno_is_a680(const struct adreno_gpu *gpu)
+{
+	return adreno_is_revn(gpu, 680);
+}
+
 static inline int adreno_is_a690(const struct adreno_gpu *gpu)
 {
 	return adreno_is_revn(gpu, 690);
@@ -328,6 +332,11 @@ static inline int adreno_is_a650_family(const struct adreno_gpu *gpu)
 		adreno_is_a660_family(gpu);
 }
 
+static inline int adreno_is_a640_family(const struct adreno_gpu *gpu)
+{
+	return adreno_is_a640(gpu) || adreno_is_a680(gpu);
+}
+
 u64 adreno_private_address_space_size(struct msm_gpu *gpu);
 int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx,
 		     uint32_t param, uint64_t *value, uint32_t *len);
-- 
GitLab


From 20c8e39985b9a4442123e0fa6496183d9418e422 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:21:00 +0200
Subject: [PATCH 1786/1791] drm/msm/a6xx: Add A619_holi speedbin support

A619_holi is implemented on at least two SoCs: SM4350 (holi) and SM6375
(blair). This is what seems to be a first occurrence of this happening,
but it's easy to overcome by guarding the SoC-specific fuse values with
of_machine_is_compatible(). Do just that to enable frequency limiting
on these SoCs.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542772/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 31 +++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index d7139eae0f732..ff9a8d342c779 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2216,6 +2216,34 @@ static u32 a618_get_speed_bin(u32 fuse)
 	return UINT_MAX;
 }
 
+static u32 a619_holi_get_speed_bin(u32 fuse)
+{
+	/*
+	 * There are (at least) two SoCs implementing A619_holi: SM4350 (holi)
+	 * and SM6375 (blair). Limit the fuse matching to the corresponding
+	 * SoC to prevent bogus frequency setting (as improbable as it may be,
+	 * given unexpected fuse values are.. unexpected! But still possible.)
+	 */
+
+	if (fuse == 0)
+		return 0;
+
+	if (of_machine_is_compatible("qcom,sm4350")) {
+		if (fuse == 138)
+			return 1;
+		else if (fuse == 92)
+			return 2;
+	} else if (of_machine_is_compatible("qcom,sm6375")) {
+		if (fuse == 190)
+			return 1;
+		else if (fuse == 177)
+			return 2;
+	} else
+		pr_warn("Unknown SoC implementing A619_holi!\n");
+
+	return UINT_MAX;
+}
+
 static u32 a619_get_speed_bin(u32 fuse)
 {
 	if (fuse == 0)
@@ -2276,6 +2304,9 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_gpu *adreno_gpu, u3
 	if (adreno_is_a618(adreno_gpu))
 		val = a618_get_speed_bin(fuse);
 
+	else if (adreno_is_a619_holi(adreno_gpu))
+		val = a619_holi_get_speed_bin(fuse);
+
 	else if (adreno_is_a619(adreno_gpu))
 		val = a619_get_speed_bin(fuse);
 
-- 
GitLab


From cd036d542afb82adfbbd43c5dbeb7010e8e91ee7 Mon Sep 17 00:00:00 2001
From: Konrad Dybcio <konrad.dybcio@linaro.org>
Date: Fri, 16 Jun 2023 01:21:01 +0200
Subject: [PATCH 1787/1791] drm/msm/a6xx: Add A610 speedbin support

A610 is implemented on at least three SoCs: SM6115 (bengal), SM6125
(trinket) and SM6225 (khaje). Trinket does not support speed binning
(only a single SKU exists) and we don't yet support khaje upstream.
Hence, add a fuse mapping table for bengal to allow for per-chip
frequency limiting.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/542780/
Signed-off-by: Rob Clark <robdclark@chromium.org>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index ff9a8d342c779..b3ada1e7b598b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2204,6 +2204,30 @@ static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	return progress;
 }
 
+static u32 a610_get_speed_bin(u32 fuse)
+{
+	/*
+	 * There are (at least) three SoCs implementing A610: SM6125 (trinket),
+	 * SM6115 (bengal) and SM6225 (khaje). Trinket does not have speedbinning,
+	 * as only a single SKU exists and we don't support khaje upstream yet.
+	 * Hence, this matching table is only valid for bengal and can be easily
+	 * expanded if need be.
+	 */
+
+	if (fuse == 0)
+		return 0;
+	else if (fuse == 206)
+		return 1;
+	else if (fuse == 200)
+		return 2;
+	else if (fuse == 157)
+		return 3;
+	else if (fuse == 127)
+		return 4;
+
+	return UINT_MAX;
+}
+
 static u32 a618_get_speed_bin(u32 fuse)
 {
 	if (fuse == 0)
@@ -2301,6 +2325,9 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_gpu *adreno_gpu, u3
 {
 	u32 val = UINT_MAX;
 
+	if (adreno_is_a610(adreno_gpu))
+		val = a610_get_speed_bin(fuse);
+
 	if (adreno_is_a618(adreno_gpu))
 		val = a618_get_speed_bin(fuse);
 
-- 
GitLab


From 21f773515902d8b303df650674ad1c5243beb245 Mon Sep 17 00:00:00 2001
From: Alan Previn <alan.previn.teres.alexis@intel.com>
Date: Thu, 8 Jun 2023 16:07:16 -0700
Subject: [PATCH 1788/1791] drm/i915/gsc: take a wakeref for the
 proxy-init-completion check

Ensure intel_gsc_uc_fw_init_done and intel_gsc_uc_fw_proxy_init
takes a wakeref before reading GSC Shim registers.

NOTE: another patch in review also adds a call from selftest to
this same function. (https://patchwork.freedesktop.org/series/117713/)
which is why i am adding the wakeref inside the callee, not the
caller.

v2: - add a helper, 'gsc_uc_get_fw_status' for both callers
      (Daniele Ceraolo)

Fixes: 99afb7cc8c44 ("drm/i915/pxp: Add ARB session creation and cleanup")
Signed-off-by: Alan Previn <alan.previn.teres.alexis@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230608230716.3079594-1-alan.previn.teres.alexis@intel.com
(cherry picked from commit 8c33c3755b75c98d8eb490df345b4187a295a1a8)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
index f46eb17a7a987..60e9c6c9e7756 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c
@@ -24,21 +24,27 @@ static bool gsc_is_in_reset(struct intel_uncore *uncore)
 	       GSC_FW_CURRENT_STATE_RESET;
 }
 
-bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc)
+static u32 gsc_uc_get_fw_status(struct intel_uncore *uncore)
 {
-	struct intel_uncore *uncore = gsc_uc_to_gt(gsc)->uncore;
-	u32 fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG);
+	intel_wakeref_t wakeref;
+	u32 fw_status = 0;
 
-	return REG_FIELD_GET(GSC_FW_CURRENT_STATE, fw_status) ==
+	with_intel_runtime_pm(uncore->rpm, wakeref)
+		fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG);
+
+	return fw_status;
+}
+
+bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc)
+{
+	return REG_FIELD_GET(GSC_FW_CURRENT_STATE,
+			     gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore)) ==
 	       GSC_FW_PROXY_STATE_NORMAL;
 }
 
 bool intel_gsc_uc_fw_init_done(struct intel_gsc_uc *gsc)
 {
-	struct intel_uncore *uncore = gsc_uc_to_gt(gsc)->uncore;
-	u32 fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG);
-
-	return fw_status & GSC_FW_INIT_COMPLETE_BIT;
+	return gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore) & GSC_FW_INIT_COMPLETE_BIT;
 }
 
 static int emit_gsc_fw_load(struct i915_request *rq, struct intel_gsc_uc *gsc)
-- 
GitLab


From cb359c639dc099ce4316cec9013fd4b2ebeb990c Mon Sep 17 00:00:00 2001
From: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Date: Wed, 14 Jun 2023 15:36:46 -0700
Subject: [PATCH 1789/1791] drm/i915/huc: Fix missing error code in
 intel_huc_init()

Smatch warns:
	drivers/gpu/drm/i915/gt/uc/intel_huc.c:388
	    intel_huc_init() warn: missing error code 'err'

When the allocation of VMAs fail: The value of err is zero at this
point and it is passed to PTR_ERR and also finally returning zero which
is success instead of failure.

Fix this by adding the missing error code when VMA allocation fails.

Fixes: 08872cb13a71 ("drm/i915/mtl/huc: auth HuC via GSC")
Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230614223646.2583633-1-daniele.ceraolospurio@intel.com
(cherry picked from commit ce432fd34cc6c7b7af06d1403ec0be19d1e518dc)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/gt/uc/intel_huc.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
index e0afd8f895026..ddd146265beb4 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c
@@ -384,6 +384,7 @@ int intel_huc_init(struct intel_huc *huc)
 
 		vma = intel_guc_allocate_vma(&gt->uc.guc, PXP43_HUC_AUTH_INOUT_SIZE * 2);
 		if (IS_ERR(vma)) {
+			err = PTR_ERR(vma);
 			huc_info(huc, "Failed to allocate heci pkt\n");
 			goto out;
 		}
-- 
GitLab


From d57ba095e4f170963ec420d6cd780aa19459bc65 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 16 Jun 2023 11:31:47 +0200
Subject: [PATCH 1790/1791] drm/i915: make i915_drm_client_fdinfo() reference
 conditional again

The function is only defined if CONFIG_PROC_FS is enabled:

ld.lld: error: undefined symbol: i915_drm_client_fdinfo
>>> referenced by i915_driver.c
>>>               drivers/gpu/drm/i915/i915_driver.o:(i915_drm_driver) in archive vmlinux.a

Use the PTR_IF() helper to make the reference NULL otherwise.

Fixes: e894b724c316d ("drm/i915: Use the fdinfo helper")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org> # build-tested
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230616093158.3568480-1-arnd@kernel.org
(cherry picked from commit 8084c63743a88472af0a34ba209eebf9caea1dae)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/i915_driver.c     | 2 +-
 drivers/gpu/drm/i915/i915_drm_client.h | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 75cbc43b326dd..0ad0c5885ec27 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -1816,7 +1816,7 @@ static const struct drm_driver i915_drm_driver = {
 	.open = i915_driver_open,
 	.lastclose = i915_driver_lastclose,
 	.postclose = i915_driver_postclose,
-	.show_fdinfo = i915_drm_client_fdinfo,
+	.show_fdinfo = PTR_IF(IS_ENABLED(CONFIG_PROC_FS), i915_drm_client_fdinfo),
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
diff --git a/drivers/gpu/drm/i915/i915_drm_client.h b/drivers/gpu/drm/i915/i915_drm_client.h
index 4c18b99e10a4e..67816c912bca1 100644
--- a/drivers/gpu/drm/i915/i915_drm_client.h
+++ b/drivers/gpu/drm/i915/i915_drm_client.h
@@ -47,8 +47,6 @@ static inline void i915_drm_client_put(struct i915_drm_client *client)
 
 struct i915_drm_client *i915_drm_client_alloc(void);
 
-#ifdef CONFIG_PROC_FS
 void i915_drm_client_fdinfo(struct drm_printer *p, struct drm_file *file);
-#endif
 
 #endif /* !__I915_DRM_CLIENT_H__ */
-- 
GitLab


From 274d4b96b12f78cef4f72a97a4967032233f6cae Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 6 Jun 2023 11:23:56 +0300
Subject: [PATCH 1791/1791] drm/i915: Fix a NULL vs IS_ERR() bug

The mmap_offset_attach() function returns error pointers, it doesn't
return NULL.

Fixes: eaee1c085863 ("drm/i915: Add a function to mmap framebuffer obj")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZH7tHLRZ9oBjedjN@moroto
(cherry picked from commit 3a89311387cde27da8e290458b2d037133c1f7b5)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_mman.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
index 4e7a838ab7bd9..aa4d842d4c5a8 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c
@@ -1085,8 +1085,8 @@ int i915_gem_fb_mmap(struct drm_i915_gem_object *obj, struct vm_area_struct *vma
 		/* handle stolen and smem objects */
 		mmap_type = i915_ggtt_has_aperture(ggtt) ? I915_MMAP_TYPE_GTT : I915_MMAP_TYPE_WC;
 		mmo = mmap_offset_attach(obj, mmap_type, NULL);
-		if (!mmo)
-			return -ENODEV;
+		if (IS_ERR(mmo))
+			return PTR_ERR(mmo);
 	}
 
 	/*
-- 
GitLab